臺灣博碩士論文加值系統

由於馮紐曼架構無法快速處理大量數據以符合人工智慧運算需求，因此利用金氧半場效電晶體和非揮發性記憶體作為神經元和突觸以仿造大腦神經架構提升運算速度。電阻式記憶體為作為電子突觸應用之最具潛力元件，因此本論文利用氧化鈦製作電阻式記憶體元件，探討其作為電子突觸之特性，並藉由雙層結構改善其非理想特性。
第一部分以氧化鈦作為介電層製作Ni/TiOx/W結構，透過循環掃描確認其具備突觸增益與抑制特性，並由電壓電流曲線擬合確認其機制為蕭特基能障調變。經由不同脈衝波形量測元件突觸增益與抑制、長短期記憶特性與成對脈衝促進特性，並計算其非線性度、記憶時間與成對脈衝促進。第二部分添加氧化矽層製作Ni/SiOy/TiOx/W雙層結構元件，並量測其突觸特性與探討操作機制。
Ni/SiOy/TiOx/W元件相較於Ni/TiOx/W元件，抑制部分蕭特基能障改變幅度由0.824 eV~0.881 eV提升至1.02 eV~1.20 eV，而增益幅度由1.02~0.82eV提升至1.28 eV~1.02 eV，增益與抑制非線性度由10.8 %與22.5 %改善至6.7 %與19.7 %，長短期記憶皆呈現與生物特性相似，成對脈衝促進時間由0.5 s提升至0.6 s。Ni/TiOx/W元件突觸增益特性主要歸因於鎢與氧化鈦間的能障變化所致，而突觸抑制特性則為鎳與氧化鈦間能障，因此Ni/TiOx/W元件突觸抑制特性的能障變化不佳。而Ni/SiOy/TiOx/W元件添加的氧化矽層因其自由能與氧化鈦相近可利於氧空缺累積與游離調變蕭特基能障高度，故添加氧化矽層於鎳與氧化鈦間可改善突觸特性。

The Von Neumann architecture cannot quickly process large amounts of data to meet the requirement of artificial intelligence computing. Therefore, MOSFET and non-volatile memory are used as neurons and synapses to imitate the neural architecture of the brain to increase the computing speed. Resistive memory is the most potential device for electronic synapse applications. In this thesis, titanium oxide is used to fabricate resistive memory devices, to explore its characteristics as an electronic synapse. In addition, a double-layer structure was used to improve its non-ideal characteristics.
In the first part, a Ni/TiOx/W structure is fabricated and its synaptic potentiation and depression was demonstrated by consecutive voltage sweepings. The fitting result of voltage-current curve indicated that the synaptic behavior was due to Schottky barrier modulation. Different pulse waveforms were used to characterize the synaptic potentiation and depression, long short-term memory and paired-pulse facilitation (PPF) of the devices, and its nonlinearity, retention and PPF index were also calculated. In the second part of the thesis, a silicon oxide layer was added to fabricate a Ni/SiOy/TiOx/W double-layer structure, and its synaptic characteristics were measured and discussed. Compared with the Ni/TiOx/W device, adding the SiOx layer led to a larger change in Schottky barrier for depression (1.02 eV~1.20 eV) and potentiation (1.28~1.02 eV). The Ni/SiOy/TiOx/W device had nonlinearities of 6.7 % (potentiation) and 19.7 % (depression), PPF index of 0.6 s, and similar long short-term memory characteristics with the Ni/TiOx/W device. The potentiation behavior was dominated by the Schottky barrier between W electrode and TiOx layer, and the depression behavior was dominated by the Schottky barrier between the Ni electrode and TiOx layer, which led to small barrier change. The free energy of the silicon oxide layer is similar to that of titanium oxide, which facilitates the accumulation and release of oxygen vacancies and helps to adjust the Schottky energy barrier. Therefore, adding a silicon oxide layer between nickel and titanium oxide can improve the synaptic properties.

