Growth of β-Ga₂O₃ Single-Crystal Microbelts by the Optical Vapor Supersaturated Precipitation Method

Abstract

Monoclinic β-Ga₂O₃ microbelts were successfully fabricated using a one-step optical vapor supersaturated precipitation method, which exhibited advantages including a free-standing substrate, prefect surface, and low cost. The as-grown microbelts possessed a well-defined geometry and perfect crystallinity. The dimensions of individual β-Ga₂O₃ microbelts were a width of ~50 μm, length of ~5 mm, and thickness of ~3 μm. The SEM, XRD, HRTEM, XPS, and Raman spectra demonstrated the high single-crystalline structure of β-Ga₂O₃ microbelts. Twelve frequency modes were activated in Raman spectra. The optical band gap of the β-Ga₂O₃ microbelt was calculated to be ~4.45 eV. Upon 266 nm excitation, 2 strong UV emissions occurred in photoluminescence spectra through the radiative recombination of self-trapped excitons, and the blue emission band was attributed to the presence of donor-acceptor-pair transition. The individual β-Ga₂O₃ microbelt was employed as metal-semiconductor-metal deep-ultraviolet photodetector, which exhibits the photoresponse under 254 nm. This work provides a simple and economical route to fabricate high-quality β-Ga₂O₃ single-crystal microbelts, which should be a potential synthetic strategy for ultra-wide bandgap semiconductor materials.

1. Introduction

Deep-ultraviolet (DUV) photodetectors that operate in the solar-blind spectrum range (190–280 nm) have attracted much attention due to their potential applications in various fields such as missile warning, biochemical detection, and security communication [1,2,3]. These applications have been demonstrated in semiconductor materials with various device architectures, including ultraviolet enhanced Si-based photodiodes [4], ZnO [5], diamond [6], AlN [7], and AlGaN [8]. Nevertheless, currently available Si-based DUVs require optical filters to block out visible and infrared light, leading to the low penetration depth of high-energy UV photons. Moreover, the high-defect density and complex alloying process in alloy wide bandgap semiconductors can affect the high-quality epitaxy, cut-off wavelength, and signal-noise ratio, etc. As an alternative, monoclinic β-Ga₂O₃ is an attractive class of ultra-wide bandgap semiconductor material due to its excellent physical properties, such as suitable electronic bandgap (4.2~4.9 eV), large breakdown electric field (8 MV/cm) [9,10], and superior room-temperature PL. Furthermore, β-Ga₂O₃ exhibits high optical transparency in both the DUV and visible wavelength regions, which makes it a promising candidate for exploring DUV photodetectors [11]. These distinct properties offer an ideal platform for various unique civil and military applications. To date, several morphologies, including bulk, epitaxial layers, and micro-/nanostructures have been demonstrated on β-Ga₂O₃-based optoelectronic devices. Among them, micro-/nanostructures with a large amount of multiformity, large surface-to-volume ratio, and shorter effective conductive channels were considered an ideal system for investigating dimensionally confined carrier transport phenomena, as well as building blocks for optoelectronic devices. Therefore, it would be greatly advantageous and attractive to explore novel desirable solar-blind photodetectors via an easy fabrication process and low-cost strategy.

Well-established techniques of physical evaporation [12,13], thermal evaporation [14,15], and chemical vapor deposition (CVD) [16,17,18] have been carried out to fabricate abundant β-Ga₂O₃ structures including nanowires, nanobelts, and microwires. Several studies have reported the achievement of preparing β-Ga₂O₃ micro-/nanostructures. The β-Ga₂O₃ nanowires have been grown using a thermal evaporation method, in which metallic gallium was placed on the substrate formed by compact powders of Ga₂O₃ and then sintered at a high temperature of 1150 °C for 8 h [19]. The growth time of β-Ga₂O₃ nanowires was shortened to 3 h through the metal-assist multistep process [20]. Although the catalyst-free CVD method could be applied to fabricate the β-Ga₂O₃ microbelt in 30 min, the complex process of inductively coupled plasma dry etching on sapphire substrate was required [21]. Similarly, the β-Ga₂O₃ microwires could be synthesized in 30 min, for which the clean silicon substrate needed to be covered with 100 nm β-Ga₂O₃ film as a buffer layer [22]. Therefore, the sophisticated fabrication processes in the above techniques limit the development of the practical routes; for example, the time-consuming growth process lowers the growth efficiency, the complexity of the fabricating procedure for the substrate increases the cost, and the metal-assisted catalytic method can lead to the unintentional doping of the β-Ga₂O₃ to form deep-level impurities inside the micro-/nanostructures. Compared with these approaches, the one-step optical vapor supersaturated precipitation (OVSP) method, which was able to adjust the gas flux, growth time, growth power, and other parameters during the growing process in real time, was impressive [23,24,25,26]. During the OVSP process, the growth driving force is based on vapor supersaturated precipitation in the homogeneous temperature field, which is conducive to avoiding the generation of compensation defects. This is obviously different from the previous studies where temperature gradients were used as the growth driving force [27,28], which may lead to thermal-fluctuation-induced lattice mismatching. Moreover, the quasi-equilibrium OVSP method was free of substrate, which could further improve the crystallinity of micro-/nanostructures and reduce cost.

