I am trying to analyse some water samples that may be contaminated with MTBE. I need some tips on how to go about proving the presence or absence of MTBE using LC. Any ideas would be appreciated.

Why LC, I would think that GC (GC/MS for positive compound ID) would be the way to go here. By the way, most methods can not determine the absolute absence of a given compound. They can determine that amounts found are more than the detection limit, but not less. This is a very important distinction to make when doing the kind of environmental contamination analysis that you are attempting.

For the detection of trace MTBE in gasoline contaminated soils or water, I recommend EPA Method 8260, which utilizes GCMS. GCMS gives unequivocal results for MTBE due to its unique mass spectrum, which is easily distinguishable from potentially interfering matrix components.

I too think this is a natural for cap. GC. There is an EPA method for volatile organics, I think it is 8021A that uses a PID. You could also use a CP-SIL5 CB column with a FID to detect volitle organics. If you want details on either of these methods you can email me. Goo Luck.