Fig. 1. UV spectral region and its subdivisions.
Fig. 2. Schematic structures of different semiconductor photodetectors.
Fig. 3. (a) Energy band diagram of a metal and a semiconductor before contact; (b) ideal energy band diagram of a metal/n–semiconductor junction for ϕm>ϕs.
Fig. 4. Dependence of the Schottky barrier height on the metal work function of different metals, and the ideal Schottky barrier height based on the Schottky–Mott model (short dashed curve).
Fig. 5. Chart illustrating transformation relationships among the forms of
Ga2O3 and its hydrates. Redraw with permission from Roy
et al. [
93]. Copyright 1952 American Chemical Society.
Fig. 6. Various
Ga2O3 nanostructures. (a)
γ-Ga2O3 nanocrystals. Reprinted with permission from Wang
et al. [
196]. Copyright 2010 American Chemical Society. (b)
β-Ga2O3 nanowires. Reprinted with permission from Nogales
et al. [
229]. Copyright 2007 American Institute of Physics. (c)
Ga2O3/GaN:Ox@SnO2 shell@core NWs. Reprinted with permission from Lupan
et al. [
234]. Copyright 2015 Elsevier B.V. (d)
β-Ga2O3 nanorods. Reprinted with permission from Vanithakumari
et al. [
244]. Copyright 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. (e) As-synthesized
β-Ga2O3 nanobelt. Reprinted with permission from Zou
et al. [
256]. Copyright 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. (f) Branched multiwire nanostructures (left) and its X-ray fluorescence (XRF) map (right). Reprinted with permission from Martinez-Criado
et al. [
224]. Copyright 2014 American Chemical Society.
Fig. 7. The crystal structure of β-Ga2O3 and its (2¯01) surfaces.
Fig. 8. Band structure of
β-Ga2O3. (a) At the GGA–DFT (PBE) level and (b) at the hybrid HF–DFT (Gau–PBE) level. Reprinted with permission from Mock
et al. [
275]. Copyright 2017 American Physical Society.
Fig. 9. (αhν)2 versus
hν plots for (a) (010) Mg-doped substrate for
E||c (closed circles) and
E||a* (open circles) at RT. (b) Data set of (001) undoped substrate for
E||a (closed squares) and
E||b (open squares). Dotted lines represent the energies of the direct absorption edge. Partially polarized reflectance spectra at RT are shown in the upper part of the figures. Energies of the dips and the shoulder are indicated by the vertical arrows. Reprinted with permission from Onuma
et al. [
12]. Copyright 2015 The Japan Society of Applied Physics.
Fig. 10. (a)
1 μm×1 μm scan image of an annealed surface observed by tapping mode atomic force microscopy. (b)
I–
V characteristics of a photodetector. Closed (black) and open (red) symbols represent current in the dark condition and current in the presence of 250 nm light irradiation, respectively. (c) Photocurrent spectral response (blue) and photoresponsivity of the photodetector (black) at a reverse bias of 10 V. The dashed line indicates the photoresponsivities expected in the case without carrier multiplication. Reprinted with permission from Oshima
et al. [
304]. Copyright 2008 The Japan Society of Applied Physics. (d) Transient response of the detector. (e) Signal from the flame detection system during the demonstration. Reprinted with permission from Oshima
et al. [
305]. Copyright 2009 The Japan Society of Applied Physics.
Fig. 11. (a) Dark
I–
V characteristics of the
Au-Ga2O3 Schottky photodiode annealed at various temperatures. The inset shows the device configuration. (b) Spectral responsivities of the photodiode annealed at 400°C and the as-fabricated photodiode at a reverse bias of 3 V. The inset shows the photocurrent of the devices under reverse bias voltage to 5 V. Reprinted with permission from Suzuki
et al. [
87]. Copyright 2009 American Institute of Physics. (c) Time-dependent photoresponse of the
β-Ga2O3 thin films prototype photodetector to 254 nm illumination: (top) the Ohmic-type device; (bottom) the Schottky-type device. Reprinted with permission from Guo
et al. [
310]. Copyright 2014 American Institute of Physics. (d) Temporal response tests of the PDs with KrF pulse laser illumination at 10 V bias. Reprinted with permission from Cui
et al. [
315]. Copyright 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.
