Talanta 190 (2018) 321–326

Contents lists available at ScienceDirect

Talanta
journal homepage: www.elsevier.com/locate/talanta

Fast extraction of cannabinoids in marijuana samples by using hard-cap T

espresso machines
Kristiina Leiman, Lorenzo Colomo, Sergio Armenta, Miguel de la Guardia,
⁎
Francesc A. Esteve-Turrillas
Department of Analytical Chemistry, University of Valencia, Jeroni Muñoz Building, 50th Dr. Moliner St., 46100 Burjassot, Valencia, Spain

A R T I C LE I N FO A B S T R A C T

Keywords: A simple, quick and low cost procedure was developed for the extraction of Δ9-tetrahydrocannabinol, canna-
Δ9-tetrahydrocannabinol bidiol, and cannabinol from marijuana samples, based on the use of a hard-cap espresso extraction with 2-
Cannabidiol propanol. After extraction, cannabinoids were directly determined after appropriate dilution by gas-chromato-
Cannabinol graphy-mass spectrometry, reaching a limit of detection from 0.03 to 0.05 mg g−1. Extraction efficiency was
Hard-cap espresso extraction
evaluated by the comparison of results obtained for seized samples by the proposed method and a reference
Gas chromatography-mass spectrometry
Ion mobility spectrometry
methodology based on ultrasound-assisted extraction. Moreover, ion mobility was proposed for the rapid and
sensitive determination of Δ9-tetrahydrocannabinol and cannabidiol providing a quick response for the analysis
of seized marijuana samples in 1 min, including extraction, dilution and determination.

1. Introduction included in commercial mass libraries. Sample injection in GC techni-

Cannabis sativa has been extensively used in the Middle East and
Asia, since the 6th century B.C. due to its hallucinogenic and medicinal
properties [1]. Physiological effects in humans include changes in heart
and respiratory rate, altered blood pressure, dry mouth and increased
appetite, but also feeling of euphoria, relaxation, altered time percep-
tion, lack of concentration and impaired learning [2]. Because of that,
marijuana is frequently used for recreational purposes and it is highly
regulated all around the world.
The primary effects of marijuana are caused by the cannabinoids
present in the plant, which act on cannabinoid receptors altering brain
neurotransmissions [3]. Cannabinoids are present in fresh plants in
form of their acid precursors, but a decarboxylation to neutral forms
happens in course of time, upon heating or under alkaline conditions
[4]. Thus, Δ9-tetrahydrocannabinol (THC), cannabidiol (CBD), and the
THC degradation product cannabinol (CBN) are the most abundant
cannabinols analysed in harvested plants all seized drugs.
The United Nations Office on Drugs and Crime (UNODC) re-
commends different methods for the identification and analysis of
cannabis products, from presumptive tests like colour tests and im-
munoassays, to ion mobility spectrometry (IMS) and chromatography
methods [5]. Gas chromatography with flame ionization (GC-FID) or
mass spectrometry (GC-MS) detection is the most employed approach,
and reference MS spectra of the most common cannabinoids are usually
ques is typically carried out at temperatures higher than 150 °C and, as
consequence, a decarboxylation of acid cannabinoids occurs during
analytical measurement [5]. High-performance liquid chromatography
(HPLC) can be also employed for the analysis of cannabinoids, allowing
the determination of acid cannabinoids [6]. Extraction of cannabinoid
compounds from plant samples has been carried out by several ap-
proaches, including dynamic maceration [7], ultrasound-assisted ex-
traction (UAE) [8], supercritical fluid extraction [9], and microwave-
assisted extraction [7], using methanol, ethanol, acetone or hexane as
extraction solvents.
The use of hard cap espresso machines has been recently proposed
for analytical extractions due to its easy use, speed, availability and low
price, providing efficient extraction of organic compounds from solid
samples in few seconds. Hard cap espresso machines have been pre-
viously employed for the extraction of polycyclic aromatic hydro-
carbons [10] and polychlorinated biphenyls from soils [11], bioactive
compounds from vegetables and spices [12], and pesticides from air-
borne filters [13]. Moreover, matrix co-extraction is reduced because
extraction is made at mild conditions of 19 bar and 72 °C for a very
short period of time [14].
The main objective of this study has been the development of a
method for the quick determination of cannabinoids from marijuana
samples based on the use of hard cap espresso extraction and GC-MS
determination. Extraction efficacy was compared to that obtained by a

⁎
Corresponding author.
E-mail address: francesc.a.esteve@uv.es (F.A. Esteve-Turrillas).

