split peaks

[davo_m]

Hi, I am having some troubles with split peaks which I cannot solve and I am hoping that someone may have some suggestions.

I am running digested DNA (deoxyribonucleotides) on a Supelco LC-18-DB column with a guard column. As a reference I have been using dNTP's that are treated to resemble the digested DNA.

I use an isocratic run of 0.03M NH4H2PO4 with 2.5% methanol. In between each sample I wash with 66% methanol and 34% water.

I now seem to get split peaks, and broad peaks with a shoulder on every single peak.

Longer washing periods didn't seem to help so I upped the wash to 100% methanol each time, and then re-equilibrated my column with the running buffer. The next sample worked brilliantly, however it the broad and shouldered peak returned. 100% methnanol washes are no longer effective! I have tried very long wash periods, temporarily reversing the flow through the column and lowering the injection volume (only 10ul to begin with). The sample is mixed with the mobile phase prior to injection, so it shouldn't be caused by differences with the injection solution.

Is it possible that the column is no good? It has seen a fair few hours of work now.

I am stumped as to what would have caused this after running smoothly for some time. If anyone has any suggestions I am more than willing to give them a try.

Many thanks,
David

[james little]

I have had a lot of bad experiences with guard columns.

Why don't you remove the guard column and try with just the column.

[HW Mueller]

james,
I am also no great friend of guards, but this type of substrate seems to be THE reason for using them. Thus: You have examples where the guard gets fouled but not the column? Over-guarded?

davo_m,
the column works if you replace the guard column?
I have no experience with DNA, etc., but proteins can be cemented onto the column by organics like methanol. Surely there must be a better dissolution media for DNA also.

[tom jupille]

David, what you're describing is a textbook example of either a partially-plugged frit or a void space at the head of the column or guard.

Does the problem go away when you put in a new guard column? If so, then the guard column is doing its job.

If the problem persists with a new guard column, but you have already gone through several guards, then your column may simply be at the end of its useful life, as you suggested.

[davo_m]

Hello everyone.

Thanks for the suggestions! I have removed the guard column completely and that has fixed the problem nicely.

I will replace and re-attach the guard soon because it might have saved the column from plugging up.

Many thanks for the help. I thought the guard column was there to help me.

David

[HW Mueller]

Didn´t it?

[WK]

I use a disposable 0.20um frit in front of the column and this traps most particles - never used a guard column. But I suppose your samples might be nasty which will require regular guard column replacement. You could try running with just a frit. Do you/are you able to filter your samples?

[tom jupille]

Many thanks for the help. I thought the guard column was there to help me. Rolling Eyes

It did. The whole point to a guard column is that it "takes the hit" instead of the analytical column. It's not unusual to change the guard many times during the life of an analytical column.

[davo_m]

Yes, I suppose it did help me.

I had only recently replaced it, so perhaps there was something 'chunky' in a sample that caused the problem. I will try using a disposable frit as WK suggests as it is likely to be plugged again soon.

My samples are 10ul of a DNA digestion so filtering is nigh impossible.

Thanks yet again for the helpful insight, I appreciate it very much.

Dave