Microwave air and microwave finish drying of banana (Maskan).pdf

Journal of Food Engineering 44 (2000) 71±78

www.elsevier.com/locate/jfoodeng

Microwave/air and microwave ®nish drying of banana

Medeni Maskan

Department of Food Engineering, Engineering Faculty, University of Gaziantep, 27310 Gaziantep, Turkey

Received 16 July 1999; accepted 6 December 1999

Abstract

Banana samples (4:3 0:177; 7:4 0:251 and 14 0:492 mm thick) were dried using the following drying regimes; convection

(60°C at 1.45 m/s); microwave (350, 490 and 700 W power) and convection followed by microwave (at 350 W, 4.3 mm thick sample)

®nish drying. The drying of banana slices took place in the falling rate drying period with convection drying taking the longest time.

Higher drying rates were observed with the higher power level. Microwave ®nish drying reduced the convection drying time by

about 64.3%. A physical model was employed to ®t the experimental data and gave good ®t for all experimental runs except mi-

crowave ®nish data. Microwave ®nish dried banana was lighter in colour and had the highest rehydration value. Ó 2000 Published

Notation

drying constant (min ÿ1 )

rious damage to the ¯avour, colour, nutrients, reduction

in bulk density and rehydration capacity of the dried

product (Lin, Durance & Scaman, 1998; Drouzas,

Tsami & Saravacos, 1999). Major disadvantages of hot

air drying of foods are low energy eciency and lengthy

drying time during the falling rate period. Because of the

low thermal conductivity of food materials in this peri-

od, heat transfer to the inner sections of foods during

conventional heating is limited (Adu & Otten, 1996;

Feng & Tang, 1998). The desire to eliminate this prob-

lem, prevent signi®cant quality loss, and achieve fast

and eective thermal processing has resulted in the in-

creasing use of microwaves for food drying. Microwave

drying is rapid, more uniform and energy ecient

compared to conventional hot air drying. In this case,

the removal of moisture is accelerated and, furthermore,

heat transfer to the solid is slowed down signi®cantly

due to the absence of convection. And also because of

the concentrated energy of a microwave system, only

20±35% of the ¯oor space is required, as compared to

conventional heating and drying equipment. However,

microwave drying is known to result in a poor quality

product if not properly applied (Yongsawatdigul &

Gunasekaran, 1996a; Adu & Otten, 1996; Drouzas &

Schubert, 1996).

It has also been suggested that microwave energy

should be applied in the falling rate period or at a low

moisture content for ®nish drying (Prabhanjan, Ra-

maswamy & Raghavan, 1995; Kostaropoulos & Sarav-

acos, 1995; Funebo & Ohlsson, 1998). The reason for

moisture ratio

microwave

r 2

coecient of determination

S.E.

standard error

drying time (min)

W d

weight of dried sample (g)

W t

weight of rehydrated sample (g) at any time

moisture content (kg H 2 O/kg dry solids) at any time

X e

equilibrium moisture content (kg H 2 O/kg dry solids)

X o

initial moisture content (kg H 2 O/kg dry solids)

1. Introduction

Banana is one of the important high sugar containing

tropical fruit crops grown commercially in many coun-

tries. It is very susceptible to deterioration and consid-

erable amounts of this fruit is wasted due to the lack of

ecient preservation techniques that are unique to ba-

nana. An alternative method seems to be drying to ob-

tain a stable banana for further use.

Drying is one of the oldest methods of food preser-

vation and it is a dicult food processing operation

mainly because of undesirable changes in quality of the

dried product. High temperatures and long drying times,

required to remove the water from the sugar containing

fruit material in conventional air drying, may cause se-

E-mail address: maskan@gantep.edu.tr (M. Maskan).

PII: S 0 2 6 0 - 8 7 7 4 ( 9 9 ) 0 0 167-3

M. Maskan / Journal of Food Engineering 44 (2000) 71±78

this is essentially economic. Due to high cost, microwave

can not compete with conventional air drying. However,

microwaves may be advantageous in the last stages of

air drying. Because the least ecient portion of a con-

ventional drying system is near the end, when two-thirds

of the time may be spent removing the last one-third of

the moisture content (Al-Duri & McIntyre, 1992).

shorter drying times compared with hot air drying

alone.

The objectives of this study are to: (1) compare drying

characteristics, colour and rehydration of banana sam-

ples dried by hot air, microwave and hot air followed by

a microwave ®nish drying; (2) determine the drying

constants, using a simple diusion model (Eq. (1)) and

assess the eect of selected parameters such as drying

temperature, sample thickness and microwave power.

