Home TLC Thin layer chromatography

[sadpanda]

so Joe just my FINAL QUESTION i promise!!! (THANKYOU for your time and generosity answering all my previous questions, which has allowed me to get to a position where sometime later this week I can start doing experiments and sharing results, im getting excited!)

For convenience sake is it ok if i add the NaOH in with the Fast Blue bath, or do they need to be two separate baths?

 

[G.O. Joe]

That's part of the directions to follow, see visualization. The basic directions for TLC, GC, and HPLC are in the UN guide in full - hence the title Recommended methods for the identification and analysis of cannabis and cannabis products. If your tap water is as alkaline as mine, you may find that sufficient - but try it according to the directions first except maybe doing the dipping and the BB salt.

 

[sadpanda]

ok cool yes the UN visualization doc recommends two methods, one with methanol, but the one i'll use is "50mg (Fast Blue B) in 20ml of NaOH (0.1 N)" - although i'll have to figure out myself if 50mg B == 50mg BB (not sure how exactly though! so i'll just have to experiment with different levels). And it seems "0.1M" and "0.1N" are the same, at least in regards to sodium hydroxide solution... that is, a ratio of 4.0 grams to 1.0 litres water (i'll obviously be cutting that down quite a lot as i only need ~20ml for the bath ... hope it's not too much hassle crushing the NaOH pellets!). It's good to know tap water is generally fine to use but I figured i'd play it safe by using distilled water as it's cheap and i figured the less unwanted crap/molecules getting in the way the better, especially for somebody like me lol. (I'm guessing there wouldn't be much of a problem anyway due to the specificity of Fast Blue, but i might as well play it safe!) Anyway it's Monday now at last and I'm expecting most of the few remaining parts to arrive this week! I do have the main parts so I could start TLCing now, but until i get some glass measuring syringes (ebay says arrival 12th-16th ie this week) to know exact quantities, plus lightproof freezer container for Fast Blue storage, working with it will just be a waste as i'll get no solid data. Patience will be a virtue!

Planned experiments:
1) with and without NaOH.
2) varying ratios of hexane:diethyl-ether, to help confirm the suggested 4:1 as best.
3) varying ratios of Fast Blue to determine optimal.
4) non-cannabis tests:
4a) I have a strawberry which i managed to not eat, now frozen: in 2014 (using a technique they say was developed in 2011) scientists determined, thanks to Fast Blue, that strawberries have a lot more phenolics than previously thought so itll just make for an interesting organic non-cannabis test. Just don't ask me what any of the resulting dye spots are... "hey check out all the THC in this strawberry!".
4b) Also I think it'd be cool to try some random leaf from a non-cannabis plant, a tree leaf or something.
4c) piss test!? You know you want me to. I'm pretty sure i don't want me to. Fast Blue is used to detect the THC metabolite THC-COOH in urine. THC-COOH = 11-nor-delta(9) tetrahydrocannabinol-9-carboxylic acid.
5) leaving the developed plates out for various number of days for scans to measure the deterioration of the dye (my current understanding is it fades 'in a few weeks', which is why i was hoping to avoid NaOH if it was only being used as a preserver, but clearly the reading states alkalinity is important with Fast Blue). Will try with and without NaOH.
6) figure out best temperature and duration for on-plate decarboxylation in oven

All these experiments to try, at my personal expense, because despite RTFM'ing for a full month there's still nothing concrete, with so many differing measurements. So that's where we are ... this amazing and relatively inexpensive technique to determine THC:CBD ratios, yet hardly anybody is doing it because there's just not enough specific information available! hopefully we can change that and open some doors. And yes, it surprises me that most breeders don't seem to use it.

anyway i can do all these tests because it's easy to cut the alfoil TLC plates for individual runs - we don't need the more-expensive glass plates for this (and i dont have anything to cut glass ones with either), another money-saving and convenience-improving tip thankyou again

Watch this space! The band at Experiment City are already doing sound checks and should be jamming by the weekend

 

[sadpanda]

actually rather than trying to make 25ml of sodium-hydroxide:water 0.1N solution each time, it seems easiest (and more accurate and consistent) to simply make 1L worth and then keep that stored aside for easy use whenever its needed.

