The 3D Monolithically Integrated Hardware Based Neural System with Enhanced Memory Window of the Volatile and Non-Volatile Devices

被引:2
|
作者
Jeon, Yu-Rim [1 ]
Seo, Donguk [2 ]
Lee, Yoonmyung [2 ]
Akinwande, Deji [1 ]
Choi, Changhwan [3 ]
机构
[1] Univ Texas Austin, Dept Elect & Comp Engn, Austin, TX 78172 USA
[2] Sungkyunkwan Univ, Dept Elect & Comp Engn, Suwon 16419, South Korea
[3] Hanyang Univ, Div Mat Sci & Engn, Seoul 04763, South Korea
基金
新加坡国家研究基金会;
关键词
3D neuromorphic system; CMOS integration; convolutional neural network; high k metal oxide; RRAM; synaptic device; transistor; wafer bonding;
D O I
10.1002/advs.202402667
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
3D neuromorphic hardware system is first demonstrated in neuromorphic application as on-chip level by integrating array devices with CMOS circuits after wafer bonding (WB) and interconnection process. The memory window of synaptic device is degraded after WB and 3 Dimesional (3D) integration due to process defects and thermal stress. To address this degradation, Ag diffusion in materials of Ta2O5 and HfO2 is studied in a volatile memristor, furthermore, the interconnection and gate metal Ru are investigated to reduce defective traps of gate interface in non-volatile memory devices. As a result, a memory window is improved over 106 in both types of devices. Improved and 3D integrated 12 x 14 array devices are identified in the synaptic characteristics according to the change of the synaptic weight from the interconnected Test Element Group (TEG) of the Complementary Metal Oxide Semiconductor (CMOS) circuits. The trained array devices present recognizable image of letters, achieving an accuracy rate of 92% when utilizing a convolutional neural network, comparing the normalized accuracy of 93% achieved by an ideal synapse device. This study proposes to modulate the memory windows up to 106 in an integrated hardware-based neural system, considering the possibility of device degradation in both volatile and non-volatile memory devices demonstrated by the hardware neural system. Monolithically 3D integrated on-chip neural system is demonstrated using CMOS circuits, considering the degradation of memory window after wafer bonding and interconnection processes. Ag diffusion in bi-layer is studied in volatile memristors. Additionally, interconnection and gate metal are investigated to reduce defective traps in non-volatile transistor. A memory window is improved over 106 in both types of devices. image
引用
收藏
页数:10
相关论文
共 50 条
  • [21] Technology CAD of non-volatile SONOS memory devices
    Chakraborty, P.
    Mahato, S. S.
    Maiti, T. K.
    Saha, S.
    Maiti, C. K.
    PROCEEDINGS OF THE 2007 INTERNATIONAL WORKSHOP ON THE PHYSICS OF SEMICONDUCTOR DEVICES: IWPSD-2007, 2007, : 164 - +
  • [22] Nanoimprint for future non-volatile memory and logic devices
    Meier, M.
    Nauenheim, C.
    Gilles, S.
    Mayer, D.
    Kuegeler, C.
    Waser, R.
    MICROELECTRONIC ENGINEERING, 2008, 85 (5-6) : 870 - 872
  • [23] Resistive non-volatile memory devices (Invited Paper)
    Waser, Rainer
    MICROELECTRONIC ENGINEERING, 2009, 86 (7-9) : 1925 - 1928
  • [24] Non-volatile hybrid memory devices with excellent reliability
    Chung, Dae Sung
    Kang, Il
    Kwon, Soon-Ki
    Kim, Yun-Hi
    RSC ADVANCES, 2013, 3 (32) : 13156 - 13162
  • [25] Ferroelectric polymers for non-volatile memory devices: a review
    Li, Huilin
    Wang, Ruopeng
    Han, Su-Ting
    Zhou, Ye
    POLYMER INTERNATIONAL, 2020, 69 (06) : 533 - 544
  • [26] SCALING DOWN MNOS NON-VOLATILE MEMORY DEVICES
    YATSUDA, Y
    HAGIWARA, T
    MINAMI, SI
    KONDO, R
    UCHIDA, K
    UCHIUMI, K
    JAPANESE JOURNAL OF APPLIED PHYSICS, 1982, 21 (01) : 85 - 90
  • [27] SINGLE SHOT 3D SHAPE MEASUREMENT OF NON-VOLATILE DATA STORAGE DEVICES
    Balasubramaniam, Badrinath
    Li, Beiwen
    PROCEEDINGS OF ASME 2023 18TH INTERNATIONAL MANUFACTURING SCIENCE AND ENGINEERING CONFERENCE, MSEC2023, VOL 2, 2023,
  • [28] Non-Volatile FPGAs based on Spintronic Devices
    Goncalves, Olivier
    Prenat, Guillaume
    Di Pendina, Gregory
    Dieny, Bernard
    2013 50TH ACM / EDAC / IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2013,
  • [29] Analog architectures for neural network acceleration based on non-volatile memory
    Xiao, T. Patrick
    Bennett, Christopher H.
    Feinberg, Ben
    Agarwal, Sapan
    Marinella, Matthew J.
    APPLIED PHYSICS REVIEWS, 2020, 7 (03):
  • [30] Exploring the Vulnerability of CMPs to Soft Errors with 3D Stacked Non-Volatile Memory
    Sun, Guangyu
    Kursun, Eren
    Rivers, Jude A.
    Xie, Yuan
    2011 IEEE 29TH INTERNATIONAL CONFERENCE ON COMPUTER DESIGN (ICCD), 2011, : 366 - 372