As I understand things they don't work like that.The only way to cancel the natural internal EMF of a metallic crystal lattice is to cool it. As you cool it, there will be a temp. where all of the free electrons return to a home atom. Only then will the lattice not have an internal EMF. As you warm the lattice up...one electron will be freed. As soon as the first free electron starts moving about the lattice, you have an internal EMF.
The 'natural' thermal movement is randomly oriented so that, over time, there is no preferred direction of drift or build up.
If however you introduce a preference/selection mechanism then you will introduce a concentration gradient which produces a voltage.
This can apply to random drift phenomena in general such as osmosis not only to electric phenomena such as semiconductor junctions, electrophoresis, electro-osmosis , the Hall effect and others.