As an increasing number of U.S. states and countries enact laws to legalize the use of medicinal or recreational marijuana, there is a critical need to ensure product quality and safety. Similar to other consumer products such as foods and pharmaceuticals, cannabis testing needs to include the analysis of metals, some of which may be toxic if ingested or inhaled. Existing analytical methods used for plant-based samples can be applied to cannabis plant material and other cannabis-based products. A microwave acid digestion sample preparation procedure was verified using appropriate reference materials, and the digested cannabis samples were analyzed using both inductively coupled plasma–mass spectrometry (ICP-MS) and ICP–optical emission spectrometry (OES). The ICP-MS method used aerosol dilution to provide a highly robust plasma suitable for extended analysis of high-matrix sample digests. A fast ICP-OES method was also developed to analyze the same cannabis samples.
Cannabis-based products are available in a wide variety of formulations ranging from dry plant material, plant concentrates including waxes and distillates, and infused products such as foods and candies. Given the variety of sample matrices, existing sample preparation procedures developed for inductively coupled plasma–based techniques can be applied to cannabis products. For example, trace-element analysis of plant and nutritional supplement materials is a well-established application (1). Following acidic digestion to break down the primary components of the plant-based samples, inductively coupled plasma–mass spectrometry (ICP-MS) or ICP–optical emission spectrometry (OES) is often used for quantitative analysis because of the multielement capability, speed, and robustness of each technique. ICP-OES is suited to the analysis of mineral and micronutrients such as K, Ca, Mg, Cu, Fe, Mn, Zn, Cu, Mo, and Ni-vital elements required for plant growth. When the required analytes also include trace elements such as As, Se, Cd, Pb, and Hg, which may require lower detection limits, ICP-MS offers greater sensitivity, delivering detection limits and accurate analysis down to nanogram-per-liter (part-per-trillion) levels.
Most of the states that have legalized the use of marijuana for either medicinal or recreational use have enacted regulations for acceptable limits of toxic elements (Cd, Pb, As, and Hg) in cannabis and cannabinoid products (2). As shown in Table I (3), the limits can vary among states, and regulations governing the safety and quality of cannabis-based products are likely to evolve to include more elements. (See upper right for Table I; Table I: Example U.S. state regulations for heavy metals .)
Preparation of Cannabis Samples
Two cannabis plant samples were analyzed in this study. Approximately 0.15 g of buds from each cannabis plant was weighed into a quartz vessel. Then 4 mL of nitric acid (HNO3) and 1 mL of hydrochloric acid (HCl) were added and the samples were microwave digested using a one-step program: ramp time of 20 min to a temp of 240 °C and hold time of 15 min. Hydrochloric acid was included to ensure the stability of Ag and Hg in solution.
The digested samples were diluted using a mix of 1% HNO3 and 0.5% HCl. National Institute of Standards and Technology (NIST) 1547 Peach Leaves and NIST 1573a Tomato Leaves standard reference materials (SRMs) were prepared using the same method to verify that the digestion was complete and confirm the quantitative recovery of the analytes.
The ICP-MS system used for the analysis was a standard 7800 (Agilent), which includes the High Matrix Introduction (HMI) system. The ICP-OES system used was a standard 5110 SVDV (Agilent) fitted with an Advanced Valve System (AVS) six-port valve. Both instruments were used with an SPS 4 autosampler (Agilent). The ICP-MS system was configured with the standard sample introduction system consisting of a Micromist glass concentric nebulizer, a quartz spray chamber, and a quartz torch with a 2.5-mm i.d. injector. The interface consisted of a nickel-plated copper sampling cone and a nickel skimmer cone. The ICP-OES sample introduction system consisted of a SeaSpray nebulizer, a double-pass cyclonic spray chamber, and a 1.8-mm i.d. injector torch. Table II lists the operating conditions used for the ICP-MS and ICP-OES systems. (See upper right for Table II, click to enlarge; Table II: ICP-MS and ICP-OES operating conditions.)
The settings for the ICP-MS HMI were autotuned using an aerosol dilution factor of 4x, as appropriate for the matrix level of the sample digests. HMI enables the routine analysis of samples that contain high and variable matrix levels, while minimizing the need for conventional liquid dilution. By automating dilution in the aerosol phase, manual sample handling steps and the potential for contamination during sample preparation can be reduced, producing more accurate results. The 7800 ICP-MS system uses helium (He) collision–reaction cell (CRC) gas and kinetic energy discrimination (KED) to control all common polyatomic interferences. In plant samples such as cannabis, all the target metals can be measured at the required levels using He cell gas, so a very simple, single-mode method can be applied.
