Marijuana Mining: Solventless vs. Solvent-Based Extraction

killykoon-Stock.adobe.com

When any state adopts new recreational cannabis laws, it often feels like a gold rush. As new legal markets open up, new processors always seem to initially favor solventless methods of making concentrates like rosin, hashish, and kief, like panners drawn en masse to gold-bearing streams.

The reasons are plain to see for the uninitiated. All three have very low equipment costs, require little or no training, and avoid the need for expensive “blast-proof” rooms. Like panning for gold, it’s so simple anyone can do it. There’s also a perception that solventless methods are “cleaner” than solvent-based extractions. Given their popularity and impending rescheduling, it’s worth taking a deeper dive into the benefits and drawbacks of all these solventless cannabis extraction systems from the more commercial perspective of experienced lab operators in mature recreational cannabis markets.

Let’s start with the fact that cannabis concentrates are the oldest known manufactured drug on earth. Hash-making predates the fermenting of wine by over 1,000 years, and the ancient process was simple. Put cannabis in a burlap bag and beat the bag with a broom until the trichome heads fall out. This time-honored process is still employed today in places like Morocco and Lebanon, with the first famous for strains that make rich, dark-colored hash, and the latter known for strains that produce a blonder hash, both with THC potencies around 50%.

Courtesy of SciPhy Systems

Water-hash methods are commonly used commercially since it’s more efficient to run larger batches quickly. Water processes involve using different mesh sizes of bags to filter out unwanted or lower-grade particles. The end result is 3-5 different grades of concentrate, with the final filter producing the highest grade referred to as a “full-melt” hash that can be over 60% pure THC. Full melt is highly prized since yields are small. Lower-grade hash from the upper mesh bags is typically used in edibles, and any water-soluble flavors and aromas are lost to the water. This method of classification based on size exclusion is not very different from mining “placer” gold in a stream.

Industrial chemists classify the first method of beating the burlap bag as agitation, and the water method as washing. Even though chemists refer to water as the “universal solvent” neither method can be called a true extraction since the separation is purely mechanical. The cannabinoids and terpenes contained within the trichome aren’t dissolved and separated from the waxy shell. Any good oil on the surface of the bud gets left behind too. Herein lies the first significant drawback to solventless systems; you only get about half the cannabinoids and terpenes out, much like the gold miners of yesteryear, who only pulled half the gold out of the rock they worked so hard to mine.

Rosin is typically made with a heated press, and kief is made by shaking or beating the cannabis to “dry sift”, just like the old Lebanese and Moroccan hashers have for eons, although these days processors typically use dry ice tumblers to improve yields. Rosin makers looking for a blonder product will combine these methods, first kiefing the biomass and pressing kief to make rosin. This process retains more flavorful terpenes than water hash since you're not washing away water-soluble flavonoids, and potency tends to be around 60%, Rosin and Kief also avoid significant mold issues with water hash methods. However, yields are meager by comparison. Often below 40%, meaning the process wastes up to 60% of the precious cannabinoid oil and terps. This defines the biggest problems with these solventless methods. High labor and production costs, long timelines, and net yields lower than 50% of the total cannabinoids and terpenes in the plant.

The analogy of gold mining is extremely relevant to cannabis extraction. When industrial companies mine gold, they don’t waste precious time and money sifting through tons of river gravel for a few ounces of placer gold anymore. (1) It’s simply not commercially viable to do it this way these days, even though it's much easier to get river gold out. Plus, most reachable placer deposits are depleted worldwide. Instead, large-scale gold mining companies blast out large amounts of rock from the earth referred to as ore, then they use chemical (solvents) processes in huge foundries (labs) to extract minerals from the rock. Ore veins tend to contain relatively small amounts of gold and other trace minerals next to larger amounts of copper. It’s not uncommon for a rich ore vein to have 0.4% gold and 5% copper, but that smaller percentage of gold is worth 800 times the copper’s value! Even though there’s over 12 times more copper than gold in the ore. Both have value of course, and neither is discarded if a mine is to remain profitable.

Marijuana extractors are quite literally in the “marijuana mining” industry. Our “ore” is the whole-plant biomass. Our “copper” consists of cannabinoids like THC and CBD, typically making up 60% or less of the crude oil once extracted and dewaxed. The “gold” in cannabis is the terpenes since they are worth 15 times more kilo-for-kilo than cannabinoids, but they make up 10% or less of the crude. (2) If you discovered your mining process was only recovering half the copper, and worse yet, destroying most or all the valuable gold, one would think you’d change your SOPs to fix your mining process quickly, or you wouldn't be in business very long. This is the challenge all extractors face once markets mature, and prices compress across the board.

