n-ZnO Nanotube Heterojunctions with

Sep 19, 2017 - We measure the current–voltage characteristics of the device under dark and illuminated conditions and demonstrated the high stabilit...
0 downloads 0 Views 5MB Size
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces 2017, 9, 37120-37127

A Photodetector Based on p‑Si/n-ZnO Nanotube Heterojunctions with High Ultraviolet Responsivity Tahani H. Flemban, Md Azimul Haque, Idris Ajia, Norah Alwadai, Somak Mitra, Tom Wu, and Iman S. Roqan* King Abdullah University of Science and Technology (KAUST), Physical Sciences and Engineering Division, Thuwal 23955-6900, Saudi Arabia S Supporting Information *

ABSTRACT: Enhanced ultraviolet (UV) photodetectors (PDs) with high responsivity comparable to that of visible and infrared photodetectors are needed for commercial applications. n-Type ZnO nanotubes (NTs) with high-quality optical, structural, and electrical properties on a p-type Si(100) substrate are successfully fabricated by pulsed laser deposition (PLD) to produce a UV PD with high responsivity, for the first time. We measure the current−voltage characteristics of the device under dark and illuminated conditions and demonstrated the high stability and responsivity (that reaches ∼101.2 A W−1) of the fabricated UV PD. Time-resolved spectroscopy is employed to identify exciton confinement, indicating that the high PD performance is due to optical confinement, the high surface-to-volume ratio, the high structural quality of the NTs, and the high photoinduced carrier density. The superior detectivity and responsivity of our NT-based PD clearly demonstrate that fabrication of high-performance UV detection devices for commercial applications is possible. KEYWORDS: photodetectors, oxides, time-resolved measurements, nanotubes, heterojunctions, ultraviolet, pulsed laser depositions



applied fields and no oxygen dependency compared to ZnObased Schottky contact UV detectors, which require a high applied field and rely on the oxygen adsorption/desorption process.16 This adsorption/desorption process can be explained as free electrons from the surface being trapped by oxygen molecules in the dark, to form a chemically adsorbed surface state (O2 + e− → O2−). Under illumination at a photon energy above the band gap, electron−hole pairs are photogenerated, holes are captured in the surface, and the negatively charged adsorbed oxygen ions are discharged (h+ + O2− → O2). The free electrons produced during the photodesorption process will increase the photoconductivity.2,4 Wang and co-workers achieved ultrafast UV photoresponse in p-Si/n-ZnO, which is compatible to Si-based PDs.15 However, the UV responsivity of presently available ZnO-based p−n devices currently remains low.16−23 Enhancing the responsivity of UV PDs to a level comparable to that of visible-light PDs is thus needed before inexpensive and commercially viable UV PD devices can be fabricated. Here, we report on optimized ZnO nanotubes (NTs) grown on p-Si by pulsed laser deposition (PLD) and demonstrate their high-quality optical, structural, and electrical properties suitable for UV photodetection. This is the first work reporting on a ZnO NTs-based PD fabricated by direct growth, without a

INTRODUCTION Room-temperature (RT) ultraviolet (UV) photodetectors (PDs) have applications in astronomy, chemical and biological detection, optical communications, flame sensing, and national defense.1 However, obtaining high responsivity and stability in the UV range of the spectrum that is comparable to that in the visible and infrared ranges remains a challenge. Commercially available Si-based PDs have low UV responsivity due to several limitations, such as the indirect and small band gap energy of Si (1.1 eV), the effect of temperature variations on PD characteristics, and the need for high-pass optical filters and phosphors to stop low-energy photons.2,3 Therefore, the performance of UV PDs based on Si can be reduced due to the low efficiency and greater dark currents of Si. Consequently, the development of UV PDs using wide band gap semiconductors (ZnO, for example) would overcome these limitations. Wurtzite ZnO is an ideal RT UV PD4−6 due to its wide and direct band gap (3.3 eV at RT); high exciton binding energy (60 meV) compared to the RT thermal energy (25 meV), which provides stability against thermal dissociation of excitons;2 environmentally friendly properties; and chemical and thermal stability.7,8 Furthermore, ZnO has unique electric and piezophototronic properties that can be exploited in PD technologies.5,6,9,10 The high stability of ZnO polar surface has been used to induce the formation of many nanostructures using different techniques.11 Therefore, significant efforts have been devoted to the development of ZnO-based UV PDs.12−15 ZnO-based p−n heterojunctions have the advantage of low © 2017 American Chemical Society

Received: July 4, 2017 Accepted: September 19, 2017 Published: September 19, 2017 37120

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces

Figure 1. ZnO NTs grown on p-Si(100) under different laser fluences at P(O2) = 100 mTorr: (a) 6.2 J cm−2, (b) 4.4 J cm−2, (c) 4.0 J cm−2, (d) 3.6 J cm−2, (e) 2.8 J cm−2, and (f) 1.6 J cm−2. variable-temperature closed-helium cryostat for measurements in the 4−300 K temperature range. Device Fabrication. To fabricate the device, we deposited silver as a bottom electrode. A 150 nm indium−tin oxide (ITO) contact serving as a top electrode was deposited using a shadow mask (Figure S1 in the Supporting Information) by radio frequency magnetron sputtering during in situ thermal annealing at 150 °C (to increase the ITO transparency and avoid high resistivity), using argon plasma at a constant current of 0.15 A for 85 min and a working pressure of 4 mTorr. We measured all device photoresponse characteristics using a probe station connected to a Keithley 4200 semiconductor analyzer under dark and UV-illuminated conditions (λ = 365 nm, whereby the light intensity was tuned using neutral density filters and was calibrated using an optical power meter). The responsivity (R) was calculated as R = ΔI/PS, where ΔI is the difference between the photocurrent and the dark current, P is the incident power density, and S is the effective area illuminated by the UV lamp (λ = 365 nm). The external quantum efficiency (EQE) was calculated from the responsivity using the expression EQE = Rhc/eλ, where h is Planck’s constant, c is the speed of light, e is the electronic charge, and λ is the incident light wavelength.29 Specific detectivity (D*) was calculated using the equation D* = R/(2eJdark)1/2, where Jdark is the dark current density. We assumed that the noise was dominated by shot noise.29

seed layer or the need for costly and complex fabrication processing, leading to superior performance UV PDs. We also show a high-performance p-Si/n-ZnO NT heterojunction device with UV responsivity. We grew one-dimensional (1D) ZnO NTs on p-Si to fabricate a p−n heterojunction with superior responsivity and high quantum efficiency. PDs based on 1D nanostructured materials usually exhibit high responsivity and quantum efficiency due to their large surface-to-volume ratio.24−26 In addition, these 1D devices showed significant carrier transport owing to discontinuous interfaces, low grain boundary densities, and limited two-dimensional (2D) charge transport.4,27,28



