Ad

CGa

b

c

d

e

a

ARR1AA

KHCLD

1

“aisln

fstda

e

h0

Journal of Pharmaceutical and Biomedical Analysis 149 (2018) 532–540

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

j o ur nal ho me page: www.elsev ier .com/ lo cate / jpba

nalysis of cannabinoids in commercial hemp seed oil andecarboxylation kinetics studies of cannabidiolic acid (CBDA)

inzia Citti a,b,c, Barbara Pacchettid, Maria Angela Vandelli c, Flavio Fornic,iuseppe Cannazzab,c,e,∗

Dipartimento di Scienze e Tecnologie Biologiche ed Ambientali, Universitad̀el Salento, Via per Monteroni, 73100 Lecce, ItalyCNR NANOTEC, Campus Ecoteckne dell’Università del Salento, Via per Monteroni, 73100 Lecce, ItalyDipartimento di Scienze della Vita, Università di Modena e Reggio Emilia, Via Campi 103, 41125 Modena, ItalyLinnea SA, Via Cantonale, 6595 Riazzino, SwitzerlandFEDERCANAPA Federazione Italiana Canapa, via Genova 23◦ , 00184 Rome, Italy

r t i c l e i n f o

rticle history:eceived 4 September 2017eceived in revised form4 November 2017ccepted 15 November 2017vailable online 20 November 2017

eywords:emp seed oil

a b s t r a c t

Hemp seed oil from Cannabis sativa L. is a very rich natural source of important nutrients, not onlypolyunsaturated fatty acids and proteins, but also terpenes and cannabinoids, which contribute to theoverall beneficial effects of the oil. Hence, it is important to have an analytical method for the determi-nation of these components in commercial samples. At the same time, it is also important to assess thesafety of the product in terms of amount of any psychoactive cannabinoid present therein. This workpresents the development and validation of a highly sensitive, selective and rapid HPLC-UV methodfor the qualitative and quantitative determination of the main cannabinoids, namely cannabidiolic acid(CBDA), tetrahydrocannabinolic acid (THCA), cannabidiol (CBD), tetrahydrocannabinol (THC), cannabinol

annabidioliquid chromatographyecarboxylation kinetics

(CBN), cannabigerol (CBG) and cannabidivarin (CBDV), present in 13 commercial hemp seed oils. More-over, since decomposition of cannabinoid acids generally occurs with light, air and heat, decarboxylationstudies of the most abundant acid (CBDA) were carried out in both open and closed reactor and the kinet-ics parameters were evaluated at different temperatures in order to evaluate the stability of hemp seedoil in different storage conditions.

© 2017 Elsevier B.V. All rights reserved.

. Introduction

Cannabis sativa L. produces small fruits, botanically defined asachenes”, although they are usually named “seeds”, which are vari-ble in size depending on the cultivar. The actual “seed” is enclosedn the pericarp, which represent the protective “shell”. Most of theeed consists of an embryo, mainly the two cotyledons (embryoniceaves), rich in oils, proteins and carbohydrates, that represents theourishment of the plant during germination [1].

According to an old legend, Buddha (Prince Siddharta Gautama),ounder of Buddhism, was able to survive eating only one hempeed each day for six years. This apocryphal story reveals a seed of

ruth since hemp seeds possess an extremely high nutritional valueue to a high content of unsaturated fatty acids (about 80% of fattycids) and proteins (about 25%) [2]. Moreover, it has been suggested

∗ Corresponding author at: Dipartimento di Scienze della Vita, Università di Mod-na e Reggio Emilia, Via Campi 103, 41125 Modena, Italy.

E-mail address: giuseppe.cannazza@unimore.it (G. Cannazza).

ttps://doi.org/10.1016/j.jpba.2017.11.044731-7085/© 2017 Elsevier B.V. All rights reserved.

that other minor components, such as terpenes and cannabinoids,could contribute to the surprising beneficial effects of hemp seeds[3].

After removal of the bracts, the seed is squashed or macerated,and finally pressed at high pressure to give the hemp seed oil and amushy residue, the “husk”, which is used as fertilizer or compressedinto tablets and used for cattle feed. The hemp seed oil obtained bycold pressing finds its main use as food. The hemp seed oil hasa pleasant nutty flavour, sometimes with a hint of a bitter after-taste. Polyunsaturated fatty acids (PUFA) constitute about the 80%of the total fatty acids, including linoleic, �- and �-linolenic acid,which offer the best omega-6 to omega-3 PUFA ratio consideredoptimal for nutrition [4]. Cold-pressed hemp seed oil may also serveas dietary source of natural antioxidants (�-tocopherol, vitamin E,etc.) for disease prevention and health promotion [4,5]. Therefore,the presence of antioxidants is a warranty of a cold-pressed oil, thus

of chemical stability against oxidative rancidity and a consequentrelatively short shelf life.

