Analysis of Pesticides and Mycotoxins in Cannabis Brownies

Abstract

Government regulations in California require the cannabis industry to test edible products for an extensive list of pesticides and mycotoxins. Here, an effective workflow for this complex analysis was developed in brownies, and optimization strategies are detailed in order to provide a starting point for similar matrices. Both LC-MS/MS and GC-MS/MS were employed and excellent results for LOQ, linearity, accuracy, and precision were attained for all the target compounds.

Introduction

Legalization of cannabis (marijuana) for recreational and medical purposes demands accurate, reliable analytical methods to assess the quality and safety of any cannabis-derived product. Developing effective methods requires careful consideration of analyte properties and matrix effects, and testing is further complicated because requirements vary by country and state for different product types. Among the various commercially available cannabis products, edibles constitute one of the most popular categories. Edibles encompass a broad variety of products, including several types of beverages, chocolates, baked goods, and candy among others. Currently, the state of California demands the analysis of an extensive list of pesticides and mycotoxins not only in cannabis flower but also in cannabis-derived goods [1]. For this reason, methods capable of tackling challenges for different matrices are highly desired.

The work presented here describes the development of a test method for the analysis of pesticides and mycotoxins in cannabis brownies using the California list. We selected brownies as a model matrix due to their popularity among cannabis edibles users, and also because they contain high levels of potential interferences (carbohydrates and fats). We discuss the development work that was necessary to optimize this method for brownies as a point of reference for labs developing methods for similar edibles (e.g., cookies or other baked goods). The final method established here for the analysis of pesticides and mycotoxins in cannabis brownies provided excellent results in terms of linearity, accuracy, precision, and limits of quantitation (LOQs).

Experimental

Initially, method development work was performed to optimize the sample preparation procedure and to assess the impact of matrix effects on quantitative analysis. The outcome of those experiments is discussed in the results section, and the final recommended methodology for the analysis of pesticides and mycotoxins in cannabis brownies is presented here.

Sample Preparation

For a solid sample like brownies, the first step in sample preparation is matrix homogenization. We used a SPEX Freezer/Mill grinder to pulverize the sample into a fine powder. Samples were precooled for 2 min, then, three cycles of 2 min grinding (at 15 cps) and 1 min cooling produced a loose, homogeneous powder that was easy to work with. (Alternatively, a food processor with dry ice could be used.)

Pulverized blank brownie matrix (0.5 g) was weighed into 4.0 mL glass vials (cat.# 24654) and fortified with pesticides and mycotoxins at the levels defined in Table I. A mixture of internal standards was added at 200 ng/g (compounds listed in Table II). 1.5 mL of acetonitrile acidified with 1% acetic acid was added to the sample. The sample was vortexed and sonicated for 5 min (no centrifugation required), and then the supernatant was passed through a 100 mg Resprep C18 SPE cartridge (cat.# 26030). An additional 1.5 mL of extraction solvent (acidified acetonitrile) was added to the sample pellet, and then the sample was vortexed again. The supernatant was passed through the same C18 cartridge.

For LC-MS/MS analysis, 750 µL of supernatant was mixed with 250 µL of water, and then centrifuged for 5 min at a low temperature (~7 ⁰C). A 2 μL aliquot of final extract was injected into the LC-MS/MS system.

For GC-MS/MS analysis, the remaining supernatant was transferred to a Q-sep QuEChERS dSPE tube containing pre-weighed magnesium sulfate and PSA (cat.# 26215). After vortexing and centrifuging, 500 μL of extract was mixed with 500 μL of acidified acetonitrile. A 1 μL aliquot of final extract was injected into the GC-MS/MS system.

