Chromatography Forum

Did some digging around but haven't quite found an answer yet so figured I would post here while I am still looking in case anyone has any experience.

I am looking to derive free fatty acids to a pentafluorobenzyl ester derivatives. I know the PFB reagent is best suited for ECD detector but an FID should still be able to pick this derivative up, correct?

Next issue is column, I am currently only using a SP-2560 for FAMEs. I assume this column would not be sufficient for separation of fatty acids with a PFB derivative. Don't necessary want to blind test as the columns aren't cheap, I suppose I could just test with an older column we have laying around that is no longer giving good separation (just thought of this).

If a SP-2560 is not a good column choice, does anyone have any that would be good?

Thanks in advance.
jp

This derivative is not optimal for an FID. Reflux for 3 hours in acetone? Why not just use a methyl ester?

The higher mol wt of the PFB ester makes it harder to elute for the longer FA and short FA esters could easily be lost during the reaction.

best wishes,

Rod

Hi Rod Thanks for the reply.

The PFB Derivatives are done at room temp for 20 mins in acetonitrile. Reason we are not wanting to use methyl esters is we were trying to remove the separation step (SPE or TLC) to isolate only FFA from plasma to increase throughput.

PFB-bromide is selective (supposedly) for phenols, thiols, and carboxylic acids. This should be selective for just creating FFA derivatives. There are some other concerns I have on this though in that the extraction steps leave the triglycerides and cholesterol esters in tack and being injected into column, which I'm worried about junking up column or interfering with analysis. Also slightly worried about creating FFA from other lipids with water in extraction process since there is also a base involved.

At the moment we have moved on from this and are just using SPE to isolate the FFA and then create methyl esters. It however is something we are researching more.

20 min at ambient? That is cool if it does work.

Good luck and keep all of us informed on your progress.

Rod

I cannot answer your question, but I have a question related to your topic.

What type of SPE column are you using? Aminopropyl bonded silica? What brand? What are the material for frits and column housing?

What I have read is that plasticizers and fatty acid contamination in frits and column housing can bury your real signal, especially in the free fatty acid portion.

I just tested several SPE columns yesterday. All portions in the blank run showed many peaks, and significant amount in the free fatty acid part.

I am trying to find some prepacked GLASS SPE columns, but it seems that I could not find any.

Hi Rod Thanks for the reply.

At the moment we have moved on from this and are just using SPE to isolate the FFA and then create methyl esters. It however is something we are researching more.

I don't know if they would consider it, but Supelco (where I used to work) sometimes would do special items on a custom basis.

Call the company and ask for the SPE product manager and ask.

best wishes,

Rod

Not sure which columns you tested or what solvent washings and volumes you are using.

We are currently testing the BondElut Aminopropyl bonded silica columns from Agilent. Currently we are using the standards ones, but we have inquired about getting these columns with stainless steel frits, which you just need a special part number for (they are much more expensive). Our blank runs are only showing two very small peaks that would interfere with C16:0 and C18:0, we are estimating it would be less then 0.1% of an actual sample peak and at the moment are contemplating considering whether or not this contaminant would make a difference, leaning strongly towards no as these are not our main peaks of interest. These contaminant peaks are also very reproducible from blank run to blank run, so we are also considering subtracting out a blank sample with each run.

Hope this helps.

Thanks you for your input.

I tested 100mg/1ml Hypersep NH2 columns.

I am doing lipid class separation using the conventional protocol: After washing the column twice with 1 ml n-heptane, elute with 1 ml chlorofrom:2-proh (2:1), 1ml 2% acetic acid in ether, and finally 1ml methanol twice.

Then do a derivation reaction by adding acetyl chloride before injection on GC.

I see contamination peaks in all 3 eluants, also around 16:0, 18:0 (have not put any efforts to identify what exactly they are). They are other smaller late peaks also.

The contamination problem from plastic housing are reported in some articles. Particularly, this article, http://www.ncbi.nlm.nih.gov/pubmed/3145937, has data for Bond Elut Aminopropyl bonded silica column, they mentioned the contamination in the free fatty acid part is equivalent to the amount of FFA present in 5 mg of myocardial tissue or 10-15 ml of plasma, which is crazy.

Unfortunately, when I did the very limited number of controls, the peaks showed different peak profile in both heights and peak numbers.

I am only interested in the last elution, the supposedly phospholipid portion. I have not seen any reports talking about contamination in this part, but none of the reports used plastic columns . I never used plasticware before this for any lipid work, I am just afraid of plastics.

I am seriously considering to pack my own column using disposable glass pastuer pipet at this point.

Not sure which columns you tested or what solvent washings and volumes you are using.

We are currently testing the BondElut Aminopropyl bonded silica columns from Agilent. Currently we are using the standards ones, but we have inquired about getting these columns with stainless steel frits, which you just need a special part number for (they are much more expensive). Our blank runs are only showing two very small peaks that would interfere with C16:0 and C18:0, we are estimating it would be less then 0.1% of an actual sample peak and at the moment are contemplating considering whether or not this contaminant would make a difference, leaning strongly towards no as these are not our main peaks of interest. These contaminant peaks are also very reproducible from blank run to blank run, so we are also considering subtracting out a blank sample with each run.

Hope this helps.

Thanks for the suggestion, chromatographer1.

I do see Sigma sells a 6ml glass column with Teflon frits(maybe a Supelco brand?), but its size is a bit too large for me.

I am only dealing with ~ 1mg total lipids to get the phospholipid part. 50mg to 100 mg adsorbent is what I needed. That little adsorbent may only fill a very short portion of the column.

However, if I go that route, I might as well just do everything by self and pack column in a disposable glass pastuer pippet.

I don't know if they would consider it, but Supelco (where I used to work) sometimes would do special items on a custom basis.

Call the company and ask for the SPE product manager and ask.

best wishes,

Rod

Tiggeria -
Just wondering if you found a solution to your issue? Did you get special order glass parts or did you just pack your own in disposable glass pastuer pipettes? We tested glass pipettes by cutting open a plastic cartridge and packing our own columns and found the contamination decreased significantly, almost to nothing. I'm still thinking the majority of the contamination is coming from the plastic frits in the SPE columns though. Through agilent we can order these aminopropyl with stainless steel frits but we haven't went that far yet.

I'm assuming packing these SPE materials are no different than packing a normal column for flash chromatography. Or is there a standard method you should follow?
Thanks.