I'm in the process of trying to reproduce an proceedure that uses a 5 cm guard column and a 25 cm "working column." For various reasons, we're not using a guard column. We have a 25 cm column (also have the option of getting the 30 cm column) and cant get anywhere near their resolution. Every, and I mean every otherparameter is the same (linear flow rate, gradient composition, gradient time, dead time etc). Is our resolution so much different just because of the 5 cm guard column. I would have thought that the extra connections would have increased B.B. and not added enough of a difference. Has anybody seen anything like this or have any ideas? Any advice would be helpful.
Thank you

Dear Anonymus;

I very much doubt that a 5cm guard column can make such a difference. The only way I could understand this is if the guard column contains a totally different adsorbent from that in the analytical column.

Most likely your problem is somewhere else. If you are getting basically the expected retention, then most likely the lack of resolution you observe is because the LC system has too much dead volume (particularly true if you are trying to resolve small signals), or because the sample has been improperly prepared.

Another possibility is that you have encountered a common problem in LC. This is, column reproducibility. If the procedure you are following is more than 10 years old, then it is likely that new columns are just too different now, and quite probably the method was not very robust to start with.

Try other columns of the same type, perhaps a different supplier. Investigate if it is possible to find an old column. Try some mobile phase and temperature changes.

Good Luck;

Benjamin

First, let me ask you if the 5-cm precolumn and the 25-cm main column contained the same packing. If this is not the case, then you are getting the influence of the different selectivities of the precolumn and the main column and there is no way that you will get the separation from the main column alone.
If the precolumn and the main column are made from the same packing, the issue might be the scaling of the gradient. Before blaming the column, I recommend that you scale the gradient properly. Assuming that the precolumn and the main column have the same i.d., you need to scale the flow rate by 5/6 of the original flow rate. This gives you the identical elution profile (and hopefully identical retention times) as you had on the 30 cm column.

Uwe,
If you're scaling the flow rate by 5/6 to accomodate the new column size, shouldn't you also scale the gradient times by 6/5 to ensure continuity? You won't get the same run elution time, but you should get the same relative k's this way, no?
Chris