The measurement techniques employed to assess hot-carrier degradation are very similar to those used for bias temperature instability. The degradation is often measured as a change in drain current ΔId or as a threshold voltage shift ΔVth [170]. Additionally, recovery of hot-carrier stressed devices is currently of increasing scientific interest [171]. Measuring the recovery after hot-carrier stress is very similar to BTI recovery experiments. Recovery can be observed by switching all voltages to their respective relaxation levels and recording the drain current as it is done for stress. However, often it is necessary to accelerate recovery to collect more data points in less time. This can be achieved by switching the device temperature to higher temperatures while measuring recovery.
A typical stress or stress-measure technique to assess HCD is shown in Figure 7.2, where the drain current is recorded for a preset pair of gate Vg and drain voltages Vd at a fixed temperature. Before the hot-carrier experiment starts, the fresh device is characterized by measuring for example the Id - Vg, Id - Vd or CV curves. During this initial measurement great care not to BTI stress the device in this phase has to be taken. Throughout the hot-carrier experiment the drain current is constantly recorded, while gate and drain voltage are regulated by a controller to cycle between the various stress conditions. Hot-carrier stress is defined as the set of voltages (Vg, Vd) at constant device temperature, which cause a prominent drain current shift |ΔId| or an equivalent absolute threshold voltage shift |ΔVth| in a given time ts.