Pyrolytic formation of polycyclic aromatic hydrocarbons from sesquiterpenes

Article abstract of DOI:10.1016/j.foodchem.2012.06.033

The products of the pyrolysis of four sesquiterpenes, β-caryophyllene, α-cedrene, longifolene and valencene, have been examined. Pyrolysis was carried out at 300, 400 and 500 °C, the products determined by GC-MS and then examined for similarities and diff

Full text of DOI:10.1016/j.foodchem.2012.06.033

Food Chemistry 135 (2012) 1316–1322
Contents lists available at SciVerse ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Pyrolytic formation of polycyclic aromatic hydrocarbons from sesquiterpenes
b
b
George W. Francis a, , Alfred A. Christy , Jostein Øygarden
⇑
a Department of Chemistry, University of Bergen, Allégaten 41, 5007 Bergen, Norway
b Department of Science, Faculty of Engineering and Science, University of Agder, Serviceboks 422, N-4604 Kristiansand, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
The products of the pyrolysis of four sesquiterpenes, b-caryophyllene, a-cedrene, longifolene and valen-
Received 20 March 2012
Received in revised form 18 May 2012
Accepted 18 June 2012
cene, have been examined. Pyrolysis was carried out at 300, 400 and 500 °C, the products determined by
GC–MS and then examined for similarities and differences using multivariate data analysis. Analysis
showed that longifolene was most resistant and caryophyllene least resistant to pyrolysis with cedrene
and valencene occupying intermediate positions. While the compounds were largely unchanged at
300 °C, polycyclic aromatic hydrocarbons (PAHs) were major components of the pyrolysates at 400
and 500 °C. No less than nine of the 16 EPA priority pollutants were present in the pyrolysates at the
higher temperatures.
Available online 28 June 2012
Keywords:
Pyrolysis
Sesquiterpene
Polycyclic aromatic hydrocarbon
b-Caryophyllene
Ó 2012 Elsevier Ltd. All rights reserved.
a-Cedrene
Longifolene
Valencene
PAH
1. Introduction
Chempakam, & Zachariah, 2008). In this way they are deliberately
introduced to the food chain in addition to their occurrence in min-
Sesquiterpenoids are products of the terpenoid, or isoprenoid,
pathway and have been identified in many plants. This biosyn-
thetic pathway is a requirement for photosynthesis since both
chlorophyll and carotenoids, the pigments involved in light har-
vesting, depend on the availability of geranylgeranyl pyrophos-
phate or its equivalent for their assembly (Bogorad, 1976 &
Britton, 1976). All green plants must therefore contain at least
transiently the C-15 precursor, farnesyl pyrophosphate, which is
also the precursor for the sesquiterpenoids which are commonly
identified in these organisms.
The sesquiterpenoids are the most varied in structure of the ter-
penoids and even forty years ago over 1000 members of the category
were identified as belonging to 30 main skeletal classes and some 70
less common types (Devon & Scott, 1972). New skeletons and indi-
vidual compounds continue to be reported (Fraga, 2011). This diver-
sity results from a variety of sesquiterpene synthases which utilise
farnesyl pyrophosphate as a precursor for cyclisation, methyl shifts,
hydride shifts, Wagner–Meerwein rearrangements etc., in different
combinations to produce everything from acyclic to tetracyclic
structures, which are then liable to changes in oxidation level and
functionalisation (Degenhardt, Köller, & Gershenzon, 2009).
Sesquiterpenoids are important components of many spices
where they contribute to both the taste and aroma (Parthasarathy,
or amounts in a fortuitous manner in other foods. Indeed, genetic
manipulation to increase plant sesquiterpenoid biosynthesis with
a view to improving plant traits and crop protection has recently
been reviewed (Yu & Utsumi, 2009). A good example of this ap-
proach is to be found in a study of sesquiterpenoids in rice (Cheng
et al., 2007).
A previous study (Christy, Lian, & Francis, 2011) showed that
the pyrolysis of steroid hormones resulted in the formation of
polycyclic aromatic hydrocarbons (PAHs) and that the pyrolysates
varied according to the steroid involved. It was decided to extend
this work to sesquiterpenes on the basis they are common compo-
nents in foods and the fact that they might contribute in a similar
way to the formation of PAHs. Most sesquiterpene skeletons are
either bicyclic or tricyclic and two compounds of each type were
included in this study. Caryophyllene and valencene have bicyclic
structures, while cedrene and longifolene have tricyclic structures
(Fig. 1). The choice of sesquiterpene hydrocarbons rather than of
oxygen-containing sesquiterpenoids was intended to simplify the
results.
