The effect of temperature on void volume

[olufadi1]

Hello all. I have recently started doing some HPLC experiments. I had a general question about void volume determination.

I have read that void volume does not depend on temperature (ideally I guess). My void volume marker's (benzene, stationary phase is silica gel) unretained time decreases with temperature. One dissertation I saw mentioned that this may be due to swelling of the stationary phase because of its coefficient of thermal expansion.

My HPLC goes has a limited temperature range (up to 50 C). So I was wondering would it be valid to use the smallest unretained time at (50 C) when determining retention times of other probes at different temperatures? Or do I need to measure unretained times at all my operating temperatures for comparison (maybe due to thermal expansion or some other reason the void time may be changing). Thanks for the help.

[tom jupille]

Void volume determination is a can of worms, as witness several threads from early in the Forum's history:
http://www.lcresources.com/discus/messa ... 20040107am
http://www.lcresources.com/discus/messa ... 20000443pm
Most manufacturers pump specs for flow accuracy are around +/- 0.25%, and bear in mind that most pumps use some form of compressibility compensation to make fine adjustments to the flow. At higher temperature, the back pressure will be lower, which may introduce some flow variation (could be either higher or lower, depending on how the compressibility compensation is set).

Additional thoughts:

Those threads dealt mainly with reversed-phase, but in your case are you sure that benzene is actually unretained? A shift to earlier elution at higher temperature suggests that there might be some chemical interaction going on.

A second consideration is the pore size distribution of your silica. If there is a substantial population of small pores, the benzene may be partially excluded. The use of D2O has been suggested as the optimum (at least for reversed-phase) because it can penetrate all the pores (scroll down about 1/3 of the way through this thread: http://www.lcresources.com/discus/messa ... 20040107am ). As you imply, it's hard to imagine silica swelling enough over a 20 degree temperature range to make a measurable difference.

Bottom line on all of this is that you need to ask how much accuracy you really need for your purpose.

[olufadi1]

Thanks for the helpful response. Just as you said I think the benzene may be interacting with my stationary phase to some degree. From the reading that I had done for normal-phase benzene and toluene seemed to be unretained markers that are used so I decided to go with benzene, since it is less polar. I have seen the used of deuterated solvents also, but my system is only equipped with a UV detector.

In regards to your pore size comment, the silica gel I used has a pore size of 70 angstroms (I am not sure about the size of benzene but I did a quick search and I think its substantially smaller? So maybe this is not an issue?) For my application and calculations I believe I just need something that is consistent and reproducible. Thanks again.

[danko]

What is the mobile phase?

Best Regards

[Peter Apps]

Is the pump itself heated to the same temperature as the column ? If not, then the mobile phase expands as it reaches the hot column, and the volume flow rate increases, giving a smaller unretained hold up time and apparently a smaller void volume.

Peter

[olufadi1]

@danko: The mobile phase is cyclohexane.

@peter: I see. No, my hplc only has a column oven.

[danko]

I think your probe is somewhat retained.
To test this hypotesis, mix some ethanol in the mobile phase 30 - 40% and I believe you'll see retention time decrease.
Ideally I would just choose ethanol for a mobile phase in this particular experiment - if I were you.

Best Regards