Ultrasonics Sonochemistry 36 (2017) 286–300

Contents lists available at ScienceDirect

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

Effect of ultrasound-assisted osmotic dehydration pretreatment on the

convective drying of strawberry
Ezzeddine Amami a,⇑, Wissal Khezami a, Salma Mezrigui a, Laxmikant S. Badwaik b, Asma Kammoun Bejar c,
Carmen Tellez Perez d, Nabil Kechaou c
a
Université de Tunis El Manar, Unité de recherche «Chimie des matériaux et de l’Environnement», Institut Supérieur des Sciences Biologiques Appliquées de Tunis (ISSBAT), 9,
rue Zouhair Essafi, Tunis 1006, Tunisia
b
Department of Food Engineering and Technology, School of Engineering, Tezpur University, Napaam, Assam, India
c
Université de Sfax, Groupe de recherche en Génie des Procédés Agroalimentaires, Laboratoire de recherche en Mécanique des Fluides Appliquée – Génie des
Procédés–Environnement, Ecole Nationale d’Ingénieurs de Sfax (ENIS), Tunisia
d
University of La Rochelle-Faculty of Science and Technology, Laboratory of Engineering Science for Environment (LaSIE UMR 7356 CNRS), Avenue Michel Crépeau, 17042 La
Rochelle, France

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

Article history:
Received 29 September 2016
Received in revised form 4 December 2016
Accepted 5 December 2016
Available online 7 December 2016
Keywords:
Ultrasound assisted osmotic dehydration
Convective air drying
Dietetic dehydrated strawberry
Moisture diffusivity with shrinkage
Color and phytochemical analysis
X-ray diffraction
The work was planned to produce dietetic dehydrated strawberries with application of ultrasound osmo-
tic dehydration (UOD) as pretreatment. A Box-Behnken design (BBD) were employed to characterize the
UOD strawberry process, optimize and investigate the effect of independent variables like ultrasound
time (10, 20 and 30 min), concentration of osmotic solution (distilled water, 32.5 and 65 °Brix) and tem-
perature (20, 30 and 40 °C) on the water loss (WL), solid gain (SG) and weight reduction (WR). The same
BBD were used to estimate the main effects of OD without ultrasound treatment. A multi-criteria opti-
mization based on maximal possible values of WL and WR and minimal value of SG was achieved.
This was 20.5 min of ultrasound time, 47.5 °Brix osmotic solution concentration and 31 °C medium tem-
perature. The effects of UOD carried out under the optimized operating conditions on kinetics of convec-
tive air drying of strawberry at 40, 50 and 60 °C, and velocity of 1 m/s, using the phenomenological model
of Coupled Washing/Diffusion (CWD). Starting accessibility, moisture effective diffusivity Deff value and
activation energy (Ea) were calculated from similar-Fick’s law and analogous Arrhenius equation, respec-
tively with and without shrinkage correction. Color difference, Chroma and hue angle of fresh and UOD-
air dried strawberries were evaluated. Quality attributes were estimated through the assessment of the
phenolic content, antioxidant activity, rehydration capacity, and X-ray powder diffraction characteristics.
Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction Drying is a complex operation, which usually includes simulta-

Due to its outstanding wealth of nutrient, delicious flavor and
health benefits strawberry (cv. camarosa) is consumed to maintain
good health by minimizing the risk of chronic illnesses such as the
cardiovascular disease, hypertension, obesity and type 2-diabetes
[1]. According to FAOSTAT (Food and Agriculture Organization
2013), the international production of strawberry was over 4.3 mil-
lion tons in 2011 [2]. Dehydration could preserve nutritional com-
ponents and provide extension of shelf-life of strawberry [3].
Consumption of healthy dehydrated strawberry as a part of food-
stuffs (e.g., ice creams, cookies, breakfast cereals, cakes, energy
bars and dairy products), has gained a noticeable increase [4].
⇑ Corresponding author.
E-mail address: ezzeddineamami@yahoo.fr (E. Amami).
neous phenomena of heat and mass transfers. Models of drying
kinetics have been studied through different fundamental, phe-
nomenological, kinematical, and empirical ways [5]. Recently, a
Coupled Washing/Diffusion CWD model was adequately phe-
nomenologically defined [5]. When one can assume that external
transfers are not the limiting phenomena, the main part of the dry-
ing kinetics is revealed by the diffusion of liquid water within the
matrix. Thus, air flow velocity is assumed to be higher than the
Critical Airflow Velocity CAV. Another key-point of CWD model is
to consider the specific initial short time stage of interaction
between hot air and the superficial zone of the product, which
should be a diffusion-free washing stage; this brings together the
concept of starting accessibility dXs. This Coupled Washing/
Diffusion CWD model allows determining the effective moisture
diffusivity Deff based on similar second Fick’s law with or without
shrinkage consideration [6], by using the only information

