Open Access
Add to BibSonomy
Abstract
Effect of oxygen on the physical properties of GaTe is investigated by optical pump–terahertz probe spectroscopy. Oxygen erosion in GaTe leads to significant reductions in lifetime and transient conductivity. Terahertz emission is also evaluated. Terahertz radiation intensity gradually dies away with increasing air exposing time. Such phenomena are attributed to the trapping of photogenerated carriers by oxygen–introduced defects, which can result in the reduction of the carrier density, acceleration of the photocarrier relaxation, and impediment of the terahertz emission.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Gallium telluride (GaTe) belongs to the III–VI group layered metal chalcogenides. The layers are connected by the covalent bonds of Te–Ga–Ga–Te [1,2], and the adjacent layers interact with each other through van der Waals forces. GaTe has a direct band gap of ∼ 1.65 eV regardless of thickness [3]. This is different from the few–layer MoS2 and WS2, whose bands change with the number of layers. Additionally, regardless of the number of layers, GaTe displays broken inversion symmetry [3], leading to strong second–harmonic and third–harmonic generation in layered GaTe [4]. GaTe has attracted growing interest in the field of nanoelectronics and optoelectronics, which has the potential to design phototransistor, photodetector [5,–8] and highly efficient terahertz emitters [9–11]. However, GaTe is very susceptible to oxygen [12], Te vacancies promote the rapid degradation of GaTe in air, resulting in performance changes. Therefore, it is necessary to study the action of oxygen in GaTe. Zhang et al. found that self–driven oxygen intercalation reduced the bandgap of GaTe from 1.75 eV to 1.19 eV [13]. Bondino et al. investigated the transfer of narrow–bandgap GaTex to wide–bandgap Ga2O3 after exposure to air, which is beneficial for applications in electrocatalysis, photocatalysis and gas sensing [14]. So far, the influence of oxidation on GaTe has been studied mainly by Raman, photoluminescence (PL), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and other conventional methods [12,15]. However, the change in the terahertz band of oxidized GaTe is little understood.
Terahertz spectroscopy is an advanced non-destructive test method used to study the physical properties of materials, especially the ultrafast dynamics process two–dimensional semiconductor materials. In this paper, the optical pump–terahertz probe (OPTP) spectra of GaTe under different oxidation periods are studied, the lifetime and transient photoconductivity are analyzed to explain the effect of oxygen in GaTe. We focus on the effects of oxidation on physical properties of GaTe and analyze the reasons. The impact of oxygen adsorption on terahertz emission efficiency in GaTe is further described. Our work provides important support for the application of GaTe as a new generation of optoelectronic and terahertz emission devices.
2. Methods
GaTe single crystals were generated by the Bridgeman method in tubular furnaces. Te and Ga elements with a ratio of 1: 1.2 were used as raw materials. The growth temperature was 850°C and nitrogen acted as a shielding gas. The sample was dissociated into micron–thick sheets according to crystal planes (210) [16]. The dissociation face of the grown GaTe crystal exhibits a metallic luster. An 80 µm thick GaTe single crystal sheet was selected for the experiment as shown in Supporting Information S1. To investigate the role of oxygen in GaTe, the sample was prepared and tested in freshly cleaved, one week and one month oxidation, respectively. The specific synthesis process, details on the oxidation environment of the sample, and the SEM characteristics refer to Supporting Information S1.
The Raman and PL spectra were measured by a Confocal Raman microscopy (Renishaw Co., inVia–Reflex). The wavelength of the excitation laser is 532 nm. A femtosecond amplifier (100 fs, 1 kHz, 800 nm) was used as the laser source of OPTP system. ZnTe crystals were used to generate and detect terahertz radiation, and the excitation wavelength is 800 nm. A 200 µm thick, type–I β–barium borate (BBO) crystal was added to generate second harmonic at 400 nm, which was used to excite the samples. For the terahertz emission measurements, we employed the same OPTP system that blocks the laser in front of the ZnTe emitter, so that the excitation light used to excite the sample becomes the pump light that induces terahertz radiation. So the pump beam wavelength before GaTe crystal was still 400 nm.
