Research for Agricultural Sciences:" Effect of high temperature fluidized drying and tempering on head rice yield and mechanical strength of Vietnamese rice varieties " doc

SECTION 2

HIGH TEMPERATURE FLUIDISED BED DRYING
(Experimental study on high moisture paddy- Vietnamese rice varieties)

75


Effect of high temperature fluidized drying and tempering on head rice yield
and mechanical strength of Vietnamese rice varieties

Introduction
Being as a major exporter of rice all over the world, the quality of rice has become a central issue
for Vietnamese farmers, particularly in rainy cropping season. Therefore, it is important to dry rice
as quickly as possible to prevent spoilage but maintain quality. The conventional drying technique
such as flat bed drying can take up to 8 hrs (or even more) to dry the paddy to safe moisture level.
High temperature drying can allow to dry the paddy faster, therefore, the space and time taken to
dry the paddy can be much shorter. This can serve as a compact drier. High temperature fluidized
bed drying technique has been investigated to be an effective method for drying high moisture rice
grain which is easily deteriorated in humid tropic climate (Taechapairoj et al. 2003). In this
method, the rice grain are suspended by the upward-moving drying air with high velocity 2-3 m/s,
thus paddy and air are mixed vigorously (Kunze and Calderwood 2004). Fluidized bed (or fluid
bed) drying is applicable in the first stage of drying when paddy is required to decrease high
moisture content down to 18% (wet basis). Paddy then is continued to be dried at low temperature
or by ambient air in storage bin (Proctor 1994; Taechapairoj et al. 2003). Alternately, a multi-pass
fluidized bed drying can be applied as a compact drying process.
High temperature (over 1000C) fluidized bed drying of paddy has been reported by a few
researchers. But, normally it is recommended that hot air drying temperature should not exceed
1500C to avoid the adverse effect on the whiteness of rice. The lower range of drying temperature
40-1500C has been also investigated by Tirawanichakul et al. (2004). They reported an
improvement on the head rice yield when the drying temperature over 800C was used for the high

moisture content (32.5% db) paddy. This was probably contributed by the occurrence of partial
gelatinization of starch at high temperature. Some researchers suggested the need for tempering
the grain for 25-30 minutes if high temperature drying is used (Poomsa-ad et al. 2005;
Prachayawarakorn et al. 2005). Although there are a few research works being reported on high
temperature fluidized bed drying, no real application has been reported. Therefore, more
understanding on this subject is required.

76


This study aims to enhance the knowledge on the effect of high temperature drying and tempering
on head rice yield, fissured kernels, mechanical strength and quality changes of Vietnamese rice
varieties.

The specific objectives of this study are:
To study the feasibility of high temperature fluidized bed drying of high moisture content
Vietnamese rice varieties (two rice varieties were considered in this study).
To study the optimum drying temperature and tempering duration as reflected by the
lowest number of fissured kernel after the completion of drying followed by tempering.
To study the mechanical strength of the rice kernels under different drying and tempering
regimes.
To study the effect of high temperature drying and tempering on the whiteness, pasting
properties and crystallinity of rice starch. These properties can reflect the cooking quality
of rice.

Materials and Methods

Fluidized bed drier:
A batch lab-scale drier (HPFD150) developed at Chemical Engineering Department, Nong Lam
University, Vietnam was used in this experiment (Figure 1). This dryer consists of three main

components, namely (i) a cylindrical shaped drying chamber (40 cm in height x 15 cm diameter)
(ii) 8 kW electric heating units and (iii) a centrifugal fan driven by a 0.75 kW motor. Inlet drying
temperature ranging from 20-1000C was monitored by a Hanyoung DX7 temperature controller.
Outlet air temperature was monitored by a Daewon thermometer (Korea).

