pubs.acs.

org/ac Article

Raman-Based Differentiation of Hemp, Cannabidiol-Rich Hemp, and

Cannabis
Lee Sanchez, David Baltensperger, and Dmitry Kurouski*
Cite This: Anal. Chem. 2020, 92, 7733−7737 Read Online

ACCESS Metrics & More Article Recommendations *

sı Supporting Information
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: Hemp (Cannabis sativa) has been used to treat pain as far back as 2900 B.C. Its
pharmacological effects originate from a large variety of cannabinols. Although more than 100
different cannabinoids have been isolated from Cannabis plants, clear physiological effects of
only a few of them have been determined, including delta-9 tetrahydrocannabinol (THC),
Downloaded via UNIV FED DE GOIAS on April 29, 2025 at 11:35:52 (UTC).

cannabidiol (CBD), and cannabigerol (CBG). While THC is an illicit drug, CBD and CBG are
legal substances that have a variety of unique pharmacological properties such as the reduction
of chronic pain, inflammation, anxiety, and depression. Over the past decade, substantial efforts
have been made to develop Cannabis varieties that would produce large amounts of CBD and
CBG. Ideally, such plant varieties should produce very little (below 0.3%) if any THC to make their cultivation legal. The amount of
cannabinoids in the plant material can be determined using high performance liquid chromatography (HPLC). This analysis,
however, is nonportable, destructive, and time and labor consuming. Our group recently proposed to use Raman spectroscopy (RS)
for confirmatory, noninvasive, and nondestructive differentiation between hemp and cannabis. The question to ask is whether RS can
be used to detect CBD and CBG in hemp, as well as enable confirmatory differentiation between hemp, cannabis, and CBD-rich
hemp. In this manuscript, we show that RS can be used to differentiate between cannabis, CBD-rich plants, and regular hemp. We
also report spectroscopic signatures of CBG, cannabigerolic acid (CBGA), THC, delta-9-tetrahydrocannabinolic acid (THCA),
CBD, and cannabidiolic acid (CBDA) that can be used for Raman-based quantitative diagnostics of these cannabinoids in plant
material.

O ver the past decade substantial efforts have been made to

develop CBD- and CBG-rich hemp varieties.1 After
plant harvesting, these compounds can be extracted and used
Scheme 1. Structures of CBG-A, CBG, CBD-A, CBD, THC-
A, and THC Cannabinols Present in Cannabis Plants

in a large number of commercial products ranging from oils

and gums to alcoholic beverages and sprays. CBG has a large
spectrum of pharmacological activity. It is known to kill or
decelerate bacterial growth, reduce inflammation, and inhibit
tumor cell growth.2,3 CBG has also been found to be
particularly effective in glaucoma treatment because it reduces
intraocular pressure.4
In Cannabis plants, CBGA can be converted to CBD and
THC, as well as the acid derivatives of these cannabinols
(cannabidiolic acid (CBDA) and THCA),5 Scheme 1. CBD
relieves chronic pain, lowers blood pressure, and reduces
inflammation, anxiety and depression. It can also alleviate side
effects related to cancer treatment, such as nausea and
vomiting.1 Lastly, CBD is capable of mitigating symptoms of
try.8−11 These sophisticated tests are destructive, time-
neurological disorders, such as reduction of muscle spasticity in consuming and can only be performed in certified laboratories.
people with multiple sclerosis.6,7 This drastically complicates hemp farming.11
THC is considered to be an illicit drug due to its well-
defined psychoactive properties. On a federal and state-specific
level, its trafficking and possession are considered a felony Received: February 24, 2020
offense that carries serious consequences such as prison time Accepted: May 13, 2020
and significant monetary fines. Therefore, hemp farmers are Published: May 13, 2020
constantly seeking hemp varieties that would produce high
concentrations of CBD and CBG and little if any THC. Such
analysis is primarily done by HPLC and mass spectrome-

