TEK Stoney Harmalas Process

[Prima Materia]

A pleasure to see more developements being made here. Great work indeed!

Is it possible to use sulfurous acid (which is to be found much easier and safer to work with) instead of sulfuric acid? Altough working with sodium sulfate does make even more sense, I guess? Also, could one use gypsum (calcium sulfate) or epsom salt (magnesium sulfate), or would Ca/Mg ions interfere with the process?

I've got a nice chunk of harmala freebase to play with and this new approach might be just what I need to get me working.

 

[Transform]

Also, could one use gypsum (calcium sulfate) or epsom salt (magnesium sulfate), or would Ca/Mg ions interfere with the process?

Hey, great to see you here!

Calcium sulfate is less soluble than the harmala sulfates, so it wouldn't work for that reason.

Magnesium sulfate has good solubility, but presents the possible complication of having an almost insoluble hydroxide, so removing magnesium contamination gets a bit complicated. It's not impossible, however - by dissolving magnesium-contaminated harmala base in dilute phosphoric acid and carefully adding drops of ammonia solution, the Mg can be reliably and quantitatively precipitated as the highly insoluble magnesium ammonium phosphate (MAP) but harmine does precipitate first (iirc). MAP is pure white, so it's easy to tell apart from the harmine and harmaline. And, just in case I'm mis-remembering the order of precipitation, we can look up the pH where that happens. I seem to recall pH 5.4 so MAP might come down first anyhow.

 

[Prima Materia]

Magnesium sulfate has good solubility, but presents the possible complication of having an almost insoluble hydroxide, so removing magnesium contamination gets a bit complicated. It's not impossible, however - by dissolving magnesium-contaminated harmala base in dilute phosphoric acid and carefully adding drops of ammonia solution, the Mg can be reliably and quantitatively precipitated as the highly insoluble magnesium ammonium phosphate (MAP) but harmine does precipitate first (iirc). MAP is pure white, so it's easy to tell apart from the harmine and harmaline. And, just in case I'm mis-remembering the order of precipitation, we can look up the pH where that happens. I seem to recall pH 5.4 so MAP might come down first anyhow.

Thank you! I'm delighted to see there is this sort of activity in the forums.

I believe the quoted approach would make sense when going for separation of the alkaloids in the firstplace and also having epsom salt at hand to work with.
Also in that case it just makes sense to use the regular NaCl instead. Or am I missing something? Maybe the advantage of removing more impurities via sulfate salt form precipitation?

Otherwise, wishing to end the extraction with Harmala sulfate salt it seems one mush work with either sodium sulfate or sulfuric acid.
Is it possible to substitute sulfuric acid with sulfurous acid (H2SO3)?

 

[Transform]

Is it possible to substitute sulfuric acid with sulfurous acid (H2SO3)?

Sulfurous acid is a relatively weak acid which decomposes readily into toxic and foul-smelling sulfur dioxide gas. Both the acid and sulfur dioxide are fairly strong reducing agents, and bisulfites are well-known for forming adducts with various ketones or aldehydes. The solubility profiles of sulfites will not necessarily show any analogy to those of sulfates.

All of the above, as well as a few further points such as sulfites autoxidising to sulfate, sulphide, elemental sulfur, thiosulfate and hydrogen sulfide, means that an experimentation would carry something of a risk of getting weird, complicated, stinky and generally annoying. Hydrogen sulfide is rather dangerously toxic and sulfur dioxide is particularly terrible for asthma sufferers.

If this hasn't entirely put you off, you might want to experiment by using sodium metabisulfite, which is sold as a sterilising agent for use in brewing and preserving. But again, it will emit sulfur dioxide if mixed with most other acids, even fairly weak ones.

How else would you plan on using sulfurous acid? I'd suggest that attempting to make your own by burning sulfur and dissolving the resulting sulfur dioxide may prove to be unnecessarily cumbersome, not to mention potentially dangerous.

 

[Prima Materia]

Sulfurous acid is a relatively weak acid which decomposes readily into toxic and foul-smelling sulfur dioxide gas. Both the acid and sulfur dioxide are fairly strong reducing agents, and bisulfites are well-known for forming adducts with various ketones or aldehydes. The solubility profiles of sulfites will not necessarily show any analogy to those of sulfates.