Paddy Samples:
Long-grain rice cultivars A10 and OM2717 were collected in fields of local farmers at Tien Giang
Province and Hochiminh City in May 2007. Freshly harvest paddy (24-33% moisture, wb) was
immediately transported to the laboratory and kept in cold storage maintained at 50C. The A10
variety of paddy had 31-33% initial moisture (wb), whereas OM2717 had 25-26% moisture (wb)
content. Rice samples were allowed to equilibrate at room temperature before subjecting to drying.

77


Approximately 200 gram of paddy (thickness of bed 2 cm) was dried in fluidized bed dryer at 80
and 90oC for the drying periods of 2.5 and 3.0 min. The experimental procedure is illustrated in
Figure 2. Samples were transferred to a sealed glass jar immediately after drying and then
tempered in an incubator set at 75 and 860C, which were equivalent to the grain temperatures after
fluidized bed drying at 80 and 900C respectively. The tempering duration used was 0, 30, 40 and
60 mins. A sealed glass jar was warmed up to set temperature and stored in a foam box while
taking out from the incubator to prevent the loss of heat.
After tempering, the samples were thin layer dried at 350C to safe storage level of moisture content
(below 14% wb). The dried rice was sealed in plastic bags and kept at room temperature for 3 days
before determining the head rice yield (HRY), cracked grain ratio and mechanical strength of rice
kernels. 200 gram of paddy was also dried in the thin layer dryer at 350C for 16 h down to 14% wb
and used as control sample. All treatments were undertaken in triplicate.

Moisture content determination
The moisture contents before and after tempering and final moisture content after thin layer drying

of each drying run were determined by drying 5-10 gram of rough rice (in duplicate) in an oven at
1300C for 16h.

Head rice yield
100 gram of paddy was dehusked/dehulled and milled by laboratory milling system and whole
kernel was separated from broken kernel to determine the head rice yield, which is defined as the
ratio of the mass of unbroken kernel to the mass of paddy.

Cracked grain ratio
Fissure enumeration was carried out in duplicate on 50 manually dehulled brown rice kernels per
each measurement by visual observation with the assistance of a light box.

78


Mechanical strength of kernels
The breaking force of each individual brown rice grain was measured by three-point bending test
with the test device (developed at the University of Queensland, Australia) attached to Universal
Texture Analyser (Micro Stable Systems, UK). 50 rice grains of each lot were randomly selected
and dehulled by hand. Chalky and fissured kernels were discarded. The pre-test, test and post test
speeds of the probe used were 1mm/s, 2 mm/s, and 10 mm/s respectively. From the force-distance
curve obtained by Texture Analyzer software, the peak force at which rice kernel failure occurred
was regarded as breaking force of the rice kernel.

Colour determination
The objective of the color measurement was to evaluate the effect of high temperature drying and
tempering on the whiteness of the rice. It was undertaken on only A10 rice variety. Milled rice
sample of each treatment was placed in a clear Petri dish and the colour parameters were measured
by a Minolta Chroma Meter CR-200 (Japan) in CIE 1976 L*, a*, b* colour space. Parameter L*,
+a*, -a*, +b* and -b* describe the brightness, red, green, yellow and blue colour, respectively. The

total colour difference, ∆E* was also calculated.

Statistical analysis
A multilevel factorial design, consisting of two levels of drying temperature (80 and 900C), two
levels of drying time (2.5 and 3.0 min) and four levels of tempering time (0, 30, 40, and 60 min),
was chosen in this experiment. The data was analyzed by using statistical package MINITAB®
Release 14 (Minitab Co., USA) and GLM (General Linear Model) procedure.

79


Fluidized bed dryer (developed at Chemical
Engineering Dept., Nong Lam University,
Vietnam)

Tempering of rice in sealed glass jars

Thin layer dryer used in this study

Dried rice of replicate 1 (27 treatments)

Figure 1: Equipment used in drying and tempering of the rice samples

80


Fresh paddies
(A10 30-33%wb, OM2717 24-26%wb)

Moisture

determination

Fluidized bed drying

800C
2.5 & 3.0 min

90 0C
2.5 & 3.0 min

Moisture
determination

Tempering in incubator

750C

86 0C

Tempering time 0, 30, 40 and 60 min

Thin layer drying at 350C

Moisture
determination

Moisture
determination

Head rice yield (%)

Fissured kernels (%)
Breaking force (N)

Figure 2. Schematic diagram of experimental and measured parameters.