MR X ÿ X e

2. Microwave principles and application to drying of foods

X o ÿ X e expÿkt:

Microwaves are electromagnetic waves in the fre-

quency range of 300 MHz to 300 GHz (equivalent to a

wavelength of 1±0.01 m), generated by a magnetron-

type vacuum tube. Electromagnetic energy at 915 and

2450 MHz can be absorbed by water containing mate-

rials or other ``lossy'' substances, such as carbon and

some organics, and converted to heat (Khraisheh,

Cooper & Magee, 1997a). Because the waves can pene-

trate directly into the material, heating is volumetric

(from the inside out) and provides fast and uniform

heating throughout the entire product. The quick energy

absorption by water molecules causes rapid evaporation

of water (results in higher drying rates of the food),

creating an outward ¯ux of rapidly escaping vapour. In

addition to improving the rate of drying, this outward

¯ux can help to prevent the shrinkage of tissue structure,

which prevails in most conventional air drying tech-

niques. Hence better rehydration characteristics may be

expected in microwave dried products (Prabhanjan et al.,

1995).

In recent years, microwave drying has gained popu-

larity as an alternative drying method for a wide variety

of food products such as fruits, vegetables, snack foods

and dairy products. Several food products have been

successfully dried by the microwave-vacuum application

and/or by a combined microwave assisted-convection

process. These authors included Kim and Bhowmik

(1995) for plain yogurt, Yongsawatdigul and Gunasek-

aran (1996a) for cranberries, Lin et al. (1998) for carrot

slices, Drouzas et al. (1999) for model fruit gels, Al-Duri

and McIntyre (1992) for skimmed milk, whole milk,

casein powders, butter and fresh pasta, Bouraout,

Richard and Durance (1994) for potato slices, Prab-

hanjan et al. (1995) for carrots, Tulasidas, Raghavan

and Norris (1996) for grapes, Funebo and Ohlsson

(1998) for apple and mushroom, and Ren and Chen

(1998) for American ginseng roots.

Another group of researchers proposed a two-stage

drying process involving an initial forced air convective

drying, followed by a microwave ®nish drying (Prab-

hanjan et al., 1995; Feng & Tang, 1998). Microwave

application has been reported to improve product

quality such as better aroma, faster and better rehy-

dration, considerable savings in energy and much

For microwave drying it can be assumed that X e 0.

3. Materials and methods

3.1. Material

Ripe bananas (Musa species) with an initial moisture

content of 3.1 kg H 2 O/kg dry solids were obtained from

a local supermarket and stored at 4 0:5°C. Prior to

drying, samples were taken out of storage, hand peeled,

cut into 4:3 0:177; 7:4 0:251; 14 0:492 mm thick

and 30 0:901 mm diameter slices (where shows

standard deviation of measurements) with a cutting

machine. At least 10 measurements of the thickness were

made at dierent points with a dial micrometer; only

slices that fell within a 5% range of the average thickness

were used. All bananas used for drying were from the

same batch.

3.2. Drying equipment

A programmable domestic microwave oven (Arßelik

ARMD 580, TURKEY), with maximum output of

700 W at 2450 MHz. was used. The oven has the facility

to adjust power (wattage) supply and the time of pro-

cessing. The hot air drying experiments were performed

in a pilot plant tray dryer (UOP 8 tray dryer, Arm®eld,

UK). The dryer (Fig. 1) consisted of a proportional (P)

controller controlling the temperature. Air was drawn

into the duct through a mesh guard by a motor driven

axial ¯ow fan impeller whose speed can be controlled in

the duct.

3.3. Drying procedure

The drying regimes were as follows:

(1) Hot air drying: The dryer was operated at an air

velocity of 1.45 m/s, parallel to the drying surface of the

sample, 60°C dry bulb and 30°C wet bulb temperatures.

The three sample thicknesses were studied at constant

temperature. Moisture loss was recorded at 10 min inter-

vals during drying for determination of drying curves

by a digital balance (Avery Berkel, CC062D10ABAAGA).

M. Maskan / Journal of Food Engineering 44 (2000) 71±78

Fig. 1. Schematic diagram of the hot air drying equipment (not to scale).

Bananas were dried until equilibrium (no weight change)

was reached.

(2) Microwave drying: Factors investigated in micro-

wave drying were microwave power intensity (350, 490

and 700 W) at constant sample thickness of 4.3 mm, and

sample thickness/load (4.3 mm/3.56 g, 7.4 mm/5.60 g

and 14.0 mm/11.85 g) at constant microwave power

output of 490 W. One glass petri dish (7.1 cm diameter

1:2 cm deep), containing the sample, was placed on the

centre of a turntable ®tted inside (bottom) the micro-

wave cavity during treatment for even absorption of

microwave energy. The presence of the turntable was

necessary to achieve the optimum oven performance and

to reduce the levels of re¯ected microwaves onto the

magnetron (Khraisheh et al., 1997b). The drying was

performed according to a preset power and time

schedule. Moisture loss was measured by taking out and

weighing the dish on the digital balance periodically.

