Harmine, harmaline and THH from Syrian Rue. Verification and finetuning of the VDS-protocols

[pitubo]

You mustn't heat a solution of sodium bicarbonate.

I was thinking of heating water to say 80 deg C and poring the sodBIcarb in, would it also start to transform in sodcarb on that modest level of heating? I would it that to minimize the needed volume of basing liquid.

Nice grafix, Jees. In theory, practice and theory are alike. In practice however...

I'm afraid that at 80 degrees significant decomposition is happening already. You'll have to try what works, I don't have hard numbers - just a recollection of fizzing and bubbles when bicarbonate hits hot water of unspecified temperature.

 

[Jees]

...You'll have to try what works,..

So I did by heating water and a temp probe in it and a chunk of sodbicarb, and from 40 deg C there was the very first signs of few and tiny CO2 escaping but majority of the chunk stayed calm. At 50 deg C the CO2 escaping was undeniable happening while there was still the majority not converting. I tend to believe we can use the CO2 escaping bubbles as an ideal parameter for the conversion of sodbicarb to sodcarb.

I realize that by heating water, once nearing 40 deg C involves the water oxygen to be released and also forming bubbles. But there was a difference noticeable between those bubbles coming from that or those coming from the sodbicarb --> CO2 + sodcarb.

I guess that graph-picture is theoretical, in practical terms you simply cannot heat water and poor sodbicarb in to make a more concentrated solution of it, unchanged.

Classified 8)

PS: I thought to skip the oven trick and get immediately concentrated sodcarb water by boiling sodbicarb. Was worth a try though, but not good enough.
This is VDS' recipe:

A batch of sodium carbonate was synthesized by heating sodium bicarbonate at 200°C for 2 hours.

* * *

DreaMTripper thank you for the pdf's, big ones, will dig into them later and I hope to understand to content.
:thumb_up:

 

[An1cca]

Concerning the use of sodium bicarbonate: as you can see in pic1 of my previous post, the precipitate from the ammonia-only basification is clearly more coloured than those where bicarbonate was used first. I think this is because the first precipitates take with them a lot of impurities. This same thing I noticed in the different fractions while pH-separating harmine from harmaline. Impurities get caught by precipitation. I think VDS wrote about this as well in the article.

So, the use of sodium bicarbonate might be helpful in pre-purifying the THH mother liquor. This step does not seem to lead to lower yields, however it takes more work.

Next time, I think I'll try the old activated charcoal way of purifying the DHH-solution before reducing.

Jees, the pictures I made were at magnification 150-640x.

As for the Manske: I have some beautiful clusters growing, will post soon!

 

[Jees]

An1cca for president! :roll:

...Hypothesis at hand: the moment you are sure that some undissolved layer of sodbicarb forms at the bottom, then you can also be sure that all ZnCO3 has precipitated. An1cca's test was indicating OK, but I would like to redo this test to re check.

I came at an attempt to this finally.

But I used sodcarb and stopped at pH 7.55 to simulate a pH level of sodbicarb. I would really weight the ZnXX precipitates only and not undissolved sodbicarb as this seems to happen at 7.55. So sodcarb at 7.55 would do that simulation as a completely dissolved base.

First pic is the mist at pH 9.2 after that the precipitates were already filtered off at 7.55, so it's the precipitates formed between 7.55 and 9.2, being 37% of total.

Second pic, the filter above is the 63% weighting precipitates that were filtered of at 7.55.
Filter below the 37% of total weight precipitates between 7.55 and 9.2

I realize that sodbicard is not sodcarb, and therefore this test is only indicative to an extend. Since the above post mentioning, the first precips catching impurities is another motivation to separate before reducing.

Love adding pieces to the puzzle

 

[ganesh]

So i presume that you have now mastered a method to seperate Harmine, Harmaline, and THH?

Are you sure it is so? Is it GC-MS TESTED?

If so, will you be making a tek?

 

[An1cca]

Nice work Jees, well I have a new batch of DHH that is being reduced at the moment and will precipitate with a definite excess of bicarbonate this time. However, as can be derived from the weights of the aforementioned precipitates, there appears to be no remaining carbonate/bicarbonate in the precipitates when compared to the ammonia-only route. This finding is confirmed by the fact that no fizzing was observed when the 2034mg of crude THH were dissolved in vinegar for the Manske.

