SO2 is highly aggresive gas. Make sure you use SufInert deactivated lsample lines, loops, injection port liners. Also consider to use a Siltek deactivated Metal column. I have used MXT 0.53mm with 7 microns for that. On PDMS SO2 will elute very close to COS and for separation, you may have to go subambient.
If No COS is there, teh MXT xcolumn will work.
alternative, you can use a Rt-U-BOND porous polymer. Here SO2 has higher ertention and also gives a good peak shape. Last solution is to use a 0.53mm micropacked Shincarbon in Siltek/sulfnert treated MXT. Here SO2 also elutes as a sharp peak, (but no other sulfur elutes). Very specific and high retention..
detection: FPD, PFPD or SClD.
Injection: use a split, but note that because the SO2 elutes fast, you can make a big error: see also a recent blog on that.
http://blog.restek.com/?p=8610
Jaap de zeeuw, Restek Corporation