by
AA » Mon Jul 05, 2010 1:59 pm
Generally speaking, unmodified HPLC systems are not suitible for SFC. It is not as simple as putting together a list of "change this to that" either but it is possible (I'v done it), I will try to hit some of the highlights.
1. Pump heads need to be cooled to remove the heat generated by the compression of the carbon dioxide.
2. Some materials in some systems are not compatable with carbon dioxide (teflon comes to mind). Some seal materials tend to shrink causing leaks. Some check valve materials are prone to this as well.
3. Some system use manifold valves (like GPV's in Quartanary systems or A1/A2 switching in high pressure binary systems). These valves are on the low pressure side of the system and usually cant take more than 100 psi with out leaking. The input pressure of CO2 is ~ 800 psi.
4. Some pumps use pre-compression algorythyms in the pumps. This kind of firmware sometimes does not like the compresibilty of the CO2. It is quite compressible compare to LC solvents.
5. Depending on the kind of autosampler, modification may need to be made. Loop filling styles seem to work better than "needle part of the loop" types. It's all about where any CO2 in the loop goes when the loop is decompressed. One option is to modify the AS to do what is known as "modifier stream injections", this can help.
6. Detector flow cells need to be modified to handle very high system pressures (4000 psi at least). Your average UV flow cell is usually rated at 1000 psi. Some flow cell gasket materials are compatible with CO2.
7. You would have to construct some kind of back pressure generating device after the detector to keep the CO2 liquid untill after it passes through the flow cell. A simple needle valve will often do the trick.
I didnt hit all the issues, it can be done if you are a bit of a mechanic. So, to simply answer your question, no you cant drag an old LC out of storage and hook the tank of CO2 up to it and run SFC.
