Ultrasonics Sonochemistry 20 (2013) 595–602

Contents lists available at SciVerse ScienceDirect

Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultson

Optimization of olive leaf extract obtained by ultrasound-assisted extraction with

response surface methodology
Selin Sß ahin a,⇑, Rüya Sß amlı b
a
Istanbul University, Engineering Faculty, Department of Chemical Engineering, 34320 Avcilar, Istanbul, Turkey
b
Istanbul University, Engineering Faculty, Department of Computer Engineering, 34320 Avcilar, Istanbul, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: In the present article, ultrasound-assisted extraction (UAE) of polyphenols from agricultural and indus-
Received 29 March 2012 trial waste of olive oil and table oil productions, olive tree (Olea europaea) leaves were investigated.
Received in revised form 25 May 2012 The aim of the study is to examine the extraction parameters such as solvent concentration (0–100% eth-
Accepted 31 July 2012
anol (EtOH), v/v), the ratio of solid to solvent (25–50 mg/mL) and extraction time (20–60 min), and to
Available online 11 August 2012
obtain the best possible combinations of these parameters through response surface methodology
(RSM). The extract yield was stated as mg extract per g of dried leaf (DL). Total phenolic content was
Keywords:
expressed in gallic acid equivalent (GAE) per g of dried leaf. Free radical scavenging activity for the anti-
Olive leaves
Ultrasound-assisted extraction
oxidant capacity was tested by 1,1-diphenyl-2-picryl hydrazyl (DPPH) radical. The second order polyno-
Polyphenols mial model gave a satisfactory description of the experimental data. 201.2158 mg extract/g DL,
Antioxidant activity 25.0626 mg GAE/g DL, and 95.5610% in respect to inhibition of DPPH radical were predicted at the opti-
Optimization mum operating conditions (500 mg solid to 10 mL solvent ratio, 60 min of extraction time and 50% EtOH
RSM composition), respectively.
 2012 Elsevier B.V. All rights reserved.

1. Introduction parameters affecting the industrial processes employed for extrac-

tion such as preliminary preparations, particle size of the extracted
Being in search of inexpensive, renewable and abundant material, solvent type, solvent composition, solid to solvent ratio,
sources of polyphenol is attracting world-wide interest. Industries extraction temperature and pressure, extraction time and pH.
generate a large quantity of by-products as a waste, containing Acquiring the optimal operating conditions for an extraction meth-
high-added value chemical compounds, which are known as phy- od is a must for commercial applications of the process. In order to
tochemicals, occurring naturally in plants. These biologically active optimize a process, one-factor at-a-time approach is generally ap-
compounds are applied for the preparation of dietary supplements, plied. This classical method is time consuming and expensive. Fur-
nutraceuticals, functional food ingredients or cosmeceuticals [1]. thermore, possible interactions among other operating parameters
The largest group of phytochemicals comprises polyphenols. Plant are ignored, which gives rise to misleading conclusions. Actually,
polyphenols are secondary metabolites, which are distinguished by the response of a process occurs by the interactions of different
their water solubility, high molecular weights (ranging from 500 to variables affecting the operation. Response surface methodology
3000–4000 D), 12–16 phenolic groups and 5–7 aromatic rings per (RSM) considers the probable interactions between operation
1000 relative molecular mass [2]. parameters. RSM is a collection of statistical and mathematical
In this study, Olive tree (Olea europaea) leaf native to the Med- techniques used for developing, improving and optimizing pro-
iterranean countries was selected as the potential antioxidant cesses. This methodology has been used to optimize the extraction
source rich in polyphenol. As an agricultural and industrial waste, processes of polyphenols from various plant materials through sev-
olive leaves are a cheap, renewable and abundant source of poly- eral conventional and novel methods like conventional solid–liquid
phenols. However, polyphenolic profile of olive leaves is known extraction (SLE), microwave-assisted extraction (MAE), ultra-
to be affected by several factors such as leaf age, degree of ripeness, sound/microwave-assisted extraction (UMAE), supercritical fluid
geographical origin, cultivar, phonological stage during sampling, extraction (SFE), subcritical extraction (SCE) and UAE [5–14] (see
proportion of brunches on the tree, moisture content and degree Table 1).
of contamination with soil [3,4]. There are also technological Since there is a diversity in composition of polyphenols depend-
ing on the plant structure, cultivar and harvesting time as men-
⇑ Corresponding author. Tel.: +90 212 4737070x17656; fax: +90 212 4737180. tioned above, a universal extraction protocol does not exist.
E-mail address: [email protected] (S. Sßahin). Extraction methods for each natural polyphenol source must be

1350-4177/$ - see front matter  2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ultsonch.2012.07.029
596 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602

Table 1
Summary of the researches published on the extraction of natural polyphenol sources, optimized by RSM.

