Journal of Chromatographic Science 2015;53:1222– 1231
doi:10.1093/chromsci/bmu216 Advance Access publication January 29, 2015
Article
Application of Optimized Vortex-Assisted Surfactant-Enhanced DLLME for Preconcentration
of Thymol and Carvacrol, and Their Determination by HPLC-UV: Response Surface
Methodology
Mehrorang Ghaedi1, Mostafa Roosta2*, Saeid Khodadoust3 and Ali Daneshfar4
1
Chemistry Department, Yasouj University, Yasouj 75918-74831, Iran, 2Young Researchers and Elite Club, Sepidan Branch, Islamic Azad
University, Sepidan, Iran, 3Behbahan Khatam Alanbia University of Technology, Behbahan, Iran, and 4Department of Chemistry,
University of Ilam, Ilam 65315-516, Iran
*Author to whom correspondence should be addressed. Email: [email protected]
Received 16 February 2014; revised 14 November 2014
A novel vortex-assisted surfactant-enhanced dispersive liquid–liquid
microextraction combined with high-performance liquid chromatography (VASEDLLME–HPLC) was developed for the determination of thymol and carvacrol (phenolic compound). In this method, the extraction
solvent (CHCl3) was dispersed into the aqueous samples via a vortex
agitator and addition of the surfactant (Triton X-100). The preliminary
experiments were undertaken to select the best extraction solvent and
surfactant. The influences of effective variables were investigated
using a Plackett–Burman 27 – 4 screening design and then, the significant variables were optimized by using a central composite design
combined with desirability function. Working under optimum conditions specified as: 140 mL CHCl3, 0.08% (w/v, Triton X-100), 3 min extraction time, 6 min centrifugation at 4,500 rpm, pH 7, 0.0% (w/v) NaCl
permit achievement of high and reasonable linear range over 0.005–
4.0 mg L21 with R 2 5 0.9998 (n 5 10). The separation of thymol and
carvacrol was achieved in <14 min using a C18 column and an isocratic
binary mobile phase acetonitrile–water (55:45, v/v) with a flow rate of
1.0 mL min21. The VASEDLLME is applied for successful determination
of carvacrol and thymol in different thyme and pharmaceutical samples
with relative standard deviation <4.7% (n 5 5).
Introduction
Thymol (2-isopropyl-5-methylphenol) and carvacrol (5-isopropyl2-methylphenol) are two major constituents of thyme oil, essential oil of Origanum vulgare (oregano), wild bergamot and plants
including Thymus vulgaris (1, 2). These compounds as natural additives, applied in many foods (as flavorings), perfumes and pharmaceuticals due to their antitussive, antibacterial, antifungal,
antioxidant, anticancer and anti-carcinogenic properties (3 –6).
As they are used for standardization of pharmaceutical compounds
based on their thymol or carvacrol contents, the development a
new validated method for their extraction and determination is
a challenging requirement (7).
Various analytical techniques such as thin-layer chromatography (TLC) combined with densitometry (8), gas chromatography
(GC) (4, 9, 10), gas chromatography–mass spectrometry (GC-MS)
(11, 12) and high-performance liquid chromatography (HPLC)
with fluorimetric detection (13) have been applied for their determination in various matrixes. In some cases, their lower content in
complicated matrices make an emergency task to preliminary application of separation and preconcentration techniques (14).
This goal can be achieved by combination of a novel and
environment-friendly sample preparation method with advanced
instruments to analysis the analytes content with higher accuracy
(15, 16).
Various procedures such as dispersive liquid –liquid microextraction (DLLME) (17, 18), supported liquid membrane (SLM)
(19), hollow fiber liquid-phase microextraction (HF-LPME)
(20), solid-phase microextraction (SPME) (21), liquid-phase
microextraction (LPME) (22) and matrix solid-phase dispersion
(MSPD) (23) are described and developed for prior separation
and/or preconcentration of target compounds. DLLME as a miniaturized sample pretreatment technique using microliter volumes of the extraction solvent along with few milliliters of
dispersive solvents was developed by Assadi et al. (24).
Because of relatively high volume of disperser solvent in
DLLME, the distribution and extraction of the analyte into the
extractant phase were decreased. This limitation can be solved
by the development of new methods and reducing and/or elimination of the content and amount of disperser solvent (25 –27).
For this purpose, some novel microextraction methods such as
vortex-assisted liquid – liquid microextraction (VALLME) (28)
and vortex-assisted dispersive liquid – liquid microextraction
(VADLLME) have been developed (29). These methods are
based on dispersion of the extraction solvent into aqueous samples by vortex mixing as a mild emulsification procedure. In
these methods, the main restriction of DLLME (the application
of dispersive solvent) and the analytes degradation was resolved
by omitting the dispersive solvent, reducing extraction time and
solvent consumption and finally increase the extraction yields
and improve the quality of the extracts. In recent decades, the
uses of the surfactants (amphiphilic molecules), as green extraction solvents, have been developed (30, 31). These compounds
have unique physicochemical properties such as high solubility
in water and organic solvents, good solvating ability of organic
and inorganic compounds, high thermal stability, as emulsifier to
enhance the dispersion of water-immiscible phases and accelerate
the formation of fine droplets of the extraction solvent in an aqueous sample solution. The application of a surfactant as emulsifier
decreases the interfacial tension between two phases (bridging
between them) and can contribute in dispersion of organic solvent into aqueous phase (32–34).
There are several experimental variables affecting the vortexassisted surfactant-enhanced dispersive liquid – liquid microextraction (VASEDLLME) procedure. In a one-variable-at-a-time
(OVAT) approach, every related single variable varied, while all
other variables are kept fixed at a specific set of conditions
# The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
without attention to the any interaction between the variables.
However, this procedure requires a high number of experiments
(high consumption of reagent) and it is also time consuming.
