GC/MS or just GC for biodiesel feedstock analysis

[AZBiodiesel]

Hey guys, Im pretty new to the world of chromatography so here goes...I would like to compile as much data as possible from a wide variety of triacylglycerols and fatty acids that we could possibly use to make biodiesel (FAME). Not only to ID and characterize the oils themselves but other possible contaminants found in vegetable oils and animal fats, finished fuel product, and possibly even petroleum products in the future.

My question is how practical is a MS compared to just a GC with FID? Should I be focusing more on column selection rather than the hardware? My one concern is that on this GC system we will be running the ASTM D6584 free and total glycerol method most of the time which uses a 5 % phenylpolydimethylsiloxane coated column. Would it be worth my while to use a MS in conjunction with this column?

Also, If we were to buy a GC/MS system, what are some of the more common tests environmental labs do that use an MS? I would like to think of this as an investment and try to outsource some kind of testing in the future.

Currently I use an HP 5890 with CoC, FID and manual gas regulators. We would keep the old GC running. I was thinking a 6890 FID with the 5973.

Any thoughts or comments are greatly appreciated.

[Don_Hilton]

A GC/MS system trades off some sensitivity for the ability to identify compunds by mass spectral characteristics. If I read your description correctly, you are talking about setting up a system with MS and an FID. Using both MS and FID at the same time in a GC can be a bit tricky. I've had more than one GC/MS with a FID installed - just in case we wanted to use FID. I believe that I did use the FID on one such instrument once - to trouble shoot. On the other instruments, the instruments were finally surplused - with FID's never used.

You do not give dimensions of the column you will be using - there are some columns that work quite well with FID that do not have sufficient pressure drop across the column to use with a MS detector.

For environmental analyses that use the MS, there are the EPA volatile and semivolatile methods. However, if you plan to switch between triglyceride and environmental samples, you may find that the issues associated with switching between methods and passing QC tests to be challanging.

[AZBiodiesel]

Yes it would be equipped with an FID. Normal operation would be using the FID. When the MS is hooked up is there a way to switch between FID and MS relatively quickly?

For example the GC would normally be running quantitation of free and total glycerol in FAME, but if we got a new type of oil in, i would make a small test batch into methyl esters copying our plant process and then try to ID the different FAMEs using the mass spec.

I guess the important part is having the library to ID the major constituents, but I could do that with just the FID and different columns and standards? What about the unknowns?

I want to be able to determine degree of saturation of the oils as well, can this be done with an MS? I know I would still have to swap out a correct column to separate based on this criteria right? So is there justification on using an MS which ID's by m/z rather than FID which is carbon number?

The column i usually use is an agilent db5ht 15m 0.32 ID connected to a 5m guard column 0.53 ID.

[Don_Hilton]

It has been a number of years since I ran a 0.32 ID column into a MS - and I believe it was longer than 15 m - and it was marginal for operation of the MS.

To switch a column between the FID and MS would require physically moving the column. The length between the end of the column and the ferrule differs.

And what do you plan on the MS doing when not in use? You do not want to just shut it down and let it sit. MS systems do not seem to "like" sitting being turned off. The do much better when left running.

For what you want to do, I would suggest the best solution is two separate GC systems. If you can not do that, the second option is two inlets on the same GC, one with a column to the FID and the other (with something like a 20m to 30m 0.2 mm or 0.25 mm id column) to the MS. This will allow for proper flow to each detector and will allow you to keep the MS under vaccum at all times. Check to see if you can install a COC inlet in both positions.

To run the GC with the same inlet for both FID and MS you will need to change columns - and associated parts in the COC inlet.

And, as you say that you are new to chromatography, I will point out that the MS detector can be very helpful in identifying compounds - but there are a few things to keep in mind:

- If you use a mass spectral library to identify compounds, the compound has no chance of being correctly identified if it is not in the library. And most of us who look at complex mixtures of natural origin keep running into things that are not in the librarry.

- Spectra of some compunds look similar to each other. The mass spectrum of compunds depends on the fragmentation chemistry of the compunds. In some cases, the molecuar ion is formed and it does not fragment - at least not enough to give usable information, as is common in some PAHs. In other cases, the molecuar ion is so unstable that you never see that ion, and you see a stable fragment - which in some series of molecues, like the gamma lactones, is common to a number of compunds.

- When you have coeluting peaks, you have mixed spectra. These can be sorted out by programs like AMDIS, but such spectra must be used with caution because noise can affect the deconvolution process.

Expect some time to learn how to interpret resuts, even when using mass spectral libraries.

[AICMM]

AZBiodiesel,

I worked with one customer who had initially had an MS installed for biodiesel and it was a disaster. Column parameters for typical biodiesel are not really compatible with MS. If I were forced to do it, I would probably look at doing a split injection at the injector to 0.32 and 0.25 columns. Problem is, you are going to have a hard time doing on-column in this scenario although you could do a short pre-column and splitter. Then the challenge is how the splitter holds up to the temperatures you are getting up to AND how inert the splitter will stay. Perhaps the Agilent micro-plate could survive these conditions and stay leak free?