Plant material Extraction Process parameters The most effective parameter Refs.
method
Wheat Conventional Solvent concentration, extraction temperature, extraction time. Solvent concentration [5]
SLE
Pine sawdust/ Conventional Extraction temperature, solvent to solid ratio, extraction time. Solvent to solid ratio [6]
almond SLE
hulls
Wheat bran UAE Solvent concentration, extraction temperature, extraction time. Extraction time [7]
Vetiveria SFE Extraction temperature, extraction pressure, extraction time. Extraction pressure [8]
zizanioides
Vetiver grass SFE Co-solvent concentration, extraction temperature, extraction pressure. Co-solvent concentration and extraction [9]
pressure
Longan fruit UAE Ultrasonic power, extraction temperature, extraction time. Ultrasonic power, extraction temperature, [10]
pericarp extraction time
Inga edulis Conventional Solvent concentration, extraction temperature, extraction time. Solvent concentration, and extraction [11]
leaves SLE temperature
Black currants Conventional Solvent concentration, extraction temperature, solvent to solid ratio. Solvent to solid ratio [12]
SLE
Tomatoes UMAE/UAE Microwave power/extraction temperature, extraction time, solvent to Microwave power and solvent to solid ratio. [13]
solid ratio.
Grape pomace Conventional Extraction temperature, extraction time, solvent to solid ratio/extraction Extraction temperature, extraction time, solvent [14]
SLE/SFE temperature, extraction pressure, presence of co-solvent. to solid ratio, presence of co-solvent

designed and optimized. This research aims to be a useful engi- the procedure of Malik and Bradford [15]. Folin–Ciocalteu reagent
neering tool to develop an extraction process from olive leaves, was used as oxidizing agent. To 10 Folin–Ciocalteu reagent of ex-
considering the difficulty of obtaining natural plant extracts due tract, 190 lL of water was added. 1 mL of Folin–Ciocalteu reagent
to the lack of an extraction protocol of these specific products. and 800 lL of Na2CO3 (75%, w/v) were added. The samples were
incubated for 30 min. The absorbance was measured at 760 nm.
2. Materials and methods The amount of total phenolic content was expressed in gallic acid
equivalent per g of dried leaf (mg GAE/g dried matter). A calibra-
2.1. Plant material tion curve was calculated using pure gallic acid concentrations
ranging from 0.065 to 1.04 mg/mL with a regression coefficient
Tavsßan yüreği variety of olive tree leaves was picked randomly of 0.9977 in 100% EtOH, 0.07–1.13 mg/mL with a regression
from the same tree grown in Kasß in Mediterranean area of Turkey, coefficient of 0.9986 in 50% EtOH, 0.175–1.4 mg/mL with a regres-
presenting Mediterranean climate at pretty high altitude. After col- sion coefficient of 0.9963 in water.
lecting, they were both dried and stored at ambient temperature in
the dark. Before extraction processes, they were ground into parti- 2.5. Determination of antioxidant capacity
cles whose average diameters were between 0.9 and 2.0 mm.
The 1,1-diphenyl-2-picryl hydrazyl (DPPH) assay was done
2.2. Chemicals according to a method [16] with some modifications. Samples
were tested individually at a final concentration of 100 lM by
Ethanol and methanol were provided from Merck and were of diluting a methanolic solution of DPPH radical (500 lM). The mix-
>99.5% and >99.8% mass fraction purity, respectively. Folin–Ciocal- tures were vigorously mixed and allowed to stand in the dark for
teu reagent, sodium carbonate, gallic acid and 1,1-diphenyl-2-pic- 30 min at 25 C. The absorbance of the resulting solution was mea-
ryl hydrazyl (DPPH) were purchased from Sigma–Aldrich. 18 mO sured using a spectrophotometer at 517 nm against a blank sample
deionised water from a Millipore Milli-Q water purification system without DPPH. The ability to scavenge the DPPH radical was calcu-
was used to prepare mixtures analyses. lated by using the equation below:

2.3. Ultrasound-assisted extraction Antioxidant capacity ð% inhibitionÞ

¼ ½ðAcontrol  Asample Þ=Acontrol   100 ð1Þ
Ultrasound-assisted extraction was conducted in an ultrasonic
bath (Protech, 220 V and 50 Hz) at 25 C. Five hundred milligram
where Acontrol is the absorbance of the control (DPPH solution with-
of dried and ground leaves and solvent were sealed in an erlenma-
out sample) and Asample is the absorbance of the test sample.
yer flask and placed into the bath. The mixture was centrifuged
(Nüve, CN 180) at 5000  g for 25 min. After centrifugation, the
supernatant was filtered through a 0.45 lm syringe filter and 2.6. Response surface methodology
stored at 80 C until analysis for the biochemical measurements.
For the extract yields, the solvent was removed from a certain Response surface method (RSM) is an experimental optimiza-
quantity extract in a rotary evaporator (Buchi, Switzerland). tion procedure based on physical experiments or computer exper-
iments (simulations) [17,18] and experimented observations. It
2.4. Determination of total polyphenols was first developed in 1951 by G.E.P. Box and K.B. Wilson [19].
In most of the methods, the effects of parameters to each other
The concentration of total phenols in extracts was measured by are ignored, and only the effects of distinct parameter to the results
UV-spectrophotometry (Optima, SP-300), based on colorimetric are calculated. However, parameters usually affect each other,
oxidation/reduction reaction. The total phenolic content was especially in physical experiments. For example if the parameters
determined according to the Folin–Ciocalteu method by following are density and temperature, handling only the distinct effect
 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602 597

Table 2 In this study, three independent parameters were used (Table 2).
Values of the independent parameters and their coded forms with their symbols Those were X1 (solid/solvent ratio, mg/mL), X2 (time, min) and X3
employed in RSM for optimization of olive leaf extraction through UAE.
(solvent concentration, %, v/v).
Independent Symbols of the Original values of Coded forms of On the other hand, the responses were Y1 (extract yield, mg ex-
parameters parameters the parameters the parameters tract/g dried leaf), Y2 (total polyphenol content, mg GAE/g dried
Solid/solvent ratio X1 500/10 1 leaf) and Y3 (antioxidant activity, % inhibition). Thus, the function
(mg/mL) 500/15 0 of this study containing three independent variables is expressed
500/20 1
Time (min) X2 20 1
with Eq. (4):
40 0
60 1 Y n ¼ b0 þ b1 X 1 þ b2 X 2 þ b3 X 3 þ b11 X 21 þ b22 X 22 þ b33 X 23
EtOH concentration X3 0 1 þ b12 X 1 X 2 þ b13 X 1 X 3 þ b23 X 2 X 3 ð4Þ
(%, v/v) 50 0
100 1

2.7. Statistical analysis

cannot be enough, since temperature can also affect density. Three replicate extractions were carried out for each of the sam-
Accordingly, RSM is a method that aims to remove this drawback. ples followed by a minimum of three spectrophotometric measure-
A function is supposed to be obtained for prediction of the re- ments from each extract. Statistical analysis on the means of
sults in response surface method. The general formula of this func- triplicate experiments was carried out using the ANOVA procedure
tion is of the InStat software, version 3.0 (GraphPad, San Diego, CA, USA).
Tukey’s test of significance between means was used for illustra-
X
k X
k X
k
tion of significance.
Y ¼ b0 þ bi X i þ bi X i X j þ bii X 2i þ . . . ð2Þ
The regression coefficients in the equation of RSM of this study
i¼1 i<j i¼i
were calculated with MATLAB (Mathematical Laboratory) simula-
where Y represents the result (or with the name in RSM – response), tion program, version 2009b. The coefficients are expressed as sur-
bi, "i are regression coefficients, Xi, "i are the independent face plots using RSM to visualize the relationship between the
parameters. response and experimental results of each variable.
If the response is defined by a linear function of independent
parameters, then it is a first-order function that can be expressed 3. Results and discussions
as Eq. (3):
3.1. Effect of process parameters on the UAE performance
Y ¼ b0 þ b1 X 1 þ b2 X 2 ð3Þ
Generally first-order equations cannot give the desired approx- The experimental design was presented in Table 2. Table 3 indi-
imation to the real experiments, so second-order models or models cates the influence of EtOH concentration (0%, 50%, 100%; v/v), the
with more parameters are used. ratio of solid to solvent (500/10–20 mg/mL), and extraction time

Table 3
Experimental values of the extract yield, total polyphenolic content and antioxidant activity of the olive leaf extracts obtained by UAE at various conditions.*

