REV. CHIM. (Bucharest) ♦ 63 ♦ No. 3 ♦ 2012 http://www.revistadechimie.ro 293
Long - term Storage and Cannabis Oil Stability
IRENNE GABRIELA TROFIN1*, GABRIEL DABIJA1, DANUT–IONEL VAIREANU2, LAURENTIU FILIPESCU3
1General Inspectorate of Romanian Police, Central Laboratory for Drug Analysis and Profiling, Bucharest, Romania.
2University Politehnica of Bucharest, Faculty of Applied Chemistry and Materials Science, Department of Applied Physical
Chemistry and Electrochemistry, 313 Splaiul Independentei, 060042, Bucharest, Romania
3University Politehnica of Bucharest, Faculty of Applied Chemistry and Materials Science, Department Technology of Inorganic
Substances and Environmental Protection, 313 Splaiul Independentei, 060042, Bucharest, Romania
The paper presents the results of an experimental study regarding the stability of cannabis oil during its long-
term storage in different conditions. The content of major cannabinoids, namely tetrahydrocannabinol (
Δ
9–
THC), cannabinol (CBN), and cannabidiol (CBD) contained in two batch samples of cannabis oil seizures
made by criminal prosecution authorities from Romania was measured during their storage over a period of
four years in darkness at 4°C and in laboratory light at 22°C. The results revealed a steadily decay of
Δ
9–THC
over the entire storage period from a very high initial content up to a relatively low final content. A slight
difference regarding the degree of decay of
Δ
9–THC between the two storage conditions was recorded,
meaning that this is more pronounced when the samples were exposed to light at 22°C. The same trend was
recorded for CBD. As expected, the content of CBN increases during storage and the increase is higher when
the samples were exposed to light at 22°C.
Keywords: cannabis, oil, decay, cannabinoid
Cannabis oil, often called hashish oil, is a liquid cannabis
product with a high content of tetrahydrocannabinol (Δ9–
THC) obtained by extraction from either herbal cannabis
or from cannabis resin. Although such a product requires
more complex preparation methods than other cannabis
products, it is often preferred by drug dealers because they
may traffic more psychoactive material in a smaller
quantity of cannabis product [1].
As with other cannabis products, cannabis oil has
different stability in various environmental conditions [2-
4]. The psychoactive component, namely Δ9–THC, has a
relatively high instability when the cannabis products are
exposed to air, light or acidic environments [5-7]. Its
instability to heat has been also demonstrated [8]. It is
widely accepted that the main pathway of cannabis
products deactivation is the conversion of Δ9–THC to
cannabinol (CBN). After five years of storage in an ethanol/
propylene glycol solution of a Δ9–THC sample, it was
proposed a correlation between the oxidative derivatives
of this cannabinoid formed during its degradation to the
less psychoactive component CBN [9].
* email: irennetrofin@yahoo.com
REV. CHIM. (Bucharest) ♦ 63 ♦ No.3 ♦ 2012http://www.revistadechimie.ro294
Other major cannabinoid present in cannabis products, namely cannabidiol may also suffer changes depending on
storage conditions. In this respect, under acidic conditions it may transform into Δ9–THC by acid-catalyzed cyclization
and, in the presence of oxygen, is oxidized to monomeric and dimeric hydroxyquinones [10].
Although the cannabinoids stability was intensively
studied previously, most of the researches were done on
pure solutions, which can have a different behaviour
towards actual cannabis products. Therefore, the objective
of this paper is to explore experimentally the influence of
storage conditions such as temperature and light on the
stability of the major cannabinoids in the cannabis oil.
Experimental part
Chemicals and reagents
All chemicals and reagents used for samples preparation
and analysis were of analytical grade from Merck
(Darmstadt, Germany). The etalons of Δ9–tetrahydro-
cannabinol (Δ9–THC), cannabidiol (CBD), and cannabinol
(CBN) were purchased from Switzerland. The ultrapure
water used in HPLC analyses was prepared in-house using
a Millipore system, model Milli-Q Integra 3.
Cannabis oil samples
Cannabis oil from two different seizures (marked with
symbols U1 and U2) made by criminal prosecution
authorities from Romania and provided by Central
Laboratory for Drug Analysis and Profiling were subject to
experimental investigation. The samples of cannabis oil
were supplied in small bottles and have a black-brown
colour and a high consistency comparable to that of a paste.
The samples were stored in the darkness at 4°C and in the
laboratory light at 22°C for four years. At regular intervals,
namely at every three months, samples were taken for
analysis in order to determine the content of their major
cannabinoids (Δ9–THC, CBD, and CBN).
