Studii şi Cercetări Ştiinţifice 2016, 17 (4), pp.

335 – 347
Chimie şi Inginerie Chimică, Biotehnologii, Industrie Alimentară

Scientific Study & Research

Chemistry & Chemical Engineering, Biotechnology, Food Industry ISSN 1582-540X

ORIGINAL RESEARCH PAPER

OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION

OPERATING CONDITIONS USING THE RESPONSE
SURFACE DESIGN METHODOLOGY

Haitham Osman1*, Ihab Shigidi1,2 , Abubakr Elkhaleefa1

1
King Khalid University, Department of Chemical Engineering, P.O. Box
9146, Abha 61413, Saudi Arabia
2
Al-Neelain University, Department of Chemical Engineering, P.O. Box
10179, Khartoum, Sudan

*Corresponding author: [email protected]

Received: June, 27, 2016

Accepted: December, 06, 2016

Abstract: This paper applies Response Surface Design (RSD) to

model the experimental data obtained from the extraction of sesame seeds
oil using n-hexane, chloroform and acetone as solvents under different
operating conditions. The results obtained revealed that n-hexane
outperformed the extraction obtained using chloroform and acetone.
The developed model predicted that n-hexane with a rotational speed of 547
rpm and a contact time between the solvent and seeds of 19.46 hours with
solvent: seeds ratio of 4.93, yields the optimum oil extracted of 37.03 %,
outperforming chloroform and acetone models that gave prediction for 4.75
and 4.21 respectively.
While the maximum predictions yield for chloroform is 6.73 %, under the
operating conditions of 602 rpm, and 24 hours contact time, with a ratio of
solvent: seeds of 1.74. On the other hand the acetone maximum prediction is
only 4.37 %, with operational conditions of 467 rpm, and 6.00 hours contact
time, with a ratio of solvent: seeds of 1.
It is has been found that the maximum oil extraction yield obtained from the
chloroform (6.73 %) and Acetone (4.37 %) is much lower than that
predicted by n-hexane 37.03 %.

Keywords: ANOVA statistical test, operating conditions, process

optimization, sesame seeds oil extraction, response
surface design
© 2016 ALMA MATER Publishing House, “VASILE ALECSANDRI” University of Bacău. All rights reserved. 335
 OSMAN, SHIGIDI and ELKHALEEFA

INTRODUCTION

The use of solvents in oil extraction has been a subject that gained the attention of
researchers’ worldwide for a very long time. The need for a clean, high-efficiency
process that provides higher extraction yield was the motivation for most of the work
carried in this area.
Sesame seeds have been used as a healthy food and for disease prevention in many
countries for centuries. The literature reports indicate many health benefits associated
with the consumption of sesame seeds; for example, they significantly increase plasma
γ-tocopherol and enhance vitamin E activity, which are believed to prevent human
aging-related diseases such as cancer and heart disease. Studies have also shown that
including sesame in the diet can improve antioxidant capacity [1].
Sesame seeds have higher oil content (around 42-50 %) than most of the known oil
seeds, see Table 1.

Table 1 .Oil content of selected seeds/fruits

Product name Oil content in [%] Water content in [%]
1 Mustard seeds 15.0 - 35.0 5.0 - 9.0
2 Sesame seed 42.0 - 50.0 5.0 - 13.0
3 Shea nuts 46.0 - 48.0 5.0 - 8.0
4 Soybeans 13.0 - 24.0 8.0 - 10.0
5 Sunflower seeds 19.0 - 56.0 7.0
6 Olives, black 15.0 - 31.0 52.0
7 Olives, kernels 12.0 - 15.0 8.0