http://dx.doi.org/10.1016/j.ultsonch.2016.12.007
1350-4177/Ó 2016 Elsevier B.V. All rights reserved.
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
gathered from experimental drying data excluding the part able to
be attributed to the starting accessibility stage.
Shrinkage is generally closely related to the amount of water
released during drying before the material crosses glass transition
level. It normally leads to a modification in functional behavior
such as the diffusivity value [5]. The shrinkage in biological tissues
is predicted with preferably simple empirical models considering
the volume in the product with respect to their initial values. Ideal
shrinkage model is normally based on assuming that the volume
loss of the product during drying should correspond to the volume
of the evaporated water [7]. A linear relationship between shrink-
age and moisture content during drying is reported in several
works using different drying procedures of numerous fruits and
vegetables [7–10].
Since the drying is highly energy consuming operation, nowa-
days it is very important to act in reducing the energy consump-
tion, increasing the energy efficiency and reducing the processing
time [11]. Besides, normally drying can promote total phenolic
degradation, antioxidant activity reduction and undesirable
changes of color [12].
Osmotic dehydration (OD) with or without ultrasound has
received considerable attention in recent years as one of the sim-
plest and inexpensive pre-drying treatment of strawberries in
order to reduce energy consumption, improve food quality, and
speed up the drying time [13]. Fruits osmotic dehydration (OD)
occurs naturally, following their placement in hypertonic solutions
presenting a high osmotic pressure and low water activity. Conse-
quently a simultaneous diffusion phenomenon takes place where
water outflows from the food to the solution and the solute from
the solution to the food. These mechanisms lead to a water loss
(WL) and solids gain (SG) from and into the food respectively. It
is often desired in this case to maximize the WL and minimize
the SG [14]. The efficiency of the OD process depends upon many
parameters, such as the concentration and temperature of the
osmotic solution, the immersion time, agitation (mechanical, cen-
trifugal or ultrasonic waves), the structure of the fruit and its
geometry, as well as the liquid/solid ratio [15].
New technologies such as ultrasonic (US) waves are recently
used to enhance food osmotic dehydration with tendency to opti-
mize nutrients intake in human diets [16,17]. Ultrasonic waves in a
frequency range of 20–100 kHz can generate quick series of alter-
native compressions and expansions in solid media in an analogue
to a sponge when it is squeezed and relaxed repeatedly. Thus lead-
ing to the deformation of internal microscopic channels porous
fruits that facilitate the convective water removal [18]. Wave fre-
quency, power and treatment time of the US greatly affect the
osmotic dehydration rate and the final product’s structure. Fernan-
des and Rodrigues [17] reported soluble solid losses (mostly sugar)
for several fruits in ultrasound-assisted osmotic dehydration
experiments. Due to the discrepancy in the results obtained for dif-
ferent fruit, more studies on the effect of ultrasound on osmotic
dehydration are still needed.
Response Surface Methodology (RSM) is widely used as a statis-
tical technic in process optimization and product quality ameliora-
tion in a short time period and minimum experiences [19]. RSM
based on the principles of design of experiments (DOE) and opti-
mization techniques is characteristically used for mapping of the
response over the experimental domain to select the optimum
conditions for desirable responses [20–22]. The most frequently
utilized Box–Behnken quadratic design (BBD) was specifically
selected since it requires fewer design points than a Central Com-
posite Design in cases of the same number of factors and operates
at a high efficiency [19]. Several studies have been carried out in
the area of osmotic dehydration of strawberry [13,23,24].
However, to the best of our knowledge RSM is still lacking as a
modeling and optimization tool for ultrasound osmotic dehydration
287
of strawberries. The rehydration is a complex process aims at the
restoration of raw material properties and is widely used as a qual-
ity indicator. Rehydration indicates the chemical and physical
changes caused by drying and pretreatments [25].
X-ray powder diffraction analysis is a common technique of
fundamental importance used to confirm the crystalline–amor-
phous state of sugars of dried products in a powder form. The
amorphous materials (diffuse and large peaks due to the disorderly
displayed molecules producing disperse bands) [26] show no X-ray
patterns that are typical of crystalline materials (yield a series of
sharp and defined peaks since they are presented in a highly
ordered state) [27]. Analysis of Crystalline state is of great impor-
tance for the stability of powder usually used for juice preparation.
Obtaining powders containing crystalline state sugars is of funda-
mental importance for their stability [26]. The high sugar content
of dried powders, mostly present in amorphous state, are very
hygroscopic, having a stickiness tendency and forming high
agglomerates. This tendency to agglomeration may become accen-
tuated as the amorphous state sugar transforms into crystalline
sugar through adsorption of small amounts of water [26]. The
glassy or amorphous powders are considered as melted liquids,
characterized by the absence of a lucid fusion temperature.
Thereby, the induction of sugar crystallization during drying
through physical methods such ultrasound may make the dehy-
drated powders steady vis-à-vis their functional properties.
The physicochemical quality of strawberry such as color, total
phenolic, antioxidant activity and rehydration capacity can be
affected during drying. The antioxidant activity and total phenolics
fruit contents were investigated by Alonzo-Macıas [28] in swell
and convective drying strawberry fruits, in power ultrasound
assisted convection drying of strawberries by Gamboa-Santos
et al. [29], in acoustically assisted drying treatments kinetics of
pineapple by Rodríguez et al. [12], in osmotic concentration
assisted continuous high frequency ultrasound and dehydration
of rabbiteye blueberries by Stojanovic and Silvain [30] and recently
in immersion and contact ultrasound treatment of the apple tissue
by Artur Wiktor et al. [31]. Although, ultrasound assisted osmotic
dehydration has been applied to strawberry [13] and freeze dried
strawberry [32], however, According to our knowledge, there are
few scientific work has as yet been reported worldwide concerning
the effect of UOD-convective air drying processing on the overall
quality of strawberries.
This study has investigated the use of ultrasound for strawberry
osmotic dehydration. Precisely, a Box-Behnken design (BBD) con-
jugated with Derringer’s desired function methodology was
employed to optimize and study the induction of the key process
variables such as ultrasound time, and concentration and temper-
ature of osmotic solution (independent variables) on WL, WR and
SG (dependent variables). Thus, The specific objectives taken were
(1) to determine the drying, rehydration characteristics and X-ray
diffraction of UOD-convective dried strawberry and to compare
and contrast the quality of these dehydrated strawberries in terms
of their polyphenol compounds, the antioxidant activity, and color
to that convective dried strawberries., (2) to study the effective
moisture diffusivity with and without shrinkage consideration
during the convective drying of strawberry tissue pretreated by
UOD using the phenomenological model of Coupled Washing/Dif-
fusion and to investigate the effect of air drying temperature on
drying time.
2. Materials and methods
2.1. Fresh materials
Fresh strawberries of the same size, color and ripeness were
purchased from a local market in Tunis (Tunisia) and stored in a
288 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
refrigerator at 4 °C for maximum 4 days until use. The strawberries
were removed from refrigeration and left for equilibrium at room
temperature before experimentations. Then, the fruits were
washed and dried with absorbent tissue paper. Following this step,
the strawberries were manually cut parallel to the main axis into
semi-spherical halves of average 3.7 ± 2 cm length, 3.6 ± 2 cm
width and 1.5 ± 0.1 cm thickness with a stainless steel knife and
weighted. The °Brix (11 ± 1) and moisture content of fresh straw-
berry (89 ± 1% w/w moisture content, wet basis) was observed in
the fresh samples.
2.2. Osmotic treatment
A solution of distilled water (0 °Brix) and two different osmotic
solutions were prepared using commercial sugar and distilled
water to give a concentration of 32.5 and 65 °Brix. The osmotic
dehydration was carried out in separate 100 ml Erlenmeyer flasks.
2.3. Ultrasound assisted osmotic dehydration
The halves strawberries were loaded into the Erlenmeyer flask
sealed by plastic film to avoid loss of solvent solution/fruit ratio
was set at 4:1 w/w. The sample flask was immersed into the ultra-
sonic bath (Skymen JP-020S, volume: 3L, China) for irradiation dur-
ing a period of 10 to 30 min [13]. The frequency of the ultrasonic
frequency was 40 kHz and the intensity was 0.2 W/cm2. The ultra-
sound intensity was determined by calorimetric method described
by Löning et al. [33]. The experiments were carried out at 20, 30
and 40 °C in the ultrasound bath. To evaluate the effect of ultra-
sound, the same procedure was repeated in the ultrasonic free
environment (OD). The experiments were carried out in triplicates
and the mean values were used to fit the parameters.
When the immersion time was reached (respectively 10, 20,
and 30 min), sample was withdrawn from the glass beakers and
quickly rinsed with fresh water to eliminate excess solution. Sub-
sequently, it was slightly wiped with an absorbent paper and
weighted using an analytical balance (FA 1004, China) with an
accuracy of ±0.0001 g. Moisture content of the samples was deter-
mined by a gravimetric method, according to AOAC 930.04 [34].
Weighed Sample was placed in a drying oven at 105 °C until
weight stabilization. The samples were cooled down in desiccators
and reweighed. Moisture content and solid content of the samples
were then calculated from the sample weights before and after
drying. The °Brix of osmotic solution was measured simultane-
ously with a digital refractometer (Reichert Leica AR200). The
water loss and solid gain were calculated according to the method
proposed by Amami et al. [25].
2.4. Experimental design and statistical analysis
BBD with three variables at three levels (Table 1) was used to
investigate responses and then to characterize the optimal combi-
nation of variables for UOD process. The three independent
variables used in this study were ultrasound time, and concentra-
tion and temperature of osmotic solution. The dependent variables
were WL, WR and SG. The complete design generated 15
Table 1
The level of actual and coded values used for ultrasound osmotic dehydration process.
Actual Factor Coded Factor
x1 X1
x2 X2
x3 X3
experimental points (Table 2). Experiments were carried out in
random order. Obviously, the same Box–Behnken design and the
same methodology were used to estimate the main effects of OD
without ultrasound treatment (Table 2).
Therefore, the general designs are given in terms of standard-
ized coded variables (Xi), which in any particular application are
linearly related to actual factor (xi) by the following equations
[19–21]:
x1  20 x2  32:5 x3  30
X1 ¼ ; X2 ¼ ; X3 ¼
10 32:5 10
2.5. Optimization
The RSM was applied to the experimental data using the soft-
ware STATISTICA (Trial Version Statsoft. Inc, 2010). The method
of linear regression of the polynomial model with 10 coefficients,
including quadratic and linear effects of factors and linear effect
of interactions, was supposed to depict relationships between
response (Y) and the independent variables (X1, X2 and X3) as given
in Eq. (1).
Y ¼ b0 þ b1 X 1 þ b2 X 2 þ b3 X 3 þ b12 X 1 X 2 þ b13 X 1 X 3 þ b23 X 2 X 3
þ b11 X 21 þ b22 X 22 þ b33 X 23 ð1Þ
where, Y is the response variable of removal efficiency (WL, WR and
SG (%)), b0 is the estimated coefficient of fitted response regression
at the center point, b1, b2 and b3 are linear coefficients, b12, b13 and
b23 are cross product coefficients, b11, b22 and b33 are quadratic coef-
ficients. The b coefficient is the amplitude of these values and gives
ideas about the contribution of each factor in the prediction of the
response. For a positive value of b, the impact of the correspondent
parameter is higher if b is greater, while negative term indicates an
antagonistic effect upon the response.
Numerical optimization using the Derringer’s desirability func-
tion (DF) method was carried out by defining constraint criteria at
maximum and minimum ranges of responses to determine the
optimum values of independent variables. DF distinguishes and
creates a function for each individual response (di). The geometric
mean of individual desires was used to determine the overall desir-
ability function (D) that should be maximum following selection of
optimum value of affective variables with considering their
interaction. This algorithm varies on a scale of 0–1, where 0 repre-
sents a completely undesirable response, and 1 depicts the most
desirable one [21,22].
2.6. Air drying
The amount of sample is two strawberries halves of 30 ± 1 g.
Fresh and UOD pretreated strawberry halves at optimized condi-
tions were placed in petri dishes (flat surface up) in a single-
layer arrangement and dried in a forced circulation air-drying oven
(Binder FD 53, Tuttlingen, Germany). Air temperature in the oven
was set at 40, 50 and 60 °C and an air velocity of 1.0 m/s. Cross-
flow air moved from side to side of the dryer at 1 m/s, flowing par-
allel to the width of the dryer shelves. The amount of water
removed during the drying process was recorded at 5 min intervals
Independent Variables Coded levels of variables
1 0 +1
Ultrasound-Osmotic time (min) 10 20 30
°Brix (%) 0 32.5 65
Temperature (°C) 20 30 40
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300 289

Table 2
Box-Behnken design matrix with experimental values of response variables for ultrasound and without ultrasound osmotic dehydration process.

X1 X2 X3 UOD OD
SG (%) WL (%) WR (%) SG (%) WL (%) WR (%)
10 65 30 0.56 5.714 6.275 0.38 5.230 5.608
30 65 30 0.31 13.106 12.797 0.39 13.12 12.73
20 65 20 0.31 6.539 6.848 0.136 5.768 5.632
20 65 40 0.17 8.102 7.933 0.262 9.319 9.057
10 0 30 0.60 5.167 5.770 0.40 4.642 5.044
30 0 30 0.80 2.600 3.404 0.51 4.470 4.979
20 0 20 0.39 1.318 1.707 0.16 4.054 4.214
20 0 40 1.09 4.408 5.504 0.78 6.394 7.173
20 32.5 30 0.55 8.341 8.897 0.38 10.45 10.83
20 32.5 30 0.57 7.775 8.348 0.36 10.31 10.67
20 32.5 30 0.53 8.246 8.782 0.36 8.888 9.250
10 32.5 20 0.01 3.107 3.119 0.089 4.110 4.021
30 32.5 20 0.14 4.016 4.158 0.500 4.404 3.905
10 32.5 40 1.13 2.323 3.460 0.70 4.941 5.640
30 32.5 40 0.29 8.542 8.253 0.474 9.613 9.139