Run no X1 (mg/mL) X2 (min) X3 (%, v/v) Extract yield** (mg/g DL) Total polyphenol content*** (mg GAE/g DL) Antioxidant activity*** (% inhibition)
1 500/10 20 100 84.39 ± 2.39 a 7.88 ± 0.12 a 95.62 ± 0.03 a
2 500/15 20 100 77.78 ± 2.87 a 6.47 ± 0.13 b 95.32 ± 0.06 b
3 500/20 20 100 73.11 ± 2.13 a 5.18 ± 0.32 c 95.75 ± 0.04 c
4 500/10 40 100 78.37 ± 3.38 a 9.29 ± 0.17 d 96.99 ± 0.02 d
5 500/15 40 100 79.07 ± 3.07 a 7.29 ± 0.21 ab 96.22 ± 0.03 e
6 500/20 40 100 36.16 ± 3.84 b 5.93 ± 0.04 bc 96.35 ± 0.04 f
7 500/10 60 100 115.06 ± 5.94 c 10.15 ± 0.10 d 94.16 ± 0.01 g
8 500/15 60 100 55.17 ± 4.19 d 8.45 ± 0.47 ad 95.23 ± 0.03 b
9 500/20 60 100 71.57 ± 3.43 a 6.45 ± 0.15 b 96.52 ± 0.04 h
10 500/10 20 50 158.33 ± 6.42 ef 14.18 ± 0.18 e 93.14 ± 0.03 i
11 500/15 20 50 142.88 ± 7.12 e 15.64 ± 0.36 f 93.60 ± 0.02 j
12 500/20 20 50 158.59 ± 6.42 ef 10.20 ± 0.15 d 93.24 ± 0.02 k
13 500/10 40 50 165.24 ± 4.74 f 20.37 ± 0.63 g 94.11 ± 0.02 g
14 500/15 40 50 200.33 ± 6.28 gh 17.88 ± 0.13 h 95.73 ± 0.02 c
15 500/20 40 50 248.83 ± 6.17 i 12.42 ± 0.23 i 97.97 ± 0.01 l
16 500/10 60 50 203.89 ± 6.11 g 19.51 ± 0.49 g 94.92 ± 0.01 m
17 500/15 60 50 171.75 ± 5.26 f 19.58 ± 0.43 g 95.22 ± 0.01 b
18 500/20 60 50 206.01 ± 6.00 g 15.82 ± 0.18 f 98.73 ± 0.01 n
19 500/10 20 0 169.68 ± 6.24 f 9.98 ± 0.49 d 86.31 ± 0.02 o
20 500/15 20 0 171.20 ± 5.81 f 7.92 ± 0.08 a 87.71 ± 0.04 p
21 500/20 20 0 167.85 ± 3.16 f 6.58 ± 0.22 b 88.07 ± 0.04 r
22 500/10 40 0 189.05 ± 5.95 h 8.67 ± 0.23 ad 87.54 ± 0.02 s
23 500/15 40 0 196.64 ± 3.37 gh 9.27 ± 0.28 d 90.05 ± 0.04 t
24 500/20 40 0 202.14 ± 7.34 gh 8.75 ± 0.23 ad 89.43 ± 0.02 u
25 500/10 60 0 213.47 ± 3.47 g 12.99 ± 0.02 i 89.19 ± 0.01 v
26 500/15 60 0 212.49 ± 6.52 g 9.45 ± 0.20 d 89.53 ± 0.02 y
27 500/20 60 0 207.51 ± 4.37 g 8.75 ± 0.30 ad 87.82 ± 0.01 z
*
Means within the same column not sharing a common letter indicate significant difference at p < 0.05.
**
Data are expressed as the mean (n = 3)±S.D.
***
Data are expressed as the mean (n = 9)±S.D.
598 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602

