臺灣博碩士論文加值系統

目前氧化物薄膜電晶體(TFT, thin film transistor)主要應用在非揮發性記憶體(nonvolatile memory)、生物感測器(biosensor)、紫外光光偵測器(UV photodetector)和透明可撓性面板(transparent flexible display)上，最近幾年除了上述的應用外，氧化物薄膜電晶體開始作為突觸元件被應用在類神經型態運算(neuromorphic computing)上，主要是利用氧化物薄膜電晶體在照射次能隙光源(sub-bandgap light source)時會產生持續性光電導效應(persistent photoconductivity)，所以在重複照光時氧化物薄膜電晶體會達到不同的導電狀態，所以作為突觸元件應用在類神經型態運算上時便可以用來控制輸入資料所佔的權重(synaptic weight)，本研究利用溶液法製備氧化鋅錫為基底的薄膜電晶體(ZTO-base TFT)，並以波長為405 nm及520 nm雷射光作為次能隙光源。
本研究第一部分主要探討的是持續性光電導效應的產生原因跟影響帶正電氧空缺和光電子復合速率的因素，從紫外光-可見光穿透圖譜的結果中我們可以計算出氧化鋅錫薄膜的能隙為3.9 eV，在照射次能隙光源時，氧化鋅錫薄膜裡的中性氧空缺會吸收光子的能量並游離成帶正電的氧空缺，在此過程中會伴隨著光電子的產生。當光源被關閉時，帶正電氧空缺跟光電子復合並轉換成原本的中性氧空缺過程中有需要克服的能障存在，所以帶正電氧空缺無法快速地跟光電子復合並轉換成中性氧空缺，導致在光反應量測中會有殘留的光電流在通道中(持續性光電導效應)，能障的大小我們可藉由變溫動態光反應的量測結果經由公式計算可得到氧化鋅錫薄膜帶正電氧空缺跟光電子復合並轉換成原本的中性氧空缺過程中的能障為0.294 meV，接下來我們使用405 nm的雷射光作為光源照射氧化鋅錫薄膜並改變光源的功率密度和施加不同的閘極偏壓來探討量測參數對帶正電氧空缺和光電子復合速率的影響，為了充分的討論不同量測參數帶來的影響，吾人在這一部份使用了兩種不同的光反應量測方式，第一種是長時間的量測，光源會持續打開直到光電流達到飽和，發現在長時間穩定的光源照射下，光源的功率密度增加和施加正的閘極偏壓會導致飽和的光電流值增加，帶正電氧空缺和光電子復合的速率變快。第二種是固定開關光源的時間，反覆開關光源的量測，光源的功率密度增加會使得元件在相同的操作次數下更快到達飽和電流，施加正的閘極偏壓除了使得元件在相同的操作次數下更快到達飽和電流還可以持續性光電導效應。
本研究第二部分則是在探討氧化鋅錫為基底的薄膜電晶體作為人工突觸元件(artificial synaptic device)的相關生物體突觸行為模擬和實際的應用量測。在這一個部分，吾人將製備氧化鋅錫薄膜光電晶體和紅汞染料敏化(mercurochrome-sensitized)後的氧化鋅錫薄膜光電晶體，並使用波長為405 nm及520 nm雷射光作為做為增益型突觸前神經元刺激(excitatory presynaptic neuron spike)，正閘極偏壓脈衝作為抑制型突觸前神經元刺激(inhibitory presynaptic neuron spike)，汲極電流(光電流)作為突觸後電流(postsynaptic current)，電流(導電率)的變化則被視為突觸權重(synaptic weight)。重要的生物體突觸及記憶行為包含短期記憶(STM, short-term memory)、長期記憶(LTM, long-term memory)、短期記憶轉換成長期記憶(STM-to-LTM transition)和學習經驗相關的行為(learning-experience behaviors)，藉由改變施加光刺激的次數並量測施加完光刺激後電流自然衰退的行為，可以模擬人腦短期記憶、長期記憶、短期記憶轉換成長期記憶的行為，學習經驗相關的行為則可藉由反覆施加光刺激和讓光電流自然衰退的操作來模仿。實際的應用量測為突觸後電流(權重)增幅/抑制行為，突觸後電流(權重)增幅行為(potentiation)跟氧化鋅錫薄膜的持續性光電導效應有關，突觸後電流(權重)抑制行為(depression)則是跟消除氧化鋅錫薄膜的持續性光電導效應有關，突觸後電流(權重)增幅行為可藉由改變光刺激的功率密度和施加的光刺激次數來調控元件的導電狀態的增加情形，突觸後電流(權重)抑制行為則是藉由改變正閘極偏壓的大小和施加的脈衝次數來調控元件的導電狀態的減少情形。本研究提供了一個有效的方法使得光電晶體可以做為突觸元件使用並應用在下一代的類神經型態運算系統。

The application of oxide TFT mostly in nonvolatile memory、biosensor、UV photodetector and transparent flexible display. In recently, oxide TFT is used as synaptic device in neuromorphic computing. Because oxide TFT would have PPC effect with sub-bandgap light source, oxide TFT would have different conductive states. These states could use as synaptic weight to modulate the input date. In this study, ZTO thin film was fabricated by solution process and the light source was 405 nm laser and 520 nm laser.