for the first 30 min then at 30 min intervals to 3 h of drying and where: X, Xt1, and Xeq are the water content dry basis (g H2O/g d.m.)
then every hour by means of an analytical balance (FA 1004, in the solid matrix at time t, at the initial time range chosen for
China), with an accuracy of ±0.0001 g. Drying is stopped when studying diffusion stage t = t1, and at the equilibrium time
the sample weight shows to be fairly constant in time. (t ? 1), respectively.
Approximations to the analytical solution of diffusion model
2.6.1. Determination of bulk shrinkage Fick’s second law can be arbitrarily classified into constant volume
Determination of bulk shrinkage of each strawberry halves was and variable volume models.
achieved by using a slide gauge of 102 mm accuracy. Volumetric 2.6.2.1.1. Approximate model 1: Constant radius (without shrinkage).
shrinkage (Sb) of semi-spherical strawberry was calculated accord- The diffusion model Fick’s second law in spherical coordinates (Eq.
ing to Eq. (2): (6)) with the assumptions of i/ constant internal water diffusivity,
ii/ constant internal temperature, and iii/ negligible shrinkage dur-
Sb ð%Þ ¼ ðV f =V 0 Þ  100 ð2Þ
ing long drying times has been applied to fit drying data within the
considered range (t > t1) [35,36].
where V0 and Vf are the volume of the semispherical strawberry
!
before and after drying, respectively. 6 X1
1 n2 p2 Deff1=2 ðt  t 1 Þ
The shrinkage during thin layer convective air drying of straw- X rt1 ¼ exp  ð6Þ
p n¼1 n2
2 r21=2
berry half is described as a linear relationship between volume (V/
V0) and moisture ratios (X/X0) and calculated using Eq. (3) [7,10]: where Deff1/2 is the effective diffusion coefficient for the semispher-
V=Vo ¼ AðX=X0 Þ þ B ð3Þ ical fruit (m2/s); r1/2 is the radius of the semispherical fresh fruit
(m).
From the afore-mentioned relationship, Eq. (4) was obtained to For long drying times Eq. (6) can be simplified by limiting this
determine the change of the characteristic diffusion dimension expression to its first term. Logarithmic transformation allows
(r1/2) during drying and, afterwards, included in the modeling of implementing a straight-line equation. The effective diffusivity
the drying kinetics to provide a more realistic estimation of the value is typically determined by plotting experimental drying data
effective moisture diffusivity. in terms of ln (Xrt1) versus time [36]. A plot of ln (Xrt1) versus time
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   should give a straight line slope of:
3 X
r 1=2 ¼ A þ B  r 30 ð4Þ p2 Deff1=2
X0 Slope ¼ ð7Þ
r 21=2
where subscript 0 refers to the initial time.
2.6.2.1.2. Approximate model 2: Radius variable with moisture content
(with shrinkage). This model integrates the radius change during
2.6.2. Mass transfers: Coupled Washing/Diffusion CWD model
drying as a function of moisture content. To take the shrinkage
2.6.2.1. Effective moisture diffusivity Deff. The effective moisture dif-
observed during strawberry drying into consideration in Eq. (6),
fusivity Deff should reveal various mass transfer phenomena of liq-
the strawberry half equivalent radius must be account as a
uid water within the solid matrix, which has gradient of water
function of time in Eq. (6), the sample equivalent radius must be
concentration dry basis X as driving force. Deff should be calculated
considered as a function of time (r1/2 = r1/2(t)) and is obtained from
on the base of similar-second Fick’s law from the data gathered
Eq. (4).
from experimental drying results excluding those able to be
Eq. (6) becomes [35]:
inserted in the specific range of initial airflow/superficial interac- !
tion (washing stage t < t1). Crank method used to determine effec- 6 X
1
1 n2 p2 Deff1=2 ðt  t 1 Þ
tive diffusivity implies defining the reduced dimensionless water X rt1 ¼ 2 exp  ð8Þ
p n¼1 n2 r 1=2 ðtÞ2
content Xrt1 ratio relative to the time t1 as a dependent variable
of time [6]. It was calculated for the range of t > t1 as established For long drying times Eq. (8) can be simplified to a straight-line
by Allaf et al. [5]: equation and become:
!
X  X eq 6 p2 Deff1=2 ðt  t1 Þ
X rt1 ¼ ð5Þ X rt1 ¼ exp  ð9Þ
X t1  X eq p2 r 1=2 ðtÞ2
290
The moisture diffusivity is thus estimated using method of
slopes. Its values are generally determined from the slopes of ln
(Xrt1) vs. t/[r1/2(t)]2 straight-line. These values identified for differ-
ent temperatures: 40, 50 and 60 °C for strawberry sample of 14.40–
17.65 mm initial equivalent spherical radius.
The drying data from the drying tests were then expressed as
the dimensionless moisture ratio (Xr) versus drying time or drying
rate versus moisture ratio at different air temperatures.The tem-
perature effect can be assessed by a fitting of calculated Deff1/2
Ea
values to the Arrhenius law (Deff ¼ D1 e RT ) as mentioned by Amami
et al. [11]. The activation energy Ea can be determined from the
plot of ln(Deff) versus 1/T. The slope of the line is (Ea/R) and the
intercept equals ln(D1),where Ea is the activation energy (J/mol),
D1 is the diffusion coefficient corresponding to infinite tempera-
ture (m2/s.), T is the air flow temperature (K) and R is the universal
gas constant (R = 8.3145 J mol1 K1).
In order to analyze the influence of shrinkage on the values of
water diffusivity and activation energy for a non-shrinking and
for a shrinking models were evaluated and compared.
2.6.2.2. Starting accessibility dXs. Since the diffusion model concerns
are only for t > t1, the extrapolated value computed at t = 0 has a
water content Xo with a distinct value from the real initial water
content Xi. The difference is defined as the starting accessibility
dXs expressed as % db or g H2O/g db [6]: dXs = Xi-Xo.
dXs reflects the amount of water available on the surface and
extracted from it in a very short time independently from the dif-
fusion process.
2.6.3. Quality analyses
2.6.3.1. Water activity. The water activity (Aw) was measured
before and after drying experiments. The determination of Aw
value was directly measured using a ‘‘NOVASINA” apparatus (Aw
Sprint TH-500) at 25 °C with the accuracy of 0.001. The fresh/dried
samples were placed in a converter chamber and kept there until
equilibrium was reached. Aw measurements were carried out in
triplicate and the arithmetic mean was calculated.
2.6.3.2. Color analysis. Another important quality parameter was
the color measurement. The color of the fresh and treated (dried
and UOD/dried) strawberry were measured from external and
internal side using a handheld tristimulus colorimeter (Minolta
Chroma Meter CR-300, CIE, 1976) and the color was recorded using
CIE L⁄a⁄b⁄ color spaces. L⁄ indicates lightness, and its value ranges
from 0 (black) to 100 (white), a⁄ is a measure of greenness (a⁄)/
redness (+a⁄) and b⁄ is the grade of blueness (b⁄)/yellowness
(+b⁄). The colorimeter was calibrated against a Minolta standard-
white reflector plate before each actual color measurement. In
addition, for derived color parameters the total color difference
(DE), Chroma or color saturation value (C), hue angle (h°), and
the (DL) were calculated using the following equations [32]:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðL0  L Þ þ ða0  a Þ2 þ ðb0  b Þ
2
Total colordifference ðDEÞ ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Chroma ðC  Þ ¼ ða Þ2 þ ðb Þ
 2
 was then measured by the decrease in absorption at 517 nm using
 b
hue angleðh Þ ¼ arctan
a
DL ¼ L  L0
where L⁄0, a⁄0, b⁄0 are the values for fresh strawberry. Three replicates
were measured and results averaged. The C axis represents Chroma
E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
(or ‘‘saturation”). It ranges from 0, which is completely unsaturated
(neutral grey, black or white) to 100 for very high Chroma (satura-
tion) or ‘‘color purity”. The hue (h) is the angle that represents the
saturated color in the color space. It is represented in degrees,
which ranges from 0° (red), through 90° (yellow), 180° (green),
270° (blue) to 360° (red). The results were presented as DL, DC
and Dh, because each samples had slight differences in color.
2.6.3.3. Determination of rehydration ratio. Rehydration ratio (RR)
was evaluated by soaking one quart of the dried strawberry (about
0.2–0.5 g) [28] in a sufficient volume of distilled water (approxi-
mately 200 times the weight of dried strawberry) in a Stainless
Steel Mesh Infuser in 100 ml flask beaker at 20 °C for 4 h [28].
The samples were then removed after each specific time intervals
(0, 5, 10, 15, 20, 25, 30, 60, 90, 120, 180 and 240 min), drained
for 30 s, and weighed. All measures were done in triplicate. The
water content was expressed as grams of water absorbed per hun-
dred gram dry matter.
2.6.3.4. Phytochemical analysis.
2.6.3.4.1. Sample extraction preparation. After dehydration, each
sample (air dried or UOD-air dried) was ground with a grinder
(AR110510, Moulinex, France) into powders. These powders were
also used for Analysis of X-ray powder diffraction in the following
sections. Subsequently, 1 g of sample aliquots was weighed and
placed in a 100 ml Erlenmeyer flasks then extracted with 20 ml
50% ethanol at 31 °C for 21 min in the ultrasonic cleaning bath
(Skymen JP-020S, China, 40 kHz, power 120 W and volume: 3L).
Afterward, the extract was centrifuged for 10 min (15000 rpm)
for further extraction. Subsequently, the supernatant was filtered
using a medium-speed filter under a vacuum at room temperature,
and stored at 4 °C to be used within 24 h.
2.6.3.4.2. Determination of total phenolic content (TPC). The total
phenols content was determined according to the Folin-Ciocalteu
method [37] with slight modifications. Briefly, 100 lL of filtered
ethanolic extract added with 100 lL of EtOH (50%) and 100 ll of
Folin’s solution (2 N, Sigma) were incubated for 2 min then 2 ml
of Na2CO3 (75 g/l) is added. The mixtures were vortexed briefly
and left in the dark for 30 min at room temperature. The absor-
bance of the solutions was measured using UV–Visible spectropho-
tometer at a wavelength of 720 nm against sample without extract.
The total phenolic content was calculated based on the calibration
curve build using increasing concentrations (10–500 lg/ml) of Gal-
lic acid described by the equation y = 0.0033x + 0.0368 (R2 = 0.99).
The results were expressed as milligrams of Gallic Acid Equivalent
(mg GAE/mg of dry matter (dm)). All measurements were done in
triplicate for each extract.
2.6.3.4.3. Determination of antioxidant activity (DPPH Assay). The
antioxidant activity of the ethanol extracts was evaluated spec-
trophotometrically by the modified DPPH method [38] which is
  2 based on the quantification of free radical-scavenging. An ethanolic
solution (7.3 mg DPPH in 180 mL ethanol) was prepared. After
ð10Þ adjusting the blank with ethanol, an aliquot of 15 ll of dried straw-
berry extract was added to 1500 ll of a (100 mM) DPPH solution. A
ð11Þ volume of the mixture was vigorously shaken and left to stand in
the dark at room temperature for 30 min. The antioxidant activity
ð12Þ UV–Visible spectrophotometer against 50% ethanol in DPPH solu-
tion. The percent inhibition of the DPPH radical was calculated
thus:
ð13Þ
ðAc  AsÞ
Ið%Þ ¼  100 ð14Þ
Ac
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
where I%: Percentage of the anti-radical activity; Ac and As are the
absorbance values of the control and test sample, respectively. For
each sample, three replications were done.
2.6.3.4.4. Analysis of X-ray powder diffraction. The X-ray diffraction
patterns of air-dried, powdered strawberry were used to monitor
the crystallization process with an X-ray diffractometer equipped
with a Cu-Ka source at 1.540 Å (XRD, Bruker-AXS D8 Advanced
Bragg-Brentano X-ray Powder Diffractometer, GMBH (Germany))
with an input energy of 40 kV and 30 mA. The powder was placed
and slightly pressed on an aluminum sample tray using a glass
slide and scanned at a diffraction angle (2h) between 0° and 60°,
with a stepped increase of 0.02° 2h and 0.5 s/step at room temper-
ature. The XRD patterns of all the samples were plotted using Win-
PLOTR 2015 software for comparison [39].
3. Results and discussion
3.1. Effect of ultrasound assisted osmotic dehydration on WL and SG of
strawberry
The non-parametric Wilcoxon Matched Pairs Test at a 95% level
of confidence was carried out to compare the results of the osmotic
dehydration experiments with and without the application of
ultrasound. The results showed a considerable difference signified
by a positive significant Marked tests (2215 < Z < 3407) was
observed between these two treatments considering SG
(p = 0.00065), WL (p = 0.0076) and WR (p = 0.0267).
3.2. Statistical analysis
The multiple regressions testing was carried out on trial data.
The quadratic model in coded unit shows the role of every variable
and their interactions in independent variables. The mathematical
equation in terms of coded factors for ultrasound assisted osmotic
dehydration of strawberry on SG, WL and WR model are shown in
Eqs. (15)–(17).
SG ¼ 0:555 þ 0:245X 1 þ 0:312X 2  0:115X 3 þ 0:267X 1 X 2
þ 0:389X 1 X 3 þ 0:296X 2 X 3 þ 0:148X 21  0:007X 22
þ 0:156X 23
WL ¼ 8:12 þ 1:49X 1 þ 2:496X 2 þ 1:049X 3 þ 2:489X 1 X 2
þ 1:327X 1 X 3  0:381X 2 X 3  1:034X 21  0:439X 22
 2:589X 23
WR ¼ 8:675 þ 1:248X 1 þ 2:183X 2 þ 1:164X 3 þ 2:222X 1 X 2
þ 0:938X 1 X 3 þ 0:296X 2 X 3  0:678X 21 þ 0:431X 22
 2:745X 23
