-
Members of the previous forum can retrieve their temporary password here, (login and check your PM).
~Phalaris = The Way Of The Future~
aizoaceous
Titanium Teammate
As I wait for my larger crop, I was able to scrounge another 106 g of P. brachystachys, mostly outdoor-grown. I used this for another batch with an improved method, ending in crystals of nearly gramine-free 5-MeO-DMT oxalate.
Extraction of Grass into Cold Water
As usual, the grass was frozen for storage, then chopped coarsely and placed in a 500 mL beaker. In a first pull, I simply covered it with 242 g of DI water and stirred by hand for a minute or so. No acid was used, but natural pH was 6.0 so that didn't seem obviously necessary. The grass was initially still frozen, so temperature was just above 0 C. This yielded 43% of my final total. I then sonicated for 20 min in an ice bath, increasing the yield in that first pull to 53% of total.
I did a second pull, also in an ice bath, this time with 200 mg citric acid, reducing pH to 3.9, and again sonicated. This yielded 27% of total.
I did a third pull, this time at 50 C, with 100 mg citric acid reducing pH to 3.6. This yielded 14% of total. A fourth identical pull yielded 6% of total.
All the pulls were light green in color. In the first two, fine green sediment was easily distinguishable at the bottom, perhaps implying that the chlorophyll and other fats came off in bigger pieces at lower temperature? I combined the first three pulls and tried to filter under vacuum, but the filtrate was still green and the paper clogged. I tried a few mm of Celite 545, and for once that worked--the filtrate went from cloudy green to a beautiful pale red-brown, crystal clear.
I reduced to 85 mL. This precipitated some light brown stuff with a cottage cheese consistency, but that stuck together in clumps and the solution remained clear. After three days in the fridge (which I'd guess were unnecessary; I just was busy), that was easily filtered.
I'm pleased to have separated off those fats, which were going into an emulsion for me before. I suspect those are much easier to separate before reduction. (I made cacio e pepe last night, confirming that vigorous boiling is a great way to form and stabilize an emulsion.) I need to experiment more to determine what combination of the cold initial extraction vs. filtration accounts for this improvement.
Extraction of Water into Limonene
I titrated with solid potassium carbonate to pH 11.1. The solution turned darker and cloudy. I stirred it to suspend the solids as uniformly as possible and took a 20 uL sample for analysis, which I added to 1 mL of mobile phase and 80 uL of 10% citric acid (since the eluent is acidic, but weakly enough that the large amount of potassium carbonate overcomes that). This showed 83 mg 5-MeO-DMT, plus the usual smaller areas of DMT and gramine.
I extracted once with 25 mL limonene and re-tested the water, which dropped to 6.4 mg, 8% of initial. I extracted again with another 25 mL, and it dropped to 0.6 mg, 1%.
I combined the two limonene pulls, and extracted with 20 mL DI water. This initially went to pH 9.6, and pulled about 4 mg 5-MeO-DMT. (I was initially undecided as to whether I was washing or extracting; but that pulled enough I chose "extracting"). I titrated with oxalic acid, eventually reaching pH 4.5 with 30 mg acid. This would be about right for 5-MeO-DMT oxalate dibasic, so I stopped there. A sample of the water showed 69 mg 5-MeO-DMT.
The extraction from limonene back into water proceeds rather slowly. I'd thought I was done when I reached my target pH after a few minutes of stirring, but five minutes later it had risen back to 7.8. If mechanical stirring is unavailable then this step will require some endurance. The initial extraction from water into limonene goes much faster. I was stirring pretty vigorously, with a big vortex and an apparently homogenous mixture. Since filtering off the fats per above, I've had zero trouble with emulsions, with clear separation within seconds of stopping stirring and a perfect interface after just a few minutes.
I pulled the limonene with another 20 mL water. pH fell to 2.9 with just 5 mg oxalic acid, suggesting very little alkaloids remained. A sample of that water showed 3 mg 5-MeO-DMT.
I evaporated the first water pull on a 75 C water bath under an air stream, yielding an orange-yellow resin with slight odor.
I redissolved that in 99% food-grade ethanol, which was cloudy even with 7 mL solvent and long sonication. I let that settle and decanted, leaving a very thin film of sediment. Potassium oxalate is reported to be insoluble in ethanol, so maybe some potassium carbonate carried over?
A sample of this ethanolic solution shows 71 mg 5-MeO-DMT. Comparing the chromatograms of this product and the initial aqueous extract, we see that our gramine is depleted (versus 5-MeO-DMT), but not significantly so. DMT is significantly depleted, as expected from my smaller-scale experiments earlier. If we were targeting DMT, then this method would actually enrich our extract in gramine, the wrong direction.
I like the limonene, though. The toxic and fire hazards aren't zero, but they're much lower than other options. It's volatile enough that my product has no significant odor. It should be reusable too, though I haven't tried that yet.
Crystallization
I reduced the ethanolic solution to about 0.5 mL, then left it to cool and evaporate at room temperature. This resulted in the same resin as I'd obtained when I evaporated the water, no crystals.
I tried redissolving in isopropanol, but it was only slightly soluble (about 3 mg/mL 5-MeO-DMT) so I abandoned that and evaporated. I redissolved in 1 mL ethanol, and titrated with a 5% solution of oxalic acid in water to pH 2.0, 800 uL for an additional 40 mg acid. I evaporated this to dryness at 50 C, and got the usual resin.
I'd intended to redissolve that in ethanol and slowly evaporate, so I added 0.5 mL and sonicated; but instead the whole thing abruptly formed tiny crystals, I guess since the ethanol instead acted as an antisolvent.
I added 3 mL more ethanol, but very little dissolved, in contrast to the easy solubility at higher pH. I added 2 mL methanol but it still didn't dissolve. I evaporated off the mixed alcohols at 50 C, and the alkaloids returned to the amorphous yellow resin. I added pure methanol, just barely redissolving all of it in 2 mL with sonication at 50 C.
I allowed the methanolic solution to cool and evaporate at room temperature for about five hours, after which I observed larger crystals at the bottom.
A sample of the supernatant still showed about half the alkaloids in solution. I added 99.9% isopropanol dropwise, and saw rapid crystallization with each drop. That stopped a little before 2 mL. I carefully syringed off the supernatant, washed the crystals in 2 mL ethyl acetate, syringed that off, and dried the crystals, obtaining 53 mg. The product lost most of its luster after washing and drying, but was still visibly crystalline.
I dissolved the crystals in water, obtaining pH around 3.0, roughly as expected for the monobasic salt of oxalic acid. This took considerable effort, about ten minutes sonicating at 50 C. The gramine and DMT are both depleted to almost nothing. Comparing against my standard, the peak area implies 45 mg 5-MeO-DMT. I've unfortunately forgotten whether my standard was a freebase or salt, so I need to go back and check if that's reasonable; but regardless, I think this looks good.
I'd taken samples of the supernatant after this crystallization from methanol, and then also after adding isopropanol. They showed increasing enrichment in gramine, consistent with a pure precipitate. I also sampled the ethyl acetate wash, and it showed negligible alkaloids. Most of the remaining color was removed with the ethyl acetate.
I'd saved my supernatant and it continued to crystallize, so I probably should have waited longer here (unless that was gramine...). I evaporated that to dryness at 50 C, and it returned to the resin. I redissolved in 2 mL methanol and left it to crystallize overnight. That was too long, and the solvent evaporated completely; but the residue looked crystalline. I tried to suspend that in ethyl acetate, but the whole thing returned to resin. Nothing visibly dissolved with scraping or sonication, and the solvent didn't turn yellow; so I don't think anything good was happening. I evaporated that, and redissolved in methanol. I'll keep experimenting and see what I can learn from that remaining product.
Conclusions
I believe that:
Smoking the oxalate seems like an unusually bad idea, but the gramine-free salt could of course be returned to freebase. The pH seems unpleasantly low for sublingual too, so I might try that next, following Ott and @neurobloom. I guess I could also just titrate up with baking soda. I recently tried some unfermented kanna, which is high in natural oxalates (~0.5% fresh weight) around pH 5, with no irritation. This extract would be a lower total dose of oxalate, though about 3x that concentration.
As to subjective effects, I took the last of my previous batch in doses of ~10 mg 5-MeO-DMT with a similar amount of mixed harmala alkaloids sublingually. I felt the potentiation that Ott reported, a much stronger euphoria, intensely and purely physical. The psychological effects are harder to articulate, but certainly present. It's a less popular drug than DMT for good reason; but since I'm currently well-equipped to study it, I plan to proceed.
This method won't work well for DMT, since the limonene depletes that significantly. A similar method might work, with more and hotter limonene pulls, and a different salt and solvents for crystallization.
Edit: Simplify discussion of crystallization, which was rather confused before.
Extraction of Grass into Cold Water
As usual, the grass was frozen for storage, then chopped coarsely and placed in a 500 mL beaker. In a first pull, I simply covered it with 242 g of DI water and stirred by hand for a minute or so. No acid was used, but natural pH was 6.0 so that didn't seem obviously necessary. The grass was initially still frozen, so temperature was just above 0 C. This yielded 43% of my final total. I then sonicated for 20 min in an ice bath, increasing the yield in that first pull to 53% of total.
