JFS E: Food Engineering and Physical Properties

Catalytic Infrared Dehydration of Onions

MICHAEL M. GABEL, ZHONGLI PAN, K.S.P. AMARATUNGA, LINDA J. HARRIS, AND JAMES F. THOMPSON

ABSTRACT: Dehydrated onions are commonly dried with convection heating, which is inefficient and costly. This
study compared the drying and quality characteristics of onion dried with catalytic infrared (CIR) heating and forced
air convection (FAC) heating. Sliced high-solids onions were dehydrated under 9 conditions: CIR heating with and
without air recirculation, and FAC each operated at 60, 70, and 80 ◦ C. In general, CIR both with and without air
recirculation had higher maximum drying rates, shorter drying times, and greater drying constants than FAC at
moisture contents greater than 50% (d.b.). Dried onion quality, measured as pungency degradation, was similar for
both the drying methods at 60 and 70 ◦ C. The color analysis showed better product color (whiter and less yellow) at
lower temperatures for CIR and higher temperatures for FAC. The browning could have been caused by the higher
surface heat flux of the CIR heating and longer process times of FAC drying. Aerobic plate counts and coliform counts
were not significantly different for either product from the CIR or FAC drying. However, samples dried by the CIR
had significantly lower yeast and mold counts than those dried by the FAC. It is recommended that CIR be used in
the early stages of onion drying.
Keywords: dehydration, drying, infrared, onion, pungency

Introduction water, which could result in rapid moisture removal. Because the CIR

E: Food Engineering & Physical Properties

O nion, an important commercial food product and common
household seasoning, is considered the most important spice
bulb crop in the world (Pruthi 1980). Onions stored for long peri-
directly converts natural gas to radiant energy, it is more energy ef-
ficient than the typical infrared emitters using electricity. Relatively
little literature is available regarding infrared dehydration of onion
ods are subject to storage rot, sprouting, rooting, and loss of water (Wang 2002; Kumar and others 2005; Sharma and others 2005) and
making dehydration necessary (Akbari and Patel 2003). Many breeds there has been no published research of catalytic infrared dehydra-
of high-solids dehydrator onions are grown for the sole purpose of tion of onions.
making dried onion products. In 2001, California produced approxi- Final product quality is very important when evaluating a dry-
mately 1/4 of all the onions in the United States. Of those produced in ing method. High quality dried onion products should be white in
California, 66% were storage onions, which are primarily used for de- color, rehydrate easily, have high pungency, and low microbial loads
hydration (USDA 2001). Dehydrated onions are normally produced (Saimbhi and others 1970). The objective of this research was to
using forced convection tunnel dryers with variable temperatures compare the drying characteristics of thin-layer onion dehydration
at different drying states. Convective drying has been researched in- using CIR drying and forced air convection (FAC) drying with various
tensively, but this method has low energy efficiency and drying rate drying conditions. The studied drying characteristics included the
and also adversely affects the quality of the final products (Mazza drying rate, drying time, drying constant, and quality of dried onion
and LeMaguer 1980; Gowda and others 1986; Elustondo and oth- product. The measured quality characteristics included pungency
ers 1996; Markowshi and Zielonka 1996; Laguerre and others 1999; degradation, color changes, and microbial load reductions.
Adam and others 2000; Akbari and Patel 2003).
Infrared drying offers many advantages over convection drying,
Materials and Methods
including greater energy efficiency, heat transfer rate, and heat flux
Materials
which results in shortened drying time and higher drying rate (Sandu
Southport White Globe onions were supplied by Gilroy Foods
1986; Dostie 1992; Navarri and others 1992; Afzal and Abe 1997;
Inc. (Gilroy, Calif., U.S.A.) and were representative of onions used
Skjoldebrand 2002). Recent introduction of catalytic infrared (CIR)
in their commercial operation for the 2003 and 2004 seasons. The
dryers for dehydration of agricultural materials has increased the in-
onions had diameter of 40 to 70 mm and solids content ranged be-
terest in infrared drying for onion (Macaluso 2001). CIR uses natural
tween 24.3% and 29.3%. Moisture content of onion samples was
gas or propane, which is passed over a mesh catalyst pad to produce
measured according to the American Dehydrated Onion and Garlic
thermal radiant energy through a catalytic reaction. This reaction
Assn. (ADOGA) Official Standards and Methods (1997) (70 ◦ C for 6 h
occurs below the ignition temperature of gas so that no flame is
at 26.1 mmHg vacuum). Moisture measurements were performed
produced. The electromagnetic radiant energy from CIR has peak
in duplicate and reported as average on a dry basis unless otherwise
wavelengths in the range of medium to far infrared. The peak wave-
mentioned. Onions were cleaned by removing the tops and bot-
lengths match reasonably well with the 3 absorption peaks of liquid
toms along with the outer dry layers and the 1st fleshy layer, which
MS 20050650 Submitted 10/31/2005, Accepted 8/21/2006. Authors Gabel and accounted for about 20% of each onion. Onions were sliced perpen-
Harris are with Dept. of Food Science and Technology, Univ. of California, dicular to the axis into 2.5 ± 0.2 mm thick pieces using an industrial
Davis, One Shields Ave., Davis, CA 95616. Author Pan is with Processed Foods food slicer.
Research Unit, Western Regional Research Center, USDA-ARS, 800 Buchanan
St., Albany, CA 94710. Authors Pan, Amaratunga, and Thompson are with
Dept. of Biological and Agricultural Engineering, Univ. of California, Davis, CIR dryer setup
One Shields Ave., Davis, CA 95616. Direct inquiries to author Pan (E-mail:
zpan@pw.usda.gov).
The CIR dryer arrangement (Figure 1) consisted of a drying
chamber (95 × 65 × 65 cm) with a CIR emitter (Catalytic Drying
No claim to original US government works