Sesame oil is generally regarded as high-priced and high-quality oil. It is one of the
most stable edible oil despite its high degree of unsaturation [2]. Sesame oil is rich in
monounsaturated and polyunsaturated fatty acids [1]. The most abundant fatty acids in
sesame oil are oleic, linoleic, palmitic, and stearic acids, which together comprised
about 96 % of the total fatty acids.
Extraction of sesame oil has developed significantly over the years; mechanical
methods are an early means of separation by physical pressure to ‘squeeze the oil out’.
The most energy efficient, practical embodiment of that method is the modern screw
press. This is a conveyor screw with a slotted cage surrounding it and a screw with
diminishing space for the solid material as the material proceeds from pitch to pitch of
the screw. Eventually, as the free space is progressively restricted, the oil is squeezed
out of the solids and through the slots. More than half of the oil is easily removed in this
way, but at least 7 - 18 % residual oil is left in the solids, the process uses considerable
horsepower, there is considerable wear and maintenance, and it takes many machines
for high capacity.
In a nutshell such techniques are no longer prevailing currently due to the higher cake
oil contents. Solvent extraction, providing higher yields 98 - 99 %, is nowadays the
dominant technique applied in most extraction processes. Solvent extraction became the
commonly used commercial technique to recover oil from oilseeds, presently n-hexane
is the preferred solvent throughout the world due to its extraction efficiency and ease of
availability [3]. In comparison, solvent extraction with hexane will leave about 0.5 % of
residual oil. In addition, it uses less horsepower, and requires less maintenance; hence it
is more efficient and reliable.
336 St. Cerc. St. CICBIA 2016 17 (4)
 OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION OPERATING CONDITIONS
USING RESPONSE SURFACE DESIGNING

Over the years, researchers investigated different solvents for sesame seeds oil
extraction, Durma et al.[4] investigated the impact of roasting seeds on the oil yield and
has concluded that antioxidant capacity of the roasted seeds and oxidative stability of
the extracted oil could be greater than that of the unroasted counterpart. Elkhaleefa et al.
[5] have studied the operating condition of sesame seeds oil extraction in more
comprehensive details as they examined the effect of sesame seeds particle sizes, the
ratio of solvent to seeds mass, contact time, stirring effect, roasting impact and
extraction temperatures.
The application of experimental design using response surface methodology (RSM) has
been widely used for optimization of experimental parameters. RSM is a collection of
mathematical and statistical techniques widely applied in the food industry to evaluate
the relationship between the predicted values of independent and dependent variables.
RSM is not only time and reagent saving, but it also provides information about the
interactions of parameters [6, 7]. Experimental design for process optimization via
response surface methodology has been studied by many researchers [8 – 12].
The objective of this study was to analyze the effects of: seeds/solvent ratio (1 to 5),
rotation speed (0, 150, 300 and 700 rpm) and contact time (6, 12, 24 hours) on the oil
extraction yields. An effective technique to study and evaluate the effect of some of the
above variables on the extraction yield and extract composition, which allows
identification of the main effects with limited experimental works, is response surface
analysis of results of carefully planned experiments, or so-called response surface
design (RSD) of experimental data. Data obtained by Elkhaleefa et al. [5] are used as
operational conditions parameters.

MATERIALS AND METHODS

The experimental design is based on a face centered (FCD), three-variable (operation

conditions in this case), three-level 33 fractional factorial design with star points and
triplicates at the center [13]. Experiments were carried out in a randomized order to
minimize the effect of unexpected variability in the observed response due to extraneous
factors. A second-order (quadratic) polynomial equation was used to express each
response variable Y as a function of the independent variables X in the general form:
(1)
where: X0 (dummy variable) equals 1; A01, A02, and A03 are linear coefficients; A12, A13,
and A23 are cross-product coefficients; A11, A22 and A33 are quadratic coefficients, and
the super index y defines the dependent variables (e. g. concentration of A, B, and C).
The best fitting of the model was evaluated by determining the coefficient of
determination R2 and by the analysis of the variance (ANOVA). The coefficients of the
response surface equation were estimated by using Design-Expert Software. The
statistical significance was based on the total error criteria with a confidence level of
95 %.

