Drying kinetics of apple cylinders under combined

hot air–microwave dehydration

Ana Andr

ees

*

, Cristina Bilbao, Pedro Fito

Department of Food Technology, Polytechnic University of Valencia, Camino de Vera s/n, 46022 Valencia, Spain

Received 6 April 2003; accepted 21 July 2003

Abstract

Apple cylinders (Granny Smith) were dried in a combined hot air–microwave system. Drying experiments were carried out at

various air temperatures (25, 30, 40 and 50

°C) combined with different levels of microwave incident power (0, 3, 5, 7 and 10 W/g)

until 0.11 (d.b.) moisture content was obtained. Vacuum impregnation with an isotonic solution was used as a pretreatment before
drying. Sample weight changes were determined for different drying conditions. When drying with combined air–microwave system,
the drying rate curve could be divided in four periods limited by four critical points. An empirical model was proposed to estimate
the drying kinetic constants as a function of the air temperature and the microwave power level for both sorts of samples, fresh
apples and impregnated apples. At the end of the process different tissue matrices were obtained depending on the different drying
conditions.
Ó 2003 Elsevier Ltd. All rights reserved.

Keywords: Apple; Microwave; Kinetic; Dehydration; Vacuum impregnation

1. Introduction

It is usual to combine hot air with the microwave

system when drying with microwave energy. Hot air is,
by itself, relatively efficient at removing free water at or
near the surface, whereas the unique pumping action of
microwaves energy provides an efficient way of remov-
ing internal free water (Schiffmann, 1995). Combining
properly both unit operations in a unique way, it is
possible to improve the efficiency and the economics of
the drying process.

There are three ways in which microwave energy may

be combined with hot air drying (Schiffmann, 1995):

1. By applying the microwave energy at the beginning of

the dehydration process. The interior of the sample is
quickly heated to evaporation temperature; the vapor
is forced outwards permitting the hot air to remove
water from the surface. The improved drying rate is
ascribed to the creation of a porous structure of the
food material (puffing), which facilitates the transport

of the water vapor (Drouzas, Tsami, & Saravacos,
1999).

2. It is also possible to apply microwave energy when

the drying rate begins to fall. In this case the surface
of the material is dry, and moisture is concentrated in
the center. Applying microwaves at this moment, the
generation of internal heat and therefore vapor pres-
sure force the moisture to the surface, where it is
readily removed.

3. Finally, it has also been suggested that microwave en-

ergy should be applied in the falling rate period or at
low moisture content to finish drying when it is the
least efficient portion of a conventional drying system.
Taking into account that samples suffer shrinkage of
the structure when dried with air, which restricts diffu-
sion and cause decrease in the drying rate, explains the
length of the time required. The outward flux of vapor
can help to prevent the shrinkage of tissue structure
(Maskan, 2000). In some cases applying microwave
drying in the last stage of the dehydration process
can also be very efficient in removing bound water
from the product. With conventional heating, it is nec-
essary to expose the product to high temperature in
order to break the bonds that keep the water absorbed
in the material.

Journal of Food Engineering 63 (2004) 71–78

www.elsevier.com/locate/jfoodeng

*

Corresponding author. Fax: [phone].

E-mail address:

[e-mail]

(A. Andr

ees).

0260-8774/$ - see front matter

Ó 2003 Elsevier Ltd. All rights reserved.

doi:10.1016/S0260-8774(03)00284-X

Several food products have been successfully dried

with the employment of microwave energy combined
with hot air (Drouzas et al., 1999; Funebo & Ohlsson,
1998; Khraisheh, Cooper, & Magee, 1995). Some au-
thors propose the use of both systems applying first
conventional hot-air drying followed by microwaves for
final drying (Cheng, Ye, & Chen, 1999; Litvin, Mann-
heim, & Miltz, 1998; Maskan, 2000). There are also
some researches who have shown that pretreatment of
food materials with microwave energy can improve the
air drying rate (Funebo, Ahrn

ee, Kidman, Langton, &

Skj€

o

oldebrand, 2000).

