Renewable Energy 115 (2018) 954e968

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

The experimental new hybrid solar dryer and hot water storage
system of thin layer coffee bean dehumidification
S. Deeto a, S. Thepa a, *, V. Monyakul b, R. Songprakorp a
a
Division of Energy Technology, School of Energy, Environment and Materials, King Mongkut's University of Technology Thonburi, Bangkok 10140, Thailand
b
Biochemical Engineering and Pilot Plant Research and Development Unit (BEC), King Mongkut's University of Technology Thonburi (Bangkuntien),
Bangkok 10150, Thailand

a r t i c l e i n f o a b s t r a c t

Article history: The thin layer coffee beans dehumidification and hot water storage were investigated in solar green-
Received 3 June 2017 house dryer simultaneously. The thermal energy was stored for use in the absence of sunlight. The
Received in revised form conditions were studied the flow rate of water circulating, model of installation for solar collector
31 August 2017
assistance, the area ratio of product dehumidification to solar hot water producing (Ad:Ac), flow pattern
Accepted 4 September 2017
Available online 8 September 2017
of water circulation and capacity tank of water circulating in the system. The experiments were found the
flow pattern of water circulation within the force flow 0.20 kg/s-m^2, tray products temperature (45 deg.
C) at (Ad:Ac) 1:1, capacity tank of water circulating (60 L). The thermal energy can be used and stored in
Keywords:
Solar greenhouse dryer
the form of hot water and reused at a time without sunlight as well. The initial coffee beans moisture
Solar hot water storage content was dropped from 55 to below 12 (%w.b.) in 12 h drying time. The optimum of mathematical
Coffee bean dehumidification equations for thin layer coffee beans drying based on Midilli mathematical model i.e. (MR ¼ a exp(-kt^n)
Effective moisture diffusivity coefficient e bt); k ¼ 0.03838, n ¼ 1.56771, a ¼ 1.03046, b ¼ 0.00477, R2 ¼ 0.9896, RMSE ¼ 0.0420. Moreover, the
effective moisture diffusivity coefficient of 9.754  10^-11 m^2/s.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction drying is highly dependent on the a-vailability of sunshine or un-

controllable the environment weather and is susceptible to
Water content is the most important quality parameter of contamination from foreign materials and animals (i.e. dust, sand
green coffee bean, as it can control fermentation and mould as well as insect and fungal infestations), which grow in moist
growth during storage and transport, which lead to the develop- conditions. Such contaminations become the products unusable.
ment of flavors at the cup quality or the formation of contami- Moreover, most agricultural and fisheries products require drying
nation. It is also important for economical reasons that coffee is to preserve the quality of the final product, which open sun drying
commercial and more interesting to buy solid coffee materials [1]. increase the risk of low-quality products [3]. However, there are
Solar drying is commonly in practice since a long past time for some drawbacks associated with solar drying i.e. unsteady of
preservation of food and agricultural products. This was done environment weather, reliability of solar radiation during rainy
particularly with open sun drying under the open sky. However, period or cloudy days and its unavailability at night time. In a
the traditional sun drying is a relatively slow process which hybrid solar dryer, drying is continued during off sunshine hours
considerable losses can happen during the process. In addition, a by back-up heat energy or storage heat energy was selected [4].
reduction in the product quality takes place due to contaminate or Contact dryers with a high efficiency of heat transfer, have been
insect infestation, enzymatic reactions, microorganism growth used successfully to dry a wide variety of products ranging from
and mycotoxin development [2]. Furthermore, open sun drying food and dairy products such as baby food, potato flakes, butter-
requires a large open space and long drying times. Although this milk powder, and coffee grounds sludge to chemical and other
traditional method requires only a small investment, open sun miscellaneous products such as carbon black, colloidal clay, pig-
ments, peat, and many kinds of sludge [5].
This paper presents the experimental performances and
* Corresponding author. developed the greenhouse dryer for drying of thin layer coffee bean
E-mail address: [email protected] (S. Thepa).

