PO-CON1752E

Pesticide analysis in Hopsand Cannabis by GC-MS-MS

ASMS 2017 TP-191

Jeffrey H. Dahl,1 Riki Kitano,1 Vikki Johnson,1

and Julie Kowalski2

1Shimadzu Scienti�c Instruments; Columbia, Maryland2Trace Analytics; Spokane, Washington

2

Pesticide analysis in Hops and Cannabis by GC-MS-MS

IntroductionThe rising popularity of medical and recreational cannabis has dramatically increased the demand for this plant. Cultivators of the closely related hops plant and cannabis may use pesticides or other chemical residues to protect their plants from mold and insects, however few of these substances may be legally used in cannabis. The possible adverse health effects of unapproved chemicals have

drawn signi�cant public attention. Many chemical residues may be measured using LC-MS however GC-MS is also needed for some compounds. In this work we used GC-MS-MS with a modi�ed QuEChERS extraction and cleanup to rapidly measure pesticides in hops and cannabis with high performance.

MethodTest portions of dried hops or cannabis �ower were homogenized by grinding and extracted using QuEChERS extraction with SPE cleanup. Detection was carried out by GC-MS and GC-MS-MS using a GCMS-TQ8040 triple quadrupole mass spectrometer. Pesticide recovery was

determined using spiking experiments and calibration curves were prepared using matrix matched standards. All cannabis analysis was carried out in state-certi�ed testing labs with proper licenses in force.

Figure 1 Modi�ed QuEChERS Extraction

Grind weighed test portion (1.5 g)

SPEX GenoGrinder,2 metal balls, 5 minat 1500 rpm

Hydrate 15 mLwater, shake 30 min

15 mL 1%acetic acid in ACNshake 30 min

Add AOACQuEChERS saltsvortex 2 mincentrifuge

Clean up aliquotof supernatantby dSPE

Table 1 Method conditions

Inj. Temp : 250 ºC

Inj. Mode : Splitless (High Press. Inj. 250 kPa, 1.5 min)

GC Column : Rxi-5MS (30 m x 0.25 mm ID, df = 0.25µm)

with Rxi-Guard (5 m x 0.25 mm ID)

GC Oven Temp : 105 ºC (3 min), 10 ˚C/min to 130,

4 ˚C/min to 200, 8 ˚C/min to 290 (6 min)

Flow Mode : Constant linear velocity (44.1 cm/sec)

Interface Temp : 280 ºC

Ion Source Temp : 230 ºC

3

Pesticide analysis in Hops and Cannabis by GC-MS-MS

Figure 2 Various sample cleanup techniques investigated. Z-Sep+ is silica-zirconia-C18 material.

Z-Sep+ (silica-zirconia-C18)

QuEChERS followed byhexane layer appliedpost salting

C18, PSA, GCB

C18, Z-Sep+ PSA, GCB, Z-Sep+

PSA, C18,GCB

Stan

dard

dSP

E

ZSep

+ dSP

E

Verd

e dSP

E

C18 SP

E (pa

ss th

roug

h)

PSA SP

E (pa

ss th

roug

h)

Combo

SPE

Figure 3 Representative GCMS chromatogram of pesticides spiked into cannabis matrix blank

5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

(x100,000)

170417.G3-DEVx05_L3_VwH_A21_39

4

Pesticide analysis in Hops and Cannabis by GC-MS-MS

Results and DiscussionQuEChERS extraction with dispersive SPE cleanup were tested to determine which provided the best combination of pesticide recovery and cleanup for dried �ower cannabis samples. A wide variety of dSPE sorbent brands were tested and the best results were obtained with Sigma Verde dSPE, which removed a large amount cannabinoids as shown in Figure 7. Hexane treatment resulted in removal of a large portion of the early eluting background, which we determined contained low molecular weight compounds such as terpenes. The recoveries of some pesticides were slightly reduced in the hexane treatment, therefore we plan to test different volumes of hexane

treatment to �nd an optimum amount. Matrix matched calibration curves were linear within the quantitation limits established for each compound, which was compound dependent, but ranged from as low as 20 ppb or lower to greater than 500 ppb for a few substances. Detection limits and quantitation limits were required to have 3:1 and 10:1 signal to noise respectively. Recovery was compound dependent however the majority were within the range of 70-120% while outliers above and below the range were observed. In a random sampling of 15 dried �ower cannabis samples offered for retail sale, pesticides were detected in three samples as shown below.