The sufficiency of the model was evaluated through analysis of
variance (ANOVA). Moreover, the variation of the data around the
fitted model designed by the Lack of Fit was also checked. In the
present study, with regards to WL and WR, the Lack of Fit is not sig-
nificant relative to the pure error, indicating good response to the
model. Only the fitted SG model exhibits lack of fit (p = 0.012 for
both UOD and OD) at the confidence level indicates that there is
still some statistically significant variability left that cannot be
accounted by the factors and their interactions. ANOVA for SG,
WL and WR, respectively indicated that the second-order polyno-
mial model (Eq. (1)) was adequate to represent the actual relation-
ship between the response and the variables, with a high value of
coefficient of determination (R2 = 0. 9714 for SG, R2 = 0. 989 for WL
and R2 = 0.9897 for WR). This means that the model did not explain
291
only about 2.86 and 1.03% of sample variation for SG and WL,
respectively. The value of the adjusted determination coefficient
(adjusted R2 = 0. 9201 for SG, adjusted R2 = 0. 969 for WL and
adjusted R2 = 0. 9713 for WR) was also high to indicate a high sig-
nificance of the model.
Fig. 1 shows the closeness of the observed SG of ultrasound-
assisted OD of strawberry versus those from the empirical model
Eq. (15). Point above or below the diagonal line represented areas
of over or under prediction. It is apparent from these figures that
the predicted data closely banding around the straight line, which
proves that no significant violations of the model were found in the
analysis in the range of the operating variables.
All linear terms have significant effect at 5% level of significance
except for the process temperature no significantly affect SG. The
main effects of UOD time and °Brix on all independent variables
were highly significant as was evident from their respective P-
values (0.00005 < p < 0.00285). When the concentration of sugar
in the osmotic solution is increased, WL rate (p = 0.00005) is more
affected than SG (p = 0.0009), promote an important apparent
selectivity of transfer [40]. The relative magnitude of b values
(Eq. (16)) indicates the maximum positive effect of osmotic solu-
tion concentration (b = 2.496) followed by process duration
(b = 1.49) and temperature (b = 1.049) on water loss.
The Response surface and contour plots were generated as a
function of two independent variables at a time with the third
being maintained at the central level are more helpful in. The
three-dimensional surface curves were plotted to understand both
the main and the interactive effects of these two factors and pro-
vide useful information about the behavior of the system within
the experimental design. The response surface plots for water loss
(WL) are shown in Fig. 2, for significant factor interaction resulted
from the ANOVA.
Fig. 2(A) shows the increased water loss with increase in osmo-
tic solution concentration and UOD time. This is mainly because of
the increase in the osmotic pressure gradient. A similar variation in
water loss with UOD time and temperature has also been observed
in Fig. 2(B). This might be because of the decreases of the viscosity
of the osmotic solution and thus reduces the external resistance to
mass transfer at product surface and yield better water transfer
characteristics [41].
ð15Þ Only the quadratic term of UOD time and temperature has sig-
nificant effect on WL and WR at 5% level. The temperature has sig-
nificant negative effect on WL and WR in quadratic term. The result
also indicated that temperature could act as limiting factors, and
increase in their values cause changes in the structure of cell mem-
ð16Þ branes, resulting in loss of selectivity with an increase of the
0,6
0,3
ð17Þ
Predicted Values
0,0
-0,3
-0,6
-0,9
-1,2
-1,2 -0,9 -0,6 -0,3 0,0 0,3 0,6
Observed Values
Fig. 1. The predicted vs. observed values of SG of ultrasound-assisted OD of
strawberry using RSM.
292 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
Fig. 2. Influence of process variables on water loss (A) UOD time and °Brix for 30 °C of temperature process (B) UOD time and temperature at 32.5°Brix of osmotic solution
concentration.
permeability of the cell wall, leading to greater solute uptake. Thus
explain the significant positive effect of temperature at 7% level of
significance on the SG. Also the interaction between °Brix and
Temperature had significant effects (PX2X3 = 0.00563).
The result also could be explained by the frequency histogram
Pareto chart of effects (Fig. 3). Pareto analysis is a technique that
helps to prioritize and focus resources visually. It shows the influ-
ence of each factor on the response in decreasing order. As it can be
seen (Fig. 3 (A)), the most important linear effect for SG were the
°Brix followed by the UOD time. The efficiency of SG was also
affected significantly by all factor interactions such as UOD
time  Temperature and UOD time  °Brix. Nevertheless, an excess
of °Brix producing negative influence on the SG but is it statically
non-significant.
A Pareto chart of the main effects for SG of experiments carried
out without application of ultrasound is shown in Fig. 3(B). It was
observed the same tendency as obviously and the most important
linear effect for SG were the °Brix followed by the OD time, mean-
while with slightly inferior coefficients. This difference may be
explained by the effect of ultrasound treatment. The use of ultra-
sound also decreases the amount of soluble solids, especially sugar,
of the fruit and produces a dried low-sugar fruit. Same tendency
have been shown in the ultrasound-assisted osmotic dehydration
of pineapple and papaya [42]. All the results mentioned above
are in agreement with many study which state that by using highly
concentrated solutions (60–70% by weight), a considerable product
weight loss can be achieved along with a low solute gain [40].
3.3. Optimization and validation of the models
The choice of the optimum process conditions for osmotic dehy-
dration depends on the application that would be given to the
product. During osmotic dehydration, it is essential to achieve high
(A)
MS Residual=,0162974
X2 6,921
X1*X3 6,095
X1 5,440
X2*X3 4,639
X1*X2 4,191
X3 -2,554
X3^3 2,356
X1^2 2,232
X2^2 -,118
p=,05
Standardized Effect Estimate (Absolute Value)
Fig. 3. The Pareto chart of standardized effect of the main effects for SG: (A) of UOD strawberry and (B) of OD strawberry.
rates of water loss and, depending on the final product application;
it would be also desirable to minimize the uptake of solutes. The
aimed of our study was the use of strawberry for production of diet
foods for diabetic, for thus it is essential to have a low incorpora-
tion of sugar, for the maintenance of nutritional and sensory char-
acteristics of fresh fruit.
The Derringer’s desirability function method was employed to
optimize the process variables, which would maximize WL, WR
and minimize SG. According to the software optimization step,
the desired goal for each operational condition (UOD time, solution
concentration and temperature) was chosen ‘‘within the range’’,
while the responses WL and WR was defined as ‘‘maximum’’ and
the response SG was defined as ‘‘minimum’’. The second-order
polynomial models obtained in this study were utilized for each
response in order to obtain specified optimum conditions. As
shown in Fig. 4 the optimum conditions were found to be UOD
time of 20.5 min, osmotic solution concentration of 47.5 °Brix,
and osmotic temperature of 31 °C, respectively with overall desir-
ability value of 0.62. At this optimum conditions, WL, WR, and SG
were found to be 9.38%, 9.77%, and 0.387%. The optimized water
loss observed for strawberry in this study was very higher than the
water loss observed for other strawberry study [13], which have
lost only 2.3% weight basis under approximatively the same condi-
tions; 40 kHz ultrasonic waves immersed in an osmotic solution of
50° Brix, also at 30 °C during 20 min.
In order to verify the predictive capacity of the response surface
models, three experiments were run under the recommended opti-
mum conditions. The observed experimental values for the WL,
WR, and SG were found to be 9.80%, 10.16%, and 0.36%, respec-
tively. The difference between experimental and predicted was less
than 7% of deviation, thus indicating the adequacy of models fitted
by RSM and confirm the predictability of SG and WL in the exper-
imental condition used.
(B)
MS Residual=,0282162
X2 4,75
X1 4,72
X3^2 3,13
X3 -2,74
X1*X2 2,60
X1*X3 2,27
X2*X3 2,21
X1^2 2,11
X2^2 -,47
p=,05
Standardized Effect Estimate (Absolute Value)
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
UOD time (min)
,80000
-,3871
-1,600
16,000
9,3834
-2,000
16,000
9,7705
-2,000
,62097
10, 20,5 30, 0,
Fig. 4. Desirability charts of variables for maximum WL and WR and minimal SG response for UOD strawberry.
For the osmotic dehydration experiments without ultrasound
the Derringer’s desirability function method was used at optimum
conditions to determine the WL, WR and SG parameters. This
parameters reach the respective value of 10.69% (WL), 10.92
(WR) and 0.23% (SG). The high obtained WL for OD can be
explained as reported that the effect of the ultrasound-assisted
osmotic dehydration on the water diffusivity depended on the
degree of breakdown of cells in the fruit tissue [17].
Osmotic dehydration can’t obtain totally dried products and is
usually applied as pretreatment for other drying process. Further
study, was done to investigate the effect of ultrasound assisted
osmotic dehydration carried out under the optimized operating
conditions followed by hot-air drying process as a pretreatment
for drying on the mass transfer properties (moisture content, water
activity, drying rate, water accessibility, diffusivity and activate
energy with and without shrinkage consideration) and on quality
properties (rehydration ratio, color, total phenolic content, antiox-
idant activity and X-ray diffraction powder).
3.4. Drying operation
3.4.1. Shrinkage
Table 3 shows the constants A and B of linear function (Eq. (3))
of the volume change at different temperatures and pretreatment
with a coefficient of determination (R2) which presented higher
values (>0.99) for all the conditions. This equation is used to esti-
mate the particle radius of the strawberry in the numerical solu-
tion described previously. According to the literature, similar
values have been reported describing the shrinkage in strawberry
and other fruit. Thus, the slope values (0.628–0.676) of this rela-
tionship for strawberry air dried at temperatures of 40–60 °C was
in the range previously reported by Piotrowski et al. [43] for straw-
berry convectively dried at temperatures of 45–65 °C (0.62–0.68),
by Gamboa-Santos et al. [10] for strawberry convectively dried at
temperatures of 70 °C and 2 m/s (0.692), by Ramallo and Mascher-
oni [44] for pineapple samples dried at temperatures of between
293
∞Brix (%) Temperature (∞C) Desirability
,30883
0,
SG (%)
,5
-1,137
1,
13,106
1,
WL (%)
,5
1,3176
0,
12,797
1,
WR (%)
,5
1,7068
0,
Desirability
47,5 65, 20, 31 40,
45 and 75 °C (0.652–0.785) and recently reported by Dehghannya
et al. [7] for ultrasound-assisted osmotic dehydration pretreatment
of the drying Mirabelle plum at 80 °C and 1.4 m/s (0.640–0.764).
The increase of air temperature increases the amount of shrink-
age. The shrinkage of food materials increases with the volume of
water removed, since the more the water removed the more con-
traction stresses are originated in the material. Table 3 indicates
an increase in shrinkage with UOD pretreatment. This increase
may be explained by the low osmotic solution concentration
(47.5%) used and essentially the less immersion time in osmotic
solution (20.5 min). This result agree with thus of Dehghannya
et al. [7] who advanced that shrinkage of pretreated plum samples
at constant ultrasonication time and osmotic solution concentra-
tion, increasing immersion time from 60 to 240 min decreased
the shrinkage. With respect to the solution concentration, a smaller
moisture content, and a consequently higher shrinkage were
observed for samples with lower osmotic solution concentrations.
The Mirabelle plums treated with ultrasound for 30 min and dehy-
drated at osmotic solution concentration of 70% for 240 min, prior
to drying, were found to have the lowest shrinkage (64.1%) com-
pared to control (76.4%) [7].
3.4.2. CWD model and activation energy
3.4.2.1. The results of effective diffusivity. Moisture diffusivities were
evaluated from the slopes of ln[(X-Xeq)/(Xt1-Xeq)] vs. ratio time/[r
(t)]2 fitted straight-line (as shown in Fig. 5 for UOD-air dried straw-
berry) and for both constant sample equivalent radius. This linear
data fitting have a good linear fitting with a determination coeffi-
cient R2 > 0.995.
Identified diffusivities of fresh and UOD-air dried samples were
summarized in Table 4. Equally shrinkage, temperature and pre-
treatment influences on their values were illustrated.
The effective diffusivities of UOD-air dried samples were higher
than fresh-air dried samples. In addition, the higher the tempera-
ture, the higher the Deff values. Indeed, increasing temperature
normally greatly activates water molecules, speeding up the water
294 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300