© 2020 American Chemical Society https://dx.doi.org/10.1021/acs.analchem.0c00828

7733 Anal. Chem. 2020, 92, 7733−7737
Analytical Chemistry pubs.acs.org/ac Article

Recently, our group demonstrated that cannabis/hemp Certificates of analyses are provided in the Supporting
diagnostics can be done by noninvasive and nondestructive Information.
spectroscopic analysis of fresh plant material.12 Using a hand- Raman Spectroscopy. Raman spectra from plants were
held Raman spectrometer, we were able to distinguish hemp taken with a hand-held Resolve Agilent spectrometer equipped
from cannabis with 100% accuracy, as well as differentiate with 830 nm laser source (beam diameter ∼2 mm). Resolve
three different cannabis varieties with on average 97% spectral resolution was 15 cm−1. Samples were brought in a
accuracy. Raman spectroscopy (RS) is based on inelastic direct contact with the spectrometer for spectral acquisition.
light scattering of photons, which excite molecules in the The following experimental parameters were used for all
sample to higher vibrational or rotational states.9 After these collected spectra: 10s acquisition time, 495 mW power. The
inelastically scattered photons are collected by a spectrometer, spectra were automatically baselined by the instrument
the change in the photon energy is determined. Since the software. In total, 20−40 spectra were collected from each
change in the photon energy will directly depend on the sample type (hemp, CBD-rich hemp and cannabis). Spectra
vibrational properties of molecules in the sample, RS can be shown in the manuscript are raw baseline corrected, without
used to probe structure and composition of analyzed smoothing.
specimens.13 Raman spectra from CBG, CBGA, CBDA, THCA, and THC
The question to ask is whether RS can be used to distinguish oils deposited on microscope coverslips were collected on a
between hemp, cannabis, which is used in this paper to refer to home-built inverted microscope (Nikon TE-2000U) with 20×
hemp plants rich in THCA, as well as CBD-rich hemp. To dry Nikon objective (NA = 0.45). A single longitudinal diode
answer this question, we have collected Raman spectra from mode laser (Cobolt, Germany) was used to generate a 488 nm
four different CBD-rich hemp varieties. We compared these laser light that was directed to the microscope and reflected
spectroscopic signatures to the Raman spectra collected from through 50/50 beam splitter to the sample. The scattered light
cannabis and hemp reported in our previous study (Sanchez et was directed to a confocal IsoPlane SCT 320 Raman
al., RSC Advances, 2020)12. spectrometer (Princeton Instruments, NJ, U.S.A.) equipped

■ EXPERIMENTAL SECTION
Materials. Cannabinol standards CBG, CBGA, CBDA,
with a 1200 groove/mm grating blazed at 500 nm. Prior to
entering the spectrograph, Rayleigh scattering was filtered with
a LP02-488RE-25 long-pass filter (Semrock, NY, U.S.A.). The
THCA, and THC were received as a gift from CannID spectrograph dispersed light was then sent to PIXIS:400BR
(Austin, TX). These standards were originally manufactured CCD (Princeton Instruments, NJ, USA). A motorized stage
by Cerilliant (Round Rock, TX). Chemical structures of these H117P2TE (Prior, MA, U.S.A.) controlled by Prior Proscan II
compounds are shown in the Scheme 1. Solvents were was used to move the sample relative to the incident laser
evaporated from these standards by leaving uncapped vials for beam. Raman spectrum from CBD isolate was collected using a
several hours in a fume hood. Next, residual cannabinol oils hand-held Resolve Agilent spectrometer. All data were
were redissolved in ethanol and deposited onto clean processed using GRAMS/AI 7.0 (Thermo Galactic, NH,
microscope coverslips and dried under room temperature. U.S.A.).
CBD isolate (Scheme 1) was received from Texas Farm & Multivariate Data Analysis. SIMCA 14 (Umetrics, Umeå,
Process, LLC. Sweden) was utilized for statistical data treatment of the
Plants. Hemp plants with the following amounts of THCA, Raman spectra collected in this study. Imported spectra were
CBDA, CBD, CBGA, and CBG were used: (1) “T5−005”, truncated so as to only include wavenumbers 701−1700 cm−1
THCA: 0.09%, THC: 0%, CBDA: 1.68%, CBD: 0.64%, and scaled to unit variance via standard normal variate (SNV)
CBGA: 0.1%, CBG: 0.02%; (2) “Trump sauce-006 correction in order to give all spectral regions equal
(TS006)”, THCA: 0.1%, THC: 0%, CBDA: 2.27%, CBD: importance. Spectra were then normalized by the total area
0.67%, CBGA: 0.05%, CBG: 0%; (3) “T5-Joey-008”, THCA: followed by a first derivative application. classes. Next, with
0.13%, THC: 0%, CBDA: 2.48%, CBD: 0.34%, CBGA: 0.15%, orthogonal partial least-squares discriminant analysis (OPLS-
CBG: 0%; and (4) “Hawaii haze-010”, THCA: 0.13%, THC: DA), we determined the number of predicting and orthogonal
0%, CBDA: 2.15%, CBD: 0.25%, CBG: 0.03%; CBGA: 0.18%. significant components and identified spectral regions that best
explain the separation between the classes.