St. Cerc. St. CICBIA 2016 17 (4) 337

 OSMAN, SHIGIDI and ELKHALEEFA

The mathematical model

Experimental data from literature [5] were modelled using the Design-Expert Software.
Response Surface Design (RSD) was used to evaluate and find the optimum model, the
three variables; seeds/solvent ratio, contact time (hour) and stirring speed (rpm) were
coded to ensure dimensionless of the variables by applying the general equation:
variables value mean variable
The dimensionless variables  between variables int ervals 
range
Thus obtain: Seeds/solvent ratio X1 = (A-3)
Contact time X2 = (B-12)/6
Stirring speed X3 = (C-150)/150
as. The design model can be written in a function of each independent variable (X1, X2
and X3) as follows:
(2)
where: β0, βi, βii, and βij are the regression coefficients for intercept, and the notations X1
= A, X2 = B and X3 = C are the dimensionless independent variables as presented in
Table 2.

Table 2. Experimental design variables used for optimizing sesame seeds oil yield.
Coded levels
Independent variables Xn Symbol
-1 0 1
Seeds to solvent ratio in [%] (r) A 1 3 5
Contact time in [h] (t) B 6 12 24
Stirring speed in [rpm] (S) C 0 350 750

RESULTS AND DISCUSSION

In this paper the Response Surface Design (RSD) is used as an effective technique to
study, model and compare between the effects of various variables. These variables are
the operating conditions for oil extraction using three different solvents; n-hexane,
chloroform and acetone, the variables are: seeds/solvent ratio (1 to 5), rotational speed
(0, 150, 300 and 700 rpm) and contact time (6, 12, 24 hours) and their effect on the
overall oil extraction yield.

n-Hexane solvent extraction model

The operating conditions of sesame seeds oil extraction by hexane solvent have been
studied by Elkhaleefa et al. [5] in a more comprehensive details as they examined the
effect of sesame seeds particle sizes, the ratio of solvent to seeds mass, contact time,
stirring effect, roasting impact and extraction temperatures. The experimental data
consist of 60 runs. This paper models the experimental data using the above regression
for the three variables.

338 St. Cerc. St. CICBIA 2016 17 (4)

 OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION OPERATING CONDITIONS
USING RESPONSE SURFACE DESIGNING

Pre-model results for n-hexane

When applying the developed model using the previously defined data, the following
Pre-model has been obtained:

(3)
The model has been verified using ANOVA statistical test where Table 3 was obtained.

Table 3. ANOVA Test for response surface quadratic model analysis of variance
F-Value P-Value
Model 1263.63 < 0.0001
A - seeds/solvent 9975.61 < 0.0001
B - contact time 136.80 < 0.0001
C - stirring speed 39.83 < 0.0001
AB 27.63 < 0.0001
AC 3.13 0.0830
BC 0.075 0.7857
A2 178.63 < 0.0001
B2 0.72 0.3997
C2 21.48 < 0.0001

From Table 3 it has been found that the calculated F-value of 1263.63 implies that the
model is significant (as the largest value is desirabled). However, the P-value should be
less than 0.0500 to indicate that model terms are significant. In this case A, B, C, AB,
A2, C2 are considered significant model terms. Values greater than 0.1000 indicate that
the model terms are not significant, the model reduction will improve this model.

Table 4. R2 Coefficient values

Std. Dev. 0.23 R2 0.9956
Mean 32.87 Adj R2 0.9948
C.V. % 0.68 Pred R2 0.9938
PRESS 3.59 Adeq Precision 109.090

The "Pred R2" of 0.9938 is in reasonable agreement with the "Adj R2" of 0.9948
as presented in Table 4. "Adeq Precision" measures the signal to noise ratio. A ratio
greater than 4 is desirable. The model ratio of 109.090 indicates an adequate signal and
this model can be used to navigate the design space. The determination coefficient (R2)
and the adjusted determination coefficient (Adj. R2) were 0.9956 and 0.9948,
respectively, which implied that the model could explain 99.56 % variability of the
response variable.
The experimental and predicted values were compared in order to determine the validity
of the model, the values of "Prob > F" less than 0.0500 indicate model terms are
significant. In this case A, B, C, AB, A2, C2 are significant model terms. This model can
be visualised in 3D graphical shape as in Figure 1A and 1B.