Table 3
Constants of Equation AX + B (Eq.3), which describes the Volume Change during Drying of strawberry at Different Temperatures and pretreatment.

Fresh-AD 40 °C Fresh-AD 50 °C Fresh-AD 60 °C UOD-AD 40 °C UOD-AD 50 °C UOD-AD 60 °C

r1/2i (cm) 1.633 1.439 1.483 1.764 1.498 1.62
r1/2eq (cm) 1.174 1.004 1.018 1.151 0.962 0.980
V0 (cm3) 9.127 6.248 6.840 11.500 7.041 8.924
Veq (cm3) 3.394 2.114 2.214 3.193 1.864 1.965
A 0.628 0.661 0.676 0.722 0.735 0.780
B 0.372 0.338 0.323 0.277 0.265 0.220

transfer from the core to the surface of the matrix. The effective
diffusivities varied from 8.42  1010 to 23.27  1010 m2/s for
fresh-air dried samples and from 12.41  1010 to
25.96  1010 m2/s for UOD-air dried samples over the tempera-
ture range 40–60 °C for non-corrected drying for shrinkage. These
diffusivities values were in the same range as those obtained by
García-Noguera et al. [13] during the drying (60 °C, 0.5 m/s) of
UOD strawberry halves (50 °Brix and 20 min immersion) when
the shrinkage was not considered. They obtained Moisture diffu-
sivity values of 33.16  1010 m2/s.
Values of Deff were somewhat smaller when taking shrinkage
into consideration. This difference increased with drying tempera-
ture, from 38% lower at 40 °C until 49% at 60 °C. These diffusivities
varied from 5.05  1010 to 12.20  1010 m2/s for fresh-air dried
samples and from 7.66  1010 to 13.20  1010 m2/s for pre-
treated samples over the temperature range 40–60 °C. Effective
diffusivity coefficients considering shrinkage but not varying with
moisture content were reported by Simal et al. [45], who obtained
diffusivities of around 4.7  10–10 to 1  10–9 m2s–1 for grapes [46].
These diffusivities values were in the same range as those
obtained by Gamboa-Santos et al. [10] during the drying of straw-
berry cubes of 8.5 mm when considering the phenomenon of
shrinkage, moreover the external resistance (ER) to mass transfer
and diffusion equation of nonlinear character. They obtained Mois-
ture diffusivity values of 0.763–1.305  1010 m2/s over the same
temperature range of 40–60 °C.
3.4.2.2. Comparison between prediction models. Evaluation of diffu-
sional model (Eq. (9) with shrinkage) for moisture prediction can
be wrought comparing the experimental dimensionless moisture
Fig. 5. The natural logarithm of the dimensionless moisture versus time to square
equivalent radius ratio of UOD pretreated half strawberry dried at 40 (square), 50
(triangle) and 60 °C (circle).
values with their predicted values. The experimental and predicted
Xr values for UOD-air dried strawberry at 40, 50 and 60 °C lay
around the straight line (Fig. 6). The coefficient of correlation (R2)
between the experimental and predicted value of the model is usu-
ally >99.4%. There is a better accordance between experimental
and predicted Xr values obtained by the model 2 with shrinkage
consideration. This favorable effect is more important for high
temperature of drying.
3.4.2.3. Activation energy. Activation energy estimated from the
Arrhenius type equation is the amount of energy required for
removing moisture from the materials. The graphic representation
for the different assays carried out is presented in Fig. 7 with Linear
correlation coefficients higher than 0.979. The activation energy
values without shrinkage correction were found to be 44.4 ± 1
and 32.02 ± 2 kJ/mol for fresh-air dried and UOD-air dried straw-
berry halves, respectively. The values obtained was within the
range with reported values by other researches [11,13,36,46]. The
identified Ea values with shrinkage correction decreased to
38.5 ± 1.1 and 23.47 ± 0.8 kJ/mol for fresh-air dried and UOD-air
dried strawberry halves, respectively.
3.4.2.4. Starting accessibility dxs. Fig. 8 represents a typical kinetics
of airflow drying at 60 °C from fresh-air dried (open circles) and
UOD-air dried (full symbols) strawberry. As seen the extrapolation
of the diffusion model can be performed to determine Xo as the
theoretical value of X at t = 0. The experimental value of initial
water content Xi and the extrapolated calculated water content
Xo at t = 0 from the diffusion model corresponds to the starting
accessibility dry basis dXs = Xi  Xo. It represents that the water
quickly removed from the surface independently from the diffu-
sion process.
The starting accessibility for the control sample (Fresh-air dried
strawberry) in our study was in the range of the value 0.0030 g wa-
ter/g d.b found by Alonzo-Macías [28] for the control Hot Air Dried
strawberry at 65 °C and 1.2 m/s.
As presented in Table 4, after UOD pretreatment, the starting
accessibility was systematically greater to 1.65 times more than
fresh-air dried sample at 60 °C (CWD model with shrinkage). These
increases indicate that the acoustic energy contributes to decrease
the internal mass transfer resistance and it could be linked to the
alternative compressions and decompressions produced by the
ultrasound waves at matrix surface, which could make easier the
removal of superficial water [47].
As expected and mentioned by Allaf et al. [5] the higher the
temperature, the higher the starting accessibility. The starting
accessibility was 1.85 and 2.85 higher at 60 °C then 40 °C for the
fresh-air dried and UOD-air dried strawberry, respectively.
Our objectives were to identify the strict conditions of adopting
hot air drying HAD as diffusion-controlled operation. The values
obtained with these experiments have been very interesting. The
fact that the initial accessibility is very low shows that we are
not at all far from a complete diffusion process over the operation.
In this study the Airflow Velocity of 1 m/s can be assumed very
near of the value of Critical Airflow Velocity CAV defined by Allaf
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300 295

Table 4
Modeling of drying kinetics of air dried and UOD-air dried strawberry. Identified Parameters: effective diffusivity and starting accessibility with shrinkage and drying
temperature.

CWD model without shrinkage CWD model with shrinkage

Starting accessibility dXs (% db) Deff0 (1010 m2/s) Starting accessibility dXs (% db) DeffS (1010 m2 s1) Relative error E (%) (Deff0  DeffS)/Deff0
Fresh-AD 40 °C 0.163 8.42 0.565 5.05 40.04
Fresh-AD 50 °C 0.231 14.82 0.803 8.58 42.09
Fresh-AD 60 °C 0.361 23.27 0.998 12.20 47.37
UOD-AD 40 °C 0.201 12.41 0.541 7.66 38.30
UOD-AD 50 °C 0.258 18.26 0.978 9.75 46.62
UOD-AD 60 °C 0.957 25.96 1.614 13.20 49.29

Fig. 6. Experimental (Xrexp) vs. calculated with CWD model with shrinkage (Xrcalc) dimensionless moisture ratio of UOD pretreated half strawberry dried at 40 (square), 50
(triangle) and 60 °C (circle).

1/T (k-1)
0.00295 0.003 0.00305 0.0031 0.00315 0.0032 0.00325
-19.6
-19.8
-20
y = -5341.3x - 3.8014
Ln (Deff)

-20.2 R² = 0.9791
y = -3851.3x - 8.207
R² = 0.9999
-20.4
-20.6
-20.8
-21 y = -2823.6x - 11.986
R² = 0.9936
-21.2
-21.4 y = -4630.7x - 6.5955
-21.6 R² = 0.9908

Fig. 7. Identification of the Arrhenius’ equation parameters for air dried (circles) and UOD-air dried strawberry (triangles); Discontinuous lines (– – –) Deff from the CWD
model without shrinkage; Continuous line (—) Deff from the CWD model with shrinkage.

et al. [5]. As long as hot air velocity is lower than CAV, drying kinet- 3.4.3. Effect of temperature and UOD on the drying kinetics
ics depends on airflow velocity and drying operation cannot be a Fig. 9 presents the kinetic curves Xr(t), obtained for the fresh
diffusion controlled process. In their study for untreated carrot, and UOD pretreated strawberry tissue, dried at temperatures 40,
CAV was 1 m/s. 50 and 60 °C. The experimental results showed that the drying
296 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300

Dimensionless moisture rao,Xr

0,8

0,6

0,4

0,2

0
0 500 1000 1500 2000
Time (min)

Fig. 8. Typical kinetics of Airflow drying at 60 °C from untreated (open circles) and 1
UOD pre-treated (full symbols) strawberry: experimental data and diffusion model

Dimensionless moisture rao,Xr

used to calculate effective diffusivity Deff and starting accessibility dWs = Wi  Wo.
(—) represents the diffusion model and discontinuous lines (– – –) represent the 0,8
extrapolation of diffusion model.