Fig. 12. (a) Photoresponsivity of metal–semiconductor–metal deep ultraviolet photodetectors. (b) Low-frequency noise power density as a function of frequency of metal–semiconductor–metal deep-ultraviolet photodetectors. Reprinted with permission from Lee
et al. [
320]. Copyright 2018 IEEE. (c) EDX line scan along the In gradient (the black dashed lines separate different crystallographic phases); (d) single XRD patterns selected from each phase. (e) Responsivity versus photon energy of 10 MSM-PDs along the In gradient. (f) Spectrally resolved
J–
V measurements of an SC [the analyzed
ϕBn,eff are shown as squares in (e)]. Reprinted with permission from Zhang
et al. [
326]. Copyright 2016 American Institute of Physics.
Fig. 13. (a) Real-time photoresponse of the detector to 254 nm light. (b) Enlarged rise and decay edges for the first “ON” and “OFF”, respectively. Reprinted with permission from Feng
et al. [
266]. Copyright 2006 American Institute of Physics. (c) Enlarged SEM images of the photo switch made of an individual
Au-in-Ga2O3 peapod nanowire. (d) Photoresponse behaviors during illumination ON and OFF cycles between electrodes 11–13. Reprinted with permission from Hsieh
et al. [
207]. Copyright 2008 American Chemical Society.
Fig. 14. (a) Time-dependent photoresponse of the bridged
β-Ga2O3 NWs grown at 925°C (sample 2, solid line) and at 800°C (sample 3, dotted line) measured in dry air under a bias voltage of 5 V and a UVC (254 nm) irradiance of
∼2 mW· cm−2. The photocurrent to dark current ratios for sample 2 and sample 3 were
∼5×104 and
∼5×106, respectively. (b) Spectral responses of sample 1 (squares), sample 2 (circles), and sample 3 (triangles). (c) Room-temperature PL spectra of sample 1 (solid line), sample 2 (dashed line), and sample 3 (dotted line), revealing an increase in defect emissions with decreasing growth temperature. (d) Time-dependent photoresponse of the bridged
β-Ga2O3 NWs (sample 1). Reprinted with permission from Li
et al. [
331]. Copyright 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. (e) Schematic illustration of the fabrication of the
β-Ga2O3 nanowire array film and its vertical Schottky photodiode. (f)
I−
V characteristics of the device in the dark and under the illumination of 254 nm light in the logarithmic scale. Inset shows the photovoltaic characteristic of the device near zero bias. (g) Time-dependent photocurrent response of an
Au/β-Ga2O3 nanowire array film Schottky photodiode measured under 254 nm light illumination with the intensity of
2 mW·cm−2 at 0 V. (h) Decay edge of the current response at a reverse bias of 10 V. Reprinted with permission from Chen
et al. [
203]. Copyright 2016 American Chemical Society.
Fig. 15. (a) Atomic force microscopy image of a 2D
β-Ga2O3 photodetector; inset, the corresponding height profile of the photodetector. Reprinted with permission from Feng
et al. [
258]. Copyright 2014 Royal Society of Chemistry. (b)
I–
V characteristics of the device in the dark. (c)
I–
V characteristics of the device under irradiation with 254 nm light at different light intensities. Reprinted with permission from Zhong
et al. [
259]. Copyright 2015 Elsevier B.V. All rights reserved. (d) The ON/OFF cycle of the
β-Ga2O3 nanobelt device upon 250 nm light illumination from the RT to 433 K under the intensity of
9.835×10−5 W/cm2 at a bias of 6.0 V. Reprinted with permission from Zou
et al. [
256]. Copyright 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. (e) Time-dependent photoresponse of the fabricated photodetector under various illumination conditions (254, 365, 532, and 650 nm light exposure). (f) Experimental and fitted curves of the current rise and decay process at 254 nm illumination under a gate bias of 0 V. Reprinted with permission from Oh
et al. [
169]. Copyright 2016 Royal Society of Chemistry.
