Transparent conductive silver nanowire electrodes with high resistance to oxidation and thermal shock

Abstract

Oxidation is the main issue for silver-nanowire electrodes when bulk silver is reduced to nano-scale size, restricting its practical application due to the significant rise in sheet resistance. We report a simple method by virtue of incorporating polyethoxysiloxane (PES) into a silver-nanowire layer to enhance resistance to oxidation and thermal shock. As a result of the PES incorporation, the temperatures at which oxidation occurs were raised from 250 and 300 °C to 300 and 500 °C under air and N₂ atmospheres, respectively. Moreover, after PES incorporation the silver nanowires came into tighter contact at the cross connections due to sol–gel shrinkage, resulting in a nearly 2-fold enhancement in the conductivity and superior bend-test performance compared to the pristine AgNW electrodes. We used the silver-nanowire electrodes incorporated with PES to prepare a G/F/F-stack dual-touch sensor, revealing a rapid response and stable performance for more than 3 months.

---

1. Introduction

Transparent conductive electrodes are key components of optoelectronic products, such as touch panels, liquid crystal displays, solar cells, and organic light emitting diodes. The most commonly used material in the manufacture of transparent conductive electrodes is indium tin oxide (ITO), as a result of its low electrical resistance and high transparency. As optoelectronic products have become popular and cheap, low-cost or new functional materials for transparent conductive electrodes have received increasing attention and have been sought over the last two decades. Several new materials, including carbon nanotubes,^1,2 graphene nanosheets,³ silver-particle grids⁴ and silver nanowires (AgNWs),^5,6 have been investigated widely as potential replacements for ITO in transparent conductive electrodes to solve the major problems: inflexibility and the limited supply of indium.^7,8 Moreover, ITO coated on plastic substrates possesses poor conductivity due to being incapable of high-temperature annealing, restricting the development of ITO plastic films. Although all of these candidates possess low-cost and flexible performance, silver materials show the best ratio of DC conductivity to optical conductivity (σ_DC/σ_OP),⁹ which is a key figure of merit for transparent conductive electrodes. Very recently, transparent conductive electrodes made from AgNWs and metal mesh have been applied practically in touch panels by touch-sensor manufacturers, such as TPK Film, O Film, and Young Fast Optoelectronics.

A number of methods have been reported to successfully fabricate AgNWs, for example, template,^10,11 hydrothermal,^12,13 low-temperature growth,¹⁴ thermally-induced formation,¹⁵ microwave assistance^16,17 and photo-reduction,¹⁸ but the salt-mediated polyol method reported by Xia’s group^19–23 is the most impressive and popular because of its simplicity and effectiveness. In the method, the aspect ratio of AgNWs depends strongly on the kinds of inorganic salts,^24,25 ratio of salt to silver nitrate,²⁶ molecular weight of PVP (polyvinylpyrrolidone),²⁷ ratio of PVP to silver nitrate,²⁵ reaction temperature,^26,28 reaction time,²⁹ atmosphere,^30,31 and reagent-addition procedure.^23,28,32 Based on these parameters, many efforts have been exerted to control the length and diameter of AgNWs to enhance performance in the application of transparent conductive electrodes.^29,33 However, some critical issues such as low adhesion on substrates and high junction resistance between wires prevent AgNWs from practical application. A simple method to improve the AgNW adhesion is to cover AgNWs with a resin^6,34 or stick/bury AgNWs on the surface of a resin layer.^9,35 As to the junction issue, the post-thermal treatment at ca. 200 °C has been used to improve the conductivity of AgNW electrodes. Lee et al.³⁶ supposed that the reduction in junction resistance was attributed to the flowing and partial decomposition of PVP which allowed AgNWs to contact tightly and fuse together; too-long thermal-treatment time or too-high thermal-treatment temperature would result in AgNWs becoming disconnected droplets, leading to an increase in sheet resistance. Recently, some studies showed that the network-junction resistance could be reduced through sol–gel shrinkage to make AgNWs with tighter contacts³⁷ or galvanic displacement of Au to connect the wires.³⁸ These methods significantly improve the electrical conductivity of transparent conductive electrodes of AgNWs, providing performance comparable to that of commercial ITO films. However, it remains a challenge to develop a facile approach to simultaneously solve the issues of adhesion and junction resistance.

