So this is cool.. Thought I'd share a novel method of extraction using a super wicked acoustic technology. I too am excited because this is the very first application of using resonant acoustic technology for means of extraction. It represents one of the most efficient and fastest extracting technology out there although I cannot provide the evidence for this yet. Undoubtedly this use will surface, but for here, now, let me be the first to demonstrate:

I have in my lab a mixer called a Resonant Acoustic Mixer, or RAM for short. Applications are vast, from mixing of pharmaceutical powders, to mixing of concrete, but also its uses encompass ASD creation through acoustic fusion, nanoscale particle size reduction, kind of you name it.

Obviously, when I first heard about it I thought of extraction scenarios, and now that I have one to play around with I thought I'd test out the waters.
Oh by the way, it oscillates up and down at 60 times per second (resonant frequency ~60Hz) and up to 100g of acceleration as its moving. Think of it as 60fps at 980 m/s per second.


I did 3 things:

1. Demonstrated the use of Resonant Acoustic Mixing for extraction purposes. You have your typical extraction brew (solids filtered out), basic pH, and on top you can have your solvent, here toluene. The collage below represents 8 seconds of acoustic mixing at 100g. The photos say it best, but what's happening is "Faraday instabilities" are being created through the oscillating action, which leads to finger-like projects extending upwards and downwards like stalactites and stalagmites. The surface area of the liquids is amplified to a crazy degree, and micromixing occurs in the bulk. All energy is directed towards the brew and a computer automatically monitors the power and adjusts for the constant resonance.
I did not have anything left to extract in this case so I don't have color changes in solvent to show, but it's suitable enough to show what could be possible

2. I thought maybe I could mix 2 immiscible liquids, methanol and heptane. I used a small bit of methanol to solibulize some spice and added heptane on top. I hoped that the mixing would transfer some of the solids to the heptane. The collage in this case shows 30 seconds of mixing with complete binding into a single phase of solution.

3. Supersaturation: Last scenario is I have some crystals contained with an amount of heptane. Here I hoped that the mixing would be so efficient that supersaturation would happen, and crystallization would happen without any heating. There was particle size reduction but not any supersaturation after 30 seconds. Mixing for longer could definitely make a difference, though. For obvious reasons there is concern when mixing any solvent that's heated, but there are workarounds for that that could be explored further.
Nothing much to report on the visuals before and after, but I have a super slow-mo video that I took of the crystals being mixed.
Just wait until the slow-mo stops


Hope everybody finds this as cool as I do !

Thanks for sharing, any novel type of experiment in this area is cool and appreciated

What would be the advantage/difference between this and say, an ultrasonic bath?

Wouldn' t it create emulsions when mixing the non-polar solvent with the aqueous plant solution? Maybe it would be more appropriate to use during the aqueous extraction of the plant, when a thorough mixing is actually desired and emulsions aren't an issue?

Hey endlessness,

Good question on the acoustic vs ultrasaonic energy difference. Acoustic energy as represented here can be considered as a vastly superior form, in several ways:
* Ultrasonic energy operates in the 1000's of KHz (MHz) range whereas the operational frequency for the former is much lower, in the 10's - 100's of KHz. Sonication provides highly localized regions of distribution but the acoustic energy in the case I'm describing here is transferred entirely through the bulk mixture. Inertial and stored forces are exchanged during resonance and all the input and mixing forces are put directly into the material.
* While the impedance of the vibrating system is continuously being matched to the load and resonance maintained, this energy transport reaches the microscale eddies which makes solute diffusion orders of magnitude higher. The acoustic streaming causes these microscale vortices to be created everywhere in the mixture, uniformly.
* In the case I'm showing, apart from the distribution of the acoustic energy, there is mechanical oscillation which increases the interfacial area enormously and reduces the mass-transfer resistance. The way all these forces and factors come together changes the thermodynamic properties (activity, chemical potential, entropy etc.) in favor of extraction and the kinetics of solute transfer are exponentially higher than what sonication can provide.


In terms of emulsions, I have not found it to be an issue. When I saw the mixture after shaking my heart kind of sank because it looked like it just made a mess of emulsions, but I believe it was just the visual result of the intimate level of mixing, because full separation occurred by 20 minutes later.
Generally, nowadays I try and filter every last bit of solid out after the aqueous (or, in my case 70:30 H2O:MeOH) acid extraction from the bark, first by transferring out with a bottle top dispenser, then decanting as needed, then filtering through like a 3 um filter via vacuum filtration. When defatting, I've found that it really helps by letting the defat process (2 is good) to occur over at least a day, better two. By the time you basify, separation is always rapid with never an issue of emulsions because there's no source of them anymore.

I too think it would be good to do the extraction from the bark to aqueous itself here. I think it would be crazy efficient. Not sure if this is exactly related, but the machine can super-load solutions more than is possible through other means, one example is polymer loading in water, you can reach 80% solids content when mixing this way which simply can't be done when anything else. But I bet this doesn't hurt the case that it would be good for this situation!


