Dear everyone,

I am trying to use a gradient method (according to to Synder's Practical HPLC method development)as an initial trial method to see if my analytes could be measured by isocratic or gradient elution. However, I have a problem: For Mobile Phase B, I added 10mM phosphate buffer with water, pH adjusted 3.0. That is okay, but for Mobile Phase A, should I add 10mM phosphate buffer into the 100% acetonitrile? I tried, but the phosphate would salt out.(Even if I made a 1M stock, and dilute with water to 10mM, and then add acetonitrile). In addition, how could I adjust pH with pure acetonitrile? My colleague said she has read some paper, and people uses 80% acetonitrile, with no added buffer, and no pH adjusted in Mobile Phase A. However, I am wondering, how would this be correct since there is no buffer, and pH not adjusted in the organic solvent, but there is buffer and pH adjusted in the water part (Mobile Phase B)?

Please advise me of the correct way to do it.

Thank you!

You do not need to add any buffer to Acetonitrile.
I run a similar gradient method for my very polar analytes. Since Acetonitrile is neutral it will not effect the pH of your H20+buffered solvent as it goes threw the column, it will simply dilute it as your gradient increases. The more polar your analyte the sooner it will elute on the column because polar analytes are more attracted to the H20 than the column packing. If you run a H20+0.1% formic acid for Solvent A and Acetonitrile +0.1% formic acid for Solvent B and run a 10-95% B gradient you can see where your analyte will elute. If your analyte is acid sensitive you can run the above solvent without formic acid. If analyte is very close to solvent front, a phosphate or formate buffer will increase your analytes attraction to the column vs solvent, therefore giving a longer RT. A note: the pH of H20 is around 7, by adding 0.1% formic acid/1L to your solvent the pH of H20 is lowered to 2.7. Be sure your column is stable at these pH's. Most RP columns are stable from pH 2-9, but you should check your column description sheet.
For help making buffered solvents check out this web site www.embl-heidelberg.de/JaMBW/5/4/

Personally, I would ramp from 10% mpA to 70%mpB at a shallow (20mins) and steep gradient(60mins). Make sure all the compds elute within these gradients. Evaluate using DryLab (above 2 runs) or visually examine to determine the appropriate condition.

Sorry, I would use 10% organic/90% aqueous to 70%organic/30% aqueous (buffer should not ppt at low ionic conc).

Is this June or 1st of April? Are you starting with B? Organics and pH was extensively discussed earlier. If you get your sequence straight you have to make sure that you stay far away from precipitating mixtures, thats all.

I was going to sit this one out but HWM encouraged me.

The addition of something other than water will most definitly affect the proton activity, and pH of a buffer/organic liquid mixture. Whether it affects the separation remians to be demonstrated.

For example, in water the strongest acid one can have is H3O+, protonated water. In a solvent that is a weaker base than water, a stronger acid than H3O+ can exist. (Silica columns are not run under such conditions so this tid bit is irrelevant to most chromatography.)

Dielectric strength of the organic solvent is important. So for example, HSO4- is a fairly strong acid in water. In low dielectric solvents this salt is nearly neutral because in a low dielectric solvent it is energetically unfavorable to dissociate the H+ for the double minus SO4--. Imagine the situation with phosphate.........

And then refer back to previous discussion for other reasons mixing an organic solvent with an aqueous buffer will affect pH, proton activity, etc. It probably won't affect the separation because in most cases one adjusts conditions way beyond the minimum to fully protonate or neutralize the analyte.

Possibly, I was misunderstood. My last sentence above refers to the original question as to whether one ALWAYS needs added buffer in the organic modifyer (second moph in a gradient) ....... One does not if one stays far away from precipitating mixtures. For instance: if one starts with aqu. buffer and needs to go to a max. of 20% aqu. buffer + 80% CH3CN one can attain this with the second moph being pure CH3CN, but ending the gradient at 80% CH3CN.
As far as pH and organic modification is concerned there exists a nice illustration in a slide seminar? by Uwe Neue, et. al. In the Waters web page I found only a strongly abbreviated form of this (pdf file 148K).
Hans

HWM.....just kidding. Should have added a :)