UPDATE 2023: This thread now also includes some experiments that were conducted, added after that theoretical ideas. But the first analyze of the Spectra had an error: DMT-N-Oxide was mistaken with the DMSO-Signal. That gave a wrong amount of signals which would have indicated a polymerization. Sadly the corrected spectrum does not give any strong signs of polymerization and therefore the title was also changed back, to account for a still more theoretical discussion about polymers





So over the years there was a huge talk about DMT Polymerization. When heating pure white Spice on a hot plate, most of it vaporizes, but still there is some orange residue, which will not vaporize above 180 °C. This residue seems to be dissolve slower in Acetone than regular DMT. Also I've seen some Spice crashing out very fast at Room Temperature, while another portion just came out in the Freezer. Ok ... yes this is normal for Re-X, but I mean it felt like these were different batches of Stuff.
So either its just a sort of combustion with charred products left behind, or indeed DMT may form bigger structures that dont vaporize readily.

A possible solution to distinct between combustion products and polymers would be vaporization in N2-Atmosphere obviously.

But how could any possible Multimer of DMT look like? Amines would be the most reactive part of the molecule obviously, but when looking into other examples in literature it may be something different:

1. The enzymatic polymerization of Psilocybin uses that 5-position and 2-position. Now as this is an enzymatic reaction, there is no direct translation, but if Nature is doing stuff, then it mostly means something - or in other words, it indicates that both positions may be most promising if also using heat. (source)



2. Fischer-Indole Synthesis is using Hydrazine precursors and electrophile fragments to build tryptamines. While performing such a reaction it may be possible to get fragments with higher mass which evolve by excessive nucleophile attack of that 2-position Carbon. I think this also indicates that this 2-position may be indeed quite reactive. (source)



3. Another paper even specificially talks about this pathway of general indole polymerization, so I guess that this may really be the most meaningfull way of also polymerizing (by accident) DMT. (source)



In principle that tryptamine skeletton could act as either nucleophile or electrophile, with electron-donating substitutens even increasing nucleophile character:



But this way only Dimers could be formed. To get up to multimers with more than 2x of DMT mass like Benzyme saw in MS this could then only happen by 2,3-addition. The 3-position would also be a good nucleophile, due to forming a tertiary carbo cation in acidic environments. In this picture these 2,3-polymers are shown. (Yet for substituted indols like DMT, this 3-position would be super hindered). (source)



Any easy way to verify that would be to retrieve a sample with possible polymers ...
... and extract just the DMT by pulling with a low amount of Room Temperature Naphtha for 5 min and then throw the residual stuff in NMR.
If indeed this is the way of DMT-Polymerization then the

Signals at 1H 7,1 ppm (2-Position) --> should be (nearly) gone or at least not fit anymore to stochiometry
Signals at 13C 122 ppm (2-Position) --> 127 ppm - ok ... this will be super hard to see/distinguish/verify ...










Sadly there was an error in the discussion of both NMR. It is important to have a signal which can be normalized to the spectrum in a way that the spectrum depicts exactly 1 molecule. This way it is possible to measure a decrease of certain protons and therefore a successive polymerization of molecules. Sadly I was pretty dumb and mistook the solvent signal (DMSO) for being the N-Oxide. Now it's corrected and means we have much less N-Oxide and therefore also much less reacted DMT. The new text reflects the proper results.

I tried to induce DMT-Polymer formation and then analyze with NMR. 2 possible ways were examined: base-catalyzed polymerization and radical-induced polymerization. I have no idea if any of these will indeed induce a polymerization, but that's why I did those experiments obviously ... It does not reflect how DMT may polymerize inside of a cooking pot while extracting, but I thought it could be a more direct patchway and make it easier to find any clues of it's structure.

