Nathan,

Refer to

E. Leclerc, Y. Sakai,
and T. Fujii, “ Cell culture in 3-Dimensional microfluidic structure of PDMS,”
Biomedical Microdevices, vol. 5, pp. 109-114, 2003
- Harsh.D.Sundani.

Graduate Research Assistant,

EECS Department,
The University of Toledo,
Toledo, Ohio.

> From: kpnichols@uchicago.edu
> Date: Sat, 27 Mar 2010 11:22:39 -0500
> To: mems-talk@memsnet.org
> Subject: Re: [mems-talk] Best way to spin PDMS for making sheets?
>
> Nathan,
>
> To get PDMS to come off, you don't need a very well controlled
> hydrophobic layer. Just use perfluorodecyltrichlorosilane (FDTS)
> applied in a vacuum desiccator, and then baked at ~110. You can find
> FDTS suppliers here: http://www.chemexper.com/
>
> 1) Place your Si wafers (with no PDMS on them, this is just so you end
> up with a long fluorinated molecule sticking up from the surface so
> the PDMS will peel off easily later) in a vacuum desiccator connected
> to house vacuum (do a google image search for vacuum desiccator to see
> what these look like). 2) Place 10-20 uL of FDTS in a small container
> (the top of an eppendorf tube works well) in the bottom of the chamber
> (location doesn't matter much). 3) Pull house vacuum, and leave them
> for ~2 hours (exact time will be more or less depending on the vacuum
> strength and container size). 4) Bake the wafers at 110 for ~2 hours.
>
> There are lots of methods out there to get higher contact angles, but
> that will be good enough.
>
> If you have a goniometer, measure the static contact angle and check
> that it's at least 90 with DI water. If you don't have a goniometer
> handy, just pipette (or somehow dispense) a few microliters of DI
> water on the surface, and make sure that it forms tight balls that
> roll around easily.
>
> Also, keep in mind that most people use PDMS precisely because you
> don't need to ablate, etch, or otherwise chemically attack it. You can
> just use "soft lithography" which uses a stamp to mold the PDMS as it
> cures. Si molds are common, but glass and metal (even CNC machined
> brass molds are good enough for some microfluidic applications). There
> are hundreds of papers out there describing it, if you're interested.
>
> Kevin P Nichols, Ph.D.