Does ambipolar transport when an electric field is applied to the semiconductor imply zero drift current? I am reading Neaman's book on semiconductor physics and devices and one of the more important conclusions are that excess holes and electrons drift and diffuse together due to internal electric fields. However, when holes and electrons move in the same direction, wouldn't that mean zero net current?
Additionally, if this is true, why would the electrons/holes move at all if the result in terms of current is the same.
I find it very hard to grasp why the electrons and holes move in the same direction. As the external field tries to separate the electrons from the holes, doesn't the internal electric field get weaker as the electrons move further away from the holes? And wouldn't the internal field be formed locally in the semiconductor, near the excess carriers, resulting in the external field to be dominating in most location in the lattice? This would cause the carriers to separate in the end. However, Neaman's explanation indicates otherwise.
Thanks in advance
 A: The easiest way to understand this is to consider the situation of low injection where the excess minority concentration is much smaller than the majority concentration. Let us assume that you have a homogeneous one-dimensional n-type semiconductor and introduce a localized spatial pulse of excess minority carrier holes into it, first without external field. The positive charge pulse produces an electric field which produces a majority carrier electron current in the direction of the minority carrier holes which, essentially, leads to a very fast shielding of the hole field. This is a very fast process because the internal electric field of the holes produces a strong electron current due to the relatively high concentration of electrons. The end effect is that the positive excess hole pulse is practically neutralized by a minor change in electron concentration relative to its equilibrium concentration which mirrors the excess hole distribution. Therefore the neutrality approximation can be assumed.
When the spatial excess hole pulse disperses due to diffusion, the excess electron concentration follows its shape. When you move the positive excess hole pulse by any means along the semiconductor the shielding negative excess electron pulse follows but no net electron transport occurs. The electron concentration accumulates at the location where the hole pulse is and returns to the equilibrium when it has passed. This is similar to the local accumulation of charge at a metal surface by electrostatic induction when you move an isolated charged object along it. No net charge transport occurs in the metal. When you move the hole pulse by applying an external electric field, the hole charge will be transported by drift and a much larger electron drift current will be produced in opposite direction due to the much larger majority concentration. The shielding electron pulse will nor separate from the hole pulse. It consists of different electrons that accumulate along the location and  following the shape the hole pulse takes.     
