🚀 go-pugleaf

From: Robert Erck
bob_erck@qmgate.anl.gov

I have a question concerning microelectronic fabrication.  The questions
spring from an article recently published in Science. (12 June, p. 1537.)  
In the article, the authors describe a vibrating cantilever beam apparatus
that they constructed by etching Si which they use to measure crack
propagation.  As a crack slowly proceeds into the arm supporting the
device, the resonant frequency falls.  Crack growth rates in the supporting
arm <<1 angstrom/hr can be studied in this manner. 
When the device is exposed to humid air (70%RH), the resonant frequency
drops quickly (~hours), and they attribute this to "immediate crack growth"
upon exposure to moisture.  The resonant frequency subsequently decreases
at a much slower rate.  
I have concerns about this interpretation however.  The method of
fabrication of the cantilever beam was specified as an anisotropic etch of
B doped Si.  I know that exposure of boron to the moisture in humid air
results in the formation of visible films of boron oxide/ boric acid.  If
there is any boron on the cantilever beam, it might react with the moisture
and form an oxide.  A back-of-the-envelope calculation shows that only a
few dozen layers  of oxide would account for the frequency drop on this
micron-thick device.  I wonder if that's what's happening?  
So my three questions are: 
1) does an anisotropic etch of B doped Si leave unetched material that
contains B?  2) would boron oxide/acid would form?  
3) or, alternately, does semiconductor-grade Si itself have a tendency to
adsorb a film (10 nm) that could cause such a weight gain?