Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century by Research Signpost

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications Editor

Agnieszka Iwan Electrotechnical Institute, Division of Electrotechnology and Materials Science M. Sklodowskiej-Curie 55/61 Street, 50-369 Wroclaw, Poland

Co-editor

Ewa Schab-Balcerzak Institute of Chemistry, University of Silesia, Szkolna 9, 40-006 Katowice, Poland

Transworld Research Network, T.C. 37/661 (2), Fort P.O., Trivandrum-695 023 Kerala, India

Published by Transworld Research Network 2011; Rights Reserved Transworld Research Network T.C. 37/661(2), Fort P.O., Trivandrum-695 023, Kerala, India Editor Agnieszka Iwan Co-editor Ewa Schab-Balcerzak Managing Editor S.G. Pandalai Publication Manager A. Gayathri Transworld Research Network and the Editors assume no responsibility for the opinions and statements advanced by contributors ISBN: 978-81-7895-523-0

Preface Liquid crystals (LC) constitute an fascinating class of materials in terms of their academic fundamental properties, and their tremendous success in commercial applications. The liquid crystal state exist between the crystalline solid and the amorphous liquids. The unique combination of flow and molecular ordering result in many valuable properties. Despite the fact that liquid crystal have been discovered more than one hundred years ago in 1888 by F. Rheintzerâ&#x20AC;&#x2122; and found many applications they are very interesting as materials for modern technologies: organic electronics and photonics and can be considered as materials of XXI century. Serious efforts have been made to exploit LC in such devices as transistors, photovoltaics, organic light emitting diodes and lasers. This is a multidisciplinary field of research, evidenced by the pervasive nature of LC science, constantly expanding and challenging with new applications, new molecular architecture design and new phase types discovery. Seven chapters of this book have been contributed by internationally recognized scientists carrying out their research concern of various liquid crystals low molecular compounds and polymers through their molecular structure design, synthesis and mesomorphic behavior towards to application of them in modern technology. The topics addressed in this book include: synthesis and mesomorphic properties study of liquid crystals containing benzothiazole core; an overview of Schiff base liquid crystals; liquid crystalline compounds with inorganic p-carborane cages along with perspective applications of them for thermal neutron radiation; dendrimers with liquid crystal properties; LC epoxy resins and their potential applications; polymers for LC photoalignment and utilization of atomic force microscopy techniques in study of LC compounds. Finally, we would like to thank all the contributors who make this work possible. Agnieszka Iwan Ewa Schab-Balcerzak

Contents

Chapter 1 Benzothiazole as structural components in liquid crystals Sie-Tiong Ha, Teck-Ming Koh, Siew-Teng Ong, Yip-Foo Win and Yasodha Sivasothy Chapter 2 An overview of liquid crystals based on Schiff base compounds Shankar B. Rananavare and V.G.K.M. Pisipati Chapter 3 p-Carborane liquid crystals: Nematogenic properties and potential application Adam Januszko Chapter 4 Liquid crystalline dendrimer: Towards intelligent functional materials Amrit Puzari Chapter 5 Liquid crystallinity in polymers â&#x20AC;&#x201C; Liquid crystalline epoxy resins Beata Mossety-Leszczak and Magdalena WĹ&#x201A;odarska

125

Chapter 6 New strategy in development of liquid crystal photoaligning materials with reactive C=C bonds Lyudmyla Vretik, Oleg Yaroshchuk, Valentyna Zagnii Vasyl Kyrychenko and Volodymyr Syromyatnikov Chapter 7 Utilization of various atomic force microscopy techniques in investigation of liquid crystal compounds Andrzej Sikora

153

191

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 1-17 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

1. Benzothiazole as structural components in liquid crystals 1

Sie-Tiong Ha1, Teck-Ming Koh2, Siew-Teng Ong1, Yip-Foo Win1 and Yasodha Sivasothy3 Department of Chemical Science, Faculty of Science, Universiti Tunku Abdul Rahman, Jln Universiti Bandar Barat, 31900 Kampar, Perak, Malaysia; 2Department of Chemistry, National University of Singapore, 3 Science Drive 3, 117543, Singapore; 3Chemistry Department, Faculty of Science, Universiti Malaya, 50603 Kuala Lumpur, Malaysia

Abstract. Four new homologous series of liquid crystals comprising a benzothiazole core were prepared and studied. These homologous series are 4-[(1,3-benzothiazol-2-ylimino)methyl]phenyl alkanoates, 4-[(1,3-benzothiazol-2-ylimino)methyl]-3-hydroxyphenyl alkanoates, 4-{[(6-methoxy-1,3-benzothiazol-2-yl)imino]methyl}phenyl alkanoates, and 3-hydroxy-4-{[(6-methoxy-1,3-benzothiazol-2-yl)imino]methyl}phenyl alkanoates. The members in each series differed in the length of their alkanoyloxy chain. The mesomorphic properties were determined by differential scanning calorimetry, optical polarizing microscopy and X-ray diffraction techniques. The effect of the introduction of the benzothiazole moiety into the molecular structure of the liquid crystals on the mesomorphic properties are discussed and compared with that of other wellknown molecular fragments. The influences of the change in the substituents attached to the benzothiazole core on the mesomorphic properties are also reported. Correspondence/Reprint request: Dr. Sie-Tiong Ha, Department of Chemical Science, Faculty of Science Universiti Tunku Abdul Rahman, Jln Universiti, Bandar Barat, 31900 Kampar, Perak, Malaysia E-mail: hast_utar@yahoo.com, hast@utar.edu.my

Sie-Tiong Ha et al.

1. Heterocyclic liquid crystals Molecular order in liquid crystal phases depends mainly on the mesogenic core structure, its geometry, polarizability, molecular conformation, length-to-breadth ratio as well as the number and the position of permanent dipole moments in the core [1]. Interest in the study of mesomorphic heterocycles has dramatically increased in the recent years due to their wider range of structural templates as well as their optical and photochemical properties [2]. The significance of the heterocyclic core in determining the properties of liquid crystals have been reported in a series of review papers [3]. Heterocyclic liquid crystals can be synthesized into having high dielectric biaxiality built into the compact core unit which is essential in technological devices. These materials possess great potential application in spatial light modulation, all-optical signal processing, optical information storage, organic thin-film transistors, fast switching ferroelectric materials, fluorescent probes for the detection and analysis of biomolecules [4]. Heterocyclic mesogens are usually incorporated with heteroatoms, such as N, O and S, resulting in a reduced symmetry in the overall molecule as well as the generation of a stronger polar induction. The inclusion of the heteroatom can considerably change the polarity, polarizability and to a certain extent the geometry of a molecule, thus influencing the type of mesophase, the phase transition temperatures, dielectric constants and other properties of the mesogens [5]. Examples of liquid crystals with incorporated heterocyclic rings are pyridine [1], thiophene [6], oxadiazole [7] and benzoxazole [8]. Benzothiazole liquid crystals Although many compounds having a heterocyclic core exhibit mesomorphic properties, mesogenic examples derived from benzothiazole are relatively rare. Therefore, the benzothiazole ring is chosen as the mesogenic core in this study. Conventionally, liquid crystalline organic compounds have been widely used as materials in liquid crystal displays. Recently, a few series of smectic reactive benzothiazole mesogens having non-conjugated diene end groups were reported owing to their potential light-emitting and charge-transporting behavior in organic light-emitting devices (OLEDs) [9]. Besides that, assembling in smectic liquid crystalline phase can also induce the overlapping of aromatic cores, hence facilitating the hopping of charge carriers between the molecules. Owing to this phenomenon, Hanna and

Benzothiazole as structural components in liquid crystals

co-workers also studied carrier transport properties of some smectic benzothiazole liquid crystalline derivatives [10-13]. Thus, based on the above mentioned interesting results, this has prompted us to study benzothiazolebased liquid crystals. Hence, we report on four new homologous series of liquid crystals consisting of a benzothiazole core and their liquid crystalline properties were also investigated.

2. Liquid crystalline compounds containing benzothiazole core 4-[(1,3-Benzothiazol-2-ylimino)methyl]phenyl alkanoates, 4-[(1,3-benzothiazol-2-ylimino)methyl]-3-hydroxyphenyl alkanoates, 4-{[(6-methoxy-1,3benzothiazol-2-yl)imino]methyl}phenyl alkanoates, and 3-hydroxy-4-{[(6methoxy-1,3-benzothiazol-2-yl)imino]methyl}phenyl alkanoates were synthesized and their liquid crystalline behaviors are discussed. The members in each of the series differed in the length of their alkanoyloxy chain. The synthetic protocol mainly involved the condensation between benzothiazole amino and aromatic aldehydes followed by the Steglich esterification of the intermediate compound (Schiff base) with fatty acids. The synthetic scheme is depicted in Scheme 1. O

N NH2 R1

C2 H5 OH

N R2

R2 1. DCM, DMF 2. DCC, DMAP 3. C n-1H2n-1COOH

OOC n-1H2n-1

N N R1

Compound nBSP nBSP-OH CH3O-nBSP CH3O-nBSP-OH

R1 H H OCH3 OCH3

R2 H OH H OH

n 2 to 6, 8, 10, 12, 14, 16, 18 8, 10, 12, 16, 18 2 to 6, 8, 10, 12, 14, 16, 18 2 to 6, 8, 10, 12, 14, 16, 18

Scheme 1. Synthetic routes towards four homologues series of benzothiazole derivatives.

Sie-Tiong Ha et al.

Series nBSP and nBSP-OH The main structural difference between both series is the presence of the lateral hydroxyl group in nBSP-OH and the absence of it in nBSP. The liquid crystalline behaviour of the compounds was first examined through differential scanning calorimetry. The graph of the phase transition temperatures against the number of carbons in the alkanoyloxy chain of nBSP and nBSP-OH were illustrated in Figures 1a and 1b, respectively.

(a)

Transition temperature C

130

120

Cr-I

110

100

90 6

Number of carbon atoms (n) in alkanoyloxychain

(b) Figure 1. Graph of transition temperatures against the number of carbon atoms (n) in the alkanoyloxy chain of (a) series nBSP and series nBSP-OH. Cr, SmA and I denote the crystalline, smectic A, and isotropic phases, respectively.

Benzothiazole as structural components in liquid crystals

Liquid crystal phase was only appeared in series nBSP (higher members). The presence of the liquid crystal phase was further characterized via optical polarizing microscopy technique. Observation under the polarizing microscope revealed that the higher members (n = 10 to 18) in series nBSP exhibited a SmA phase. Optical photomicrographs of 10BSP and 16BSP are shown in Figure 2. As for series nBSP-OH, no liquid crystal phase was detected. For 10BSP, upon cooling from its isotropic liquid state, the SmA phase emerged as b창tonnet (Figure 2a) and then coalesced to form a fan-shaped focal conic texture. As for compound 16BSP, the co-existence of the fan-shaped and homeotropic (dark region) textures (Figure 2b) were observed. In the homeotropic region, the director of the phase is orthogonal to the layer planes. Consequently, the observed phase is assigned as a SmA phase. Varying geometric anisotropy (ratio between length and breadth of a molecule) by increasing the alkyl chain length is an important factor for the diverse properties among the members. From the graph (Figure 1a), it is clearly noticed that no liquid crystal phase was observed for those short chain derivatives of nBSP (n = 2, 3, 4, 5, 6, 8). In general, a rigid molecular structure is not favoured in generating a liquid crystal phase. However, the liquid crystal phase starts to emerge from the n-decanoyloxy derivative onwards as a monotropic (metastable) SmA phase. Thus, it was confirmed that a certain length of the flexible chain (C10) is a prerequisite in promoting liquid crystal phase formation for this homologous series. The length of the alkanoyloxy chain also plays an important role in determining the stability of the mesophase. As can be seen from the graph, when the alkanoyloxy chain length was increased by two methylene groups, from the C10 to the C12 member, it caused the compound to exhibit an enantiotropic (stable) SmA phase. Instead of showing a SmA phase during the cooling scan alone (monotropic), the SmA phase was observed during both the heating and cooling scans (enantiotropic). Further lengthening of the carbon chain by another two methylene groups stabilized the SmA phase where the SmA phase range increased from 4.8 oC for the C12 member to 8.4 oC for the C14 member. The tetradecanoyloxy chain was found to generate the widest SmA phase range in this series as further lengthening of the chain to the C16 member resulted in the SmA phase range decreasing to 5.0 oC. The decrease of the phase range was noticed as the chain length kept increasing. This was proven by the monotropic SmA phase which was observed in the C18 member. Certain extent of flexibility is essential for promoting liquid crystal phase, however, the continuous increasing of the carbon chain length of a compound will cause the molecule to be too flexible hence reducing the stability of its mesophase (in term of phase range) or even completely diminishing the mesophase [14].

Sie-Tiong Ha et al.

(a)

(c)

(e)

(b)

(d)

(f)

Figure 2. Optical photomicographs of benzothiazole liquid crystals obtained from microscopy technique. (a) 10BSP exhibiting smectic A phase, emerging as b창tonnet upon cooling from isotropic liquid at 69 oC; (b) 16BSP exhibiting smectic A phase with fan-shaped and homeotropic textures at 74 oC; (c) CH3O-10BSP exhibiting nematic phase with disclination lines at 89 oC; (d) CH3O-10BSP exhibiting transition from nematic phase (top right corner) to SmC (bottom left) phase upon further cooling at 55 oC; (e) CH3O-16BSP-OH exhibiting nematic droplets at 122 oC; (f) CH3O16BSP-OH exhibiting nematic phase with marble texture at 97 oC.

Benzothiazole as structural components in liquid crystals

Series CH3O-nBSP and CH3O-nBSP-OH In order to vary the mesomorphic behaviours of the previous series, a terminal methoxyl group was introduced into the molecular fragment. Here, two homologues series of benzothiazole derivatives comprising a terminal 6-methoxyl substituent were synthesized. A plot of the transition temperatures against the number of carbons in the alkanoyloxy chain for series, CH3O-nBSP and CH3O-nBSP-OH, are given in Figures 3a and 3b, respectively.

(a)

(b) Figure 3. Graph of transition temperatures against the number of carbon atoms (n) in the alkanoyloxy chain of (a) series CH3O-nBSP and (b) series CH3O-nBSP-OH. Cr, SmC, N and I denote the crystalline, smectic C, nematic and isotropic phases, respectively.

Sie-Tiong Ha et al.

In the case of series nBSP and CH3O-nBSP, liquid crystal phase was observed beginning from the shorter chain derivatives (n ≥ 4) in CH3O-nBSP as compared to series nBSP (whereby the liquid crystal phase was only observed for the longer chain derivatives, n ≥ 10). As for series nBSP-OH and CH3O-nBSP-OH, the effect of introducing a terminal methoxyl group was rather obvious. In the previous section, it has been mentioned that series nBSP-OH was non-mesogenic. However, liquid crystal phase was successfully generated in CH3O-nBSP-OH by incorporating a methoxyl group into the parent molecular fragment. For series CH3O-nBSP, short chain members (n = 2 and 3) were nonmesogenic compounds while the medium chain members (n = 4 to 8) exhibited a nematic phase. Both nematic and SmC phases were observed for the long chain members (n = 10, 12, 14, 16, 18). For series CH3O-nBSP-OH, no liquid crystal phase was observed for the C2 and C3 members, however, a nematic phase was observed in the remaining members of this series. Optical photomicrographs of CH3O-10BSP taken during the cooling cycle are depicted in Figures 2c-d. Meanwhile, optical photomicrographs of CH3O-16BSP-OH showing nematic phase with marble texture are given in Figures 2e-f. According to Figure 3a, the odd-even effect on the mesomorphic properties of CH3O-nBSP was not obvious in this series. The short chain derivatives (n = 2 and 3) were non-mesogenic compounds owing to the excessive rigidity in their molecules thus unfavouring mesophase formation. As for the C4, C5, C6, C7 and C8 members, their terminal chain lengths are long enough for promoting mesophase (nematic phase) formation. As the alkyl chain length increased by another two methylene groups to the C10 member, a monotropic (metastable) SmC phase was observed. Similarly, once the carbon chain reaches a certain length (n ≥ 12), enantiotropic (stable) SmC phase which accompanied by nematic phase at a higher temperature was induced. The melting points exhibited a descending trend from the C2 to the C12 member, and then increased from the C12 to the C18 members. As for the clearing temperature, it demonstrated a descending trend owing to the mesogenic core dilution resulting from the flexibility provided by the increased terminal alkanoyloxy chain [15]. On the other hand, the nematic phase range (ΔN) apparently decreased as the alkyl chain length increased. This resulted from the long carbon chain being attracted and intertwined which in turn facilitated the lamellar packing causing a decrease in the nematic phase range. However, by increasing the length of the carbon chain, the SmC phase range (ΔSmC) did not exhibit the usual trend whereby ΔSmC should increase. The ΔSmC was found to be 7.6 oC for the C12 member and it decreased to 0.9 oC

Benzothiazole as structural components in liquid crystals

for the C14 member, and increased again to 4.6 oC and 9.2 oC for the C16 and C18 members, respectively. The increase in the Van der Waals forces led to the increase of the melting temperature from the C12 to the C18 members which in turn caused an unusual smectogenic phase range in CH3O-nBSP. Nevertheless, it still can be deduced that the smectic phase stability was enhanced where the SmC-to-N transition temperature increased from the C10 to the C18 derivative. Accordingly (Figure 3b), the melting temperature (TCr-N) of CH3OnSBP-OH was considerably reduced with ascending chain length owing to the increase in its structural flexibility. On the other hand, the clearing temperatures (TN-I) descended as the number of carbon atoms increased, resulting from the dilution of the core system. A noticeable reduction in the nematic phase range was observed as the alkyl chain length increased. The nematic phase was generally exhibited by compounds possessing short to medium length chains. As the chain length increased, the nematogenic properties decreased and were accompanied by reduced nematic phase stability, therefore leading to a decrease in the phase range. The nematic phase will diminish if the length of the chain kept increasing.

3. XRD studies of benzothiazole liquid crystals The presence of SmA phase in series nBSP and CH3O-nBSP was further studied through power X-ray diffraction (XRD) analysis. The XRD pattern of the representative compounds 14BSP and CH3O-12BSP are shown in Figure 4 while their XRD data are given in Table 1. Table 1. Power X-ray diffraction data of 14BSP and CH3O-12BSP.

2 theta (°) d-spacing L d/L Phase

14BSP 2.35 35.52 Ǻ 31.57 Ǻ 1.13 SmA

CH3O-12BSP 2.48 30.81 Ǻ 31.48 Ǻ 0.98 SmC

In Figure 4a, the XRD patterns of 14BSP showed a sharp diffraction peak at 2.35 o, implying the formation of a layered structure which is typically characteristic of a layer structure observed for a smectic phase [16]. Generally, a sharp and strong peak at a low angle (1o < 2θ < 6o) in a small angle X-ray scattering curve is observed for smectic structures, but not seen in nematic and cholesteric structures.

Sie-Tiong Ha et al.

It is common that when the reflection within the SmA layer corresponds to d~L (d is layer spacing and L is molecular length), the smectic A layer is called the monolayer [17]. However, when the d-layer spacing is an intermediate between L and 2L, then it is called a partial bilayer phase. The d-layer spacing upon cooling of 14BSP from its isotropic liquid is 35.52 Ǻ whereas the molecular length (L) obtained by MM2 molecular calculation is 31.57 Ǻ. This provides the evidence for a partial bilayer structure of the smectic layers (L < d < 2L). The d/L ratio of 14BSP (1.13) falls within the range between 1.12 to 1.20 for partial bilayer arrangement [18]. The smectic phases of compounds CH3O-nBSP were further studied by temperature-dependent XRD analysis to confirm their SmC nature as well as

(a)

(b) Figure 4. XRD diffractograms of (a) 14BSP and (a) CH3O-12BSP.

Benzothiazole as structural components in liquid crystals

to determine the layer spacing at the operating time upon cooling. In Figure 4b, the presence of a sharp diffraction peak at a low angle (2.48o) confirmed that the layer periodicity is present in the mesophase structure whilst the broadly diffuse signal at the wide angle region indicated the short range order typical of the nematic phase [19]. However, for a nematic phase, no peak appears at a small angle and a broad peak at 2Î¸ â&#x2030;&#x2C6; 20o can be observed in the XRD diffractogram. Upon combining the results from the polarized optical microscopy and XRD analysis, the presence of the SmA and SmC phases in both the series can been confirmed.

4. Structure-liquid crystal property relationships Here, comparisons between the title compounds with other structurally related compounds are presented and the structure-property relationships were discussed. The phase transition temperatures of benzothiazole derivatives and reference compounds are presented in Tables 2-5, where Cr, SmA, SmC, N, I, denote the crystalline, smectic A, smectic C, nematic and isotropic phases, respectively. Values given in parentheses refer to monotropic phase transitions. Table 2. Effect of linking group on mesomorphic properties. Compounds

Structures and transition temperatures (oC) O

S N

12BSP

C11H23

Cr 80.8 SmA 85.6 I N OC 12H25

A [20]

Cr 86.9 (SmA 81.2) I O

S N

CH3O-12BSP

H3CO

C11H23

Cr 65.6 SmC 73.1 N 114.3 I S N

[21]

H3CO

OC12H25

Cr 113.0 (SmA 108.0) N 135.0 I

Sie-Tiong Ha et al.

Table 3. Effect of lateral hydroxyl group on mesomorphic properties. Average mesophase range (oC)

Compounds

Average thermal stability (oC)

SmA

SmC

6.0

97.1

89.5

86.7

35.0

5.6

89.0

119.3

11.1

136.8

142.2

N N

nBSP O

N N

nBSP-OH O

N N H3CO

CH3O-nBSP O

N H3CO

CH3O-nBSP-OH

Effects of linking group on mesomorphic properties A summary of the structures and transition temperatures of compounds 12BSP, A, CH3O-12BSP and B are tabulated in Table 2. Upon comparison between the structures of 12BSP and compound A, differences are in the Schiff base linkage between the mesogenic cores and the ester linkage in the terminal chain. The stepped core structure in 12BSP caused by the Schiff base linkage resulted in a more stable SmA phase whereby an enantiotropic SmA was observed in 12BSP while monotropic SmA was found in compound A. Another example can be seen from the comparison between CH3O-12BSP and compound B. Although both compounds have the stepped core structure due to either the Schiff base or azo linkage, an enantiotropic smectic phase was only observed in CH3O-12BSP. This was probably due to the presence of the ester linkage in the terminal chain. The existence of the ester linkage instead of the ether linkage increases the length of the core unit which in turn favors the lamellar packing in the liquid crystal phase, hence stabilizing the smectic phase.

Benzothiazole as structural components in liquid crystals

Table 4. Effect of terminal substituent group on mesomorphic properties. O

S N

C11H23

12BSP O

N N H3CO

C11H23

CH3O-12BSP N N Cl

OC12H25

compound C N N O 2N

OC12H25

compound D Compounds

Transition temperatures ( C)

Mesophase range (oC) Sm

12BSP

Cr 80.8 SmA 85.6 I

4.8

CH3O-12BSP

Cr 65.6 SmC 73.1 N 114.3 I

7.5

41.2

C [22]

Cr 80 SmA 172 I

38.0

D [23]

Cr 156 SmA 194 I

92.0

Effects of lateral hydroxy group on mesomorphic properties In order to understand the effect of the lateral hydroxyl group on the mesomorphic properties, the molecular structures and transition temperatures of series nBSP, nBSP-OH, CH3O-nBSP and CH3O-nBSP-OH are listed in Table 3. It can be noticed that series nBSP exhibited a SmA phase while series nBSP-OH is a non-mesogenic series. The presence of the lateral hydroxy group in series nBSP-OH prohibited the formation of mesophase by increasing the molecular broadness. It is known that the length-to-breadth ration is crucial in generating a liquid crystal phase. The presence of the lateral hydroxyl group has broadened the molecule, hence decreasing the molecular length-to-breadth ratio.

Sie-Tiong Ha et al.

Table 5. Effect of mesogenic core on mesomorphic properties. Compounds

Structures and transition temperatures (oC) O

S N

14BSP

C13H27

Cr 82.3 SmA 90.7 I N

E [24]

OC14H29

Cr 97 I O

S N

CH3O-16BSP-OH

H3CO

C15H31

Cr 123.2 N 129.7 I O H3CO

C15H31 O

F [25] HO

Cr 98 SmA 111 I

Sometimes, the presence of a lateral hydroxyl group may also inhibit the formation of smectic phase while favoring the formation of other phases. By comparing series CH3O-nBSP and CH3O-nBSP-OH, we found that the analogues bearing the lateral hydroxy group exhibited only a nematic phase. A broader core unit (CH3O-nBSP-OH) is more favourable in nematic phase formation. Compounds with a lateral hydroxyl group normally exhibited higher melting and clearing temperatures. From the comparison between series CH3O-nBSP and CH3O-nBSP-OH, it can be seen that both melting (TM) and clearing (TC) temperatures are higher in series CH3O-nBSP-OH. The Schiff base linkage may form intramolecular hydrogen bonding with the o-hydroxyl group, hence, increases the melting and clearing temperatures [26]. Effects of terminal substituent group on mesomorphic properties A terminal group can be introduced into the molecules to alter their mesomorphic properties. A summary of the molecular structures and transition temperatures of compound 12BSP, CH3O-12BSP, C and D are tabulated in Table 4.

Benzothiazole as structural components in liquid crystals

The unsubstituted benzothiazole derivative, 12BSP, possessed the shortest mesophase range (4.8 oC) among these compounds. The introduction of a polar methoxyl terminal group into the benzothiazole derivative (CH3O-12BSP), increased the mesophase range drastically from 4.8 oC to 48.7 oC (total for smectic and nematic phases). Incorporating a smaller and more polar chlorine atom into compound C further increased the melting (80 oC) and clearing (172 oC) temperatures of the compound to a certain extent. The highest melting and clearing temperatures were observed for the nitro-substituted benzothiazole derivative (compound D), exhibiting 156 oC and 194 oC as melting and clearing temperatures, respectively. Effects of mesogenic core on mesomorphic properties Table 5 summarizes the molecular structures and transition temperatures of compounds 14BSP, E, CH3O-16BSP-OH and F. Comparison between these compounds enables us to understand the effects of the benzothiazole core on their mesomorphic properties. According to Table 5, 14BSP, bearing a benzothiazole core unit, exhibits an enantiotropic SmA phase while compound E with a naphthalene core is a non-mesogenic compound. The benzothiazole ring structure, incorporated with electronegative heteroatoms N and S, results in a reduced symmetry in the overall molecule, thus generating a stronger polarity. The induction of polarity by the N and S atoms on these heterocyclic cores may be responsible for the formation and enhancement of mesophases [3]. The comparison between CH3O-16BSP-OH and compound F has revealed that the benzothiazole core was also responsible in the formation of the nematic phase. It can be seen that CH3O-16BSP-OH exhibited only a nematic phase whereas compound F exhibited a SmA phase. Compared to the single benzene ring, the fused-ring structure of the benzothiazole core enhanced the molecular polarizability in turn increasing the intermolecular cohesive forces which induced the formation of the nematic phase [23]. Besides, the melting and clearing temperatures of CH3O-16BSP-OH were found to be slightly higher than that of compound F. As a result of its enhanced polarizability and increased intermolecular cohesive forces, the thermal stability of CH3O-16BSP-OH was increased, resulting in higher melting and clearing temperatures.

4. Summary Four homologues series of compounds were successfully synthesized. Only one series did not exhibit liquid crystal phase. The liquid crystalline

Sie-Tiong Ha et al.

properties of the remaining three series were thoroughly investigated where by nematic, SmA and SmC phases were observed in those compounds. Through a series of comparisons, structure properties relationship was established in which the effects of the linking group, lateral hydroxyl group, terminal group and mesogenic core were discussed. The information presented here may lead to a better understanding of the structure-properties relationship, making it possible to fine tune the desired properties of the compounds.

Acknowledgements The authors would like to thank the Universiti Tunku Abdul Rahman, Malaysia Toray Science Foundation, MOHE and MOSTI for the financial supports.

References 1. 2.

4. 5. 6.

Collings, P.J., Hird, M. 1998, Introduction to Liquid Crystals: Chemistry and Physics, Taylor & Francis Ltd., London, UK; Khoo, I.C. 2007, Liquid Crystals, John Wiley & Sons Inc., New Jersey, USA. Srividhya, D., Manjunathan, S., Thirumaran, S. 2009, E-Journal of Chem., 6, 928; Srividhya, D., Manjunathan, S., Thirumaran, S., Saravanan, C., Senthil, S. 2009, J. Mol. Struct., 2009, 927, 7; Tsai, H.-H. G., Chou, L.C., Lin, S.C., Sheu, H.S., Lai, C.K. 2009, Tetrahedron Lett., 50, 1906; Huang, R.T., Wang, W.C., Yang, R.Y., Lu, J.T., Lin, I.J. 2009, Dalton Trans., 35, 7121; Kozhevnikov, V.N., Cowling, S.J., Karadakov, P.B. , Bruce, D.W. 2008, J. Mater. Chem., 18, 1703; He, C.F., Richards, G., Kelly, S., Contoret, A., Oâ&#x20AC;&#x2122;Neill, M. 2007, Liq. Cryst., 34, 1249; Kozhevnikov, V.N., Whitwood, A.C., Bruce, D.W. 2007, Chem. Commun., 3826; Fan, R., Malliaras, G., Sukhomlinova, L., Gu, S., Twieg, R.J. 2000, American Physical Society, Annual March Meeting, Minneapolis, abstract #G21.009; Petrov, V.F. 2006, Mol. Cryst. Liq. Cryst., 457, 121; Petrov, V.F. 2005, Mol. Cryst. Liq. Cryst., 442, 51; Petrov, V.F., Pavluchenko, A.I., 2003, Mol. Cryst. Liq. Cryst., 393, 1; Petrov, V.F., Pavluchenko, A.I., 2003, Mol. Cryst. Liq. Cryst., 393, 15; Petrov, V.F. 2001, Liq. Cryst., 28, 217; Bezborodov, V.S., Petrov, V.F., Lapanik, V.I. 1996, Mol. Cryst. Liq. Cryst., 20, 785; Titov, V.V., Pavlyuchenko, A.I. 1980, Chemistry of Heterocyclic Compounds, 16, 1. Seed, A. 2007, Chem. Soc. Rev. 36, 2046; Lai, C.K., Liu, H.C., Li, F.J., Cheng, K.L., Sheu, H.S. 2005, Liq. Cryst. 32, 85. Zhang, B.Y., Jia, Y.G., Yao, D.S., Dong, X.W. 2004, Liq. Cryst., 31, 339. Majumdar, K.C., Mondal, S., Ghosh, T., 2010, Mol. Cryst. Liq. Cryst., 524, 17; Gipson, R.M., Sampson, P., Seed, A.J., 2010, Liq, Cryst., 37, 101; Hird, M., Toyne, K.J., Goodby, J.W., Gray, G.W., Minter,V., Tuffin, R.P., McDonnell

Benzothiazole as structural components in liquid crystals

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

D.G. 2004, J. Mater. Chem., 14, 1731; Eichhorn, S.H., Paraskos, A.J., Kishikawa, K., Swager, T.M., 2002, J. Am. Chem. Soc., 124, 12742; Campbell, N.L., Duffy, W.L., Thomas, G.I., Wild, J.H., Kelly, S.M., Bartle, K., O'Neill, M., Minter, V., Tuffin, R.P. 2002, J. Mater. Chem., 12, 2706. Prajapati, A.K., Modi, V. 2010, Liq. Cryst., 37, 407; Han, J., Zhang, Y.F., Wang, J.Y., 2010, Key Eng. Mater., 428, 52; Parra, M.L., Elgueta, E.Y., Jimenez, V., Hidalgo, P.I., 2009, Liq. Cryst., 36, 301; Zhang, P., Bai, B., Wang, H., Qu, S., Yu, Z., Ran, X., Li, M., 2009, Liq. Cryst., 36, 7; Parra, M.L., Hidalgo, P.I., Elgueta, E.Y., 2008, Liq. Cryst., 35, 823; Sparavigna, A., Mello, A., Montrucchio, B., 2008, Phase Trans., 81, 471; Karamysheva, L.A., S.I. Torgova, Agafonova, I.F., Petrov, V.F. 2000, Liq. Cryst. 27, 393. Liao, C.C., Wang, C.S., Sheu, H.S., Lai, C.K., 2008, Liq. Cryst., 64, 7977; Wang, H.C., Wang, Y.J., Hu, H.M., Lee, G.H., Lai, C.K., 2008, Tetrahedron, 64, 4939; Lai, C.K., Liu, H.C., Li, F.J., Cheng, K.L., Sheu, H.S. 2005, Liq. Cryst., 32, 85; Wang, C.S., Wang, I.W., Cheng, K.L., Lai, C.K., 2006, Tetrahedron, 62, 9383. Aldred, M.P., Vlachos, P., Dong, D., Kitney, S.P., Tsoi, W.C., Oâ&#x20AC;&#x2122;Neill, M., Kelly, S.M. 2005, Liq. Cryst., 32, 951. Funahashi, M., Hanna, J. 1997 Phys. Rev. Lett., 78, 2184. Tokunaga, K., Takayashiki, Y., Iino, H., Hanna, J. 2009, Phys. Rev. B, 033201, 1. Tokunaga, K., Iino, H., Hanna, J. 2009, Mol. Cryst. Liq. Cryst., 510, 241. Tokunaga, K., Takayashiki, Y., Iino, H., Hanna, J. 2009, Mol. Cryst. Liq. Cryst., 510, 250. Koden, M., Yagyu, T., Takenaka, S., Kusabayashi, S. 1983, J. Phys. Chem., 87, 4730. Berdague, P., Bayle, J.P., Ho, M.S., Fung, B.M. 1993, Liq. Cryst., 14, 667. Wang, Y., Zhang, B.Y., He, X.Z. 2007, Colloid Polym. Sci., 285, 1077. Liao, C.C., Wang, C.S., Sheu, H.S., Lai, C.K. 2008, Tetrahedron, 64, 7977. Reddy, R.A, Sadashiva, B.K. 2004, J. Mater. Chem., 14, 310. Pal, S.K., Raghunathan, V.A., Kumar, S. 2007, Liq. Cryst., 34, 135. Ha, S.T., Koh, T.M., Yeap, G.Y., Lin, H.C., Boey, P.L., Win, Y.F., Ong, S.T., Ong, L.K. 2009, Mol. Cryst. Liq.Cryst., 506, 56. Prajapati, A.K., Bonde, N.L. 2006, J. Chem. Sci., 118, 203. Prajapati, A.K., Bonde, N.L. 2005, Proc. of 11th National Conference on Liquid Crystals, Allahabad, 114. Prajapati, A.K., Bonde, N.L. 2009, Mol. Cryst. Liq.Cryst., 501, 72. Vora, R.A., Prajapati, A.K. 1998, Liq. Cryst., 25, 567. Yeap, G.Y., Ha, S.T., Boey, P.L., Mahmood, W.A.K., Ito, M.M., Youhei, Y. 2006, Mol. Cryst. Liq.Cryst., 452, 73. Yeap, G.Y., Ha, S.T., Ishizawa, N., Suda, K., Boey, P.L., Mahmood, W.A.K. 2003, J. Mol. Struct., 658, 87.

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 19-52 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

2. An overview of liquid crystals based on Schiff base compounds 1

Shankar B. Rananavare1 and V.G.K.M. Pisipati2 Department of Chemistry, Portland State University, Portland, OR 97206, USA 2 Liquid Crystal Research Centre, ECE Department, Koneru Lakshmaiah University, Vaddeswaram, 522 502, India

Abstract. In the mid sixties, the pioneering research efforts at RCA and Xerox brought attention to possibility of using a relatively unknown class of organic materials, i.e., liquid crystals (LC’s), for modulating electro-optic properties of light for fabricating a revolutionary new flat panel display. Today the flat panel liquid crystal display devices, (LCD’s) have replaced the bulky and power-hungry cathode ray tubes (CRTs) and thus changed the face of the modern office. In this review, we focus on the first class of pure synthetic liquid crystals based on N-(p-nalkoxy benzylidene)-p-n-alkyl anilines, popularly known as nO.m Schiff’s base compounds, where the n and m represent the chain lengths on either side of the rigid core. This homologous series exhibits rich but subtle poly mesomorphism (exhibition of different liquid crystalline phases) as functions of chain length, isotopic substitution of hydrogen with fluorine, insertion of optically active groups and metals. Further, it has been observed that these materials exhibit different clearing temperatures as well as liquid crystal phase modifications depending on the placement and/or the

11 Correspondence/Reprint request: Dr. Shankar B. Rananavare, Department of Chemistry, Portland State University, Portland, OR 97206, USA. E-mail: ranavas@pdx.edu

Shankar B. Rananavare & V.G.K.M. Pisipati

removal of the electronegative oxygen atom on either side of the rigid benzylidene core. That is, these compounds can be identified as nO.m, n.Om, nO.Om and n.m series. In this review, we briefly cover interesting phase variants, the nature of different phase transitions and phase diagrams of the above four series of homologous series. The main focus lies on the nature of isotropic – nematic (IN), nematic smectic-A, (NA), smectic–A – smectic-C, (AC) and other higher order smectic to smectic phase transformations. In addition, we provide discussions on multicritical points such as NA tricritical point (TCP) and ACC* Lifshitz point. The occurrence of such multi-critical points permits predictions of physical properties that are important in material optimization for device applications. Finally we conclude with a new class of amphitropic liquid crystals.

Introduction I. Historical perspective Today the use of LCDs is norm than rarity as it was even 10 years ago. The technological breakthroughs that have led to replacement of ubiquitous cathode ray tubes (CRTs) stretch as far as the mid-sixties when the notion of flat panel display that would consume far less power and space appeared in Radio Corporation of America (RCA) labs [1, 2]. Here, the basic idea was to modulate linearly polarized light transmission through an optically anisotropic medium such as nematic liquid crystal by means of applied electric fields. Thus, these liquid crystals acted as voltage controlled optical valve similar to MOSFET (metal oxide semiconductor field effect transistors). Early segmented direct-drive liquid crystal displays [3] found immediate home in calculators due to their low power consumption. To improve upon cumbersome direct drive approach, a matrix addressing scheme was proposed by Alt and Pesko [4] at IBM and was subsequently elegantly implemented [5] in the STN (super twisted nematic) type of devices developed by Scheffer et al [6]. However, the mid-nineties saw a rapid decline in use of STN displays owing to their inability to address larger area displays arising from inter-pixel cross-talk effects. Today, the key to the dominance of LCD technology is its marriage with the classic semiconductor technology through the use of thin film transistors (TFTs) to address individual pixels. TFT based addressing makes it possible design an entirely digital display. The development of thin film transistors for this application traces its origin to back to the legendary Palo Alto Research Center (PARC) at XEROX [7]. As matter of fact, Xerox had an essential prototype of modern digital LCD device incorporating TFTs as early as early eighties. The main technological hurdle, at the time, was the ability to mass-produce thin film transistors on large area substrates. Herein lies the crucial difference in

An overview of liquid crystals based on Schiff base compounds

semiconductor and display industry; the former employs die sizes of few square inches while for the latter the die size is in square feet. Early nineties saw a brief but unsuccessful resurgence of interest in the LCD manufacturing [8] in US guided by ARPA (advanced research projects agency) to establish strategic alternatives to Asian manufacturers. The display market still continues to be dominated by Asian electronic giants such as Sony and Sharp. Mid nineties also saw entry of Samsung in this market who now has now established the largest market share in LCD market [9]. To be successful display manufacturer, clearly, requires not only an expertise in liquid crystals cell processing but also requires an expertise in semiconductor manufacturing. Besides information displays, that thrive on high pixel density, liquid crystals have found many wide ranging applications spanning non-linear optical devices [10, 11] to anisotropic solvents in chemistry [12-14]. Displaytech has commercialized fast spatial light modulators using ferroelectric liquid crystals (FLCs) [15, 16]. In addition, Kent system developed an ultrathin newspaper//book using polymer dispersed and cholesteric liquid crystals [17]. It is not too difficult to imagine the future applications of liquid crystals for bottom-up self assembly methods for nanotechnology [14, 18, 19] as the limitations of currently actively pursued self-assembled monolayer (SAM) methods [20] such as high defect density and inability to work over large surface area are fully realized. Although in all fairness, besides LCD displays, the second commercially successful application of liquid crystal based materials from a technological viability perspective is yet to emerge. Therefore, in this review we will focus mainly on materials properties that are useful in developing better displays while noting along the way potential future applications in other areas.

II. Display relevant LC properties The fundamental properties of display relevance of the most commonly used nematic liquid crystals are their orientational elasticity and alignability on wide area substrates [21]. To align liquid crystal molecules parallel to surface (the so called homogeneous alignment) rubbed polymer surfaces or grooved surfaces are used; the macroscopic alignability of LC is a result of elastic energy minimization on such surfaces. In the absence of electric and magnetic fields, lying parallel to grooves is preferred on a macroscopic scale. This molecular level alignment can be perturbed, nonlinearly (switching) through application of modest electric field. Interplay of surface and electrically field (controlled alignment due to dielectric constant anisotropy

Shankar B. Rananavare & V.G.K.M. Pisipati

(Δε)) permits fabrication of display devices of directionally aligned fluid on length scales of feet! There exists a threshold voltage at which molecular director begins to align with respect to (or normal to) the applied electric field at the Freederick’s transition. For example, a 90 degree twisted nematic LCD cell has a threshold voltage that is fortunately independent of electrode spacing (LC film thickness) and is give by VTh = π [(K11+K33-2K22)/(ε0Δ ε )]1/2; where K11, K33, K22, are elastic constant associated splay, bend and twist distortion of LC director [22] and ε0 is the permittivity of vacuum. The optical transmission through LC cells can be turned on or off by electrically switching molecular alignment with respect to incoming plane polarized light. This can be modeled using extended Jones matrix formalism. For a simple homogenously aligned LC cell the optical path length (~Δn(birefringence)) affecting the device contrast is highly dependent on the viewing angle. This effect was responsible for poor viewing angle characteristics of early LCDs which have been overcome in last two decades using vertical nematic cells and in-plane switching [23, 24]. One of the main limitations of LCD is their relatively low contrast due to light loss in polarizers that affect their bright on state and leakage of light through the off state. Compared to CRTs or LED displays optical switching of the LCDs is slow. In the absence of electric field, pixel switching time scales directly as the viscosity(η) of nematic liquid crystal and inversely as the orientational elastic constant (s), Kii [22]. Geometrically, the switching time, τ, varies as a square of the gap thickness (d), so overall τ~ ηd2/Ki) [22]. Making devices thinner does enhance switching speed but too small thickness lead to degradation of the device contrast which is related to reduction in optical path length. To reduce viscosity, use of fluorinated functional groups [25] as well as chain unsaturation [26] has been explored. These important considerations dictate the materials and device optimization programs.

III. Molecular engineering of LCs for flat panel display applications Molecular factors affecting elastic constants, optical birefringence, switching time have been continually explored both theoretically and experimentally. As it turns out many of these fundamental properties of nematic liquid crystals, Δn, Δε , Kii can be related to the degree of molecular orientational order (S=<P2(cos(θ)>, where P2 is a second Legendre polynomial; and θ is the angle describing instantaneous orientation of molecular long axis and the direction of average alignment of molecules, the director ň [21]). Orientational order, S, exhibits a strong temperature

An overview of liquid crystals based on Schiff base compounds

dependence near the nematic to isotropic phase transition (to the so called clearing temperature as nematic liquid crystals in bulk are cloudy and become transparent upon undergoing a phase transition into an isotropic phase). From device engineerâ&#x20AC;&#x2122;s point of view, in addition to having a wide operational temperature range, the clearing temperature must be much higher than room temperature in order to have stable switching performance. Lower temperatures operation of the organic liquid crystals molecules is limited by phase transition into solid or other low symmetry smectic phases. Thus, unlike EL, LED, plasma or field emission displays, the range of operational temperatures of LCDs is narrower and is continually enhanced through strong synthesis efforts and through use of mixtures of liquid crystals. The study of the phase diagram of mixtures of LC focuses on locating the lowest temperature eutectic in mixture of 6-12 compounds! The compositional parameter space has to be large so that the process window, encompassing temperature range, contrast ratio, stable threshold voltage and switching speed, can be optimized. The molecular aspects of phases and phase diagrams of liquid crystals are the primary focus of this review. Although this review focuses on LC materials and properties it should borne in mind that LC-substrate interaction are also of paramount importance for device performance optimization. Use of rubbed polyimide allows the so called planar alignment (molecular long axis lying parallel to surface) while surfactants enable vertical alignment (homeotropic alignment) of molecules with respect to electrodes [21]. Introduction of surface coatings over the electrode also helps to overcome a major issue in device incorporation of liquid crystals: electrolysis/electrochemistry at the electrode-LC interface. During the sixties and seventies chemical degradation issues of liquid crystals were overcome through novel synthetic methods. Schiffâ&#x20AC;&#x2122;s bases were one of the first pure single component room temperature nematic liquid crystals to be synthesized [27]. Their synthesis involves a simple condensation reaction between an amine and an aldehyde. Typically both the species contain aromatic benzene ring and also alkyl chains connected to the aromatic rings either directly or through an ether linkage. For a variety reasons these interesting class of compounds were quickly replaced by cyano-biphenyls developed by Gray and coworkers [28] for LCD device applications. The cyano-biphenyls have better chemical stability especially with respect to hydrolysis and oxidation. Nevertheless, the ease of synthesis and hence low cost of the Schiff base compounds made them very attractive for the fundamental studies of liquid crystals. These materials have been proving grounds for many spectacular theoretical prediction of new classes of liquid crystals such as ferroelectric liquid crystals [29, 30]. This review therefore focuses on the Schiff base liquid crystals where we discuss

Shankar B. Rananavare & V.G.K.M. Pisipati

how the fundamental phases are modulated as a function of molecular architecture, more specifically length of alkyl and alkoxy chain lengths [31] as well as derivatives containing only alkyl chains [32], alkoxy chain length in both segments of molecule [33, 34], in the aldehydic and amine portions of molecules. What is clear is that presence of alkoxy chain [35] creates a distinct tilt of the chain axis along with introduction of significant off- axis dipole moment.

IV. Theoretical aspects Molecular level interactions that lead to formation of nematic phases are molecular anisotropy in shape and induced dipolar interactions [21, 36-38]. Onsager developed a statistical mechanical theory [21, 39] based on molecular packing considerations for rod-like molecules. Onsagerâ&#x20AC;&#x2122;s theory has been extremely successful in predicting lyotropic nematic phase made of stiff rod like molecules such as DNA[40-42], peptide/polymers[43, 44] etc. For smaller size molecules, which exhibit thermotropism, molecular field theory of nematic phase was first developed by Mayer-Saupe [36-38]. A deeper molecular level connection to the elastic constants to the mean field potential was developed later [45-47]. Subsequent exposition of mean filed model for lower temperature smectics by McMillan [48, 49] in mid seventies led to the foundation of molecular engineering aspects of liquid crystals. McMillan and de Gennes emphasized coupling of oriental and positional ordering of molecules which leads to interesting crossover in the order of the nematic to smectic A (NA) phase transition from first order to second order through a tricritical point. Our experimental studies [50-52] in eighties verified this basic theoretical picture although the issue of true second order NA transition has been controversial as the nematic order fluctuations lead to a cubic term in the order parameter expansion as was pointed in early seventies [53-55]. Practically, the nature of this transition is important is due to fact that it is common to find nematic phase transforms into a smectic phase at low temperature as opposed to freezing into solid phase. Observed divergence of twist and bend elastic constant as the second order smectic A phase is approached is problematic from device physics point of view as it tends to affect threshold voltage and switching speed of LC devices. Further richness of phase topology of liquid crystals was realized with an introduction of chiral center in alkyl chain portion which Meyer predicted should lead a ferroelectric liquid crystals phase of C2 symmetry. This phase was discovered in Schiff base compounds during seventies [29, 30, 56]. These ferroelectric liquid crystals exhibit net non zero molecular dipole in the plane of smectic liquid crystal layers. Prost developed beautiful theory on

An overview of liquid crystals based on Schiff base compounds

liquid crystal structures that emphasized longitudinal dipole moments [57]. However, these theoretical developments were strictly based on Landau or order parameter based description of phases and phase transitions of liquid crystals. Once again these considerations lead us away from molecular features to a more general symmetry arguments as developed by De-Gennes which are beyond the intended focus of this narrow review. The following is the brief outline of the review. We begin with a description of phase behavior in classic Schiff base based liquid crystals [58-60] architectures as illustrated in figure 1. We first consider the phase variants observed as a function of chain length and compare the results with molecular field theoretic predictions. This is followed by a brief discussion of phase transitions that affect display related properties. Thermodynamics of phase transition is developed in light of prevailing theories especially emphasizing the Landau-De Gennes approach. Interesting symmetry arguments that led to prediction and discovery of ferroelectric liquid crystals based on Schiff base molecules are presented next along with our contributions to the field. A new synthetic variation of nO.m series is outlined based on introduction of highly polarizable SF5 group in the alkyl chain that features partial fluorination. Here the goal was to reduce LC viscosity and improve their optical birefringence properties. Finally we give an example, where classical thermotropic (nO.mâ&#x20AC;&#x2122;s) with lyotropic liquid systems are combined into the so called amphitropic liquid crystals.

Results and discussion The molecular structures of the liquid crystals reviewed are shown in Figure 1.

1.0. nO.m compounds Synthesis of these series of molecules is relatively straightforward. Alcoholic solutions of n-alkyl amine and m-alkoxy aldehyde are refluxed in presence of few drops of glacial acetic acid. After couple of hours of refluxing, the reaction mixture is stored overnight in refrigerator to crystallize the product which is filtered and washed with cold alcohol to purify the liquid crystal.

