Photoluminescence, luminescence excitation spectra, and electrical conductivity of β-Ga2O3-In2O3 solid solutions were studied. For this purpose, polycrystalline samples of unintentionally doped (UID) and doped with Ca or Zr β-Ga2O3-In2O3 solid solution with 20% In were synthesized and characterized. All samples were obtained by the high-temperature solid-phase method from appropriate oxides at 1300 °C at low and high oxygen partial pressure. It was established that UID and doped with Ca2+ or Zr4+ samples synthesized in an oxygen atmosphere were highly resistive, while the samples synthesized in an argon atmosphere had high conductivity. The conductivity was the lowest in the samples doped with Ca2+ and was 10−13 Ω−1 cm−1, while in the samples doped with Zr4+, the electrical conductivity was the highest and reached 10−3 Ω−1 cm−1. The broadband luminescence of β-Ga2O3-In2O3 solid solution is a superposition of three elementary bands with maxima in the violet 3.08 eV, blue 2.73 eV, and green 2.45 eV regions of the spectrum. Doping with Ca2+ or Zr4+ impurities and varying the synthesis atmosphere led mainly to a redistribution of intensities between the elementary luminescence bands. The luminescence arises from the radiative recombination of charge carriers through donor–acceptor pairs and self-localized holes. Donors and acceptors are formed by native defects such as (Gai, VGa, VGaVo) or doping impurities (Zr4+, Ca2+). Unlike the luminescence spectra, the luminescence excitation spectra change significantly when the synthesis conditions vary or when doping with divalent impurities. The excitation band at 4.46 eV is due to electron transitions from the VGa or VGaVO acceptor levels to the conduction band. Electron transitions from acceptor levels of Ca2+ impurities are manifested in the intense excitation band at 4.1 eV.

Gallium oxide β-Ga2O3 is an attractive ultra-wide bandgap (∼4.8 eV) semiconductor for power electronics1,2 and UV optoelectronics.3–5 Solid solutions with the β-Ga2O3 structure are formed in the Ga2O3-In2O3 and Ga2O3-Al2O3 systems.6–9 An increase in the indium concentration is accompanied by a decrease in the bandgap,6,7 while an increase in the concentration of Al, on the contrary, leads to its increase.8,9 The ability to easily change the bandgap in solid solutions opens up additional opportunities for varying the material’s electrical, optical, and photoelectric properties, creating heterostructures, and, as a result, can be helpful for the most widely practical use of the materials.9–13 

The successful use of β-Ga2O3 became possible due to producing quality and high-conductive crystals and films of β-Ga2O3 doped with Si, Sn, Ge, or Zr.1,2,14 On the other hand, doping β-Ga2O3 with Mg or Fe impurities made it possible to obtain high-resistance substrates to manufacture power diodes and transistors.2 Doping impurities create energy levels in the bandgap of β-Ga2O3, acting as donors, acceptors, traps, and recombination centers for charge carriers. Many works devoted to the study of energy levels in β-Ga2O3 (Refs. 15–19) have been published. Most of these works are dedicated to studying shallow and deep energy levels, located near the bottom of the conduction band and are donors or traps for electrons. At the same time, significantly fewer works are devoted to the study of energy levels that are created by acceptor impurities and are located near the top of the valence band. Such levels in oxides, as a rule, serve as traps for holes and recombination centers for electrons. To investigate such deep energy levels, studies of photoconductivity,20,21 EPR,22,23 deep-level optical spectroscopy, DLTS,19 and positron annihilation spectroscopywere used.24 

Since solid solutions in the β-Ga2O3-In2O3 system have attracted the attention of researchers quite recently, the effect of impurities on the optical and electrical properties and the spectrum of traps in the material have not been studied to date. In one of the first papers,25 it was shown that β- Ga2O3-In2O3 possesses intense luminescence. An increase in indium concentration is accompanied by changes in the bandgap width7 and the position of the energy levels of impurities and intrinsic defects in the forbidden band of the material.25 Studying electrical conductivity, luminescence, and photoluminescence excitation spectra will allow us to detect and identify the energy levels created by impurities and intrinsic defects in the β-Ga2O3-In2O3 bandgap. Such studies are highly relevant to the practical application of the material. Therefore, this work is devoted to the study of recombination luminescence, luminescence excitation spectra, and conductivity of polycrystalline samples of β- Ga2O3-In2O3 solid solution with 20 mol. % In2O3 content synthesized in an oxygen atmosphere or an inert argon atmosphere and doped with potential donor (Zr4+) or acceptor (Ca2+) type impurities.