Although hemp seeds do not contain any cannabinoid, their con-tact with the resin secreted by the epidermal glands located on

https://doi.org/10.1016/j.jpba.2017.11.044http://www.sciencedirect.com/science/journal/07317085http://www.elsevier.com/locate/jpbahttp://crossmark.crossref.org/dialog/?doi=10.1016/j.jpba.2017.11.044&domain=pdfmailto:giuseppe.cannazza@unimore.ithttps://doi.org/10.1016/j.jpba.2017.11.044

and B

flnprdsacaTaTiieHhh

glstebot

nspbsacfap

bm

Fn

C. Citti et al. / Journal of Pharmaceutical

owers and leaves and/or a bad selection of the bracts of the perigo-ium, which have the highest cannabinoid content, can cause theresence of the latter in the hemp oil. Hence, cannabinoids actuallyepresent “impurities” of the hemp seed oil. Their concentrationepends from both the cultivar and the cleaning process of theeed. The concentrations of cannabinoids can be extremely variablemong different oil varieties. Cultivars of Cannabis sativa (CS) thatan be used in Europe for seed production are those authorized with

level of tetrahydrocannabinol (THC) lower than 0.2%. As a result,HC contamination in hemp seed oil is generally extremely lownd only exceptionally it exceeds the limit of 5 mg/kg (maximumHC limit in food imposed by German legislation in 2000) [6–8]. Its reasonable that such limit can be met in northern Europe, but its known that thermal stress that occurs in southern Europe couldasily raise the value of 5 mg/kg. In fact, the European Industrialemp Association has proposed a less strict THC limit of 10 mg/kg inemp seed oil and specified that the oil must be consumed withouteating, since heat would raise the THC content [8].

CS cultivars employed for seed production with a low THC levelenerally contain a high concentration of cannabidiol (CBD). Theatter is void of the THC psychotropic activity and is one of the mosttudied cannabinoids with proven anti-inflammatory, antiepilep-ic, analgesic, neuroprotective, antipsychotic and anticonvulsantfficacy [9–13]. Besides, it is also a potent antioxidant [14]. It haseen hypothesized that the nutraceutical properties of hemp seedil are not due only to its equilibrated PUFA content but mainly tohe presence of CBD [3].

Anyway, a purely chemical clarification is needed: CBD isot a biosynthetic product of CS. In fact, CS plant synthe-ises cannabigerolic acid (CBGA) from olivetolic acid and geranylyrophosphate [15]. CBGA can undergo transformations catalysedy enzymes such as tetrehydrocannabinolic acid-synthase (THCA-ynthase), which leads to tetrahydrocannabinolic acid (THCA),nd cannabidiolic acid-synthase (CBDA-synthase), which leads toannabidiolic acid (CBDA) [15]. There exist varieties of CS selectedor the lack of THCA-synthase, which produce only CBDA. Theyre called “legal”, thus they can be cultivated for fibre and seed

roduction.

Cannabinoid acids like THCA, CBDA, CBGA, etc. undergo a decar-oxylation process whose rate depends on different factors, butainly the higher the temperature the faster the process. This

ig. 1. Molecular structure of ibuprofen (IS), cannabidivarin (CBDV), cannabidiolic acid (CBol (THC) and tetrehydrocannabinolic acid (THCA).

iomedical Analysis 149 (2018) 532–540 533

decarboxylation finally leads to the formation of the correspondingneutral cannabinoids, THC, CBD, CBG, etc., which is not due to theaction of enzymes, but to a simple chemical reaction or, more pre-cisely, to a decomposition catalysed by heat. The conversion alsotakes place at room temperature, but it is much slower.

Anyway, the ratio between CBDA and CBD concentration inhemp seed oil could represent a valuable marker of both the storageconditions and the production process. To this end, 13 differ-ent commercial hemp seed oils obtained by cold pressing wereanalysed in order to determine the content of CBDA, CBD, cannabi-nol (CBN), cannabidivarin (CBDV), cannabigerol (CBG), THC andTHCA. CBN originates from cannabinolic acid (CBNA), which in turnderives from the oxidation of THCA as a consequence of an incorrectstorage (exposure to air) [16].

The analysis of cannabinoids in hemp food products is of greatimportance in order to guarantee the absence of any psychotropicor intoxicant component for human health. The analysis can be per-formed by several methods, but the most widely employed are gaschromatography coupled to flame ionization detector (GC-FID) ormass spectrometry detector (GC–MS) and high performance liquidchromatography coupled to UV (HPLC-UV) or mass spectrometrydetector (HPLC–MS) [17]. In particular, very few works have beenpublished on the analysis of cannabinoids in hemp food products,most of which carried out by GC–MS [18–21].

The present work focuses on the development and validation ofa rapid and sensitive HPLC-UV for the identification and quantifi-cation of the main cannabinoids, namely CBDA, THCA, CBD, THC,CBG, CBN and CBDV, in commercially available hemp seed oils. Themolecular structure of all the aforementioned cannabinoids andIS is reported in Fig. 1. Moreover, since CBDA to CBD ratio can betaken as a marker of cold pressing and good storage conditions, thedecarboxylation kinetics of CBDA to CBD in hemp seed oil has beeninvestigated in both open and closed reactor.