3. Experiment
3.1 Raman and PL spetra
The Raman lattice vibration mode of GaTe under different oxidation degrees were measured in the range of 50 to 350 cm-1, results are shown in Fig. 1(a). For freshly cleaved GaTe, there are typical Raman characteristic peaks located at 54 cm-1, 66 cm-1, 115 cm-1, 163 cm-1, 176 cm-1, 207 cm-1 and 268 cm-1, which is consistent with previous studies [3,17]. For GaTe exposed to air for one week, the oxidation peak at 130 cm-1 appears and grows up. For GaTe exposed to air for one month, strong Raman oxygen–chemisorbed peaks at 123 cm-1 and 140 cm-1 merges to one broad hump, and other peaks are too weak to observed. In addition, there are two inconspicuous oxidation peaks in the freshly cleaved sample located at 130 cm-1 and 140 cm-1, demonstrating that GaTe is susceptible to oxidation in the natural environment. Pure GaTe spectrum can be found in Ref. [3]. The PL spectra also exhibit significant oxidation–dependent behaviors as shown in Fig. 1(b). The fresh cleaved sample exhibits a strong PL peak at 1.62 eV, which is near the bandgap of GaTe. This PL peak weakens with increasing degree of oxidation and is almost invisible one month after oxidation (as shown in the insert figure of Fig. 1(b)). This is due to the interaction between oxygen and Te, the band gap reduces to less than 1.5 eV [13–15], and the exciton recombination path is changed. The intensity of PL peak at 1.62 eV reflects the degree of oxidation of GaTe exposed to air after different oxidation times. Figure 1(c)–(d) present the schematic diagrams of the atomic structure for the freshly cleaved, one week, and one month oxidized GaTe, respectively. These graphs were generated by the software of Mercury. It is only for illustrative purposes and does not fully reflect the actual situation of the crystal lattice. Oxygen gradually permeates from the surface to the gaps among layers. As the oxidation process deepens, oxygen even reacts with Ga to form Ga2O3 [13,14]. During this process, Ga2Te3 and GaTexO are formed as intermediate steps, and finally Ga2O3 and Te are formed in the surface of the sample.
Fig. 1. (a)–(b) Raman and PL spectra of GaTe at different oxidation time in air: freshly cleaved (black), one week (red), and one month (blue). The right insert in (b) is the zoom-in PL spectra of GaTe exposed to air for a month. The left insert in (b) is the picture of the sample. (c)-(e) Schematic diagrams of the atomic structures for freshly cleaved, one week, and one month oxidized GaTe, respectively.
3.2 Lifetime of GaTe affected by oxygen
Air–exposing time dependent lifetime of GaTe is measured by OPTP. A schematic illustration of the principle of OPTP spectroscopy used for the GaTe experiment is shown in the Fig. 2(a). The pump fluence is 50 uJ/cm2. Figure 2(b) shows the normalized transient differential transmissivity, -ΔT/T, of GaTe at different exposure times to air. The relaxation curves are fitted by double exponential function:
Fig. 2. (a) A schematic illustration of the principle of OPTP spectroscopy used for the GaTe experiment. (b) Normalized transient differential transmissivity of the GaTe at different exposure times to air. The points are experimental data and the lines are fitting curves.
With the increase of oxidation time, the lifetime of GaTe decreased. The fast and slow decay times shorten by 83.4% and 66.2%, respectively. Compared with other data, the slow relaxation time of the fresh cleaved GaTe has a large error range. However, within the error fluctuation range, the change law and the relative relationship of the curves are not affected. So the results are credible. The fast relaxation process originates from the capture of the photocarriers by the trap states [18,19], which exist at the grain boundaries and defects in GaTe. The slow relaxation process is attributed to the photocarrier recombination [20] and exciton recombination, accompanied by the carrier–phonon scattering [21] and exciton–phonon scattering [22]. More defect states are introduced in the GaTe duo to the oxygen errosion. These defect states can effective capture photo–generated carriers, resulting in the acceleration of the fast relaxation process. Then defect–induced recombination and scattering enhancement effect accelerate the slow relaxation process.