I did a second pull, also in an ice bath, this time with 200 mg citric acid, reducing pH to 3.9, and again sonicated. This yielded 27% of total.
I did a third pull, this time at 50 C, with 100 mg citric acid reducing pH to 3.6. This yielded 14% of total. A fourth identical pull yielded 6% of total.
All the pulls were light green in color. In the first two, fine green sediment was easily distinguishable at the bottom, perhaps implying that the chlorophyll and other fats came off in bigger pieces at lower temperature? I combined the first three pulls and tried to filter under vacuum, but the filtrate was still green and the paper clogged. I tried a few mm of Celite 545, and for once that worked--the filtrate went from cloudy green to a beautiful pale red-brown, crystal clear.
I reduced to 85 mL. This precipitated some light brown stuff with a cottage cheese consistency, but that stuck together in clumps and the solution remained clear. After three days in the fridge (which I'd guess were unnecessary; I just was busy), that was easily filtered.
I'm pleased to have separated off those fats, which were going into an emulsion for me before. I suspect those are much easier to separate before reduction. (I made cacio e pepe last night, confirming that vigorous boiling is a great way to form and stabilize an emulsion.) I need to experiment more to determine what combination of the cold initial extraction vs. filtration accounts for this improvement.
Extraction of Water into Limonene
I titrated with solid potassium carbonate to pH 11.1. The solution turned darker and cloudy. I stirred it to suspend the solids as uniformly as possible and took a 20 uL sample for analysis, which I added to 1 mL of mobile phase and 80 uL of 10% citric acid (since the eluent is acidic, but weakly enough that the large amount of potassium carbonate overcomes that). This showed 83 mg 5-MeO-DMT, plus the usual smaller areas of DMT and gramine.
I extracted once with 25 mL limonene and re-tested the water, which dropped to 6.4 mg, 8% of initial. I extracted again with another 25 mL, and it dropped to 0.6 mg, 1%.
I combined the two limonene pulls, and extracted with 20 mL DI water. This initially went to pH 9.6, and pulled about 4 mg 5-MeO-DMT. (I was initially undecided as to whether I was washing or extracting; but that pulled enough I chose "extracting"). I titrated with oxalic acid, eventually reaching pH 4.5 with 30 mg acid. This would be about right for 5-MeO-DMT oxalate dibasic, so I stopped there. A sample of the water showed 69 mg 5-MeO-DMT.
The extraction from limonene back into water proceeds rather slowly. I'd thought I was done when I reached my target pH after a few minutes of stirring, but five minutes later it had risen back to 7.8. If mechanical stirring is unavailable then this step will require some endurance. The initial extraction from water into limonene goes much faster. I was stirring pretty vigorously, with a big vortex and an apparently homogenous mixture. Since filtering off the fats per above, I've had zero trouble with emulsions, with clear separation within seconds of stopping stirring and a perfect interface after just a few minutes.
I pulled the limonene with another 20 mL water. pH fell to 2.9 with just 5 mg oxalic acid, suggesting very little alkaloids remained. A sample of that water showed 3 mg 5-MeO-DMT.
I evaporated the first water pull on a 75 C water bath under an air stream, yielding an orange-yellow resin with slight odor.
I redissolved that in 99% food-grade ethanol, which was cloudy even with 7 mL solvent and long sonication. I let that settle and decanted, leaving a very thin film of sediment. Potassium oxalate is reported to be insoluble in ethanol, so maybe some potassium carbonate carried over?
A sample of this ethanolic solution shows 71 mg 5-MeO-DMT. Comparing the chromatograms of this product and the initial aqueous extract, we see that our gramine is depleted (versus 5-MeO-DMT), but not significantly so. DMT is significantly depleted, as expected from my smaller-scale experiments earlier. If we were targeting DMT, then this method would actually enrich our extract in gramine, the wrong direction.
I like the limonene, though. The toxic and fire hazards aren't zero, but they're much lower than other options. It's volatile enough that my product has no significant odor. It should be reusable too, though I haven't tried that yet.
Crystallization
I reduced the ethanolic solution to about 0.5 mL, then left it to cool and evaporate at room temperature. This resulted in the same resin as I'd obtained when I evaporated the water, no crystals.
I tried redissolving in isopropanol, but it was only slightly soluble (about 3 mg/mL 5-MeO-DMT) so I abandoned that and evaporated. I redissolved in 1 mL ethanol, and titrated with a 5% solution of oxalic acid in water to pH 2.0, 800 uL for an additional 40 mg acid. I evaporated this to dryness at 50 C, and got the usual resin.
I'd intended to redissolve that in ethanol and slowly evaporate, so I added 0.5 mL and sonicated; but instead the whole thing abruptly formed tiny crystals, I guess since the ethanol instead acted as an antisolvent.
I added 3 mL more ethanol, but very little dissolved, in contrast to the easy solubility at higher pH. I added 2 mL methanol but it still didn't dissolve. I evaporated off the mixed alcohols at 50 C, and the alkaloids returned to the amorphous yellow resin. I added pure methanol, just barely redissolving all of it in 2 mL with sonication at 50 C.
I allowed the methanolic solution to cool and evaporate at room temperature for about five hours, after which I observed larger crystals at the bottom.
A sample of the supernatant still showed about half the alkaloids in solution. I added 99.9% isopropanol dropwise, and saw rapid crystallization with each drop. That stopped a little before 2 mL. I carefully syringed off the supernatant, washed the crystals in 2 mL ethyl acetate, syringed that off, and dried the crystals, obtaining 53 mg. The product lost most of its luster after washing and drying, but was still visibly crystalline.
I dissolved the crystals in water, obtaining pH around 3.0, roughly as expected for the monobasic salt of oxalic acid. This took considerable effort, about ten minutes sonicating at 50 C. The gramine and DMT are both depleted to almost nothing. Comparing against my standard, the peak area implies 45 mg 5-MeO-DMT. I've unfortunately forgotten whether my standard was a freebase or salt, so I need to go back and check if that's reasonable; but regardless, I think this looks good.
I'd taken samples of the supernatant after this crystallization from methanol, and then also after adding isopropanol. They showed increasing enrichment in gramine, consistent with a pure precipitate. I also sampled the ethyl acetate wash, and it showed negligible alkaloids. Most of the remaining color was removed with the ethyl acetate.
I'd saved my supernatant and it continued to crystallize, so I probably should have waited longer here (unless that was gramine...). I evaporated that to dryness at 50 C, and it returned to the resin. I redissolved in 2 mL methanol and left it to crystallize overnight. That was too long, and the solvent evaporated completely; but the residue looked crystalline. I tried to suspend that in ethyl acetate, but the whole thing returned to resin. Nothing visibly dissolved with scraping or sonication, and the solvent didn't turn yellow; so I don't think anything good was happening. I evaporated that, and redissolved in methanol. I'll keep experimenting and see what I can learn from that remaining product.
Conclusions
I believe that:
- With careful filtration (or other mechanical separation, maybe a week or so settling), chlorophyll and other plant fats can be removed from the aqueous extract before boiling to reduce. This fixes all trouble with emulsions, and improves the color from dark brown to orange-yellow. I don't know whether this is necessary for the subsequent crystallization.
- Room-temperature limonene effectively extracts 5-MeO-DMT from water. Gramine is depleted slightly, but not usefully so. DMT is extracted less effectively, significantly depleted with respect to 5-MeO-DMT or gramine.
- Oxalic acid will indeed crystallize 5-MeO-DMT from Phalaris extract. This may occur only at pH ~ 2.0, I guess implying crystallization only as the monobasic salt. I can't find anything useful in the literature on this.
- At this low pH, the 5-MeO-DMT oxalate is somewhat soluble in methanol and water. It's not very soluble in ethanol, isopropanol, or ethyl acetate. Oxalic acid is quite soluble in the latter three, so excess free acid should be effectively removed if those are used to wash or as an antisolvent. Most of the color is also removed.
- This crystallization separates the gramine, because gramine oxalate is some combination of more soluble and less initially abundant.
- A simpler crystallization might also work, like precipitation directly from the limonene, perhaps followed by washing in ethyl acetate. I just didn't want to try that first, since it's more trouble to recover from if it fails. If anyone wants a project, then TLC is sufficient to judge success.
- I'll try crystallizing more until I get gramine, and that should give a sense of how much initial gramine we can tolerate.
Smoking the oxalate seems like an unusually bad idea, but the gramine-free salt could of course be returned to freebase. The pH seems unpleasantly low for sublingual too, so I might try that next, following Ott and @neurobloom. I guess I could also just titrate up with baking soda. I recently tried some unfermented kanna, which is high in natural oxalates (~0.5% fresh weight) around pH 5, with no irritation. This extract would be a lower total dose of oxalate, though about 3x that concentration.
As to subjective effects, I took the last of my previous batch in doses of ~10 mg 5-MeO-DMT with a similar amount of mixed harmala alkaloids sublingually. I felt the potentiation that Ott reported, a much stronger euphoria, intensely and purely physical. The psychological effects are harder to articulate, but certainly present. It's a less popular drug than DMT for good reason; but since I'm currently well-equipped to study it, I plan to proceed.
