[Garthh]

I'm producing biodiesel, I'm looking for ways to test for mono, di & triglycerides.  A quick & dirty method is mix 1 part biodiesel, 9 parts methanol  Shake well. the results should be clear!  If there is noticable drop out, there are several possible causes:
incomplete reaction, not enough caustic [potassium hydroxide] in the catalyst
overeaction too much caustic or water [leading to soap]
Adjusting the ratio [methanol/bio], doesn't refine the results.

I understand to get highly accurate results, I need some sort of spectrum analysis. Is there any sort of reagent testing I could do, even being able to differentiate between the 3 types of glycerides would be a good troubleshooting tool.

where would I find the wavelengths of various compounds, as they relate to spectral analysis?

Another test I use is washing w/water & measuring the disolved solids, salts of metals [potassium].  Most easily available test equiptment [probes] are designed for aquous solutions.  I need a way to quantify the indirect results, or suggestions for direct methods.

Thanks for your consideration.

[macman104]

I'm just going off of previous experience.  My experience has nothing to do with biodiesels, but the method may work for you, or it may not be applicable at all.

mono, di and tri *should* all react differently in terms of solubility to pH adjustments, right?  If so, you can try altering the pH of the solution in a separatory funnel and running TLC methods to see what various parts are present.  Just so you know where I'm coming from, we use this method to distinguish between mono, di, and tri version of the compound below.  Obviously, once we have the one we want, we run NMR to verify, but this is good for quick work to see which one we actually have.

So...whether it has any bearing on what you do or not, I dunno, but I thought I'd throw it out there.

[Garthh]

Thanks macman104,
I could use a little further clarfication. TLA's [3 letter acronym's], I need to know what TLC methods mean? separatory funnel I understand.
I think that you are suggesting that I alter the ph of my sample & mix w/a solvent [to be determined], observe for drop out?

[Yggdrasil]

TLC stands for thin-layer chromatography.

http://en.wikipedia.org/wiki/Thin_layer_chromatography

[macman104]

Thanks macman104,
I could use a little further clarfication. TLA's [3 letter acronym's], I need to know what TLC methods mean? separatory funnel I understand.
I think that you are suggesting that I alter the ph of my sample & mix w/a solvent [to be determined], observe for drop out?

First off, I have to apologize, I had the incorrect structure in mind.  Instead of it being 3 groups that have a t-butyl attached, they are the hydrolyzed version as a carboxylic acid instead.  But anyway, I'm not 100% sure what you mean by dropout, if by drop-out you mean enter the aqueous layer then yes.  If not, read on.  If I take the acid mentioned above, dissolved in say...ethyl acetate, and I add an aqueous layer.  Then, if I add NaOH and make the solution more basic, the carboxylic acids will change over to their water soluble salts.  I can then have them dissolve into the aqueous layer, and check for their presence in the aqueous layer through TLC.

Like I said, though, I'm not sure if this type of procedure is applicable to your case, but it may work along the same method of turning the alcohols into their water soluble form.

Hope the above was clear.

[Garthh]

Thanks Yggdrasil & Macman

I'll do some research on TLC, may be a test we could economically implement.

Now I'll clarify, when we run a 9:1 test I should refer to results as separation, not drop out.