GCMS of Vegetable Cooking Oil

[ukstudent]

Hi, im a student doing gcms of vegetable oil that had been used to fry chips but I cant find an esterification technique that works. Does anyone know suitable solvents?
Many thanks

[Don_Hilton]

Given that MSTFA or BSTFA work rather well, perhaps the question it what have you tried? And, what is happening that is indicating to you that the seterification is not working?

[Consumer Products Guy]

It sounds like you are not doing intact triglycerides (which wouldn't react with MSTFA, etc.).

We saponify in a volumetric flask using either KOH or NaOH in methanol on a steam bath. Then we acidify with either BF3/methanol or H2SO4/methanol and heat on steam bath five minutes. Then we cool, add hexane and mix, then float hexane-methyl ester layer to top by adding saturated sodium chloride and shaking (less messy if you stopper that). We inject that upper hexane-methyl ester layer into GC. Easy.

[ZeroAir]

sounds like CPGs post above has you covered. if its any help, we do analysis of plant lipid mixtures and I will provide you our procedure:

dissolve maybe 10 or 20uL of oil in maybe 1 mL or so of chloroform, then dry the solution under nitrogen with heat. Add 10uL BSTFA and 10uL pyridine or similar base to aid in deprotonation of active hydrogens. React in a capped vial for 30 minutes, then dry again under nitrogen, redissolve in chloroform or similar volatile solvent, then inject into GC-FID (if you have one available), verify your concentrations are correct so you dont mess the column on the GCMS. Then inject on GCMS.

hopefully this works,

good luck,

Luke

[ukstudent]

Many thanks for the replies guys, I really appreciate it

I'm a total rookie when it comes to Gcms. Im not too sure what Mstfa and Bstfa is?

I used a journal procedure which told me to methylate 50ml of the oil sample in 4ml of 1M methanoic acid KOH for 1 hour at 23 degrees celcius. It then told me to simply extract with hexane in a seperating funnel. This was actually for the flame gc but I assume it would not make a difference with the gcms?

As I only have 2ml samples of oil I divided the volumes to match. So instead of 50ml I used 2ml of oil and instead of 4ml of KOH I used 0.16ml.

I used this method and my supervisor told me it was too concentrated and that I could do damage to the machine. I'm really apprehensive to keep coming up with different techniques in case it could do further damage

If you could tell me what I'm doing wrong that would be a real help!

Many thanks and regards

Steve

[ZeroAir]

Hey,

MSTFA and BSTFA are silylating agents. GC columns are susceptible to degradation when exposed to active protons (-OH, -NH, etc.) Thus you can protect your column by substituting these protons with trimthylsilyl groups (TMS). you just put your sample in some silylating agent in the presence of some base for a while, cook it a bit and you should be set. I am not sure about the Methanoic Acid but maybe that will do the same thing?...

As far as concentration: we face the same problems in our lab. We run everything over a GC-FID (FID detector is not harmed by concentrated samples, and we use a poorer quality column on the GC) to check the concentration is within appropriate range for the GCMS. We usually look for a reading of 400-800uA on the FID, but it may be difference for you. 400-800 is probably a good starting place. Basically run your sample then adjust the concentration then run on GCMS.

If you have more questions just ask, I am happy to help!

Luke

[Don_Hilton]

The methanolic KOH is to obtain transesterification making fatty acid methyl esters rather than triglycerides. The addition of MSTFA or BSTFA would derivatize the remaining glycerin, monoglycerides, and diglycerides - wich typically results in sharper chromatographic peaks for these compunds. The active hydrogen in glycerin is not going to do any significant damage to a GC column - and as glycerin is practically insoluable in hexane, it will form a second phase, which you get rid of with the excess methanol and KOH.

[Peter Apps]

Hey,

MSTFA and BSTFA are silylating agents. GC columns are susceptible to degradation when exposed to active protons (-OH, -NH, etc.) Thus you can protect your column by substituting these protons with trimthylsilyl groups (TMS)There is another thread on columns (not) being damaged by polar analytes, I am pretty sure that the purpose of silylation is to improve the chromatographability of the analytes by removing poalr groups that are susceptible to adsorption, not to protect the column. you just put your sample in some silylating agent in the presence of some base for a while, cook it a bit and you should be set. I am not sure about the Methanoic Acid but maybe that will do the same thing?...

As far as concentration: we face the same problems in our lab. We run everything over a GC-FID (FID detector is not harmed by concentrated samples high concentrations of silylating reagents will cause silica deposits inthe FID, and we use a poorer quality column on the GC) to check the concentration is within appropriate range for the GCMS. We usually look for a reading of 400-800uA on the FID, but it may be difference for you. 400-800 is probably a good starting place. Basically run your sample then adjust the concentration then run on GCMS.good idea if you have the time and a spare instrument - another approach is to first run the sample split 30:1 on the GC-MS then if the peaks are too small run it splitless

If you have more questions just ask, I am happy to help!

