Chromatography Forum

This may seem like a naive question but is it possible to use 1.8um silica columns on a 'conventional' LC system - or am I forced to purchase a UPLC system. That's my second question - so people think it is worthwhile to purchase UPLC at this time. I can see how added sensitivity would be an advantage but the very rapid analyses is not such a big issue.

Many thanks

John

Hello John,

You can use these columns without any problem on a conventional system. The only drawback is that you’re not able to get fully benefit of the features of sub 2um material.

It is like putting a Ferrari engine in a Ford.

The favourable flow area at the end of the van deemter cuve is now neglected.

Nevertheless if you are interested in higher sensitivity, the trick is to let your compound elute in a smaller volume. This is what happening, most of the times, when using shorter columns or at elevated temperature.

In my opinion that the gain in sensitivity between a 3.5um and 1.8um of the same dimensions is marginal.

The other thing you have to be very careful about is extra-column volume. The obvious concerns are the injector loop (it should roughly be scaled to the column volume) and the detector cell (in principle it should also be scaled to the column volume; in practice you're stuck with 8 μL or so), but people often lose sight of things like connecting tubing, preheater tubing, etc. Standard 0.010" id tubing has about 0.5 μL of volume per cm of tubing length. No big deal with a conventional 150 X 4.6 mm column, but it can kill a separation carried out on a small column.

We presented a paper at PittCon earlier this year that had examples of some of the problems. It's still available for download from our web site (http://www.lcresources.com ; click on the "Resources" link, then on "Exchange"; download "paper 2002-6").

It also depends on what type of system you are running. The 1100 can be tightened to really cut down on dead volume. The Alliance system really won't do it. Tom pointed most of this out, but you can specifically, 1) remove the mobile phase mixer; 2) use red peek tubing from pump to autosampler; from autosampler to column; from column to detector; 3) use 20uL sample loop; 4) Use 0.007â€

I all agree with all of you, especially the presentation of Tom is crucial.

Nevertheless, you have to think over what is a fast analysis?

1. Are these sub 1min analysis;
2. Is this just reducing the analysis time of an existing method.

For sub 1 min analysis every extra nL is killing. This is less the case if you are able to go down from 30 min to 6 minutes.

Thank you all very much for that information. I will try what you have suggested.

Tom - I've downloaded your paper which is exactly what I needed and I will be passing on the link to our group - thank you.

I will probably try to get my bosses to invest in a UPLC next year. I'll use the argument that we don't want to get left behind.

Thanks

John

I think you may have missed some of the details here. You can achieve much of the UPLC/1200 SL performance with 2-2.5u materials with out the backpressure you get with sub-2u. You don't need high pressure, just better performance. You can do this already (or get pretty close).

I wouldn't invest in the UPLC until I/you have explored all the other options first. Most I know aren't even using the UPLCs they were "sold" on anymore.

We need to be clear about what is possible and what is not.

In order to take full advantage of either the resolving power or the speed that comes with the use of sub-2-micron particle columns with a rational length (5 cm and more), you will need a UPLC system. A simple calculation of performance vs backpressure will tell you that. Larger particles can be used with a lower pressure system, but this is going at the expense of either run time for equal performance or performance at equal run. Both are worse with larger particles.

Sensitivity is not so much a question of particle size, but a question of short thin columns in combination with a detector that is designed for that purpose. The UPLC system has been designed to combine all these features, including a unique detector design that maximizes the sensitivity despite a small sample volume. Of course, one does gain a bit from the higher performance of the columns as well.

Have you tried 3um technology? You can get really - really fast
separations using our 3um technology (without sacrificing resolution or gaining all that backpressure). These columns are manufactured in Japan by Imtakt - Check this out:

http://www.imtakt.com/TecInfo/TI202E.pdf

Whatever one can do with 3 micron in 20 seconds, one can do with 1.5 micron in 10 seconds. Please, let us not get confused!

>one can do with 1.5 micron in 10 seconds. Please, let us not get confused!
I'm curious about the performance of 1.5um in 10 seconds. Could you let me know the chromatogram ?

The principle is clear: you can change the speed of any separation by reducing the particle size.

Here is the example requested:

It looks nice! Thank you.

>The principle is clear: you can change the speed of any separation by reducing the particle size.

For customers, practice is more important than principle. So, if you know, please provide this 1.5um product information and analytical conditions of this data. In addition, please let us know how many injections we can operate under this condition.

Yazawa,

This chromatogram was executed on a 1.7 micron ACQUITY UPLC column with the dimensions of 2.1mm x 20 mm at a flow rate of 1.9 mL/min with a gradient profile from 25% to 95% MeCN over 6 seconds. The purpose of this chromatogram was a demonstration what can be done. It is not a recommendation what one should do for a routine method, as was inquired above.

The thought behind all of this can be found in the following publications:

U. D. Neue, Y.-F. Cheng, Z. Lu, “Fast Gradient Separationsâ€

Neue,

Thank you for a lot of information.
But ...

>Whatever one can do with 3 micron in 20 seconds, one can do with 1.5 micron in 10 seconds. Please, let us not get confused!

I am also confused where is 1.5um in 10seconds. Because Acquity is 1.7um.