Solar-blind photodetector based on Ga₂O₃ nanowires array film growth from inserted Al₂O₃ ultrathin interlayers for improving responsivity

Abstract

Due to the confinement of scale and dimension, the photon–electron interactions in nanostructures can be effectively manipulated for excellent photoelectric properties of materials and devices. β-Ga₂O₃ with a direct bandgap of ∼4.9 eV has a great potential in solar-blind ultraviolet detection. Photodetectors based on nanostructures of β-Ga₂O₃ material usually show a high photoelectric performance. However, low device fabrication repeatability and low effective utilization of light hinder the practical application of the photodetectors. In this work, we propose a preparation method to obtain a Ga₂O₃ nanowires film which combines the benefits of nanowires and thin films by alternative depositing of Ga₂O₃ and Al₂O₃ ultrathin layers. The vertical nanowire structured films are formed at low temperature with an appropriate stress relaxation. Then prototype photodetectors of a metal–semiconductor–metal structure are fabricated using the as-grown films. The nanowire films-based photodetectors exhibit higher responsivities than the smooth film-based photodetector, which is attributed to their larger surface-to-volume ratio and stronger light scattering effects on the nanostructured film surface. Besides, the excellent stability and tolerable response time of Ga₂O₃/Al₂O₃ nanowires based photodetectors indicate that the photodetectors can be developed for the future applications in solar-blind ultraviolet communication.

---

Introduction

Due to the strong absorption of deep-ultraviolet (DUV) light by stratospheric ozone, solar irradiation wavelengths shorter than 280 nm, called the solar-blind region, do not exist at the surface of the earth. PDs with DUV solar-blind sensitivity can detect very weak signals accurately under sun and artificial illuminations due to the “black background”. DUV photodetectors (PDs) have been intensively studied as important applications in missile tracking, space communication, photolithography, and so on could be found.^1–4 Recently, some wide bandgap semiconductors such as AlGaN,^5,6 ZnMgO,⁷ diamond,⁴ and β-Ga₂O₃ have been developed for DUV solar-blind PDs.^8–10 However, high quality epitaxial AlGaN film is difficult to be prepared due to high growth temperature,^5,9 single wurtzite phase ZnMgO,^7,11 and diamond are not possible to be used to detect entire DUV region because of their mismatched bandgap.^4,12 Therefore, β-Ga₂O₃, with ∼4.9 eV direct bandgap and tunable by alloying with Al₂O₃ (high to 7 eV) or In₂O₃ (low to 3.8 eV),^13,14 is considered as one of the ideal candidates to fabricate solar-blind PDs.

For the development of solar-blind PDs, bulks and films β-Ga₂O₃ based-devices have been vastly studied,^1–3,8,15 but the long response time, low responsivity of two-dimensional (2D) films and high cost of bulk materials-based devices hinder their practical applications.^8,15 To improve the applicability and photoconductivity performance of PDs based on Ga₂O₃, one-dimensional (1D) nanostructures Ga₂O₃ of low fabrication cost have been used as an alternative due to their low dimensional and large surface-to-volume ratio, resulting in shortening of response and decay times, enlargement of responsivity.^16,17 However, the low device fabrication repeatability and low effective utilization of light hinder the practical application of the photodetectors.¹⁸ Recently, it has been found that vertical nanostructures array film is advantageous in many aspects, such as easier fabrication, lower cost, more flexible, and higher performance.^19,20 Nanowires (NWs) film combines the benefits of NWs and thin films, while an easy and low cost fabrication method of Ga₂O₃ NWs films is rarely reported.^19,21–23 It is well known that the proper interlayers can efficiently modify the surface states, band gap, and conductivity of nanostructured films.

In this work, we propose a Ga₂O₃ and Al₂O₃ alternative deposition technique to prepare Ga₂O₃/Al₂O₃ NWs film by laser molecular beam epitaxy (LMBE) on the α-Al₂O₃ (0001) substrates. And then the prototype solar-blind PDs are fabricated by depositing Ti/Au interdigital electrode on Ga₂O₃/Al₂O₃ films to construct a metal–semiconductor–metal (MSM) structure. The PDs show spectra selectivity and high sensitivity. In comparing with smooth film-based PD, the NWs films-based PD exhibits a much higher responsivity. Moreover, the excellent stability and tolerable response time of NWs films-based PD are hopeful to their potential applications in solar-blind ultraviolet communication.

