Enhancing LiAlOX synaptic performance by reducing the Schottky barrier height for deep neural network applications.

Although good performance has been reported in shallow neural networks, the application of memristor synapses towards realistic deep neural networks has met more stringent requirements on the synapse properties, particularly the high precision and linearity of the synaptic analog weight tuning. In this study, a LiAlOX memristor synapse was fabricated and optimized to address these demands. By delicately tuning the initial conductance states, 120-level continuously adjustable conductance states were obtained and the nonlinearity factor was substantially reduced from 8.96 to 0.83. The significant enhancements were attributed to the reduced Schottky barrier height (SBH) between the filament tip and the electrode, which was estimated from the measured I-V curves. Furthermore, a deep neural network for realistic action recognition task was constructed, and the recognition accuracy was found to be increased from 15.1% to 91.4% on the Weizmann video dataset by adopting the above-described device optimization method.

Reducing Interface Traps with High Density Hydrogen Treatment to Increase Passivated Emitter Rear Contact Cell Efficiency.

In this work, a high-density hydrogen (HDH) treatment is proposed to reduce interface traps and enhance the efficiency of the passivated emitter rear contact (PERC) device. The hydrogen gas is compressed at pressure (~ 70 atm) and relatively low temperature (~ 200 °C) to reduce interface traps without changing any other part of the device's original fabrication process. Fourier-transform infrared spectroscopy (FTIR) confirmed the enhancement of Si-H bonding and secondary-ion mass spectrometry (SIMS) confirmed the SiN/Si interface traps after the HDH treatment. In addition, electrical measurements of conductance-voltage are measured and extracted to verify the interface trap density (Dit). Moreover, short circuit current density (Jsc), series resistance (Rs), and fill factor (F.F.) are analyzed with a simulated light source of 1 kW M-2 global AM1.5 spectrum to confirm the increase in cell efficiency. External quantum efficiency (EQE) is also measured to confirm the enhancement in conversion efficiency between different wavelengths. Finally, a model is proposed to explain the experimental result before and after the treatment.