81


Results and Discussion
The evolution of moisture content during fluidized bed drying and subsequent tempering for both
rice varieties is presented in Figure 3. The percentage of moisture content removal of the paddy
was found to be in the range of 7.7 to 12.0%. As can be seen in Figure 3, increasing drying
temperature led to larger percentage of moisture removal.

Tempering period not only helped moisture redistribution inside the rice kernel but also caused
slight loss of moisture during prolong tempering duration (40 min as shown in Figure 3). Within
2.5 min of fluidized bed drying, the moisture content for A10 variety dropped by 8.7-9.4% at 800C
and 11.0-12.0% at 900C, while it dropped by 7.7-8.6% at 800C and 9.7-11% at 900C for OM2717
variety. Extending drying time to 3.0 min did remove further 1% moisture content in both drying
temperatures for A10 rice. However, there was only 0.1-0.3% moisture content drop for OM2717
rice. It is suggested that the amount of moisture that can be removed during fluidized bed drying
depends on the initial moisture content of paddy. Note that, the initial moisture content of A10
(31-33% wb) was higher than that of OM2717 (25-26% wb). An extension of drying time to 3.0
min did not remove further moisture because the diffusion of moisture inside the kernel is time
dependent. If the drying proceeds further moisture removal will continue from outer surface of the
grain which will result in to physical stress due to increased differential moisture between the
interior and exterior layers of rice kernels. Therefore, the tempering step is necessary to be
employed to allow moisture to diffuse from interior to exterior prior to further drying.

82

30

A10

Drying duration

20
OM2717

M
oisture content, %wb

40

10
Tempering duration

0
0

10

20

30

40

50

60

70

Ope ration tim e, m in
80C, 2.5 min

80C, 3.0 min

90C, 2.5 min

90C, 3.0 min

80C, 2.5 min

80C, 3.0 min

90C, 2.5 min

90C, 3.0 min

Figure 3. The reduction of moisture contents during fluidized bed drying and tempering at high
temperatures for two Vietnamese rice varieties (A10 and OM2717).
Head rice yield
Drying time effect: Increased drying time reduced the head rice yield significantly (Figure 4). In
our other experiments we found that a longer than 3 minutes drying time caused a remarkable
reduction in moisture content below 17.5% wb but caused higher fissures in the grain (results not
presented here). As mentioned earlier, a longer exposure of grain at higher temperature causes

strain on the grain due to the rapid removal of moisture from the surface while it takes time to
diffuse water from the interior. The large differential moisture will cause more fissures in the
kernels. According to this result, not more than 2.5 minutes should be the drying time for the rice
at these high temperature conditions, although no experiment was done for less than 2.5 min
assuming that we will not achieve enough moisture removal for shorter time than this (Figure 3).

83


Tempering time effect: The experimental result in this study showed that at both drying
temperatures with drying time of 2.5 min, extending tempering duration to 40 min can improve the
head rice yield (Figure 4). The same trend was observed for both rice varieties. There was a clear
trend of increasing head rice yield with tempering time up to 40 minutes. It indicates that 30-40
minutes can be an optimal tempering time for both varieties. It should be noted that the tempering
temperatures used in this work are above the glass transition temperature of rice. This means that
the rice were in rubbery state during the duration of tempering. This will be discussed more in
further work in future.

It was found that the drying temperature, tempering time and interaction between drying
temperature and tempering time had significant effect on HRY (P <0.05) as represented in Table 1.

