No chemical solvent goes into the extraction. Ice water separates the trichome heads, then heat and pressure release what's inside them. Fewer steps, gentler handling, and an extract that keeps more of what the plant made.
What are the methods that count as solventless?
Three, and they share a principle: physical force does the separating rather than chemistry.
Method - | Ice water hash | Dry sift | Rosin pressing
How it works - | Chilled trichome heads snap off and sink through progressively finer screens | Dried material is agitated over screens so trichome heads fall through | Hash or flower goes between heated plates and pressure releases the resin
What it produces - | Wet hash, graded by wash and screen size | Loose kief, sorted by grade | Rosin, the pressed oil that fills a cartridge
Speaking from the press room: rosin pressing is where most buyers meet the category, because it's the form that fits a cartridge. The other two are usually a stage on the way there rather than a destination. Dry sift deserves a word, because it's the oldest of the three and the one you'll see least. It skips water entirely: dried material is agitated over screens until the trichome heads fall through, and what collects underneath is kief. It's simple and it keeps well, but it's harder to get clean, since anything else brittle enough to break off comes through the same screen. That's why the modern premium tier runs on water rather than sift.
How ice water hash works
Cannabis is a plant covered in trichomes, the resin heads that hold nearly all the cannabinoids and aromatic compounds. Ice water hash exploits a simple physical fact: chilled trichome heads become brittle and snap off, and they sink while plant matter floats.
Fresh-frozen material goes into cold water and is agitated gently. The trichome heads break away, fall through a stack of progressively finer screens, and get collected wet before being dried. Nothing is dissolved. The water is a medium for moving material around, and it never becomes part of the product.
Grading happens at the screens. The earliest, cleanest fraction is the most prized, and material that melts completely under heat gets called full melt. That grade is what our Live Rosin line is pressed from, and it's the reason the flavor carries the way it does. Grades get talked about in stars, and the scale isn't certified by anyone. Six star, or full melt, describes material clean enough to melt away completely on a hot surface with no residue. Lower grades carry more plant matter and behave differently under heat. It's a useful shorthand between people who work with the stuff, and it's a marketing term the moment it reaches a package, so treat the number as a claim rather than a grade.
How rosin pressing works
Rosin is the press step. Ice water hash goes between heated plates, pressure is applied, and the resin flows out as an amber oil. It's closer to an olive press than a laboratory.
Two variables do the work: temperature and time. Lower and slower preserves more of the aromatic fraction and yields less. Hotter and faster yields more and costs you flavor. A producer's choice between those two tells you what they're optimizing for, and it's the single most useful thing to ask about a rosin product.
We press once. A first press pulls the cleanest material, and running the same hash again returns more volume and less character. That decision is why our yields are lower and our price sits where it does. There's a practical reason the format works in a cartridge at all. Rosin is thick and it behaves differently from a refined oil in hardware, which is why the pairing between the material and the ceramic matters as much as the pressing does. A press that produces beautiful rosin and hardware that can't wick it gives you a frustrating product. Temperature at the coil matters for the same reason. Rosin carries more of the volatile aromatic fraction than a refined oil does, and those compounds are the first thing heat destroys. Run it hot and you'll taste the loss immediately, which is why the format rewards a lower setting and a slower draw.
Why water counts as a solvent
Because chemically it is one, and this is the piece of vocabulary worth getting right.
A solvent is any substance that dissolves another. Water dissolves a great many things, which makes it the most common solvent on earth. So a strict reading of solvent-free would exclude ice water hash, and that reading is technically correct and practically useless.
The distinction the industry actually means is chemical solvent: butane, propane, ethanol, carbon dioxide under pressure. Those dissolve the resin itself and then have to be removed, and the removal is regulated as a manufacturing step, then verified on the residual solvents panel of a lab report. Water never dissolves the resin. It moves brittle trichome heads from one place to another and then leaves.
So solventless is the accurate term and solvent-free is the loose one. When a package says solvent-free, it's usually reaching for the same idea with worse precision. There's one more wrinkle worth knowing, since it comes up whenever someone quotes the dictionary at you. Carbon dioxide extraction is sometimes marketed as clean or natural, and it isn't solventless. Under pressure, carbon dioxide behaves as a solvent and dissolves the resin, which is precisely the thing ice water never does. The test isn't whether a substance sounds benign. It's whether the resin was dissolved. The same test settles the other borderline cases people raise. Pressing flower directly, without washing it into hash first, is still solventless, because heat and pressure did all the work. Adding a food-grade oil to thin a finished rosin happens after extraction rather than during it, which is why it is generally treated as leaving the solventless description intact. It still changes what you're inhaling and belongs on the label. Method claims describe how the resin came out of the plant. They say nothing about what happened to it afterward, and that gap is where most misleading packaging lives.
Does solventless mean zero solvent on the report?
No, and a brand that promises that is promising something the paperwork won't support.
Every batch sold in California clears a residual solvents panel regardless of how it was made, and solventless products earn no exemption. Trace amounts can and do appear on those panels, because hardware, cleaning and handling all contribute their own residue long after extraction is finished. Every result still has to sit under the state's action level to be sold.
This matters for how you read a claim. The defensible statement is that no chemical solvent is used to extract it. A blanket zero is a different claim, and the document behind it usually says otherwise. We'd rather tell you where the line sits than let a certificate of analysis surprise you later. None of that undermines the category. It's a statement about how sensitive modern instruments are and how many surfaces a product touches between the press and the shelf. A panel that reports a trace at a small fraction of the action level is telling you the system works, not that something went wrong.
How do you check a solventless claim?
Three questions, in order of how much they tell you.
Ask which method was used, because a producer running ice water hash and a single press can say so in one sentence. Ask what the press temperature was, since the answer separates people optimizing for flavor from people optimizing for yield. Then ask for the batch document and read the residual solvents panel, which is where the claim meets an outside referee.
A label alone settles none of this. Solventless carries no certification and nobody audits the word, so the method description and the document are what give it weight. If you only have time for one of the three, make it the last one. Method claims are easy to make and hard to check from the outside, and a batch document is the one artifact in this whole process that somebody else signed. Reading it takes less skill than people expect. Find the residual solvents panel, check that every line reports below its action level, and match the batch code on the document to the code on your package. That's the whole exercise. You don't need to interpret the numbers, only confirm that a licensed lab looked and that the document in front of you describes the thing in your hand.
