LC-MS/MS split peak early eluting analytes

[admartin6614]

Hello,

I'm currently running PFAS analysis with a Thermo Vanquish HPLC - TSQ Altis mass spec. I've been running this method for many years now and in the past few months the first eluting peak (PFBA) has started splitting significantly. It looks like there's too much organic coming from somewhere. I'm also getting irregularities in analyte response causing my instrument QC to fail very close to running an analyte calibration. It may be just coincidental, but the peak splitting started occurring around the same time we experienced a 400 psi drop in back pressure.

We are running a gradient that starts 95:5 Water: Methanol, both organic and aqueous mobile phase contain 2 mM ammonium acetate.

Troubleshooting/Maintenace done to date:

1. Both pump heads were replaced - this resolved the back pressure issue.
2. New guard cartridge/column was installed.
3. New tubing was installed on the instrument.
4. All mobile phases, seal washes, needle washes were replaced.
5. The metering head on the sample manager was replaced.
6. The needle on the sample manager was replaced with the needle seat.
7. The seal on the divert valve was replaced.
8. I've verified that the final extracts contain the correct ratio of water: methanol.
9. I've tried reducing the injection volume and this does correct the chromatography which really makes me think too much organic is coming from somewhere.

I have noticed that some of our samples that have more matrix interference seem to resolve the splitting as well.

I've had a service engineer onsite, but nothing appears diagnostically wrong with the instrument.

It must be something subtle that I'm overlooking, hoping maybe someone has some insights.

Thanks
Allen

[Hollow]

8. I've verified that the final extracts contain the correct ratio of water: methanol.
9. I've tried reducing the injection volume and this does correct the chromatography which really makes me think too much organic is coming from somewhere.

I have noticed that some of our samples that have more matrix interference seem to resolve the splitting as well.

First guess would have also been too much of organic in your sample.
Or a clogged inlet frit of your column.
Now you confirmed that the chromatography improves with lower injection volume.
So what about pH/ionisation effects? If there's too much of your acid which your mobile phase can't buffer correctly. That could introduce a similar problem as too much organic.

Have you tried to adjust the pH of your sample before injection? Maybe add of the ammonium acetate buffer (maybe higher in concentration)
Verified that the buffer pH and concentration is correct?
What if you increase the buffer concentration in your mobile phase from 2 to 5 or 10 mM?

[admartin6614]

Thank you for your reply.

I have tried increasing the concentration of ammonium acetate to 5 and 10 mM. It doesn't help the peak splitting, but it makes the chromatography of some of my other peaks worse.

I've spent some time going through older data and subtle peak splitting of the early eluters started months ago, 2 months after it started the degasser went out. I had that repair performed and the chromatography after looked normal and symmetrical but within a few weeks it was starting to split again.

The problem is there's nothing diagnostically wrong with the instrument. As such I'm having trouble getting support from the service engineers.

[Multidimensional]

Did you install a brand new HPLC column ??? One of the most common reasons for the observations made would also be due to a fouled column. Columns are consumable items, requiring regular washing/cleaning, but they eventually fail and must be replaced. Column lifetime is a complex subject as it depends on the mobile phased used, training of the operator, sample types, sample load etc. Some methods may see no more than 200 injections before a column must be disposed of. Other examples could last 2000 injections.

[Multidimensional]

"HPLC Peak Splitting. Common Reasons For It"; https://hplctips.blogspot.com/2016/09/h ... s-for.html