Controlling the dimensionality and crystal quality of β-Ga₂O₃ micro-/nanostructures is not only synthetically interesting, but also could promote the practical application of ultra-wide bandgap semiconductor photodetectors. The optical properties of β-Ga₂O₃ mainly focus on spectroscopic identification of the carrier trapped by intrinsic defects, which were crucial to influencing the carrier concentrations or acting as recombination centers [29,30]. Typical photoluminescence (PL) spectra of β-Ga₂O₃ are acknowledged to be in UV and blue (BL) bands [31,32]. The UV band is traditionally recognized as being impurity-independent, and experiments accompanied by theoretical studies have attributed this to the recombination of a self-trapped exciton (STE) from the free electrons to the self-trapped hole (STH) states [33]. The BL band originates from the donor-acceptor-pair (DAP) recombination [34]. These emission properties could be affected by the deep-level defects. A variety of defect features that vary with growth and processing have been reported, but the specific defect origins of PL bands in β-Ga₂O₃ is still a controversial topic. Further investigation is needed to correlate the photoelectric performance with intrinsic defects.

In this work, we proposed a one-step OVSP method for the fabrication of high-quality β-Ga₂O₃ microbelts with a free-standing substrate. The growth mechanism of β-Ga₂O₃ microbelts was revealed, which depended on a quasi-equilibrium process supported by a uniformed temperature field originating from the homogeneous optical heating zone, rather than temperature gradients. The SEM, XRD, HRTEM, and XPS spectra demonstrated the high single-crystalline structure of β-Ga₂O₃ microbelts. The optical and electrical properties of a single β-Ga₂O₃ microbelt were demonstrated. The optical band gap of the β-Ga₂O₃ microbelt was calculated to be ~4.45 eV. Upon 266 nm excitation, two strong UV and blue emission bands were attributed to the radiative recombination of STE and DAP transition, respectively. The β-Ga₂O₃ microbelt exhibited photoresponse at 254 nm illumination. This work provides a facile approach to prepare high-quality β-Ga₂O₃ microbelts for a variety of optoelectronic applications in micro/nanophotonics.

2. Experimental Section

2.1. Fabrication of β-Ga₂O₃ Microbelts

The mixture of Ga₂O₃ (Alfa Aesar, Ward Hill, MA, USA, 99.999%) and graphite powders with a weight ratio of 1:1 was pressed by isostatic pressing at 68 MPa for 30 min as a reactive precursor rod, the dimensions of which were ~6 mm in diameter and ~3 cm in length. The precursor rod was produced in an optical floating zone furnace (Crystal Systems Co., Ltd, 10000H–HR-I-VPO-PC, Kobuchisawa, Hokuto, Yamanashi, Japan) equipped with four of 1500 W halogen lamps. The lamp power was first increased to 52%@1500 W within 0.3 h. Then the lamp power was held for 0.5 h for the growth of β-Ga₂O₃ microbelts using the OVSP method. Finally, the power was reduced to zero within 0.3 h to release the residual stress of the microbelts. The precursor rod was rotated at 10 rpm during OVSP growth. The carrier gas of O₂ was pumped into the chamber at 50 mL/min. As-grown β-Ga₂O₃ microbelts were cooled to the room temperature naturally.

