Glass, Metal, Ceramic, Wood: Does Your Inhalation Device Material Change What You Inhale?

The device material that has been measured in published research is not the bowl or the mouthpiece. It is the heating element, the connector pin, and the other metal components inside electronic cartridges. A 2025 study from the National Research Council of Canada and Health Canada, published in Scientific Reports, analysed six legal Canadian cannabis vape liquid products and found cobalt, nickel, and zinc from device components driving variability within a single batch.

Single-particle mass spectrometry detected metal particles of aluminium, cobalt, chromium, copper, nickel, tin, and zinc in the liquids. When the cartridges were vaped on a machine, the resulting aerosols contained particles of cobalt, chromium, nickel, lead, tin, and zinc. Scanning electron microscopy of an unused device showed cracking on the connector pin, a plausible route for contamination before the product is ever opened.

Patients choosing between a glass piece and a metal one are asking a reasonable question, and answering it with folklore does them no favours. The evidence supports a different framing: the parts of the system with published contamination data are the metal contacts and heating elements of electronic cartridges, and the variables shown to drive leaching are operating temperature, device age, structural integrity, liquid pH, and cutting agents.

That reframing is actionable. It points toward questions a patient can actually ask about a product, such as whether it has been tested for metals in the aerosol rather than only in the oil, and it points away from choices that have no data behind them at all.

The familiar claim runs like this: glass is chemically inert and preserves terpenes, metal reacts with volatile compounds and dulls flavour, wood is porous and absorbs cannabinoids. Each part of that sounds plausible from general chemistry. None of it has been tested in a published, peer-reviewed comparison of cannabis inhalation devices.

There is no indexed study comparing cannabinoid or terpene delivery from a glass pipe against a metal pipe against a wooden one. There is no published head-to-head comparison of ceramic against metal heating elements in cannabis vaporizers on either cannabinoid yield or emissions. When a page states that glass gives you a clearer effect and higher bioavailability, it is stating a preference, not a finding.

That absence is worth saying out loud rather than filling with confident-sounding text. It is also not the end of the story, because the device material question has been studied seriously, just from a different angle.

A 2025 scoping review by Gaur and Agnihotri in The Scientific World Journal followed PRISMA extension methods and identified nine studies published through February 2024 on metal release from cannabis vapes. Across those studies, the structural elements of the devices leached nickel, chromium, lead, cobalt, cadmium, and copper, which were carried into the aerosol as small particles.

The review identified which variables drive that leaching: the structural integrity of device components, device age, operating temperature, the pH of the vape liquid, and the viscosity cutting agents present in it. The authors noted that none of the included studies evaluated the direct health impact of these metals in users, which is an important limit on how far the finding can be pushed.

Mechanistic work in nicotine devices fills in the picture. A 2022 study in the International Journal of Environmental Research and Public Health by Rastian and colleagues measured chromium, nickel, copper, and lead during simulated use of a mod-type device and found metal concentrations rising with both puff count and coil power. Chromium and nickel were higher in the aerosol than in the tank liquid, which the authors read as direct transfer from the heated nichrome coil, while copper and lead accumulated in the liquid first.

This literature would be easy to overstate. A 2021 study from the Centers for Disease Control and Prevention, published in Toxics by Gonzalez-Jimenez and colleagues, analysed aerosols from 50 vaping products for ten metals. Only five of the 50 devices produced aerosols with detectable metal concentrations. Cannabinoid-based aerosols were below reportable levels with a single exception, a sample at 16.08 nanograms of copper per 10 puffs.

Part of that is chemistry rather than safety. Cannabis concentrates are hydrophobic, and the aqueous trapping methods developed for nicotine e-liquids do not capture nonpolar aerosols well. A 2021 paper in ACS Omega by Mallampati and colleagues addressed exactly that problem and showed that not all metals present in a concentrate are recovered in the aerosol under standard voltage settings, and that a nonpolar solvent in the collection step improves recovery for some metals.

The reasonable reading of these papers together is that metals from device hardware do reach the aerosol, that the amounts vary widely across products, and that the measurement methods are still being standardized. That is a genuine safety signal with real uncertainty attached, not a settled exposure estimate.

The variable with the strongest published effect on what comes out of a cannabis device is heat. In a 2009 study in Inhalation Toxicology, Pomahacova, Van der Kooy, and Verpoorte compared vapor from a Volcano device at several temperatures against cannabis cigarette smoke. The ratio of cannabinoids to byproducts was significantly higher in vapor produced at 200 and 230 degrees Celsius than in smoke, and the worst ratio in the entire experiment came from vapor at 170 degrees Celsius.

Earlier work by Hazekamp and colleagues in the Journal of Pharmaceutical Sciences in 2006 characterized the same device for clinical use and found that about 54 percent of loaded THC reached the balloon reproducibly, while roughly 35 percent of inhaled THC was directly exhaled again. Those numbers describe the delivery efficiency of a specific device operated a specific way, which is the kind of detail that actually determines a dose.

