sample traps versus columns

[Wolfgang]

hello all,


I was wondering what the rationale is for using sample traps when using LC/MS. What is the advantage of having a C18 sample trap prior to a C18 column? Why are the dimensions of a sample trap so small? Are appropriate flow rates applicable to sample traps as they are for HPLC columns? I just read a paper where they used 100um i.d. x 10mm length sample traps and their flow rate for sample loading was 3uL/min. That sounds way too fast for a 100um i.d. column. Any help will be appreciated.

[Kostas Petritis]

Wolfgang,

May I ask what was the I.D. of the separation column? There is a rational behind it (at least for the nano-LC-MS for proteomics applications).

Kostas

[Uwe Neue]

I do not know the context of the publication that you are referring too, but we have worked extensively with something that might fall under the category of sample traps. We use very short and thin columns, at very high velocities to either enrich a very dilute sample or to clean up a sample to reduce background and improve sensitivity. Then the sample can be eluted either directly into the MS or onto an analytical column ofr further separation.

[Kostas Petritis]

Wolfgang,

Uwe partially covered the subject.

In addition if you are working in the field of proteomics then:

We tend to inject large volumes of proteins digests of a certain concentration (about 1 ug/uL) because of peptide solubility reasons etc. In 50 um ID columns we inject about 3 uL. If you use a flow rate of about 200 nL/min it will take you about 15 min to load all this amount and you will further need time to wash in order to get rid of salts and other non retained compounds (we tend to do so even if we do SPE clean-up).

In our case where we are doing proteomics in very high pressures (10000-20000 psi) we pack very long columns (i.e. 80 cm) where the "void volume" starts to be significant (unretained compounds may take 15 min or more to go through the column) so if you want to wash your column your throughput goes significally down.

Finally, by using the trapping column you significally incrase your separation column lifetime (especially if you are not doing SPE clean up before).

For the above reasons (high throughput, high injection volumes, separation column lifespan, sample clean-up) people use trapping columns where as they do not care about separation in that step they can operate in higher than the optimum linear velocities. This works for small molecules and peptides, I do not know about proteins that have much smaller mass transfer properties...

The industry standard right now is 300 um trapping columns and 75 um separation column but this will further go down. We were able to go down to 15 um ID by using 3 um silica particle size but the packing of these materials is quite difficult (especially if you want to pack very long columns). Hopefully, smaller particle size materials (<2 um) with narrow particle diameter distribution will make the packing of these columns easier).

See Anal. Chem 2002 74 4235-4249 for relevant information (MS sensitivity, small column ID, injection amounts etc...)

Kostas