Food

Chemistry
Food Chemistry 92 (2005) 143–149
www.elsevier.com/locate/foodchem

Extraction of fennel (Foeniculum vulgare Mill.) seeds with

supercritical CO2: Comparison with hydrodistillation
a,*
Biljana Damjanović , Žika Lepojević b, Vladimir Živković c, Aleksandar Tolić a

a
Faculty of Metallurgy and Technology, University of Montenegro, Cetinjski put bb., 81000 Podgorica, Serbia and Montenegro
b
Department of Pharmaceutical Engineering, Faculty of Technology, University of Novi Sad, Bulevar Cara Lazara 1,
21000 Novi Sad, Serbia and Montenegro
c
Center for Ecotoxicological Researches of Montenegro, Put Radomira Ivanovica 2, 81000 Podgorica, Serbia and Montenegro

Received 7 June 2004; accepted 16 July 2004

Abstract

Ground fennel (Foeniculum vulgare Mill.) seeds, growing wild in Montenegro, were extracted with supercritical CO2 (SC-CO2) at
a flow rate of 0.2 kg CO2/h under varying extraction conditions in order to determine yield, composition and organoleptic charac-
teristics of extract. The extracts obtained were compared to fennel seed oil isolated by hydrodistillation. In the SC-CO2, extracts as
well in the hydrodistilled oil, the major compounds were trans-anethole (68.6–75.0%) and (62.0%), methylchavicol (5.09–9.10%) and
(4.90%), fenchone (8.40–14.7%) and (20.3%), respectively. With pressure varying from 80 to 150 bar and temperature varying from
40 to 57
C, it was found that, for the selected herb material, the optimal conditions of SC-CO2 extraction (high percentage of trans-
anethole, with significant content of fenchone and reduced content of methylchavicol and co-extracted cuticular waxes) are: pres-
sure, 100 bar; temperature, 40
C; extraction time, 120 min. Organoleptic tests confirmed that hydrodistilled oil possessed a less
intense fennel seed aroma than extracts obtained by SC-CO2.
 2004 Elsevier Ltd. All rights reserved.

Keywords: Foeniculum vulgare Mill.; Supercritical CO2; Hydrodistillation; Essential oil; Cuticular wax; Anethole; Fenchone

1. Introduction cia-Jiménez, Pérez-Alonso, & Velasco-Negueruela,

2000).
Fennel (Foeniculum vulgare Miller) is an annual, The third European Pharmacopoeia reports two fen-
biennial or perennial plant, depending on the variety, nel drugs, from bitter fennel fruits (F. vulgare Miller
belonging to Apiaceae family and is native to the Med- spp. vulgare var. vulgare) and sweet fennel fruits [F.
iterranean area (Piccaglia & Marotti, 2001). It has been vulgare Miller spp. vulgare var. dulce (Miller) Thel-
cultivated and introduced into many regions outside lung]. The drug consists of the dry, ripe, whole cremo-
that zone; it is grown commercially in some of them, carps and mericarps (commonly called seeds), which
such as Russia, India, China and Japan (Volák & Sto- contain 1–4% (v/w) of volatile oil whose major constit-
dola, 1998). Fennel has been known, since antiquity, as uents are the phenylpropanoid derivative trans-ane-
a medicinal and aromatic herb, commonly used to fla- thole and fenchone (Miraldi, 1999). The oil is used as
vour liqueurs, breads, fishes, salads and cheeses (Gar- an ingredient of cosmetic and pharmaceutical products
for its balsamic, cardiotonic, digestive, lactogogue and
tonic properties.
*
Corresponding author. Tel./fax: +381 81 245 406. The essential oil of aromatic herbs has usually been
E-mail address: bibana@cg.yu (B. Damjanović). isolated either by hydrodistillation (Bilia, Fumarola,

0308-8146/$ - see front matter  2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2004.07.019
144 B. Damjanović et al. / Food Chemistry 92 (2005) 143–149

