Late to the show, but this only works because the deposited charge always has the same sign, correct? For instance, if the ions used are positive icons (electrons stripped), the surface elements get positively charged, and the diodes drain that charge to the substrate. But, in operation, the surface elements are negatively charged, so the diodes block. (Right? Or is it the other way around?)

The diodes break down nondestructively under the voltage, getting rid of the charge regardless of polarity. During operation of the chip, the diodes are reverse biased, so they have no effect.