Cannabis Oil QWET Extraction Battle of the Wash: Dry Ice vs. Freezer

By Troy Ivan


QWET cannabis oil extraction involves freezing cannabis and ethanol before combining them to “wash” and harvest the cannabinoids and terpenes from the cannabis plant to craft a concentrate containing the maximum amount of cannabinoids and terpens and minimum amount of undesirable components like fats, lipids, and chlorophyll. Once the combination of ethanol and cannabis has soaked for a predetermined amount of time it must be strained to separate the plant matter from the resulting ethanol solution we will now call the “wash”. The wash contains all the components that have been harvested from the cannabis plant and will later make up the final concentrate once the ethanol has been removed by forced evaporation.

Previously, I wrote a blog post “Super-Cooled QWET Wash for Cannabis Extraction Using Dry Ice” explaining the process when I was first experimenting with it and using badly beaten up trim with some good success. As soon as I finished that post I wanted to revisit the dry ice exercise to find out how far I could push the wash with better material and how it compared to simply using the freezer, so here we are with the follow up post and the epic Battle of the Wash: Dry Ice vs. Freezer!


There are many different methods of infusion and extraction, all with a different set of advantages and disadvantages. With QWET the solvent involved is food-grade 95% ethanol and it’s advantage is a high efficiency in collecting cannabinoids and terpenes. Unfortunately, that high efficiency of collection also extends to aggressively attaching to undesirable water based components like fats, lipids, and chlorophyll, ethanol’s only disadvantage. Fats, lipids and chlorophyll from the cannabis may be acceptable to varying degrees when making oil for edibles or medicinal full extract cannabis oil (FECO), but they are very undesirable for a high quality, clean, concentrates intended for smoking. Therefore, to craft a concentrate free from undesirable components the undesirable plant components must be neutralized by freezing and the alcohol must be super-cooled to keep them frozen during the wash. In simple terms, when the plant material is frozen and the ethanol temperature is below freezing the polar attraction of the ethanol and undesirables in the plant matter are unable to latch onto each other and the undesirables stay with the plant matter when the freezing ethanol is strained away, ideally leaving only terpenes and cannabinoids in the wash. You will know success when the resulting, well filtered, wash is a beautiful, crystal clear, light golden color with no trace of a green hue.

The difference between green on the left and gold on the right


For this battle my plan was to use high quality indoor dried buds to achieve nice, high yields. Unfortunately, the material I used ended up being quite crusty, not great quality, and contained quite a few seeds. Using lower quality buds may have actually been a blessing in disguise. In using lower quality buds I set the bar low for expected yields, knowing high quality dank will surpass it.

The goal in preparing the buds was to break them up as much as possible by hand to allow the ethanol to flow freely over all surfaces and at the same time cause as little damage as possible. Ripping, tearing, cutting, and trimming plant material causes cell wall damage and allows undesirables to flow unimpeded into the wash contaminating it and turning it green.



Pitting dry ice against the freezer required numerous runs of differing duration to collect enough data points and information for meaningful comparison. To this end I decided to do 3 runs in the freezer and 3 on dry ice. The washes would be 5 , 10, and 15 minutes long for the freezer compared to 15, 20, and 30 minutes on dry ice. Each run was allotted a ½ ounce (oz). After removing the seeds and stems, and breaking up the buds by hand, I ended up with 6 plastic bags containing 13 grams (g) each. The 6 happy bags parted ways with 3 going in the freezer for 24-hours and 3 in a cooler with dry ice for 2 hours. The ethanol was also split up, putting half in the freezer for 24-hours and half in the cooler with dry ice until it reached -30°F to  -40°F.

Once the materials were properly cooled and frozen it was go-time. The process was to take each bag of the frozen cannabis, pour enough chilled ethanol into that bag to allow for free movement of the material in the solution, and quickly return the bag to the freezer or the dry ice cooler from which it came to soak for the predetermined period.  For the 5-min wash I lightly agitated the it every 1-minute and for the longer washes I did the same every few minutes. When time was up I quickly separated the plant material from the wash using 2 wire mesh coffee filters with a 150 micron screen sandwiched in between. The fast, rough filtered wash was then put through the Buchner funnel filter a 2-3 times until it was crystal clear, then into a mason jar.fullsizeoutput_1539

To purge the ethanol and claim the oil I worked so diligently for I used the Source Turbo by ExtractCraft ( I used this tool for many reasons beginning with it’s the perfect size for this kind of single batch production. Secondly, because it works under vacuum and operates at around 100°F the terpenes will show little to no conversion and the color won’t be adulterated by the open atmosphere and higher temperatures giving me a true color comparison of what actually happened in the wash when inspecting at the final concentrate. Lastly, it’s going to save me a lot of money in ethanol because I will reclaim almost all of it for reuse.