Methods
The procedure that led to the sample preparation for
analysis consisted of extracting 0.05 g of cannabis oil in 20
mL of a methanol-chloroform (9:1, v/v) mixture. Thus, the
samples were shaken for 30 min and then placed in an
ultrasonic bath at ambient temperature for 15 min in order
to increase the cannabinoids extraction rates. The extracts
were filtered and some aliquots (0.6 mL) of the filtrates
were transferred to a 4 mL vials and then evaporated to
dryness by oven evaporation, only up to 80°C for prevention
of any decomposition reactions. Then, the vials were put
into a heating unit at 220°C for 12 min when the traces of
tetrahydrocannabinolic acid (THCA) are decarboxylated.
Decarboxylation is particularly required for the
determination of the entire content of Δ9–THC of the
sample. Before analyses, the residues were extracted in
1.5 mL extraction solvent (methanol-chloroform 9:1, v/v).
After this, the samples were subject to analyses of the
major cannabinoids content (Δ9–THC, CBD, and CBN) [1].
Analytical protocol
Extracts obtained by procedure described above have
been subject to analytical investigations through
instrumental methods (GC–FID – Gas Chromatography–
Flame Ionization Detector and HPLC – High Performance
Liquid Chromatography) in order to find out the content in
major cannabinoids (Δ9–THC, CBD, and CBN).
GC–FID analyses were carried out on a 7890A gas
chromatograph with a flame ionization detector.
Separation was achieved on a fused silica capillary column
(HP-5MS, 30 m×0.32 mm i.d., 0.25μm film thickness, J&W
Scientific, Folsom, CA, USA). Temperature program: 150°C
hold for 1 min, 10°C/min to 280°C, hold for 5 min. The
injection port and interface temperature were 250°C and
300°C, respectively. Split injection mode was used (20:1)
and hydrogen, with a flow rate of 30 mL per min, was used
as carrier gas [11].
HPLC analyses were carried out on an Agilent 1100
Series HPLC chromatograph equipped with a quaternary
pump, autosampler, column oven and diode-array detector
(DAD) UV Lamp ON (223 nm). Chromatography was
achieved on a 250 mm × 4.6 mm i.d., 5 μm Hypersil ODS
column. The HPLC operates with constant flow at 1 mL
mobile phase (acetonitrile 37.5% and ultrapure water) per
minute.
Results and discussions
GC–FID and HPLC chromatograms revealed a very high
content of ΔΔ
ΔΔ
Δ9–THC in the cannabis oil samples and
subsequently, a very high potency of these type of cannabis
product compared to other cannabis products such as
herbal cannabis and cannabis resin. The difference
between the two samples regarding their potency is
significant and comes probably from the use of different
procedures for their obtaining. The difference could be due
to the type of solvent used for extraction, number of
extractions, conditions of extractions etc. These results
indicate, on the one hand, that the samples were prepared
in different clandestine laboratories, improvised by different
drug traffickers and, on the other hand, that the samples
went through different routes of trafficking originating from
different geographical areas. Following a detailed analysis
of such data, the prosecution authorities could identify
trafficking routes, and finally the places where this type of
drugs were made.
REV. CHIM. (Bucharest) ♦ 63 ♦ No. 3 ♦ 2012 http://www.revistadechimie.ro 295
The experimental results concerning the stability of the
major cannabinoids indicate a small but constant
difference between cannabinoids content of the cannabis
oil depending on storage conditions. Figures 1 and 2 show
the variation of the major cannabinoids content in the two
samples of cannabis oil as a function of time and storage
conditions. As one can see, in both samples the Δ9–THC
content decreases during storage and is always higher in
the samples stored in the darkness at 4°C than in the
samples stored in the laboratory light at 22°C. The same
trend can be also observed for CBD content. As it was
expected, the CBN content increases during storage period
and is always higher in the samples stored in laboratory
light at 22°C than in the samples stored in darkness at 4°C.
The results obtained for the two samples derived from the
two different seizures regarding the evolution of the major
cannabinoids content during storage in different conditions
are presented in table 2. The results revealed a steadily
decay of Δ9–THC over the entire storage period. Moreover,
the decay of Δ9–THC in the samples exposed to light at
22°C is higher than in the samples stored in the darkness at
4°C. In this respect, when the samples from seizure U1
were stored in the darkness at 4°C, 21.6% of Δ9–THC (fig. 1
a) was lost in the first year with an average loss of 4.4%
every tree months, 21.83% in the second year with an
average loss of 5.46%, 21.64% in the third year with an
average loss of 5.41%, and 18.69% in the fourth year with
an average loss of 4.67%. When the samples from the same
seizure were stored in the laboratory light at 22°C, 23.16%
of Δ9–THC was lost in the first year with an average loss of
5.79% every tree months, 23.25% in the second year with
an average loss of 5.81%, 22.99% in the third year with an
average loss of 5.75%, and 20.45% in the fourth year with
an average loss of 5.11%. Finally, after four years of storage,
the samples stored in the darkness at 4°C lost 83.75% of
Δ9–THC and the samples stored in the laboratory light at
22°C lost 89.85% of Δ9–THC ( 6.1% higher).
