I have worked with anionic PFAS surfactants for the last 3 years. The first thing that jumps out to me is that you are having cosolvent effects. I would switch to 10mM ammonium acetate in your aqueous phase and use 10mM ammonium acetate in methanol or just use straight acetonitrile. Mixing solvents is generally not a good idea and if you are doing so, never above 5%. Depending on the analyte, it is very likely that your surfactant's ability to form hydrophobic bonds >>> polar bonds. So water in the organic mobile phase weakens the affinity of the surfactants to the solvent. Even though miscibility charts say ACN is miscible in H2O you still want to avoid going over 5%. And I can say from personal experience of using 90/10 ACN/water that I saw miscibility issues in my research. Needle wash MUST be 100% organic and given your clean air injections this could very well be the issue. Acetone might work if you want something more polar but I would use whatever solvent your samples are in. Are you ramping your mobile phase gradients?

The other aspect you can address is by injecting blanks in between samples. When working with surfactants, I always injected twice after any kind of cal curve or QC before sample analysis. The first blank cleans it and the second blank confirms the system is clean. You probably see carryover in blank injections but not air because something about the needle is dirty. When it dips into a solvent, some analytes may be partitioning from the needle to the mobile phase into the system. Air injections can be helpful for troubleshooting, but try to avoid it as injecting air into an LC is generally not good. Do you purge your pumps every morning before use?

The last suggestion I have for reducing carryover is to add an extra 2-4 minutes to your run at 100% organic.

Oliver Roe Oliver - thank you so much for these tips. I am running isocratically at 90/10/0.15 ACN/H2O/Formic acid as this combination has given overally the best peak shape. This morning, I did some injections of the standard in 100% organic solvent instead of 50/50 water/organic, and I've noticed less carryover when using the 100% organic solvents. Odd.

If I've already purged the pumps the day before I don't generally have the habit of purging them again unless it's been a while or the solvents have been changed.

I am going to tinker with the method based on the suggestions that you've proposed here. It is my first time developing a method for surfactants, so I do appreciate advice coming from experts!

Tijana Vasiljevic I hope the suggestions help!

In general, purging the pumps before you begin in the day is useful in ensuring there are no air bubbles in the system as well as eliminating anything like your salts that might build up in your lines pre-injection if you don't have a delay column.

Please keep us updated if you need anymore assistance troubleshooting.

Oliver Roe Thanks for your tips Oliver!

I did the following:

- Changed the mobile phase to 95/5/0.15 ACN/H2O/Formic acid.

- Changed the needle wash to 100% ACN.

I did 10 injections yesterday - no carryover.

I did 10 injections this morning - no carryover.

Then I prepared my actual samples, started injecting blank (fresh) and boom! Carryover again :(

I checked the injection without a column (just union) and indeed there is a peak. It appears that it's related to injector/needle.

Tijana Vasiljevic It's wonderful to hear there is some reduction in carryover. I suspect there is still something gunked up on the needle. If you can easily detach the needle, I would soak it in ACN or MeOH in a sonication bath for at least an hour. What needle rinse settings are you using in your LC method?

Oliver Roe So it's the waters ftn-h, where your mobile phase literally continuously washes the needle from within and that's 95/5/0.15 ACN/H2O/Formic acid. You can only control the external wash (which just washes the outside of the needle) and for that I've tried 100 % methanol, 100% acetonitrile, 50/25/25 MeOH/ACN/IPA and even water! Now I'm trying acetone, but it doesn't seem to be doing a whole lot.

Tijana Vasiljevic Have you tried methylene chloride? If you were using a mixed aqueous needle rinse there may be contamination built up a bit in the lines. Do you see a decrease in the carryover or is it constant? Did you purge your lines prior to running this morning?

Oliver Roe The only time I saw a decrease in carryover was after running 1 hour (at 1 ml/min) of 50/25/25 MeOH/ACN/IPA and then 30 minutes of pure water (at 1 ml/min). This was done with a union. It appears this combo cleans it but I need to figure out a way to incorporate it in my method.

Thanks so much for reading my issues!

Tijana Vasiljevic I would be careful running 100% aqueous through a RP column as it can destroy the stationary phase. For incorporating a flush step, there should be a way in your LC software to build in into a pre-run step.

May I ask why you are using a mix of 3 solvents?

USE Needle wash is METHANOL : H20 : FORMIC ACID (50:50:0.2),

Strong wash is ACN:H20:FORMIC ACID (80:20:0.2)

Week wash is ACN : H2O (50:50)

INTIATION THIS TRAIL

Oliver Roe Hi Oliver! That mix of three solvents is a cleaning solution we used to use at a lab I previously worked at - the mixture did an excellent job of cleaning ppb-level analytes from microextractive devices like SPME. We'd do 1 hour sonication of the SPME devices in the cleaning solution and voila - clean as a whistle!