Polycrystalline undoped and Ca2+ or Zr4+ impurities doped β-Ga2O3-In2O3 solid solution of nominal compositions β-(Ga1−xInx)2O3 (x = 0.2) were obtained by the high-temperature solid-state chemical reaction method. To study the role of native defects in creating energy levels in the forbidden band of the material, β-(Ga0.8In0.2)2O3 solid solution samples with various concentrations of defects were synthesized. The intrinsic point defects in β-(Ga0.8In0.2)2O3 solid solution are oxygen vacancies, metal vacancies, and interstitial atoms. The concentration of such defects depends on the partial pressure of the components during the synthesis of samples. Therefore, part of the samples was synthesized in an oxygen atmosphere at a pressure of ∼1 atm. Another part of the samples was prepared under conditions of low partial pressure of oxygen, for which synthesis was used in an atmosphere of inert argon gas. Gallium oxide (β- Ga2O3), indium oxide (In2O3), CaCO3, and ZrO2 of at least 4N purity grade were used as starting materials. The concentration of indium oxide in the initial mixture of solid solution components was 20 mol. %, while the concentration of Ca and Zr impurities was 0.05 at. %. According to the supplier’s specification, used oxides contain the following main donor and acceptor impurities in ppm (Max): Si ≤ 10, Ge ≤ , Sn ≤ , Fe ≤ 5, Cr ≤ 5, Mg ≤ 0, Ca≤2, Zr ≤ 1.

First, the powder mixture of oxides, weighted in stoichiometric amounts, was mechanically mixed in an agate mortar until a homogeneous mass was obtained. Samples for the synthesis were obtained as plane-parallel tablets with a diameter of 8 mm and a thickness of 1 mm by mechanical pressing. After that, tablets were wrapped in platinum foil and placed in quartz ampoules. Samples in ampoules were subjected to synthesis in an electric furnace at a temperature of 1300 °C and treated during 12 h. Before the synthesis, air was removed from the ampoule, and a cylinder with oxygen or argon gas was connected. All undoped and Ca- or Zr-doped β-(Ga0.8In0.2)2O3 ceramic samples synthesized in an oxygen atmosphere were white. Undoped and Zr-doped β-(Ga0.8In0.2)2O3 ceramic samples synthesized in an argon atmosphere were bluish. At the same time, Ca-doped ceramics samples of β-(Ga0.8In0.2)2O3, synthesized in argon, had a white color. XRD examination revealed that all synthesized ceramic samples show a monoclinic structure isotypic with β-Ga2O3.

The presented study involves measuring the luminescence spectra under optical excitation and studying photoluminescence excitation spectra at room temperature, utilizing a Solar CM2203 spectrofluorimeter. A Hamamatsu R928 photomultiplier registered experimental photoluminescence data with a resolution of 1 nm. Appropriate corrections were applied to the photoluminescence spectra to account for the system’s spectral response. The measured luminescence excitation spectra were corrected by the xenon lamp emission spectrum.

The two-probe method was used to measure the electrical conductivity of polycrystalline samples of the above-mentioned solid solutions. Indium contacts were formed on the front and back surfaces of the samples. Preliminary analysis of the current-voltage characteristics revealed that indium establishes good ohmic contacts with high-resistance samples. Currents in the range from 10−3 to 10−14 A were measured using a B7-30 electrometer.

Polycrystalline samples of β-(Ga0.8In0.2)2O3 possess luminescence when excited by light from the region of the fundamental absorption edge. Figures 1(a)1(c) show the photoluminescence spectra of unintentionally doped (UID) as well as Zr- or Ca-doped β-(Ga0.8In0.2)2O3 ceramics. The photoluminescence of β-(Ga0.8In0.2)2O3 UID samples synthesized in an oxygen atmosphere [Fig. 1(a)] is broad and extends from 1.5 to 3.5 eV with an emission maximum near 2.63 eV. The complex luminescence spectrum of the solid solution was decomposed into three elementary curves using Gaussian functions. It was taken into account previously obtained data that the luminescence spectrum of gallium oxide is complex and contains at least three distinct bands with known peak positions,15,26 the relative contribution of which is determined by the presence of impurities and synthesis conditions. A slight shift of the maxima of the elementary luminescence bands in solid solutions is due to decreased bandgap width when indium ions replace gallium ions. The luminescence spectrum was decomposed into elementary curves with maxima in violet 3.08 eV, blue 2.74 eV, and green 2.47 eV region of the spectrum. The half-width of elementary luminescence bands was ∼0.4 eV. The blue and green luminescence bands were the most intense for ceramics synthesized in oxygen. Figure 1(b) shows the photoluminescence spectra of UID β-(Ga0.8In0.2)2O3 ceramics synthesized in an argon atmosphere. A slight decrease in the intensity of luminescence and a shift of the integral maximum of emission to the short-wavelength spectrum region to 2.9 eV [Fig. 1(b)] were observed. Figure 1(c) shows the normalized difference spectrum of luminescence for samples synthesized in an atmosphere of oxygen and argon. The given spectrum shows that a decrease in the oxygen concentration in the atmosphere during the synthesis of ceramics leads to an increase in the intensity of the short-wave violet luminescence band and a reduction in the intensity of the green luminescence band.

FIG. 1.

Luminescence spectrum of UID β-(Ga0.8In0.2)2O3 [(a)–(c)], β-(Ga0.8In0.2)2O3:Zr ceramics [(d)–(f)], and β-(Ga0.8In0.2)2O3:Ca [(g)–(i)] ceramics synthesized in oxygen [(a), (d), and (g)] and argon [(b), (e), and (h)], and the normalized difference spectrum of luminescence for samples synthesized in an oxygen and argon atmospheres [(c), (f), and (i)].

