Metals in Cannabis and Related Substances—Regulations and Analytical Methodologies: Page 2 of 4

March 1, 2018
Abstract / Synopsis: 

Evolution in the legislation regarding the production and consumption of cannabis and related products for medicinal and recreational uses has led to emerging regulations regarding the potency, terpene profiles, and impurity content of such products. Among the impurities, or contaminants, that require testing are metals. Because of their toxicity and carcinogenicity, most jurisdictions are, at a minimum, requiring testing for arsenic, lead, cadmium, and mercury. Metals accumulate in plant material through normal metabolic processes. Furthermore, some plants are capable of hyperaccumulating metals to concentrations well above those in the soils and waters that nourish them. Some such metals are beneficial or nutritional in nature, whereas others show varying levels of toxicity. As such, testing for metals, particularly ones that are toxic, are important for products destined for human consumption. Quantitation of metals within cannabis materials can be accomplished through a variety of analytical techniques. Such amenable technologies include atomic absorption (AA) spectroscopy, inductively coupled plasma-optical emission spectroscopy (ICP-OES), and inductively coupled plasma-mass spectrometry (ICP-MS). As with all aspects of analytical instrumentation, each method has its advantages and disadvantages. Here, we present an overview of analytical methodologies, challenges facing the analyst, and notes on regulatory stipulations.

Current Regulations

One of the primary challenges facing cannabis analysis is non-uniform regulations. Because cannabis remains illegal at the federal level, oversight on regulations is left to the state’s discretion. As such, limits on metal concentrations, as well as which metals require testing, vary from state to state. For example, most states require testing for the “big four” heavy metals: arsenic, cadmium, lead, and mercury. Certain states go further, such as Maryland, requiring testing for barium, chromium, selenium, and silver in addition to the big four. Adding further complication, the concentration limits for various products can also vary state by state, with some jurisdictions taking a stricter approach and others more relaxed. There is also variation by method of ingestion (for example, inhalational versus oral administration) with tolerances of metals for inhalational products being stricter than for those that are orally administered (that is, eaten or ingested).

Sample Preparation

Metals analysis is carried out on aqueous samples in which the material of interest is dissolved. The dissolution process can be carried out by a few different methods, the most simple of which is conventional hot-plate or hot-block digestion. Samples are placed into a vessel with solvents, typically strong acids such as nitric acid (HNO3) or hydrochloric acid (HCl). For organic matrices such as cannabis, hydrogen peroxide (H2O2) is often used to increase the oxidation potential of the solvents. The vessels are capped and refluxed at a moderate temperature, often for ≥8 h, to ensure complete digestion.

A more common method for sample preparation for elemental spectroscopy analysis is microwave digestion. A small amount of solid sample is placed into a digestion vessel along with solvents, similar to a hot-plate digestion procedure. The vessels are sealed and placed into the microwave, which allows the solvents within the vessels to reach very high temperatures and pressures, increasing the efficiency of dissolution. A typical microwave digestion of cannabis products is complete within ~45 min.

In the case of either digestion method, the molecular constituents of the digested material are destroyed during dissolution, but the constituent elements remain in solution. After dissolution, the resulting solution is typically diluted with ultrapure water to an appropriate matrix acid concentration and is ready for analysis.

Critical to carrying out digestions for metals analysis is the use of clean labware, “trace metals grade” reagents, and ultrapure (typically >18 MΩ•cm) deionized water. As we explore in the sections below, common elemental spectroscopy methods are sensitive to metals ranging in concentration from parts per billion down to parts per trillion. Thus, reagents, such as acids, containing 1 ppm of various heavy metals may contain 50 ppb of said metals after they are diluted to the appropriate concentration, which is much higher than detection limits and, often, regulatory limits.

Analytical Methods

Atomic Absorption Spectroscopy

Long a staple of analysis labs, atomic absorption (AA) spectroscopy, is perhaps the most common analytical technique for metals analysis. Atomic absorption operates under the principle of the Beer-Lambert law, with gaseous free atoms absorbing light of a characteristic wavelength. The ratio of the intensity of the light shone through a blank sample to the intensity of light shone through one containing the analyte is the absorbance, and corresponds to the concentration of the element within the sample.

Atomic absorption spectroscopy is often attractive because of its relatively low cost compared to other elemental techniques. The instruments themselves as well as cost of operations (that is, consumables) tend to be far less that inductively coupled plasma (ICP) techniques. Additionally, with the use of a graphite furnace, detection limits can be in the low parts-per-billion to high parts-per-trillion range; however, with a graphite furnace comes increased costs.

Despite the low cost and potential for low detection limits, AA is notably limited in terms of sample throughput; these limitations arise from a few fundamental aspects of the technique. First, each element analyzed requires the use of a hollow cathode lamp, many of which are “single element” lamps, though “multielement lamps” do exist. As such, the number of elements per sample is limited by the number of lamps that the instrument can hold. Second, each element quantitated requires its own analysis. For example, during a typical analysis, the instrument will perform an injection-aspiration of the sample while transilluminating with the first hollow cathode lamp. Then, the instrument will switch lamps and repeat the injection-aspiration for any subsequent elements.

As a final limitation to the technique, elements such as As, Bi, Sb, Se, and Bi are simply not very sensitive by flame AA and benefit from the use of a hydride vapor generator to form volatile hydrides before sample injection. This approach necessitates the use of an additional piece of laboratory equipment as well as reagents. Additionally, it may require separate sample preparations because, often, potassium iodide (KI) is used to reduce As in solution from As(V) to As(III) for vapor generation; however, the use of KI may induce the precipitation of mercury as potassium tetraiodomercurate(II) from the sample solution. Thus, separate sample preparations for As and Hg when using a hydride vapor generator for analysis may be required.

Laboratories on a tight budget, testing only a few elements, or those that are not as concerned with sample throughput may benefit from the use of AA with a hydride vapor generator as their analytical technique for metals.

  1. S. Kuzdzal, R. Clifford, P. Winkler, and W, Bankert, “A Closer Look at Cannabis Testing,” Shimadzu Scientific Instruments Whitepaper (2017). Retrieved 16 January 2018:
  2. A. Hazekamp, Cannabinoids 1(1), 1-9 (1996).
  3. C. Reilly, The Nutritional Trace Metals (Wiley-Blackwell, Oxford, UK, 2004).
  4. L. Järup, Br. Med. Bull. 68(1), 167-182 (2003).
  5. I. Raskin, R.D. Smith, and D.E. Salt, Curr. Opin. Biotechnol. 8, 221-226 (1997).
  6. E. Charkowski, “Hemp ‘Eats’ Chernobyl Waste, Offers Hope for Hanford,” in Central Oregon Green Pages (1998). Republished:
  7. M. Girdhar, N.R. Sharma, H. Rehman, A. Kumar, and A. Mohan, Biotech. 4(6), 579-589 (2014).

How to Cite This Article

A.P. Fornadel, D.L. Davis, R.H. Clifford, and S.A. Kuzdzal, Cannabis Science and Technology 1(1), 36-41 (2018).