Version 2

The oil produced in the Source was then lined up on silicone mats in order of extraction duration and grouped together with the freezer and dry ice samples together. Finally, the oil was placed in a vacuum oven for post process purging to make the final product into shatter and ready for comparison.




My expectation for the freezer washes was that as the soak time lengthened I would see more and more green with the 5-min wash looking good and the 15-min wash looking glaringly green. I was shocked and very pleased to see the 15-min wash come out nearly the same color as the 5-min wash. There was only a slight difference between the 5 and 15-min while virtually no difference between the 10 and 15-min wash.

From left to right: 5, 10, 15 minute freezer washes

Once the final product was finished purging in the vacuum oven it was apparent that the color of the shortest wash was only slightly lighter than the longest, but not by much.  The final products were quite consistent in color for all three samples with the shortest wash showing only a marginal advantage with lighter color.   The yields of the three washes from shortest to longest were 12.5%, 15%, 15%.

Oil from freezer washes before final purge. Top to bottom: 5, 10, 15 min washes
Freezer wash final product.  Top to bottom: 5, 10, 15 min washes



The shortest dry ice wash was 15-min so I could directly compare a freezer and dry ice run of the same duration, then followed with the 20 and 30-min washes. Again, by the time I got to the 30-min wash I was definitely expecting to see obvious signs of green in the wash, but again I was happily surprised to see almost no difference in color between the shortest and longest soaks.

From left to right: freezer 5, 10, 15 min washes then dry ice 15, 20, 30 min washes

All three final purged products came out very similar with the shortest soak again showing only a marginal advantage in quality while the 30-min soak still looked amazing.  The yields from shortest to longest were 11.5%, 14%, 15.5%

Oil from dry ice washes before final purge. Top to bottom: 15, 20, 30 min washes
Dry ice wash final product.  Top to bottom: 15, 20, 30 min washes

Emboldened by the beauty of the 30-min wash success I was determined to find the limit and see where the undesirables would finally make their way into my wash and contaminate it with the green leak. I decided to perform a 1-hour dry ice wash in the same manner as the others. Again, a beautiful wash resulted and the final product looked nearly identical to the 30-min wash and yielded 16%. I was quite shocked again.


The shorter washes in each group were more clear than the longer ones which is commonly expected. The unexpected was the very narrow margin of clarity and color quality of the finished concentrates between the shortest and longest washes of each group.

I believe three factors made the very clean, long washes possible. First, the material, while broken up thoroughly by hand suffered only minor cell wall disruption keeping the undesirables well contained. Second, through the entire wash cycle of each run everything was kept well chilled and not allowed the chance to warm or vary in temperature. Third, the alcohol and plant material was separated with great expediency, and never squeezed nor pressed.

In general, the yields of both groups increased with longer soak times as expected. The 10 and 15-min washes from the freezer were identical, both 1.96 grams. I think the 15-min wash would have actually been a bit more if I had been more disciplined and consistent with the amount of alcohol used with every wash. It is clear in the photos that the 15-min freezer wash was 2 fl oz short of the other samples. I believe this reduced the final yield of that wash to some extent, but how much I don’t know. The dry ice samples showed an increase in yield all the way to the 1-hour mark but increased at a slower rate between the 30-min and 1-hour washes. I’m convinced the yields in this experiment were low due to the quality of bud used and can be improved with higher quality buds.

The stark difference between the freezer and dry ice runs, in terms of clarity and color, clearly demonstrated that the lower temperature of the dry ice QWET very effectively immobilizes undesirables and increases color quality. I think using the dry ice performs two processes at the same time. Normally QWET is done in the freezer, then to clean up further the wash can be put back in the freezer to winterize. Using dry ice essentially performs both of these tasks at the same time.

Freezer group on top, then dry ice group in middle, and bottom is the 1-hour dry ice run.

Playing devils advocate and a brief word of caution. Many people associate winterization, with some loss of flavor pointing to a possible loss of some terpenes and/or flavonoids in the process. If the dry ice QWET does simulate winterization then this loss of flavor could happen here as well. Therefore, I would encourage those that try this to experiment with both the freezer and the dry ice, using smaller quantities and exploring which method and soak time suits your personal preference. There is no right or wrong in this craft, it’s only what makes you happy and your friends pass out. Whatever you choose, crafting with the Source by ExtractCraft is going to help you make FIRE with ease and win bragging rights!





