Nick,
   I would guess you have an adhesion problem caused by the thermal
expansion mismatch of the film on the silicon or a relatively dirty
interface.  First, I would make sure the silicon is very clean before
deposition.  If it is PVD, start with a back sputter etch to clean the
surface before depositing.  Check the stress properties of the deposited
film.  Usually, the deposition parameters can significantly influence the
stress in the thin film.  the good news is that you have two components
that will form silicides.  This can be used to your advantage by reacting
this film with the substrate to get good adhesion.  Unfortunately, you have
to control how much ni and ti get consumed, the phase that results from the
thermal reaction, the stress of the as deposited film,...  Another idea is
to deposit a small amount of titanium and then form a disilicide (the low
resistivity phase above 650C or so).  Then deposit your Ni Ti on top of
this.  This may give you a more thermally stable film (kinetics and
thermodynamics) with better adhesion.
At 09:40 AM 7/22/99 GMT0BST, Nicholas Botterill wrote:
>Dear all,
>
>As this is my first posting to this group, I'd like to introduce
>myself, my name is Nick Botterill, and I'm a PhD student at the
>University of Nottingham, U.K., investigating thin film shape memory
>alloys.
>
>At present, I'm depositing equiatomic Nickel Titanium onto (100)
>Silicon wafer substrates, but am experiencing problems when I attempt
>to recrystallise the film (it is amporphous, or nano-crystalline upon
>deposition).  I expect the film to recrystallise at around 500-600
>degrees Celsius, however the film simply peels off the substrate.  In
>anyones oppinion, do you think that this is due to mismatches in
>thermal expansivity of NiTi and Si, or due to the stresses in the
>films ?
>
>Any input would be very much appreciated.  I look forward to hearing
>from some of you in the near future.
>
>Many thanks,
>
>Nick Botterill
>
>[email protected]
>[email protected]
>
>