The SDS just suggests storage as "Keep container tightly closed in a dry and well-ventilated place" and is "Stable under recommended storage conditions", and Sigma sells 0.1N premade solutions, so pre-making 1L and storing it aside seems fine

http://www.instructables.com/answers/Soduim-hydroxide-container/

Sodium hydroxide should NOT be stored in glass. Do not use polyethylene terephthalate (aka PET or PETE) type plastic either which is probably the most common household plastic (water bottles, etc).

I use high-density polyethylene (aka HDPE) or polypropylene (aka PP) which are somewhat common around the house. Other plastics work too but you will have to look at its chemical resistance chart, not all plastics will work. The sodium hydroxide I have came in a #2 HDPE bottle.

Certain metal containers are OK but I tend to avoid metal because there are so many different compositions and chemicals that can't be stored in or even near any kind of metal container. As mentioned, use an airtight container because it is extremely hygroscopic.

and http://www.ineos.com/globalassets/i...ents/ineos-hdpe-chemical-resistance-guide.pdf (chemical resistance chart for HDPE) for sodium hydroxide concentrate gives it the maximum "Satisfactory" rating (the only other ratings are Some Attack, and Unsatisfactory).

Im pretty sure cola etc drinks come in clear PET bottles in which case we can't use those, but i just checked in my fridge and ive got a 2L iced coffee flavored milk, as well as orange juice, and checking the base they both say HDPE, with a 2 in the recycling logo, so it sounds perfect no need to buy off ebay, ill just rinse it out with tap water and then a little bit of distilled water and then a little bit of high quality isopropyl alcohol that ive already got nice and easy to pour too with handle

 

[sadpanda]

Something else i've seen a lot on TLC plate images is a CURVE on the outer runs, resulting in a "° o o °" kinda shape rather than the expected flat "o o o o", yet they still have the same Origin, so the Rf values are seemingly affected too:

It seems this is because of using jars that don't have flat bottoms. Not only do jars without flat bottoms result in that above effect, it also means you need more eluent than you would if the base was flat, and that also means having to use a bit more eluent (the eluent wont even be touching the base of the plate until its ~3mm deep), although the latter isn't so much of a problem as i think we can retain and reuse the leftover.

But anyway i just invested $2 in a 50pcs bag of 4mm (smallest, probably best) stainless steel ball bearings, hopefully sitting a ring or two worth of those around the outer base will result in less eluent + a more desirable flatter result (but hopefully without a 'wavey' effect from the balls). Interesting experiment anyway ... and of course I will keep my eye out for elusive flat bottom jars (have to be the right dimensions too of course), but hey... $2.

 

[Chimera]

Go to Lowes's or Home Depot and find a large glass brick used in construction. Take it to your local glass cutting shop and have them carefully cut the top off of it- voila a perfect developing vessel. I saw them in use at Arno Hazekamp's lab in Leiden, a nice little money saver tip. Shopping via science catalogs is worse than buying at grow stores, they think everyone is spending grant money so prices are always high high high!

 

[sadpanda]

a 'glass brick' so one of these?

and then take it somewhere and pay them to cut it for me? i dunno that sounds a lot more effort and expense than just buying a flat bottom jar or using a filler like ball bearings, and no airtight seal either

 

[sadpanda]

Another experiment i'll have to do is try to figure out whats best for on-plate oven decarboxylation of 2µl extract dots ... (takes less time and won't stink up the place like decarbing buds does)

Montana Biotech - "Heat plates in a 300°F/149°C preheated oven for ~3.5 minutes. Remove from heat and let cool."

Alpha Cat - "Heat the plate for 40 seconds in the oven at 100°C/212°F"

Alchimia - "we place the plates in a Pyrex tray which we put in the oven at 120-140°C/248-284°F during 5-10 minutes."

Journal of Chromatography 1990 tested the "Effect of heating time and temperature on the THC content of an n-hexane marihuana extract after heating on the glass surface in an open reactor", with a graph demonstrating 252°F/122°C for 27 minutes is optimal (but doesn't say it was on plates, and doesn't give quantity)

Cannalytics - "Heat the spots by holding the tip of the smallest possible flame of a cigarette lighter on the glass side of the TLC plate for 10 seconds. Hold the plate at a 30 degree angle to avoid blackening."

