Fabrication of Periodic Silicon Nanopillars in a Two-Dimensional

Mar 17, 2015 - We present a method to fabricate well-controlled periodic silicon nanopillars (Si NPs) in hexagonal arrays using silica nanosphere (SNS...
0 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/Langmuir

Fabrication of Periodic Silicon Nanopillars in a Two-Dimensional Hexagonal Array with Enhanced Control on Structural Dimension and Period Jea-Young Choi,† T. L. Alford,† and Christiana B. Honsberg*,‡ †

School for Engineering of Matter, Transport and Energy, and ‡School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona 85287, United States S Supporting Information *

ABSTRACT: We present a method to fabricate well-controlled periodic silicon nanopillars (Si NPs) in hexagonal arrays using silica nanosphere (SNS) lithography (SNL) combined with metal-assisted chemical etching (MaCE). The period of the Si NPs is easily changed by using our silica nanosphere (SNS) spin-coating process, which provides excellent monolayer uniformity and coverage (>95%) over large surface areas. The size of the deposited SNS is adjusted by reactive ion etching (RIE) to produce a target diameter at a fixed period for control of the surface pattern size after a gold metal mask layer deposition. The Si NPs are etched with the MaCE technique following introduction of a Ni interfacial layer between the Si and Au catalyst layer for adhesion and improved lithographical accuracy. The result is a fast, convenient, and large-area applicable Si surface nanolithography technique for accurate and reproducible Si NP fabrication.

■

INTRODUCTION The fabrication of Si nanostructures is important due to its potential use in advanced optical and electrical device applications. Many studies report possible applications of Si nanostructures in the fields of nanoelectronics,1,2 optoelectronics,3 photovoltaics,4−9 fuel cells,10 and chemical sensors.11 Thus, a variety of nanolithography and fabrication processes have been explored and developed. However, despite these efforts, a well-controlled and reproducible Si nanostructure fabrication method is still very challenging and requires relatively time-consuming and expensive processes (e.g., electron beam lithography) creating a barrier preventing widespread use. Therefore, research is ongoing to find a cost-effective, less time-consuming and large-area nanofabrication process that provides acceptable structural accuracy and reproducibility. As a part of this effort, various novel nanolithography and fabrication techniques have been developed and investigated. Nanoimprint lithography is one of these, and it has proven its potential as a highly accurate nanofabrication approach that also offers enhanced capability for large-area applications.12−14 However, nanoimprinting still needs an expensive and time-consuming nanolithography process to fabricate the master mold with a well-defined surface nanopattern over a large area.15 Nanosphere lithography (NSL), therefore, has been widely studied since NSL provides convenient control of the lithographical scale just by introducing different sizes of nanospheres directly on the target surface to serve as etch masks. Unfortunately, the most widely available polystyrene sphere (PS) NSL process has limitations, which are (1) insufficient etching selectivity to Si © XXXX American Chemical Society

restricting the structure aspect-ratio, (2) a nonuniform PS mask size reduction which is needed for precise dimension control of Si nanostructures in a fixed period, and (3) inadequate coverage for large area applications. To overcome the limits of the PS NSL process, we recently reported a “solvent-controlled spin-coating method” to deposit a uniform silica nanosphere (SNS) monolayer on large Si surfaces (e.g., a 4-in. round Si wafer) that produces more than 90% SNS monolayer coverage.16 Using our novel SNS spincoating method, we here introduce various sizes of SNS (i.e., 310, 600, and 840 nm in diameter) on 2-in. Si wafers with more than 95% monolayer coverage and thus provide the capability to fabricate Si nanostructures with different periods. Utilizing SNS for NSL (SNL) offers a well-defined and reproducible SNS diameter reduction using reactive ion etching (RIE) to realize various dimensions of nanostructures in a certain period. The dimension control of the Si nanostructures in a chosen period is a key consideration in fields like optoelectronics for effective light diffraction and/or scattering.17,18 The growing interest in thin-film Si solar cells has motivated efforts to develop an effective nanofabrication process with enhanced lithographical accuracy of both period and dimension. This is because controlled subwavelength scale periodic nanostructures can provide greatly enhanced light absorption in a thin Si absorber beyond the theoretical absorption limit of Si.19−21 Received: January 19, 2015 Revised: March 12, 2015

A

DOI: 10.1021/acs.langmuir.5b00128 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 1. Different size of SNS in (a) 310 nm, (b) 600 nm, and (c) 840 nm diameters, spin-coated on 2-in. Si wafers. (Top, middle, and bottom of panels a−c represent the whole 2-in. wafer with each SNS, × 2k, and ×20k magnification SEM images, respectively.) (d) Optimum SNS solution concentration vs SNS diameter for spin-coating, and (e) table representing the ratio of SNS diameter (Dia., nm) to solution concentration (Conc’, mg/mL) for spin-coating.