EXPEREMINTAL SECTION

Sample Preparation. The ZnO target for PLD was conventionally prepared from ZnO (99.99% pure) powder (Sigma-Aldrich). A pellet of 2.5 cm diameter was pressed under 100 bar pressure and sintered at ∼1100 °C. n-Type ZnO NTs were grown on p-Si without seeding or catalysts by a one-step PLD method. We used a krypton fluoride (krF) excimer laser with 248 nm wavelength. The laser frequency, targetsubstrate vertical distance, and partial oxygen pressure were fixed at 10 Hz, ∼9 cm, and 100 mTorr, respectively, at a deposition temperature of 650 °C. We used a p-type Si (100) wafer doped on a p-type boron substrate with very low resistivity (0.001−0.005 Ω cm). Structural Characterizations. We used scanning electron microscopy (SEM) (FEI Nova Nano 630), high-resolution transmission electron microscopy (HR-TEM) (Titan), and X-ray diffraction (XRD) (Bruker D8 with Cu Kα and λ = 1.5406 Å) for structural characterizations. Optical Characterization and Carrier Dynamics. We investigated the optical properties of ZnO NT through photoluminescence (PL) measurements conducted using a 325 nm He−Cd laser with a beam power of 8 mW. The samples were mounted in a He closedcycle cryostat system for low-temperature PL measurements (6 K). The spectra were detected by an Andor spectrograph with a UV− visible charge coupled device (CCD) camera, allowing the PL spectra to be detected and analyzed. We performed temperature-dependent time-resolved PL (TRPL) experiments by third harmonic UV (λ = 266 nm) pulses of a mode-locked Ti:sapphire femtosecond pulsed laser with a pulse width of ∼150 fs and an average pulse power intensity of ∼2.1 kW cm−2 (with a 76 MHz repetition rate). We used a coherent Verdi-V18 diode-pumped solid-state continuous wave laser to pump the Ti:sapphire laser and detected the emission emanating from the sample with a Princeton Instruments spectrograph attached to a UVsensitive Hamamatsu C6860 streak camera with a temporal resolution of ∼2 ps, using synchroscan mode. We mounted the sample in a



RESULTS AND DISCUSSION Nanorod (NR) and NT formation occurs initially though material nucleation as a result of the initial textured layer that is formed on the Si substrates during deposition due to the high lattice mismatch between Si and ZnO.30 Depending on the surface energy of the layers, this gives rise to Stranski− Krastanov (SK) nucleation, followed by NR and NT growth.31,32 In our work, this textured layer was formed during ZnO NT growth on p-Si using a one-step method. We found that the laser fluence can affect the nanostructure type (such as NRs, NTs, or quantum dots) formed on a p-Si substrate using PLD due to the changes in the kinetic energy E of the charged species upon arrival at the substrate surface, leading to NT formation. This relationship is described by the following equation33 ⎛ d⎞ E = Eο exp⎜ − ⎟ ⎝ λ⎠

(1)

where Eο is the initial energy of the charged species escaping from the target and d is the distance between the target and the substrate, while λ is the mean free path of the ablated species 37121

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces (the plasma) traveling toward the substrate in the PLD chamber. We studied the effect of the growth temperature, oxygen partial pressure [P(O2)], laser fluence, and the number of laser pulses to optimize the growth. We found that 650 °C and 100 mTorr were the optimal growth temperature and P(O2), respectively, for forming ZnO NTs. Then, we investigated the effect of laser fluence by depositing samples by laser fluence ranging from 1.6 to 6.2 J cm−2 and 30 000 pulses. Figure 1 shows SEM images of samples deposited at different fluence conditions. Samples deposited using high laser fluence (in the 4−6.2 J cm−2 range) exhibit dense, filmlike ZnO NRs, the size of which decreases with the decrease in laser fluence (Figure 1a−c). When the laser fluence decreases to 3.6 J cm−2 (Figure 1d), NT formation commences. Figure 1e shows that 2.8 J cm−2 is the optimal fluence for obtaining high NT density, as further decreasing the fluence to 1.6 J cm−2 results in a decline in ZnO NT density (Figure 1f). We observed that NT formation requires slightly lower energy, as shown in Figure 1. We also studied the effect of the number of pulses by increasing the number of pulses from 30 000 to 40 000 at the laser fluence of 2.8 J cm−2 to optimize the NT quality (Figure 2) (Figure S2 in Supporting Information). We

Figure 3. HR-TEM of a cross section of ZnO NTs near the interface, indicating that the hollowing emerged from the interface.

Figure 4. XRD pattern of ZnO NTs grown on p-Si substrate (log scale).