A good selection of design and optimization models makes possible to simultaneously evaluate the variables influence during
the extraction process (35). Multivariate approaches based on
experimental design are applicable to simultaneously evaluate
the main effect and variables interaction with at least number
of runs (36 –38).
In this study, VASEDLLME as a novel, simple, sensitive, inexpensive and rapid/assisted extraction method has been developed
for preconcentration of thymol and carvacrol and then HPLC
and UV detection was applied for their determination in thymes
and pharmaceutical samples. The influence of variables and their
interactions on VASEDLLME were investigated by experimental
design. A Plackett – Burman (P – B) screening design was used
to study the main variables [pH, concentration of surfactant, volume of extraction solvent, ionic strength (NaCl %), vortex time,
centrifugation speed and centrifugation time] that affect the
VASEDLLME process and then the response surface methodology
(RSM) was used to optimize the significant variables by using the
desirability function (DF). To the best of our knowledge, this study
is the first report describing the application of an optimized
VASEDLLME liquid chromatography method for determination
of thymol and carvacrol in thymes and pharmaceutical samples.
Experimental
Reagents and material
Carvacrol (.99%) as a liquid and thymol (.99%) as a crystal
solid were purchased from Sigma-Aldrich (Tulsa, OK, USA).
HPLC-grade acetonitrile and reagent-grade sodium dodecyl sulfate (SDS), sodium dihydrogen phosphate, sodium monohydrogen phosphate, phosphoric acid and sodium hydroxide were
supplied from Merck (Darmstadt, Germany). Triton X-100 (isooctyl phenoxy polyethoxy ethanol), Triton X-114 (octyl phenol
polyethylene glycol ether) and Triton X-405 were purchased
from Aldrich (USA). Cetyl trimethylammonium bromide
(CTAB) was supplied from Acros Organics (Geel, Belgium).
Doubly distilled, deionized water was used in all experiments.
A stock solution of 1,000 mg L21 of carvacrol and thymol was
prepared by dissolving 0.1 g of the reagents in 100.0 mL of acetonitrile in a volumetric flask. All of the standard solutions were
stored at 48C and brought to ambient temperature just prior to
use. A solution of Triton X-100 (0.2 mol L21) was prepared by
dissolving 6.25 g of Triton X-100 in water and diluting to
50.0 mL in a volumetric flask. Filtering of all solutions was carried
out using 0.45 mm membranes (Millipore, Bedford, MA, USA).
Apparatus
The chromatographic measurements were carried out with an
Agilent Technologies (Wilmington, DE, USA), 1100 HPLC system
equipped with Standard Micro Auto Sampler (model G1313A),
Micro Vacuum Degasser (model G1379A), Quaternary Pump
(model G1311A), Series Multiple Wavelength Detector (model
G13658) and a Zorbax-eclipse XDB-C18 (250 mm 4.6 mm
ID, 5 mm) column. The chromatographic calculations were performed using a Chemstation data handling system. The mobile
phase was 45.0% (v/v) water in acetonitrile with a flow rate of
1.0 mL min21 and peaks were detected at 210 nm. A vortex agitator system (Retsch, Germany) at 2,500 rpm was used for the
vortex-assisted extraction. A Hermle Labortechnik GmbH centrifuge (model Z206A, Germany) was used to accelerate the phase
separation.
Sample preparation
Antitussive syrup samples such as Thymian (200 mL; mina laboratory, Tehran, Iran) and Broncho T.D. (200 mL; Tolid Daru
Company, Tehran, Iran) were purchased from a pharmacy and
stored in the dark medium at 48C. The distillate thyme sample
with edible quality purchased from a local supermarket and
stored in dark medium. All samples were filtered using 0.45 mm
Millipore membranes and then 50 mL of these samples were used
for the extraction procedure.
VASEDLLME
The steps of the VASEDLLME procedure (Figure 1) are carried
out as follows: 140 mL of CHCl3 (as extraction solvent) was rapidly injected into the 5.0 mL of aqueous sample that contained
0.02 mg L21 of each compound and 0.08 mg mL 21 of Triton
X-100 (as emulsifier) in a 10-mL screw cap glass tube with conical bottom. The tube was capped immediately, and the mixture
was then vigorously shaken using a vortex agitator for 3.0 min at
2,500 rpm. Fine droplets were formed during the vortex agitation process which facilitated mass transfer of the analytes
from the aqueous sample to the extraction solvent. Phase separation was achieved by centrifugation for 6.0 min at 4,500 rpm. The
CHCl3 phase (130 mL) sedimented at the bottom of the centrifuge tube completely transferred to a HPLC sample vial using
200.0 mL HPLC syringe (Hamilton) and blown to dryness with
a mild nitrogen stream. The residue (as settled phase) was dissolved in 250 mL of acetonitrile and 5.0 mL of this sample was
injected into the HPLC system for subsequent analysis. In all
real samples, the thymes content was evaluated by a standard
addition method.
Experimental design
Experimental design was developed to decrease the number of
experimental runs and consider simultaneous interaction of variables to achieve true optimum points (39, 40). An experimental
Plackett – Burman (P – B) design was built for the screening of
the main factors affecting the extraction recovery (ER) of thymol
and carvacrol. A P – B design can examine up to N 2 1 factors
(f N 2 1) in N experiments (N is a multiple of 4) (41, 42).
