PERSISTENT GHOST PEAK NEED HELP!

[ARM]

Hi there,

Im trying to put into work a HPLC method that consist on a gradient with water,acetonitrile and a constant concentration of 200 mM sodium acetate buffer. I have a peak appearing at the same retention time that one of my compounds of interest,even when I run a blank injection. I have HPLC grade reagents,I passivated the system,changed the injection needle and rotor,changed the MCGV,put a new column on,changed solvents,bottles and inline filter with no effect. Could you help please?Thanks

[tom jupille]

Probably garbage in the "A" solvent reservoir. There is a diagnostic procedure in the "Troubleshooting Wizard" on the LC Resources web site that can help confirm it:
http://www.lcresources.com/resources/TSWiz/hs400.htm

If you confirm that it is garbage, you have to find the source. In addition to the things you have already changed, potential sources include:
- the sodium acetate
- the pH electrode (if you are adjusting pH and put the electrode in the mobile phase)
- residual detergent on glassware

[Rob Burgess]

What happens when you set your injector to inject 0 (zero) µL? Do you still see a "ghost peak"?

Also do you still see a ghost peak when you run your gradient profile without the buffer?

[ARM]

Thanks for your replies...

When I dont inject anything the peak is still there. I even did a run bypassing the injector and connecting the pump capillary directly to the column compartment and the peak is still there,so the injector is not the source.

The sample is not the source neither as I did injections in brand new columns without any previous injections and the peak is there

When I run the gradient profile substituting the buffer for water I do not see the ghost peak. Does this mean that the buffer is definitively the cause of the peak? or is it possible that the peak doesnt elute the system without the buffer?

When I increased the equilibration time before the blank injection (with 75% water,25% buffer) the area of the peak increased considerably. So it seems it is either the water or the buffer,isnt it?but i do not understand because Im using HPLC grade water from Fisher, freshly filtered, I never rinsed with soap the bottles just with water,I autoclaved bottles from time to time. I prepared the buffer exactly in the same way that some colleagues in other lab with the same reagents,acetic acid,water and methanol HPLC grade from Fisher and sodicum acetate enzyme grade high purity from Fisher too,and they dont have this peak coming..i cannot find the source of the contamination....if you have any other suggestion you are more than welcome,its really getting on my nerves!!Thanks a lot!!

[AdrianF]

What wavelength are you using? I would guess its fairly low.

Try a higher grade of Sodium acetate or clean up the buffer solution by pumping it down a column which has previously washed with a high percentage of organic/water say 90% acetonitrile. This should remove the absorbing impurity. Alternatively you could use an Empore filter to clean up your buffer.

It only takes a very small amount of UV absorbing impurity to cause a problem

[ARM]

Sorry I didnt say it, Im using fluorescence detection (Ex 276,Em 316)
although im also monitoring 280 nm and the peak is also in the UV chromatogram,but much smaller than in fluorescence.

I just did another run equilibrating the system only with water for 45 minutes and the peak became huge,another reason for suspect of the water?im confused...

[project44]

I had the same problem using 100mM of phosphate bufer. Try to reduse the buffer concentration.

[tom jupille]

If longer equilibration = larger garbage peak, the garbage is coming from your A reservoir.

You have two choices: clean it from the solvent (per AdrianF's suggestion) or find out how it got there and eliminate the source.

Try changing things one at a time:
- water source (if you were using an in-house water system, try buying a bottle of HPLC grade water; if you were using bottled water, try from an in-house source)
- pH electrode (if you were putting the electrode in the mobile phase to adjust the pH, stop). To measure pH, take an aliquot of your mobile phase in a small beaker, measure the pH, and then throw away the aliquot.
- sodium acetate (try a different lot and/or a different supplier)
- residual detergent (try exhaustively rinsing all your glassware with organic solvent before use)
- latex (are you wearing gloves? are you touching the inside of a beaker by any chance?)

By the way, 200 mM is a very high concentration for a buffer in HPLC (something like 25 mM would be more common); are you sure you need it that high?