Figure 2. Three different sizes of SNS in (a) 310 nm, (d) 600 nm, and (g) 840 nm in diameter, and RIE etched with two different etch times. Panels (b) and (c) are 1.5 min and 2.5 min on 310 nm SNS, (e) and (f) are 4 min and 6 min on 600 nm SNS, and (h) and (i) are 6 min and 8 min on 840 nm SNS. (j) SNS diameter change vs RIE time, and (k) table showing etch rates for each SNS. The solutions were sonicated for 5 h to produce complete dispersion of the SNS. Test substrates consisted of 2-in. round polished n-type Si (100) wafers (280 μm thickness) cleaned in a piranha solution [H2SO4 (96%):H2O2 (30%) = 4:1] for 15 min to form a hydrophilic Si surface followed by a 10 min deionized (DI)-water rinse. SNSs were spin-coated under common ambient lab conditions at 2000 rpm (80 rpm/sec acceleration) for 120 s after dropping 300 μL of solution on the wafer surface. The uniformity of SNS spin-coated samples was observed by scanning electron microscopy (SEM, JEOL XL-30), and the coverage of the SNS monolayer was calculated by direct counting of the SNS covered area on the Si surface through image analysis software, “Image J” (National Institutes of Health, USA)22 as shown in our previous report.16 SNS-coated Si substrates were then transferred to the reactive ion etching chamber (Oxford PlasmaLab 80+, RIE), and the SNS size reduction was performed using a CHF3/Ar gas combination with 200 W RF plasma power at 25/25 sccm gas flow rate and 75 mTorr

Here, we report our novel fabrication process for Si NPs with control over both period and dimension. A metal-assisted chemical etching (MaCE) technique is introduced to provide a fast and low-cost etching process that achieves effective vertical etching with minimal size variation of the NPs. Various interface metal layers (Cr, Ti, and Ni) are investigated as Au adhesion layers and Ni is determined to produce enhanced lithographical accuracy for highly uniform Si NP fabrication of an intended dimension. Moreover, for MaCE, the influence of fractional catalyst coverage on etch rate and controlled vertical etching is investigated.

■

EXPERIMENTAL SECTION

The SNS solution was prepared with powders of 310, 600, or 840 nm diameter SNS (Bang Lab.), which were dispersed in N,N-dimethylformamide (DMF) (Sigma-Aldrich) at an optimized concentration. B

DOI: 10.1021/acs.langmuir.5b00128 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir chamber pressure followed by a 2 min O2 plasma clean with 100 W at 45 sccm gas flow rate and 150 mTorr chamber pressure to remove surface organic contamination resulting from the CHF3/Ar RIE. (Note: The O2 cleaning step after RIE is highly crucial for later complete SNS removal.) Metal layers for the subsequent MaCE process were deposited with an electron-beam evaporator (Lesker PVD-75); then the SNS were removed by sonication in water. The MaCE solution was prepared by mixing 2.5 vol % hydrogen peroxide (H2O2) with 10 vol % hydrofluoric acid (HF, 49%) in DIwater. Etching was performed by immersing samples into the etching solution while stirring. After MaCE, the Au layer was removed by dipping in a Au etchant for 5 min.

