Thank you for this SOP! I just did my first run and had some questions, maybe you can help. Do you put your ultrasonic homogenizer on full power or do you start at 80%? Also what pump would you recommend to work with the filters you suggested? During filtration it took quite a while so I’m not sure if my pump is not strong enough or the particle size was not small enough.
If it didn’t go through the filter, then the average particle size was larger that the pores in the filter…
2 Likes
If he is filtering the final product (drink), there could be other ingredients that are causing the pores to clog. If he is filtering the concentrated water soluble, then he is overwhelming the pores and experiencing nanoparticle aggregation.
2 Likes
Thank you! That’s what I figured. So should I just run it longer? And should I be using the homogenizer at max power?
You dont need to use max power, but I would still use a good amount of power. You can process the stuff multiple times.
1 Like
I would suggest sending a sample to the lab for particle size analysis. If you are just starting to do some production and want to understand the formula you are working with and its properties, lab analysis would be the best first step.
This is also a great way to bring efficiencies to your workflow as well as understand your parameters and their effects on the formulation.
For example, if this is the first time you ran production, I would write down the time you were sonicating as well as the RPM of your stir bar if you had one, the temperature your solution was held at, etc… then take a sample. Go ahead and adjust one of the parameters, maybe higher power on the sonicator? Run a second run with that single parameter adjusted.
Submit them both for particle size analysis and that should help you understand the impact of the power level of your homogenizer.
This will also tell you if you should expect your solution to easily filter through your filtration membrane or if you get a result back that says your average particle size distribution is at X and your filtration membrane has a pore size that is less than X, it’s clear it is the size of your particles that are inhibiting filtration and not some potential secondary effect from the matrix.
3 Likes
You should run it at the lowest percentage possible. Not saying 10%. But shouldn’t need to go 80-100%, 75% will probably suffice if you have enough resonancy time on the horn
1 Like
What state are you located in?
I recommend running at 70-80% amplitude to reduce the degradation of the horn. That being said, your horn is still going to look like swiss cheese after a few dozen hours; this is the consequences of intense ultrasound on metal.
Now you are in the weeds of optimizing your solution so it can be sterile filtered. This is actually quite necessary since the titanium/stainless in solution isn’t what I’d call food grade.
When I developed my formulation I had no money so I used the data I had. The fact that it didn’t sterile filter is actually useful data. No filtrate means your median particle size is >220nm. If it filters at first but then fouls your around 160-200nm. If its cloudy but filters, you’re around 100-125nm. Once it starts getting translucent, it should filter without much issue and you can make educated guesses with 10nm being transparent, 30nm being very translucent, 50nm beginning to become translucent (translucent in a pipette), 100nm is cloudy.
Also, what is the membrane material you’re using? I’ve found PTFE don’t filter well even with translucent formulations and it doesn’t filter with larger particle emulsions.
4 Likes
Thank you! I am using nanostabilizer LT. My formulation was 11 grams of distillate (testing at 930mg/gram) to 55 grams of nanostabilizer mixed on a hot plate at 50 degrees celcius for 30 minutes. Then added 134 grams of distilled water for a total of 200 grams. After the water was added I left on the hot plate mixing for an additional 30 minutes. Once fully mixed and a thick milk color I ran in the homogenizer at 80% for intervals of 15 seconds on, 3 seconds off until a temperature of 48 degrees celcius was reached. At that point I cooled the solution in an ice bath and mixed it until the temp came back down below 20 degrees celcius. I repeated this process for a total of 5 times and the solution still wasn’t fully translucent, it still had a slight cloudiness to it. I left the solution in the fridge overnight and came back the next morning to try and run it again. It immediately became a lot more transluscent. I ran it two more times after that with an ice water bath in between. It seemed pretty translucent at that point so I put it through the same filter mentioned in this SOP. At first it filtered ok but slow. After about half of it was filtered it slowed down dramatically to the point where no more was filtering. The homogenizer I used is the 600w ultrasonicstor cell disruptor from ussolid.
One other thing I noticed is after homogenizing it starts to smell like burnt rubber so I’m not sure if it’s heating too much and burning or if that is normal. It is soluble in water (has been stable for two days now) and can definitely feel the effects (put .2mg in a bottle of water to get around a 10mg serving). It does have a bitter taste to it.
The last thing I can think of is should I have it on a stirring plate the entire time it is being homogenized? I couldn’t fit it inside the homogenizer box.
Thank you everyone!!
1 Like
New York
You can use the lso stabilizer, dry or out and dose with that. I can give you the ratios I use (for a small fee) to use so you’ll be around 100mg per g of powder. In higher doses, it doesn’t taste nearly as bad.
1 Like
25mg/mL is quite concentrated, you can get away with it but you generally get better results when running more dilute. 10mg/mL is a good rule of thumb for non-optimized formulation, then once you tune it, concentrated runs are not really an issue. I tuned mine so it can go up to 50mg/ml, but I typically run it at 30 mg/mL.
You also likely need to run longer. I have that same homogenizer and I run 120g of distillate for 15 hours, sometimes as much as 30 hours; for you that’ll translate to 1.5-3 hrs depending on your reactor geometry. I don’t stir mine since the ultrasound does that, but you can if you wish. It sounds like your formulation is decently optimized, if you’re using the ISM pre-mix, a 6:1 ratio should get rid of that last bit of cloudiness according to their marketing and webinars.
1 Like
One thing to keep in mind, if you are using a sonication horn, there is often microscopic or sometimes visible with the naked eye, flakes of metal from the sonication horn that will be in your sonicated solution.
This can also clog the filter and slow down filtration, in that case if you can filter half through one filter membrane, just filter the other half through a fresh membrane.
2 Likes
What is the small fee? And don’t I need a spray dryer?
1 Like
You are all awesome! Thank you all so much for the help!
1 Like
I’m going to throw some interesting technology here because i’ve wanted to develop it into a product for years and simply have not been able to find the bandwidth to.
This technique has been tested in thc/quillaja system and produced a lower particle size emulsion with a single pass than a microfluidizer unit did with several passes.
To run this, take a tubular membrane with a particle size somewhat smaller than your desired nanoemulsion. (for instance, if you want 50nm, use a 30nm membrane). The membrane has two sides – the side that has the nanopores, and the side that has the bulk support.
Using two pumps, introduce very, very high crossflow on both sides of the membrane. Have water running on the nanoporous side, and a premix of ethanol, cannabinoids, and emulsifier on the support side.
Then, introduce very minimal pressure on the support side. The premix will begin to go backwards through the membrane, and extruding through the 30nm pores. As it extrudes, tiny 30nm droplets are produced, and the fast crossflow of the water immediately washes them into the bulk solution where immediately configure as liposomes.
The limits of how small of droplets can be produced by this method have not been explored by me, but I know at least 50nm is achievable and I think no doubt much lower could be as well.
This method is not appropriate for use with spiral wound membranes as it requires back-flowing – however, tubular, flat sheet, and hollow fiber types may all be potential ways to scale this up.
One can also make a micro emulsion in advance using just a blender or similar, and extrude that, with slightly more pressure the micelles will be sheared through the tiny pores, reducing their size. This method has the advantage of only needing one pump.
9 Likes
Thanks for your post want to get this setup for my gummies 
It a expensive setup but the effect of the nano thc must be great!
would it not easier to have it in powderform so it can be used for edibles tablets water? What is the + of make it in liqued form?
cheers
It is easier to use in a powder form. You can dose stronger and it tastes better. But requires more equipment to be done and a more expensive surfactant. Liquid surfactants are extremely cheap if you find it at the right place.
For example:
2 Likes