目錄
中文摘要 I
英文摘要 II
誌謝 IV
目錄 V
圖目錄 VII
表目錄 X
第一章 緒論 1
1-1 前言 1
1-2 研究動機與目的 1
1-3 論文架構 2
第二章 文獻回顧與基礎理論 3
2-1 電阻式記憶體簡介 3
2-2 電阻式記憶體發展 6
2-3 電阻式記憶體之人工神經網路應用 7
2-4 電阻式記憶體與電子突觸差別 11
2-5 電子突觸操作機制比較 15
2-6 電流傳導機制 19
2-7 類比式電阻記憶體改善 22
第三章 實驗步驟 26
3-1 實驗流程說明 26
3-2 實驗設備 27
3-3 元件製備 29
3-4 材料分析 31
3-5 電性分析 31
第四章 結果與討論 37
4-1 Ni/TiOx/W元件類比特性與機制探討 37
4-1-1元件製備流程與材料化學鍵結分析 37
4-1-2電導切換機制 39
4-1-3增益與抑制分析 44
4-1-4長短期記憶分析 49
4-1-5成對脈衝促進分析 51
4-2 Ni/SiOy/TiOx/W元件類比特性與機制探討 51
4-2-1元件製備流程與材料化學鍵結分析 51
4-2-2電導變化機制 53
4-2-3增益與抑制分析 57
4-2-4長短期記憶分析 63
4-2-5成對脈衝產生分析 64
第五章 結果與未來展望 65
5-1 結論 65
5-1-1電導變化 65
5-1-2非線性度 66
5-1-3長短期記憶 66
5-1-4成對脈衝促進 67
5-2未來展望 67
參考文獻 69

[1]C. Szegedy, W. Liu, and Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, A. Rabinovich, "Going Deeper with Convolutions," IEEE Conference on Computer Vision and Pattern Recognition 2015, Massachusetts, United States of America, Jun. 7-12, 2015.
[2]O. Krestinskaya, K. N. Salama, and A. P. James, "Learning in Memristive Neural Network Architectures Using Analog Backpropagation Circuits," IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 66, no. 2, pp. 719–732, Feb. 2019.
[3]L. Michalas, S. Stathopoulos, A. Khiat, and T. Prodromakis, "An Electrical Characterisation Methodology for Identifying the Switching Mechanism in TiO2 Memristive Stacks," Scientific Reports, vol. 9, no. 1, pp. 1-8, Dec. 2019.
[4]L. Tai, X Xu, P Yuan, J. Yu, Q. Luo, H. Lv, and M. Liu, "A CMOS Compatible, Forming Free TaON-based ReRAM with Low Soft Errors and Good Retention!!," IEEE International Conference on Integrated Circuits, Technologies and Applications 2018, Beijing, China, Nov. 21-23, 2018.
[5]C. H. Cheng, C. Y. Tsai, A. Chin, and F. S. Yeh, "High Performance Ultra-Low Energy RRAM with Good Retention and Endurance," IEEE International Electron Devices Meeting 2010, California, United States of America, Dec. 6-8, 2010.
[6]M. Ismail, I. Talib, A. M. Rana, E. Ahmed, and M. Y. Nadeem, "Performance Stability and Functional Reliability in Bipolar Resistive Switching of Bilayer Ceria Based Resistive Random Access Memory Devices," Journal of Applied Physics, vol. 117, no. 8, Feb. 2015.
[7]H. Wang and X. Yan, "Overview of Resistive Random Access Memory (RRAM): Materials, Filament Mechanisms, Performance Optimization, and Prospects," Physica Status Solidi – Rapid Research Letters, vol. 13, no. 9, pp. 1900073, May 2019.
[8]R. Waser, R. Dittmann, G. Staikov, and K. Szot, "Redox-Based Resistive Switching Memories - Nanoionic Mechanisms, Prospects, and Challenges," Advanced Materials, vol. 21, no. 25-26, pp. 2632–2663, Jul. 2009.
[9]J. Park, K. P. Biju, S Jung, W. Lee, J. Lee, S. Kim, S. Park, J. Shin, and H. Hwang, "Multibit Operation of TiOx-Based ReRAM by Schottky Barrier Height Engineering," IEEE Electron Device Letters, vol. 32, no. 4, pp. 476–478, Apr. 2011.
[10]I. Valov, "Interfacial Interactions and Their Impact on Redox-based Resistive Switching Memories (ReRAMs)," Semiconductor Science and Technology, vol. 32, no.9, pp. 093006, Sep. 2017.
[11]A. Sebastian, M. Le Gallo, and E. Eleftheriou, "Computational Phase-Change Memory: Beyond Von Neumann Computing," Journal of Physics D: Applied Physics, vol. 52, no. 44, pp. 443002, Oct. 2019.