Fig. 16. (a) Schematic diagram of the APD device. (b)
I–
V characteristics of the photodetector under dark and illumination conditions with 254 nm light of
1.67 mW/cm2. (c)
I–
V characteristics at 300, 330, and 370 K in the reverse voltage under the dark condition; the inset shows the dependence of the avalanche breakdown voltage on the recording temperature. (d) Spectral response of the device at
−6 V bias. (e) Transient response of the device at
−6 V bias and a second-order exponential fit of the data. (f) Energy band diagram of the APD device in reverse bias. Reprinted with permission from Zhao
et al. [
240]. Copyright 2015 American Chemical Society.
Fig. 17. (a) Dependence of current–voltage characteristics on intensity of UV light illumination of a deuterium lamp. (b) Top, reverse-current response of a photodiode to deep-UV light pulses (reverse-bias voltage is 2 V); bottom, light waveform measured by a silicon pin photodiode for the same light pulses. Reprinted with permission from Nakagomi
et al. [
342]. Copyright 2013 American Institute of Physics. (c) Spectral response of photodiodes based on a
β-Ga2O3/GaN heterojunction with 116 nm and 175 nm thick
β-Ga2O3 layers and based on a
β-Ga2O3/SiC heterojunction with a 116 nm
β-Ga2O3 layer. (d) Top, response of a photodiode to deep-UV light pulses with a reverse-bias voltage of 2 V; bottom, light waveform measured with a silicon APD module for the same light pulses. Reprinted with permission from Nakagomi
et al. [
340]. Copyright 2015 Elsevier B.V. All rights reserved.
Fig. 18. (a)
I–
V characteristics of the
β-Ga2O3/SnO2 bilayer under the dark and illumination conditions (inset shows the construction of the developed device). (b) Dark current versus voltage at various temperatures. (c) The spectral response of the device at
−5.5 V bias. (d) The transient response of the device at
−5.5 V bias under pulsed 254 nm light illumination and a second-order exponential fit of the data. Reprinted with permission from Mahmoud [
352]. Copyright 2016 Elsevier B.V. All rights reserved. (e) Spectral photoresponse of a Schottky diode under different biases and the transmittance spectra of the
Ga2O3 epilayer and ZnO substrate. (f) The normalized transient photoresponse characteristics measured under reverse biases of
−34.8 V (calculated) at room temperature. (g) and (h) Schematic energy diagrams at high reverse bias under 254 and 365 nm illumination, respectively. Reprinted with permission from Chen
et al. [
181]. Copyright 2017 American Chemical Society.
Metal Stack | Doping () | Treatments | (Ω · )/Method | Reference | Ti/Au (50/300 nm) | | Si implantation in contact region 950°C annealing for implant activation 450 °C RTA for contact | | Sasaki et al. [48] | Ti/Au | | Si implantation in contact region 925°C annealing for implant activation 470°C RTA for contact | | Higashiwaki et al. [51] | Ti/Au (20/230 nm) | | Si implantation in contact region 950°C annealing for implant activation ICP etching prior 470°C RTA for contact | | Wong et al. [52] | Ti/Au | | diffusion of Sn from the SOG layer ICP etching prior 450°C RTA for contact | | Zeng et al. [53] | Ti/Au (30/130 nm) | | degenerately doped contact layer with a high Si doping concentration above 470°C RTA for contact | | Zhang et al. [54] | Ti/Au/Ni | | ICP/RIE etching prior 470°C RTA for contact | | Krishnamoorthy