Vol.3, No.1B, 1-4 (2014) Journal of Agricultural Chemistry and Environment
http://dx.doi.org/10.4236/jacen.2014.31B001
Cassava slices drying by using a combined hot-air
single-plane microwave dryer
Patomsok Wilaipon
Department of Mechanical Engineering, Faculty of Engineering, Naresuan University, Phitsanulok, Thailand;
Received October 2013
ABSTRACT
Characteristics of cassava drying were investi-
gated by using a microwave hot-air drying sys-
tem. Two waveguides were installed on a single
plane of the microwave cavity. The drying expe-
riments w ere carried out at two levels of sample
surface temperature set-points, 70˚C and 80˚C
respectively. Cassava (Rayong-9) with 2.5 kg
weight and 61% moisture content on wet basis
was dried in the dryer for about 300 - 340 mi-
nutes until the final values of moisture content
of about 20% db were achieved. It was found
that the drying time decreased with an increase
in sample-surface temperature set point. Ap-
proximately 87% of the moisture was removed
during the drying period. It was found that there
was a rapid decrease in moisture ratio values
followed by the gradual decline period in all ex-
periments. With regard to drying kinetics, 5
commonly used mathematical models were ex-
amined w ith the experimental data. It was found
that Page’s and diffusion models provided a
good agreement between the experimental and
predicted moisture ratio values for all tempera-
ture set-points. The regression results indicated
that highest values of coefficient of determina-
tion and adjusted coefficient of determination as
well as low est value of standard error of estima-
tion were reported for the case of Page’s model
at 80˚C temperature set-point.
KEYWORDS
Micro wave Dry ing; Cassava; Dryin g
1. INTRODUCTION
Thermal drying may be defined as the process of ther-
mally removing moisture to yield a solid product. Two
processes simultaneously take place during thermal dry-
ing. These processes are the evaporation of surface mois-
ture and the internal moisture transferring of the drying
product. The energy from the surrounding environment,
as a result of convection, conduction, radiation, or the
combination of these effects, is transferred to the drying
product to evaporate the surface moisture. In addition,
the movement of internal moisture to the surface of the
drying product may occur through several mechanisms
such as diffusion, capillary effect as well as an increase
in internal pressure of the material.
According to the conventional hot-air dryer and solar
drying, several kinds of materials have been investigated
with regard to their drying characteristics. The effect of
drying time on chili-pepper moisture ratio was studied by
a researcher from Nigeria [1]. Four drying models viz.
Newton, Henderson and Pabis, Page, and Logarithmic
models were fitted in the experimental data by using re-
gression technique. The solar drying characteristics of 3-
shapes strawberry, whole, halves quarter and 3 mm discs,
were investigated by Egyptian researchers [2]. Moreover,
the effect of drying temperature, 50˚C to 80 ˚C, on dr ying
rates of apple slice was studied [3]. Several thin-layer
drying equations were fitted to the experimental drying
data in order to examine the most appropriate equation.
For the case of banana drying, empirical and diffusion
models were utilized to describe the characteristics of
intermittent and continuous solar drying [4]. A least
square method was used for drying model fitting in order
to minimize the standard error between the experimental
data and the calculated values.
One of the promising drying techniques is microwave
method, which is considerably different from the conven-
tional drying. The electromagnetic field in microwave
drying interacts with the drying material as a whole
while the hot-air drying depends on the rate of heat prop-
agation from higher-temperature material surface to the
inside. The researchers from Turkey investigated the dry-
ing characteristics of spinach using 8 microwave power
levels. It was found that the drying process was com-
pleted between 290 to 4005 s depending on the value of
microwave power [5]. Microwave drying in combination
with hot air drying was also used for pumpkin slices

P. Wilaipon / Journal of Agricultur al C hemistry and Environment 3 (2014) 1-4
2
drying [6]. Drying periods for the case of microwave, hot
air, and combined microwave-air drying methods were
studied. It was reported that the latter was accounted for
the shortest drying period.
The aim of this study was to evaluate the drying cha-
racteristics of combined microwave-air method for the
case of cassava. Furthermore, the mathematical model
parameters were also calculated by using regression
technique.