2.2. Characterization

The morphology and elemental mapping of β-Ga₂O₃ microbelt was obtained by scanning electron microscopic (SEM, Shimadzu SSX-550, Kyoto, Japan), which was equipped with an energy-dispersive spectrometer (EDS). High-resolution transmission electron microscopy (HRTEM) was performed using JEOL JEM-F200. The TEM image of the β-Ga₂O₃ microbelt was taken from a thin edge by grinding the sample. The crystal structure was determined by X-ray diffraction (XRD) (Bruker D8 DISCOVER), which operated at 40 kV and 40 mA with a step size of 0.02° in the diffraction angle range from 20° to 80°. The Raman and PL spectra were acquired by a SmartRaman confocal-micro-Raman system (developed by Institute of Semiconductors, Chinese Academy of Sciences, Beijing, China) under the backscattering geometry. The collected Raman and PL signals were analyzed using a Horiba LabRAM iHR550 spectrometer. For Raman spectral analysis, a 633 nm He-Ne laser with linear polarization (Thorlabs, Newton, NJ, USA, HNL210) was used as the excitation source, equipped with a 2400 lines/mm grating and a 10×/NA 0.25 objective (Olympus MPlan N, Shinjuku, Japan). For PL spectra acquisition, a 266 nm line CW laser (CRYTU FQCW266, Berlin, Germany) was used as the excitation source, equipped with a 600 lines/mm grating and a 20×/NA 0.40 objective (Thorlabs, LMU-20×-UVB). X-ray photoelectron spectroscopic (XPS) analysis was conducted using a photoelectron spectrometer (Thermo Fisher Scientific, Waltham, MA, USA, ESCALAB 250Xi) with monochromatic Al-Kα radiation. The binding energies were then calibrated using the C 1s peak at 284.80 eV to compensate for the surface charges. The ultraviolet-visible (UV-vis) absorption spectrum was recorded by an UV-vis spectrophotometer (UV-3600IPLUS, Shimadzu, Kyoto, Japan), in which the BaSO₄ powder was used as the reference.

2.3. Device Construction

The configuration of the photodetector was based on a single microbelt with (In/Ga)/β-Ga₂O₃/(In/Ga) as the metal-semiconductor-metal (MSM) DUV structure. The In/Ga electrodes were fabricated using a manual droplet deposition technique. Firstly, the β-Ga₂O₃ microbelt was transferred onto a cleaned glass substrate. Then, a needle tube was employed to carefully deposit In/Ga electrodes, approximately 1 mm in diameter, onto both ends of the β-Ga₂O₃ microbelt to establish ohmic contacts. The distance between the In/Ga electrodes was approximately 267 μm, as determined by an optical microscope. The effective working area of the β-Ga₂O₃ microbelt-based DUV detector was measured to be 1.4 × 10⁻⁴ cm². The current-voltage (I-V) curves were measured by a source meter (Keithley, Cleveland, OH, USA, SMU 2600B) equipped with a tip-probe platform and the Auto Curve Analysis Tracer Software v. 2.1.4. The photoresponse curve of the photodetector was recorded using a Xenon lamp with a monochromator sorting filter. All the characterizations and measurements were performed at room temperature and ambient conditions. The standard undoped n-β-Ga₂O₃ single crystal bulk (purchased from Hefei Kejing Material Technology Co., Ltd., Anhui, China) with orientation of <100> was used as control, of which the size was 10 × 10 × 0.5 mm³.

3. Results and Discussion

3.1. Growth of β-Ga₂O₃ Microbelts

Figure 1a shows a schematic diagram of the OVSP process for β-Ga₂O₃ microbelt growth. The growth was carried out in a quartz chamber with controllable gas atmospheres. Light from four halogen lamps was focused on the chamber center by ellipsoidal mirrors in order to form a homogenized optical heating zone, where the Ga₂O₃ precursor rod was positioned and rotated for uniformed heating. Figure 1b demonstrates the imaged OVSP growth process of β-Ga₂O₃ microbelts, which can be divided into two stages. In the first growth stage, cracks emerged in the surface of the precursor rod, accompanied by volatile white smoke, which originated from the decomposition of Ga₂O₃ into Ga₂O and Ga vapors through the carbothermal reduction reaction. When the growth temperature was greater than the threshold for β-Ga₂O₃ microbelts, the concentrations of Ga₂O and Ga vapors were dramatically increased. In the second stage, the Ga₂O and Ga vapors were supersaturated and deposited directly on the precursor rod with a Ga₂O₃ crystal nucleus to form a belt-like architecture. Finally, hundreds of free-standing transparent individual Ga₂O₃ microbelts were obtained from the cluster architectures on the precursor rod. The growth mechanism of Ga₂O₃ microbelts can be specified as:

$${2\mathit{Ga}}_2{O}_3+5C\to {\mathit{Ga}}_{2}O+2\mathit{Ga}+5\mathit{CO}$$

(1)

$${\mathit{Ga}}_2{O+O}_2\to {\mathit{Ga}}_2{O}_3$$

(2)

$${4\mathit{Ga}+3O}_2\to {2\mathit{Ga}}_2{O}_3$$

(3)

