Chromatography Forum

JM might be on to a global solution for problems like this - i.e. those that are caused by column deterioration due to something in the mobile phase.

Conventionally we protect out columns by putting a guard column in the sample stream - the weakness of this approach is that whatever bad things were happening to the column now happen to the guard column, and they stilll affect the separation.

If we put the guard column upstream of the injector (and still downstreamof the pump in order to get the necessary flow and pressure) the guard column will still "tame" the mobile phase but if it gets clogged, cavities or channeled this will not affect the separation.

This is exactly the approach that was used in the form of saturator columns to stop aggressive mobile phases from dissolving silica columns.

It might be worth a shot ............

Peter

You said that you damaged several guard columns - do you use the old style (where the guard column is actually a column)

The newer ones use a cartridge system where the 'column' is about the size and shape of a watch battery. Not too $ and they are generally discarded after every run.

Only questions, no answers I'm afraid.

Do you premix the solvents, or rely on the pump?. What temperature is your column and lab solvents ( eg solvents brought in from stores and used before warming up )?. Are any of the solvent handling systems different ( polymeric, metal rather than glass )?. Did the column backpressure increase the same for good and bad batches, and was it increased when peak splitting occured. Does increasing the column temperature by 10C change the split peak behaviour of good or bad solvent?.

If your earlier methods worked fine, what were the sample solvents and mobile phase systems, and if they included THF, does using a bad batch also kill those analyses?. Were the sample concentrations the same?. Have you tried diluting the sample and injecting more?. If you perform multiple injections of blank sample solvent, after a couple of samples, then repeat those samples, is there any indication that the quantity of bad solvent will change the peak splitting onset.

Can you analyse the rubbish on the front of the columns - eg back flush then use nmr or MS. Was there any comment on the nature of the rubbish ( particulate, goo) ?. Is the rubbish soluble in a slightly more aggressive mobile phase, or if the pH is adjusted?.

The peroxide value of the "good" batch obviously has climbed, as it's exceeds the Merck spec ( 0.02 ), so it's hard to make any meaningful comparison, but I'd be asking Merck to check inventory for lots more .

My recommendations would be to premix the mobile phase, allow it to stand for an hour and then filter. I'm dubious that it's the THF quality ( but it's possible ). I think something else is marginal, and you may have to modify your method systematically to find it. I'd start with sample solvent/concentration, mobile phase composition, and column temperature, as split peaks can often result from one of those.

Good luck

Bruce Hamilton

Eventually the idea brought up by JM is a good solution - clean the THF with an online C18 SPE cartridge and see what happens afterwards...

Lot of answers, observations, suggestions... I will try to answer them.

Bruce,

We premix the solvents; column at 40ºC and solvents at ambient temperature. The backpressure of the columns increased with the bad THF, not with the good THF. We did not try to increase column temperature to see the effect on the peak splitting (an idea to try!).

The old methods have the same concentration, they worked fine, I have not tried lately with bad THF (another idea).

I did not try either to inject blanks and repeat samples to see if they get worst. I suppose that tyey do, but have not tried.

We have no possibility to analyse the rubbish or dirt.

When we filter mobile phase, we usually filter the individual solvents and mix them after filtration not to change the composition because of the different volatilisation of the component solvents. However, we can also try the idea of filtering one hour after premixing the solvents.

Albany,

The guard-columns we used are about 1cm long, not the watch battery type...

Peter,

Another good idea to put guard column upstream the injector.

JM,

The bad and good THf worked on the same equipments, and on several same equipments.

ym3142,

Last time after 4-5 injections the peaks began to split.



So, a lot of suggestions to try. We are entering in holiday period, so only in september I will be able to try them. i will get back with the results.

thanks everybody!

Kati

Still would like to know what happens to column performance if you pass quite a bit of bad THF through a good (new) column as well as through a bad (split peaks) column. (That is, I am still skeptical as to the THF being the source of your problems, if it really is then you should have purified THF after those experiments).

Here is something to consider that nobody has mentioned until now: how about if the contamination comes from the vial caps?

Some types of caps may be more sensitive than others to the sample solvent. Your original problem went away when you changed the sample solvent. Why? Maybe because the THF dissolved stuff out of the caps and deposited it on the column, and MeCN did not do this. But maybe you have now inadvertently changed the vial caps again, and the new caps have a problem with MeCN...

Just a thought...

Uwe,

We have been using the same type of vial caps for months, I don't think it would be the problem. Especially because we used the same vials with good and bad THF and had no problems with the good one while had problems with the bad THF.

Thanks!

Too bad, ...... we still don´t know whether you can remove the cause of peak splitting of that column(s) by washing it extensively with what you think is bad THF.

Dear H.W. Mueller,

If I understand well, your suggestion is to pass some bad THF through a column and utilize it to prepare a new mobile phase and use this new mobile phase with a good column and see what happens. Right?

Unfortunately we have lot of people on holidays, till september we have personal problemas, but in september I will make the experience and get back with the results.

Thank you very much for being interested in this "phenomen".

Kati

Partly. The more promient suggestion is that what you think is bad THF will clean your splitting peak column to perform almost like new, takes 30 min to 1h. This guess comes from using commercial solvents since ~ 1958. Though I have seen gradient grade organics do some reactions due to impurities I still hope that manufacturers are not so incompetent (or devious?) as to deliver a HPLC solvent that leaves a black, chrom influencing gunk at the column head.

Let me see if I understood well.

I take a bad column that already splits the peak, pass bad THF (100%) during 1h 1ml/min and than try to perform an analysis with a mobile phase prepared with fresh bad THF (not that that passed through the column).

Is this your suggestion?

As for gradient grade solvents, my first hope was that the good THF was gradient grade and the bad ones only HPLC grade - some years ago Merck had gradient grade and HPLC grade solvents. But unfortunately now they only have HPLC grade THF, they do not have gradient grade THF.

Thanks!

Yes, then you can vary the theme a bit, depending... would almost willing to bet, well lets wait.

Sometimes the tubing of the instrument is not compatable. I suggest you using stainless steel tubing and column fittings. Also using freshly opened un-stableized HPLC grade THF (one of our method is specified this way).

Good job, Bruce Hamilton