Figure 1a−c, photographs and SEM images of three 2-in. Si wafers are shown with three different SNS sizes: 310, 600, and 840 nm in diameter, respectively, and all have more than 95% monolayer coverage. The solution concentration for each SNS size was optimized to produce maximum SNS monolayer coverage and uniformity as shown in Figure 1a−c, and interestingly, from Figure 1d, the optimal SNS concentration just proportionally increases with the SNS diameter. From Figure 1e, the ratio of SNS diameter (Dia.) to solution concentration (Conc’) is around “2” for all three diameters of SNS. Therefore, with our spin-coating approach, uniform coatings of various sizes of SNS are achieved by simply changing the SNS solution concentration based on the SNS diameter. This provides the user with the capability to control the period (or density) of fabricating Si nanostructures. SNS Size Reduction by RIE. In addition to their period, the fabrication of Si nanostructures of a certain dimension is also highly crucial to realize certain optical, electrical, and physical phenomenon. For instance, there is growing interest in the use of Si nanostructure with an optimized period and size for solar cells. The study by Spinelli et al. has demonstrated that well-defined Si NPs with optimized period and dimension can offer enhanced forward scattering of incident light providing an improved omnidirectional antireflection effect over a broad wavelength range.17 Furthermore, the study by Sang Eon et al. emphasized the importance of Si nanostructure size (or Si filling fraction) in producing strong transmitted light diffraction for enhanced light absorption.23 Thus, effective SNS size control is important to fabricate the most ideal Si nanostructures. The advantage of using SNS for the NSL process as opposed to PS is that SNS (i.e., SiO2) offers sufficient etching selectivity to Si, which allows uniform SNS size reduction with an RIE process. Size reduction of PS is possible, but the soft material nature inevitably produces a nonuniform result; consequently, the fabricated nanostructure has a rough surface.24,25 This rough surface is especially problematic for surface dominant device application like solar cells since it generally increases the surface recombination velocity and consequently degrades solar cell performance.26 In Figure 2a−i, plan view SEM images of RIE etched SNS with different initial diameters prior to and after etching clearly reveal uniform SNS etching retaining a smooth circular shape. In Figure 2j,k, the SNS horizontal etching rates are shown to be relatively constant, ranging from 41.2−46.5 nm/min as the RIE etching time increases regardless of initial SNS size. Therefore, effective size control of the SNS mask layer can be achieved with well-defined shapes. Metal Cathode Layer Deposition for Surface Patterning. The above-described SNS nanolithography technique has been combined with MaCE to fabricate well-defined Si NPs. The MaCE process requires a metal catalyst layer to reduce the Si surface, forming SiO2, which is etched away by HF mixed in the MaCE solution. An Au cathode layer was chosen since Au offers a slower etch rate compared to other noble metals (i.e., Pt, and Ag) and thus provides controllable etching for Si NPs fabrication.27 In addition, the process with Au etches Si vertically without a porous layer formation, and avoids unexpected surface morphology changes.7 Au, however, inherently has weak adhesion to the Si surface, which is problematic in our process. After Au layer deposition, a lift off process is used to remove the SNS mask layer to avoid etching of the Si NPs related to the Au-covered SNS, as shown in

■

RESULT AND DISCUSSION SNS Monolayer Spin-Coating. Prior to Si NP fabrication, deposition of a uniform high coverage SNS monolayer must be

Figure 3. (a) Si surface with SNS etched to 611 nm average diameter (840 nm period); (b) Cr/Au, (c) Ti/Au, and (d) Ni/Au deposited Si surface after SNS removal, which have 722, 667, and 622 nm diameter of pattern, respectively. (Scale bar = 500 nm).

Figure 4. Scheme of reaction processes for metal-assisted chemical etching based on the galvanic process. ① The oxidant (i.e., H2O2) is reduced by the catalytic activity of noble metal. ② The generated holes from reduction of oxidant consequently diffuse to Si/metal interface through metal layer. ③ The oxidation of Si occurs by the injected holes, and ④ SiO2 is reacted with HF. ⑤ Finally, SiO2 is dissolved and etched away into the solution.30

realized. In our previous report,16 we have demonstrated that a hexagonal arrayed SNS monolayer with enhanced uniformity could be achieved with more than 90% surface coverage on large Si surface areas (4-in. Si wafer) by using an N,N-dimethylformamide (DMF) solvent SNS spin-coating method. In this report, we demonstrate that this spin-coating method also works well to deposit various sizes of SNS on the Si surface. In C

DOI: 10.1021/acs.langmuir.5b00128 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 5. Panels (a), (d), and (g) are RIE etched SNS with 840, 600, and 310 nm initial diameter. (b), (e), and (h) are plan view SEM images, and (c), (f), and (i) are 80°-angled SEM images for Si NPs.