Temperature also governs degradation chemistry. A 2019 ACS Omega study by Meehan-Atrash and colleagues found that vaporizing and dabbing cannabis extracts generated methacrolein, benzene, and methyl vinyl ketone, and that adding terpenes produced higher levels of gas-phase products than THC alone. The same work found gas-phase degradant levels substantially lower in vaping and dabbing than in cannabis smoking.

The most severe documented harm from cannabis vaping was not caused by device material. The 2019 outbreak of e-cigarette or vaping product use-associated lung injury, which the CDC linked to 2,807 hospitalizations and 68 deaths, has been strongly associated with vitamin E acetate used as a diluent in illicit THC products.

A 2024 study in Chemical Research in Toxicology by Wang and colleagues at the California Department of Public Health measured the mechanism in real devices. Ketene, a highly toxic gas, formed when vitamin E acetate was vaped, increased with repeat puffs, and tracked with measured internal device temperatures between 200 and 500 degrees Celsius. Battery power strength, which sets the heating temperature, played an important role. Ketene did not form when vitamin E without the acetate group was vaped under the same conditions.

This matters for the material question because it shows where the risk hierarchy sits. What is in the liquid, and how hot the device runs, have produced measurable clinical harm. What the mouthpiece is made of has not been shown to produce anything.

For the proposition that metal components of electronic cannabis cartridges contaminate the liquid and the aerosol, the evidence is moderate and converging. Independent laboratories using different analytical approaches, including a national metrology institute, have detected the same metals and traced them to cartridge hardware by elemental matching.

For the proposition that glass, wood, or ceramic bowls and mouthpieces meaningfully change vapor quality, cannabinoid delivery, or effect, there is no direct evidence. The claim appears widely in consumer writing and does not appear in the indexed research literature.

For the proposition that metal exposure from cannabis vaping causes specific clinical harm in users, the evidence is absent. The 2025 scoping review states plainly that none of the included studies evaluated direct health impact, and inference to organ-level harm comes from general heavy metal toxicology rather than from studies of vape users.

The anchor study analysed six products from one regulated market at one point in time. Cartridge hardware is sourced globally and changes constantly, so these results characterize a sample rather than a market.

Emitted particle counts in the aerosol fell below the limit of quantitation, which means the study establishes presence more firmly than amount. Presence without a quantified dose cannot support an exposure or risk estimate.

The CDC analysis pulls in the opposite direction, with only five of 50 devices producing detectable metals in aerosol and cannabinoid products generally below reportable levels. Reconciling these results requires attention to trapping method, device type, and analytical sensitivity, and the field has not settled on a standard.

None of this work shows that a metal pipe delivers a weaker effect than a glass one, that wood absorbs a meaningful fraction of cannabinoids, or that ceramic heating elements produce cleaner vapor than metal ones. Those comparisons have not been published.

None of it shows that cannabis vaping is safe, and none of it shows that any particular product is contaminated. Batch and product variability was one of the loudest findings in the anchor study.

It also does not establish a threshold. Without quantified aerosol concentrations and inhaled dose estimates, there is no basis for saying how much metal exposure a given pattern of use produces.

Cannabis product safety conversations have moved through pesticides, residual solvents, and microbial contamination, and hardware is the current frontier. Unlike the plant-side contaminants, hardware metals are not addressed by testing the flower or the oil before it is filled, because the contamination happens in the cartridge and during use.

That is a regulatory gap rather than a scientific mystery. Testing regimes in most jurisdictions require heavy metal testing of the input material, not of the assembled device or of the aerosol it produces. The anchor study’s finding that an unused connector pin was already cracked is the kind of observation that argues for device-level standards.

For clinicians, this connects to the older and better-established conversation about route of administration. Inhalation has advantages for onset and titration that oral routes do not, and those advantages are real. They come with an aerosol the patient’s lungs have to handle, and hardware is part of what is in that aerosol.

When a patient asks me whether glass is better, I tell them I do not have data for that, and then I tell them what I do have data for. The metal inside a disposable cartridge is measurable in the oil and in the aerosol. That is worth knowing, and it is not what most people are asking about when they ask about materials.

In practice my advice has less to do with the material and more to do with the source. Buy from a regulated market with published testing. Keep the temperature low enough to avoid combustion and high enough to actually vaporize, which the older Volcano work puts in the range around 200 degrees Celsius rather than 170. Be sceptical of anything with an unfamiliar diluent in it, because the worst outcome anyone has documented in this space came from an additive, not from hardware.

I would also say that a reusable device you can inspect and clean is easier to reason about than a sealed disposable you cannot. That is a judgement about transparency rather than a claim about chemistry, and I try to label it as such.

There is no published evidence that glass, wood, or ceramic bowls change vapor quality or cannabinoid delivery. There is converging evidence that metal components inside electronic cartridges shed nickel, chromium, lead, cobalt, and other metals into the liquid and the aerosol, with temperature, device age, and liquid composition driving how much. Focus on product source, testing transparency, temperature control, and what is in the liquid, rather than on the material of the piece.

The useful distinction here is between the part of a device that touches vapor and the part that heats it. Published contamination work concerns the heated and current-carrying metal components of electronic cartridges. Claims about inert bowls and porous wood are extrapolations from general material properties, presented as if they were measurements.