C 2006 Institute of Food Technologists Vol. 71, Nr. 9, 2006—JOURNAL OF FOOD SCIENCE E351
doi: 10.1111/j.1750-3841.2006.00170.x
Further reproduction without permission is prohibited
 Catalytic infrared dehydration . . .

Technologies LLC., Independence, Kans., U.S.A.) mounted from the perature. The ratio of emitter on and off time decreased with the
top of the chamber. The sample was placed on a drying tray (84 × increase of moisture removed and drying time. The emitter surface
53 cm), which consisted of a fine mesh aluminum screen stretched temperature decreased rapidly during the intermittent heating pe-
across a strip steel frame. At the product surface, IR intensity de- riod, which lasts about 20 to 30 min before the emitter temperature
creased away from the center of the drying tray. When the drying tray was stabilized at about 100 ◦ C. In the temperature control, when
was positioned 20 cm below the emitter, the intensity at the tray’s the emitter temperature was below 150 ◦ C, the natural gas may not
center was 4818 W/m2 . Average intensity at 15 points across the tray be effectively catalyzed and emitter temperature was maintained by
at this distance was 2226 W/m2 measured using an Ophir FL205A the electric heating unit in the emitter.
Thermal Excimer Absorber Head (Ophir Optronics Inc., Wilming- Thermocouples and balance inputs were processed by using a
ton, Mass., U.S.A.). An aluminum waveguide (48 × 30 cm, upper data acquisition system, which was developed at Univ. of California,
rim; 42 × 22 cm, base perimeter), used to achieve a uniform heating Davis and consisted of a personal computer with Test Point software
of the entire product, rested on top of the drying tray and surrounded (Capitol Equipment, Bedford, N.H., U.S.A.).
the product. A balance (Ohaus Adventurer Pro; 8 kg capacity, 0.1 g
accuracy) was placed beneath the drying tray and measured prod- FAC dryer setup
uct weight during drying. A 1/100 HP exhaust fan (Dayton Electric
The FAC dryer used in the tests was an electrically heated column
Mnfg.) was located on the top of the drying chamber for ventilation.
dryer with diameter of 33 cm. A fan powered by a 3/4 HP permanent
Two 1/10 HP fans (Dayton Electric Mnfg.) mounted on each side of
magnet DC motor (Dayton Electric Mnfg.) blew heated air through
the dryer were used to recirculate part of the warm air in the drying
an electric coil heater and then through the column. Product was
chamber. These fans pulled air from the top of the drying chamber
placed in a circular mesh tray near the bottom of the column and
and fed it back into the chamber through slits running the entire
suspended by wires to the Ohaus balance to record product weight
length of the drying chamber.
change during drying. The temperature of heated air was also con-
The CIR emitter was preheated by an electric element located
trolled by the same computer control system as the CIR dryer using
inside the emitter. The natural gas intake was regulated by a gas
a Type T thermocouple to measure the temperature of the air before
control valve controlled by a computer system. Two Type-T thermo-
it reached the product. The on and off cycles of the electric heating
couples (response time 0.15 s) were used to measure the product
coils were controlled automatically to maintain the set point tem-
temperature. Each thermocouple was placed inside the flesh of an
E: Food Engineering & Physical Properties