Initial experiments showed that the pyrosylates contained ma-
jor amounts of PAHs, the carcinogenic and mutagenic properties of
which have been established (Wornat, Braun, Hawiger, Logwell, &
Sarofim, 1990). Mechanisms for the PAH induced mutogenesis and
tumorigenesis have been proposed and in most cases the first step
is the oxidation of the PAH by cytochrome 450 enzymes to provide
electrophilic species, diolepoxides, that then react to form DNA
⇑
Corresponding author.
E-mail address: george.francis@kj.uib.no (G.W. Francis).
0308-8146/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2012.06.033

G.W. Francis et al. / Food Chemistry 135 (2012) 1316–1322
1317
2. Materials and methods
2.1. Samples, pyrolysis and GCMS analysis
b-Caryophyllene, -cedrene, logifolene and valencene were ob-
a
tained from commercial sources. All solvents were of HPLC quality
and used as received.
The ampoules used were made from borosilicate glass and the
main body had approximate internal dimensions of 17 Â 80 mm
with a wall thickness of 1.2 mm. Ampoules had initially an exten-
sion which provided access to the body of the ampoule through a
narrow neck. Each sample was prepared by introducing the com-
pound (3–4 mg) to be pyrolysed through the extension to the am-
poule, replacing the air in the ampoule with nitrogen and then
closing it by melting the neck to provide a gas-tight seal. The use
of a nitrogen atmosphere in the ampoule prevents oxidation and
thus simplifies the analysis of the pyrolysate.
Sealed ampoules were placed directly into an electric oven pre-
heated to the intended pyrolysis temperature, 300, 400 or 500 °C,
where they were allowed to remain for a period of 4 h. After this
time the ampoules were removed and cooled in the dark overnight.
The cooled ampoules were carefully opened and extracted with
1 ml portions of hexane. Hexane has been shown to be a suitable
solvent for PAHs (García-Falcón, Cancho-Grande, & Simal-Gándara,
2005; Khatmullina, Safarova, Kudasheva, & Kitaeva, 2012) and pre-
liminary experiments indicated that this solvent dissolved all of
the pyrolysates in the present case.
The GC–MS analysis was carried out on a Hewlett Packard (HP)
5890 series II gas chromatograph connected to a HP 5971A quad-
rupole mass spectrometer. The components were separated on a
Fig. 1. (a) Structures of the sesquiterpenes in the present study. (b) Indication of
how bond scission might lead to variable 57. (c) Variables found to correlate with
specific pyrolysis substrates. (d) Ring systems (group variables) found in pyroly-
sates at 500 °C.
30 m  0.25 mm I.D fused silica column coated with 0.25
lm
100% dimethyl polysiloxane stationary phase. The oven was pro-
grammed from 50 °C, after a 5 min hold time, at 4 °C/min up to
250 °C where the temperature was held for a further 10 min.
Adjustments were made to the program to improve resolution
where required.
adducts which can induce both preneoplastic and tumorigenic
mutations (Chakravarti et al., 2008). The mutagenicity of the ad-
ducts has been related to their relative repair efficiencies in mam-
malian cells (Lagerqvist et al., 2011).
The total ion chromatograms of the pyrolysates were acquired
using the HP G1034C chemstation program. The peaks in the
While the amounts of spices added to individual foods sub-
jected to roasting, frying, grilling and other forms of heat treatment
may be limited, the continuing exposure to such foods over ex-
tended times must clearly be a cause for concern. It has been
shown that lifetime, rather than recent, intake of cooked meat
and the PAHs therein, increase the risk of breast cancer (Steck
et al., 2007). Lifetime consumption of meat and cooking practises
has shown that regular consumption of well-done red meat in-
creases the risk of prostrate cancer (John, Stern, Sinha, & Koo,
2011). Higher cooking temperatures and longer cooking times
are known to increase mutagenicity in cooked meat (Barrington
et al., 1990). While the temperatures met with in normal cooking
methods should not exceed 300 °C, the occurrence of local hot
spots under non-ideal conditions and/or open flame cooking make
it interesting to study the effects at this and higher temperatures.
The aim of this work was thus to establish which PAHs and in what
amounts these were formed at temperatures of 300, 400 and
500 °C. It was also regarded as being important to see to what
extent the sesquiterpene structure affected the PAH formation.
Pyrolysis may be defined as the transformation of materials by
heat treatment and a variety of techniques are used to accomplish
this under controlled conditions (Voorhees, 1984). While the initial
process in pyrolysis involves the cleavage of molecules into smaller
unstable species, these over time react in a process called pyrosyn-
thesis to form molecular products of greater mass (Lee, Novotny, &
Bartle, 1981). It is this latter process that results in the formation of
PAHs and investigation into the mechanisms involved remains
under active investigation (Shukla & Koshi, 2010).
total ion chromatogram were then identified using
a mass
spectral library (NIST 98). Library identifications were further
investigated and confirmed by examining and comparing the
fragmentation patterns obtained with the suggested molecular
structure.