FIG. 1.

Luminescence spectrum of UID β-(Ga0.8In0.2)2O3 [(a)–(c)], β-(Ga0.8In0.2)2O3:Zr ceramics [(d)–(f)], and β-(Ga0.8In0.2)2O3:Ca [(g)–(i)] ceramics synthesized in oxygen [(a), (d), and (g)] and argon [(b), (e), and (h)], and the normalized difference spectrum of luminescence for samples synthesized in an oxygen and argon atmospheres [(c), (f), and (i)].

Close modal

The luminescence spectrum of β-(Ga0.8In0.2)2O3:Zr polycrystalline samples is shown in Figs. 1(d)1(f). The highest emission yield was also observed for samples synthesized in an oxygen atmosphere [Fig. 1(d)]. The luminescence maximum was located near 2.64 eV. The emission band was also decomposed well into the same elementary Gaussian curves with maxima at 3.08, 2.74, and 2.44 eV. The β-(Ga0.8In0.2)2O3:Zr ceramic shows a very slight decrease in the relative intensity of the blue and violet luminescence bands compared to the UID ceramic. The synthesis in an argon atmosphere [Fig. 1(e)] led also to a decrease in the integrated luminescence of the β-(Ga0.8In0.2)2O3:Zr compared to undoped ceramics. In addition, an even more significant increase in the relative intensity of the violet luminescence band was observed in Zr-doped ceramics [Fig. 1(f)].

The luminescence spectrum of β-(Ga0.8In0.2)2O3:Ca ceramics [Figs. 1(g)1(i)] is also similar to the luminescence spectrum of undoped ceramics. The highest luminescence yield was also observed for samples synthesized in an oxygen atmosphere. The integral luminescence maximum was observed at 2.64 eV. There is an increase in the green luminescence band and a slight decrease in the intensity of the violet luminescence band [Fig. 1(g)]. Synthesis in an argon atmosphere led to a decrease in the integrated luminescence yield by approximately three times [Fig. 1(h)] and a slight increase in the relative intensity of the violet luminescence band compared to the β-(Ga0.8In0.2)2O3:Ca sample synthesized in an oxygen atmosphere [Fig. 1(i)].

Figures 2(a) and 2(b) show the photoluminescence excitation spectra for the green, blue, and violet luminescence bands of UID β-(Ga0.8In0.2)2O3 samples. It can be seen that the excitation spectra of all luminescence bands are similar in the energy range of 3.5–5.5 eV. Since the excitation spectrum of the blue-green luminescence covers a broader range of energies from 2.7 to 5.7 eV, further decomposition of the photoluminescence excitation spectrum into elementary bands and their analysis is given [Figs. 2(c) and 2(d)] for the luminescence registration at 480 nm.

FIG. 2.

Photoluminescence excitation spectra of elementary emission bands of the UID β-(Ga0.8In0.2)2O3 ceramics synthesized in oxygen (a) and argon (b) atmospheres. The shaded Gaussians that illustrate the computer approximation of the excitation spectrum [(c) and (d)] and a schematic image of the emission corresponding to the spectral range for which the excitation spectra were recorded.

FIG. 2.

Photoluminescence excitation spectra of elementary emission bands of the UID β-(Ga0.8In0.2)2O3 ceramics synthesized in oxygen (a) and argon (b) atmospheres. The shaded Gaussians that illustrate the computer approximation of the excitation spectrum [(c) and (d)] and a schematic image of the emission corresponding to the spectral range for which the excitation spectra were recorded.

Close modal

The excitation spectrum of the green luminescence band of the UID β-(Ga0.8In0.2)2O3 ceramic sample synthesized in an oxygen atmosphere covers a wide range of energies from 2.7 to 5.5 eV and overlaps the fundamental absorption region of 4.7–5.5 eV and the transparency region of 2.7–4.7 eV of the material [Fig. 2(a)]. The total maximum of the excitation band is located at energies near ∼4.5 eV. This broad luminescence excitation band was decomposed into elementary Gaussian bands taking into account the position of the fundamental absorption edge21,27 and the photoconductivity spectra of gallium oxide and the solid solution,5 and the estimation of the bandgap of the solid solution based on the results of diffuse reflection spectra.28 As a result of the decomposition, two intense elementary excitation bands were obtained with maxima near 4.9 and 4.44 eV and a low-intensity excitation band at about 3.3 eV [Fig. 2(c)]. The elementary excitation bands are shown in the figure by shaded Gaussians. The 4.9 eV excitation band is located in the fundamental absorption region, and its half-width is ∼0.7 eV. The 4.44 and 3.3 eV bands are located in the material transparency region. The half-width of the 4.44 eV excitation band is smaller and is about ∼0.4 eV. The most intense is the excitation band at 4.9 eV. The intensity of the excitation band at 4.44 eV is approximately 0.65 of the intensity of the band at 4.9 eV. The relative excitation intensity of the 3.3 eV band is approximately 0.1 of the excitation intensity at the maximum. For the solid solution samples synthesized in an argon atmosphere [Fig. 2(b)], a slight increase in the relative intensity of the excitation bands of 4.44 and 3.3 eV was observed [Fig. 2(d)].

