Iso and Terephthalic Acids Derivitisation

[mbdixon]

Hi,

I have developed a method for quantifying Iso and terephthalic acids in iso-octane. I basically just derivatise with BTFSA and I get great chromatography on an Agilent HP6-MS column.

However, I have to also do the same for the acids in an aqueous matrix. Without success I have so far tried:

• Direct usage of BTFSA
  Evaporating to dryness, dissolving in methanol and using BTFSA
  Evaporating to dryness, dissolving in isooctane and using BTFSA - not sure how soluble the acids are in isooctane
  direct use of BF3
  Refluxing in KOH and then using BF3

Any suggestions?

[Consumer Products Guy]

We assayed fatty acids routinely after making methyl esters using BF3/CH3OH or H2SO4/CH3OH.

We also used BSTFA to derivatize analytes of interest even more, even in consumer products containing up to 90% water. We diluted samples and standards in DMF or pyridine, added BSTFA and shook, then ready for injection.

When we had -NH groups to derivative, we used pyridine solvent and also heated capped autosampler vials 15 minutes on steam bath.

[mbdixon]

Thanks, I'll give the pyridine dilution a shot and see what happens.

I only tried the BF3 as we had it in the cupboard and I thought why not!

[Consumer Products Guy]

Pyridine acts as a hydrogen receptor/scavenger

[rb6banjo]

I have had good success getting benzoic acid out of a substantially aqueous matrix by using solid-phase extraction (C18). Be sure to dry the packing well by pulling air through it. Then, I just pass 2 mL of MeOH:BF3 (Sigma-Aldrich) through the packing and collect it in a reaction vial. This isolates the extract and gets it ready for derivatization. I heat to 100 °C for an hour. Cool it to RT. Dilute the mixture to 20 mL (to make 10% alcohol in water) and then analyze the headspace using SPME - looking for methyl benzoate (GCMS).

It might work for the methyl esters of your iso and terephthalic acids. You may have to go longer on the derivatization time to get both acid groups derivatized.
The methyl-ester derivatives might be too heavy for reliable headspace-SPME analysis. I haven't tried it before. I just thought I'd mention it.