220 thoughts on “Cannabis Oil QWET Extraction Battle of the Wash: Dry Ice vs. Freezer

  1. In my own experiment, I found that not only was the color brighter doing the dry ice method, it had more flavor from terpenes. Dry ice imho, is superior to freezer method. Thank you for your excellent review

  2. Great article, keep them coming! You mention a vacuum oven. Is this only for make dabs? Or does it have other uses? Can you recommend a product that works for this process.

    1. The vacuum oven, also known as a ‘drying oven’ is a standard piece of lab equipment used for de-gasing. For concentrates it’s used to remove residual contamination like butane, water, and ethanol. They are quite expensive and the pumps you need to run them are quite expensive as well. If you are just doing small batches for fun a simple single stage 3 CFM pump and 1.5g tank with a controlled heater is fine. You can find a package deal on Amazon usually with all of that together for $200-$300. The most common brand used is Best Value Vac

      1. Can you vape it (qwet) without this final purge to become shatter?

      2. Hi, could you supply alink to the pump and heater you mentioned. Thank you for these articles, super informative.

      3. I don’t have a link or promote any vac brands. If you google vacuum chambers, probably 1g or 1.5g and a 3cfm pump with a heat pad you should find some options

  3. Wow, I was thingking on doing the same data collection experiment once I got my Source in! Cheers for sharing the love!

    Currently testing the process with room temp 192 proof 10min soak (I distill my own EtOH) on some wild sage for my 1st run just to test it out.

    QUESTION: One can get down to nearly -80F with dry ice IN the EtOH (I have done so with a cold trap); so I was thinking about doing a 1hr QWET soak with dry ice in the EtOH. Wondering if -40F degrees more will make a difference (decrease the polar pickup of undiziribles) with cannabis?

    1. Around -40 the “polar pickup” grinds almost to a halt. In this post I stopped at a 1-hour test but I think it could actually go on almost indefinitely. The colder it is the slower the pick up of the desirable as well. If you set two jars next to each other, like with orange peel or something, and do one with frozen ethanol and one with room temp ethanol you can visually see the difference in the speed of pickup. It is drastic.

  4. Wonderful experiment! After all is said and done…. what do you end up doing most often…. did you find after reflecting more that dry ice really is worth it? Seems like the dry ice is interesting and useful but the cost might be an issue if you produce often? Perhaps one should do all their bud runs and save all trim – then run all trim in a narrow time period to save on dry ice. Also – have you ever had samples tested to see the differences that wash time and method on cannabinoid %s? I know we can extrapolate on relative differences based on color but would be interested what your 15% yield means in terms of % THC…. is that oil/shatter 50% THC or 80% THC. Such great work!

    1. Hi, thank you for your kind words. I use my stand alone freezer mostly. Dry ice takes a little more preparation and the majority of my exercises are testing and experimenting so I’m normally not aiming for absolute perfection more than to gain more knowledge about the process. Having said that, almost everything I make comes out awesome using either the freezer or dry ice. You ask, “is dry ice worth it”. I think I explain that pretty well in the article and what I said there is exactly what I still think is true. Dry ice where I am is only about $1.60/lb so I haven’t been real thoughtful about maximizing the use of it, but for somewhere that its not available the efficiency with its use would be important, I just haven’t put much thought into it.

      In Colorado there are very restrictive regulations on lab testing for unlicensed citizens, meaning I don’t have access to testing reliable testing facilities. The only tests that I have been able to get done have been in the mid 80’s and I don’t think there will be much difference between the dry ice and freezer, though just by the difference in color with the same material you would have to assume the dry ice would perform better. Hopefully, I answered your question satisfactorily, and thank you again for reaching out

  5. How do you feel the return difference is from Ethanol to Butane?

    With the quality I see here, I almost want to work towards using a less volatile solvent (Ethanol) than Butane, exe. What are your thoughts?

    1. Oh man, I have seen some killer fights start like this, its a sensitive subject for many people, so I’d like to first say I am a firm believer in to each his own and everyone has their own path. For me its ethanol because I feel it is safer, cleaner, and less hassle. ‘Safer’ I think need no explination. ‘Cleaner’ as well I think is obvious especially as the ethanol I use is food grade. ‘Less hassle’ is debatable actually. I see butane as more of a hassle because blasting is pretty messy and dealing with the fumes can be problematic, but getting a good result is not so difficult. On the other hand, some people may say dealing with ethanol is more of a hassle because getting a good result takes more attention and can be unforgiving if not done properly. There are many other benefits of ethanol but this answer will get too long and I am currently working on a blog post to that end anyway.