2.2. Data analysis
The large number of compounds in the pyrolysates makes it ex-
tremely difficult to carry out comparisons between the different
starting materials and different temperatures used. Similarities,
dissimilarities and relationships between the products formed
were thus identified by means of principal component analysis
(PCA), a chemometric technique for which the theoretical basis
for which can be found elsewhere (Christy, Ozaki, & Gregoriou,
2001).
The peaks in the total ion chromatograms from all pyrolysates
were area-integrated and recorded. A table was then prepared uti-
lising the component peak areas as variables and the pyrolysis
temperatures as samples. This table was then used to obtain infor-
mation about the products found in the pyrolysates and as a data
matrix for the multivariate data analysis which was carried out
using the SIRIUS multivariate data analysis programme (Kvalheim
& Karstang, 1997) on the data matrix of area percentages of the
components. Since the table already contains the percentage data
for the individual components found in the total ion chromato-
grams pre-processing of the data was unnecessary. Additional

1318
G.W. Francis et al. / Food Chemistry 135 (2012) 1316–1322
analysis was carried out after grouping compounds into classes
according to structural criteria.
from this analysis to avoid domination of the results. The two
principal components identified explained a total of 98.9% of the
variance. PC 1 clearly separates caryophyllene from the other com-
pounds and can thus be identified with the fact that the pyrolysis
has proceeded further in this case than for the other compounds.
Cedrene is clearly associated with variable 57 and well separated
from the other components along PC 2.
3. Results
3.1. General
No unchanged starting material was present in the pyrolysates
at 500 °C. A bi-plot including a score plot and a loading plot of the
pyrolysis products at 500 °C is also provided in Fig. 4b. PC 1 sepa-
rates valencene and cedrene from each other where variable 89 is
associated with valencene while variable 47 is connected to ced-
rene. Caryophyllene and longifolene are separated from the other
compounds along PC 2 and are associated with the variables 40
and 41.
The results of pyrolysis were also examined by classifying the
products of pyrolysis into a total of 17 structural groups. The first
of these groups corresponded to acyclic compounds, while the
groups 2–15 were based on the main ring structure present in each
case (see Table 3). The final two groups were group 16 which con-
sisted of rearranged sesquiterpoid structures (C15H24) while group
17 was assigned to unchanged starting material. A bi-plot using
starting compound as samples and integrated group peak areas
as variables was prepared for each individual temperature. While
the bi-plots at 300 and 400 °C provided little additional informa-
tion, that for 500 °C was somewhat more informative and is repro-
duced as Fig. 5. This showed that caryophyllene had a central
position on the plot and also fell close to many of the variables.
Longifolene, valencene and cedrene on the other hand are found
at positions towards the edge of the plot at well separated angular
positions. Longifolene (variable 14), valencene (variables 6 and 9)
and cedrene (variable 4) were clearly associated with particular
variables.
The gas chromatograms obtained increased in complexity
according to the pyrolysis temperature involved. At the one ex-
treme the chromatogram obtained for valencene after applying
the conditions of pyrolysis at 300 °C showed only unchanged start-
ing material, while at the other the chromatogram obtained on
pyrolysis of longifolene at 500 °C, reproduced in Fig. 2, included
some 100 different peaks.
PAHs were not observed in any of the pyrolysates at 300 °C, or
in that of longifolene at 400 °C. In all other cases considerable
amounts of PAH were found in the pyrolysates. Table 1 shows
the polycyclic aromatic hydrocarbons present in greater than
0.1% of the total pyrolysates. Compounds identified by the EPA as
priority pollutants are italicised in the table.
Table 2 provides the amounts of substituted and unsubstituted
PAHs found in the pyrolysates at the different temperatures. The
increased formation of unsubstituted PAHs at higher temperatures
is significant since these are known to carry greater mutagenic
activity (Wornat et al., 1990).
3.2. Chemometric results
Analysis of the total data matrix using temperatures as samples
and peak areas as variables revealed that two principal compo-
nents could explain 98.9% of the total variance. A scoreplot show-
ing the scores of all of the samples is shown in Fig. 3. It is
immediately obvious that the objects are divided into two groups
along the x-axis (PC 1). All objects at 500 °C plus caryophyllene
at 400 °C are collected loosely on the right side of the plot, while
the remaining objects are tightly grouped to the left of the origin.