ANOTHER IMPORTANT FINDING is that microscopically, I could confirm the completeness of separation of harmine and harmaline by the VDS pH-metric method. I already had a good feeling about this when the titration curve matched the one in the article, and now I could check for myself that no 'leaves' can be observed in recrystallized harmine fractions and that no 'needles' can be observed in the harmaline fraction. Even without GC-MS, I feel convinced we're dealing with two separate compounds now. To see is to believe :lol: I attached some relevant images.

As you can see, there are quite some impurities showing up in the recrystallized harmine slide. It was the first fraction of a batch and therefore contains the most impurities.

 

[Jees]

Nice pics!

Your post #87 proved not needing more overkill on sodbicarb, and perhaps proof that sodbicarb takes the ZnXX down sooner (pH wise) / better than my attempts with sodcarb. I will confirm that soon. It's nice to simply know what works best, and have it checked.
Kind of exiting to see if sodcarb at 7.5 can equal or fails in mimicking sodbicarb.

About separating harmine-harmaline pH wise:
- did you use ammonia as suggested by VDS?
- what did you do with the 2 % mixed fraction (VDS's fractions 11 & 12)?

Thank you so much, you're rocking with this

 

[Jees]

Experiment done:

I confirm everything An1cca said in post #87: I based til 7.56 with sodbicarb, ran liquid 3 times trough same coffee filter (first run had still a slight mist but 3rd run was clear as water), then based further with saturated sodcarb water till 9.3 and it stayed water clear.

* So there is no need to base further than 7.6 with the sodbicarb, there's a slight layer undissolved sodbicarb at the bottom. By then ALL ZnXX will be catched in the filter. Since THH will start it's earliest precipitation at 8.8 (post #97) we can be sure that no THH gets prematurely lost by the filtering off at 7.5 - 7.6

* Except for confirming what was done already, the new element here is that sodcarb cannot do what sodbicarb does at pH 7.5. The pH at which both start to precipitate the ZnXX is the same like 5.8. Both sodcarb and sodbicarb are fully dissolved at this point. Only sodbicarb has the job done at 7.5 while sodcarb needs a higher pH to precipitate all ZnXX. Tadaa.

The reason for this could possibly be that sod-bicarb becomes partly undissolved at 7.5 (a little layer at the bottom) and this hardware might act as a kind of promoter for the ZnXX crystals to grow under lower pH (lower pH than compared when everything is liquid such as in the case of sodcarb).

So both An1cca and me have confirmed:

...Hypothesis at hand: the moment you are sure that some undissolved layer of sodbicarb forms at the bottom, then you can also be sure that all ZnCO3 has precipitated. An1cca's test was indicating OK, but I would like to redo this test to re check.

We could now discuss how thick that layer must be instead of targeting a particular pH, just for the sake of a low tech visual reference as being the easiest way of determining "enough". We could make for instance a 10% rule: if give or take 10% of the liquid level is undissolved sodbicarb then you're set. As a matter of fact I gonna give that a try and see what pH that delivers. To be continued...

 

[Jees]

As a side note: this knowledge suits the reduction of deems n-oxide also.
The ZnXX could be taken out at 7.5 with sodbicarb as mentioned, then base further and see the deems precipitate start at pH of 9. The advantage of having the ZnXX out is legio if you have done this already, the separation of the pulls is better without the ZnXX flocculants that used to make fun right on the interface level between base phase and pull-solvent. There is no more annoying place for them to reside than there, and guess what they do?


Next time get them out before the deems precipitating.

 

[Jees]

...If so, will you be making a tek?

For now it's still searching for confirmation of the paper and the best hands on routine, it might end as a tek. But with the OP paper and the hints here you could get along already.
The last pages protocols of the OP paper say a lot but they lack some details here and there, which is what we try to nail. But this you probably know.