perature. The air velocity for all of the tests was maintained at 0.5
onion slice, which was placed within the innermost 10 cm2 of the
m/s.
onion bed. The average temperature was used as input to control
the product temperature by turning the emitter on or off, which was
achieved by opening and closing the gas supply valve of the emitter. Drying trials
For example, the temperature of the emitter surface was 752 ± 22 ◦ C Nine conditions were tested: CIR heating with and without air
in the early heating stage. It took 8, 11, and 13 min to heat the product recirculation, and FAC each operated at 60, 70, and 80 ◦ C. An onion
to the set temperatures of 60, 70, and 80 ◦ C, respectively, without air sample, 250 g for CIR and 150 g for FAC, of intact slices was arranged
circulation. A large amount of moisture had been removed during in a single layer on the drying tray at a loading rate of 2.5 kg/m2 . For
this drying period. After the product reached the set temperature, CIR the drying tray was placed in the preheated CIR drying cham-
the emitter was automatically turned off and the emitter tempera- ber and the thermocouples were positioned as described above. Dis-
ture started to decrease. Once the product temperature was below tance between the emitter and drying tray was 15 cm with maximum
the set temperature, the emitter was turned on again to maintain intensity of 4752 W/m2 . CIR drying tests with air recirculation had
the product temperature. During the intermittent heating period, both the recirculation fans on during the entire test. Average air ve-
because a large amount of moisture had been removed, a relatively locity was set at 0.5 m/s. Each weight point was averaged with the
small amount of energy was needed to maintain the product tem- 2 prior and 2 consecutive points to correct for noise. Weight data
were also corrected for lift created by air inside the dryers based on
the weights with and without air flow. Onion weight and temperature
were recorded every 60 s with the aforementioned data acquisition
3 4 and control system.
Targeted final moisture content (MC) of the dried onion was set
at 10% (d.b.) in this study. The final weight of dried onion sample
2 5 was determined based on the initial and final MC and initial sample
weight. The experiments were replicated 2 or more times.

1
Drying kinetics
Drying rate was calculated in grams of moisture loss per kilogram
of initial weight of onion sample per minute (g/kg initialweight min). The
6 exponential model was chosen to describe the drying process. Model
curves were fitted to the experimental data and the performance of
7 8 11 the model was determined by the correlation coefficient (R2 ).
10 The exponential model is as follows:
9
M − Me
MR = = exp [−kt] (1)
Figure 1 --- CIR dryer: 1 --- natural gas; 2 --- gas flow M0 − Me
control; 3 --- air exhaust; 4 --- blower; 5 --- CIR emitter;
6 --- recirculation warm air; 7 --- computer controller;
8 --- wave guide; 9 --- thermocouple, 10 --- onion sample; where MR is the moisture ratio; M is the moisture content (% d.b.) at
11 --- balance any given time during drying; M 0 is the initial moisture content; M e

E352 JOURNAL OF FOOD SCIENCE—Vol. 71, Nr. 9, 2006 URLs and E-mail addresses are active links at www.ift.org
Catalytic infrared dehydration . . .

is the equilibrium moisture content; k is the drying constant (h−1 ); Statistical analysis
and t is time in h. Data from the quality test experiments and drying times were sta-
Moisture ratio (MR) was determined using the moisture con- tistically evaluated in Excel using the t test with the assumption of
tent data collected in the drying experiments. The fixed 4% (d.b.) equal variances. The pungency values for FAC and CIR were com-
equilibrium moisture content (EMC) was estimated from findings pared at each time interval for each temperature. Any value that
in literature (Wang 2002). For each drying condition, MR data was statistically significant (P < 0.05) is indicated in Table 2. The
were plotted on semilogarithmic axis versus the time (h) and the measured color results were compared for all 3 drying methods and
slope of the fitting line was the constant k. Correlation coefficients, any method that was significantly different was indicated in Table 3.
means, and standard deviations were also calculated for all 9 drying
conditions. Results and Discussions
Quality tests Drying rates and kinetics
Pungency. The effect of drying on pungency was determined Drying rates. When the drying rates were calculated and plotted
with 4 trays each containing 40 g of sliced onions, which were dried against moisture content (Figure 2) for each of the 3 drying temper-
simultaneously. Trays were removed at 10, 20, 30, and 40 min for atures, the CIR tests showed much higher drying rates throughout
80 ◦ C experiments; 10, 20, 40, and 60 min for 70 ◦ C experiments; and the course of drying than the FAC drying before the MC reached
30, 60, 120, and 180 min for 60 ◦ C experiments. These times corre-
spond to the approximate times that were required to achieve a 10%
MC (d.b.) at each temperature. After removal, sample weight was
measured and corresponding MC was calculated. Deionized wa-
ter was added to the dried products until the total weight of water
plus product was 90 ± 1 g. After rehydrating for 5 min, the samples
were homogenized for 30 s at 7000 rpm and 30 s at 10000 rpm us-
ing a hand-held homogenizer (Bahmix Bio-Mixer, Bartlesville, Okla.,
U.S.A.). Slurries were held for 30 min at room temperature before
filtering through 2 layers of cheesecloth.