Calibration Standards and Quality Control (QC) Samples

A nine-point calibration curve ranging from 5 to 700 ng/g was prepared in matrix in triplicate to assess linearity and allow quantitative analysis. To evaluate method accuracy and precision, three different QC concentration levels (10, 100, and 500 ng/g) were prepared in quadruplicate. For both the calibration standards and the QC samples, 0.5 g of pulverized blank brownie samples were weighed in 4 mL vials. Table I summarizes the µL of stock solution added to each vial to obtain different concentrations in matrix.

Table I: Preparation of calibration standard and QC sample concentration levels.

Instrument Conditions

LC-MS/MS analysis of pesticides and mycotoxins in cannabis brownies was performed using a 2.7 µm x 100 mm x 2.1 mm Raptor ARC-18 analytical column (cat.# 9314A12) and a Shimadzu LCMS-8060 LC-MS/MS. General instrument conditions, compound-specific ESI polarity, and analyte transitions are provided in Figure 1.

GC-MS/MS analysis was performed using a 30 m x 0.25 mm x 0.25 µm Rxi-5ms (cat.# 13423) analytical column and a Thermo Scientific TSQ 8000 triple quadrupole GC-MS/MS. Instrument conditions and analyte transitions are provided in Figure 2. Note that it was important to use a single taper inlet liner with wool (cat.# 23447) for this application. The wool packing enhances vaporization and mixing/homogenization with the carrier gas for better reproducibility, and the taper at the bottom of the liner funnels analytes onto the column, reducing the potential for interactions with the inlet seal. In addition, a 10 min hold time was used at the end of each run in order to ensure that compounds would not carry over and interfere with subsequent analyses.

Results and Discussion

Chromatographic Performance

In order to cover all of the compounds regulated by the state of California, we developed both LC-MS/MS and GC-MS/MS methods. Although the majority of pesticides are easily analyzed using LC-MS/MS, hydrophobic pesticides, such as chlordane, methyl parathion, captan, chlorfenapyr, and pentachloronitrobenzene (PCNB), are either not detectable or are very poorly ionized under electrospray ionization conditions. Other pesticides, such as cyfluthrin and cypermethrin, have rather low responses in ESI, but they can be analyzed using both GC-MS/MS and LC-MS/MS instrumental methods. In order to achieve an acceptable ESI response for these two analytes, ionization source temperatures should be set at relative low values. In our case, the interface, desolvation line, and heating block temperatures were all set at 100 °C. In this work, chlordane, methyl parathion, captan, chlorfenapyr, and pentachloronitrobenzene (PCNB) were analyzed using GC-MS/MS; cyfluthrin and cypermethrin were analyzed using both instrumental platforms; and the rest of the pesticides and mycotoxins were run using only LC-MS/MS. Figures 1 and 2 show the LC and GC chromatograms for the target analytes, respectively. One of the main advantages of our LC gradient conditions is that all mycotoxins elute between 2.4 and 3.6 minutes, while the main cannabinoids, such as CBD, CBG, CBN, THC, and THCA-A, will start to elute after 5.6 min. Based on a comparison to a mixed cannabinoids standard analyzed under the same conditions used here, the only coelutions between the target analytes and the main cannabinoids would be between CBD and piperonyl butoxide (5.7 min); CBG and chlorpyrifos (5.8); and cyfluthrin and CBN (6.5 min). THC (delta 8 and 9) and THCA-A did not coelute with any of the target compounds.