Figure 3 shows the excitation spectra of the luminescence bands of β-(Ga0.8In0.2)2O3:Zr samples. The green luminescence band has an excitation spectrum similar to the excitation spectrum of undoped ceramics, with a maximum near 4.5 eV. The main elementary bands are also located at ∼4.9 and 4.44 eV for β-(Ga0.8In0.2)2O3:Zr. In these samples, the relative intensity of the 4.44 eV excitation band increased compared to undoped samples. More significant changes in the excitation spectrum are observed for β-(Ga0.8In0.2)2O3:Zr samples synthesized in an argon atmosphere [Fig. 3(b)]. In particular, the 4.9 eV luminescence excitation band is the main one, and the intensity of the 4.44 eV excitation band is reduced. In addition, for samples synthesized in an inert argon atmosphere, the relative intensity of the 3.3 eV excitation band increased 10 times compared to samples synthesized in oxygen.

FIG. 3.

Photoluminescence excitation spectra of elementary emission bands of the β-(Ga0.8In0.2)2O3:Zr ceramics synthesized in oxygen (a) and argon (b) atmospheres. The shaded Gaussians that illustrate the computer approximation of the excitation spectrum [(c) and (d)] and a schematic image of the emission corresponding to the spectral range for which the excitation spectra were recorded.

FIG. 3.

Photoluminescence excitation spectra of elementary emission bands of the β-(Ga0.8In0.2)2O3:Zr ceramics synthesized in oxygen (a) and argon (b) atmospheres. The shaded Gaussians that illustrate the computer approximation of the excitation spectrum [(c) and (d)] and a schematic image of the emission corresponding to the spectral range for which the excitation spectra were recorded.

Close modal

The β-(Ga0.8In0.2)2O3:Ca ceramic synthesized in an oxygen atmosphere [Fig. 4(a)] also has an excitation spectrum similar to the excitation spectrum of luminescence of UID β-(Ga0.8In0.2)2O3 [Fig. 2(a)]. The blue-green luminescence decomposed elementary excitation peaks are also located at 4.9 and 4.44 eV [Fig. 4(c)]. In β-(Ga0.8In0.2)2O3 samples doped with Ca impurity, the intensity of the 4.44 eV excitation band decreased relative to the 4.9 eV excitation band. A low-intensity broad excitation band with a maximum near 3.3 eV is also observed for these samples. Its intensity is only 0.01 of the intensity at the maximum of the excitation band.

FIG. 4.

Photoluminescence excitation spectra of elementary emission bands of the β-(Ga0.8In0.2)2O3:Ca ceramics synthesized in oxygen (a) and argon (b) atmospheres. The shaded Gaussians illustrate the computer approximation of the excitation spectrum [(c) and (d)] and a schematic image of the emission corresponding to the spectral range for which the excitation spectra were recorded.

FIG. 4.

Photoluminescence excitation spectra of elementary emission bands of the β-(Ga0.8In0.2)2O3:Ca ceramics synthesized in oxygen (a) and argon (b) atmospheres. The shaded Gaussians illustrate the computer approximation of the excitation spectrum [(c) and (d)] and a schematic image of the emission corresponding to the spectral range for which the excitation spectra were recorded.

Close modal

The synthesis of polycrystalline samples of β-(Ga0.8In0.2)2O3:Ca ceramics in an inert argon atmosphere also led to drastic changes in the form of the excitation spectrum [Fig. 4(b)]. The maximum of the excitation spectrum has shifted to the low-energy region and is located near 4.12 eV. It is located far in the region of crystal transparency. The band at 4.12 eV became the main one in the excitation spectrum [Fig. 4(b)]. Its intensity is several times higher than that of higher energy excitation bands at 4.46 and 4.9 eV. A broad low-intensity excitation band is also visible in the transparency region at 3.3 eV.

Figure 5 shows the temperature dependence of the electrical conductivity of UID as well as Ca- or Zr-doped β-(Ga0.8In0.2)2O3 ceramic samples. The electrical conductivity of undoped and doped β-(Ga0.8In0.2)2O3 samples synthesized in an oxygen atmosphere was relatively low at 295 K. It varied from ∼10−13 Ω−1 cm−1 for samples doped with Ca to ∼2 × 10−11 Ω−1 cm−1 for samples doped with Zr. The activation energy of conductivity in such high-resistance samples was considerable and lay in the energy range of 0.8–0.98 eV.

FIG. 5.

Dependence of the conductivity of different β-(Ga0.8In0.2)2O3 samples on the reciprocal temperature. Empty polyhedra illustrate the conductivity of samples synthesized in an oxygen atmosphere, and filled polyhedra demonstrate the conductivity of samples synthesized in an inert argon atmosphere.

FIG. 5.

Dependence of the conductivity of different β-(Ga0.8In0.2)2O3 samples on the reciprocal temperature. Empty polyhedra illustrate the conductivity of samples synthesized in an oxygen atmosphere, and filled polyhedra demonstrate the conductivity of samples synthesized in an inert argon atmosphere.

