Open Access
Add to BibSonomyAbstract
Al(Ga)InP subwavelength structures (SWS) were fabricated and optimized through thermally dewetted Au nanotemplate and ICP pattern-transfer. When λ< 900 nm, most AlGaInP nanostructures exhibit the reflectivity of less than 2% and insensitive to the incident angle up to 45°. When λ extends to 1800 nm, the reflectivity of less than 5% over 0°-45° is achieved in the optimized nanostructure, which benefits III-V multi-junction solar cells to improve their efficiency. Moreover, not only is such cost-effective nano-fabrication process completely compatible with the other processing of III-V solar cells, but their defined disordered SWS benefit the antireflection performance over broadband and wide view according to the comparison between the measurement and simulation results from AlGaInP SWS.
©2012 Optical Society of America
1. Introduction
The marriage of high-efficient III–V multi-junction solar cells (MJ-SCs) and advanced concentrator systems makes concentrated photovoltaics become attractive in the terrestrial solar power plants and attract increasing interests in the recent years. Conversion efficiency of more than 40% under concentrated condition has been demonstrated for III-V solar cells through material growth technique and structure optimization [1–5]. Higher efficiencies are expected for cells with more than three junctions [5,6]. However, higher reflection loss due to their high refraction index of III-V material limits their efficiency further improvement, especially for the MJ-SCs due to such devices exhibit a wide absorption spectrum ranging from 300nm to beyond 1700nm [7]. In addition, MJ-SCs are more sensitive to the reflectivity variation than single-junction cells due to their current matching requirements. The optimization of dielectric multilayer interference structures as antireflection coatings in such solar cells is therefore challenging [8].
Meanwhile, it is well known that sub-wavelength structures (SWS), including disordered SWS, can exhibit ultralow reflectivity over broadband and wide view once their high order diffraction is cut-off. Therefore, various cost-effective self-assembly techniques were employed to fabricate such nanostructures on different materials, such as Si, GaN, GaP and glasses [9–13]. However, most of them are not available for AlGaInP material as these fabrication processing are not compatible with the conventional GaAs processing. Up to date, AlGaInP broadband antireflection coating has been demonstrated by nanoimprint technology [14] and thermal dewetted Ag nanoparticles [15]. However, master template is too expensive for nanoimprint litheography to be suitable for wafer-scaled solar cell application. In addition, nanoimprint facility must be extra added. Even though thermal dewetted Ag nanotemplate is easily distributed on the whole wafer, Ag easily contaminates the chamber during the following dry etching process, by which, nano-pattern is transferred to the required semiconductor materials. In addition, thermally dewetted Pt/Pd and Ni nanotemplate have been successfully employed in nano-fabrication of Si [9] and GaN [11], however, their thermal dewetted temperatures are too high for III-V material to be suitable for AlGaInP SWS fabrication.
Recently, broadband antireflection nanostructures of GaAs and Si have been reported based on thermally dewetted Au nanotemplates [13,14], by which, wafer-scaled Au nano-pattern with size of 26-225 nm dependent on Au thickness were achieved at 500°C and excellent antireflection performance have been observed on these GaAs and Si nanostructured surface fabricated through these Au nanopattern and thereafter pattern-transfer process. The most important is that such cost-effective nano-pattern technique is completely compatible with the other process of III-V optoelectronic devices.
In this paper, nanopattern on Al(Ga)InP/ GaAs substrate were formed by the above-mentioned thermally dewetted Au technique. Thereafter, AlGaInP disordered subwavelength structures have been achieved and optimized by inductively coupled plasma (ICP) etching. Finally, excellent antireflection performance is observed over broadband and wide view, which should be widely employed in the III-V multi-junction solar cells and effectively improve their efficiencies.
2. Fabrication and characterization
AlGaInP subwavelengh structure fabrication procedure is shown in Fig. 1 . In order to obtain excellent antireflection performance over the whole MJ-SCs absorbed spectrum (300-1700 nm) on GaAs substrate, 900 nm thick Al(Ga)InP material was firstly grown on p-type GaAs substrate by MOCVD facility. In order to avoid Al(Ga)InP surface oxidization, 50 nm and 100 nm thick SiN were deposited on the AlGaInP sample surface by plasma enhanced chemical vapor deposition (PECVD). Thereafter, 10 nm thick Au was sputtered on such sample surface. In terms of Lee's reported condition [16,17], the samples were heated at 500°C by rapid thermal annealing (RTA) process in N2 environment. RTA time was fixed at 100s. As show in Fig. 2 , such 10 nm thick Au film was changed into nano-sized particles through the above-mentioned RTA process. During this process, Au thin film were agglomerated into the nanoparticles to contain the minimum value of surface energy for the whole material system. Thereafter, the underneath SiN layer was etched through Fluorine-based plasma by reaction ion etching (RIE) facility under the Au nanopattern. And then, antireflectional AlGaInP nanostructures were formed and optimized through ICP process under the SiN/Au nanomask. Finally, the residue mask of SiN and Au was removed by the wet etchant and Al(Ga)InP nano textured surfaces appeared.
