Hf1-xZrxO2忆阻器的研究与应用

基于掺杂金属氧化物基的忆阻器因其出色的开关性能和多比特信息存储潜力，在人工神经网络（ANN）硬件系统实现中脱颖而出，成为备受瞩目的候选者。然而，考虑到实际应用的性能和制造要求，如开关行为和CMOS工艺的兼容性，其广泛应用仍面临挑战。本论文深入研究了通过共溅射技术制造的非晶Zr掺杂HfO2（HZO）忆阻器，其性能相较于传统的单氧化物忆阻器有了显著提升。值得一提的是，这种材料的选择不仅与CMOS工艺兼容，还满足了工业界对忆阻器性能的严苛要求。与对照的HfO₂忆阻器相比，优化后的HZO忆阻器展现出更低的操作电压和更快的切换速度等优势。此外，本研究还提出了一种创新的电压脉冲幅值逐渐增加的编程方案，成功实现了高线性度的模拟态调谐。通过模拟评估，本研究进一步探索了HZO忆阻器在自组织映射（SOM）网络中的应用，特别是在Fashion MNIST数据库分类任务中的表现。结果表明，在较短的训练周期内，HZO忆阻器就实现了高达92%的识别准确性，充分展示了其在神经网络计算领域的巨大潜力。本研究针对HZO忆阻器的探索，不仅为开发CMOS工艺兼容的忆阻器提供了坚实的理论基础，还为未来高性能存储和神经网络计算中的实际应用铺设了道路。

Doped-Metal oxide-based memristors, with potential for improved switching performance and capability for multi-bit information storage, are attractive candidates in the implementation of artificial neural network (ANN) hardware systems. However, performance and process considerations such as switching behavior and CMOS process compatibility remain a challenge. In this thesis, we investigated amorphous Zr-doped HfO₂ (HZO) memristors fabricated via co-sputtering approach that boasts performance boost over conventional mono-oxides memristors. At the same time, the choice of material is CMOS process compatible for industry adoption. HZO memristors with optimized stoichiometry exhibit reduced switching voltages and faster switching as compared to control HfO₂ memristors. Concurrently, it is shown that high linearity analog states tuning is achievable via a programming scheme that utilizes voltage pulses with increasing amplitudes. We further show via simulation evaluation that HZO memristors implemented in a self-organizing-map (SOM) network for Fashion MNIST database classification, achieving accuracy of 92% with short training cycles. Our study thus paves a potential pathway for further development of CMOS process compatible HZO memristors for use in future storage and computing applications.