air temperature enhances considerably the drying kinetics for both
fresh and UOD pretreated strawberry tissue. The UOD pre-
treatment enhances further the drying kinetics. In this study the
liquid water diffusivity was considered constant independently
on water content value. It would be possible that, for this reason
the CWD model without shrinkage describe more the experimental
results (smaller diffusivity normally linked to more compact struc-
ture should be compensated by smaller radius).
Fig. 10 shows the drying curves presented in coordinates dXr/
dt versus Xr for the same experimental data. Drying rate decreases
continuously with decreasing moisture content. As expected from
Fig. 10 increasing the air temperature and UOD pretreatment
increased the drying rate, and consequently decreased drying time.
In the experimental and the CWD model without shrinkage curves,
the constant rate period was not visible or completely absent.
Thus, falling rate period drying occurred for strawberry samples
for both conventional air drying and UOD-air drying. Such behavior
agrees with the observations of many studies [9,11,13,36]. In order
to distinguish correctly between the different drying periods, the
drying rate has been calculated with the CWD model with shrink-
age consideration. The plots obtained with the current radius plot
exhibits a very short rising temperature period followed by a short
constant-rate drying period and a long falling-rate drying period
(diffusion phase) from which diffusivity was estimated. The most
accurate values of drying rate were obtained with moisture con-
tent estimated by Fick’s– type diffusion with shrinkage (Model 2).
3.4.4. Optimal conditions
For an osmotic dehydration process followed by hot air drying
as suggested by Garcia-Noguera et al. [13] based in previously
studies, the optimal processing time results when ultrasound
assisted osmotic dehydration allows increasing the drying rate of
fruit.
Table 5 shows the values of total processing (pretreatment + air
drying) time to achieve 90% water content removal in strawberry
halves. These values were compared to results of Garcia-Noguera
et al. [13]. In confirmation of their study it was observed that
strawberry submitted to ultrasound-assisted osmotic dehydration
dries in less time during the air-drying stage than fresh fruit due to
changes in intercellular tissue as a consequence of the combined
effects of osmotic pressure and fatigue induced by ultrasonic
waves [13]. Ultrasound assisted osmotic dehydration during
0,6
0,4
0,2
0
0 500 1000 1500 2000
Time (min)
Fig. 9. Xr during the drying of untreated (open symbols) and UOD pre-treated (full
symbols) strawberry samples subjected to air temperature of 40 °C (diamonds);
50 °C (triangles); 60 °C (circles); Discontinuous lines (– – –) represent the CWD
model without shrinkage; Continuous line (—) represents the CWD model with
shrinkage.
pretreatment generally reduced the total processing time. The
difference in total time from the two works might be linked to
the different product characteristics (initial moisture content and
°Brix), to the low drying speed used (only 0.5 m/s) and to the for-
mation of a small sucrose layer on the surface of the strawberries
due to positive SG in their cases.
The percentage of time reduction compared to fresh-air dried
strawberries (total processing time) showed a decrease with tem-
perature increase. Many previous works [10,13] have highlighted a
similar tendency for ultrasound effect with temperature; it was
observed that ultrasound application increased the drying rate.
However, as the drying temperature increases the influence of
the ultrasound application on the drying rate decreases.
Ultrasound-assisted osmotic dehydration of strawberry halves
at 40 kHz for 20.5 min in a 47.5% sucrose solution needed the
shortest total processing time of 401 min to achieve 90% reduction
in moisture content. When compared to fresh-air dried berries
(450 min of processing time), these times resulted in a 49 min
reduction (10.88%) in total processing time.
3.4.5. Assessment quality
3.4.5.1. Water activity (Aw). The results of the water activity are
presented in Table 6. After both drying method applied, the final
water activity of the dried fruits obtained ranged between 0.452
and 0.481. These values show that a microbiologically stable pro-
duct (Aw < 0.6) for each dried product was achieved independently
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300 297

0,006 Table 6
Results of water activity and phytochemicals analysis of fresh and dried strawberry.

0,005 Water TPC Anti-radical activity

activity (mg GAE/g DM) (I %)
0,004 Fresh 0.915 27.77 95.00
-dXr/dt (min-1)

Fresh-AD 40 °C 0.474 13.41 67.95

0,003 Fresh-AD 50 °C 0.464 17.24 76.19
Fresh-AD 60 °C 0.466 14.80 53.29
UOD-AD 40 °C 0.452 17.31 65.66
0,002 UOD-AD 50 °C 0.479 14.99 75.27
UOD-AD 60 °C 0.481 13.34 67.03
0,001

0
The color comparisons of the dried strawberries produced by
0 0,2 0,4 0,6 0,8 1
UOD-air drying and by conventional air drying are shown in
Dimensionless moisture rao, Xr
Table 7. Drying had a major influence on color as all values (L⁄,
a⁄, b⁄) were affected. It is evident that Chroma of dehydrated mate-
0,006 rials is not strongly affected by air temperature. Chroma changes
are more intense during conventional than during UOD-drying at
0,005 the same temperature, indicating less saturation and a pale
appearance. In general evaluation, the color of fruits was found
to be better in UOD pretreatment in terms of brightness and red-
0,004
-dXr/dt (min-1)

ness. Only at 60 °C the UOD-air dried strawberries were harder

0,003
0,002
0,001
0 As seen in Table 7, the decrease of the hue (Dh) of internal side
0 0,2 0,4 0,6 0,8
Dimensionless moisture rao, Xr
Fig. 10. Drying rates curves during drying of untreated (open symbols) and UOD
pre-treated (full symbols) strawberry subjected to air temperature of 40 °C
(diamonds); 50 °C (triangles); 60 °C (circles). Discontinuous lines (– – –) represent
the CWD model without shrinkage; Continuous line (—) represents the CWD model
with shrinkage.
from the drying conditions used. However, the lowest value was
achieved after convective drying at 40 °C with UOD pretreatment.
3.4.5.2. Color assessment. The chromatic parameters L⁄, a⁄, and b⁄ of
internal side of fresh strawberry were 70.2, 13.47 and 2.24, respec-
tively, while they were 69.15, 9.12 and 2.2, respectively for exter-
nal side. The L, C and h mean values for fresh strawberry were
70.44, 9.89 and 13.50, respectively. In comparison to the fresh fruit,
the dried strawberry had higher L⁄ (lightness) and lower +a⁄ (red-
ness) and b⁄ (blueness) values. Moreno et al. [48] observed that
OD increased the illuminance of pretreated strawberry compared
with the fresh fruit.
than the fresh air-dried strawberries. Similar tendency were
observed by Garcia-Noguera et al. [32], which showed that hue val-
ues decreased and the parameter L⁄ (lightness) have increased dur-
ing the osmosonication processes of strawberry. They explained
that by the increases of Anthocyanins color (responsible of red
color of strawberry upon the removal of water by displacement
of the hydration/dehydration equilibrium toward colored species
[32].
1 was found significant and pre-treatments demonstrated effective-
ness on saving the redness except at 60 °C. Therefore, it can be con-
cluded that the decrease of anthocyanin in fleshy part of fruit was
significantly affected by the drying process. Similarly, the drying
process affected the decrease of the external red color, but the
effect of UOD pretreatments was found to be more significant.
The brightness (DL) of the dried fruits increased both on external
and internal sides and they were differently affected by UOD pre-
treatments. It is shown that the increase in internal side brightness
was higher than in external side brightness. In terms of pretreat-
ments, it can be seen that UOD-AD 50 °C saved the brightness of
both side color higher than UOD-AD 60 °C. UOD-air dried samples
had the lowest difference chrome, indicating more saturation and
an appearance that near to the vivid color of fresh strawberries.
By reducing drying time, UOD minimized color losses during con-
vective air-drying.
Color change between raw material and dried samples was
mainly due to the differences in chrome, hue and lightness proper-
ties. Color changes may be attributed to thermal degradation of
anthocyanins during strawberry dehydration [48]. The total color
change (DE) of the dried samples varied from 8.75 to 11.66 units

Table 5
Total processing time (pretreatment + forced convective oven drying) to achieve 90% moisture content reduction at different conditions.

Treatment Pretreatment time Air drying time (min) Total processing time % time reduction compared to fresh, untreated
(min) (min) strawberries (total processing time)
AD 40 °C 0 1200 1200
AD 50 °C 0 540 540
AD 60 °C 0 450 450
AD 60 °C [13] 0 612 612
UOD (40 KHZ, 47.5 °Brix)+ AD 40 °C 21 840 861 (28.25%)
UOD (40 KHZ, 47.5 °Brix)+ AD 50 °C 21 420 441 (18.33%)
UOD (40 KHZ, 47.5 °Brix)+ AD 60 °C 21 380 401 (10.88%)
UOD (40 KHZ, 50 °Brix)+ AD 60 °C [13] 20 474 494 (19.28%)
298 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300

Table 7
Color change (DL, DC, Dh and DE) of dried and UOD-dried strawberries as a function of air temperature processing.

Treatment Internal side External side

DL DC Dh DE DL DC Dh DE
Fresh-AD 40 °C 3.45 6.12 30.93 8.75 6.60 3.25 25.38 5.98
Fresh-AD 50 °C 7.59 6.28 31.44 11.66 7.69 3.26 24.95 7.58
Fresh-AD 60 °C 5.21 6.55 32.91 8.97 3.01 4.13 20.42 6.32
UOD-AD 40 °C 4.93 6.48 19.71 8.83 6.55 2.14 12.89 7.66
UOD-AD 50 °C 8.48 4.22 20.79 11.93 7.66 2.95 10.94 7.38
UOD-AD 60 °C 5.98 4.14 28.85 12.73 8.17 1.35 9.68 10.21

