Chromatography Forum

Hi all GC folks!
I'm trying to develope a method for quantification of Octamethylcyclotetrasiloxane by GCMS. I use Octamethylcyclotetrasiloxane in IPA as standard (calibration curve, 0,002 µg/ml-0,2 µg/ml). My biggest problem is that Octamethylcyclotetrasiloxane seem to accumulate somewhere in the instrument and after a few runs I can't get a blank blank! Does anyone know what to do? I have tried to wash the syringe, the liner, cut the column, clean the ion source, column burn out, nothing works. Also the repetability is very poor, the peaks change in shape and intensity, they not only grow due to the accumulation, now and then they get really small too (from the same concentration).
Help please!

Hi

please post information about the method such as instrument parameters (type of liner, injection volume, injetion/oven temps, split/splitless? column etc).

this give people a better chance to pinpoint your problem.

You most likely have quite high and eratic background levels from vial and inlet septa and the stationarry phase on the column. Eliminating all of these is likely to be quite a challenge.

Peter

You most likely have quite high and eratic background levels from vial and inlet septa and the stationary phase on the column. Eliminating all of these is likely to be quite a challenge.

Peter

Agree with Peter, if one looks low enough on the GCMS, plenty of cyclomethylsiloxanes will be present. Your path may be to make your centrations of sample and standard high enough so that such backgound peaks are negligible. Zero can be dificult or impossible to obtain.

Yes, I'm starting to realise that it is not so easily done... But thanks for the response!
Instrument parameters are:
GCMS: Shimadzu QP2010+
Column: Rxi-5ms, 30m, 0,25mm, ,26µm
Deactivated glass split liner with silanised wool plug. (Could this be a problem? OMCTS has maybe a high affinity to both glass and the wool?)
Oven: 70C-280C 20C/min, hold 2 min at 280C.
Ion source: 250C, Interface: 250C, Injector: 250C
Split: 1:10, Inj Vol: 1,0 µl, Column flow: 1,2 ml/min

My analyte is in a water/protein buffer, I percipitate the protein with IPA and filter the solution, then analyse. It is very difficult to concentrate the analyte and it is really trace amounts. But even when I use higher concentrated standards and spike IPA the reproducibility is poor. I get a feeling that OMCTS lives a life of its own... The calibration curve looks good, but if I spike a solution (only IPA and OMCTS) to about the midpoint of the calibration curve, the results are quite erratic!

You are right, it's not going to be easy, so before you start focussing on the OMCTS per se you need to ensure that the poor repeatability is not due to a problem with the GC or MS, by running some alkanes or something else tractable.

If the easy samples come out OK, then OMCTMS itself probably is the problem. You need a non-silicone phase in the column, and to eliminate the deactivated glass wool in the inlet. Use vials without silicone seals or septa. Replace the inlet septum with a septumless injection head such as a Jade Valve.

Just for interest, why do you need to know OMCTS levels ? - your lowest standard is 2 ng/ml, what effects is it likely to have at these levels (not to mention that you would have to detect 2 pg with a 1 ul injection).


Peter

Agree with Peter that you should not be using a siloxane column. Perhaps a squalane (from the old days.) Problem is that you are using a GC/MS and squalane is typically packed column and finding a non-siloxane capillary column at the temperatures you want will be challenging.

Best regards.

May have some luck with a molecular sieve column - haven't tried it though..

Bruce Hamilton

You are right, it's not going to be easy, so before you start focussing on the OMCTS per se you need to ensure that the poor repeatability is not due to a problem with the GC or MS, by running some alkanes or something else tractable.

If the easy samples come out OK, then OMCTMS itself probably is the problem. You need a non-silicone phase in the column, and to eliminate the deactivated glass wool in the inlet. Use vials without silicone seals or septa. Replace the inlet septum with a septumless injection head such as a Jade Valve.

Just for interest, why do you need to know OMCTS levels ? - your lowest standard is 2 ng/ml, what effects is it likely to have at these levels (not to mention that you would have to detect 2 pg with a 1 ul injection).


Peter

Well Peter,

I agree with you, these levels are of no danger to anyone, but I work at at contract laboratory and my customer wants this analysis. I was able to talk them into a limit method, so I don't have to quantify the samples below a certain concentration. But the problem is that I need to validate the method for quantification of higher concentrated samples. By the way, OMCTS comes from Antifoam C used in the production of the product I'm analysing. The finished product should contain little or no OMCTS and, "little" according to my customer has a big span... Customer, U gotta love them!
Thank you all for helping me! I have absolutely no possibility to use packed columns, does anyone have any other suggestion on what phase I should go for?