Following preliminary evaluation of the significant factors, the
optimum working conditions attained using central composite
designs (CCDs) (43, 44). A CCD combines a 2f factorial design
with additional points (star points) and at least one point at the
center of the experimental region to obtain properties such as
rotatability or orthogonality, in order to fit quadratic polynomials
(45). The star points are located at þ a and – a from the center of
the experimental domain (46). An axial distance, a, was selected
with a value of 2.0 in order to establish the rotatability condition
of the CCD. The total number of design points needed (N) is
determined by the following equation
N ¼ 2f þ 2f þ N0 ;
ð1Þ
Response Surface Methodology 1223
Figure 1. Schematic diagram of the VASEDLLME procedure.
where f and N0 are the number of variables and center points
(f ¼ 4, N0 ¼ 6), respectively (47). The mathematical relationship
between the four independent variables can be approximated by
the second-order polynomial model (48)
y ¼ b0 þ
4
X
i¼1
bi xi þ
4 X
4
X
i¼1 j ¼1
bij xi xj þ
4
X
dfi ¼
bii xi2 ;
ð2Þ
i¼1
where y is the predicted response (ER); and Xi s are the independent variables (surfactant, modifier, sonication time and centrifugation time) that are known for each experimental run. The
parameter b0 is the model constant; bi is the linear coefficient;
bii are the quadratic coefficients and bij are the cross-product
coefficients.
DF
DF is a common and established technique for concurrent
evaluation of input variables to create a function for each individual response di and finally determined a global DF (49, 50).
The desirability procedure involves three steps: (i) predicting
responses on the dependent variable by fitting the observed
responses using an equation based on the levels of the independent variables, (ii) finding the levels of the independent
variables that simultaneously produce the most desirable
predicted responses on the dependent variables and (iii) maximize the overall desirability with respect to the controllable
variables. In DF, the response (U) is converted into a particular desirability function (df i ) in the range of 0 to 1 (it is
1224 Ghaedi et al.
better that value limited to one). The individual DF for
the ith characteristic is computed via the following equation
(49, 50)
U a wi
; aU b
ba
dfi ¼ 1; U . b
ð3Þ
dfi ¼ 0; U , a
In Equation (3), a and b are the lowest and highest obtained
values of the response and wi is the weight. The individual
desirability scores for the predicted values combined into
overall DF by computing their geometric mean of different
dfi values.
DF ¼ ½df1v1 df2v2 dfnvn 1=n ;
0 vi 1 ði ¼ 1; 2; : : :; nÞ
n
X
ð4Þ
vi ¼ 1;
i¼1
where dfi indicates the desirability of the response Ui (i ¼ 1, 2,
3,. . . , n) and vi represents the importance of responses.
Calculation of ER
The ER% was defined as the percentage of the total analyte,
which was extracted in the settled phase and can be calculated
according to following equation
Cset Vset
ER% ¼
100;
C0 Vaq
ð5Þ
where Vset and Vaq are the volume of the settled phase and the
volume of the aqueous sample, respectively. C0 and Cset are the
concentration of the aqueous sample and the settled phase,
respectively.
Results and Discussion
The VASEDLLME efficiency can be influenced by several variables
such as pH, ionic strength, type and concentration of extraction
solvents, type and concentration of the surfactant, vortex time,
speed and time of centrifuge. One of the major challenges in
the utilization of microextraction is that the selection of experimental conditions provides acceptable response at low analyte
concentration. To obtain the optimum conditions for the extraction of thymol and carvacrol, a combination of P – B screening
design and CCD followed by DF was used for the optimization
of experimental variables. For this purpose, STATISTICA 7.0 statistical package was used to generate the experimental matrix and
to evaluate the results.
Selection of surfactant and extraction solvent
Before transmission of P – B design, the preliminary experiments were undertaken to select the best surfactant. The selected surfactant must to have the properties such as high
solubility in all phases, miscible with the both organic solvent
and aqueous sample and accelerate the emulsification of the
organic solvent into the aqueous samples. Surfactants decreased the interfacial tension between two liquids and their
amphipathic structure and adjust hydrophilicity as well as lipophilicity of solution to serve as an emulsifier. The surfactants
CTAB, SDS, Triton X-100, Triton X-114 and Triton X-405 were
considered for this purpose at the condition of 0.02 mg L 21
each analyte, 4.0 min of vortex time, centrifugation time
7.0 min, 200.0 mL of extraction solvent, 0.05 mg mL21 of each
surfactant, 2% NaCl and pH 7. Non-ionic surfactants show the
best extraction efficiencies (Table I). It seems that non-ionic
surfactants have a larger solubilization capacity and appropriate hydrophobicity for the target analytes compared with
ionic surfactants (30). Among non-ionic surfactants, Triton
X-100 shows better efficiency than Triton X-114 and/or
Triton X-405.
The selection of an appropriate extraction solvent is of great
importance in VASEDLLME to obtain high extraction efficiency.
The extraction solvent should have a high capacity for extraction
of target components, immiscibility with water and additionally
has density higher than the aqueous phase. To obtain a satisfactory preconcentration and extraction efficiency, 200.0 mL of
three organic solvents including CHCl3 (density 1.48 g mL21),
CH2Cl2 (density 1.33 g mL21) and CCl4 (density 1.59 g mL21)
were applied, and the results are shown in Table I. It is clear
that the higher recovery were achieved using CHCl3 as an extraction solvent. This may be attributed to the same polarity of target
analytes with CHCl3 and higher solubility of them in CHCl3.
Therefore, according to the obtained results CHCl3 was chosen
as the extraction solvent in subsequent studies.
Table I
Mean Recoveries and RSDs Obtained from VASEDLLME Technique Using Different Surfactants and
Extraction Solvent (n ¼ 3)a
Surfactant type
Extraction solvent
Volume of sedimented
phase (mL)
CHCl3
CCl4
CHCl2
CHCl3
CCl4
CHCl2
Triton X-100
Triton X-114
Triton X-405
CTAB
SDS
83.2b + 3.8c
68.7 + 4.2
65.3 + 3.7
38.0 + 4.1
27.5 + 3.2
76.5 + 2.8
64.2 + 3.4
63.3 + 4.1
31.0 + 3.8
25.0 + 3.6
71.0 + 4.6
62.7 + 3.8
56.3 + 4.3
30.4 + 2.7
23.0 + 2.8
165
160
160
135
120
165
155
160
135
110
165
155
160
130
110
a
Conditions: 0.02 mg L21 each analyte; vortex time 4 min; centrifugation time 7 min; volume of
extraction solvent 200 mL; 0.05 mg mL21 surfactant; NaCl (2%); pH 7.
b
ER (%).
c
RSD.