The variation of CBN (fig. 1 b) of the same samples over
the storage period indicates that when these were stored
in the darkness at 4°C, 59.03% of CBN was formed in the
first year with an average gain of 14.76% every three
months, 16.49% in the second year with an average gain
of 4.12%, 4.76% in the third year with an average gain of
1.19%, and 3.01% in the fourth year with an average gain of
0.75%. When the samples from the same seizure were
stored in the laboratory light at 22°C, 65.29% of CBN was
formed in the first year with an average gain of 16.32%
every three months, 14.18% in the second year with an
average gain of 3.54%, 3.72% in the third year with an
Table 1
THE INITIAL CONTENT OF MAJOR CANNABINOIDS
IN CANNABIS OIL
Table 2
EVOLUTION OF THE MAJOR CANNABINOIDS CONTENT IN
CANNABIS OIL DURING STORAGE IN DIFFERENT CONDITIONS
Fig. 1 Variation of cannabinoids content in
cannabis oil: (a) Δ9–THC decay;
(b) CBN formation; (c) CBD decay
REV. CHIM. (Bucharest) ♦ 63 ♦ No.3 ♦ 2012http://www.revistadechimie.ro296
Fig. 2 Decay degree of Δ9–THC in
cannabis oil derived from:
(a) seizure U1; (b) seizure U2
average gain of 0.93%, and 3.95% in the fourth year with an
average gain of 0.99%. Finally, after four years, the samples
stored in the darkness at 4°C gained 83.29% of CBN and
the samples stored in the laboratory light at 22°C gained
87.15% of CBN (3.86% higher). The same trend was also
recorded for the second cannabis oil sample.
Comparing the decay degree of Δ9–THC in the first year
of storage period of cannabis oil with the formation degree
of CBN in the same year, it can be seen that the latter yield
is with about 35% higher than the first yield. In this respect,
the changes regarding the content of CBN during the
storage period can not be entirely correlated with the
chemical and/or biochemical decay processes of Δ9–THC
to CBN. Some other unknown variables beyond the control,
such as the origin place, degradation already started during
the trafficking transports play also an important role. Also,
other degrading routes of other cannabinolic compounds
must be considered as contributors to the overall increase
of CBN content upon long-term storage.
The variation of CBD (fig. 1 c) in the same samples over
the entire storage period indicates that in the case of
samples from seizure U1 stored in the darkness at 4°C,
11.03% of CBD was lost in the first year with an average
loss of 2.76% every tree months, 12.15% in the second
year with an average loss of 3.04%, 7.47% in the third year
with an average loss of 1.87%, and 9.54% in the fourth year
with an average loss of 2.39%. When the samples from the
same seizure were stored in the laboratory light at 22°C,
13.45% of CBD was lost in the first year with an average
loss of 3.35% every three months, 12.05% in the second
year with an average loss of 3.01%, 6.81% in the third year
with an average loss of 1.7%, and 8.5% in the fourth year
with an average loss of 2.12%. Finally, after four years of
storage, the samples stored in the darkness at 4°C lost
40.18% of CBD and the samples stored in the laboratory
light at 22°C lost 44.8% of CBD (with less 0.62%). The same
trend was recorded for all cannabis oil samples.
Analyzing these results it can be seen that the decay
degree of CBD in the first year of storage period of cannabis
oil is about 10% (approximately quarter the difference of
35% between the decay degree of Δ9–THC and respectively,
the formation degree of CBN) in the case of the samples
stored in the darkness at 4°C and about 19%
(approximately half the difference of 40% between the
decay degree of Δ9–THC and respectively the formation
degree of CBN) was observed in the case of the samples
stored in the laboratory light at 22°C. These results suggest
a different degrading rout of CBD, probably dependent on
the storage conditions. When the samples were stored in
the darkness one of the degradative route could be the
biochemical cyclization of CBD to Δ9–THC, followed by
the oxidative decay of Δ9–THC to CBN. In addition, when
the samples were exposed to light, CBD might achieve
photo-reactive properties and transforms into Δ9–THC [10].
