Organic Memcapacitors Based On Charge Trapping Mechanism

Due to the development of artificial intelligence,the demand for information processing is increasing rapidly.Meanwhile,considering energy and environmental issues,people require the efficiency of a computer system to be higher and higher.Since the memory and processing units of a traditional computer are separated,there is an upper limit to its operating speed and energy efficiency.As the size and performance of a transistor gradually approach to its physical limits,it is necessary to improve the structure of a computer system.Based on the study of the nervous system in a human brain,new memory devices,represented by memristors and memcapacitors,have started to gain more and more attention.Thanks to the ability of a synapse to store and process information at the same time,the human brain can perform complex tasks with very low energy consumption.Memristors/memcapacitors are ideal devices for synaptic simulation because they have both the ability of resistance/capacitance transformation and the ability of non-volatile storage.In this paper,memcapacitors are chosen because of their theoretical advantages,such as lower power consumption and no sneak-path current,compared to memristors.Organic materials have more advantages over inorganic materials in terms of flexibility,biocompatibility and so on.Therefore,the main research goals of this paper are to design and fabricate a stable memcapacitor using organic materials and to investigate its mechanism and potential applications.The details are as follows:(ⅰ)Design,fabrication,and characterization of an organic memcapacitor.Inspired by an organic semiconductor field effect transistor,a charge-trapping-based memcapacitor,whose structure was Cu/Pentacene/PS/SiO2/Si(n++),had been successfully fabricated.With the lower electrode grounded and the upper electrode biased,the device always remained in its smallest capacitance state when the applied voltage was lower than the threshold voltage.While the applied voltage was higher than the threshold voltage,the device capacitance could gradually rise from the minimum to the maximum value.If the device was pre-processed with a relatively high bias voltage,the threshold voltage of the device would be shifted.The device state could be preserved even after the bias voltage was removed.By varying the ratio of the area of the copper electrode layer to that of the pentacene layer,the switching ratio of the memcapacitor was significantly increased,reaching a maximum of about 34.(ⅱ)Mechanism study of the organic memcapacitors.By testing the pentacene-removed device and comparing the theoretical values with the experimental values,the capacitance transformation was caused by the conversion between the depletion and accumulation states of holes in the pentacene layer.The memory characteristic was determined to come from the charge trapping and storge in the PS material by testing a non-PS-layer device.In addition,the AC response process in the pentacene-naked device part was analyzed by using admittance model,and the trend of the linear region in the C-V loop was fitted well.The state variable of the organic memcapacitors was also be shown.(ⅲ)Application practice for the organic memcapacitors.An artificial neural network for performing simple logic operations is constructed,in which a component consisting of two organic memcapacitors is used to simulate one synaptic weight.The network successfully implemented three different logic operations with low energy consumption(～100 pJ).In summary,an organic memcapacitor with a large switching ratio and good retention was designed and fabricated.By changing the threshold voltage of the device,the readout value can be continuously adjusted.This device fills the lack of research on organic memcapacitors and has potential on the field of artificial neural networks.