prediction of retention for gradient

[Supernova]

Hello,

I am trying to predict how will change my chromatogram during a gradient analysis using the well-know equation:

k* = (Tg x F)/(Delta Phi*Vm*S)

where Tg is the gradient time
F is the flow rate
Delta Phi is percent range in strong solvent divided by 100
Vm is the volume column
S is a constant depending on the MW of the analytes

This equation works very well for predicting the relationship between gradient capacity factor (k*) and gradient steepness (Delta Phi/Tg).
While gradient steepness becomes shallow, retention times increase. However, this reasoning does not work anymore when flow rate is increasing.

Indeed, I was expecting a shorter run time when the flow rate was increasing (or a shorter column was used) but it does not match with the equation since k* is directly proportionnal to F.

I do not know where my understanding is wrong.
Has someone any thought about?

Thanks to all

[tom jupille]

The catch is that k* doesn't correlate exactly with retention time in a gradient system. If you double the flow rate only (i.e., no other change), and if k* stayed constant, the retention time would be cut in half. In practice, the increase in k* partially counteracts that, so the net effect is that retention time is reduced, but not cut in half.

The math involved is daunting. If you want to wallow in it ( ), get a copy of the Snyder and Dolan book on linear solvent-strength gradients; here's a link to it on Amazon:
http://tinyurl.com/2ao2zl

I've always cheated and used a chromatography modeling program like DryLab (or ChromSword, or ACD Simulator, just to be fair) that has the appropriate calculations built in.