Table 4 extract and total phenolic content, as a result of its lower viscosity
Regression coefficients of the RSM for three response parameters. and also its altering the plant structure by swelling the matrix,
Coeffiient Predicted coefficients which enables the solvent to more completely penetrate the
Extract Total polyphenol Antioxidant leaves. Accordingly, water is acting as the plant swelling agent,
yield content activity while ethanol is believed to disrupt the bonding between the sol-
b0 140.6288 1.5586 83.3700
utes and plant matrices [20,21]. Therefore, the mixture of water
b1 6.1846 0.0029 0.1893 and EtOH as solvent agent exhibited the best performance to ex-
b2 3.0813 0.3018 0.0778 tract polyphenols of all the extractants used. Another explanation
b3 1.7527 0.4685 0.1904 might be the high dielectric constant of water, which leads to in-
b11 0.2991 0.0330 0.0043
crease polarity indices of EtOH with its water solution [22].
b22 0.0182 0.0015 0.0003
b33 0.0202 0.0039 0.0008 Generally, pure EtOH extracts had higher antioxidant capacity
b12 0.0287 0.0255 0.0038 with the values of 94.16–96.99%. On the other hand, pure water ex-
b13 0.0341 0.0075 0.0023 tracts exhibited relatively lower antioxidant capacity, showing dif-
b23 0.0099 0.0007 0.0001
ferent values at p < 0.001. Degradation of polyphenols might occur
by hydrolysis with water, and also by oxidation of the antioxidant
components with formed peroxyl and hydroxyl radicals [23].
(20–60 min) on the extract yield, total phenolic content and anti-
oxidant activity of olive leaves extracted by UAE with the statistical
analysis of the related experiments. The extract yield of UAE ran- 3.1.2. The ratio of solid to solvent
ged from 36.16 to 248.84 mg/g DL through various EtOH ratio. As By decreasing the solid to solvent ratio, the amounts of extracts
for total polyphenols, the quantity changed between 5.18 and and their total polyphenols both decreased, while the antioxidant
20.31 mg GAE/g DL. Antioxidant capacity of the extracts ranged activities of the related extracts did not show any alteration. The
from 86.31 to 97.97% with respect to inhibition of DPPH radical. decrease of the yields with the increase of solvent quantity is
inconsistent with mass transfer principles, since the concentration
3.1.1. Solvent concentration gradient which is driving force is supposed to be higher when a
While 50% EtOH as extraction solvent was distinguished by the lower solid to solvent ratio used, leading to higher diffusion. The
highest level of extract yield, pure ethanol (100% EtOH) showed the same tendency of the same parameter was also acquired for the
poorest level of all the other solvent systems. In addition, pure pectin extraction from apple pomace [24], a kind of Chinese herbal
EtOH extracts were found significantly different (p < 0.001) from medicine extraction [25], and felodipine tablet extraction through
the extracts obtained by pure water and those of 50% EtOH. microwave [26]. On the other hand, consuming less solvent for the
Regarding total polyphenols, the best performance was achieved extraction is extremely reasonable and practical state from the
by using 50% EtOH as extractant under different extraction condi- economical point of view.
tions. Although pure water (0% EtOH) is the most polar solvent of As seen in Table 3, the extracts obtained through the conditions
all, it did not yield the best results with respect to extract and total of 500 mg solid to 10 mL, 50% EtOH solution ratio at 40th and 60th
phenolic content. This phenomenon might be attributed to the minutes shared the highest total polyphenols with the values of
higher viscosity of water than that of the other solvents, which is 20.37, 19.51 and 19.58 mg GAE/g DL, which are not significantly
a matter of mass transfer. 50% EtOH showed the greatest yield of different at p > 0.05.

Table 5
Predicted values of the extract yield, total polyphenolic content and antioxidant activity of the olive leaf extracts obtained by RSM at
various conditions.

Run X1 X2 X3 Y1 Y2 Y3
no (mg/mL) (min) (%, v/v) (mg/g DL) (mg GAE/g DL) (% inhibition)
1 500/10 20 100 76.6688 8.1156 95.4790
2 500/15 20 100 63.2133 5.9261 95.4330
3 500/20 20 100 64.7128 5.3866 95.6020
4 500/10 40 100 90.9148 12.2516 96.4350
5 500/15 40 100 74.5893 12.6121 96.0090
6 500/20 40 100 73.2188 4.4226 95.7980
7 500/10 60 100 90.6008 17.5876 97.6310
8 500/15 60 100 71.4053 10.2981 96.8250
9 500/20 60 100 67.1648 4.6586 96.2340
10 500/10 20 50 167.4838 16.9906 93.2090
11 500/15 20 50 162.5533 16.6761 93.7380
12 500/20 20 50 172.5778 18.0116 94.4820
13 500/10 40 50 191.6298 20.4266 94.2650
14 500/15 40 50 183.8293 17.5621 94.4140
15 500/20 40 50 190.9838 16.3476 94.7780
16 500/10 60 50 201.2158 25.0626 95.5610
17 500/15 60 50 190.5453 19.6481 95.3300
18 500/20 60 50 194.8298 15.8836 95.3140
19 500/10 20 0 157.2988 6.3656 86.9390
20 500/15 20 0 160.8933 7.9261 88.0430
21 500/20 20 0 179.4428 11.1366 89.3620
22 500/10 40 0 191.3448 9.1016 88.0950
23 500/15 40 0 192.0693 8.1121 88.8190
24 500/20 40 0 207.7488 8.7726 89.7580
25 500/10 60 0 210.8308 13.0376 89.4910
26 500/15 60 0 208.6853 9.4981 89.8350
27 500/20 60 0 221.4948 7.6086 90.3940
 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602 599