Table II
Factors, Codes, Low and High Levels in 27 – 4 Plackett –Burman Design Matrix
Factors
(X1)
(X2)
(X3)
(X4)
(X5)
(X6)
(X7)
pH value
Ionic strength (NaCl concentration; w/v) (%)
Vortex time (min)
Centrifugation time (min)
Centrifugation speed (rpm)
Volume of CHCl3 (extraction solvent) (mL)
Triton X-100 (mg mL21)
Levels
Low (21)
High (þ1)
0
1
1
1500
50
0.03
5
9
11
5500
200
0.07
Plackett– Burman design
The development of an optimized method requires plenty of
experiments that increase exponentially with the number of independent variables. To decrease the number of experiments, a
decrease in dimensions of independent variables was considered
in a series of preliminary-screening experiments. As shown in
Table II, a 27 – 4 P– B design with two levels were undertaken to
evaluate the significance of seven variables (36). The analysis of
these results produced the standardized main effect Pareto
charts (P ¼ 95%) that are shown in Figure 2. The bar length is
proportional to the significance of the variables for ER. The results indicate that the vortex time, volume of extraction solvent
(CHCl3) and surfactant (Triton X-100) were the most significant
variables with a positive effect on the ER. Other significant variable was the ionic strength (NaCl %) that has negative effect on
the ER. In this procedure, addition of ionic strength causes inhibition of dispersion of extraction solvent by surfactants in aqueous samples and the ER is low in the presence of the salt.
According to the obtained results, centrifugation time, pH and
centrifugation speed had no significant effect on the ER and
are not considered for further studies in CCD.
CCD
In the CCD step, the plan of experiments was run in a random
manner in order to minimize the effect of uncontrolled variables.
As presented in Tables III and IV, four independent variables [the
extraction solvent (X6) and surfactant volume (X7), vortex time
(X3) and ionic strength (X2)] were prescribed in three levels
(low, basal and high) with coded value (21, 0, þ1) and the
star points of þ2 and 22 for þa and – a, respectively, were selected for each set of experiments (Table III). The total 30
Response Surface Methodology 1225
Figure 2. Standardized main effect Pareto chart for the Plackett–Burman design of screening experiment. The vertical line in the chart defines 95% confidence level.
Table III
Experimental Factors and Levels in the CCD
Factors
(X2)
(X3)
(X6)
(X7)
Table V
ANOVA for CCD
Levels
NaCl concentration (w/v, %)
Vortex time (min)
Volume of CHCl3 (mL)
Triton X-100 (mg mL21)
Star point,
a ¼ 2.0
Low (21)
Central (0)
High (þ1)
2a
þa
1.5
2.0
100
0.03
3.0
3.5
150
0.05
4.5
5.0
200
0.07
0
0.5
50
0.01
6
6.5
250
0.09
Table IV
Experimental Conditions and ER Values Obtained Through the CCD
Runs
X6
X7
X2
X3
ER (%)
24
16
15
18
22
5
23
29 (c)
26
12
2
10 (c)
27
6
17
13
20 (c)
9 (c)
11
8
19 (c)
14
1
7
30 (c)
4
28
3
21
25
150
200
200
200
250
200
150
150
150
100
100
150
150
200
200
100
150
150
100
200
150
100
100
200
150
100
150
100
50
150
0.09
0.03
0.03
0.07
0.05
0.03
0.01
0.05
0.05
0.03
0.03
0.05
0.05
0.03
0.07
0.07
0.05
0.05
0.03
0.07
0.05
0.07
0.03
0.07
0.05
0.07
0.05
0.05
0.05
0.05
3.0
4.5
1.5
4.5
3.0
1.5
3.0
3.0
6.0
4.5
4.5
3.0
3.0
4.5
1.5
1.5
3.0
3.0
1.5
4.5
3.0
4.5
1.5
1.5
3.0
4.5
3.0
1.5
3.0
0.0
3.5
2.0
5.0
5.0
3.5
2.0
3.5
3.5
3.5
5.0
2.0
3.5
0.5
5.0
3.5
5.0
3.5
3.5
2.0
2.0
3.5
2.0
5.0
5.0
3.5
5.0
6.5
2.0
3.5
3.5
93.7
88.5
89.8
84.2
89.4
89.5
89.2
87.9
85.7
77.2
80.0
88.4
85.5
90.6
89.7
85.5
92.3
87.5
80.5
84.5
90.7
86.5
74.0
93.3
92.2
72.5
93.0
79.9
57.0
92.5
(c), Center point.
