1 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com Doping Modulated Carbon Nanotube Synapstors for a Spike Neuromorphic Module Alex Ming Shen, Kyunghyun Kim, Andrew Tudor, Dongwon Lee, and Yong Chen* amplitudes are usually less than 1 nA, which enables the spikes to be processed with extremely low power consump- tion in a neural network. Due to the synaptic plasticity, the amplitude of a PSC triggered via a synapse can be increased or decreased by the spikes, which establishes the foundation for learning and memory in a neural network. Silicon (Si)-based circuits have been utilized to emulate neural networks, [1,3–5] but the Si circuits consumed consider- ably more energy than a biological network and were unable to be integrated at a scale comparable with the biological neural network. [6–11] There have been extensive attempts to emu- late the functions of synapses and neurons utilizing various electronic devices, such as floating gate silicon transistors, [12] nanoparticle organic transistors, [13] resistive switches, [14] memristors, [15,16] phase change memory, [17] and carbon nanotube (CNT) transistors. [2,18–20] However, the devices lack the analog memory, [12–14,21] plasticity, [10] or the function for spikes to trigger PSCs for spike signal processing. [6,15,17] Spikes can trigger PSCs in other synaptic devices, but the PSC amplitudes range between 10 -3 –10 -6 A, [14,15,21] resulting in significantly higher power consumptions of the devices than those of biological synapses. Due to their nanoscale dia- meter, high carrier mobility, and sensitivity to environmental changes, CNTs have been extensively explored as an alterna- tive to Si in nano-electronic devices. [22] Large-scale devices DOI: 10.1002/smll.201402528 A doping-modulated carbon nanotube (CNT) electronic device, called a “synapstor,” emulates the function of a biological synapse. The CNT synapstor has a field-effect transistor structure with a random CNT network as its channel. An aluminium oxide (Al 2 O 3 ) film is deposited over half of the CNT channel in the synapstor, converting the covered part of the CNT from p-type to n-type, forming a p–n junction in the CNT channel and increasing the Schottky barrier between the n-type CNT and its metal contact. This scheme significantly improves the postsynaptic current (PSC) from the synapstor, extends the tuning range of the plasticity, and reduces the power consumption of the CNT synapstor. A spike neuromorphic module is fabricated by integrating the CNT synapstors with a Si-based “soma” circuit. Spike parallel processing, memory, and plasticity functions of the module are demonstrated. The module could potentially be integrated and scaled up to emulate a biological neural network with parallel high-speed signal processing, low power consumption, memory, and learning capabilities. Synapstors A. M. Shen, Dr. K. Kim, A. Tudor, D. Lee, Prof. Y. Chen Department of Mechanical and Aerospace Engineering California Nano Systems Institute University of California Los Angeles, California 90095, USA E-mail: yongchen@seas.ucla.edu 1. Introduction A neural network can process massive amounts of informa- tion at an extremely high speed with low power consumption. The signals, in the format of millisecond-long potential spikes, route through billions of neurons via trillions of synapses – the junctions between neurons. A spike from a presynaptic neuron can trigger a dynamic analog postsynaptic current (PSC) in a postsynaptic neuron via a synapse. Spikes from thousands of presynaptic neurons can trigger multiple PSCs via multiple synapses, connected in parallel, in a postsyn- aptic neuron. When the total PSC collected by the soma of the postsynaptic neuron drives its potential above a threshold value, a spike will be fired from the neuron, which in turn triggers PSCs in other neurons. [1,2] The synapses process mas- sive spikes in a parallel mode in a neural network. The PSC small 2014, DOI: 10.1002/smll.201402528