1.1. General comments In the molecular skeleton of nO.m homologous series, there are two electronegative atoms namely oxygen and nitrogen, which lead to non- uniform

Shankar B. Rananavare & V.G.K.M. Pisipati

Figure 1. Generic structures of Schiff base liquid crystals.

charge distribution that can be approximated as localized dipoles. It has been noted that molecules with polar alkoxy chain generally give rise to higher NI transition temperatures than one without (n.m series [35], see below) as has also been observed in the cyano-biphenyl series [22]. However, in the present series, magnitude as well as relative location of dipole moments is constant, thus dispersive and steric interactions are modulated by the variation in chain lengths at either terminus of the effectively rod-like molecules. This homologous series exhibits rich mesomorphism as shown in Table 1 below. The observed phases include nematic (N), smectic A (A), smectic C(C), smectic B(B), smectic F (F) and smectic G (G). The alkoxy and alkyl chain numbers are varied from 1 to 18 and 1 to 10, 12, 14, 16 respectively. The richest mesomorphism is exhibited by 5O.6. The salient and interesting features observed from the systematic study presented in Table 1 are [31]: 1. All the phase variants are drawn from the common phase sequence exhibited by the compound 5O.6, NACBFG (Figure 2) and this is the only compound which exhibits this hexa phases variant, 2. These compounds exhibit twenty one different types of phase sequence variants. These include different types of mono, di, tri, tetra, penta and hexa variants, There are three types of mono variants (N,A, F), five types of di variants (NA, NG, AB, AF, FG), five types tri variants (NAB,

An overview of liquid crystals based on Schiff base compounds

4. 5. 6.

NAG, ABF, ABG, AFG), five types of tetra variants (NACB, NACG, NABG, ACBG, ACFG), two types of penta variants (NACFG, NACBG) and one type of hexa variant (NACBFG) totaling twenty one different types of liquid crystalline phase sequence variants, Four different types of phase sequences are exhibited by single compounds viz, NACB by 4O.7, NACBG by 5O.7, NACFG by 5O.5 and NACBFG by 5O.6. The above compounds differ in one alkyl chain number. Addition and subtraction of one alkyl chain to 5O.6 gives two different types of penta phase variants which differ in B and F smectic where both are present in 5O.6, The dominant phase sequence variants are (â&#x2030;Ľ 20 compounds which exhibit the same sequence of phases) N (20), F (19), AB (30), FG (25), NAB (19), and ABG (29), For n and m<6, i.e., in the top left corner of the table corresponding to smaller molecular lengths nematic and smectic G phases predominate, For n>10 no nematic is seen while for m>10 nematic phase is observed as long as n<6. As m is varied in this series, the onset of orthogonal smectic-A at 6O.10 quenches the nematic phase for higher homologues, Table 1. phase variants exhibited by nO.m homologues series.

n/m 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 18

NA NG NAB NAB NAB NAB NA A A A A

4 N N N N NG NG NABG NG NG NAG NG NAG NABG NAB NAB NACG NAB NABG ABG AB AB AFG ABG AB ABG AB ABG AB AB AB ABG AB AB ABG AB ABG AB AB AB ABG AB AB AB AB AB AB

NAG NABG NACFG NABG NACG ABG ABG ABG ABG ABG ABG ABG ABG F F

NAG NAB NACBFG NABG ACFG ACBG ACFG ACFG AFG ABG AB ABG ABG F F

N N NACB NACBG NAB NACG ACBG ACBG ABG AFG AFG AB AB FG F FG

8 N N NA NAB NABG NABG ACBG ABG ACFG ACFG AFG AFG AF F F F FG

9 N N N NAB NAB NA ABG ABG ACBG ACBG ACBG FG FG F F FG F

10 N N NA NAB NAB AB AB ABF ACBG ACFG AFG AFG F F F F F

12 N N N NAB NAB ABG AB AB ABG AFG FG FG FG F F FG F

14 N NA N NAB NAB AB ABG ABG ABG FG FG FG FG FG FG FG F

16 N A N NAB NA ABG AB AB AB AFG FG FG FG FG FG FG FG

Shankar B. Rananavare & V.G.K.M. Pisipati

Figure 2. nO.m Phase sequence variant tree.

7. The 5O.m homologues are unique as all phase variants exhibited by nO.m compounds happen to be subset of their phase variants. This argument further derives support from the largest phase variant, NACBFG exhibited by 5O.6. This is depicted in the Figure 2, 8. The dominant single phase variants N and F are concentrated on the top right and down right portion of the table respectively. Mono-variant A phase is less prominent, 9. The phase variants with tilted phases, C and F are concentrated in the middle of the table, For n and m> 10, only tilted smectic mesophases F and/or G are observed. 10. The smectic-G phase is distributed throughout, 11. Like the single mono variants, N and F, the di phase variants AB and FG are equally populated among the nO.m compounds, 12. Even though the AB sequence is distributed on the left down and right up positions of the table while the FG is more concentrated on the down right of the table, 13. Orthogonal smectic phases, A and B are dominant compared to the tilted C and F phases,

An overview of liquid crystals based on Schiff base compounds

14. With the increase of alkoxy chain number the onset single phase variant smectic-F is observed with the small number of alkyl chain length. The minimum n and m numbers required for the manifestation single phase variant F are 13 and 10 respectively, while the onset smectic-F from isotropic melted with smectic-G occurred with the n and m values equal to 10 and 14 respectively, 15. The twenty one different phase sequence variants and their degeneracy in nO.m compounds are – N (20), A(5), F (19), NA (7), NG (6), AB (30), AF(2), FG (24), NAB (19), NAG (4), ABG (29), ABF (1), AFG(10), NACB (1), NACG (3), NABG (8), ACBG (8), ACFG (6), NACBG (1), NACFG (1), NACBFG (1) and 16. Most of the phase transitions in all the compounds are enantiotropic except in few cases and the clearing temperatures are < 100oC.

1.2. Specific comments 1.2.1. Nematic phase The early molecular theory of nematic phase, due to Meier-Saupe [36-38] introduces an induced dipole-induced dipole type interaction (having 1/r6 dependence) between smaller size nematogens. In addition, this interaction defines the orientational ordering in terms of second Legendre polynomial (S=<P2(cos(θ)>=<3cos2(θ)-1>/2)) in orientation (θ) of molecules with respect to the direction of average molecular alignment, the nematic director. For most of the nO.m compounds, this interaction should be essentially constant, as the polarizable portions of molecule C-O, C=N and aryl groups are fixed leading to a relatively constant TNI≈70-90˚C for most of the homologous series except for very small size molecules (n<2). It should be noted (see below) that generally a presence of longitudinal dipole moment (μL) on the molecular long axis (see below) leads to higher TNI. For example 5CB vs. 5OCB (pentyl vs. pentoxy chain attached to cyanobiphenyl group) differ in their respective clearing temperatures (TNI) by 35C, with the latter having the higher clearing temperature [22]. The odd and even chain effect predicted by Mercelja is also observed in variation of NI transition temperature [61]. Nematic phase disappears more rapidly with increase in the alkoxy chain length (n) than the alkyl chain length (m).

1.2.2. Smectic A phase For fixed n (or m) variation in the alkyl chain length m (n) leads to a reduction in the nematic phase extent and also to the appearance of an

Shankar B. Rananavare & V.G.K.M. Pisipati

orthogonal smectic A phase. In Smectic A phase molecules exhibit positional ordering that can be described as a one dimensional mass density wave, i.e., tendency to form layers. This propensity is consistent with the McMillanâ&#x20AC;&#x2122;s model of the smectic A phase [48]. McMillan described the positional ordering in terms of average values of Fourier component describing crystallographic positional order. Lowest order component for Smectic A phase is <Cos(2Ď&#x20AC;z/d>, where d is the layer thickness. This positional order disappears at the nematic to smectic A (NA) phase transition. This transition is perhaps most the celebrated phase transition in liquid crystal research as it can be either first or second order. It is possible to realize even higher order NA phase transition, beyond first or second order, a tricritical phase transition (see below). Both McMillan and De Gennes pointed out that the nematic to smectic A phase transition can go through a tricritical point driven by the coupling of nematic and smectic order parameters and their fluctuations [21]. McMillan designated a simple criterion based on the ratio of TNA/TNI, now commonly referred to as the McMillan ratio (M). When M is less than 0.88 his mean field theory predicts a second order NA transition while for larger values it is first order thus defining at tricritical point at M=0.88. McMillan also predicted dependence of M with molecular layer thickness scaled with respect to the rigid core which is shown in Figure 3.

Figure 3. McMillan parameter M as a function of dimensionless measure of layer thickness (l).

An overview of liquid crystals based on Schiff base compounds

Our extensive DSC and ESR and X-ray results for nO.m series are summarized in Table 2 below where we outline the M value as well as the nature of the NA transition. Tricritical value of M for the nO.m series is 0.95-0.96 (see Table 2), but in general the value appears to depend on a given LC homologous series. For liquid crystals of cyano-biphenyl homologous series, which have longitudinal dipole along the molecular long axis, compound 9CB is close to the tricritical NA point and it exhibits a MTCP value of 0.994. This value is significantly higher than the McMillanâ&#x20AC;&#x2122;s prediction. Similarly, different values of MTCP have been reported for other LC series [53]. To gain further molecular level insight, setting the coupling coefficient between nematic and smectic order parameters to be inversely proportional to the longitudinal molecular dipole moment; we find that MTCP~1-K/ Îź2L [62]. Although it should be noted that the increase in the alkoxy chain length affects the order of the NA transition more strongly than increase m the alkyl chain length. Our systematic study of the layer thickness in the smectic A phase revealed that the incremental increase in d spacing is smaller for alkoxy chain than alkyl chain [63]. It implies that the alkoxy chain is tiled with respect to the central rigid benzylidene core of the molecule. For all the molecules at or above n value of 6, only a first order NA transition is observed. For highly asymmetric chain length differences, i.e., m-n>5, the smectic A phase may exhibit interdigitation [64]. High degree of interdigitation appears to suppress the formation of phases with hexatic ordering, perhaps due to enhanced in-plane packing disorder. See for example a disappearance of smectic B phase in 5O.m (B phase disappears after m>15) and nO.1 (B phase disappears after n>9) series. Similarly, tilted hexatic phases like F or C phases do not appear in this region. Table 2. Nematic-Smectic A (NA transition in nO.m series: the McMillan ratio M and the order of transition. n/m 4 5

6 7

4 4O.4 M=0.916 II order 5O.4 M=0.951 II order 6O.4 M=0.976 I order 7O.4 M=0.994 I order

5 4O.5 M=0.898 II order 5O.5 M=0.930 II order 6O.5 M=0.972 I order 7O.5 M=0.990 I order

6 4.O.6 M=0.936 II order 5O.6 M=0.966 I order 6O.6 M=0.990 I order 7O.6 M=1.0 No nem.

7 4.O7 M=0.925 II order 5O.7 M=0.956 II order 6O.7 M=0.990 I order 7O.7 M=0.997 I order

8 4O.8 M=0.956 II order 5O.8 M=0.979 II order 6O.8 M=0.956 I order 7O.8 M=1.0 No nem.

Shankar B. Rananavare & V.G.K.M. Pisipati

We have studied the order of the NA transition as function of n and m using ESR [51, 52, 65, 66], optical birefringence [50], X-ray [67] and DSC [51, 68] techniques. 1.2.2.1. Nematic-Smectic A phase (NA) transition As discussed early on, this is one of the most intriguing liquid crystal phase transitions where first manifestation of solid-like positional order (one dimensional) appears. The transition is an example of 1D melting/freezing. We still do not have a satisfactory theoretical picture of the nature of this transition. De Gennes and McMillan pointed out the possibility of having both first or second order transitions while Lubensky [54] and Anisimov [55] asserted that the director fluctuations make this transition always first order due to the presence of cubic term in the order parameter expansion for the Landau free energy. An important consequence of the second order nature of this transition is the divergence of spatial fluctuation in order parameter. The effect can be probed by analyzing the X-ray scattering line shape of oriented samples. Anisotropic coherence lengths and their temperature dependence have been investigated over orders of magnitudes in reduced temperature (t). Such studies revealed that many compounds do show a second order NA transition as reflected in diverging coherence lengths both parallel and perpendicular to the smectic layer normal [53, 69, 70]. The twist and bend elastic constants depend on the coherence length; so the occurrence of second order NA transitions near the room temperature is not desirable for display device applications of LCs. Diverging elastic constants increase the threshold voltage for the device switching. This makes pixel addressing difficult. Furthermore, extreme susceptibility of coherence lengths, hence the elastic constants to the variation in temperature are also undesirable. Lower NA transition temperatures, i.e. smaller values of M, are realized by using lower chain homologues and selecting molecular structure with smaller longitudinal dipole moments. This valuable guidance for synthesis became only available after development of molecular field theories for nematic and smectic phases. 1.2.2.1.1. ESR studies. To further investigate how the orientational order parameter varies in the mixture of compounds (4O.6 and 6O.4) that exhibit first and second order NA transition, we employed electron spin resonance technique. Here, ESR active nitroxide radical in very small concentration was dissolved in liquid crystals. Detailed analysis of ESR line shape provided information on molecular ordering and dynamics near phase transitions [51, 52]. The orientational order parameter, S, and its enhancement in the

An overview of liquid crystals based on Schiff base compounds

smectic A phase as predicted by McMillan and De Gennes is observed [51, 52]. By looking at disappearance of discontinuity in S at the first order NA transition we measured the critical exponent for this secondary order parameter (ΔS~|TTCP-TNA |β2, with β2≈1) consistent with the prediction of a tricritical point. When the data in table 2 is analyzed, it turns out that the conjugate field variable for the secondary order parameter, i.e. discontinuity in S at the NA transition, is M. Interestingly, the entire nematic order parameter versus temperature data can be placed on a single universal curve if the a new reduced temperature variable is defined as │ (TNI-T)/ (T*NI-T*NA) │; i.e., scaling the temperature deviation by the effective nematic phase extent defined as difference between hypothetical second order NI and NA phase transitions. Furthermore, the phase topology as a function of order parameter S with respect to M establishes the similarity of this phase transition to the super-fluid to normal fluid phase transition observed in the mixtures of Helium isotopes (3He+4He). This was pointed out by de Gennes by noting the similar symmetry of the order parameters for these two transitions, which is a complex two component vector (ψ=│ψ│e(-iφ)) [21]. The ESR probe orientational order parameter measurements of S have now been confirmed by direct order parameter measurement through optical birefringence [50]. Our studies also explore quasi-critical dynamics near the NI and NA phase transitions [51, 65, 66] in nO.m compounds. At the NI transition, the orientational fluctuations give rise to divergence in EPR spin probe linewidth given by a mean field like exponent (w~ξ~~│T-TNI│1/2 here w refers to homogeneous linewidth of EPR spin label undergoing rapid molecular reorientation, yet experiencing slow dynamic fluctuation modes of the nematic orientational order. At the NA transition, probe expulsion from LC core to alkyl region in layered smectics couples orientational fluctuation of probe with the smectic order parameter. It leads to a slower linewidth divergence (w~ξ1/2~~│T-TNI│1/3). This latter observation is consistent with the De Gennes super-fluid analogy where coherence length is predicted to diverge with 2/3 power-law exponent. 1.2.2.1.2. X-ray studies. Our unpublished data as well as other data from literature pertaining to the critical exponents associated with coherence length (ξ) in directions parallel (ν⎜⎜) and perpendicular (ν⊥) to smectic layer and X-ray quasi Bragg peak intensities (γ) for nO.m homologous series are presented in Table 3. These results are in good agreement with compilation of results provided by Garland et al [53]. Away from the tricritical NA point the results are consistent with a 3D-XY model while for tricritical and weakly first order NA transition the exponents correspond to the mean field values.

Shankar B. Rananavare & V.G.K.M. Pisipati

Table 3. Critical exponents determined from X-ray studies; definitions:

I ∝ t −γ ; ξ Π ∼ t −ν Π ; ξ ⊥ ∼ t −ν ⊥ ; t = Liquid crystal

TNA − T TNA

υ|| 4O.8 0.958 1.31 0.70 4O.7 0.926 1.46 0.78 4O.6 0.936 -0.64±0.09 6O.4 0.976 -0.43±0.08 Tricritical 4O.6:6O.4 mixture (81:19 w/w) 0.955 -0.60±0.05 3D-X-Y 1.32 0.67 Tricritical (Theory) 0.87 1 0.5

Reference υ⊥ 0.57 [69] 0.65 [70] 0.46±0.06 [67] 0.4±0.1 “ 0.55±0.05 “ 0.67 [53] 0.5 [53]

1.2.3. Smectic B phase Given that the observed molecular tilts of smectic F and G phases are small, formation of crystalline B phase appears to be a general rule than exception. Smectic B phase that is commonly observed in nO.m compounds tends to be of crystalline variety with a saturated hexatic order parameter. An interesting consequence, as remarked before, is that when there is significant interdigitation this phase is quenched. Generally in nO.m compounds this phase melts into either Smectic A or smectic C phase and it freezes into a tilted smectic G phase. This phase is almost like a three dimensional crystal, melting or freezing through strong first order transitions. In 2D monolayer, however this transition can be potentially second order of the type predicted by Kosterlitz –Thouless [71].

1.2.4. The smectic C phase This tilted phase is mainly observed in the central region of Table 1, where n ≈ m ≈ 6. In this region, the molecules should be more or less straight and have lateral dipole moment components. Observation of smectic C phase is also consistent with the McMillan’s model [49]. In the McMillan’s mean field model [49], the transition to smectic C phase ( as opposed to C1 and C2 phases ) from smectic A is governed by transverse dipoles (TAC ~μ22 where μ2 is the molecular dipole away from the center of mass (CM), presumably corresponding to the dipole moment associated with the alkoxy group (1.15D)). Formation of non-ferroelectric smectic C phases requires 2 μ22> μ12

An overview of liquid crystals based on Schiff base compounds

where μ1 is a dipole moment at the center of mass of the molecule corresponding to the dipole moment of Schiff base C-N group, ≈1.57D [49]. According to McMillan “The structural factors favoring the smectic C phase relative to smectic A phase are: (i) approximate center of symmetry; (ii) large outboard parallel dipole moments; (iii) zigzag (trans) gross shape of molecules.” Thus when n and m differ by large number (⎜n-m⎜> 3) approximate center of molecular symmetry is lost leading to suppression of the smectic C phase. When n ≈ m for small m (<4) it may be more difficult to maintain the zigzag or trans shape of the molecule, due internal segmental rotation around carbon bonds in the alkyl chains. While for large n, but with n≈m, an incipient in-plane hexatic order might lead to the smectic F phase directly as described below. Chu and McMillan [72] have described a Landau theory for the NAC multicritical point and it suggests a possibility of direct transition from nematic into smectic C phase which is not observed in the nO.m series (but see below). Chen and Lubensky [73] have described the NAC multicritical point as a Lifshitz point [74, 75] where one of the Lifshitz invariants vanishes. X-ray diffraction studies of Birgenau et al [76] do indeed observe Lifshitz point like fluctuations in nematic phase as smectic C phase is approached, although experimentally the N-C transition always appears to be a first order transition. Given that the nO.m. series does not exhibit a direct nematic to smectic C phase transition, we will not discuss here the NAC multicritical point. 1.2.4.1. Smectic A to smectic C (AC) phase transition. In nO.m series, smectic C phase is reached by cooling from the Smectic A phase. This phase transition in this series tends to be a weakly first order and appears to be close to a tricritical point. This finding is consistent with calorimetric studies of Garland et al [21, 77]. What is typically observed is a mean field behavior with unusually large sixth order term for the Landau expansion. Most of our studies have been performed with low temperature resolution hence we were unable to corroborate MIT work for all the AC transitions in this series. This is especially true for compounds such as 10O.8 which appears to exhibit a weakly first order AC transition [63]. This AC transition has become very important in understanding behavior ferroelectric liquid crystals which are realized by doping smectic C phase with chiral solutes. The order parameter for this transition is a complex order parameter similar to NA transition (θ=│θ│e(-iφ)) [21, 78, 79], where θ is a molecular tilt angle with respect to layer normal and phi φ the azimuthal angle, also called as phase angle. Above the AC phase transition, in smectic A phase the fluctuation in the magnitude of tilt angle dominate while below the transition, the phase fluctuation dominate. The latter are called as the Goldstone modes.

Shankar B. Rananavare & V.G.K.M. Pisipati

In ferroelectric materials the switching involves change of Ď&#x2020; by 180 degrees, which can occur in microseconds, a much faster response than millisecond switching speeds commonly found for the nematic based LCDs. Therefore, studies of phase transition and switching dynamics in chiral smectic C as well as bent-core liquid crystals have come under intense scrutiny for their potential display applications. An Achilles heel of these devices based on chiral smectics or smectics in general is the ability to realize defect free alignment over a large area. This is much more challenging than nematic LCs owing to low free energy of focal conic defects which are difficult to anneal away, requiring higher temperature realignment process.

1.2.5. The smectic F phase The smectic F is characterized by in-plane hexatic order. In our studies [63, 80] of few representative nO.m compounds, we noted that both F and C phases exhibit relatively small molecular tilt angles. This is again consistent with the notion that the F phase is reached either from isotropic, or Smectic A or Smectic C phase, which (in nO.m series) have either have no tilt or very small tilt angles. De Gennes and Prost describe this phase transition from C-F as a transition without change in symmetry, i.e. hexatic bond orientational order in the smectic planes varies continuously across the transition [81-83]. Theoretically this is very intriguing transition in that it is believed to be similar to Kosterlitz -Thouless transition [71] in 2D for thin films. The lower temperature transition from the smectic F phase is mostly to smectic G phase or in some instances directly into a solid crystal phase. In this respect, the low temperature transition represents a 2D freezing whereas the interlayer smectic order becomes long range. We found that the intensity of the Bragg peaks increases upon cooling from an istropic to F phase following a power-law behavior. Invariably this transition is weakly first order. Simple molecular mean field theory for either F or G phase does not exist. But a cursory look at the Table 1 indicates that the phase is prevalent in bottom right section, where n and m values are large. The conditions for the occurrence of F phase are therefore very similar to the smectic C phase as discussed above. The primary difference between them is the magnitude hexatic bond orientational order parameter, just as the difference between liquid and gas phases is the magnitude of density of the two phases. It would be interesting to study the C-F transition in thin films and bulk to see if they follow the same trend as TB9 based compounds which also have Schiff base linkage [83]. Finally just as in the case of smectic C phase, the F is not present on the off-diagonal region (bottom left and top right regions). Here the hydrocarbon chain

An overview of liquid crystals based on Schiff base compounds

length differential |m-n| is large, presumably leading to interdigitation and suppression of the smectic F phase.

1.2.6. Smectic G phase Smectic G phase is almost always the lowest temperature mesomorphic phase found in nO.m series. It can be induced in binary mixtures of mesogenic molecule exhibiting strong intermolecular hydrogen bonding [84, 85]. The hydrogen bonding interaction is directional and allows only fixed number of orientational sites for hydrogen bond formation. This presumably enhances the positional ordering i.e. crystallinity, which is the hallmark of the smectic G phase. Therefore, it is more aptly described as 3D pseudocrystalline phase. Consistent with this finding is the X-ray study of Noh et al who observed that the F-G transition is first order. Although a 3D crystal, elastic constants in Smectic G phase tend to be smaller [83]. These authors noted diffuse scattering with coherence length approaching few hundred nanometers along the direction of the director. Our studies of smectic G phase in 10O.8 and 10O.14 also found that the nature of the transition from hexatic F phase to G is first order [63]. Interestingly, Bragg peaks showed power-law increase in the peak intensity upon cooling into the smectic G phase indicating development of a long range crystalline order parallel to layer normal as the in-plane hexatic order saturates. Early NMR studies of 5O.7 and the mixtures of 5O.6 and 9O.4. due to Doane et al established that the hydrocarbon chains maintain at least some degree of mobility in this phase [86] with high degree molecular ordering.

2. nO.Om series 2.1. General comments This series [34] has been recently attracting greater attention. Phase variants observed in this series are shown in Table 4 (see below). Interestingly, the phase variants are not symmetrically disposed along n=m diagonal although having two alkoxy chains confers a pseudo center of inversion to these class of molecules. Replacement of alky group with alkoxy group seems to have increased propensity for appearance of nematic phases in all the compounds except the few located in the right bottom corner. Galewski et al report that most of these transitions are first order as indicated by thermal calorimetric studies. It would be interesting to study the orientational ordering either through birefringence or magnetic resonance lineshape studies near NA transitions in case of compounds that exhibit the transition.

Shankar B. Rananavare & V.G.K.M. Pisipati

The compound showing largest phase variants is 10O.O5, exhibiting NACIG. Another striking feature of the phase diagram is the complete disappearance of the smectic F phase. The reasons for this phenomenon are not well understood at the moment. The hexatic smectic F phase is replaced with another similar hexatic phase, the smectic I phase. As both the chain lengths are increased the tendency to transform directly from isotropic phase to the smectic C is phase is noted. One significant difference compared to nO.m series is the propensity of nematic to smectic C phase transition as compared to the NA phase transition. Furthermore, neither smectic A, nor Smectic B or smectic G phases are as prevalent.

2.2. Specific comments 2.2.1. Nematic phase Presence of polarizable alkoxy group in both chains improves the appearance of nematic phase for a large region of nXm matrix as can be seen in Table 4. Furthermore, these multiple longitudinal dipoles appear to increase the TNI to above 100C or so, approximately 30-40 degrees above the Table 4. Phase variants in nO.Om series.

n/m 1 2 3 4 5

1 N N N N N

2 N N N N N

3 N N N N

4 N N N N N*

5 N N N N -*

6 N N N N N

7 N N N N N

8 N N N N NC

9 N N N N NC

10 N N N NC NC

12 N N N NC NC

N N

NCI

NACI

N N

NCI

N N

NCI

NCIG

NCI

N NACB

NACB

NACI

NCIG

NCI

N NAB

NACB

NACB NACIG NACIG CI

12 N N`AB NAB NACB ACB ACB CI CI CI CI C Our extensive studies on 4O.Om and 5O.Om series have shown that the compound 4O.O5 exhibits no liquid crystalline phases while 5O.O5 exhibits a narrow range nematic phase.

An overview of liquid crystals based on Schiff base compounds

clearing temperatures commonly found in the nO.m series. This is consistent with the Maier-Saupe theory [36-38, 87] results:

TC ≈ 4.55U / k B ∝ f ( μ 2 ) where μ is the induced molecular dipole moment. The importance of these types of molecules is that they allow a design of Schiff base liquid crystals with a much wider nematic phase range than was accessible with the nO.m series alone. Thus, mixtures of these molecules with high NI transition temperature with those Schiff base molecules that have much lower transition temperature such as n.m series discussed below would be useful. Such mixtures allow careful studies of critical phenomena of crossover from a 3D-XY model to the tricritical point in a chemically similar set of molecules.

2.2.2. Smectic A phase Unlike the nO.m series (Table 1) this phase is only observed in the bottom left portion of the table 4 and altogether there are only 11 instances of the compounds exhibiting the NA transition. In Figure 4 we plot M, the McMillan ratio vs. the sum n+m corresponding to total alkyl chain length by realizing that the molecules in this homologous series are effectively centrosymmetric. Clearly, as McMillan predicted, an increase in M occurs as a function of size of the molecule. A similar plot for nO.m series, showing much higher scatter appears in Figure 4 (right side)1. It should be noted that an even-odd effect [61] in hydrocarbon chain length on the NI transition temperature does make the scatter bit high but qualitatively the results are consistent with the McMillan’s model for the smectic A phase as discussed before. Careful studies of orientational order parameter discontinuities (ΔS) or critical exponents at the NA transition are yet to be performed. It would be interesting to know whether this homologous series exhibits a TCP and the corresponding value of M and how it compares with MTCP (≈ 0.95) for the nO.m series. Empirical prediction based on nCB and nOCB series would be that the corresponding value of MTCP would be higher. A compound 9O.O2 does exhibit M=0.92 and shows a weak or no peak in DSC is likely to be the best candidate for the study of a second order NA phase transition. Coincidently this M value is similar to the one exhibited by 4O.7 (with n+m=11) (M=0.925) which displays a second order NA phase transition. Note, the slopes of M vs n+m (≈ 0.0135) are also comparable, across these two series. 1

Although if M is plotted as a function n much better quality linear fit is observed of this series

Shankar B. Rananavare & V.G.K.M. Pisipati

Figure 4. Variation of the McMillan parameter M in nO.Om and nO.m series.

2.2.3. Smectic C phase Given the (almost) center of symmetry of these molecules, the observed preponderance of smectic C phase should be taken as a further confirmation of the McMillanâ&#x20AC;&#x2122;s model of the smectic C phase (see above). It would be interesting to investigate the critical behavior at N-C phase transition as both McMillan and De Gennes models predict a divergence of all three elastic constants at this transition. Blending mixtures [34, 88] with compounds exhibiting NA transition would also permit studies of the NAC multicritical point [44, 76, 89], for example, compounds like 10O.O6 with 6O.O10. Advantage is that chemically similar molecules of similar dimensions should be more compatible and presumably form ideal solutions. This was not possible with nO.m series as no compounds in that series exhibits a NC transition. Another interesting possibility would be to mix a tricritical NA mixture with a compound exhibiting NC transition, to see how or if phase topology changes and bears resemblance to the classic Lifshitz point topology. Similarly it would be worthwhile to establish the nature of AC transition whether it exhibits mean field behavior with large sixth order term or not.

3.0. n.m compounds and 4.0. n.Om series Few studies of these homologous series [35] have been reported. For n.m series the LC transition temperatures are significantly suppressed along with the diversity LC phases. A comparison of phase variants appear below.

An overview of liquid crystals based on Schiff base compounds

Table 5. Comparison of phase variants in nO.m, n.Om, nO.Om and n.m homologous series [35]. Compound 1O.5 1.O5 1O.O5 1.5 2O.5 2.O5 2O.O5 2.5 5O.5

Phase variant N Non-liquid crystalline nature. N Liquid crystal below room temp. N Non-liquid crystalline nature. N Liquid crystal below room temp. NACFG

5.O5 5O.O5 5.5 1.O16

NG N NG Non-liquid crystalline nature.

1O.O16

Non-liquid crystalline nature.

1.16 2O.16A 2.O16 2O.O16

Non-liquid crystalline nature. A Non-liquid crystalline nature. Non-liquid crystalline nature.

2.16

Non-liquid crystalline nature.

4O.16

I—92.1—N—83.5--K I—90.4—N—63.3--K I—119.3—N—88.5--K I—77.8—N—54.5—A—53.1—C—49.9— F—47.0—G—28.0--K I—76.8—N—51.6—G—40.7--K I—109.1--N I—40.8—N—23.9--G

I—70.1—A—52.2--K

I—69.2—N—68.5—A—51.3--B

4O.O16

Non-liquid crystalline nature.

4.16

I—38.1—A—35.0--K

5O.16

I—68.8—N—67.4—A—55.1--K

5.O16

I—79.2—N—72.3—A—51.3--K

5O.O16

I—87.4—N—87.0--K

5.16

I—52.9—A—40.2--K

I—75.5—A—71.5—B

8O.16

NAB

Transition temperatures (OC) I—62.6—N—39.7--K

8.O16

Non-liquid crystalline nature.

8O.O16

Non-liquid crystalline nature.

8.16

Non-liquid crystalline nature.

P.A. Kumar, M.L.N. Madhu Mohan and V.G.K.M. Pisipati, Liq.Cryst., 27, 727 (2000). N.V.S.Rao, D.M. Potukuchi and V.G.K.M. Pisipati, Mol.Cryst.Liq.Cryst., 196, 71 (1991). C M.Jitendranath, C.G.Rama Rao, M.Srinivasulu and Venkata G.K.M.Pisipati Mol.Cryst.Liq. Cryst., 366, 47 (2001). B

Shankar B. Rananavare & V.G.K.M. Pisipati

The majority of the present benzylidene aniline compounds exhibit nematic phase except 1.O5 and 2.O5. The presence of oxygen atom on both sides of the benzylidene aniline (1O.O5 and 2O.O5 compounds) and the removal of oxygen from both sides of the compounds (1.5 and 2.5) results in significant changes in melting and clearing temperatures, and their liquidcrystalline nature. As discussed before the molecular dipoles due C-O groups lead nO.Om compounds exhibit their nematic phase at highest temperature, while compounds lacking such polar groups viz, n.m series, exhibits overall lower clearing temperatures. The take home lesson from these synthetic experiments is that molecules of high and low clearing temperatures, that are needed for wide operational range of LCDs, can be synthesized using this strategy of introduction or removal of oxygen in the alky chain region flanking the mesogenic core of the liquid crystal molecules. The n.Om series also illustrates important effect of the reversal of central linkage with respect to the rigid core regions of the molecule (CH=N or N=CH). It shows that in 5O.5 and 5.O5), small and insignificant changes occur on the clearing temperatures as well as melting temperature. However, the variety of LC phases they exhibit is found to be drastically different. In conclusion, the experimental studies of these classic systems illustrates the extent of success of mean field theories as well as challenges in establishing refinement of theoretical models to predict the phase variants. It is clear that besides the molecular shape and mean/induced dipoles, the effects of relative orientations and spacing of functional group dipole moments is very important and need to be investigated theoretically to have better predictability of phase variants.

5.0. Other variations Introduction of a chiral functional group or geometric variation in molecular architecture of nO.m has led to discovery of new liquid crystals such as ferroelectric [29, 30, 56] or bent core liquid crystals [90]. In addition, to improve device switching speeds by reducing viscosity (molecular friction) through introduction of fluorinated chain has also been pursued. This is a very large area of active research and here we focus on few selected molecular structural variants that we have investigated [91, 92]/synthesized [62]. In addition, we present studies of classic thermotropic nO.m of series molecules mixed with traditional surfactant based liquid crystals [93-95]. The so called amphitropic liquid crystals are still in the early stages of development especially with respect to their potential applications in lithography or so called bottom up approach in assembling nanostructures.

An overview of liquid crystals based on Schiff base compounds

5.1. Chiral centers and ferroelectric liquid crystals The prediction and discovery of the first ferroelectric liquid crystals by Meyer et al provided a first technologically significant, new class of liquid crystals based on Schiff base architecture [29, 30]. Here the molecules, arranged in smectic C phase have a chiral center attached to them which reduces the overall symmetry of smectic C phase from C2h to C2. Along the C2 axis, in the plane of layer, a non-vanishing component of polarization is generated. The coupling of tilt (primary order parameter) and polarization has been included in the Landau model for the smectic A-chiral C(C*) transition [96]. The direction of the polarization vector can be rapidly changed through application of an electric field. In the bulk, the ferroelectric smectic C* phase exhibits a helical modulation of polarization in direction normal to smectic layers. Switching entails rotation of molecules along the cone of the fixed tilt angle. The switching time ( Ď&#x201E; = Îł / P.E , where Îł is rotational viscosity, P the polarization and E is the in-plane applied electric field [78]) is at least three orders of magnitude shorter than typical milliseconds switching times observed in nematic liquid crystal based devices. Unlike nematics, which do not have any positional order, the smectic C* phase has positional, orientational as well as hexatic bond orientational order. Following the analogy to dilute magnetism, we [91, 92] investigated ferroelectric phase behavior in a mixture of chiral nO.m, p-(n-decyloxy-benzylidene)-p-amino(2-methylbutyl) cinnamate (DOMBABC [79]) and an achiral 10O.8 molecule to probe if the ferroelectric phase persists to low concentrations of the chiral molecule. The temperature composition phase diagram appears in Figure 5.

Figure 5. Temperature composition phase diagram of DOBAMBC and 10O.8 mixtures.

Shankar B. Rananavare & V.G.K.M. Pisipati

Surprisingly, the observed the chiral-achiral phase transitions appear at finite and significant concentration of DOMBABC (chiral molecule). This finding is contrary to what would be expected based on results of induction of cholesteric phases by chiral solutes in the nematic phase. The smectic C* phase disappears at about 36% dopant (chiral molecule) concentration! The total phase diagram shows remarkable topology similar to a Lifshitz point as predicted by Michaelson et al [74, 75, 97]. At the ACC* Lifshitz point, induced by concentration, pitch length diverges and polarization disappears with power laws that can be rationalized in terms of disappearance of Lifshitz invariants. Nonetheless, the precise chemico-physical principle behind this phenomenon remains to be uncovered. Similar ACC* Lifshitz points have been observed as a function of the helix unwinding fields such as electric, magnetic and confining geometries (so called surface stabilized FLCs discovered by Clark and Lager wall [98]). A fundamental difference between these results and our results lies in that in these studies of mixtures chiral and achiral LCs we observe a simultaneous disappearance of switchable polarization and helical pitch modulation.

5.2. SF5 based nO.m liquid crystals Search for ever increasing polarization (dipole moment per unit area) lead to discovery of anti-ferroelectric phases in the early nineties [99, 100]. Similarly if the molecule is bent, banana-like, then the breakdown of axial symmetry can lead to ferroelectric phases that do not have a local chiral group as was discovered in mid nineties [90]. In the late nineties, in pursuit of higher longitudinal dipole moments to realize the exotic devil staircase type structures in the electric switching led one of us to conjecture that perhaps a highly polarizable and highly polar group such as SF5 might allow one to create such a phase. The structure of the synthesized nO.m, analogs is shown in Fig. 6: In these compounds we observe monotropic liquid crystals which when mixed produce a stable nematic phase [62]. More interestingly, these compounds exhibit remarkably high values of McMillan parameter and yet exhibit a second order NA phase transition. This is similar to result observed in compound, N-p-cyano-benzylidene-p-octyloxy-aniline CBOOA [101], which has an axial cyano group in addition to the benzylidene group in structure. Despite its M=0.934 it exhibits a second order NA transition implying a higher value for its MTCP. These SF5-based compounds also feature per-fluorinated alkyl chains, but not the alkoxy chains. Our expectation was that the longitudinal dipole moment might give rise to longitudinal ferroelectric order. Single crystal analysis revealed that the

An overview of liquid crystals based on Schiff base compounds

Figure 6. SF5 based partially fluorinated nO.ms.

Figure 7. Ball-stick model of single crystal SF5 based 7O.2.

fluorocarbon and the hydrocarbon chains mixed, leading to a formation of longitudinal and in-plane anti-ferroelectric order as shown in Fig. 7. We also explored possibility of bent core liquid crystals with meta location of amine groups without success perhaps due to limited range of n and m variations.

Shankar B. Rananavare & V.G.K.M. Pisipati

5.3. Amphitropic liquid crystals In past we have extensively investigated how bilayer smectic A phase of lyotropic, surfactant-based systems interacts with additives, such as water and hydrocarbons [102-105]. Here the main effect is an increased interlayer spacing as a function of additive concentration till a phase boundary is reached. In cases where additive gets embedded in the layered structure significant changes in molecular orientational ordering in bilayers [106, 107] have been noted. It was therefore of interest if a nematic forming nO.m compound could be dispersed using surfactant and to examine the resulting model ternary phase diagram of such a system. Most common surfactants such as anionic SDS (sodium dodecyl sulfate), cationic CTAB (cetyl tetraammonium bromide) or nonionic Brij 30 (C12H25(EO)4OH) molecules exhibited formation of milky emulsions. In these emulsions water droplets tend to form pearl garland-like chains [108-110] are believed to be important for chemo-responsive configurable assemblies [110]. We focused exclusively on a nonionic Triton analog surfactant which exhibited rich phase mesomorphism with MBBA (1O.4) as shown in figure 8 below [111]. The surfactant is sparingly soluble in the nematic phase;

Figure 8. (Left) A partial three component phase diagram of MBBA, Triton X114 and water is shown. Multiphase coexistences as viewed between crossed polarizers (right). Note the inverted test tube in the bottom right portion of the figure corresponding to an isolated critical point between lamellar LCs. L2 and L3 are lamellar LCs, I is an isotropic phase. Shaded triangular region shows coexistence of three phases, I+L2+L3.

An overview of liquid crystals based on Schiff base compounds

however, very small amounts dissolved in the nematic phase of MBBA lead to destruction the nematic phase. The NI phase transition is very rapidly quenched at low concentration of surfactant. Even where a small amount of surfactant is present in the nematic phase the maximum amount of water it accommodates is even smaller <0.5%. We rationalize the instability of nematic phase in presence of inverse micellar structures to creation of axial +1 type of defects compensated by creation of corresponding â&#x20AC;&#x201C;1 type of defect in the surrounding nematic phase. As more water and surfactant are added, the deformation elastic energy associated with micellar structures drives the transition from the nematic phase containing micelles to an isotropic phase. Note that the dimensions of micelles are comparable to the core dimensions of defects found in nematics [87]. This proposed mechanism borrows from the well understood mechanism of homeotropic alignment of nematic liquid crystals by surfactant derivatized surfaces. In the lamellar region of the phase diagram there appears an extension towards water corner in a manner reminiscent of nonionic surfactantbenzene- water phase diagrams [105, 112]. A noteworthy feature of this ternary phase diagram is a three phase region where two lamellar phases of different densities coexist with an isotropic phase as shown in above (top right portion of) figure 8. Following the two lamellar phase coexistences into a single phase lamellar region we were able to locate a critical point where the bulk viscosity diverges. More studies are needed to establish the nature of this critical divergence. The remarkable features of the phase diagram are: 1. The nematic phase of MBBA is completely quenched with very small addition of surfactant molecules; 2. We find that the lamellar phase accommodates small amount of MBBA but does not lead to formation of a nematic microemulsion; 3. Interestingly, we find a novel three phase region where two lamellar phases coexist with an isotropic phase. 4. Along the side of the Gibbs triangle, which shows coexistence of lamellar phases, the viscosity exhibits dramatic divergence hinting a formation of an isolated critical point. At this concentration, the system has 80% water and yet it does not flow freely but forms a birefringent gel. Away from this region, the lamellar phase L1 as well as coexisting lamellar phases (L2 and L3) and I phases flow freely. More details of the work will be published elsewhere. However, the general area of phase behavior of surfactants and the classic thermotropic liquid crystals remains mysterious and poorly investigated. One deficiency of nO.ms for this application is that they tend to hydrolyze over time.

Shankar B. Rananavare & V.G.K.M. Pisipati

Summary and future prospects The classic nO.m series has provided a novel platform for development of modern liquid crystal physics and chemistry. Some of the key achievements are: the discovery of new classes of liquid crystals such as ferroelectric and bent core liquid crystals. The potential of these mesogen in polymeric systems as well as their incorporation in amphitropic liquid crystals should be contemplated. Such materials will provide a suitable medium for aligning nanowire and nanotubes. Once alignment is achieved and locked these materials are relatively easy to chemically degrade. In general, the use of liquid crystals for lithographic application is an area that has received no attention but holds potential of improving resist sensitivity as well as providing novel non linear optic material of controllable birefringence for immersion lithography.

Acknowledgements It is a great pleasure to acknowledge many stimulating discussion with Professor Jack Freed at Cornell University. Professor Anthony J Ward provided a careful review of the manuscript. Financial support was provided to SBR from National Institute of Health grant No. HL 54209 and to VGKMP by Department of Science and Technology, New Delhi, India through the grant No.SR/S2/CMP-0071/2008.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

H. Kawamoto, Proceedings of the IEEE 2002, 90, 460. R. Williams, The Journal of Chemical Physics 1963, 39, 384. M. Schadt, W. Helfrich, Applied Physics Letters 1971, 18, 127. P. M. Alt, P. Pleshko, Electron Devices, IEEE Transactions on 1974, 21, 146. T. J. Scheffer, B. Clifton, D. Prince, A. R. Conner, Displays 1993, 14, 74. T. Scheffer, J. R. Nehring, Annual Review of Materials Science 1997, 27, 555. J. A. McDonald, Microelectronics Journal 1994, 25, xvii. http://business.highbeam.com/industry-reports/equipment/special-industrymachinery-not-elsewhere-classified. http://www.samsung.com/ae/news/newsRead.do?news_seq=19410&gltype=local news. T. Larsen, A. Bjarklev, D. Hermann, J. Broeng, Opt. Express 2003, 11, 2589. F. Simoni, Nonlinear optical properties of liquid crystals and polymer dispersed liquid crystals, Vol. 2, World Scientific, New York, 1997. C. Aroulanda, V. Boucard, F. GuibĂŠ, J. Courtieu, D. Merlet, Chemistry â&#x20AC;&#x201C; A European Journal 2003, 9, 4536. R. Yamamoto, Physical Review Letters 2001, 87, 075502.

An overview of liquid crystals based on Schiff base compounds

14. M. D. Lynch, D. L. Patrick, Nano Letters 2002, 2, 1197. 15. L. K. Cotter, T. J. Drabik, R. J. Dillon, M. A. Handschy, Opt. Lett. 1990, 15, 291. 16. Y. W. Zhang, M., O'Callaghan M. J., Walker, C. M., Thurmes W, Ferroelectrics 2006, 344, 11. 17. J. W. Doane, A. Khan Cholesteric liquid crystals for flexible displays, chapter 17 pg 332, Wiley, Hoboken, NJ 07030, 2005. 18. M. Shimomura, T. Sawadaishi, Current Opinion in Colloid & Interface Science 2001, 6, 11. 19. B. D. Gates, Nat. Nano., 2010, 5, 484. 20. J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M. Whitesides, Chemical Reviews 2005, 105, 1103. 21. P. G. De Gennes, P. J., Prost The physics of liquid crystals, Vol. 83, second edition ed., Clarendon Press, Oxford Science Publications, Oxford, 1993. 22. I.C. Khoo, S.T. Wu, Optics and nonlinear optics of liquid crystals, Chapt. 2,, Vol. 1, World Scientific, River Edge, NJ 07661, 1993. 23. R. Lu, X. Zhu, S. T. Wu, Q. Hong, T.X. Wu, Journal of Display Technology 2005, 1, 3. 24. M. Koden, in http://sharp-world.com/corporate/info/rd/tj2/pdf/12.pdf, Sharp Technology. 25. P. Detlef, T. Kazuaki, Journal of Materials Chemistry 2004, 14, 1219. 26. G. Gauza, C. H. Wen, B. Wu, S. T. Wu, A. Spadlo, R. Dabrowski, Liquid Crystals 2006, 33, 705. 27. H. Kelker, B. Scheurle, Angewandte Chemie International Edition in English 1969, 8, 884. 28. G. W. Gray, K. J. Harrison, J. A. Nash, Electronics Letters 1973, 9, 130. 29. R. B. Meyer, Ferroelectrics 1980, 28, 319 30. R. B. Meyer, L. Libert, I. Strzeleck, P. Keller, J. Phys. (Paris) Lett 1975, 36 69. 31. V. Pisipati, Zeitschrift Fur Naturforschung Section a-a Journal of Physical Sciences 2003, 58, 661. 32. N. Ajeetha, V. Pisipati, Zeitschrift Fur Naturforschung Section a-a Journal of Physical Sciences 2005, 60, 207. 33. K. Fakruddin, R. J. Kumar, P. V. D. Prasad, V. Pisipati, Molecular Crystals and Liquid Crystals 2009, 511, 1616. 34. J. Godzwon, M. J. Sienkowska, Z. Galewski, Acta Physica Polonica A 2008, 113, 1145. 35. N. Ajeetha, D. M. Potukuchi, V. Pisipati, Phase Transitions 2005, 78, 369. 36. W. Maier, A. Saupe, Z. Naturforsch 1958, 13A, 554. 37. W. Maier, A. Saupe, Z. Naturforsch 1959, 14A, 882. 38. W. Maier, A. S. Z., Z. Naturforsch, 1960, 15A, 287. 39. L. Onsager, Ann. N. Y. Acad. Sci. 1949, 51, 627. 40. M. Nakata, G. Zanchetta, B. D. Chapman, C. D. Jones, J. O. Cross, R. Pindak, T. Bellini, N. A. Clark, Science 2007, 318, 1276. 41. H.-S. Park, S.-W. Kang, L. Tortora, Y. Nastishin, D. Finotello, S. Kumar, O. D. Lavrentovich, The Journal of Physical Chemistry B 2008, 112, 16307. 42. G. Zanchetta, et al., Journal of Physics: Condensed Matter 2008, 20, 494214.