In this study, we incorporated polyethoxysiloxane (PES) into an AgNW network and showed significant improvement to both mechanical properties and electrical conductivity. Moreover, resistance to oxidation at elevated temperature increases remarkably, extending applicability in practical products. The effects of PES incorporation on surface-chemical information, mechanical properties, and optical performance of AgNWs were further investigated and analyzed.

2. Experimental

2.1 Materials

Silver nitrate (AgNO₃), PVP (MW: 40 000), and ethylene glycol (EG) were purchased from Sigma-Aldrich. PES (Shin-Etsu), tetrabutyl titanate (TBT, Fluka), sodium chloride (Katayama Chem., 99.5%) and 2,4-pentanedione (Alfa Aesar) were used as received. Reagent-grade solvents and deionized water (DI water, >18 MΩ cm) were used throughout the experiments.

2.2 Preparation of AgNWs

AgNWs were synthesized using a two-step polyol reduction method as reported in the previous study.³⁹ Briefly, 36 mL of 0.3 M PVP and 80 μL of 0.2 M NaCl EG solution were mixed in a reaction vessel and heated at 160 °C. 20 μL of 1 M AgNO₃ EG solution was then added into the mixture. After 5 min, a further charge of 1 M AgNO₃ (4 mL) was added slowly using a peristaltic pump; when the color of the solution turned to a misty auburn, the residual AgNO₃ solution was added into the vessel immediately. After the color of the solution became silver-whitish, the as-prepared AgNWs were washed three times with ethanol through centrifugation. The AgNWs had lengths of 10–15 μm and diameters of 60–110 nm.

2.3 Preparation of AgNW layer and thermal treatment

Glass or polyethylene terephthalate (PET) substrates were washed with ethanol and cleaned using an O₂ plasma cleaner (PDC-23G, Harrick Plasma). AgNW solution (2.3 mg mL⁻¹) was spin-coated on a treated substrate and then dried at room temperature. The AgNW-coated substrates were placed under a N₂ or an air atmosphere at various temperatures (200, 250, 300, 400, and 500 °C) for 20 min.

2.4 PES incorporation

3 g of 0.1 M hydrochloric acid was added into 30 g of PES ethanol solution with various PES contents (0.16, 0.24, 0.38, 1.6, 3.2, 4.8, 6.24, 8, 11.2 wt%). The mixing solution was heated at 70 °C for 3 h, and then cooled to room temperature. The AgNW-coated substrates were dipped in the mixing solution, and subsequently raised immediately from the dipping bath. Upward and downward dipping rates in all experiments were 110 mm min⁻¹. The coated samples were placed in an oven at 70 °C for 10 min and then at 100 °C for 3 h to post-cure the PES.

In order to compare the effect of the incorporation treatment, a treatment was also prepared using TiO₂ as suggested elsewhere.³⁷ TBT and 2,4-pentanedione were mixed in ethanol solution with a weight ratio of 8 : 1. The total mixture solution was 30 g with various TiO₂ contents (0.16, 0.24, 0.38, 1.6, 3.2, 4.8, 6.24, 8, 11.2 wt%). To the solution was added a drop of 61% nitric acid, and then it was heated at 70 °C for 3 h. The coating procedure is the same as done for PES incorporation.

