The effect of light and humidity on the stability of silver nanowire transparent electrodes

Abstract

Transparent electrodes based on silver nanowire (AgNW) films have attracted considerable attention owing to their high electrical and thermal conductivity, optical transparency, and flexibility. However, the long-term reliability of AgNW electrodes, has seldom been studied. In this work, the effects of storage environment, such as, natural light and humidity, on the long-term reliability have been investigated in detail. The increase of electrical resistance with storage time greatly depended on the natural light illumination, especially ultraviolet (UV) illumination, and high humidity seriously accelerated the failure of AgNW electrodes. All the results were dependent on the storage temperature. Some over-coating layers have been used to protect the AgNW electrodes from light and humidity in environment and the results showed that an epoxy resin protecting layer exhibited the longest lifetime: over 40 days at 85 °C and 85% relative humidity.

---

1. Introduction

In 2008, Lee et al. fabricated the first silver nanowire (AgNW) transparent conductive films.¹ Many attempts have been made to use AgNW films in solar cells,^2–5 OLEDs,^6–8 touch screens,^9,10 and sensors.^11,12 Various detailed post-treatment techniques have been designed and developed to improve the conductivity and transparency of AgNW films.^3–16 Hundreds of papers/books about AgNW films are published annually; however, there have been no systematic studies about the reliability of AgNW films although it is known that bulk silver is easily corroded or tarnished in air and sensitive to some gases.

Under ambient conditions, bulk silver reacts strongly with gaseous sulfur-containing compounds to form a Ag₂S corrosion layer, the reactions are accelerated if water and oxygen are also present.^17–20 Similarly, silver nanoparticles (AgNPs) are also easily corroded via sulfidation/oxidization because, compared with bulk silver, silver nanomaterials have a much higher surface-to-volume ratio, which enhances surface reactivity. For example, the sulfidation rate due to chemisorption of sulfides on NPs is 7.5 times higher than that of bulk silver under the same conditions.^21,22

As an emerging transparent conductive film, AgNW films with low resistance, high transmittance and high reliability are very important for future applications. Some researchers have realized the serious problem of silver corrosion²³ and developed some barrier layers to prevent slow degradation of AgNW films over time and enhance the film reliability. For example, AgNW–rGO (reduced graphene oxide: r-GO) hybrid electrodes were exposed to 70 °C/70% RH for 8 days leading to a sheet resistance increase of only 50%.²⁴ Jin et al. reported high performance of a AgNW film coated by a non-conductive binder of poly(dopamine), which included a chemical-resistant substance (alginic acid); the AgNW film exhibited long-term stability during 85 °C/85% RH aging for 30 days.²⁵ However, pristine AgNWs without any protective layer also showed long-term stability under the same conditions, which contradicts other observations of silver corrosion.^17–23 AgNW–PEDOT:PSS composite films showed a 25% increase in sheet resistance when the film was exposed to 25 °C/50% RH for only 40 h.²⁶ With UV–ozone treatment, a AgNW–PDMS (poly(dimethylsiloxane)) film maintained its original sheet resistance for only 300 min.²⁷ Other AgNW–metal oxide (SiO₂ or ZnO) films and Au-coated AgNW films only showed short lifetimes below one week.^28,29 The results suggest that the reliability of AgNW films remains an important issue, and most protective layers are only effective for very short periods. To our knowledge, efficient protective layers have not been developed yet, and the failure mechanism of AgNW films remains unclear. Very recently, the Simonato group reported that AgNW films remained stable in ambient atmosphere over two years however, the exact nature of the storage conditions is unknown.³⁰ They claimed concentrated H₂S or light exposure did not cause sheet resistance change, however, an oxide shell around a metallic silver particle core has been observed under ambient conditions after light illumination for several hours.³¹ These reports left many unsolved issues. In this work, AgNW films were fabricated and coated with various materials; then the coated films were stored in ambient air under various conditions, including natural daylight, UV light, light shielding, high humidity, and vacuum. The microstructure and resistance of the AgNW films were observed to identify failure factors and clarify failure mechanisms, and provided real feedback concerning the stability of AgNW films in practical applications.

