Chromatography Forum

Hi all,
I'm working on an LCMS method for THCCOOH (carboxy-THC) in urine and I noticed that I get dramatically different signals depending on the pH of the buffer I add during sample prep. At pH7 signals are about 25 fold lower than they are with the pH5 buffer. Does anyone know what could be causing this?
My first thought was THC sticking to glass/plastic to a higher extent at pH7, or maybe some kind of improved ionisation effect at the source (ESI negative). Any ideas?

Thanks for your help

H

Is your mobile phase buffered?

pKa affecting whether the carboxy group is protonated or not?

Yes the mobile phase is buffered (0.1% formic/10mM ammonium formate).

The pKa of THCCOOH is about 4.6 I think so at pH7 it will be ionised (negatively charged). I get 25 fold higher signals at pH5. THC is pretty notorious for sticking to plastic and glass since it's so lipophillic but 25 fold difference seems pretty dramatic.

I'm using negative mode ESI so wouldn't I expect higher signals if the molecule is already negatively charged before it reaches the source (pH7). Instead I get higher signals at pH5 which makes me think it's a sticking/adsorption issue.

The sample buffers are pretty strong compared to mobile phase buffer- about 150mM.

You could spike the analyte post-sample prep if you want to check whether it's a recovery or an ionization effect.

About the sample buffer versus the mobile phase buffer: I think it is not as straightforward to say the sample buffer is stronger so that will determine the pH. Remember that by injecting sample you're diluting it in mobile phase. It depends on various parameters (injection volume, sharpness of the peak,..) what the pH will be when it reaches the MS. By then it also has organic mixed in, i assume, which makes it incorrect to talk about a pH.

What is your sample prep?

I get dramatically different signals depending on the pH of the buffer I add during sample prep. At pH7 signals are about 25 fold lower than they are with the pH5 buffer. Does anyone know what could be causing this?

If LLE/MLLE/SSLLE or SPE-RP (Sil-C18 etc.) are extraction/purification methods then its normal (pH decrease > logD increase > recovery increase) and sample pH must be < pKa of analytes.

Hey guys, thanks for the replies

The sample prep is a simple protein precipitation (protein crash with MeOH, centrifugation). The supernatant is transferred to a glass LC vial and mixed with a very dilute buffer, and injected into the mass spec.

Spiking the analyte post sample prep would be a good option, I might try that next time I'm in the lab.
THC elutes at high %B so yeah by the time it hits the mass spec, who knows haha (not me anyway)