Shankar B. Rananavare & V.G.K.M. Pisipati

43. S. Bucak, C. Cenker, I. Nasir, U. Olsson, M. Zackrisson, Langmuir 2009, 25, 4262. 44. Z. Y. Chen, Macromolecules 1994, 27, 2073. 45. L. R. Evangelista, I. Hibler, H. Mukai, Physical Review E 1998, 58, 3245. 46. J. Nehring, A. Saupe, The Journal of Chemical Physics 1971, 54, 337. 47. J. Nehring, A. Saupe, The Journal of Chemical Physics 1972, 56, 5527. 48. W. L. McMillan, Physical Review A 1971, 4, 1238. 49. W. L. McMillan, Physical Review A 1973, 8, 1921. 50. V. G. K. M. Pisipati, P. V. D. Prasad, Molecular Crystals and Liquid Crystals 2009, 506, 13. 51. S. B. Rananavare, V. G. K. M. Pisipati, J. H. Freed, Liquid Crystals 1988, 3, 957. 52. S. B. Rananvare, V. G. K. M. Pisipati, J. H. Freed, Chemical Physics Letters 1987, 140, 255. 53. C. W. Garland, G. Nounesis, Physical Review E 1994, 49, 2964. 54. B. I. Halperin, T. C. Lubensky, S.-k. Ma, Physical Review Letters 1974, 32, 292. 55. M. A. Anisimov, Mol. Cryst. Liq. Cryst. 1988, 162A, 1. 56. R. B. Meyer, Molecular Crystals and Liquid Crystals 1977, 40, 33 57. P. Barois, J. Pommier, J. Prost, Phase Transitions: A Multinational Journal 1991, 33, 183 58. D. Demus, S. Diele, S. Grande, H. Sackmann, Advances in Liquid Crystals 1983, 6, 1. 59. H. Sackmann, D. Demus, Molecular Crystals and Liquid Crystals 1973, 21, 239. 60. A. Wiegeleben, L. Richter, J. Deresch, D. Demus, Molecular Crystals and Liquid Crystals 1980, 59, 329. 61. Marcelja, S. The Journal of Chemical Physics 1974, 60, 3599. 62. J. A. Smith, R. A. DiStasio, N. A. Hannah, R. W. Winter, T. J. R. Weakley, G. L. Gard, S. B. Rananavare, The Journal of Physical Chemistry B 2004, 108, 19940. 63. V. G. K. M. Pispati, S. B. Rananavare, Molecular Crystals and Liquid Crystals Science and Technology. Section A. Molecular Crystals and Liquid Crystals 1994, 238, 207 64. V. G. K. M. Pisipati, S. B. Rananavare, Liquid Crystals 1993, 13, 757. 65. J. H. Freed, A. Nayeem, S. B. Rananavare, in Molecular Dynamics of Liquid Crystals, Vol. 431 (Ed.: G. R. Luckhurst, C. A. Veracini), 1994, pp. 335. 66. A. Nayeem, S. B. Rananavare, V. S. S. Sastry, J. H. Freed, The Journal of Chemical Physics 1992, 96, 3912. 67. S. B. Rananavare, R. C. Creapau, V. G. K. M. Pisipati, J. H. Freed, 1988. 68. V. G. K.M. Pisipati, S.B. Rananavare, J.H. Freed, Molecular Crystals and Liquid Crystals 1987, 4, 181. 69. R. J. Birgeneau, C. W. Garland, G. B. Kasting, B. M. Ocko, Physical Review A 1981, 24, 2624. 70. C. W. Garland, M. Meichle, B. M. Ocko, A. R. Kortan, C. R. Safinya, L. J. Yu, J. D. Litster, R. J. Birgeneau, Physical Review A 1983, 27, 3234. 71. J. M. Kosterlitz, D. J. Thouless, Journal of Physics C: Solid State Physics 1973, 6, 1181. 72. K. C. Chu, W. L. McMillan, Physical Review A 1977, 15, 1181.

An overview of liquid crystals based on Schiff base compounds

73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96.

J.-h. Chen, T. C. Lubensky, Physical Review A 1976, 14, 1202. A. Michelson, Physical Review Letters 1977, 39, 464. A. Michelson, Physics Letters A 1977, 60, 29. C. R. Safinya, R. J. Birgeneau, J. D. Litster, M. E. Neubert, Physical Review Letters 1981, 47, 668. R. J. Birgeneau, C. W. Garland, A. R. Kortan, J. D. Litster, M. Meichle, B. M. Ocko, C. Rosenblatt, L. J. Yu, J. Goodby, Physical Review A 1983, 27, 1251. Carlsson, T. Zeks, B., Filipic, C., Levstik, A., Blinc, R. Blinc, Molecular Crystals and Liquid Crystals Incorporating Nonlinear Optics 1988, 163, 11 S. Dumrongrattana, C. C. Huang, G. Nounesis, S. C. Lien, J. M. Viner, Physical Review A 1986, 34, 5010. P. B. Rao, N. V. S. Rao, V. G. K. M. Pisipati, Molecular Crystals and Liquid Crystals 1991, 206, 9 A. Aharony, R. J. Birgeneau, J. D. Brock, J. D. Litster, Physical Review Letters 1986, 57, 1012. J. D. Brock, A. Aharony, R. J. Birgeneau, K. W. Evans-Lutterodt, J. D. Litster, P. M. Horn, G. B. Stephenson, A. R. Tajbakhsh, Physical Review Letters 1986, 57, 98. D. Y. Noh, J. D. Brock, J. D. Litster, R. J. Birgeneau, J. W. Goodby, Physical Review B 1989, 40, 4920. R. Suriyakala, V. G. K. M. Pisipati, Phase Transitions: A Multinational Journal 2004, 77, 281 P. Swathi, S. S. Sastry, P. A. Kumar, V. G. K. M. Pisipati, Molecular Crystals and Liquid Crystals 2001, 365, 523. M. J. Vaz, Z. Yaniv, R. Y. Dong, J. W. Doane, Journal of Magnetic Resonance (1969) 1985, 62, 461. S. Chandrasekhar, Liquid Crystals, pg. 41, 2 nd ed., Cambridge University Press, Cambridge, 1992. J. Godzwon, M. J. Sienkowska, Z. Galewski, Journal of Molecular Structure 2007, 844-845, 259. L. J. Martnez-Miranda, A. R. Kortan, R. J. Birgeneau, Physical Review A 1987, 36, 2372. T. Niori, T. Sekine, J. Watanabe, T. Furukawa, H. Takezoe, Journal of Materials Chemistry 1996, 6, 1231. S. B. Rananavare, V. G. Pisipati, E. W. Wong, in Liquid Crystal Materials, Devices, and Applications III, Vol. 2175, 1 ed., SPIE, San Jose, CA, USA, 1994, pp. 108. S. B. Rananavare, V. G. K. M. Pisipati, E. W. Wong, Physical Review Letters 1994, 72, 3558. G. S. Attard, S. Fuller, O. Howell, G. J. T. Tiddy, Langmuir 2000, 16, 8712. D. Blunk, N. Bongartz, C. Stubenrauch, V. GaĂ&#x152;Ë&#x2020;rtner, Langmuir 2009, 25, 7872. S. Fuller, N. N. Shinde, G. J. T. Tiddy, G. S. Attard, O. Howell, Langmuir 1996, 12, 1117. T. Carlsson, B. Zeks, A. Levstik, C. Filipic, I. Levstik, R. Blinc, Physical Review A 1987, 36, 1484.

Shankar B. Rananavare & V.G.K.M. Pisipati

97. I. Musevic, B. Zeks, R. Blinc, T. Rasing, P. Wyder, Physical Review Letters 1982, 48, 192. 98. N. A. Clark, S. T. Lagerwall, Applied Physics Letters 1980, 36, 899. 99. H. Takezoe, E. Gorecka, M. Ccaronepiccaron, Reviews of Modern Physics, 82, 897. 100. A. D. L. Chandani, E. Gorecka, Y. Ouchi,. H. Takezoe, A. Fukuda, Jpn. J. Appl. Phys 1989, 28, L1265. 101. J. Als-Nielsen, R. J. Birgeneau, M. Kaplan, J. D. Litster, C. R. Safinya, Physical Review Letters 1977, 39, 352. 102. S. B. Rananavare, A. J. I. Ward, D. W. Osborne, S. E. Friberg, H. Kaiser, The Journal of Physical Chemistry 1988, 92, 5181. 103. A. J. I. Ward, S. E. Friberg, D. W. Larsen, S. B. Rananavare, The Journal of Physical Chemistry 1984, 88, 826. 104. A. J. I. Ward, S. E. Friberg, D. W. Larsen, S. B. Rananavare, Langmuir 1985, 1, 24. 105. A. J. I. Ward, S. B. Rananavare, S. E. Friberg, Langmuir 1986, 2, 373. 106. E. Meirovitch, E. T. Samulski, A. Leed, H. A. Scheraga, S. Rananavare, G. Nemethy, J. H. Freed, The Journal of Physical Chemistry 1987, 91, 4840. 107. S. E. Friberg, S. B. Rananavare, D. W. Osborne, Journal of Colloid and Interface Science 1986, 109, 487. 108. P. Poulin, H. Stark, T. C. Lubensky, D. A. Weitz, Science 1997, 275, 1770. 109. T. C. Lubensky, D. Pettey, N. Currier, H. Stark, Physical Review E 1998, 57, 610. 110. G. M. Koenig, I.-H. Lin, N. L. Abbott, Proceedings of the National Academy of Sciences 2010, 107, 3998. 111. S. B. Rananavare, M. Silverberg, E. W. Wong, C. W. Sheeleigh, Oregon Graduate Institute, 1992. 112. H. Christenson, S. E. Friberg, D. W. Larsen, The Journal of Physical Chemistry 1980, 84, 3633.

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 53-93 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

3. p-Carborane liquid crystals: Nematogenic properties and potential application Adam Januszko Military Institute of Engineer Technology, 136 Obornicka, 50-961 Wroclaw, Poland

Abstract. A series of 10-vertex or 12-vertex p-Carborane containing liquid-crystalline compounds have been investigated and compared with their carbocyklic analogues: bicycle [2.2.2] octane (Bco), cyclohexyl (CHx) and benzene (Ph). The presence of inorganic p-Carborane cages has a vital impact on the change of thermal, dielectric and optical properties of calamitic liquidcrystalline materials. The impact of p-Carborane cage on the generation of nematic phase has been discussed. The perspective application of p-Carborane liquid crystals for thermal neutron radiation is presented.

1. Foreword The analysis of scientific literature which covers the study of liquid crystal compounds reveals that with each decade the number of new organic compounds exhibiting liquid-crystalline properties raises by from a few Correspondence/Reprint request: Dr. Adam Januszko, Military Institute of Engineer Technology, 136 Obornicka, 50-961 Wroclaw, Poland. E-mail: januszko@witi.wroc.pl

Adam Januszko

Figure 1. P-Carborane cages: 1) C2B8H8 (ten-vertex cage – 10V), 2) C2B10H10 (twelve-vertex cage – 12V). Top vertexes are marked with a gray dot corresponding with C–H bond; the remaining vertexes correspond with B–H.

thousand to several thousand. The annually updated Vill’s database on liquidcrystalline compounds [1] is supplemented with 10 00 compounds each year. The latest LiqCryst 4.6 lists over 120 000 compounds, whereas 20 years ago only mere 10 00 were known [2]. Such wide interest in compounds representing “unusual state of the matter” [3] is rooted both in human curiosity, of the urge to discover the secrets of the world and also stems from the resulting development of technologies which utilize liquid-crystalline properties in information imaging equipment and in LCD among other. The search for materials which would fulfil the requirements of advanced technology, the new phenomena being discovered and the effects found in liquid-crystalline materials cause the research to turn to previously uncharted areas. Material engineering is focused not only on modelling and synthesis, but also on utilizing the physico-chemical properties of the products. One can distinguish several groups of liquid-crystalline materials, either according to their structural features, or on their properties. Among other, particular attention should be paid to the materials which contain carborane cages, namely: 1,10-dicarbo-closo-decaborane (10-vertex – 10V in short) and 1,12dicarbo-closo-dodecaborane (12-vertex –12V in short). The location of the above-mentioned cages in calamitic (i.e. rod-shaped) liquid-crystalline molecules results in generation of mostly the nematic phase (only a few carborane compounds are known to produce smectic phases). Utilising these materials in neutron radiation measurement can be viewed as a chance for these materials to be used in neutron radiation dosimetry.

1.1. Introduction From the moment they had been discovered in 1888 [4], liquid crystals had to wait a long time to find application. Researches have stated to become interested in liquid crystals only in the recent forty years. It was not only due

p-Carborane liquid crystals: Nematogenic properties and potential application

to their unique physico-chemical properties but due to development of technology which made application of liquid crystals in optoelectronic devices possible. Widely used nowadays in LCD, for many years liquid crystals could not be practically applied because of technological difficulties. Many other applications have been searched for. During the Cold War the Great Powers with vast arsenal of nuclear weapons fought about the dominion over the world, there existed a potential danger to the civil population, a threat caused by ionizing radiation. The possibility of developing a dosimeter, a device measuring ionizing radiation enabling direct readout of the results using liquid-crystalline material picked up interest of a number of scientists on both the sides of the Iron Curtain. The works on using a cholesterol derivative for the detection of gamma radiation conducted by the Alfassi group indicated that cholesteric LCs can be utilized in the detection of gamma radiation in the spectrum ranging from tens of krads to Mrad [5,6]. The impact of proton radiation on cholesteryl nonanoate and cholesteryl chloride was studied by Lavrentovich [7]. The findings were that the temperature of the material changes by 0.01oC (measures under 600nm) after it had absorbed the radiation dose of 1kGy. Further works pointing to lack of significant sensitivity of cholesteric materials to gamma radiation suggest continuation of the research towards increasing this sensitivity by two orders of magnitude [8]. The results of radiation sensitivity evaluation, this time to electronic radiation (beta radiation) were presented by Dhar [9]. The maximal dose of the absorbed radiation which was 33 kGy and the tests resulted in total destruction of nematic phase of the studied 4-cyano-4’pentylobiphenyl (5CB) and 4-cyano-4’-octylobiphenyl (8CB). The dose was determined using liquid-crystalline compound dielectric anisotropy inversion [10]. In 8CB liquid-crystalline material such a dose resulted in drop of Δε to 0. What results from the above analysis is that liquid-crystalline materials which considering their chemical structure are organic compounds consisting mainly of carbon (C) (on the average 60-85% by weight) and H, O, N, F atoms, and which had been subject to ionizing radiation (gamma-, beta- or proton radiation (γ, β, p) do not exhibit changes of physico-chemical properties in absorbed radiation range event up to a few kGy (103 Gy). Further, they are chemically stable substances, the value of annual radiation absorbed dose limit is 5 cGy) [11]. In order to utilize any liquid-crystalline material for measurements of the ionizing radiation absorbed doses its sensitivity should be increased by five orders of magnitude. One can state that the resistance of liquid crystals to gamma radiation is their advantage. The commercially available, AN/UDR-13 Military Pocket Radiac [12], military dosimeter detects dose rate of residual radiation units from 0.001 cGy/hr to 999 cGy/hr utilises a standard LC display for the readout. The search for

Adam Januszko

liquid-crystalline materials sensitive to radiation should therefore focus on the atomic level. When the chemical structure of liquid-crystalline molecules is analysed with regard to their potential application in ionizing radiation detection, one should search for such atoms whose ionization radiation cross section would be much greater than the cross section of atoms from which they are composed, that is: C, H, O, N (the standard composition for organic compounds). The method of destroying brain cancer cells used in medicine, the so-called Boron Neutron Capture Therapy (BNCT) utilizes neutron radiation, together with chemical compounds which contain boron 10B isotope [13÷16]. In BNCT alpha particles are used in cancer treatment. They are the products of boron 10 B isotope decay and are formed following the neutron capture according to the nuclear reaction (1):

Boron 10B isotope is the active element of chemical compounds used in this sort of therapy. Its thermal neutron cross section equals 3700b (1b=10-24cm2). For comparison purposes: C, H, O thermal neutron cross section equals circa 1b. The analysis of other studies on the characteristics of carborane liquid crystals [17÷22] revealed the possibility of developing carborane liquid-crystalline compounds with such physico-chemical properties that they could be applied in neutron radiation detection. The condition indispensable for such an application is the existence of mesophase (preferably the nematic phase) in wide temperature range and as high as possible number of boron atoms in a liquid-crystalline molecule (in the form of carborane cages). This chapter is dedicated to the modelling, development and tests of such materials for the purpose of perspective neutron radiation dosimetry.

1.2. Modifications of a liquid-crystalline molecule (molecule structure) Liquid-crystalline compounds which belong to the calamitic group are aromatic by nature and assume the shape of a rod. Their simplified structure is illustrated in figure 2. R1 and R2 terminal substituents are included in groups, such as the alkyl group –CnH2n+1, the alkoxy group –OCnH2n+1, the nitro group –NO2, the cyano group –CN, and the isothiocyanate group –NCS, whereas the letters “L1” and “L2” represent the linking groups, such as: CH=N, N=CH, COO, OOC, N=N and others [23÷26].

p-Carborane liquid crystals: Nematogenic properties and potential application R1

Figure 2. Schematic shape of a calamitic liquid crystal molecule. Y1 R1

L1 n

Y2 Y3 K3

Y4 K4

Figure 3. Possible modification of the shape of a calamitic liquid crystal molecule.

Subsequent years of search for liquid-crystalline materials with new properties lead to significant modification of the molecules themselves. An exemplary complex structure is presented in figure 3. Cages from K1 to K4 can represent benzene rings (Ph), cyclohexyl rings (CHx), bicycle [2.2.2] octane (Bco) or p-Carborane 10-vertex or 12-vertex. Substituents Y1 to Y4 represent H or F atoms, while the values of n and m multipliers can equal 0,1,2. The cyano or alkoxy terminal substituents raise the temperature of mesophaseâ&#x20AC;&#x201C;isotropic liquid phase transition (TM-I â&#x20AC;&#x201C; isotropisation temperature) more than any other substituents. The ratio of alkyl- or alkoxy chain length to the length of the molecule is also a deciding factor, as far as isotropisation temperature is concerned. Increase of the length of n-alkyl chain broadens the range of mesophase formation: with shorter homologous series, nematic phase appears, while with increase of the chain length, the tendency to produce smectic phase A (SmA) appears [27,28]. The properties of L1 and L2 linking groups are key elements in maintaining the stability of the mesophase. The linking group containing multiple bonds increases the rigidity and linearity of the molecule. The study of carborane-containing liquid-crystalline materials started with the derivatives of the ester bond group due to planar properties and highly nematogenic tendencies exhibited by the ester bond.

2. Carborane-containing liquid-crystalline compounds 10- and 12-vertex (fig. 1) p-Carboranes form an inorganic cage structure which is an effective element of rod-shaped liquid-crystalline molecules. The presence of inorganic p-Carborane cages has a vital impact on the change of thermal-, dielectric- and optical properties of liquid-crystalline materials.

Adam Januszko

Figure 4. Structure of the tested liquid-crystalline compounds with ester linking group. Denotation of R, R1 and R2 substituents is explained in the text.

While terminal substituents (R1, R2 â&#x20AC;&#x201C; fig.3) reduce the melting point of a liquid-crystalline compound, it is the rigid mesogenic core which, by introducing anisotropic interaction, decides upon the macroscopic properties of the material, such as its mesophase stability and photochemical stability.

2.1. Three- and four-ring p-Carborane mesogenic esters The studies of liquid-crystalline compounds containing a 10-vertex or 12-vertex p-Carborane cage were conducted relative to their carbocyclic analogues: bicyclo[2.2.2] octane (Bco), benzene (Ph) and cyclohexyl rings

p-Carborane liquid crystals: Nematogenic properties and potential application

(CHx). This indicates that the efficiency of mesophase stabilization obtained by introducing p-Carborane cage to the rod-shaped molecule is generally lower than the efficiency of Bco, Ph, CHx analogues [15]. The sequence is justified by the size, conformation and quadrupole properties of the above-mentioned carbocyclic cages which decide about the intermolecular interactions and molecular weight distribution. The closest topological analogue of a 12-vertex p-Carborane cage is bicyclo[2.2.2] octane [29, 30]. This dependence was also observed in carborane mesogenic esters, the structure of which is illustrated in Fig. 4. All the carborane analogues in group one exhibit only the nematic phase. Nematic phase range (ΔTN) changes from 50 ºC to 94ºC, and isotropisation temperature (TN-I) is between 96 and 211ºC. Carborane analogues in group two (Figs. 4 II) exhibit strong nematogenic tendency with nematic phase range being from 70 ºC to 89ºC (additionally, generation of monotropic smectic phase is observed), while the isotropisation temperature ranges between 129 and 167ºC. Carborane analogues in group three (Figs. 4 III) are characterised by mesogenic polymorphism in temperatures above 114ºC which in one of the tested compounds dwindles, generating nematic phase only in temperature of 84ºC and above. The impact of 10- and 12-vertex carborane cage on the properties of group I, II, and III liquid-crystalline compounds (Fig. 4) was investigated on exemplary compounds (carborane-containing carboxy derivatives) collected in Tables 1 and 2. The change of thermodynamic properties of the analysed mesogenic esters (Tab. 1 and Tab. 2) depending on the location of the p-Carborane cage in the molecule as expressed by temperature ranges of crystal, nematic, and smectic SmA phases are illustrated in Fig. 5 and Fig. 9. The studied p-Carborane mesogens (Fig. 5) which contain a carboxy group −COO− generate only the enantiotropic nematic phase whose range is between 35ºC (Ic4) and 94ºC (Ib3), whereas the melting point for a compound containing two 12-vertex p-Carborane cages (Ia4/Ib1) changes from 41ºC (Ic2) to 177ºC. Ia4/Ib1 compound shows only monotropic nematic phase and that is at two degrees below its melting point. The greatest range of nematic phase (94ºC and 92ºC) can be observed in compounds (Ib3 and Ib4, respectively) in molecules of which three different rings: 12-vertex, Ph and CHx are simultaneously present.

Adam Januszko

Table 1. Group I of the studied p-Carborane derivatives with ester linking group.

In a molecule, the presence of two 12-vertex rings simultaneously or of a 12-vertex ring with a Bco ring, which is the closest topological analogue of a 12-vertex ring, significantly increases the melting point and decreases the phase range. In Ia4/Ib1 compounds, substitution of a 12-vertex cage with a

p-Carborane liquid crystals: Nematogenic properties and potential application

Table 2. Group II and III of the studied p-Carborane derivatives with linking ester group.

Bco cage, either on the side of C atom in −COO− (Ia3) ester bond or on the side of O-atom (Ib2), significantly decreases the melting point by 20ºC and 16ºC, respectively. This leads to increase of the nematic range by 53ºC and 52ºC, respectively.

Adam Januszko

Figure 5. Temperature ranges for carborane derivatives with COO linking group: crystalline (Cr) and nematic phases (N). The compound symbols (Y axis) are consistent with Table 1.

Further substitution of a Bco cage with a CHx cage in Ia2 p-Carborane analogue further decreases the melting point by additional 56ºC, and with Ib3 analogue − even by 82ºC. The mesophase range increases by additional 22ºC and 44ºC. Another substitution of a CHx ring with a benzene (Ph) ring effects in both the decrease of the melting point of Ia1 and Ib4 analogues by 13ºC and 8ºC, respectively, and slight change of the nematic phase range by +3ºC and by -2ºC. Nematic phase stability depends from molecular structure and position of p-Carborane cage to −COO− group (Fig. 7). Despite the fact that in compounds belonging to the Ia group where 12-vertex cage is present on the side of −COO− bond (it is connected to it via a Ph ring) a similar trend towards nematic phase stability is observed, when compared with the Ib group whose 12-vertex cage is located on the opposite side of the −COO− bond (on the side of C atom), the difference in nematic phase range is observed. The Ia compounds exhibit higher melting point TCr-N, as compared with the compounds of Ib group, but lower clearing temperature TN-I (Fig. 7). No significant difference has been observed, as far as the mesophase stability is concerned in compounds where one of 12-vertex cages has been substituted with a Bco ring (Ia3 and Ib2 compounds). Regardless of the Bco position in relation to –COO– bond, in both these cases what can be seen is

p-Carborane liquid crystals: Nematogenic properties and potential application

the same nematic phase range (50ºC) and similar melting points (157ºC and 161ºC). Definite broadening of the nematic phase range is observed, however, when 12-vertex cage, located on the side of O-atom in the −COO− bond, is substituted with CHx or Ph rings (compounds: Ib3, Ib4). In spite of the fact that a similar trend is observed in Ia1 and Ia2 compounds where CHx or Ph rings are located on the side of C atom in the −COO− bond, in Ib3 and Ib4 compounds the difference in the mesophase range is greater by 20ºC. The analysis of mesogenic properties based on compound structure with two 12-vertex cages (Ia4/Ib1) indicates that the following nematogenic trend is exhibited by the rings/cages: 12V < Bco < CHx < Ph (Fig. 7). The transfer of one phenyl ring so that it is in direct vicinity of a 12-vertex cage (Ic compound group) results in decrease of melting point by 7ºC (Ic1, Ic3), as compared with the Ib4 compound. One witnesses here a drastic R

COO

Figure 6. Simplified structural diagram of liquid-crystalline compounds belonging to group I, (R = C5H11, A and B = 12V, Ph, CHx, Bco).

Figure 7. Difference in Cr-N phase transition temperatures of compounds from group I, including Ia4/b1 compound. The compound symbols are consistent with those in table 2. A and B correspond with structural elements presented in Figure 6.

Adam Januszko

drop of nematic phase range by 35ºC and 48ºC (Fig. 8), respectively. The increase of alkyl chain length from C5 to C7 markedly decreases the melting point of Ic2 compound by 19ºC. This relationship is not observed in Ic3, where alkyl chain is lengthened and O-atom is connected with the phenyl ring. Substitution of benzene ring with cyclohexyl ring Ic4 is responsible for a considerable increase of melting point: by 22ºC, in relation to Ib4; by 29ºC in relation to Ic1 and Ic3 phenyl analogue, and by 48ºC in case of Ic2. This last compound is characterised by the lowest melting point, as far as this group is considered. The analysis conducted confirms a general trend towards destabilization of nematic phase which accompanies extension of the alkyl chain for compounds which belong to groups II and III (Fig. 9), as compared with Ib4 compound (wide nematic phase range) and Ic2 compound (low melting point). The extension of alkyl chain from C5 (Ib4 compound) to C8 (IIa1 compound)

Figure 8. Difference in Cr-N and N-I phase transition temperatures of compounds from group Ic including Ib4 compound. The compound symbols are consistent with those in Table 1.

p-Carborane liquid crystals: Nematogenic properties and potential application

Figure 9. Temperature ranges for carborane derivatives with COO linking group: crystalline (Cr), smectic (SmA) and nematic (N) phases. The compound symbols (Y axis) are consistent with those in Table 1.

decreases the melting point by 17ºC with simultaneous minor (3ºC) decrease of mesophase range (compound IIa1). The extension of Ib4 compound molecule by additional cyclohexyl ring changes the mesomorphic properties of the IIIa compound entirely. In IIIa and IIIb four-ring esters what was observed in relation to IIa1 compound, was increase of melting point by 55ºC and 59ºC (respectively) with simultaneous broadening of the nematic phase by 94ºC and 86ºC and the appearance of enantiotropic smectic SmA phase whose range is 8ºC and 13ºC, respectively (Tab. 2 and Fig. 9). The substitution of hydrogen atom with fluoride atom connected with the carboxyl group of a four-ring ester (IIIb compound) has a minor impact on both the melting point and mesophase range in case of the smectic SmA (decrease by -4ºC) and nematic phase (increase by 4ºC). The substitution of the middle ring in phenyl ring of IIIa compound with ethyl bond −CH2=CH2− (IIIc) causes the total decline of the smectic SmA phase with accompanying lowering of the melting point by 42ºC. This also has an impact on significant destabilization of the nematic phase, whose temperature range is depressed by 123 ºC. Among the three-ring esters, IIa1 compound is characterised by the lowest melting point and at the same time by broad nematic phase range:

Adam Januszko

89ºC. When this compound is altered and modified by replacing one of the phenyl rings with a pyrimidine ring (IIa3 compound) this results in a significant, 66ºC, increase of the melting point with simultaneous decrease of nematic phase by -54ºC. The same dependence is observed among the 10vertex carborane derivatives (IIb compounds), where the substitution of the phenyl ring with a pyrimidine ring results in increase of the melting point by 36ºC and decrease of nematic phase range 39ºC (IIib3 and IIb4 compounds). There is a general trend of melting point increasing with simultaneous decrease of nematic phase in pyrimidine derivatives, also in compounds where alkyl chains was substituted with alkoxy chains (−O−C8). The substitution of the phenyl ring in IIa compound with pyrimidine ring causes increase of the melting point by 24ºC and drop of the phase range width by 16ºC. Similarly as in case of 10-vertex derivative (IIb2), the introduction of a pyrimidine ring (IIb4) increases the melting point by 20ºC and decreases the phase range by -19ºC. Notwithstanding all pyrimidine derivatives are characterised by high melting point, one should note that the introduction of an O-atom between the phenyl-alkyl chain link significantly broadens nematic phase range with

Figure 10. Difference in Cr-N and N-I phase transition temperatures T(IIb)–T(IIa) of compounds from group II, series 1, 2,3 and 4. The compound symbols are consistent with those in Table 2.

p-Carborane liquid crystals: Nematogenic properties and potential application

coinciding decrease of melting point. The substitution of octyl ring with octyloxyl ring in a 12-vertex derivative (IIa4) decreases the melting point by 12ºC with simultaneous considerable (even by 36ºC) broadening of the nematic phase range. In case of a 10-vertex derivative (IIb4) these values are 13ºC and 39ºC, respectively. A contrary tendency is seen in compounds with no pyrimidine ring (IIa1, IIb1). The substitution of octyl rings with octyloxyl ring increases the melting point by 30ºC in IIa2 compound, and by 3ºC in case of IIb2 compound. The substitution of 12-vertex cage with 10-vertex cage in group II compounds with terminal octyl and octyloxyl substituents depresses both the melting point and isotropisation temperature (series 1 and 2). However, this hardly results in temperature changes in compounds with pyrimidine ring in series 3 and 4 (Fig. 10). The above analysis of carborane mesogen structures visualizes the dependence of mesophase stability on the location of p-Carborane cages in a given molecule. What is observed here is decidedly a more positive impact of the terminal location of p-Carborane cage on nematic phase stability than the one exerted when the cage is located within the rigid molecular core. Steric properties of p-Carborane cages and their location in rigid molecular core play a pivotal role in mesophase stability.

2.2. Analysis of p-Carborane mesogenic nematic phase stability with respect to carbocyclic analogues The analysis of nematic phase destabilization in mesogenic ester derivatives with respect to the type of spatial cage or flat ring and their location in the molecule is presented in Fig. 11. Unlike carborane derivatives, all the other analogues without a 12-vertex cage are characterised by rich smectic polymorphism. All the other carbocyclic derivatives, apart from a relative broad nematic phase range, generate enantiotropic orthogonal smectic phases [31]. Only in the compound where A = B = Ph generation of monotropic tilted smectic phases: SmC and G was noted. Due to diversity of smectic phases generated by carbocyclic analogues, only the nematic phase underwent mesophase stability analysis. The most stable nematic phase (TN-I=262ºC) can be observed in compounds which contain two Bco rings (A = B = Bco, Fig. 11) and which contain isolated benzene ring in the central area of the molecule. In spite of the preferred arrangement of electrons π, the analogue containing in its structure all the aromatic rings A = B = Ph has a decidedly lower clearing temperature, that is 87ºC. Thus it seems that the deciding role in nematic

Adam Januszko

COO

Figure 11. Isotropisation temperatures for C5−A−COOPh−B−C5 esters. A = X axis, B = Z axis.

phase stability belongs to the steric effect and not the electronic effect. The observed prevailing trend of clearing temperatures of individual analogues from series C5−A−COOPh−B−C5 is as follows: Bco>CHx>Ph>12V (Fig. 11). The analysis of the dependence of liquid-crystalline properties on their structure is illustrated in a Fig. 12, where the differences in clearing temperatures of analogues are compared with the ones characteristic for the biphenyl series, where B = Ph. Replacement of one Ph ring with a Bco ring in a biphenyl derivative results in increase of clearing temperature by over 40ºC in all the analogues, whereas substitution using a 12-vertex cage decreases the clearing temperature on the average by 15ºC. What is surprising, in carbocyclic analogues such substitution results in an increase by 20ºC. One should note here no impact of the substitution of benzene ring (Ph) with a cyclohexyl ring (CHx) on clearing temperature which increased by 10ºC only in an analogue where A=12V. The isotropisation temperature behaviour illustrated in Fig. 12 characteristic for liquid-crystalline compounds as dependent on the location

p-Carborane liquid crystals: Nematogenic properties and potential application

Figure 12. The difference of isotropisation temperatures for C5−A−COOPh−B−C5 compounds in function of change of substituent A of the series of analogues for B = CHx, 12V, Bco in relation to the series, where B = Ph. The lines are guides for the eye.

of a large substituent suggests that the stability of nematic phase increases along with increase of cage/ring volume. This arises due to high filling factor also known as molecular weight distribution defined as the relation of van der Waals molar volume to the geometric volume of the cylinder circumscribed on a molecule [34]. Its influence on the mesophase stability is most clearly visible when di-p-Carborane analogues are compared with biphenyl ones [30,35]. Figure 13 illustrates the impact of the filling factor on the nematic phase stability. In a compound of the following structure C5−A−COOPh−B−C5, the substitution of a phenyl ring with a cyclohexyl ring in location A increases the stability of the nematic phase on the average by 13ºC and by 45ºC when the phenyl ring is substituted by the Bco ring. When substituent A assumes a 12-vertex form in the above series of analogues, the impact on mesophase stability in function of the cage type is much more complex. A compound comprising 12-vertex cage and two aliphatic rings (Bco, CHx) has a moderate impact on decrease of nematic phase stability (by 7ºC and 5ºC, respectively) in relation to the phenyl analogue. For comparison purpose: a compound in the structure of which two large 12-vertex cages are present

Adam Januszko

Figure 13. The difference of isotropisation temperatures for C5−A−COOPh−B−C5 compounds in function of change of substituent A series of analogues for B =CHx, 12V, Bco in relation to the phenyl analogue, where A=Ph. The lines are guides for the eye.

(C5−12V−COOPh−12V−C5) increases nematic phase stability by 20ºC. This results from the difference of filling factor in the nematic phase which, in turn, results from the difference of cage/ring size, the cages/rings being structural components of the molecules. The same relationship can be observed in the series of compounds II (tab.2). The expansion via carbocyclic analogues (Bco and Ph) allows for analysis of the impact of the kind of spatial substituent on the mesophase stability. Each of the four analysed analogue compounds of the C5−A−COOPh−B (B= Ph−C8; Ph−O−C8; Py−C8; Py−O−C8, where Py is the pyrimidine ring) on the whole maintains the same sequence of clearing temperatures depending on the kind of the ring/cage present A: Bco>Ph>12V>10V (Fig. 14). The only exception is series 3 where the clearing temperature of the 12-vertex derivative is 2ºC lower than the one of the Ph derivative. All the carborane derivatives (the 10- and 12-vertex ones) containing a pyrimidine ring generate the enantiotropic nematic phases only. Biphenyl carborane derivatives (series 1 and 2, Fig. 14), apart from the nematic phase, generate monotropic smectic phases and, furthermore, in −4ºC the G phase which can be observed in C5−10V−COOPh− Ph−C8 analogue.

p-Carborane liquid crystals: Nematogenic properties and potential application

Figure 14. Isotropisation temperatures for C5−A−COOPh−B esters (group II tab.23) A= Z axis, B= X axis, where 1=Ph−C8; 2=Ph−O−C8; 3=Py−C8; 4=Py−O−C8 (Py is the pyrimidine ring).

Figure 15 presents graphic comparison of clearing/isotropisation temperature TN-I of series 1÷4 in relation to isotropisation temperature of biphenyl derivative (A = Ph). The observed impact of structural changes on mesophase stability (in relation to Ph) seems to be similar for all the analysed three-ring ester derivates: 12-vertex, 10-vertex and Bco ones. On the average, the substitution of a Ph ring with a 10-vertex cage lowers the clearing temperature by 30ºC in all the series. Lowering of clearing temperature by 20ºC is observed in the carborane derivatives containing biphenyl in their structure (series 1 and 2), whereas no difference in clearing temperatures has been noticed for carborane derivatives containing pyrimidine rings (series 3 and 4, Fig. 15). Substitution of the benzene ring (Ph) with a Bco ring clearly increases clearing temperature in all the analysed series, 1 through 4, by 40ºC on the average. The substitution of one of the benzene rings with a pyrimidine ring, what can be presented as substitution (in one of the biphenyl rings) of two CH groups (series 1 and 2) with N-atoms (series 3 and 4) has minor impact on the difference in clearing temperatures for 10-vertex analogues (10ºC) and only a little greater in case of 12-vertex analogues (20ºC). This very weak

Adam Januszko

Figure 15. Differences in isotropisation temperatures for C5−A−COOPh−B esters belonging to group II, as compared with isotropisation temperature of phenyl analogue. Axis X: 1= Ph−C8; 2= Ph−O−C8; 3= Py−C8; 4= Py−O−C8. The lines are guides for the eye.

tendency towards forming lamellar phases by carborane derivatives (they generate monotropic phases only) indicates a significant difference of thermotropic properties in comparison with the carbocyclic analogues (these generate four enantiotropic smectic phases). The difference in properties of structurally similar compounds is the result of steric differences in hydrocarbon rings and carborane cages and the dynamic intermolecular interactions. The size and the shape of individual substituent determine the rigidity of the molecular core, while the magnitude of the barriers to internal rotation leads to energy minimum conformers (Fig. 16). The comparison of four-ring carborane esters C5−A−COO−Ph−Ph−CHx−C5 belonging to IIIa group (Tab. 2) and this group’s modification in the form of IIIb and IIIc, in relation to its carbocyclic analogues (A= Bco, CHx, Ph) draws attention to the fact that on the whole all the esters from his group exhibit strong polymorphism (apart from the nematic phase they generate orthogonal smectic phases and the so-called soft crystals). Only 12-vertex analogue (IIIc compound) exhibits the nematic phase only, even having been cooled down by 50°C below the melting point. All the ester analogues from series IIIa and IIIb are characterised by high clearing temperatures – above 290°C.

p-Carborane liquid crystals: Nematogenic properties and potential application

Figure 16. Ring/cage symmetry.

The highest clearing temperatures was noted for the Bco derivatives and they are as high as over 350°C. Decidedly lower clearing temperature (by ca. 160°C) is characteristic for the three-ring esters from the IIIc group, in the structure of which one of the benzene rings was substituted with the −CH2−CH2− ethyl bond (Fig. 17). The comparison of clearing temperatures indicates that Bco derivatives strongly stabilise nematic phase, on the average by 45°C, as compared with the benzene derivatives (Ph). The clearing temperatures of IIIa and IIIc carborane esters are akin to their phenyl analogues (the difference is 8°C). The introduction of fluoride atom to one of the benzene rings (IIIb compound) influences the stability of the carborane analogues’ nematic phase in a dissimilar way when compared with the carbocyclic analogues (Fig. 18). On the whole mono-fluorination of one of the benzene rings lowers clearing temperature for the Bco, CHx, Ph carbocyclic analogues (the strongest destabilisation is visible in phenyl analogue and equals -21°C), whilst in a 12-vertex carborane analogue it is increased by 2°C. What is interesting is the similarity of the Bco curve shape to the one illustrating the 12-vertex derivative (Fig. 18) which is lowered only for all the compounds of group III, by 45°C on the average. This dependence clearly indicates that 12-vertex carborane substituent has greater impact on the drop of clearing temperature than the Bco or CHx substituents do. The observed difference of isotropisation temperatures in individual analogues changes inversely proportionally to the size o the ring/cage (Fig. 19).

Adam Januszko

Figure 17. Sm-N and N-I phase transition temperatures for analogues belonging to group III (C5−A−COO−Ph−Ph−CHx−C5 where A = 12V, Bco, CHx, Ph). ΔTN-I [°C] 75 12V

Bco CHx

T [°C]

-15 0

III1a

III2b

III3c

Figure 18. The difference in clearing temperatures of 12-vertex, Bco and CHx derivatives in relation the Ph derivative IIIa, IIIb and IIIc group of compounds.

The analysis of geometric sizes of individual rings/cages points to the steric effect where they shield the lateral substituents [36], among which a 12-vertex cage is the most efficient one. Due to the rich smectic polymorphism of group III compounds, what is necessary here is the analysis of individual spherical substituents on the destabilisation of the smectic phase. The trend towards destabilization of the smectic phase for compounds

p-Carborane liquid crystals: Nematogenic properties and potential application

0,0

7,5 1,0

7,0 2,0

6,5 3,0

° [A] 5,0

6,0 4,0

Δ T N-I [°C]

-40 IIIb-IIIa IIIc-IIIa

-80

-120

-160

-200

12V

Bco

CHx

Figure 19. Differences in isotropisation temperatures for C5−A−COOPhPh−C analogues belonging o group III, as compared with isotropisation temperature of the phenyl analogue. The values on top of the diagram are the diameters of Van der Waals cylinders formed by rings/cages rotating around the axis by lateral substituents.

which belong to group III (Fig. 20) in relation to the phenyl analogue resembles the trend of nematic phase destabilisation (Fig. 18). The smectic phase is much more stable for the Bco and CHx substituents than it is for the Ph substituent, while a 12-vertex substituent destabilises the smectic phase entirely. The substitution of the middle Ph ring with a −CH2=CH2− bond totally disturbs the occurrence of the smectic phase of a carborane derivative with simultaneous noteworthy decrease of the nematic phase range by 110°C. In the Bco and Ph analogues the same manipulation causes the nematic phase to broaden. The analysis of four-ring esters from group III (C5−A−COO−PhPhCHx−C5) indicates the following sequence of nematic phase destabilisation in relation to individual analogues among the IIIa and IIIc compounds: Bco>CHx>Ph>12V. Among the IIIb compounds, this sequence looks different: Bco>12V∼CHx>Ph. The smectic phase destabilisation sequence for all the esters in group III assumes the following order: CHx>Bco>Ph>12V. This trend is also visible among the three-ring esters.

Adam Januszko ΔT Sm-N [°C] 150 12V 100

Bco CHx

T [°C]

-50

-100 0

III1a

III2b

III3c

Figure 20. Sm-N phase transition temperatures for 12-vertex, Bco and Chx analogues in relation to the Ph analogue IIa, IIIb and IIIc group of compounds.

2.3. Modification of p-Carborane mesogens via fluorination and extension of the alkyl terminal substituents The presence of smectic phases among several of the liquid-crystalline carboranes (see: appendix 1) inspired the search for induction of the smectic phases in carborane mesogens via modification of terminal substituents. The impact of two- and three-ring esters (Fig. 21) on mesogenic properties exerted by extension of the alkyl chain and its pre-fluorination (partial fluorination) has been analysed for all the 12-vertex, Bco and CHx analogues and for the Ph analogues [37, 38]. While in carbocyclic group of nematogenic two-ring esters both the extension of the terminal alkyl chain [39, 40] and its pre-fluorination [41, 42] strongly generates smectic phase, in their carborane analogues these procedures prove ineffective. With the exception of the CHx analogue where a very narrow (1.1°C) smectic SmA phase range is noted [43], in all the two-ring ester analogues where R1 = C5H11, only nematic phase has been observed. If the alkyl chain is extended twice, that is from R1 −C5H11 to −C10H21, not only is the nematic phase not eliminated in the 12-vertex and Ph analogues, but additionally the extension results in broadening of the phase range by 12°C and 6°C.

p-Carborane liquid crystals: Nematogenic properties and potential application

Appendix 1. Phase transition temperatures (oC) and enthalpies (kJ/mol) of described compounds. X 12V

Bco

C5H11

COO

C5H11

COO

C5H11

COO

C5H11

COO

C5H11

Cr1 79 Cr2 88 N 158 I

Cr 101 N 174 I

Cr1 130 Cr2 157 N 208 I

Cr 177 (N179) I

(0,1) (29,8) (1,3)

(25,7) (1,2)

(1,3) (16,7) (1,7)

(19,7) (1,0)

Cr 119 N 218 I

Cr 59 SE 199 N 227 I

Cr1 144 Cr2 161 N 211 I

(27,4) (1,5)

(7,4) (9,5) (1,6)

Cr1 97 Cr2 159 E 231 N 262 I

(11,7) (15,1) (1,1)

(9,1) (5,4) (8,5) (1,3)

CHx

Cr1 76 Cr2 89 (SB 73) N 174 I (1,5) (3,6) (23,1) (1,6)

Cr98(G 81SmC82SmA 84) N175 I (26,5) (0,3) (1,0) (1,8) (1,3)

Cr 40 (SmX 30)a E 157 N 188 I (6,6) (24,6) (8,7) (1,4)

Cr 72 (SmX 57) E 176 N 219 I

Cr 75 N 169 I (24,2) (1,2)

(3,1) (10,4) (8,0) (1,9)

Cr44E61SmB138SmA15 Cr 85 SmB 125SmA141 N 3N188I (31,7) 218 I (1,1) (4,4) (0,8) (1,6) (26,3) (2,8) (0,1) (1,1)

Cr1 61 Cr2 67 N 159 I (0,7) (21,2) (1,0)

a) phase H or K

COO

R1 X

Figure 21. The structure of the two- and three-ring esters. R1 assumes the form: C5H11, C10H21, C3H6−C2F5 while R2= C8H17, C2H4-C6F13. X = 12V, Bco, CHx, Ph.

Identical extension of the alkyl chain in the CHx analogue causes the smectic SmA phase to broaden by 10°C, and a 20°C wide SmB phase to be generated. In the Bco the extension of alkyl ring results in decrease of nematic phase range by 22.5°C at the cost of the generated enantiotropic SmA phase (12.6°C) and monotropic smectic SmB phase. This confirms the results of homological series tests, where the increase of tendencies towards SmA phase generation is observed along with the increase of the alkyl ring length [27, 28]. The introduction of five fluorine atoms (R1=C3H6−C2F5) into the alkyl chain –C5H11 has a much stronger impact on the generation of the smectic phase than the extension of the alkyl chain by one additional length. In all the carbocyclic analogues of the two-ring esters the nematic phase has been completely replaced by the smectic SmA phase. What is more, smectic SmB

Adam Januszko

phase is generated in the CHx analogue. Perfluorination of the alkyl chain in a 12-vertex p-Carborane analogue, as differentiated from the cyclic hydrocarbon analogues, does not lead to smectic phase generation but results in a considerable destabilization of the nematic phase. The impact of individual alkyl chain modifications on the mesophase stability is presented in figure 22. Perfluorination of the alkyl chain in 12-vertex and CHx analogues has lesser impact on the mesophase stability than its extension has. Only in the Ph analogue a few-degrees increase of isotropisation temperature has been noticed in case of both chain modifications. Depending on the cage/ring type, two different trends can be seen, as far as the impact of terminal substituent change on mesophase stability. If the alkyl chain is twice as long, it has the following impact on isotropisation temperature: 12V竏ｼPh>CHx>Bco, whereas its perfluorination is most phase-destabilizing in the 12-vertex analogue. The trend then is as follows Ph>CHx>Bco>12V (Fig. 23).

Figure 22. Isotropisation temperatures (mesophase竏段sotropic phase) in function of the structural change introduced into the alkyl chain of 12-vertex, Bco, CHx, Ph two-ring ester analogues.

p-Carborane liquid crystals: Nematogenic properties and potential application

Figure 23. The difference in phase transition temperatures of 12-vertex, Bco, CHx, Ph two-ring ester analogues resulting from the extension of the alkyl chain and its partial fluorination.

The first trend reflects the general rule of asymptotic dependency between isotropisation temperatures for homology sequence up to the limiting value of 70°C [38]. The second one presents an exact opposite to the previously-presented trend connected with the cage/ring size [31, 32]. The above-presented impact of the change of terminal alkyl substituents which has been observed in the two-ring esters is not visible in the analysed group of the three-ring esters (Fig. 21). Perfluorination of the octyl chain (C8H17→C2H4−C6F13) increases both the melting point and clearing temperatures in all the analogues (12V, Ph, Bco) and additionally generates smectic phase in the 12-vertex analogue while broadening the smectic phases in Ph and Bco carbocyclic analogues. In a three-ring 12-vertex analogue perfluorination of the octyl chain significantly destabilizes the nematic phase, changing the phase range from 71°C to 4°C concurrently generating enantiotropic smectic SmA phase (whose width equals 61oC) and monotropic SmC smectic phase (8°C below the melting point). Just to compare: in a Ph analogue perfluorination of the terminal alkyl chain leads to the complete substitution of the nematic phase with the smectic SmA phase and to a significant broadening of the smectic

Adam Januszko

SmC phase. The Bco analogue shares the properties common for the two molecular analogues. On the one hand perfluorination of the Bco analogue (similarly as of the 12-vertex analogue) leads to complete substitution of nematic phase with smectic SmA phase, and on the other it increases the mesophase stability by 66°C (the value approximate to the values in the Ph analogue). The studies of the miscibility of the fluorinated and the non-fluorinated series in phase diagram of mixture of the 12-vertex and Ph analogues reveal the ideal miscibility previously observed in other carbocyclic series [19,20,52]. The results reveal compatibility of the perfluorinated carborane esters with carbocyclic hydrocarbon analogues. They can serve as functional phase in polar liquid-crystalline mixtures.

2.4. Modification of the properties of p-Carborane mesogens via the change of the linking group The diversity of L linking groups (Fig. 2) which are but small structural elements has a significant impact on the change of properties in mesogenic molecules. This occurs due to the molecules’ conformational properties, rigidity, polarity, electronic structure and the ability to form hydrogen bonds. This impact is moderated through electronic and steric properties of the rings connected to such a group. The studies on the impact of an L bond inserted between two Ph−L−Ph, CHx−L−CHx and Ph−L−CHx rings on mesogenic properties of the molecules indicate that aromatic rings prefer unsaturated bonds which transport electron couplings, whereas saturated cyclohexan rings are highly compatible with dimethyl bonds and ester group [45÷48]. p-Carboranes are a relatively large, almost spherical inorganic δ-aromatic cage [49] which is neither a Ph ring, nor a CHx ring analogue. The tests of the impact of the L linking groups directly in contact with the p-Carborane cage relative to the phenyl analogue demonstrated a considerable impact on mesogenic properties. These tests based on the three-ring series of compounds (fig.24). All the p-Carboranes derivatives (X = 12V) in both (n=0 and n=1) series are purely nematic, whereas the phenyl analogues are characterised by rich polymorphism. Only the 12-vertex derivatives with a four-atom CH2CH2COO linking group have no mesogenic properties, even having been cooled down by 50°C below their melting point. However, phenyl analogues of these compounds generate both the nematic and smectic phases (see appendix 2) [50]. The analysis of isotropisation temperatures of 12-vertex analogues proves that they are slightly influenced by the linking group L (Fig. 25).

p-Carborane liquid crystals: Nematogenic properties and potential application

Average isotropisation temperature of a series where n=0 equals 109±9°C, whilst in the series where n=1 average TM-I equals 132±8°C. Therefore introduction of an oxygen atom increases isotropisation temperature on the average by 22±2°C (series 12V n=1, fig.25). The behaviour of p-Carborane amide analogues (CONH bond) is an exception in this series. Their clearing temperatures are markedly lower than those of the average TM-I in both the series: by 56°C for n=0 and by 41°C for n=1, whereas introduction of an oxygen atom increases isotropisation temperature on the average even by 39°C, which is almost twice the value of the Ph analogue. As distinguished from the p-Carboranes in all the analysed biphenyls (X = Ph, fig.24) the impact of the linking group L on Appendix 2. Phase transition temperatures (oC) of described compounds.