Experimental

All the samples are deposited on α-Al₂O₃ (0001) substrates by LMBE. The base pressure in chamber is 1 × 10⁻⁶ Pa. The growth temperature is 650 °C (750 °C, 850 °C) and the oxygen pressure is 5 × 10⁻³ Pa. The laser ablation is carried out at a laser fluence of ∼3 J cm⁻² with a repetition rate of 1 Hz using a KrF excimer laser with a wavelength of 248 nm. The distance between target and substrate is 4 cm. The Ga₂O₃/Al₂O₃ layers are alternately deposited and both depositions are repeated 20 times. The thickness of the films is controlled by changing the pulse numbers. Ga₂O₃ layers are fixed at 300, and Al₂O₃ layers are fixed at 30. Also, 6000 pulses sample of pure β-Ga₂O₃ have been grown at 650 °C (750 °C, 850 °C) and all other conditions are the same as comparison. To fabricate a PD, an interdigital Ti/Au electrode is deposited on the films using a shadow mask by radio frequency magnetron sputtering. The electrode fingers are 200 μm wide, 2800 μm long, and with a 200 μm spacing gap.

The crystallinity of the as-grown films are investigated by D8 Advance X-ray diffraction (XRD) using Cu Kα (λ = 1.5405 Å) radiation. The surface morphology and chemical composition are characterized by a Hitachi S-4800 field emission scanning electron microscope (FE-SEM) equipped with energy-dispersive X-ray spectrometer (EDX). UV-visible (UV-vis) absorption spectrum is taken using a Hitachi U-3900 UV-vis spectrophotometer. The current–voltage (I–V) and time-dependent photoresponse of the Ga₂O₃/Al₂O₃ films-based PDs are measured by Keithley 2450 source meter. A UV lamp (∼7 W) with the wavelength of 254 nm and 365 nm, and light irradiation power density is tuned through adjusting the distance between the light source and the samples. The time-dependent photoresponse measurement is performed at a constant voltage of 10 V.

Results and discussion

Fig. 1(a) shows the XRD patterns of the as-grown Ga₂O₃/Al₂O₃ films compared with the pure β-Ga₂O₃ films in different temperatures. Diffraction peaks locates at around 19°, 38°, and 59° are observed, except diffraction peaks of the Al₂O₃ substrate. These peaks are ascribed to the patterns of β-Ga₂O₃ and can be assigned as the (2̄01), (4̄02) and (6̄03) lattice plane (PDF# 43-1012), respectively. With decreasing growth temperature of Ga₂O₃/Al₂O₃ films, the intensity of the diffraction peaks decreases. The degradation of crystallinity is contributed to the low substrate temperature for the higher inclusion level of Al₂O₃, which weakens the epitaxial relation with sapphire substrate.^13,24 Compared with the pure β-Ga₂O₃ samples, the (6̄03) peaks of the Ga₂O₃/Al₂O₃ XRD patterns display a gradually shift to higher 2θ with the increasing growth temperature. It is attributed to the increasing content of Al ions incorporated into the Ga site with the Al₂O₃ layers (ALs) diffusing driven by the thermodynamic power, as the radii of Al ions are smaller than that of Ga ions.^13,24,25

Fig. 1 Crystal structure and morphology of as-grown Ga₂O₃/Al₂O₃ films prepared at different temperatures: (a) the XRD patterns contrast with pure β-Ga₂O₃ growth in the same conditions, (b) the top view of FE-SEM images and the corresponding enlarged image (inset). (c) Schematic diagram of the proposed growth mechanism of Ga₂O₃/Al₂O₃ nanowires film.