This method won't work well for DMT, since the limonene depletes that significantly. A similar method might work, with more and hotter limonene pulls, and a different salt and solvents for crystallization.
Edit: Simplify discussion of crystallization, which was rather confused before.
Last edited:
Grasshoppers
Titanium Teammate
First off, I want to acknowledge the level of detail and precision in your work. The process you’ve described for cleaning the plant extract is impressive and demonstrates your deep expertise in analytics. It’s clear that this approach yields high-quality extracts, ensures purity, and provides a reliable way to control dosage.
That said, I wanted to share a slightly different perspective that we’ve been exploring. From what we’ve observed, there’s significant natural variation within Phalaris subspecies in terms of their biochemical profiles. Some wild populations already contain potent individuals that are low in undesirable constituents. Our thinking is that by carefully selecting and breeding these plants, we could potentially reduce or even eliminate the need for cleaning procedures.
In other words, with the right selection, the natural diversity within the Phalaris might offer cleaner, more potent plants from the start. Selection could pave the way for a more efficient approach enabling traditional ayahuasca preparations.
Certain varieties described in the literature—such as cv. Australian, cv. Australian II, cv. Uneta, and older strains like cv. Stenoptera already show promise. Additionally, wild samples from the Mediterranean region appear to hold great potential. Your expertise in analysis would be invaluable in identifying and verifying these naturally cleaner, highly potent individual plants.
That said, I wanted to share a slightly different perspective that we’ve been exploring. From what we’ve observed, there’s significant natural variation within Phalaris subspecies in terms of their biochemical profiles. Some wild populations already contain potent individuals that are low in undesirable constituents. Our thinking is that by carefully selecting and breeding these plants, we could potentially reduce or even eliminate the need for cleaning procedures.
In other words, with the right selection, the natural diversity within the Phalaris might offer cleaner, more potent plants from the start. Selection could pave the way for a more efficient approach enabling traditional ayahuasca preparations.
Certain varieties described in the literature—such as cv. Australian, cv. Australian II, cv. Uneta, and older strains like cv. Stenoptera already show promise. Additionally, wild samples from the Mediterranean region appear to hold great potential. Your expertise in analysis would be invaluable in identifying and verifying these naturally cleaner, highly potent individual plants.
aizoaceous
Titanium Teammate
I see improved genetics and improved extraction as entirely complementary (and anyone with an interest in Phalaris should be following your work; your collection seems large and diverse even by the standards of commercial or academic breeding, perhaps the largest in the world targeting high rather than low alkaloids). No method can extract what isn't present, so higher tryptamines are always better. The unwanted alkaloid oxalates also aren't infinitely soluble, so that content limits the pure yield achievable when crystallizing as above.
I'm concerned that my post above makes the extraction look more difficult than it is. Almost all of the steps that I list above are useless or worse; I just gave a full description for the benefit of anyone else attempting method development. I'll aim to post an optimized method once I've run it a few times, and I don't think it should be much more complicated than cactus, MHRB, etc. Those have naturally clean alkaloid profiles (at least by Phalaris standards), but most people still prefer to consume them as the purified extract.
P. aquatica seems like a more attractive breeding target than P. brachystachys, perhaps due to greater heterozygosity from the lower degree of self-compatibility that we discussed above. I feel like P. brachystachys from seed is the most expedient way to grow a useful amount of 5-MeO-DMT, while a selected P. aquatica clone holds the greatest long-term promise.
A useful source of pure DMT is a greater challenge. Crystallization as above would remove small amounts of 5-MeO-DMT, but it's hard to judge how far to evaporate without at least TLC. Smoking is lower risk, since an experienced user can titrate the dose. I'd want pretty long experience before I took unanalyzed ayahuasca from allegedly DMT-dominant Phalaris, though. I don't think the lethal dose of 5-MeO-DMT is known very accurately. The 2005 decedent had 201.6 mg/L in his stomach contents, but that's hard to convert to a dose. I've seen speculation that probably involved synthetic 5-MeO-DMT, since traditional plant sources don't contain nearly so much; but strong extracts may present the same risk.
A DMT salt with much lower solubility than the corresponding 5-MeO-DMT salt would be very helpful here. I feel like 5-MeO-DMT's reputation as difficult to crystallize may be a sign for optimism as to that, and I might screen some acids and solvents. If anyone has relevant literature (or justification for a guess) then I'd love to see it.
neurobloom
Established member
Tired my first vaporized aq (CV tanit ) freebase extract + harmala acetate combined together in one hit. I used a low dose aq extact that normally would bring about threshold effects alone but was shocked at how powerful the same dose felt with a low dose harmala. The two have potentiated each other extremely well I had to quickly exhale prematurely to avoid a hard lunch Into a different galaxy.
Effects also lasted a lot longer than usual and brought more nausea but besides the powerful intensity from a low dose for this combination it felt safe and the harmala landed a new quality to the experience that I really enjoyed. The harmala helped with the 5-meo-dmt onset anxiety and even my heart rate didn't increase as it would normally does with 5-meo-dmt vaporized extracts alone.
Overall it's a nice combination but it poses more risk for overdose if no accurate scale is used. I still don't recommend this combination but in my personal experience it seems fine it's just a matter of dosage.. have to be really careful how much you add to your vaped...it's ridiculous how little you need when you mix these two together ..I think 3mg worth of pure 5-meo-dmt with a low dose harmala is already potent enough for a medium to above medium dose.
Below I attached a pic of how much I need of the aq freebase alone to get a sub breakthrough dose. I used maybe 1/4th of that with harmala and got stronger effects then the whole amount in the picture.
Effects also lasted a lot longer than usual and brought more nausea but besides the powerful intensity from a low dose for this combination it felt safe and the harmala landed a new quality to the experience that I really enjoyed. The harmala helped with the 5-meo-dmt onset anxiety and even my heart rate didn't increase as it would normally does with 5-meo-dmt vaporized extracts alone.
Overall it's a nice combination but it poses more risk for overdose if no accurate scale is used. I still don't recommend this combination but in my personal experience it seems fine it's just a matter of dosage.. have to be really careful how much you add to your vaped...it's ridiculous how little you need when you mix these two together ..I think 3mg worth of pure 5-meo-dmt with a low dose harmala is already potent enough for a medium to above medium dose.
Below I attached a pic of how much I need of the aq freebase alone to get a sub breakthrough dose. I used maybe 1/4th of that with harmala and got stronger effects then the whole amount in the picture.
Attachments
MagyarHungary1111
Rising Star
Well done !
aizoaceous
Titanium Teammate
Crystallization Update
In my previous extraction, I'd yielded 71 mg 5-MeO-DMT, from which I crystallized 45 mg of gramine-free 5-MeO-DMT. This meant there was still 26 mg in solution, so I tried crystallizing further. As mentioned previously, I'd evaporated that to dryness and redissolved in 1 mL methanol. The following day, I saw crystals in perhaps 100 uL solvent. I added 1 mL absolute food-grade ethanol as an antisolvent, and didn't observe further crystallization. I swirled gently, syringed off the liquid, and let the crystals dry.
This yielded an additional 10 mg, but it contained gramine, slightly below the crude extract's ratio. So that gives a rough idea of how far we can push the crystallization while still purifying. I don't yet see how to exactly judge that stopping point without chromatography, though TLC would do the job easily.
The ethanolic solution that I'd syringed off eventually crystallized further, so I'd been impatient. I think those will contain gramine, though.
I also tried crystallizing the oxalate from water, by redissolving my original crystals in 1 mL. This took extended effort (by sonication for 8 min at 50 C), but the solution eventually went clear. I then left it to cool and evaporate. This took a couple weeks, but crystals did eventually appear. I saw no crystallization until the water was almost completely gone. I therefore suspect that 5-MeO-DMT oxalate is significantly more soluble in water than in methanol, or prone to supersaturation or something. This is probably undesired when attempting to purify by crystallization (since the window between dissolving the product and crystallizing the impurities is thus very small), though perhaps a method could be found. For now I'll probably stick with the methanol, accepting the toxicity in exchange for quicker and better separation.
Solvent Choice, New Study
I did more small-scale extractions with multiple different solvents. I used a mixture of Phalaris aquatica and P. brachystachys extracts in acidified water. This gave strong peaks of gramine, DMT, and 5-MeO-DMT, plus some unidentified peaks from the P. aquatica. All detection below is by fluorescence, in the same chromatographic system described earlier. Additional unidentified peaks were present by absorption, and not everything eluted.
For analytical convenience I ran backwards from before, measuring the depletion of alkaloids out of the original aqueous extract, rather than enrichment of alkaloids into the nonpolar solvent (since many of those solvents can't be injected directly into my HPLC). I also worked at larger scale to improve accuracy, enabling absolute comparisons of the same peak across different solvents, and not just relative comparisons of different peaks within the same solvent like before.
To each of six 15 mL tubes, I added 5 mL of the acidic aqueous extract. I made each tube basic with 1 mL concentrated ammonia. I processed dichloromethane, ethyl acetate, d-limonene, and naphtha at room temperature. I processed pharmaceutical mineral oil and an additional naphtha on a 50 C water bath. To each tube I added 2 mL of its solvent. I capped the tube, and shook vigorously to mix. I centrifuged to break any emulsion, then removed the solvent with a glass syringe and discarded it. I stirred with a spatula to resuspend any sediment. I repeated the liquid-liquid extraction a second time. So I pulled each tube with its solvent twice, and discarded both pulls.
