Samsung Advanced Institute of Technology

Unique on-current reduction under off-scenario is observed in InGaZnO thin-film transistors (IGZO TFTs). For memory applications,

the programming transistors are predominantly exposed to asymmetric off-state biases, which may result in programming current

variation. The current decrease is investigated varying the applied field. Further investigation of the phenomenon is conducted

with transmission line-like method and degradation recovery tests, and current reduction can be attributed to contact resistance increase

by charge trapping in the source and drain electrode and the channel region. The current decrease is subsequently formulated

with a stretched exponential function for quantitative analysis of off-state degradation, and the bias dependencies of parameters are

modeled. A neural network hardware acceleration simulator is utilized to assess the complicated impact the off-state current degradation

could instigate on on-chip trainable IGZO TFT-based synapse arrays. The simulation results generally show a deterioration of

training accuracy with aggravated off-state stability, and mitigation methods are suggested.