Photoresponses of InSnGaO and InGaZnO thin-film transistors

Abstract

The photoresponses of thin-film transistors (TFTs) using indium-based oxide semiconductors have been studied. The devices were fabricated using amorphous InSnGaO (ITGO) or InGaZnO (IGZO) as the active semiconducting layer. ITGO and IGZO TFTs showed typical electrical characteristics including high on/off ratios and low off currents. Both devices induced photocurrents upon exposure to ultraviolet light due to their wide band gaps. However, the recovery time of IGZO TFTs was almost 1 h due to the slow recombination of trapped charges in the oxide semiconductors. In contrast, the recovery time of ITGO TFTs was significantly reduced compared to that of IGZO TFTs. We found that the origin of the shorter recovery time of ITGO TFTs was the low electron binding energy of indium, which was obtained by replacing zinc with tin and by increasing the composition ratio of indium. This method may be a useful way to fabricate high-speed optoelectronics based on oxide semiconductors.

---

Introduction

Indium-based oxide semiconductors are considered as an alternative material to amorphous silicon for a wide range of electronic devices, such as thin-film transistors (TFTs) and memory.^1–3 TFTs, which use oxide semiconductors, show high field-effect mobilities (μ_eff), high on/off ratios, and low off currents that are comparable to those of amorphous silicon.^4,5 Moreover, oxide semiconductors have wide band gaps, which are appropriate for the fabrication of transparent electronics.⁶ Recently, transparent TFTs using indium-based oxide semiconductors, such as InGaZnO (IGZO), were successfully demonstrated as switching devices that can be used to run active-matrix liquid crystal displays (AMLCDs) and organic light-emitting diodes (AMOLEDs).^7,8 There are many ways to fabricate devices based on oxide semiconductors; these range from high-temperature vacuum processes to low-temperature solution processes.^9,10 Many research groups have successfully fabricated oxide semiconductor devices on flexible plastic substrates via solution processing.^11–13 Therefore, indium-based oxide semiconductors are the most attractive materials in both research and industrial fields.¹⁴

Similarly to TFTs and memory, photosensors are also devices that can be fabricated using indium-based oxide semiconductors to produce next-generation transparent and flexible optoelectronics. Phototransistors, which can be modulated by high energy photons, have already been reported by several research groups.^15–17 IGZO phototransistors have shown good photoresponsivity in the ultraviolet (UV) light region due to their wide band gaps; however, these devices do not respond to low-energy visible or infrared (IR) light.¹⁵ Moreover, IGZO phototransistors that induce a high photocurrent when illuminated by low-energy photons (e.g., visible and IR light) have been reported. These devices are made by doping metal nanoparticles, quantum dots, and polymers on the IGZO surface.^18–21 Due to these previous successful demonstrations, IGZO is considered to be a good candidate for fabricating photosensors that can respond to a wide range of light. However, although the photoresponses of IGZO phototransistors have successfully been demonstrated, the recovery time (i.e., the time from the on state to the off state) of IGZO phototransistors is too long to be considered useful for optoelectronic devices. It is well known that the slow recombination of trapped charges in oxide semiconductors leads to the slow recovery time of IGZO phototransistors.²² Additionally, the presence of unexpected trapped charges can be reduced by controlling the composition and ratio of the oxide semiconductors.

Therefore, we have increased the ratio of indium and replaced zinc (Zn) with tin (Sn) in IGZO. We fabricated InSnGaO (ITGO) phototransistors to improve the recovery time and photoresponsivity of the device. IGZO phototransistors were also fabricated for comparison. In this way, we were able to improve the photoresponsivity and reduce the recovery time. Also, X-ray photoelectron spectroscopy (XPS) was used to find the origin of the observed improvement. These results, along with the origin of the improvement, suggest a useful way to improve the photoresponsivity and reduce the recovery time of phototransistors based on oxide semiconductors.

Experiments

A 10 nm-thick ITGO film was deposited on a SiO₂ (100 nm)/Si wafer by a radio-frequency sputtering system using a target of In : Ga : Sn (7 : 1 : 2 mol ratio).²³ A vacuum of 5 mTorr was maintained during the deposition, and the film was post annealed at 350 °C for 1 h under ambient conditions. The thickness of the ITGO film was measured and confirmed by an ellipsometry measurement. Typical photolithography and etching processes were used to pattern the source and drain electrodes (100 nm-thick aluminum). A photoresist and buffered oxide etchant were used to define the active channel area of the device. The ITGO TFTs have a channel length of 25 μm and a width of 200 μm. The same procedures were used to fabricate the IGZO TFTs, with the exception that a single target of In₂O₃ : Ga₂O₃ : ZnO (1 : 1 : 1 mol ratio) was used. The device properties were characterized using a semiconducting parameter analyzer (HP 4145B), and the photocurrent was measured upon exposure to various light sources (lasers with wavelengths of 405, 532, and 635). The compositions and ratios of the oxide semiconductors were investigated with XPS to determine the origin of the improvement.