ZnO planes, respectively. The ZnO NT lattice constants were calculated using Bragg’s law, whereby the spacing between the planes in hexagonal structures is given by Miller indices h, k, and l.38 The a and c lattice parameters are calculated as a = 3.14 Å and c = 5.20 Å. To investigate the material quality, the optical properties of ZnO NTs were studied. Low-temperature (6 K) and RTPL spectra reveal strong and sharp band gap peaks at 3.36 and 3.27 eV, respectively, with a very weak defect band in the visible region (Figure 5a), confirming the high quality of the ZnO NT structures.39 The weak defect band indicates that the density of the trap states in the band gap introduced by defects is low, resulting in a significant improvement in the photocurrent that rapidly reaches a steady state, as will be shown later. This phenomenon leads to substantial PD UV responsivity.26,40,41 As 1D confinement leads to high PD responsivity, we studied the temperature-dependent PL and TRPL to investigate the confinement behavior of the NTs. Figure 5b shows the temperature-dependent PL of ZnO NTs. As the temperature increases, the near-band-edge (NBE) emission shows a red shift. The dependence of the NBE exciton peak energy on temperature is determined by Varshni’s formula42 and indicates that NBE is dominated by free excitons; similar behavior was observed in 1D confinement in ZnO quantum wells.43 In order to elucidate the confinement behavior of the ZnO NTs, we studied temperature-dependent TRPL to investigate the carrier dynamics (Figure S3 in the Supporting Information). Figure 5c shows the TRPL spectra of ZnO NT samples at 4 K and RT. The TRPL decay lifetime of the sample followed a nonexponential trend in all temperature ranges. The TRPL spectra were fitted, with excellent convergence, using the following biexponential decay model44

Figure 2. ZnO NTs grown by increasing the number of pulses (up to 40 000) with a laser fluence of 2.8 J cm−2: (a) SEM top view, where the inset is a closeup image of a single NT, and (b) cross-sectional SEM image.

found that the optimal NT structure was achieved at 2.8 J cm−2 laser fluence and 40 000 pulses at a growth temperature of 650 °C and 100 mTorr pressure. Figure 2 shows optimized ZnO NTs with a well-shaped hollowing that is on average ∼650 nm long with a wall thickness (d) of ∼10−25 nm and an external radius (r) of ∼90−110 nm (including the hollowing). In this case, the ratio of the wall thickness to the outer radius (d/r) is smaller than (3/8)1/2. At this ratio, light is confined via total internal reflection at the interface between the external and internal surface of the tube.34,35 This internal reflection can improve PD responsivity.36 HR-TEM images of a cross-section taken parallel to the [0001] c-axis near the interface between a ZnO NT and the pSi substrate shown in Figure 3 confirm that the hollowing starts from the bottom of the NT near the interface, resulting in enhanced light absorption of ZnO NT in comparison to ZnO NR due to the high aspect ratio.37 The SEM and HR-TEM results also reveal a significant surface-to-volume ratio, which ensures that a highly active area is exposed to light to produce sufficient photoinduced carrier density. Figure 4 shows XRD measurements that reveal the highquality crystalline structure of the n-ZnO NTs. A sharp and intense peak [at 2θ = 32.94° with a small full width at halfmaximum (fwhm) of 0.011°] and a minor peak (at 2θ = 34.42° with a fwhm of 0.213°) correspond to the (100) and (002)

It = A1e−t / τ1 + A 2 e−t / τ2

(2)

where A1 and A2 are the relative initial intensities of the fast and slow decay components, respectively, τ1 and τ2 are the fast and slow lifetimes, respectively, and It is the time-dependent 37122

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces

Figure 5. (a) PL of ZnO NTs at 6 K and RT. (b) Temperature-dependent PL, where the inset shows the photon energy as a function of temperature. (c) TRPL of ZnO NTs at RT and 4 K with a double-exponential decay fit (blue and magenta lines) (log scale). (d) Temperature dependence of radiative and nonradiative recombination lifetimes. The dotted line represents the bulk behavior.

Figure 6. (a) A schematic illustration of the structure of our p-Si/n-ZnO NTs UV PD. (b) An energy band diagram of our PD.

intensity of the photoemission from the sample. As τ1 and τ2 at 4 K (RT) were 10.7 and 35.7 ps (7.9 and 25.2 ps), respectively, our results are in good agreement with the previously reported values for ZnO nanostructures.45,46 The nonexponential decay of nanostructures has been extensively discussed in pertinent literature.39,46,47 Fast decay, τ1, has been attributed to the effects of surface recombination states when the surface-to-volume ratio is high, whereas slow decay, τ2, has been ascribed to the bulklike states in the thicknesses of NT walls,46 leading to multiexcitation centers. Figure 5d shows PL lifetimes as a function of temperature and reveals the carrier confinement behavior of the NTs. We extracted the radiative (τrad) and nonradiative (τnr) lifetimes using the following expressions48 τrad(T ) = τPL(T )/ηth(T )

where τPL(T) is the PL lifetime of the NTs at any given temperature and ηth(T) is the ratio between the PL intensity at any given temperature and at 4 K. The validity of this method is based on the assumption that internal quantum efficiency is at its maximum at 4 K.48 We observed that the radiative lifetime seems to increase in an approximately linear fashion as the temperature increases; that is, τrad ∝ T, which is a signature of 1D-like confinement of carriers.49,50 The three-dimensional (3D) behavior exhibited by bulk materials (represented by the dotted line in Figure 5d) diverges significantly from the experimentally observed dependence of the radiative lifetime on temperature. As the NT wall thickness (10−25 nm) is greater than the Bohr diameter (2aB), 1D-like confinement can be ascribed to the quantum-confined Stark effect (QCSE) induced by both spontaneous and piezoelectric polarizations in wurtzite ZnO.51 In addition, the reduction in the dimensionality of the thickness of NT walls decreases the photoinduced carrier transit time, which can result in improving responsivity.4,41

(3)

and τnr(T ) = τPL(T )/[1 − ηth(T )]

(4) 37123

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces

Figure 7. (a) I−V characteristics of our p-Si/n-ZnO PD under dark and 365 nm illumination conditions with different power densities. (b) Photoresponse under 365 nm UV illumination of 1 mW cm−2 and bias voltage of ±2 V. (c) Rise and fall photoresponse time of PD fit by doubleexponential functions under 365 nm illumination with an intensity of 1 mW cm−2 and a bias of −2 V. (d) The responsivity and detectivity under different bias voltages for a fixed illumination power of 365 nm with 1 mW cm−2. (e) Responsivity and detectivity under 365 nm illumination with different light intensities and a bias of −2 V.