Gallori, Mazzi, & Vincieri, 2000; Guillén, Cabo, & Mur- matography (GC) and gas chromatography-mass
illo, 1996) or extraction with classical solvents such as spectrometry (GC–MS).
hexane or ethanol (Damjanović, Skala, Petrović-Dja-
kov, & Baras, 2003), as traditional spice-processing
methods. These techniques present serious drawbacks: 2. Materials and methods
Low extraction efficiency, long extraction time, toxic
residual solvent in the products and deterioration of 2.1. Collection and preparation of herb samples
the thermally sensitive materials (Gámiz-Gracia & Lu-
que de Castro, 2000). Much work has recently been Many factors can influence the amount of essential
done on the composition of fennel essential oil obtained oil in aromatic herbs, e.g., climate and environmental
by traditional techniques (Barazani et al., 2002; Bernath conditions, season of collection, age of plants and, for
et al., 1999; Lawrence, 1995; Ruberto, Tiziana Baratta, fennel especially, the stage of ripening of the fruits. To
Deans, & Damien Dorman, 2000). avoid this influence in the present work, the ripe, green-
The advantages of the supercritical fluid extraction ish-brown seeds of growing wild fennel (Foeniculum
(SFE) technique are well known by now and it is often vulgare Mill.) were collected manually from the same
regarded as an alternative to the traditional methods. collection site in the Podgorica region (central south
It has been established as an environmentally benign Montenegro), within two days, at the end of September,
technique for separating essential oil from the vegetable 2002. A voucher specimen was deposited in the Herbar-
substrates. SFE exploits the unique properties of gases ium, Department of Biology, Faculty of Natural Sci-
above their critical values: Liquid like density, high rates ences and Mathematics, University of Montenegro.
of mass transfer and tuneable selectivity that arises by The seeds were air-dried and stored in double layer pa-
varying temperature and pressure. SFE of flavours and per bags at ambient temperature, protected from direct
fragrances has been reviewed by Kerrola (1995) and light until further analysis.
Reverchon (1997). The initial water inherent in the herb seed was found
Carbon dioxide is the most widely used compressed to be 9.6% (w/w), using a Dean and Stark apparatus
fluid, especially for the extraction of natural products, with n-heptane as the reflux solvent. Typically, a batch
because it is non-toxic, non-explosive, readily available, of 240 g of material was milled in a coffee mill and, after
easily removable from the product and possesses con- sieving, a mean particle diameter size of 0.9 mm was ob-
venient critical properties (Tc = 31.1
C, Pc = 73.8 bar). tained. A typical particle size distribution is shown in
Moreover, obtained extract retains the organoleptic Fig. 1. A prepared batch was kept in an air-tight resala-
characteristics of the starting spice material (Oszagyan ble polypropylene bag and stored at 8
C for a maxi-
et al., 1996; Reverchon & Senatore, 1992). mum of three days before use, in order to avoid losses
An attempt to extract fennel oil using liquid CO2 in of volatile compounds.
a Soxhlet type apparatus has been described by Naik,
Lentz, and Maheshwari (1989). Subcritical water 2.2. Reagents
extraction (Gámiz-Gracia & Luque de Castro, 2000),
SFE of Hungarian fennel (Simándi et al., 1999), and Commercial carbon dioxide (99% purity, Tehno-gas,
SFE extraction and mathematical modelling of fennel Novi Sad, Serbia and Montenegro) was used for extrac-
seed oil and essential oil (Reverchon, Daghero, Mar- tions. HPLC grade chloroform and methanol were pur-
rone, Mattea, & Poletto, 1999)have been reported chased from Riedel-de Haën, Germany. Standard
previously. samples for GC analyses were purchased from Fluka,
The essential oil or oleoresins from fennel seeds are Great Britain.
important ingredients for flavouring cosmetics, phar-
maceuticals and food products; hence, the improve-
50
ment of the quality of fennel oil is of economic 43.58
Distribution (%)

importance. For this reason, the extraction of fennel 40 34.44

seed oil using SC-CO2 could have an industrial 30
impact.
20 18.48
The aim of this work was to compare the SC-CO2
extraction of fennel seed with hydrodistillation and to 10
optimize the oxygenated compound contents, e.g., 2.34 1.02 0.14
0
trans-anethole and fenchone. The influence of parame- 2.575 1.4 0.557 0.237 0.13 0.05
ters, such as temperature, pressure and extraction time,
Particle size (mm)
on the extraction rate of essential oil and cuticular waxes
was also studied. The compositions of the extracts and Fig. 1. Typical particle size distribution of the ground fennel seeds
hydrodistilled oil obtained were analyzed by gas chro- used in this work.
 B. Damjanović et al. / Food Chemistry 92 (2005) 143–149 145