I want to over-decarb at least once anyway to hopefully see the THC spot degrading into the CBN spot

Also some interesting notes about it in this patent https://www.google.com/patents/US20070077660

In an embodiment, the sample may be placed on a little piece of aluminum foil or on a small dish or in a crucible, etc. Then the sample can be heated in e.g. an oven for about 2-8 minutes, preferably about 3-6 minutes at about 80-200° C., preferably about 120-180° C. By trial and error (i.e. varying heating temperature, heating time; and developing the TLC plate 11 and checking the result), the appropriate conditions can be found. If e.g. different reddish spots appears between the CBN spot and the natural smear of acids on the fingerprint (the not heated version) the temperature was too high and/or the heating period too long. Heating for about 4 minutes at about 150° C. provided good results.

Alternatively, in a specific embodiment, the method of the invention further includes:
spotting on the TLC plate an amount of the extraction liquid with the extracted cannabinoid(s), followed by a heating of the plate.

Referring to FIG. 5, after providing one or more spots 12 to the plate 11, as described above, TLC plate 11 can be heated. The same heating conditions as described above can be used, e.g. 4 minutes in an oven at about 150° C. Alternatively, the TLC plate 11 may be carefully heated with a candle or lighter 21. To this end, the plate 11 may be seized between thumb and index-finger or with a tweezers, etc, preferably such that only the edges of the plate 11 are touched. The spots 12 are carefully heated, about 2.5-4 cm (h4) over the flame of candle or lighter 21 for about 1-3 minutes, at both sides of the plate 11 (silica side/glass side). Also for this method applies that some trial and error may be necessary to find the right heating conditions. Good results were obtained with a distance h4 of about 3 cm and a heating time of 1½ minutes. Heating is usually performed either directly after weighing the cannabis sample 2 or after spotting the extraction liquid with the extracted cannabinoid(s) 7 on the TLC plate 11. Hence, for determining the cannabinoid content in the cannabis according to the method of the invention, until development, only one of the optional heating steps as mentioned above is chosen.

 

[G.O. Joe]

Silica is acidic. This is a much less popular graph from the same report. Aluminas are more reactive. Anyone in the cannabis analysis business who doesn't know this, maybe they're not a better source of information than forensic chemists.

 

[sadpanda]

awesome pic Joe, thankyou!!! Im quite perplexed by it though! ...

As soon as the material gets heated all its Delta9 THC starts to degrade!? rather than increase!? its just Delta8 increasing?? its almost as if it's saying "if you want delta-9 THC don't decarboxylate at all, not even for a second" !?

and i thought the THC would degrade into CBN, yet the CBN level constantly decreases???!?

now i'm even more confused/not sure how many mins to go for... is the chart suggesting 10 mins @ 145C is best for highest THC? but... highest D8 THC, lowest D9 THC!?

 

[G.O. Joe]

The graph is showing the progress of isomerization of THC over time by the silica. At 106°C the same effect was found - that's the first line after the graph. This likely happens in some GC columns during the run and it's a bit questionable whether the d8 isomer is produced at all by any strain.

 

[sadpanda]

what would your recommendation be for temperature and duration in fan-forced oven for a 2uL droplet on silica plate? for highest THC/CBD (best general purpose decarboxylation i guess)

ps. just waiting on NaOH to arrive and a glass measuring beaker, all should be here by mid next week at the latest, then i begin!!!
Also I found a better jar at the grocery store - just look at all the pickled/preserves/jams/etc, there's heaps obviously, and some have reasonably flat bases

 

[sadpanda]

AND SO MY THIN LAYER CHROMATOGRAPHY JOURNEY BEGINS!!! ...

Well, after 6 fricken weeks of googling, scratching my head at conflicting information, reverse engineering commercial narc/canna detection kits by their documentation, trying to get answers out of G.O. Joe, figuring out what parts are needed, annoying G.O. Joe (but i can't thank you enough!), and waiting for free but slow shipping from Asia (i'm trying to do this on a budget!), everything has come together and today I received the last components I was waiting for including safety glasses -- because safe science is fun science!™

Ive spent the last 4 hours doing my first run, lol... didn't want to rush it, plus i had to decarboxylate etc. Not sure how long i'll be able to get it down to, probably around an hour maybe half.