	Peaks	tR (min)	Precursor Ion	Product Ion 1	Product Ion 2	Polarity
1.	Daminozide-d6	0.6	167.0	149.3	49.3	+
2.	Daminozide	0.7	161.1	44.1	143.2	+
3.	Acephate	1.5	184.0	143.1	95.1	+
4.	Oxamyl	1.8	237.1	72.1	90.1	+
5.	Flonicamid	1.9	230.1	203.1	174.1	+
6.	Methomyl	1.9	163.1	88.1	106.1	+
7.	Thiamethoxam	1.9	292.0	211.1	181.1	+
8.	Imidacloprid	2.2	256.1	209.1	175.1	+
9.	Mevinphos I	2.2	225.1	127.1	193.2	+
10.	Acetamiprid	2.2	223.0	126.1	56.1	+
11.	Dimethoathe-d6	2.2	236.1	205.1	-	+
12.	Dimethoate	2.3	230.0	199.1	125.1	+
13.	Thiacloprid	2.4	253.0	126.0	90.1	+
14.	Mevinphos II	2.4	225.1	127.1	193.2	+
15.	Aflatoxin G2	2.4	331.2	189.3	115.2	+
16.	Aflatoxin G1	2.4	329.2	243.2	215.3	+
17.	Aldicarb	2.5	116.0	89.2	70.2	+
18.	Aflatoxin B2	2.5	315.3	287.2	243.3	+
19.	Dichlorvos	2.6	220.9	109.1	79.2	+
20.	Dichlorvos-d6	2.6	227.0	115.1	-	+
21.	Aflatoxin B1	2.6	313.2	241.2	128.2	+
22.	Imazalil	2.6	297.0	159.0	201.0	+
23.	Carbofuran	2.6	222.1	123.1	165.2	+
24.	Propoxur	2.6	210.1	111.1	93.1	+
25.	Carbaryl-d7	2.7	209.2	152.2	-	+
26.	Carbaryl	2.7	202.1	145.1	127.1	+
27.	Diuron-d6	2.9	239.1	78.2	-	+
28.	Atrazine-d5	2.9	221.2	179.1	-	+
29.	Naled	2.9	397.8	127.1	109.1	+
30.	Metalaxyl	2.9	280.2	220.2	192.2	+
31.	Spiroxamine	2.9	298.3	144.2	100.2	+
32.	Chlorantraniliprole	3.0	483.9	452.9	285.9	+
33.	Phosmet	3.0	318.0	160.1	77.2	+
34.	Azoxystrobin	3.1	404.0	372.1	344.1	+
35.	Linuron-d6	3.1	255.1	160.1	-	+
36.	Fludioxonil	3.2	247.0	180.0	126.0	-
37.	Methiocarb	3.2	226.1	169.1	121.1	+
38.	Dimethomorph I	3.2	388.2	301.2	165.3	+
39.	Boscalid	3.2	342.9	307.1	140.1	+
40.	Paclobutrazol	3.3	294.3	70.1	125.1	+
41.	Malathion	3.3	331.0	127.2	285.2	+
42.	Dimethomorph II	3.4	388.2	301.2	165.3	+
43.	Myclobutanil	3.4	289.1	70.1	125.1	+
44.	Bifenazate	3.4	301.0	198.1	170.2	+
45.	Ochratoxin A	3.5	404.2	239.1	358.3	+
46.	Fenhexamid	3.5	302.1	97.1	55.2	+
47.	Spirotetramat	3.7	374.2	302.1	216.1	+
48.	Ethoprophos	3.8	243.1	131.1	97.1	+
49.	Fipronil	3.8	436.8	331.8	251.9	-
50.	Fenoxycarb	3.9	302.1	88.1	116.1	+
51.	Kresoxim-methyl	4.1	314.2	267.2	222.2	+
52.	Tebuconazole	4.2	308.1	70.1	125.1	+
53.	Diazinon-d10	4.2	315.2	170.2	-	+
54.	Spinosyn A (spinosad)	4.3	732.4	142.2	98.1	+
55.	Diazinon	4.3	305.1	169.2	153.2	+
56.	Coumaphos	4.4	363.1	227.1	307.1	+
57.	Pyridaben	4.4	365.1	309.2	147.2	+
58.	Propiconazole	4.4	342.0	159.0	69.2	+
59.	Clofentezine	4.5	303.0	138.1	102.1	+
60.	Spinosyn D (spinosad)	4.8	746.5	142.3	98.4	+
61.	Spinosyn J (spinetoram)	4.8	748.5	142.3	98.3	+
62.	Trifloxystrobin	4.9	409.2	186.1	145.1	+
63.	Prallethrin	4.9	301.2	123.2	105.2	+
64.	Pyrethrin II	5.2	373.1	161.1	133.2	+
65.	Spinosyn L (spinetoram)	5.4	760.5	142.2	98.1	+
66.	Piperonyl butoxide	5.7	356.3	177.2	119.2	+
67.	Chlorpyrifos	5.8	349.9	198.0	97.1	+
68.	Hexythiazox	5.9	353.1	228.1	168.1	+
69.	Etoxazole	6.4	360.2	141.1	304.2	+
70.	Spiromesifen	6.4	273.2	255.2	187.2	+
71.	Pyrethrin I	6.6	329.2	161.2	105.2	+
72.	Cyfluthrin (qualifier)	6.6	453.1	193.2	-	+
73.	Cyfluthrin	6.6	451.1	191.2	-	+
74.	Cypermethrin	6.8	433.1	191.0	416.0	+
75.	(E)-Fenpyroximate	6.8	422.2	366.1	138.1	+
76.	trans-Permethrin	7.4	408.3	183.2	355.1	+
77.	cis-Permethrin	7.7	408.3	183.2	355.1	+
78.	Avermectin B1a	7.7	890.5	305.4	567.4	+
79.	Etofenprox	7.8	394.3	177.2	359.3	+
80.	Bifenthrin	8.0	440.0	181.2	166.2	+
81.	Acequinocyl (precursor ion 1)	9.3	402.3	343.2	189.0	+
82.	Acequinocyl (precursor ion 2)	9.3	386.0	344.2	189.1	+

