Optical and Electronic Energy Band Properties of Nb-Doped β-Ga2O3 Crystals

Abstract

Systemic investigations are performed to comprehend the structural, optical, and electrical characteristics of four niobium (Nb) doped β-Ga₂O₃ crystals (β-Ga₂O₃:Nb) grown by the optical floating zone (OFZ) method. All of the β-Ga₂O₃:Nb crystals revealed monoclinic phases and good crystalline qualities. While the different Nb doped (i.e., 0.0001 mol%, 0.01 mol%, 0.1 mol% and 0.5 mol%) samples exhibited slightly changed bandgap energies E_g (≡ 4.72 eV, 4.73 eV, 4.81 eV, 4.68 eV)—the luminescence features indicated distinctive defect levels—affecting the electronic energy structure significantly. By increasing the Nb doping level from 0.0001 mol% to 0.1 mol%, the Fermi level (E_F) moves closer to the bottom of the conduction band. For the sample with Nb doping 0.5 mol%—no further improvement is noticed in the electronic properties. Finally, the energy band diagrams of four samples are given.

1. Introduction

Among the five commonly identified polymorphs of the gallium oxide (Ga₂O₃)—the β-Ga₂O₃ phase is the most thermodynamically stable (melting point of 1795 °C) exhibiting a monoclinic crystal structure, wider band gap, high breakdown electric field, and many other excellent properties [1,2,3,4]. In recent years, β-Ga₂O₃ has gained considerable interest for the development of Schottky diodes, high breakdown voltage switches, solar- and visible-blind photodetectors, random-access memory gas sensors, and high-power electronic devices [5,6,7,8,9]. However, due to ultra-wide bandgap E_g (4.7–4.9 eV), the electrical conductivity of pure β-Ga₂O₃ is quite poor, which has limited its use in many device engineering projects. Therefore, it is of great interest to improve the electric conductivity of β-Ga₂O₃ by doping. It has been established that Si, Sn, and Ge are the most effective n-type dopants for β-Ga₂O₃ materials [10,11,12]. By first principle calculations, Peelaers and Van de Walle [13] explored the possibilities of using W, Mo, Re, and Nb as n-type dopants in β-Ga₂O₃. Among them, Nb dopant has emerged as the best candidate for n-type doping because the atomic radius of Nb is very close to that of the Ga atom. Moreover, as Nb belongs to group V elements in the periodic table, it can provide twice as many electrons as the most commonly used group IV dopants (i.e., Si, Sn, and Ge) in β-Ga₂O₃. Earlier Zhou et al. [14] reported Nb doping as an effective approach to improve the resistivity and carrier density of β-Ga₂O₃. However, many fundamental properties of β-Ga₂O₃:Nb materials still remained insufficiently unknown. Moreover, there exist no systematic studies on the defect energy levels and electronic energy band structures of β-Ga₂O₃:Nb. In this paper, we report the results of a systematic study on the optical characteristics and electronic energy band structures of Nb-doped β-Ga₂O₃ crystals grown by the optical floating zone (OFZ) method. We strongly feel that the outcomes described here are of vital importance for many optoelectronic device applications.

2. Materials and Methods By using high purity gallium oxide (Ga₂O₃) powder (99.9999%) and niobium pentoxide (Nb₂O₅) powder (99.99%) as the raw materials and a [010] oriented β-Ga₂O₃ single crystal as a seed, we grew Nb doped β-Ga₂O₃ crystals in a Quantum Design IRF01-001-00 infrared image furnace through the OFZ technique. The nominal Nb doping concentrations are about 0.0001 mol%, 0.01 mol%, 0.1 mol%, and 0.5 mol% (labeled as samples S1, S2, S3, and S4), respectively, calculated by a mixing ratio of Nb₂O₅ and Ga₂O₃ raw materials. The OFZ-grown crystals were cut into bulk samples along the (100) crystal plane by a STX-202A diamond wire cutting machine. Each sample size was 5 × 5 × 1 mm³ and its two (100) plane surfaces were polished for optical measurements and analysis. To comprehend the crystal structure of the samples, we used high-resolution x-ray diffraction (HR-XRD) with a wavelength of 1.546 Å (D8 Discover, Advance Bruker, Karlsruhe, Germany). The optical transmission (OT) and photoluminescence excitation (PLE) spectra were recorded by exploiting an ultraviolet/visible/near infrared (UV/Vis/NIR) spectrometer (Zolix Omni-λ 300i spectrometer). The results were used to analyze the bandgap properties. Room temperature photoluminescence (PL) measurements were performed by using a spectrometer (Zolix Omni-λ 750i) with a wavelength of 213 nm (FL-213-Pico laser, Changchun New Industries, CNI., Changchun, China) to explore the luminescent features. We also employed the X-ray photoelectron spectroscopy (XPS) and an ultraviolet photoelectron spectroscopy (UPS) with a monochromatic Al Kα X-ray source having a photon energy of 1486.7 eV (ESCALAB 250XI+, Thermo Fisher Scientific, Waltham, MA, USA) for measuring the electronic band structures of the four samples. 3. Results and Discussion 3.1. Crystal Quality

