Journal of Food Engineering 103 (2011) 2937

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Shrinkage, density, porosity and shape changes during dehydration

of pumpkin (Cucurbita pepo L.) fruits
L. Mayor a,, R. Moreira b, A.M. Sereno c
a

Instituto Universitario de Ingeniera de Alimentos para el Desarrollo, Universidad Politcnica de Valencia, Camino de Vera, s/n, 46022 Valencia, Spain
Departamento de Enxeara Qumica, Universidade de Santiago de Compostela, Ra Lope Gmez de Marzoa s/n, E-15782 Santiago de Compostela, Spain
c
REQUIMTE, Department of Chemical Engineering, Faculty of Engineering, University of Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal
b

a r t i c l e

i n f o

Article history:
Received 2 March 2010
Received in revised form 27 August 2010
Accepted 29 August 2010
Available online 8 October 2010
Keywords:
Air drying
Bulk density
Modelling
Osmotic dehydration
Particle density
Shape factors

a b s t r a c t
The aim of this work was to study the changes in volume, density, porosity and shape factors of pumpkin
tissue during osmotic dehydration (OD) and air drying (AD). Pumpkin cylinders with length/diameter
ratio of 5/3 were used. OD experiments were carried out with solutions of sucrose, sodium chloride
and mixtures of both solutes at different temperatures. AD experiments were conducted at 70 C. Volume
of samples decreased linearly with weight reduction (WR). Bulk density varied in a restricted range (5
13%) during dehydration and for all the methods maximum values were found. Particle density increased
during both processes. Porosity increased at advanced degrees of dehydration, showing a minimum value
at the beginning of OD and AD. The proposed models to evaluate shrinkage, bulk and particle densities
and porosity from WR were satisfactorily applied. Image analysis showed that shrinkage of samples during OD was isotropic. Pumpkin cylinders increased elongation and decreased roundness and compactness
during osmotic dehydration.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
The knowledge of physicochemical properties of food materials
is important for an adequate design of food operations as well as
for the control and improvement of the quality of the nal product
(Rahman, 2005).
Particularly in dehydration processes, the heat and mass transfer ows can modify physicochemical properties of the material
such as chemical composition (McLaughlin and Magee 1998),
mechanical properties (Lewicki and Lukaszuk, 2000) and volume
and porosity. The quality of the dehydrated product depends on
the extension of these changes. Regarding to the changes in volume and porosity, high shrinkage and low porosity lead to products with poor rehydration capability (McMinn and Magee,
1997). Furthermore, the changes in volume and dimensions must
be considered for mass transfer modelling during dehydration
(Simal et al., 1998; Khallou et al., 2009). In fact, experimental
shrinkage data during air drying (AD) of foods is relatively abundant (Mayor and Sereno, 2004).
Knowledge of the bulk density of food materials is an important
parameter in storage, transport, mixing and packaging operations
(Rahman, 2005). Heat and mass transfer in solids depend on density and porosity values (Rahman, 2001). Textural and sensorial

Corresponding author.
E-mail address: luimalo@iad.upv.es (L. Mayor).
0260-8774/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jfoodeng.2010.08.031

properties of foods are also related to density and porosity

(Rahman, 2001). Porosity is related to the chemical stability of
dried products; degradation of sugars (White and Bell, 1999) and
lipid oxidation (Shimada et al., 1991) depended on this property.
Food shape is one of the main quality attributes perceived by
the consumer (Fernandez et al., 2005). Drying not only causes volume changes but also may cause changes in shape. In this sense,
product deformation is not fully described by the evaluation of
volumetric shrinkage (Panyawong and Devahastin, 2007).
Osmotic dehydration (OD) is a non-thermal process that consists
in the immersion of a food material in a hypertonic solution. The
difference of the chemical potential between the material and the
solution promotes two main uxes: the outcome of water from
the material to the osmotic solution, and the income of soluble solids from the osmotic solution to the material. As osmotic agents are
often used sugars (sucrose or glucose) and salts (sodium chloride).
Volume changes during OD are mainly due to compositional
changes and mechanical stresses associated to mass uxes. These
changes have been analyzed as variations in the volumes of solid,
liquid and gas phases of the food material during the process (Barat
et al., 2001), and have been correlated with changes in moisture
content and WR (Moreira and Sereno, 2003), or with WL (Nieto
et al., 2004). These three aforementioned works studied shrinkage
phenomena during OD of apples. Volumetric shrinkage during OD
of other food products has also been reported, as the case of
strawberries (Viberg et al., 1998), mangos (Giraldo et al., 2003)
and tomatoes (Souza et al., 2007; Bui et al., 2009). In spite of these

30

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

Nomenclature
a
AD
ARD
b
c
C
d
D
E
L
Lm
m
N
NMC
OD
p
R
R2
Re
s
S
SG

parameter of Eqs. (11) and (12)

air drying
average relative deviation
parameter of Eq. (11)
parameter of Eq. (11)
compactness
diameter of the stirrer (m)
diameter (m)
elongation
length (m)
length of major axis (m)
sample mass (kg)
number of revolutions per second (s1)
normalized moisture content
osmotic dehydration
perimeter (m)
roundness
coefcient of determination
Reynolds number
solid mass (kg)
surface area (m2)
solids gain (kg/kg)