NMR spectroscopy results in a spectrum which can be used to idenfity certain structural features of a molecule depending on the protons attached to it. Every proton will give a signal in 1H-NMR and the X-axis position of those is depending on the chemical environment. Different environments will cause signals at different X-axis positions, which is labelled at [ppm] as unit. Now when looking at the respective unsaturated DMT signals (the one which are believed to may cause polymerization) a decrease of these signals will show that the total amount of those protons was decreased per average molecule. Now the signal which has reduced the most indicates a place where a reaction may have taken place to the highest amount, ultimately leading to the possibly most reactive site of the molecule regarding a possible polymerization reaction.

Comment: Sadly I'm not using the correct nomenclature to numberize the atoms - my NMR Analysis Prediction is determining the numbering and I have no idea how to change it, so I'm just going with that. To make things worse this numbering seems to change from time to time, but well then you can play Sherlok Holmes trying to follow these numbers.




Idea:
DMT is used without solvent to maximize concentration and therefore polymerization chance. The sample is heated above melting point to speed up reaction of course, but also to give DMT enough mobility to undergo a reaction. As a base Triethylamin (Et3N) is added, which is a much stronger base than DMT and might accelerate the formation of nucleophiles that could be the initial step of forming a polymer, as well as for potential propagation reactions.

Procedure:
- Heat DMT for 10 days at 70 °C + 1 wt-% Triethylamin added every 3 days

Resulting dark oil was extracted 2x with boiling Heptane to keep only what is insoluble in Alkanes to ensure not picking up un-polymerized stuff.

Resulting Polymer: 5 mg (from 100 mg)



N-Oxide: (3,13 ppm)
We can see some N-Oxide, which is in a ratio of roughly 1,80 : 4,18 = 43 %. But that is not the N-Oxide in the whole DMT, but the one that was left when extracting. To calculate inside the whole batch:

100 mg DMT heated
5 mg DMT-Oil remained which is 43 % N-Oxide
= 2,5 mg N-Oxide = 2,5 % N-Oxide

So actually a pretty low amount of N-Oxide considering that it was heated for so many days. Still in that old experiment when heating for 2 h at 100 °C there was no N-Oxide. So it seems if you totally exaggerate with heating you indeed get quite some N-Oxide, even when there is no light present. Still those are just traces and probably not the reason why the rest did not dissolve in Heptane.

Aromatic Signals: (6,98 - 7,50 ppm)
Sadly they look exactly like before. No signal reduction. That would indicate none of these indolic sp2-carbons underwent any reaction and therefore probably also no polymerization to be seen here ...

Indolic NH: (10,84 ppm)
Here we have a signal that is only at 68 % strength of what it should be based on former measurements. Still that proton is somewhat acidic / loosely bound and therefore might exchange with Deuterium in the solvent. That again means that the magnitude of integral of those acidic protons are something not to be taken too seriously when comparing the absolute values. Still there is a definite decrease to be observed. That would mean this N has reacted in some kind of way, probably around 1/3 of them inside of the whole sample that was left over after Heptane boiling. But as there is no reaction observable within the same magnitude of any other carbons of the aromatic signals, there was probably no coupling with any of these. Only viable option would be coupling both N with each other, which would not happen in a base-catalyzed reaction. It can be simply seen that we have 3 different types of this signal:

NH a) = 10,66 ppm (0,05 = 11 %)
NH b) = 10,84 ppm (0,27 = 61 %) - this is unreacted DMT (at least pos. 5 & 6)
NH c) = 10,96 ppm (0,12 = 28 %)

A signal which is probably the same, but shifted a little left or right indicates a very similar structure with some minor differences within close proximity (1-3 bonds remote of the corresponding Proton). In NMR closely bound substituents will have an impact on the chemical shift and shape of peaks. If that peak is shifted, it could indicate that there was a reaction at position C5 (only option within max. 3 bonds distance to NH6) taking place, that shifted the signal either left (higher ppm) or right (lower ppm). In general it is possible to explain the chemical shift based on chemical environment like this in an easy way:

Higher ppm: Proton has a low electron density, with high electron density neighbour atoms and polar bond
Example = Acetic Acid H-OOC-R

Lower ppm: Proton does not have reduced electron density, with no increased electron density neighbour atoms and not polar bond
Example = Methan H-CH3

So a shift that could have taken place, when considering options for polymerization would be as follows:



Higher ppm could be induced by reaction of C5 with N6 of another molecule.
Lower ppm could be induced by reaction of C5 with any other sp2-carbon of another molecule.
10,84 ppm is DMT without any reaction at C5 or N6
What is missing now in total to Integral = 0,87 has reacted with N6 and is therefore invisible.

Sadly we dont have a sign for any C5 reaction and so there is also not too much as an insight from here. Still a reaction of N6 with other sp2-Carbons would be possible as also discussed in the big thread of DMT-Polymerization by Loveall, see post of Benzyme.



Impurities:
We can see some impurities in the aliphatic range at 2,07 ppm and lower. But these are nothing that would be displayed by DMT or a DMT polymer that would exist based on the theoretical thoughts above, so also no further hint.

Summary:
Sadly no sign of polymerization. Indeed the signal of the NH at Position 6 seems reduced. But there is no hint what kind of coupling could have happened, as there is no corresponding C-atom which also has a decreased signal.


The first evaluation was performed wrong and assumed that another signal was the N-Oxide. Therefore the overall amount of signals was reduced much more and this lead to a heavy decrease of those aromatic sp2 R-CH=R bonds. The NMR based on this assumption is still seen here. Maybe someone with more training would see if there is not maybe still some hint of a reaction. This probably wrong assumption lead to the polymer structure shown here, but this one is probably not true or at least not a consequence of the newly-discussed NMR-result.





Idea:
DMT is used without solvent to maximize concentration and therefore polymerization chance. The sample is heated above melting point to speed up reaction of course, but also to give DMT enough mobility to undergo a reaction. As an Initiator the classic thermally-activatable AIBN was used. This one needs also 70 °C for decomposition and will create radicals.

Procedure:
Heat DMT for 1 night at 70 °C + 1 wt-% Azobis(isobutyronitril) (AIBN)

Resulting dark oil was extracted 2x with boiling Heptane to keep only what is insoluble in Alkanes to ensure not picking up un-polymerized stuff.

Resulting Polymer: 4 mg (from 50 mg)



N-Oxide: (3.20 ppm)
We see much less N-Oxide, only like 5 %. But as this is only the sample with the unreacted DMT removed by Heptan the calculation of the whole content is:

50 mg DMT heated
4 mg DMT-Oil remained which is 5 % N-Oxide
= 0,2 mg N-Oxide = 0,4 % N-Oxide

That is actually much less than in the base-catalyzed setup. Of course this one was heated much shorter, but AIBN is a radical initiator and thus I would have expect to produce quite a bunch of N-Oxide. So formation is much lower than in the first experiment and even negligible in total. A foolproof way to create N-Oxide with Radicals is H2O2, which will go to 100 % conversion within just 1 night. But here you create Oxygen Radicals, which are maybe not so prone to evolve if you start with AIBN in concentrated DMT.

Aromatic Signals: (6,95 - 7,53 ppm)
As before basically no reduction in the total signal count (5 -> 4,74). This drop could also be within the uncertainty of NMR-measurement with these type of sp2-Carbons. So no hint of polymerization as above ...

Indolic NH: (10,79 ppm)
This time only 1 signal, but the reduction is also probably not within the error range as it is - 28 % (100 % should be ~ 0,87). So probably something happened here again and it is nearly at the same magnitude like for the reaction above. But also here we dont have any signs what could be the corresponding coupling.