Fig. 1 The schematically diagram of Al(Ga)InP subwavelength antireflection structure fabrication through thermally dewetted Au mask and ICP etching.
Fig. 2 The morphology of Au nanoparticles after thermally dewetted Au/SiN/Al(Ga)InP/GaAs structure at 500°C and 100 s.(a) top view of Au/SiN(100 nm) (b) top view of Au/SiN(50 nm) (c) side view of Au/SiN(100 nm) (d) side view of Au/SiN(50 nm).
The Au nanoparticles morphology, Al(Ga)InP etched profiles and morphologies have been characterized by cold filed-emission scanning electron microscope (SEM) (Hatachi, S4800, cold field electron gun). The specular reflectivities of all samples have been characterized by UV-VIS-NIR spectrophotometer (PerkinElmer, Lambda 750) after the residue removal.
3. Results and discussion
3.1 Nanoparticle pattern preparation
The surface morphologies of Au nanoparticles after 500°C 100 s annealing for two kinds of Au/SiN(50 or 100 nm) /Al(Ga)InP/GaAs samples are shown in Fig. 2. As all metal thermal dewetting process, Au agglomeration happens once Au thin film surface energy is larger than its interface energy and the surface energy of underlying substrate in such high temperature heating process. 10 nm thick Au film was thermally dewetted into Au nanoparticles during such RTA process as shown in [16,17]. Moreover, it is clearly shown in Figs. 2(a) and 2(c) that such RTA process cannot only change the morphology of Au film, but change the morphology of the underneath dielectric film also if such film is quite thick (such as 100 nm), which is assumed that the weak thermal conductivity and different thermal expansion coefficient of SiN material degrades the heat aggregation during thermal dewetting process. Therefore, it is necessary to deposit suitable thick SiN film to obtain high quality Au nanopatterned SiN/Al(Ga)InP/GaAs samples. For the sample with 50 nm thick SiN, the distributed Au nanoparticles and smooth SiN underlying layer has been clearly observed over the whole wafer as shown in Figs. 2(b) and 2(d). Au nanoparticles with semisphere shape, diameter of 50-150 nm, height of 70-120 nm, and density of 12.77/μm2 are formed as shown in Figs. 2(b) and 2(d), which is suitable for the following antireflection SWS fabrication.
3.2 Al(Ga)InP anitreflection subwavelength structure fabrication
After Au nanoparticle mask was successfully transferred onto the underlying SiN layer through Fluorine-based plasma etching, Al(Ga)InP material was etched based on Cl2/N2/Ar = 10/15/2 sccm plasma at low pressure as did in [14]. As shown in Fig. 3 , nanocone structures with diameter of about 150 nm and height of about 300 nm have been formed on the sample surface after employing ICP etching 3 min. Their anti-reflection performance is shown in Fig. 4 . As a reference, the reflectivity of bulk Al(Ga)InP/GaAs material is also added. Clearly, the sample with SWS surface exhibits lower reflectivity, especially in the short wavelength domain (λ<900 nm), even though Al(Ga)InP/GaAs has extremely high refractive index in the short wavelength domain, which indicates that such nanocone structure is quite closely distributed and their diameters are small enough to cut off their high-order diffraction and suppress the reflection of the short wavelength range. However, the reflectivity of this sample with such nanocone structures increases with the incident wavelength and finally up to 19% when λ = 1500 nm, which indicates that such nanocone height is not high enough to suppress the reflection of longer wavelength derived from their evanescent-wave coupling. Therefore, the SWS with higher aspect ratio are required to achieve low reflection over broadband (300-1800 nm).