In this study, microwave energy and air have been

simultaneously applied during the whole drying process.
The purpose was to better understand the effect of both
heating systems in the drying process of apple cylinders.
An empirical model was proposed to assess the effect of
selected parameters such as drying temperature and
microwave power.

2. Materials and methods

Apples (Granny Smith) were used as test material.

Samples were obtained by cutting the apples into cylin-
drical pieces with a diameter of 20 mm and a height of 20
mm. The axis of each cylinder was parallel to the main
axis of the apple. Half of the samples were vacuum im-
pregnated prior drying. Impregnated samples were pre-
pared by immersing them in apple juice (isotonic

solution) at 50 mbar pressure for 10 min plus 10 min
more at atmospheric pressure (Fito & Pastor, 1994). This
pretreatment was carried out in order to replace the
productÕs internal gas by the impregnation solution
(Salvatori, Andr

ees, Chiralt, & Fito, 1998) and to study

the influence of these structural and compositional
changes on drying kinetics and volume changes during
drying.

The drying experiments were performed in a specially

designed hot air–microwave oven equipped with con-
tinuous output-power microwave energy (Fig. 1). Each
time the oven temperature and the incident microwave
power were prefixed. The microwave power levels were
set at 0, 3, 5, 7 and 10 W/g combined with air at 25, 30,
40 and 50

°C. Air velocity was set at 1.0 ± 0.1 m/s in all

cases and the relative humidity of the ambient was
62 ± 8%. Sample was suspended from a balance and the
kinetics of the drying were studied by continuous
weighing the sample during drying process.

The initial moisture content of fresh apples was de-

termined with an A&D Model AE100 Infra Red balance
varying between 4.95 and 7.13 (d.b.). The resulting dry-
ing curves were modelled and the kinetic parameters
obtained were related with process variables (air tempera-
ture, microwave power and vacuum impregnation pre-
treatment) by multiple regression analysis performed with
the Statgraphics Plus 4 Software (Statgraphics, 1998).

In order to evaluate volume changes, samples were

photographed before and after the drying process. The
height and the diameter were estimated using the

Nomenclature

x

w

moisture content (wet basis)

X

w

moisture content (dry basis)

X

0

w

initial moisture content (dry basis)

dX

w

=

dt drying rate (g water/g dry mater s)

e

porosity

q

r

solid density (kg/m

3

)

q

a

bulk density (kg/m

3

)

D

diffusivity (m

2

/s)

r

radius of the cylinder (m)

l

height of the cylinder (m)

t

time (s)

V

f

=V

0

residual volume (dimensionless)

VI

vacuum impregnated samples

No VI non-vacuum impregnated samples
T

(

°C)–MW combined hot air (°C)–microwave power

P

microwave power (W/g)

T

temperature (

°C)

Fig. 1. Schematic illustration of the combined hot air–microwave drying equipment.

72

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78

ADOBE PHOTOSHOP 5.0.2. program. An Anova was
performed to compare volume changes under different
process variables.

2.1. Cryo-SEM observations

Some samples were dried until 0.11 (d.b.) by applying

different drying conditions in order to observe the in-
fluence of process variables on microstructure changes.
Each sample was frozen in slush nitrogen and attached
to the specimen holder of a CT-1000C Cryo-transfer
system (Oxford Instruments, Oxford, UK) interfaced
with a JEOL JSM-5410 scanning electron microscope
(SEM). The sample was then fractured and transferred
from cryostage to the microscope sample stage, where
the condensed surface water was sublimed by controlled
warming to

)85 °C. Afterwards, the sample was trans-

ferred again to the cryostage in order to gold coat it
by sputtering. Finally the sample was put back to the
microscope sample stage to be viewed at an accelerating
voltage of 15 keV and at different magnifications.