Table 1. Progress in the ZnO-NT-Based PD and Other High-Performance ZnO-Based PDs structure

electrode

p-Si/n-ZnO NWs p-Si/n-ZnO NWs p-Si/n-ZnO p-Si/n-ZnO NRs ZnO film (MSM) ZnO NWs (MSM)

Au/Ti/ITO Ti Ti/Al Ag Au/Pt Au

ZnO nanoparticles film (MSM) p-Si/n-ZnO NTs

Au ITO

bias voltage (V) −1 −20 −2 −4 3 10 (5) 120 −5 −3 −2

power of illumination xenon lamp at 480 nm 25 μW at 365 nm 650 μW at 365 nm 150 W xenon lamp at 360 nm 15 mW/cm2 at 325 nm 0.5 mW/cm2 at 370 nm 1.06 mW/cm2 at 370 nm 1 mW/cm2 at 365 nm

responsivity (A/W) 0.01 0.07 0.34 0.38 24 40 (1.2) 61 101.2 40 21.5

rise time (s)

fall time (s)

ref

11 1 0.67

14 45 1.02

17 19 22 20 23 18

0.1 0.44

1 0.59

21 this work

At −2 V bias voltage, the output current increases from 7.19 (dark) to 176.06 μA at a power density of 1 mW cm−2. We observe rectifying behavior in the dark, which is a p−n junction characteristic. However, under illumination, electrons and holes generated in the depletion region of the p-Si/n-ZnO NT heterojunctions drift in opposite directions in the presence of an electric field operating under a reverse bias, leading to the collection of more photocurrent than dark current and the disappearance of the rectification characteristics of the device. Such behavior has been observed in p-n PDs.19,20,22,40,53,54 A reverse bias is thus advantageous for PDs because it increases the depletion width, while decreasing the photoinduced carrier transit time and the carrier loss.19 Under a forward bias, a photoresponse to UV light is also possible. The responsivity of our device was examined using a power density of 1 mW cm−2 and 365 nm UV source under ±2 V bias. Figure 7b shows the device photoresponse obtained by

To fabricate the PD device, we deposited ITO as an ncontact on NTs and Ag as a p-contact on p-Si. Figure 6a presents a schematic diagram of the vertically injected p-Si/nZnO-NT-based PD. Relevant energy levels of the materials used in this device structure are shown in Figure 6b. The valence band (VB) and conduction band (CB) of Si are −5.17 and −4.05 eV, respectively, while the VB and CB of ZnO are −7.71 and −4.35 eV, respectively.15,52 Upon 365 nm UV illumination, photoinduced carriers are generated in ZnO NTs. The energy band alignment of the device ensures that electrons are injected from the photosensitive ZnO NTs and are collected by the ITO electrode, while the VB holes reach the Ag electrode through the p-Si substrate. We measured the I−V characteristics of our p-Si/n-ZnO-NTbased PD under 365 nm illumination at different power densities (1−100 mW cm−2), as shown in Figure 7a, revealing the high UV photoresponse at the reverse bias voltage of 2 V. 37124

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces

and the incident power received over the device effective area, given by the relation R = IPh/P.64 It can be clearly seen that, at a fixed bias voltage, the responsivity decreases with the increasing illumination power. In addition, for a fixed illumination power, the detector exhibits a higher responsivity at a greater bias voltage. We demonstrate that the photodetecting performance was significantly increased in ZnO-NT-based PDs. We ascribe this superior performance compared to those obtained in previous studies to several factors. First, the superior crystal quality of ZnO NTs with low defect densities, as shown by PL, leads to a remarkable reduction in the density of traps induced by defects in the band gap and allows sufficient photoinduced carrier transport. Second, the large surface-to-volume ratio of ZnO NTs (as the hollowing emerges from the ZnO−Si interface) boosts the responsivity due to the surface states. This, in turn, enhances the density of trapped carriers (mainly holes), leading to an increase in the carrier lifetime accompanied by a reduction in the carrier transient time due to the low dimensionality of the NTs compared to NWs.4 As most photons are absorbed at the surface before arriving at the internal surface of the NT,40,67 the above band gap excitation (using UV light) leads to induced surface states. In addition, the oxygen adsorption/desorption process may occur in the surface state, leading to more generated carriers.2,4,68 Third, the nanodimensions of ZnO NTs allow 1D-like confinement, which is essential for enhancing absorption during illumination and reducing the temperature dependence of the devices, while also diminishing carrier scattering, thereby enhancing carrier transport. Fourth, as the d/r ratio is smaller than (3/8)1/2, it allows internal reflection34,35 light to be trapped inside the active layer, resulting in a reduction in the transit time of the photoinduced carriers and consequently an increase in the responsivity.36 Therefore, the high-quality structure, with a large surface-to-volume ratio, leads to a strong carrier transport process, which results in the superior responsivity of the NTs. In sum, the high responsivity and stability of the ZnO NT structures suggest that they are impressive candidates for use in UV PD devices compared to nanowires.