E: Food Engineering & Physical Properties

Pungency was measured using a chemical pyruvic acid assay out-
lined by Anthon and Barrett (2003). Filtrate (25 µL), deionized water
(1.0 mL), and 1.0 mL of 0.25 g L−1 DNPH in 1 M HCl were placed in
13 mm × 100 mm test tubes, and tubes were heated in a 37 ◦ C water
bath for 10 min. Upon removal, 1.0 mL of 1.5 M NaOH was added.
Absorbance at 515 nm was measured on a Beckman DU 7500 spec-
trophotometer. Inherent, nonenzymatically formed pyruvate was
measured after heating a fresh 40-g sample in an 800 W microwave
oven for 1 min. Standards were prepared by adding 25 µL of sodium
pyruvate solution in concentrations of 0, 2, 4, and 8 mM instead
of the onion filtrate. Enzymatically formed pyruvate was calculated
from the difference between the amount of total pyruvate and the
nonenzymatically formed pyruvate. The average results of duplicate
tests were reported as percentage losses in pungency from a fresh
onion sample at various moisture contents.
Color measurement and microbial load reduction tests. L and
b color measurements of Lab color were performed for milled dried
onion samples from the drying rate trials using a Minolta CM-508
spectrophotometer. The a measurement was not reported because
it is not often used to indicate dried onion quality. The average of 5
readings from each sample was reported.
For determining the microbial load reduction, six 50 g sliced sam-
ples were dried with CIR and FAC at 60, 70, and 80 ◦ C to 10% MC.
Dried sample (10 g) was stored for 5 d. The sample was added to a
90 mL Butterfield’s buffer dilution blank and allowed to rehydrate
for 10 min at 4 ◦ C. Sample was homogenized for 1 min in a stom-
acher blender (Stomacher 400, Seward, Thetford, U.K.) and serial
dilutions were made and plated on Tryptic Soy Agar (TSA) (Difco,
Becton Dickinson, Sparks, Md., U.S.A.)) for aerobic plate counts, Di-
cholroan Rose Bengal Chloramphenicaol agar (DRBC, Difco, Becton
Dickinson) for yeast and mold counts, and Coliform Petrifilm (3M,
St. Paul, Minn., U.S.A.) for coliform counts. Duplicates of each di-
lution were made. TSA and Coliform Petrifilms were incubated at
35 ◦ C for 24 ± 2 h. DRBC plates were held at room temperature (24
± 2 ◦ C) for 5 d. The average results from duplicate test samples were
Figure 2 --- Drying rates of different drying methods and
recorded as colony forming units (CFU)/10 g dried sample. The tests conditions at various drying temperatures: (a) 60, (b) 70,
were performed in duplicate. and (c) 80 ◦ C

URLs and E-mail addresses are active links at www.ift.org Vol. 71, Nr. 9, 2006—JOURNAL OF FOOD SCIENCE E353
 Catalytic infrared dehydration . . .

Table 1 --- Result summary of onion drying characteristics

Drying time to Max Drying Correlation
reach 50% MC (d.b.) drying rate constant k coefficient
Drying condition (min) (g/kg initialweight min) (h−1 ) (R 2 )
60 ◦ C CIR–recirculation 54.5 ± 7.8 46.9 ± 8.4a 1.88 ± 0.42 0.988
70 ◦ C CIR–recirculation 32.0 ± 1.0a 59.9 ± 4.2a 3.23 ± 0.39 0.995
80 ◦ C CIR–recirculation 24.0 ± 4.6a 81.5 ± 6.6a 4.47 ± 0.67 0.997
60 ◦ C CIR 47.3 ± 7.5 47.5 ± 10.5a 2.04 ± 0.24 0.992
70 ◦ C CIR 36.5 ± 7.8a 67.7 ± 9.2a 2.87 ± 0.83 0.991
80 ◦ C CIR 18.0 ± 2.8a 83.4 ± 4.6a 5.95 ± 1.25 0.991
60 ◦ C FAC 56.0 ± 8.5 20.7 ± 3.1 2.23 ± 0.26 0.927
70 ◦ C FAC 51.0 ± 1.4 23.9 ± 3.9 2.45 ± 0.16 0.937
80 ◦ C FAC 33.0 ± 5.7 43.2 ± 22.1 3.82 ± 0.94 0.844
a
Significantly different (P < 0.05) when compared to FAC value at same temperature.

50%. Below this moisture content the decreased drying rates for CIR
drying could be due to lower moisture transfer rate in the onion slices
for the CIR than the FAC. The air flow might help with the moisture 1

removal at the low moisture range. Increasing the drying temper- 0.9
(a) 60°C
ature in the CIR drying trials increased the drying rate. A minimal
0.8
increase in drying rate was noted for FAC drying as air temperature 60°C CIR- Recirculation

was increased. 0.7

60°C CIR

Moisture Ratio, MR
It is apparent that air recirculation in the CIR drying caused lower 0.6 60°C FAC

drying rates, especially for the 60 and 80 ◦ C trials although an slight 0.5
Predicted

increase was seen in the 70 ◦ C trial. The recirculation of warm air

0.4
did not improve the drying rate under CIR drying as expected. Re-
0.3
circulation air decreased the drying rate and increased drying time,
E: Food Engineering & Physical Properties

which has been reported by Sandu (1986) and Paakkonen and others 0.2

(1999). 0.1

Drying rates changed with the drying temperature as expected.

