Brain‐inspired neuromorphic computing is a new paradigm that holds great potential to overcome the intrinsic energy and speed issues of traditional von Neumann based computing architecture. With the ability to perform vector‐matrix multiplications and flexible tunable conductance, the memristor crossbar array (CBA) structure is one of the most promising candidates to realize neural cognitive systems. The boom in the development of memristive synapses and neurons has propelled the developments of artificial neural networks (ANNs) to emulate the highly hierarchically organized network of human brain in the past decade. To achieve this, realizing large scale, high‐density memristive CBAs is a prerequisite to constructing complex ANNs. Herein, the stringent requirements in device performance and array parameters for hardware ANNs are analyzed, and the efforts in addressing the associated challenges are discussed. Recent progress on the experimental demonstration of neuromorphic computing systems (NCSs) is presented. Recommendations for further performance optimization at the device, circuit, and algorithm levels are proposed. This Report serves as a guide for the hardware implementation of NCS based on large‐scale CBAs.
19 September 2019 in Advanced Electronic Materials
Realization of memristors capable of storing and processing data on flexible substrates is a key enabling technology toward “system‐on‐plastics”. Recent advancements in printing techniques show enormous potential to overcome the major challenges of the current manufacturing processes that require high temperature and planar topography, which may radically change the system integration approach on flexible substrates. However, fully printed memristors are yet to be successfully demonstrated due to the lack of a robust printable switching medium and a reliable printing process. An aerosol‐jet‐printed Ag/MoS2/Ag memristor is realized in a cross‐bar structure by developing a scalable and low temperature printing technique utilizing a functional molybdenum disulfide (MoS2) ink platform. The fully printed devices exhibit an ultra‐low switching voltage (0.18 V), a high switching ratio (107), a wide range of tuneable resistance states (10–1010 Ω) for multi‐bit data storage, and a low standby power consumption of 1 fW and a switching energy of 4.5 fJ per transition set. Moreover, the MoS2 memristor exhibits both volatile and non‐volatile resistive switching behavior by controlling the current compliance levels, which efficiently mimic the short‐term and long‐term plasticity of biological synapses, demonstrating its potential to enable energy‐efficient artificial neuromorphic computing.