In order to obtain the Al(Ga)InP SWS with higher aspect ratio, Ar flow rate increases to 3.5 and 4 sccm is adopted in the ICP pattern-transfer process. Their etched profiles are shown in Fig. 5 . It is clearly observed both the nanocone height and their diameter increases with the Ar flow rate. As expected, the aspect ratio of the plasma etched nanocone structure increases with Ar flow rate. And the value of about 2.3 and 3.2 for Ar flow rate of 3.5 and 4 sccm is observed as shown in Figs. 5(a) and 5(b). The specular reflectivities of these samples have been measured at oblique angle up to 45° and shown in Fig. 5(c). Compared with Fig. 4, the reflectivity reduction is clearly observed for any samples over the whole spectrum especially for the long wavelength domain, due to more Ar gas enhancing nanostructure height. When λ <900 nm, all etched Al(Ga)InP nanocone samples exhibit the extremely low reflectivity of less than 2%. When λ >1500 nm, their reflectivity decreases with the nanostructure height and their aspect ratio, where all their diameters are within subwavelength domain, but the nanostructure height is more or less than the effect characteristic length, which is related with the incident wavelength and incident angle [18]. Moreover, the higher the aspect ratio, the broader the wavelength span of the reflectivity insensitive to the incident angle. For Ar = 4 sccm etched sample, the wavelength span of the reflectivity insensitive to the incident angle extends to 1400nm. When incident wavelength is less than 900 nm, the reflectivities of these two samples are less than 2% over the wide view up to 45°, which benefits GaInP/GaAs solar cells to improve the conversion efficiency and operate without sun-tracking system.
Fig. 5 Al(Ga)InP subwavelength structure profile etched by reaction gas (a) Cl2/N2/Ar = 10/15/3.5; (b) Cl2/N2/Ar = 10/15/4; (c) the specular reflectivity of the above two types of samples.
In order to further optimize the Al(Ga)InP nanostructure profile, Ar = 4 sccm etching processes with etching time of 3 min and 4 min are employed. And the etched Al(Ga)InP SWS profiles are shown in Figs. 6(a) and 6(b). Even though the similar slope profile can be observed in the images of Fig. 6, the nanostructure in Fig. 6(a) exhibits flatter top, looser distribution, and lower height, compared with the nanostructure in Fig. 6(b), due to its short etching time and Au nanopattern mask, which indicates that longer etching time is beneficial to forming the nanocone structure and enhancing the graded variation of effective refractive index from air to Al(Ga)InP/GaAs material. Finally, dense nanocone forest as shown in Fig. 6(b) has been achieved through the optimized recipe etching 5 min. And the photo of the whole sample (1/2 2 inch wafer) is shown in inset of Fig. 6(b). Their specular reflectivities measured at different incident angle are shown in Fig. 7 . As expected, the specular reflectivity decreases much with the etching time, especially in the longer wavelength domain. For the sample with the optimized subwavelength structure, low reflectivity of less than 5% is observed over the whole spectrum (200-2000 nm) when the incident angle is less than 30°. When the incident angle increases to 45°, the reflectivity of less than 5% can be achieved over the wavelength span of 200-1800 nm. The reflectivity insensitive to the incident angle up to 45° is observed in this material when the incident wavelength span is 200-1400 nm, which means sun-tracking system should not be required for current GaInP/GaAs/GaInAs triple-junction solar cells provided that such Al(Ga)InP SWS window is successfully employed. Moreover, when the incident wavelength is less than 900 nm, most AlGaInP SWS samples exhibit the reflectivity of less than 2% and insensitive to the incident angle up to 45° as shown in Fig. 4, Fig. 5(c), Figs. 7(a) and 7(b), which means the Al(Ga)InP SWS suitable for the GaAs/GaInP dual-junction solar cells have a big fabrication tolerance and should be easy and effectively to improve their conversion efficiency. Moreover, such fabrication process is cost-effective and compatible with the other processing of III-V solar cells, which benefits it is employed in the process line of III-V solar cells to improve their efficiency.
Fig. 6 Al(Ga)InP subwavelength structure profile etched through the optimized recipe, but different time (a) t = 3 min; (b) t = 5 min.
Fig. 7 The specular reflection spectra measured over wide view for the samples etched (a) t = 3 min and (b) t = 5 min.
3.3 The comparison between the measurement and simulation results
In order to better understand the behavior of such Al(Ga)InP SWS, the simulation has been conducted based on rigorous coupling-wave analysis method. The geometry parameters were extracted from Fig. 6(b) and averaged as diameter of 205.8 nm and height of 668 nm. Moreover, the close-packed periodic nanocone profile has been adopted in the simulation in terms of the whole profile of Fig. 6(b) and its large area view. Both the measurement and the simulation results at incident angle of 8° and 45° are shown in Fig. 8 .
