It may be worth trying 1,3-propanediol at a higher concentration by adding it at 2%, 3%, and 4% calculated by weight/weight. Then use the lowest effective concentration. Per FDA GRAS notice GRN. 302, the maximum concentration in an alcoholic beverage is 5%. That would also be at least the maximum concentration for non-alcoholic beverages.
Also, someone may have mentioned this, but temperature control is essential. I have found bitterness increases under higher temperature processes liquid.
Are you sterile filtering your nanoemulsion?
All grades of polysorbates are generally going to have bitterness, IME. Have you found otherwise? It’s too bad synthetic surfactants and emulsifiers are not an option for people who want to make a clean label product because they work so well.
That’s interesting! I use an overhead impeller mixer in the process tank tank for agitatation and use a high-shear roto-stator for the rough emulsion stage. I haven’t thought about using a high-shear rotor-stator homogenizer instead. Compared to the impeller mixer you wrote about, does the roto-stator improve your process?
A few questions:
What is the throughput of your hybrid system to produce a transparent emulsion, what is your flow rate, and what is your oil load?
Does the rotor-stator mixer increase the process liquid temperature?
It looks like you’re using the top port on the reactor chamber as the inlet. If so, have you tried using the bottom port as the inlet? I ask because it’s generally preferred to have the process liquid flowing toward the sonotrode’s horn tip to increase cavitation and mixing within the chamber.
What are the dimensions of your sonotrode’s (aka probe’s) horn? You should post a picture so we can see what you’re using. An off-the-shelf flow cell reactor vessel may work for you. I know someone on the forum who has a custom fabrication machine shop that I’m using for custom machine SS tubes. LMK if you want his info. To save money, you could use an in-line tube heat exchanger rather than a jacketed flow cell reactor vessel.
Are you using a jacketed batch reactor vessel now?
You’re correct; the ultrasonic transducer/converter frequency should be 20 kHz, and ideally, the sonotrode’s (probe) amplitude at full power is 120 μm (a minimum of 100 μm).
So, I personally think that the super refined polysorbate that I use is less bitter than other options. I agree that it will probably never be completely devoid of bitterness, but I would say I felt the impact was a bit more mild using a super refined version. I source my super refined polysorbate 80 from Croda, so there may be variations between manufacturers or suppliers and I’m sure it’s one of those YMMV situations, especially because everyone has such a different perspective on taste.
I do feel that using the high shear homogenizer as my initial mixing and homogenizing apparatus does reduce the amount of time I need to recirculate the formulation through the flow cell for a consistent and reproducible particle size of my final product. I also think that the high sheer homogenizer does a much better job of mixing my bulk formulation to begin with vs an impeller. Now as I mentioned, the formulation that I normally use contains water, liquid coconut (mct) oil and glycerin as the formulation base. These three liquids are already a bit more difficult to mix together as it is, and that’s not even getting into the dissolution of the solid ingredients that I have to add as well. So based on visual observations and my own non-data based assessment, the high shear homogenizer does provide me an overall better formulation experience vs an impeller.
Here are my answers:
1). Right now I can run about 6kg of my formulation and I run my peristaltic pump at “50”. I would have to go back and double check what that means as I don’t think it’s really a specific unit of measurement, it’s a value that will equate to a certain flow rate. I just don’t know what that actually is. With that in mind, I’ve found that after about 20-25min max of recirculation, I have been able to get a very consistent particle size result. My formulation is a concentration of 1% w/w D9 and I often use oil between 85-90% D9 purity. Keep in mind that my product is not designed to be an aqueous product and must have a water activity level below 0.85, so this formulation does NOT produce a transparent emulsion, but it looks like milk. Even with that in mind, you can see below that my particle size is still below 200um, I have a very smooth particle size distribution curve, and a consistent zeta potential (although I don’t believe the zeta potential means much in this case because I am using non-ionic surfactants, so the value itself may not necessarily dictate stability). I have posted the results I normally obtain from my finished product below.
With all of that in mind, when I was making solely aqueous based nanoemulsions, I was always doing small batch sizes and I haven’t run that type of formulation on this setup. I would have to dig through my data but I’ve been able to create some formulation mixtures that are transparent/translucent at about 20mg/mL but this is using a high energy method and I never really tried to go above that because I never needed to for any R&D application I was doing. So I am unfortunately not able to really comment on or describe how well my system works with a fully aqueous based =< 50um particle size formulation.
I have not found that the high sheer homogenizer has as big of an impact on temperature increase as the circulation through the sonication horn does. I do get some heat generation from the homogenizer, absolutely, but I never really get close to my 50C upper limit until I turn on the sonication horn and it’s been running for a few minutes. Luckily the flow cell is jacketed so I can help control the temperature a bit more directly at the flow cell, but in general I rarely ever hit my upper temperature limit unless I let the horn run for like 10min non-stop and it starts to produce more heat than the jacket can offset.