0
For each of the plots from the CIR drying tests there was an absence 0 20 40 60
Time, min
80 100 120

of or just a very brief appearance of a constant rate period because 1

onions are hygroscopic and hygroscopic foods tend to quickly enter 0.9
(b) 70°C
the falling rate period (Rahman and Perera 1999). Additionally, sur-
0.8
face drying may occur more rapidly with CIR drying, which results 70°C CIR- Reciruclation

70°C CIR
in quicker entrance to the falling rate period due to slow water dif- 0.7
Moisture rat io, MR

70°C FAC
fusion to the surface of the onion. It is similar to many other foods 0.6
Predicted
that do not exhibit a constant rate-drying period due to the colloidal 0.5

and hydrophilic nature (Mazza and LeMaguer 1980; Baker 1997).

0.4
The FAC drying tests showed more of a distinct constant rate
◦
period at each of the 3 temperatures tested although the 80 C rate 0.3

period was not as profound as the other 2 temperatures. This might 0.2

be due to the lower heat flux, which resulted in a longer time to reach 0.1

the critical moisture content. Almost all of all tests showed linear
0
relationships between the drying rates and moisture contents for 0 10 20 30 40 50 60 70 80 90 100
Time, min
the period of moisture content from 50% to 225% (d.b.). 1

Based on the drying rate results, IR drying is recommended to dry 0.9

(c) 80°C
onion until about 50% MC, then followed by FAC drying in the latter
0.8
stage if a combined IR/convection drying system is used for drying 80°C CIR- Recirculation

80°C CIR
onion. For existing drying facilities, IR drying could be considered 0.7
Moisture Rati o, MR

80°C FAC
by adding a unit at the front of current conventional drying systems 0.6
Predicted
to take advantage of the high drying rate of IR for improving the 0.5

overall rate of drying.

0.4
The maximum drying rates and times required to reach 50% MC
(d.b.) under various conditions are summarized in Table 1. The max- 0.3

imum drying rates of CIR were significantly higher (P < 0.05) than 0.2

that of FAC drying at corresponding temperatures. But no significant 0.1

difference (P > 0.05) was observed between the CIR drying with and
0
without air recirculation at each corresponding temperature. 0 10 20 30 40 50 60
Time, min
Drying modeling. Based on the modeling results, the exponen-
tial model fits well with the experimental data from CIR drying Figure 3 --- Predicted and measured moisture ratio at dif-
(Table 1 and Figure 3). The correlation coefficients (R2 ) were greater ferent drying time and temperatures: (a) 60 ◦ , (b) 70 ◦ , and
◦
than 0.988. It has been reported that the Page model can be used for (c) 80 C

E354 JOURNAL OF FOOD SCIENCE—Vol. 71, Nr. 9, 2006 URLs and E-mail addresses are active links at www.ift.org
Catalytic infrared dehydration . . .

Table 2 --- Pungency changes during drying at different drying temperatures (percent pyruvate of fresh sample). Mois-
ture content is percent dry basis (d.b.).
Minutes of heating
30 60 120 180
60 C ◦
CIR 89.1 ± 15.7 79.1 ± 3.2 62.4 ± 7.9 59.3 ± 1.0
Moisture content (%) 142.4 69.2 22.5 14.4
FAC 80.7 ± 8.5 70.2 ± 3.7 63.9 ± 7.0 65.1 ± 0.4a
Moisture content (%) 106.2 39.3 20.1 15.6
Unheated sample at 180 min: 99.2 ± 7.2
10 20 40 60
70 ◦ C CIR 86.1 ± 5.4 80.7 ± 19.4 73.6 ± 1.9 83.6 ± 4.9
Moisture content (%) 135.7 69.5 19.6 15.7
FAC 115.2 ± 11.2a 70.0 ± 17.0 74.9 ± 0.2 74.2 ± 18.2
Moisture content (%) 148.6 69.3 21 9.7
Unheated sample at 60 min: 104.0 ± 2.3
10 20 30 40
80 ◦ C CIR 93.7 ± 32.5 69.9 ± 25.0 73.0 ± 6.2 61.3 ± 4.6
Moisture content (%) 81 30.5 21.1 11.2
FAC 89.9 ± 20.4 96.2 ± 18.8 97.3 ± 14.8a 94.0 ± 3.8a
Moisture content (%) 151.3 73.4 41.6 24.6
Unheated sample at 40 min: 89.4 ± 2.8
a
Significantly greater value (P < 0.05) for same time interval and temperature.

modeling the drying characteristics of onions under infrared heating inactivation and/or precursor degradation (Brewster and Rabinow-
(Wang 2002; Sharma and others 2005). After examining the fitness itch 1990). Significant color changes were also noted during this time
of the Page model for the CIR drying, it was found that the drying ex- (Table 3).
ponents of the Page models were close to 1 and the improvement in The large variability of measured pungency results could be