Column	Raptor ARC-18 (cat.# 9314A12)
Dimensions:	100 mm x 2.1 mm ID
Particle Size:	2.7 µm
Pore Size:	90 Å
Guard Column:	Raptor ARC-18 EXP guard column cartridge 5 mm, 2.1 mm ID, 2.7 µm (cat.# 9314A0252)
Temp.:	40 °C
Standard/Sample	California pesticide standard #1 (cat.# 34124)
	California pesticide standard #2 (cat.# 34125)
	California pesticide standard #3 (cat.# 34126)
	California pesticide standard #4 (cat.# 34127)
	California pesticide standard #5 (cat.# 34128)
	California pesticide standard #6 (cat.# 34129)
	Dimethoate-d6 (cat.# 31988)
	Dichlorvos-d6 (cat.# 31987)
	Carbaryl-d7 (cat.# 31985)
	Diazinon-d10 (cat.# 31986)
	Atrazine-d5 (cat.# 31984)
	Diuron-d6 (cat.# 31989)
	Liuron-d6 (cat.# 31990)
	Aflatoxins standard (cat.# 34121)
	Ochratoxin A (cat.# 34122)
	Compounds not present in these mixes were obtained separately.
Diluent:	75:25 Acetonitrile:water
Conc.:	5-15 ng/mL (Expected concentration range in extract of brownie initially spiked at 100 ng/g.)
Inj. Vol.:	2 µL
Mobile Phase
A:	Water, 2 mM ammonium formate, 0.1% formic acid
B:	Methanol, 2 mM ammonium formate, 0.1% formic acid
	Time (min) Flow (mL/min) %A %B 0.00 0.5 95 5 1.5 0.5 35 65 8.5 0.5 5 95 9.5 0.5 0 100 10.5 0.5 0 100 10.6 0.5 95 5 12 0.5 95 5