So I pump from the bottom port of my batch mixing vessel into the top port of my flow cell, which means the liquid should be running from the top of the horn down to the tip, as you described. Maybe I’m a bit confused? In any event, I can certainly play around with the flow direction setup, it’s not something I really considered.
It may be worth trying 1,3-propanediol at a higher concentration by adding it at 2%, 3%, and 4% calculated by weight/weight.
I appreciate the tips there. I will certainly try it more before I write it off. I also have b-cyclodextrin on the way that may have some synergy.
Are you sterile filtering your nanoemulsion?
So I do have sterile filters. I have not filtered it yet. I have not consumed more than 5mL of it but I will certainly filter it before I get heavy metals screening and most certainly for production batches. Any idea if filtering could improve taste? I have noticed a very small amount of foam on the surface of the nano mother liquor (on the order of <1-2mL).
What are the dimensions of your sonotrode’s (aka probe’s) horn?
Bit of a personal question don’t you think?? My probe is ~22mm diameter and ~4in long. The horn is ~4.5in. The only issue I am seeing is that the tip of my probe is flared. May prove challenging to schlide a flow cell over and get it to seal…
Are you using a jacketed batch reactor vessel now?
I just got my jacketed 1L glass beaker. I was using a waterbath to control the temps before. At first I didn’t think to use the waterbath, and I think I got the solution too hot. Any idea if that could be the source of at least some of the bitterness?
I’m sure this is dabbling into some IP/NDA, but it can’t hurt to ask… Is this for cannabinoids in general? Or are we specifically talking about THC? Or are these derivatives of cannabinoids with additional polar side chain groups such as Carmagerol (Polar CBG)?
One of the ingredients is fairly expensive, we have different grades as we do wholesale the delta 8 powder. We have a buick version and our ferrari which is the clear tasteless version.
Sell clear tasteless for 1-1.8 cents per mg for CBD
Cool. From the image it looked it your inlet was the top port. My bad. And thanks for the detailed response.
Depending upon your formulation and nanoemulsion quality, filtering may reduce bitterness. I’m not sure what you mean by “nano mother liquor,” but a finished nanoemulsion shouldn’t have foam (it should break during processing).
IME high temperature does affect flavor and bitterness, as does long processing time (over-processing). I keep the process liquid at 30-40’C.
I completely agree. @MedicineManHempCo Temperature can definitely impact the formulation, going as far as even disrupting the stability of the particle size if the temperature gets too high. I usually keep my cutoff temperature at 50C before I cease all aspects of the process that generate heat. If you were doing some work with a non-regulated water bath or didn’t keep an eye on the actual temperature there is certainly a chance the temperature could have had a negative impact on the formulation which may or may not mean a change in flavor.
Last night I just tried the n-zorbit maltodextrin from modernist pantry and it worked great, made a stable emulsion with no bitterness and pretty quick onset (within 10-20 minutes.) My concentration was 50mg THC and 100mg CBN in a 4oz syrup.roughly 5% maltodextrin by weight
Not sure if the discussion is still open. But if you are using cyclodextrin, why look to the surfactant? Just the cyclodextrin itself might be able to serve your purpose.
Cyclo dextrin is more of a “stabilizer” for emulsions. The action of encapsulating the active ingredient takes multiple surfactant/emulsifier/biopolymer combinations so even cyclodextrin on its own would macroemulsify at best.
I understand. I was referring to making the standard cyclodextrin-adduct of the cannabinoid concerned, and leaving that in the aqueous solution. That would effectively be a solution, not an emulsion. But yes, you can possibly put only around 20 mg cannabinoid in about 500 mL water. I’ve never gone beyond 10 mg though.
Hey @Ralf your DM’s are closed but I’m currently using live resin in my gummies and bitter blockers to mask the flavor. The bitter blocker helps but he issue is that the live resin sticks to my tongue long after the bitter blocker has gone away. Would love to chat with you regarding this if you think you can help.
SoluScience specializes in water soluble ingredients for beverages, however, we do have a team that has created some amazing gummies as well with great taste. I can put you in touch with them if you send me a DM. - Travis
I think there is just a lack of understanding when it comes to food science and flavor modification. I believe thinks that he can mask the bitterness of the terps from the live resin when in reality it is used to help with masking the bitterness of emulsions which for the most part do not include aromatic compounds, just the cannabinoids.
@Littletree you should look into some of the thread of developing an emulsion before trying to figure out your flavor modification.