and from 5.98 to 7.58 in internal and external side, respectively
depending on applied drying temperature. Except at 60 °C only
minor color difference could be observed between fresh-air dried
and UOD-air dried samples. For UOD-air dried samples at 60 °C
DE values among greater than fresh-air dried, which indicated that
samples were turning darker than the fresh.
3.4.5.3. Total phenol content and antioxidant activity of ethanolic
extract of dried strawberry. The initial phenolic content was
27.77 ± 0.16 mg Gallic acid g1 dry strawberry sample. Levels of
phenols for unprocessed strawberry is in the range of those
reported by other authors [49,50].
The present experiments showed that temperature and UOD
pretreatment had an important effect on the total phenolic content
compared to the fresh sample. During drying, TPC levels were gen-
erally decreased. We observed losses of 51.70%, 37.91% and 46.68%
at 40, 50 and 60 °C, respectively, as compared to fresh sample. The
longer processing time in 40 °C associated higher processing tem-
perature in 60 °C as compared to 50 °C contribute to diminish the
protective effect against oxidative damage to cells. This is in line
with results reported by Alonzo-Macías [28] they reported a
decrease in phenolic content during air drying (16.8 mg Gal-
lic acid g1 dry strawberries at 50 °C). For UOD- air dried straw-
berry differences in phenolic stability occurred between air
temperatures, being higher at higher temperatures, they lost
37.65%, 46.01% and 51.94% of TPC over the same range of temper-
ature, respectively. The highest total phenolic content (TPC) was
recorded in plants dried by AD 50 °C and UOD-AD 40 °C and was
38% low than those of fresh plants whereas the lowest content
was found in the case of plants dried by UOD-AD 60 °C (51.94%
low than fresh strawberry). The increase in total phenolic content
for UOD-AD 40 °C may be related to an increased extractability of
some of the antioxidant components following ultrasound process-
ing who can give rise to pores in the vegetal tissue and, conse-
quently, improve the extraction of polyphenols during sample
preparation. However highest shrinkage related to negative quality
characterized by increased hardness, biochemical reactions and
brown appearance can explained values for UOD-AD 60 °C [50].
The DPPH scavenging activity of fresh fruit was determined to
be 95.03 ± 1.18% at 1000 lg/mL concentration. On the other hand,
the scavenging activity decreased significantly after the drying
process, and it was found to be 53.29% at 60 °C. As shown in Table 6,
there was no significant difference in (% I) of fresh-air dried and
UOD-air dried samples at 40 and 50 °C. Whereas, it is decreased
by 44% in air dried samples compared to 29.44% in UOD-air dried
samples. In this aspect, some authors reported same results. Woj-
dylo et al. [50] and Alonzo-Macías, [28] reported that convection
drying lowered antioxidant activity and total phenol content in
strawberry. Regardless of the processing method employed, the
temperature of 50 °C had greater antioxidant activity than the
others temperature. Too numerous factors are implied in antioxi-
dant activity and It is difficult to explain their evolution according
to heat process conditions or phenolic content. Alonzo-Macías, [28]
reported that a decrease in phenolic content does not lead system-
atically to a decrease of the antioxidant activity. Indeed, with lower
phenolic content, one can also have a higher antioxidant activity.
From these results, it can be concluded that total phenolic content
might not be directly correlated with total antioxidant activity.
3.4.5.4. Rehydration. Rehydration, often used to evaluate instant
products, is one of the most important quality parameters of dried
strawberries. For rehydration ratio, curves were plotted for g
water/100 g dry matter versus time for each dried conditions
(Fig. 11). For all experiments the amount of moisture uptake
increased with rehydration time, but with decreasing rate up to
saturation level (Fig. 11). The same behavior was observed by
Alonzo-Macıas [28] during rehydration of swell and convective
dried strawberry fruits and by Jambrak et al. [51] during rehydra-
tion of mushrooms, brussels sprouts and cauliflower. The same
range of RR data was reported by El-Beltagy et al. [52] (within
the range 257–344) for strawberry samples of different geometries
subjected to solar drying for up to 24 h.
Fig. 11 indicates increase in rehydration ratio with increased
temperature. In cases when OD proceeded, the UOD-air dried sam-
ples had higher water content than those of fresh-air dried sam-
ples. The lowest solid uptake might be the reason of maximized
sample shrinkage during drying. This effect may be responsible
for the better water content of ultrasound osmotically treated sam-
ples. Noshad et al. [53] reported that the lower rehydration ratios
of the osmotic-ultrasonic convective dried Quince osmo- blueberries
were attributed to the added the sugar on the dehydrated fruits,
which reduced the amount of water that could be absorbed during
rehydration. Jambrak et al. [51] concluded that the rehydration
properties for ultrasound treated button mushrooms were higher
than for untreated samples, so rehydration behavior of plant food
could also be enhanced by the use of ultrasound.
Fig. 11. Rehydration kinetics of untreated (open symbols) and UOD pre-treated
(full symbols) strawberry half subjected to air temperature of 40 °C (diamonds);
50 °C (triangles); 60 °C (squares).
 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300
3.4.5.5. X-ray powder diffraction. Fig. 12 shows XRD profile of the
analysis of X-ray diffraction of the convective air dried fresh straw-
berry at 40, 50 and 60 °C temperature. The air dried strawberry
powder showed amorphous characteristics at 40, 50 and 60 °C air
drying temperature. This meant that the relatively low air temper-
ature used in drying there no crystallization of sugars of the con-
vective air-dried strawberry powder. Mazza [54] attributed the
lower rates of moisture transport observed in air drying of sucrose
treated carrots to the crystallization of sucrose during the subse-
quent air-drying process.
The diffractogram of air-dried strawberry powder obtained in
this study was comparable to the one reported by Haque et al.
[55]. The rapid drying of low molecular weight sugars present in
strawberry (sucrose, fructose and glucose) and organic acids that
happened under air drying processes tend to produce amorphous
metastable state dried products due to the fact that during drying,
the material did not reach the conditions necessary for occurrence
of crystallization of sugars (above the glassy transition tempera-
ture) [26,56].
Xray diffraction patterns of air-dried and UOD-air dried
strawberry powders at 60 °C clearly presented amorphous
Fig. 12. X-ray diffraction patterns of convective air-dried fresh strawberry at 40 °C
(cross), 50 °C (squares) and 60 °C (circles).
Fig. 13. X-ray diffraction patterns of convective air-dried strawberry at 60 °C
(circles) and UOD + convective air dried strawberry at 60 °C (full circles).
299
characteristics and exhibited no crystalline peak (Fig. 13). The
absence of crystalline peaks confirmed that no substantive changes
occurred on the hygroscopicity of air dried and UOD-air dried
strawberry powders obtained. The X-ray patterns and shapes for
all the strawberry powders tested were similar to spray-dried
sucrose showing the predominance of sucrose sugars present in
strawberry [56,57].
4. Conclusion
This work evaluated the production of osmotic-dehydrated
strawberry with low sugar content with applying ultrasound to
decrease the sugar uptake from the fruit. The influence of opera-
tion parameters such as ultrasound and OD time, solution concen-
tration and system temperature on water loss and solid gain are
discussed. The effect of ultrasonic osmosis on weight reduction is
also studied. OD concentration, ultrasound treatment duration
and temperature increase the amount of water removed during
OD. However the ultrasound treatment decreases the intake of
solids by the product. Then ultrasound OD is appropriated at
higher WL and limited solids (sugar) uptake (dietetic products).
BBD and RSM are demonstrated to be an effective technic for
modeling the effect of sonication on OD while minimizing the
number of experiments required. The second-order polynomial
models for all the response variables were found to be statistically
significant. The optimal conditions for maximum water loss and
weight reduction and minimum solid gain, were 20.5 min for
UOD time, 47.5 °Brix osmotic solution concentration and 31 °C
medium temperature in order to obtain water loss of
9.38 g/100 g fresh sample, weight reduction of 9.77 g/100 g fresh
sample and solid gain of 0.387 g/100 g fresh sample.
The result of this study indicate that increasing drying temper-
ature during the drying of strawberry increases the drying rate and
drying coefficient and consequently decreases drying time. UOD
before the air drying of strawberry tissue resulted in a spectacular
reduction of drying time and an increase in effective water diffu-
sivity because it is not influenced by slight sugar gain on the sur-
face of the fruit, which makes it easy for the water removal. The
Coupled Washing/Diffusion CWD model provides a simple mathe-
matical calculation tool for taking into account the shrinkage dur-
ing drying of fruits with high water content value. The differences
between the data corrected and non-corrected for shrinkage were
very significant, nearly one order of half magnitude. These results
showed that strawberry diffusivities were much overestimated
when shrinkage was not taken into account in drying data process-
ing. The drying time of strawberry halves was shortened by about
28% when the samples were previously treated with US (40 kHz) in
a 47.5% sucrose solution (31 °C for 21 min) prior to drying.
UOD for a short contact time minimized color losses during con-
vective air-drying. In terms of pretreatments, it can be seen that
UOD-AD 50 °C saved the brightness of both side color higher than
UOD-AD 60 °C. The highest total phenolic content (TPC) was
recorded in plants dried by AD 50 °C and UOD-AD 40 °C and was
38% low than those of fresh plants Regardless of the processing
method employed, the temperature of 50 °C had greater antioxi-
dant activity than the others temperature. The X-ray diffraction
patterns of air-dried and UOD-air dried strawberry powders clearly
presented amorphous characteristics and exhibited no crystalline
peak formation.
This method could make a significant contribution to the fruit
processing industry and confer a higher quality to the finished
products. Strawberries dehydrated by means of UOD can be an
attractive component of popular breakfast cereals such as
muesli.
300 E. Amami et al. / Ultrasonics Sonochemistry 36 (2017) 286–300

