Open Access
Add to BibSonomyAbstract
Light-confinement phenomena on semiconducting ZnO nanograting structures were directly observed by means of confocal microscopy-based scanning photocurrent microscopy (SPCM), exhibiting a high spatial resolution distinguishing the 200 nm width of the ZnO nanostructure. Its ability to map the reflectance and photocurrent at the same time enables spatially resolved multiple light-confinement phenomena to be exhibited, such as the diffraction mode and the localized cavity-like resonant mode. Through diverse periods of the nanograting, in this case 600, 800 and 1000 nm, and various incident light intensity levels, we confirmed the period-dependent confined modes and thus established the ratio of the photocurrent change according to the incident intensity. Our study can provide accurate and comprehensive information regarding light confinement depending on the nanostructured geometry compared to conventional methods. This can assist those involved in creating effective designs of light management systems in nanostructured optoelectronics.
© 2016 Optical Society of America
1. Introduction
Light management is an indispensable factor in optoelectronics because it significantly influences their efficiency. Generally, the management has been achieved by geometrical effects, such as structural changes of the active layers and/or electrodes [1, 2], and by embedding additional particles in the interface layer and/or in the active layer [3, 4]. Such physically altered geometries induce certain optical resonances in the far-field or near-field, such as Mie, Fabry-Perot, waveguide-mode and diffraction-mode resonances, as well as localized surface plasmons and surface plasmon polaritons [5–7]. These light-management techniques have largely been studied in an effort to obtain an optimized efficiency for each type of optoelectronic device, such as LEDs (light-emitting diodes) [8], photodetectors [9] and solar cells [2, 3]. However, designs and evaluations of the geometrical effects have mostly been carried out by indirect methods, such as numerical simulations and average measurements of the reflectance (or absorbance) and external quantum efficiency (EQE), among others, though such tasks do not provide precise information about the local light confinement [10]. Furthermore, with regard to simulations, many optoelectrical parameters of nanostructured materials are still unknown. To solve these problems, direct observations using a scanning system with a focused beam are required. In this respect, scanning photocurrent microscopy (SPCM), also known as optical beam-induced current (OBIC) microscopy or laser beam-induced current (LBIC) microscopy, can be considered [11, 12]. This technique has been used mainly to examine certain characteristics of semiconductor devices, such as localized defects, the internal electric field distribution, the minority carrier diffusion lengths, and the doping concentration profiles [13, 14]. Although previous studies analyzing light confinement using an SPCM have been reported, photovoltaic devices [15, 16] and photodetectors [17] have their own designs. Therefore, they only exhibited specifically confined light phenomena depending on the individually designed devices and therefore provide limited information regarding the confined light on local geometries.
Here, we investigate light-confinement phenomena on a periodic ZnO nanograting structure using a SPCM based on confocal microscopy. Given that the periodic grating is a basic structure which provides fundamental light-confinement effects, comprehensive information regarding this type of confinement can be obtained. To ensure optimized measurement conditions for spatial mapping, we fabricated lateral-type devices (of the metal-semiconductor-metal type) with a semiconducting ZnO nanograting. The experiments were carried out systematically using three types of samples with 600, 800 and 1000 nm periods, along with various intensities of the incident light. The simultaneously measured reflectance and photocurrent images enabled us to observe the spatial distribution of the multiple instances of confined light, such as the diffraction mode and the localized cavity-like resonant mode. Furthermore, we confirmed the period-dependent confined modes and the resulting ratio of the photocurrent change according to the incident intensity level.