Close modal

At the same time, the conductivity was much higher for β-(Ga0.8In0.2)2O3 ceramics samples that were synthesized in an argon atmosphere. It varied greatly depending on the type of doping impurity, from ∼4 × 10−7 Ω−1 cm−1 for samples doped with Ca to ∼10−3 Ω−1 cm−1 for samples doped with Zr. The activation energy of electrical conductivity in β-(Ga0.8In0.2)2O3:Ca samples was 0.68 eV. The activation energy of such highly conductive β-(Ga0.8In0.2)2O3:Zr samples was ∼0.15 eV. It should be noted that the UID ceramic samples synthesized in an argon atmosphere had electrical conductivity higher than calcium-doped samples but lower than zirconium-doped samples.

Note that in all cases, the conductivity of samples synthesized in an argon atmosphere was higher than the conductivity of samples synthesized in an oxygen atmosphere. The main electrical characteristics of different β-(Ga0.8In0.2)2O3 ceramic samples are presented in Table I.

TABLE I.

Value of electrical conductivity of polycrystalline β-(Ga0.8In0.2)2O3 samples at a temperature of 300 K and the activation energy of electrical conductivity in the temperature range of 300–500 K.

Annealing atmosphereUID β-(Ga0.8In0.2)2O3β-(Ga0.8In0.2)2O3:Zrβ-(Ga0.8In0.2)2O3:Ca
ArgonOxygenArgonOxygenArgonOxygen
Electrical conductivity at 300 K, Ω−1 cm−1 4 × 10−6 6 × 10−12 1 × 10−3 2 × 10−11 4 × 10−7 ∼10−13 
Activation energy, eV ∼0.15 0.9 ∼0.15 0.8 0.68 0.98 
Annealing atmosphereUID β-(Ga0.8In0.2)2O3β-(Ga0.8In0.2)2O3:Zrβ-(Ga0.8In0.2)2O3:Ca
ArgonOxygenArgonOxygenArgonOxygen
Electrical conductivity at 300 K, Ω−1 cm−1 4 × 10−6 6 × 10−12 1 × 10−3 2 × 10−11 4 × 10−7 ∼10−13 
Activation energy, eV ∼0.15 0.9 ∼0.15 0.8 0.68 0.98 

The luminescence spectra of β-(Ga0.8In0.2)2O3 polycrystalline samples (Fig. 1) resemble the luminescence of gallium oxide.26,29–33 Elementary luminescence bands are shifted to longer wavelengths in β-(Ga0.8In0.2)2O3. As in gallium oxide, the luminescence yield and spectral composition in the β-(Ga0.8In0.2)2O3 solid solution are almost independent of dopants. Doping with divalent Ca2+ or tetravalent Zr4+ impurities did not lead to the appearance of new luminescence bands but only to a change in the shape of the luminescence spectrum. Doping and varying the synthesis atmosphere, led mainly to a redistribution of intensities between the elementary luminescence bands, also observed for UID samples. Synthesis in an oxygen atmosphere and doping with divalent acceptor impurities always led to an increase in the relative intensity of long-wave luminescence. Synthesis in an atmosphere of inert argon gas, as well as doping with donor-type impurities, on the contrary, led to an increase in the intensity of short-wavelength luminescence bands. The luminescence of β-(Ga0.8In0.2)2O3 solid solutions is well explained based on previously proposed models of luminescence centers for gallium oxide.26,29–33 In particular, luminescence in the violet, blue, and green parts of the spectrum occurs due to the radiative recombination of carriers through DAP. Background impurities of tetravalent metals, hydrogen, interstitial metals, and oxygen vacancies with two trapped electrons act as shallow or deep donors. The acceptors component of DAP is formed by native acceptor-type defects, such as gallium vacancies VGa, bivacancies (VGaVO), or divalent metal impurities Ca2+. Since there are shallow (Si4+, Ge4+, Zr4+, Gai) and deep (VO + 2e) donors in the β-(Ga0.8In0.2)2O3 samples, the maxima of the DAP luminescence bands can be located in a wide range of wavelengths from 500 to 350 nm. Violet luminescence in β-(Ga0.8In0.2)2O3 can also arise due to the recombination of electrons with self-trapped holes or holes localized on defects.32 

β-(Ga0.8In0.2)2O3 has a β-Ga2O3 crystal structure where part of the octahedrally coordinated gallium atoms are replaced by indium atoms.6 The high partial pressure of oxygen during the synthesis leads to an increase in the energy of the formation of oxygen vacancies,34,35 i.e., to a decrease in their concentration. At the same time, the formation energy of gallium vacancies decreases, and their concentration increases. The low partial pressure of oxygen, on the contrary, leads to a decrease in the energy of the formation of oxygen vacancies, and an increase in the energy of the formation of gallium vacancies. Therefore, the dominant defects will be oxygen vacancies.

Excitation bands at 4.9, 4.44, and 3.3 eV were found in the excitation spectra of doped and undoped β-(Ga0.8In0.2)2O3 samples. The excitation band with a maximum of 4.9 eV is located in the fundamental absorption region. It is observed in all samples and corresponds to interband transitions in β-(Ga0.8In0.2)2O3. Excitation bands at 4.44 and 3.3 eV are located before the fundamental absorption edge in the region transparency of β-(Ga0.8In0.2)2O3. Their relative intensities change with doping and varying synthesis conditions.

The UID β-(Ga0.8In0.2)2O3 samples synthesized in an oxygen atmosphere were highly resistive. At the same time, similar samples, but synthesized in an atmosphere of argon, were lower resistive. As in gallium oxide, there is a relatively high ∼10–100 ppm concentration of background donor impurities (Si, Sn, etc.) in β-(Ga0.8In0.2)2O3. The low conductivity of the β-(Ga0.8In0.2)2O3 samples synthesized in the oxygen atmosphere indicates that the total concentration of compensating acceptors is close to the concentration of background donors. As can be seen from the raw material specification, the concentration of background impurity acceptors in the raw material is lower than that of donors. Therefore, samples synthesized in an oxygen atmosphere additionally have a sufficiently high concentration of native acceptors. Such native acceptors in β-(Ga0.8In0.2)2O3 are gallium vacancies VGa (Ref. 24) or complexes based on them [bivacancies VGaVO, vacancy-interstitial complexes 2VGa-Gai (Ref. 36)], the concentration of which is always higher in samples synthesized in oxygen than in samples synthesized in argon. The acceptors create energy levels near the top of the valence band. Electron transitions from electron-filled acceptor levels to the conduction band will manifest as excitation bands. The luminescence excitation band with a maximum near 4.44 eV is observed in both undoped and doped samples. Therefore, this excitation band at 4.44 eV may be due to electronic transitions from native acceptor defect levels to the conduction band. The number of acceptors with trapped electrons is higher in conductive samples, so the 4.44 eV band has a slightly higher relative intensity in samples synthesized in argon.

Zr creates shallow donor levels near the bottom of the conduction band. The concentration of the doping Zr impurity was high and amounted to 0.05 at. %. However, β-(Ga0.8In0.2)2O3:Zr samples synthesized in an oxygen atmosphere were also quite highly resistive. It is obvious that such a low conductivity of β-(Ga0.8In0.2)2O3 samples doped with a zirconium donor impurity is caused by a high concentration of compensating acceptors. The concentration of background acceptor impurities in the raw material was 2 orders of magnitude lower than the concentration of Zr donors, so the Zr donors are compensated by the intrinsic acceptor defects. Gallium vacancies or their complexes can be such intrinsic defects of the acceptor type. In particular, these acceptor levels can be formed by intrinsic defects such as VGa gallium vacancies or complexes based on them, for example, (VGaVO) or vacancy-interstitial complexes (2VGa Gai).36 This assumption is confirmed by the results presented in paper,37 where it was found that the annealing of β-Ga2O3:Zr samples in oxygen leads to an increase in the resistance of the samples due to the formation of gallium vacancies. Since the samples had a high resistance, the concentration of acceptors was close to the concentration of the zirconium impurity. In such high-resistance samples, most of the acceptor levels are filled with electrons. Accordingly, transitions of electrons from these levels to the conduction band should be observed in the luminescence excitation spectrum. That is, the main luminescence excitation band in the β-(Ga0.8In0.2)2O3:Zr samples, synthesized in an oxygen atmosphere at 4.44 eV, is due to transitions from such acceptor levels. The excitation band in the region of shorter wavelengths at 4.9 eV is caused by interband transitions.

Divalent Ca2+ ions replace Ga3+ ions in the crystal lattice of the β-(Ga0.8In0.2)2O3. When doping with Ca2+ impurities, to ensure the electrical neutrality of the material, a sufficiently high concentration of positively charged oxygen vacancies VO is additionally formed. A high concentration of VO, in turn, leads to an increase in the probability of the formation of a higher concentration of bivacancies in such samples. Ca2+ creates acceptor levels near the top of the valence band. The transition of electrons from Ca acceptors to the conduction zone is manifested in the luminescence excitation band.

β-(Ga0.8In0.2)2O3:Ca samples synthesized in an oxygen atmosphere always had very high resistance. In such samples, the concentration of intrinsic donors is low. Therefore, only part of the Ca acceptors will be filled with electrons. This is because Ca2+ ions form impurity acceptor levels near the top of the valence band.

β-(Ga0.8In0.2)2O3:Ca samples synthesized in an argon atmosphere had higher electrical conductivity than samples synthesized in an oxygen atmosphere (Table I). This is primarily due to additional donors, which were formed during the synthesis under conditions of low oxygen partial pressure. Since the concentration of Ca impurity is much higher than the concentration of intrinsic acceptor defects, impurity acceptors will dominate in compensating additional donors. That is why the transition of electrons from the Ca acceptor levels to the conduction zone gives the most intense luminescence excitation band. It is located in the region of transparency of the material in front of the fundamental absorption edge at an energy of 4.12 eV (303 nm).

Figure 6 shows the band diagram, the basic energy levels, and possible channels of excitation of host emission in UID and doped with Ca or Zr, β-(Ga0.8In0.2)2O3 ceramics. Previously, a similar band diagram was proposed for β-Ga2O3:Cr and β-Ga2O3:Cr,Mg crystals based on a detailed study of optical-luminescent and electrical properties.16 

FIG. 6.

