DIY Spectrometry Kit

[Zapa]

I was wondering, would the DIY Spectrometry Kit from Kickstarter be any useful in drug analysis?
I would think not, since it's meant to be used with visible light, but maybe an Infrared light could be attached to it for our purpose.

What do you great minds of ADD think about it?

 

[Zapa]

Yeah, that's the one I was referring to.

after further reading into it, i had a glimpse of a "marquis/mecke/mandelin 2.0" tool kit there. and it looks quite interesting. problem is:
where you get the trustworthy references from, that your test will apply to?

Maybe we could create a "Bluelight Spectrometry Chart" using substances already tested by other means!

 

[Zapa]

Yes indeed. I'll probably get one before the end of the Kickstarter (a spectrometer for less than an hundred bucks, why the hell not?!) and I will try it on a few substances.
I'm gonna post the results here if people are interested. I don't have NMR or HPLC so it might not be worth much, but it could give a good idea of what is achievable with this kit.

 

[Roger&Me]

Visible spectrometry isn't of value in qualitative analysis because the spectra are too broad. Its really only useful for generating a beer's law plot to calculate the concentration of a colored solution.

 

[sekio]

Rog&Me is right. Colorimetric analysis is a really really poor method of analysing drugs.

 

[sekio]

From my POV, all this can do is tell you what color your drugs are. It's not an IR or UV spectroscope! All it will tell you is "this drug reflects yellow light, and blue light".
So this tool would probably be most useful for checking the colors developed from a Mecke/Maruqis/Mandelin.

It's not going to be a revlution in drug testing, sorry... Can you tell if your MDMA is cut with PMA just by looking at the color? Because if you can't this machine certainly won't be able to.

 

[Zapa]

It's not an IR or UV spectroscope!

They state in the FAQ that you can choose your own light source and they cite UV light as being a possibility.
I suppose IR could work too, but I could be wrong.

Would it be worth something using these light sources?

 

[sekio]

The CCD in a typical mobile phone is not very sensitive in the UV or IR ranges that are used for chemical "fingerprinting". Depending on the lens system, composition of the lens and CCD, ambient temperature, & presence or absence of a filter, you may be able to get some UV or IR data, (more likely IR) but it is unlikely.

You certainly cannot do UV-VIS or FTIR. with this set up. One would need a specialised detector, laser source and a beamsplitter to do "proper" fourier-transform spectroscopy.

The spectra used in UV/IR chemical fingerprinting are very far away from the visible light spectrum.

IR spectroscope measured wavelengths are typically around 3000-15000 nM. The average IR LED is around 800 to 900 nM. Maybe you could find a 1000nM one. A typical CCD or photocell also won't respond to those high wavelengths.

UV-VIS usually uses wavelengths of below 250 nm,LED's are 240-360 nm. So you *might* be able to squeeze out one end of a UV spectrum, but it's not going to be enough...

 

[Zapa]

Thanks a lot for your answers sekio!

So I guess if I can't find some cheap ≈150nm UV light it is not worth purchasing...

 

[sekio]

Assuming the plastic lens in your camera-detector transmits UV light at 150nm (probably won't - UV/VIS usually uses things like sapphire lenses), and the CCD also detects the UV and doesn't just get fried. 150nm is into mercury disinfection light territory - and is the "bad" UV that you block with sunscreen!

 

[Zapa]

150nm is into mercury disinfection light territory - and is the "bad" UV that you block with sunscreen!

You've got a point here, and it's not even the UVB blocked with sunscreen, it's the more nocive UVC blocked by the ozone layer.

 

[tryp2fun]

Thanks a lot for your answers sekio!

So I guess if I can't find some cheap ≈150nm UV light it is not worth purchasing...

150 nm is in the vacuum UV region. It is absorbed so strongly by oxygen that it is not transmitted through air. The UV region is 200-400 nm, and most of that does not make it to the ground. Only the UV A, which is 320-400 nm, actually makes it to the ground and is responsible for sunburn. UV is also absorbed by glass, so you would need to use quartz or sapphire optics to see it. The most common UV light source is a mercury lamp, which has a strong emission line at 254 nm.

 

[Ceres]

The most accessible analytical technique available to people on a low budget is thin layer chromatography really.

 

[Roger&Me]

But what can it really tell you in the case of an ecstasy pill or something like that? I mean, you will likely get a bunch of spots with different Rf values even if MDMA is the only active ingredient in the pill. Not to mention, you need access to something like phosphomolybdic acid to see spots that aren't UV active. Less accessible in practice than in theory.

 

[Ceres]

But what can it really tell you in the case of an ecstasy pill or something like that? I mean, you will likely get a bunch of spots with different Rf values even if MDMA is the only active ingredient in the pill. Not to mention, you need access to something like phosphomolybdic acid to see spots that aren't UV active. Less accessible in practice than in theory.

Well, firstly if you get multiple spots then you know you have impurities or adulterants...with a good seperation spots can be cut out and redissolved in solvent then subjected to say microcrystalline tests or simple stains like mecke/marquis. Running plates with reference standards of common cutting agents and so on can further clarify the picture. Staining can get quite esoteric but then again there are simple techniques like charring that can reveal more information than just UV or iodine...TLC can be pretty powerful if you go about it the right way, and is considerably cheaper than any HPLC setup with UV/VIS or GC/MS gear...

edit> you can also infer further data based on the molecules affinity for the silica and how far it might move up the plate, given london forces/hydrogen bonding...running multiple samples together, even 2d TLC can give you a lot more information that a straightforward spot test of a pill with marquis. Also how you treat the sample before spotting the plate can help, e.g use a spotting solvent that wont dissolve sugars or starches and so on.

 

[Roger&Me]

Well, firstly if you get multiple spots then you know you have impurities or adulterants...

Yeah, I know... I guess I didn't phrase that very clearly. I was just saying that even in a good pill with MDMA as the only active compound, there's likely a lot of inactive stuff that will show up on your plate. Pills are such a complex mixture of binders, fillers, and crap.

But I do agree its far more accessible than HPLC, GC/MS, etc, which are basically off limits to the layperson due to their prohibitive cost.

 

[Maus.]

I think you could tell from the relative size and intensity of the spots on the TLC plate how pure the sample is. Sounds pretty straightforward.

 

[Ceres]

Well it's interesting because you can image the plates in controlled conditions, then use image analysis software to process the results and it starts to become almost quantatitve

 

[sekio]

How many people have access to coated plates, developing agents, and chromatographic solvents, though?

 

[skillet]

How many people had access to Marquis/Mandelin etc. before distributors started packaging them for pill testing? I don't see any reason why the same couldn't be done for tlc. If you could just spray the plate with Marquis it would be a fairly easy and informative test. I'm not sure that would work, but I don't really see why it wouldn't.