[sofian.fst]

Does anyone knows about the internal standards if we want to analyze nicotine in hair?

[Archer]

Nicotine-2,4,5,6-d₄ will give you the best results.

[Arkcon]

Nicotine-2,4,5,6-d₄ will give you the best results.

Although it would be best, sofian.fst:, if you told us what your analytical method was.  You might not be able to resolve the deuterated analogue from the active with your method.  Also, why do you need an internal standard?  If you just need something to cancel out loss on handling issues, then many compounds may be similar enough, again, depending on the analytical method.

[Archer]

Hair is not a particularly simple matrix to work with.

Once it has been digested there are so many compounds present that only Gas Chromatography Mass Spectrometry (GC-MS) or Liquid Chromatography (LC)-MS is sufficiently selective and sensitive to detect a single compound.

If you are using a mass selective detector then deuterated analogues are without doubt the best internal standard to use as a one point calibration is sufficient.

GC-FID (Flame Ionising Detector) or NPD  (Nitrogen - Phosphorus detector) or even LC-DAD (diode array detector) produce too much noise when analysing hair so are not viable techniques, in my experience.

There are no commercially available molecularly imprinted polymers, that I know of, for selective separation of nicotine from complex matrices and even if there were these are not great for analysis of drugs in hair.

[Arkcon]

If you are using a mass selective detector then deuterated analogues are without doubt the best internal standard to use as a one point calibration is sufficient.

Granted, assuming the ionization procedure doesn't promote hydrogen-deuterium transfer.  The "softer" the ionization, the more likely it is to happen:  http://en.wikipedia.org/wiki/Hydrogen%E2%80%93deuterium_exchange#Mass_spectrometry

That's a pretty crappy reference, but I'll try to look for another.

[Archer]

I've never come across that as being an issue in small molecule analysis.

I suppose that maybe the increased sensitivity and selectivity outweighs the problems associated with this phenomenon?   

As you are looking, in this case, for an M : M+4 ratio I would expect non-linear responses as this ratio changes, this would make the single point calibration very inaccurate indeed.

This is certainly food for thought!