So im testing two types of DMT fumarate conversion

Batch 1:

1g DMT fumarate






is mixed with excess sodium carbonate





Water is added to make paste (ended up adding a bit to much water but no problem, just mix well and let dry)





After drying and powderizing, I pulled 5x with acetone, its now evapping, results will come later today or tomorrow.





Batch 2:

1g DMT fumarate is dissolved in pure warm water. Glass on the left is the DMT fumarate water, glass on the right is sodium carbonate saturated solution.





Some of the sodium carb solution is added to the DMT fumarate water, immediate precipitation clouds form:






I put the glass on the fridge, and after 3 days water is poured out through filter, part of the freebase is on the glass:






This was scraped and put together in the filter, which is now drying:






Yield also still unknown, both will be weighed later today or tomorrow and ill post back the results. The fumarate from the water precipitation looks beautiful and only has a mild dmt smell, not nearly as sharp as naphtha extracted product. I think this pretty much shows how one can for sure get quite pure white dmt crystals without having to resort to petrochemicals, only using food-grade chems.

Note: The water precipitation does not work with DMT acetate dissolved in water (neither with filtered mimosa brew), as tested by myself and others, probably for the reasons explained by Infundibulum in another thread, so it has to be redissolved dmt fumarate

Whoa, nice to see that people try this thing again!

Endlessness, please note that the water precipitates from the basic solution will have traces of sodium fumarate as well as sodium carbonate that are inevitable on the watery dmt crystals. SWIM had to wait for the dmt to dry, then dissolve it in acetone to get rid of the contaminating salts.

Yeah thank you for the heads up in this water freebase dmt crystalization method!

So how comes there would be traces of sodium fumarate and sodium carbonate? Arent they very soluble in water? Considering I filter dry the dmt crystals, do you think it would still be a significant amount of sodium fumarate and carbonate?

BTW, yield of the water crystalization came through... not very good but not incredibly bad: 450mg (from 1g dmt fumarate)

the acetone pulls from the other method are still evapping.

If the mimosa brew used vinegar, it would most probably not work. If it used another acid such as fumaric acid and was cleaned enough such as you suggest, who knows... Would be wonderful if someone tested that!

I dont understand the skim milk idea, how does that work? where does the milk come into play?

If there are traces of water in the crystals then there will also be traces of sodium carbonate and sodium fumarate, unless you vacuum filtered to deduce the amount of water to the absolute minimum. Yet, as I said it is traces and possibly negligible; it may even be comparable to traces that can be carried over if, say, 99% or less acetone is used for the "paste" method.

Endlessness, I also want to ask, was there the water after the water precipitation yellowish? The dmt fumarate seems to have a fair amount of n-oxide, whiel the precipitated crystals are very white. dmt-n-oxide is much more water-soluble and will stay in suspension. When SWIM did these experiments there was a significantly yellow supernatant left atop of the precipitates.

The latter is an issue since some n-oxide in suspension will create "hydrophobic pockets" (or in other words it will make the solution more hydrophobic-friendly) and prevent precipitation of a significant amount of nn-dmt. Did you try to stick in the fridge to see if more precipitation occurs? Or even let it dry and pull with acetone to see how much is left behind unprecipitated??

Interesting Dagger.... Be sure to tell us if you ever try that1

Ok so, final results are in. The evap from the acetone pulls on the dry sodium carb+fumarate paste looked like this on the evap dish, white crystals with some yellow in between;






but when scrapped together it was very waxy and yellow, even some browner darker spots:





the final yield from both methods, each one separately starting with 1g dmt fumarate (of the same batch):

Water crystalization: 450mg white crystals
Sodium carb paste + acetone pulls: 470mg yellow wax

Here's both side-by-side to see the difference in quality:





Conclusion: Quantitative yields are very similar but water crystalization yields purer product and avoids the need for acetone/IPA pull.

Question that remains: Why is there a 50-60% loss in conversion of FASI fumarates regardless of method used, while the theoretical weight loss of the fumaric acid molecule should be around 30%? If it was FASW it would be understandable because of excess fumaric acid, but FASI should theoretically consist of only dmt fumarate.

Again very nice demonstration of principles!

There is another conclusion, that dmt n-oxide is much more water soluble and remains in suspension - the 20mg that are absent in the water precipitation but present in the acetone pulls are n-oxide.

In either case, the expected highest yield in each method is ~760mg freebase dmt so with a real-life yield of 450 there seems to be a loss of 40%

The reasons for this may be the following:

1. There is contamination with fumaric acid, but this alone cannot account for such a yield loss.

2. It is likely that there is more n-oxide not pulled by acetone? Could you try to pull with another solvent like ethanol, and see if you can get more out of the dried paste? SWIM did this in the past and when acetone stopped pulling he pulled more with dcm with no increase of yield though.