works, few experimental data on shrinkage during OD of fruits and

vegetables are found in the literature, and the variety of food materials studied is restricted.
Pumpkin (Cucurbita pepo L.) is a seasonal crop, which has been
used for human and animal feed. OD of pumpkin, as a single or
combined process, can be an alternative way to develop new food
products (Kowalska et al., 2008; Konopacka et al., 2010). OD kinetics of pumpkin were already determined under several experimental conditions by using binary solutions of sucrose and glucose
(Kowalska et al., 2008) or sodium chloride (Mayor et al., 2006)
and ternary solutions of sucrose and sodium chloride (Mayor
et al., 2007).
The aim of this work was to present experimental data on
changes in volume, bulk and particle densities and porosity during
OD of pumpkin fruits. The obtained results were compared with
the changes observed during air drying (the most conventional
method of dehydration) of this food material. Models to estimate
these physical properties during dehydration were proposed.
Dimensional and shape changes were also studied by means of image analysis.
2. Materials and methods
2.1. Sample preparation
Pumpkin fruits ( C. pepo L.) were purchased from a local producer, and stored at 1520 C in a chamber until processing. Pumpkins with similar initial moisture content (9597 kg water/100 kg
product) and soluble solids (24 Brix) were selected for the experiments. Cylinders (25 mm length, 15 mm diameter) from the
parenchyma tissue were obtained employing a metallic cork borer
and a cutter. In order to obtain a good structural and compositional
homogeneity in the samples, the cylinders were taken from the
middle zone of the mesocarp, parallel to the major axis of the fruit.
2.2. Dehydration experiments
Pumpkin cylinders were dehydrated with sucrose solutions at
different concentrations and temperatures selected using a uni-

suc
T
V
WL
WR
X
Y

sucrose
temperature (C)
volume (m3)
water loss (kg/kg)
weight reduction (kg/kg)
variable of Eq. (11)
coded variable

Greek symbols
e
porosity
m
kinematic viscosity (m2/s)
q
density (kg/m3)
Subscripts
b
bulk
is
insoluble solids
o
initial
p
particle
ss
initial soluble solids
suc
sucrose
w
water

form shell design (Doehlert, 1970), as observed in Table 1. These

solutions were prepared with distilled water and commercial
sucrose. The cylinders were put in baskets, which were fully
immersed into stirred glass vessels (diameter 15 cm, height
25 cm) containing the osmotic solution. Then, the vessels were
hermetically closed to avoid water evaporation during the osmotic
treatments. Agitation was conducted using a magnetic stirrer; the
speed was chosen according to the kinematic viscosity of the
osmotic solution to obtain a constant Reynolds number (ca.
3000) (Mayor et al., 2006). Reynolds number was calculated
according Eq. (1) (Perry and Green, 1999)
2

Re

Nd

where kinematic viscosities of the osmotic solutions were obtained

from other authors (Chenlo et al., 2002), and ranged from
0.94  106 up to 37.5  106 m2/s.
The weight ratio of osmotic solution to pumpkin cylinders was
20:1 to maintain a constant concentration of the osmotic solution
during OD. Thermoregulation was obtained by means of a thermostatic bath (0.2 C). Some dehydrated samples were removed
from the vessels at different process times, immediately they were
gently blotted with paper to remove the excess of osmotic solution
and kept in plastic boxes till experimental determinations.
The same experimental procedure was followed in the OD
experiments with sodium chloride solutions at concentrations of

Table 1
Experimental design for dehydration of pumpkin with sucrose solutions.
Coded experimental plan

Actual experimental plan

Y1

Y2

Sucrose (kg/100 kg)

T (C)

1
0
1
0.5
0.5
0.5
0.5

0
0
0
0.866
0.866
0.866
0.866

30
45
60
37.5
37.5
52.5
52.5

25
25
25
12
38
12
38

31

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

Image analysis was performed in samples osmodehydrated

with 60% sucrose solutions at 25 C. Six samples were removed
at different process times for the analysis of size and shape parameters. From each sample, one rectangular slab of ca. 0.51 mm of
thickness was gently cut parallel to the height of the cylinders at
the maximum section area with a razor blade. One face of the slab
was stained with a solution of methylene blue 0.1% (Mayor et al.,
2005) during 15 s. After that, samples were ready for observation
under the stereomicroscope (Olympus SZ-11, Tokyo, Japan) working in transmitted light mode. A digital video camera (Sony SSCDC50AP, Tokyo, Japan) was attached to the microscope and connected to a computer. Image acquisition was carried out with an
interface (PCTV videocard, Pinneacle Systems GmbH, Munich,
Germany). Images were calibrated with a stage micrometer of
2 mm length and divisions of 0.01 mm intervals (Leitz Wetzlar,
Germany). Image analysis of the isolated sample contour was
performed using free software UTHSCSA Image Tool v.2.0 (Health
Science Centre, University of Texas, San Antonio, TX). Several geometrical parameters of the samples were analyzed (Lewicki and
Pawlak, 2003; Mayor et al., 2005): surface area (S); perimeter of
the contour (p), length of the major axis (Lm), dened as the length
of the longest line that can be drawn through the object; length of
the minor axis, dened as the length of the longest line that can be
drawn through the object perpendicular to the major axis; elongation (E), dened as the ratio of Lm to the length of the minor axis;
roundness (R) and compactness (C) dened by the Eqs. (9) and (10),
respectively:

5, 10 and 20 kg/100 kg at 25, 38 and 12 C, respectively; and with

ternary solutions 3.75% NaCl58% sucrose and 7.5% NaCl45% sucrose at 25 C.
AD of fresh samples was performed in an oven at 70 C under
natural convection and an air relative humidity of 6 2% (determined by wet bulb thermometer). A vent in the oven avoided
humidity build-up during drying. Pumpkin cylinders were put on
a plastic wire net to allow heat and mass transfer through the
whole surface of the samples. At different process times some cylinders were removed from the oven and kept in plastic boxes till
experimental determinations.
2.3. Experimental determinations
At each selected process time, four samples were used for gravimetric determinations. The bulk volume (V) of each cylinder was
calculated from the resultant buoyant force of the sample when
immersed in n-heptane (Lozano et al., 1980). After this measurement, particle volume (Vp) was obtained with a gas pycnometer
specially designed to measure this property in moist products
(Sereno et al., 2007). Sample porosity (e) was calculated by Eq. (2)

