Light‐Stimulated Synaptic Transistor with High PPF Feature for Artificial Visual Perception System Application (Adv. Funct. Mater. 22/2022)

Abstract

Light‐Stimulated Synaptic Transistors Optoelectronic synaptic devices with a high paired pulse facilitation index are essential for constructing high‐precision artificial visual perception systems. In article number 2113053, Jun Wang and co‐workers develop a light‐stimulated synaptic transistor with an ultra‐high PPF index (≈196%) by introducing hexagonal boron nitride into a classic graphene‐based hybrid transistor framework, which provides an effective method for constructing artificial visual perception systems in the future.

Artificial Visual Systems With Tunable Photoconductivity Based on Organic Molecule‐Nanowire Heterojunctions

“…The maximum PPF index value is 178%, obtained at ΔTime = 1.67 s. The pulse interval‐dependent decay of the PPF index with two different light pulses can be expressed utilizing a double exponential decay function as presented in the inset of Figure 2h. [ 46 ] The y 0 parameter is the PPF index when the pulse interval approaches infinity. The t 1 and t 2 parameters are the decay constants of the rapid and slow phases, respectively.…”

Printable Coffee‐Ring Structures for Highly Uniform All‐Oxide Optoelectronic Synaptic Transistors

“…Figure S13 (Supporting Information) shows the change in EPSC (ΔEPSC) when triggered by optical pulses of varying UV intensities; it can be seen that ΔEPSC increased from 0.9 to 6.2 nA as the intensities of optical pulses increased from 2.2 to 6.5 mW cm −2 , which is related to more photoexcited electrons are generated under illuminating a higher intensity light. [ 47,48 ] Upon the termination of the UV light pulse, the EPSC decayed slowly over time and produced a persistent photoconductivity (PPC) behavior common to metal oxides. This variation of EPSC can be attributed to the ionization and recombination of oxygen vacancies in the ITO film.…”

“…[ 53 ] PPF is determined by the time interval (Δ t ) between two pulses, and the EPSC generated by the latter pulse (A 2 ) is higher than the former one (A 1 ), which is related to the reorganization of photoelectrons. [ 54 ] The PPF index (A 2 /A 1 ) was well fitted by the printed coffee‐ring transistors with a series pulse interval (Δ t ), as shown in Figure 3g, which can be simulated by the double exponential function: [ 48,55 ] $$\[ \begin{array}{*{20}{c}}{{\rm{PPFindex}} = {A_2}{\rm{/}}{A_1} = 1 + {C_1}\,\cdot\,{\rm{e}}( - \Delta t{\rm{/}}{\tau _1}) + {C_2}\,\cdot\,{\rm{e}}( - \Delta t{\rm{/}}{\tau _2})}\end{array} \]$$where C 1 and C 2 represent the initial facilitation magnitudes of the rapid and slow phases, and τ 1 and τ 2 are the characteristic relaxation times of the rapid and slow phases, respectively. The inset of Figure 3g shows the EPSCs triggered by a pair of successive UV light presynaptic spikes with a Δ t of 1 s, corresponding to the calculated PPF ratio of 146% at ∆ t = 1 s. The τ 1 and τ 2 extracted from the fitting curve are 0.09 and 5.8 s, respectively.…”

“…To further demonstrate the pattern recognition capability of the printed coffee‐ring transistor synapse, a three‐layer artificial neural network (ANN) was constructed to perform supervised learning of handwritten digits (28 × 28 pixels) adopted from the Modified National Institute of Standards and Technology (MNIST) database. [ 48,58 ] As illustrated in Figure a, the ANN consists of 784 input neurons, 300 hidden neurons, and 10 output neurons, which are connected by synaptic weight. Each input neuron receives information from the corresponding pixels of the 10 digital images from inputs “0” to “9”.…”

Recent Progress in 2D Inorganic/Organic Charge Transfer Heterojunction Photodetectors

“…Along with the high‐efficient convolution and spike network integrated with these chips, the low‐energy consumption optical chip realizes a series of in‐computing function containing the learning, recognition, and classification. [ 248–251 ] Also in machine vision, the retina like photodetector are developed with visual adaption, [ 234 ] convolutional processing [ 232 ] and spike encoding processing functions. [ 252 ] All these recent advances of 2D based photodetectors ( Figure ) provide the reference for future guidance of 2D inorganic/organic photodetectors.…”