臺灣博碩士論文加值系統

電阻式隨機存取記憶被視為最有潛力的非揮發性記憶體，因為其具備低功耗、高操作速度和可做成三維結構……等優點。電阻式記憶體另外一個有潛力的應用是在類神經網路的運算當中，作為類神經突觸，此類神經突觸需要電阻式記憶體的類比電阻切換特性，且若要能夠當作突觸來用有幾項電性要求及指標，例如:線性度、對稱性、動態範圍、能耗……等。眾多文獻指出，電阻類比切換的線性度越好，或是更低的非線性度會導致未來將其導入類神經運算後，執行機器學習任務時，有更好的學習正確率。
本文主要研究以氧化鋯為主的電阻式記憶體如何在結構、製程上做改進，以得到更低的非線性度。第一部分討論將惰性下電極(Pt)改為較活性的電極(TiN)，成功製作出具有類比電阻切換特性的電阻式記憶體，此外，我們提出了一種機制來解釋其差異。第二部分著重於透過熱退火的方式使得類比電阻切換的非線性度及耐久度獲得改善；電導增強和抑制的非線性度分別從大於9進步到4.45和5.29，這可以從經退火後得到TiON層來解釋。第三部分則是討論藉由氧化鋁作為額外的阻擋層，沉積在氧化鋯下，得到雙層結構的電阻式記憶體，且因氧化鋁具有低離子遷移率等的特性，相比於單層氧化鋯結構，可以使電導增強和抑制非線性度改善到3.94和2.42，此外藉由更進一步的製程參數優化，成功製造出高線性度的類神經突觸，其電導增強和抑制的非線性度低至1.3和1.82，未來可將此結構應用於類神經運算的電子突觸。

Resistive random access memory (RRAM) is the most promising nonvolatile memory in the future, due to its serval advantages, low power consumption, high operation speed, and 3D compatible architecture……etc. Another potential application of RRAM is to implement it to neuromorphic computing. To use RRAM as an electronic synapse, it should perform the capability of gradual resistance change. Furthermore, some electrical properties and metrics need to be considered, e.g., linearity, symmetry, dynamic range, etc. Many papers conclude that the higher linearity of resistance change, the better the learning accuracy we can achieve in the machine learning task.
In this thesis, we mainly focus on ZrOx-based RRAM, trying to improve the nonlinearity by device design engineering. Firstly, by changing the bottom electrode from Pt to TiN, we successfully demonstrate ZrOx-based RRAM with gradual resistivity change. In addition, we propose a mechanism to explain the difference. Secondly, by introducing additional post-deposition annealing, the nonlinearity of the weight update is further improved from >9 to 4.45 for potentiation; >9 to 5.29 for depression. This can be explained by interface oxygen vacancies due to the formation of the TiON layer after annealing. In the third part, based on the previous report, a methodology to improve the nonlinearity, we used the AlOx as a barrier layer, because AlOx has low ion mobility due to the ALD process. By stacking AlOx under ZrOx, we obtained the bilayer structure RRAM. Compared to the single layer (ZrOx) device, the nonlinearity was further improved to 3.94 and 2.42 for potentiation and depression, respectively, and the methodology was confirmed. Additionally, with process parameter optimized, we have fabricated a synaptic RRAM with high linearity weight update, which nonlinearity is 1.3 for potentiation, and 1.82 for depression. In the future, this can be further applied to the neuromorphic computing system to serve as the electronic synapse.
 

摘要 i
Abstract ii
Acknowledgement iii
Table Caption vii
Figure Caption viii
Chapter 1 Introduction 1
1.1 Introduction of Random Access Memory 1
1.2 Volatile Memory 1
1.2.1 DRAM 1
1.2.2 SRAM 1
1.3 Nonvolatile Memory 2
1.3.1 Flash Memory 2
1.3.2 MRAM 3
1.3.3 PCRAM 4
1.3.4 FeRAM 4
1.3.5 RRAM 4
1.4 Neuro-Inspired Computing Using Synaptic RRAM 5
1.4.1 Hardware Acceleration for AI 5
1.4.2 Synaptic RRAM and Crossbar Array 6
Chapter 2 Overview of Synaptic RRAM 14
2.1 Electric Characteristics in RRAM 14
2.1.1 Forming Process 14
2.1.2 Set and Reset Process 15
2.1.3 Operation Methods 15
2.2 Carrier Conduction Mechanisms 15
2.2.1 Ohmic Conduction 15
2.2.2 Space Charge Limit Current 16
2.2.3 Schottky Emission Current 16
2.2.4 Frenkel-Pool Emission 17
2.2.5 Fowler-Nordheim Tunneling 17
2.2.6 Direct Tunneling 17
2.2.7 Hopping Conduction 18
2.3 Resistive Switching Mechanism 18
2.3.1 Filaments Generation and Rupture 19
2.3.2 Oxygen Vacancy Migration 19
2.3.3 Cation Migration 20
2.3.4 Joule Heating 20
2.4 Electrical characteristics of Synaptic RRAM 20
2.4.1 Potentiation and Depression 20