This may be interpreted as reflecting the fact that while all com-
pounds are extensively decomposed after pyrolysis at 500 °C,
caryophyllene alone is seriously changed by pyrolysis at 400 °C.
There appears to be only limited difference in the final products
of the pyrolysis of the different compounds at 500 °C.
4. Discussion
4.1. General
The results of the principal component analysis may be dis-
cussed in relation to Table 1 which examines the PAHs found at
the different temperatures. The finding of very large conversions
to PAH at 500 °C is disturbing in view of the fact that these com-
pounds are known to be carcinogenic (Luch, 2005). Even the results
from the pyrolysis of the compounds other than longifolene at
400 °C are disquieting. Fig. 1 indicates clearly that caryophyllene
is much more sensitive to pyrolysis than the other sesquiterpenes
examined since it reacts at 400 °C, while the others appear to be
much less affected at this temperature. A further conclusion from
Fig. 4 is that in general terms there is little difference in the
pyrolysates obtained at 500 °C, although the valencene pyrolysate
is somewhat separated and thus somewhat different from the
others.
A bi-plot from the principal component analysis combining a
score plot and a loading plot of the pyrolysis products at 400 °C
is shown in Fig. 4a. The data for unchanged material was excluded
The fact that caryophyllene is more sensitive to pyrolysis paral-
lels similar findings that this compound is among the most sensi-
tive sesquiterpenes to ozonolysis (Shu
& Atkinson, 1994).
Caryophyllene has long been known to be readily rearranged under
a variety of conditions (Aebi, Barton, Burgstahler, & Lindsey, 1954;
Parker, Raphael, & Roberts, 1965; Schulte-Elte & Ohloff, 1971) and
a general review on the chemistry of cayophyllene is available
(Collado, Hanson, & Macias-Sanchez, 1998).
The caryophyllene pyrosylate at 400 °C is well separated from
the other compounds along PC 1 (Fig. 5a) corresponding to the
much greater conversion seen in Table 2. The identification of ced-
rene with variable 57 may be related to the fact that the compound
contains a cyclohexene ring annealed to two five-member rings.
Fig. 2. Total ion chromatogram (TIC) for longifolene pyrolysed at 500 °C.

G.W. Francis et al. / Food Chemistry 135 (2012) 1316–1322
1319
Table 1
PAH compounds identified in the pyrosylates of caryophyllene (K), cedrene (S), longifolene (L) and valencene (V) at prolysis temperatures of 400 and 500 °C. Structural groups are
indicated (Group) as are individual identification numbers (ID No.) as used in the multivariate data analysis. Molecular ions are indicated in brackets after the compound names.
The numbers in the vertical columns indicate the percentage area of the peaks. The column for longifolene at 400 °C is empty but included for clarity.
Group
ID No.
Compound name
400 °C
500 °C
L
K
V
S
L
K
V
S
5
85
87
88
89
90
91
92
93
94
Azulene (128)
0.80
1.58
0.72
0.36
0.41
7-Ethyl-1,4-dimethylazulene (184)
4,8-Dimethyl-6-phenylazulene (232)
Benzcycloheptatriene (142)
Indane (118)
Indene (116)
X-Methylindane (132)
X-Methylindene (140)
X,Y-Dimethylindene (144)
X,Y-Dimethylindane (146)
X,Y,Z-Trimethylindane (160)
2-Ethylindane (146)
3-(2-Methylpropenyl)-1H-indene (170)
1,23-Trimethylindene (158)
1,2-Dihydro-3-methylnaphthalene (144)
Tetrahydro-1,1,6-trimethylnaphthalene (174)