Fig. 8 The measured and simulated reflection spectra comparison for Al(Ga)InP SWS as shown in Fig. 6(b); (a) incident angle is 8°, (b) incident angle is 45°.
As shown in Fig. 8, each figure is divided into three sections. In the 1st section of each figure, the simulation results are well consistence with their measurement partner. Compared with the measurement results, there is some undulation in the simulated curve, especially in the Fig. 8(b), which is due to the coherent coupling of their high-order diffraction in such periodic nanocone structures as discussed in [18,19]. That is to say, disordered SWS benefit to obtain flatter reflection in such domain. In the 2nd section, back-reflection from GaAs substrate makes contribution to the measured reflection, meanwhile, high-order diffraction is gradually cut off and reduces the simulated reflection. As a result, the difference between the simulated and measured results increases and finally up to 3.3% as shown in Figs. 8(a) and 8(b). From another point of view, such analysis indicates that the real surface reflection of our fabricated SWS is lower than the measured one in this wavelength span. In the 3rd section, even though back-reflection still makes contribution to the measured results, Figs. 8(a) and 8(b) shows that the simulated value gradually approaches the measured one and then surpasses it, where evanescent-wave coupling of zero-order diffraction increases and makes stronger contribution to the surface reflection. Such phenomenon further indicates that disordered SWS can effectively degrade evanescent-wave coupling of zero-order diffraction and reduce the surface reflection in this long-wavelength domain and wide view. Overall, our fabricated disordered SWS benefit to achieve broadband quasi-omnidirectional antireflection performance, compared with the periodic partner.
4. Conclusion
In summary, Al(Ga)InP SWS with different profile have been fabricated and optimized through thermally dewetted Au nano-template and thereafter ICP etching. Combined with their reflection performance, the optimization of their subwavelength profile is performed and discussed. The reflectivity of less than 5% over 200-1800 nm and wide view up to 45° has been achieved in the optimized SWS. Such broadband and quasi-omnidirectional low reflection loss should benefit the III-V MJ-SCs to improve their conversion efficiency and operate without sun-tracking system. Especially when λ<900 nm, the extremely low reflectivity of less than 2% over the incident angle up to 45° is achieved for this GaAs absorption domain, which is much better than the conventional antireflection coating based on interference effect and makes it enhance the conversion efficiency of GaAs/GaInP solar cells once such SWS are well passivated. In addition, the comparison between the simulated reflection spectrum and the measured partner further indicates that such disordered SWS is more suitable for broadband and quasi-omnidirectional antireflection operation than its periodic partner due to they can effectively destroy the high-order diffraction coupling in the short-wavelength span and evanescent-wave coherent coupling in the long-wavelength span.
Acknowledgments
This work is funded by National Basic Research Program (Grant No. 2007CB936701), the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry, the Jiangsu Province Project (No. BE2009056), and Suzhou City Project (No. SG201020).
References and links
1. L. D. Partain, M. S. Kuryla, R. E. Weiss, R. A. Ransom, P. S. Mcleod, L. M. Fraas, and J. A. Cape, “26.1% solar cell efficiency for Ge mechanically stacked under GaAs,” J. Appl. Phys. 62(7), 3010–3015 (1987). [CrossRef]
2. K. A. Bertness, S. R. Kurtz, D. J. Friedman, A. E. Kibbler, C. Kramer, and J. M. Olson, “29.5% efficient GaInP/GaAs tandem solar cells,” Appl. Phys. Lett. 65(8), 989–991 (1994). [CrossRef]