St. Cerc. St. CICBIA 2016 17 (4) 339

 OSMAN, SHIGIDI and ELKHALEEFA

Design-Expert® Software

oil yield
37

27
37.1
X1 = A: SEEDS/SOLVENT
X2 = B: CONTACT TIME
34.75
Actual Factor
C: STIRRING SPEED = 350.00

oil yield
32.4

30.05

27.7

24.00 5.00
19.50 4.00
15.00 3.00

B: CONTACT TIME10.50 2.00

A: SEEDS/SOLVENT
6.00 1.00

Figure 1A. 3D Prediction of oil extraction yield using n-hexane at different

seeds/solvent ratio vs. contact time at constant stirring speed of 350 rpm

Design-Expert® Software oil yield

24.00
oil yield
37

27
19.50
X1 = A: SEEDS/SOLVENT
X2 = B: CONTACT TIME

Actual Factor
C: STIRRING SPEED = 350.00
15.00
29.2859 30.8475 32.4091 33.9707 35.5322

10.50

6.00
1.00 2.00 3.00 4.00 5.00

X1: A: SEED S/SOLVENT

X2: B: C ON TAC T TIME

Figure 1B. Contour prediction of oil extraction yield using n-hexane at different
seeds/solvent ratio vs. contact time at constant stirring speed of 350 rpm

The experimental data and predicted values were compared in order to determine the
validity of the model and have showed good agreement as presented in Figure 2.

340 St. Cerc. St. CICBIA 2016 17 (4)

 OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION OPERATING CONDITIONS
USING RESPONSE SURFACE DESIGNING

Design-Expert® Software Predicted vs. Actual

oil yield
38.00
Color points by value of
oil yield:
37

27 35.25

32.50

29.75

27.00

27.00 29.52 32.03 34.55 37.07

X: Actual
Y : Predicted

Figure 2. Oil extraction yield prediction vs. oil extraction yield experimental data
using n-hexane

Model reduction and modification for n-hexane

Table 5 below summarizes the statistical parameters for the three Response Surface
Design (RSD), it shows that the parameters are estimated adequately because of
(p-value < 0.05). High coefficients of determination (R2 > 0.90) for all Response
Surface Desing models, and high signal-to-noise (> 4) for all responses, indicate that the
models can be used to navigate the design space. This statistical indicators are
complemented in Figure 2 by scatter plots demonstrating the correlation between
predicted and experimental responses.

Table 5. ANOVA for response surface reduced quadratic model analysis

of variance table [partial sum of squares - type iii]
F-Value P-Value
Total Model Variables 1862.24 < 0.0001
A - seeds/solvent 10442.82 < 0.0001
B - contact time 144.81 < 0.0001
C - stirring speed 40.28 < 0.0001
AB 27.15 < 0.0001
A2 175.57 < 0.0001
C2 21.11 < 0.0001

Model verification

The experimental and predicted values were compared in order to determine the validity
of the model in the (Figure 2). The researchers [14] found that solvent concentration
plays a critical role in the extraction of soluble solids; Also, Kwon et al. [15] reported
that solvent concentration was the most important factor contributing to the extraction
components using RSM. The plot of predicted versus experimental values for the
concentration shows more correlations, due to the quadratic equation.
St. Cerc. St. CICBIA 2016 17 (4) 341
 OSMAN, SHIGIDI and ELKHALEEFA

The F-value model of 1862.24 implies that the model is significant. There is only a
0.01% chance that this large "F-Value" could occur due to noise. Moreover the
P-values of less than 0.0500 indicate that the model terms are significant. In this case A,
B, C, AB, A2 and C2 are the significant model terms. Values greater than 0.1000
indicate that the model terms are not significant, concluding that this is the best model
terms that can be achieved using ANOVA test since all terms are significant.
Table 6, presents the "Pred R-Squared" of 0.9941 which is in reasonable agreement with
the "Adj R2" of 0.9947. On the other hand, "Adeq Precision" measures the signal to
noise ratio where ratio value greater than 4 is desirable. This model ratio of 128.012
indicated an adequate signal, which it higher than the previous pre-model.