switching the light source on and off six times in order to test its repeatability in the UV region under a forward and reverse bias of 2 V. As shown in Figure 7b, the photocurrent was consistent and repeatable under both reverse (red curve) and forward (black curve) bias. The curve in Figure 7c shows the photoresponse time pertaining to a reverse bias of 2 V when the PD was operated under UV illumination (with a power density of 1 mW cm−2). Here, to determine the changes in the photoresponse time during the rise and decay, we fitted the data using various functions, whereby the best fit was achieved with a double-exponential function4,55 (Figure S4 in the Supporting Information). This behavior is prevalent and characterizes many materials, including ZnO NR/nanowire (NW) based UV photodetectors4,55,56 due to the presence of the defect traps, which delays charge carrier evacuation.57,58 On the basis of the fitting functions and parameters (blue curve), the time constants for the photocurrent rise τr1 and τr2 and decay τd1 and τd2 are estimated at 0.44, 3.19, 0.59, and 7.18 s, respectively. This indicates a rapid rise, followed by a slow rise and a rapid decay, followed by a slow decay. We derived the weighted-average rise and fall times of our device at 0.44 and 0.59 s, respectively. A typical photoresponse and recovery time is observed compared to high-performance PDs based on ZnO reported in previous work, as shown in Table 1. Constant photocurrent over 3600 s demonstrates the stability of the operational device after a storage time of 5 months (Figure S5 in the Supporting Information). Visible blindness characteristics are crucial in UV PDs. Therefore, we demonstrate that the I−V measurements of our PD under white light show a very weak response (Figure S6 in the Supporting Information), confirming the visible blindness characteristics of our PD.20,59,60 R and D* are the most critical parameters determining PD performance. Previous reports indicated that R and specific D* are calculated on the basis of the assumption that noise in the current is primarily dominated by shot noise.29,61,62 The photocurrent response of our PD leads to very high R (in the A W−1 range), D*, and EQE. Specifically, under 365 nm UV illumination of 1 mW cm−2 and −2 V, R = 21.51 A W−1 is obtained in the UV region, corresponding to D* = 1.26 × 1012 cm Hz1/2 W−1 and EQE = 73.1 × 102% with an effective area of ∼0.79 mm2. When, the voltage bias increases to −5 V, the responsivity increases to 101.2 A W−1, as shown in Figure 7d. Thus, we achieved an R value that was greater than that of commercial UV Si-based PDs (mostly in the 0.1−0.2 A W−1 range, with effective areas of 0.8 mm2).1,63 Table 1 provides a comparison of the responsivity, bias voltage, power density of illumination, and photoresponse time of ZnO-NW-based photoconductors reported in extant publications with those obtained in our work, indicating the superiority of the responsivity of our ZnO/p-Si NTs. To confirm that this high responsivity is attributed to the NTs and not the nanostructures between the NTs, the I−V measurement of a ZnO layer without NTs grown on p-Si was carried out (Figure S7, Supporting Information). No p−n characteristic is observed, and very low responsivity (0.48 A W−1) is obtained, indicating that the high responsivity of our PD is due to the NTs. Figure 7e shows that R decreased as the UV intensity increased, which is consistent with the observations made in previous studies.16,24,64,65 At relatively high intensities, the decrease in R is attributed to a manifestation of hole-trap saturation4 and the shrinking of the depletion region.66 Therefore, R is defined as the ratio of the photocurrent (IPh)



CONCLUSIONS We optimized our 1D p-Si/ZnO-NT-based UV PD to obtain high UV responsivity (fabricated by direct growth without a need for costly and complex fabrication processing). SEM and TEM confirmed the high quality of the vertical n-ZnO NTs (with well-defined hollowing) grown on p-Si. TRPL revealed that the radiative carrier lifetime is due to 1D-like carrier confinement. We fabricated p-Si/n-ZnO NT heterojunctions using a process that can potentially serve as a viable approach for enhancing the performance of UV PD devices due to high photoinduced carrier density, the high surface-to-volume ratio of the ZnO NTs, and the extremely low density of defect trapping states. These characteristics result in superior responsivity in the hundreds of amperes per watt range (101.2 A W−1), as well as high external efficiency. Using this process, PDs based on both Si and ZnO can be produced at low cost for biological, environmental, and industrial applications. In addition, the junction devices based on Si are compatible with the complementary metal-oxide−semiconductor (CMOS) technology. 37125

