Experiment design

[kobie]

My project is to optimize the conditions of analyzing B vitamins and I am in the process of experiment design. I have some questions that need suggestions from the expert on the forum.
So the parameters for the optimization includes columns (I would like to test about 5-6 columns available in the lab), pH from 2.5-6 unbuffered v.s buffered and finally mobile phase solvent (methanol, acetonitrile and isopropyl alcohol)
-starting with buffered mobile phase at different pH (2.5, 3.0, 3.5, 4.0, 5.5 and 6.0) and B as Acetonitrile for all columns. Then from the results, I will decide the pH range that produces the best results and then rerun with equivalent unbuffered mobile phase for all columns.
-Compare the chromatograms under buffered and unbuffered conditions within the same pH range.
-Determine which conditions (buffered or unbuffered) and which column produce the best results.
-Try other combinations of solvents (methanol, acetonitrile and isopropyl alcohol) and acids (TFA, formic and acetic)
So if I am following this experiment design, right in the beginning I assume that ACN, MeOH and isopropyl alcohol behave the same for all columns, which is not true.
This is the only design in my mind right now. I am still confused about a lot of things so I would like to receive insight and suggestion from the forum. Thanks a lot!

[HPLCaddict]

Without being an expert in analyzing B vitamins some general hints:

- (Most important one ) See if you can get access to a good book about HPLC method development. "Practical HPLC method development" by Snyder & co. is THE ressource for your task. "High performance gradient elution" (Snyder & Dolan, I think) is another one.
- You said you're task is to optimize the analysis. So there is a present method? Maybe you could start from there...no need to reinvent the wheel.
- You've got equipment capable of gradient elution, I suppose? At the beginning, don't bother with isocratic elution, try wide range gradients (something like 5-80% organic) for scouting.
- For buffers of different pHs you should use different buffer salts (e.g. phosphate vs. acetate)
- If you see that your analytes are sensitive towards pH, don't even think of using unbuffered eluents. Lacking of robustness is a nightmare for everyday practical work.

[kobie]

Thank you for your reply HPLCaddict,
Available methods I have found involve the use of ion pair agents which I try to avoid while some others run at very low aqueous phase (down to 0.5%) with C18 columns (which are the only kind of columns I have the access to). So I want to see how other factors like pH, solvent or buffered conditions improve the separation without using that low aqueous phase.
I will find the book you recommend for more insight. Thanks for the suggestion.
My HPLC system has the capability of gradient elution so I will take advantage of that.
You mention the sensitivity of the analytes towards pH. Would you mind elaborating some more about this?

[HPLCaddict]

The B vitamins comprise a class of substances that have very different properties. Some of them are very polar which means they are hard to retain on C18. That's why those methods use ion-pairing or mobile phases with very low organic content. Be careful with C18 columns and low organic mobile phases. Generally, "normal" C18 columns should not be used with mobile phases of <5% organics. There are C18s which can tolerate even purely aqueous phases. Check with the column manual or the manufacturer if you're unsure.
Concerning the pH, if your analytes are ionazible (meaning they contain acidic and/or basic groups) their retention time may change strongly with differing pHs. That's why you should use a buffer to stabilize the mobile phase's pH and thus the retention times.