2. Materials and methods

2.1. Chemicals and reagents

Acetonitrile, water, 2-propanol, formic acid LC–MS grade werepurchased from Carlo Erba (Milan, Italy). THCA, CBDA, CBDV, THC,CBD, CBG and CBN were purchased from Cerilliant (Sigma Aldrich,

DA), cannabigerol (CBG), cannabidiol (CBD), cannabinol (CBN), tetrahydrocannabi-

5 and B

Rtf

2

(aa(lfgfieruw1i

t(opt(eviaMwssawmiTIC3

2

oco<owimCeaboTCaq

34 C. Citti et al. / Journal of Pharmaceutical

ound Rock, Texas). Organic hemp seed oils were bought fromhe normal commercial cycle of the Italian territory and numberedrom H1 to H13.

.2. HPLC-UV and HPLC-MS analysis

HPLC analyses were performed on an Agilent TechnologiesWaldbronn, Germany) modular model 1200 system, consisting of

vacuum degasser, a binary pump, a thermostated autosampler, thermostated column compartment and a diode array detectorDAD). A Poroshell 120 EC-C18 column (3.0 × 100 mm, 2.7 �m, Agi-ent, Milan, Italy) was used with a mobile phase composed of 0.1%ormic acid in both (A) water and (B) acetonitrile. The chromato-raphic conditions were set as follows: 0.0-10.0 min linear gradientrom 70 to 80% B; 10.0-10.1 min from 80 to 95% B; 10.1-11.0 minsocratic elution with 95% B; 11.0-11.1 min from 95 to 70% B forquilibration of the column with the initial conditions. The totalun time was 15 min. The flow rate was set at 0.4 mL/min. The col-mn temperature was set at 25◦ C. The sample injection volumeas 5 �L. The UV/DAD acquisitions were carried out in the range

90–500 nm and chromatograms were acquired at 228 nm. Threenjections were performed for each sample.

HPLC-MS analyses were carried out in order to ensure the iden-ity of the peaks and were performed on an Agilent TechnologiesWaldbronn, Germany) modular model 1200 system, consistingf a vacuum degasser, a binary pump, a thermostated autosam-ler, a thermostated column compartment and a 6540 quadrupoleime of flight (QToF) mass spectrometer operating in both positiveESI + ) and negative ionization (ESI − ) mode. The mass spectrom-ter experimental parameters were set as follows: the capillaryoltage was 3.5 kV, the nebulizer (N2) pressure was 35 psi, the dry-ng gas temperature was 350 ◦C, the drying gas flow was 11 L/minnd the skimmer voltage was 40 V. Data were acquired by Agilentass Hunter system software (version 6.0). The mass spectrometeras operated in full-scan mode in the m/z range 50–700. MS/MS

pectra were automatically performed with nitrogen as the colli-ion gas in the m/z range 50–700, using the auto MS/MS functionnd a collision energy of 20 eV. Extracted ion chromatograms (EICs)ere obtained with an accuracy of 10 ppm m/z from total ion chro-atogram (TIC) employing the m/z corresponding to the molecular

ons [M + H]+ 359.2217 for CBDA and THCA, 315.2300 for CBD andHC, 287.1998 for CBDV, 317.2468 for CBG and 311.2024 for CBN.n ESI− the molecular ions [M − H]− considered were 357.2164 forBDA and THCA, 313.2012 for CBD and THC, 285.1830 for CBDV,15.2385 for CBG and 309.1902 for CBN.

.3. Preparation of standard solutions

Blank matrix was obtained by washing hemp seeds (22 g clearedf bracts) with ethyl alcohol (3 × 100 mL) in order to removeannabinoids. Then the seeds were cold squeezed to obtain 4 mLf cannabinoid-free hemp seed oil (concentration of cannabinoidsLOD). The final blank matrix (20 mL) was obtained by diluting theil with 16 mL of 2-propanol. A stock solution of IS (5000 �g/mL)as prepared by dissolving 50 mg in 10 mL of 2-propanol. IS work-

ng solution was prepared by further diluting the stock solution inobile phase to 500 �g/mL. Stock solution of CBDV, CBDA, CBG,

BD, CBN, THC and THCA (1000 �g/mL) in methanol were prop-rly diluted in blank matrix to obtain samples of 10, 25, 50, 75nd 100 �g/mL. 100 �L of each sample were diluted in 890 �L oflank matrix and 10 �L of IS (500 �g/mL) to the final concentrationsf 1.0, 2.5, 5.0, 7.5 and 10 �g/mL for CBDV, CBDA, CBG, CBD, CBN,

HC and THCA and 5 �g/mL for IS. Independently prepared CBDV,BDA, CBG, CBD, CBN, THC and THCA stock solutions of 12.5, 50nd 80 �g/mL were used for the preparation of low concentrationuality control (LQC) (1.25 �g/mL), medium concentration quality

iomedical Analysis 149 (2018) 532–540

control (MQC) (5.00 �g/mL), and high concentration quality con-trol (HQC) (8.00 �g/mL) samples, respectively. QC samples wereprepared as for calibration standards.

2.4. Method validation

In order to demonstrate the reliability and robustness of theHPLC-UV method, full method validation was carried out based onEMEA guidelines and in agreement with international guidelinesfor analytical techniques for the quality control of pharmaceuti-cals (ICH guidelines) [22]. Specifically, the method was validatedin terms of specificity and selectivity, linearity, limit of detection(LOD) and limit of quantification (LOQ), precision and accuracy,recovery and matrix effect, dilution integrity and stability.