http://dx.doi.org/10.1016/j.renene.2017.09.009
0960-1481/© 2017 Elsevier Ltd. All rights reserved.
 S. Deeto et al. / Renewable Energy 115 (2018) 954e968 955

covered with 1 mm thickness a clear plastic sheets. The front side

wall of the dryer has a door 0.15 m wide, 0.15 m long for loading of
products. As shown Fig. 2 the 0.5 kg of products are placed as a
thin layer on tray which made of a black aluminum plate with
0.5 m wide and 0.5 m long. Also, tray was designed with tubes
under it for the water circulation in case of heat transfer on plate.
The 5 inches DC fan was installed in the wall to suck out moisture
air on outlet channel with a same size of inlet channel in opposite
dryer wall.

2.1.2. Solar hot water collector

The solar hot water collector consists of transparent cover,
absorber, tubes and construction. The transparent cover is a
5 mm thickness clear glass with 70 mm aluminum frame and
10 mm insulation thickness polyurethane foam inside. The
absorber plate is made of 2 mm black color steel on the insu-
lating polyurethane foam 10 mm thickness. The absorber is fixed
at 25 mm below the clear grass cover and the tubes are con-
nected around the absorber plate for water circulation. The
construction is made of steel to install 4 square meters hot water
collector area. The solar hot water collector is south face direc-
Fig. 1. The schematic of the new solar dryer system.
tion with 14 slope angle.

2.1.3. Water storage tank

A cubic water storage tank of 180-L in maximum volume is
and solar hot water producing for storage heat energy. The effective
completely insulated with 10 mm thickness polyurethane foam all
moisture diffusivity and a mathematical drying model of thin layer
side. It made of 5 mm stainless steel and three channels port for K-
coffee bean dehumidification were investigated.
type thermocouple at three levels of water storage tank: top,
middle and bottom.
2. Materials and methods
2.1.4. Water circulation system
2.1. Model description of solar dryer system The water circulation system consists of water pump, pipes and
joint system, and water flow meter. The water pump (Wizz,
The new solar dryer system was designed, constructed and WCP130, kW 0.37, HP 0.5, In 2.5 A, Wmax 550, Hmax 22 m, Qmax
tested at the Division of Energy Technology, School of Energy, 90 L/min) is installed to provide a water flow circulation in system;
Environment and Materials, King Mongkut's University of Tech- solar greenhouse dryer, solar hot water collector and water storage
nology Thonburi, Bangkok, Thailand. The solar dryer system con- tank. The pipes and joint system are completely insulated with
sists of several parts: solar greenhouse dryer, solar hot water 10 mm thickness polyurethane foam. The water flow rate is
collector, water storage tank and water circulation system. The controllable by adjusting the water valve and monitoring the water
schematic of the new solar dryer system is shown in Fig. 1. flow meter.

2.1.1. Solar greenhouse dryer 2.2. Experimental procedure

The solar greenhouse dryer consists of a gable roof structure
made of PVC tubes, a clear plastic sheets cover and a black color Coffees products in Thailand were almost decreased the mois-
flat plate PVC floor. The dimension of solar greenhouse dryer is ture content with a traditional drying. The continuous dehumidi-
0.5 m wide, 0.6 m long and 0.7 m height. All sides of dryer are fication of products was a fermentation protection. The solar hot

Fig. 2. The product tray dimension.

956 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Table 1
The experimental condition setting.

Conditions Tank Volume (liters) Area Ratio (Ad:Ac) Flow Type Flow Rate (kg/s-m^2)

Without- Solar Collector Assistance 60 Force 0.20

0.13
0.06
0.03
Solar Collector Assistance 60 1:1 Thermosyphon
1:2
1:4
60 1:1
1:2
1:4
120 1:1 Force 0.20
1:2
1:4
180 1:1
1:2
1:4