■
These hemp varieties have been grown on a plant farm located
in Delta, CO. Spectra were collected from 10 to 15 buds of 2−
3 fresh plants of each variety. RESULTS AND DISCUSSION
Hemp and cannabis plants were grown at Evergreen Raman spectra of CBD-rich hemp (for clarity, spectra of only
Enterprises LLC located in Denver, CO. Cannabis variety T5-005 and TS006 are shown in Figure 1) exhibited
known as “twisted sherbert” (TS) contained THCA: 4.05% vibrational bands that can be assigned to cellulose, xylan,
THCA, THC: 0.04%, CBDA: 0%, CBD: 0%, CBGA: 0.23%, carotenoids, and lignin, Figure 1 and Table 1.14 These
CBG: 0%. Hemp contained: THCA: 0% THCA, THC: 0%, vibrational bands were also present in the spectra of hemp
CBDA: 1.08%, CBD: 0.07%, CBGA: 0%, CBG: 0%.12 Hemp and cannabis (TS). We have found substantial difference in the
spectra were collected from buds of 5−10 fresh plants; intensities of vibrational bands of cellulose and carotenoids in
cannabis spectra were collected from buds of 10−15 fresh- these three groups of plants. Specifically, T5-005 and TS006
frozen plants. Fresh-freezing of plants was performed by exhibited the highest intensity of a vibrational band at 1525
placement of plant buds into freezer at −10 to −15 °C. Fresh- cm−1, which can be assigned to carotenoid. At the same time,
freezing is a standard procedure in cannabis farming that is intensity of this band is lower in the spectrum of hemp and
used to preserve cannabinol content of plants during their cannabis. Similar changes have been observed for vibrational
postharvest processing. Based on visual examination, fresh- bands at 1155−1228 cm−1, which can be assigned to cellulose
freezing does not result in any noticeable changes in plant and xylan. These spectral changes suggest a substantial
appearance or texture. difference in scaffold molecules in hemp, cannabis, and
7734 https://dx.doi.org/10.1021/acs.analchem.0c00828
Anal. Chem. 2020, 92, 7733−7737
Analytical Chemistry pubs.acs.org/ac Article

Figure 1. Top: Raman spectra collected form T5−005 (purple) and

TS006 (blue), hemp (green) and TS (red). Bottom: Raman spectra of
THCA (black) and CBDA (orange). Spectra normalized on CH2
vibrations (1440 cm−1) that are present in nearly all classes in
biological molecules (marked by asterisks (*)).