      Now for the yield question. This one is tricky, but by your question I am assuming you are quite familiar with BHO extraction. People report 20% and over for BHO, and I would say that is probably high, especially after dewaxing/winterizing. Ethanol is really dependent on exactly 3 things. 1) quality of starting material (same as bho) 2) quality of process 3) what you are making 4) how many washes you are doing. Obviously, quality material gives higher yields. The quality of the person executing the process is very important because if they are too fast, efficiency will suffer along with yield, or yield will be high because they left it go too long and picked up too many undesirables. What you are making impacts yield because if you are making a clean smoke then your yield will be lower than if you are making oil for edibles or FECO. Now here is where ethanol shines, you can do more than one wash. You can do a single wash to make clean smoke and pull 15%-20% then do a deep-long wash to make some FECO and grab another 10-15%. So the range for that entire spectrum is something like 7%-30%. Hopefully that makes sense?

      1. Rad Info! Thanks

        Your doing a great job for this plant! Keep it up ‘Crafter

      2. Thank you for sharing this information. After having a real challenge removing the chlorophyll, lipids and waxes from an ethanol extraction made by a MB machine, someone was kind enough to share this article with me.

        I’m using this method solely for the purpose of making FECO to use in edibles and suppositories to treat intestinal disease. If my goal is to make FECO, not something to dab or vape, how would you recommend I wash it differently to maximize yield, without picking up excessive chlorophyll, lipids or waxes, which might be causing irritation and inflammation in my intestines?

        Also, do you think a 2.5 micron, 94sec/100mL, grade 5 filter be too slow for this application?

        Thanks again!

      3. Mark, yes the MB pulverizes everything and is not designed for the cannabis plant. I would suggest reading my posts on making RSO and FECO as well as my QWET posts, all of the information is provided there ( I use both med and slow flow which are 12 micron and 3 micron. 2.5 micron is fine if you are doing a proper extraction, but if you are using the MB then you will have problems as the mud will just clog it up.

      4. Apologize for the double post. I didn’t think the first one made it through, so I logged on and tried to retype from memory. I no longer use the MB machine. Hopefully, my second reply was worded better. Thank you

      5. ah bummer, I just saw the “apologize for the double post” so I deleted it without reading, I assumed it was exactly the same. Was there anything important I missed?

      6. I think the terminology was my problem. I was making my own FECO; but, after feeling like the chlorophyll was causing problems, I wanted to avoid the impurities inherent to FECO, which probably means I need a different end product at that point.

        If I used more agitation or a warmer solvent, I could pull more out of the material, but I would gain the impurities with the increased yield. If that’s the case and I wanted a very high quality oil for edibles and suppositories, without chlorophyll, etc., would I follow the dry-ice cold ethanol extraction and just do a 2nd cold ethanol wash and combine it (post filter) with the first wash to maximize yield, as there wouldn’t be any need to separate the washes for my purposes?

      7. For your purposes dry ice is probably overkill. I might do like a 1-hour freezer wash with agitation every 10 minutes and see what you think of that and go from there. Then do a room temp 2nd wash to make oil for cooking or topicals or whatever?

  6. Thanks for sharing, super helpful info. I was wondering if you had tried maxing this out yet as previously mentioned? Soaking for 1 hr, 2 hrs, 3 hrs, etc… while maintaining -40 temps and seeing where the yield ceases to increase.

    Thanks again,

    1. Hi. I have not personally had time to go back to this yet. However, I have heard other followers have successfully left in overnight. Unfortunately, it’s not possible to compare those yields to mine because there are too many variables.

  7. HI there, great article, i’ve been doing the dry ice extraction and evaporating the alcohol with a heating and stirring plate, but every time i get this amber transparent thick oil, never the solid yellow shatter, is this because of the temps i need to evaporate at normal Atm presure? how do you get the ethanol to reduce into that piece in the center of the extractcraft source paper?

  8. Really great article,
    I have been looking at the methods of reducing chlorophyll and strong taste of oral extracts and gummies with full plant benefits. You put clarity into how cold! Freezer will get me started. Need to save my limited resources so I may obtain the turbo extractcrafter! Folks I know growing for themselve home thc and indy hemp are fine with rice cookers evap etoh into the air!
    I am looking to extract industrial hemp as co now allows, and would like to create edible oil for gummies, and dropper bottle with MCT oil, and topical salves. I have never thought of smoking the extract.
    I come from a nutritional herbal point of view and stopped smoking 30 years ago because it was not beneficial for me and thc to dance together, recent thc edibles experiences suggest still not in my best interest.
    So I love cbd rich hemp products, sleep better, a bit of a relaxing effect…
    I really resonate with the freezer EtOH method as I live in central co mountains, and want a good cold extract at 0 F and seems like lower temp ethanol reclamation with Exactcrafter will save terpenes and offer a more palatable cbd edible experience.
    Again, thanks for your diligent article!