3.2. Microstructures of β-Ga₂O₃ Microbelts

Figure 2a shows the typical optical and SEM images of an individual β-Ga₂O₃ microbelt. The β-Ga₂O₃ microbelts were rectangular with a width of ~50 μm and a length of ~5 mm. The thickness of about 3 μm indicated the belt-like architecture of β-Ga₂O₃ microbelts grown by OVSP, as captured in the lower left inset of Figure 2a. The repeatability was verified by the process of multiple growths of β-Ga₂O₃ microbelts under the same optimum conditions. The distribution of diameter (w) measured from 36 microbelts is shown in the upper right inset of Figure 2a, where w = 52 ± 15 nm. The EDS mapping spectra demonstrated the uniform spatial distribution of the Ga and O elements, as shown in Figure 2b. Elements C and Au in the EDS spectrum were assigned to the specimen holder. Figure 2c shows the periodic atomic arrangement of the β-Ga₂O₃ microbelt captured by HRTEM. The well-resolved lattice fringes of 0.297 nm corresponded to the (400) plane [35]. The fast Fourier transition (FFT) pattern taken from the HRTEM image clearly showed the symmetrical sharp diffraction spots, indicating the single-crystal microstructure of the β-Ga₂O₃ microbelt, as shown in the inset in Figure 2c [36]. Figure 2d shows the vertical XRD pattern of an individual β-Ga₂O₃ microbelt, where the growth orientation of the microbelt was placed perpendicularly along the X-ray, as shown in the inset of Figure 2d. All diffraction peaks were indexed as the monoclinic-type structure (JCPDS No. 43-1012). Three dominant peaks located at 30.03°, 45.77°, and 62.45°; corresponded to (400), (600), and (800) crystal planes, respectively. The full width of half peak (FWHM) of the (400) plane was estimated to be ~0.06°, which was ~1.9-fold narrower than that previously reported by Shen et al., indicating the high quality of the β-Ga₂O₃ microbelt [37]. The average crystallite size (D) was calculated using the Scherrer equation:

$$D=\frac{0.89\lambda }{\beta \mathit{cos}\theta }$$

(4)

where λ is the X-ray wavelength of 0.15406 nm, β and θ are the FWHM and diffraction angle of the (400) peak, respectively [38]. The D of β-Ga₂O₃ microbelts was calculated to be 136 nm.

Figure 3a illustrates the micro-Raman spectra of an individual β-Ga₂O₃ microbelt in the temperature range of 300 K to 120 K. The room-temperature Raman spectrum of a standard β-Ga₂O₃ bulk single crystal was set as control. The low-frequency Raman active modes (R1: 113, 144, 169, 200 cm⁻¹) were assigned to the vibration and translation of the tetrahedra-octahedra chains, the mid-frequency modes (R2: 320, 346, 417, 476 cm⁻¹) were attributed to the deformation of Ga₂O₆ octahedra, and the high-frequency modes (R3: 630, 653, 659, 767 cm⁻¹) were related to the stretching and bending phonon frequencies of Ga₁O₄ tetrahedra, respectively [39]. When the temperature increased from 120 K to 300 K, the Raman shift and FWHM of the mode at A_g10 (767 cm⁻¹) exhibited a linear variation, including a blueshift of ~3 cm⁻¹ and a FWHM widening of~2 cm⁻¹, which was attributed to the phonon-phonon interaction under the effect of temperature [40,41], as shown in Figure 3b.

Table 1. Experimentally determined Raman modes at 300 K for the β-Ga₂O₃ microbelt.

Mode Symmetry	Ag1	Bg1	Ag2	Ag3	Ag4	Ag5	Ag6	Ag7	Ag8	Bg5	Ag9	Ag10
β-Ga2O3 bulk single crystal		145	169	200		346	417		630		659	767
β-Ga2O3 microbelt	113	144	169	200	320	346	417	476	630	653	659	767
Dohy et al.	111	147	169	199	318	346	415	475	628	651	657	763

summarizes the position of Raman peaks in this work and previous work reported by Dohy et al. [42]. The existence of only relatively small deviations reveals that the as-grown β-Ga₂O₃ microbelt was a monoclinic single crystal with strain and defect levels close to those of the crystalline bulk [22].