2.5 Characterization

The morphologies of the AgNWs and the PES incorporation were examined using an optical microscope (M835, M&T Optics) and a field-emission scanning electron microscope (JSM-7401F, JEOL). The elemental analysis and binding energy of the surface of the AgNWs were determined using X-ray photoelectron spectroscopy (XPS) (PHI 5000 VersaProbe, ULVAC-PHI) at a take-off angle of 90°. The sheet resistances of the transparent conductive electrodes were determined using a four-pin probe meter (Loresta-GP, Mitsubishi Chemical) with an MCP-T610 probe. The transmittance of each sample was measured at four points using an UV-Vis spectrophotometer (Lambda 850, Perkin-Elmer). The anti-scratch properties of the AgNW electrodes were determined using a motorized abrasion tester (Model 339, Fu-Chien Enterprise Co) with a cotton mat for 10 rounds in a 400 g load. For a bend test, a thin-film sample of 1 cm × 5 cm was made on the PET substrate. It was regarded as a cycle in the bend test each time the short edge of the sample was bent to touch the other short edge and then relaxed until the sample was flat.

3. Results and discussion

The pristine AgNW electrodes show a trade-off characteristic that sheet resistance increases with transmittance. As shown in Fig. 1, the performance can be improved by thermal treatment, which may lead to the flowing or partial decomposition of PVP and therefore reduces the junction resistance among AgNWs. Considering the need for gathering basic information on AgNW electrodes for practical applications, performing a systematic analysis is highly desirable. To this end, the AgNW electrodes were treated under N₂ and air atmospheres at various temperatures. As shown in Fig. 2, the sheet resistance of AgNW electrodes can be reduced by thermal treatment with the temperature below 250 °C (air) and 400 °C (N₂). With increasing temperature, the sheet resistance shows a significant reduction, reaching a minimum value at 250 and 300 °C under air and N₂ atmospheres, respectively, but then increases upon increasing the temperature further. No matter what kind of atmosphere is used, the sheet resistance of the AgNW electrodes increases remarkably at elevated temperatures. However, the increase in sheet resistance is more moderate under an N₂ atmosphere than an air one. Fig. 3 displays the SEM images of the AgNW electrodes under air and N₂ thermal treatments at various temperatures. Under an air atmosphere, the AgNWs start to be broken down into nanorods at 250 °C and are further granulated into nanoparticles with rising temperature, whereas the AgNWs still maintain their integrity below 400 °C under a N₂ atmosphere. The SEM observation is in good agreement with the results for sheet resistance. Therefore, the rise of sheet resistance induced by thermal treatment may result from the breakdown of AgNWs.

Fig. 1 Variation of the sheet resistance of the AgNW electrodes as a function of transmittance.

Fig. 2 Variation of the relative sheet resistance of the AgNW electrodes as a function of the thermal-treatment temperature under various atmospheres. R₀ is the initial sheet resistance, and R is the sheet resistance after thermal treatment.

Fig. 3 SEM images of the AgNW electrodes treated thermally under (a–e) air and (f–j) N₂ atmospheres at various temperatures: (a and f) 200 °C, (b and g) 250 °C, (c and h) 300 °C, (d and i) 400 °C, and (e and j) 500 °C.

Although the melting point of silver declines with decreasing particle size, why are there different results for air and N₂ atmospheres? To determine the cause, the chemical compositions on the surface of AgNWs under different conditions of thermal treatment were analyzed (Table 1). Compared to other conditions, thermal treatment in air results in the oxygen content on the surface of AgNWs increasing remarkably and the binding energy of Ag(3d_5/2) shifting from 368.2 to 367.2 eV (Fig. 4), where the negative binding-energy shift indicates that the metallic state Ag⁰ is oxidized to Ag²⁺ due to the non-electronegativity effect.^40–42 We speculate the reason that the AgNWs are broken more easily under air than N₂ is mainly because of the oxidation of AgNWs. In fact, while the electrodes are exposed to an air atmosphere at room temperature, even without thermal treatment, the sheet resistance of pristine AgNW electrodes increases gradually (Fig. 5) and their color becomes dark brown (Fig. 6).

