See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/289659988

Box-Behnken design for investigation of

microwave-assisted extraction of patchouli oil

Conference Paper · September 2015

DOI: 10.1063/1.4938350

CITATIONS READS

0 75

2 authors:

Heri Kusuma Mahfud Mahfud

Institut Teknologi Sepuluh Nopember Institut Teknologi Sepuluh Nopember
55 PUBLICATIONS 51 CITATIONS 61 PUBLICATIONS 91 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Chemical Compounds View project

Extraction assisted by microwave of cananga flowers View project

All content following this page was uploaded by Heri Kusuma on 09 January 2016.

The user has requested enhancement of the downloaded file.

 Box-Behnken design for investigation of microwave-assisted extraction of patchouli oil
Heri Septya Kusuma and Mahfud Mahfud

Citation: AIP Conference Proceedings 1699, 050014 (2015); doi: 10.1063/1.4938350

View online: http://dx.doi.org/10.1063/1.4938350
View Table of Contents: http://scitation.aip.org/content/aip/proceeding/aipcp/1699?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in

Response surface methodology (RSM) modeling of microwave-assisted extraction of natural dye from Swietenia
mahagony: A comparation between Box-Behnken and central composite design method
AIP Conf. Proc. 1699, 050009 (2015); 10.1063/1.4938345

Central composite rotatable design for investigation of microwave-assisted extraction of ginger (Zingiber
officinale)
AIP Conf. Proc. 1699, 050008 (2015); 10.1063/1.4938344

Microwave-assisted synthesis and characterization of nickel ferrite nanoparticles

AIP Conf. Proc. 1675, 020029 (2015); 10.1063/1.4929187

Doping optimization of polypyrrole with toluenesulfonic acid using Box-Behnken design

AIP Conf. Proc. 1571, 782 (2013); 10.1063/1.4858750

Observation of microwave-assisted magnetization reversal in perpendicular recording media

Appl. Phys. Lett. 103, 042413 (2013); 10.1063/1.4816798

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 Box-Behnken Design for Investigation of Microwave-
Assisted Extraction of Patchouli Oil
Heri Septya Kusuma1, a), Mahfud Mahfud2, b)
1,2
Department of Chemical Engineering, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia
a)
Corresponding Author: heriseptyakusuma@gmail.com
b)
mahfud@chem-eng.its.ac.id

ABSTRACT. Microwave-assisted extraction (MAE) technique was employed to extract the essential oil from patchouli
(Pogostemon cablin). The optimal conditions for microwave-assisted extraction of patchouli oil were determined by
response surface methodology. A Box-Behnken design (BBD) was applied to evaluate the effects of three independent
variables (microwave power (A: 400–800 W), plant material to solvent ratio (B: 0.10–0.20 g mL -1) and extraction time (C:
20–60 min)) on the extraction yield of patchouli oil. The correlation analysis of the mathematical-regression model indicated
that quadratic polynomial model could be employed to optimize the microwave extraction of patchouli oil. The optimal
extraction conditions of patchouli oil was microwave power 634.024 W, plant material to solvent ratio 0.147648 g ml-1 and
extraction time 51.6174 min. The maximum patchouli oil yield was 2.80516% under these optimal conditions. Under the
extraction condition, the experimental values agreed with the predicted results by analysis of variance. It indicated high
fitness of the model used and the success of response surface methodology for optimizing and reflect the expected extraction
condition.
Keywords: Microwave-assisted extraction; Patchouli oil; Extraction; Box-Behnken design; Pogostemon cablin