https://doi.org/10.1016/j.talanta.2018.08.009
Received 28 March 2018; Received in revised form 30 July 2018; Accepted 3 August 2018
Available online 06 August 2018
0039-9140/ © 2018 Elsevier B.V. All rights reserved.
K. Leiman et al. Talanta 190 (2018) 321–326

Table 1
Gas-chromatography-mass spectrometry (GC-MS) and ion mobility spectrometry (IMS) conditions and analytical figures of merit for the determination of canna-
binoids in marijuana samples.
GC-MS

Analyte RT (min) Target ions (m/z) LOD LOQ RSD (%)

mg L−1 mg g−1 mg L−1 mg g−1

TPP 9.70 325 + 326 – – – – –

CBD 9.95 231 + 246 0.09 0.05 0.30 0.15 7.4
THC 10.90 299 + 314 0.05 0.03 0.17 0.08 6.3
CBN 11.20 238 + 295 0.08 0.04 0.27 0.13 5.6

IMS

Analyte Drift time (ms) K0 (cm2 V−1 s−1) LOD LOQ RSD (%)

mg L−1 mg g−1 mg L−1 mg g−1

CBD + THC 16.860 1.0515 0.2 0.1 0.5 0.3 4.2

Abbreviations: K0, reduced mobility; LOD,limit of detection; LOQ, limit of quantification; RSD, relative standard deviation; RT, retention time.

reference UAE for the analysis of seized samples. Moreover, the use of with 200 mL deionized water to avoid damages in system tubing or
IMS for the determination of cannabinoids in sample extracts has been heating block. Following the aforementioned recommendations the
also proposed in order to achieve its determination extremely fast in coffee machine was used in our laboratory during the last two and a
less than 1 min, including extraction and analysis steps. half years without observing any damage or incident.

2. Experimental 2.3. Extraction reference method

2.1. Reagents and samples
Δ9-tetrahydrocannabinol, cannabidiol and cannabinol standard so-
lutions of 1 mg mL−1 in methanol were purchased from Cerilliant
(Round Rock, TX, USA). Triphenyl phosphate (TPP, 99%), provided by
Sigma (St. Louis, MO, USA), was employed as internal standard for GC-
MS determinations and HPLC grade 2-propanol was provided by
Scharlab (Barcelona, Spain).
Cannabis plant samples, obtained from local seizures, were kindly
supplied by Unidad de Inspección de Farmacia y Control de Drogas del
Área de Sanidad de Valencia (Valencia, Spain). Samples were chopped
and homogenized in order to have as representative as possible aliquots
of the sample. Extraction efficiency was assessed using buds, flowers
and stems by separate.
2.2. Hard cap espresso extraction
A Nespresso Essenza Manual XN2003 Krups (Barcelona, Spain)
coffee machine was used for the extraction of cannabis samples. No
modification was made to the extraction system and it was placed in a
fume hood to exhaust solvent vapours. Additional safety conditions can
be seen in the literature [12]. 0.2 g of sample was homogenized with 2 g
Speed matrix dispersing agent for Applied Separations (Allentown, PA,
USA) and placed inside a Nespresso compatible stainless steel refillable
capsule from Mycoffestar (Zurich, Switzerland). Capsule volume was
completed with additional dispersing agent and a 47 mm borosilicate
filter (Scharlau, Barcelona, Spain) was placed on the top to filter the
extract and to avoid any sample release. Filled cap was inserted into the
coffee machine and it was extracted with 100 mL 2-propanol in 40 s. No
additional filtration was required for extract analysis. Sample extract
was diluted in 2-propanol appropriately and TPP was added as internal
standard for GC-MS determination at a final concentration of 1 mg L−1.
Before starting the extraction of samples, the internal parts of the
hard cap espresso machine were heated and purged with 100 mL 2-
propanol using refillable capsule filled with dispersing agent.
Additionally, 20 mL 2-propanol was employed between samples to
clean the system and to avoid any cross contamination. After the ex-
traction of samples, internal parts of the espresso machine were cleaned
Ultrasound-assisted extraction (UAE) was chosen as reference
method for the extraction of cannabinoids due to the fact that it has
been widely used in previous studies for these extractions [15,16].
Thus, a Branson 2510 ultrasound cleaner water bath from Branson
(Hampton, NH, USA) was used for the extraction of samples using the
conditions detailed next. 0.2 g homogenized sample was introduced in a
100 mL glass beaker closed and extracted with 25 mL 2-propanol for
25 min at room temperature. Then, sample extract was filtrated using a
0.22 µm polytetrafluoroethylene (PTFE) syringe filter and, after ap-
propriate dilution, TPP was added at a final concentration of 1 mg L−1.
2.4. GC-MS procedure
GC-MS determination of cannabinoids was performed by
using an Agilent Technologies 7890A series GC system (Palo alto,
CA, USA) equipped with a Zebron ZB-5MS capillary column
(30 m x 0.32 mm x 0.25 µm), and coupled with Agilent Technologies
5975C inert XL EI/CI MSD triple axis single quadrupole detector
equipped with an Agilent Technologies 7683 Series Autosampler. A
volume of 1 µL sample extract was injected using splitless mode and
1 mL min−1 helium carrier flow. Oven conditions were: 150 °C hold
1 min, then heated at 10 °C min−1 to 250 °C, and finally hold at 250 °C
for 5 min. The analysis time by GC-MS was estimated in 20 min, in-
cluding the oven cooling time and the autosampler injection. Electron
impact ionization at 70 eV was employed using the selected ion mon-
itoring (SIM) acquisition mode. Selected ions for TPP, CBD, THC and
CBN are shown in Table 1.
2.5. IMS procedure
An IONSCAN-LS 400B from Smiths Detection (Morristown, NJ,
USA) IMS equipped with a 63Ni foil radioactive ionization source, was
used to determine THC in this study. Smiths Detection IM station v.
5.389 software was used for data acquisition and processing. 4 µL of
sample extract was placed on a 2 µm x 46.2 mm PTFE filter from
Whatman Inc. (Florham Park, NJ, USA) and thermally desorbed in the
IMS. Plasmagram was obtained in positive mode, using nicotinamide as
internal calibrant and reactant with a reduced mobility (K0) of

