Figure 1 - Mesopores in aqu. electrolyte
Figure 1 - Mesopores in aqu. electrolyte
According to this "current-burst-model", macropores form if the balance between direct dissolution and oxidation in
a current burst is within a certain bandwidth, and mesopores, generally observed on highly doped p- and n-Si, are observed
if the oxidation component is too small. Accordingly, an enhancement of the oxidizing power of an electrolyte, should have
the potential to produce macropores under conditions where otherwise only mesopores occur. Strong oxidizing compontes were
added to the electrolyte. The macorpores were etched in the dark n-Si: (0.020-0.060 Wcm).
Figure 2 - Macropores
No backside illumination could be used because the small diffusion length of highly doped Si would prevent the hole
diffusion to the front side. The holes needed thus must be generated by electric field effects (avalanche break-through)
which is easy in highly doped materials. But avalanche break-through so far has been considered to be prime the effect for
mesopore generation and thus should not be seen as a prime reason for macropore formation. According to the current burst model,
the prime reason for macropore formation in this (and other) cases is the decreased probability for current bursts on
H-passivated surfaces coupled with the smoothing action provided by a minimum of oxidation following direct dissolution,
and this accounts at least qualitatively for the experimental observations. Other factors, as, e.g., the shape of the space
charge region (SCR), or the supply of holes may influence macropore stability, too, by influencing the nucleation probability
of current bursts, but the interplay between H-passivation and oxidation must be seen as the prime parameters.
Figure 3 - The nucleation takes place at {111} pyramides