Fabrication And Performance Research Of CNT Synapses TFT

Along with the rapid development of science and technology, the problems that people are facing are becoming more and more complex. Also the amount of data that people need to handle is rising in the form of exponential growth. Therefore people's requirement on the performance of computer also is creasing rapidly. So computers need to deal with more and more problems with unclear data structure and logic, such as image identification, voice recognition and so on.However, traditional computers that based on the von Neumann's architecture are very weak and inefficient when dealing with these kinds of problems. Therefore, in this case, the concept of a computer that can mimic the behavior of human brain's computing has attracted wide attention all over the world. This new type of brain-like computer can quickly grasp the inherent laws of a big amount of data by repeated learning, which may effectively solve the problem that traditional computer cannot. So, it is remarkably promising and widely considered as the future of computer.Neuron is the basic unit of brain. And synapse, which connects neurons, is the key of realizing the function of signal processing and memory. Similarly, as a basic unit of brain-like computation, synaptic transistor surely is the underlying base of brain-like computer.In this work, thin film transistor based on carbon nanotube network is deeply researched,and a new type of carbon nanotube transistor with similar function of synapse is realized. It is a kind of junctionless thin film transistor without PN junction in drain and source. The benefit of this structure includes the reduced short channel effect, lower threshold voltage,and simple manufacture process. The electric-double-layer gate dielectric is applied. The phosphorus doped silicon dioxide films is made by room temperature PECVD. So the membrane is embedded with a large number of ions, which is the reason of electric-double-layer effect. This thin film transistors based on carbon nanotube network can mimic the basic functions of biological synapse, such as excited postsynaptic currents and memory etc. Because of these characteristics, the carbon nanotube transistor has the advantages of high open current, high on-off ratio, small noise, simple preparation, large memory window, and so on.