All of the above, as well as a few further points such as sulfites autoxidising to sulfate, sulphide, elemental sulfur, thiosulfate and hydrogen sulfide, means that an experimentation would carry something of a risk of getting weird, complicated, stinky and generally annoying. Hydrogen sulfide is rather dangerously toxic and sulfur dioxide is particularly terrible for asthma sufferers.

If this hasn't entirely put you off, you might want to experiment by using sodium metabisulfite, which is sold as a sterilising agent for use in brewing and preserving. But again, it will emit sulfur dioxide if mixed with most other acids, even fairly weak ones.

How else would you plan on using sulfurous acid? I'd suggest that attempting to make your own by burning sulfur and dissolving the resulting sulfur dioxide may prove to be unnecessarily cumbersome, not to mention potentially dangerous.

Thank you for the explanations, much appreciated.

 

[Transform]

Thank you for the explanations, much appreciated.

I'm also wondering whether bisulfite can form an adduct with harmaline, as it's technically a cyclic imine of an aromatic ketone. If this was to be the case, it could even be useful for separate harmaline from harmine.

I'm not sure if bisulfite is a strong enough reducing agent to form THH, but your idea has proven unexpectedly interesting given this further consideration. If you happened to have a pile of mixed harmala alkaloids, I'd suggest some careful experimentation using sodium metabisulfite.

If anyone should choose to follow this up, please keep in mind the safety issues outlined above!

 

[Prima Materia]

I'm also wondering whether bisulfite can form an adduct with harmaline, as it's technically a cyclic imine of an aromatic ketone. If this was to be the case, it could even be useful for separate harmaline from harmine.

I'm not sure if bisulfite is a strong enough reducing agent to form THH, but your idea has proven unexpectedly interesting given this further consideration. If you happened to have a pile of mixed harmala alkaloids, I'd suggest some careful experimentation using sodium metabisulfite.

If anyone should choose to follow this up, please keep in mind the safety issues outlined above!

I have got a pile of mixed harmalas and potassium metabisulfite (just regular brewing stuff I guess) at hand. I'm planing to get some glauber salt (sodium sulfate) to try out the harmala sulfate precipitation as well. Are there any further considerations I should keep in mind?

Will report here when something happens.

 

[Transform]

I have got a pile of mixed harmalas and potassium metabisulfite (just regular brewing stuff I guess) at hand. I'm planing to get some glauber salt (sodium sulfate) to try out the harmala sulfate precipitation as well. Are there any further considerations I should keep in mind?

Will report here when something happens.

Work out your molar ratios and you should be able to precipitate harmine while a putative harmaline bisulfite complex/adduct would possibly remain dissolved at a higher than usual pH.

I'd need to check up on the actual likelihood of an imine-bisulfite adduct even forming. The cyclic structure of harmaline, along with its steric hindrance and aromatic conjugation are all factors that could count against this hypothetical possibility.

If you end up mixing potassium bisulfite and sodium sulfate, there's a likelihood that potassium sulfate will crystallise out at lower temperatures. This is something to bear in mind if you get some lovely crystals.

Keep accurate records of everything!

 

[Jan e Kharabat]

I have got a pile of mixed harmalas and potassium metabisulfite (just regular brewing stuff I guess) at hand. I'm planing to get some glauber salt (sodium sulfate) to try out the harmala sulfate precipitation as well. Are there any further considerations I should keep in mind?

Will report here when something happens.

For doing Manske with harmala acetate and sodium sulfate, try to avoid excess acetic acid in your solution. The conversion would be more complete if there aren't many competing acetate ions.

 

[Transform]

For doing Manske with harmala acetate and sodium sulfate, try to avoid excess acetic acid in your solution. The conversion would be more complete if there aren't many competing acetate ions.

In a similar vein, use of sodium bisulfate (NaHSO₄) solution to dissolve harmala freebase would be a way of side-stepping the issue with acetic acid. And again, using only enough of this acidic salt to dissolve the harmalas, and no more, is likely to be desirable, pH-dependency notwithstanding.

Sodium bisulfate = Pool "pH down" crystals.