I then made each tube acidic with 6 mL 10% citric acid, resuspended, shook to mix, and centrifuged. I diluted that water into mobile phase and injected. I've superimposed those chromatograms to produce this figure:
These are chromatograms of the water after liquid-liquid extraction. So the presence of a peak means it wasn't extracted into the solvent, so a flatter chromatogram is a stronger solvent, backwards from before.
Dichloromethane and ethyl acetate are almost perfectly flat, showing that they pulled essentially all the alkaloids. Limonene is a little higher, but not by much. Hot naphtha is much higher. Hot mineral oil and room-temperature naptha are yet higher and similar. That seems like the expected order to me. The mineral oil worked better than I'd expected, though it's probably impractically viscous.
I should have run a control tube that got basified, centrifuged, and acidified but not extracted. I didn't, so I've included a chromatogram of the original aqueous extract for the starting reference. I've scaled that by a factor of 5/12, to account for the dilution from the additional 6+1 mL added to the tubes.
I plotted the same data, scaling each trace vertically to make the 5-MeO-DMT peaks the same height:
We see that the gramine peaks are very close to the same height, implying that gramine was pulled in roughly equal proportion. I've hidden the DCM, ethyl acetate, and limonene traces here since they're too small in absolute terms, in the noise.
Scaling to the DMT peak, we see enrichment of gramine in the water:
This implies depletion from the solvent, the good direction. Take care that the depletion ratio in the solvent can't be calculated from this curve alone; we need the absolute change from the starting point too, and relative error gets amplified when we subtract. I don't think my chromatography is good enough to justify a real calculation, but the ratio of DMT to gramine is probably improved by a factor of less than two. More pulls will worsen that ratio.
The limonene pulled DMT more effectively than I expected from the extraction that I crystallized. After more careful review of the samples I took then, I think the DMT did actually make it into the limonene, but roughly half stayed in the limonene when I extracted back into water, under the same conditions that pulled almost all of the 5-MeO-DMT. I think the rest got separated when I evaporated, redissolved into ethanol, and filtered. (So would washing DMT oxalate with ethanol remove 5-MeO-DMT impurities? Obviously don't trust any such separation without extensive verification.)
I don't think the gramine is particularly harmful in my application. If it's undesired then choice among these solvents under these conditions provides limited benefit in rejecting it, though. It's possible that other solvents or other conditions would reject better. My solutions were somewhat weak (around 0.5 mg/mL alkaloids), and stronger might behave differently.
Based on current knowledge, I would choose my solvent primarily based on factors other than selectivity against unwanted alkaloids. Limonene seems like the winner to me--cheap, almost perfectly nontoxic, flash point (just) above room temperature, pretty good absolute strength. It was also much less prone to emulsions than any other solvent, forming a sharp interface with no centrifuging.
My Phalaris brachystachys is growing outdoors, and I'll aim to extract a larger amount once I can harvest that. I'll perhaps then try to crystallize as the fumarate or succinate, since those are friendlier. This could possibly also improve the gramine-free yield if I got lucky on the relative solubilities, though I doubt it.
In my previous extraction, I'd yielded 71 mg 5-MeO-DMT, from which I crystallized 45 mg of gramine-free 5-MeO-DMT. This meant there was still 26 mg in solution, so I tried crystallizing further. As mentioned previously, I'd evaporated that to dryness and redissolved in 1 mL methanol. The following day, I saw crystals in perhaps 100 uL solvent. I added 1 mL absolute food-grade ethanol as an antisolvent, and didn't observe further crystallization. I swirled gently, syringed off the liquid, and let the crystals dry.
This yielded an additional 10 mg, but it contained gramine, slightly below the crude extract's ratio. So that gives a rough idea of how far we can push the crystallization while still purifying. I don't yet see how to exactly judge that stopping point without chromatography, though TLC would do the job easily.
The ethanolic solution that I'd syringed off eventually crystallized further, so I'd been impatient. I think those will contain gramine, though.
I also tried crystallizing the oxalate from water, by redissolving my original crystals in 1 mL. This took extended effort (by sonication for 8 min at 50 C), but the solution eventually went clear. I then left it to cool and evaporate. This took a couple weeks, but crystals did eventually appear. I saw no crystallization until the water was almost completely gone. I therefore suspect that 5-MeO-DMT oxalate is significantly more soluble in water than in methanol, or prone to supersaturation or something. This is probably undesired when attempting to purify by crystallization (since the window between dissolving the product and crystallizing the impurities is thus very small), though perhaps a method could be found. For now I'll probably stick with the methanol, accepting the toxicity in exchange for quicker and better separation.
Solvent Choice, New Study
I did more small-scale extractions with multiple different solvents. I used a mixture of Phalaris aquatica and P. brachystachys extracts in acidified water. This gave strong peaks of gramine, DMT, and 5-MeO-DMT, plus some unidentified peaks from the P. aquatica. All detection below is by fluorescence, in the same chromatographic system described earlier. Additional unidentified peaks were present by absorption, and not everything eluted.
For analytical convenience I ran backwards from before, measuring the depletion of alkaloids out of the original aqueous extract, rather than enrichment of alkaloids into the nonpolar solvent (since many of those solvents can't be injected directly into my HPLC). I also worked at larger scale to improve accuracy, enabling absolute comparisons of the same peak across different solvents, and not just relative comparisons of different peaks within the same solvent like before.
To each of six 15 mL tubes, I added 5 mL of the acidic aqueous extract. I made each tube basic with 1 mL concentrated ammonia. I processed dichloromethane, ethyl acetate, d-limonene, and naphtha at room temperature. I processed pharmaceutical mineral oil and an additional naphtha on a 50 C water bath. To each tube I added 2 mL of its solvent. I capped the tube, and shook vigorously to mix. I centrifuged to break any emulsion, then removed the solvent with a glass syringe and discarded it. I stirred with a spatula to resuspend any sediment. I repeated the liquid-liquid extraction a second time. So I pulled each tube with its solvent twice, and discarded both pulls.
I then made each tube acidic with 6 mL 10% citric acid, resuspended, shook to mix, and centrifuged. I diluted that water into mobile phase and injected. I've superimposed those chromatograms to produce this figure:
These are chromatograms of the water after liquid-liquid extraction. So the presence of a peak means it wasn't extracted into the solvent, so a flatter chromatogram is a stronger solvent, backwards from before.
Dichloromethane and ethyl acetate are almost perfectly flat, showing that they pulled essentially all the alkaloids. Limonene is a little higher, but not by much. Hot naphtha is much higher. Hot mineral oil and room-temperature naptha are yet higher and similar. That seems like the expected order to me. The mineral oil worked better than I'd expected, though it's probably impractically viscous.
I should have run a control tube that got basified, centrifuged, and acidified but not extracted. I didn't, so I've included a chromatogram of the original aqueous extract for the starting reference. I've scaled that by a factor of 5/12, to account for the dilution from the additional 6+1 mL added to the tubes.
I plotted the same data, scaling each trace vertically to make the 5-MeO-DMT peaks the same height:
We see that the gramine peaks are very close to the same height, implying that gramine was pulled in roughly equal proportion. I've hidden the DCM, ethyl acetate, and limonene traces here since they're too small in absolute terms, in the noise.
Scaling to the DMT peak, we see enrichment of gramine in the water:
This implies depletion from the solvent, the good direction. Take care that the depletion ratio in the solvent can't be calculated from this curve alone; we need the absolute change from the starting point too, and relative error gets amplified when we subtract. I don't think my chromatography is good enough to justify a real calculation, but the ratio of DMT to gramine is probably improved by a factor of less than two. More pulls will worsen that ratio.The limonene pulled DMT more effectively than I expected from the extraction that I crystallized. After more careful review of the samples I took then, I think the DMT did actually make it into the limonene, but roughly half stayed in the limonene when I extracted back into water, under the same conditions that pulled almost all of the 5-MeO-DMT. I think the rest got separated when I evaporated, redissolved into ethanol, and filtered. (So would washing DMT oxalate with ethanol remove 5-MeO-DMT impurities? Obviously don't trust any such separation without extensive verification.)
I don't think the gramine is particularly harmful in my application. If it's undesired then choice among these solvents under these conditions provides limited benefit in rejecting it, though. It's possible that other solvents or other conditions would reject better. My solutions were somewhat weak (around 0.5 mg/mL alkaloids), and stronger might behave differently.
Based on current knowledge, I would choose my solvent primarily based on factors other than selectivity against unwanted alkaloids. Limonene seems like the winner to me--cheap, almost perfectly nontoxic, flash point (just) above room temperature, pretty good absolute strength. It was also much less prone to emulsions than any other solvent, forming a sharp interface with no centrifuging.
My Phalaris brachystachys is growing outdoors, and I'll aim to extract a larger amount once I can harvest that. I'll perhaps then try to crystallize as the fumarate or succinate, since those are friendlier. This could possibly also improve the gramine-free yield if I got lucky on the relative solubilities, though I doubt it.
