Ghost peak in reversed phase chromatography

[toconnor]

Hello. For the past 2 months I have been having trouble with the a reversed phase HPLC method (see conditions below). It is running on Agilent system. A ghost peak has been appearing with very varying retention time of between 2 and about 30 minutes (It is not caryover from previous injection- run time has been extended to up to 1 hour). Peak shape varies from good peak shape to 'a big blob'. Peak height varies from 10mAU to about 30. Peak area of about 1200mAUs to 2000.

Column- Phenomenex Luna C18-(2) 5 μm x 250 mm x 4.6 mm
Also used Hicrome and Waters columns, old and new columns, guard columns used and not used
Column temperature 25 oC
Mobile Phase Acetonitrile: deionised water (50:50).
Different suppliers and grades of ACN and water used

Flow rate 1.0 ml/min
Runtime 10 min
Injection volume 100 l
Detection UV at 210 nm

Samples are prepared in acid at pH 1.2. In all acid blanks and samples (dissolution samples including tablet) the ghost peak appears.
Acid makeup has been investigated without using pH meter and only glassware used. Different grades and suppliers of HCL and water have been used for this prep.

Different injector vials used
Different HPLC systems used
Different analysts
Solvent inlet filters on HPLC changed

No contamination peak from water blank, no contaminant peak for a zero mico L injection on HPLC

Systems cleaned down with ACN, Water, IPA

Any help would be greatly appreciated

[danko]

Hi Toconnor,

It is not caryover from previous injection- run time has been extended to up to 1 hour

Try and run a gradient (e.g. from 50:50% to 20:80% Water:Acetonitrile in 10 min) following the 10 min separation time and see what happens.

Best Regards

[Jade.Barker]

Samples are prepared in acid at pH 1.2. In all acid blanks and samples (dissolution samples including tablet) the ghost peak appears.

I am new to HPLC, so this might be an elementary question. I have read it is best to run pH 2-8. Is it possible the pH is outside the range for the column and you are detecting the column degrading?

[toconnor]

Thanks Danko and Diagnostic Lab for the comments and ideas. One thing I failed to mention is that this is a validated method with about 2 years of perfect chromatography. If the ghost peak was present in earlier chromatography, then it was at a very small to negligable peak height and peak area. The peak has only been seen about 2 months ago and is running on the same system with same batch numbers of ACN H2O and HCL. Also same column and different brands and batches of columns. More than one system tried all Agilent. The problem has not been seen before at least not at a sizable area and height. Thanks again for the ideas and comments.

[Bruce Hamilton]

I would focus on the acid sample, as it appears that the peak correlates with an acid injection?. I suspect something is being deposited on the column, and the acid injection is partially redissolving it, and the most likely source is your samples. I assume you can't use a DAD to look at the UV spectrum?.

I would try injecting several different volumes of acid blank, and perhaps differing reduced concentrations of acid, up to 100 ul. I would also try different equilibrium times between injections, extending up to an hour, after first having ensured you had a typical area range for the peak by repeated injections of acid blank. I would also try increasing column temperature to see if peak changes size.

If the peak size increases with equilibrium time. I would suspect degasser contamination, mobile phase contamination ( filtration, reservoir sinters etc. component ( esp water, or any filtration you may been performing ).

I would try flushing the system with an acidic mobile phase ( eg gradient with 0.1% TFA, with 100ul injections of 1:1 mobile phase ) and then revert back to your analysis to see if the peaks have gone. If they have then focus on your sample preparation.

Your second comment about the same batch numbers refers to normal use,? as you also said you had tried different suppliers - water would be my main suspect, after any filtration equipment in filters in the instrument, and the equilibrium time experiment should help sort that out.

I would also very carefully examine the sample preparation, and ensure that you are not adding the peak, such as by immersing the pH electrode into the solution.

I would be a little concerned about acid-soluble material depositing in a non acid mobile phase, if the peaks are from your sample. You need to find what has changed from historical procedures. Changes such as increased mixing, warming, or ultrasonication could dissolve material that previously was insoluble.

Good luck,

Bruce Hamilton

[mohan_2008]

Bruce has given a good description of the problem solution.

I want to add a few more lines.

For, Ghost peaks appearing from nowhere.

"When performing sonication, keep an eye on the temperature of the sonication bath". Some of these drug compounds are very sensitive to heat and can break up into components that can UV absorbe".

Use only HPLC grade components, as impurities can lodge on the head of your column, and can elute at a later time.

Try running your preparations with/without sonication and observe the difference.

[toconnor]

Thanks for good ideas. Just to clarify a few things

Yes it is only in the samples that are prepared in pH 1.2 HCL in which the ghost peak appears.

To clarify the method has been ran over 2 years with various columns and various batch numbers of ACN Water and HCl. My point was that the ghost peak has been seen with solvents and columns that previously gave perfect chromatography.

The pH probe has not been used in some of the acid preps in order to isolate that as not being the contaminant

In relation to sonication mohan_2008, sonication is not used in the acid prep only stirring. Sonication is only used for the mobile phase ACN and water. All components are HPLC grade

The peak may be gone at last. After mobile premixing of mobile using supergradient ACN and degassing with helium instead of sonication, the peak has not re-appeared the last 2 days. However it may re-appear/ is being masked by solvent front and is not moving. If it does re-appear I will work on your ideas Bruce.

Thanks
Tom

[DM909]

Watch out for the suction frits in your mobile phase reservoirs. They are often a source of gunk, especially the ones placed in aqueous phases. Someone suggested to me to first dry the frits (if they're SS) in an oven, then sonicate with water - EtOH - THF - EtOH - water or similar, to clean them up.