et al. [55] | Ti/Al/Au (15/60/50 nm) | | Ar plasma treatment | | Zhou et al. [63] | Ti/Al/Ni/Au | | ICP etching prior 470°C RTA for contact | | Chabak et al. [56] | Ti/Al/Ni/Au (20/100/50/50 nm) | | ICP/RIE etching prior 470°C RTA for contact | | Green et al. [57] | Ti/Al/Ni/Au | | ICP etching prior | – | Moser et al. [58] | AZO/Ti/Au (10/20/80 nm) | [48] | Si implantation in contact region 950°C annealing for implant activation 400°C RTA for contact | | Carey et al. [59] | ITO/Ti/Au (10/20/80 nm) | [48] | Si implantation in contact region 950°C annealing for implant activation 600°C RTA for contact | | Carey et al. [61] | ITO/Pt (140/100 nm) | | 800–1200°C RTA for contact | not measured but Ohmic for annealing above 900°C | Zhou et al. [62] | Ti, In, Ag, Sn, W, Mo, Sc, Zn, and Zr (20 nm), with Au (100 nm) overlayers | | 400–800°C RTA for contact | not measuredIn and Ti produced linear I–V curves after anneals | Yao et al. [60] |
|
Table 1. Summary of Ohmic Contact Properties on β-Ga2O3
Metal | Barrier Height (eV) | Ideality Factor | Doping () | Process/Measurement | Comments | Reference | Ni | 1.25 | 1.01 | | evaporation Ni/Au, I–V and C–V | built-in voltage is 1.18 V from C–V and 1.0 V from I–V | Oishi et al. [68] | Ni | 1.05 | not measured | | DC–sputtering, I–V and C–V | higher barrier measured with C–V | Armstrong et al. [69] | Ni | 1.08–1.12 | 1.05–1.10 | – | evaporation Ni/Au, I–V | sample etched with at 140°C | Kasu et al. [74] | Ni | 0.95 | 3.38 | UID | e-beam deposition Ni/Au, I–V | barrier height increase, and ideality factor decrease with temperature | Oh et al. [75] | Ni | 0.8–1.0 | 1.8–3.2 | UID | evaporation, I–V | with up to 0.164 | Ahmadi et al. [71] | Ni | 1.07 | 1.3 | | e-beam deposition Ni/Au, I–V | barrier height increase, and ideality factor decrease with temperature | Ahn et al. [76] | Ni | 1.54 | 1.04 | | e-beam deposition, I–V | similar value to that from C–V | Farzana et al. [77] | Ni | 1.10 | 1.05 | | evaporation Ni/Au, I–V and C–V | sample etched with at 140°C | Kasu et al. [78] | Ni | 1.2 | 1.00 | | e-beam deposition Ni/Au, I–V | dry etch damage to Schottky contact | Yang et al. [79] | Ni | 0.99–1.02 | 1.05–1.09 | | evaporation Ni/Au, I–V | (001) substrate | Oshima et al. [80] | Ni | | | – UID – | e-beam evaporation, I–V | cleaned with HCl and , vertical SBD | Yao et al. [72] | Ni | 0.81 | 2.29 | UID | I–V | with x to about 0.08 | Qian et al. [81] | Pt | 1.35–1.52 | 1.04–1.06 | – | evaporation Pt/Ti/Au, I–V and C–V | sample etched with 85 wt.% at 135°C, higher barrier measured with C–V | Sasaki et al. [70] | Pt | 1.15 | | | evaporation Pt/Ti/Au, I–V and C–V | was calculated to be | Higashiwaki et al. [82] | Pt | 1.04 | 1.28 | | e-beam deposition Pt/Au, I–V | barrier height increase, and ideality factor decrease with temperature | Ahn et al. [76] | Pt | 1.58 | 1.03 | | e-beam deposition, I–V | similar value to that from C–V | Farzana et al. [77] | Pt | 1.39 | 1.1 | | sputtering Pt/Ti/Au, I–V | barrier height stable up to at least 150°C | He et al. [83] | Pt | 1.46 | | | evaporation Pt/Ti/Au, I–V and C–V | barrier height may be increasing by the presence of F | Konishi et al. [84] | Pt | 1.01 | 1.07 | | e-beam evaporation Pt/Au, I–V | comparison to TiN | Tadjer et al. [73] | Pt | | | – UID | e-beam evaporation, I–V | bulk and epilayer SBD, cleaned with HCl and | Yao et