Ok, lets say you have n-type impurities in a semiconductor. So there is an equilibrium number of electrons in the conduction and impurity bands given by the temperature and the typical semiconductor statistics. Now, you excite electrons from the valence to the conduction band. Here are my questions: 1. Will some of the resultant recombination be impurity band to valence band? (Assume low temperatures and small binding energy if you must). 2. Will some conduction electrons that have not yet recombined into the valence band, thermalize into the impurity band (that has some vacancies due to donor electrons recombining). I have read a lot on semiconductors but I have not seen anyone discuss if this occurs yet I have been told that it is true. Thanks.