Grasshoppers
Titanium Teammate
Thank you so much for sharing your fascinating report. I really appreciate your insights and contribution to our understanding of the impact of different solvents.
Based on your findings, I’m planning to experiment with ethyl acetate instead of dichloromethane for TLC extractions—it seems like a promising alternative.
Regarding gramine, I’d love to dive deeper into the potential toxic compounds in Phalaris. For instance, @neurobloom has reported acute cardiocirculatory effects after consuming Phalaris extracts, both in combination with reversible monoamine oxidase inhibitors (RIMAs)[1] and even when smoked without RIMAs[2]. Though these events seem rare, they do raise questions about the safety profile of Phalaris extracts, especially as more potent clones become widely available.
We’ve internally discussed the following compounds as potential causes of these acute effects:
Do you have any insights into how tyramines might be extracted during acid-base extraction and their occurrence in P. aquatica?
I’m also curious whether you’ve encountered beta-carbolines (like 2MeTHBC and 26DiMeTHBC) in your research. We’ve observed signs of THBCs, particularly in P. arundinacea and P. cerulescens.
Additionally, we’ve noticed a fluorescent compound in some P. canariensis seedlings, with an RF similar to DMT, under 275nm UVC light. It has appeared rarely in P. aquatica. If this rings any bells, I’d be keen to hear your thoughts on what it could be:
[3]
As we continue to collect potent clones, the focus on ensuring low toxicity is becoming more critical. I’d love to hear your perspective on this.
Looking forward to your thoughts!
Based on your findings, I’m planning to experiment with ethyl acetate instead of dichloromethane for TLC extractions—it seems like a promising alternative.
Regarding gramine, I’d love to dive deeper into the potential toxic compounds in Phalaris. For instance, @neurobloom has reported acute cardiocirculatory effects after consuming Phalaris extracts, both in combination with reversible monoamine oxidase inhibitors (RIMAs)[1] and even when smoked without RIMAs[2]. Though these events seem rare, they do raise questions about the safety profile of Phalaris extracts, especially as more potent clones become widely available.
We’ve internally discussed the following compounds as potential causes of these acute effects:
- Tyramine
- N-Methyltyramine
- Hordenine (N,N-Dimethyltyramine)
- Gramine
Do you have any insights into how tyramines might be extracted during acid-base extraction and their occurrence in P. aquatica?
I’m also curious whether you’ve encountered beta-carbolines (like 2MeTHBC and 26DiMeTHBC) in your research. We’ve observed signs of THBCs, particularly in P. arundinacea and P. cerulescens.
Additionally, we’ve noticed a fluorescent compound in some P. canariensis seedlings, with an RF similar to DMT, under 275nm UVC light. It has appeared rarely in P. aquatica. If this rings any bells, I’d be keen to hear your thoughts on what it could be:
[3]As we continue to collect potent clones, the focus on ensuring low toxicity is becoming more critical. I’d love to hear your perspective on this.
Looking forward to your thoughts!
Last edited:
aizoaceous
Titanium Teammate
I haven't studied any of this toxicology that much. Tyramine without an MAOI seems completely fine here, since guidelines permit up to 600 mg per day. That goes down to 6 mg per day with classical pharmaceutical MAOIs, though even that seems fine with typical doses of 5-MeO-DMT and harmalas should be safer. I feel like hordenine is pretty nontoxic, given its high concentration in barley (especially when malted) and the apparent absence of concern in MAOI diets.
All the above assumes absorption in the stomach and intestines. I don't think there's any literature at all on effects when vaporized and inhaled. There's not on sublingual either, though that feels closer at least. Livestock chewing Phalaris grass are presumably absorbing there too, though I believe Phalaris staggers still haven't been convincingly explained.
An old paper (available on Sci-Hub) prepared food samples for analysis of tyramine by extraction with ethyl acetate, so I guess that pulls it. I don't see why an acid-base extraction wouldn't pull it in general, though perhaps specific conditions would reject it. From anecdotal reports on mescaline, I'd guess that naphtha would reject tyramine better than it rejects gramine, but not perfectly.
Unless we're unlucky on the solubilities, crystallization should reject small amounts of any of these near-perfectly. The challenge is in judging where to stop crystallization though, since evaporation all the way to dryness just recovers the original mixture. We'd ideally find a solvent system and salt where the desired product is much less soluble than the impurities, so that an excess of solvent doesn't lose too much product. That feels hard, though. I'll continue to experiment.
From my earlier experiment, I believe beta-carbolines don't elute in my usual tryptamine method, and thus aren't analyzed here. A quick fluorescence screen reported in that same post showed nothing in my P. brachystachys; but you're working with a far bigger gene pool than I am, with correspondingly greater potential for both positive and negative surprises.
I have no guesses for that mystery spot, beyond the obvious that it's probably a tryptamine. It's sure convenient that they're fluorescent, but the visible fluorescence on TLC is unfortunately less reported than the emission peak in the UV. There's that Japanese paper with the oxidizing stain, but that's forensic so they mostly report Shulgin's synthetics.
All the above assumes absorption in the stomach and intestines. I don't think there's any literature at all on effects when vaporized and inhaled. There's not on sublingual either, though that feels closer at least. Livestock chewing Phalaris grass are presumably absorbing there too, though I believe Phalaris staggers still haven't been convincingly explained.
An old paper (available on Sci-Hub) prepared food samples for analysis of tyramine by extraction with ethyl acetate, so I guess that pulls it. I don't see why an acid-base extraction wouldn't pull it in general, though perhaps specific conditions would reject it. From anecdotal reports on mescaline, I'd guess that naphtha would reject tyramine better than it rejects gramine, but not perfectly.
Unless we're unlucky on the solubilities, crystallization should reject small amounts of any of these near-perfectly. The challenge is in judging where to stop crystallization though, since evaporation all the way to dryness just recovers the original mixture. We'd ideally find a solvent system and salt where the desired product is much less soluble than the impurities, so that an excess of solvent doesn't lose too much product. That feels hard, though. I'll continue to experiment.
From my earlier experiment, I believe beta-carbolines don't elute in my usual tryptamine method, and thus aren't analyzed here. A quick fluorescence screen reported in that same post showed nothing in my P. brachystachys; but you're working with a far bigger gene pool than I am, with correspondingly greater potential for both positive and negative surprises.
I have no guesses for that mystery spot, beyond the obvious that it's probably a tryptamine. It's sure convenient that they're fluorescent, but the visible fluorescence on TLC is unfortunately less reported than the emission peak in the UV. There's that Japanese paper with the oxidizing stain, but that's forensic so they mostly report Shulgin's synthetics.
Grasshoppers
Titanium Teammate
DCM could not be substituted by EtOAc for TLC analysis, as the spots became distorted and split. This suggests a possible unknown interaction or contamination in the EtOAc used.
The presence of potential toxins must be carefully considered for several reasons. We aim to explore Phalaris as a source of N,N-DMT, which will require higher doses and thus increase the potential impact of any toxins present. The oral use of Phalaris in combination with RIMAs, similar to its use in ayahuasca preparations, is not uncommon, with many reports already available. However, this necessitates a thorough understanding of any toxic compounds.
Crystallization methods for purification may be beyond the reach of many users, which makes addressing these concerns even more important.
It's still possible that the reported cardiocirculatory effects mentioned by @neurobloom were caused by unknown compounds, or they might have been purely psychological "bad trips." Nonetheless, screening for potential toxins remains essential.
Thank you for referencing N.P. Sen (1969). We will begin investigating tyramine. Even though tyramines are probably non-fluorescent, they should be detectable on TLC plates with iodine staining.
I greatly appreciate your pioneering experiments on purifying DMT from Phalaris. Your work clearly paves the way for its future use and provides a foundation for further exploration.
The presence of potential toxins must be carefully considered for several reasons. We aim to explore Phalaris as a source of N,N-DMT, which will require higher doses and thus increase the potential impact of any toxins present. The oral use of Phalaris in combination with RIMAs, similar to its use in ayahuasca preparations, is not uncommon, with many reports already available. However, this necessitates a thorough understanding of any toxic compounds.
Crystallization methods for purification may be beyond the reach of many users, which makes addressing these concerns even more important.
It's still possible that the reported cardiocirculatory effects mentioned by @neurobloom were caused by unknown compounds, or they might have been purely psychological "bad trips." Nonetheless, screening for potential toxins remains essential.
Thank you for referencing N.P. Sen (1969). We will begin investigating tyramine. Even though tyramines are probably non-fluorescent, they should be detectable on TLC plates with iodine staining.
I greatly appreciate your pioneering experiments on purifying DMT from Phalaris. Your work clearly paves the way for its future use and provides a foundation for further exploration.
aizoaceous
Titanium Teammate
Were they distorted before elution? Ethyl acetate is higher-boiling and possibly a stronger eluent, so the alkaloids might just be spreading more as you spot them. A higher temperature on the hotplate, an air stream (like from a hair dryer), or slower spotting might help in that case. Water dissolved in the ethyl acetate would cause similar and yet worse trouble. Your sodium carbonate should remove that, but more will be required than with the DCM. You could decant and add more anhydrous sodium carbonate, to confirm it doesn't clump up (indicating the presence of residual water).