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces



(11) Gomez, J. L.; Tigli, O. Zinc Oxide Nanostructures: From Growth to Application. J. Mater. Sci. 2013, 48, 612−624. (12) Yu, J.; Tian, N. High Spectrum Selectivity and Enhanced Responsivity of a ZnO Ultraviolet Photodetector Realized by the Addition of ZnO Nanoparticles Layer. Phys. Chem. Chem. Phys. 2016, 18, 24129−24133. (13) Zhao, Q.; Wang, W. J.; Shao, J. Y.; Li, X. M.; Tian, H. M.; Liu, L.; Mei, X. S.; Ding, Y. C.; Lu, B. H. Nanoscale Electrodes for Flexible Electronics by Swelling Controlled Cracking. Adv. Mater. 2016, 28, 6337−6344. (14) Hu, L. F.; Yan, J.; Liao, M. Y.; Xiang, H. J.; Gong, X. G.; Zhang, L. D.; Fang, X. S. An Optimized Ultraviolet-A Light Photodetector with Wide-Range Photoresponse Based on ZnS/ZnO Biaxial Nanobelt. Adv. Mater. 2012, 24, 2305−2309. (15) Wang, Z. N.; Yu, R. M.; Wang, X. F.; Wu, W. Z.; Wang, Z. L. Ultrafast Response p-Si/n-ZnO Heterojunction Ultraviolet Detector Based on Pyro-Phototronic Effect. Adv. Mater. 2016, 28, 6880−6886. (16) Hatch, S. M.; Briscoe, J.; Dunn, S. A Self-Powered ZnONanorod/CuSCN UV Photodetector Exhibiting Rapid Response. Adv. Mater. 2013, 25, 867−871. (17) Sun, K.; Jing, Y.; Park, N.; Li, C.; Bando, Y.; Wang, D. L. Solution Synthesis of Large-Scale, High-Sensitivity ZnO/Si Hierarchical Nanoheterostructure Photodetectors. J. Am. Chem. Soc. 2010, 132, 15465−15467. (18) Guo, L.; Zhang, H.; Zhao, D. X.; Li, B. H.; Zhang, Z. Z.; Jiang, M. M.; Shen, D. Z. High Responsivity ZnO Nanowires Based UV Detector Fabricated by the Dielectrophoresis Method. Sens. Actuators, B 2012, 166-167, 12−16. (19) Luo, L.; Zhang, Y. F.; Mao, S. S.; Lin, L. W. Fabrication and Characterization of ZnO Nanowires Based UV Photodiodes. Sens. Actuators, A 2006, 127, 201−206. (20) Al-Hardan, N. H.; Jalar, A.; Abdul Hamid, M. A. A.; Keng, L. K.; Ahmed, N. M.; Shamsudin, R. A Wide-Band UV Photodiode Based on n-ZnO/p-Si Heterojunctions. Sens. Actuators, A 2014, 207, 61−66. (21) Jin, Y. Z.; Wang, J. P.; Sun, B. Q.; Blakesley, J. C.; Greenham, N. C. Solution-Processed Ultraviolet Photodetedtors Based on Colloidal ZnO Nanoparticles. Nano Lett. 2008, 8, 1649−1653. (22) Singh, S.; Hazra, P.; Tripathi, S.; Chakrabarti, P. Performance Analysis of RF-Sputtered ZnO/Si Heterojunction UV Photodetectors with High Photo-Responsivity. Superlattices Microstruct. 2016, 91, 62− 69. (23) Liu, C.; Zhang, B. P.; Lu, Z. W.; Binh, N. T.; Wakatsuki, K.; Segawa, Y.; Mu, R. Fabrication and Characterization of ZnO Film Based UV Photodetector. J. Mater. Sci.: Mater. Electron. 2009, 20, 197− 201. (24) Ji, L. W.; Peng, S. M.; Su, Y. K.; Young, S. J.; Wu, C. Z.; Cheng, W. B. Ultraviolet Photodetectors Based on Selectively Grown ZnO Nanorod Arrays. Appl. Phys. Lett. 2009, 94, 203106. (25) Kathalingam, A.; Kim, H. S.; Park, H. M.; Valanarasu, S.; Mahalingam, T. Effect of Indium on Photovoltaic Property of n-ZnO/ p-Si Heterojunction Device Prepared using Solution-Synthesized ZnO Nanowire Film. J. Photonics Energy 2015, 5, 053085. (26) Hu, L. F.; Yan, J.; Liao, M. Y.; Wu, L. M.; Fang, X. S. Ultrahigh External Quantum Efficiency from Thin SnO2 Nanowire Ultraviolet Photodetectors. Small 2011, 7, 1012−1017. (27) Xie, Y.; Madel, M.; Li, Y. J.; Jie, W. Q.; Neuschl, B.; Feneberg, M.; Thonke, K. Polarity-Controlled Ultraviolet/Visible Light ZnO Nanorods/p-Si Photodetector. J. Appl. Phys. 2012, 112, 123111. (28) Chen, Y. W.; Liu, Y. C.; Lu, S. X.; Xu, C. S.; Shao, C. L. Photoelectric Properties of ZnO: In Nanorods/SiO2/Si Heterostructure Assembled in Aqueous Solution. Appl. Phys. B: Lasers Opt. 2006, 84, 507−510. (29) Saidaminov, M. I.; Haque, M. A.; Savoie, M.; Abdelhady, A. L.; Cho, N.; Dursun, I.; Buttner, U.; Alarousu, E.; Wu, T.; Bakr, O. M. Perovskite Photodetectors Operating in Both Narrowband and Broadband Regimes. Adv. Mater. 2016, 28, 8144−8149. (30) Gopalakrishnan, N.; Shin, B. C.; Lim, H. S.; Kim, G. Y.; Yu, Y. S. Comparison of ZnO: GaN Films on Si(111) and S(100) Substrates by Pulsed Laser Deposition. Phys. B 2006, 376-377, 756−759.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b09645. The microscope image of ITO electrode, a tilted SEM image of ZnO NTs, the TRPL spectra for different temperatures, the method of the double-exponential function fitting of the response time, photocurrent stability of the device over time, I−V characteristics of our p-Si/n-ZnO PD under visible light conditions, and SEM images of the wetting layer without NTs and its I− V characteristics (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Tom Wu: 0000-0003-0845-4827 Iman S. Roqan: 0000-0001-7442-4330 Author Contributions

All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Tahani Flemban is grateful for a scholarship from Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia. The authors wish to thank Ms. Ecaterina Ware from Imperial College London, UK, for TEM sample preparation and Dr. M. A. Roldan from KAUST core lab for HR-TEM measurements. The authors thank KAUST for the financial support.



REFERENCES

(1) Monroy, E.; Omnes, F.; Calle, F. Wide-Bandgap Semiconductor Ultraviolet Photodetectors. Semicond. Sci. Technol. 2003, 18, R33−R51. (2) Sang, L. W.; Liao, M. Y.; Sumiya, M. A Comprehensive Review of Semiconductor Ultraviolet Photodetectors: From Thin Film to OneDimensional Nanostructures. Sensors 2013, 13, 10482−10518. (3) Wang, J. A.; Lee, S. Ge-Photodetectors for Si-Based Optoelectronic Integration. Sensors 2011, 11, 696−718. (4) Soci, C.; Zhang, A.; Xiang, B.; Dayeh, S. A.; Aplin, D. P. R.; Park, J.; Bao, X. Y.; Lo, Y. H.; Wang, D. ZnO Nanowire UV Photodetectors with High Internal Gain. Nano Lett. 2007, 7, 1003−1009. (5) Wang, Z. N.; Yu, R. M.; Wen, X. N.; Liu, Y.; Pan, C. F.; Wu, W. Z.; Wang, Z. L. Optimizing Performance of Silicon-Based p-n Junction Photodetectors by the Piezo-Phototronic Effect. ACS Nano 2014, 8, 12866−12873. (6) Yang, Q.; Guo, X.; Wang, W. H.; Zhang, Y.; Xu, S.; Lien, D. H.; Wang, Z. L. Enhancing Sensitivity of a Single ZnO Micro-/Nanowire Photodetector by Piezo-phototronic Effect. ACS Nano 2010, 4, 6285− 6291. (7) Ozgur, U.; Alivov, Y. I.; Liu, C.; Teke, A.; Reshchikov, M. A.; Dogan, S.; Avrutin, V.; Cho, S. J.; Morkoc, H. A Comprehensive Review of ZnO Materials and Devices. J. Appl. Phys. 2005, 98, 041301. (8) Nagaraju, G.; Ko, Y. H.; Yu, J. S. Effect of Diameter and Height of Electrochemically-Deposited ZnO Nanorod Arrays on the Performance of Piezoelectric Nanogenerators. Mater. Chem. Phys. 2015, 149150, 393−399. (9) Wang, Z. L.; Song, J. H. Piezoelectric Nanogenerators Based on Zinc Oxide Nanowire Arrays. Science 2006, 312, 242−246. (10) Wang, Z. L. Progress in Piezotronics and Piezo-Phototronics. Adv. Mater. 2012, 24, 4632−4646. 37126