Table 1 XPS analysis of the surfaces of the AgNWs thermally treated at 300 °C under different atmospheres

Samples	Compositions, at%		Ag(3d5/2), eV
	O	Ag
Pristine AgNWs	10.2	89.8	368.2
AgNWs under air atmosphere	28.1	71.9	367.2
AgNWs under N2 atmosphere	15.7	84.3	368.2

---

Fig. 4 XPS spectra of Ag_3d for the AgNW electrodes treated in various atmospheres.

Fig. 5 Variation of the relative sheet resistance of the AgNW electrodes with various extents of PES incorporation exposed to air at room temperature over time. R₀ is the initial sheet resistance, and R is the sheet resistance after exposure over time.

Fig. 6 Photographs of AgNW (left) and 3.2 wt% PES–AgNW (right) electrodes exposed to air at room temperature over 15 days.

In order to prevent oxidation and improve the poor mechanical properties, we incorporated PES into the AgNW layer (PES–AgNW). We found that the PES incorporation significantly improved the resistance to oxidation/thermal attack on the AgNW electrodes, making them thermally stable at temperatures up to 300 °C and 500 °C for air and N₂, respectively (Fig. 7). Table 2 shows the electrical properties and the anti-scratch properties of the AgNW electrodes incorporated with various PES contents. With increasing PES content (over 3.2 wt%), the PES–AgNW electrodes show superior anti-scratch properties (Table 2) and resistance to oxidation (Fig. 6) compared to the pristine AgNW electrodes. Most notably, PES incorporation also reduces sheet resistance by 1–2 orders of magnitude, similar to the effect of thermal treatment. The significant reduction in sheet resistance does not occur with resin incorporation.^6,9 As can been seen in the cross-sectional SEM images of the AgNW electrodes (Fig. 8), the AgNWs come into closer contact at the cross connections after the PES incorporation, which may be caused by sol–gel shrinkage. The closer contact may lead to lowering of the resistance at the junctions of the AgNWs to enhance the conductivity of the AgNW electrodes, which is not observed in the carbon-nanotube electrodes.⁸ It has been reported that TiO₂ also induces sol–gel shrinkage to improve the connection of AgNWs (TiO₂–AgNW).³⁷ Comparing PES with TiO₂, Table 2 shows that both incorporations have a similar effect on the enhancement of the conductive and mechanical properties . However, TiO₂ content that is too high results in a rise in electric resistance and a decline in anti-scratch performance. Fig. 9 shows the surface morphologies of PES–AgNW and TiO₂–AgNW. In low solid content, PES forms a uniform layer, whereas TiO₂ appears as granulated particles. When the solid content is high, obvious cracks appear in the TiO₂ layer and cracks even break the AgNWs (Fig. 9g and h). This might explain why the electrical and mechanical properties become poorer with high TiO₂-content incorporation.

Fig. 7 Variation of the relative sheet resistance of the 3.2 wt% PES–AgNW electrodes as a function of the thermal-treatment temperature under various atmospheres. R₀ is the initial sheet resistance, and R is the sheet resistance after thermal treatment.

Table 2 Variation of the relative sheet resistance of the AgNW electrodes after various incorporations and anti-scratch testing^b

Sample no.		1	2	3	4	5	6	7	8	9
a The AgNW layer was wiped off.b R0: the surface resistance of the pristine AgNW sample. R: the surface resistance of the AgNW sample incorporated with PES or TiO2 before the anti-scratch test. RA: the surface resistance of the AgNW sample incorporated with PES or TiO2 after the anti-scratch test.
Concentration of coating solution (wt%)		0.16	0.24	0.38	1.6	3.2	4.8	6.24	8	11.2
PES incorporation	R/R0	0.058	0.181	0.075	0.015	0.123	0.085	0.063	0.089	0.039
	RA/R0	—^a	—^a	—^a	0.543	0.142	0.083	0.066	0.097	0.041
TiO2 incorporation	R/R0	0.090	0.152	0.041	0.138	0.219	0.206	0.051	0.056	1.948 × 10^4
	RA/R0	—^a	—^a	—^a	1.499	0.220	0.189	0.047	0.063	1.139 × 10^9

---

Fig. 8 Cross-sectional SEM images of the AgNW electrodes (a) without and (b) with PES incorporation.