Table 1. Vibrational Bands and Their Assignments for

Hemp, Cannabis Species, and THCA Figure 2. Raman spectra of THC (red), THCA (maroon), CBD
band vibrational mode assignment (black), CBDA (green), CBG (blue), and CBGA (violet).
780 TBA cannabinoids12
Table 2. Vibrational Bands Observed in the Raman Spectra
835 TBA cannabinoids12
of THC, THCA, CBD, CBDA, CBG, and CBGA
916 ν(C−O−C) in plane, symmetric cellulose, lignin15
993−1000 ν3 (C−CH3 stretching) and carotenoids, compound vibrational bands, cm−1
phenylalanine protein16,17
THC 733, 775, 780, 835, 891, 908, 950, 1011, 1037, 1080, 1114, 1129,
1084 ν(C−O)+ν(C−C)+δ(C−O−H) carbohydrates18 1185, 1236, 1255, 1276, 1295, 1321, 1365, 1372, 1440, 1462, 1570,
1114 νsym(C−O−C), C−OH bending cellulose19,20 1600, 1623, 1666
1155 νasym(C−C) ring breathing carbohydrates, THCA 715, 733, 775, 780, 835, 891, 908, 950, 1011, 1037, 1080, 1114, 1129,
cellulose15 1185, 1236, 1255, 1276, 1295, 1321, 1365, 1372, 1440, 1462, 1570,
1600, 1623, 1666
1185 ν(C−O−H) next to aromatic xylan21,22
ring+σ(CH) CBD 775, 802, 865, 901, 924, 965, 985, 1012, 1080, 1104, 1136, 1150,
1176, 1230, 1262, 1274, 1302, 1340, 1370, 1437, 1451, 1515, 1585,
1212−1228 δ(C−C−H) aliphatic,23 xylan21 1623, 1643, 1663
1267 C−O stretching (aromatic) lignin24 CBDA 775, 802, 843, 901, 1176, 1262, 1307, 1370, 1437, 1451, 1585, 1602,
1285 δ(C−C−H) aliphatic23 1623, 1643, 1663
1295 TBA cannabinoids12 CBG 725, 745, 768, 778, 800, 810, 835, 860, 879, 902, 918, 968, 988, 998,
1321 δCH2 bending vibration cellulose, lignin15 1033, 1060, 1100, 1114, 1129, 1161, 1182, 1219, 1275, 1295, 1309,
1328, 1345, 1365, 1381, 1440, 1452, 1516, 1587, 1623, 1627, 1670
1376 δCH2 bending vibration aliphatic23
CBGA 759, 810, 860, 879, 923, 998, 1080, 1114, 1172, 1287, 1295, 1381,
1440 δ(CH2) + δ(CH3) aliphatic23 1440, 1452, 1500, 1623, 1670
1455 δCH2 bending vibration aliphatic23
1525 CC (in plane) carotenoids25,26 Table 3. Accuracy of Classification by OPLS-DA for
1610 ν(C−C) aromatic ring +σ(CH) lignin27,28 Cannabis, Industrial, and CBD-Rich Hemp
1623−1666 aromatic cannabinoids12,29
CBD-risch
members correct hemp cannabis hemp
CBD-rich plants. Such changes can be used to identify plant hemp 22 100% 22 0 0
varieties or even predict geographical origin of their growth. cannabis 64 100% 0 64 0
We have also observed substantial differences in the CBD-rich 126 100% 0 0 126
vibrational bands that can be assigned to THCA in hemp, hemp
cannabis, and CBD-rich plants. As was previously discussed, Fisher’s prob. 9.3 × 10−12
TS exhibits vibrational bands at 780, 1295, 1623, and 1666
cm−1, which can be assigned to THCA. We have not observed spectra of T5−005 and TS006 relative to the intensities of
vibrational bands at 780 and 1623 cm−1 in the spectra of T5- these bands in the spectra of hemp. This spectral difference can
005 and TS006, whereas the intensity of 1666 cm−1 band was be used to enable confirmatory differentiation between
nearly identical as in the spectrum of TS. Thus, this spectral industrial hemp and CBD-rich hemp.
difference can be used to distinguish between CBD-rich hemp Although spectroscopic analysis of plant material reveals
and cannabis. At the same time, we have observed an increased striking differences between cannabis and CBD-rich hemp,
intensity of vibrational bands at 1295−1300 cm−1 in the Raman signatures of THCA and CBDA are very similar.
7735 https://dx.doi.org/10.1021/acs.analchem.0c00828
Anal. Chem. 2020, 92, 7733−7737
Analytical Chemistry pubs.acs.org/ac Article