Fig. 1(b) represent top view FE-SEM images of the Ga₂O₃/Al₂O₃ films. The NWs with long length and high density are spreading over the entire film surface of these samples grown in 650 °C and 750 °C, but the sample grown in 850 °C is a 2D film. The surface morphology of the samples show a transformation from 1D to 2D depended on growth temperature. The average radius is ∼50 nm, and the density is ∼2 × 10¹⁴ cm⁻² of NWs film growth in 650 °C. The average radius is ∼100 nm, and the density is ∼2.5 × 10¹³ cm⁻² of NWs film growth in 750 °C. We think the vanishing of NWs is contributed to no stress Ga₂O₃/Al₂O₃ islands formed with drastic diffusion of Al in high temperature, which induces the growth of Ga₂O₃/Al₂O₃ NWs. Perhaps, the epitaxial relation with the substrate is much stronger with the Al₂O₃ interlayers diffusing, as a result the stress relaxation process of the ALs free and the Ga₂O₃/Al₂O₃ nanostructures on the film surface are emerged to a film.^26,27 Compared with sample 650 °C, sample 750 °C with a higher substrate temperature may form smaller aspect ratios (high/width) stress islands to induce larger but lower density NWs growth, which may have a weaker relaxation with the decreased thickness of ALs by thermal diffusion. Let us summarize the most important points regarding the formation of NWs on Ga₂O₃/Al₂O₃ film surface corresponded with GaN-interlayers NWs growth: the formation of the islands is dependent on the stress relaxation process of interlayers, and the transmission of the relaxation from weak to strong is dependent on the temperature of the substrate with the interlayers diffusion;^28–30 surface diffusion of adatoms on the substrate and along the NWs vertical side wall can account for the NWs lengthening during the growth phase once NWs nuclei have been formed.^26–28 We can expect that the aspect ratio of the stress islands can be tuned by changing the growth temperature and interlayers thickness. In other words, the NWs radius on the film surface can be tuned which acts stress islands as a seed in the growth.

From EDX pattern, there is no impurity in the NWs films except Al, Ga, and O. We propose the Ga₂O₃ NWs film with ALs growth mechanism is shown in Fig. 1(c): first, the Ga₂O₃/Al₂O₃ clusters formed on the substrate of the epitaxial relation is much stronger.²⁶ second, whereby the misfit dislocations are developed, Ga₂O₃/Al₂O₃ stress islands are formed.³⁰ Third, the islands grows with a fixed aspect ratio before their transformation to NWs.²⁶ Fourth, the islands transformed to NWs (the epitaxial constraint should be very weak) with ALs relaxation accumulating enough.^26–28 Fifth, with the ALs relaxation process, the Ga₂O₃/Al₂O₃ nanowires could grow vertically and radially without forming a smooth film.^26–30 Although the mechanism is not clear enough, the method can be regarded as a new attempt to obtain Ga₂O₃ NWs film.

Fig. 2 shows the UV-vis absorbance spectrum of the Ga₂O₃/Al₂O₃ films prepared at different temperatures. It is evident that all the samples have significant absorption edges at ∼250 nm, near the lower edge of the solar-blind region, and those edges shift to lower wavelength with the increase of growth temperature. For the direct transition semiconductors, the absorption followed a power law of the form:⁸

αhν = B × (hν − E_g)^1/2 (1)

where α is the absorption coefficient, h is the Planck's constant, ν is the frequency of the incident light, and B is a constant. Then the optical band gap E_g of the samples can be obtained by plotting (αhν)² versus hν and extrapolated the straight-line portion of this plot to the energy axis. The estimate E_g values are 4.88 eV, 5.04 eV, and 5.11 eV for the samples of 650 °C, 750 °C and 850 °C, respectively. With the increasing of the growth temperature, the E_g values blue shift. It was ascribed to the wide band gap of Al₂O₃, and the content of Al incorporated into the Ga site increased with Al ions thermal diffusion of the ALs, then the E_g values shift further.^13,25 These could also be verified by the XRD patterns of the Ga₂O₃/Al₂O₃ films.

Fig. 2 UV-vis absorbance spectrum of the Ga₂O₃/Al₂O₃ films prepared at different temperatures and the plot of (αhν)² versus hν (inset).

In order to check the DUV photoelectric properties of the Ga₂O₃/Al₂O₃ nanowires films compared with Ga₂O₃/Al₂O₃ film, a three-pair interdigital Au/Ti electrode was deposited through a shadow mask to serve as contact electrodes, the schematic diagram of the fabricated prototype photodetector device is shown in the bottom right inset of Fig. 3(a). Fig. 3(a)–(c) show the room temperature I–V characteristics in dark and different light illuminations for the samples 650 °C, 750 °C and 850 °C based devices, respectively. It is obvious that all the devices are much more sensitive for 254 nm than 365 nm light illuminations. At room temperature (300 K), the devices exhibit a pronounced photoresponse to the 254 nm light. For the light intensity varying from 10 to 70 μW cm⁻² of 254 nm light illuminations, photocurrent ranges from 1966.35 to 4637.74 nA, 260.68 to 657.35 nA, 1.86 to 2.82 nA for samples of 650 °C, 750 °C and 850 °C based devices, respectively (at 10 V bias). In detail, the dark current and photocurrent/dark-current ratio are 512.78 nA/9.04, 109.35 nA/6.01, 1.24 nA/2.95 under 254 nm light illuminations (70 μW cm⁻²) at 10 V for samples of 650 °C, 750 °C and 850 °C based devices, respectively.

