Available online at [Link].

com

Procedia Engineering 42 (2012) 247 260

20th International Congress of Chemical and Process Engineering CHISA 2012

25 29 August 2012, Prague, Czech Republic

Cinnamon bark volatile oils separation and determination

using solid-phase extraction and gas chromatography
F. Golmohammad a*, M. H. Eikani, H. M. Maymandi
Department of Chemical Technologies, Iranian Research Organization for Science and Technology (IROST),
P.O. Box: 15815-3538, Tehran, Iran

Abstract
One of the SPE method is reversed phase separation which involve a polar usually aqueous or moderately polar
sample matrix (mobile phase) and a nonpolar stationary phase. The analyte of interest is typically mid- to non polar.
Several SPE materials are in the reversed phase category including C18. The hydrophilic silanol groups at the surface
of the raw silica packing have been chemically modified with hydrophobic alkyl functional groups by reaction with
the corresponding silanes. Retention of organic analytes from polar solutions on to the SPE materials is due primarily
to attractive forces between the carbon-hydrogen bonds in the analyte and the functional group in the silica surface.
These nonpolar-nonpolar attractive forces are van der waals forces. In this work, at first the aqueous solutions of
Cinnamomum zeylanicum Blume bark were obtained by hydrodistillation and superheated water extraction (SWE)
methods, then main the volatile compounds were extracted from aqueous solutions by solid phase extraction (SPE)
method. This method involves adsorption of the desired components on C18 SEP PAK (Monofunctional -Si
(CH3)2C18H37), followed by desorption in the desorption chamber, using minimum volume solvent as the mobile
phase. Solid phase extraction conditions optimized on C18 SEP PAK classic cartridge. Different solvent including
acetone, hexane, ethyl acetate, ethanol, and mix methanol: water were examined. A good separation of Ecinnamaldehyde was obtained on the adsorbent column. The amount of extracted essential oil composition of C.
zeylanicum by SPE, evaluated in different conditions. Gas chromatography/mass spectrometry and gas
chromatography were used to identify and quantity the volatile compound composition. GC analysis showed that the
maximum yield of E-cinnamaldehyde obtained by methanol: water in (70:30 v/v) ratio by SPE extraction. At the
optimum condition, the GC results indicated that E-cinnamaldehyde was the highest area percentage of 79.60% in the
volatile oils extracted by hydrodistillation and 88.50% by superheated water extraction.

2012 Published by Elsevier Ltd. Selection under responsibility of the Congress Scientific Committee
(Petr Kluson)

* Corresponding author. Tel.: +982282275537; fax: +982282276265.

E-mail address: fgolmohamady@[Link]

1877-7058 2012 Published by Elsevier Ltd.

doi:10.1016/[Link].2012.07.416

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

248

Keywords: Cinnamon bark; cinnamomum zeylanicum Blume; solid phase extraction(SPE); hydrodistillation; superheated water
extraction(SWE); e-cinnamaldehyde