2.4.2 Important parameter for Synaptic RRAM 21
Chapter 3 Experimental Details 30
3.1 Experimental process flow 30
3.2 Substrate Preparation 30
3.3 Binary RRAM Fabrication 30
3.3.1 TaN/Ta/ZrOx/Pt RRAM Fabrication 30
3.4 Synaptic RRAM Fabrication 31
3.4.1 TaN/Ta/ZrOx/TiN Synaptic RRAM device with Annealing 31
3.4.2 TaN/Ta/annealed-ZrOx/AlOx/TiN Synaptic RRAM device 31
3.5 Electrical analysis 32
Chapter 4 From Binary to Synaptic RRAM 35
4.0 Motivation 35
4.1 Experimental results 35
4.1.1 Electrical Property of Binary RRAM: TaN/Ta/ZrOx/Pt and Synaptic RRAM: TaN/Ta/ZrOx/TiN 35
4.1.2 Synapse characteristic of Synaptic RRAM: TaN/Ta/ZrOx/TiN 37
4.1.3 Discussion of different bottom electrode 38
4.1.4 Discussion of Post Deposition Annealing 40
Chapter 5 Enhanced LTP and LTD linearity Using Bilayer Structure 49
5.0 Motivation 49
5.1 Experimental results of synaptic RRAM 50
5.1.1 TaN/Ta/annealed-ZrOx/AlOx/TiN devices 50
5.1.2 Discussion of Filament Mechanism 51
Chapter 6 Conclusions 59
References 60

[1] Chen, Jian. “Selective operation of a multi-state non-volatile memory system in a binary mode.” U.S. Patent No. 6,456,528. 24., 2002.
[2] Bez, R., Camerlenghi, E., Modelli, A., & Visconti, A. “Introduction to the flash memory” Proceedings of the IEEE, 91(4), pp. 489-502, 2003.
[3] Naoi, K. and Simmon, P. “New Materials and New Configurations for Advanced Electrochemical Capacitors” Journal of The Electrochemical Society, vol. 17, pp. 34-37, 2008.
[4] Bandiera, S., & Diney, B. “Magnetic Random Access Memories” Nanomagnetism: Applications and Perspectives, 2016.
[5] Raoux, Simone. “Phase change materials.” Annual Review of Materials Research, vol. 39, pp. 25-48, 2009.
[6] Slaughter, J.M., Chen, E.Y. , Whig, R., Engel, B.N. , Janesky, J., and Tehrani, S. “Magnetic Tunnel Junction Materials for Electronic Applications” JOM (USA), 52(6), pp. 11, 2000.
[7] X. Q. Wei et al., “Universal HSPICE model for chalcogenide based phase change memory elements” Proceedings. 2004 IEEE Computational Systems Bioinformatics Conference, Stanford, CA, pp. 88-91, 2004.
[8] Scott, James F., and Carlos A. Paz De Araujo. “Ferroelectric memories.” Science, 246.4936, pp. 1400-1405, 1989.
[9] Tan, Scott H., et al. “Perspective: Uniform switching of artificial synapses for large-scale neuromorphic arrays.” APL Materials, 6.12, pp. 120901, 2018.
[10] Ligang Gao, et al. “Fully parallel write/read in resistive synaptic array for accelerating on-chip learning’’ Nanotechnology, 26, pp. 45, 2015.
[11] S. Yu, “Introduction to Neuro-Inspired Computing Using Resistive Synaptic Devices.” Neuro-inspired Computing Using Resistive Synaptic Devices. Springer, Cham, pp. 1-15, 2017.
[12] S. Yu, et al. “An Electronic Synapse Device Based on Metal Oxide Resistive Switching Memory for Neuromorphic Computation,” in IEEE Transactions on Electron Devices, vol. 58, no. 8, pp.2729-2737, Aug. 2011.
[13] Hong, XiaoLiang, et al. “Oxide-based RRAM materials for neuromorphic computing.” Journal of materials science, 53.12, pp. 8720-8746, 2018.
[14] S. Yu, “Neuro-inspired computing with emerging nonvolatile memorys,” Proceedings of the IEEE, vol. 106, no. 2, pp. 260-285, Feb. 2018.
[15] Janousch, Markus, et al. “Role of oxygen vacancies in Cr‐doped SrTiO3 for resistance‐change memory.” Advanced materials, 19.17, pp. 2232-2235, 2007.
[16] A. Belmonte et al., “Analysis of the Excellent Memory Disturb Characteristics of a Hourglass-Shaped Filament in Al2O3/Cu-Based CBRAM Devices,” IEEE Transactions on Electron Devices, vol. 62, no. 6, pp. 2007-2013, June 2015.
[17] Lanza M. “A Review on Resistive Switching in High-k Dielectrics: A Nanoscale Point of View Using Conductive Atomic Force Microscope.” Materials, 7(3), pp. 2155-2182, 2014.