Figure 4a shows the XPS spectrum of the β-Ga₂O₃ microbelt, confirming the existence of main elements, i.e., Ga and O. As shown in Figure 4b, the two peaks of the Ga 2p spectrum were located at ~1145.9 eV and ~1119.0 eV with 26.9 eV separation, which were identified as Ga 2P_1/2 and Ga 2P_3/2, respectively, indicating the presentence of Ga³⁺ ions [43]. The binding energies around 20.79 eV corresponded to Ga 3d electrons [44], as shown in Figure 4c. The O 1s spectrum was fitted by two Gaussian peaks at binding energies of 531.6 eV and 532.5 eV, corresponding to the lattice oxygen (O_I) and the chemisorbed oxygen (O_II) [45], respectively, as shown in Figure 4d. The average Ga/O ratio in Ga₂O₃ can be estimated by the integrated intensity ratio of the deconvoluted XPS components calibrated by the relative sensitivity factor (R) using:

$$R=\frac{I_{{\mathit{Ga}2p}_{3/2}}{/F}_{{\mathit{Ga}2p}_{3/2}}}{I_{O1s}{/F}_{O1s}}$$

(5)

where I is the integrated intensity and F is the sensitivity factor (F_Ga2p3/2 = 2.751 and F_O1s = 0.733). Compared with the previously reported R value of 0.62 in β-Ga₂O₃ single crystal [46], the R value of the β-Ga₂O₃ microbelt was calculated to be 0.68, which was closer to the stoichiometric ratio of Ga₂O₃. The morphological and structural results revealed that the high-quality β-Ga₂O₃ single-crystal microbelts were grown by the OVSP method.

3.3. UV-Vis Absorption and PL Spectra of β-Ga₂O₃ Microbelts

Figure 5 shows the UV-vis absorption spectrum of β-Ga₂O₃ microbelts in the wavelength range of 200–700 nm. Strong absorption was observed in 200–280 nm region, indicating the ultra-wide bandgap nature of β-Ga₂O₃. The Tauc’s fitting was used to identify the band gap of the β-Ga₂O₃ microbelt as follows [47]:

$$\alpha h\upsilon =C{(h\upsilon -E_g)}^{1/2}$$

(6)

where α is the absorption coefficient, h is the Planck’s constant, ν is the frequency of incident photon, and E_g is the optical bandgap. Plotting (αhν)² as a function of photon energy, and extrapolating the linear portion of the curve to absorption equal to zero as shown in the inset of Figure 5, gave the E_g value of the β-Ga₂O₃ microbelt to be 4.45 eV, in good agreement with the reported value [48]. This feature makes the β-Ga₂O₃ microbelt possess the selective response to the solar-blind UV band.

Figure 6a,b shows the PL spectra of an individual β-Ga₂O₃ microbelt and the high-quality bulk β-Ga₂O₃ single crystal. The broad UV emission band of the individual β-Ga₂O₃ microbelt peak could be fitted by three Gauss peaks at 332 nm (3.73 eV, named UV-1), 368 nm (3.37 eV, named UV-2), and 402 nm (3.08 eV, named BL), which are close to those of the bulk β-Ga₂O₃ single crystal at 329 nm, 366 nm, and 403 nm respectively. The UV-1 and UV-2 emissions were attributed to the recombination of STE from the free electrons to the STH states, which were localized on the single oxygen atom in the lattice. The oxygen atom at the O I site contributed the UV-1 emission, while the one at the O II site activated the UV-2 emission [49,50]. The BL band appearing at 2.8–3.0 eV was related to the DAP transitions involving the deep donors (e.g., oxygen vacancies (V_O) and interstitial Ga) and acceptors (e.g., gallium vacancies (V_Ga), and V_O + V_Ga complexes) [51,52,53]. Figure 6c shows the schematic band diagram and emission behavior in the β-Ga₂O₃ microbelt. Upon 266 nm illumination, the UV-1 and UV-2 emissions occurred through the radiative recombination of STE. Meanwhile, the DAP emissions for the BL band were excited by the transition from the V_O defect to the (V_Ga + V_O) complex, of which the charge state was (1−), as calculated by Deák et al. [54]. The FWHM of PL emissions has been regarded as a well-accepted criterion to evaluate the crystal quality of semiconductor materials. As the temperature increased from 80 K to 300 K, the FWHM of BL emissions in the β-Ga₂O₃ microbelt and bulk β-Ga₂O₃ single crystal expanded from 88–94 nm and 83–87 nm, respectively, as illustrated in Figure 6d. The broader FWHM of β-Ga₂O₃ should be ascribed to the large Stokes shift of PL bands for optical absorption, caused by the strong electron-phonon coupling and the localization of recombining electrons and holes [55,56].