3. Results and discussion

Fig. 2 shows the experimental drying rate curves for

all experimental conditions, where the great influence of
microwave energy can be observed (Khraisheh et al.,
1995; Maskan, 2000). The drying rate curves obtained
using only hot air show a falling rate period and the
drying rate was slower for vacuum impregnated samples
due to the greater amount of liquid phase in these
samples (Fito et al., 2001). However, the shape of the
drying curves changes dramatically when combined
conditions (hot air–microwave) are used. The typical
pathway of these curves can be generalized as shown in
Fig. 3 where four drying periods can be observed:

Period 0: This first period can be considered a

warming up step where the temperature rises quickly
without significant mass losses.

Period I: It can be assumed that liquid water flows

from the inner to the cylinder surface where it is re-
moved by evaporation due to convective heat transfer.
The length of this period is quite different comparing
impregnated and non-impregnated samples curves,
probably due to the higher amount of liquid phase in the
impregnated samples and because free water probably
has a higher value of the diffusion coefficient compared
with the water included in the plasmalemma of the cells.
The evaporation of free water will contribute to the
cooling of the sample and consequently a slight reduc-
tion of the drying rate is observed during this period in
impregnated samples.

Period II: It is assumed that when the sample tem-

perature exceeds the boiling point the sudden vapor-
ization of liquid water is promoted and the drying rate

significantly increases reaching a maximum (point 3).
For lower microwave power, the boiling temperature is
not reached and the acceleration of the drying rate can
be explained by the simultaneous action of the temper-
ature and moisture content gradients (Constant, Moyne,
& Perre, 1996).

Period III is characterized in all cases by a linear

decrease of the drying rate with the moisture content of
the sample.

Period IV: This final step has the typical pathway of

those processes where mass transfer is explained by
diffusional mechanisms. Internal diffusion was assumed
to be the mechanism responsible for the lost of water
during Period IV of the dehydration process and the
water diffusivity (D) can be calculated from the experi-
mental data using a solution of FickÕs second law.

The periods described above are defined by four

critical points (Fig. 3) which location depends on the
process variables (temperature and microwave power).
These critical points were identified in all the drying
curves and a multiple regression analysis was performed
to relate their coordinates with the process variables
(Table 1). In all cases the most relevant variable that
affects the location of the critical points is the microwave
power. On the other hand, the analysis of variance
showed that air temperature did not have a significant
effect at the 95% confidence level on the drying rate for
vacuum impregnated samples. However the drying rate
at the critical points for non-impregnated samples in-
creased with the drying temperature.

The equations proposed to model the drying kinetics

for the periods mentioned above are presented in Table
2. The multiple regression analysis of the parameters
obtained yielded the equations to predict them as a
function of process variables (Table 2). It is observed for
the whole drying process that an increase in the intensity
of microwaves resulted in acceleration of the drying rate
of the apple cylinders. However, no large differences were
observed due to the air temperature until the last period
(p > 0:05). As a consequence of the above mentioned
observations, rapid drying can be achieved for high levels
of microwave power while air temperature will not
greatly facilitate drying until the last falling rate period.

All these equations establish an empirical model that

allows predicting experimental drying rate curves for
samples dried under a combined hot air–microwave
system, in the interval of temperatures and microwave
power tested in this work. A close agreement between
experimental data and predicted values was observed as
in the example illustrated in Fig. 4.

3.1. Changes in sample volume

Volume changes were compared and the analysis

of variance showed that air temperature did not have
a significant effect at the 95% confidence level. The

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78

73

influence of microwave power and vacuum impregna-
tion pretreatment on the residual volume is shown in
Fig. 5. Comparison between both sorts of samples
shows that greater volume reduction was promoted in

impregnated samples (76.1%) than in non-impregnated
samples (67.9%). The slower drying kinetics of impreg-
nated samples can be the reason for its greater volume
reduction, as mass transfer and deformation-relaxation

VI-25˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

VI-30˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

VI-40˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

VI-50˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

No VI-25˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

No VI-30˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

No VI-50˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

No VI-40˚C-MW

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

0 W/g

3 W/g

5 W/g

7 W/g

10 W/g

Fig. 2. Drying rate curves for apple cylinders, impregnated (VI) and non-impregnated (No VI), for different combinations of air temperature and
microwave power.