3.3 Transient photoconductivity of GaTe affected by oxygen
In the OPTP spectra (Fig. 2), the time delays at 0 ps, 1 ps, 10 ps, 20 ps, and 40 ps are selected to measure the time–domain spectroscopy (TDS) signals of the samples with and without pump, respectively. Then the transient conductivity of the GaTe under different time delays can be obtained. The GaTe at different oxidation times is investigated. The calculation of the complex refractive index of GaTe is described in Supporting Information S3, and the refractive index is 3. The results are fitted by the Drude–Smith model:
Figure 4 shows the change in transient conductivity of GaTe with different oxidation degrees at 0 ps delay. The experimental data fit well with the Drude–Smith model in the range of 0.5-2 THz. The real part and the absolute value of the imaginary part of the transient conductivity decrease significantly with the increase of oxidation time, which means that the material's ability to conduct current is reduced and the energy dissipation is increased. The charge carrier density, back–scattering time, DC conductivity, and carrier mobility with error bars of GaTe at different time delays and oxidation degrees are summarized in Fig. 5. The decrease of the transient conductivity can be explained by the decrease in carrier concentration as show in Fig. 5(a). With the increase of oxidation time, oxygen atoms gradually intrude into the GaTe layer gap, and even chemically react with Ga atoms, introducing a large number of defects inside the material. More carriers are bound by defects, thus the carrier number reduces. As a result, the transient charge transfer caused by the pump light is reduced, resulting in a decrease in the transient conductivity.
Fig. 4. (a)–(c) Transient photoconductivity of the freshly cleaved GaTe, GaTe exposed to air for one week, and one month at 0 ps delay time. The black hollow squares represent the real part, and the red hollow dots represent the imaginary part. The yellow and blue lines represent the real and imaginary parts of the fitted data, respectively.
Fig. 5. (a)–(d) Charge carrier density, back–scattering time, DC conductivity, and mobility of the GaTe with error bars varies with delay times at different of oxidation times in air, respectively.
The increasing defect states can also enhance the scattering probability among carriers, leading to an increase in the scattering rate and a decreases in the scattering time, as shown in Fig. 5(b). In addition, in freshly cleaved and one week oxidized GaTe, with the increase of time delay, the scattering time shows a trend of increasing first and then decreasing, while the one month oxidized GaTe decreased continuously. The increase in scattering time mainly occurs during the fast relaxation process (Table 1). When the sample is excited by the pump light, the photogenerated carriers accumulate rapidly in a short period of time. In freshly cleaved and one week oxidized GaTe, the carriers likely become more mobile because there are fewer present to annihilate one–another after a brief relaxation period. A best spot is formed between dissipation to prevent overcrowding and limitations to thermal relaxation by these moderate delays. This isn't observed in the most oxidized sample because there are fewer carriers in the photoconductive peak and expedited pathways (nearby traps/lower–energy sites) for thermal relaxation, eliminating the effect. The change of DC conductivity and carrier mobility are easy to understand with Eq. (3). With the deepening of the oxidation process, reducing carrier density and scattering time result in lower DC conductivity and carrier mobility, as shown in Fig. 5(c) and (d). In addition, the DC conductivity decreases with the increase of delay time, which is consistent with the relaxation regularity of OPTP spectra [25,26]. And the change in mobility with increasing delay time is consistent with the scattering time. The details for parameters calculation and the backscattering coefficient c1 are shown in Supporting Information S4. The value of c1 is between -0.69 to -0.86, which means that strong coherent backscattering leads to localization. However, c1 does not exhibit significant oxidation-dependent properties, possibly due to the complex phonon dispersion process.
3.4 GaTe terahertz emission affected by oxygen
GaTe is expected to be a highly efficient terahertz radiation source. Terahertz radiation intensity for semiconductor materials can be described by the following formula [27,28]:
The terahertz electric field arises from two parts, one is the nonlinear polarization (P) caused by optical rectification (OR) effect and the other is the transient photocurrent (J). For GaTe, if the excitation energy is less than the band gap (for example, 800 nm pump excitation), the rectification current from the first term on the right–hand side plays a dominant role in THz radiation. If the photon energy is above the band gap (take 400 nm pump excitation as an example), both the rectification current and the transient photocurrent are involved in terahertz emission. However, compared to transient photocurrent, the terahertz radiation generated by OR is weak and negligible [9]. In addition, the intensity and polarity of terahertz radiation generated by GaTe is affected by pump polarization [9–11]. To explore the terahertz emission intensity of the samples under different oxidation conditions, we measure the maximum terahertz emission spectra by rotating the samples, results display in Fig. 6(a). For more information about the rotation experiment, see Support Information S5. The pump laser wavelength is 400 nm and the pump fluence is 50 uJ/cm2. The corresponding Fourier transform spectra are shown in Fig. 6(b). The central frequency of the frequency–domain spectra is around 0.5 THz, and the bandwidth is close to 2.7 THz, mainly due to the limitation of ZnTe probing.