1. Introduction
The principle of SPE is similar to liquid-liquid extraction (LLE), involving a partitioning of solutes between two phases. However, instead of two immiscible liquid phases, as in LLE, SPE involves
partitioning between a liquid (sample matrix or solvent with analytes) and a solid (sorbent) phase. This
sample treatment technique enables the concentration and purification of analytes from solution by
sorption on a solid sorbent and purification of extract after extraction [1]. An SPE device consists of a
resin bed packed into a small extraction tube, usually made of plastic. The resin is packed between two
frits to hold the resin bed securely in place. A liquid sample is passed through the resin bed by applying
either positive pressure or vacuum to the column [2].
A typical procedure of solid phase extraction involves four steps. First, the column is conditioned with
an appropriate solvent to solvate functional groups of the sorbent. After the sorbent is further conditioned
with the sample matrix solvent, the sample solution is forced through the sorbent by aspiration or positive
pressure. The column containing retained analyte is subsequently washed with an appropriate solvent that
selectively elutes impurities but leaves the analyte on the column. The purified analyte is finally eluted
with a solvent strong enough to displace the analyte from the sorbent.
The mechanisms involved in solid phase extraction are normal phase, reversed phase and Ion
exchange chromatography that are explained respectively.
Normal phase chromatography whose procedures typically involve a polar analyte, a mid-to nonpolar
matrix (e.g. acetone, chlorinated solvents, and hexane), and a polar stationary phase. Polar-functionalized
bonded silicas (e.g. LC-CN, LC-NH2, and LC-Diol), and polar adsorption media (LC-Si, LC-Florisil,
ENVI-Florisil, and LC-Alumina) are typically used under normal phase conditions. Retention of an
analyte under normal phase conditions is primarily due to interactions between polar functional groups of
the analyte and polar groups on the sorbent surface. These include hydrogen bonding, pi-pi interactions,
dipole-dipole interactions, and dipole-induced dipole interactions, among others. A compound adsorbed
by these mechanisms is eluted by passing a solvent that disrupts the binding mechanism usually a
solvent that is more polar than the samples original matrix.
Reversed phase chromatography whose procedures involve a polar (usually aqueous) or moderately
polar sample matrix (mobile phase) and a nonpolar stationary phase. The analyte of interest is typically
mid-to nonpolar. Several SPE materials, such as alkyl-or aryl-bonded silicas (LC-18, ENVI-18, LC-8,
ENVI-8, LC-4, and LC-Ph) are in the reversed phase category. Ion exchange chromatography that can be
used for compounds which are charged when in a solution (usually aqueous, but sometimes organic) [3].
As far as essential oils are concerned, solid phase extraction (SPE) is used for their pre-separation and
fractionation prior to gas chromatographic analysis as well as after superheated water extraction (SWE).
According to Antonelli and Fabbri [4], essential oils can be divided in two categories the former one
includes essential oils which are very complex mixtures whose components belong to different classes of
compounds and the latter one consists of essential oils with one major component comprising up to 90%
with a few other minor compounds which can be important for the quality and difficult to identify. That is
why silica SPE is needed for pre-separation of essential oils which consequently leads to formation of
three fractions: the first one containing all hydrocarbons, both saturated and unsaturated, the second one
which contains carbonyl compounds, ethers, esters and tertiary alcohols and the third one containing
primary alcohols, acids and diols. Such a fractionation helps to avoid problems with peak overlapping or
co eluting substances as well as to facilitate the concentration of remaining fractions after removal of the

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

major compound in the latter group of essential oils. As a frequent aim of SWE is to avoid the use of
organic solvents, it is suggested to apply extraction methods that are solvent free or use minimal amounts
of solvents. SPE is exactly such a technique which enables extraction of retained analytes with a small
volume of elution solvent [5].
An endcapped C18 SPE cartridge was used after superheated water extraction of fragrance compounds
from Rosa canina. In this case, a solvent used for elution was hexane and identified compounds were
mainly benzyl alcohol, benzaldehyde, phenylethyl alcohol, 2,6,11-trimethyl dodecane, eicosane,
tetrahydroional and limonene [6].
The most frequently used sorbent in the analysis of volatile components is octadecyl (C18) that
enables reversed phase extraction of mid-to nonpolar analytes.
A column with such a stationary phase was utilized in the analysis of volatile compounds of Rosa
damascena after extraction with superheated water, which is obligatory when extracting essential oils on a
laboratory scale. For elution, a mixture of hexane and ethyl acetate was used and there were 37 eluted and
identified components with percentage higher than 0.05%. The major ones were linalool, phenylethyl
alcohol, citronellol, nerol and geraniol [7].
An identical procedure with the same sorbent and eluent was applied for re extracting analytes from
the aqueous extract of Origanum onites, which are mainly carvacrol, borneol, terpinen-4-ol, -terpineol,
thymol and linalool [8], as well as in the work concerning essential oils of Achillea monocephala leaves
and flowers. The major compounds were camphor and borneol for the leaf oil and camphor, borneol, 1,8cineole and -campholenal for the flower oil [9]. Dichloromethane was used for elution of compounds
retained on C18 cartridge while isolating the volatile and semi-volatile compounds of Salvia officinalis
leaves infusion. These compounds were -thujone, camphor, 1,8-cineole, 6-oxobornyl acetate, -thujone,
1-borneol, exo-2hydroxycineole acetate [10].
Octadecyl sorbent was also proved to be the best sorbent for retaining hexanal and hexanol the
compounds responsible for beany flavour of soymilk. Many sorbents were tested considering polarity
of hexanol and hexanal reverse-phase sorbents were preferable, but only one adsorbed both mentioned
compounds, which were then eluted with methanol [11].
A solid phase extraction on a trifunctional silane SPE C18 cartridge was utilized to evaluate residues
of essential oil components in honey. Some of essential oils have been tested successfully against the mite
and are applied to control varroosis but the drawback of such a natural treatment is a high level of
residues found in honey that may change its taste. Retained compounds such as thymol, menthol,
eucalyptol, and camphor were eluted with acetone [12].
An endcapped C18 SPE cartridge and hexane as eluent were used for re extracting the analytes after
subcritical water extraction of essential oils from Thymbra spicata, which were carvacrol and thymol
more than 90%, (E)-car-3-en2-ol and enantiomers of -pinene. In this work, also the efficiency of the
C18 material was tested with steam-distilled sample of essential oil of T. spicata with a known
composition, which showed that there were almost no changes in the sample composition after the
application of SPE of C18 material [13].
Another cartridge used in the analysis of volatile compounds is a florisil cartridge with MgO/SiO2
phase that enables reversed phase extraction of polycyclic aromatic hydrocarbons. An example of such a
use can be a separation of Smyrnium olusatrum stem essential oil from aqueous phase after
hydrodistillation. Three fractions were collected using dichloromethane, dichloromethanemethanol and
methanol in turn. As NMR and GC-MS analysis showed, the first fraction was constituted by
furanodiene, while the second fraction was constituted by another compound which remained unknown
[14].
On-line coupling of solid-phase extraction or liquid chromatography with gas chromatography for the
analysis of biological samples reported [15]. Selection of a suitable on-line SPEGC or LCGC method