2.4.1. Specificity and selectivityMoreover, identification was assessed by comparing accurate

(within 5 ppm error) m/z of [M+H]+ and MS/MS spectra of authenticstandards with those obtained by HPLC-MS and MS/MS of hempseed oil samples.

2.4.2. LinearityCalibration curve was constructed at five calibration levels 1.0,

2.5, 5.0, 7.5 and 10 �g/mL of CBDV, CBDA, CBG, CBD, CBN, THCand THCA, and 5 �g/mL of ibuprofen (IS). Peak area ratios ofcannabinoid-to-IS were plotted versus nominal concentrations. Cal-ibration curve was evaluated at the beginning of each validationday of five consecutive days (n = 5). Linearity was assessed throughevaluation of the coefficient of determination (R2), which shouldbe greater than 0.998 using weighted regression method (1/x).

2.4.3. Limit of detection (LOD) and limit of quantification (LOQ)Limit of detection (LOD) was estimated based on a 3:1 signal-

to-noise ratio. Standard stock solutions of all analytes wereappropriately diluted at the levels of these estimated values andwere analysed repeatedly six times. Limit of quantification (LOQ)was estimated based on a 10:1 signal-to-noise ratio. Standard stocksolutions of all analytes were appropriately diluted at the levels ofthese estimated values and were analysed repeatedly six times.

2.4.4. Precision and accuracyThe precision and accuracy were evaluated at three levels, LQC

(1.25 �g/mL), MQC (5.00 �g/mL), and HQC (8.00 �g/mL). Each sam-ple was analysed repeatedly three times within a single day todetermine the intra-day precision and accuracy, and three timesa day for five successive days (n = 15) to determine the inter-dayprecision and accuracy. The precision was expressed as the coef-ficient of variance (RSD), and the accuracy was expressed as thepercentage of mean calculated to nominal concentration.

2.4.5. RecoveryRecovery was evaluated using authentic hemp seed oil samples

spiked at three different levels. To this end, a standard solutionat three different concentrations was added to the authentic oilsample. Recovery (R) was calculated by the following formula:

R = 100 ×[(CS − CH) /CSTD

]

where CS is the concentration of the sample after spiking with thestandard solution, CH is the concentration without spiking the stan-dard solution and CSTD is the concentration of the standard solution.

2.4.6. Matrix effectMatrix effect was evaluated for each analyte analyzing the QC

samples spiked at low, medium and high concentrations. ME valueswere calculated by dividing the analyte peak area in QC samples

and Biomedical Analysis 149 (2018) 532–540 535

srf

M

Wio

2

ardac

2

sTw

2

am

2

2

aldw

2

floaU

2o

kev

l

Wdtttb

t

B(w

Fig. 2. A) HPLC-UV chromatogram of blank sample spiked with IS (5 �g/mL). B)HPLC-UV chromatogram of analytical standards and IS mixture at LOQ (IS 5 �g/mL,analytes standards 1 �g/mL). C) HPLC-UV chromatogram of an authentic hemp seedoil sample (H8).

C. Citti et al. / Journal of Pharmaceutical

piked after the SPE procedure (post-extraction addition) to theesponse for neat standards. Matrix effect was calculated as matrixactor (MF) by the formula:

F% =(

AM − ASAS

)x100

here AM is the area of the peak (normalized for the area of the IS)n the matrix and AS is the area of the peak (normalized for the areaf the IS) of neat standards.

.4.7. Dilution integrityDilution integrity was demonstrated by spiking the blank matrix

t 2, 10 and 25 times the HQC concentration (16, 80 and 200 �g/mL,espectively). Six replicates per dilution factor were prepared andiluted with blank matrix. The concentrations were calculated bypplying the dilution factor 2, 10 and 25 against freshly preparedalibration curve.

.4.8. StabilityThe autosampler stability was determined for LQC and HQC

amples in blank matrix diluted samples for 24 h at 10 ◦C (n = 3).he drugs were considered stable if the mean concentration wasithin ±10% of the original concentration.

.5. Authentic hemp seed oil sample preparation

100 �L of hemp seed oil were diluted with 395 �L of 2-propanolnd 5 �L of IS solution. The solution was vortex mixed for 1 min ataximum speed and directly injected into the HPLC system.

.6. Decarboxylation studies

.6.1. Open reactor500 �L of hemp seed oil was placed in 10 open vials and put in

n oven at 80 ◦C. One vial was taken out every 15 min. The oil waset to cool down to room temperature and 100 �L were diluted asescripted above and analysed by HPLC-UV (n = 3). The experimentas repeated at 90, 100, 110 and 120 ◦C.

.6.2. Closed reactor500 �L of hemp seed oil was placed in 10 ampoules, which were

ame sealed and put in an oven at 120 ◦C. One ampoule was takenut every 15 min. The oil was let to cool down to room temperaturend 100 �L were diluted as descripted above and analysed by HPLC-V (n = 3).

