This is a hopefully valuable archive of a Reddit reply I made containing intricate details of DMT crystallography and spectroscopy. It is addressed to Reddit moderator ClobWobbler who claims reliance on Benzyme posts. I hope a mod can place it for a wider audience here, thanks, as I’m still a junior member. Next post will be about FASA method being obsolete folly.
ClobWobbler, misguided by Benzyme: this is a difficult post to write so sorry to be so blunt. You’ve been simply wrong about a yellow polymorph dimer that can depolymerize during smoking for years now due to lack of advanced training in organic and physical chemistry.
I have deduced the exact structure of the colored dimer involved and it’s far too covalently tough to split back into monomer DMT under mild heat of vaporization because it’s not only coupled in two but then it naturally oxidizes to create a more stable fully conjugated dimer that is highly colored but amorphous at room temperature so quite sticky. It’s also not very soluble so it oils out with or before DMT.
Besides a covalently linked “dimer” there’s no other type of dimer appropriate to claim for DMT, and has never been mentioned in the alkaloid literature nor the entire corpus of organic chemistry. Hydrogen bonding is awkward for DMT as the only donor is the bulky flat indole ring and the only acceptor has bulky methyl groups so a dimer would rip apart by entropy under Brownian motion, not exist as a lingering 14 membered ring of DMT backbones via a mere two hydrogen bonds. Charge transfer doesn’t exist with a sole boring alkyl indole. DNA is a hydrogen bonded dimer, designed for donor/acceptors to perfectly line up.
Trace amounts of dimer that still afford 99.99% pure DMT can readily taint full DMT crystals, but never does the yellow go away upon melting or dissolution of the crystals. The only “yellow polymorph” mentioned in the classic paper on colored DMT was indeed cited as amorphous. They only characterized three colorless polymorphs, discarding the waxy solidified yellow mass.
The molecular orbital energy levels involved in DMT itself make it impossible that mere crystal packing forces can shift the mere two ring indole group all the way from the deep UV region of absorption over to the visible region. A mere two neutral aromatic rings show a spectrum peak nowhere near the visible region. So just like highly sensitive drug tests can detect nanograms by forming strong visible dyes, yellow DMT is from an impurity that can’t be removed easily since even fractions of 1% of it colors crystals like food coloring colors sugar crystals on a stick:
Laboratory students out there who test yellowed DMT simply cannot find the impurity since it’s still at only trace levels and their samples are not analytical standards to begin with so 0.1% impurity that happens to be a strong dye goes undetected. Yet they still get odd little baseline peaks. That’s the yellow, you just don’t have a reference to assign them exactly, yet. But they integrate to only trace mass? Yup.
The extended pi-system of carbazoles like harmine allows them to just barely exit the UV band and enter the visible (blue) band which lets them reflect yellow to us. The DMT dimer that oxidatively couples then as is typical reforms a double bonded system by oxidizing again to loose H2 is an even bigger chromophore.
Philosophically you have been correct that yellow color doesn’t really matter for DMT purity because that yellow represents a very strong dye you can see at even trace levels. The giveaway is how the exact same yellow taint fully remains upon crystal melting. Polymorphism effects should disappear where crystals are melted down, but the color remains!
You also now have my report of ridding yellow via purification. Tough, drawn out, multi-stage, multi-strategy purification. I made the FASA method obsolete in the process, learning that hot DMSO and acetone could recrystallize the fumarate at very compact volume what acetone alone required gallons. It was the fumarate steps that let me achieve real progress in ridding straw color traces.
Here is the actual very likely dimer structure based on the known reactivity of indole derivatives similar to DMT:
2,2-bisindole itself is a brown solid, consistent with being a very strong light absorber that if diluted would show as orange or eventually yellow.
Or there just might be a CH methylene group coupling them at the same points if the first step is reversible reaction with carbon dioxide in the air, prior to dehydration to a double bond ethylene group then coupling to a second DMT. Tryptamines in general do this with aldehyde impurities and the paper on yellow DMT only got their yellow sample from acetonitrile which is notoriously hard to purify of acetaldehyde content.
And here is the UV/VIS spectra of DMT and harmalas showing how the yellow harmalas just barely reach a peak edge into the visible region (400-700nm) whereas DMT has a huge electronic gap from interacting with visible light:
Mere crystal packing forces on such a boringly neutral molecule cannot shift the fundamental absorption energies of the isolated indole ring. Color polymorphism comes from long extended dye molecules being readily twisted due to crystal packing forces. DMT isn’t a chromophore at all.
The DMT dimer however is another significant step larger in unconjugated size so becomes a much stronger dye than the harmalines. But DMT itself is just an indole lacking but a boring alkyl side chain and so is just too small in size with a way too large band gap between LUMO and HOMO orbitals to absorb light at all:
Finally, the arcane term “auto oxidation” that keeps getting thrown around isn’t a proper term used in organic chemistry. It’s from some era of field chemistry like natural products activity search, and the “auto” part means in air. It’s just air oxidation. There’s nothing “auto” about it, nothing involving two DMT molecules cooperating to swap electrons.
Here are some found images of DMT crystals that support the trace dye impurity claim, where white crystals grow out of yellow syrup, eventually filling the syrup with hard DMT needles:
Uniformly yellow crystals are even lower in impurity so the dye just gives a general rock candy effect rather than crashes out in dots or films. Permitted dye levels in sugar candies are typically 50mg/kg or an impurity level of a mere 0.005%. Color thus has little to do with practical purity levels since highly colored candy is regulated to be 99.995% pure, which is beyond even pharmaceutical grade up into analytical grade standards.