Dimethylamine

[Mach]

How would you determine traces of dimethyl amine (<0.1% w/w) in 1,1-Dimethylbiguanide HCl (metformin)? Derivatisation/HPLC? I have access to ASPEC, any idea for an automated analysis with derivatisation?
Would a GC method be better?

[Mark Tracy]

If you have an ion chromatograph, the Dionex IonPac CS18 can do dimethylamine. I'm not sure where it elutes, but metformin does come out and is detectable (I'll ask one of the other chemists who actually did it). Take a look at the examples around page 18: http://www1.dionex.com/en-us/webdocs/38 ... 18_V28.pdf

[chromatographer1]

Your dimethylamine will be present as a chloride salt. You would have to convert the salt to the free base in order to do GC analysis.

If you could put the sample into a basic pH system you might be able to do GC headspace or direct GC injection, ............ OR NOT.

I would stay with IC and not try to do GC.

best wishes,

Rod

[Mach]

The trouble with making it basic is that metformin will split into dicyandiamide and dimethylamine, and that is why I have considered derivatisation. We do not want to form the substance we are looking for.

[Mark Tracy]

I asked one of my colleagues, and on the CS18 column, dimethylamine elutes in 5 mM methanesulfonic acid, and metformin elutes in 25 mM acid. Also, the CS17 column is a reasonable alternative. Detection is by conductivity.