DIY Spectrometry Kit

[sekio]

The M-reagents are all freaky heteroatom oxoacids (or their salts) in sulphuric acid. I can't imagine they would have been widely available.

Even if you choose a M-stain as a developer, most people aren't gonna have 2% ether in pentane to run a TLC, nor coated TLC plates. I guess it's up to Dancesafe to figure it out.

(You're right that a Marquis type stain would be best for ID)

 

[skillet]

That's my point, they're only available now because some people decided to make them available. You can buy tlc plates on a certain online marketplace, it just needs someone to sell smaller quantities. As for the eluent, I'd try DCM/MeOH(5-10%)/conc. aq. NH4OH(0.5-1%). If you can sell vanadate/formaldehyde in conc. sulfuric acid, I don't really see why you couldn't sell that. If the DCM is a problem toluene/xylene (with some adjustment) could probably be used instead.

I'm sure there're even more innocuous solvent systems that would work too, that's just what I'd use in a lab.

 

[Ceres]

Selecting an eluent depends on the nature of the analyte, but you can go a long way with stuff like MeOH/IPA/acetone/toluene........

 

[sekio]

Xylene/ethanol or even limonene/ethanol plus ammonia would be ideal solvent systems - DCM is not exactly easily available

Limonene/ethanol/ammonia TLC would be very interesting. One could even develop the plates with vanillin/sulfuric acid. Nontoxic green reagents all round.

 

[Ceres]

UV245nm reactive plates make exposure a breeze....

 

[Blue Druid]

Most of this discussion is way over my head, but i get the impression that the kickstarter spectrometer isn't useful for drug analysis. Is this solely because of lack of uv light & poor mobile phone camera quality or is there something more fundamental to it.

Sorry if i'm bringing down the intellectual standard of ADD, just want to understand if the diy spectrometer can be improved upon for drug analysis purposes (or if countertop version will work) or if there's some deeper issue rendering it any diy/home spectrometer unsuitable.

Spectrometers in question are at http://www.kickstarter.com/projects/jywarren/public-lab-diy-spectrometry-kit

 

[sekio]

The reason the DIY spectrophotometer can't do shit for drug ID is because it is a visible-light spectrophotometer, and not a UV/IR one.
(The spectra commonly used for structural deduction and chemical ID are not acquired from the visible area.)

Analogy: This is like an AM radio, whereas chemists use FM radios for their work. They are both radios, sure, but they operate on two seperate signals.

 

[electrodevo]

Most of this discussion is way over my head, but i get the impression that the kickstarter spectrometer isn't useful for drug analysis. Is this solely because of lack of uv light & poor mobile phone camera quality or is there something more fundamental to it.

The executive summary I get is:
(*) Range of UV and IR LEDs is not broad enough to ID chemicals
(*) CCDs are not sensitive enough to ID chemicals

Another issue I see with spectrum is that even with the ability to fingerprint, from what I understand you have to match a reference compound. It's not the best tool for IDing unknown compounds. Also the compound would have to be reasonably pure from what I understand which is not a guarantee with random powders. Quite a headache.

That being said, although they aren't used for drug ID I'm sure, there are CCD based spectrometers on the market that are $1000-$4000. It would be interesting to know the difference between that and the $10 one.

The TLC discussion is fascinating to me because there's currently no easy way for the average person to determine compound purity. A "mass market" TLC would be probably the only feasible way to do that.

Apart from reagents, I wonder if there may be a way to cobble up melting point and freezing point deprivation tests at home. Though it would probably require a bit of equipment investment to get any sort of accuracy. Certainly less equipment than GCMS/NMR though.