2.3. Hydrodistillation extracts were identified by comparing their retention

times with those of available standards and their mass
Raw, as well as residual (after SC-CO2 extraction), was calculated from a predetermined peak area response
samples of ground fennel seed were subjected to hydro- factor.
distillation according to the fourth Yugoslav Pharmaco- The GC–MS analyses were carried out using a Shim-
poeia in a Clevenger-type apparatus. The obtained oil adzu QP 5050 GC–MS equipped with a PTE-5 capillary
was separated from water and the yield was determined column with the same characteristics as the one used in
in terms of dry basis yield, as g/100 g of dry herb; it was GC. The column temperature was maintained at 60
C
stored in a freezer at
4
C until analyzed by GC and for 2 min and than programmed to increase as follows:
GC–MS. 60–300
C at 4
C/min. The temperature was 260
C at
the injection port and 300
C at the interface. The sam-
2.4. Supercritical fluid extraction ples, previously dissolved in chloroform:methanol (3:1),
were injected (1 ll) in split mode with split ratio of 1:20
Supercritical fluid extraction (SFE) was carried out and the flow rate of the carrier gas (helium) was 1.5 ml/
with a laboratory-scale high-pressure extraction plant min; inlet pressure, was 16.8 kPa. The MS conditions
(NOVA – Swiss, Effretikon, Switzerland), described pre- were: The ionization voltage, 70 eV, scanning interval
viously (Pekić, Zeković, Petrović, Lepojević, & Tolić, 0.5 s, detector voltage 1.3 kV and m/z range of 40–
1995). The main part and characteristics (manufacturer 500. The components were identified by computer
specification) of the plant were as follows: The dia- searching (using data base programme Class 5000) and
phragm-type compressor (up to 1000 bar), extractor comparing their mass spectral data with available stand-
with an internal volume of 200 ml (pmax = 700 bar), sep- ards and those in the WILEY229 and the NIST107 mass
arator with internal volume of 200 ml (pmax = 250 bar) spectra libraries.
and a maximum CO2 mass flow rate of approximately
5.7 kg/h. The amount of fennel seed sample in the
extractor was 80 g; pressure and temperature were inves- 3. Results and discussion
tigated values; CO2 flow rate was 0.2 kg CO2/h; and the
total extraction time was 4 h (samples were taken every In order to obtain the best conditions for the SC-CO2
half hour). Separator conditions were 14 bar and 25
C. extraction of fennel seeds, experiments were performed
The obtained extracts were measured and stored in a at different pressure–temperature settings of the extrac-
freezer at
4
C until analyzed by GC and GC–MS. tor and results were compared to hydrodistilled oil.
The extraction degree (Ed) of essential oil obtained by The selection of pressure and temperature ranges is
SC-CO2 was evaluated using Eq. (1) based on the fact that a great change in the density
M CO2 ðtÞ M CO2 ðtÞ and dielectric constant of CO2 occurs between 80 and
Ed ¼ 100 ¼ 100 ; ð1Þ 150 bar, and, in order to prevent the thermal degrada-
M0 M CO2 þ M HD Res
tion of some volatile oil compounds, the temperatures
where M0 refers to the initial mass (content) of essential up to 57
C are applied. The pressure varied from 80
oil in the herb sample. MCO2(t) is the mass of essential oil to 150 bar and temperature from 40 to 57
C and vari-
extracted by supercritical CO2 at time t, MCO2 is the to- ations in the solvent density and the transport properties
tal essential oil mass extracted by CO2 after 4 h, and of the system were observed.
MHD Res is the residual mass of essential oil isolated The total yields of fennel seed hydrodistilled oil and
by hydrodistillation after CO2 extraction. For cuticular SC-CO2 extracts at different temperatures and pressures
waxes, the extraction curve points were calculated on are shown in Fig. 2.
dry basis yield, as g/100 g of dry herb. Each experiment
was repeated twice.

