臺灣博碩士論文加值系統

人工智慧的運算在近日由於電腦效能的提升而日漸興起，而類神經網路學習則是其中一樣熱門的學習模型，其利用元件模仿人類大腦接受外部訊號，改變神經元間突觸的權重的方式學習，與以往人工運算相比從軟體上的改變轉移到硬體上的改變，而達到更低功耗、更快速的運算。電阻式記憶體可以透過外加偏壓改變其電導值，以改變電導的方式達到改變不同的權重，被視為模仿神經突觸的熱門元件之一。
在本文中研究氧化鉿基電阻式記憶體當作突觸元件的電性，透過增加一層富含氧空缺的氧化鈦，可幫助元件在從高組態轉換成低組態時的組態變化呈現緩和的變化，而在添加摻雜氧化鋁層於HfOx元件時可更進一步提升元件的性能，使組態變化更緩和，且穩定性也提升，此元件可以連續操作400次以上的組態轉換，而在室溫環境下維持其組態而不變化長達104秒，除此之外，其在脈衝量測下觀察電導的增益和減益特性，此元件在20us甚至縮短至1us皆可得到優良的電導增/減益特性，而其最好的非線性度更可達到1.52/2.15的效能。

Due to that the device toward more powerful, Artificial Intelligence (AI) computing become more and more popular. Neuromorphic computing is one of the most popular model of AI computing. By simulating the weight update of synapses between the neuro cells when the human brain accepts outside signal, we can use a new way to update the hardware condition instead of software. It is expected that the AI computing become much faster and much lower power consumption.
According to oxygen vacancy rich layer model and low oxygen vacancy mobility model, we add AlOx layer between TiOx and HfOx for improving the device’s performance. Based on TEM and EDX analyses, we find that Al doped into HfOx layer to form HfAlOx compound film. Based on such the oxygen vacancy mobility of HfAlOx layer formation, would lead to narrow the second filament. Through experiments, 1nm thick AlOx layer employed in the TiN/TiO/HfAlOx/TiN device exhibits the best property. Such device obtains excellent properties such as faster speed device (both set and reset pulse width is 1us) with good nonlinearity (3.39 for potentiation and 2.87 for depression behavior) and best nonlinearity (2.15 for potentiation and 1.52 for depression behavior with 10us pulse width) with 500 conductance states and retention with more than 104 s.

摘要 I
Abstract II
Contents V
Table captions VII
Figure captions VIII
Chapter 1 Introduction 1
1.1 Introduction of Random Access Memory 1
1.2Volatile Memory 1
1.2.1 SRAM 1
1.2.2 DRAM 2
1.3 Nonvolatile Memory 3
1.3.1 Nor/NAND Flash 4
1.3.2 PCRAM 5
1.3.3 MRAM 5
1.4 Neuromorphic computing 7
Chapter 2 Model of neuromorphic RRAM device 11
2.1 Filament formation and rupture 11
2.2 Different model for synapse device 11
2.2.1 Low mobility of oxygen vacancy layer 12
2.2.2 Oxygen vacancy rich layer 12
Chapter 3 Experimental Details 17
3.1 Fabrication of device 17
3.2 Nonlinearity fitting 18
Chapter 4 Result and discussion 22
4.1 Motivation 22
4.2 Electrical property of single HfO2 layer device 22
4.3 Electrical property of TiOx 3nm/HfO2 layer device 26
4.4 Electrical property of TiOx 1nm/HfO2 layer device 34
4.5 Improve pulse and DC measurement 44
4.5 Doped Al into HfOx layer and device performance 51
4.6 TEM and XPS analyses and model switching mechanism 60
Chapter 5 Conclusions 66
References 68

[1] Chen, Jian. “Selective operation of a multi-state non-volatile memory system in a binary mode.” U.S. Patent (2002) No. 6,456,528. 24.

[2] Sohyeon Kim, Yawar Abbas, Yu-Rim Jeon, Andrey Sergeevich Sokolov, Boncheol
Ku and Changhwan Choi. “Engineering synaptic characteristics of TaOx/HfO2 bi
-layered resistive switching device” Nanotechnology 29 (2018) 415204 (8pp)

[3] Park, Jaesung, Myunghoon Kwak, Kibong Moon, Jiyong Woo, Dongwook Lee and Hyunsang Hwang. “TiOx-Based RRAM Synapse With 64-Levels of Conductance and Symmetric Conductance Change by Adopting a Hybrid Pulse Scheme for Neuromorphic Computing.” IEEE Electron Device Letters 37 (2016): 1559-1562.