Fig. 6. (a) Time–domain terahertz emission waveforms for different samples. (b) Corresponding frequency–domain spectra.
It can be seen that the presence of oxygen severely affects the terahertz radiation of GaTe. In Fig. 6(a), the terahertz intensities of GaTe exposed to air for one week and one month decrease by 23.3% and 94.5% with comparing to the freshly GaTe, respectively. The emitted terahertz pulse generates from the surface of GaTe, transmitted through the crystal and radiated outward from the other side. The rectification current derives from polarization, which means that the nonlinear crystals is excited with strong photon energy to produce an ultrafast shift current. Introduction of the oxygen defects breaks the original asymmetric structure and polarization state of GaTe, resulting in the diminishment of the terahertz emission. The transient photocurrent results from the drift of the photogenerated carriers under the action of the surface electric field and the photo–Dember effect, which is caused by the difference in diffusion coefficients for electrons and holes [29,30]. After ultrafast femtosecond excitation, electron charge displacement occurs in GaTe. The charge displacement creates a shift current that emits terahertz radiation. As the air–exposure time increases, the oxidation depth of the GaTe gradually increases and a large number of defects accumulate in the crystal lattice. The electron charge displacement is disturbed by oxygen, and the generation of the shift current is hindered, so that the emission of terahertz radiation from GaTe is weakened.
4. Conclusion
Oxidation effect on the ultrafast dynamics of GaTe is studied by the time–resolved terahertz and terahertz emission spectroscopies. We find that the intrusion of oxygen into GaTe leads to a significant decrease in hot carrier lifetime and transient photoconductivity. Concurrently, with the increase of oxidation time, the carrier density, back–scattering time, DC conductivity, and carrier mobility of GaTe decrease. It can be explained that due to the surface oxidation, a number of defects are introduced into the GaTe lattice. The hot carriers are trapped by those defects, accelerating the decay process of the carriers. Capture of carriers by defects also can efficiently impede the transient charge transfer, so the transient conductivity falls. Terahertz emission intensity also shows significant attenuation. This is because the electron charge displacement is blocked by oxygen, the shift current is reduced, resulting in reduced terahertz radiation from the GaTe. This work provides fundamental understandings for the potential application of GaTe as a candidate of the atomically thin terahertz emission and optoelectronic devices.
Funding
The National Natural Science Foundation of China (11904056, 12274424, 61988102); The Key Research and Development Program of Guangdong Province (2019B090917007, 2022YFA1203500); The Science and Technology Planning Project of Guangdong Province (2019B090909011); The Guangzhou basic and applied basic research Project (202102020053).