Tetrahydro-X,Y-dimethylnaphthalene (160)
1,4-Dihydro-1,4-methanonaphthalene (142)
Tetrahydro-tetramethylnaphthalene (186)
3-(1,1-Dimethylethyl)-1,2-dihydronaphthalene (186)
Naphthalene(128)
X-Methylnaphthalene (142)
X,Y-Dimethylnaphthalene (156)
X,Y,Z-Trimethylnaphthalene (170)
1,2,3,4-Tetramethylnaphthalene (184)
1-Ethylnaphthalene (156)
X-Isopropenylnaphthalene (170)
1-Isopropenylnaphthalene (168)
1,2,5,6-Tetramethylacenaphthene (208)
Acenaphthene (154)
Biphenyl (154)
X-Methylbiphenyl (168)
0.39
0.27
7.24
6
7
0.81
9.13
0.71
1.25
0.89
3.31
1.63
0.28
7.76
0.39
1.00
0.48
3.60
1.35
0.23
19.46
0.71
1.02
0.95
2.29
1.02
0.86
0.78
3.20
1.59
0.23
0.52
1.73
3.62
1.87
1.83
0.54
0.21
0.69
95
96
98
0.15
0.12
0.11
0.11
100
102
103
104
105
106
108
109
110
111
112
113
114
115
116
118
119
120
125
126
127
128
130
133
134
137
138
139
140
141
143
0.19
0.21
8
9
0.37
0.39
0.49
0.29
0.55
0.46
0.40
0.55
2.49
0.17
0.60
1.58
5.37
0.85
0.18
0.46
0.49
0.14
10.14
4.50
8.80
4.91
14.36
7.85
0.18
1.53
0.34
1.39
7.33
3.13
1.19
0.11
1.21
0.10
0.44
0.96
0.44
1.33
0.32
0.55
0.16
0.58
0.17
0.70
0.63
0.51
0.52
0.27
0.33
0.11
1.22
1.17
0.63
0.11
0.33
0.20
0.54
0.18
0.64
10
11
0.57
0.48
X,Y-Dimethylbiphenyl (182)
a
-Methylstilbene (194)
o-Terphenyl (230)
1-Tolyl-2-phenylethene (194)
5,6,11,12-Tetrahydro-dibenzo[a,e]cyclooctene (208)
Fluorene (166)
0.15
1.29
3.44
1.45
0.10
0.62
0.90
1.22
2.91
1.01
0.45
0.25
0.85
0.90
2.71
0.78
0.29
0.26
0.86
0.71
2.99
1.06
0.19
0.57
0.32
X-Methylfluorene (180)
2,3-Dimethylfluorene (194)
Benz[a]fluorene (216), Benz[b]fluorene (216)
Fluoranthrene (202)
12
13
Anthracene (178)
Phenanthrene (178)
144
1-Methylanthracene (192)
2.68
2.82
2.40
2.39
2-Methylanthracene (192)
1-Methylphenanthrene (192)
2-Methylphenanthrene (192)
X,Y-Dimethylanthracene (206)
X,Y-Dimethylphenanthrene (206)
2,3,5-Trimethylphenanthrene (220)
Phenalene (166)
1a,9b-Dihydro-cyclopropa[1]phenanthrene (192)
Pyrene (202)
X-Methylpyrene (216)
X,Y-Dimethylpyrene (230)
Benzo[a]pyrene (252)
X-Methylchrysene (242)
Triphenalene (228)
145
1.30
1.59
0.39
1.20
2.11
0.12
147
149
150
151
152
153
154
156
157
158
159
160
161
0.21
0.24
0.25
1.08
3.52
1.72
14
15
2.21
3.85
2.35
0.31
0.68
2.33
3.31
1.28
0.84
2.50
1.61
0.24
0.76
0.32
0.38
0.26
0.18
0.25
0.72
2-Methyltriphenalene (242)
Benz[a]anthracene (228)
1-Methylbenz[a]anthracene (242)5-Methylbenzo[c]phenanthrene (242)
7,12-Dimethylbenz[a]anthracene (256)
5,8-Dimthylbenzo[c]phenanthrene (256)
0.34
0.19
0.16
0.46
0.21
0.21
0.86
0.10
The original data show that variable 57 (2-methyl-6-(p-tolyl)hept-
2-ene) is produced at 300 °C (0.21%) and at 400 °C (9.10%), but was
absent at 500 °C. It was not observed in any of the other
pyrosylates.
Fig. 5b shows the results of the PCA at 500 °C and indicates that
PC 1 separates valencene and cedrene from each other where var-
iable 89 (benzcycloheptatriene) is associated with valencene, while
variable 47 (ethyl-methylbenzene) is connected to cedrene. PC 2

1320
G.W. Francis et al. / Food Chemistry 135 (2012) 1316–1322
Table 2
Compounds identified by structural groups in the pyrosylates of caryophyllene (K), cedrene (S), longifolene (L) and valencene (V) at pyrolysis temperatures of 300, 400 and 500 °C.
Group structure type is based on the main skeleton of the compound. Group 16 refers to rearranged starting material, i.e. C15H24 compounds other than the original compound.
Group 17 refers to unchanged starting material.