My first experiment: determine how much Fast Blue BB is needed!
An absolute must before any other experiments. Being the first run I wanted to ensure a good strong positive reaction, so I got together a 1) few different bud samples, 2) decarb'd tiny amounts @ 120C for 5min, 20min, 35min, 3) some Bluebird Botanical CBD oil, 4) some local CBD oil, 5) some decarb'd CBD oil, and 6) some high THC RSO... and mixed them all together into one! Hedged bets pretty well there. This filled the 1.5ml eppendorf tube about a third full, and i topped it up to about two thirds full with hexane, the solvent im using for extraction. Left to do its thing for at least half an hour or so while i was setting up other things, but shook it vigorously quite often.

Also I wasnt interested in things like "how much THC or CBD is there", i just wanted to determine 1 thing from this first run: how much Fast Blue BB is needed.

So I decided to turn the 10cm x 5cm TLC plate from usual portrait to landscape - i'd get TLC runs half as short, but twice as many of them... then I could cut them up and dunk each one individually in the dye bath, adjusting the strength as I go.

It actually only took around 4 minutes for the solvent (4:1 hexane:diethyl ether) front to reach within a millimeter of the top. So now boys and girls we have standard "leaf chromatography" where we're separating chlorophyll etc - for some reason my science teacher never taught me this! ripped off.

Here's a scan a few minutes after the TLC run. My scanner is decent at 1200dpi and i then reduce to 800px wide in Photoshop.

So as you can see instead of the usual 5x runs up the 10cm plate ive opted for 9x 5cm runs.

We can see chlorophyll because well, im guessing because it is a pigment afterall! no dye needed for that.

But where are the cannabinoids!?

Enter Fast Blue BB... (zomg omg omggggg)

So I cut the above TLC plate up into 9 pieces 5cm tall each (aluminium plates are awesome, don't even need tinsnips - just regular scissors! and cheaper than glass)

I quickly decided that weighing the Fast Blue was going to be tricky, so instead I just opted for "how many of these 3mg measuring scoops will i need?"

So I made up my 1.0N NaOH solution (4gms sodium hydroxide pellets in 1L distilled water), and had previously (just using water) determined that 25mls was ideal for my bath.

To that 25mls i then added 1 flat scoop of Fast Blue BB salt powder (which is orangey brown), and the bath quickly went a mild urine yellow color, lol... whoa, ok so this stuff is pretty sensitive it seems.

Anyway I dunked the first of the nine TLC cut plates in, gave it a couple seconds and removed to dry. To my pleasant surprise, in just a couple seconds i was seeing those glorious cannabinoid colors shine through -- not overly strong, but a lot stronger than i thought ... hey, just a tiny '3mg' scoops worth. Because i was already getting a good reaction I decided to put two more in at this level.

I then tried with 2, 3, and 4 scoops worth. At 4 scoops the solution was really getting quite dirty, because i was only stirring it in and its a bit clumpy, so next time i'll just be putting it in a jar for a good quick shake first, simple.

So anyway it seems 2-3 scoops of Fast Blue BB is ideal
So that's probably about 8-15mgs Fast Blue BB to 25mls solution, which would easily do at least 2 full 10cm x 5cm plates.

The plates after a two second bath with Fast Blue BB:

WHOA!!! Strong response! Well they did say Fast Blue BB was more vibrant than Fast Blue B! check out those beautiful ORANGES and REDS and SCARLETS and VIOLETS!!! looks like there's a lot of CBD in there from all that orange hehe (well i did put quite a bit in!)

So again, DISREGARD THE "DIRTYNESS" of the plates - i just need to shake the Fast Blue in a jar next time, rather than stirring (its a bit too clumpy). Also ignore the poor separation - i didnt filter the extract first, I probably made dots larger than intended (just getting used to the micropippette device), and i only used tiny 5cm runs, as again I only had one thing i wanted to determine: quantity of Fast Blue required. (Success)

Time for a short break now, some iced coffee and relax with a couple bowls of the good stuff (after momentarily putting away all the flammable solvents lol). Then time for more experiments, and actual PROPER FULL 10CM RUNS!

Just a special THANKYOU once again to everyone who's helped me so far, you're the reason my first experiment was a success

 

[sadpanda]

2nd run - testing various substances (cos it can!)
Now that i know how much Fast Blue to use i can start experimenting with proper 10cm runs!