2.5. GC and GC–MS analyses

GC analyses were performed using Ai Cambridge

GC-94 gas chromatograph equipped with a FID and a
DB-5 capillary column (30 m · 0.32 m, film thickness
0.25 lm). Oven temperature was programmed at 60
C
for 2 min, and then increased to 250
C at a rate of
8
C/min; injector and detector temperature were 120
and 270
C, respectively. Helium as carrier gas was ad-
justed to a flow rate of 1.5 ml/min. A split flow of 11 Fig. 2. Total yield of the hydrodistilled oil (HD) and SC-CO2 extracts
ml/min was used. The constituents of essential oil and after 4 h of extraction.
146 B. Damjanović et al. / Food Chemistry 92 (2005) 143–149

At 80 bar, the total amount of the obtained extract is mally labile active ingredients present in the final
unusually small but, as pressure increases the amount of product (e.g., 1.8-cineole).
extracted material increases, (Fig. 2). It was found that The chemical compositions of the SC-CO2 extracts
at higher pressures (100 and 150 bar) the solubility of and hydrodistilled oil are shown in Table 1.
some compounds from vegetable matter pass from neg- In the hydrodistilled oil, 19 compounds were identi-
ligible to significant (Reverchon, 1997). This is explained fied, with major compounds being: trans-Anethole
by the fact that the density and viscosity of supercritical (62.0%), fenchone (20.3%) and methylchavicol (4.90%).
CO2 change and, therefore, its extracting power in- In the SC-CO2 extracts, 28 compounds were identified
creases. However, the high pressure is bound to result with major compounds being: trans-Anethole (68.6–
in greater cost for the extraction operating system, as 75.0%), fenchone (8.40–14.7%) and methylchavicol
well as stringent operation design and increased energy (5.09–9.10%).
demand. Hydrodistilled oil contained considerable higher per-
It should also be noted that the appearance of the ex- centages of fenchone (20.3%), a-pinene (2.81%), myr-
tracts changed slightly with the increase of temperature cene (1.68%) and limonene (3.15%) than SC-CO2
and pressure. The colour, which is straw yellow when extracts, (8.40–14.7%, 0.22–0.43%, 0.30–0.57% and
lower pressure and temperature are used, becomes dar- 0.36–0.61%, respectively). In contrast, the SFE extracts
ker, particularly with increasing temperature. Increased contained a higher percentage of oxygenated com-
temperature, at 100 and 150 bar, has a favourable effect pounds, especially methylchavicol and trans-anethole.
on the extraction efficiency although the change is not Some oxygenated compounds, not found in hydrodis-
significant. However, when the temperature reaches tilled oil, were identified in insignificant amounts in
57
C, extract quality probably begin to decrease SFE extracts, e.g., terpinen-4-ol (0.05–0.09%) and anis-
according to the colour of the extracts. Moreover, ele- aldehyde (0.05–0.06%), while monoterpene b-ocimene
vated temperature could lead to decomposition of ther- was identified only in hydrodistilled oil (0.22%). This