Fig. 3 I–V characteristics curve of the photodetector based on Ga₂O₃/Al₂O₃ films prepared at different temperatures of (a) 650 °C, (b) 750 °C and (c) 850 °C under dark, 365 nm and 254 nm (various light intensities from 10–70 μW cm⁻²) light illuminations; the inset (bottom right of (a)) shows the schematic diagram of MSM structure. (d) Corresponding responsivity dependence on the bias for samples 650 °C, 750 °C and 850 °C under 254 nm illuminations (70 μW cm⁻²).

Although NWs film-based PD (samples of 650 °C and 750 °C) exhibits much higher photocurrent/dark-current ratio than smooth film-based PD (samples of 850 °C) and the dark current is also higher. The dark current rising is attributed to a lower substrate temperature for higher density of defects and misfit dislocation induced by higher inclusion level of ALs, which agrees with the analysis of NWs film growth mechanism above.^31,32 Therefore, it is not clear enough that Ga₂O₃/Al₂O₃ nanowires film with a larger surface-to-volume ratio would take on a higher response to 254 nm light illumination.

Thus, the responsivity (R_λ) is critical for evaluating a photosensitive device. Fig. 3(d) shows the R_λ of the three samples under 254 nm light illuminations, calculated by:¹⁶

R_λ = (I_light − I_dark)/(P_λ × S) (2)

where I_light is the light current, I_dark is the dark current of the device when illuminated with a light source, P_λ is the light intensity illuminated on the device, S is the effective illuminated area, and λ is the wavelength of illuminating light. The R_λ values of the devices under 254 nm light illuminations (70 μW cm⁻²) are 1.40 A W⁻¹, 0.19 A W⁻¹, 5.83 × 10⁻⁴ A W⁻¹ at bias 10 V for the samples 650 °C, 750 °C and 850 °C based devices, respectively. It is apparent that the NWs films-based PDs take on a much higher responsivity (samples of 650 °C and 750 °C) than smooth film-based PD (sample of 850 °C), and smaller radius NWs film-based device is (samples of 650 °C) higher than larger radius NWs film-based device (samples of 750 °C). Due to the surface morphology of the films transformed from 1D to 2D with the growth temperature increasing, the surface-to-volume ratio has decreased, which would leads to the light harvest area decreasing, as a result photogenerated carriers reduction.^32–35 Moreover, small NWs with size comparable to the wavelength of solar blind light are either mixed into the NWs film as light scattering centers or forming a scattering layer on the top of the film to reflect the incident light, aiming to extend the traveling distance of the incident light in the device.^36,37 It implies that the NWs film-based PDs R_λ can be enhanced by tune the nanowires size on the film surface.

To evaluate the performance of the smaller- and larger radius NWs film-based devices comparatively, the I–t photoresponse stability under 254 nm illuminations (70 μW cm⁻²) for more than 600 s had been investigated. They show excellent photoresponse stability to 254 nm light illuminations. The results show in Fig. 4(a) and (b), for smaller- and larger radius NWs film-based devices, respectively.

Fig. 4 Time-dependent photoresponse of Ga₂O₃/Al₂O₃ nanowires film-based photodetectors (a) and (b) the photodetector maintain its stability during 630 s under 254 nm light illuminations; (c) and (d) experimental curve and fitted curve of the current rise and decay process to 254 nm (70 μW cm⁻²) illuminations; at bias of 10 V, for smaller- and larger radius nanowires film based devices, respectively.

The response (rise) edges and the recovery (decay) edges usually consist of two components with a fast-response component and a slow-response component. The rise edges and the decay edges in smaller radius NWs film-based devices are obviously different from those in larger radius NWs film-based devices. For a more detailed comparison of the response time of the two NWs film-based devices, the quantitative analysis of the process of the current rise and decay process involves the fitting of the photoresponse curve with a bi-exponential relaxation equation as following:¹⁰