[18] Murgatroyd, P. N. “Theory of space-charge-limited current enhanced by Frenkel effect.” Journal of Physics D: Applied Physics, 3.2, pp. 151, 1970.
[19] Matsumura, M., Jinde, Y., Akai, T. and Kimura T. “Analysis of Current-Voltage Characteristics of Organic Electroluminescent Devices on the Basis of Schottky Emission Mechanism.” Japanese Journal of Applied Physics, 35(11R), pp. 5735, 1996.
[20] Lim, E. W. and Ismail, R. “Conduction Mechanism of Valence Change Resistive Switching Memory: A Survey” Electronics, 4(3), pp. 586-613, 2015.
[21] Chiu, F. C. “A Review on Conduction Mechanisms in Dielectric Films.” Advances in Materials Science and Engineering, 2014.
[22] Kim, J. H., Choi, J. B., Shin, B. J. and Park, K. H. “An empirical model for charge leakage through oxide–nitride–oxide interpoly dielectric in stacked-gate flash memory devices.” Semiconductor Science and Technology, 18(2), pp. 158, 2003.
[23] Yang, L., Kuegeler, C., Szot, K., Ruediger, A. and Waser R. “The influence of copper top electrodes on the resistive switching effect in thin TiO2 films studied by conductive atomic force microscopy.” Applied Physics Letters, 95(1), pp. 013109, 2009.
[24] Yu, Shimeng, et al. “Improved uniformity of resistive switching behaviors in HfO2 thin films with embedded Al layers.” Electrochemical and Solid-State Letters, 13.2, pp. H36-H38, 2010.
[25] Sawa, A. “Resistive switching in transition metal oxides.” Materials today, 11(6), pp. 28-36, 2008.
[26] Szot, K., Dittmann R., Speier, W. and Waser, R. “Nanoscale resistive switching in SrTiO3 thin films.” Physical status solidi (RRL)-Rapid Research Letter, 1(2), pp. R86-R88, 2007
[27] H. -. P. Wong et al., "Metal–Oxide RRAM," in Proceedings of the IEEE, vol. 100, no. 6, pp. 1951-1970, June 2012.
[28] Tsuruoka, T., Terabe, K., Hasegawa, T. and Aono M. “Forming and switching mechanisms of a cation-migration-based oxide resistive memory.” Nanotechnology, 21(42), pp. 425205, 2010.
[29] U. Russo, D. Ielmini, C. Cagli and A. L. Lacaita, “Self-Accelerated Thermal Dissolution Model for Reset Programming in Unipolar Resistive-Switching Memory (RRAM) Devices,” in IEEE Transactions on Electron Devices, vol. 56, no. 2, pp. 193-200, Feb. 2009.
[30] Chen, Pai-Yu, et al. “Mitigating effects of non-ideal synaptic device characteristics for on-chip learning.” Proceedings of the IEEE/ACM International Conference on Computer-Aided Design. IEEE Press, 2015.
[31] I.-T. Wang, C.-C. Chang, L.-W. Chiu, T. Chou, and T.-H. Hou, “3D Ta/TaOx/TiO2/Ti synaptic array and linearity tuning of weight update for hardware neural network applications,” Nanotechnology, vol. 27, no. 8, pp. 365204, 2016.
[32] Wei Wu, et al. “A Methodology to Improve Linearity of Analog RRAM for Neuromorphic Computing”, 103-104, 2018 IEEE Symposium on VLSI Technology, Honolulu, HI, USA
[33] Tsung-Ling Tsai, et al. “Unipolar resistive switching behaviors and mechanism in an annealed Ni/ZrO2/TaN memory device” J. Phys. D: Appl. Phys., 48, pp. 035108, 2015.
[34] Umesh Chand, et al. “Mechanism of High Temperature Retention Property (up to 200 °C) in ZrO2-Based Memory Device with Inserting a ZnO Thin Layer” IEEE Electron Device Letters, vol. 35, no. 10, pp. 1019-1021, Oct. 2014.
[35] L. Goux, et al. “Evidences of oxygen-mediated resistive-switching mechanism in TiN/HfO2/Pt cells” Appl. Phys. Lett., 97, pp. 243509, 2010.
[36] Jonggi Kim, et al. “Effect of thermal annealing on resistance switching characteristics of Pt/ZrO2/TiN stacks” Materials Chemistry and Physics, 142(2-3), pp. 608-613, 2013.
[37] 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.
[38] Huang, Chun-Yang, et al. “Switching mechanism of double forming process phenomenon in ZrOx/HfOy bilayer resistive switching memory structure with large endurance.” Applied Physics Letters, 104.6, pp. 062901, 2014.
[39] Sohyeon Kim, et al. “Engineering synaptic characteristic of TaOx/HfO2 bi-layered resistive switching device” Nanotechnology, 29, pp. 415204, 2018.