Figure 4 Representative calibration curves.

0 500 1000 1500 2000 2500 3000 3500 Conc.0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

Area(x100,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Area(x100,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.0

0.5

1.0

1.5

2.0

2.5Area(x1,000,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.00

0.25

0.50

0.75

1.00

1.25

1.50

Area(x1,000,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.00

0.25

0.50

0.75

1.00

1.25

1.50Area(x100,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.0

0.5

1.0

1.5

2.0

2.5

Area(x1,000,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.0

1.0

2.0

3.0

4.0

5.0

Area(x1,000,000)

0 500 1000 1500 2000 2500 3000 3500 Conc.0.0

2.5

5.0

7.5

Area(x100,000)

0 2500 5000 7500 Conc.0.0

1.0

2.0

3.0

4.0

5.0

6.0

Area(x100,000)

Naledr2 > 0.998

Quintozener2 > 0.999

Spiroxaminer2 > 0.999

MGK-264r2 > 0.999

Captanr2 > 0.998

Imazalilr2 > 0.999

Myclobutanilr2 > 0.999

Cis-Permethrinr2 > 0.996

Cy�uthrinr2 > 0.996

5

Pesticide analysis in Hops and Cannabis by GC-MS-MS

Figure 5 Chromatogram of cannabis matrix blank with hexane treatment (black) and without (blue) prepared using standard dSPE.

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

(x10,000,000)

3 mL hexane layer reduced low MWbackground considerably but no changeto THC/cannabinoids

Hexane, Verde dSPE

No hexane, Verde dSPE

GCMS Scan (TIC) of matrix blank

6

Pesticide analysis in Hops and Cannabis by GC-MS-MS

Figure 6 (Left) Table of recoveries for selected cleanup techniques. (Right) Images of various cleanups in progress. (Below) Dirty GCMS liner.

Standard dSPE (Restek C18, PSA, GCB)

Verde dSPE(Sigma PSA, GCB, Z-Sep+)

Z-Sep+(Sigma Z-Sep+)

Hexane layer, 3 mL post-salt

Combo cartridges (Agela Cleanert TPT)

Contaminated liner

Recovery Std SPE Std SPE withHexane

Verde only Verde withHexane

LOQ

Chlofentezine deg.

Dichlorvos

Propoxur

Ethoprophos

Naled

Dimethoate

Carbofuran

Quintozene

Diazinon

Parathion-methyl

Spiroxamine-1

Carbaryl

Metalaxyl (Mefenoxam)

Spiroxamine-2

Methiocarb

Malathion

Chlorpyrifos

MGK 264-1

Thiamethoxam deg.