[12]A. Chattopadhyay, C. H. Chang, and H. Yu, Emerging Technology and Architecture for Big-data Analytics, Springer International Publishing, pp. 77-81, 2017.
[13]Y. Chen, "ReRAM: History, Status, and Future," IEEE Transactions on Electron Devices, vol. 67, no. 4, pp. 1420–1433, Apr. 2020.
[14]Y. C. Chen, C. C. Lin, S. T. Hu, C. Y. Lin, B. Fowler, and J. Lee, "A Novel Resistive Switching Identification Method through Relaxation Characteristics for Sneak-path-constrained Selectorless RRAM Application," Scientific Reports, vol. 9, no. 1, pp. 12420, Dec. 2019.
[15]J. B. Reece, F. Rawle, R. Jackson, and P. V. Minorsky, S. Wasserman, S. J. Walde, M. L. Cain, L. A. Urry, C. Moyes, and D. G. Durnford, Campbell Biology, Pearson Publishing, pp. 154-161, Aug. 2018.
[16]S. Yu, Neuro-inspired Computing Using Resistive Synaptic Devices, Springer International Publishing, pp.227-231, 2017.
[17]B. Kasap and A. J. Van Opstal, "Dynamic Parallelism for Synaptic Updating in GPU-accelerated Spiking Neural Network Simulations," Neurocomputing, vol. 302, pp. 55–65, Aug. 2018.
[18]Y. Li, Z. Wang, R. Midya, Q. Xia, and J. J. Yang, "Review of Memristor Devices in Neuromorphic Computing: Materials Sciences and Device Challenges," Journal of Physics D: Applied Physics, vol. 51, no. 50, pp. 503002, Dec. 2018.
[19]B. Pernici, Special Topics in Information Technology, Springer International Publishing, pp. 784-788, 2020.
[20]T. Wang, Y. C. Lin, Y. F. Wang, C. W. Hsu, and T. H. Hou, "3D Synaptic Architecture with Ultralow Sub-10 fJ Energy per Spike for Neuromorphic Computation," IEEE International Electron Devices Meeting, California, United States of America, Dec. 15-17, 2014.
[21]M. Ismail, S.-U. Nisa, A. M. Rana, T. Akbar, J. Lee, and S. Kim, "Enhancement of Resistive Switching Performance by Introducing a Thin Non-stoichiometric CeO2-x Switching Layer in TiO2-based Resistive Random Access Memory," Applied Physics Letters, vol. 114, no. 1, pp. 012101, Jan. 2019.
[22]M. Zhao, B. Gao, J. Tang, H. Tang, and H. Wu, "Reliability of Analog Resistive Switching Memory for Neuromorphic Computing," Applied Physics Reviews, vol. 7, no. 1, pp. 011301, Mar. 2020.
[23]H. J. Kim, M. Kim, K. Beom, H. Lee, C. J. Kang, and T. S. Yoon, "A Pt/ITO/CeO2/Pt Memristor with an Analog, Linear, Symmetric, and Long-term Stable Synaptic Weight Modulation," APL Materials, vol. 7, no. 7, pp. 071113, Jul. 2019.
[24]W. Zhang, B. Gao, and J. Tang, X. Li, W. Wu, H. Qian, H. Wu, "Analog‐Type Resistive Switching Devices for Neuromorphic Computing," Physica Status Solidi – Rapid Research Letters, vol. 13, no. 10, pp. 1900204, Oct. 2019.
[25]W. Wu, H. Wu, B. Gao, N. Deng, S. Yu, and H. Qian, "Improving Analog Switching in HfOx-Based Resistive Memory with a Thermal Enhanced Layer," IEEE Electron Device Letters, vol. 38, no. 8, pp. 1019–1022, Aug. 2017.
[26]B. Gao, B. chen, and F. Zhang, P. Huang, L. Liu, X. Liu, and J.Kang, "Multi-bit Nonvolatile Logic Implemented with Metal-oxide Based Resistive Switching Device," Solid State Communications, vol. 205, pp. 51–54, Mar. 2015.
[27]J. Woo, K. Moon, J. Song, S. Lee, M. Kawk, J. Park, and H. Hwang, "Improved Synaptic Behavior Under Identical Pulses Using AlOx/HfO2 Bilayer RRAM Array for Neuromorphic Systems," IEEE Electron Device Letters, vol. 37, no. 8, pp. 994–997, Aug. 2016.