Impurities:
Nothing to be seen except water and the solvent (DMSO). Looks like some pretty pure DMT, which is in this rare case not really what I was hoping for

The first evaluation was performed wrong and assumed that another signal was the N-Oxide. Therefore the overall amount of signals was reduced much more and this lead to a heavy decrease of those aromatic sp2 R-CH=R bonds. The NMR based on this assumption is still seen here. As these results were kind of similar for both reactions, a reactivity pattern was also assumed based on how strongly the signal of each proton decreased and shown here. As also with the polymer structure this is both not a consequence of the newly analyzed and corrected spectrum, so probably outdated.



Summary:
So after correcting the formerly probably wrong evaluation, there is no sign anymore of DMT-coupling. It was a pretty much fitting theory that was put up in the first time and without contradictions, but as now there is no real proof of this anymore, it's better to drop what can't be proven Whatever is left from these reactions, it is impossible to dissolve it anymore in Heptane at any temperature. Therefore I strongly believe this is some kind of heavily altered Tryptamine. Still, it just shows up as regular DMT once again

Not sure which other experiment could be done. But it's just weird to me that no matter which Polymer is analyzed it always shows up as just DMT. Benzyme got some perfect proofs of polymerization with the Mass Spec, but even this experiment right here could not give any answers HOW these molecules might be attached to each other, of course probably just because the conditions did not create any polymers in the first place. Still these conditions were thought to be really effective at it, which makes me wonder how to strongly force that polymerization alltogether.

"That 1-position" is actually the 2-position, as outlined in the bottom of the last diagram you posted
That is actually the normal way of numbering an indole - the pyrrole nitrogen is the 1-postion and the atoms shared between two rings are not counted because they don't normally bear substituents. If the conjugation is disrupted they can get numbered 3a and 7a.


Interesting nonetheless - I hope you or someone else with access to the machinery gets to run the NMR on some goo soon.

Oh well ok so that may also be a way of counting. I just checked 2-Methyl-Tryptamine and so youre totally right. Guess I should now changed that "1-position" to 2



Technically I would have access, but right now I just dont have the balls to do.

As written here I think that upon heating DMT partially polymerizes as you will always have trace amounts of a dark/black residue, even when using pure Spice. This residue does not vaporize further even when heating excessively above 200 °C.

So either this is just a regular sort of combustion / charred products or this is a sign of trace formation of multimers or otherwise just single-molecule DMT combustion / degradation which forms this black stuff.

An meaningful way to check whether this is just a charred product from a combustion reaction would be to heat DMT in an inert atmosphere - then there should be no chance of forming oxidated combustion products. If that coloration still occurs, this may at least be another hint that also thermally polymerization could happen.

Here is what I tried to come as close as possible to this:

2 portions of DMT were placed on a inert surface (no metal, to prevent any oxidating effects of metal surface) which was just a petri dish - then this was covered with a beaker and with a needle there was a constant flow of N2 into that chamber (left). Also 1 portion of Spice was placed besides the beaker in full O2-contact as a reference.



Now the chamber was heated from below with the hot plate to above 160 °C. Both the reference and the N2-Spice vaporized and you can see that both produced the same pattern of black residue (right). So there is the same result in both cases with the chamber (hopefully) excluding any contact with oxidation agents. So maybe the residue may not or not fully be caused by oxidated charred combustion products, but by something else and this could be thermally-initiated oligo-/polymerization.

Also very often I saw that when freeze precipitating there was a quite fast precipitation of more yellow stuff already at room temperature, causing not that nice crystals. Then the real white stuff only comes out when temperature reaches -20 °C and this takes much more time, but will give much nicer crystals.

I always thought: Ah ok ... maybe thats just "non-pure" DMT coming out first by too fast precipitation and this always has this ugly morphology. Also this yellow stuff tends to be more sticky, no matter how long you fan it. The white goodies are the only ones, who will tend to be super dry and nice in shape.

Now all the time when recrystalizing this ugly yellow stuff to convert it to white, then it will always stay in its yellow slightly sticky form. So it seems this is not connected to impurities.