INTRODUCTION
Patchouli oil is obtained from the leaves of Pogostemon cablin (patchouli), a plant of the Lamiaceae family,
originating from Malaysia and India [1]. It is an important essential oil (EO) in the perfume industry, used to
give a base and lasting character to a fragrance [2,3,4]. The EO is very appreciated for its characteristic pleasant
and long lasting woody, earthy, and camphoraceous odor, as well as for its fixative properties, being suitable for
use in soaps and cosmetic products [5,6]. It is also on the FDA’s (Food and Drug Administration) list of
substances approved for human consumption, in section 172.510, as a natural additive for food flavoring [7].
Moreover, the plant (Pogostemon cablin) is widely used in traditional Chinese medicine as it offers various
types of pharmacological activity according to the composition of the oil [1,8].
Since there is no synthetic replacement for patchouli oil, the demand is very high in the international
markets. The world demand of patchouli oil is currently stood at 587 tons per annum. The demand of patchouli
has never stopped increasing. It is anticipated that the price of patchouli will remain high in the future because
of high demand of patchouli worldwide and introduction of new application of patchouli oil. In view of the
importance of patchouli oil to many sectors of industries, it would be benefitial to improve the production of
patchouli oil. As a result of this, prospects for discovering method of inducing patchouli production are very
promising. It is expected that through understanding of the mechanism of patchouli extraction, there will be
sufficient supply to meet market demands. Therefore, there is strong incentive to optimize the patchouli oil
extraction yield.
In recent years, the use of microwave-assisted extraction (MAE) of constituents from plant material has
shown tremendous research interest and potential [9]. The principle of heating during MAE is based on the
direct effect of the microwaves on molecules by ionic conduction and dipole rotation within the processed
materials. Ionic conduction caused by the electrophoretic migration and the dipole rotation of realignment of
ions under an applied electromagnetic field results in the rise of temperature within the extraction solution. This
heating mode obviously enhances the transfer progress of objective solute within material matrix toward
solution [10].

International Conference of Chemical and Material Engineering(ICCME) 2015

AIP Conf. Proc. 1699, 050014-1–050014-11; doi: 10.1063/1.4938350
© 2015 AIP Publishing LLC 978-0-7354-1346-7/$30.00

This article is copyrighted as indicated in the article. Reuse of AIP content050014-1

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 Therefore, in this study used methods microwave-assisted extraction (MAE) to obtain patchouli oil. In
addition, the use of methods microwave-assisted extraction (MAE) is also based on the availability of sufficient
microwave readily available in the public.
Response surface methodology (RSM) was a collection of statistical and mathematical techniques that has
been successfully used to determine the effects of several variables and optimize processes. The main advantage
of RSM was to reduce number of experimental trials needed to evaluate multiple variables and their interactions
[11]. Response surface methodology (RSM) was applied to optimize the conditions for oil extraction from hemp
seed (Cannabis sativa L.) by supercritical carbon dioxide extraction with three independent variables namely
operating temperature, pressure and particle diameter [12] and Phaleria macrocarpa seed by solvent extraction
technique using n-hexane with three independent variables namely extraction temperature, time and solvent-to-
feed ratio [13].
However, there were no reports available in the literature regarding the optimization of microwave-assisted
extraction of essential oils from the patchouli by RSM. In this study, the microwave-assisted extraction
parameters of essential oils from the patchouli (microwave power, plant material to solvent ratio and extraction
time) was firstly investigated and optimized using a three factors three level Box-Behnken response surface
experimental design (BBD).
The objectives of this study were to determine the effect of MAE parameters including microwave power,
plant material to solvent ratio and extraction time on the yield of patchouli oils; and to optimize processing
conditions of MAE for the highest yield of patchouli oils.

MATERIALS AND METHODS

Material and chemicals

Dried leaves of Pogostemon cablin was collected from Tulungagung, East Java, Indonesia. The leaves then
were chop to a size around 5-10 mm using a commercial grade blender (Arte Blender, BL-001) and stored at
room temperature until required. Aquadest and anhydrous sodium sulphate used in the experimental work were
all of analytical grade.

FIGURE 1. Schematic representation of the microwave-assisted extraction apparatus used in this study.

Microwave extraction of patchouli oil

A domestic microwave oven (EMM-2007X, Electrolux, 20 L, 800 W; variable in 200 W increments, 2.45
GHz) was modified for MAE operation shown in Fig. 1. The dimensions of the PTFE-coated cavity of the
microwave oven were 46.1 cm x 28.0 cm x 37.3 cm. Three plant material to solvent ratio of Pogostemon cablin
leaves samples (0.10, 0.15 and 0.20 g ml-1) were placed in a 1 l flask containing deionized water (400 ml). The
flask was setup within the microwave oven cavity and a liebig condenser was used (outside the oven) to collect
the extracted essential oils. The microwave oven was operated at three power level (400 W, 600 W and 800 W)
for a period of 1 h. To remove water, the extracted essential oils were then dried over anhydrous sodium sulfate,
weighed and stored in amber vials at 4 oC until they were used for analysis. The yield of patchouli oil was
calculated as follows

This article is copyrighted as indicated in the article. Reuse of AIP content050014-2

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 x
y  100 (1)
z
where X is the weight of extracted patchouli oil (g) and Z is the weight of Pogostemon cablin leaves (g).