[1] ZIDAN M A, STRACHAN J P, LU W D. The future of electronics based on memristive systems [J]. Nature Electronics, 2018, 1(1): 22-9.
[2] CHUA L. Memristor-The missing circuit element [J]. IEEE Transactions on Circuit Theory, 1971, 18(5): 507-19.
[3] STRUKOV D B, SNIDER G S, STEWART D R, et al. The missing memristor found [J]. Nature, 2008, 453(7191): 80-3.
[4] LEE M-J, LEE C B, LEE D, et al. A fast, high-endurance and scalable non-volatile memory device made from asymmetric Ta2O5−x/TaO2−x bilayer structures [J]. Nature Materials, 2011, 10(8): 625-30.
[5] GERGEL-HACKETT N, HAMADANI B, DUNLAP B, et al. A Flexible Solution-Processed Memristor [J]. IEEE Electron Device Letters, 2009, 30(7): 706-8.
[6] BORGHETTI J, SNIDER G S, KUEKES P J, et al. ‘Memristive’ switches enable ‘stateful’ logic operations via material implication [J]. Nature, 2010, 464(7290): 873-6.
[7] GOVOREANU B, KAR G S, CHEN Y Y, et al. 10×10nm2 Hf/HfOx crossbar resistive RAM with excellent performance, reliability and low-energy operation; proceedings of the 2011 International Electron Devices Meeting, F 5-7 Dec. 2011, 2011 [C].
[8] SHAN X, ZHAO C, WANG X, et al. Plasmonic Optoelectronic Memristor Enabling Fully Light-Modulated Synaptic Plasticity for Neuromorphic Vision [J]. Advanced Science, 2022, 9(6): 2104632.
[9] WU F, CAO P, PENG Z, et al. Memristor Based on TiOx/Al2O3 Bilayer as Flexible Artificial Synapse for Neuromorphic Electronics [J]. IEEE Transactions on Electron Devices, 2022, 69(1): 375-9.
[10] LIU L, XIONG W, LIU Y, et al. Designing High-Performance Storage in HfO2/BiFeO3 Memristor for Artificial Synapse Applications [J]. Advanced Electronic Materials, 2020, 6(2): 1901012.
[11] LIU T Y, YAN T H, SCHEUERLEIN R, et al. A 130.7mm2 2-layer 32Gb ReRAM memory device in 24nm technology; proceedings of the 2013 IEEE International Solid-State Circuits Conference Digest of Technical Papers, F 17-21 Feb. 2013, 2013 [C].
[12] WANG Z, LI C, SONG W, et al. Reinforcement learning with analogue memristor arrays [J]. Nature Electronics, 2019, 2(3): 115-24.
[13] CHEN J, PAN W Q, LI Y, et al. High-Precision Symmetric Weight Update of Memristor by Gate Voltage Ramping Method for Convolutional Neural Network Accelerator [J]. IEEE Electron Device Letters, 2020, 41(3): 353-6.
[14] HUANG J, YANG S, TANG X, et al. Flexible, Transparent And Wafer‐scale Artificial Synapse Array Based on TiO x /Ti 3 C 2 T x Film for Neuromorphic Computing [J]. Advanced Materials, 2023, 35.
[15] LI C, HU M, LI Y, et al. Analogue signal and image processing with large memristor crossbars [J]. Nature Electronics, 2018, 1(1): 52-9.
[16] YAO P, WU H, GAO B, et al. Face classification using electronic synapses [J]. Nature Communications, 2017, 8(1): 15199.
[17] DESHPANDE V, NAIR K S, HOLZER M, et al. CMOS back-end-of-line compatible ferroelectric tunnel junction devices [J]. Solid-State Electronics, 2021, 186: 108054.
[18] GRENOUILLET L, FRANCOIS T, COIGNUS J, et al. Nanosecond laser anneal (NLA) for Si-implanted HfO2 ferroelectric memories integrated in back-end of line (BEOL); proceedings of the 2020 IEEE Symposium on VLSI Technology, F, 2020 [C]. IEEE.
[19] HUR J, LUO Y C, TASNEEM N, et al. Ferroelectric Hafnium Zirconium Oxide Compatible With Back-End-of-Line Process [J]. IEEE Transactions on Electron Devices, 2021, 68(7): 3176-80.
[20] KIM K-H, OH S, FIAGBENU M M A, et al. Scalable CMOS back-end-of-line-compatible AlScN/two-dimensional channel ferroelectric field-effect transistors [J]. Nature Nanotechnology, 2023.
[21] CHEN H, LI L, WANG J, et al. Performance Optimization of Atomic Layer Deposited HfOx Memristor by Annealing With Back-End-of-Line Compatibility [J]. IEEE Electron Device Letters, 2022, 43(7): 1141-4.
[22] ZAHEER M, BACHA A-U-R, NABI I, et al. All Solution-Processed Inorganic, Multilevel Memristors Utilizing Liquid Metals Electrodes Suitable for Analog Computing [J]. ACS Omega, 2022, 7(45): 40911-9.
[23] LAN J, ZHU Q, ZHANG Y, et al. Zinc-Alloyed HFO2 Synaptic RRAM with Operating Voltage and Switching Energy Enhancement; proceedings of the 2022 China Semiconductor Technology International Conference (CSTIC), F 20-21 June 2022, 2022 [C].
[24] GUO T, TAN T, LIU Z. Enhanced resistive switching behaviors of HfO2:Cu film with annealing process [J]. Vacuum, 2015, 114: 78-81.
[25] TAN T, GUO T, CHEN X, et al. Impacts of Au-doping on the performance of Cu/HfO2/Pt RRAM devices [J]. Applied surface science, 2014, 317: 982-5.
[26] CHOI B J, TORREZAN A C, NORRIS K J, et al. Electrical performance and scalability of Pt dispersed SiO2 nanometallic resistance switch [J]. Nano letters, 2013, 13(7): 3213-7.
[27] BOUAZIZ J, ROMEO P R, BABOUX N, et al. Huge reduction of the wake-up effect in ferroelectric HZO thin films [J]. ACS Applied Electronic Materials, 2019, 1(9): 1740-5.
[28] CAO R, SONG B, SHANG D, et al. Improvement of Endurance in HZO-Based Ferroelectric Capacitor Using Ru Electrode [J]. IEEE Electron Device Letters, 2019, 40(11): 1744-7.
[29] HU Y, RABELO M, KIM T, et al. Ferroelectricity Based Memory Devices: New-Generation of Materials and Applications [J]. Transactions on Electrical and Electronic Materials, 2023.
[30] BÉGON‐LOURS L, HALTER M, PUGLISI F M, et al. Scaled, Ferroelectric Memristive Synapse for Back‐End‐of‐Line Integration with Neuromorphic Hardware [J]. Advanced Electronic Materials, 2022, 8(6): 2101395.