3.1.3. Extraction time Washing is happening at the beginning of the extraction with a
The mechanism of the UAE process here has two main stages. rapid increase. After 20 min, the slow extraction is observed by a
First, dissolution of soluble components on surfaces of the plant low raise in both concentrations. Sixty minutes are accepted as
matrix occurs, which is called also as ‘‘washing’’. Secondly, mass the optimum time in this process, giving the equilibrium concen-
transfer of the solute from the plant matrix into the solvent by dif- tration of the extracted leaves.
fusion and osmotic processes, which is known as ‘‘slow extraction’’ The total phenolic content of all samples increased steadily as
[27–30]. These two phenomena are clearly observed in Table 3. a function of time. There was a tendency to drop of the

Fig. 1. Response surface plots for the extract yields of olive leaves as a function of (a) solvent concentration to time (solid to solvent ratio = 500 mg/15 mL); (b) solvent
concentration to solid/solvent ratio (time = 40 min); (c) solid/solvent ratio to time (solvent concentration = 50%).
600 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602

Fig. 2. Response surface plots for the total polyphenol contents of the extracts as a function of (a) solvent concentration to time (solid to solvent ratio = 500 mg/15 mL); (b)
solvent concentration to solid/solvent ratio (time = 40 min); (c) solid/solvent ratio to time (solvent concentration = 50%).

antioxidant capacity after 40 min. This can be explained by the 3.2. Optimization of UAE by RSM
heating effect of overexposure to ultrasound treatment for longer
extraction time, which leads to the degradation of antioxidant Table 4 shows the coefficient values of extract yield, total poly-
property of the extracts. However, the extending ultrasonic time phenol content and antioxidant activity of the olive leaf extracts.
could result in a higher extract yield. It is also obvious that con- Those values were used for a final predictive equation for each
suming much longer time is an impractical situation from the eco- independent parameter. Predicted values of the extraction condi-
nomical point of view. tions of EtOH concentration (0%, 50%, 100%; v/v), the ratio of solid
 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602 601

to solvent (500/10–20 mg/mL), and extraction time (20–60 min) Y 1 ¼ 140:6288  6:1846X 1 þ 3:0813X 2 þ 1:7527X 3
are presented in Table 5. In order to estimate the relationship be-
þ 0:2991X 21  0:0182X 22  0:0202X 23  0:0287X 1 :X 2
tween independent and dependent parameters, three dimensional
plots were illustrated (Figs. 1–3).  0:0341X 1 :X 3  0:0099X 2 :X 3 ð5Þ
The predicted models for the yield, total polyphenol content
Y 2 ¼ 1:5586  0:0029X 1 þ 0:3018X 2 þ 0:4685X 3 þ 0:0330X 21
and antioxidant activity of the extracts are presented by the fol-
lowing equations, respectively: þ 0:0015X 22  0:0039X 23  0:0255X 1 :X 2  0:0075X 1 :X 3
þ 0:0007X 2 :X 3 ð6Þ

Fig. 3. Response surface plots for the antioxidant activities of the extracts as a function of (a) solvent concentration to time (solid to solvent ratio = 500 mg/15 mL); (b) solvent
concentration to solid/solvent ratio (time = 40 min); (c) solid/solvent ratio to time (solvent concentration = 50%).
602 S. Sßahin, R. Sßamlı / Ultrasonics Sonochemistry 20 (2013) 595–602