74

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78

are coupled phenomena. In addition, a linear relation-
ship with microwave output level was found. Impreg-
nated samples showed a higher residual volume as
microwave level increased, this means that dried im-

pregnated samples are smaller for the apples dehydrated
in air only, than for microwave treated apples. When
drying non-impregnated samples, the relationship was
also linear but in the opposite tendence, the volume
decreased as microwave level increased.

These results could be explained considering the

values of the porosity. The porosity of the samples (e)
was calculated from Eq. (1) in terms of bulk density (q

r

)

and apparent density (q

a

). For the estimation of the bulk

density Eq. (2) was applied as a function of the mass
fraction of water in the sample. Bulk density was cal-
culated assuming the vegetal tissue as constituted by
water and carbohydrates.

e

¼

q

r

q

a

q

r

ð1Þ

q

r

¼

1

x

w

1000

þ

ð1 x

w

Þ

1590

ð2Þ

IV

III

I

1

3

2

4

II

0

dX

w

/dt

X

w

/ X

0

w

Fig. 3. Drying rate curve shape for apple cylinders under combined hot
air–microwave dehydration.

Table 1
Empirical equations to predict the location of the critical points that define the periods in the drying rate curves

VI

No VI

Point 1: lower border for Period I
dX

w

dt

¼ 0:054 þ 0:014 P

dX

w

dt

¼ 0:045 þ 0:003 T þ 0:010 P

Point 2: upper border for Period I and lower border for Period II
dX

w

dt

¼ 0:001 þ 0:011 P

dX

w

dt

¼ 0:001 þ 0:011 P

Point 3: upper border for Period II and lower border for Period III
dX

w

dt

¼ 0:089 þ 0:050 P

dX

w

dt

¼ 0:241 þ 0:006 T þ 0:052 P

Point 4: upper border for Period III and lower border for Period VI
dX

w

dt

¼ 0:017 þ 0:011 P

dX

w

dt

¼ 0:017 þ 0:011 P

Table 2
Predicted equations of the drying rate for apple cylinders dehydrated with forced air combined with microwaves

Period

Equations

Parameters

I

a

dX

w

dt

¼ a

1

X

w

X

wo

b

1

a

1

¼ 0:008 P þ 0:123

b

1

¼ 0:042, P 2 ½3–7 W=g

b

1

¼ 0:013, P ¼ 10 W/g

II

dX

w

dt

¼

b

2

X

w

X

wo

a

2

X

w

a

2

¼ 1:89

1

P

þ 1:43 (No VI)

a

2

¼ 9:13

1

P

þ 1:21 (VI)

b

2

¼ 0:089 (No VI)

b

2

¼ 0:060 (VI)

III

dX

w

dt

¼ a

3

X

w

b

3

a

3

¼ 0:056 þ 0:025 VI þ 0:001 T þ 0:024 P

b

3

¼ 0:042 P 0:0365

IV

dX

w

dt

¼ f ðD; r; l; tÞ

b

D

¼ 1:657 0:562 VI

c

+ 0:089

T þ 0:497 P

b

4

¼ 0:687 (No VI)

b

4

¼ 0:944 (VI)

b

4

¼ 0 (air drying)

a

Only for vacuum impregnated samples.

b dX

w

dt

¼ X

wo

D

p

2

4l

2

þ

5:783

r

2

exp

D

p

2

4l

2

þ

5:783

r

2

t

þ b

4

.

c

VI takes value 0 for non-impregnated samples and 1 for impregnated samples.