Energy diagram and possible electronic transitions in the luminescence and photoluminescence excitation spectra.

FIG. 6.

Energy diagram and possible electronic transitions in the luminescence and photoluminescence excitation spectra.

Close modal

Polycrystalline samples of undoped and doped with Ca2+ or Zr4+ ions β-(Ga0.8In0.2)2O3 were obtained by the method of high-temperature solid-phase synthesis. Doped and undoped samples synthesized in an oxygen atmosphere were always highly resistant, while the samples synthesized in an argon atmosphere with a low partial pressure of oxygen had high conductivity. The value of conductivity also depended on the type of doping impurity. The conductivity was the highest in samples doped with zirconium and synthesized in an argon atmosphere, reaching 10−3 Ω−1 cm−1. UID and doped with Ca2+ or Zr4+ ions β-(Ga0.8In0.2)2O3 solid solution samples show enough intense luminescence in the spectral region from 2.0 to 3.5 eV. This luminescence yield and spectral composition are practically independent of doping impurities. Doping and changing the synthesis atmosphere led mainly to a redistribution of intensities between the main elementary luminescence bands, also observed for UID ceramics. Luminescence arises due to the radiative recombination of carriers through DAP and self-localized holes. Donors and acceptors are formed by native defects such as (Gai, VGa) or impurities (Zr4+, Ca2+). Unlike the luminescence spectra, the luminescence excitation spectra change significantly when the synthesis conditions vary or when doping with divalent impurities. The intensive excitation band observed in the transparency region of β-(Ga0.8In0.2)2O3 at 4.44 eV corresponds to electron transitions from acceptor levels created by natural defects such as VGa or VGaVo to the conduction band. Electron transitions from acceptor levels created by Ca2+ impurities are manifested in the form of intense excitation bands located at 4.1 eV.

The work was supported by the Ministry of Education and Science of Ukraine (Project No. 0122U001702).

The authors have no conflicts to disclose.

A. Luchechko: Conceptualization (equal); Formal analysis (equal); Funding acquisition (equal); Methodology (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). V. Vasyltsiv: Conceptualization (equal); Data curation (equal); Investigation (equal); Validation (equal); Writing – original draft (equal); Writing – review & editing (equal). M. Kushlyk: Formal analysis (equal); Investigation (equal); Validation (equal); Visualization (equal); Writing – review & editing (equal). L. Kostyk: Methodology (equal); Validation (equal); Writing – review & editing (equal). D. Slobodzyan: Methodology (equal); Validation (equal); Writing – review & editing (equal).

