Polystyrene is good plastic for your experiments because it very easy dissolved in acetone as well in other well-known and more strong solvents (toluene and so on). 

Dear all,

I agree with Dr. Serge V. Shil'ko, however, in polymer mixtures or 'blends' it is rarely common that two polymers will maintain good dispersion or distribution of their phases relative to each other. Always the problem of compatibility arises and leads to phase separation while cooling down, unless a third component is added which insures good compatibility between PDMS and PS (compatiblizer).

So, the point I want to add is that you should first find the conditions to maintain the PS phase within the PDMS one, and after that solvant extraction of PS is applied. Regards

Dear Yannapol Sriphutkiat, the original method of microchannel structure obtaining (alternative to dissolving)  has been proposed  before. This method based on ability of microorganisms (bacteria) to gnaw out selectively some polymer phase in composites. The differentiation of their "tastes" relative to various polymers allow us generate porous structure with contiguous microchannels. 

An other idea is to use PVA (polyvinyl alcohol) that can be dissolved using water.

Be carefull with acetone because it soaks PDMS

As said Polystyrene is a good choice. I was also thinking about acrylics (perspex/PMMA) as acetone is effective in dissolving superglue which is chemically similar. Looking at the table in the link below it seems I was right but I don't have direct experience of this.

http://www.plasticsintl.com/plastics_chemical_resistence_chart.html

I am doing my master's project on microfluidics too, I use ABS to print my structure and dissolve it easily with Acetone. But nowadays some ABS filaments don't dissolve quickly enough.

Yannapol Sriphutkiat