[28]H. Jiang, L. Han, P Lin, Z. Wang, M. H. Jang, Q. Wu, M. Barnell, J. J. Yang, H. L. Xin, and Q. Xia, "Sub-10 nm Ta Channel Responsible for Superior Performance of a HfO2 Memristor," Scientific Reports, vol. 6, no. 1, pp. 28525, Jun. 2016.
[29]S. Park, J. Noh, M. Choo, A. M. Sheri, M. Chang, Y. B. Kim, C. J. Kim, M. Jeon, B. G. Lee, B. H. Lee, and H. Hwang, "Nanoscale RRAM-based Synaptic Electronics: Toward a Neuromorphic Computing Device," Nanotechnology, vol. 24, no. 38, pp. 384009, Sep. 2013.
[30]F.-C. Chiu, "A Review on Conduction Mechanisms in Dielectric Films," Advances in Materials Science and Engineering, vol. 2014, pp. 1–18, 2014.
[31]S. Chandrasekaran, F. M. Simanjuntak, R. Saminathan, D. Panda, and T.-Y. Tseng, "Improving Linearity by Introducing Al in HfO2 as a Memristor Synapse Device," Nanotechnology, vol. 30, no. 44, pp. 445205, Nov. 2019.
[32]T. Ioroi, H. Kageyama, T. Akita, and K. Yasuda, "Formation of Electro-Conductive Titanium Oxide Fine Particles by Pulsed UV Laser Irradiation," Physical Chemistry Chemical Physics, vol. 12, no. 27, pp. 7529, Jun. 2010.
[33]H. Khan and I. K. Swati, "Fe3+-doped Anatase TiO 2 with d–d Transition, Oxygen Vacancies and Ti3+ Centers: Synthesis, Characterization, UV–vis Photocatalytic and Mechanistic Studies," Industrial & Engineering Chemistry Ressearch, vol. 55, no. 23, pp. 6619–6633, Jun. 2016.
[34]H. J. Kim, D. Park, and P. Yang, K. Beom, M. J. Kim, C. Shin, C. J. Kang, and T.-S. Yoon, "Synaptic Characteristics with Strong Analog Potentiation, Depression, and Short-term to Long-term Memory Transition in a Pt/CeO2/Pt Crossbar Array Structure," Nanotechnology, vol. 29, no. 26, pp. 265204, Jun. 2018.
[35]Y. N. Zhong, T. Wang, X. Gao, J.-L. Xu, and S.-D. Wang, "Synapse-Like Organic Thin Fil m Memristors," Advanced Functional Materials, vol. 28, no. 22, pp. 1800854, May 2018.
[36]J. J. Yang, J. P. Strachan, F. Miao, M.-X. Zhang, M. D. Pickett, W. Yi, D. Ohlberg, G. Medeiros-Ribeiro, and R. S. Williams, "Metal/TiO2 Interfaces for Memristive Switches," Applied Physics A, vol. 102, no. 4, pp. 785–789, Mar. 2011.
[37]J. HoIzl and F. K. Schulte, Work Function of Metals, Springer, pp. 150, 1979.
[38]Y. Itzhaik, T. Bendikov, D. Hines, P. V. Kamat, H. Cohen, and G. Hodes, "Band Diagram and Effects of the KSCN Treatment in TiO2/Sb2S 3/CuSCN ETA Cells," The Journal of Physical Chemistry C, vol. 120, no. 1, pp. 31–41, Jan. 2016.
[39]S. Seetharaman, A. McLean, R. Guthrie, and S. Sridhar, Treatise on Process Metallurgy Elsevier Publishing, pp. 452, 2014.
[40]D. Acharyya, A. Hazra, and P. Bhattacharyya, "A Journey Towards Reliability Improvement of TiO2 Based Resistive Random Access Memory: A review," Microelectronics Reliability, vol. 54, no. 3, pp. 541–560, Mar. 2014.
[41]K. E. González–Flores, B Palacios-Márquez, J Álvarez–Quintana, S A Pérez–García, L Licea–Jiménez, P Horley, and A Morales-Sánchez, "Resistive Switching Control for Conductive Si-nanocrystals Embedded in Si/SiO2 Multilayers," Nanotechnology, vol. 29, no. 39, pp. 395203, Sep. 2018.