In Figure 1, we showed the HR-XRD spectra of four samples where the scanning range was set between 10° and 110° with a step of 0.05°. The perusal of Figure 1 revealed four strong diffraction peaks and three weak diffraction peaks assigned as the (400), (600), (800), (1200), and (200), (512), (603) crystal planes, respectively (referring to the Ga₂O₃ XRD Joint Committee on Powder Diffraction Standards (JCPDS) card 04-003-1858). The positions of four main diffraction peaks and their full width at half maxima (FWHM) are reported in Table 1. The results of HR-XRD showed that all four β-Ga₂O₃:Nb crystals with different Nb doping levels exhibited good monoclinic phases and the crystal qualities have not been lowered with the doping. Among the four samples, S4 has minimum values of FWHM for four diffraction peaks of (400), (600), (800), (1200).

The full XPS spectra of four samples are displayed in Figure 2a. To avoid surface contamination, we used an Ar ion beam to clean the sample surface for three minutes. The perusal of full spectrum (as shown in Figure 2a) revealed both Ga and O elements. Due to low Nb doping and the sensitivity limit of the XPS, it is difficult to trace the Nb element in the full spectrum. As shown in Figure 2b, there are two fitting peaks of the O 1s obtained by the peak-differential-imitating analysis. Among them, the part with lower binding energy (O(I)) was attributed to the oxidation (Ga-O-Ga) bond [15], while the part with higher binding energy (B.E.) was related to the oxygen vacancies (O(II)) in the Ga₂O₃ film [6]. The fitting results of Ga 2p 1/2 and Ga 2p 3/2 are shown in Figure 2c. The peak positions of Ga 2p 1/2 are 1118.7 eV (S1), 1118.5 eV (S2), 1118.6 eV (S3), and 1118.9 eV (S4), respectively. For Ga 2p 3/2, the peak positions are 1145.7 eV (S1), 1145.4 eV (S2), 1145.5 eV (S3), and 1145.8 eV (S4), respectively. The XPS spectral analysis results of the four samples are reported in Table 2. It was noted that the ratio of Ga and O in each sample is around 2:3 (i.e., Ga:O = 0.706484 (S1), 0.712328 (S2), 0.709421 (S3), 0.733137 (S4), respectively). The atomic ratio of Ga and O is slightly higher than 2:3, which is possibly due to the presence of O vacancies in Ga₂O₃. It should be noted that the content of Ga in the S4 sample is higher than that in the other samples, which indicates that S4 has more O vacancies to affect the carrier concentration and defect-related luminescence properties.

3.2. Bandgap and Defect-Related Luminescence

The OT measurements were performed to obtain the bandgap energies of the four β-Ga₂O₃:Nb samples and the recorded transmission spectra is shown in Figure 3. The bandgap energies are calculated by using the following expression [16]

α = Β′^h_ν−Eg12

where α represents the absorption coefficient, ℎν denotes the incident photon energy, B′ is related to the valence and conduction bands and E_g signifies the bandgap energy. In Equation (1), if ℎν is zero, the square of absorption coefficient becomes equal to the E_g. By linear extrapolation, the bandgap energies of the four samples are obtained as 4.72 eV (S1), 4.73 eV (S2), 4.81 eV (S3), 4.68 eV (S4), respectively. The absorption shoulder located near 4.5 eV is usually considered as the perturbation caused by Ga³⁺ vacancies (V_Ga³⁺) [17].