Thank you kindly for the support!

Thanks for the explanation!

It does sound very interesting! Would be awesome if you could do a side by side test in some extraction with normal stirring or letting it sit or sonication versus RAE.

Another thing I wonder is how well it would do in extractions where the plant material isn't powdered but rather shredded and therefore there isn't so much surface area to expose the plant material to the solvent.

One more question: What is the volume capacity of this resonant acoustic mixer ?

Let us know if you run any more experiments with this

Hi Samvidbuho,
Serious apparatus such an entry model mixer @ 70kg (155lbs) for a 500grams processing ability.
https://resodynmixers.com/product/labram-1/
I guess modding a hifi woofer won't have the structural integrity and necessary power delivery

Haha Jees, well you're off to a good start! You got the right idea anyway, acoustic energy can be of tremendous assistance. A simple and common phenomena which is becoming increasingly well harnessed (nowadays they even have scanning *acoustic microscopy for 3D structure visualization, the potential is wild)

endlessness,
like Jees is saying, Resodyn makes some serious equipment that can handle up to nearly a ton of material (920 lbs). I'm using the 500g capacity one, but the unit is bigger than one might expect. Any sized or shaped container that can fit can work, so for small-scale R&D type stuff it's very sufficient.
As far as using shredded bark instead of powdered, hm I am thinking it maybe would not be so well. The mixing is high intensity but low shear (it was actually invented for working with explosives!), and reducing hard agglomerates typically requires milling media such as zirconium beads. In such respect, I suppose the initial effects of size of the bark would eventually be eliminated and overall extraction efficiency would probably be higher since you've reduced the particles to micron-sized pieces. But without doing this you'd probably be best off working with powdered material.

Lastly, I'd be curious to see SBS comparisons too. Ok! I am wondering if there's anything I could simulate the extraction with sans root bark. Do you have any ideas? It would probably be worth ordering a little bit just for these tests. Thanks for expressing the interest. It's refreshing to have encouragement in experimentation. There's just, no better way to gain knowledge am I right?

Sounds like interesting equipment for sure.

As for simulating extraction, can you not do an extraction but of a legal compound? Caffeine, piperine, curcumin or whatever ? Loads of legit reasons to extract any of these.

And yeah definitely no better way to gain knowledge than experimenting, and it's great when you share it because we all learn together, whether it turns out as expected/desired or not

Ah, so, I too wondered about caffeine since it's actually something I'm working with anyway, but I'm hesitant on the usefulness of any generated data. Similarly too when I threw some sugar into water and mixed it - all the sugar dissolved, but then it becomes more a matter of dissolution or solibulization and not quite extraction per se. Even trying to supersaturate caffeine in water might not be sufficient to show extraction power because the environments aren't the same, in that you know the extraction media contains loads of other compounds which interact in the microenvironment and complicate things just because of their presence which a binary mixture couldn't mimic.

So I was thinking like dried and diced orange peels in an organic solvent or something like that. Or a common alkaloid-containing culinary plant?

Ok I decided:
I'll extract caffeine from coffee grounds with water and run samples at different time points on HPLC. That way we can see extraction power for other lurking analytes too.
I'll try and keep the volume of water saturated with respect to caffeine content

Sound good? (endlessness) let me if there's something else you see fit. I should be able to throw this on by the end of the week

Alright, I was able to squeeze in some extraction tests today for 25g of coffee in 100mL filtered (18.2 MΩ) water.

All just for fun, I'm not working with a validated coffee extract method or anything but the caffeine method is. What else was detected just happened to come out at the wavelength I was testing, so there's definitely more to it but the purpose is still served

Because of limitations in timing, I just did a quick 5 minute extraction, taking samples at 15s, 30s, 60s, 120s and 300s. Resonant Acoustic Mixer (RAM) side by side with sonication and stirring with stir bar. Centrifuged the samples and diluted them before analysis.

As can be seen, RAM (or RAE here) extracted almost 1.5x more substances in the first 15s and continued to extract both more caffeine and other detected substances (12 total RS) throughout the entire extract period. At 2 minutes it extracted as much as the other two methods did in 5 minutes. After this it looks like the resonant method is still extracting on a first order basis while the stir bar method may be approaching some sort of limit. Sonication performed the lowest in this case, probably because of the limit of bulk motion. I hand swirled the sonication series, but the stir bar I put into a rapid vortex. The particle size of the coffee grounds is another factor. I imagine the extracting power of the RAE method would be more enhanced at smaller particle sizes than the others. In fact, the samples I took from the acoustic mixer were clearly of a significantly darker and more complex color than the rest, and I bet through some other lens that would have shown more.

Anyway - there you have it. Only 25g and 5 minutes for the study but you can imagine that with longer times and larger amounts the differences would really show. In fact I'd like to show this myself some time coming up

Update: It's been a long day, the graphs showing the minutes are wrong, should be 2 min not 2.5

Excellent work friend! I love the science

Im surprised the sonication did the worse but your explanation makes sense due to the lack of bulk motion.