E: Food Engineering & Physical Properties

correlation coefficients was limited. Therefore, it is concluded that caused by nonuniform drying among the samples, difficulties in
the exponential model can be used to predict reasonably well the achieving a homogenous sample, and human error during the as-
drying characteristics of onions under the CIR drying. However, the say. Based on the obtained results, if 80 ◦ C drying temperature of CIR
correlation coefficients of the exponential model for the FAC drying drying is used, it is recommended to use it in the early drying stage
were in the range of 0.844 to 0.927, which indicates that the expo- before the moisture content of the samples reaches 75%. Then low
nential model may not be an appropriate model for describing the temperature drying can be used in the latter drying stage to prevent
drying characteristics. The predicted and experimental data of the severe browning and pungency degradation.
FAC drying in Figure 3 have shown that they did not fit well at the Color. The color measurement results of onion samples with
middle of the FAC drying process (Figure 3), which could be due to 10% MC are summarized in Table 3. The L values showed a decrease
the long constant rate periods observed. This was most apparent for with increasing temperatures for the CIR drying and the opposite
the 80 ◦ C FAC trial. effect for the FAC drying. The FAC results were opposite findings
from those of Lee and others (1995) where L values decreased with
Quality of dried onions higher air temperatures for convection drying of onion. The low L
Pungency. Both CIR and FAC drying methods had similar trends values for the 60 ◦ C FAC sample may be a result of extended drying
in pungency changes of onion samples, especially at 60 or 70 ◦ C time resulting from low drying temperatures. For the CIR drying, a
(Table 2). Pungency of FAC dried samples at the latter drying stage high drying temperature could increase browning and result in a
showed a greater decrease at the lower temperatures (60 and 70 ◦ C) dark color.
when compared to the 80 ◦ C test. This result could be due to longer The 60 ◦ C FAC dried sample was significantly (P < 0.05) less white
drying times causing more degradation of the product. The 80 ◦ C than either of the CIR samples. Samples from the other drying tem-
test with both drying methods showed that the pungency did not peratures are not significantly different from each other. The L color
change significantly until the moisture reached approximately 75%. parameter alone may not describe well the color changes occurring
Below 75% MC the pungency of FAC dried samples did not change during the drying process. The b parameter of the samples was also
significantly. This is consistent with the findings of Lee and oth- compared for evaluating the color data in its entirety.
ers (1995), in which high drying temperature had high pungency in
the dried onion. It can also be explained that accelerated drying in Table 3 --- L and b values of onion dried with different meth-
the initial stages would retain volatiles (Mazza and Maguer 1979; ods and temperatures
Brewster and Rabinowitch 1990). This is because the volatiles be- Drying condition L values b values
come “locked” into the product when it reaches the critical moisture ◦
60 C CIR- Recirculation 92.5 ± 2.7 8.6 ± 0.5
content. 70 ◦ C CIR- Recirculation 93.4 ± 3.2 6.2 ± 0.5
In general, the pungency of CIR dried samples decreased with 80 ◦ C CIR- Recirculation 88.1 ± 2.4 13.1 ± 1.0
moisture reduction. Although the measured pungencies varied with 60 ◦ C CIR 93.9 ± 2.8 5.2 ± 2.1
drying conditions and moisture during drying, the pungencies in the 70 ◦ C CIR 92.9 ± 2.3 7.7 ± 0.6
80 ◦ C CIR 89.4 ± 2.1 11.1 ± 1.5
samples with the longest drying time and equivalent moisture con-
60 ◦ C FAC 83.9 ± 3.5a 12.2 ± 0.7a
tent were similar, except for the 80 ◦ C trial. The 80 ◦ C trial showed a 70 ◦ C FAC 91.0 ± 4.0 9.9 ± 0.6a
greater loss in pungency in the samples dried in the CIR, especially 80 ◦ C FAC 93.1 ± 2.7 7.8 ± 0.4
at the latter drying stage. This decrease may be caused by the large a
Significantly different value (P < 0.05) from other drying methods at same
heat flux delivered to the product and may have resulted in alliinase temperature.

URLs and E-mail addresses are active links at www.ift.org Vol. 71, Nr. 9, 2006—JOURNAL OF FOOD SCIENCE E355
 Catalytic infrared dehydration . . .

Table 4 --- Microbial data for aerobic plate counts, coliform counts, and yeast and mold counts of CIR and FAC dried
samples (reported as log CFU/10 g dried sample)
Trial 1 Trial 2 Average 3
Fresh 5.24 Fresh 5.54 Fresh 5.39 ± 0.22
Aerobic plate counts CIR FAC CIR FAC CIR FAC
60 ◦ C 4.21 4.08 60 ◦ C 3.23 3.41 60 ◦ C 3.72 ± 0.70 3.75 ± 0.47
70 ◦ C 3.98 4.05 70 ◦ C 3.71 3.36 70 ◦ C 3.85 ± 0.19 3.71 ± 0.49
80 ◦ C 3.97 4.03 80 ◦ C 3.37 3.39 80 ◦ C 3.67 ± 0.42 3.71 ± 0.45
Fresh 5.27 Fresh 5.51 Fresh 5.39 ± 0.17
Coliform counts CIR FAC CIR FAC CIR FAC
60 ◦ C 3.04 3.95 60 ◦ C 2.40 2.40 60 ◦ C 2.72 ± 0.45 3.18 ± 1.10
70 ◦ C 2.18 3.93 70 ◦ C 2.40 2.18 70 ◦ C 2.29 ± 0.16 3.05 ± 1.24
80 ◦ C 1.70 3.00 80 ◦ C 1.00 1.00 80 ◦ C 1.35 ± 0.49 2.00 ± 1.41
Fresh 4.56 Fresh 4.84 Fresh 4.70 ± 0.20
Yeast and mold counts CIR FAC CIR FAC CIR FAC
60 C◦
4.16 4.72 ◦
60 C 4.12 4.81 ◦
60 C 4.14 ± 0.03 4.77 ± 0.07
70 ◦ C 3.83 4.69 70 ◦ C 4.00 4.31 70 ◦ C 3.92 ± 0.12 4.50 ± 0.27
80 ◦ C 3.44 4.10 80 ◦ C 3.44 4.00 80 ◦ C 3.44 ± 0.00 4.05 ± 0.07