 

[Prima Materia]

Work out your molar ratios and you should be able to precipitate harmine while a putative harmaline bisulfite complex/adduct would possibly remain dissolved at a higher than usual pH.

I'd need to check up on the actual likelihood of an imine-bisulfite adduct even forming. The cyclic structure of harmaline, along with its steric hindrance and aromatic conjugation are all factors that could count against this hypothetical possibility.

That's interesting. If I understand correctly, your hypotesis is that upon mixing mixed harmala freebase with bisulfite, a harmaline-sulfite complex would form that would be more soluble at higher pH which would make it easier to precipitate only harmine when attempting the separation. Is that correct?

Or rather would one mix a bisulfite solution with mixed harmala freebase and possibly only dissolve the harmaline leaving harmine behind?

I believe that acidic pH is needed to start converting the bisulfite into SO2. Therefore this hypotetical reaction should be done at maybe slighty acidic to neutral to basic pH values.

FYI: above are just some speculations from a non-educated non-chemist.

 

[Transform]

That's interesting. If I understand correctly, your hypotesis is that upon mixing mixed harmala freebase with bisulfite, a harmaline-sulfite complex would form that would be more soluble at higher pH which would make it easier to precipitate only harmine when attempting the separation. Is that correct?

Or rather would one mix a bisulfite solution with mixed harmala freebase and possibly only dissolve the harmaline leaving harmine behind?

I believe that acidic pH is needed to start converting the bisulfite into SO2. Therefore this hypotetical reaction should be done at maybe slighty acidic to neutral to basic pH values.

FYI: above are just some speculations from a non-educated non-chemist.

This is indeed my hare-brained hypothesis for which I'd give odds of about 20% for it being viable. Good enough for a test-tube experiment, like I say, preferably preceded by a thorough search of the chemical literature.

Looking at it logically, imine formation is already energetically favourable, especially when stabilised by a six-membered ring as we have in harmaline. Conjugation of the imine double bond into the delocalised π-bonding of the indole aromatic system stabilises this formal imine of 2-acetylindole yet further. It becomes a question of balancing bond enthalpies - the formation of the carbon-sulfur bond versus the breaking of the carbon-nitrogen bond, the latter of which also involves shunting a negative charge onto the nitrogen atom. This last requirement is also disfavoured in comparison with sulfite addition to a plain carbonyl on account of nitrogen's lower electronegativity compared to oxygen.

I think tha harmaline would already need to be protonated for there to be any chance of sulfite addition, but as you point out, the pH mustn't get too low otherwise the bisulfite starts decomposing.

I kind of wish I hadn't eaten all my harmalas now, although making some more wouldn't be too difficult.

 

[Transform]

LLM output appears to confirm my suspicions:

[…] — I need the ESSENTIAL-INFO CHECKLIST before I can fully coach you on this:
- Area of focus (career, wellness, relationships)
- Current challenge
- Desired outcome and timeline

That said, you asked a technical chemistry question. I'll answer it concisely and helpfully now (scientific tone), then we can switch to coaching once you give the checklist.

Examples of bisulfite (HSO3−) adducts with imines and optimal formation conditions:

- **Existence**: **Yes — imines can form bisulfite adducts**, though they are generally less common and often less stable than carbonyl (aldehyde/ketone) bisulfite adducts. Adduct formation is favored with activated/iminium-like substrates and under conditions that protonate the imine to increase electrophilicity.

Representative examples and notes
- Iminium salts (e.g., N-protonated imines or stabilized iminium species) readily form bisulfite addition products; such adducts have been isolated and characterized in some studies.
- Imines conjugated to electron-withdrawing groups (e.g., tosyl- or sulfonyl-imines, or imines adjacent to carbonyls) show higher tendency to add bisulfite.
- Schiff bases derived from aromatic aldehydes (electron-poor) and primary amines can give isolable bisulfite adducts under protonating conditions, but these are more labile than aldehyde–bisulfite salts.