identical Raman spectra, Figure 2. These results suggest that

both THCA and THC can be detected in the plant using RS.
We have also observed high similarity between spectra
collected from CBDA and CBD. Both spectra exhibited
vibrational bands at 775, 802, 901, 1176, 1370, 1437, 1585,
1623, 1643, and 1663 cm−1. However, CBD had peaks at 865,
985, 1080, 1104, 1136, 1150, 1274, and 1515 cm−1 that were
not observed in the spectra of CBDA. At the same time, CBDA
has peaks at 843, 1262, and 1602 cm−1 that were not observed
in the spectrum of CBD. We have also observed that a peak
centered at 1307 in the spectrum of CBD was 5 cm−1 blue-
shifted in the spectrum of CBDA.
Figure 3. Loading plot of two predictive components (component 1 Interestingly, CBGA and CBG exhibited greater spectro-
(blue) and component 2 (orange)) in the Raman spectra of cannabis, scopic difference relative to those observed between THCA
hemp, and CBD-rich hemp. and THC, as well as between CBDA and CBD, Figure 2.
Although nearly all vibrational bands were present in both
CBGA and CBG spectra, their relative intensities were
drastically different, Figure 2 and Table 2. For instance,
vibrational bands at 1287−1295 cm−1 were found to be the
most intense in the spectrum of CBGA. However, the band
centered at 1129 cm−1 were the most intense in the spectrum
of CBG. It should be noted that this band was not present in
the spectrum of CBGA.
Although hemp plants typically have very small quantities of
CBG/CBGA (“T5-005”; CBGA: 0.1%, CBG: 0.02%; ‘TS006′,
CBGA: 0.05%, CBG: 0%), our results demonstrate that RS can
be used for identification of this cannabinol in CBG/CBGA-
rich hemp. Specifically, we found that bands at 725−768, 879,
998, and 1060 cm−1 are unique for CBG/CBGA and are not
present in the Raman spectra of THC/THCA and CBD/
CBDA.
To further prove our expectation that RS can be used for
Figure 4. OPLS-DA variant component plot of Raman spectra highly accurate differentiation between hemp, CBD-rich hemp,
collected from hemp (green), cannabis (blue), and CBD-rich hemp
and cannabis, we used OPLS-DA analysis. The final model,
(red).
containing one predictive component, 2 orthogonal compo-
nents, and 1001 (701−1700 cm−1) out of 1651 original
THCA and CBDA exhibit very similar structures (Scheme 1), wavenumbers, was used to generate the misclassification table
consequently reflected in the profiles of vibrational bands in (Table 3) and the loadings plot (Figure 3). It should be noted
1570−1691 and 1212−1366 cm−1 spectral regions. At the that for the reported OPLS-DA model, in addition to the
same time, we found that THCA has a vibrational band Raman spectra of TS, we have used spectra collected from two
centered at 1295 cm−1, which has been found to be 5 cm−1 other cannabis varieties known as “triple chocolate chip
red-shifted in the spectrum of CBDA. Also, Raman spectrum (TCC)” and “gelato cake (GC)”. Detailed spectroscopic
of THCA exhibits a vibrational band at 1185 cm−1, which was analysis of these cannabis varieties, as well as their cannabinoid
not evident in the spectrum of CBDA. Raman spectrum of profiles have been previously reported by our group study
THCA has a doublet at 775 and 780 cm−1, whereas only one (Sanchez et al., RCS Advances, 2020). The spectra of GC and
vibrational band at 775 cm−1 has been detected in the TCC have been included in the reported model to increase the
spectrum of CBDA. The spectroscopic analysis of vibrational number of Raman spectra of cannabis.
bands of THCA and CBDA suggests that only the band at 780 The first predictive component (PC; Figure 3) explains 92%
cm−1 can be used for confirmatory differentiation between of the variation between classes. Absolute intensities in the
THCA and CBDA. One can also envision that an intensity of loading spectrum are proportional to the percentage of the
the vibrational band at 1623 cm−1 is present in the spectrum of total variation between classes explained by each wavenumber.
cannabis (4.05% of THCA) and absent in the spectrum of The model identified the peak at 780 cm−1, which could be
CBD-rich hemp (T5−005; CBDA 1.68%, TS006; CBDA assigned to cannabinoids (Table 1), cellulose, and lignin peaks
2.27%) due to differences in the amount of THCA and CBDA at ∼916 cm−1 and the bands at 1260−1321 cm−1, which
in these two groups of plants. Alternatively, such difference in correspond to both THCA and cellulose. Also, the model
the intensity of 1623 cm−1 can be attributed to the difference identified peaks at 1440 cm−1, which could be assigned to
in the Raman cross-section of these THCA and CBDA. aliphatic vibrations and bands at 1623−1666 cm−1, which
Additional studies are required to fully elucidate Raman cross originate from cannabinoids, to be the strongest spectral
sections of THCA and CBDA to enable quantitative prediction markers of cannabis, which supports the conclusions of our
of their content in plants. qualitative spectral analysis above. This indicates that coupling
The question to ask is how different are the spectroscopic of OPLS-DA with RS allows for a 100% accurate differ-
signatures of CBDA and THCA from their decarboxylated entiation between cannabis, hemp and CBD-rich hemp, Tables
derivatives. We found that THCA and THC exhibit nearly 3 and 4.
7736 https://dx.doi.org/10.1021/acs.analchem.0c00828
Anal. Chem. 2020, 92, 7733−7737
Analytical Chemistry