2. Experimental section

Preparation of AgNW films

AgNWs were synthesized on a large scale using a modified polyol process.³² The prepared AgNWs were dispersed in ethanol at a concentration of about 2.5 wt% to fabricate AgNW films by drop-coating onto 3 × 3 cm polyethylene terephthalate (PET) or glass substrates, which were allowed to dry naturally at room temperature. These AgNW films were treated with high-intensity pulse light illumination (PulseForge 3300, Novacentrix, Austin, TX, USA) at a fixed light intensity of 1.0 J cm⁻², similarly to a recently reported method.³³ Finally, the films were over-coated and stored under real laboratory conditions including seasonal changes. The detailed formation process for the AgNW films is shown in a schematic diagram (Fig. S1a†), unless stated otherwise. When a coating was too thick to measure sheet resistance, Au electrodes (each 0.3 × 3 cm) were placed on two sides of the film by sputtering at RT. These electrodes were used to evaluate the film’s conductive performance. For each set of conditions, three to five samples were tested to confirm the reproducibility.

Optical, electrical, and microscopic characterization

Transmittance spectra were recorded using a Jasco UV-visible-near-infrared spectrophotometer (V670, JASCO Corp.) with the corresponding substrate used as a reference. The films had 82–87% transmittance, unless stated otherwise. The sheet resistance of the AgNW films was measured using the four-probe method with a surface resistivity meter (Loresta GP MCP-T610, Mitsubishi Chemical Analytech Co., Ltd.) and resistance was obtained by a two-probe method with a multimeter (Sanwa, PC5000). Morphologies of the AgNW films were observed and analyzed by SEM (JEOL JSM-6700F). Long-term reliability tests of the AgNW films followed test standards in microelectronics in which the test temperature and humidity are defined as 85 °C and 85% RH (ESPEC SH-240S3).

3. Results and discussion

Effects of natural daylight

As a transparent conductive film, a AgNW film is always exposed to natural daylight; therefore, the effects of light on the failure of AgNW films should be understood. Here, bare AgNW films without any coating were fabricated and kept under laboratory conditions. Some films were directly exposed to air and others were wrapped with Crecia Kim towels to avoid natural light but allow exposure to the same air environment. Sheet resistance was measured periodically to monitor temporal changes in conductivity (Fig. 1). The figure shows that sheet resistance generally increased with storage time although changes were small in the 1st week. After the 1st week, the resistance began to sharply increase, with the increase rate highly dependent on the AgNWs film’s state. For example, in the 2nd week, the resistance of films directly exposed to natural light increased by 3–5 factors compared with the original resistance (Fig. 1a). For all samples, the resistance increased by factors of over 25 in the 3rd week; this agrees with most reports on the stability of pristine AgNW films.^24,26–29 After 30 days, most samples were completely destroyed. Conversely, the films covered with a Kim towel showed slow increases in resistance. Under exposure to air, but not light, the increase in resistance was less that a factor of five over 30 days (Fig. 1b). This indicates that shaded storage decelerated the failure of AgNW films.

Fig. 1 Temporal changes in sheet resistance for naked AgNW films (a) exposed to air and (b) wrapped in a Crecia Kim towel, and AgNW films coated with SP-1509 vinyl ester resin and (c) exposed to air and (d) wrapped in a Crecia Kim towel.