Optimal conditions for formation
- Solvent: **aqueous or mixed aqueous-organic (water, MeOH/H2O, EtOH/H2O)**. Bisulfite salts are water-soluble; some procedures use MeOH/H2O to improve substrate solubility.
- Bisulfite source: **sodium bisulfite (NaHSO3)** or **sodium metabisulfite (Na2S2O5, pre-dissolved)**. Typical stoichiometry: 1–2 equiv of HSO3− per imine, often using slight excess.
- pH/acid: **mildly acidic conditions** (pH ≈ 3–6) favor protonation of the imine to an iminium-like species and promote nucleophilic attack by HSO3−. Strongly basic conditions disfavor adduct formation. Buffering or gentle addition of acid (e.g., acetic acid) is commonly used.
- Temperature: **0–25 °C** usually sufficient; some sluggish substrates benefit from gentle warming (up to ~40 °C). Lower temperature can stabilize labile adducts.
- Time: reaction times from minutes to several hours depending on substrate and concentration. Monitoring by NMR or TLC recommended.
- Workup: bisulfite adducts are often water-soluble; isolate by removal of organic solvent and extraction, or characterize in situ. Reversal (liberation of the imine/aldehyde) is achieved by basification and extraction or by heating/acid workup depending on substrate.

Practical tips and cautions
- Imines that are freshly formed and not strongly conjugated are less likely to form stable bisulfite adducts — consider protonation or converting to an iminium salt first (e.g., add HCl or use a Lewis acid) to improve yield.
- Protecting groups on nitrogen (e.g., N-tosyl) that stabilize the imine can both help and hinder: they can make the carbon more electrophilic (promoting addition) but may also reduce reversibility.
- Analytical confirmation: 1H NMR (disappearance/shifts of imine CH), 13C NMR (change in C=N signal), and HRMS can confirm adduct formation. IR: loss/shift of C=N stretch.
- If isolation is desired, tailor solvent partitioning because many bisulfite adducts are water-soluble.

If you want, I can:
- Search the literature for specific published examples (e.g., named authors, substrate structures, characterization data).
- Propose an experimental procedure for a specific imine (give me its structure).

Accountability/next step (life-coach style): do you want me to find concrete literature examples and an experimental protocol for a particular imine within the next 24–48 hours? Answer yes/no and provide the imine structure if yes.

Encouragement mantra: You’ve got the curiosity — let’s pin the chemistry down like proper lab-grade mischief.

[ ]
Here's a suggested lab procedure, which I wouldn't entirely trust in some of the finer details (e.g., risk of indole polymerisation under acidic conditions, use of 'greater than' symbol when referring to pH with "greater acidity", i.e. "avoid pH > 3" should read "avoid pH < 3"):

[…] — thanks, that’s enough ESSENTIAL-INFO for this task (Area of focus: chemistry; Current challenge: make a harmaline–bisulfite adduct; Desired outcome & timeline: implicit — get conditions/protocol). I’m [LLM name] — here’s a focused, practical plan.

GOAL STATEMENT (SMART)
- Convert harmaline to a stable bisulfite adduct in lab-scale (50–200 mg) within 24–48 hours, achieving >50% isolated or high-conversion recoverable adduct suitable for characterization (NMR/HRMS).

ACTION PLAN — 3 bite-sized steps with timeline
1. Prepare reagents and solvent (0–1 hour)
- Dissolve harmaline (50–200 mg) in 3–5 mL methanol or ethanol. If solubility is poor, use 1:1 MeOH/H2O or EtOH/H2O.
- Prepare an aqueous solution of sodium bisulfite: dissolve NaHSO3 (1.5–3.0 equiv relative to harmaline) in minimal water (1–2 mL). If NaHSO3 is not available, dissolve Na2S2O5 (approx. 1.25 equiv) in water — it generates HSO3−.

2. Run the addition under mildly acidic conditions (0–24 hours)
- Cool the harmaline solution to ~0–5 °C (ice bath) to reduce side reactions; then add the aqueous bisulfite solution dropwise with stirring.
- Keep reaction pH mildly acidic: target pH ~4–5. If needed, add 0.1–0.2 equiv acetic acid to maintain this pH. Do not go strongly acidic (avoid > pH 3) which can hydrolyze or polymerize indolic systems.
- Stir at 0–5 °C for 1–2 hours, then allow to warm to 20–25 °C and stir for another 2–12 hours. Monitor by TLC (silica may be problematic for charged species — monitor disappearance of harmaline spot) and by 1H NMR (disappearance/shift of imine proton/changes in aromatic region) using small aliquots: extract aliquots into DMSO-d6 or CDCl3 after basification if needed for clearer spectra.

