Several things could be an issue.

First, wafers are not uniformly thick, but it should be easy to tell if this
is an issue.

Second, if the pits are oriented in a way that they can accumulate bubbles
inside of them, this can reduce the etch rate.

Third, the strirring of the bath can cause different flow of etchant accross
different parts of the wafer, which can effect the etch rate.

I'd look at the stirring/agitation and wafer orientation.

David Nemeth
Senior Engineer
Sophia Wireless, Inc.

-----Original Message-----
From:  Duan
Subject: [mems-talk] The uniformity of TMAH etching

I did several experiments with TMAH etching, but I found that, the ER
of TMAH was variable, even on the same wafer.  For example, I was
trying to etch through a 370um thick silicon wafer in 25%TMAH at 85
degree(fresh solution, and typical etch rate is around
30microns/hour).  After 14hours' etching, part of the wafer had been
etched through, but half of them were not.  It happened several times.

Could someone tell me why?

I used wet thermal oxide as mask(1.5 micron thick) and  I am sure that
the exposed oxide had been completely removed after RIE.  Also, the
wafer was prime wafer.