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78

75

Fig. 6 shows apple sample porosity, for pretreated
samples and for non-pretreated samples, for samples
with a moisture content about 0.11 (d.b.), as a function
of the process conditions. Higher porosity was obtained
for non-impregnated samples (71%) than for impreg-
nated ones (42%). Vacuum impregnation with isotonic
solutions modifies the pathway of sample porosity
change along the drying process, as sample porosity
reduction occurs when a vacuum pulse is applied at the
beginning of the impregnation process. In this opera-
tion, as mentioned above, impregnation solution is in-
troduced in the fruit pores replacing the initial gas (Fito,
Chiralt, Barat, & Martıınez-Monz

o

o, 2002).

In Fig. 6, it is also observed that the porosity

values for non-impregnated samples only decreased
significantly when high microwave power was applied
(10 W/g). However, for impregnated samples, the
porosity values increased as microwave power in-
creased, maybe related with a ‘‘puffing effect’’, the
water vapor generated inside the material that could
produce a porous structure (Drouzas et al., 1999).

3.2. Microstructure changes

Fig. 7 shows micrographics of apple tissue dried by

combining air (40

°C) with different levels of microwave

incident power. It is observed that air dried samples
(Fig. 7A) present a porous structure, and cell walls are
greatly shrunk, which leaves wide spaces between
neighbouring cells (Bilbao, Albors, Gras, Andr

ees, &

Fito, 2000). Samples dried at low microwave power
(Fig. 7B) also reveal a porous structure but these pores
are much smaller. Finally, the sample dried at higher
microwave power contained a hollow center that could
be macroscopically observed. In the micrographics of
these samples (Fig. 7C), two zones can be distinguished,
a porous tissue corresponding to the outer zone where
the effect of microwave heating occurred to a lesser ex-
tent and an inner zone, next to the hole, where the tissue
appeared more compact, due to the combined effect of
the high temperatures (that promote cell membranes
denaturation and phase transitions), and the pressure

y = 0.8659x + 19.593

R

2

= 0.948

y = -0.6602x + 35.421

R

2

= 0.9379

0

5

10

15

20

25

30

35

40

45

50

0

1

2

3

4

5

6

7

8

9

10

P (W/g)

V

f

/V

o

(%)

No VI

VI

Fig. 5. Relationship between residual volume and microwave incident
power.

No VI

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

observed

predicted

VI

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.0

0.2

0.4

0.6

0.8

1.0

X

w

/X

w

o

dX

w

/dt

observed

predicted

Fig. 4. Observed drying rate curve vs predicted one for apple cylinders dried at 40

°C air temperature assisted with 7 W/g incident microwave power.

No VI

P (W/g)

Porosity

0

3

5

7

10

0.66

0.68

0.7

0.72

0.74

VI

P (W/g)

Porosity

0

3

5

7

10

0.42

0.46

0.5

0.54

0.58

0.62

0.66

Fig. 6. Means and 95.0% LSD intervals of porosity of apple cylinders dried until 0.11 (d.b.) water content as a function of the incident power.

76

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78

gradient created as a consequence of the vapor genera-
tion inside the material.

Vacuum impregnation treatment causes differences in

the structural features as can be observed in Fig. 8. It is
difficult to differentiate between intra and extra cellular
volume in vacuum impregnated samples dried without
microwave power since both are full of water and solids
vitrified together due to the high solids concentration
(Fig. 8A). However, cell limits can be identified since cell
walls appear in a brighter colour, and cells appear
greatly shrunk in irregular forms. Impregnated samples
dried with the aid of microwaves also show a continuous
aspect of the cellular tissue interrupted by the formation
of some pores during the drying process and holes or
cracks due to the disruption of the tissue (Fig. 8B and

C). These microscopic observations are in accordance
with the tendency observed in porosity values and re-
sidual volume mentioned above.

4. Conclusions

Drying kinetic in combined hot air–microwave sys-

tems is affected by process variables as microwave power
and air temperature and product variables as density and
porosity affected by vacuum impregnation pretreatments.
Microwave power effect was higher than air temperature,
decreasing significantly the drying time. The higher den-
sity and lower porosity of vacuum impregnated samples
implied slower kinetics and higher volume reduction as a

Fig. 7. Apple tissue dried by combining air with different levels of
microwave incident power for No VI samples.

