Hi Willie,

I had good success in bonding PDMS chip to glass (alumina surface) not on
silicon though. The bonding is due to the -OH linkage between substrate and
PDMS surface. I had to do an oxygen ash (or you can do RIE at very low
power) to clean the surface, dip in water to create -OH bonds on the
surface. Similarly you have to clean the surface of the silicon too, and
gently place the PDMS chip on the substrate. Baking at lower temperatures
may also help. I also put some weight on PDMS channels to ensure that whole
surface is in contact with the substrate. I am not sure how long it takes
for good bond, but I used to take it out next day and they were pretty good.

I completely agree with Dirk and Joe too.

Regards,
Pavan

On Fri, Apr 3, 2009 at 7:33 AM, Dirk Renckens - TNW <
T.J.A.Renckens@tudelft.nl> wrote:

> Dear Willie,
>
> PDMS bonded to silicon can generally withstand only a few bar
> overpressure before delamination. Depending on the flow the flow you
> want to use in your channels, the pressure drop generated may very well
> exceed the delamination pressure. I would advise estimating the pressure
> drop.
>
> Pressure drop correlations for rectangular geometries can be found
> (amongst others) in:
> Shah - Laminar Flow Forced Convection in Ducts (1978), Chapter 5
>
> Because I do not know the exact geometry you are using, a paper on the
> deformation of PDMS channels by pressure drop due to flow might also
> interest you:
> Gervais - Lab Chip 6 500 (2006)
>
> Good luck. If you are having trouble, feel free to contact me.
>
> Kind regards,
>
> Dirk Renckens

--
Pavan Samudrala