Last updated on August 19, 2026 · Originally published February 18, 2017
In 2017 this article asked what the water in bubble hash actually does and admitted it could not find a satisfactory answer. Nine years later, the physics is clearer. The terpene comparison is not.
The difference between dry sift and bubble hash is the addition of water. The late Frenchy Cannoli regarded water as the only major development in sieving since the technique was invented.
That sounds like a small change. It is not — but what the water does turns out to be different from what most people assume, and the comparison everyone draws between the two methods is confounded in a way that almost no one mentions.
What actually separates dry sift from bubble hash?
Both methods aim at the same thing: removing intact trichome heads from plant material without solvents.
Dry sift uses agitation and screens to detach and grade trichome heads. Plant fragments of similar dimensions can pass through with them, which is why high-grade sift requires careful refinement.
Bubble hash uses cold water as its working medium. Low temperature makes glandular trichomes easier to detach; agitation breaks them free; and the water carries the detached material away from the bulk plant tissue. Larger plant material floats or remains in the work bag while detached heads tend to settle, but the collected fractions are ultimately graded by mesh size.
Water therefore adds temperature control, transport and settling behavior to the same basic act of sieving. How much each contributes to purity has not been measured.
What is the water actually doing?
Water has several clear physical roles in a bubble hash wash and one plausible chemical effect.
The first role is as an agitation medium. Moving water carries the mechanical energy that breaks trichome heads from their stalks and transports them away from the plant material.
The second is temperature control. Cold keeps the glands brittle enough to snap cleanly rather than smear, and slows evaporation throughout.
Water may also have a chemical effect, and this is the one the original article guessed at. Terpene hydrocarbons — myrcene, limonene, the pinenes — are effectively insoluble in water. Oxygenated terpenoids such as linalool and terpineol are meaningfully more soluble. In principle, water could therefore remove oxygenated terpenoids preferentially if the resin is exposed, shifting the aroma rather than simply reducing it. But solubility falls in cold water, and no published bubble-hash experiment has demonstrated that this occurs to a meaningful degree.
Drake reached that hypothesis in 2017 while explicitly disclaiming his own chemistry. It was plausible then and remains plausible now. The effect is simply unmeasured.
Does washing strip terpenes?
Probably, but less is established than the confidence around it suggests.
Ruptured heads can release resin that remains on the water, vessel or equipment rather than entering the collected fraction. But a visible surface sheen is not a measurement of terpene loss — it may contain waxes, lipids and plant material as well as resin — and the composition of wash water has not been adequately studied.
Wet hash must then be dried, introducing another stage at which its volatile profile can change. Air-drying takes longer and exposes the material to air; freeze-drying is faster and removes ice by sublimation under vacuum. Neither process is chemically neutral.
Freeze-drying is commonly assumed to preserve more terpenes, and vendors routinely claim as much. The limited cannabis research is not that simple. One 2024 study of flower found 45 compounds at lower abundance after freeze-drying than after tray drying, 28 of them terpenes, and proposed that ice-crystal damage exposed the volatiles and allowed them to escape under vacuum. Hash has not been tested the same way, so we do not know whether that result carries across — but it rules out treating freeze-drying as automatically superior. The wider problem of where volatiles go after harvest is covered in what is lost after harvest.
Why do sources disagree about which preserves more terpenes?
Sources disagree because the comparison is confounded, and almost nobody says so.
Search the question and the results split neatly in half. One group holds that dry sift preserves more, because water never touches the resin. The other holds that bubble hash preserves more, because it retains volatile compounds that dry sift loses. Both cite the same physical principles. Both sell equipment for the method they recommend.
The reason they can both marshal evidence is that they are not comparing the same thing. Conventional dry sift generally requires dried material. Premium “live” bubble hash is commonly made from fresh-frozen material, harvested and frozen within hours and never dried at all.
Drying and curing together represent one of the largest terpene-loss stages in the entire chain. So a comparison of sift against bubble hash is substantially a comparison of dried input against fresh-frozen input, plus the drying step that follows a wash — with the water riding along and collecting the credit or the blame.
To isolate the effect of water, researchers could split the same dried batch between dry sifting and ice-water separation, then measure each material before separation, after separation and after any required drying. A separate experiment could compare bubble hash made from matched fresh-frozen and dried material. As far as we can establish, neither comparison has been published.
What should you actually take from the comparison?
Three variables travel with the method and are difficult to separate: the condition of the starting material, the way the resulting hash is dried, and the operator’s technique. Over-agitation ruptures heads in a wash; insufficient refinement leaves plant matter in sift. Neither outcome is inherent to the method; both depend heavily on the operator.
Water clearly controls temperature, transmits agitation and carries detached trichomes through the process. Settling may assist separation, and differential solubility may alter some exposed volatiles. What remains unknown is how large any of those effects are compared with drying, curing, input material and technique.
So the answer to the 2017 question is not quite “here is what the water does.” It is more interesting than that. Water has several clear physical roles and one plausible chemical role, and it receives credit or blame for differences that are confounded by everything surrounding it.
Nine years on, there is still no controlled comparison of the two methods on matched material. Most of the readily available yield and equipment figures come from vendors rather than controlled studies, and vary widely enough to be worth treating with caution. That is an unsatisfying answer, but it is the accurate one.
For how solventless processing fits together as a category, see our guide to what “solventless” actually means, and for what happens when this resin is pressed, BHO vs. rosin covers the chemistry that separates the routes.
References
- Spadafora ND, Felletti S, Chenet T, et al. The influence of drying and storage conditions on the volatilome and cannabinoid content of Cannabis sativa L. inflorescences. Analytical and Bioanalytical Chemistry. 2024. doi:10.1007/s00216-024-05321-w
- Martins MAR, Silva LP, Ferreira O, Schröder B, Coutinho JAP, Pinho SP. Terpenes solubility in water and their environmental distribution. Journal of Molecular Liquids. 2017;241:996-1002. doi:10.1016/j.molliq.2017.06.099
- Pourseyed Lazarjani M, Young O, Kebede L, Seyfoddin A. Processing and extraction methods of medicinal cannabis: a narrative review. Journal of Cannabis Research. 2021;3:32. doi:10.1186/s42238-021-00087-9
- Wanas AS, Radwan MM, Chandra S, et al. Chemical composition of volatile oils of fresh and air-dried buds of cannabis chemovars, their insecticidal and repellent activities. Natural Product Communications. 2020;15(5). doi:10.1177/1934578X20926729
Originally published February 18, 2017. Updated August 18, 2026 to address the question the original article raised, describe what water contributes to separation, and identify the confound in published comparisons of the two methods.
