Kitchen Chromotography - Qualitative or Quantitative

[mongman]

OK I have a question for those with a little chemistry in their background.

I remember in highschool chem. doing some basic cromotography using a simple strip of paper and solution containing the chemicals in question. Would it be possible to use a similar method to assess the content of a pill. And would this be a quantative or qualitive analysis?

Obviously some relativly pure MDMA would need to be procured to calibrate the test (so you know how far along the strip that particular molecule traveled) but relatively pure MDMA is not that hard to come across.

 

[BigTrancer]

There are things called Thin Layer Chromatography (TLC) plates which someone was looking into a while (maybe 18-24 months) back. I'm not sure what the results were, or if they never arrived.

BigTrancer

 

[Fetish Jester]

Well, looking back at my days in analytical chem, it would still be qualitative. You would have the calibrated strip, but you can still only test for the presence of MDMA. What TLC does is check for the presence of a certain chemical.

If you want a quantative result, the only chromatography that I could recommend is column chromatography, and even with that it's either via titration or mass spec (maybe) that will provide a result. Both of these either require some hard core equipment and calibration or access to a proper lab. Also, for a proper TLC test, you'll need some real equipment (standardised TLC strips and the such)

My thoughts is just use an EZ test, cause it's as hard as hell to find out how much of a substance is in a sample.

 

[cuddlefish]

It is very difficult to get reliable quanitative results and qualitative separation without something along the lines of GC/MS (Gas Chromatography/Mass Spectroscopy) or HPLC (High Performance Liquid Chromatography). The equipment used in both of these techniques is rather pricey, to the tune of $10K. Thin Layer Chromatography on the other hand is effective in qualitative analysis (separating the components of a mixture) and is very cheap. Of course there is a catch, one has to use the right solvent combination for the types of compounds you're trying to separate and this requires a bit of experimentation. There is no reference guide for solvent combinations.

Once the components are separated and dried on the plate several methods can be used to show up the spots, under UV light, charing the plate (burning the components on the plate), and applying a reagent tester(ie. mandelein or marquis) to the plate to observe the colour change on the spots. That way the tester won't mask the presence of other compounds in the mixture because they've all been separated.

The problem is the solvent combination. I don't have any standardised chems and don't particularly want to go wasting pills feeling around for the right combo. I tried to get a thread going in pill testing a while ago for the BL'er chemists to brainstorm different combos and their pros and cons. but it just didn't take off.

If we did get the right combo we could put EZ test out of busness so I wouldn't expect any help from them. :D

Cuddles

 

[nanobrain]

show me where ya can get an HPLC or a GC/MS for 10 grand, 50-70,000 USD more likely. and ya gotta have an internal standard and calibrate the equipment for the specific reagents.

oh to op, its chromAtography

 

[cuddlefish]

^^^ I didn't exactly go researching prices. I'm not in the market for one and its not an item which is advertised with a price tag attached. I've been told 10K if you look around for bit (ie scrounge second hand non ISO equip.) and put it together yourself which is not that big a deal. This is still out of the price range of most people and if it was just found in someones house, I think a few questions would be asked.

Anyway can we think about possible solvent combos? Personally I rekon something along the lines of a methanol/acetone combo would be an interesting start.

Cuddles

 

[phase_dancer]

nanobrain: you can build a reasonable HPLC from new parts for $10K AUS or less.

Pump - ~ $1,500
Injector~ $2,500
Column/s < $1,000
UV/VIS detector ~ $3,000
integrator/chart plotter ~$1,500

A second hand device could obviously be purchased for less. GC is also not that expensive, although a new unit would set you back at least $25K. Second hand units however can go for as little as a $1-2K. MS is the big cost. A new unit from Varian will set you back around $70K, and this unit is not a stand alone, but works on the back of it's companion GC. A stand alone, state of the art MS can cost upwards of $500K

But the important thing to remember with any of these devices in relation to cost, it that apart from initial outlay, ongoing operational costs can be considerable. Columns in HPLC *wear out* and different types are required in different applications when isolating the same substance from different mixtures. Solvents must be of the highest purity (costly) to avoid any interference to results and to avoid contamination of the column. GC requires gases which also must be of high purity.