Group
Group structure type
300 °C
400 °C
500 °C
L
K
V
S
L
K
V
S
L
K
V
S
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Acyclic
3.48
0.83
0.93
5.19
48.08
0.80
Cyclopentene
Cyclohexene
Benzene
Azulene
Benzocycloheptatriene
Indene
Tetralin
Naphthalene
Biphenyl
0.32
0.37
5.32
0.12
0.81
2.09
0.17
1.34
0.64
0.21
0.35
10.45
0.58
0.08
39.67
1.73
9.13
8.26
0.18
18.37
1.23
5.59
0.30
5.27
8.73
1.54
42.61
1.09
7.76
7.28
0.06
18.98
2.03
6.29
0.62
6.02
6.31
0.96
22.42
0.39
19.46
6.31
50.62
0.79
7.23
6.53
9.56
2.09
4.09
31.78
0.97
4.68
0.26
4.56
6.98
2.15
13.57
3.64
4.95
0.57
5.02
4.95
2.13
Fluorene
Fluoranthene
Anthracene/Phenanthrene
Pyrene
Tetracyclic PAH
Sesquiterpene
Unchanged
0.50
95.37
0.86
99.14
0.08
99.71
0.68
98.97
24.17
4.26
4.67
84.79
1.08
87.77
100
Fig. 3. Scoreplot for showing the scores of samples of the sesquiterpenes pyrolysed
at temperatures of 300, 400 and 500 °C. Identities of samples as follows: 1 – cedrene
(300), 2 –valencene (300), 3 – longifolene (300), 4 – caryophyllene (300), 5 –
cedrene (400), 6 – valencene (400), 7 – longifolene (400), 8 – caryophyllene (400), 9
– cedrene (500), 10 – valencene(500), 11 – longifolene (500), 12 – caryophyllene
(500).
separates caryophyllene and longifolene from the other com-
pounds and indicates their association with variables 40 and 41
(dimethylbenzene and trimethylbenzene).
Since unsubstituted PAHs are known to be more dangerous to
life than the substituted compounds a table indicating the amounts
of both substituted and unsubstituted compounds is provided (Ta-
ble 3). It is immediately clear that as the temperature is increased
the amounts of unsubstituted PAHs increases in both absolute
amount and proportion, thus demonstrating the deleterious effects
of increasing pyrolysis temperatures.
Fig. 4. (a) Bi-plot, combined score plot and loading plot, of samples pyrolysed at
400 °C. (b) Bi-plot, combined score plot and loading plot, of samples pyrolysed at
500 °C. Variables refer to individual compounds.
4.2. Mechanism
The formation of PAHs from organic materials have been the
subject of investigations over a number of years and it has been
shown that almost any organic compound or mixture of com-

G.W. Francis et al. / Food Chemistry 135 (2012) 1316–1322
1321
Table 3
Substituted, unsubstituted and total PAH at 400 and 500 °C pyrolysates of longifolene, caryophyllene, valencene and cedrene.
Compound and pyrolysis temperature (°C)
Unsubstituted PAH
Substituted PAH
Total PAH
% of total PAH that is unsubstituted
Longifolene 400
Caryophyllene 400
Valencene 400
Cedrene 400
0.0
1.3
0.9
0.0
14.9
3.4
0.0
16.1
4.3
–
8.1
20.9
0.0
0.0
0.6
0.6
Longifolene 500
Caryophyllene 500
Valencene 500
Cedrene 500
19.8
17.1
32.8
12.7
40.5
40.4
44.9
36.7
60.3
57.5
77.6
49.4
32.8
29.7
42.3
25.7
penes with caryophyllene clearly being the most susceptible of the
compounds examined. Longifolene, on the other hand was most
resistant to pyrolysis. Increasing the pyrolysis temperature from
400 to 500 °C not only increased the amounts of PAHs formed,
but also led to an increasing proportion of the more dangerous
unsubstituted PAHs.
The pyrolysates at 500 °C show much similarity and two factors
may be contributing to this: the known propensity of the sesqui-
terpenes to rearrange and the fact that pyrosynthesis requires
breakdown to low weight species prior to reassembly.
It is apparent from the present results that the presence of ses-
quiterpenes in spices and other vegetable food products and the
resulting PAH production represent a real risk factor when both
high temperatures and long cooking/heat treatment times are in-
volved in the preparation of food.
Fig. 5. Bi-plot, combined score plot and loading plot, of samples pyrolysed at
500 °C. Variables refer to structural groups.
References
Aebi, A., Barton, D. H. R., Burgstahler, A. W.,
& Lindsey, A. S. (1954).
Sesquiterpenoids. V. Stereochemistry of tricyclic derivatives of caryophyllene.
Journal of the Chemical Society, 4659–4665.
Badger, G. M., Donnely, J. K., & Spotswood, T. M. (1965). The formation of aromatic
hydrocarbons at high temperatures. The pyrolysis of some tobacco constituents.
Australian Journal of Chemistry, 18, 1249–1266.
Barrington, P. J., Baker, R. S. U., Truswell, A. S., Bonin, A. M., Ryan, A. J., & Paulin, A. P.