Moreover, to further verify the effects of natural light and look for suitable coating materials, AgNW films were coated with a Ripoxy SP-1509 vinyl ester resin (Showa Denko, Japan³⁴). The resin was dissolved in ethanol with a concentration of 10 wt% and then coated onto AgNW films, which were left directly in air or wrapped with a Kim towel. The SP-1509 layer is transparent in the visible range and cut out a part of UV-light (Fig. S2†), and serves as a barrier to moisture and gas to some extent. Sheet resistivity generally increased with storage time (Fig. 1c and d). The resistance of the protected AgNW film remained low for 22 days, but afterward, it quickly increased by a factor of over 15 (Fig. 1c). In contrast, the AgNW films wrapped in the Kim towel almost maintained low resistance for up to 50 days (Fig. 1d). This confirms that natural light truly accelerated the failure of AgNW films even when they were coated. However, compared with the unprotected AgNW films (Fig. 1a), the protected ones showed a significant clear delay in failure, which might imply an effect of the coating layer that will be discussed later. Although it has been claimed that corrosion of silver is related to water and gases in the air,^17–23 in the tests described above, the water and gas conditions were the same, therefore, the effect of natural light on silver corrosion cannot be ignored. Moreover, we also found that films with small values of initial sheet resistance always failed more slowly than those with high initial resistance (Fig. S3†). Films with higher sheet resistance have sparser NW networks, so wire failure significantly depleted conductive paths, leading to the quick increase of resistance.³⁵

SEM images give detailed information about AgNW evolution over time. Fresh AgNW films, coated or uncoated, always showed a normal random network structure with sharp wires (Fig. 2a). After 2 months, the wire shapes had changed from sharp to blurred (Fig. 2b–d). The films without coatings were broken into short thin rods, with some clear irregular particles between or around the rods. Some clear traces of the original wires remained (Fig. 2b). Apparently, thin NWs were more easily broken than thick ones. For example, a thick NW marked by the arrow in Fig. 2b remained a complete structure although some small particles appear on its surface. No clearly broken parts were observed for the coated films (Fig. 2c and d). These NWs maintained their original linear structures, although the original smooth surfaces were replaced by rougher surfaces. The high-magnification image in Fig. 2d shows that many tiny particles formed and surrounded these wires, and EDS analysis suggested the presence of elemental sulfur (Fig. S5†). These results agree well with other observations of many small protuberances (i.e., particles) after only 3 weeks of exposure to ambient air.^23,35 Our results also confirm that AgNW films prepared by the polyol process are unstable and undergo notable degradation in ambient air even when coated with a protective layer.^26–29

Fig. 2 SEM images of pristine AgNW films exposed to air for (a) 1 day and (b) 60 days. (c) SEM image of AgNW film coated with SP-1509 resin and exposed to air for 60 days. (d) High-magnification image of a region from panel (c).

Yu et al. has reported that AgNP aggregation is accelerated by light irradiation due to the inherent redox instability of silver.³⁶ Grillet et al., have confirmed that a photo-assisted oxidation process always occurred on the surface of Ag nanoparticles to form an oxide shell.³¹ Recently, a broadband, continuous light source was used to weld AgNWs via plasmonic mediated absorption and surface diffusion melting, indicating that light affects the AgNW structure.^37,38 In our experiments, the films shaded with Kim towels showed slower increases in sheet resistance than those exposed to light, implying that natural light accelerated structural changes in AgNWs. However, shading and coating did not completely stop the failure of AgNW films, strongly suggesting that ambient environment should be carefully considered. In the following sections, we individually address these factors under the same natural light.

Effects of thickness and coating layer material

Failure was rapid when AgNW films were directly exposed to air. However, the fact that a coating delayed the failure implied that the thickness and material of the coating are importance. We firstly used two concentrations of SP-1509 in ethanol, 5 or 20 wt%, to control the coating thickness. Fig. 3a shows changes in film resistance with storage time under exposure to laboratory air and natural light. The films coated with a thin layer showed low resistance only over the first 2 weeks; thereafter, the resistance increased drastically by a factor of over 20 within 20 days. In comparison, films with thick coatings maintained the resistance near the initially low level for over 2 months. Then, the resistance started to increase with time. These results strongly suggest that a thick layer prolonged the lifetime of AgNW films under natural light. The thin and thick layers of SP-1509 have the same transmittance in the visible and UV range. The light affect can be excluded. The resin SP-1509 is used as a matrix in conductive adhesives and can protect devices in ambient air by shielding them from moisture and gases. Hence, a coating of SP-1509 initially protected the AgNW films. However, over time, some harmful gases or moisture might penetrate the resin layer and reach the AgNW surfaces depending on the thickness of the coating. It should be stressed that a thin coating, such as GO, PEDOT, might be the main corrosion reason why the AgNW films showed very short lifetimes in those studies.^24–29 This implies that more attention should be given to the design of the protecting layers for AgNW films.