In the first part, we would discuss the generation of PPC effect and the influence parameter. From UV-Vis transmission spectrum, we could get the bandgap of ZTO was 3.9 eV. When ZTO TFT was illuminated by sub-bandgap light, neutral oxygen vacancies would ionize to positively-charged oxygen vacancies and released photoelectrons. When light was turned off, there was an energy barrier needed to be overcome while positively-charged oxygen vacancies recombined with photoelectrons and transformed to neutral oxygen vacancies. There would have persistent photocurrent in the channel (persistent photoconductivity). By calculating the data which measured at different temperature, we could get the energy barrier was 0.294 eV. We changed power density of light source and applied different gate bias to discuss the influence of measurement parameter to the recombination rate of photoelectrons and positively-oxygen vacancies with 405 nm laser as light source. We would use two measurement method to discuss the influence of different parameter. First was long time measurement, the light source would persistent turn on until photocurrent reached saturation. It is noted that the saturation photocurrent would increase with the increase of power density and positive gate bias. The increasement of photocurrent would cause the increase of recombination rate. The second one would set the light on and light off time were 10s and measure with multioperation. The photocurrent would faster reach saturation in the same operation number with the increase of power density and stop the increase of persistent photocurrent. The photocurrent would faster reach saturation in the same operation number with the positive gate bias and elimination PPC effect.
In the second part, we would discuss the biological synaptic behavior intimated with artificial ZTO-based TFT and the applied behavior measurement in neuromorphic computing. We would fabricate ZTO TFT and mer/ZTO TFT. With intimating the biological synapses, it was suggested that the 405nm and 520nm light spike, the drain current (photocurrent), the positive gate bias pulse and the change in current (conductivity) could be represented as the excitatory presynaptic neuron spike, the postsynaptic current, the inhibitory presynaptic neuron spike and synaptic weight respectively. The important synaptic functions and memory behaviors are investigated, including potentiation/depression, STM, LTM, STM-to-LTM transition and learning-experience behaviors. STM, LTM, STM-to-LTM transition could be intimated by change light stimulation number. The learning-experience behaviors could be intimated with repeat operation about light stimulation and spontaneous retention. The applied behavior measurement in neuromorphic computing meant the potentiation and depression behavior. Potentiation behavior was associated with the PPC effect of ZTO and depression behavior was associated with the elimination of PPC effect of ZTO. Potentiation behavior could modulate the increase of conductivity in ZTO TFT with the power density of light source and the spike number. Depression behavior could modulate the decrease of conductivity in ZTO TFT with the positive gate bias and pulse number. This study provides a simple and effective method to make phototransistors being synaptic device, which supply a new strategy for developing next generation neuromorphic computation system.