249

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

250

for the analysis of biological samples must take into account the type of sample matrix and the purpose of
the analysis.
In the present work solid-phase extraction followed by GC-mass spectrometry and gas
chromatography was used to re-extract and determined E-cinnamaldehyde, the main component of
Cinnamon bark oil. Thus at first response surface methodology (RSM) was used to optimize the
extraction yield and at the highest obtainable extract concentration. The extraction yield was defined as
the amount of E-cinnamaldehyde. In this regard, initially aqueous extract from cinnamon (C. zeylanicum)
bark as an essential oil-bearing material extracted by superheated water extraction (SWE) technique at
different temperatures, flow rates and particle sizes. Then essential oil extracted from aqueous extract by
solid phase extraction method and optimized the suitable elution solvent.
2. Material and Methods
2.1. Sample preparation
Cinnamon barks imported bulk from India were taken from the Medicinal Research Laboratory of
Iranian Research Organization for Science and Technology (IROST, Tehran, Iran). The moisture content
of the bark was 13.4% (dry basis). The samples were stored in polyethylene bags at 70oC until used.
The samples were ground within 3 min using a laboratory mill immediately prior to extraction to
minimize loss of volatiles. Ground cinnamon barks were sieved into mean particle size from 0.25 to 1.00
mm using ASTM standard sieves within 10 to 15 min.
2.2. Chemicals
Methanol and n-hexane (HPLC grade), ethyl acetate, acetone and ethanol (Aldrich Chemical Co.,
Milwankee, WI) were used in solid phase extraction (SPE) experiments. E-cinnamaldehyde (Merck,
Darmstadt, Germany) was used as authentic component for GC calibration purposes. Doubly distilled
water purified through a Milli-Q deionizing unit (Millipore, Bedford, MA) was used as extractant.
2.3 Hydrodistillation
The barks (50.0 g) were ground and extracted using Clevenger type hydrodistillation extractor to yield
(v/w, dry basis) 3.6% essential oils (mean of two replications). 200 mL of water for 3 h was used. To be
able to compare the results with SWEs and based on density calculation, 0.062 g of essential oil (1:25
dilution ratio) was dissolved in 2 mL n-hexane prior to the GC.
2.4 SWE and SPE systems
SWEs were carried out in a laboratory-built apparatus shown in Fig. 1 Detailed description of the
apparatus has been presented elsewhere [16]. It was mainly consists of 5 L de-ionized water feed tank
with N2 gas entering and purging capability (nearly 20 min), a high pressure metering pump (Eldex,
Optos series, Model 2S, USA, a cylindrical extraction chamber (103 mm 16 mm i.d.), and a double pipe
heat exchanger (cooling surface area: 240 cm2, cooling media: tap water, flow rate: 2 L/min). At the end
of each extraction run, 10 mL of the obtained extract was used for SPE and pre-concentration before GC