OM2717

A10
70

60

Head Rice Yield, %

70

60
50

50

40
40

30
30

20

80oC, 3.0 min
90oC, 3.0 min

80oC, 2.5 min
90oC, 2.5 min

20

80oC, 2.5 min

Reference

10

80oC, 3.0 min

90oC, 2.5 min

10

90oC, 3.0 min

Reference

0
0

0
0

10

20

30

40

50

60

70

20

40

60

80

T emp er ing t ime, min

Te m pe r i n g t i m e , mi n

a) A10 variety

b) OM2717 variety

Figure 4. Effect of tempering time on HRY at drying temperature of 800C and 900C for 2.5
and 3.0 min. Reference HRYs of A10 and OM2717 were 54.5% and 43.26%, respectively
(triangular dot).

84


Table 1. Main effects and interaction effects between drying temperatures, drying time and
tempering time on HRY, fissured kernels (FK) , and breaking force (BF) for A10 and OM2717
cultivars.

HRY

P value
FK

0.003
0.089
0.000
0.010
0.580
0.493
0.891

0.000
0.000
0.000
0.000
0.023
0.000
0.002

0.000
0.003
0.038
0.000
0.073
0.007
0.027

0.000
0.000
0.000
0.960
0.000

0.471
0.961

0.000
0.001
0.000
0.524
0.486
0.967
0.295

0.000
0.000
0.029
0.091
0.014
0.597
0.213

Interaction factor(s)
A10
DryingTemp
DryingTime
TemperingTime
DryingTemp*DryingTime
DryingTemp*TemperingTime
DryingTime*TemperingTime
DryingTempt*DryingTime*TemperingTime
OM2717
DryingTemp

DryingTime
TemperingTime
DryingTemp*DryingTime
DryingTempt*TemperingTime
DryingTime*TemperingTime
DryingTemp*DryingTime*TemepringTime

BF

Fissured kernels:
Drying temperature and time effect: Drying temperature, drying time and tempering time had a
very significant effect on fissured grain percentage. Without tempering, the percentage of fissured
kernels was 20-40% for A10 variety and more than half of OM2717 white rice kernels were
broken (60-86%). Increase in drying time also increased the amount of damaged rice kernels
(Figure 5).

Tempering time effect: Figure 5 indicates that the tempering of the rice significantly decreased
the amount of fissured kernels. This demonstrates how important it is to temper the paddy for an
optimal time if a high drying temperature is used. There was a continuous decrease in the fissured
kernels as the tempering duration was increased particularly for OM2717 rice variety, however,
based on the A10 variety we can see that a time of 30-40 minutes is required to achieve a low
fissured kernels. The tempering has two simultaneous effects, one is to allow moisture to
equilibrate (diffuse from interior to exterior part of the grain) and also relax the molecules making
85


the structure more rigid. Both effect will reduce the fissured kernels and improve the head rice
yield.

In general, a decrease in fissured kernel as the tempering time increased correlated very well with

the head rice yield (Figure 6). It should be noted that all the fissured kernels may not result in to
broken grain during milling. While the HRY values at various drying times and tempering times
from 30-60 min remained unchanged in OM2717 variety (Figure 4), the amount of fissured
kernels continued to decrease (Figure 5). The identical values of HRY may be explained by the
fact that some fissured kernels, which were not broken during milling, contributed to the amount
of head rice.

Various patterns of cracks were observed within fluidized dried rice kernels as shown in Figure 7.
The pattern (a), (b), and (c) may be counted to the head rice because they have a potential to
maintain ¾ length of kernel after undergoing milling procedure. It may also be possible that the
grains with the cracks close to two ends of the kernel may be more resistant to breakage during
milling than those who have cracks at the middle. Therefore, we believe that the enumeration of
fissured kernel is a better indicator of rice cracking than the head rice yield. In addition, we
experienced with our laboratory milling system that the head rice yield was also highly dependent
on settings and length of operation of the rice mill. We had to make many repetitions of
experiments and readjustments to the milling system to achieve an optimum operating condition.