(O)n

Figure 24. Three-ring series structure. X=12V or Ph; L=(CH2CH2; COO; CH=CHCOO; CH=N; CH=CH; CONH; CH2CH2OOC), n=0,1; R = C5H11.

Adam Januszko

300 n=1 n=0

250

T M-I [ºC]

200 Ph 150 12V 100 n=1 50

n=0

CONH

CH=CH

CH=N

CH=CHCOO

COO

CH2CH2

Figure 25. Dependence of isotropisation temperature TM-I for 12-vertex and Ph analogues in function of linking groups L (Fig. 24).

isotropisation temperature is significant (Fig. 25). In the phenyl analogue, the lowest isotropisation temperature TM-I = 158°C has been noted for the ethyl linking group CH2−CH2, and the highest TM-I = 267°C for the ethylene group CH=CH. Introduction of an oxygen atom into the terminal pentyl chain increases the clearing temperature of the researched series on the average by 16.5±1.6°C. The difference in isotropisation temperatures for the p-Carborane series in relation to their phenyl analogue is strongly dependant on the structure of the linking group (Fig. 26). The smallest difference ΔTM-I= -50°C has been noted for the dimethyl bond (CH2−CH2) only, whereas the highest ΔTM-I = -208°C for the amide bond (CONH) and n=0. In the studied compounds, mesophase destabilising effect of the substitution of the benzene ring with a p-Carborane cage is on the average higher in compounds with terminal pentyloxyl substituent (n=1) by 6±2°C than in case of compounds with pentyl substituent (n=0). A pair of compounds with the CONH linking group, where the difference is 22°C, is the exception to the rule. Such behaviour could be caused by the difference in intermolecular quadruple interaction of the carborane cage with pentyloxyphenyl and pentylphenyl [51, 52].

p-Carborane liquid crystals: Nematogenic properties and potential application

Δ T M-I [ºC]

-50

-100

-150 n=1 -200 n=0

CONH

CH=CH

CH=N

CH=CHCOO

COO

CH2CH2

-250 7

Figure 26. Difference in isotropisation temperatures for 12-vertex analogues, as compared with the Ph for different linking groups L. (ΔTM-I = TM-I(12V) - TM-I(Ph)).

The efficiency of individual linking groups’ (fig.19) mesophase destabilization in Ph analogues: CH=CH > CONH > CH=N ∼ CH=CHCOO > COO > CH2CH2 > CH2CH2OOC is analogical to the trend observed in simple two-ring phenyl derivatives, such as R1−Ph−L−Ph−R2 [36÷38]. The sequence reflecting the efficiency of mesophase destabilization in individual linking groups in 12-vertex analogues is slightly different: CH=CHCOO ∼ COO ∼ CH=CH > CH=N ∼ CH2CH2 > CONH > CH2CH2OOC. This order illustrates the rigidity of the linking group and its electronic interaction with the ring/cage. The importance of linking group rigidity and the expansion of its electronic impact is visible when CH=CHCOO group is compared with the CH2−CH2OOC group. The reduction of alkyl group double bond increases molecular elasticity which results in depression of the clearing temperature by 133°C in the Ph analogue, and by 26°C in the 12-vertex analogue. The increase of molecular elasticity of compounds containing a CH2−CH2OOC linking group is proved by the observed remarkably high enthalpy of N-I phase transition of both Ph analogues (n=0.1) 5.3 and 6.0 kJ/mol, respectively. These enthalpies, four times as high as average enthalpies of

Adam Januszko

isotropic phase transitions, correspond to significant changes in entropy and can be identified (during the N-I phase transition) with vital conformational changes resulting from excessive molecular elasticity. What should be noted here is that 12-vertex analogues with CH2−CH2OOC linking group do not exhibit any mesogenic properties whatsoever. An amide group (−CONH−) has been fund to have particularly interesting properties among the testes linking groups. Its geometry is similar to the one of the ester group (COO) but due to the presence of a hydrogen atom in the compound, it can also form a strong hydrogen bond. The literature on this topic states that the amide group in hydrocarbon liquidcrystalline compounds increases both the melting point, and isotropisation temperature and, in case of the ester analogues, stabilizes smectic phases [1, 53÷58]. Additional diffraction tests of amide mesogens revealed the occurrence of intermolecular hydrogen bonds in the smectic phase which stabilize this sort of molecular ordering [59]. Similar behaviour has been observed in the tested groups of compounds as well (Fig. 25). The substitution of carboxylic group (COO) with amide group (CONH) in both the benzene analogues increases the melting point by 100°C and the isotropisation temperature by 57°C entirely eliminating the nematic phase in favour of the smectic SmA phase. The same procedure of exchanging the linking group COON with the CONH one in both the p-Carborane analogues yields a contrary result. Whereas the melting point increases by less than 30°C, the isotropisation temperature drops drastically by 46°C (n=1) and by 65°C (n=0) entirely eliminating the presence of the smectic phase. Such behaviour may be caused by the impossibility of forming intermolecular hydrogen bond which in turn results from the steric and electronic effects between the p-Carborane and the carbonyl group. The volume of the carborane cage limits the ability of the two molecules to come close to each other thus weakening the nucleophilic properties of the carbonyl group (hydrogen-bond acceptor ability) [60]. The test results indicate an inconsiderable impact of the type of linking group L on mesogenic properties of p-Carborane compounds. Due to limiting effect exerted on the hydrogen bond generation by the p-Carborane cage, the cage has a marked impact of the efficiency of amide CNOH group mesophase stabilization. The above-analysed changes of properties of a liquid-crystalline chemical compound realized via the “shift” of carborane cage from one area in its molecule’s structure to another often requires conducting a complex synthesis. The next chapter is dedicated to the synthesis of the discussed carborane mesogens.

p-Carborane liquid crystals: Nematogenic properties and potential application

3. Synthesis of carborane mesogens All the mesogenic carborane derivatives are a new group of liquidcrystalline compounds. They have been obtained from initial substrates, such as 1,10-dicarbo-closo-decaborane (10-vertex) or dicarbo-closo-dodecaborane (12-vertex) which have been used most often. Some carbocyclic mesogenic analogues have been synthesized already and the results have been issued in a number of publications. Due to the differences in the liquid-crystalline synthesis methods, as compared to the ones applied in case of their hydrocarbon analogues, synthesis diagrams for all the carborane derivatives obtained and described here are provided below.

H + I

1. BuLi

DME, Pyrimidine

2. CuCl

H 1. BuLi

1. BuLi

2. ClCOOMe 2. C5H11I

3. KOH/THF

HOOC 1. (COCl)2, PCL5 or SOCl2

C5H11

BBr3; R = Me

2. R1−X−OH; X = Ph, Bco, CHx

R1−X

C5H11 OR

OOC

ClOC − X − (O)n− R1

X = Ph, Bco, CHx

X = Ph, 12V; n = 0,1

OOC −X− (O)n− R1

C5H11 X = Ph, 12V; n = 0,1

Figure 27. Diagram of p-Carborane esters synthesis.

Adam Januszko

3.1. Synthesis of p-Carborane esters The group of carborane esters I, II and III (Fig. 4) has been obtained as the result of a complex synthesis from p-dicarbo-closo-dodecaborane in accordance with the diagram below (Fig. 27) [31].

3.2. Synthesis of mesogenic p-Carborane compounds which differ in the linking group Synthesis of mesogenic p-Carborane compounds which differ in the linking group (Fig. 24) has been conducted by the authors [50] according to the diagrams below (Figs.28-30). H11C5 − O

CH = N

(O)n

H11C5 − O

C5H11

CH=CH

(O)n

C5H11

ArNH2 H2 / Pd ArCHPPh3

H11C5 −O

−CH2−CH2 −

(O)n−C5H11

1. BuLi

H11C5 −O

CHO

2. HCOOEt

H11C5 −O

X = 12V

X = Ph, 12V (EtO)2P(0)CHCOOR

H11C5 − O 1. NaOH

CH=CH

COO R

CH=CH

COOH

CH=CH − COO

R=Et

2. HCl

H11C5 − O

1. (COCl)2 2. ArOH

H11C5 − O

(O)n− C5H11

Figure 28. Synthesis of mesogenic p-Carborane differing in the linking group and obtained from an aldehyde. 1) Initial substrate has been synthesized according to the procedure depicted in Fig. 27.

p-Carborane liquid crystals: Nematogenic properties and potential application H11C5 − O

CH2−CH2−OTHP

CH2−CH2−OH

1. BuLi 2. THPOCH2CH2 Br

H11C5 − O H+/ H2O

H11C5 − O

ClOC

H11C5 − O

(O)n− C5H11

CH2−CH2−OOC

(O)n− C5H11

Figure 29. Synthesis of mesogenic p-Carborane differing in the linking group (12vertex, Ph). 1) Initial substrate has been synthesized according to the procedure depicted in Fig. 27.

Figure 30. Synthesis diagram for mesogenic p-Carborane differing in the linking group and obtained from carboxylic acid. 1) Initial substrate has been synthesized according to the procedure depicted in Fig. 27.

4. The impact of neutron radiation on the properties of carborane-containing nematogens The possibilities of applying carborane-containing, nematogenic liquid crystals neutron radiation dosimetry have been studied using one of the

Adam Januszko

compounds from ester group: C5-V12-COO-Ph-O-C6. Natural source of neutral radiation, 252Cf, has been used in the experiment. The source emitted neutron flux on the order of 10-4 n/cm2s [61]. In order to thermalise the neutrons, the source has been placed in a sphere filled with paraffin which, additionally, was covered with a layer of cadmium (Fig. 31). Carborane liquid-crystalline material has been inserted into a liquidcrystalline cell with thickness equal d = 100μm 50cm away from the source of radiation on a (PMMA) polymethacrylate methyl plate, which was to perform the role of a phantom. Using Monte Carlo method, the number of neutrons captured by the tested liquid-crystalline material has been calculated to be 52% [62]. Figure 31 presents the results of isotropisation temperature changes for the tested carborane-containing nematogen, whose initial isotropisation temperature equalled TN-I = 36.5ºC. What is visible here is the change in isotropisation temperature by ΔTN-I = 2.51ºC following 36 hours of neutron flux expose [63]. The same material was used as the R reference sample but it has not been previously subjected to neutron radiation. The observed result indicates destructive character of neutron radiation on carborane-containing liquid-crystalline material. The course of the reaction proceeding according to formula (1), lead to the destruction of calamitic molecules, which resulted in the drop of isotropisation temperature TN-I. The application of neutron radiation phenomenon on carboranecontaining liquid-crystalline materials still requires further studies [64].

Figure 31. Schematic diagram of the experiment consisting in the exposure of liquid crystal (LC) to thermal neutron flux.

p-Carborane liquid crystals: Nematogenic properties and potential application

Figure 32. Change of isotropisation temperature for sample (S) containing C5-V12COO-Ph-O-C6 in function of thermal neutron radiation exposure time. The neutrons have been generated by 252Cf source. Reference sample R has not been previously subjected to radiation.

Summary The charter presented structural analysis of compounds and this analysis the search for new liquid-crystalline materials. The properties of calamitic liquid-crystalline compounds have been studied considering the type of the carborane cages and carbocyclic rings and the role they perform depending on their location in a molecule. The results of mesogenic analyses of the three- and four-ring carborane esters in relation to the cyclic hydrocarbon analogues have been presented. The comparison of mesophase stability of carborane compounds and their carbocyclic analogues has been researched, taking into account the structural changes which included: -

fluorisation and extension of the alkyl chains; change of the linking group (bond) or its elimination;

Adam Januszko

introduction of molecular symmetry; substitution of terminal alkyl chain with an alkoxy chain.

The impact of the type of central substituent (carborane cage in relation to the carbocyclic analogues) on the generation and stability of mesophase has been investigated. The following conclusion can be drawn from the findings: −

− −

− − −

definitely most calamitic liquid-crystalline compounds containing in their structure a 10-vertex or a 12-vertex carborane cages generate nematic phase (only one compound with a 10-vertex cage generates enantiotropic smectic SmA phase; one generates monotropic SmC phases and a G phase; two compounds with a 12-vertex cage generate enantiotropic smectic phase and two compounds – monotropic smectic SmC and SmA phases); due to particular properties of carborane cages, their position in terminal location exerts a positive impact on nematic phase stability, as compared to the one they have when located in rigid molecular core; unlike what could be observed in the carbocyclic analogues, prefluorisation of a short (n ≤ 5) alkyl chain in carborane-containing analogues does not lead to generation of smectic phase, but results in significant destabilization of the nematic phase; rigidity of the linking group and its electronic interaction with carborane cage plays a vital role in stabilization of mesophase in carboranecontaining nematogens; mesophase stability is largely governed by the type of ring/cage (located in the molecular core) and to a lesser degree by the properties of terminal group; p-Carborane liquid crystals, due of their specific properties, are perspective materials for thermal neuron detection.

Acknowledgements I would like to thank Prof. Piotr Kaszynski, head of Organic Material Research Group (Chemistry Department, Vanderbilt University, Nashville, TN) for his support and collaboration. This research is supported by NSF.

References 1. 2.

Vill V., LiqCryst 4.6 database, Hamburg, LCI Publisher, 2006. Januszko A., Master Degree thesis, WAT, 1990.

p-Carborane liquid crystals: Nematogenic properties and potential application

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.

Adamczyk A., Niezwykły stan materii – ciekłe kryształy, 1979, Wiedza Powszechna, Warszawa. Reinitzer F., Monatshefte für Chemie, 1888, 9, 421. Alfassi Zeev B., Radiation Effects, 1977, 32-1, 2. Alfassi Zeev B., Kłosowicz S.J., Mol. Cryst. Liq. Cryst., 1994, 239, 181. Lavrentovich O., Ukrainski Fizycheskii Zhurnal, 1988, 33-6, 859-9. Kłosowicz S.J., 1991, WAT, wewn. 2058/92. Dhar R., Verma R., Rath C.M., Sarkar S.K., Wadhawan V.K., Dabrowski R., Tykarska M., Appl. Phys. Lett., 2008, 92, 014108. Dhar R., CLC 2009 Augustów. Prawo atomowe, Dziennik Ustaw 2001, Nr 3 poz. 18, Ustawa z dnia 29 listopada 2008 r. US Patent 4,605,859. Asbury A.K., Ojean R.G., Nielsen S.L., Sweet W.H., J. Neuropathol. Exp. Neurol., 1972, 31. Brownell G., Zamenhof, R.G., Murray B.W., Wellum G.R., Therapy in Nuclear Medicine, R.P. Spencer (ed.), Grune and Stratton, Inc., New York, 1978. Coderre J.A., Strahlenther. Onkol., 1989, 165, 215-217. Barth R.F., Solloway A.H., Fairchild R.G., Cancer Res., 1990, 50, 1061-1070. Kaszynski P., Huang J., Jenkins G.S., Bairamov K.A., Lipiak D., Mol. Cyst. Liq. Cyst., 1995, 260, 315-332. Douglass A.G., Mierzwa M.L., Kaszynski P., SPIE, 1998, 3319, 59-62. Douglass A.G., Czuprynski K., Mierzwa M.L., Kaszynski P., Chem. Mater., 1998, 10, 2399-2402. Douglass A.G., Czuprynski K., Mierzwa M.L., Kaszynski P., J. Mater. Chem., 1998, 8, 2391-2398. Kaszynski P., Douglass A.G.; J. Organomet. Chem., 1999, 581, 28. Kaszynski P., „closo-Boranes as Structural Elements for Advanced Anisotropic Material” [w:] Anisotropic Organic Materials – Approaches to Polar Order, R. Glaser and P. Kaszynski Eds., ACS Symposium Series, vol. 798, American Chemical Society, Washington, D.C., 2001, 68-82. Gray G.B., 1982, Polymer Liquid Crystals, edited by Ciferri A., Kingbaum W.E. and Meyer R.B., (Academic Press). Klingbil R.T., Genowa D.J., Criswell T.R., Van Meter J.P.; JACS, 1974, 96, 7651. Demus D., Demus H., Zaschke H.; Fluessige Kristalle in Tabellen, VEB Deutsche Verlag fur Grundstoff Industrie, Lipzig, 1974. Gray G.W., Goodby J.W.G., Smectic Liquid Crystals – Textures and Structures, 1984. Griffin A.C., Fisher R.F., Havens S.J., JACS, 1978, 100, 6329. Gray G.W., Hartly J.B., Ibbotson A., Jones B.; J. Chem. Soc., 1955, 4359-4368 Kaszynski P., Pakhomov S., Tesh K.F., Young V.G. Jr.; Inorg. Chem., 2001, 40, 6622-6631. Piecek W., Kaufman J.M., Kaszynski P.; Liq. Cryst., 2003, 30, 39.

Adam Januszko

31. Ohta K., Januszko A., Kaszyński P., Nagamine T., Sasnouski G., Endo Y.; Liq. Cyst., 2004, 31/5, 671-682. 32. Januszko A., Kaszynski P., Wand M.D., More K.M., Pakhomov S., O’Neill M., J. Mater. Chem., 2004, 14, 1544-1553. 33. Ringstrand B., Vroman J., Jensen D., Januszko A., Kaszynski P., Dziaduszek J., Drzewinski W., Liq. Cyst., 2005, 32/8, 1061-1070. 34. Demus D., Hauser A., 1990, Selected Topics in Liquid Crystals Research, edited by H.D. Koswing, (Berlin: Academie − Verlag), 19-44. 35. Czuprynski K., Kaszynski P., Liq. Cyst., 1999, 26, 775. 36. Toyne K.J., Thermotropic Liquid Crystals, 1987, (Ed.) Gray, G. W., Wiley, New York, 28-63. 37. Januszko A., Glab K.L., Kaszynski P., Liq. Cyst., 2008, 35/5, 549-553. 38. Januszko A., Kaszynski P., Liq. Cyst., 2008, 35/6, 705-710. 39. Demus D., Goodby J.W., Gray G.W., Spiess H.-W., Vill V. (Eds.), Handbook of Liquid Crystals; 1998, vol. 2A., 411-440, Wiley-VSH, New York, 40. Neubert M.E., Liquid Crystals, Kumar S. (Ed.), 2001, Cambridge University Press; Cambridge, 393-476. 41. Janulis E.P., Novack J.C., Papapolymerou G.A., Tristani-Kendra M., Huffman W.A., Ferroelectrics, 1998, 85, 763-772. 42. Guittard F., de Givenchy E.T., Garibaldi S., Cambon A., J. Fluorine Chem., 1999, 100, 85-96. 43. Pradhan N.K., Paul, R., Mol. Cryst. Liq. Cryst., 2001, 366, 157-164. 44. Czuprynski K., Douglass A.G., Kaszynski P., Drzewinski W., Liq. Cryst., 1999, 26, 261-269. 45. Singh S., Liquid Crystals: Fundamentals, Word Scientific: New Jersey, 2002, 101-105. 46. Knaak L.E., Rosenberg H.M., Servé M.P., Mol. Cryst. Liq. Cryst., 1972, 17, 171185. 47. Demus D.Z., Chem., 1975, 15, 1-13. 48. Praefcke K., Schmidt D., Heppke G., Chem. Ztg., 1980, 104, 269-271. 49. Bregadze V.I., Chem. Rev., 1992, 92, 209-223. 50. Nagamine T., Januszko A., Ohta K., Kaszynski P., Endo Y., Liq. Cyst., 2008, 35/7, 865-884. 51. Januszko A., Glab K., Kaszynski P., Patel K., Lewis R.S., Mehl G.H., Wand M.D., J. Mater. Chem., 2006, 16, 3183-3192. 52. Nagamine T., Januszko A., Kaszynski P., Ohta K., Endo Y., J. Mater. Chem., 2006, 16, 3836-3843. 53. Arora S.L., Fergason J.L., Taylor T.R., J. Org. Chem., 1970, 35, 4055-4058. 54. Schroeder D.C., Schroeder J.P., J. Org. Chem., 1972, 37, 1425-1428. 55. Kalyvas V., McIntyre J.E., Mol. Cyst. Liq. Cyst., 1982, 80, 105-118. 56. Vora R.A., Gupta R., Mol. Cyst. Liq. Cyst., 1981, 67, 215-220. 57. Mori A, Uno K., Kato N., Takeshita H., Hirayama K., Ujmie S., Liq. Cyst., 2004, 31, 285-294. 58. Mori A, Hirayama K., Kato N., Takeshita H., Ujmie S., Chem. Lett., 1997, 509510.

p-Carborane liquid crystals: Nematogenic properties and potential application

59. Kajitani T., Kohmoto S., Yamamoto M., Kishikawa K., J. Mater. Chem., 2004, 14, 3449-3456. 60. Zakharkin L.I., Kalinin V.N., Res E.G., Bull. Acad. Sci. USSR, Div. Chem. Sci., 1974, 2543-2545. 61. Bogard J.S., DOSAR Calibration Laboratory, ORNL, Oak Ridge, Tennessee, USA. 62. Veimont K., Obliczenia MCNP5, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA. 63. Kaszynski P., Januszko A., Detection of Neutron Radiation Using Liquid Crystalline Materials, Office of Technology Transfer and Enterprise Development, 2003, Vanderbilt University, TN, USA. 64. Januszko A., Piliszek P., Jaroszewicz L., Kaszynski P., 2009, CLC August贸w (POLAND).

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 95-124 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

4. Liquid crystalline dendrimer: Towards intelligent functional materials Amrit Puzari Girijananda Chowdhuri Institute of Management & Technology, Hatkhowapara Azara, Guwahati-17, Assam, India

Abstract. The structural organization of various types of dendritic systems that form liquid crystalline mesophases have been analysed with special emphasis on their relevance towards developing intelligent functional materials. Liquid crystalline dendrimers has been gaining considerable interest in recent years for development of newer functional materials because of their mesomorphic properties of the mesogenic subunits associated with the supermolecular versatile architecture. Combination of these features enhances their prospect as potential candidate for developing functional materials. Structural control achieved with such type of molecular architecture by tuning the terminal mesogenic groups, dendritic core and dendrimer generation, for appropriate molecular design are also reviewed. The influence of the dendritic matrix, particularly the factors like multivalency of the dendritic core, multiplicity of the branches etc. on their supramolecular organization are analysed. Some basic concepts about thermotropic liquid crystals as well as dendrimer along with various phases of liquid crystalline dendrimers are provided in the

11 Correspondence/Reprint request: Dr. Amrit Puzari, Girijananda Chowdhuri Institute of Management & Technology, Hatkhowapara, Azara, Guwahati-17, Assam, India. E-mail: amrit09us@yahoo.com or amrit09us@gmail.com

Amrit Puzari

introductory chapter. Description on selected examples of liquid crystalline dendrimer such as side-chain, main-chain, fullerodendrimers, shape-persistent dendrimers, supramolecular dendromesogens and metallodendrimers etc. as representative families of liquid crystalline dendrimers along with their utility towards designing of intelligent functional materials are made in the proceeding sections. In conclusion an overall rational analysis has been made on the prospect of further development of liquid crystalline dendrimer.

Introduction Liquid crystalline molecules are â&#x20AC;&#x2DC;soft materialsâ&#x20AC;&#x2122;, representing a special states of matter (also called the fourth state of matter) which is recognized by itâ&#x20AC;&#x2122;s long range order (as in crystals) and mobility (as in liquid) [1-3]. These mesophases can occur in pure materials with dependence on temperature when it is referred to as thermotropic liquid crystals or can occur in multicomponent system with dependence on composition when it is referred to as lyotropic liquid crystals. This unusual feature of such molecules attribute anisotropic physical properties to liquid crystalline compounds and therefore have found enormous application in various fields, more specifically in electronics, display devise, mobile telecommunication and computing devices. Thermotropic liquid crystals are used in electrooptical displays, temperature sensors etc [4]. While lyotropic systems are mostly used in cosmetic industries [5] besides their use as templates for preparation of mesoporous materials. Thermotropic liquid crystals are of three types namely nematic, cholesteric and smectic and all these are distinguished by the different kinds of molecular order they exhibit. For example nematic structure maintains a parallel or nearly parallel arrangement to each other along the long molecular axes. They can rotate about one axis and the structure is one dimensional. Cholesteric is in fact a chiral nematic liquid crystal where direction of the long axis of the molecule in a given layer is slightly displaced from the direction of the molecular axes of the molecules in an adjuscent layer. All non nematic liquid crystals are included in smectic phase. In smectic phase liquid crystals retain some amount of orientational order that is present in nematic along with small amount of positional order. In smectic structures, the molecules are free to bounce around randomly and tend to point along a specific direction and arrange themselves in layers, either in neat rows or randomly distributed. Mesomorphism is exhibited by materials having liquid crystalline properties (which are referred to as mesogens) and shape of the mesogen, that form the mesophases like enantiotropic or monotropic mesophase may guide to the structure of liquid crystalline compound.

Liquid crystalline dendrimer: Towards intelligent functional materials

Also liquid crystalline materials can self organize into various nanostructured phases, such as layered, columnar, micellar, and bicontinuous cubic which are formed by segregation of immiscible parts in molecules on nanometer scale. Columnar phases are obtained by packing of disc like molecules into columns and are characterized by the symmetry of the arrangement, for example hexagonal (Colh), rectangular (Colr), square (Cols) or oblique (Colo). This phase is obtained with many non-discotic molecules such as polycatenar, polyphilic, and bent-shape mesogens. Diffusion between and within the column readily occurs and the phase are fluid. Chiral modifications for calamitic mesogens (rod like species) are also known which are referred to as chiral mesophase (for example SmC*, SmI* etc.). These are exhibited by either a pure enantiomer or by a non chiral compound doped with little amount of chiral additives [6]. Additionally, less common liquid crystalline phases such as miceller cubic phases (CubI), bicontinuous cubic phases (CubV) are known to exhibit by some classical rod like mesogens. Laminated type phases (Lam) are also recognized with some smectic phases which are consists of sheets in which aromatic cores lie parallel to the layered planes, separated from each other by lipophilic sublayers [7]. Self organized structure of lyotropic liquid crystals are even compatible with those of living systems. Liquid crystals have numerous functional utility such as information and mass transport, sensing, catalysis, templates, electro optic displays and so on. Therefore these materials have outstanding potential for advance application and therefore towards development of intelligent materials. Dendrimers are highly branched three dimensional macromolecules with a branch point at each monomer unit and have attracted interest of both industrial and academic chemists. The most important feature of dendritic molecules is their well defined shape, symmetric nature and molecular architecture, which is not observed in case of hyper branched polymers [8]. Also unlike hyper branched polymers, dendrimers are obtained by careful stepwise growth of successive layers of generations. It is possible to functionalise dendrimers at the periphery of the globular structure instead of insulating a reactive site at the centre of a dendrimer. Besides amphiphilic dendrimers containing an extended rigid block represents a class of self assembling systems those are increasingly used for the construction of supramolecular architecture with well defined shape. Nowadays a broad range of dendrimers are available and some of them are even available commercially, and have found to be promising towards important chemical processes and also as drug or gene delivery devices, as carriers for catalytically active site in flow reactors and also as chiral auxiliaries for asymmetric synthesis. The field of research on the look for newer dendritic

Amrit Puzari

species has also been boosted up by the enhanced possibility of their potential in wide spread application. Their resemblance and similar dimensions to some living components and to molecular functional materials projects them as potential candidate for even biological science too, provided these are suitably functionalised. High molecular weight monodisperse dendrimers are promising candidates for the development of new liquid-crystalline materials for specialized application. The pervasive nature of liquid crystalline science along with the mesomorphic behaviour of dendritic architecture projects liquid crystalline dendrimers as interesting materials for basic studies and development of newer materials for organic molecular electronics and other allied fields of science. The notable features of liquid crystalline dendrimers is that the liquid crystalline properties of the compounds can be controlled as a function of dendrimer generation and can be tuned by careful selection of the mesogenic moiety. Therefore molecular engineering of liquid crystalline dendrimers through understanding of the structure-organization relationship is highly significant for design of mesomorphic with tailor made properties for application in various fields. Thus LC dendrimers projects them as most versatile dendritic architecture where mesomorphism can be modulated by very subtle modifications of the dendritic connectivities. These branch of science can be regarded as a multidisciplinary field of research with broader prospect for design of new novel molecular architecture.

1. Structural organization of liquid crystalline dendrimer The evolutionary concept of dendritic architecture and more specifically dendritic nanostructure has already been able to introduce a new dimension in the hierarchical complexity of matter. The three dimensional branched macromolecular architecture is one of the most pervasive topologies observed in nature at the macro- and microdimensional length scales and consisting of three topologically distinct regions: multivalent surface, branched interior and encapsulated core (figure 1). This has also been recognized as fourth major class of macromolecular architecture with distinct characteristic properties [9]. Numerous synthetic strategies developed so far has led to four architectural subclasses for dendritic polymers, namely random hyperbranched polymers, dendrigraft polymers, dendrons and dendrimers. Therefore dendrimers and dendrons are potentially useful versatile scaffolds for development of new liquid crystalline materials as the interesting structural framework which will allow modulation of the mesomorphism in the structural organization through appropriate modification of the dendritic connectivity. They thus constitute

Liquid crystalline dendrimer: Towards intelligent functional materials

Encapsulated core

Branched interior

Multivalent surface

Figure 1. Topological regions of a dendrimer.

an important class of mesogens where new types of mesophases and original morphologies may be discovered [7]. Overall structure of a liquid crystalline dendrimer influenced by the mesogenic or pro-mesogenic promoters attached to the peripheral groups of the flexible dendritic network, tendency of the dendritic core to adopt a globular isotropic conformation and also by the chemical and structural compatibilities between the dendritic core and the peripheral groups. Depending on the number of generation of the LC dendrimer, their structural organization may have different characteristics. For example higher generation dendrimers have strong tendency to self organize into supramolecular columnar mesostructure [10]. The strong tendency of the core to phase separate with increasing number of generation also led to formation of strongly segregated smectic structure for some LC dendrimers. These supramolecular organizations of LC dendrimers also changes with change in temperature.

1.1. Influence of dendritic matrix The matrix of the liquid crystalline dendrimer is highly influential in determining the material properties of the complex structure. The complex

100

Amrit Puzari

structure generated can be one dimensional (lamellar phases), two dimensional (columnar phases) or three dimensional (micellar or bicontinuous phases). It has been observed from X-ray crystallography and molecular modeling that 1st and 2nd generation dendrimers normally possess planar structure while structural complicacy arises for dendrimers with higher generations due to overcrowding of the terminal groups. With increasing generations the shape of the dendrimer become increasingly globular in order to occupy larger molecular structure with minimized repulsion between the segments [11]. In case of globular macromolecular structures, the core region of the dendritic matrix becomes increasingly shielded off from the surroundings by the dendritic wedges from large number of compartments created in the interior. For liquid crystalline dendrimers with increasing generation number the isotropization temperature increases and phase transition behaviour also changes. The strongest influence of spherical molecular architecture on the phase behaviour of the LC dendrimers appears at higher generations which regulates the structural organization of LC dendrimers. Depending on the type and nature of intra molecular forces the resultant dendritic structure can be rigid or flexible. In case of flexible structures backfolding may occur as a consequence of weak forces between surface functionalities or dendrons and relatively higher mobilities leading to more disordered conformation where the molecular density is spread out over the entire molecular area. However back-folding may also resulted from attractive forces such as ion-pairing, hydrogen bonding etc. between functional groups at the inner part of dendrons and the surface functional groups. Thus the structural organization of LC dendrimers upon growing to higher generations depends upon its ability of the surface groups to form network with each other. In the case where hydrogen bonding is possible with the surface groups, a dendrimeric motif with a very dense periphery and hollow core may be resulted. Self assembly of such supramolecular dendritic architecture into bulk phase to produce periodic nanostructures renders possible the design of newer intelligent organic matter having multidimensional utility. This fact encourages further research activities on liquid crystalline bulk nanoscale organic matters.

1.2. Influence of multivalency of dendritic core on supramolecular organization The core structure of dendrimer which is some times denoted as generation â&#x20AC;&#x153;zeroâ&#x20AC;? thus presents no focal point. The multiplicity of the

Liquid crystalline dendrimer: Towards intelligent functional materials

101

branched cell and directional characteristics of the core influences certain features of the dendritic architecture. Some details about core cell architectural components are available in literature [12]. The concentric monomer shells (generations) surrounding the core have well defined monomer shell saturation levels analogous to electron shells in atoms. Structural organization and building events in case of liquid crystalline dendrimer too is dependent on the architectural behaviour of the nucleus (core) including directional characteristics. Both the core multiplicity and branch multiplicity can be considered as a measure of precise number of terminal groups and mass amplification as a function of dendrimer generation. The core thus can be considered as the molecular information center containing the information about size, shape, multiplicity and directionality expressed as a function of covalent connectivity with the outer shells.

1.3. Influence of multiplicity of the branches Each architectural component of a dendritic architecture manifests a special function characteristic of its own and at the same time also defines the properties of the whole structure. Interior of dendritic architecture can be considered as the branched cell amplification region and it also provides an idea of the type and volume of interior void space that may be enclosed by the terminal groups as the dendrimer is grown. Therefore the multiplicity of the branches of the interior region also influences the structural organization of dendritic liquid crystals since the multiplicity and nature of the interior region will influence the process of molecular self assembly and formation of quantized bundles of nano scale building units. Bulk nanoscale self-assembly of organic matter is promising for using as scaffolds for photonic materials and other intelligent components of molecular electronics. Dendrons and dendrimers are particularly versatile in generating such periodic nanostructures. Elemental compositions of the core as well as hydrophilic and hydrophobic nature of the core group are also decisive factors in the process of structural organization of liquid crystalline dendrimers.

1.4. Influence of terminal functional group Surface groups in dendritic architecture manifests several characteristic functions. Appropriate functionalization of the surface groups might enable one to design newer intelligent functional material with novel utility. Further to this the surface functional groups also serve as a template polymerization

102

Amrit Puzari

region as each generation is amplified and covalently attached to the precursor generation. For dendritic architecture, the terminal functional groups witness an exponential growth as the dendrimer generation increases which may lead to â&#x20AC;&#x153;tethered congestionâ&#x20AC;? due to over crowding of the end groups. The consequence of such congestion is that while lower generation of the dendrimers are obtained as open floppy structures, the higher generations become increasingly robust. Molecular design of liquid crystalline dendrimer thus influenced by the chemical nature and structure of both the functional groups and the dendritic matrix. Intrinsic connectivity of the dendrimer such as the multivalency of the focal core, multiplicity of the branches has significant influence on the geometric growth of the dendrimer generation, supramolecular organization, stability and mesomorphic structure. Tuning of the mesomorphic structure can be achieved through appropriate design of molecular architecture with concomitant change of the terminal mesogenic groups and similar change in dendritic core and interior will produce analogous effect. Liquid crystalline dendrimers are attractive candidates for design of newer liquid crystalline materials containing active molecular units for specific physical properties. Mesogenic group itself shows mainly a nematic phase which disappears quickly on getting attached to a dendritic scaffold. Also owing to the fact that the branched and the aromatic parts are chemically linked, the smectic phase represents the most stable thermodynamic state of organization in a lamellar morphology.

2. Selected examples of liquid crystalline dendrimers 2.1. Side-chain liquid crystalline dendrimer The chain conformation of liquid crystalline dendrimers is an important factor controlling their phase behaviour as well as material properties. Sidechain liquid crystalline dendrimer is the term applied to those flexible dendritic network containing mesogenic or pro-mesogenic groups attached laterally (side-on) or terminally (end-on) to the termini of the branches. Both liquid crystalline dendrimers with terminal mesogenic groups and side chain liquid crystalline polymers consists of molecules possessing the following structural units: polymeric chain, spacer and terminal (or side chain) mesogenic groups. Of course liquid crystalline polymers possess linear polymer chain while liquid crystalline dendrimers possess dendritic architecture. A schematic representation of such side-chain liquid crystalline dendrimer is given in figure 2. The tendency of the dendritic core to adopt globular isotropic conformation and probable microphase separation due to

Liquid crystalline dendrimer: Towards intelligent functional materials

103

Multivalent core Connectivity, Nc=4 Branching point, Multiplicity,Nb=3 Mesogenic groups

(a) (b)

Figure 2. Schematic 2D representation of a first generation side-chain dendrimer. The mesogens is terminally (a) or laterally (b) connected.

chemical and structural incompatibilities between the flexible dendritic core and the terminal groups collectively contribute towards stabilization of the mesogens and determines the mesomorphic properties of the entire molecule. The type of mesophase of such compounds depends on the chemical nature of the mesogenic groups. It is important to know whether a dendrimer skeleton containing mesogenic group is able to impede liquid crystalline properties or whether the high concentration of mesogenic units in a dendritic molecule facilitate some pre-organization leading to the formation of mesophases. 2.1.1. Poly(propyleneimine) (PPI) LC dendrimers Dendritic motif of poly(propyleneimine)(PPI) dendrimers contain a tetravalent core and a binary branching point and between two junctions there are three CH2 groups. For this series of LC dendrimers (figure 3) Meijer et al. showed that for the two series of dendrimers functionalized with pentyloxy and decyloxy cyanobiphenyl mesogens, all the higher generation dendrimers with pentyl spacer exhibit liquid crystalline phases between glass transition temperature and the isotropic liquid, while those with decyloxy spacer also exhibit liquid crystalline behaviour but with the presence of a crystalline or semicrystalline state at low temperature and with higher transition temperatures [13]. All these dendrimers showed a SmA phase. PPI dendrimer with the nonmesogenic 3,4-bis(decyloxy) benzoate end group, induction of a Colh phase was systematically observed from G0 to G4 [14]. The orientational order of

104

Amrit Puzari RN H

R HN

NH H RN

R NH

R R HN

H N

R NH NHR R NH R NH N

N N

HN R

N N

HN RHN R

N N

N N NH R

HN R

N N

HN R HN R

R NH R NH

N H N

R R HN

R NH

NH R NR H N NH R R N N NHH R NR H NH R H NR

O O R=

Figure 3. Poly(propyleneimine) dendrimers with terminal mesogenic groups.

the peripheral mesogenic group also determines the possible interaction between terminal groups and thus influencing the layer spacing of the LC dendrimer. Layer spacing does not vary as a function of the number of peripheral mesogenic units. 2.1.2. Polyamidoamine (PAMAM) LC dendrimers For this series of LC dendrimers also the dendritic motif is same as that of PPI dendrimers but between two junctions there are â&#x20AC;&#x201C;(CH2)2-CONH-(CH2)2segment in place of three CH2 groups and between two central N, there are two CH2 groups. Microphase separation between the mesogenic rigid unit and flexible dendritic skeleton leads to formation of smectic phases. True smectic phases are formed by calamitic mesogens through superposition of equidistant molecular layers and are characterized along orientational correlation of the principal axis. On the other hand columnar phases are

Liquid crystalline dendrimer: Towards intelligent functional materials

105

obtained by stacking of disc-like molecules into columns, which are packed parallel into 2D ordered lattice and are characterized by symmetry of the arrangement, for example for hexagonal case it is Colh [15]. It is to be noted that the arrangement of the mesogenic end groups is predominantly perpendicular with respect to the layer planes, with the dendritic core located between these mesogenic sublayers. For PAMAM LC dendrimers (figure 4), the layer periodicity is almost independent of generation number and probably the molecular volume expands within the plane of the layer as we move from one generation to the next. Grafting of additional terminal chains at the periphery of the anisotropic matrix influences the phase behaviour of the liquid crystalline dendrimer. Dendrimers functionalized by anisotropic units bearing one terminal alkoxy chain led to formation of smectic phase whilst the same matrices functionalized with mesognic units bearing two or three terminal alkoxy chains exhibit Colh phase (figure 4) [16]. This happens due to the fact that the terminal chains prevents parallel disposition of the pro-mesogenic group and are forced to arrange radially about the central moiety. The process of molecular self-assembly induces supramolecular columns which further self organize into columnar hexagonal phases. The dendritic core in such conformation forced to deform anisotropically in one main direction, corresponding to the molecular axis. The presence of amide linkers in PAMAM based LC dendrimers enhances intra- and intermolecular hydrogen-bonds interactions and thereby accounting for greater thermodynamic stability. 2.1.3. Carbosilane liquid crystalline dendrimers Silicon-containing dendrimers, depending on the nature of the linkage at each junction, can be derivatized into carbosilane (Si-C), siloxane (Si-O), and carbosilazane (Si-N) sub-systems. Among these carbosilane dendrimers constitute one of the most important class of dendrimers due to their excellent chemical and thermal stability and versatility of the core connectivity which gives rise to a rich Si-C chemistry. For carbosilane LC dendrimers [17] the dendritic motif is based on a tetravalent core (Nc= 4) coupled to a binary branching multiplicity (Nb=2). Nature of the mesogen and generation number of the dendrimer are necessarily been involved in determining the phase of the LC dendrimer. Thus dendrimers up to 4th generation functionalized by calamitic units (for example cyanobiphenyl, methoxyphenyl benzoate or anisic acid drived mesogens) exhibit solely SmecticA and SmecticC phases between room temperature and 90oC and the smectic phases stability and order improve upon dendrimerization due because of enhanced microphase

106

Amrit Puzari R

R NH

O O

O N

R HN

R NH

N O

NH O

R HN

N H N

N O

R HN

H NR

O N

RN H

O O N

N H

N O

N H

N O

NH R

O HN

O NH

O N

NH R

O N

HN R

H N

R' O R= L1: R'=R''=H R=L2: R'=H, R''=OC10H21 R=L3: R'=R''=OC10H21

OC10H21 HO

(L)

R''

Figure 4. Representative structure of PAMAM LC dendrimer and peripheral mesogens, R

separation between the mesogenic units and the flexible dendritic Sicontaining skeleton. The dendritic core in such compounds is probably in a distorted conformation with a 2D expansion in a plane parallel to the smectic layers. The supramolecular shape of the dendrimer varies significantly with increase in temperature. Above the SmA phase, a Colr phase is first formed, and on further heating a Colh phase is formed. This happens because as the

107

Liquid crystalline dendrimer: Towards intelligent functional materials

carbosilane dendrimer continues its expansion in the third direction leading to the curvature of the mesogen/dendrimer interface, mesogen must compensate the increase of the cross sectional area of the core. So further increase of temperature results more cylindrical shape leading to columnar structure [15]. Carbosilane LC dendrimers G-n(Und-CB)m (figure 5) of 1, 2, 3, 4, and 5 generations with 8, 16, 32, 64, and 128 terminal cyanobiphenyl mesogenic groups were synthesized via hydrosilylation reaction using Pt-catalyst [18]. q NC q CN

O O O

q NC

O CN q

O Si O Si Si O Si

p H3C Si

Si O

p O

Si CH3

O Si

CH3

Si O

Si H3C

p O Si

O Si Si

q NC

Si O Si

O Si

CN q

O O O

NC q CN q

n=1, 2, 3, 4, 5 p=1, 3, 7, 15, 31 q=1, 2, 4, 8, 16 m=8, 16, 32, 64, 128

Figure 5. Carbosilane LC dendrimer G-n(Und-CB)m; In the formula n is the generation number and m is the number of terminal groups.

108

Amrit Puzari

Mesogenic groups were linked to the carbosilane dendritic matrix through an undecylenic spacer and that facilitate mesophase formation. Consideration of the influences of the factors such as generation number of the dendritic matrix, spacer length between the mesogenic units and the dendritic matrix and chemical nature of the terminal mesogenic group are very important in assigning the phase behaviour of such compounds. Thus in case of carbosilane LC dendrimers, for higher generations the influence of spherical molecular architecture become more significant which might pursue even formation of different supramolecular nanostructures of columnar type in addition to the smectic type arrangement. 2.1.4. Polyester and polyether LC dendrimers LC dendrimers of these types are obtained by functionalizing amphiphilic polyol monodendrons with alkyl chain and then connecting those dendrons to a linear polyethylene oxide chain (figure-6). These dendrimers on increasing the polyethylene oxide portion and temperature, self assemble into various supramolecular architecture such as lamellar, micellar cubic, Colh etc. [19]. Mesomorphism can be induced in such dendrimers through microphase separation between the hydrophobic dendron and hydrophilic linear polymer. Dendritic core is built by various generations of 2,2-bis(hydroxyphenyl) ethane. Ferroelectric behaviour is known to exhibit by such LC dendrimers when appropriate chiral mesogenic groups are being attached to the amphiphilic polyester dendritic core [20]. Ionic conductivity of ion doped samples of such dendrimers in different phases can eventually be correlated with the mesomorphic behaviour. OC22H45 O O

OC22H45 OC22H45 OC22H45

O n

O O O

OC22H45 O

O O

OC22H45 OC22H45 OC22H45

Figure 6. Structure of the polyether dendrimer.

109

Liquid crystalline dendrimer: Towards intelligent functional materials

2.2. Main-chain liquid crystalline dendrimer Unlike side chain liquid crystalline dendrimers, where the junctions are consisting of single atoms, main chain liquid crystalline dendrimers contain anisotropic molecular moieties at every branching points, i.e. within such dendritic architectures at every level of dendritic hierarchy anisotropic groups are present. Consequently these molecular architectures possess less conformational freedom and also presence of these anisotropic segments forces the dendrimer to adopt more extended conformation. Conformational restriction occur due to squeezing of the dendritic part within the peripheral cell. Branches for such architectures do not radiate isotropically as we observe with side-chain LC dendrimers and rather they favour an anisotropic order by a gain in enthalpy. Such system possess larger number of functional units compared to side-chain LC dendrimers and a proportionate amplification of properties also expected. B. Donnio et. al. have synthesized main chain LC dendrimers with homolithic systems containing building blocks of identical dendritic branch [21] and heterolithic systems with different anisotropic cores [22]. They have synthesized the dendrimers by a convergent/divergent approach consisting in synthesizing first the acidic branches and then grafting onto the anisotropic core in presence of diphenyl(2,3-dihydro-2-thioxo-3-benzoxazolyl)phosphonate. (scheme 1) [23]. The perspective structures of few such LC dendrimers are represented in figure 7. Main chain LC dendrimers with eight functional arms are also some times referred to as â&#x20AC;&#x2DC;octopus dendrimerâ&#x20AC;&#x2122;. H2N

H2N C12H25O C12H25O

H2N O

O O

O OH

PhO N O P PhO O S

NH O O NH

HN O O HN

H2N

NH2

B NH2

H2N H2N

C12H25O C12H25O

NH2

C H2N N H2N

NH2

Scheme 1. Synthesis of main chain LC dendrimers A, B, C and D (refer to fig. 7 for structure of A, B, C and D)

110

Amrit Puzari

Depending on the chain substitution pattern of the terminal units, octopus dendrimers with terminal mesogenic groups may exhibit smectic mesomorphism (SmB and SmA phases) due to parallel disposition of the mesogenic groups on both sides of the tetravalent core, or columnar mesophase. Dendrimers with one terminal aliphatic chain per unit adopt a prolate conformation and this unique structural pattern with highly segregated molecular sublevels favours formation of smectic mesophases. On the other hand dendrimers with two or more than two aliphatic chains disfavours parallel disposition of the terminal mesogenic groups. The dendrimer adopts an oblate conformation which leads to induction of a Colh phase which can arguably possess onion morphology. The situation can pictorially be represented as shown in scheme 2. The columnar mesostructure in case of the later are indeed resulted from mismatch between the surface areas of the aromatic cores and the cross section of the aliphatic chains and thereby imparting a wedge-like conformation to the octopus dendrimer with the mesogenic groups radially distributed. Figure 7

C12H25O C12H25O

O O

OC12H25 OC12H25

O NH

O C12H25O C12H25O

O O

(A)

OC12H25

C12H25O C12H25O

OC12H25

O O

O OC12H25

C12H25O C12H25O

O O

HN C12H25O

NH O O

NH O O NH

O N H

O OC12H25 OC12H25

OC12H25

C12H25O O O

C12H25O C12H25O

(B)

OC12H25 OC12H25

111

Liquid crystalline dendrimer: Towards intelligent functional materials OC12H25 OC12H25

C12H25O C12H25O

O O

OC12H25 O

HN O

HN N

C12H25O

OC12H25

C12H25O C12H25O

OC12H25 OC12H25

C12H25O

O O

C12H25O

(C)

OC12H25 OC12H25

C12H 25O

OC12H25 C12H25O C 12H25O

OC12H25

O O

O OC12H25

C12H25O C12H25O

O O

OC12H25

HN HN

NH O

O O

C12H 25O

(D)

C12H25O

OC12H25

Figure 7. Main chain liquid crystalline dendrimers.

With such a conformation the dendrimer can self-assemble into supramolecular discs or columns and even to hexagonal net consistent with the lattice parameters (9-10 nm). These dendrimers having onion morphology for the columns in fact can have a wide range of possible interesting structures. Several of such dendrimers were prepared in practice [24-25] with the multicomponent supermolecules including homolithic systems (i.e. the building blocks constituting the dendritic branch are all identical) [figure 8(A)] and alternate heterolithic systems (i.e. containing different anisotropic units) [figure 8(B-D)].

