Self-powered solar blind ultraviolet photodetector based on amorphous (In_0.23Ga_0.77)₂O₃/bixbyite (In_0.67Ga_0.33)₂O₃ heterojunction

Solar-blind ultraviolet (200–280 nm) photodetectors have attracted significant attention because of their potential applications in many important fields and activities such as missile tracking, flame detection, nonline-of-sight optical communication, biomedicine, ultraviolet radiation monitoring below the ozone hole, and so on.^1–3 Particularly, they attracted attention for their high precision and accuracy of weak signal detection ability in sunlight due to the absence of solar-blind region irradiation at the Earth’s surface.^4,5 Commercial solar-blind ultraviolet photodetectors are mainly based on the photoelectric effect (photomultiplier tubes), which are limited due to their bulky structure and fragility, along with the requirement of a large bias voltage. Current research studies in solar-blind ultraviolet photodetectors focus mostly on wide bandgap semiconductors such as AlGaN, ZnMgO, diamond, and Ga₂O₃.^6–9 However, there are obvious disadvantages in these structures; for example, the epitaxial quality of the AlGaN film deteriorates dramatically with increasing Al concentration, the single wurtzite phase of the MgZnO film breaks down with increasing Mg content, and the bandgap of diamond cannot be tuned. Ga₂O₃ is considered as an ideal candidate for its wide bandgap (4.5–5.1 eV), which lies sharply in the solar-blind ultraviolet region and exhibits a flexible tunability in the bandgap by alloying with different materials.¹⁰ Furthermore, self-powered and zero power consumption type photodetectors are strongly desirable within the background of a green, resource-saving society. Certainly, homostructured p-n junctions are the first choice. However, due to the self-compensation effect of wide bandgap semiconductors and the high ionization energy of acceptor dopants, the preparation of high-quality p-type wide bandgap semiconductors still remains a challenge. As an alternative, heterostructured photodetectors have gradually entered the public domain. Heterostructured photodetectors with different structures and materials have been widely studied. Currently, Ga₂O₃ and various materials are being used to prepare heterostructured self-powered photodetectors, such as GaN, ZnO, SiC, MoS₂, Si, etc., which are widely reported in the field of solar-blind ultraviolet detection and visible-blind ultraviolet detection.^11–20 Indium doping can reduce the bandgap of Ga₂O₃ and expand its absorption range, making Ga₂O₃ more suitable for solar-blind ultraviolet detection. However, only a few works have reported about an (In_xGa_1−x)₂O₃ heterojunction photodetector and its advantages, such as that the (In_xGa_1−x)₂O₃/Si heterostructure exhibits self-powered photodetection ability in the solar-blind ultraviolet range.²¹ 

The bandgap of the (In_xGa_1−x)₂O₃ ternary alloy films can be modified by different In concentrations and various crystal phases. Therefore, constructing a heterostructure using two different In concentrations and various crystal phase (In_xGa_1−x)₂O₃ thin films also gives rise to the possibility of self-powered photodetection. More importantly, different In concentrations and various crystal phase (In_xGa_1−x)₂O₃ heterostructured photodetectors have their unique advantages: on the one hand, (In_xGa_1−x)₂O₃ heterostructured photodetectors with different In concentrations and various crystal phases are easy to prepare. Particularly, for magnetron sputtering, In content in the films can be easily controlled by the sputtering parameters of In₂O₃ and Ga₂O₃ simultaneously without replacing other targets.²² On the other hand, the crystal phase of (In_xGa_1−x)₂O₃ could be controlled by In content and postannealing conditions. The design of the detector is also flexible. According to different wavelength detection requirements, two suitable In concentrations and various crystal phase of materials can be selected to form a heterostructured photodetector.

In this work, an amorphous (In_0.23Ga_0.77)₂O₃/bixbyite (In_0.67Ga_0.33)₂O₃ heterostructured photodetector [a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ PDs] has been prepared using magnetron sputtering, which exhibits excellent self-powered solar-blind ultraviolet detection performance. The peak wavelength of the photodetector can be adjusted by changing the thickness of the a-(In_0.23Ga_0.77)₂O₃ layer. This work has verified the feasibility of the (In_xGa_1−x)₂O₃ ternary alloy heterojunction with different compositions, providing new ideas for the design of (In_xGa_1−x)₂O₃ heterostructured self-powered solar-blind ultraviolet detectors.