Fig. 3 Comparison of temporal changes in sheet resistance of AgNW films coated with a thick layer of SP-1509 resin and a thin layer (a), gelatin, PVP and PVA (b) and epoxy resin (c).

In addition to the thickness, to look for suitable materials and confirm the effects of water^18–20 on AgNW films in air is also an issue. Several kinds of gelatin, polyvinyl alcohol (PVA), polyvinylpyrrolidone (PVP) and epoxy resin (Showa Denko, Japan) were selected to form a thick coating layer on the AgNW films. Fig. 3b and c shows temporal changes in the resistance of the coated AgNW films in ambient air. The increase in resistance over time depended on the material used for the coating. For the films coated with gelatin, the resistance increased by a factor of three when the AgNWs films were kept in air for only 1 week; this increase was faster compared with films without coating (Fig. 1a). The films coated with gelatin completely failed after only 1 month (Fig. 3b). Films coated with PVP showed low rates of increase in resistance. After 20 days in air, these films had resistances that had increased by less than a factor of four; this was much lower than those for films coated with gelatin at the same ambient conditions. However, the PVP-coated films were generally damaged over time, and resistance was nearly 10 times higher after the films were stored in air for 30 days. Thereafter, resistance increased drastically and some films quickly failed. The PVA-coated films maintained low resistance over 45 days. It is found that these results are consistent with the properties of these coating materials. Gelatin is a widely used green food additive that can dissolve in hot water and has high water retention. When gelatin is coated on AgNW films and dried, some moisture remains; this retained water is slowly released and contributes to the failure of NWs. PVP and PVA are both water-soluble polymers with a plurality of polar groups, so they are relatively strong hydrophilic materials. When the same amounts of PVA and PVP were kept in air at RT, the weight gain of PVP was always more than that of PVA. This water absorption capability might accelerate the failure of AgNW films coated with PVP.

Epoxy resin was coated on surfaces of AgNW films and cured with imidazole at 150 °C for 30 min to form a uniform and hard layer in order to observe changes in resistance over time. Fig. 3c shows a very stable resistance, except for deviations in measurements, without any change over 6 months. Owing to its special waterproof and chemical-proof properties, epoxy resin is widely used as a sealing agent for electronic devices. On the AgNW films, the epoxy resin layer blocked penetration of moisture and protected the films from failure. A pure coating of epoxy resin on glass was placed in air at RT, and changes in weight were measured over time. The weight remained almost the same for over 2 months, demonstrating that the water-absorbing ability of epoxy resin is very poor. These results imply that blocking water from air might be a suitable method to extend AgNW films’ lifetimes. In summary, the epoxy resin has shielding ability compared with other layers, which might protect the AgNW films from failure to some extent (Fig. S2†).

Effects of water

Water is a key factor in the corrosion of bulk Ag, and the results for the materials discussed above also imply that moisture in air affected the reliability of AgNW films.^24–29 In order to further confirm the effects of water, uncoated AgNW films were stored in a dry glass desiccator, which was periodically evacuated to remove as much moisture as possible, and a glass vessel containing hot water was introduced to generate a high humidity of about 70–80% by evaporation of water under natural light. Fig. 4 shows the resulting changes in resistance with storage time. In the first 2 weeks, the AgNW films in the high-humidity vessel maintained low resistance without clear changes, but thereafter, three samples showed rapid increases in resistance and completely failed in the following 3rd week. Only one sample avoided failure until the 4th week. In contrast, the AgNW films stored in dry vacuum almost maintained their initial low resistance without any change for over 2 months; thereafter, the resistance slowly increased. These results confirm that water significantly enhances the failure of AgNW films stored in ambient air, just as water enhances corrosion of bulk silver.^17–20 It should be mentioned that the AgNW films rapidly failed when the temperature and humidity were increased at the same time, which agrees with a previous report.³⁰ However, although vacuum storage significantly decreased the effects of water vapor and other gases, the films exposed to the same natural light still deteriorated over time. This strongly confirms that natural light is an inevitable cause of corrosion in AgNW films again.³¹

Fig. 4 Comparisons of temporal changes in sheet resistance of AgNW films stored in a vessel with hot water or under vacuum.