[AdrianF]

I have seen this kind of problem before with fluorescence detection. I developed a method for PAHs in water which was then transferred to another lab where they got massive interference because of the water they were using.

With fluorescence it is essential to use the highest possible purity water - use a polisher eg from Elga or Millipore which has a UV lamp to reduce TOC(Total Organic Carbon). The water should also be freshly prepared if at all possible.

The buffer salt should be of the highest grade available eg Aristar and used at the lowest possible concentration. Use a fresh bottle for this project and make sure no-one else can get their spatulas in there!

If you still have a problem, this web address is a reference from LC-GC for cleaning up mobile phases using Empore Disks.

http://solutions.3m.com/3MContentRetrie ... on=current

[ARM]

Thanks everybody for your replies, I just joined this forum but I wish I have found it before! I have followed all your advice...I have tried with different type of water,different sodium acetate of different brands,with no sucess. I filtered the water and the sodium acetate buffer through an old HPLC column after being flushed with 90% acetonitrile, then I cleaned the system and put another column,and the peak is still there! I dont use latex gloves,dont use the pH electrode or soap,I only use water to rinse my glassware. I tried with different glassware too...still there. The only thing I have left to try is the Empore filters,but they are quite expensive to use them all the time..what I dont understand is why my colleagues dont have this problem using the same reagents...the only difference is that they use 0.1 micron nylon filters to filter the water instead of 0.2 micron that I use, do you think that could make a difference?thanks again

[AdrianF]

The fact that your colleages don't have the same problem with the same reagents points to another source.

To eliminate problems I sugggest making up your mobile phase as normal and then asking them to use it.

If they get the ghost peak it looks as if it is coming from your filtration step - personally I never filter mobile phase. Filtration steps can sometimes introduce impurities. Try not filtering or use their filtration equipment.

If they don't get a ghost peak the problem must be in your HPLC system eg the injector. Try another system.

[Kostas Petritis]

Filtering your mobile phase A through a C18 column should have taken care of the problem if you had impurities in water or sodium acetate so I bet that your collegues will report no problems with your mobile phases (and the other way round if they prepare your mobile phase, you will always have the same problem).

The only other option is that your HPLC system introduces somewhere that impurity(ies), maybe in the pistons?

[Bruce Hamilton]

You may be able to sell a haunted HPLC for enough money to buy a new one, but if not...

Ghost busting is unfortunately seldom first time lucky, it takes a systematic approach, but you have the advantage that your colleagues' system is OK. Ideally it woul be nice to swap mobile phases with them, but they may not be willing to try yours.....

If the ghost peak has the same fluorescence/absorption as your target compound, then you have contamination, and that could be from laboratory glassware, HPLC, or dispensing devices. If the optical behavior is different, consider all possible sources, especially if the HPLCs have different roles and histories.

Assuming that the problem is not the quality of your solvents - the most likely option, and well discussed by others, I would check the solubility behaviour of the sodium acetate during the gradient and conditioning. You want to ensure that the mobile phase doesn't precipitate at any stage or temperature.

I would start with the HPLC solvent handling system.

I would thoroughly clean all sinters and filters with warm water and/or solvent, and wash out each solvent line ( from reservoir to injector ) with both 100% water and 100% methanol - that is intended to thoroughly rinse all material from degasser membranes as well as any polymeric lines/fittings.

If your system has a seal wash, ensure the seal wash solution is fresh and clean.

I'd then make fresh mobile phase and perform 3-4 injections of blank, to see what trends in ghost peak area appears. The information then helps you choose what next to try.

Please keep having fun,

Bruce Hamilton

[HW Mueller]

Maybe degassing is insufficient? Is your peak really a fluorescence?

[wanda50]

I recently had that problem. Troubleshooting I injected a strong solvent repeatedly followed by passiving the system. Still had a problem, so I had to rebuild the injection system - needle, line, seat, seals. I still had carryover, so the pump had to have pm done. I haven't started it up yet, but in my experience a regular pm will do the trick.