References [27] Y.H. Roos, Phase Transitions in Foods, Academic Press, New York, 1995, p. 360.
[28] M. Alonzo-Macıas, Comparative Study of Strawberry Drying Processes: Hot Air
Drying, Freeze-Drying and Swell Drying, Application on the Biological
[1] F. Giampieri, S. Tulipani, J.M. Alvarez-Suarez, J.L. Quiles, B. Mezzetti, M. Battin,
Compounds Preservation, University of La Rochelle, La Rochelle, 2013.
The strawberry: composition, nutritional quality and impact on human health,
[29] J. Gamboa-Santos, A. Montilla, A. Cortijo Soria, J.A. Carcel Carrión, J.V. García
Nutrition 28 (2012) 9–19.
Pérez, M. Villamiel, Impact of power ultrasound on chemical and
[2] FAOSTAT. Core production data database, Food and Agriculture Organization of
physicochemical quality indicators of strawberries dried by convection, Food
the United Nations, 2013, <http://Faostat.Fao.Org/>.
Chem. 161 (2014) 40–46, http://dx.doi.org/10.1016/j.foodchem.2014.03.106.
[3] D. Knorr, M. Zenker, V. Heinz, D.U. Lee, Applications and potential of
[30] J. Stojanovic, J.L. Silva, Influence of osmotic concentration, continuous high
ultrasonics in food processing, Trends Food Sci. Technol. 15 (2004) 261–266.
frequency ultrasound and dehydration on antioxidants, colour and chemical
[4] J. Gamboa-Santos, R. Megías-Pérez, A. Cristina Soria, A. Olano, A. Montilla, M.
properties of rabbiteye blueberries, Food Chem. 101 (2007) 898–906.
Villamiel, Impact of processing conditions on the kinetic of vitamin C
[31] A. Wiktor, M. Sledz, M. Nowacka, K. Rybak, D. Witrowa-Rajchert, The influence
degradation and 2-furoylmethyl amino acid formation in dried strawberries,
of immersion and contact ultrasound treatment on selected properties of the
Food Chem. 153 (2014) 164–170.
apple tissue, Appl. Acoust. 103 (2016) 136–142.
[5] T. Allaf, K. Allaf, Instant Controlled Pressure Drop (DIC) in Food Processing,
[32] J. Garcia-Noguera, F.I.P. Oliveira, L. Curtis, W.S. Rodrigues, F.A.N. Fernandes,
Springer-Verlag, New York, United States, 2014, p. 183.
Effect of ultrasonic and osmotic dehydration pre-treatments on the colour of
[6] T.H. Nguyen, J.L. Lanoisellé, T. Allaf, K. Allaf, Experimental and fundamental
freeze dried strawberries, J. Food Sci. Technol. 51 (2014) 2222–2227.
critical analysis of diffusion model of airflow drying, Drying Technol. (2016),
[33] J.M. Löning, C. Horst, U. Hoffmann, Investigations on the energy conversion in
http://dx.doi.org/10.1080/07373937.2016.1155052.
sonochemical process, Ultrason. Sonochem. 9 (2002) 169–179.
[7] J. Dehghannya, R. Gorbani, B. Ghanbarzadeh, Shrinkage of mirabelle plum
[34] Association of Official Analytical Chemists. Moisture in dried fruits. In Official
during hot air drying as influenced by ultrasound-assisted osmotic
methods of analysis (16th edition). Washington, USA: Association of Official
dehydration, Int. J. Food Prop. 19 (2016), http://dx.doi.org/10.1080/
Analytical Chemists, 1990.
10942912.2015.1055362.
[35] J. Crank, The Mathematics of Diffusion, Clarendon Press, Oxford, 1975.
[8] J.E. Lozano, E. Rotstein, M.J. Urbicain, Shrinkage, porosity and bulk density of
[36] I. Doymaz, Convective drying kinetics of strawberry, Chem. Eng. Process. 47
foodstuffs at changing moisture contents, J. Food Sci. 48 (1983) 1497–1502.
(2008) 914–919.
[9] P.S. Madamba, R.H. Driscoll, K.A. Buckle, Shrinkage, density and porosity of
[37] V.L. Singleton, R. Orthofer, R.M. Lamuela-Raventos, Analysis of total phenols
garlic during drying, J. Food Eng. 23 (1994) 309–319.
and other oxidation substrates and antioxidants by means of Folin-Ciocalteu
[10] J. Gamboa-Santos, A. Montilla, J.A. Cárcel, M. Villamiel, J.V. Garcia-Perez, Air-
reagent, Methods Enzymol. 299 (1999) 152–178.
borne ultrasound application in the convective drying of strawberry, J. Food
[38] W. Brand-Williams, M.E. Cuvelier, C. Berset, Use of a free radical method to
Eng. 128 (2014) 132–139.
evaluate antioxidant activity, LWT – Food Sci. Technol. 28 (1) (1995) 25–30.
[11] E. Amami, A. Fersi, L. Khezami, E. Vorobiev, N. Kechaou, Effect of pulsed electric
[39] A. Ghouili, J. Rohlicek, T. Ben Ayed, Ben Hassen, Crystal structure
field and osmotic dehydration pretreatment on the convective drying of carrot
determination from powder diffraction data of the coumarin vanillin
tissue, Drying Technol. 26 (2008) 231–238.
chalcone, Powder Diffr. 29 (2014) 361–365.
[12] O. Rodríguez, W. Gomes, S. Rodrigues, F.A.N. Fernandes, Effect of acoustically
[40] J.M. Conway, F. Castaigne, G. Picard, X. Voxan, Mass transfer considerations in
assisted treatments on vitamins, antioxidant activity, organic acids and drying
the osmotic dehydration of apples, Inst. Food Sci. Technol. 16 (1983) 25–29.
kinetics of pineapple, Ultrason. Sonochem. 35 (2017) 92–102, http://dx.doi.
[41] M.B. Uddin, P. Ainsworth, S. Ibanoglu, Evaluation of mass exchange during
org/10.1016/j.ultsonch.2016.09.006.
osmotic dehydration of carrots using response surface methodology, J. Food
[13] J. Garcia-Noguera, F.I.P. Oliveira, M.I. Gallão, C.L. Weller, S. Rodrigues, F.A.N.
Eng. 65 (2004) 473–477.
Fernandes, Ultrasound-assisted osmotic dehydration of strawberries: effect of
[42] F.A.N. Fernandes, F.I.P. Oliveira, S. Rodrigues, Use of ultrasound for dehydration
pre-treatment time and ultrasonic frequency, Drying Technol. 28 (2010) 294–
of papayas, Food Bioprocess Technol. 1 (2008) 339–345.
303.
[43] D. Piotrowski, A. Gołoś, P. Grzegory, Shrinkage and mechanical properties of
[14] E. Amami, E. Vorobiev, N. Kechaou, Effect of pulsed electric field on the osmotic
defrosted strawberries dried by convective, vacuum and convective-vacuum
dehydration and mass transfer kinetics of apple tissue, Drying Technol. 23
methods, Acta Agrophysica. 21 (2014) 193–204.
(2005) 581–595.
[44] L.A. Ramallo, R.H. Mascheroni, Effect of shrinkage on prediction accuracy of the
[15] E. Amami, E. Vorobiev, N. Kechaou, Modelling of mass transfer during osmotic
water diffusion model for pineapple drying, J. Food Process Eng., 1745-4530
dehydration of apple tissue pre-treated by pulsed electric field, LWT–Food Sci.
(2011), http://dx.doi.org/10.1111/j.1745-4530.2011.00654.x.
Technol. 39 (2006) 1014–1021.
[45] S. Simal, A. Mulet, P.J. Catalá, J. Cañellas, C. Rosselló, Moving boundary model
[16] N. Barman, L.S. Badwaik, Effect of ultrasound and centrifugal force on
for simulating moisture movement in grapes, J. Food Sci. 61 (1996) 157–160.
carambola (Averrhoa carambola L.) slices during osmotic dehydration,
[46] A.L. Gabas, F.C. Menegalli, J. Telis-Romero, Determination of Concentration-
Ultrason. Sonochem. 34 (2017) 37–44.
Dependent Effective Moisture Diffusivity of Plums Based on Shrinkage
[17] F.A.N. Fernandes, S. Rodrigues, Application of ultrasound and ultrasound-
Kinetics, in: J. Welti-Chanes, J.F. Velez-Ruiz (Eds.), Transport Phenomena in
assisted osmotic dehydration in drying of fruits, Drying Technol. 26 (2008)
Food Processing. Food Preservation Technology, CRC Press, Boca Raton, 2003,
1509–1516.
http://dx.doi.org/10.1201/9781420006261.ch10.
[18] S. Fuente-Blanco, S E.R.F. Arabia, V.M. Acosta-Aparicio, A. Blanco-Blanco, J.A.
[47] J.V. García-Pérez, J.A. Cárcel, S. de la Fuente, E. Riera, Ultrasonic drying of
Gallego-Juarez, Food drying process by power ultrasound, Ultrason.
foodstuff in a fluidized bed: parametric study, Ultrasonics 44 (2006) 539–543.
Sonochem. 44 (2006) 523–527.
[48] J. Moreno, A. Chiralt, I. Escriche, J.A. Serra, Effect of blanching/osmotic
[19] D. Yongbo, Z. Jiong, X. Xuejuan, R. Tingyuan, K. Jianquan, Boxe Behnken design
dehydration combined methods on quality and stability of minimally
for the optimization of nanoscale retrograded starch formation by high-power
processed strawberries, Food Res. Int. 33 (2000) 609–616.
ultrasonication, LWT–Food Sci. Technol. 67 (2016) 206–213.
[49] Y. Nuñez-Mancilla, M. Pérez-Won, E. Uribe, A. Vega-Gálvez, K. Di Scala,
[20] P. Dayal, V. Pillay, R.J. Babu, M. Singh, Box-Behnken experimental design in the
Osmotic dehydration under high hydrostatic pressure: Effects on antioxidant
development of a nasal drug delivery system of model drug hydroxyurea:
activity, total phenolics compounds, vitamin C and colour of strawberry
characterization of viscosity, in vitro drug release, droplet size and dynamic
(Fragaria vesca), LWT – Food Sci. Technol. 52 (2013) 151–156.
surface tension, AAPS PharmSciTech 6 (2005) 573–585.
[50] A. Wojdyło, A. Figiel, J. Oszmian’Ski, Effect of drying methods on strawberry
[21] M. Roosta, M. Ghaedi, A. Daneshfar, R. Sahraei, A. Asghari, Optimization of the
fruits, J. Agric. Food Chem. 57 (2009) 1337–1343.
ultrasonic assisted removal of methylene blue by gold nanoparticles loaded on
[51] A. Rezek Jambrak, T.J. Mason, L. Paniwnyk, V. Lelas, Accelerated drying of
activated carbon using experimental design methodology, Ultrason.
button mushrooms, brussels sprouts and cauliflower by applying power
Sonochem. 21 (2014) 242–252, http://dx.doi.org/10.1016/j.
ultrasound and its rehydration properties, J. Food Eng. 81 (2007) 88–97.
ultsonch.2013.05.014.
[52] A. El-Beltagy, G.R. Gamea, A.H. Amer Essa, Solar drying characteristics of
[22] F. Dranca, M. Oroian, Optimization of ultrasound-assisted extraction of total
strawberry, J. Food Eng. 78 (2007) 456–464.
monomeric anthocyanin (TMA) and total phenolic content (TPC) from
[53] M. Noshad, M. Mohebbi, F. Shahidi, S. Ali Mortazavi, Multi-objective
eggplant (Solanum melongena L.) peel, Ultrason. Sonochem. 31 (2016) 637–
optimization of osmotic-ultrasonic pretreatments and hot-air drying of
646.
quince using response surface methodology, Food Bioprocess Technol. 5
[23] K.A. Taiwo, M.N. Eshtiaghi, B.I.O. Ade-Omowaye, D. Knorr, Osmotic
(2012) 2098–2110.
dehydration of strawberry halves: Influence of osmotic agents and
[54] G. Mazza, Dehydration of carrots: Effects of pre-drying treatments on moisture
pretreament methods on mass transfer and product characteristics, Int. J.
transport and product quality, J. Food Technol. 18 (1983) 113–123.
Food Sci. Technol. 38 (2003) 693–707.
[55] M. Haque, Y. Roos, Crystallization and X-ray diffraction of spray-dried and
[24] H. Allali, L. Marchal, E. Vorobiev, Blanching of strawberries by ohmic heating:
freeze-dried amorphous lactose, Carbohydr. Res. 340 (2005) 293–301.
effects on the kinetics of mass transfer during osmotic dehydration, Food
[56] O.A. Caparino, J. Tang, C.I. Nindo, S.S. Sablani, J.R. Powers, J.K. Fellman, Effect of
Bioprocess Technol. 3 (2010) 406–414.
drying methods on the physical properties and microstructures of mango
[25] E. Amami, A. Fersi, L. Khezami, E. Vorobiev, N. Kechaou, Centrifugal osmotic
(Philippine ‘Carabao’ var.) powder, J. Food Eng. 111 (2012) 135–148.
dehydration and rehydration of carrot tissue pre-treated by pulsed electric
[57] K.A. Athmaselvi, C. Kumar, M. Balasubramanian, I. Roy, Thermal, structural,
field, LWT–Food Sci. Technol. 40 (2007) 1156–1166.
and physical properties of freeze dried tropical fruit powder, J. Food Process.
[26] M. Cano-Chauca, P.C. Stringheta, A.M. Ramos, J. Cal-Vidal, Effect of the carriers
(2014) 10, Article ID 524705.
on the microstructure of mango powder obtained by spray drying and its
functional characterization, Innovative Food Sci. Emerg. Technol. 6 (2005)
420–428.