■
pubs.acs.org/ac Article

CONCLUSIONS (3) Borrelli, F.; Fasolino, I.; Romano, B.; Capasso, R.; Maiello, F.;
Coppola, D.; Orlando, P.; Battista, G.; Pagano, E.; Di Marzo, V.; Izzo,
Our results clearly demonstrate that RS can be used for A. Biochem. Pharmacol. 2013, 85 (9), 1306−16.
confirmatory, noninvasive, and nondestructive differentiation (4) Nadolska, K.; Gos, R. Klin Oczna 2008, 110 (7−9), 314−7.
between hemp, CBD-rich hemp, and cannabis (THCA-rich (5) Havelka, J. What is CBG (cannabigerol) & what does this
hemp) with 100% accuracy. We also reported Raman spectra cannabinoid do? Leafly, 2017; https://www.leafly.com/news/
of six major cannabinoids, THC, THCA, CBD, CBDA, CBG, cannabis-101/what-is-cbg-cannabinoid, accessed March 2020.
and CBGA. Analysis of their vibrational fingerprint shows that (6) Mouslech, Z.; Valla, V. Neuro. Endocrinol. Lett. 2009, 30 (2),
RS can be used for identification of these chemicals in plant 153−79.
material. We also provided experimental evidence that allows (7) Darkovska-Serafimovska, M.; Serafimovska, T.; Arsova-
for differentiation between THC/THCA vs CBD/CBDA vs Sarafinovska, Z.; Stefanoski, S.; Keskovski, Z.; Balkanov, T. J. Pain
Res. 2018, 11, 837−842.
CBG/CBGA, as well as between CBD-CBDA and CBG- (8) Burnier, C.; Esseiva, P.; Roussel, C. Talanta 2019, 192, 135−
CBGA. These results substantially expand applicability of RS as 141.
a tool for hemp cultivation and breeding. (9) Zivovinovic, S.; Alder, R.; Allenspach, M. D.; Steuer, C. J. Anal.

■
*
ASSOCIATED CONTENT
sı Supporting Information
Sci. Technol. 2018, 9, 1−10.
(10) Patel, B.; Wene, D.; Fan, Z. T. J. Pharm. Biomed. Anal. 2017,
146, 15−23.
(11) Nie, B.; Henion, J.; Ryona, I. J. Am. Soc. Mass Spectrom. 2019,
The Supporting Information is available free of charge at 30 (5), 719−730.
https://pubs.acs.org/doi/10.1021/acs.analchem.0c00828. (12) Sanchez, L.; Filter, C.; Baltensperger, D.; Kurouski, D. RSC
Adv. 2020, 10, 3212−3216.
HPLC results of cannabis analyses that demonstrate (13) Farber, C.; Mahnke, M.; Sanchez, L.; Kurouski, D. TrAC,
cannabinoid content of analyzed plant material (PDF) Trends Anal. Chem. 2019, 118, 43−49.