So it turns out just shaking it harder in a jar doesn't make some of the clumpy bits dissolve ... I wonder if it's because im trying to dissolve them in NaOH 0.1N solution - perhaps dissolving in water first is best? might have to try.

Anyway I tried 8 different substances ... 3 of them returned completely clear readings (probably correctly). It's important to note however that I haven't found my feet yet when it comes to how much solvent to use, and I used quite a bit with these ones (nearly full 1.5ml eppendorf) so perhaps theyre a little more dilute than they should be. Each of these 8 tests have different ratios too! And also remember im a complete newb at this just starting

Here's how my 2nd run turned out:

Note1: yes still a bit 'dirty' from undissolved clumpy Fast Blue, ill filter it next time if i cant get a full dissolve.
Note2: because i thought the 1-4 plate which i did first wasnt as bright as it should be i added one more scoop of Fast Blue before dunking the second 5-8 plate.
Note3: the darker areas eg between 2 and 3 are just liquid touching up against the flatbed scanner glass (plates not fully dried). I use cards on the side to ensure the TLC plates aren't squashed down.

LANES:
1. CBD oil from Bluebird Botanical, this has been in my fridge over 1yr now too but still strong CBD reading, and with just a light hint of THC.
2. [CLEAR] Urine, because science. Google assures me that Fast Blue is used to detect the metabolite THC-COOH in urine, but somehow i managed to pass the test - i shouldve failed. I dont know or really care much about this test though, im guessing to do it properly requires another step or two i didnt do, but I still find it interesting nonetheless, to help get an idea about what is and isnt detected.
3. [CLEAR] Strawberry fruit, a small chunk from one anyway, with science dictating i eat the rest to ensure <LD50. I know that Fast Blue has been used to analyse phenolics in strawberries, but my TLC came up clear. Perhaps the fruit is meant to be dried, crushed, decarboxylated etc first i dont know, but a fresh piece of fruit shaken about in hexane for half an hour yielded a clear TLC.
4. [CLEAR] Hair. I pulled about 10 hairs out of my head to ensure i got some roots and stuck them in hexane for half an hour. Again i passed this test, but shouldnt have, but probably didn't do it properly.

5. The 'multimix' is the exact same extract i used in the first run when i was determining Fast Blue quantity. It's a mixture of half a dozen different things including CBD oils and freshly oven-decarboxylated buds over three time scales.
6. A full-spectrum CBD paste, glorious! check out that bright YELLOW cannabinoid! and then the dark red one above that! then just a hint of THC, being CBD paste afterall! then LOADS of CBD! and a few other cryptic cannabinoids above it! beautiful spectrum
7. chewy caramel brownies -- the solvent really struggled to make any progress into the bit i cut off, so the solvent was still very clear when i tested, so the small THC dot isn't being given much of a fair chance!
8. some RSO of strain Chronic in a syringe. No CBD in this baby, just a THC bomb!

 

[sadpanda]

in regards to the Fast Blue powder clumping, perhaps i just need to mix it in with a tiny amount of water first instead of big amount ... i thought 25ml already was a tiny amount though, lol

https://www.quora.com/Why-does-coco...a-little-bit-of-milk-but-not-in-a-lot-of-milk

When the cocoa granules get wet, the starches expand and intertwine with one another. Given enough fluid, some clump will get wet around its entire surface. Because the starches are fully hydrated, and the fat in the cocoa is hydrophobic, water doesn't pass through them, and the inside remains dry.

You can try stirring it, but when there's enough fluid, the capsules will tend to be pushed around rather than burst. It's like trying to pick up a marble with a chopstick: too many degrees of freedom.

When there's only a little bit of fluid, any capsules that form are easier to burst because they're sitting on powder rather than liquid. They don't move as readily and it's easier for your spoon or whisk to push through them rather than pushing them aside.

You can get similar effects with other powders, especially when they're hot and gelatinize the outside. Both flour and cornstarch are dissolved in small amounts of cold water before being added to large amounts of liquid. Once the granules are fully separated by the liquid, there's less risk that they'll end up linking with each other to form the skin.

I could easily use some filter paper but id rather try for max dissolving first. I think what'll probably end up happening is i'll just use the end of the scoop to squash the clumps, that breaks them apart and most of their interior is able to dissolve then... still end up with a few specks of dirt but at least getting most of it to good use. Then filter.

