The International Journal of Engineering And Science (IJES) ||Volume|| 1 ||Issue|| 1 ||Pages|| 09-12 ||2012|| ISSN: 2319 – 1813 ISBN: 2319 – 1805
Enhancement of different electrolytes on Porous In0.27Ga0.73N using photo-electrochemical etching A. Y. Hudeish1, A. Mahgoob 2 1, 2
Physics Department, Hodeidah University, Hodeidah, Yemen.
--------------------------------------------------- Abstract -------------------------------------------------This article reports the properties and the behavior of In 0.27 Ga0.73 N during the photoelectrochemical etching process using four different electrol ytes. The measurements show that the porosity strongly depends on the electrolyte and highl y affects the surface morphology of etched samples, which has been revealed by scanni ng electron microscopy (S EM) i mages. Peak intensity of the photoluminescence (PL) spectra of the porous In0.27 Ga0.73 N samples was observed to be enhanced and strongly depend on the electrolytes. Among the samples, there is a little di fference in the peak position indicati ng that the change of porosity has little influence on the PL peak shift, while it highly affecting the peak intensity. Raman spectra of porous In0.27 Ga0.73 N under four different solution exhi bit phonon mode E2 (high), A1 (LO), A1 (TO) and E2 (low). There was a red shift in E2 (high) in all samples, indicating a relaxati on of stress in the porous In0.27 Ga0.73 N surface with respect to the underlyi ng single crystalline epitaxial In 0.27 Ga0.73 N . Raman and PL intensities were high for samples etched in H2SO4:H2O2 and KOH followed by the samples etched in HF:HNO3 and in HF:C2 H5 OH. Keywords— Electrolyte; InGaN; Photo-electrochemical etching; Porosity.
----------------------------------------------------------------------------------------------------------------Date of Submission: 20, October 2012, Date of Publication: 10, November 2012, ----------------------------------------------------------------------------------------------------------------I.
INTRODUCTION
The In GaN ternary alloy system receives a great deal of attention among III-nit ride co mpound semiconductors because of its direct band gap tuning fro m 0.7 eV for InN to 3.4 eV for GaN, giv ing In GaN great potential for the design of high efficiency optoelectronic devices that operate in the IR, v isible, and UV reg ions of the electro magnetic spectrum [1]. Porous III-n itride co mpounds are considered as pro mising materials for optoelectronics [2] and chemical and biochemical sensors [3] because of their unique optical and electronic properties compared with bulk materials [4,5]. The formation of a porous nanostructure has been widely reported for crystalline silicon [6]. In addition to porous silicon research, attention has also been focused on other porous semiconductors, such as GaAs [7] and GaN [8-10]. Interest in porous semiconductor materials arises fro m the fact that these materials can act as sinks for threading dislocation and are able to acco mmodate strain. Po rous semiconductor materials are also useful for understanding the fundamental properties of nanoscale structures for the development of nanotechnology. Research on porous GaN is strongly driven by the robustness of porous GaN, including its excellent thermal, mechanical, and chemical stabilities that make it h ighly desirable for
www.theijes.com
optical applicat ions [11]. Many researchers [4,12-14] have used the photoelectrochemical etching (PEC) technique to synthesize porous GaN. The PEC technique is more suitable and cheaper compared with other techniques for producing high density nanostructures with controlled pore size and shape [4]. Electrolyte, current density and illu mination are the main factors that affect electrochemical etching. Hydrofluoric acid (HF) is the most common ly used material in etching GaAs and GaN [15]. In the current study, the PEC technique is used to synthesize porous InGaN nanostructures at various current densities. To the best of our knowledge, this study is the first to report of porous In GaN by using four different electrolytes as the PEC technique.
II.