I also wonder how passive extraction (simply letting it sit in the solvent) would do compared to others, so if you ever run that experiment again try to consider adding this too.

How much does one of these RAM cost? Hehe

Thanks for all the work and sharing your results!

Thanks friend! I'm glad you liked this. This was pretty fun to do.

Wonderful idea to compare all these to simple [maceration?]. Definitely need to add that control you're totally right. Next time I want to extend the time longer too, say 15 - 30 minutes. Hehe, I kind of made a mistake this time too, the pipette tip I was trying to use to draw in the centrifuged liquid didn't fit, so I used a different one with a different tip which actually fit, but it pulled up such a small amount I was utterly confused but I carried on. Only later it hit me that the volumes were measured on a 10x smaller scale. Lol. Next time I'd like to dilute what I intended to (1mL / 100mL not 0.1mL / 100mL)

And, as far as the sonicated sample goes, I have a few extra comments,
1. There was about a minute time between the last 2 time points that I wasn't swirling it by hand. That could have decreased the extraction maybe?
2. It maybe would be good to compare a simple caffeine + water solution afterall in order to test a situation where the advantage of sonication in reducing particle size is evinced. I'd like to see if RAM still outperforms. In this case, the sonmicator didn't have enough energy to break up the grounds very well I don't think, so that and the lack of any complete mixing of bulk fluid with bulk grounds probably lead to these results.
3. Mildly triturating with mortar and pestle might be interesting too. In my visual observations it seemed like the resonant mixer actually reduced the size of the grounds considerably, but I'd have to take a better look next time.
4. Lastly, also speaking of visuals, I had thought the sonicator had pulled up a lot more than the stir bar, simply because of the coloration of the flasks. The RAEd flasks were still quite browned in color, the sonication less so, and the stir bar I thought looked the least discolored. I needed to use amber flasks for 4/5 of those samples, and since I was kind of in a rush with other things I didn't check. But I think this was the case. So could be the sonicated samples did pull more other compounds besides caffeine than stir bar. Hard to say

Lemme see if I can take this study a bit further next week.. you have me curious too

Price is 37K. Which is kind of a steal if you have a bunch of items on your wish list like I do that kind of total over a million... I'd highly recommend. You can rent one for a month for just $3,000 and put that cost towards the purchase if you like it


Appreciation your input as always

P.S. Fixed the graphs

Quick little update,

I was curious to compare sonication to RAE on the caffeine standard itself. Thought this would be the closest match up so the results were a little surprising, but not really unexpected.

In short, caffeine in excess was added to 20mL water, and extracted for 5 minutes via sonicator (ultrasonic method) or resonant acoustic extractor (acoustic method).

Visually it was striking because the caffeine solid in the sonicated jar was chunky / fluffy, and when shaken caused a great deal of particles to briefly suspend in solution, very opaque. In contrast, after 5 min with the resonant mode, the bulk solution was entirely clear, with a ring of caffeine tightly bunched and stuck together touching the solution and above it. When this jar was shaken, only a few particles were seen whirling around. I tried to capture this is the attached photos.

Results are clear though: Resonant Acoustic Extraction dissolved close to 70% of the maximum capacity water has for caffeine (~21.7 mg/mL at saturation) while the Sonication only extracted ~45% of the maximum. Basically, RAE extracted over 100 mg (1.57x) more caffeine than the sonicated solution in just 5 minutes. Dope.

What's cool though too is how the RAE changed the caffeine that wasn't extracted. Sonicator didn't do much to reduce the particle size at all of this excess caffeine, but the acoustic energy really made a huge difference

Thanks for the descriptions and sharing further experiment results!

Id love to experiment with one, who knows maybe in the future it will be possible to have it in our lab...

Let us know if you do further tests!

Be well

Brother, you'd have so much fun too. The possibilities are just like, endless, and it is so as of yet unexplored.

At your continued encouragement, I do have something to add from yesterday's study which is really cool:

Today when I looked at the two jars with the caffeine, the RAE solution was completely clear. Maybe 25-50mg stuck to the sides, but it became incredibly supersaturated (at least 1.5x the literature saturation value of caffeine in water). But super interesting that the was more extraction even after the mixing. Maybe the molecules were so agitated that they were still in motion for a while?
The sonicated jar was completely cloudy, full of crashed out material. Photo attached.

I tested a new sample of the RAE today and it exceeded the method limits of detection.
Most curious of all maybe is: in one vial I had the filtrate I just tested. In the other I had a portion I drew by pipet. After I sampled the filtrate, I combined the solutions, and the next time I checked over 150mg, at least, crashed out of <10 mL of water. The vial was nearly completely caffeine crystals and hardly any liquid. It was nuts, I think a small particle of solid got added to the filtrate and that little seed was enough to precipitate out a huge amount. The sonicated solution must have crashed from self-seeding.

But this opens up a lot of interesting scenarios with seeded crystallization.