.6.3. Calculation of kinetic parameters and free energy barriersf decarboxylation

The rate constant of decarboxylation (k) refers to a first-orderinetic process. Therefore, k was calculated by plotting the lin-ar logarithm values of unconverted CBDA (in percentage) (x axis)ersus time (y axis), according to the following equation Eq. (1):

n [CBDA] = −kt (1)here k is the decarboxylation rate constant (s−1) and t is the

ecarboxylation time (s). The slope of linear equation correspondso −k. The half-life of decarboxylation is the time in which 50% ofhe molecules of CBDA are decarboxylated into CBD (CBDA concen-ration is reduced by 50%). For a first-order reaction the half-life cane calculated by Eq. (2):

= ln 2 (2)

1⁄2 ky plotting the logarithmic values of k on the y axis and 1/TK−1) on the x axis, the slope of the equation corresponds to − EAR ,here EA is the activation energy and R the universal gas constant

536 C. Citti et al. / Journal of Pharmaceutical and Biomedical Analysis 149 (2018) 532–540

Table 1Amount of the main cannabinoids in 13 different commercial hemp seed oils. The values are expressed in ppm as mean ± standard deviation (n = 3).

Oil mg/Kg

CBDV CBDA CBG CBD CBN THC THCA CBDA/CBD

H1 8.769 ± 0.080

C. Citti et al. / Journal of Pharmaceutical and Biomedical Analysis 149 (2018) 532–540 537

F open( re als

sOfGlimpcpOs1T1

ig. 3. Decarboxylation of CBDA at 80 (A), 90 (B), 100 (C), 110 (D) and 120 ◦C (E) in�g/mL) vs time (min). Non-linear fitting curves of CBD and sum of CBDA and CBD a

eed oil should be sufficiently low to not have intoxicating effect.nly three countries so far have imposed a national limit for THC in

ood. The lowest THC limit in Europe of 5 mg/kg is recommended byermany [6–8], whereas Switzerland and Belgium maintain higher

evels of THC, 20 and 10 mg/kg, respectively [8]. In this context, it ismportant to develop a highly sensitive, but at the same time rapid

ethod in order to detect even low traces of cannabinoids. In theresent work, 13 different hemp seed oils were analysed for theirannabinoid content. In order to achieve a high sensitivity with ahotodiode array detector, we attempted different dilution options.ur previously developed method for medicinal cannabis was not

uitable for this purpose since the sensitivity reached was only of80 mg/kg [23], which is far beyond the European average limit.herefore, by gradual addition of 100 �L aliquots of 2-propanol to00 �L of hemp seed oil, the optimal dilution resulted to be a 1:5

reactor and at 120 ◦C in closed reactor (F). Values are expressed as concentrationo shown.

(v/v) ratio of oil:2-propanol (400 �L total volume of 2-propanol) toreach a sensitivity of 1 mg/kg in real oil sample. By injecting 5 �L ofthe oil/alcohol solution into the HPLC system a good peak shape wasobtained. The results of the analyses of the cannabinoids in hempseed oils are reported in Table 1. A representative chromatogramof an authentic hemp seed oil sample spiked with IS is reportedin Fig. 2 panel C. The data obtained indicated that oils H1, H4, H5,H6, H7 and H9 contained a low amount of CBDA and CBD, whichmeans an amount of total CBD [calculated as (CBDA × 0.877)+CBD]lower than 5 mg/kg. On the other hand, these oils, with the excep-tion of H9, which is poor of cannabinoids, contained between 5 and

10 mg/kg of total THC [calculated as (THCA × 0.877)+THC]. Thesevalues exceeded the German THC limits in hemp seed oil, but onlyif the limit refers to the sum of THC and THCA. The abovemen-tioned oils did not contain a high content of neutral cannabinoid

538 C. Citti et al. / Journal of Pharmaceutical and Biomedical Analysis 149 (2018) 532–540

in hemp seed oil in open and closed reactor.

TbohLasHiici

3

dTstwdrgootsocwwcfpawrtAeo

Fig. 5. Eyring plot for the calculation of the thermodynamic parameters of CBDA

Fig. 4. Decarboxylation kinetics of CBDA

HC ( < 2 mg/kg), but a high content of THCA, which should note considered intoxicating if the oil is consumed as such (with-ut heating). H2, H3, H8, H10, H11, H12 and H13 resulted with aigh CBDA and CBD content and an amount of total THC below theOQ. Looking at the CBDA/CBD ratio, which can be considered as

marker of cold pressing and good storage conditions, most oilshowed a low value in the average of 3.7:1. Conversely, H4, H5 and6 showed a very high CBDA/CBD ratio in the average of 14:1. H11,

n particular, was stored for one year at room temperature; it fact,t displayed a CBDA/CBD value below 1. These results suggested aorrelation between CBDA/CBD ratio and the oil age, but furthernvestigations are needed to have the scientific proof.