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127

Research Article

ACS Applied Materials & Interfaces (31) Cao, G. Nanostructures & Nanomaterials; Imperial College Press: London, 2004. (32) Flemban, T. H.; Singaravelu, V.; Devi, A. A. S.; Roqan, I. S. Homogeneous Vertical ZnO Nanorod Arrays with High Conductivity on an In Situ Gd Nanolayer. RSC Adv. 2015, 5, 94670−94678. (33) Yan, M.; Zhang, H. T.; Widjaja, E. J.; Chang, R. P. H. SelfAssembly of Well-Aligned Gallium-Doped Zinc Oxide Nanorods. J. Appl. Phys. 2003, 94, 5240−5246. (34) Zhan, J. X.; Dong, H. X.; Sun, S. L.; Ren, X. D.; Liu, J. J.; Chen, Z. H.; Lienau, C.; Zhang, L. Surface-Energy-Driven Growth of ZnO Hexagonal Microtube Optical Resonators. Adv. Opt. Mater. 2016, 4, 126−134. (35) Nobis, T.; Kaidashev, E. M.; Rahm, A.; Lorenz, M.; Grundmann, M. Whispering Gallery Modes in Nanosized Dielectric Resonators with Hexagonal cross Section. Phys. Rev. Lett. 2004, 93, 103903. (36) Chen, E. L.; Chou, S. Y. High-Efficiency and High-Speed Silicon Metal-Semiconductor-Metal Photodetectors Operating in the Infrared. Appl. Phys. Lett. 1997, 70, 753−755. (37) Israr, M. Q.; Sadaf, J. R.; Yang, L. L.; Nur, O.; Willander, M.; Palisaitis, J.; Persson, P. O. A. Trimming of Aqueous Chemically Grown ZnO Nanorods into ZnO Nanotubes and their Comparative Optical Properties. Appl. Phys. Lett. 2009, 95, 073114. (38) B D Cullity, S. R. S. Elements of X-ray Diffraction, 3rd ed.; Prentice-Hall: New York, 2001. (39) Feng, Z. C. Handbook of Zinc Oxide and Related Materials; CRC Press: Boca Raton, FL, 2013. (40) Bie, Y. Q.; Liao, Z. M.; Zhang, H. Z.; Li, G. R.; Ye, Y.; Zhou, Y. B.; Xu, J.; Qin, Z. X.; Dai, L.; Yu, D. P. Self-Powered, Ultrafast, VisibleBlind UV Detection and Optical Logical Operation based on ZnO/ GaN Nanoscale p-n Junctions. Adv. Mater. 2011, 23, 649−653. (41) Jie, J. S.; Zhang, W. J.; Jiang, Y.; Meng, X. M.; Li, Y. Q.; Lee, S. T. Photoconductive Characteristics of Single-Crystal CdS Nanoribbons. Nano Lett. 2006, 6, 1887−1892. (42) Varshni, Y. P. Temperature Dependence of the Energy Gap in Semiconductors. Physica 1967, 34, 149−154. (43) Beaur, L.; Bretagnon, T.; Gil, B.; Kavokin, A.; Guillet, T.; Brimont, C.; Tainoff, D.; Teisseire, M.; Chauveau, J. M. Exciton Radiative Properties in Nonpolar Homoepitaxial ZnO/(Zn,Mg)O Quantum Wells. Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 84, 165312. (44) Jung, S. W.; Park, W. I.; Cheong, H. D.; Yi, G. C.; Jang, H. M.; Hong, S.; Joo, T. Time-Resolved and Time-Integrated Photoluminescence in ZnO Epilayers Grown on Al2O3(0001) by Metalorganic Vapor Phase Epitaxy. Appl. Phys. Lett. 2002, 80, 1924−1926. (45) Kwok, W. M.; Djurišić, A. B.; Leung, Y. H.; Chan, W. K.; Phillips, D. L. Time-Resolved Photoluminescence from ZnO Nanostructures. Appl. Phys. Lett. 2005, 87, 223111. (46) Zhao, Q. X.; Yang, L. L.; Willander, M.; Sernelius, B. E.; Holtz, P. O. Surface Recombination in ZnO Nanorods Grown by Chemical Bath Deposition. J. Appl. Phys. 2008, 104, 073526. (47) Hauswald, C.; Flissikowski, T.; Gotschke, T.; Calarco, R.; Geelhaar, L.; Grahn, H. T.; Brandt, O. Coupling of Exciton States as the Origin of their Biexponential Decay Dynamics in GaN Nanowires. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88, 075312. (48) Gurioli, M.; Vinattieri, A.; Colocci, M.; Deparis, C.; Massies, J.; Neu, G.; Bosacchi, A.; Franchi, S. Temperature dependence of the Radiative and Nonradiative Recombination Time in GaAs/AlxGa1‑xAs Quantum-Well Structures. Phys. Rev. B: Condens. Matter Mater. Phys. 1991, 44, 3115−3124. (49) Akiyama, H.; Koshiba, S.; Someya, T.; Wada, K.; Noge, H.; Nakamura, Y.; Inoshita, T.; Shimizu, A.; Sakaki, H. Thermalization Effect on Radiative Decay of Excitons in Quantum Wires. Phys. Rev. Lett. 1994, 72, 924−927. (50) Lee, W.; Kiba, T.; Murayama, A.; Sartel, C.; Sallet, V.; Kim, I.; Taylor, R. A.; Jho, Y. D.; Kyhm, K. Temperature Dependence of the Radiative Recombination Time in ZnO Nanorods under an External Magnetic Field of 6T. Opt. Express 2014, 22, 17959−17967. (51) Morhain, C.; Bretagnon, T.; Lefebvre, P.; Tang, X.; Valvin, P.; Guillet, T.; Gil, B.; Taliercio, T.; Teisseire-Doninelli, M.; Vinter, B.;