A heterostructure with different In concentrations (In_xGa_1−x)₂O₃ was prepared using magnetron sputtering. Before sputtering, the base vacuum chamber pressure was depressed to 6 × 10⁻⁴ Pa, and the target-substrate distance was kept at 10 cm. First, a layer of the (In_0.67Ga_0.33)₂O₃ film was deposited on a single-side polished c-plane sapphire substrate. During sputtering, the substrate was heated under 600 °C to fabricate bixbyite (In_0.67Ga_0.33)₂O₃ (b-(In_0.67Ga_0.33)₂O₃). Then, a mask was used to cover a part of the film, and a layer of an amorphous (In_0.23Ga_0.77)₂O₃ [a-(In_0.23Ga_0.77)₂O₃] film was grown. During deposition for a-(In_0.23Ga_0.77)₂O₃ and b-(In_0.67Ga_0.33)₂O₃, sputtering power was kept at 100 W, while sputtering pressure was maintained at 1.0 Pa, with O₂ and Ar gas mixed with flow rates of 8 and 40 SCCM, as reported in our previous works.^23,24 Finally, a semitransparent Au and In electrode with a diameter of ∼3 mm was deposited on the a-(In_0.23Ga_0.77)₂O₃ and b-(In_0.67Ga_0.33)₂O₃ films, respectively, using vacuum evaporation, thus constructing the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction photodetector.

The crystal structure and quality of the (In_xGa_1−x)₂O₃ films were investigated by x-ray diffraction (XRD, Rigaku D/max-2600/PC) using Cu-Kα radiation (λ = 0.15418 nm). The film thicknesses were analyzed by using a scanning electron microscope (SEM, Hitachi, SU70). The optical transmittance spectra were characterized by using a TU-1901 double-beam UV-vis spectrophotometer. The compositions of the films were obtained by x-ray photoelectron spectroscopy (XPS, ThermoFisher, ESCLAB 250Xi). The photoelectric test system, comprising a Xe lamp, a monochromator, an optical chopper, and the FS-pro380 semiconductor analyzer, was used to measure the light response of the devices (I-V and I-t characteristics).

The a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction formed on the sapphire substrate is a bixbyite phase (In_0.67Ga_0.33)₂O₃ and amorphous (In_0.23Ga_0.77)₂O₃, determined by XRD (JCPDS, No. 06-0416) and shown in Fig. 1(a). Because of the concentration of In in the (In_0.67Ga_0.33)₂O₃ film is greater than 50%, the formation enthalpy of the monoclinic phase is larger than that of the bixbyite phase. Thus, (In_0.67Ga_0.33)₂O₃ films could only form a bixbyite structure, which is consistent with previous research studies.^25,26 The obtained heterojunction has a thickness of 350 nm for a-(In_0.23Ga_0.77)₂O₃ and 280 nm for b-(In_0.67Ga_0.33)₂O₃ estimated from the cross-sectional FE-SEM image of the heterojunction, as shown in Fig. 1(b). The a-(In_0.23Ga_0.77)₂O₃ shows a strong absorption in the solar-blind ultraviolet region and b-(In_0.67Ga_0.33)₂O₃ shows obvious absorption in the ultraviolet-visible region, as indicated in Fig. 1(c). The bandgap of the (In_xGa_1−x)₂O₃ films was calculated by extrapolating the linear region of the plot of (αhν)² versus hν and taking into account the intercept on the hν-axis, as shown in Fig. 1(d). The estimated bandgap was about 4.06 eV for b-(In_0.67Ga_0.33)₂O₃ and 4.84 eV for a-(In_0.23Ga_0.77)₂O₃.

The structure of the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction photodetector is shown in Fig. 2(a). The I-V curves of Au-(In_0.23Ga_0.77)₂O₃ and In-(In_0.67Ga_0.33)₂O₃ contacts are shown in Fig. 2(b) with a linear relationship, indicating that Au and a-(In_0.23Ga_0.77)₂O₃ as well as In and b-(In_0.67Ga_0.33)₂O₃ are ohmic contacts. Figures 2(c) and 2(d) present significant rectification characteristics of I-V curves for the photodetector under dark and solar-blind ultraviolet light illumination conditions with a self-powered characteristic. The photocurrent of the photodetector is 35 nA under 280 nm light illumination at zero bias, and the photo-to-dark current ratio (PDCR) is 158, indicating that the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction detector exhibits excellent self-powered detection performance.

The bandgaps of a-(In_0.23Ga_0.77)₂O₃ and b-(In_0.67Ga_0.33)₂O₃ thin films are 4.84 and 4.04 eV as shown in Fig. 1(d), respectively. Then, the calculated conduction band offset $Δ E c$ of the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction is 0.29 eV. Combined with $Δ E c$⁠, $Δ E v$⁠, and the bandgaps of the thin films, the energy band diagram of the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction is plotted as shown in Fig. 4(d). The Fermi level of the b-(In_0.67Ga_0.33)₂O₃ thin film is higher than that of the a-(In_0.23Ga_0.77)₂O₃ thin film, and electrons in the b-(In_0.67Ga_0.33)₂O₃ thin film diffuse into the a-(In_0.23Ga_0.77)₂O₃ thin film, forming a built-in electric field from the b-(In_0.67Ga_0.33)₂O₃ thin film to the a-(In_0.23Ga_0.77)₂O₃ thin film. When incident light is irradiated into the built-in electric field region, the photogenerated electrons and holes separate and drift toward the b-(In_0.67Ga_0.33)₂O₃ thin film and the a-(In_0.23Ga_0.77)₂O₃ thin film under the driven by built-in field, respectively, which leads to excellent solar-blind ultraviolet self-powered detection performance.