3.4. I-V Characteristics

Figure 7a illustrates the schematic of a single β-Ga₂O₃ microbelt with In/Ga electrodes, where the UV signal was illuminated (100 mW/cm²) vertically on the MSM DUV detector. An optical microscopy image of the MSM architecture is demonstrated in the upper left of Figure 7a. The current-voltage (I–V) curves of the β-Ga₂O₃ microbelt measured under light illumination (254 nm) and in the dark are depicted in Figure 1b. The linear I-V relationships indicated ohmic contacts between In/Ga electrodes and the β-Ga₂O₃ microbelt [57]. For the belt-like architecture, the electrical resistivity (ρ) of β-Ga₂O₃ microbelt was calculated by:

$$\rho =R\times \frac{s}{l}=R\times w$$

(7)

where R is the resistance, s is the cross-sectional area of the microbelt, l is the length, and w is the diameter. Based on the equation, the ρ value was calculated to be $1.79\times {10}^7\Omega \cdot \mathrm{cm}$, which was in the same order of magnitude as the value reported in the previous literature [58]. Upon 254 nm illumination, the current of the MSM DUV detector increased from 5.8 nA to 7.5 nA. The positive photocurrent was attributed to the solid-state process where the photogenerated electron-hole pairs instantaneously boosted the carrier density contributing the photoconduction [29]. The responsivity (R) of the β-Ga₂O₃ photodetector under a voltage of 20 V and at a power density of 100 mW/cm² was calculated as follow:

$$R=\frac{I_P}{P_{UV}\times S}$$

(8)

where I_P was the photocurrent minus the dark current, P is power density, and S is effective illumination area. In this work, λ was 254 nm, I_λ was 7.5 × 10⁻⁹ A, I_d was 5.8 × 10⁻⁹ A, and S was 1.4 × 10⁻⁴ cm². The calculated R value was found to be 0.12 mA/W, which is comparable to the R value reported in previous studies on β-Ga₂O₃-based DUV photodetectors [59]. The inset of Figure 7b presents the stability characteristics of the β-Ga₂O₃ microbelt at room temperature. After the stability of 100 s, no obvious degradation was found. However, the rejection ratio of our β-Ga₂O₃ photodetector is lower than that reported in the literature [60]. One possible approach is to optimize the synthesis parameters of the β-Ga₂O₃ microbelts to enhance their crystalline quality and minimize defects, which may improve the UV performance of the photodetectors. Another approach is to modify the device structure, such as by introducing the functional layer, to improve the separation efficiency of photogenerated charge carriers and enhance the rejection ratio. These efforts will lead to improvements in the performance of the β-Ga₂O₃ photodetector, and a discussion of solutions will certainly be included in our future works. Because of the resulting photoelectric properties and the easy fabrication process, the OVSP method is a promising tool for preparing ultra-wide bandgap semiconductor materials in future optoelectronic system applications.

4. Conclusions

The β-Ga₂O₃ microbelts were grown by the OVSP method. The vapor supersaturated precipitation and uniformed temperature field provided the growth driving force, which is key to the growth process of β-Ga₂O₃ microbelts. The structural investigations of SEM, XRD, HRTEM, XPS and Raman spectra demonstrated the single-crystal character of the β-Ga₂O₃ microbelts. Twelve active frequency modes were observed in Raman spectra. With the temperature increasing, the Raman shift and FWHM of the mode at A_g10 exhibited a redshift of ~3 cm⁻¹ and a FWHM widening of~2 cm⁻¹, respectively. The optical band gap of the β-Ga₂O₃ microbelt was found to be ~4.45 eV. Two strong UV emissions and blue emissions in the PL spectra could be attributed to the recombination of STE from the free electrons to the STH states and the presence of DAP transition, respectively. Under illumination of 254 nm, the MSM UV detector of the β-Ga₂O₃ microbelt possessed DUV sensing properties. This work provides an available method to achieve ultra-wide bandgap semiconductor materials for micro-/nanophotonic applications.