Amperometric detector question

[cygg168]

I am performing a test using a USP HPLC method, which requires to use Amperometric Electrochemical Detector. The detector model is LC-4C made from BAS (bioanalytical systems, inc), with dual series glassy carbon electrode. The column is L29, GammaBond Alumina RP1 column.

My problem is the active peak is very broad and can not pass the system suitability requirement of 1000 minimum plate count. I never worked with electrochemical detector before and tried everything I can by reading the operation manual.

Any one has experience with this detector? Any suggestions that can reduce the peak broadness and make it sharper?

Any suggestions are greatly appreciated.

Thanks

[danko]

Is there any particular reason for you focusing on the detector? I can’t think of many situations where the detector causes band/peak broadening.

Best Regards

[aceto_81]

If you are using any post column reagent with a postcolumn mixer, you could try to lower the volume of the postcolumn mixer.

Ace

[cygg168]

Thanks all.

The reason we are using this particular detector is bacause it is a USP method. If we use a different detector, we have to develop a new method and then validate the method.

Our instrument do not use post column reagent with a postcolumn mixer.

[quote="danko"]Is there any particular reason for you focusing on the detector? I can’t think of many situations where the detector causes band/peak broadening.

Best Regards[/quote]

[bisnettrj2]

cygg168 - that's not what danko meant. His question is, why do you assume the detector is causing your bad peak shape? It's more likely to be a column issue, or perhaps you have too much wide-bore tubing (a la aceto-81's suggestion) after your column, causing these broad peaks?

[cygg168]

bisnettrj2:

I know the first thing to suspect is the column when the efficiency is bad. However, we used exactly the same column (GammaBond Alumina from ES Industries), all HPLC conditions are the same as stated in the USP method. We have tried several new columns from this vendor, and switched to another vendor when we could not get good result with ES Industries columns. We had tried about 8 new columns and all gave broad peak. I called USP monograph contact info, and was told that there was no issue reported regarding to this test method.

I have worked with HPLC for a long time but never had any experience with this Amperometric detector before. The detector was newly purchased and connected to our HPLC system (by a different person). That is why I am asking for help.

Thanks

[Malarac]

I agree with what everyone has said. I have used an amperometric detector before and generally the detector is not at fault regarding peak broadening. However, having said that, incorrect setup can result in altered responses.

For example, if someone else setup the detector did they use the correct tubing (i.e. correct i.d.) from the column to the detector? How long is the tubing?

Alternatively, it may be that the incorrect settings are being used for the ECD. What is the time filter constant set at (i.e . 1, 2 or 5 sec)? If you are smoothing the signal too much it will result in broader, shorter peaks (worse tailing).

Other than these factors I would think something else is too blame rather than the detector...

[bisnettrj2]

Perhaps, cygg168, you could go through the manual and make sure everything has been set up as advised? Perhaps go through the detector electronic troubleshooting (pg 45 of the pdf below) to make sure the detector is functioning as it should be.

http://www.basinc.com/mans/lc4c.pdf

[danko]

It might be a good idea to start checking the fittings (especially the ones connected to the column). Any chance of dead volume or leak there? And again, tubing – diameter, length? Finally, column temperature – is it prescribed?

Best Regards

[HW Mueller]

How can a time constant increase tailing?

[cygg168]

Thanks every one.

I have tried everything I can think of. The tubing into the column and out to the detector has been changed 0.007 and 0.005 respectively, and as short as possible. One thing puzzles me is that we use the same instrument and detector (even the same electrode settings) but different mobile phase and column for another impurity test method of the same product, and there is no problem for that test at all.

The method requires to use a 50 cm guard column for a 150 cm analytical column. Without the guard column, the retention time is about 28 minute, but the plate count is only 800. With the guard column, the retention time is almost 50 minutes and can pass the plate count requirement but the peak shape is still broad. Increasing the column temperature from 22C to 25 will increase the retention time from 47 minutes to 67 minutes.

The system takes very long time to equilibrate each time when a condition changed. For example, when the column temperature changed from 22C to 25C, it took more than 6-8 hours for the retention time became consistant. Is this normal for the Amperometric electrochemical detector?

I will check the electronic troubleshooting. I have to consultant with the chemist who setup the instrument.

[tom jupille]

The detector can do nothing but "see" what comes out of the column. If you are seeing retention time changes, those are *NOT* due to the detector, but to column and/or solvent and/or temperature issues.

[ivanvins]

[quote="HW Mueller"]How can a time constant increase tailing?[/quote]

Time constant too high will considerably broaden the peaks, resulting in low plate counts. Depending on the filter used, some tailing can arise too, but this is not so common.

[HW Mueller]

How that?

[Uwe Neue]

Hans... A classical time constant causes an exponential decay of the signal. Input: rectangular pulse: output: exponential tail. Modern filters are digital filters that do not create a tail, but simply make the peaks wider.