A higher b value indicates a higher degree of browning and in samples dried in the CIR dryer, which may have been a result of
other color developments caused by enzymatic and nonenzymatic greater heat fluxes from the CIR emitter. The reduction of yeast and
browning reactions. Thus there was more color development in the mold in the dried samples with either drying method was no greater
CIR dried samples at higher temperatures due to the aforemen- than 1.4 log.
tioned reasons. Likewise, FAC drying has less browning at higher
temperatures due to a shorter drying time. The higher b values of
samples dried at 60 and 70 ◦ C with CIR plus air recirculation could be Conclusions
T hese experiment results indicate that CIR heating is an effec-
E: Food Engineering & Physical Properties

due to increased drying times compared to using CIR drying without

air recirculation. tive method for onion drying. Greater drying rates and shorter
To produce dried onion with a desirable light or white color CIR drying times were seen in CIR drying compared to FAC drying but
drying is recommended to dry the product at a mild temperature, only at MC greater than 50%. For achieving high quality product
such as 70 ◦ C, to take advantage of higher drying rates than observed and drying rate, a recommended combination of CIR and convec-
at 60 ◦ C drying and avoid the increased amount of browning seen tion drying is to use CIR to achieve 75% MC and then use convection
in 80 ◦ C drying. for later stages of drying. This type of processing would be beneficial
for commercial dehydrators who would not have to replace existing
Microbial load reduction. The results of aerobic plate counts equipment to incorporate a CIR unit. The recommended product
(APC) for all of the dried samples were similar (Table 4). For both temperature for CIR drying is 70 and 80 ◦ C. The higher tempera-
drying methods, the APC decreased by an average of about 1.7 log ture (80 ◦ C) should be used at the beginning of drying to achieve
from the fresh sample. There was no significant difference between maximum drying rates while product degradation is minimal. The
the 2 drying methods nor was there a difference in counts at different lower temperature (70 ◦ C) should be used for the remainder of dry-
drying temperatures. The composition of the surviving population ing because it achieves high drying rates but does not have the same
was not determined. The predominant microorganisms in dehy- adverse effects on quality factors, especially pungency and color, as
drated spices are aerobic sporeformers (Gray and Pinkas 2001) and does continual 80 ◦ C heating. Further research is needed to deter-
it is possible that this was true for the dried onion samples. The APC mine the point at which the temperature should be reduced from 80
of the dried product was not significantly different (P < 0.05) for any to 70 ◦ C. Additional studies are also necessary to determine drying
of the drying temperatures or either drying method. characteristics and quality changes that occur to onions below 10%
It was difficult to achieve a homogeneous sample of sliced onions MC.
to be used for the microbiological testing. The variation between
the trials was a result of different amounts and type of microflora Acknowledgments
represented on each of the samples used. Large standard deviations The authors are grateful for the financial support from the Cali-
were a result of averaging the results of the variable trials. fornia Inst. for Energy and Environment and the California Energy
Coliform levels in fresh samples were 5.39 log CFU/10 g dried sam- Commission (Grant nr MR-03-07) and thank Catalytic Drying Tech-
ple. The coliform count for fresh sample was not significantly differ- nologies LLC for providing the infrared dryer, Gilroy Food Inc. for
ent from the APC, suggesting that this group of bacteria dominated supplying the onion samples, and Dr. Paul Singh of Univ. of Califor-
the fresh aerobic microflora. As drying temperature increased, col- nia, Davis, for providing facility support and technical advice.
iform counts decreased in the final product. Coliform counts were
reduced during the drying process by over 2 logs at 60 ◦ C for both References
drying methods and by 3 logs at 80 ◦ C. There was no significant dif- Adam E, Muhlabauer W, Esper A, Wolf W, Spiess W. 2000. Quality changes of onion
ference between the FAC and CIR dried samples at corresponding (Allium cepa L.) as affected by the drying process. Nahrung 1(S):32–7.
[ADOGA]American Dehydrated Onion and Garlic Assn. Standard methods 1997. San
drying temperatures. Francisco, Calif.: ADOGA.
Yeast and mold counts, unlike coliforms, were not accounted for Afzal TM, Abe T. 1997. Modeling far infrared drying of rough rice. J Microw Power
Electromagn Energy 32(2):80–6.
on the APC (Gray and Pinkas 2001). Yeasts and molds do not grow fast Akbari SH, Patel NC. 2003. Studies on drying characteristics of onion. NHRDF News
enough to appear on APC agar. It is important not to correlate APC Letter 23(4):7–12.
Anthon GE, Barrett DM. 2003. Modified method for the determination of pyruvic acid
to yeast and mold counts. Yeast and mold counts were significantly with dinitrophenylhydrazine in the assessment of onion pungency. J Sci Food Agric
different for the 2 drying methods. Greater reductions were observed 83:1210–3.

