"Optimun" Gradient and isocratic condition

[Broesen]

Is there an optimum gradient time? (the time before the compounds starts to elute under isocratic condition.

If yes, isn't it just a waste of time to run and optimize an isocratic HPLC run.

[tom jupille]

There are people who believe that all LC separations should be run under gradient conditions. I am not one of them. Gradient equipment is more complex than isocratic, you have to allow sufficient time for system reequilibration before you can inject another sample, and results can vary due to instrument configuration differences.

In general, if you can accommodate all of your analytes (and "garbage") within a reasonable range of k' values (say, 2-10), then you will usually be better off with an isocratic separation. If not, then a gradient is preferred.

[Noser222]

If you have a method for which an isocratic run would work, then always running a gradient can be a waste of time because you have re-equilibration time after each run for a gradient.

I'm not sure what you meant by the first statement, but depending on the delay volume of your HPLC, you will get some isocratic elution at the beginning of the gradient (analytes with very little retention).

[Broesen]

If you have a method for which an isocratic run would work, then always running a gradient can be a waste of time because you have re-equilibration time after each run for a gradient.

I'm not sure what you meant by the first statement, but depending on the delay volume of your HPLC, you will get some isocratic elution at the beginning of the gradient (analytes with very little retention).

Hi again

My dead volume before the column is around 20 ul and the Flow is 3.3ml/min , so the gradient starts allmost immediately.
Under isocratic condition you will be geting better separation by time but you peaks will also get broader.
So, my point is that, if you can't separate the two peaks by an optimum gradient, there will be no chance that you will be able to do this under isocratic condition.

[unmgvar]

Hello Broesen,

you wrote that you are working with a 3,3 flow/min.
are you working with a monolith column, or with a UPLC?

[HW Mueller]

Broesen, for small molecules with nearly Gaussian behavior the optimum gradient is perfectly flat, that is, no gradient.