■
(14) Sanchez, L.; Ermolenkov, A.; Tang, X. T.; Tamborindeguy, C.;
Kurouski, D. Planta 2020, 251, 64.
AUTHOR INFORMATION
(15) Edwards, H. G.; Farwell, D. W.; Webster, D. Spectrochim. Acta,
Corresponding Author Part A 1997, 53A (13), 2383−92.
Dmitry Kurouski − Department of Biochemistry and Biophysics (16) Tschirner, N.; Brose, K.; Schenderlein, M.; Zouni, A.;
and The Institute for Quantum Science and Engineering, Texas Schlodder, E.; Mroginski, M. A.; Hildebrandt, P.; Thomsen, C.
A&M University, College Station, Texas 77843, United States; Phys. Status Solidi B 2009, 246 (11−12), 2790−2793.
(17) Kurouski, D.; Van Duyne, R. P.; Lednev, I. Analyst 2015, 140
orcid.org/0000-0002-6040-4213; Phone: 979-458-3778;
(15), 4967−80.
Email: [email protected] (18) De Gussem, K.; Vandenabeele, P.; Verbeken, A.; Moens, L.
Spectrochim. Acta, Part A 2005, 61 (13−14), 2896−2908.
Authors
(19) Edwards, H. G. M.; Farwell, D. W.; Webster, D. Spectrochim.
Lee Sanchez − Department of Biochemistry and Biophysics, Acta, Part A 1997, 53 (13), 2383−2392.
Texas A&M University, College Station, Texas 77843, United (20) Agarwal, U. P.; Ralph, S. A. Appl. Spectrosc. 1997, 51 (11),
States 1648−1655.
David Baltensperger − Department of Soil and Crop Sciences, (21) Agarwal, U. Front. Plant Sci. 2014, 5, 1−12.
Texas A&M University, College Station, Texas 77843, United (22) Mary, Y. S.; Panicker, C. Y.; Varghese, H. Orient. J. Chem. 2012,
States 28 (2), 937−941.
(23) Yu, M. M.; Schulze, H. G.; Jetter, R.; Blades, M. W.; Turner, R.
Complete contact information is available at: Appl. Spectrosc. 2007, 61 (1), 32−7.
https://pubs.acs.org/10.1021/acs.analchem.0c00828 (24) Cao, Y.; Shen, D.; Lu, Y.; Huang, J. Ann. Bot. 2006, 97, 1091−
1094.
Notes (25) Devitt, G.; Howard, K.; Mudher, A.; Mahajan, S. ACS Chem.
The authors declare no competing financial interest. Neurosci. 2018, 9, 404.

■
(26) Adar, F. Spectroscopy 2017, 32 (6), 12−20.
(27) Kang, L.; Wang, K.; Li, X.; Zou, B. J. Phys. Chem. C 2016, 120
ACKNOWLEDGMENTS (27), 14758−14766.
We are grateful to Cree Crawford, Co-Founder and President (28) Agarwal, U. Planta 2006, 224 (5), 1141−53.
of Cann-ID, as well as to all members of this company for (29) Sivashanmugan, K.; Squire, K.; Tan, A.; Zhao, Y.; Kraai, J. A.;
providing cannabinoid standards. We are also trateful to Conor Rorrer, G. L.; Wang, A. ACS Sens. 2019, 4 (4), 1109−1117.
Filter and Haden Shibley for the provided access to help and
cannabis plants, as well as for providing CBD isolate. We
are grateful to AgriLife Research of Texas A&M for the
provided financial support. We also acknowledge Governor’s
University Research Initiative (GURI) grant program of Texas
A&M University, GURI Grant Agreement No. 12-2016,
M1700437.

■ REFERENCES
(1) Hartsel, J. A.; Eades, J.; Hickory, B.; Makriyannis, A.
Nutraceuticals 2016, 735−754.
(2) Appendino, G.; Gibbons, S.; Giana, A.; Pagani, A.; Grassi, G.;
Stavri, M.; Smith, E.; Rahman, M. M. J. Nat. Prod. 2008, 71 (8),
1427−30.

7737 https://dx.doi.org/10.1021/acs.analchem.0c00828
Anal. Chem. 2020, 92, 7733−7737