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

analysis. For all experiments, 2.0 g of ground cinnamon bark was used at 20 bar pressure to maintain the
water as a liquid at the extraction temperatures and the extraction time was set to 120 min.
Before GC analysis essential oil components from C. zeylanicum were extracted from SWE extracts
using SPE procedure. Solid phase extraction manifold made in our laboratory shown in Fig. 2 four
channel device, equipped with Waters Co. vacuum pump. The experiments were carried out using 360
mg cartridge Sep-Pak Classic C18 (Waters Co., UK). In SPE method each cartridge was conditioned by
passing 2 mL methanol and 2 mL of water, respectively and dried completely using vacuum. Then, 10
mL of superheated water extract was loaded on it and essential oil components were eluted with 2 mL of
elution solvent to 5 mL calibrated flask prior to GC.

Fig. 1. Schematic diagram of SWE extraction system; BA: balloon; BPR, back pressure regulator; BU: burette; C: N2 cylinder; EC:
extraction cell; HX: heat exchanger; MF, micro filter; NV: needle valve, OV: oven; P: high pressure pump; PI, pressure indicator;
TI, temperature indicator; V: ball valves; W: Water Vessel; WI, Water inlet; WO: Water Outlet.

251

252

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

Fig. 2. Solid Phase Extraction four channel device.

2.5 Chromatographic conditions

The extracts were analyzed using a PU-4500 gas chromatograph (Phillips Co., Cambridge, UK)
equipped with flame ionization detector and a 25-m PB-1 fused silica column (0.53 mm i.d., 1.0 mm film
thickness). The oven temperature program was a 3 oC/min temperature ramp from 55 to 200oC. The
carrier gas was helium (99.999%, Roham Gas Co., Tehran, Iran). The column head pressure was 0.27 bar.
The detector and injector temperatures were 250 and 240 oC, respectively. External standard solution was
E-cinnamaldehyde in methanol: H2O (70:30) and the concentration range was 50 to 3000 ppm.
GC-mass spectrometry (MS) analysis was conducted on a Varian Saturn model 3400 GC-MS system
(Varian Co., Palo Air., CA) equipped with a DB-5 fused silica column (30 m 0.25 mm, film thickness
of 0.25 m) and interfaced with a Varian ion trap detector. The GC conditions were: oven temperature
from 60 to 240 oC at 3 oC/min; injector and transfer line temperature, 250 and 260 oC; carrier gas, helium
at a flow rate of 1 mL/min; splitting ratio, 1:60. The detector temperature was maintained at 240 oC. The
MS conditions were: ionization energy, 70 eV; mass range, 40400 amu and scan mode electron impact
(EI). The percentage composition of the identified components was calculated from the GC peak area.
The components were identified by comparing their retention times and mass spectra with those of pure
reference components. Mass spectra were also compared with those in the NIST (National Institute of
Standards and Technology), WILEY5 and TERPENOIDES mass spectra libraries and our own created
library.

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

253

2.6 Experimental design

RSM was used to evaluate the effects of extraction temperature (T), superheated water flow rate (F)
and mean particle size (D) on the yield of essential oils in terms of E-cinnamaldehyde quantity. The
coded and uncoded independent variables used in the RSM design [17] and their respective levels were
listed in Table 1.
Table 1. Codes, ranges and levels of independent variables of temperature (T), superheated water flow rate (F) and mean particle
size (D) in RSM design
Symbols

Independent variables

Coded levels
-1

X1

T (oC)

100

130

160

X2

F (mL/min)

1.0

2.5

4.0

X3

D (mm)

0.250

0.625

1.000

The experimental design was based on the face centered central composite design using 8 cube points,
3 center points in cube and 6 axial points as shown in Table 2.

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

254

Table 2. Experimental and predicted data for the yield and extract concentration obtained from the central composite experimental
design
Run No.