The band-to-band transitions and defect-related luminescence were also investigated by using PLE and PL techniques. As shown in Figure 4a, the band-to-band absorptions of the four samples are about 4.7 eV, which coincide with the results of OT measurements. For the S3 sample, the PLE spectrum revealed two peaks with an energy level difference of 0.23 eV, suggesting that this is either caused by the split of an energy level or related to the energy levels of a shallow donor.

To further analyze the defect-related luminescence features of the four Nb doped samples, we used a high power with a laser source of wavelength 213 nm to excite the samples. We used the Gaussian peak fitting method to perform analyses of the PL spectra and reported the best fitting results in Table 3. The defect-related luminescence spectra of β-Ga₂O₃:Nb mainly consist of three parts: the first one is an ultraviolet band with a relatively higher intensity. According to earlier reports [18,19,20], Ga₂O₃ luminescence in the ultraviolet band is independent of the impurity, which is related to the combination of free electrons and self-trapping holes (STHs). Another one is in the blue light region, which is mainly caused by the recombination of the deep-level donor-acceptor-pairs. These deep-level donors are attributed to the material’s intrinsic defects, such as the oxygen vacancies (V_O), gallium interstitials (Ga_i), and the acceptors are gallium vacancies (V_Ga), gallium-oxygen vacancy pairs complex (V_O-V_Ga), respectively [19,21,22]. It can be noted from Table 3 that the peak positions of ultraviolet and blue lights in S1 and S2 samples are almost the same, while the peaks of S3 showed blue shift and the peaks of S4 displayed red shift. The last part is red light near 700 nm with relatively weak intensity. In references [23,24], the red band was attributed to the nitrogen impurities but the study in references [25] suggested that the oxygen vacancies could participate in the PL as a hole trap, which give rise to emission between 1.6 and 2.0 eV. Since there was no N element in our samples from the XPS results, the red band might originate from holes trapped at the intrinsic defects of O vacancies at trigonal sites [25]. Unlike several reports, we noticed no green bands in the PL spectra of our samples [18,23,24,26]. We also calculated the intensity ratio of the three peaks in our samples, which are 1:0.8:0.41 (S1), 1:0.63:0.33 (S2), 1:0.29:0.17 (S3), and 1:1.15:0.102:0.73 (S4), respectively. It was noted that the intensity ratio of the blue light and red light decreased with the increase of Nb doping for samples S1–S3, for sample S4; however, the intensity ratio of red light increased significantly, which we attribute it to the interstitial oxygen.

3.3. Electronic Energy Band Structures

In order to determine the E_F and work function, the UPS study was performed on the four β-Ga₂O₃:Nb samples. By calibrating the Fermi level (E_F) of sample and testing stage in the same level, we measured the ultraviolet photoelectron spectra of four samples and the results were shown in Figure 5. Using the linear extrapolation method, we calculated the distances between the E_F and the top of valance band (i.e., the valance band minimum, VBM) for four samples, which were 4.05 eV, 4.35 eV, 4.44 eV, and 4.29 eV, respectively. The cutoff edges (E_cutoff) of the four samples were 16.24 eV, 16.23 eV, 16.24 eV, 16.26 eV, respectively, as shown in Figure 5. The work functions are calculated by the following expression [27]

_Wf=hυ−_Ecutoff

where W_f is the work function, represented the ionization energy of the electrons from E_F to the vacuum level (E_vacuum); andhυis the photon energy of light source (hυ =21.22 eV) [28,29]. The calculated W_f values of the four samples by expression (2) were 4.98 eV (S1), 4.99 eV (S2), 4.98 eV (S3), and 4.96 eV (S4), respectively. Obviously, the W_f of four samples varied slightly with the increase of Nb contents. The band diagrams of the four samples were drawn in Figure 6. It was noted that the distance between the Fermi level and the top of valance band first decreases and then increases with the increase of Nb dopants. The energy difference between conduction band minimum (CBM) and E_F (ΔE_C) for samples S1, S2, S3, and S4 were 0.67 eV, 0.38 eV, 0.37 eV, and 0.39 eV, respectively. The sample S3 has the highest E_F and the lowest ΔE_C.