.4. Decarboxylation studies

Very few studies have been reported in the literature on theecarboxylation kinetics of cannabinoids in CS derived products.he decarboxylation kinetics of THCA in Cannabis Flos has beentudied by Perrotin-Brunel et al. in a vacuum system [24]. The reac-ion was described as a pseudo-first order catalysed by formic acid,hich encounters a keto-enol structure in the transition state. Theecarboxylation of THCA and CBDA was also investigated in openeactor in the absence of solvent by T. Veress in 1990 [25]. In order toive an estimation of the stability of cannabinoid acids in hemp seedil, a part of the work was focused on the evaluation of the influencef the temperature on the decarboxylation rate in this matrix. Givenhe very low amount of THCA or any other cannabinoid acid in hempeed oil, we focused our attention on the most abundant of this classf compounds, CBDA. Oil H5 was chosen for the experiments as itontained a CBD amount below the LOD. Decarboxylation studiesere conducted in both open and closed reactor. As shown in Fig. 3,hile the decrease of CBDA due to the decarboxylation process

an be assumed as a pseudo-first order reaction, the consequentormation of CBD does not appear to follow the same trend as itsrecursor. In details, at temperatures below 100 ◦C the sum of CBDnd CBDA in concentration remained constant (the registered lossas only 1% and 2% for the decarboxylation of CBDA at 80 and 90 ◦C

espectively). It could be hypothesised that below this temperature

he decarboxylation of CBDA leads only to the formation of CBD.t 100 ◦C or above, other processes could be involved, includingither the formation of unknown side products or the evaporationf the neutral cannabinoid. In fact, the higher the temperature the

decarboxylation.

lower the CBD formed (and the sum of the two forms). At 100 ◦Cthe loss of the total molar concentration was about 20%. At 110and 120 ◦C in open reactor, there was no linear increase of CBD ordecrease of CBDA, suggesting a more complex chemistry. The lossof the total molar concentration was about 30% at 110 ◦C, whereas itwas two times higher at 120 ◦C. Decarboxylation in closed reactordisplayed a linear decrease of CBDA and simultaneous formationof CBD, but the sum of the concentrations showed a loss of about11%. Fig. 4 represents the linear regressions of the decarboxylationkinetics at each temperature in open reactor (80, 90, 100, 110 and120 ◦C) and at 120 ◦C in closed reactor (sealed ampoules). From theobtained equations, the kinetic parameters have been calculated asdescribed in the methods section and reported in Table 2.

The Eyring equation (Eq. (5)) was then employed to calcu-lated the thermodynamic parameters �H# and �S# (enthalpy andentropy) of the decarboxylation process in open reactor, whichresulted to be about 86.4 ± 3.4 kJ mol−1 and −85.9 ± 9.7 J mol−1K−1,respectively. The contribution of �S# is very low (negative value)with respect to that of �H#, indicating that the free energy barrierdepends exclusively on the enthalpic factor (Fig. 5).

The Arrhenius plot was built to calculate the activation energy,which resulted 89.5 ± 3.6 kJ mol−1, and A, which was 6.88 × 108 s−1(Eq. (3)).

C. Citti et al. / Journal of Pharmaceutical and Biomedical Analysis 149 (2018) 532–540 539

Table 2Kinetic parameters (k and t1/2), linear regressions with R2, free energy barriers (�G#(T)) and activation entropy (�S#(T)) of the decarboxylation of CBDA in hemp seed oil ateach temperature in open reactor (80, 90, 100, 110 and 120 ◦C) and at 120 ◦C in closed reactor.

T (◦C) Linear regression R2 k (s−1) t1/2 (min) �G#(T) (kJ mol−1) �S#(T) (J K−1 mol−1)

80 y = −3.37 × 10−5x + 4.61 0.9695 3.37 ± 0.20 × 10−5 385 117.2 ± 0.2 −77.1 ± 9.690 y = −1.02 × 10−4x + 4.58 0.9811 1.01 ± 0.04 × 10−4 115 117.3 ± 0.1 −77.4 ± 9.6100 y = −2.55 × 10−4x + 4.70 0.9947 2.55 ± 0.08 × 10−4 38 117.8 ± 0.1 −77.6 ± 9.6110 y = −4.61 × 10−4x + 4.52 0.9943 4.61 ± 0.20 × 10−4 23 119.1 ± 0.1 −77.8 ± 9.6120 y = −7.46 × 10−4x + 4.51 0.9770 7.46 ± 0.41 × 10−4 14 120.7 ± 0.2 −78.0 ± 9.6120a y = −1.21 × 10−3x + 4.91 0.9657 1.21 ± 0.91 × 10−4a Experiment carried out in closed reactor.

Fl

Bwrf[treor

dtcwtr9tik5

4

atdocdam

[

[

[

[

[

[

[

[17] C. Citti, D. Braghiroli, M.A. Vandelli, G. Cannazza, Pharmaceutical and

ig. 6. Arrhenius plot for the calculation of the activation energy and the extrapo-ation of k at 25, 20 and 5 ◦C.

Our results were compared with those reported by Perrotin-runel et al. for THCA [24]. The rate constant k at 100 ◦C and 120 ◦Cas about 5 × 10−4 s−1 and 21 × 10−4, indicating a decarboxylation

ate respectively twice and three times higher than those obtainedor CBDA in this study. Perrotin-Brunel et al. [24] and Wang et al.26] reported a value of A of 3.7 × 108 s−1 and 8.7 × 108 s−1, respec-ively. Our results for A (6.88 × 108 s−1) confirmed that the lowerate constant for the decarboxylation of CBDA is not due to thentropy factor (to which A is related). The data obtained for kf CBDA decarboxylation are also in accordance with the valueseported by Wang et al. [26] at the temperatures considered.