I = I₀ + Ae^−t/τ₁ + Be^−t/τ₂ (3)

where I₀ is the steady state photocurrent, t is the time, A and B are constant, τ₁ and τ₂ are two relaxation time constants. As the results show in Fig. 4(c) and (d) the photoresponse processes are well fitted, for smaller- and larger radius NWs film-based devices, respectively. τ_r and τ_d are the time constants for the rising edge and fall edge, respectively. We note that the rise edge and decay edge consisting of two components, (τ_r1 and τ_r2) and (τ_d1 and τ_d2) respectively. The rise edge constants τ_r1 and τ_r2 are estimated to be 1.26 s/11.38 s and 0.44 s/4.96 s for smaller- and larger radius NWs film-based devices, respectively. The decay edge constants τ_d1 and τ_d2 are estimated to be 1.99 s/26.15 s and 1.64 s/27.40 s for smaller- and larger radius NWs film-based devices, respectively. Generally, the fast-response component could be attributed to the rapidly change of the carrier concentration as soon as the light was turned on or off, while the slow-response component is caused by the carrier trapping/releasing owing to the existence of oxygen vacancies defects in the nanowires. It is evident that the smaller radius NWs film-based devices provides a much slower response speed to 254 nm than that of larger NWs film-based device. As the smaller radius NWs film has a higher defects and oxygen vacancies density (the dark current higher than larger radius NWs film), the carrier concentration changes slower and gentler than the larger radius NWs film-based device when the light on.¹⁰ In addition, the defects and oxygen vacancies would always capture photogenerated carriers and the extending traveling distance of the incident light in smaller radius NWs film-based device, causing a persistent rise and decay photoresponse.^16,17 As a result, with a lower defects and oxygen vacancies density, enhanced light scattering effects would lead to a slower response speed.

The Ga₂O₃/Al₂O₃ NWs films show significant high responsivity in the solar-blind PDs. All characteristic parameters of the NWs based devices (samples 650 °C and 750 °C) and smooth film-based device (sample 850 °C) are showed in Table 1. Although the obtained Ga₂O₃/Al₂O₃ NWs films-based solar-blind PDs still exhibit lower response speed and higher dark current, it could be greatly improved by optimizing NWs film growth technology and device structure.^19–21 Our findings propose a possible technique to realize solar-blind PDs with high responsivity.

Table 1 The characteristic parameters of Ga₂O₃/Al₂O₃ nanowires films-based (samples 650 °C and 750 °C) and smooth films-based (sample 850 °C) photodetectors under dark and 254 nm illuminations (70 μW cm⁻²) at bias of 10 V

Items	650 °C (NWs film)	750 °C (NWs film)	850 °C (film)
Dark current (nA)	512.78	109.35	1.24
Photocurrent (nA)	4637.74	657.34	2.95
Photocurrent/dark-current ratio	9.04	6.01	2.38
Responsivity (A W^−1)	1.40	0.19	5.83 × 10^−4
τr1/τr2 (s)	1.26/11.38	0.44/4.96	—
τd1/τd2 (s)	1.99/26.15	1.64/27.40	—

---

Conclusions

In conclusion, Ga₂O₃ nanowires array films are successfully grown on the α-Al₂O₃ (0001) substrates by inserting Al₂O₃ ultrathin interlayers through the LMBE technique. The crystallinity, surface morphology, band gap and photoconductivity of the as-grown films depend on the ion thermal diffusion and stress relaxation of Al₂O₃ layer insertion. Compared with smooth film-based device, the NWs films-based PDs exhibit much higher responsivity, which is attributed to that NWs films holding a larger surface-to-volume ratio and stronger light scattering effects. Our findings propose a possible technique to improve the responsivity of solar-blind PDs with excellent stability and tolerable response time. Ga₂O₃/Al₂O₃ nanowires films-based devices can be developed for future applications in solar-blind ultraviolet communication.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 51572241, 51572033, 11404029, 51172208), Beijing Natural Science Foundation (Grant No. 2154055), China Postdoctoral Science Foundation Funded Project (Grant No. 2014M550661).