Detector	MS/MS
Ion Mode:	ESI+/ESI-
Mode:	MRM
Instrument	UHPLC
Sample Preparation	Brownies were pulverized using a SPEX Freezer/Mill grinder and 0.5 g samples were fortified with pesticides and mycotoxins at 100 ng/g. A mix of internal standards was added at 200 ng/g. 1.5 mL of acetonitrile acidified with 1% acetic acid was added to the sample. The sample was vortexed and sonicated for 5 min, and then the supernatant was passed through a 100 mg Resprep C18 SPE cartridge (cat.# 26030). An additional 1.5 mL of extraction solvent (acidified acetonitrile) was added to the sample pellet, and then the sample was vortexed again. The supernatant was passed through the same C18 cartridge. 750 µL of extract was mixed with 250 µL of water, and then centrifuged for 5 min at low temperature (~7 ⁰C). 2 μL of final extract was injected into the LC-MS/MS system.
Notes	Want even better performance when analyzing metal-sensitive compounds? Check out Inert LC columns at www.restek.com/inert.

Télécharger le PDF

	Peaks	tR (min)	Polarity	Precursor Ion	Product Ion	Transition Type
1.	Atrazine-D5	7.5	Positive	220.0	58.0	Quantifier
2.	Atrazine-D5	7.5	Positive	205.0	127.0	Qualifier
3.	Quintozene	7.8	Positive	294.9	236.9	Quantifier
4.	Quintozene	7.8	Positive	236.8	118.9	Qualifier
5.	Methyl parathion	8.2	Positive	263.0	109.0	Quantifier
6.	Methyl parathion	8.2	Positive	263.0	79.0	Qualifier
7.	Captan	9.1	Positive	184.0	149.1	Quantifier
8.	Captan	9.1	Positive	184.0	134.1	Qualifier
9.	trans-Chlordane	9.1	Positive	271.9	237.0	Quantifier
10.	trans-Chlordane	9.1	Positive	372.9	265.9	Qualifier
11.	cis-Chlordane	9.3	Positive	372.9	265.9	Quantifier
12.	cis-Chlordane	9.3	Positive	271.9	237.0	Qualifier
13.	Chlorfenapyr	9.5	Positive	247.1	227.1	Quantifier
14.	Chlorfenapyr	9.5	Positive	59.1	31.1	Qualifier
15.	Cyfluthrin	11.5	Positive	163.0	127.1	Quantifier
16.	Cyfluthrin	11.5	Positive	199.1	170.1	Qualifier
17.	Cypermethrin	11.7	Positive	163.0	127.1	Quantifier
18.	Cypermethrin	11.7	Positive	181.1	152.1	Qualifier

Column	Rxi-5ms, 30 m, 0.25 mm ID, 0.25 µm (cat.# 13423)
Standard/Sample	California pesticide standard #1 (cat.# 34124)
	California pesticide standard #2 (cat.# 34125)
	California pesticide standard #3 (cat.# 34126)
	California pesticide standard #4 (cat.# 34127)
	California pesticide standard #5 (cat.# 34128)
	California pesticide standard #6 (cat.# 34129)
	Atrazine-d5 (cat.# 31984)
Diluent:	Acetonitrile
Conc.:	5-7.5 ng/mL Expected concentration range in extract after extracting from brownie fortified at 100 ng/g (final extract was diluted in half with acetonitrile).
Injection
Inj. Vol.:	1 µL splitless
Liner:	Topaz 4.0 mm ID single taper inlet liner w/wool (cat.# 23447)
Inj. Temp.:	250 °C
Purge Flow:	5 mL/min
Oven
Oven Temp.:	90 °C (hold 1 min) to 310 °C at 25 °C/min (hold 10 min)
Carrier Gas	He, constant flow
Flow Rate:	1.4 mL/min

