1. We are currently experiencing issues with viewing and uploading images, our team is working on the issue.
    Dismiss Notice

Using a prism to monitor cannabinoid distillation

Discussion in 'Concentrates and Extracts' started by eugeneoregon, Sep 12, 2017.


    eugeneoregon Active Member

    It works great as a periscope. It is crystal and does not mind the heat. Cheapest piece of glass in my house!!!

    BobbleHead Member

    Great work EO.
    I saw your stuff on youtube too and posted a comment but all was disappointed it was all deleted before I had your reply.
    Was wondering about the actual pressure within the still
    At the low Pressures you are working (molecular flow), I was wondering if you had tried any experiments with widening and shortening the connection between the bellows and still. The pressure in the still is very likely higher than where you have the gauge located, but there is no way to insert the gauge into lab glassware. All you can do is make the connection as wide and short as possible.
    The fact that you are able to distill the good stuff at about 138 degree C likely means your are attaining a high vacuum within the still. From a practical viewpoint that is a good enough vacuum, as it is below the degradation point of THC which begins around 200C
    On one of BP open sources videos, they are at around 200 microns (by their gauge) and the head temp is 211C when the good stuff comes over. The published value is 155-157C at 50 microns. If these are correct (a big IF), the pressure in your still is 34 microns. It also gives the heat of evaporation as 31.8 kJ/mole.
    Keep up the good work

    eugeneoregon Active Member

    I believe your pressure prediction to be innacurate and at odds with the observable facts. Let me explain.

    The published boiling temp of Delta-9 THC is accurately described on the opensource website used by chemists and in place by the government - Pubchem. They accurately list the boiling point of Delta-9 THC at 392° F (200C) at 0.02 mm Hg (20 microns). Here is the page link.


    Since Pubchem is peer reviewed and deemed accurate for government labs let me assume for this post that they are correct. If correct, then the only way for the temperatures in my process to be lower than pubchems temp is if the pressure at the boiling puddle is also lower. They had to pull 20 microns to get it to boil at 200C. I have demonstrated distillation at temperatures now 60C less than that listed in Pubchem.

    This information along with two other data points and using a single point conversion chart (a very crude method of predicting pressures/temps) and averaging I can detect no deviation from the pressure curve that suggests the indicated reading of ¾ micron is in error, and in fact all the readings seem to line up properly with what the single point chart suggests. Your prediction of pressure rise of 35 microns easily puts the boiling point over 200C, if Pubchem is to be trusted. My videos seem to confirm the data Pubchem has published.

    I turned off comments on youtube because it began to distract me. After about the third really rude comment I concluded that the distraction of unkind people would surely detract from the enjoyment I get so the comments amount to bad medicine for me. I just want to make good medicine.

    BobbleHead Member

    It is apparent to me that there is no reliable data on THC which is why I find your work so interesting and valuable. I appreciate your efforts.
    The value I gave at 50 microns is also a published value

    I can't post links so look up "Isolation, Structure, and Partial Synthesis of an Active Constituent of Hashish" in the Journal of the American Chemical Society

    One of these published values must be wrong as the BP must be lower at a lower pressure, not higher. Who knows which is correct.
    And to me, all of these values are suspect due to the method of pressure measurement. They are all measured on the other side of a small barbed connection which has a very high resistance at very low pressures so the pressure is undoubtedly somewhat higher in the still.

    In the molecular flow regime you can use this table to calculate the conductance of any tubes. Valves and elbows are also a problem

    I cant post links so google Lesker Dushman's Table

    as you can see a connector with a 0.4 cm inner diameter and 1 cm long will only allow a pumping speed of about 0.3 l/sec, no matter what your pump speed is, and a small amount of water vapour, alcohol, off-gassing, or tiny leaks will make the pressure in the still higher than at your gauge.

    You may be able to check this by attaching your bellows directly to one of the pigs (or Kugelrohr bulb) ground glass joints, with as large a quality tube as you have. This would give a shorter wider connection and would decrease the pressure in your still and make the gauge a more accurate representation of the pressure in it.

    The one from open sources may be accurate because at 200 microns they are in viscous and not molecular flow. They had 211C at about 200 microns.

    there is a useful tool if you google BOILING POINT CALCULATOR from trimen.pl to see which points fit

    It would be great if you could solve this question once and for all. No one else has been able to.

    And it is all about good medicine my friend.

    eugeneoregon Active Member

    Using the single point conversion chart I computed the predicted temperature for boiling at 75 microns based on mynown measurements I observed and cross referenced to the official Pubchem citation which is the current government information.

    At that pressure of just 75 microns the computed boiling temp is easily over 300C. VERY EASy. The reference you cite is tyypically used by very old publications. Even Wikipedia has it wrong and disagrees with Pubchem which IS the defiinitive and correct valure maintained by our government. I have studied this in depth for years and recognize your reference. As I recall they arrived at that bogus value after testing some hash and the reference you cite was published in 1964, the first year THC had been isolated and identified. This value represented their best estimate and was wrong.