322
K. Leiman et al.
1.860 cm2 V−1 s−1. The shutter grid width was 0.2 ms and plasma-
grams were collected for 40 ms. The desorber, inlet and drift tube
temperatures were set to 255, 290 and 237 °C, respectively. Total time
of measurement was 20 s. A 300 mL min−1 counterflow of dry air was
introduced as drift gas at the end of the drift region. The electric field
strength in the drift region was 252 V cm−1, with a total drift voltage of
1763 V and a drift-tube length of 7 cm.
3. Results and discussion
3.1. Cannabinoids determination by GC-MS
Filtered sample extracts were directly analysed by GC-MS after ap-
propriate dilution. Table 1 shows selected ions for the SIM acquisition
of the studied cannabinoids and the internal standard. As it has been
commented in the introduction section, decarboxylation of acid can-
nabinoids may occur during their analysis by GC with injection tem-
peratures higher than 150 °C [5]. Thus, major acid cannabinoids pre-
sent in the sample, such as tetrahydrocannabinolic and cannabidiolic
acids will be transformed in the GC injector and quantified as THC and
CBD, respectively.
Analytical figures of merit of the GC-MS determination are sum-
marized in Table 1. Calibration curves were prepared from 0.3 to
5.0 mg L−1 solutions in 2-propanol. Linear relationships were obtained
for calibration curves with correlation coefficients R2 higher than 0.992
for all studied cannabinoids. Limits of detection (LOD) and quantifi-
cation (LOQ) were calculated as 3 and 10 times, respectively, the
standard deviation of the intercept divided by the calibration slope. The
obtained LODs and LOQs were from 0.05 to 0.09 mg L−1 and from 0.17
to 0.30 mg L−1, respectively. Precision was established as the relative
standard deviation (RSD) of 0.5 mg L−1 standards prepared in 2-pro-
panol measured 5 consecutive times. Adequate precision was obtained
with RSD values lower than 8%.
3.2. Hard cap espresso extraction of cannabinoids
Cannabinoids are relatively thermal stable compounds, with the
aforementioned exception of acid ones. However, some studies in the
literature indicate that little decomposition of THC may occur at 65 °C
and that it can be significant at temperatures from 85° to 100 °C [17].
Thus, thermal stability of the studied cannabinoids was assessed in
order to check that they resist the extraction at 72 °C for few seconds.
To do this study, an empty capsule filled only with dispersing agent was
spiked with 100 µL of a 500 mg L−1 cannabinoids standard and ex-
tracted with 100 mL 2-propanol by hard cap espresso extraction and
then analysed by GC-MS. Quantitative recoveries were obtained for
CBD, THC and CBN compounds, ranging from 91% to 106%, being
confirmed the stability of these compounds under the employed ex-
traction conditions.
Acetonitrile:water and ethanol:water solutions were employed as
extractant solvents in precedent studies carried out with hard cap
espresso machines. Nevertheless, the presence of high percentages of
water in the extracts may unfeasible the analytical determination by
using GC-MS or IMS. Thus, pure acetonitrile and 2-propanol were se-
lected and evaluated for the first time as solvents for the hard cap
espresso extraction of cannabinoids from marijuana samples. 0.2 g
sample was mixed with dispersing agent, placed inside the capsule,
extracted five consecutive times with 50 mL of each one of the two
considered solvents, and analysed by GC-MS. Sample extracts were
adequately diluted when required, due to the high concentration of
cannabinoids present in the analysed samples. Fig. 1 shows that can-