3. How about the stoichiometry of the salt? We assume to be 2 molecules of dmt per fumaric acid, but in FASI and limo a good amount may be mono-dmt fumarate or one molecule of dmt to one molecule of fumaric acid. In the latter case 1 gram of dmt mono fumarate has an expected yield of 610mg as freebase.

what is also interesting though is that whenever SWIM does the freebasing he goes FASA in xylene and his acetone-washed fumarates often give a yield in freebase that is at least 700mg+ freebase from 1 gram fumarates. On the other hand, the only times he used limonene + FASI was on some chaliponga extractions and he got very similar yield losses (~50% less than expected) in the freebasing.

hey inf!



How can there be any significant contamination of fumaric acid? Remember this is FASI not FASW. Fumaric acid is soluble in the IPA+limo mix (which one can easily test by adding FASI to the limo after precipitation of alkaloids, and there is no precipitation of the fumaric acid). So the precipitation should theoretically only be alks+attached fumaric molecule, no?



Correct me if im wrong but I thought there couldnt be any n-oxide present. How could it be present if it is a zwitterion, therefore not having a salt form (so how could it have precipitated when adding FASI to limo) ? In FASW or vinegar pulls it would be more probable to be present because it would simply migrate to the acidic layer, but in FASI we're talking about precipitated crystals... Or is there something Im missing?

Also another thing to question.. Is it SURE that acetone does not pick up n-oxides? And how sure are we that there are n-oxides at all to begin with? I ask because I already made a side-by-side test, I had freebased some dmt acetate with sodium carb, divided the dried product in two halves, and one half I pulled with IPA, the other I pulled with acetone. I did 5 pulls in each and the yield was exactly the same! So either acetone does pull n-oxide, or I had no n-oxide at all in my bark and neither it formed when evapping the dmt acetate with heat and fan.

but yeah I cant pull anymore because I threw the (supposed) spent product away, Im out of IPA and no money so thats why I used acetone, and with the test and for the reasons I mentioned above I thought I would be getting all the actives.




This is VERY interesting! So how feasible chemically is it that the solvents used (xylene+acetone vs limonene+IPA) will affect the stoichiometry and that limo+fasi would create mono-dmt fumarate? Is there any test without advanced lab equip that could be made to determine in which form of fumarate it is, or how can we know if mono-dmt fumarate exists at all?

This is one of the most useful posts! thank you endless

If you follow the Acetone conversion method exactly, but instead of using acetone, use heptane and freeze precip- the yields for SWIM have always been as expected (700 to 750mg of very pure freebase per gram of Fumarates). Whenever SWIM attempts the conversion using IPA or Acetone, the yields are far less than expected (similar to your results). Why this is, SWIM is not sure.

so what you are saying is that the step where the dmt seems to be lost is when pulling from the sodium carb+dmt dried paste? So if one pulled with heptane instead of acetone it would come out more? Thats interesting.

But then this also defeats the whole purpouse of avoiding petrochemicals and using food grade/usp grade chems like i was trying, but its worth it for someone to test it I guess. Side-by-side with heptane vs acetone pulls.

In any case, I would have thought that 5 warm acetone pulls would have exhausted the paste. How many heptane pulls do you do?

Of course, but on the other hand same goes for xylene + FASA; Taking xyelne and adding FASA does not show any precipitation. However, bufomas nass-speced FASA precipitates and he found a significant amount of fumaric acid, he also made a tek to clean up fumaric acid by water crystallisation of dmt-fumarate.

The explanation for this possibly lies on the way the extraction is performed; sure, FASA added in clean xylene and FASI added to clear limonene does not produce any precipitation. What happens though when FASI is added in limop that has already been used for 4-5 pulls, that has been saturated with all sorts of fatty and lipid impurities? it is likely that the change in teh chemical composition of the solvent (or "solvent ageing" may cause precipitation of a significant amount of fumaric acid as well as fumarate salts.

Could one test this by adding some FASI in limonene that has been used for many pulls and cleaned with acidic water?


There is n-oxide; and it dissolves in acetone - If there's no oxide then what could give the yellow colour? This implies (but not specifically indicates!) presence of the N-oxide. N-oxide is not exactly a zwitterion and it can form salts, so it can be present in FASI pulls from limo. It is a weirder-than-normal salt, and I should make a good drawing at some point if anyone is interested in extreme details!

(on a sidenote, it is wrong to say that zwiterrions do not dorm salts. there are indications that psilocybin forms salts and amino acids, the most abundant zwiterrions in biology also form salts!)


It is feasible, chemistry is so complex and we should not be surprised! All these hypotheses about the different are testable but tough!

I had missed this message:



No, the water after precipitation just a bit opaque, white-ish colour. and yeah I did stick in the fridge for another couple of days and there's a bit more that precipitated and the water got clearer. I think it might be those 20mg that there were of difference, didnt weigh it yet, its finishing drying. It is also not as white as the rest, its a bit more tan/pinkish. Ill take pic and weigh it

so doing an acetone wash would pick up the dmt and not the salts, correct?



but if that was the explanation, it wouldnt explain my case because all the limo pulls were fresh! So it has to be something else.... Any more ideas?