Luke

[ZeroAir]

I will have to search for the thread on silylation/column quality/chromatographability, thanks for the heads up!

I guess I say that because FIDs are easy to clean not like mass spec, and the latter because we use on column injection , I guess I hadn't considered split/splitless injectors. Thanks for the clarification!

[Consumer Products Guy]

What I posted up above has worked great for 30 years for me.

We've also used trimethylsilyl derivatization for glycerin, intact monoglycerides, intact diglycerides for almost that long.

This is not new stuff....no need to reinvent the wheel....

[ukstudent]

I am ever so grateful for the replies guys, cant thank you enough!

Consumer Products guy, If I had 2ml of oil sample could you tell me what volumes of the other reagents I would need to use in the method you provided please?

'We saponify in a volumetric flask using either KOH or NaOH in methanol on a steam bath. Then we acidify with either BF3/methanol or H2SO4/methanol and heat on steam bath five minutes. Then we cool, add hexane and mix, then float hexane-methyl ester layer to top by adding saturated sodium chloride and shaking (less messy if you stopper that). We inject that upper hexane-methyl ester layer into GC. Easy.'

I cant express how much of a help this forum, I felt I was sinking for a while.

Many many thanks

Steve

[Consumer Products Guy]

I am ever so grateful for the replies guys, cant thank you enough!

Consumer Products guy, If I had 2ml of oil sample could you tell me what volumes of the other reagents I would need to use in the method you provided please?

'We saponify in a volumetric flask using either KOH or NaOH in methanol on a steam bath. Then we acidify with either BF3/methanol or H2SO4/methanol and heat on steam bath five minutes. Then we cool, add hexane and mix, then float hexane-methyl ester layer to top by adding saturated sodium chloride and shaking (less messy if you stopper that). We inject that upper hexane-methyl ester layer into GC. Easy.'

I cant express how much of a help this forum, I felt I was sinking for a while.

Many many thanks

Steve

'We saponify 1 gram fat/oil in a 100 ml volumetric flask using 10 ml 0.5 N KOH or NaOH in methanol on a steam bath 20-30 minutes. Then we acidify with either 12% BF3/methanol or 12% H2SO4/methanol and heat on steam bath five minutes. Then we cool, add 10 ml hexane, cap and mix, then float hexane-methyl ester layer to the NECK by adding saturated sodium chloride and shaking (less messy if you stopper that). We inject that upper hexane-methyl ester layer into GC. Easy.'

For fatty acid samples or soap samples, just dissolve in 12% BF3/methanol or 12% H2SO4/methanol and proceed as above.

Inject about 0.5 µl or more if using capillary column. I like cyanopropyl silicone solumns better than DEGS or PEG columns for this, depends on how much separation you require.

I'm not at work today, but after one has done this 1037 times, it sticks in the memory....save the other 1 ml for use if you accidentally knock over the first sample !!!

[ukstudent]

Many many thanks for the reply, it was a great help. However, I have a two queries if you could answer them please.
1) As we don't have standard 12% H2SO4/Methanol, is the H2SO4 solution measured in V/V or W/V? And how much of each would I need to make the solution?
2) Regarding acidifying process is there a specific pH to which I should acidify?

Once again, many thanks for your help, I really appreciate it.
Steve

[Consumer Products Guy]

Many many thanks for the reply, it was a great help. However, I have a two queries if you could answer them please.
1) As we don't have standard 12% H2SO4/Methanol, is the H2SO4 solution measured in V/V or W/V? And how much of each would I need to make the solution?
2) Regarding acidifying process is there a specific pH to which I should acidify?

Once again, many thanks for your help, I really appreciate it.
Steve

1. We add 140 ml concentrated H2SO4 to 860 ml cold CH3OH, carefully, in a fume hood, swirl to mix. It doesn't have to be precise, the acid is to catalyze the esterification. We're going to try VWR #101372-776 (sulfuric acid already made up in methanol) so our safety peoples are happy we don't have concentrated sulfuric acid around. Pre-made BF3-methanol is nice, but more expensive (e.g. Sigma).

2. "Regarding acidifying process is there a specific pH to which I should acidify?" No, just excess of acid. There is no real pH as it's not an aqueous solution. If you feel you need to neutralize first, go to neutral and then add a few ml excess acid-methanol.

[ukstudent]

Thankyou once again for the reply

I tried the method and it worked but the concentration was too high in the gcms machine once again

My supervisor said I had a concentration of 25 million when it should be no more than 1 million?

To make the h2so4/methanol I mixed 14ml h2so4 with 86 of methanol

To make the saturated sodium chloride I dissolved sodium nacl powder in water untill no more dissolved

I added 10 ml hexane as you said but still the results are way too concentrated

My supervisor told me as im damaging the machine if it comes out like this again I will not be able to use the machine again, please could you help me out, this project is getting really frustrating and im running out time. Thanks again

Steve