nabinoids were mostly extracted in the first 50 mL; nevertheless,
100 mL were required to obtain a quantitative extraction, being the
total extraction time less than 40 s. The obtained recovery values were
98 ± 3, 99 ± 7, and 97 ± 8% for buds, leaves and stems, respec-
tively. Similar recovery results were obtained for pure acetonitrile
323
Talanta 190 (2018) 321–326
Fig. 1. Recovery of the three evaluated cannabinoids in five consecutive ex-
tractions of 0.2 g marijuana samples with 50 mL 2-propanol analysed by gas-
chromatography-mass spectrometry.
extracts. However, 2-propanol was selected in this study based on its
environmentally friendly behaviour as compared with acetonitrile [18].
Moreover, 2-propanol presents a higher compatibility than acetonitrile
in the analysis of extracts by GC-MS and IMS, being the main organic
solvent employed for IMS analysis due to its high spreading capacity
onto the thermal desorption PTFE membrane.
The proposed procedure provided a 500 times dilution, considering
that 0.2 g sample was extracted with 100 mL solvent. Nevertheless, it
was not a drawback because of cannabinoid concentration in marijuana
is typically in the percentage range. Thus, the obtained LODs, which
were 0.05, 0.03 and 0.04 mg g−1, for CBD, THC and CBN, respectively,
were low enough to reach the required analytical sensitivity.
The proposed methodology was compared to those found in the
literature for the extraction of cannabinoids in marijuana samples.
Table 2 shows that a wide variability of organic solvent mixtures have
been employed for the extraction of cannabinoids, being used mixtures
of methanol:chloroform for vortex assisted extraction [19–21], but also
acetonitrile:methanol [22], methanol:n-hexane [23], and hexane:ethyl
acetate [24] for sonication. A centrifugation step is typically carried out
to clarify the extracts before chromatography analysis. The involved
total analysis time of reported methods was in the 30–60 min interval,
which highlights the speed and simplicity of the proposed extraction
method based on the use of hard-cap espresso machines. Moreover,
filtration or centrifugation steps were not required because an on-line
filtration is carried out by placing a borosilicate filter in the cap output.
Regarding sensitivity, no great differences were observed from litera-
ture studies with LOQ values for cannabinoids determination in mar-
ijuana from 0.05 to 2.4 mg g−1.
3.3. Determination of cannabinoids in seized samples
The concentration of the studied cannabinoids in seized marijuana
samples was determined in triplicate using hard cap espresso extraction
and GC-MS. Fig. 2 shows the obtained GC-MS chromatograms for a
sample and, as it can be seen, the three evaluated cannabinoids were
easily identified and determined. Table 3 shows the obtained results for
five buds, two leaves and one stem samples. The obtained concentration
range for buds (n = 5) varied from 16 to 95 mg g−1 for THC, from 0.15
to 0.24 mg g−1 for CDB, and from 4.3 to 21.0 mg g−1 for CBN. In the
case of leaves (n = 2) and stems (n = 1) a low cannabinoid content was
mainly obtained, from 0.87 to 7.2 mg g−1 for THC, from < LOD to
0.076 mg g−1 for CBD, and from 0.9 to 9.7 mg g−1 for CBN.
The aforementioned samples were analysed in triplicate by the re-
ference UAE and obtained results were compared with those found by
the proposed extraction method. Table 3 shows the results for both,
hard cap espresso extraction and UAE, and as it can be seen, similar
K. Leiman et al. Talanta 190 (2018) 321–326

