TO DOWNLOAD A COPY OF THIS POSTER, VISIT WWW.WATERS.COM/POSTERS ©2016 Waters Corporation

INTRODUCTION

Cannabinoids, such as tetrahydrocannabinol (THC), have gained considerable attention over the past decade for use in treating various conditions. Depending on how it is derived and handled, THC exists in many isomeric forms.1 Four (trans) isomers are predominant including (+)-Δ8-THC, (‑)-Δ8-THC, (+)-Δ9-THC, and (-)-Δ9-THC (structures in Figure 1). The major naturally-occurring isomer is (-)-Δ9-THC; considered the most therapeutically active cannabinoid present in cannabis.1 Under acidic conditions, cannabidiol (CBD) can convert to Δ9

-THC and other THC isomers.2,3 The FDA requires that stereoisomeric composition be quantified for active chiral

pharmaceutical compounds. Consumable products also need to be monitored for mixtures of positional and stereoisomers that can form resulting in changes in potency, pharmacological activity, or toxicity.3 Chiral analysis of THC is also applicable in forensic drug profiling.4 To that end, the separation of Δ8-THC, Δ9-THC and their stereoisomers, was investigated using Waters ACQUITY UPC2 technology and Trefoil chiral columns. After optimization, calibration and repeatability was performed for each isomer. The method was then used to determine the THC composition of a commercially available CBD product, before and after undergoing acidic conversion to THC.

THE SEPARATION OF ∆8-THC, ∆9-THC, AND THEIR STEREOISOMERS BY UPC2 USING TREFOIL CHIRAL COLUMNS Jacquelyn Runco, Andrew Aubin, Catharine Layton

Waters Corporation, Pittsburgh, PA, USA

CONCLUSION

The separation of (±)Δ8-THC and (±)Δ9-THC and

their stereoisomers was accomplished in less than 3

minutes on the Waters ACQUITY UPC2 system, using

the Trefoil AMY1 column at 15% ethanol mobile

phase conditions.

The THC isomers were well separated on all three

Waters Trefoil chiral stationary phases. The

columns also exhibited different selectivity, which is beneficial when resolving these THC isomers from

matrix interferences.

The ACQUITY UPC2 separation of the four THC

isomers showed excellent repeatability and the

calibration curves showed good linearity. As a result, the methodology can be used for

quantitative analysis of THC in cannabis products.

METHODS

Standards (Obtained from Cerilliant, exempt standards)

(±)Δ8-THC at 0.1 mg/ml in heptanes

(±)Δ9-THC at 0.1 mg/ml in heptanes

(-)Δ8-THC at 1 mg/ml in methanol

(-)Δ9-THC at 1 mg/ml in methanol

A 50:50 mix of the (±)Δ8-THC and (±)Δ9-THC standards was

used for method development.

The (-)Δ8-THC and (-)Δ9-THC standards were diluted 1:10 in

200 proof ethanol and were used to determine peak order.

Serial dilutions of the (±)Δ8-THC and (±)Δ9-THC standards in 200 proof ethanol were used for calibration.

CBD Acid Conversion A commercial CBD oil extract was used as an example product. Three aliquots of the CBD oil were treated using the

following conditions: Sample 1: 15.65 mg of CBD Oil in 3mL 200-proof ethanol,

heated overnight at 55°C Sample 2: 15.26 mg of CBD Oil in 3mL 0.1M HCl in 200-

proof ethanol, heated overnight at 55°C Sample 3: 15.40 mg of CBD Oil in 3mL 200-proof ethanol,

room temperature

All three samples were filtered and diluted 1:10 in ethanol

before injection on the UPC2

References

1) S. Levin, S. Abu-Lafi, J. Zahalka, and R. Mechoulam, Resolution of chiral cannabinoids on amylose tris(3,5-dimethylphenylcarbamate) chiral stationary phase: effects of structural features and mobile phase additives, Journal of Chromatography A, 654 (1993) 53-64

2) Y. Gaoni, R. Mechoulam, The isomerization of cannabidiol to tetrahydrocannabinols,

Tetrahedron, 22 (1966) 1481-1488

3) C. Layton, J. Runco, A. Aubin, Forced degradation of cannabidiol, Application Note, August 2016, 720005766EN

4) L. Ambach, F. Penitschka, A. Broillet, S. Konig, W. Weinmann, W. Bernhard,

Simultaneous quantification of delta-9-THC, THC-acid A, CBN and CBD in seized drugs using HPLC-DAD, Forensic Science International, 243 (2014) 107-111

RESULTS & DISCUSSION

Initially, the three Trefoil columns were screened, which are amylose

(AMY1) and cellulose (CEL1 & CEL2) based chiral stationary phases

that have a wide range of applicability. The results of the screen (Figure 2) showed that all three columns provided good separation of

the THC isomers. Different elution order and selectivity were observed, which can be beneficial when separating these compounds

from matrix interferences. Even though all three separations could be optimized, the separation obtained on the AMY1 column was selected

due to higher retention and resolution than on the two CEL columns.