I thought the yellow colour could come from, for example, plant oils. But then if you pressed me and asked me how the plant oils got into the precipitations in the first place I wouldnt know how to answer you. Do you think thats possible?

I also thought about these 'other' alkaloids (NMT and 2-methy-1,2,3,4-tetrahydro-beta-carboline ?), and whether they might be yellow? Lastly, does dmt exhibit polymorphism? Is it possible the yellow is related to how the crystals form and not only in impurities?




I would love to see a drawing of that molecule! How stable are these salts, and is specifically dmt fumarate a possible salt for dmt n-oxide? and if so, do you think dmt n-oxide-fumarate would be a crystaline solid?

Just as a side note, I'm interested in the details about N-oxide, and I'm sure I wouldn't be the only one!

Yup, exactly!


Aw, crap... I think then I run off of explanations. Either case, it is unlikely that any fumaric acid contamination would be the prime reason for yield loss.


All these explanations are plausible but isn't the presence of the oxide the more "intuitive" at the moment? Of course there must be other alkaloids who may also contribute to the yellowiness and of course the presence of the oxide in FASI precipitates has not been demonstrated by an analytical method that can detect the oxide, so I may as well be totally wrong here.


OK, the drawing is attached, but it is of dmt-n-oxide hydrochloride instead of teh fumarate for simplicity! The correct structure of dmt-n-oxide is also given. Please note the "arrow" bond between the nitrogen and the oxygen. This denotes a dipolar bond. This special bond means that in the interaction of nitrogen and oxygen there is not mutual share of electrons but rather the nitrogen donates the electrons of the bond. Because of this, nitrogen is slightly more positive and oxygen slightly more negative. This is typical of amine oxides. Amine oxides can steal protons (or "hydrogens" ) that stick on the oxygen to form hydrozylamines . During hydroxylamine formation of a tertiary amine like dmt the nature of the dipolar bond changes and nitrogen becomes positively charged, or a cationic hydroxylamine. An example of something that approximates the structure of a dmt-n-oxide salt is the structure of hydroxylamonium chloride. In the wikipedia article just imagine the hydrogens of the nitrogen substitute for 2 methyls and the rest of the dmt molecule.

EDIT: I forgot to address a couple of questions, DMT-n-oxide fumarate is a possible salt but I do not know how stable it is. I do not know either whether it is a crystalline solid, but it might be; dmt n-oxide freebase is oily, but it can be frozen at low temperatures to produce a crystalline substance. Same goes with mescaline freebase which is oily at room temperature but freezes to crystalline. And as the oily mescaline freebase makes crystalline salts, it is possible that dmt-n-oxide fumarate may be a crystalline substance.

SWIM would like to try making some n-oxide and further experiment when he has time for experiments, for the time he's horrendously busy...

taste it - if it's crazy bitter, it may be fumarate - if it's crazy sour, it's probably just fumaric acid - if it's bland and tasteless, some crazy biproduct of the skim milk i'd imagine.

I'm trying to understand why Vinegar evaporates produce a very hard-to-handle runny dark goo freebase; compared to when one takes fumerates and does the sodium carbonate conversion to freebase? I believe this can be done with 99% IPA as well. Maybe I am missing that the full range is not completely present in the original fumerate step, as compared to salting out with vinegar.
IS this the reason?

Thanks for helping me understand this. I will try to look over the threads to find the answer.

It could be that I was under the impression that FASI yielded a full range fumerate product, but this may not be the case.

One of the reasons could be more n-oxide present in acetates than in fumarates from FASI for example. In this very thread if you read further back I had a discussion with infundibulum about n-oxide. He affirms it could be possible to have n-oxide fumarate, though I wonder how easily it is formed, in, for example FASI, maybe only small amounts? While chemically it will definitely be in the evapped vinegar acetate if it was there in the first place. So I think thats one main reason.

My mind was slow initially; I just grasped the beauty of the water precipitation clean up. Excellent!
Maybe some loss is worth it to keep it clean/green.
If the yield is the same as IPA clean up, and the product cleaner and purer, then no doubt this seems like the best system for pure spice yet.

Endless, I know you like to do some things by eye, and I'm sure you've earned your right with your experience

Would you please consider suggesting a basic formula, i.e., a ratio of water and then how much saturated Sodium Carbonate solution to add to the water.

If per chance there was any residue present in the drying water, then keeping the Sodium Carbonate water to a minimum would be helpful toward this end.
And, is there a chance that keeping the water volume to a minimum would be of any benefit toward minimizing loss during process?

I would have to imagine that if this was done with a gallon of water with one gram fumarate spice, then the loss would have to be more?

Judging by your pictures -- and those look like Pint drinking glasses, you dissolved 1 gram spice in about 4 oz hot water.
Then you added 1-2 oz saturated sodium carbonate solution.
And then you waited a couple days but did not refrigerate because this would be unnecessary.

And so filtering out the acetone through coffee filter to keep the fumaric and carbonate traces in the filter, and recollecting the crystal in hte evaporation dish ?