References

1. L. Sang, M. Liao and M. Sumiya, Sensors, 2013, 13, 10482–10518.
2. F. Omnes, E. Monroy, E. Munoz and J. L. Reverchon, Proc. SPIE, 2007, 6473, 64730E.
3. E. Monroy, F. Omnes and F. Calle, Semicond. Sci. Technol., 2003, 18, R33–R51.
4. M. Y. Liao, L. W. Sang, T. Teraji, M. Imura, J. Alvarez and Y. Koide, Jpn. J. Appl. Phys., 2012, 51, 090115.
5. R. Mcclintock, A. Yasan and K. Minder, Appl. Phys. Lett., 2005, 87, 241123.
6. M. Razeghi, Proc. IEEE, 2002, 90, 1006–1014.
7. K. Koike, K. Hama and I. Nakashima, J. Cryst. Growth, 2005, 278, 288–292.
8. D. Y. Guo, Z. P. Wu, P. G. Li, Y. H. An, H. Liu, X. C. Guo, H. Yan, G. F. Wang, C. L Sun, L. H. Li and W. H. Tang, Opt. Mater. Express, 2014, 4, 1067–1076.
9. Y. Li, T. Tokizono and M. Liao, Adv. Funct. Mater., 2010, 20, 3972–3978.
10. D. Y. Guo, Z. P. Wu, Y. H. An, X. C. Guo, X. L. Chu, C. L. Sun, L. H. Li, P. G. Li and W. H. Tang, Appl. Phys. Lett., 2014, 105, 023507.
11. T. Takagi, H. Tanaka and S. Fujita, Jpn. J. Appl. Phys., 2003, 42, L401–L403.
12. M. Liao, Y. Koide and J. Alvarez, Appl. Phys. Lett., 2005, 87, 022105.
13. F. Zhang, K. Saito and T. Tanaka, Appl. Phys. Lett., 2014, 105, 162107.
14. F. Zhang, K. Saito and T. Tanaka, Solid State Commun., 2014, 186, 28–31.
15. T. Oshima, T. Okuno and N. Arai, Appl. Phys. Express, 2008, 1, 011202.
16. L. Li, E. Auer and M. Liao, Nanoscale, 2011, 3, 1120–1126.
17. C. Yan, N. Singh and P. Lee, Appl. Phys. Lett., 2010, 96, 053108.
18. W. Tian, C. Zhi and T. Zhai, J. Mater. Chem., 2012, 22, 17984–17991.
19. X. Chen, K. W. Liu, Z. Z. Zhang, C. R. Wang, B. H. Li, H. F. Zhao, D. G. Zhao and D. Z. Shen, ASC Appl. Mater. Interfaces, 2016, 8, 4185–4191.
20. X. Wang, W. Tian and M. Liao, Chem. Soc. Rev., 2014, 43, 1400–1422.
21. W. Weng, T. Hsueh and S. Chang, IEEE Trans. Nanotechnol., 2011, 10, 1047–1052.
22. D. Y. Guo, Z. P. Wu, P. G. Li, Q. J. Wang, M. Lei, L. H. Li and W. H. Tang, RSC Adv., 2015, 5, 12894–12898.
23. J. Li, X. Chen, Y. Zhi, M. He and H. Li, J. Phys.: Condens. Matter, 2001, 13, L937–L941.
24. K. Kaneko, H. Kawanowa and H. Ito, Jpn. J. Appl. Phys., 2012, 51, 020201.
25. D. Y. Guo, Z. P. Wu, Y. H. An, X. J. Li, X. C. Guo, X. L. Chu, C. L. Sun, M. Lei, L. H. Li, L. X. Cao, P. G. Li and W. H. Tang, J. Mater. Chem. C, 2015, 3, 1830–1834.
26. V. Consonni, M. Hanke and M. Knelangen, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 035310.
27. V. Dubrovskii, V. Consonni and L. Geelhaar, Appl. Phys. Lett., 2012, 100, 153101.
28. V. Consonni, Phys. Status Solidi RRL, 2013, 7, 699–712.
29. C. Cheze, L. Geelhaar and B. Jenichen, Appl. Phys. Lett., 2010, 97, 153105.
30. K. Kolasinski, Curr. Opin. Solid State Mater. Sci., 2006, 10, 182–191.
31. T. Oshima, T. Okuno and N. Arai, Jpn. J. Appl. Phys., 2009, 48, 070202.
32. Z. Galazka, R. Uecker and K. Irmscher, Cryst. Res. Technol., 2010, 45, 1229–1236.
33. J. Zhang and F. Jiang, Chem. Phys., 2003, 289, 243–249.
34. S. Bergin, Y. Chen and A. Rathmell, Nanoscale, 2012, 4, 1996–2004.
35. C. Kuo, J. Wu and S. Lin, Nanoscale Res. Lett., 2013, 8, 327.
36. O. Muskens, J. Rivas and R. Algra, Nano Lett., 2008, 8, 2638–2642.
37. D. Li, X. Fu and Y. Shao, Environ. Sci. Technol., 2013, 47, 9911–9917.

---