The silica gel that you are using is very fine one. Hence, the flow will not be very fast.

You can use solvents that are not very volatile to maintain the elution and flow rate.

Also, the column should never be left dry.

The silica gel that you are using is very fine one. Hence, the flow will not be very fast.

You can use solvents that are not very volatile to maintain the elution and flow rate.

Also, the column should never be left dry.

Consider the Van Deemter equation. If you are set on your solvent and column, but the flow rate is the last variable. The only definitive approach is to run a set of standard trials at varying flow rates. The resulting plot, example below, will allow you to make the decision of the balance between speed and resolution that suites your purposes. If you're going to do lots of runs, it may be worth the investment of time.

Image taken from: (https://teaching.shu.ac.uk/hwb/chemistry/tutorials/chrom/chrom1.htm)

Conflict of interest disclaimer: I do work for Teledyne Isco, mentioned below.

That silica is the same as that packed into the flash columns such as those sold by Teledyne Isco and others. If you look at their column dimensions and flow rates, you'll probably see a solvent velocity of ~7 cm/minute, this will be fairly close to the van Deempter flow rate since they want their users to have good resolution. This way of stating the flow rate allows one to more easily scale the column as the column width changes, and you'll note the reply above mine uses solvent velocity as well. For a 5 cm diameter column, that is around 137 mL/min assuming I did my algebra and math correctly. Make sure you run the column isocratically to determine the optimal flow rate.

If you can get a copy of How to Use Excel in Analytical Chemistry and General Scientific Data Analysis by Robert de Levie, Cambridge University Press, you'll find the calculations that you need (Chapter 6.9 in the 2001 edition).

SolventVelocity = (FlowRate * 4)/( PI * ColumnDiameter^2)

Or use this link: https://www.sensorsone.com/volumetric-flow-rate-and-diameter-to-flow-velocity-calculator/

Along with the Van Deempter curve, the best way I have seen is to do a number of sequential runs at increasing flow rates. Once you have those data, you can make a better decision about the flow rate you want to choose.

Hello, are you sure that your column 5cm in diameter? Maybe 0.5cm. ? Can you explain more about your chromatography system with 5cm diameter column. I know glass column with 0.5cm diameter with packing like you mentioned which require 50ml flow rate.

If 5cm column with mesh 200-400 you need at least 700ml/min gas flow. But for pressure stability and separation efficiency calculation you need more and it should be calculated experimentally.

Thank you all for your excellent suggestions.

Jack Silver and Karen A. Darbinyan : I am using a gravity column. I was wondering how to set the flow? Completely open the bottom tap or not? Or maybe it doesn't matter at all? I've read that if the flow is too low, it causes diffusion and if its too fast it there is not enough time for equilibration.

For that type of column, my opinion is to open the flow completely. I usually associate "equilibration" with the solvent and column being in an equilibrium; a fast flow might reduce interaction with the column. You'll probably find, under gravity conditions, the flow will be somewhat less than the ~140 mL/min mentioned earlier. My experience is that how well the column is packed will be far more important than the flow rate. Sample loading is very important too. Many people dissolve their sample in a strong solvent which "drags" the sample down the column and reduces resolution. In general, "dry" loading gives better resolution.

Thank you for your advice. It helps a lot.

Though I've never had a gravity column with a flow rate of 140 mL/min. It is much much lower even with a completely open flow.

Zahra Nazemosadat usually, one needs to supply air pressure or a pump to get the flow that high. Ideally, you compound should elute at ~2 to 5 column volumes (CV). If you use a TLC plate with the same solvent system as your column, that would be an Rf of 0.2 (5 CV) to 0.5 (2 CV). You might have noticed that CV = 1/Rf.

Thank you Jack Silver .