Table 1
Relative percentage composition of fennel seed SC-CO2 extracts and hydrodistilled oil
Compound 80 bar 100 bar 150 bar HD
40
C 47
C 57
C 40
C 47
C 57
C 40
C 47
C 57
C
a-Thujone tr tr – tr – – 0.05 tr tr 0.05
a-Pinene 0.42 0.24 0.36 0.22 0.25 0.43 0.29 0.32 0.41 2.81
Camphene 0.08 – – 0.07 0.06 0.11 0.05 0.07 0.12 0.34
Sabinene 0.19 0.21 0.29 0.18 0.19 0.18 0.25 0.21 0.19 0.56
Myrcene 0.41 0.50 0.57 0.42 0.30 0.46 0.35 0.35 0.37 1.68
a-Phellandrene 0.13 0.16 0.22 0.12 0.18 0.38 0.10 0.18 0.18 0.73
p-Cymene 0.05 tr tr 0.05 – tr tr – – 0.28
Limonene 0.61 – – 0.54 0.36 0.54 0.39 0.61 0.41 3.15
1.8 Cineole 0.71 0.69 0.70 0.20 0.17 0.03 0.24 0.25 0.13 1.20
b-Ocimene – – – – – – – – – 0.22
c-Terpinene 0.47 0.71 0.75 0.37 0.32 0.45 0.55 0.40 0.43 1.05
Fenchone 10.2 14.7 14.5 11.4 9.38 8.57 8.97 8.40 8.68 20.3
Camphor Tr tr – tr 0.05 – tr tr – 0.58
Terpinen-4-ol 0.07 tr – 0.09 – – – 0.05 – –
Methyl chavicol 6.60 6.81 9.10 6.74 6.70 6.63 5.90 5.81 5.09 4.90
cis-Anethole tr tr – tr tr tr tr 0.05 tr tr
Anisaldehyde 0.06 0.05 tr 0.06 tr – – 0.05 0.05 –
trans-Anethole 72.8 71.8 69.3 73.3 75.0 74.3 68.6 69.1 69.9 62.0
Germacrene D 0.31 0.46 0.20 0.24 0.42 0.36 1.26 0.54 0.56 0.18
Tetradecanoic acid 0.14 0.52 0.46 0.13 0.60 0.45 0.99 0.67 1.27 –
Hexadecanoic acid 0.72 0.08 0.45 0.65 0.82 0.82 1.41 1.43 1.44 –
Tetradecane 2.05 0.44 0.44 1.58 1.17 1.07 1.43 2.50 1.50 –
1,2 Benzenedicarboxylic acid, dioctyl ester 0.17 0.14 0.46 0.35 0.16 0.84 0.66 0.90 1.10 –
Pentadecane 0.08 0.10 0.14 0.07 0.13 0.19 0.88 1.41 0.88 –
Pentacosane 0.12 0.54 0.52 0.12 0.17 0.71 1.11 1.04 1.09 –
1–Hexadecanol 0.40 0.05 0.05 0.4 0.83 0.68 0.95 1.05 1.07 –
Hexacosane 0.25 0.05 0.05 0.23 0.83 0.68 0.80 1.17 1.26 –
1–Octadecanol 0.37 0.06 0.05 0.36 0.67 0.75 1.42 0.64 1.17 –
7–Octadecanone 1.34 0.41 0.34 1.78 1.15 1.32 2.48 1.94 1.73 –
Total identified 98.7 98.7 99.0 99.6 99.9 99.9 99.1 99.1 99.0 100
tr < 0.05; HD, hydrodistilled oil.
 B. Damjanović et al. / Food Chemistry 92 (2005) 143–149 147

Fig. 3. Yield (%) of fennel seed extracts isolated by SC-CO2 extraction and hydrodistillation (HD) with respect to grouped components.

indicates that degradation of the fennel essential oil dur- ponents of fennel oil, which occurred in the first hour of
ing the hydrodistillation process was minimal. Only one extraction, were lost in further stages (e.g., sabinene,
sesquiterpene hydrocarbon with insignificant percentage camphene, limonene). At 80 bar, an increase in temper-
was detected in fennel seeds, which is comparable with ature led to a rather significant decrease of the extrac-
previously reported results (Ruberto et al., 2000). tion degree due to decline of the solvent density,
A comparison between the essential oil obtained by probably dominating the increase of the solute vapour
hydrodistillation and SC-CO2 extracts showed that, in pressure (Fig. 4(a)). An opposite outcome was verified
the latter technique some undesired higher molecular for 100 and 150 bar where the temperature alteration
weight compounds (waxes) were co-extracted with the had a rather negligible influence on the essential oil yield
essential oil (Table 1). (Figs. 4(b) and (c)).
The co-extracted waxes consisted of paraffins as well At 100 and 150 bar, nearly 90% of the essential oil
