Sulphur coumpounds using GC with a PID

[Heikki]

I'm supposed to setup a method for the determination of sulphur compounds from different rotten foods. The sulphur compounds are produced by microbial fermentation and we have no idea of their concentration.

I'm using an old HP5890 series II GC, with a HNU PI 52-02A Photo Ionizing detector (PID) and a FID. My sampler is a Tekmar 7000 Automatic headspace sampler. Although the machines are old, and little or no support at all is available, they work and have performed for me in multiple other analyses before. Except the PID, I just installed it and I haven't got much experience of using it.

I've been trying to setup the method using a Restek 59% Carbowax 80/120 Carboblack-B 2mx2mm packed column, with the owen at 60 degrees Celsius.
The carrier gas is Helium and the flowrate is around 20ml/min.

The PID is set at the lowest attenuation and temperature set to 110 degrees C. The inlet and FID are both set to 110 degrees C.

The Headspace parameters are following:
Plateau 60
Loop 60
Transferline 80
Sample equil time 5min.
Loop size 250ul
Sample vial size 22ml

Now I've encountered few problems. First of all, I have to admit, that I don't exactly know what I'm doing, this methos is more like a trial and error setup, than a proper method creation. I'm more of a biochemist, with the opportunity and, should I say, responsibility of doing our chemical analyses as well.

My first problem is, that I get a mysterious peak from the PID at around 0,5min whatever I inject (even room air). I've cleaned the inlet and the ionization chamber. I've tried to clean the headspace as well, but I don't know if I did it properly. I get the peak even when skipping the HS and doing a manual injection.

My second problem is, that I'm doubting my method very much. I haven't been able to find much information on using a PID in sulphur determination.

The question that I'd like to present, is that is it possible for me to do sulphur analyses using this equipment? Should I change the column? Is there something critically wrong in my method?

[AICMM]

Heikki,

I would say that sulfur compounds by GC-PID is very doable. You probably won't be able to do COS because the IP is too high and H2S will not respond great but the good news is that it will probably be at pretty high concentration from what you describe. You should be able to do most of the others at pretty low levels because they will have pretty low IP's. Your biggest problem might be separations so I would appreciate knowing why you selected the column you did.

Also, raise the temp of you detectors at least 15-30 C to ensure that no water condensation anywhere. Regarding the stray peak, do you see it on the FID side and if not, do you have any kind of valve event happening around that time and is the peak positive, negative or both?

Finally, keep in mind that sulfur compounds want a clean, inert system and that may take you a while to get to.

Best regards,

[Heikki]

Thanks for the response.

I've realized as much about column and hence changed the column to a sulphur specific Restek Rt-sulphur 2mx2mm.

I also realized what the mysterious peak was, as it was oxygen. My samples were just too low on concentration, as they were injected into a 22ml bottle full of air. Tha sample size was from 0,5 to 2ml. Hence the sample diluted very much upon injecting into the bottle.

Heikki

[jmartin]

I use this exact setup for sulfur analysis in hydrocarbons. It works well and is much more cost-effective than SCD, but you need to know which sulfur species you're looking for. I disagree with the above post concerning COS and H2S. I have no trouble seeing those at ppm-level with an 11.7eV lamp, however this limits the max temp of the detector to 120 C. I hope the development goes (or went) well.

JM

[parvers]

I use this exact setup for sulfur analysis in hydrocarbons. It works well and is much more cost-effective than SCD, but you need to know which sulfur species you're looking for. I disagree with the above post concerning COS and H2S. I have no trouble seeing those at ppm-level with an 11.7eV lamp, however this limits the max temp of the detector to 120 C. I hope the development goes (or went) well.JM

Hey jmartin,

With the PID being selective for unsaturated compounds, do you manage to see any of the saturated Sulfur compounds, or is the PID limited to unsaturated ones?

[jmartin]

All I can tell you is that we look for 4 sulfur components; H2S, COS, methylmercaptan, ethylmercaptan. We can see all four with an 11.7eV lamp, but have some issue with hydrocarbon quenching, depending on the sample matrix. What unsats are you referring to?

JM

[AICMM]

With a PID with an 11.7 eV lamp you are going to see many of the saturated hydrocarbons in addition to the unsaturated hydrocarbons. Starting at propane or so the 11.7 lamp will ionize hydrocarbons. Does not necessarily preclude the use of the lamp since many of the hydrocarbons that will respond will also elute later than the light sulfur compounds so it really depends on your target compound list and your separations. Also, be aware that the 11.7 eV lamp really likes to run much cooler due to the window used.

Best regards.