DCM is a great solvent, low-boiling, low solubility with water, strong, not too flammable. It's just unfortunately toxic, and may become broadly unavailable in the USA soon due to targeted environmental restrictions. So replacement is beneficial where possible.
Ayahuasca with pure DMT is indeed both the most dangerous and the most desired use of Phalaris. The promise is clearly there, just not yet realized. I guess a clone in your collection might be the answer, and another few clones might strongly resemble the answer but in fact be a recipe for poisoning in certain cases, "just" a question of sorting out which is which.
Tyramine should be nonfluorescent. In addition to stains, it should be detectable by UV absorption, though with peaks far from the usual 254 nm so the LOD may be degraded. I found some other papers that used complex solvent systems to extract it (like a mixture of ethyl acetate and acetone), possibly implying low recovery with pure ethyl acetate.
If @neurobloom's bad experiences weren't from contaminants, then I'd guess they were an overdose of 5-MeO-DMT, which sure seems common even in reports from otherwise smart and cautious people (including Shulgin; I think it's literally the worst experience he reports, in all his history with new drugs). That's another benefit of purification, allowing the dose to be meaningfully weighed, though you're correct that no easy method exists for that yet. I'll keep screening salts and solvents, and maybe I'll get lucky.
Grasshoppers
Titanium Teammate
The spots became distorted during elution. I spotted with DCM at 45°C and EtOAc at 80°C, and in both cases, the initial spots were narrow. DMT should not undergo significant oxidation or degradation at 80°C. Therefore, the spotting process itself likely wasn’t the cause.
On the following TLC plate, you'll see two Phalaris aquatica specimens: the one on the left is 5-MeO-DMT dominant, while the one on the right is N,N-DMT dominant. EtOAc and DCM were used for extraction and spotting.
Both the 5-MeO-DMT and N,N-DMT spots extracted/spotted with EtOAc split, suggesting that the DMT is either partially altered or interacting with something else.
Regarding DCM toxicity: For one plate, I used 4 mL of DCM (4x1 mL). After evaporation, this wouldn’t come close to the permissible exposure limit over 8 hours (PEL) of 25 ppm in my workspace.
The unavailability of DCM in the U.S. is unfortunate. I could try petroleum ether for extraction instead; although it has a lower pulling capacity, that might not be a major issue here. I typically use 1 mL of organic solvent to pull from 100 μL of basified water extract, which contains about 100 μg of leaf sample equivalent.
At this point, we’re not yet able to determine if any of our clones are suitable for direct oral ayahuasca use. First, we haven’t sufficiently screened for toxins. Second, we haven’t observed the temporal stability of their alkaloid profiles long enough, especially since some of the most promising clones were only recently discovered and are still in the seedling stage.
Thank you for the Yigit & Ersoy (2003) paper. Unfortunately, understanding tyramine will require experimentation with a reference standard, as the existing literature is somewhat limited.
So far, we’ve only worked with wild variants and cultivars of P. aquatica. Hybridization of genetically unrelated P. aquatica specimens, as discussed in Putievsky et al. (1980), has resulted in extreme traits. While we don’t yet know if this also affects the alkaloid profiles, it seems likely. So the really exciting part of the project still lies ahead. Once we reach the stage of growing and testing hybrids, our focus will shift more towards identifying potential toxins.
On the following TLC plate, you'll see two Phalaris aquatica specimens: the one on the left is 5-MeO-DMT dominant, while the one on the right is N,N-DMT dominant. EtOAc and DCM were used for extraction and spotting.
Both the 5-MeO-DMT and N,N-DMT spots extracted/spotted with EtOAc split, suggesting that the DMT is either partially altered or interacting with something else.
Regarding DCM toxicity: For one plate, I used 4 mL of DCM (4x1 mL). After evaporation, this wouldn’t come close to the permissible exposure limit over 8 hours (PEL) of 25 ppm in my workspace.
The unavailability of DCM in the U.S. is unfortunate. I could try petroleum ether for extraction instead; although it has a lower pulling capacity, that might not be a major issue here. I typically use 1 mL of organic solvent to pull from 100 μL of basified water extract, which contains about 100 μg of leaf sample equivalent.
At this point, we’re not yet able to determine if any of our clones are suitable for direct oral ayahuasca use. First, we haven’t sufficiently screened for toxins. Second, we haven’t observed the temporal stability of their alkaloid profiles long enough, especially since some of the most promising clones were only recently discovered and are still in the seedling stage.
Thank you for the Yigit & Ersoy (2003) paper. Unfortunately, understanding tyramine will require experimentation with a reference standard, as the existing literature is somewhat limited.
So far, we’ve only worked with wild variants and cultivars of P. aquatica. Hybridization of genetically unrelated P. aquatica specimens, as discussed in Putievsky et al. (1980), has resulted in extreme traits. While we don’t yet know if this also affects the alkaloid profiles, it seems likely. So the really exciting part of the project still lies ahead. Once we reach the stage of growing and testing hybrids, our focus will shift more towards identifying potential toxins.
Last edited:
Grasshoppers
Titanium Teammate
I'm excited to share these highly potent 5-MeO-DMT crystals, extracted from Phalaris aquatica (tanit accession)!
For those who enjoy the science behind the sparkle, here’s a quick breakdown of the extraction process:
Starting Material: 30g of dried Phalaris leaves
Acid Extraction: Boiled in 1% acetic acid to help release the alkaloids
Filtration: Passed through cotton for initial purification
Basification : Added sodium carbonate (NaCO₃) to basify
Solvent Extraction: Pulled with a mix of petroleum ether and ethyl acetate, then dried
Re-acidification: Dissolved again in 1% acetic acid for further refinement
Filtration: Passed through cotton for final purification
Final Pull: Extracted one last time with petroleum ether
Hope you all enjoy the photos as much as I enjoyed creating them.
For those who enjoy the science behind the sparkle, here’s a quick breakdown of the extraction process:
Starting Material: 30g of dried Phalaris leaves
Acid Extraction: Boiled in 1% acetic acid to help release the alkaloids
Filtration: Passed through cotton for initial purification
Basification : Added sodium carbonate (NaCO₃) to basify
Solvent Extraction: Pulled with a mix of petroleum ether and ethyl acetate, then dried
Re-acidification: Dissolved again in 1% acetic acid for further refinement
Filtration: Passed through cotton for final purification
Final Pull: Extracted one last time with petroleum ether
Hope you all enjoy the photos as much as I enjoyed creating them.
Last edited:
aizoaceous
Titanium Teammate
Very nice. I believe your method is similar to what @fourthripley did back in 2014 or earlier, summarized in the attachment to this post that I only recently discovered. You have less color than either my oxalate or @fourthripley's freebase, maybe due to the cleaner genetics.
Since you're crystallizing as freebase, there's no question of excess acid, and the use of petroleum ether avoids any problem of base carrying over with water dissolved in more polar solvents. (Though maybe with use of calcium carbonate or hydroxide that would be fine regardless, due to its low solubility in water?) This seems certainly preferable to crystallization as the oxalate for anyone intending to vaporize the product, and likely preferable even for oral or sublingual use. I'll attempt a similar method with some of my P. brachystachys, and try to quantify the yield per step.
I guess a TLC of your crystals would look perfect. You could evaporate down to see at what point impurities become detectable (like I did), though that might be pointless when your 'Tanit' is so clean to start with.
As to tyramine, I think my statement above that it's nonfluorescent isn't quite right. That barley paper does show a signal (e.g. Figure 2(a)), just much weaker than the indole alkaloids. As usual, the peak emission is in UV. The visible tail of the tryptamines is already pretty faint and the tyramine will be fainter, but perhaps it would be detectable. I plan to prepare a crude standard by pyrolysis of tyrosine and measure recovery under various extraction conditions (especially over pH; what's the pKa for that hydroxy?), and possibly that will identify one of my unknown peaks. Suggestions for better standards are welcome. It's oddly hard to buy tyramine here, and a proper synthesis like by heating the tyrosine in diphenylamine seems unnecessarily toxic.
Since you're crystallizing as freebase, there's no question of excess acid, and the use of petroleum ether avoids any problem of base carrying over with water dissolved in more polar solvents. (Though maybe with use of calcium carbonate or hydroxide that would be fine regardless, due to its low solubility in water?) This seems certainly preferable to crystallization as the oxalate for anyone intending to vaporize the product, and likely preferable even for oral or sublingual use. I'll attempt a similar method with some of my P. brachystachys, and try to quantify the yield per step.
I guess a TLC of your crystals would look perfect. You could evaporate down to see at what point impurities become detectable (like I did), though that might be pointless when your 'Tanit' is so clean to start with.
As to tyramine, I think my statement above that it's nonfluorescent isn't quite right. That barley paper does show a signal (e.g. Figure 2(a)), just much weaker than the indole alkaloids. As usual, the peak emission is in UV. The visible tail of the tryptamines is already pretty faint and the tyramine will be fainter, but perhaps it would be detectable. I plan to prepare a crude standard by pyrolysis of tyrosine and measure recovery under various extraction conditions (especially over pH; what's the pKa for that hydroxy?), and possibly that will identify one of my unknown peaks. Suggestions for better standards are welcome. It's oddly hard to buy tyramine here, and a proper synthesis like by heating the tyrosine in diphenylamine seems unnecessarily toxic.