3. Workup and isolation (same day or next)
- Expect the adduct to be water-soluble. Remove organic solvent under reduced pressure. Partition between water and an organic solvent (EtOAc); the adduct will remain in aqueous layer.
- To isolate the adduct as a solid, concentrate the aqueous layer under reduced pressure (lyophilization preferred) or precipitate by adding a nonpolar solvent (e.g., addition of cold acetone or Et2O) if precipitation occurs.
- To regenerate harmaline for analysis, basify the aqueous solution (pH >9 with NaHCO3 or NaOH) and extract into DCM/EtOAc; dry and analyze.

OPTIMAL PARAMETER SUMMARY (quick reference — bold key facts)
- Solvent: **MeOH/H2O or EtOH/H2O (1:1)** for best substrate solubility and reactivity.
- Bisulfite source: **NaHSO3 (1.5–3.0 equiv)** or **Na2S2O5 (pre-dissolved)**.
- pH: **mildly acidic, ~4–5** (add catalytic AcOH if needed).
- Temperature: **0–5 °C initially, then ambient 20–25 °C**.
- Time: **total 4–24 h** (monitor by NMR).

Rationale and practical notes
- Harmaline is a relatively electron-rich indolic cyclic imine; direct addition of HSO3− to a neutral imine is slow. Protonation (mild acidity) increases electrophilicity of the C=N and favors nucleophilic attack by bisulfite.
- Keep conditions gentle to avoid indole ring protonation and side-reactions (polymerization, rearrangements). Very strong acid or heat risks decomposing the indolic scaffold.
- Use cold start to minimize competing addition to other positions (e.g., electrophilic attack on aromatic ring).
- If reaction is sluggish, convert harmaline to its iminium salt first (e.g., treat with 1 equiv HCl in EtOH to form harmaline·HCl) then add bisulfite solution — this can greatly increase adduct formation. Beware isolation complications with different counterions.
- Analytical confirmation: 1H NMR (loss/shift of the imine-like proton and upfield shifts for the adjacent carbon), 13C NMR (C=N shift), HRMS (adduct mass = harmaline + HSO3 − H), and possibly elemental analysis if isolating solid.

ACCOUNTABILITY PROMPT
- Will you try the cold addition protocol above or the iminium-salt-first variant? Reply “cold” or “iminium” and I’ll draft a step-by-step bench procedure with exact masses/volumes for a 100 mg trial.

Encouragement mantra
- You’ve got the lab chops and the curiosity — treat each tricky little adduct like a puzzle box and you’ll pop it open with a polite bit of chemistry and a stiff cup of determination.

(Yeah, the model is setup as a life coach )

 

[Transform]

I think we'd best take further discussion of the bisulfite experiment over to its own thread, so I don't end up spamming this thread with AI slop and ending up having to ban myself

I'm thinking of highlighting the role that LLMs could play in generating useful novel procedures when subjected to informed criticism.

 

[Prima Materia]

@Transform Great! Let the topic be moved then and save yourself of the dreaded ban!

 

[Transform]

@Transform Great! Let the topic be moved then and save yourself of the dreaded ban!

Well, here it is:

Developing a possible experimental harmaline separation method, with a little assistance from a LLM - "AI chemistry"

So we were chatting over in Stoney Broski's sulfate Manske thread and @Prima Materia asked about using sulfurous acid instead of sulfuric, which led me down a bit of an ideas rabbithole. While sulfurous acid is weak and unstable, bisulfites are a fairly readily available form of its salts. I...

forum.dmt-nexus.me

I might have to ban myself anyway

 

[Jan e Kharabat]

So... I was finally able to make pure white crystals, but it's hermine freebase, not acetate or sulfate. I did multiple recrystallizations with IPA. It is possible to completely remove the pigment and using multiple fractions, you can concentrate the impurities and calculate their percentage by weight. I didn't do any calculations but I think it must have been around 0.5% of the total weight (around 200mg out of 5g). Anyways, I'm quite surprised by the pretty needless, they look nice, but there's no practical need to purify and isolate to this extent.