112

Amrit Puzari

Dendrimerization (Mesogenic moiety)

Adopts prolate conformation

Adopts oblate conformation

Multilayered smectic mesophase

Onion columnar mesophase

Scheme 2. Conformational anisotropy in Octopus dendrimers leading to multilayered smectic mesophase and onion morphology.

(A)

(B)

(C)

(D)

Figure 8. Schematic representation of G-1 octopus dendrimer with a homolithic (A) heterolithic alternated (B), segmented (C) and alternated-segmented (D) structure.

2.3. Liquid crystalline fullerodendrimers The covalent attachment of dendritic addends to the [60] fullerene (C60) was first reported by Frechet et al. [26] and that was proved to be an effective method for the design of C60 dendrimers with liquid crystalline properties. Fullerene has interesting physical properties and therefore fullerene containing liquid crystalline dendrimers can encourage enthusiastic studies in the field of supramolecular chemistry and materials science through design of newer self organized structures containing the fullerene unit. Especially this process of self-assembly using supramolecular interactions, might help generation of nanoscale architecture from functional groups which are otherwise not well adapted. Although the large C60 unit disturbs the mesogenic interactions,

Liquid crystalline dendrimer: Towards intelligent functional materials

113

presence of enough mesogenic subunits in the dendritic addend, can compensate these effect to obtain self-assembled structures with liquid crystalline properties. This necessitates proper understanding of the structureorganization relationship for fullerene based supramolecular assemblies. Control of dendritic generations will help tuning of the properties for such supramolecular architecture. Addition reaction of mesomorphic malonate-based dendrimers with fullerene (C60) produces liquid-crystalline fullerodendrimers where the terminal cyanobiphenyl units act as liquid crystalline promoter. The dendrimers were prepared by applying a convergent synthetic methodology [27]. Except the second generation dendrimer, all the fullerenes showed only smectic A phases. Figure 9 (A) and (B) represents the perspective structures of the first generation and second generation of methanofullerenes obtained from the addition of malonate based dendrimers to C60 while figure 9 (C) constitutes a representative example of a fulleropyrollidines. Fulleropyrrolidines are important class of C60 derivatives which are even advantageous than the methanofullerenes as they can produce a stable reduced species having potential towards development of fullerene based redox molecular switches. This family of LC dendrimers also exhibit smectic A phases and stability of the dendrimers increases with increasing number of generations. Of course fulleropyrrolidines have slightly lower stability than that of the analogous methanofullerodendrimers. Studies carried out on the supramolecular organization of these class of LC dendrimers indicated that in case of lower generation of dendrimers, the supramolecular organization is determined by the steric factors. To account for the steric constraint and efficient space filling we need to take into account the cross sectional areas of the fullerene moiety and terminal mesogenic group. The fullerene moiety has a cross sectional area of about 90-100 Ă&#x2026;2 whereas the mesogenic group has a cross sectional area of about 22-25 Ă&#x2026;2. Apart from these other factors like strong attraction between the fullerene moieties, dipolar interaction between the mesogenic groups, amphilic nature of the macromolecules and natural tendency of the mesogenic groups to form anisotropic organization within micro-domains must have to be emphasized to describe the supramolecular organization. For higher generations the supramolecular organization is being influenced by the mesogenic groups which arrange in a parallel fashion, analogous to classical smectic A phase. This happens due to lateral extension of the branching part of the molecules. The rest of the macromolecules being located between the mesogenic sub layers [28]. Thus functionalization of C60

114

Amrit Puzari O

O O

GnO

O 6

O O

OGn

malonate based dendritic addend, Gn= 'n' th Generation O

O RO

O O

O OR

(A) = First generation (G1) methanofullerenes OR

RO O

O O

O O 6

O O

O OR

(B) = Second generation (G2) methanofullerenes OR OR

O O O

OR O

X N

O O 10

O O O

O O

X= CH3 , H

OR OR

R=*

O 10

Figure 9. Examples of LC fullerodendrons.

115

Liquid crystalline dendrimer: Towards intelligent functional materials

with liquid crystalline addends constitutes an appropriate method for the elaboration of fullerene containing thermotropic liquid crystals. Although the bulky size and shape of the fullerene moiety have a tendency to decrease the liquid crystalline tendency as it can act as bulky spacer between the molecular units, this effect can be counteracted by providing a large number of mesogenic units through an increase in the generation number of the dendrimer. Therefore appropriate selection of the dendritic addends can provide a mean to prepare tailor made fullerodendrons through proper control of the dendrimer generation.

2.4. Shape-persistent liquid crystalline dendrimers Shape persistent liquid crystalline dendrimers are another interesting family of liquid crystalline compounds which should find important applications in the field of material science. The dendritic matrix consisting of a conjugated rigid structure with electron-rich core. This family of dendritic matrix possess interesting photochemical and photophysical properties due to the presence of conjugate structure and electron rich core [15]. The matrices are intrinsically discotic in nature. R

R' X

R' R

R' X

R R

X R R

X R'

R' R

R R'

First generation, G1

R R'

Second generation, G2

Tolanoid: R=CO2(CH2CH2O)3Me, R'=H; Stilbenoid: R=R'= OCnH2n+1, n=6,12

Figure 10. Structure of G-1 and G-2 shape persistent dendrimer.

116

Amrit Puzari

Tolane-based dendrimers up to third generation, with oligo(ethylene oxide) chain as the surface groups were reported to exhibit wide temperature range Colh phases while fourth generation were found to be amorphous [29]. Stilbenoid dendrimers up to second generation showed Colh and Colo mesophases [30]. These dendrimers up to second generation, consisting of trans stilbene building blocks and synthesized through a convergent synthetic pathway [30], are shown in figure 10. Lower generation of these dendrimers (first and second generation) possess a planar conformation which can be stacked into columns without any steric hindrance as revealed by XRD pattern and molecular modeling [31]. In case of the higher generations steric crowding of the terminal groups restricts the dendrimers from such planar structure. The columnar mesophases are formed by face to face stacking of the unimolecular discs and an increased interaction between successive macrodiscs offers positive contribution towards the collective stability of the columnar mesophase.

2.5. Supramolecular dendromesogens Supramolecular dendromesogens are consisting in dendritic branches which self-assembled together into supramolecular columns (cylindrical) and/or spherical dendrimers which further self-organize into various liquid crystalline mesophases. Since because unlike dendritic branches (which grow linearly with increasing generation number of the dendrimer) the number of promesogenic unit grows geometrically, therefore there is a limit for the growth of the diameter of the supramolecule. To avoid exceeding this limit, which is marked by the length of the repeating unit of the dendrimer and the branch multiplicity, the supramolecule undergoes conformational changes leading to various phases. Supramolecular dendritic concept was developed by Percec through exploration of self-assembling nature of perfect dendritic moieties [32]. In his attempt to find out the criteria those governing the shape and size of supramolecular dendromesogens he investigated libraries of monodendrons whose structures were rationally designed. He observed that variation of the number of peripheral aliphatic chains and the position of grafting influences the rate of growth or deformation of the dendromesogens and hence the structural organization. The structural organization also depends up on the generation number of the dendrimer. Generally cylindrical or pseudo-cyllindrical supramolecular dendromesogens are obtained by the self assembly of the G2 monodendrons and some times with G3 and even with G4 depending on the connectivity of the terminal end-groups while G1 monodendrons are devoid of mesomorphism [32]. On an average they exhibit a flat tapered fan or semi-discoid shape. The influence of the apical function,

Liquid crystalline dendrimer: Towards intelligent functional materials

117

which is usually hydrophilic in nature, is also crucial for the stabilization of the mesophase. These tapered dendritic molecules through a process of self assembly may organize into infinite supramolecular columns with a polar interior. These columns, separated from each other by the aliphatic medium, then self-organize into rectangular and/or hexagonal lattices (figure 11). Upon increasing dendritic connectivity and the number of terminal chains the flat tapered conformation might deform to important 3D structures like conical or even to pseudo-spherical supramolecular entities and cubic self-assemblies. These spherical motives contain the polar portions within which in turn self assemble into miceller cubic phase. By controlling the overall molecular shape various mesophases like smectic, Colh, Colr, CubI, CubP, Tet (a LC phase with a tetragonal three dimensional unit cell), LQC (Liquid quasi crystal) etc. can be obtained from these dendromesogens [33].

Figure 11. Schematic representation of the self-assembly of dendromesogens from flat tapered to cylindrical shape.

118

Amrit Puzari

Eventually these supramolecular self-organizations are driven by intramolecular micro-segregation, steric factors as well by H-bonding interactions. X-ray crystallographic data revealed that the cubic phases result from the self-assembly of these spheroid dendrimers with Polar Regions embedded at the crystallographic position of the group [34]. These supramolecular spheres are also deformable, interacting one with the other through relatively soft pair potential and thus can be distorted into an oblate shape. The skeleton of these self-assembling monodendrons are valuable building blocks used in the construction of nanoscale objects with novel functional utility and more complex functional architectural motives where micro segregation between polar and apolar regions plays a significant role in attributing the observed mesophase along with flexibility of the backbone [35]. The intermolecular interactions among the promesogenic units also determine the conformation of the dendromesogens which adapts to their requirements as long as it is conformationally and sterically possible. For example a cylindrical symmetry in the molecule might lead to supramolecular columnar arrangement.

2.6. Liquid crystalline metallodendrimers Although vast majority of dendrimers are purely organic systems, but metallodendrimers are not few in numbers. Incorporation of the metal into the dendritic matrix opens up the possibility of application of metallodendrimers in various areas of science. Examples of such applications includes as efficient catalysts due to the high concentration of active sites, electro-active molecules due to the multiredox center, sensor applications and other photo-physical applications. Therefore metallodendrimers are supramolecular entities with greater potential for multidisciplinary areas [36]. More interestingly the properties of such macromolecules can be tuned and modulated by the specific location of the active moieties within the dendrimer. The metal can be incorporated at the core or even at the periphery of the dendrimer. LC metallodendrimers are very few in numbers and usually contain minidendrons of first generation with exception to ferrocenyl dendrimers. A few of such liquid crystalline metallodendrimers are known and upon complexation with suitable ligands they can exhibit mesomorphic behaviour. Structure of a dendritic oxovanadyl compound complexed with salen based ligand is shown in figure 12, which exhibits columnar mesophase (Colr and Colh) [37]. Rigidification of the central chelating part probably leads to occurrence of mesophase.

119

Liquid crystalline dendrimer: Towards intelligent functional materials

Coordination of other metal centers such as Co(II), Ni(II), Cu(II), Zn(II) etc. to dendritic imine ligands also produces dendrometallomesogens with stabilized mesomorphism [38]. Examples of lower generation PPI dendrimers involved in the formation of such metallodendrimers through coordination with MX2 (M is the metal center) type of salts are shown in figure 13. These dendrimers those are end functionalized with mesogenic salicylaldimine are considered as potential multicoordinative organic ligands R2

R1 R1

O O

(CH2)n

R2 R2

R3 R3

n=2,3 R1, R2, R3 =H/OCnH2n+1

R2 R1

= G1

Figure 12. Structure of an oxovanadyl dendritic complex. R

H R N R

N R H

N R

H N

HN N

C10H21O

Figure 13. Structure of dendritic imine ligands D1 and D2 involved in metallodendrimer formation.

120

Amrit Puzari

for obtaining LC metallodendrimers. Quasi metallodendrimers involving one, two or three metallic centers were thus obtained from a mesomorphic multicoordinative adequately functionalized dendritic imine ligands [39].

3. Utility of LC dendrimers as functional materials Dendritic architecture, the most pervasive topology has been assigned as the unique one in terms of its capability for total control over molecular design structural parameters, control over branching pattern, molecular size, structure and morphology. These features project them as a newer class of potential material for development of intelligent functional material. The ability of dendritic motifs to self assemble to well defined and nanosized dendritic macromolecules will continually increase their importance in material science and serve as an interesting and fascinating motif in nanoscience, nanotechnology and other interdisciplinary field. Current investigation on LC dendrimers mainly focuses on the application prospects. For example due to micro phase separation between polar and apolar regions different mesophases are observed for ionic derivatives of linear carboxylic acid with PAMAM or PPI liquid crystalline dendrimers. Such dendrimers are modified to obtain birefringent glasses at room temperature and viscous smectic A phase at higher temperature [40]. These dendrimers possess terminal flexible alkyl carboxylic acids separated by a region of high density of the ionic ammonium carboxylates and these LC dendrimers may find interesting application as anisotropic ionic conductive material [41]. The spin functional soft dendritic materials capable of undergoing spin transition at specific temperature range have potential for sensor application because of their quick response of the spin cross over phenomena to temperature change. Fujigaya et al. has reported about some dendritic triazole Fe(II) complex [obtained from Fe(II) ion, dendritic triazole ligands with polybenzyl ether dendrons] having alkyl sulphonate as counter ion, which can undergo thermal induced discolouration â&#x20AC;&#x201C; colouration cycle and that can be repeated without any sign of deterioration [42]. Phenomena of self-assembly in some of these dendrimers can attribute characteristic texture due to liquid crystalline mesophase. For such dendrimers the spin-transition and phasetransition events are perfectly synchronous to each other and show quick response to temperature change and hence are projected as promising materials for sensor application. Astruc et al. has reviewed the application prospects of liquid crystalline dendrimer [43]. Percec and his group observed the fact that when electron acceptor or donor organic groups are placed at the focal point of suitable dendrons, the supramolecular organization in such cases produced liquid crystalline

Liquid crystalline dendrimer: Towards intelligent functional materials

121

dendrimer with high charge carrier mobilities. The materials produced could have interesting electronic and opto-electronic applications [44]. Deschenaux et al. have argued that properties of fullerene-containing thermotropic liquid crystals could be of interest for applications like molecular switches; solar cells etc. and hence find importance in nanotechnology. They designed fulleropyrrolidines representing family of electroactive macromolecules. Different substituents, for example ferrocene units, can be attached to the N-atom of the fulleropyrrolidine which can make them suitable candidates for photovoltaic applications or supramolecular switches [45]. The ferrocene and the fullerene groups constitute a well known redox system with marked electrochemical properties. The supramolecular organization promoted by the mesogenic group enhances the photophysical and electrochemical properties of the supramolecule. We can generalize that fullerenes covalently functionalized with mesomorphic dendrimers can produce self assembled liquid crystalline architecture that can be tuned by molecular design to obtain tailor made characteristics for application purpose. Thermotropic liquid crystals are already been used in most of our daily used items like watches, calculators, mobile telephones etc. Therefore liquid crystalline dendrimers are certainly interesting materials for application studies. The beauty of such compounds is that liquid crystalline properties can be controlled as a function of dendrimer generation and can be tuned by careful selection of mesogenic group. For example ferroelectric properties were obtained for dendrimers functionalized with chiral mesogenic moieties [46]. Metallocenyl dendrimers due because of their electrochemical properties have attracted much attention in the field of molecular electronics. Thus smectic C and smectic A liquid crystalline phase has also been observed with LC dendrimers such as ferrocenyl and C60 terminated dendrimer [47]. Photosensitive ionic nematic liquid crystal complexes are also known [48]. PAMAM and PPI dendrimers can be decorated with appropriated groups to obtain liquid crystalline dendrimers having smectic and columnar mesophase and with interesting photophysical properties [49]. The development attained so far in the field of liquid crystalline dendrimers throws light on the bright prospect of emergence of newer intelligent opto-electronic materials for novel applications. Furthermore dendrimers containing electroactive groups at the center or on the periphery can help understanding biological electron transfer and are excellent candidates for applications like as catalyst, electron transfer mediator, ion sensors etc. Proper understanding of the connection between structure and function will provide deeper insight into the behaviour of biological systems too, so that the scope of the application can also be extended to real world problems.

122

Amrit Puzari

Conclusion Liquid crystalline dendrimers constitute the most interesting topology for generation of varieties of nanoscopic devices, intelligent materials with designer structural and functional properties. This molecular architecture might led to newer ideas for the development of materials for which their design is only limited by human imagination. With the high molecular weight monodisperse dendrimers and dendrons, liquid crystalline dendritic motifs can be generated containing the conventional liquid crystalline phases like nematic, lamellar, columnar, cubic phase as well less conventional mesophases such as onion, segregated and porous columnar mesophases, multilayerd and dark smectic phases. Chemical nature and connectivity of the mesogenic groups greatly influence the mesomorphic properties and appropriate selection of such mesogenic groups will make possible the design of newer functional materials with tailor made properties. The selforganization of these polymeric structures are gaining significant importance for obtaining intelligent nanoscale architectures with specific physical properties like magnetic, electronic or optoelectronics which in turn find important application in the fields like supramolecular chemistry, nanotechnology, biotechnology and other relevant fields.

Acknowledgements I gracefully acknowledge the timely help and encouragement offered by Dr. Agnieszka Iwan, the editor of the book, in making this effort a successful one.

References 1.

2. 3. 4. 5. 6.

a) Handbook of Liquid crystals (Eds.: Demus, D. 1998,. Goodby, J. W. Gray, G. W., Spiess, H. â&#x20AC;&#x201C;W. Vill, V. Eds) Wiley-VCH, Weinheim, 1998; b) Gray, G.W. 1987, Thermotropic Liquid Crystals, Wiley, Chichester. De Gennes, P.G. 1993, The Physics of Liquid Crystals, 2nd ed., Oxford University Press, New York. Kato, T. 2002, Science 295, 2414. Bahadur, B. 1992 Liquid Crystal â&#x20AC;&#x201C; Applications and Uses, Vol. 1-3, World Scientific Publishing Co. Pte. Ltd., Singapore. Schambil, F., Paasch, S., Schwuger, M.J. 1988, Fett Ole Fettderivate Folgeprodukte, 114, 295. Goodbye, J. W., Slancy, A. J., Booth, C. J., Nishiyama, I., Vuijik, J. D., Styring P., Toyne, K. J. 1994, Mol. Cryst. Liq. Cryst., 243, 231.

Liquid crystalline dendrimer: Towards intelligent functional materials

7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

27.

28. 29. 30.

123

Tschierske, C. 2001, J. Mater. Chem., 8, 2647. Tomalia, D. A. 2004, Aldrichimica Acta, 37(2), 39. Tomalia, D. A., Frechet, J.M.J. 2002, J. Polym. Sci., Part A : Polym. Chem. 40, 2719. Richardson, R. M., Ponomarenko, S. A., Boiko, N. I., Shibaev, V. P. 1999, Liq. Cryst., 26, 101. Percec, V., Cho, W. â&#x20AC;&#x201C;D. Mosier, P. E., Ungar, G., Yeardley, D. J. P. 1998, J. Am. Chem. Soc., 120, 11061. Durand-Reville, T., Gobbi,. L. B., Gray, B. L., Ley, S.V., Scott, J.S. 2002, Org. Lett. 4, 3847. Baars, M.W.P.L., Sontjens, S.H.M., Fischer, H.M., Peerlings, H.W.I., Meijer, E.W. 1998, Chem. Eur. J. 4, 2456. Cameron, J.H., Fischer, A., Lattermann, G., Diele, S. 1997, Adv. Mater., 9, 398. Donnio, B., Buathong, Bury, S. I., Guillon, D. 2007, Chem. Soc. Rev., 36, 1495. a) Barbera, J., Marcos, M., Serrano, J.L. 1999, Chem.-Eur. J., 5, 1834 b) Marcos, M., Gimenez, R., Serrano, J. L., Donnio, B., Heinrich, B., Gullion, D. 2001, Chem.-Eur. J., 7, 1006 c) Donnio, B., Barbera, J., Gimenez, R., Guillon, D., Marcos, M., Serrano, J. L. 2002, Macromolecules, 35, 370. a) Donnio, B., Guillon, D. 2006, Adv. Polym. Sci., 201, 45 b) Ponomarenko, S.A., Boiko, N.I., Shibaev, V.P. 2001, Polym. Sci. C, 43, 1. Ponomarenko, S. A., Boiko, N.I., Shibaev, V.P. Richardson, R. M., Whitehouse, I. J., Rebrov, E.A., Muzafarov; A.M. 2000, Macromolecules 33, 5549. Cho, B.K., Jain, A., Mahajan, S., Ow, H., Gruner, M., Wiesner, U. 2004, J. Am. Chem. Soc. 126, 4070. Busson, P., Ortegren, J., Ihre, H.,Gedde, U.W., Hult, A., Andersson, G., Eriksson, A., Lindgren, M. 2002, Macromolecules, 35, 1663. Gehringer, L., Guillon, D., Donnio, B. 2003, Macromolecules, 36, 5593. Gehringer, L., Bourgogne, C., Guillon, D., Donnio, B. 2004, J. Am. Chem. Soc. 126, 3856. Gehringer, L., Bourgogne, C., Guillon, D., Donnio, B. 2005, J. Mater. Chem., 15, 1696. Kato, T., Mizoshita, N., Kishimoto, K. 2006, Angew. Chem., Int. Ed., 45, 38. Saez, I. M., Goodby, J. W. 2005, J. Mater. Chem., 15, 26. a)Wooley, K. L., Hawker, C. J., Frechet, J. M. J., Wudl, F., Srdanov, G., Shi, S., Kao, M. 1993, J. Am. Chem. Soc., 115, 9836 b) Hawker, C. J., Wooley, K. L., Frechet, J. M. J. 1994, J. Chem. Soc., Chem. Commun, 925. a) Hawker, C. J., Frechet, J. M. J. 1990, J. Chem. Soc., Chem. Commun., 1010 b) Hawker, C. J., Frechet, J. M. J. 1990, J. Am. Chem. Soc., 112, 7638 c) Frechet, J. M. J. 1994, Science, 263, 1710. Dardel, B., Guillon, D., Heinrich, B., Deschenaux, R. 2001, J. mater. Chem. 11, 2814. Pesak, D. J., Moore, J. S. 1997, Angew. Chem. Int. Ed. Engl., 36, 1636. Meier, H., Lehmann, M., Kolb, U. 2000, Chem.-Eur.J., 6, 2462 b) Lehmann, M., Schartel, B., Hennecke, M., Meier, H. 1999, Tetrahedron, 55, 13377.

124

Amrit Puzari

31. Rosenfeldt, S., Karpuk, E., Lehmann, M., Meier, H., Lindner, P., Harnau, L., Ballauff, M. 2006, ChemPhysChem, 7, 2097. 32. a) Percec, V., Cho, W.D., Ungar, G., Yeardley, D.J.P. 2001, J. Am. Chem. Soc. 123, 1302 b) Percec, V., Cho, W.D., Mosier, P.E., Ungar, G., Yeardley, D.J.P. 1998, J. Am. Chem. Soc. 120, 11061 c) Percec, V., Cho, W.D., Ungar, G. 2000, J. Am. Chem. Soc. 122, 10273 d) Percec, V., Cho, W.D., Ungar, G., Yeardley, D.J.P. 2000, Angew. Chem. Int. Ed., 39, 1597. 33. Percec, V., Mitchell, C. M., Cho, W. D., Uchida, S., Glodde, M., Unger, G., Zeng, X., Liu, Y., Balagurusamy, V. S. K., Heiney, P. A. 2004, J. Am. Chem. Soc. 126, 6078. 34. a) Hudson, S. D., Jung, H. T., Percec, V., Cho, W. D., Johansson, G., Ungar, G., Balaguruswamy, V. S. K. 1997, Science, 278, 449 b) Ungar, G., Percec, V., Holerca, M. N., Hudson, Johansson, G., Heck, J. A. 2000, Chem.-Eur. J. 6, 1258. 35. a) Frechet, J.M.J. 1994, Science 263, 1710 b) Tomalia, D. A. 1994, Adv. Mater. 6, 529 c) Percec, V. Ahn, C. –H., Barboru, B. 1997, J. Am. Chem. Soc. 119, 12978 d) Fischer, M., Vogtle, F. 1999, Angew. Chem. Int. Ed. 38, 884 e) Bosman, A. W., Janssen, H. M., Meijer, E. W. 1999, Chem. Rev. 99, 1665. 36. Newkome, G. R., He, E., Moorefield, C. N. 1999, Chem. Rev. 99, 1689. 37. Serrette, A. G., Swager, T. M. 1993, J. Am. Chem. Soc., 115, 8879. 38. Stebani, U., Lattermann, G., Wittenberg, M., Wendorff, J. H. 1996, Angew. Chem., Int. Ed. Engl., 35, 1858. 39. Marcos, M., Omenat, A., Barberá, J., Durán, F., Serrano, J. L. 2004, J. Mater. Chem., 14, 3321. 40. Felekis, T., Tsiourvas, D., Tziveleka, L., Paleos, C. M. 2005, Liq. Cryst. 32, 39. 41. Martín-Rapún, R., Marcos, M., Omenat, A., Barberá, J., Romero P., Serrano, J. L. 2005, J. Am. Chem. Soc., 127, 7397. 42. Fujigaya, T., Jiang, D. –L., Aida, T. 2007, Chem. Asian J., 2, 106. 43. Astruc, D., Boisselier, E., Ornelas, C., 2010, Chem. Rev. 110, 1857. 44. Percec, V., Glodde, M., Peterca, M, Rapp, A., Schnell, I., Spiess, H. W., Bera, T. K., Miura, Y., Balagurusamy, V. S. K., Aqad, E., Heiney, P. A., 2006, Chem.– Eur. J., 12, 6298. 45. Campidelli, S., Lenoble, J., Barberá, J., Paolucci, F., Marcaccio, M., Paolucci, D., Deschenaux, R. 2005, Macromolecules, 38, 7915. 46. a) Busson, P., Ihre, H., Hult, A. 1998, J. Am. Chem Soc, 120, 9070 b) Boiko, N., Zhu, X., Vinokur, R., Rebrov, E., Muzafarov, A., Shibaev, V. 2000, Mol Cryst Liq Cryst, 352, 343. 47. Percec, V., Chu, P. W., Ungar, G., Zhou, J. P. 1995, J. Am. Chem. Soc. 117, 11441. 48. Marcos, M., Alcala, R., Barbera, J., Romero, P., Sanchez, C., Serrano, J. L. 2008, Chem. Mater. 20, 5209. 49. Antharjanam, P. K., Sudhadevi, J., Jaseer, M., Ragi, K. N., Prasad, E. J. 2009, Photochem. Photobiol. A: Chem. 203, 50.

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 125-152 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

5. Liquid crystallinity in polymers – Liquid crystalline epoxy resins 1

Beata Mossety-Leszczak1 and Magdalena Włodarska2 Faculty of Chemistry, Rzeszów University of Technology, W. Pola 2, 35-959 Rzeszów, Poland 2 Institute of Physics, Technical University of Łódź, Wólczańska 119, 90-924 Łódź, Poland

Abstract. The present day materials science and materials technology are directed towards the development of task-oriented structures. Liquid crystalline polymers (LCP) are examples of such materials. Most of them contain mesogenic moieties essentially identical with those of the low-molecular weight liquid crystals. The properties of LCP depend on the structure of these polymers and the position of mesogenic groups within polymer molecules. The mesogens can be located along the main chain or extend sidewise of the chain. Combinations of these two have also been reported. It is possible to cross-link the liquid crystalline monomers or oligomers into a network that either retains the capability of undergoing phase transition involving a liquid crystalline state, or have the mesomorphic order frozen in at the curing stage. In this report we discuss briefly the history of development and properties of thermoplastic liquid crystalline polymers and liquid crystalline polymer networks. The main part of our study is related to synthesis, properties and potential applications of liquid crystalline epoxy resins. Correspondence/Reprint request: Dr. Beata Mossety-Leszczak, Faculty of Chemistry, Rzeszów University of Technology, W. Pola 2, 35-959 Rzeszów, Poland. E-mail: mossety@prz.edu.pl

126

Beata Mossety-Leszczak & Magdalena Włodarska

Abbreviations LCE LCP LCPN MC-LCP SC-LCP MC/SC-LCP MC-LCE SC-LCE MC/SC-LCE SmC Tg

– – – – – – – – – – –

liquid crystalline elastomer liquid crystalline polymer liquid crystalline polymer network main chain liquid crystalline polymer side chain liquid crystalline polymer main chain/side chain liquid crystalline polymer main chain liquid crystalline elastomer side chain liquid crystalline elastomer main chain/side chain liquid crystalline elastomer smectic C phase temperature of glass transition (vitrification)

Introduction Materials capable of self-organization on the molecular level are of great importance in modern technology. An example of such materials are liquid crystalline polymeric systems, which combine typical features of liquid crystalline compounds, i.e. anisotropy of optical, dielectric, magnetic and other properties, with properties typical for polymers. Thermoplastic polymers built from linear and comb-like macromolecules are already relatively well known [1-10]. They are characterized by high impact resistance and high elastic modulus along the direction of molecular orientation. A few of them, despite of high price, found practical applications. Liquid crystalline polymer networks are another group of materials with interesting properties and many potential applications. Many studies have been devoted to elastomeric polymers of low cross-linking density, containing mesogenic groups in side chains. They are relatively easy to produce with a wide possibility of introducing significant structural modifications. Considerable attention is paid to cross-linked polymers with mesogenic groups in main chains and also to polymers obtained by curing liquid crystalline monomers, in which dense cross-linking results in ‘freezing’ the liquid crystalline phase. The thus obtained materials have a very high degree of order as well as mechanical and thermal resistance. Uncrosslinked liquid crystalline polymers can form both thermotropic and lyotropic mesophase. Combination of properties of the anisotropic liquid crystalline systems with typical properties of polymers creates possibility of obtaining new materials having exceptional toughness needed in many special applications. Liquid crystalline polymers are characterized by several distinguishing features:

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

− − − − − − −

127

very high strength, particularly in the direction of mesogen orientation, high elastic modulus in the direction of mesogen orientation, relatively low viscosity of precursors, melts or solutions, high chemical and fire resistance - due to ordered, compact structure, low solubility and resistance to solvents, high temperature of vitrification/melting, low thermal expansion coefficient resulting in high stability of shape and possibility of molding products with high dimensional precision.

Liquid crystalline monomers and polymeric precursors are also used to obtain anisotropic polymer networks. Much like in classic systems, permanent connections between chains cause significant changes in physical properties of the polymer. These properties depend on the structure of starting reagents and on the density of cross-linking. The mechanism and methods of obtaining such networks are analogous to those applied for classical polymer networks. Well-known curing reactions are used, with typical monomers and polymeric precursors. Depending on the type of reactive functional groups, curing can be initiated photochemically, thermally or by reactive curing agents. The main difference is that mesogenic groups are introduced into the molecules in the stage of precursor synthesis. The presence of these groups in the final product substantially affects the properties of the resulting material. This is because the precursor fragments can relatively easily be oriented and create ordered structures, which leads to anisotropy of physical properties. Similarly to the classical covalent polymer networks, the LCPN have been divided into two groups, taking the physical state of polymeric chains as the criterion: − −

polymers used in high elasticity region (liquid state), usually weakly cross-linked, referred to as the liquid crystalline elastomers (LCE), polymers used in glassy state (LCPN – liquid crystalline polymer networks), with ‘frozen’ liquid crystalline ordering, practically unchanging up to the degradation temperature.

1. Uncrosslinked liquid crystalline polymers Liquid crystalline polymers are relatively new group of materials, though, according to Brostow [11], it was Vorländer, who first synthesized and studied many low molecular weight liquid crystalline systems, stating already in 1923 that polymeric systems can also have mesomorphic properties. Even earlier, in turn of the 19th and 20th century, Klepl, and

128

Beata Mossety-Leszczak & Magdalena Włodarska

later Fischer, obtained a liquid crystalline polymer, a derivative of p-hydroxybenzoic acid, but they were not aware of unique properties of the obtained material. Only Vorländer, studying this polymer, discovered its mesophase. This is why Vorländer is often called the ‘father’of liquid crystalline polymeric systems. Next information about liquid crystalline macromolecules was reported in 1937 by Bawden and Pirie. They studied solutions of tobacco mosaic virus and observed that elongated molecules a few thousand angstroms long form two separated phases, one of which, with higher polymer concentration, was characterized by optical birefringency [12]. However, studies on liquid crystalline polymers, both theoretical and experimental, have only started to be carried out more intensively in the 1950s. Flory in 1956 gave criteria of anisotropic phase formation, theoretically analyzing the behavior of rigid molecules in solutions. These criteria were relevant mainly to formation of mesophase in lyotropic solutions, but, by assuming the solution concentration of zero, they could also be applied to thermotropic polymers [13,14]. At the same time, a series of lyotropic polyglutamates was obtained at Courtaulds laboratory in Maidenhead, England, during research on synthesis of artificial silk, primarily on poly(γ-benzyl-L-glutamate) [15]. These systems, however, did not find wide practical applications, though a process of producing fibers from them was patented [16]. A breakthrough in the synthesis of liquid crystalline polymers was the achievement of Stephanie Kwolek from DuPont, who synthesized and managed to process lyotropic aromatic polyamides, poly(p-benzamide) [17] and poly(p-phenyleneterephthalamide) [18]. The latter is the well known Kevlar. Fibers formed from lyotropic solutions of these polymers have extraordinary mechanical properties, in particular high elastic modulus and high tensile strength. They are used for fabrication of protective wear including bulletproof vests and composites of very high endurance, which are key components of helmets and brake pads [9]. Initially, attention was focused on the properties of lyotropic polymeric systems. The most important group of such polymers are polypeptides and aromatic polyamides (besides the synthetic lyotropic polymers, also natural high molecular weight compounds soluble in water are known, as well, e.g. nucleic acids: DNA and RNA, collagen, cellulose and some viruses). First reports on thermotropic polymers were published in 1960s, but mesomorphic character of the described systems did not stimulate a lot of interest at that time [19-21]. Intensity of studies on these polymers grew considerably in 1970s. They concerned the synthesis of linear copolyesters of ethylene terephthalate and 4-hydroxybenzoic acid [22] and those polyester

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

129

obtained in reaction of dicarboxylic acid chlorides with bis(4-hydroxyacetophenone)azine [23]. The properties of these polymers confirmed the predictions by de Gennes, who suggested that main chain thermotropic liquid crystalline polymers could be obtained by introducing elastic fragments between rigid mesogenic groups [24, 25]. In the end of 1970s, further reports came out about the synthesis of comb polymers with mesogens in side chains [26-28]. Since then, the synthesis and studies on the properties and applications of thermotropic polymeric liquid crystals became more intense. Numerous types of such polymers were reported and described in many books and review papers published on the topic [2, 6-10, 29-34]. Most of the systems contained typical mesogenic groups which appeared also in the low molecular weight liquid crystals. Polymers not containing typical mesogenic groups were also found to form crystalline phases, often in a wide range of temperature. The examples are: polyphosphazenes (–[RP=N]n–, R= ClC6H4O), polysiloxanes (–[R2SiO]n–, R= Et, n-Pr) and polysilanes (–[R2Si]n–, R = n-Bu) [35, 36]. A classifying criterion of liquid crystalline polymeric systems is the location of mesogenic group in the macromolecule. Mesogens can be introduced into the polymeric main chain (MC-LCP - main chain liquid crystalline polymer) or be present in side groups (SC-LCP - side chain liquid crystalline polymer). Synthesis of liquid crystalline polymers with complex structure has also been described, with mesogens built into both main chain and side chains (MC/SC-LCP). Schematic structure of some of them is shown in Figure 1. The liquid crystalline order can also be used as a criterion of classification of the materials, similarly as in the case of low-molecular weight compounds. However, the microscopic structure of smectic polymers is considerably different from classic smectics. A few examples of these structures, including different ordering schemes depending on the length of mesogens and flexible spacers between them, are shown in Figure 2 [7]. 1

MC-LCP

MC/SC-LCP

- mesogenic group

Figure 1. Possible location of mesogens in the macromolecule.

130

Beata Mossety-Leszczak & Magdalena WĹ&#x201A;odarska Â

Nematic phase; linear polymer

Smectic phase; linear polymer

Smectic phase; mesogens in side groups

Various packing schemes of mesogenic side groups in polymeric smectic phase Figure 2. Basic types of ordering in liquid crystalline polymers.

Detailed classification, including mesogen types (discotic, rod-like) and their placement both in macromolecule and in ensemble, can be found in the literature [7, 11, 34].

1.1. Main chain liquid crystalline polymers (MC-LCP) Polymers with mesogenic groups in the main chain are obtained mainly in polycondensation or polyaddition processes. These are most often liquid crystalline polyesters [1, 34, 37-41], polyamides [42-44], polyurethanes [4547], less often polyethers [48-50], polycarbonates [51, 52], polyacethylenes [53, 54], or polyimides [55]. The fundamental structural components of main chain liquid crystalline polymers are rigid, rod-like and disc-like aromatic mesogenic groups similar to those present in the low molecular weight liquid

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

131

crystals. Direct linking of mesogenic groups would lead to materials of very high temperature of mesophase transition, typically very close to, or exceeding, the thermal degradation limit. For example, homopolymer of 4-hydroxybenzoic acid undergoes thermal degradation above 500°C, before even reaching the melting point [9, 35]. The main problem in the synthesis of such systems is to find a proper modification of their structure that would decrease the temperature of phase transitions without losing liquid crystalline nature of the product. One possible method of such a modification is copolymerization of liquid crystalline monomers with comonomers which disrupt the regular structure of the macromolecule. Irregularity in the chain hinders crystallization, reduces crystallite sizes and partially hinders ordering of macromolecules, thereby lowering the temperature of devitrification/melting. This is a trick used in the production of a commercial copolyester of 4-acetoxybenzoic acid (ABA) and 2,6-acetoxynaphthene acid (ANA). The latter is known under commercial name Vectra A and manufactured by Ticona (formerly Celanese). Softening temperature of this copolymer is 280°C, i. e. lower than softening temperature of both ABA and ANA homopolymers. As the comonomers disrupting linearity, aromatic compounds substituted at ortho- or metapositions, are frequently used, e.g. isophthalic acid or 2,5-substituted thiophene derivatives [9]. The nonlinearity and lack of structural regularity of the macromolecule may, however, adversely affect thermal stability of the polymer. Another method of improving flexibility of macromolecules is to separate mesogens by flexible subchains (so-called spacers) which preserve linerality and structural regularity of macromolecules. These polymers are referred to as semi-rigid ones. Typical spacers are aliphatic hydrocarbon chains of various lengths, –(CH2)n–. Introduction of substituents breaking linearity of para-substituted aromatic chains increases the distance between chains and weakens interactions between them. The density of chain packing is reduced. Random placement of the substituent can, however, disturb liquid crystallinity of the polymer. Apart from the already mentioned polymer of the Vectra group (several types of aromatic copolyesters composed of 4-hydroxybenzoic acid and 6-hydroxy-2-naphthene acid have been commercialized, all of which described with a common name Vectra [34]), large significance has a commercial polymer named Xydar, manufactured by BP-Amoco. Chemically, it is a copolyester containing terephtalic acid, 4-hydroxybenzoic acid and 4,4’-biphenol units, which can be processed by injection molding [34]. These polymers are characterized by high mechanical durability,

132

Beata Mossety-Leszczak & Magdalena WĹ&#x201A;odarska

stretching resistance (due to chain orientation), low water sorption, impermeable by gases and high chemical resistance. Processing of polymers in liquid crystalline phase by extrusion or injection molding induces orientation of polymer chains in the direction of the flow. After cooling down, the product has well-organized structure and arrangement of molecules causes self-reinforcement. Extrusion method is used to produce plates, rods and fibers, whereas by press or injection molding one obtains construction supports, joints and switches used in microelectronics.

1.2. Side chain liquid crystalline polymers (SC-LCP) Side chain liquid crystalline polymers were first obtained over 30 years ago. Of key significance for the synthesis of such systems were works of Finkelmann, Ringsdorf and coworkers [26, 27] and Shibaev [28], who were the first to obtain and study these polymers. A lot of essential information concerning side chain liquid crystalline polymers can also be found in the works of McArdle [5]. A flexible polymer chain in amorphous phase tends to attain coil-like conformation. In order to enable orientation of mesogenic groups introduced as branches, they have to be attached to the main chain by means of flexible spacers, which ascertain freedom of conformational motions of the parts of molecule and reduce stress between main chain and rigid groups in branches. Final properties of the system depend on the structure of the polymer main chain, the type of mesogenic groups, their density, and length of flexible spacers. The discussed systems usually have liquid crystalline properties similar to the properties of low molecular weight mesogenic precursors. Formation of polymeric structure often stabilizes the mesomorphic phase or increases the molecular order. Liquid crystalline comb-like polymers are obtained primarily by homopolymerization of monomers containing side mesogenic groups, copolymerization of mesogenic and non-mesogenic monomers or copolymerization of different mesogenic monomers. Another method involves attaching low molecular weight mesogenic groups to the polymer chain by means of appropriate reactions, e.g. by hydrosilylation [4]. According to published data, rigid mesogenic groups are the most often attached to polysiloxane, polyacrylic or polymetacrylic chains, less often to chains such as polycarbosilane, polystyrene, polyether, polyester, polyurethane, polyphosphazene, etc. [3-7, 31, 32, 56-63]. Practical applications of liquid crystalline systems can be determined by their susceptibility to the orienting effect of an external force field. It turns out that polymers with side chain mesogens can be oriented, e.g. in a

Liquid crystallinity in polymers â&#x20AC;&#x201C; Liquid crystalline epoxy resins

133

magnetic field, much easier than the main chain polymers [34]. To date, many different comb-like systems have been described. They became the most popular objects of scientific research in the field. They can find applications in many areas of technology; as components of information storage systems, high-tech electrooptics, components of optical memory, optical filters and polarizers, in non-linear optics, holographic imaging or stationary phases in gas chromatography [5].

2. Liquid crystalline polymer network 2.1. Liquid crystalline elastomers (LCE) Liquid crystalline elastomers, introduced as a theoretical concept by de Gennes [24, 25], are systems having low cross-linking density, with reversible phase transitions during heating or cooling. Above the glass transition (Tg), often equal to the temperature of transition into the liquid crystalline state, they become elastic and can be subjected to large mechanical deformations, during which mesogen orientation occurs or liquid crystalline order disappers. This is usually accompanied by changes in physical properties, e.g. transparency. In mesomorphic state, magnetic or electric fields have also the ordering effect. Preservation of the internal order can be achieved by lowering the temperature below Tg. The synthesis and properties of LCE has already been a subject of several reviews and books [10, 64-71]. LCE are obtained from various liquid crystalline monomers. They primarily comprise polymeric systems into which reactive functional groups are introduced in quantities ensuring relatively low cross-linking density. Similarly to liquid crystalline polymers, LCEs are divided into elastomers with mesogenic groups in (i) side chains (SC-LCE), (ii) main chain (MC-LCE) and (iii) combined systems (MC/SCLCE). In the last one the mesogenic groups (not necessarily the same) are found both in the main chain and in the branches (cf. Figure 3) [67, 70, 71]. In all liquid crystalline systems, of both low and high molecular weight, practical significance and applications rely on the possibility to enforce a long-range alignment of mesogenic groups and obtain a material with monodomain order. In the case of liquid crystalline elastomers, the parameter important for potential applications is the response to mechanical strain (e. g. change in transparency) resulting from changes in the arrangement of mesogenic groups. In branched systems with mesogens in side chains (SC-LCE), orientation of mesogenic groups can be modified relatively easily. The ability of mesogens to align and the dynamics of this process depends

134

Beata Mossety-Leszczak & Magdalena WĹ&#x201A;odarska

SC-LCE

MC-LCE

MC/SC-LCE

Figure 3. Types of structures in liquid crystalline elastomers.

mainly on the length of spacers and cross-link density. At low cross-link density, the side chains retain considerable freedom of motion. In elastomers with mesogens built into the main chain, mechanical strains also change mesogen orientation, like in SC-LCE, but these systems are harder to rearrange. In combined systems, the presence of mesogens both in the main chain and in side branches leads to a complex behavior under the influence of temperature and mechanical strains. These elastomers can be seen as a combination of two subsystems, whose properties strictly depend on the structure of the main polymer chain, mesogenic groups in branches and mutual interactions. The earliest synthesized elastomers had mesogens in side chains. Continuing studies on the side chain liquid crystalline polymers, Finkelmannâ&#x20AC;&#x2122;s group obtained a liquid crystalline elastomer using hydrosilylation reaction to produce polymeric polysiloxanes with mesogens in side chains. At room temperature, the product was opaque, typical for liquid crystals, and was also elastic. Mechanical deformation, e.g. stretching, caused reorientation of mesogenic groups, disappearance of turbidity and transition to transparent form [72]. Some liquid crystalline elastomers exhibiting internal order of SmC type can be heated up, stretched up to 300% of original length and cooled down, to retain the deformation. The system returns to the equilibrium state after re-heating which breaks the smectic order [73]. Liquid crystalline comb elastomers (SC-LCE) are also the systems with mesogens belonging to side branches in polyacrylate or polymethacrylate chains. Zentel and Reckert obtained such polymers by free

Liquid crystallinity in polymers â&#x20AC;&#x201C; Liquid crystalline epoxy resins

135

radical polymerization of acrylate and methacrylate esters containing attached mesogens and some amount of hydroxyl groups (5-8%) which were used to build cross-linking bonds in reaction with hexamethylene diisocyanate or 4,4'-methylene diphenyl diisocyanate. The liquid crystallinity did not change in the course of cross-linking. The elastomer exhibited the same type of mesophase as the polymeric precursor [74]. The majority of liquid crystalline elastomers with side chain mesogens produced nowadays have similar structure. The skeleton is polysiloxane, polyacrylic and/or polymethacrylic chain [5, 67, 69, 75-79]. In comparison with SC-LCE, the liquid crystalline elastomers with mesogens in the main chain were less studied. This is probably due to the necessity of running multi-stage reactions and a smaller range of potential applications. These systems were produced from linear liquid crystalline polymers obtained by solution polycondesation of mesogenic diols with allylmalonic acid or by melt polycondensation of the diols and diethyl allylmalonate. Cross-linking was carried out through double bonds in allyl groups, which reacted with Si-H groups in bifunctional Îą,Ď&#x2030;-oligosiloxanes. The final elastomer featured a wide temperature range of mesophase and smectic order [74, 80, 81]. A technique of LCE synthesis with the use of liquid crystalline diepoxies was described by Giamberini and coworkers. It is based on a reaction of liquid crystalline diepoxy oligomers with dicarboxylic acids [82-85] or with aromatic amine [84]. Methods of synthesis of other main chain LCE were also given in a work by Xie and Zhang [67]. The first combined liquid crystalline elastomer systems with mesogenic groups appearing both in the main chain and in side chains were synthesized using a polymer precursor obtained by melt polycondensation or copolycondensation of mesogenic diols and diethyl malonate. This linear homo- or copolyester was then cross-linked by hydrosilylation, similarly to MC-LCE. The flexible aliphatic chains introduced during the cross-linking lowered the phase transition temperatures. The presence of mesogenic groups in the main chain and in side chains provided several types of liquid crystalline phases [74, 80, 81]. Synthesis of other combined LCE and their properties were also reported [70, 71]. The largest and most important group of liquid crystalline elastomers is that with mesogens constituting side branches. This is a result of relatively easy synthetic route and wide range of potential applications. Elastomeric polymer systems with introduced photoactive and photosensitive chromophore groups may also have large practical significance. Introducing chiral carbon atoms gives elastomers with ferroelectric and piezoelectric properties. Liquid crystalline elastomers may be potentially used as gas-separating membranes whose properties (gas permeability) strictly depend on temperature and

136

Beata Mossety-Leszczak & Magdalena Włodarska

T<Tisotropisation Mesogen ordering in the nematic phase forces the polymeric chain to align perpendicularly to the mean direction of mesogen orientation.

T>Tisotropisation The nematic order disappears after heating up above the isotropisation temperature thus enabling relaxation of the polymer backbone to the coillike conformation.

Figure 4. Schematic example of elastomer ordering below and above the temperature of transition to the nematic phase.

change in particular during transition from and to the mesomorphic phase. The sensitivity of LCE towards temperature, strain and magnetic or electric field enables application of these materials as waveguides, actuators, switches in optical devices, piezo- and pyroelectric sensors. The change of molecular arrangement in different temperatures is schematically shown in Figure 4. Attempts are made to use the effect of shape change due to heating (e.g. by an IR laser) or applying an electric field to create ‘artificial muscles’ for manipulating micro-objects attached to them [86]. The ability to form monodomain structures may be used in integrated optics, electrooptics, to obtain materials with nonlinear optical properties. Such materials can also be used in information storage systems, as polarizers, optical filters or oriented layers in displays [9, 64, 67, 71, 87].

2.2. Rigid liquid crystalline polymer networks (LCPN) The interest in densely cross-linked liquid crystalline polymer materials results from the possibility of obtaining anisotropic solid systems with highly organized structure at the molecular level. This applies to polymers with Tg above the room temperature. Even though the term ‘liquid-crystalline’ suggests that these materials should be liquid, the term actually refers mainly to similarity of physical properties such as optical anisotropy or birefringence, which are characteristic for low molecular weight liquidcrystalline ordered structures. This order can be preserved in a reaction creating cross-linking bonds, amount of which is large in this case. As a

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

137

Figure 5. Schematic structure of a cross-linked liquid crystalline resin.

result rigid materials are obtained, stable practically up to the temperature of degradation. The presence of mesogenic groups also increases the mechanical durability. First practical attempts to obtain polymer networks aimed at preserving the order characteristic for soaps. They were conducted by Herz with coworkers [88], by radical polymerization of a system containing natrium styryl undecanoate and water, using divinyl benzene as curing agent. In this process, a thin film was prepared with preserved smectic-like soap order. Similar attempts were described by Blumstein [89], who preserved pseudo (quasi) smectic order in a single layer of polar molecules by polymerizing tetraethylene glycol dimetacrylate. The first ‘true’ liquid crystalline network, however, was obtained by Liebert and Strzelecki [90, 91]. Polymerization of mesogenic diacryl monomer of a type of Shiff base yielded a densely crosslinked material with preserved the structure characteristic for liquid crystals. It was stable up to degradation temperature [90, 91]. Further works of Blumstein’s group dealt with polymerization of systems similar to those of Liebert and Strzelecki, but in a magnetic field. Cross-linking reaction produced material with oriented structure, as confirmed by X-ray studies [92]. In parallel with the synthesis of LCPN-type systems, theoretical research was carried out, to elucidate the structure and properties resulting from crosslinking. As it was mentioned already, deGennes predicted the possibility of obtaining cross-linked systems in his early papers on the structure of liquid crystalline thermotropic polymers [24, 25]. These studies were later extended e.g. by Warner and Edwards [93, 93]. Theoretical analysis led to the conclusion that rigid molecules linked by transverse bonds can form unique structures and have special mechanical properties. Interesting properties and potential applications in high-tech technologies caused that many various LCPNs were designed and studied [9, 95-97].