[update] i read this - "The chromogens Fast Red TR and Fast Blue BB produce a bright red or blue end product, respectively. Both are soluble in alcoholic and other organic solvents, so aqueous mounting media must be used."
So maybe i just need a few drops of hexane. Another fricken experiment to try, lol

 

[BigWillie]

Wow, I wish this thread was on column chromatography. I have been, like sadpanda on TLC, working on the details of column distillation.

I did some of the TLC as described minus the Fast Blue BB, mainly 'cause I don't have any, and got colors at various locations which changed with my ratio of the solvents. But I can't find anything on identifying the constituents without any color identification help. LOL. I guess that's why there is Fast Blue BB, eh?

 

[sadpanda]

Wow, I wish this thread was on column chromatography. I have been, like sadpanda on TLC, working on the details of column distillation.

lol those damned DETAILS!!! there's a lot of information to wade through, and a lot of it conflicting lol ... it makes for a very frustrating process just trying to get set up! i wish you good luck with your column chromatography!!!

I did some of the TLC as described minus the Fast Blue BB, mainly 'cause I don't have any, and got colors at various locations which changed with my ratio of the solvents. But I can't find anything on identifying the constituents without any color identification help. LOL. I guess that's why there is Fast Blue BB, eh?

well i guess if you do TLC with cannabis but no Fast Blue you basically just have a chlorophyll test like they do in elementary schools (see my first TLC scan in previous post where all you can see are the chlorophyll dots), lol

But ... just add a tiny sprinkle of Fast Blue to your TLC plate bath and its HELLLLOOO CANNABINOIDS!!! an extraordinary chemical

ps. I also have a small bottle of CHLOROFORM to try as the eluent!! Or will - should be here tomorrow or Friday. Will be interesting to see how it works out vs my current eluent of 4:1 hexane:diethyl ether, which is already producing excellent cannabinoid separation. (Perhaps it wont be as nasally offensive, omg that diethyl ether is terrible!). Here's a good example of using chloroform as the eluent, producing a good cannabinoid separation also, so i look forward to doing a side-by-side comparison of both by the weekend

 

[sadpanda]

btw one thing that immediately became apparent to me was that no way is anyone gonna be doing this at home accurately enough to do HPTLC-style quantitative analysis (which is what the commercial kits advertise as - "find out exactly how much THC%!"). There's just too many variables, and with our cheap little disposable universal plastic micropipette tips, ensuring the exact same quantity of Fast Blue salt each time, ensuring 1uL is always exactly 1uL, and other variables like how much solvent was used, and for how long, how hot/humid it is, how dry the bud was, how decarb'd it was etc etc, it just seems pointless. Apparently there's a ~10% error margin anyway, so im not sure why anyone would bother.

BUT ... for what it is - a qualitative test, wow ... my friend going through cancer but responding really well to CBD oil will now be able to test CBD oils/pastes to ensure they're good and not a scam, as well as find out which plants theyre growing have CBD, which have THC, and a pretty accurate representation of the CBD:THC ratio. This is why i came to TLC, because well... short of sending samples to a lab, what other option is there!? (i guess there's also Beam's test with potassium hydroxide which seems like a cool test for CBD, but nowhere near as useful/helpful as TLC)

NO LONGER BLIND when it comes to finding CBD thanks to these cool Fast Blue glasses
Night #2 of experiments starts in a few hours! just have to enqueue some mp3s...

 

[BigWillie]

Bought 25mg -$140 Fast Blue BB Salt

We will see, with the help of this thread.

I need to identify the fractions in bulk as the come down the column. I will continue to define the process.

 

[sadpanda]

only 25mg? i see you didn't shop around! I'd cancel that order if it's not too late. That's about all you get in those commercial test kits and they're about the same price! That amount is only good for about two baths or ~3-10 plates. Should be able to get at least 250mg for $100 - I got 5000mgs for just a few hundred and that was imported from a Japanese lab and freighted with dry ice. Take some time and shop around with google also email some local labs and ask them how much to get some for you (worked for me). Dont forget Fast Blue B is the other option to BB, its usually a bit cheaper too and only requires refrigeration not freezer, but its more of a carcinogen concern and results not as vibrant (please read this thread in full if you havent yet)