Figure S1a. The SNS removal is performed with sonication, which leads to Au layer lift-off from the Si when the adhesion is poor, as shown in Figure S1b. Therefore, an Au adhesion layer must be introduced, and, here, three different interfacial metal layers, Cr, Ti, and Ni, are tested. For the test, 3 nm of each adhesion layer is deposited before a 30 nm Au layer deposition. Figure 3 shows SEM images of patterned Au layers with different adhesion layers taken before and after the 1-h sonication in DI-water for SNS removal. The diameter of the SNS before metal layer deposition was an average of 611 nm (Period: 840 nm) as shown in Figure 3a. After SNS removal by sonication, the diameter of each pattern is measured, and, as shown in Figure 3b−d with red numbers, Cr and Ti produced a significantly larger diameter, 722 and 667 nm, respectively, than the initial SNS diameter, 611 nm, but the Ni/Au metal layer combination resulted in an average pattern diameter of 622 nm and this is very close to the initial SNS size. Therefore, the Ni interfacial layer for Au offers the best lithographical accuracy for Si NPs fabrication in a desired diameter. Fabrication of Si NPs. The MaCE technique is used to fabricate Si NPs since it provides vertical etching with minimal diameter variation from top to bottom. One of the widely accepted mechanisms for the MaCE galvanic process is summarized by the two half-cell reaction below:24,27−30 Cathode reaction at the metal: H 2O2 + 2H+ → 2H 2O + 2h+

(R1-1)

2H+ + 2e− → H 2

(R1-2)

Si + 2H 2O + 2h+ → SiO2 + 4H+

(R2-1)

SiO2 + 6HF → H 2SiF6 + 2H 2O

(R2-2)

Overall reaction: Si + H 2O2 + 6HF → 2H 2O + H 2SiF6 + H 2

(R3)

The galvanic process was schematically illustrated in Figure 4 in steps 1 through 4. We found that a Ni/Au layer provided a precise control on surface pattern size and subsequent Si NPs dimension as fabricated with MaCE. In Figure 5, plan view and angled SEM images of three different dimensions of the SNS features and the resulting MaCE fabricated Si NPs are presented. Figure 5a,d,g represents the SNS etched to 600, 445, and 223 nm average diameters (Note: these diameters represent 50% Si filling fraction in the structured layer) arrayed at 840, 600, and 310 nm periods, respectively. Figure 5b,e,f are plan view SEM images and c,f,i are 80°-angled SEM images of the NPs. The red solid-arrow for each period is the identical length and indicates good lithographic accuracy. In addition to period and diameter control, a reproducible vertical etching rate for MaCE is crucial for Si NP fabrication. Therefore, the etching rates with different fractional Ni/Au catalyst coverage were investigated. Three different fractional coverages (30%, 50%, and 70%) of the Ni/Au layer were deposited as shown in Figure 6a−c and dipped in the MaCE solution for 5 min, with the results as shown in Figure 6d−f. From the height of the etched Si NP, a gradual increase of etch rate was observed as the catalyst fractional coverage increased. The plot of the etch rate versus catalyst coverage indicated that

Anode reaction at Si: D

DOI: 10.1021/acs.langmuir.5b00128 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 6. Ni/Au catalyst deposited Si surface with (a) 30%, (b) 50%, and (c) 70% fractional coverage, and (d−f) Si NPs fabricated from each catalyst fractional coverage after 5 min etching in MaCE solution. (g) Etch rates from each Ni/Au fractional surface coverage. (h−j) ∼500 nm height of Si NPs from panels (a)−(c) after dipping in the etching solution for 8 min, 6.5 min, and 5 min, respectively.

deposited SNS. Under optimized RIE conditions, a constant SNS horizontal etch rate regardless of the initial SNS size is confirmed that retains a well-defined etched SNS shape. For Si NP fabrication, MaCE is combined with the SNL process. A Au catalyst layer with Ni inserted as an adhesion layer provides enhanced lithographic accuracy of the metal mask layer. Further, we were able to observe that the measured etch rates gradually increase as the fractional coverage of catalyst increases and successfully estimate the etch time to produce a desired height of Si NPs with various diameters. As a result, well-controlled Si NPs can be fabricated in a desired period and dimension which potentially provides a powerful approach to realize enhanced optical and electrical properties of Si. Further, the SNL process could be used to fabricate nanoimprint master molds19 to extend this nanofabrication capability to different materials (e.g., InP, InGaN) rather than Si.

30%, 50%, and 70% Ni/Au fractional coverage produced 63, 77, and 104 nm/min etching rates, respectively (see Figure 6g). The faster etch rate, with the larger fractional coverage of catalyst, can be explained by the enhanced penetration of etching solution with larger catalyst coverage due to wider interspacing between etched Si NPs. Since the etch depth increases linearly with etch time for MaCE,30 the etch times were estimated for a desired height of Si NP. In Figure 6h−j, we fabricated approximately 500 nm height Si NPs having 70%, 50%, and 30% Si filling fractions in the structured layer. The Si NPs were produced after dipping in the etching solution for 8 min, 6.5 min, and 5 min, respectively, which were determined based on the measured etch rates in Figure 6g. (Note: We also confirm that our process offers a great etching uniformity over the surface area even for higher aspect-ratio Si NP as shown in Figure S2.) As a result, fabrication of Si NP in a desired period and dimension was successfully achieved.

■

■

ASSOCIATED CONTENT

S Supporting Information *

CONCLUSION An improved method for fabricating desired aspect ratio Si NPs using silica nanosphere lithography in combination with metalassisted chemical etching has been developed. Using our novel solvent-controlled spin-coating method, we have demonstrated that various sizes of SNS can be deposited with enhanced monolayer coverage (>95%) and uniformity on relatively large surface areas. Different sizes of SNS can be successfully coated simply by adjusting the SNS concentration in the solution, which is easily estimated from the diameter of the SNS. To produce a desired dimension for the SNS mask layer in a fixed period, an RIE process is used for size reduction of the

The unexpected structure damage or etch on Si NPs during MaCE by incomplete SNS removal, which are covered by a Au catalyst and a peel-off Au catalyst layer during SNS removal due to weak Au adhesion on a bare Si surface. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. E

DOI: 10.1021/acs.langmuir.5b00128 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

■

(19) Yu, Z.; Raman, A.; Fan, S. Fundamental limit of nanophotonic light trapping in solar cells. Proc. Natl. Acad. Sci. U.S.A. 2010, 107 (41), 17491−17496. (20) Yu, Z.; Raman, A.; Fan, S. Fundamental limit of light trapping in grating structures. Opt. Express 2010, 18 (103), A366−A380. (21) Wellenzohn, M.; Hainberger, R. Light trapping by backside diffraction gratings in silicon solar cells revisited. Opt. Express 2012, 20 (101), A20−A27. (22) Rasband, W., ImageJ; U.S. National Institutes of Health: Bethesda, MD, 1997−2006; http://rsbinfonihgov/ij. (23) Han, S. E.; Chen, G. Optical absorption enhancement in silicon nanohole arrays for solar photovoltaics. Nano Lett. 2010, 10 (3), 1012−1015. (24) Wang, H.-P.; Lai, K.-Y.; Lin, Y.-R.; Lin, C.-A.; He, J.-H. Periodic Si nanopillar arrays fabricated by colloidal lithography and catalytic etching for broadband and omnidirectional elimination of Fresnel reflection. Langmuir 2010, 26 (15), 12855−12858. (25) Teng, D.; Wu, L.; He, W.; Ye, C. High-density silicon nanowires prepared via a two-step template method. Langmuir 2014, 30 (8), 2259−2265. (26) Oh, J.; Yuan, H.-C.; Branz, H. M. An 18.2%-efficient blacksilicon solar cell achieved through control of carrier recombination in nanostructures. Nat. Nanotechnol. 2012, 7 (11), 743−748. (27) Li, X.; Bohn, P. Metal-assisted chemical etching in HF/H2O2 produces porous silicon. Appl. Phys. Lett. 2000, 77 (16), 2572−2574. (28) Peng, K.; Hu, J.; Yan, Y.; Wu, Y.; Fang, H.; Xu, Y.; Lee, S.; Zhu, J. Fabrication of single-crystalline silicon nanowires by scratching a silicon surface with catalytic metal particles. Adv. Funct. Mater. 2006, 16 (3), 387−394. (29) Harada, Y.; Li, X.; Bohn, P. W.; Nuzzo, R. G. Catalytic amplification of the soft lithographic patterning of Si. Nonelectrochemical orthogonal fabrication of photoluminescent porous Si pixel arrays. J. Am. Chem. Soc. 2001, 123 (36), 8709−8717. (30) Huang, Z.; Geyer, N.; Werner, P.; de Boor, J.; Goesele, U. Metal-assisted chemical etching of silicon: A review. Adv. Mater. 2011, 23 (2), 285−308.

ACKNOWLEDGMENTS This material is based upon work primarily supported by the National Science Foundation (NSF) and the Department of Energy (DOE) under NSF CA No. EEC-1041895. Any opinions, findings and conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect those of NSF or DOE. We thank Dr. Clarence Tracy for his valuable discussions on this research.

