Computational Investigation of the Potential and Limitations of Machine Learning with Neural Network Circuits Based on Synaptic Transistors.

Synaptic transistors have been proposed to implement neuron activation functions of neural networks (NNs). While promising to enable compact, fast, inexpensive, and energy-efficient dedicated NN circuits, they also have limitations compared to digital NNs (realized as codes for digital processors), including shape choices of the activation function using particular types of transistor implementation, and instabilities due to noise and other factors present in analog circuits. We present a computational study of the effects of these factors on NN performance and find that, while accuracy competitive with traditional NNs can be realized for many applications, there is high sensitivity to the instability in the shape of the activation function, suggesting that, when highly accurate NNs are required, high-precision circuitry should be developed beyond what has been reported for synaptic transistors to date.

Anion Exchange Membranes for Fuel Cells Based on Quaternized Polystyrene-b-poly(ethylene-co-butylene)-b-polystyrene Triblock Copolymers with Spacer-Sidechain Design.

This work studied the polystyrene-b-poly(ethylene-co-butylene)-b-polystyrene (SEBS) triblock copolymers functionalized by butyl quaternary ammonium (C4Q) groups and alkyl side chains of different chain lengths (Cn, n = 0 to 24). The hydrated membrane morphology was modeled by dissipative particle dynamics simulation at hydration levels from 10 to 30. A hydroxide model was devised to characterize the diffusivity of anions under the coarse-grained framework. In general, the ionomers with alkyl side chains provided ion conductivity of a similar level at a lower ion exchange capacity. All hydrated SEBS-C4Q-Cn ionomers showed clear phase separation of the hydrophobic and hydrophilic domains, featuring 18.6 mS/cm to 36.8 mS/cm ion conductivity. The hydrophilic channels expanded as the water content increased, forming more effective ion conductive pathways. Introducing excess alkyl side chains enhanced the nano-segregation, leading to more ordered structures and longer correlation lengths of the aqueous phase. The membrane morphology was controlled by the length of alkyl side-chains as well as their tethering positions. Ionomers with functionalized side chains tethered on the same block resulted in well-connective water networks and higher conductivities. The detailed structural analysis provides synthesis guidelines to fabricate anion exchange membranes with improved performances.