2. MATERIAL AND METHOD
2.1. Material
Rayong-9 cassavas with an initial moisture content of
61% on wet basis were obtained from a local factory in
Phitsanulok, Thailand. T heir initial moisture content val-
ue was examined, according to ASAE S358.2 DEC99
standard, by using a cabinet hot-air dryer (Memmert 600,
30˚C - 350˚C, 2400 W) and a digital balance (accuracy
0.001 g). Then, the material was cut into 10 mm thick
and 25 - 50 mm diameter with the cutting machine. All
cassavas used in the experiment were from the same
batch.
2.2. Drying Experiment and Data Analysis
The drying s ystem was comprised of two 86 × 43 mm
rectangular waveguides, two air-cooled magnetrons, and
a 44 × 51 × 93 cm cavity. Two 800 W-magnetrons used
in the experiments work at the frequency of 2.45 GHz.
They were installed in the waveguides mounted on the
same plane, the top of the cavity. Four heaters, 2 kW
each, were installed at the air inlet duct. A temperature
controller (Shimax MAC5D) and type K thermocouple
were utilized for temperature control purpose. In order to
record the sample weight loss, a 15 kg single-point loa d-
cell coupled to a load cell indicator (Primus CM 013)
was installed on the top of the cavity. Additionally, A
Testo 435, accuracy ±0.25˚C and ±2% RH, was used for
measuring the temperature and relative humidity of inlet
air.
In all experiments, approximately 2.5 kg of samples
were used. The samples were uniformly spread on a dry-
ing tray and placed in the drying cavity. The temperature
and velocity of hot air were set at 60 ˚C and 1 m/s respec-
tively. A temperature sensor was utilized for measuring
surface temperature of the sample. It was used as an in-
put for microwave power operation control. The experi-
ments were investigated at two levels of sample temper-
ature viz. 70˚C and 80˚C respectively.
The values of moisture ratio were calculated using the
following equation:
() ()
te ie
MRMMM M=−−
(1)
where :
MR is the moisture ratio;
Mt is the moisture content at 1 hour (%);
Me is the equilibrium moisture content (%);
Mi is the initial moisture content (%).
Several conventional drying models have been pro-
posed for determining the moisture ratio as a function of
drying time. In this research, the drying models of cas-
sava drying by using 2 planes magnetron microwave-air
drying system were investigated. Newton model [7],
Page’s model [8], logarithmic model [9], Henderson &
Pabis model [10], and diffusion model [3] were applied
to describe the characteristics of cassava drying.
( )
MR exp kt= −
(2)
( )
n
MR exp kt=−
(3)
( )()
MRa expktb= −+
(4)
( )()
MRa expkt= −
(5)
() ()() ()
MRa expkt1aexpktb=− +−−
(6)
where :
k is the drying constant;
n is the power parameter;
a and b are parameters;
t is drying time (hour).
Coefficient of determination (R2), adjusted coefficient
of determination
( )
2
adjusted
R
, and standard error of esti-
mation (SEE) were utilized to evaluate the goodness of
fit of the tested drying models to the experimental data.
3. RESULT AND DISCUSSION
Effective mathematical model of drying characteristic
is crucial for cassava microwave-air drying kinetics in-
vestigation. The combination of microwave and hot-air
energy were able to reduce the sample moisture content
from 61% to 8% db in 300 - 340 minutes depending on
the levels of sample temperature set point. It was found
that as the set-point temperature increased, the drying
time was decreased. By using non-linear regression tech-
nique, the drying constants and coefficients of the five
models obtained are shown in Table 1.
In order to evaluate goodness of fit, coefficient of de-
termination (R2), adjusted coefficient of determination
( )
2
adjusted
R
, and standard error of estimation (SEE) were
also computed. The goodness of fit was determined by
the higher R2 and
2
adjusted
R
values as well as the lower
SEE values. For all cases, it was found that R2 and
2
adjusted
R
values were higher than 0.98, and SEE values
were lower than 0.029.
Furthermore, it was found that diffusion and Page’s
models gave the excellent fit results for all the experi-
mental data. For the case of diffusion model regression,
the values of R2,
2
adjusted
R
and SEE for 70˚C - 80˚C set-

P. Wilaipon / Journal of Agricultural Chemistry and E nvironment 3 (2014) 1-4
3
Table 1. The drying constants and parameters of five drying
models.