Effects of UV light

It has been shown above that films kept in the dark always showed slower rates of failure and partly cutting out UV-light seems prolong the lifetime. To confirm the effect of UV-light, sheets of three different materials (glass, PET, and polyamide (PI)), were used to cover AgNW films. These materials exhibit completely different transmittance (Fig. S2†). Glass is almost transparent and cuts out UV-light below 280 nm. PET has a slightly lower transmittance of about 75% at 550 nm and cuts out wavelengths below 310 nm. PI has the lowest transmittance and cuts out light below 480 nm. Each of these three sheets were placed on the surface of a AgNW film and sealed with commercial glue along the sides of the films; then the films were stored in air. Fig. 5a shows that the increase in resistance over time depended on the material used for the sheets. The AgNW films covered with glass showed slight changes in the first 30 days; thereafter, the resistance suddenly increased with failure. The films covered with PET maintained high conductivity after 30 days, but the resistance generally increased after that and failed until 60 days. More interesting are the results for PI sheets. Although the resistance increased similar to that for the glass and PET sheets, the rate of increase was relatively slow; after 60 days, the resistance had increased only by a factor of two. This implies that the PI sheet effectively prevented the failure of the films; this might be due to the shading provided by PI below 480 nm. Reducing or eliminating UV light might be a suitable method to extend the reliability of AgNW films in air. However, silver corrosion always involves many factors such as water, oxygen, light, and temperature. The three materials (glass, PET, and PI), provide different barriers for gas and water in air, and these differences are difficult to quantify in the present case. An accelerated test was therefore designed to confirm the influence of UV light.

Fig. 5 (a) Temporal changes in sheet resistance of AgNW films stored in air but covered with glass, PET, or PI. (b) Temporal changes in resistance of AgNW films exposed to UV radiation with no coating (blank) and covered with a PI sheet.

Two types of AgNW films were fabricated: an uncoated AgNW film and an AgNW film covered by a PI sheet. These films were illuminated with a UV lamp at 365 nm (SLUV-8, Asone Co. Ltd.) in ambient air. In the test range, no large changes in resistance were observed, but these films were easily and quickly destroyed in a short time compared with those stored in natural light (Fig. 5b). One blank (uncovered) AgNW film was damaged in the 3rd day, although it had a low resistance in the first 2 days. Another blank AgNW film showed decreased resistance in the 2nd day, and then the resistance slowly increased until the 9th day. Similarly, the AgNW films covered by the PI sheets showed decreasing resistance in the first 2 days, which might correspond to the UV light curing that enhances contacts between AgNWs.^27,37,38 Thereafter, the resistance only increased slightly and remained almost stable until the 15th day. Similar to the blank AgNW films, the covered films were completely destroyed. Compared with films stored in natural light, the short lifetimes observed here suggest that UV light is a strong failure factor. It has previously been observed that due to oxidation, the resistance of AgNW films treated with UV–ozone was unstable after as little as 1 h.^27,31 Therefore avoiding exposure to UV light should be considered in developing applications for AgNW films.