Table 6. Modified model R2 coefficient values

Std. Dev. 0.23 R2 0.9953
Mean 32.87 Adj R2 0.9947
C.V. % 0.69 Pred R2 0.9941
PRESS 3.43 Adeq Precision 128.012

From what has been found, the final equation model for the n-hexan in terms of coded
factorsis presented in the following format:

(4)
The final equation in terms of actual factors is:

(5)
Optimization of the final model resulted in (Figures 3A and 3B), where it can be
revealed that the optimum value for the maximum oil yield is 37.0458, the seeds to
solvent ratio of 4.93 and with contact time of 19.46 h and strring speed of 547 rpm as
summarised in Table 7.

Table 7. n-Hexane optimised extraction prarmeters

Solvent Contact Stirring speed Seeds:
Oil yield [%]
time t [h] S [rpm] Solvant ratio r
n-Hexane 19.46 547 4.93 37.04

342 St. Cerc. St. CICBIA 2016 17 (4)

 OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION OPERATING CONDITIONS
USING RESPONSE SURFACE DESIGNING

Design-Expert® Software oil yield

24.00
oil yield
37
Prediction 37.0344
27
19.50
X1 = A: SEEDS/SOLVENT
X2 = B: CONTACT TIME

Actual Factor
C: STIRRING SPEED = 374.78 35.5753
15.00
29.3103 30.8766 32.4428 34.0091

10.50

6.00
1.00 2.00 3.00 4.00 5.00

X1: A: SEEDS/SOLVENT
X2: B: CONTACT TIME

Figure 3A. Contour optimum predicted extracted oil value with respect to seeds:
n-hexane solvant ratio vs. contact time

Design-Expert® Software

oil yield
37

27
37.2
X1 = A: SEEDS/SOLVENT
X2 = B: CONTACT TIME
34.825
Actual Factor
C: STIRRING SPEED = 374.78
oil yield

32.45

30.075

27.7

24.00 5.00
19.50 4.00
15.00 3.00

B: CONTA CT TIME10.50 2.00

A : SEEDS/SOLV ENT
6.00 1.00

Figure 3B. 3D optimum predicted extracted oil value with respect to seeds:
n-hexane solvant ratio vs. contact time

Chloroform extraction model

Similar to the work done on the n-hexane the results for the final chloroform extraction
model is presented as:

(6)
The above equation predict the oil yield in term of (as function of) all independent
variables the response surface as can be seen from the 3D Figure 3B below is semi-flat.

St. Cerc. St. CICBIA 2016 17 (4) 343

 OSMAN, SHIGIDI and ELKHALEEFA

Optimization of chloroform model

The maximum value for the oil yield is 6.72259 at the ratio of 1.59, contact time of
23.97 and stirring speed of 310.12 as summarised in Table 8 and (Figure 4A and 4B).