V  Vp
Vp
1
V
V

After that, the same samples used for volume measurements

were weighed to determine its weight reduction (WR), Eq. (3),
and the solids gain (SG), Eq. (4), was evaluated after vacuum drying
at less than 104 Pa at 70 C till constant weight (AOAC, 1984)
according to:

mo  m
WR
mo
s  so
SG
mo

m  s mo
mo  so m

Particle (qp) and bulk (qb) densities were calculated from Eqs.
(7) and (8), respectively:

m
Vp
m
qb
V

qp

10

Lm

R and C parameters give an idea of the circularity of the object. Both

shape factors range from 0 to 1; when the value is one, the object is
a perfect circle, when their value decrease the object becomes less
circular and less round.
In order to determine the linear dimensions, average values of
ve measurements at different zones of the cylinders were performed for diameter (D) and length (L). Table 2 shows a summary
of the experimental determinations performed in this work, as well
as the sampling times selected for each experiment.

Water loss (WL) and normalized moisture content (NMC) were

determined by Eqs. (5) and (6), respectively:

NMC

4S

WL SG WR

4pS
p2
q

7
3. Results and discussion

8
3.1. Shrinkage during dehydration

Since the soluble solids of pumpkin esh are mostly sugars, soluble solids of fresh samples were determined by refractometry
(Abbe-3L refractometer, Bausch and Lomb, Rochester, NY, USA) at
20 C. The clear juice was extracted by manually pressing raw
pumpkin esh (ca. 2 g) between to plastic discs (diameter 3 cm)
and analyzed directly in the refractometer. Insoluble solids were
obtained from a mass balance with the values of soluble solids
and total solids in fresh material.

Fig. 1 shows experimental shrinkage data of pumpkin cylinders

versus WL, WR and NMC, respectively. Plots on the left correspond
to OD with sucrose solutions at different conditions, whereas plots
on the right correspond to OD with different osmotic agents and
AD. Osmodehydrated samples shrank up to 27% of their initial volume, depending on the process conditions used. Air dried samples
shrank at the end of drying up to 5% of the initial volume.

Table 2
Summary of experimental determinations.
Dehydration treatment

Process conditions

Experimental determinations

Sampling times (h)

Osmotic dehydration

Sucrose, Table 1
60% Sucrose, 25 C
NaCl
NaCl/sucrose

a,
a,
a,
a,

0,
0,
0,
0,

Air drying

Natural convection (oven drying) at 70 C

a, b, c

b, c
d
b
b

0.08, 0.25, 0.5, 1, 1.5, 2, 2.5, 3, 4, 6 and 9

0.5, 1, 3, 6 and 9
0.08, 0.5, 1, 2, 2.5, 3, 4, 5, 6 and 8
0.08, 0.25, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5 and 6

0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5, 6 and 8

(a) Kinetic parameters (water loss, solids gain, weight reduction, normalized moisture content); (b) bulk volume; (c) particle volume; and (d) image analysis.

32

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

(a)

1.1

1.1

1.0

1.0

0.9

0.9

0.8

0.8

0.7

0.7
0.6

0.4
0.3
0.2
0.1
0.0
0.0

V/Vo

V/Vo

0.6
0.5

OD, sucrose solutions

OD, NaCl/sucrose solutions
OD, NaCl solutions
Air drying

30% suc, 25C

45% suc, 25C
60% suc, 25C
37.5% suc, 12C
37.5% suc, 38C
52.5% suc, 12C
52.5% suc, 38C
Eq. (11)

0.1

0.2

0.5
0.4
0.3
0.2
0.1

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0.0
0.0

1.0

0.1

0.2

0.3

(b)

1.1

1.1

1.0

1.0

0.9

0.9

0.8

0.8

0.7

0.7

0.3
0.2
0.1
0.0
0.0

V/Vo

V/Vo

0.4

30% suc, 25C

45% suc, 25C
60% suc, 25C
37.5% suc, 12C
37.5% suc, 38C
52.5% suc, 12C
52.5% suc, 38C
Eq. (11)

0.8

0.9

1.0

0.1

0.2

0.5

0.3
0.2
0.1

0.3

0.4

0.5

0.6

0.7

0.0
0.0

0.8

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

WR (kg/kg)
1.1

1.0

0.8
0.7

0.9
0.8
0.7
0.6

V/Vo

0.6
0.5

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0.9

0.8

0.7

0.6

NMC

OD, sucrose solutions

OD, NaCl/suc solutions
OD, NaCl solutions
Air drying

1.0

30% suc, 25C

45% suc, 25C
60% suc, 25C
37.5% suc, 12C
37.5% suc, 38C
52.5% suc, 12C
52.5% suc, 38C
Eq. (11)

0.9

V/Vo

0.7

0.4

1.1

0.0
1.0

0.6

OD sucrose solutions
OD NaCl/sucrose solutions
OD NaCl solutions
Air drying
Eq. (11), all treatments

WR (kg/kg)

(c)

0.5

0.6

0.6
0.5

0.4

WL (kg/kg)

WL (kg/kg)

0.5

0.0
1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

NMC

Fig. 1. Shrinkage during dehydration of pumpkin cylinders versus (a) water loss, (b) weight reduction, and (c) normalized moisture content. Left gures correspond to
osmotic dehydration with sucrose solutions, whereas right gures correspond to osmotic dehydration with different solutions and air drying.