Detector	MS/MS
Transfer Line Temp.:	290 °C
Analyzer Type:	Quadrupole
Source Temp.:	330 °C
Electron Energy:	70 eV
Tune Type:	PFTBA
Ionization Mode:	EI
Instrument	Thermo Scientific TSQ 8000 Triple Quadrupole GC-MS
Sample Preparation	Brownies were pulverized using a SPEX Freezer/Mill grinder and 0.5 g samples were fortified with pesticides and mycotoxins at 100 ng/g. A mix of internal standards was added at 200 ng/g. 1.5 mL of acetonitrile acidified with 1% acetic acid was added to the sample. The sample was vortexed and sonicated for 5 min, then the supernatant was passed through a 100 mg Resprep SPE C18 cartridge (cat.# 26030). An additional 1.5 mL of extraction solvent (acidified acetonitrile) was added to the sample pellet, and the sample was vortexed again. The supernatant was passed through the same C18 cartridge. After reserving 750 μL for LC-MS analysis, the remaining supernatant was transferred to a Q-sep QuEChERS dSPE tube containing pre-weighed magnesium sulfate and PSA (cat.# 26215). After vortexing and centrifuging, 500 μL of extract was mixed with 500 μL of acidified acetonitrile. 1 μL of final extract was injected into the GC-MS/MS system.

Télécharger le PDF

Method Development Considerations: Sample Preparation Optimization

Because our list of target compounds comprises a broad range of analytes with diverse physical/chemical properties and polarities, it is essential that the sample preparation methodology be nonselective and to also ensure that major interferences are effectively removed without losing any of the target analytes. Acetonitrile, which is the extraction solvent traditionally used in QuEChERS methods, was employed here because it allows for the recovery of a broad range of polar and nonpolar compounds. Initially, we conducted experimental trials using 1 g of sample and at least 5 mL of acetonitrile per extraction. However, by reducing the sample size to 0.5 g, we were able to cut the extraction solvent volume to 3 mL per sample and still comply with the California regulations. Decreasing solvent waste allows for greener analytical practices and reduces the costs associated with sample analysis and waste disposal. Based on these benefits, acetonitrile extraction solvent and 0.5 g samples were used for further optimization experiments.

Of all the target analytes, daminozide is one of the most challenging, and additional adjustment of the extraction procedure was required to accommodate this pesticide. Daminozide is highly polar (log P = -1.5), so it is difficult to extract completely from the brownie matrix, and it is not easily retained under reversed-phase conditions. Additionally, low molecular weight compounds, such as daminozide, are more prone to interferences and high background noise when using MS, especially when having poor chromatographic retention. Several strategies were tested to enhance the recovery of daminozide at conditions that were also favorable for the rest of the target analytes.

The first parameter that was evaluated was the acidification of the extraction solvent. To this end, we compared the effect of using acetonitrile with acetic acid (AA) at 1% (v/v) against extracting with pure acetonitrile. Figure 3 shows the results for those target compounds that exhibited at least a 40% difference between treatments. When extracting with acidified acetonitrile, the relative response of daminozide increased by about 80%, whereas spiroxamine and ochratoxin A showed an improvement of 45 and 47% in response, respectively. Based on these findings, and the fact that none of the target compounds showed a significant decrease in response, we used acetonitrile with 1% acetic acid as the extraction solvent in our final sample preparation workflow. This provides an additional benefit in that acidification of the extracts avoids the degradation of captan, a pesticide suitable for GC-MS/MS analysis that easily degrades at neutral to high pH values [2].

The second sample preparation parameter we evaluated was the application of a single 3 mL extraction versus a two-step (1.5 mL each) extraction. Figure 4 summarizes the results for those analytes that exhibited statistical differences (t-test, p-value <0.05 at a 95% confidence) between the two extraction conditions (both of which used acidified acetonitrile). Interestingly, daminozide exhibited the most significant improvement, with 40% increase in its relative response. The other analytes in the figure showed improvements in response ranging from 10 to 20%. The increase in relative responses seen here when using a second extraction step should result in higher sample extraction recoveries in real-world samples. As this effect depends on the analyte, matrix, and extraction solvent, an evaluation at the desired experimental conditions is always recommended.