138

Beata Mossety-Leszczak & Magdalena Włodarska

Thanks to high thermal and mechanical stability, such systems can be used to produce foils for electronic integrated circuits and as matrices for advanced composites. They can become parts of optical switches, materials with nonlinear optical properties, waveguides, etc. In particular, their increased durability and impact resistance were found excellent. LCPN are obtained mainly through cross-linking of low molecular weight liquid-crystalline monomers. The structure of a monomer corresponds, essentially, to the structure of low molecular weight liquid crystals. The molecule consists of rigid mesogenic groups and flexible side groups which must end with reactive functional groups to enable cross-linking. Length of the chains (spacers) linking the functional groups with the mesogen affects the rigidity of resulting networks. The length of spacers determines the elasticity and may lead to formation of the so-called semirigid-rod networks. Sometimes, the networks are also produced from monomers or oligomers which do not have liquid-crystalline properties by themselves, but contain rigid fragments which can induce anisotropic properties of the product through cross-linking [98, 99]. Cross-linking bonds can be formed during polymerization (self crosslinking), reactions initiated photochemically or thermally, or in reactions of monomers with curing agents. Photopolymerization is limited to materials of small thickness. Selection of curing conditions, particularly the temperature of the process, is limited by the range of liquid crystallinity of monomers. The largest amount of published information refers to liquid crystalline polymer networks obtained in reactions involving acrylates and methacrylates [96, 100-108], as well as isocyanate or cyanate [99-112], maleimide or nadimide [113-116], vinyl [117, 118], acethylene [119-122] and epoxy groups. In this report, special attention is given to liquid crystalline epoxy resins. 2.2.1. Liquid crystalline epoxy resins Anisotropic polymer networks obtained by cross-linking of epoxy precursors fall among the most extensively studied groups of these materials. In the 1980s, methods of synthesis of several kinds of cross-linked systems were patented [123-125]. All were obtained from mesogenic monomers with terminal epoxy groups. These were diglycidyl ethers of 4-hydroxyphenyl 4-hydroxybenzoate, biphenyl-4,4’-diol and 4,4’-dihydroxy-α-methylstilbene, which were cured, like classic bisphenol A epoxy resins, using amines, sulphamides or anhydrides. The networks were reported to retain liquidcrystalline properties of the monomers, despite the presence of stoichiometric amounts of non-mesogenic curing agents.

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

139

Up to now, the synthesis of many epoxy monomers and oligomers with embedded mesogenic group was elaborated, their curing reactions were studied and formation of anisotropic structures was analyzed [84, 85, 112, 126-179]. Epoxy monomers used in the synthesis of LCPN-type systems have the structure typical for low molecular weight liquid crystals. The mesogen is usually formed by rod-like aromatic groups, to which flexible chains with terminal epoxy groups are attached (Figure 6). Synthesis of the so-called twin mesogens epoxy monomers has also been described [95, 148, 156, 171], in which two mesogenic groups are separated by a flexible chain (Figure 7). Examples of most common mesogenic groups found in liquid-crystalline diepoxy monomers are given in Table 1. Methods of synthesis of the precursors, like those of bisphenol A epoxy resins, are based on the reactions of diol mesogens with epoxy compounds [84, 126-130, 137, 138, 140, 148, 152, 154-158, 161, 164-172], mainly epichlorohydrin. Less often, condensation is carried out of mesogenic dicarboxylic acids (usually, their chlorides) with glycidyl alcohol [135, 136]. These reactions are usually accompanied by formation of oligomeric homologues, but the oligomers can also be used to obtain LCPNs [140, 142]. Another method of introducing the epoxy ring is oxidation of mesogenic diolefins [132, 139, 141, 143, 144, 151, 159-163, 173-179]. The liquid crystalline properties depend on the length and flexibility of spacers connecting the mesogens with epoxy groups. The spacer length affects also the properties of the resulting polymer networks. In the case of systems produced in polycondensation reaction with epichlorohydrin, the length of flexible spacer is fixed. The method involving oxidation of double bonds provides greater flexibility of introducing spacers of different structure (e.g. aliphatic, oxyethylene) and arbitrary length. The length of the flexible chains facilitates orientation of molecules and lowers the phase transition temperature, but also reduces the degree of cross-linking and rigidity of

Figure 6. Molecular structure of a mesogenic epoxy monomer.

O Figure 7. Structure of a twin liquid crystalline molecule.

140

Beata Mossety-Leszczak & Magdalena WĹ&#x201A;odarska

Table 1. Mesogenic groups in epoxy monomers. Structure

References [112]

CH=N

Structure N=HC

CH=N

O C O

[112, 126, 128, 132, 141, 151, 153, 173, 178] [112, 126, 127, 138, 142, 143, 160, 165, 173]

O C O

References [112, 152, 156] [126]

O C O

O O C

R= H, Cl, CH3

[126]

O O C

O C O

[126]

O CH=CH C

N=HC

CH=N

[132-134, 137, 138, 144-146, 150, 159, 162, 173-175, 178, 179] [135]

[138, 158]

157,

[138, 168]

161,

R= H, CH3

[126, 127, 140, 146, 172]

CH3

CH=N

C=CH

N=HC R=H or CH3

CH3

[126, 129]

CH3

O C O

N=C

C=N

O O C

[126, 130] N=HC

[163]

N N

R= -O(CH2)n-1CH3; n=6-9

H 3C

CH3

H 3C

CH3

O C O

[166]

O O C

O S O

O C O

[177, 178]

[154] CH=N

R O S O

O O C

O C O

N=N

[143, 176]

O O C

[155]

[163]

[164]

R= -COOCnH2n+1; n=4,6,8,10,12 O O C O R O C

O R:

[84]

or O

CH3

H3C

N=HC

[156]

CH=N

O (CH2) O n

n=3-6

CH=N

[161, 170] N=HC

R= H, CH3

Liquid crystallinity in polymers â&#x20AC;&#x201C; Liquid crystalline epoxy resins

141

Table 1. Continued [167] N=N

CH=N

N=HC

[169]

CH=N N=HC

N=HC

O O C O (CH2) O C n

[171] CH=N

n=3-8,10,12

network. By convention, the networks with aliphatic chains longer than 5 carbon atoms or with oxyethylene fragments are called the semi-rigid networks [9]. This classification is not, however, unequivocal, because the elasticity of a system depends also on the size and rigidity of mesogenic groups, the type of cross-linking reaction and kind of curing agents, hence on the density of cross-linking. These factors affect the concentration of mesogenic groups in the structure of the network. The majority of the known mesogenic epoxy monomers form nematic phase. This order is usually retained in the course of curing, though increase in the degree of ordering was also reported. Some monomers are capable of forming networks with smectic structure [95, 139, 140, 146, 180]. Beside the monomer structure, other factors have also an influence on the type of order emerging in the product. These are the temperature of the process (some reactions are exothermic), type of curing reaction and accompanying conditions, particularly the presence of external orienting force field (magnetic, electric) or surface properties of a mold (rubbing). Cross-linking of mesogenic systems follows the same reaction paths as the curing of typical commercial epoxy resins. In the classical epoxy resins there are two kinds of functional groups which can take part in reactions leading to cross-linking: epoxy groups and secondary hydroxyl groups. The curing is usually carried out by introducing curing agents (hardeners) which become embedded into the network structure by means of polyaddition reactions. This role is played first of all by primary and secondary aliphatic amines, aromatic amines and dicarboxylic or multicarboxylic acids. Diphenols, polyphenols and polymercaptans were also used [181]. Functionality of hardeners affects the degree of cross-linking in the networks. It is also possible to use, as hardeners, compounds not containing active hydrogen atoms: Lewis acids (e. g. boron fluoride) or Lewis bases (e. g. tertiary amines). They act as initiators of cationic or anionic polymerization, respectively, but are not embedded in the network structure or used up during

142

Beata Mossety-Leszczak & Magdalena WĹ&#x201A;odarska

the process. Acid anhydrides react with hydroxyl groups thus increasing the cross-linking density [181]. Similar methods are used for curing mesogenic epoxies, although the selected curing agent should ensure preserving the special nature of liquid crystalline systems. While selecting a hardener, one should bear in mind that it dilutes the polymer network particularly when it becomes a part of the network structure and may reduce Tg of the network. This is why aromatic hardeners are usually preferred. The majority of mesogenic epoxy monomers have a high phase transition temperature, which also determines the choice of appropriate hardeners. Curing agents should have rather a moderate reactivity. The reason is that conversion degree after heating the system up to the curing temperature should be low enough to enable spontaneous or external field enforced ordering of monomer molecules. Then, the curing reaction, usually exothermic, should not lead to heating up the system above the temperature of isotropisation or decomposition. The methods of producing rigid epoxy LCPN described in the literature encompass almost entirely curing with hardeners reacting at high temperatures (aromatic amines, phthalic anhydride and trimellitic anhydride), whereas curing with dicarboxylic acids yields products of low cross-link density and the properties typical for elastomers [82-85, 179] or systems resembling the main chain liquid crystalline oligoesters [182]. Cross-linked polymers were also obtained by anionic polymerization, using dimethyl dibenzyl ammonium chloride as initiator, or by thermal polymerization at high temperatures [136]. Another initiator used in anionic polymerization was 4-(N,N-dimethylamine) pyridyne [167, 169, 170, 175-177, 179]. In papers [169, 170] it was also reported that such a method of curing liquid crystalline epoxy precursors produces networks with anisotropic properties only when the precursor contains sufficiently long aliphatic spacers. Cationic photopolymerization of epoxy groups was also reported [132, 141, 151]. It provides a good control of the process and facilitates enforced orientation of mesogens (the molecules are oriented before polymerization). In ionic polymerizations, small quantities of initiators are used, usually not disturbing formation of the mesophase. In the majority of the studies on curing of mesogenic epoxy monomers, primary aromatic amines were used [84, 112, 126-130, 133, 134, 137-140, 142-146, 148, 149, 152-172, 174-179]. They are rigid and have lower reactivity compared to aliphatic amines. The amines are built into the network structure and have to be added in stoichiometric amounts with respect to the epoxy groups. This is a significant mass contribution which can disturb the liquid crystalline order. Often, the system composed of a monomer and amine does not attain liquid crystalline state at all upon

Liquid crystallinity in polymers â&#x20AC;&#x201C; Liquid crystalline epoxy resins

143

heating, but in the course of the reaction one can observe, e. g. by polarized optical microscopy, the appearance of textures characteristic for mesophase. Tetrafunctional amines are most often used with two equivalent primary amino groups. Application of amines with two amine groups of different reactivity favors a change of order (e.g. a nematic monomer can yield a network with smectic order alignment [145]). The most frequently used aromatic amines used as curing agents are shown in Table 2. Anhydrides of dicarboxylic acids were used less often [130, 142]. By changing the curing conditions, the network structure and size of domains could be controlled. It is generally assumed that curing of liquid crystalline systems should be carried out within the temperature range of the mesophase. One must take into account here, that increasing the temperature within the mesophase range results in reduction of order. In many cases, at the beginning of the process, liquid crystalline phase may not appear, of all. This happens when stoichiometric amount of the hardener is added. However, during the reaction, the liquid crystalline structures eventually reappear. Sometimes, it is assumed that formation of a cross-linked structure encompasses three stages: propagation of linear chains, formation of branched structures and network formation (gelation). Upon heating, this process commences practically just after melting and it is a complex dynamic system, also depending on the homogeneity of the sample and individual properties of the components. Usually, the sample needs a post-curing at elevated temperature in order to increase the conversion of functional groups and reach the desired cross-linking density. This is typical for all types of epoxy resins. In photochemically induced crosslinking, the temperature of the process can be changed within the range of mesophase existence. The temperature applied is limited by the thermal stability of photoinitiator. Samples cured without a force field yields mainly systems with polydomain liquid crystalline structure [158]. In order to take advantage of the properties resulting from anisotropy caused by the presence of mesogenic groups, the curing process should preferably be carried out by placing the sample in a magnetic or electric field, or on a specially prepared substrate. The curing process and formation of a rigid, solid structure preserves that order. Mechanical orientation, achieved in special cells with rubbing, also provides oriented materials, but of limited small thickness, only [141]. Lee and his group demonstrated that curing of liquid crystalline epoxy resin on carbon fiber induced an anisotropic alignment in the cured product [146]. The degree of order can be further improved by applying additional electric field [183]. Curing in a magnetic field was also carried out. Fields of strength varied from 1.45 to 13.5 T were used [133, 134, 136, 150, 157, 168, 174, 180, 184]. The magnetic field ensures orientation of mesogens in the entire sample

144

Beata Mossety-Leszczak & Magdalena WĹ&#x201A;odarska

Table 2. Amines used as curing agents of liquid crystalline epoxy precursors. Structure H2N

CH3 NH2

H2N

NH2

References [84, 112, 126, 127, 167-171, 175, 177] [112, 133, 134, 137, 150]

Structure CH3 H2N

References [126, 129]

CH3 N=C

C=N

NH2

H2N

O S O

NH2

H2N

CH2

NH2

[126, 128]

H2N NH2

[126, 165]

O C O

H2N

NH2 NH2

[138, 154, 175]

C2H 5

[145, 168, 172]

O S NH2 O

H2N

[157] H2N

C2H5

NH2

[158]

H2N

H 2N

148, 161,

(CH2)2

NH2

[162] CH 3

[126, 130, 142, 143, 146, 148, 149, 154, 155, 159, 161, 166, 168] [134, 137, 138, 140, 148, 149, 152154, 156, 159-161, 164, 166, 168, 174176, 178, 179] [137, 143-146]

H2N

NH2

[159, 166]

161,

163,

NH2

H2N (CH2)6 NH2

[168]

CH3 CH3

[167, 169, 170, 175-177, 179]

and yields systems with order parameter of around 0.8-0.9. The system cured in magnetic field has a monodomain structure which is arranged along the applied field [157, 178]. These systems have excellent thermal conductivity in comparison with other polymer materials [157]. The fracture toughness of the polydomain system was higher than that of the isotropic one. The bigger was the domain size and higher order parameter, the better was the fracture toughness [158]. On the other hand it was also reported that lateral substituents in mesogens can decrease some mechanical properties including tensile strength, Young modulus and elongation at break [164]. The liquid crystalline diepoxy precursors can also be used to prepare advanced composites with selected fillers. Synthesis and properties of materials prepared from liquid crystalline epoxy resins and fillers, e.g. diphenyl aluminum phosphate nanorods [176], polyhedral oligomeric silsesquioxane [179], carbon fibres and nanotubes [185, 186], organoclays [187] and polyaniline nanorods [188] were described. The polyaniline

Liquid crystallinity in polymers â&#x20AC;&#x201C; Liquid crystalline epoxy resins

145

nanorods also played a role of curing agent. It was shown that addition of this nanofiller improved thermal stability and electrical conductivity of composites. By blending liquid crystalline epoxies with anisotropic fillers, one can induce preordering of the filler particles and equip the resulting materials with unique physical (optical, electrical) properties. Properties of composites with carbon fiber prepared from a liquid crystalline epoxy resin were compared with those from bisphenol A diglycidyl ether [185]. The hardener was 2,4-diaminotoluene. The dynamic-mechanical studies showed that the composite with liquid crystalline precursor had higher Tg, superior fracture resistance and higher elastic modulus. Jang and Lee reported that addition of liquid crystalline epoxy resin to diglycidyl ether of bisphenol A caused reinforcement of the latter. Tg was also higher [189]. The reactive filler from the polyhedral oligomeric silsesquioxane group increased rigidity of the composite in glassy state and reduced Tg [179]. A small quantity (2%) of diphenyl aluminum phosphate nanorods added to a liquid crystalline epoxy matrix resulted in substantial growth of its vitrification temperature [176]. These results suggest that liquid crystalline epoxy resins can be used as matrices for advanced composites. An example of such a composite provides the material with uniform, ordered structure of nanofillers, obtained by monodomain orientation of the polymer matrix.

Summary Special properties of epoxy LCPN described here result from a combination of liquid crystalline features with properties typical for crosslinked epoxies, i.e. low thermal expansion coefficient, high durability and chemical stability, as well as very good mechanical properties. This is why the liquid crystalline epoxy resins have been extensively studied as materials for high-performance composites, electronic packaging, insulating layers, and for nonlinear optics. Their potential practical significance stems from applicability in electronics, optical, aviation and space industry. Epoxy resins are obtained in most cases from low molecular weight monomers. When melted, they have low viscosity and can be used for filling complex molds, and also as protective coatings for electronic chips. The use of low molecular weight monomers to obtain polymer networks facilitates ordering the molecules in a force field during the process of curing and manufacturing of materials having high degree of uniform order.

Acknowledgements We thank Prof. Henryk Galina for helpful discussions on the manuscript.

146

Beata Mossety-Leszczak & Magdalena Włodarska

Financial support of Structural Funds in the Operational Program Innovative Economy (IE OP) financed from the European Regional Development Fund - Project Modern material technologies in aerospace industry, No. POIG.01.01.02-00-015/08-00 is gratefully acknowledged.

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.

Ober, C.K., Jin, J.I., and Lenz, R.W. 1984, Adv. Polym. Sci., 59, 103. Dobb, M.G. and McIntyre, J.E. 1984, Adv. Polym. Sci., 60/61, 61. Finkelmann, H. and Rehage, G. 1984, Adv. Polym. Sci., 60/61, 99. Shibaev, V.P. and Platé N.A. 1984, Adv. Polym. Sci., 60/61, 173. McArdle, C.B. (ed.) 1989, Side Chain Liquid Crystal Polymers, Blackie and Son, Ltd., Glasgow. Ciferri, A. (ed.) 1991, Liquid Crystallinity in Polymers, Principles and Fundamental Properties, VCH Publisher, Inc., New York. Donald, A.M. and Windle, A.H. 1992, Liquid Crystalline Polymers, Cambridge University Press, Cambridge. Shibaev, V.P. and Lam, L. (ed.) 1994, Liquid Crystalline and Mesomorphic Polymers, Springer-Verlag, New York, Inc., New York. Chung, T.S. (ed.) 2001, Thermotropic Liquid Crystal Polymers: Thin-film Polymerization, Characterization, Blends, and Applications, Technomic Publising Company, Inc., Lancaster, PA. Demus, D., Goodby, J.W., Gray, G.W., Spiess, H.W. and Vill, V. (ed.) 1998, Handbook of Liquid Crystals Vol. 3: High Molecular Weight Liquid Crystals, Wiley-VCH, Weinheim. Brostow, W. 1990, Polymer, 31, 979. Bawden, F.C. and Pirie, N. W. 1937, Proc. R. Soc. Lond. B, 123, 274. Flory, P.J. 1956, Proc. R. Soc. Lond. A, 234, 60. Flory, P.J. 1956, Proc. R. Soc. Lond. A, 234, 73. Robinson, C. 1956, Trans. Faraday Soc., 52, 571. British Patent GB864692 1958 (Ballard, D.G.H.; Courtaulds Ltd.). British Patent GB1198081 1967 (Du Pont). British Patent GB1283064 1969 (Kwolek, S.L.; Du Pont). British Patent GB989522 1962 (Goodman, I., McIntyre, J.E. and Stimpson, J.W.; ICI Ltd.). British Patent GB993272 1962 (Goodman, I., McIntyre J.E. and Aldred, D.H.; ICI Ltd.). U.S. Patent US3637595 1969 (Cottis, S.G., Economy, J. and Nowak B.E.). Kuhfuss, H.F. and Jackson, W.I. 1976, J. Appl. Polym. Sci., Polym. Chem. Ed., 14, 2043. Roviello, A. and Sirigu A. 1975, J. Polym. Sci. Polym. Lett. Ed., 13, 455. De Gennes, P.G. 1969, Phys. Lett., A28, 725. De Gennes, P.G. 1975, C. R. Acad. Sci. Paris, 281B, 101. Finkelmann, H., Ringsdorf, H. and Wendorff, J.H. 1978, Makromol. Chem, 179, 273.

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

147

27. Finkelmann , H., Happ, M., Portugall, M. and Ringsdorf, H. 1978, Makromol. Chem, 179, 2541. 28. Shibaev, V.P., Platé, N.A. and Freidzon, Y.S. 1979, J. Polym. Sci., Polym. Chem. Ed., 17, 1655. 29. Blumstein, A. (ed.) 1978, Liquid Crystalline Order in Polymer, Academic Press, Inc., New York. 30. Chapoy, L. (ed.) 1985, Recent Advances in Liquid Crystalline Polymers, Elsevier, London. 31. Platé, N.A. and Shibaev V.P. (ed.) 1987, Comb-like Polymers and Liquid Crystals, Plenum Press, New York. 32. Nöel, C. and Navard, P. 1991, Prog. Polym. Sci., 16, 55. 33. Watanabe, J., Hayashi, M., Nakata, Y, Niori, T. and Tokita, M. 1997, Prog. Polym. Sci., 22, 1053. 34. Jin, J.I. and Kang, C.S. 1997, Prog. Polym. Sci., 22, 937. 35. Florjańczyk, Z., and Penczek, S. 1998, Polymer Chemistry Vol. III (in Polish), Oficyna Wydawnicza Politechniki Warszawskiej, Warszawa. 36. Sołtysiak, J. and Stańczyk, W. 1993, Wiad. Chem. (in Polish), 47, 217. 37. Noël, C. 1988, Makromol. Chem. Macromol. Symp., 22, 95. 38. Jin, J.I., Kang, C.S., Lee, I.H. and Yun, Y.K. 1994, Macromolecules, 27, 2664. 39. Chen, B.K., Yuan, S.Y. and Chen, J.Y. 2005, Polymer, 46, 8624. 40. Jeong, K.U., Knapp, B.S., Ge, J.J., Graham, M.J., Tu, Y., Leng, S., Xiong, H., Harris, F.W., and Cheng, S. Z. D. 2006, Polymer, 47, 3351. 41. Bagheri, M. and Rad. R.Z. 2008, React. Funct. Polym., 68, 613. 42. Guadiana, R.A., Minns, R.A., Sinta, R., Weeks, N. and Rogers, H.G. 1989, Prog. Polym. Sci., 14, 47. 43. Vogit-Martin, I.G., Simon, P., Bauer, S. and Ringsdorf, H. 1995, Macromolecules, 28, 236. 44. De Ruijter, C., Mendes, E., Boerstoel, H. and Picken S. J. 2006, Polymer, 47, 8517. 45. Penczek, P., Frisch, K., Szczepaniak, B. and Rudnik, E. 1993, J. Polym. Sci. Polym., Chem. Ed., 31, 1211. 46. Lee, J.B., Kato, T., Ujiie, S., Iimura, K. and Uryu, T. 1995, Macromolecules, 28, 2165. 47. Padmavathy, T. and Srinivasan, K.S.V. 2002, J. Polym. Sci. Part A: Polym. Chem., 40, 1527. 48. Percec, V. and Yourd, R. 1989, Macromolecules, 22, 3229. 49. Ungar, G., Percec, V. and Zuber, M. 1992, Macromolecules, 25, 75. 50. Srinivasa Rao, V. and Samui, A. B. 2009, J. Polym. Sci. Part A: Polym. Chem., 47, 2143. 51. Roviello, A. and Sirigu, A. 1978, Eur. Polym. J., 15, 423. 52. Baker, C.D., Beylen, M.V., Ottenburgs, R. and Samyn, C. 1995, Macromol. Chem. Phys., 196, 1495. 53. Lam, J.Y.W., Kong, X.X., Dong, Y.P. and Tang, B. Z. 2000, Macromolecules, 33, 5027. 54. Lam, J.Y.W. and Tang, B.Z. 2005, Acc. Chem. Res., 38, 745.

148

Beata Mossety-Leszczak & Magdalena Włodarska

55. Kricheldorf, H.R. and Linzer, V. 1995, Polymer, 36, 1893. 56. Ganicz, T., Stańczyk, W.A., Bialecka-Florjańczyk, E. and Śledzińska, I. 1999, Polymer, 40, 4733. 57. Maganini, P.L., Andruzzi, F. and Bennetti, G.F. 1980, Macromolecules, 13, 12. 58. Blumstein, A. (ed.) 1985, Polymeric Liquid Crystals, Plenum Press, New York. 59. Collyer, A.A. (ed.) 1992, Liquid Crystal Polymers: From Structure to Applications, Elsevier Applied Science, London. 60. Tsukruk, V.V. and Bliznyuk, V.N. 1997, Prog. Polym. Sci., 22, 1089. 61. Han, X., Shanks, R.A. and Pavel, D. 2005, Europ. Polym. J., 41, 984. 62. Bagheri, M. and Pourmoazzen, Z. 2008, React. Funct. Polym., 68, 507. 63. Itahara, T. and Inadome, J. 2009, React. Funct. Polym., 69, 851. 64. Warner, M. and Terentjev, E. M. 2003, Liquid Crystal Elastomers, Oxford University Press, Oxford. 65. Terentjev, E.M. 1999, J. Phys. Condens. Matter., 11, R239. 66. Mayer, S. and Zentel, R. 2002, Curr. Opin. Soild. State and Mater. Sci., 6, 545. 67. Xie, P. and Zhang, R. B. 2005, J. Mater. Chem., 15, 2529. 68. Brand, H. R., Pleiner, H. and Martinoty, P. 2006, Soft Matter, 2, 182. 69. Urayama, K. 2007, Macromolecules, 40, 2277. 70. Bispo, M., Guillon, D., Donnio, B. and Finkelmann H. 2008, Macromolecules, 41, 3098. 71. Ohm, C., Brehmer, M. and Zentel, R. 2010, Adv. Mater., 22, 3366. 72. Finkelmann, H., Kock, H.J. and Rehage, G. 1981, Makromol. Chem., Rapid Commun., 2, 317. 73. Rousseau, I.A. and Mather, P.T. 2003, J. Am. Chem. Soc., 125, 15300. 74. Zentel, R. and Reckert, G. 1986, Makromol. Chem., 187, 1915. 75. Schätzle, J. and Finkelmann, H. 1987, Mol. Cryst., Liq. Cryst., 142, 85. 76. Zentel, R. 1989, Angew. Chem., Adv. Mater., 101, 1437. 77. Degert, C., Richard, H. and Mauzac, M. 1992, Mol. Cryst., Liq. Cryst., 214, 179. 78. Davis, F.J. and Mitchell, G.R. 1996, Polymer, 37, 1345. 79. Brown, D. H., Das, V., Alle, R.W.K. and Styring, P. 2007, Mater. Chem. Phys., 104, 488. 80. Bualek, S. and Zentel, R. 1988, Makromol. Chem., 189, 791. 81. Zentel, R., Schmidt, G.F., Meyer, J. and Benalia, M. 1987, Liq. Cryst., 2, 651. 82. Giamberini, M., Amendola, E. and Carfagna, C. 1995, Makromol. Chem., Rapid Commun., 16, 97. 83. Giamberini, M., Ambrogi, V., Cerruti, P. and Carfagna, C. 2006, Polymer, 47, 4490. 84. Espinosa, M. A., Cádiz,V. and Galiá, M. 2001, J. Polym. Sci. Part A: Polym. Chem., 39, 2847. 85. Ribera, D., Giamberini, M., Serra, A. and Mantecón, A. 2006, J. Polym. Sci. Part A: Polym. Chem., 44, 6270. 86. Shenoy, D.K., Thomsen, D.L., Keller, P. and Ratna, B.R. 2003, J. Phys. Chem. B, 107, 13755. 87. Barclay, G.G. and Ober, C.K. 1993, Prog. Polym. Sci., 18, 899.

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

149

88. Herz, J., Reiss-Husson, F., Rempp, P. and Luzzati, V. 1963, J. Polym. Sci., Part C: Polym. Symp., 4, 1275. 89. Blumstein, A., Blumstein, R.B. and Vanderspurt T.H. 1969, J. Colloid Interface Sci., 31, 236. 90. Strzelecki, L. and Liebert, L. 1973, Bull. Soc. Chim. Fr., 597. 91. Bouligand, Y., Cladis, P.E., Liebert, L. and Strzelecki L. 1974, Mol. Cryst. Liq. Cryst., 25, 233. 92. Clough, S.B., Blumstein, A. and Hsu, E.C. 1976, Macromolecules, 9, 123. 93. Warner, M., Gelling, K.P. and Vilgis, T.A. 1988, J. Chem. Phys., 88, 4008. 94. Edwards, S.F. 1992, Polym. Prepr. (Am. Chem., Soc., Div. Polym. Chem.) 33, 150. 95. Shiota, A. and Ober, C.K. 1997, Prog. Polym. Sci., 22, 975. 96. Hu, J.S., Zhang, B.Y., Yao, D.S., Xiao, W.Q. and Dong, Y.L. 2005, J. Polym. Sci. Part A: Polym. Chem., 43, 4478. 97. Douglas, E.P. 2006, J. Macromol. Sci. Part C: Polym. Rev., 46, 127. 98. Gamberini. M., Amendola. E. and Carfagna, C. 1999, Polym. Engn. Sci., 39, 534. 99. Barclay, G.G., Ober, C.K., Papathomas, K.I. and Wang, D.W. 1992, Macromolecules, 25, 2947. 100. Litt, M.H., Whang, W.T., Yen, K.T. and Qian, X.J. 1993, J. Polym. Sci. Part A: Polym. Chem., 31, 183. 101. Hikmet, R.A.M., Lub, J. and Tol, A.J.W. 1995, Macromolecules, 28, 3313. 102. Broer, D.J., Mol, G.N. and Challa, G. 1991, Makromol. Chem., 192, 59. 103. Arslanov, V.V. and Nikolajeva, V.I. 1984, Vysokomol. Soed., 189B, 185. 104. Broer, D.J., Hikmet, R.A.M. and Challa, G. 1989, Makromol. Chem., 190, 3201. 105. Hikmet, R.A.M., Lub, J. and Broer, D.J. 1991, Adv. Mater., 3, 392. 106. Hoyle, C.E., Watanabe, T. and Whitehead, J.B. 1995, Macromolecules, 27, 6581. 107. Hsu, Ch.S. and Chen, H.L. 1999, J. Polym. Sci. Part A: Polym. Chem., 37, 3929. 108. Broer, D.J., Lub, J., Van Nostrum, C.F. and Wienk, M.M. 1998, Recent Res. Devel. in Polym. Sci., 2, 313. 109. Körner, H. and Ober, C.K. 1995, Polym. Mater. Sci. Eng. Prepr., 73, 456. 110. Mormann, W. and Zimmermann, J. 1995, Macromol. Symp., 93, 97. 111. Wang, Y.H., Hong, Y.L., Yang, F.S. and Hong, J.L. 1994, Polym. Mater. Sci. Eng. Prepr., 71, 678. 112. Mormann, W., Kuckertz, C. and Bröcher, M. 2010, Macromol. Symp., 290, 70. 113. Hoyt, A.E. and Benicewicz, B.C. 1990, J. Polym. Sci. Part A: Polym. Chem., 28, 3403. 114. Hoyt, A.E. and Benicewicz, B.C. 1990, J. Polym. Sci. Part A: Polym. Chem., 28, 3417. 115. Kallal-Bartolomeo, K., Milano, J.C., Vernet, J.L., and Gallot, B. 2000, Macromol. Chem. Phys., 201, 2276. 116. Ahn, Y.H., Jung, M.S. and Chang, J.H. 2010, Macromol. Chem. Phys., 123, 177. 117. Hikmet, R.A.M., Lub, J. and Higgins, J.A. 1993, Polymer, 34, 1736. 118. Lub, J., Omenat, A., Ruiz, M. A. and Artal, M.C. 1997, Mol. Cryst. Liq. Cryst., 307, 111. 119. Melissaris, A.P. and Litt, M.H. 1994, Macromolecules, 27, 2675.

150

Beata Mossety-Leszczak & Magdalena Włodarska

120. Melissaris, A.P., Sutter, J.K., Litt, M.H., Scheiman, D.A. and Scheiman, M. 1995, Macromolecules, 28, 860. 121. Dougals, E.P., Langlois, D.A. and Benicewicz, B.C. 1994, Chem. Mater., 6, 1925. 122. Langlois, D.A., Benicewicz, B.C., and Dougals, E.P. 1998, Chem. Mater., 10, 3393. 123. European Patent EP0379057 1990 (Earls, J.D., Hefner, R.E. Jr. and Puckett, P.M.; Dow Chemical CO). 124. European Patent EP0445401 1990 (Kirchmeyer, S., Müller, H.P. and Karbach A.; Bayer AG). 125. European Patent EP0475238 1991 (Hefner, R.E. Jr. and Earls, J.D; Dow Chemical CO). 126. Giamberini M., Amendola, E. and Carfagna, C. 1995, Mol. Cryst. Liq. Cryst., 266, 9. 127. Carfagna, C., Amendola, E. and Giamberini, M. 1993, Liq. Cryst., 13, 571. 128. Carfagna, C., Amendola, E. and Giamberini, M. 1994, J. Mater. Sci. Lett., 13, 126. 129. Carfagna, C., Amendola, E. and Giamberini, M. 1994, Macromol. Chem. Phys., 195, 279. 130. Carfagna, C., Amendola, E. and Giamberini, M. 1994, Macromol. Chem. Phys., 195, 2307. 131. Carfagna, C., Amendola, E. and Giamberini, M. 1997, Prog. Polym. Sci, 22, 1607. 132. Jahromi, S., Lub, J. and Mol, G.N. 1994, Polymer, 35, 622. 133. Jahromi, S. 1994, Macromolecules, 27, 2804. 134. Jahromi, S., Kuipers, W.A.G., Norder, B. and Mijs, W.J. 1995, Macromolecules, 28, 2201. 135. Rozenberg, B.A. and Gur’eva, L.L. 1992, in Synthesis, Characterization, and Theory of Polymeric Networks and Gels; Aharony, S.M. (ed.), Plenum Press, New York, 147. 136. Rozenberg, B.A. and Gur’eva, L.L. 1996, in Liquid-Crystalline Polymer Systems, Technological Advances; Isayev, A.I, Kyu, T. and Cheng, S.Z.D. (ed.), ACS, Washington, DC, 372. 137. Mormann, W. and Bröcher, M. 1998, Macromol. Chem. Phys., 199, 853. 138. Mormann, W. and Bröcher, M. 1998, Macromol. Chem. Phys., 199, 1953. 139. Mallon, J.J. and Adams, P.M. 1993, J. Polym. Sci.: Part A: Polym. Chem., 31, 2249. 140. Barclay, G.G., Ober, C.K., Papathomas, K.I. and Wang, D.W. 1992, J. Polym. Sci.: Part A: Polym. Chem., 30, 1831. 141. Broer, D.J., Lub, J. and Mol, G.N. 1993, Macromolecules, 26, 1244. 142. Su, W-F.A. 1993, J. Polym. Sci.: Part A: Polym. Chem., 31, 3251. 143. Lee, J.Y., Jang, J., Hwang, S.S., Hong, S.M. and Kim, K.U. 1998, Polymer, 39, 6121. 144. Lee, J.Y., Jang, J., Hong, S.M., Hwang, S.S. and Kim, K.U. 1999, Polymer, 40, 3197. 145. Lee, J. Y. 2006, Polym. Bull., 57, 983.

Liquid crystallinity in polymers – Liquid crystalline epoxy resins

151

146. Lee, J. Y. and Jang, J. 2007, Polym. Bull., 9, 261. 147. Liu, J., Wang, Ch., Cambell, G.A., Earls, J.D. and Priester, R.D. Jr. 1997, J. Polym. Sci.: Part A: Polym. Chem., 35, 1105. 148. Choi, E.J., Ahn, H.K., Lee, J.K. and Jin, J.I. 2000, Polymer, 41, 7617. 149. Farren, C., Akatsuka, M., Takezawa, Y. and Itoh, Y. 2001, Polymer, 42, 1507. 150. Tan, Ch., Sun, H., Fung, B. and Grady, B.P. 2000, Macromolecules, 33, 6249. 151. Wöhrle, D., Hartwig, A., Schnurpfeil, G., Harder, A. and Schröder, H. 2000, Polym. Adv. Technol., 11, 739. 152. Ochi, M. and Takashima, H. 2001, Polymer, 42, 2379. 153. Strehmel, V. 1997, J. Polym. Sci.: Part A: Polym. Chem., 35, 2653. 154. Choi, E.J., Seo, J.C., Bae, H. and Lee, J.K. 2004, Europ. Polym. J., 40, 259. 155. Cai, Z.Q., Sun, J., Zhou, Q. and Xu, J. 2007, J. Polym. Sci.: Part A: Polym. Chem., 45, 727. 156. Ochi, M., Hori, D., Watanabe, Y., Takashima, H. and Harada, M. 2004, J. Appl. Polym. Sci., 92, 3721. 157. Harada, M., Ochi, M., Tobita, M., Kimura, T., Ishigaki, T., Shimoyama, N. and Aoki, H. 2004, J. Polym. Sci.: Part B: Polym. Phys., 42, 758. 158. Harada, M., Sumitomo, K., Nishimoto, Y. and Ochi, M. 2009, J. Polym. Sci.: Part B: Polym. Phys., 47, 156. 159. Gao, J., Hou, G., Wang, Y., Li, H. and Liu, Y. 2006, Polym. Plast. Tech. Engn., 45, 947. 160. Liu, G., Gao, J., Song, L., Hou, W. and Zhang, L. 2006, Macromol. Chem. Phys., 207, 2222. 161. Gao, Z., Yu, Y., Xu, Y. and Li, S. 2007, J. Appl. Polym. Sci., 105, 1861. 162. Liu, G., Zkou, B., Zhao, D., Li, Q. and Gao, J. 2008, Macromol. Chem. Phys., 209, 1160. 163. Huo, L., Gao, J., Du, Y. and Chai, Z. 2008, J. Appl. Polym. Sci., 110, 3671. 164. Zheng, Y., Lu, M., Ren, S., Liang, L. and Lan, Y. 2007, J. Polym. Sci.: Part B: Polym. Phys., 45, 2835. 165. Mititelu, A. and Cascaval, N.C. 2005, Polym. Plast. Tech. Engn., 44, 151. 166. Balamurugan, R. and Kannan, P. 2010, J. Mater. Sci., 45, 1321. 167. Castell, P., Galiá, M. and Serra, A. 2001, Macromol. Chem. Phys., 202, 1649. 168. Castell, P., Serra, A., Galiá, M., Giamberini, M. and Carfagna, C. 2003, J. Polym. Sci.: Part A: Polym. Chem., 41, 1. 169. Castell, P., Serra, A. and Galiá, M. 2003, J. Polym. Sci.: Part A: Polym. Chem., 41, 1536. 170. Castell, P., Serra, A. and Galiá, M. 2004, J. Polym. Sci.: Part A: Polym. Chem., 42, 3631. 171. Ribera, D., Mantecón, Aand , Serra, A. 2001, Macromol. Chem. Phys., 202, 1658. 172. Cho, S. and Douglas, E.P. 2002, Macromolecules, 35, 4550. 173. Włodarska, M., Mossety-Leszczak, B., Galina, H., Bąk, G.W. and Pakuła, T. 2004, Liq. Cryst., 31, 525. 174. Mossety-Leszczak, B., Włodarska, M., Galina, H., Bąk, G.W. and Pakuła, T. 2005, Macromol. Symp., 227, 149.

152

Beata Mossety-Leszczak & Magdalena Włodarska

175. Galina, H. and Mossety-Leszczak, B. 2007, J. Appl. Polym. Sci., 105, 224. 176. Mossety-Leszczak, B., Galina, H., Włodarska, M., Kowalik, M., Łokaj, K. and Florjańczyk, Z. 2009, Polimery (in Polish), 54, 719. 177. Włodarska, M., Mossety-Leszczak, B., Bąk, G.W., Galina, H. and Ledzion, R. 2010, Opto-Electron. Rev., 17, 105. 178. Włodarska, M., Bąk, G.W., Mossety-Leszczak, B. and, Galina H. 2009, J. Mat. Pro. Tec., 209, 1662. 179. Mossety-Leszczak, B., Galina, H., Włodarska, M., Szeluga, U. and Maciejewski, H. 2010, Macromol. Symp., 291/292, 127. 180. Benicewicz, B.C., Smith, M.E., Earls, J.D., Priester, R.D. Jr., Setz, S.M., Duran, R.S. and Douglas E.P. 1998, Macromolecules, 31, 4730. 181. Ellis, B. (ed.) 1993, Chemistry and Technology of Epoxy Resins, Chapman & Hall, London. 182. Hirn, B., Carfagna, C. and Lanzetta, R. 1996, J. Mater. Chem., 6, 1473. 183. Amendola, E., Carfagna, C., Giamberini, M. and Komitov, L. 1996, Liq. Cryst., 21, 317. 184. Barclay, G.G., McNamee, S.G., Ober, C.K., Papathomas, K.I. and Wang, D.W. 1992, J. Polym. Sci.: Part A: Polym. Chem., 30, 1845. 185. Carfagna, C., Acierno, D., Di Palma, V., Amendola, E. and Giamberini M. 2000, Macromol. Chem. Phys., 201, 2639. 186. Bae, J., Jang, J. and Yoon, S.H. 2002, Macromol. Chem. Phys., 203, 2196. 187. Shen, M.M., Lu, M.G., Chen, Y.L. and Ha, C.Y. 2005, Polym. Int., 54, 1163. 188. Jang, J., Bae, J. and Lee, K. 2005, Polymer, 46, 3677. 189. Jang, J. and Lee, J.Y. 2007, J. Appl. Polym. Sci., 106, 2198.

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 153-189 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

6. New strategy in development of liquid crystal photoaligning materials with reactive C=C bonds Lyudmyla Vretik1, Oleg Yaroshchuk2, Valentyna Zagnii1, Vasyl Kyrychenko1 and Volodymyr Syromyatnikov1 1

Department of Macromolecular Chemistry, Kyiv National Taras Shevchenko University Volodymyrsâ&#x20AC;&#x2122;ka St. 60, 01033 Kyiv, Ukraine; 2Department of Physics of Crystals Institute of Physics, NASU, prospect Nauky 46, 03680 Kyiv, Ukraine

Abstract. We review our recent approaches in development of polymers for liquid crystal (LC) photoalignment. The designed polymers contain side photosensitive chains with aromatic core and terminal tetrahydrophthalic/maleimide/methacrylamide groups. Under the action of actinic UV light these materials undergo Fries rearrangement and crosslinking due to photoaddition/ photopolymeryzation that in case of polarized light illumination results in highly efficient LC alignment. It is shown that Fries rearrangement causes pronounced photoalignment effect, while the photocrosslinking determines the alignment stability.

Introduction Liquid crystal (LC) alignment is one of the key issues of LC displays and LC photonic devices in general. For operation of these devices in a certain 111 Correspondence/Reprint request: Dr. Oleg Yaroshchuk, Department of Physics of Crystals, Institute of Physics, NASU, prospect Nauki, 46, 03680 Kyiv, Ukraine. E-mail: o.yaroshchuk@gmail.com

154

Lyudmyla Vretik et al.

mode, LC alignment should be thoroughly optimized; it should demonstrate adjusted pretilt angle and anchoring energy, high macroscopic and microscopic uniformity, and outstanding stability. Quite often, for operation in multidomain modes, the alignment parameter should be spatially modulated. LC alignment is commonly achieved by a proper treatment of substrates confining LC layer. This treatment provides surface anisotropy of the substrates and, as result, anisotropic LC-surface interaction. Due to long-range orientational interaction in the liquid crystal phase, LC alignment given by the surface extends into the liquid crystal bulk on a macroscopic scale. It allows one to obtain liquid “single-crystals” in cells with a thickness varied from a few to hundreds of microns. The conventional treatment procedure of the aligning substrates is rubbing. The substrates unidirectionally rubbed provide tilted LC alignment with a pretilt angle at the substrate mainly determined by the chemical composition of the aligning material. This procedure is quite reliable, provides strong anchoring and reproducible pretilt angle, as well as high thermal stability of LC alignment. At the same time, the rubbing method has number of intrinsic drawbacks, which quite often seriously hamper its application in an industry. The major problem is a direct mechanical contact with the aligning surface resulting in formation of scratches and large static charges causing surface dusting, cross-track shorts, and failure of thin film transistors [1]. The inherent problems of rubbing alignment technique stimulated development of alternative alignment methods capable to eliminate the rubbing problems and compete with it in industry. Among these alternative methods, the photoalignment technique [2-5] shows especial promise, because of very favorable set of properties. In contrast to rubbing, this technique provides soft treatment of the aligning surfaces. Avoiding mechanical contact with the substrate, it minimizes mechanical damage and electric charging, provides excellent alignment uniformity, and an easy way for controlling alignment parameters. It is irreplaceable in a number of new LC devices, in which the LC alignment should be induced in closed volumes, on non-flat surfaces, and on the surfaces of microscopic scale used, for example, in optical communication devices. However, having these advantages, photoalignment technique suffers from other problems such as insufficient alignment stability, relatively weak anchoring, and pronounced image sticking. Despite some new approaches, such as use of deep UV irradiation [6], optical “rubbing” [7], and passivation of photoaligning layers with the thin layers of reactive mesogens [8], the

New photoaligning materials with reactive C=C bonds

155

main efforts to improve the alignment properties are focused on finding new materials suiting better for application in photoalignment. This paper describes some advances in this direction. Although it is mainly based on results of our researches, it also gives a brief review of other recent achievements in this field. The paper is organized as follows. At first we give brief introductory section, which covers principles and challenges of photoalignment technique and the idea of the present research. Section 2 gives the experimental details, and sections 3-6, the main sections of this work, describe newly synthesized polymers and their photoaligning characteristics. Finally, section 7 presents concluding remarks.

1. Background information Despite of nice books and reviews on photoalignment [1, 9-11], we decided to give concise introduction to this phenomenon with making focus on the challenges of the state-of-the-art photoalignment materials and modern trends in their improvement. Also, we give minimal information about aligning parameters discussed in the next sections.

1.1. LC alignment characterization Usually, preferable direction of LC alignment (LC director, d) near the aligning substrate, ds, is characterized by azimutal angle φ and polar angle θ, usually called pretilt angle (Fig. 1). Depending on value of pretilt angle, planar (θ =0), homeotropic (θ =90o) and tilted (0o<θ<90o) alignment is distinguished. The alignment direction ds is determined by a balance of anchoring and elastic forces, as well as external forces applied to the LC layer. For this reason, ds does not necessarily coincide with the direction set by the alignment process (easy axis direction, l). The deviation of ds from l determines degree of orientational anchoring of LC with the aligning substrate. If this deviation is not so strong, orientational part of surface free energy of LC, Fs , can be presented in the Rapini-Papoular form

rr 1 Fs = W (n l ) , 2

(1)

where W is a parameter with dimension of energy. This prameter is called anchoring coefficient or, simply, anchoring energy. It can be interpreted as energy per unit area needed to deviate LC director from the easy direction l.

156

Lyudmyla Vretik et al.

Figure 1. Characterization of LC alignment near the aligning surface.

It can also be considered as a surface elastic constant, material constant depending on the properties of aligning surface and LC. Considering the in-plane and the out-of-plane deviation of d (with regard to the plane of aligning substrate) one can select azimuthal and polar anchoring coefficients, Wa and Wp. These parameters determine director field in the cell, electro- and magneto-optical response, anchoring transitions and many other effects.

1.2. Photoalignment procedure Film of photoaligning material is exposed to light, which excites photochemical transformations and, in this way, generates surface anisotropy in the film. Commonly, the exciting light is linearly polarized (Fig. 2a), but LC photoalignment with unpolarized light is also described [12,13]. In the latter case, the induced anisotropy is caused by oblique exposure (Fig. 2b). In the majority of photoaligning materials easy axis of LC alignment is induced in the direction perpendicular to the light polarization. However, there are groups of materials providing LC alignment in the direction of light polarization, or in both these directions [14-17]. The conventional polarized light procedure (Fig. 2a) does not set direction of LC inclination. As a result, the induced alignment shows twofold degeneration of pretilt angle. To set the inclination direction unambiguously, the polarized light exposure (Fig. 2a) is usually combined with the oblique illumination with unpolarized light (Fig. 2b) or polarized light with perpendicular direction of polarization [18,19]. These two stages can be realized simultaneously by employing oblique illumination with partially polarized light. Alternatively to the oblique exposure, the pretilt angle direction can be unambiguously set in so named photorubbing process recently

New photoaligning materials with reactive C=C bonds

157

Figure 2. Exposure schemes used in photoalignment process. Cases (a) and (b) correspond to polarized and unpolarized illumination, respectively, ls marks in-plane projection of light induced easy axis.

developed [7]. In this process, the photoaligning layer is unidirectionally scanned during illumination with periodically modulated light resulting in the LC inclination towards the scanning direction. Illumination through photo-masks is commonly used to realize patterned alignment. The amount of mask illumination processes can be reduced by using materials allowing optical rewriting.