Accelerating test

Before AgNW films can be used in devices, they must be subjected to a standard microelectronics test to verify the film reliability. Here, AgNW films with selected coatings were placed in an oven at 85 °C/85% RH in the absence of natural light. Fig. 6 shows temporal changes in resistance of AgNW films. Except for the sample coated with epoxy resin, the resistance always increased with storage time. The rate of increase was largest for the sample coated with SP-1509 resin, followed by those coated with PVP and PVA, which showed almost the same rates of increase. The sample coated with epoxy resin gave the best results and maintained an unchanging resistance for over 40 days. To our knowledge, this is the longest lifetime that has been reported for any AgNW film stored in a harsh environment. In comparison, films stored in air with the same coating materials (PVP, PVA, and SP-1509) always increased faster in resistance than when they were placed in the high humidity oven. One reason might be the blocking of natural light, now UV-light, due to the thick metal container. Another reason could be heat treatment, which might densify coatings and prevent penetration of water vapor and other gases. Nevertheless, failure occurrence is only a matter of time; after some period of time, these films showed sudden increases in resistance and the SEM image shows that many wires have been broken (Fig. S4†). Moreover, SP-1509 and epoxy resin are similar resins that are used as sealing materials in many devices. The huge difference in failure for AgNW films coated with these resins might be related to differences in curing methods. The SP-1509 resin was dried and hardened in air, which might result in incomplete solidification and faulty sealing. In contrast, the epoxy resin was cured at 150 °C for 30 min to form a complete and dense layer on the surface of the AgNW film, thereby giving better protection. Furthermore, compared with the better adhesion of epoxy resin, SP-1509 always adhered weakly to the substrates; this might also affect the rapid failure of AgNW films. However, the resistance of AgNW films protected by epoxy resin increased with time over 60 days in the present test, which suggested that humidity has a large impact on AgNW reliability, as reported elsewhere.^18–23 Finally, in the accelerated test, all AgNW films were protected from natural light, which has been confirmed to affect the failure of AgNW films. Hence, for extensive applications of AgNW films over long durations, the accelerating test conditions, which are more suitable for AgNW films considering the discussed failure factors, should be further investigated in future studies for the evaluation of high-performance coatings.

Fig. 6 Temporal changes in resistance of AgNW films coated with selected materials and stored at 85 °C and 85% RH.

It has been found that AgNW films are always destroyed after a period of time even in totally different environments. EDS measurements (Fig. S5†) show that sulfide is the main final product, which agrees with the reaction between silver and H₂S gas/OCS to form Ag₂S, and the reaction is significantly enhanced at high RH.^17–23 We also found that the stability of AgNW films was worse in summer than in winter due to the difference of RT and humidity. Except for H₂S sources, OCS plays a larger role because 3 million tons of OCS are released into the atmosphere each year. And sulfide formation can be enhanced by the presence of nitrogen dioxide and water,³⁹ which strongly indicates that the corrosion reaction of silver is ubiquitous. Moreover, we have to mention that ubiquitous natural light, especially UV-light, also unusually accelerated the silver corrosion, therefore, complicating further the silver corrosion. On the other hand, we should also consider the Rayleigh instability effect. Many phenomena in nature exhibit Rayleigh instabilities, i.e. a solid cylinder will finally break up into a row of spheres, such as the formation of water drops, fiber spinning, and fission of charged finite systems.^40–42 Recently, silver nanobelts and NWs of other metals have been shown to fragment at temperatures far below their bulk melting temperatures because of Rayleigh instabilities.^43–45 Although our data supported the effect of natural light, UV light, temperature, gas and humidity on the reliability of AgNW films in ambient air, it does not exclude the possible effects of Rayleigh instabilities. More studies are required to determine whether the instability of AgNWs in air is due to corrosion, Rayleigh instabilities, or some other mechanism.

4. Conclusions

We studied how natural light and humidity environmental factors affect failure of AgNW films and found that natural light, especially UV light, temperature, and humidity often shorten the lifetime and decrease the reliability of AgNW films. Shading the film from natural light was an effective method of enhancing the reliability of AgNW films. Coating the films with a protective layer always decreased the failure rate. Dry air also prolonged the lifetime. Those materials with good waterproofing ability always extended the stability of AgNW films. AgNW films coated with epoxy resin showed the longest lifetime of over 40 days under an atmosphere at 85 °C/85% RH; this is longer than any lifetime previously reported for AgNW films. Many factors always simultaneously or separately affect the reliability of AgNW films, it may be useful to develop a multilayer structure in which each layer blocks one of the factors contributing to corrosion such as water vapor, gases, and light.