X1

X2

X3

Yield (%), dry basis

Extract concentration (mg/mL)

Experimental

Predicted

Experimental

Predicted

100

1.0

1.000

1.29

1.42

0.186

0.193

100

1.0

0.250

0.72

0.71

0.104

0.114

160

4.0

1.000

2.49

2.49

0.090

0.081

130

2.5

0.625

1.32

1.68

0.076

0.101

130

2.5

1.000

2.51

2.68

0.145

0.157

160

1.0

1.000

1.89

1.87

0.273

0.273

130

2.5

0.625

2.12

1.68

0.122

0.101

160

1.0

0.250

1.45

1.73

0.209

0.221

130

1.0

0.625

1.63

1.25

0.235

0.206

10

160

2.5

0.625

1.43

1.30

0.083

0.084

11

130

2.5

0.625

1.65

1.68

0.095

0.101

12

100

2.5

0.625

0.83

0.98

0.048

0.041

13

130

2.5

0.250

2.26

2.12

0.131

0.114

14

100

4.0

0.250

1.89

1.90

0.068

0.069

15

130

4.0

0.625

1.75

2.16

0.063

0.087

16

100

4.0

1.000

3.16

2.88

0.114

0.103

17

160

4.0

0.250

2.23

2.09

0.081

0.075

All experimental trials were randomly performed to minimize the effect of unexplained variability in
the observed responses due to extraneous factors [18]. Three replicate runs at the center of the design
were performed to allow the estimation of pure error. Extraction yield (Y) and extract concentration (C)
were considered as the response functions (RE). It means that obtaining the essential oils at the highest
value and at the same time, collecting the most concentrated extract is desirable. A second order
polynomial equation (1) was used to express the response as a function of the independent variables,
RE = 0 +
+
+
(1)
where RE represents the response variables, 0 is a constant, i the linear coefficient, ii the quadratic
coefficient and ij the cross product coefficient. Xi and Xj are levels of the independent variables. The

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

model was built based on the variables with confidence levels of 95%. The coefficients of the response
surface equations were estimated by Minitab Version 15.1.30.
For separation part, in all experiments aqueous solution from superheated water extraction were used
to evaluate SPE method for the highest yield of E-cinnamaldehyde essential oils in cinnamon bark. Based
on E-cinnamaldehyde solubility in different solvents, a preliminary study for E-cinnamaldehyde
extraction by SPE method was performed. Solvents were selected to be methanol: water mixture (at
different ratios), n-hexane, ethyl acetate (EA), acetone and ethanol. GC analysis showed in Fig.3 that the
highest yield of E-cinnamaldehyde was obtained by methanol: water (60:40) mixture.

Fig. 3. GC overlay for different solvent for SPE extraction: line 5 Mix methanol:H2O (60:40), line 4 EtOH, Line3 Mix Hexane: EA,
line 2 EA, and line 1 Aceton.

Different ratio of methanol: water 80:20, 70:30, 60:40 (v/v) also studied. GC results showed in Fig. 4
that methanol: water ratio (70:30) (v/v) extracted the highest yield of E-cinnamaldehyde. In the finalized
procedure, each cartridge was washed with 2 mL methanol and 2 mL of water, respectively and dried
completely using vacuum. Then, 10 mL of superheated water extract was loaded on it and essential oil
components were eluted with 2 mL of methanol-water mixture (70:30) to 5 mL calibrated flask. GC
analyses were done by external standard method using standard calibration curve.

255

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

256

Fig. 4. GC overlay for different Mix methanol:H2O for SPE extraction: line1 methanol:H2O (80:20), line2 methanol:H2O (70:30)
and line3 methanol:H2O (60:40)(V/V)

3. Results and Discussion

3.1 Fitting the model
The predicted values of E-cinnamaldehyde yield and its concentration in the extract were calculated
using the regression model and compared with the experimental values, Table 2. The estimated
coefficients of the variables in the models are given in Table 3.

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

257

Table 3. Regression coefficients of the fitted quadratic equation and standard errors for the yield (Y) and extract concentration (C)

Yield, Y
Value

Standard error

-10.57870

4.00362

0.17960

0.06483

0.77610

Extract concentration, C
Value

Standard error

-0.70581

0.25072

0.028

0.01375

0.00406

0.012

0.63626

0.262

-0.05345

0.03985

0.222

-4.29130

2.54503

0.136

-0.11210

0.15938

0.505

11

-0.00060

0.00025

0.046

-0.00004

0.00002

0.026

22

0.01080

0.09836

0.916

0.02003

0.00616

0.014

33

5.1145

1.57372

0.014

0.23876

0.09855

0.046

12

-0.00460

0.00285

0.149

-0.00057

0.00018

0.016

13

-0.01270

0.01138

0.303

-0.00060

0.00071

0.425

23

0.11560

0.22768

0.627

-0.02018

0.01426

0.200

Regression

coefficient
0
Linear

Quadratic

Interaction

R2

0.853

0.946

R2 (adj)