In Fig. 1a can be observed a linear decrease of volume with WL

during OD. The decrease is more accentuated in the case of NaCl
solutions, followed by sucrose solutions and NaCl/sucrose solutions. No effect of process conditions concentration and temperature is observed for the same osmotic agent, as shown in Fig. 1a
for sucrose solutions. Nieto et al. (2004), observed a linear decrease
of sample volume with WL during OD of apple; the decrease was
more accentuated at the same WL for the samples osmodehydrated with sucrose solutions compared with glucose solutions due to
the corresponding lower SG value. Mavroudis et al. (1998) during
OD of apples (var. Granny Smith) with sucrose solutions also observed a linear decrease of volume with the decrease of water in
the material; they observed no effect of process temperature on
shrinkage.

Air dried samples also showed a linear volume decrease with

WL, but in this case the decrease is higher than the shrinkage of
OD samples. At the same WL, the SG during OD decreases the volume reduction promoted by the water removal. Volume decreases
linearly with increasing WR (Fig. 1b), independently on the process
conditions and dehydration methods. Moreira and Sereno (2003)
during OD of apple with sucrose solutions found linear relationships between WR and shrinkage independently on concentration,
temperature and hydrodynamic conditions of the osmotic solutions. During AD, this linear behaviour of shrinkage against WR
(and consequently against WL) is often reported (Lozano et al.,
1983; Zogzas et al., 1994). Shrinkage of osmodehydrated and air
dried samples shows a non-linear decrease with moisture content
(Fig. 1c). This decrease is faster for pumpkins submitted to AD, due

33

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

V
2
1 aX bX cX 3
Vo

11

where X can be WL, WR, or NMC. Eq. (11) was tted to experimental
shrinkage data obtained in this work. The corresponding ts were
carried out considering each osmotic agent alone, considering all
the osmotic treatments together, AD alone, and all the dehydration
treatments together (AD and OD). Table 4 shows the values of the
parameters after tting (only were considered acceptable the ttings with ARD < 10% and R2 > 0.9). Fig. 1 shows some of these ttings, as examples. For WL and WR the linear t is satisfactory in
the case of OD treatments and AD, separately. For all the dehydration methods together only WR t (linear) is acceptable. For NMC,
cubic models are acceptable (except for AD and for the global analysis of all treatments). These equations are useful because allow the
prediction of shrinkage data independently on the process conditions (concentration and temperature) used.
3.2. Bulk density, particle density and porosity
Table 5 shows the values of some physicochemical properties of
raw pumpkin parenchymatic tissue obtained in this work. Particle
and bulk densities and porosity present the typical variability ob-

Table 3
Density, q, for fresh and dehydrated pumpkin components at 25 C.
Component

q (kg/m3)

Reference

Water
Fructose
Glucose
NaCl
Sucrose
Cellulose

997
1665
1562
2170
1581
1550

Lide (2005)
Lide (2005)
Lide (2005)
Lide (2005)
Lide (2005)
Lozano et al. (1980)

1.0

O.D. Sucrose solutions

O.D. NaCl solutions
Air drying
Diagonal

0.9
0.8
0.7

Ideal volume loss/Vo

to the aforementioned explanations given for the analysis of

shrinkage against WL.
Variation in the volume of the gas phase inside the vegetable
tissue also can contribute to increase or decrease the theoretical
shrinkage produced by the removal of water. An ideal shrinkage
could be dened as the shrinkage due to mass uxes (WL and
SG), assuming that the volume of the initial solids of the material
remains constant. Considering volume additivity, the changes in
volume were calculated from experimental data of WL and solute
gain. The density values of water, sucrose and NaCl at 25 C, are
shown in Table 3. The assumption that volumes are additive is
acceptable for the calculations, since excess volume of binary solutions of sucrose and NaCl and ternary NaCl/sucrose solutions obtained comparing experimental data from other authors is near
to 1%, rarely exceeding 2% (Chenlo et al., 2002; Lide 2005).
Fig. 2 shows the ideal shrinkage of pumpkin cylinders dehydrated with binary solutions of sucrose and NaCl and air dried versus the experimental shrinkage data. It is observed that, for each
dehydration method, experimental shrinkage is higher than the
ideal shrinkage given by the mass uxes during dehydration.
These differences increase at high degrees of shrinkage, and consequently with the volume of removed water, and are related to the
decrease of the air volume and collapse of the material (Khallou
et al., 2009). Barat et al. (2001), during OD of apples with sucrose
solutions, observed that the decrease of total volume was higher
than the decrease of the liquid phase volume in the samples. Several phenomena can promote the collapse, such as capillary forces
caused by the water removal and loss of turgor pressure in the cells
(Prothon et al., 2003).
Shrinkage of pumpkin during dehydration can be correlated
with the kinetic parameters, X, by means of empirical polynomial
equations:

0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

(Vo-V)/Vo
Fig. 2. Ideal volume loss versus actual volume loss during osmotic dehydration of
pumpkin fruits with binary sucrose and NaCl solutions and air drying.

Table 4
Parameters of Eq. (11) for osmotic dehydration (OD) and air drying.
X

R2

ARD (%)

OD sucrose solutions
WL (linear)
WR (linear)
NMC (cubic)

0.88
1.01
1.74

0.18

1.94

0.99
0.99
0.97

1.65
1.44
5.26

OD NaCl solutions
WL (linear)
WR (linear)
NMC (cubic)

0.94
1.09
16.04

37.82

21.74

0.96
0.99
0.91

2.40
1.19
6.38

OD NaCl/sucrose solutions
WL (linear)
0.82
WR (linear)
1.02
NMC (cubic)
2.17

1.50

0.66

0.99
0.99
0.99

1.18
1.96
2.67

All OD treatments
WL(linear)
WR(linear)
NMC (cubic)

0.87
1.02
1.74

0.22

1.49

0.99
0.99
0.91

3.13
1.81
8.09

Air drying
WL, WR (linear)

1.00

0.99

2.86

All dehydration methods

WR (linear)
1.02

0.99

2.43

WL = water loss, WR = weight reduction, NMC = normalized moisture content,

R2 = coefcient of determination, ARD = average relative deviation.