Table 2
Comparison of analytical methodologies for the extraction and quantification of CBD, CBN and THC in marijuana samples.
Sample (mg) Extraction procedure Detection LOQ (mg g−1) Ref.

200 2× HPLC-DAD 0.05 [19]

40 mL methanol:chloroform (9:1)
Shaken 30 min
Centrifuged 10 min, 3000 rpm
200 20 mL methanol:chloroform (9:1) HPLC-DAD 0.2–0.7 [20]
Shaken 30 min
Evaporated N2 stream
Dissolved in100 µL water:methanol (1:1)
100 2 mL methanol:chloroform (9:1) GC-MS 0.5 [21]
Shaken 20 min
Centrifuged 5 min, 300 g
50 4× UHPSFC 2.0 [22]
2.5 mL acetonitrile:methanol (8:2)
Sonicated
Centrifuged
500 10 mL methanol:n-hexane (9:1) HPLC -DAD 2.0–2.4 [23]
Sonication 20 min
Rest 7 min
10 2× GC–MS 0.1 [24]
3.0 mL hexane:ethyl acetate (6:4)
Sonication 30 min
Centrifuged 5 min, 3000 rpm
200 100 mL 2-propanol GC-MS 0.08–0.15 This study
Hard-cap espresso extraction 40 s

Abbreviations: GC-MS, gas chromatography-mass spectrometry, HPLC-DAD, high-performance liquid chromatography-diode array detector, LOQ, limit of quanti-
fication, UHPSFC, ultra-high-performance supercritical fluid chromatography and diode array/mass spectrometric detection.

Table 3
Cannabinoids concentration found in seized marijuana samples extracted by
hard cap espresso extraction (HCEE) and ultrasound-assisted extraction (UAE)
and analysed by gas-chromatography-mass spectrometry.
Sample Cannabinoid [Cannabinoid] (mg g−1 ± s, n = 3) Difference

HCEE UAE

Bud 1 THC 70 ± 14 65 ± 8 5.00

CBD 0.195 ± 0.002 0.178 ± 0.008 0.02
CBN 6.2 ± 1.0 9.7 ± 0.2 −3.50
Bud 2 THC 95 ± 7 82 ± 4 13.00
CBD 0.24 ± 0.02 0.236 ± 0.009 0.00
CBN 5.4 ± 0.4 7.6 ± 0.6 −2.20
Bud 3 THC 52 ± 6 67 ± 9 −15.00
CBD 0.18 ± 0.03 0.20 ± 0.02 −0.10
CBN 5.2 ± 0.8 6.8 ± 0.9 −1.60
Fig. 2. Gas-chromatography-mass spectrometry chromatogram obtained for (A)
Bud 4 THC 37.7 ± 1.8 41.8 ± 1.9 −4.10
1 mg L−1 cannabinoids standard and (B) 0.2 g marijuana bud sample extracted
CBD 0.148 ± 0.012 0.133 ± 0.015 0.02
by the proposed hard cap espresso extraction with 100 mL 2-propanol. CBN 4.3 ± 0.3 4.0 ± 0.5 0.30
Bud 5 THC 15.7 ± 0.5 11.10 ± 0.3 4.60
CBD 0.191 ± 0.014 0.24 ± 0.05 −0.05
concentration values were obtained by using the two extraction
CBN 21.0 ± 1.6 26 ± 3 −5.00
methods. Results were compared by using the t-student value for paired Leaves 1 THC 1.4 ± 0.6 1.52 ± 0.08 −0.12
samples to compare concentrations found after hard cap espresso ex- CBD 0.059 ± 0.003 0.065 ± 0.007 −0.01
traction with those obtained by the UAE method. The calculated t-va- CBN 9.7 ± 0.3 8.0 ± 0.7 1.70
lues were 0.225, 1.055, and 1.661 for THC, CBD and CBN, respectively, Leaves 2 THC 7.2 ± 0.8 5.5 ± 1.4 1.70
CBD 0.076 ± 0.002 > LOD –
being lower than tabulated ones, 2.365 (two tailed, α = 0.05 and 7 CBN 2.0 ± 0.2 2.1 ± 0.2 − 0.10
freedom degrees) for THC and CBN, and 2.571 (two tailed, α = 0.05 Stems 1 THC 0.87 ± 0.04 0.8 ± 0.3 0.07
and 5 freedom degrees) for CBD. Thus, it indicated that both extraction CBD > LOD > LOD –
methods assayed provided statistically comparable results for the ex- CBN 0.9 ± 0.2 0.88 ± 0.13 0.02
traction of cannabinoids in marijuana samples.