Due to the bandgap of amorphous (In_0.23Ga_0.77)₂O₃ could be adjusted by film thickness, a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction detectors containing different thicknesses of a-(In_0.23Ga_0.77)₂O₃ were prepared, as shown in Figs. 5(a)–5(d). The thicknesses of the a-(In_0.23Ga_0.77)₂O₃ thin films were approximately 223, 350, 416, and 522 nm, respectively, by adjusting different growth time. The response spectra of a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction detectors are shown in Figs. 5(e)–5(h). The peak responsivities of the heterojunction detectors with different thicknesses of a-(In_0.23Ga_0.77)₂O₃ films are 3.73, 5.78, 0.31, and 0.03 mA/W, respectively. It can be seen that when the thickness of the a-(In_0.23Ga_0.77)₂O₃ thin film increases from 223 to 350 nm, the peak responsivity of the detector increases. However, when the thickness of the a-(In_0.23Ga_0.77)₂O₃ film further increases to 416 and 522 nm, the peak responsivity of the detector decreases significantly, only 5.4% and 0.5% of the responsivity of the detector with 350 nm-thickness a-(In_0.23Ga_0.77)₂O₃ film. Besides the influence of a-(In_0.23Ga_0.77)₂O₃ thickness on responsivity, the peak wavelength of the heterojunction detectors also shifts to longer wavelength, respectively, 240, 248, 260, and 272 nm for the thickness of a-(In_0.23Ga_0.77)₂O₃ with 223, 350, 416, and 522 nm.

To explain its mechanism, the schematic diagrams of the built-in electric field distribution in the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ heterojunction with different thicknesses of a-(In_0.23Ga_0.77)₂O₃ thin films are shown in Fig. 6. The effective absorption thickness of the detector increases, and the peak responsivity of the detector increases from 3.73 to 5.78 mA/W, which indicates that the 223 nm film thickness of a-(In_0.23Ga_0.77)₂O₃ film is entirely within the built-in electric field region, as shown in Fig. 6(a). When the thickness of the a-(In_0.23Ga_0.77)₂O₃ film is further increased, the responsivity of the detector decreases sharply, which indicates that 416 and 552 nm have exceeded the critical thickness. It demonstrates that the nonbuilt-in electric field region already exceeds the critical thickness and weakens the light entering the built-in electric field region by acting as a natural filter, as shown in Figs. 6(c) and 6(d), resulting in the peak responsivity of the detector decreases.

A self-powered solar-blind ultraviolet photodetector based on (In_xGa_1−x)₂O₃ alloy heterojunction with different indium composition and various crystal phase was presented. The a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ detector exhibits an excellent solar-blind selectivity and self-powered detection performance with a responsivity of 5.78 mA/W and a detectivity of 1.69 × 10¹¹ cm Hz^1/2 W⁻¹ under 248 nm illumination. Besides, the peak responsivity of the a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ detectors could be shifted from 240m to 272 nm with enhanced film thickness from 223 to 522 nm of a-(In_0.23Ga_0.77)₂O₃, indicating the influence of a-(In_0.23Ga_0.77)₂O₃ film thickness on the performance of a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ detectors. Thus, this work provides a new strategy to achieve a broadband a-(In_0.23Ga_0.77)₂O₃/b-(In_0.67Ga_0.33)₂O₃ ultraviolet photodetector with flexible and easy fabrication.

This work was financially supported by the National Key Research and Development Program of China (Nos. 2019YFA0705201 and 2019YFA0705204) and the National Natural Science Foundation of China (Grant No. 62174042).

The authors have no conflicts to disclose.

Yiyin Nie and Hongliang Lu contributed equally to this work.

Yiyin Nie: Data curation (equal); Formal analysis (equal); Investigation (equal); Writing – original draft (equal). Hongliang Lu: Data curation (equal); Formal analysis (equal); Investigation (equal). Shujie Jiao: Conceptualization (lead); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (lead); Validation (equal); Writing – original draft (equal); Writing – review & editing (lead). Xianghu Wang: Conceptualization (equal); Data curation (equal); Resources (equal); Validation (equal); Writing – original draft (equal). Song Yang: Formal analysis (equal); Writing – review & editing (equal). Dongbo Wang: Formal analysis (equal); Writing – review & editing (equal). Shiyong Gao: Formal analysis (equal); Investigation (equal); Visualization (equal). Zhendong Fu: Formal analysis (equal); Investigation (equal); Visualization (equal). Aimin Li: Formal analysis (equal); Investigation (equal); Visualization (equal). Jinzhong Wang: Funding acquisition (lead); Methodology (equal); Project administration (lead); Supervision (equal).

The data that support the findings of this study are available from the corresponding authors upon reasonable request.