as fatty acids, alcohols, esters, aldehydes and ketones. was extracted rather easily and quickly (after 120
The major compounds were hexadecanoic acid (0.08– min) whereas the last fractions were extracted very
1.44%), tetradecane (0.44–2.50%) and 7-octadecanone slowly (Figs. 4(b) and (c)). The fractions extracted in
(0.41–2.48%). the first stage corresponded to the essential oil liberated
It can be seen from Fig. 3 that the content of non-vol- from the glands damaged during the mechanical pre-
atile compounds increased with pressure (at 47
C), treatment of the fennel seed. On the other hand, the
from a minimum of 2.39% at 80 bar, over 6.53% at slow extraction of the last fractions indicated that a
100 bar to a maximum of 12.8% at 150 bar. The total portion of the essential oil was hidden within internal
percentage of monoterpene hydrocarbons in hydrodis- tissues or located within glands that survived pre-treat-
tilled oil is more than triple that of SC-CO2 extracts, ment. This stage showed typical diffusion-controlled
(10.8% against 1.66–2.55%, respectively). These are less characteristics.
valuable components since they contribute, only to a The CO2 flow rate and the extraction time were the
minor extent, to the aroma and tend to oxidize because main parameters that contributed to optimum extrac-
of their unsaturated character. In contrast, the SC-CO2 tion, i.e., a maximum yield at a minimum extraction
extracts, except at 150 bar, contained higher percentage time. At the selected flow rate of 0.2 kg CO2/h, it was
of oxygenated compounds that strongly contribute to found experimentally that an extraction period of 120
the fragrance. Therefore, the SC-CO2 should give better min was sufficient for isolation of more than 90% of
reproduction of the natural aroma of the fennel seeds the essential oil available from the herb (except at 80
than the hydrodistilled oil. Organoleptic tests confirmed bar, as previously mentioned).
that hydrodistilled oils possess a less intense fennel seed The cuticular waxes were extracted in insignificant
aroma than extracts obtained by supercritical CO2. amounts from the herb material (Fig. 5). The rate of
Hydrodistilled oil aroma was also considered to be extraction was negligible when lower density CO2 was
rather different from that of the starting herb material. used (Fig. 5(a)). However, the increase of pressure (in-
The effect of the SC-CO2 extraction pressure, temper- crease of CO2 density) was accompanied by a signifi-
ature and time on the essential oil extraction curves are cant increase of the extraction rate. The crossover
shown in Fig. 4. At lower solvent densities (80 bar), very pressures of typical cuticular waxes (octacosane and
low extraction rates were observed. The extraction de- triacontane) were found within the range of 100–110
gree at this pressure was very low, less than 50% (Fig. bar (Papamichail, Louli, & Magoulas, 2000). There-
4(a)) Extraction rate was obviously limited by solubility fore, it was not surprising that the yield of cuticular
of the essential oil in the supercritical CO2. It was also waxes increased at 100 bar and particularly at 150
noted that peaks corresponding to the low-boiling com- bar (Figs. 5(b) and (c)).
148 B. Damjanović et al. / Food Chemistry 92 (2005) 143–149