[4] Jiyong Woo, Kibong Moon, Jeonghwan Song, Sangheon Lee, Myounghun Kwak,Jaesung Park, and Hyunsang Hwang ”Improved Synaptic Behavior Under Identical Pulses Using AlOx/HfO2 Bilayer RRAM Array for Neuromorphic Systems”, IEEE ELECTRON DEVICE LETTERS, AUGUST 2016 VOL. 37, NO. 8

[5] P. Chen et al., "Mitigating effects of non-ideal synaptic device characteristics for on-chip learning," 2015 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), Austin, TX, 2015, pp. 194-199.

[6] H. -. P. Wong et al., "Metal–Oxide RRAM," in Proceedings of the IEEE, June 2012 vol. 100, no. 6, pp. 1951-1970.
[7] W. Wu et al., "A Methodology to Improve Linearity of Analog RRAM for Neuromorphic Computing," 2018 IEEE Symposium on VLSI Technology, Honolulu, HI, 2018, pp. 103-104.

[8] J. Woo et al., "Improved Synaptic Behavior Under Identical Pulses Using AlOx/HfO2 Bilayer RRAM Array for Neuromorphic Systems," in IEEE Electron Device Letters, vol. 37, no. 8, pp. 994-997, Aug. 2016.

[9] Huang C-Y, Huang C-Y, Tsai T-L, Lin C-A and Tseng T-Y 2014 “Switching mechanism of double forming process phenomenon in ZrOx/HfOy bilayer resistive switching memory structure” with large endurance Appl. Phys. Lett. 104 062901

[10] Y. Y. Chen et al., "Tailoring switching and endurance / retention reliability characteristics of HfO2 / Hf RRAM with Ti, Al, Si dopants," 2014 Symposium on VLSI Technology (VLSI-Technology): Digest of Technical Papers, Honolulu, HI, 2014, pp. 1-2.

[11] B. Traoré et al., "On the Origin of Low-Resistance State Retention Failure in HfO2-Based RRAM and Impact of Doping/Alloying," in IEEE Transactions on Electron Devices, vol. 62, no. 12, pp. 4029-4036, Dec. 2015.

[12] S. Park et al., "RRAM-based synapse for neuromorphic system with pattern recognition function," 2012 International Electron Devices Meeting, San Francisco, CA, 2012, pp. 10.2.1-10.2.4.

[13] S. Kim et al., " Neuronal dynamics in HfOx/AlOy-based homeothermic synaptic memristors with low-power and homogeneous resistive switching" Nanoscale, 2019,11, 237-245

[14] S. Yu, "Neuro-inspired computing with emerging nonvolatile memorys," in Proceedings of the IEEE, vol. 106, no. 2, pp. 260-285, Feb. 2018.

[15] J. Fu, Z. Liao, N. Gong and J. Wang, "Mitigating Nonlinear Effect of Memristive Synaptic Device for Neuromorphic Computing," in IEEE Journal on Emerging and Selected Topics in Circuits and Systems, vol. 9, no. 2, pp. 377-387, June 2019.

[16] P. Chen et al., "Mitigating effects of non-ideal synaptic device characteristics for on-chip learning," 2015 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), Austin, TX, 2015, pp. 194-199.

[17] P. Chen, X. Peng and S. Yu, "NeuroSim+: An integrated device-to-algorithm framework for benchmarking synaptic devices and array architectures," 2017 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, 2017, pp. 6.1.1-6.1.4.

[18] A. Padovani, J. Woo, H. Hwang and L. Larcher, "Understanding and Optimization of Pulsed SET Operation in HfOx-Based RRAM Devices for Neuromorphic Computing Applications," in IEEE Electron Device Letters, vol. 39, no. 5, pp. 672-675, May 2018.

[19] E. Covi, S. Brivio, A. Serb, T. Prodromakis, M. Fanciulli and S. Spiga, "HfO2-based memristors for neuromorphic applications," 2016 IEEE International Symposium on Circuits and Systems (ISCAS), Montreal, QC, 2016, pp. 393-396.
d

[20] J. Woo, K. Moon, J. Song, M. Kwak, J. Park and H. Hwang, "Optimized Programming Scheme Enabling Linear Potentiation in Filamentary HfO2 RRAM Synapse for Neuromorphic Systems," in IEEE Transactions on Electron Devices, vol. 63, no. 12, pp. 5064-5067, Dec. 2016.

[21] Mengxing Wang et. al.“Tunnel Junction with Perpendicular Magnetic Anisotropy: Status and Challenges” Micromachines 2015, 6(8), 1023-1045;

[22] Nagarajan Raghavan et. al. “Performance and reliability trade-offs for high-K RRAM” Microelectronics Reliability 54 (2014) 2253–2257

[23] 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," in IEEE Electron Device Letters, vol. 38, no. 8, pp. 1019-1022, Aug. 2017.