Chromatography Forum

I've read in several post across the forum that buffers concentration should be between 10 and 50 mM and that a good starting point is 25 mM.
These concentrations are referred to the full eluent or to the aqueous portion of it?
Means to prepare an eluent 25 mM Acetate buffer 50:50 MeOH/Water, Do I mix 500ml of MeOH and 500 ml of 25 mM Acetate buffer or 500 ml of MeOH and 500 ml of 50 mM Acetate buffer?

Usually (but, alas, not always!) buffer concentration and pH are stated for the aqueous portion of the mobile phase.

By the way, there is nothing magic about that 10-50 mM range nor about the 25 mM. Those just happen to be where most people work in reversed phase with UV detection. You need sufficient buffer to keep ionization of your analytes and residual silanols under control. Increasing the concentration past that point provides no chromatographic advantage, is tougher on the hardware, and increases the risk of precipitation problems.

In principle, you want to use the lowest concentration that gives adequate pH control. In practice, most people don't attempt to optimize; they just use a "safe" 25 mM and let it go at that.

I agree with you. Buffer conc. should be as low as possible.
But optimizing that conc. can be a very long process.
This is the reason I did the question, at least minimize the range of concentrations to test.
Is anywhere a systematic approach to do that optimization?
Thanks you

Start at 0 and do a series of runs increasing the concentration in geometric progression (e.g., 0, 1, 2, 4, 8, . . . mM, or 0, 1, 3, 10 . . .mM). The runs should use "real" samples or spiked matrix blanks.

Talking about real samples... as soon as I lowered the buffer concentration (Acetate pH4.6) to 10 mM. the result for my wine control sample lowered in sorbic content from 186 to 171 ppm (the certified value is 190)
I don't understand what happens.
I use typical conditions: 1 ml/min of 50:50 MeOH/buffer in a C18 250x 4.6 column at 30ºC. Readings at 235 nm.
The most incredible is that I shoot a 100 mg/l Standar before and a 50 mg/l later and recoveries for them were 98-102%. over several injections...

I want to add that peak shapes were perfect in every injection, even in wine

Sometimes subtle peak shape variations can change the integration. If this were my problem, I would try going up to 50 and see if the recovery increases or if it plateaus (I'm assuming the initial condition was 25 mM).

I will work on it. Next week I hope I will have the data.
Thank you again