Following extraction with acidified acetonitrile, a simple cleanup step with C18 SPE cartridges was used for all samples to remove major hydrophobic interferences that were coextracted from brownies. After this, samples for LC-MS/MS and GC-MS/MS analysis were processed differently. LC-MS/MS samples were diluted in water and centrifuged, which permitted the separation of any remaining lipids that were not retained in the SPE cartridge because they are less soluble in the final extract once water is added. GC-MS/MS samples were treated with magnesium sulfate and primary secondary amine (PSA) dSPE cleanup to remove moisture and sugars prior to analysis. Note that because PSA can bind to pesticides such as daminozide and spiroxamine, dSPE was conducted only for the GC amenable pesticides.

Method Development Considerations: Minimizing Matrix Effects

It is well known that electrospray ionization is prone to suppression/enhancement effects caused by coextracted matrix interferences. These issues can be overcome by adjusting the sample preparation method towards more selective conditions, using an appropriate internal standard, and/or adjusting the chromatographic method to resolve analytes from coeluting interferences. For this reason, after selecting the experimental conditions for the analysis of pesticides and mycotoxins in cannabis brownies, an assessment of absolute matrix effects was conducted according to the methodology proposed by Matuszewski et al. [3]. To estimate absolute matrix effects, we compared the response of all target analytes fortified in blank matrix extract versus their response in neat solvent fortified at the same concentration (15 ppb). As can be seen in Figure 5, only daminozide showed a dramatic enhancement in its response. Considering the poor retention of daminozide at the LC gradient conditions selected for this work, this pesticide is likely coeluting with multiple unretained interferences normally occurring in matrices like brownies. Based on these results, daminozide-D6 was added to the group of internal standards to specifically correct for any variations associated with daminozide’s response.

Absolute matrix effect (%) is calculated as: (response of fortified sample extract/response of fortified solvent) x 100.

In addition to investigating absolute matrix effects, we assessed absolute analyte recoveries under the final method conditions. For this purpose, responses corresponding to brownie extracts obtained from samples fortified at 100 ng/g (prior to extraction) were compared to blank brownie extracts that were then fortified (post extraction) with the same amount of analytes. As shown in Figure 6, except for daminozide, the lowest analyte recoveries were 68% for spiroxamine and ochratoxin A. Despite the multiple efforts to enhance daminozide’s recovery, it was only possible to collect 30% of the originally spiked amount. This demonstrates that the best approach to obtain reliable and accurate results when analyzing pesticides and mycotoxins in cannabis brownies is using a surrogate (blank) matrix to prepare different calibration levels, and using a deuterated analogue for daminozide. The use of calibrators prepared in solvent and a lack of appropriate internal standards will likely lead to biased measurements.

Absolute analyte recovery (%) is calculated as: (response of fortified matrix sample/response of fortified sample extract) x 100.

Method Verification Results

Method performance for the analysis of pesticides and mycotoxins in cannabis brownies was evaluated in terms of linearity, accuracy, and precision. Summary results are presented in Table III.

Calibration curves were constructed by plotting analyte area/internal standard area ratios versus the analyte concentrations in fortified brownie matrix standards across a range of 5–700 ng/g. A weighing factor of 1/x was applied to all the calibration curves. The majority of compounds exhibited excellent linearity as demonstrated by R2 values of ≥0.9990 (values ranged from 0.9939 to 0.9999).

Accuracy and precision were assessed at three concentration levels that were selected to cover the linear range: low (10 ng/g); medium (100 ng/g); and high (500 ng/g). Excellent results were obtained for all the target analytes because the use of standards prepared in matrix minimized matrix enhancement and suppression. Recovery values ranged from 71.2 to 116% and RSD values were all under 25%, indicating good accuracy and precision.

The majority of analytes showed LOQ values under 10 ng/g, and it was demonstrated that the proposed methodology allows for the quantification of all the target compounds at concentrations below the action levels requested by the state of California in cannabis goods. Cyfluthrin and cypermethrin exhibited lower LOQ values via GC-MS/MS in comparison to LC-MS/MS; however, these two pesticides can be analyzed at the requested action levels using either of the two instrumental platforms. The LOQ for each compound was defined as the lowest concentration with a signal-to-noise ratio of at least 10, a difference of <25% between the fortified concentration and the estimated concentration, and a <25% replicate precision value.