When the material reached a constant weight, equilib-

rium moisture content was assumed to be reached. At-

tention was paid to ensure that the sample was not

charred.

(3) Air followed by a microwave ®nish drying:A4.3mm

thick banana sample was dried at 60°C and 1.45 m/s air

velocity to 1.25 kg H 2 O/kg dry solids moisture content,

the point where drying slows down. Then, sample was

taken out and dried in the microwave oven. Some pre-

liminary tests conducted on partially air dried sample

resulted in burning of sample at high microwave power

levels. Hence, a microwave power of 350 W was selected

for ®nish drying purpose.

colour and rehydration was performed (within one week

after the drying).

3.4.1. Colour

Sample colour was measured before and after drying

by a HunterLab ColorFlex, A60-1010-615 model col-

ormeter (HunterLab., Reston, VA). The colour values

were expressed as L (whiteness/darkness), a (redness/

greenness) and b (yellowness/blueness). And also, the

total colour dierence from the fresh bananas DE,as

de®ned the following, was used to describe the colour

change during drying:

L o ÿ L

ÿ 2

a o ÿ a

b o ÿ b

; 2

where subscript ``o'' refers to the colour reading of fresh

banana, L ; a and b indicate brightness, redness and

yellowness of dried samples respectively. Fresh banana

was used as the reference and a larger DE denotes

greater colour change from the reference material.

3.4.2. Rehydration

The dried samples were manually ground and im-

mediately loaded (about 0.5 g each) into small alumi-

nium sample dishes. 100 ml of distilled water was

transferred into a glass jar and a tripod was also placed

in the jar. The dishes were placed on tripod in the jar

which was then tightly closed and kept at 20°C for

equilibration. The dishes were periodically weighed until

equilibrium was reached. The rehydration ratio was

determined by

Weight gain % W t ÿ W d

W d

100:

3.4. Quality evaluation

For quality evaluation, similar drying experiments

were conducted separately under the same microwave,

hot air and microwave ®nish drying conditions. Drying

was terminated when moisture content reached about

0.1 kg H 2 O/kg dry solids. After the drying tests, samples

were kept in air tight glass jars until measurements of

3.5. Statistical analysis

ANOVA ) was conducted to de-

termine the eect of variable factors on drying param-

eters using Statgraphics software (1991). Least-squares

multiple range test was performed to dierentiate the

Analysis of variance ( ANOVA

M. Maskan / Journal of Food Engineering 44 (2000) 71±78

signi®cant eect of drying methods on drying rate. The

parameter of non-linear model (Eq. (1)) was calculated

by the NLIN procedure of the SigmaPlot (Scienti®c

Graph System, version 4.00, Jandel).

4. Results and discussion

4.1. Hot air drying

The moisture content versus time curves for hot air

drying of banana samples as in¯uenced by thickness are

shown in Fig. 2. Obviously, as the thickness of the

sample increased, the time required to achieve a certain

moisture content increased. For example, the drying

times for reaching about 0.1 kg H 2 O/kg dry solids

moisture content of 4.3, 7.4 and 14 mm thick samples

were about 482, 610 and 777 min respectively at 60°C air

temperature. These results were in agreement with pre-

vious literature studies (Yusheng & Poulsen, 1988;

Madamba, Driscoll & Buckle, 1996; Maskan & Ibano-

glu, 1998). The drying rate was calculated at dierent

times and plotted against average moisture content as

shown in Fig. 3. A constant rate period was not ob-

served in hot air drying of banana samples at 60°C.

Hence, the entire drying process for the samples oc-

curred in the range of falling rate period in this study.

However, it might be possible to have a short constant

rate period using lower temperatures such as 40±50°C.

It has been reported that almost all of the drying of

biological products takes place in the falling rate period

(Madamba et al., 1996). However, Mowlah, Takano,

Kamoi and Obara (1983) have found both constant and

falling rate periods in dehydration of bananas at 60°Cin

an air circulated oven. Air in the oven is saturated, by

the time, and forms a thick ®lm around the food that

prevents eective separation of the evaporated moisture

from the food. This may be the reason for existence of a

Fig. 3. Drying rate curves of banana slices dried by hot air (60°C and

1.45 m/s).

constant rate period in their study. A high initial drying

rate, with higher rates at lower thicknesses, was ob-

served (Fig. 3). All the samples tended to dry slowly at

the last stages of drying, presumably due to collapse

(shrinkage) of the banana structure resulting in low

transport rate of water and prolonged drying time

(Kostaropoulos & Saravacos, 1995).

