contamination?

[vvmm]

We encounter big problems when we analyzed the chloramphenicol (CAP) using LC/MS/MS. We find two peaks in the chromatography, which cannot separate completely. The RT of two peaks is 7.90 (CAP) and 8.10(contamination), respectively. The peak (RT 8.10) can be found in each blank, and have same transitions as CAP (321>152 321>257 and 323>152, Scan mode: Multiple Reaction Monitor, Ion mode: ES-). When using syringe pump inject water and MeCN into MS/MS, respectively, we also can find the peak, but the response are different (MeCN>Water). But when injected nothing, just gas, this peak disappeared. Is the MS/MS contaminated? We have cleaned every part we can clean, but not improved. I don’t know which part is contaminated? If the contamination hinds in quadrupole, why the water and MeCN have different response? And why cannot find the peak when injected nothing, just gas? If the contamination existed in the part before quadrupole, where does it hind in? Anyone has any ideas?

[MG]

If the contaminant shows up as a chromatographic peak, then most likely the contamination is in your mobile phase or autosampler. If running a gradient, try running it without injecting anything, i.e. disable the autosampler if possible. If you still see the peak, the problem is with your mobile phase(s).