water producing in the same time of drying was selected as inter- simultaneously. All experimental were conducted during summer
esting method to storage the heat energy for drying the products season of Thailand (MarcheMay, 2016) and the experimental po-
during off-sun. The coffee beans in wet process for experimental sition of measurements were shown in Fig. 1.
materials were approximately 7.3 mm diameter that had initial Solar radiation was measured by a pyranometer (Kipp & Zonen
moisture content of 55% (w.b.) with calculation from equation (6). model CM 11, accuracy ±0.5%) which was installed on a collector
The coffee beans were placed in the product tray with thin layer. plane to monitoring the global solar radiation incident. Tempera-
The experiments were started from 7.00 a.m. to 5 p.m. for a tures were measured by Type K thermocouples (±2%) which were
continuously 34 h. The drying was also continued until the final calibrated by checking the boiling and freezing temperature of
moisture content dropped to less than 12% (w.b.) and this moisture water. The relative humidity of ambient air and drying air were
content corresponded to keeping a high quality dried products measured by hygrometer (Electronik, model EE23, accuracy ±2%).
available from local markets. About 50 g of products were randomly The voltage signals of pyranometer and thermocouple were
selected at various positions and weighed by a digital balance recorded every 10 min with a multi-channel data logger (Yogogawa
(Mettler, Type PJ3000) at each two-hour intervals that it was the model HR 1300). The air flow rate at inlet and outlet channels of
representative of moisture content of product at any time. Finally, dryer was manually checked during the drying experiments with
the moisture content of products was calculated with equation (6) anemometer (DIGICON, DA-43A).
by the mass of products samples (w) and dry mass of products
samples (d) which were treated with hot air oven method using 103
2.4. Energy and moisture analysis
deg. C, for 72 h (Memert, model UFE700, accuracy ±0.5%). Total
experiments were tested with at least three repetitions by each
The energy absorption of solar collector was evaluated accord-
condition setting as illustrated in Table 1.
ing to ASHRAE standard [6,7]; fluid inlet, outlet temperature (Ti,To)
and fluid flow rate (m) _ were recorded accordingly. The amount of
2.3. Instrumentation energy (Qu) is calculated by using the following equation [8e11].

Coffees were dried to demonstrate the potentials of the solar _ p ðTo  Ti Þ

Qu ¼ mC (1)
greenhouse dryer for drying these products with water heating

Table 2
Mathematical model of drying [13e15].

No. Model Name Model

1 Newton MR ¼ exp(-kt)
2 Page MR ¼ exp(-ktn)
3 Modified Page MR ¼ exp(-(kt)n)
4 Henderson and Pebis MR ¼ a exp(-kt)
5 Wang and Singh MR ¼ 1 þ at þ bt2
6 Logarithmic MR ¼ a exp(-kt) þ c
7 Two term MR ¼ a exp(-k1t) - b exp(-k2t)
8 Two term exponential MR ¼ a exp(-kt) þ (1-a) exp(-kat)
9 Modified Henderson and Pebis MR ¼ a exp(-kt) þ b exp(-gt) þ c exp(-ht)
10 Midilli MR ¼ a exp(-ktn) e bt
11 Approximation of diffusion MR ¼ a exp(-kt) þ (1-a) exp(-kbt)
12 Verma et al. MR ¼ a exp(-kt) e (1-a) exp(-gt)
13 Logistic MR ¼ a0/(1 þ a exp(kt))
14 Thomson t ¼ a ln(MR) þ b (ln (MR))2
S. Deeto et al. / Renewable Energy 115 (2018) 954e968 957

Fig. 3. The experimental weather.

958 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Fig. 3. (continued).

The specific heat (Cp ) was given by 4.189 kJ/kg K in these case.  
The thermal efficiency of the solar collector (hc) which found from a mo Mi  Mf
general solar collector thermal efficiency equation was the ratio of W¼ (5)
100  Mf
useful energy (Qu) and the solar radiation incident on the plane of
collector (AcI). The thermal efficiency of the solar collector was The moisture content of product (M) was calculated by refer-
expressed that; ring equations (6) and (7) in wet basis and dry basis, respectively.
The mass of products samples (w) and dry mass of products
_ p ðTo  Ti Þ samples (d) were determined. The AOAC method [12] was
Q u mc
hc ¼ ¼ (2) suggested:
Ac I Ac I
The efficiency of drying system was the ratio of the energy ðw  dÞ
M¼ x100 (6)
required to evaporated moisture (WhL) and the heat supplied to w
dryer. The heat suppliers were the solar radiation incident on the
solar collector, heat energy of fan (Qf) and heat energy of pump ðw  dÞ
M¼ x100 (7)
(Qp). The efficiency of drying system was calculated for day time (3) d
and night time (4).

