Sudden SELECTIVE peak TAILING (carbohydrates)!?

[lc_k]

Dear members,

I would be very grateful for any advice on our problem concerning suddenly occured peak tailing in carbohydrate analysis (NaOH solvent). Only two compounds are affected (glucose and fructose), the other peaks continue to be normal. The affected compounds are at the end part of the chromatogram. We have been using the same technique for a few months now and suddenly this has started to occur.

Since the injection of a single standard compound gave the same results (tested for glucose only), I suspected the guard column or the analytical column as the cause. However, I got excatly the same results, when I isntalled a new guard and analytical column. I replaced most of the tubing, cleaned autosampler parts, did long rinsings and finally glucose is back to normal. However, fructose, the later eluting compound, has still bad tailing. Also, the retention times at this late part of the chromatogram continue to be affected.

So what is going on?! Many thanks for any help!

[Gerhard Kratz]

Did you check also the flow rate and temperature?

[aim7811]

I kind of had the same problem on my standards. (Glucose and gluconic acid)
My problem dissapeared after I re-made my standard stock solution.
I do think that there was some problem with my standard solution.

[lc_k]

Many thanks for you replies!

I have now checked both the column temperature as well as the flow speed of the solvent but they seem fine.

The problem is there not only for the dilutes of the stock standard solution but also the samples are affected in the same manner. The problem has 100% repeatability... So far nothing I have done has had any impact on the fructose problem.

Any more ideas...?

[WillyOne]

Can you give us some more details? means column, LC equippment, Detector etc...

[lc_k]

Yes, certainly! The column is the CarboPac PA20 (3 × 150 mm, Dionex) preceeded by the corresponding guard column, earlier also a borate trap. The peaks are detected in IRMS for d13C, no other detection is used. HPLC separation system is Thermo Surveyor, which consiss of a MS pump and a HTC PAL autosampler. Between the HPLC and IRMS is the LC Isolink interface. No alterations have been made to the system, since the establishment of this analyses method...

I have now tested the injection of fructose only, as opposed to a mix of standards, and the result is the same - fructose does not look good. Glucose is still looking normal...

Could 1 mM NaOH (also 200 mM NaOH, used for column reconditioning) leach something from the HPLC unit (or the solvent bottle, borosilicate glass) that could explain this..?

Thank you again for any ideas!

[WillyOne]

It seems I can't help you nor me. I will explain it.
I have a Dionex ICS-3000 with a Pulsed amperiometric Detector and a Carbopack PA-10 column. The equippment was working quite well until we stopped it for around six months.
When we restarted it the peak symmetry was a joke. I regenerated the column with lots of NaOH 200mM without any effect.
I only have found amelioration with a regeneration with ClH 1N (65 ml, see the column manual) followed by a really long washing with NaOH 200 mM. But the effect remains a few days and it returns to the initial asymmetric peaks.
I don't have budget to buy a new column. So I saw the heaven when read your post because you changed the column and it seemed that it could be a system problem that we could find together.
But your system and mine are really differents.
The conclusion is that is a column problem, probably a void volume at the column's head and the second one you used can be a defective one.
I'm very sorry

[lc_k]

Thank you for you reply!

Have you tried to remove the possible void volume from your column? This is not something I have attempted before...are the chances low for success? What would be the best approach?

[WillyOne]

This is near impossible. Columns are packed with a slurry, that keeps the particles in suspension, at high pressure. If you open the column, atmospheric pressure, normally the packing overflows resulting in a loss of packing volume that gives more void volume as soon as you pressurize the column when reach the flow analytical conditions.
I tryed it several times with damaged columns and I have never obtained good results.
I think void volume is one of the way the columns die. It can caused by a bad proceeding to fill them or simply as a result of sudden pressure changes that are not unusual when working in HPLC.
Regards

[HW Mueller]

A silica based bed expanding when opening a column? I have seen that only when there was air on the column. Certainly, new column conditions are most likely not attained by filling voids, but I had some success in extending life time by filling voids at the time when it was still possible to get "repair" stationary phases. Sometimes, better results were obtained by running the column backwards after a "repair". Some people have reported that improved results obtained by reversing, even without filling small voids.

[Gerhard Kratz]

I agree with HW and want to add that it is possible to get packing material by the following procedure: open the column at the end and switch on the pump, packing material will come out and the first 5 cm can be in most cases used to repair analytical columns. To store the packing material I would recommend methanol. Reversed flow after repair is highly recommended. But first flush the column without connected to the detector.

[WillyOne]

we were talking about a polymer based column (Carbopack) and, as you indicate, there is not available material for refilling

[lc_k]

Thank you for your valuable comments on dealing with a column void volume. If my column is in reality in fine condition, can I do damage by trying out to simply reverse the flow direction (CarboPac column)?

I agree that it is possible that the suggested void volume is the problem but I am not convinced yet at this point. This is, because the problem remained exactly(!) the same, when I replaced the old with a brand new column.

Has anyone observed that metals could be relased from stainless steel pumps, when low concentration NaOH solvent (1mM & 200 mM) is used in room temperature? Have you tried to fix this issue with oxalic acid rinse? I would not want to risk blocking the pumps for nothing...

[WillyOne]

Well the Dionex IC 3000 is metal free: PEEK, ceramics etc. Because that I can do the 65 ml ClH 1N wash as explained in the column manual

[WillyOne]

The ICS-3000 (Dionex) I have is metal free.
This the reason I reccomended you to wash the column with 65 ml of HCl 1M (65ml) as explained in the column manual