86


A10

OM2717

90

70
60

80oC, 2.5 min

80oC, 3.0 min
90oC, 2.5 min
90oC, 3.0 min
Reference

80

Fissured kernels, %

Fissured kernels, %

90

80oC, 2.5 min
80oC, 3.0 min
90oC, 2.5 min
90oC, 3.0 min
Reference

80

50
40
30
20
1
0

70
60

50
40
30
20
10
0

0
0

20

40

60

80

0

20

Tem pering tim e, m in

40

60

80

Tempering time, min

a) A10 variety

b) OM2717 variety

Figure 5. Effect of tempering time on fissured kernels at drying temperatures of 800C and 900C
for 2.5 and 3.0 min. Reference fissured kernels of A10 and OM2717 were all 3%.

80
70

R2 = 0.8464

HRY, %

60
50
40
30
20
10

R2 = 0.6154

0
0

20

40

60

80

100

Fissured kernels, %
OM2717

A10

Reference-A10

Reference-OM2717

Figure 6. Relationship between head rice yield (HRY) and fissured kernels for A10 and
OM2717 varieties.

87


(a)

(b)

(c)

(d)

(e)

(f)

Figure 7. Types of cracks on rice kernels observed in this study.

Mechanical strength of rice:
With reference to the low temperature drying at 35oC, the mechanical strength of the rice kernels
was higher at high temperature drying (Table 2, Figure 8). This value was significantly higher at
90oC drying temperature than at 80oC. This can be attributed to the partial gelatinization of the
starch at the surface of the rice kernel. The gelatinization of starch makes the surface dense and
some of the surface cracks can be fused and disappear during this process.

Surprisingly, the breaking force of the rice kernels was not seen influenced by the tempering
durations. There was a large variability (30 to 100 N) of the measured force of the grains,
indicating that each kernel is different in terms of strength. The breaking strength of the kernels of
both varieties did not show much difference either. We can see from Figure 8 that the average
kernel strength of both varieties ranged from 30 to 55 N. Probably this range of strength is enough
to resist the breakage during milling. The presence of cracks or fissure in the rice kernels is the
only factor which could contribute to breakage. In our previous studies we had found that the
fissured grain (with different degree of fissures) and chalky grains had lower mechanical strength
than the unbroken kernels.

88


Table 2: Mechanical breaking strength of low temperature (35oC) and high temperature
dried rice (dried for 2.5 and 3 minutes)

Rice variety

Average mechanical strength of rice kernels (N)
80oC

Reference drying

90oC

(35oC)
2.5 min

3 min

2.5 min

3 min

A10

41.2

42.1

41.5

44.5

49.3

OM2717

39.7

33.7

35.9

47.2

52.6

A10

OM2717

50
Breaking force, N

60

50
Breaking force, N

60

40
30
20

80oC, 2.5 min
90oC, 2.5 min
Reference

10

80oC, 3.0 min
90oC, 3.0 min

40
30
20

80oC, 2.5 min
90oC, 2.5 min
Reference

10

80oC, 3.0 min
90oC, 3.0 min

0

0
0

20

40

60

Tempering time, min

(a) rice variety A10

80

0

20

40

60

80

Tempering time, min

(b) rice variety OM2717

Figure 8. Effect of tempering time and drying temperature on breaking force (mechanical
strength) of rice kernels.

Colour of milled rice
The color of the milled rice was measured expecting that the high temperature drying and
tempering can influence on the yellowness of the rice. This is due to the browning reaction at
higher temperature. The results indicated that the yellowness value was higher for the samples

dried at 90oC than those dried at 80oC. This value was also increased by the tempering time. The

89


total difference which is the combination of all principal colour parameters also changed as the
tempering time is increased. The actual observation by an eye showed some differences on the
untreated and treated samples, but all these changes were still at acceptable level as a colour of
commercial rice. This will be further investigated.

Some literatures suggest that the high temperature drying has a similar effect as an ageing of rice.
In our initial work we have found some changes on the cooking property (instrumental analysis) of
rice due to tempering. The samples are being analyzed now at UQ (Australia) since no facility is
available at Nong Lam University. This will be reported later.