1.3. Photoalignment mechanisms Illumination with exciting light stimulates photochemical transformations of photosensitive fragments of photoaligning materials. Usually, these absorbing parts have elongated shape and are characterized by anisotropy of absorption. Due to these properties, illumination with polarized light results in transformation of initially isotropic orientational distribution of photosensitive fragments into the anisotropic one. Depending on the nature of photochemical transformations, this occurs by angular photoselection (angular hole burning) or angular redistribution (molecular photoreorientation) [20-22]. The non-photosensitive units of photoaligning materials can also be involved in this ordering process [23, 24]. As result, the exposed film becomes anisotropic (Weigert effect [25]). The orientational order induced in the bulk of photoaligning film and on its surface well correlate, but this correlation can be distorted by the processes of selforganization (self-assembly) occurring on the surface of the photoaligning films [26,27]. The LC molecules adjacent to the orientationally ordered photoaligning film reproduce to certain extend this order due to anisotropic interaction with

158

Lyudmyla Vretik et al.

the molecules or anisotropic fragments from the surface of this film [28]. The van der Waals, dipole-dipole, Ď&#x20AC;-Ď&#x20AC; stacking, hydrogen bonding or steric interaction may dominate depending of molecular structure of LC and photoaligning material, as well as the nature of their self-organization. The processes of anisotropic LC adsorption/desorption coupled with surface memory may also be important in the alignment mechanism. Finally, some authors support the view that the alignment direction is determined mainly by the morphological anisotropy, while the magnitude of the anchoring energy depends on the LC and aligning material interaction [29,30].

1.4. Photoalignment materials The photoaligning materials are usually classified according to the prevailing photochemistry. Initially, the photoalignment phenomenon was observed for the materials with photosensitive species undergoing trans-cis photoisomerization (Fig. 3a). This group primarily includes azocompounds: chemically [2] and physically [31] adsorbed azodyes, azodye blends and azopolymers [5]. The second group comprises materials susceptible to photodestruction, such as photooxidation, chain scission, etc. (Fig. 3b). The examples of these materials are photosensitive polyimides [32], polysilanes [33] and polystyrene [34]. Third, the most studied group includes materials undergoing photo-crosslinking of cycloaddition type. The materials containing cinnamate [3,4], coumarin [35] and chalconyl [36] chromophore undergo [2+2] cycloaddition (Fig, 3c). The [4+4] cycloaddition reaction is typical for anthracenyl chromophore [37] (Fig. 3d). These photocrosslincable moieties can be physically deposited or grafted to the substrate, covalently linked to polymer backbones or blended in polymers. Each of these types of materials has its advantages and disadvantages. The azo compounds usually give excellent LC alignment low exposure dose, which can be less than 0.05 J/cm2. However, this alignment is not sufficiently stable against heat and light, because of reversible photochemistry and orientational disordering of azo fragments. The thermal stability of the photoalignment is essentially higher for sulfuric azodyes, apparently due to strong adsorption of the molecules on the substrate. However, the photostability of these samples is not sufficient. The photodestructive polyimides commonly require high exposure dose (more than 10 J/cm2). This photochemical process creates net surface charge, which causes enhanced image sticking and display flicker. Moreover, thermal stability of photoalignment is insufficient, because the photodestructed fragments relax under annealing at high temperature.

New photoaligning materials with reactive C=C bonds

159

Figure 3. Examples of photochemical reactions leading to photoalignment: a – transcis isomerization of azobenzene fragments, b – photodecomposition of polyimide backbone, c – [2+2] cycloaddition of cinnamate fragments, d – [4+4] cycloaddition of anthracenyl fragments.

Comparing with the material discussed above, the photo-crosslinking materials demonstrate the best promise. They commonly combine rather good photosensitivity and high resistance of LC alignment to heat and light, because of irreversible photochemistry and strongly restricted molecular motions. Apparently, the aligning materials presented in the market [38, 39] belong to this class of materials.

1.5. Current trends Recently, a thirty-year effort in development of photoalignment culminated in the first industrial application; based on photoalignment effect and proprietary materials, Sharp Corp. developed and commercialized method for production of LCD panels with multidomain vertical alignment (so called UV2A technology) [40]. The merits of UV2A panels are claimed to be the lower power consumption, higher contrast ratio, faster response and lower manufacturing cost.

160

Lyudmyla Vretik et al.

The prospects of photoalignment technique for the displays with lowpretilt alignment, e.g., TN, STN, IPS LCDs, are not so clear. The earlier announced advances are still not industrialized [7,41]. Among the most urgent problems hindering application of photoalignment in this field is the material problem. The most important material issue is still alignment stability, such as photo- and thermal stability, and alignment aging understood as gradual alignment decay with a storage or operation time of LC cells. Besides, research efforts are also aimed at strengthening of sensitivity and improvement of dielectric characteristics of the photoaligning layers, to eliminate the residual DC (RDC) voltage and increasing the voltage holding ratio (VHR) of LC cells. To solve these problems, different approaches are being currently tested. For commercial reason, not all of them are reflected in public literature. According to published data, several important trends can be selected. First of all, there is a tendency to search materials bearing photosensitive groups with a basically new photochemistry. For this purpose, new organic materials [42-45], plasma coatings [46], and even biomaterials [47] are being investigated. In addition, there is a clear trend in the use of materials capable of several types of chemical transformations. Typically, these transformations play different role in photoalignment. At least one of them is rather sensitive to light polarization so that it causes efficient photoordering. The other reaction(s) serves to stabilize this ordering. For example, paper [48] reports polyamic acid - azocopolymer configuration, in which orientational order achieved by the trans-cis transformations of the azo fragments is fixed by simultaneous imidization of the polyamic acid. Similarly, in papers [49,50] azomonomer is copolymerized with cinnamate monomer. The first of them provides pronounced photoordering via trans-cis isomerization, while the second one fixes this order via photo-crosslinking of the type of [2+2] cycloaddition. The same approach is applied to improve stability of LC photoalignment on sulfuric diazodyes [51], with the difference that the attached terminal groups of the dye were capable to photopolymerization. The combination of cinnamate and alkenyl reactive groups capable, correspondingly, of [2+2] cycloaddition and polymerization, is realized in [44]. Combining these groups, authors attained essential increase of light sensitivity and anchoring energy.

1.6. Our concept The approach outlined in this article is based on several ideas. Firstly, we draw attention to a new type of photo-transformations â&#x20AC;&#x201C; Fries rearrangements, which appear in a big number of organic materials bearing

New photoaligning materials with reactive C=C bonds

161

aromatic groups. This type of photochemical reactions was first reported by Kobsa [52], and Anderson and Reese [53] upon UV irradiation of substituted phenyl acetates. Figure 4 schematically presents a currently accepted mechanism of the photo-Fries rearrangement. On the first stage, UV irradiation leads to a homolytic cleavage of the acyloxy bond resulting in a pair of radicals trapped in a solvent cage. The radical pair then diffuses away from one another (“cage escape”) leading to the phenolic products or forms an enol, which undergoes tautomerization to yield hydroxyacetophenone products. The photo-Fries rearrangement occurring in functional polymers bearing aryl ester moieties has also been intensively studied [54-57]. Guillet observed that photo-Fries active moieties in poly(phenyl acrylate) [55] and poly(naphthyl acrylate) [56] rearrange upon exposure to UV light in the range of 220-340 nm to give a polymer with pendant ortho and parahydroxyacetophenone groups. The rearrangement was confirmed by the appearance of a new band in the absorption spectrum of the polymer at 340 nm corresponding to the n→π* excitation of the hydroxyacetophenone group. Inherent change of polymer refractive index in photo-Fries rearrangement has recently led to the application of this rearrangement in developing polymers capable of adjustable refractive indices [57].

Figure 4. Mechanism of photo-Fries rearrangement.

162

Lyudmyla Vretik et al.

The angularly selective photo-Fries conversion, occurring in polarized light, results in orientational ordering, which manifests itself in optical birefringence and dichroism [58]. We demonstrate that this ordering results in LC photoalignment. Since products of Fries reaction are irreversible and photostable [59], we have good chances to improve alignment stability. Besides, since presence of reactive double bonds increasing Ď&#x20AC;-conjugation is not required, it is possible to shift molecular absorption to deep UV range. Thus, consideration of photo-Fries rearrangement as an order-generating reaction allows one to essentially expand the search field of photoaligning materials for industrial applications involving materials with totally new properties. The second idea forming the basis of these studies is to combine photoFries rearrangement with other reactions stabilizing induced orientational order. This idea is realized by introducing tetrahydrophthalic/maleimide/ methacrylamide groups capable of photo-crosslinking due to photocycloaddition or photopolymerization. It is important to note that these reactive groups are linked directly to photosensitive fragment that can fix better their orientational ordering. The classes of polymers falling under this concept are discussed in our earlier works [60-64]. The present paper reviews and systematizes these data and brings number of new results confirming effectiveness of our strategy.

2. Experimental details 2.1. Material synthesis and characterization The brief description of the synthesis of each type of materials is presented in sections 3-6. The details of these syntheses can be found in our earlier works [60-66]. Thin layer chromatography and 1H NMR spectroscopy were employed to determine the purity and the chemical structure of the obtained materials. Thin layer chromatography was performed on Merck Kieselgel plates 60-F254. 1H NMR spectra were recorded by a Varian 400 NMR spectrometer with the use of tetramethylsilane in a DMSO-d6 solvent as an internal standard. In some cases IR spectroscopy studies were additionally involved to identify and characterize structures of synthesized compounds (see part 2.3). The polymerization kinetics was monitored in dilatometry studies. The kinetic curves were obtained at the following conditions assumed as standard ones for all monomers discussed: 5 wt. % solution of monomer in DMF, presence of 1 wt. % of AIBN initiator in the monomer, inert atmosphere (argon), temperatures 60o, 70o and 80o C.

New photoaligning materials with reactive C=C bonds

163

2.2. UV spectroscopy studies These studies were conducted to elucidate mechanisms of photochemical transformations in monomers and polymers based on them. Besides, polarized UV/Vis spectra were measured to characterize photoinduced ordering in these materials. Spectra of monomers were measured for the solutions of these monomers in ethanol (concentration of monomers 10-4-10-5mol/l) with Specord UV/Vis spectrophotometer. As a source of irradiation of examined compounds we used a high-pressure Hg lamp DRS-500 (Russia). The integral light intensity was about 80 mW/cm2, and the exposure time was 15-30 min. The formation of oxyketone structures was identified with a new long-wave absorption maximum. The UV/Vis spectra of polymers were measured for both solutions and films using an S2000 diode array spectrometer from Ocean Optics. The polymers were dissolved in DMF, THF, dichloroethane or toluene. The spectra in solution were measured for the solutions with polymer concentration 10-4-10-5mol/l placed in a quartz cuvette (l=1 cm). The polymer films were prepared by spin coating 5-10 wt. % solution of chosen polymer onto slides of fused quartz. Subsequently, the films were baked at 100oC for 1 h to complete evaporation of solvent remained after coating procedure. For UV illumination we used the full emission spectrum of a high-pressure mercury lamp DRS-500. To study the photochemical changes in polymers, the samples were illuminated by a non-polarized light with an integral intensity of 105 mW/cm2. For the anisotropy induction, the polymer films were illuminated with the light linearly polarized by Glan–Thompson prism. The polarized light intensity was about 40 mW/cm2. The exposure time was varied from 5 to 120 min.

2.3. IR spectra measurements The FTIR spectra of polymer films were measured to characterize synthesized structures and clarify mechanisms of photochemical transformations. The spectra were measured in the spectral range 380–4000 cm−1 with a resolution of 2 cm−1 by using Bruker IFS-66 FTIR spectrometer. The polymer films were obtained by spin coating the polymer solutions on the KBr plates and subsequent backing at 100oC over 1 h for the completion of solvent evaporation. The films were exposed to unpolarized UV light as described in the previous subsection.

164

Lyudmyla Vretik et al.

2.4. LC cells and photoalignment tests The LC cells have been made in a standard way. The 1 wt. % solution of photoaligning polymer in DMF was spin coated on ITO covered glass plates. The films were subsequently backed at 150oC over 1 h. The photoalignment processing was realized by using high-pressure mercury lamp DRS-500, whose irradiation was linearly polarized by a Glan-Thompson prism. To impart unambiguous direction of LC pretilt, the substrates were irradiated in two steps: first with polarized UV light (15-40 mW/cm2, 5-30 min) and then with non-polarized UV light (7-100 mW/cm2, 0.5-1.0 min) by rotating the sample at 90o around its normal. In the first phase, sample was irradiated at normal light incidence, while in the second one the light was directed obliquely at the incidence angle of 45o. Typically, the substrates were irradiated through a mask opening only the central rectangular area of the substrates. This made it possible to compare the alignment in exposed and unexposed areas. Two types of LC cells were constructed. In the most common case, LC cell was made by sandwiching LC between a pair of glass/ITO slides (substrates) coated with photoalignemnt layer and irradiated as described above. To obtain a uniform director orientation across the cells, the substrates were assembled in an antiparallel fashion meaning that directions of irradiation with non-polarized light were antiparallel to each other. Cell thickness was adjusted by spacers with a diameter of 20 μm. These cells, called symmetrical cells, were used to determine the type of LC alignment (homeotropic, planar or tilted), and also to measure pretilt angle of LC. In addition to antiparallel symmetric cells, twisted symmetric cells were made, in which easy alignment directions on the opposite substrates were perpendicular and a cell thickness was 6 μm. These cells were used for electro-optic tests. To define alignment direction in the cell plane and the value of azimuthal anchoring energy, we also constructed cells consisting of rubbed polyimide substrate and a photoaligned substrate (asymmetrical cells). The easy axis of the photoalignment substrate was turned in 90o with respect to the rubbing direction of the polyimide substrate. The rubbed substrate was used as a reference one with predetermined alignment direction and strong azimuthal anchoring. The thickness of these cells was reduced to 6 μm to increase higher measurable value of azimuthal anchoring energy. The cells were filled at room temperature or in isotropic phase with various nematic LCs from Merck with positive (5CB, E7, ZLI2293, ZLI4801-000) and negative (MJ961180 and MLC6609) dielectric anisotropy. We judged the alignment quality by cell observation in a light box and

New photoaligning materials with reactive C=C bonds

165

polarizing microscope using evaluation scale with five grades: (1) excellent; (2) good (single alignment faults); (3) satisfactory (minor alignment faults); (4) bad (big number of alignment faults in a form of inversion walls, flowing patterns, etc.) and (5) no alignment. Also, similarly to [44], alignment quality of antiparallel cells was additionally evaluated by measuring transmittance oscillations of these cells placed between crossed polarizers during their rotation around the axis normal to the cell plane. The parameter

Tmax â&#x2C6;&#x2019; Tmin , Tmax

(2)

where Tmax and Tmin are maximal and minimal values of oscillating transmittance, was used as a measure of alignment quality. It will be called below the quality parameter. It describes the LC alignment quantitatively, but does not give its full assessment. Because of this, for a more complete description, we usually combine q and alignment grade according to the scale above. The pretilt angle was measured by a conventional crystal rotation method, while azimuthal anchoring coefficient was estimated by spectroscopic method described in [67].

3. Poly(1-naphthyl methacrylate) In the next four sections, we successively describe the classes of materials developed under the above concept. The order of discussion of these classes corresponds to the historical sequence of their investigation. At the same time, this order reflects milestones and logic of our research aimed at improvement of photoaligning properties. We begin by considering the aligning properties of poly(1-naphthyl methacrylate) (pNMA). The discovered good photoaligning ability of this material hinted at an important role of photo-Fries rearrangements in the photoordering process and inspired us to extensively explore the area of materials capable of Fries photoreaction.

3.1. Polymer syntheses and photochemistry The pNMA (Tg = 135oC) was obtained by simple radical polymerization of 1-naphthyl methacrylate in DMF solution with AIBN as initiator [60]. The UV/Vis spectra of pNMA in toluene solution and films were measured before and after irradiation. The pNMA spectra in the solution are presented in Fig. 5. There is evident that UV exposure leads to a decrease in

166

Lyudmyla Vretik et al.

Figure 5. UV/Vis spectra of pNMA in toluene solution: 1 – before irradiation; 2 – after irradiation (25 mW/cm2, 15 min). Inset: scheme of photo-Fries rearrangement.

the most intensive band centred at about 290 nm and associated with π→π* and n→π* transitions of naphthyl fragments, and, simultaneously, an increase in absorption in the long-wave region of the spectrum. These changes are typical for photo-Fries photorearrangement in organic materials [55,56].

3.2. LC alignment In the cells of various type based on pNMA aligning layers we observed LC alignment of rather high alignment quality (excellent/good grade, q=0.85-0.95). As an example, Fig. 6 shows the alignment of LC ZLI 4801-000 in antiparallel symmetric cell. Depending on LC and exposure dose, the induced pretilt angle was 0.5o-1.0o. The azimuthal anchoring energy showed growth with exposure dose with a clear tendency of saturation. The saturation value was (5-10)x10-6 J/m2, i.e., rather high for photoaligning materials. Unfortunately, along with high quality, LC alignment on pNMA demonstrated rather poor thermal stability. Even upon aging the cells at temperature 90oC, which is lower than the temperature of nematic-isotropic transition, and significantly lower than the glass transition temperature of pNMA, we observed gradual decay of LC alignment. The full decay was observed for about 20 min of aging. The degradation of LC alignment can not be caused by reversible photochemistry, since Fries isomers are very stable [59]. The possible reason of this can be orientational disordering of the initially photoordered naphthyl fragments and their Fries isomers at elevated temperatures. Indeed, at temperatures close to Tg the frozen potoinduced

New photoaligning materials with reactive C=C bonds

167

Figure 6. Photograph of symmetric antiparallel cell based on pNMA aligning layers and filled with nematic LC ZLI4801-000. The scheme on the right shows position of polarizer, P, analyzer, A, and alignment direction, l.

order melts, since increasing free volume of polymer provides new degrees of freedom for the polymer chains. To eliminate this disordering process and thus to stabilize LC alignment, we have developed multifunctional monomers with several active groups in the molecule; along with the groups capable of Fries rearrangements we introduced the groups with reactive C=C bonds capable of dimerization or polymerization. The key role of the latter groups was to fix orientational order induced mainly due to Fries photo-conversion. The substituted arylmethacrylates with reactive double bond in tetrahydrophthalic (section 4), maleimide (section 5) and methacrylamide group (section 6) were developed and used as starting monomers for the syntheses of advanced photoactive polymers.

4. Tetrahydrophthalimidophenyl- and naphthyl methacrylates These monomers are aryl methacrylates with an additional reactive C=C bond located in tetrahydrophthalimide fragments I- II:

168

Lyudmyla Vretik et al.

The polymers corresponding to monomers 1, 2, 3 and 4 will be denoted poly-1, poly-2, poly-3, and poly-4, respectively. In the monomers above, more active C=C bond in methacryloyl group can be polymerized by common radical process. The obtained soluble linear polymers can be further crosslinked due to photoreaction of C=C bonds in tetrahydrophthalimide fragments. Besides, under irradiation, photo-Fries reaction occurs in these polymers.

4.1. Syntheses Synthesis of the monomers was carried out in two stages. Starting hydroxyphenyl/naphthyl imides/amides were synthesized by condensation reaction of 4-aminophenole or 4-aminonaphthole and 1,2,3,6-tetrahydrophthalic anhydride/ 5-methyl-3,4,7,7-tetrahydro-phthalic anhydride at presence of sodium acetate in acetic acid as a solvent. Condensation was carried out at the temperature of boiling point of acetic acid. 5-Hydroxynaphthylamide was obtained in 1,4-dioxane solution at boiling point of 1,4-dioxane. On the second stage, monomers 1-4 were obtained by heating the appropriate initial imides/amides with excess of methacrylic anhydride and catalytic amount of H2SO4 and phenotiazine as an inhibitor of polymerization. The structures of the synthesized substances were characterized using 1H NMR, IR and UV-spectroscopy. Monomers 1-4 were polymerized by thermoinitiated radical polymerization. The polymerization was carried out in N,Ndimethylformamide (DMF) solutions (at 10 wt. % concentration of monomer) at 80oC with AIBN as initiator. It was found that during the polymerization the non-substituted tetrahydrophthalimide units (R2=I) in monomers 1, 3, and 4 readily crosslink so that linear soluble polymer products could only be obtained at low monomer conversions (~20 wt. %). In a case of monomer 2, one can reach higher yield of soluble polymer (up to ~40 wt. %) due to less activity of methyl-tetrahydrophthalimide unit. Molecular weights estimated for soluble poly-(tetrahydrophthalimidophenyland naphthyl methacrylates) by GPC (eluent chloroform:ethanol = 95:5, polystyrene standards) were in the ranges: Mn= 140 000 â&#x20AC;&#x201C; 199 000, Mw = 216 000 -266 000. As an illustrative example, detailed description and characterization of monomer 1 and its polymer poly-1 is given at the end of this section. We have further recognized a direct influence of temperature on the yield of non-crosslinked polymer products: at lower temperature (60o C) soluble polymer products were obtained with higher yields: ~25 wt. % for poly-1, and ~35 wt. % for poly-3 and poly-4). To get more details of this dependence,

New photoaligning materials with reactive C=C bonds

169

Figure 7. The kinetic curves of thermally initiated radical polymerization of monomer 4 at 60o, 70o and 80o C (a) and schematic representation of polymerization/crosslinking processes corresponding to A and B parts on the curves.

thermally initiated polymerization of the monomers was investigated with dilatometry. For illustration, Fig. 7 presents kinetic curves obtained for monomer 4. Two quasilinear parts of these curves, A and B, can be selected. The part A corresponds to linear polymer product formation, while the part B reflects formation of crosslinked products due to involvement of second C=C double bond of tetrahydrophthalimide fragment in the polymerization process. This shows that it is very important to interrupt in time the polymerization reaction to prevent formation of a large number of crosslinked products. In conclusion, it should be noted that recently we increased the polymerization yield of monomers 1 and 2 to ~40 wt. % by photopolymerization at room temperature [64]. However, this method is not universal. For example, it is ineffective for monomers 3 and 4, because of radical trapping by naphthalene fragments. Nevertheless it shows promising way towards effective synthesis procedure for this class of photoaligning polymers. Synthesis example Synthesis of monomer 1. Starting 4-hydroxyphenylimide was synthesized by condensation reaction of the equimole amounts of 4aminophenole and 1,2,3,6-tetrahydrophthalyc anhydride at the presence of the equimole amount of sodium acetate in acetic acid as a solvent. The condensation was carried out at the temperature of boiling point for acetic

170

Lyudmyla Vretik et al.

acid in glass reactor with a condenser during 5 hours. After crystallization in isopropyl alcohol, colourless crystals of 4-hydroxyphenylimide were obtained (m.p.=242o-244o C, yield of the monomer 25.9 wt. % (with respect to started 4-aminophenole)). 4-Hydroxyphenylimide was heated to 80o-90o C with 30 wt. % excess of methacrylic anhydride and catalytic amounts of H2SO4 and phenotiazine. After completion of this reaction, the resulting mixture was poured into water. Crystallization was carried out in a toluenehexane (1:1) mixture with silica gel. Colourless amorphous monomer 1 was obtained (m.p. = 137-138oC). 1H NMR identification (100 MHz, acetone-d6), ppm: 7.32 (s, 4H, Ar), 6.32 (s, 1H, =CH2), 5. 98 (m, 2H, =CH), 5.85 (s, 1H, =CH2), 3.34 (s, 2H, CH), 2.54 (d, 4H, CH2), 2.06 (s, 3H, CH3). IR identification (KBr), cm-1: broad 3080 (C-H), 3040 (C-H), 1730 (C=O), 1690 (N-C=O), 1630 (C=CH2), 1635 (HC=CH), 1600 (Aryl), 1510(Aryl), 1455 (Aryl), 1290, 1140 (C-O), 890 (=CH), other signals 820, 710. UV spectrum (ethanol, c= 2x 10-5 mol/l): λmax 1 = 220 nm, ε max 1= 2.05x 104 (l/(molxcm) and λmax 2 = 282 nm, ε max 2= 0.525x 104 (l/(molxcm). Synthesis of polymer poly-1. 0.14 g of monomer 1 and 0.0014 g of azobisisobutyronitrile (AIBN) were dissolved in 1.5 ml of DMF. The reaction vessel with the solution was subsequently closed air-tight. The solution was heated to 60oC for 60 min. Thereafter, the reaction vessel was opened and the solution was added dropwise to 20 ml of ethanol while stirring at room temperature. The separated polymer was filtered off, dried, dissolved in 1 ml of DMF and this solution was added dropwise to 10 ml of ethanol. Filtration and drying at 40oC in a vacuum gave 0.032 g (23 wt.%) of poly-1. GPC (eluent chloroform:ethanol = 95:5, polystyrene standards): Mn= 199 000, Mw = 256 000. Mw/Mn =1.3.

4.2. Photochemistry: UV/Vis spectroscopy studies Photo-Fries rearrangement in the monomers was confirmed by the appearance after UV-irradiation of a new band at around 370 nm in the absorption spectra of their ethanol solutions, both for phenylene- and naphthalene containing monomers (R1= 1,4-Ph, 1,4-Nph, 1,5-Nph).

4.3. LC alignment The photoinduced LC alignment with a quality grade “excellent” or “good” in both parallel and 90o twist configuration was recognized only for materials with 1,4-substituted aromatic core (R1= 1,4-Ph, 1,4-Nph) presented in Table 1. As well as for pNMA, easy axis of LC alignment was induced

New photoaligning materials with reactive C=C bonds

171

Table 1. Characteristics of LC alignment for the best polymers with tetrahydrophthalimide fragments.

Monomer structure

Alignment quality

Alignment quality parameter

good/excellent

0.88-0.95

good

0.85-0.90

good

0.85-0.92

O O

H3C C

C O

H2C O

O O

H3C

C C O

H2C O

O H3C

O C C O

CH3 N

H2C O

perpendicularly to the direction of light polarization E. The pretilt angle and anchoring energy values were similar to pNMA. The induced alignment was quite stable at 90oC; for LCs with the nematic-isotropic phase transition above 90oC, it was kept for 3-5 hours of our monitoring. However, the alignment degraded rather fast at 110oC, when LC was in isotropic phase. In this case, the decay time of LC alignment was only about 10 min. Thus, comparing with pNMA, this class of materials demonstrates improved thermal stability. However, the stability was still rather poor for practical use of these polymers. The lack of stability can be explained by low crosslinking rate, caused by insufficient reactivity of C=C bonds in tetrahydrophthalimide fragments.

5. Maleimidoaryl methacrylates To increase crosslinking efficiency, tetrahydrophthalimide group was replaced by another active group, maleimide group, which also contains reactive C=C bond. The maleimide and methacryloyl groups were linked through aryl spacer. The naphthyl spacer giving worse result (see 4.3) was not used in these syntheses. The general formula of the synthesized monomers is as follows:

172

Lyudmyla Vretik et al.

As in the previous case, these materials contain two C=C bonds with different reactivity. The more active C=C bond in methacryloyl group can be polymerized in vitro, in a process of polymer synthesis. In turn, the weaker C=C bond in maleimide group can be photopolymerized in situ, i.e., during photoalignment processing of the polymer film.

5.1. Syntheses Synthesis of the monomers was carried out in two stages [66], just as for monomers of previous class. The structures of the synthesized substances were characterized using 1H NMR spectroscopy. In order to obtain polymers poly5-poly10, the corresponding monomers 5-10 were polymerized at 80oC in DMF solutions at 10 wt. % concentration of monomer with AIBN as initiator. The monomers 5 and 6 were found to form polymers with crosslinked structure at low rate of monomer conversion preventing the formation of linear (soluble) polymer. For the monomer 5 soluble polymer fraction was obtained with very low yield and in only some syntheses. For the compound 6 linear polymer was reproducibly obtained at the monomer conversions less than 10 wt. %. In case of monomers 7 and 8 soluble polymer products were obtained at higher monomer conversions, 15-20 wt. %. Monomer 9 with 2,3-diphenylmaleimide fragment could be converted into soluble polymer with even higher conversion ~40 wt.%. Conversely, for monomer 10, which releases Cl during polymerization and traps free radicals, soluble polymer poly-10 was obtained at monomer conversion less than 10 wt. %. Based on results of GPC (eluent chloroform:ethanol = 95:5 or DMF, polystyrene standards) measurements for soluble poly-6-poly-10 molecular weights were estimated; Mn= 102 000 â&#x20AC;&#x201C; 461 000, Mw = 130 000 - 928 000. The example below illustrates synthesis procedure for this class of polymers. Synthesis example Synthesis of monomer 6: Starting 4-hydroxyphenylimide was synthesized by condensation reaction of the equimole amounts of 4-aminophenole and

New photoaligning materials with reactive C=C bonds

173

citraconic anhydride in acetic acid at boiling point of acetic acid in glass reactor with a condenser during 1 hour. After crystallization in isopropyl alcohol colourless crystals of 4-hydroxyphenylmaleimide were obtained (m.p.= 175oC, yield ~80 wt.%). 4-Hydroxyphenylmaleimide was heated at 90o C with double excess of methacrylic anhydride with catalytic amounts of H2SO4 and phenotiazine. After 4 hours, the resulting mixture was poured into water. Crystallization was carried out in a toluene-hexane (1:1) mixture with silica gel. Colourless amorphous monomer 6 was obtained (m.p. = 128oC, yield ~ 60 wt.%). 1H NMR identification (100 MHz, acetone-d6), ppm: 7.52-7.25 (m, 4H, Ar), 6. 68 (s, 1H, =CH), 6.33 (s, 1H, =CH2), 5.88 (s, 1H, =CH2), 2.12 (s, 3H, CH3), 2.05 (s, 3H, CH3). Synthesis of polymer poly-6: 1g of monomer 6 and 0.001g of AIBN were dissolved in 10 ml of DMF. The solution was heated at 60oC for 60 minutes in the subsequently closed air-tight vessel. Thereafter, the reaction solution was added dropwise to 20 ml of acetone while stirring at room temperature. Filtration and drying at 40oC in a vacuum gave 0.1535 g (~15 wt. %) of poly-6. GPC (eluent DMF, polystyrene standards): Mn= 461 000, Mw = 928 000. Mw/Mn =2.0.

5.2. Photochemistry: UV/Vis and IR spectroscopy studies Spectra of monomers were studied by UV spectroscopy. In this case, we measured the spectra of monomers, dissolved in ethanol before and after irradiation with UV light. After the irradiation, broad absorption band around 320-330 nm was observed, which indicated the photo-Fries rearrangement.

Figure 8. Schemes of (a) [2+2] photocycloaddition and (b) Fries rearrangement of the molecules of maleimidoaryl methacrylates.

174

Lyudmyla Vretik et al.

The long-wave absorption increase after UV irradiation was also observed in the polymer films. This indicated Fries reaction in a solid state of the polymers. The changes in polymer films under UV light were also studied by IR spectroscopy. The photo-crosslinking reaction was detected by changes in vibration bands of maleimide moieties. In particular, the band at 850 cm-1 demonstrated pronounced decrease giving evidence of [2+2] photocycloaddition occurring between two maleimide moieties. The appearance of new absorption band at 1680 cm-1 indicated formation of ketone structures as products of Fries reaction. Thus, both [2+2] photocycloaddition and Fries rearrangement reactions, theoretically possible in these polymers, actually realize under UV irradiation. Schemes of these reactions are shown in Fig. 8.

5.3. LC alignment The alignment quality results for these polymers are summarized in Table 2. It can be seen that the range of the alignment quality parameter is rather broad. One can observe some correlation between the alignment quality and the end substituents R3 and R4 in maleimide groups. For short substituents (R3,4=H, CH3) the alignment quality is better than for the longer one. Furthermore, for asymmetric maleimide structures (R3â&#x2030; R4) the alignment is better than for symmetric one (R3=R4). The former observation can be explained by steric hindrance caused by long and bulky substituents, which lowers yield of [2+2] photocycloaddition. This is indirectly confirmed by the fact that the crosslinking reaction is much more intensive in polymerization of monomers 5 and 6, having small H and CH3 substituents, than in monomer 9 with bulky aryl substituents (see section 5.1). The bulky terminal groups may also reduce yield of Fries reactions, because under conditions of limited free volume of polymers, they hamper molecular rearrangements associated with significant steric changes. The higher alignment grade in case of asymmetric maleimide structures might be explained assuming that R3 and R4 fragments influence C=C bond energy. Possibly, at R3â&#x2030; R4, this bond is weaker so that the yield of [2+2] photocycloaddition increases. This hypothesis will be tested in the future studies. It should be noted especially high alignment quality obtained for poly-6. The end substituents R3 and R4 in this polymer seem to be well optimized from the view point of photoreactivity and LC photoalignment. The excellent alignment capability of this polymer is additionally illustrated by Fig. 9. The pretilt angle on this aligning polymer falls in the range 0.2-1.0o, and azimuthal anchoring coefficient exceeds 10-5 J/m2. The alignment endured 5 h

New photoaligning materials with reactive C=C bonds

175

backing at 100oC. This altogether makes this polymer and similar structures rather promising for industrial applications. At the same time, these materials have some weak points. One of them is insufficient reproducibility of synthesis; the materials are usually obtained with different polymerization yield that influences LC alignment. The most promising structure poly-6 cannot be sufficiently modified that makes difficult further improvement of this polymer. Table 2. Characteristics of LC alignment for some polymers with maleimide fragments. Monomer structure and polymer marking

Alignment quality

Alignment quality parameter

O N

CH3

CH2

excellent/ good 0.9-0.96

poly-5 O

N H3C

excellent

0.95-0.98

good/ excellent

0.88-0.96

good

0.86-0.94

CH 2

bad/satisfact.

0.6-0.8

CH2

bad

0.4-0.6

CH2

poly-6 O H 3C N

CH3

CH2

H 3C O

poly-7 O

CH3

CH 3

CH2

poly-8 O

poly-9 O Cl N Cl O

poly-10

CH3

176

Lyudmyla Vretik et al.

Figure 9. Photographs of symmetric antiparallel cell based on poly-6 aligning layers and filled with nematic LC E7. The schemes below show position of polarizer, P, analyzer, A, and alignment direction, l.

6. Methacrylamidoaryl methacrylates Development of these materials is a culmination point of the present research. In the materials above the photoalignment treatment resulted in conversion of C=C bonds located in the bulky cyclic groups. We believed that the bulky photoproducts are less mobile than the small one, leading to an increase in thermal stability of the induced alignment. As shown above, this idea has been partially correct. However, the above discussed synthesis problems and constrains in polymer modification stimulated us to change direction of research. The cycling crosslinkable groups we decided to replace by small methacryloyl groups capable of radical polymerization. In other words, we switched to the class of polyarylmethacrylates containing nonpolymerized methacryloyl groups. The general formula of these polymers and their possible photoproducts are shown in Fig. 10. Normally, these polymers were obtained by radical polymerization of corresponding monomers, same as the polymers of other classes discussed above. The monomers were methacryloylamidoaryl-methacrylates having two methacryloyl groups of different reactivity (Fig. 11). The more active O-methacryloyl group was subjected to polymerization in a synthesis process, while the less active NH-methacryloyl group underwent crosslinking reaction due to polymerization process under UV irradiation or heating.

New photoaligning materials with reactive C=C bonds

177

Figure 10. General formula of methacrylic polymers with free methacryloyl group.

Figure 11. Structures of synthesized methacrylamidoaryl methacrylates.

In addition to direct polymerization, some of these structures were obtained by polymer-analogue reaction. The advantages and weak points of these two synthesis approaches are discussed in the next subsection.

6.1. Syntheses Synthesis of the monomers was carried out in one or two stages by acylation of 1,4-aminophenole (monomer 11), 1,4- and 1,5-aminonaphthole (monomers 12 and 13), tyramine (monomer 14) or 4-amino-phenethyl alcohol (monomer 15) with methacrylic anhydride or methacryloyl chloride. Monomer 16 was prepared in three stages: p-hydroxy benzoic acid and 1,4-

178

Lyudmyla Vretik et al.

aminophenole were acylated with methacryloyl chloride and obtained intermediates were condensed using DCC/DMAP activating reagents. The structures of the synthesized substances were characterized using 1H NMR, IR and UV-spectroscopy. Monomers 11-16 were polymerized by radical polymerization in DMF solutions (10 wt. % concentration of monomers) with AIBN as initiator at 80oC. It was found that naphthalene-containing monomers 12, 13 could be radically polymerized into soluble polymer products poly-12, poly-13 up to 70 wt. % monomer conversion. Instead, maximum yield of soluble polymers poly-11, poly-14 and poly-16 for phenylene-containing monomers 11, 14, 16 falls down to 30 wt. % of monomer conversion. Soluble polymer poly-15 could be obtained at 15 wt. % conversion of monomer 15. For all polymers obtained the molecular weight distribution is bimodal consisting of a higher molecular weight fraction (for which Mw, Mn and Mw/Mn values are derived) and a broad lower molecular weight fraction. The GPC results presented in Table 3 for poly-11, show quasilinear grows of polymer weight with the yield of polymerization. The values of Mn and Mw obtained for poly-11–poly-16 were in the ranges: Mn= 16 500–33 600, Mw =49 400–60 000. Monomers 11 and 16 were also copolymerized with monomers bearing hydrophobic tails using the same method; one of these monomers was dissolved in DMF together with methylmethacrylate, buthylmethacrylate, 2,2,3,3-tetrafluoropropylmethacrylate (MF-1) or 2,2,3,3,4,4,5,5-octafluoropenthylmethacrylate (MF-2) in appropriate molar ratios (1:1, 1:2, 2:1, etc.), doped by AIBN and kept at 80oC up to 10-30 wt. % of monomer conversion. Table 3. GPC data, polymerization yield, and alignment quality grade for poly-11. Polymer yield, wt.%

Mw/Mn

Low molecular weight fraction, wt. %

Alignment grade

29 400

12 600

2.3

excellent

50 500

28 600

1.8

good

60 000

33 600

1.9

excellent/ good

* GPC-measurements were performed using PSS-SDV columns and DMF eluent containing LiBr at a flow rate of 1.0 mL/min. The calibration curves for GPC analysis were obtained using PSS polystyrene standards (1000 – 400000 D).

New photoaligning materials with reactive C=C bonds

179

Given that selective polymerization of O-methacryloyl groups in monomers 11-16 is limited by low yield of soluble polymer product, we have also tried to explore another synthesis way; polymer materials with analogues structural elements were synthesised by polymer analogue reactions starting from poly(methacrylic acid). However, we realized that transformation of free carboxylic -COOH groups in poly(methacrylic acid) (Mn=43 400, Mw = 75 600) is limited in this reaction to 40 mol %. This means that no more then 40 mol % of photoactive fragments can be introduced. Representative of such polymers is a polymer P1 the synthesis of which is described below. Each of these synthesis methods has strengths and weaknesses. The monomer polymerization method provides maximal concentration of photosensitive fragments. However, because of similar reactivity of methacrylate and methcrylamide groups, it is hard to prepare soluble polymer with high monomer conversion yield; in fact, it is less than 30-40 wt. %. Moreover, the obtained structures are not pure; according to 1H NMR spectra, even at monomer conversion 15-20 wt. %, these polymers have free C=C bonds of both methacrylamide and methacrylate types. Also, these polymer products contain fractions with high and low molecular weight. One way to improve synthesis of these polymers, which we now develop, is the use of monomers with increased difference in reactivity of terminal groups containing C=C bond. The polymers prepared by polymer analogues reactions obviously have only one type of C=C bonds. The disadvantage of this approach is insufficient replacement of –COOH groups with photosensitive fragments (monomer conversion is ≤40 mol %). The non-reacted –COOH groups may cause undesired types of polymer-LC interaction. New approaches are needed to increase concentration of photosensitive fragments in these polymers. Synthesis examples Synthesis of monomer 11. In the solution of 54.5 g (0.5 mol) of 4-aminophenole in 250 ml of dry tetrahydrofurane 50.5 g (1 mol) of triethylamine was added. Then 104.5 g (1.1 mol) of methacryloylchloride was added dropwise to the solution at room temperature with stirring and permanent cooling. The reaction mixture was kept 3 days at room temperature and then poured into 1 l of distilled water. The obtained residue was filtered off and washed with water till neutral reaction in litmus. Crystallization in toluene mixture with silica gel gave 70.3 g of monomer 11 (m.p.=103o-105o C, yield 70.3 wt. %). 1H NMR identification (400 MHz, CDCl3), ppm: 7.73 (d, 2H, Ar), 7.0 (d, 2H, Ar), 6.34 (s, 1H, =CH2), 5.78 (m, 2H, =CH2), 5.47 (s, 1H, =CH2), 2.06 (s, 6H, CH3). UV spectrum (ethanol, c= 5x 10-5 mol/l): λmax = 260 nm, ε max= 1.15x 104 (l/molxcm).

180

Lyudmyla Vretik et al.

Synthesis of polymer poly-11. 1.4 g of monomer 11 and 0.014 g of AIBN were dissolved in 17 ml of DMF. The reaction vessel with the solution was subsequently closed air-tight and heated to 80o C for 35 minutes. Thereafter, the reaction vessel was opened and the solution was added dropwise to 150 ml of ethanol while stirring at room temperature. The separated polymer was filtered off, dried, dissolved in 10 ml of DMF, and this solution was added dropwise to 50 ml of ethanol. Filtration and drying at 40oC in a vacuum gave 0.323 g (23 wt. %) of poly-11. GPC (PS standard, DMF as mobile phase): Mn = 28 600, Mw= 50 500, Mw/Mn =1.8. Synthesis of copolymer-1. 0.91 g (0.004 mol) of monomer 11 and 1g of 2,2,3,3-tetrafluoropropylmethacrylate (0.004 mol) with 0.0191 g of AIBN were dissolved in 24 ml of DMF. The reaction vessel with the solution was subsequently closed air-tight and solution was heated to 80oC for 40 minutes. Thereafter, the reaction vessel was opened and the solution was added dropwise to 200 ml of diethyl ether. Filtration gave 0.24 g (12 wt. %) of copolymer 1. IR identification (KBr), cm-1: broad 3392 (NH), 1752 (-O-CO-), 1668 (amide I), 1508, 1253, 965, 904, 664 (aryl), other signals 2937,1408, 1392, 1170, 1156, 1128, 1016, 805, 712. Synthesis of polymer P1. 1 g (0.012 mol) of poly(methacrylic acid) in 5 ml of DMF was stirred during 24 hours. To the resulting solution were added 2.12 g (0.012 mol) of 4-methacryloylamino phenole, 2.47 g (0.012 mol) of DCC and 0.49 g of DMAP. This mixture was stirred at 20оС during 5 days, then filtered off and poured into isopropanol. The obtained residue was isolated by filtration and dried in vacuum, giving 0.8 g (28 wt. %) of polymer P1. GPC (PS standard, DMF as mobile phase): Mn=54 000, Mw=86 300, Mw/Mn =1.6.

6.2. Photochemistry: UV/Vis and IR spectroscopy studies The polymers of this series demonstrate similar changes under UV irradiation. As an example, Fig. 12 presents spectra of P1 in THF solution (a) and film (b) before irradiation and after successive steps of irradiation. It is seen that the spectra of these polymers contain only one intensive band with a maximum at 240-265 nm. This band is caused by π- conjugated side chain containing various double bonds, such as C=C (aromatic rings and terminal methacryloyl groups) and C=O (carbonyl groups). According to Fig. 12, the photoinduced spectral changes are similar for solution and film. UV irradiation leads to a gradual decay of the main absorption band and, simultaneously, the growth of the broad band in the long-wave region (300-360 nm). The former change is caused by reduction of

New photoaligning materials with reactive C=C bonds

181

Figure 12. UV/Vis spectra of P1 in (a) THF solution and (b) film for a sequence of exposure times at light intensity 80 mW/cm2. (c) Photochemical transformations of polymer molecules. The upper and lower products are formed in result of photopolymerization and photo-Fries rearrangements, respectively.

Ď&#x20AC;-conjugation in the molecule and can be caused by saturation of C=C bonds in NH-methacryloyl groups and Fries rearrangements. The latter change is typical for Fries rearrangements. Detecting changes typical for Fries photoreaction, UV/Vis spectra are unable to distinguish different Fries transformations. As demonstrated in Fig. 12 (c), two types of Fries rearrangements, in arylester Ph-O-CO and arylamide Ph-NH-CO groups, are possible in P1. Moreover, because of complexity of polymer structure, UV/Vis spectroscopy cannot unambiguously confirm the polymerization of NH-methacryloyl groups. To verify realization of each of these possible reactions, several auxiliary compounds were synthesized. These compounds had simplified structure so

182

Lyudmyla Vretik et al.

Table 4. Peak positions of IR absorption bands related to photosensitive groups of P1 [70]. Structural element Ar-NH-C=O

Ar-O-C=O C=CH2

Peak position, cm-1 Experimental Ref. [68,69] 1668

1680-1630

1528

1570-1515

1322 1750 1196 and 1166

1330-1200 1750-1735 1200 and 11501100 1680-1620 930-945

1628 930

Assignment Amide I, ν (C=O) Amide II, ν (C-N)+ δ(C-N-H) Amide III, δ(NH)+ δ(OCN) ν (C=O) ν (COC) ν (C=C) δ(C-H)

that some types of photochemical transformations possible for P1 were excluded [63]. These studies showed that all three reactions theoretically predicted for P1 are actually realized under irradiation. This conclusion was also supported by IR studies. First of all, we identified vibration bands corresponding to arylester, arylamide and methacryloyl groups assuming that these bands are sensitive to the above discussed reactions (Table 4). For the majority of these bands we observed changes after UV irradiation that gave clear evidence for all three photochemical reactions theoretically possible in this polymer [70].

6.3. Photoinduced ordering in polymer films The photoinduced orientational ordering in polymer films under irradiation with polarized light was studied by polarization UV/Vis spectroscopy and null ellipsometry [63]. For illustration, Fig. 13 presents the data obtained by polarization spectroscopy for P1. It can be seen that under irradiation the initially isotropic angular distribution of light absorption is transformed to the anisotropic one with the absorption maximum in the direction perpendicular to the direction of light polarization E (D⊥ absorption coefficient) and the absorption minimum in the E direction (D⎜⎜). This implies that after irradiation the photoinduced fragments are orientationally ordered in the direction perpendicular to light polarization. Seemingly, due to anisotropic interaction with these fragments, LC molecules are aligned in the same direction. According to Fig. 13 (b), the photoinduced ordering demonstrates saturation trend. Assuming uniaxial ordering in the saturation, the order parameter of photosensitive polymer chains can be calculated

New photoaligning materials with reactive C=C bonds

183

Figure 13. (a) Angular dependence of the optical absorption of the film of P1 at 265 nm, maximum of the main absorption band, for exposure times 0, 30, 180, 300 min (curves 1, 2, 3 and 4, respectively). E marks direction of polarization of exciting light. (b) Optical densities D⎜⎜,⊥ and dichroism ΔD= D⊥– D⎜⎜ as functions of exposure time for the film of P1. I=45 mW/cm2 (modified of ref. [63]).

as S = ( D⊥ − DII ) /( D⊥ + 2 DII ) . The obtained value is S ≈ 0.11 . The changes in the coefficients of optical absorption under irradiation correspond to angular photoselection mechanism [20-22]. The photoordering features for other polymers of this series are rather similar to those of P1.

6.4. LC alignment The quality of LC alignment with these polymers is strongly dependant on polymerization yield. At high yields (usually, higher than 70 wt. %) the polymer products are highly crosslinked and so insoluble in organic solvents used for film coating purpose. At lower yields (typically, 40-70 wt. %) the materials can be dissolved, but quality of LC alignment is rather poor. Finally, if the yield is lower than some critical value, acceptable LC alignment is detected. The critical values of polymerization yield for these polymers are presented in Table 5. One can see, that the highest yield is obtained for naphthalene containing polymers (R5=1,4-Nph, 1,5-Nph). However, these polymers give somewhat worse alignment then the homologues with phenyl core (R5=1,4-Ph, 1,4-OPh, 1,4-NHPh and 1,4-COOPh). This may be due to the lower shape anisotropy of photosensitive fragments. Finally, the alkyl tail between O or NH methacryloyl

184

Lyudmyla Vretik et al.

Table 5. Polymerization yield, solubility and characteristics of LC ZLI2293 alignment for poly(methacrylamidoaryl methacrylates). Aromatic core, R5

Yield, wt.%

Solubility

Alignment grade

LC pretilt angle

1,4-Ph 1,4-Nph 1,5-Nph 1,4-OPh 1,4-NHPh 1,4-COOPh

35 48 50 29 30 30

DMF DMF DMF DMF DMF DMF

excellent good/excellent good excellent excellent /good excellent

0.4o-1.2o 0.3o-0.8o 0o-0.6o 0o-0.5o 1.0o-2.2o 0o-0.8o

group and aromatic core (monomers with R5= 1,4-OPh, 1,4-NHPh) slightly reduces yield of soluble polymer, but gives rise to some enhancement of LC pretilt angle. Even below the critical yield, the quality of LC alignment decreases with the yield. As was noted in the synthesis section 6.1, the polymers obtained by way of monomer polymerization have bimodal distribution of molecular weight. The relative content of the latter fraction increases with the polymerization yield and, simultaneously, the quality of LC alignment decreases (Table 3). Thus, perhaps, the low molecular weight fraction plays an essential destructive role in LC alignment. For the polymers synthesized by polymer analogue reaction we obtained rather encouraging result: these polymers provide high-quality alignment at even rather low degree of attachment of photosensitive fragments. The increase of this parameter above 10 mol % mainly accelerates generation of photoalignment and modifies pretilt angle, but does not significantly affect azimuthal anchoring. This gives us hope that the industrial demands can be met at even low concentration of photsensitive fragments. For both types of polymers, synthesis conditions can be optimized for the LC alignment of excellent quality. For illustration, Fig. 14 (a) demonstrates alignment in the cell based on poly-11. One can judge the high quality of alignment in the areas with parallel and twisted director configuration. Fig. 14 (b) shows that LC uniformly switches in an electric field. The azimuthal anchoring energy grows and saturates with an exposure dose approaching 10-4 J/m2 in the saturation state. This is one of the best results claimed for photoaligning materials. The pretilt angle for homopolymers was lower then 1.5o. However, it can be dramatically increased by copolymerization of the monomers 11-16 with the monomers containing hydrophobic chains. Figs. 14 (d)-(f) demonstrate properties of the films based on 11-MF-2 series of copolymers. The increase

New photoaligning materials with reactive C=C bonds

185

in the concentration of MF-2 in the copolymer leads to an increase of its hydrophobicity. This, in turn, results in an increase in pretilt angle. As there is obvious from Fig. 14 (f), full range controlling of pretilt angle can be realized.