Deparis, C. Internal Electric Field in Wurtzite ZnO/Zn0.78Mg0.22O Quantum Wells. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 72, 241305R. (52) Gayen, R. N.; Bhattacharyya, S. R. Electrical Characteristics and Rectification Performance of Wet Chemically Synthesized Vertically Aligned n-ZnO Nanowire/p-Si Heterojunction. J. Phys. D: Appl. Phys. 2016, 49, 115102. (53) Yin, B.; Qiu, Y.; Zhang, H.; Luo, Y.; Zhao, Y.; Yang, D.; Hu, L. Improved Photoresponse Performance of A Self-Powered Si/ZnO Heterojunction Ultraviolet and Visible Photodetector by the Piezophototronic Effect. Semicond. Sci. Technol. 2017, 32, 064002. (54) Hazra, P.; Singh, S.; Jit, S. Ultraviolet Photodetection Properties of ZnO/Si Heterojunction Diodes Fabricated by ALD Technique without using A Buffer Layer. J. Semicond. Sci. Technol. 2014, 14, 117. (55) Wang, Z. N.; Yu, R. M.; Pan, C. F.; Liu, Y.; Ding, Y.; Wang, Z. L. Piezo-Phototronic UV/Visible Photosensing with Optical-Fiber-Nanowire Hybridized Structures. Adv. Mater. 2015, 27, 1553−1560. (56) Ni, P. N.; Shan, C. X.; Wang, S. P.; Liu, X. Y.; Shen, D. Z. SelfPowered Spectrum-Selective Photodetectors Fabricated from n-ZnO/ p-NiO Core-Shell Nanowire Arrays. J. Mater. Chem. C 2013, 1, 4445− 4449. (57) Ahn, S. E.; Lee, J. S.; Kim, H.; Kim, S.; Kang, B. H.; Kim, K. H.; Kim, G. T. Photoresponse of Sol-gel-synthesized ZnO Nanorods. Appl. Phys. Lett. 2004, 84, 5022−5024. (58) Ghosh, R.; Basak, D. Electrical and Ultraviolet Photoresponse Properties of Quasialigned ZnO Nanowires/p-Si Heterojunction. Appl. Phys. Lett. 2007, 90, 243106. (59) Um, H. D.; Moiz, S. A.; Park, K. T.; Jung, J. Y.; Jee, S. W.; Ahn, C. H.; Kim, D. C.; Cho, H. K.; Kim, D. W.; Lee, J. H. Highly Selective Spectral Response with Enhanced Responsivity of n-ZnO/p-Si Radial Heterojunction Nanowire Photodiodes. Appl. Phys. Lett. 2011, 98, 033102. (60) Zhang, T. C.; Guo, Y.; Mei, Z. X.; Gu, C. Z.; Du, X. L. VisibleBlind Ultraviolet Photodetector Based on Double Heterojunction of nZnO/insulator-MgO/p-Si. Appl. Phys. Lett. 2009, 94, 113508. (61) Gong, X.; Tong, M. H.; Xia, Y. J.; Cai, W. Z.; Moon, J. S.; Cao, Y.; Yu, G.; Shieh, C. L.; Nilsson, B.; Heeger, A. J. High-Detectivity Polymer Photodetectors with Spectral Response from 300 to 1450 nm. Science 2009, 325, 1665−1667. (62) Velusamy, D. B.; Haque, Md. A.; Parida, M. R.; Zhang, F.; Wu, T.; Mohammed, O. F.; Alshareef, H. N. 2D Organic−Inorganic Hybrid Thin Films for Flexible UV−Visible Photodetectors. Adv. Funct. Mater. 2017, 27, 1605554. (63) Zou, J. P.; Zhang, Q.; Huang, K.; Marzari, N. Ultraviolet Photodetectors Based on Anodic TiO2 Nanotube Arrays. J. Phys. Chem. C 2010, 114, 10725−10729. (64) Cai, Y. H.; Tang, L. B.; Xiang, J. Z.; Ji, R. B.; Lai, S. K.; Lau, S. P.; Zhao, J.; Kong, J. C.; Zhang, K. High Performance Ultraviolet Photodetectors Based on ZnO Nanoflakes/PVK Heterojunction. Appl. Phys. Lett. 2016, 109, 073103. (65) Dang, V. Q.; Trung, T. Q.; Kim, D. I.; Duy, L. T.; Hwang, B. U.; Lee, D. W.; Kim, B. Y.; Toan, L. D.; Lee, N. E. Ultrahigh Responsivity in Graphene-ZnO Nanorod Hybrid UV Photodetector. Small 2015, 11, 3054−3065. (66) Zhang, H.; Babichev, A. V.; Jacopin, G.; Lavenus, P.; Julien, F. H.; Egorov, A. Yu.; Zhang, J.; Pauporte, T.; Tchernycheva, M. Characterization and Modeling of a ZnO Nanowire Ultraviolet Photodetector with Graphene Transparent Contact. J. Appl. Phys. 2013, 114, 234505. (67) Sze, S. M.; Ng, K. K. Physics of Semiconductor Devices, 3rd ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2006. (68) Li, Q. H.; Wan, Q.; Liang, Y. X.; Wang, T. H. Electronic Transport through Individual ZnO nanowires. Appl. Phys. Lett. 2004, 84, 4556−4558.

37127

DOI: 10.1021/acsami.7b09645 ACS Appl. Mater. Interfaces 2017, 9, 37120−37127