Also when heating the DMT it is vice-versa: The white stuff may even dissolve at Room Temperature, while that yellow stuff will only become a liquid. So refluxing the Naphtha is needed to get this stuff in solution.

But as I always thought this is equal DMT I never bothered to separate both by their different dissolving behaviour in Naphtha.

So regarding the Multimer-Question I now think that this sticky yellow stuff are the Dimers/Oligomers (and also still some real DMT) and the white stuff is the one-and-only DMT monomer. Thats why it takes much more temp to dissolve in Naphtha and also why it crashes out much faster in Naphtha - solubility is just even worse than Monomer. Also thats why it is sticky no matter what you do and why it is yellow and not white - no matter regarding purity.

So after ages I finally did a fast test to confirm this - I had 3 different samples and made a TLC (MeOH + 2,5 % Ammoniak). You can see that all 3 samples contained DMT, but also a spot at half-height. This should be a Dimer (no idea if Trimers etc. would show up at same height, so I can just confirm Dimer-presence). Also all these 3 samples looked different and some were really white, so it seems that even good-purity DMT may have Dimers, as long as its derived by boiling-Naphtha-re-x.



What you can see here: When you dont heat the Naphtha to boiling temperature, then you will not dissolve the brown oil. This means that possible multimers stay on the bottom. So once you freeze that surfactant naphtha, you will only get the white crystals, not the yellow sticky stuff.
I would advise doing this anyways, because the Multimers will vaporize at higher temperatures and if you vaporize correctly at 165 - 200 °C, then the Multimers will only clogg your device with brown sticky residue over time. This could also happen with DMT, but then at a much lower rate.

These multimers may also be active, so just use them for anything which does not involve proper vaporizing, like making Changa (over-torching your stuff anyways) and Pharmahuasca.

Thanks for doing this - that's a great post.

Could you perhaps try a 2D tlc with an acidic solvent system on the second run to see if the putative dimer/oligomer spot breaks up into any monomer? It might require increased loading to give a visible amount of the 2nd spot, or of course just use the brown oil from a re-x.

Someone (you? Benzyme?) has suggested that an A/B breaks up the oligomers again, hence the suggestion for an acidic tlc run. This could also be behind how the oligomer is still (seemingly) orally active.

I guess the time scale for backconversion would be much longer than TLC running speed :/

But indeed would be interesting if that works. But actuall I have a quite big batch of mostly *pure whatever*, what is left after doing several pulls with 40 °C Naphtha. I guess even then this will contain leftover DMT, but would be a better starting material for conversion tests.

If I still had HPLC/GC access then it would be just too easy to simply measure detector signal before / after. Maybe there will be a MS soon here, but with no chromatograph prior. Still I could check any intensity of multimers before / after conversion? But I havent used MS in years so lets see how that instrument would even perform.

I could check if that leftover stuff is mostly free of DMT, then stirr it in Trifluoroacetic acid and then just do a DC of this. Acid will evaporate super fast, but not sure if it will depolymerize completely without water.

That lower spot is NMT, you'll see it in pretty much every single DMT sample from mimosa.. If you don't see it, increase concentration and you likely will.

Well that would be plausible regarding polarity - but its an MHRB extract so I guess there shouldnt be any NMT?

I've seen it repeatedly in my tests.. If you have some NMT, or an acacia confusa sample to compare, I'd bet you'll see it's that very spot.

MHRB does have NMT, it's in small amounts though, I've seen it in GC/LC-MS spectra, I remember burnt seeing in his samples too.

I posted this in the reddit sub, regarding NMT



Correct me if I'm wrong on this; some species may have higher NMT conc. than others.
I've done analysis on 12 samples purportedly being a NMD/DMT mix, with 4 of them claiming to be only NMT. Three of them only had 189->145 m/z. The fourth had a small 175 m/z.