Stability of the target compounds in LC extracts after 24 and 48 hours of storage in the autosampler at 10 ⁰C (n=4) was assessed relative to the peak intensities of freshly prepared samples (0 hours). Most compounds showed a change in response of less than 10% over the entire study period.

Table III: Verification results for the optimized final method for the analysis of pesticides and mycotoxins in cannabis brownies. Results were determined by LC-MS/MS, except for those compounds that were reported by GC-MS/MS as indicated.

^†Reported results were determined by GC-MS/MS.
^*Category I pesticides, LOQ is ≤100 ng/g
**Action level is the total concentration of both isomers together.
Results are calculated based on relative contribution of each isomer to the overall fortification levels as follows:
^aMevinphos I: low: 8 ng/g; medium: 79 ng/g; high: 395 ng/g
^bMenvinphos II: low: 2 ng/g; medium: 21 ng/g; high: 105 ng/g
^cDimethomorph I: low: 4 ng/g; medium: 39 ng/g; high: 195 ng/g
^dDimethomorph II: low: 6 ng/g; medium: 61 ng/g; high: 305 ng/g
^eSpinosad- spinosyn A: low: 7 ng/g; medium: 71 ng/g; high: 355 ng/g
^fSpinosad - spinosyn D: low: 3 ng/g; medium: 29 ng/g; high: 145 ng/g
^gSpinetoram - spinosyn J: low: 8 ng/g; medium: 80 ng/g; high: 400 ng/g
^hSpinetoram - spinosyn L: low: 2 ng/g; medium: 20 ng/g; high: 100 ng/g
ⁱPyrethrin I: low: 5 ng/g; medium: 54 ng/g; high: 270 ng/g
^jPyrethrin II: low: 3 ng/g; medium: 34 ng/g; high: 170 ng/g
^kPermethrin-cis: low: 4 ng/g; medium: 41 ng/g; high: 205 ng/g
^lPermethrin-trans: low: 6 ng/g; medium: 59 ng/g; high: 295 ng/g

Conclusion

An effective workflow for the analysis of pesticides and mycotoxins in cannabis brownies was developed and optimization strategies were detailed in order to provide a starting point for similar matrices. LC-MS/MS and GC-MS/MS amenable compounds were all extracted using acidified acetonitrile and then cleaned up using C18 SPE cartridges. GC-MS/MS samples required an additional dSPE cleanup with magnesium sulfate and PSA. The use of acidified acetonitrile was proven to be crucial in the recovery of analytes such as daminozide, spiroxamine, and ochratoxin A. Excellent results in terms of LOQ values, linearity, accuracy, and precision were attained for all the target compounds. Since this methodology only uses 3 mL of extraction solvent per sample, a significant reduction in solvent usage/waste is also possible. Extracts showed acceptable stability even after 48 hours in the LC autosampler.

Want even better performance when analyzing mycotoxins and other metal-sensitive compounds? Learn more at www.restek.com/inert

References

1. Text of Regulations, Bureau of Cannabis Control, California Code of Regulations, https://cannabis.ca.gov/wp-content/uploads/sites/13/2019/01/Order-of-Adoption-Clean-Version-of-Text.pdf, (accessed 8 November 2019).
2. Quantification of residues of folpet and captan in QuEChERS extracts, EU Reference Laboratories for Residues of Pesticides, http://www.eurl-pesticides.eu/userfiles/file/EurlSRM/meth_CaptanFolpet_EurlSRM.pdf, (accessed 7 November 2019).
3. B. K. Matuszewski, M. L. Constanzer, C. M. Chavez-Eng, Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC−MS/MS, Anal. Chem. 75 (2003) 3019–3030. https://pubs.acs.org/doi/10.1021/ac020361s