Table 5
Some physicochemical properties of raw pumpkin parenchymatic tissue.
Property

Average value

Range

Moisture content (%)

Soluble solids (%)
Insoluble solids (%)
Bulk density (kg/m3)
Particle density (kg/m3)
Porosity (%)

95.57
3.22
1.21
890
1040
14.79

[94.4496.92]
[2.143.63]
[0.781.97]
[860920]
[10031070]
[10.2218.18]

served in other vegetables; for this reason, the changes in these

properties during dehydration are presented as reduced values.
The change of bulk density during OD with sucrose or NaCl
solutions was restricted (ca. 5%). Higher change was obtained for
ternary NaCl/sucrose solutions (ca. 10%) and AD (ca. 13%). Similar
trend is observed for all the treatments during dehydration; bulk
density initially increases, reaching a maximum value and then decreases or uctuates till the end of the process (data not shown).

34

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

Nieto et al. (2004), during OD of apples with sucrose and glucose solutions, observed an increase of bulk density of apple samples in the beginning of the treatments (around the rst hour of
treatment) for both osmotic agents, and then bulk density uctuated until the end of the processes. During AD of different fruits
and vegetables, Lozano et al. (1983) and Krokida and Maroulis
(1997) observed that the change in bulk density was different for
each food material tested. In some cases bulk density increased
during dehydration (banana, carrot), in other cases decreased
(apple), and still in other cases initially increased, reached a maximum value and then decreased (sweet potato, garlic). This different behaviour could be associated with different physicochemical
characteristics of the raw material, such as chemical composition,
initial porosity or the presence of soft/rigid structures, which can
lead to different type of stresses during processing (Rahman,
2001). It is reasonable to think that for OD the characteristics of
the raw material also can inuence in the change in bulk density
as observed for AD of vegetables. The type and amount of infused
solids are other important factors in the change of bulk density
during OD.
Bulk density changes can be predicted by means of Eq. (12):

qb 1  WR 1  WR

V=V o
1 aWR
qbo

12

where a is the corresponding regression coefcient of Eq. (11)

(Table 4), for shrinkage and WR correlations for each osmotic agent
and for AD. Average and maximum relative deviations between
experimental and predicted density values were, respectively,
2.4% and 5.6% for sucrose solutions; 1.3% and 4.8% for NaCl solutions; 3.2% and 6.0% for NaCl and sucrose solutions, and 2.9% and
6.1% for AD.
Fig. 3 shows the changes in normalized particle density (qp/qpo)
and normalized porosity (e/eo) against WR. Similar trends of bulk
and particle densities and porosity were observed with WL and
WR of samples. For NMC the observed trend was not as clear as
in the case of WL and WR (data not shown).
Particle density (Fig. 3a) increases slowly at the beginning of OD
with sucrose solutions, but above WR = 0.5 the increase is more
pronounced. No signicant differences are observed among the
process conditions tested. Particle density increases by the compositional change of the wet solid matrix during dehydration. Initially, water content is high, but during dehydration the content
of more dense substances (sucrose, cellulose) increases leading to
the increase of particle density. For air dried samples, the behaviour is similar; initially particle density increases very low and
above WR = 0.6 increases more pronouncedly up to a value around
40% higher at the end of the process. During OD of apple with sugar
solutions, a progressive increase of particle density along the process was observed (Nieto et al., 2004). For vegetables, the same
behaviour of particle density along drying was also observed
(Krokida and Maroulis, 1997).
The modelling of particle density, Eq. (7), during dehydration
can be performed from the composition of the material. The particle volume can be calculated from the masses and densities of each
component. For pumpkin parenchyma, fresh material is composed
by water, insoluble and soluble solids, and gas phase. So the particle volume (without the gas phase) can be dened as

V po V w V is V ss

13

In terms of masses of the components and densities, Eq. (13) can be

rewritten as

V po

mw

qw

mis

qis

mss

qss

14

For osmodehydrated pumpkin with sucrose and NaCl solutions,

the gained solids must be taken into account, so the particle volume can be dened, as

Vp

mw

qw

mis

qis

mss

qss

msuc

qsuc

mNaCl

qNaCl

15

Lozano et al. (1980) considered the insoluble solids of apple tissue as cellulose. In this work the same assumption was considered.
Soluble solids in pumpkin are mainly fructose and glucose in the
same proportion, so an average value of the densities of fructose
and glucose was taken as the density of initial soluble solids. Density values used in the calculations are shown in Table 3. Predicted
values of particle volume for fresh and dehydrated pumpkin fruits
with sucrose solutions and AD were compared with experimental
data obtained with the gas pycnometer. This prediction can be considered adequate, leading to a relative deviation of the predicted
values of 2.63% on average.
Finally, normalized particle density can be obtained by means
of Eq. (16)