4. Conclusions In summary, comprehensive investigations of the structural, optical, and electronic traits are reported on a series of Nb doped β-Ga₂O₃ crystals by using XRD, OT, PLE, PL, XPS, and UPS techniques. Four β-Ga₂O₃:Nb crystals grown by the OFZ method exhibited good monoclinic structure. While the Nb doping slightly affected the bandgap and work function characteristics of the four samples, it certainly played an important role influencing the defect luminescence and energy band structures. The PL spectra have revealed Nb doping inhibited the green light luminescence. With the increase of Nb doping, the distance between Fermi level and the bottom of the conduction band first decrease and then remain almost unchanged. We strongly feel that the structural, optical, and electronic features reported here are quite useful to help design Nb doped β-Ga₂O₃ optoelectronic devices.

Figure 1. High-resolution x-ray diffraction (HR-XRD) 2θ-ω patterns for the Nb doped β-Ga2O3.

Figure 2.(a) The wide-scanning X-ray photoelectron spectroscopy (XPS) spectra of samples S1, S2, S3, S4. (b) O 1s fitting results. (c) Ga 2p fitting results.

Figure 3. The optical transmission spectra of four samples of Nb-doped β-Ga2O3 at room temperature, and the bandgap of the four samples calculated results.

Figure 4.(a) Room Temperature photoluminescence excitation (PLE) and photoluminescence (PL) spectra of samples S1, S2, S3, and S4. (b) Deconvoluted PL spectra of samples S1, S2, S3, and S4.

Figure 5. The ultraviolet photoelectron spectroscopy (UPS) spectra of S1, S2, S3, S4 samples.

Figure 6. Schematics of energy level structure for Nb doped β-Ga2O3 samples.

Samples	S1		S2		S3		S4
Crystal faces	2θ	FWHM	2θ	FWHM	2θ	FWHM	2θ	FWHM
(400)	30.12	0.110	30.12	0.108	30.20	0.094	30.14	0.072
(600)	45.89	0.130	45.9	0.139	45.99	0.174	45.85	0.095
(800)	62.58	0.190	62.57	0.158	62.68	0.153	62.57	0.09
(1200)	102.29	0.281	102.26	0.203	102.49	0.299	102.32	0.21

Sample	Ga 2p 1/2		Ga 2p 3/2		Ga	O 1s		O
	B.E. (eV)	Area	B.E. (eV)	Area	Ratio%	B.E. (eV)	Area	Ratio%
S1	1145.7	18250	1118.7	36435	41.4	530.7	77453	58.6
S2	1145.4	23873	1118.5	47134	41.6	530.48	99503	58.4
S3	1145.5	22035	1118.6	43616	41.5	530.52	92349	58.5
S4	1145.8	23697	1118.9	46686	42.3	530.86	96144	57.7

Simples	S1			S2			S3			S4
Peaks (eV)	3.25	2.69	1.77	3.23	2.78	1.77	3.04	2.62	1.67	3.41e	3.10	1.90	1.73
FWHM (nm)	80.8	14	183	80.6	124	174	93	110	223	60	85	70	142
Ratio of Area	1:0.8:0.41			1:0.63:0.33			1:0.29:0.17			1:1.15:0.102:0.73

Author Contributions

Conceptualization, X.L., W.N., C.X., L.W., and Z.F.; data curation, X.L. and W.N.; formal analysis, X.L. and W.N.; funding acquisition, L.W., Q.S., and C.X.; investigation, W.N. and X.C.; project administration, L.W.; resources, H.C., C.X., L.W., and Z.F.; supervision, L.W.; writing-original draft, X.L.; writing-review and editing, L.W., C.X., D.N.T., and Z.F. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China [51802327, 51972319]; and the Science and Technology Commission of Shanghai Municipality (No. 19520744400) and the Guangxi Natural Science Foundation (No. 2018GXNSFAA138127).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data available in a publicly accessible repository.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (51802327, 51972319), the Science and Technology Commission of Shanghai Municipality (No. 19520744400), and the Guangxi Natural Science Foundation, China (No. 2018GXNSFAA138127).