    There is no error in my Pirani gauge nor in the proper way it is installed. As you point out there is no way to test and observe your theory which does not seem to agree with the known and observable data that I presented. There is absolutely no way that a distillation carried out at 200 microns was done at 211C IF they were distilling delta 9 THC. It cannot fit the equation for predicting pressures even close IF the two temps of Pubchem and the one I posted are accurate.

    What is not in dispute is the measured temp I observed, and there is simply no way that the data you present fits in the same data set while at the same time that temp and pressure agree with the Pubchem data (by computation). Your error is simply because you have not observed those other values for yourself and so you rely on anecdotal evidence. No harm there but your evidence is outdated and unreliable as a practical matter. I will leave my configuration as it is because it works.

    BobbleHead Member

    Yep. Can't argue with your success so I understand why you don't want to change it.
    The values at the lower pressures are highly suspect for the reasons I stated. I worked as a research chemist for years so I know how these things work.
    I am gathering the equipment to allow a larger bore tube from my bellows to my microscale distillation setup and I will try and repeat your work.
    I have considered adding a turbomolecular pump but then really you need at least a 7 cm diameter tube between the pump and the still for it to be useful. Getting low pressures actually into the still really does require a large connection when you get to these really low pressures.
    And by the way, using the boiling point calculator I mentioned to you, I put in the data you have produced at 0.75 microns and the Marinol data you presented, and then I put in the pressure that open point was using (200 microns). This calculator predicted the boiling point they should have observed was 229C, based on your data and the Marinol, so that is pretty close. So maybe all three of you are right. Let's hope so. And lets try and refine the temp/pressure curve for THC.
    (I just read in a news article that pure THC for clinical research is $2,000 a gram.)
    I am still unable to post any links so you would have to look up the link for the calculator as previously indicated.

    BobbleHead Member

    I was just reading "Cannabis and Cannabis Extracts:Greater Than the Sum of Their Parts?" by John M. McPartland Ethan B. Russo from a book on HIV and AIDS. They give references to the BP of THC as being 157C at 760 mmHg along with that of other cannabanoids and terpenes. This is an often referenced book, but I'm sure we both agree this is way too low. I'm not sure why this number seems so elusive or why 157C seems the most often referenced number on the net. It is clearly wrong.

    Chemspider has a predictive method of calculating the BP of THC and they say it should be 390.4±42.0 °C at 760 mmHg. If I plug this in and the published Marinol value into that BP calculator along with your P of 0.75 microns it predicts your BP to be 147 which is pretty close to your 138 given all the uncertainties. In fact if you use the upper limit of chemspider's estimate they are within 3 degrees of your findings.
    I will still perform the experiment with tube size but so far the experimental evidence points to your pressure reading being adequate and the following as the best available data points:

    a BP at 760,000 microns of about 390C
    a BP at 50 of about 200C
    a BP at 0.75 of about 140C

    which yields a heat of evaporation of 116Kj/mol

    as a final check this would predict (again using the BP calculator I referenced above) that open source should see a BP of 223 at their P of about 200 microns and they get 211, pretty close. ( In a comment on their youtube video they say "Full vacuum on our Welch 1400b Duo Seal. With vapor pressure of cannabis @ 0.9cfm on the pump we see around 175-250 micron" and again you can see around 211 on the head thermometer )

    Anyway we are in the right ballpark

    BobbleHead Member

    well I messed that up it should have said BP at 20 of 200C

    eugeneoregon Active Member

    Nice work.

    The problem with using my data set is the inherent and known limitstions of the pirani sensor. They cannot provide accurate measurment below one micron, so when the gauge reads ¾ of one micron it means "less than one" and not necessarily ¾ of one. If the actual vacuum is ½ of one micron we would still see ¾.
    So the data I present has that limitation. plus I am unaware of the protocols used to set up an experiment for official measurement of BP/Pressure. My protocol relies on visual indication of distillate, and not "first drop" from the thermometer bulb in the boiling flask. So the angle of my rig certainly impacts how the distilation proceeds and when "first distillilate" is detected. I have no doubt this is not the standard against which your calculator computes these things.

    That seems to me to be a great effort on your part. The differences that you noted are possibly (likely) owing to my deviation from a standardized protocol, and the fact that the true vacuum pulled can only be assumed at ¾ and not lower.

    BobbleHead Member

    Hello EO
    I know you have experimented with your equipment quite a bit, for instance different tubing, gauges and greases.
    I was wondering if you have found the need for expensive pump oils, they go to insane prices.
    The pump I have (Edwards RV-3) goes down to about 50 microns right now and it's ultimate is advertised as 3 microns. I've changed the oil, checked for leaks etc. to no avail. It may need a rebuild, but I just want to make sure that this oil I am using is adequate.
    I'm using JB Gold DVO-24, which isn't cheap, but I haven't been able to find the vapor pressure nor how good this is at the micron / submicron level or if it could be preventing the RV-3 from getting as good a vacuum as it might with a different oil. 15-20 dollar Vacuum oil looks to be capable of 0.2 microns, whereas some of the Dow Corning, silicone based stuff is $200+ a qt but goes down to 10-7 micron range..
    I have a Welch Chemstar 11402 N on its way so the oil may be even more important then, as it has an ultimate Vacuum of 0.1 micron, below my gauge as well.
    Any comments, especially with regard to oil, would be appreciated

Share This Page