Reverse Recovery
The inability of a conducting diode to turn off immediately when it is exposed to reverse voltage, so that it passes current in the reverse direction for a short time. This period is typically 100 ns-1 µs in standard diodes and much shorter in fast and Schottky types; it produces loss and noise.
While a diode conducts, charge carriers accumulate in its structure. When the voltage is reversed, clearing these carriers takes time; during this period the diode conducts in the reverse direction and an unwanted current pulse appears in the circuit. The peak of the pulse depends on the forward current and on how fast the current is interrupted (di/dt).
The consequences fall under three headings. First, this reverse current flows through the opposite switch, and since the full supply voltage is across that switch at that moment it adds directly to the switching loss. Second, the abrupt interruption of the current starts oscillation with the parasitic inductances and radiates broadband noise. Third, in severe cases the pulse can behave like a current rise similar to shoot-through in the leg.
MOSFET body diodes are generally not good in this respect. That is why in high-frequency, high-current designs either synchronous rectification is used to minimise the diode conduction time, or a parallel fast/Schottky diode is added so that most of the current passes through the better-recovering device.
The context in which this term is used: DC Motor Driver Selection Guide