1. Retention can sometimes change dramatically with the degree of ionization of the analytes, and this changes most rapidly when pH = pKa. Because pH is a logarithmic function, +/- 1 pH unit changes the %ionization from 10% to 90%.; +/- 2 pH units changes the %ionization from 1% to 99%. Therefore, the best general recommendation for initial work is to stay at least 1.5 pH units away from the pKa of your analyte(s) if possible.
1a. In reversed-phase chromatography, hydrophobic ("non-polar", "greasy", . . . take your pick) molecules are more strongly retained than hydrophilic molecules. Given a choice between the ionized and neutral forms of the molecule, the neutral form (free acid or free base) will be more hydrophobic and therefore more strongly retained than the ionized form (anion or cation). Therefore, it is usually best to start with a pH at which the neutral form predominates (relatively low for acids; relatively high for bases). In ion-exchange, neutral species are unretained, so you would go the opposite direction on pH.
1b. Once you have decided on a pH, you need to find a good buffer at that pH. Remember that buffers are themselves weak acids or bases; they are most effective at their own pKa, and become ineffective more than 1 pH unit or so away from their pKa.
1c. Once you have a list of candidate buffers based on their pKa, you can weed out choices based on detector compatibility. Many organic buffers, for example, have fairly high UV cutoffs. If you are chromatographing proteins and detecting at 280 nm, this is no problem. If you have to detect at 210, then it is a problem. If you will be doing mass spec, then the buffer must be volatile. Most of this aspect is simply applied common sense.
2a. Yes, mM = millimolar. Be careful, however, because there is a bit of ambiguity here. I have always interpreted the buffer concentration and pH as being defined with respect to the aqueous part of the mobile phase (i.e., before any organic solvent is added). A substantial number of people interpret concentration and pH as being defined with respect to the final mobile phase (i.e., after addition of organic). Hopefully, the interpretation will be clear from context (e.g. "50% acetonitrile + 50% of a 25 mM, pH 2.5 potassium phosphate buffer" is different from "25mM, pH 2.5 potassium phosphate buffer in 50% acetonitrile/water"). If it's not clear, you may have to find the original method and look at the detailed procedure. Which brings up another point: if you are documenting a procedure, be as explicit as possibe in describing the mobile phase preparation!
2b. The buffer concentration needs to be high enough to prevent pH changes caused by the presence of the sample. This is usually in the range of 10 to 50 mM (in the aqueous part of the mobile phase!). I'd say the most common values are 25 mM for UV detection and 10 mM for mass spec. You could often get by with less, and you may occasionally need more, but most people don't attempt any optimization of the buffer concentration unless they have problems.
3. If your sample contains only neutral (non-ionizable) compounds, then a buffer is usually not necessary in reversed-phase. Otherwise, use of a buffer is highly recommended.
4. Buffer pH is usually selected on the basis of separation chemistry (see responses 1, above). One good rule for HPLC is "to get the best results, disturb the equilibrium of your system as little as possible". This suggests that the sample should ideally be dissolved in your mobile phase. If the sample must be dissolved in something other than the mobile phase (a buffer at different pH, in your example), you should not have any problems so long as:
- the injection volume is low and/or
- the pH difference is not too great and/or
- you have sufficient buffer capacity in your mobile phase to deal with the upset.
Finally, don't apologize. Basic questions are what this section is all about. We all had to learn either by asking questions or making mistakes (the former is a lot easier and cheaper!).