 

[Jees]

I think the visual 10% rule works fine. More than 10% doesn't hurt but it's not needed at all and wasteful on resources.
Filtered the un-reacted Zn off;
(added some left over dry ZnOH/ZnCO3 that I had, to make the test "extra", it dissolved.)
Start adding sodbicarb (careful at start, then it fizzles like crazy), as soon as one starts to get milky liquid then after each sodbicarb add, let is settle for 60 seconds, this allows the undissolved sodbicarb to settle while the ZnOH/ZnCO3 still floats. Repeat until the undissolved layer is say 10% of the total liquid level, see picture (will add when I'm back home). You can also just monitor the liquid top level that rises: in this case start was 100 ml and on the picture we're at say 110. Remember to swirl with a spoon for at least 30 seconds after each sodbicarb add, I think that is important to do.
(sneakingly I checked Ph which was 7.8, but we suppose to do all this without pH meter too)
Run trough a coffee filter, first time some "milk" trace remains, just run 2 x extra trough same coffee filter (which gets more clogged/finer on each run), the result should be clear.
BTW remember to fold your coffee filter seams. Then I pushed pH up with saturated sodcarb water til 9.8 and it remained clear, proving again all/noticeable ZnOH/ZnCO3 was out.

* * *

This technique of which the principle was laid out by Ar1cca is to get the ZnOH/ZnCO3 out before precipating THH start. If I'm correct the reason is to have precipating THH in a solution that has no floating ZnOH/ZnCO3 in it. If I'm correct again this is to eliminate all chances for entaglement between ZnOH/ZnCO3 while the forming of THH. Please An1cca chime in if I'm mistaken.
We know that excess base does liquify ZnOH/ZnCO3 but this happens at +/- pH 9 (for ammonia), pH 10.60 for sodcarb, and astronomical high pH for lye. So IF there is an entanglement between forming THH (start at pH 8.8 ) and the yet non liquified ZnOH/ZnCO3 at that pH level, then the liquification act at 10 - 10.8 might come too late.
Whether the concern is legit is yet to be seen, but it sure NEVER hurts to get the ZnOH/ZnCO3 out in advance by such a stupid simple sodbicarb act that even doesn't involve pH metery if using the 10% rule.

This all applies to harmaline that was separated from harmine in advance, a route mainly done by An1cca so far.

* * *

I would like to incorporate the 10% rule to reduction result of a mix harmine+harmaline, so without a former separation of the two.
The 10% rule would (hypothesis so far) separate the THH from a mixture of {harmine + (ZnOH/ZnCO3) + undissolved sodbicarb} completely and brain dead simply. What to do with that {...} is for later but here we just focus on the THH especially. After filtering off the {...} then we should (is yet to be proven) just dump sodcarb in the liquid to gain THH solely and quite pure.
If this works, going to check out this, then an ammonia free route is established and even without needing a pH meter at all, because there's no previous separating act between harmine+harmaline. I still believe this separating act needs care to be done properly and the least kitchen level act of all, also needs microscope or TLC to check method. So if we can avoid this altogether it would simplify things.
To be continued...

 

[An1cca]

Jees, I think the 10% rule you describe will be very practial indeed. For me, it was difficult to make the difference between precipitating Zn-salts and excess bicarbonate, but as you say, the bicarbonate falls to the bottom rather quickly, while the zinc salts stay floating for a longer time.

One thing we have to consider with this method is the fact that THH-rich mother liquor can stay behind in the 10% sedimentation at the bottom. How will we 'wash' the bicarbonate without getting the zinc salts? Perhaps we should just add some water (10% ?) to the sediment that remains after the mother liquor was poured through the filter. Then we stir until this water is saturated with bicarbonate as well before we pour it over our filter. Just a suggestion.

Anyways, I will give some more details about the Manske that I did on the THH:

2034mg THH were dissolved in 16ml vinegar 7% + 24ml water. The solution was warmed and a hot solution of 12g NaCl in 40ml water was poured into it (pic1). Soon clouding appeared (pic2) and after a few hours, beautiful THH.HCl crystals emerged from the solution, looking like sea urchins. When this solution was cooled, more clouding appeared that transformed into more crystals (pic4). Pictures 5-9 show some more THH.HCl crystals and pic8 sepcifically shows dark-brown impurities at the bottom. So these could not be removed by Manske-salting . Afterwards, I filtered the THH.HCl, dissolved it in a small amount of water and precipitated with ammonia 12% (pic 13 right sample, 1568mg). Then I added ammonia to the mother liquor and recovered 94mg more of THH.