BBD response surface design

In this present study, Design-Expert version 9.0.4.1 (State-Ease Inc., Minneapolis, MN, USA) statistical
package was employed for modeling and optimization of MAE with independent variables namely microwave
power (A), plant material to solvent ratio (B) and extraction time (C). Three factors three level Box-Behnken
response surface experimental design (BBD) was applied to investigate the individual and interactive effects of
process variables on the yield of essential oil (%). For statistical calculations, the process variables were coded
at three levels (-1, 0 and +1) and the coding was done by the following equation [14]

xi  xz
xi  , i = 1,2,3...k (2)
xi

where xi, is the dimensionless value of an independent variable; Xi, the real value of an independent variable; Xz,
the real value of an independent variable at the center point; and ∆Xi, step change of the real value of the
variable i. Seventeen experiments were designed with five replications by BBD using the following equation
(Table 1), [15,16]

N  2 K ( K  1)  Co (3)

where K is number of factors and C0 is the number of central point. The correlation between the response and
independent variables was evaluated by developing the second order polynomial mathematical model and the
generalized form of equation is given below [17]

Y   o    i X i    i i X 2i   
k k k

ii X i X j  e (4)
i 1 i 1 i 1

Later, developed mathematical models were used to plot the 3D response surface contour graphes in order to
study the interactive effect of independent variables on the response and validation of developed models was
carried out by plotting actual versus predicted graph and ANOVA. Finally, numerical optimization was
employed to optimize the process variables for higher betalain content from dragon fruit peel using MAE.
TABLE 1. Ranges of independent variables and their levels.
Factor level
Independent variables
-1(-α) 0 1(+α)
Microwave power (W) (A) 400 600 800
Plant material to solvent ratio (g mL-1) (B) 0.10 0.15 0.20
Extraction time (min) (C) 20 40 60

RESULTS AND DISCUSSION

Effect of single factor on the extraction yield of patchouli oil

1. Effect of microwave power
Microwave power is the key parameter which influence the yield of essential oil from Pogostemon cablin
leaves significantly and directly related to the quantity of sample and the extraction time required. So,
experiments were carried out to study the effect of microwave power (400–800 W) on the yield of patchouli oil.
From the results, it is observed that the yield of patchouli oil is increased linearly with increasing microwave
power from 400 to 600 W (Fig. 4a and b). This is due to the fact that microwave power provides localized
heating in the sample, which acts as a driving force for MAE to destroy the plant matrix so that the solute can
diffuse out and dissolve in the solvent. Therefore, increasing the power will generally improve the extraction
yield and result in shorter extraction time [18,19].
When microwave power than 600 W, the yield of patchouli oil decreased with the further increase of
microwave power. These data suggest that applying a higher microwave power for a short time maybe the most

This article is copyrighted as indicated in the article. Reuse of AIP content050014-3

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 effective way to extract essential oil from Pogostemon cablin leaves using MAE. However, a higher microwave
power may lead to decrease the yield of patchouli oil because of the loss of chemical structure of the active
compounds and thermal degradation of the essential oil. This result indicates that microwave power of 600 W is
enough to extraction of patchouli oil in the present work.

2. Effect of plant material to solvent ratio

Plant material to solvent ratio is one of the crucial parameters which influence the extraction of essential oil
from Pogostemon cablin leaves. In order to investigate the effect of plant material to solvent ratio on the yield of
patchouli oil, experiments were carried out in various plant material to solvent ratio ranges from 0.10 to 0.20 g
ml-1 and the results are shown in Fig. 4a and c. From the results, it is observed that the yield of patchouli oil is
increased linearly with increasing plant material to solvent ratio up to 0.15 g ml-1. This can be explained by the
fact that increases in plant material to solvent ratio would increase the larger surface area and enhances the
extraction efficiency. The solvent volume must be sufficient to guarantee that the entire sample is immersed in
the solvent throughout the entire irradiation process, especially when using a matrix that will swell during the
extraction [20,21,22].
In case of higher plant material to solvent ratio, where amount of water is less, the plant material might have
experienced the overheating or charring, resulting in the decreased rate and yield. For high water content, the
heat could be wasted in heating up the water which might have reduced the efficiency of the process. Also,
hydrolytic effect might have contributed to the lower yield [23]. However, plant material to solvent ratio beyond
0.15 g ml-1 shows negative impact on the yield of patchouli oil.