HPLC often requires the use of low flow rates to ensure adequate separation of a mixture. This can mean waiting hours for a single run. At other times, different solvent combinations must be trailed to obtain adequate separation; necessary to obtain a quantitative result. In all, great volumes of solvents can be used establishing optimum machine parameters for that mixture before you actually do the analytical run. If it were rushed, that big single peak which is taken to represent MDMA could in fact be two peaks of different substances superimposed.


cuddlefish: In regards to colour markers for TLC, have you seen this article from microgram?

Microgram Journal, Volume 1, Numbers 1-2 (January - June 2003) 23
Evaluation of Ninhydrin Analogues and Other Electron-Deficient
Compounds as Spray Reagents for Drugs on Thin Layer Chromatograms
Myriam Azoury*, Avraham Zelkowicz, and Zafrir Goren
Division of Identification and Forensic Sciences
Analytical Chemistry Laboratory
Israel Police National Headquarters
Jerusalem 91906, Israel
[email: [email protected]]
Joseph Almog
Casali Institute of Applied Chemistry
The Hebrew University of Jerusalem, Israel

ABSTRACT: Twenty-four electron-deficient compounds were evaluated as potential spray color-reagents for
basic drugs on TLC plates. Two of them, 4-chloro-7-nitro-2,1,3-benzoxadiazole and 5,6-dimethoxyninhydrin,
were superior to ninhydrin with respect to sensitivity and selectivity, and offer considerable potential.

KEYWORDS: Thin Layer Chromatography, TLC, Spray Reagents, Ninhydrin, Illicit Drugs

Introduction
Since the discovery by Dutt and Teo1 that spraying thin layer chromatographic (TLC) plates bearing drug spots with ninhydrin produces a variety of colors that can distinguish between many drugs, this reagent has been intensively used in this laboratory. The colors that are produced with ninhydrin, when correlated with the specific migration values (Rf) for each spot on specific TLC plates and using select solvent systems, greatly enhance the specificity of TLC for various drugs.

In forensic laboratories, the main use of ninhydrin as a spray reagent has been for detection of fingerprints, especially on porous surfaces such as paper and cardboard.2-4 However, despite its great utility, research has
continued to develop even more sensitive or selective reagents. Over the last two decades a significant number of ninhydrin analogous and similar, electron deficient compounds have been synthesized and evaluated as fingerprint reagents. Some of these new reagents have displayed superior properties versus ninhydrin in their sensitivity to amino acids and latent fingerprints, particularly in the fluorescence mode.2-7

The aim of the present study was to evaluate some of these new fingerprint detection reagents for drug detection
on TLC plates. The development of new, more intense, or fluorescent colors for various drugs would increase the
overall specificity and sensitivity of drug-screening

Volume 1
Numbers 1-2 Posted On-Line At:
January - June 2003

Evaluation of Ninhydrin Analogues and Otheir Electron-Deficient Compounds as Spray Reagents for Drugs on Thin Layer Chromatograms

I am particularly interested in compound H (4-chloro-7-nitro-2,1,3- benzoxadiazole) as it possesses interesting fluorescence properties when bound to amino acids (and yes, with some amines as well!!) I'm hope to look into this at depth at a later date (along with some other very hopeful candidates).


As for standards being required to identify drugs, there are some ways around this, particularly when levels are in concentrations typically found in pills. Briefly, another similar chemical is used as a reference. To confirm a positive, the separated sample is run again with the reference and retention variations should be spot on. A couple of these non-active amines are sold by Aldrich for this purpose.

 

[cuddlefish]

Thank you Phase Dancer, I'm very impressed with all that. I don't usually have enough time to go through the journals but that article is very interesting. I didn't know that microgram was easily available to anyone except those practicing forensics.