2. Experimental
Sample preparation
The deposition of the ZnO nanograting on two-inch SiO2/Si substrates was achieved by ultraviolet-nanoimprint lithography (UV-NIL) with hydrothermal growth [18, 19] via the following steps: (1) the solution of the ZnO precursor was spin-coated onto a substrate at 3500 rpm for 1 minute. (2) It was then prebaked on a hot plate at 80°C. (3) The three prepared polyurethane acrylate (PUA) molds, classified by 600, 800 and 1000 nm periods of a nanoline pattern, were attached to individual substrates, followed by the illumination of UV light onto the molds to cure the resin under a pressure level of 0.02 MPa. Consequently, a ZnO thin-film layer with a line nanopattern remained on the substrate. (4) The samples were then annealed in a furnace at 350°C for 1 hour to remove the organics and to crystallize the seed layer. (5) Finally, the three samples underwent wet etching with a 0.25% HNO3 solution such that only the line-nanopatterned ZnO remained. (6) In addition, hydrothermal growth was carried out by immersing the samples in a prepared solution at 60°C for 5 min. (7) The substrates were subsequently rinsed with deionized water and dried with blowing N2 gas. The solution process information of the ZnO precursor resin and the hydrothermal synthesis are described in detail in the literature [19]. To measure the photocurrent, metal-semiconductor-metal (MSM) devices were obtained by a shadow mask (gap size between electrodes: 50 μm) using a thermal evaporation process.
Characteristics
The morphologies of the synthesized ZnO nanogratings were confirmed by cold-type scanning electron microscopy (SEM) (S-4800, Hitachi). The crystallinity characteristics of the samples were determined by means of X-ray diffraction (XRD) analyses (D8 Advance, Bruker) carried out with Cu Kα radiation (wavelength, λ = 1.5418 Å). The 2θ range was 30−38°, and the scan speed was 0.2°/min. The energy band levels of the semiconducting ZnO nanogratings were characterized by the micro-photoluminescence (micro-PL) spectra using a Peltier-cooled charge-coupled device detector (Newton, Andor), a Plan Fluor 100 × objective lens (numerical aperture of 0.9) and a randomly polarized excitation laser with a wavelength of 355 nm (intensity: 7 kW/cm2, spot size on the samples: 481 nm). When measuring the micro-PL, we only select the highest intensity of the near-band edge emission to exclude the influence of periodic structures among the samples. The confocal microscopy-based SPCM, which is a custom-made setup, is separated into two measurement parts: the reflectance image and the photocurrent image. The light source of a randomly polarized continuous wave (CW) laser exhibiting a wavelength of 405 nm is illuminated onto the samples through a Plan Fluor 100x objective lens (NA 0.9). The spot size of the laser on the sample is 540 nm, and various intensities of light, in this case 34, 207, 387 and 570 μW/cm2 were used to examine the light confinement phenomena in more detail. The reflectance image resulting from normal incident light was measured by a photodiode after passing through a pin hole which eliminated the out-of-focus light. At the same time, the photocurrent image was measured by a lock-in amplifier to reduce environmental noise during the measurement step. The scanning step was set to 8 nm by means of a 2D piezo-stage. The two images were recorded simultaneously and sent to a computer.
3. Results and discussion
Figure 1(a) shows a photograph of a ZnO nanograting on a two-inch substrate synthesized by UV-NIL with hydrothermal growth, which can be adopted for roll-to-roll printing capable of a continuous process of large-area deposition. Figures 1(b)-1(e) shows SEM images of the three samples, exhibiting individually 1000, 800 and 600 nm periods of nanobeams (width: 200 nm, height: 165 nm). The average error in the dimensions is approximately ± 15 nm. The formation of crystalline ZnO was examined in the XRD patterns, as shown in Fig. 1(f). The patterns clearly present the (100), (002) and (101) planes of the wurtzite crystal structure of ZnO. However, the natively doped n-type semiconducting ZnO nanorods on the nanoline pattern have irregular morphology caused by the subtle conditions of the seed layer in each case and by hydrothermal growth, as shown in Fig. 2(a-c). Because the surface defect states of ZnO nanostructures are easily affected by their morphology, size and growth conditions, the optoelectrical properties of the samples can have unequal characteristics [20]. The micro-PL presenting the energy states of semiconductors have been frequently used in research of ZnO nanostructures. Normally, the spectral region of ZnO is largely divided into two regions: the near-band edge (NBE) emission related to good crystallinity and the deep-level emission (DLE) arising from the defect energy states. In our experiments, Fig. 2(d) showing the micro-PL spectra of the samples also presents the two peaks. However, due to the dissimilar morphologies, the energy states of each sample are unequal. Therefore, with respect to this situation, another objective of this study is to examine the influence of the somewhat dissimilar ZnO nanostructures on the locally confined light paths according to the reference to the periodicity because it is important for optoelectronic applications with ZnO nanograting structures prepared by UV-NIL with hydrothermal growth. Figure 3(a) shows a SPCM based on confocal microscopy. The wavelength of the incident beam is 405 nm, indicating that it excites the defect energy states regarding DLE, enabling the measurement of photocurrent images in our samples. This is because the photocurrent generated from the defect-energy states of the ZnO nanostructures has lower persistent photoconductivity than the photocurrent arising from the illumination of UV light. However, due to the resulting low photocurrent level, we applied a 20 V bias voltage to the devices. The experimentally estimated spatial resolution is less than 200 nm owing to the high scanning resolution (scan step of 8 nm) with the focused beam having a Gaussian profile, thus allowing the observation of localized light-confinement, as shown in the schematic diagram of Fig. 3(b).