EXPERIMENTAL
In 0.27 Ga0.73 N/GaN/AlN epitaxial layers were grown on Si(111) substrate by using a plasma assisted mo lecular beam ep itaxy (PA-M BE) system (Veeco Gen II). High-purity sources, such as galliu m (7N), alu minu m (6N5), and indiu m (7N), were installed in the Knudsen cells. Reactive nitrogen species were generated by channeling high-purity nitrogen to a radio frequency (RF) source. The resultant nitrogen plasma was at a nitrogen pressure of
The IJES
Page 9
Enhancement of different electrolytes on Porous In 0.27 Ga 0.73N using photo-electrochemical etching 1.5×10−5 Torr under 300W rated discharge power. The III. RESULTS AND DISCUSSION Si(111) wafer (3inches) was cleaned using the Radio Corporation of A merica (RCA) method prior to the loading of the substrate to the PA-MBE system. The substrate was outgassed in the load-lock and buffer chamber after being loaded into the PA-MBE system, and then it was transferred to the growth chamber. Surface treat ment was conducted on the substrate to ensure the absence of SiO2 , leaving only clean Si substrate. This process was done by depositing a few monolayers of Ga on the substrate at 750°C, resulting in the format ion of Ga 2 O3 . The clean Si surface was further confirmed by the presence of prominent Kikuchi lines that were observed using reflection high energy diffraction. A few monolayers of Al were deposited on the Si surface prior to the growth of nitride layers to inhib it the formation of Si xNy , which is detrimental to the g rowth of subsequent epitaxy layers. An AlN buffer layer was deposited by setting Al and N shutters to open simu ltaneously for 15 min. The GaN layer was then deposited on the buffer layer at a substrate temperature of 800°C for 33 min. Finally, the substrate temperature was reduced to 700°C, and the In and Ga effusion cells were heated to 925 and 930°C, respectively, to initiate the growth of In GaN. The final growth process lasted for approximately 30 min. Porous In 0.27 Ga0.73 N was produced using the UV-PEC method. The etching cell was made fro m Teflon with platinu m wire as cathode and In 0.27 Ga0.73 N film as anode. The native oxide of the samples was init ially removed using NH4 OH:H2 O (1:20), followed by HF:H2 O (1:50). Boiling aqua regia HCl:HNO3 (3:1) was subsequently used to clean the samples. The samples were then etched by four different solutions to produce porous InGaN. The first electrolyte was a mixture of aqueous HF solution and absolute ethanol C2 H5 OH (1:4) by volu me with pH of 4.3. The second electrolyte was a mixture of HF solution and nitric acid HNO3 (1:4)with pH of 1.66. The third solution was potassium hydro xide KOH with pH o f 14, and the fourth electrolyte was a mixtu re of sulphuric acid H2 SO4 and H2 O2 (3:1) with pH of 4.5. In the electrochemical etching process, we used constant current density of J = 15mA/cm2 for 10min and power UV lamp (∼10W). After etching, the samples were rinsed in deionized water and dried in amb ient air. All samples were rinsed with ethanol after the etching process, and were then dried using nitrogen gas. All experimental processes were conducted at room temperature.
www.theijes.com
First, the mechanism of UV-enhanced electrochemical etching on In 0.27 Ga0.73 N layer will be is cussed, followed by growth of In 0.27 Ga0.73 N on the porous template. In 0.27 Ga0.73 N is a monocrystalline epilayer and consists of sub-grain boundaries that contain only slight twist or tilts. Discreet threading dislocations, as observed by SEM and AFM studies, define the sub-grain boundary or the colu mnar structure in In 0.27 Ga0.73 N epilayers [16]. The threading dislocations serve as traps for the photo-generated carriers, creating reg ions of reduced etch rate [17].
Fig. 1 SEM images of samples etched under different electrolytes (a) HF:C2H5OH, (b) HF:HNO3, (c) KOH and (d) H2 SO4 :H2 O2 .