The 5110 ICP-OES system uses a vertical torch and solid-state RF generator to ensure robust, stable analysis of complex samples over extended run times. The integrated AVS valve system was used to improve the sample throughput and minimize argon gas usage.
Both instruments can perform a rapid screening measurement together with the quantitative analysis. Using a standard feature of the respective software, semiquantitative results can be reported for elements not included in the calibration standards.
Representative ICP-MS and ICP-OES calibration curves for the critical trace elements As, Cd, Pb, and Hg are shown in Figure 1. (See upper right for Figure 1, click to enlarge; Figure 1: (a) ICP-MS and (b) ICP-OES calibration curves for As, Cd, Pb, and Hg.) All curves show excellent linearity across the respective calibration ranges.
To verify the digestion procedure used for the cannabis samples, two plant material SRMs were analyzed by ICP-MS and ICP-OES. Some plant samples may contain high levels of rare earth elements (REEs), which can be problematic for the analysis of low concentrations of As and Se by ICP-MS. REEs have low second ionization potentials, so they readily form doubly charged ions (REE++). As the quadrupole mass spectrometer separates ions based on their mass-to-charge ratio (m/z), these doubly charged ions appear at half their true mass. Doubly charged ions of the REEs 150Nd, 150Sm, 156Gd, 156Dy, 160Gd, and 160Dy therefore appear at m/z 75, 78, and 80, potentially causing overlaps that can bias the results for As and Se in samples that contain high levels of the REEs. The ICP-MS system corrects for these interferences using the “half mass correction” setting in the ICP-MS MassHunter software. The mean results shown in Table III were in good agreement with the certified concentrations, where provided, including for As in NIST 1547 and Se in both NIST 1547 and 1573a. (See upper right for Table III, click to enlarge; Table III: Mean concentrations (n = 3, ppm) in two plant SRMs measured using ICP-MS and ICP-OES. Results include percent recovery compared to certified or uncertified (information) values, where available.)
Quantitative results for the two cannabis samples showed that the concentrations of As (160.0 ppb), Cd (11.33 ppb), Pb (24.00 ppb), and Co (162.1 ppb) were relatively high in cannabis sample 1. Pb and Co were also high in cannabis sample 2, at 55.40 and 143.4 ppb, respectively These concentrations were well below existing regulatory or guideline levels for As, Hg, Pb, and Cd, so a spike recovery test was carried out to check the accuracy of the ICP-MS and ICP-OES methods at the higher concentrations that may be encountered in actual sample analysis. The two cannabis sample digests were spiked with a premixed standard (Environmental Mix Spike, Agilent) containing multiple elements at 200 ppb, Na, Mg, K, Ca, and Fe at 2000 ppb, and Hg at 4 ppb.
Using the ICP-MS and ICP-OES systems’ direct analysis methods, excellent spike recoveries were achieved for most elements in the spiked samples. All recoveries were within ±20% for Cd, Pb, As, and Hg, as shown in Table IV. (See upper right for Table IV, click to enlarge; Table IV: Quantitative and spike recovery results for two cannabis samples.) The spike recovery results for K, Ca, and Mn were invalid because the concentration levels of these elements in the mixed spike solution were much too low (20 times lower) relative to the levels present in the unspiked cannabis digest samples.
Both ICP-MS and ICP-OES can be used for the quantitative analysis of multiple elements-including the four target toxic metals Cd, Pb, As, and Hg-in cannabis samples following acid digestion. The choice of which technique to use will depend on the required method detection limits, level of experience of staff in the laboratory, and available budget.
Both techniques are suitable for trace metal screening of medicinal and recreational cannabis, as well as related products. ICP-OES has a lower capital cost and is somewhat easier to use, while ICP-MS offers greater sensitivity and is more suitable for the ultratrace level analytes. To ensure that each method was simple enough to be applied to routine quality control (QC) and safety testing, the ICP-MS was operated in a single mode (helium collision mode) for all measurements. The ICP-OES was used with a 6-port valve system, suitable for high throughput applications. Based on the findings of the spike recovery test of two cannabis plant samples, both methods were found to be accurate for multiple elements over a wide concentration range. Suitability of the microwave-assisted sample preparation method was demonstrated by the good recovery results obtained for two plant-based SRMs.
Jenny Nelson is an application scientist with Agilent Technologies in Santa Clara, California. Craig Jones is an application engineer with Agilent Technologies in Santa Clara. Neli Drvodelic is an application engineer with Agilent Technologies in Melbourne, Australia. Direct correspondence to: firstname.lastname@example.org
J. Nelson, C. Jones, and N. Drvodelic, Cannabis Science and Technology1(2), 20-25 (2018).