I am attaching here some SEM pictures of structures etched in Si by DRIE. The process is far from being optimized, e.g., inadequate sidewall passivation, huge underetching of the mask. Aside from that, these etching defects in the Si that appear to be randomly spread but somewhat regular in shape, this rectangular shape (as can be seen in the top view of the microscope picture), I am wondering what the explanation for them, the physical explanation of their propagation or development, can be.

By the way, the density of their occurrence is not the same on the wafer, it is location dependent, i.e., on one side of the wafer, they do not exist, then they increase as you move to the other side, where at the end almost all open edges of the mask are completely affected by them.

I have never seen such defects of the DRIE being described in literature before, so I thought, it would be interesting to know what others with longer experiences have to say about this. Thanks!