Fig. 9 SEM images of the AgNW electrodes dipped in various (a–d) PES and (e–h) TiO₂ concentrations: (a and e) 0.16 wt%, (b and f) 1.6 wt%, (c and g) 8 wt%, and (d and h) 11.2 wt%. The inset image in (g) and the arrows in (f) indicate the broken AgNWs.

In a previous study of CNT electrodes,⁸ a near 100 nm incorporation layer whose thickness conforms to the requirement of an anti-reflection structure will reduce reflectance and consequently increase transmittance. According to Fresnel theory, to achieve optimal transmittance with respect to the incident wavelength λ, the incorporation layer of thickness d (if its extinction coefficient is negligible) should satisfy the criteria: d = λ/4n_c and , where n_c, n₀, and n_s are the refractive indices of the overcoat, air, and substrate, respectively.^43,44 In the present case, the value of substrate (glass) is near 1.5, and thus the optimal refractive index of the incorporation layer is about 1.22; refractive index that is too low or too high will lead to high reflectance and low transmittance. The corresponding optimal coating thickness is near 113 nm if the incident wavelength is 550 nm. For most optical materials, their refractive indices are greater than 1.22, and thereby low-refractive-index-material hybrids or homogenous pore formation (n₀ = 1) have been widely utilized to reduce the refractive index. In contrast, the refractive index of silver is a function of the incident wavelength and is very low in the visible-light region (0.07–0.09; much lower than that of air).⁴⁵ For an incorporated material having a higher refractive index, it is easier to make the incorporation layer reach the optimal value. The refractive index of TiO₂ is much higher than that of PES. However, Fig. 10 shows that the electrodes incorporated with PES reveal superior transmittance compared to those incorporated with TiO₂. It may be explained by the fact that numerous air voids among TiO₂ particles (Fig. 9f–h) deteriorate the light transmission due to refractive-index reduction. Besides, a great difference between the refractive index of TiO₂ and that of AgNWs will induce more significant light scattering compared to the PES–AgNWs. Observing Fig. 8b, the diameter of AgNWs is comparable to the layer thickness or the optimal thickness. Therefore, the effective homogenous model is inappropriate for the present cases and the light scattering is inevitable. It may explain that, regardless of what the composition is, both PES and TiO₂ incorporations on AgNW electrodes decrease the transmittance (Fig. 10).

Fig. 10 Transmission spectra of the (a) PES–AgNW and (b) TiO₂–AgNW electrodes made from various dipping solution concentrations.

Based on the aforementioned results, the PES incorporation exhibits superior characteristics in electrical, optical, and mechanical aspects compared to the TiO₂ incorporations. A bend test was also conducted to investigate the effect of PES incorporation on the flexible application. Fig. 11 shows that the sheet resistance of the PES–AgNW electrode remained almost unchanged at the end of 500 cycles, whereas that of the pristine AgNW electrode (without PES incorporation) increased nearly 3 times. The result indicates that PES solidifies the AgNWs and thus increases the reliability of electrodes. We used the PES–AgNW electrodes (transmittance = 89%; sheet resistance = 100 Ω sq⁻¹) to prepare a glass/film/film-stack (G/F/F-stack) dual-touch sensor, as shown in Fig. 12. Two PES–AgNW PET films were stacked under a cover glass for a capacitive sensor. The circuit patterning of the PES–AgNW electrodes was carried out by laser etching. The as-prepared touch sensor was evaluated using the eGalaxTuner software (see ESI†). The touch sensor reveals a rapid response and stable performance for more than 3 months.