3. Effect of extraction time

In MAE the period of heating is another important factor to be considered. Extraction times in MAE are very
short compared to conventional techniques and usually higher extraction time tends to increase the extraction
yield. However, this increase was found to be very small with longer time [24]. In this study, the yield of
patchouli oil increased significantly or rapidly as the increasing of extraction time from 1 to 20 min, and then
the yield of patchouli oil slightly increased with extraction time beyond 20 min.
It appears in Fig. 4b and c which show that along with increasing extraction time, the yield of patchouli oil
produced will increase. The yield of patchouli oil couldn’t reach its maximum in 60 min during the MAE
process. If the extraction time was more than 60 min, the extraction yield of patchouli oil slightly increased with
the increase of extraction time. Moreover, the excessive time exposure in the microwave field may lead to
decrease the yield of patchouli oil because of the loss of chemical structure of the active compounds and thermal
degradation of the patchouli oil.

Statistical analysis and model fitting

The results of the RSM analysis carried out as shown in Section 2.3 are shown in Table 2. Experimental
extraction yields were used to determine the coefficients of the response surface equation (Eq. (4)). The
regression coefficients of the intercept, linear, quadratic, and interaction terms of the model were calculated
using the least square technique and are presented in Table 3. Obtained second-order polynomial equation (Eq.
(5)) was found well to represent the experimental data (R2 = 0.9478).
The analysis of variance for the experimental results of the BBD is also shown in Table 4. The R2 and adj-R2
values were 0.9478 and 0.8808 respectively (Table 4). A high R2 indicates that the variation could be accounted
for by the data satisfactorily fitting the model. Since CV is a measure expressing the standard deviation as a
percentage of the mean, smaller values of CV give better reproducibility. The coefficient of variation (CV) of
less than 10 indicated that the model was reproducible [25]. The Predicted Residual Sum of Squares (PRESS)
for the model, which is a measure of how a particular model fits each point in the design, was 2.7900. The
model F-value, 14.1400, implied that the model was significant. Adequate precision measures the signal-to
noise ratio. A ratio greater than 4 is desirable [26]. For the proposed models, this value was 10.9600, a very
good signal-to-noise ratio. All these statistical parameters show the reliability of the models.
The p-values were used as a tool to check the significance of each coefficient, which in turn may indicate the
pattern of the interactions between the variables. The smaller was the value of p, the more significant was the
corresponding coefficient. It can be seen from this table that the linear coefficients (B,C), quadratic coefficients
(C2) and cross product coefficients (AB, AC, BC) were not significant (p > 0.05). The other term coefficient (A,
A2, B2) was significant, with very small p values (p < 0.05).

This article is copyrighted as indicated in the article. Reuse of AIP content050014-4

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 TABLE 2. Response surface central composite design and results for extraction yield of patchouli oil.
Actual variables Yield of EO (%)
Run
A (W) B (g mL-1) C (min) Experimental Predicted Residue
1 600 0.15 40 2.7615 2.7618 -0.0003
2 800 0.15 20 2.2264 2.2035 0.0229
3 400 0.15 60 1.8328 1.8562 -0.0234
4 400 0.10 40 1.4547 1.5460 -0.0913
5 800 0.15 60 2.4139 2.3436 0.0703
6 600 0.15 40 2.7615 2.7618 -0.0003
7 800 0.20 40 2.0053 1.9145 0.0907
8 800 0.10 40 1.8899 2.0752 -0.1852
9 600 0.20 60 2.2048 2.3668 -0.1620
10 600 0.15 40 2.7615 2.7618 -0.0003
11 600 0.10 20 2.5070 2.3455 0.1615
12 600 0.15 40 2.7615 2.7618 -0.0003
13 400 0.15 20 1.6110 1.6818 -0.0760
14 600 0.10 60 2.5554 2.4414 0.1140
15 400 0.20 40 1.6193 1.4346 0.1847
16 600 0.20 20 2.0335 2.1481 -0.1145
17 600 0.15 40 2.7615 2.7618 -0.0003

TABLE 3. Analysis of variance for the fitted models.