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Nos. 61988102, 12274424 and 11904056), the Key Research and Development Program of Guangdong Province (No. 2019B090917007, 2022YFA1203500), the Science and Technology Planning Project of Guangdong Province (No. 2019B090909011), and the Guangzhou Basic and Applied Basic Research Project (No. 202102020053).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. U. Shenoy, U. Gupta, D. Narang, D. Late, U. Waghmare, and C. Rao, “Electronic structure and properties of layered gallium telluride,” Chem. Phys. Lett. 651, 148–154 (2016). [CrossRef]
2. V. Vi, N. Hieu, B. Hoi, N. Binh, and T. Vu, “Modulation of electronic and optical properties of GaTe monolayer by biaxial strain and electric field,” Superlattices Microstruct. 140, 106435 (2020). [CrossRef]
3. J. Susoma, J. Lahtinen, M. Kim, J. Riikonen, and H. Lipsanen, “Crystal quality of two–dimensional gallium telluride and gallium selenide using Raman fingerprint,” AIP Adv. 7(1), 015014 (2017). [CrossRef]
4. J. Susoma, L. Karvonen, A. Säynätjoki, S. Mehravar, R. Norwood, N. Peyghambarian, K. Kieu, H. Lipsanen, and J. Riikonen, “Second and third harmonic generation in few–layer gallium telluride characterized by multiphoton microscopy,” Appl. Phys. Lett. 108(7), 073103 (2016). [CrossRef]
5. P. Hu, J. Zhang, M. Yoon, X. Qiao, X. Zhang, W. Feng, P. Tan, W. Zheng, J. Liu, X. Wang, J. C. Idrobo, D. B. Geohegan, and K. Xiao, “Highly sensitive phototransistors based on two–dimensional GaTe nanosheets with direct bandgap,” Nano Res. 7(5), 694–703 (2014). [CrossRef]
6. Z. Wang, K. Xu, Y. Li, X. Zhan, M. Safdar, Q. Wang, F. Wang, and J. He, “Role of Ga vacancy on a multilayer GaTe phototransistor,” ACS Nano 8(5), 4859–4865 (2014). [CrossRef]
7. F. Liu, H. Shimotani, H. Shang, T. Kanagasekaran, V. Zólyomi, N. Drummond, V. Faĺko, and K. Tanigaki, “High–sensitivity photodetectors based on multilayer GaTe flakes,” ACS Nano 8(1), 752–760 (2014). [CrossRef]
8. M. Karmakar, P. Kumbhakar, T. Singha, C. Sekhar Tiwary, D. Chanda, and P. Datta, “Anomalous indirect carrier relaxation in direct band gap atomically thin gallium telluride,” Phys. Rev. B 107(7), 075429 (2023). [CrossRef]
9. M. Tong, Y. Hu, W. He, X. Yu, S. Hu, X. Cheng, and T. Jiang, “Ultraefficient terahertz emission mediated by shift–current photovoltaic effect in layered gallium telluride,” ACS Nano 15(11), 17565–17572 (2021). [CrossRef]
10. J. Dong, K. Gradwohl, Y. Xu, T. Wang, B. Zhang, B. Xiao, C. Teichert, and W. Jie, “Terahertz emission from layered GaTe crystal due to surface lattice reorganization and in–plane noncubic mobility anisotropy,” Photonics Res. 7(5), 518–525 (2019). [CrossRef]
11. G. Xu, G. Sun, Y. Ding, I. Zotova, K. Mandal, A. Mertiri, G. Pabst, R. Roy, and N. Fernelius, “Investigation of terahertz generation due to unidirectional diffusion of carriers in centrosymmetric GaTe crystals,” IEEE J. Select. Topics Quantum Electron. 17(1), 30–37 (2011). [CrossRef]
12. M. Kotha, T. Murray, D. Tuschel, and S. Gallis, “Study of oxidation and polarization-dependent optical properties of environmentally stable layered GaTe using a novel passivation approach,” Nanomaterials 9(11), 1510 (2019). [CrossRef]
13. R. Zhang, Y. Wei, Y. Kang, M. Pu, X. Li, X. Ma, M. Xu, and X. Luo, “Breaking the cut–off wavelength limit of GaTe through self–driven oxygen intercalation in air,” Adv. Sci. 9(9), 2103429 (2022). [CrossRef]