Acknowledgements

We thank Yasuha Izumi for help in sample preparation and SEM observation. The Support Office for Large-Scale Education and Research Projects, Osaka University is thanked for their support on copy-editing. The work was partly supported by Showa Denko Co. Ltd, and COI Stream project.

References

1. J. Y. Lee, S. T. Connor, Y. Cui and P. Peumans, Nano Lett., 2008, 8, 689–692.
2. H. Y. Jong, S.-Y. Joe, C. Pang, K. M. Lee, H. Jeong, J. K. Park, Y. H. Ahn, J. C. Mello and S. Lee, ACS Nano, 2014, 8, 2857–2863.
3. B. H. Hardin, W. Gaynor, I. K. Ding, S. B. Rim, P. Peumans and M. D. McGehee, Org. Electron., 2011, 12, 875–879.
4. C. H. Chung, T. B. Song, B. Bob, R. Zhu, H.-S. Duan and Y. Yang, Adv. Mater., 2012, 24, 5499–5504.
5. T. Tokuno, M. Nogi, M. Karakawa, J. Jiu, T. T. Nge, Y. Aso and K. Suganuma, Nano Res., 2011, 4, 1215–1222.
6. X. Zeng, Q. Zhang, R. Yu and C. Lu, Adv. Mater., 2010, 22, 4484–4488.
7. H. Lee, D. Lee, Y. Ahn, E. Lee, L. Park and Y. Lee, Nanoscale, 2014, 6, 8565–8570.
8. W. Gaynor, S. Hofmann, M. C. Greyson, C. Sachse, S. Mehra, A. Salleo, M. D. McGehee, M. C. Gather, B. Lüssem, L. Müller-Meskamp, P. Peumans and K. Leo, Adv. Mater., 2013, 25, 4006–4013.
9. A. R. Madaria, A. Kumar and C. Zhou, Nanotechnology, 2011, 22, 245201.
10. J. Lee, P. Lee, H. Lee, D. Lee, S. S. Lee and S. H. Ko, Nanoscale, 2012, 4, 6408.
11. Y. Zhu, C. M. Hill and S. L. Pan, Langmuir, 2011, 27, 3121.
12. S. Yao and Y. Zhu, Nanoscale, 2014, 6, 2345–2352.
13. Y. Liu, Q. Chang and L. Huang, J. Mater. Chem. C, 2013, 1, 2970.
14. W. Gaynor, G. F. Burkhard, M. D. McGehee and P. Peumans, Adv. Mater., 2011, 23, 2905–2910.
15. C. H. Chung, T. B. Song, B. Bob, R. Zhu and Y. Yang, Nano Res., 2012, 5, 805–814.
16. A. Kim, Y. Won, K. Woo, S. Jeong and J. Moon, Adv. Funct. Mater., 2014, 24, 2462–2471.
17. C. Kleber, R. Wiesinger, J. Schnoller, U. Hilfrich, H. Hutter and M. Schreiner, Corros. Sci., 2008, 50, 1112–1121.
18. J. P. Franey, G. W. Kammlot and T. E. Graedel, Corros. Sci., 1985, 25, 133–143.
19. T. E. Graedel, G. J. Franey, G. W. Gualtieri, D. L. Kammlott and D. L. Malm, Corros. Sci., 1985, 25, 1163–1180.
20. L. Volpe and P. J. Peterson, Corros. Sci., 1989, 29, 1179–1196.
21. C. Levard, B. C. Reinsch, F. M. Michel, C. Oumahi, G. V. Lowry and G.E. Brown, Environ. Sci. Technol., 2011, 45, 5260–5266.
22. M. D. McMahon, R. Lopez, H. M. Meyer III, L. C. Feldman and R. F. Haglund, Appl. Phys. B, 2005, 80, 915–921.