Fig. 5. Effects of the extraction pressure and temperature on the

Fig. 4. Effects of the extraction pressure and temperature on the
cuticular wax extraction curves.
extraction degree of the essential oil.

The optimal chemical composition of fennel seed with a high percentage of trans-anethole, responsible
essential oil depends on the desired oil application. In for fennel seed aroma, while the pharmaceutical indus-
the food industry, the most suitable type is fennel oil try prefers essential oil with a higher percentage of fenc-
 B. Damjanović et al. / Food Chemistry 92 (2005) 143–149 149

hone. In both cases, the content of methylchavicol, Damjanović, B. M., Skala, D., Petrović-Djakov, D., & Baras, J.
which is assumed to be toxic to a certain degree, should (2003). A comparison between the oil, hexane extract and
supercritical carbon dioxide extract of Juniperus communis L..
be reduced. The optimal ratio of these compounds was Journal of Essential Oil Research, 15, 91–93.
obtained at 100 bar and 40
C (Table 1). Garcia-Jiménez, N., Pérez-Alonso, M. J., & Velasco-Negueruela, A.
(2000). Chemical composition of fennel oil, Foeniculum vulgare
Miller, from Spain. Journal of Essential Oil Research, 12, 159–162.
4. Conclusions Gámiz-Gracia, L., & Luque de Castro, M. D. (2000). Continuous
subcritical water extraction of medicinal plant essential oil:
Comparison with conventional techniques. Talanta, 51, 1179–1185.
Fennel seeds were extracted by varying CO2 pressure Guillén, M. D., Cabo, N., & Murillo, J. (1996). Characterization of the
from 80 to 150 bar and temperature from 40 to 57
C. essential oils of some cultivated aromatic plants of industrial
Supercritical CO2 extraction of essential oil is accompa- interest. Journal of the Science of Food and Agriculture, 70,
nied by the co-extraction of unwanted cuticular waxes. 359–363.
Kerrola, K. (1995). Literature review: Isolation of essential oils and
However, SC-CO2 extraction allowed optimization of flavor compounds by dense carbon dioxide. Food Reviews Interna-
the experimental conditions for selection of substances tional, 11, 547–573.
of interest, namely the oxygenated compounds, trans- Lawrence, B. M. (1995). In Essential Oils 1992–1994. Carol Stream:
anethole and fenchone. It was found that SC-CO2 Allured Publishing Corporation.
extracts generally contained higher percentages of oxy- Miraldi, E. (1999). Comparison of the essential oils from ten
Foeniculum vulgare Miller samples of fruits of different origin.
genated compounds that strongly contribute to the fra- Flavour and Fragrance Journal, 14, 379–382.
grance. Therefore, the SC-CO2 extraction gives better Naik, S. N., Lentz, H., & Maheshwari, R. C. (1989). Extraction of
reproduction of the natural aroma of the fennel seeds perfumes and flavours from plant materials with liquid carbon
than does the hydrodistilled oil. For the selected herb dioxide under liquid–vapor equilibrium conditions. Fluid Phase
material and CO2 flow rate, the optimal conditions of Equilibrium, 49, 115–126.
Oszagyan, M., Simándi, B., Sawinsky, J., Kery, A., Lemberkovics, E.,
SC-CO2 extraction (high percentage of trans-anethole, & Fekete, J. (1996). Supercritical fluid extraction of volatile
with significant content of fenchone and reduced compounds from lavandin and thyme. Flavour and Fragrance
amount of methylchavicol and co-extracted cuticular Journal, 11, 157–165.
waxes) were: Pressure, 100 bar, temperature, 40
C Papamichail, I., Louli, V., & Magoulas, K. (2000). Supercritical fluid
and extraction time, 120 min. extraction of celery seed oil. Journal of Supercritical Fluids, 18,
213–226.
Pekić, B., Zeković, Z., Petrović, L., Lepojević, Ž., & Tolić, A. (1995).
Behavior of (
)-a-Bisabolol and (
)-a-Bisabololoxides A and B in
Acknowledgements camomile flower extraction with supercritical carbon dioxide.
Separation Science and Technology, 30(18), 3567–3576.
Piccaglia, R., & Marotti, M. (2001). Characterization of some Italian
We thank Mrs. Danijela Šuković (Center for Ecotox- types of wild fennel (Foeniculum vulgare Mill.). Journal of Agricul-
icological Research, Podgorica, Serbia and Montene- tural and Food Chemistry, 49, 239–244.
gro) for GC–MS analyses and Mr. Milan Janković for Reverchon, E., & Senatore, F. (1992). Isolation of rosemary oil:
supplying the fennel seeds. Comparison between hydrodistillation and supercritical CO2
extraction. Flavour and Fragrance Journal, 7, 227–230.
Reverchon, E. (1997). Supercritical fluid extraction and fractionation
of essential oils and related products. Journal of Supercritical
References Fluids, 10, 1–37.
Reverchon, E., Daghero, J., Marrone, C., Mattea, M., & Poletto, M.
Barazani, O., Cohen, Y., Fait, A., Diminshtein, S., Ravid, U., (1999). Supercritical fractional extraction of fennel seed oil and
Putievsky, E., et al. (2002). Chemotypic differentiation in indige- essential oil: Experiments and mathematical modelling. Industrial &
nous population of Foeniculum vulgare var. vulgare in Israel. Engineering Chemistry Research, 38, 3069–3075.
Biochemical Systematics and Ecology, 30, 721–731. Ruberto, G., Tiziana Baratta, M., Deans, S. G., & Damien Dorman,
Bernath, J., Nemeth, E., Patheo, F., Mihalik, E., Kalman, K., & H. J. (2000). Antioxidant and antimicrobial activity of Foeniculum
Franke, R. (1999). Regularities of the essential oil accumulation in vulgare and Crithmum maritimum essential oils. Planta Medica, 66,
developing fruits of fennel (Foeniculum vulgare Mill.) and its 687–693.
histological background. Journal of Essential Oil Research, 11, Simándi, B., Deák, A., Rónyai, E., Yanxiang, G., Veres, T.,
431–438. Lemberkovics, É., et al. (1999). Supercritical carbon dioxide
Bilia, A. R., Fumarola, M., Gallori, S., Mazzi, G., & Vincieri, F. F. extraction and fractionation of fennel oil. Journal of Agricultural
(2000). Identification by HPLC-DAD and HPLC-MS analyses and and Food Chemistry, 47, 1635–1640.
quantification of constituents of Fennel teas and decoctions. Volák, J., & Stodola, J. (1998). The illustrated book of herbs. London:
Journal of Agricultural and Food Chemistry, 48, 4734–4738. Caxton Editions.