Y 3 ¼ 83:3700 þ 0:1893X 1 þ 0:0778X 2 þ 0:1904X 3 References

þ 0:0043X 21 þ 0:0003X 22  0:0008X 23  0:0038X 1 :X 2 [1] W. Peschel, F. Sánchez-Rabaneda, W. Diekmann, A. Plescher, I. Gartzía, D.
 0:0023X 1 :X 3  0:0001X 2 :X 3 ð7Þ Jiménez, R. Lamuela-Raventós, S. Buxaderas, C. Codina, An Industrial approach
in the search of natural antioxidants from vegetable and fruit wastes, Food
Chemistry 97 (2006) 137–150.
[2] C.J. Dillard, J.B. German, Phytochemicals: nutraceuticals and human health –
Response surface plots were constructed according to the Eq.
review, Journal of the Science of Food and Agriculture 80 (2000) 1744–1756.
(5) in order to determine the optimum conditions for the extract [3] N.S. El, S. Karakaya, Olive tree (Olea europaea) leaves: potential beneficial
yield of olive leaves extracted by UAE. Fig. 1a shows the effect of effects on human health, Nutrition Reviews 67 (2009) 632–638.
[4] V.T. Papoti, M.Z. Tsimidou, Impact of sampling parameters on the radical
EtOH concentration and extraction time on the yield of the olive
scavenging potential of olive (Olea europaea L.) leaves, Journal of Agricultural
leaf extract. The extract yield increased with the increase of water and Food Chemistry 57 (2009) 3470–3477.
concentration in the solvent solution at a fixed time. Fig. 1b dem- [5] C. Liyana-Pathirana, F. Shahidi, Optimization of extraction of phenolic
onstrates the influence of EtOH concentration and solid to solvent compounds from wheat using response surface methodology, Food
Chemistry 93 (2005) 47–56.
ratio. The response showed the same tendency with those of [6] M. Pinelo, M. Rubilar, J. Sineiro, M.J. Nunez, Extraction of antioxidant phenolics
Fig. 1a. Rather, the amount of extract rised by water concentration from almond hulls (Prunus amygdalus) and pine sawdust (Pinus pinaster), Food
at a fixed solid to solvent ratio. Fig. 1c shows the effect of time and Chemistry 85 (2004) 267–273.
[7] J. Wang, B. Sun, Y. Cao, Y. Tian, X. Li, Optimisation of ultrasound-assisted
solid to solvent ratio on the olive leaf extraction. Similarly, it has extraction of phenolic compounds from wheat bran, Food Chemistry (2008)
the same trend as those of Fig. 1a and b. 804–810.
Fig. 2, illustrated by means of Eq. (6) represents the effect of [8] L.T. Danh, R. Mammucari, P. Truong, N. Foster, Response surface method
applied to supercritical carbon dioxide extraction of Vetiveria zizanioides
solid/solvent ratio, time and EtOH concentration on the total essential oil, Chemical Engineering Journal 155 (2009) 617–626.
polyphenols of olive leaves extracted. Fig. 2a shows the effect of [9] L.T. Danh, P. Truong, R. Mammucari, N. Foster, Extraction of vetiver essential oil
solvent concentration and extraction time on the total polyphe- by ethanol-modified supercritical carbon dioxide, Chemical Engineering
Journal 165 (2010) 26–34.
nols. The plot of total polyphenols, affected by EtOH concentra-
[10] B. Yang, M. Zhao, J. Shi, N. Yang, Y. Jiang, Effect of ultrasonic treatment on the
tion and time shows a marked increase at the point, where both recovery and DPPH radical scavenging activity of polysaccharides from longan
solvent concentration and time have the average values. Fig. 2b fruit pericarp, Food Chemistry 106 (2008) 685–690.
[11] E.M. Silva, H. Rogez, Y. Larondelle, Optimization of extraction of phenolics
demonstrates the influence of solvent concentration and solid to
from Inga edulis leaves using response surface methodology, Separation and
solvent ratio on polyphenol extraction. Similarly, the increase in Purification Technology 55 (2007) 381–387.
EtOH concentration at a fixed time gave rise to an increase in [12] J.E. Cacace, G. Mazza, Optimization of extraction of anthocyanins from black
the total phenolic content, and reached a maximum at the aver- currants with aqueous ethanol, Journal of Food Science 68 (2003) 240–248.
[13] Z. Lianfu, L. Zelong, Optimization and comparison of ultrasound/microwave
age conditions of solvent concentration and solid to solvent ratio. assisted extraction (UMAE) and ultrasonic assisted extraction (UAE) of
Fig. 2c exhibits the effects of extraction time and solid to solvent lycopene from tomatoes, Ultrasonics Sonochemistry 15 (2008) 731–737.
ratio on the total polyphenol yields. The response showed that the [14] M. Pinelo, A. Ruiz-Rodŕiguez, J. Sineiro, F.J. Se~noŕans, G. Reglero, M.J. Ńu~nez,
Supercritical fluid and solid–liquid extraction of phenolic antioxidants from