Table 8. Chloroform optimised extraction prarmeters

Solvent Contact Stirring Seeds:
Oil yeild [%]
time t [h] speed S [rpm] Solvant ratio r
Chloroform 23.97 310.12 1.59 6.722

Design-Expert® Software Predictio 6.72259 oil yield

24.00
oil yield
6.70869
6.7

6 6.64279

19.50
X1 = A: SOLVENT /SEEDS
6.5769
X2 = B: CONT ACT T IME
B: CONTACT TIME

Actual Factor
C: ST IRRING SPEED = 310.12 6.51101
15.00

10.50

6.44512

6.00
1.00 2.00 3.00 4.00 5.00

A: SOLVENT/SEEDS

Figure 4A. Contour optimum oil extracted yield with respect to constant contact
time and chloroform solvent to seeds ratio

Design-Expert® Software

oil yield
6.7

6
6.79
X1 = A: SOLVENT/SEEDS
X2 = B: CONTACT TIME
6.6875
Actual Factor
C: STIRRING SPEED = 637.32
6.585
oil yield

6.4825

6.38

24.00 5.00
19.50 4.00
15.00 3.00

B: CONTACT TIME10.50 2.00

A: SOLVENT/SEEDS
6.00 1.00

Figure 4B. 3D optimum oil extracted yield with respect to constant contact time and
chloroform solvent to seeds ratio at different stirring speeds

Acetone extraction model

Similarly when applying the same technique on acetone extraction experimental data,
the following final equation in terms of coded factors is obtained.
344 St. Cerc. St. CICBIA 2016 17 (4)
 OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION OPERATING CONDITIONS
USING RESPONSE SURFACE DESIGNING

(7)
Final equation in terms of actual factors is:

(8)
The 3D model can be visualised at this point as in (Figure 5).
Design-Expert® Software

oil yield
4.4

4
4.32
X1 = A: SEEDS/SOLVENT
X2 = B: CONTACT TIME
4.28
Actual Factor
C: STIRRING SPEED = 350.00
4.24
oil yield

4.2

4.16

24.00 5
19.50 4
15.00 3

B: CONTACT TIME10.50 2
A: SEEDS/SOLVENT
6.00 1

Figure 5. 3D optimum oil extracted yield with respect to constant contact time and
acetone solvent to seeds ratio at different stirring speeds

The best model achived a maximum value for the oil yield of 4.31772 the ratio 1.00 the
contact time is 6 h the strring speed is 468.64, Table 9 and Figure 6.

Table 9. Acetone optimised extraction prarmeters

Solvent Contact Stirring Seeds:
Oil yeild [%]
time t [h] speed S [rpm] solvant ratio r
Acetone 6 468.64 1.00 4.31772
Design-Expert® Software oil yield
24.00
oil yield 4.22231
4.4 4.24616

4
19.50
X1 = A: SEEDS/SOLVENT
X2 = B: CONTACT TIME

Actual Factor
C: STIRRING SPEED = 467.79
15.00

4.19846
10.50 4.22231

4.27001
4.29387
Prediction 4.31772
4.24616
4.27001
6.00
1.00 2.00 3.00 4.00 5.00

X1: A: SEED S/SOLVEN T

X2: B: C ON TAC T TIME

Figure 6. Contour optimum oil extracted yield with respect to constant contact time
and acetone solvent to seeds ratio at different stirring speeds
St. Cerc. St. CICBIA 2016 17 (4) 345
 OSMAN, SHIGIDI and ELKHALEEFA

CONCLUSIONS

This study showed that RSM is a powerful modelling technique capable of describing
and predicting solvent extraction operational process parameters. Contours and surface
plots in RSM estimated the effects of three independent variables- contact time, stirring
speed and seeds to solvent ratio- efficiently.
Table 10 summarizes the optimised predictions for the model using the solvents:
n-hexane, chloroform and acetone.

Table 10. Prediction of the extraction for the solvents

Solvent Contact time Stirring speed Seeds:
Oil yield [%]
[h] [rpm] Solvent ratio r
n-Hexane 19.46 547 4.93 37.0063
Chloroform 23.94 602 1.74 6.73632
Acetone 6.00 467 1.00 4.31772

In a nutshell, hexane has shown significant extraction yields in comparison with other
solvents. The developed model was able of predicting extraction yields that are in good
agreement with experimental data.