Fig. 11 Variation of the relative sheet resistance of the pristine AgNW electrode and the 3.2 wt% PES–AgNW electrode as a function of bending cycles. R₀ is the initial sheet resistance, and R is the sheet resistance after the bending test.

Fig. 12 Photograph of the touch sensor made from the PES–AgNW electrodes.

4. Conclusion

We have fabricated highly conductive PES–AgNW electrodes by virtue of incorporating PES into AgNWs. The PES incorporation results in a nearly 2-fold enhancement in conductivity compared to the pristine AgNWs electrodes. The improvement may be attributed to the closer contact due to the sol–gel shrinkage. Compared with TiO₂–AgNW electrodes, the PES–AgNW ones show superior optical and mechanical properties as a result of formation of a compact layer. The PES–AgNW electrodes also reveal good resistance to oxidation (300 °C in air; 500 °C in N₂) and excellent bend-test performance (500 cycles), and have been utilized successfully for practical application in a touch sensor.

Acknowledgements

This study was supported financially by the Ministry of Science and Technology of the Republic of China.

References

1. Z. Wu, Science, 2004, 305, 1273–1276.
2. B. T. Liu, C. H. Hsu and W. H. Wang, J. Taiwan Inst. Chem. Eng., 2012, 43, 147–152.
3. L. G. De Arco, Y. Zhang, C. W. Schlenker, K. Ryu, M. E. Thompson and C. W. Zhou, ACS Nano, 2010, 4, 2865–2873.
4. B. Y. Ahn, D. J. Lorang and J. A. Lewis, Nanoscale, 2011, 3, 2700–2702.
5. A. R. Madaria, A. Kumar, F. N. Ishikawa and C. W. Zhou, Nano Res., 2010, 3, 564–573.
6. C. H. Liu and X. Yu, Nanoscale Res. Lett., 2011, 6, 75.
7. A. Rahy, P. Bajaj, I. H. Musselman, S. H. Hong, Y.-P. Sun and D. J. Yang, Appl. Surf. Sci., 2009, 255, 7084–7089.
8. B. T. Liu and C. H. Hsu, J. Colloid Interface Sci., 2011, 359, 423–427.
9. X. Y. Zeng, Q. K. Zhang, R. M. Yu and C. Z. Lu, Adv. Mater., 2010, 22, 4484–4488.
10. T. A. Crowley, K. J. Ziegler, D. M. Lyons, D. Erts, H. Olin, M. A. Morris and J. D. Holmes, Chem. Mater., 2003, 15, 3518–3522.
11. E. Braun, Y. Eichen, U. Sivan and G. Ben-Yoseph, Nature, 1998, 391, 775–778.
12. T. Tetsumoto, Y. Gotoh and T. Ishiwatari, J. Colloid Interface Sci., 2011, 362, 267–273.
13. J. Xu, J. Hu, C. J. Peng, H. L. Liu and Y. Hu, J. Colloid Interface Sci., 2006, 298, 689–693.
14. S. H. Kim, B. S. Choi, K. Kang, Y. S. Choi and S. I. Yang, J. Alloys Compd., 2007, 433, 261–264.
15. T. K. Chen, W. T. Chen, M. C. Yang and M. S. Wong, J. Vac. Sci. Technol., B, 2005, 23, 2261–2265.
16. M. Tsuji, Y. Nishizawa, K. Matsumoto, N. Miyamae, T. Tsuji and X. Zhang, Colloids Surf., A, 2007, 293, 185–194.
17. L. F. Gou, M. Chipara and J. M. Zaleski, Chem. Mater., 2007, 19, 1755–1760.