Fig. 8. Apple tissue dried by combining air with different levels of
microwave incident power for VI samples.

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78

77

consequence of the coupling of mass transfer and defor-
mation-relaxation phenomena. An empirical model was
proposed to predict drying rate curves taking into ac-
count the studied variables and a good agreement is
obtained when experimental and predicted data are com-
pared. The tissue characteristics observed in the micro-
graphics point out that process variable not only affect
the drying kinetics, they also lead to different macro and
microstructure of the final product.

References

Bilbao, C., Albors, A., Gras, M., Andr

ees, A., & Fito, P. (2000).

Shrinkage during apple tissue air-drying: macro and microstruc-
tural changes. In Proceedings of the 12th international drying
symposium. Elsevier Sciences, IDS2000 (ISBN 0-444-50422-2).

Cheng, W., Ye, Y., & Chen, Z. (1999). Microwave drying of foods with

high humidity. Microwave and Optical Technology Letters, 22(3,
August 5), 205–207.

Constant, T., Moyne, C., & Perre, P. (1996). Drying with internal heat

generation: theoretical aspects and application to microwave
heating. AIChE Journal, 42(2), 359–368.

Drouzas, A. E., Tsami, E., & Saravacos, G. D. (1999). Microwave/

vacuum drying of model fruit gels. Journal of Food Engineering, 39,
117–122.

Fito, P., Chiralt, A., Barat, J. M., Andr

ees, A., Martıınez-Monz

o

o, J., &

Martıınez-Navarrete, N. (2001). Vacuum impregnation for devel-
opment of new dehydrated products. Journal of Food Engineering,
49(4), 297–302.

Fito, P., Chiralt, A., Barat, J. M., & Martıınez-Monz

o

o, J. (2002). Mass

transport and deformation relaxation phenomena in plant tissues.
In J. Welti-Chanes, G. Barbosa-C

a

anovas, & J. M. Aguilera (Eds.),

Engineering and food for the 21st century (pp. 235–252). Lancaster:
CRC Press.

Fito, P., & Pastor, R. (1994). On some non-diffusional mechanism

occurring during vacuum osmotic dehydration. Journal of Food
Engineering, 21, 513–519.

Funebo, T., Ahrn

ee, L., Kidman, S., Langton, M., & Skj€

o

oldebrand, C.

(2000). Microwave heat treatment of apple before air dehydra-
tion––Effects on physical properties and structure. Journal of Food
Engineering, 46, 173–182.

Funebo, T., & Ohlsson, T. (1998). Microwave-assisted air dehydration

of apple and mushroom. Journal of Food Engineering, 38(3), 353–
367.

Khraisheh, M. A. M., Cooper, T. J. R, & Magee, T. R. A. (1995).

Investigation and modelling of combined microwave and air
drying. Transactions of the Institution of Chemical Engineers,
73(C), 121–126.

Litvin, S., Mannheim, C. H., & Miltz, J. (1998). Dehydration of

carrots by a combination of freeze drying, microwave heating and
air or vacuum drying. Journal of Food Engineering, 36, 103–111.

Maskan, M. (2000). Microwave/air and microwave finish drying of

banana. Journal of Food Engineering, 44, 71–78.

Salvatori, D., Andr

ees, A., Chiralt, A., & Fito, P. (1998). The response

of some properties of fruits to vacuum impregnation. Journal of
Food Process Engineering, 21, 59–73.

Schiffmann, R. F. (1995). Microwave and dielectric drying. In A. S.

Mujumdar (Ed.), Handbook of industrial drying-1 (pp. 345–372).
New York: Marcel Dekker Inc.

Statgraphics (1998). Statgraphics plus for Windows. Rockville, MD:

Manugistics Inc.

78

A. Andr

ees et al. / Journal of Food Engineering 63 (2004) 71–78