Fig. 1 The photograph (a) shows the patterned ZnO nanograting on a two-inch wafer. The cross-sectional view of the SEM image (b) shows the nearly rectangular structure of the nanobeam. The scale bar in the inset is 300 nm. Top view of the SEM images shows the ZnO nanograting structures exhibiting periods of 1000 nm (c), 800 nm (d) and 600 nm (e). Though these samples have few fragments throughout the area, this factor did not significantly affect our experiment. (f) XRD patterns of the three samples.
Fig. 2 SEM images showing top views of samples, in this case with 1000 nm (a), the 800 nm (b) and the 600 nm periods, presenting the dissimilar morphologies of the samples resulting from the subtle conditions of the hydrothermal growth and the seed layers prepared by UV-NIL. The scale bar is 800 nm. The micro-PL spectra (d) show the energy states of semiconducting ZnO nanorods. In the normalized graph, the dissimilar positions of peaks in the NBE spectra indicate the nonidentical crystallinity among samples. In addition, the different intensities of the DLE spectra present the different defect energy states.
Fig. 3 The schematic (a) of the SPCM based on confocal microscopy. The reflectance coming from the reflected light has a high optical resolution due to the pin hole. After fabricating MSM-type devices with the semiconducting ZnO nanograting, a photocurrent of several nA was measured by means of a current preamplifier and a lock-in technique, which together remove s environmental noise during the measurement. The chopping frequency depending on the carrier dynamics and noises in the device was set to 2 kHz in our experiments. The reflectance and photocurrent were mapped simultaneously by a 2D piezo-stage. The schematic diagram (b) represents the experimental environment of the mapping on the ZnO nanograting structure.
Figure 4 shows the reflectance and photocurrent data on the ZnO nanograting structures. The dark line patterns in the reflectance images in Fig. 4(a) indicate ZnO nanostructures, as the incident light was absorbed at these regions, indicating that our mapping system can distinguish ZnO nanobeams with a width of 200 nm. To facilitate the analysis of the images, we summated each pixel of the two-dimensional (2D) image data in the vertical direction and converted it to one-dimensional (1D) data, as shown in Fig. 4(b). The black arrows indicate the sunken area of the reflectance in the space between the nanobeams, clearly indicating the light-trapping effect caused by adjacent ZnO nanobeams. In other words, both side regions of the arrows have partially confined light, inducing the maximum peak intensity of reflectance. The peak is at a distance of 270 nm from the edge of the nanobeam, which is nearly half of the spot size in the experiments. Regarding this and considering the Gaussian beam profile, the confined light is added to the center of the beam profile exhibiting high intensity, which leads to the maximum reflectance peaks. Since the added light is in a direction normal to the substrate because it was detected through an objective lens, the positions of the peaks could not involve a localized resonant mode, instead showing scattering. For the 800 and 600 nm periods, due to the reduced period, one peak was observed. Considering this situation, the reflectance data depends considerably on the diffraction mode arising from the periodic nanograting, inducing scattering. One possible reason to explain the existence of the diffraction mode in our experiments is the scanning spot size of 540 nm obtained from the full-width-at-half-maximum (FWHM) of the Gaussian profile, which could induce the (quasi)-diffracted mode. Interestingly, we also observed that when the distance between the nanobeams becomes narrower, the reflectance (the black curves with the gold arrows) in the space is brighter at the intensity level of 387 μW/cm2, whereas the reflectance (the gray curves with the green-yellow arrows) showed an opposite result under low intensity at 34 μW/cm2 (we describe this later in more detail).