Source DF Coefficient Sum of square Mean square F-value p-Value
Model 9 3.1700 0.0350 14.1400 0.0010
Residual 7 0.0170 0.0250
Lack of fit 3 0.0170 0.0580
Pure error 4 0.0000 0.0000
Total 16 3.3500
R2 0.9478
Adj-R2 0.8808
CV 7.0300
PRESS 2.7900
Standard deviation 0.1600
Adequate precision 10.9600

TABLE 4. The significance of each response variable effect showed by using F ratio and p value in the nonlinier second
order model.
Variables DFa SSb MSc F-value p-Valued
Linier effects A 1 0.5100 0.5100 20.4200 0.0027
B 1 0.0370 0.0370 1.4800 0.2625 nse
C 1 0.0490 0.0490 1.9800 0.2018 nse
Quadratic effects A2 1 1.8400 1.8400 73.9600 <0.0001
B2 1 0.5400 0.5400 21.5900 0.0024
C2 1 0.0260 0.0260 1.0500 0.3398 nse
Interaction effects AB 1 0.0006 0.0006 0.0240 0.8806 nse
AC 1 0.0003 0.0003 0.0120 0.9165 nse
BC 1 0.0038 0.0038 0.1500 0.7087 nse
a
Degree of freedom
b
Sum of squares
c
Mean sum of squares
d
p values <0.05 were considered to be significant
e
ns: not significant

The predicted values calculated from Eq. (5) were in very good agreement with the experimental values, as
shown in Fig. 2. Hence, this quadratic model is well suited for this experimental set up.

This article is copyrighted as indicated in the article. Reuse of AIP content050014-5

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 FIGURE 2. Experimental extraction yield vs. the predicted extraction yield under optimum extraction conditions.

Perturbation plot
Perturbation plot shows the comparison between all factors at a selected point in the considered design
space. The perturbation plot for the yield of patchouli oil is shown in Fig. 3. The yield response was drawn by
changing only one factor over its range while the other factors were held constant. The plot demonstrates the
effect of all factors at a central point in the design space (e.g., microwave power, plant material to solvent ratio
and extraction time). All factors indicated a positive effect on the yield of patchouli oil. The relatively flat line
of extraction time shows lower effect of this factor on the yield of patchouli oil in the design space. It can be
seen from Eq. (5) and Table 4 the perturbation plot that microwave power and plant material to solvent ratio had
significant curvature effect. The steep curvature in plant material to solvent ratio behavior demonstrated the
response of yield of patchouli oil % was very rapid to these factors. Through the comparison of coefficients in
Eq. (5), the most significant parameter was determined. In this manner, the order of positive influence of the
individual terms on the obtained yield of patchouli oil response was plant material to solvent ratio, microwave
power and extraction time.

FIGURE 3. Pertubation plot for rate response (for A: microwave power, B: plant material to solvent ratio and C: extraction
time).

This article is copyrighted as indicated in the article. Reuse of AIP content050014-6

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 Yield of patchouli oil (%) = –7.41415 + 0.021382A + 41.05175B
+ 0.016369C – 0.00123AB – 0.00000214375AC + 0.030725BC
– 0.0000165419A2 – 143.01B2 – 0.0001976C2 (5)

Optimization of extraction conditions of patchouli oil

The best way of expressing the effect of any parameter on the yield within the experimental space under
investigation was to generate response surface plots of the equation. Response surfaces were plotted by using
Design-Expert software (version 9.0.4.1) to study the effects of parameters and their interactions on yield of
patchouli oil. The results of yield of patchouli oil affected by microwave power, plant material to solvent ratio
and extraction time are presented in Figs. 4 and 5.

