Robert Thomas, Scientific Solutions

The recent announcement by the state of Colorado proposed that all marijuana concentrates in electronic cannabais delivery systems (ECDS) must be tested for heavy metal contaminants in the emitted aerosol

(1). This has been long overdue, since it was first reported in 2017 that vaping liquids were showing high levels of metals in the vaped aerosols

(2). It caused some concern, but the problem wasn’t fully investigated, except that the metals were likely coming from the internal components. However the dangers of vaping resurfaced again in 2019 when e-cigarettes and vaping pens were the source of 67 deaths as a result of severe respiratory failure from vaping devices containing vitamin E acetate, one of the diluents used in these systems

(3). The problem eventually got resolved by removing this compound from the ingredients in all legally produced vaping products. Colorado has been very proactive in this field and as a result became the first state to require testing cannabinoid vaping aerosols for a panel of metals contaminants. This announcement has stimulated a great deal of interest from other state regulators to see how this eventually gets resolved, so let’s take a closer look at the challenges in implementing these regulations.

Fundamentals of Vaping

When cannabis (extract or flower) is heated to 200-300°C the psychoactive and therapeutic cannabinoid components volatilize into tiny airborne droplets that form an aerosol called a vapor, which is inhaled by the user. The process is carried out using a battery-powered element to heat the cannabis product in a tank that is often mixed with diluent liquids like vegetable glycerin to decrease its viscosity and sometimes flavorings to increase its appeal to younger users. In single-use cartridges, the liquid is often soaked into fibers and a heating element is wound around a wick. In pods, the liquid is usually in direct contact with the heating element, whereasin refillable systems, users add their desired liquid into a tank. In some devices, the temperature can be selected using variable voltage control. As a result, it is extremely difficult to predict what metals will find their way into the aerosol. A schematic of two typical vaping pen are shown in Figure 1.


Figure 1: Basic components of two common cannabis vaping pen designs (www.

There have been a number of studies in the public domain describing the analysis of e-liquids (4). However very little has appeared on charactering aerosols, mainly because of all the different materialsused inside commercially-available systems. For example, some of the common components being used for tanks include plastic, glass (Si, Na, B) or stainless steel (Fe, Cr, Ni, Co, Mn), while coils and atomizers come in a variety of materials including nichrome (Ni, Cr), kanthal (Fe, Cr, Al), sometimes with brass electrical connectors (Cu, Zn, Pb) or other materials. Some of the newer atomizers are made from ceramic materials (Si, Al, Ca).

Analyzing Vaping Liquids for Heavy Metals

It’s important to emphasize that the Colorado directives specifically stipulates characterizing the vaping aerosol for heavy metals, not just the liquid. Sampling an aerosol without contaminating the sample presents an order of magnitude greater complexity, because the aerosolization mechanism and transport of metallic particles at vaping temperatures around 200-300° C are very different to sampling metal ions in solution. Moreover, once the aerosol has been generated, there has to be a way to trap and collect the aerosol without additional contamination in order to measure the metal conten by inductively coupled plasma mass spectrometry (ICP-MS). In comparison, sampling a vaping liquid is fairly straight forward, as it can just be sampled directly from the refillable pod or vaping device tank. The diluents used in the device will dictate what types of solvents to use for dilution. With electronic nicotine delivery systems (ENDS), the diluents are typically propylene glycol or vegetable glycerin, which are water soluble (hydrophillic).This means that they can be extracted and dissolved using a mixture of mineral acids and as long as the calibration standards are made up in a similar matrix, they can be presented to the ICP-MS for analysis (5). However vaping devices used for cannabinoids are a little more complex to sample because the diluents might also include medium chain triglyceride (MCT) oil which is not water soluble (hydrophobic), in addition to the cannabinoid extract itself which is oily and very viscous in nature. This means the liquid either has to be extracted and dissolved in an organic solvent, which will present problems for the ICP-MS instrument unless an optimized sample introduction system is used. The alternative approach is to use a microwave digestion system to redissolve the residue in an acid mixture, before being presented to the instrument for analysis.

Characterizing Vaping Aerosols for Heavy Metals

First it’s important to emphasize that the transport of the metal particles in a vaping aerosol is not fully understood, but will be different based on the metal compositions of vaping device components and their heating characteristics. The temperatures at which ENDS/ECDS liquids are heated (200-300 °C) are insufficient for vaporization of most metals and metal oxides; therefore the mechanism of transport in the aerosol is likely to be independent of metal volatility (6). Moreover, limited entrainment of metal particles, or dissolved metal compounds in aerosol droplets formed by passage of air over the liquid surface will have an impact on the metal transport mechanism. As a result, the principal forms of the metals would not vaporize at the relatively low heating element temperatures, but rather be inefficiently entrained and transported in the aerosol (6, 7). For this reason, spiking trace elements into the liquid and measuring recoveries in the collected aerosol as a way of validating the method will unlikely be a useful exercise.