A10

A10

40

14

80oC, 2.5 min
13

37
36

80oC, 2.5 min

34

80oC, 3.0 min

90oC, 2.5 min

35

Y ello wness, b*

Delta E*

38

90oC, 3.0 min

Reference

33

90oC, 2.5 min

80C, 3.0 min

90oC, 3.0 min

39

Reference

12
11
10
9

32
0

10

20

30

40

50

60

70

8

Tempering time, min

0

10

20

30

40

50

60

70

Tempering time, min

OM2717

OM2717
14
13

38

80oC, 2.5 min
90oC, 2.5 min

36

80oC, 3.0 min
90oC, 3.0 min

Yellowness, b*

Total color difference, Delta E*

40

34
32

12
11
10
9

28

80oC, 2.5 min

80oC, 3.0 min

90oC, 2.5 min

30

90oC, 3.0 min

8

0

10

20

30

40

50

60

70

0

Tempering time, min

(a) Total colour difference

10

20

30

40

50

60

70

Tempering time, min

(b) Yellowness

Figure 8. Effect of tempering time on total colour difference and yellowness of A10 and OM2717
varieties at various drying temperatures and drying times.

90


Conclusions
With the studies being done to date, the following conclusions can be drawn:
Within the investigated drying and tempering conditions, a fluidized bed drying
temperature of 80oC and tempering duration of 30-40 minutes can reduce the drying
time without significantly affecting the quality of rice (some quality aspects are still
being investigated).
Fluidized bed drying can significantly reduce the drying time used in flat bed driers.
The actual drying time used by flat bed driers ranges from 8-10 hours for wet paddy if
the farmers practice to dry down to safe level of moisture. Interestingly if the paddy
needs to be dried until 15-16% moisture, this type of drying system can be used as a
compact drier. If lower moisture content is required, say 14%, this drying system
working as a multi-pass principle can be used as a compact dryer.
The high temperature fluidized bed drying of high moisture paddy can be feasible. But
many factors should be considered such as variability of moisture of the paddy, effect
of impurities, investment cost, farmers’ adoptability, scale-up problem etc. We will

consider some of these aspects in future experiments.
Bibliography
Cnossen AG, Siebenmorgen TJ, Yang W, Bautista RC (2001) An application of glass transition
temparture to explain rice kernel fissure occurence during the drying process. Drying Technology
19, 1661-1682.
Kunze OR, Calderwood DL (2004) Rough-rice drying-Moisture adsorption and desorption. In
'Rice Chemistry and Technology'. (Ed. ET Champagne) pp. 223-268. (American Association of
Cereal Chemists, Inc.: St. Paul, Minnesota, USA).
Perdon AA (1999) Amorphous state transition in rice during the drying process University of
Arkansas.
Poomsa-ad N, Soponronnarit S, Prachayawarakorn S, Terdyothin A (2002) Effect of tempering on
subsequent drying of paddy using fluidisation technique. Drying Technology 20, 195-210.
Poomsa-ad N, Terdyothin A, Prachayawarakorn S, Soponronnarit S (2005) Investigations on headrice yield and operating time in the fluidised-bed drying process: experiment and simulation.
Journal of Stored Products Research 41, 387-400.
91


Prachayawarakorn S, Poomsa-ad N, Soponronnarit S (2005) Quality maintenance and economy
with high-temperature paddy-drying processes. Journal of Stored Products Research 41, 333-351.
Sutherland JW, Ghaly TF (1990) Rapid fluidised bed drying of paddy rice in the humid tropics. In
'Proceeding of the 13rd ASEAN Seminar on Grain Post-harvest Technology'.
Taweerattanapanish A, Soponronnarit S, Wetchakama S, Kongseri N, Wongpiyachon S, (1999)
Effects of drying on head rice yield using fluidization technique. Drying Technology 17, 345-353.
Tirawanichakul S, Prachayawarakorn S, Varanyanond W, Tungtrakul P, Soponronnarit S (2004)
Effect of fluidized bed drying temperature on various quality attributes of paddy. Drying
Technology 22, 1731-1754.

92