1226 Ghaedi et al.
Source of
variation
Sum of
square
Degree of
freedom
Mean
square
F-value
P-value
X6
X7
X2
X3
X62
X72
X22
X32
X6X7
X6X2
X6X3
X7X2
X7X3
X2X3
Lack of fit
Pure error
Total error
709.595
519.676
13.316
1.470
46.693
4.354
0.301
1.833
43.701
0.844
41.348
38.609
1.444
7.668
198.098
23.673
1757.159
1
1
1
1
1
1
1
1
1
1
1
1
1
1
10
5
29
709.5948
519.6761
13.3162
1.4698
46.6930
4.3545
0.3011
1.8334
43.7013
0.8444
41.3476
38.6088
1.4439
7.6676
19.8098
4.7347
149.8722
109.7598
2.8125
0.3104
9.8619
0.9197
0.0636
0.3872
9.2301
0.1783
8.7329
8.1545
0.3050
1.6195
4.1840
0.000064
0.000137
0.154378
0.601439
0.025656
0.381597
0.810952
0.561037
0.028813
0.690351
0.031696
0.035589
0.604557
0.259133
0.063833
experiments were performed according to the CCD, which were
determined based on preliminary experiments and their responses are presented in Table IV. To find the main, interaction and
quadratic effects, analysis of variance (ANOVA) was calculated
using STATISTICA 7.0 (Table V). A P-value ,0.05 in the ANOVA
table indicates the statistical significance of an effect at 95% confidence level. The ‘lack of fit (LOF) P-value’ of 0.06 implies that
the LOF is not significant relative to the pure error. The F-test
was used to estimate the statistical significance of all terms in
the polynomial equation within 95% confidence interval. Data
analysis gave a semi-empirical expression of ER% with the
following equation
ER% ¼ 89:42 þ 5:49x1 þ 0:79x2 1:44x4 4:32x12
þ 2:75x42 1:74x1 x2 1:71x1 x3 þ 1:70x1 x4
0:75x3 x4 :
ð6Þ
In the next step of the design, RSM was developed by considering
all the significant interactions in the CCD to optimize the critical
Figure 3. Response surface plots of ER (%) versus significant variables. These plots were obtained for a given pair of factors at fixed and optimal values of other variables: (A) Triton
X-100–CHCl3; (B) NaCl– CHCl3; (C) Vortex time–CHCl3; (D) Triton X-100–NaCl; (E) Triton X-100–Vortex time and (F) NaCl–Vortex time.
factors. Figure 3 shows the most relevant fitted response surfaces
for the design and depicts the response surface plots of ER (%)
versus significant variables. These plots were obtained for a given
pair of factors at fixed and optimal values of other variables. The
curvatures of these plots indicate the interaction between the
variables. The surface plots (Figures 3A – E) show that at low
CHCl3 volume and low percent of Triton X-100 the ER% is low
(40 – 80%). It seems that at low volumes of CHCl3, complete
Response Surface Methodology 1227
Figure 4. Profiles for predicated values and desirability function for ER of thymol and carvacrol. The dashed line indicates current values after optimization.
phase separation was not formed. As can be seen from
Figure 3A – E with increase of Triton X-100, the ER% increases
and reaches a maximum value at 140 – 190 mL of CHCl3.
Figures 3C and F indicate that increasing the vortex time leads
to slow the increase in ER% and with elevating the NaCl percent,
the ER% decreases rapidly. The NaCl percent has negative correlation with the ER% (Figure 3B, D and F).
Optimization of CCD by DF for the extraction procedure
The profile for predicted values and desirability option in the
STATISTICA 7.0 software was used for the optimization process
(Figure 4). Profiling the desirability of responses involves specifying the DF for each dependent variable (ER%) by assigning predicted values. The scale in the range of 0.0 (undesirable) to 1.0
(very desirable) is used to obtain a global function (D) that
should be maximized according to efficient selection and optimization of designed variables. The CCD design matrix results
(Table IV) show the maximum (93.7%) and minimum (57.0%)
ER of thymol and carvacrol. According to these values, DF settings for each dependent variable of ER% are depicted on the
right hand side of Figure 4: desirability of 1.0 was assigned for
maximum ER% (93.7%), 0.0 for minimum (57.0%) and 0.5 for
middle (75.4%). Because desirability 1.0 was selected as the target value, the overall response (ER%) obtained from these plots
with the current level of each variable in the model is depicted
at the top (left) of Figure 4. These figures show that variables affect simultaneously the response (ER%) and its desirability. On
the basis of these calculations and the desirability score of 1.0,
maximum recovery (97.6%) was obtained under optimized
1228 Ghaedi et al.
conditions set as follows: 0.08 mg mL21 Triton X-100, 140 mL
CHCl3, 3.0 min of vortex time and no addition of salt (0.0%
NaCl). The validity of duplicate assenting experiments at the optimized value of all parameters was investigated. The results are
closely co-related with the data obtained from desirability optimization analysis using CCD. It was seen that the CCD with DF
is efficiently applied for optimization of the ER% of the target
analytes.
Analytical performance of VASEDLLME
At optimum experimental conditions, the validity of the proposed
method was examined by conducting a set of similar experiments
at 10 concentration levels of 0.005, 0.01, 0.02, 0.04, 0.08, 0.2, 0.4,
0.8, 2.0 and 4.0 mg L21 and the calibration curve was plotted at
three replicated extractions. The limit of detection (LOD) and
limit of quantification (LOQ) were calculated as 3 and 10 times
the standard deviation of 10 replicate runs of samples spiked
with low concentration of analytes (0.005 mg L21). The calibration plots were linear in the range of 0.005 –4.0 mg L21 with a
LOD of 1.6 mg L21and the LOQ of 5.0 mg L21 in water samples
with correlation coefficients (r 2) ranged from 0.9997 to 0.9999.
The repeatability study was carried out by performing three
parallel replicated extractions at the intermediate concentration
level (0.05 mg L21 for each analyte) under the optimal conditions. Repeatability and reproducibility studies were carried out
under the optimal conditions. The actual amounts of thymol and
carvacrol in all samples were evaluated by the standard addition
method at three replicate experiments, and the results are given
in Table V. The repeatability resultant expressed as relative
Figure 5. Chromatograms of carvacrol and thymol by HPLC at optimum extraction conditions: column, Zorbax-eclipse XDB-C18 (250 mm 4.6 mm, 5 mm); mobile phase,
acetonitrile – water (55:45, v/v); flow rate 1.0 mL min21. (a) Broncho T.D. sample; (b – e) Broncho T.D. sample spiked with 0.8, 2.0 and 4.0 mg L21 of carvacrol and thymol,
respectively. Peaks: carvacrol (1) and thymol (2).