Oh yeah...and in case it hasn't been mentioned before... 145 m/z appears for most simple tryptamines, as an ion transition species. It represents 3-vinylindole....unless there is N-substitution on the pyrrole nitrogen, then it's 158 m/z.

(this is for LCMS)

Hmmm well both points taken, not sure whether it is some or not, but maybe I can check by Thermogravimetry (NMT should still vaporize mostly below 200 °C) or MS.

Acacia confusa def has large amounts of NMT, for example, together with DMT. Not sure what kind of biological mechanisms are at work, if it's a unique thing or not.

In my own tests on Acacia confusa, NMT went from 22 to 55% of alk content, the rest being DMT.

As its referenced in Trout's Notes, some samples of Acacia confusa show around half half NMT: DMT ratio, or another test showed 3:1.

I haven't found references for Mimosa hostilis having NMT in literature but burnt detected it and so did I, although in very small/trace amounts only.

I've noticed depending on method used, it co-chromatographs with DMT.

I had NMT reference standard to compare in all tests, TLC and for the GC/LC-MS.

On the polymerization... I initially postulated that it may have been an artifact created in the ion source, as it is sometimes seen in single-quad ion traps.. but if this is also seen in GCMS, then that's great news, it means it's reproducible data. I suspect it is occurring during basification, and yes, they can be cleaved via acidic hydrolysis.

quick question guys, but we have to assume a couple variables. the question is what kind of time frame would i be looking at to diy learn this stuff to be able to participate and contribute in discussion like this. and in what direction, or via what material would i study to fast track the process..

the variables we have to make assumptions about are my capacity to learn and retain the material i study, and to understand it without needing someone to explain some of it, or put perspectives on it for me.

i dont wanna sound arrogant, but go ahead and assume that i would be one of the first to complete an exam, i would do it without notes mostly, and the teacher would think i must have cheated because i never studied, i just read and retained the material. and understood how it all relates and fits together.

with a present knowledge of nearly zero chemistry. just learning what magnesium sulphate is, starting to understand pka, no periodic table understanding, cant write a formula or reaction, etc.

I have no chemistry background, I learned all I know over the years mostly here in the Nexus reading threads, googling stuff and asking in the chat or forum when I couldn't find answers on my own. And then of course, experimenting, doing extractions, and eventually I was lucky enough to work with analysis at an NGO where I also learned a bunch. Even though Benzyme and Brennendes Wasser would run laps around me easily in terms of real chemistry knowledge, I'm happy to have gotten this far as an interested hobbyist that simply put in my time reading and experimenting.

There's endless information out there, you just have to take it step by step and dedicate time to it. I can't say how long but the longer you're at it, the better you'll understand things.

If you're interested specifically about the analysis part, I've written a megathread specifically about TLC and another older one on many different analytical methods.

Also make sure to read the FAQ , there's a lot of info written or linked there regarding general extraction chemistry and related subjects.

Outside the Nexus, there are many resources, im sure Benz and Brennendes can say some good ones too.. Khan Academy has great free chemistry courses, I'm sure there are many others out there too.

Good luck and be sure to ask if you do your research and can't find answers.

thank you, at present, im interested in the pharmacology.

You can find some interesting articles if you search for "tryptophan decomposition". Tryptophan is obviously not DMT, but it is structurally similar and may decompose along the same pathways.

The first image below is from the first paper attached, specifically chapter 2.3 on heat-induced decomposition. It appears to me that such decomposition for DMT is quite likely, and I would not be shocked to learn that there's indole and skatole in that brown mess.

The second image is from the second paper attached. The authors show that decomposition in air begins around 90°C.

Offtopic: the attached old paper is not related to thermal decomposition, but is interesting nevertheless. The authors are observing a particular detrimental effect of HCl on tryptophan in aerobic conditions.

One cannot be sure without testing, but I would personally recommend against using HCl for tryptamine workup or DMT extraction if other acids are available. Boiling root bark in HCl solution, for hours, under air and light, is probably the worst-case scenario.