Results and discussion

Fig. 1 shows the electrical characteristics of the IGZO and ITGO TFTs. Fig. 1(a) shows the transfer curve of the IGZO TFT, which was measured with a drain voltage (V_d) of 2 V. The on/off ratio was ∼10⁻⁶ and the off current was 9 × 10⁻¹² A; these are typical values for an oxide semiconductor. The μ_eff and subthreshold swing (SS) were calculated as 0.26 cm² V⁻¹ s⁻¹ and 6.96 V dec⁻¹, respectively. The inset shows the output curves of the IGZO TFT, where the saturation properties of the current can be clearly observed. Fig. 1(b) shows the transfer curve of the ITGO TFT (V_d = 2 V). This curve is also typical of an oxide semiconductor. The saturations of the output curves were also observed, as shown in the inset. However, the current between the source and drain of the ITGO TFTs, with a gate voltage of 30 V, increased by almost 150 times compared to that of the IGZO TFTs. Moreover, the μ_eff was increased to 1.51 cm² V⁻¹ s⁻¹ and SS was 2.21 V dec⁻¹. The increase in the current and mobility indicate that the conductivity of the ITGO film is higher than that of the IGZO film due to the higher carrier concentration. The carrier concentration of the ITGO film (30 nm) was measured as 5.15 × 10¹⁶ cm⁻³, while that of the IGZO film (30 nm) was measured as 1.84 × 10¹⁵ cm⁻³. Although the higher carrier concentration increased the off current, we expect a reduced number of trapped charges due to the high conductivity in the ITGO TFTs.

Fig. 1 Typical transfer and output curves of (a) IGZO and (b) ITGO TFTs. While measuring the transfer curves, the drain voltage was set to 2 V.

Fig. 2 shows the photoresponses of IGZO and ITGO TFTs. Fig. 2(a) shows the n-type transfer curves of the IGZO TFT with (dashed line) and without (solid line) exposure to a 405 nm wavelength of light. The light was illuminated on the channel area with a laser diode. An increase in the photocurrent was observed; this occurred because the energy of the photon (λ = 405 nm) was high enough to excite electrons from the valence band to the conduction band (band gap = ∼3 eV). The current between the drain and source electrodes (I_DS) at a gate voltage of 30 V was increased by 92.9% after exposure to the light. The inset shows the schematic of light exposure on the active channel area of the device. The thickness of the IGZO film was 10 nm, which was used as the active channel material. Fig. 2(b) shows the n-type transfer curves of the ITGO TFT with (dashed line) and without (solid line) exposure to 405 nm light. The photocurrent was also observed. The increase of I_DS at a gate voltage of 30 V was 17.8%, which is smaller than the increase that was observed for the IGZO TFT. Furthermore, it is observed that the photocurrent was even induced at a negative gate voltage, which was not observed for the IGZO TFT. The presence of a photocurrent at a negative gate voltage indicates that ITGO TFTs can be modulated with photons to function as phototransistors. The photoresponses under various photon sources are summarized in the Fig. 2(c). Since both IGZO and ITGO are wide-band gap semiconductors, most of the photoresponses occurred with high-energy photons (e.g., light with a wavelength of 405 nm). The responsivity and external quantum efficiency (EQE) of the IGZO TFT were 2.6 A W⁻¹ and 7.97 (V_d = 2 V, V_g = 6 V), respectively, when a 405 nm wavelength light source was illuminated on the device. The device shows a typical performance of IGZO phototransistors with a UV-light.^15,24 Meanwhile, the photoresponse of the ITGO phototransistor was increased dramatically. The responsivity and EQE were 4.7 × 10⁴ A W⁻¹ and 1.45 × 10⁵ (V_d = 2 V, V_g = −5 V), respectively, upon exposure to the 405 nm light source. The high values of responsivity and EQE are due to the high carrier concentration and mobility of ITGO TFT. Since both IGZO and ITGO are wide-band gap semiconductors, the responsivity and EQE were negligible with lower energy photons, such as 532 and 635 nm light sources, as shown in Fig. 2(c) and S1.† The transmittances of ITGO (10 nm) and IGZO (10 nm) are shown in Fig. 2(d). Both ITGO and IGZO films showed similar transmittance properties at the wavelength larger than 450 nm. However, the transmittance of IGZO film were slightly smaller than the transmittance of ITGO film at the high-energy photon region (<450 nm).

