Why is the drift current in a PN junction independent of electric field? I've learnt that under the forward biased condition in PN junction, the diffusion current increases slowly and after surpassing the cut-in voltage, the diffusion current increases exponentially, and under reverse biased condition, the diffusion current approaches zero. 
In either of the conditions, the drift current is said to remain almost constant. 
Why is it so? Why is the drift current independent of the electric field (by the battery)? Why did my textbook emphasized on the word "almost" when talking about the drift current to be constant? Shouldn't the movement of minority charge carriers (drift current) increase under the reverse biased condition in PN junction?
 A: it could be that the number of minority charge carriers only depends on temprature(it does for a fact), so applying an electric field to move those minority charge carriers wouldnt help much if there are no minority charge carriers left.
A: Current in reverse bias comes primarily from injecting minority carriers into the depletion region. There aren't a lot of minority carriers to start with, so you don't have a lot of current. Changes in diode bias show up as a change in electric field (almost) exclusively across the depletion region. The current in reverse bias is limited by how quickly those minority carriers are injected into the depletion region, which is not a function of electric field within the depletion region. Therefore current in reverse bias is not a function of reverse bias.
In reverse bias you also have generation current where electron/hole pairs are spontaneously generated within the depletion region. The generation rate is not a function of electric field, however increasing the reverse bias will increase the width of the depletion region, increasing the current. Generation current is typically small, and the width does not typically increase very quickly with increasing reverse bias, and there is a square root dependence. Therefore this affect is typically small.
The fact that changes in reverse bias will show up as a small change in electric field within the diode neutral regions means the rate of minority carrier injection will change as a function of bias. Its a very small change, but this accounts for the "almost" your book emphasized.