E356 JOURNAL OF FOOD SCIENCE—Vol. 71, Nr. 9, 2006 URLs and E-mail addresses are active links at www.ift.org
Catalytic infrared dehydration . . .

Baker CGL. 1997. Industrial drying of foods. London: Chapman and Hall. Mazza G, Maguer ML. 1979. Volatiles retention during the dehydration of onion (Allium
p 309. cepa L.). Lebensm- Wiss U- Technol 12:333–7.
Brewster JL, Rabinowitch HD. 1990. Onions and allied crops, Vol. 3 biochemistry, food Navarri P, Andrieu J, Gevaudan A. 1992. Studies on infrared and convective drying
science and minor crops. Boca Raton, Fla.: CRC Press, Inc. p 265. of nonhygroscopic solids. In: Mujumdar AS, editor. Drying ’92. Elsevier Science
Dostie M. 1992. Optimization of a drying process using infrared, radio frequency and Publishers. p 685–94.
convection heating. In: Mujumdar AS, editor. Drying ’92. Elsevier Science Publishers. Paakkonen K, Havento J, Galambosi B, Pyykkonen M. 1999. Infrared drying of herbs.
p 679–84. Agric Food Sci Finland 8:19–27.
Elustondo MP, Pelegrina AH, Urbicain MJ. 1996. A model for the dehydration rate of Pruthi JS. 1980. Spices and condiments: chemistry, microbiology, and technology. New
onions. J Food Engr 29:375–86. York: Academic Press, Inc. p 449.
Gowda SJ, Gupta CP, Ojha TP. 1986. Studies on dehydration of onion. Mysore J Agric Rahman MS, Perera CO. 1999. Drying and food preservation. In: Rahman MS, editor.
Sci 20:186–94. Handbook of food preservation. Auckland, New Zealand: Horticulture and Food
Gray RJH, Pinkas JM. 2001. Gums and spices. In: Downes FP, Ito K, editors. Com- Research Inst. of New Zealand. p 173–207.
pendium of methods for the microbiological examination of foods, 4th ed. Wash- Saimbhi MS, Padda DH, Singh G. 1970. Evaluation of onion varieties for dehydration
ington, D.C. APHA. p 533–540. traits. J Food Sci Technol 8:149–50.
Kumar DGP, Hebbar HU, Sukumar D, Ramesh MN. 2005. Infrared and hot-air drying Sandu C. 1986. Infrared radiative drying in food engineering: a process analysis.
of onions. J Food Proc Pres 29(2):132–50. Biotechnol Progr 2:109–19.
Laguerre JC, Tauzin V, Grenier E. 1999. Hot air and microwave drying of onions: a Sharma GP, Verma RC, Pathare P. 2005. Mathematical modeling of infrared radiation
comparative study. Drying Technol 17(7&8):1471–80. thin layer drying of onion slices. J Food Engr 71(3):282–6.
Lee J, Kang H, Chang K, Kim S. 1995. Drying of onion and ginger using drying system Skjoldebrand C. 2002. Infrared processing. In: Henry CJK, Chapman C, editors. The
controlled by microcomputer. Agric Chem Biotech 38(1):78–82. nutrition handbook for food processors. Cambridge, UK: Woodhead Publishing. p
Macaluso V. 2001. Infrared drying technology applications. Cereal Foods World 423–32.
46(8):355–6. [USDA]U.S. Dept. of Agriculture, Natl. Agricultural Statistical Service. 2001. Commod-
Markowski M, Zielonka G. 1996. Heat transfer coefficient during the air drying of ity Report. Available from: http://www.usda.gov/nass/pubs/estindx.htm.
onions. Pol J Food Nutr Sci 5(4):97–104. Wang J. 2002. A single-layered model for far-infrared radiation drying of onion slices.
Mazza G, Lemaguer M. 1980. Dehydration of onion: some theoretical and practical Drying Technol 20(10):1941–53.
considerations. J Food Technol 15:181–94.

E: Food Engineering & Physical Properties

URLs and E-mail addresses are active links at www.ift.org Vol. 71, Nr. 9, 2006—JOURNAL OF FOOD SCIENCE E357