The bottom line, literally, is rosin, kief, and hash only get about half the copper (THC) out. Water strips and the heat of rosin presses cook off many of the (gold) terpenes, which boil off at around 70 degrees F. In contrast, while ethanol and CO₂-based extractions are highly efficient at pulling out over 90% of the (THC) copper, they tend to obliterate the (gold) terpenes in the process, even though it’s worth over 15 times what the copper is worth! It should be noted that experienced CO₂ operators can do a sub-critical terp stripping before extraction, but the efficiency is low, flavors can be degraded, and most don’t do this extra step. The painful irony of labs using ethanol and CO₂ is they are often forced to buy terpenes (gold) from outside suppliers to make flavorful end products like vape carts.

Commercial cannabis processors, much like commercial gold miners, might start with panning, but only to test and find rich ore veins. To be commercially viable and extract all the valuable elements from ore bodies, they all turn to chemical solvents, and the hard truth is some chemicals are much more efficient than others, or as chemists say, selective. Scale is also key: larger equals lower costs. With ethanol and most CO₂ processes destroying the (gold) terpenes, when market prices drop, and they do, the problem is clear. Only one solvent class can run fresh, terp-rich material and preserve those terpenes intact while selectively pulling out cannabinoids: light hydrocarbons like butane. Moreover, the THC potency and yields of hydrocarbon diamonds can approach 100% pure.

So why don't new labs start where experienced operators end up? There are two main reasons, and cost tops the list. Where someone can squeeze out a few grams of rosin with a hair curler in their living room or hash in a bathtub, safely using hydrocarbons at scale requires commercial-grade extractors housed in permitted Class-1/Div-1 blast rooms. Just like a commercial gold mine, this equipment is not cheap and permits do not come easy. But like those large-scale gold mines, they capture everything more efficiently and are infinitely more profitable than panning in a stream.

Perceived safety is often the other reason big reason people go solventless. In terms of product safety, the reality is hash can be a mold-making machine without significant additional and costly steps. Even Kief and Rosin are unpurified, which will likely not meet the upcoming GMP requirements once federal rescheduling takes place next year. (3) Only solvent-based extractions meet GMP-purity standards by filtering concentrates to 0.3 microns; fine enough filtration to eliminate mycotoxins. In terms of solvent safety, it’s worth noting that iso-butane has been studied as a propellant in metered-dose inhalers to deliver albuterol and steroidal medications into asthma patients’ compromised lungs, with results demonstrating its safety as an alternative to chlorofluorocarbons. (4, 5, 6)

To put it simply, while solventless extracts will always be a great choice for boutique extractors or hobbyists making a few grams of personal stash, they will never be able to compete with commercial-scale extractors pumping out 10 liters (10,000 grams) an hour or more. With the anticipated industry consolidation U.S. federal rescheduling will likely bring very soon, the only labs that stand to survive will be those who "mine" the full value of cannabis to GMP standards at scale.

It is worth adding that it’s not all about doubling yields or tripling scale in this business. In taste tests, most daily dabbers admit (BHO) Live-Resin mined from fresh bio has superior flavor too!

References

1. How Gold is Mined: The Lifecycle of a Gold Mine. World Gold Council. https://www.gold.org/gold-supply/gold-mining-lifecycle (accessed 2025-03-14).
2. Fuhr, S. Terpenes: Capturing Cannabis’ Highest Value Fraction. Marijuana Venture Magazine. October 13, 2023. https://www.marijuanaventure.com/terpenes-capturing-cannabis-highest-value-fraction/ (accessed 2025-03-14).
3. Facts About the Current Good Manufacturing Practice (CGMP). U.S. Food and Drug Administration. January 21, 2025. https://www.fda.gov/drugs/pharmaceutical-quality-resources/facts-about-current-good-manufacturing-practice-cgmp (accessed 2025-03-14).
4. Ding, L.; Zhang, J.S. Isobutane driven salbutamol sulfate metered dose inhaler: formulation selection and respiratory tract absorption in guinea pigs. Yao Xue Xue Bao. 2009, 44 (9), 1040-5. PMID: 20055182
5. Myrdal, P.B.; Sheth, P.; Stein, S.W. Advances in Metered Dose Inhaler Technology: Formulation Development. AAPS PharmSciTech. 2014, 15 (2), 434-455. DOI: 10.1208/s12249-013-0063-x
6. Final Report of the Safety Assessment of Isobutane, Isopentane, n-Butane, and Propane. Journal of the American College of Toxicology. 1982, 1 (4). DOI: 10.3109/10915818209021266