3.4. Determination of cannabinoids by IMS
The IMS plasmagrams of THC, CBD and CBN standards of 1 mg L−1
in 2-propanol are shown in Fig. 3A. The most intense peak at 9 ms drift
time corresponds to the internal calibrant nicotinamide with a reduced
mobility (K0) of 1.860 cm2 V−1 s−1 and a proton affinity of 918.3 kJ
mol−1 used to compensate the effect of small changes in temperature
and pressure during IMS determinations. CBN did not provide any IMS
signal, even at 50 mg L−1 concentration level. The absence of signal for
CBN in IMS systems that use nicotinamide as calibrant has been also
reported in literature [25], but CBN has been detected in IMS system
that use a calibrant with a low proton affinity (878.6 kJ mol−1) such as
isobutyramide [26]. Regarding CBD and THC two overlapped peaks
were observed at 16.860 ms drift time, which corresponded to a K0
value of 1.0515 cm2 V−1 s−1. This obtained K0 value was consistent
with previously reported values for THC (1.050 ± 0.004 cm2 V−1 s−1)
[27] and CBD (1.07 ± 0.02 cm2 V−1 s−1) [28]. Thus, CBD and THC
were quantified together by using a THC standard because of the

324
K. Leiman et al. Talanta 190 (2018) 321–326

carried out using the hard cap espresso extracts analysed by IMS.
Fig. 3B shows the obtained plasmagram for a bud sample, diluted 100
times to fit the IMS linear range, and as it can be seen CBD/THC signal
was identified. Table 4 shows that the obtained results and concentra-
tion values found were similar to those obtained by GC-MS. It must be
indicated that IMS quantifies the sum of CBD and THC compounds, but
as it can be seen in Table 3, CBD concentration levels were negligible in
comparison with THC ones. The obtained results by IMS and GC-MS
analysis of field bud samples were compared by using the t-student
value for paired samples with a calculated t-value of 2.54 that is lower
than the tabulated one (2.78, two tailed, α = 0.05 and 4 freedom de-
grees). Thus, it can be concluded that IMS determination provided re-
sults statistically comparable to GC-MS for the determination of THC in
marijuana samples.
Therefore, THC can be accurately determined by using a procedure
based on a previous hard cap espresso extraction followed by IMS de-
termination, involving a total analysis time lower than 1 min, providing
an ultrafast response method for the analysis of seized samples.

4. Conclusions

It has been evidenced that the developed method for the major
cannabinoids extraction is a really encouraging example of the wide
range of possibilities that a conventional and low cost hard cap espresso
assisted extraction could offer in analytical laboratories. The quantita-
tive extraction of THC, CBD and CBN from buds, leaves and stems has
been achieved in a single and fast extraction of 40 s, using 100 mL 2-
propanol and GC-MS determination. The effects of solvent nature and
extraction volume have been evaluated to reach quantitative re-
coveries. Moreover, the proposed method allows a LOD of 0.05, 0.03
and 0.04 mg g−1 for the determination of CBD, THC and CBN in mar-
ijuana samples, respectively. Samples, obtained from local seizures,
Fig. 3. Ion mobility spectrometry plasmagrams of (A) 1 mg L−1 standard of were also extracted by a UAE reference method and the obtained results
THC, CBD and CBN in 2-propanol and (B) a marijuana bud sample extracted by were statistically comparable to those found by the proposed proce-
hard cap espresso (HCEE) and ultrasound-assisted extractions (UAE). dure.
Moreover, a quick response method has been developed for the
Table 4 determination of THC in marijuana samples by using hard cap assisted
Δ9-tetrahydrocannabinol (THC) concentration found in seized marijuana buds extraction and IMS determination. The proposed method involves an
extracted by hard cap espresso extraction and analysed by gas-chromatography- analysis time of 1 min and provided results statistically comparable to
mass spectrometry (GC-MS) and ion mobility spectrometry (IMS). those found by GC-MS, and a LOD of 0.1 mg g−1 THC, being an inter-
Sample [THC] (mg g−1 ± s, n = 3) Difference esting alternative to rapid presumptive tests that allows the identifi-
cation of THC from seizures, and also an accurate determination of its
a
GC-MS IMS concentration level.
Bud 1 70 ± 14 83 ± 12 12.7
Bud 2 95 ± 7 114 ± 8 18.9 Acknowledgements
Bud 3 52 ± 6 63 ± 2 10.9
Bud 4 37.7 ± 1.8 38 ± 10 −0.1 Authors gratefully acknowledge the financial support of
Bud 5 15.7 ± 0.5 18 ± 2 2.1
the Ministerio de Economía y Competitividad (CTQ-2014-52841-P). K.
a Leiman also thanks to the European Union for the Erasmus+ Mobility
Expressed as the sum of CBD and THC.
Traineeship Grant.
similar sensitivity provided by both compounds in IMS. Nevertheless,
Conflict of interest disclosure
concentration levels of THC in marijuana are typically higher than for
CBD. So, no interferences are expected to be found in the analysis of
The authors declare no competing financial interest.
seized samples.
The peak area obtained from the average plasmagram of an 11 s
References
window around the cannabinoid desorption time was employed as
analytical signal. Injection volume was evaluated from 1 to 10 µL using
[1] B. Thomas, M. ElSohly, The botany of cannabis sativa L, in: B. Thomas (Ed.), The
1 mg L−1 THC standard in 2-propanol, and 4 µL was selected because it Analytical Chemistry of Cannabis, Elsevier, Amsterdam, Netherlands, 2015, pp.
provided the maximum IMS signal that remained constant at increased 1–26.
volumes. Linearity was assessed using a calibration line of THC from 0.5 [2] P. Sharma, P. Murthy, M.M.S. Bharath, Chemistry, metabolism and toxicology of
to 3.0 mg L−1 prepared in 2-propanol with an obtained R2 higher than :cannabis: clinical implications, Iran. J. Psychiatry 7 (2012) 149–156.
[3] L. Console-Bram, J. Marcu, M.E. Abood, Cannabinoid receptors: nomenclature and
0.996. LOD and LOQ were calculated as aforementioned in the use of pharmacological principles, Prog. Neuropsychopharmacol. Biol. Psychiatry 38
GC-MS, being obtained values of 0.2 and 0.5 mg L−1, respectively. A (2012) 4–15.
high precision of analysis was obtained with a RSD around 4%. [4] J. Broséus, F. Anglada, P. Esseiva, The differentiation of fibre and drug type can-
nabis seedlings by gas chromatography/mass spectrometry and chemometric tools,
Determination of THC in marijuana bud samples (n = 5) was Forensic Sci. Int. 200 (2010) 87–92.