3. R. R. King, D. C. Law, K. M. Edmondson, C. M. Fetzer, G. S. Kinsey, H. Yoon, R. A. Sherif, and N. H. Karam, “40% efficient metamorphic GaInP/GaInAs/Ge multi-junction solar cells,” Appl. Phys. Lett. 90(18), 183516 (2007). [CrossRef]
4. W. Guter, J. Schone, S. P. Philipps, M. Steiner, G. Siefer, A. Wekkeli, E. Welser, E. Oliva, A. W. Bett, and F. Dimroth, “Current matched triple-junction solar cell reaching 41.1% conversion efficiency under concentrated sunlight,” Appl. Phys. Lett. 94(22), 223504 (2009). [CrossRef]
5. M. Stan, D. Aiken, B. Cho, A. Cornfeld, V. Ley, P. Patel, P. Sharps, and T. Varghese, “High-efficiency quadruple junction solar cells using OMVPE with inverted metamorphic device structures,” J. Cryst. Growth 312(8), 1370–1374 (2010). [CrossRef]
6. C. H. Henry, “Limitation efficiencies of ideal single and multiple energy gap terrestrial solar cells,” J. Appl. Phys. 51(8), 4494–4500 (1980). [CrossRef]
7. W. Guter, J. Schone, S. P. Philipps, M. Steiner, G. Siefer, A. Wekkeli, E. Welser, E. Oliva, A. W. Bett, and F. Dimroth, “Current-matched triple junction solar cell reaching 41.1% conversion efficiency under contrated sunlight,” Appl. Phys. Lett. 94(22), 223504 (2009). [CrossRef]
8. M. F. Schubert, F. W. Mont, S. Chhajed, D. J. Poxson, J. K. Kim, and E. F. Schubert, “Design of multilayer antireflection coatings made from co-sputtered and low-refractive-index materials by genetic algorithm,” Opt. Express 16(8), 5290–5298 (2008). [CrossRef] [PubMed]
9. Y. Lee, K. Koh, H. Na, K. Kim, J.-J. Kang, and J. Kim, “Lithography-free fabrication of large area subwavelength antireflection structures using thermally dewetted Pt/Pd alloy etch mask,” Nanoscale Res. Lett. 4(4), 364–370 (2009). [CrossRef] [PubMed]
10. Y. Kanamori, K. Hane, H. Sai, and H. Yugami, “100nm period silicon antireflection structures fabricated using a porous alumina membrane mask,” Appl. Phys. Lett. 78(2), 142–143 (2001). [CrossRef]
11. C. H. Chiu, P. C. Yu, H. C. Kuo, C. C. Chen, T. C. Lu, S. C. Wang, S. H. Hsu, Y. J. Cheng, and Y. C. Chang, “Broadband and omnidirectional antireflection employing disordered GaN nanopillars,” Opt. Express 16(12), 8748–8754 (2008). [CrossRef] [PubMed]
12. S. L. Diedenhofen, G. Vecchi, R. E. Algra, A. Hartsuiker, O. L. Muskens, G. Immink, E. P. A. M. Bakkers, W. L. Vos, and J. G. Rivas, “Broadband and omnidirectional antireflection coatings based on semiconductor nanorods,” Adv. Mater. 21(9), 973–978 (2009). [CrossRef]
13. Y. F. Li, J. H. Zhang, S. J. Zhu, H. P. Dong, F. Jia, Z. H. Wang, Z. Q. Sun, L. Zhang, Y. Li, H. B. Li, W. Q. Xu, and B. Yang, “Biomimetic surfaces for high-performance optics,” Adv. Mater. 21(46), 4731–4734 (2009).
14. J. Tommila, V. Polojarvi, A. Aho, A. Tukiainen, J. Viheriala, J. Salmi, A. Schramm, J. M. Kontio, A. Turtiainen, T. Niemi, and M. Guina, “Nanostructured broadband antireflection coatings on AlInP fabricated by nanoimprint lithography,” Sol. Energy Mater. Sol. Cells 94(10), 1845–1848 (2010). [CrossRef]
15. Y. M. Song, E. S. Choi, J. S. Yu, and Y. T. Lee, “Light-extraction enhancement of red AlGaInP light-emitting diodes with antireflective subwavelength structures,” Opt. Express 17(23), 20991–20997 (2009). [CrossRef] [PubMed]
16. J. W. Leem, J. S. Yu, Y. M. Song, and Y. T. Lee, “Antireflective characteristics of disordered GaAs subwavlength structures by thermally dewetted Au nanoparticles,” Sol. Energy Mater. Sol. Cells 95(2), 669–676 (2011). [CrossRef]
17. J. W. Leem and J. S. Yu, “Broadband and wide-angle antireflection subwavelength structures of Si by inductively coupled plasma etching using dewetted nanopatterns of Au thin films as masks,” Thin Solid Films 519(11), 3792–3797 (2011). [CrossRef]
18. S. J. Wilson and M. C. Hutley, “The optical properties of ‘Moth eye’ antireflection surfaces,” Opt. Acta (Lond.) 29(7), 993–1009 (1982). [CrossRef]
19. R. Y. Zhang, B. Shao, J. R. Dong, J. C. Zhang, and H. Yang, “Absorption enhancement analysis of crystalline Si thin film solar cells based on broadband antireflection nanocone grating,” J. Appl. Phys. 110(11), 113105 (2011). [CrossRef]