By the way,
My intrument works fine, it is definately the OMCTS that is the problem!

You are right, it's not going to be easy, so before you start focussing on the OMCTS per se you need to ensure that the poor repeatability is not due to a problem with the GC or MS, by running some alkanes or something else tractable.

If the easy samples come out OK, then OMCTMS itself probably is the problem. You need a non-silicone phase in the column, and to eliminate the deactivated glass wool in the inlet. Use vials without silicone seals or septa. Replace the inlet septum with a septumless injection head such as a Jade Valve.

Just for interest, why do you need to know OMCTS levels ? - your lowest standard is 2 ng/ml, what effects is it likely to have at these levels (not to mention that you would have to detect 2 pg with a 1 ul injection).


Peter

Peter,
When you say non-silicone phase column, do you mean that polymer (stationary phase) should not contain silicone and that the tubing still can be fused silica, or do I have to go for a different tubing, like stainless steel?

I mean a non-silicone stationary phase. Silica is not silicone (or silicon) so a silica column will be fine. As a straightforward first shot try a polyethylene glycol phase, most manufacturers put "wax" in the column name e.g. RtxWax, InnoWax etc.

Insist that the customer specifies a lower limit of detection and quantitation, and then validate only down to that level - what would be the point of going lower ?

Peter

Does anyone know if there are any type of septumless injection heads which are compatible with a Shimadzu 2010? Jade valve only seem to be compatible with earlier Shimadzu models, like 14 and 17. I've asked "my dealer" but they are still checking it out. Maybe someone out there knows?

/Kristin

A quick Google check reveals that there is plenty of literature available for monitoring trace levels of OMCTS. After all, it and it's metabolites were a significant part of the ruptured breast implant scare of a decade or so ago - hence the low acceptable concentrations. The column you have, even though a silicone stationary phase, appears suitable, but probably not ideal.

I would spend a little time looking at what others have done, and would be rather loathe to use a polar column, and would especially try for the thinnest film possible, even down to 0.1 um, to keep temperatures low.
I would consult column manufacturers, and find which stationary phase they woudl recommend. PLOT columns are suitable alternatives to packed columns.

Withe regard to septum, I would just use a cork borer and a crimp vial cap liner made from butyl rubber or viton, and accept it will only last a few injections.

OMCTS does react with active glass surfaces, so you need to ensure your inlet system is as inert as possible, with minimal surface area.

However, the column probably should be run isothermally ( 75 - 90C ) for the analytical sequence, every time you ramp up the temperature, you may create more unwanted background.

Also, because your molecule reacts with glass, then you should try to minimise injector temperature ( I'd start with 150C, changing sample solvent if necessary ), and perhaps reducing interface line temperature -after all it's fairly volatile ( ~175C ).

Bruce Hamilton

A quick Google check reveals that there is plenty of literature available for monitoring trace levels of OMCTS. After all, it and it's metabolites were a significant part of the ruptured breast implant scare of a decade or so ago - hence the low acceptable concentrations. The column you have, even though a silicone stationary phase, appears suitable, but probably not ideal.

I would spend a little time looking at what others have done, and would be rather loathe to use a polar column, and would especially try for the thinnest film possible, even down to 0.1 um, to keep temperatures low.
I would consult column manufacturers, and find which stationary phase they woudl recommend. PLOT columns are suitable alternatives to packed columns.

Withe regard to septum, I would just use a cork borer and a crimp vial cap liner made from butyl rubber or viton, and accept it will only last a few injections.

OMCTS does react with active glass surfaces, so you need to ensure your inlet system is as inert as possible, with minimal surface area.

However, the column probably should be run isothermally ( 75 - 90C ) for the analytical sequence, every time you ramp up the temperature, you may create more unwanted background.

Also, because your molecule reacts with glass, then you should try to minimise injector temperature ( I'd start with 150C, changing sample solvent if necessary ), and perhaps reducing interface line temperature -after all it's fairly volatile ( ~175C ).

Bruce Hamilton

I was afraid that was the case, that OMCTS reacts with glass surfaces. How do you think the sample and standard preparation is effected by that? I naturally use glass wares for that, and the sample vials are glass too.

Kristin