14. F. Bondino, S. Duman, S. Nappini, G. D’Olimpio, C. Ghica, T. Menteş, F. Mazzola, M. Istrate, M. Jugovac, M. Vorokhta, S. Santoro, B. Gürbulak, A. Locatelli, D. Boukhvalov, and A. Politano, “Improving the efficiency of gallium telluride for photocatalysis, electrocatalysis, and chemical sensing through defects engineering and interfacing with its native oxide,” Adv. Funct. Mater. 32(41), 2205923 (2022). [CrossRef]
15. J. Fonseca, S. Tongay, M. Topsakal, A. Chew, A. Lin, C. Ko, A. Luce, A. Salleo, J. Wu, and O. Dubon, “Bandgap restructuring of the layered semiconductor gallium telluride in air,” Adv. Mater. 28(30), 6465–6470 (2016). [CrossRef]
16. Q. Zhao, T. Wang, Y. Miao, F. Ma, Y. Xie, X. Ma, Y. Gu, J. Li, J. He, B. Chen, S. Xi, L. Xu, H. Zhen, Z. Yin, J. Li, J. Ren, and W. Jie, “Thickness–induced structural phase transformation of layered gallium telluride,” Phys. Chem. Chem. Phys. 18(28), 18719–18726 (2016). [CrossRef]
17. T. Wang, Q. Zhao, Y. Miao, F. Ma, Y. Xie, and W. Jie, “Lattice vibration of layered GaTe single crystals,” Crystals 8(2), 74 (2018). [CrossRef]
18. S. Kar, Y. Su, R. Nair, and A. Sood, “Probing photoexcited carriers in a few–layer MoS2 laminate by time–resolved optical pump terahertz probe spectroscopy,” ACS Nano 9(12), 12004–12010 (2015). [CrossRef]
19. H. Wang, C. Zhang, and F. Rana, “Ultrafast dynamics of defect–assisted electron–hole recombination in monolayer MoS2,” Nano Lett. 15(1), 339–345 (2015). [CrossRef]
20. S. Ghosh, A. Winchester, B. Muchharla, M. Wasala, S. Feng, A. Elias, M. Krishna, T. Harada, C. Chin, K. Dani, S. Kar, M. Terrones, and S. Talapatra, “Ultrafast intrinsic photoresponse and direct evidence of sub–gap states in liquid phase exfoliated MoS2 thin films,” Sci. Rep. 5(1), 11272 (2015). [CrossRef]
21. H. Shi, R. Yan, S. Bertolazzi, J. Brivio, B. Gao, A. Kis, D. Jena, H. G. Xing, and L. Huang, “Exciton dynamics in suspended monolayer and few–layer MoS2 2D Crystals,” ACS Nano 7(2), 1072–1080 (2013). [CrossRef]
22. T. Korn, S. Heydrich, M. Hirmer, J. Schmutzler, and C. Schüller, “Low-temperature photocarrier dynamics in monolayer MoS2,” Appl. Phys. Lett. 99(10), 102109 (2011). [CrossRef]
23. C. Manfredotti, R. Murri, A. Rizzo, L. Vasanelli, and G. Micocci, “Electrical propertlies of GaTe grown by various methods,” Phys. Stat. Sol. (a) 29(2), 475–480 (1975). [CrossRef]
24. L. Huang, Z. Chen, and J. Li, “Effects of strain on the band gap and effective mass in two–dimensional monolayer GaX (X = S, Se, Te),” RSC Adv. 5(8), 5788–5794 (2015). [CrossRef]
25. J. Lu and L. Hongwei, “A critical review on the carrier dynamics in 2D layered materials investigated using THz spectroscopy,” Opt. Commun. 406, 24–35 (2018). [CrossRef]
26. H. Liu, X. He, J. Ren, J. Jiang, Y. Yao, and G. Lu, “Terahertz modulation and ultrafast characteristic of two–dimensional lead halide perovskites,” Nanomaterials 12(20), 3559 (2022). [CrossRef]
27. Z. Fan, M. Xu, Y. Huang, Z. Lei, L. Zheng, Z. Zhang, W. Zhao, Y. Zhou, X. Wang, X. Xu, and Z. Liu, “Terahertz surface emission from MoSe2 at the monolayer limit,” ACS Appl. Mater. Interfaces 12(42), 48161–48169 (2020). [CrossRef]
28. Z. Lei, Y. Huang, W. Du, Z. Fan, J. Chang, H. Wang, Y. Jin, and X. Xu, “Nonlinear optical response on the surface of semiconductor SnS2 probed by terahertz emission spectroscopy,” J. Phys. Chem. C 124(39), 21559–21567 (2020). [CrossRef]
29. M. Johnston, D. Whittaker, A. Corchia, A. Davies, and E. Linfield, “Simulation of terahertz generation at semiconductor surfaces,” Phys. Rev. B 65(16), 165301 (2002). [CrossRef]
30. M. Reid and R. Fedosejevs, “Terahertz emission from (100) InAs surfaces at high excitation fluences,” Appl. Phys. Lett. 86(1), 011906 (2005). [CrossRef]