MGK 264-2

Captan

Fipronil

Paclobutrazol

Fludioxonil

Myclobutanil

Kresoxim-methyl

Chlorfenapyr

Propiconazole-1

Propiconazole-2

Tri�oxystrobin

Tebuconazole

Spiromesifen

Acetamiprid

Phosmet

Fenoxycarb

Bifenthrin

Bifenazate

Chlorantraniliprole

Etoxazole

cis-Permethrine

Pyridaben

trans-Permethrine

Cy�uthrin-1

Cypermethrin-1

Boscalid

Etofenprox

Azoxystrobin

104.1%

92.6%

94.0%

88.1%

56.7%

89.6%

95.2%

78.3%

82.5%

84.8%

57.8%

89.9%

85.8%

57.5%

90.3%

89.5%

79.0%

80.4%

84.3%

81.1%

107.8%

87.1%

87.3%

81.8%

NA

85.9%

83.8%

76.3%

40.8%

89.9%

83.0%

113.1%

87.6%

NA

83.9%

76.9%

NA

75.9%

80.6%

81.9%

73.2%

75.4%

84.8%

82.5%

85.6%

71.3%

90.8%

76.6%

91.1%

103.4%

80.8%

85.1%

89.1%

112.1%

49.4%

70.1%

77.5%

55.9%

102.4%

86.1%

54.2%

105.0%

87.4%

63.7%

69.9%

80.2%

75.0%

87.7%

84.0%

89.5%

77.2%

NA

84.5%

79.3%

76.9%

76.4%

83.7%

80.6%

78.5%

82.5%

94.0%

81.8%

47.0%

NA

71.2%

62.1%

88.6%

63.6%

56.3%

81.7%

78.0%

82.9%

54.2%

88.0%

90.0%

42.0%

91.8%

76.4%

7.7%

80.1%

102.0%

75.1%

76.3%

81.1%

7.3%

106.6%

77.2%

8.2%

101.1%

81.5%

74.6%

77.5%

79.2%

77.3%

96.6%

78.1%

71.5%

74.0%

76.3%

79.8%

75.6%

63.2%

70.1%

83.5%

59.0%

76.0%

77.1%

82.6%

79.7%

72.7%

68.4%

68.5%

70.8%

62.0%

67.6%

69.5%

79.8%

76.5%

77.9%

69.1%

81.1%

44.5%

59.8%

91.5%

76.1%

22.7%

82.8%

93.8%

44.6%

69.6%

79.5%

6.4%

73.3%

73.3%

7.7%

76.7%

86.7%

62.0%

66.9%

78.9%

71.7%

72.0%

84.3%

75.5%

77.8%

77.5%

82.4%

78.1%

61.0%

68.5%

84.2%

64.5%

76.1%

91.2%

112.5%

87.3%

46.2%

71.5%

71.2%

64.2%

70.9%

61.0%

54.9%

77.3%

72.0%

80.4%

50.9%

86.7%

100

20

50

50

50

20

100

20

50

20

50

50

50

20

50

20

20

20

50

20

200

20

20

20

20

20

20

100

50

40

20

50

500

500

20

100

40

20

100

200

50

50

100

100

20

20

50

Pesticide analysis in Hops and Cannabis by GC-MS-MS

7

Figure 7 Chromatogram of cannabis matrix blank with standard dSPE and Verde dSPE. The Verde dSPE reduced the cannabinoid level signi�cantly and returned proper retention times to the compounds.

Figure 8 (Left) Chromatogram of hops matrix blank with standard and Verde dSPE. No hexane was used in either sample. (Right) Table of results of pesticide analysis in randomly selected cannabis samples.

Dramatic decrease in THC/cannabinoid amount with Verde dSPE. Proper retention time restored.

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

(x10,000,000)

29.50 29.75 30.00 30.25 30.50 30.75 31.00 31.25 31.50 31.75 32.00 32.25 32.50 32.75 33.00 33.25

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5(x10,000,000)

GCMS Scan (TIC) of matrix blank

Hexane, Verde dSPE

Hexane, Standard dSPE

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6(x100,000,000)

123 ng/g cy�uthrin, 99 ng/g cypermethrin

139 ng/g chlorfenapyr1064 ng/g chlorfenapyr

Verde dSPE also improved background in hops

Verde dSPE

Standard dSPE

Sample

BA01

BA02

BA03

BA04

BA05

BA06

BA07

BA08

BA09

BA10

BA11

BA12

BA13

BA14

BA15

Pesticides detected

First Edition: June, 2017

© Shimadzu Corporation, 2017

Pesticide analysis in Hops and Cannabis by GC-MS-MS

A method for detection of chemical residues in dried cannabis �ower samples by GC-MS-MS was developed. Our method can detect low levels of common pesticides in samples offered for retail sale with excellent selectivity and speed. Measurements of a larger selection of commercially available cannabis samples are being carried out.

Conclusion