qp 1  WR

qpo V p =V po

16

where Vpo and Vp are obtained by means of Eqs. (14) and (15),
respectively.
Average relative deviation (ARD) between experimental data
and predicted values obtained with Eq. (15) was 2.02%, indicating
that the model gives a good prediction of experimental data. Fig. 3a
shows predicted values of normalized particle density for OD with
60% sucrose solutions at 25 C and AD at 70 C.
Porosity of dehydrated pumpkin with sucrose solutions (Fig. 3b)
slightly decreases up to intermediate WR values (ca. 0.4); after that
point porosity increases till the end of the process. No effect of the
process conditions on porosity trend with WR or WL is observed.
For AD the behaviour is similar; at the beginning of the process
porosity uctuates but above WR = 0.5, porosity increases and at
the end of the process almost triplicates (e/eo = 2.8, WR = 0.95) its
initial value.
Mavroudis et al. (1998) observed an increase of porosity in
osmodehydrated apples at the end of the process. Giraldo et al.
(2003) showed that during OD of mango porosity of dehydrated
samples initially decreased and after that increased. The initial decrease of sample porosity can be explained by the fast initial
impregnation of the tissue with the osmotic solution, which penetrates into the external pores by capillary forces and other mass
transfer mechanisms. The accumulation of sucrose in the external
surface of the material generates a dense layer that hinders the further penetration of the osmotic solution and simultaneously minimizes the gas ow from the material to the solution. These
combined phenomena increase the food porosity.
Porosity during AD of foodstuffs can follow different behaviours
(Lozano et al., 1983; Krokida and Maroulis, 1997); in some cases it
decreases (sweet potato); in others it initially decreases and then
increases (pear) and still in other cases it increases during the
whole drying process (apple, banana). As commented for bulk density change, this different behaviour can be associated with the initial structural and compositional characteristics of the raw
material, as well as the process conditions.
Fig. 4 shows the changes in total volume, particle volume and
air volume during OD of pumpkin with 60% sucrose solutions
and AD. In the initial stage (up to WR = 0.5) in both treatments
the three volumes decrease during dehydration and the relative
decrease of air volume is the highest. The relative decrease of particle volume and total volume is practically the same for OD. For
AD, total volume decreases more than particle volume. Above
WR = 0.5, in OD the particle volume starts to decrease in percentage more than total volume, and air volume uctuates and practically remains constant until the end of the process. In the same
range for dried samples, the air phase volume remains constant
up to WR = 0.7 and then decreases until the end of drying; total

35

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

(a)

1.3

1.5
1.4

1.2

OD sucrose solutions
Air drying
Predicted, air drying

1.3
1.1

1.2
1.1

1.0
30% suc, 25C
60% suc, 25C
37.5% suc, 38C
52.5% suc, 38C

0.9

0.8
0.0

0.1

0.2

1.0

45% suc, 25C

37.5% suc, 12C
52.5% suc, 12C
Predicted, 60% suc

0.3

0.4

0.5

0.6

0.9

0.7

0.8
0.0

0.8

0.1

0.2

0.3

WR (kg/kg)

(b)

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.7

0.8

0.9

1.0

WR (kg/kg)

2.0

3.0

1.8

2.5

1.6
1.4

OD sucrose solutions
Air drying
Predicted, air drying

2.0

1.2
1.0

1.5

0.8

1.0

0.6
30% suc, 25C
60% suc, 25C
37.5% suc, 38C
52.5% suc, 38C

0.4
0.2
0.0
0.0

0.1

0.2

45% suc, 25C

37.5% suc, 12C
52.5% suc, 12C
Predicted 60% suc

0.3

0.4

0.5

0.6

0.5

0.7

0.0
0.0

0.8

0.1

0.2

WR (kg/kg)

0.3

0.4

0.5

0.6

WR (kg/kg)

Fig. 3. Changes in particle density (a) and porosity (b) during dehydration of pumpkin cylinders versus weight reduction. Left gures correspond to osmotic dehydration with
sucrose solutions at different process conditions, whereas gures on the right correspond to osmotic dehydration with sucrose solutions and air drying.

1.1

1.1
1.0
0.9

Total volume
Particle volume
Air volume

0.9

0.8

0.8

0.7

0.7

0.6

0.6

V/Vo

V/Vo

1.0

Total volume
Particle volume
Air volume

0.5

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1
0.0

0.0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

WR (kg/kg)

0.0

0.2

0.4

0.6

0.8

1.0

WR (kg/kg)

(a)

(b)

Fig. 4. Relative volume changes for total volume, particle volume and air volume during dehydration of pumpkin cylinders in (a) 60% sucrose solutions at 25 C and (b) air
drying at 70 C.

volume decreases less than particle volume like in OD samples. In

this way, it seems that the gas phase is better retained during OD
than in AD. As commented before, the formation of a dense layer of
osmotic agent in the material surface can be the cause of this
phenomenon.
Porosity, Eq. (2), can be predicted by means of Vp calculated
from Eqs. (14) and (15), and V with Eq. (11). ARD in the prediction
of porosity values was 14.5%. Fig. 3b shows experimental and pre-

dicted values of normalized porosity for air dried and osmotic

dehydrated pumpkin with 60% sucrose solutions.
3.3. Sample shape analysis
Fig. 5 shows the contour of cylinders dehydrated with sucrose
solutions (60%, 25 C), at different process times. It is observed a
size decrease during the process. The shape also changes during

36

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

1.4

Normalized shape factor

1.2

1.0

0.8

0.6

0.4

Elongation
Roundness
Compactness

0.2

0.0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

WR (kg/kg)
Fig. 7. Changes in shape factors during osmotic dehydration (60% sucrose, 25 C) of
pumpkin cylinders versus weight reduction.

1.2

diameter decrease was higher than the length decrease for both
products. In the case of potato cylinders, the authors suggested
that the high L/D ratio favoured the formation of an inner core
along the axis length maintaining the shape along this axis. In
the case of cauliower stems, the presence of oriented bres made
the product stiffer in a preferential orientation and the shrinkage
was more pronounced in the diameter. Based on those results,
two important factors can affect the shrinkage isotropicity: the
existence of preferential pathways of mass transfer (due to geometric and structural features) and the homogeneity of the material structure (due to structural features).
Fig. 7 shows the relative changes in the shape factors as a function of WR. The average initial values of the shape factors were
1.657, 0.670 and 0.749 for elongation, roundness and compactness,
respectively. Elongation slightly increases, whereas roundness and
compactness decrease during dehydration. It is often reported a
decrease of roundness during dehydration of foods, as in the case
of AD of apple discs (Mayor et al., 2005; Fernandez et al., 2005)
or apricot cubes (Riva et al., 2005). The tissue suffers deformations
as a consequence of the water removed in the material; in this way
roundness and compactness decrease during dehydration. Elongation increases mainly due to the corner effect that reduces the
minor axis length (minimum value of the diameter of the cylinder)
but maintains the major axis length (distance between two opposite corners of the cylinder).

