spectrometry for machine 'smell'?

[insane_alien]

Could a digital spectrometer be used to detect compounds in the local atmosphere? i'm not sure if the air is dense enough to work in such a small space or even the tiny amount of aroma chemicals in the air that make a smell. just wondering if it would be possible to use this to make a sort of universal machine nose.

 

probably not, it popped into my head while writing stuff about heat exchangers and thought i'd throw it out there see if anything happens.

[Klaynos]

I doubt n will change enough for it to work on just air.

 

Ramen spectroscopy is I think what the biophysicists call it, a couple of my friends are working on using gold nanospheres and tiny quantities of impurities in liquid to see how much the impurities alther the SPP mode.

[John Cuthber]

Raman spectroscopy might work but, unless you stick the molecules to something solid, the Raman effect is very weak.

Mass spectroscopy gets used for this sort of thing.

I don't understand the reference to "n".

[insane_alien]

okay, i had doubts it would work.

[Klaynos]

n, the complex refractive index. It's how spectroscopy methods work...

 

So if two samples have the same n (where n=n+ik) values they will produce the same results as each other. So if the variation is small your resolution in spectrum shape may not be good enough...

[insane_alien]

doesn' spectroscopy measure absorbtion/emission lines? thats what i remember from my spectroscopy class a couple of years ago.

[Klaynos]

doesn' spectroscopy measure absorbtion/emission lines? thats what i remember from my spectroscopy class a couple of years ago.

 

Indeed, if we consider reflection, (which is R = 1 - A - T, where A is absorbtion and T is transmission)...

 

when you measure a result you are measuring (for s polarised light, you can formulate it for any polarisation state but I chose s for simplicity) R = r_s* r_s

 

Where r_s is the Fresnel coefficient of s polarised light, and r_s* is it's complec conjugate. The coefficients are depending on, n in the incident medium (1 in air) and n of the material, and the angle of incidence. And n is dependent on the photon energy.

[John Cuthber]

OK so for UV, vis, IR spectroscopy oyou have to measure the change in the of the refractive index. But what people actually measure and generally report is the absorbance. For a lot of things this will be near zero but a few things can definitely be measured this way. Mercury sniffers, for example, work on the absorbtion of light from a mercury lamp.

I'm pretty sure you could get some sort of spectrum out of the "smelly" air that was different from clean air. What would be tricky would be making any sense of mixtures of compounds. That might be part of the reason Mass spec gets used

[insane_alien]

also, isn't this the method they use to remotely detect atmospheric compositions on other planets. i was thinking that if it worked on a large thickness of air, it might be possible to do the same on a small thickness of air but with more sensitive equipment.

[John Cuthber]

Yes, you can and they do. In particular it's used to measure ozone but it's used for a lot of other things like measuring along the perimeter of a manufacturing site so they know if there's a leak.

The ozone measurements are UV I think but a lot of other stuff is done with IR.

None of these seem to be a chemical nose to me because my nose isn't a hundred metres long.

[insane_alien]

just think of it as smell-o-vision