(1990). Mutagenicity of basic fractions derived from lamb and beef cooked by
common household methods. Food and Chemical Toxicology, 28, 141–146.
Bogorad, L. (1976). Chlorophyll biosynthesis. In T. W. Goodwin (Ed.), Chemistry and
Biochemistry of Plant Pigments (pp. 64–148). (Vol. 1, 2nd ed.) London:
Academic Press.
Britton, G. (1976). Biosynthesis of carotenoids. In T. W. Goodwin, Chemistry and
Biochemistry of Plant Pigments (pp. 262-327) (Vol. 1, 2nd ed.). London:
Academic Press.
Chakravarti, D., Venugopal, D., Mailander, P. C., Meza, J. L., Higginbotham, S.,
Cavalieri, E. L., et al. (2008). The role of polycyclic aromatic hydrocarbon-DNA
adducts in inducing mutations in mouse skin. Mutation Research, 649, 161–178.
Cheng, A.-X., Xiang, C.-Y., Li, J.-X., Yang, C.-Q., Hu, W.-L., Wang, L.-J., et al. (2007). The
rice (E)-b-caryophyllene synthase (OsTPS3) accounts for the major inducible
volatile sesquiterpenes. Phytochemistry, 68, 1632–1641.
Christy, A. A., Lian, M. I., & Francis, G. W. (2011). Pyrolytic formation of polyaromatic
hydrocarbons from steroid hormones. Food Chemistry, 124, 1466–1472.
Christy, A. A., Ozaki, Y., & Gregoriou, V. G. (2001). Modern Fourier Transform Infrared
Spectroscopy. Amsterdam: Elsevier Science, p 285.
Collado, I. G., Hanson, J. R., & Macias-Sanchez, A. J. (1998). Recent advances in the
chemistry of caryophyllene. Natural Products Review, 15, 187–204.
Cypres, R. (1987). Aromatic hydrocarbon formation during coal pyrolysis. Fuel
Processing Technology, 15, 1–15.
Devon, T. K., & Scott, A. I. (1972). Handbook of Naturally Occurring Compounds, Vol. 2.
Terpenes. New York and London: Academic Press, p. 55.
pounds can provide PAHs under suitable conditions (Badger,
Donnely, & Spotswood, 1965; Cypres, 1987; Wornat, Sarofim, &
Longwell, 1987). Three mechanisms are currently thought to con-
tribute to some degree in the production of PAHs (Ross et al.,
2011; Shukla, 2010; Shukla, Miyoshi, & Koshi, 2010; Slavinskaya
& Frank, 2009). These mechanisms are known by the acronyms
HACA (Hydrogen Abstraction Carbon Addition), MAC (Methyl
Addition Cyclisation) and PAC (Phenyl Addition Cyclisation), and
are believed to act in concert to provide PAHs.
The production of PAHs from sesquiterpenes by the above
mechanisms requires their breakdown by bond scissions, among
others to 1- and 2-carbon species. Some specific support for initial
bond scission of the required type may be found in the fact that at
300 and 400 °C cedrene provided 2-methyl-6-(p-tolyl)-hept-2-ene
which might be formulated by such bond cleavage and partial aro-
matisation (see Fig. 2b). This might be taken together with the
finding in the 500 °C pyrolysate of cedrene to further bolster this
suggestion. However, the generality of such fission to smaller frag-
ments is accepted as a formation mechanism for PAH from many
materials and is thus not controversial.
However, it should be remembered that the aromatisation of
sesquiterpenes by heat fusion in the presence of sulphur or sele-
nium was much used to establish both carbon skeletons and sub-
stitution patterns (Nigam & Levi, 1965; Stahl & Müller, 1974).
Thus some aromatisation may be occurring without prior fragmen-
tation and this should be borne in mind.
Degenhardt, J., Köller, T. G.,
& Gershenzon, J. (2009). Monoterpene and
sesquiterpene synthases and the origin of terpene skeletal diversity in plants.
Phytochemistry, 70, 1621–1637.
Fraga, B. M. (2011). Natural sesquiterpenoids. Natural Product Reports, 28,
1580–1610.
García-Falcón, M. S., Cancho-Grande, B., & Simal-Gándara, J. (2005). Minimal clean-
up and rapid determination of polycyclic aromatic hydrocarbons in instant
coffee. Food Chemistry, 90, 643–647.
5. Conclusions
John, E. M., Stern, M. C., Sinha, R., & Koo, J. (2011). Meat consumption, cooking
practices, meat mutagens and risk of prostrate cancer. Nutrition and Cancer, 63,
525–537.
Khatmullina, R. M., Safarova, V. I., Kudasheva, F. Kh., & Kitaeva, I. M. (2012).
