Hi Eloi,

I am personally not aware of any method without derivatisation.

Probably, one of these might be helpful in your case where the method of estimation after derivatisation is a little more easier.

https://pdfs.semanticscholar.org/6cf0/967c20bbc4bc4fd11a5a88d85202d39f34d5.pdf

https://pubs.acs.org/doi/abs/10.1021/ac50082a008?journalCode=iecac0

https://pdfs.semanticscholar.org/e849/7c7a6084005d7af73fc25fc09f1bce30f863.pdf

https://www.researchgate.net/post/Does_anyone_know_of_a_method_for_determination_of_formaldehyde_using_titration

https://pdfs.semanticscholar.org/e849/7c7a6084005d7af73fc25fc09f1bce30f863.pdf

GC head space?

GC headspace could be used with mass spectrometer in SIM mode or you can try ECD (it is also giving nice aignal for -oxygen containing compounds).

regards

G

There are quite a few gas sensors known for formaldehyde:

Article Formaldehyde Gas Sensors: A Review

Some years ago it was shown that it is possible to quantify formaldehyde even in low concentrations by NMR, see:

Emeis D, Anker W, Wittern K-P. Quantitative 13C NMR Spectroscopic Studies on the Equilibrium of Formaldehyde with Its Releasing Cosmetic Preservatives. Anal Chem. 2007;79(5):2096-100.

Thank you people!! This will certainly keep me entertained for a while.

For precise and accurate results you should prefer Pre column derivatisation with 2,4 Dinitrophenyl Hydrazine.

Long path cell with FTIR is one recommended technique, with some limitations, here:

https://pdfs.semanticscholar.org/3f57/08385dc91beecfd0f50997cdd38d37e3377f.pdf

Not all of the derivatization methods are long and complex. Acetyl acetone is very fast and easy, plus it gives you the advantage of being able to measure the concentrations colorimetrically. Since your target range is low mg/L you'd be able to dilute your samples to get around any interferences and still have plenty of sensitivity.

Wolfgang suggested C13 NMR for this analysis. Since I am not clear as to your matrix, I will throw out the idea of using the more sensitive H1 NMR for the formaldehyde peak using an internal standard. No derivatization needed. Just need to add an internal standard to your sample and a deuterated solvent.

Thanks Douglas. I feel lured by how convenient 1H NMR is but perhaps the fact that my solutions (perhaps 'suspensions' would be a better word) will contain rather high quantities of Fe(Ni)S precipitates (and I guess that's paramagnetic) would make NMR too noisy. Unless I figure out a way of getting rid of the Fe-based precipitates beforehand. Perhaps ultra-centrifugation could be enough?

We used the DNPH method at UCL, it works well and is quite simple if I remember. You could try adding potassium ferricyanide solution to a sample and running NMR as it chelates the Fe but not sure about the Ni. May be worth a try perhaps.

Eloi, that does complicate matters, you are right. After you centrifuge the suspension to remove the Fe(NI)S, you might be able to add EDTA to complex any dissolved Fe. If you are real lucky, you might then be able to quantify the formaldehyde if the EDTA removes the iron from interacting with the formaldehyde.