18. Z. Xie, Z. Y. Wang, Y. X. Ke, Z. G. Zha and C. Jiang, Chem. Lett., 2003, 32, 686–687.
19. Y. G. Sun, B. Gates, B. Mayers and Y. N. Xia, Nano Lett., 2002, 2, 165–168.
20. Y. G. Sun and Y. N. Xia, Adv. Mater., 2002, 14, 833–837.
21. K. E. Korte, S. E. Skrabalak and Y. N. Xia, J. Mater. Chem., 2008, 18, 437–441.
22. Y. G. Sun, B. Mayers, T. Herricks and Y. N. Xia, Nano Lett., 2003, 3, 955–960.
23. Y. G. Sun, Y. D. Yin, B. T. Mayers, T. Herricks and Y. N. Xia, Chem. Mater., 2002, 14, 4736–4745.
24. C. Chen, L. Wang, G. H. Jiang, J. F. Zhou, X. Chen, H. J. Yu and Q. Yang, Nanotechnology, 2006, 17, 3933–3938.
25. D. P. Chen, X. L. Qiao, X. L. Qiu, J. G. Chen and R. Z. Jiang, J. Mater. Sci.: Mater. Electron., 2011, 22, 6–13.
26. X. L. Tang, M. Tsuji, M. Nishio and P. Jiang, Bull. Chem. Soc. Jpn., 2009, 82, 1304–1312.
27. Y. C. Lu and K. S. Chou, Nanotechnology, 2010, 21, 215707.
28. S. Coskun, B. Aksoy and H. E. Unalan, Cryst. Growth Des., 2011, 11, 4963–4969.
29. S. M. Bergin, Y. H. Chen, A. R. Rathmell, P. Charbonneau, Z. Y. Li and B. J. Wiley, Nanoscale, 2012, 4, 1996–2004.
30. X. L. Tang, M. Tsuji, P. Jiang, M. Nishio, S. M. Jang and S. H. Yoon, Colloids Surf., A, 2009, 338, 33–39.
31. B. Wiley, T. Herricks, Y. G. Sun and Y. N. Xia, Nano Lett., 2004, 4, 1733–1739.
32. M. J. Hu, J. F. Gao, Y. C. Dong, S. L. Yang and R. K. Y. Li, RSC Adv., 2012, 2, 2055–2060.
33. S. Sorel, P. E. Lyons, S. De, J. C. Dickerson and J. N. Coleman, Nanotechnology, 2012, 23, 185201.
34. M. J. Hu, J. F. Gao, Y. C. Dong, K. Li, G. C. Shan, S. L. Yang and R. K. Y. Li, Langmuir, 2012, 28, 7101–7106.
35. W. Gaynor, G. F. Burkhard, M. D. McGehee and P. Peumans, Adv. Mater., 2011, 23, 2905–2910.
36. J. Y. Lee, S. T. Connor, Y. Cui and P. Peumans, Nano Lett., 2008, 8, 689–692.
37. R. Zhu, C. H. Chung, K. C. Cha, W. B. Yang, Y. B. Zheng, H. P. Zhou, T. B. Song, C. C. Chen, P. S. Weiss, G. Li and Y. Yang, ACS Nano, 2011, 5, 9877–9882.
38. L. B. Hu, H. S. Kim, J. Y. Lee, P. Peumans and Y. Cui, ACS Nano, 2010, 4, 2955–2963.
39. B.-T. Liu and H.-L. Kuo, Carbon, 2013, 63, 390–396.
40. J. F. Weaver and G. B. Hoflund, J. Phys. Chem., 1994, 98, 8519–8524.
41. S. W. Gaarenstroom and N. Winograd, J. Chem. Phys., 1977, 67, 3500–3506.
42. P. Sangpour, A. Babapour, O. Akhavan and A. Z. Moshfegh, Surf. Interface Anal., 2009, 41, 157–163.
43. H. A. Macleod, Thin-film optical filters, Institute of Physics Pub., Philadelphia, 2001.
44. B.-T. Liu and W.-D. Yeh, Colloids Surf., A, 2010, 356, 145–149.
45. G. Hass and L. Hadley, Optical constants of metals, in American Institute of Physics Handbook, ed. D. E. Gray, McGraw-Hill, New York, 1972.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra11660g

---