摘要 I
Abstract III
List of Figures VIII
List of Tables XII
Chapter 1 Introduction 1
1-1 Motivation 1
1-2 Introduction of oxide TFT 2
1-3 Important parameters for evaluating characteristics of TFT 3
1-4 Important parameters for evaluating characteristics of photodetector 6
1-5 Application of synaptic device in artificial neutral network3-5 7
Chapter 2 Photoresponse of zinc sin oxide (ZTO) thin film transistor 10
2-1 Fabrication of ZTO thin film transistor 10
2-2 TFT characteristics 10
2-2.1 Basic TFT characteristics 11
2-2.2 Photo-TFT characteristic 11
2-3 Photoresponse behavior of oxide semiconductor 15
2-3.1 Photoresponse behavior 15
2-3.2 Energy barrier for photoelectrons and positively-charged oxygen vacancies recombination process 16
2-3.3 Photoresponse formula derivation 16
2-4 Photoresponse behavior of ZTO transistor 19
2-4.1 Energy barrier 19
2-4.2 Long time photoresponse behavior of ZTO TFT 22
2-4.3 Short time photoresponse behavior of ZTO TFT 26
Chapter 3 Synaptic behavior of ZTO-base thin film transistor 32
3-1 Fabrication of ZTO-based thin film transistor 32
3-2 Basic TFT characteristics 33
3-3 Synaptic behavior of ZTO and mer/ZTO transistor 34
結論 63
Conclusion 65
Reference 67

1.H. Choi, S. Seo, J. H. Lee, S. H. Hong, J. Song, S. Kim, S. Y. Yim, K. Lee, S. J. Park and S. Lee, Solution-processed ZnO/SnO2 bilayer ultraviolet phototransistor with high responsivity and fast photoresponse, J. Mater. Chem. C 6, 6014-6022 (2018).
2.H. Hosono, Ionic amorphous oxide semiconductors: Material design, carrier transport, and device application, J. Non-Cryst. Solids 352, 851-858 (2006).
3.D. Kuzum, S. M. Yu and H. S. P. Wong, Synaptic electronics: materials, devices and applications, Nanotechnology 24, 22 (2013).
4.S. Agatonovic-Kustrin and R. Beresford, Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research, J. Pharm. Biomed. Anal. 22, 717-727 (2000).
5.D. P. Kumar, T. Amgoth and C. S. R. Annavarapu, Machine learning algorithms for wireless sensor networks: A survey, Inf. Fusion 49, 1-25 (2019).
6.R. K. Mehra, H. Duan, S. J. Luo and F. H. Ma, Laminar burning velocity of hydrogen and carbon-monoxide enriched natural gas (HyCONG): An experimental and artificial neural network study, Fuel 246, 476-490 (2019).
7.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, Metal Oxide Thin Film Phototransistor for Remote Touch Interactive Displays, Adv. Mater. 24, 2631-2636 (2012).
8.A. Janotti and C. G. Van de Walle, LDA+ U and hybrid functional calculations for defects in ZnO, SnO2, and TiO2, Phys. Status Solidi B-Basic Solid State Phys. 248, 799-804 (2011).
9.A. Janotti and C. G. Van de Walle, Oxygen vacancies in ZnO, Appl. Phys. Lett. 87, 3 (2005).
10.A. Janotti and C. G. Van de Walle, Native point defects in ZnO, Phys. Rev. B 76, 22 (2007).
11.I. Tanaka, F. Oba, K. Tatsumi, M. Kunisu, M. Nakano and H. Adachi, Theoretical formation energy of oxygen-vacancies in oxides, Mater. Trans. 43, 1426-1429 (2002).
12.D. A. Neamen, Semiconductor Physics And Devices. (McGraw-Hill:New York, 2011).
13.K. Sakaguchi, K. Shimkawa and Y. Hatanaka, Transport and recombination of photo‐carriers under potential fluctuation in TiOx films prepared by rf magnetron sputtering method, Phys. Status Solidi C
8, 2796-2799 (2011).
14.L. J. A. Koster, V. D. Mihailetchi and P. W. M. Blom, Bimolecular recombination in polymer/fullerene bulk heterojunction solar cells, Appl. Phys. Lett. 88, 3 (2006).
15.C. Groves and N. C. Greenham, Bimolecular recombination in polymer electronic devices, Phys. Rev. B 78, 8 (2008).
16.M. Lee, M. Kim, J. W. Jo, S. K. Park and Y. H. Kim, Suppression of persistent photo-conductance in solution-processed amorphous oxide thin-film transistors, Appl. Phys. Lett. 112, 5 (2018).
17.K. Ghaffarzadeh, A. Nathan, J. Robertson, S. Kim, S. Jeon, C. Kim, U. I. Chung and J. H. Lee, Persistent photoconductivity in Hf-In-Zn-O thin film transistors, Appl. Phys. Lett. 97, 3 (2010).
