Peaks in Blank/Diluent

[AngelicaV]

Greetings fellow chromatographers!

New to forum and relatively new to GC, I thought I'd ask about issues with peaks in my blank. Let me start from the beginning.

I was working on transfering a method on DMF detection. The solvent the method calls for is Acetonitrile, the column: an Agilent RES-SOLV column. Anyway, my first trial injection of Acetonitrile was clean. Next one (my standard) showed a peak (yay!). After performing a sample set of my standards, my %rsd didn't pass and my blanks showed a "DMF" peak. After conditioning the column for a few hours and cleaning out the syringe, there's still a peak in my blank. I obtained another blank sample to inject; still a peak. I tried a different solvent manufacturer; still a peak. Okay, maybe there's just an impurity in Acetonitrile that elutes at the same retention time as DMF.

Unfortunately I'm pressed for time, so I was instructed to move on to another method that uses a different column and solvent. This one uses DMSO as the solvent and a DB-WaxPlus column. I've ran this method before to test a different product and it worked beautifully. Now everything's set up, cleaned out my syringe, my solutions are ready to go and... guess what... there's a peak in the blank!! Again?! So I changed my liner (which was clean but I changed it anyway), and ran a fresh solvent blank. Peak again.

Currently I'm conditioning the column; hopefully I won't have a peak again. My next step (I think) is to change the septum (which was changed about two weeks ago). But there must be something I'm clearly missing. Has anyone else ran into this problem before? If so, what steps did you take to clear it up?

[JTM]

After conditioning the column, these are some basic things to try:

Replace:

Septum
Liner
Gold Seal

Trim the column

Check for leaks

Make sure your detector is clean (depending on what detector you have)

Try a new stock solution of solvent (different lot)

Clean all your glassware again

Try a new stock of sample vials

Clean your injection syringe thoroughly.

[aldehyde]

You should also try to half split the system. It is unlikely the detector is contaminated, but you could plug it and make a run. See if you get the peak even with no column flow into the detector.

Does the system have two inlets? Try hooking the column up to the other inlet. Are you using an auto sampler? What happens if you make a manual injection? If you are using wash vials they could be contaminated, they should be completely emptied when the solvent reaches the fill line rather than just filled up. Otherwise, sample on the needle may eventually contaminate it.

When does the peak elute? What is your inlet temp, oven temp, detector temp, flow rate?

[AICMM]

AngelicaV,

The ghost peak shows up at the RT of DMF every time? If that is the case, it cannot be detector since the only way to get the same time is to chromatograph. I am assuming you are using a temperature program? If that is the case, then I would suspect the split vent or septum purge lines. If you flashed into the split line then it may be contaminated. If it is contaminated and you are using a temperature program then the column may be acting as a trap and the peak is collecting there until the next full run.

You might try rinsing the split line or baking it out.

Best regards.

[AngelicaV]

Thank you all for you responses! Yesterday, I changed the septum, liner and gold seal and trimmed about a cm off both ends of the column. I injected my blank twice, then my standard and then a more concentrated DMF solution (to make sure I identified the peak correctly). The peaks eluted later this time (from 3.8 min to 6.2 min) but the same issue remained; the ghost peak consistently showed up at the same time as my DMF peak. So I conditioned the column overnight.

This morning, I had the same problem, even after getting a fresh vial of blank. I was instructed to trim more of my column on the inlet side (about another 5 cm) and replace all the wash and waste vials. Cutting the column didn't seem to help (although I thought it was very odd that the peaks shifted back to 3.8 min). I know the wash vials were brand new from the box, but I didn't change the waste vials so I replaced those. I also opened up a different box of vials and reprepped my samples. Sure enough, the ghost peak is gone and my working standard is showing my DMF peak, which is great! Thanks again for everyone's responses!

[JTM]

Perhaps there's a problem with that lot? It's rare, but it can happen.

The other thing you can try is running the sample on another instrument if you have it available. That will eliminate any possibility of it being a problem with the instrument.

[aldehyde]

Thats great that it is fixed. A small critique: 1 cm is too little to cut to make a difference, the purpose of trimming the column is: a) anything that doesn't pass through the column gets stuck in the first few inches. Trimming it can help clean things up. b) getting a better cut can remove active sites or remove obstruction.

[TheBupKing]

Agreed, when someone recommends trimming a column, it's usually implied to be measured in meters, not centimeters. Of course you will have retention time shifts when you change the length of the column.

Glad the problem was fixed, at first it sounded like you might be overloading your injection port and getting blowback into the injector.

[JTM]

Some programs "prefer it" when you tell it when you trimmed it as well.