Wouldn't tyrosine be amenable to carvone-catalysed decarboxylation, just like tryptophan? Here's a great post, btw, on the general aspects of ketone-catalysed decarboxylation:
This may also be worthy of perusal:
US20180312461A1 - Method for solvent-free decarboxylation of amino acids via imine formation - Google Patents
The present application provides solvent-free methods for decarboxylation of amino acids via imine formation with a ketone, enone, enal, aldehyde co-reagent or combination thereof, under heated conditions, with optional recovery of the co-reagent and/or co-reagent byproduct.
patents.google.com
Useful refs from the SM post:
Rapid Conventional and Microwave-Assisted Decarboxylation of L-Histidine and Other Amino Acids via Organocatalysis with R-Carvone Under Superheated Conditions
Douglas M. Jackson, Robert L. Ashley, Callan B. Brownfield, Daniel R. Morrison & Richard W. Morrison
Synthetic Communications Volume 45, 2015 - Issue 23 Pages 2691-2700
DOI:10.1080/00397911.2015.1100745
Schiff bases. Part I. Thermal decarboxylation of α-amino-acids in the presence of ketones
A. F. Al-Sayyab and Alexander Lawson
J. Chem. Soc. C, 1968,0, 406-410
DOI:10.1039/J39680000406
Rapid Conventional and Microwave-Assisted Decarboxylation of L-Histidine and Other Amino Acids via Organocatalysis with R-Carvone Under Superheated Conditions
Douglas M. Jackson, Robert L. Ashley, Callan B. Brownfield, Daniel R. Morrison & Richard W. Morrison
Synthetic Communications Volume 45, 2015 - Issue 23 Pages 2691-2700
DOI:10.1080/00397911.2015.1100745
Schiff bases. Part I. Thermal decarboxylation of α-amino-acids in the presence of ketones
A. F. Al-Sayyab and Alexander Lawson
J. Chem. Soc. C, 1968,0, 406-410
DOI:10.1039/J39680000406
aizoaceous
Titanium Teammate
Synthesis of Tyramine
The carvone-catalyzed decarboxylation looks great, easily available reagents with low toxicity. Reported yields (mostly for amino acids other than tyrosine) vary greatly, but I don't care much about that here. My skills and facilities for synthesis are weaker than for analysis, but it still seemed within reach.
I first tried solvent-free like their patent, with 1.5 g l-tyrosine and 5.0 g d-carvone. Lacking proper equipment, I used a 10 mL beaker on a magnetic stir plate, with my heat gun's narrow nozzle blowing at the base of the beaker. (My stir plate does heat, but it doesn't get hot enough for a proper oil bath.) I placed a thermocouple in the beaker and adjusted the heat gun for 180 C. This bubbled and appeared to be working, but unfortunately soon turned dark, yielding an intractable tar upon cooling. I used d-carvone (smells like caraway) and not the patent's l-carvone (spearmint), but the d-carvone worked in their paper.
I repeated the process with a solvent, using 7 mL d-limonene, 0.9 g l-tyrosine, and 1.5 g d-carvone. I kept the temperature just below the boiling point of limonene, but still had to top it off a couple times. That looked better, but there was still a lot of tar.
I repeated the process with a long test tube, loosely capped with aluminum foil. The reaction mixture is thus blanketed with carbon dioxide and other evolved pyrolysis gases, excluding air. Most papers seem to be relying on that effect, though one ran under nitrogen. The upper cool portion of the test tube also serves to condense evaporated limonene, maintaining a crude reflux. I again started with 7 mL d-limonene, 0.9 g l-tyrosine, and 1.5 g d-carvone, stirring magnetically and heating with my heat gun. I saw bubbles, though it's hard to distinguish carbon dioxide bubbles from local boiling from the applied heat. It may be easiest to track the reaction progress by the exotherm--the temperature was steady with a steady heat gun setting for the first forty minutes or so, then began to drop abruptly around the same time the lower phase of the reaction mixture looked entirely liquid. The reaction mixture never became totally homogenous, but the lower white phase melts around 160 C, as expected for tyramine. The upper phase turned a dark red, but there was nothing like the previous tar. The lower phase hardens into a powdery solid, luckily not one big rock.
I extracted with 6 mL 2M hydrochloric acid, then heated the aqueous phase on an 80 C water bath for 30 min. The Georgia authors recommend this because they found an unfortunately stable Schiff base intermediate, which didn't hydrolyze into the amine without that heating. I feel like that should have released a noticeable extra layer of carvone, though. The paper says three cycles of heating and extraction are necessary to recover 80% of the carvone, but I gave up after one with no obvious change. There was still some red color, which may be the Schiff base.
I extracted the aqueous phase with 3 mL ethyl acetate (they used ether but I don't have that) anyways, then centrifuged to separate. There were still some undissolved solids; maybe tyramine's solubility is low enough that I needed more acid? I don't care much about yield so I just discarded the solids and proceeded. I neutralized the acid with sodium bicarbonate. The solution turned milky, then to a foam like shaving cream that required dilution to filter. The filter cake was cohesive after drying, but easily broken up with a spatula. Melting point was wide, perhaps from 155 C to 165 C. Retention time in my usual HPLC system was 3.2 min, same as for tyrosine.
Yield was 490 mg, which would be 72% if it were pure. The wide melting point means it's not, with a faint pink color and a strong smell of caraway. Crystallization as the hydrochloride would probably be good idea. The co-elution with tyrosine is apparently expected per that barley paper; they got separation only with almost pure water as the eluent. I don't see what else would elute at that time except unreacted tyrosine, and that melts far higher (actually decomposes, at 343 C). So I'm pretty sure this worked.
For clarity, tyramine is not a desirable constituent. I synthesized it only because I needed it as a standard and couldn't purchase it. I don't know why it's so hard for a non-institution to purchase tyramine here, possibly because it's a controlled substance in Florida. Consumption results only in a headache or worse, so that scheduling is probably just carelessness from whoever Florida consulted, swept up with other phenethylamines.
Tyramine in Phalaris
With that peak now identified, I looked back at my old chromatograms. None show any evidence of tyramine. The P. aquatica has some unidentified early eluters, but the retention times look too far off. As always, results hold only for the exact genetics that I tested, and P. aquatica in particular shows big intraspecific variation. As for all undesired constituents, tyramine may present a greater risk to users of DMT-dominant varieties due to the much higher dose.
Tyramine may exhibit complex solubility behavior, since its acidic phenol has pKa close to its basic amine. This discussion started with the question of whether tyramine will get pulled in a typical acid-base extraction. @neurobloom found a paper that said it won't, while others seem to imply it will. To study this, I spiked about 500 ug/mL of tyramine into 1 mL aqueous Phalaris extract, and took a sample for analysis. I then made it basic with an excess of potassium carbonate, extracted once with 0.5 mL ethyl acetate, and resampled the aqueous extract. That's a lot of tyramine (there's only ~100 ug/mL 5-MeO-DMT), but it won't be a very big peak since the fluorescence is weaker than for the indole alkaloids. Here's the result:
The tyramine peak has roughly the same height before and after (unlike all the others), so it's not getting pulled. I also tried soy sauce; a peak appears with expected time and height when the sauce is simply diluted and injected, but does not appear in the ethyl acetate extract of the basified sauce. I tried tyramine hydrochloride in plain water, and it also didn't get pulled.
I'm not sure what I'm doing differently from Sen 1969. I tried exactly replicating their conditions, using borate buffer prepared with an electronic pH meter, and I still got nothing. I'd love to know what I'm missing here, but for now I think this further emphasizes that tyramine isn't a big concern:
Outdoor Culture
As I'd already noted elsewhere, my P. brachystachys is growing vigorously outdoors. I've harvested more than 2 kg by now, from just a couple square feet. My first harvest was unfortunately very weak though, around 200 ug 5-MeO-DMT per g fresh weight vs. 1000 ug/g for material grown indoors.
I'd considered low fertility as the cause, since the grass was growing in natural soil with only sporadic hand-fertilizing. I'm now injecting fertilizer into my irrigation water, but a second harvest yielded almost exactly the same. Remaining explanations for the difference include temperature, sunlight, and soil conditions other than fertility (pH in calcareous soil?). I have a small hydroponic system in part shade outdoors, and may try that to compare.
The outdoor grass is also harder to extract, with concentration dropping more slowly on successive pulls than with the indoor grass, implying there's more left unextracted. So the drop in total alkaloids is actually smaller than the numbers I'm giving above, though I think it's still dramatic and the more difficult extraction is undesirable. The outdoor leaves are thicker and stiffer, so this may not be surprising.
For the second outdoor harvest, I tried to macerate the grass with an immersion blender to release more alkaloids. That didn't feel very promising, since the fibrous grass tended to jam the blender. I see no evidence that it helped, since neither the total yield nor the slope of decrease with additional pulls improved. An auger-type wheatgrass juicer also jammed constantly, though an auger built to better tolerances might have worked better. Sonication or boiling in large amounts of water may still be the best option, but reducing off that excess water later is slow. I might try going straight to base, like from a paste of dried and powdered grass with water and calcium hydroxide.