4.2. Microwave drying

The eect of changing the power output in the mi-

crowave oven on the moisture content curve of 4.3 mm

thick banana sample is shown in Fig. 4. At all power

levels, drying curves were steeper and tended to end at

about the same time. The observed initial acceleration of

drying may be caused by an opening of the physical

structure allowing rapid evaporation and transport of

water (Kostaropoulos & Saravacos, 1995). The

ANOVA

Fig. 2. Air drying curves for banana slices (air at 60°C and 1.45 m/s).

Fig. 4. Drying curves of banana slice (4:3 0:177 mm) dried by mi-

crowave method with dierent microwave power levels.

ANOVA

showed no eect of the intensity of power on moisture

M. Maskan / Journal of Food Engineering 44 (2000) 71±78

loss (P > 0:05). Similar results were obtained by Walde,

Balaswamy, Sivaswamy, Chakkaravarthi and Rao

(1995) for microwave drying of gum karaya and

Yongsawatdigul and Gunasekaran (1996b) for micro-

wave-vacuum drying of cranberries. However, several

investigators have reported the eect of power output on

drying time of food materials (Al-Duri & McIntyre,

1992; Prabhanjan et al., 1995; Drouzas & Schubert,

1996). Fig. 5 shows the eect of drying conditions on the

drying rate. Although high moisture foods can be ex-

pected to have a period of constant rate drying, this was

not observed in the present study under any of the test

conditions. The total drying times required to reach a

®nal moisture content of about 0.1 kg H 2 O/kg dry solids

were 13, 18 and 27 min at 700, 490 and 350 W, respec-

tively. These results show that the drying time of the

4.3 mm thick sample was shortened from 482 min by hot

air drying to the 13±27 min range when dried by mi-

crowave energy.

karan, 1996b). Due to that, the 14 and 7.4 mm thick

samples spread on the bottom of the dishes as a thin

layer, a large drying surface area formed hence, drying

accelerated (data were not shown). Only the thin sample

(4.3 mm) maintained its shape without spreading and it

took much time to dry this sample compared to the

others. The drying times were found to be about 18, 13

and 10 min at 490 W for 4.3, 7.4 and 14 mm thick

samples to reach a moisture content of about 0.1 kg

H 2 O/kg dry solids.

4.3. Microwave ®nish drying

ANOVA results showed a signi®cant

dierence (P < 0:05) between drying rates of hot air and

microwave drying techniques. Average drying rate over

the drying periods was 0.0088 kg water/kg dry solids/

min for air dried, and between 0.0880 and 0.2027 kg

water/kg dry solids/min for microwave dried 4.3 mm

thick banana sample in the 350±700 W power range,

respectively. The results indicated that mass transfer

within the sample is rapid during microwave heating

because heat is generated within the sample, creating a

large vapour pressure dierential between the centre and

the surface of products (Lin et al., 1998).

Eorts were made to study the eect of the sample

thickness (4.3, 7.4, 14 mm) on drying at constant power

output (490 W). In contrast to hot air drying, the thicker

sample dried more rapidly than the thin one. It is be-

cause of sudden and volumetric heating, generating high

pressure inside the bananas, resulted in boiling and

bubbling of the samples (Yongsawatdigul & Gunase-

Banana sample (4.3 mm thick) was hot-air dried at

60°C initially, then microwave energy was applied (to

the point where conventional drying is very slow) for

®nish drying. The drying rate versus average moisture

content curve was presented in Fig. 6. It can be seen that

microwave ®nish application increased the drying rate

signi®cantly (over 0.8 kg water/kg dry solids/min). The

same sample has an initial drying rate value of about

0.035 kg water/kg dry solids/min (Fig. 3) when air dried,

0.4 kg water/kg dry solids/min when microwave dried at

350 W microwave power output (Fig. 5). This applica-

tion also reduced the drying time from 482 to 172 min

(64.3% reduction in drying time) when dried by air and

®nish dried respectively.

4.4. Modelling drying curves

In this work, Eq. (1) was employed as a physical

model for description of the drying processes, rather

than a mathematical model (Prabhanjan et al., 1995;

Madamba et al., 1996; Drouzas et al., 1999). Drying

data were used to test the applicability of this model.

The parameter k together with S.E. and r 2 were evalu-

ated using nonlinear regression. The results are

Fig. 5. Drying rate curves of banana slice (4:3 0:177 mm) under

various microwave power levels.

Fig. 6. Drying rate of banana slice (4:3 0:177 mm) dried by hot air

(60°C and 1.45 m/s) followed by microwave at 350 W power output.

ANOVA

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: