(1)trabajos astrobiologia

Research Papers on Astrobiology

CĂŠsar Menor SalvĂĄn May 2015

Summary The present volume comprehends the research papers published by Dr. Cesar Menor Salvan regarding Astrobiology and specifically the question of Origins of Life, Chemical Evolution and Prebiological Chemistry.

Papers published as main contributor or principal researcher

Ibáñez de Aldecoa AL, Velasco F, Menor-Salván C (2013). Natural pyrrhotite as catalyst in prebiotic chemical evolution. Life; 3:502-517. DOI: 10.3390/life3030502 Menor-Salván, C., & Marín-Yaseli, M. R. (2013). A new route for the prebiotic synthesis of nucleobases and hydantoins in water/ice solutions involving the photochemistry of acetylene.

Chemistry A European Journal, 19, 6488–6497. doi:10.1002/chem.201204313 Menor-Salván, C., & Marín-Yaseli, M. R. (2012). Prebiotic chemistry in eutectic solutions at the water-ice matrix. Chemical Society Reviews, 41, 5404–5415. doi:10.1039/c2cs35060b Menor-Salván C, Ruiz-Bermejo M, Guzmán M, Osuna-Esteban S, Veintemillas-Verdaguer S (2009) Synthesis of pyrimidines and triazines in ice: implications for the prebiotic chemistry of nucleobases. Chemistry-A European Journal, 15: 4411-4418. DOI: 10.1002/chem.200802656 Menor-Salván C, Ruiz-Bermejo M, Osuna-Esteban S, Muñoz-Caro G, Veintemillas-Verdaguer S (2008). Abiotic synthesis of polycyclic aromatic hydrocarbons in aqueous environment: a new prebiotic scenario. Chemistry and Biodiversity, 12:2729-2739. DOI: 10.1002/cbdv.200890228 Ruiz-Bermejo M, Menor-Salván C, Osuna-Esteban S, Veintenillas-Verdaguer S (2007). The effects of ferrous and other ions on the abiotic formation of biomolecules using aqueous aerosols and spark discharges. Origins of Life and Evolution of Biospheres, 37(6):507-521.

DOI: 10.1007/s11084-007-9107-0 Ruiz-Bermejo M, Menor-Salván C, Osuna-Esteban S, Veintenillas-Verdaguer S (2007) Prebiotic Microreactors: A Synthesis of Purines and Dihydroxy Compounds in Aqueous Aerosol. Origins of Life and Evolution of Biospheres , 37:123-142. DOI: 10.1007/s11084-006-9026-5

Reprint

Supported by

www.chemeurj.org

Prebiotic Chemistry C. Menor-Salv n,* M. R. Mar n-Yaseli . . . . . . . . . . 6488 – 6497 A New Route for the Prebiotic Synthesis of Nucleobases and Hydantoins in Water/Ice Solutions Involving the Photochemistry of Acetylene

2013 – 19/20

Greatness from small beginnings: The ultraviolet irradiation of a urea solution subjected to freeze–thaw cycles under an acetylene atmosphere leads to the formation of pyrimidines, purines, and hydantoin (see scheme).

This work describes the possible pathways to different nitrogen heterocycles from these precursors. This reaction completes the model of the abiotic origin of nucleobases in our solar system and on prebiotic Earth.

DOI: 10.1002/chem.201204313

A New Route for the Prebiotic Synthesis of Nucleobases and Hydantoins in Water/Ice Solutions Involving the Photochemistry of Acetylene C¦sar Menor-Salv n* and Margarita R. Mar n-Yaseli[a] Dedicated to Professor Juan P¦rez-Mercader Abstract: The origin of nucleobases and other heterocycles is a classic question in the chemistry of the origins of life. The construction of laboratory models for the abiotic synthesis of nitrogen heterocycles in plausible natural conditions also aids the understanding and prediction of chemical species in the Solar System. Here, we report a new explanation for the origin of hydantoins, purines, and pyrimidines in eutectic water/ice/urea solutions driven by ultraviolet irradiation (in the 185– 254 nm range, UVC) of acetylene

under anoxic conditions. An analysis of the products indicates the synthesis of hydantoin and 5-hydroxyhydantoin, the purines uric acid, xanthine, and guanine, and the pyrimidines uracil and cytosine. The synthesis occurred together with the photo-oxidation of bases in a complex process for which possible pathways are proposed. In conclusion, Keywords: heterocycles nucleobases · photochemistry prebiotic chemistry · urea

Introduction It is likely that the genetic code has been a feature of life on Earth since the existence of the first organisms,[1] but the origin of the nucleic acids in the first life-forms remains unknown. From a chemical perspective, we could surmise that regardless of whether RNA is prebiotic,[2] the origin of informational polymers was preceded by the synthesis of a stock of nucleobases and related molecules in a chemical evolutionary process. This premise is attractive because the retrosynthetic analyses of nucleobases leads to small precursors whose prebiotic existence is plausible. The prebiotic stock of nitrogen heterocycles could also have been imported from extraterrestrial sources by meteoritic bombardment. From this point of view, models for the natural abiotic synthesis of nitrogen heterocycles could explain the origin of the nucleobases found in meteorites and their potential presence as icy grains in dense molecular clouds or planetary environments.[3] Since the prebiotic synthesis of adenine, achieved in 1961 by hydrogen cyanide polymerization,[4] three main routes for

· ·

an acetylene-containing atmosphere could contribute to the origin of nucleobases in the presence of a urea/water system by an HCN-independent mechanism. The presence of ice has a dual role as a favorable medium for the synthesis of nucleobases and protection against degradation and as a source of free radicals for the synthesis of highly oxidized heterocycles. A mechanism for the origin of hydantoins and uracil from urea in plausible conditions for prebiotic chemistry is also proposed.

the prebiotic synthesis of pyrimidines and purines have been identified (Scheme 1): 1) the formation of purines by HCN oligomerization,[5] 2) the formation of pyrimidines from cyanoacetylene or b-alanine and urea or guanidine[6] and 3) the synthesis of both purines and pyrimidines by using formamide as a precursor in the presence of mineral catalysts.[7] If we support the idea of a prebiotic origin of nucleobases, a

[a] Dr. C. Menor-Salv n, M. R. Mar n-Yaseli Centro de Astrobiolog a (CSIC-INTA) Ctra. Torrejýn-Ajalvir km. 4 28850 Torrejýn de Ardoz (Spain) E-mail: menorsc@cab.inta-csic.es Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201204313.

6488

Scheme 1. Summary of the main currently identified prebiotic pathways to purines and pyrimidines.

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2013, 19, 6488 – 6497

FULL PAPER “cold origin” or “ice world” is the preferred prebiotic environment because the nucleobases would be stable under cold conditions, and such an environment supports the relevance of eutectic water solutions of reactants in an ice matrix.[8] In a previous publication, we demonstrated that the synthesis of pyrimidines and purines is possible in an ice matrix using eutectic urea solutions by sparking a methane/nitrogen atmosphere.[9] We explained this synthesis of nitrogen heterocycles by invoking the two classic mechanisms of cyanoacetylene and hydrogen cyanide formation because both reactive molecules are products of the heating or sparking of CH4/N2 atmospheres.[10] In our analysis of the gas mixture obtained after sparking a CH4/N2 mixture, the main product was acetylene, followed by unsaturated hydrocarbons and HCN, but no cyanoacetylene was found.[11] Additionally, acetylene could be formed by the photolysis of methane[12] and is present as one of the minor components in the atmosphere of Titan[13] and possibly in other icy bodies in the solar system.[14] Acetylene could thus be a source material for the abiotic formation of polycyclic aromatic hydrocarbons and polymeric photopolymers that could constitute the haze observed in the atmospheres of Titan and Jupiter.[15] Surprisingly, acetylene has received comparatively less attention in the construction of laboratory models for prebiotic chemical evolution.

The photochemistry of acetylene is a complex process in which ultraviolet irradiation in the presence of water leads to the formation of glyoxal, formaldehyde, oxalic acid, glyoxylic acid, and formic acid in a mechanism driven by hydroxyl radicals generated from the photolysis of water.[16] Glyoxal is a precursor of hydantoin, a common prebiotic molecule, and glyoxylic acid is a prebiotic precursor of increasing interest.[17] Despite these findings, the photolysis of acetylene in the presence of water under anoxic conditions has not been widely studied, and we cannot disregard the formation of other precursors that could lead to nitrogen heterocycles by condensation with urea and ring-closure.

In this work, the potential relevance of acetylene photooxidation in the origin of nucleobases was explored. We investigated if an ice matrix could be a favorable environment for the synthesis of nucleobases and explored the pathways for their abiotic formation in the context of secondary photo-oxidation processes in a plausible prebiotic scenario.

Results and Discussion Following the previous experimental model,[9] a solution of urea (0.1 m) in ultrapure water was subjected to freeze–thaw cycles (beginning from the eutectic point of the urea/water system, ¢21 C, to 5 C) in a sealed reactor under a pure acetylene atmosphere for three weeks (Figure S1, the Supporting Information). The system was irradiated with UV light (254 nm) for the first 72 h. After the end of the experiment, the reactor content was allowed to warm to room temperature, collected in sealed vials and stored at ¢80 C. Chem. Eur. J. 2013, 19, 6488 – 6497

For the analysis of nitrogen heterocycles, the remaining urea was separated, and the fractions containing heterocycles were collected using cation exchange resin. The retained compounds were eluted and analyzed by gas chromatography-mass spectrometry (GC-MS) after derivatization as trimethylsilyl ethers/esters using total ion count (TIC) mode and selected ion mode (SIM). The ions at m/z 254, 257, and 270 were monitored for cytosine and the hydroxylated and reduced forms of pyrimidines.

The reaction products (Figure 1 and Figure 2) contain hydantoins, pyrimidines and purines, including uracil, uric acid, xanthine, guanine, and adenine. The observation of purines as their 8-hydroxy derivatives did not necessarily indicate a photo-oxidation process and could be an artifact from the formation of trimethylsilyl derivatives,[18] The highest yields corresponded to 5-hydroxyhydantoin, 2-oxo-4,5-dihydroxyimidazolidine, hydantoin, uracil, parabanic acid, and uric acid (Table 1). Additionally, cytosine, 6-methyluracil (not Table 1. Selected yields and mass spectrometric data for purines, pyrimidines, and hydantoins as their trimethylsilyl derivatives.

Product

Yield[a] m/z ([%])

hydantoin

66

5-hydroxyhydantoin uracil

128 27

cytosine[b]

3.6

dihydroorotic acid[c] uric acid[d] xanthine guanine

4.1 33 16 14

244(20), 229(100), 147(88), 100(15), 86(40), 73(50) 332(3), 317(24), 202(20), 189(33), 174(9), 147(100), 100(12), 73(80) 256(52), 255(50), 241(100), 147(42), 113(26), 99(60), 73(41) 254(100), 240(89), 170(19), 147(12), 112(25), 98(39), 73(41) 359(10), 331(5), 257(28), 147(30), 100(59), 73(100) 456(51), 441(57), 383(15), 147(18), 73(100) 368(12), 353(30), 294(9), 147(31), 73(100) 367(15), 352(34), 264(9), 147(15), 99(9), 73(100)

[a] Yield defined as mg of product per gram of acetylene used in the experiment. [b] Expressed as sum of cytosine and dihydrocytosine. [c] TriTMS derivative. [d] Tetra-TMS derivative.

visible in the TIC analysis due to coelution), and iso-orotic and orotic acids were found in significant quantities by using SIM analysis at m/z 254 (Figure S2, the Supporting Information). The formation of dihydrocytosine and dihydroorotic acid was observed in the SIM GC/MS analysis at m/z 257 (Figure S2, the Supporting Information). The expected formation of solid photopolymers[16] was observed in an acetylene/water irradiation control experiment (Figure S3, the Supporting Information) but was inhibited in the presence of urea, indicating that urea acted as a sink for reactive species generated during irradiation. The major aqueous products found, after UV irradiation of acetylene in the presence of water subjected to freeze–thaw cycles under anoxic conditions, were glycolic acid and succinic acid. Significant quantities of acetic acid, glyoxylic acid, oxalic acid, and benzoic acid were also observed.

A control experiment was performed entirely at low temperature in eutectic conditions of the urea/water system

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

6489

C. Menor-Salv n and M. R. Mar n-Yaseli

Figure 1. Total ion chromatogram of the unambiguously identified TMS-derivatized products after UV irradiation of a urea solution subjected to freeze– thaw cycles under an acetylene atmosphere.

(¢21 C, a solid solution at our standard urea concentration). Under these conditions, we found that yields of hydantoins and uracil, the major products identified, were significantly lower than those found for the freeze–thaw cycle experiments. Glycolic, glyoxylic and benzoic acids, together with evidences of uracil photoproducts, were also found. Cytosine and purines fell under the detection limit.

The synthesis of purines (Scheme 2) can be achieved following two pathways. In the first pathway, guanine and adenine are formed first, followed by deamination to yield xan-

Scheme 2. Synthesis of purines by irradiation of urea/acetylene.

6490

www.chemeurj.org

thine and hypoxanthine. Then, uric and parabanic acids are produced as the final photoproducts of guanine[19] .

We hypothesized a second pathway involving hydantoin as the precursor of the purine ring. To test this possibility, an experiment was performed by using urea (0.1 m) and hydantoin (0.025 m) subjected to freeze–thaw cycles and UV irradiation under an inert (argon) atmosphere (Figure S4, the Supporting Information). The products found in the TIC GC-MS chromatogram were 5-hydroxyhydantoin, an oxidation product of hydantoin, and parabanic acid. Significant quantities of uric acid and its photoproducts, allantoin, 5-hydroxy-hydantoin-5-carboxamide, and alloxanic acid were also found.[20] Based on mass spectrum interpretation, we also tentatively identified 5-hydroxyisouric acid, which yields allantoin or 5-hydroxy-hydantoin-5-carboxamide upon hydrolysis and loss of ammonia and carbon dioxide. This result suggested the formation of the purine ring from hydantoin followed by a purine photo-alteration pathway similar to the non-enzymatic uric acid degradation associated with biological processes.[20] This pathway is initiated by uric acid, an efficient scavenger of the free radicals generated by UV irradiation. The hydroxyl radical-addition to the purine

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2013, 19, 6488 – 6497

Prebiotic Synthesis of Nucleobases and Hydantoins

FULL PAPER

Figure 2. Continuation of the chromatogram in Figure S2 (the Supporting Information), showing the TMS derivatives of purines obtained after UV irradiation of a urea solution subjected to freeze–thaw cycles under an acetylene atmosphere. Allantoin was detected as tri- and tetra-(trimethylsilyl) derivatives. Adenine was not found, however, a peak for 8-hydroxyadenine was noted instead. 8-hydroxyadenine, along with 8-oxoguanine, could be derivatization artifacts of adenine and guanine and not real compounds synthesized in the experiment.[18] These compounds are therefore considered to be indirect evidence of the formation of purine nucleobases.

ring then yields the hydroxylated form of uric acid, which degrades to allantoin through ring-opening and decarboxylation, or to alloxanic acid through oxidation, hydrolysis, and loss of urea (Scheme 3).

Parabanic acid has not yet been observed in the irradiation of hydantoin solution subjected to freeze–thaw cycles under an inert atmosphere (Figure S5, the Supporting Information). In the absence of urea, hydantoin irradiation mainly afforded 2-oxo-5-hydroxyimidazole, and 5-hydroxyhydantoin. This finding suggests that parabanic acid was generated in the water/ice solution as the final photo-alteration product of purines. The mechanisms of formation of uric acid from urea and hydantoin in our experimental model are not currently well-understood. However, we identified the formation of glycine (a hydrolysis product of hydantoin) and a very small proportion of N-formylglycine.

This result indicates the formation of single carbon active molecules or an organocatalytic carbodiimide/urea cycle.[21] The formation of carbodiimide by dehydration of urea in the eutectic, concentrated brines in an ice matrix and the acidic medium generated during the reaction leads to a highly electrophilic reactant that activates the carboxyl of glycine and favors formylation. How the formation of urea Chem. Eur. J. 2013, 19, 6488 – 6497

Scheme 3. Synthesis of uric acid from hydantoin and urea under an inert atmosphere.

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

6491

C. Menor-Salv n and M. R. Mar n-Yaseli

derivatives, single carbon active species, and carbodiimide dantoin identified in the experiment performed by using drives the formation of the purine ring from hydantoin is deuterated acetylene were originated by the photo-oxidation unclear and is currently under investigation. of pyrimidines, a major proportion of deuterium-free 5-hyHydantoin, together with 5-hydroxyhydantoin and paradroxyhydantoin should be observed. However, the photobanic acid, is a product of the degradation of pyrimidines by oxidation of thymine leads to the formation of 5-methyl-5hydroxyl radicals[22] or other reactive-oxygen species[23] genhydroxyhydantoin,[26] indicating that C5 in the hydantoin ring comes from C5 in the pyrimidine ring. If the real mecherated by the UV irradiation of water.[16c,d] Under our condianism of photo-oxidation of pyrimidines preserves the hytions, the irradiation of aqueous cytosine and uracil soludrogen at the C5 position, then the high yield of 5-hydroxytions did not yield hydantoin (Figure S6, the Supporting Inhydantoin-5[D] obtained would be consistent with the degformation). As a consequence, photo-oxidation of previously radation of pyrimidines. By comparison, the photo-oxidation generated nucleobases does not explain the presence of a of hydantoin leads preferentially to the formation of 2-oxorelatively high amount of hydantoin and 2-oxo-4,5-dihydrox5-hydroxyimidazole and a minor amount of deuterium-free yimidazolidine. However, given that glyoxal is one of the 5-hydroxyhydantoin, possibly originating from the photo-oxmajor products of acetylene irradiation through the addition idation of uric acid or other purines. Anyway, the yield of 5of hydroxyl radicals,[16, 24] the formation of hydantoin by the hydroxyhydantoin observed after the irradiation of cytosine reaction of glyoxal with urea could be achieved under our and uracil is low (estimated 13 % of transformation of cytoexperimental conditions. To provide evidence supporting sine and uracil) compared with the yield of the main experithis hypothesis, two experiments were performed. In the ment, suggesting a direct synthetic pathway. Consequently, first, a solution of urea (0.1 m) and glyoxal (0.01 m) (simulatthe main possible source for 5-hydroxyhydantoin-5[D] is ing the quantitative transformation of acetylene into glyoxring-closure of the a-hydroxyhydantoic acid, produced by al) was submitted to freeze–thaw cycles under an acetylene condensation of urea and glyoxylic acid, favored by the atmosphere without UV irradiation. After the given reaction acidic medium generated by acetylene irradiation time, high yields of 2-oxo-4,5-dihydroxyimidazolidine and (Scheme 5). Glyoxylic acid is a product of the UV irradiahydantoin were found, but the photo-oxidation products 5tion of acetylene or acetic acid in presence of water, as well hydroxyhydantoin and 2-oxo-5-hydroxyimidazole were not as a product of oxidation of glyoxal. By using deuterated observed (Figure S7, the Supporting Information). Second, acetylene, the reaction would result in the a-deuterated the irradiation of dideuteroacetylene over a solution of urea form of a-hydroxyhydantoic acid and, hence, deuterated 5(0.1 m) subjected to freeze–thaw cycles yielded 2-oxo-4,5-dihydroxyhydantoin should be obtained. To confirm this pathhydroxyimidazolidine-4,5[D2] and hydantoin-5,5[D2] with a way, a solution of urea (0.1 m) and glyoxylic acid (0.01 m) minor amount of deuterium-free hydantoin. On the basis of was submitted to freeze–thaw cycles under an inert (argon) these experiments, the probable mechanism for the synthesis of hydantoin is through a pinacol/pinacolone-type rearrangement of 2-oxo-4,5-dihydroxyimidazolidine (Scheme 4).

The experimentation with deuterated acetylene led to the formation of 5-hydroxyhydantoin-5[D], which could be the result of a direct synthesis by condensation of acetylene photoproducts and urea or the result of the photo-oxidation of pyrimidines and, to a lesser extent, hydantoin (See Figure S8 (the Supporting Information) for a summary of the deuterated products found).

The mechanism for the formation of 5-hydroxyhydantoin by photo-oxidation of pyrimidines is unclear, but the most accepted explanation is the formation of dialuric acid or alloxan by hydroxyl radical-addition, followed by loss of carbon.[25] In that case, if the 5-hydroxyhy- Scheme 4. Synthesis of hydantoin from acetylene and a water/urea solution.

6492

www.chemeurj.org

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2013, 19, 6488 – 6497

Prebiotic Synthesis of Nucleobases and Hydantoins

Scheme 5. Synthesis of hydantoins in a urea/ice matrix under an acetylene atmosphere and UV irradiation. The route to 5-hydroxyhydantoin by ring-closure of 2-hydroxyhydantoic acid is prevalent against the formation of pyrimidine ring in the isolated glyoxylic-urea system.

atmosphere with UV irradiation. After the given reaction time, high yields of 5-hydroxyhydantoin and possible photooxidation products were found (Figure S9, the Supporting Information).

The reaction between urea and glyoxylic acid could also be the origin of uracil by the Davidson and Baudisch synthesis.[27] Although the reaction originally occurred between malic acid and urea in concentrated sulfuric acid, the key step in uracil formation was the condensation of urea and glyoxylic acid, generated from malic acid dehydration in a sulfuric acid solution. The freezing cycles leading to brines with concentrated organic solutes would favor these final condensation steps. The experiment performed by ultraviolet irradiation of a water/ice solution of urea and glyoxylic acid under an inert atmosphere led to the formation of a low yield of uracil, partially confirming this scheme. However, the low quantity of uracil found does not explain the yield observed in the main experiment, suggesting additional formation mechanisms or the lack of key reactants or catalysts in the complex mixture formed by the irradiation acetylene atmosphere.

The formation of cytosine-5,6[D2] was observed in the same experiment as evidenced by the two mass unit increases of the [M¢1] + and [M¢15] + ions in the mass spectrum.

The presence of uracil-6[D] was inferred on the basis of the net shift of the same ions by one unit and the unaltered m/z 99 ion in the silylated uracil mass spectrum (C4H7OSi + , the fragment includes the C4- and C5-ring carbons). These results indicate that the C5 position was unsubstituted in the pyrimidine ring. Additionally, the formation of deuterated 5-hydroxyuracil and undeuterated 6-hydroxyuracil confirmed this structure. This conclusion suggests that the synthesis of pyrimidines follows three pathways. The first pathway is the synthesis of cytosine and subsequent formation of 5,6-dihydrocytosine and 5-hydroxycytosine (although this could be an artifact, see ref. [18]). The second pathway is the aforementioned synthesis of uracil from glyoxylic acid and urea by the Davidson and Baudich process. Third, the uracil synthesis could proceed in a pathway parallel to the Chem. Eur. J. 2013, 19, 6488 – 6497

FULL PAPER biological synthesis of orotic acid:[28] ureido-intermediates are formed by the reaction of urea with reactive carbonyl compounds generated by acetylene/water irradiation[29] and the ring is closed to produce dihydroorotic acid and dihydrouracil. This reaction would be followed by dehydrogenation to orotic acid and photochemical decarboxylation to dihydrouracil and uracil,[30] which coexist in an equilibrium.[31] The detection of undeuterated orotic acid, dihydroorotic6[D], and malic acid support this mechanism.

Other pathways[32] that could yield uracil are the condensation of a,b-unsaturated acids with urea or the Fischer and Roeder synthesis of dihydrouracil from urea and acrylic acid, which could be a photochemical product of acetylene or its derivatives acetaldehyde and acrolein,[33] followed by photochemical dehydrogenation to uracil. Therefore, one possible mechanism for the formation of iso-orotic acid and 6-methyluracil, deuterated at C6 and the methyl group, respectively, is that both compounds could be derivatives of uracil after the addition of formaldehyde or acetate at C6.[34, 6b] Thymine was not found in the mixture, and the reason for the preference for addition at C6 was not clear. Thus, the direct synthesis of iso-orotic acid from urea and other intermediates derived from acetylene, followed by decarboxylation to uracil cannot be disregarded. Taking these results together with the addition of the hydroxyl radical and the photochemical alteration of the products,[20a] a general scheme could be suggested for the chemistry of pyrimidines in an acetylene/urea/ice system (Scheme 6).

Overall, the low temperature of the reaction and the formation of eutectic urea solutions in the ice matrix favored the formation of nitrogen heterocycles through the photolysis of acetylene, formation of carbonyl compounds and condensation with urea, and the protection of both purines and pyrimidines against degradation (Figure S10, the Supporting Information). The polymerization of acetylene with incorporation of nitrogen from cyanic acid has been hypothesized for the origin of nitrogen heterocycles in condensates under interstellar space conditions.[16d] Here, we demonstrate that nitrogen heterocycles could be formed by ultraviolet irradiation of acetylene-containing atmospheres. This formation depends on the presence of an efficient condensing agent and a nitrogen source, such as urea, whose presence in a planetary environment at a pre-biochemical evolutionary stage is supported by its facile synthesis from plausible precursors.[35] The difference between this route and that from synthesis under astrophysical conditions, in which the formation of pyrimidines could proceed through hydroxyl addition to pyrimidine or the reaction of cumulene species with urea,[36] is the role of water. UV irradiation of acetylene in the presence of water leads to reactive carbonyl compounds even in the absence of atmospheric oxygen.

A major difference with spark-discharge experiments[9] when considering the role of urea is the lack of s-triazines.

We explained the formation of s-triazine by the decomposition of urea in ammonia and isocyanate and reaction of the isocyanate with biuret, also generated by sparking. This find-

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

6493

C. Menor-Salv n and M. R. Mar n-Yaseli

the radiation source.[38] Another possible route is the uptake of acetylene into the water/ice phase, in which the generation of reactive-oxygen species under UV radiation could be favored at lower energies. The high-energy ultraviolet radiation, that is, vacuum ultraviolet radiation (VUV) or extreme ultraviolet radiation (EUV), has been used usually in the study of the photochemical origin of organic molecules from an astrochemical point of view. The irradiation with VUV of acetylene/water mixtures at ¢263 C and ultrahigh vacuum leads to hydroxylated photoproducts.[39] However, the production of organic molecules driven by the Scheme 6. Summary of prebiotic chemistry of pyrimidines in a urea/ice matrix initiated by the UV irradiation of acetylene. photodissociation of water at lower photon energies and higher pressures has been scarcely studied. The ultraviolet irradiation of water/ice at ing is consistent with the results observed under UV irradiaatmospheric pressure and at wavelengths of more than tion, in which triazines were absent, indicating that the for200 nm generates molecular oxygen, hydrogen peroxide, hymation of ammonia, isocyanate, and biuret from urea was droxyl radicals, and atomic hydrogen.[38, 40] At higher presnegligible. This fact also discards possible HCN formation [37]

by UV irradiation of ammonia and acetylene. To test this sures, the energy necessary for the water photodissociation shows a redshift and, in the GPa range, the generation of possibility, a control experiment was performed with a soluhydroxyl radical from water/ice has been reported at 350 nm tion of ammonia subjected to freeze–thaw cycles and UV irirradiation.[41] In our conditions, the combination of higherradiation under acetylene. Analysis of the products did not reveal purines or pyrimidines but did contain 1H-imidazole, energy photons in the gas phase and the photoreactions 1H-imidazole-2-carboxaldehyde, and a small amount (< 1 % driven by lower energy photons absorbed in the condensed of total imidazoles) of hydantoin (Figure S11, the Supportphase, which could generate radical initiators from water ing Information). Other indications of the unavailability of and/or organic solutes, could explain the importance of the HCN from irradiation of urea/acetylene is the low yield of reactive-oxygen species in the system. One of these species adenine with respect to guanine and uric acid and the lack is molecular oxygen, which could be photochemically generof 2,4-diaminopyrimidine, a common product found in preated in water/ice, as well as being present as a trace in the biotic chemistry experiments involving the cyanide pathexperimental setup. Oxygen is an interesting initiator; way.[3a, 9] In addition, we did not detect any 4-aminoimidathrough the O2 dissociation by irradiation at 185 nm, followed by hydroxyl radical generation in presence of water zoles, such as 4-aminoimidazole-5-carboxamide (AICA), or ozone formation and subsequent photolysis at which are classic intermediates in the synthesis of purines by 254 nm.[16e, 42] It could be interesting from a prebiotic evoluHCN polymerization.

Water also plays an essential role in the formation of photion point of view to establish how small proportions of motoproducts under UV irradiation. The atmospheric processlecular oxygen[43] influenced the photochemical production ing of acetylene in the presence of water in gas phase leads of relevant organic molecules. to the formation of reactive intermediates as glyoxal, which Although the reaction products studied in the water solucan enter the liquid/solid water pool readily due to its high tion were dominated by the reaction between acetylene and HenryÏs law constant. Once in the water pool, glyoxal could reactive-oxygen species, the photodissociation of acetylene react with urea or could be oxidized by hydroxyl radical has been evidenced by the presence of benzoic acid in the generated in the ice matrix to yield glyoxylic acid. This reaction products. Further work is necessary to study the alroute depends on the reactive-oxygen species generated by ternative routes initiated by acetylene-derived reactive spephotodissociation of water, which, moreover, are involved to cies, as ethynyl or vinylidene radicals. Independently of the explain the deuterated products found by using C2D2. Water first photochemical steps, our overall results show that the water/ice system irradiation with UVC constitutes a source photodissociation yielding the hydroxyl radical could take of reactive-oxygen species, evidenced by the generation of place in the gas phase, initiated by the 185 nm photons from 6494

www.chemeurj.org

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2013, 19, 6488 – 6497

Prebiotic Synthesis of Nucleobases and Hydantoins

FULL PAPER Conclusion

Scheme 7. Reconstruction of the photochemistry of acetylene in the water/ice system under UVC irradiation, based on the products found in water/urea- and pure water experiments.

oxalic, glyoxylic and glycolic acids, which have been regarded as products of the hydroxyl radical-mediated photooxidation of glyoxal and acetic acid[44] (Scheme 7). Our results show that the glyoxal is a precursor of larger dicarboxylic acids, as succinic acid (Figure S7, the Supporting Information). An alternative route could be the recombination product of radicals generated by UVC irradiation of the organic acids previously formed.[45] The production of glyoxylic acid from glyoxal (or by the photooxidation of other acetylene-derived precursors, as acetic acid), in the presence of urea leads to the formation of 5-hydroxyhydantoin and pyrimidines in our experimental model. This result could be connected with the glyoxylate scenario proposed by Eschenmoser, wherein the glyoxylic acid is the starting point towards several molecules with prebiotic interest, including sugars.[46, 17c] The possible role of glyoxylic acid as a mimic of phosphate in a plausible RNA ancestor[47] and as starting point to a primordial metabolism, strengthens the prebiotic interest of the production of glyoxylic acid and nucleobases in the same experimental model.

This paves the way to further research in the glyoxylate/ice scenario. The photochemical behavior of the nucleobases and hydantoin in the water/ice matrix, which has not been studied apart from experiments under astrophysical conditions, showed that the photoproducts found are comparable to those obtained in the biochemistry of bases and reveals pathways to form nucleobases that are compatible with the chemomimesis concept,[48] such as the synthesis of uric acid or uracil through dihydroorotic acid or iso-orotic acid. This new alternative mechanism minimizes the contribution of cytosine deamination to uracil synthesis and instead indicates a role for acetylene, a molecule thought to be accessible to planetary atmospheric chemistry and astrochemistry.

Chem. Eur. J. 2013, 19, 6488 – 6497

A cold environment (liquid water and ice) with cyclic freezing and thawing was exposed to an acetylene-rich atmosphere under anoxic conditions. Ultraviolet irradiation of the system led to the formation of reactive intermediates that enriched the water pool in organic acids including acetic, glyoxylic, glycolic, and succinic acids. If the water solution contained urea, UV irradiation led to the formation of nitrogen heterocycles such as hydantoin, pyrimidines, and purines, with 5-hydroxyhydantoin, hydantoin, uric acid, and uracil being the most abundant products.

Analysis of the reaction products suggests that hydantoin was a product of the direct synthesis from acetylene-derived glyoxal and urea. 5-hydroxyhydantoin was a product of both the direct synthesis from glyoxylic acid and urea, and, to a lesser extent, from the photodegradation of purines and pyrimidines. The water/ice matrix played a dual role as a protective medium and a source of radicals for the photo-oxidation of purines and pyrimidines. The results further suggest that the synthesis of the nucleobases proceeded through a HCN- or nitrile-independent pathway. Also, the formation of cytosine and uracil could run by independent pathways, being negligible the deamination of cytosine in our experimental conditions.

The availability of acetylene and urea in a prebiotic world and the yields of hydantoins recorded here could explain the origin of these molecules in prebiotic and astrochemistry laboratory simulations. Our results also contribute to the understanding of chemical evolution in cold, water-rich planetary environments.

Experimental Section Urea solution freeze–thaw cycles under an UV-irradiated acetylene atmosphere: Urea (0.1 m, 50 mL; Fluka Biochemika, Sigma Aldrich GmbH, Germany) in ultrapure, degassed water was frozen at ¢5 C in a sealed and thermostated glass reactor (Figure S1, the Supporting Information) under acetylene at atmospheric pressure. The acetylene was generated by reaction of calcium carbide (Panreac SA, Spain) with ultrapure water. The acetylene generated contained significant quantities of vinyl acetylene, divinylsulfide, and acetic acid, as well as traces of phosphine, sulfur dioxide, and other unsaturated hydrocarbons. The acetylene stream was purified by passing through a glass column filled with granular silica gel impregnated with 60 % phosphoric acid and heated to 175 C in a custom-made cylindrical furnace. Next, acetylene was deoxygenated and acetic and phosphoric acids removed by bubbling in a solution of pyrogallol in KOH (1 m). The purity of acetylene in the reactor was assessed by the injection of the gas (50 mL) into a PLOT-Q column mounted on a Perkin–Elmer Autosystem XL-Turbomass Gold quadrupole GC/MS system in scan EI mode in the range 20–200 amu. The analysis showed pure acetylene with traces of vinylacetylene and acetic acid.

Once the reactor was filled with purified acetylene (approximately 2.3 g or 0.09 mol of gas), freeze–thaw cycles were established by varying the temperature between ¢21 and 5 C (two hours at ¢21 C, ramping to 5 C at 0.1 C min¢1 and two hours at 5 C) by using a Haake Phoenix II programmable cryostat (Thermo Electron Corporation). The acetylene atmosphere was irradiated with UV radiation by using a Spectroline 11SC1 Hg vapor discharge lamp. This lamp emits photons corresponding to Hg emission lines with a main wavelength centered at (253.7 10) nm

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

6495

C. Menor-Salv n and M. R. Mar n-Yaseli

with an estimated flux of 5.8 × 1014 photons cm¢2 s¢1 from 1“ of the lamp surface and a wavelength centered at (185 5) nm that constitutes approximately 5 % of the UVC emission. The irradiance spectrum of the lamp shows small peaks in the near-UV and visible regions. The irradiation was performed for 72 h. After this period, the lamp was disconnected and the reactor was kept sealed, which then underwent active freeze– thaw cycles for three weeks. The system was then allowed to warm to room temperature and the yellowish and slightly acidic (mean pH 3.5) aqueous solution was stored in sealed headspace vials under an inert atmosphere and frozen in liquid nitrogen. The frozen samples were freezedried and stored at ¢20 C until analysis. For the control experiments using an inert atmosphere, argon of the maximum purity available (99.9995 %, Praxair SA, Spain) was used.

Deuterated-acetylene experiments: To explore the mechanism involved in our experiments, identical conditions were employed, but C2D2 was used instead of acetylene. Deuterated acetylene was generated by reaction of calcium carbide with deuterium oxide (containing 99 % deuterium; Aldrich Chemical Company, Inc. WI, USA).

Ion-exchange separation: The experiment products contained an excess of urea, which could interfere with uracil determination by GC-MS. To remove excess urea, the dried sample was dissolved in HCl (10 mL, 0.1 m) and stirred for 1 hour with Dowex 50W × 8–400 cation exchange resin in the H form. After stirring, the solution containing pure urea was removed and the resin, suspended in water, was transferred to a glass column. The solution was eluted and column washed sequentially with water, NH3 (0.1 m), and HCl (0.1 m) until the effluent was slightly acidic. The eluted fractions were combined and freeze-dried prior to GCMS analysis.

Gas chromatography/mass spectrometry: The solid, freeze-dried samples were derivatized with N,O-bis(trimethylsilyl) trifluoroacetamide (BSTFA) with 1 % trimethylchlorosilane (TMCS, provided by Pierce, Rockford, IL, USA). Briefly, the dried sample (1 mg) was combined with BSTFA TMCS (0.1 mL) in a dry glass vial, stirred and heated at 60 C for 3 h. After that, the sample (1 mL) was transferred to an Agilent GC 6850A gas chromatograph in the splitless mode with the injection port at 290 C. The analysis was performed using an HP-5 MS (Agilent), 5 % phenyl-95 % methylsiloxane capillary column (30 m × 0.25 mm i.d., 0.25 mm film). Helium was used as the carrier gas at a flow rate of 1.1 mL min¢1. The oven temperature, initially 40 C for 1.5 min, was ramped to 130 C at a rate of 5 C min¢1. Then, the temperature was ramped to 180 C at 10 C min¢1 and held for 10 min. Next, the temperature was raised to 220 C at 20 C min¢1 and held for 15 min. In the last step, the temperature was raised to 300 C at 10 C min¢1 and held for 18 min.

Mass spectrometric analysis was performed using an Agilent 5975C VL MSD quadrupole in EI mode with an ionization energy of 70 eV. The ion source temperature was set at 230 C and the quadrupole at 290 C. Identification of compounds was performed in scan mode with a range of 45–650 amu. The determination of cytosine and related compounds was performed by using scan fragmentograms and selected ion monitoring (SIM) mode at 254 and 257 amu.

When available, the identified compounds were confirmed against authentic standard (provided by Sigma–Aldrich) mass spectra and retention times. Controls were performed by using mixtures of standards to identify possible derivatization artifacts and to avoid ambiguities. Other organic compounds were identified by searching their mass spectra in the NIST database, or comparison of published mass spectra data and mass spectrum interpretation. For identification purposes, we considered only peaks with a signal-to-noise ratio over 20. Those peaks whose match probability in the database were below 90 % and/or tentatively or ambiguously identified were considered unidentified and not discussed in this paper.

6496

www.chemeurj.org

Acknowledgements This work was supported by grant AYA2009–13920-C02–01 from the Ministerio de Ciencia e Innovaciýn (Spain) and the Instituto Nacional de T¦cnica Aerospacial “Esteban Terradas”.

[1] J. P. Dworkin, A. Lazcano, S. L. Miller, J. Theor. Biol. 2003, 222, 127 – 134.

[2] a) J. L. Bada, Earth Planet. Sci. Lett. 2004, 226, 1 – 15; b) C. Anastasi, F. F. Buchet, M. A. Crow, A. L. Parkes, M. W. Powner, J. M. Smith, J. D. Sutherland, Chem. Biodiversity 2007, 4, 721 – 739.

[3] a) M. P. Callahan, K. E. Smith, H. J. Cleaves, J. Ruzicka, J. C. Stern, D. P. Glavin, C. H. House, J. P. Dworkin, Proc. Natl. Acad. Sci. USA 2011, 108, 13995 – 13998; b) Z. Martins, O. Botta, M. L. Fogel, M. A. Sephton, D. P. Glavin, J. S. Watson, J. P. Dworkin, A. W. Schwartz, P. Ehrenfreund, Earth Planet. Sci. Lett. 2008, 270, 130 – 136; c) S. A. Sandford, M. P. Bernstein, L. A. Allamandola, Astrophys. J. 2004, 607, 346 – 360.

[4] a) J. Orý, A. P. Kimball, Arch. Biochem. Biophys. 1961, 94, 217 – 225; b) J. Orý, A. P. Kimball, Arch. Biochem. Biophys. 1962, 96, 293 – 313.

[5] a) A. W. Schwartz, H. Joosten, A. B. Voet, BioSystems 1982, 15, 191 – 193.

[6] a) K. E. Nelson, M. P. Robertson, M. Levy, S. L. Miller, Orig. Life Evol Biosph. 2001, 31, 221 – 229; b) A. W. Schwartz, G.JF. Chittenden, Biosystems 1977, 9, 87 – 92.

[7] a) R. Saladino, M. Barontini, C. Cossetti, E. Di Mauro, C. Crestini, Orig. Life Evol. Biosph. 2011, 41, 317 – 330; b) R. Saladino, C. Crestini, F. Ciciriello, G. Constanzo, E. Di Mauro, Chem. Biodiversity 2007, 4, 694 – 720.

[8] a) A. Eschenmoser, Angew. Chem. 2011, 123, 12618 – 12681; Angew.

Chem. Int. Ed. 2011, 50, 12412 – 12472; b) C. Menor-Salv n, M. R. Mar n-Yaseli, Chem. Soc. Rev. 2012, 41, 5404 – 5415 and references therein.

[9] C. Menor-Salv n, M. Ruiz-Bermejo, M. I. Guzm n, S. Osuna-Esteban, S. Veintemillas-Verdaguer, Chem. Eur. J. 2009, 15, 4411 – 4418.

[10] L. E. Orgel, Orig. Life Evol. Biosph. 2004, 34, 361 – 369.

[11] M. Ruiz-Bermejo, C. Menor-Salv n, J. L. de La Fuente, E. MateoMart , S. Osuna-Esteban, J. A. Mart n-Gago, S. Veintemillas-Verdaguer, Icarus 2009, 204, 672 – 680.

[12] A. Bar-Nun, Icarus 1979, 38, 180 – 191.

[13] a) W. Zheng, D. Jewitt, R. I. Kaiser, J. Phys. Chem. A 2009, 113, 11174 – 11181; b) R. K. Khanna, Icarus 2005, 178, 165 – 170; c) D. W. Clarke, J. P. Ferris, Icarus 1997, 127, 158 – 172.

[14] C. Tornow, E. Kîhrt, U. Motschmann, Astrobiology 2005, 5, 632 – 650.

[15] a) C. Menor-Salv n, M. Ruiz-Bermejo, S. Osuna-Esteban, G. MuÇoz-Caro, S. Veintemillas-Verdaguer, Chem. Biodiversity 2008, 5, 2729 – 2739; b) F. Cataldo, J. Photochem. Photobiol. A 1996, 99, 75 – 81.

[16] a) D. Cremer, R. Crehuet, J. Anglada, J. Am. Chem. Soc. 2001, 123, 6127 – 6141; b) S. Hatakeyama, N. Washida, H. Akimoto, J. Phys.

Chem. 1986, 90, 173 – 178; c) D. J. Peeble, S. Das, C. von Sonntag, J. Phys. Chem. 1985, 89, 57 – 84; d) M. Nuevo, S. N. Milam, S. A. Sandford, Astrobiology 2012, 12, 295 – 314; e) A. G. Carlton, B. J. Turpin, K. E. Altieri, S. Seitzinger, A. Reff, H. J. Lim, B. Ervens, Atmos. Environ. 2007, 41, 7588 – 7602.

[17] a) P. D. de Marcellus, Astrobiology 2011, 11, 847 – 854; b) A. Eschenmoser, Chem. Biodiversity 2007, 4, 554 – 573; c) E. Ware, Chem. Rev. 1950, 46, 403 – 470.

[18] a) J. L. Little, J. Chromatogr. A 1999, 844, 1 – 22; b) T. G. England, A. Jenner, O. I. Aruoma, B. Halliwell, Free Rad. Res. 1998, 29, 321 – 330.

[19] C. Vialas, C. Claparols, G. Pratviel, B. Meunier, J. Am. Chem. Soc. 2000, 122, 2157 – 2167.

[20] a) K. J. Volk, R. A. Yost, A. Brajler-Toth, Anal. Chem. 1989, 61, 1709 – 1717; b) K. Kahn, P. Serfoto, P. A. Tipton, J. Am. Chem. Soc. 1997, 119, 5435 – 5442; c) S. K. Tyagi, G. Dryhurst, J. Electroanal.

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2013, 19, 6488 – 6497

Prebiotic Synthesis of Nucleobases and Hydantoins

[21] [22]

[23]

[24]

[25] [26]

[27] [28]

[29] [30] [31] [32]

Chem. 1987, 216, 137 – 156; d) H. Kaur, B. Hallywell, Chem-Biol. Interactions 1990, 73, 235.

R. Saladino, G. Botta, S. Pino, G. Constanzo, E. Di Mauro, Chem.

Soc. Rev. 2012, 41, 5526 – 5565. a) J. Hong, D. G. Kim, C. Cheong, K. J. Paeng, Microchem. J. 2001, 68, 173 – 182; b) T. J. Yu, R. G. Sutherland, R. E. Verrall, Can. J. Chem. 1980, 58, 1909 – 1915. a) J. R. Wagner, J. Cadet, Acc. Chem. Res. 2010, 43, 564 – 571; b) J. RiviÀre, F. Bergeron, S. Tremblay, D. Gasparutto, J. Cadet, J. Am. Chem. Soc. 2004, 126, 6548 – 6549; c) E. Gilbert, S. HoffmannGlewe, Water Res. 1992, 26, 1533 – 1540. a) R. Volkamer, P. J. Ziemann, M. J. Molina, Atmos. Chem. Phys.

2009, 9, 1907 – 1928; b) F. Wittrock, A. Richter, H. Oetjen, J. P. Burrows, M. Kanakidou, S. Myriokefalitakis, R. Volkamer, S. Beirle, U. Platt, T. Wagner, Geophys. Res. Lett. 2006, 33, L 16804.

G. Baccolini, C. Boga, C. Delpivo, G. Micheletti, Tetrahedron Lett.

2011, 52, 1713 – 1717.

M. Redrejo-Rodr guez, C. Saint-Pierre, S. Couve, A. Mazouzi, A. A. Ishchenko, D. Gasparutto, M. Saparbaev, PLoS ONE 2011, 6, e21039.

D. Davidson, O. Baudisch, J. Am. Chem. Soc. 1926, 48, 2379 – 2383.

Y. Yamagata, K. Sasaki, O. Takaoka, S. Sano, K. Inomata, K. Kanemitsu, Y. Inoue, I. Matsumoto, Orig. Life Evol. Biosph. 1990, 20, 389 – 399.

M. Terasaki, S. Nomoto, H. Mita, A. Shimoyama, Orig. Life Evol.

Biosph. 2002, 32, 91 – 98. a) A. Lazcano, Orig. Life Evol. Biosph. 2010, 40, 161 – 167; b) J. A. Ferris, P. C. Joshi, J. Org. Chem. 1979, 44, 2133 – 2137. a) G. J. F. Chittenden, A. W. Schwartz, Nature 1976, 263, 350 – 351; b) A. J. Varghese, Biochemistry 1971, 10, 4283 – 4290.

I. M. Lagoja, Chem. Biodiversity 2005, 2, 1 – 50.

Chem. Eur. J. 2013, 19, 6488 – 6497

FULL PAPER [33] J. P. Ferris, Orig. Life Evol. Biosph. 1992, 22, 109 – 134.

[34] R. F. Cline, R. M. Fink, K. Fink, J. Am. Chem. Soc. 1959, 81, 2521 – 2527.

[35] a) R. Navarro-Gonz lez, A. Negrýn-Mendoza, E. Chacýn, Orig.

Life Evol. Biosph. 1989, 19, 109 – 118.

[36] J. A. Kaye, D. F. Strobel, Icarus 1983, 54, 417 – 433.

[37] a) P. P. Bera, M. Nuevo, S. N. Milam, S. A. Sandford, T. J. Lee, J. Chem. Phys. 2010, 133, 104303; b) T. Wang, J. H. Bowie, Org. Biomol. Chem. 2012, 10, 652 – 662.

[38] A. Yabushita, T. Hama, M. Kawasaki, J. Photochem. Photobiol C: Photochem. Rev. 2013, DOI:10.1016/j.jphotochemrev.2013.01.001.

[39] C. Y. Robert Wu, D. L. Judge, B. M. Cheng, W. H. Shih, T. S. Yih, W. H. Ip, Icarus 2002, 156, 456 – 473.

[40] a) P. Kl n, I. Holoubeck, Chemosphere 2002, 46, 1201 – 1210; b) C. C. Wang, L. I. Davis Jr, J. Chem. Phys. 1975, 62, 53 – 55.

[41] M. Ceppatelli, R. Bini, V. Schettino, Proc. Natl. Acad. Sci. USA 2009, 106, 11454 – 11459.

[42] M. I. Stefan, J. R. Bolton, Environ. Sci. Technol. 1999, 33, 870 – 873.

[43] J. Kasting, Nat. Geosci. 2013, 6, 9 – 10.

[44] D. H. Parker, Acc. Chem. Res. 2000, 33, 563 – 571.

[45] K. E. Altieri, S. P. Seitzinger, A. G. Carlton, B. J. Turpin, G. C. Klein, A. G. Marshall, Atmos. Environ. 2008, 42, 1476 – 1490.

[46] V. N. Sagi, P. Venkateshwarlu, F. Hu, G. Meher, R. Krishnamurthy, J. Am. Chem. Soc. 2012, 134, 3577 – 3589.

[47] H. D. Bean, F. A. L. Anet, I. R. Gould, N. V. Hud, Orig. Life Evol.

Biosph. 2006, 36, 39 – 63.

[48] J. Peretý, Chem. Soc. Rev. 2012, 41, 5394 – 5403.

Õ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Received: December 4, 2012 Revised: February 15, 2013 Published online: March 27, 2013

www.chemeurj.org

6497

View Online / Journal Homepage / Table of Contents for this issue

This article was published as part of the

Downloaded by Centro de AstrobiologĂ­a on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Prebiotic chemistry themed issue Guest editors Jean-François Lambert, Mariona Sodupe and Piero Ugliengo

Please take a look at the issue 16 2012 table of contents to access other reviews in this themed issue

Chem Soc Rev

Dynamic Article Links

Cite this: Chem. Soc. Rev., 2012, 41, 5404–5415 www.rsc.org/csr

TUTORIAL REVIEW

Prebiotic chemistry in eutectic solutions at the water–ice matrixw Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Ce´sar Menor-Salva´n* and Margarita R. Marı´ n-Yaseli Received 29th February 2012 DOI: 10.1039/c2cs35060b A crystalline ice matrix at subzero temperatures can maintain a liquid phase where organic solutes and salts concentrate to form eutectic solutions. This concentration effect converts the confined reactant solutions in the ice matrix, sometimes making condensation and polymerisation reactions occur more favourably. These reactions occur at significantly high rates from a prebiotic chemistry standpoint, and the labile products can be protected from degradation. The experimental study of the synthesis of nitrogen heterocycles at the ice–water system showed the efficiency of this scenario and could explain the origin of nucleobases in the inner Solar System bodies, including meteorites and extra-terrestrial ices, and on the early Earth. The same conditions can also favour the condensation of monomers to form ribonucleic acid and peptides.

Together with the synthesis of these monomers, the ice world (i.e., the chemical evolution in the range between the freezing point of water and the limit of stability of liquid brines, 273 to 210 K) is an under-explored experimental model in prebiotic chemistry.

Introduction Life as we know it depends on interfacial redox and transport processes between liquid water and a system of lipid membranes with the associated protein machinery. It seems logical to assume that life emerged from liquid water solutions where relatively simple raw materials were synthesised or accumulated.

These solutions could be subjected to water–mineral matrix

Centro de Astrobiologı´a (INTA-CSIC), INTA, E-28850 Torrejo´n de Ardoz, Spain. E-mail: menorsc@cab.inta-csic.es; Tel: +32 91520 6458 w Part of the prebiotic chemistry themed issue.

Cesar Menor-Salvan studied Chemistry at the University of Alcala´ and obtained his PhD degree in Biochemistry in 2004, working on the metabolism and toxicology of thiolated purine bases. Since 2007 he has been a Research Scientist at the Centro de Astrobiologia (CAB) and started a line devoted to the prebiotic chemistry of nitrogen heterocycles and the origin of cofactors and proto-metabolic pathways.

His research Ce´sar Menor-Salva´n interests include Prebiotic Chemistry, the origins of biochemistry and the organic markers of biological evolution on Earth.

5404

Chem. Soc. Rev., 2012, 41, 5404–5415

interfacial chemistry or concentration and compartmentalisation processes, which ultimately leads to the emergence of life in a complexity increasing process. Consequently, to determine the possible compositions of the raw materials for the plausible first steps of abiotic evolution, pioneering experiments on prebiotic chemistry have been conducted in watersaturated atmospheres and liquid solutions,1 which are largely supported by a reductive atmosphere model.

The criticisms regarding an efficient atmospheric-liquid water origin for the organic components of the first biochemical processes on Earth arise from the lack of a universally accepted geochemical model for the Archean atmosphere.

Additionally, the classic prebiotic chemistry approach deals

Margarita Roig Marı´n-Yaseli obtained her degrees in Pharmacy and Biochemistry at the University of Zaragoza. Now she is a PhD student at the Centro de Astrobiologia, focused on the Prebiotic Chemistry of nitrogen heterocycles.

Margarita R. Marı´ n-Yaseli

This journal is

c

The Royal Society of Chemistry 2012

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

with the problem of the concentration and stability in liquid water of the plausible prebiotic reactants. These criticisms and the lack of experimental evidence supporting a model for the origin of biochemical pathways have led to two main schools of thought.

The first concept is the possibility of an in situ origin on Earth, which focuses on either water–mineral interfacial processes as a way for concentration and compartmentalisation of environmentally synthesised reactants2 or on the origin of chemoautotrophic pre-biochemical systems.3 The second concept argues that amino acids, nitrogen heterocycles and simple organic molecules and monomers could be synthesised by irradiation at very low temperatures in extra-terrestrial ice layers composed of water and other condensates.4 Ice is the most abundant form of water beyond the asteroid belt.5 The chemistry of ices at low temperatures followed by the delivery of the organic molecules on Earth by comets, meteorites and dust particles could have been an important source of organics on the prebiotic Earth and could have played a key role in early chemical evolution. The photochemistry and radiochemistry of outer solar system bodies and interstellar ices have received substantial attention.6 Despite the research into the photochemical transformations in ice from an astrochemical point of view, the study of the chemistry in the range of stability of the ice–water interface has not received much attention. This may be due to the scarcity of the defined conditions in the Solar System during the epoch of active prebiotic chemistry or the difficulties for demonstrating that these cold conditions existed in Hadean Earth.

The evidence for a liquid water subsurface ocean on Saturn’s moon Europa7 and the possible presence of water– ammonia eutectic brines or even a subsurface ocean in other outer giant planet satellites such as Titan8 or Enceladus9 rekindled the interest in liquid water prebiotic chemistry.

Moreover, the subsequent proposed steps for the emergence of cellular life have a limited temperature range, and a hot prebiotic Earth was regarded to be an unlikely environment for the origin of life by some authors.10 Miller and Orgel stated in 1974 that the emergence of biological organisation could only occur at temperatures below the melting point of the polynucleotide structure. After observing the instability of organic compounds in the prebiotic stages, these authors concluded that a temperature of 273 K would have been beneficial and that temperatures near the eutectic point of NaCl solutions (251.3 K) would have been even better.11 The low temperatures in planetary surface ices could be more conductive to the origin and the preservation of molecules that could be relevant for the emergence of life. In 1994, in one of the first explorations of the idea of an ice world-based origin of the life raw materials, Bada12 suggested that ice formations on early Earth could have preserved organic compounds against hydrolysis or photochemical degradation. Under plausible planetary conditions, the presence of liquid water at T o 273 K within an ice matrix creates a potential reactor where the synthesis or polymerisation of molecules of biological interest could occur. Herein, we will review our current knowledge of the chemical models that simulate possible prebiotic synthetic pathways in liquid water interfacial ice. The experimental This journal is

c

The Royal Society of Chemistry 2012

approaches developed in the literature are primarily focused on the RNA-world hypothesis of an abiotic origin of nucleic acids, as these studies provide experimental evidence for the abiotic synthesis and polymerisation of nitrogen heterocycles and nucleotides. Apart from the molecular evolutionary perspective for the emergence of life, exploring the chemistry in liquid inclusions confined in an ice matrix could explain and predict the composition of objects in the inner Solar System and icy planetary bodies.

The ice–liquid water system and its presence on the early Earth and in the Solar System The ice–liquid water system has not received much attention in the literature, including the chemical physics and astrochemical/ astrobiological literature. In the latter case, the experimental efforts are focused on low temperature condensates, where there is no evidence of a liquid interface and the ice is in its amorphous crystalline state. In the inner Solar System, including on Earth, ice occurs naturally in the crystalline form with two primary polymorphs, which are cubic and hexagonal. The crystallisation of water under current Earth surface conditions results in hexagonal ice Ih. The ice formed from liquid or heated from amorphous ice at temperatures between 100 and 130 K is crystalline, with a diamond-type cubic structure Ic.13 Cubic ice is metastable at T o 70 K and undergoes a transformation to the amorphous state (the stable form at these temperatures) via cosmic ray bombardment and ultraviolet irradiation.14 The irradiation diminishes the kinetic barrier between the metastable cubic ice form and stable amorphous ice form at lower temperatures.15 The crystallisation of ice Ih leads to the formation of various interfaces, such as ice–ice, ice–atmosphere and water–ice, as well as water–ice–mineral, which results from crystallisation of solutes by ice matrix exclusion or the presence of suspended mineral grains.16 The ice–ice and ice–atmosphere interfaces are not a distinct transition. Nuclear magnetic resonance studies of ice crystals indicate the existence of a liquid transition between the crystals or between the ice and the atmosphere. The thickness of this liquid phase becomes monomolecular at T o 243 K and is thickened by dissolved solutes excluded from the ice matrix to the interface during crystallisation.16 The unexpected presence of crystalline ice in the Quaoar object at the Kuiper Belt, on Enceladus and its suggested presence in Titan17 imply that the evolution of ices is subject to occasional heating events. If crystalline ice and if even fluid water solutions are unambiguously present, the conditions for the increase in organic complexity from reactions between precursors such as cyanide or cyanoacetylene may exist. The young and active surface of the Jovian moon Europa suggests the possibility of a subsurface water ocean from the observations of the Voyager mission and strengthened by the observations with the Galileo spacecraft.18 Recently, it has been stated that Europa possesses an active dynamic ice–water system with cycles of melting and refreezing. In addition, a lenticular body of liquid brine in the Thera Macula region of approximately 20 000–60 000 km3 has been predicted.19 The composition of Europa’s subsurface water, underlying an ice crust, could be rich in sulphate salts, the source of surface evaporite deposits.20 Chem. Soc. Rev., 2012, 41, 5404–5415

5405

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

The details on water composition and temperature are unknown, but estimations suggest a Mg–SO4–Na(K) rich water with temperatures in the range 210–270 K.21 A model for the formation of liquid ammonia–water pockets that cause episodic cryomagmatism and a subsurface eutectic water– ammonia solution has been proposed for the Saturn moon Titan.22 Within this context, both Titan and Europa constitute important astrobiological targets for direct exploration and laboratory simulations to predict the chemistry that will be found and to test our experimental prebiotic chemistry models.23 A complex prebiotic chemistry has been predicted for Titan that includes the formation of nucleobases24 and the possibility of a methane–acetylene based chemical or biochemical evolution.25 From these hypotheses based on atmospheric or surface chemistry, the prebiotic possibilities of liquid water brines entrapped under ice have received less attention and are the object of speculative discussion regarding possible biochemical evolution and the presence of chemoautotrophic life.26 Some models suggest a Hadean terrestrial atmosphere composed primarily of high pressure carbon dioxide. If liquid water were present in oceans over a basaltic crust, a CO2 atmosphere would be unstable and could be depleted as carbonates in a period of approximately 10 million years due to hydrothermal circulation and reaction of the CO2 with the crustal rock. Under these conditions, together with the Hadean faint Sun, the model developed by Sleep and Zahnle27 agrees with the ideas suggested by J. Bada in 1994,12 predicting ice-covered oceans and an average surface temperature of approximately 220 K, with freeze–thaw episodes motivated by occasional warming provoked by high energy impacts.

These cold conditions would be prevented if a methane-rich atmosphere were present during the Hadean, as methane is a potent greenhouse gas. Evidence thus far does not support an atmosphere with a high enough concentration of methane to avoid freezing of the ocean surface. This model would be amenable for the development of prebiotic chemistry in an ice matrix based on HCN, cyanoacetylene, acetylene, urea or cyanate precursors synthesised on Earth or brought in via extraterrestrial input.28 The freezing of ocean water is a complex process. Modern sea water begins to freeze at 271.2 K and crystals of pure ice (Ih) begin to grow, surrounded by liquid brine with sodium chloride concentrations up to 25%. The liquid solution is concentrated within the ice structure in channels, which have been observed in stained samples under the microscope, with diameters ranging from 10 to 100 mm.29 Based on observations of microscopic ice layers, it is estimated that 1 m3 of sea ice has a network of channels with a combined surface area of 105 to 106 m2. The volume of ice occupied by the brine channels and the brine conditions within the channels are directly proportional to the temperature; at 267 K, the brine salinity in sea ice is 100 (on the practical salinity scale, i.e., dimensionless units that are equivalent to the ratio between the sample solution and a standard KCl solution; normal ocean water has a salinity range of 30–35); at 263 K, the salinity rises to 145, and at 252 K, the salinity reaches a maximum of 216.30 In sea ice, the presence of interstitial channels filled with liquid water and concentrated solutes has been observed over a range of 5406

Chem. Soc. Rev., 2012, 41, 5404–5415

temperatures down to 243 K. Sea ice can lead to the formation of solid mineral phases from the crystallisation of dissolved salts.

During freezing or thawing events, the temperature gradients and density changes in the ice matrix lead to pressure gradients and motion of the trapped liquid water that fills the channels and pores. The freezing process led to the formation of potential gradients, with pH variations of up to 3 units.31 The boundary between liquid and solid water has a different refractive index and reveals an interface. Measurements of the zeta potential (electric potential difference between the fluid brine and the stationary liquid layer attached to the ice crystals) showed that the interfacial properties of an ice–water system are comparable to the interface with hydrophobic and nonionogenic solids, such as diamond or hydrocarbons.32 These properties could be essential for the solute exclusion from the interstitial brines in ice and the formulation of a freezeconcentration model for explaining the prebiotic chemistry observed in the ice matrix.

Observation of the behaviour of stains in ice shows that organic molecules are excluded from the ice matrix and concentrated in the interstitial brine, where chromatographic separation has been noted. Another important property of the behaviour of organic molecules in ice is that a dilute starting solution of a given solute always reaches the same molal concentration in the interstitial solution, which is determined by the final incubation temperature.33 For example, a freezing dilute urea solution tends to form an interstitial eutectic 8 m solution with a melting point of 261 K. These properties of the ice–water interface convert the ocean ices, at temperatures within the range of existence of the interface with liquid brines, into a potential reactor for the first steps responsible for the emergence of life.

Prebiotic synthesis of nucleobases and other nitrogen heterocycles in the ice matrix Nucleobases are a small group of one-ring (pyrimidines) and two-ring (purines) nitrogen heterocycles that, together with sugars and phosphate, compose nucleic acids. The pyrimidines include uracil, thymine and cytosine and purines include adenine and guanine. Other heterocycles belonging to both groups are important intermediates in the biochemistry, including xanthine, hypoxanthine and orotic acid. It is generally assumed that the earliest living forms on Earth used a genetic code based on nucleobases.34 In addition, nitrogen heterocycles could have been involved in the first metabolic pathways as cofactors.35 Regardless of the controversy regarding whether life began with a replicator, as suggested by the RNA-world hypothesis, or with metabolism, as suggested by later authors,36 there is no evidence to discard the hypothesis of a prebiotic source of nucleobases or cofactors for the first living system. The first logical hypothesis considers that the prebiotic synthesis took place on Earth, although it is not clear if the environmental conditions were consistent with efficient in situ synthesis.37 The second logical hypothesis is the delivery of nitrogen heterocycles to Earth by comets, meteorites and dust particles. This extra-terrestrial delivery could compensate for a possible lack of availability from in situ synthesis. Analysis of carbonaceous chondrites, This journal is

c

The Royal Society of Chemistry 2012

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

a class of meteorites rich in organic carbon and water,38 has demonstrated the presence of N-heterocycles. These heterocycles include adenine, guanine and triazines (ammeline and melamine), which were found in the Orgeil meteorite by Hayatsu in 1964.39 Subsequent analyses performed from 1965–197540 show that the extraction conditions and sample treatments determine the analytical results. However, the presence of nucleobases in carbonaceous chondrites is widely accepted. In 2008, Martins et al. demonstrated41 the extraterrestrial origin of xanthine and adenine in a Murchinson meteorite sample using carbon isotope measurements.

Recently, Callahan et al. demonstrated that the suite of purines found in carbonaceous chondrites is consistent with those obtained using ammonium cyanide chemistry.42 The questions that arise from these results include how were the nitrogen heterocycles synthesised on Earth or other bodies in Solar System, and how could the ice–water interface play a role in this process?

Synthesis based on hydrogen cyanide The synthesis of nucleobases and other nitrogen heterocycles in the parent body of a meteorite could be a process that is dependent on the water content and irradiation of precursors.

The seminal work of Juan Oro´ and co-workers demonstrated that adenine can be easily synthesised from hydrogen cyanide (Scheme 1).43 A prebiotic origin for the nucleobases was

thereafter regarded as a realistic possibility.44 Additionally, a Fischer–Tropsch type synthetic mechanism catalysed by mineral phases at high temperature has been suggested for the origin of N-heterocycles in meteorites,45 but its actual significance is unclear46 and currently is not a widely accepted route.

Cyanide is the primary precursor involved in our current models for prebiotic synthesis of nitrogen heterocycles and a possible precursor to the organic molecules that gave rise to biochemistry. Cyanide could be generated photochemically or by spark discharges in methane/nitrogen planetary atmospheres.47 In addition, free HCN and cyanide polymers have been observed in comets, dust particles48 and in the Titan atmosphere.49 The mechanism of synthesis of adenine from HCN implies that the first step is polymerisation to the HCN-tetramer diaminomaleonitrile (DAMN; Scheme 1). This intermediate could undergo further polymerisation to form dark brown solid polymers, which upon hydrolysis release nitrogen heterocycles, including adenine.50 This hydrolysis could take place in the ice–water interface in the parent body of comets or meteorites during their journey in the inner Solar System or after these objects impacted the Earth. Another possible mechanism is the reaction of DAMN with formamidine51 to afford a 4-amino-5-cyanoimidazole (AICN) intermediate. This reaction yields adenine through the coupling of HCN or formamidine.

The hydrolysis of AICN leads to 4-aminoimidazole-5-carboxamide (AICA), which could be a xanthine and hypoxanthine precursor52 (Scheme 1). Formamidine has also been identified

Scheme 1 Synthesis of purines by polymerisation of cyanide to the HCN tetramer and formation of cyanoimidazole derivatives. The related formation of glycine, formamidine and glycolonitrile was observed in ice–water experiments.

This journal is

c

The Royal Society of Chemistry 2012

Chem. Soc. Rev., 2012, 41, 5404–5415

5407

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

as an organic precursor found in comets53 and prebiotic chemistry laboratory simulations.54 A possible major mechanism for the formation of adenine from HCN, which was elucidated by Voet and Schwartz in 1982, is the reaction between the HCN tetramer and its cyanoimino tautomer or diiminosuccinonitrile (an oxidation product of the HCN tetramer) to yield the carbamimidoyl cyanide derivative. This molecule cyclises to 4-amino-2-cyanoimidazole-5-carbimidoylcyanide. Further addition of the cyanoimino derivative and ring closure affords adenine-8-carboxamide (Scheme 1).55 This product is quantitatively converted to adenine by hydrolysis.

The above mechanism was supported by the structural elucidation of 4-amino-2-cyanoimidazole-5-carboxamide and its hydrolysis product, 4-aminoimidazole-2,5-dicarboxamide.

However, the adenine-8-carboxamide has not yet been identified in HCN oligomerisation experiments.

The last proposed mechanism is the UV-induced photoisomerisation of the HCN tetramer to 4-amino-5-cyanoimidazole.

The reaction of this imidazole with HCN or with its hydrolysis product ammonium formate in a melt directly yields adenine.56 Because it is the key reaction in the pathway, the formation of the HCN tetramer requires a high HCN concentration to avoid the volatilisation or hydrolysis to ammonium formate, which competes with the formation of diaminomaleonitrile in dilute solutions. Therefore, it would have been impossible to reach sufficiently high HCN concentrations in the open oceans or by water evaporation.57 One solution to this problem could be to consider alternatives to aqueous HCN chemistry. The formation of nucleobases from formamide in the presence of inorganic catalysts at high temperature creates a robust pathway for adenine, hypoxanthine, uracil and cytosine among other N-heterocycles.58 One solution to this problem could be concentrating HCN using the liquid–ice interfacial properties. During the first attempt to test this possibility, Sanchez et al. (1966) showed that HCN concentrates in a frozen eutectic solution. The eutectic solution, which has a mole fraction of 70 to 80% in HCN, is formed at 249 K and deposits a dark HCN polymer.59 Considering the activation energy of the HCN polymerisation and the rate constants, the formation of the HCN tetramer in eutectic fluids should be complete in a few years. At 173 K, the reaction occurs over the order of hundreds of millions of years.60 The advantageously stable conditions in a water–ice interface could surpass the handicap of prebiotic synthesis at low temperatures and the problem of concentration and stability at high temperatures.

Additionally, the freezing of dilute glycolonitrile solutions, produced by addition of HCN and formaldehyde, produces adenine in low yield (0.004%).61 In a long duration experiment, Miyakawa et al. maintained a frozen solution of ammonium cyanide at 195 K over 27 years and at the end of this time period, identified adenine as well as other purine and pyrimidine products.62 Although the HCN pathway has been extensively studied for the synthesis of purines, it has been demonstrated that the polymerisation of cyanide could provide a pathway for the formation of the pyrimidines including uracil, 5-hydroxyuracil and orotic acid.63 The freezing of cyanide solutions could also provide a source of amino acids.

In 1972, another long-term experiment involved a solution of 5408

Chem. Soc. Rev., 2012, 41, 5404–5415

NH4CN prepared from HCN and NH3. These reagents were frozen and subjected to variable temperatures of 253 K and 195 K for 25 years. The analysis indicated the formation of glycine and small amounts of alanine and aspartic acid.64 The mechanism for the cold synthesis of amino acids from HCN has not been elucidated, but may include the hydrolysis of HCN polymers65 and the hydrolysis of 2-aminoacetonitrile, which is formed during HCN tetramer evolution, to glycine (Scheme 1).

Prebiotic laboratory synthesis from frozen cyanide solutions could be a model for the prebiotic synthesis of nucleobases.

This synthesis could also explain the chemistry observed in icecovered objects within the inner Solar System, such as asteroids and comets during their closest passage to the Sun, and in objects with complex chemistry, including Titan or Enceladus. To efficiently serve both goals, more experimental work should be performed to elucidate the mechanisms involved in frozen HCN solution, to test if the classic pathway through cyanoimidazole derivatives is reproducible in the ice matrix scenario and to determine if alternative pathways should also be examined.

Synthesis based on cyanoacetylene/acetylene and the role of urea Cyanoacetylene is the other primary precursor considered for the synthesis of nucleobases. Cyanoacetylene can be obtained in the laboratory from methane–nitrogen mixtures by spark discharges66 by irradiation with short-wave ultraviolet radiation at 185 and 254 nm;67 the spectrum of this molecule has been observed in the interstellar medium68 and by the Voyager mission in Titan’s atmosphere,69 where crystalline condensates of cyanoacetylene with acetylene may exist.70 The potential prebiotic relevance of cyanoacetylene in origin of life studies was pointed out by Ferris, Sanchez and Orgel in 1968. They observed that the reaction of cyanoacetylene with aqueous 1 M sodium cyanate or 1 M urea gave cytosine in up to 5% yield (Scheme 2).71 The prebiotic availability of cyanate could be explained by the hydrolysis of cyanogen and urea, which may also be present in cometary and interstellar ices.72 The mechanism of this reaction could be explained by cyanoacetaldehyde, generated by hydrolysis of cyanoacetylene.

The Miller research demonstrated the eutectic concentration and reaction of cyanoacetaldehyde with urea in an ice matrix at 253 K to give cytosine and uracil in 0.005% and 0.02% yields, respectively.73 In the same report, cyanoacetaldehyde reacted with guanidine at 253 K to give cytosine in 0.05% yield and uracil in 10.8% yield, as well as lesser amounts of isocytosine and 2,4-diaminopyrimidine after 2 months.74 This reaction may proceed through the cyanoacetaldehyde dimer, 4-(hydroxymethylene) pentenedinitrile, easily formed by concentrating the cyanoacetaldehyde solutions (Scheme 2).72 The basis of these experiments is the freezing of a urea or guanidine solution. This process provides a concentration mechanism because the crystalline ice excludes the solute and a eutectic solution is formed. At 262 K, urea forms an 8 m eutectic solution in water. This effect could be significant from a prebiotic point of view, despite the slower reaction rates, as has been shown in recent experiments.

This journal is

c

The Royal Society of Chemistry 2012

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Scheme 2 Cyanoacetylene as a precursor for pyrimidines. The reaction of cyanoacetylene with urea or ammonium cyanate yields cytosine, whose deamination leads to uracil. The reaction with guanidine directly forms 2,4-diaminopyrimidine and goes through a pentanedinitrile intermediate.

An unresolved issue with the cyanoacetylene pathway in the synthesis of nucleobases is its reactivity to nucleophiles,75 which suggests a high number of competitive reactions that lead to the formation of amino- or hydroxyacrylonitriles and subsequent polymers or hydrolysis products; on the other hand, the prebiotic origin of cytosine was questioned, at least in the liquid water medium, because its spontaneous and rapid deamination to uracil.76 In part, the reactions in the water–ice interface could overcome the problem of dilution and degradation associated with solutions in liquid water pools.

Although much time has elapsed since the first proposal in 1966 of a low temperature prebiotic environment for the origin of nucleobases, it was not until 2000 that the product of the classic approach of spark discharges in a methane/nitrogen based atmosphere was subjected to eutectic freezing77 at 253 K for 5 years. The frozen spark discharge product showed a more extensive mixture of amino acids and the presence of adenine, which was absent in the control experiment at room temperature.

The first experimental simulation of prebiotic synthesis in ice–liquid water directly from a nitrogen/methane atmosphere by spark discharges was performed in 2009.78 The sparking on a freezing dilute urea solution under a nitrogen/methane atmosphere leads to the formation of cytosine, uracil and 2,4,6-trihydroxypyrimidine (barbituric acid) as the main identified pyrimidines, in addition to adenine. The experiments showed that using the freeze–thaw conditions, the observed sequence of pyrimidine yield obtained was cytosine > uracil > 2,4-diaminopyrimidine > 2,4,6-trihydroxypyrimidine.

This journal is

c

The Royal Society of Chemistry 2012

The formation of pyrimidines by oxidative alteration of cytosine (UV irradiation, hydroxyl radical addition or other free radical mechanisms and further oxidation to barbituric acid) could explain the results observed.79 The formation of cytosine as the main pyrimidine suggests that the low temperature conditions could reduce the rate of deamination to uracil and favour subsequent chemical evolution steps, as suggested by Bada.12 The triazine series (cyanuric acid, ammelide, ammeline and melamine) are also obtained in high yields (Scheme 3).

The formation of triazines appears to be dependent on the freezing of urea solution. The triazines are not biological compounds, but they could mimic nucleobases behaviour in nucleic acids and their potential prebiotic role has been discussed.80 Their presence in meteorites remains contentious.81 The key factor appears to be the freezing process itself and not the temperature of the final ice obtained, as the temperature was selected to be right below the freezing point of 0.1 M urea.

In a liquid urea solution at room temperature, there is no evidence of nucleobases. Instead, the formation of hydantoins, nitriles and tholins (reddish-brown, insoluble, heteropolymeric or macromolecular materials formed by sparking or irradiation of simple carbon sources, such as methane) is prevalent.

The behaviour of urea in the ice–water interface is the key factor because urea tends to form dimers or oligomers in a concentration-dependent manner.82 Urea molecules in aqueous fluids tend to form hydrogen bonds with neighbouring water molecules at both the amino and the carbonyl groups.83 Spectroscopic studies show that at urea concentrations higher than 1 M, the urea–urea molecular interactions are significant.

The urea–urea molecular interaction with subsequent formation of dimers or clusters of urea molecules becomes dominant at eutectic concentration.84 During freezing, the urea is segregated from pure ice to accumulate in supercooled microfluid inclusions of a supersaturated solution. This system is governed by dehydration and association of solute molecules.85 Thus, the extent of urea dimerisation (18% in 0.1 M urea solution at standard temperature86) is expected to increase and to become quantitatively a few degrees below the onset of freezing.

Consequently, we expect an apparently paradoxical similarity between the process observed in molten urea84 and urea clusters entrapped in an ice matrix when the latter are subjected to direct sparking or irradiation.

This behaviour could explain the sequence of products obtained (cyanuric acid > ammelide > ammeline > melamine), which is the same sequence observed when urea is heated above its melting point. The spark discharges into the ice, then, could thermally decompose urea clusters into ammonium cyanate. Further decomposition of ammonium cyanate leads to cyanic acid. The cyanic acid reacts with urea to form the biuret and with the formed biuret to form cyanuric acid (a cyanic acid trimer), which is the main triazine observed.84 Several routes to ammelide are possible: reaction of cyanuric acid and ammonia or cyanic acid and urea or biuret. The process, in which the decomposition products accelerate the formation of triazines, could explain the high concentration of cyanuric acid obtained in these conditions. Another parallel pathway is the formation of melamine by cyanamide polymerization.

Chem. Soc. Rev., 2012, 41, 5404–5415

5409

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Scheme 3 Urea as precursor of nitrogen heterocycles. Possible pathways to pyrimidines, hydantoins and triazines in frozen urea solution under a methane/nitrogen atmosphere.

The melamine hydrolysis yields cyanuric acid (Scheme 3).

These pathways and the same reaction sequence, with the same relative abundance of triazines, have been studied in molten urea87 at temperatures between 406 and 460 K. In this case, an alternative route for forming purines could result from the condensation of amino acids and biuret, a reaction that occurs at high temperature;88 however, this alternative still has not been studied in ice–water systems and could be an unlikely possibility because of the high activation energy of such condensations. We also cannot discard other alternative pathways parallel to the polymerisation of concentrated urea solutions. For example, the production of cyanic acid during atmospheric discharges or thermal alteration of tholins89 and the subsequent reaction in freezing urea solutions could be an alternative source of cyanuric acid. Additional laboratory studies are necessary for clarifying the mechanisms involved in the cold synthesis of triazines and purines in the ice matrix.

The effect of concentration of solutes on the ice matrix, together with the low availability of water vapour, could explain the preferential synthesis of polycyclic aromatic hydrocarbons (PAHs) by sparking a methane/nitrogen atmosphere over an ice matrix.90 The model of PAH synthesis is interesting because it could confirm the theoretical synthesis of aromatics by acetylene insertion mechanisms proposed for Titan’s atmosphere.91 In laboratory experiments at sub-zero temperatures,65 the acetylene addition mechanism could explain the preferential formation of aromatics and poly(triacetylene) 5410

Chem. Soc. Rev., 2012, 41, 5404–5415

polymers (Scheme 4) by two possible mechanisms. First, a single aromatic ring could be generated from acetylene and vinyl radical and PAH growth by H abstraction and acetylene addition (Berthelot synthesis, similar to PAHs formation in flames). The second mechanism involves polyyne growth.

The presence of water ice induces oxidations leading to the formation of aromatic polar species such as benzaldehyde or acetophenone. The reaction in ice, in contrast to the dry high temperature synthesis of PAHs, leads to hydroxyl-rich poly(triacetylene) based polymers. Overall, these ice–water laboratory experiments reveal the expected chemical species in surface or subsurface ices on solar system objects or extrasolar planetary bodies.

The activation of methane/nitrogen atmospheres by spark discharges could lead to various chemistries involving reactive intermediates, including HCN, cyanoacetylene and acetylene.

The preference for the hydantoins in liquid urea solutions at room temperature versus pyrimidines in frozen solution experiments could be due to the acetylene formation and subsequent alteration by means of ozone and hydroxyl radicals at higher temperatures to form a-dicarbonyl compounds such as glyoxal.92 The reaction of glyoxal with urea under mild acidic conditions yields hydantoin,93 whose further oxidation yields 5-hydroxyhydantoin and parabanic acid (Scheme 3). These three hydantoins are always found together in all the experiments reported in the literature. Its formation could be explained also as alteration products of uracil by hydroxyl and other free radicals This journal is

c

The Royal Society of Chemistry 2012

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Scheme 4 matrix.

Polycyclic aromatic hydrocarbons and acetylene polymers detected from sparking a methane/nitrogen atmosphere on the water–ice

generated in water solutions by photolysis or irradiation.79,94 At lower temperatures, the degradation of pyrimidines to hydantoins and the oxidation of acetylene could be diminished, due to the lower availability of reactive oxygen species generated from the excitation of water. As a consequence, hydantoins could be the final products of alteration of pyrimidines under prebiotic conditions subjected to UV-irradiation or other energetic processes. Regarding acetylene, the polymerisation could be the preferred reaction pathway, as shown by the formation of poly-triacetylene and aromatic hydrocarbons at the ice–water matrix previously described. In this environment, the HCN or cyanoacetylene pathways could dominate other alternative mechanisms as the synthesis of uracil by reaction of urea with acetylene dicarboxylic acid.95 This acid is the aqueous hydrolysis product of dicyanoacetylene, which is an exotic product of methane/nitrogen atmospheres observed in the Titan atmosphere.96 The role of acetylene derivatives has not been studied in the ice–water scenario, and further experiments are necessary to explore the possible alternative pathways related to acetylene in prebiotic synthesis in an ice matrix and to put it in context with the classic mechanisms involving cyanide and cyanoacetylene. The products identified in the simulations of methane/nitrogen atmospheres over the ice–water interface include dicarboxylic and hydroxycarboxylic acids, amino acids and pyrazines, suggesting an additional mechanism to those suggested above.

In summary, the advantages of an ice–water interface in prebiotic synthesis include the reduction in the formation of polymers and tholins with a preference for ring systems (nitrogen heterocycles or aromatic rings) by the effect of concentration of diluted reactants such as HCN, urea or cyanate. Combined with other rocks or minerals, the freezing of liquid water solutions could favour mineral surface–organic solute interactions.97 This journal is

c

The Royal Society of Chemistry 2012

The ice–water system in the origin of nucleic acids The ice matrix is an appropriate environment for the synthesis of nitrogen heterocycles, as demonstrated by the synthesis of triazines and nucleobases in freezing urea solutions. Could the ice–water interface be a favourable environment for the assembly of the first biologically relevant informational polymers?

The success in the synthesis of nucleobases from a feedstock of active nitrogen species available prebiotically led to the establishment of a similar retrosynthetic analysis for RNA and to the search for prebiotically plausible syntheses of a primordial informational, self-replicating polymer. If the discovery of an abiotic pathway to the origin of the first nucleotides and the constitutional self-assembly of RNA is achieved, the RNAworld hypothesis (a term coined by Walter Gilbert in 1986),98 which proposes a molecular evolutionary step involving autocatalytic RNA molecules prior to the origin of protein synthesis and metabolic machinery, will be strengthened.

The current state of prebiotic chemistry does not provide a complete model for an abiotic origin of RNA, and the first formulations of an RNA world have been re-evaluated.99 However, some argue that it may be premature to conclude that the prebiotic RNA world is unlikely to be a step in the emergence of life.100 In this context, the ice–water interface has been evaluated thoroughly as a matrix for the polymerisation of highly activated nucleotides. The first demonstration of this possibility was performed by Gryaznov and Letsinger in 1993.101 In their experiment, the coupling of an a-bromoacyl-activated oligonucleotide (bromoacetylamino-3 0 -desoxythimidine in the 3 0 -terminus) with another oligonucleotide with a phosphorothioate group in the 5 0 -terminus proceeded without a template in a frozen saline solution at 255 K in 5 days. The reaction was explained as a result of the high local concentration of reactants in the fluid cavities in the ice matrix.

Chem. Soc. Rev., 2012, 41, 5404–5415

5411

Scheme 6

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Scheme 5

The enhancing effect of the ice matrix on the formation of RNA oligomers was demonstrated by Kanavarioti et al. in a very remarkable experiment in which oligouridylates up to 22 bases long were synthesised by incubating a uridine 5 0 -monophosphorimidazolide solution at 255 K at a pH range between 6 and 8 in the presence of magnesium and lead cations (Scheme 5).102 The study of the ribonuclease A digestion products showed that the oligomers obtained are mainly linear and that 30% carry at least one 3 0 –5 0 linkage. The fluorescence microscopy observation of an ice layer under the experimental conditions with acridine orange staining indicated that the organic solutes were concentrated in the eutectic lattice structure included in the ice matrix. The authors concluded that the formation of eutectic solutions of reactants in the ice matrix facilitated the oligomerisation. The polymerisation most likely occurs in the liquid concentrated solutions between the ice crystals, and not by the adsorption of reactants onto the ice surfaces, as previously suggested by Stribling and Miller,103 who studied the template directed synthesis of poly(U) in diluted solutions concentrated by freezing close to the NaCl eutectic. The ice also has an effect on the metal catalysis. The reaction in the ice–water medium requires Pb2+ as a catalyst and not Mg2+.

This phenomenon is different from reactions in solution, which require both magnesium and lead cations. A possible interpretation of this observation is that the molecular associations in an ice matrix tend to be more stable than the corresponding ones in solution. An open question that arises is the role of certain metal cations (for example lead) as prebiotic catalysts.

The lead catalysis in the polymerisation of activated nucleotides could be related to the mechanism of leadzymes104 and suggests that metal ion catalysis is central in a hypothetical RNA world.

If pyrimidine and purine-activated nucleotides are used in the water–ice interface at 255 K during 38 days in the presence of Mg2+ and Pb2+, a mixed-sequence polynucleotide with approximately the same proportion of purine and pyrimidine residues is obtained.105 Monnard and Szostak106 studied the template-directed RNA polymerisation in water–ice at 256.4 K, a temperature that permits the maintenance of a stable water–ice interface for long periods of time. They found that lead and magnesium ions catalyse the elongation of a RNA hairpin with a 5 0 -overhang as a template.

Similarly, the non-enzymatic synthesis of polyadenosine in a sea–ice matrix, directed by poly(U), was performed, using adenosine-50 -monophosphate (2-methyl) imidazolide as monomer.

Temperature fluctuations established the freeze–partial thaw cycles during one year. The results show high molecular weight poly(A) formation, with chain lengths of as many as 420 residues (Scheme 6).107 The freezing-concentration model could also govern the conformational rearrangement pathway of the formed polymers.

5412

Chem. Soc. Rev., 2012, 41, 5404–5415

Freezing a 21-nt RNA hairpin solution at 203 K followed by incubation at 263 K results in the conversion to the duplex dimer form.108 The formation of frozen microenvironments during prebiotic evolution could be a key factor in the possible prebiotic evolution of informational polymers.

The formation of peptides in the ice matrix The linking of monomer units to form simple polymers likely defined an important step in the origins of life, and many conditions have been proposed, including dehydration agents,109 sulphide minerals,110 melting111 or hydrothermal systems.112 Further studies suggest important roles of catalytic surfaces, such as clays, or interfaces created by wet–dry cycling of monomers on mineral surfaces.113 Based on this idea, Schwendinger and Rode found a particularly simple process of salt-induced peptide formation, using 40–50 mM amino acid solutions where NaCl at concentrations above 3 M can act as a dehydrating or condensation agent, using dissolved Cu(II) as a catalyst.114 Experiments carried out

by Fox demonstrated that the melting of amino acids at temperatures in the range of 400 to 433 K, to allow melting without decomposition, produces a type of polymer called ‘proteinoids’. This phenomenon will occur provided that acidic or basic amino acids are present in excess.115 However, the so-called ‘proteinoids’ are mainly heteropolymers containing only very small quantities of peptide bonds.116 The melting of a mixture of urea and alanine yields the dipeptide Ala-Ala.81 The largest number of proposals and related experiments performed in order to model the prebiotic peptide formation in solution involves the postulated existence of coadjutant condensation reagents in a homogenous catalytic process.

These reagents include cyanamide and cyanoguanidine, which may act as prebiotically plausible condensing agents.117 A problem associated with high temperature processes is the decomposition of amino acids and the hydrolysis of peptides, which constitutes a limitation for the organisation of larger polymers.118 The synthesis in freezing solutions could prevent undesirable side reactions, hydrolysis of the formed peptide bond, and the decomposition of amino acids as well as reduce the rate of amino acid racemisation.119 This idea is connected to a different approach to the problem of amino acid condensation that was introduced years ago: the salt-induced peptide formation reaction. Salty brines could have played a role in the polymerisation of amino acids. However, the formation of a peptide bond is not straightforward at low temperatures without condensing agents, and the experiments performed were carried out at high temperatures under drying conditions.

Could freezing of the primitive oceans have produced the concentrated salty brines with the associated condensing agents needed to promote the salt-induced polymerisation process?

This journal is

c

The Royal Society of Chemistry 2012

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

Liu and Orgel studied the oligomerisation of b-amino acids in aqueous solutions under eutectic conditions using activation by the water-soluble reagents EDAC (1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide) and carbonyl-diimidazole.120 The oligomerisation of b-amino acids (L-aspartic acid, b-amino adipic acid, b-glutamic acid) using these condensing agents proceeds efficiently at 253 K (under eutectic freezing), even from dilute solutions of the substrates. This reaction produces peptides in the range 15 to 20 units (maximum: 45) in length with a yield of over 50%. The efficiency of polymerisation and the length distribution of the oligomers were almost unaffected by the solute concentration over a broad range of 0.1 to 100 mM at 253 K. According to these results, the EDAC reagent constitutes the model of a group of activating agents whose function is the direct reaction with the carboxyl group of amino acids. Cyanogen, cyanamide and cyanoguanidine are prebiotically plausible members of this group. The elucidation of the pathway shows that the first step is the direct attack of the carboxyl group on the carbodiimide to form an O-acylisourea.

The free amino group of another amino acid attacks this activated species to form a peptide bond. In the case of a-amino acids, the carboxyl group of the dipeptide can be activated and then cyclise efficiently to give a diketopiperazine, thus inhibiting oligomerisation.121 Cyclisation of an activated dimer of b-amino acids is not straightforward because an eight-membered ring does not form readily.

In 1996, Vajda et al. synthesised four protected dipeptides and a protected tripeptide in frozen dioxane and other organic solvents.122 The data demonstrated that the coupling rates in frozen dioxane at 254 K exceed by approximately one order of magnitude of the rates in liquid solution at 313 K. Vajda suggested that enhanced reaction rates and/or yields, diminution of racemisation, and the suppression of side reactions can be expected in frozen systems, and these possibilities substantially increase the importance of peptide formation in eutectic frozen solutions.123 However, no further investigation on these possibilities has been performed.

Concluding remarks Prebiotic chemistry in the range of stability of a liquid water–ice interface (277 to 243 K under common laboratory conditions) has been proposed since the pioneering experiment in the field. These ideas were proposed to overcome the concentration and stability problems associated with liquid water prebiotic chemistry. The experiments performed demonstrated that the synthesis of aromatic hydrocarbons, purines and pyrimidines and other nitrogen heterocycles of potential prebiotic interest (such as triazines) is favoured in the ice matrix by classic cyanide and cyanoacetylene pathways following a freezing-concentration model. Despite these results, the experimental prebiotic chemistry in the solute-concentrated solutions that fill the space confined by the ice matrix has received relatively little attention in the elaboration of the models for the origin of organics in Solar System bodies and prebiotic evolution. Consequently, it is necessary to clarify the mechanisms involved and the role of reactants as well as to perform more experiments under plausible prebiotic conditions, especially if geochemical models support stable icy environments on the prebiotic Earth.

This journal is

c

The Royal Society of Chemistry 2012

The concentration of reactant solutions by freezing also enhances the polymerisation of activated nucleotides and the formation of small peptides in the presence of an activating agent. The prebiotic relevance of these polymerisation reactions and the gap between the nucleobase synthesis and the organisation of the first biopolymers is a matter for discussion.

Nevertheless, the ice world constitutes an interesting prebiotic chemistry scenario that awaits further investigation.

Acknowledgements We acknowledge the Centro de Astrobiologia (CSIC-INTA) and the grants of the project AYA2009-13920-C02-01 from the Ministerio de Ciencia e Innovacio´n (MICINN, Spain).

Notes and references 1 S. Miller, Science, 1953, 117, 528.

2 J. P. Ferris, R. A. Sanchez and L. E. Orgel, J. Mol. Biol., 1968, 33, 693; D. Clarke and J. Ferris, Icarus, 1997, 127, 158.

3 W. Martin and M. J. Russell, Philos. Trans. R. Soc., B, 2006, 362, 1887.

4 G. M. Mun˜oz Caro, U. Meierhenrich, W. A. Schutte, W. H. P. Thiemann and J. M. Greenberg, Astron. Astrophys., 2004, 413, 209.

5 J. I. Lunine, Meteorites and the early solar system, 2006, vol. II, p. 309.

6 P. Klan and I. Holoubek, Chemosphere, 2002, 46, 1201 and references therein.

7 C. F. Chyba and C. B. Phillips, Origins Life Evol. Biospheres, 2002, 32, 47.

8 G. Tobie, O. Grasset, J. I. Lunine, A. Mocquet and C. Sotin, Icarus, 2005, 175, 496.

9 F. Postberg, S. Kempf, J. Schmidt, N. Brilliantov, A. Beinsen, B. Abel, U. Buck and R. Srama, Nature, 2009, 459, 1098.

10 V. Moulton, P. P. Gardner, R. F. Pointon, L. K. Creamer, G. B. Jameson and D. Penny, J. Mol. Evol., 2000, 51, 416.

11 S. L. Miller and L. Orgel, The Origins of Life on the Earth, Prentice Hall, New Jersey, 1974.

12 J. L. Bada, Earth Planet. Sci. Lett., 1994, 226, 1.

13 M. Blackman and N. D. Lisgarten, Proc. R. Soc., Ser. A, 1957, 239, 93.

14 A. Kouchi and T. Kuroda, Nature, 1990, 344, 134.

15 H. E. Stanley, MRS Bull., 1999, 24, 22.

16 D. M. Anderson and A. Banin, Origins Life Evol. Biospheres, 1975, 6, 23; D. M. Anderson, Life in the Universe, ed. J. Billigham, MIT Press, Cambridge, Massachusetts, 1981.

17 W. Zheng, D. Jewitt and R. I. Kaiser, J. Phys. Chem. A, 2009, 113, 11174.

18 M. G. Kivelson, K. K. Khurana, C. T. Russell, M. Volwerk, R. J. Walker and C. Zimmer, Science, 2000, 289, 1340.

19 B. E. Schmidt, D. D. Blankenship, G. W. Patterson and P. M. Schenk, Nature, 2011, 479, 502.

20 T. B. McCord, G. B. Hansen, F. P. Fanale, R. W. Carlson, D. L. Matson, T. V. Johnson, W. D. Smythe, J. K. Crowley, P. D. Martin, A. Ocampo, C. A. Hibbitts, J. C. Granahan and the NIMS Team, Science, 1998, 280, 1242.

21 G. M. Marion, C. H. Fritsen, H. Eicken and M. C. Payne, Astrobiology, 2003, 3, 785.

22 G. Mitri, A. P. Showman, J. I. Lunine and R. M. C. Lopes, Icarus, 2008, 196, 216.

23 R. Shapiro and D. Schulze-Makuch, Astrobiology, 2009, 9, 1.

24 S. Pilling, D. P. Andrade, A. C. Neto and R. Rittner, J. Phys.

Chem., 2009, 113, 11161.

25 C. P. McKay and H. Smith, Icarus, 2005, 178, 214; R. S. Oremland and M. A. Voytek, Astrobiology, 2008, 8, 45.

26 D. Schulze-Makuch and L. N. Irwin, EOS Trans., Am. Geophys.

Union, 2001, 82, 150; C. F. Chyba and C. B. Phillips, Proc. Natl.

Acad. Sci. U. S. A., 2001, 98, 801.

27 K. Zahnle, L. Schaeffer and B. Fegley, Cold Spring Harbor Perspect. Biol., 2010, 2, a004895; N. H. Sleep, K. J. Zahnle and

Chem. Soc. Rev., 2012, 41, 5404–5415

5413

28 29 30 31

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

51 52

53 54 55 56 57 58

59 60 61 62 63

P. S. Neuhoff, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 3666; E. G. Nisbet and N. H. Sleep, Nature, 2001, 409, 1083.

C. F. Chyba, P. J. Thomas, L. Brookshaw and C. Sagan, Science, 1990, 249, 366.

H. Trinks, W. Shro¨der and C. K. Biebricher, Origins Life Evol.

Biospheres, 2005, 35, 429.

H. Eicken, J. Kolatschek, J. Freitag, F. Lindemann, H. Kassens and I. Dmitrenko, Geophys. Res. Lett., 2002, 27, 1919.

V. K. Bronshteyn and A. A. Chernov, J. Cryst. Growth, 1991, 112, 129.

J. Drzymala, Z. Sadowski, L. Holysz and E. Chibowski, J. Colloid Interface Sci., 1999, 220, 229.

P. A. Monnard and H. Ziock, Chem. Biodiversity, 2008, 5, 1521.

J. P. Dworkin, A. Lazcano and S. Miller, J. Theor. Biol., 2003, 222, 127.

B. E. H. Maden, Trends Biochem. Sci., 1995, 20, 337; A. Eschenmoser, Angew. Chem., 1988, 27, 5.

S. A. Benner, A. D. Ellington and A. Tauer, Proc. Natl. Acad. Sci. U. S. A., 1989, 86, 7054.

J. F. Kasting and L. Brown, Origins Life Evol. Biospheres, 1996, 26, 219; R. Stribling and S. Miller, Origins Life Evol. Biospheres, 1986, 17, 261.

O. R. Norton, The Cambridge Encyclopedia of Meteorites, Cambridge University Press, Cambridge, 2002.

R. Hayatsu, Science, 1964, 146, 1291.

R. Hayatsu, M. H. Studier, L. P. Moore and E. Anders, Geochim.

Cosmochim. Acta, 1975, 39, 471.

Z. Martins, O. Botta, M. L. Fogel, M. A. Sephton, D. P. Glavin, J. S. Watson, J. P. Dworkin, A. W. Schwartz and P. Ehrenfreund, Earth Planet. Sci. Lett., 2008, 270, 130.

M. P. Callahan, K. E. Smith, H. J. Cleaves, J. Ruzicka, J. C. Stern, D. P. Glavin, C. H. House and J. P. Dworkin, Proc.

Natl. Acad. Sci. U. S. A., 2011, 108, 13995.

J. Oro´ and A. P. Kimball, Arch. Biochem. Biophys., 1961, 94, 217.

L. E. Orgel, Origins Life Evol. Biospheres, 2001, 34, 361.

R. Hayatsu and E. Anders, Top. Curr. Chem., 1981, 99, 1–37.

A. W. Schwartz and G. J. F. Chittenden, BioSystems, 1977, 9, 87–92.

R. A. Sanchez, J. Ferris and L. E. Orgel, J. Mol. Biol., 1967, 30, 223.

C. N. Matthews and R. D. Minard, Faraday Discuss., 2006, 133, 393.

D. E. Shemansky, A. I. F. Stewart, R. West, L. W. Esposito, J. T. Harlett and X. Liu, Science, 2005, 308, 978.

S. Miyakawa, H. J. Cleaves and S. Miller, Origins Life Evol.

Biospheres, 2002, 32, 209; M. Ruiz-Bermejo, J. L. de la Fuente, C. Rogero, C. Menor-Salvan, S. Osuna-Esteban and J. A. Martin-Gago, Chem. Biodiversity, 2012, 9, 25.

R. A. Sanchez, J. P. Ferris and L. E. Orgel, J. Mol. Biol., 1967, 30, 223.

R. Saladino, C. Crestini, F. Ciciriello, G. Constanzo and E. Di Mauro, Chem. Biodiversity, 4, 694; S. Yuasa, D. Flory, B. Basile and J. Oro´, J. Mol. Evol., 1984, 21, 76; J. Oro´ and P. Kimball, Arch. Biochem. Biophys., 96, 293.

G. F. Joyce, Nature, 1989, 338, 217.

A. W. Schwartz, A. B. Voet and M. Veen, Origins Life Evol.

Biospheres, 1984, 14, 91.

A. B. Voet and A. W. Schwartz, Bioorg. Chem., 1983, 12, 8.

G. Zubay and T. Muy, Origins Life Evol. Biospheres, 2001, 31, 87; R. A. Sanchez, J. P. Ferris and L. E. Orgel, J. Mol. Biol., 1968, 38, 121.

L. E. Orgel, Crit. Rev. Biochem. Mol. Biol., 2004, 39, 99.

R. Saladino, C. Crestini, V. Neri, F. Ciciriello, G. Constanzo and E. Di Mauro, ChemBioChem, 2006, 7, 1707; H. L. Barks, R. Buckley, G. A. Grieves, E. Di Mauro, N. V. Hud and T. M. Orlando, ChemBioChem, 2010, 11, 1240.

R. A. Sanchez, J. Ferris and L. E. Orgel, Science, 1966, 153, 72.

F. Raulin, P. Brunston, P. Paillous and R. Sternberg, Adv. Space Res., 1995, 15, 321.

A. W. Schwartz, H. Joosten and A. B. Voet, BioSystems, 1982, 15, 191.

S. Miyakawa, H. J. Cleaves and S. L. Miller, Origins Life Evol.

Biospheres, 2002, 32, 209.

A. B. Voet and A. W. Schwartz, Origins Life Evol. Biospheres, 1981, 12, 45.

5414

Chem. Soc. Rev., 2012, 41, 5404–5415

64 M. Levy, S. L. Miller, K. Brinton and J. L. Bada, Icarus, 2000, 145, 609.

65 J. P. Ferris, P. C. Joshi, E. H. Edelson and J. G. Lawless, J. Mol. Evol., 1978, 11, 293; J. Oro´ and S. Kamat, Nature, 1961, 190, 442.

66 R. A. Sanchez, J. P. Ferris and L. E. Orgel, Science, 1966, 154, 784.

67 B. N. Tran, J. C. Joseph, M. Force, R. C. Briggs, V. Vuitton and J. P. Ferris, Icarus, 2005, 177, 106.

68 B. E. Turner, Astrophys. J., 1971, 163, L35; W. J. Lafferty and F. J. Lovas, J. Phys. Chem. Ref. Data, 1978, 7, 441; M. Morris, B. E. Turner, P. Palmer and B. Zuckerman, Astrophys. J., 1976, 205, 82; Y. Osamura, K. Fukuzawa, R. Terzieva and E. Herbst, Astrophys. J., 1999, 519, 697.

69 P. S. Monks, P. N. Romani, F. L. Nesbitt, M. Scanlon and L. J. Stief, J. Geophys. Res., 1993, 171, 15; D. W. Clarke and J. P. Ferris, Icarus, 1995, 115, 119.

70 R. K. Khanna, Icarus, 2005, 178, 165.

71 J. P. Ferris, R. A. Sanchez and L. E. Orgel, J. Mol. Biol., 1968, 33, 693.

72 M. Nuevo, J. H. Bredeho¨ft, U. J. Meierhenrich, L. d’Hendecourt and W. H.-P. Thiemann, Astrobiology, 2010, 10, 245.

73 K. E. Nelson, M. P. Robertson, M. Levy and S. L. Miller, Origins Life Evol. Biospheres, 2001, 31, 221.

74 H. J. Cleaves, K. E. Nelson and S. L. Miller, Naturwissenschaften, 2006, 93, 228.

75 A. Benidar, J. C. Guillemin, O. Mo and M. Yan˜ez, J. Phys. Chem.

A, 2995, 109, 4705; H. Mo¨llendal, B. Khater and J. C. Guillemin, J. Phys. Chem. A, 2007, 111, 1259.

76 R. Shapiro, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 4396.

77 M. Levy, S. L. Miller, K. Brinton and J. L. Bada, Icarus, 2000, 145, 609.

78 C. Menor-Salvan, M. Ruiz-Bermejo, M. I. Guzman, S. OsunaEsteban and S. Veintemillas, Chem.–Eur. J., 2009, 15, 4411.

79 J. Hong, D. G. Kim, C. Cheong and K. J. Paeng, Microchem. J., 2001, 68, 173.

80 G. K. Mittapalli, K. R. Reddy, H. Xiong, O. Mun˜oz, B. Han, F. De Riccardis, R. Krishnamurthy and A. Eschenmoser, Angew.

Chem., 2007, 119, 2522; M. Hysell, J. S. Siegel and Y. Tor, Org. Biomol. Chem., 2005, 3, 2946.

81 Z. Martins, O. Botta, M. L. Fogel, M. A. Sephton, D. P. Glavin, J. S. Watson, J. P. Dworkin, A. W. Schwartz and P. Ehrenfreund, Earth Planet. Sci. Lett., 2008, 270, 130.

82 Y. Hayashi, Y. Katsumoto, S. Omori, N. Kishii and A. Yasuda, J. Phys. Chem. A, 2007, 111, 1076.

83 A. K. Soper, E. W. Castner and A. Luzar, Biophys. Chem., 2003, 105, 649.

84 Y. Kameda, M. Sasaki, S. Hino, Y. Amo and T. Usuki, Bull.

Chem. Soc. Jpn., 2006, 79, 1367; G. Grdadolnik and Y. Marechal, J. Mol. Struct., 2002, 615, 177.

85 M. I. Guzma´n, L. Hildebrandt, A. J. Colussi and M. R. Hoffmann, J. Am. Chem. Soc., 2006, 110, 931.

86 R. H. Stokes, J. Phys. Chem., 1965, 69, 4012.

87 P. M. Schaber, J. Colson, S. Higgins, D. Thielen, B. Anspach and J. Brauer, Thermochim. Acta, 2004, 424, 131.

88 I. M. Lagoja and P. Herdewijn, Chem. Biodiversity, 2007, 4, 818.

89 J. L. de la Fuente, M. Ruiz-Bermejo, C. Menor-Salvan and S. Osuna-Esteban, J. Therm. Anal. Calorim., 2012, DOI: 10.1007/s10973-011-2141-1.

90 C. Menor-Salvan, M. Ruiz-Bermejo, S. Osuna-Esteban, G. Mun˜oz-Caro and S. Veintemillas, Chem. Biodiversity, 2008, 5, 2729.

91 E. H. Wilson, S. K. Atreya and A. Coustenis, J. Geophys. Res., 2003, 108, 1.

92 D. Cremer, R. Crehuet and J. Anglada, J. Am. Chem. Soc., 2001, 123, 6127.

93 E. Ware, Chem. Rev., 1950, 46, 403.

94 G. A. Infante, P. Jirathana, E. J. Fendler and J. H. Fendler, J. Chem. Soc., Faraday Trans. 1, 1974, 70, 1162.

95 A. S. Subbaraman, Z. A. Kazi, A. S. U. Choughuley and M. S. Chadha, Origins Life Evol. Biospheres, 1980, 10, 343.

96 Z. Guennoun, N. Pietri, I. Couturier-Tamburelli and J. P. Aycard, Chem. Phys., 2004, 300, 23.

97 M. D. Brasier, R. Matthewman, S. McMahon and D. Wacey, Astrobiology, 2010, 11, 725; N. Lahav and S. Chang, J. Mol. Evol., 1976, 19, 36.

This journal is

c

The Royal Society of Chemistry 2012

Downloaded by Centro de Astrobiología on 14 September 2012 Published on 01 June 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35060B

98 W. Gilbert, Nature, 1986, 319, G; F. Joyce, Nature, 2002, 418, 214.

99 S. A. Benner, A. D. Ellington and A. Tauer, Proc. Natl. Acad. Sci. U. S. A., 1989, 86, 7054.

100 C. Anastasi, F. F. Buchet, M. Crowe, A. L. Parkes, M. W. Powner, J. M. Smith and J. D. Sutherland, Chem.

Biodiversity, 2007, 4, 721.

101 S. M. Gryaznov and R. L. Letsinger, J. Am. Chem. Soc., 1993, 115, 3808.

102 A. Kanavarioti, P. A. Monnard and D. W. Deamer, Astrobiology, 2001, 1, 271.

103 R. Stribling and S. L. Miller, J. Mol. Evol., 1991, 32, 282.

104 W. G. Scott, Curr. Opin. Chem. Biol., 1999, 3, 705.

105 P. A. Monnard, A. Kanavarioti and D. W. Deamer, J. Am. Chem.

Soc., 2003, 125, 13734.

106 P. A. Monnard and J. C. Szostak, J. Inorg. Biochem., 2008, 102, 1104.

107 H. Trinks, W. Shro¨der and C. K. Biebricher, Origins Life Evol.

Biospheres, 2005, 35, 429.

108 X. Sun, J. M. Li and R. M. Wartell, RNA, 2007, 13, 2277.

109 J. Hulshof and C. Ponnamperuma, Origins Life Evol. Biospheres, 1976, 7, 197.

This journal is

c

The Royal Society of Chemistry 2012

110 C. Huber and G. Watchtershauser, Science, 1998, 281, 670.

111 H. Mita, S. Nomoto, M. Terasaki, A. Shimoyama and Y. Yamamoto, Int. J. Astrobiol., 2005, 4, 145.

112 E. Imai, H. Honda, K. Hatori, A. Brack and K. Matsuno, Science, 1999, 283, 831; Y. Ogata, E. Imai, H. Honda, K. Hatori and K. Matsuno, Origins Life Evol. Biospheres, 2000, 30, 527.

113 J. Bujda´k and B. M. Rode, Origins Life Evol. Biospheres, 1999, 29, 451.

114 M. G. Schwendinger and B. M. Rode, Anal. Sci., 1989, 5, 411.

115 S. W. Fox and H. J. Harada, J. Am. Chem. Soc., 1960, 82, 3745.

116 B. M. Rode, Peptides, 1999, 20, 773.

117 T. Vajda, Cell. Mol. Life Sci., 1999, 56, 398.

118 C. P. Ivanov and Slavcheva, Origins Life Evol. Biospheres, 1977, 8, 13.

119 J. L. Bada and G. D. McDonald, Icarus, 1995, 114, 139.

120 R. Liu and L. E. Orgel, Origins Life Evol. Biospheres, 1998, 28, 47.

121 P. Greenstein and M. Winitz, Chemistry of the amino acids, John Wiley & Sons, New York, 1961, vol. 1.

122 T. Vajda, CryoLetters, 1996, 17, 295–302; T. Vajda, G. Szo´ka´n. and M. Hollo´si, J. Pept. Sci., 1998, 4, 300.

123 T. Vajda, Cell. Mol. Life Sci., 1999, 56, 398.

Chem. Soc. Rev., 2012, 41, 5404–5415

5415

FULL PAPER DOI: 10.1002/chem.200802656

Synthesis of Pyrimidines and Triazines in Ice: Implications for the Prebiotic Chemistry of Nucleobases C sar Menor-Salv n,*[a] Dra. Marta Ruiz-Bermejo,[a] Marcelo I. Guzm n,[b] Susana Osuna-Esteban,[a] and Sabino Veintemillas-Verdaguer[a] Abstract: Herein, we report the efficient synthesis of RNA bases and functionalized s-triazines from 0.1 m urea solutions in water after subjection to freeze–thaw cycles for three weeks.

The icy solution was under a reductive, methane-based atmosphere, which was subjected to spark discharges as an energy source for the first 72 h of the experiment. Analysis of the products indicates the synthesis of the s-triazines cyanuric acid, ammeline, ammelide, and melamine, the pyrimidines cyto-

sine, uracil, and 2,4-diaminopyrimidine, and the purine adenine. An experiment performed as a control at room temperature, with the urea solution in the liquid phase and with the same atmosphere and energy source, led to the synthesis of hydantoins and insoluble tholin, but there was no evidence of Keywords: heterocycles nucleobases · origin of life prebiotic chemistry · triazines

Introduction The prebiotic synthesis of nucleic acid bases is a central issue in the RNA-world hypothesis, one of the main proposals for the origin of life, based on the self-assembly of nucleic acid monomers. Possible scenarios for the synthesis of nucleic acids are still under debate, and despite the abiotic synthesis of several nucleobases, the relevance of these syntheses to the origin of life is not well established.

In a pioneering work, Ferris et al. obtained cytosine from urea and cyanoacetylene and from sodium cyanate solution

[a] Dr. C. Menor-Salv n, D. M. Ruiz-Bermejo, S. Osuna-Esteban, Dr. S. Veintemillas-Verdaguer Centro de Astrobiolog a Consejo Superior de Investigaciones Cient ficas– Instituto Nacional de T cnica Aeroespacial (CSIC–INTA) Carretera Torrej n-Ajalvir, Km. 4,2 28850 Torrej n de Ardoz, Madrid (Spain) E-mail: menorsc@inta.es [b] Dr. M. I. Guzm n School of Engineering and Applied Sciences and Department of Earth and Planetary Sciences Harvard University Cambridge, MA 02138 (USA)

Chem. Eur. J. 2009, 00, 0 – 0

· ·

the synthesis of pyrimidines or triazines. The synthesis of pyrimidines from urea is possible under a methane/ nitrogen atmosphere only at low temperature, in the solid phase. The generation of both pyrimidines and triazines in comparable yields from urea, together with a possible role for triazines as alternative nucleobases, opens new perspectives on the prebiotic chemistry of informational polymers.

and cyanoacetylene.[1] The isolation of the latter from spark discharges in methane/nitrogen mixtures suggests the relevance of this synthesis to the origin of life.[1, 2] Cyanoacetylene is present in the atmosphere of Titan, in comets, and in the interstellar medium and, thus, is indeed of prebiotic relevance.[3] A quarter of a century later, Robertson and Miller reported the synthesis of cytosine in high yields (30–50 %) upon heating various concentrations of urea and cyanoacetaldehyde in a sealed ampoule.[4] Recently, the prebiotic relevance of pyrimidine synthesis from urea and cyanoacetaldehyde was discussed.[5, 6] The main concerns are the availability and instability of the reactants. In the case of cyanoacetaldehyde, it could 1) react with amino acids, 2) undergo hydrolysis to generate formate and acetonitrile, or 3) form a dimer. Consequently, any cyanoacetaldehyde is unlikely to survive long enough to be available in sufficient quantity to produce the necessary concentration for cytosine synthesis. On the other hand, urea decomposes to ammonia and carbon dioxide at pH < 5 and hydrolyses at pH > 5 with a gradual decrease in concentration.[7] Urea could also react with amino acids to form Ncarbamoyl amino acids or hydantoins.

A way to overcome these problems, along with the problem of cytosine instability through deamination to form uracil, has been hypothesized in which ice–water solutions

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&1&

These are not the final page numbers! ÞÞ

are used to generate the appropriate conditions for the reaction through the exclusion of solutes from concentrated interstitial brines in the ice matrix. This exclusion leads to a concentration enhancement in the microenvironments in the ice that has been shown to be of up to six orders of magnitude upon slow freezing of aqueous solutions with an organic molecule like methylene blue.[8] In addition, freezing extends the lifetime of labile molecules and gives time for them to be processed further, due not only to the protective effect of the ice but also to the low temperatures, which tend to retard reaction mechanisms that are prevalent in the liquid phase. Indeed, the unique environment generated in the interstitial liquid channels in ice, with high pressures and strong gradients, could favor certain reactions, such as nucleotide polymerization.[9] On the basis of this hypothesis, cytosine and uracil were synthesized from cyanoacetaldehyde and urea in a frozen solution at 11 8C, the eutectic point of the urea/water system.[10] However, the constraints on the prebiotic validity of a synthesis from two components, one of them very unstable, mean that the problem of the origin of pyrimidine nucleobases persists. Thus, an experimental prebiotic simulation that produces bases with a good yield and under a plausible combination of prebiotic components and energy is still necessary. When it is taken into account that no other synthesis of cyanoacetaldehyde besides that from cyanoacetylene has been demonstrated, along with the reported syntheses of cytosine from cyanoacetaldehyde and of cyanoacetylene from methane/nitrogen mixtures, it is theoretically possible to perform the synthesis of pyrimidines from a methane atmosphere.[11] However, to date, this synthesis has not been described in the literature.

This work explores the prebiotic importance of the suggested cyanoacetylene/urea pathway to cytosine and offers an icy scenario for the synthesis of pyrimidines and triazines from methane/nitrogen and urea. We also suggest a model for the possible chemical pathways generated by spark discharges in methane/nitrogen atmospheres and how the presence or absence of icy surfaces, in conjunction with urea as a reactant, drives the course of the reaction through these proposed chemical routes.

Results A freeze–thaw cycle (repetition of 5 to 5 8C variations) was generated in a degassed, sterile 0.1 m urea solution (pH 7.1) under a nitrogen/hydrogen/methane (30:30:40) atmosphere by using a thermostatic reactor (Figure 1). This atmosphere is reducing in nature and constitutes the primary carbon source for the experiment.[12] Once the freeze–thaw cycle was established in the entire solution pool, the system was energized by means of spark discharges during the solid phase of the cycle. A tungsten electrode connected to a high-frequency and high-voltage generator induced a discharge that impacted the water/ice surface. As a reference, the solution is fully frozen at 5 8C. The colligatively de-

&2&

www.chemeurj.org

Figure 1. Illustration of the experimental system used in this study. A thermostated reactor equipped with an energy source was filled with the gas mixture used in the experiment and with the liquid urea solution.

After sealing, the temperature cycles were established by using a programmable cryostat filled with silicon oil. 1) and 2) Tungsten electrodes connected to a high-voltage (50 KV) generator and to the ground, respectively; 3) urea solution in water.

pressed freezing temperature (estimated from the cryoscopic constant of H2O, l = 1.86 K mol 1) was Tf 273.15–1.86ACHTUNGRE[urea] = 272.96 K for 0.1 m urea.

After 72 h, the voltage generator was disconnected, and the reactor was maintained with active freeze–thaw cycling.

After 3 weeks, the reactor content, a yellowish solution with a pH value in the range 8.4–8.8 (3 experiments performed) and without any suspended organic solid, was allowed to warm to room temperature and was collected as fast as possible in sealed headspace vials under a nitrogen atmosphere to avoid excessive manipulation and to minimize the contamination risk. The solution was freeze dried, and the dry yellowish solid was analyzed.

The reaction product (Figure 2) contained unreacted urea and the triazines cyanuric acid (2,4,6-trihydroxy-s-triazine), ammelide (6-amino-2,4-dihydroxy-s-triazine), ammeline (4,6-diaminotriazin-2-ol), and melamine (2,4,6-triamino-s-triazine). These compounds were identified as tri(trimethylsilyl) (TMS) derivatives, by using the agreement of the retention times with those of standards as the criterion for identification, together with the mass spectra, by using the following ions for determination: m/z 345 and 330 for cyanuric acid; m/z 344, 329, and 171 for ammelide; m/z 343, 328, and 171 for ammeline; and m/z 342, 327, and 171 for melamine.

The reaction also yielded the pyrimidines cytosine, uracil, and 2,4,5-trihydroxypyrimidine. These pyrimidines, 2,4-diaminopyrimidine, and the purine adenine were identified as TMS derivatives by using the following ions for determination: m/z 254, 240, and 170 for cytosine; m/z 256 and 241 for uracil; m/z 279 and 264 for adenine; m/z 254 and 239 for 2,4-diaminopyrimidine; and m/z 344 and 329 for 2,4,6-

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Chem. Eur. J. 0000, 00, 0 – 0

Synthesis of Pyrimidines and Triazines in Ice

FULL PAPER

Figure 2. Gas chromatogram showing the trimethylsilyl derivatives of pyrimidines and triazines obtained from spark discharges between a CH4/N2/H2 atmosphere and a 0.1 m urea solution subjected to freeze–thaw cycles. Residual urea (not shown) appears as a strong peak (urea–TMS derivative; m/z 204, 189, 147, and 73) at a retention time of 19 min.

trihydroxypyrimidine. The major products were cyanuric acid and cytosine. Overall, the ratio of cytosine/hydroxypyrimidines was 1 (Table 1). The only amino acids identified in the reaction product were glycine (not shown in the chromatogram), asparagine, and aspartic acid. The hydantoins, which were the major products in the control experiment (see below), were also present in minor quantities.

To test the effect of the freeze–thaw cycles on the chemistry of the system, one control experiment was performed at room temperature with the same experimental setup. Under

Table 1. Yields of triazines and nucleobases obtained by sparking of a 0.1 m urea solution subjected to freeze–thaw cycles under a CH4/N2/H2 atmosphere (experiment 1), subjected to freeze–thaw cycles under an inert atmosphere (experiment 2), or under the atmosphere of experiment 1 with the urea solution maintained in the solid phase (experiment 3).

Yields are calculated based on the total urea amount introduced into the experiment.

Compound

Experiment 1 Experiment 2 Experiment 3 Yield [%] Yield [%] Yield [%]

cyanuric acid ammelide ammeline melamine cytosine uracil 2,4,6-trihydroxypyrimidine adenine

7.1 2.8 1 0.02 4.2 1.9 0.9 0.15

Chem. Eur. J. 2009, 00, 0 – 0

2.8 1.6 – 0.1 – – – –

2.1 0.8 detected – 1.6 1 – –

these conditions, the reaction product contained a significant quantity of insoluble organic material (32.5 % of the carbon introduced as methane and urea in the experiment). The solution contained unreacted urea, glycine (not shown in the chromatogram), hydantoin, substituted hydantoins, parabanic acid, and carbamoyl-glycine (Figure 3). The pyrimidines and purines were undetected and the triazines were detected with low yields (< 0.01 %).

Due to the known behavior of urea in an aqueous solution subjected to a phase change, we took into account the possibility of the generation of triazines directly from urea by a mechanism of polymerization similar to those proposed for triazine generation from molten urea. A control experiment was performed by sparking the urea solution with the same conditions as the freeze–thaw cycles but under an inert atmosphere (argon). Under these conditions, we saw the formation of triazines but not pyrimidines (Figure 4).

The yields of the triazines obtained in this experiment were significantly lower than those under a methane/nitrogen atmosphere. This could be due to argon ionization, which diverts the electric discharge in all directions around the electrode and diminishes the energy input into the ice–water pool. It is interesting to note that we found melamine but not ammeline under these conditions. Further study of the chemistry of urea solutions with energy sources and the effect of concentration and temperature on this is in progress in our laboratory.

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&3&

These are not the final page numbers! ÞÞ

C. Menor-Salv n et al.

Figure 3. Gas chromatogram showing the trimethylsilyl derivatives of hydantoins obtained from spark discharges between a CH4/N2/H2 atmosphere and a 0.1 m urea solution at room temperature. Residual urea (not shown) appears as a strong peak at a retention time of 19 min.

Figure 4. GC/MS chromatogram (total-ion-count mode) showing the trimethylsilyl derivatives of triazine derivatives obtained from a urea solution subjected to freeze–thaw cycles and spark discharges under an inert atmosphere (argon). Due to the lower yields obtained under these conditions, the residual urea appears as a strong peak that saturates the system. To avoid damage to the mass spectrometer, the ion registration begins at 26 min.

The effect of freeze–thaw cycles is a key phenomenon, as we showed by comparison with an experiment performed entirely at low temperature with water in the solid phase.

Under these conditions, we found that mostly triazine deriv-

&4&

www.chemeurj.org

atives were formed. Also, we found that the concentration of pyrimidines and the overall yields were significantly lower than those found for the freeze–thaw-cycle experiments.

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Chem. Eur. J. 0000, 00, 0 – 0

Synthesis of Pyrimidines and Triazines in Ice

To test whether urea suffers degradation or biological contamination in solution when subjected to freeze–thaw cycles in our reactor design, a third control experiment was performed without spark discharges under the methane/nitrogen/hydrogen atmosphere. After three weeks of freeze– thaw cycles, the solution content was analyzed; no evidence of degradation, microbial contamination, or the generation of nitrogen heterocycles was found.

Discussion Our experiments show that the synthesis of pyrimidines under a methane/nitrogen atmosphere is possible with high yields if a urea source is present. In this process, the presence of frozen water or ice is a decisive factor. With water subjected to freeze–thaw cycles, the synthesis of pyrimidines and triazines is strongly favored and the generation of insoluble or polymeric organic materials or tholins is negligible. Previously reported experiments on the synthesis of cytosine from cyanoacetaldehyde and guanidine or urea in ice at 11 8C suggested a preference for uracil generation and the deamination of aminopyridines or their precursors during formation.[10] Under our conditions, the observed preference sequence of pyrimidines was cytosine > uracil > 2,4-diaminopyrimidine > 2,4,6-trihydroxypyrimidine. Further work is in progress to establish whether long-term storage in ice under freeze–thaw cycles alters the concentrations of the resultant pyrimidines.

The behavior of urea in an aqueous solution subjected to a phase change could be a key factor in the results observed with water in different phases. The urea molecules in fluidwater solutions form pronounced hydrogen bonds with the neighboring water molecules at both the amino and the carbonyl groups.[13] The number of hydration water molecules per molecule of urea has been reported as approximately 2 at concentrations of less than 5.0 m urea.[14] Neutron diffraction measurements at 25 8C on aqueous 15 mol % ( 10 m) urea showed that approximately 4.3 water molecules are hydrogen bonded to the carbonyl oxygen atom.[15] Infrared and dielectric spectroscopy studies showed that the two predominant interactions are those of urea–urea (observed at urea concentrations higher than 1 m) and water–urea. At a concentration of 11 m, nearly all of the urea molecules have other urea molecules as their nearest neighbors, because few water molecules remain to hydrate the urea molecules at high concentrations. Concentration-dependence studies showed that this is due to the aggregation of urea molecules in dimers and/or oligomers at higher concentrations.[14, 16] There is a considerable amount of urea dimer or clusters present in aqueous solutions.[13] Upon freezing of an aqueous solution of a polar organic compound, the solute is segregated from the pure ice to accumulate in supercooled microfluids. The solute retains a certain amount of water in the form of a supersaturated solution. Under these conditions, the system is governed by dehydration and the association of the solute molecules.[17]

Chem. Eur. J. 2009, 00, 0 – 0

FULL PAPER In analogy to another polar organic compound, such as pyruvic acid, in frozen aqueous solutions,[18] we can expect that the urea remains monomeric in frozen aqueous urea solutions. The extent of urea dimerization ( 18 % in 0.1 m urea at 25 8C[19]) is expected to increase at lower temperatures and higher urea concentrations and to approach completion a few degrees below the onset of freezing.[8, 17] Thus, in our frozen-water experiment, urea is present as hydrogenbonded urea dimers or higher association clusters and is randomly distributed within the ice. In this environment, water can still hydrate the polar organic molecules,[17, 18] but the urea–urea interactions will predominate.

Triazine derivatives are obtained in the laboratory and industry as products of the pyrolysis of urea.[20] These compounds were found in the Murchinson and Orgueil meteorites, and it has been argued that their origin could be the hydrolysis of macromolecular material present in the meteorite[21] or the polymerization of hydrogen cyanide,[22] which could form the unsubstituted s-triazine. We have shown that the polymerization of HCN is not a necessary condition for the generation of triazines when the urea–water system is present at low temperatures (Figure 4). This implies that several alternative pathways could be active, dependent on the temperature and the presence of the reactants. As in the case of the synthesis of polycyclic aromatic hydrocarbons (PAHs)[23] and as stated in the previous paragraph, we could expect that the ice medium, due to its surfaces and the freezing process, would favor a mechanism similar to those of the pyrolysis processes of urea, which could explain the synthesis of triazines.

In fact, the control experiment performed under an inert atmosphere shows the formation of triazines in the same relative abundance as the main experiment. During pyrolysis, part of the urea undergoes isomerization to ammonium isocyanate and decomposition, which releases ammonia. The reaction of the isocyanic acid, also released, with urea generates biuret (aminocarbonyl urea). The biuret reacts with more isocyanic acid and generates cyanuric acid and ammonia, ammelide, or ammeline. The latter could react with ammonia to form melamine. This reaction sequence, proposed for the pyrolysis of urea, is not observed by heating below the urea melting point and is consistent with the relative abundance of s-triazines found with our conditions: cyanuric acid > ammelide > ammeline > melamine.[20] The photodecomposition of urea under UV radiation is a possible source of isocyanic acid and ammonia as an alternative to thermal decomposition.[24] Under our conditions, dissipation of spark energy in the form of ultraviolet radiation and the associated photochemical processes could be one of the sources of isocyanic acid, together with the local effect of the spark and the behavior of urea in frozen solutions. In fact, the fluid microenvironments or urea oligomers generated in the ice matrix could be comparable to the molten urea and could facilitate the condensation reactions.

With regard to the relevance of our results to prebiotic evolution, the importance of functionalized s-triazines would not be diminished, because they could act as purines

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&5&

These are not the final page numbers! ÞÞ

C. Menor-Salv n et al.

or pyrimidines in nucleic acid molecules,[25] a role studied by gine or aspartic acid under hydrolysis-favorable conditions, Mittapalli et al.,[26] who observed that oligopeptoids tagged as a byproduct.[5] We found both asparagine and aspartic with triazines as potential primordial informational polyacid, which suggests that, when a urea solution undergoes mers showed an imbalance of pairing behavior and weak freeze–thaw cycles, the generation of pyrimidine nucleobasbase pairing. On the other hand, 2,4-diaminotriazines in es is the preferential reaction. peptidic oligomers establish strong base pairing with 2,4-diThe urea solution subjected to freeze–thaw cycles favors oxopyrimidines in nucleic acid backbones.[26] We obtained the generation of nucleobases, diminishes the formation of insoluble tholin and other polymeric organic material, and pyrimidines and triazines as the main products and, taking reduces the importance of side reactions between urea and into account the low yields of purine bases in the synthesis, amino acids or their precursors generated in classic spark we do not preclude the possibility of a primordial informaexperiments. Moreover, the pyrimidines are more stable at tional polymer containing triazines that was gradually rejectlow temperatures, and the ice matrix gives them a chance to ed during evolution due to weak base pairing and instability undergo further chemical evolution to higher complexity. issues.

The microenvironments in which the molecules are located The control experiment performed with water in the are surrounded by a medium that is more reflective and can liquid phase at ambient temperature with the same atmosphere and urea concentration did not show the formation of either triazines or pyrimidines (Figure 3). The main organic material obtained in the liquidphase experiment was insoluble tholin, with similar properties to the material obtained in previously reported experiments performed in our laboratory at room temperature.[27] The soluble organic material obtained in the control experiment in the liquid phase contained mainly unreacted urea and hydantoins.

The generation of hydantoins could be related to reactions between urea and amino acids, amino nitriles, or carbonyl compounds, generated in the spark experiment at room temperature.[26, 27] The Urech synthesis from a-amino acids and cyanate (which could be generated by urea isomerization) and the Siemonson synthesis or Heintz synthesis from urea yield 5functionalized hydantoins. The Bucherer–Bergs synthesis yields 5,5’-functionalized hydantoins.[28] We also identified carbamoyl-glycine at ambient temperature. This product, formed by the reaction of urea with glyScheme 1. Possible synthetic pathways in a urea solution subjected to freeze–thaw cycles under a methane/nicine, appears to be absent in trogen atmosphere, with spark discharges between an electrode and the ice/liquid surface as an energy source. the experiment performed in The pathways lead to those chemical families found in the reaction product: the synthesis of aromatics by the ice. However, given that cya- generation and polymerization of acetylene, the synthesis of pyrimidines from cyanoacetylene and urea, the noacetylene and hydrogen cya- synthesis of triazines from the possible isocyanic acid generated during urea decomposition, and the synthesis of hydantoins from urea and amino acids, cyanohydrins, hydroxy acids, or other carbonylic precursors. These nide are the main products of suggested pathways are based on the final products obtained. The pathway leading to hydantoin synthesis and spark discharges, we must the generation of insoluble organic material (tholin) prevails in the liquid phase. By contrast, the frozen expect the formation of aspara- medium favors the generation of pyrimidines and triazines.

&6&

www.chemeurj.org

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Chem. Eur. J. 0000, 00, 0 – 0

Synthesis of Pyrimidines and Triazines in Ice

be protective towards photodimerization. It also provides higher stability against deamination due to lower temperatures, which prevent thermal decomposition within the timescale of our experiment. Our results confirm the potential importance of ice in planetary environments, such as Mars, Saturn’s icy moons, or the primitive Earth, as previously suggested.[11] In summary, the spark discharges in a methane/nitrogen atmosphere and freeze–thaw cycles operate by two basic mechanisms: first, the generation and polymerization of acetylene, which leads to aromatic compounds and acetylene-derived polymers, as previously reported in experiments with pure water under freeze–melt conditions,[23] and second, the generation of cyanide derivatives, such as cyanoacetylene. Urea could act as a cyanoacetylene scavenger, favoring the mechanism of formation of nitrogen derivatives (Scheme 1). Therefore, the presence of urea reduces the availability of reactants for the generation of cyanide or acetylene polymers and explains the lack of insoluble organic materials obtained in experiments with urea/ice. At room temperature, the classic mechanism of atmospheric generation of tholins and amino acid precursors appears to prevail and leads to the formation of hydantoins without the formation of pyrimidines or triazines, despite the presence of excess urea in the solution.

Conclusions The freezing process offers an interesting environment that favors the abiotic synthesis of molecules of biochemical interest and could be a favorable scenario for the laboratory exploration of prebiotic chemistry in planetary environments. The ice matrix plays the role of a protective medium that avoids the degradation of molecules such as the pyrimidines, enhances the yields, and diminishes the side reactions, which constitute the constraints for the actual prebiotic relevance of cyanoacetylene, acetylene, or urea.

The classical synthesis of cytosine and uracil from cyanoacetylene/cyanoacetaldehyde and urea could only be performed by using a urea solution under a methane/nitrogen atmosphere if the solution is subjected to freeze–thaw cycles.

The high-yield synthesis of triazines opens a way to explore these molecules as mimics of prebiological nucleobases in the early stages of evolution and could help to explain their presence in meteorites.

FULL PAPER (electrically earthed with a secondary tungsten electrode). Freeze–thaw cycles were established by varying the temperature between 5 8C and 5 8C (2 h at 5 8C, ramping to 5 8C at a rate of 0.1 8C min 1, and 2 h at 5 8C) with a Haake Phoenix II programmable cryostat (Thermo Electron Corporation). The system was maintained under spark discharge for 72 h.

After this period, the high-voltage generator was disconnected and the reactor kept sealed with active freeze–melt cycles for three weeks. After that, the system was allowed to warm to room temperature, and the yellowish aqueous solution was stored in sealed headspace vials under an inert atmosphere.

For the control experiment with an inert atmosphere, argon of the maximum purity available (99.9995 %) was used.

Gas chromatography–mass spectrometry: The solid, freeze-dried sample was derivatized with N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) with 1 % trimethylchlorosilane (TMCS; provided by Pierce, Rockford, IL, USA) according to this protocol: dried sample (1 mg) was combined with BSTFA/TMSC (0.1 mL) in a dry glass vial, and the mixture was stirred and heated at 60 8C for 3 h. After that, a sample (1 mL) was transferred to the injection port of a GC apparatus in the splitless mode, and the analysis was performed by using an Elite-5 (Perkin–Elmer) 5 % phenyl–95 % methylsiloxane capillary column (30 m 0.25 mm; 0.25 mm film).

Mass spectrometry analysis was performed by using a Perkin–Elmer Autosystem XL-Turbomass Gold quadrupole apparatus in the EI + scan mode. Organic compounds were identified by a search for their mass spectra in the NIST database, and the identified compounds were confirmed against authentic standards (provided by Sigma–Aldrich) by comparison of mass spectra and retention times. For identification purposes, we considered only peaks with a signal-to-noise ratio over 20. Those peaks for which the match probability in the database was below 90 %, those that were matched but for which authentic samples were commercially unavailable, and those that were only tentatively identified were considered unidentified and are not discussed in this paper.

Acknowledgements We are grateful to James P. Ferris for his helpful comments and critical revision of our results. We are thankful for the research facilities of the Centro de Astrobiolog a (CAB) and for grants from the Instituto Nacional de T cnica Aeroespacial “Esteban Terradas” (INTA) and project AYA2006-15648-C02-02 of the Ministerio de Educaci n y Ciencia (Spain).

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

Experimental Section Urea-solution freeze–thaw cycles: 0.1 m Urea solution (50 mL) in ultrapure, degassed water was frozen at 5 8C in a sealed and thermostatized glass reactor under an atmosphere of nitrogen (30 %), hydrogen (30 %), and methane (40 %) at atmospheric pressure. The system was energized with a high-voltage generator (Model BD-50E, Electrotechnic Products Inc., IL, USA) by high-frequency spark discharges (50 KV, 0.5 MHz) directly into the water, through a tungsten electrode attached to the reactor

Chem. Eur. J. 2009, 00, 0 – 0

[11] [12] [13] [14] [15]

J. Ferris, R. Sanchez, L. Orgel, J. Mol. Biol. 1968, 33, 693 – 704.

R. Sanchez, J. Ferris, L. Orgel, Science 1966, 154, 784 – 785.

D. W. Clarke, J. P. Ferris, Icarus 1995, 115, 119 – 125.

M. Robertson, S. L. Miller, Nature 1995, 375, 772 – 774.

R. Shapiro, Proc. Natl. Acad. Sci. USA 1999, 96, 4396 – 4401.

R. Shapiro, Origin Life Evol. Biochem. 2002, 32, 275 – 278.

R. C. Warner, J. Biol. Chem. 1941, 137, 705 – 723.

D. Heger, J. Jirkovsky, P. Klan, J. Phys. Chem. A 2005, 109, 6702 – 6709.

H. Trinks, W. Schrçder, C. K. Biebricher, Origin Life Evol. Biochem.

2005, 35, 429 – 445.

H. J. Cleaves II, K. E. Nelson, S. L. Miller, Naturwissenschaften 2006, 93, 228 – 231.

L. Orgel, Origin Life Evol. Biochem. 2002, 32, 279 – 281.

M. Ruiz-Bermejo, C. Menor-Salvan, S. Osuna-Esteban, S. Veintemillas, Origin Life Evol. Biochem. 2007, 37, 123 – 142.

A. K. Soper, E. W. Castner, A. Luzar, Biophys. Chem. 2003, 105, 649 – 666.

Y. Hayashi, Y. Katsumoto, S. Omori, N. Kishii, A. Yasuda, J. Phys.

Chem. A 2007, 111, 1076 – 1080.

Y. Kameda, M. Sasaki, S. Hino, Y. Amo, T. Usuki, Bull. Chem. Soc. Jpn. 2006, 79, 1367 – 1371.

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&7&

These are not the final page numbers! ÞÞ

C. Menor-Salv n et al.

[16] G. Grdadolnik, Y. Marechal, J. Mol. Struct. 2002, 615, 177 – 189.

[17] M. I. Guzm n, L. Hildebrandt, A. J. Colussi, M. R. Hoffmann, J. Am. Chem. Soc. 2006, 128, 10 621 – 10 624.

[18] M. I. Guzm n, A. J. Colussi, M. R. Hoffmann, J. Phys. Chem. A 2006, 110, 931 – 935.

[19] R. H. Stokes, J. Phys. Chem. 1965, 69, 4012 – 4017.

[20] P. Schaber, J. Colson, S. Higgins, D. Thielen, B. Anspach, J. Brauer, Thermochim. Acta 2004, 424, 131 – 142.

[21] Z. Martins, O. Botta, M. L. Fogel, M. A. Sephton, D. P. Glavin, J. S. Watson, J. P. Dworkin, A. W. Schwartz, P. Ehrenfreund, Earth Planet. Sci. Lett. 2008, 270, 130 – 136.

[22] C. N. Matthews, R. D. Minard, Proc. IAU Symposium 2008, 4, 453 – 458.

[23] C. Menor-Salvan, M. Ruiz-Bermejo, G. MuÇoz-Caro, S. Osuna, S. Veintemillas, Chem. Biodiv. 2009, 5, 2729 – 2739.

&8&

www.chemeurj.org

[24] F. Duvernay, T. Chiavassa, F. Borget, J. P. Aycard, J. Phys. Chem. A 2005, 109, 208 – 218.

[25] M. Hysell, J. S. Siegel, Y. Tor, Org. Biomol. Chem. 2005, 3, 2946 – 2952.

[26] G. K. Mittapalli, K. R. Reddy, H. Xiong, O. Munoz, B. Han, F. De Riccardis, R. Krishnamurthy, A. Eschenmoser, Angew. Chem. 2007, 119, 2522 – 2529; Angew. Chem. Int. Ed. 2007, 46, 2470 – 2477.

[27] M. Ruiz-Bermejo, C. Menor-Salvan, E. Mateo-Mart , S. Osuna, J. A. Martin-Gago, S. Veintemillas, Icarus 2008, 198, 232 – 241.

[28] E. Ware, Chem. Rev. 1950, 46, 403 – 470.

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Received: December 17, 2008 Published online: && &&, 2009

Chem. Eur. J. 0000, 00, 0 – 0

Synthesis of Pyrimidines and Triazines in Ice

FULL PAPER Origins of Life C. Menor-Salv n,* D. M. Ruiz-Bermejo, M. I. Guzm n, S. Osuna-Esteban, S. Veintemillas-Verdaguer . . . &&&&—&&&&

From urea to nucleobases: Freeze– thaw cycles in urea (1) solutions under methane/nitrogen atmospheres lead, with application of an energy source, to the synthesis of pyrimidines (mainly cytosine (2) and uracil (3)), triazines (such as cyanuric acid (4)), and ade-

Chem. Eur. J. 2009, 00, 0 – 0

nine. This synthesis appears to be dependent on the atmosphere and the freezing conditions. At room temperature, hydantoin (5) is obtained. However, a freezing urea/water system subjected to an energy source under an inert atmosphere generates s-triazines.

2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Synthesis of Pyrimidines and Triazines in Ice: Implications for the Prebiotic Chemistry of Nucleobases

www.chemeurj.org

&9&

These are not the final page numbers! ÞÞ

Life 2013, 3, 502-517; doi:10.3390/life3030502 OPEN ACCESS

life ISSN 2075-1729 www.mdpi.com/journal/life Article

Natural Pyrrhotite as a Catalyst in Prebiotic Chemical Evolution Alejandra López Ibáñez de Aldecoa 1, Francisco Velasco Roldán 2 and César Menor-Salván 1,* 1

2

Centro de Astrobiología (CSIC-INTA), 28850 Torrejón de Ardoz, Madrid 28850, Spain; E-Mail: lopezima@cab.inta-csic.es Departamento de Mineralogía y Petrología, Facultad de Ciencia y Tecnología, Universidad del País Vasco, Bilbao 48080, Spain; E-Mail: francisco.velasco@ehu.es

* Author to whom correspondence should be addressed; E-Mail: cmenor@mtiblog.com; Tel./Fax: +34-646-321-775.

Received: 3 June 2013; in revised form: 19 August 2013 / Accepted: 20 August 2013 / Published: 28 August 2013

Abstract: The idea of an autotrophic organism as the first living being on Earth leads to the hypothesis of a protometabolic, complex chemical system. In one of the main hypotheses, the first metabolic systems emerged from the interaction between sulfide minerals and/or soluble iron-sulfide complexes and fluids rich in inorganic precursors, which are reduced and derived from crustal or mantle activity. Within this context, the possible catalytic role of pyrrhotite, one of the most abundant sulfide minerals, in biomimetic redox and carbon fixation reactions was studied. Our results showed that pyrrhotite, under simulated hydrothermal conditions, could catalyze the pyruvate synthesis from lactate and that a dynamic system formed by coupling iron metal and iron-sulfur species in an electrochemical cell could promote carbon fixation from thioacetate esters.

Keywords: pyrrhotite; life origin; pyruvate; lactate; reductive carboxylation; thioesters

1. Introduction Since the relationship between hydrothermal systems and the origins of life was first proposed [1,2] the complete geochemical perspective of this relationship has been described in detail [3].

The experiments that have replicated sulfide rich hydrothermal solution chemistry have confirmed the

Life 2013, 3

503

potential of these systems to synthesize organic molecules through carbon fixation. Ferrous sulfide minerals, which usually precipitate in the submarine hydrothermal systems, that are candidate environments for the origin of life, play a key role in this process. These iron minerals include pyrrhotite, as well as mackinawite, greigite, and violarite [4], which are unstable minerals under aerobic conditions.

The hypothesis of a chemoautotrophic origin of metabolism, proposed by G. Wächtershäuser, suggested that iron sulfide surfaces could support an autocatalytic chemolithotrophic metabolism driven by the exergonic formation of pyrite (FeS2) from more reduced minerals, such as pyrrhotite or pentlandite, in the presence of free sulfide [5]. The result that synthetic (Fe,Ni)S catalyzed the synthesis of methyl mercaptan and acetic acid, using CO as an inorganic carbon source, reinforced the idea of a prebiotic chemistry associated with iron sulfide minerals, although the mineral phases involved in this experiment have not been identified [6]. In a parallel elaboration of the iron sulfide scenario, the M. Russell model [7] incorporated the idea that compartmentalization (with membranous FeS precipitates as ancestral compartments) is essential for the reactive concentration and the origin of a gradient dissipation based carbon fixation metabolism, which is driven by chemiosmotic and proton-motive forces [8,9]. The carbon fixation of organic material is an obvious prerequisite for life, and Earth’s current biochemistry maintains five autotrophic pathways in a metabolic strategy that is highly conservative in anaerobic autotrophic prokaryotes [10]. The characteristics and preservation of carbon fixation pathways in the extant biochemistry caused us to look back to the beginning of biological evolution associated with the development of an autocatalytic autotrophic biochemical system, in which transition metal complexes containing proteins play an essential role [11]. In particular, the role of iron sulfur proteins, which contain Fe-S clusters as active centers in the electron transfer reactions in the CO2 fixation pathways, and their occurrence in what are possibly the most primitive steps of oxidation of organic substrates, such as succinate, in the cell energy transduction machinery, have been the key evidence used to connect the origin of biochemistry and the geochemistry of the origin of life [12]. Thus, considering the structural similarity between the biological iron-sulfur clusters and the crystal structure of iron sulfide minerals [13], the biomimetic activity of synthetic soluble Fe-S clusters [14] and the highly preserved and ancient biochemical reactions involved could explain why Fe-S clusters are found in all biological systems and why iron sulfide clusters were chosen by nature rather than other metal clusters or organometallic catalysts. In this sense, Fe-S clusters could be “living fossils” that trace the origin of life to the mineral roots of biochemistry.

Using the idea of the construction of a model of protometabolism catalyzed by ancient iron-sulfur active centers, we focused on the origination of pyruvate metabolism on pyrrhotite matrix. We selected pyrrhotite, due to its stability relative to more reduced iron sulfide minerals and due to the central role that this mineral plays in the iron-sulfur geochemistry: its transformation in pyrite could provide redox energy to the system, could be a source of soluble FeS clusters [15], and its surface could have efficient catalyst and electron transfer properties.

Recently, it has been demonstrated that the interconversion of hydroxyl acids and keto acids could be catalyzed by the FeS/FeS2 system [16]. Using this reference, we tested the feasibility of using natural iron sulfide pyrrhotite to mimic oxidoreductase activity in the interconversion of pyruvic and lactic acids. The pyruvate is a central metabolite in the Archaea, Bacteria, and Eukarya kingdoms, and iron-sulfur enzymes are involved in the reactions that link pyruvate with carbon fixation pathways and with thioester biochemistry. In many anaerobic autotrophic organisms, the reductive carboxylation

Life 2013, 3

504

of acetyl-CoA to pyruvate could be catalyzed by pyruvate-ferredoxin oxidoreductase (PFOR), which acts as pyruvate synthase using the iron-sulfur protein ferredoxin as an electron donor [17].

This reaction connects the Wood-Ljundhal pathway with the reductive Krebs cycle to generate biosynthetic intermediates in the anabolic metabolism of Archaea, such as Methanobacterium thermoautotrophicum. The PFOR is an ancient heterotetrameric molecule present in all Archaea, autotrophic bacteria and anaerobic protozoa. The ancestral subunit is ferredoxin-like and binds iron-sulfur clusters [18], reinforcing the idea of its prebiotic origin.

However, the reactivity of thioesters, their biochemical role, and their plausible prebiotic formation suggest that primitive analogs of coenzyme A thioesters could have been involved in the origin of biochemistry [19]. Similarly, it has been proposed that the reductive carboxylation of thioacetic acid using FeS as an electron donor to form pyruvic acid is a possible starting point for a prebiotic thio-analog of the reductive Krebs cycle [20]. The standard reduction potential of pyruvate synthesis by reductive carboxylation is Eº = −500 mV [10]. To drive the reductive carboxylation forward, low potential electron donors are needed. In this work, we explore the possibility of connecting the biochemical reductive carboxylation of thioesters to the geochemistry of the iron-sulfide system.

Considering iron sulfide clusters and minerals as prebiotic analogs of ferredoxin iron-sulfur clusters and the iron metal/pyrrhotite/pyrite system as a source of reducing power, we performed the reductive carboxylation of ethyl thioacetate to model the more complex acetyl-CoA thioester. This constitutes a test for the pyrite-pulled intermediary metabolism proposed by Wächtershäuser and the suggested synthesis of pyruvic acid from thioacetic acid and CO2 [5,20,21].

2. Experimental Section 2.1. Pyrrhotite Mineral Characterization All of the pyrrhotite samples used were from the Gualba deposit (Barcelona, Spain), a Fe-Cu polymetallic skarn formed by the replacement of the Cambrian-Ordovician carbonate rocks after intrusion of a body of granodioritic composition of Hercynian age [22]. The pyrrhotite was characterized by electron and optical microscopy, electron microprobe analysis (EMPA-WDS), and X-ray diffraction (XRD). The composition was determined to be Fe0.89S, and the mineral consists of a mixture of hexagonal and monoclinic polytypes. The ore was selected for its massive character and purity, which minimized the trace mineral and organic carbon contents. The Gualba pyrrhotite is strongly ferrimagnetic; this characteristic has been used for the isolation of pyrrhotite from the ore paragenesis, which includes traces of pyrite, chalcopyrite, and very subordinate amounts of arsenides and non-metallic minerals of the mica and chlorite groups (Figure 1). The nickel content of the pyrrhotite from the Gualba quarries was 0.012–0.018 wt%, with negligible copper and cobalt.

The ore samples were crushed and pyrrhotite was magnetically separated; the pyrrhotite powder was analyzed by XRD before the experiments to assure that it was pyrite free (Figure 2). To eliminate potential organic contaminants, the nearly pure pyrrhotite concentrate was extracted first with dichloromethane and then with methanol and then dried and stored in an anoxic atmosphere.

Life 2013, 3

505

Figure 1. Texture of the pyrrhotite ore assemblage from the Gualba deposit used in the experiments after separation of the accompanying minerals. Po: pyrrhotite; Ccp: chalcopyrite; Py: pyrite; Chl: chlorite minerals. Note the appearance of products of alteration after polishing close to the microfractures.

Figure 2. Powder X-ray diffraction of the Gualba pyrrhotite concentrate after magnetic separation and washing. The presence of pyrite and other minerals is negligible.

Po Po Po

Intensity

Po

40

20

60

2- (deg) CuKα

2.2. Pyrrhotite in Lactate-Pyruvate Oxidoreduction All of the experiments were conducted in sealed 22 mL headspace vials and in a glove box with a nitrogen atmosphere. Each vial was filled with 1 mmol of crushed pyrrhotite, 1 mmol of elemental sulfur (S), and 5 mM of buffer phosphate (with different pH: 2, 5 and 8). Subsequently, 100 µmol of pyruvic acid or 200 µmol of lactic acid were added, together with 0.5 mmol of sodium sulfide and 0.5

Life 2013, 3

506

mmol of sulfuric acid, which served to provide the H2S. The reactions were performed at 130 °C for a period of 5 h. Finally, the aqueous phase was separated and analyzed by gas chromatography-mass spectrometry. The remaining mineral fraction was characterized by XRD to determine the mineral phase transformations.

2.3. Metallic Iron/Pyrrhotite in Lactate-Pyruvate Synthesis An electrochemical cell has been constructed using a cylindrical graphite reactor filled with granulated iron metal under a 10 mM sodium bisulfate solution (pH 5.5), a microporous clay barrier and pyrrhotite wet paste formed by pyrrhotite powder, a 50 mM ammonium bicarbonate solution containing 1 mmol of sodium sulfide (pH 9) and, optionally, 1 mmol of hydroquinone. Previously, 1 mmol of ethyl thioacetate was adsorbed by the pyrrhotite powder. A graphite electrode inserted in the pyrrhotite constitutes the cathode. The constructed cell showed a direct measured potential of 0.400 V to 0.835 V, depending on the composition and the pH of the solutions, using the reactor as the anode and the graphite electrode as the cathode. The cell was connected to the circuit depicted in Figure 3. The value of resistor R1 is 1.5 K and the value of resistor R2 is 100 K. The voltmeter measures the external voltage source plus the voltage provided by the iron/pyrrhotite cell. The system was connected under an anoxic (N2) atmosphere in a glove box; all of the instruments and materials used were first sterilized to minimize bacterial contamination. After 72 h, the reactor content was extracted with hot water, and the solution was filtered through a cation exchange resin column (Dowex 50w X8) in H+ form to remove dissolved iron and other metallic cations. The clear, slightly yellowish solution was freeze-dried and stored at −20 °C until the chromatographic analysis was performed.

Figure 3. Electrochemical cell designed to perform the reductive carboxylation of thioacetate ester. See text for details and resistor values.

R1

V

+

Graphite electrode Pyrrhotite

1.1V

Clay

R 2

Iron metal ‐ Graphite reactor

A

2.4. Gas Chromatography-Mass Spectrometry The solid, freeze-dried samples were derivatized with N,O-bis(trimethylsilyl) trifluoroacetamide (BSTFA) with 1% trimethylchlorosilane (TMCS, provided by Pierce, Rockford, IL, USA). Briefly, 1 mg of dried sample was combined with 0.1 ml of BSTFA+TMSC in a dry glass vial, stirred, and heated at 60 °C for 3 h. Following that, 1 µL of sample was transferred to an Agilent GC 6850A gas

Life 2013, 3

507

chromatograph in the splitless mode with the injection port at 290 °C. The analysis was performed using an HP-5MS (Agilent), 5% phenyl-95% methylsiloxane capillary column (30 m x 0.25 mm i.d., 0.25 µm film). Helium was used as the carrier gas at a flow rate of 1.1 ml/min. The oven temperature, initially at 40 °C for 1.5 min, was ramped up to 130 °C at a rate of 5 °C/min. Following this, the temperature was ramped up to 180 °C at 10 °C/min and held there for 10 min. Subsequently, the temperature was raised to 220 °C at a rate of 20 °C/min and held there for 15 min. In the final step, the temperature was raised to 300 °C at a rate of 10 °C/min and held there for 18 min.

Mass spectrometric analysis was performed using an Agilent 5975C VL MSD quadrupole in EI+ mode with an ionization energy of 70 eV. The ion source temperature was set to 230 °C and the quadrupole was set to 290 °C. Identification of compounds was performed in scan mode with a range of 45–650 amu.

The identified compounds were confirmed against authentic standard (provided by Sigma-Aldrich) mass spectra and retention times. Controls performed using mixtures of standards were used to identify possible derivatization artifacts and to avoid ambiguities. Those peaks with less than a 90% match probability in the database and/or tentatively or ambiguously identified were considered unidentified and are not discussed in this paper.

3. Results and Discussion 3.1. Oxidoreductase Activity of Pyrrhotite In a prebiotic scenario, the synthesis of pyruvic acid is of great interest. Many studies have focused on the role of this keto acid in the synthesis of other organic compounds with metabolic relevance, such as acetic acid, methylsuccinic acid and other cyclized compounds [23], oxalacetic acid, acetoacetic acid, fumaric acid, and succinic acid [24]. In the first steps toward the formation of the protometabolic system prior to the origin of living biochemistry, the redox reactions catalyzed by transition metals, such as those in the iron-sulfur minerals, were necessary, as is exhibited by the conserved organometallic enzymes of the most primitive metabolic pathways for carbon fixation and energy generation. This supports the iron-sulfur world hypothesis, wherein the formation of pyrite from pyrrhotite may have provided enough reducing power to catalyze these fundamental redox reactions [20,21]. However, the pyrrhotite-pyrite transformation coupled with redox reaction is not a sufficient condition for the building of the iron-sulfur world. As Schoonen et al. pointed out [25], the pyrrhotite-pyrite couple is not capable of reducing CO2, although it is termodinamically favorable, due to the energetically unfavorable electron transfer from the pyrrhotite valence band to the LUMO (lowest unoccupied molecular orbital) of CO2, especially at higher temperatures. In consequence, to evaluate the pyrrhotite system in the context of iron-sulfur world hypothesis, we studied its redox properties and design an experimental system that overcome the limitations of the FeS/FeS2 couple.

In our experiments, the initial goal was to mimic the oxidoreductase activity of the fermentative enzyme lactate dehydrogenase (LDH) and to connect the pyruvate-lactate redox reaction to the FeS/pyrite system (1). This reaction is essential for explaining the behavior of lactate-pyruvate in sulfur rich mediums.

Pyruvate + FeS + H2S → Lactate + FeS2 + H2

(1)

Life 2013, 3

508

In a biochemical system, the LDH enzyme catalyzes the redox interconversion between pyruvate and lactate. In bacteria, there are two types of LDH: NADH-dependent and NADH-independent. In the first group, the pyruvate resulting from the glycolysis is reduced to lactate, regenerating the nicotinamide adenine dinucleotide (NAD+) from the reduced nicotinamide adenine dinucleotide (NADH) produced in the earlier step. In the second group, which mostly consists of anaerobic bacteria, the LDH does not need the cofactor NADH, and the lactate is used as a carbon source in the subsequent steps of metabolism [26]. One of these non-NADH dependent LDH is the lactate cyctochrome c reductase, which uses a Fe(III)-porphyrin complex as an electron acceptor to yield pyruvic acid and Fe(II)-porphyrin.

To test the activity of the FeS/H2S system as a redox catalyst, we performed two different sets of experiments. First, we used pyruvic acid as a reagent to see whether this system can reduce it to lactic acid under different pH conditions. All of the reactions were carried out at 130 °C for 5 h in an anaerobic atmosphere. Figure 4 shows the percentage of lactic acid synthesis related to the initial amount of pyruvic acid present in the mixture.

If we look at the third column in Figure 4, we note that pyrrhotite per se is not able to promote the synthesis of lactic acid. However, when it is mixed with H2S at either an acidic or basic pH, there is a large reduction in pyruvic acid, greater than 50%. These results differ from those of Wang et al. [16], who found that the in situ precipitation of FeS reduced the pyruvic acid and that its mixture with H2S decreased the yield of the reaction.

Figure 4. Lactic acid synthesis by pyruvic acid reduction coupled to iron sulfide/sulfur system. Role of pH and sulfur oxidation state.

Pyruvic acid

+

+

+

+

+

+

+

+

S0

-

+

-

+

-

+

-

+

Pyrrhotite

-

-

+ -

+ -

+

+

+ +

+ +

SH2

The highest rate of lactic acid production (100%) is obtained by mixing elemental sulfur (Sº) and H2S. If we compare these results with mixtures 2 and 4, where these reagents were used separately, we can see that the activity of the Sº and the H2S might be coupled, thereby enhancing the reaction. It is interesting to note that in mixture 2, Sº can reduce the pyruvic acid mainly at an acidic pH. Wang et al. [16] determined that Sº, at temperatures higher than 113 °C and at neutral pH, acts as an oxidant.

Life 2013, 3

509

Nevertheless, according to our results, Sº is not able to oxidize lactic acid (Figure 5), but it is a reducing agent at lower pH. We performed four reactions using pyrrhotite to clarify its role in the pyruvic acid reduction. The reactions correspond to conditions 3, 4, 7, and 8. Comparing them, we note that the highest yields of lactic synthesis are obtained in pyrrhotite/H2S. In this case, the amount of lactic synthesis is higher than in reaction 5, where only H2S was added. If the reaction medium contains pyrrhotite with Sº and H2S, the rate of pyruvic reduction is decreased compared with reaction 6. However, lactic acid synthesis in reaction 4, pyrrhotite + Sº, is enhanced at neutral and basic pH compared with reaction 2.

Overall, we observe that in the presence of pyrrhotite at an acidic pH, the addition of Sº to the mixture decreases the rate of pyruvic acid reduction, though the rate is enhanced at neutral and basic pH. Taken together, our results demonstrate that the pyrrhotite is able to reduce pyruvic acid into lactic acid only in the presence of Sº and/or H2S. This result is important for explaining the reaction product in Section 3.2.

Figure 5. Pyruvic acid synthesis by lactic acid oxidation coupled with iron sulfide/sulfur system. Role of pH and sulfur oxidation state.

Lactic acid

+

+

+

+

+

+

+

+

S0

-

+

-

+

-

+

-

+

Pyrrhotite

-

-

+ -

+ -

+

+

+ +

+ +

SH2

In the second set of experiments, we tested the reverse reaction, that is, the oxidation of lactic acid into pyruvic acid in sulfur rich systems. Figure 5 shows the percentage of pyruvic acid synthesis related to the initial amount of lactic acid. First, we note that the oxidation rates are lower than the reduction rates. In addition, the oxidation takes place only in the presence of pyrrhotite and Sº (reaction 4 and 8). There is no detectable pyruvic acid synthesis at reaction 7 (pyrrhotite + H2S); however, in the presence of elemental sulfur (reaction 8), the presence of H2S enhances the lactic acid oxidation compared with reaction 4 (pyrrhotite + Sº). In this case, our results are in agreement with those of Wang et al. [16] and support the idea of a coupled reaction system between FeS/Sº/H2S, mainly at acidic pH, and with a significant formation of pyrite, which is identified by XRD analysis of the solid material after the reaction (Figure 6). According to the results, the presence of H2S decreases the rate of oxidation at basic pH compared to acidic pH., The hypothesis used to explain this pH dependence is that the protonated state of the lactic acid affects the reaction. However, in our experiments, when we

Life 2013, 3

510

combined Sº with pyrrhotite, the effect of pH was the opposite of that in reaction 8; thus, it seems more reasonable to conclude that H2S is the factor causing the pH-dependent behavior of the reaction.

In fact, at pH 2 and 5, the free sulfide is in the form of H2S, whereas at pH 8, it is in SH− form. If we consider that there is a coupled reaction between FeS/Sº/H2S, the form of all the components under different pH conditions will affect the overall reaction. Additionally, the lactic acid could form stable complexes with iron, e.g., Fe(Lac)+ and Fe(Lac)2. The formation of iron-lactate complexes is an important factor in the explanation of the differences in the reactions. Lactic acid promotes iron mobilization from the mineral; simulations performed using Geochemist’s Workbench showed that the formation of Fe(III) lactate complexes is associated with pyrrhotite transformation in pyrite in the presence of Sº. This could explain the lack of oxidation to pyruvate.

Considering the role of elemental sulfur in the oxidation of lactic acid, it is interesting to highlight that in the presence of pyrrhotite, it promotes the oxidation at low pH, which is in contrast with the results obtained for the reduction, where elemental sulfur decreases the yield of lactic synthesis.

Thus, these results suggest that at 130 °C the Sº acts as an oxidant when the pH is acidic and as a reductant at neutral and basic pH. Figure 6. Powder X-ray diffraction of pyrrhotite ore after experiments. The identified phases include pyrite and siderite.

Py

Intensity

Py

Po

Po

Po Sd

20

Po

40

60

2- (deg) CuKα

Overall, our results demonstrate that the pyrrhotite/Sº/H2S system has the ability to mimic the oxidoreductase activity and that it drives lactate-pyruvate redox chemistry in low temperature hydrothermal systems. The smaller yield of lactic synthesis in the presence of pyrrhotite, Sº and H2S in comparison with the same mixture without pyrrhotite (Figure 4) might be explained by the fact that in the presence of pyrrhotite the coupling reaction system is promoted at the same time as the reduction, and the oxidation is pH-dependent.

3.2. Pyrrhotite/Iron Metal in Thioester Reductive Carboxylation: a Ferredoxin Mimic?

Metallic iron is an extremely rare mineral in the modern Earth’s crust, mainly due to its instability in oxic conditions, but it may have been abundant in the early Earth before the rise of an oxygen-rich atmosphere. Iron is a good source of electrons, through its oxidation to Fe2+ (standard reduction

Life 2013, 3

511

potential, Eº = −440 mV). We constructed an electrochemical cell using a powder iron metal paste in water-HSO4− at pH 5 as the anode, separated by a clay barrier from the cathode, which was formed by pyrrhotite containing 1 mmol of ethyl thioacetate and wetted with a NH4+ HCO3− 50 mM solution (pH 9) containing 0.15 mmol of Na2S; we used the cell to test whether iron metal could be an additional source of reducing power and a supply of mobile iron under mild conditions. The system was connected to a power supply at 1.1 V under a nitrogen atmosphere in the circuit depicted in Figure 3; the current began at 17.4 µA, rose to a maximum of 23 µA and 1.87 V after 6 h and then decayed to a constant current of 10.5 µA. The organic solutes were analyzed after three days of standing in anoxic conditions at room temperature; the analysis shows a significant quantity of lactic acid (Figure 7) with an estimated yield of 6.5% of the added ethyl thioacetate. Pyruvic acid was also detected, as well as glycolic acid and glycine. Aliphatic amines and unidentified organic compounds, including sulfur-containing molecules, were formed.

Figure 7. GC/MS chromatogram of the TMS derivatives of identified products obtained by carboxylation of ethylthioacetate coupled to the iron/pyrrhotite/sulfide electrochemical system.

Lactic acid

O

Relative abundance

OH O Pyruvic acid

Glycolic acid

12

15

20

24

Retention time (min)

No carbon chains greater than C3 were detected. The X-ray diffraction of pyrrhotite conducted after the experiment shows peaks of remaining pyrrhotite, pyrite, siderite and unidentified iron oxides, possibly wustite, ferrihydrite, and “green rust” (iron hydroxycarbonate). A control experiment performed without an external voltage source showed a significantly lower yield of lactic acid (less than 1%). An additional control experiment without thioester does not show formation of C2 or C3 organic compounds, suggesting that lactate/pyruvate are formed after carboxylation of thioacetate ester and that the formation of organic compounds directly from carbon dioxide is not possible under these conditions.

The voltage source generates an electrochemical gradient and can supply additional electrons to the pyrrhotite-S2−/pyrite redox couple, which favors CO2 reduction and is not an efficient electron acceptor due to its large overpotential (−2.22 V vs. SCE). The presence of iron-sulfur clusters can significantly reduce the CO2 reduction overpotential, promoting carbon fixation [20]. Under our chosen conditions, the anodic oxidation of iron metal could supply Fe2+ cations to the system, with

Life 2013, 3

512

subsequent formation of FeS(aq), which constitutes the building blocks of crystalline forms of ferrous sulfides [27]. The soluble iron-sulfur complexes that constitute the aqueous form of FeS could have similar properties to the iron-sulfur clusters found in the metalloprotein ferredoxin, which may mediate the electron transfer reaction for the reductive carboxylation of thioacetic acid [28]. The formation of these protoferredoxins [8] could be favored by the formation of organic sulfur species, as ethanethiol, derived from ethyl thioacetate, which could form complexes with the formula [Fe4S4(SC2H5)4]3−.

Under these conditions, the external source and the iron could constitute the electron donors necessary to keep the iron-sulfur complexes in a reduced state and to overcome the reduction potential of the pyruvate synthesis from CO2 and thioester (Eº = −500 mV). Hence, our experiment can be considered to be a biomimetic pathway to the biosynthesis of pyruvate from acetyl-CoA, promoted by PFOR and reduced ferredoxin [17]. The strongly favored presence of lactic acid, which is the major organic product of the experiment, can be explained by the formation of stable Fe-lactic acid complexes, which stabilize both the lactic acid and the iron in solution.

The role of iron-sulfur clusters in the reductive carboxylation were demonstrated in the seminal work of Nakajima et al. [29], in which an alpha-keto acid is formed under mild conditions as an intermediate in the synthesis of phenylalanine from n-octyl-phenylthioacetate, catalyzed by a synthetic iron-sulfur cluster that models ferredoxin. Although the Nakajima reaction helped to conceptualize the possibility of iron-sulfur clusters as non-enzymatic catalysts and its potential role during life origin, the reaction is not geochemically extrapolable, as it has been performed in non-aqueous conditions and with hydrosulfite as electron donor (Eº = −0.660V). The electron donor role, as suggested in the Iron-Sulfur world theory, could be assumed by the pyrite formation from pyrrhotite in a geochemical environment [30]. In fact, the control experiment without an external voltage source shows a significantly lower yield in the formation of lactic acid, suggesting that a low potential electron donor could be necessary for the process. To test this possibility, we performed an experiment using the same electrochemical cell design but without using an external voltage source and adding 1 mmol of hydroquinone (Eº = −0.699V). Hydroquinone can act as analog of the biological ubiquinol and can perform electron transfer reactions on the surface of minerals [31]. The model biochemical reaction that motivates the selection of hydroquinone as an electron donor is the formation of pyruvate by direct carboxylation of acetic acid, promoted by (quinone) pyruvate dehydrogenase. The presence of hydroquinone promotes the synthesis of lactic acid, increasing the yield to 10.5% and suggesting that electrons can be transferred through iron sulfur clusters or surfaces, similar to the ubiquinol/iron-sulfur system in biochemistry (Figure 8).

Our experiments show that iron metal is essential to the reaction, possibly as an electron source and soluble iron source. However, what is the role of pyrrhotite? A control experiment performed using silica sand instead of pyrrhotite does not show the formation of C3 derivatives, which indicates that pyrrhotite plays an essential role in the formation of pyruvate/lactate. Although the synthesis of pyruvate, promoted by FeS, has been reported by G.D. Cody et al. [32], the high CO pressure experimental conditions suggest a different mechanism. Under our mild conditions, two facts could help to explain the role of pyrrhotite. (i) The pyrrhotite structure is complex; it is a non-stoichiometric, iron defective sulfide, with surface vacancies and distortions [33]. The surface chemistry of pyrrhotite, although not widely studied, should be different than of troilite (its stoichiometric equivalent) and the synthetic FeS phases usually used in prebiotic chemistry. The electronic properties of pyrrhotite

Life 2013, 3

513

particles could allow them to be used as electron transfer conduits to the species in solution or to newly formed membraneous precipitates, in a electrochemical induced pathway similar to the photoelectrochemical formation of alpha-ketoglutaric acid from pyruvate, catalyzed by sphalerite particles [34]. (ii) The formation of pyrite in the presence of hydrogen sulfide, which, in synergy with the formation of soluble protoferredoxin clusters, could supply reducing power in the form of molecular hydrogen. This pyrite-pulled reaction connects the prebiotic chemistry with the biochemistry; this is an equivalent mechanism to that in which ferredoxin is maintained in a reduced state by hydrogen in chemolithotrophs [12]. Although the system shows considerable chemical complexity, and further work is necessary to explain the products obtained, the common key feature is the disequilibrium. We established an electrochemical reactor, with dissipation of a potential and pH gradient, together with the formation of new mineral precipitates, following the theory postulated by M. Russell [35,36]. The main difference is that the Russell theory proposed an exhalative submarine model that involves the mixture of two fluids at different temperature and pH: The Hadean ocean water and the hydrothermal fluid. The formation of membranous iron sulfide precipitate in the fluid interface promoted an environment where accumulation of clusters and organics and the dissipation of electrochemical gradients through membrane, promotes the protometabolic reactions. We propose in this experiment a fluid-rock interaction model in which iron metal and iron sulfur minerals promoted the formation of new species and the adequate environment for the carbon fixation, by formation of new iron solid species and soluble clusters. The presence of previously formed organic precursors (such as thioesters or alkyl thiols, in the reported experiment) is necessary to the increase of organic complexity in the mild conditions tested.

Figure 8. GC/MS chromatogram of the TMS derivatives of products obtained by carboxylation of ethylthioacetate coupled to the iron/pyrrhotite/sulfide electrochemical system, with hydroquinone as an additional electron donor. \ O OH OH OH

Lactic acid HO

Hydroquinone

Relative abundance

Pyruvic acid O HO

OH

Glycolic acid O H2N

Glioxylic acid

12

OH

Glycine

20

15

Retention time (min)

An interesting feature of the reaction is the formation of glycolic acid and glycine. The reductive amination of alpha-keto acids in the presence of ammonium to yield amino acids has been reported

Life 2013, 3

514

under mild conditions in the presence of Fe2+ [37]. In our results, we did not identify any amino acids larger than glycine. The glycine could be synthesized from glyoxylic acid by transamination using other amino acids as nitrogen donors [38] and by reductive amination in the presence of strong reducing agents [39]. The possible presence of glyoxylic acid as an intermediate under our conditions connects the pyrrhotite/electrochemical reduction with the synthesis of lactate, which is promoted by the highly reducing photo-generated conduction band electrons in the zinc sulfide mineral sphalerite [40].

This photo-electrochemical synthesis leads to the direct formation of lactate by a reaction between carbon dioxide and glyoxylic acid. It is possible that the electronic properties of pyrrhotite favor a similar pathway in the electrochemical cell (Figure 9), although oxalic acid was not identified and there was no evidence of thioacetate independent carbon fixation. It is interesting to note the lack of organic acids greater than C3, which suggests that there are no further reactions involving pyruvic acid or that reduction to lactic acid or degradation (for example, formation of oxaloacetate and degradation to pyruvate) are kinetically favored.

Figure 9. Proposed routes to the synthesis of lactic acid by carboxylation of an alkyl thioacetate ester. The synthesis of pyruvic acid could be regarded as a prebiotic analog of reductive carboxylation of acetyl CoA promoted by PFOR, with iron-sulfur clusters of ferredoxin as electron donors.

In summary (Figure 9), the formation of lactic acid under our experimental conditions can be explained through pyruvate synthesis by the reductive carboxylation of thioacetate ester, followed by a reduction to lactic acid (see Section 3.1). Additionally, the presence of glycolic acid and glycine could suggest alternative pathways, including the glyoxylate intermediary. Further experiments are necessary to explore the potential of electrochemical gradients in combination with sulfide minerals and soluble iron sulfide complexes in the origin of a biochemical system. The identification of unidentified aminoand thio- derivatives found in our experiments could help to explain the complex mechanism implicated in the biomimetic reductive carboxylation of a thioester. Aside from establishing the precise mechanism that this system uses for the formation of C3 acids from C2 precursors and carbon dioxide, our experiments show that iron sulfides in a non-equilibrium environment, characterized by the

Life 2013, 3

515

presence of electrochemical gradients and soluble and solid iron species in diverse oxidation states, could support biologically useful reactions. This supports the idea that transition metal sulfides play a role in the development of protometabolism.

4. Conclusions The natural iron sulfide pyrrhotite, in a soluble sulfur-rich environment and in the presence of newly formed mineral precipitates and soluble iron-sulfur clusters (whose formation was induced by iron metal anodic oxidation), can promote the reductive carboxylation of simple thioacetic acid esters to form pyruvate/lactate under mild conditions. The reaction could be regarded as an abiotic analog of the pyruvate synthesis promoted by pyruvate ferredoxin oxidoreductase and favored by the presence of low potential electron donors, such as hydroquinone, which suggests that the origin of ancient organic cofactors boosted the emergence of simple protometabolic systems. Additionally, the complexity of the resultant system does not exclude alternative pathways, such as reductive amination or alternative carbon fixation mechanisms. The results are consistent with the theories that place the iron sulfide species at the center of the geochemistry-biochemistry transition and show that the diversity of oxidation states of iron and sulfur and the morphologies of the solid iron phases, in a scenario that favors the electrochemical coupling by the proximity of diverse mineral forms, is fundamental for carbon fixation to the C3 compounds lactate/pyruvate from the C2 precursors (thioacetate or ethylthiol). This suggests that the dissipation of electrochemical gradients and the electronic properties of pyrrhotite and other iron sulfur species could promote the emergence of a thioester based biochemical system.

Acknowledgments We thank Frederic Varela Balcells and the friends of FMF forum, who kindly supplied the mineral samples used in this work, to Sabino Veintemillas Verdaguer for his support in the laboratory and to Isabel Robledo for her support in the manuscript edition. First author thank the financial support of the JAE-pre program for PhD students of the Consejo Superior de Investigaciones Científicas (CSIC, Spain). This work has been performed without other financial support and the corresponding author would like to thank the Spanish SEPE (Servicio Público de Empleo Estatal) for its unemployment compensation, which made possible the publication of this paper.

Conflicts of Interest The authors declare no conflict of interest.

References 1.

2.

3.

Holm, N.G., Ed. Marine Hydrothermal Systems and The Origin of Life; Springer: Berlin, Germany, 1992; pp. 1–242.

Harvey, R.B. Enzymes of thermal algae. Science 1924, 50, 481–482.

Martin, W.; Baross, J.; Kelley, D.; Russell, M.J. Hydrothermal vents and the origin of life.

Nat. Rev. Microbiol. 2008, 6, 805–814.

Life 2013, 3 4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

516

Cody, G.D.; Boctor, N.Z.; Brandes, J.A.; Filley, T.R.; Hazen, R.M.; Yoder, H.S. Assaying the catalytic potential of transition metal sulfides for abiotic carbon fixation. Geochim. Cosmochim. Acta 2004, 68, 2185–2196.

Wächtershäuser, G. Before enzymes and templates: theory of surface metabolism. Microbiol. Rev. 1988, 52, 452–484.

Huber, C.; Wächtershäuser, G. Activated acetic acid by carbon fixation on (Fe,Ni)S under primordial conditions. Science 1997, 276, 245–247.

Russell, M.J.; Hall, A.J.; Turner, D. In vitro growth of iron sulphide chimneys: possible culture chambers for origin-of-life experiments. Terra Nova 1989, 1, 238–241.

Russell, M.J. The emergence of life from iron monosulphide bubbles at a submarine hydrothermal redox and pH front. J. Geol. Soc. London 1997, 154, 377–402.

Martin, W.; Russell, M.J. On the origin of biochemistry at an alkaline hydrothermal vent.

Philos. T. Roy. Soc. 2007, 362, 1887–1925.

Fuchs, G. Alternative Pathways of Carbon Dioxide Fixation: Insights into the Early Evolution of Life? Ann. Rev. Microbiol. 2011, 65, 631–658.

Eck, R.V.; Dayhoff, M.O. Evolution of the structure of ferredoxin based on living relics of primitive amino acid sequences. Science 1966, 152, 363–366.

Beinert, H. Iron-sulfur proteins: ancient structures, still full of surprises. J. Boil. Inorg. Chem.

2000, 5, 2–15.

Russell, M.J.; Martin, W. The rocky roots of the acetyl-CoA pathway. Trends Biochem. Sci. 2004 29, 358–363.

Holm, R.H. Electron Transfer: Iron–Sulfur Clusters. In Comprehensive Coordination Chemistry II; McCleverty, J.A., Meyer, T.J., Eds.; Pergamon: Oxford, UK, 2003; pp. 61–90.

Russell, M.J.; Hall, A.J. The onset and early evolution of life. In Evolution of Early Earth’s Atmosphere, Hydrosphere and Biosphere-Constraints from Ore Deposits; Kesler, S.E., Ohmoto, H., Eds.; Geological Society of America: Boulder, CO, USA, 2006; pp. 1–32.

Wang, W.; Yang, B.; Qu, Y.; Liu, X.; Su, W. FeS/S/FeS2 redox system and its oxido reductase-like chemistry in the iron-sulfur world. Astrobiology 2011, 5, 471–476.

Furdui, C.; Ragsdale, S.W. The role of pyruvate ferredoxin oxidoreductase in pyruvate synthesis during autotrophic growth by the Wood-Ljungdahl pathway. J. Biol. Chem. 2000, 275, 28494–28499.

Ragsdale, S.W.; Pierce, E. Acetogenesis and the Wood-Ljundahl pathway to CO2 fixation.

Biochim. Biophys. Acta 2008, 1784, 18973–18980.

De Duve, C. Blueprint for a Cell: The Nature and Origin of Life; Neil Patterson Publishers: Burlington, NC, USA, 1991.

Cody, G.D. Transition metal sulfides and the origins of metabolism. Ann. Rev. Earth Planet. Sci. 2004, 32, 569–599.

Wächtershäuser, G. Evolution of the first metabolic cycles. Proc. Natl. Acad. Sci. USA 1990, 87, 200–204.

For a detailed overview of the Gualba deposit (In Spanish). Available online: http:// www.foro-minerales.com/forum/viewtopic.php?p=82307&highlight=gualba#82307 (accessed on 3 June 2013).

Life 2013, 3

517

23. Hazen, M.R.; Deamer, W.D. Hydrothermal Reactions of Pyruvic acid: Synthesis, Selection, and Self-Assembly of Amphiphilic Molecules. Origins Life Evol. B 2007, 37, 143–152.

24. Cooper, G.; Reed, C.; Nguyen, D.; Carter, M.; Wang, Y. Detection and formation scenario of citric acid, pyruvic acid, and other possible metabolism precursors in carbonaceous meteorites.

Porc. Natl. Acad. Sci. USA 2011, 108, 14015–14020.

25. Schoonen, M.A.A.; Xu, Y.; Bebie, J. Energetics and kinetics of the prebiotic synthesis of simple organic acids and amino acids with the FeS-H2S/FeS2 redox couple as reductant. Origins Life Evol. B 1999, 29, 5–32.

26. Garvie, E.I. Bacterial Lactate Dehydrogenases. Microbiol. Rev. 1980, 44, 106–139.

27. Rickard, D.; Luther, G.W. Chemistry of Iron Sulfides. Chem. Rev. 2007, 107, 514–562.

28. Bonomi, F.; Werth, M.T.; Kurtz, D.M. Assembly of FenSn(SR)2- (n = 2,4) in aqueous media from iron salts, thiols and sulfur, sulfide, thiosulfide plus rhodonase. Inorg. Chem. 1985, 24, 4331–4335.

29. Nakajima, T.; Yabushita, Y.; Tabushi, I. Amino acid synthesis through biogenetic-type CO2 fixation.

Nature 1975, 256, 60–61.

30. Drobner, E.; Huber, H.; Wächtershäuser, G.; Rose, D.; Stetter, K.O. Pyrite formation linked with hydrogen evolution under anaerobic conditions. Nature 1990, 346, 742–744.

31. Kung, K.; Mcbride, M.B. Electron Transfer Processes Between hydroquinone and iron oxides.

Clay. Clay Miner. 1988, 36, 303–309.

32. Cody, G.D.; Boctor, N.Z.; Filley, T.R.; Hazen, R.M.; Scott, J.H.; Sharma, A.; Yoder, H.S., Jr. Primordial car- bonylated iron-sulfur compounds and the synthesis of pyruvate. Science 2000, 289, 1337–1340.

33. Rosso, K.M.; Vaughan, D.J. Sulfide mineral surfaces. Rev. Miner. Geochem. 2006, 61, 505–556.

34. Guzman, M.I.; Martin, S.T. Prebiotic metabolism: production by mineral photoelectrochemistry of alpha-ketocarboxylic acids in the reductive tricarboxylic acid cycle. Astrobiology 2009, 9, 833–842.

35. Russell, M.J. The alkaline solution to the emergence of life: Energy, entropy and early evolution.

Acta biotheor. 2007, 55, 133–179.

36. Mielke, R.E.; Robinson, K.J.; White, L.M.; Mcglynn, S.E.; Mceachern, K.; Bhartia, R.; Kanik, I.; Russell, M.J. Iron-Sulfide-Bearing Chimneys as Potential Catalytic Energy Traps at Life’s Emergence. Astrobiology 2011, 11,933–950.

37. Huber, C.; Wächtershäuser, G. Primordial reductive amination revisited. Tetrahedron Lett. 2003, 44, 1695–1697.

38. Nakada, H.I.; Weinhouse, S. Non-enzymatic transamination with glyoxylic acid and various amino acids. J. Biol. Chem. 1953, 204, 831–836.

39. White, R.H. A simple synthesis of (RS)-[2–2H] glycine by the reductive amination of glyoxylic acid. J. Labelled Compd. Rad. 1983, 20, 787–790.

40. Guzman, M.I.; Martin, S.T. Photo-production of lactate from glyoxylate: how minerals can facilitate energy storage in a prebiotic world. Chem. Commun. 2010, 46, 2265–2267. © 2013 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

1309

Thermal Wet Decomposition of Prussian Blue: Implications for Prebiotic Chemistry

by Marta Ruiz-Bermejo* a ), Celia Rogero a ), Ce´sar Menor-Salva´n a ), Susana Osuna-Esteban a ), ´ ngel Mart n-Gago a ) b ), and Sabino Veintemillas-Verdaguer a ) b ) Jose´ A a ) Centro de Astrobiolog a (Consejo Superior de Investigaciones Cient ficas-Instituto Nacional de Te´cnica Aeroespacial (CSIC-INTA)), Carretera Torrejo´n-Ajalvir, Km. 4.2, ES-28850 Torrejo´n de Ardoz, Madrid (phone: þ 34 91 520 6402/6458; fax: þ 34 91 5206410; e-mail: ruizbm@inta.es) b ) Instituto de Ciencias de Materiales de Madrid (CSIC), C/ Sor Juana Ine´s de la Cruz, 3, Cantoblanco, ES-28049 Madrid

The complex salt named Prussian Blue, Fe4[Fe(CN)6 ]3 · 15 H2O, can release cyanide at pH > 10. From the point of view of the origin of life, this fact is of interest, since the oligomers of HCN, formed in the presence of ammonium or amines, leads to a variety of biomolecules. In this work, for the first time, the thermal wet decomposition of Prussian Blue was studied. To establish the influence of temperature and reaction time on the ability of Prussian Blue to release cyanide and to subsequently generate other compounds, suspensions of Prussian Blue were heated at temperatures from room temperature to 1508 at pH 12 in NH3 environment for several days. The NH3 wet decomposition of Prussian Blue generated hematite, a-Fe2O3 , the soluble complex salt (NH4 )4[Fe(CN6 )] · 1.5 H2O, and several organic compounds, the nature and yield of which depend on the experimental conditions. Urea, lactic acid, 5,5dimethylhydantoin, and several amino acids and carboxylic acids were identified by their trimethylsilyl (TMS) derivatives. HCN, cyanogen (C2N2 ), and formamide (HCONH2 ) were detected in the gas phase

by GC/MS analysis.

Introduction. – It has been suggested that HCN could act as an important prebiotic precursor of purines, pyrimidines, and amino acids, as well as of other compounds such as oxalic acid and guanidine [1]. In the presence of a base media, such as amine or ammonia [2], concentrated solutions of HCN ( < 0.01m) can polymerize and produce nucleic acid bases and amino acids, whereas, in dilute solutions, hydrolysis becomes dominant [3]. Taking into account the production rates of HCN in the primitive atmosphere and the experimental hydrolysis rates, the steady-state concentration of HCN in the primitive ocean could be in the range of 4 10 6 – 2 10 8 m at pH 8 between 08 and 258 [4]. Miyakawa et al. [5] estimated this concentration to be ca. 2 10 6 m at pH 8 and 08. Therefore, if HCN polymerization was actually important for the production of the first and essential biomolecules, there must have been routes by which diluted HCN solutions were efficiently concentrated. Since HCN is more volatile than H2O, it cannot be concentrated by evaporation, if the pH is lower than the pKa of HCN (9.2 at 258). An alternative and plausible mechanism is, therefore, eutectic freezing. Miller and co-workers achieved the synthesis of several purine and pyrimidine bases from a frozen NH4CN solution (HCN 0.15m plus NH3 0.1m) that had been held at 788 for 27 years [5].

2009 Verlag Helvetica Chimica Acta AG, Z rich

1310

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

Another route for the concentration of cyanide was proposed by Arrhenius et al.

[6]. Cyanide may be concentrated by reaction with Fe2 þ to form ferrocyanide, with the subsequent precipitation as an insoluble complex salt called Prussian Blue, Fe4[Fe(CN)6 ]3 · 15 H2O (ferric hexacyanoferrate(II), crystal structure reported by Buser et al. [7]). Indeed, ferrocyanide anion ([Fe(CN)6 ]4 ) has been mentioned as a possible abundant component of the primitive ocean [8], and ferrocyanides and ferricyanides have been suggested as compounds of prebiotic interest [9]. In a recent work, we demonstrated that Prussian Blue is easily formed in spark-discharge experiments using saline aerosols of ferrous salts [10]. Therefore, the formation of Prussian Blue offers a mechanism for the concentration of cyanide in the form of the ferrocyanide ion.

Subsequent reactions of the Prussian Blue might lead to the production of several compounds of interest in prebiotic chemistry. Thus, the ferrocyanide moiety can be photochemically oxidized in the near-UV range by H2O, liberating HCN, with a- and gFeOOH as oxidation products [11]. In this context, Tiwari [12], and Tiwari and Sharma [13] established the formation of amino acids from soluble K4Fe(CN)6 and HCHO in aqueous solution using UV-light irradiation. Additionally, it is known that the thermal decomposition of Prussian Blue in vacuum does not lead to significant changes in the coordination environment of Fe ions at temperatures below 1758 [14], and that thermally induced oxidative decomposition of Prussian Blue produces amorphous Fe2O3 , and b- and g-Fe2O3 , depending on the temperature (2508 and 3508, resp. [15] ).

The thermal decomposition in vacuum of complex type KMII[Co(CN)6 ] · H2O provides cyanogen gas (CN)2 by diffusion of free radical CN toward the surface of the solid and produces amorphous metal powder as a final decomposition product [16]. Additionally, the decomposition of Fe chelates in alkaline media (pH > 12) at 1008 to produce hematite, a-Fe2O3 , has been studied [17]. However, as far as we know, no work has been reported on the wet decomposition of Prussian Blue. At 4 < pH < 10, Prussian Blue is a stable and insoluble compound; however, outside that interval, it could undergo solubilization, releasing cyanide (these values were calculated on the data reported by Meeussen et al. [18]).

The hypothesis is that, first, the Prussian Blue could act as a reservoir of HCN, and second, triggered by pH and heating in local environments (e.g., volcanic eruption with the subsequent production of ammonium in high concentrations [19]), it might lead to the production of compounds of interest from the point of view of the origin of life.

Thus, the goal of the present work is to explore the decomposition products, inorganic and organic, of the Prussian Blue in alkaline media. For this purpose, suspensions of Prussian Blue were heated at different temperatures (from room temperature to 1508) in NH3 solutions (pH 12), and the collected products were analyzed by a combination of spectroscopic and analytical techniques. The detection of some compounds might have implications for prebiotic chemistry.

Results. – Thermal Wet Decomposition of Prussian Blue. To test the above hypothesis about the relevance of Prussian Blue in prebiotic chemistry, suspensions of this salt were heated in NH3 medium at different temperatures for 24 h, 48 h, and 1 week under inert atmosphere (see Table 1 for a summary of the experiments). In all experiments, after the reaction time a yellow solution and an insoluble solid were observed. Both phases were separated by centrifugation. The supernatants were

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

1311

Table 1. Reaction Conditions and Organic Compounds Identified as MSTFA Derivatives from the Heating of Prussian Blue in Ammonia Medium (pH 12) Entry

1 a) 2 3 4 5 6 7 8 9 10 11 12 13 a

Temp. [8]

r.t.

70 70 70 100 100 100 130 130 130 150 150 150

Time [h]

168 24 48 168 24 48 168 24 48 168 24 48 168

MSTFA Derivatives [mmol] Urea

Lactic acid

Dimethylhydantoin

Malic acid

Glycine

2-Aminoisobutyric acid

0.013 0.042 0.045 0.109 0.010 0.013 0.643 0.158 0.158 0.014 0.028 0.028

0.042 0.131 0.0203 0.018 0.001 0.019 0.020 0.002 0.169 – – –

– – – 0.408 0.023 – 0.210 – – – – –

– – – 0.008 – – – – – – – –

– – – – – 0.084 – – – 0.047 – –

– – – – – – 0.325 – – – – –

) Only the gas phase was analyzed.

chromatographed on silica gel, and, in all samples, two fractions were collected, a white solid (solid 1) for the first fraction and a yellow solid (solid 2) for the second one. Fig. 1 depicts the general scheme for the workup of the different samples produced.

Fig. 1. General scheme for the workup of the samples

1312

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

Study of the Supernatants. Solids 1 and 2. Both solids 1 and 2 were characterized by spectroscopic and analytical methods. Independently from the preparation conditions, solid 1 was found to be a mixture of H2O-soluble organic compounds, and solid 2 was mainly constituted by the complex salt (NH4 )4[Fe(CN)6 ] · 1.5 H2O (crystal structure reported by Morosin [20]). Based on the results of elemental analysis, the conversion of Prussian Blue into soluble organic compounds was ca. 0.2 – 0.7%, and its conversion into (NH4 )4[Fe(CN)6 ] · 1.5 H2O was ca. 4 – 7% with respect to the amount of carbon present in the starting Prussian Blue.

The IR spectra of solids 1 and 2 showed clear differences with respect to Prussian Blue (see Fig. 2). Solid 1 presents bands related to imine, isonitrile, and carbonyl groups, C¼C bonds, and ammonium cation, NH þ4 . The peak at 1762 cm 1 raised its complexity in the experiments performed at the highest temperature. Therefore, solid 1 seems to be a complex organic solid formed by compounds rich in nitrogen and unsaturations, but not in nitriles. Solid 2 presents secondary bands at 2116 cm 1 associated with N¼C¼O, N¼C¼N, or N C groups (see Table 2 for the details of the assignment of the bands).

Table 2. Characteristic Frequencies of IR Absorption Spectra Obtained from the Solids 1 and 2 Solids 1

Solids 2

3120 3039 2170 2116 2050 2047 2031 1762 1639 1399 1075 960

1399

Vibrational groups identity NH stretch ( NH þ4 ) ¼CH2 stretching (vinylene group) N C N¼C¼O/N¼C¼N/ N C C¼C¼N C N R C¼C¼NH C¼O/C¼N C¼C strech (vinylene group) NH2 deformation (NH þ4 ) C O vibrational mode/C¼C¼ strech (symmetry) CH out-of-plane deformation (vinylene group)

X-Ray photoelectron spectroscopy (XPS) of solid 1 presents an overview that reveals the presence of only C, O, and N. This indicates that solid 1 could be related to the formation of organic molecules. The core-level peak of N 1s (Fig. 3, a) can be decomposed only using one component centered at 402.5 eV, a value compatible with C NH C and O¼C NH C groups and also with the presence of NH þ4 . The XPS overview spectrum of solid 2 contains: Fe, N, O, and C. The ratio N/Fe (N/Fe)xps ¼ 11.66 matched with the empirical formula of (NH4 )4[Fe(CN)6 ] · 1.5 H2O, and it was in good agreement with the XPS spectrum of the reference (NH4 )4[Fe(CN)6 ] · 1.5 H2O standard sample 1). Fig. 3, b, shows the high-resolution Fe 2p and N 1s core-level spectra recorded for solid 2 and includes the comparison with the reference (NH4 )4[Fe(CN)6 ] · 1.5 H2O (dashed lines). In both cases, spectra show peaks at the same positions. The Fe 1)

As a reference, the Prussian Blue and (NH4)4[Fe(CN)6] · 1.5 H2O spectra were registered.

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

1313

Fig. 2. Representative transmission FT-IR spectra. a) Prussian Blue, b) (NH4 )4Fe(CN)6 · 1.5 H2O, c) white solid (H2O-soluble organic compounds from the first fraction), d) yellow solid (second fraction). Main features of (NH4 )4Fe(CN)6 · 1.5 H2O.

1314

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

Fig. 3. a) N 1s Core-level spectra for the solid 1-KBr sample. b) Fe 2p and N 1s core-level spectra for the solid 2-KBr sample. Dots correspond to the experimental data, red line to the fit, and the blue curves correspond to the curve components used for the deconvolution. Inset: Comparison with the reference NH4[Fe(CN)6 ] · 1.5 H2O standard sample.

2p core level presents two peaks (separated by 13 eV) associated with the spin-orbit splitting, Fe2p3/2 and Fe 2p1/2 . Based on the position of Fe 2p3/2 (709.7 eV), Fe-atoms are in a Fe2 þ chemical state, which corresponds to (NH4 )4[Fe(CN)6 ] · 1.5 H2O. For the N 1s core level, two curve components are clearly observed. The first one, centered at 399.1 eV, is related to C N, and the second one, centered at 402.4 eV, is assigned to the NH þ4 of this compound (for the assignment of the binding energy, see Table 3).

Since the spectroscopic and elemental analyses indicate that the solids 1 of each experiment were constituted only by organic matter, GC/MS analyses were performed, revealing formation of several molecules (Fig. 4). For details of the organics formed and their yield, see Table 1.

Study of the Insoluble Solids. The insoluble solids separated by centrifugation were characterized by IR spectroscopy and X-ray diffraction. The insoluble solids of all experiments comprised hematite (a-Fe2O3 ) and unreacted Prussian Blue, and likely other amorphous iron oxides. The portion of unreacted Prussian Blue was 25 – 45%,

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

1315

Table 3. Summary of the Decomposition of the Fe 2p and the N 1s Core-Level Spectra Fe 2p

N 1s

Binding Chemical Binding Chemical state energy [eV] state energy [eV] 709.4 711.0 ( NH4 )4[ Fe(CN)6 ] · 1.5 H2O 709.7

Fe2 þ Fe3 þ Fe2 þ

Solid 1 Solid 2

– Fe2 þ

Prussian Blue

– 709.7

399.4

C N

399.1 402.4 402.5 399.1 402.4

C N ( NH4 ) þ C NH C/O¼C NH C/( NH4 ) þ C N ( NH4 ) þ

depending on the experimental conditions. The detection of hematite is very relevant from the point of view of astrobiology, as we will discuss later.

Study of the Gas Phase. The formation of volatile species and gases was monitored

by GC/MS, as aliquots were collected every 24 h for each temperature of reaction assayed. Additionally, an experiment at room temperature (Table 1, Entry 1) was carried out in order to check the decomposition of Prussian Blue without heating. In all cases, the formation of HCN was observed, achieving the highest concentration at 1308 for a period of 3 d (as measured by the area of the chromatographic peak).

Additionally, in all cases, the maximum concentration of HCN in the gas phases also was observed after 3 d, after which it decreased. Cyanogen, (CN)2 , and formamide, HCONH2 , were also detected in all experiments (Fig. 5).

Discussion. – The experimental evidences reveal the plausible implications of Prussian Blue for the primitive Earth and for prebiotic chemistry. As a first consideration, we would expect that, under the conditions assayed during this work, Prussian Blue would be able to release its entire reservoir of cyanide. If this were the case, the concentration of cyanide should attain 6.40m, sufficiently high to polymerize in NH3 medium. However, the yields for the solids 1 indicate a concentration of cyanide in solution of ca. 0.01m, and the formation of HCN oligomers [21] was not observed. At concentrations above 0.01m, the polymerization of HCN occurs, whereas, below that, hydrolysis takes place. Since our results were obtained on the boundary, a competition between the two processes occurred. Thus, under the conditions assayed, one can achieve the formation of discrete organic molecules, and the presence of HCONH2 in the gas phase is directly related to the hydrolysis of HCN. The best experimental conditions that we have found in order to obtain organic compounds from Prussian Blue were attained at 100 – 1308 for short periods of reaction (see Table 1). This is due to several concurrence factors: i) the decomposition of Prussian Blue is favored at high temperature, ii) HCN solubility decreases with increasing temperature, and iii) the ratio of hydrolysis of HCN to polymerization increases for higher temperatures and long periods of time.

The formation of organic molecules such as urea, 5,5-dimethylhydantoin, and lactic acid from Prussian Blue might be of interest from the point of view of prebiotic chemistry. At high concentrations, urea is considered an important prebiotic reactant in

1316

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 6 (2009)

Fig. 4. Chromatograms showing the non-volatile organic-compound profile found for the solids 1 from experiments shown in Table 1: a) Entry 2, b) Entry 5, c) Entry 8, d) Entry 11

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 6 (2009)

1317

Fig. 4 (cont.)

Fig. 5. Representative GC/MS chromatogram in full-scan mode for the analysis of the gas phases (in this picture, experiment at 708 and 48 h of reaction time, Table 1, Entry 3). a) SIM Mode m/z 52 GC/MS chromatogram, b) SIM mode m/z 45 GC/MS chromatogram.

the synthesis of pyrimidines [22] [23]. Its role as a prebiotic reagent remains doubtful due to its instability at the high concentrations needed to act as a prebiotic intermediate [24]. Urea could be formed directly by oligomerization of HCN or by hydrolysis of

1318

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

(CN)2 [25], and, usually, it is obtained in spark-discharge experiments [26]. Urea was detected in almost all our experiments, demonstrating that Prussian Blue could be an alternative source of this compound under different conditions.

From an astrobiological point of view, hydantoins are also important, since they have been suggested as precursors for the emergence of prebiotic peptides and amino acids [27], and it was hypothesized that primitive microorganism on Earth may be able to use hydantoins as C or N sources [28]. 5,5-Dimethylhydantoin was detected under several conditions of heating of Prussian Blue (Table 1, Entries 5, 6, and 8). The XPS spectroscopic data indicates that this molecule can be formed under moderate temperatures, although its low concentration at the end of the reaction hinders its detection by GC/MS. In one experiment (Table 1, Entry 8), 2-aminoisobutyric acid was detected. This compound is the hydrolysis product of 5,5-dimethylhydantoin. This result was checked by heating of standard solutions of 5,5-dimethylhydantoin in NH4OH (pH 12). 5,5-Dimethylhydantoin was previously detected in acid-hydrolyzed HCN polymers [29] and in carbonaceous chondrites such as Murchinson [30] and Yamato-791198 [31]. In those cases, the formation of hydantoins is explained by reaction between cyanate and amino acids to form the N-carbamoyl derivatives, which, in turn, may cyclize to hydantoins. The formation of cyanate from Prussian Blue seems to be easy in basic media, but the direct formation of amino acids is not, as we demonstrated.

Moreover, different hydroxy acids were also detected under several conditions (Table 1, Entries 2 – 5 and 7 – 10), especially lactic and malic acids. They are of biological interest, because they can act as donor – acceptor electron systems.

Finally, the thermal wet alkaline decomposition of Prussian Blue also leads to the formation of inorganic compounds (Scheme). We can see that, under anoxic conditions and up to pH 10, Prussian Blue is an excellent precursor of hematite, a-Fe2O3 . Hematite is the most frequent and the most thermally stable polymorph of iron(III) oxides that exists in the Earth s surface [32], and it is related to banded iron formations (BIFs). BIFs are defined as chemical sediments, typically thinly-bedded or laminated, whose principal chemical characteristic is an anomalously high content of iron, commonly but not necessarily containing layers of chert [33]. The iron occurs as oxides, principally magnetite (Fe3O4) and hematite, and the chert is composed principally of quartz [34]. From an astrobiological point of view, BIFs are of interest because: i) iron oxides, including hematite, are among a limited number of minerals known to preserve microscopic evidence of life over geologic time, ii) several of the mechanisms proposed for hematite formation may include microbial mediation, and iii) they have been proposed as a candidate site for the emergence of life. At present, the mechanism of formation of BIFs is controversial. Castro [35] has proposed four possible models for their formation: I) bioticoxygenic model, II) biotic-anoxygenic model, III) abiotic-oxygenic model, and IV) abiotic anoxygenic model. In the last one, UV photooxidation of ferrous to ferric oxides at the ocean surface occurred because of the lack of an ozone layer. The formation of Prussian Blue under abiotic conditions and its subsequent transformation to hematite might offer an alternative to the abiotic anoxygenic model for BIFs. Conclusions. – We can conclude that FeII could act as a scavenger and concentrator of carbon in the prebiotic hydrosphere. Wet Prussian Blue decomposition by local pH

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

1319

Scheme. General Reaction Scheme Showing the Main Products Identified in the Thermal Wet Decomposition of the Prussian Blue

fluctuations under anoxic conditions could release HCN and (CN)2 gas with the subsequent formation of discrete organic molecules and iron oxides.

The authors have used the research facilities of Centro de Astrobiolog a (CAB) and have been supported by Instituto Nacional de Te´cnica Aeroespacial Esteban Terradas (INTA) and the project AYA2006-15648-C02-02 of the Ministerio de Educacio´n de Ciencia, Spain. We thank M. T. Ferna´ndez for recording the IR spectra and the DRX measurements, and J. M. Sobrado for his help during the XPS experiments.

Experimental Part Sample Preparation. Prussian Blue was prepared from aq. solns. of K4Fe(CN)6 · 3 H2O and FeCl3 · 6 H2O [36]. Ammonium hexacyanoferrate(II), (NH4 )4Fe(CN)6 · 1.5 H2O, was prepared according to Brauer [37].

1320

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

Suspensions of Fe4[Fe(CN)6 ]3 · 15 H2O (400 mg) in 5n NH4OH (1 ml, pH 12) were heated at r.t., 708, 1008, 1308, and 1508 for 24 h, 48 h, and 1 week in sealed vials under inert atmosphere (see Table 1 for a summary of the experiments). All samples were centrifuged after the indicated reaction time, and the supernatant was separated from the insoluble solid. The supernatants were chromatographed using flashgrade silica gel with i-PrOH/NH3 (30%)/MeCN 17 : 5 : 5 as eluant. In all cases, two fractions were collected. Both fractions were freeze-dried, to yield a white solid (solid 1) and a yellow solid (solid 2).

Fig. 1 shows the general scheme for the workup of the different samples produced.

Elemental Analysis. The elemental analyses of the samples were performed in Unidad de Instrumentacio´n Cient fica (Universidad de Alcala´ de Henares, 28801 Alcala´ de Henares, Spain) using a CHN-O-rapid (Heraeus) elemental analyzer.

IR Spectroscopy. IR Spectra were obtained using a Nexus Nicolet FTIR spectrometer. The spectra were obtained in CsI pellets on the reflectance mode of operation, and they were registered in the range from 4000 to 500 cm 1. The assignment of the bands was achieved using the tables reported by Pretsch et al. [38].

XPS Spectroscopy (X-Ray photoelectron spectroscopy). XPS Spectra were collected in an ultrahigh-vacuum (UHV) system equipped with a multi-channeltron hemispherical electron-energy analyzer, using an AlKa X-ray source. For the XPS experiments, KBr pellets were prepared at 10 – 20% in weight for all the analyzed samples. During measurement, samples presented some charge effects induced by the X-ray radiation. Therefore, each peak has been energy-corrected in order to maintain the energy position of the first scan (overview in all cases). High-resolution Fe 2p and N 1s spectra were recorded within an estimated resolution of ca. 0.9 eV. The spectra are well-described by the superposition of several Doniach – Sunjic curve components. The intensities of the XPS core levels were evaluated by the peak areas, after a standard background subtraction according to the Shirley procedure. The assignment of the binding energy was achieved using standard spectra from the Handbook of X-Ray Photoelectron Spectroscopy [39].

X-Ray Powder Diffraction. The X-ray analysis was performed using an X-ray-diffraction system Seifert, model XRD 3003 TT (from 58 until 708, step 0.18, step time 3 s).

GC/MS. GC/MS Analyses of the TMS derivatives in the full-scan mode were carried out on an Autosystem XL-Turbo Mass Gold (Perkin-Elmer) with an Elite-5 column (crossbond 5% diphenyl/95% dimethylpolysiloxane; 30 m 0.25 mm i.d. 0.25 mm film thickness) and using He as carrier gas. The gasphase analyses were conducted with the same chromatograph using a HP-Plot/Q column (30 m 0.32 mm i.d., 20.0 mm film).

Analytic Procedure for GC/MS. The solids 1 of each reaction were analyzed for org. compounds by GC/MS after treatment with 2,2,2-trifluoro-N-methyl-N-(trimethylsilyl)acetamide (MSTFA) containing 1% Me3SiCl and anh. pyridine 1 : 2 at 1508 for 30 min. The GC oven was programmed as follows: 608 (initial temp.), heated to 1308 at 58/min, held for 11 min, heated to 1808 at 108/min. held for 20 min, and heated to 2808 at 208/min, held for 5 min. The temp. of the injector was 2808, and the flow rate was 2.5 ml/ min. Org. compounds were identified by searching for their mass spectra in the NIST database, and identified compounds were confirmed against authentic standards by comparison of mass spectra and retention times.

Additionally, samples of the gas phases were collected every 24 h and analyzed by GC/MS. The GC oven was programmed as follows: 608 (1 min) to 1508 at a rate of 58/min, held for 5 min, and from 1508 to 1808 at 58/min, held for 10 min. The flow rate was 1.3 ml/min (pressure 10 psi). The mass spectrometer was operated under pos. electron-ionization mode at ionization energy of 70 eV, m/z range 35 – 350 and the transfer line at 1808. 100 ml of gas was injected.

REFERENCES [1] E. Borquez, H. J. Cleaves, A. Lazcano, S. L. Miller, Orig. Life Evol. Biosphere 2005, 35, 79; R. Saladino, C. Crestini, G. Costanzo, E. DiMauro, Curr. Org. Chem. 2004, 8, 1425; M. Levy, S. L. Miller, J. Oro´, J. Mol. Evol. 1999, 49, 165; A. B. Voet, A. W. Schwartz, Bioorg. Chem. 1983, 12, 8; J. P. Ferris, P. C. Joshi, E. H. Edelson, J. G. Lawless, Earth J. Mol. Evol. 1978, 11, 293; J. P. Ferris, D. B.

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

[2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

[22]

[23] [24] [25] [26] [27] [28] [29] [30] [31] [32]

[33]

[34] [35] [36] [37]

1321

Donner, A. P. Lobo, J. Mol. Biol. 1973, 74, 499; J. Oro´, Biochem. Biophys. Res. Commun. 1960, 2, 407.

O. Kikuchi, T. Watanabe, Y. Satoh, Y. Inadomi, J. Mol. Struct. (Theochem) 2000, 507, 53.

R. A. Sanchez, J. P. Ferris, L. E. Orgel, J. Mol. Biol. 1967, 30, 223.

R. Stribling, S. L. Miller, Orig. Life 1987, 17, 261.

S. Miyakawa, H. J. Cleaves, S. L. Miller, Orig. Life Evol. Biosphere 2002, 32, 209.

T. Arrhenius, G. Arrhenius, W. Paplawsky, Orig. Life Evol. Biosphere 1994, 24, 1.

H. J. Buser, D. Schwarzenbach, W. Petter, A. Ludi, Inorg. Chem. 1977, 16, 2704.

L. E. Orgel, The origin of life and the evolucinary biochemistry , Plenum Publising Corporation, New York, 1974, p. 369.

A. D. Keefe, S. L. Miller, Orig. Life Evol. Biosphere 1996, 26, 111.

M. Ruiz-Bermejo, C. Menor-Salva´n, S. Osuna-Esteban, S. Veintemillas-Verdaguer, Orig. Life Evol.

Biosphere 2007, 37, 507.

G. Arrhenius, Sources and geochemical evolution of cyanide and formaldehyde . Fourth Symposium of Chemical Evolution, NASA Ames Research Center, Ed. S. E. Bzik, 1990, p. 31.

V. K. Tiwari, Nat. Acad. Sci. Lett. (India), 1983, 6, 159.

V. K. Tiwari, R. K Sarma, Himalayan Chem. Pharm. Bull. 1985, 2, 32.

H. Inoue, T. Nakazawa, T. Mitsuhashi, T. Shirai, E. Fluck, Hyperfine Interact. 1989, 46, 723.

R. Zboril, L. Machala, M. Mashlan, V. Sharma, Cryst. Growth Des. 2004, 4, 1317.

H. Inoue, S. Yanagisawa, J. Inorg. Nucl. Chem. 1973, 35, 2561.

R. S. Sapieszko, E. Matijevic´, J. Colloid Interface Sci. 1980, 74, 405.

J. C. L. Meeussen, M. G. Keizer, F. A. M. De Haan, Environ. Sci. Technol. 1992, 26, 511.

A. P. Johnson, H. J. Cleaves, J. P. Dworkin, D. P. Glavin, A. Lazcano, J. L. Bada, Science 2008, 322, 404.

B. Morosin, Acta Crystallogr., Sect. B 1978, 34, 3730.

C. N. Matthews, R. D. Minard, Faraday Discuss. 2006, 133, 393; C. N. Matthews, Planet. Space Sci. 1995, 43, 1365; J. P. Ferris, W. J. Hagan Jr., Tetrahedron 1984, 40, 1093; J. P. Ferris, E. H. Edelson, J. M. Auyeung, P. C. Joshi, J. Mol. Evol. 1981, 17, 69.

K. E. Nelson, M. P. Robertson, M. Levy, S. L. Miller, Orig. Life Evol. Biosphere 2001, 31, 221; Y. Yamagata, K. Sasaki, O. Takaoka, S. Sano, K. Inomata, K. Kanemitsu, Y. Inoue, I. Matsumoto, Orig.

Life Evol. Biosphere 1990, 20, 389; A. W. Schwartz, G. J. F. Chittenden, Biosystems 1977, 9, 87.

C. Menor-Salva´n, M. Ruiz-Bermejo, M. I. Guzma´n, S. Osuna-Esteban, S. Veintemillas-Verdaguer, Chem. – Eur. J. 2009, 15, 4411.

R. Shapiro, Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4396.

A. W. Schwartz, M. van der Veen, T. Bisseling, G. J. F. Chittenden, Biosystems 1973, 5, 119.

S. L. Miller, H. C. Urey, Science 1959, 130, 245.

J. Taillades, I. Beuzelin, L. Garrel, V. Tabacik, C. Bied, A. Commeyras, Orig. Life Evol. Biosphere 1998, 28, 61.

C. Syldatk, O. May, J. Altenbuchner, R. Mattes, M. Siemann, Appl. Microbiol. Biotechnol. 1999, 51, 293.

J. P. Ferris, J. D. Wos, A. P. Lobo, J. Mol. Evol. 1974, 3, 311.

G. W. Cooper, J. R. Cronin, Geochim. Cosmochim. Acta 1995, 59, 1003.

A. Shimoyama, R. Ogasawara, Orig. Life Evol. Biosphere 2002, 32, 165.

R. M. Cornell, Schwertmann, The iron oxides. Structure, properties, reactions, occurrences and uses , Wiley-VCH, Weinheim, 2006; R. Zboril, M. Mashlan, D. Petridis, Chem. Mater. 2002, 14, 969.

C. K. Klein, N. J. Beukes, Time distribution, stratigraphy, and sedimentologic setting, and geochemistry of Precambrian iron-formations , in The Proterozoic Biosphere , Ed. J. W. Schopf, C. K. Klein, Cambridge University Press, New York, 1992, p. 139.

C. C. Allen, F. Westall, R. T. Schelble, Astrobiology 2001, 1, 111.

L. O. Castro, Econ. Geol. 1994, 89, 1384.

A. Ludi, J. Chem. Educ. 1981, 58, 1013.

G. Brauer, Handbook of preparative inorganic chemistry , Academy Press, London, 1963.

1322

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)

[38] E. Pretsch, T. Clerc, J. Seibl, W. Simon, Tablas para la elucidacio´n estructural de compuestos orga´nicos por me´todos espectrosco´picos , Alhambra Longman, Madrid, 1980.

[39] J. F. Moulder, W. F. Stickle, P. E. Sobol, K. D. Bomben, Handbook of X-ray Pholoelectron Spectroscopy , Perkin-Elmer Corporation, Physical Electronics Division (USA), 1992.

Received February 2, 2009

Orig Life Evol Biosph DOI 10.1007/s11084-007-9107-0

The Effects of Ferrous and other Ions on the Abiotic Formation of Biomolecules using Aqueous Aerosols and Spark Discharges M. Ruiz-Bermejo & C. Menor-Salván & S. Osuna-Esteban & S. Veintemillas-Verdaguer

Received: 21 December 2006 / Accepted: 14 July 2007 # Springer Science + Business Media B.V. 2007

Abstract It has been postulated that the oceans on early Earth had a salinity of 1.5 to 2 times the modern value and a pH between 4 and 10. Moreover, the presence of the banded iron formations shows that Fe+2 was present in significant concentrations in the primitive oceans. Assuming the hypotheses above, in this work we explore the effects of Fe+2 and other ions in the generation of biomolecules in prebiotic simulation experiments using spark discharges and aqueous aerosols. These aerosols have been prepared using different sources of Fe+2, such as FeS, FeCl2 and FeCO3, and other salts (alkaline and alkaline earth chlorides and sodium bicarbonate at pH=5.8). In all these experiments, we observed the formation of some amino acids, carboxylic acids and heterocycles, involved in biological processes. An interesting consequence of the presence of soluble Fe+2 was the formation of Prussian Blue, Fe4[Fe(CN)6]3, which has been suggested as a possible reservoir of HCN in the initial prebiotic conditions on the Earth.

Keywords Aerosol chemistry . Ancient sea . Biomolecules . Ferrous ion .

Prebiotic chemistry . Prebiotic enviroment . Seawater . Ferrocyanide

Introduction Simulation experiments aiming at the synthesis of organic molecules under prebiotic conditions have been carried out using a diversity of gas mixtures and different external energy sources (i.e., electric discharges, ultraviolet irradiation, high-energy particle or photon beams, etc). In most cases, water was an essential reactant, usually present in vapour form (e.g., Utsumi and Hattori 2002; Miller 1955). Other experiments incorporated liquid water (e.g., Takahashi

M. Ruiz-Bermejo (*) : C. Menor-Salván : S. Osuna-Esteban : S. Veintemillas-Verdaguer Consejo de Investigaciones Científicas-Instituto Nacional de Técnica Aeroespacial (CSIC-INTA), Centro de Astrobiología, Carretera Torrejón-Ajalvir, Km. 4,2, 28850 Torrejón de Ardoz, Madrid, Spain e-mail: ruizbm@inta.es

M. Ruiz-Bermejo, et al.

et al. 2005; Plankensteiner et al. 2004; Takano et al. 2003). Solid water was present in experiments dealing with the irradiation of interstellar ices analogues (e.g., Bernstein et al.

2002; Muñoz-Caro et al. 2002; Kobayashi et al. 1995) and in experiments at low temperature using a cold plasma as the energy source (e.g., Simionescu et al. 1974). One experiment was reported with an extremely low amount of water, less than 10 ppm (Miyakawa et al. 1998).

Water based aerosols are ubiquitous in the actual troposphere (Ellison et al. 1999) and it has been suggested that they could play an important role as prebiotic microreactors in the origin of life (Donaldson et al. 2004; Tervahattu et al. 2004; Tuck 2002; Dobson et al.

2000). Additionally, differences in reactivity between bulk water in any physical state and water present in air–water interfaces are expected (Donaldson and Vaida 2006). In our previous work, where aqueous aerosols played an active role in the synthesis of organic molecules under simulated prebiotic conditions these differences have been experimentally observed (Ruiz-Bermejo et al. 2007). Under those conditions, we found amino acids, monodi- and tri- carboxylic acids, purine bases and other heterocycles in significant amounts and diversity. This result supports the early hypothesis made by Shah (1970), suggesting that aerosol droplets behave as microscopic chemical reactors that offer a number of potential advantages for prebiotic synthesis.

Pure water was employed in the majority of the prebiotic chemistry simulation experiments. Only a few works describe the effect of saline solutions in this type of experiment (Plankensteiner et al. 2006; Manolache et al. 1997; Kobayashi et al. 1990). Nonetheless, it has been postulated that the ancient sea had a salinity of the 1.5 to 2 times the modern value (Knauth 2005; Morse and Mackenzie 1998) and a pH=4–10, although the pH value of early ocean water is under debate (see Kempe and Kazmierczak 2002; Morse and Mackenzie 1998; Russell and Hall 1997; Williams and Frausto da Silva 1996; Bada et al.

1994; Macleod et al. 1994; Grotzinger and Kasting 1993; Kasting 1993; Gregor et al. 1988; Holland et al. 1986; Walker 1985). The presence of banded iron formations (Poulton et al.

2004; Barley et al. 1997; Holland 1989; Walker and Brimblecombe 1985) strongly suggests that dissolved iron (ferrous ion) was present in significant quantities in the ocean water under the reducing atmosphere of the Archean epoch. Finally it has been hypothesized that the concentration of iron could decrease by precipitation as iron carbonate (siderite) (Macleod et al. 1994) or iron sulphide (pyrite) (Summers 2005).

The effects of aerosols and salinity on the prebiotic synthesis of organic compounds are probably correlated. The salinity and pH of the aqueous phase may have an important influence on gas–liquid interfaces, a likely site for relevant prebiotic reactions to occur (Tervahattu et al. 2004). It has been proposed that a solid core, organic or inorganic, formed by partial evaporation of the saline aerosol droplets, could improve the catalytic properties of the aerosol droplets (Lerman and Teng 2004).

The goal of this work was to study the role played by ferrous ions and other ions on the production of organics that result from our bubble–aerosol–droplet (bubble–sol cycle) experiments.

Materials and Methods Simulation of Prebiotic Marine Aerosols A 500 ml glass reactor was filled with a gas mixture containing CH4–N2–H2 (40:30:30), purchased from Praxair, and 5 ml of the following solutions: experiment 1, high salinity

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols Table 1 Solutions or suspensions used in simulations experiments with spark discharges and CH4 +N2 +H2 atmosphere with aqueous aerosol

Experiment

Solution or suspension used

1

0.63 M NaCl, 0.18 M MgCl2, 0.015 M CaCl2, 0.015 M KCl, 0.02 M NaHCO3. pH=5.8 0.88 g/mL FeS. pH=7 0.63 M NaCl, 0.18 M MgCl2, 0.015 M CaCl2, 0.015 M KCl, 0.02 M NaHCO3 +0.88 g/mL FeS 0.01 M FeCl2. pH=7 2 mg/mL FeCO3 (natural siderite). pH=7

2 3 4 5

solution 1.5 times the concentration of actual sea water adjusted to pH=5.8 with HCl (Morse and Mackenzie 1998); experiment 2, suspension of FeS in water; experiment 3, suspension of FeS in the same solution as in experiment 1; experiment 4, a dilute solution of FeCl2 (Mahajan et al. 2003); experiment 5, a suspension of siderite in water. See Table 1 for the detailed concentrations. NaCl, MgCl2, CaCl2 and KCl analytical grade from Panreac, FeS and FeCl2·4H2O from Sigma-Aldrich, NaHCO3 from Fluka and siderite was a natural sample, of hydrothermal origin, collected from Hiendelaencina Mines (Guadalajara, Spain). The solutions were bubbled with dried N2 for 1 h, and prior to use, all glassware and the electrodes were heated in a high-temperature oven (Nabertherm Labotherm L5) at 400°C in air for 2 h, in order to eliminate any possible organic contaminants. Before each experiment the system was successively evacuated with a membrane pump and purged with the reaction gas mixture for 4×. After this process the reactor was filled until a pressure of 1,200 mbar. An ultrasonic aerosol generator (BONECO model 7035) working at 1.8 MHz and 33 W was used for the generation of the bubble–sol cycle that generated the saline aerosol in a few minutes (Ruiz-Bermejo et al. 2007). Two tungsten electrodes attached to the reactor were used with a high voltage generator (Model BD-50E, Electrotechnic Products, Illinois, USA) to produce the spark discharges (50 kV). The system was maintained at constant temperature (38°C) with active aerosol and electric discharge during 72 h. After this period, the liquid solution and the solid material were recovered for analytical study.

Instrumental Analyses Infrared Spectroscopy IR spectra were obtained using a Nexus Nicolet Fourier transform infrared spectrometer. The spectra were obtained in CsI pellet on the reflectance mode of operation.

X-Ray Diffraction The X-ray analysis was done using a X-Ray Diffraction System Seifert, model XRD 3003 TT, (from 5° until 70°, step 0.1°, step time 3 s).

Gas Chromatography-Mass Spectrometry Gas chromatography-mass spectrometry (GC-MS) analyses in the full-scan mode were carried out on an Autosystem XL-Turbo Mass Gold (Perkin Elmer) with an Elite-5 column (crossbond 5% diphenyl–95% dimethyl polysiloxane, 30 m × 0.25 mm i.d.×0.25 μm film thickness) and using He as carrier gas.

High-Performance Liquid Chromatography High-performance liquid chromatography (HPLC) analyses were carried out on a Surveyor (ThermoFinnigan) with a PDA detector using a Kromasil 100 C18 5 μm 25×0.46 column.

Amino acids (μmol) Glycine Aminomalonic acid Alanine b-Alanine Sarcosine Isoserine Serine Aspartic acid Iminodiacetic acid 2-Aminobutyric acid 3-Aminobutyric acid 3-Aminoisobutyric acid N-methylalanine Proline Glutamic acid Valine Ornithine Histidine 2-Aminoadipic acid Cystine t-Leucine Isoleucine Alloisoleucine Total Hydroxy acids (μmol) Glycolic acid 2,3-Dihydroxypropanoic acid Malic acid Tartatic acid 11.435 0.141 11.083 0.848 1.512 0.770 0.899 1.256 – 0.669 3.454 4.707 – 0.372 1.130 – 0.074 d – – 0.422 0.785 0.240 39.77 0.507 0.017 0.018 0.019

C2H5NO2 C3H5NO4 C3H7NO2 C3H7NO2 C3H7NO2 C3H7NO3 C3H7NO3 C4H7NO4 C4H7NO4 C4H9NO2 C4H9NO2 C4H9NO2 C4H9NO2 C5H9NO2 C5H9NO4 C5H11NO2 C5H12N2O2 C6H9N3O2 C6H11NO4 C6H12N2O4S2 C6H13NO2 C6H13NO2 C6H13NO2

C2H4O3 C3H6O4 C4H6O5 C4H6O6

Experiment 1

0.011 0.002 0.029 0.003

7.875 – 13.972 0.270 0.526 0.321 0.293 0.694 – 0.062 0.718 0.844 – – 0.302 d 0.135 d – d – – – 26.01

Experiment 2

0.478 0.010 0.007 0.003

8.427 – 16.118 0.425 0.542 0.515 0.456 0.731 – 0.069 0.891 0.761 – d 0.507 d 0.071 d d d 0.190 – – 29.72

Experiment 3

Table 2 Identification and quantification of molecules of biochemical interest obtained in experiments 1 to 5

0.327 0.008 0.011 0.003

3.434 0.565 4.630 0.126 0.240 0.105 0.074 0.220 – d 0.245 0.387 0.194 – 0.154 – – – – – 0.150 – – 10.52

Experiment 4

0.140 0.028 0.019 0.044

5.091 – 17.738 0.182 – 0.235 0.168 0.259 – 0.049 0.128 0.166 0.170 – – – – – – – – – – 24.19

Experiment 5

0.17 0.02 0.03 0.01

3.93 1.06 12.98 1.16 0.45 0.34 0.14 0.34 0.03 0.18 0.68 1.65 d – 0.08 – 0.43 d – – 0.19 – – 23.64

Control experimenta

M. Ruiz-Bermejo, et al.

C4H6O4 C4H8O4 C5H6O4 C5H8O4 C6H8O6 C6H10O4 C6H10O4

C4H8O4 C5H8O5 C5H8O6 C6H12O3 C7H12O5

0.020 0.003 – 0.002 0.002 – – 0.03

0.05

0.58 0.040 – – 0.003 0.003 0.002 0.001 0.05

0.004 0.002 – 0.015 –

– – – 0.001 0.007

0.003 – – 0.015 –

0.032 0.011 – 0.003 0.001 – – 0.05

0.52

Experiment 3

Experiment 2

Experiment 1

0.028 – – 0.002 0.003 – – 0.03

0.36

0.007 – – 0.003 –

Experiment 4

0.187 0.013 d 0.020 0.030 d – 0.08

0.30

– d d 0.055 0.009

Experiment 5

0.05 0.02 – – 0.01 – – 0.08

0.26

0.02

– 0.01

Control experimenta

a

Data from Ruiz-Bermejo et al. (2007)

d = detected (noise/signal>0.1)

Values are expressed as the mean number of micromoles obtained in each experiment (five repetitions of each experiment were performed). Amino acids were identified and confirmed using authentic standards and two techniques (HPLC and GC-MS). Ornithine, hystidine and cystine were tentatively assigned by comparison with the retention of authentic standard by HPLC only

2,3-Dihydroxybutanoic acid 2-Hydroxypentanodioic acid 2,4-Dihydroxypentanodioic acid Hydroxycaproic acid 2,3-Dimethyl-3-hydroxyglutaric acid Total Carboxylic acids (μmol) Succinic acid But-2-enedioic acid 2-Methyl maleic acid 2-Methylsuccinic acid Tricarballylic acid 2-Methylglutaric acid Adipic acid Total

Table 2 (continued)

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols

M. Ruiz-Bermejo, et al.

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols

R

Fig. 1 HPLC chromatogram showing the amino acid profile obtained in a CH4–N2–H2 atmosphere in the conditions described for experiments 1–5. The analysis of the residue of the water soluble organic fractions after hydrolysis was performed as stated in “Materials and Methods.” All amino acid were identified by two different methods (HPLC and GC-MS) against authentic standards. 1 Aspartic acid. 2 Glutamic acid. 3 Serine. 4 Glycine. 5 Isoserine. 6 β-Alanine. 7 Histidine. 8 Alanine. 9 3-Aminoisobutyric acid. 10 3Aminobutyric acid. 11 Proline. 12 2-Aminobutyric acid. 13 t-Leucine. 14 Isoleucine. 15 Alloisoleucine. 16 Sarcosine. 17 Ornithine. 18 Cystine. 19 Valine. 20 2-Aminoadipic acid. 21 N-Mehtylalanine. Asterisks, compounds identified by HPLC, unconfirmed by GC-MS

Analytical Procedure In all experiments, we collected a water soluble fraction and an insoluble fraction. Both fractions were separated by centrifugation, and then immediately freeze dried by means of a standard lyophilizer and stored at −20°C under nitrogen atmosphere. Prior to analysis the water soluble fractions of each experiment was desalted by ion exchange chromatography [Dowex 50 W X 8-400 (H+)], using water as eluant, hydrolyzed with 6 M HCl at 110°C for 24 h, and freeze dried again to remove water, HCl, and any volatile organics. As a rule, the identification of the mass peaks attributed to organic compounds was verified by comparison with the retention times, mass spectra and UV absorbance spectra of external standards, purchased from Sigma-Aldrich and Fluka. Amino acid analysis: the identification and quantification of amino acids were done by HPLC and GC-MS according to RuizBermejo et al. 2007. Hydroxy- and carboxylic acids analysis: the hydroxy acids and carboxylic acids were identified and quantified by GC-MS using the method described in Ruiz-Bermejo et al. 2007. Heterocycles analysis: Crude and hydrolyzed samples were analyzed using the method described in Ruiz-Bermejo et al. 2007.

Results Analysis of Molecules with Biological Interest Amino Acids The acid hydrolysates of all non-volatile water soluble fractions (experiments 1–5) provide a series of proteinaceus and non-proteinaceus amino acids (Table 2, Fig. 1). The amount and number of amino acids detected in experiment 1 are greater than in the control experiment with pure water (see Table 2). Consequently, the presence of chloride and bicarbonate salts seems to improve the production of amino acids. A similar result was obtained previously in simulation experiments using proton irradiation, CH4/N2 atmosphere and solutions of chlorides and Fe(NH4)2(SO4)2 at pH=7 (Kobayashi et al. 1990). The presence of insoluble salts such as FeS or with low solubility such as FeCO3 (experiment 2 and 5) does not affect significantly the synthesis of amino acids (Table 2). Nevertheless, in the experiment with soluble Fe+2 (FeCl2, experiment 4) the total amount of amino acids detected is lower than in the control experiment (Table 2). The presence of FeS leads to the formation of amino acids containing sulphur. Cystine was tentatively detected in experiments 2 and 3.

M. Ruiz-Bermejo, et al.

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols

R

Fig. 2 Gas chromatograms showing the carboxylic acids and heterocyclic compounds of water soluble fractions after acidic hydrolysis and derivatization with N-methyl-N-trimethylsilyl-trifluoroacetamide. 1 Succinic acid. 2 Methylsuccinic acid. 3 2,3-Dihydroxypropanoic acid. 4 2,3-Dihydroxybutanoic acid. 5 Hydroxycaproic acid. 6 2-Methylglutaric acid. 7 Aminomalonic acid. 8 Malic acid. 9 Adipic acid. 10 Tartaric acid. 11 Tricarballylic acid. 12 But-2-enedioic acid. 13 2,3-Dimethyl-3-hydroxyglutaric acid. 14 2Hydroxypentanodioic acid. 15 2-Methyl maleic acid. 16 [1,3,5]Triazine-2,4,6-triol (tentatively). 17 2,4Dihydroxypentanodioic acid. Single asterisks, compound not identified but with the same mass spectrum as tartaric acid. Double asterisks, compound not identified, it might be a thiol. Triple asterisks, compound not identified but with the same mass spectrum that 2,3-dihydroxybutanoic acid

Hydroxy Acids and Carboxylic Acids We have found hydroxy acids and carboxylic acids in the hydrolysates of the non-volatile water soluble fractions of the experiments 1–5 (Table 2, Fig. 2). Biologically interesting compounds, such as tartaric acid and glyceric acid, as well as succinic acid, malic acid and tricarballylic acid were identified in all experiments. These molecules were also detected in our control experiment with pure water aerosols, but the presence of dissolved salts further increases the total amount of these products (see Table 2). In experiments 1 and 5, an additional amount of carbon is present in the form of bicarbonate and carbonate, respectively, and this fact could explain the yield enhancement in these polar compounds.

Siderite (experiment 5) seems to be the most effective in the increase of the production of these molecules. On the other hand, in experiment 4 the presence of soluble Fe+2 does not remarkably influence the formation of hydroxy acids and carboxylic acids.

Heterocycles Containing Nitrogen We have also studied the non volatile water soluble fractions from experiments 2, 3, 4 and 5 before acid hydrolysis (Fig. 3). In experiments 2, 3 and 5, we detected hydantoin which is another molecule with a possible prebiotic potential. It seems that the presence of salts does not have a significant influence in its formation. However, analysis of the non-hydrolyzed water soluble fraction analysis from experiment 4 did not show the presence of any identified organic compounds.

In contrast with our control experiment, where adenine and 2,6-diaminopurine were formed with significant yields [0.11 μmol and 0.02 μmol, respectively, (Ruiz-Bermejo et al.

2007)], none of the salt bearing experiments reported in this work showed the formation of these compounds.

Formation of Prussian Blue, Fe4[Fe(CN)6]3 The most relevant finding obtained in this work was the formation of the insoluble complex salt Prussian Blue [ferric hexacyaneferrate (II), Fe4[Fe(CN)6]3] (Fig. 4) in experiment 4.

Prussian Blue co-precipitated with the water insoluble organic matter. These fractions could not be separated by standard techniques (Fig. 4a). The amount of water insoluble organic fraction obtained in this experiment was estimated by subtraction assuming that all the iron originally present was in the form of Prussian Blue at the end of the process (roughly 6.15 mg), see Table 3.

Iron (II) sources with solubility lower that FeCl2 such as FeS, FeCO3 (values of pKs 18.1 and 10.5, respectively, experiments 3 and 5), as expected, did not lead to the formation of Prussian Blue.

M. Ruiz-Bermejo, et al.

Fig. 3 Gas chromatograms of water soluble fractions before hydrolysis, derivatized under the same conditions after the acidic hydrolysis. 1 Urea. 2 Oxalic acid. 3 Succinic acid. 4 2, 3-Dihydroxypropanoic acid. 5 1, 3-Dihidro-2H-imidazol-2-one. 6 Methylmalonic acid. 7 Hydantoin. 8 2-Methylglutaric acid. 9 Parabanic acid. 10 Malic acid. 11 [1,2,3]-Triazine-2,4,6-triol. Asterisks, compounds tentatively assignment by NIST library

The value of pH obtained in experiment 4 was significantly lower than in the rest of the experiments (Table 3). The alkalinity of the solutions was due to the presence of ammonium ions tested by Nessler assay.

Discussion Influence of the Formation of Prussian Blue on the Synthesis of Organics in Prebiotic Simulation Experiments using Spark Discharges and Aqueous Aerosols In prebiotic synthesis experiments using spark discharges and CH4, as carbon source, HCN is one of the main products and it is known that when solutions of Fe (II) and CNâ&#x2C6;&#x2019; are mixed, Fex(CN)y precipitates are formed (Sharpe 1976). In the case of experiment 4, the presence of soluble Fe (II) leads to the formation of Prussian Blue. In order to form this complex salt, partial oxidation of Fe (II) to Fe (III) ions is required. The oxidation of free

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols Fig. 4 a Reflectance Fourier transform infrared (FTIR) spectra of the water insoluble fraction from experiment 4. b Reflectance FTIR spectra of Prussian Blue. c Diffractogram of water insoluble fraction from experiment 4. The signals observed were assigned to Prussian Blue by reference to ASTM file number 73-068

ferrous ion is easily performed even in anoxic conditions. Aqueous Fe(II) is oxidized to Fe (III) simply by exposure to light. Additionally, it is well known that in spark discharge experiments radicals are formed and these radicals, as in the biological chemistry of iron, can act as oxidants for Fe (II).

The precipitation of Prussian Blue steals carbon and nitrogen from the system and this fact influences in the end the pH of the raw reaction and reduces the production of the total organic matter (see Table 3). In particular, the formation of amino acids in experiments with spark discharges and CH4 is thought to follow a mechanism similar to that of the Strecker synthesis Table 3 Data of pH and amount of non-volatile organic matter obtained in each experiment

Values are expressed as mean ± SEM of five experiments

Experiment

pH (solution)

S fraction, amount obtained (mg)

I fraction, amount obtained (mg)

1 2 3 4 5

8.6 8.0 8.6 7.4 8.2

68.8±8.3 51.3±2.4 60.1±8.6 38.6±5.2 64.1±4.9

9.7±1.1 11.8±1.0 15.0±3.4 13.9±1.7 10.0±0.4

M. Ruiz-Bermejo, et al.

(Miller 1957). In such a process, any reduction in HCN concentration leads to a reduction in the yield in amino acids. The correlation observed between the decrease of amino acid production and the removal of HCN by precipitation as Prussian Blue, in experiment 4, is in general agreement with this hypothesis. Our results also suggest that metal-bound cyanide was unable to participate in the Strecker synthesis, apparently contradicting the results of Summers and Lerner (1998). These authors indicate that the Strecker synthesis has little sensitivity toward whether the cyanide it uses is free or bound to Fe(II).

The insensitivity of hydroxy and carboxylic acids to Prussian Blue precipitation indicates that HCN would not be implicated in the formation of these compounds. In accordance with the previous observations, we suggest that the formation of hydroxy acids and carboxylic acids under the conditions assayed is through a radical pathway. This hypothesis is in agreement with the results reported by Donaldson and coworkers (Donaldson et al. 2004).

This group indicates that carboxylic acids are produced by reaction of OH with hydrocarbons at the air–water interface, especially in aerosols. On the other hand, Tuck (2002) argues that chain lengthening in hydrocarbons in the anaerobic prebiotic atmosphere would have resulted from gas phase reaction with methane, with further oxidation to alcohols, aldehydes and acids. Thus, the presence of aqueous aerosol and, additionally, the presence of salts would allow the stabilization of the radical intermediate species, giving them a better chance for their evolution to produce the compounds indicated above.

Possible Role of Prussian Blue as Reservoir of HCN It has been suggested that polymerization of HCN is fundamental in the prebiotic synthesis of organic compounds (e. g., Kikuchi et al. 2000; Ferris and William 1984; Miller and Orgel 1974). Following this hypotheses the HCN is implicated in the formation of the precursors of amino acids and heterocycles such as purine and pyrimidines bases (Borquez et al. 2005; Saladino et al. 2004; Levy and Miller 1999; Voet and Schwartz 1982; Oró 1960).

If HCN polymerization was actually important for the production of the first and essential biomolecules, there must have routes by which dilute HCN solutions were concentrated (Chen and Chen 2005). The formation of Prussian Blue offers an alternative mechanism for the concentration of cyanide in the form of the ferrocyanide, [Fe(CN)6] 4− ion (Arrhenius et al. 1994). Other authors have suggested that ferrocyanide could be an abundant component in the primitive sea (Orgel 1974) and ferrocyanides and ferricyanides have been mentioned as compounds with prebiotic interest (Keefe and Miller 1996).

It is important to remark that solubility of Prussian Blue is highly dependent on pH and redox potential of the environment. At 4<pH<10, Prussian Blue is a stable and insoluble compound but outside that interval it undergoes solubilization, releasing cyanide [these values were calculated on the data reported by Meeussen et al. (1992)]. Oxidizing or highly reductive mediums could also lead to unstabilization of Prussian Blue.

We have demonstrated that Prussian Blue is easily formed under plausible prebiotic conditions. Subsequent reactions of the Prussian Blue, triggered by pH fluctuations, might lead to the production of compounds of interest from the point of view of origins of life (experiments currently in progress).

Conclusions We can conclude that, using CH4 as carbon source and spark discharges for the prebiotic synthesis of organic compounds, many aqueous environments can be considered as relevant

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols

sites for prebiotic synthesis of amino acids. The presence of soluble ferrous salts partially inhibit the amino acid synthesis, probably due to oxidation of ferrous to ferric ions and the precipitation of Prussian Blue. We are currently investigating a possible role for this compound as an intermediate in prebiotic chemistry.

Acknowledgements The authors have used the research facilities of Centro de Astrobiología and have been supported by Instituto Nacional de Técnica Aeroespacial “Esteban Terradas” and the project AYA200615648-C02-02 of the Ministerio of Educación y Ciencia (Spain). We acknowledge the constructive revision and comments of Alan W. Schwartz and Guillermo Muñoz Caro.

References Arrhenius T, Arrhenius G, Paplawsky W (1994) Archean geochemistry of formaldehyde and cyanide and the oligomerization of cyanohydrin. Orig Life Evol Biosph 24:1–17 Bada JL, Bigham C, Miller SL (1994) Impact melting of frozen oceans on the early Earth. Implications for the origin of life. Proc Natl Acad Sci U S A 91:1248–1250 Barley ME, Pickard AL, Sylvester PJ (1997) Emplacement of a large igneous province as a possible cause of banded iron formation 2.45 billion years ago. Nature 385:55–58 Bernstein MP, Dworkin JP, Sandford SA, Cooper GW, Allamandola LJ (2002) Racemic amino acids from the ultraviolet photolysis of interstellar ice analogues. Nature 416:401–403 Borquez E, Cleaves HJ, Lazcano A, Miller SL (2005) An investigation of prebiotic purine synthesis from the hydrolysis of HCN polymers. Orig Life Evol Biosph 35:79–90 Chen QW, Chen CL (2005) The role of inorganic compounds in the prebiotic synthesis of organic molecules.

Curr Org Chem 9:989–998 Dobson CM, Ellison GB, Tuck AF, Vaida V (2000) Atmospheric aerosols as prebiotic chemical reactors.

Proc Natl Acad Sci U S A 97:11864–11868 Donaldson DJ, Vaida V (2006) The influence of organics films at the air–aqueous boundary on atmospheric processes. Chem Rev 106:1445–1461 Donaldson DJ, Tervahattu H, Tuck AF, Vaida V (2004) Organic aerosols and the origin of life: an hypothesis.

Orig Life Evol Biosph 34:57–67 Ellison GB, Tuck AF, Vaida V (1999) Atmospheric processing of organic aerosols. J Geophys Res 104:11633–11641 Ferris JP, William Jr JH (1984) HCN and chemical evolution: the possible role of cyano compounds in prebiotic synthesis. Tetrahedron 40:1093–1120 Gregor CB, Garrels RM, Mackenzie FT, Maynard JB (1988) Chemical cycles in the evolution of the Earth.

Wiley, New York Grotzinger JP, Kasting JF (1993) New contrains on Precambrian ocean composition. J Geol 101:235–243 Holland HD (1989) The chemical evolution of the atmosphere and oceans. Princeton University Press, Princeton Holland HD, Lazar B, McCaffey M (1986) Evolution of the atmosphere and oceans. Nature 320:27–33 Kasting JF (1993) Evolution of the Earth’s atmosphere and hydrosphere: Hadean to recent. In: Engel MH, Macko SA (eds) Organic geochemistry: principles and applications. Plenum, New York, pp 611–623 Keefe AD, Miller SL (1996) Was ferrocyanide a prebiotic reagent? Orig Life Evol Biosph 26:111–129 Kempe S, Kazmierczak J (2002) Biogenesis and early life on Earth and Europa: favored by an alkaline ocean? Astrobiology 2:123–130 Kikuchi O, Watanabe T, Satoh Y, Inadomi Y (2000) Ab initio GB study of prebiotic synthesis of purine precursors from aqueous hydrogen cyanide: dimerization reaction of HCN in aqueous solution. J Mol Struct Theochem 507:53–62 Knauth LP (2005) Temperature and salinity history of the Precambrian ocean: implications for the course of microbial evolution. Palaeogeogr Palaeoclimatol Palaeoecol 219:53–59 Kobayashi K, Tsuchiya M, Oshima T, Yanagawa H (1990) Abiotic synthesis of amino acids and imidazole by proton irradiation of simulated primitive Earth atmospheres. Orig Life Evol Biosph 20:99–109 Kobayashi K, Kasamatsu T, Kaneko T, Koike J, Oshima T, Saito T, Yamamoto T, Yanagawa (1995) Formation of amino acid precursors in cometary ice environments by cosmic radiation. Adv Space Res 16:221–226

M. Ruiz-Bermejo, et al.

Lerman L, Teng J (2004) In the beginning. In: Seckbach J (ed) Origins: genesis and diversity of life, vol 6.

Springer, Berlin Heidelberg New York, pp 35–55 Levy M, Miller SL (1999) Production of guanine from NH4CN polymerization. J Mol Evol 49:165–168 Macleod G, Mckeown C, Hall AJ, Russell MJ (1994) Hydrotermal and oceanic pH conditions of possible relevance to the origin of life. Orig Life Evol Biosph 24:19–41 Mahajan TB, Elsila JE, Deamer DW, Zare RN (2003) Formation of carbon–carbon bonds in the photochemical alkylation of polycyckic aromatic hydrocarbons. Orig Life Evol Biosph 33:17–35 Manolache S, Romanescu C, Cobialeac G, Simonescu CI (1997) Some terpenes and related derivatives synthesized in cold plasma systems. Molecules 2:158–164 Meeussen JCL, Keizer MG, de Haan FAM (1992) Chemical stability and decomposition rate of iron cyanide complexes in soil solutions. Environ Sci Technol 26:511–516 Miller SL (1955) Production of some organic compounds under possible primitive Earth conditions. J Am Chem Soc 77:2351–2361 Miller SL (1957) The formation of organic compounds on the primitive Earth. Ann N Y Acad Sci 69:260–274 Miller SL, Orgel LE (1974) The origin of life on the Earth. Prentice Hall, Englewood Cliffs Miyakawa S, Tamura H, Sawaoka AB, Kobayashi K (1998) Amino acids synthesis from an amorphous substance composec of carbon, nitrogen, and oxygen. Appl Phys Lett 72:990–992 Morse JW, Mackenzie FT (1998) Hadean ocean carbonate geochemistry. Aquat Geochem 4:301–319 Muñoz-Caro GM, Meirhenrinch UJ, Schutte WA, Barbier B, Arcones-Segovia A, Rosenbauer H, Thiemann WH-P, Brack A, Greenberg JM (2002) Amino acids from ultraviolet irradiation of interstellar ice analogues. Nature 416:403–406 Orgel LE (1974) In: Dose K, Fox SW, Deborin GA, Pavlovskaya TE (eds) The origin of life and the evolutionary biochemistry. Plenum, New York, pp 369–371 Oró J (1960) Synthesis of adenine from ammonium cyanide. Biochem Biophys Res Commun 2:407–412 Plankensteiner K, Reiner H, Schranz B, Rode BM (2004) Prebiotic formation of amino acids in a neutral atmosphere by electric discharge. Angew Chem Int Ed Engl 43:1886–1895 Plankensteiner K, Reiner H, Schranz B, Rode BM (2006) Amino acids on the rampant primordial Earth: electric discharges and the hot salty ocean. Mol Divers 10:3–7 Poulton SW, Fralick PW, Canfield DE (2004) The transition to sulphidic ocean ≈1.84 billions years ago.

Nature 431:173–177 Ruiz-Bermejo M, Menor-Salván C, Osuna-Esteban S, Veintemillas-Verdaguer S (2007) Prebiotic microreactors: a synthesis of purines and dihydroxy compounds in aqueous aerosol. Orig Life Evol Biosph 37:123–142 Russell MJ, Hall AJ (1997) The emergence of life from iron monosulphide bubbles at a submarine hydrothermal redox and pH front. J Geol Soc Lond 154:377–402 Saladino R, Crestini C, Costanzo G, DiMauro E (2004) Advances in the prebiotic síntesis of nucleic acids bases: implications for the origin of life. Curr Org Chem 8:1425–1443 Shah DO (1970) The origin of membranes and related surface phenomena. In: Ponnamperuma C (ed) Exobiology. North Holland, Amsterdam, pp 235–265 Sharpe AG (1976) The chemistry of cyano complexes of transition metals. Academic, London Simionescu CR, Denes F, Onac D, Bloos G (1974) Síntesis of some amino acids, sugars, and peptides in cold plasma. Abiotic synthesis of some high-molecular-weight proteid-like structures (V). Biopolymers 13:943–954 Summers DP (2005) Ammonia formation by the reduction of nitrite/nitrate by FeS: ammonia formation under acidic conditions. Orig Life Evol Biosph 35:299–312 Summers DP, Lerner N (1998) Ammonia from Iron (II) reduction of nitrite and the Strecker synthesis: Do iron (II) and cyanide interfere with each other? Orig Life Evol Biosph 28:1–11 Takahashi J, Masuda H, Kaneko T, Kobayashi K, Saito T, Hosokawa T (2005) Photochemical abiotic synthesis of amino-acid precursors from simulated planetary atmospheres by vacuum ultraviolet light. J Appl Phys 98:024097 Takano Y, Ohashi A, Kaneko T, Kobayashi K (2003) Abiotic synthesis of high-molecular-weight organics from an inorganic gas mixture of carbon monoxide, ammonia, and water by 3 MeV proton irradiation.

Appl Phys Lett 84:1410–1412 Tervahattu H, Tuck A, Vaida V (2004) Chemistry in prebiotic aerorsols: a mechanism for the origin of life.

In: Seckbach J (ed) Origins: genesis, evolution and diversity of life, vol 6. Springer, Berlin Heidelberg New York, pp 153–165 Tuck A (2002) The role of atmospheric aerosols in the origin of life. Surv Geophys 23:379–409

Effects of ferrous and other ions on biomolecule formation in aqueous aerosols Utsumi Y, Hattori T (2002) Synthesis of ammonium and organic compounds from N2, H2O vapour, and CO2 gas mixture by synchrontron radiation induced photochemical reactions at atmospheric pressure and at room temperature. Rev Sci Instrum 73:1387–1389 Voet AB, Schwartz AW (1982) Uracil synthesis via HCN oligomerization. Orig Life 12:45–49 Walker JCG (1985) Carbon dioxide on the early Earth. Orig Life Evol Biosph 16:117–127 Walker JCG, Brimblecombe P (1985) Iron and sulphur in the pre-biotic ocean. Precambrian Res 28:205–222 Williams RJP, Frausto da Silva JJR (1996) The natural selection of the chemical elements. Claredon, Oxford p. 305 ff.

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited.

In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevierâ&#x20AC;&#x2122;s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy

Icarus 204 (2009) 672–680

Contents lists available at ScienceDirect

Icarus journal homepage: www.elsevier.com/locate/icarus

CH4/N2/H2-spark hydrophobic tholins: A systematic approach to the characterisation of tholins. Part II Marta Ruiz-Bermejo a,*, César Menor-Salván a, José Luis de la Fuente b, Eva Mateo-Martí a, Susana Osuna-Esteban a, José Ángel Martín-Gago a,c, Sabino Veintemillas-Verdaguer a,c a

Centro de Astrobiología [Consejo Superior de Investigaciones Científicas-Instituto Nacional de Técnica Aeroespacial (CSIC-INTA)], Carretera Torrejón-Ajalvir, Km. 4.2. Torrejón de Ardoz, E-28850 Madrid, Spain Instituto Nacional de Técnica Aeroespacial (INTA), Carretera Torrejón-Ajalvir, Km. 4.2. Torrejón de Ardoz, E-28850 Madrid, Spain c Instituto de Ciencia de Materiales de Madrid (CSIC), C/Sor Juana Inés de la Cruz, 3. Cantoblanco, E-28049 Madrid, Spain b

a r t i c l e

i n f o

Article history: Received 2 March 2009 Revised 25 May 2009 Accepted 2 July 2009 Available online 10 July 2009 Keywords: Prebiotic chemistry Prebiotic environments Organic chemistry Spectroscopy

a b s t r a c t Two different simulation experiments of prebiotic synthesis were carried out in a CH4/N2/H2 atmosphere with spark discharge activation of aqueous aerosols and liquid water. In both cases, a hydrophilic tholin and a hydrophobic tholin were obtained. The methodology developed by our group for the characterisation of hydrophilic tholins [Ruiz-Bermejo, M., Menor-Salván, C., Mateo-Martí, E., Osuna-Esteban, S., Martín-Gago, J.A., Veintemillas-Verdaguer, S., 2008. Icarus 198, 232–241] was used in order to study the hydrophobic tholins. The gas precursors of the tholins from mixtures containing CH4, with and without H2, were studied. We propose that the formation of the hydrophobic tholins involves the formation of unsaturated oligomeric hydrocarbon chains from vinyl and acetylene monomers, as well as allene derivatives formed in the gas phase after the incorporation of polar groups into these hydrocarbon chains.

Finally, we compare our results concerning hydrophobic tholins with HCN polymers, since it is generally suggested that the polymeric material formed in spark experiments are possible oligomers of HCN, and that Titan’s tholins could be poly-HCN. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction Tholins are complex organic materials obtained by the irradiation of gas mixtures containing CH4, CO2 or CO and N2 or NH3 (Sagan et al., 1993). These substances could have played an important role in the origin of life on Earth (Sagan and Khare, 1979). The hydrolysis of synthetic tholins yields important bioorganics such as amino acids, purine and pyrimidine bases and carboxylic acids (see e.g. Ruiz-Bermejo et al., 2007; Koike et al., 2003; McDonald et al., 1994, 1991; Khare et al., 1986). The pyrolysis of tholins also gives organic compounds (see e.g.

McGuigan et al., 2006; Takano et al., 2004; Pietrogrande et al., 2001; Coll et al., 1999). On the other hand, several aspects related to the formation of tholins during simulation experiments can help us understand the chemical atmospheric evolution of the planets and the deposition and concentration of organic compounds on their surfaces.

In the last few decades, interest in Saturn’s Moon Titan has increased. Titan has a thick atmosphere containing CH4 and N2.

This atmospheric composition and the presence of different sources of energy led to the formation of tholins (Imanaka et al., * Corresponding author. Fax: +34 91 520 6410.

E-mail address: ruizbm@inta.es (M. Ruiz-Bermejo).

0019-1035/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.icarus.2009.07.001

2004). Long linear chains or aromatic structures, as well as a mixture of both, have been proposed for the nature of Titan’s tholins (Lebonnois et al., 2002). Also, it has been suggested that the base of Titan’s tholins could be poly-HCN (Coll et al., 1999; Matthews, 1995), and the same idea has been proposed for the insoluble solids in simulation spark experiments (Cleaves et al., 2008). Since on Titan water seems to be scarce, we hypothesise that the formation of hydrophobic tholins should be preferential with respect to the formation of hydrophilic tholins.

Two different simulation experiments of prebiotic synthesis were carried out in a CH4/N2/H2 atmosphere with spark discharge activation of aqueous aerosols and liquid water. In both cases, a hydrophilic tholin and a hydrophobic tholin were obtained. In order to obtain information about the formation, structure and nature of the hydrophobic tholins, we used the methodology previously developed by us for the characterisation of hydrophilic tholins (Ruiz-Bermejo et al., 2008). Additionally, we studied the nature of the gas precursors of hydrophobic tholins using different CH4 mixtures, both with or without H2, to elucidate a plausible mechanism for their formation. This information together with the development of a systematic characterisation of the tholins is interesting since this methodology can help to resolve some open questions about the structure of Titan’s tholins. Thus, the goal of this paper is to compare our experimental results on the hydrophobic

Author's personal copy

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

tholins with the data previously reported in the bibliography to infer how the possible mechanisms of formation have influence on the nature of different tholins.

673

dried supernatant was derivatizated for GC–MS analysis by adding pyridine (150 lL) and MSTFA + 1% TMCS (50 lL) and heating the mixture at 150 °C for 30 min. Details on the carboxylic acid analysis by GC–MS are described in Ruiz-Bermejo et al. (2007).

2. Experimental 2.6. Molecular weight estimation 2.1. Production of tholins The hydrophobic tholins were obtained from an CH4/N2/H2 atmosphere with spark discharge activation of aqueous aerosol (tholin 1) and liquid water (tholin 2), 5 mL of water were used in each experiment. The reaction times were 3 days, the temperature of the system 38 °C and the initial pressure 1 atm. The aqueous aerosol were regenerated with the helpful of a ultrasonic aerosol generator (BONECO model 7035). The complete details about the preparation and separation of the hydrophilic and hydrophobic tholins are given in Ruiz-Bermejo et al. (2007).

MALDI mass spectra were obtained in a Reflex III (Bruker) using ditranol, tetracyanequinodimethane (TCNQ) or 2,5-dihydroxybenzoic acid (DHB) as the matrix for the hydrophobic tholins. The hydrophobic tholins were partially solvated in DMSO and the samples were analysed by ESI-TOF mass spectrometry using an Agilent 6210 Time-of-Flight LC/MS TOF MS. Mass spectrometry analyses were performed at the Servicio Interdepartamental de Investigación, (Universidad Autónoma de Madrid, Ciudad Universitaria de Cantoblanco, 28049 Madrid, Spain) and at Unidad de Instrumentación Científica (Universidad de Alcalá de Henares. 28801 Alcalá de Henares, Spain).

2.2. Elemental analysis 2.7. Thermal analysis Elemental C, H and N analyses were performed using a ‘‘CHN-Orapid” (Heraeus) elemental analyser at Unidad de Instrumentación Científica (Universidad de Alcalá de Henares. 28801 Alcalá de Henares, Spain).

2.3. Standard spectroscopy techniques IR spectra were obtained using a Nexus Nicolet FT-IR spectrometer. The spectra were obtained in CsI pellets using the reflectance mode of operation.

13 C CP MAS-NMR spectra were obtained on a Bruker Advance 400 spectrometer using a standard cross-polarisation pulse sequence. Samples were spun at 10 kHz. The spectrometer frequency was set to 100.62 MHz. A contact time of 1 ms and a period between successive accumulations of 5 s were used. The number of scans was 1600. Chemical shift values were referenced to TMS (trimethylsilane).

2.4. XPS spectroscopy (X-ray photoemission spectroscopy) The XPS analysis was carried out with a hemispherical electron analyser using an MgKa X-ray source (1253.6 eV). The sample, a tholin–CsI pellet, was introduced in an ultra high vacuum chamber with a base pressure in the 10 9 mbar pressure range. Experiments were carried out at room temperature. The XPS spectra of the different regions were decomposed in Gaussian curve components.

We applied the criterion of using the lowest number of components for the analysis of the N(1s) and C(1s) peaks. The XPS assignment of the different nitrogen chemical groups into the tholins was made by recording XPS spectra of simple pattern molecules as references. These reference samples, such as alanine (amine species), adenine (nitrogen heterocycles and amine species) and nucleic peptide acids, PNAs (Briones and Martín-Gago, 2006) (amides species), were measured under the same experimental conditions to calibrate the binding energies of each nitrogen group involved with a different chemical species. The error between the curve fitting and the XPS spectra is about 0.1–0.5%.

2.5. Reactivity assays Oxidative cleavage of double bonds. A suspension of the hydrophobic tholin (20 mg), NaIO4 (134 mg) and RuCl3 (2 mg) in acetonitrile/hexane/water (7:7:1) was refluxed for 24 h. The crude reaction was centrifuged and washed with water (3 ). The supernatant was collected and dried under reduced pressure. Finally, the

Differential scanning calorimetry (DSC) measurements on selected samples were made using a DSC 2920 TA Instrument. Temperature and heat flow were calibrated with indium and an empty pan was used as a reference. We used a hermetically sealed aluminium pan. Sample and reference pans were heated from ambient temperature to 500 °C at 5 °C/min.

2.8. Gas precursor analysis by GC/MS Analysis of the gas-phase precursors of the tholins. GC–MS analyses in the full-scan mode were carried out on an Autosystem XL-Turbomass (Perkin Elmer) with a HP-Plot/Q column (30 m 0.32 mm i.d., 20.0 lm film) using He as the carrier gas.

Method 1: the temperature was raised from 60 °C (1 min) to 150 °C at a rate of 5 °C/min, held for 5 min, and then from 150 °C to 180 °C at 5 °C/min and held for 10 min. The flow rate was 1.3 mL/min (pressure 10 psi). The mass spectrometer was operated under EI+ mode at ionisation energy of 70 eV, m/z range of 35–350, and the transfer line was at 150 °C. 100 lL of gas was injected.

Method 2: the temperature was raised from 40 °C (1 min) to 100 °C at a rate of 2.5 °C/min, held for 5 min, then from 100 °C to 180 °C at 5 °C/min, held for 15 min and then from 180 °C to 200 °C at 10 °C/min, and held 1 min. The flow rate was 1.8 mL/ min (pressure 12 psi). The mass spectrometer was operated under EI+ mode at an ionisation energy of 70 eV, m/z range 21–350 and the transfer line was at 180 °C. 100 lL of gas was injected.

3. Results The empirical formulas of the hydrophobic tholins [from here to the end of the text we will refer to the hydrophobic tholin from experiment 1 (aqueous aerosol) as tholin 1, and the hydrophobic tholin from experiment 2 as tholin 2 (liquid water)] determined by elemental analysis were C30H45NO3 and C106H157N5O14, respectively. Tholin 1 constitutes 10.5% of the total C input in the system, while tholin 2 constitutes 18.5%. In other words, in relation to the total non-volatile organic material collected in experiments 1 and 2, tholin 1 constitutes 38% of C and tholin 2 constitutes 47%.

Therefore, the presence of aqueous aerosols partially inhibits the formation of hydrophobic tholins. The hydrophobic tholins 1 and 2 are not soluble in water and have very low solubility in common organic solvents such as hexane, toluene and DMSO. The two hydrophobic tholins are thermically stable solids in the range of 25–500 °C, and phase changes were not observed in the

Author's personal copy

674

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

range indicated. The thermal stability was established by DSC measurements.

3.1. Determination of functional groups The IR spectra of hydrophobic tholins 1 and 2 were recorded, and no significant differences were observed between them (Fig. 1). The two solids presented strong absorptions related to the asymmetric and symmetric stretching and bending modes for –CH3 and –CH2-groups. The IR spectra also showed bands that can be associated with the O–H stretching mode, the N–H stretching mode, C@O bending, alkene, alkyne, nitrile or imine groups, and features related to C@C@C (cumulene system) or C@C@CHCONH2. The IR spectrum of the hydrophobic tholins after acid hydrolysis were also recorded (Fig. 1c). The bands related to double bonds and conjugated systems disappeared after acid hydrolysis. The band at 1540 cm 1, usually associated with NH II of the amide bonds, remained after acid hydrolysis, so there was no evidence of peptide bonds in these solids, but N–H groups were clearly present. For a complete assignment of all bands for the tholins and their hydrolysis products, see Table 1.

In the solid-state 13C NMR spectra of tholins 1 and 2, broad resonances were observed (Fig. 2). Signals that could be due to nitriles (–C„N) and alkenes (C@C) were overlapping (centred resonances around 129 and 115 ppm, respectively). The resonances near 70 ppm were assigned to hydroxylic carbons (C–OH), but an alkyne group (C„C) contribution is also possible. A very broad resonance was observed in the range of 60–10 ppm. In this signal, it is possi-

ble that there are overlapping resonances corresponding to amine (–CNH–R) and –CH, –CH2– and –CH3 groups. An additional resonance at 173 ppm was observed in tholin 1, which could be assigned to C@O of either carboxylic acids or esters. These data generally agree with the IR spectra.

The C1(s) XPS core level peak of both tholins 1 and 2 showed two components around 285 eV and 288 eV. We assigned the first one to C–H, C–C, and C@C groups, and the second one to carbonyl, carboxylic and C@N bonds, respectively. This second component was about 18% for tholin 1; whereas it was 5% for tholin 2. Fig. 3 shows the N1(s) XPS core level peak. It can also be resolved into two components around 398 eV and 400 eV binding energy, which are related to nitrogen with unsaturated bonds (–N@) and to NH amine groups and nitriles, respectively (see Table 2 for details on the assignment). The ratio of the two nitrogen components for the tholins was almost similar: tholin 1 presented 82% unsaturated bonds and 18% amines and nitriles, and tholin 2 presented 76% unsaturated bonds and 24% amines and nitriles. Therefore, both tholins had similar chemical compositions with respect to the nitrogen functional groups.

The spectroscopy data of the hydrophobic tholins indicated conjugated structures. In order to confirm this result, an assay for double bonds was performed (Kimberley, 2002). Oxidative cleavage with RuCl3 in the presence of NaIO4 led to the formation of succinic acid and glutaric acid, as identified by GC–MS (Fig. 4). This result confirmed the presence of double bonds and structures such as –C@CH–CH2–CH2–HC@C– and –C@CH–CH2–CH2–CH2–HC@C– in the hydrocarbon chain of the hydrophobic tholins.

3.2. Assays to determine molecular weight

a 2730

The mass spectra of the hydrophobic tholins were obtained using MALDI-TOF. In the case of hydrophobic tholin 1, peaks separated by 44 amu were observed in the range of 600–900 amu when the sample was prepared in the solid state by grinding together with the TCNQ matrix. These results showed a low reproducibility, likely due to the heterogeneity of the sample, the limitations of the preparation technique and the low volatility of these tholins. However, the ESI-TOF mass spectra of the tholins suspended in DMSO indicated a maximum range of mass about 2800 amu.

2235 1956

3435 3309 1377 1458

2873

1706

2960 2931

b

2729 1955 2235 2216

3.3. Analytical study of the gas-phase precursors of hydrophobic tholins

1379 1460

3310

c

1711 2730

3434

2213 1379 1460 1707

2957

d 2187 3444 3191 3330 4000

3500

3000

1645 2500

2000

1417

1500

1000

Wavenumber (cm -1) Fig. 1. Transmission FT-IR spectra. (a) Tholin 1. (b) Tholin 2. (c) Hydrolyzed tholin 1. (d) Insoluble HCN polymer.

In a recent work, we proposed that during the prebiotic simulation experiments in the presence of water, there are competitive mechanisms for the formation of organic compounds, and that the reactions in the gas phase may favour the production of hydrophobic tholins (Ruiz-Bermejo et al., 2008). In this regard, the formation of gas species was studied. After two hours of spark discharges, the gas phases of experiments 1 and 2 were analysed

by GC–MS. In both cases, the composition of the gas mixture was the same, formed mainly of propane, pentane, hexane and a set of alkenes and alkynes from C2 to C5 (see Fig. 5). It is interesting to point out the presence of propene, propyne, 1-buten-3-yne and 1,3-butadiyne due to their relevance in the formation of hydrophobic tholins, as one will see below. Oxygenated compounds such as acetone and acetaldehyde were also observed. Acetylene and ethylene were also detected, as well as hydrogen cyanide (Fig. 6). Additionally, experiments under the same conditions but without hydrogen were carried out, and the same mixture of gas precursors was obtained. As a control experiment, a reaction without water was performed, which gave the same results.

As a summary of the previous discussion, we can conclude that: (i) the tholins 1 and 2 are water insoluble solids with the same heterogeneous nature. (ii) They seem to be composed of hydrocarbon

Author's personal copy

675

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680 Table 1 Characteristic frequency of IR absorption spectra obtained from the tholins 1 and 2 and from the acid hydrolyzed tholin 1. The frequency is expressed in cm (b) = broad, (m) = medium, (s) = strong, (v s) = very strong, (sh) = shoulder, (w) = weak, (v w) = very weak.

1

. Intensity code:

Tholin 1

Tholin 2

Hydrolyzed tholin 1

Insoluble HCN polymer

Vibrational groups identity

3435 (b, m)

(b, m) (b, m) (m) (v w)

3433 (b, m)

3444 (sh) 3330 (b, s)

3309 (m) 3075 (sh)

3437 3389 3310 3078

2960 2931 2873 2730 2235 2209

(v s) (v s) (v s) (w) (w) (w)

2961 2932 2873 2729 2235 2216

(vs) (v s) (v s) (w) (m) (m)

2959 (v s) 2927 (v s) 2872 (v s) 2730

2112 1956 1706 1671 1536

(v w) (w) (v s) (v s) (v w)

2111 1955 1711 1676 1529

(w) (m) (v s) (sh) (w)

2155 (v w)

N–H stretching (single bond) or NH2 asym. stretching or O–H stretching NH2 asym. stretching or O–H stretching C„CH C@CH2 stretching –NH2 sym stretching (may be –NH2 in primary amides) C–H stretching (CH3 asym.)

C–H stretching (CH2) C–H stretching (CH3 sym.)

N–CH3 or N–CH2– R–CN stretching or R–C„CAR C„C or conjugated nitriles C„N stretching or/and C„C stretching N@C@O or N@C@N C@C@C or C@C@CHCONH2 C@O stretching C@O bending or N–C@C, O–C@C or C@C (stretching) or CH@N–N@CH N–H bending (possible amide) or aromatics (quadrant stretch) or C@C (+contribution C@N, N@N) CH2 asym. bending C–N stretching (amide III bands) CH3 sym. bending C–O stretching (C–OH in alcohols) C–O vibrational mode C–O vibrational mode or C@C@stretching sym.

Vynil C–H bending

3308 (v w) 3190 (b, s)

2213 (w) 2187 (m)

1707 (v s) 1645 (b, s) 1540 (v w)

1458 (v s)

1460 (m)

1459 (s)

1377 (v s)

1379 (m)

1377 (m)

1417 (s) 1185 (w) 1082 (b, w) 1073 (b, m) 970 (w)

969 (w)

970 (v w)

1009 (w)

chains containing double C@C and triple C„C bonds, occasionally containing carbonyl (C@O), hydroxyl (–OH), nitrile (C„N), imine (C@N), and amine (–NH) groups. (iii) No positive results for tholins 1 and 2 were obtained in the bioorganic compound analysis by HPLC, GC–MS and SPME/GC–MS [these results were not commented on above because they are negative. The organic analysis methodology was the same as described in Ruiz-Bermejo et al.

(2008) for hydrophilic tholins]. (iv) Under the conditions reported here, hydrophobic tholin formation was independent of the water state (aerosol or liquid). (v) The presence of aqueous aerosols does not favour the formation of hydrophobic tholins. (vi) The production of olefinic and alkynes precursors of hydrophobic tholins seems to be independent of the presence of hydrogen.

34 22 14 129

a

72

173 31 14

4. Discussion 4.1. Formation of hydrophobic tholins

127 71

b 159 168

79

c 300

250 250

200

150

100

50

0

-50

δ (ppm) Fig. 2. Solid-state 13C NMR spectra from: (a) Tholin 1. (b) Tholin 2. (c) Insoluble HCN polymer.

4.1.1. Structure. Mechanistic proposal Taking into account the physical and chemical properties of hydrophobic tholins, it is logical to think that they are formed by a cross-linked network of hydrophobic polymers with a high degree of structural heterogeneity. The tholins’ nature makes them extremely difficult to study and characterise, as was demonstrated in the previous sections of this work and in the literature (Lebonnois et al., 2002). This difficulty also is well-known in the field of polymer chemistry when the objective is the characterisation of cross-linked polymer networks (Kannurpatti et al., 1997).

The synthetic methods used in polymer chemistry to prepare a cross-linked polymeric network can be classified into two groups according to a free-radical mechanism (Moad and Solomon, 1995). In the first, a network is produced by vulcanisation, peroxidation, or irradiation of a pre-existing linear polymer, generally with double bonds along its primary structure. In the second, network formation involves the copolymerisation of a vinyl monomer with a small amount of a multifunctional monomer.

Author's personal copy

676

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

lene derivatives contain a cumulated double bond, they have the potential to provide olefin-containing polymers. In the same manner, some diene derivatives, such as 1,4-pentadiene, have a similar function.

From the polymeric structures illustrated in the Scheme 1, it is possible to think that the first synthetic approach to give a network structure, indicated above, can yield tholins as final products. Additionally, it is important to take into account that when multifunctional monomers are copolymerised, certain additional aspects must be considered. Thus, the kinetics and polymer architecture are strongly influenced by the presence of pendant double or triple bonds in the polymer chain. These bonds are involved in the following reactions (Dusek, 1998; Matsumoto, 1995) (Scheme 2): (a) Primary cyclisation (intramolecular cross-linking), occurring when a growing radical reacts with a double or triple bond of the same growing primary chain, creating loops or rings. (b) Secondary cyclisation (intramolecular cross-linking), occurring when a growing radical reacts with a double or triple bond of the same macromolecule but belonging to a different primary chain. (c) Cross-linking (intermolecular cross-linking), occurring when a growing radical reacts with a pendant double or triple bond belonging to another chain. A functionalisation of the remaining double or triple bonds can be accomplished subsequently.

The entire processes of the reactions previously described offer little control over the network structure, leading to a heterogeneous network macrostructure, as is well-known in the chemistry of free-radical polymerisation. However, it has many advantages over other polymerisation techniques, such as its versatility of monomer type, mild reaction conditions and tolerance of water (Moad and Solomon, 1995).

It is known that cross-linked polymeric materials can be studied for their properties and thus, significant conclusions regarding their heterogeneous structure can be drawn. Thus for example, further analysis dynamic mechanical measurements provides interesting information about the heterogeneous nature of the distribution of relaxation times of the segments of these networks polymers. Solid-state NMR spectroscopy and swelling techniques have also extensively used to gather evidence regarding the heterogeneity of these polymeric systems. A detailed study of the degree or extend of cross-linking (upon time of exposure to the spark) of these tholins will be described in the near future.

Fig. 3. XPS spectra of the N1(s) core level peak.

Table 2 X-ray photoelectron spectroscopy data for the tholins 1 and 2.

Group

BE (eV)

Tholin 1

Tholin 2

CH/C–C/–C@C CO/CN/C@O/C@N –N@/–N@C (heterocycles) NH–C@O (amides) NH (amines)/–CN (nitriles)

285 288 398 399 400

285 288 398.1 – 400.3

285.3 288.1 398.4 – 400.2

In order to understand the chemical nature of the hydrophobic tholins under study, in Scheme 1 (in this scheme, the stereochemistry and configurational aspects are not considered for simplicity) is shown the different homopolymers and the structure of a possible resulting copolymer that can be obtained from the distinct vinyl monomers and the acetylene derivatives found in a CH4 atmosphere for the experiments of prebiotic synthesis.

The incorporation of the polymeric main chain of a multifunctional compound, such as vinylacetylene, leads to the formation of a macromolecular system with vinyl and/or alkynyl substituents as side groups along the main chain. In addition, it is important to note that the different allenes found, such as propa-1,2-diene and 1,2-butadiene, play a significant role in the polymer chemistry, although they are not considered in this scheme for simplicity (Schuster and Coppola, 1984; Landor, 1982; Patai, 1980). Since al-

4.1.2. Influence of hydrogen and temperature on the nature of the tholins In general, the properties and elemental composition of the tholins depend on the composition of the initial gas mixture and the temperature (see e.g. Imanaka et al., 2004). Thus, it is well-known that the presence of hydrogen is determinant in the formation of amino acids in spark discharge experiments in atmospheres rich in CO and CO2 (Schlesinger and Miller, 1983). However, the presence of hydrogen is not critical to the formation of biomolecules in atmospheres containing CH4. The simulation experiments of Titan’s tholins in the absence of hydrogen show the production of organic molecules with biological interest in relatively good yields (e.g. McDonald et al., 1994). The formation of our hydrophobic tholins using CH4 as a carbon source seems to be independent of the presence of H2, since in its absence, the gas-phase precursors were the same. Further experiments are needed to determine the true role of hydrogen in the formation and composition of the tholins in atmospheres containing methane. To our knowledge, few efforts have been made towards this goal (Sekine et al., 2008).

On the other hand, temperature seems to play an important role in the formation of tholins. Whether Titan’s haze contains a significant amount of aromatic macromolecules is still an open question (Lebonnois et al., 2002). The formation of PAHs (polycyclic aromatic hydrocarbons) in Titan’s haze is related to the polymerisa-

Author's personal copy

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

677

Fig. 4. GC–MS chromatogram showing the oxidative cleavage products of the tholin 1.

Fig. 5. GC–MS chromatogram showing the gas-phase precursors of hydrophobic tholins from a based CH4 atmosphere after two hours of activation by spark discharges, using the method 1 for GC–MS. The chromatogram was registered in the range m/z from 35 to 350. The gas mixture showed is independently of the presence of H2 and/or water.

Author's personal copy

678

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

Fig. 6. GC–MS chromatogram showing the gas-phase precursors of hydrophobic tholins from a based CH4 atmosphere after two hours of activation by spark discharges, using the method 2 for GC–MS. The chromatogram was registered in the range m/z from 21 to 350.

R

R

R

R

n

R

R n

R = H, CH 3 R

R

R

R

3,4-addition

1,2-addition

n n

n 1,4-addition

n Scheme 1. Polymeric structures which can be originated from unsaturated compounds found in the gas-phase precursors of the hydrophobic tholins according to Figs. 5 and 6.

Scheme 2. Different reactions for polymeric chains containing double bonds.

tion of acetylene to poly-ynes, and subsequent formation of PAHs (Wilson and Atreya, 2003). The production of acetylene under the conditions of experiments 1 and 2 was low, and PAHs were not detected. Nevertheless, in experiments using the same conditions and atmospheres to produce tholins 1 and 2, but working below 0 °C, the formation of PAHs were detected by SPME/GC–MS. Additionally, a hydrophilic acetylene-based polymer was also collected (Menor-Salván et al., 2008). In this case, acetylene generation in a methane/nitrogen atmosphere and subsequent

polymerisation to PAHs and poly-ynes was favoured by the presence of water freeze-melt cycles. Therefore, there are competitive mechanisms for the formation of tholins depending on the temperature gradient, the interface processes and the formation of acetylene. Thus, part of Titan’s haze could be formed by aromatic structures in the region of its atmosphere where there is an adequate temperature gradient that favours the formation of acetylene, while in other regions other mechanisms could

Author's personal copy

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

preferentially form aliphatic hydrocarbons. Titan’s haze may be a mixture of tholins with different natures and distributions throughout the atmosphere according to the gradient of temperature and pressure. The formation of PAHs in the regions richer in acetylene would be favoured.

4.2. Insoluble HCN polymers and hydrophobic tholins HCN is a ubiquitous molecule in the universe and it is a significant product in prebiotic simulation experiments (see e.g. Stribling and Miller, 1987; Ferris and Hagan, 1984). It has been proposed that HCN polymers are important substances in the first stages of chemical evolution leading to the emergence of life (see e.g. Matthews and Minard, 2006; Chen and Chen, 2005; and Saladino et al., 2004 and internal references). In the literature, different models have been reported for the structure of HCN polymers (Umemoto et al., 1987; Ferris et al., 1981; Matthews and Moser, 1967; Völker, 1960). HCN polymers have been suggested as the main constituents of the complex organic products collected in spark discharge experiments (Matthews and Moser, 1967). Indeed, laboratory simulations of Jupiter’s and Titan’s tholins (Sagan and Khare, 1979; Khare et al., 1984; and McDonald et al., 1991) showed IR reflectance spectra that were similar to the spectra of HCN polymers (Matthews and Ludicky, 1986; Cruikshank et al., 1991). Coll et al. (1999) have indicated that the IR spectrum of poly-HCN prepared under anhydrous conditions (Völker, 1960) had a similar appearance to Titan’s tholins synthesised at low temperature; thus, poly-HCN could be the base of the tholins. However, McDonald et al. (1991) concluded that the IR similarities between the IR spectra of Jupiter’s tholins and HCN polymers were only due to a similar distribution of functional groups rather than a detailed structural identity. We applied our methodology to the characterisation of the insoluble HCN polymers (‘‘HCN black polymer”, obtained from filtration of the resulted polymerisation suspension and washed with water several times in a similar way to the used for the isolate of the tholins 1 and 2). The HCN polymer was prepared according to Borquez et al. (2005), and its elemental analysis was comparable to that previously reported in the bibliography (Eastman et al., 2003; Garbow et al., 1987; Umemoto et al., 1987), 36.5% C, 4.4% H, 40.8% N, 17.4% O. Considering the IR and the 13C NMR spectra (see Figs.

1d and 2c and Table 1) of the HCN polymers, we can say that our hydrophobic tholins were not analogous solids to the insoluble HCN polymers. Tholins 1 and 2 contained a high number of aliphatic carbons (very sharp features in the IR spectra at 2960, 2931, 2873 and at 1458, 1377 cm 1 indicates the presence of – CH2 and –CH3 groups as well as the wide resonance centred at 34 ppm in the solid-state 13C NMR spectra), indicating the formation of hydrocarbon chains. Conversely, the insoluble HCN polymer presented a totally different structure. No aliphatic carbons were observed in the 13C NMR spectrum, which seems to indicate a structure likely formed by heterocycles containing nitrogen.

5. Conclusions In simulation experiments with an external energy source and a CH4/N2-based atmosphere, it is possible to obtain tholins with different natures in relation to their hydrophilic or hydrophobic character. This character depends directly on the experimental conditions such as temperature, but also on the water interfacial processes (liquid–gas and solid–gas). We observed that physic state of the water is notable important in the nature of the hydrophilic tholins but it seems to have other role in the formation of the hydrophobic tholins. The tholins are not homogeneous substances, and they cannot be considered as true polymers. The formation of aliphatic hydrocarbon chains or aromatic compounds depends

679

on the gas-phase precursors and the competitive mechanisms favoured by the experimental conditions.

Hydrophobic tholins 1 and 2 are materials constituted of hydrocarbon chains containing unsaturations and polar groups. These tholins are formed by means of polymerisation or the addition of unsaturated units during the first stages of the reaction in the gas phase. Subsequently, the chains can react with each other to form network structures. On the other hand, hydrophobic tholins 1 and 2 contain a notable amount of non-volatile organic matter.

Moreover, they can release important bioorganics such as dicarboxylic acids. Therefore, it would be interesting to investigate how these hydrophobic substances could evolve under geological conditions.

Indeed, under the same reaction conditions, it is possible to collect hydrophilic and hydrophobic tholins whose natures are different. The study of all of these complex substances involves the use of complementary spectroscopy and analytical techniques, and at least a previous separation based on water solubility.

Acknowledgments The authors have used the research facilities of Centro de Astrobiología (CAB) and have been supported by Instituto Nacional de Técnica Aeroespacial ‘‘Esteban Terradas” (INTA) and the project AYA2006-15648-C02-02 of the Ministerio of Educación y Ciencia (Spain). We thank M.T. Fernández for obtaining the IR spectra and the DSC assays, J. Sobrado for the XPS measurement, I. Sobrados from ICMM (Instituto de Ciencia de Materiales de Madrid. CSIC. Spain) for the measurement the 13C NMR spectra and D. Nna Mvondo for her help in the analysis of the gas phase by GC–MS. References Borquez, E., Cleaves, H.J., Lazcano, A., Miller, S.L., 2005. An investigation of prebiotic purine synthesis from the hydrolysis of HCN polymers. Orig. Life Evol. Biosph.

35, 79–90.

Briones, C., Martín-Gago, J.A., 2006. Nucleic acids and their analogs as nanomaterials for biosensor development. Curr. Nanosci. 2, 257–273.

Chen, Q.W., Chen, C.L., 2005. The role of inorganic compounds in the prebiotic synthesis of organic molecules. Curr. Org. Chem. 9, 989–998.

Cleaves, H.J., Chalmers, J.H., Lazcano, A., Miller, S.L., Bada, J.L., 2008. A reassessment of prebiotic organic synthesis in neutral planetary atmospheres. Orig. Life Evol.

Biosph. 38, 105–115.

Coll, P., Coscia, D., Smith, N., Gazeau, M.-C., Ramírez, S.I., Cernogora, G., Israel, G., Raulin, F., 1999. Experimental laboratory simulations of Titan’s atmosphere: Aerosol and gas phase. Planet. Space Sci. 47, 1331–1340.

Cruikshank, D.P., Hartmann, W.K., Tholen, D.J., Allamandola, L.J., Brown, R.H., Matthews, C.N., Bell, J.F., 1991. Solid C„N bearing material on outer Solar System bodies. Icarus 94, 345–353.

Dusek, K., 1998. Network formation involving polyfunctional polymer chains in polymer network. In: Stepto, R.F.T. (Ed.). Blackie Academic & Professional, Thomson Science, London (Chapter 3).

Eastman, M.P., Helfrich, F.S.E., Umantsev, A., Porter, T.L., Weber, R., 2003. Exploring the structure of hydrogen cyanide polymer by electron spin resonance and scanning force microscopy. Scanning 25, 19–24.

Ferris, J.P., Hagan, W.J., 1984. HCN and chemical evolution: The possible role of cyano compounds in prebiotic synthesis. Tetrahedron 40, 1093–1120.

Ferris, J.P., Edelson, E.H., Auyeung, J.M., Joshi, P.C., 1981. Structural studies on HCN oligomers. J. Mol. Evol. 17, 69–77.

Garbow, J.R., Schaefer, J., Ludicky, R., Matthews, C.N., 1987. Detection of secondary amides in HCN polymers by dipolar rotational spin-echo 15N NMR. Macromolecules 20, 305–309.

Imanaka, H., Khare, B.N., Elsila, J.E., Bakes, E.L.O., McKay, C.P., Cruikshank, D.P., Sugita, S., Matsui, T., Zare, R.N., 2004. Laboratory experiments of Titan tholin formed in cold plasma at various pressures: Implications for nitrogencontaining polycyclic aromatic compounds in Titan haze. Icarus 168, 344–366.

Kannurpatti, A.R., Anderson, K.J., Anseth, J.W., Bowman, C.N., 1997. Use of living radical polymerization to study the structural evolution of highly crosslinked polymer networks. J. Polym. Sci. Part B: Polym. Phys. 35, 2297–2307 (and references herein).

Khare, B.N., and 10 colleagues, 1984. The organic aerosols of Titan. Adv. Space Res. 4, 59–68.

Khare, B.N., Sagan, C., Ogino, H., Nagy, B., Er, C., Schram, K.H., Arakawa, E.T., 1986.

Amino acids derived from Titan tholins. Icarus 68, 176–184.

Kimberley, M., 2002. Oxidative cleavage of an alkene with catalytic ruthenium tetraoxide; 2-acylphenylacetic acid. SyntheticPage 186.

Author's personal copy

680

M. Ruiz-Bermejo et al. / Icarus 204 (2009) 672–680

Koike, T., Kaneko, T., Kobayashi, K., Miyakawa, S., Takano, Y., 2003. Formation of organic compounds from simulated Titan atmosphere: Perspectives of the Cassini mission. Biol. Sci. Space 17, 188–189.

Landor, S.R. (Ed.), 1982. The Chemistry of Allenes. Academic, London.

Lebonnois, S., Bakes, E.L.O., McKay, C.P., 2002. Transition from gaseous compounds to aerosols in Titan’s atmosphere. Icarus 159, 505–517.

Matthews, C.N., Moser, R.E., 1967. Peptide synthesis from hydrogen cyanide and water. Nature 215, 1230–1234.

Matsumoto, A., 1995.

Free-radical cross-linking polymerization and copolymerization of multivinyl compounds. Adv. Polym. Sci. 123, 41–80.

Matthews, C.N., 1995. Hydrogen cyanide polymers: From laboratory to space.

Planet. Space Sci. 43, 1365–1370.

Matthews, C.N., Ludicky, R.A., 1986. The dark nucleus of Comet Halley: Hydrogen cyanide polymers. In: Battrick, B., Rolfe, E.J., Reinhard, R., Noordwijk, E.S.A. (Eds.), 20th ESLAB Symposium on the Exploration of Halley´s Comet, vol. 1, pp.

273–277.

Matthews, C.N., Minard, D., 2006. Hydrogen cyanide polymers, Comets and the origin of life. Faraday Discuss. 133, 393–401.

McDonald, G.N., Khare, B.N., Thompson, W.R., Sagan, C., 1991. CH4/NH3/H2O spark tholin: Chemical analysis and interaction with jovian aqueous clouds. Icarus 94, 354–367.

McDonald, G.N., Thompson, W.R., Heinrich, M., Khare, B.N., Sagan, C., 1994.

Chemical investigation of Titan and Triton tholins. Icarus 108, 137–145.

McGuigan, M., Waite, J.H., Imanaka, H., Sacks, R.D., 2006. Analysis of Titan tholin pyrolysis products by comprehensive two-dimensional gas chromatography– time-of-flight mass spectrometry. J. Chromatogr. A 1132, 280–288.

Menor-Salván, C., Ruiz-Bermejo, M., Osuna-Esteban, S., Muñoz-Caro, G., Veintemillas-Verdaguer, S., 2008. Synthesis of polycyclic aromatic hydrocarbons and acetylene polymers in ice: A prebiotic scenario. Chem.

Biodiv. 5, 2729–2739.

Moad, G., Solomon, D.H., 1995. The Chemistry of Free Radical Polymerization.

Pergamon, Oxford, England.

Patai, S., 1980. The Chemistry of Ketenes, Allenes, and Related Compounds. Wiley, New York.

Pietrogrande, M.C., Coll, P., Sternberg, R., Szopa, C., Navarro-Gonzalez, R., VidalMadjar, C., Dondi, F., 2001. Analysis of complex mixtures recovered from space

missions. Statistical approach to the study of Titan atmosphere analogues (tholins). J. Chromatogr. A 939, 69–77.

Ruiz-Bermejo, M., Menor-Salván, C., Osuna-Esteban, S., Veintemillas-Verdaguer, S., 2007. Prebiotic microreactors: A synthesis of purines and dihydroxy compounds in aqueous aerosol. Orig. Life Evol. Biosph. 37, 123–142.

Ruiz-Bermejo, M., Menor-Salván, C., Mateo-Martí, E., Osuna-Esteban, S., MartínGago, J.A., Veintemillas-Verdaguer, S., 2008. CH4/N2/H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus 198, 232–241.

Sagan, C., Khare, B., 1979. Tholins: Organic chemistry of interstellar grains and gas.

Nature 277, 102–107.

Sagan, C., Khare, B.N., Thompson, W.R., McDonald, G.D., Wing, M.R., Bada, J.L., VoDinh, T., Arakawa, E.T., 1993. Polycyclic aromatic hydrocarbon in the atmosphere of Titan and Jupiter. Astrophys. J. 414, 399–405.

Saladino, R., Crestini, C., Costanzo, G., DiMauro, E., 2004. Advances in the prebiotic synthesis of nucleic acids bases: Implications for the origin of life. Curr. Org. Chem. 8, 1425–1443.

Schlesinger, G., Miller, S.L., 1983. Prebiotic synthesis in atmospheres containing CH4, CO and CO2. J. Mol. Evol. 19, 376–382.

Schuster, H.F., Coppola, G.M., 1984. Allenes in Organic Synthesis. Wiley, New York.

Sekine, Y., Imanaka, H., Matsui, T., Khare, B.N., Bakes, E.L.O., McKay, C.P., Sugita, S., 2008. The role of organic haze in Titan’s atmospheric chemistry I. Laboratory investigation on heterogeneous reaction of atomic hydrogen with Titan tholin.

Icarus 194, 186–200.

Stribling, R., Miller, S.L., 1987. Energy yields for hydrogen cyanide and formaldehyde synthesis: The HCN and amino acid concentrations in the primitive ocean. Origins Life 17, 261–273.

Takano, Y., Tsuboi, T., Kaneko, T., Kobayashi, K., Marumo, K., 2004. Pyrolysis of highmolecular-weight complex organics synthesized from a simulated interstellar gas mixture irradiated with 3 MeV proton beam. Bull. Chem. Soc. Jpn. 77, 779– 783.

Umemoto, K., Takahashi, M., Yokota, K., 1987. Studies on the structure of HCN oligomers. Origins Life 17, 283–293.

Völker, T., 1960. Polymeric hydrogen cyanide. Angew. Chem. 72, 379–384.

Wilson, E.H., Atreya, S.K., 2003. Chemical sources of haze formation in Titan’s atmosphere. Planet. Space Sci. 51, 1017–1033.

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.1 (1-10) Icarus ••• (••••) •••–•••

Contents lists available at ScienceDirect

Icarus www.elsevier.com/locate/icarus

CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins Marta Ruiz-Bermejo a,∗ , César Menor-Salván a , Eva Mateo-Martí a , Susana Osuna-Esteban a , José Ángel Martín-Gago a,b , Sabino Veintemillas-Verdaguer a,b a

Centro de Astrobiología [Consejo Superior de Investigaciones Científicas-Instituto Nacional de Técnica Aeroespacial (CSIC-INTA)], Carretera Torrejón-Ajalvir, Km. 4.2, E-28850 Torrejón de Ardoz, Madrid, Spain b Instituto de Ciencia de Materiales de Madrid (CSIC), C/ Sor Juana Inés de la Cruz, 3, Cantoblanco, E-28049 Madrid, Spain

a r t i c l e

i n f o

a b s t r a c t

Article history: Received 25 April 2008 Revised 18 July 2008

Tholins are complex organic materials produced by irradiation of several carbon and nitrogen rich atmospheres. They could have played a key role in the origin of life on Earth because their ability to release important bioorganics, which are nowadays present in proteins, nucleic-acid bases and other important biostructures. Usually, the yield of these compounds is higher after acid hydrolysis, however little is known about the structure and chemical composition of the tholins. In this work, we propose the use of different spectroscopic and separation techniques, which are not usually applied in this field, in order to obtain complete information about the tholin structure and behavior. Two different simulation experiments of prebiotic synthesis were carried out in CH4 /N2 /H2 atmosphere out from spark discharge activation of aqueous aerosols and liquid water, respectively. In both cases, a hydrophilic tholin and a hydrophobic tholin were obtained. Herein, we report the application of this methodology to our hydrophilic tholins and we review, briefly, some astrobiological aspects related to these complex substances. © 2008 Elsevier Inc. All rights reserved.

Keywords: Prebiotic chemistry Prebiotic environments Organic chemistry Spectroscopy

1. Introduction Nowadays, the term tholin is mainly used to designate the haze in the stratosphere of Saturn’s moon, Titan, and the solids obtained in simulation experiments of Titan’s atmospheric processes. In a general way, tholins are complex organic materials obtained by irradiation of atmospheres that contain CO, CO2 or CH4 , as carbon sources, and N2 or NH3 , as nitrogen sources (Sagan et al., 1993).

Taking into account the possible sources of carbon and nitrogen and the potential external energy sources (i.e. spark discharges, UV irradiation, cold plasma, high energy particles or soft X-rays), the possibilities to prepare tholins are numerous. Moreover, the structure and properties of tholins depend on the energy source employed, the ratios C/N and C/H of the initial gas mixture, the pressure and the temperature (Imanaka et al., 2004; Coll et al., 1999, 1998). Therefore, the number of possible tholins with different properties is very large. Nonetheless, several aspects related to the formation of tholins during simulation experiments can help us to understand the chemical atmospheric evolution of the planets and the deposition and concentration of organic compounds over their surfaces.

*

Corresponding author. Fax: +34 91 520 6410.

E-mail address: ruizbm@inta.es (M. Ruiz-Bermejo).

0019-1035/$ – see front matter doi:10.1016/j.icarus.2008.07.008

©

It has been proposed that tholins could have played an important role in the origin of life on Earth (Sagan and Khare, 1979). Tholins usually provide amino acids and, in some cases, nucleic-acid bases, hydroxy acids, carboxylic acids and other organic molecules (e.g. Koike et al., 2003; McDonald, 1994, 1991; Khare et al., 1986). In most cases, amino acids and nucleic-acid bases were only detected after acid hydrolysis of tholins, and the nature of their precursors is still unclear. Mainly, aminonitriles or short peptide-like molecules have been proposed as plausible parent compounds for amino acids, and several mechanisms have been indicated for their formation (Miyakawa et al., 1999; Matthews and Ludicky, 1992; Khare et al., 1989a; Thompson and Sagan, 1989; Miller, 1957). Indeed, after pyrolysis, the tholins produced a great diversity of organics such as lineal and branched hydrocarbons (saturated and unsaturated), aromatic hydrocarbons, nitriles and heterocycles containing nitrogen (see e.g. McGuigan et al., 2006; Takano et al., 2004a; Pietrogrande et al., 2001; Coll et al., 1999).

In spite of the many efforts made to elucidate the structure of tholins (see, for example, McDonald et al., 1994; McDonald et al., 1991), their exact nature remains in shadows. For instance, whether the products obtained in the Titan simulation experiments [using mixtures of N2 (99–90%), CH4 (10–01%), energy source: electric discharges, UV irradiation, cold plasma] are long chains, aromatics or a mixture of both, is in debate (Lebonnois et

2008 Elsevier Inc. All rights reserved.

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

2

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.2 (1-10) M. Ruiz-Bermejo et al. / Icarus ••• (••••) •••–•••

al., 2002). To our understanding, the main difficulty in the structural and analytical study of tholins is the lack of a general, systematic and complete methodology accepted by the entire scientific community that allows for the comparison of tholins obtained under different experimental conditions.

In this work, two different simulation experiments of prebiotic synthesis were carried out using a CH4 /N2 /H2 atmosphere and spark discharges either in aqueous aerosols or in liquid water (experiment 1 and experiment 2, respectively). In both experiments, we obtained a hydrophilic tholin and a hydrophobic tholin (Ruiz-Bermejo et al., 2007a). The tholins obtained were analyzed in the same way using a set of complementary techniques and assays. Among these, we coordinated the use of standard spectroscopic techniques, such as FT-IR (Fourier Transform-Infrared Spectroscopy), 13 C NMR (13 C Nuclear Magnetic Resonance) and UV–vis (Ultraviolet–visible spectroscopy), along with others not usually used in these analyses, such as XPS (X-ray Photoemission Spectroscopy), which, to our knowledge, only has been reported by Tran et al. (2003). Due to the heterogeneous nature of the tholins, these techniques were used in combination with other powerful analytical tools, such as GC–MS (Gas Chromatography–Mass Spectrometry), SPME/GC– MS (Solid Phase Microextraction/Gas Chromatography–Mass Spectrometry) and HPLC (High Performance Liquid Chromatography), as well as biochemical techniques, such as electrophoresis and ultrafiltration-dialysis. Finally, classic qualitative assays and other techniques, such as elemental analysis, thermal analysis and mass spectrometry, were employed in order to get complementary information. Herein, we report the application of this approach to the study of the hydrophilic tholins prepared by us and extending our previous work (Ruiz-Bermejo et al., 2007a). Moreover, we compare our data with the previous results from other authors in a brief review pointing out the astrobiological aspects related to the main organic molecules detected in tholins and their possible formation.

2. Experimental 2.1. Production of tholins The details about the preparation and separation of the hydrophilic and hydrophobic tholins are given in Ruiz-Bermejo et al.

(2007a).

2.2. Elemental analysis Elemental C, H and N analyses were performed at Unidad de Instrumentación Científica (Universidad de Alcalá de Henares, 28801 Alcalá de Henares, Spain) using a “CHN-O-rapid” (Heraeus) elemental analyzer.

2.3. Standard spectroscopy techniques IR spectra were obtained using a Nexus Nicolet FT-IR spectrometer. The spectra were obtained in CsI pellets on the reflectance mode of operation.

13 C CP MAS-NMR spectra were obtained in a Bruker Advance 400 spectrometer using a standard cross-polarization pulse sequence.

Samples were spun at 10 kHz. Spectrometer frequency was set to 100.62 MHz. A contact time of 1 ms and a period between successive accumulations of 5 s were used. The number of scans was 1600. Chemical shift values were referenced to TMS. Standard 13 C NMR spectra were obtained using a Mercury 400 Varian NMR spectrometer. The spectra were recorded in D2 O. The number of scans was 50000. Chemical shift values were referenced to TMS.

UV–vis spectra were obtained using an Agilent 8453 spectrophotometer. All spectra were recovered in H2 O. The assignment of adsorption bands in the UV–vis spectra (190–750 nm) was made using previous data reported in the literature (McDonald et al., 1996, 1994, 1991; Simionescu et al., 1974).

2.4. XPS spectroscopy The XPS analysis was carried out with a hemispherical electron analyzer (Phoibos 150 MCD) using a MgK α X-ray source (1253.6 eV). The sample, tholin-CsI pellet, was introduced in an ultra high vacuum chamber with a base pressure in the 1 × 10−9 mbar pressure range. Experiments were carried out at room temperature. The XPS spectra of the different regions were decomposed in Gaussian curves components. We applied the criterion of using the lowest number of components for the analysis of the N(1s) and C(1s) peaks. The XPS assignment for the different nitrogen chemical groups into the tholins was made recording XPS spectra of simple pattern molecules as a reference. These reference samples such as alanine (amine species), adenine (N-cycles and amine species) and peptides acids (amides species), where measured in the same experimental conditions to calibrate the binding energies of each nitrogen group involved with different chemical species.

2.5. Reactivity assays Classical reactivity tests are often ruled out in favor of modern spectroscopic techniques, but in this work, we proposed that some of them were especially relevant if results were negative. The tests for peptides were performed according to the method of Bradford (1976) and the copper–bicinchoninic acid (BCA) assay according to Stoscheck (1990). The presence of ammonia was detected by the colorimetric method using Nessler reagent. Finally, the formation of carbon dioxide during the experiments was confirmed by passing the remnant gas in the reactor through a saturated solution of barium hydroxide.

2.6. HPLC separation of the tholin components The tholins were separated by reverse-phase HPLC and normalphase HPLC. In the first case, a Kromasil 100 C18 5 μm 25 × 0.46 column was used with the following conditions: Solvent A: MeOH (methanol), Solvent B: 0.1% TFA (trifluoroacetic acid) in water, 0 min 0% B (100% A), 15 min 50% B (50% A), 19 min 0% B (100% A), 20 min 0% B (100% A). The flow was 1.75 mL/min, the column temperature was 45 ◦ C and the chromatogram was registered at 220 nm. In the second case, a BetaBasic-8 (Thermo HypersilKeystone) 5 μm 100 × 0.46 column was used and the mobile phases were CH3 CN (acetonitrile), isocratic method. The flow was 1.25 mL/min, the column temperature was 35 ◦ C and the chromatogram was registered at 206 nm.

2.7. Molecular weight estimation The tholins were fractionated by an YM-3 Microcon centrifugal filter device (Millipore Corp.) to retain molecules above 3 kDa. Subsequent separations were completed by dialysis using MWCO 1 kDa “B” size and MWCO 500 Da “B” size cellulose acetate membranes (Harvard Apparatus).

Electrophoretic techniques were applied under the assumption that the hydrophilic tholins could behave as protein analogs, and consequently, could migrate under an electric field at different velocities depending upon their molecular weight. 40 μg of sample were loaded onto 16% sodium dodecyl sulfate–polyacrylamide (SDS–PAGE) gels in non-reducing conditions, according to the

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.3 (1-10) CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins

method of Laemmli (1970). Samples were dissolved in sample buffer containing 0.05 mol/L Tris (pH 6.8), 10% glycerol and 3% SDS. Molecular weight standards were myosin (212 kDa), alphamacroglobulin (170 kDa), beta-galactosidase (146 kDa), transferrine (76 kDa), glutamate deshydrogenase (53 kDa), and cytochrome c (8 kDa). Electrophoresis was carried out with 100 V at room temperature. After that, the gel was fixed in 50% methanol/12% acetic acid/0.05% formaldehyde. The gel was pretreated with 0.2 mg/mL sodium thiosulfate for 1 min and washed three times with water.

It was incubated with a solution containing 0.02 mg/mL silver nitrate and 0.075% formaldehyde and washed twice with water. The gel was developed with a solution containing 0.06 g/mL sodium carbonate, 0.004 mg/mL sodium thiosulfate and 0.05% formaldehyde and washed twice with water. The development was stopped by the addition of 50% methanol/12% acetic acid.

Measurements of electrophoretic mobility were made by dynamic light scattering (DSL) using a Zetasizer Nano ZS (Malvern Instruments). Each DLS measurements correspond to the average of four 1 min measurements. Autocorrelation functions were fit to a cumulated analysis (Gaussian distribution of particle size). Although, a fit to an intensity-weighted two-population size distribution model provided similar trends between samples (not shown), the tholin samples possessed an average hydrodynamic diameter of 104.4 nm (lysozme 199.8 nm). Samples were measured at a constant temperature of 25 ◦ C in polystyrene cells using water as a dispersant. Values of z-potential and electrophoretic mobility of the samples were determined in a water solution at pH 7 and Tris–HCl–SDS 0.1% at pH 8.8 in order to determine if Laemmli SDS–PAGE electrophoresis was a valid technique for the molecular weight estimation and purification of tholins of MW >1 kDa. Mass spectrometry analyses were performed at the Servicio Interdepartamental de Investigación, SIdI (Universidad Autónoma de Madrid, Ciudad Universitaria de Cantoblanco, 28049 Madrid, Spain). MALDI mass spectra were recovered in a Reflex III (Bruker) using α -cyane-4-hydroxycinamic acid (ACC) and 2,5-dihydroxybenzoic acid (DHB) as the matrix for the hydrophilic tholins. ESI mass spectra were recovered in an API Q Star Pulsar I (Applied Biosystems).

dimethylpolysiloxane, 30 m × 0.25 mm i.d., 0.25 μm film) and using He as the carrier gas. The temperature was raised from 40 ◦ C (4 min) to 150 ◦ C at a rate of 15 ◦ C/min, held for 2 min, 150 ◦ C to 255 ◦ C at 5 ◦ C/min, held for 15 min, and 255 ◦ C to 300 ◦ C at 10 ◦ C/min, and held for 1 min. The mass spectrometer was operated under EI+ mode at an ionization energy of 70 eV, m/ z range 30–600 and transfer line at 300 ◦ C. 3. Results The empirical formulas of the hydrophilic tholins (from here to the end of the text we will call as tholin 1 the hydrophilic tholin from experiment 1 and as tholin 2 the hydrophilic tholin from experiment 2) determined by elemental analysis are C9 H18 N5 O5 and C9 H16 N5 O5 , respectively. Both tholins are highly soluble in water and partially soluble in methanol and ethanol.

3.1. Determination of functional groups The IR spectrum of tholin 1 showed features which can be associated to the O–H stretching mode, to the N–H stretching mode, to the C=O bending mode, to the R–C≡N stretching mode and to the asymmetric and symmetric stretching and bending modes for –CH3 and –CH2 groups (see Table 1 for the details in the assignment of the bands). Some authors (McDonald et al., 1991; Khare et al., 1986) assigned, tentatively, the band at 1530–1560 cm−1 to a possible amide II or N–H bending mode indicating the possibility of peptides in the spark tholins. In the tholin 1, this feature was observed as a weak band at 1541 cm−1 . The IR spectrum of acid hydrolyzed tholin 1 did not show the feature at 2250 cm−1 or the feature at 1541 cm−1 , and the band related to –COOH was enhanced (continuous signal among 3350 and 1700 cm−1 ) confirming Table 1 Characteristic frequency of IR absorption spectra obtained from the tholins 1 and 2 and the fractions from hydrolyzed tholin 1 Tholin 1

Tholin 2

Hydrolyzed tholin 1

3448 (sh)

2.8. Thermal analysis Differential scanning calorimetry (DSC) measurements on selected samples were made using a DSC 2920 TA Instrument. Temperature and heat flow were calibrated with indium and an empty pan was used as a reference. We used a hermetically sealed aluminum pan. Sample and reference pans were heated from ambient room temperature to 500 ◦ C at 5 ◦ C/min.

3

3321 (b, s) 3207 (b, s)

3347 (b, s) 3223 (sh)

2977 2933 2880 2250

2971 2936 2878 2248

(v w) (v w) (sh) (w)

(w) (w) (w) (w)

3027 (sh) 2974 (sh) 2807 (sh)

Vibrational groups identity N–H stretching (single bond) or NH2 asym. stretching or O–H stretching O–H stretching N–H stretching or overtone NH2 bending C=CH stretching or CH aromatic C–H stretching (CH3 asym.)

C–H stretching (CH2 ) C–H stretching (CH3 sym.)

R–CN stretching or R–C≡C–R C≡C or conjugated nitriles

2143 (w)

2.9. Amino acids, heterocycles, hydroxy- and carboxylic acids analysis

by GC–MS, and HPLC The amino acids, heterocycles, hydroxy acids and carboxylic acids analysis were done by GC–MS and HPLC according to RuizBermejo et al. (2007a).

2.10. Organic analysis by SPME-GC/MS The raw mixture was analyzed by solid phase microextraction (SPME) coupled with GC–MS. The samples were heated in a closed vial with a septum at 180 ◦ C for 45 min. A 100 μm CARdimethylpolysiloxane (CAR-PDMS) fiber was then exposed to the headspace, keeping the sample at the same temperature for a further 45 min. Analytes on the fiber were then thermally desorbed in the injection port of a Perkin–Elmer Autosystem XL-Turbomass GC–MS instrument at 290 ◦ C for 4 min (splitless mode). The analysis was performed using a capillary column (5% diphenyl–95%

1671 (v s)

1728 (sh) 1672 (v s)

1627 (s) 1541 (w)

1615 (sh) 1554 (sh)

1452 (w)

1452 1410 1382 1234 1199 1083

1379 1234 1207 1096

(b, m) (w) (v w) (w)

(sh) (m) (m) (w) (w) (b, m)

998 (sh)

994 (w)

776 (w)

782 (w)

2006 (m) 1735 (v s)

1623 (sh)

1409 (s) 1212 (s) 1091 (m) 989 (w) 859 (w)

C=C=N stretching (asym.)

C=O stretching C=O bending or N–C=C, O–C=C or C=C (stretching) or CH=N–N=CH N–H bending N–H bending (possible amide) or aromatics (quadrant stretch) or C=C (+ contribution C=N, N=N) CH2 asym. bending CH3 –C=O, CH3 –C=C CH3 sym. bending C–O st (arc–OH) C–C/C–N stretching C–O vibrational mode C–O vibrational mode or C=C=st si Vinyl C–H bending or C–N–C (saturated heterocycles) C–N stretching (aliphatic) Aromatics/N–H bending

The frequency is expressed in cm−1 . Intensity code: (b) = broad, (m) = medium, (s) = strong, (v s) = very strong, (sh) = shoulder, (w) = weak, (v w) = very weak.

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.4 (1-10) M. Ruiz-Bermejo et al. / Icarus ••• (••••) •••–•••

4

the presence of –C≡N groups and amides (see Table 1 and Fig. 1).

In addition, this enhancement also could be related to an ester groups.

The IR spectrum of tholin 2 was very similar to tholin 1, but the band related to the –COOH group was less intense (see Table 1).

The solid-state 13 C NMR spectrum of the tholin 1 showed four groups of broad resonances about 166, 126, 78 and 47 ppm. In order to resolve these bands, the spectrum was recorded in a solution of deuterium oxide, and in this case, the previous broad resonances could be resolved. As a result, in the 13 C NMR (D2 O, 100 MHz) spectrum of hydrophilic tholin 1, a great number of resonances were observed but as a first approximation, it was possible to group all resonances in six main types of carbons (see Table 2). All these resonances were consistent with the observed features in the IR spectrum. It was remarkable that resonances at 210–190 ppm, related with ketones and aldehydes, were absent.

The solid-state 13 C NMR spectrum of the tholin 2 was similar to

Fig. 1. Transmission FTIR spectra. (a) Tholin 1. (b) Hydrolyzed tholin 1.

Table 2 Functional groups assignment for the tholins 1 and 2 from

13

tholin 1, but the number of resonances related to carboxylic acids was smaller, see Table 2.

The UV–vis spectra of the hydrophilic tholins 1 and 2 are shown in Fig. 2. These tholins present a maximum absorption close to 200 nm that can be attributed to conjugated double bonds, to α , β unsaturated nitriles, to carboxylic acids and to carbonyl groups.

Fig. 2. Absorbance spectra. (a) Tholin 1. (b) Tholin 2. Solid line: registered in water.

Dashed line: registered in methanol. Dot line: registered in ethanol.

C NMR (D2 O, 100 MHz) spectra

Carbon type

Shift (δ ) (ppm) Tholin 1

Tholin 2

Carboxylic, amide, ester (–COOH, –CONHR , –COOR)

184.12, 184.72, 181.18, 180.80, 179.93, 178.20, 177.38, 174.61, 173.90, 173.26, 172.99, 171.79, 171.02, 167.80, 166.79, 166.33, 165.00, 164.67, 163.33, 162.79, 161.04, 160.62

183.25, 179.98, 179.87, 179.22, 179.10, 171.78, 170.97, 167.84, 167.80, 166.31, 164.63, 162.84, 160.65

Nitrile (–C≡N)

119.36, 119.32, 118.26, 118.22, 117.89, 117.85, 117.27, 117.19, 116.93, 116.45, 115.63, 110.07

120.92, 120.35, 119.24, 119.06, 118.60, 118.52, 118.41, 117.84, 117.77, 117.21, 117.14, 115.63

Hydroxyle (–C–OH)

83.38, 82.60, 81.90, 76,58, 73.71, 68.49, 68.02, 67.53, 66.88

82.62, 81.92, 76.57, 73.70, 68.45, 67.53

Amine –(C–NHR )

62.49, 62.64, 61.88, 61.64, 61.30, 60.72, 58.84, 57.71, 49.73, 47.96, 47.93, 43.83, 42.39, 41.56, 40.75

62.64, 61.24, 60.66, 59.92, 57.61, 56.07, 54.70, 50.87, 49.24, 49.58, 47.89, 44.39, 43.76, 43.29, 41.52, 40.76

–CH2 –, –CH–

36.83, 36.53, 36.24, 35.99, 35.73, 35.05, 30.18, 28.64, 28.55, 27.96, 27.72, 26.29, 23.04, 21.66, 21.45, 21.66, 21.45, 20.68, 20.18, 19.73, 18.41, 17.80, 17.16

37.40, 36.48, 36.22, 35.91, 35.64, 34.77, 34.48, 28.58, 27.66, 25.68, 25.57, 23.79, 21.60, 21.38, 20.12, 19.70, 18.37, 18.14, 17.76, 16.08

–CH3

13.30, 9.98, 9.50, 9.12

13.11, 9.48, 9.32, 9.21, 7.82

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.5 (1-10) CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins

Fig. 3. XPS Spectra of the N1(s) core level peak of tholins 1 and 2. The experimental data are presented by points, overlapped with a continuous curve corresponding to the best fit obtained, the different components used for the fit are represented by dashed curves.

Table 3 X-ray photoelectron spectroscopy data for the tholins 1 and 2 Group

BE (eV)

Tholin 1

Tholin 2

CH/C–C/–C=C CO/CN/C=O/C=N –N=/–N=C (heterocycles) NH–C=O (amides) NH (amines)/–CN (nitriles)

284–285 287 397–398 399 400

285.1 288.4 397.2 398.6 400

285.6 287.9 – – 399.9

A little hypsochromic shift with the polarity of solvent was observed, suggesting a material with net dipolar momentum.

The C1(s) XPS core level peak (data not shown) of both hydrophilic tholins 1 and 2 showed two components that were assigned to C–H, C–C and C=C groups, and to carbonyl, carboxylic and C=N bonds, respectively. Fig. 3 shows the XPS core-level spectra of N1(s) peaks and the components of the tholins 1 and 2 samples. The N(1s) peak of the hydrophilic tholin 1 was resolved in three components, which were assigned to the N cycles [unsaturated chemical bonds (–N=)], amide groups and amine groups plus nitriles, respectively. However, in the case of tholin 2, the nitrogen peak shows a unique component that was assigned to the amines plus nitriles compounds (see Table 3 for the details in the assignment).

Functional group assays were performed with tholins 1 and 2.

The Nessler assay for the determination of ammonium cations and the Ba(OH)2 assay for the determination of carbonate anions were positive. These results were in agreement with the final pH of the reaction raw products (pH ∼ 8.7, in both cases). With respect to the peptide bonds, the BCA assay was positive, but not conclusive, because it merely indicated the presence of reductive groups capable to reduce Cu(II) to Cu(I) at a pH 11. However, the Bradford assay was negative for tholins 1 and 2. Therefore, the presence of peptides could be safely excluded in our tholins.

3.2. Fractionation of the tholins Tholins 1 and 2 were analyzed by reverse-phase HPLC and normal-phase HPLC (Fig. 4) in order to separate their possible components according to their polarities. The HPLC chromatograms seemed to indicate that tholin 1 and tholin 2 are similar complex

5

substances. The differences in polarities of the different fractions were reflected in the UV–vis spectra (see Table 4). The absorptions in the range of 240–290 nm are related to conjugated double bonds or aromatic structures, thus, the tholin 2 seems to present a higher amount of unsaturated bonds than the tholin 1.

The fractionation of the tholins according to the molecular weight by ultrafiltration and dialysis, using centrifugal filter devices and cellulose acetate membranes for biodialysis, indicated that the aqueous aerosols improved the formation of lighter materials (Table 5).

In other attempt to estimate the molecular weight of the tholins, electrophoretic techniques were tested. The measurement

by DLS of the fraction of tholin separated by ultrafiltration at a cut-off value of 3 kDa in the presence of SDS at a pH 8.8 showed an increase of electrophoretic mobility versus the value obtained in the water solution (Table 6).

The values of the electrophoretic mobility in SDS–PAGE conditions for the tholin samples were similar to the values of a real protein, which was used as a control. The increase of electrophoretic mobility and the differences in mobility in water for different samples suggest a binding of the anionic detergent SDS to the tholin structure, impairing a negative charge to the complex.

The possibility of SDS binding to the tholin structure and the similar values of mobility in Laemmli’s SDS–PAGE conditions open up the possibility to use this electrophoretic technique in purification and molecular weight estimation of tholins with molecular weights over 1 kDa. Regarding this result, tholins 1 and 2 were separated

by SDS–PAGE electrophoresis. Overall, we confirmed that the electrophoretic behavior of the hydrophilic tholins was comparable to those of proteins, which suggests the presence of chargeable amine and carboxyl groups. We separated a polymer in the range of 10– 27 kDa and a small band with a molecular weight of 790 Da. Although the distribution of molecular weights by ultrafiltration-dialysis and electrophoresis indicates fractions heavier than 0.5 kDa, the MALDI mass spectra of tholins 1 and 2 recovered in DHB as matrix (2,5-dihydroxy benzoic acid) presents only peaks between 104 and 469 amu. All attempts to achieve mass spectrometric data using MALDI with other matrix or other mass spectroscopy techniques (FAB, EI and ESI) were unsuccessful.

3.3. Thermal stability The hydrophilic tholins 1 and 2 presented a well differentiated thermal behavior. The tholin 1 showed two phase transitions at 110 and 205 ◦ C and a continuous loss of weight. The tholin 2 presented three phase transitions at 205 about 300 and 390 ◦ C and a moderate weight loss in the studied range (from 25 to 500 ◦ C).

3.4. Analytical study for organic compounds The tholin 1, as well as the tholin 2, yielded amino acids only after acid hydrolysis. The amino acid profile was similar in both cases, although, the amount and diversity was greater for the tholin 1. In contrast, the raw tholin 1 presented a series of free carboxylic acids, not shown for the tholin 2, which was strongly enhanced after acid hydrolysis. Indeed, tholin 1 showed, before hydrolysis, heterocycles such as parabanic acid, hydantoin, adenine and 2,6-diaminopurine, not present in tholin 2 (Ruiz-Bermejo et al., 2007a). These analytical results were in excellent agreement with all spectroscopic data shown above.

In addition, PAHs were analyzed by SPME/GC–MS techniques.

Recently, we tested, for the first time in this context, the ability of this tool to study a raw reaction from an experiment with the same gas mixture and energy source described in this paper, but working below 0 ◦ C and using melting-freezing cycles (MenorSalván et al., 2008, 2007). In that case, PAHs were detected in a

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.6 (1-10) M. Ruiz-Bermejo et al. / Icarus ••• (••••) •••–•••

6

Fig. 4. HPLC chromatograms. Solid line: Tholin 1. Dashed line: Tholin 2. (a) Separation by reverse-phase conditions: Kromasil 100 C18 column, MeOH-0.1% TFA in water, detector wavelength 220 nm. (b) Separation by normal-phase conditions: BetaBasic-8 column, CH3 CN, detector wavelength 206 nm.

Table 4 Wavelengths of the absorbance spectra of each fraction separated by reverse-phase HPLC from tholins 1 and 2

Table 6 Dynamic light scattering results of protein lysozyme and tholin 1 in water at pH 7 and SDS solution at 0.1% in Tris–HCl buffer at pH 8.8

Tholin 1

Sample

Diameter (nm)

Z-potential (mV)

Mobility (μm cm/V s)

Lysozyme–water Lysozime–SDS Tholin 1–water Tholin 1–SDS

80.3 199.8 439.8 104.4

5.13 15.60 2.71 12.5

0.402 −1.339 −0.212 −1.222

Tholin 2

RT (min)

λmax (nm)

RT (min)

λmax (nm)

1.2 1.4 1.5 3.0 3.2 3.7

212, 257 (sh), 298 217 215 218, 235 (sh), 257 (sh) 215, 257 (sh) 227

1.1 1.4 1.6 3.0 3.2 3.6 3.9

215 215 215, 250 (sh) 220, 257 (sh) 220, 260, 285 (sh) 217, 252, 292 (sh) 218

RT = retention time of the each fractions observed by photodiodes array detector.

Table 5 Fractions separated by ultrafiltration and dialysis using filter devices and cellulose acetate membranes according to molecular weight Fraction

Tholin 1 (%)

Tholin 2 (%)

<0.5 kDa

67.8 4.6 16.3 7. 2 4.0

52.7 2 .0 34.8 7 .5 3 .0

0.5–1 kDa 1–3 kDa >3 kDa (soluble) >3 kDa (non-soluble)

relatively high yield showing the excellent capacity of this technique for the detection of PAHs in tholins (detection limit about ppb). However, no positive results were obtained for the tholins 1 and 2 by SPME/GC–MS. Moreover, the tholins 1 and 2 were ex-

Data presented as mean of two experiments.

Table 7 Functional groups present in the tholins 1 and 2 Tholin 1

Tholin 2

–CH3 , –CH2 –, –CH– –C=C– (no aromatic) –COOH, –COOR –OH –NH2 , –NH– –C≡N –N=/–N=C (heterocycles) –NH–C=O (amides)

–CH3 , –CH2 –, –CH– –C=C– (no aromatic) –COOH, –COOR –OH –NH2 , –NH– –C≡N

tracted with cyclohexane in a similar way to that described by Sagan et al. (1993), and the extracted sample analyzed by conventional GC–MS techniques did not detect PAHs. As a summary of the previous discussion we can conclude that: (i) The aqueous aerosols increase the diversity of functional groups

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.7 (1-10) CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins

in the formation of the hydrophilic tholins. (ii) Tholin 2 present a greater grade of unsaturation than tholin 1. (iii) Water aerosols favor the formation of lighter material. (iv) The hydrophilic tholins present an electrophoretic behavior similar to the common proteins although they do not contain peptide bonds in their structure. In Table 7 are indicated the total functional groups found for the tholins 1 and 2, using the indicated techniques.

4. Discussion 4.1. Organic molecules detected in tholins: Implications for astrobiology 4.1.1. Amino acids Acid hydrolysis of tholins usually releases amino acids. Different precursors and mechanisms have been proposed to explain their formation. In the experiments using spark discharges and CH4 and N2 , the main products of reaction are HCN, NH3 and HCOH. These molecules then could react to form aminonitriles according to the well-known Strecker synthesis, and finally, yield amino acids after hydrolysis (Miller, 1957). Another possibility is the HCN polymerization, which could lead to the formation of heteropolypeptides according to Matthews’s models (Matthews and Ludicky, 1992).

In this context, Coll et al. (1999) have suggested that the base of Titan’s tholins could be poly-HCN and McDonald et al. (1991) proposed that spark tholins contained short peptide-like fractions.

Thompson and Sagan (1989) also have suggested the existence of another mechanism of amino acid formation on Titan tholin synthesis. It is based on the polymerization reaction of nitriles with unsaturated and radical species. An alternative mechanism has been proposed in absence of HCN (Miyakawa et al., 1999).

The tholin 1 and the tholin 2 both provided amino acids only after acid hydrolysis. The lighter fractions <0.5 kDa were major components of the tholins. The IR, 13 C NMR and XPS spectra showed clearly the presence of nitrile and amine groups, suggesting aminonitriles as precursors of amino acids. In addition, in a recent work, we have shown that the Strecker mechanism was implicated in the formation of amino acids under our experimental prebiotic conditions (Ruiz-Bermejo et al., 2007b).

On the other hand, in the hydrophilic tholin 1, hydantoin and parabanic acid were detected. Thus, the amide groups detected by spectroscopy techniques can be related to this set of compounds and not with peptide bonds. Besides, the Bradford assay was negative suggesting that the broad band obtained in the SDS–PAGE electrophoresis was not peptidic in nature.

Therefore, we proposed the Strecker synthesis as the most favorable mechanism, but not the only one for the production of amino acids under our experimental conditions. The aqueous aerosols improved the formation of low molecular weight compounds, such as aminonitriles, and improved the yield in amino acids. The aqueous aerosols also allowed the formation of molecules, such as hydantoin, increasing the yield of amino acids and offering an alternative mechanism for the production of amino acids (Taillades et al., 1998).

4.1.2. PAHs PAHs (polycyclic aromatic hydrocarbons) are important components in the interstellar medium (see e.g. Lovas et al., 2005; Sloan et al., 1999; Salama, 1998; Snow et al., 1998; d’Hendecourt and Ehrenfreund, 1997). On the other hand, it has been suggested that the optical properties of Titan’s haze may be controlled by the size and the abundance of aromatic structures.

The differences observed in the tholins formed under different experimental conditions may be due to the variations of the aromatic structures (Imanaka et al., 2004). Moreover, recently it has been reported that PAHs could be key molecules in the origin of life due to their photochemical properties (Segré et al., 2001;

7

Deamer, 1992). The “Aromatic World” hypothesis suggested that assemblies based on aromatic hydrocarbons could serve as components of informational polymers, containers and mediators in protometabolic pathways (Ehrenfreund et al., 2006).

Aromatic compounds have been detected in synthetic tholins.

After pyrolysis of synthetic tholins, the most abundant aromatic compounds detected by GC–MS were benzene derivatives and some heteroaromatic compounds such as pyrrole, pyridine or pyridazine derivatives (see e.g. McGuigan et al., 2006; (Pietrogrande et al., 2001), Coll et al., 1999, 1998; Ehrenfreund et al., 1995; Khare et al., 1984). Coll et al. (1998) suggested that monoaromatic compounds could be the major constituents of Titan’s tholins and that these compounds could be the products of pyrolysis of the PAHs. It was difficult to determinate whether the source of these aromatic compounds was thermal desorption from solid phase or thermal decomposition of heavier molecules. However, Sagan et al.

(1993) indicated that intermediate-sized molecules, such as PAHs, were produced simultaneously with tholins and some fraction of them were adsorbed into the tholin matrix. These authors directly detected PAHs, containing from two to four fused rings, by L2 MS (two-step laser mass spectrometry) and by SF (synchronous fluorescence) in laboratory Titan’s and Jupiter’s tholins. Imanaka et al.

(2004) also detected PAHs in Titan’s tholins by μL2 MS (microprobe laser-desorption, laser-ionization mass spectrometry) formed under low pressure conditions (13–160 Pa). The μL2 MS data showed a complex organic mixture containing a variety of aromatics up 300 amu. This suggested a predominant amount of one- to two-ring compounds with extensive side-chain addition, but simple PAHs, such as naphthalene and phenanthrene, were not detected. Trainer et al. (2004a, 2004b) identified PAHs in several tholins by AMS (aerosol mass spectrometer) showing that the initial concentration of CH4 and the ratio C/O were decisive in the formation of PAHs. In the case of the Titan’s tholins, CH4 concentrations of upwards of 10% lead to formation of PAHs. In addition, the presence of aromatic compounds in tholins also can be determined by several spectroscopic techniques. Some assignment of aromatic groups by FT-IR spectroscopy are shown in Table 8, although it is not easy to distinguish between C=C double bonds and aromatic rings.

Using UV–vis spectroscopy, McDonald et al. (1994) indicated that the absorption bands at 280–290 nm and at 240 nm were either conjugated double bonds or aromatic structure. Finally, using solid state 13 C NMR, Sagan et al. (1993) observed resonances in Titan’s and Jupiter’s tholins consistent with aromatic and/or alkene groups. The solid state 13 C NMR spectrum of the insoluble organic residue in the Murchison meteorite showed a broad resonance at 129 ppm that was assigned to aromatic carbon. In this case, the width of the peak indicated that aromatic carbons existed in many different local electronic environments (Cody et al., 2002).

In our experiments, aromatics and PAHs were not detected in the tholin matrix, although the amount of CH4 was up to 10%, and the experimental conditions were similar to Jupiter’s tholin (Sagan et al., 1993). The IR spectra and UV–vis spectra of tholins 1 and 2 showed bands related to C=C and aromatic groups, but the 13 C NMR spectra did not present resonances in the range of 150– 120 ppm associated with aromatic carbons. These results were in good agreement with the analytical data. Therefore, tholins 1 and 2 presented conjugated double bonds in their structures, but neither contained substituted benzenes nor PAHs. Additionally, the conjugated unsaturated bonds were not widespread in the fractions separated by HPLC from these tholins (see Table 4), and the distribution of the unsaturated bonds was different in both. The UV–vis spectra of the fractions from tholin 2 showed a greater degree of conjugation than tholin 1. This result was in agreement with the empirical formulas of the bulk tholins.

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.8 (1-10) M. Ruiz-Bermejo et al. / Icarus ••• (••••) •••–•••

8

Table 8 Vibrational frequencies that suggest aromatic compounds or alkenilic groups measured by FT-IR spectroscopy

ν (cm−1 )

Assignment

Ref.

1636–1617 1579–1565a 1450–1440 760–740 650–560

C=C stretching conjugated to –C≡N/aromatics/group containing N Aromatics (quadrant strecthing) Aromatics (semicircle strecthing) Aromatics Aromatics

Imanaka et al. (2004)

1596

C=C stretching conjugated to alkenes/aromatics

Khare et al. (1989a, 1989b)

3005 1600 1450

C=C or aromatic C–H groups Aromatic C=C Aromatic C=C

McDonald et al. (1991, 1994)

1650–1640 1600–1500

C=O, N–H stretching, C=C stretching C–C vibrations associated with aromatic groups

Nascimiento et al. (1998)

1650–1600 860–690

Alkenilic C=C Aromatic molecules

Sagan et al. (1993)

1600 970 880 825

C=C stretching conjugated to C=C, C≡C/–C≡N CH=CH, trans C=CH2 C=CHR

Tran et al. (2003)

a

Imanaka et al. point out that an increase in the band at 1580 cm−1 is related to case of conjugation with fused ring systems.

In summary, the formation of PAHs, in simulation experiments to produce tholins, is very sensitive to the environmental conditions, such as pressure, temperature, and source and proportion of carbon in the initial gas mixture, and it is independent of the energy source used.

On the other hand, taking into account that the presence of aromatic and/or PAHs in the Titan’s tholin is still an open question, we propose the use of the SPME/GC–MS techniques as a useful tool to study the atmosphere of Titan. The GC–MS systems are widely used as flight instrumentation and could be easily implemented with a fiber of SPME to detect, with a relatively high sensibility, aromatic and polycyclic compounds.

4.1.3. Heterocycles containing nitrogen Adenine and 2,6 diaminopurine were found free before hydrolysis in the tholin 1, and this result is in agreement with the XPS data where a signal related to –N=C (heterocycles) was observed.

In contrast, in the tholin 2 heterocycles were not detected. We suggest a similar mechanism in the formation of adenine to the previously proposed by Yuasa et al. (1984) in which the HCN was implicated. Although a radical mechanism cannot be ruled out, the aqueous aerosols seemed to improve the stabilization of polar radicals (Ruiz-Bermejo et al., 2007b). Cyanoacetylene was produced in substantial yield from electric discharges acting on CH4 + N2 mixtures (Sanchez et al., 1966), and Ferris et al. (1968) obtained heterocycles from the cyanoacetylene and urea trough radical pathways.

4.2. Structure and formation of hydrophilic tholins 4.2.1. Distribution of molecular weights in the tholins A great variety of molecular weight distributions for several tholins, using gel filtration chromatography, has been previously reported. McDonald et al. (1991) found fractions in the range of weights between 200 and 700 for a Jupiter’s tholin. Triton’s and Titan’s tholins presented a molecular weight distribution between 200 to 600 Da that consisted of polar molecules with only ∼10 to 50 (C + N) atoms (McDonald et al., 1994). Takano et al.

(2004b) obtained high molecular weight organic compounds from CO/NH3 /H2 O and proton irradiation. In this case, the molecular weight distribution ranged between several hundred and ∼3 kDa. In the all examples reported, the data were recovered using UV

detectors in a range of 195–220 nm, therefore, the information obtained only corresponded to the fraction containing UV-detectable material.

Sarker et al. (2003), using high-resolution Fourier-transform ion cyclotron resonance mass spectrometry (FT-IC/MS), showed simple and regular patterns in the molecular weight distribution of the organics in regularly spaced clusters from a Titan’s tholin with a range of weights between 120 and 800 Da. The authors indicated that the C/N ratio was a smoothly decreasing function of atomic mass, and the degree of unsaturated bonds scaled with molecular complexity.

The mass spectrometry techniques applied to the tholins 1 and 2 did not provide good results, likely due to their heterogeneous nature. However, the alternative ultrafiltration-dialysis method showed that the tholin 1 contained a greater amount of light molecules than the tholin 2. This result was in agreement with the thermal behavior of both tholins and with the presence of a greater grade of unsaturated bonds in tholin 2. In addition, tholins 1 and 2 seemed to present a Gaussian distribution of molecular weights up 0.5 kDa, typical for the formation of polymers.

Another simple and successful technique for the determination of molecular weight of polar tholins was electrophoresis, but this biochemical tool is poorly reported for the determination of tholins and other similar chemical species (Draganic et al., 1980; Ferris et al., 1981). Although, we do not know the exact nature of the chemical bonds responsible for the skeleton of the hydrophilic tholins, we detected polar groups such as carboxylates, nitriles and amino. At this basic level, the hydrophilic tholins can develop a net electric charge in water by the same acid base reactions that the proteins undergo and should migrate under an electrical field under similar conditions. We proved in this work that the effective migration of hydrophilic tholins in an electric field is exactly the same as that of common proteins (meaning under the same values of electric field and gel type). We detected high molecular weight polymers not observed in previous work. A comprehensive characterization applying the methodology herein developed of these heavier polar fractions requires further studies.

4.2.2. No homogeneous structure of hydrophilic tholins: Competitive mechanisms The molecular weight distribution and the organic molecule analysis indicate that the tholins 1 and 2 are formed mainly by discrete molecules and small polar oligomers as well as by poly-

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.9 (1-10) CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins

mers/heteropolymers, which can get to a remarkable molecular weight. Taking into account the spectroscopy data, we suggest that these polymers/heteropolymers were constituted by hydrocarbon chains containing a notable number of unsaturated bonds and chargeable polar groups such as –OH, –NH and –COOH. Our proposal for the structure for the heavier fraction was similar to that reported by McDonald et al. (1991).

The nature and ratio between the discrete molecules and the polymers/heteropolymers detected in tholins 1 and 2 suggest different mechanisms of formation that compete with each other. The aqueous aerosols improved the formation of light polar compounds and molecular diversity, but partially inhibited the formation of the polymers/heteropolymers. In this regard, we propose a set of three pot reaction mechanisms: one on the bulk water, one in the water–gas interface and the last in the gas phase. The hydrophilic character was clearly enhanced by the increase of the water–gas interface. This suggested to us a mechanism in which a hydrocarbon chain was formed on the gas phase by means of radical pathway. Then, it located on the surface that incorporated polar radicals coming also from the gas phase. This could explain the molecular weight distribution in tholin 1 since under gas-surface enhanced conditions a higher density of polar groups was incorporated. These groups (for example, the carboxylates) were, in fact, terminal groups and favored oligomers or small polymers. In addition, the fast incorporation of hydrophilic groups favored the solubilization of the molecule/polymer that stops growth. The formation of a considerable amount of carboxylic acids in experiment 1 supported this hypothesis. It was in agreement with the results reported by other authors (Donaldson et al., 2004; Tuck, 2002; Ellison et al., 1999). These authors indicated that carboxylic acids were produced by the reaction of OH with hydrocarbons at the water–gas interface, especially in aerosols. In this context, the presence of aqueous aerosols did not lead to the formation of PAHs since the enhanced interfacial attachment of polar groups mentioned actively competed with the formation of unsaturated compounds that could polymerize by radical pathway (Wilson and Atreya, 2003).

Another important aspect of the mechanism to consider is why, in our experiments, there are one hydrophilic tholin and one hydrophobic tholin formed instead of a tholin with a mixed nature. This result was important with respect to the origin of the first membranes because it was clear that the self-assembly phenomenon responsible for membrane formation was possible only if both hydrophobic and hydrophilic molecular components were present.

Also it could be interesting to take in consideration that the presence of water in Titan seems to be scarce and that the formation of aqueous aerosols is unlikely in this moon and therefore the formation of hydrophobic tholins may be preferential respect on the formation of hydrophilic tholins. In this case, the reactions in the gas phase may be dominant. The hydrophobic tholin formed in experiments 1 and 2 were studied with the same techniques and methodology described in this work and the results indicate that differences with the hydrophilic tholins are notable (manuscript in preparation).

5. Conclusions The parallel use of different spectroscopic, spectrometric, analytical and biochemical techniques applied to our hydrophilic tholins provide global and useful knowledge about their nature.

With the obtained structural information, it is possible to infer that tholins 1 and 2 are complex mixtures that include carboxylic (acids and esters), amine, hydroxy, nitrile and amide groups. In these raw mixtures, other functional groups like ketones, aldehydes or aromatics and peptide-like fractions were not observed. These bulk

9

tholins are not homogeneous, being formed mainly by low molecular weight compounds and other fractions with predictable polymeric nature. These heavier fractions seem to be formed by hydrocarbon chains containing unsaturated bonds, which were present in large amounts in tholin 2, as well as polar groups. Therefore, an important effect of the aqueous aerosols in the prebiotic simulation experiments with spark discharges is to instigate the formation of bioorganics in a high relatively yield and improves the formation of discrete multifunctional precursors (aminonitriles for amino acids and ester for carboxylic acids). By the contrary the presence of aerosols did not favor the polymerization processes, which implicated the formation of C=C. This fact could be due to the special properties of the water–gas interphase of the aerosol’s droplets. We propose that hydrocarbon chains located at the water–gas interface were more exposed to the bombardment of the polar molecules generated in the spark, whereas bulk polymerization took place independently the presence of aerosol by a different mechanism, originating the more hydrophobic fraction.

Finally, we remark that there is a great potential for this multitechnical approach to obtain exhaustive information about tholins, and therefore, it can be used for the study of the heavier fractions of the tholins. These fractions are very interesting due to the similarities between them and proteins with respect to their electrokinetic behavior in water. Further studies may prove that these similarities could be extended to other biochemical aspects.

Acknowledgments The authors have used the research facilities of Centro de Astrobiología (CAB) and have been supported by Instituto Nacional de Técnica Aeroespacial “Esteban Terradas” (INTA) and the project AYA2006-15648-C02-02 of the Ministerio of Educación y Ciencia (Spain). We thank M.T. Fernández for recovering the IR spectra and the DSC assays, J. Sobrado for the XPS measurement and I. Sobrados from ICMM (Instituto de Ciencia de Materiales de Madrid.

CSIC. Spain) for recovering the 13 C NMR spectra. We acknowledge the constructive revision and comments of M.-P. Zorzano.

References Bradford, M.A., 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal.

Biochem. 72, 248–254.

Cody, G.D., Alexander, C.M.O’D., Tera, F., 2002. Solid-state (1 H and 13 C) nuclear magnetic resonance spectroscopy of insoluble organic residue in the Murchison meteorite: A self-consistent quantitative analysis. Geochim. Cosmochim. Acta 66, 1851–1865.

Coll, P., Coscia, D., Smith, N., Gazeau, M.-C., Guez, L., Raulin, F., 1998. Review and latest results of laboratory investigations of Titan’s aerosols. Orig. Life Evol.

Biosph. 28, 195–213.

Coll, P., Coscia, D., Smith, N., Gazeau, M.-C., Ramírez, S.I., Cernogora, G., Israel, G., Raulin, F., 1999. Experimental laboratory simulations of Titan’s atmosphere: Aerosol and gas phase. Planet. Space Sci. 47, 1331–1340.

Deamer, D.W., 1992. Polycyclic aromatic hydrocarbons: Primitive pigment systems in the prebiotic environment. Adv. Space Res. 12, 183–189.

Donaldson, D.J., Tervahattu, H., Tuck, A.F., Vaida, V., 2004. Organic aerosols and the origin of life: An hypothesis. Orig. Life Evol. Biosph. 34, 57–67.

Draganic, Z.D., Kinetic, V., Jonanovic, S., Draganic, I.G., 1980. The radiolysis of aqueous ammonium cyanide—Compounds of interests to chemical evolution studies.

J. Mol. Evol. 15, 239–260.

Ehrenfreund, P., Boon, J.J., Commandeur, J., Sagan, C., Thompson, W.R., Khare, B., 1995. Analytical pyrolysis experiments of Titan aerosol analogues in preparation for the Cassini Huygens mission. Adv. Space Res. 15, 335–342.

Ehrenfreund, P., Rasmussen, S., Cleaves, J., Chen, L., 2006. Experimentally tracing the key steps in the origin of life: The aromatic world. Astrobiology 6, 490–520.

Ellison, G.B., Tuck, A.F., Vaida, V., 1999. Atmospheric processing of organic aerosols.

J. Geophys. Res. 104, 11633–11641.

Ferris, J.P., Sanchez, R.A., Orgel, L.E., 1968. Studies in prebiotic synthesis. 3. Synthesis of pyrimidines from cyanoacetylene and cyanate. J. Mol. Evol. 33, 693–704.

Ferris, J.P., Edelson, E.H., Auyeung, J.M., Joshi, P.C., 1981. Structural studies on HCN oligomers. J. Mol. Evol. 17, 69–77.

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

ARTICLE IN PRESS JID:YICAR 10

AID:8729 /FLA

YICAR:8729

[m5G; v 1.64; Prn:29/09/2008; 13:14] P.10 (1-10) M. Ruiz-Bermejo et al. / Icarus ••• (••••) •••–•••

d’Hendecourt, L., Ehrenfreund, P., 1997. Spectroscopic properties of polycyclic aromatic hydrocarbons (PAHs) and astrophysical implications. Adv. Space Res. 9, 1023–1032.

Imanaka, H., Khare, B.N., Elsila, J.E., Bakes, E.L.O., McKay, C.P., Cruikshank, D.P., Sugita, S., Matsui, T., Zare, R.N., 2004. Laboratory experiments of Titan tholin formed in cold plasma at various pressures: Implications for nitrogen-containing polycyclic aromatic compounds in Titan haze. Icarus 168, 344–366.

Khare, B.N., and 10 colleagues, 1984. The organic aerosols of Titan. Adv. Spaces Res. 4, 59–68.

Khare, B.N., Sagan, C., Ogino, H., Nagy, B., Er, C., Schram, K.H., Arakawa, E.T., 1986.

Amino acids derived from Titan tholins. Icarus 68, 176–184.

Khare, B.N., Sagan, C., Thompson, W.R., Flynn, L., Morrison, M.A., 1989a. Amino acids and their polymers in the lower clouds of Jupiter? Preliminary findings. Orig.

Life 19, 495–496.

Khare, B.N., Thompson, W.R., Murray, B.G.J.P.T., Chyba, C.F., Sagan, C., 1989b. Solid organic residues produced by irradiation of hydrocarbon-containing H2 O and H2 O/NH3 ices: Infrared spectroscopy and astronomical implications. Icarus 79, 350–361.

Koike, T., Kaneko, T., Kobayashi, K., Miyakawa, S., Takano, Y., 2003. Formation of organic compounds from simulated Titan atmosphere: Perspectives of the Cassini mission. Biol. Sci. Space 17, 188–189.

Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.

Lebonnois, S., Bakes, E.L.O., McKay, C.P., 2002. Transition from gaseous compounds to aerosols in Titan’s atmosphere. Icarus 159, 505–517.

Lovas, F.J., McMahon, R.J., Grabow, J.U., Schnell, M., Mack, J., Scott, L.T., Kuczkowski, R.L., 2005. Interstellar chemistry: A strategy for detecting polycyclic aromatic hydrocarbons in space. J. Am. Chem. Soc. 127, 4345–4349.

Matthews, C.N., Ludicky, R., 1992. Hydrogen cyanide polymers on comets. Adv. Space Res. 12, 21–32.

McDonald, G.N., Khare, B.N., Thompson, W.R., Sagan, C., 1991. CH4 /NH3 /H2 O Spark tholin: Chemical analysis and interaction with jovian aqueous clouds. Icarus 94, 354–367.

McDonald, G.N., Thompson, W.R., Heinrich, M., Khare, B.N., Sagan, C., 1994. Chemical investigation of Titan and Triton tholins. Icarus 108, 137–145.

McDonald, G.N., Whited, L.J., DeRuiter, C., Khare, B.N., Patnaik, A., Sagan, C., 1996.

Production and chemical analysis of cometary ice tholins. Icarus 122, 107–117.

McGuigan, M., Waite, J.H., Imanaka, H., Sacks, R.D., 2006. Analysis of Titan tholin pyrolysis products by comprehensive two-dimensional gas chromatography-timeof-flight mass spectrometry. J. Chromatogr. A 1132, 280–288.

Menor-Salván, C., Ruiz-Bermejo, M., Osuna-Esteban, S., Veintemillas-Verdaguer, S., 2007. Generating the aromatic world: Synthesis of aromatic compounds in icy environments. Geochim. Cosmochim. Acta 71, A653.

Menor-Salván, C., Ruiz-Bermejo, M., Osuna-Esteban, S., Muñoz-Caro, G., Veintemillas-Verdaguer, S., 2008. Synthesis of polycyclic aromatic hydrocarbons and acetylene polymers in ice: A prebiotic scenario. Chem. Biodiv., in press.

Miller, S.L., 1957. The mechanism of synthesis of amino acids by electric discharges.

Biochim. Biophys. Acta 23, 480–489.

Miyakawa, S., Sawaoka, A.B., Ushio, K., Kobayashi, K., 1999. Mechanism of amino acids formation using optical emission spectroscopy. J. Appl. Phys. 85, 6853– 6857.

Nascimiento, L.F.C., Mota, R.P., Valenca, G.P., Teixeira, J.C., Algatti, M.A., Honda, R.Y., Bortoleto, J.R.R., 1998. An N2 :CH4 :H2 O glow discharge plasma probed by optical and electric techniques: Significance to the radiation chemistry of Titan’s upper atmosphere in the presence of meteorite water. Planet. Space Sci. 46, 969–974.

Pietrogrande, M.C., Coll, P., Sternberg, R., Szopa, C., Navarro-Gonzalez, R., VidalMadjar, C., Dondi, F., 2001. Analysis of complex mixtures recovered from space missions: Statistical approach to the study of Titan atmosphere analogues (tholins). J. Chromatogr. A 939, 69–77.

Ruiz-Bermejo, M., Menor-Salván, C., Osuna-Esteban, S., Veintemillas-Verdaguer, S., 2007a. Prebiotic microreactors: A synthesis of purines and dihydroxy compounds in aqueous aerosol. Orig. Life Evol. Biosph. 37, 123–142.

Ruiz-Bermejo, M., Menor-Salván, C., Osuna-Esteban, S., Veintemillas-Verdaguer, S., 2007b. The effects of ferrous and other ions on the abiotic formation of biomolecules using aqueous aerosols and spark discharges. Orig. Life Evol.

Biosph. 37, 507–521.

Sagan, C., Khare, B., 1979. Tholins: Organic chemistry of interstellar grains and gas.

Nature 277, 102–107.

Sagan, C., Khare, B.N., Thompson, W.R., McDonald, G.D., Wing, M.R., Bada, J.L., VoDinh, T., Arakawa, E.T., 1993. Polycyclic aromatic hydrocarbon in the atmosphere of Titan and Jupiter. Astrophys. J. 414, 399–405.

Salama, F., 1998. The diffuse interstellar bands: A tracer for organics in the diffuse interstellar medium? Orig. Life Evol. Biosph. 28, 349–364.

Sanchez, R.A., Ferris, J.O., Orgel, L.E., 1966. Cyanoacetylene in prebiotic synthesis.

Science 154, 784–785.

Sarker, N., Somogyi, A., Lunine, J.I., Smith, M.A., 2003. Titan aerosol analogues: Analysis of the non-volatile tholins. Astrobiology 3, 719–726.

Segré, J.H.D., Ben-Eli, D., Deamer, D.W., Lancet, D., 2001. The lipid world. Orig. Life Evol. Biosph. 31, 119–145.

Simionescu, C.R., Denes, F., Onac, D., Bloods, G., 1974. Synthesis of some amino acids, sugars, and peptides in cold plasma: Abiotic synthesis of some high-molecular weight proteid-like structures (V). Biopolymers 13, 943–954.

Sloan, G.C., Hayward, T.L., Allamandola, L.J., Bregman, J.D., DeVito, B., Hudgins, D.M., 1999. Direct spectroscopic evidence for ionized polycyclic aromatic hydrocarbons in the interstellar medium. Astrophys. J. 513 (1 Pt 2), L65–L68.

Snow, T.P., Le Page, V., Keheyan, Y., Bierbaum, V.M., 1998. The interstellar chemistry of PAH cations. Nature 391, 259–260.

Stoscheck, C.M., 1990. Quantitation of protein. Meth. Enzymol. 182, 50–69.

Taillades, J., Beuzelin, I., Garrel, L., Tabacik, V., Bied, C., Commeyras, A., 1998.

N-carbamoyl-α -amino acids rather than free a-amino acids formation in the primitive hydrosphere: A novel proposal for the emergence of prebiotic peptides. Orig. Life Evol. Biosph. 28, 61–77.

Takano, Y., Tsuboi, T., Kaneko, T., Kobayashi, K., Marumo, K., 2004a. Pyrolysis of highmolecular-weight complex organics synthesized from a simulated interstellar gas mixture irradiated with 3 MeV proton beam. Bull. Chem. Soc. Jpn. 77, 779– 783.

Takano, Y., Ohashi, A., Kaneko, T., Kobayashi, K., 2004b. Abiotic synthesis of highmolecular-weight organics from an inorganic gas mixture of carbon monoxide, ammonia, and water by 3 MeV proton irradiation. Appl. Phys. Lett. 84, 1410– 1412.

Thompson, W.R., Sagan, C., 1989. Atmospheric formation of organic heteropolymers from N2 + CH4 : Structural suggestions for amino acids and oligomer precursors.

Orig. Life 19, 503–504.

Trainer, M.G., Pavlov, A.A., Curtis, D.B., McKay, C.P., Worsnop, D.R., Delia, A.E., Toohey, D.W., Toon, O.B., Tolbert, M.A., 2004a. Haze aerosols in the atmosphere of early Earth: Manna from heaven. Astrobiology 4, 409–419.

Trainer, M.G., Pavlov, A.A., Jimenez, J.L., McKay, C.P., Worsnop, D.R., Toon, O.B., Tolbert, M.A., 2004b. Chemical composition of Titan’s haze: Are PAHs present?.

Geophys. Res. Lett. 31, doi:10.1029/2004GL019859. L17S08.

Tran, B.N., Ferris, J.P., Chera, J.J., 2003. The photochemical formation of Titan haze analog. Structural analysis by X-ray photoelectron and infrared spectroscopy.

Icarus 162, 114–124.

Tuck, A., 2002. The role of atmospheric aerosols in the origin of life. Surv. Geophys. 23, 379–409.

Wilson, E.H., Atreya, S.K., 2003. Chemical sources of haze formation in Titan’s atmosphere. Planet. Space Sci. 51, 1017–1033.

Yuasa, S., Flory, D., Basile, B., Oró, J., 1984. Abiotic synthesis of purines and other heterocyclic compounds by the action of electrical discharges. J. Mol. Evol. 21, 76–80.

Please cite this article in press as: Ruiz-Bermejo, M., et al CH4 /N2 /H2 spark hydrophilic tholins: A systematic approach to the characterization of tholins. Icarus (2008), doi:10.1016/j.icarus.2008.07.008

PRL 100, 146102 (2008)

PHYSICAL REVIEW LETTERS

week ending 11 APRIL 2008

Asymmetric Chiral Growth of Micron-Size NaClO3 Crystals in Water Aerosols S. Osuna-Esteban,1,* M.-P. Zorzano,1 C. Menor-Salva´n,1 M. Ruiz-Bermejo,1 and S. Veintemillas-Verdaguer1,2 1

2

Centro de Astrobiologı´a (CSIC-INTA), Carretera de Ajalvir km 4, 28850 Torrejo´n de Ardoz, Madrid, Spain Department of Particulate Materials ICMM-CSIC, Sor Juana Ine´s de La Cruz 3, Cantoblanco 28049, Madrid, Spain (Received 10 June 2007; published 9 April 2008) We describe an aerosol-liquid cycle that launches the autocatalytic amplification of any initial imbalance of the order of 10 7 % (1 ppb) up to total chiral purity in a single step process. Crystal nucleation of NaClO3 is initiated at the aerosol air-water interface where, due to the accumulation of ambient chiral impurities or added hydrophobic chiral aminoacids in tiny concentrations (ppb), the initial levorotatory (l) and dextrorotatory (d) excess will not be produced with equal probability. The enantiomeric yield is then enhanced up to homochirality by recycling the crystallites through a liquid phase. In the absence of added catalysts this process leads to preferential (d) homochiral crystallizations in a ratio of 4:1 which is due to ambient contamination. By adding only 2 ppb of (L) or (D) Phe, we induce a final preferential homochiral crystallization of (d) or (l) handedness, respectively, in a ratio of 2:1.

DOI: 10.1103/PhysRevLett.100.146102

PACS numbers: 82.65.+r, 33.15.Bh, 61.50. f, 61.66. f

The homochirality of natural amino acids (L) and sugars (D) remains a puzzle for theories of the chemical origin of life. Mirror-image asymmetric molecules, i.e., chiral isomers or enantiomers, are classically considered as chemically identical. The tiny excess of one enantiomer in a racemic mixture due either to natural fluctuations, external fields, or small energy imbalances must have been amplified by the joint effect of an autocatalytic process and some type of competition between the left- and right-handed products. Laboratory experiments demonstrate how in specific autocatalytic reaction systems the presence of a small amount of reaction product with enantiomeric excess (ee) at the beginning of the reaction can result in a much larger ee at the end of the reaction. In organic synthesis, in 1995 Soai showed how an organic reaction with initial 5% ee in the catalyst produced an ee of 55% in the product [1]. In a later modified version, he showed that a catalyst with ee as low as 5 10 5 % can breed its own chirality to 99.5% ee in a three-cycle synthesis [2]. A number of mathematical models have also shown the ability to generate extensive ee from racemic initial conditions without the influence of any chiral field, simply amplifying the initial random fluctuations by an autocatalytic process and including some competing process between the left- and righthanded enantiomers [3–5]. In inorganic crystallization processes, back in 1990, Kondepudi studied the crystallization of the sodium chlorate NaClO3 salt which is now a classical experiment of spontaneous homochirality [6].

The undisturbed supersaturated solution yields statistically equal numbers of levo (l) and dextro (d)-rotary crystals nuclei [7,8]. Stirring the solution produces secondary nucleation and causes the breakdown of chiral symmetry [9].

The ee of either (l) or (d) nuclei can be greater than 99.8%.

In a number of independent stirred crystallizations, the final homochiral handedness seems to be random. Thus it is equally probable to find a (l) or (d) homochiral situation.

Recently, Viedma has proposed a variation of this experi0031-9007=08=100(14)=146102(4)

ment: a 100% symmetry breaking may be obtained from systems initially containing both (l) and (d) crystals of NaClO3 if the stirred solutions contained certain amount of 3 mm-sized glass balls that continuously crushed the crystals keeping their maximum size to about 200 m [10]. In particular, the system with initial 5% ee achieved 100% ee of the same sign within 8 h. Under these conditions a continuous dissolution-crystallization phenomenon is enhanced by the abrasion-grinding process. This enhanced dissolution leads to the inhibition of the less favored chirality and favors the autocatalytic crystallization of the dominant one.

In this context, it is a challenge to amplify the tiniest possible enantiomeric imbalance up to enantiomerically pure products using a process which does not involve harsh interferences with the products, such as rotating grinders.

In this Letter we propose the air-water interface as a potential surface for enhanced isomerization kinetics.

The air-water interface enhances the effective molecular ordered packing and increases the local concentration and the influence of short-range forces. To increase the surfacevolume ratio and explore the surface-induced phenomena we propose to generate an aerosol-liquid cycle, see Fig. 1(a), of a saturated sodium chlorate NaClO3 solution where dissolution of the smallest crystals will also take place. Here it is shown how the aerosol-liquid cycle will launch the autocatalytic amplification of any initial imbalance of the order of 10 7 % (1 ppb) up to total chiral purity in a one-pot, single step process.

Experimental setup and observations.—All glass material was steamed for sterilization during 2 h. The experiments were run at room temperature in an ultraclean ventilation system of laminar airflow. We used sodium chlorate from Aldrich S.A. (ACS quality) and ultrapure water (Millipore Milli-Q Q. Gard). For each crystallization process 5 ml of a sodium chlorate aqueous solution (4.88 g NaClO3 in 10 g of dissolution, saturated 51.2%, 9.86 m

146102-1

© 2008 The American Physical Society

PHYSICAL REVIEW LETTERS

PRL 100, 146102 (2008)

week ending 11 APRIL 2008

aerosol

liquid bulk

(a)

(b)

(c)

FIG. 1 (color online). (a) Scheme of aerosol-liquid cycle and aerosol surface-induced phenomena. (b) Glass reactor with aerosol-liquid cycle of a supersaturated NaClO3 solution. (c) SEM image of the first crystals formed in the aerosol phase: spherical aggregates of 100 m diameter and individual crystals of 10 m. The reference bar is 100 m-sized.

saturated at 30 C) was prepared at 45 C under vigorous stirring (500 rpm). The hot solution was filtered, through a 0:2 m-pore size sterile nylon, directly to the 500 ml glass reactor and placed within a bath of the ultrasonic aerosol generator (a commercial, small ultrasonic generator of 10 8 MHz, 30 W). The initial temperature was 30 C but during operation the water bath of the ultrasonic generator heated up and raised the temperature of the liquid in the flask up to 38 C. The experiment was run with 5 ml of saturated solution of sodium chlorate under five different initial conditions: one without any added catalyst, and the rest with added traces of D or L amino acids that will act as chiral catalyst (see below for details). After 2 min, the ultrasonic device produces an optically thick cloud of aerosols in suspension above the liquid bulk, see Fig. 1(b). For this solution the minimum theoretical aerosol diameter is 1:2 m but aerosol droplets may be greater due to droplet coalescence. Typical sizes of the droplet salt shell relics seen by scanning electron microscopy (SEM) imaging were about 30 m (not shown). After 1â&#x20AC;&#x201C;2 h (depending on the experiment) of an experimental run, we take a sample of the products in the aerosol phase in an aluminum SEM stub. In Fig. 1(c) we show a SEM image of the NaClO3 crystals. One can clearly distinguish two crystal aggregates with the spherical shape of two independent aerosol droplets of about 100 m size (notice the curvature of this shell). Smaller, 10 m-sized crystals can also be distinguished and they seem to have octahedral form, see the isolated crystals on the right-hand-most side of Fig. 1(c). In Fig. 2 we show a microscope image of crystals taken from the aerosol phase roughly half an hour later (depending on the experiment), just previous to the onset of crystallization. There are crystals of about 100 m with octahedral or mixed cubo-octahedral form, instead of the characteristic cubic crystals that are usually formed in bulk solution. It is known that habit modification of crystals is mainly related with the effect of impurities. In Fig. 3 we show the measured curve of concentration in the

FIG. 2 (color online). Microscope image of a sample taken from the aerosol phase right before the onset of crystallization: individual 100 m-sized octahedral crystals. The reference bar is 100 m-sized.

flask liquid versus time. For completeness we include the solubility during this process (see the straight lines), which depends on the liquid temperature. As evaporation takes place, both on the liquid and aerosol surface, the concentration increases [11]. When the concentration is well above the solubility curve, a sharp peak appears which corresponds to the supersaturation state. Then the crystal nucleation is launched. This process is first initiated in the air-water interface of aerosols, where the local concentration is enhanced. The growth of these crystals in the flask leads to a rapid drop in the chlorate concentration. After, typically, 4 h there are crystals in the flask of observable sizes (ranging between 1.75â&#x20AC;&#x201C;1.80 mm as measured in 50 crystals) and show the characteristic cubic form of liquid crystallization with perfect facets (i.e., with no abrasion or crystal breakage in contrast to the classical stirred Kondepudi and the Viedma experiments). The chirality of the crystals was then determined from their optical activity using a stereomicroscope and two linear polarizers.

For each experimental run there are three possible outcomes of the crystallization process: total symmetry breaking with either (l) or (d) homochiral crystallization or racemic crystallization with equal population of (l) and (d) crystals. See Figs. 4(a) and 4(b) for a total homochiral case (l) and a racemic result, respectively. The racemic crystallization results are associated with experiments where the crystallization initiation took place in the liquid bulk and not in the aerosol phase. In fact in most of these

FIG. 3. Time evolution of the flask chlorate concentration showing a rapid drop due to crystal growth. The solubility evolution due to temperature variation is indicated as reference.

146102-2

PHYSICAL REVIEW LETTERS

PRL 100, 146102 (2008)

(a)

(b)

FIG. 4 (color online). Crystal samples from the liquid bulk, taken after 4 h. The reference bar is 2 mm-sized. (a) Homochiral case showing (l) polarization. (b) Racemic crystallization with roughly equal populations of (l) and (d) crystals (light and dark cubes, respectively).

cases we noticed that the aerosol phase was optically thin (indicating that the aerosol forming process was not effective) and crystallization took much longer than the usual 4 h.

Next we summarize the results of 176, 4 h long, crystallization processes under different initial conditions, and the ratio of the preferred crystallization form to the nonpreferred one, see Table I. For completeness we include the counting of experiments which lead to racemic distributions, but we focus on the homochiral cases which are related to aerosol crystallization. If the experiment is run with no added catalyst, see case (i), the result is a homochiral preferential amplification of (d). Since this experimental setup is so sensitive to tiny amino acid impurities, the aerosols amplify the influence of natural, biological air contamination which consists of natural (L) amino acids [12]. The ratio of this preferential homochiral crystallization is 4. This result is consistent with a recent work on the sodium chlorate crystallization where air-water crystal nucleation is enhanced with a wind generator acting on a set of Petri dishes. Then, in an independent amplification step (the glass-ball grinding process), the initial ambientinduced ee is amplified to total purity [13] (notice that that work uses the opposite optical activity convention). Next we add traces of amino acids of known chirality. If we use (D) or (L) Phe, which is one of the most hydrophobic amino acids, we see that with only 2 ppb the ratio of

TABLE I. Number of experiments that lead to final homochiral (d), (l) or racemic (l d) crystallizations under different initial conditions and ratio of preferred crystallization.

Chiral catalyst (i) Ambient contamination (ii) 2 ppb D-Phe (iii) 2 ppb L-Phe (iv) 200 ppb D-Phe (v) 2 ppb D-Tyr (vi) 2 ppb L-Tyr

(d)

(l)

Ratio

Racemic

16 8 13 8 13 13

4 16 7 16 10 15

4 2 1.85 2 1.3 1.15

7 6 9 4 4 7

week ending 11 APRIL 2008

preferential crystallization is of the order of 2, for either (l) or (d), see cases (ii) and (iii), respectively. If the concentration of the amino acid is increased (200 ppb) we obtain the same final homochiral ratio, see case (iv) with D-Phe. Finally for less hydrophobic amino acids, such as Tyr, the (D) or (L) added catalyst is not able to drive the preferential homochiral crystallization processes and the final homochiral handedness seems to be random.

Physical interpretation. 窶認rom what has been described above we conclude that each aerosol is a micron-sized reactor that produces chiral nanocrystals of the same handedness. As can be seen in Fig. 2, the aerosol crystallization process takes place on the droplet surface. The Na cations tend to form hydrated clusters in which the ion binds to water oxygen atoms, staying in the homogeneous environment of the aerosol liquid whereas the anion is closer to the surface [14]. In addition, intensive evaporation takes place at the air-water interface. Thus there is a net increment in the local concentration of salt at this interface, and the crystallization process is launched here which is the first environment to reach supersaturation. Theory has suggested that discrimination between hetero- and homochiral ordering becomes more likely when short-range repulsive forces (such as those experienced by molecules in closepacked crystals) are significant. On the surface, short-range repulsive forces induce the production of homochiral crystal aggregates. As for the influence of amino acids: hydrophilic amino acids are solubilized primarily in the water aerosol volume, whereas hydrophobic molecules migrate to the air-water interface [15]. The fact that hydrophobic molecules, located at the air-water interface, are more effectively driving the chirality of the crystallization process is consistent with our hypothesis that the crystallization takes place in this environment. Furthermore, the octahedral form of the initial aerosol induced crystal nuclei proves the presence of impurities during the crystallization process. The autocatalytic process responsible for the huge chirality imbalance is based on the following cycle: first homochiral crystal aggregates are very efficiently formed on the droplet surface, and put back into the liquid bulk where they split into smaller units, see Fig. 1(a). Whenever the liquid flask solution is below saturation the smallest crystals will be dissolved in the liquid. Some crystals are lifted again in aerosols where rapid evaporation and increased local concentration continuously brings the solution to supersaturation conditions producing new aggregates of the same handedness in the aerosol phase.

Thus crystals which grow first will be slightly greater and survive while the opposite handedness crystals, of smaller size, redissolve. By this crystallization-dissolution process the production of nuclei of the opposite handedness is indirectly suppressed, similarly to what was observed in [10,16]. The growth of these crystal seeds in the flask leads to a final rapid drop in the solution concentration which brings the liquid bulk nucleation rate to zero. From this

146102-3

PRL 100, 146102 (2008)

PHYSICAL REVIEW LETTERS

moment on, new crystals will only be formed in the aerosol phase. Any initial bias (ambient chiral impurities or added hydrophobic chiral amino acids in concentrations of ppb) that may support the faster crystallization of one of the chiralities in the aerosol air-water interface, will make this final homochiral configuration more probable. However if, due to the initial unavoidable reaction fluctuations, the initial enantiomeric excess overcame the catalyst influence and favored the opposite crystallization, it would also be amplified to total purity. Thus, in the experiments where nucleation has been induced in the aerosol phase, the only possible final outcomes are (d) or (l) homochiral states.

Why is it that the addition of L-Phe produces a ratio of 1.85 in favor of (d) instead of the ratio 4 found with ambient (L) contamination? Added, internal impurities and ambient contamination effects do not act simultaneously, and thus they do not add their effects. Added internal impurities, such as L-Phe, which are present from the beginning in the aerosol droplets, act first launching the crystal nucleation process. Once the nucleation has been induced by this catalyst the sign and ratio of the final bias in the homochiral crystallizations is already determined (in this case, 1:85:1). Ambient contamination acts later, feeding slowly but continuously the surface of the aerosol phase with impurities. In case no nucleation has been launched by internal catalysts, this infinite pool of ambient impurities drives very efficiently the nucleation process towards a strongly biased ratio (4:1). In this Letter we have shown how the aerosol air-water interface and this aerosol-liquid cycling seems to be very efficient in ordering and packing organic and inorganic molecules, increasing the local concentration and enhancing the influences of short-range forces, evaporating beyond supersaturation and thus favoring certain reactions such as crystallization.

On Earth, the natural water cycle goes through the aerosol phase and the liquid bulk phase on a regular basis.

It has been speculated that the aerosol cycle on the primordial Earth served as a natural reactor for prebiotic chemistry, offering an integrated and cyclic network of microenvironments which strongly supported organic chemical self-organization [17,18]. Aerosol induced phenomena such as this one have been proven to have influence on atmospheric chemistry [19]. In this work we have demonstrated the relevance of the aerosol-liquid cycle on the enhancement of an initial chiral excess. It also constitutes an experimental proof of how tiny (ppb) initial imbalances can be amplified by an autocatalytic cycle to total homochirality. This technique provides promising results and can be refined for further applications.

week ending 11 APRIL 2008

The authors wish to acknowledge the scientific advice and constant support of Professor Juan Pe´rez-Mercader and D. Hochberg. This research was supported by the Instituto Nacional de Te´cnica Aerospacial (INTA) and by Grant No. AYA2006-15648-C02-02 of the Ministerio de Educacio´n y Ciencia (Spain).

*osunaes@inta.es http://www.cab.inta.es [1] K. Soai, T. Shibata, H. Morioka, and K. Choji, Nature (London) 378, 767 (1995).

[2] Sato, H. Urabe, S. Ishiguro, T. Shibata, and K. Soai, Angew. Chem., Int. Ed. 42, 315 (2003).

[3] J. Rivera Islas et al., Proc. Natl. Acad. Sci. U.S.A. 102, 13 743 (2005).

[4] D. Hochberg and M.-P. Zorzano, Chem. Phys. Lett. 431, 185 (2006).

[5] T. Buhse, J. Mex. Chem. Soc. 49, 328 (2005).

[6] D. K. Kondepudi, R. J. Kaufman, and N. Singh, Science 250, 975 (1990).

[7] F. Donald Bloss, Crystallography and Crystal Chemistry, Mineralogical Society of America, Monograph Series (Holt, Rinehart and Winston, Washington, DC, 1994).

[8] As the light travels toward the observer, the plane of polarization of a linearly polarized beam is rotated by the crystal either to the left (counterclockwise) or to the right (clockwise) relative to its plane upon entry. If the former holds, the crystal is levorotatory (l); if the latter, it is dextrorotatory (d) [7]. This is the convention we have followed here.

[9] S. Veintemillas-Verdaguer, S. Osuna-Esteban, and M. A. Herrero, J. Cryst. Growth 303, 562 (2007).

[10] C. Viedma, Phys. Rev. Lett. 94, 065504 (2005).

[11] At the onset of crystallization the viscosity is lowered.

This further increases the aerosol production and the evaporation increases nonlinearly. Furthermore there are rapid concentration fluctuations right at this point.

[12] D. W. Armstrong, J. P. Kullman, X. Chen, and M. Rowe, Chirality 13, 153 (2001).

[13] C. Viedma, Cryst. Growth Des. 7, 553 (2007).

[14] B. C. Garrett, Science 303, 1146 (2004).

[15] J. Yano, H. Furedi-Milhofer, E. Wachter, and N. Garti, Langmuir 16, 9996 (2000).

[16] J. Crusats, S. Veintemillas-Verdaguer, and J. M. Ribo´, Chemistry 12, 7776 (2006).

[17] L. Lerman, Orig. Life Evol. Biosph. 26, 369 (1996).

[18] C. M. Dobson, G. Barney Ellison, A. F. Tuck, and V. Vaida, Proc. Natl. Acad. Sci. U.S.A. 97, 11 864 (2000).

[19] M. Ruiz-Bermejo, C. Menor-Salva´n, S. Osuna-Esteban, and S. Veintemillas-Verdaguer, Orig. Life Evol. Biosph.

37, 123 (2007).

146102-4

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

2729

Synthesis of Polycyclic Aromatic Hydrocarbons and Acetylene Polymers in Ice: A Prebiotic Scenario

by Ce´sar Menor-Salva´n*, Marta Ruiz-Bermejo, Susana Osuna-Esteban, Guillermo Mun˜oz-Caro, and Sabino Veintemillas-Verdaguer Centro de Astrobiolog a (Consejo Superior de Investigaciones Cient ficas-Instituto Nacional de Te´cnica Aeroespacial (CSIC-INTA)), Carretera Torrejo´n-Ajalvir, Km. 4,200, E-28850 Torrejo´n de Ardoz, Madrid (e-mail: menorsc@inta.es)

The recent evidences of presence of subsurface oceans of liquid water and ice on Saturn s moons, and the possible presence and astrobiological importance of polycyclic aromatic hydrocarbons (PAHs) in these environments, provide strong motivation for the exploration of the prebiotic chemistry in ice and to test if PAHs could be experimentally synthesized in ice surfaces under atmospheres containing methane as carbon source. In this work, we present a new design for prebiotic-chemistry experiments in ice matrix.

Using this design, a mixture of products including PAHs, polar aromatic compounds, and hydrophilic acetylene-based polymers was obtained. We propose that acetylene generation in a methane/nitrogen atmosphere and subsequent polymerization to PAHs and polyynes could be a favored pathway in the presence of water freeze – melt cycles. These results shed light on the processes involved in PAH synthesis in icy environments and on the physical factors that drive the different competing pathways in methane/ nitrogen atmospheres.

Introduction. – Recently, polycyclic aromatic hydrocarbons (PAHs) were proposed as key molecules in the study of the origin of life. Due to their photochemical properties, PAHs could play the role of a primitive pigment system that drove synthesis of amphiphilic compounds and contributed to prebiotic chemical evolution [1] [2]. The aromatic world hypothesis suggests that assemblies based on aromatic hydrocarbons could serve as components of informational polymers, or as containers and mediators in protometabolic pathways [3]. The stability of such aromatic compounds and their capacity for self-assembly, driven by the p – p stacking interaction and weak forces, could make aromatic hydrocarbons potential building blocks for protocellular structures. Indeed, it has been demonstrated that amphiphilic polycyclic aromatic compounds are capable of self-assembly and the formation of bilayer structures [4].

PAHs represent a substantial portion of the interstellar carbon, constituting ca. 20% of the available carbon [5]. The aromatic compounds present in interstellar and circumstellar ices may contribute to the materials incorporated into planets, asteroids, and comets. Aromatic hydrocarbons have been identified in meteorites [6] and interplanetary dust particles [7], and PAH derivatives constitute a high percentage of the carbon in chondrites [8]. Despite the astrobiological and astrochemical interest in PAHs and the abundance of studies conducted at high (combustion) temperatures, we could not find an efficient synthesis of aromatics under conditions applicable to planetary atmospheres. The theoretical synthesis of aromatics by acetylene-insertion mechanism at Titan atmosphere highlights the need for an experimental confirmation 2008 Verlag Helvetica Chimica Acta AG, Z rich

2730

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

of the synthesis of PAHs at lower temperatures [9]. Concerning the presence of prebiotic aromatic compounds on a planet/asteroid, we pose two questions: could the synthesis of aromatics take place in a planetary environment, or does this process require cosmic sources? Furthermore, are the PAHs found in meteorites, and potentially present in Earth-like planets such as Mars, diagenetic products of biological compounds or abiotic in origin? [10] The study of prebiotic chemistry in cold environments recently gained significance with the investigation of Saturn s icy moons. The possible presence of liquid water plumes (with eruption of methane, carbon dioxide, and nitrogen) and recycled ice in Enceladus and a liquid briny ocean in Europa suggest the presence of a water freeze – melt cycle [11] [12]. Recently, Cassini-mission observations suggested the presence of a subsurface liquid ocean of H2O/NH3 in Titan [13]. The freeze-cycling process is a unique environment that generates a concentrated liquid phase with strong electric and temperature gradients, and high pressure providing an efficient reactor [14]. This work explores the possibilities of prebiotic chemistry in icy environments using methane as carbon source, and the chemical evolution in presence of water freeze – melt cycles in order to establish predictable pathways for the chemistry in subsurface water oceans in liquid – solid equilibrium.

Results. – Experiments in Water Freeze – Melt Cycles. To clarify the above questions and to simulate an icy planetary body with a freeze – melt cycle active under primordial atmosphere, the following experiment was performed: a freeze – melt cycle was generated on a pure and sterile water pool under a N2/H2/CH4 30 : 30 : 40 atmosphere.

This atmosphere is adequately reducing in nature and constitutes the primary carbon source for the experiment [15]. Once the freeze – melt cycle was established in the entire water pool during the ice phase, the system was energized by means of spark discharges. A tungsten electrode connected to a high-frequency and high-voltage generator induced a discharge that impacted the water/ice surface. After 72 h, the voltage generator was disconnected, and the reactor was maintained under active freeze – melt cycling. After three months, the reactor contents, a yellowish solution with suspended organic solid, were allowed to warm up to room temperature and finally collected as fast as possible in sealed headspace vials under N2 atmosphere, in order to avoid excessive manipulation and minimize the contamination risk.

The bulk sample was analyzed by solid-phase microextraction (SPME) coupled with gas chromatography/mass spectrometry (GC/MS) to determine the presence of organic compounds (Fig. 1, a). We found a rich mixture of benzene derivatives, with a significant abundance of phenylethyne and biphenyl compounds. We identified PAHs, with significant quantities of naphthalene, alkyl-naphthalenes, acenaphthene, fluorene, and anthracene (Fig. 1). Also, we identified diphenylmethane, acenaphthylene, pyrene, anthracene, phenanthrene, and benzil, and the single-ring aromatics benzene and toluene were identified. The main aromatics synthesized were naphthalene, methylnaphthalenes and biphenyl, containing 0.01% of the total carbon introduced in the experiment (Table). Moreover, the organic speciation indicates that some aromatic carbonyl derivatives (acetophenone, benzaldehyde, tolualdehydes, benzoic acid, and geranylacetone) were also synthesized. All compounds were identified by comparison of their mass spectra and retention times with those of pure commercial standards.

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

2731

Fig. 1. a) GC/MS Total ion chromatogram (TIC) of the SPME extract of bulk sample obtained after 72 h of energization by spark discharges and let stand for three months while maintaining the water freeze – melt cycles under N2 /CH4 /H2 30 : 40 : 30 atmosphere. Chromatogram shows only compounds confirmed against authentic standards. b) Same type of extract from an experiment initiated under the same conditions (freeze – melt cycles, N2/CH4/H2 atmosphere) but stopped after 72 h of spark discharges. The chromatogram shows the GC/MS TIC of the SPME extract of the sample. .: Detected in Murchison meteorite; *: detected in Allende meteorite; ^: detected in Cold Bokkeveld meteorite.

After analysis of the bulk sample, we performed the analysis of the suspended solid

by APCI-TOF-MS and NMR. The total mass of sample was 22.5 mg, representing 29%

2732

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

Table. Quantitative Estimation of the Main Aromatic Compounds Obtained in the Experiment Using Water Freeze – Melt Cycles under N2 /CH4 /H2 30 : 40 : 30 Atmosphere Compound

C [ppb] a )

Total amount [mg]

Total amount [nmol]

Total carbon [%] b )

Naphthalene 2-Methylnaphthalene 1-Methylnaphthalene 1,1’-Biphenyl Acenaphthene Fluorene Anthracene

588 182 202 404 16 48 58

17.64 5.46 6.06 12.12 0.48 1.26 1.02

130 38 43 79 3 8 6

0.00412 0.00118 0.00128 0.00264 0.00010 0.00026 0.00024

a

) Concentration of each compound in bulk solution obtained in the experiment (30 ml). b ) Percentage of total carbon introduced in the experiment (428 mg).

of the total organic material obtained in the experiment and 2.6% of the total carbon introduced. This solid is partially soluble in H2O, and if left to stand, coalesces into micron-sized particles that could absorb dyes and can be separated by centrifugation (Fig. 2). The atmospheric pressure chemical ionization (APCI) fragmentation showed a polymeric nature for the solid, displaying several significant fragments consisting of successive losses of triacetylene (C6H2 , m/z 74) and H2O (m/z 18). The polymer appears to be constituted by chains of repetitive units of OH-substituted triacetylene with amine and nitrile terminations. Further work is in progress to elucidate its structure. No evidence of aromatic carbon was found in polymers or solids formed during the experiment, and the largest PAHs found were three-ring species.

Fig. 2. Methylene blue-stained vesicles of hydroxylated poly(triacetylene), collected after coalescence during 24 h and centrifugation at 10000 rpm (400 )

Control Experiments. Four control experiments were performed using the same reactor: I) the same atmosphere and the same reaction times with H2O in liquid phase

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

2733

at room temperature. In this experiment, we did not find evidence of the synthesis of aromatic compounds, and aliphatic alcohols were identified in the reaction mixture (Fig. 3); II) performed under the same conditions but without energization of the system; in this case, neither organic material nor contamination was observed in the water pool; III) same conditions of freeze – melt cycles and spark discharges for 72 h, but the samples were analyzed immediately after this period. We found benzene and naphthalene derivatives (Fig. 1, b); and IV) performed under the same conditions as the main experiment in terms of time and energy, and H2O in the ice phase without the melt protocol. Significant quantities of conjugated dienes, diynes, and nitriles were identified, with no evidence of PAH synthesis (Fig. 4). The formation of solid material with low solubility was increased in this control experiment.

Discussion. – We have shown that the excitation of a system by penetration of spark discharges in ice/water surfaces leads to the synthesis of a mixture of reactive compounds. This synthesis could be the consequence of four effects: i) cleavage of H2O to OH . radical, and subsequent generation of H2O2 by recombination of OH . ; ii) other reactive species derived from electrically activated H2O: oxygen, singlet oxygen, and possibly O3 [15]; iii) dissipation of spark energy in the form of ultraviolet radiation and associated photochemical processes [16] [17]; iv) atmospheric generation of CH3 . radical and acetylene. The direct conversion of CH4 to acetylene using electric arc, flame pyrolysis, or plasma discharges is a well-known process [18]. The generated acetylene could follow two main pathways: PAH growth mechanism by H abstraction – acetylene addition and/or polyyne generation model. Under electric arc or pyrolysis conditions, these mechanisms lead to the formation of large PAHs and soot particles [19] [20]. Under the conditions of our experiments, the temperature could be a decisive factor as it affects the amount of H2O in the gas phase (Fig. 3). At temperatures over 58, the atmospheric generation of OH . radicals could lead to the generation of alcohols, and further oxidation to carbonyl compounds [21]. At lower temperatures, the increased significance of acetylene-addition mechanisms could explain the presence of aromatic rings and the presence of a poly(triacetylene) polymer by two complementary mechanisms: first, generation of single aromatic ring from acetylene and vinyl radical or vinylacetylene (Berthelot synthesis) and growth of PAHs by H abstraction plus acetylene addition, and second, polyyne growth and generation of polymer particles (Scheme). The model of benzene generation in the Titan atmosphere proposed by Wilson et al. [9] could be compatible with our experimental data and suggests that the mechanism responsible for PAH formation at temperatures under combustion could be valid in a wide range of pressure and temperatures between the triple point of H2O and below 200 K. The generation of geranylacetone (6,10-dimethylundeca-5,9-diene-2-one) is an interesting issue. This molecule is an acyclic isoprenoid, classical intermediate in squalene metabolism. We confirmed twice the synthesis of geranylacetone in the freeze – melt cycle experiments only, with no detectable presence in any of the control experiments performed. We suggest two possible routes: successive ethynylation and hydrogenation [22], coherent with the observation of secondary alcohols in control experiments without ice, that could be possible intermediates, and variation on the Frenklach mechanism of acetylene addition, followed by evolution in the ice matrix [19] [23].

2734

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 5 (2008)

Fig. 3. GC/MS Chromatogram showing the SPME extract of the mixture obtained in control experiment I, performed under the same conditions of atmosphere (N2/CH4/H2 30 : 40 : 30), spark discharges, and reaction time of the main experiment, but with the system maintained at room temperature. Under these conditions, there is no evidence of aromatic-ring-containing molecules. Instead, we obtain a complex mixture composed basically of aliphatic alcohols, suggesting that T > 58 increases the significance of atmospheric generation of HO . radical.

The presence of H2O induces oxidation of aromatics parallel to PAH growth. This effect occurs at the phenyl stage of the PAH-growth mechanism, leading to the formation of benzaldehyde, tolualdehydes, acetophenone, and benzoic acid (Fig. 4).

The oxygen incorporation at this stage diverts carbon from the PAH-growth pathway and could explain the lack of higher PAHs. The freeze â&#x20AC;&#x201C; melt cycles (that, from an astrobiological point of view, could be generated by thermal variations on a planetary surface with water-ice and liquid phases in equilibrium) convert the ice surface from a trap or condenser of reactive molecules to a reactor that leads to the synthesis of a complex mixture of aromatic compounds and unsaturated molecules. This role of ice could be caused by the generation of concentrated brines during freezing, subjected to high pressure and gradients, and/or catalytic effect of the ice surface. In this ice reactor , reactions could take place that lead to more stable and complex products such as PAHs, and to the formation of carbonyl compounds and prevent the formation of soot or kerogen-type material by the generation of a hydrophilic poly(triacetylene) material that could form particles and vesicles by coalescence. The polyynes and cyanopolyynes have been detected in some astrophysical objects, and the presence of triacetylene and more complex polyynes is expected in Titan [24] [25]. These compounds are stable at low temperature and evolve to polymer chains at standard temperature. Under our conditions, the generation of acetylenic reactants in presence of ice leads to the formation of a OH-rich, polar poly(triacetylene) derivatives. This result is interesting for the expected chemistry in

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 5 (2008)

2735

Scheme. Possible Pathways of Reaction in Freeze â&#x20AC;&#x201C; Melt Ice Cycles Starting from Methane That Could Explain the Main Products Obtained in Experiment. The boxed zone of pathway represents alternative route favored at room temperature with generation of saturated alcohols. The presence of nitrogen made possible a mechanism driven by HCN polymerization and nitrile generation at room temperature. Low temperature favors the mechanism based on acetylene polymerization that generates PAHs and the poly(triacetylene) polymer, and room temperature favors the nitrogen incorporation pathways and generation of primary/secondary alcohols.

possible surface or subsurface ices on planetological objects. Further work is in progress to explain the formation and astrobiological importance of this polymer.

It is interesting to note that a significant fraction of the aromatic hydrocarbons detected in carbonaceous chondrites are common with the products reported here [26] [27]. In particular, all the alkyl derivatives of naphthalene found in our experiments were detected in meteorites (Fig. 5). The alkyl derivatives of benzene and

2736

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

Fig. 4. Control experiment performed under the same conditions of Fig. 1, a, but using water in the ice phase without freeze – melt cycles. The figure shows the TIC of the SPME extract of bulk sample. No PAHs were found in this control experiment. The molecules containing conjugated C¼C or C C bonds, and reactive groups (aldehyde, nitrile) could be precursors that undergo cyclization reactions to produce the more stable products shown in Fig. 1.

naphthalene are compounds usually found in bitumens and oils on Earth, and considered of biological origin. Our results indicate that these compounds could be of abiotic origin in extraterrestrial media and the potential occurrence of alkyl-aromatics should be taken with caution.

Of course, our results do not prove that aromatics found in meteorites or interstellar space are produced by the route described here, but they represent a plausible pathway for the chemistry in presence of H2O and CH4 at low temperature, and an alternative route for the classic HCN pathways described for N2/CH4 mixtures [28].

Conclusion. – I) Icy surfaces favor the synthesis of molecules with conjugated C¼C/ C C bonds, and the freezing process favors the cyclization processes and polar aromatic compounds. II) Aromatic compounds could be synthesized in planetary/ asteroidal environments, including terrestrial planets or Saturn s icy moons, from CH4 by the acetylene-polymerization mechanism. III) Irradiation of CH4-rich ices or waterice under CH4/N2 atmosphere generates a poly(triacetylene) derived tholin and discrete aromatic molecules, but not polyaromatic kerogen-type material or HCN polymers.

IV) The aromatic compounds and PAHs found in chondrites and the potential presence of unsaturated/aromatic molecules in Earth-like planets may not be indicators of extant/extinct life.

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 5 (2008)

2737

Fig. 5. GC/MS Chromatograms at specific mass/charge relations for different aromatic compounds. a) Fragmentogram at m/z 141 (characteristic of alkyl-naphthalenes) of the SPME extract of bulk sample obtained in the main experiment: 72 h of excitation by spark discharges plus three months maintaining the water freeze â&#x20AC;&#x201C; melt cycles under N2/CH4/H2 30 : 40 : 30 atmosphere. We found 1- and 2-methylnaphthalenes, 1- and 2-ethylnaphthalenes, and 1,3-dimethylnaphthalene. The presence of alkyl derivatives of naphthalene has been verified in carbonaceous chondrites. b) Fragmentogram at m/z 152 of the same sample showing alkyl-benzenes (compound class identified by their fragmentation pattern) and other important aromatic compounds generated in the experiment.

2738

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

Experimental Part Simulation of Water-Ice Prebiotic Conditions. Ultrapure, sterile, and oxygen-free H2O (30 ml) was frozen at 58 in a sealed and thermostatized glass reactor under an atmosphere of N2 (30%), H2 (30%), and CH4 (40%). The system was energized with a high-voltage generator (model BD-50E, Electrotechnic Products Inc. Illinois, USA) by high-frequency spark discharges (50 kV, 0.5 MHz) directly in H2O, through a W electrode attached to reactor (electrically earthed with a secondary W electrode). Freeze – melt cycles were established by varying the temp. between 58 and 58. The system was maintained under spark discharge for 72 h. After this period, the high-voltage generator was disconnected, and the reactor was kept sealed with active freeze – melt cycles over three months. Then, the system was allowed to warm up to r.t., and the yellowish aq. soln. was stored in sealed headspace vials under an inert atmosphere.

Solid-Phase Microextraction. Bulk sample was analyzed using headspace solid phase microextraction (SPME), followed by GC/MS. An aliquot (15 ml) of the bulk sample was heated at 708 for 30 min in a vial closed with a PTFE septum cap. A 75-mm carboxen-polydimethylsiloxane fibre (provided by Supelco) was then exposed to the headspace while keeping the sample at 708 for another 10 min. To avoid memory effect of the SPME fiber and to avoid environmental pollution, a new unused SPME fiber is used for each analysis.

Quant. estimation was performed using external calibration curves for each compound in H2O soln. in order to compensate the different fiber response. 15 ml of calibration standard was heated at 708 for 30 min in a vial closed with a PTFE septum cap. A 75-mm carboxen-polydimethylsiloxane fiber was then exposed to the headspace while keeping the sample at 708 for another 10 min.

GC/MS. The analytes collected by the fiber were thermally desorbed in the injection port of the GC in the splitless mode, and the analysis was performed using an Elite-5 (Perkin-Elmer), 5% phenyl-95% methylsiloxane cap. column (30 m 0.25 mm i.d., 0.25-mm film).

MS Analysis was performed with a Perkin-Elmer Autosystem XL-Turbomass Gold quadrupole in scan EI þ mode. Org. compounds were identified by search of their mass spectra in the NIST database, and identified compounds were confirmed against authentic standards (provided by Sigma-Aldrich) by comparison of mass spectra and retention times. For the identification purposes, we considered only peaks with a signal-to-noise ratio over 20, which corresponds to the signal-to-noise ratio of the peak for 1 ng/ml of hexachlorobenzene, which was used as a surrogate standard. Those peaks of which the match probability in the database was below 90% and/or matched but were commercially unavailable were considered as unidentified.

APCI-TOF-MS. The TOF-MS analysis was conducted using a Micromass mass spectrometer equipped with an atmospheric pressure chemical ionization interface (APCI). Measurements were performed in positive-ion mode with 30-V fragmentor voltage, 4508 APCI probe temp., 1308 source temp., 2.5 ml · min 1 nebulizer gas flow (N2 ), 5 ml · min 1 desolvation gas flow and 3600-V corona voltage.

We are grateful to Prof. Bernd R. T. Simoneit and Prof. A. M. Echevarren for their thorough and constructive reviews. We thank Jose Maria Arribas and Servicio de Espectrometr a de Masas (Universidad de Alcala´ ) for APCI-TOF analysis support and, also, we thank the Centro de Astrobiolog a (CAB) for the research facilities. The grants from Instituto Nacional de Te´cnica Aeroespacial Esteban Terradas (INTA) and the project AYA2006-15648-C02-02 of the Ministerio de Educacio´n y Ciencia (Spain) are gratefully acknowledged.

REFERENCES [1] [2] [3] [4] [5] [6]

D. Segre´, D. Ben-Eli, D. W. Deamer, D. Lancet, Orig. Life Evol. Biosphere 1988, 31, 57.

D. W. Deamer, Adv. Space Res. 1992, 12, 183.

P. Ehrenfreund, S. Rasmussen, J. Cleaves, L. Chen, Astrobiology 2006, 6, 490.

L. Chen, C. Geiger, J. Perlstein, D. G. Whitten, J. Phys. Chem., B 1999, 103, 9161.

L. J. Allamandola, A. G. Tielens, J. R. Barrer, Astrophys. J., Suppl. Ser. 1989, 71, 733.

M. A. Sephton, I. Gilmour, Astrophys. J. 2000, 540, 588.

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

2739

[7] S. J. Clemett, C. R. Maechling, R. N. Zare, P. D. Swan, R. M. Walker, Science 1993, 262, 721.

[8] C. Sagan, B. N. Khare, W. R. Thompson, G. D. McDonald, M. R. Wing, J. L. Bada, T. Vo-Dihn, E. T. Arakawa, Astrophys. J. 1993, 414, 399.

[9] E. H. Wilson, S. K. Atreya, A. Coustenis, J. Geophys. Res. 2003, 108, 1.

[10] L. Becker, D. P. Glavin, J. L. Bada, Geochim. Cosmochim. Acta 1997, 61, 475.

[11] J. S. Kargel, Science 2006, 311, 1389.

[12] M. G. Kivelson, K. K. Khurana, C. T. Russell, M. Volwerk, R. J. Walker, C. Zimmer, Science 2000, 289, 1340.

[13] R. D. Lorenz, B. W. Stiles, R. L. Kirk, M. D. Allison, P. Persi del Marmo, L. Iess, J. I. Lunine, S. J. Ostro, S. Hensley, Science 2008, 319, 1649.

[14] H. Trinks, W. Schrçder, C. K. Biebricher, Orig. Life Evol. Biosphere 2005, 35, 429.

[15] M. Ruiz-Bermejo, C. Menor-Salvan, S. Osuna-Esteban, S. Veintemillas, Orig. Life Evol. Biosphere 2007, 37, 123.

[16] B. Sun, S. Kunitomo, C. Igarashi, J. Phys. D: Appl. Phys. 2006, 39, 3814.

[17] G. Schlesinger, S. L. Miller, J. Mol. Evol. 1983, 19, 383.

[18] J. R. Fincke, R. P. Anderson, T. Hyde, B. A. Detering, R. Wright, R. L. Bewley, D. C. Haggard, W. D. Swank, Plasma Chem. Plasma Proc. 2002, 22, 105.

[19] M. Frenklach, Phys. Chem. Chem. Phys. 2002, 4, 2027.

[20] A. V. Krestinin, Combust. Flame 2000, 121, 513.

[21] A. F. Tuck, Surv. Geophys. 2002, 23, 379.

[22] W. Bonrath, M. Eggersdorfer, T. Netscher, Catal. Today 2007, 12, 45.

[23] C. W. Bauschlicher Jr., A. Ricca, Chem. Phys. Lett. 2000, 326, 283.

[24] F. Cataldo, Orig. Life Evol. Biosphere 2006, 36, 467.

[25] A. Aflalaye, D. Andrieux, Y. Be´nilan, P. Bruston, P. Coll, D. Coscia, M.-C. Gazeau, M. Khlifi, P. Paillous, R. Sternberg, E. de Vanssay, J.-C. Guillemin, F. Raulin, Adv. Space Res. 1995, 15, 5.

[26] J. R. Cronin, S. Chang, in The Chemistry of Life s Origins , Eds. J. M. Greenberg, C. X. MendozaGo´mez, V. Pirronello, Kluwer, Dordrecht, 1993, p. 209 – 258 and refs. cit. therein.

[27] M. A. Sephton, Nat. Prod. Rep. 2002, 19, 292.

[28] C. N. Matthews, Faraday Discuss. 2006, 133, 393.

Received June 5, 2008

Orig Life Evol Biosph DOI 10.1007/s11084-006-9026-5

Prebiotic Microreactors: A Synthesis of Purines and Dihydroxy Compounds in Aqueous Aerosol M. Ruiz-Bermejo & C. Menor-Salván & S. Osuna-Esteban & S. Veintemillas-Verdaguer

Received: 12 June 2006 / Accepted: 10 August 2006 # Springer Science + Business Media B.V. 2006

Abstract We report the synthesis of purine bases and other heterocycles and the formation of amino acids, hydroxy acids and dihydroxy compounds by the spark activation of an atmosphere of methane, nitrogen and hydrogen, in the presence of an aqueous aerosol. With the aid of the interface air–water, the organic material obtained shows greater amounts and diversity of molecules with biological interest than the products obtained in the absence of an aerosol. Our results support the suggestion that aerosols may have played a significant role in the prebiotic origin of molecular diversity and evolution.

Keywords prebiotic chemistry . prebiotic environments . aerosol chemistry . purine bases . amino acids

Introduction According to modern ideas on the origin of life, organic compounds derived from chemical reactions in primitive Earth-like environments. These reactions took place in the transition period from abiotic, inorganic, or simple organic compounds to autonomous self-replicating molecules capable of evolving by natural selection, generally called the ‘prebiotic epoch’ (Bada 2004). Our vision of the prebiotic epoch is underpinned in Miller’s classic experiment (Miller 1953). There has been considerable discussion concerning the nature of the primitive atmosphere since Miller’s original publication. While arguments have been advanced that the primitive atmosphere would not have been as reducing as some have supposed (Delano 2001) there are no data available concerning the geochemistry of the period in question and the possibility of a methane-containing atmosphere has been left open. Consequently, syntheses have been modelled in a variety of atmospheres ranging

M. Ruiz-Bermejo : C. Menor-Salván (*) : S. Osuna-Esteban : S. Veintemillas-Verdaguer Centro de Astrobiología [Consejo Superior de Investigaciones Científicas-Instituto Nacional de Técnica Aeroespacial (CSIC–INTA)], Carretera Torrejón-Ajalvir, km 4,2, 28850 Torrejón de Ardoz, Madrid, Spain e-mail: menorsc@inta.es

Orig Life Evol Biosph

from strongly reducing (Ring et al. 1972) through weakly reducing (Schlesinger and Miller 1983) to redox neutral (Civis et al. 2004).

In this work, we examine the possible effect of a greatly increased gas–water interface on the synthesis of organic compounds in a reducing atmosphere, by using aerosols as prebiotic microreactors.

The possible importance of aerosols in the origin of life on Archean Earth has recently been emphasized (Donaldson et al. 2004; Tuck 2002; Dobson et al. 2000; Lerman 1994).

The first proposal of a possible role of aerosols as important reactors in prebiotic chemistry was made by Shah (1970). This suggestion is supported by physical–chemical observations which demonstrate that the formation of an aerosol depends only on a liquid water–air interface and a physical mechanism that ejects bubbles into the atmosphere, such as wind, sea waves or shock waves (Lerman 1996; Ellison et al. 1999).

On modern Earth, aerosols are ubiquitous in the troposphere and play an important role in climate and atmospheric chemistry (Ellison et al. 1999). Recent laboratory work (Donaldson and Anderson 1999) has shown that small organic molecules tend to concentrate at the water–air interface. Similarly, recent observations of the composition of natural aerosol particles have shown that at least 10% of the aerosol is organic matter, with more than 10.000 peaks in gas chromatography (Murphy et al. 1998; Murphy 2005) and organic coating of surfactants in marine aerosol particles (Tervahattu et al. 2002). Due to these facts, we could assume that the bubble–aerosol–droplet cycle (the bursting of bubbles that injects into the atmosphere the aerosol particles and subsequent condensation in droplets), was active in the Archean epoch. Therefore, it could be functionally relevant to chemical evolution on the early Earth or elsewhere. It is important to note that aerosol particles are different from cloud droplets and raindrops. The latter are larger hydrometeors with droplet radii greater by one to four orders of magnitude than aerosol droplets. These have a fractional organic content minuscule compared with aerosol particles (Tuck 2002).

The bubble–aerosol–droplet or ‘bubblesol’ cycle offers the possibility to overcome a number of problems with regard to prebiotically synthesized organic matter (Dobson et al.

2000; Lerman 1996). The goal of our work was to experimentally generate the bubble– aerosol–droplet cycle in putative prebiotic conditions and test the above hypothesis.

Although the mechanism would presumably be effective in atmospheric models of a large range of compositions, we chose a mixture of CH4, N2, and H2, since this composition is known to yield amino acids in sufficient amounts for comparative quantitative analysis (Schlesinger and Miller 1983).

Materials and Methods Simulation of the prebiotic conditions A 500 ml glass reactor is filled with 5 ml of ultrapure water (Millipore Milli-Q Q.Gard®.

Additionally the ultrapure water was bubbled with dried N2 for 1 h in order to remove the residual traces of oxygen) and a gas mixture containing CH4:N2:H2 (40:30:30), purchased from Praxair SA. Prior to use, all glassware and the electrodes was heated in a hightemperature oven (Nabertherm L5) at 400°C for 2 h in order to eliminate any possible contaminants. Before the experiment the system is successively evacuated with a pump and purged with the reaction gas mixture for four times. The system working pressure is

Orig Life Evol Biosph

1,200 mbar at 38°C. An ultrasonic aerosol generator (BONECO model 7035) was used for the generation of the bubble–aerosol–droplet cycle with the liquid water in reactor. Two tungsten electrodes attached to the reactor were used with a high voltage generator (Model BD-50E, Electrotechnic Products Inc. Illinois, USA) to produce the spark discharges (50 KV) in the aerosol. The system is maintained at constant temperature with active aerosol and electric discharge during 72 h. After this period, the yellow-brown liquid solution and the solid material were recovered for analytical study.

Another experiment was carried out as experimental control in the same conditions of pressure and temperature but without the aerosol.

Instrumental analysis Elemental analysis. The elemental analysis of each fraction was made in Unidad de Instrumentación Científica (Universidad de Alcalá de Henares. 28801 Alcalá de Henares.

Spain) using a ‘CHN-O-rapid’ (Heraeus) elemental analyzer.

Infrared spectroscopy. IR spectra were obtained using a Nexus Nicolet FTIR spectrometer.

The spectra of water-soluble and water-insoluble fractions were obtained in CsI pellets.

The assignment of adsorption bands was made using previous data reported in the literature, (McDonald et al. 1991, 1994; Nascimento et al. 1998; Khare et al. 2002; Imanaka et al. 2004).

Nuclear Magnetic Resonance (NMR). 13C CP MAS-NMR spectra were obtained in a Bruker Advance 400 spectrometer, using a standard cross-polarization pulse sequence.

Samples were spun at 10 kHz. Spectrometer frequencies were set to 100.62 and 400.13 MHz for 13C and 1H, respectively. A contact time of 1 ms and a period between successive accumulations of 5 s were used. The number of scans was 1,600. Chemical shift values were referenced to TMS. 13C NMR spectra of water soluble fractions were obtained using a Mercury 400 Varian NMR spectrometer. The spectra were recorded in D2O. The number of scans was 50,000. Chemical shift values were referenced to TMS. Gas Chromatography–Mass Spectrometry (GC–MS). GC–MS analyses in the full-scan mode were carried out on an Autosystem XL-Turbo Mass Gold (Perkin Elmer) with an Elite-5 column (crossbond 5% diphenyl-95% dimethyl polisiloxane, 30 m×0.25 mm i.d.× 0.25 μm film thickness) and using He as carrier gas.

High-Performance Liquid Chromatography (HPLC). HPLC analyses were carried out on a Surveyor (ThermoFinnigan) with a PDA detector using a Kromasil 100 C18 5 μm 25×0.46 column.

Analytical procedure In the two types of experiments (aqueous aerosol and control experiment without aerosol), we obtained one water soluble fraction and one water insoluble fraction. These fractions were separated by centrifugation and independently analyzed. They were freeze dried using a lyophilizer Cryodos 80 (Telstar). Each fraction was hydrolyzed with 6 M HCl at 110°C for 24 h, freeze dried again to remove water, HCl and any volatile

Orig Life Evol Biosph

organics, and then analyzed as follows. The identification of the peaks of organic compounds were verified by comparing the retention times, mass spectra and UV absorbance spectrum with external standards, purchased from Sigma-Aldrich and Fluka.

Amino acid analysis. The amino acids were characterized by GC-MS after esterification with 2-propanol and acylation with trifluoroacetic anhydride. The GC oven was programmed as follows: 60°C (initial temperature), heated to 130°C at 5°C/min with a hold time of 11 min and heated at 180°C at 10°C/min with a final hold time of 20 min. The temperature of injector was 220°C and the flow rate was 10 ml/min.

Quantitative analysis was done by HPLC after derivatization with phenylisothiocyanate (PITC) and using the following conditions: Solvent A: 50 mM NH4OAc buffer, pH 6.5.

Solvent B: 100 mM NH4OAc-CH3CN (50:50), pH 6.5. 0 min 0% B (100% A), 45 min 70% B (30% A), 46 min 70% B (30% A), 48 min 100% B (0% A). The flow was 2 ml/min, the column was thermostated at 52°C and the chromatogram registered at 254 nm. Purine base analysis: the purine bases (adenine and 2,6-diaminopurine) were characterized by GCMS after treatment with N-methyl-N-trimethylsilyl-trifluoroacetamide containing 1% trimethylchlorosilane and anhydrous pyridine in a ratio 1:3 at 150°C for 30 min. The GC oven was programmed as follows: 60°C (initial temperature), heated to 130°C at 5°C/min with a hold time of 11 min and heated at 180 °C at 10 °C/min with a final time hold 80 min.

The temperature of injector was 220°C and the flow rate of 2.5 ml/min. Quantitative analysis was done by HPLC using the following conditions: Solvent A: 50 mM KH2PO4 buffer, pH 3.

Solvent B: MeOH. 0 min 1% B (99% A), 20 min 95% B (5% A), 40 min 30% B (70% A) at a flow of 1 ml/min. The column was thermostated at 30°C and the chromatogram was recorded at 260 nm. Hydroxy- and carboxylic acids analysis: the hydroxy acids and carboxylic acids were identified by GC-MS by the same method using for the detection of purine bases. Quantitative analysis was done by GC-MS. Other heterocycles analysis: parabanic acid and hydantoin were analyzed before acid hydrolysis. One portion of the water soluble fraction obtained in the aerosol experiment was extracted with chloroform and centrifuged. The pellet is a dark solid that was analyzed by GC-MS using the purine bases analysis protocol explained above.

Results and Discussion Our experimental set-up consists of a glass reactor filled with a gas mixture containing CH4:N2:H2 40:30:30 over a liquid water pool. The reactor is set up in an ultrasonic device that generates an aerosol, by focusing the ultrasound wave (1.8 MHz, 30 W) onto the liquid surface (Figure 1). The estimated aerosol droplet size of about 3 μm, was calculated accordingly with the equation deduced by Lang (1962):

d¼

π σ 4ρ f 2

1= 3

Where σ is water surface tension at 20°C (72.75 erg/cm2): ρ is water density at 20°C (0.99821 g/cm3) and f is the frequency of the ultrasonic ceramic transductor (1.768 MHz).

Two tungsten electrodes attached to a high-voltage generator produce the spark discharge in the aerosol. The generation of aerosol by low power ultrasonic transductors

Orig Life Evol Biosph

Figure 1 Experimental setting for bubble sol simulation: (1) piezoelectric transductor, (2) ultrasonic waves generated in the water cuvette, (3) aerosol fountain induced in the water surface inside the reaction recipient, (4) aerosol cloud, (5) spark discharge, (6) tungsten electrodes, (7) triple way valve.

is extensively used in simulations of aqueous aerosols (King et al. 2004) and in pharmaceutical field (Steckel and Eskandar 2003). This kind of ultrasonic device generates a continuous and stable bubbleâ&#x20AC;&#x201C;aerosolâ&#x20AC;&#x201C;droplet cycle by capillary wave formation. The nebulizer does not generate cavitation bubbles in the bulk, the main mechanism for the sonochemical processes (Dalby et al. 1997; Singh et al. 1998).

With these conditions for the simulation, the spark discharge leads to the formation of organic material in different non volatile phases: a soluble fraction that remains in the liquid water pool as a yellow-brown solution of pH 8.6 (S fraction) and an insoluble fraction (I fraction) isolated by centrifugation. Both phases were analyzed separately.

Elemental analysis. Amount of C and N fixed in the aerosol experiment The S fraction was freeze dried, giving a non-volatile water soluble residue with empirical formula C9H18N5O5 (Table I). This fraction represents the 16% of the total carbon introduced in the experiment (Table I). The I fraction, with an empirical formula C30H45NO3, represents 10.48% of the total carbon. In the aerosol experiment the non volatile organic matter fixed was the 27.46% of C and 6.19% of N of the total C and N input in the system.

Orig Life Evol Biosph TABLE I C and N fixed and other data of the organic material obtained in aqueous aerosol experiment and control experiment S fraction (aerosol) I fraction (aerosol) S fraction (control) I fraction (control) pH (solution) Amount obtained (mg) C fixed (% control) N fixed (% control) C/N C/O Empirical formula

8.65 60.7±4.4 16.34±1.35 5.98±0.52 1.8 1.8 C9H18N5O5

Insoluble 17.8±1.9 10.48±0.34 0.25±0.03 43 5.4 C30H45NO3

8.73 43.3±3.1 16.94±1.21 10.93±0.78 1.8 1.8 C9H16N5O5

Insoluble 25.8±2.8 18.46±2.07 0.99±0.11 21.2 7.6 C106H157N5O14

Values are expressed as mean ± SEM of ten experiments performed in aerosol and control conditions.

The ratios C/N, C/O and C/H are lower in the S fraction than in I fraction. Moreover, the empirical formulas of these fractions suggest important structural differences. These differences are related to the presence of a higher number of functional groups in the S fraction than in the I fraction. In order to demonstrate this, the FTIR and 13C NMR spectra of each fraction were recorded.

Comparison among the organic materials obtained in aqueous aerosol experiment and control experiment The control experiment carried out without aerosol yields a soluble fraction, (pH=8.7), and one insoluble fraction that is collected from the surface of liquid and from the wall of reaction vessel as a thin film. The two phases were analyzed separately in the same way indicated for the experiment with aerosol.

The results of the elemental analysis and the empirical formula for the fractions of control experiment are showed in Table I. The amount of S fraction obtained in the aqueous aerosol experiment is significantly higher than the S fraction in the control experiment (Figure 2). Analogous results were obtained if the amount is expressed as percentage of input C and N fixed in S fraction. This result suggests that the presence of aqueous aerosol increases significantly the amount of soluble organic material synthesized abiotically and the fixation of nitrogen and carbon from atmosphere in form of organic compounds.

On the other hand, the empirical formulas of both S fractions are very similar but it is possible to see clear differences amount their respective FTIR spectra and 13C NMR spectra (see Appendix). In both bulk water soluble fractions are observed the same functional groups but the ratio each other is different. The 13C chemical shift indicates different species with the same kind of functional groups. In particular the presence of aqueous aerosol seems to improve the formation of carboxylic acids.

Nevertheless, the differences among the I fraction in both experiments are not obvious.

The aerosol does not seem to influence in the formation of insoluble solid under the conditions above explained. This could be related with the lack of recirculation of the insoluble solid fraction in the bubblesol cycle.

Significant differences were found in amount of atmospheric elements fixed and spectra among control and aerosol experiment. Taking into account these differences, we performed

Orig Life Evol Biosph Figure 2 Effect of aqueous aerosol in the production of solid organic material by spark discharge activation of a N2-CH4-H2 atmosphere. (a) Amount obtained of each fraction. (b) Carbon fixed in each fraction (% total input carbon in experiment). (c) Nitrogen fixed in each fraction (% total input nitrogen). Values are mean Âą SEM from ten experiments performed in duplicate (aerosol vs. control without aerosol). Significant differences from control were statistically analyzed by t-Student method: **P<0.01.

Orig Life Evol Biosph

the chemical analysis of these fractions, to check the effect of aerosol in the synthesis of molecules of biological interest.

Analysis of molecules with biological interest The acid hydrolysate of the non-volatile S fraction generates a complex mixture of compounds, essentially amino acids (7.5% respect to the total of non volatile organic material obtained), but also carboxylic acids and heterocyclic compounds (Table II, Figures 3 and 4). A significant number of the molecules obtained contain chiral centres, but in our synthesis we do not observe enantiomeric excess and we report the racemic mixture of asymmetric molecules.

Adenine is the most abundant heterocyclic compound with biological interest identified in our aerosol experiment. Adenine has been synthesized using spark discharges in CH4: C2H6:NH3 atmospheres (Yuasa et al. 1984), and with lower yield using proton irradiation of CO:N2:H2O mixtures (Miyakawa et al. 2002a, b). We report a yield of 0.0521% in adenine, in respect to the total non volatile organic matter in a single 72 h spark discharge experiment (Figure 3). This corresponds to 0.015% yield in adenine respect to the total carbon introduced in the system. We show that presence of an aerosol increases the yield of adenine by two orders of magnitude, compared to previous works (Yuasa et al. 1984). Our yield is comparable to the yield obtained in NH4CN solutions (Miyakawa et al. 2002a, b; Levy et al. 2000).

An important remark is that free adenine is present in the soluble fraction, not in the form of an unknown precursor, as stated in previous works. In such experiments adenine was only generated after acid hydrolysis. With respect to the mechanism of adenine formation, it was demonstrated that the key step to the adenine synthesis is the formation of HCN and the tetramer diaminomaleodinitrile, which isomerises when irradiated with UV light, to 4-amino-5-cyanoimidazole (Orgel 2004). Further addition of HCN to this compound yields adenine directly (Sanchez et al. 1968). The synthesis of the HCN tetramer and the formation of adenine compete with the reaction between HCN and formaldehyde, which would lead to the formation of glycolic acid or glycine, and possibly other reactions that imply HCN (Orgel 2004). The detection of glycolic acid and ammonium formiate in our experiment supports the stated mechanism. Moreover, we detect 2,6-diaminopurine (Figure 5), consistent with the synthesis of both purine derivatives by polymerization of ammonium cyanide solutions (Borquez et al. 2005).

We interpret that the increased yield of adenine obtained in our experiments is due to the local enrichment of this compound or its immediate precursors in the liquidâ&#x20AC;&#x201C;gas interface (Donaldson et al. 2004; Dobson et al. 2000). This represents a reduction of importance of bulk-solution interfering reactions. A parallel argument was employed in the effect of montmorillonite on the increase purine yield through its selective adsorption on the clay surface (Cohn et al. 2001). Aerosols have the advantage of the ubiquitous presence in planetary environments with surface liquid pools (Lerman 1992).

We identified other heterocyclic compounds in the sample (Figure 3b). One of the most interesting, from a biological point of view, is the imidazoline-2,4-dione or hydantoin.

Recently, hydantoin has been suggested as a precursor for the emergence of prebiotic peptides and amino acids, under similar conditions of pH and temperature to those of our experiment (Taillades et al. 1998). Moreover, it was hypothesised that primitive microorganism on Earth may be able to use hydantoins as C or N sources. This idea is

Orig Life Evol Biosph TABLE II Identification and quantification of molecules of biochemical interest obtained in spark discharge activation of CH4-N2-H2 atmosphere with or without aqueous aerosol

AMINO ACIDS Glycine Aminomalonic acid Alanine β-Alanine Sarcosine Isoserine Serine Aspartic acid Iminodiacetic acid 2-Aminobutyric acid 3-Aminobutyric acid 3-Aminoisobutyric acid N-Methylalanine Glutamic acid Ornithine Histidine t-Leucine TOTALa HYDROXY ACIDS Glycolic acid 2,3-Hydroxypropanoic acid Malic acid Tartaric acid 2-Hydroxypentanodioic acid 6-Hydroxycaproic acid TOTALa CARBOXYLIC ACIDS 2-Methylbutanodioic acid Succinic acid Glutaric acid Tricarballylic acid TOTALa HETEROCYCLES Parabanic acid Hydantoin Adenine 2,6-diaminopurine TOTALa

Aqueous aerosol experiment Amount (μmol)

Control experiment Amount (μmol)

C2H5NO2 C3H5NO4 C3H7NO2 C3H7NO2 C3H7NO2 C3H7NO3 C3H7NO3 C4H7NO4 C4H7NO4 C4H9NO2 C4H9NO2 C4H9NO2 C4H9NO2 C5H9NO4 C5H12N2O2 C6H9N3O2 C6H13NO2

3.93 1.06 12.98 1.16 0.45 0.34 0.14 0.34 0.03 0.18 0.68 1.65 d 0.08 0.43 d 0.19 1.63a

2.89 0.001 8.68 0.21 1.01 0.12 0.07 0.10 d 0.40 1.02 1.34 0.09 0.16 0.24 – – 1.19a

C2H4O3 C3H6O4 C4H6O5 C4H6O6 C5H8O5 C6H12O3

0.17 0.02 0.03 0.01 0.01 0.02 0.018a

– d d – – – d

C5H8O4 C4H6O4 C5H8O4 C6H8O6

0.02 0.05 d 0.01 0.007a

d 0.01 – – 8 × 10−5

C3H2N2O3 C3H4N2O2 C5H5N5 C5H6N6

0.02 0.06 0.11 0.02 0.022a

– – d – d

Values are expressed as mean of μmol obtained in one experiment (ten experiments performed). Amino acids were identified and confirmed using authentic standards and two techniques (HPLC and GC-MS). Ornithine and histidine are tentatively assigned by HPLC only, using authentic standards. d detected (Noise/Signal>0.1). a

yield (%) respect to total carbon in the system.

Orig Life Evol Biosph

Figure 3 Gas chromatograms showing the carboxylic acids and heterocyclic compounds obtained in a CH4: N2:H2 atmosphere with water aerosol. (a) Analysis of the residue of the water soluble organic fraction after hydrolysis with HCl 6 M and derivatization as stated in the methods section. In min 83 appears the 2,4diaminopurine peak, not shown in the figure. *: this peak corresponds to a possible isomer of tartaric acid.

Both molecules have the same mass spectra. (b) Analysis of the water soluble organic fraction after CHCl3 extraction and before acid hydrolysis.

Orig Life Evol Biosph

based on studies on the hydantoinases group, enzymes that catalysed the hydrolysis of hydantoins, (Syldatk et al. 1999). We demonstrate that the availability of hydantoins is a possibility that could occur on the early stages of the biochemical or biological evolution.

In addition, we have found dihydroxy acids in the aerosol experiments (absent in the control experiments without aerosols). We report the prebiotic synthesis of 2,3dihydroxybutanedioic acid (tartaric acid) and 2,3-dihydroxypropanoic acid (glyceric acid). These compounds have not yet been identified in experiments in prebiotic synthesis.

Other biologically relevant molecules identified in the products of spark discharge in the aerosol environment are the propane-1,2,3-tricarboxylic acid (tricarballylic acid, Figure 3a) and succinic and malic acids. The succinic and malic acids are members of the Krebs cycle and the tricarballylic acid is structurally related to citric and isocitric acid that are also members of the Krebs cycle, as well. It has been suggested the possibility that these carboxylic acids, including tricarballylic acid, could take part in a primordial variant of the Krebs cycle (MelĂŠndez-Hevia et al. 1996). However, the hypothesis on self-organizing protometabolic cycles is a controversial issue (Orgel 2000). Anyway, our experiment demonstrates that aerosol chemistry could have contributed to the production of the raw materials for this suggested primordial metabolism, in addition to previous work in synthesis of dicarboxylic acids by irradiation of acetic acid solutions (Negron-Mendoza and Ponnamperuma 1976). Further work must be performed to explore and optimise the potential of the aerosol as a prebiotic reactor. In fact, our future prospects include testing the

Figure 4 HPLC chromatogram showing the amino acid profile obtained in a CH4:N2:H2 atmosphere with water aerosol. The analysis of the residue of the water soluble organic fraction after hydrolysis with HCl 6 M and derivatization with PITC as stated in the methods section. All amino acid were identified by two different methods (HPLC and GCâ&#x20AC;&#x201C;MS) against authentic standards, with the exception of histidine and ornithine, identified tentatively. We confirm the synthesis of t-leucine, but we can not assign any peak to Leu, Ile or Val.

Orig Life Evol Biosph 90

N

80

N

N

N H NH2 Adenine

70

60

mAU

50

40

NH2 N

30

H2N

N

H N N

2,6-Diaminopurine

20

10

0 0

2

4

6

8

10 Minutes

12

14

16

18

20

Figure 5 HPLC chromatogram showing the purine bases profile obtained in a CH4:N2:H2 atmosphere with water aerosol. The analysis of the residue of the water soluble organic fraction after hydrolysis with HCl 6 M was performed in C18 RP column after previous separation of heterocyclic fraction by 2D-TLC.

effects of a solid, organic or inorganic, core in the aerosol (formed by partial evaporation of aerosol solution droplets), because the core could improve the catalytic properties of the droplet (Lerman and Teng 2004).

In summary, the chemical analysis of the S fractions of aerosol and control experiments are consistent with the structural data. The formation of polar compounds with biological interest (i.e., amino acids, purine bases and carboxylic acids) is improved in the presence of aqueous aerosols.

Conclusions We can conclude that organic molecules of potential biochemical interest obtained in the presence of an aqueous aerosol show greater amounts and diversity than the material obtained using the same conditions without aerosol (see Figure 6). The experimental evidence obtained in this work support the hypothesis made by Shah and developed by Lerman, in which the aerosol droplet behaves as a microscopic chemical reactor that offers a number of potential advantages for prebiotic synthesis.

Acknowledgments The authors acknowledge the Centro de Astrobiología (CAB) for the research facilities and the grants from Instituto Nacional de Técnica Aeroespacial ‘Esteban Terradas’ (INTA). We especially thank Juan Pérez Mercader for his continuous efforts for the development of Prebiotic Chemistry in the CAB. We acknowledge the thorough and constructive revision and comments of Alan W. Schwartz, James H. Cleaves, Antonio Lazcano, Adrian Tuck, Louis Lerman, Guillermo Muñoz Caro and M. Paz Zorzano.

Orig Life Evol Biosph Figure 6 Effect of aerosol in the production of chemical families with astrobiological interest. (a) Yield of amino acids in the total amount of solid organic material. (b) Yield of carboxylic acids in the total amount of S fraction. (c) Yield of heterocycles obtained in the total amount of S fraction. The control value for heterocycles is zero because the detected compounds in the experiment without aerosol are under quantification limit of the technique. Values are expressed as mean Âą SEM of ten experiments. Significant differences were determined by Student t-test (***P<0.001).

Orig Life Evol Biosph

Appendix Spectroscopic analysis of the organic material in aerosol simulation experiments

FTIR Spectra The IR spectrum of S fraction (Figure 7a) shows features at 3,321 and at 3,207 cm−1 which are assigned to O-H stretching mode and N-H stretching mode, respectively (Table III). At 1,671 cm−1 a very strong band is observed, which is associated to the C=O bending mode.

At 2,250 cm−1 the spectrum shows a weak band assigned to R C N stretching. The bands characteristic of the asymmetric and symmetric stretching and bending modes, respectively, for –CH3 and –CH2 groups are observed like weak bands at 2,977, 2,933, 2,880 and 1,452, 1,379 cm−1, respectively.

The IR spectra of the I fraction also show (Figure 7c) bands that could be assigned to NH or O-H bonds (3,435 cm−1). Indeed, the IR spectrum of I fraction show a feature at 1,707 cm−1 associated with C=O bonds, a band at 1,536 cm−1 that may be correlated with amide II bonds or N-H bending mode (also present in the S fraction as a weak band at 1,541 cm−1) and bands that probably are due to alkenes, alkynes, nitriles and imines groups.

In the spectra of I fractions a band at 1,955 cm−1 can be seen that may be due to C=C=C (cumulene system) or C=C=CHCONH2. Finally, it is important to indicate that the features related to –CH3 and –CH2 groups are actually enhanced in the I fraction with respect to the S fraction. For a complete assignation of all bands in S and I fractions, see Table III.

NMR Spectra The solid-state 13C NMR spectrum of the S fraction shows four groups of broad resonances about 166, 126, 78 and 47 ppm (Figure 8a) In order to resolve these bands, the spectra were recorded in solution of deuterium oxide, and in this case the previous broad resonances could be resolved. As a result in the 13C NMR (D2O, 100 MHz) of S fraction a great number of resonances are observed in a very complex spectrum (Figure 9a). Nevertheless, as a first approximation, it is possible to group all resonances in six main types of carbons.

The resonances about 180–160 ppm indicate the presence of amides and/or carboxylic groups (carboxylic acid or esters). It is remarkable that resonances at 210–190 ppm related with ketones and aldehydes were absent. The resonances about 120–110 ppm can be assigned to nitriles groups and in the region 80–65 ppm are observed resonances that may be due to C-OH groups. The possible amine groups are observed about 65–40 ppm. The resonances in the 37–15 ppm range indicate likely –CH and –CH2 groups and finally the resonances about 14–9 ppm may be assignment to –CH3. All these resonances are consistent with the observed features in the IR spectrum; in particular, the assignation of the continuous signal among 3,350 and 1,700 cm−1 (Figure 7a) to the presence of carboxylic acid is supported by the high number of resonances about 180–160 ppm.

In the solid-state 13C NMR spectra of I fraction, broad resonances are observed (Figure 8b). The signals that could be due to nitriles (–C≡N) and alkenes (C=C) are overlapped (centred resonances around 129 and 115 ppm). The resonances near 75 ppm can be assigned to hydroxylic carbon (C-OH) but also it is possible an alkyne group (C≡C) contribution. A very broad resonance is observed in the range 60–10 ppm. In these signal is

Orig Life Evol Biosph

a 2250 1541 1627

1671 3321

2143

Reflectance(a.u)

b

2248

1554

3223

3448

1615 1672

3347 1728

c

2730

1956

2235 3309 3435

1706

2873

2960

1536

2931

1377

1458

d

2729

2111 969

1955 2235 2216

1529

3310 1746 3437

1676

3389 1711

4000

3000

2000

1000

Wavenumber (cm-1) Figure 7 Transmission FTIR spectrum obtained from water soluble and insoluble fractions of CH4–N2–H2 aqueous aerosol spark experiments and CH4–N2–H2 control experiments. (a) Water soluble fraction, S, from aqueous aerosol experiment; (b) Water soluble fraction, S, from control experiment; (c) Insoluble fraction, I, from aqueous aerosol experiment; (d) Insoluble fraction, I, from control experiment.

Orig Life Evol Biosph TABLE III Characteristic frequency of IR absorption spectra obtained from water soluble and insoluble fractions of CH4-N2-H2 aqueous aerosol spark experiments and CH4-N2-H2 control spark experiment D fraction (aerosol)

D fraction (control)

S fraction (aerosol)

S fraction (control)

Vibrational groups identity

3448 (sh)

3435 (b, m)

3437 (b, m)

N-H stretching (single bond) or NH2 asym stretching or O-H stretching NH2 asym stretching or O-H stretching O-H stretching C≡CH N-H stretching or overtone NH2 bending C=CH2 streching C-H stretching (CH3 asym) C-H stretching (CH2) C-H stretching (CH3 sym) N-CH3 or N-CH2R-CN stretching or R-C≡C-R C≡C or conjugated nitriles

3389 (b, m) 3321 (b, s)

3347 (b, s)

3207 (b, s)

3223 (sh)

2977 (v w) 2933 (v w) 2880 (sh)

2971 (w) 2936 (w) 2878 (w)

2250 (w)

2248 (w)

3309 (m)

3310 (m)

3075 2960 2931 2873 2730 2235 2209

3078 2961 2932 2873 2729 2235 2216

(sh) (v s) (v s) (v s) (w) (w) (w)

(v w) (v s) (v s) (v s) (w) (m) (m)

2143 (w) 2112 (v w) 1956 (w)

1671 (v s)

1728 (sh) 1672 (v s)

1706 (v s) 1671 (v s)

2111 (w) 1955 (m) 1746 (sh) 1711 (v s) 1676 (sh)

1627(s) 1541 (sh)

1615 (sh) 1554 (sh)

1536 (v w)

1529 (w)

1452 (w)

1452 (sh) 1410 (m) 1382 (m) 1234 (w) 1199 (w) 1083 (b, m)

1458 (v s)

1460 (m)

1377 (v s)

1379 (m)

1379 1234 1207 1096

(b, m) (w) (v w) (w)

1082 (b, w) 1073 (b, m)

998 (sh)

994 (w)

776 (w)

782 (w)

970 (w)

969 (w)

N=C=O or N=C=N C=C=C or C=C=CHCONH2 C=O stretching C=O bending or N-C=C, O-C=C or C=C (stretching) or CH=NN=CH N-H bending N-H bending (possible amide) or aromatics (quadrant stretch) or C=C (+ contribution C=N, N=N) CH2 asym bending CH3 sym bending C-O st (arc-OH) or C-C / C-N stretching C-O vibrational mode C-O vibrational mode or C=C= st si Vinyl C-H bending or C-N-C (saturated heterocycles) Vynil C-H bending Aromatics/N-H bending

The frequency is expressed in cm−1 . Intensity code: (b) = broad, (m) = medium, (s) = strong, (v s) = very strong, (sh) = shoulder, (w) = weak, (v w) = very weak.

Orig Life Evol Biosph

a

S fraction (aerosol experiment) 166 ppm

61 ppm

47 ppm 24 ppm

78 ppm 126 ppm

31 ppm

b

I fraction (aerosol experiment) experiment

12 ppm 129 ppm

171 ppm

400

300

200

76 ppm

100

0

-100

-200

13

C Chemical Shift

Figure 8 Solid-state 13C NMR spectra from water soluble (a) and insoluble fractions (b) of CH4-N2-H2 aqueous aerosol spark discharge experiments.

likely that they are overlapping the resonances corresponded to amine (–CNH-R) and –CH, –CH2 and –CH3 groups. An additional resonance at 171 ppm is observed in the I fraction (Figure 8) that could be assigned to C=O of either carboxylic acids or esters.

With the structural information present, we could infer that the water soluble (S) fraction is a complex mixture, where there are carboxylic acids, amines, alcohols, and nitriles and it is possible the presence of amides and esters. In this raw mixture we could not observe other functional groups like ketones, aldehydes or C=C from alkenes or aromatics.

Therefore, in any case the S fraction obtained from aqueous aerosol is constituted by different polar units. By the contrary, insoluble (I) fractions are apolar solids, insoluble in water and in the common organic solvents. The fraction I seem to consist in large and rigid

47.9

Orig Life Evol Biosph

S fraction (aerosol experiment)

225

225

200

175

150

125

200

175

150

125

100

100

60.7

76.6

119.3 110.1

162.8

171.0

a

75

50

50

75

25

S fraction (control experiment)

175 175

28.6 47.9

118.4

150 150

0

b

60.7

200 200

160.6

179.1 11.9

225 225

0

25

125 125 13C

100 100

75 75

50 50

25 25

00

Chemical Shift

Figure 9 13C NMR (D2O) spectra from water soluble fractions (S) from CH4-N2-H2 aqueous aerosol spark experiments and from the control experiment. (a) Water soluble fraction from aqueous aerosol experiment; (b) Water soluble fraction from control experiment.

hydrocarbon chains. The occasional presence of unsaturated bond (C=C, Câ&#x2030;ĄC) increase the rigidity of the chains and the amine and hydroxyl groups could favour the hydrogen bond interactions among them. Both factors contribute to the low solubility of I fraction.

Orig Life Evol Biosph

References Bada JL (2004) How life began on Earth: a status report. Earth Planet Sci Lett 226:1–9 Borquez E, Cleaves HJ, Lazcano A, Miller SL (2005) An investigation of prebiotic purine synthesis from the hydrolysis of HCN polymers. Orig Life Evol Biosph 35:79–84 Civis S, Juha L, Babankova D, Cvacka J, Frank O, Jehlicka J, Kralikova B, Krasa J, Kubat P, Muck A, Pfeifer M, Skala J, Ullschmied J (2004) Amino acid formation induced by high-power laser in CO2/CON2-H2O gas mixtures. Chemical Physics Letters 386:169–173 Cohn CA, Hannson TK, Larrson HS, Sowerby SJ, Holm NG (2001) Fate of prebiotic adenine. Astrobiology 1:477–480 Dalby R, Tiano S, Hickey A (1997) Medical devices for the delivery of therapeutic aerosols to the lungs.

In: Hickey A (ed) Inhalation aerosols vol. 94. Dekker, New York Delano J (2001) Redox history of the Earth’s interior since ∼3900 Ma: implications for prebiotic molecules.

Orig Life Evol Biosph 31:311–341 Dobson CM, Ellison GB, Tuck AF, Vaida V (2000) Atmospheric aerosols as prebiotic chemical reactors.

Proc Natl Acad Sci USA 97:11864–11868 Donaldson DJ, Anderson D (1999) Adsorption of atmospheric gases at the air–water interface. 2: C1–C4 alcohols, acids and acetone. J Phys Chem, A 103:871–876 Donaldson DJ, Tervahattu H, Tuck AF, Vaida V (2004) Organic aerosols and the origin of life: an hypothesis.

Orig Life Evol Biosph 34:57–67 Ellison GB, Tuck AF, Vaida V (1999) Atmospheric processing of organic aerosols. J Geophys Res 104:11633–11641 Imanaka H, Khare BN, Elsila JE, Bakes ELO, McKay CP, Cruikshank DP, Sugita S, Matsui T, Zare RN (2004) Laboratory experiments of Titan tholin formed in cold plasma at various pressures: implications for nitrogen-containing polycyclic aromatic compounds in Titan haze. Icarus 168:344–366 Khare BN, Bakes ELO, Imanaka H, McKay CP, Cruikshank DP, Arakawa ET (2002) Analysis of the timedependent chemical evolution of Titan haze tholin. Icarus 160:172–182 King MD, Thompson KC, Ward AD (2004) Laser tweezers Raman study of optically trapped aerosol droplets of seawater and oleic acid reacting with ozone: implications for cloud–droplet properties.

J Am Chem Soc 126:16710–16711 Lang RJ (1962) Ultrasonic atomization of liquids. J Acoust Soc Am 34:6–8 Lerman L (1992) The role of the air–water interface in prebiotic chemistry. PhD thesis, Stanford University Lerman L (1994) The bubble–aerosol–droplet cycle as a natural reactor for prebiotic chemistry (I). Orig Life Evol Biosph 24:111–112 Lerman L (1996) The bubble–aerosol–droplet cycle: a prebiotic geochemical reactor (that must have been).

Orig Life Evol Biosph 26:369–370 Lerman L, Teng J (2004) In the beginning. In: J Seckbach (ed) Origins: genesis and diversity of life vol. 6.

Springer, Berlin Heidelberg New York Levy M, Miller SL, Brinton K, Bada JL (2000) Prebiotic synthesis of adenine and amino acids under Europalike conditions. Icarus 145:609–613 McDonald GN, Khare BN, Thompson WR, Sagan C (1991) CH4/NH3/H2O Spark tholin: chemical analysis and interaction with jovian aqueous clouds. Icarus 94:354–367 McDonald GN, Thompson WR, Heinrich M, Khare BN, Sagan C (1994) Chemical investigation of Titan and Triton tholins. Icarus 108:137–145 Meléndez-Hevia E, Wadell TG, Cascante M (1996) The puzzle of the Krebs citric acid cycle: assembling the pieces of chemically feasible reactions and opportunism in the design of metabolic pathways during evolution. J Mol Evol 43:293–303 Miller SL (1953) A production of amino acids under possible primitive Earth conditions. Science 117:528– 529 Miyakawa S, Cleaves HJ, Miller SL (2002a) The cold origin of life: B. Implications based on pyrimidines and purines produced from frozen ammonium cyanide solutions. Orig Life Evol Biosph 32:209–218 Miyakawa S, Yamanashi H, Kobayashi K, Cleaves HJ, Miller S (2002b) Prebiotic synthesis from CO atmospheres: implications for the origins of life. Proc Natl Acad Sci USA 99:14628–14631 Murphy DM (2005) Something in the air. Science 307:1888–1889 Murphy DM, Thomson DS, Mahoney MJ (1998) In situ measurements of organics, meteorite material, mercury, and other elements in aerosols at 5 to 19 kilometers. Science 282:1664–1667 Nascimento LFC, Mota RP, Valenca GP, Teixeira JC, Algatti MA, Honda RY, Bortoleto JRR (1998) An N2:CH4:H2O glow discharge plasma probed by optical and electric techniques: significance to the radiation chemistry of Titan’s upper atmosphere in the presence of meterorite water. Planet Space Sci 46:969–974

Orig Life Evol Biosph Negron-Mendoza A, Ponnamperuma C (1976) Formation of biologically relevant carboxylic acids during the gamma irradiation of acetic acid. Orig Life 7:191–196 Orgel LE (2000) Self-organizing biochemical cycles. Proc Natl Acad Sci USA 97:12503–12507 Orgel LE (2004) Prebiotic adenine revisited: eutectics and photochemistry. Orig Life Evol Biosph 34:361–369 Ring D, Wolman Y, Friedmann N, Miller SL (1972) Proc Natl Acad Sci USA 69:765–769 Sanchez RA, Ferris JP, Orgel LE (1968) Studies in prebiotic synthesis IV. Conversion of 4-aminoimidazole5-carbonitrile derivatives to purines. J Mol Biol 30:223–253 Schlesinger G, Miller SL (1983) Prebiotic synthesis in atmospheres containing CH4, CO and CO2. J Mol Evol 19:376–382 Shah DO (1970) The origin of membranes and related surface phenomena. In: C Ponnamperuma (ed).

Exobiology, chapter 7. North-Holland, Amsterdam, pp 235–265 Singh V, Kaur KP, Khurana A, Kad GL (1998) Ultrasound: a boon in the synthesis of organic compounds.

Resonance 3:56–61 Steckel H, Eskandar F (2003) Factors affecting aerosol performance during nebulization with jet and ultrasonic nebulizers. Eur J Pharm Sci 19:443–455 Syldatk C, May O, Altenbuchner J, Mattes R, Siemann M (1999) Microbial hydantoinases – industrial enzymes from the origin of life? Appl Microbiol Biotechnol 51:293–309 Taillades J, Beuzelin I, Garrel L, Tabacik V, Bied C, Commeyras AN (1998) N-carbamoyl-alpha-amino acid rather than free alpha-amino acid formation in the prebiotic hydrosphere: a novel proposal for the emergence of prebiotic peptides. Orig Life Evol Biosph 28:61–77 Tervahattu H, Juhanoja J, Kupiainen K (2002) Identification of an organic coating on marine aerosol particles

by TOF-SIMS. J Geophys Res 107: http://dx.doi.org/10.1029/2002JD001403.

Tuck A (2002) The role of atmospheric aerosols in the origin of life. Surv Geophys 23:379–409 Yuasa S, Flory D, Basile B, Oro J (1984) Abiotic synthesis of purines and other heterocyclic compounds by the action of electrical discharges. J Mol Evol 21:76–80

Orig Life Evol Biosph (2011) 41:331–345 DOI 10.1007/s11084-010-9232-z PREBIOTIC CHEMISTRY

Prebiotic Synthesis of Protobiopolymers Under Alkaline Ocean Conditions Marta Ruiz-Bermejo & Luis A. Rivas & Arantxa Palacín & César Menor-Salván & Susana Osuna-Esteban

Received: 22 September 2010 / Accepted: 29 November 2010 / Published online: 16 December 2010 # Springer Science+Business Media B.V. 2010

Abstract Clasically, prebiotic chemistry has focused on the production and identification of simple organic molecules, many of them forming part of “intractable polymers” named tholins. In a previous work, we demonstrated that in experiments using an external energy source and inorganic carbon the aqueous aerosols improved the formation of hydrophilic tholins. Herein, we elucidate the role of pH (from 4 to 12) in prebiotic experiments using saline aqueous aerosols, spark discharges and an atmosphere containing CH4. At all values of pH, the saline aqueous aerosols increased the production of a significant variety of carboxylic acids that could have been present in a primitive Krebs cycle. Moreover, the study for the first time of hydrophilic tholins by 2-D electrophoresis revealed that these are formed by a set of unexpected heavy polymeric species. The initial alkaline conditions significantly increased both the apparent molecular weight of polymeric species up to 80 kDa and their diversity. We propose the term of protobiopolymers to denote those polymeric species fractionated by 2-D electrophoresis since these are formed by biomolecules present in living systems and show diversity in length as well as in functional groups. Thus, aerosols formed in simulated alkaline ocean conditions could provide an optimal medium for the formation of the primeval materials that could be precursors to the emergence of life.

Keywords Prebiotic synthesis . Polymer condensation . Aerosol chemistry . Tholins .

Alkaline oceans . Prebiotic environments

M. Ruiz-Bermejo (*) : L. A. Rivas : C. Menor-Salván : S. Osuna-Esteban Departamento de Evolución Molecular, Centro de Astrobiología [Consejo Superior de Investigaciones Científicas-Instituto Nacional de Técnica Aeroespacial (CSIC-INTA)], Carretera Torrejón-Ajalvir, Km 4, 28850 Torrejón de Ardoz, Madrid, Spain e-mail: ruizbm@inta.es A. Palacín Unidad de Bioquímica, Departamento de Biotecnología, E.T.S. Ingenieros Agrónomos, UPM, Ciudad Universitaria 28040 Madrid, Spain

332

M. Ruiz-Bermejo et al.

Introduction Traditionally, simulation experiments aimed at the synthesis of simple bioorganic molecules under prebiotic conditions have been carried out using many different gas mixtures containing carbon (i.e., CH4, CO2 or CO) and nitrogen (i.e., NH3 or N2) along with different external energy sources (i.e., electric discharges, ultraviolet irradiation, highenergy particle, photon beams, etc.) to simulate the conditions of the early Earth, other planetary systems or the interstellar medium. In most of these prebiotic simulation experiments, water has been an essential reactant, because it appears that early life would need to exist in a liquid medium. Water-based aerosols are ubiquitous in the current troposphere (Ellison et al. 1999), and it has been suggested that they may have played an important role as prebiotic microreactors for the origin of life (Dobson et al. 2000; Donaldson et al. 2004; Tervahattu et al. 2004; Tuck 2002). Additionally, differences in reactivity between bulk water in any physical state and water present in air-water interfaces have been suggested (Donaldson and Vaida 2006). Recently, we observed these differences experimentally (Ruiz-Bermejo et al. 2007a). Under simulated prebiotic conditions using aqueous aerosols, spark discharges and an atmosphere of CH4/N2/H2, we found amino acids, mono- di- and tri- carboxylic acids, purine bases and other heterocycles in greater amounts and diversity than those detected in control experiments using a pool of liquid water. This result supports the hypothesis proposed by Shah (Shah 1970) suggesting that aerosol droplets act as microscopic chemical reactors that offer a number of potential advantages for prebiotic synthesis. Moreover, the formation of aerosols is a common process because it depends only on a liquid water-air interface and a physical mechanism that ejects bubbles into the atmosphere, such as wind, sea waves or shock waves (Ellison et al. 1999).

In addition, it has been proposed that an organic or inorganic solid core, formed by partial evaporation of the saline aerosol droplets, could improve the catalytic properties of the aerosol droplets (Lerman and Teng 2004). In such a case, aerosols and salts would act in synergy in the prebiotic synthesis of organic compounds. In this context, we tested the effect of saline aqueous aerosols in the formation of bioorganic compounds under plausible prebiotic conditions (Ruiz-Bermejo et al. 2007b). For that test, aqueous solutions with a salinity 1.5 times the concentration of actual sea water, the likely concentration of the ancient sea, and a pH=5.8 (Knauth 2005; Morse and Mackenzie 1998) were used. In those experiments, we observed that the presence of saline aerosols enhanced the yield of hydroxy- and carboxylic acids and increased the amount and variety of amino acids.

In spite of fact that the salinity and pH of the aqueous phase may have an important influence on gas-liquid interfaces (Tervahattu et al. 2004), to our knowledge no systematic studies have been performed to elucidate the role of pH in prebiotic experiments, using an external energy source and inorganic carbon. This lack of study could be due to the controversial question of the oceanic pH on the early Earth. To address this question, Grotzinger and Kasting (1993) provided arguments for a relative constancy of pH, Kempe and Kazmierczak (1994) favour a so-called soda ocean (alkaline oceans analogous to modern terrestrial soda lakes) and Macleod et al. (1994) and Russell and Hall (1997) concluded that the early ocean was acidic.

The goal of this work was to study the role of pH in the production of simple bioorganic compounds and in the formation of high molecular weight polymeric species obtained from them under plausible prebiotic conditions. Although it is considered that the early Earthâ&#x20AC;&#x2122;s atmosphere was dominated by carbon dioxide (Walker 1985), for the simulation experiments we chose a mixture containing CH4 for comparative proposes with our previous

Prebiotic Synthesis of Protobiopolymers

333

results (Ruiz-Bermejo et al. 2007a, b; 2008) and because the presence of CH4 favours the formation of greater amount of organic material (Schlesinger and Miller 1983).

Furthermore, the aerosol mechanisms would presumably be effective in other atmospheres.

The difficulties in finding significant differences between the bulk organic materials, commonly named “intractable polymers” or tholins (Sagan et al. 1993; Khare et al. 1986; Sagan and Khare 1979), obtained in several sets of prebiotic experiments, by conventional techniques such as IR spectroscopy or GC-MS were overcome, in this work, by the use of 2-dimensional (2-D) electrophoresis. This 2-D classical technique, which is widely used for the analysis of complex protein mixtures, has not been previously used for the study of tholins. Thus, using this technique with the experiments performed under alkaline ocean conditions, we detected polymeric species with very high molecular weights not previously reported in the literature. The nature of these polymeric species is discussed, and we propose the name of “protobiopolymers” for them due to the nature of their monomers and their potential properties.

Methods Simulation of Prebiotic Marine Aerosols at Different pH Values A 500-ml glass reactor was filled with a gas mixture containing CH4:N2:H2 (40:30:30), purchased from Praxair, and 5 mL of a saline solution 1.5 times the concentration of sea water [0.63 M NaCl, 0.18 M MgCl2, 0.015 M KCl, 0.02 M NaHCO3 (Morse and Mackenzie 1998; Ruiz-Bermejo et al. 2007b) adjusted to various pH levels: Experiment 1, pH=4 with HCl (Russell and Hall 1997). Experiment 2, reference experiment, pH=5.8 with HCl (Morse and Mackenzie 1998). The details of this experiment are reported in RuizBermejo et al. (2007b). Experiment 3, pH=7.8, final pH of the saline solution indicated above; Experiment 4, pH=9.8 with NaOH (Kempe and Kazmierczak 2002). Experiment 5, pH=12 with NaOH (Holm and Neubeck 2009). A control experiment with pure water was performed; the details of this control experiment are reported in Ruiz-Bermejo et al.

(2007a). NaCl, MgCl2, CaCl2 and KCl, and NaOH and analytical grade HCl (36%) were obtained from Panreac and NaHCO3 was obtained from Fluka. The solutions or suspensions were bubbled with dried N2 for 1 h, and prior to use, all glassware and electrodes were heated in a high-temperature oven (Nabertherm Labotherm L5) at 400°C in air for 2 h, to eliminate any possible organic contaminants. Before each experiment, the system was evacuated with a membrane pump and purged with the reaction gas mixture 4 times. After this process, the reactor was filled until a pressure of 1,200 mbar was reached.

An ultrasonic aerosol generator (BONECO model 7035) working at 1.8 MHz and 33 W was used for the bubble-sol cycle that generated the saline aerosol in a few minutes (RuizBermejo et al. 2007a and b). Two tungsten electrodes attached to the reactor were used with a high voltage generator (Model BD-50E, Electrotechnic Products Inc. Illinois, USA) to produce spark discharges (50 kV). The system was maintained at constant temperature (38°C) with active aerosol and electric discharge for 72 h. After this period, the liquid solution and the solid material were recovered for analytical and spectroscopic studies.

Instrumental Analyses Infrared Spectroscopy IR spectra were obtained using a Nexus Nicolet FTIR spectrometer.

The spectra were obtained in CsI pellets using the reflectance mode of operation.

334

M. Ruiz-Bermejo et al.

Gas Chromatography–Mass Spectrometry (GC-MS) GC-MS analyses in the full-scan mode were carried out on an 6850 network GC system coupled to a 5975 VL MSD with a tripleaxis detector operating in electronic impact (EI) mode at 70 eV (Agilent), using an HP-5MS column (crossbond 5% diphenyl-95% dimethyl polysiloxane, 30 m×0.25 mm i.d.×0.25 μm film thickness) and He as the carrier gas.

Analytical Procedure In all experiments, we collected a solution and an insoluble solid. These were separated by centrifugation and immediately freeze-dried using a standard lyophiliser; in this way, we obtained a hydrophilic tholin and a hydrophobic tholin. All samples were stored at −20°C under a nitrogen atmosphere. Prior to the analysis of organics, the hydrophilic tholins from each experiment were desalted by ion exchange chromatography [Dowex 50 W X 8-400 (H+)], using 5 N NH4OH as a first eluant and water as a second eluant, obtaining in all cases an ammonium fraction (FNH3) and an aqueous fraction (FH2O). Both fractions were lyophilised and subsequently weighed.

For the identification of organic molecules with biological interest in all freeze-dried fractions, FNH3 and FH2O, the following protocol was used: i) The samples were hydrolysed with 6 M HCl at 110°C for 24 h and then freeze dried to remove water, HCl and any volatile organics. ii) Approximately six milligrams of each hydrolysed sample in 120 μL of BSTFA with 1% TMCS [N,O-bis(trimehtylsilyl)trifluoroacetamide with Trimethylchrolosilane, from Thermo Scientific] was heated at 70°C for 19 h, to obtain the respective TMS derivatives. iii) The derivatised samples were analysed by GC-MS using the following GC oven program: 60°C (initial temperature) with a hold of 1.5 min, heated to 130°C at 5°C/min with a hold time of 11 min, heated to 180°C at 10°C/min with a hold time of 10 min and heated to 220°C/min at 20°C/min with a final hold time of 15 min. One microlitre of each sample was injected. The temperature of the injector was 220°C, and the injections were performed in splitless mode. The detector temperature was 300°C. The flow rate was 1.1 mL/min. As a rule, the identification of the GC-MS peaks attributed to organic compounds was verified by comparison with the retention times and mass spectra of external standards (purchased from Sigma-Aldrich and Fluka).

Tricine-SDS Polyacrylamide Gel Electrophoresis One hundred micrograms of dried FNH3 and FH2O fractions were fractionated by the method of Laemmli (1970) in 10%– 20% Tris-Tricine gels using the XCell SureLock™ Mini-Cell (Invitrogen, Carlsbad, CA, USA).

Two-Dimensional Gel Electrophoresis Two-dimensional electrophoresis (2-D electrophoresis) of FNH3 fractions was performed as described by Görg et al. (2000), using an Immobiline DryStrip gradient (pH 3–10; 7 cm; Amersham Biosciences) for the first dimension (Isoelectric focusing; IEF). Two hundred fifty micrograms of dried polymeric FNH3 fractions were dissolved in 200 μL DeStreak Rehydration Solution with 0.5% IPG buffer pH 3–10 and loaded onto the IPG strip. IEF was performed in IPGPhor-I (Amersham Biosciences), finally reaching 8000 Vh. After the separation, the first dimension strip was equilibrated twice with equilibration buffer (50 mM Tris–HCl pH 8, 6 M urea, 30% glycerol, 2% SDS and 0.02% Bromophenol Blue). SDS-PAGE in the second dimension was performed with 15% polyacrylamide gels. After electrophoresis, the gels were stained

Prebiotic Synthesis of Protobiopolymers

335

with the PlusOne™ Silver Staining Kit (BioRad, Hercules, CA, USA) following the manufacturer’s instructions.

Results Prebiotic Experiments Performed at Different Values of pH Using Saline Water Aerosols.

Total Amount of Non Volatile Organic Material Formed Several experiments were performed to test the effect of pH on the production and nature of the organic material obtained under the plausible prebiotic conditions of the ancient sea. To that end, we used spark discharges, a reductive atmosphere of CH4 and saline aqueous aerosols at different pH values in unbuffered medium. We carried out the experiments at pH=4 (Experiment 1), pH=7.8 (Experiment 3), pH=9.8 (Experiment 4) and pH=12 (Experiment 5) to cover a wide range of pH values. As a reference experiment, we used saline aerosols at pH=5.8 (Experiment 2), and as control experiments, we used aqueous aerosols of pure water at neutral pH. Details about the reference and control experiments are reported in Ruiz-Bermejo et al. (2007b) and RuizBermejo et al. (2007a), respectively. Five experiments for each pH value were carried out (Fig. 1). Independent of the initial pH in each experiment, the solutions had a pH between 8.0 and 8.8 when the experiment was complete. The alkalinity of the final solutions was due to the presence of ammonium ions, which was tested by the Nessler assay [see details about this assay in Ruiz-Bermejo et al. (2008)]. In every experiment, after a 72-h reaction time, a brownish solution and an insoluble solid were collected; both were separated by centrifugation and then freeze-dried. In this way, a water-soluble organic material, hydrophilic tholin, and an insoluble water solid, hydrophobic tholin, were collected and weighed (Fig. 1a). The experiments showed that the higher initial pH value led to the higher amount of hydrophobic tholins. In contrast, a higher amount of hydrophilic tholin was produced in those experiments with an initial pH closer to neutral (Fig. 1a).

Taking into account the total non-volatile carbon fixed for each set of experiments, a pH of ~ 7–8 seemed to be most favourable for the formation of the non-volatile organic material.

Fractionation of the Hydrophilic Tholins by Ion Exchange Resin To determine the content of organic molecules and the presence of possible polymeric species in the hydrophilic tholins, these bulk materials were desalted using an ion exchange resin with NH4OH and water as eluants. The ammonium and aqueous fractions, FNH3 and FH2O, respectively, separated by this method were freeze-dried, and the solids were collected and were weighed. The FNH3/FH2O ratio increased notably as the pH increased. At an acidic pH, the dried-weight of FH2O was roughly twice that of FNH3, while at highly alkaline pH, the dried-weight of FH2O was practically the same as that of FNH3 (Fig. 1b).

Identification of Organic Molecules of Biological Interest by GC-MS To identify organic molecules of biological interest, the FNH3 and FH2O fractions were analysed for amino acids, carboxylic acids and heterocycles containing nitrogen by GCMS. To accomplish this goal, first the FNH3 and FH2O fractions from each set of experiments were acid hydrolysed, and then the freeze-dried and lyophilised materials were derivatisated

336

M. Ruiz-Bermejo et al.

Experiment 1

2(a)

3

4

5

Control (b)

8.4 ± 0.1

8.6 ± 0.1

8.0 ± 0.1

8.0 ± 0.1

8.8 ± 0.1

8.7

Total soluble material (mg)

59 ± 5

69 ± 8

71 ± 5

57 ± 7

55 ± 6

61 ± 4

Total insoluble material (mg)

7±1

10 ± 2

14 ± 3

14 ± 3

24 ± 4

18 ± 2

66

79

85

71

79

79

0.56

-

0.69

0.75

1.08

-

pH(final)

Total organicmaterial (mg) FNH3/FH2O (mg)

a)

120 Hydrophilic tholin

Weight (mg)

100

Hydrophobic tholin

80

Total organic Material

60 40 20 0 5.8

4

7.8

9.8

12

b) 1.2

FHN3/FH2O

1 0.8 0.6 0.4 0.2 0 4

7.8

9.8

12

pH Fig. 1 In the table, values are expressed as the mean±SEM for five experiments performed at different pH values under the conditions described in the experimental procedures section. (a), (b) Reference values previously reported: (a) reference experiment at pH=5.8, Ruiz-Bermejo et al. 2007a; (b) experimental control with aerosols and pure water, Ruiz-Bermejo et al. 2007b. a. The graph indicates that: i) at pH~7, a maximum of non volatile water-soluble material (hydrophilic tholin) was produced; ii) high pH values led to an increase in the insoluble material collected (hydrophobic tholin); iii) the amount of organic material (hydrophilic tholin plus hydrophobic tholin) fixed as non-volatile species in each set of experiments suggests that at pH values between 7 and 8, it is possible to obtain the greatest amount of organic species. b. The graph indicates that high pH values favour the formation of polymeric species. The FNH3 fractions contain polymeric materials with molecular weights up to 80 kDa

with BSTFA to obtain the corresponding TMS derivatives. This derivatisation method is not specific for each kind of molecule indicated, but for comparative purposes, it provides an excellent general overview of the molecules present in the FNH3 and FH2O fractions. GCMS analysis revealed that no remarkable differences were found in the identities of organic molecules in the FNH3 and FH2O fractions obtained at the different pH values (Table 1). In addition, the composition of organics was very similar between FNH3 and FH2O for both

Prebiotic Synthesis of Protobiopolymers

337

Table 1 Identification of molecules of biochemical interest obtained in spark activation of CH4-N2-H2 atm with aqueous saline aerosols at different values of pH. Note that the molecules identified in the FNH3 fractions are practically the same molecules found in the FH2O fractions FNH3 fractions

FH2O fractions

Amino acids

Molecular formula

pH=4.0

pH=7.8

pH=9.8

pH=12

✓

✓

Glycine

C2H5NO2

✓

✓

Aminomalonic acid

C3H5NO4

✓

✓

Alanine

C3H7NO2

✓

✓

✓

β-Alanine

C3H7NO2

✓

✓

✓

Isoserine

C3H7NO3

✓

Serine

C3H7NO3

✓

3-amino-2hydroxypropanoic acid

C3H7NO3

pH=4.0

pH=7.8

pH=9.8

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓ ✓

✓

✓

✓

✓

✓

✓

C3H8N2O2 C4H9NO3

✓

Threonine

C4H9NO3

✓

✓

Aspartic acid

C4H7NO4

✓

✓

✓

✓

✓

✓

✓

Iminodiacetic acid

C4H7NO4

✓

✓

✓

✓

✓

✓

✓

2-Aminobutyric acid

C4H9NO2

✓

✓

✓

✓

C4H9NO2 C4H9NO2

✓

✓

2,3-diaminopropanoic acid

2-Aminoisobutyric acid

✓ ✓

Homoserine

3-Aminoisobutyric acid

pH=12

✓

✓

✓ ✓ ✓ ✓

✓ ✓ ✓

N-Methylalanine

C4H9NO2

N-ethylglycine

C4H9NO2

Glutamic acid

C5H9NO4

✓

✓

✓

✓

✓

✓

✓

✓

Glycolic acid

C2H4O3

✓

✓

✓

✓

✓

✓

✓✓

✓

Lactic acid

C3H6O3

✓

✓

✓

✓

✓

✓

✓

✓

2-Hydroxypropanoic acid

C3H6O3

✓

3-Hydroxypropanoic acid

C3H6O3

✓

✓

✓

✓

Glyceric acid

C3H6O4

✓

✓

✓

✓

✓

✓

Malic acid

C4H6O5

✓

✓

✓

✓

✓

✓

✓

✓

Tartatic acid

C4H6O6

✓

✓

✓

✓

✓

✓

✓

✓

2-hydroxybutyric acid

C4H8O3

✓

✓

✓

✓

2-hydroxyisobutyric acid

C4H8O3

✓

✓

3-hydroxybutyric acid

C4H8O3

✓

✓

✓

2,3-Dihydroxybutanoic acid

C4H8O4

✓

✓

✓

✓

2,4-Dihydroxybutanoic acid

C4H8O4

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓ ✓

✓

HYDROXY ACIDS

2-Hydroxyglutaric acid

C5H8O5

2,4-Dihydroxypentanodioic acid

C5H8O6

✓

✓

✓

✓

✓ ✓

✓

Carboxilic acids Oxalic acid

C2H2O4

Fumaric acid

C4H4O4

Succinic acid

C4H6O4

2-Methylsuccinic acid

C5H8O4

Tricarballylic acid

C6H8O6

Isocitric acid

C6H8O7

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓ ✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

338 2-Methylglutaric acid

M. Ruiz-Bermejo et al.

C6H10O4

✓

✓

HETEROCYCLES Parabanic acid

C3H2N2O3

✓

✓

✓

✓

✓

✓

✓

2,4,6-trihydroxy-1, 3,5-triazine

C3H3N3O3

✓

✓

✓

✓

✓

✓

✓

✓

5-hydroxy-hydantoin

C3H4N2O3

✓

✓

✓

✓

✓

✓

✓

✓

Imidazole

C3H6O3

✓

✓

✓

✓

✓

2,6-dihydroxy pyrimidine

C4H4N2O2

Uracil

C4H4H2O2

2,4,5-trihydroxy pyrimidine

C4H4N2O3

2,3,5-trihydroxypyrazine

C4H4N2O3

Alloxanoic acid

C4H4N2O4

Cytosine

C4H5N3O

Orotic acid

C5H4N2O4

✓ ✓ ✓ ✓

✓

✓ ✓

✓ ✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

Other organics Urea

CH4N2O

Dihydro-pyrimidine2,4-dione

C4H6N2O2

✓

✓

✓

✓ ✓

different and for the same pH values. Figure 2 shows a representative chromatogram of these GC-MS analyses.

Structural Characterisation of the Tholins by IR Spectroscopy The IR spectra of the FNH3 and FH2O fractions obtained under the different pH values were recorded to elucidate any possible structural differences between them (Fig. 3). No significant differences were found for the FNH3 fractions (Fig. 3a) obtained under all pH conditions. The same results were obtained for the FH2O fractions (Fig. 3b). All FH2O fractions obtained from these experiments had similar IR spectra. We conclude that no notable spectroscopic differences exist among the hydrophilic tholins obtained under conditions of different pH. The IR spectrum of the hydrophilic tholin from the control experiment was very similar to those shown here. Therefore, the primary macromolecular structures of the water-soluble materials obtained under the conditions assayed seemed to contain the same functional groups (−OH, −NH−, −CO, −COOH, R-C≡N, −CH3 and −CH2). In relation to the IR spectra of the hydrophobic tholins (Fig. 3c), again, no remarkable differences were detected in the different experiments. Moreover, the IR spectra of the hydrophobic tholins were very similar to the hydrophobic tholin obtained from the control experiment. For a complete assignment of the IR bands, see Ruiz-Bermejo et al. (2009, 2008 and 2007a), who reported an exhaustive structural study of the tholins obtained from aqueous aerosols.

Detection of High Molecular Weight Polymeric Species in Hydrophilic Tholins by 2D-Electrophoresis In a previous work (Ruiz-Bermejo et al. 2008), we demonstrated that SDS-PAGE could be used for the determination of molecular weights of hydrophilic tholins. Therefore, we conducted an analysis of the FNH3 and FH2O fractions using electrophoresis techniques.

Previously, both fractions from each experiment were centrifuged in an YM-3 Microcon centrifugal filter device (Millipore Corp) to retain and concentrate molecules above 3 kDa.

Prebiotic Synthesis of Protobiopolymers

339

5 3 1 7 14 45 11

16 19 15 6 8 10 13 9 12

18 17

10

34 36 26

20 2

32

22

11

23 24 12

37

27 29 25

21

30 28

35 33

39

31

20

38

30

Time (min) Fig. 2 GC-MS chromatogram showing the organic molecule profile from a FNH3 fraction (polymeric fraction) obtained in a CH4:N2:H2 atm with saline aqueous aerosols under alkaline ocean conditions (pH=12). 1.

Lactic acid; 2.2-hydroxybutyric acid; 3. glycolic acid; 4. alanine; 5. glycine; 6. 2-hydroxyisobutyric acid; 7. oxalic acid; 8. 3-hydroxypropanoic acid; 9. 3-hydroxybutyric acid; 10. 2-aminobutyric acid; 11. urea; 12. β-alanine; 13. N-ethylglycine; 14. succinic acid; 15. 2-methylsuccinic acid; 16. glyceric acid; 17. fumaric acid; 18. 2,3-dihydroxybutanoic acid; 19. serine; 20. 3-amino-2-hydroxypropanoic acid; 21. threonine; 22.dihydro-pyrimidine-2,4-dione; 23. 2-methylglutaric acid; 24. 2,4-dihydroxybutanoic acid; 25. parabanic acid; 26. malic acid; 27. 5-hydroxyhydantoin; 28. aspartic acid; 29. iminodiacetic acid; 30. 2-hydroxyglutaric acid; 31. 2,4,5-trihydroxypyrimidine; 32. 2,4,6-trihydroxy-1,3,5-triazine; 33. glutamic acid; 34. tartaric acid; 35. alloxanoic acid. 36. 2,4-dihydroxypentanodioic acid; 37. tricarballylic acid; 38. orotic acid; 39. isocitric acid

Tricine-SDS electrophoresis fractionation of the concentrated samples demonstrated the presence of silver-stained polymeric species only in the FNH3 fractions (data not shown).

Although we do not know the exact nature of the skeleton of the FNH3 fractions, in these fractions, we detected polar groups such as carboxylic acids, nitriles and amines by GC-MS and IR spectroscopy. Based on those results, we conducted an analysis of the polymeric FNH3 fractions by 2-D electrophoresis to fractionate the net charge polymeric species and determine their pIs and apparent molecular weights (Fig. 4). At this basic level, these FNH3 fractions can develop a net electric charge in water by the same acid–base reactions that proteins undergo and should also migrate in an electric field under similar conditions. Prior to isoelectrofocusing, the same amount (250 μg of dried weight) of each concentrated sample was dissolved in a rehydration buffer that enhances solubility and minimises possible aggregation due to charge-charge and hydrophobic interactions. Next, the gradient immobiline dry strips were fractionated in 15% polyacrylamide gels and silver stained. As shown in Fig. 4, and in good agreement with the tricine-SDS electrophoresis analysis results, silver-stained polymeric species can be detected in FNH3 fractions from all experiments. In the samples with a higher initial pH, however, greater amounts and more types of silver-stained polymeric species were detected. In all experiments, independent of their initial pH, three anionic spots focused at pI 3.0 were detected with apparent molecular weights of 6.5, 7.5 and 8.0 kDa (a, b and c spots, respectively, Fig. 4), although there was much more 8.0 kDa-polymer in the most alkaline initial pH experiment (Fig. 4a). In addition, heavier anionic spots were detected at the same pI in alkaline

340

M. Ruiz-Bermejo et al.

Fig. 3 Transmission FTIR spectra obtained from the experiment performed at pH=4. a FNH3 fraction; b FH2O fraction; c hydrophobic tholin

a) 2063

1053 1410

1736

2855 3320

b)

3191

1614

2925 1666

Reflectance (a.u.)

2091 1088 1410 590 1679 1728 3056 3183

c)

2242

875 1418 3390

2874

1701

2932 2961 3500

2500

1500

500

Wavenumber(cm-1)

experiments with an initial pH of 9.8 and 12. Their approximate molecular weights were 80 kDa and 65 kDa in the pH 9.8 experiments (f and e spots, respectively, Fig. 4b) and 45 kDa in the pH 9.8 and pH 12 experiments (d spots, Fig. 4a and b). Interestingly, from the highest initial pH assayed (pH 12), other polymeric species with different pIs were silver-stained. Five 50-kDa spots were focused at pI 3.5, 4.7, 5.0, 5.5 and 5.7 (g, h, i, j and k spots in Fig. 4a), and in contrast to the other experiments, 4 cationic species were focused at pI 9.0 of 45, 38, 28 and 14 kDa (l, m, n and o spots, respectively, in Fig. 4a).

Discussion The most important finding of this work is that experimental conditions that resemble those of alkaline oceans, which have been proposed as plausible prebiotic scenarios, enhanced the

Prebiotic Synthesis of Protobiopolymers

341

Fig. 4 2-D fractionation of hydrophilic polymeric species from FNH3 fractions. Two hundred fifty micrograms of dried FNH3 fractions from experiments at an initial pH=12 (a), pH=9.8 (b), pH=7.8 (c), pH= 4 (d), were separated by isoelectric focusing in a 3–10 pH gradient. After IEF, the samples were further separated in 15% SDS-polyacrylamide gels and visualised by silver staining. Polymeric spots are highlighted with a white open circle and named

production and diversity of polymeric species with high molecular weights. Moreover, we used 2-D electrophoresis to characterize the “intractable polymers” or tholins for the first time.

The traditional approach for the determination of the molecular weights of the tholins has been gel filtration chromatography. Using this technique, McDonald et al. (1991) reported that analogs of Jupiter’s tholins contain fractions with weights ranging between 200 and 700 Da. Triton’s and Titan’s tholins (CH4/N2 atm) presented a molecular weight distribution between 200 and 600 Da (McDonald et al. 1994). Takano et al. (2004) obtained organic compounds with a molecular weight distribution ranging between several hundred and ~3 kDa from a gas mixture of CO/NH3/H2O and proton irradiation. In all of these reported examples, the data were recovered using UV detectors in a range of 195–220 nm, and therefore, the information obtained corresponded only to the fraction containing UV-detectable material; this information is also limited by the type of chromatographic column used. On the other hand, Sarker et al.

(2003) used high-resolution Fourier-transform ion cyclotron resonance mass spectrometry (FTIC/MS) to show simple and regular patterns in the molecular weight distribution of the organics from Titan’s tholin, with a range of weights between 120 and 800 Da. The common mass spectrometric techniques, such as MALDI and ESI, are unsuccessful for the determination of molecular weights of bulk tholins, likely due to their heterogeneous nature (Ruiz-Bermejo et al. 2008). However, classical and widely used biochemical tools such as

342

M. Ruiz-Bermejo et al.

electrophoresis are poorly suited for the determination of molecular weights of tholins and other similar chemical species (Draganic et al. 1980; Ferris et al. 1981; Ruiz-Bermejo et al.

2008). The analysis of the FNH3 fractions by 2-D electrophoresis revealed a well-defined distribution of polymeric species in relation to the initial pH of the experiments. Thus, the alkaline pH values (pH 9.8 and 12) led to the formation of heavier anionic polymeric species, up to an apparent molecular weight of 80 kDa, as compared to those obtained at a lower initial pH (pH 4 and 7.8). Strikingly, only an initial pH of 12 promoted the formation of cationic major species that focused around pI 9.0 and, in lower amounts, 45-kDa polymeric species that focused at a pI range between 3.5 and 5.7. It is interesting to point out that the polymeric species present in the FNH3 fractions were mainly formed by the same discrete organic molecules present in the lighter material of the FH2O fractions and that the molecular weights of the polymeric species increased remarkably with the initial value of the pH. In addition, a pH increase led to the formation of a greater amount of water-soluble polymeric materials.

Further analyses are needed to determine the mechanism underlying the polymerisation of cationic species only at a very alkaline pH (pH 12).

In a previous work (Ruiz-Bermejo et al. 2008), we suggested that the nature and ratio of the discrete molecules and the polymeric species detected in hydrophilic tholins formed from aqueous aerosols, indicate different mechanisms of formation that compete with each other. In this context, we proposed a set of three reaction scenarios: one using bulk water, one at the water-gas interface and one in the gas phase. For the present case, reactions in the gas phase are not interesting because these reactions lead to the formation of hydrophobic tholins, and the nature of these tholins does not seem to depend on salinity, the initial pH values or on the presence of water (Ruiz-Bermejo et al. 2009). In contrast, the increase in water-gas interfaces promotes the formation of a considerable diversity and amount of carboxylic acids at any pH value, in good agreement with the results reported by other authors (Donaldson et al. 2004; Ellison et al. 1999; Tuck 2002) and with our previous data (Ruiz-Bermejo et al. 2007a; b). Therefore, independent of the initial pH, the main condition for a substantial formation of carboxylic acids under plausible prebiotic conditions is the presence of aqueous aerosols.

In general, the presence of aqueous aerosols allows for the formation of a great diversity of polar organic molecules and, therefore, the production of hydrophilic tholins. Because the final pH was practically the same in all experiments carried out in this work and hydrophilic tholins were formed at a high ratio due to the presence of aqueous aerosols, it is straightforward to suppose that the formation of the soluble polymeric species was favoured by the increase of the water-air interface in environments with high initial pH values and that the processes in the bulk solution were secondary. These results support the hypothesis proposed by Donaldson and Vaida (2006), which suggests that atmospheric aerosols could be the most favourable available chemical space where organics were concentrated, selected and transformed in early biochemical reactions contributing to non-enzymatic biopolymer formation.

In this work, we demonstrated the formation, in one pot, of polymeric species of high molecular weight formed by building blocks present in modern biological systems (amino acids, hydroxy- and carboxylic acids and several heterocycles containing nitrogen). This first identification of polymeric species of very high molecular weights formed under alkaline ocean conditions opens up a new pathway in the development of Prebiotic Chemistry. We propose the name â&#x20AC;&#x153;protobiopolymersâ&#x20AC;? for these polymeric species consisting of biomolecules and formed concomitantly under prebiotic conditions. It is a fact that polymerisation increases complexity, which is a pre-requisite for the emergence of life. The size and three-dimensional conformation of the biopolymers shape their potential intermolecular interactions. Analyses of the actual monomer composition of these

Prebiotic Synthesis of Protobiopolymers

343

polymeric species as well as their chemical bonds and structures are in progress. Further analyses could be carried out to determine potential catalytic and informational properties and to generate a hypothesis about the roles of these compounds in the first prebiotic systems. Although the hypothesis of self-organizing protometabolic cycles is a controversial issue (Orgel 2000), our experiments show that aerosol chemistry favours the production of potential actors in such a hypothesis. Aqueous aerosols formed from analogues of alkaline oceans lead not only to the production of the raw materials necessary for a primordial variant of the Krebs cycle (Meléndez-Hevia et al. 1996), such as succinic acid, malic acid, isocitric acid and tricarballylic acid (Table 1), but also to the production of a chemically diverse group of protobiopolymers that could potentially interact with and transform those biomolecules. These materials could establish a cycle of reactions as a precursor to intermediate metabolism. This cycle would later lead to the increase of the molecular complexity and to an increase in structural organisation, which could lead ultimately to the emergence of life. In this way, the results of this work are in good agreement with the hypothesis that alkaline environments and alkaline oceans could favour the early biogenesis of life on Earth (Kempe and Kazmierczak 2002; Russell 2003).

In addition, it has recently been suggested that alkaline environments may have been much more common on the early Earth than considered previously and that these high pH environments could promote the abiotic and simultaneous formation of pentoses and purines (Holm et al. 2006; Holm and Neubeck 2009), the basic elements in the “RNA world” hypothesis (see e.g., Joyce and Orgel 1999). Therefore, it seems very interesting to study in depth the implications of alkaline environments as plausible scenarios for the development of the first living system.

Conclusions The identification of high molecular weight polymeric species by 2-D electrophoresis as components of hydrophilic tholins formed from highly alkaline aqueous aerosols, under a CH4 atm, and the notable production of carboxylic acids under this type of environment suggests that the water aerosols formed from alkaline oceans could have provided the optimal medium for the simultaneous formation of protobiopolymers and protometabolic systems on the early Earth.

Acknowledgements The authors have used the research facilities of the Centro de Astrobiología (CAB) and have been supported by the Instituto Nacional de Técnica Aeroespacial “Esteban Terradas” (INTA) and by the projects AYA2009-13920-C02-01 of the Ministerio de Ciencia e Innovación (Spain). We thank Dr. S. Veintemillas for his useful comments. We thank Profesor G. Salcedo and Dr. A. Díaz-Perales for the use of their research facilities in their laboratory in the Department of Biotechnology at the Escuela Técnica Superior de Ingenieros Agrónomos (UPM).

References Dobson CM, Ellison GB, Tuck AF, Vaida V (2000) Atmospheric aerosols as prebiotic chemical reactors.

Proc Nat Acad Sci USA 97:11864–11868 Donaldson DJ, Tervahattu H, Tuck AF, Vaida V (2004) Organic aerosols and the origin of life: a hypothesis.

Orig Life Evol Biosph 34:57–67 Donaldson DJ, Vaida V (2006) The influence of organic films at the air-aqueous boundary on atmospheric processes. Chem Rev 106:1445–1461

344

M. Ruiz-Bermejo et al.

Draganic ZD, Kinetic V, Jonanovic S, Draganic IG (1980) The radiolysis of aqueous ammonium cyanideCompounds of interests to chemical evolution studies. J Mol Evol 15:239–260 Ellison GB, Tuck AF, Vaida V (1999) Atmospheric processing of organic aerosols. J Geophys Res 104:11633–11641 Ferris JP, Edelson EH, Auyeung JM, Joshi PC (1981) Structural studies on HCN oligomers. J Mol Evol 17:69–77 Görg A, Obermaier C, Boguth G, Harder A, Scheibe B, Wildgruber R, Weiss W (2000) The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 21:1037–1053 Grotzinger JP, Kasting JF (1993) New constrains on Precambrian ocean composition. J Geol 101:235–243 Holm NG, Dumont M, Ivarsson M, Konn C (2006) Alkaline fluid circulation in ultramafic rocks and formation of nucleotides constituents: a hypothesis. Geochem Trans 7:7 Holm NG, Neubeck A (2009) Reduction of nitrogen compounds in oceanic basement and its implications for HCN formation and abiotic organic synthesis. Geochem Trans 10:9 Joyce GF, Orgel LE (1999) In: Gesteland RF, Cech TR, Atkins JF (eds) The RNA World. Cold Spring Harbor Laboratory Press, New York, pp 49–77 Khare BN, Sagan C, Ogino H, Nagy B, Er C, Schram KH, Arakawa ET (1986) Amino acids derived from Titan tholins. Icarus 68:176–184 Knauth LP (2005) Temperature and salinity history of the Precambrian ocean: implications for the course of microbial evolution. Palageo Palaeoclimat Palaeoecol 219:53–59 Kempe S, Kazmierczak J (1994) The role of alkalinity in the evolution of ocean chemistry, organization of living systems and biocalcification processes. Bull’InstiOcéanogr Monaco special 13:61–117 Kempe S, Kazmierczak J (2002) Biogenesis and early life on Earth and Europa: favored by an alkaline ocean? Astrobiology 2:123–130 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4.

Nature 227:680–685 Lerman L, Teng J (2004). In the beginning. In: Seckbach J (ed) Origins: Genesis and Diversity of Life.

Springer, pp 35–55 Macleod G, Mckeown C, Hall AJ, Russell MJ (1994) Hydrotermal and oceanic pH conditions of possible relevance to the origin of life. Orig Life Evol Biosph 24:19–41 McDonald GD, Khare BN, Thompson WR, Sagan C (1991) CH4/NH3/H2O Spark tholin: chemical analysis and interaction with jovian aqueous clouds. Icarus 94:354–367 McDonald GD, Thompson WR, Heinrich M, Khare BN, Sagan C (1994) Chemical investigation of Titan and Triton tholins. Icarus 108:137–145 Meléndez-Hevia E, Wadell TG, Cascante M (1996) The puzzle of the Krebs citric acid cycle: assembling the pieces of chemically feasible reactions and opportunism in the design of metabolic pathways during evolution. J Mol Evol 43:293–303 Morse JW, Mackenzie FT (1998) Hadean ocean carbonate geochemistry. Aquat Geochem 4:301–319 Orgel LE (2000) Self-organizing biochemical cycles. Proc Nat Acad Sci USA 97:12503–12507 Ruiz-Bermejo M, Menor-Salván C, Osuna-Esteban S, Veintemillas-Verdaguer S (2007a) Prebiotic microreactors: a synthesis of purines and dihydroxy compounds in aqueous aerosol. Orig Life Evol Biosph 37:123–142 Ruiz-Bermejo M, Menor-Salván C, Osuna-Esteban S, Veintemillas-Verdaguer S (2007b) The effects of ferrous and other ions on the abiotic formation of biomolecules using aqueous aerosols and spark discharges. Orig Life Evol Biosph 37:507–521 Ruiz-Bermejo M, Menor-Salván C, Mateo-Martí E, Osuna-Esteban S, Martín-Gago JA, VeintemillasVerdaguer S (2008) CH4/N2/H2 spark hydrophilic tholins: a systematic approach to the characterization of tholins. Icarus 198:232–241 Ruiz-Bermejo M, Menor-Salván C, de la Fuente JL, Mateo-Martí E, Osuna-Esteban S, Martín-Gago JA, Veintemillas-Verdaguer S (2009) CH4/N2/H2 Spark hydrophobic tholins: a systematic approach to the characterization of tholins. Part II. Icarus 204:672–680 Russell MJ, Hall AJ (1997) The emergence of life from iron monosulphide bubbles at a submarine hydrothermal redox and pH front. J Geol Soc Lond 154:377–402 Russell MJ (2003) The importance of being alkaline. Science 302:580–581 Sagan C, Khare BN, Thompson GD McDonald, Wing MR, Bada JL, Vo-Dinh T, Arakawa ET (1993) Polyciclic aromatic hydrocarbons in the atmosphere of Titan and Jupiter. Astrophys J 414:399–405 Sagan C, Khare BN (1979) Tholins: organic chemistry of interstellar grains and gas. Nature 277:102–107 Sarker N, Somogyi A, Lunine JI, Smith MA (2003) Titan aerosol analogues: analysis of the non-volatile tholins. Astrobiology 3:719–726 Schlesinger G, Miller SL (1983) Prebiotic synthesis in atmospheres containing CH4, CO and CO2. J Mol Evol 19:376–382

Prebiotic Synthesis of Protobiopolymers

345

Shah DO (1970) The origin of membranes and related surface phenomena. In: Ponnamperuma C (ed) Exobiology. North Holland, pp 235–265 Takano Y, Ohashi A, Kaneko T, Kobayashi K (2004) Abiotic synthesis of high-molecular-weight organics from an inorganic gas mixture of carbon monoxide, ammonia, and water by 3 MeV proton irradiation.

Appl Phys Lett 84:1410–1412 Tervahattu H, Tuck A, Vaida V (2004). Chemistry in prebiotic aerorsols: A mechanism for the origin of life.

In: Seckbach J (ed) Origins: Genesis, Evolution and Diversity of Life. Springer, pp 153–165 Tuck A (2002) The role of atmospheric aerosols in the origin of life. Surv Geophys 23:379–409 Walker JCG (1985) Carbon dioxide on the early Earth. Orig Life Evol Biosph 16:117–127

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

25

New Insights into the Characterization of Insoluble Black HCN Polymers

by Marta Ruiz-Bermejo* a ), Jose´ L. de la Fuente b ), Celia Rogero a ), Ce´sar Menor-Salva´n a ), Susana Osuna-Esteban a ), and Jose´ A. Mart n-Gago a ) c ) a

) Centro de Astrobiolog a (Consejo Superior de Investigaciones Cient ficas – Instituto Nacional de Te´cnica Aeroespacial (CSIC-INTA)), Carretera Torrejo´n-Ajalvir, Km 4, E-28850 Torrejo´n de Ardoz, Madrid (phone: þ 34-91-520-6402/6458; fax: þ 34-91-520-6410; e-mail: ruizbm@inta.es) b ) Instituto Nacional de Te´cnica Aeroespacial (INTA), Carretera Torrejo´n-Ajalvir, Km 4, E-28850 Torrejo´n de Ardoz, Madrid c ) Instituto de Ciencia de Materiales de Madrid (CSIC), C/Sor Juana Ine´s de la Cruz, 3, E-28049 Cantoblanco, Madrid

The data presented here provide a novel contribution to the understanding of the structural features of HCN polymers and could be useful in further development of models for prebiotic chemistry. The interpretation of spectroscopic and analytical data, along with previous results reported by other authors, allowed us to propose a mechanism for the aqueous polymerization of HCN from its primary and simplest isolated oligomer, the diaminomaleonitrile (DAMN) tetramer. We suggest that insoluble black HCN polymers are formed by an unsaturated complex matrix, which retains a significant amount of H2O and important bioorganic compounds or their precursors. This polymeric matrix can be formed by various motifs of imidazoles and cyclic amides, among others. The robust formation of HCN polymers assayed under several conditions seems to explain the plausible ubiquity of these complex substances in space.

Introduction. – HCN is ubiquitous in the universe and is a significant product in prebiotic simulation experiments [1 – 4]. HCN Polymers may be the major components of dark matter, which could be present in objects such as asteroids, moons, planets, and, in particular, comets [1] [5 – 7]. It has been proposed that the reddish haze (tholins) present in the atmosphere of Titan, the largest moon of Saturn, could be due to the presence of HCN polymers [8]. In addition, it has been suggested that HCN polymers may be important substances in the first stages of the chemical evolution of life [9]. This hypothesis is based on the fact that HCN polymers are precursors of important bioorganic compounds such as purines, pyrimidines, and amino acids, as well as other biological compounds such as oxalic acid and guanidine [10 – 13]. HCN can spontaneously polymerize in the presence of bases such as NH3 and free radicals from ionizing radiation, and occurs over a wide range of temperatures and pressures in both polar (water) and non-polar (hydrocarbon) solvents and surfaces [1]. The HCN polymers, also known as HCN oligomers, azulmic acid or azulmin , are heterogeneous solids ranging in color from yellow or orange to brown or black, depending on the degree of polymerization and/or cross-linking processes. The structures of HCN polymers have not been fully characterized and remain controversial due to their complex and heterogeneous nature.

2012 Verlag Helvetica Chimica Acta AG, Z rich

26

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

Several models attempting to explain the complex structure of HCN polymers have been proposed [14 â&#x20AC;&#x201C; 19]. Scheme 1 compiles these models.

Scheme 1. Different Models Proposed in the Literature for HCN Polymers

During the polymerization of HCN in aqueous environments, a H2O-soluble and an insoluble solid product are formed. Our study is focused on the insoluble solid, commonly named insoluble black HCN polymers or black azulmic acids . These polymers were prepared from solutions of equimolar amounts of NH4Cl and NaCN in pure H2O at concentrations of 1 and 10m using different reaction times. To further

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

27

elucidate the structure of this polymer, we used the methodology we developed previously for the characterization of tholins [20] [21]. Data from the elemental analysis, FT-IR, 13C-CP-MAS-NMR spectra, and GC/MS analysis of the insoluble black HCN polymers were compared to results previously reported in the literature.

Additionally, we used X-ray photoelectron spectroscopy (XPS) to quantify various chemical species in powder samples [20 â&#x20AC;&#x201C; 22]. Additionally, to obtain complementary information, the insoluble black HCN polymers were acid hydrolyzed. The hydrolyzed supernatants and black insoluble residues were also analyzed. Scheme 2 outlines the general treatment of the samples, the separation in fractions and the techniques used for this study.

Scheme 2. Reaction Conditions, General Treatment of Samples, Fractionation, and Techniques Used

28

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

The results presented herein complete the data reported in a previous paper [20] where some spectroscopic data were shown. The strength of HCN polymerization reactions in an aqueous medium and the proposed structures for insoluble black HCN polymers are discussed.

Results. – The results of the elemental analyses of the insoluble black HCN polymers are compiled in Table 1. Our results are comparable to those reported by Eastman et al. [23] and Labadie et al. [24] for short reaction times (1 – 10 d). For longer reaction times (30 d), black insoluble solids richer in oxygen were obtained. However, a longer reaction time did not have a significant influence on the yield of these insoluble products, 29 4%. In this work, the yields were calculated from the initial amount of C in the NaCN used as a reactant. The estimated empirical formulae, C3H4N3O and C5H6N4O2 , indicate the formation of highly unsaturated structures with a backbone based on C and N. The elemental analyses of the insoluble black residues after acid hydrolysis provided the following results: C, 36.6 0.6%; H, 4.1 0.2%; N, 33.2 0.8%; and O, 22.3 0.9%. Thus, the estimated empirical formula for the hydrolyzed residues was C2H3N2O. Table 1. Experimental Conditions, Yields, and Elemental Analyses for the Series of Insoluble Black HCN Polymers Prepared in This Work and by Other Authors. The yields in this work were calculated from the initial amount of carbon in the initial NaCN. N.R., not reported; c, initial concentration; T, reaction temp.; t, reaction time.

Reactants

c [m]

T [8]

t

Yield [%] C H N O Empirical Ref. [%] [%] [%] [%] formula

KCN (aq) þ NH4Cl(aq) 1 – 1.5 70 HCN(l) þ NH3(aq) þ Et3N N.R. r.t.

N.R. N.R.

36.5 35

4.2 39.2 N.R. 38

15.7 12

C3H4N3O [23] – [25]

HCN(g) þ NH3(aq) HCN(aq) þ NH4OH HCN(l) þ NH3(l) HCN(aq) þ NH3(aq) NaCN(aq) þ NH4Cl(aq) NaCN(aq) þ NH4Cl(aq) NaCN(aq) þ NH4Cl(aq) NaCN(aq) þ NH4Cl(aq)

N.R. 31 N.R. 51 31 25 27 33

39.6 38.77 44.5 40.2 36.2 36.5 36.5 36.2

4.0 3.96 4.0 3.8 4.3 4.4 4.4 3.9

10.4 16.42 N.R. N.R. 17.7 17.4 17.2 19.2

C5H6N5O C3H4N3O – – C3H4N3O C3H4N3O C3H4N3O C5H6N4O2

12.5 1 N.R. N.R. 10 1 1 1

80 h 3 d, after water 30 d r.t.

4d 90 16 h r.t.

30 d 40 – 50 5 h 38 3d 38 3d 38 10 d 38 30 d

46.1 40.85 52.5 41.8 40.8 40.8 41.9 39.6

[16] [24] [18] [19] This work This work This work This work

Determination of Functional Groups. The FT-IR spectra of the insoluble black HCN polymers synthesized in our laboratory were similar, with no significant differences among them (Fig. 1, c). These spectra highly resemble those reported in the literature for HCN polymers prepared under different experimental conditions.

Quirico et al. [8] reported an IR spectrum of an HCN polymer prepared from pure NH3 and liquid HCN that was similar to the IR spectrum in this study. Similar IR data were reported by Liebman et al. [26] and Umemoto et al. [16] for HCN polymers obtained from liquid HCN and Et3N in MeCN, and from gaseous HCN and aqueous NH3 , respectively. The IR spectra do not provide sufficient information to identify the

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

29

Fig. 1. Transmission FT-IR spectra of a) hydrolyzed supernatant, b) hydrolyzed insoluble black residue, and c) insoluble black HCN polymers

differences between HCN polymers obtained in H2O solutions and those prepared in aprotic mediums or without solvents. Therefore, the IR spectra must be deconvoluted to identify the functional groups in the HCN polymers. The features in the IR spectra of our insoluble black HCN polymers can be assigned to N-containing groups: primary and secondary amines (3444 cm 1, NH stretch; 3330 cm 1, NH2 antisym. stretch, and 3191 cm 1 NH2 sym. stretch), CN and carbodiimide groups (2187 cm 1, C N and

30

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

=

=

=

N¼C¼N ), amides, urea and triazines (1655 cm 1, C¼O stretch (amide I band); 1560 cm 1, NH deformation (amide II band) plus ring stretch in triazine compounds; 1395 cm 1, C N stretch (amide III band)), and azoxy groups (1490 cm 1, N¼N O antisym. stretch).

All of the insoluble black hydrolyzed residues presented similar bands in their IR spectra (Fig. 1, b). These IR bands were assigned to the following functional groups: 3084 cm 1 (very broad), primary and secondary amines (OH groups of carboxylic acids, and ¼CH in aromatic and unsaturated hydrocarbons may overlap); 1725 and 1665 cm 1, C¼O stretch in carbonyl and amide compounds; 1590 cm 1, NH2 primary alkyl amide; 1515 cm 1, NH in secondary amides and triazine compounds; and 1420 cm 1 OH in carboxylic acids and C N in primary amides.

The IR spectra of the soluble hydrolyzed materials exhibit bands that may correspond to the NH þ4 cation (3050 and 1410 cm 1), CH3 and CH2 groups in aliphatic compounds (2895 cm 1), and C¼O in carbonyl, carboxylic, and urea compounds (1735 and 1670 cm 1).

As in the FT-IR analysis, the 13C-CP-MAS-NMR spectra of all of our samples were similar and resembled spectra obtained by Garbow et al. [25] for the water-insoluble fraction as well as spectra reported by Mamajanov and Herzfeld [14] [15] for an HCN polymer prepared from gas HC15N and Et3N. The deconvoluted solid-state 13C-NMR spectrum of our insoluble black HCN polymers displayed the following resonances: i) resonance at 168 ppm, which may correspond to amide groups ( CONH2 ); ii) resonance at 159 ppm, which may correspond to imine and/or heterocyclic groups ( C¼N ); iii) a group of resonances at 154, 149, and 139 ppm, which may correspond to C-atoms of heterocyclic compounds containing N; iv) a group of low-intensity resonances between 130 and 100 ppm, which may correspond to alkenes ( C¼C ) and nitriles ( C N); and v) a third group of unresolved resonances between 100 and 60 ppm, which may correspond to C-atoms bound to a heteroatom, such as C (N) of amines (Fig. 2, b). The 13C-CP-MAS-NMR spectra of the insoluble hydrolyzed residues (Fig. 2, a) exhibit resonances similar to those of the insoluble black HCN polymers , but with a well-defined signal at 171 ppm, which may correspond to carboxylic acids, as was observed in IR analysis.

XP Spectra were recorded to obtain further structural data for the HCN polymers.

XP Spectra of the prepared samples recorded from 1200 to 0 eV provided an overview of the main elements in the insoluble black HCN polymers , as well as any trace contaminants (data not shown). As expected, the HCN polymers consisted mainly of C and N with a lower contribution of O. It is important to note that C and O were the main contaminants in samples prepared in air and were also the main contaminants of the KBr pellets prepared. On the other hand, the C bonds were relatively well-defined by 13C-CP-MAS-NMR and IR spectra. Therefore, only the high-resolution N 1s core level spectra were relevant for elucidation of the structures of the controversial Ncontaining functional groups in the insoluble black HCN polymers , which can be considered the distinctive fingerprint of the possible structures of the insoluble black HCN polymers .

Fig. 3 and Table 2 present the binding energies of the different components of the N 1s core level peak, as well as their assignment and quantitative results. The two main components are centered at 397.6 and 398.7 eV, and can be associated with the N¼C =

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

13

C-CP-MAS-NMR Spectra of a) hydrolyzed insoluble black residue and b) insoluble black HCN polymers

=

Fig. 2.

31

=

bonds (imine and/or heterocyclic groups) and the CO N group (amides), respectively. The small contribution detected at higher binding energies, 400.5 eV, corresponds to the NH groups in pyrrolic structures, as well as CN and amine groups.

This assignment of the binding energies for the N 1s core level peak was checked against standard samples prepared in our laboratory [21]. The functional groups identified by XPS were in agreement with the IR and 13C-NMR assignments . The advantage of the XPS technique is that it provides quantitative information about each N-containing group in the insoluble black HCN polymers . Additionally, the XP spectra of the insoluble hydrolyzed residues were recorded (Table 2). For the insoluble hydrolyzed residues, the absence of a component signal at 400 eV, the increase in

32

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

signals for amide groups in the XP spectra, and the presence of resonances assigned to carboxy C-atoms in the 13C-NMR spectra indicated that the component with the signal at 400.5 eV in the insoluble black HCN polymers can be assigned to nitrile groups.

Fig. 3. High-resolution N 1s core level spectra of insoluble black HCN polymers . The dots correspond to experimental data, the solid line corresponds to the fit, and the filled gray-scale curves correspond to the curve components used for deconvolution of the spectra.

Table 2. X-Ray Photoelectron Spectroscopy ( XPS) Data of the N 1s Core Level, Indicating the Nitrogen Chemical Environments in Black HCN Polymers and in the Hydrolyzed Insoluble Residue Group

BE [eV]

N¼ (imines)/ N¼C (heterocycles) CO N (amides) NH (amines)/ C N (nitriles)/pyrrolic NH =

=

397 – 398 399 400

BE (% of the component) Insoluble black HCN polymers

Hydrolyzed insoluble residues

397.6 (47.4) 398.7 (47.4) 400.5 (5.1)

398.0 (51.9) 399.4 (48.1)

Additionally, the UV/VIS spectra were recorded, and one sample is shown in Fig. 4.

The UV/VIS spectra of the all samples showed two shoulders at ca. 345 and ca. 465 nm that can be related to N-heterocyclic macromolecular systems with p-extended conjugation [27] [28].

Identification of Small Molecules Absorbed in the Insoluble Black HCN Polymers .

The ability of the insoluble black HCN polymers to retain H2O and other lowmolecular weight molecules was confirmed by preliminary thermogravimetric (TG) and GC/MS analyses.

The TG measurements (Fig. 5) indicated that the amount of adsorbed H2O was 8 – 10% depending on the reaction times. Long reaction times led to a greater amount of adsorbed H2O in the matrix of the insoluble black HCN polymers , which was consistent with data from the elemental analysis. A complete thermal study of these

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

33

Fig. 4. UV/VIS Spectrum of HCN polymers in DMSO

Fig. 5. Thermogravimetry (TG) Curve for HCN polymers. Heating rate was 108/min under N2 .

samples was described in an other publication, dealing with the structure property relationships in this complex polymeric system [29].

About 30 â&#x20AC;&#x201C; 40% of the insoluble black HCN polymers was released as H2O-soluble organic material after acid hydrolysis. This is consistent with the results reported by Labadie et al. [24]. The dried residues of the hydrolyzed supernatants were analyzed for polar organic compounds by GC/MS (Scheme 2). The chromatograms of all the samples showed similar profiles (Fig. 6) independent of the reaction times. In general,

Fig. 6. GC/MS Chromatogram showing the profile of organic molecules absorbed by the matrix of the insoluble black HCN polymers . These compounds are released after acid hydrolysis and were detected in the hydrolyzed dried supernatants independently of the initial reaction conditions. The GC/MS analyses of the insoluble black HCN polymers before hydrolysis and of the hydrolyzed insoluble residues were negative; no small polar molecules were detected.

34 CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

35

all samples contained large amounts of oxalic acid, urea, and glycine, as expected [13] [24]. These three compounds comprised ca. 10 – 15% of the soluble hydrolyzed material. The purines, such as adenine, hypoxanthine, and guanine, and the pyrimidines, such as uracil, 5-hydroxyuracil, 5-aminouracil, and orotic acid, were previously identified by Miyakawa et al. [10] in a low-temperature experiment at the eutectic point of HCN. Negron-Mendoza et al. [30] quantified the dicarboxylic, malonic, succinic, and maleic acids in experiments involving the irradiation of NH4CN. However, to the best of our knowledge, these acids were not previously reported in H2O polymerization experiments with HCN. Furthermore, for the first time, the organic compounds pyrazine-2,5-diol, hydroxymalonic acid, aminomalonic acid, 2,4,5trihydroxypyrimidine, and 2,4,7-trihydroxypteridine were detected in HCN polymerization experiments. Further analyses are in progress to identify organic compounds in the HCN polymers, because the detection of these molecules seems to depend on sample preparation and analytical tools rather than experimental conditions of polymerization [31].

In summary, we can conclude that insoluble black HCN polymers are formed by a highly unsaturated C N matrix, which can absorb H2O and low-molecular-weight polar molecules. This matrix likely consists of macrostructures with a high content of Ncontaining heterocycles and non-hydrolyzable amides. Temperature and reaction times did not seem to affect the synthesis of insoluble black HCN polymers in an aqueous medium. Therefore, the formation of insoluble black HCN polymers is robust and is expected to occur under diverse environmental conditions.

Discussion. – Proposal of Structures for Insoluble Black HCN Polymers . The initial steps in the oligomerization of HCN are well-understood [9]. The rate-determining step is the nucleophilic attack of the CN ion at the C N bond of HCN, which leads to the formation of imino-acetonitrile. This step is followed by the stepwise condensation of HCN to form aminomalonitrile (AMN) and diaminomaleonitrile (DAMN). DAMN is readily formed at room temperature in 0.1 – 1.0m aq. HCN solutions [32]. The lowestorder oligomer is isolated from aqueous solutions. The mechanism of formation of higher order HCN oligomers is much less clear. In this study, we propose a possible formation mechanism and structures for the insoluble black HCN polymers starting with DAMN (Scheme 3; new structures are framed).

DAMN is a weakly basic amine, and its chemistry has been explored in detail [33].

This reagent has been used extensively in the preparation of heterocyclic molecules, such as dicyanoimidazoles, dicyanopyrazines, and purines. Furthermore, Johnson et al. have used DAMN in the preparation of different polymeric systems [34] [35].

DAMN is an A2B2 monomer. The linear polyamine shown in Scheme 3, particularized as a 12-mer, could be obtained by an addition reaction between the amine groups, the nucleophilic agent, and the CN groups of the other molecule of this reactive monomer, which acted as the electrophilic agent. Subsequent elimination reactions, such as decyanation (Pathway a) and/or deamination (Pathway b), for this polyamine may allow formation of new extended-conjugation macrostructures. The formation of stable five- or six-membered rings is possible from the decyanation of this linear polyamine. Pathway a-1 is consistent with the recent work of Mamajanov and Herzfeld [14], who have studied the solid-state reaction of crystalline DAMN (Scheme 1). The

Scheme 3. Mechanistic Proposal for the Formation of Insoluble Black HCN Polymers from DAMN. The framed structures are indicated for the first time.

36 CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

CHEMISTRY & BIODIVERSITY â&#x20AC;&#x201C; Vol. 9 (2012)

37

linear polyamide (Pathway a-2) displayed a high tendency to form cyclic components, because the polymerization reactions in solution favor intramolecular cyclizations. For the first time, the resulting cyclic polyamide was considered as a motif of the insoluble black HCN polymers . Taking into account the quantitative results of XPS, this cyclic polyamide system appeared to be prevalent in the insoluble black HCN polymers .

On the other hand, the formation of five- and six-membered rings would also be expected by a deamination reaction (Pathway b) from the polyamine and intramolecular cyclization between CN groups, which may yield the heterocyclic polymer indicated in Scheme 3. This is, to the best of our knowledge, the first time that this novel macrostructure based on pyrrole and dihydropyrazine rings has been proposed for HCN polymers.

These heterocyclic macrostructures indicate the presence of a non-hydrolyzable matrix for the insoluble black HCN polymers , according to the spectroscopic data presented above. This polymeric matrix can be formed by different motifs of imidazoles and cyclic amides, as well as other groups.

Astrobiological Implications of HCN Polymers. The study of HCN polymers is relevant to prebiotic chemistry and astrobiology. It has been suggested that primitive Earth may have been covered by water and carbonaceous materials originating from bolide bombardment or from photochemical reactions in the atmosphere, including HCN polymers, which would have supplied essential components for establishing protein/nucleic acid life [1]. As mentioned above, it is well-known that HCN polymers release nucleic acid bases and amino acids. In addition to those species, dicarboxylic acids were detected in this work. Some of these acids could have been involved in a primordial variant of the Krebs cycle [36]. Metabolic-type reactions could have played a central role in the processes that led to the origin of life. Although the hypothetical existence of self-organizing proto-metabolic cycles is controversial [37], our preliminary GC/MS analysis of the insoluble black HCN polymers demonstrated that the materials required for the development of a proto-metabolic system, and structural and informational materials could be synthesized simultaneously. Recently, polycyclic aromatic hydrocarbons (PAHs) have been suggested as energy transduction elements [38] in possible proto-metabolic systems, because they absorb light in the near-UV and blue region, and can capture light energy either by donating electrons to produce molecules with higher chemical potential or by generation of ionic gradients. Thus, HCN polymers may also be capable of capturing or transducing energy due to the unsaturated nature of their matrix (Scheme 3).

Conclusions. â&#x20AC;&#x201C; This study offers new insights into the structural characterization of HCN polymers based on the complementary use of several spectroscopic and analytical tools. Based on the data reported here, we conclude that insoluble black HCN polymers are formed by an unsaturated complex matrix, which can retain H2O, and important bioorganic compounds or their precursors. We propose that this insoluble matrix is composed of several polyheterocyclic motifs, mainly imidazoles and cyclic polyamides. HCN Polymerization in H2O yields many compounds considered to be potential life precursors. Furthermore, insoluble black HCN polymers can be obtained under several environmental conditions, which explains their ubiquity in a large diversity of space environments.

38

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

The authors used the research facilities of Centro de Astrobiolog a (CAB) and were supported by the Instituto Nacional de Te´cnica Aeroespacial Esteban Terradas (INTA), and the projects AYA2009-13920C02-01 and BIO2007-67523 of the Ministerio de Ciencia e Innovacio´n (Spain). We thank M. T. Ferna´ndez for recording the IR spectra, I. Sobrados from ICMM (Instituto de Ciencia de Materiales de Madrid, CSIC, Spain) for recording the 13C-NMR spectra, and S. Veintemillas and G. Mun˜oz-Caro for useful comments.

Experimental Part Preparation of Insoluble Black HCN Polymers . Insoluble black HCN polymers were prepared according to a method described by Borquez et al. [31] using solns. of equimolar concentrations of NH4Cl and NaCN in pure H2O (MilliQ grade) at concentrations of 1 and 10m. The 1m solns. were left to stand for 3 d, one week, and four weeks, and the 10m soln. was left to stand for 3 d. All of the solns. were heated to 388 for the duration of the reaction for comparison to previous work concerning the formation of tholins under a CH4 atmosphere and spark discharges [20] [21] [39]. After the reaction, the samples were filtered using a glass fiber filter (0.25 mm) and washed with dist. H2O (4 ) to collect the black solids, which were dried under reduced pressure. The results of the elemental analyses and the XPS recordings indicated the absence of salts in all of the samples. NH4Cl and NaCN were obtained from Sigma Aldrich.

Hydrolysis of the Insoluble Black HCN Polymers . For hydrolysis, 6n HCl at 1108 for 24 h was used.

The hydrolyzed samples were centrifuged. The insoluble black residues were collected, washed with H2O (3 ), dried under reduced pressure, and studied by the same techniques as the insoluble black HCN polymers . The supernatants were collected, freeze-dried, and analyzed.

Standard Spectroscopy Techniques. UV/VIS Spectra were obtained using an Agilent 8453 spectrophotometer. All spectra were recorded in DMSO. All samples were partially soluble. IR Spectra were obtained using a Nexus Nicolet FT-IR spectrometer. The spectra were obtained in CsI pellets in the reflectance mode and were recorded from 4000 to 500 cm 1. 13C-CP-MAS-NMR Spectra were obtained using a Bruker Advance 400 spectrometer and a standard cross-polarization pulse sequence. Samples were spun at 10 kHz, and the spectrometer frequency was set to 100.62 MHz. A contact time of 1 ms and a period of 5 s between successive accumulations were used. The number of scans was 5000, and the chemical shift values were referenced to TMS. Elemental Analysis. Elemental C, H, and N analyses were performed using a LECO CHNS-932 elemental analyzer. For the elemental O analyses, a LECO VTF-900 analyzer was used.

XP Spectroscopy. XP Spectra were collected in an ultra-high vacuum (UHV) system equipped with a multi-channeltron hemispherical electron energy-analyzer (Phoibos, Specs GmbH) using an Al Ka X-ray source, and recorded at normal emission (incident angle 458). For the XPS experiments, 25% KBr pellets were prepared for all the samples analyzed. During the recording, the samples showed a small charge effect induced by the X-ray radiation. Therefore, to maintain the energy position of the first scan, the peaks were energy-corrected. High-resolution N 1s, C 1s, and O 1s core level spectra were recorded with an estimated resolution of 0.9 eV. The spectra were well described by the superposition of several Doniach Sunjic curve components. Data were analyzed with CASAXPS and FittXPS software. The intensities of the XPS core levels were evaluated by the peak areas after subtraction of the standard background area according to the Shirley procedure. Assignment of the binding energy was performed using standard spectra from the Handbook of X-ray Photoelectron Spectroscopy [40] according to previous experimental data for which reference samples were used [21] [22] [41].

Organic-Compound Analysis by GC/MS. To identify polar organic compounds of biological interest in hydrolyzed supernatants, the following protocol was used: i) hydrolyzed samples were freeze-dried to remove H2O and HCl for 48 h; ii) ca. 6 mg of the dried residues in 100 ml of BSTFA þTMCS (N,Obis(trimethylsilyl)trifluoroacetamide with Me3SiCl (Thermo Scientific)) were maintained at 608 overnight to obtain the corresponding TMS derivatives; iii) the derivatized samples were analyzed by GC/MS using the following GC oven program: 608 (initial temp.) held for 1.5 min, heating to 1308 at 58/min and holding for 11 min, heating to 1808 at 108/min and holding for 10 min, heating to 2208 at 208/min and holding for 15, and heating to 3008 at 108/min and holding for 10 min. An injection volume of 2 ml was

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

39

used for each sample. The temp. of the injector was 2208, and the injections were performed in splitless mode. The detector temp. was 3008. The flow rate was 9.2 psi. The GC/MS analyses were performed in full-scan mode on a 6850 network GC system coupled to a 5975 VL MSD with triple-axis detector operating in electron-impact (EI) mode at 70 eV (Agilent), using an HP-5 MS column (30 m 0.25 mm i.d. 0.25 mm film thickness) and He as a carrier gas. Identification of the GC/MS peaks was confirmed by comparison of retention times and mass spectra with those of external standards purchased from Sigma Aldrich and Fluka.

Thermogravimetry (TG) Analysis. A Perkin-Elmer TGA-7 was used for TG measurements. The instrument was calibrated both for temp. and weight by standard methods. Non-isothermal experiments were performed over a temp. range of 25 – 10008 at a heating rate of 108/min. The average sample weight was ca. 10 mg, and the dry N2 flow rate was 100 cm3/min.

REFERENCES [1] [2] [3] [4] [5] [6]

[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]

C. N. Matthews, R. D. Minard, Faraday Discuss. 2006, 133, 393.

Q. W. Chen, C. L. Chen, Curr. Org. Chem. 2005, 9, 989.

R. Saladino, C. Crestini, G. Costanzo, E. DiMauro, Curr. Org. Chem. 2004, 8, 1425.

R. Stribling, S. L. Miller, Origins Life 1987, 17, 261.

C. N. Matthews, R. D. Minard, Astrobiology 2007, 7, 490.

P. Ehrenfreund, S. Charnley, D. Wooden, From ISM Material to Comet Particles and Molecules , in COMETS II , Eds. M. Festou, H. U. Keller, H. Weaver, University of Arizona Press, Tucson, 2004, p. 115.

N. Fray, Y. Be´nilan, H. Cottin, M.-C. Gazeau, R. D. Minard, F. Raulin, Meteorit. Planet. Sci. 2004, 39, 581.

E. Quirico, G. Montagnac, V. Lees, P. F. McMillan, C. Szopa, G. Cernogora, J.-N. Rouzaud, P. Simon, J.-M. Bernard, P. Coll, N. Fray, R. D. Minard, F. Raulin, B. Reynard, B. Schmitt, Icarus 2008, 198, 218.

J. P. Ferris, W. J. Hagan Jr., Tetrahedron 1984, 40, 1093.

S. Miyakawa, H. J. Cleaves, S. L. Miller, Origins Life Evol. Biosphere 2002, 32, 209.

A. B. Voet, A. W. Schwartz, Bioorg. Chem. 1983, 12, 8.

J. P. Ferris, P. C. Joshi, E. H. Edelson, J. G. J. Lawless, J. Mol. Evol. 1978, 11, 293.

J. P. Ferris, D. B. Donner, A. P. Lobo, J. Mol. Biol. 1973, 74, 511.

I. Mamajanov, J. Herzfeld, J. Chem. Phys. 2009, 130, 134504.

I. Mamajanov, J. Herzfeld, J. Chem. Phys. 2009, 130, 134503.

K. Umemoto, M. Takahashi, K. Yokota, Origins Life 1987, 17, 283.

J. P. Ferris, E. H. Edelson, J. M. Auyeung, P. C. Joshi, J. Mol. Evol. 1981, 17, 69.

C. N. Matthews, R. E. Moser, Nature 1967, 215, 1230.

T. Vçlker, Angew. Chem. 1960, 72, 379.

M. Ruiz-Bermejo, C. Menor-Salva´n, J. L. de la Fuente, E. Mateo-Mart , S. Osuna-Esteban, J. A. Mart n-Gago, S. Veintemillas-Verdaguer, Icarus 2009, 204, 672.

M. Ruiz-Bermejo, C. Menor-Salva´n, E. Mateo-Mart , S. Osuna-Esteban, J. A. Mart n-Gago, S. Veintemillas-Verdaguer, Icarus 2008, 198, 232. ´ . Mart n-Gago, S. M. Ruiz-Bermejo, C. Rogero, C. Menor-Salva´n, S. Osuna-Esteban, J. A Veintemillas-Verdaguer, Chem. Biodiversity 2009, 6, 1309.

M. P. Eastman, F. S. E. Helfrich, A. Umantsev, T. L. Porter, R. Weber, Scanning 2003, 25, 19.

M. Labadie, R. Jensen, E. Neuzil, Biochim. Biophys. Acta 1968, 165, 525.

J. R. Garbow, J. Schaefer, R. Ludicky, C. N. Matthews, Macromolecules 1987, 20, 305.

S. A. Liebman, R. A. Pesce-Rodriguez, C. N. Matthews, Adv. Space Res. 1995, 15, 71.

M. Toba, T. Nakashima, T. Kawai, Macromolecules 2009, 42, 8068.

M. S. Rodr guez-Morgade, G. D. Pantos, E. Caballero, J. L. Sessler, T. Torres, Macroheterocycles 2008, 1, 40.

J. L. de la Fuente, M. Ruiz-Bermejo, C. Menor-Salva´n, S. Osuna-Esteban, Polym. Degrad. Stab.

2011, 96, 943.

40

CHEMISTRY & BIODIVERSITY – Vol. 9 (2012)

[30] A. Negro´n-Mendoza, Z. D. Draganic´, R. Navarro-Gonza´lez, I. G. Draganic´, Radiat. Res. 1983, 95, 248.

[31] E. Borquez, H. J. Cleaves, A. Lazcano, S. L. Miller, Origins Life Evol. Biosphere 2005, 35, 79.

[32] R. A. Sanchez, J. P. Ferris, L. E. Orgel, J. Mol. Biol. 1967, 30, 223.

[33] C. J. Hawker, Adv. Polym. Sci. 1999, 147, 113.

[34] D. M. Johnson, S. E. Reybuck, R. G. Lawton, P. G. Rasmussen, Macromolecules 2005, 38, 3615.

[35] D. M. Johnson, P. G. Rasmussen, Macromolecules 2000, 33, 8597.

[36] E. Mele´ndez-Hevia, T. G. Waddell, M. Cascante, J. Mol. Evol. 1996, 43, 293.

[37] L. E. Orgel, Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 12503.

[38] P. Ehrenfreund, S. Rasmussen, J. Cleaves, L. Chen, Astrobiology 2006, 6, 490.

[39] M. Ruiz-Bermejo, C. Menor-Salva´n, S. Osuna-Esteban, S. Veintemillas-Verdaguer, Origins. Life Evol. Biosphere 2007, 37, 123.

[40] J. F. Moulder, W. F. Stickle, P. E. Sobol, K. D. Bomben, Handbook of X-Ray Photoelectron Spectroscopy , Perkin-Elmer Corporation, Physical Electronics Division (USA), 1992.

[41] K. Y. Kang, B. I. Lee, J. S. Lee, Carbon 2009, 47, 1171.

Received February 2, 2011

Polymer Degradation and Stability 96 (2011) 943e948

Contents lists available at ScienceDirect

Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab

Thermal characterization of HCN polymers by TGeMS, TG, DTA and DSC methods José L. de la Fuente a, *, Marta Ruiz-Bermejo b, César Menor-Salván b, Susana Osuna-Esteban b a

Instituto Nacional de Técnica Aeroespacial “Esteban Terradas” (INTA), Carretera Torrejón-Ajalvir, Km. 4, E-28850 Torrejón de Ardoz, Madrid, Spain Centro de Astrobiología [Consejo Superior de Investigaciones Científicas-Instituto Nacional de Técnica Aeroespacial “Esteban Terradas” (CSIC-INTA)], Carretera Torrejón-Ajalvir, Km. 4, E-28850 Torrejón de Ardoz, Madrid, Spain

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 December 2010 Received in revised form 17 January 2011 Accepted 28 January 2011 Available online 22 February 2011

This paper presents a thermogravimetry (TG) study of hydrogen cyanide polymers, synthesized from the reaction of equimolar aqueous solutions of sodium cyanide and ammonium chloride. Differential thermal analysis (DTA) and differential scanning calorimetry (DSC) were also used to evaluate the thermal behaviour of these black polymers, which play an important role in prebiotic chemistry. A coupled TGemass spectrometer (MS) system allowed us to analyze the principal volatile thermal decomposition and fragmentation products of the isolated HCN polymers under dynamic conditions and an inert atmosphere. After dehydration, a multi-step decomposition occurred in this particular polymeric system, due to the release of ammonia, hydrogen cyanide (depolymerization reaction), isocyanic acid (or cyanic acid) and formamide; these two latter species allow us identify bond connectivities.

Finally, data collected from TG experiments in an oxidative atmosphere showed significant differences at higher temperatures, above 400 C. According to these results, the different techniques of thermal analysis here applied have demonstrated to be an adequate methodology for the study and characterization of this complex macromolecular system, whose structure remains controversial even today. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: HCN polymers Coupled TGeMS Thermal decomposition DSC DTG

1. Introduction Hydrogen cyanide (HCN) is a ubiquitous molecule in the universe. HCN, whether as gas, a pure liquid, or in solution, polymerizes easily over a wide range of temperature and pressure in the presence of amounts of a base catalyst to yield a dark complex solid, known as “HCN polymers” [1e3]. In addition, recently HCN polymers have been synthesized from the thermal decomposition of formamide [4].

The HCN polymers may be among the organic macromolecules most readily formed within the solar system. The HCN polymers could be the major component of the dark matter observed on many bodies of the outer solar system including asteroids, moons, planets and, especially, comets [5,6]. It has been also proposed that the reddish haze (tholin) present in the atmosphere of Titan could be analogous to HCN polymers [7].

Besides the interest in this polymeric system from a prebiotic perspective, the HCN polymers can be used as a starting material for the synthesis of carbon nitride materials, expected to have unique mechanical properties such as wear, resistance coating and hardness [8e18]. However, certain structural aspects of these HCN

* Corresponding author. Tel.: þ34 91 5206841; fax: þ34 91 5206611.

E-mail address: fuentegj@inta.es (J.L. de la Fuente).

0141-3910/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymdegradstab.2011.01.033

polymers remain controversial, despite extensive efforts over the last 50 years. Part of this is due to the various analytical techniques different authors have used, part of it is due to the manner in which the polymers were synthesized.

Thermogravimetry (TG) is one of the most commonly used thermal analysis techniques for the characterization of both inorganic and organic materials, including polymers. It provides quantitative results regarding the weight loss of a sample as a function of temperature or time. Moreover, TG measurements give basic information about the thermal properties of the material and its composition. The derivative thermogravimetry (DTG) can be used to investigate the differences between thermograms.

In recent years, multi-purpose thermal analysis coupled with evolved gas analyzers has become very popular, as it can be used to carry out further analysis of evolved gases during TG measurements; producing details of thermal decomposition processes, which in turn facilitate estimation of sample structure and composition. Regarding instrumentation, a great number of gas detectors or analyzers have been utilized in evolved gas analysis (EGA), being the most important analyzer Fourier transform infrared (FTIR) spectrometers, and pre-eminently mass spectrometers (MS). Both methods can be used to record spectra repetitively, thereby producing a time-dependent record of the composition of the gas phase, from which EGA curves can be constructed for selected species [19e22].

944

J.L. de la Fuente et al. / Polymer Degradation and Stability 96 (2011) 943e948

In the present work, a systematic thermal degradation study stability of the HCN polymers has been carried out both in argon (thermal stability) and in oxygen (thermal-oxidative stability) atmosphere. A coupled TGeMS system was used to analysis the principal species evolved during dynamic thermal decomposition or fragmentation processes for the HCN polymers synthesized. In addition, differential thermal analysis (DTA) and differential scanning calorimetry (DSC) were also employed to evaluate different thermal features of these black polymers. The results obtained from this thermal analysis provide useful information in drawing a correlation between their properties and chemical structure, and to understand their decomposition mechanism.

2. Experimental section Three HCN polymer samples were synthesized using different reaction times. The polymers were prepared from solutions of equimolar amounts of ammonium chloride (NH4Cl) and sodium cyanide (NaCN) in pure water (MilliQ grade) at 1 M concentration, in a similar way to the one described by Borquez et al. [23]. The solutions were left standing for three, ten days, and one month without stirring. All solutions were heated at 38 C during the reaction time for comparative purposes with respect to our previous works about the formation of tholins under CH4 atmosphere [24e26]. Samples were filtered using glass fibre filters after reactions, and washed with distillate water ( 4) collecting black solids which were dried under reduced pressure. The yields in this work were calculated from the initial amount of carbon contained in the initial NaCN. The NH4Cl and NaCN were obtained from SigmaeAldrich.

Elemental C, H and N analyses were performed using a LECO CHNS-932 elemental analyzer. For the elemental O analyses a LECO VTF-900 analyzer was used. Thermogravimetry (TG), derivative thermogravimetry (DTG) and differential thermal analysis (DTA) measurements were preformed with a simultaneous thermal analyzer model Q600 of TA Instruments. Non-isothermal experiments were performed in the temperature range from 25 to 1000 C and at a heating rate of 10 C/min. The average sample weight was w10 mg, and the argon and oxygen flow rate was 100 ml/min. A coupled TGeMS system was used to make an analysis of principal species evolved during the dynamic thermal decomposition of fragmentation processes of all the polymers samples; with an electron-impact quadruple mass-selective detector, model Thermostar QMS200 M3. A differential scanning calorimetry (DSC; Perkin Elmer DSC/TA7DX, PC series with liquid nitrogen for low temperatures) was also used. The temperature and heat flow were calibrated with common standards, such as indium. Samples (w5 mg) were scanned at 10 C/min under dry nitrogen (100 ml/min).

Table 1 Experimental conditions, yields (calculated from the initial amount of carbon containing in the initial NaCN) and elemental analysis for the series of HCN black polymers synthesized.

Polymer sample

Reaction time (days)

Yield (%)

C (%)

H (%)

N (%)

O (%)

Empirical formula

1 2 3

3 10 30

25 27 33

36.5 36.5 36.2

4.4 4.4 3.9

40.8 41.9 39.6

17.4 17.2 19.2

C3H4N3O C3H4N3O C5H6N4O2

3.1. Thermogravimetric and DSC analysis The TG thermograms of HCN polymers were recorded under dynamic conditions from room temperature to 1000 C at a heating rate of 10 C/min and under argon atmosphere. Representative thermogravimetric curves are shown in Fig. 1a. As can be seen, the thermal degradation of the different samples shows practically an identical behaviour, which demonstrates the high degree of similarity between them; in spite of the fact that they were obtained with different reaction times.

From this figure, it is evident that the thermal degradation of all samples can be divided into three stages: drying stage (<150 C), main pyrolysis stage (150e500 C) and carbonization (>500 C).

The first one, between 25 and 150 C, involves a mass loss of around 9 wt%, and corresponds to the vaporization of moisture, to the desorption of water and to the possible emission of volatile organic compound. This initial step of weight loss at low temperatures indicates the hydrophilicity of the HCN polymers. The degradation starts around 150 C and distributes over a broader temperature range. The second stage, between 150 and 500 C, corresponds with

3. Results and discussion The yield found in the polymerization reactions as well as the results of the elemental analysis for the “HCN polymers” obtained under the experimental conditions described above are shown in Table 1. Our results are comparable to those reported by Eastman et al. [27] for short reaction times (1e10 days). For longer reaction times (30 days) a black, insoluble solid richer in oxygen was obtained. However, a longer reaction time does not have a significant influence on the yield of these insoluble products (29 4%).

The analytical and spectroscopic data, elemental analysis, FTIR, and 13C CP-MAS-NMR, of the HCN polymers synthesized are very similar, indicating that basically the same macrostructure is present in all of them (see Figs. 1 and 2 in reference [24]). These spectra greatly resemble those reported in the literature for HCN polymers prepared under different experimental conditions [5,28e30].

Fig. 1. (a) TG and (b) DTG curves for HCN polymer samples synthesized. Heating rate was 10 C/min and the argon atmosphere.

J.L. de la Fuente et al. / Polymer Degradation and Stability 96 (2011) 943e948

945

peak at 80 is attributed to the evaporation of absorbed water, and the rest of the observed peaks, at 265, 675 and 885 C, reflect decomposition.

The DSC thermograms of the samples run in nitrogen atmosphere, where the first endothermic peaks are clearly observed at 90 and 255 C, are in good agreement with the DTA and DTG peaks found (the small differences are due to the technique considered).

3.2. Mass spectrometric thermal analysis

Fig. 2. DTA curve for an HCN polymer sample. Heating rate was 10 C/min and the argon atmosphere.

a mass loss of about 25 wt%. This broad stage may therefore be rationalized to be a composite of two thermal events as described below. The final stage occurs between approximately 500 and 1000 C. This last step is the major thermal decomposition stage, and the samples lost a maximum of ca. 50% of weight. At 1000 C, the char residue is another characteristic of these HCN polymers, being equal to 13, 17 and 15% for the samples 1e3, respectively (Table 1).

The first derivatives of thermograms were calculated to highlight the inflection points that indicate thermal transitions. These DTG curves are shown in Fig. 1b, where a deconvolution of one of these curves into an individual Gaussian peak has been made assuming a linear background over the temperature range of the fitting. The temperatures of the DTG maxima with the corresponding rates of weight loss, dW/dT, for the three samples are collected in Table 2. Fig. 1b illustrates more clearly the small differences in the thermal decomposition behaviour of the three samples. The first DTG peak appears at 80 C, probably resulted from the desertion of water and/or organic volatile material as it was mentioned before. At the second stage, a double peak observed, the first one around 260 C and the last one appears between 334 and 371 C. A more significant change due to thermal degradation takes place from the third peak, and continues with a main degradation at the last stage, with a peak of maximum loss weight between 651 and 668 C, and finally one last shoulder around 810 C. The three samples follow globally the same stages of pyrolysis. However, some small differences appear, thus for the sample 1, obtained to the lowest reaction time, its first three maxima decomposition rates take place at slightly lower temperatures than the polymer samples prepared with longer reaction times.

The DTA thermogram under argon of HCN polymer samples under study exhibits different stages according with the previous DTG analysis, with a very broad endotherm between 300 and 950 C, as is shown in Fig. 2. The first and well defined endothermic

MS coupled with TG system has been used to study volatile species of thermal decomposition and fragmentation processes for all HCN polymer samples obtained. Dynamic measurements were carried out in argon atmosphere. Fig. 3 (as an example) presents the ion current for m/z detected in the MS versus temperature for the sample 2 (identical results were obtained for the other samples). In this inert atmosphere major signals are observed in the range 50e550 C with an MS peak for H2Oþ (with m/z ¼ 18) that appeared at 85 C (these data coincide with loss observed on TG curves and DSC data). The second major signal corresponds with m/z ¼ 17, which is attributed to OHþ or/and ammonia NHþ 3 , with maximum rates of forming centred at 90 C (probably for OHþ), and a following broad peak (extended between 155 and 530 C) centred around 280 C. This second broad MS peak presents the same profile than the third major signal for m/z ¼ 16 (NHþ 2 ).

In addition to the mentioned products, H2O and NH3, other significant components are released from the polymer such as: hydrogen cyanide, isocyanic acid HeN]C]O (or cyanic acid HeOeCN), and formamide HCONH2. The HCN evolution (thermal depolymerization) with m/z ¼ 26 for CNþ and 27 for HCNþ, starts at 120 C and takes place over all the temperature range under study.

The shape of the main peaks of mass spectrometry resembles those of DTG curves. These profiles are very similar to those ion currents corresponding to Cþ and CHþ with m/z ¼ 12 and 13, respectively.

On the other hand, the release of gases contributing to m/z ¼ 43, associated to isocyanic acid (or cyanic acid), begins at 150 C, and it also happens on a large range of temperature with a plateau form.

The profiles for m/z ¼ 42 (NCOþ) and m/z ¼ 30 (NOþ) show the same tendencies due to their ion fragmentation. The release of formamide is followed with the profile of m/z ¼ 45 and 44 (CONHþ 2 ), where two peaks around 255 and 400 C can be observed.

Finally, at temperatures above 550 C, the gaseous species can be mainly due to the tar cracking of the HCN polymer main chain, observing TGeMS profiles for the m/z ¼ 51 and 52 and probably with ion fragmentations producing a signal on the profiles of m/ z ¼ 24, 13 and 12. At these high temperatures the cleavage of the polymer chain can originate, for example, the formation of NCeC^CHþ (m/z ¼ 51) and NCeCH]CHeþ (m/z ¼ 52) type structures.

Fig. 4 compares the ion current curve for the main thermal decomposition gases for an HCN polymer sample under study, with a signal at m/z ¼ 17, 18, 27, 43 and 45. As is shown in this figure, the intensity order of MS signal is m/z ¼ 18 (8.5$10 9) > 17 (2$10 9) > 27 (5$10 10) > 43 (15$10 12) > 45 (<10 12). This indicates the majority of thermal decomposable components in the HCN polymers can release H2O, ammonia, HCN, isocyanic acid and

Table 2 Characteristic temperatures for the thermal decomposition of synthesized HCN black polymers in argon atmosphere, DTG maxima with the corresponding rates of weight loss, dW/dT. Polymer sample Tmax1 ( C) dW1/dT (wt%/ C) Tmax2 ( C) dW2/dT (wt%/ C) Tmax3 ( C) dW3/dT (wt%/ C) Tmax4 ( C) dW4/dT (wt%/ C) Tmax5 ( C) dW5/dT (wt%/ C) 1 2 3

68.1 81.0 78.1

0.101 0.087 0.093

252 260 259

0.103 0.085 0.083

334 368 371

0.093 0.084 0.082

651 653 668

0.148 0.126 0.133

818 807 815

0.0102 0.094 0.089

946

J.L. de la Fuente et al. / Polymer Degradation and Stability 96 (2011) 943e948

Fig. 3. Ion intensity curves for an HCN polymer sample, heating at 10 C/min in argon atmosphere.

formamide. These results are in agreement with the scarce data of the literature concerning the thermal decomposition of the system under study [7,31,32]. On one hand, the recent work of Cataldo and co-worker [31], where these authors analyze HCN polymers obtained from thermal decomposition of formamide, and they ďŹ nd in their pyrolysis a large amount of gaseous HCN in a wide range of temperatures together with ammonia and isocyanic acid. On the other hand, the pioneering study of Ferris et al. where the major pyrolytic reaction products were identiďŹ ed as CO2, H2O, HCN, CH3CN, HCONH2, and pyridine [32].

The hydrophilic character of these HCN polymers demonstrated in this paper is also a feature of other well-known polymers, as for example polyamides based on cyanoimidazoles as has been welldescribed by Thurber and Rasmussen [33]. This fact is very relevant, so that the last studies about the structure of this polymeric system, performed Mamajanov and Herzfeld and based on solid state NMR, propose a polyaminoimidazole as the main chain for HCN polymers [30]. On the other hand, the pyrolytic formation of some of these products, such as formamide and isocyanic acid, is consistent with the presence of amide groups in these HCN polymers. The presence

J.L. de la Fuente et al. / Polymer Degradation and Stability 96 (2011) 943e948

947

Fig. 4. (a) DTG curve and (b) MS signal comparison of m/z ¼ 17, 18, 27, 42 and 44 for an HCN polymer sample, heating at 10 C/min in argon atmosphere.

of secondary amide groups has been established by solid state NMR in HCN polymers obtained from equimolar amounts of H13CN and HC15N by Matthews and co-workers [34,35].

It should be mentioned that the intensity of m/z signal in this technique is related to the amount of species, but not decisive. It is also determined by different ionized energies of the distinct gas species. Therefore, the quantitative comparison of MS signals of different samples and components is not performed in this paper.

On the other hand, it is also important to note that pyrolysis of these HCN polymers releases gaseous hydrogen cyanide, isocyanic acid and formamide in a wide temperature range. However, some of these molecules (as well as other oligomers from them) could not be thermal decomposition products, they can simply be side products, formed during the polymerization reaction, and trapped on the polymeric matrix. Thus for example, urea could be one of these side products, however it is thermally unstable, forming ammonia and cyanic acid (or isocyanic acid) in a simple decomposition reaction [36,37].

3.3. Effect of oxygen on the thermal decomposition The effect of the oxygen on the thermal degradation of the HCN polymers was also analyzed. The HCN polymers thermal-oxidative degradation has a different behaviour than in argon degradation.

Fig. 5a and b shows the TG and DTG curves, respectively, for an HCN polymer sample degraded in argon and oxygen environments from room temperature to 1000 C. A detailed analysis of the curves allow us to conclude that the two first degradation stages, until temperature around 400 C, are not apparently affected by the type of environment under study. Below this temperature the oxygen notably decreases the stability of the HCN polymers. The first DTG peak, caused by the presence of water and/or volatile compounds is shifted to lower temperatures, and now appears around 50 C in the thermo-oxidative degradation. At the second stage, where two separated DTG maxima under argon flux can be registered, it is unaffected by the change of the atmosphere. This fact suggests that these peaks, which correspond to the main pyrolysis stage, do not arise from the thermally weak structures sensitive to oxidation.

Fig. 5. (a) TG and (b) DTG curves for an HCN polymer sample. Heating rate was 10 C/ min and the oxygen atmosphere.

However, it is noticeable the great difference observed from ca.

400 C, where the HCN polymers present higher stability in argon than in an oxygen atmosphere at higher temperatures. Besides, the TG analyses of all HCN polymers under oxidative atmosphere provide no residues. Thus, it is very illustrative the clear change at the third decomposition stage, which is now quite narrow, between 400 and 600 C (versus 500 and 1000 C in argon). In an inert atmosphere, the DTG maximum is observed around 660 C, and in this case it appears at lower temperatures, 540 C. In addition, the second of DTG peaks, which appear as a shoulder around 815 C in argon, it is not observed in the presence of oxygen. The dW/dT values of this peak significantly increase under oxygen atmosphere, as this figure clearly shows. In this range of temperatures decompositionecarbonization reactions normally take place. Thus the presence of oxygen diminishes the stability of the macrostructure of the HCN polymer, and leads to intense thermo-oxidation processes. This means that the oxygen accelerated the mass loss of this polymer through oxidation.

4. Conclusions Various samples of HCN polymers were synthesized from the reaction of equimolar amounts of NaCN and NH4Cl in water, and characterized by FTIR and solid state NMR for a study of their thermal properties using different analytical techniques such as TG, DTA, DSC and TGeMS. The TG curves showed that the mass loss percentages for the decomposition were practically the same for each sample; therefore their thermal stability is not essentially influenced by the reaction time used in their syntheses. The thermal degradation of

948

J.L. de la Fuente et al. / Polymer Degradation and Stability 96 (2011) 943e948

all the samples was divided into three stages: drying stage (<150 C), main pyrolysis stage (150e500 C) and carbonization (>500 C). The first one with a mass loss of w9 wt%, indicated the high hydrophilicity of the HCN polymers. The endothermic decomposition that began at about 150 C is characterized by a continuous release of HCN (with the maxima of ion currents associated with the maxima of loss weight observed on DTG curves) and isocyanic acid (or cyanic acid) up to temperature of 900 C. Together these species, also release ammonia and formamide, however, the maxima of elimination of these volatile species takes place at lower temperature, in the range 150e450 C. The pyrolytic formation of some of these products, such as formamide and isocyanic acid, is consistent with the presence of amide groups in these HCN polymers. Also MS peak for ion fragments with m/ þ þ þ z ¼ 12 (Cþ), 13 (CHþ), 16 (NHþ 2 ), 24 (C2 ), 26 (CN ), 30 (NO ), 42 þ þ (NCO ), 44 (H2NCO ), 51 and 52 were detected for the investigated polymers. DTA and DSC measurements revealed the exclusive endothermic nature for all the thermal decomposition processes.

The oxidative decomposition of our HCN polymers showed the same stages, in the same manner as in argon. The first peak is slightly displaced to lower temperature, as well as the pyrolytic reactions that take place at the third degradation stage. This last stage is narrowed, and the corresponding rate of weight loss is considerably enhanced. In this final high temperature region, oxygen reacts readily with the char residue, resulting in rapid char oxidation.

To our knowledge this is the first reported application of TGeMS technique to explore the thermal stability of this complex, heterogeneous and particular macromolecular system. These polymers have been suggested to be the key starting point for the origin of protein/nucleic acid based life, and therefore could be the critical link connecting cosmochemistry and biochemistry, their study being a topic that continues to be investigated in many research centres. It is believed that the results shown in this paper represent a non-negligible advancement in the structural characterization of HCN polymers and in the understanding of their thermal decomposition mechanism.

Acknowledgements The authors have used the research facilities of Centro de Astrobiología (CAB) and have been supported by Instituto Nacional de Técnica Aeroespacial “Esteban Terradas” (INTA) and the project AYA2009-13920-C02-01 of the Ministerio de Ciencia e Innovación (Spain). We thank R. Rojas and S. Veintemillas from ICMM (Instituto de Ciencia de Materiales de Madrid. CSIC) for their useful comments.

References [1] Minard RD, Yang W, Varma P, Nelson J, Matthews CN. Heteropolypeptides from poly-a-cyanoglycine and hydrogen cyamide: a model for the origin of proteins. Science 1975;90:387e9.

[2] Ferris JP, Hagan WJ. HCN and chemical evolution: the possible role of cyano compounds in prebiotic synthesis. Tetrahedron 1984;40:1093e120.

[3] Matthews CN, Minard RD. Hydrogen cyanide polymer, comets and the origin of life. Faraday Discuss 2006;133:393e401.

[4] Cataldo F, Patane G, Compagnini G. Synthesis of HCN polymer from thermal decomposition of formamide. J Macromol Sci A Pure Appl Chem 2009;46: 1039e48.

[5] Quirico E, Montagnac G, Lees V, McMillan PF, Szopa C, Cernogora G, et al. New experimental constraints on the composition and structure of tholins. Icarus 2008;198:218e31.

[6] Matthews CN, Minard RD. Hydrogen cyanide polymers connect cosmochemistry and biochemistry. Organic matter in space. In: Proceedings IAU Symposium No. 251. 2008. p. 453e457.

[7] Fray N, Bénilan Y, Cottin H, Gazeau M-C, Minard RD, Raulin F. Experimental study of degradation of polymers: application to the origin of extended sources in cometary atmosphere. Meteorit Planet Sci 2004;39:581e7.

[8] Subrayan RP, Rasmussen PG. An overview of materials composed of carbon and nitrogen. Trends Polymer Sci 1995;3:165e72.

[9] Shiao J, Hoffman RW. Studies of diamond-like and nitrogen-containing diamond-like carbon using laser Raman spectroscopy. Thin Solid Films 1996; 283:145e50.

[10] Chowdhury AKMS, Cameron DC, Hashmi MSJ. Vibrational properties of carbon nitride films by Raman spectroscopy. Thin Solid Films 1998;332:62e8.

[11] Komatsu T, Samejima M. Preparation of carbon nitride C2N by shock-wave compression of poly(aminomethineimine). J Mater Chem 1998;8:193e6.

[12] Komatsu T. Attempted preparation of diamond-like carbon nitride by explosive shock compression of poly(methineimine). J Mater Chem 1998;8: 2475e9.

[13] Yap YK, Kida S, Aoyama T, Mori Y, Sasaki T. Binding state transformation in high temperature synthesized CN thin films. Diam Relat Mater 1999;8:614e7.

[14] Hsu C-Y, Hong FC-N. The effect of substrate temperature on the growth of CNx films with b-C3N4-like microcrystallites by an inductively coupled plasma (ICP) sputtering method. Diam Relat Mater 1999;8:1315e23.

[15] Weich F, Widani J, Frauenheim Th. Paracyanogen-like structures in highdensity amorphous carbon nitride. Carbon 1999;37:545e8.

[16] Collins C, Thadhani N, Iqbal Z. Shock-compression of CeN precursors for possible synthesis of b-C3N4. Carbon 2001;39:1175e81.

[17] Roy D, Chhowalla M, Hellgren N, Clyne TW, Amaratunga GAJ. Probing carbon nanoparticles in CNx thin films using Raman spectroscopy. Phys Rev B 2004; 70:35406e12.

[18] Hinago H, Nagahara H. European patent; 2008. No. 1,939,141 A1.

[19] Warrington SB. Thermal analysis and calorimetry. In: Günzler H, Willimas A, editors. Handbook of analytical techniques. Weinheim, Germany: Wiley-VCH Verlag GmbH; 2001. p. 83344.

[20] Warrington SB. Evolved gas analysis. In: Charsley EL, Warrington SB, editors.

Thermal analysis-techniques and applications. London: Royal Society of Chemistry; 1992.

[21] Holdiness MR. Evolved gas analysis by mass spectrometry: a review. Thermochim Acta 1984;75:361e99.

[22] Szekely G, Nebuloni M, Zerilli LF. Thermal analysis-mass spectrometry coupling and its applications. Thermochim Acta 1992;196:511e32.

[23] Borquez E, Cleaves HJ, Lazcano A, Miller SL. An investigation of prebiotic purine synthesis from the hydrolysis of HCN polymers. Orig Life Evol Biosph 2005;35:79e90.

[24] Ruiz-Bermejo M, Menor-Salván C, de la Fuente JL, Mateo-Martí E, Osuna-Esteban S, Martín-Gago JA, et al. CH4/N2/H2-spark hydrophobic tholins: a systematic approach to the characterisation of tholins. Part II. Icarus 2009; 204:672e80.

[25] Ruiz-Bermejo M, Menor-Salván C, Mateo-Martí E, Osuna-Esteban S, Martín-Gago JA, Veintemillas-Verdaguer S. CH4/N2/H2-spark hydrophylic tholins: a systematic approach to the characterisation of tholins. Icarus 2008; 198:232e41.

[26] Ruiz-Bermejo M, Menor-Salván C, Osuna-Esteban S, VeintemillasVerdaguer S. Prebiotic microreactors: a synthesis of purines and dihydroxy compounds in aqueous aerosol. Orig Life Evol Biosph 2007;37:123e42.

[27] Eastman MP, Helfrich FSE, Umantsev A, Porter TL, Weber R. Exploring the structure of a hydrogen cyanide polymer by electron spin resonance and scanning force microscopy. Scanning 2003;25:19e24.

[28] Liebman SA, Pesce-Rodriguez RA, Matthews CN. Organic analysis of hydrogen cyanide polymers: prebiotic and extraterrestrial chemistry. Adv Space Res 1995;15:71e80.

[29] Umemoto K, Takahashi M, Yokota K. Studies on the structure of HCN oligomers. Origins of Life 1987;17:283e93.

[30] Mamajanov I, Herzfeld J. HCN polymers characterized by SSNMR: solid state reaction of crystalline tetramer (diaminomaleonitrile). J Chem Phys 2009; 130:134504/1e134504/5.

[31] Cataldo F, Lilla E, Ursini O, Angelini G. TGA-FT-IR study of pyrolysis of poly (hydrogen cyanide) synthesized from thermal decomposition of formamide.

Implication in cometary emissions. J Anal Appl Pyrolysis 2010;87:34e44.

[32] Ferris JP, Edelson EH, Auyeung JM, Joshi PC. Structural studies of HCN oligomers. J Mol Evol 1981;17:69e77.

[33] Thurber EL, Rasmessen PG. Approaches to novel AB heteroaromatic polyamides based on cyanoimidazoles. J Polym Sci A Polym Chem 1993;31: 351e64.

[34] McKay RA, Schaefer J, Stejskal EO, Ludicky R, Matthews CN. Double-crosspolarization detection of labeled chemical bonds in HCN polymerization.

Macromolecules 1984;17:1124e30.

[35] Garbow JR, Schaefer J, Ludicky R, Matthews CN. Detection of secondary amides in HCN polymers by dipolar rotational spin-echo 15N NMR. Macromolecules 1987;20:305e9.

[36] Kamimoto M, Sakamoto R, Takahashi Y, Kanari K, Ozawa T. Investigation of latent heat-thermal energy storage materials. II. Thermoanalytical evolution of urea. Thermochim Acta 1984;74:281e90.

[37] Chen JP, Isa K. Thermal decomposition of urea and urea derivatives by simultaneous TG/(DTA)/MS. J Mass Spectrom Soc Jpn 1998;46:299e303.

Exp Astron DOI 10.1007/s10686-008-9103-z ORIGINAL ARTICLE

TandEM: Titan and Enceladus mission A. Coustenis · S. K. Atreya · T. Balint · R. H. Brown · M. K. Dougherty · F. Ferri · M. Fulchignoni · D. Gautier · R. A. Gowen · C. A. Griffith · L. I. Gurvits · R. Jaumann · Y. Langevin · M. R. Leese · J. I. Lunine · C. P. McKay · X. Moussas · I. Müller-Wodarg · F. Neubauer · T. C. Owen · F. Raulin · E. C. Sittler · F. Sohl · C. Sotin · G. Tobie · T. Tokano · E. P. Turtle · J.-E. Wahlund · J. H. Waite · K. H. Baines · J. Blamont · A. J. Coates · I. Dandouras · T. Krimigis · E. Lellouch · R. D. Lorenz · A. Morse · C. C. Porco · M. Hirtzig · J. Saur · T. Spilker · J. C. Zarnecki · E. Choi · N. Achilleos · R. Amils · P. Annan · D. H. Atkinson · Y. Bénilan · C. Bertucci · B. Bézard · G. L. Bjoraker · M. Blanc · L. Boireau · J. Bouman · M. Cabane · M. T. Capria · E. Chassefière · P. Coll · M. Combes · J. F. Cooper · A. Coradini · F. Crary · T. Cravens · I. A. Daglis · E. de Angelis· C. de Bergh · I. de Pater · C. Dunford · G. Durry · O. Dutuit · D. Fairbrother · F. M. Flasar · A. D. Fortes · R. Frampton · M. Fujimoto · M. Galand · O. Grasset · M. Grott · T. Haltigin · A. Herique · F. Hersant · H. Hussmann · W. Ip · R. Johnson · E. Kallio · S. Kempf · M. Knapmeyer · W. Kofman · R. Koop · T. Kostiuk · N. Krupp · M. Küppers · H. Lammer · L.-M. Lara · P. Lavvas · S. Le Mouélic · S. Lebonnois · S. Ledvina · J. Li · T. A. Livengood· R. M. Lopes · J.-J. Lopez-Moreno · D. Luz · P. R. Mahaffy · U. Mall · J. Martinez-Frias · B. Marty · T. McCord · C. Menor Salvan · A. Milillo · D. G. Mitchell · R. Modolo · O. Mousis · M. Nakamura · C. D. Neish · C. A. Nixon · D. Nna Mvondo · G. Orton · M. Paetzold · J. Pitman · S. Pogrebenko · W. Pollard · O. Prieto-Ballesteros · P. Rannou · K. Reh · L. Richter · F. T. Robb · R. Rodrigo · S. Rodriguez · P. Romani · M. Ruiz Bermejo · E. T. Sarris · P. Schenk · B. Schmitt · N. Schmitz · D. Schulze-Makuch · K. Schwingenschuh · A. Selig · B. Sicardy · L. Soderblom · L. J. Spilker · D. Stam · A. Steele · K. Stephan · D. F. Strobel · K. Szego · C. Szopa · R. Thissen · M. G. Tomasko · D. Toublanc · H. Vali · I. Vardavas · V. Vuitton · R. A. West · R. Yelle · E. F. Young

Received: 14 December 2007 / Accepted: 27 May 2008 © Springer Science + Business Media B.V. 2008

A. Coustenis (B) · M. Fulchignoni · D. Gautier · E. Lellouch · M. Hirtzig · B. Bézard · M. Combes · C. de Bergh · D. Luz · B. Sicardy Laboratoire d’Etudes Spatiales et d’Instrumentation en Astrophysique (LESIA), Observatoire de Paris-Meudon, 92195, Meudon Cedex, France e-mail: athena.coustenis@obspm.fr

Exp Astron

Abstract TandEM was proposed as an L-class (large) mission in response to ESA’s Cosmic Vision 2015–2025 Call, and accepted for further studies, with the goal of exploring Titan and Enceladus. The mission concept is to perform in situ investigations of two worlds tied together by location and properties, whose remarkable natures have been partly revealed by the ongoing Cassini– Huygens mission. These bodies still hold mysteries requiring a complete exploration using a variety of vehicles and instruments. TandEM is an ambitious mission because its targets are two of the most exciting and challenging bodies in the Solar System. It is designed to build on but exceed the scientific and technological accomplishments of the Cassini–Huygens mission, exploring Titan and Enceladus in ways that are not currently possible (full close-up and in situ coverage over long periods of time). In the current mission architecture, TandEM proposes to deliver two medium-sized spacecraft to the Saturnian system. One spacecraft would be an orbiter with a large host of instruments which would perform several Enceladus flybys and deliver penetrators to its surface before going into a dedicated orbit around Titan alone, while the S. K. Atreya · M. Hirtzig Department of Atmospheric, Oceanic, and Space Sciences, University of Michigan, Ann Arbor, USA T. Balint · K. H. Baines · T. Spilker · R. M. Lopes · G. Orton · K. Reh · L. J. Spilker · R. A. West JPL, Caltech, Pasadena, USA R. H. Brown · C. A. Griffith · J. I. Lunine · P. Lavvas · C. D. Neish · M. G. Tomasko · V. Vuitton · R. Yelle LPL, University of Arizona, Tucson, AZ, USA M. K. Dougherty · I. Müller-Wodarg · C. Bertucci · C. Dunford · M. Galand Imperial College, London, UK F. Ferri CISAS, Università di Padova, Padua, Italy R. A. Gowen · A. J. Coates · N. Achilleos · A. D. Fortes Mullard Space Science Laboratory, University College London, Bloomsbury, UK L. I. Gurvits · S. Pogrebenko Joint Institute for VLBI in Europe, Dwingeloo, The Netherlands R. Jaumann · F. Sohl · M. Grott · H. Hussmann · M. Knapmeyer · L. Richter · N. Schmitz · K. Stephan German Aerospace Center (DLR), Institute of Planetary Research, Berlin, Germany Y. Langevin IAS, Université Paris-Sud, Orsay, France M. R. Leese · A. Morse · J. C. Zarnecki Open University, Milton Keynes, UK C. P. McKay NASA/AMES, Palo Alto, CA, USA

Exp Astron

other spacecraft would carry the Titan in situ investigation components, i.e. a hot-air balloon (Montgolfière) and possibly several landing probes to be delivered through the atmosphere.

Keywords TandEM · Titan · Enceladus · Saturnian system · Landing probes

1 Context TandEM was proposed in response to ESA’s Cosmic Vision 2015–2025 Call.

The proposal was constructed around an L-class mission to explore in situ Titan and Enceladus, two of the most interesting bodies in the Saturnian system, indeed in the Solar System. Titan’s dense and optically thick atmosphere has been penetrated by the Cassini–Huygens mission (http://saturn.jpl.nasa. gov/mission/index.cfm) to reveal a complex world with diverse geophysical and

X. Moussas Faculty of Physics, University of Athens, Athens, Greece F. Neubauer · T. Tokano · J. Saur · M. Paetzold University of Cologne, Cologne, Germany T. C. Owen Institute for Astronomy, University of Hawaii, Honolulu, USA F. Raulin · Y. Bénilan · P. Coll LISA, Universités Paris 12 and Paris 7, Créteil, France E. C. Sittler · G. L. Bjoraker · J. F. Cooper · D. Fairbrother · F. M. Flasar · T. Kostiuk · P. R. Mahaffy · C. A. Nixon · P. Romani NASA/Goddard Space Flight Center, Greenbelt, MD, 20771, USA C. Sotin · G. Tobie · O. Grasset · A. Herique · S. Le Mouélic Lab. de Planétologie et de Géodynamique, Faculté des Sciences, Nantes, France E. P. Turtle · T. Krimigis · R. D. Lorenz · D. G. Mitchell · D. F. Strobel The Johns Hopkins University Applied Physics Laboratory, Laurel, MD, 20723, USA J.-E. Wahlund · R. Modolo Swedish Institute for Space Physics, Uppsala, Sweden J. H. Waite · F. Crary · E. F. Young Southwest Research Institute, San Antonio, USA J. Blamont · L. Boireau Centre National d’Etudes Spatiales, Paris, France I. Dandouras · D. Toublanc Centre d’Etude Spatiale des Rayonnements, Toulouse, France T. Krimigis Academy of Athens, Athens, Greece

Exp Astron

atmospheric processes reminiscent of those on Earth, but operating under very different conditions from our home world. Cassini and Huygens will leave us with an outstanding list of questions about Titan that are directly relevant to our understanding of the nature of planetary evolution, planetary processes, and planetary habitability. Cassini discovered that Enceladus is an active world with plumes spewing water, nitrogen and organic molecules into space; how such a tiny moon could be so active and whether the source of the activity is an organic-bearing liquid water region under the surface are questions that will require direct penetration through the surface for their answers.

The questions that will remain to be answered at the close of the Cassini– Huygens mission have led this Consortium of Titan and Enceladus experts to propose a new mission to the Saturnian system, and to outline a technology development path for achieving Technology Readiness Level (TRL) 6 and for enhancing and enabling technologies so that such a mission can go forward in the Cosmic Vision timeframe. This mission concept addresses directly several

C. C. Porco CICLOPS, SSI, Boulder, CO, USA E. Choi Bombardier Aerospace, Toronto, ON, Canada R. Amils · J. Martinez-Frias · C. Menor Salvan · D. Nna Mvondo · O. Prieto-Ballesteros · M. Ruiz Bermejo Centro de Astrobiologia, CSIC-INTA, Madrid, Spain P. Annan Sensors and Software, Mississauga, Canada D. H. Atkinson NASA, University of Idaho, Moscow, USA M. Blanc Ecole Polytechnique, Paris, France J. Bouman · R. Koop · A. Selig · D. Stam SRON, Netherlands Institute for Space Research, Utrecht, The Netherlands M. Cabane · E. Chassefière · G. Durry · C. Szopa Université Pierre et Marie Curie-Paris6, Service d’Aéronomie, Paris, France M. T. Capria · A. Coradini · E. de Angelis · A. Milillo Istituto Nazionale di Astrofisica, Rome, Italy T. Cravens University of Kansas, Lawrence, USA I. A. Daglis National Observatory of Athens, Athens, Greece I. de Pater · S. Ledvina Astronomy Department, University of California at Berkeley, Berkeley, USA

Exp Astron

of the scientific questions highlighted in the ESA Cosmic Vision 2015–2025 call, particularly: 1.3 “Life and habitability in the Solar System” and 2.2 “The giant planets and their environments”, but also 2.1 “From the Sun to the edge of the Solar System”.

TandEM is a multi-component exploration system to be built with international partners for a launch around 2018 or later. The mission concept has been developed through the expertise of 155 scientists and engineers from all over the world, many of whom are actively involved in the exploration of the Saturn system through Cassini–Huygens and/or have participated in mission designs for existing and future missions. Details on the mission can be found on http://www.lesia.obspm.fr/cosmicvision/tandem/index.php.

2 Introduction Titan and Enceladus as revealed by the Cassini–Huygens mission are enigmatic objects, introducing extraordinary challenges for geologists,

O. Dutuit · W. Kofman · B. Schmitt · R. Thissen Lab. Planétologie Grenoble, Univ. J. Fourier, CNRS, Grenoble, France R. Frampton Boeing, Chicago, USA M. Fujimoto · M. Nakamura Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency (JAXA), Chôfu, Japan T. Haltigin · W. Pollard · H. Vali McGill University, Montreal, Canada F. Hersant Univ. de Bordeaux, Bordeaux, France W. Ip Institute of Astronomy, National Central University, Jhongli City, Taiwan R. Johnson University of Virginia, Charlottesville, USA E. Kallio Finnish Metereological Institute, Helsinki, Finland S. Kempf · N. Krupp · M. Küppers · U. Mall Max Planck Institute, Lindau, Germany H. Lammer · K. Schwingenschuh Space Research Institute, Graz, Austria L.-M. Lara · J.-J. Lopez-Moreno · R. Rodrigo Instituto de Astrofísica de Andalucía, CSIC, Granada, Spain

Exp Astron

astrobiologists, organic chemists, and planetologists; dreamworlds for those who yearn to explore mysterious new realms.

Cassini has shown that Enceladus possesses active plumes, the sources of which may be pockets of liquid water near the surface, an intriguing possibility for such a small and distant satellite. Little is known today about Enceladus, and after several flybys during Cassini’s life we shall still lack precise knowledge on how this body’s interior works or how it affects its surrounding environment (magnetosphere, rings, satellites), a serious drawback considering that we know that Enceladus is the primary source of mass loading of Saturn’s magnetosphere (like Io for Jupiter).

Titan, on the other hand, Saturn’s largest satellite, is unique in the Solar System with its extensive atmosphere made mostly of N2 , with a column density ten times that of Earth’s, and possessing a rich organic chemistry thanks to abundant methane. Titan’s atmosphere is not in chemical equilibrium. It is a chemical factory initiating the formation of complex positive and negative ions

P. Lavvas · I. Vardavas FORTH, University of Crete, Heraklion, Greece S. Lebonnois Labarotoire de Météorologie Dynamique, Paris, France J. Li LASG, IAP, Chinese Academy of Sciences, Taipei, China T. A. Livengood USRA National Center for Earth and Space Science Education, Columbia, USA D. Luz CAAUL-Observatório Astronómico de Lisboa, Tapada da Ajuda, 1349-018, Lisboa, Portugal B. Marty CRPG, Nançy, France T. McCord Bear Fight Center, Winthrop, WA, USA O. Mousis Institut UTINAM, Université de Franche-Comté, CNRS/INSU, Besançon, France C. A. Nixon Department of Astronomy, Univesity of Maryland, College Park, MD, 20742, USA J. Pitman Lockheed Martin Sensing and Exploration Systems, Denver, CO, USA P. Rannou Univ. Reims Champagne-Ardennes, Reims, France F. T. Robb University of Maryland Biotechnology Institute, Baltimore, MD, 21202, USA

Exp Astron

in the high thermosphere as a consequence of magnetospheric–ionospheric– atmospheric interactions involving solar EUV, UV radiation, energetic ions and electrons. The second most abundant constituent, methane, is dissociated irreversibly to produce hydrocarbons (e.g. C2 H6 and C2 H2 ) and nitriles, (e.g.

HCN), from the coupled nitrogen chemistry. The energetic chemistry produces large molecules like benzene, naphthalene, etc., which begin to condense out at ∼950 km, are detectable in solar and stellar UV occultations, and initiate the process of haze formation. As the haze particles fall through the atmosphere and grow, they become detectable with imaging systems such as the Cassini ISS at ∼500 km altitude and are ubiquitous throughout the stratosphere. They are strong absorbers of solar UV and visible radiation and play a fundamental role in heating Titan’s stratosphere and driving wind systems in the middle atmosphere, much as ozone does in the Earth’s middle atmosphere. Eventually, these complex organic molecules are deposited on Titan’s surface in large quantities, where data from Cassini’s instruments hint at their existence. Hence the upper thermosphere is linked intimately with the surface and the intervening atmosphere.

On Titan, methane can exist as a gas, liquid and solid. Playing a role similar to that of water on the Earth, methane is cycled between the atmosphere, surface and the interior. Cloud systems, the size of terrestrial hurricanes (∼up to 1,000 km across), appear occasionally, while smaller transient systems form on a daily basis. The smaller clouds dissipate quickly suggesting the presence of precipitation, carving out the fluvial features that cover much of the equatorial landscape and are common also in the vicinity of lakes seen in the northern polar regions. Near-IR observations indicate that clouds exist mainly south of 60◦ S, and in a band at 40◦ S latitude. Titan’s cloud coverage appears to be less than that of the Earth, but highly variable. Titan’s atmospheric methane may

S. Rodriguez AIM/CEA, Saclay, France E. T. Sarris Demokritos University of Thrace, Xanthi, Greece P. Schenk Lunar and Planetary Institute, Houston, USA D. Schulze-Makuch School of Earth and Environmental Sciences, Washington State University, Pullman, USA L. Soderblom US Geological Survey, Tucson, AZ, USA A. Steele Carnegie Inst, Washington, DC, USA K. Szego KFKI Research Institute for Particle and Nuclear Physics, Budapest, Hungary

Exp Astron

be supplemented by high latitude lakes and seas of methane and ethane [41], which over time cycle methane back into the atmosphere where it rains out, creating fluvial erosion over a wide range of latitudes.

After several close Cassini flybys, in January 2005, the Huygens Probe became the first human artifact to descend through Titan’s atmosphere, reach the surface and return several hours of data from an exotic landscape cut by channels and apparently soaked, near the surface, with methane, ethane and other hydrocarbons (http://www.esa.int/SPECIALS/Cassini-Huygens/index. html). Several years of flybys by the Cassini orbiter have led to radar and near-IR maps that suggest surface liquids poleward of the 70◦ latitude. Dunes made of frozen organic fine grains extend for thousands of kilometres. In the absence of a massive surface ocean, but with analogues to all other terrestrial hydrological phenomena present, Titan’s methane cycle is indeed exotic.

The N2 –CH4 byproducts in Titan’s atmosphere eventually end up as sediments on the surface, where they accumulate presently at a rate of roughly 0.5 km in 4.5 Gyr. Since no large source was detected by Cassini to resupply methane, cryovolcanic outgassing has been hypothesized [42], yet over what timescales and through which internal processes are unknown. Cassini– Huygens also found that the balance of geologic processes—impacts, tectonics, fluvial, aeolian—is somewhat similar to the Earth’s, more so than for Venus or Mars. Titan may well be the best analogue to an active terrestrial planet in the sense of our home planet, albeit with different working materials.

The Cassini–Huygens mission is a remarkable success, answering many outstanding questions about the Saturn system and Titan in particular. As for many successful missions, the key contributions of Cassini may be the questions raised rather than those answered. An important limitation of Cassini, with respect to Titan science, is the insufficient spatial coverage allowed by its orbit around Saturn. While measurements have highlighted the complexity of Titan’s atmosphere and magnetic environment, the coverage has been insufficient to achieve a full understanding. The minimum possible flyby altitude of 950 km and the uneven latitudinal coverage have limited our ability to explore the full set of atmospheric chemical processes. Opportunities for occultations have been rare, thus gaps remain in the atmospheric structure when in the magnetospheric downstream region. The single vertical profile of the atmosphere taken by Huygens limits our understanding of horizontal transport and latitudinal variations.

The surface of Titan, as revealed by the Huygens probe and the Cassini orbiter, offers us an opportunity to stretch our current models in an effort to explain the presence of dunes, rivers, lakes, cryovolcanoes and mountains in a world where the rocks are composed of water ice rather than silicates and the liquid is methane or ethane rather than liquid water, but the limited spatial coverage of high resolution imaging (corresponding to 25–30% with RADAR and much less with VIMS) limits our view of the range of coupled geologic, geochemical or energetic processes still going on Titan’s surface and the interior. The exciting results from the Huygens post-landing measurements

Exp Astron

are limited to a fixed site, short timescales, and do not allow for direct subsurface access and sampling.

The two major themes in Titan exploration—the methane cycle as an analogue to the terrestrial hydrological cycle and the complex chemical transformations of organic molecules in the atmosphere and the surface—make Titan a very high priority if we are to understand how volatile-rich worlds evolve and how organic chemistry and planetary evolution interact on large spatial and temporal scales. Both are of keen interest to planetology and astrobiology.

The intriguing discoveries of geological activity, excess warmth and outgassing on Enceladus (due perhaps to the ejection of water and organics from subsurface pockets bathed in heat, or by some other mechanism), mandate a follow-up investigation to that tiny Saturnian world that can only be achieved with high resolution remote observations, and detailed in situ investigations of the near-surface south polar environment.

TandEM will investigate the science goals described in Section 3, with the mission scenario and the payload detailed in Sections 4, 5 and 6. Some outreach possibilities are given in Section 7.

3 Scientific objectives The primary science goals of this mission are to understand Titan’s and Enceladus’ atmospheres, surfaces and interiors, to determine the pre- and proto-biotic chemistry that may be occurring on both objects, and to derive constraints on the satellites’ origins and evolution, both individually and in the context of the complex Saturnian system as a whole (Table 1).

To achieve these goals, we need a mission that can visit both satellites and (a) study the atmosphere of Titan (including the “agnostosphere” a gap in Titan’s atmosphere between 400 and 950 km altitude that will remain unexplored after Cassini), (b) image in detail, access, sample and analyze Table 1 Characteristics of Titan and Enceladus Characteristic

Titan

Enceladus

Distance from Saturn Period (days) Orbit inclination (◦ ) Excentricity of orbit Mass (1022 kg) Radius (km) Density (kg/m3 ) Gravity acceleration (m/s2 ) Escape velocity (km/s) Geometric albedo Temperature at surface (K) Pressure at surface (bar) Main atmospheric components

20.25R Sat 15.95 0.28 0.029 13.5 2,575 1,880 1.35 2.64 0.2 94 1.5 N2 , CH4

3.95R Sat 1.37 0.009 0.0047 0.011 252 1,606 0.12 0.235 1.0 114–157 10−10 − 10−13 H2 O, N2 /CO, CH4 , CO2

Exp Astron

the various surface features (such as the hydrocarbon lakes, the dunes, river systems, impact craters, mountain ranges and volcanoes), aerosol deposits on the surface and any upwelled subsurface material, (c) determine whether Titan has a sub-surface liquid ocean, and, (d) investigate the atmosphere and penetrate or sample the surface of Enceladus.

In the following, we discuss the TandEM objectives for understanding Titan and Enceladus, their relationship and connection to the origin and evolution of the Solar System and their astrobiological implications.

3.1 Titan as a system The Earth is studied as a complex system of coupled parts. And so is Titan.

More than any other natural satellite in the Solar System, Titan exhibits somewhat similar coupling between magnetosphere, upper atmosphere and ionosphere, lower atmosphere, surface and the interior as the Earth. Indeed, because Titan possesses a cycle of condensation and evaporation of liquids from the surface, it is more like the Earth in this regard than present-day Venus or Mars.

The unifying working fluid in the terrestrial system is water. Water is present in the Earthâ&#x20AC;&#x2122;s mantle and crust, modifying the rheology and chemistry of rock in profound ways. Plate tectonics recycles water in the crust back into the atmosphere and oceans. Water passes through liquid, solid and gas phases as a function of position and time, setting the basic climate state of the Earth through clouds and vapour and, via rainfall and ice, carving the landscape and altering climate. Water is photolyzed in the upper atmosphere, creating products that absorb and emit solar energy. Prior to the origin of life, Earthâ&#x20AC;&#x2122;s atmospheric oxygen budget was driven by water photochemistry.

Cassini and Huygens investigations have suggested a role for methane on Titan equally profound as that of water on the Earth. Methane is likely present in the interior, dissolved in a liquid mantle and trapped in ice as clathrate hydrate, affecting the mechanical, thermal and chemical properties of the ice crust. Methane may be injected from the deep crust via volcanism and geysering, some of which could be explosive. These processes may be facilitated by the presence of a liquid water mantle under the crust. Surface (and potentially subsurface) reservoirs of liquid methane at high latitudes in the north, and likely in the south, affect the global methane cycle in ways that have yet to be fully understood. It is thought that on century to millennium timescales, convective downpours at mid-latitudes carve channels, erode the landscape, and move organic particulates into the lowlands. Methane in Titanâ&#x20AC;&#x2122;s atmosphere forms clouds and helps drive the basic radiative balance of the dense atmosphere. High in the upper atmosphere of Titan, methane is broken apart by ultraviolet light and energetic particles and undergoes chemical processes, together with nitrogen, to form a host of complex organic molecules and dense aerosol layers. These compounds and aerosols fall to the surface where the liquids find their way into methane aquifers or into the polar

Exp Astron

lakes and seas, while the solid hydrocarbons and nitriles coat the surface and agglomerate into particles that are blown by the winds into vast fields of dunes.

This beautiful portrait of a planetary body as comprehensively active as the Earth, yet different in its working materials and details, is incomplete and still speculative. Cassini–Huygens has shown us direct and indirect evidence for many of the component parts. But confirmation of key players and how they fit together still eludes us.

We are motivated to ask the following key questions that we expect will require TandEM for their definitive answers, despite additional progress that Cassini and Huygens data may provide: 1. Are the “lakes and seas” filled with methane and ethane, and do they extend to a subcrustal hydrocarbon “methanofer” system over a larger area of Titan? Where is all the ethane? Are these processes affected by a deep water ocean, e.g. through fissures by tidal flexing?

2. What is the composition of the dune particles, and of the bright and dark materials on the surface?

3. Does Titan contain ammonia and to what extent has it been deposited on the surface?

4. What, if any, are the types of cryovolcanism on Titan?

5. What are the origins of the various mountain systems and, in general, what is the crustal history of Titan?

6. What are the magnitude and time variations of an internally generated magnetic field, if there is one, due to induction in an ocean or a weak dynamo?

7. What are the abundances of krypton and xenon, and hence what is the origin of Titan’s methane?

8. What is the role of the essentially unexplored ‘agnostosphere’ in the chemistry and dynamics?

9. How is Titan’s upper atmosphere coupled to the magnetosphere of Saturn?

10. What are the seasonal- and longer-scale dependencies of the distribution of methane among vapour, clouds and rain?

11. What are the temporal and spatial dependencies of the upper and lower atmospheric winds?

3.1.1 Titan’s upper atmosphere and magnetospheric interactions The proposed mission will investigate Titan’s atmosphere, its ionosphere and exosphere as well as its plasma and magnetic field environment. Titan’s upper atmosphere (>400 km) plays a key role for the chemistry and physics of the entire atmosphere. It is the region where solar EUV radiation is absorbed, dissociating molecular nitrogen and methane and initiating a chain of highly complex photochemistry that affects all layers. This is where Titan’s atmosphere interacts with Saturn’s magnetosphere. It is where Titan’s induced magnetosphere forms and energy and particles from Saturn’s magnetosphere

Exp Astron

are deposited, dissociating and ionizing nitrogen and methane, contributing to the complex atmospheric chemistry (http://saturn.jpl.nasa.gov/multimedia/ images/image-details.cfm?imageID=2176). Here, atmospheric and ionospheric escape processes also occur, directly affecting atmospheric evolution. The altitude range from ∼400 to 950 km of the atmosphere is below the reach of Cassini in situ measurements but above the region of intense remote sensing and most Huygens measurements, and can justifiably be called the “agnostosphere”.

Our knowledge of Titan’s upper atmosphere and magnetospheric interaction is due to Voyager (1980), Earth-based and the recent Cassini/Huygens observations, which have given us an unprecedented insight into Titan’s complex upper atmospheric chemistry and its relationship to the magnetosphere.

However, Cassini can only sample the atmosphere in situ down to 950 km altitude, while Huygens obtained a single vertical profile of total atmospheric density from below 1,400 km and a much more detailed set of measurements below ∼140 km to the surface. Further information on Titan’s upper atmosphere was obtained from ground-based and Cassini observations of stellar occultations. The major recent discoveries in Titan’s upper atmosphere include its chemical complexity (∼100 inferred organic molecules), the presence of haze at high altitudes which are at different altitudes from those at which Voyager observed haze layers, strong horizontal variations, vigorous dynamics and intimate coupling of the ionosphere to the magnetic field. The observed heavy negative ions (∼5,000 amu) in Titan’s ionosphere [7] indicate that the ionosphere may provide the seed population for aerosol formation deeper down in the atmosphere. For the exosphere, neutral gas escape rates have been found to be much larger than previously assumed. A neutral hydrogen cloud was detected, extending out to many Titan radii. Observations of energetic neutral atoms (ENA) have been used to study magnetosphere dynamics near Titan. New results have been obtained regarding the incident magnetospheric flow conditions at Titan, the plasma composition outside and inside Titan’s induced magnetosphere and the magnetotail-wake.

The prime objectives for the upper atmosphere/magnetosphere science by TandEM are: 1. Investigate the structure, chemistry, dynamics and energy balance of the largely unexplored agnostosphere between about 400 and 950 km and investigate the coupling of dynamics and chemistry. Understanding this broad transition region is a key component in understanding the entire atmosphere together with its relevance for astrobiology.

2. Investigate in detail the chemistry of Titan’s upper neutral atmosphere and ionosphere, the formation of heavy hydrocarbons, positive and negative ions and aerosols due to ionization and dissociation by EUV and energetic particle radiation. Cassini instruments are not capable of detecting species at high mass and at the accuracy needed. A key suite of measurements involve the precipitated aerosols, starting with their composition, including isotopic ratios. This will require a clever GCMS outfitted with getters

Exp Astron

3.

4.

5.

6.

that can remove various species and concentrate others. It must have the capability of reaching high atomic weights—beyond 150 amu.

Analyze the horizontal structure and dynamics of Titan’s upper atmosphere and investigate the effects on the chemistry.

Investigate the induced magnetosphere of Titan with its unique incident flow conditions. Study the magnetoplasma environment at different positions of Titan around Saturn.

Investigate the coupling of Titan’s upper atmosphere to Saturn’s magnetosphere. Study correlation between the plasma and neutrals in Titan’s corona in order to reveal the mechanisms for atmospheric escape of both neutrals and ion.

Derive the characteristics of a possible internal magnetic field and evaluate the implications in the context of other large icy satellites, e.g., Jupiter’s satellite Ganymede, which has a magnetic field, and Callisto which does not.

These goals are important contributions to themes 2.2 but also 2.1 of ESA’s Cosmic Vision Plan. Our improved understanding of bodies within our Solar System can also be applied to exo-planets.

To achieve these objectives measurements described in Section 5.1 will be carried out. A dedicated Titan orbiter is essential to accomplish these tasks, and some of the required measurements prefer a spinning orbiter element.

3.1.2 Titan’s neutral atmosphere One overarching goal of the TandEM mission is to understand the workings of Titan’s atmosphere, which, similar to Earth, is coupled to the interior (e.g. through outgassing), the surface (e.g. through evaporation and precipitation), and the ionosphere (e.g. through the chemistry).

A fundamental outstanding question about Titan pertains to its circulation, and the role played by the heating and cooling of the globally varying composition of Titan’s atmosphere. Titan’s temperature profile follows similar trends to that of Earth. At the surface, both atmospheres are warmed by greenhouse gases (∼20 K for Titan and ∼35 K for Earth), mainly water (30 K) and CO2 on Earth, and pressure-induced N2 , H2 and CH4 absorption on Titan. The temperature decreases with altitude from the surface up to the tropopause (∼10–15 km on Earth and ∼40 km on Titan). Above the tropopause, the atmosphere heats up due to ozone on Earth and to a balance between warming by haze and methane, and cooling due mainly to haze, C2 H2 , C2 H6 and HCN on Titan. Here the atmospheres differ significantly in structure. Titan’s haze, which enshrouds the moon and limits remote-sensing investigations of Titan’s lower atmosphere and surface, absorbs ∼30% of the incoming solar radiation, acting like an “anti-greenhouse gas”, and diminishing the illumination of the lower atmosphere [25]. Furthermore, the heating and cooling of the atmosphere by haze and other photolysis by-products (e.g. C2 H2 , C2 H6 and HCN) are complicated because their abundances differ dramatically

Exp Astron

with latitude and season [8], due to their entrainment in seasonal winds.

Titan’s atmospheric circulation affects the distribution of gases and haze, and their condensation and deposition on Titan’s surface. In parallel, resultant latitudinal abundance variations affect the circulation. To understand Titan’s circulation, seasonal measurements of the stratospheric composition are needed, which Cassini will sample only from roughly southern summer solstice to northern spring equinox. At the close of the Cassini mission, we will still not know unambiguously the haze densities in winter and summer polar regions, which differ, perhaps by an order of magnitude, from equatorial values.

Another intriguing question concerns the origin of winds in the upper stratosphere which rotate much faster than the surface, coupled with an unexpected layer of low wind speeds in the lower stratosphere as well as several reversals of the wind direction detected by Huygens near the surface.

Elucidating the details of the transport pattern of atmospheric angular momentum is crucial in understanding the mechanism by which super-rotation is maintained. To this purpose, measurements of wind profiles at different places and seasons and a better characterisation of the nature of eddies, waves and tides are needed.

Titan’s methane cycle, the source of methane and the processes that govern its surface and atmospheric distribution, are poorly understood. Despite the irreversible attrition of methane, Huygens measured a largely methane-saturated troposphere [49]. The methane inventory on Titan differs significantly from that of water on Earth. Titan’s atmosphere has ∼5 m of precipitable methane, whereas Earth’s has 2.5 cm of precipitable water vapour. Titan’s known lakes and seas cover ∼1% of the surface, whereas oceans cover 70% of the Earth. It is thus extraordinary that Titan’s lower atmosphere holds as much methane as it does, with a 45% relative humidity in an equatorial environment that geologically is dominated by dunes. While the polar lakes and seas might contain enough methane to periodically saturate the atmosphere, allowing violent rainstorms to carve river channels, they cannot be the ultimate source of the methane.

Perhaps volcanism in recent times has released just enough methane to saturate the atmosphere and leave the surface fairly dry. Or perhaps there is a steady-state source communicating to the surface in a way that has not been identified. Such questions require detailed measurements of the tropospheric methane content (globally) and the composition of the surface. Long radiative and dynamical timescales in the lower atmosphere make answering these questions with a single mission difficult, and hence a return to Titan imperative.

Another open question pertains to the formation of Titan’s clouds. On Titan, only 10% of incident sunlight, compared to Earth’s 60%, reaches Titan’s surface to drive weather. Despite the fact that the radiative time constant (∼140 Earth years) exceeds Titan’s year (29.5 Earth years), large convective storms are observed to develop in Titan’s atmosphere, apparently varying with season [13]. Images of clouds from Earth-based observatories and by Cassini, indicate a peculiar and unexpected pattern: over the approximate decade of observations, clouds have appeared most regularly south of 60◦ S latitude,

Exp Astron

in the North polar region (http://saturn.jpl.nasa.gov/multimedia/images/ image-details.cfm?imageID=2470 ) and in a band around 40◦ S latitude.

Various theories have been proposed to explain their formation, involving atmospheric circulation, surface heating and cryo-volcanism. Yet these models cannot be tested because the structure of the atmosphere (e.g. humidity, winds and temperature profile) and the surface conditions (e.g. its dampness, temperature and evaporation rates) near storms will not be measured by Cassini. In particular, in situ measurements are needed at high latitudes (where clouds and lakes reside) to further understand Titan’s atmospheric engine.

Titan’s neutral atmosphere parallels Earth’s, with a complex meteorology that cycles methane from the surface to the atmosphere, and a circulation pattern and atmospheric dynamics that affects the surface composition and geology (Fig. 1). Cassini, while it has provided the first glimpses into the complexity of the surface and lower atmosphere, will not obtain significant information on the surface humidity, the temperature profile and its variations in the boundary layer (near the surface), the atmospheric winds, the composition of the surface, and the source of atmospheric methane. In addition the seasonal structure (composition and temperature) will be sampled only from northern winter to spring. TandEM will address these omissions by (a) sampling a different season from Cassini or Voyager, (b) remaining operational across a significant fraction of a Saturnian (Titanian) year.

The prime objectives for the lower atmospheric science by TandEM are: 1. Determine the near surface temperature and temperature profile in the polar troposphere.

2. Search for evidence of atmospheric tides in the thermal profile.

3. Map out the meridional circulation of the troposphere, and its change with seasons.

4. Seek evidence for orographic winds and clouds, and winds associated with tropospheric convective clouds.

Fig. 1 Main couplings between Titan’s atmospheric processes

Exp Astron

5. Characterise the structure and seasonal evolution of the polar vortex and its relevance for Titan’s meteorology.

6. Map the seasonal and latitudinal variation in tropospheric methane abundance.

7. Determine the time-dependent physical and chemical properties of tropospheric methane/nitrogen clouds, and possible ethane ground fog (including electrical properties).

8. Search for evidence of methane outgassing on the surface, and methane evaporation from lakes.

9. Determine the nature of polar winter tropospheric cloud and precipitation.

10. Quantify the coupling of the surface and atmosphere in terms of mass and energy balance. Do further chemical reactions among aerosols and atmosphere take place on the surface? In the lakes/seas? On shorelines?

In playas? Is their any surface segregation of these precipitated aerosols, driven by aeolian or fluvial processes?

The objectives that we delineate point to the set of measurements described in Section 5.1.

3.1.3 Titan’s surface The combination of geomorphologic information from high-resolution RADAR coverage (few hundred meters/pixel: http://saturn.jpl.nasa.gov/ multimedia/images/image-details.cfm?imageID=1176 ) with spectroscopic information from VIMS at more moderate resolution (few hundred meters/pixel to a few kilometres/pixel) have addressed the first order geological and chemical processes in a limited geographical area on Titan’s surface.