Fig. 2 Transfer curves of (a) IGZO and (b) ITGO TFTs with/without exposure to 405 nm wavelength light (1 mW). Insets show the device structures. (c) Responsivity (top) and EQE (bottom) of the ITGO and IGZO TFTs upon the exposure to 405, 532, and 635 nm wavelengths of light. (d) Transmittances of ITGO (10 nm) and IGZO (10 nm) films.

To see the recovery characteristics from the on state to the off state, we tracked changes in the transfer curves as a function of the time after the light source was turned off. Fig. 3(a) shows the change of the transfer curve of the IGZO TFT, which was measured at the dark state, with the illumination of a light source, and waiting 30 and 60 min after turning the light source off. Due to the slow recombination of trapped charges in oxide semiconductors, it takes almost 1 h to recover to the original transfer curve. Alternatively, the recovery time was reduced to 10 min for the ITGO TFT, as shown in Fig. 3(b). The recombination of trapped charges in the ITGO film appears to be faster compared to the IGZO film; this is due to the higher conductivity and carrier concentration of the ITGO film, as suggested in Fig. 1(b). To find the origin of the shorter recovery time of ITGO TFTs (compared to IGZO TFTs), the composition and ratio of both 10 nm-thick ITGO and IGZO films were investigated using XPS (Fig. 4). In Fig. 4, one can see that the In 3d spectrum of the ITGO film has been shifted to a lower binding energy (as compared to the spectrum of the IGZO film). Ga 2p and O 1s peaks of ITGO film have been also shifted to a lower binding energy as shown in Fig. S2.† Also, the amount of indium in the ITGO film was larger than in the IGZO film. The composition ratio of indium in the ITGO film was measured to be 22.13%, while that of the IGZO film was only 3.8% (Tables 1 and 2). These measurements indicate that the ITGO film has more indium and that the electrons are bound at a lower energy compared to the IGZO film. This provides evidence of the higher conductivity and carrier concentration of the ITGO film. Therefore, by increasing the amount of indium and replacing Zn with Sn, the switching speed of phototransistors based on oxide semiconductors can be improved.

Fig. 3 Changes in the transfer curves according to the time after turning off the light sources: (a) IGZO and (b) ITGO TFTs.

Fig. 4 In 3d spectra of 10 nm-thick ITGO and IGZO films, as measured by XPS. The binding energies of In 3d_5/2 are 444.5 eV (ITGO) and 445.5 eV (IGZO), and the binding energies of In 3d_3/2 are 452 eV (ITGO) and 453 eV (IGZO).

Table 1 Composition and ratio of the IGZO film

IGZO	At%
In 3d	3.8
Ga 2p	14.1
Zn 2p	7.52
O 1s	74.58

---

Table 2 Composition and ratio of the ITGO film

ITGO	At%
In 3d	22.13
Sn 3d	1.44
Ga 2p	11.62
O 1s	64.81

---

Conclusion

ITGO and IGZO TFTs were fabricated to study the photoresponses of indium-based oxide semiconductors. Both ITGO and IGZO TFTs showed an induced photocurrent upon exposure to a 405 nm light source due to their wide band gaps. Alternatively, a photocurrent was not induced upon exposure to 532 and 635 nm light sources. Meanwhile, ITGO TFTs induced much higher photocurrents, and photocurrents could even be induced at a negative gate voltage. This phenomenon was not observed in the IGZO TFT. These results indicate that ITGO TFTs can be modulated by photons to function as phototransistors. The responsivity and EQE of the ITGO TFTs, upon exposure to a 405 nm light source, were 2.2 × 10⁴ A W⁻¹ and 6.8 × 10⁴, respectively. Moreover, we observed that the recovery times of ITGO TFTs were shorter than those of IGZO TFTs. We found that the cause for the shorter recovery times in ITGO TFTs was the increased amount of indium and the lower electron binding energy, which lead to the increased carrier concentration and conductivity of the film. Therefore, although the recovery time of ITGO phototransistors was not as fast as silicon-based devices, this method provides a useful way to fabricate high-speed optoelectronics based on oxide semiconductors.

Acknowledgements

This work was supported by the Center for Advanced Soft-Electronics funded by the Ministry of Science, ICT, and Future Planning as a Global Frontier Project (CASE-2014M3A6A5060946). It was also partially supported by a research project of the National Research Foundation of Korea (NRF-2013R1A1A1A05007934).