Best regards,

AICMM

[Karen01]

Be aware of something if you DO have both an FID and an MS on an Agilent GC. You can not use the GC/MS ChemStation software for ONLY the FID. You have to have the MS running else it won't work.

Also You can't install the LC/GC version of ChemStation on a computer that has the GC/MS software on it. There are conflicts.

I had a situation where I needed to use to use a 0.32 column with a proprietary phase not meant for MS and at a flow rate that was not compatible with MS. So I had an FID installed ... and ran into those issues.

While I don't know for sure, if you install two inlets and 2 columns, you MIGHT be able to inject into and use the FID and just ignore the MS data (as nothing was injected anyway).

Two systems would be much simpler.

- Karen

[Amirav]

The answer to GC-FID or GC-MS depends on the type of information you are looking for. If you wish to identify the di and triglycerides you may wish to learn about GC-MS with Supersonic Molecular Beams and read J. Chromatogr. A, 1195 (2008) 127–135. or visit http://www.avivanalytical.com
With standard GC-MS you will not be able to elute these compounds and other heavy compounds and if you will, the absense of molecular ions will impede identification. Note that the GC-FID elution peak(s) of triglycerides is of all the saturated and various unsaturated with the same number of carbon atoms.
What type of information are you after?

[Consumer Products Guy]

In the soap business, we assay for free glycerol, mono-, di-, and triglycerides, and usually use GC-FID with a metal capillary designed for high-temperature assays up to 400C (thin film, 10m, not fused silica, etc.).

The trimethylsilyl-derivatized mono- and di-glycerides fall within their own retention regions, and before the triglycerides. Good reference standards can be obtained from Nu Chek Prep.

We have 4 GCMS, but we don't feel we have need for them for this project. Fatty acids/soaps are assayed either as the trimethylsilyl derivatives, or by fatty acid methyl esters on cyanopropylsilicone columns such as SP-2330, SP-2340. Again, good reference standards of fatty acids and/or FA methyl esters can be obtained from Nu Chek Prep.

[AZBiodiesel]

wow this is great, thank all of you for the info. Well it looks like it will be better to just have have separate systems, and for the time being just focusing on another GC-FID setup.

I think the other part of the problem is that I'm not really sure what kind of information I am looking for. I want to check the purity of our incoming feedstock oils. We use everything from camelina oil, soy, canola, fish oil, waste veggie oil, used veg oil hydraulic fluid, all kinds of stuff to make fuel out of. If for example we have a feedstock that causes us to come out of spec on the finished fuel end, then i would like to be able to look for differences in that feedstock compared to the other oils. For example I would like to see what kind of identifiable cellular material carries over into the feedstock oil, and subsequently into our reactors, and then into the finished fuel and even into the glycerin waste stream. Things like carotenoids, antioxidants, any aromatics that could be produced in the process, by products of rancid oil etc. things not removed by refining, bleaching, deodorizing...

The second part is characterizing the fatty acids in their methyl ester form. Saturation, chain length, anything else that is detectable via chromatography (need to do more homework here haha)

Also a sidenote: Consumer products guy, I have been working with a client who has his own soap company, currently they have been outsourcing their quality control tests to a one person lab in new york. We have been discussing me taking over and developing a testing schedule for their products in our lab which is in the same city. Any information you have on testing of soap products would be greatly appreciated. I can't imagine it would be much different than some of the tests we do in our biodiesel lab.

Thanks, and happy new years to all...

[Consumer Products Guy]

Also a sidenote: Consumer products guy, I have been working with a client who has his own soap company, currently they have been outsourcing their quality control tests to a one person lab in new york. We have been discussing me taking over and developing a testing schedule for their products in our lab which is in the same city. Any information you have on testing of soap products would be greatly appreciated. I can't imagine it would be much different than some of the tests we do in our biodiesel lab...

You are correct: you probably can do all their needed tests with GC and wet chemistry methods. For soaps, we only used HPLC for germicide/antibacterial agent analysis or for mold inhibitors in the wrappers/cartons. Most of our test procedures are not proprietary or that involved. Post the type of tests this soap company will want to have run. Likely these will include %free fatty acids, % volatile, % NaCl, % glycerol (especially if this company uses kettle process as opposed to simple neutralization of fatty acids), possibly fatty acid distribution of incoming fats, oils, and/or fatty acids. We rarely assay for % sodium soap in our own products. If you would like specifics, please post your E-mail address.

[AZBiodiesel]

Thanks CPG i would love more info!

[Consumer Products Guy]

I E-mailed you.

[petegcms]

If you are interested in identifying the double bond position in lipids you could use the methods established by Brenna et al. We use a Varian Saturn 2000 GCMSn with CIMS capability. You can pick one up on the used market for ~20k. I have been doing this work for 12 years so if you have any question e-mail me.

Pete