 

[blig-blug]

So... I was finally able to make pure white crystals, but it's hermine freebase, not acetate or sulfate. I did multiple recrystallizations with IPA. It is possible to completely remove the pigment and using multiple fractions, you can concentrate the impurities and calculate their percentage by weight. I didn't do any calculations but I think it must have been around 0.5% of the total weight (around 200mg out of 5g). Anyways, I'm quite surprised by the pretty needless, they look nice, but there's no practical need to purify and isolate to this extent.

Beautiful crystals. When in salt form, do they remain white (even if not as a crystal)? I'm curious.

 

[Jan e Kharabat]

Beautiful crystals. When in salt form, do they remain white (even if not as a crystal)? I'm curious.

Yes, they remain white, the color definitely is not of harmine itself. However, the crystallization behavior changes as the color fades. Pure harmine becomes more difficult to neatly crystallize as compared to the reddish brown one, in terms of premature crystallization. I add equal volume of sodium sulfate and it crashes out too soon, I add 80% and nothing happens, then two days later due to some disturbance, it crashes out all at once. Then there's also the problem of sodium sulfate itself, it tends to make super saturated solutions which remain stable and liquid, and one touch turns the whole thing solid, meaning that you can't say how much sodium sulfate there is in a given "seemingly saturated" solution. And the sulfate salt of harmine still behaves better than the HCl, the HCl just makes pudding like blobs.
Anyways, with much pain, I was able to make fine needles of sulfate, very much colorless, but too fine. Also, the concentrations of sodium sulfate that work best for crystallization, don't crystallize everything. In one case something even weirder happened, I had added an overly super saturated solution of sodium sulfate and left the solution to crystallize over a week. The needles were fine and pretty, but when I filtered them, I found quartz like chunks of sodium sulfate buried within the harmine crystals. Weirder still, the harmine crystallization was still not complete, and the mother liquor clouded again upon adding NaOH. The amount of freebase weighed about 20-30% of the total amount. Also, even if you get it all right, the excess salt in the final product is still a pain, and the excess sodium sulfate leads to a more powdery texture of the crystal surface, which I also don't like.
So, to sum it all up...
a) the "impurities" contribute a good deal towards the structure of the crystals, making them more stable, the pure crystals are much more delicate.
b) the "impurities" also allow for a more stable and cleaner crystallization.
c) sodium sulfate has its pros and cons, it works well and produces prettier crystals when the harmine is sufficiently impure, the crystallization process becomes more unpredictable with pure harmalas. The concentration of harmine also has a significant effect. With high enough concentration, it crashes out with the addition of even 70-80% by volume of saturated sodium sulfate solution. If the concentration is low enough, nothing happens, and if you further increase the concentration of sodium sulfate, it begins to crash out along with, or even before harmine.
d) The sulfate salt of harmine doesn't crystallize well via slow evaporation of aqueous solution, same goes for HCl as well.
e) the fluorescence of pure harmine is actually deep blue, very much or perhaps the same as that of THH, the greenish fluorescence of the "impure" harmine likely comes from the "impurities".
f) Adding NaCl to harmine sulfate solution gives you harmine HCl ( I had hoped that the sulfate salt would crash out). I concluded this from the appearance of the resulting crystals, so I could be wrong about it. It is also quite likely that the result would be a mixture of sulfate and HCl.
Anyways, I'm tired of this project, I'll wrap this up soon and continue on with harmaline next. May be a thread of it's own would be good for this topic.

 

[modern]

I have returned my interest in playing with harmalas again and want to try different acids. Here is the Merck index reference on these... when pure freebase slow crystallization in HOT ethanol seems can make some pretty crystals. Mainly for harmaline but harmine doesn't seem too different. “rhombic octahedra crystals”

Rereading this I think the red crystals I got by adding to a heavy citric acid solution may be HARMALOL "red crystals from water"... It is a small amount compared to the rest of the crystals I got from my caapi.

Lactate monohydrate Also sublimate harmine possible it seems but at 260C need vacuum... maybe one day I'll mess around with vacuum distillation.