1.0

4. Conclusions

Fig. 5. Changes in shape and size during osmotic dehydration of pumpkin cylinders
in 60% sucrose solutions at 25 C, at different process times (h). (a) 0, (b) 0.5, (c) 1,
(d) 3, (e) 6, and (f) 9. The horizontal line at the bottom of images corresponds to
2 mm.

Normalized dimension

dehydration; shrinkage is more accentuated at mid-length and

mid-thickness of the cylinder, whereas at the edges shrinkage is
less pronounced; this corner effect is clearly observed in the
most dehydrated cylinder (Fig. 5f). Del Valle et al. (1998) also reported this effect during OD of apple cylinders, and Mulet et al.
(2000) during AD of potato cubes.
As observed in Fig. 6, no signicant differences were found in
the decrease of length and diameter during the process, indicating
an isotropic shrinkage. Trujillo et al. (2007) found similar results
during the AD of beef meat discs (L/D  0.25). However, Mulet
et al. (2000) showed an anisotropic shrinkage in the AD of potato
(L/D  4.6) and cauliower stem (L/D  1.7) cylinders, where the

0.8

0.6

0.4

0.2

Diameter
Length

0.0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

WR (kg/kg)
Fig. 6. Relative changes in dimensions of osmodehydrated pumpkin cylinders (60%
sucrose, 25 C) versus weight reduction.

Shrinkage of samples was observed for all the dehydration

methods studied. Independently on the process conditions (concentration and temperature of the osmotic solution), volume decreased linearly with WL and WR during OD. Air dried samples
showed the same behaviour but shrinkage was more accentuated
for the same WL.
Bulk density varied in a restricted range during dehydration;
differences between maximum and minimum values were around
5% for OD with binary sucrose and NaCl solutions, 10% for ternary
NaCl/sucrose solutions and 13% for AD. For all the dehydration
methods, bulk density initially increased, then reached a maximum
value and after that decreased or uctuated till the end of the
process.
For all the treatments, particle density increased slowly at the
beginning of the process, and at certain WR value (0.5 for OD with
sucrose solutions and 0.6 for AD) the increase was more pronounced.

L. Mayor et al. / Journal of Food Engineering 103 (2011) 2937

The change in porosity also followed the same behaviour during

OD and AD. Initially, porosity decreased (up to WR = 0.4 and 0.5 for
OD and AD, respectively) and then porosity started to increase till
the end of the treatment, increasing twofold and threefold the
initial value for OD with sucrose solutions and AD, respectively.
The proposed models to evaluate shrinkage, bulk and particle
densities and porosity from WR were satisfactorily applied.
Experimental results indicated that shrinkage of pumpkin cylinders was isotropic. During dehydration, elongation increased and
roundness and compactness decreased.
Acknowledgement
The author Luis Mayor wishes to acknowledge SFRH/BD/3414/
2000 PhD Grant to Fundaao para a Cincia e a Tecnologia,
Portugal.
References
AOAC, 1984. Ofcial Methods of Analysis, 14th ed. Association of Ofcial Analytical
Chemists, Gaithersburg, MD.
Barat, J.M., Fito, P., Chiralt, A., 2001. Modeling of simultaneous mass transfer and
structural changes in fruit tissues. Journal of Food Engineering 49, 7785.
Bui, H.T., Makhlouf, J., Ratti, C., 2009. Osmotic dehydration of tomato in sucrose
solutions: Ficks law classical modelling. Journal of Food Science 74, E250E258.
Chenlo, F., Moreira, R., Pereira, G., Ampudia, A., 2002. Viscosities of aqueous
solutions of sucrose and sodium chloride of interest in osmotic dehydration
processes. Journal of Food Engineering 54, 347352.
Del Valle, J.M., Cuadros, T.R.M., Aguilera, J.M., 1998. Glass transitions and shrinkage
during drying and storage of osmosed apple pieces. Food Research International
31, 191204.
Doehlert, D.H., 1970. Uniform shell designs. Applied Statistics 19, 231239.
Fernandez, L., Castillero, C., Aguilera, J.M., 2005. An application of image analysis to
dehydration of apple discs. Journal of Food Engineering 67, 185193.
Giraldo, G., Talens, P., Fito, P., Chiralt, A., 2003. Inuence of sucrose solution
concentration on kinetics and yield during osmotic dehydration of mango.
Journal of Food Engineering 58, 3343.
Khallou, S., Almeida-Rivera, C., Bongers, P., 2009. A theoretical model and its
experimental validation to predict the porosity as a function of shrinkage and
collapse phenomena during drying. Food Research International 42, 11221130.
Konopacka, D., Seroczynska, A., Korzeniewska, A., Jesionkowska, K., NiemirowiczSzczytt, K., Plocharski, W., 2010. Studies on the usefulness of Cucurbita maxima
for the production of ready-to-eat dried vegetable snacks with a high
carotenoid content. LWT Food Science and Technology 43, 302309.
Kowalska, H., Lenart, A., Leszczyk, D., 2008. The effect of blanching and freezing on
osmotic dehydration of pumpkin. Journal of Food Engineering 86, 3038.
Krokida, M.K., Maroulis, Z.B., 1997. Effect of drying method on shrinkage and
porosity. Drying Technology 15, 24412458.
Lewicki, P.P., Lukaszuk, A., 2000. Effect of osmotic dewatering on rheological
properties of apple subjected to convective drying. Journal of Food Engineering
45, 119126.
Lewicki, P.P., Pawlak, G., 2003. Effect of drying on microstructure of plant tissue.
Drying Technology 21, 657683.
Lide, D.R., 2005. CRC Handbook of Chemistry and Physics, 85th ed. CRC Press, Boca
Raton, FL.
Lozano, J.E., Rotstein, E., Urbicain, M.J., 1980. Total porosity and open-pore porosity
in the drying of fruits. Journal of Food Science 45, 14031407.
Lozano, J.E., Rotstein, E., Urbicain, M.J., 1983. Shrinkage, porosity and bulk density of
foodstuffs at changing moisture contents. Journal of Food Science 48, 1497
1502 (1553).