Fig. 1 shows a high magnificat ion image of fabricated porous In 0.27 Ga0.73 N by scanning electron microscope (SEM). Fig. 1(a) shows a random formation, which indicates a slow reaction between HF:C2 H5 OH(1:4) and In 0.27 Ga0.73 N and this was confirmed by a lo w measured porosity with size ranges from < 20 to 25 n m. Fig. 1(b) shows new corallike pore mo rphologies, may be due to the presence of ammon ia in HF:HNO3 (1:4). Fig. 1(c) shows a high effect of KOH on In 0.27 Ga0.73 N , the sample shows some large holes somewhere and no other pores anywhere else. This could be understood in principle by assuming that avalanche breakdown starts at the weakest point where subsequently most of the current will be drawn into it. Fig. 1(d) shows that H2 SO4 :H2 O2 give well-defined layers of pores with size ranges fro m < 50 to 250 n m, which gives a clear indicat ion that the centres of the grains are attacked by a higher etch rate as co mpared to the grain boundaries, which contain the threading dislocations. Similar features have been observed in GaN [18] and GaAs [19]. Upon removal of the materials of the grains at the top layer, subsequent etching takes place at the sub-grains at the lower layer and so on, creating layered nano-pores structures. Bardwell et al. [16] reported that when the etch rate is too large the grain boundaries are etched significantly slo wer than the center of the crystals. This leads to a hexagonal, rough morphology. However, s mooth surfaces can be obtained when the etch electrolytes in the crystal centres is slow enough so the grain boundaries are etched at a sufficient electrolytes. In other word, the final morphology of the etched surface will depend on the electrolytes of the centres of the grains and the grain boundaries.
The IJES
Page 10
Enhancement of different electrolytes on Porous In 0.27 Ga 0.73N using photo-electrochemical etching
Fig. 2 AFM micrographs of samples etched under different electrolytes (a) HF:C2 H5 OH, (b) HF:HNO3 , (c) KOH and (d) H2 SO4 :H2 O2 .
Figs. 2(a)–2(c) show 3D-AFM images of the samples etched in HF:C2 H5 OH, HF:HNO3 ,KOH and H2 SO4 :H2 O2 respectively. This observation is in agreement with the SEM results in Fig. 1(a). Figs. 2(a) Figs. 2(b) Figs. 2(c) and 2(d) show the effect of porosity strongly dependent on the electrolytes and thus affecting the surface morphology of the samples. However, the root mean square (RMS) increases from 12.2 n m to 15.3 n m for the etched film using on the electrolytes of HF:C2 H5 OH and HF:HNO3 . Fig. 2(c) shows the considerable influence of the electro lyte KOH on the surface mo rphology of the etched film when the RMS increased to 19.2 n m. Fig. 2(D) using H2 SO4 :H2 O2 electrolyte on the surface morphology the RMS increased to 24.2 n m. This finding indicates that the roughness of the surface increases depend on the electrolytes. PL is one of the properties of nanostructure materials. In order to explain the broad PL band, it is assumed that the porous In 0.27 Ga0.73 N contains wide distribution of nanocrystallites with d ifferent sizes. This means that the position and shape of PL band also depend on the size d istribution of the nanocrystallites. Fig. 3 shows the room temperature photoluminescence (PL) spectra of nanoporous In 0.27 Ga0.73 N samples etched using different electrolytes. Fig. 3 shows the photoluminescence (PL) spectra of the as-grown and porous films, d isplaying the near band-edge emission of In 0.27 Ga0.73 N at a wavelength of 365 n m. The spectra were observed to be slightly b lue-shifted in the samp les etched by KOH and H2 SO4 :H2 O2 ind icating an increasing stress in the samples, while a slightly red shift observed in the samples etched by HF:C2 H5 OH and HF:HNO3 (relative to the spectrum of the as -grown sample), indicating a reduce in the stress. Similar b lue-shifted PL fro m porous In 0.27 Ga0.73 N has been reported before [22]. Appearance of the b lue shifted PL emission is may be correlated with the development of highly anisotropic structures in the morphology. Among the samples, there was a little difference in the peak position indicating that the change of porosity has a litt le influence on the PL peak shift. Ho wever, the PL peak intensity of the porous samples has increas ed compared to that of the as -grown, wh ich indicates that
www.theijes.com