RADAR and VIMS together with Huygens high-resolution imaging and in situ studies have completed a preliminary survey of the chemical and physical processes in an area of order 100 km2 . Combined ISS, VIMS and RADAR data [38] have provided near-global surface coverage at a variety of wavelengths and resolutions (http://saturn.jpl.nasa.gov/multimedia/images/imagedetails.cfm?imageID=2385).

Together, Cassini and Huygens have established a diversity of geologic features comparable to that on the Earth: •

•

erosional features such as channels and dendritic networks (http://saturn. jpl.nasa.gov/multimedia/images/image-details.cfm?imageID=1326 ), possible lakes and seas, fluvial erosional deltas and other erosional and depositional constructs, possible glacial-flow constructs, as well as widespread aeolian, erosional and depositional features such as dunes: (http://saturn. jpl.nasa.gov/multimedia/images/image-details.cfm?imageID=2260); impacts: the very low crater frequency [21] is indicative of active geological surface processes;

Exp Astron

•

volcano-tectonic features (http://saturn.jpl.nasa.gov/multimedia/images/ image-details.cfm?imageID=1558): domes, possible cryovolcanic flows, and bright spots as well as mountain chains (http://saturn.jpl.nasa.gov/ multimedia/images/image-details.cfm?imageID=2381).

However, detailed knowledge of the global distribution of these features, their possible associations and causal relationships, their ages and the geophysical processes that are responsible for their origin and evolution remain poorly understood, even where Huygens landed [39], http://saturn.jpl.nasa.gov/ multimedia /images/image-details.cfm?imageID = 1881, http://saturn.jpl.nasa. gov/multimedia/images/image-details.cfm?imageID=1310.

The deposition of CH4 –N2 photochemical materials, including liquid ethane, suggest that Titan’s surface will have lakes and seas of liquid hydrocarbons, which may have been found already in the high latitude regions, making a uniquely evocative landscape. At high spatial resolution Titan’s landscape may be quite exotic, because the working materials differ so much from those on the Earth.

The prime objectives remaining for the surface and interior science with TandEM are: 1. Determine the ages and temporal relationships of various geologic units and constructs.

2. Seek the extent, timescale and mechanisms of volcanism and tectonism.

3. Determine the nature of erosion on Titan, its spatial extent and the relative balance of aeolian, fluvial, chemical, and glacial processes. Are there active fluvial channels, carrying liquid and eroding the surface?

4. Determine Titan’s interior structure to a level of accuracy and confidence higher than that of Cassini, to assess the size of any internal ocean, core, and extent of differentiation.

5. Determine the age of the surface, the structure of the crust and mantle, and how the interior has interacted with the surface and atmosphere over time.

6. Determine the composition, distribution and physical state of materials on and beneath Titan’s surface and how are they related to geology. Search for ammonia as an indicator of volcanism, and the presence of a liquid water-ammonia mantle. Does Titan still contain ammonia in its ice? If yes, does it reach the surface by itself or mixed with ice? if so, what is its lifetime? Seek evidence of sulfur, carbon dioxide, and other constituents, key to understanding Titan’s origin and evolution.

7. Determine the source of the atmospheric methane, its outgassing history, and the surface/interior sinks for ethane and other photochemical products.

8. Determine the origin of Titan’s major volatiles, such as methane and nitrogen, and structural materials such as ice and rock through noble gas abundances and isotopic ratios of hydrogen (D/H in H2 O-ice), carbon and nitrogen in abundant organic and inorganic material, sulfur (if present), and at a level much more sensitive than provided by Cassini/Huygens.

Exp Astron

TandEM will have the requisite capabilities to make the measurements described in Section 5.1 that are required to address these fundamental objectives 3.2 Enceladus as a system The discovery by the Cassini spacecraft of a large plume emanating from surface cracks near the south pole of Enceladus [9] venting water vapour (http:// saturn.jpl.nasa.gov /multimedia /images/image -details.cfm? imageID = 1629), fine icy particles (or dust) and organics implies an underground heat source.

Whether the plumes emissions arise directly from chambers of pressurized liquid water, or evaporation from warm ice, or the explosive release of gases from clathrate ices is unknown. The origin and time history of the internal heating source is a major puzzle. Ions formed from the emissions are driven out into the magnetosphere of Saturn, and are implanted into the E ring grains, the other icy satellites, and even into Titan’s atmosphere.

Fine icy particles escaping from the jets (http://saturn.jpl.nasa.gov/ multimedia/images/image-details.cfm?imageID=2027) also drift outward forming the E-ringmake their way into the ring; confirming Enceladus as the main source for this ring. Cassini observations (http://saturn.jpl.nasa.gov/ multimedia/images/image-details.cfm?imageID=2779), including extended mission observations, will allow direct plume sampling as well as further remote sensing observations of the south polar regions and high phase plume imaging. With the TandEM mission we plan to extend both remote sensing and in situ measurements beyond Cassini capabilities, with more advanced instrumentation. We will remotely monitor the satellite and its activity over time, at high spatial resolution and over a range of different wavelengths and also obtain in situ measurements via plume fly-throughs. We also plan to deploy penetrators to the south polar terrain for making direct measurements of particular regions in the geologically active zone.

Our primary science objectives include: • • • • • •

Origin, nature and properties of the plume (including particle dynamics, temporal variability, spatial distribution and production rates of gas/dust, association with local hot spots, etc.)

Existence, depth, extent and composition of sub-surface liquid water (including degree of tidal flexing in the liquid zone, vent structure, liquidwater/clathrates ratios, overall composition, see Section 3.3.2) Signs of past/present life (including organic inventory, molecular chirality, etc. see Section 3.4) Some secondary objectives would be to: Characterization of the surface (including resurfacing processes, geometry of fractures, crater morphology, see Section 3.2.2) Characterization of the interior (including structure and mass distribution, gravity field, global topography, endogenic and exogenic dynamics, see Section 3.3.2)

Exp Astron

•

The impact of Enceladus on the magnetosphere (including magnetospheric processes, plasma loading effects)

In the process of addressing the above objectives we would also discover answers to important unresolved questions such as the: • • •

Influence of Enceladus on other satellites Influence of Enceladus on the ring structure Determination of dust flux into system

Some of the detailed science questions linked to these priorities to be addressed by TandEM are given below, some are addressed in Sections 3.3 and 3.4, and the requisite measurements are given in Section 5.2.

3.2.1 Nature of the plume source, variability and composition Single point measurements from the only close Cassini flyby of Enceladus to date have allowed us to gain a basic understanding of the composition of the plume, incorporating water vapour, dust and organics. In order to ascertain the variability of the plume, which has been suggested by the Cassini magnetometer and energetic particle observations, numerous fly-throughs of the plume separated both temporally and spatially as well as visual temporal monitoring of the plume over a range of phase angles and distances will be required. Understanding specific properties of the plume, its variability and its composition will lead to an understanding of the nature of its source.

Essential objectives include exploration of the: • • • • • • •

Surface distribution of temperatures Radiation environment Dynamic and physical properties of the plume ‘dust’ particles Spatial distribution of the gas and dust and its temporal variability Gas phase composition; noble gas determination, gas/dust determination Gas and particle production as function of time Pick-up ion production from solar UV and magnetospheric irradiation

3.2.2 Characterise the surface of Enceladus Although Cassini/ISS has produced superb images of Enceladus’ surface (http://saturn.jpl.nasa.gov/multimedia/images/image-details.cfm? imageID = 2032), at present, large regions of the surface of Enceladus have still not been imaged at a resolution sufficient to determine precise geological relationships and geologic history in detail. In order to be able to characterize large-scale, small-scale, and dynamic geological features on the surface we require repeated coverage at a variety of latitudes and longitudes at high resolution.

Coverage spanning a variety of temporal scales—from hours, to days, weeks, months and even years—would be especially important for evaluating models for the geologically active south polar region.

Exp Astron

Specific science objectives related to the characterization of surface features include: • • • • • •

Resurfacing processes; viscous relaxation; tectonic overprinting; deposition from plumes Surface ages and composition Vent structure, composition, temperature in the south polar region Craters: inventory, morphology, terrain dating, thermal gradient, spatial variability Magnetospheric irradiation environment Biological potential: organic inventory, molecular chirality, carbon and oxygen isotropic ratios

3.2.3 The impact of Enceladus on its environment We will investigate: • •

• •

How the structure of the E ring is driven by and dependent on plume and outgassing processes from Enceladus.

The magnetospheric processes. Measurements of the composition arrival directions and energy distribution of neutral atoms will yield information about internal composition, erosion of the upper surface, emission rate and the interaction with the magnetosphere.

How does Enceladus populate Saturn’s magnetosphere with neutrals, plasma and dust particles.

The influence on other satellites via implantation of ions to their surfaces and driving many of the processes of Saturn’s magnetosphere and its interactions with the other icy satellites. It may be the dominant source of water and oxygen delivered to Titan’s atmosphere.

The full description of the measurements to be performed in order to achieve the science goals presented here are given in Section 5.2.

3.3 Origin, interior and evolution The dense nitrogen–methane atmosphere of Titan and the spectacular waterrich, gas plumes of Enceladus offer some key challenges: Why should Saturn host such volatile and active worlds? How did they form? What internal processes make Titan and Enceladus so different from the other icy moons?

The fact that Titan has an atmosphere, something that the Galilean satellites do not possess, as well as the Enceladus ejecta furthermore suggest that the former is a sort of end-point for satellite formation around a giant planet, while the latter is a body from which primordial satellite materials are still being emitted. These key elements give us the opportunity to enhance our understanding of ours and other planetary systems (even satellites around giant exoplanets). The Cassini–Huygens mission has started to answer these questions, but key measurements are still missing to draw a complete picture

Exp Astron

of Titan’s and Enceladus’ origin and evolution. In the following sections, we detail the key objectives and measurements to be acquired in the frame of the TandEM mission to address those questions.

3.3.1 Origin of Titan and Enceladus Connecting a model for the solar nebula to a model of giant planet formation that includes the formation of regular satellite systems is quite a difficult problem. Using a 3-D hydro-dynamical model, [22] showed that in the last stage of giant planet formation, a protosatellitary disk of captured material gradually emerges from the contracting atmosphere. Satellite formation could have taken place in this disk. Any successful model for satellite formation must include predictions for the variety of satellite densities and the probable compositions of atmospheres and surfaces. These are fundamental properties that can be deduced from in situ and orbital observations to better constrain satellite formation models. Unfortunately, critical observations of key cosmochemical elements required for the discrimination between competing formation models of Titan and Enceladus are rare or even lacking.

Titan We are far from understanding the origin of Titan. Scenarios may be envisaged for the origin of the planetesimals that formed Titan, based on two end-member type hypotheses: 1. Planetesimals were produced early, just after or even before the end of the formation of Saturn, within a relatively warm subnebula. In this case, only silicates and moderately refractory volatiles could have been incorporated in the planetesimals that formed Titan and the other indigenous regular satellites. It is then difficult to trap the noble gases and CH4 in icy grains but the trapping of H2 O, NH3 , CO2 and non-volatile compounds would be facilitated [1, 5, 30, 31]. The small amount of 36 Ar detected by the Huygens GCMS (36 Ar/N < 1/1000 × Earth’s ratio, [27]) implies an ambient formation temperature of ∼100 K in the sub-nebula [30, 31], under the assumption that Argon was trapped in amorphous ice. It has been proposed that methane on Titan was not initially incorporated, but was produced within the satellite from chemical (catalysed) reactions between carbon compounds such as CO2 and H produced from H2 O by water–rock reactions [5].

2. Alternatively, the primary sub-nebula may have dissipated rapidly (possibly due to turbulence). The planetesimals were then produced in the cold, gas-dominated solar nebula near the end of its existence [16]. This is consistent with observations of circumstellar disks around young stars showing that their temperature dramatically decreases with time in no more than 10 Myr. Like in the former scenario, NH3 and CO2 condensed, while CH4 , H2 S and Xe were trapped in the form of clathrate hydrate, being incorporated into in the migrating planetesimals that eventually formed Titan [15, 16]. The deficiency in water ice at late epochs severely

Exp Astron

limited the trapping of less stable species such as CO and Ar, clathrated at low temperatures (around 30 K), much later than CH4 , thus explaining their low abundances in Titan’s atmosphere. In this scenario, the nondetection of Xe by the Huygens GCMS would be explained by preferential sequestration within Titan’s crust and deep interior [16].

In both of these hypotheses, NH3 was released from the planetesimals during accretion, forming the early atmosphere. Subsequently, the conversion of NH3 into N2 may have been caused by photolysis in Titan’s hot protoatmosphere [4] or impact-induced high temperature chemical processes [24], whereas CH4 should have been outgassed from the interior, as is 40 Ar [27, 47].

Although the process of atmospheric formation would be similar in the two scenarios, the resultant deuterium-to-hydrogen (D/H) ratio in vaporized water ice and methane should be different. The detection of Kr and Xe or sulfur compounds on the surface, which could not be achieved by the Huygens probe, will help discriminate between these two hypotheses.

Enceladus A key question is whether Enceladus formed from primordial bricks similar to those of Titan. In contrast to the Jovian system, which is characterized by a negative gradient in the satellite uncompressed density with distance from Jupiter, the Saturnian system exhibits large variations in mean density and mass of the satellites. So we should expect great variability also in the satellites’ bulk chemical composition. The INMS instrument on board Cassini has found non-condensible volatile species (e.g. N2 , CO2 , CH4 ) in jet-like plumes over Enceladus’ geologically active South Polar Terrain [50].

Some species, like for instance CO2 , might be primordial, whereas others, like N2 , might require a liquid water reservoir in the interior at elevated temperatures, enabling aqueous, catalytic chemical reactions [23]. Taken together, the evidence strongly suggests that Enceladus’ interior is at least partially differentiated into a silicate-metal core overlain by a water ice-liquid water shell [37]. However, the energy source required to initiate and preserve activity at the level comparable to that observed today over geologic periods and the concentration of geologic and thermal activity towards the south-polar region are still not well understood.

Determining the composition of Enceladus’ plumes will provide strong constraints on the origin of volatiles on Enceladus, and more generally in the entire Saturnian system. Even though the two satellites formed from satellitesimals with slightly different compositions, the detection of CO2 emanating from Enceladus’ interior suggests that CO2 may also be present within Titan’s interior. Comparison between Enceladus and Titan’s composition will help constrain their origin and also bring some insight on the origin of the Galilean satellites.

Measurements that are required to better constrain the formation scenario of Titan and Enceladus are delineated in Section 5.3.1. A key measurement in this is the value of D/H in Titan’s ice. It can tell us the source of the ice, meaning the process(es) that formed it, hence conditions in Saturn’s subnebula. It will

Exp Astron

also help us determine the source of Titan’s methane—whether captured by ice forming in the subnebula, the solar nebula, cometary impacts, or produced in situ.

3.3.2 Interior and early evolution Many internal processes play crucial roles in the evolution of Titan and Enceladus. The formation and replenishing of Titan’s atmosphere and the jet activity at Enceladus’ South Pole are intimately linked to the satellite’s interior structures and dynamics. The primary objectives of the TandEM mission regarding Titan and Enceladus’ interiors will be (1) to determine their presentday structures and levels of activity (2) to determine whether the satellites underwent significant tidal deformation, and whether they possess intrinsic or induced magnetic fields and significant seismicity, (3) to identify heat sources, internal reservoirs of volatiles (in particular methane and ammonia) and eruptive processes.

Present-day interior structure: rocky core and liquid water/ice shells The Radio Science Subsystem on board Cassini, by measuring the principal components of Titan’s and Enceladus gravitational potential [34, 35], will provide important constraints on the satellites’ internal differentiation. However, the sizes of their cores and the thicknesses of their ice mantles will remain uncertain [40]. Determining the present-day structure of the outer ice mantle and of the innermost core is crucial to determining how a satellite’s interior has differentiated over time and the effects on the evolution of the surface and atmosphere (http://www.sciences.univ-nantes.fr/geol/UMR6112/Nature% 20Titan%20outgassing_short_english.htm).

Therefore, the next step will be to characterize the sizes and states of the rock-metal cores, and the structure of the outer H2 O layers. On Titan, joint measurements of large-scale and meso-scale topography and gravitational field anomalies from an orbiter and from an aerial platform would impose important constraints on the thickness of the lithosphere, the presence of mass anomalies at depth and any lateral variation of the ice mantle thickness. On Enceladus, altimetric and gravimetric profiles performed during close flybys (<100 km) over the South Pole will be used to retrieve the internal structure below the active region (depth of liquid reservoirs, lithosphere thickness, existence of thermal plumes in the ice mantle, in the rocky core) (e.g. [32]).

These data will provide fundamental constraints on the origin of the South Pole Hotspot on Enceladus.

Tidally-induced deformation, magnetic field and seismicity The Radio Science measurements performed by Cassini at Titan should be a priori capable of detecting any tidal fluctuations of the gravitational potential [34]. The detection of amplified periodic signals would suggest the existence of a liquid layer at depth, most likely composed of liquid water [40, 46]. However, even if these tidal fluctuations were detected, they would not provide firm constraints

Exp Astron

on important structural parameters such as the depth and radial extent of the liquid water reservoir, the thickness of the overlaying icy shell, and any lateral variations. The TandEM mission will provide such constraints by monitoring the tidally induced fluctuation of the gravity field and topography from an orbiter and from the ground.

Seismicity on both Titan and Enceladus may be associated with tidal fluctuations, as on the Moon. Various sources such as tidally induced fracturing events, cryovolcano-tectonic activity, global contraction, trapped surface waves, natural impacts etc. may generate seismic events on Titan and Enceladus. The detection of seismic activity and localization of seismic sources would require the emplacement of seismometers by surface probes (penetrators and/or landers) that could acquire data over a sufficiently long period of time (at least a few tidal cycles: one tidal cycle ∼16 days for Titan, 1.37 days for Enceladus).

Monitoring tidally induced gravity changes and seismic activity, in addition to confirming whether the satellite is tectonically active, would provide an opportunity to sound its deep interior and help characterize any threedimensional internal heterogeneity. The detection and characterization of induced magnetic fields from Titan’s and Enceladus’ interiors in addition to tidally-induced topographic, gravimetric and seismic signatures would provide evidence for the presence of internal liquid water layers, and place constraints on depth and composition in terms of electrolyte content. The detection of an intrinsic magnetic field on Titan (which must be quite weak based on upper-limits imposed by Cassini data [6]) would also imply operation of a selfsustained magnetic dynamo, providing important constraints on the nature of an iron-rich core.

Heat sources, cryovolcanism and eruptive processes Even though there are indications that internal outgassing has occurred on Titan and is currently occurring on Enceladus, the sources of energy and the eruptive mechanisms still remain poorly constrained [14, 29, 33, 43, 48]. Possible venting mechanisms are sudden decompression of near-surface reservoirs of liquid water [33], clathrate decomposition [20], or other cryovolcanic processes [12].

Measuring the near-surface thermal gradients and mapping the surface thermal emission on both bodies will provide key constraints on the heat flow through the icy crust. These data, combined with estimates of the ice shell thickness and tidal fluctuations, and supplemented with high resolution imaging of the icy particle jets and plume-will help determine the origin and locales of energy dissipation on Enceladus and the heat transfer mechanism [33, 43].

Temporal monitoring of the icy particle jets over time through repeated high resolution imaging will ascertain if the jetting activity is associated with tidal flexing [17, 44].

Knowledge of the composition, particularly the amount of highly volatile ices, such as ammonia, will help in understanding Enceladus’ thermal history and how and when differentiation took place (http://saturn.jpl.nasa. gov/multimedia/images/image-details.cfm?imageID=2148). On both Titan and

Exp Astron

Enceladus, sampling of surface materials and eruptive gases will provide information on the composition of cryovolcanic materials. These data as well as remote sensing visual, infrared and radar data, will be used to assess the depth and composition of the cryomagmatic reservoir. In the case of Titan, ammonia, methane and carbon dioxide mixing ratios will help distinguish between different hypotheses about replenishing volatiles in its atmosphere [5, 12, 47]. Subsurface sounding and gravity measurements will also be used to identify cryomagmatic chambers below cryovolcanic edifices.

Early evolution: crust and atmosphere formation On Titan, the formation of the crust and of the atmosphere is intimately linked to the differentiation of the interior, which led to the formation of a discrete rock-metal core, and the crystallization of the post-accretional ammoniaâ&#x20AC;&#x201C;water ocean [47]. A signature of the internal differentiation can be searched for in the isotopic ratios of noble gases (Ar, Kr, Xe, Ne) in the atmosphere and in surface materials (icy materials, deposited aerosols, hydrocarbon mixtures). On both Titan and Enceladus the abundance of radiogenically derived noble gas isotopes in the atmosphere and the gaseous plumes, respectively, will constrain the chemical exchange between the rock-metal core and the outer H2 O layers.

This will constrain the likelihood of silicate volcanism linked to the core and waterâ&#x20AC;&#x201C;rock interactions during early epochs. Stable isotope ratios of actively venting volatiles bearing carbon, nitrogen and oxygen are also important in constraining the nature of volatile replenishment and atmospheric evolution.

Induction studies from combined in situ and orbital magnetometric measurements will provide information on the electric conductivity of the ocean combined with a range of depths. This will constrain the concentration of electrolytes in a current internal ocean and hence will provide some information on the early history of leaching through waterâ&#x20AC;&#x201C;rock interactions [10, 12].

Subsurface sounding (e.g. via ground penetrating radar, seismic monitoring) in the oldest terrains (identified by impact crater statistics) may also reveal crustal layering remnant from early crust formation and subsequent recycling processes. Compositions of materials excavated by impacts will also constrain the formation process of the crust.

Measurements that are required to better constrain the deep interiors of Titan and Enceladus are delineated in Section 5.3.2.

3.4 Astrobiological potential of Titan and Enceladus The Saturnian System is a key target for astrobiology. The key components of life, liquid water and organic material, are all present on moons of Saturn.

Titan has an environment very rich in organics and it is often considered as one of the best targets to study prebiotic chemistry at a full planetary scale.

Studies of the synthesis of organic molecules and solids on Titan may help us understand how these processes occurred on the early Earth. With the possibility of a liquid water subsurface (and the melt sheets produced by large impacts), Titan is even considered as a possible habitat for life [11, 26, 36].

Exp Astron

The recent discovery on Enceladus of water ice plumes with methane and the possible presence of a large internal reservoir of liquid water containing active organic chemistry also make this small Saturnian satellite a new important planetary target for astrobiology.

3.4.1 Similarities of Titan with the early earth Retracing the processes that allowed the emergence of life on Earth around four billion years ago is a difficult challenge. Our planet has drastically evolved since then, and most traces of the environmental conditions at that time have been erased. It is thus crucial for astrobiologists to find extraterrestrial locales with similarities to our planet. Such environments will provide a way to study some of the processes that occurred on the primitive Earth, when prebiotic chemistry was active.

Although Titan is much colder than the Earth, it does present many similarities with our planet (http://saturn.jpl.nasa.gov/multimedia/images/imagedetails.cfm?imageID=1105). Titan’s atmosphere is made of the same main constituent, dinitrogen. It also has a similar structure from the troposphere to the ionosphere, and a surface pressure of 1.5 bar—the only case of an extraterrestrial planetary atmospheric pressure close to that of Earth. As noted in the introduction, many analogies can also be made between the role of methane on Titan and that of water on the Earth. Methane on Titan seems to play the role of water on the Earth, with a complex cycle that has yet to be fully understood.

Analogies can also be made between the current organic chemistry on Titan and the prebiotic chemistry that was active on the primitive Earth. In spite of the absence of permanent bodies of liquid water on Titan’s surface, the chemistry is similar. Moreover, Titan is the only planetary body, other than the Earth with long-standing bodies of liquid on its surface (although direct observational evidence of the longevity of Titan’s surface liquids remains to be obtained). Several of the organic processes which are occurring today on Titan form organic compounds which are considered as key molecules in terrestrial prebiotic chemistry, such as hydrogen cyanide (HCN), cyanoacetylene (HC3 N) and cyanogen (C2 N2 ). In fact, with several percent of methane in dinitrogen, the atmosphere of Titan is one of the most favourable atmospheres for prebiotic synthesis. Until recently, such an atmospheric composition was thought to be different from that of the primitive Earth. However, new modelling of the hydrogen escape in the Earth’s primitive atmosphere indicates that it may have been much richer in hydrogen and methane than previously thought. This result suggests that Titan may be even more similar to the primitive Earth than expected [45].

The degree of complexity that can be reached from organic chemistry in the absence of permanent liquid water bodies on Titan’s surface is still unknown, but it could be quite high.

Exp Astron

3.4.2 A complex prebiotic-like chemistry In the atmosphere of Titan, CH4 chemistry is coupled with N2 chemistry.

This combination should produce many organics in the gas and particulate phase: in particular, hydrocarbons, nitriles and complex refractory organics, The latter seem to be well modeled by the solid products—usually named “tholins”—formed in laboratory experiments ([28]; and refs included therein).

We also need to consider the addition of water and oxygen ions into Titan’s atmosphere. These come from a magnetospheric source that can be traced to Enceladus’ plumes, and they participate in the chemical processes, forming in particular CO and CO2 . Could these water–oxygen compounds then be locked up into aerosols?

Several organic compounds have already been detected in Titan’s stratosphere. The list includes hydrocarbons (both with saturated and unsaturated chains) and nitrogen-containing organic compounds, exclusively nitriles, as expected from laboratory experiments. Since Cassini’s arrival in the Saturnian system, the presence of water vapour and benzene has been unambiguously confirmed by the CIRS instrument (first detected by ISO).

Surprisingly, GCMS on board Huygens did not detect a large variety of organic compounds in the low atmosphere. The mass spectra collected during the descent show that the medium-altitude and low stratosphere and the troposphere are poor in volatile organic species, with the exception of methane [27].

Condensation of such species on the aerosol particles is a probable explanation for these atmospheric characteristics. These particles, for which no direct data regarding chemical composition were previously available, were analyzed by the Huygens ACP instrument. ACP results show that the aerosol particles are made of refractory organics which release HCN and NH3 during pyrolysis. This supports the tholin hypothesis. From these new in situ measurements it seems very likely that the aerosol particles are made of a refractory organic nucleus, covered with condensed volatile compounds [18].

However, the nature and abundances of the condensates have not been measured. Even more importantly for astrobiology, neither the elemental composition nor the molecular structure of the refractory part of the aerosols has been determined. The potential chirality of its complex organic part is unknown.

The direct analysis of the ionosphere by the INMS instrument during the closest Cassini flybys of Titan shows the presence of many organic species, in spite of the very high altitude (1,100–1,300 km) [51]. Extrapolation of the INMS measurements (limited to mass up to 100 Da) and of CAPS data, strongly suggests that high-molecular-weight species (up to several 1,000 Da) may be present in the ionosphere. These new data—if confirmed— revolutionize the understanding of the organic processes occurring in Titan’s atmosphere, with a strong implication that ionospheric chemistry plays a role in the formation of complex organic compounds in Titan’s environment, which was predicted [3] but whose extent was not envisioned before [2, 51].

Exp Astron

It is thus essential to determine the ion and neutral composition of the ionosphere with a mass range and resolution allowing the detection and characterisation of a very wide range of compounds.

In the current picture of Titan’s organic chemistry, the chemical evolution of the main atmospheric constituents—dinitrogen and methane—produces mainly ethane, which accumulates on the surface or the near subsurface, and is eventually dissolved to form methane–ethane lakes and seas, and complex refractory organics which accumulate on the surface, together with condensed volatile organic compounds such as HCN and benzene. In spite of the low temperature, Titan is not a congealed Earth: the chemical system is not frozen.

Titan is an evolving planetary body and so is its chemistry. Once deposited on Titan’s surface, the aerosols and their complex organic content may follow a chemical evolution of astrobiological interest. Laboratory experiments show that, once in contact with liquid water, Titan tholins can release many compounds of biological interest, such as amino acids [19]. Such processes could be particularly favourable in regions on Titan’s surface where cryovolcanism is occurring. Thus one can envision the possible presence of such compounds on Titan’s surface or near subsurface. Long-term chemical evolution is impossible to study in the laboratory: in situ measurement of Titan’s surface thus offers a unique opportunity to study some of the many processes which could have been involved in prebiotic chemistry, including isotopic and enantiomeric fractionation [28].

It is crucial to be able to perform a detailed chemical analysis (at the elemental, molecular, isotopic and chiral levels) of the various types of surface zones, particularly those where cryovolcanism and impact ejecta (or melt sheets) are present.

3.4.3 Potential habitats All ingredients that are supposed to be necessary for life to appear and develop—liquid water, organic matter and energy—seem to be present on Titan. Indeed, interior structure models suggest that Titan, as well as Europa, Ganymede and Callisto, have maintained internal liquid water reservoirs (probably mixed with some ammonia and more speculatively sulfur). At the beginning of Titan’s history, this hypothetical subsurface ocean may have been in direct contact with the atmosphere and with the internal bedrock, offering interesting analogies with the primitive Earth, and the potential implication of hydrothermal vents in terrestrial prebiotic chemistry. Consequently, it cannot be excluded that life may have emerged on or in Titan. In spite of the extreme conditions in this environment, life may have been able to adapt and to persist. Even the possible current conditions (pH, temperature, pressure, salt concentrations) are not incompatible with life as we know it on Earth [11].

However, the detection of a potential biological activity in the putative liquid mantle would be very challenging.

Nevertheless it seems astrobiologically essential to confirm the presence of such an internal ocean.

Exp Astron

3.4.4 Enceladus The jets emanating from the South Pole of Enceladus are probably the most accessible samples from a liquid water environment in the outer Solar System. In addition to water ice the jets include methane, propane, acetylene, and N2 or CO. The likely source of Enceladus’ jets is a pressurized subsurface liquid reservoir (http://saturn.jpl.nasa.gov/multimedia/images/ image-details.cfm?imageID=2026). If N2 is present it may reflect thermal decomposition of ammonia associated with the subsurface liquid reservoir and may imply that the water is in contact with hot rocks—providing a source of heat as well as mineral surfaces for catalyzing reactions. If this scenario proves correct, then all the ingredients are present on Enceladus for the origin of life by chemoautotrophic pathways—a generally held model for the origin of life on Earth in deep sea vents. In this model, life on Earth began in deep sea hot springs where chemical energy was available from a mix of H2 , S and Fe compounds. The fact that the branches of the tree of life that are closest to the common ancestor are thermophilic has been used to argue that this reflects a thermophilic origin of life—although other explanations are possible. In the material emanating from Enceladus we can search for the specific molecules associated with such systems, including H2 H2 S, FeS, etc. and the expected organic byproducts of the ecosystems that would inhabit such systems-biomarkers. Interestingly, CH4 is a key expected biological byproduct, although there are many non-biological ways CH4 can be produced.

The low molecular weight organics detected by Cassini may be just one part of a suite of organics present in the plume and on the surface. Studies of the nature of these organics can tell us whether or not they are biogenic.

The molecular species likely to be produced by such a prebiotic or biotic chemistry—such as amino-acids, lipidic compounds and sugars—could be detected in the plumes of Enceladus using in situ techniques. It is also crucial to confirm the presence of liquid water reservoirs, both by remote sensing measurements and measurements made on the surface. Most of the important astrobiological questions to be covered will not be answered by Cassini– Huygens and require a dedicated new mission to Titan and Enceladus.

4 Mission profile/scenario 4.1 Mission concept The baseline mission concept is for two moderately-sized spacecraft to provide both in situ (near or at the surface) orbital and surface science measurements at both Titan and Enceladus, via establishment of a novel, but practical, orbit, which cycles around these two moons (Fig. 2). This mission architecture could provide a cost-effective solution, which is efficient in launch mass, very flexible, and provides a significant element of fault tolerance. In addition, this

Exp Astron Fig. 2 TandEM flight elements with orbital information

architecture—through its multi-element nature—lends itself to international collaboration.

4.1.1 Baseline architecture While TandEM could be implemented through a number of propulsion system options, in our proposal, we concentrated on technologies which provide low risk to the mission. The baseline architecture uses chemical propulsion, while future studies could investigate other options (SEP, aerocapture, aerobraking, etc.).

Our baseline mission concept then considers two spacecraft to be launched around 2018 or later on one or two launch vehicles (to be assessed): 1. the Titan–Enceladus orbiter (carrying the Enceladus penetrators), and 2. the carrier for the Titan in situ investigation elements (the Titan Montgolfière/balloon and the three lander/mini-probes) A possible TandEM mission timeline chart is given hereafter in Table 2.

4.1.2 Key mission phases The orbiter will perform several flybys of the two moons (Titan and Enceladus) early in the mission and will deliver the Enceladus landers/penetrators. The carrier spacecraft would deliver the Titan in situ assets, including the Titan hot air balloon with a mass allocation of ∼265 ± 30 kg, and up to three entry probes with a mass allocation of ∼500 kg. Future studies could consider converting this carrier spacecraft into an orbiter, thus allowing for telecom redundancy

Exp Astron Table 2 TandEM mission possible timeline Possible timeline Proposal pre-selected Internal assessment phase in parallel with Laplace Industrial assessment phase and TDP definition SPC down selection to two missions Competitive definition phase Technology development Consolidation of international collaboration SPC selection of first mission Start of industrial phase Development (7 years) Launch Interplanetary trajectory Arrival at Saturn Phasing for Enceladus mission phase Prime Titan mission phase Extended mission phase at Titan

Oct. 2007 Nov. 2007–Oct. 2008 Aug. 2008–June 2009 Nov. 2009 Jan. 2010–June 2011 2009–2014 2009–2014 Nov. 2011 2011/2013 2011/2013–2018/2020 L L to L + 7–9 years A A–A + 1 year Duration ∼2 years (TBD) 2 years

2018 or later 2025–2027 or later 2027 or later 2027–2028 or later 2028–2030 or later 2030–2032 or later

from the surface assets and some interesting science measurements (e.g., occultations using the telecom link between the two orbiting spacecraft).

The orbiter would provide telecom relay between the Titan in situ elements and Earth.

Launch and cruise phase for both spacecraft In our baseline architecture, both spacecraft would be sent to Saturn on a multiple inner planets gravity assist, using chemical propulsion for the cruise phase, employing swing-bys of inner Solar System planets, and the possibility for cruise phase science. An ExoMarslike launch but with Atlas V-521 could deliver ∼1,700 kg mass to Saturn. The orbiter, initially acting as the Titan–Enceladus cycler, then as a Titan science and telecom orbiter, would deliver up to two Enceladus penetrators, with an assumed mass allocation of ∼100 kg each. As for the second spacecraft, the most effective scenario is that planned for the ExoMars mission, which would allow a 1,660 kg mass at Saturn with ∼800 kg still available for the entry probes.

This large mass margin would allow for spacecraft growth and/or payload increase, or alternatively could result in a smaller launch vehicle—an option, which is based on mission cost and programmatics.

Insertion into Titan–Enceladus cycling orbit establishment The baseline architecture assumes chemical propulsion for both the cruise phase and for following Saturn orbit insertion (SOI). It is then followed by four Titan swingbys to bring the period down to that of the cyclical phase (3:7 resonance between Titan and Enceladus). For SOI, the option for aerocapture at Titan will be studied as an alternative to chemical propulsion, for example with respect to fuel mass savings (∼1.8 km/s), technology impact, reduced time to be captured into a 2:1 Titan–Enceladus 32-days resonant orbit in a single pass

Exp Astron

in comparison to ∼130 days with chemical propulsion (Fig. 3). Aerobraking is another flight-qualified technique which could be envisaged to save mass.

Titan–Enceladus cycling orbit operations The Titan–Enceladus Orbiter would deliver up to two Enceladus landers/penetrators, which would perform in situ seismic, temperature, and possibly compositional measurements. The collected data would be relayed back to Earth through the T–E orbiter during its Enceladus flybys. Around the beginning of this phase the orbiter releases (simultaneously or in separate flybys) the Enceladus mini-probes/penetrators to provide the Enceladus surface investigations via data which are relayed to the Orbiter during its Enceladus flybys. Regular flybys of Enceladus would occur every ∼8.25 days, allowing for considerable targeting flexibility (e.g. some or all could encounter the South Pole plume). The baseline scenario assumes 12 Enceladus and correspondingly two Titan swing-bys, followed by three to four swing-bys of Titan on the way to Titan orbit insertion (Fig. 4).

This phase provides coordinated Enceladus surface and flyby science; flyby Titan science; other Saturnian science, and the serendipitous ability to directly investigate the possible transport of (organic) material from Enceladus to Titan and indirectly to other parts of the Saturnian system.

Titan orbit Coordinated Titan orbital science, Montgolfière operations and surface probe investigations (for ∼1–2 years) will take place after the orbital insertion around Titan. Indeed, at the end of the previous phase, the Orbiter is manoeuvred to a polar orbit solely around Titan to perform optimised Titan orbital science and provide telecom support to the Titan in situ elements.

Specifically, the orbiter will provide high data rate relay communications from the Titan balloon and the small probes to the Earth. The mission is designed so that the Titan Orbiter is established in Titan polar orbit before the flyby carrier delivers the Titan probes. This would leave time for the Orbiter observations of Titan and orbit optimization, in preparation for the in situ measurement support.

Fig. 3 TandEM orbit insertion: chemical (left) and using aerocapture (right)

Exp Astron

Fig. 4 Enceladus–Titan phase of the TandEM mission. Following insertion, a series of three Titan gravity assist injects the spacecraft into an orbit with a period of 6.85 days, which is five times the period of Enceladus (1.37 days). After seven encounters with Enceladus (Ea), a small adjustment makes it possible to encounter Titan again, as the elapsed time (47.95 days) is very close to three Titan periods (47.84 days). A second series of encounters (Eb) can be implemented, extending the Enceladus phase to 96 days with up to 14 fly-bys. A series of five Titan fly-bys and two small deep space manoeuvres implementing a DVGA strategy reduce the relative velocity to 965 m/s, at which stage a 420 m/s insertion manoeuvre results in a 1 day elliptical orbit around Titan. The period can be further reduced by aerobraking

4.2 Earth-based science segment of the mission Earth-based and Earth-bound observations will be conducted on two distinctive sets of targets relevant to TandEM: the mission objects (Titan, Enceladus and their environment) and the TandEM mission spacecraft. The former will help to prepare the mission science operations and then interpret the data from in situ observations and experiments. These “traditional” astronomical observations will be conducted in a broad range of wavelengths, from highenergy domain to low radio frequencies using a variety of Earth-based and Earth-bound facilities.

The latter will include VLBI and Doppler tracking of the spacecraft aimed at achieving very high accuracy of trajectory characterization of all mission spacecraft in various phases of their operations and localisation of in situ measurements required for the mission objectives. In addition, the radio telescopes involved in VLBI tracking will provide a backup “eavesdropping” support to the nominal mission communication scenario enabling receipt of low data-rate signals during critical and/or high scientific value events of the mission.

4.2.1 Astronomy support observations of the TandEM mission objects The TandEM Earth-based observing campaign will follow the example of Cassini–Huygens mission. The data obtained by these observations will

Exp Astron

provide synergic to in situ TandEM measurements. While the specific scientific programme of astronomical observations in support to the TandEM mission is to reflect the mission programme, one should anticipate particular attention to the following topics: • • • •

Long-term seasonal monitoring of all objects of the Saturnian system; Characterisation of Titan atmosphere using continuum (visual, IR) and spectral line (UV, visual, IR, millimetre and sub-millimetre domains) of the Titan atmosphere; Visual and IR characterisation of the surface of Enceladus; Multi-band monitoring of non-thermal radiation of the iono- and magnetospheres of Saturn and other constituencies of the Saturnian system.

The astronomical observations will be conducted in many cases using national and international facilities, accessible via open proposal peer-review mechanism. The TandEM mission will coordinate Earth-based astronomical observing campaign within its remit and in compliance with the access mechanisms of the observing facilities.

4.2.2 VLBI and Doppler tracking segment of the TANDEM mission The Earth-based global network of radio telescopes and processing facilities (correlators) will conduct Planetary Radio Interferometry and Doppler Experiment (PRIDE) aimed at providing ultra-precise estimate of the state-vector of the TandEM mission spacecraft. It will be based on the heritage of the Huygens Doppler Wind Experiment (DWE) and Very Large Baseline Interferometry (VLBI) tracking experiments. Today’s technology and very conservative projection of capabilities of VLBI radio telescopes for the next two decades lead to the following guaranteed 1σ accuracy of positional measurements: 500 m based on S-band (2 GHz signal), 100 m at X-band (8 GHz) and 30 m at Ka-band (32 GHz). VLBI tracking of all mission space-based elements poses minimal requirements on the on-board instrumentation, namely (1) optimisation of the transmitters signal spectra and (2) availability of a reasonably stable on-board local oscillator within the on-board radio transmitting sub-system.

4.2.3 Direct-to-earth data transmission The nominal mission scenario assumes transmission of the science and housekeeping data from the Titan and Enceladus probes via relay by the orbiter (carrier spacecraft). Indeed, the amount of data produced by the probes, or the Titan balloon (e.g. images) will require a high-capacity radio relay system.

However, as an efficient backup able to provide support to critical mission operations and experiments, a low data-rate link can be achieved with the nominal transmission from the probe(s) and received by the large Earth-based radio telescopes. The most attractive option of DtE would involve the Square Kilometre Array (SKA) as the Earth-based facility able to operate at the S band (2.3 GHz). The facility is expected to be fully operational in 2020.

Exp Astron

As shown by preliminary assessment estimates, SKA will be able to receive data streams from the TandEM mission spacecraft at the rate of 30–100 bps.

Implementation of the TandEM DtE using SKA does not require any specific modifications of the on-board radio systems, but does require optimisation of the mission scenario in order to achieve direct visibility “S/C–SKA” during critical mission events. Detailed trade-offs of the TandEM DtE regime will be investigated during the project assessment study.

5 Proposed payload instrument complement This study will investigate strategies to combine instrument concepts to minimize resource requirements and achieve multiple measurement requirements.

Miniaturization strategies that maintain functionality will be investigated.

Heritage will of course be important, but new technology and innovative instrument concepts should be encouraged. A list of a baseline strawman payload is given in this section for each component. Note however, that this is a preliminary list that will be refined and justified during the next study phase, if the mission is selected. We will also look into instrument strategies that will fulfil multiple science goals of both Titan and Enceladus science. More work is required from the science community to arrive at well-defined strawman payload for each element of the mission.

5.1 Titan as a system For the upper atmosphere and magnetospheric interactions scientific objectives, the required measurements are described in Table 3. Some of the required orbiter in situ measurements need coverage of 4π viewing angle (pitch-angle) with high time resolution (6 rpm). It is therefore necessary to put the orbiter in a spinning science phase for at least two Titan days (32 days) together with an additional commissioning period just before this spin-phase.

This spin-phase should be part of the Titan polar and elliptical orbit phase when the periapsis is lowered to 800 km (or as low as possible).

For Titan’s neutral atmosphere the objectives that we delineate in Section 3.1.2 point to the measurements and the instrumentation proposed in Table 4.

The objectives of Titan Surface Sciences (Section 3.1.3) are not being addressed by Cassini. Indeed, with the exception of the small region around the Huygens landing site, Cassini’s optical and radar resolution is inadequate to characterize small scale and possibly dynamic geological features (methane geysers, cryo-flows, stream deposits, dunes etc.) See Fig. 5.

Hence the measurements required to investigate the above objectives include: •

high-resolution global optical infrared stereo mapping and radar surface detection (resolution <100 m)

Exp Astron Table 3 Instrumentation required for the Titan upper atmosphere and magnetospheric measurements Key measurements

Instruments

Densities of major neutral species (N2 , CH4 , H2 )

UV spectrometer Ion and neutral mass spectrometer (INMS) Accelerometer Radio science Auroral and airglow photometer INMS Aerosol analyser Millimetre and sub-millimetre spectrometer

Densities of aerosols, positive and negative ions (10–10,000 amu), stable and reactive H, C, N containing neutral species with resolution of 0.01 amu Direct measurements of neutral winds ion velocities

Vertical profiles of electron and ion temperatures Vertical profiles of neutral temperature (above 400 km)

Exospheric structure including transition region, corona, spatial and temporal variability, escape kinetics and rates of important atmospheric species, ENA formation Ionisation and heating processes of Titan’s atmosphere/ionosphere (HCN)

Photochemistry during solar eclipse of Titan

by Saturn

Measurements of global plasma and magnetic field structure of induced magnetosphere for different conditions of Saturn Local Time and magnetospheric dynamics out to ∼10 RT Measurements concerning ion loss rates for different chemical species through the tail/ wake region under varying magnetospheric condition Unexplored low-altitude structure of the magnetic ionopause and below Study particle populations near magnetic ionopause

Langmuir probe Electric fields and plasma waves Ion drift meter Millimetre and sub-millimetre spectrometer Langmuir probe Radio science Millimetre and sub-millimetre spectrometer UV spectrometer INMS Accelerometer Radio science Energetic neutral atom imager and composition analyser (ENA) UV spectrometer INMS Plasma packagea INMS Auroral and airglow photometer Radio science ENA UV spectrometer INMS Plasma packagea Radio science Plasma package ENA Plasma packagea INMS Radio science Dual magnetometer Plasma analyser Langmuir probe

Atmosphere measurements must cover the globe and differing solar illumination. a The Plasma package can be a highly integrated payload suite (HIP) that consists of a plasma analyser (electrons and ions), dual magnetometer, energetic ion and electron detector, dual Langmuir probe, electric field and plasma waves, ion drift meter and a search coil magnetometer.

Exp Astron Table 4 Instrumentation required for the Titan neutral atmosphere measurements Key measurements

Instruments

Temperature, pressure and methane and ethane humidity at different latitudes and seasons, with high fidelity near the surface Wind from the surface to the thermosphere

GCMS ASI/MET Radio science

Microwave sounder (100–700 km, error 10 m/s) IR spectrometer (70–500 km) Accelerometer (lower troposphere, error <1 m/s) ASI/MET (surface, error <1 m/s) Structure of clouds and precipitation Visible and NIR camera and mapping spectrometer Nephelometer Optical rain gauge Electric field sensor Opacity structure and scattering Nephelometer characteristics of haze Visible and NIR camera and mapping spectrometer Evaporation rates, temperature GCMS and winds over lakes ASI/MET Composition of the surface and its Heat flow/physical properties probe thermal properties Visible and NIR camera and mapping spectrometer Tunable laser spectrometer Accelerometer Subsurface camera/microscope (mineralogy) Subsurface spectrometer Composition of the haze and its Aerosol analyser building molecules Electric field sensor Solar partitioning of sunlight and thermal Visible and NIR camera and mapping spectrometer cooling at different latitudes and seasons IR spectrometer Latitudinal and vertical distribution of minor IR spectrometer hydrocarbons, nitriles and oxygen-bearing Microwave sounder species and its temporal variation GCMS Visible and NIR camera and mapping spectrometer

•

highest-resolution (<1 m) infrared imaging from a near-surface platform (required in any case to select appropriate sampling sites for surface chemistry)

Furthermore, no detailed surface composition mapping was possible with Cassini–Huygens, nor are there measurements of the depth of deposits available to permit a good organic inventory to be determined. Therefore, the following measurement scenario is required: • • •

global mapping from an orbiter regional surface investigations from a near surface platform—balloon in situ measurements from a set of local landed stations/micro-penetrators or using tethers to transport to the balloon subject to further study

with the instrumentation given in Table 5.

Exp Astron

Fig. 5 The diversity of Titan’s surface: geological features discovered by Cassini/Huygens which remain to be explained. Future landing locations will be selected among such sites (Credit: Univ.

Nantes and CIGAL/LESIA)

5.2 Enceladus Our prime science objectives to achieve with regards to Enceladus are to: • • • • •

understand the origin, nature and properties of the plume reveal the existence, depth and extent of sub-surface liquid water search for signs of past/present life and in the process of resolving these issues to also: characterize the surface and the interior characterize the impact of Enceladus on the magnetosphere, other satellites and the ring structure

In order to achieve these science objectives the following measurements via the orbiter are required (Table 6).

5.3 Origin, interior and evolution 5.3.1 Origin and evolution Several observational tests may help address the question of Titan and Enceladus’ origin and evolution. They are described hereafter in Table 7.

Exp Astron Table 5 Instrumentation required for the Titan surface measurements Key measurements

Instruments

High-resolution global optical infrared stereo mapping and radar surface detection (resolution <100 m) Highest-resolution (<1 m) infrared imaging from balloon Composition of the surface, structure, thermal, and electrical properties

Infrared multispectral stereo camera Radar imaging NIR-microscope/close-up imager

Heat flow/physical properties probe Accelerometer Subsurface camera/microscope (mineralogy) Spectrometer Descent camera High-mass-resolution in situ measurements X-ray spectrometer of the surface material (molecular and isotopic) Raman-LIBS spectrometer from a variety of locations Subsurface spectrometer Compositional context mapping from balloon NIR camera and mapping spectrometer Global compositional mapping (resolution <1 km) NIR camera and mapping spectrometer Depth of surface deposits Radar sounder Accelerometer Seismometer Global wind pattern and its seasonal variation ASI/MET Physical properties of the interior (internal Heat flow/physical properties probe structure, intrinsic heat flow, thermal and Seismometer chemical boundary layers) Radiation dosimeter Magnetometer Surface energy exchange (radiation, temperature, ASI/MET chemical boundary GCMS Optical rain gauge Heat flow/physical properties probe Astrobiology detections (habitable subsurface Seismometer biotic and pre-biotic water; chemistry, Subsurface camera/microscope life forms) Spectrometer

5.3.2 Interiors The required measurements for addressing the internal structure goals (as presented in Section 3.3.2) are given in Table 8.

This series of measurements will permit to retrieve (a) the depth and composition of the putative interior liquid layer, (b) the size and state of the rock-metal core, (c) the structure of the crust and the depth of liquid reservoir, (d) the internal differentiation process as well as the formation and recycling process of the atmosphere and the crust.

5.4 Astrobiology measurements Instrument suggestions for astrobiological studies (to be implemented for both satellites) are given in Table 9. Other techniques, similar to those under development for Mars organics analysis (such as the Urey experiment) can be envisaged.

Exp Astron Table 6 Instrumentation required for Enceladus orbiting, and Enceladus and Titan surface and interior measurements Key measurements

Instruments

Map the physical and chemical surface properties and relate them to surface processes, organics and volatiles and physical properties of water ice; passive multispectral remote sensing observations of the surface and plumes; infrared remote sensing temperature maps of cracks and cervices with vents Remote imagery over time of the plume at a variety of phase angles and distances; stereo imaging to measure shape of Enceladus; long term imaging of Enceladus from a stable orbit around Titan with 50–250 m resolution In situ particle, dust, magnetic field observations with composition capability of high sensitivity and mass resolution

UV-IR imaging spectroscopy (0.2–5 μm)

High resolution camera (10 m/pixel)

Magnetometer Energetic particle detector Energetic neutral atom imager Plasma+ion neutral mass spectrometer Dust detector Gas analyser LENA detector Plasma waves/Langmuir probe

Table 7 Instrumentation required for the Titan and Enceladus origin measurements Key measurements

Instruments

Isotopic ratios in major species (H, C, N, O) in surface solid materials (water ice, deposited aerosols, hydrocarbon mixtures) and in gaseous compounds Noble gas abundances and isotopic ratio (Ar, Kr, Xe, Ne) with 100 to 1,000 times higher sensibility than those measured by Cassini–Huygens Sulfur compounds in surface materials (H2 S, OCS)

Gas chromatograph mass spectrometer (GCMS), capable of measuring stable isotopes and incl. shared pyrolysis and chemical extraction Ion and neutral mass spectrometer GCMS (as above)

GCMS (as above)

Table 8 Instrumentation required for the Titan and Enceladus interior measurements Key measurements

Instruments

Mapping topography, gravity and magnetic fields on global and local scales Seismic and tidal survey

Radar or laser altimeter Radio science subsystem, gravity gradiometer Magnetometer Seismometer Gravity gradiometer Radar or laser altimeter Ground penetrating radar

Subsurface sounding, crustal layering and liquid reservoirs Near-surface thermal gradient and thermophysical properties Composition of surface materials, of cryovolcanic magma and gases (especially C, N, O isotopic ratios) Radiogenically-derived noble gas abundances

Thermal sensing package Surface sample analysis package GCMS Vis-NIR mapping spectrometer Radar imaging GCMS

Exp Astron Table 9 Instrumentation required for the Titan and Enceladus astrobiology measurements Key measurements

Instruments

Ionosphere: ion and neutral composition Orbiter within a high mass range (several 1,000 Da) 1–2,000 Da high resolution MS (TOF) Stratosphere–troposphere: high Particle collector and analyser ? sensitivity (sub-ppb level) molecular and Balloon isotopic analysis of the gas phase. Chemical Altimeter composition of the aerosols: organic and Stable isotope mass spectrometer inorganic analysis, elemental, molecular and Gas chromatograph–high resolution isotopic analysis of the aerosols—vertical and mass spectrometer with capabilities for latitudinal variations analyzing refractory materials (laser Surface: organic and inorganic analysis-elemental, desorption, chemical derivatization, molecular, isotopic and chiral analysis differential thermal analyser–pyrolyser, of the surface materials chemolysis) Subsurface: from penetrators and/or Subcritical water extractor–microcapillary by analyzing surface materials electrophoresis system ejected from subsurface Probes/landers Hydrocarbon related mineralogy for surface IR stereo camera/spectrometer for context and subsurface X-ray fluorescence spectrometer Need to analyze different areas: bright Stable isotope mass spectrometer and dark regions, lake/damp playa/ Gas chromatograph mass spectrometer with shoreline/dune field capabilities for analysing refractory material, and chiral GC columns Subcritical water extractor–microcapillary electrophoresis system Raman/LIBS spectrometer Drilling capability to analyse the subsurface and/or melting system Molecular analysis of the plumes and Orbiter search for a subsurface ocean 1–2,000 Da high resolution MS (TOF) Particles collector and analyser?

See Section 3.2 Search for evidences of internal water ocean and See Sections 3.1 and 3.3 information on its properties Search for episodical liquid water bodies on the surface Search for molecular, isotopic (C, S and O) Penetrators/probes and chiral biosignatures Stable isotope mass spectrometer GC-MS with chiral columns Search for molecular, isotopic and chiral (?)

Orbiter biosignatures 1–2,000 Da high resolution MS (TOF) Particles collector and analyser?

5.5 Preliminary strawman payload A traceability matrix was constructed from the science goals prioritization and the associated instrumentation. The strawman instrument payload proposed hereafter provides a strong set of crosscutting complementary observational capabilities, as determined by our traceability matrix. Cameras, spectrometers, magnetometers, radar, radio science, seismometers, organic matter and surface composition analyzers, and a host of new conceptual instruments will scan all

Exp Astron

spectral ranges and return data of high level of detail and quality. The combination of orbiter and in situ elements provides opportunities for both large and small payloads, engaging a potentially wider community of instrument providers from a variety of member states. The heritage of previously successful missions such as Cassini–Huygens, Rosetta, Venus and Mars Express, etc. as well as new ones currently under study (such as ExoMars, etc.) will be extremely beneficial to the definition of the technological feasibility and maturity of the proposed concept.

A TandEM orbiter strawman payload could include such instruments as an advanced ion and neutral mass spectrometer, a thermal IR spectrometer, a dual magnetometer, a microgradiometer, a sub-surface radar, Radar and laser altimeters, and SAR, a radio science experiment, a UV spectrometer, a visible and near-IR camera and mapping spectrometer (in the 0.9–6 μm range).

Other possibilities include a Dual Langmuir probe, a microwave (submm) sounder/spectrometer, a neutral analyser (LENA) and a plasma analyser (for electrons and ions).

For a Titan aerial platform, a GCMS (with stable isotope measuring capabilities), a visible and near-IR camera and mapping spectrometer, an atmospheric structure instrument (MET) and a sub-surface Radar (GPR) would be— according to the traceability matrix—among the main instruments to be on board.

On the Titan mini-Huygens landing probes one would obviously want to have a gas analyzer with chemical sensors, a gas chromatograph mass spectrometer (with stable isotope) and a radio science experiment, as well as a sub-surface radar, a microscope/lose-up imager and a visible and nearIR high-resolution camera and mapping spectrometer depending on the mass allocation permitted.

For the Enceladus penetrators and in order to perform such key measurements as sub-surface chemistry, searching for water and organics, internal body and core structure, surface activity, morphology and landing site context, etc. one might envisage chemical and thermal sensors, seismometers, descent camera, accelerometers, etc.

6 Basic spacecraft key factors and technology issues The TandEM mission will be built around a central core of an orbiter and a number of in situ elements, accommodating a core group of instruments.

Specifically, our baseline architecture assumes the use of two smaller launch vehicles. The first one—a Titan–Enceladus orbiter—would also deliver the Enceladus penetrator(s), while a second Titan flyby/carrier spacecraft would accommodate the Titan hot air balloon and the Titan mini-probes. Alternatively, a large single launch vehicle could be used, but this introduces other complexities to the mission. Aerobraking within Titan’s atmosphere will be given serious consideration to minimize resource requirements and risk. Our baseline approach will allow the orbiter to be used initially for both Enceladus

Exp Astron

and Titan science, with a terminating orbit around Titan. A common set of instruments will be used for both Enceladus and Titan orbiter science to reduce mission cost.

The initial orbiter phase would be optimized for Enceladus science via numerous flybys (i.e., cycler orbits) of Enceladus’ southern plumes, using advanced remote sensing, and particle and field instrument packages. Additionally, such a cycler orbit will allow direct in situ study of the possible transport of (organic) material between Enceladus and Titan, and indirectly to other parts of the Saturnian system. During this phase, the Enceladus penetrator(s) would be released to perform Enceladus surface investigations via data relayed to the orbiter on its flybys. The Titan balloon and miniprobes would be phased to arrive after the orbiter achieves circular Titan orbit, and the orbiter has first completed an initial reconnaissance phase to support balloon and probe science; study Titan’s upper atmosphere and winds remotely; measure the internal structure using altimetry and radio science data and finally study the Titan-magnetosphere interaction using both in situ and remote sensing techniques.

The Titan aerial probe is conceived as a Montgolfière hot air balloon concept, which could stay aloft for several years within Titan’s atmosphere and perform touch-and-go sample acquisition from various surface locations.

Combining the RTG-heated hot air balloon with a tether-suspended sampling mechanism—attached to the gondola—would allow for global in situ measurements of Titan’s lower atmosphere and surface. Based on topographical maps of the surface, atmospheric winds and global compositional maps of the surface from the orbiter, the hot air balloon will be able to navigate safely and have its science mission operations optimized by the time it reaches the surface. The aerial probe, once in Titan’s atmosphere, will navigate nominally at altitudes of ∼10 km, with ceilings at 3 and 25 km above the surface, and provide global measurements of Titan’s lower atmosphere and winds, as well as close up images of the surface and its composition.

Mini-probes (either descent and landing modules launched from the carrier or landers/penetrators deployed from the Montgolfière) will be used for a serendipitous investigation of surface features, as well as atmospheric investigations at different locations than covered by the balloon. Specific terrain targeting and co-temporality will be assured to allow for a small seismic network to be established. Micro-penetrators are also well suited to investigating terrains, which are difficult for soft landers. Further study is required to determine exact probe numbers and configurations according to available mass, capabilities and risk.

This mission will require new technologies and capabilities so that the science goals can be achieved within the cost cap, schedule, and acceptable risks. Miniaturization, while maintaining functionality of spacecraft and science package, will be part of our Technology Development Plan (TDP).

International participation will play a key role in achieving all the science goals of this mission within acceptable cost to all partners. We will sub-divide

Exp Astron

the science payloads into remote sensing packages and in situ packages with instrument PIs for each to simplify management structure and allow for optimization of resources and cost within each of the instrument package teams. For example, we will give serious consideration to the development of an integrated remote sensing package for the orbiter with a common telescope to minimize resource and cost, and allow long-term observations of the Saturn system to be made from a stable Titan orbit. Such observations could be performed through the extended mission phase.

Therefore, in this section, we identify a baseline launch mass and sub-system structure to provide an overall mass, volume and power budget from which rough order of magnitude (ROM) costs can be identified. Approved engineering models will be used to assess resource requirements for such a scenario and additional capabilities, if possible within acceptable cost and resource margins, will be added to the payload. This will then provide a framework around which our proposed mission concept will be built. The above mission concept will allow us to identify key areas for technology development and corresponding development of a technology plan.

6.1 Orbiter Here, we identify the key pieces of the Titan–Enceladus orbiter with a baseline science payload. Our baseline configuration uses chemical propulsion only for the cruise phase and for Saturn orbit insertion (SOI) with a V of ∼2 km/s. In addition, the Titan–Enceladus cycling phase, plus Titan orbit insertion (TOI) would require an additional V of ∼1 km/s. As a mission architecture trade option, we will also consider using an aeroshell to enable Saturn orbit insertion (SOI) via a Titan aerobraking maneuver (TAM), and then chemical propulsion techniques for the remainder of the mission. The aeroshell is deployed after SOI to reduce orbiter fuel mass for chemical propulsion maneuvers during the Enceladus phase of the mission and Titan orbit insertion (TOI). As another option, we will also consider retaining the aeroshell to put the orbiter in orbit around Titan. The aeroshell is considered a technology under development in the US, so cannot be considered as baseline at this time.

The orbiter is a key TandEM element, since it represents a significant increase in science return, but also serves as communication link between Earth and the balloon and probes/landers on the ground, while the Direct to Earth (DtE) communication possibility has to be assessed and can only be complementary. We will also take advantage of the low radiation environment, and reduce risk by miniaturizing the payload mass and power and thus cost.

Structural mass reductions using new composite materials being developed independently by industry will be surveyed during the proposal period. Deployable high gain antenna (HGA) options will be considered as another mass reduction approach.

The baseline orbiter sub-system package includes dry spacecraft mass and V ∼ 3,000 m/s chemical propulsion fuel mass. The launch mass is ∼2,586 kg.

Exp Astron

The mass of a solar electric propulsion (SEP) module would be 1,080 kg for 18.7 kW of power (at 1 AU). The numbers are based on present mass, volume and power numbers for typical spacecraft and science instrument (SI) estimates and we expect with focused and independent technology developments these numbers will come down. Margins ∼20–30% have been added. Smaller margins ∼10% are used for fuel allocations. The data does show that by using the aerobraking option to put the orbiter in orbit around Saturn, with a launch mass of ∼1,653 kg one can use the Atlas V-521 launcher for the orbiter payload and as we discuss below can use a separate Atlas V-521 launcher for the Titan balloon, probes and carrier. For the baseline and the aerobraking option one can make ROM cost estimates of orbiter and mission operations (MO) during the study phase. Clearly, the most cost effective option is aerobraking so that our TDP will make the aeroshell technology a critical element for development.

6.2 Titan Montgolfière Among different types of balloons, the most appropriate for Titan’s case appears to be the Montgolfière—or hot air balloon—heated by the excess heat of two RTGs (Radioisotope Thermoelectric Generators).

The Titan Montgolfière has been studied by the Balloon Department of the CNES Toulouse Center which possesses an extensive knowledge of such vehicles as proved by the successful program called MIR (Montgolfière Infra Rouge) launched on a routine basis in the Earth’s atmosphere since 1980. The studies show that the atmosphere of Titan appears as the most favourable of all the planetary atmospheres for balloon (especially Montgolfière) flight, because it is rather dense, since mainly composed of nitrogen with a surface pressure of 1.5 bar, and very cold (less than 100 K in the 0–70 km altitude range).

Compared to MIR on Earth, with the same temperature ratio and for the same lift, the balloon volume can be reduced because of a larger pressure on Titan (∼1 bar) with respect to Earth (∼0.07 bar) at float altitude.

An orbiter is required as a data relay to Earth with a transmission rate of 20 kbs−1 between the orbiter and the buoyant station(s) and two MultiMission Radioisotope Thermoelectric Generators (MMRTGs) are needed to produce a power of about 3.5 kW after 8 years of planetary cruise and assumed 2 years of Earth storage. In this case, the sustentation is provided by Titan’s gas, filling the balloon and heated by an external energy source. It is therefore not necessary to carry the heavy tanks, whose mass create a major constraint on the balloon mission.

The drawback of the Montgolfière is that the difference in density between the aerostat gas and the ambient atmosphere is so small that a large volume is required. At Titan, a Montgolfière requires 1% of the heat required by Earth Montgolfières due to a combination of reduced cryogenic radiation, reduced heat transfer, and buoyancy inversely proportional to the square of

Exp Astron

the absolute temperature. Solar or IR atmospheric heating is too weak on Titan to be considered, catalytic decomposition of hydrazine and RTG heating are possibilities (preferably the latter).

The principle of a flight in Titan’s atmosphere is to use the local gas heated by the excess thermal flux of the RTG system, which is baselined for providing electrical power to the balloon payload. The thermal effects coming from the Sun on the planet’s atmosphere are negligible. The major question in the analysis of the flight is the convective exchanges with ambient; this has been treated by CNES with the help of the model of convection well validated by the MIR programme.

The results have been applied to the example of a 1,000 m3 balloon (Fig. 6).

With the MMRTGs as a heat source, the heat balance leads to a temperature difference of 3◦ C to 4◦ C providing sufficient lift for a typical payload mass.

The balloon would have a radius of 6.5 m, a mass of 28 kg for a skin thickness of 30, 5 μm made of a film of Kapton coated by FEP on both sides. This materiel is assembled by thermosoldering. Because of the expected instability of the vehicle in the temperature gradient, it is necessary to add a small 2.3 m radius balloon (mass 4 kg) filled with hydrogen (11 kg).

The system would be deployed starting around 30 km of altitude and the balloons inflated while descending down to 3 km where the tanks and inflation system (60 kg) used to carry hydrogen would be jettisoned. Then the vehicle would climb back to a nominal float altitude of 10 km (Fig. 7). A valve placed

Fig. 6 Titan Montgolfière system description

Exp Astron

Fig. 7 Titan deployment and flight simulation

on top of the Montgolfière (Fig. 7) would be used for emptying partially the main balloon in order to obtain excursions in altitude (up to 25â&#x20AC;&#x201C;30 km).

Nominal ceiling could be achieved around 10 km by using the active control of the valve (mass of control system = 10 kg). Another possibility is a ceiling around 3 km. Complete filling corresponds to an infinite flight around 30 km.

The total mass breakdown is 263 kg including 30 kg of margin. The mass available below the balloon is 130 kg, including 90 kg of the RTGs; a gondola for housekeeping, data telecommunication and scientific instruments of 40 kg.

It is estimated that a scientific payload of at least 20 kg would be available out of the 40 kg gondola.

The lifetime is anticipated to be long (months to years) since leaks do not impair performances in a Montgolfière. However the lifetime would be reduced due to H2 leakage from the secondary balloon.

6.3 Probes Probes provide in situ measurements not possible with orbiting or airborne instruments; ground truth; and support interpretation of orbiting instrument data with hard ground information.

Exp Astron

We propose a baseline of Titan descent and landing mini-probes, in line with the Huygens heritage, as well as micro-penetrators and lander probes for Enceladus to address the mission requirements in complementary ways. Their selection will ultimately depend on available mass, capability and risk.

6.3.1 Titan mini-probes The Titan mini probe design will benefit from the Huygens probe experience, and from the GEP-ExoMars (Geophysics Package) development (itself originated from the NETLANDER project).

The arrival on Titan will be on very similar conditions to the Huygens one. The aeroshell and the parachute system will hence be derived from the Huygens one, downscaled to produce the same ballistic coefficient. An aeroshell diameter of 1.3 m will provide the same entry as the Huygens one.

The surface module design will be strongly inherited from the GEPExoMars, which has similar environment and mission constraints: • • •

Autonomy; Cold thermal environment; Low solar power availability.

The negligible solar flux on Titan’s surface led to implement a radioisotopic device for electrical power generation. The GEP development of a European RTG could be used to satisfy the Titan mini-probes’ power requirement. This RTG could be based on a Russian RHU (radioisotope heater unit) design, with a European thermoelectric generator (TEG) developed in Europe, which is expected to provide an increased system efficiency, compared to the RITEC TEG used on the (unsuccessful) MARS-96 mission. The electrical power will be in the order of 4 W, which leads to a time-sharing payload management.

The power system will be based on the coupled RTG/secondary battery with an operational schedule taking into account science measurements campaign, telecommunication session to the Orbiter and battery charging.

The Titan mini-probes will benefit from electronic integrated design in order to reduce the mass, and to share the thermal dissipation to the overall probe elements. The RTG will be embedded within the overall equipments.

The avionics will benefit from the ExoMars GEP development. The thermal control will inherit from the Huygens design which has demonstrated a very good behaviour. A high TRL level is applicable for the Entry and Descent system, thanks to the strong heritage from Huygens. A TRL level of eight is hence considered.

For the parachute system, a resizing is necessary considering the lower mass with regard to the Huygens probe. A TRL of eight is considered. The surface module will be strongly derived from the GEP-ExoMars module.

Considering the specific operational concept required by the electrical power management (time sharing between instruments and telecommunication), the data handling will necessitate specific development in terms of operation. A TRL level of seven could be considered for the data handling.

Exp Astron

6.3.2 Micro-penetrators We propose low mass kinetic penetrators capable of withstanding impact at velocities up to several hundred m/s, and embed themselves under the surface to a depth of up to a few metres.

The whole system comprises three major elements: (a) carrier accommodation and ejection system; (b) a descent probe which contains a thruster and attitude control assembly, a possible descent camera; and also includes (c) the penetrator itself (if penetrators were also used for Titan, we could propose a largely common penetrator design and payload). Figure 8 shows a conceptual design of such a penetrator.

A preliminary mass estimate for such a common penetrator is â&#x2C6;ź5 kg, which includes a â&#x2C6;ź1.7 kg core instrument suite comprising seismometers, geochemistry package, thermal sensors, accelerometers and descent camera (see Table 10). Complementary payload instruments (e.g. subsurface mineralogy/astrobiology camera, magnetometer, etc.) will be considered for inclusion after further study of science priorities, technical maturity, and available mass.

The only payload item which may not be common to both Enceladus and Titan is that of the chemical sensor because of the atmospheric presence on Titan.

Though there is space qualification heritage for such systems with Lunar-A and Deep Space-2 (DS2) at impact velocities proposed here, together with extensive military experience with instrumented shells, some particular adaptions to this mission will be required e.g. (ruggedisation of new technology science instruments; possible effects on long cruise phase on thrusters (solid) fuel; low mass attitude control system; extending battery power system lifetime by employing RHUâ&#x20AC;&#x2122;s; and possible need for a trailing aerial for communications through subsurface material).

Fig. 8 Penetrator descent module

Detachable Propulsion Stage Point of Separation Payload Instruments

Penetrator Delivery System Penetrator

Exp Astron Table 10 Possible penetrator science payload Penetrator payload instrument

Mass

Heritage

Descent camera (g) Accelerometer and tiltmeter (g) Geochemistry package (g) Water-volatile experiment (g) Seismometer (g) Thermal/heat flow (g) Permittivity (g) Mineralogy/astrobiology camera (g) Magnetometer (g) Total penetrator payload mass (kg)

10 66 260 750 300 300 ∼ 100 ∼ 200 60 ∼2

Beagle2, ExoMars DS2, Lunar-A Beagle-2 XRS DS2, Ptolemy (Rosetta) ExoMars Lunar-A IWF (Graz) Ground based (USA) Various space missions

Enceladus penetrators A two penetrator system is proposed to allow investigation of two surface terrains and provide some fault tolerance. Deployment from the Titan–Enceladus orbiter requires a thruster to reduce the ∼3.8 km/s V to an impact velocity around 200–300 m/s. Preliminary calculations for solid fuel system (such as one employed for Lunar-A), with an ejection velocity of ∼2,600 m/s, equates to a thruster system comprising ∼26.8 kg fuel + 2.5 kg for associated tanks, structure and motor. Use of a bi-propellant system, could result in a slightly lower separation mass and provide additional control. The result is a total separation mass for a two penetrator system estimated at around 77 kg. Thus, a nominally allocated 100 kg for this system provides a 23 kg contingency which could be used to enable additional complementary payload instruments if sufficient mass is available.

Titan penetrators These may be deployed either from orbit or from the hot air balloon. In either case, descent will be significantly affected by Titan’s thick atmosphere. This may allow fin-based ACS which have commonly been used by the defence sector on Earth. Here, thermal insulation from the cold landed phase may also be useful for protection against atmospheric heating.

Communications could be to the orbiter or balloon, depending on balloon location with time. The descent probe subsystems will be largely identical with those for Enceladus except for the ACS to reduce costs.

Orbit deployment Allows deployment of multiple penetrators to pre-selected different regions/terrains, at closely spaced times to also allow a simultaneous small seismic network to be operational. There is a heritage of direct atmospheric entry with the Titan Huygens probe. A preliminary analysis using comparison with Huygens indicates a mass per descent probe of around 12 to 30 kg respectively, which is also in agreement with simple use of thrusters and ACS system to ensure correct entry orientation and sufficient penetration.

Balloon deployment This allows low altitude and perhaps lower mass deployment, but targeting will depend upon the vagaries of Titan’s winds, and consequent ability to ensure a contemporary seismic network is unlikely if

Exp Astron

specific terrain targeting is a priority. A mass estimate for a thruster based balloon deployment is ∼8.5 kg/descent probe, based on a reduced 1.5 kg probe accommodation and ejection system, and ∼1.0 kg thruster mass.

6.4 Key technology areas This future ambitious program will require ESA and its international partners to investigate innovative and critical technologies, such as: consolidating and improving upon the Huygens entry technology and extending it to controlled dips in the atmosphere which are required for aerocapture; the orbital stability of Titan’s aerocaptured orbiter; technology development for the balloons the mini-probes and the penetrators; DtE communications, etc.

A technology implementation plan, for infusing these and other technologies into the TandEM mission, will be developed during the pre-phase A study.

The list of study tasks also includes: 1. RTG heat exchanger, material development (two-layer balloon concept) and drop and deployment test for Montgolfière balloon. Small RTGs enable many new options e.g. long-lived seismic stations).

2. Microelectronics development which can be done under low radiation specification for mission.

3. Develop tether system and surface-liquid sampling capabilities.

4. Trade studies on solar electric propulsion.

5. Trajectory designs for probe/landers/ penetrators releases on Titan and Enceladus.

6. On-board science autonomy: data selection, compression and storage C&DH and Telecom systems.

7 Communications and outreach Astronomy plays an important role in every day life at all latitudes and longitudes and even the layperson is very much interested, as perhaps it seems that this science is imprinted in our genes and we have an atavistic interest for it. It is recognised in this proposal that communication and outreach is an important part of this mission. The purpose is the rapid dissemination of information to scientists as well as information to the general public. Titan and Enceladus are complex systems with many interacting components like the Earth.

The interest of the informed and layman public in the Saturnian system has steadily increased in the past, with a peak in the current Cassini–Huygens era.

Images from the Saturnian system have regularly been featured as the Astronomy Picture of the Day and have attracted international media broadcasting.

This can only be amplified by a return to Saturn’s neighbourhood. TandEM will be appealing for educational institutions at different levels.

Exp Astron

There are several points about the TandEM mission that make it unique and of high interest to the general public. Titanâ&#x20AC;&#x2122;s being another world with many similarities to an early Earth is one. Although it is a moon, it is larger than Mercury, has an active geology and a significant atmosphere. The geology has Earth-like scientific and educational analogues despite its alien environment: Mountains of ice, lakes of methane/ethane, dunes where the dust is ice or even some organic material. The atmosphere is probably similar in composition and density to that of the early Earthâ&#x20AC;&#x2122;s. Although Titan is extremely cold, there are complex organic chemical reactions occurring in the atmosphere, which may be similar to those that occurred on Earth.

The TandEM mission makes full use of Titanâ&#x20AC;&#x2122;s atmosphere for aerobraking/aero-capture and to enable a highly mobile instrument platform, i.e. a hot air balloon. The physics of a hot air balloon demonstrate practical aspects of physics used in school curricula e.g. mass, density and the ideal gas laws.

There are many websites describing design and construction of model hot air balloons, showing that there is a general interest in hot air balloons.

During pre-launch and cruise phases there can be competitions to name and design logos for the balloon and probes/landers for Titan and Enceladus (similar to the Philae lander for the Rosetta mission). Gradually building media awareness through briefings and pres releases. Design educational programs. Enable people to add their names on a CD that will be carried on the gondola.

During the Titan/Enceladus encounter: Dissemination of information on web. News alerts to subscribers on mobile phones by text messaging. Enabling the general public to help choose where to land the probes and penetrators and capture balloon samples.

Public outreach activities will include: internet sites, leaflets, a comic, 3D constructions with cardboard or plastic, public lectures, TV and radio programs, CD and DVDs, exhibitions, articles in magazines and newspapers, theatrical plays for Saturn and its satellites with emphasis on Titan and Enceladus. The supports could be produced and distributed by ESA and all its partners to media and newspapers and possibly through newspapers to the general public.

Interactive exhibitions are considered to spread around European institutions and planetariums with the collaboration of the local astronomy and physicists societies. their main subject will be a virtual tour, beginning with the evolution of Astronomy from the ancient years including the famous Antikythera mechanism, the oldest astronomical computer, and ending with a travel to the Saturniann system with final destinations Titan and Enceladus.

After having watched the show, children will be involved at a game, which concern the design of a mission to Titan and Enceladus and a balloon trip to fly over and study it.

Moreover, several contests will be organized by ESA among the European schools, in order to collaborate each other and create mission posters, log and flag. An expedition to name the TandEM instrumentation could be also held.

Ideas and resources will naturally flow in regard to such a mission.

Exp Astron

8 Conclusions and recommendations The TandEM mission, currently under study by ESA and its international partners will make an enormous impact on the planetary science community.

Indeed, the list of questions pertaining to Titan as a system is lengthy as befits a world perhaps second only to Earth in its level of geologic and atmospheric activity (as described in Section 3.1). The origin and evolution of the puzzling moon, Enceladus, is also a critical end point for understanding the formation of planetary systems and its astrobiological potential, as explained in Section 3.2.

In addition, the variability of the magnetospheric plasma, neutral gas, Ering ice grain density, radio emissions, and co-rotation of Saturn’s planetary magnetic field in response to Enceladus’ plume activity need to be addressed to understand the present Saturnian system of satellites, rings and plasma.

Such a mission will also certainly stimulate important technological advances through the challenging new components required for this investigation. In particular, we have begun to develop conceptual designs for delivering the science payload included in the orbiters, aerial platforms and probes, and to define a launch/delivery/communication management architecture. The TandEM mission concept will allow us to identify key areas for technology development and corresponding development of a technology plan.

It is highly recommended if not mandatory for the international space agencies to combine efforts and collaborate in order to make such an ambitious endeavour a reality. ESA and its partners will share the mission operations according to a plan to be defined during the study phases. TandEM will require new technologies and capabilities so that the science goals can be achieved within a cost cap defined with acceptable risk for a class-L mission (similar to the one of Cassini–Huygens). The cost target considered here is widely recognized in all studies for a return to Titan and Enceladus to be reasonable given the cost of the highest priority science. International participation among ESA, NASA and other potential partners will play a key role in achieving all the science goals of this mission, which will revolutionize our understanding not only of the Saturnian system, but of the entire Solar system.

Acknowledgements We want to thank Ivan Juiz who first put together the TandEM web site, and our current webmaster: Renaud Romagnan. We are also grateful to Sylvain Cnudde for considerable artistic work on TandEM, and to SIGAL@LESIA.

References 1. Alibert, Y., Mousis, O.: Astron. Astrophys. 465, 1051–1060 (2007) 2. Atreya, S.: Science 316, 843–845 (2007) 3. Atreya, S.K.: Atmospheres and ionospheres of the outer planets and their satellites, pp. 183– 186. Springer, New York (1986) 4. Atreya, S.K., et al.: Science 201, 611–613 (1978) 5. Atreya, S.K., et al.: Planet. Space Sci. 54, 1177–1187 (2006) 6. Backes, H., et al.: Science 308, 992–995 (2005) 7. Coates, A.J., et al.: Geophys. Res. Lett. 34, L22103 (2007)

Exp Astron 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

Coustenis, A., et al.: Icarus 189, 35–62 (2007) Dougherty, M., et al.: Science 311, 1406 (2006) Engel, S., Lunine, J.I., Norton, D.L.: J. Geophys. Res. 99(E2), 3745–3752 (1994) Fortes, A.D.: Icarus 146, 444–452 (2000) Fortes, A.D., et al.: Icarus 188, 139–153 (2007) Griffith, C., et al.: Science 310, 474–477 (2005) Grott, M., et al.: Icarus 191, 203–210 (2007) Hersant, F., et al.: Planet. Space Sci. 52, 623–641 (2004) Hersant, F., et al.: Icarus (2007, in press) Hurford, T.A., et al.: Nature 447, 292–294 (2007) Israël, G., et al.: Nature 438, 796–799 (2005) Khare, B.N., et al.: Icarus 68, 176–184 (1986) Kieffer, S.W., et al.: Science 314, 1764 (2006) Lorenz, R.D., et al.: Geophys. Res. Lett. 34, L07204 (2007) Magni, G., Coradini, A.: Planet. Space Sci. 52, 343–360 (2004) Matson, D.L., et al.: Icarus 187, 569–573 (2007) McKay, C.P., et al.: Icarus 80, 23–53 (1989) McKay, C.P., et al.: Science 253, 1118–1121 (1991) McKay, C.P., Smith, H.D.: Icarus 178, 274–276 (2005) Niemann, H.B., et al.: Nature 438, 779–784 (2005) Nguyen, M.-J., et al.: Planet. Space Sci. 55, 2010–2014 (2007) Nimmo, F., et al.: Nature 447, 289–291 (2007) Owen, T., et al.: Phys. Uspekhi. 48, 635–639 (2005) Owen, T., et al.: Faraday Discuss. 133, 387–391 (2006) Palguta, J., et al.: Icarus 180, 428–441 (2006) Porco, C.C., et al.: Science 311, 1393–1401 (2006) Rappaport, N., et al.: Icarus 126, 313–323 (1997) Rappaport, N., et al.: Icarus 190, 175–178 (2007) Raulin, F.: Space Sci. Rev. Available at http://www.springerlink.com/content/ 0527749877335772/. (2007, in press) Schubert, G., et al.: Icarus 188, 345–355 (2007) Soderblom, L.A., et al.: Planet. Space Sci. 55, 2025–2036 (2007a) Soderblom, L.A., et al.: Planet. Space Sci. 55, 2015–2024 (2007b) Sohl, F., et al.: J. Geophys. Res. 108, 5130 (2003). doi:10.1029/2003JE002044 Sotin, C.: Nature 445, 29 (2007) Sotin, C., et al.: Nature 435, 786–789 (2005) Spencer, J.R., et al.: Science 311, 1401–1405 (2006) Spitale, J.N., Porco, C.C.: Nature 449, 695–697 (2007) Tian, F., et al.: Science 308, 1014–1017 (2005) Tobie, G., et al.: Icarus 175, 496–502 (2005) Tobie, G., et al.: Nature 440, 61–64 (2006) Tobie, G., et al.: Icarus (2007, in press) Tokano, T., et al.: Nature 442, 432–435 (2006) Waite, H., et al.: Science 311, 1419–1422 (2006) Waite, H., et al.: Science 316, 870–875 (2007)