yield reached a slight peak at the mean values of the extraction grape pomace: a comparative study, European Food Research and Technology
conditions. 226 (2007) 199–205.
Fig. 3 are the response surface plots to exhibit the effect of pro- [15] N.S.A. Malik, J.M. Bradford, Changes in oleuropein levels during differenciation
and development of floral buds in ‘Arbequina’ olives, Scientia Horticulturae
cess conditions on the antioxidant activity of the extracts obtained 110 (2006) 274–278.
by UAE, based on the Eq. (7). The antioxidant capacity of the ex- [16] J. Yu, L. Wang, R.L. Walzem, E.G. Miller, L.M. Pike, B.S. Patil, Antioxidant activity
tracts increased with the increase of EtOH quantity in the solvent of citrus limonoids, flavonoids, and coumarins, Journal of Agricultural and
Food Chemistry 53 (2005) 2009–2014.
solution at a fixed time (Fig. 3a). Similarly, the increase of EtOH
[17] G.E.P. Box, W.G. Hunter, J.S. Hunter, Statistics for Experimenters, John Wiley &
concentration led to the increase of antioxidant activity at a fixed Sons, New York, 1978.
ratio of solid to solvent (Fig. 3b). The antioxidant activity of the ex- [18] D.C. Montgomery, Design and Analysis of Experiments, fifth ed., Wiley, New
York, 2000.
tracts slightly increased with the increase of extraction time at a
[19] Response Surface Methodology. Available from: <http://en.wikipedia.org/wiki/
fixed solid to solvent ratio (Fig. 3c). Response_surface_methodology> [20 February 2012].
Optimum conditions of UAE process applied for olive leaves [20] Q. Lang, C.M. Wai, Recent advances in extraction of nutraceuticals from plants,
were found as 500 mg solid to 10 mL solvent ratio, 60 min of Trends in Food Science and Technology 17 (2006) 300–312.
[21] Q. Lang, C.M. Wai, Supercritical fluid extraction in herbal and natural product
extraction time and 50% EtOH composition, providing the pre- studies-a practical review, Talanta 53 (2001) 771–782.
dicted values of 201.2158 mg extract/g DL, 25.0626 mg GAE/g DL, [22] G. Spigno, D.M. De Favari, Microwave-assisted extraction of tea phenols: A
and 95.5610% in respect to inhibition of DPPH radical. phenomenological study, Journal of Food Engineering 93 (2009) 210–217.
[23] M. Co, C. Zettersten, L. Nyholm, P.J.R. Sjöberg, C. Turner, Degradation effects in
the extraction of antioxidants from birch bark using water at elevated
temperature and pressure, Analytica Chimica Acta 716 (2012) 40–48.
4. Conclusions [24] S. Wang, F. Chen, J. Wu, Z. Wang, X. Liao, X. Hu, Optimization of pectin
extraction assisted by microwave from apple pomace using response surface
methodology, Journal of Food Engineering 78 (2007) 693–700.
Olive tree (Olea europaea) leaf is a potential source rich in poly-
[25] Z. Guo, Q. Jin, G. Fan, Y. Duan, C. Qin, M. Wen, Microwave-assisted extraction of
phenols. In order to achieve the maximum extraction performance effective constituents from a Chinese herbal medicine Radix puerariae,
by means of UAE, a 50% EtOH, 500 mg dried leaf to 10 mL solvent, Analytica Chimica Acta 436 (2001) 41–47.
and 60 min of time should be employed as optimal operating con- [26] C.S. Eskilssona, E. Björklunda, L. Mathiassona, L. Karlsson, A. Torstensson,
Microwave-assisted extraction of felodipine tablets, Journal of
ditions. Solvent concentration was proved to be the most signifi- Chromatography A 840 (1999) 59–70.
cant parameter of all the parameters used, influencing the [27] D.T. Veličković, D.M. Milenović, M.S. Ristić, V.B. Veljković, Kinetics of ultrasonic
extraction performance obtained by UAE. Consequently, the extraction of extractive substances from garden (Salvia officinalis L.) and
glutinous (Salvia glutinosa L.) sage, Ultrasonics Sonochemistry 13 (2006) 150–
present study was carried out to contribute to the simulation 156.
and optimization of olive leaf extraction. [28] D.T. Veličković, D.M. Milenović, M.S. Ristić, V.B. Veljković, Ultrasonic extraction
of waste solid residues from the Salvia sp. essential oil hydrodistillation,
Biochemical Engineering Journal 13 (2008) 150–156.
Acknowledgement [29] M. Vinatoru, An overview of the ultrasonically assisted extraction of bioactive
principles from herbs, Ultrasonics Sonochemistry 8 (2001) 303–313.
[30] M. Toma, M. Vinatoru, L. Paniwnyk, T.J. Mason, Investigation of the effects of
The authors wish to thank for the support of the Research Fund ultrasound on vegetal tissues during solvent extraction, Ultrasonics
of Istanbul University. Project number is N-22445. Sonochemistry 8 (2001) 137–142.