I noticed in the methods that a dioxane, xylene, ethanol, ammonia solvent combination was used to elute heroin, cocaine, MDMA, diazepam, and flunitrazepam which all have rather different structures. Do you think that it would be possible to separate compounds so similar as those of MDMA, MDA, PMA and methamphetamine? Especially MDA and PMA as the only difference between these compounds is the methylene-dioxy and para-methoxy groups.

One more question-You mentioned the levels and concentrations found in pills. As the pill are illegitimatly manufactured the synthesis of MDMA would surely leave a significant amount of the precursors (safrole,isosafrole, MD-P2P) would it be likey to elute and separate these as extra spots on the plate? And another thought would any extraction/elution action on a pill elute compound from the binding material of the pill? I'm sure you can see the problems this could yield for a quick TLC separation on a pill as you could get spots not only from any active chemicals but add to that spots from precursors and bindings, and that could be one hell of a confusing plate.

I don't suppose Aldrich would make it very easy to get a handle on said reference chems?

Cuddles

 

[johnboy]

There are things called Thin Layer Chromatography (TLC) plates which someone was looking into a while (maybe 18-24 months) back. I'm not sure what the results were, or if they never arrived.

That was me, and I tried them and it was all too freakin' hard for a non-scientist like me. I'm sure if you ahve proactice, and some idea what you are doing, it would be vauable, but I had neither

 

[BigTrancer]

Cool JB thanks for the update.

BT

 

[phase_dancer]

cuddlefish: In many ways I feel TLC is limited in these applications. Not only - as JB points out - is the process fiddly, but separating structurally similar compounds and identifying them as separate chemicals is asking a lot IMO. I'm sure almost anything can be done with TLC or advanced variations, but without specific colour indicators it is difficult to to tell anything more than whether a pure single compound is present or not. I confess my experience with TLC has been limited to undergrad lab pracs, but we did some interesting things with it none the less. The problem as I see it lies with the almost endless combinations of thing which can be put into tablets/powder.

How does one test for everything?

The simple answer is that you can't. Things are made more difficult when that to be separated is a mixture of closely resembling compounds which elute at similar rates in various mediums. Elution chromatography has been suggested, but it too would require specific mediums for different mixtures and may not provide adequate purity. HPLC or GC are the next steps up, but even those find certain amphetamine combinations difficult to separate.

I believe if we can come up with specific markers for each chemical -in other words a reagent/immunolabel etc that indicates ONLY that chemical, and use that with TLC, then a great user kit could be developed using an automated form of 2D TLC. The tests still wouldn't tell you what odd or new additives may be in the pill, but it would tell you if something is there which is out of place.

The latest personal area of interest involves using fluorescent markers and a tunable UV detector. Violet & UV lasers (freq doubling a diode output) are becoming cheaper and semiconductor and nanotech grating makes affordable units a promising reality. If we work on fluorescence I believe an affordable and user friendly device could be just around the corner. MD and MeO groups are easily discernable using UV spec, so the same should apply with this approach.

As for contaminants, yes they would definitely be present. As for their effect on TLC; I would suggest that they would/should only be present in small amounts which may not great affect migration, but without doing the tests who would really know?

Regarding reference chems, I see no reason why they would not be available. The reagent H mentioned above is commercially available and is also used in fingerprint analysis. However, working with illegal/illicit chems of course is naughty, so you would need to be able to substitute some drugs for legal non-active compounds of similar structure. This is not necessarily as difficult as it seems, and needs to be done anyway before a permit is applied for to go further. I could suggest a few ways to go about this - PM if you like.

In the meantime here are a couple of papers (courtesy of Rhodium) you may find interesting.


Optimization of extraction parameters for the chemical profiling of MDMA tablets

Spectrophotometric and Liquid Chromatographic Identification of 3,4-Methylenedioxyphenylisopropylamine and Its N-Methyl and N-Ethyl Homologs