Table VI
Extraction Recoveries and RSD in Different Samples at Spiked Level by the VASEDLLME –HPLC Method
Compounds
Carvacrol
Added (mg L21)
0.00
0.8
2.0
4.0
0.0
0.8
2.0
4.0
Thymol
Broncho T.D. syrup
Thymian syrup
Distilled thyme
Found (mg L21)
RSD (%)
ER (%)
Found (mg L21)
RSD (%)
ER (%)
Found (mg L21)
RSD (%)
ER (%)
1.358
2.1189
3.289
5.266
0.392
1.214
2.323
4.257
1.66
2.86
3.27
4.23
1.11
2.39
3.14
3.95
–
103.87
96.55
97.70
–
102.75
96.55
96.62
0.917
1.677
2.853
4.670
0.345
1.106
2.308
4.151
2.66
2.86
3.47
3.93
1.11
2.15
2.78
3.45
–
94.99
96.79
93.83
–
95.13
98.15
94.13
0.123
0.964
2.093
4.061
1.032
1.822
3.054
4.906
1.02
1.83
1.95
2.85
2.35
3.12
3.58
4.85
–
105.13
98.51
98.45
–
98.75
101.09
96.85
Table VII
Comparison of the VASEDLLME Method with Reported Methods for the Determination of Thymol and Carvacrol
Methods
a
HD-HSME -GC-FID
UAE-DLLME-GC-FID
HD-HPLC
Dissolving-HPLC
VASEDLLME-HPLC
LODs (mg L21)
r2
RSDs (%)
Linear range (mg L21)
Extraction time (min)
References
1870, 230
0.2 – 29
0.6, 1.8
1.70, 1.56
1.6
0.9944 –0.9979
0.995 –0.998
0.9992 –0.9979
0.9999
0.9998
6.37– 11.80
,11
4.5 –4.7
1.9 –2.5
1.02– 4.85
6.25 –81.25, 1.25 –87.50
0.001 – 2.1
15 –90, 2 –9
8 –200
0.005 – 4
5
10
240
–
3
(4)
(6)
(7)
(13)
This work
a
Hydrodistillation – headspace solvent microextraction.
standard deviations (RSDs, n ¼ 6) was ,5.0% that shows high repeatability of the proposed method. The intra-day precisions and
the recoveries of thymol and carvacrol determined by the standard addition method (n ¼ 5) in all samples spiked with three
different concentration levels (0.8, 2.0 and 4.0 mg L21) of
analytes were in the range of 0.3 – 4.7% and 93.8 – 105.2%,
respectively.
The applicability of the proposed method in real sample analysis for the determination of carvacrol and thymol in pharmaceutical samples including Broncho T.D. and Thymian antitussive
syrups and distilled thyme were studied by the standard addition
method. The respective chromatogram of Broncho T.D. antitussive syrup is shown in Figure 5 and average concentrations of
thymol and carvacrol in all samples are presented in Table VI.
Response Surface Methodology 1229
Comparison with the literature
Table VII indicates the LOD, coefficient of determination (r 2),
RSD, extraction time and linear range using hydrodistillation –
headspace solvent microextraction (4), ultrasonic-assisted extraction –DLLME combined with gas chromatography (6), hydrodistillation–HPLC (7), dissolving with water and HPLC (13) and
VASEDLLME combined with HPLC (this work) methods for the
determination of thymol and carvacrol in some real samples.
The proposed method provides similar quantification extraction
efficiency, with advantages of being faster, elimination of disperser organic solvents and using surfactants as emulsifier and low
limit of detection compared with other methods.
7.
8.
9.
10.
Conclusion
In this study, the analytical utility of experimental design for evaluation of optimum VASEDLLME of thymol and carvacrol and their
determination with HPLC-UV has been investigated. The proposed method is fast, simple and sensitive when compared
with other methods. The results obtained from validation experiments indicate that the proposed method can be applied for the
determination of carvacrol and thymol in pharmaceutical and
thyme samples. In this work, the experimental design procedure
used first for efficiency of the methodology by P– B screening design to study the main variables and then the CCD and RSM in
order to optimize the variables using the DF. Moreover, DF was
used to identify the optimum ER% by calculating specific variable
optimization simultaneously. Application of VASEDLLME together with the desirability optimization procedure resulted in the
successful determination of carvacrol and thymol with good sensitivity, repeatability and short extraction and separation time.
Acknowledgments
11.
12.
13.
14.
15.
16.
17.
The authors express their appreciation to the Graduate School
and Research Council of the University of Yasouj and Sepidan
Azad University for financial support of this work.
18.
References
1. Rivas, L., McDonnell, M.J., Burgess, C.M., O’Brien, M., Navarro-Villa, A.,
Fanning, S., et al.; Inhibition of verocytotoxigenic Escherichia coli in
model broth and rumen systems by carvacrol and thymol;
International Journal of Food Microbiology, (2010); 139: 70 –78.
2. Ultee, A., Smid, E.J.; Influence of carvacrol on growth and toxin production by Bacillus cereus; International Journal of Food
Microbiology, (2001); 64: 373– 378.
3. Du, W.X., Olsen, C.E., Avena Bustillos, R.J., McHugh, T.H., Levin, C.E.,
Friedman, M.; Storage stability and antibacterial activity against
Escherichia coli O157:H7 of carvacrol in edible apple films made
by two different casting methods; Journal of Agricultural and
Food Chemistry, (2008); 56: 3082–3088.
4. Kiyanpour, V., Fakhari, A.R., Alizadeh, R., Asghari, B., Jalali-Heravi, M.;
Multivariate optimization of hydrodistillation – headspace solvent
microextraction of thymol and carvacrol from Thymus transcaspicus; Talanta, (2009); 79: 695– 699.
5. Bagamboula, C.F., Uyttendaele, M., Debevere, J.; Inhibitory effect of
thyme and basil essential oils, carvacrol, thymol, estragol, linalool
and p-cymene towards Shigella sonnei and S. flexneri; Food
Microbiology, (2004); 21: 33 –42.