Fig. 4 Each periodic nanograting sample was simultaneously mapped into reflectance (a) and photocurrent (c) images. The white scale bar is 300 nm and the black dashed rectangles indicate the ZnO nanobeam. The stacked graph (b) is the 1D data resulting from the summation of each reflectance images (a) in the vertical direction, with identical y-axis ranges. The stacked graph (d) represents the photocurrent subtracted from the offset value after the summation of each pixel of photocurrent image (c) in the vertical direction. The sky-blue rectangles in both graphs (b,d) indicate the ZnO nanobeam, and the high (the black curves) and low (the gray curves) intensities of the incident light are 387 and 34 μW/cm2, respectively. In figure (b), the gold (for 387 μW/cm2) and green-yellow (for 34 μW/cm2) dashed lines indicate the maximum intensity of reflectance in the 1000 nm period. That is, as the period becomes narrower, the maximum reflectance under high intensity of the incident light increases, whereas the maximum under a low intensity of the incident light is reduced.
According to the reflectance data measured by detecting normal directional light, it is difficult to express the localized resonant mode adjacent to the nanograting structures. In this respect, measurement of the photocurrent image may help to identify the probable resonances because the region in which the photocurrent is generated indicates an area of concentrated electric fields on the surface. The photocurrent data in Fig. 4(c) and 4(d) have several differences in comparison with the reflectance data: (1) The sunken level (black arrows) of the photocurrent in the middle space was observed in the 1000 nm period as well as in the 800 nm period. (2) The sunken level in the space of the 1000 nm period is deeper than the reflectance when comparing each magnitude of the maximum, the minimum, and the sunken region. (3) The distance (165 nm) of the maximum photocurrent peak from the edge of the nanobeam is shorter than that (270 nm) in the reflectance data, which indicates that the positions of the peaks are distinguished from the diffraction mode causing the reflectance. (4) Furthermore and importantly, the locations of the maximum peaks in the 1000 and 800 nm periods under high (black curves) and low (gray curves) light intensity levels are all identical. From these observations, the existence of localized cavity-like resonance arising at the corner between a nanobeam and the substrate is certain, as the features and locations of the peaks can establish a localized mode distinguished from the diffraction mode arising from the periodic nanograting structure. The spatial range of the localized confinement originating from the geometry of the nanobeam appears to be related to the size of the nanobeam because the location (165 nm) is similar to its height. In the 600 nm period, discriminatively, the maximum peak clearly exists on the top of the ZnO nanobeam. This is due to the correlation between the space distance (400 nm ± 15 nm) between the nanobeams and the wavelength (405 nm) of the incident light, which raises the position of the concentrated electromagnetic field on the top surface of the nanostructure and thus leads to a strong diffraction mode. Furthermore, we seriously considered the presence of the surface plasmon on the surface of the ZnO nanostructure. However, it is thought that the influence of the surface plasmon on the generation of photocurrent is much smaller than that of photocurrent arising from the localized mode because of surface defect states on the ZnO nanostructure as confirmed by DLE emission in the micro-PL spectra. Therefore, we ignored the consideration of the surface plasmon effect on the photocurrent. However, there is the possibility for the existence of weak surface plasmon resonance on the n-type semiconducting ZnO nanostructure.
To investigate the spatial distribution of the confined light in more detail, we measured the reflectance and photocurrent images according to the diverse intensity of the incident light, at 34, 207, 387 and 570 μW/cm2. In Fig. 5(a), the order of the reflectance magnitudes among the samples at 34 μW/cm2 was reversed above 207 μW/cm2 (the blue dashed ellipses), which is attributed to the nonlinear optical properties of the semiconducting ZnO nanostructure in combination with the illuminated intensity on the surface resulting from the light-trapping efficiency [21]. Above 150 μW/cm2, the period 600 nm has higher reflectance than other samples, implying that the 600 nm period has the strongest effect of the diffraction mode. As mentioned earlier, this is due to the distinct interaction with light as compared to the other samples. In the normalized inset, above 387 μW/cm2 (the pink dashed arrow), all of the reflectance is nearly saturated owing to the geometrical effect and the high intensity of the incident light.