The carvone-catalyzed decarboxylation looks great, easily available reagents with low toxicity. Reported yields (mostly for amino acids other than tyrosine) vary greatly, but I don't care much about that here. My skills and facilities for synthesis are weaker than for analysis, but it still seemed within reach.
I first tried solvent-free like their patent, with 1.5 g l-tyrosine and 5.0 g d-carvone. Lacking proper equipment, I used a 10 mL beaker on a magnetic stir plate, with my heat gun's narrow nozzle blowing at the base of the beaker. (My stir plate does heat, but it doesn't get hot enough for a proper oil bath.) I placed a thermocouple in the beaker and adjusted the heat gun for 180 C. This bubbled and appeared to be working, but unfortunately soon turned dark, yielding an intractable tar upon cooling. I used d-carvone (smells like caraway) and not the patent's l-carvone (spearmint), but the d-carvone worked in their paper.
I repeated the process with a solvent, using 7 mL d-limonene, 0.9 g l-tyrosine, and 1.5 g d-carvone. I kept the temperature just below the boiling point of limonene, but still had to top it off a couple times. That looked better, but there was still a lot of tar.
I repeated the process with a long test tube, loosely capped with aluminum foil. The reaction mixture is thus blanketed with carbon dioxide and other evolved pyrolysis gases, excluding air. Most papers seem to be relying on that effect, though one ran under nitrogen. The upper cool portion of the test tube also serves to condense evaporated limonene, maintaining a crude reflux. I again started with 7 mL d-limonene, 0.9 g l-tyrosine, and 1.5 g d-carvone, stirring magnetically and heating with my heat gun. I saw bubbles, though it's hard to distinguish carbon dioxide bubbles from local boiling from the applied heat. It may be easiest to track the reaction progress by the exotherm--the temperature was steady with a steady heat gun setting for the first forty minutes or so, then began to drop abruptly around the same time the lower phase of the reaction mixture looked entirely liquid. The reaction mixture never became totally homogenous, but the lower white phase melts around 160 C, as expected for tyramine. The upper phase turned a dark red, but there was nothing like the previous tar. The lower phase hardens into a powdery solid, luckily not one big rock.
I extracted with 6 mL 2M hydrochloric acid, then heated the aqueous phase on an 80 C water bath for 30 min. The Georgia authors recommend this because they found an unfortunately stable Schiff base intermediate, which didn't hydrolyze into the amine without that heating. I feel like that should have released a noticeable extra layer of carvone, though. The paper says three cycles of heating and extraction are necessary to recover 80% of the carvone, but I gave up after one with no obvious change. There was still some red color, which may be the Schiff base.
I extracted the aqueous phase with 3 mL ethyl acetate (they used ether but I don't have that) anyways, then centrifuged to separate. There were still some undissolved solids; maybe tyramine's solubility is low enough that I needed more acid? I don't care much about yield so I just discarded the solids and proceeded. I neutralized the acid with sodium bicarbonate. The solution turned milky, then to a foam like shaving cream that required dilution to filter. The filter cake was cohesive after drying, but easily broken up with a spatula. Melting point was wide, perhaps from 155 C to 165 C. Retention time in my usual HPLC system was 3.2 min, same as for tyrosine.
Yield was 490 mg, which would be 72% if it were pure. The wide melting point means it's not, with a faint pink color and a strong smell of caraway. Crystallization as the hydrochloride would probably be good idea. The co-elution with tyrosine is apparently expected per that barley paper; they got separation only with almost pure water as the eluent. I don't see what else would elute at that time except unreacted tyrosine, and that melts far higher (actually decomposes, at 343 C). So I'm pretty sure this worked.
For clarity, tyramine is not a desirable constituent. I synthesized it only because I needed it as a standard and couldn't purchase it. I don't know why it's so hard for a non-institution to purchase tyramine here, possibly because it's a controlled substance in Florida. Consumption results only in a headache or worse, so that scheduling is probably just carelessness from whoever Florida consulted, swept up with other phenethylamines.
Tyramine in Phalaris
With that peak now identified, I looked back at my old chromatograms. None show any evidence of tyramine. The P. aquatica has some unidentified early eluters, but the retention times look too far off. As always, results hold only for the exact genetics that I tested, and P. aquatica in particular shows big intraspecific variation. As for all undesired constituents, tyramine may present a greater risk to users of DMT-dominant varieties due to the much higher dose.
Tyramine may exhibit complex solubility behavior, since its acidic phenol has pKa close to its basic amine. This discussion started with the question of whether tyramine will get pulled in a typical acid-base extraction. @neurobloom found a paper that said it won't, while others seem to imply it will. To study this, I spiked about 500 ug/mL of tyramine into 1 mL aqueous Phalaris extract, and took a sample for analysis. I then made it basic with an excess of potassium carbonate, extracted once with 0.5 mL ethyl acetate, and resampled the aqueous extract. That's a lot of tyramine (there's only ~100 ug/mL 5-MeO-DMT), but it won't be a very big peak since the fluorescence is weaker than for the indole alkaloids. Here's the result:
The tyramine peak has roughly the same height before and after (unlike all the others), so it's not getting pulled. I also tried soy sauce; a peak appears with expected time and height when the sauce is simply diluted and injected, but does not appear in the ethyl acetate extract of the basified sauce. I tried tyramine hydrochloride in plain water, and it also didn't get pulled.
I'm not sure what I'm doing differently from Sen 1969. I tried exactly replicating their conditions, using borate buffer prepared with an electronic pH meter, and I still got nothing. I'd love to know what I'm missing here, but for now I think this further emphasizes that tyramine isn't a big concern:
- It's not present in any of my plants.
- If it were present, my method doesn't extract it.
- If it somehow got extracted, it's probably not too harmful.
Outdoor Culture
As I'd already noted elsewhere, my P. brachystachys is growing vigorously outdoors. I've harvested more than 2 kg by now, from just a couple square feet. My first harvest was unfortunately very weak though, around 200 ug 5-MeO-DMT per g fresh weight vs. 1000 ug/g for material grown indoors.
I'd considered low fertility as the cause, since the grass was growing in natural soil with only sporadic hand-fertilizing. I'm now injecting fertilizer into my irrigation water, but a second harvest yielded almost exactly the same. Remaining explanations for the difference include temperature, sunlight, and soil conditions other than fertility (pH in calcareous soil?). I have a small hydroponic system in part shade outdoors, and may try that to compare.
The outdoor grass is also harder to extract, with concentration dropping more slowly on successive pulls than with the indoor grass, implying there's more left unextracted. So the drop in total alkaloids is actually smaller than the numbers I'm giving above, though I think it's still dramatic and the more difficult extraction is undesirable. The outdoor leaves are thicker and stiffer, so this may not be surprising.
For the second outdoor harvest, I tried to macerate the grass with an immersion blender to release more alkaloids. That didn't feel very promising, since the fibrous grass tended to jam the blender. I see no evidence that it helped, since neither the total yield nor the slope of decrease with additional pulls improved. An auger-type wheatgrass juicer also jammed constantly, though an auger built to better tolerances might have worked better. Sonication or boiling in large amounts of water may still be the best option, but reducing off that excess water later is slow. I might try going straight to base, like from a paste of dried and powdered grass with water and calcium hydroxide.
Grasshoppers
Titanium Teammate
Thank you very much for your thorough investigation into the presence of tyramines in Phalaris species and their solubility.
While tyramine toxicity could theoretically occur in oral preparations of ayahuasca derived from Phalaris species, it appears this risk can be mitigated through careful selection of plant material.
Regarding extracted alkaloids, this is indeed excellent news—another significant step forward in understanding Phalaris as a potential source of psychedelics.
I'm particularly curious about whether the differences in potency are influenced more by the climate or the nutrient solution used.
For P. aquatica, it’s interesting to note:
While tyramine toxicity could theoretically occur in oral preparations of ayahuasca derived from Phalaris species, it appears this risk can be mitigated through careful selection of plant material.
Regarding extracted alkaloids, this is indeed excellent news—another significant step forward in understanding Phalaris as a potential source of psychedelics.
I'm particularly curious about whether the differences in potency are influenced more by the climate or the nutrient solution used.
For P. aquatica, it’s interesting to note:
- Nutrient-depleted soil seems to have a negative effect on alkaloid levels.
- Regrowth following dormancy exhibits particularly high potency.
- Silicon fertilizers, such as potassium silicate, might positively impact alkaloid production.
Certainly a plausible hypothesis. A rigid interpretation of the 1977 catchall phenethylamine amendment of the UK misuse of drugs act 1971 would have tyramine placed under class A; therefore making certain types of cheese is, technically, punishable by life imprisonment! (I'm going to have to check if tyramine is excluded through a specific amendment, but I don't recall having seen one before.)
[Edit: I was wrong - simple hydroxy substituents are not covered.
This would include, say, homovanillylamine, but not tyramine or dopamine.]
I’m mostly after 5-MeO-NMT when it comes to Phalaris arundinacea.
Plus what about the beta carbolines?
Plus what about the beta carbolines?