Figure 1: Positional and stereo isomeric structures of THC

Figure 2:Screening of the four THC isomers on the three Trefoil

chiral columns using a 2-20% gradient over 5 minutes

Figure 3: Isocratic separations of the four THC isomers on the AMY1

column at 10% co-solvent (top) and 15% co-solvent (bottom) condi-tions

Figure 4: Calibration curves for the four THC isomers on the

AMY1 column at 10% isocratic conditions. The calibration curve concentrations were 0.003125, 0.00625, 0.0125, 0.025

and 0.05 mg/mL, and the injection volume was 1 µL

Time-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

5.0e-3

1.0e-2

1.5e-2

2.0e-2

-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

2.5e-3

5.0e-3

7.5e-3

1.0e-2

1.25e-2

-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

2.0e-3

4.0e-3

6.0e-3

4.574.15

0.32

0.00 0.34

2.16

4.34

5.04

4.93 5.50

3.443.20

3.09

3.01

0.34

0.00

0.355.72

3.68

3.022.722.60

2.39

0.000.33

3.653.26 4.98 5.50

AMY1

CEL1

CEL2

(-)Δ

8

(-)Δ

9

(+)Δ

8

(+)Δ

9

(-)Δ

8

(-)Δ

9

(+)Δ

8

(+)Δ

9

(-)Δ

8

(-)Δ

9

(+)Δ

8

(+)Δ

9

Time-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

2.0e-3

4.0e-3

6.0e-3

-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

-1.0e-3

0.0

1.0e-3

2.0e-3

3.0e-3

4.0e-3

5.0e-3

6.0e-3

0.32

0.00

0.343.21

0.401.80

4.113.555.55

0.33

0.31

0.00

0.34

1.71

0.39

0.491.15

2.111.84

2.89

(-)Δ

8

(-)Δ

9

(+)Δ

8

(+)Δ

9

15% EtOH

10% EtOH

y = 11910x + 0.9861R² = 1

y = 10660x - 1.5417R² = 0.9988

y = 13137x - 14R² = 0.9993

y = 14548x - 9.2639R² = 0.9999

0

100

200

300

400

500

600

700

800

0 0.01 0.02 0.03 0.04 0.05 0.06

(-)D8-THC

(+)D8-THC

(+)D9-THC

(-)D9-THC

Linear ((-)D8-THC)

Linear ((+)D8-THC)

Linear ((+)D9-THC)

Linear ((-)D9-THC)

O

CH3

H

HOH

CH3

CH3

CH3

O

CH3

H

HOH

CH3

CH3

CH3

O

CH3

H

HOH

CH3

CH3

CH3

O

CH3

H

HOH

CH3

CH3

CH3

(+)-Δ8-THC (-)-Δ8-THC

(+)-Δ9-THC (-)-Δ9-THC

Time-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

5.0e-3

1.0e-2

1.5e-2

-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

2.0e-3

4.0e-3

-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

1.0e-2

2.0e-2

-0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00

AU

0.0

2.0e-3

4.0e-3

0.32

0.00

0.343.20

1.350.41

1.71 2.79

3.54 4.095.52

6.00

2.98

0.32

0.00 2.23 2.61

5.53553

3.2213

5.5242

(2)

(3)

(4)

(1)

(-)Δ

8

(-)Δ

9

CBD

-Δ9-THC (untreated sample)

-Δ9-THC (acid sample)

-Δ8-THC (acid sample)

Figure 5: UPC2 chromatograms showing (1) separation of

the THC isomer standards, (2) analysis of CBD oil Sample 1 (heat only), (3) analysis of CBD oil Sample 2 (heat + acid),

and (4) analysis of CBD oil Sample 3 (control). THC isomers are as indicated. The separations were achieved at 2 mL/min

and 10% ethanol on the AMY1 column using 1µL injections.

In order to optimize the separation, the co-solvent percentage at elu-

tion was determined from the screening separation. With a gradient delay of 0.34 min, gradient slope of 3.6%/min, and 2% starting per-

centage, the co-solvent percentage at elution of the first peak at 4.15 minutes was calculated using the following equation:

%Co-solvent at Elution =

(retention time – gradient delay) x gradient slope + starting % (4.15 – 0.34 min ) x 3.6%/min + 2%

15.7%

Figure 3 shows separation of the THC isomers at 15% and 10% etha-nol conditions. At 15%, the separation of the positional and stereo

isomers of THC is achievable in less than 3 minutes. Comparatively, the liquid chromatography separation of these isomers is 23 minutes

and does not achieve baseline resolution.

In this case, the separation would be used to analyze THC content in a sample containing cannabidiol (CBD). As a result, the isocratic

methods were investigated to separate the THC isomers from CBD. The 10% method provided acceptable resolution and was therefore

used for calibration and repeatability. Figure 4 shows calibration curves for the four THC isomers on the AMY1 column using the 10%

isocratic method.

UPC2 screening conditions System: ACQUITY UPC2 system with an ACQUITY PDA detector Columns: (3 mm X 150 mm, 2.5 µm)

ACQUITY UPC2 Trefoil AMY1 Column (AMY1) ACQUITY UPC2 Trefoil CEL1 Column (CEL1)

ACQUITY UPC2 Trefoil CEL2 Column (CEL2) Mobile Phase A: Carbon Dioxide

Mobile Phase B: 200-proof ethanol

Gradient: 2 to 20% B over 5 minutes Column temperature: 50°C

Injection volume: 1 µL Flow rate: 2 mL/min

ABPR: 2000 psi PDA absorbance: 228nm

Compensation reference: 500-600nm Optimized conditions: noted on figures

Repeatability was determined using the 0.025 mg/mL sam-

ple. The calibration curves were linear (R2 > 0.998), and the peak areas showed good reproducibility, with area count RSD

values less than 2% (n=7).

THC content is of particular interest for purposes of product

quality control. Under acidic conditions, CBD can convert to

Δ9-THC and other THC isomers.3 Once the method was cali-

brated, and shown to be reproducible, the separation was ap-

plied to the analysis of the three CBD oil extract samples

(Figure 5).

Results of the analysis showed a very small amount of THC in

the initial product and no change in the heated product. Un-

der acidic conditions, a significant amount of (-)Δ9-THC was

detected, calculated to be 1.16mg or approximately 7.6% of

the initial 15.26 mg sample, along with a very small detect-

able amount of (-)Δ8-THC.