Fig. 5 Simultaneously measured reflectance (a) and photocurrent (b) according to various intensities of incident light were prepared by subtracting the offset value (the minimum value) from the maximum value in order to relatively compare the geometrical effects for light confinement among different periods. The pink dashed arrows in the inset of both graphs indicate the inflection point which means that the geometrical effect is nearly saturated above 387 μW/cm2.
In Fig. 5(b), the photocurrent for the period of 800 nm is highest among the samples, while the 1000 nm period has the lowest value throughout the range of the incident intensity. Further, in the normalized inset, as the period is reduced, there exists an early slow increase (the blue dashed up arrow) and a final rapid decrease (the blue dashed down arrow) based on 387 μW/cm2 (the pink dashed arrow). These findings are attributed to the relative multiple light confinement effects associated with the localized cavity-like mode and the diffraction mode. Concerning only the localized mode, it could have limited resonant intensity owing to the restricted resonant area. On the other hand, the diffraction mode depending on the periodicity has a relatively large resonant area leading to high resonant intensity. Therefore, the diffraction mode has higher sensitivity for a broad range of incident intensity than the localized mode. In this respect, the 1000 nm period exhibits mainly localized resonance at a low intensity level because it has a weak effect of the diffraction mode, which causes the lowest photocurrent in Fig. 5(b) and an early rapid increase followed by a slow decrease according to the intensity as shown in the inset. While, the 600 nm period having a strongdiffraction on top of ZnO the nanobeam results in an early slow increase followed by a final rapid decrease, which indicates a rapid saturation due to the weakened geometric effect. On the other hand, this implies that the side wall of the nanobeam in the 600 nm period is underutilized. In case of the 800 nm period, it has the highest photocurrent at all intensity levels because both modes are present strongly, differing from the 600 and 1000 nm periods. Also, due to both modes, its rate of increase and decease according to the incident intensity is in the middle between the 1000 nm and the 600 nm periods, as shown in the inset of Fig. 5(b).
Furthermore, to elaborate on the effects of the period and the incident intensity on the localized cavity-like resonance, we compared the spatial 1D photocurrent data, as shown in Fig. 6(a). In the figure, the slopes (cyan dashed lines) of the photocurrent are lowered with an increase in the offset values according to the increase in the intensity and the reduced periods, providing evidence of the incident intensity and the diffraction mode caused by the period affecting the photocurrent in the localized resonant region. In this context, we confirmed the relative ratio of the photocurrent change according to the incident intensity among the samples, as shown in Fig. 6(b). The 1000 nm period has a maximum peak (black arrow) at a low intensity level, showing that it has a relatively dominant effect on the localized resonance, whereas the 600 nm period having a strong diffraction mode has a maximum peak (green arrow) at a high intensity level. For the 800 nm period, the photocurrent exists in a relatively wide intensity range (red arrows) owing to the mixed confinement effect.
Fig. 6 (a) Changing photocurrent shapes subtracted from offset value near the ZnO nanobeam according to the period and intensity of light. (b) The maximum photocurrent divided by the offset photocurrent. The arrows present the maximum position of the photocurrent with respect to the intensity of light. The stacked schematic (c) presents the photocurrent shape on the ZnO nanobeam according to the period. The red dashed lines indicate the light paths resulting from the diffraction mode. The yellow dashed lines represent the localized cavity-like resonances. The red dashed ellipses denote the illuminated portion on the surface of the ZnO nanobeam due to the diffraction mode. The blue arrows present the ascending illumination of the diffraction mode due to the reduced period.