Conflicts of Interest

The authors declare no conflict of interest.

References

1. Kumar, S.S.; Rubio, E.J.; Noor-A-Alam, M.; Martinez, G.; Manandhar, S.; Shutthanandan, V.; Thevuthasan, S.; Ramana, C.V. Structure, Morphology, and Optical Properties of Amorphous and Nanocrystalline Gallium Oxide Thin Films. J. Phys. Chem. C 2013, 117, 4194-4200.

2. Zhang, F.B.; Saito, K.; Tanaka, T.; Nishio, M.; Guo, Q.X. Structural and optical properties of Ga2O3 films on sapphire substrates by pulsed laser deposition. J. Cryst. Growth 2014, 387, 96-100.

3. Ueda, N.; Hosono, H.; Waseda, R.; Kawazoe, H. Synthesis and control of conductivity of ultraviolet transmitting β-Ga2O3 single crystals. Appl. Phys. Lett. 1997, 70, 3561.

4. Yoo, J.-H.; Rafique, S.; Lange, A.; Zhao, H.; Elhadj, S. Lifetime laser damage performance of β-Ga2O3 for high power applications. APL Mater. 2018, 6, 36105.

5. Rafique, S.; Han, L.; Tadjer, M.J.; Freitas, J.A.; Mahadik, N.A.; Zhao, H. Homoepitaxial growth of β-Ga2O3 thin films by low pressure chemical vapor deposition. Appl. Phys. Lett. 2016, 108, 182105.

6. Guo, D.Y.; Wu, Z.P.; An, Y.H.; Guo, X.C.; Chu, X.L.; Sun, C.L.; Li, L.H.; Li, P.G.; Tang, W.H. Oxygen vacancy tuned Ohmic-Schottky conversion for enhanced performance in β-Ga2O3 solar-blind ultraviolet photodetectors. Appl. Phys. Lett. 2014, 105, 023507.

7. Xie, C.; Lu, X.; Liang, Y.; Chen, H.; Wang, L.; Wu, C.; Wu, D.; Yang, W.; Luo, L. Patterned growth of β-Ga2O3 thin films for solar-blind deep-ultraviolet photodetectors array and optical imaging application. J. Mater. Sci. Technol. 2021, 72, 189-196.

8. Wu, J.; Li, C.; Rong, X.; Cao, P.; Han, S.; Zeng, Y.; Liu, W.; Zhu, D.; Lu, Y. Temperature-Dependent Crystallization of Ga2O3 for Ultraviolet Photodetectors. J. Electron. Mater. 2020, 49, 4581-4588.

9. Gan, K.-J.; Liu, P.-T.; Ruan, D.-B.; Hsu, C.-C.; Chiu, Y.-C.; Sze, S.M. Effect of Annealing Treatment on Performance of Ga2O3 Conductive-Bridging Random-Access Memory. J. Electron. Mater. 2020, 49, 6817-6822.

10. Víllora, E.G.; Shimamura, K.; Yoshikawa, Y.; Ujiie, T.; Aoki, K. Electrical conductivity and carrier concentration control in β-Ga2O3 by Si doping. Appl. Phys. Lett. 2008, 92, 202120.

11. Siah, S.C.; Brandt, R.E.; Lim, K.; Schelhas, L.T.; Jaramillo, R.; Heinemann, M.D.; Chua, D.; Wright, J.; Perkins, J.D.; Segre, C.U.; et al. Dopant activation in Sn-doped Ga2O3 investigated by X-ray absorption spectroscopy. Appl. Phys. Lett. 2015, 107, 252103.

12. Ahmadi, E.; Koksaldi, O.S.; Kaun, S.W.; Oshima, Y.; Short, D.B.; Mishra, U.K.; Speck, J.S. Ge doping of β-Ga2O3 films grown by plasma-assisted molecular beam epitaxy. Appl. Phys. Express. 2017, 10, 41102.

13. Peelaers, H.; Van de Walle, C.G. Doping of Ga2O3 with transition metals. Phys. Rev. B 2016, 94, 195203.

14. Zhou, W.; Xia, C.; Sai, Q.; Zhang, H. Controlling n-type conductivity of β-Ga2O3 by Nb doping. Appl. Phys. Lett. 2017, 111, 242103.