37

Mavroudis, N.E., Gekas, V., Sjholm, I., 1998. Osmotic dehydration of apples.
Shrinkage phenomena and the signicance of initial structure on mass transfer
rates. Journal of Food Engineering 38, 101123.
Mayor, L., Sereno, A.M., 2004. Modelling shrinkage during convective drying of food
materials: a review. Journal of Food Engineering 61, 373386.
Mayor, L., Silva, M.A., Sereno, A.M., 2005. Microstructural changes during drying of
apple slices. Drying Technology 23, 22612276.
Mayor, L., Moreira, R., Chenlo, F., Sereno, A.M., 2006. Kinetics of osmotic
dehydration of pumpkin with sodium chloride solutions. Journal of Food
Engineering 74, 253262.
Mayor, L., Moreira, R., Chenlo, F., Sereno, A.M., 2007. Osmotic dehydration kinetics
of pumpkin fruits using ternary solutions of sodium chloride and sucrose.
Drying Technology 25, 17491758.
McLaughlin, C.P., Magee, T.R., 1998. The determination of sorption isotherm and
isosteric heats of sorption of potatoes. Journal of Food Engineering 35, 267280.
McMinn, W.A.M., Magee, T.R.A., 1997. Physical characteristics of dehydrated
potatoes part II. Journal of Food Engineering 33, 4955.
Moreira, R., Sereno, A.M., 2003. Evaluation of mass transfer coefcients and
volumetric shrinkage during osmotic dehydration of apple using sucrose
solutions in static and non-static conditions. Journal of Food Engineering 57,
2531.
Mulet, A., Garcia-Reverter, J., Bon, J., Berna, A., 2000. Effect of shape on potato and
cauliower shrinkage during drying. Drying Technology 18, 12011219.
Nieto, A.B., Salvatori, D.M., Castro, M.A., Alzamora, S.M., 2004. Structural changes in
apple tissue during glucose and sucrose osmotic dehydration: shrinkage,
porosity density and microscopic features. Journal of Food Engineering 61,
269278.
Panyawong, S., Devahastin, S., 2007. Determination of deformation of a food
product undergoing different drying methods and conditions via evolution of a
shape factor. Journal of Food Engineering 78, 151161.
Perry, R.H., Green, D.W., 1999. Perrys Chemical Engineers Handbook, seventh ed.
McGraw-Hill Companies, New York.
Prothon, F., Ahrne, L., Sjoholm, I., 2003. Mechanisms and prevention of plant tissue
collapse during dehydration: a critical review. Critical Reviews in Food Science
and Nutrition 43, 447479.
Rahman, M.S., 2001. Toward prediction of porosity in food during drying: a brief
review. Drying Technology 19, 113.
Rahman, M.S., 2005. Mass-volume-area-related properties of foods. In: Rao, M.A.,
Rizvi, S.S.H., Datta, A.K. (Eds.), Engineering Properties of Foods. Taylor and
Francis Group, Boca Raton, FL.
Riva, M., Campolongo, S., Leva, A.A., Maestrelli, A., Torreggiani, D., 2005. Structure
property relationships in osmo-air-dehydrated apricot cubes. Food Research
International 38, 533542.
Sereno, A.M., Silva, M.A., Mayor, L., 2007. Determination of particle density and
porosity in foods and porous materials with high moisture content.
International Journal of Food Properties 10, 455469.
Shimada, Y., Roos, Y., Karel, M., 1991. Oxidation of methyl linoleate encapsulated in
amorphous lactose based food model. Journal of Agricultural and Food
Chemistry 39, 637641.
Simal, S., Rossello, C., Berna, A., Mulet, A., 1998. Drying of shrinking cylinder-shaped
bodies. Journal of Food Engineering 37, 423435.
Souza, J.S., Medeiros, M.F.D., Magalhaes, M.M.A., Rodrigues, S., Fernandes, F.A.N.,
2007. Optimization of osmotic dehydration of tomatoes in a ternary system
followed by air-drying. Journal of Food Engineering 83, 501509.
Trujillo, F.J., Wiangkaew, C., Pham, Q.T., 2007. Drying modeling and water diffusivity
in beef meat. Journal of Food Engineering 78, 7485.
Viberg, U., Freuler, S., Gekas, V., Sjholm, I., 1998. Osmotic pretreatment of
strawberries and shrinkage effects. Journal of Food Engineering 35, 135145.
White, K.L., Bell, L.N., 1999. Glucose loss and Maillard browning in solids as affected
by porosity and collapse. Journal of Food Science 64 (6), 10101014.
Zogzas, N.P., Maroulis, Z.B., Marinos-Kouris, D., 1994. Densities, shrinkage and
porosity of some vegetables during air drying. Drying Technology 12, 1653
1666.