[31] SUNG C, HWANG H, YOO I K. Perspective: A review on memristive hardware for neuromorphic computation [J]. Journal of Applied Physics, 2018, 124(15).

[32] ZHAO Z, YAN X. Ferroelectric memristor based on Hf0. 5Zr0. 5O2 thin film combining memristive and neuromorphic functionalities [J]. physica status solidi (RRL)–Rapid Research Letters, 2020, 14(9): 2000224.

[33] LAN J, LI Z, CHEN Z, et al. Improved Performance of HfxZnyO-Based RRAM and its Switching Characteristics down to 4 K Temperature [J]. Advanced Electronic Materials, 2023, 9(3): 2201250.

[34] ZHANG L, HUANG H, YE C, et al. Exploration of highly enhanced performance and resistive switching mechanism in hafnium doping ZnO memristive device [J]. Semiconductor Science and Technology, 2018, 33(8).

[35] RAJARATHINAM S, GANGULY U, VENKATARAMANI N. Impact of oxygen partial pressure on resistive switching characteristics of PLD deposited ZnFe2O4 thin films for RRAM devices [J]. Ceram Int, 2022, 48(6): 7876-84.

[36] LI X Y, WANG Y L, LIU W F, et al. Study of oxygen vacancies' influence on the lattice parameter in ZnO thin film [J]. Mater Lett, 2012, 85: 25-8.

[37] SHARMA U, SINGH C, VARMA V M, et al. Preparation and characterization of hafnium-zirconium oxide ceramics as a CMOS compatible material for non-volatile memories [J]. B Mater Sci, 2023, 46(2).

[38] RYU S W, CHO S, PARK J, et al. Effects of ZrO2 doping on HfO2 resistive switching memory characteristics [J]. Appl Phys Lett, 2014, 105(7).

[39] XU H T, WU C J, ZHAO X H, et al. Improved Resistance Switching Stability in Fe-Doped ZnO Thin Films Through Pulsed Magnetic Field Annealing [J]. Nanoscale Res Lett, 2017, 12.

[40] KU B, ABBAS Y, SOKOLOV A S, et al. Interface engineering of ALD HfO2-based RRAM with Ar plasma treatment for reliable and uniform switching behaviors [J]. J Alloy Compd, 2018, 735: 1181-8.

[41] WANG T Y, YU L J, CHEN L, et al. Atomic layer deposited Hf0.5Zr0.5O2-based flexible RRAM; proceedings of the 2017 IEEE 12th International Conference on ASIC (ASICON), F 25-28 Oct. 2017, 2017 [C].

[42] WU Z, ZHU J, ZHOU Y, et al. Bipolar Resistive Switching Properties of Hf0.5Zr0.5O2 Thin Film for Flexible Memory Applications [J]. physica status solidi (a), 2018, 215(1): 1700396.

[43] ISMAIL M, BATOOL Z, MAHMOOD K, et al. Resistive switching characteristics and mechanism of bilayer HfO2/ZrO2 structure deposited by radio-frequency sputtering for nonvolatile memory [J]. Results Phys, 2020, 18.

[44] ISMAIL M, MAHATA C, KWON O, et al. Neuromorphic Synapses with High Switching Uniformity and Multilevel Memory Storage Enabled through a Hf-Al-O Alloy for Artificial Intelligence [J]. Acs Applied Electronic Materials, 2022, 4(3): 1288-300.