The standard method used to trap and collect the vapor from ENDS or ECDS is with an aerosol generation machine, which is an adaption of traditional smoking machines used to measure contaminants in tobacco products like cigarettes and cigars. These aerosol generation testing machines work extremely well for measuring organic analytes but when it comes to evaluating inorganic pollutants they have limitations, because many of the components such as tubing, filters and impingers are not made out of high purity materials. So researchers who are interested in looking at very low levels of trace metals in e-cigarette aerosols typically have to customize the aerosol generation machines to make them better-suited for looking at trace metals.

Vaping Aerosol Collection

If vaping aerosol analyses are to provide robust data, the methodologies and instrumentation must be scientifically valid. While no vaping machine puffing regimen mimics a human vaper, it is imperative that the data collected follows a standard and reproducible procedure. There are two standard methods available. ISO Standard – Method, 20768:2018 was specifically developed for the routine testing of cannabinoid vapor products with an analytical vaping machine (8). And CORESTA Method 81 (9), which defines the requirements for the generation and collection of e-cigarette aerosol for analytical testing purposes. They are both similar in functionality, but the CORESTA method has been more widely used because of its application with tobacco and nicotine vaping products. Both methods have similar specifications, including prescribed limits for pressure drops, puff volume, puff time, and puff profile for generation and collection of reproducible aerosol results.

Note: It is important to emphasize that the CORESTA 81 Method was developed for tobacco and nicotine products, so the puff specifications reflect a typical cigarette smoker or END consumer. We know that cannabis consumers use ECDS very differently with regard to duration, volume, frequency, and number of puffs. For that reason, specifications must be developed specifically for cannabis vaping systems. to ensure the data can be validated under standardized testing conditions.

The specifications of CORESTA Method 81 are shown in Table 1.

Table 1: Major specifications defined in CORESTA Method No 81 (9)

Puff Specification Value
Puff Profile Rectangular with a pressure drop device of 1000 Pa ± 50 Pa.
Profile Maximum Flow Rate 18.5 ml/s ± 1 ml/s
Puff Duration 3 s ± 0.1s
Puff Volume 55 ml ± 0.3 ml
Puff Frequency 1 puff every 30s
Puff Number Total number of puffs collected from an e-cigarette


There are many different aerosol vaping testing systems on the market, which can deliver these specifications. An example is shown in Figure 2 which is the Cerulean CETI 8 – an 8-port aerosol testing machine that can and can accommodate up to eight different e-cigarettes or vaping devices (10).


Figure 2: A typical 8-port aerosol testing machine (


Once the device has gone through its testing protocol and the condensate has been trapped, collected and analyzed by ICP-MS, the weight of each element in nanograms (ng) in the collected aerosol is then calculated based on the number of puffs tested. A common unit to compare different ENDS or ECDS is typically ng/10 puffs or ng/50 puffs, depending on the type of product being tested and the purpose of the analysis.

Collecting Hydrophobic Liquids

As mentioned previously, analyzing liquids from ENDS and ECDS is a fairly seamless dilute and shoot method using a nitric acid/hydrochloric acid mixture, because diluents like propylene glycol and vegetable glycerin are hydrophillic. Even liquids from ECDS are not over complicated because the cannabinoid oils and diluents like MCT oil can either be diluted with a suitable organic solvent or the oils can be digested in a microwave digestion system. However, to sample an aerosol containing different oils and cannabinoids (ie hydrophillic and hydrphobic in nature) is an order of magnitude more difficult. Filters tend to become clogged by viscous condensates and there is a strong likelehood that the collected material will be oily in nature which might require an organic solvent to make sure that will also maintain the metallic contaminants in solution.

This was exemplified by the trapping methodology described in a recent publication by Pappas and coworkers, which is based on a modification of the method used for ENDS (11, 12). The modification of this method for oily ECDS aerosols includes an initial tubing rinse with diethylene glycol monoethyl ether (DEGMEE). This solvent has low volatility, is an excellent solvent for oils, and is water soluble. In this study, DEGMEE dissolves and dilutes the oil droplets from up to 50 CORESTA Method 81 vaping puffs as it passes through the condensation tube into a 50 mL polymethyl pentene (PMP) volumetric flask. The remnant in the condensation tube is then followed by multiple rinses with 2% v/v nitric acid + 1% v/v hydrochloric acid into the same 50 mL flask, and dilution to 50 mL with the same acid solution. Since DEGMEE is both oil and water soluble, it emulsifies the oil with the aqueous acid for analysis. Calibration standards are prepared in 2% v/v nitric acid, + 1% v/v hydrochloric acid + 10% DEGMEE for the purpose of matrix matching with the same solution obtained after rinsing aerosol metals from the condensation tube.

So there are definitely analytical challenges to characterizing ECDS aerosols for elemental contaminants by ICP-MS. Without question, carrying out quantitation of low-level impurities requires knowledge and experience of working in the ultra-trace environment, parrticulary when the sample matrix is hydrophillic and hydrphobic in nature, that requires specialized sampling stratergies. These are not insurmountable hurdles to overcome, but in light of the many novice users in cannabis testing labs who are not well versed in the complexities of the ICP-MS technique, these issues can easily lead to errors and inaccuracies that could result in false positive and false negative results.

Final Thoughts

Currently the regulated limits of the traditional “big four” heavy metals in inhaled cannabis products in the majority of the 36 US states where cannabis is legal are shown in Table 2 (13).


Table 2: Maximum Limits for Heavy Metal Contaminants in Inhaled Cannabis Products (13)


Heavy Metal Maximum Limit (µg/g)
Pb 0.5
Cd 0.3
As 0.2
Hg 0.1


So what should be included in a panel of elements for ECDS aerosols? Clearly the point of testing vaping devices is to find out what toxic substances are being corroded and transported to the user? Assuming the cannabinoid extracts have been tested for the state required panel and are below the maximum allowable limits, if any of these other elements show up in the device aerosol, they will most probably have come from the vaping process. So it would be perfectly valid to include all the likely metal candidates to the regulated list for the state, based on the design of the vaping device, but it is not yet clear what limits should be set for those additional elements. Only time will tell, but other states are watching these new Colorado regulations very closely and possibly looking to add this to their list of regulatations in the near future. In addition, the uncertaintly about how these devices will be tested might even encourage the industry to design vaping devices with more inert materials that are unlikely to corrode or even develop systems that are free of metal components.



This article has been summarized from the author’s recent book entitled “Measuringng Heavy Metal Contaminants in Cannabis and Hemp”, ISBN 9780367417376, with the permission from CRC Press, Boca Raton, FL.


Further Reading

  1. Colorado to Require Cannabis Vapor Testing by 2022, L. Bear-McGuiness , Analytical Cannabis, October 15, 2020,
  2. Heavy Metal Contaminants from Cannabis Vaporizer Cartridges: Valid Concern or Blowing Smoke? I. Afia, R. Weltman, K. Boyar, CannMed Conference, September 2019:
  3. Outbreak of Lung Injury Associated with the Use of E-Cigarette, or Vaping, Products, Center for Disease Control and Prevention (CDC) Website:
  4. Lead and Other Toxic Metals Found in E-Cigarette Vapors, P. Olmedo et. al., Johns Hopkins Bloomberg School of Public Health, February 7, 2017,
  5. Analysis of Toxic Metals in Liquid from Electronic Cigarettes, N. Gray, International Journal of Environmental Research and Public Health, 2019;16: doi: 10.3390/ijerph16224450.
  6. Analysis of Toxic Metals in Electronic Cigarette Aerosols Using a Novel Trap Design, M. Halstead, Journal of Analytical Toxicology, 2020;44:149–155.
  7. Toxic Metal-Containing Particles In Aerosols from Pod-Type Electronic Cigarettes. R.S Pappas, et al., Journal of Analytical Toxicology, 2020; doi: 10.1093/jat/bkaa088.
  8. ISO Method, 20768:2018, Vapor Products – Routine Analytical Vaping Machine – Definitions and Standard Conditions,
  9. CORESTA Recommended Method Number 81, Routine Analytical Machine for E-Cigarette Aerosol Generation and Collection – Definitions and Standard Conditions, June, 2015,
  10. Cerulean CETI 8 E-cigarette vaping instrument (8 Channel),
  11. Analysis of Toxic Metals in Electronic Cigarette Aerosols Using a Novel Trap Design, M. Halstead, Journal of Analytical Toxicology, 2020;44:149–155.
  12. Measurement of Elemental Constituents of Vaping Liquids and Aerosols with ICP-MS. R.S. Pappas et al. in R.J. Thomas, Measuring Heavy Metal Contaminants in Cannabis and Hemp, CRC Press, 2020, ISBN 9780367417376.
  13. The Status of US States’ Legalization of Medical Marijuana,