In-memory computing has carried out calculations in situ within each memory unit and its main power consumption comes from data writ-ing and erasing.Further improvements in the energy efficiency of in-memory computing...In-memory computing has carried out calculations in situ within each memory unit and its main power consumption comes from data writ-ing and erasing.Further improvements in the energy efficiency of in-memory computing require memory devices with sub-femto-Joule energy consumption.Floating gate memory devices based on two-dimensional(2D)material heterostructures have outstanding char-acteristics such as non-volatility,multi-bit storage,and low opera-tion energy,suitable for application in in-memory computing chips.Here,we report a floating gate memory device based on a WSe 2/h-BN/Multilayer-graphene/h-BN heterostructure,the energy consump-tion of which is in sub-femto Joule(0.6 fJ)per operation for pro-gram/erase,and the read power consumption is in the tens of femto Watt(60 fW)range.We show a Hopfield neural network composed of WSe 2/h-BN/Multilayer-graphene/h-BN heterostructure floating gate memory devices,which can recall the original patterns from incorrect patterns.These results shed light on the development of future com-pact and energy-efficient hardware for in-memory computing sys-tems.展开更多
基金This work was supported in part by the National Key Research and Development Program of China under Grant 2020YFB2008802 and Grant 2020YFB2008803in part by the Fundamental Research Funds for the Cen-tral Universities under Grant WK2100230020in part by the USTC Center for Micro and Nanoscale Research and Fabrication,in part by the USTC In-stitute of Advanced Technology,and in part by the CAS Key Laboratory of Wireless-Optical Communications.
文摘In-memory computing has carried out calculations in situ within each memory unit and its main power consumption comes from data writ-ing and erasing.Further improvements in the energy efficiency of in-memory computing require memory devices with sub-femto-Joule energy consumption.Floating gate memory devices based on two-dimensional(2D)material heterostructures have outstanding char-acteristics such as non-volatility,multi-bit storage,and low opera-tion energy,suitable for application in in-memory computing chips.Here,we report a floating gate memory device based on a WSe 2/h-BN/Multilayer-graphene/h-BN heterostructure,the energy consump-tion of which is in sub-femto Joule(0.6 fJ)per operation for pro-gram/erase,and the read power consumption is in the tens of femto Watt(60 fW)range.We show a Hopfield neural network composed of WSe 2/h-BN/Multilayer-graphene/h-BN heterostructure floating gate memory devices,which can recall the original patterns from incorrect patterns.These results shed light on the development of future com-pact and energy-efficient hardware for in-memory computing sys-tems.
