Chromatography Forum

Hi everybody,

we have to analyze the formaldehyde level in a product dissolved in water.
At acidic pH the active ingredient release formaldehyde so we have to derivatize the aldehyde in a slightly basic media (about ph 7 to 8)..

Do you know of any derivatization reagent that works at this pH and at ambient temperature?
We have previously worked with DNPH but we think that he need an acidic media to derivatize the aldehyde..

Thanks in advance for the answer,

best regards

We've used DNPH for HCHO derivatization and quantitation; however the DNPH reagent is in acidic solution. However, check this out; HPLC with post column is used quite a bit in Europe, with HPLC to separate HCHO from HCHO-donor species, then using post-column reagents, then detector.

http://www.pickeringlabs.com/catalog/pd ... dehyde.pdf

CPG has an excellent suggestion. There are also a number of other reagents available, some LC and some GC. A summary is available in:

Bicking, M.K.L., Cooke, W.M., Kawahara, F.K., and Longbottom, J.E., "Method Development for the Determination of Formaldehyde in Samples of Environmental Origin." Am. Soc. Testing Mat., Special Technical Publication 976, 1988, pg 159.

Also, DNPH will work at a pH of about 5, if that is acceptable for your samples. See:

Bicking, M.K.L., Cooke, W.M., Kawahara, F.K., and Longbottom, J.E., "The Effect of pH on the Reaction of 2,4-Dinitrophenylhydrazine with Formaldehyde and Acetaldehyde." J. Chromatogr. 1988, 455, 310.

If these options are not possible, write back and we will think of something else.

Thanks mbicking and consumer product guy,
I think that the approach of a post column derivatization will work but unfortunately we don't have the instrument to perform the post column derivatization..I don't know if the first separation could be done by SPE (diol column?) and after that make the derivatization by DNPH...

I'm trying also the DNPH approach with and without HCl to see if it works..

The problems that we have with this sample are:
1) The active is very concentrated in a water solution and could contain more than 1% formaldehyde
2) Formaldehyde is one hydrolysis product of the active compound. The concentration of formaldehyde is in equilibrium with the active cmpd.
3) As soon as we dilute the sample the hydrolysis of the active will increase rapidly and so the content of formaldehyde
4) If we increase the temperature the concentration of formaldehyde will increase (we get trouble trying the analysis by GC-headspace)
5) If we decrease the pH (below 7-8) the concentration of formaldehyde will increase..
5) If we can't dilute the sample we have to add an "huge" amount of derivatizating agent and we could have incomplete derivatization..

Sorry for the long post and thanks for any answer..

Well, you have a very difficult problem.

If the active component is indeed in equilibrium with formaldehyde, then any attempts to derivatize the free formaldehyde (or remove it by any chemical or physical method), will result in more formaldehdye being generated. This is LeChatlier's Principle - "when a system at equilibrium is subjected to a stress, the system responds to relieve the stress." In this case, removing formaldehyde will cause the equilibrium to be shifted toward formation of formaldehyde. This is true only if a true equilibrium exists; I'm not sure about that here but we don't have enough information.

I don't know if we would be able to help you without getting much more information than you would be willing to post on a public forum. If you want to set up a confidential discussion, please contact me privately.

Consumer Products Guy may have some ideas; he seems to have lots of useful suggestions. :idea:

Just to throw a few additional points into the discussion, this time in defense of Agilent's Zorbax: the outer shape of a particle has absolutely nothing to do with the performance of the column. The reason that Phenomenex found "broken particles" in the Zorbax material is that they unpacked a column (it is otherwise not possible to get a packing from the competition). I can show the same pictures for the Luna packing. And, lastly, Phenomenex does not make their own silica, they buy it, while Zorbax is home-made by Agilent (as Nova-Pak, Spherisorb, Symmetry, XTerra, XBridge, AQUITY packings etc. are home-made from the bottom up by Waters)

You could try the Nash method, it's an old method and may not be as accurate as DNPH or PFBHA but it does work (and is an old ASTM method)

http://my.att.net/p/s/community.dll?ep= ... d=8002&ck=

The link above describes performing it as post column reaction but we have done it as pre-derivitisation. The reaction is ammonium acetate/acetic acid (Not sure what pH you end up with!!!!) and pentadione reacting with the HCHO to give a lutadine species which is bright yellow!

Matt

The Nash acetylacetone procedure works at various pH, see the original publication from about 60 years ago. We have used the reagent both with and without acetic acid, both work. We found HCH to react without adding heat, in about 2 hours of sitting, where HCHO donors often required heat to break them down, so were able to estimate a "free HCHO" level.

I also looked at the Nash procedure years ago (the ASTM reference above). The pH was about 4 - 5, and at room temperature, the reaction was very slow, with complete reaction requiring 6 hours or more.

But if your "equilibrium" statement is correct, then every derivatization method is going to be an exercise in futility. As soon as you start removing formaldehyde, whether by reaction or some other means, the system is going to respond by generating more formaldehyde, and the observed levels will keep rising until either the source or the reagent is used up.

Your only hope is to separate the formaldehyde from the precursor.

First of all I would say a great thaks to everybody for the help and the share of your experience...

I've made one analysis derivatizing the formaldehyde standard and sample with DNPH at pH 1 and 5. At pH 1 almost all the active seems hydrolyzed to an impurity and formaldehyde..
At ph 5 I have found some interesting results:

sample diluted 1:10000= Form. conc. 0.3% (final pH 5.2)
sample diluted 1:1000= Form. conc. 0.15% (final pH 5.6)
sample diluted 1:100= Form. conc. 0.05% (final pH 6.4)

The DNPH concentration was in large excess and the peak of DNPH was still present also at the 1:100 dilution..
I don't have used a buffer so the final pH of the samples were different from that of the standard (about 5.1).. This maybe could cause the differences in concentration at the three dilution.
I'm planning to use a buffer to have the same pH in the standards and in the samples.
Dear mbicking which buffer have you used for the analysis at pH 5? Unfortunately I don't have access to your article (Bicking, M.K.L., Cooke, W.M., Kawahara, F.K., and Longbottom, J.E., "The Effect of pH on the Reaction of 2,4-Dinitrophenylhydrazine with Formaldehyde and Acetaldehyde." J. Chromatogr. 1988, 455, 310) so please can you tell me which buffer have you used?

Thanks again for the answers..

For a pH of 5, an acetate buffer is the best choice.

Send us results for a buffered experiment. The results you report may be a pH effect only.

Do you have the ability to determine the concentration of your active? If so, that would be useful to determine if any is hydrolyzing to formaldehyde.

Today I've tried to use an acetate buffer (50mM pH 5.0), as soon as i get something new I'll send you the information..
We are a contract lab and we don't have the method for the analysis of the active (our costumer have it), but we have to develop the method of quantification for formaldehyde.

Thanks again,

bye