325
K. Leiman et al.
[5] United Nations Office on Drugs and Crime, Recommended methods for the identi-
fication and analysis of cannabis and cannabis products, United Nations, New York,
2009.
[6] Y.H. Wang, B. Avula, M.A. Elsohly, M.M. Radwan, M. Wang, A.S. Wanas,
Z. Mehmedic, I.A. Khan, Quantitative determination of Δ9-THC, CBG, CBD, their
acid precursors and five other neutral cannabinoids by UHPLC-UV-MS, Planta Med.
84 (2018) 260–266.
[7] V. Brighenti, F. Pellati, M. Steinbach, D. Maran, S. Benvenuti, Development of a new
extraction technique and HPLC method for the analysis of non-psychoactive can-
nabinoids in fibre-type Cannabis sativa L. (hemp), J. Pharm. Biomed. Anal. 143
(2017) 228–236.
[8] C. Agarwal, K. Máthé, T. Hofmann, L. Csóka, Ultrasound-assisted extraction of
cannabinoids from Cannabis sativa L. optimized by response surface methodology,
J. Food Sci. 83 (2018) 700–710.
[9] M. Sexton, K. Shelton, P. Haley, M. West, Evaluation of cannabinoid and terpenoid
content: cannabis flower compared to supercritical CO2 concentrate, Planta Med. 84
(2018) 234–241.
[10] S. Armenta, M. de la Guardia, F.A. Esteve-Turrillas, Hard cap espresso machines in
analytical chemistry: what else? Anal. Chem. 88 (2016) 6570–6576.
[11] D. Gallart-Mateu, A. Pastor, M. de la Guardia, S. Armenta, F.A. Esteve-Turrillas,
Hard cap espresso extraction-stir bar preconcentration of polychlorinated biphenyls
in soil and sediments, Anal. Chim. Acta 952 (2017) 41–49.
[12] M.T. Martínez-Sena, M. de la Guardia, F.A. Esteve-Turrillas, S. Armenta, Hard cap
espresso extraction and liquid chromatography determination of bioactive com-
pounds in vegetables and spices, Food Chem. 237 (2017) 75–82.
[13] A. López, C. Coscollà, V. Yusà, S. Armenta, M. de la Guardia, F.A. Esteve-Turrillas,
Comprehensive analysis of airborne pesticides using hard cap espresso extraction-
liquid chromatography-high-resolution mass spectrometry, J. Chromatogr. A 1506
(2017) 27–36.
[14] S. Dold, C. Lindinger, E. Kolodziejczyk, P. Pollien, S. Ali, J.C. Germain, S.G. Perin,
N. Pineau, B. Folmer, K.H. Engel, D. Barron, C. Hartmann, Influence of foam
structure on the release kinetics of volatiles from espresso coffee prior to con-
sumption, J. Agric. Food Chem. 59 (2011) 11196–11203.
[15] A.A.M. Stolker, J. van Schoonhoven, A.J. de Vries, I. Bolbedijk-Pastorova,
W.H.J. Vaes, R. van den Berg, Determination of cannabinoids in Cannabis products
using liquid chromatography-ion trap mass spectrometry, J. Chromatogr. A 1058
(2004) 143–151.
[16] J. Omar, M. Olivares, M. Alzaga, N. Etxebarria, Optimisation and characterization
of marijuana extracts obtained by supercritical fluid extraction and focused ultra-
