Split Peaks... just sometimes?

[GoldLeader6]

Hi Folks,
I've searched but can't come up with a solid answer. Running on a Shimadzu GC-2010 with autosampler, analyte is IPA, ISTD is THF, diluent is DMSO. Out of 50 injections, we see the IPA/THF peaks split in about half a dozen seemingly random chroms. Any idea why just a few would split and not the whole sequence?
Thanks,
GoldLeader6

[Peter Apps]

Welcome to the forum.

Nobody can have any sensible ideas about why your peaks are splitting unless your provide information about your analytical conditions - temperatures, flow rates, type and dimensions of column, sample introduction (with DMSO as a diluent you are probably doing something out of the ordinary), detector etc etc.

The more you tell us the more we can help.

Peter

[GoldLeader6]

Hi Peter,
I was cautious to post details due to anonymity. But:
Column = HP-5, 30m x 0.53mm, 1.5 um film
Control = Head Pressure @ 40 kPA
Split = 1:5
Inj Vol = 1 uL
INJ = 270C
DET=FID@280C
Oven=130C, isothermal
Elution order = IPA, THF, DMSO

I've attached an example chromatogram below as well.

We're using DMSO as diluent due to the nature of the sample matrix and this is a method that was validated and transferred to us from another lab so we have little latitude to change conditions or solution preps. Unfortunately I'm unable to contact the transferring lab for answers due to reasons.

My confusion is that the splitting is intermittent and not across all injections and is seen randomly in both standards and samples.

Hopefully this will give you more information to go on.
Thanks,
GL6

[James_Ball]

Punching the numbers into a vapor volume calculator I am seeing that 1ul DMSO is expanding to almost 500ul. If you are using a very narrow injection port liner it could possibly be over filling the liner with vapor, if at the very limit it could lead to the random nature of the split peaks. If your liner has at least 700ul volume then you should be ok on that.

Another thing that can cause peaks to split is if the column is inserted too far into the liner or cut very unevenly on the inlet end.

[Peter Apps]

Intermittent peak splitting is often due to solvent condensing on the column. The high BP of DMSO makes this plausible, the 5:1 split and megabore column make it less likely.

An increase in split ratio or column temperature would solve it - but it is impossible to fix a method that cannot be changed.

Peter

[lpnian]

Hi, if the performance is good before 50 injection, i thank the reason may be column contamination or the someplace is in leaking, you can try to cut column , change septum in inlet. hope it useful.

[dblux_]

Intermittent peak splitting is often due to solvent condensing on the column. The high BP of DMSO makes this plausible, the 5:1 split and megabore column make it less likely.

An increase in split ratio or column temperature would solve it - but it is impossible to fix a method that cannot be changed.

Peter

What Peter said is probably the reason.
But it may be the autosampler, which "simulates" double injections.
To eliminate it I would try a new syringe.

[Peter Apps]

I notice your retention times are quite short and with the inlet pressure you have your carrier gas flow rate might be high. Do you use any packing in the inlet liner ? If not you might be spraying liquid droplets into the column.

What are you allowed to change ?

Peter

[GoldLeader6]

Thanks for the advice everyone. I did also have some concerns about DSMO coming out with an isothermal program at 130C and you've confirmed that for me.

The method does not specify which liner to use, so I'll check what our chemist is using and make sure it has a large enough volume and at least a glass wool plug.

Also, I found that there have been similar issues in previous runs with this method (from months ago before I was involved) so maybe it's time to send it back to development.

Again, many thanks for your quick responses. Much appreciated.

[dblux_]

Another thought, is your washing solvent the same as sample solvent ?

[Peter Apps]

Another thought, is your washing solvent the same as sample solvent ?

Good question !

Peter