The Arrhenius plot also allowed to extrapolate k and t1/2 ofecarboxylation at 25, 20 and 5 ◦C in order to provide an indica-ion of stability of CBDA in hemp seed oil during common storageonditions (Fig. 6). The decarboxylation rate constant at 25 ◦C k(25)as 1.66 ± 0.99 × 10−7 s−1 and the t1/2(25) was about 49 days (less

han 2 months). Decreasing the temperature by only 5 ◦C, theesulting k(20) was one order of magnitude lower than k(25),.20 ± 5.97 × 10−8, and t1/2(20) was about two times higher thanhat at 25 ◦C, i.e. 87 days (about 3 months). If the hemp seed oils stored in the fridge (5 ◦C), the decarboxylation rate constant(5) becomes 1.37 ± 1.07 × 10−8, which is translated in a t1/2(5) of87 days (about 20 months).

. Conclusions

A highly sensitive and rapid HPLC-UV method was developednd validated for the qualitative and quantitative determination ofhe main cannabinoids present in commercial hemp seed oils. Theeveloped method could be easily employed by any analytical lab-ratory to assess the safety and quality of hemp seed oil before itsommercialization. The analysis can provide not only quantitative

ata of the cannabinoids content, but also qualitative informationbout the purity of the oil (e.g. CBDA/CBD ratio can be taken as aarker of cold pressing and good storage conditions).

[

14 119.2 ± 0.4 –

The decarboxylation kinetics studies of CBDA, which is the mainacid cannabinoid present in hemp seed oil, provided useful sugges-tions for the best storage conditions (5 ◦C) in order to preserve thestability of hemp seed oil for longer time.

Acknowledgements

We would like to thank Crystal Hemp (Lugano, Switzerland) andFarmacia Tundo Alfredo (Alliste, Italy) for financial support.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at https://doi.org/10.1016/j.jpba.2017.11.044.

References

[1] E. Small, Cannabis: a Complete Guide, CR Press, Boca Raton Florida, 2002.[2] A. Smeriglio, E.M. Galati, M.T. Monforte, F. Lanuzza, V. D’Angelo, C. Circosta,

Polyphenolic compounds and antioxidant activity of cold-pressed seed oilfrom finola cultivar of cannabis sativa l, Phytother. Res.: PTR 30 (8)) (2016)1298–1307.

[3] J.C. Callaway, Hempseed as a nutritional resource: an overview, Euphytica140 (1) (2004) 65–72.

[4] C. Da Porto, D. Decorti, F. Tubaro, Fatty acid composition and oxidationstability of hemp (Cannabis sativa L.) seed oil extracted by supercriticalcarbon dioxide, Ind. Crops Products 36 (1) (2012) 401–404.

[5] L.L. Yu, K.K. Zhou, J. Parry, Antioxidant properties of cold-pressed blackcaraway, carrot, cranberry, and hemp seed oils, Food Chem. 91 (4) (2005)723–729.

[6] M.K.F. Grotenhermen, D. Lohmeyer, THClimits for Food: A Scientific Study,nova-Institute, Hürth, Germany, 2015.

[7] S.G.W.D.W. Lachenmeier, Current status of THC in german hemp foodproducts, J. Ind. Hemp 10 (2) (2005) 5–17.

[8] M.C.L. Sarmento, F. Grotenhermen, D. Kruse, Scientifically Sound Guidelinesfor THC in Food in Europe, 2015 (Hürth).

[9] A.W. Zuardi, J.A. Crippa, J.E. Hallak, F.A. Moreira, F.S. Guimaraes, Cannabidiol aCannabis sativa constituent, as an antipsychotic drug, Brazilian journal ofmedical and biological research, Rev. Bras. Pesqui. Med. Biol. 39 (4) (2006)421–429.

10] S. Yamaori, Y. Okamoto, I. Yamamoto, K. Watanabe, Cannabidiol, a majorphytocannabinoid, as a potent atypical inhibitor for CYP2D6, Drug Metab.Dispos. 39 (11) (2011) 2049–2056.

11] A.W. Zuardi, I. Shirakawa, E. Finkelfarb, I.G. Karniol, Action of cannabidiol onthe anxiety and other effects produced by delta 9-THC in normal subjects,Psychopharmacology (Berl) 76 (3) (1982) 245–250.

12] H.M. Harris, K.J. Sufka, W. Gul, M.A. ElSohly, Effects ofdelta-9-Tetrahydrocannabinol and cannabidiol on cisplatin-Inducedneuropathy in mice, Planta Med. 82 (13) (2016) 1169–1172.

13] D.S. Reddy, The utility of cannabidiol in the treatment of refractory epilepsy,Clin. Pharmacol. Ther. 101 (2) (2017) 182–184.