Figure 14. (a, b) Photographs of symmetric LC cells based on poly-11 photoaligning layers viewed between crossed polarizers. (a) The cell with different types of alignment; marks 1, 2 and 3 designate areas with no alignment, antiparallel and 90o twist alignment. (b) The pixelized cell with 90o twist alignment; black rectangular areas correspond to pixels being under voltage. (c) Azumuthal anchoring coefficient as a function of exposure time for the cell based on poly-11 photoaligning layers. (d-f) Figures representing properties of the films of copolymers based on monomer 11 and 2,2,3,3,4,4,5,5-octafluoropenthyl-methacrylate (MF-2). (d) Photos of water drops corresponding to 0, 40 and 100 mol % of MF-2, in case 1, 2 and 3, respectively. (e) Water contact angle as a function of concentration of MF-2 in the copolymer. (f) Pretilt angle vs. MF-2 concentration curves; 1 â&#x20AC;&#x201C; LC MJ961180, 2 â&#x20AC;&#x201C; LC ZLI 2293.

186

Lyudmyla Vretik et al.

LC photoalignment on these polymers is extremely photo- and thermally stable. The high intensity (Iâ&#x2C6;ź1 W/cm2) UV light exposure for 10 hours did not lead to any noticieable destruction of LC alignment. In the temperature tests, the alignment withstand heat 120oC for 5 hours. The set of these properties makes polymers of this series rather promising for practical applications. However, to evaluate potential of these materials for industrial use, wide range of additional tests should be held, aimed at estimation of LC alignment stability, polar anchoring, parameters of electro-optic response, dielectric properties (e.g., VHR and RDC characteristics), etc. Together with our industrial partners we are currently conducting these studies.

7. Conclusion In summary, we have proposed new strategy in development of photoaligning polymers with reactive C=C bonds. In contrast to widely used C=C groups capable of [2+2] cycloaddition, our materials contain C=C groups capable of polymerization. Due to angularly selective photo-crosslinking of these groups under polarized light, these materials acquire photoaligning function. The photoalignment effect is essentially enhanced if, in parallel with photocrosslinking, polymer structure allows for one or several photo-Fries rearrangements. Based on this concept, several classes of efficient photoaligning materials have been developed with the culmination in a class of polyarylmethacrylates containing non-polymerized methacryloyl groups. These polymers require simple synthesis procedure and low cost starting compounds. They have structures quite flexible for design. Due to position of the main absorption band in shortwave UV range and high crosslinking degree, these materials provide alignment of outstanding photo- and thermal stability. The azimuthal anchoring energy on these polymers approaches 10-4 J/m2, which is claimed for the best photoaligning polymers. The LC pretilt angle can be widely varied by introduction of hydrophobic chains in the polymer structure. These properties make the described polymers rather attractive for industrial application. The industrial tests are in progress. The problems needed to be solved are better optimization of molecular structures and synthesis ways, use of appropriate UV sources (with emission spectrum in a shortwave UV range), finding of optimal exposure procedure. Finally, it is worth noting that the idea of combining in a single sidechain of several photosensitive groups, one of which is capable of crosslinking has been very productive. Recently, we found a dramatic improvement in thermal stability of LC alignment on azopolymer films with

New photoaligning materials with reactive C=C bonds

187

introduction in the side azobenzene chains of terminal groups capable of polymerization [71]. Crosslinking of these groups prevent orientational disordering of azobenzene fragments at elevated temperatures and thus stabilize LC alignment. We are confident that the same approach can be productively used to improve the LC alignment properties of other classes of photoaligning materials.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Takatoh, K., Hasegawa, M., Koden, M., Itoh, N., Hasegawa, R., and Sakamoto, M., 2005, Alignment Technologies and Applications of Liquid Crystal Devices, Taylor & Francis. Ichimura, K., Suzuki, Y., Seki, T., Hosoki, A., and Aoki, K., 1988, Langmuir, 4, 1214. Dyadyusha, A., Marusii, T., Reshetnyak, V., Reznikov, Y., and Khizhnyak, A. 1992, JETP Lett. 56, 17. Schadt, M., Schmitt, K., Kozinkov, V., and Chigrinov, V. 1992, Jpn. J. Appl. Phys., Part 1, 31, 2155. Shannon, P., Gibbons, W., and Sun, S. 1994, Nature (London), 368, 532. Hasegawa, M. 2000, Jpn. J. Appl. Phys., 39, Part 1, No. 3A, 1272. Yokoyama, H. and Kimura, M. 2005, IDW/AD â&#x20AC;&#x2122;05, 49. Yaroshchuk, O., Kyrychenko, V., Tao, D., Chigrinov, V., Kwok, H.S., Hasebe, H., and Takatsu, H. 2009, Appl. Phys. Lett., 95, 021902. Chigrinov, V., Kozinkov, V., and Kwok, H.S. 2008, Photoalignment of Liquid Crystal Materials: Physics and Application, Wiley-SID series, UK. Oâ&#x20AC;&#x2122;Neill, M., and Kelly, S. 2000, J. Phys. D: Appl. Phys., 33, R67. Ichimura, K. 2000, Chem. Rev., 100, 1847. Hasegawa, M., and Taira, Y. 1995, Proc. 21st JLCC, 344. Yaroshchuk, O., Tereshchenko, A., Lindau, Ju., and Boehme, A. 1996, SPIE, 2795, 71. Schadt, M., Seiberle, H., and Schuster, A. 1996, Nature, 351, 212. Reznikov, Yu., Yaroshchuk, O., Gerus, I., Homuth, A., Pelzl, G., and Weissflog, W. 1998, Mol. Materials, 9, 333. Obi, M., Morino, S., and Ichimura, K. 1999, Chem. Mater., 11, 656. Kawatsuki, N., Hamano, K., Ono, H., Ssaki, T., and Goto, K. 2007, Jpn. J. Appl. Phys., 46 (1), 339. US Patent 6,226,066 (2001). Dyadusha, A., Khizhnyak, A., Marusii, T., Reznikov, Y., Yaroshchuk, O., Reshetnyak, V., Park, W., Kwon, S., Shin, H., and Kang, D. 1995, Mol. Cryst. Liq. Cryst., 263, 399. Dumont, M., and Sekkat, Z. 1992, Proc. SPIE, 1774, 188. Yaroshchuk, O., Dumont, M., Zakrevskyy, Y., Bidna, T., and Lindau, J. 2004, J. Phys. Chem. B, 108, 4647.

188

Lyudmyla Vretik et al.

22. Yaroshchuk, O., Kiselev, A., Zakrevskyy, Yu., Bidna, T., Kelly, J., Chien, L.-C. and Lindau, J. 2003, Phys. Rev. E, 68, 011803. 23. Pedersen, T., Johansen, P., Holme, N., Ramanujam, P., and Hvilsted, S. 1998, J. Opt. Soc. Am. B: Opt. Phys., 15, 1120. 24. Yaroshchuk, O., Reshetnyak, V., Tereshchenko, G., and Lindau, J. 1999, Mater. Sci. Eng., C, 8–9, 211. 25. Weigert, F., 1919, Verh. Phys. Ges., 21, 485; 1921, Z. Phys. 5, 410. 26. Yaroshchuk, O., Zakrevskyy, Yu., Kumar, S., Kelly, J., Chien, L.-C., and Lindau, J. 2004, Phys. Rev. E, 69 (1), 011702. 27. Yaroshchuk, O., Chigrinov, V., Nadtoka, O., and Kwok, H. 2006, Liq. Crystals, 33 (2), 149. 28. Usami, K., Sakamoto, K., Uehara, Y., and Ushioda, S. 2005, Appl. Phys. Lett., 86, 211906. 29. Kumar S., Kim, J.H. and Shi, Y. 2005, Phys. Rev. Lett., 94, 077803. 30. Yaroshchuk, O., Agra, D.M.G., Zakrevskyy, Y., Chien, L.-C., Lindau, J., and Kumar, S. 2001, Liq. Crystals, 28, (5), 703. 31. Chigrinov, V., Prudnikova, E., Kozenkov, V., Kwok, H. S., Akiyama, H., Kawara, T., Takada, H., and Takatsu, H. 2002, Liq. Cryst., 29, 1321. 32. West, J., Wang, X., Ji, Y., Kelly, J. R., 1995, SID 95 Digest, XXVI, 703. 33. Yaroshchuk, O., and Kadashchuk, A. 2000, Appl. Surf. Sci., 158 (3–4), 357. 34. Hasegawa, M., 1999, Jpn. J. Appl. Phys., 38, L255. 35. Jackson, P. R., and O’Neill, M. 2001, Chem. Mater., 13, 694. 36. Makita, Y., Natsui, T., Kimura, S., Nakata, S., Kimura, M., Matsuki, Y., and Takeuchi, Y. 1998, J. Photopolym. Sci. Technol., 11, 187. 37. Kawatsuki, N., Tadahiro, A., Kawakami, Y., and Yamamoto, T. 2000, Jpn. J. Appl. Phys., 39, L5943. 38. US Patent 5,389,698. 39. US Patent 6,919,404 B2. 40. Miyachi, K., Kimura, N., Yamada, Y., Mizushima, S. 2010, IDW’10, LCT1-1. 41. Maruyama, K., Hayashi, H., Ono, Y., Nozaki, A., Suzuki, Y., Naitoh, S., and Ikariya, T. 2005, IDW/AD ’05, 1411. 42. Ichimura, K. 2003, J. Photochem. & Photobiol. A: Chemistry, 158, 205. 43. Murai, H., Nakata, T., and Goto, T. 2002, Liquid Crystals, 29 (5), 669. 44. Kim, D-H., Park, S-K., Kwon, S.-B., Kurioz, Yu., Reznikov, Yu., Tereshchenko, O., and Gerus, I. 2006, SID 06 Digest, 867. 45. Wilderbeer, H., Teunissen, J.-P., Bastiaansen, C., and Broer, D. 2003, Adv. Mater., 15 (12), 985. 46. Hwang, J.-Y., Kang, H.-K., Park, C.-J., Seo, D.-S. 2005, Mol. Cryst. Liq. Cryst., 434, 151. 47. Gvozdovskyy, I., Orlova, T., and Terenetskaya, I. 2005, Mol. Cryst. Liq. Cryst., 430, 199. 48. Park, B., Jung, Y., Choi, H., Hwang, H., Kakimoto, M., and Takezoe, H. 1998, Jpn. J. Appl. Phys., 1, 37, 5663 49. Kim, J. Y., Kim, T. H., Kimura, T., Fukuda, T., and Matsuda, H. 2002, Optical Materials, 21, 627.

New photoaligning materials with reactive C=C bonds

50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71.

189

Kang, H., Kang, D., and Lee, J-C. 2009, Polymer, 50, 2104. Chigrinov, V., Muravski, A., and Kwok, H. S. 2003, Phys. Rev. E., 68, 061702. Kobsa, H. 1962, J. Organic Chemistry, 27 (7), 2293. Anderson, J. C., and Reese, C. B. 1960, Proc. Chem. Soc. 1960, June, 217. Frechet, J. and Tessier, T., 1985, Macromolecules, 18 (3), 317. Li, S., and Guillet, J. 1977, Macromolecules, 10 (4), 840. Merle-Aubry, L., Holden, D., Merle, Y., and Guillet, J. 1980, Macromolecules, 13, 1138. Hofler, T., Grieber, T., Gstrein, X., Trimmel, G., Jakopic, G., and Kern, W. 2007, Polymer, 48, 1930. Kawatsuki, N., Takatsuka, H., and Yamamoto, T. 2001, Jpn. J. Appl. Phys., 40, L.209. Allen, N. S., 1983, Polymer Photochemistry, 3, 167. Syromyatnikov, V., Vretik, L., Yaroshchuk, O., Zakrevskyy, Yu., Kim,T. M., Jo, J. H., Kim, J. Y., and Kim, S. H. 2001, Mol. Cryst. Liq. Cryst., 368, 543. Vretik, L., Syromyatnikov, V., Zagniy, V., Paskal, L., Savchuk, O., Yaroshchuk, O., Dolgov, L., Kyrychenko, V., and Lee, C.-D. 2007, Mol. Cryst. Liq. Cryst., 479, 121. Vretik, L., Paskal, L., Syromyatnikov, V., Zagniy, V., Savchuk, O., Dolgov, L., and Yaroshchuk, O. 2007, Mol. Cryst. Liq. Cryst., 468, 173. Kyrychenko, V., Smolyakov, G., Zagniy, V., Vretik, L., Paskal, L., Syromyatnikov, V., and Yaroshchuk, O. 2008, Mol. Cryst. Liq. Cryst., 496, 278. Vretik, L., Syromyatnikov, V., Zagniy, V., Savchuk, E., and Yaroshchuk, O. 2008, Mol. Cryst. Liq. Cryst., 486, 57. Gryshchuk, L., Vretik, L., Syromyatnikov, V., Novikova O., Paskal P., Voytekunas V., and Abadie M.-J. 2004, Polymer International, 53, 27. Gryshchuk, L., Vretik, L., Syromyatnikov, V. 2002, Materials Science, 20, 5. Konovalov, V., Muravski, A., and Yakovenko, S. 2000, SIDâ&#x20AC;&#x2122;00 Digest, 620. Gunzler, H., Gremlich, H.U. IR Spectroscopy. An Introduction. Wiley-VCH: Weinheim, (2002). Bellamy, L.J. The infra-red spectra of complex molecules. Wiley: New York, (1963). Vretik, L., Kyrychenko, V., Smolyakov, G., Yaroshchuk, O., Zagniy, V., Gavrilko, T., Syromyatnikov, V. 2011, Mol. Cryst. Liq. Cryst., 536, 224. Kyrychenko, V., Vretik, L., Yaroshchuk, O., Nadtoka, O., Syromyatnikov, V., Book of abstracts of ILCCâ&#x20AC;&#x2122;10, July 12-16, Krakow, Poland, p. 216.

Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Liquid Crystalline Organic Compounds and Polymers as Materials of the XXI Century: From Synthesis to Applications, 2011: 191-219 ISBN: 978-81-7895-523-0 Editors: Agnieszka Iwan and Ewa Schab-Balcerzak

7. Utilization of various atomic force microscopy techniques in investigation of liquid crystal compounds Andrzej Sikora Electrotechnical Institute, Division of Electrotechnology and Materials Science M. Skłodowskiej-Curie 55/61, 50-369 Wrocław, Poland

Abstract. Recent development of various materials is stimulated by utilization of advanced measurement tools and methods. In particular, the use of the submicron diagnostic techniques is desired, when the impact of the production process on the molecular structure of the material has to be determined in order to optimize the manufacturing technology. One of the most popular measurement techniques which delivers the information about various properties of the surface with nanometer resolution is Atomic Force Microscopy (AFM). This technique, basing on local scanning tip and sample interaction is commonly used in imaging of the topography of the surface as well as its mechanical, electrical, thermal and optical properties. Investigation of the properties of the liquid crystal compounds, which can be applied in wide area of the applications, requires utilization of various measurement techniques in order to obtain desired information. Also the data processing and interpretation is essential task, Correspondence/Reprint request: Dr. Andrzej Sikora, Electrotechnical Institute, Division of Electrotechnology and Materials Science, M. Skłodowskiej-Curie 55/61, 50-369 Wrocław, Poland. E-mail: sikora@iel.wroc.pl

192

Andrzej Sikora

therefore advanced software must be used in order to extract certain information. The topography image can be a source of the information about the roughness, size of particular features as grains or pores and the height of the deposited layer. Observation of the topography change after the exposure the surface to certain force allows to observe the wear phenomena. Mechanical properties of the surface as the stiffness and adhesion can be mapped and used for estimation of the homogeneity of the different compounds dispersion. Such kind of the analysis is essential, where certain behavior of the material is desired. As the liquid crystals are often utilized in applications where the electrical properties must be known, it is possible to verify submicron distribution of the surface potential of thin layer.

Introduction Atomic Force Microscopy is a surface imaging method allowing to obtain certain information about the surface of the sample on the micro- to nanoscale. The high resolution measurement is possible due to observation of the interaction between the scanning tip and the surface [1,2]. As the tip is very sharp [3,4] (typically few nanometers of the apex radius), the interaction area is limited to a very small area, therefore even atomic or molecular imaging is possible. Since the introduction of the AFM in 1985 [1], a wide range of measurement techniques have been developed and successfully introduced on the market. Imaging of the topography of the surface can be performed in three modes: contact [1], intermittent contact [5,6] and noncontact [7,8]. Due to amount of the force acting between tip sample, the noncontact mode is most delicate mode, as contact mode requires at least few nanonewtons of the force to work properly. Therefore one should carefully choose the measurement technique in order to provide the best resolution and not to modify/ damage the scanning tip as well as the surface. In case of liquid crystal compounds it is crucial to distinguish and identify specific components if the macromolecular features are created. In order to do that, advanced modes of imaging are useful. The easiest way to observe the presence of different materials, is to image the mechanical properties of the surface. By measurement of the lateral forces, the friction can be mapped. Such mode is known as LFM (Lateral Force Microscopy) or FFM (Friction Force Microscopy) [9, 10]. On the other hand, the measurement of the response of the surface on modulated load allows to image local stiffness. It can be performed with FMM (Force Modulation Microscopy) [11, 12]. The Phase Imaging mode can be used in local non-homogeneities of the tip-sample interaction imaging [13, 14]. This type of the information is however very difficult to interpret, therefore more advanced methods is also utilized in order to identify certain phenomena

Utilization of atomic force microscopy

193

acting between the tip and the sample [15, 16]. Also the force spectroscopy [17, 18] can be used in order to observe specific interactions as well as their changes due to wear of the nanostructures. Finally, the Kelvin Probe Force Microscopy (KPFM) [19, 20] known also as Surface Potential Imaging can reveal differences in electrical properties of the material at nanoscale. Such information related to the structure of the material can be essential when specific application of the liquid crystal compound is considered. It should be underlined, that the increase of the AFM’s popularity in various fields of science and industry is due to ability to performing the measurements in: -

vacuum, ambient and liquids [21, 22, 23], high magnetic fields [24], wide range of temperatures: lower than 100K [25, 26, 27] and over to 670K [28] (locally),

Also the ability of large objects measurement (scanning head can be placed directly on the investigated surface) as well as the simple sample preparation are significant advantages of the method. The development of the AFM techniques is still progressing bringing a new measurement modes and solutions.

1. Principles of the Atomic Force Microscopy 1.1. Tip sample interaction – The high resolution imaging method High resolution imaging of the surface, requires local measurement of the probe-surface interaction. Therefore the radius of the tip’s apex should be as small as possible. Commercially available AFM tips, integrated with cantilevers which are used to force sensing, are microfabricated from silicon and silicon nitride by processes used in semiconductor industry (photolithography and anisotropic etching) [29]. Such technology allows to provide relatively low cost in mass production, high repeatability of the parameters of the cantilevers and very sharp tip. When such tip approaches the surface (fig. 1), it senses the attractive and repulsive forces. By maintaining the interaction force at constant level and moving the tip back and forth over the surface, one can acquire the changes of the tip’s height. This data, when mapped, shows the topography of the surface. The tip-sample interaction is connected to the presence of various phenomena in such short range. The most important are: the Van der Waals forces, chemical bonds (Morse potential) (fig. 2), electrostatic force,

194

Andrzej Sikora

Figure 1. Schematically shown tip-sample interaction allowing to obtain high imaging resolution [30].

Figure 2. The plot of phenomena used for tip-sample distance detection [30].

magnetic force, adhesion force and capillary force. Although all the phenomena have an impact on the behavior of the tip, applying a certain measurement method causes specific kind of interaction, which plays main role in the measurement. When the contact mode is used, the repulsive forces (Van der Waals and chemical bonds) have a major role in the tip-sample distance control. In non-contact mode, when the tip is in attractive forces area, Van der Waals, electrostatic and magnetic forces are most significant in this process. One must also notice, that the vertical resolution of the imaging of the surface depends on the measurement technique, as different phenomena show various dynamics along the tip-sample distance. In figure 2 the tunneling current was also presented as the reference and the first method of the atomic resolution surface imaging [31]. One should be aware that the measurement environment plays essential role in the measurement. When the sample is in ambient, one must take into

Utilization of atomic force microscopy

195

account presence of the water film on the surface, therefore unlike in the vacuum, obtaining the ultimate resolution may by much more complex. In case of the measurements performed in liquids, involvement of much more phenomena must be taken into account. All this information should be considered when the measurement is planned. 1.2. Principles of AFM systems Typical AFM system contains three major components: -

scanning head with the sample, scanning tip and the cantilever, optical detection system of the bending of the cantilever and piezoelectric positioning system, controller with analogue and digital circuits, high voltage amplifiers driving the piezoelectric positioning system, computer with dedicated software for scanning process control and data processing and analysis.

Simplified diagram showing typical AFM setup is shown in figure 3. Additionally, the vibration damping system as well as the temperature and noise isolation chamber is required in order to provide stable measurement conditions. It should be noted that the scanning heads placed in high vacuum and cryogenic chambers are also commercially available. The scanning head is also equipped with the optical microscope which includes the CCD camera. This feature allows to place the scanning tip in area of interest as well as perform the laser beam alignment on the backside of the cantilever. The example of such view is shown in figure 4.

Figure 3. Principle of operation of Atomic Force Microscope with optical detection system of the cantileverâ&#x20AC;&#x2122;s displacement.

196

Andrzej Sikora

The scanning process control software allows to define such parameters as: the measurement mode, scanning speed, scanning area, the settings of the PID (proportional-integral-derivative) controller responsible for the tipsample distance maintaining, the scanning X-Y resolution and many others. Additionally to the topography image, one can acquire other signals which can be related to certain modes: friction, stiffness, surface potential. Also some specific data can be imaged as well: PID error signal, total optical signal power and many others. This information can play a significant role during the interpretation of the results, when appearance of the artifacts must be considered. A screenshot of the AFM control software is shown in figure 5.

Figure 4. The optical microscope view of the cantilever during the measurement of the liquid crystal compound.

Figure 5. The view of the AFM control software SPMLab from Veeco. On left side of the screen, the images of various signals are visible.

Utilization of atomic force microscopy

197

1.3. Data representation Once the measurement data is gathered in the memory of the computer, it needs to be displayed on the screen in order to be interpreted. As the data is X-Y two-dimensional matrix, the most appropriate method of the representation is the image with color related palette to the value of measured phenomena (fig. 6). Although the way the images are generated by various programs can differ, the general rule is that on the sides of the pictures one can find lateral scale (sometimes the ruler is placed on the picture), and additionally the Z-scale is shown on the side of the image in order to provide the height information. Therefore one can see that the darker colors represent the low areas, while the bright ones are related to the high areas of the sample. In some cases the image suffers insufficient color palette dynamics due to large amount of fine details, which canâ&#x20AC;&#x2122;t be seen in typical view. Such case can be particularly inconvenient, when the images are printed in a gray scale in the paper. Therefore the utilization of certain transform allows to overcome that problem and improve the readability of the image by revealing small details like molecular ordering of the surface. One of the useful tools in such case is the Sobel transform [32, 33], which as the edge detection filter provides desired result. The example of such image is shown in figure 6. The presence of the steps-like structure in top right corner of the image is clearly visible. It should be underlined, that transformed image does not provide the quantitative information about the height of the structure, therefore it should be presented with the original topography image.

Figure 6. The topography image (on left) and the Sobel transform of the topography (on right) of the liquid crystal compound.

198

Andrzej Sikora

+ Figure 7. The three-dimensional view of the topography (on left) and the phase image (on right) of the liquid crystal compound.

Presenting the surface in three-dimensional view (fig. 7) can be more intuitive and obvious for some readers, however due to influence of the perspective, some details and dimensions can be misread. Therefore the planar view is the most popular imaging method. As mentioned above, the AFM system can deliver additional information about the surface, therefore other images can be presented in order to image the distribution of certain phenomena on the scanned surface (fig. 7).

2. Morphology â&#x20AC;&#x201C; Dimensions and roughness of the surface Basing on the topography data, one can extract the information about the size (length, width, height) of certain structures. Also a number of the roughness parameters can be calculated in order to obtain quantitative, statistical information about the surface. It should be emphasized, that AFMâ&#x20AC;&#x2122;s due to their operational principles are used as the metrology tools in microand nanoscale calibration procedures [34, 35], therefore one can assume that obtained results are reliable. Although the uncertainty analysis of the results is a complex task and requires consideration of a number of issues including the shape of the tip [36, 37]. 2.1. Determination of size of the structures The surface of the liquid crystal compound can reveal the structures which allow to conclude the specific macromolecular ordering. Figure 8 shows the topography of the discotic-shaped azomethines with triphenylamine

Utilization of atomic force microscopy

199

moieties sample [36] measured with intermittent contact mode. One can easily note the presence of the terraces-like structures therefore high level of macromolecular ordering can be concluded. The height distribution shows presence of three major fractions of the height (fig. 9). Basing on this graph, one can easily note that single stepâ&#x20AC;&#x2122;s height is 4 nm. Additionally, the fraction of certain height can be determined. Another issue that is frequently analyzed is the impact of the deposition process on the thickness of the film. As in some applications the thickness of the film is limited (requested transparency), one is interested in optimization of the process in order to meet the requirements. Figure 10 shows the surface of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine with ester groups PAZ 1 sample, prepared using a simple one-step polycondensation procedure without solvent and catalysts [37]. In order to verify its thickness, the scratch revealing the

Figure 8. The topography of the surface of the azomethine.

Figure 9. The height distribution of the sample shown in figure 8.

200

Andrzej Sikora

surface of the substrate was made and the area of scratch was measured with intermittent contact mode. Obtained topography map can be analyzed using the profile/ multiprofile (fig. 11) and the height distribution data (fig. 12). The profile readout delivers information about certain place, so it can be not representative for all surface. The multiprofile method allows to verify the repeatability of the measurements in few places, therefore this approach is more reliable. The height distribution data gives statistical information, however selection of the certain place on the distribution peak can be difficult, when the surfaces are not flat. One can use both methods in order to obtain the data redundancy and evaluate the measurement process.

Figure 10. The topography of the scratched aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 1 sample with profile markers (on left) and height distribution selection markers (on right).

Figure 11. The set of three profiles of the sample shown in figure 10. The repeatability of the data sets is acceptable.

Utilization of atomic force microscopy

201

Figure 12. The height distribution of the sample shown in figure 10. Selected middle area of the graph is related to the measured height of the film.

2.2. Roughness statistics The roughness data is very often one of the most important information about the investigated sample. Basing on it, one can conclude the impact of the substrate on the growth of the thin film, the morphological homogeneity of the surface, the usability of the material in specific application and finally, its mechanical wear. Due to the complexity of various criteria used in description of the roughness, one needs to be aware of certain issue that must be considered, and which determines the roughness factor that should be used. Nevertheless, two main factors are given in general: Sa: roughness average and Sq: root mean square [38]. Also two another parameters are helpful during the analysis of the surface’s morphology: Ssk: surface skewness (the unbalance of height distribution maximum) and Sku: surface kurtosis (peak’s width on height distribution). The surface area ratio (Sdr) can be very useful source of the information, when one aims to create very porous material in order to obtain large area of contact (for instance for chemical processes), or very smooth in order to reduce the wettability (protecting, passive layers). Figure 13 shows the surface of the aliphatic– aromatic polyazomethine PAZ 4 measured with intermittent contact mode [37]. The roughness analysis was performed and the results were placed below. The height distribution is the basic statistical kind of information (fig. 14). The share of certain height ranges can be easily estimated. Also the presence of the plateaus can be revealed as well. As mentioned above, the determination of height of the structures can be performed either. The Sku and Ssk parameters describe the asymmetry and width of the distribution’s peak.

202

Andrzej Sikora

Figure 13. The topography of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 4.

Figure 14. The height distribution graph of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 4.

Figure 15. The Abbott-Firestone graph of PAZ 4.

Utilization of atomic force microscopy

203

The Abbott-Firestone graph also known as the Bearing Area Curve (BAC), is the cumulative probability density function of the surface profile's height. This kind of data representation is more popular among some professions than the height distribution. The set of the roughness parameters of PAZ 4 is placed in table 1. Additionally, advanced procedures of the software allows to verify the privileged direction of the structure’s orientation. The angular spectrum Table 1. The set of roughness parameters measured for PAZ 4. Scanning area: Sa Sq Ssk Sku Sy St Sz S10z Sz tph Sds Ssc Sv Sp Smean Sdq Sdq6 Sdr S2A S3A

0.41250, 7.0553E+5 11.633 nm 14.655 nm 0.014465 2.9587 90.210 nm 90.210 nm 90.210 nm 89.283 nm 89.283 nm 11122 1/µm² 0.48894 1/nm 42.517 nm 47.693 nm 9.3844E-8 nm 0.75192 0.70392 25.428 % 5.625E+5 nm² 7.0553E+5 nm²

Sbi Sci Svi Spk Sk Svk Std Stdi Srw Srwi Shw Sfd Scl20 Str20 Scl37 Str37 Sdc0_5 Sdc5_10 Sdc10_60 Sdc50_95

0.59688 1.5856 0.12003 15.010 nm 36.749 nm 14.016 nm 0 degree 0.63881 750 nm 0.017684 250 nm 2.0874 96.869 nm 0.41250 77.789 nm 0.76812 23.140 nm 5.2427 nm 19.705 nm 24.225 nm

Figure 16. The angular spectrum graph of PAZ 4.

204

Andrzej Sikora

graph shows the angle and the amplitude of the slopes of the features on the surface. Using this tool one can easily verify if the flake-like objects grew with statistically homogenous orientation, or due to some influence (gas flow, electrostatic field, temperature gradient), the orientation was induced. Such influence in case of liquid crystal compounds can be useful when properly utilized in the production process, but uncontrolled can cause considerable loses. 2.3. Grain / pores analysis The grain and pores analysis plays important role in evaluation of the properties of the surface. During deposition of the liquid crystal compounds, various forms may be created. Their size, shape, orientation, density and other parameters may be caused by specific component and can be controlled with specific process factor. In order to provide the appropriate data allowing to determine the relation between morphology of the surface and the fabrication process, certain data must be acquired. In presented example the topography of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 1 [37] is shown (fig. 17). One can easily note, that the cheese-like surface contains a number of pores. After applying the force, their amount increased significantly. The impact of the force on the quantity, shape, volume can be determined basing on statistical information about the pores properties. Figure 18 shows an example of the graph presenting the radius distribution of the pores on both surfaces. One can see the general increase of the number of the pores, however the distribution of the radius changed as well.

Figure 17. The topography of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 1 before applying the force (on left) and after applying the force (on right). The pores are detected and marked by the software.

Utilization of atomic force microscopy

205

Figure 18. The graph showing the distribution of the pores size of PAZ 1 before applying the force (on left) and after applying the force (on right).

3. Determination of local mechanical properties of the material The knowledge about the mechanical properties of the surface at the submicron scale is very often desired in order to verify if it meets specific requirements. As the new devices become smaller, the measurement has to be performed on very small surfaces. Also mapping of the mechanical properties is often performed in order to determine the uniformity of the distribution of specific parameter. 3.1. Force spectroscopy The measurement of the mechanical properties in micro- and nanoscale is a difficult and complex procedure [17, 29, 39]. It requires detailed knowledge about the measurement device, in particular the cantilever (Young modulus) and the tips apex diameter [40, 41]. However utilization of the force spectroscopy is very widely used, as it allows to determine specific properties of the surface as well as to observe various kinds of behavior of the sample. The measurement is performed as follows: when the tip approaches the surface, the bending of the cantilever is monitored. At some point the tip enters the interaction area (attractive forces), which makes the cantilever bending into the sampleâ&#x20AC;&#x2122;s direction (snap-in event). Then the tip touches the surface and as the distance reduction continues, it presses the surface. At this moment, the cantilever bends in opposite direction, as the vector of the force changed. When the force reaches desired level, the tip is retracted form the surface, and at some point one can note snap-off event, which generally allows to conclude that the cantilever is back in equilibrium state (fig. 19).

206

Andrzej Sikora

Acquired curve can be utilized in determination of the stiffness, adhesion, peak force, energy dissipation [30]. Presented examples of the force-distance curved measured on aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine samples [37] are shown in figure 20. The force curve of PAZ 3 sample shows typical shape, when solid state sample is measured. PAZ 5 sample behaves different way. One can see that when the tip was retracted, the force decreased very rapidly. This was caused due to inelastic deformation of the surface, however partially the surface recovered its shape. This behavior can indicate Bingham plastic like behavior of the material. The most complex curve was obtained on surface of PAZ 1.

Figure 19. The principles of the measurement and analysis of the force-distance curve [30].

Figure 20. The set of force-distance curves measured on aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine samples: PAZ 3 (top left), PAZ 5 (top right), PAZ 1 (bottom left).

Utilization of atomic force microscopy

207

After the snap-in event occurred, the tip was still attracted to the surface for another about 50 nm. Also the snap-out event had unusual shape, with slope decay and finally sudden out-of contact jump. Due to complex morphology of the sample, one can assume that the specific behavior could be caused by presence of the fiber-like structures, which could snap to the tip and drag it as the approach and retraction was performed. Obtained information is essential, when further processing of the film is planned in order to obtain expected product. It must be emphasized, that liquid crystal compounds very often can exhibit complex behavior. 3.2. Observation of wear phenomena of macromolecular features When the application of the liquid crystal compound as covering/ protecting film is considered, the wear resistance condition must be fulfilled. The tests of the applying the repetitive load in nanoscale can deliver interesting data about macromolecular response on acting force. Presented example shows the series of topography scans of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 1 (fig. 21) [37] measured with intermittent contact mode. As the force induced by the oscillating tip increased, one can observe the changes in the topography. In order to improve the readability, the Sobel transform images are also shown. The amplitude of the cantilever oscillation was respectively: 1, 4.5, 9 and 16 nm. In order to provide complete set of data about investigated material, the force spectroscopy measurements were performed. The changes of the forcedistance curves as the following load cycles were repeated, confirmed the presence of macromolecular fatigue phenomena. Figure 21

208

Andrzej Sikora

Figure 21. The set of topography images (left) and the Sobel transform of the topography image (right) after the force applying. The oscillation amplitudes of the cantilever, causing observed changes of the surface were as follows: 1, 4.5, 9 and 16 nm.

Obtained results allowed to conclude the active behavior of the fiber like macromolecular structures, and classify the material as the potentially applicable in sensor industry.

Utilization of atomic force microscopy

209

Figure 22. The set of force-distance curves acquired during following measurements obtained on aliphatic–aromatic polyazomethine PAZ 1.

3.3. Phase Image - Identification of non-homogeneities of the material The phase image is the additional signal, accessible in intermittent contact mode. As the scanning tip oscillates, the phase shift between the response signal and excitation signal may vary. It can indicate the nonhomogeneities of the material, as its origin is connected to the energy dissipation on the surface [16]. Figure 23 presents the surface of the aliphatic–aromatic polyazomethine PAZ 4 [37] containing flakes-like features. Additional map shows the phase image, which reveals local nonhomogeneities of the properties of investigated material. The flakes-like objects differ significantly from the substrate. It may be caused by differences of the stiffness. One should be aware, that the phase image signal is very difficult to interpret in terms of identifying of certain phenomena responsible for observed changes. It is however very common and as qualitative source of information can be effectively used. Figure 24 presents the surface of the aliphatic–aromatic polyazomethine PAZ 4 [37] containing fiber-like structures. Although the material is homogenous, the energy dissipation varies when the tip moves across the observed features. Therefore phase image data can be used in determination of the dimensions of the fibers. The average length of the objects is in the range of 50-100 nm, and the diameter – about 10 nm. One can see that the phase image can be useful tool in both: investigating the presence of the components with different mechanical properties and observation of macromolecular objects of the liquid crystal compounds.

210

Andrzej Sikora

Figure 23. The topography (left) and phase image (right) of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 4.

Figure 24. The topography (left) and phase image (right) of the aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine PAZ 3.

3.4. Mapping of mechanical properties of the surface using advanced analysis of torsional bending of the cantilever As the phase image mode canâ&#x20AC;&#x2122;t deliver desired information about certain properties of the material, the advanced mode utilizing the torsional bending of the cantilever can be used [42, 43, 44]. The high frequency response of the cantilever is analyzed and converted into the force-distance curve, and eventually used for the mechanical properties determination. By creating the maps of observed phenomena, one can obtain detailed information about the properties of certain components or features on the surface. In this case, the following maps of the properties are available: the topography, adhesion,

Utilization of atomic force microscopy

211

Figure 25. The images of the topography (top, left), adhesion (top, right), stiffness (middle, left), peak force (middle, right), energy dissipation for tip-sample separation (bottom left) and the energy dissipation for the deformation (bottom right) of the sample of aliphaticâ&#x20AC;&#x201C;aromatic polyazomethine.

212

Andrzej Sikora

stiffness, peak force, energy dissipation for tip-sample separation and energy dissipation for deformation of the surface. It should be mentioned, that some maps (peak force, energy dissipation for deformation) can deliver the information about the tip-sample interaction, allowing to verify the influence of the scanning process on the surface. It is essential, when one investigates fragile sample and requires the confirmation of minimal impact of the force applied by the scanning tip. The example shown in figure 25 allows to observe the structure of the material in much more detail than the topography itself. One can observe presence of the changes of the stiffness and the adhesion along the edges of elevated structures. It can be caused by their sandwich-like structure indicating the high level of molecular ordering. Also, as expected, the stiffness of the surface is smaller on the top of elevated structures. On the other hand, the adhesion on the top of those structures is larger than on the sides. On the flat surface however one can see also the non-homogenous distribution of the adhesion (left top corner). One can also note presence of the buried structures in the left bottom corner, revealed on the stiffness and the energy dissipation for the tip-sample separation. This presence can be also seen on the topography image after careful analysis. 3.5. Observation of the doping materialâ&#x20AC;&#x2122;s distribution using the Force Modulation Mode The Force Modulation Mode (FMM) allows to observe local changes of the stiffness of the surface [11, 12]. As the scanning tip touches the surface and presses it with certain frequency, the surface bends proportionally to its stiffness. The response can be imaged and used for the homogeneity analysis, when different materials reveal their properties. The advantage of this method is possibility of observing the features located few nanometers below the surface, as the pressing tip bends the sample with certain force. It should be noted, that the applied force is relatively large (up to tens of nanonewtons) and can cause the modification of the surface. Therefore it should be applied very carefully. Presented results show the measurement data obtained during scanning of the liquid crystal doped with fullerene (fig. 26). As the topography slightly can indicate presence of the features connected to the doping of the material, the force modulation responses (the amplitude as well as the phase) reveal very clearly the structure of the fullerene chains. In order to determine the matrix/ doping material surface ratio, the histogram of the FMM response was analyzed (fig. 27). Once the factor was determined, one can compare it to the fabrication procedures in order to optimize it. Using presented data it is possible to verify the dispersion of the

Utilization of atomic force microscopy

213

Figure 26. The result of the measurement of the liquid crystal compound doped with the fullerene acquired with Force Modulation Mode. The topography (top), the mechanical response amplitude (bottom left), the mechanical response phase (bottom right).

Figure 27. The histogram of the mechanical response amplitude of the surface presented in figure 26.

214

Andrzej Sikora

doping material. As one can see, the fullerene particles created the net-like feature made of multiple chains of the diameter approximately 6 nm. 3.6. Determining presence of the components by measurement of the friction Some material can be distinguished by the measurement of the lateral forces measured with the scanning tip in contact mode with Lateral Force Microscopy or Friction Force Microscopy mode. As the friction in submicron scale or even atomic scale can be observed [9, 10], the imaging of this property is a useful tool in determining the dispersion of the doping material.

Figure 28. The result of the measurement of the liquid crystal compound doped with the fullerene acquired with Lateral Force Microscopy. The topography (top), the friction during the forward scan (bottom left), the friction during the backward scan (bottom right).

Utilization of atomic force microscopy

215

Figure 29. The angular spectrum graph of the friction of the liquid crystal compound doped with the fullerene.

As the friction images reveal complex structure made of fullerene chains, the homogeneity of the chains orientation is one of the issues to be analyzed. The utilization of the angular spectrum tool allowed to observe that the orientation of the lines of created structure is not 360o homogenous. There are some privileged directions that could be caused by the properties of the substrate, presence of the electrostatic field, gas flow, temperature gradient or other influence.

4. Surface potential imaging using Kelvin Probe Microscopy As the liquid crystal compounds can be used in applications where their reaction on the current flow or the electrostatic field is expected, the knowledge about the relation between the molecular structure and electrical properties of such materials is required. One of the measurement modes capable of delivering such information is the Kelvin Probe Force Microscopy (KPFM) known also as the Surface Potential imaging mode [19, 20]. As in other modes the sample should be biased (Electrostatic Force Microscopy) [45, 46] or conduct the current (Conductive AFM, Scanning Spread Resistance Microscopy) [47, 48], the KPFM mode doesnâ&#x20AC;&#x2122;t require the electrical connection of the sample to any circuit, therefore the preparation of the sample for such measurement is relatively easy. The Kelvin Probe technique can provide qualitative CPD (Contact Potential Difference) imaging although one should be aware the influence of surface states if the measurement is performed in ambient. Figure 30 shows the results of the measurement of the azomethine [49]. As the material shows the terraces-like structure, the impact of the morphology on the electrical properties was

216

Andrzej Sikora

Figure 30. The topography (left) and surface potential image (right) of the azomethine.

Figure 31. The profile of the surface potential created along the line marked in surface potential map.

investigated. The surface potential map reveals the relation of the topography and SP image. In order to determine the influence of certain height, the profile of SP distribution was prepared (fig. 31). The measured relation was 30 mV/91 nm, which gives 0.3 mV/1nm. As the observed thickness of the monolayer was approximately 0.3 nm, the single monolayer causes approximately the 0.1 mV change of the contact potential.

Summary Recent progress of the material science, with strong focus on nanoscale applications, forces utilization of advanced measurement modes of Atomic Force Microscopy in order to provide the information of the properties of the

Utilization of atomic force microscopy

217

material at submicron level. By applying such methods as: contact mode with lateral force measurement or the FMM mode as well as the intermittent contact mode with Phase Image and other sophisticated measurement techniques, one can obtain interesting information about the surface. As presented in this chapter examples of the investigations of the liquid crystal compounds, the advanced measurement methods should be supported by the software solutions which are able to determine a number or the properties, which very often must be analyzed statistically. Presented results proved that AFM is very useful and versatile diagnostic tool.

Acknowledgements This work was performed within statute funds of Electrotechnical Institute and partially was supported by the Polish Ministry of Science and Higher Education (MNiSW) in the frame of the research project no. N N505 466338.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Binnig G., Quate C.F., Gerber C., 1986, Phys. Rev. Lett. 56, 930. Binnig G., Gerber C., Stoll E., Albrecht T.R., Quate C.F., 1987, Europhys. Lett., 3, 1281. Wolter O., Bayer T., Greschner J., 1991, J. Vac. Sci. Technol. B 9(2), 1353. Ohnesorge F., Binnig G., 1993, Science 260, 1451. Zhong Q., Inniss D., Kjoller K., Elings V., 1993, Surf. Sci. Lett. 290, 688 Hansma P. K., Cleveland J. P., Radmacher M., Walters D. A., Hillner P. E., Bezanilla M., Fritz M., Vie D., Hansma H. G., Prater C. B., Massie J., Fukunaga L., Gurley J., Elings V., 1994, Appl. Phys. Lett. 64(13), 1738. Marti Y., Williams C.C., Wickramasinghe H.K., 1987, J. Appl. Phys. 61(10), 4723. Sarid D., Elings V., 1991, J. Vac. Sci. Technol. B 9, 431. Mate C.M., McClelland G.M., Erlandsson R., Chiang S., 1987, Phys. Rev. Lett. 59, 1942. Marti O., Colchero J., Mlynek J., 1990, Nanotechnology 1, 141. Maivald P., Butt H.J., Gould S.A.C., Prater C.B., Drake B., Gurley J.A., Elings V.B., Hansma P.K., 1991, Nanotechnology 2, 103. Radmacher M., Tillman R.W., Fritz M., Gaub H.E., 1992, Science 257, 1900. Magonov S.N., Elings V.B., Whangbo M.H., 1997, Surf. Sci. 375, 385. Pickering J.P., Vancso G.J., 1998, Polym. Bull. 40, 549. Garcia R., San Palo A., 1999, Physical Review B 60, 4961 San Palo A., Garcia R., 2001, Physical Review B 64, 193411 Anczykowski B., Kr端ger D., Babcock K.L., Fuchs H., 1996, Ultramicroscopy 66, 251.

218

Andrzej Sikora

18. Lantz M.A., Hug H.J., Hoffmann R.,. van Schendel P.J.A, Kappenberger P., Martin S., Baratoff A., Güntherodt H.-J., 2001, Science 291, 2580. 19. Nonnenmacher M., O’Boyle M. P., Wickramasighe H. W., 1991, Appl. Phys. Lett. 58 (25), 2921. 20. Jacobs H. O., Knapp H. F., Muller S., Stemmer A., 1997, Ultramicroscopy 69, 239, 39 21. Marti O., Drake B., Hansma P.K., 1987, Appl. Phys. Lett. 51, 484. 22. Binggeli M., Christoph R., Hintermann H.E., Colchero J., Marti O., 1993, Nanotechnology 4, 59 23. Giessibl F.J., Gerber C., Binnig G., 1991, J. Vac. Sci. Technol. B 9, 984. 24. Wildoer J.W.G., Harmans C.J.P.M., van Kempen H. 1997, Phys. Rev. B 55, R16013. 25. Kirk M.D., Albrecht T., Quate C.F., 1988, Rev. Sci. Instrum. 59, 833. 26. Albrecht T.R., Grutter P., Rugar D., Smith D.P.E., 1992, Ultramicroscopy 42–44, 1638. 27. Hug H.J., Moser A., Jung T., Fritz O., Wadas A., Parashikor I., Güntherodt H.J., 1993, Rev. Sci. Instrum. 64, 2920. 28. Basire C., Ivanov D.A., 2000. Phys. Rev. Lett. 85, 5587. 29. Albrecht T.R., Akamine S., Carver T.E., Quate C.F., 1990, J. Vac. Sci. Technol. A 8, 3386. 30. Bhushan B., 2007 Handbook of Nanotechnology, 2nd edn. Berlin SpringerVerlag. 31. Binnig G., Rohrer H., Gerber C., Weibel E., 1982, Phys. Rev. Lett. 49, 57. 32. Vincent O. R., Folorunso O., 2009, Proceedings of Informing Science & IT Education Conference (InSITE) 2009, 97. 33. SPIP online manual http://www.imagemet.com/index.php?main=products&sub= features&id=120 34. Dai G., Pohlenz, F., Dziomba T., Xu M., Diener A., Koenders L., Danzebrink H.U., 2007, Meas. Sci. Technol. 18, 415. 35. Korpelainen, V., Lassila, A., 2007, Meas. Sci. Technol. 18, 395. 36. Iwan A., Sek D., Pociecha D., Sikora A., Palewicz M., Janeczek, H., 2010, Journal of Molecular Structure 981, 120. 37. Iwan A., Palewicz M., Sikora A., Chmielowiec J., Hreniak A., Pasciak G., Bilski P., 2010, Synthetic Metals 160, 1856. 38. SPIP online manual http://www.imagemet.com/WebHelp/spip.htm#roughness.htm 39. Cleveland J.P., Manne S., Bocek D., Hansma P.K., 1993, Rev. Sci. Instrum. 64, 403. 40. Williams P. M., Shakesheff K. M., Davies M. C., Jackson D. E., Roberts C. J., Tendler S. J. B., 1996, J. Vac. Sci. Technol. B 14(2) 1557. 41. Villarrubia J. S., 1996, J. Vac. Sci. Technol. B 14(2) 1518. 42. Sahin O., 2008, Physical Review B 77, 115405. 43. Sahin O., Erina N., 2008, Nanotechnology 19, 445717. 44. Sikora A., Bednarz L., Central European Journal of Physics, 9(2), 2011, 372.

Utilization of atomic force microscopy

219

45. Stern J. E., Terris B. D., Mamin H. J., Rugar D., 1988, Appl. Phys. Lett. 53, 2717. 46. Terris B. D., Stern J. E., Rugar D., Mamin H. J., 1990, Vac. J. Sci. Technol. A8, 374. 47. Thimm J., Mechler A., Lin H., Rhee S.K., Lal R., 2005, J. Biol. Chem. 280, 10646. 48. Matsuki N., Ishihara R., Beenakker K., 2009, Journal of the Society for Information Display 17(3), 293. 49. Iwan A., Bilski P., Janeczek H., JarzÄ&#x2026;bek B., DomaĹ&#x201E;ski M., Rannou P., Sikora A., Pociecha D., Kaczmarczyk B, 2010, Journal of Molecular Structure 963, 175.