6. Sereshti, H., Izadmanesh, Y., Samadi, S.; Optimized ultrasonic assisted
extraction – dispersive liquid – liquid microextraction coupled with
1230 Ghaedi et al.
19.
20.
21.
22.
23.
gas chromatography for determination of essential oil of Oliveria
decumbens Vent; Journal of Chromatography A, (2011); 1218:
4593–4598.
Hajimehdipoor, H., Shekarchi, M., Khanavi, M., Adib, N., Amri, M.A.; A
validated high performance liquid chromatography method for the
analysis of thymol and carvacrol in Thymus vulgaris L. volatile oil;
Pharmacognosy Magazine, (2010); 6: 154–158.
Bazylko, A., Strzelecka, H.; Quantitative determination of phenol derivatives from Oleum thyme; Chromatographia, (2000); 52: 112–114.
Nozal, M.J., Bernal, J.L., Jimenez, J.J., Gonzalez, M.J., Higes, M.;
Extraction of thymol, eucalyptol, menthol, and camphor residues
from honey and beeswax: determination by gas chromatography
with flame ionization detection; Journal of Chromatography A,
(2002); 954: 207–215.
Kohlert, C., Abel, G., Schmid, E., Veit, M.; Determination of thymol in
human plasma by automated headspace solid-phase microextraction/gas chromatographic analysis; Journal of Chromatography B,
(2002); 767: 11– 18.
Lodesani, M., Pellacani, A., Bergomi, S., Carpana, E., Rabitti, T., Lasagni,
P.; Residue determination for some products used against Varroa infestation in bees; Apidologie, (1992); 23: 257–272.
Abu-Lafi, S., Odeh, I., Dewik, H., Qabajah, M., Hanus, L.O., Dembitsky,
V.M.; Thymol and carvacrol production from leaves of wild Palestinian
Majorana syriaca; Bioresource Technology, (2007); 99: 3914–3918.
Vinas, P., Soler-Romera, M.J., Hernandez-Cordoba, M.; Liquid chromatographic determination of phenol, thymol and carvacrol in
honey using fluorimetric detection; Talanta, (2006); 69: 1063– 1067.
Herrero-Hernandez1, E., Carabias-Martinez, R., Rodriguez-Gonzalo,
E.; Use of a bisphenol-A imprinted polymer as a selective sorbent
for the determination of phenols and phenoxy acids in honey by liquid chromatography with diode array and tandem mass spectrometric detection; Analytica Chimica Acta, (2009); 650: 195– 201.
Ramos, L.; Critical overview of selected contemporary sample preparation techniques; Journal of Chromatography A, (2012); 1221:
84– 98.
Wang, T., Qin, Y., He, H., Lv, J., Fan, Y.; An extraction technique for
analytical sample preparation in aqueous solution based on controlling dispersion of ionic surfactant assemblies in isotachophoretic migration; Journal of Chromatography A, (2011); 1218: 185–189.
Khodadoust, S., Talebianpoor, M.S., Ghaedi, M.; Application of an optimized dispersive nanomaterial ultrasound-assisted microextraction
method for preconcentration of carbofuran and propoxur and their
determination by high-performance liquid chromatography with UV
detection; Journal of Separation Science, (2014); 37: 3117–3124.
Talebianpoor, M.S., Khodadoust, S., Rozbehi, A., Akbartabar Toori, M,
Zoladl, M., Ghaedi, M., et al.; Application of optimized dispersive liquid – liquid microextraction for determination of melatonin by
HPLC – UV in plasma samples; Journal of Chromatography B,
(2014); 960: 1– 7.
Msagati, T.A.M., Nindi, M.M.; Comparative study of sample preparation
methods; supported liquid membrane and solid phase extraction in
the determination of benzimidazole anthelmintics in biological matrices by liquid chromatography – electrospray – mass spectrometry;
Talanta, (2006); 69: 243–250.
Romero-Gonzalez, R., Frenich, A.G., Vidal, J.L.M., Aguilera-Luiz, M.M.;
Determination of ochratoxin A and T-2 toxin in alcoholic beverages
by hollow fiber liquid phase microextraction and ultra high-pressure
liquid chromatography coupled to tandem mass spectrometry;
Talanta, (2010); 82: 171–176.
Huang, J.F., Lin, B., Yu, Q.W., Feng, Y.O.; Determination of fluoroquinolones in eggs using in-tube solid-phase microextraction coupled to
high-performance liquid chromatography; Analytical and
Bioanalytical Chemistry, (2006); 384: 1228–1235.
Perez, J.F.H., Campana, A.M.G.; Determination of N-methylcarbamate
pesticides in water and vegetable samples by HPLC with post-column
chemiluminescence detection using the luminol reaction; Analytica
Chimica Acta, (2008); 630: 194– 204.
Wang, S., Mu, H., Bai, Y., Zhang, Y., Liu, H.; Multiresidue determination
of fluoroquinolones, organophosphorus and N-methyl carbamates
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
simultaneously in porcine tissue using MSPD and HPLC – DAD;
Journal of Chromatography B, (2009); 877: 2961–2966.
Rezaee, M., Assadi, Y., Milani Hosseini, M.R., Aghaee, E., Ahmadi, F.,
Berijani, S.; Determination of organic compounds in water using dispersive liquid–liquid microextraction; Journal of Chromatography
A, (2006); 1116: 1 –9.
Zhou, Q.X., Bai, H.H., Xie, G.H., Xiao, J.P.; Temperature-controlled
ionic liquid dispersive liquid phase micro-extraction; Journal of
Chromatography A, (2008); 1177: 43 –49.
Khodadoust, S., Ghaedi, M., Hadjmohammadi, M.R.; Dispersive nano
solid material-ultrasound assisted microextraction as a novel method
for extraction and determination of bendiocarb and promecarb: response surface methodology; Talanta, (2013); 116: 637–646.