Chromatographic determination of polycyclic aromatic hydrocarbons in oil
sludge. Journal of Analytical Chemistry, 67, 251–257.
Sesquiterpenes produce PAHs on pyrolysis at 400 and 500 °C. At
the lowest temperature studied, 300 °C, no PAHs were formed
although some rearrangement was observed. The temperature
required for pyrolytic reactions to occur varies between sesquiter-

1322
G.W. Francis et al. / Food Chemistry 135 (2012) 1316–1322
Kvalheim, O. M.,
&
Karstang, T. V. (1997).
A
general-purpose program for
sesquiterpenes at 296
1193–1205.
Shukla, B., & Koshi, M. (2010). Comparitive study on the growth mechanisms of
PAHs. Combustion and Flame, 158, 369–375. and references therein.
Shukla, B., Miyoshi, A., & Koshi, M. (2010). Role of methyl radicals in the growth of
PAHs. Journal of American Chemical Society, 21, 534–544.
Slavinskaya, N. A., & Frank, P. (2009). A modelling study of aromatic soot precursors
formation in laminar methane and ethane flames. Combustion and Flame, 156,
1705–1722.
2 K. International Journal of Chemical Kinetics, 26,
multivariate data analysis. Chemometrics & Intelligent Laboratory Systems, 2,
235–237.
Lagerqvist, A., Håkansson, D., Lundin, C., Prochazka, G., Dreeij, K., Segerbäck, D., et al.
(2011). DNA repair and replication influence the number of mutations per
adduct of polycyclic aromatic hydrocarbons in mammalian cells. DNA Repair, 10,
877–886.
Lee, M. L., Novotny, M. V., & Bartle, K. D. (1981). Analytical Chemistry of Polycyclic
Aromatic Compounds. Columbus, Ohio: Academic press, p. 441.
Luch, A. (2005). The Carcinogenic Effects of Polycyclic Aromatic Hydrocarbons. London:
Imperial College Press.
Nigam, I. C., & Levi, L. (1965). Essential oils and their constituents XXIV. Study of
sesquiterpene dehydrogenation reactions by gas–liquid chromatography.
Journal of Chromatography, 17, 466–471.
Stahl, E.,
& Müller, Th. K. B. (1974). Schwefel-und Selendehydrierungen im
Mikrogrammbereich. Zeitschrift fur Analytische Chemie, 271, 257–264.
Steck, S. E., Gaudet, M. M., Eng, S. M., Britton, J. A., Teitelbaum, S. L., Neuget, A. I.,
et al. (2007). Cooked meat and risk of breast cancer–lifetime versus recent
dietary intake. Epidemiology, 18, 373–382.
Parker, W., Raphael, R. A., & Roberts, J. S. (1965). Neoclovene, a novel rearrangement
product of caryophyllene. Tetrahedron Letters, 2313–2316.
Parthasarathy, V. A., Chempakam, B., & Zachariah, T. J. (2008). Chemistry of Spices.
Willingford, UK: CABI.
Ross, A. B., Lea-Langton, A., Fitzpatrick, E. M., Jones, J. M., Williams, A., Andrews, G.
E., et al. (2011). Investigation of pyrolysis of hydrocarbons and biomass model
compounds using a micropyrolysis flow cell. Energy and Fuels, 25, 2954–2955.
Voorhees, K. J. (1984). Analytical Pyrolysis (pp. 196). London: Butterworth.
Wornat, M. J., Braun, A. G., Hawiger, A., Logwell, J. P., & Sarofim, A. F. (1990). The
relationship between mutagenicity and chemical composition of polycyclic
aromatic compounds from coal pyrolysis. Environmental Health Perspectives, 84,
193–201.
Wornat, M. J., Sarofim, A. F., & Longwell, J. P. (1987). Changes on the degree of
substitution of polycyclic aromatic compounds from pyrolysis of a high-volatile
bituminous coal. Energy and Coals, 1, 431–437.
Schulte-Elte, K. H.,
isocaryophyllene. Rearrangement of their 1,5-diene systems following direct
excitation in the liquid phase. Helvetica Chimica Acta, 54, 370–397.
& Ohloff, G. (1971). Photochemistry of caryophyllene and
p,
Yu, F., & Utsumi, R. (2009). Diversity, regulation, and genetic manipulation of plant
mono- and sesquiterpenoid biosynthesis. Cellular and Molecular Life Sciences, 66,
3043–3052.
p
Shu, Y. K., & Atkinson, R. (1994). Rate constants for the gas-phase reactions of O3
with a series of terpenes and OH radical formationfrom the O3 reactions with