al. [72] | Pt | 1.05–1.20 | 1.34–1.55 | | e-beam evaporation Pt/Au, I–V | both (010) and single-crystal substrates were used; the (010) SBD had a larger | Fu et al. [64] | Au | 1.07 | 1.02 | – | e-beam deposition, I–V | affinity of , work function of Au | Mohamed et al. [33] | Au | 1.71 | 1.09 | | e-beam deposition, I–V | interface consistent with inhomogeneous barrier | Farzana et al. [77] | Au | | 1.08 | – | e-beam deposition, I–V | barrier height decrease, and ideality factor decrease to near unity after annealing at temperature above 200°C | Suzuki et al. [85] | Pd | 1.27 | 1.05 | | e-beam deposition, I–V | similar value to that from C–V | Farzana et al. [77] | Cu | 1.32 | 1.03 | | DC sputtering, I–V | low-mobility layer grown by PLD | Splith et al. [86] | Cu | 0.98, 1.07 | 1.05, 1.1 | | Cu/Au/Ni, I–V | SBD, MOSSBD | Sasaki et al. [87] | Cu | | | – | e-beam evaporation, I–V | bulk vertical SBD, cleaned with HCl and | Yao et al. [72] | W | | | – UID | e-beam evaporation, I–V | bulk and epilayer SBD, cleaned with HCl and | Yao et al. [72] | Ir | | | – UID | e-beam evaporation, I–V | bulk and epilayer SBD, cleaned with HCl and | Yao et al. [72] | TiN | 0.98 | 1.09 | | ALD 350°C, I–V | similar to Pt used as comparison | Tadjer et al. [73] | | 1.94 | 1.09 | | magnetron sputtering , I–V | bulk crystal | Müller et al. [34] | | 1.42 | 1.28 | | PLD sample |
|
Table 2. Summary of Reported Schottky Barrier Contacts to β-Ga2O3
Polymorph | Bandgap | Structure and Space Group | Lattice Parameters | Comment | Reference | | 5.3 eV [91](5.25 eV [113]) | rhombohedral | | experimental | Marezio et al. [97] | | calculated | Yoshioka et al. [98] | | 4.4–5.0 eV [12] | monoclinic | | experimental | Geller et al. [94] | | calculated | He et al. [95] | | 4.4 (indirect) [102]5.0 (direct) [102] | cubic | | experimental | Otero et al. [101]Oshima et al. [102] | | – | cubic | | experimentalcalculated | Roy et al. [93]Yoshioka et al. [98] | | 4.9 (direct) [106]4.5 (indirect) [110]5.0 (direct) [110] | hexagonal | | experimental | Playford et al. [96] | | experimental | Mezzadri et al. [104] | – | orthorhombic | | calculated | Yoshioka et al. [98]Kracht et al. [107] | | experimental | Cora et al. [108] |
|
Table 3. Summary of the Basic Parameters of Ga2O3 Polymorphs
Material | Structure | Responsivity (A/W) | Decay Time (μs) | Rejection Ratio | Reference | | | bulk | vertical–Schottky | | | | Oshima et al. [305] | bulk | vertical–Schottky | | – | | Suzuki et al. [85] | | film-based MSM | | 19.1 | 80.7 | – | Cui et al. [315] | | film-based MSM | | | | | Yuan et al. [321] | | film-based MSM | – | – | | Zhang et al. [326] | individual–NWs | NW-based MSM | – | | – | Feng et al. [266] | NWs | NW-based MSM | – | | | | Li et al. [331] | NWs film | NW-based MSM | | 64 | | Chen et al. [203] | 2D nanosheet | MSM | 3.3@254 nm | | – | Feng et al. [258] | nanosheet | MSM | 19.31@254 nm | | – | Zhong et al. [259] | nanoflake | MSM | | | | – | Oh et al. [169] | | core–shell NW heterojunction APD | | 42 | 815 | | Zhao et al. [240] | | heterojunction | 0.053@260 nm | 30 | 30 | | Nakagomi et al. [344] | | heterojunction | | 300 | | Nakagomi et al. [340] | | heterojunction APD | | 48 | 100 | | Mahmoud [352] | | heterojunction APD | | 238 | 3040 | | Chen et al. [181] |
|
Table 4. Summary of the Reported Basic Parameters of Representative Ga2O3 Photodetectors