Fixed extraction time: 40 min Fixed plant material to solvent ratio: 0.15 g ml-1

Fixed microwave power: 600 W

FIGURE 4. Response surface (3-D) showing the effect of the microwave power (A), plant material to solvent ratio (B) and
extraction time (C) on the patchouli oil extraction yield

In Figs. 4a and 5a, when the 3-D response surface plot and the contour plot were developed for the yield of
patchouli oil with varying microwave power and plant material to solvent ratio at fixed extraction time 40 min.
The yield of patchouli oil was increasing evidently as the increasing of microwave power and plant material to
solvent ratio, and nearly reached a peak at the moderate microwave power and plant material to solvent ratio
tested. The highest extraction yield occured at microwave power 634 W and plant material to solvent ratio 0.14
g ml-1.
In Figs. 4b and 5b, when the 3-D response surface plot and the contour plot were developed for the recovery
of patchouli oil with varying microwave power and extraction time at fixed plant material to solvent ratio 0.15 g
ml-1, it can be seen that maximum recovery of patchouli oil can be achieved when microwave power and
extraction time were 634 W and 52 min, respectively.
Figs. 4c and 5c shows the effect of the plant material to solvent ratio and extraction time on the yield of
patchouli oil at a fixed microwave power 600 W. At a definite extraction time, the yield of patchouli oil
increased slightly as the increasing of plant material to solvent ratio and nearly reached a peak at the moderate
plant material to solvent ratio tested. It can be seen that maximum recovery of patchouli oil can be achieved
when plant material to solvent ratio and extraction time were around 0.14 g ml -1 and 52 min, respectively.

This article is copyrighted as indicated in the article. Reuse of AIP content050014-7

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 Fixed extraction time: 40 min Fixed plant material to solvent ratio: 0.15 g ml-1

Fixed microwave power: 600 W

FIGURE 5. Contour plots showing the effect of the microwave power (A), plant material to solvent ratio (B) and extraction
time (C) on the patchouli oil extraction yield

It can be concluded that optimal extraction conditions of patchouli oil was microwave power 634.024 W,
plant material to solvent ratio 0.147648 g ml-1 and extraction time 51.6174 min (Fig. 6). At this optimized
condition, the predicted yield of the patchouli oil was 2.80516%.

FIGURE 6. Graphical optimization of extraction yield of patchouli oil (%) and its desirability

This article is copyrighted as indicated in the article. Reuse of AIP content050014-8

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 Model adequacy checking
Usually, it is necessary to check the fitted model to ensure that it provides an adequate approximation to the
real system. Unless the model shows an adequate fit, proceeding with the investigation and optimization of the
fitted response surface likely give poor or misleading results [27]. The residuals from the least squares fit play
an important role in judging model adequacy [26]. By constructing a normal probability plot of the residuals, a
check was made for the normality assumption, as given in Fig. 7. The normality assumption was satisfied as the
residual plot approximated along a straight line. Fig. 8 presents a plot of residuals versus the predicted response.
The general impression is that the residuals scatter randomly on the display, suggesting that the variance of the
original observation is constant for all values of Y. Both of the plots (Figs. 7 and 8) are satisfactory, so we
conclude that the empirical model is adequate to describe the patchouli oil yield by response surface.

FIGURE 7. Normal probability of internally studentized residuals

FIGURE 8. Plot of internally studentized residuals vs. predicted response

Validation of the model

To compare the predicted result with the practical value, the rechecking experiment was performed using
some extraction condition. The value obtained from real experiments, demonstrated the validity of the RSM
model, since there was no significant (p > 0.05) differences (Table 5). The strong correlation between the real

This article is copyrighted as indicated in the article. Reuse of AIP content050014-9

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 and the predicted results confirmed that the response model was adequate to reflect the expected extraction
condition.
TABLE 5. Predicted and experimental values of the responses at some extraction condition.
Extraction condition Yield of patchouli oil (%)
Microwave Plant material to solvent Extraction
Experimental Predicted
power (W) ratio (g mL-1) time (min)
400 0.10 60 1.5048 1.5237
600 0.15 60 2.7748 2.7616
800 0.15 40 2.3641 2.3524

CONCLUSIONS
Box-Behnken response surface experimental design (BBD) has proved to be effective in estimating the
effect of three independent variables; microwave power, plant material to solvent ratio and extraction time on
the patchouli oil yield and for predicting the optimal operational conditions. The experimental results showed
that linear terms of two independent variables (microwave power and plant material to solvent ratio) had
significant effects on the response value. Based on the analysis of variance and the agreement of the
experimental and predicted results, it can be concluded that the generated model was suitable for the simulation
of microwave-assisted extraction of patchouli oil with any combination of tested variables. The optimal
conditions were as follows: microwave power 634.024 W, plant material to solvent ratio 0.147648 g ml-1 and
extraction time 51.6174 min. At this optimized condition, the predicted yield of the patchouli oil was 2.80516%.