[45] JERRY M, CHEN P-Y, ZHANG J, et al. Ferroelectric FET analog synapse for acceleration of deep neural network training [J]. 2017 IEEE International Electron Devices Meeting (IEDM), 2017: 6.2.1-6.2.4.

[46] PARK S, SHERI A, KIM J, et al. Neuromorphic speech systems using advanced ReRAM-based synapse; proceedings of the 2013 IEEE International Electron Devices Meeting, F 9-11 Dec. 2013, 2013 [C].

[47] JO S H, CHANG T, EBONG I, et al. Nanoscale Memristor Device as Synapse in Neuromorphic Systems [J]. Nano Letters, 2010, 10(4): 1297-301.

[48] KOHONEN T. Essentials of the self-organizing map [J]. Neural Networks, 2013, 37: 52-65.

[49] KOHONEN T. The self-organizing map [J]. Proceedings of the IEEE, 1990, 78(9): 1464-80.

[50] MILJKOVIĆ D. Brief review of self-organizing maps; proceedings of the 2017 40th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), F 22-26 May 2017, 2017 [C].

[51] KOHONEN T. Exploration of very large databases by self-organizing maps; proceedings of the Proceedings of International Conference on Neural Networks (ICNN'97), F 12-12 June 1997, 1997 [C].

[52] VESANTO J, ALHONIEMI E. Clustering of the self-organizing map [J]. IEEE Transactions on Neural Networks, 2000, 11(3): 586-600.

[53] LIN Y H, WANG C H, LEE M H, et al. Performance Impacts of Analog ReRAM Non-ideality on Neuromorphic Computing [J]. IEEE Transactions on Electron Devices, 2019, 66(3): 1289-95.

[54] KOHONEN T, OJA E, SIMULA O, et al. Engineering applications of the self-organizing map [J]. Proceedings of the IEEE, 1996, 84(10): 1358-84.

[55] BAÇÃO F, LOBO V, PAINHO M. Self-organizing Maps as Substitutes for K-Means Clustering; proceedings of the Computational Science – ICCS 2005, Berlin, Heidelberg, F 2005//, 2005 [C]. Springer Berlin Heidelberg.

[56] THAI L H, HAI T S, THUY N T J I J O I T, et al. Image classification using support vector machine and artificial neural network [J]. 2012, 4(5): 32-8.

[57] BALA R, KUMAR D J I J C I R. Classification using ANN: A review [J]. 2017, 13(7): 1811-20.

[58] THOMAS A. Memristor-based neural networks [J]. Journal of Physics D: Applied Physics, 2013, 46(9): 093001.

[59] WANG Z, JOSHI S, SAVEL’EV S, et al. Fully memristive neural networks for pattern classification with unsupervised learning [J]. Nature Electronics, 2018, 1(2): 137-45.

[60] KIM H, SAH M P, YANG C, et al. Neural Synaptic Weighting With a Pulse-Based Memristor Circuit [J]. IEEE Transactions on Circuits and Systems I: Regular Papers, 2012, 59(1): 148-58.

[61] XIAO H, RASUL K, VOLLGRAF R. Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms [J]. ArXiv, 2017, abs/1708.07747.

[62] LIU Y, WEISBERG R H, MOOERS C N K. Performance evaluation of the self-organizing map for feature extraction [J]. J Geophys Res, 2006, 111(C5).

[63] GREESHMA K, SREEKUMAR K J I J O R T, ENGINEERING. Hyperparameter optimization and regularization on Fashion-MNIST classification [J]. 2019, 8(2): 3713-9.

[64] MANGIAMELI P, CHEN S K, WEST D. A comparison of SOM neural network and hierarchical clustering methods [J]. European Journal of Operational Research, 1996, 93(2): 402-17.

[65] E I KNUDSEN, S LAC A, ESTERLY S D. Computational Maps in the Brain [J]. Annu Rev Neurosci, 1987, 10(1): 41-65.

[66] GHASEMINEZHAD M H, KARAMI A. A novel self-organizing map (SOM) neural network for discrete groups of data clustering [J]. Applied Soft Computing, 2011, 11(4): 3771-8.