■

REFERENCES

(1) Schmidt, V.; Riel, H.; Senz, S.; Karg, S.; Riess, W.; Gosele, U. Realization of a silicon nanowire vertical surround-gate field-effect transistor. Small 2006, 2 (1), 85−88. (2) Goldberger, J.; Hochbaum, A. I.; Fan, R.; Yang, P. Silicon vertically integrated nanowire field effect transistors. Nano Lett. 2006, 6 (5), 973−977. (3) Tian, B.; Zheng, X.; Kempa, T. J.; Fang, Y.; Yu, N.; Yu, G.; Huang, J.; Lieber, C. M. Coaxial silicon nanowires as solar cells and nanoelectronic power sources. Nature 2007, 449 (7164), 885−U8. (4) Peng, K. Q.; Xu, Y.; Wu, Y.; Yan, Y. J.; Lee, S. T.; Zhu, J. Aligned single-crystalline Si nanowire arrays for photovoltaic applications. Small 2005, 1 (11), 1062−1067. (5) Fang, H.; Li, X.; Song, S.; Xu, Y.; Zhu, J. Fabrication of slantinglyaligned silicon nanowire arrays for solar cell applications. Nanotechnology 2008, 19 (25), 255703. (6) Garnett, E. C.; Yang, P. Silicon nanowire radial p−n junction solar cells. J. Am. Chem. Soc. 2008, 130 (29), 9224−+. (7) Peng, K.; Wang, X.; Lee, S.-T., Silicon nanowire array photoelectrochemical solar cells. Appl. Phys. Lett. 2008, 92 (16). (8) Hwang, Y. J.; Boukai, A.; Yang, P. High density n-Si/n-TiO2 core/shell nanowire arrays with enhanced photoactivity. Nano Lett. 2009, 9 (1), 410−415. (9) Sivakov, V.; Andrae, G.; Gawlik, A.; Berger, A.; Plentz, J.; Falk, F.; Christiansen, S. H. Silicon nanowire-based solar cells on glass: Synthesis, optical properties, and cell parameters. Nano Lett. 2009, 9 (4), 1549−1554. (10) Chan, C. K.; Peng, H.; Liu, G.; McIlwrath, K.; Zhang, X. F.; Huggins, R. A.; Cui, Y. High-performance lithium battery anodes using silicon nanowires. Nat. Nanotechnol. 2008, 3 (1), 31−35. (11) Patolsky, F.; Zheng, G.; Lieber, C. M. Fabrication of silicon nanowire devices for ultrasensitive, label-free, real-time detection of biological and chemical species. Nat. Protoc. 2006, 1 (4), 1711−1724. (12) Ganesan, R.; Dinachali, S. S.; Lim, S. H.; Saifullah, M.; Chong, W. T.; Lim, A. H.; Yong, J. J.; San Thian, E.; He, C.; Low, H. Y. Direct nanoimprint lithography of Al2O3 using a chelated monomer-based precursor. Nanotechnology 2012, 23 (31), 315304. (13) Han, K.-S.; Shin, J.-H.; Yoon, W.-Y.; Lee, H. Enhanced performance of solar cells with anti-reflection layer fabricated by nanoimprint lithography. Sol. Energy Mater. Sol. Cells 2011, 95 (1), 288− 291. (14) Hirai, Y.; Harada, S.; Isaka, S.; Kobayashi, M.; Tanaka, Y. Nanoimprint lithography using replicated mold by Ni electroforming. Jpn. J. Appl. Phys. 2002, 41 (6S), 4186. (15) Kuo, C. W.; Shiu, J. Y.; Cho, Y. H.; Chen, P. Fabrication of large-area periodic nanopillar arrays for nanoimprint lithography using polymer colloid masks. Adv. Mater. 2003, 15 (13), 1065−1068. (16) Choi, J.-Y.; Alford, T. L.; Honsberg, C. B. Solvent-controlled spin-coating method for large-scale area deposition of two-dimensional silica nanosphere assembled layers. Langmuir 2014, 30 (20), 5732− 5738. (17) Spinelli, P.; Verschuuren, M. A.; Polman, A., Broadband omnidirectional antireflection coating based on subwavelength surface Mie resonators. Nat. Commun. 2012, 3. (18) Kim, I.; Jeong, D. S.; Lee, W. S.; Kim, W. M.; Lee, T.-S.; Lee, D.K.; Song, J.-H.; Kim, J.-K.; Lee, K.-S. Silicon nanodisk array design for effective light trapping in ultrathin c-Si. Opt. Express 2014, 22 (S6), A1431−A1439. F

DOI: 10.1021/acs.langmuir.5b00128 Langmuir XXXX, XXX, XXX−XXX