18.Z. G. Yin, X. W. Zhang, Z. Fu, X. L. Yang, J. L. Wu, G. S. Wu, L. Gong and P. K. Chu, Persistent photoconductivity in ZnO nanostructures induced by surface oxygen vacancy, Phys. Status Solidi-Rapid Res. Lett. 6, 117-119 (2012).
19.M. Dusza, F. Granek and W. Strek, Illumination intensity dependent photoresponse of ultra-thin ZnO/graphene/ZnO heterostructure, Opt. Mater. 74, 176-182 (2017).
20.E. Di Gennaro, U. Coscia, G. Ambrosone, A. Khare, F. M. Granozio and U. S. Di Uccio, Photoresponse dynamics in amorphous-LaAlO 3/SrTiO 3 interfaces, Sci Rep 5, 6
(2015).
21.S. L. Dai, X. H. Wu, D. P. Liu, Y. L. Chu, K. Wang, B. Yang and J. Huang, Light-Stimulated Synaptic Devices Utilizing Interfacial Effect of Organic Field-Effect Transistors, Acs Applied Materials ＆ Interfaces 10, 21472-21480 (2018).
22.Q. T. Wu, J. W. Wang, J. C. Cao, C. Y. Lu, G. H. Yang, X. W. Shi, X. C. Chuai, Y. X. Gong, Y. Su, Y. Zhao, N. D. Lu, D. Geng, H. Wang, L. Li and M. Liu, Photoelectric Plasticity in Oxide Thin Film Transistors with Tunable Synaptic Functions, Adv. Electron. Mater. 4, 8 (2018).
23.M. Lee, W. Lee, S. Choi, J. W. Jo, J. Kim, S. K. Park and Y. H. Kim, Brain-Inspired Photonic Neuromorphic Devices using Photodynamic Amorphous Oxide Semiconductors and their Persistent Photoconductivity, Adv. Mater. 29, 8 (2017).
24.S. Gao, G. Liu, H. L. Yang, C. Hu, Q. L. Chen, G. D. Gong, W. H. Xue, X. H. Yi, J. Shang and R. W. Li, An Oxide Schottky Junction Artificial Optoelectronic Synapse, ACS Nano 13, 2634-2642 (2019).
25.M. G. Yun, Y. K. Kim, C. H. Ahn, S. W. Cho, W. J. Kang, H. K. Cho and Y. H. Kim, Low voltage-driven oxide phototransistors with fast recovery, high signal-to-noise ratio, and high responsivity fabricated via a simple defect-generating process, Sci Rep 6, 9 (2016).
26.S. Jeon, S. E. Ahn, I. Song, C. J. Kim, U. I. Chung, E. Lee, I. Yoo, A. Nathan, S. Lee, J. Robertson and K. Kim, Gated three-terminal device architecture to eliminate persistent photoconductivity in oxide semiconductor photosensor arrays, Nat. Mater. 11, 301-305 (2012).
27.J. Choi, J. Park, K. H. Lim, N. K. Cho, J. Lee, S. Jeon and Y. S. Kim, Photosensitivity of InZnO thin-film transistors using a solution process, Appl. Phys. Lett. 109, 5 (2016).
28.S. Lim, M. Kwak and H. Hwang, Improved Synaptic Behavior of CBRAM Using Internal voltage Divider for Neuromorphic Systems, IEEE Trans. Electron Devices 65, 3976-3981 (2018).
29.J. X. Wang, Y. Chen, L. A. Kong, Y. Fu, Y. L. Gao and J. Sun, Deep-ultraviolet-triggered neuromorphic functions in In-Zn-O phototransistors, Appl. Phys. Lett. 113, 5 (2018).
30.W. J. Cheng, R. R. Liang, H. Tian, C. C. Sun, C. S. Jiang, X. W. Wang, J. Wang, T. L. Ren and J. Xu, Proton Conductor Gated Synaptic Transistor Based on Transparent IGZO for Realizing Electrical and UV Light Stimulus, IEEE J. Electron Devices Soc. 7, 38-45 (2019).
31.H. K. Li, T. P. Chen, P. Liu, S. G. Hu, Y. Liu, Q. Zhang and P. S. Lee, A light-stimulated synaptic transistor with synaptic plasticity and memory functions based on InGaZnOx-Al2O3 thin film structure, J. Appl. Phys. 119, 5 (2016).