However, I also have some VERY good news: enter ACTIVATED CHARCOAL ! Before adding zinc, I first boiled a solution of 3012mg harmaline in 100ml vinegar 7% with 60mg activated charcoal for 10 minutes. Then I added 3030mg of zinc powder to the hot solution and let the recipient stand at room temperature for 19 hours. Then, the charcoal and zinc were filtered off. The resulting liquid can be seen in pics 11-12.
Then I just bombed the solution with a large excess (+100ml) ammonia 12%, let the precipitate settle , filtered it and washed it with 50ml of ammonia 3%. (pic 13, left sample).

Just look at the difference:shock: ! It seems activated charcoal is capable of eliminating those nasty plant matters that keep on sticking to every precipitate from rue seed to pure alkaloid. Well, now it seems they can be caught, nice!

Will clean up manske-THH (pic 13, right) also by boiling for 10 min with 2% activated charcoal. I'll post the pics!

And yes, I would very much like to collaborate on a protocol or 'tek', but as Jees very correctly mentioned, we're still in the finetuning-phase, some patience please:wink:

 

[An1cca]

Allright fellow researchers, by now it's getting clear that Van Der Sypt's paper contains a lot of correct and useful information. Up until now, the only thing that was not reproducible is the clean-up of the harmine-fraction with sodium carbonate as stated in his protocol 4. In my opinion, he used ammonia and extrapolated its washing capacities to sodium carbonate. As demonstrated above, the work of Jees and myself has proved him wrong.

However, in all other aspects, his methodology was verified (pH-metric separation of harmine and harmaline, microscopically verified), synthesis and clean-up of THH, ammonia-free route with sodium (bi)carbonate,... I might just send him a card for christmas:grin:

I have some more data on the Manske operation I did on the THH. So I started with 2034mg of THH that after Manske, filtering off THH.HCl, dissolution in 12ml vinegar + some water, basification, filtering, washing with ammonia 3% and drying, resulted in 1662mg of brownish THH.

Conclusion: Just as with harmine and DHH, one can perform a Manske with THH (and by the way do some snorkeling around the sea urchins:d ). The yield is good: 81%. SO FOR ANYONE BEING PUT OFF BY THEORETICAL CADMIUM-CONTAMINATION, this might be a step to consider.
However it IS NOT a good purification step (see pics in previous post). As I said yesterday, I would do a charcoal-purification on the brownish THH from the Manske above. Well, after 10 min of boiling with 30mg activated charcoal, filtering and addition of not too much ammonia, the dried precipipate was more fluffy (as I now sees is typical for reasonably pure THH), off white and weighed 1450mg (pic1). Yield: 87%. This finding is interesting, because the experiment I described yesterday, where I bombed 3g of reduced DHH with a LARGE excess of ammonia had a very poor yield: only 1844mg! I think yields have got a lot to do with the amount of dilution in the mother liquor. So the more concentrated the alkaloid-solutions, the higher the yield after basification. I think this is especially true for THH, having the highest pKa of all three alkaloids. This is something to integrate in a future protocol.

Another thing, on microscopic examination of recrystallized THH, there is NO DHH NOR HARMINE visible (pic2 and pic3). This implies two things: the separation of harmine and DHH was complete (as harmine will not be reduced) and the conversion of DHH to THH was complete as well (as mentioned also in Endlessness' thread about zinc-vinegar reduction). Man I love the feeling when the pieces of the puzzle fall into place .

So, some more THH-experiments: I tried my hands at a sublimation. Very basic, atmospheric pressure, glass water-filled beaker that fits inside a steel mug but does not touch the bottom. Put this in an oven at 210°C for about half an hour. Underside of beker covered with sublimated THH, even some needles can be seen! (pics 4-5). Then, I scraped some of the crystals from the bottom (there was a brownish tinge from decomposition) and put them under the microscope and shone on it with a led-light. Well guys, :shock: crystal orgasm (pic6)! Almost as beautiful as the picture in the paper!

Then I tried a salting experiment. THH is not soluble in water, fumaric acid is poorly soluble in water, but perhaps THH fumarate is very soluble. I tried to test this for myself. I weighed out 500mg of THH and 270mg of fumaric acid. This is the amount needed if it reacts in a monoprotic way. So I put the water and the THH together in a small recipient inside a hot water bath. Then I gradually added fumaric acid. After having added 138mg, the THH was totally dissolved as di-THH fumarate (pic7-9). So now it's proven as well that fumaric acid acts as a diprotic acid with THH. On evaporation, the return of the sea-urchins was observed:d (pics 10-12). However, also some discoloration was seen (pic13). Maybe because I evaporated the fumarate on a steam-bath? Or is this just the same old harmala-curse that again and again turns our purified alkaloids into brown powders? The end result was pretty nice though (pic14).