To explain the period-dependent multiple light confinements of the nanograting structures, a schematic diagram is shown in Fig. 6(c). As the period is reduced, the portion of the confined light (the red dashed arrows) resulting from the diffraction mode increases. Therefore, with the 1000 nm period, the portions illuminated by the weak diffraction mode (the red dashed ellipses) exist in the low area of the side wall due to the wide period, which brings out a relatively dominant effect of the localized mode. With the 600 nm period, because the period is most reduced in this case, the portion by the strong diffraction mode exists on the top surface of the ZnO nanobeam, which leads to the position of the maximum photocurrent peak on the top of the ZnO nanobeam. And it does not influence significantly the photocurrent in the localized resonant region as compared to the other samples. In contrast, the light portions caused by the diffraction mode in the 800 nm period exists in the middle of the side wall. Therefore, its range is widest in the side wall, leading to an increase in the photocurrent value in combination with the localized mode. Consequently, through our mapping system, we confirmed that the period-dependent multiple instances of light confinement on the nanograting structures individually determine the amount of photocurrent and the ratio of the photocurrent change relative to the incident intensity. Moreover, we confirmed that the somewhat dissimilar morphologies of the ZnO nanorods prepared by UV-NIL with hydrothermal growth do not considerably affect the geometry-dependent light confinement outcomes, suggesting the feasible application of light-confinement with the ZnO nanograting structure.
4. Conclusion
In summary, we demonstrated an experimental observation of the light-confinement phenomena on a periodic ZnO nanograting structure owing to the simultaneous mapping of the reflectance and photocurrent by means of confocal microscopy-based SPCM. The reflectance images demonstrate the confined light phenomenon regarding the diffraction mode depending on the periodicity and the photocurrent images show the localized cavity-like resonant mode associated with the geometry of the ZnO nanobeams. Furthermore, with diverse periodicity and various intensity levels of the incident light, we confirmed the period-dependent multiple light-confinement which determines the amount of photocurrent and the ratio of the photocurrent change according to the incident intensity level. We believe that our study providing an accurate distribution of confined light paths on the nanostructure suggests a novel approach to investigate effectively the light-confinement effect in nanostructured optoelectronics.
Acknowledgments
We gratefully acknowledge support from the Global Frontier R&D Program by the Center for Advanced Meta-materials (CAMM) by the National Research Foundation under the Ministry of Science, ICT and Future Planning (2014M3A6B3063707) and the Basic Research Fund of Korea Institute of Machinery and Materials (NK200L, SC1170).
References and links
1. C. W. Teplin, B. G. Lee, T. R. Fanning, J. Wang, S. Grover, F. Hasoon, R. Bauer, J. Bornstein, P. Schroeter, and H. M. Branz, “Pyramidal light trapping and hydrogen passivation for high-efficiency heteroepitaxial (100) crystal silicon solar cells,” Energy Environ. Sci. 5(8), 8193–8198 (2012). [CrossRef]
2. U. Palanchoke, V. Jovanov, H. Kurz, P. Obermeyer, H. Stiebig, and D. Knipp, “Plasmonic effects in amorphous silicon thin film solar cells with metal back contacts,” Opt. Express 20(6), 6340–6347 (2012). [CrossRef] [PubMed]
3. D. Liang, Y. Huo, Y. Kang, K. X. Wang, A. Gu, M. Tan, Z. Yu, S. Li, J. Jia, X. Bao, S. Wang, Y. Yao, H. S. P. Wong, S. Fan, Y. Cui, and J. S. Harris, “Optical absorption enhancement in freestanding GaAs thin film nanopyramid arrays,” Adv. Energy Mater. 2(10), 1254–1260 (2012). [CrossRef]
4. J. L. Wu, F. C. Chen, Y. S. Hsiao, F. C. Chien, P. Chen, C. H. Kuo, M. H. Huang, and C. S. Hsu, “Surface plasmonic effects of metallic nanoparticles on the performance of polymer bulk heterojunction solar cells,” ACS Nano 5(2), 959–967 (2011). [CrossRef] [PubMed]
5. A. P. Vasudev, J. A. Schuller, and M. L. Brongersma, “Nanophotonic light trapping with patterned transparent conductive oxides,” Opt. Express 20(10), A385–A394 (2012). [CrossRef] [PubMed]
6. H. A. Atwater and A. Polman, “Plasmonics for improved photovoltaic devices,” Nat. Mater. 9(3), 205–213 (2010). [CrossRef] [PubMed]
7. M. L. Brongersma, Y. Cui, and S. Fan, “Light management for photovoltaics using high-index nanostructures,” Nat. Mater. 13(5), 451–460 (2014). [CrossRef] [PubMed]
8. J. Hauss, T. Bocksrocker, B. Riedel, U. Geyer, U. Lemmer, and M. Gerken, “Metallic Bragg-gratings for light management in organic light-emitting devices,” Appl. Phys. Lett. 99(10), 103303 (2011). [CrossRef]
9. F. P. G. de Arquer, F. J. Beck, M. Bernechea, and G. Konstantatos, “Plasmonic light trapping leads to responsivity increase in colloidal quantum dot photodetectors,” Appl. Phys. Lett. 100(4), 043101 (2012). [CrossRef]
10. C. R. McNeill, H. Frohne, J. L. Holdsworth, and P. C. Dastoor, “Near-field scanning photocurrent measurements of polyfluorene blend devices: directly correlating morphology with current generation,” Nano Lett. 4(12), 2503–2507 (2004). [CrossRef]
11. R. Graham and D. Yu, “Mod. “Scanning photocurrent microscopy in semiconductor nanostructures,” Phys. Lett. B 27(25), 1330018 (2013).