Model The drying constants and coefficients
k a b n
Newton
80˚C
70˚C
0.54348
0.45176
Page’s
80˚C
70˚C
0.56965
0.50033
0.93884
0.89054
Logarithmic
80˚C
70˚C
0.51691
0.43869
0.98288
0.95491
−0.00936
0.00621
Henderson & Pabis
80˚C
70˚C
0.52998
0.43146
0.97663
0.95907
Diffusion
80˚C
70˚C
0.54348
0.39875
2.86555
0.88591
0.99999
7.12935
point temperature were 0.9948 - 0.99 55, 0.99 41 - 0.9948,
and 0.0204 - 0.0207 respectively. With regard to Page’s
model, the values of these three criteria were found to be
0.9935 - 0.9969, 0.9931 - 0.9966, and 0.0163 - 0.0224
respectively. Normal probability plots of residuals of
these models are shown in Figures 1-4.
Figure 1. Normal probability plot of residual of Page’s model
for the case of 70˚C.
Figure 2. Normal probability plot of residual of Page’s model
for the case of 80˚C.
Figure 3. Normal probability plot of residual of Diffusion
model for the case of 70˚C.
Figure 4. Normal probability plot of residual of Diffusion
model for the case of 80˚C.
4. CONCLUSION
Drying kinetic of cassava in a single-planes micro-
wave hot-air oven was investigated. Drying time de-
creased with an increase in sample-surface temperature
set point. Approximately 87% of the moisture was re-
moved from the sample during the 300 - 340 minutes
drying-period. The rapid decrease in moisture ratio val-
ues followed by the gradual decline period was found in
all experiments. With regard to 5 drying models applied
to describe the drying kinetic of the sample, it was found
that diffusion and Page’s models provided a good agree-
ment between the experimental and predicted moisture
ratio values. High values of coefficient of determination
and adjusted coefficient of determination as well as low
value of standard error of estimation were also reported
for the case of these two models.
ACKNOWLEDGEMENTS
The author gratefully acknowledges Naresuan Univerity for the fi-
nancial support and Energy for Environment Research Unit for the
research equipment.
REFERENCES
[1] Tunde-Akintunde, T.Y. (2011) Mathematical modeling of
sun and solar drying of chilli pepper. Renewable Energy,

P. Wilaipon / Journal of Agricultur al C hemistry and Environment 3 (2014) 1-4
4
36, 2139-2145.
http://dx.doi.org/10.1016/j.renene.2011.01.017
[2] El-Beltagy, A., Gamea, G.R. and Amer Essa, A.H. (2007)
Solar drying characteristics of strawberry. Journal of
Food Engineering, 78, 456-464.
http://dx.doi.org/10.1016/j.jfoodeng.2005.10.015
[3] Togrul, H. (2005) Simple modeling of infrared drying of
fresh apple slices. Journal of Food Engineering, 71, 31 1 -
323. http://dx.doi.org/10.1016/j.jfoodeng.2005.03.031
[4] Baini, R. and Langrish, T.A.G. (2007) Choosing an ap-
propriate drying model for intermittent and continuous
drying of banana. Journal of Food Engineering, 79, 330-
343. http://dx.doi.org/10.1016/j.jfoodeng.2006.01.068
[5] Ozkan, I.A., Akbudak, B. and Akbudak, N. (2007) Micro-
wave drying characteristics of spinach. Journal of Food
Engineering, 78, 577-583.
http://dx.doi.org/10.1016/j.jfoodeng.2005.10.026
[6] Alibas, I. (2007) Microwave, air and combined micro-
wave-air-drying parameters of pumpkin slices. Food Sci-
ence and Technology, 40, 1445-1451.
http://dx.doi.org/10.1016/j.lwt.2006.09.002
[7] Liu, Q. and Bakker-Arkema, F.W. (1977) Stochastic mo-
deling of grain drying. Part 2: Model development. Jour-
nal of Agricultural Engineering Research, 66, 275-280.
http://dx.doi.org/10.1006/jaer.1996.0145
[8] Zhang, Q. and Litchfield, J.B. (1991) An optimization of
intermittent corn drying in a laboratory scale thin layer
dryer. Drying Technology, 9, 383-395.
http://dx.doi.org/10.1080/07373939108916672
[9] Singh, S., Sharma, R., Bawa, A.S. and Saxe na D.C. (2008)
Drying and rehydration characteristics of water chestnut
(Trapa natans) as a function of drying air temperature.
Journal of Food Engineering, 87, 213-221.
http://dx.doi.org/10.1016/j.jfoodeng.2007.11.027
[10] Henderson, S.M. and Pabis, S. (1961) Grain drying theory
II: Temperature effects on drying coefficients. Journal of
Agricultural Engineering Research, 6, 169-174.