0.664

0.877

The values of the coefficient of determination, R2, as a measure of the degree of fit, for yield (0.853)
and extract concentration (0.946) are both high enough to confirm that the model represents the
experimental results adequately. As represented in Table 3, it can be found that the variables with the
most significant effect on the yield were the linear and quadratic terms of temperature (p<0.05) and the
quadratic term of particle size (p<0.05). It is in accordance with that the extraction temperature is the key
factor in SWE process [5].
3.2 Optimization of extraction condition
The optimum operating conditions was determined by defining the highest values achievable for the
yield and extract concentration. The yield and extract concentration in terms of the main component of
cinnamon bark, E-cinnamaldehyde, would be considered optimum with minimum yield of 1.8% and
minimum concentration of 0.12 mg/mL as mean values in the whole 17 experiments. The criteria used

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

258

for optimization included maximum yield and maximum extract concentration. Considering software
generated results for the responses, the optimum combined condition was found to be at 138oC, 1 mL/min
and 1 mm. The model prediction capability for optimum response values was verified using generated
operating conditions for temperature, flow rate and mean particle sizes. Experimental and predicted
values for the yield and extract concentration were 2.69 and 2.21%, and 0.318 and 0.280 mg/mL,
respectively. The results show that acceptable agreement exists between values calculated using the
model and experimental values.
After preliminary study for SPE extraction, optimum conditions on the basis of highest Ecinnamaldehyde yield determined. As GC results showed nonpolar like n-hexane and midpolar solvent
like mix (Hexane: EA) had no good separation for cinnamon bark essential oil. The best extraction
obtained by mix methanol and water. Comparison with different ratio of Polar elution solvent (Methanol:
H2O) in Fig.4 determined that the amount of extraction were close to each other, but the best one were
methanol-water mixture (70:30). For SPE the all separation carried out by washing each cartridge with 2
mL methanol and 2 mL of water, respectively and dried completely using vacuum. Then, 10 mL of
superheated water extract was loaded on it and absorbed components were eluted with 2 mL of methanolwater mixture (70:30).
3.3 Comparison with hydrodistillation method
In Table 4, the results of comparison among the SWE and hydrodistillation have been shown. As can
be seen, up to 8 components including Z-cinnamaldehyde and E-cinnamaldehyde as the main oxygenated
components (C9H8O) and 6 other sesquiterpenes (C15H24) were identified in the samples. These
components make up more than 95% of the oil. The mean relative standard deviation per peak was
calculated to be 15%.
Table 4. Comparative composition (%) of the volatiles of cinnamon bark extracted by SWE and hydrodistillation methods.
RIa

SWEb,*

Hydrodistillationc,*

Z-cinnamaldehyde
E-cinnamaldehyde
-Copaene

1216
1278
1379

0.603
88.507
8.601

0.688
79.600
7.234

E-Caryophyllene
Germacrene D
-Muurelene
-Cadinene
Cubenene

1421
1482
1501
1524
1535

t
t
t
0.668
1.620

0.737
0.895
3.058
6.298
1.490

Components

Retention indices on the DB-5 column

Sample = 2.0 g, Temperature = 138 oC, Flow rate = 1.0 mL/min, Particle size = 1.000 mm, Pressure = 20 bar, Extraction time =
120 min.
c
Extraction time = 3 hr,
*
t = trace < 0.01
b

The extracts compositions as E-cinnamaldehyde percentage were 88.507% and 79.600% for SWE and
hydrodistillation, respectively. It showed that the SWE extract was richer in E-cinnamaldehyde as the
main and desirable component of cinnamon bark essential oil. Considering cinnamon bark essential oil
density (1.05 g/mL) and E-cinnamaldehyde percentage (79.600%), it may be concluded that the
extraction yield based on that component was higher for hydrodistillation (3.00%) than that for the
optimized SWE (2.69%). On the other hand, except for -copaene and cubenene, undesirable non-