Bai, H., Zhou, Q., Xie, G., Xiao, J.; Temperature-controlled ionic liquid –liquid-phase microextraction for the pre-concentration of lead
from environmental samples prior to flame atomic absorption spectrometry; Talanta, (2010); 80: 1638– 1642.
Yiantzi, E., Psillakis, E., Tyrovola, K., Kalogerakis, N.; Vortex-assisted
liquid – liquid microextraction of octylphenol, nonylphenol and
bisphenol-A; Talanta, (2010); 80: 2057–2062.
Zhang, Y., Lee, H.K.; Determination of ultraviolet filters in water samples by vortex-assisted dispersive liquid–liquid microextraction followed by gas chromatography – mass spectrometry; Journal of
Chromatography A, (2012); 1249: 25 –31.
Yang, Z.H., Lu, Y.L., Liu, Y., Wu, T., Zhou, Z.Q., Liu, D.H.; Vortex-assisted
surfactant-enhanced-emulsification liquid – liquid microextraction;
Journal of Chromatography A, (2011); 1218: 7071–7077.
Kardani, F., Daneshfar, A., Sahrai, R.; Determination of b-sitosterol and
cholesterol in oils after reverse micelles with Triton X-100 coupled
with ultrasound-assisted back-extraction by a water/chloroform binary system prior to gas chromatography with flame ionization detection; Analytica Chimica Acta, (2011); 701: 232–237.
Wu, C., Liu, N., Wu, Q., Wang, C., Wang, Z.; Application of ultrasoundassisted surfactant-enhanced emulsification microextraction for the
determination of some organophosphorus pesticides in water samples; Analytica Chimica Acta, (2010); 679: 56 –62.
Saraji, M., Bidgoli, A.A.H.; Dispersive liquid – liquid microextraction
using a surfactant as disperser agent; Analytical and Bioanalytical
Chemistry, (2010); 397: 3107–3115.
Roosta, M., Ghaedi, M., Daneshfar, A.; Optimisation of ultrasoundassisted reverse micelles dispersive liquid – liquid micro-extraction
by Box – Behnken design for determination of acetoin in butter
followed by high performance liquid chromatography; Food
Chemistry, (2014); 161: 120– 126.
Massart, D.L., Vandeginste, B.G.M., Buydens, L.M.C., de Jong, S., Lewi,
P.J., Smeyers-Verbeke, J.; Handbook of chemometrics and qualimetrics: part A. Elsevier, Amsterdam, (1977).
Stalikas, C., Fiamegos, Y., Sakkas, V., Albanis, T.; Developments on chemometric approaches to optimize and evaluate microextraction;
Journal of Chromatography A, (2009); 1216: 175–189.
37. Khodadoust, S., Hadjmohammadi, M.R.; Determination of
N-methylcarbamate insecticides in water samples using dispersive
liquid– liquid microextraction and HPLC with the aid of experimental design and desirability function; Analytica Chimica Acta, (2011);
699: 113– 119.
38. Roosta, M., Ghaedi, M., Daneshfar, A., Sahraei, R., Asghari, A.;
Optimization of the ultrasonic assisted removal of methylene blue
by gold nanoparticles loaded on activated carbon using experimental
design methodology; Ultrasonics Sonochemistry, (2014); 21:
242–252.
39. Pizarro, C., Saenz-Gonzalez, C., Perez-del-Notario, N., Gonzalez-Saiz,
J.M.; Development of an ultrasound-assisted emulsification– microextraction method for the determination of the main compounds causing cork taint in wines; Journal of Chromatography A, (2012); 1229:
63 –71.
40. Momenbeik, F., Roosta, M., Nikoukar, A.A.; Simultaneous microemulsion liquid chromatographic analysis of fat-soluble vitamins in pharmaceutical formulations: optimization using genetic algorithm;
Journal of Chromatography A, (2010); 1217: 3770–3773.
41. Khodadoust, S., Ghaedi, M.; Optimization of dispersive liquid–liquid
microextraction with central composite design for preconcentration
of chlordiazepoxide drug and its determination by HPLC-UV;
Journal of Separation Science, (2013); 36: 1734– 1742.
42. Dejaegher, B., Dumarey, M., Capron, X., Bloomfield, M.S., Vander
Heyden, Y.; Comparison of Plackett –Burman and supersaturated designs in robustness testing; Analytica Chimica Acta, (2007); 595:
59 –71.
43. Box, G.E.P., Hunter, J.S., Hunter, W.G.; Statistics for experimenters,
2nd ed., Wiley –Interscience, New York, (2005).
44. Bruns, R.E., Scarminio, I.S., Neto, B.B.; Statistical design–chemometrics, Elsevier, Amsterdam, (2006).
45. Box, G.E.P., Wilson, K.B.; On the experimental attainment of optimum
conditions; Journal of the Royal Statistical Society: Series B
(Statistical Methodology), (1951); 13: 1– 45.
46. Myers, R.H., Montgomery, D.C.; Response surface methodology: process and product optimization using designed experiments, Wiley,
New York, (2002).
47. Morgan, E.; Chemometrics: experimental design, Wiley, London,
(1991).
48. Roosta, M., Ghaedi, M., Shokri, N., Daneshfar, A., Sahraei, R., Asghari,
A.; Optimization of the combined ultrasonic assisted/adsorption
method for the removal of malachite green by gold nanoparticles
loaded on activated carbon: experimental design; Spectrochimica
Acta Part A: Molecular and Biomolecular Spectroscopy, (2014);
118: 55 –65.
49. Harrington, E.C.; The desirability function; Industrial Quality
Control, (1965); 21: 494–498.
50. Derringer, G., Suich, R.; Simultaneous optimization of several response variables; Journal of Quality Technology, (1980); 12:
214–219.
Response Surface Methodology 1231