Outline

---

I boiled Rue seeds 5 times. After each boil, I did a simple base extraction of total alkaloids. I then washed, dried and weighed the impure freebases.

Motivation

---

To get an idea of how efficient the subsequent boils are and find a sensible trade-off between yield and energy/time consumption.


Source Material

---

300g of Iranian P. harmala seeds.

Boiling

---

I used a pressure cooker. The PC is an electric one with presets, the maximum time setting (Soup / Hi) is 37 minutes. That's the only setting I used. In the first two boils I used it once, but in boils 3 - 5 I used it twice ( = totals of 74 minutes).
In each boil I added enough vinegar to get the pH to ~4.5.

In the first two boils I used 3.5 lt water.
In boils 3 - 5 I used less, perhaps 2.5 lt.

Basing & Washes

---

I based with NaOH to pH >13. Then washed with my slightly alkaline tap water to bring the pH below 9.5, under the assumption that at that pH the weight of NaOH content is negligible and doesn't skew the results.


Results

---

1st boil, single: 11g
2nd boil, single: 6.8g
3rd boil, double: 2.4g
4th boil, double: 1.7g
5th boil, double: 0.7g


Conclusion

---

Considering how cheap Rue seeds are, the law of diminishing returns makes it not worth my time and energy to do more than two boils. A third boil adds 50% more time and energy costs, but only 13% more alkaloids.

Thanks for sharing this

I've long suspected there to still be some significant amounts of alkaloids left even after 3 boils but I never had the time or patience to do an experiment . This confirms my suspicions ..

Not sure if there have been any other threads like this but if not I think maybe this should be stickied. . Very practical information here.

That's a nice 7.5% raw alkaloids jugabe
Are you planning to manske and see what weight you have then?

It was a bit too much for me to manske the 5 boils separately as I don't have a good drying setup, I was using a heated propagator and it took ages. Might get a food dehydrator one day.

However, I did manske the first two boils together and attempted a separation. The first, harmine (with some harmaline contamination) fraction has dried and the harmaline is drying as I type these words. Once finished, I will see by how much the weight has decreased and post an update here.

I should add I didn't filter the rue tea through coffee filters, only a fine cloth. The seeds were boiled whole. Also, the raw alkaloids from the later boils came out darker, presumably because the alkaloids get boiled out of the plant material first, and the later boils pull more impurities. This only strengthens the point made in the conclusion, as after purification the weight balance should shift to favour the first boils even more.

Thanks,
once I did a bloody overkill to see when finally there would nothing base out any more, it must have been over 8 boils for sure, also PC on whole seeds. The last precipitates where no bits but more of a white slime. So yes the latter boils might yield stuff we're least interested in. The slime was not filterable as per coffee filter at all. So that was going until the end

Seeing forward to your finals.

Thanks for that, Jagube - very useful data

Update for Jees:
The crude alkaloids from the first two boils, after coffee filtering and a single manske run, went from 17.8g to 14.1g, thus a reduction to 79%. This would indicate a harmine/harmaline yield of 4.7% from the first two boils alone. But since it was only one manske round, I think they're still quite impure. I don't aim to do more manske rounds, because I don't care for high purity, so take this particular data point for what it's worth.


In my case, the last precipitates were fluffy and much of the fluff disappeared as the washes brought the pH down.

Thanks. I usually do 4 boils based on practice, but it's good to have this backed up by hard data.

I once did an experiment for finding alkaloids in the seed cores by fully extracting whole seeds and then grinding them and re-extracting. The experiment failed because the ground seed mash is practically impossible to filter (unless you have a something like a hydraulic press) but I have not had PC at the time as well as vacuum filtration and more tricks at hand.

Having your data and method, it would be interesting to predict amount extracted from say 6th boil (by interpolating the trend curve), then doing the 6th boil with ground seeds and calculating the yield minus estimate. I have a suspicion the cores might be rich in harmaline, which is valuable substance for tetrahydroharmine synthesis.

Do you think PC is necessary for the successive boils? I think pressure cooking just the first time is sufficient.

OOps posted wrong thread, cya later

I think a more accurate way would be to do two side-by-side 1st boils, one with whole and one with ground seeds. I don't think much would come out of the 6th boil, and interpolation is also unreliable, we may be looking for a pattern where none exists (a quick look at the results shows the sequence of the ratios of the weights of the (k+1)-th to the k-th boil is not monotonous).

My seeds seem to have a 1:2 harmine:harmaline ratio. The use case I'm the most interested in is the creation of a Caapi copy. In that use case, for which 1:1 is enough, I'm left with harmaline (or THH) I don't know what to do with. So I don't feel strongly motivated to get even more harmaline out, unless there is a way to convert it to harmine.


It's not necessary, I just prefer it because it's clean and efficient.

Okay. Yeah I just wanted to design experiment that gives a clue without being too labour intensive. I assume there will be a serious bump in the yield as well as alkaloid ratio is the seed cores have different composition. But doing the extraction all the way side by side is easier - I don't know why I always come up with only complicated ways

My rue seeds seem to be the opposite, I get almost 4:1 harmine to harmaline. I am therefore short of harmaline to make THH and looking different ways to get more harmaline, either by hydrogenation of harmine, sourcing other seeds or trying to grind them.

But the seeds are in the PC anyway, they don't even leave my PC to exchange liquid, here's a dirt cheap vacuum pump used but sucking on that tube works as well.
(Btw the red filter failed somehow to let the liquid trough.)

As Jugabe said they are so cheap/sustainable we shouldn't be very anal about a 1% gain here or there.

Nice report ! thanks !
It's coherent with my visual experience ;
But the question of "how many pulls" is also link to the quantity you process ; for 1kg it might worth a 3rd pull for sure. Also, it must be highly dependent of how much water is used.

Yes. My only large PC now is the 30 Quart ALL AMERICAN beast. It's a pain to strain the water from it, especially when it's still hot from the stove.

I never thought about siphoning off the liquid, good idea.

I think I will just pour off the supernatant and a little seeds into the strainer. The returns are diminishing anyway so it should be good enough

Sure good enough, my setup was just like I had those things in da house anyway

Whole syrian rue seeds: boiling seems necessary (in this one-off test).

a) Tried a cold soak of 3 months in pH 3 waters, 1 x shake per week, extract.

b) Then boiled the seeds to see if anything was left: oh yes, and the next boil again a yield, then the 3rd boil looked like near exhausted, so only 2 boiling sessions needed.

Yield of a) looked eye balled like total yield of b)
so only aprox 50% was obtained in the CWE.

Very complete results were obtained from CWE's on powdered seeds, not so on whole seeds.
I've read several times "The actives are in the outer peel of the seeds." but with these results I doubt that. Where did that "outer peel actives" stance originate from?

PS: with finely shredded MHRB there was a good result with CWE so I gave the whole seeds a similar go.

Happy trials...