sound extraction and retention time locking GC-MS, J. Sep. Sci. 36 (2013)
326
Talanta 190 (2018) 321–326
1397–1404.
[17] M.A. Repka, M. Munjal, M.A. ElSohly, S.A. Ross, Temperature stability and
bioadhesive properties of Δ9-tetrahydrocannabinol incorporated hydro-
xypropylcellulose polymer matrix systems, Drug. Dev. Ind. Pharm. 32 (2006)
21–32.
[18] C. Capello, U. Fischer, K. Hungerbühler, What is a green solvent? A comprehensive
framework for the environmental assessment of solvents, Green Chem. 9 (2007)
927–934.
[19] B. Patel, D. Wene, Z.T. Fan, Qualitative and quantitative measurement of canna-
binoids in cannabis using modified HPLC/DAD method, J. Pharm. Biomed. Anal.
146 (2017) 15–23.
[20] B. De Backer, K. Maebe, A.G. Verstraete, C. Charlier, Evolution of the content of
THC and other major cannabinoids in drug-type cannabis cuttings and seedlings
during growth of plants, J. Forensic Sci. 57 (2012) 918–922.
[21] G. Lopes de Oliveira, M.H. Voloch, G.B. Sztulman, O.N. Neto, M. Yonamine,
Cannabinoid contents in cannabis products seized in São Paulo, Brazil, 2006–2007,
Forensic Toxicol. 26 (2008) 31–35.
[22] M. Wang, Y.H. Wang, B. Avula, M.M. Radwan, A.S. Wanas, Z. Mehmedic, J. van
Antwerp, M.A. ElSohly, I.A. Khan, Quantitative determination of cannabinoids in
cannabis and cannabis products using ultra-high-performance supercritical fluid
chromatography and diode array/mass spectrometric detection, J. Forensic Sci. 62
(2017) 602–611.
[23] L. Ambach, F. Penitschka, A. Broillet, S. König, W. Weinmann, W. Bernhard,
Simultaneous quantification of delta-9-THC, THC-acid A, CBN and CBD in seized
drugs using HPLC-DAD, Forensic Sci. Int. 243 (2014) 107–111.
[24] Z. Bruci, I. Papoutsis, S. Athanaselis, P. Nikolaou, E. Pazari, C. Spiliopoulou,
G. Vyshka, First systematic evaluation of the potency of Cannabis sativa plants
grown in Albania, Forensic Sci. Int. 222 (2012) 40–46.
[25] C.W. Su, K. Babcock, M. Donahue, An investigation of the effects of temperature and
holding materials on the IMS detection of THC, Int. J. Ion. Mobil. Spectrom. 4
(2001) 31–34.
[26] A.C. Ruth, C.M. Gryniewicz-Ruzicka, M.L. Trehy, N. Kornspan, G. Coody,
Consistency of label claims of internet-purchased hemp oil and cannabis products as
determined using IMS and LC-MS: a marketplace survey, J. Regul. Sci. 4 (2016) 1–6.
[27] J.R. Verkouteren, J.L. Staymates, Reliability of ion mobility spectrometry for qua-
litative analysis of complex, multicomponent illicit drug samples, Forensic Sci. Int.
206 (2010) 190–196.
[28] V. Sedwick, M. Massey, T. Codio, A.B. Kanu, Method validation parameters for
drugs and explosives in ambient pressure ion mobility spectrometry, Int. J. Ion.
Mobil. Spec. 20 (2017) 75–86.