REFERENCES
1. L. F. Hu, S. P. Li, H. Cao, J. J. Liu, J. L. Gao, F. Q. Yang, and Y. T. Wang, J. Pharm. Biomed. Anal. 42,
200–206 (2006).
2. M. Singh, S. Sharma, and S. Ramesh, Ind. Crops Prod. 16, 101–107 (2002).
3. A. Srikrishna and G. Satyanarayana, Asymmetry 16, 3992–3997 (2005).
4. C. Wildwood, O livro dos perfumes (São Paulo, Brazil, 1994).
5. F. Deguerry, L. Pastore, S. Wu, A. Clark, J. Chappell, and M. Schalk, Arch. Biochem. Biophys. 454, 123–
126 (2006).
6. T. J. Betts, J. Chromatogr. A 664, 295–300 (1994).
7. Federal Regulations Code. Food and Drugs Administration, from the U.S. Government Printing Office via
GPO Access [CITE: 21CFR172.510], U.S.A., 3 pp. 49–52 (2002).
8. J. Wu, X. Lu, W. Tang, H. Kong, S. Zhou, and G. Xu, J. Chromatogr. A 1034, 199–205 (2004).
9. J. Song, D. Li, C. Liu, and Y. Zhang, Innov. Food Sci. Emerg. Technol. 12, 282–287 (2011).
10. Z. Xianzhe, Y. Fangping, L. Chenghai, and X. Xiangwen, J. Northeast Agric. Univ. (Engl. Ed.) 18, 79–86
(2011).
11. K. Zhong, W. Lin, Q. Wang, and S. Zhou, Int. J. Biol. Macromol. 51, 612–617 (2012).
12. C. Da Porto, D. Voinovich, and A. Natolino, J. Supercrit. Fluids 68, 45-51 (2012).
13. J. Azmir, I. S. M. Zaidul, M. M. Rahman, K. M. Sharif, F. Sahena, and M. H. A. Jahurul, Ind. Crops Prod.
52, 405-412 (2014).
14. B. Prakash, R. Shukla, P. Singh, P. K. Mishra, N. K. Dubey, and R. N. Kharwar, Food Res. Int. 44, 385–
390 (2011).
15. V. K. Raina, S. K. Srivastava, and K. V. Syamasunder, Flav. Fragr. J. 18 (1), 18–20 (2003).
16. R. Randhir, Y. I. Kwon, and K. Shetty, Innov. Food Sci. Emerg. Technol. 9, 355–364 (2008).
17. K. Thirugnanasambandham, V. Sivakumar, and P. Maran, Carbohyd. Polym. 112, 622–626 (2014).
18. Z. Hu, M. Cai, and H. H. Liang, Sep. Purif. Technol. 61 (3), 266-275 (2008).
19. C. H. Chan, R. Yusoff, G. C. Ngoh, and F. W. L. Kung, J. Chromatogr. A 1218, 6213–6225 (2011).
20. C. S. Eskilsson and E. Björklund, J. Chromatogr. A 902, 227–250 (2000).
21. V. Mandal, Y. Mohan, and S. Hemalath, Phcog. Rev. 1 (1), 7–18 (2007).
22. P. Tatke and Y. Jaiswal, Res. J. Med. Plants 5, 21–31 (2011).
23. M. A. Desai, J. Parikh, and A. K. De, Chem. Eng. Res. Des. 92, 798-803 (2014).
24. Y. Wang, J. You, Y. Yu, C. Qu, H. Zhang, and L. Ding, Food Chem. 110 (1), 161–167 (2008).
25. R. L. Mason, R. F. Gunst, and J. L. Hess, Statistical Design and Analysis of Experiments: With
Applications to Engineering and Science, Wiley Series in Probability and Mathematical Statistics (John
Wiley and Sons, New York, 1989).

This article is copyrighted as indicated in the article. Reuse of AIP content050014-10

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
 26. R. H. Myers and R. C. Montgomery, Response Surface Methodology, Process and Product Optimization
using Design Experiment (Wiley, New York, 2002).
27. H. Li, L. Pordesimo, and J. Weiss, Food Res. Int. 37, 731–738 (2004).

This article is copyrighted as indicated in the article. Reuse of AIP content050014-11

is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
125.167.33.114 On: Wed, 06 Jan 2016 04:16:50
View publication stats