A pseudo zero-order kinetic was used (fig. 3) in order to
calculate the kinetic parameters of the Δ9–THC decay such
as the rate constant (k’), the half-time (t1/2), and the decay
rate (v). The linear regression parameters and the
correlation coefficients are presented in the table 3. As
can be seen from the table 4, both rate constant and decay
rate are higher in the samples stored in the laboratory light
at 22°C than those stored in the darkness at 4°C. The values
of the half-time corresponding to the samples stored in the
laboratory light at 22°C are smaller than those stored in the
darkness at 4°C. These results suggest a higher rate of Δ9–
THC decay in the cannabis oil stored in normal conditions
(natural light and ambiental temperature) than in the case
of special storage conditions (darkness and low
temperature).
Fig. 3 Pseudo zero-order kinetic of Δ9–THC decay in the cannabis oil; the solid line represents the linear regression of data
corresponding to 4oC and darkness storage conditions and, the dashed line represents the linear regression of data corresponding
to 22oC and laboratory light storage conditions
REV. CHIM. (Bucharest) ♦ 63 ♦ No. 3 ♦ 2012 http://www.revistadechimie.ro 297
Table 4
KINETIC PARAMETERS OF Δ9-THC DECAY
CALCULATED FROM A PSEUDO-ZERO
ORDER KINETIC
Table 3
LINEAR REGRESSION PARAMETERS
AND CORRELATION COEFFICIENTS
Conclusions
Chemical characterization of the cannabis oil samples
derived from two different seizures revealed a very high
content of Δ9–THC compared with other cannabis products
such as the herbal cannabis and cannabis resin. The
difference between the potency of the two samples
suggests a different preparation method or a different route
of trafficking, which could finally indicate the places where
the samples were prepared.
The experimental results regarding the stability of the
major cannabinoids species revealed differences that
should be taken into consideration between cannabinoids
content of the cannabis oil are related to the storage
conditions. Thus, the results revealed a steady decay of
Δ9–THC over the entire storage period. Moreover, the decay
of Δ9–THC in the samples exposed to light at 22°C is more
pronounced than in the samples stored in the darkness at
4°C. The content of CBN increases during storage, and
increase is more pronounced for the samples exposed to
light at 22°C than those stored in the darkness at 4°C. These
results are in part consistent with those obtained for Δ9–
THC. The CBD content decreases during storage especially
for samples exposed to light at 22°C. This evolution could
be explained by considering the biochemical cyclization
of CBD to Δ9–THC, followed by the decay of Δ9–THC to CBN
as a degrading route for the samples stored in the darkness
at 4°C and, both biochemical and photochemical
cyclization of CBD to Δ9–THC followed by decay of Δ9–THC
to CBN as a degrading routes for the samples exposed to
light at 22°C.
The decay of Δ9-THC takes place up on a pseudo zero-
order kinetic and the calculated values of the kinetic
parameters suggest a higher rate of Δ9–THC decay in
normal storage conditions such as light and ambiental
temperature than in special ones such as darkness and
low temperature.
References
1. *** United Nations Office on Drugs and Crime (UNODC), Manual
for use by national drug analyses laboratories, 2009.
2. MARTONE G., DELLA CASA E., Forensic. Sci. Int., 47, 1990, p. 147.
3. LINDHOLST C., AUSTRALIAN J. Forensic. Sci., 42, no 3, 2010, p. 181.
4. ROSS S. A., ELSOHLY M. A., Bull. Narc., 49, no.50, 1997, p. 139.
5. MECHOULAM R., HANUŠ L., Chem. Phys. Lipids., 108, 2000, p. 1.
6. GOLDING F.S., DÍAZ-FLORES E.J., DÍAZ R.C., Ann. Clin. Lab. Sci.,
28, no.3, 1998, p. 160.
7. DUSSY F.E., HAMBERG C., LUGINBÜHL M., SCHWERZMANN T.,
BRIELLMANN T.A., Forensic Sci. Int., 149, 2005, p. 3.
8. TROFIN I.G., VLAD C.C., DABIJA G., FILIPESCU L., Rev.
Chim.(Bucharest), 62, no. 6, 2011, p. 639
9. CARBONE M., CASTELLUCCIO F., DANIELE A., SUTTON A ., LIGRESTI
A., DI MERZO V., GAVAGNIN M., Tetrahedron, 66, no. 49, 2010, p. 9497.
10. MECHOULAM R., HANUŠ L., Chem. Phys. Lipids., 121, 2002, p. 35.
11. VLAD C.C., TROFIN I.G., DABIJA G., FILIPESCU L., Rev. Chim.
(Bucharest), 62, no. 9, p. 916
Manuscript received: 16.08.2011