So, it seems we're coming to the end of the track. All that remains now is to puzzle all these observations together in clear protocols. These will be largely based on VDS's paper and will integrate the additional research done by us Nexians:wink: like the error in protocol 4, the use of activated charcoal, the 10% bicarbonate rule, keeping the amount of mother liquor low, .... Coming up: The "VDS-Nexus protocols for the isolation of harmine, harmaline and tetrahydroharmine from Syrian rue":thumb_up:

 

[An1cca]

While cleaning up, I noticed some flickering at the bottom of the steel sublimation beaker. Upon inspection, I saw that just on top of the unsublimated THH, a fine tapestry of crystals was deposited. What would that look like under the microscope?:shock:

This led me to think that the round THH-clusters that VDS shows and speaks of, are of minor interest. They form in the specific reducing/basification environment he used. When given the chance however, 'pure and free' THH wants to recrystallize as radial clusters of slender crystals. In fact, this is what I noticed when recrystallizing THH on a slide (somthing VDS never did?): the little boulders that form are primordial radial clusters (pics 2&3 of previous post).

I'm getting better at taking pictures with my phone, but just imagine having a dedicated microscope+camera..:want:

 

[Intezam]

hmmmh....there are reports of rue having a small amount of natural THH. Could natural THH crystals look different from the racemic THH ??? Or, are they both racemic (i.e. the round type and the radial type)?

note: we don't understand ----> unsublimated THH :lol:

 

[An1cca]

Intezam, I'm not capable of answering those questions, but in the light of the current procedures, the naturally occurring THH will be lost in the crude extractions from seed. This is mentioned in the paper's discussion and I tend to agree giving the conclusions from my previous post where dilution of the mother liquor causes THH to stay in solution during basification.

Is there a thread somewhere on the Nexus discussing the properties of THH's enantiomers? Endlessness?

 

[Intezam]

Is there a thread somewhere on the Nexus discussing the properties of THH's enantiomers? Endlessness?

Found this thread There is no discussion on crystal shape using microscopy. That would surely be an interesting addition.

edit: :lol: .. silly we...of course it would be lost. Intezam needs to focus moar. Must've been the canna-witch spreading the mind in unwanted retard-ways.:!:

 

[Jees]

...I think this is because the first precipitates take with them a lot of impurities. This same thing I noticed in the different fractions while pH-separating harmine from harmaline. Impurities get caught by precipitation...

... I could confirm the completeness of separation of harmine and harmaline by the VDS pH-metric method. I already had a good feeling about this when the titration curve matched the one in the article...

I said to make a reduction of a harmine+DHH mix but I could not resist making the separation act on behalf of the pH depressions per VDS, and thus not on behalf of pKa values as was advertised in the past years. I've too less time and extracted rue at hand to do both right now. So I separated. The reduction attempt on a mix will have to wait.

While An1cca is storming forward (let's cheer him up) I'll detail a bit more about how that separation went for me. Having seen pH dancing at previous extractions but never bothered about it before, this time the pH dance really becomes the protagonist, thanks to VDS.

My trigger numbers were a little bit different:
** the first pH degression point (for harmine) came at 7.00 and not 7.20 as in the paper
** second one (for harmaline) came at 8.10 and not 7.70 as in the paper
** the amount of pH dropping backwards was lesser than in the paper

Calibrating my pH meter was rigorous, I've bulk lab calibration liquids 4.00 and 7.00 and 9.00 on which the meter responds 3.90 and 7.00 and 8.90 (so my probe has a non linearity of no practical concern). I would not trust these kind of little bottles very far after they've been used once (but better than nothing I suppose):

Circumstances like used acids/bases and concentrations could skew absolute values too as suggested here:

...I think yields have got a lot to do with the amount of dilution in the mother liquor. So the more concentrated the alkaloid-solutions, the higher the yield after basification...

I've no idea how pH meters (or their calibration) differs amongst us, so lets focus on the curve shape rather than exact pH values. Once nearing a pH depression point I switched base to more dilute to be more gentle so the pH depression manifest itself better and to avoid a pH overshoot which would rape the curve. Base was always added drop wise. I think it would be harder with a 0.1 resolution pH meter though, better to have a 0.01 resolution meter.

PIC PRECIPCURVE

The moment I decided to stop adding base for harmine, was exactly same pH at which the depression began, in my case 7.00. So I filtered harmine off at point X. I realize some harmine is left between X and pH 8.10 but it would be very negligible I'm sure off. This amount missed-out harmine would be taken off from the resulting THH anyway during a sodbicarb treat (10% layer trick) together with the ZnOH/ZnCO3 after reduction.

Rounding off the second depression for harmaline, based further til 8.5 and filtered off harmaline.
(note: none were GC-MS tested so far, thus all alleged, only melting points in VDS-paper were used to identify)

In the coffee filter pics: the brown heap: harmine 4.2 gr, the flat brighter one: harmaline 2 gr. As how this post opened: first precipate is brown-est, has more possible impurities.

 

[An1cca]

So nice to see you're checking this as well, Jees! Clearly the composition of your alkaloid extract is different from the one in the paper. I myself have got about 40/60 har/DHH from my extracts. I think many factors play a role in this, starting with the differences between batches of seed.

Anyhow, I'm positive that whatever the initial composition may be, this pH-metric method will separate the two alkaloids nicely. I've done 3 separations by now and must say it is a fast method as well. I do use a simple vacuum-filtration setup however (where the vacuum comes from a manual wine-saver pump connected to a Buchner-flask through a garden hose). Recommended!

If I'm not mistaken, I can see the macroscopical difference between the more powdery harmine and flaky DHH on your pictures. A microscope would confirm this separation, as well as providing some means of identification, apart from the melting points VDS measured.

By now I'm quite a fan of activated charcoal. It just takes 15 min and 1 filtration step more to enjoy a macroscopically much cleaner product. I do think however that regular coffee filters will perhaps not retain the fine particles. I use opened Norit capsules as a source. The tablets contain fillers. Recommended:thumb_up:!

 

[Jees]

Thank you for paving the way, I'll be in your footsteps

* * *

I've been mentioning this:

Hmmm, when I rise to pH9 then there is still a lot of yellow in the 'milk'. Only when rising to at least 11.5 then all yellow start becomming white, usually I rise to just above 12. I've always thought that this yellow (above pH 9) is the harmaline (DHH) that was not readily precipated as FB.

Because we had a theoretical outline of calculated estimations here

...The intersection point corresponds at a pH of around 8.75, which is the pH where one will get the most harmine with the least harmaline contamination. At this pH one will in theory get at around 92% harmine and 8% harmaline in his precipitates. ...

So I was always thinking that whatever drops out above 8.75 must be the harmaline, it fitted so nicely the calculated estimations.

So I took a better look at what exactly drops out above 8.75.
After having the harmine & harmaline filtered out as in previous post, I continued to rise pH to near 11 with lye water (I was sure lye would not precipitate). I did harvest 400mg of dry product. What is it? Characteristics:
- nice beautiful yellow liquid until it FB as white-ish;
- start of precipitation from 9 (then a first mist appeared);
- from 10.5 and above most or all yellow has gone white FB;
- FB in the water more gell-ish, no hard structure like harmine and DHH;
- more difficult to filter than harmine & DHH, because of gell type;
- some just went trough the filter leaving the filtrate misty;
- no pH depression noticed for this precititate;
- it all came along with 2 X manskes, it seems to survive manske.

I have this product because I always base my extracts till 12. VDS only worked upto 9.6 which explains he had nothing noticed above 8.5.

I had some of that product already submitted to vinegar and Zn for 3 days now, and no color change at all, see picture before reduction attempt and after 3 days. So I believe it is no harmaline. And the blacklight color indicate (don't you think?) it is also no natural level of seeds-THH that I gained.

Whatever it is (and I thought it was the strict harmaline for years), I doubt it can serve us when we're on a harmine, harmaline, THH hunt.
Next time, when I do A/B's on syrian rue, I'll limit the highest pH to 9.00 to cut out this portion of whatever it is. All hints welcome.