14] A.J. Hampson, M. Grimaldi, J. Axelrod, D. Wink, Cannabidiol and(−)�9-tetrahydrocannabinol are neuroprotective antioxidants, Proc. Natl.Acad. Sci. 95 (14) (1998) 8268–8273.

15] S. Sirikantaramas, F. Taura, Cannabinoids: biosynthesis and biotechnologicalapplications, in: S. Chandra, H. Lata, M.A. ElSohly (Eds.), Cannabis Sativa L. –Botany and Biotechnology, Springer International Publishing, Cham, 2017, pp.183–206.

16] J.W. Fairbairn, J.A. Liebmann, M.G. Rowan, The stability of cannabis and itspreparations on storage, J. Pharm. Pharmacol. 28 (1) (1976) 1–7.

Biomedical Analysis of cannabinoids: A critical review, J. Pharm. Biomed.Anal. 147 (2018) 565–579.

18] M. Pellegrini, E. Marchei, R. Pacifici, S. Pichini, A rapid and simple procedurefor the determination of cannabinoids in hemp food products by gas

http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://dx.doi.org/10.1016/j.jpba.2017.11.044http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0005http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0010http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0015http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0020http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0025http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0030http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0035http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0040http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0045http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0050http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0055http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0060http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0065http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0070http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0075http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0080http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0085http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090

5 and B

[

[

[

[

[

[

[

[26] W.Y.-H. Wang Mei, Avula Bharathi, M. Radwan Mohamed, S. Wanas Amira,van Antwerp John, F. Parcher Jon, A. ElSohly Mahmoud, A. Khan Ikhlas,Decarboxylation study of acidic cannabinoids: a novel approach usingultra-High-Performance supercritical fluid Chromatography/Photodiodearray-Mass spectrometry, Cannabis Cannabinoid Res. 1 (1) (2016) 262–271.

40 C. Citti et al. / Journal of Pharmaceutical

chromatography-mass spectrometry, J. Pharm. Biomed. Anal. 36 (5) (2005)939–946.

19] H.C.F.E. Dussy, M. Luginbuhl, T. Schwerzmann, T.A. Briellmann, Isolation ofDelta9-THCA-A from hemp and analytical aspects concerning thedetermination of Delta9-THC in cannabis products, Forensic Sci. Int. 149 (1)(2005) 3–10.

20] D.W. Lachenmeier, L. Kroener, F. Musshoff, B. Madea, Determination ofcannabinoids in hemp food products by use of headspace solid-phasemicroextraction and gas chromatography–mass spectrometry, Anal. Bioanal.Chem. 378 (1) (2004) 183–189.

21] S. Gandolfi, G. Ottolina, S. Riva, G.P. Fantoni, I. Patel, Complete ChemicalAnalysis of Carmagnola Hemp Hurds and Structural Features of ItsComponents, 2013.

22] EMEA, ICH Topic Q2(R1) Validation of Analytical Procedures: Text andMethodology, 2006.

23] C. Citti, G. Ciccarella, D. Braghiroli, C. Parenti, M.A. Vandelli, G. Cannazza,Medicinal cannabis: principal cannabinoids concentration and their stabilityevaluated by a high performance liquid chromatography coupled to diode

iomedical Analysis 149 (2018) 532–540

array and quadrupole time of flight mass spectrometry method, J. Pharm.Biomed. Anal. 128 (2016) 201–209.

24] H. Perrotin-Brunel, W. Buijs, J. v. Spronsen, M.J.E. v. Roosmalen, C.J. Peters, R.Verpoorte, G.-J. Witkamp, Decarboxylation of �9-tetrahydrocannabinol:kinetics and molecular modeling, J. Mol. Struct. 987 (1) (2011) 67–73.

25] T. Veress, J.I. Szanto, L. Leisztner, Determination of cannabinoid acids byhigh-performance liquid chromatography of their neutral derivatives formedby thermal decarboxylation, J. Chromatogr. A 520 (1990) 339–347.

http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0090http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0095http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0100http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0105http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0110http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0115http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0120http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0125http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130http://refhub.elsevier.com/S0731-7085(17)32236-7/sbref0130

Analysis of cannabinoids in commercial hemp seed oil and decarboxylation kinetics studies of cannabidiolic acid (CBDA)1 Introduction2 Materials and methods2.1 Chemicals and reagents2.2 HPLC-UV and HPLC-MS analysis2.3 Preparation of standard solutions2.4 Method validation2.4.1 Specificity and selectivity2.4.2 Linearity2.4.3 Limit of detection (LOD) and limit of quantification (LOQ)2.4.4 Precision and accuracy2.4.5 Recovery2.4.6 Matrix effect2.4.7 Dilution integrity2.4.8 Stability

2.5 Authentic hemp seed oil sample preparation2.6 Decarboxylation studies2.6.1 Open reactor2.6.2 Closed reactor2.6.3 Calculation of kinetic parameters and free energy barriers of decarboxylation

3 Results and discussion3.1 Chromatographic conditions3.2 Method validation3.3 Authentic hemp seed oil sample preparation and analysis3.4 Decarboxylation studies

4 ConclusionsAcknowledgementsAppendix A Supplementary dataReferences