12. P. Parkinson, Y. H. Lee, L. Fu, S. Breuer, H. H. Tan, and C. Jagadish, “Three-dimensional in situ photocurrent mapping for nanowire photovoltaics,” Nano Lett. 13(4), 1405–1409 (2013). [PubMed]
13. M. Engel, M. Steiner, R. S. Sundaram, R. Krupke, A. A. Green, M. C. Hersam, and P. Avouris, “Spatially resolved electrostatic potential and photocurrent generation in carbon nanotube array devices,” ACS Nano 6(8), 7303–7310 (2012). [CrossRef] [PubMed]
14. E. J. Lee, K. Balasubramanian, J. Dorfmüller, R. Vogelgesang, N. Fu, A. Mews, M. Burghard, and K. Kern, “Electronic-band-structure mapping of nanotube transistors by scanning photocurrent microscopy,” Small 3(12), 2038–2042 (2007). [CrossRef] [PubMed]
15. S. L. Howell, S. Padalkar, K. Yoon, Q. Li, D. D. Koleske, J. J. Wierer, G. T. Wang, and L. J. Lauhon, “Spatial mapping of efficiency of GaN/InGaN nanowire array solar cells using scanning photocurrent microscopy,” Nano Lett. 13(11), 5123–5128 (2013). [CrossRef] [PubMed]
16. G. Mariani, P. S. Wong, A. M. Katzenmeyer, F. Léonard, J. Shapiro, and D. L. Huffaker, “Patterned radial GaAs nanopillar solar cells,” Nano Lett. 11(6), 2490–2494 (2011). [CrossRef] [PubMed]
17. H. Yuan, X. Liu, F. Afshinmanesh, W. Li, G. Xu, J. Sun, B. Lian, A. G. Curto, G. Ye, Y. Hikita, Z. Shen, S. C. Zhang, X. Chen, M. Brongersma, H. Y. Hwang, and Y. Cui, “Polarization-sensitive broadband photodetector using a black phosphorus vertical p-n junction,” Nat. Nanotechnol. 10(8), 707–713 (2015). [CrossRef] [PubMed]
18. S. H. Ahn and L. J. Guo, “Large-area roll-to-roll and roll-to-plate nanoimprint lithography: a step toward high-throughput application of continuous nanoimprinting,” ACS Nano 3(8), 2304–2310 (2009). [CrossRef] [PubMed]
19. E. Chong, S. Kim, J. H. Choi, D. G. Choi, J. Y. Jung, J. H. Jeong, E. S. Lee, J. Lee, I. Park, and J. Lee, “Interior-architectured ZnO nanostructure for enhanced electrical conductivity via stepwise fabrication process,” Nanoscale Res. Lett. 9(1), 428 (2014). [CrossRef] [PubMed]
20. C. Park, J. Lee, and W. S. Chang, “Geometrical separation of defect states in ZnO nanorods and their morphology-dependent correlation between photoluminescence and photoconductivity,” J. Phys. Chem. C 119(29), 16984–16990 (2015). [CrossRef]
21. M. K. Kavitha, K. B. Jinesh, R. Philip, P. Gopinath, and H. John, “Defect engineering in ZnO nanocones for visible photoconductivity and nonlinear absorption,” Phys. Chem. Chem. Phys. 16(45), 25093–25100 (2014). [CrossRef] [PubMed]
