[Babcock_Hall]

Some time ago the carotenoid samples I had prepared from microbial sources were subjected to mass spectral analysis.  Apparently they had polyethylene glycols as contaminants.  I am returning to the work now and would like to avoid this problem, obviously.  We tried to avoid plastic as much as possible in our preps (our HPLC filters have plastic housing, however), and we tried to make sure that our glassware had either been cleaned in a base bath or with a mixture of nitric and sulfuric acids.  The final step of our purification was RP-HPLC over C-30.  The carotenoids were eluted isocratically in 60/40 MTBE:MeOH.  The MTBE had been treated with alumina and filtered.  Any guesses as to where the PEG was coming from and how to avoid it?  Thanks in advance.

[discodermolide]

Some time ago the carotenoid samples I had prepared from microbial sources were subjected to mass spectral analysis.  Apparently they had polyethylene glycols as contaminants.  I am returning to the work now and would like to avoid this problem, obviously.  We tried to avoid plastic as much as possible in our preps (our HPLC filters have plastic housing, however), and we tried to make sure that our glassware had either been cleaned in a base bath or with a mixture of nitric and sulfuric acids.  The final step of our purification was RP-HPLC over C-30.  The carotenoids were eluted isocratically in 60/40 MTBE:MeOH.  The MTBE had been treated with alumina and filtered.  Any guesses as to where the PEG was coming from and how to avoid it?  Thanks in advance.

Any PEG's in the bug growth medium?

[Babcock_Hall]

We are working with a mixed population of microorganisms in an agricultural waste lagoon, and it is possible that the lagoon has PEGs.  One would think that subsequent purification steps might separate PEGs from carotenoids, which are generally less polar.  One of the carotenoids has a methoxy group at either end, but the rest is just hydrocarbon.

[discodermolide]

We are working with a mixed population of microorganisms in an agricultural waste lagoon, and it is possible that the lagoon has PEGs.  One would think that subsequent purification steps might separate PEGs from carotenoids, which are generally less polar.  One of the carotenoids has a methoxy group at either end, but the rest is just hydrocarbon.

Mass spec will see everything. They may be present in trace amounts only. Any chance of doing LC/MS to try and quantify it?

[fledarmus]

Have you tried running blanks of everything? Take a sample of purified water and subject it to all of the same treatment conditions that you use for your lagoon water samples? That could tell you if the PEG was a contaminant from your method or a component of your sample.

[Babcock_Hall]

At one point I thought that the MTBE from the LC purification might be the problem.  IIRC I took 50-100 mL MTBE and removed it by rotary evaporation, then ran some CDCl3 through the flask and checked the contents by NMR.  I don't think I saw anything.  I like the idea of blanks, but I have not run any yet.

We now have better MS facilities than we did before (when I did my previous MS work I went off site).  We do not have a C30 column for our mass spectrometer, but we might be able to do LC/MS with a C18 column.  The person in charge of the mass spectrometer would like me to check my present sample with NMR to avoid putting PEGs on the column, because the PEGs are difficult to remove.  I am also changing my base bath in case there are impurities there that have built up over time.   

[Babcock_Hall]

I took C-13 and H-1 NMR spectra of polyethylene glycol 400 off the shelf to use as a reference.  The main H-1 envelope is near 3.65 ppm, and the main C-13 peak is between 70-71 ppm in CDCl3.  In both cases there are smaller peaks on either side of the main envelope.  This is close to what I expected from the literature value in d6-DMSO (I only have access to the first page of this article):  Julian M. Dust, Zhi Hao Fang, J. Milton Harris; Macromolecules, 1990, 23 (16), pp 3742–3746.  http://pubs.acs.org/doi/pdf/10.1021/ma00218a005
I went back and checked an old NMR spectrum of all-trans spirilloxanthin from several years ago, and I now see a crosspeak in the heteronuclear correlation NMR that has chemical shift values very close to these. What I still cannot figure out is how this contaminant is getting into our samples.  We will run some TLC this week, to see whether or not silica would be expected to remove it.