Chromatography Forum

Dear,

I optimised a LC-MS/MS method on a 2.5 µm, 3.0*50mm C18 column.
I would like to compare the sensitivity of this method with some other columns (also C18, but smaller diameter and/or particle size).

To have equal separations, I would have to keep the linear velocity constant. This means adaptation of flow rate when i.d./dp are changed.

However, my MS/MS method is optimised for a flow rate of 0.5 ml/min, so I would like to keep this unchanged.
If I simply keep the flow rate and adjust the gradient (i.e. keep the amount of column volumes per minute constant for all columns), are the separations then still equal (regarding to matrix effects etc)?

Thanks!

At the first, i think the seperation is not so important in the lc/ms/ms,if you have different cloumns,even through the flow rate is same, the seperation is also different. but the ms/ms septerca is ok.

Your using gradient elution and 50mm length column. Under these conditions,
I'm not sure there's a significant difference between sub 2um and 3um particle sizes.
However, the difference in system pressure will be significant.

3mm I.D. (3um particle) is optimal for 0.5mL/min flow rate. You can try increasing inj. volume to increase sensitivity.
If you notice tailing, adjust sample diluent / try different column brands.

If you're limited in sample amount, then 2mm I.D. can be useful.

Scaling at the same flow rate is fine, but you need to consider the consequences, and check if your instrument is capable of doing this.

If you use a 2 mm column instead of a 3 mm column at the same flow rate, the backpressure will increase 2-fold, even if you use the same particle size. The analysis time will be 2x shorter. The sensitivity will increase a bit less than 2x, but it will still be higher.

If you decrease the particle size of the column, the pressure will increase with the square of the ratio of the two particle sizes. The velocity is actually correct for a change from 2.5 to 1.7 micron particles, if you stick with the same flow rate and change the column diameter as described above. However, your total pressure will be at least 4-fold higher compared to the initial pressure. Your sensitivity will increase by probably another 50%, but this depends on a lot of other things as well.

If the increased pressure does not bother you, go for it!

Just one more remark about this: I kept the flow rate and injection volume the same when comparing different columns. And I also kept the gradient & run time exactly the same (because the pumps still need to be able to get it mixed...)

Then I did the comparison, is this a good way to compare the sensitivity of the different columns?

Thanks

Some people may rip their hair out, but a comparison is a comparison.

If you give me a rough idea about the gradient range, gradient run time and the molecular weight of your analytes, I can tell you what to expect.

Also, please keep an open mind with particle size. I can show you examples of exceptions to the rule.
Meaning, a well designed 3um particle HPLC column can result in improved sensitivity over sub 3um or (sub 2um).

For example, a 3um HPLC column that shows less tailing then the sub 3um column for a particular
solute will result in improved sensitivity. It happens.

The most important difference between the two separation is that you are changing the separation regime. To compare the same gradients, you should have kept the ratio of gradient volume to column volume constant. If you are running the same gradient time, you are expanding the gradient for the smaller-volume column, which is not a valid comparison, since you are diluting your analytes more.

I had a gradient which goes 25% to 90% B in 2 minutes and stays at this level for 1.5minute. Total run time is 6'. Analyest are 200-400 Da.

If adapting the gradient for smaller columns, this would mean a shorter analysis time. I don't know if the pumps can mix the mobile phase then appropriately and if the pressure won't be too high. That's why I kept all the LC conditions equal...

If you have a well working instrument, you should get a peak capacity of around 60 for the 2.5 micron 3 mm column and one around 105 for the 1.7 micron 2 mm column. With other words, the peak width for the 2.5 micron column would be around 2 seconds, while the one for the 1.7 micron column would approach, but not quite reach one second.

The pressure is a function of the flow rate. So if you run a shorter analysis time and keep the flow rate constant, the pressure will not change for the same column.

Ok, I understand your point: by keeping flow rate and gradient the same, the separation regime is changed and cannot really be compared.

But, by keeping the injection volume the same, you could also argue that separations are not comparable then since the relative amount of analyte on column isn't the same...

If my goal is just to have the most sensitive possible method, can't I reason that since there are so many variables, I decided to keep the LC parameters unchanged and choose the optimal column for these parameters...?

Thanks for your help

And another remark: I am afraid to suffer more from matrix effects when shortening gradient...

The separation in the second case will be better, and you will have less matrix effects (if randomly distributed in the chromatogram). If you would sharpen the gradient for the narrower column, the expected matrix effects would be about the same for both columns.

But this is neither here nor there. I did not answer your orginal question completely: since you gain nearly a factor of 2 in the peak width with the narrower column, you can expect nearly a 2-fold increase in sensitivity.

The reason that you do not gain more is due to the technique used for the comparison. But, as I said above, while complicated, there is nothing fundamentally wrong with doing it this way.