the porosity has a highly influence on the PL peak intensity. The sample etched in H2 SO4 :H2 O2 shows the highest PL intensity enhancement (∼23.7) fo llo wed by that in KOH (∼13.8) and both have the highest porosities 53% and 42% respectively. The intensity of emitted light is proportional to the number of photons emitted. This means that the number of the emitted photons is much higher for porous In 0.27 Ga0.73 N than that of as-grown In 0.27 Ga0.73 N. The amp lification of porosity-induced PL intensity could be explained by the ext raction of strong PL by light scattering fro m the sidewalls of the In 0.27 Ga0.73 N crystallites. Since the surface area per unit volume is higher in porous In 0.27 Ga0.73 N, the larger surface area o f porous In 0.27 Ga0.73 N provides much more exposure of In 0.27 Ga0.73 N molecu les to the illu mination of PL excitation lights. This may result a higher number of electrons to take part in the excitation and recomb ination process in porous In 0.27 Ga0.73 N compare to the smaller surface area of as -grown In 0.27 Ga0.73 N. As a result, the number of emitted photon due to radiative reco mbination process is higher in porous In 0.27 Ga0.73 N. Additional peak like shoulder observed in the sample etched in KOH may be due to the presence of defects. Fro m the literature, many defects related to PL spectra have been reported [40]. The presence of this peak is still not clear; it could be related to exciton bound to structural defect. Ho wever, PL emissions are samp le dependent and affected by the variations of the built-in strain [23], therefore, the peak could be also due to the strain-induced structural defects, or it could be due to incorporation of impurity-induced disorder or surface defects during etching. Porous films have higher surface area per unit volume co mpared with as -grown films, and thus, the porous In 0.27 Ga0.73 N film provides much more exposure to the electrolytes of PL excitation lights for In 0.27 Ga0.73 N molecules. Th is phenomenon may result in a h igher number of electrons taking part in the excitation and recombination process in porous films co mpared with the s maller surface area of the as-grown films [18]. The relatively wide statistical size distribution of the pores can be attributed to the broadening of the linewidth of porous films.
Fig. 3 PL intensity of samples etched under different electrolytes (a) as-grown, (b) HF:C2 H5 OH, (c) HF:HNO3, (d) KOH and (e) H2 SO4 :H2 O2 .
The IJES
Page 11
Enhancement of different electrolytes on Porous In 0.27 Ga 0.73N using photo-electrochemical etching Fig. 4 shows the Raman spectra of the as grown and porous In 0.27 Ga0.73 N films. One Raman phonon mode, namely, the E1 (TO) is for bidden, this explains its absence. The spectra exhib it phononmode E2 (h igh) and A 1 (LO) and relatively s mall peaks of A 1 (TO) and E2 (lo w) [24]. While A 1 (TO) and E2 (lo w) are absent in the Raman spectra of as -grown samp le. The presence of these two peaks in porous sample shows that the change of optical properties in the porous samples has taken place, this could be attributed to the crystal disordering in the films [25]. The frequencies of all the observed modes were in good agreement with the results of other researchers [18 ,26]. Fig. 3B shows a red shift in Raman spectra on E2 (high) through all samples compared to as grown indicating that stress relaxation has taken place in the samples. The relaxation of the stress could be due to the enlargement of GaN’s lattice constant in porous structure fro m the quantum confinement effect. Using the proportionality factor of 4.2 cm−1 / GPa for hexagonal GaN [18,25] this shift corresponds to a relaxation of stress by 0.25GPa for samp le etched in KOH and in HF:HNO3 (1:4), 0.13GPa for sample etched in HF:C2 H5 OH (1:4) and in H2 SO4 :H2 O2 (3:1). In particular samp les (d) and (e) (KOH and H2 SO4 :H2 O2 ) have shown that the electrochemical etching has imp roved their crystal quality. The intensity of peaks of E2 (high) and A 1 (LO) inH2 SO4 :H2 O2 (3:1) and KO Hare observed to be higher than the other samples indicating more light interaction in these samples. Finally, photoluminescence (PL) and Raman spectroscopy were employed to demonstrate the existence of strain in GaN thin films. The strain effect was clearer in the Raman than that in the PL.
chosen suitable etching factors to enhance the structural and optical properties of thin films for optoelectronic devices. In general, the optical properties of the etched samples were improved compared to the as grown sample. The sample etched in H2 SO4 :H2 O2 (3:1) exhib ited uniform surface morphology, h ighest porosity and the highest enhancement of PL emission. Raman spectra of all the porous samples with E2 (h igh) peak has been observed to be slightly shifted to lower frequency relative to the as-grown sample, suggesting that stress relaxation has taken place in the samples.
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
Fig. 4 The Raman spectra of samples etched under different electrolytes (a) as-grown, (b) HF:C2H5OH, (c) HF:HNO3 , (d) KOH and (e) H2 SO4 :H2O2.
[21] [22]
IV.
