Effect of concentration on ion ratio value - Chromatography Forum

Effect of concentration on ion ratio value

Discussions about GC-MS, LC-MS, LC-FTIR, and other "coupled" analytical techniques.

6 posts Page 1 of 1
Hello everyone.

In our lab we use HPLC-MS/MS to quantify drugs of abuse in biological samples. In most of them (almost forty) the precursor ion breaks into three different daughter ions, i.e. we identify the quantifier ion and two confirming ions. Again, the ion ratio of these two confirming ions are within the proper range in most of the drugs. However, we notice that ion ratio is below this range at low concentrations.

We know that lower concentrations means lower sensibility but I don't think that is the cause because we sometimes get a fine confirmation of both ions. When we cannot confirm one ion, its ionic ratio is usually below the range (90% of times) but it can be higher sometimes (10% of times). Also, this problem is VERY common in THC, to the point that the confirmation peak doesn't even appear in the chromatogram.

In some occasions, the ionic ratio of the confirming ion of calibrator 1 is too low but the ionic ratio of the confirming ion of calibrator 2 is too high. In this scenario we can slightly modify the mean of ion ratios so that the range deviates towards one of them, thus confirming it. However confirming both of them is impossible without opening the window of tolerance.

In other occasions, the ion ratio in one of the calibrators is so high/low that the average of ion ratios is not representative. Thus, the ion ration range (which is calculated from the average of ion ratios in all calibrators) deviates to the point that we cannot confirm any calibrator or samples.

Sorry for the long post, I wanted to give as much information as possible. Does anyone have a similar problem? How can we get a fine confirmation at low concentrations (now we only get it from time to time)? Why does this thing happen: the detector, the lens, etc? Thanks in advance!

During method development. did you validate for interference using matrix blanks? Triple quads are not very selective towards m/z, and in complex matrices it is not unheard of that you will get a 'hit' for your MRM-transitions stemming from other compounds than what you are actually looking for. At higher concentration, it might not affect your peak intensity beyond whats normal but at low concentrations it can throw your ion ratios off completely. The solution is to either use other, more selective transitions, modify the chromatography to separate them or increase the acceptable LOD/LOQ.
Another thought: Is this happening because you're reaching the low end of the instrument's sensitivity? It may be that the peak areas are subject to random errors that become proportionally bigger when the peaks are very small? If so, there's not a lot you can do about it: when random variation becomes as large as the tolerance you permit on your qualifier ions, then the peak will begin to fail on qualifier ion ratio, just randomly.

Some instruments have thresholds below which they don't record a signal. If you're using an instrument of this type, and the qualifier ion becomes very weak indeed, you may lose the start and end of the peak where it isn't as high as the threshold, biasing it to fail low rather than high. In any case, it's hard for the integrator to see something that's barely there, so depending on your integrator settings, it is quite possible it will be biased towards giving an unduly low area for very small peaks.

But yes, confusions caused by coeluting and background ions are of big relevance too.
How do you quantify? Do you first screen and then perform standard addition or run a calibration curve on positive samples?

I would suggest to compare the ion ratios of low concentrations in unknowns to comparable concentrations of that compound spiked in the same matrix.

We (in environmental analysis) are allowed to have

1) Larger deviations in ion ratios if the qualifier is naturally lower than the quantifier
2) Larger deviations closer to the LOQ. The guidelines list that "expert judgement" applies in this case

I don't know which rules apply to your analysis! I'm guessing they would be more strict but it's not my field. Are you obliged to have 3 ions for each compound?

Which leads me to the following point: Isn't it hard to find 3 daughter ions of the same parent, which have an intensity within 2 orders of magnitude of one another?

For some (small) pesticides, I have trouble finding 2! I can image some drugs of abuse with a rather low molecular weight (MDMA, amphetamine,..).
Poly Styrene, we don't use matrix blanks but interference from other compounds is not the cause as the ionic ratio is typically lower than expected.

Imh thanks for you reply. I think that may be the problem. We use calibrators of 5 ppb prepared with a multicomponent solution. However, although the lowest concentration for all compounds is the same (5 ppb), the signal is lower or higher depending on each specific drug and the response of the instrument to it. Interestingly, THC is the compound with the most confirming problems and also one wich gives the lowest peak and thus area in the chromatogram.

Hello, Rndirk. We run a calibration curve on positive samples. "I would suggest to compare the ion ratios of low concentrations in unknowns to comparable concentrations of that compound spiked in the same matrix." We already do that as the control is processed just like a normal sample and we receive a lot of samples with low concentrations. We normally have the same difficulty in confirming either the samples or the control at these low concentration.

I find very interesting those two rules you have, they would make our work so much easier. We have an acreditation and auditors are very strict but if we find some bibliography justifying the use of a broader tolerance we could do that, of course. If you know it, I would be so so grateful to you.

I am a new employee and the method was already validated when they hired me, but yes, we get three different daughter ions from every drug of abuse, except for one metabolite. In this case we only find two daughter ions: one for quantifying and another one to confirm.

Thanks all three for your soon reply.
Carlos Salas Sierra wrote:
Poly Styrene, we don't use matrix blanks but interference from other compounds is not the cause as the ionic ratio is typically lower than expected.

If the interference is at the mass of the main quantifier ion, then it would bias the qualifier ratios lower, so it could be from a matrix interference.
The past is there to guide us into the future, not to dwell in.
6 posts Page 1 of 1

Who is online

In total there are 11 users online :: 0 registered, 0 hidden and 11 guests (based on users active over the past 5 minutes)
Most users ever online was 234 on Tue Feb 06, 2018 7:33 am

Users browsing this forum: No registered users and 11 guests