References

1. K. Takechi, M. Nakata, K. Azuma, H. Yamaguchi and S. Kaneko, IEEE Trans. Electron Devices, 2009, 56, 2027.
2. E. Chong, K. C. Jo and S. Y. Lee, Appl. Phys. Lett., 2010, 96, 152102.
3. Y.-S. Fan, P.-T. Liu, L.-F. Teng and C.-H. Hsu, Appl. Phys. Lett., 2012, 101, 052901.
4. W.-T. Chen, S.-Y. Lo, S.-C. Kao, H.-W. Zan, C.-C. Tsai, J.-H. Lin, C.-H. Fang and C.-C. Lee, IEEE Electron Device Lett., 2011, 32, 1552.
5. L.-Y. Su, H.-Y. Lin, H.-K. Lin, S.-L. Wang, L.-H. Peng and J. J. Huang, IEEE Electron Device Lett., 2011, 32, 1245.
6. J. C. Park, S. Kim, S. Kim, C. Kim, I. Song, Y. Park, U.-I. Jung, D. H. Kim and J.-S. Lee, Adv. Mater., 2010, 22, 5512.
7. N. Gong, C. Park, J. Lee, I. Jeong, H. Han, J. Hwang, J. Park, K. Park, H. Jeong, Y. Ha and Y. Hwang, SID Int. Symp. Dig. Tech. Pap., 2012, 43, 784.
8. Y. G. Mo, M. Kim, C. K. Kang, J. H. Jeong, Y. S. Park, C. G. Choi, H. D. Kim and S. S. Kim, J. Soc. Inf. Disp., 2011, 19, 16.
9. J. S. Jung, K. S. Son, K.-H. Lee, J. S. Park, T. S. Kim, J.-Y. Kwon, K.-B. Chung, J.-S. Park, B. Koo and S. Lee, Appl. Phys. Lett., 2010, 96, 193506.
10. S.-Y. Han, G. S. Herman and C. Chang, J. Am. Chem. Soc., 2011, 133, 5166.
11. Y. S. Rim, W. H. Jeong, D. L. Kim, H. S. Lim, K. M. Kim and H. J. Kim, J. Mater. Chem., 2012, 22, 12491.
12. Y.-H. Kim, J.-S. Heo, T.-H. Kim, S. Park, M.-H. Yoon, J. Kim, M. S. Oh, G.-R. Yi, Y.-Y. Noh and S. K. Park, Nature, 2012, 489, 128.
13. J.-S. Seo, J.-H. Jeon, Y. H. Hwang, H. Park, M. Ryu, S.-H. K. Park and B.-S. Bae, Sci. Rep., 2013, 3, 2085.
14. K. Nomura, H. Ohta, K. Ueda, T. Kamiya, M. Hirano and H. Hosono, Science, 2003, 300, 1269.
15. T. H. Chang, C. J. Chiu, W. Y. Weng, S. J. Chang, T. Y. Tsai and Z. D. Huang, Appl. Phys. Lett., 2012, 101, 261112.
16. S.-E. Ahn, I. Song, S. Jeon, Y. W. Jeon, Y. Kim, C. Kim, B. Ryu, J.-H. Lee, A. Nathan, S. Lee, G. T. Kim and U.-I. Chung, Adv. Mater., 2012, 24, 2631.
17. T. H. Chang, C. J. Chiu, S. J. Chang, T. Y. Tsai, T. H. Yang, Z. D. Huang and W. Y. Weng, Appl. Phys. Lett., 2013, 102, 221104.
18. S. J. Park, S. M. Lee, K. H. Lee, J.-S. Park and S. J. Kang, J. Vac. Sci. Technol., A, 2015, 33, 021101.
19. S. M. Lee, S. J. Park, K. H. Lee, J.-S. Park, S. Park, Y. Yi and S. J. Kang, Appl. Phys. Lett., 2015, 106, 031112.
20. S. W. Shin, K.-H. Lee, J.-S. Park and S. J. Kang, ACS Appl. Mater. Interfaces, 2015, 7, 19666.
21. H.-W. Zan, W.-T. Chen, H.-W. Hsueh, S.-C. Kao, M.-C. Ku, C.-C. Tsai and H.-F. Meng, Appl. Phys. Lett., 2010, 97, 203506.
22. K. Ghaffarzadeh, A. Nathan, J. Robertson, S. Kim, S. Jeon, C. Kim, U. I. Chung and J. H. Lee, Appl. Phys. Lett., 2010, 97, 143510.
23. H.-J. Jeong, K.-C. Ok, J. Park, J. Lim, J. Cho and J.-S. Park, IEEE Electron Device Lett., 2015, 36, 1160.
24. S.-J. Chang, T. H. Chang, W. Y. Weng, C. J. Chiu and S. P. Chang, IEEE J. Sel. Top. Quantum Electron., 2014, 20, 3803605.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra17896k

---