23. J. L. Elechiguerra, L. Larios-Lopez, C. Liu, D. Garcia-Gutierrez, A. Camacho-Bragado and M. J. Yacaman, Chem. Mater., 2005, 17, 6042–6052.
24. Y. Ahn, Y. Jeong and Y. Lee, ACS Appl. Mater. Interfaces, 2012, 4, 6410–6414.
25. Y. Jin, D. Deng, Y. Cheng, L. Kong and F. Xiao, Nanoscale, 2014, 6, 4812–4818.
26. D. Y. Choi, H. K. Kang, H. J. Sung and S. S. Kim, Nanoscale, 2013, 5, 977.
27. W. J. Lee, M. Y. Lee, A. P. Roy, L. S. Lee, S. Y. Park and I. In, Chem. Lett., 2013, 42, 191–193.
28. K. Zilberberg, F. Gasse, R. Pagui, A. Polywka, A. Behrendt, S. Trost, R. Heiderhoff, P. Görrn and T. Riedl, Adv. Funct. Mater., 2014, 24, 1671–1678.
29. T. Kim, A. Canlier, C. Cho, V. Rozyyev, J. Y. Lee and S. M. Han, ACS Appl. Mater. Interfaces, 2014, 6, 13527–13534.
30. C. Mayousse, C. Celle, A. Fraczkiewicz and J. P. Simonato, Nanoscale, 2015, 7, 2107–2115.
31. N. Grillet, D. Manchon, E. Cottancin, F. Bertorelle, C. Bonnet, M. Broyer, J. Lerme and M. Pellarin, J. Phys. Chem. C, 2013, 117, 2274–2282.
32. J. Jiu, T. Araki, J. Wang, M. Nogi, T. Sugahara, S. Nagao, H. Koga, S. Suganuma, E. Nakazawa, M. Hara, H. Uchida and K. Shinozaki, J. Mater. Chem. A, 2014, 2, 6326–6330.
33. J. Jiu, T. Sugahara, M. Nogi, T. Araki, K. Suganuma, H. Uchida and K. Shinozaki, Nanoscale, 2013, 5, 11820–11828.
34. H. W. Cui, J. Jiu, S. Nagao, T. Sugahara, K. Suganuma, H. Uchida and K. A. Schroder, RSC Adv., 2014, 4, 15914–15922.
35. H. H. Khaligh and I. A. Goldthorpe, Nanoscale Res. Lett., 2013, 8, 235.
36. S. Yu, Y. Yin, J. B. Chao, M. Shen and J. Liu, Environ. Sci. Technol., 2013, 48, 403–411.
37. J. A. Spechler and C. B. Arnold, Appl. Phys. A, 2012, 108, 25–28.
38. E. C. Garnett, W. Cai, J. J. Cha, F. Mahmood, S. T. Connor, M. G. Christoforo, Y. Cui, M. D. McGehee and M. L. Brongersma, Nat. Mater., 2012, 11, 241.
39. J. H. Payer, G. Ball, B. I. Rickett and H. S. Kim, Mater. Sci. Eng., A, 1995, 198, 91–102.
40. K. F. Gurski and G. B. Mc Fadden, Proc. R. Soc. London, Ser. A, 2003, 459, 2575.
41. D. Duft, T. Achtzehn, R. Müller, B. A. Huber and T. Leisner, Nature, 2003, 421, 128.
42. C. Brechignac, P. Cahuzac, F. Carlier, C. Colliex, J. Leroux, A. Masson and U. Landman, Phys. Rev. Lett., 2002, 88, 196103.
43. Y. Sun, B. Mayers and Y. Xia, Nano Lett., 2003, 3, 675–679.
44. M. E. Toimil Molares, A. G. Balogh, T. W. Cornelius, R. Neumann and C. Trautmann, Appl. Phys. Lett., 2004, 85, 5337.
45. H. Li, J. M. Biser, J. T. Perkins, S. Dutta, R. P. Vinci and H. M. Chan, J. Appl. Phys., 2008, 103, 024315.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra02722e

---