  9. Have you tried this QWET freezer or dry ice with kief extracted by the dry ice sieve method? Using the dry ice sieve technique you can produce high quality kief with little green plant from low quality trim.

      1. Do you have any additional pointers for people who might attempt with Leif as I plan to?

  10. Considering myself scientifically minded, this is EXCELLENT! Love your process and the way you iterated. Have always just used/grown flower and was recently introduced to OpenVape on a visit to CO. My goal is to use my excess flower and trim to start making my own ejuice for that pen. What would you suggest at the most efficient way to achieve this? My thought was to make QWET then thin with MCT oil. Would you suggest a better method? And do you have a “recipe” for the QWET method (or if you think another is optimal) with budget in mind? Just hoping to avoid having to buy “theSource”, vacuum oven, etc unless they are highly suggested. All for personal use only, your feedback is greatly appreciated!

  11. Do you do multiple washes or just one? I know most people do at least 2, but they also only wash for 3 minutes or so. Obviously, yours are up to 15 minutes in the freezer and even longer with dry ice.

    1. I personally would only do 2-washes. You might be able to get a little out of a third wash but I’m not that diligent, but if I had more time and it was necessary to really squeeze my budget I would consider it.

      1. Can you direct me to info on how to do multiple washes? Is this process just for cannabis as I’m looking to use for hops? TY

      2. Hops I would most likely wash efficiently the first time and wouldn’t be much value left. I’ve mostly used pellets so the collection was quite fast and easy

      3. Can you please explain “multiple washes” for me. Are you saying new material is being washed in the solution to increase its potency? Sorry, I’m just not clear on this step.

      4. It just means running the same material more than one time to make sure none of the good stuff is left behind. The first wash will only pick up around 70% or so of the cannabinoids available if you are doing a frozen or quick wash, so doing a second wash of the same material to gather what was left behind makes great economic sense. On the other hand, if you are doing an aggressive and long soak you are probably getting everything that’s available so a second wash is probably not necessary.

  12. Another great post. What speed filters are you using with the Buchner filter? I’ve got slow, medium, and fast. Is it worthwhile to run through a couple or all the filters? I’ll be using the Source’s vacumn hookup through the filters. Does the vacumn oven make that much of a difference? I thought I had everything ready to go but hadn’t seen the oven before.

    1. I use slow filters which are about 6 micron, medium flow would probably work, but I always feel happier with the results of the slow filters

      1. Thanks for the info about filter size. I want it to come out as clean as possible, so I don’t mind waiting if it’s worth it.

        Should be interesting. Using various citrus and later in the year, cherries to produce our own bitters, syrups, and other ingredients needed for craft cocktails.

    1. I’m only using paper filters. I should step up my game to some carbon or silica but I just haven’t had the time to get there yet…….

      1. I typically don’t carbon polish. I take most care in the wash process and all I do is for personal consumption and testing. If I were larger scale and producing professionally I might do, but then again I might not unless I was polishing crude to improve distillate clarity output. I think Summit Research has some good input and silica compounds if you are interested in advanced filtering mediums.

  13. Thanks so much for your time and effort to put this report together! Any comments about the moisture level of the plant material? I would imagine that drier is better, but wouldn’t want it at the crumble stage…

    1. It’s my pleasure and I hope it helped. Yes, the dryer the better, assuming you are talking about material that has been properly harvested and stored. I have seen 15% suggest and the minimum moisture content, but I don’t know exactly why that number was accepted to be the imaginary cutoff.

  14. I plan to use this frozen qwet technique real soon. I hope to use the ethanol to wash multiple batches before boiling off the solvent. Have you tried that? Do you have an idea of the solubility limit of thx in ethanol? Will save me some evaporation time.

    1. My oil is always clean because I only use food grade ethanol so I’m not worried about some residuals because it holds no adverse health effects for me in those minute amounts. I use full vac around 90F most the time. Lately, I have also been messing with 112F at full vac with the pump running for 24hrs which has been putting out some good results. I also sometimes work at 85F and leave under vac for a week or more. There are many ways to skin that cat, you just have to find the one you agree with the most.

  15. What was your typical yield variance between using dry flower and wet frozen? In regards to your yield from fresh frozen vs quality of end product, is it possible for a processor during tests to determine extraction yield upon which a wholesale producer is paid, to during the processing manipulate the yield downwards? If so could such low yield also result in actually increasing the final value of their end product?