CONCLUS IONS
[23]
The In 0.27 Ga0.73 N /Si(111) thin film with thickness of 110 n m was prepared using the PA-MBE technique, with indiu m mo le fraction of 0.27. The porous nanostructures of the In 0.27 Ga0.73 N using the UV-assisted electrochemical etching method to produce porous In 0.27 Ga0.73 N for different electrolytes. These nanostructures can open a new and promising area in ternary III-nitride materials through the
www.theijes.com
[24] [25] [26]
D. Chen, B. Liu, H. Lu, Z. Xie, R. Zhang and Y. Zheng,. Electron Device Lett., 30, 605, 2009. Z. Hassan, Y. Lee, F. Yam, K. Ibrahim, M. Kordesch, W. Halverson and P. Colter. Solid State Commun., 133, 283, 2005. F. K. Yam and Z. Hassan. Materials Letters, 63, 724, 2009. Ramizy, Z. Hassan and K. Omar. Sens. Actuators B, 155, 699(2011). F. Yam, Z. Hassan and S. Ng. Thin Solid Films, 515, 3469, 2007. Loni, L. Canham, M. Berger, R. Arens-Fischer, H. Munder, H. Luth, H. Arrand and T . Benson. Thin Solid Films, 276, 143, 1996. E. Moyen, W. Wulfhekel, W. Lee, A. Leycuras, K. Nielsch, U. Gösele and M. Hanbücken. Appl. Phys. A, 84, 369, 2006. X. Guo, T. Williamson and P. Bohn. Solid State Commun. 140,159, 2006. G. Korotcenkov and B. Cho, Crit.. Rev. Solid State Mater. Sci., 35, 1, 2010. F. Yam, Z. Hassan and A. Hudeish. Thin Solid Films, 515, 7337, 2007. F. Yam, (2007). The Study Of Gan Materials For Device Applications, in: Universiti Sains Malaysia, Malaysia. K. Beh, F. Yam, C. Chin, S. Tneh. J. Alloys Compd., 506, 343, 2010. K. Al-Heuseen, M. Hashim and N. Ali. Appl. Surf. Sci., 257, 6197, 2011. K. Al-heuseen, M. Hashim and N. Ali, (2010). Physica B, 405, 3176. O. Volciuc, E. Monaico, M. Enachi, V. Ursaki, D. Pavlidis, V. Popa and I. T iginyanu. Appl. Surf. Sci., 257, 827, 2010. J. A. Bardwell, J. B. Webb, H. T ang, J. Fraser, and S. Moisa, J. Appl. Phys. 89, 4142 , 2001. N. G. Weimann and L. F. Eastman, J. Appl. Phys. 83, 3656,1998. K. Al-Heuseena, M.R. Hashima, N.K. Ali. Applied Surface Science 257, 6197, 2011. H. Foll, J. Carstensen, S. Langa, M. Christophersen, I.M. Tiginyanu, Phys. Status Solidi (a) 197, 61, 2003. A. Vajpeyi, S. Tripathy, S. Chua and E. Fitzgerald, Physica E, 28, 141, 2005. V. Davydov, A. Klochikhin, V. Emtsev, A. Smirnov, I. Goncharuk, A. Sakharov, D. Kurdyukov, M. Baidakova, V. Vekshin and S. Ivanov, phys. status solidi B, 240 , 425, 2003. L.S. Chuah, Z. Hassan, H. Abu hassan, Mater. Sci. Poland. 26, 3, 2008. T.L. Tansley, E.M. Goldys, M. Godlewski, B. Zhou, H.Y. Zuo, (1997), The contribution of defects to the electrical and optical properties of GaN, in: S.J. Pearton (Ed.), GaN and Related Materials I, Gordon and Breach Science Publishers. M. Kuball, Surf. Interf. Anal. 31, 987, 2001. T.L.Williamson, D.J. Diaz, P.W. Bohn, R.J.Molnar, J. Vac. Sci. Technol. B 22, 925, 2004. C. Kisielowski, J. Kruger, S. Ruvimov, T. Suski, J.W. Ager, E. Jones, Z. Liliental, M.Rubin, E.R.Weber, Phys. Rev. B 54,17745, 1996.
The IJES
Page 12