Method validation

[krisradh1432]

Dear experts,

I managed to separated my eight interest compounds using standard. However, the compounds show different intensity in fluorescence and UV even the conc. are same. This makes the peaks not fully separate. So, I reduced the conc. of one compound in the standard mixture according to roughly ratio that can be found in sample. The compound usually present in low amount or high when the other two side peaks absent based on literature review. Now, the peaks are separated well.

My questions is
1) Is what I did correct and valid for next validation step?
2) Should I do the validation step by injecting the mixture of standards or by individually the compounds to determine linearity, range, LOD,LOQ and etc?

Thank You.

[DSP007]

Dear krisradh1432
You understand everything correctly. Therefore, in addition to lecture you - not reason and offensive.

Good luck in your work. You come in, if that( any problem)

[Vlad Orlovsky]

UV/fluorescent activity has nothing to do with separation. You need to improve your method and obtain full baseline separation. I would suggest to obtain a resolution of at least 2.5-3 between peaks in order to avoid potential problems in the future, due to lot-to-lot variations, sample matrix interference, and other factors. Please contact me by email and I will try to advice you on how to improve your method.

[krisradh1432]

Dear Vlad Orlovsky.

Yes I really need a guide from HPLC experts as I am new to HPLC and validation. Some more, I do not have friends/lecturer who expert in HPLC to refer whenever I have doubt except online notes and this forum.
Can I get your email. Mine is krisradh1432@yahoo.com. Hope can get your email, so I can discuss more about my work with you,
Thank you

[Rob Burgess]

If all your peaks are the same size (peak, width and area) - then yes a Rs of 2.5 should be fine. But if it is an impurity profile with closely eluting poeaks of wildly different sizes then then resolution values are not so useful. Judging by eye (i.e. baseline resolution) is the best way to "assess" your separation.

[tom jupille]

2) Should I do the validation step by injecting the mixture of standards or by individually the compounds to determine linearity, range, LOD,LOQ and etc?

You should do both.

[krisradh1432]

If all your peaks are the same size (peak, width and area) - then yes a Rs of 2.5 should be fine. But if it is an impurity profile with closely eluting poeaks of wildly different sizes then then resolution values are not so useful. Judging by eye (i.e. baseline resolution) is the best way to "assess" your separation.

http://imageshack.us/photo/my-images/35/unledsz.jpg/
The resolution of certain peaks are below 2.Tried with various mobile phase conditions (water, methanol, acetonitrle) isocratically yet failed, better resolution but the runtime more than 35min using this column.http://www.nacalaiusa.com/product.php?id=48.
All compounds in same conc. but the detection intensity for peak 4 is high. Actually the compounds are vitamin E isomers. Peak 1-delta tocotrienol, Peak 2-gamma tocotrienol, Peak 3-beta tocotrienol, Peak 4-delta tocopherol, Peak 5-alpha tocotrienol, Peak 6-gamma tocopherol,Peak 7-beta tocopherol,Peak 8-alpha tocopherol. The intensity of compound increase from alpha, beta, gamma and delta even they are in same conc. Eventually, peak 4 exists in low conc. in food matrices. That the reason why I plan to reduce the conc of peak 4 to get better chromatogram. (The scale is small, injected low conc.). Is it ok to proceed to validation as I cant further to resolve the peaks? Thanks for the guidance.
This is my post graduate research, I am worry Im revealing much the idea .

[tom jupille]

In a very general sense, validation consists of "demonstrating that the method does what it purports to do". Once you have defined what the method is supposed to do, how you go about demonstrating it depends quite a bit on who wants to know. If you are in a pharmaceutical company or an environmental lab, "who wants to know" is a regulatory agency, and they have fairly well documented rules on what you have to do. In research, the rules are more vague; ultimately they have to be based on doing good science.

Because validation is such an important part of regulatory compliance in pharmaceuticals, many of the comments that you will get on validation are specific to that environment; they may not be relevant in your case.

If I were you, I would get together with my supervisor/advisor/mentor to get clear agreement on what you are trying to accomplish and what you have to do to prove that you are accomplishing it. If you think that a resolution of 1.5 is good enough to let you quantitate *your* compounds in *your* samples, then prove it by validating the method.