SYMBOLS
Aij linear coefficient
Xij independent variables
Y oil yield [%]
β independent coefficient
A seeds/solvent ratio [%]
X1 seeds/solvent, dimensionless
t contact time, [h]
X2 contact time, dimensionless
S stirring speed, [rpm]
X3 stirring speed, dimensionless
R2 determination coefficient

REFERENCES

1. Elleuch, M., Bedigian, D., Zitoun, A., Zouari, N.: Sesame (Sesamum indicum L.) seeds in food,
nutrition and health, Nuts & seeds in health and disease prevention, 2011, 1029-1036;
2. Hwang, L.S.: Sesame oil, Bailey's Industrial Oil and Fat Products, 6th edition, 6 Volume set,
2005;
3. Saxena, D.K., Sharma. S., Sambi, S.: Comparative extraction of cottonseed oil by n-Hexane and
ethanol, ARPN Journal of Engineering and Applied Sciences, 2011, 6, 1, 84-89;
4. Durmaz, G., Gökmen, V.: Impacts of roasting oily seeds and nuts on their extracted oils, Lipid
Technology, 2010, 22,179-182;
5. Elkhaleefa, A., Shigidi, I.: Optimization of Sesame Oil Extraction Process Conditions, Advances
in Chemical Engineering and Science, 2015, 5, 305;
6. Duarte, R.M., Duarte, A.C.: Optimizing size-exclusion chromatographic conditions using a
composite objective function and chemometric tools: Application to natural organic matter
profiling, Analytica chimica acta, 2011, 688, 90-98;

346 St. Cerc. St. CICBIA 2016 17 (4)

 OPTIMIZATION OF SESAME SEEDS OIL EXTRACTION OPERATING CONDITIONS
USING RESPONSE SURFACE DESIGNING

7. Leardi, R.: Experimental design in chemistry: a tutorial, Analytica chimica acta, 2009, 652,
161-172;
8. Bingol, D., Kulcu, M.: Optimization of the solid phase extraction method for determination of
Cu(II) in natural waters by using response surface methodology, Analyst, 2011, 136, 4036-4044;
9. Liu, R.L., Zhang, J., Mou, Z.L., Hao, S.L., Zhang, Z.Q.: Microwave-assisted one-step extraction-
derivatization for rapid analysis of fatty acids profile in herbal medicine by gas chromatography-
mass spectrometry, Analyst, 2012, 137, 5135-5143;
10. Esfahani, A.R., Firouzi, A.F., Sayyad, G., Kiasat, A., Alidokht, L., Khataee, A.: Pb (II) removal
from aqueous solution by polyacrylic acid stabilized zero-valent iron nanoparticles: process
optimization using response surface methodology, Research on Chemical Intermediates, 2014,
40, 431-445;
11. Djafarzadeh, N., Zarei, M., Behjati, B., Khataee, A.: Optimization of the oxalate catalyzed
photoelectro-Fenton process under visible light for removal of reactive red 195 using a carbon
paper cathode, Research on Chemical Intermediates, 2013, 39, 3355-3369;
12. Akalin, M.K., Karagöz, S.: Optimization of ethanol supercritical fluid extraction of medicinal
compounds from St. John's Wort by central composite design, Analytical Letters, 2014, 47,
1900-1911;
13. Box, G.E., Behnken, D.W.: Some new three level designs for the study of quantitative variables,
Technometrics, 1960, 2, 455-475;
14. Park, N.Y., Lee, G.D., Jeong, Y.J., Kwon, J.H.: Optimization of extraction conditions for
physicochemical properties of ethanol extracts from Chrysanthemum boreale, Journal-Korean
Society of Food Science and Nutrition, 1998, 27, 585-590;
15. Kwon, J.H., Bélanger, J.M., Paré, J.J.: Optimization of microwave-assisted extraction (MAP) for
ginseng components by response surface methodology, Journal of Agricultural and Food
Chemistry, 2003, 51, 1807-1810.

St. Cerc. St. CICBIA 2016 17 (4) 347