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

1.
M.
Higashiwaki
,
K.
Sasaki
,
A.
Kuramata
,
T.
Masui
, and
S.
Yamakoshi
,
Phys. Status Solidi A
211
,
21
(
2014
).
2.
S. J.
Pearton
,
J.
Yang
,
P. H.
Cary
,
F.
Ren
,
J.
Kim
,
M. J.
Tadjer
, and
M. A.
Mastro
,
Appl. Phys. Rev.
5
,
011301
(
2018
).
3.
D.
Guo
,
Q.
Guo
,
Z.
Chen
,
Z.
Wu
,
P.
Li
, and
W.
Tang
,
Mater. Today Phys.
11
,
100157
(
2019
).
4.
J.
Díaz
,
I.
López
,
E.
Nogales
,
B.
Méndez
, and
J.
Piqueras
,
J. Nanopart. Res.
13
,
1833
(
2011
).
5.
A.
Luchechko
,
V.
Vasyltsiv
,
L.
Kostyk
, and
B.
Pavlyk
,
Phys. Status Solidi A
216
,
1900444
(
2019
).
6.
R. D.
Shannon
and
C. T.
Prewitt
,
J. Inorg. Nucl. Chem.
30
,
1389
(
1968
).
7.
V. I.
Vasyltsiv
,
Ya. I.
Rym
, and
Ya. M.
Zakharko
,
Phys. Status Solidi B
195
,
653
(
1996
).
8.
R.
Wakabayashi
,
K.
Yoshimatsu
,
M.
Hattori
,
J. S.
Lee
,
O.
Sakata
, and
A.
Ohtomo
,
Cryst. Growth Des.
21
,
2844
(
2021
).
9.
X.
Xia
,
N. S.
Al-Mamun
,
C.
Fares
,
A.
Haque
,
F.
Ren
,
A.
Hassa
,
H.
Von Wenckstern
,
M.
Grundmann
, and
S. J.
Pearton
,
ECS J. Solid State Sci. Technol.
11
,
025006
(
2022
).
10.
11.
J.
Jesenovec
,
B.
Dutton
,
N.
Stone-Weiss
,
A.
Chmielewski
,
M.
Saleh
,
C.
Peterson
,
N.
Alem
,
S.
Krishnamoorthy
, and
J. S.
McCloy
,
J. Appl. Phys.
131
,
155702
(
2022
).
12.
B.
Mazumder
and
J.
Sarker
,
J. Mater. Res.
36
,
52
(
2021
).
13.
H.
Peelaers
,
J. B.
Varley
,
J. S.
Speck
, and
C. G.
Van de Walle
,
Appl. Phys. Lett.
112
,
242101
(
2018
).
14.
M.
Saleh
,
A.
Bhattacharyya
,
J. B.
Varley
,
S.
Swain
,
J.
Jesenovec
,
S.
Krishnamoorthy
, and
K.
Lynn
,
Appl. Phys. Express
12
,
085502
(
2019
).
15.
A.
Luchechko
,
V.
Vasyltsiv
,
L.
Kostyk
,
O.
Tsvetkova
, and
B.
Pavlyk
,
J. Nano-Electron. Phys.
11
,
03035
(
2019
).
16.
V.
Vasyltsiv
,
A.
Luchechko
,
Y.
Zhydachevskyy
,
L.
Kostyk
,
R.
Lys
,
D.
Slobodzyan
,
R.
Jakieła
,
B.
Pavlyk
, and
A.
Suchocki
,
J. Vac. Sci. Technol. A
39
,
033201
(
2021
).
17.
A.
Luchechko
,
V.
Vasyltsiv
,
L.
Kostyk
,
O.
Tsvetkova
, and
B.
Pavlyk
,
ECS J. Solid State Sci. Technol.
9
,
045008
(
2020
).
18.
K.
Irmscher
,
Z.
Galazka
,
M.
Pietsch
,
R.
Uecker
, and
R.
Fornari
,
J. Appl. Phys.
110
,
063720
(
2011
).
19.
Z.
Zhang
,
E.
Farzana
,
A. R.
Arehart
, and
S. A.
Ringel
,
Appl. Phys. Lett.
108
,
052105
(
2016
).
20.
Z.
Galazka
et al,
J. Cryst. Growth
404
,
184
(
2014
).
21.
A.
Luchechko
,
V.
Vasyltsiv
,
L.
Kostyk
, and
O.
Tsvetkova
,
Acta Phys. Pol. A
133
,
811
(
2018
).
22.
B. E.
Kananen
,
L. E.
Halliburton
,
E. M.
Scherrer
,
K. T.
Stevens
,
G. K.
Foundos
,
K. B.
Chang
, and
N. C.
Giles
,
Appl. Phys. Lett.
111
,
072102
(
2017
).
23.
B. E.
Kananen
,
L. E.
Halliburton
,
K. T.
Stevens
,
G. K.
Foundos
, and
N. C.
Giles
,
Appl. Phys. Lett.
110
,
202104
(
2017
).
24.
J.
Jesenovec
,
M. H.
Weber
,
C.
Pansegrau
,
M. D.
McCluskey
,
K. G.
Lynn
, and
J. S.
McCloy
,
J. Appl. Phys.
129
,
245701
(
2021
).
25.
V. I.
Vasil’tsiv
,
Ya. M.
Zakharko
, and
Ya. I.
Rym
,
J. Appl. Spectrosc.
58
,
299
(
1993
).
26.
L.
Binet
and
D.
Gourier
,
J. Phys. Chem. Solids
59
,
1241
(
1998
).
27.
T.
Matsumoto
,
M.
Aoki
,
A.
Kinoshita
, and
T.
Aono
,
Jpn. J. Appl. Phys.
13
,
1578
(
1974
).
28.
V.
Stasiv
et al,
J. Alloys Compd.
982
,
173827
(
2024
).
29.
V. I.
Vasil’tsiv
,
Ya. M.
Zakharko
, and
Ya. I.
Rym
,
Ukr. Fiz. Zh.
34
,
1320
(
1988
).
30.
T.
Onuma
,
S.
Fujioka
,
T.
Yamaguchi
,
M.
Higashiwaki
,
K.
Sasaki
,
T.
Masui
, and
T.
Honda
,
Appl. Phys. Lett.
103
,
041910
(
2013
).
31.
T.
Onuma
,
Y.
Nakata
,
K.
Sasaki
,
T.
Masui
,
T.
Yamaguchi
,
T.
Honda
,
A.
Kuramata
,
S.
Yamakoshi
, and
M.
Higashiwaki
,
J. Appl. Phys.
124
,
075103
(
2018
).
32.
Y. K.
Frodason
,
K. M.
Johansen
,
L.
Vines
, and
J. B.
Varley
,
J. Appl. Phys.
127
,
075701
(
2020
).
33.
Q. D.
Ho
,
Thomas
Frauenheim
, and
Peter
Deák
,
Phys. Rev. B
97
,
115163
(
2018
).
34.
F. A.
Kröger
,
The Chemistry of Imperfect Crystals: Applications of Imperfection Chemistry, Solid State Reactions and Electrochemistry
(
North-Holland
,
Amsterdam
,
1974
).
35.
H.
Peelaers
,
J. L.
Lyons
, and
J. B.
Varley
,
APL Mater.
7
,
022519
(
2019
).
36.
J. M.
Johnson
et al,
Phys. Rev. X
9
,
041027
(
2019
).
37.
S. K.
Swain
,
M. H.
Weber
,
J.
Jesenovec
,
M.
Saleh
,
K. G.
Lynn
, and
J. S.
McCloy
,
Phys. Rev. Appl.
15
,
054010
(
2021
).