15. Hollinger, G.; Skheyta-Kabbani, R.; Gendry, M. Oxides on GaAs and InAs surfaces: An x-ray-photoelectron-spectroscopy study of reference compounds and thin oxide layers. Phys. Rev. B 1994, 49, 11159-11167.

16. Ghose, S.; Rahman, M.S.; Rojas-Ramirez, J.S.; Caro, M.; Droopad, R.; Arias, A.; Nedev, N. Structural and optical properties of β-Ga2O3 thin films grown by plasma-assisted molecular beam epitaxy. J. Vac. Sci. Technol. B 2016, 34, 02L109.

17. Víllora, E.G.; Yamaga, M.; Inoue, T.; Yabasi, S.; Masui, Y.; Sugawara, T.; Fukuda, T. Optical Spectroscopy Study on β-Ga2O3. Jpn. J. Appl. Phys. 2002, 41, L622-L625.

18. Polyakov, A.Y.; Smirnov, N.B.; Shchemerov, I.V.; Pearton, S.J.; Ren, F.; Chernykh, A.V.; Lagov, P.B.; Kulevoy, T.V. Hole traps and persistent photocapacitance in proton irradiated β-Ga2O3 films doped with Si. Apl. Mater. 2018, 6, 096102.

19. Binet, L.; Gourier, D. Origin of the blue luminescence of β-Ga2O3. J. Phys. Chem. Solids 1998, 59, 1241-1249.

20. Deák, P.; Duy Ho, Q.; Seemann, F.; Aradi, B.; Lorke, M.; Frauenheim, T. Choosing the correct hybrid for defect calculations: A case study on intrinsic carrier trapping in β-Ga2O3. Phys. Rev. B 2017, 95, 75208.

21. Kellendonk, F.; Harwig, T. Some observations on the photoluminescence of doped β-galliumsesquioxide. J. Solid State Chem. 1978, 24, 255-263.

22. Onuma, T.; Fujioka, S.; Yamaguchi, T.; Higashiwaki, M.; Sasaki, K.; Masui, T.; Honda, T. Correlation between blue luminescence intensity and resistivity in β-Ga2O3 single crystals. Appl. Phys. Lett. 2013, 103, 41910.

23. Song, Y.P.; Zhang, H.Z.; Lin, C.; Zhu, Y.W.; Li, G.H.; Yang, F.H.; Yu, D.P. Luminescence emission originating from nitrogen doping of β-Ga2O3 nanowires. Phys. Rev. B 2004, 69, 75304.

24. Zhang, T.; Lin, J.; Zhang, X.; Huang, Y.; Xu, X.; Xue, Y.; Zou, J.; Tang, C. Single-crystalline spherical β-Ga2O3 particles: Synthesis, N-doping and photoluminescence properties. J. Lumin. 2013, 140, 30-37.

25. Ho, Q.D.; Frauenheim, T.; Deák, P. Origin of photoluminescence in β-Ga2O3. Phys. Rev. B 2018, 97, 115163.

26. Víllora, E.G.a.; Atou, T.; Sekiguchi, T.; Sugawara, T.; Kikuchi, M.; Fukuda, T. Cathodoluminescence of undoped β-Ga2O3 single crystals. Solid State Commun. 2001, 120, 455-458.

27. Kahn, A. Fermi level, work function and vacuum level. Mater. Horiz. 2016, 3, 7-10.

28. Liu, P.; Liu, J.; Zhang, B.; Zong, W.; Xu, S.; Liu, Y.; Cao, S. Enhanced electroluminescent performance by doping organic conjugated ionic compound into graphene oxide hole-injecting layer. J. Mater. Sci. 2019, 54, 12688-12697.

29. Klein, A.; Korber, C.; Wachau, A.; Sauberlich, F.; Gassenbauer, Y.; Harvey, S.P.; Proffit, D.E.; Mason, T.O. Transparent Conducting Oxides for Photovoltaics: Manipulation of Fermi Level, Work Function and Energy Band Alignment. Materials 2010, 3, 4892-4914.