Simultaneous quantification of different analytes

[kobie]

Hi everyone,

In guideline for chromatographic method validation by FDA and others, it is recommended that peaks should have retention factor of more than 2 in order to be quantified. In many cases where simultaneous determination of different analytes is done, I wonder if that requirement is the same. I am doing simultaneous analysis of water soluble vitamins (with at least 4 at the same time) and due to their different chemistry and interaction with the column packing and mobile phase, it is impossible to satisfy the requirement of k>2 for quantification.
I have tried to search for answers to this question but cannot find a satisfactory one to this issue anywhere. I would like to hear insight from the experts here.

[Vlad Orlovsky]

it is possible if you are using the right column:
http://www.sielc.com/Application-HPLC-S ... ities.html
http://www.sielc.com/Application-HILIC- ... oup-B.html

[michaelbarnes42]

If you are unable to get proper seperation in one run due to varying chemistries, I would think you would have to do 2 or more runs to do a proper analysis

[kobie]

Thank you for the feedback!

@ Vlad Orlovsky:
My methods have been optimized to include more vitamins than the ones you provided in the link and I managed to avoid the elution in the void volume. Actually all of the compounds of interest come out after t0 but the early eluters have the retention factor of about 0.3-1.5 which is less than 2.

@ michaelbarnes42
I think it is a good point. I guess I am curious to know how people deal with simultaneous analysis in real world where time-efficient methods are favored. I wonder if it is a hard and fast rule to obtain peaks with k>2 before quantifying them. I mean it is not easy to achieve that for simultaneous analysis.

[Vlad Orlovsky]

can you list all vitamins you have? I would like to see if we can develop a method for your mixture (free of charge).

[michaelbarnes42]

Thank you for the feedback!

@ Vlad Orlovsky:
My methods have been optimized to include more vitamins than the ones you provided in the link and I managed to avoid the elution in the void volume. Actually all of the compounds of interest come out after t0 but the early eluters have the retention factor of about 0.3-1.5 which is less than 2.

@ michaelbarnes42
I think it is a good point. I guess I am curious to know how people deal with simultaneous analysis in real world where time-efficient methods are favored. I wonder if it is a hard and fast rule to obtain peaks with k>2 before quantifying them. I mean it is not easy to achieve that for simultaneous analysis.

I think I misunderstood you, I was thinking that your peaks were coming out with your solvent or basically unretained. If that were the case your k would approximately 0. Yes, I know solvents can be retained, just most of the methods I use, have solvents that are unretained.

I use peaks with a k<2 all the time. Unless a method specifically states k>2 I really don't see it as a requirement. As long as you get good seperating and 'clean' peaks, I see no problem using peaks at k<2, I mainly work with pharmaceuticals, so this may not apply to what you are working with.

I am not an expert, but regarding FDA and USP I don't see anything in the USP prohibiting the use of peaks with k<2.

[kobie]

@ Vlad Orlovsky
Thank you very much for your offer. I am trying to analyze all water soluble vitamins and among them, I have the most issues with vitamin B1. The purpose of my project is to develop a method that can be conveniently transferred to different detector systems. So far I have managed to have satisfactory resolution using YMC Pack Pro with 0.025%TFA and ACN as mobile phase. Even with TFA, the retention of B1 does not improve much. I got the retention factor of about 0.3, which means it just comes right after the void volume. Another concern is that I don't know if TFA at such a low concentration 0.025% is gonna suppress ionization much or not.


@ michaelbarnes42
Thanks for your input. Actually this is just a part of my project for my master degree but I am interested in working in supplement/pharmaceutical industry so your feedback is helpful to me. As to the FDA ans USP guideline, I think they are more for analysis of individual analytes. I agree with you that they don't clearly prohibit the use of peaks with k<2. The guideline I found here only states "Recommendation" with regard to those validation parameters.
But then I wonder how you decide if a peak is "clean." One of my analytes has the k value of about 0.3 which means it just comes out of the column right after the void volume. I want to quantify it but still unsure about how substantiate that this peak is pure. In your practice, do you have use any more realistic and acceptable range in terms of this requirement on k? I mean during the method development process for the problematic not-well-retained peaks, do you have any cutoff, let's say 0.5 or 1 or maybe as long as the peak comes after t0, to call it's a wrap?
Thanks again for sharing!

[Peter Apps]

[quote="kobie" with 0.025%TFA and ACN as mobile phase.
quote]
Is there a reason not to run a gradient ?

Peter

[HPLCaddict]

I see k>2 as a reasonable suggestion so you can be quite sure to have no interferences with t0 noise. When developing methods i usually shoout at k>2, with rather "clean" chromatography I'd go down to ~1.3-1.5. But I have a bad feeling in my gut to go even lower. A k of 0.3 would give me a bad headache as it's close to non-retained. Yes, it might look clean now, but you'll never know what will be in the future. And if you're in a regulated environment as I am, you're usually not able to adjust your method just in order to move you're peaks a bit.
Even with everyone screaming for fast chromatography nowadays, I'd invest that extra little run time in order to increase the robustness of the method.
Concerning analytes of widely differing elution characteristics, as Peter already pointed out, what's wrong about going the old fashioned way and using gradient elution?

[Consumer Products Guy]

We regularly assay an SPF product for 4 sunscreen actives, one component of which has two distinct peaks, isocratic in under 15 minutes. We've cGMP-validated the assay for each component, have resolution of about 3 between all the peaks.

We do system suitability on all 5 peaks.