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

oxygenated components like E-caryophyllene, germacrene D, -muurelene, and -cadinene had

significantly lower values in the SWE method. It may be the result of low solubility of such
sesquiterpenes in water. These facts showed that SWE extracted more valuable oxygenated components
selectively.
4. Conclusions
The main essential oils of cinnamon bark were successfully extracted by solid phase extraction method
followed by gas chromatography. C18 SPE cartridge and mix methanol:water as eluent were used for reextracting the essential oils after the superheated water extraction from cinnamon bark. As GC results
showed nonpolar and midpolar solvents were not good extractants for cinnamon bark essential oils. The
best extraction obtained by polar solvents like mix methanol: water (70:30). Thus results show that SPE
is a powerful method for sample separation and it may be useful in this area. The SPE method has been
applied as an alternative to liquid-liquid extraction for simplicity, low cost and has many advantages in
comparison with more traditional sample preparation techniques. Also, the SWE process was optimized
using RSM to extract essential oil components of cinnamon bark and compared with hydrodistillation
method.
Acknowledgments
The financial support of Iranian Research Organization for Science and Technology (IROST) and
Iranian National Science Foundation (INSF), contract no. 89/S/14926, is gratefully acknowledged.
References
[1] Camel V. Solid phase extraction of trace elements. Spectrochimica Acta 2003;B58:1177-1233.
[2] Guide to Solid Phase Extraction. Supelco Bulletin 1998;910.
[3] Zief M, Kiser R. Solid Phase Extraction for Sample Preparation JT Baker 1994.
[4] Antonelli A, Fabbri C. Essential Oils: SPE Fractionation. Chromatographia 1999;49:125-130.
[5] Smith RM. Extractions with superheated water. J Chromatogr A 2002;975:31-46.
[6] zel MZ, Clifford A. Superheated water extraction of fragrance compounds from Rosa canina. Flavour Fragr J 2004;19:
354-359.
[7] zel MZ, Gogus F, Lewis AC. Comparison of direct thermal desorption with water distillation and superheated water
extraction for the analysis of volatile components of Rosa damascena Mill using GCxGC-TOF/MS. Anal Chim Acta
2006;556:172-177.
[8] zel MZ, Kaymaz H. Superheated water extraction, steam distillation and Soxhlet extraction of essential oils of Origanum
onites. Anal Bioanal Chem 2004;379:1127-1133.
[9] zel MZ, Gogus F, Lewis C. Extraction of essential oils of leaves and flowers of Achillea monocephala by superheated
water. Flavour Fragr J 2006;21:122-128.
[10] Radulescu V, Chiliment S, Oprea E. Capillary gas chromatographymass spectrometry of volatile and semi-volatile
compounds of Salvia officinalis. J Chromatogr A 2004;1027:121-126.
[11] Wang ZH, Dou J, Macura D, Durance TD, Nakai S. Solid phase extraction for GC analysis of beany flavours in soymilk.
Food Res Int 1998;30:503-511.
[12] Adamczyk S, Lazaro R, Perez-Arquillue C, Conchello P, Herrera A. Evaluation of residues of essential oil components in
honey after different anti-varroa treatments. J Agric Food Chem 2005;53:10085-10090.

259

F. Golmohammad et al. / Procedia Engineering 42 (2012) 247 260

260

[13] zel MZ, Gogus F, Lewis AC. Subcritical water extraction of essential oils from Thymbra spicata. Food Chem 2003;82:
381-386.
[14] Bertoli A, Pistelli L, Morelli I, Fraternale D, Giamperi L, Ricci D. Volatile constituents of different parts (roots, stems and
leaves) of Smyrnium olusatrum L. Flavour Fragr J 2004;19:522-525.
[15] HytylinenT, Riekkola M L. Solid-phase extraction or liquid chromatography coupled on-line with gas chromatography in
the analysis of biological samples. J Chromatogr B 2005;817:1321.
[16] Mortazavi SV, Eikani MH, Mirzaei H, Jafari M, Golmohammad F. Extraction of essential oils from Bunium persicum
Boiss. using superheated water. Food Bioproducts Proc 2010;88:222-226.
[17] Liu G, Xu X, Hao Q, Gao Y. Supercritical CO2 extraction optimization of pomegranate (Punica granatum L.) seed oil
using response surface methodology. LWT- Food Sci Technol 2009;42:1491-1494.
[18] Sin HN, Yusof S, Sheikh Abdul Hamid N, Abd Rahman R. Optimization of hot water extraction for sapodilla juice using
response surface methodology. J Food Eng 2006;74:352-358.