WhL
hd ¼ (3) 2.5. Mathematical model of drying
Ac It þ Q f þ Q p
The experimental data were used to fitting drying curves by 14
different moisture content ratio models [13e15] as illustrated in
WhL
hd ¼ (4) Table 2. The moisture ratio (MR) was obtained (8):
Qf þ Qp
ðM  Me Þ
The latent heat of vaporization (hL ) was given by 2264.76 kJ/kg. MR ¼ (8)
ðMo  Me Þ
The mass of water removed from a wet product (W) was calculated
by using amount of initial mass product (mo), the initial and final Therefore, the moisture ratio (MR) was reform to M/Mo. The
moisture content of product (Mi,Mf) which were shown as (5). equilibrium moisture content (Me), that using the final moisture
 S. Deeto et al. / Renewable Energy 115 (2018) 954e968 959

Fig. 4. The solar hot water producing-without solar collector assistance.

content of the dehydrated product until constant in the chamber 3. Results and discussion
test, has a low value in compare to the initial moisture content
(Me « Mo) and the moisture content at any time (Me « M) which is 3.1. Experimental results
shown in below equation [16,17].
3.1.1. The experimental weather
The environment of experiments such as solar radiation,
ambient temperature and relative humidity were recorded during
ðMÞ investigations. The solar radiation has a same trend as mountain
MR ¼ (9)
ðMo Þ curve in total cases (Fig. 3a). The ambient temperature was almost
stable while in the morning the temperature was a little lower than
The best equation was used the correlation coefficient (R2) and
other (Fig. 3b). The relative humidity in total test was highest in
Root Mean Square Error (RMSE) which were a main criteria for
morning which decreased at noon to the lowest and increased a bit
selecting the variation in the drying curves of the coffee bean
at evening was a little (Fig. 3c).
dehydrated samples [18].

Table 3
The solar hot water producing-without solar collector assistance.

Flow Rate Ambient Temp. % RH Avg. Solar Intensity Chamber Dryer Temp. Plate Dryer Temp. Energy Storage Thermal
(kg/s-m^2) Avg. (Deg. Cel.) Avg. (W/m^2) Avg. (Deg. Cel.) Avg. (Deg. Cel.) (kJ) Efficiency

0.03 30.4 65.5 564.9 33.5 35.2 3516.2 32.0

0.06 31.2 62.1 568.8 35.0 35.3 3767.4 33.8
0.13 30.3 65.4 573.3 33.9 36.5 3767.4 34.0
0.20 30.2 63.6 566.7 33.2 35.3 4269.7 38.8
960 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Fig. 5. The solar hot water producing-with solar collector assistance (Ad:Ac) in thermosyphon flow.

3.1.2. Solar hot water producing 3.1.2.2. Solar collector assistance (Ad:Ac) in thermosyphon and force
flow. The experimental results were investigated the solar hot
water which produced with solar greenhouse dryer and solar col-
3.1.2.1. Without solar collector assistance. The solar hot water pro-
lector assistance (Ad:Ac). The area ratio of product dehumidifica-
duced with solar greenhouses dryer and without solar collector
tion to solar hot water produced (Ad:Ac) at three different ratios;
assistance which installed the product tray inside the chamber to
1:1, 1:2 and 1:4 in each flow pattern with a natural flow (Ther-
absorb the radiation. The experiments were investigated at flow
mosyphon Flow) and force flow. For thermosyphon flow (Fig. 5aec
rate of water circulation in four levels: 0.03, 0.06, 0.13 and 0.20 kg/
and Table 4), the optimum area ratio (Ad:Ac) of 1:4 was accumu-
s-m^2 (Fig. 4aed) and the results were concluded in Table 3. The
lated the thermal energy in form of hot water 5776.7 kJ. The force
drying temperature was not excess 45 deg. C [19] which is the
flow were shown in Fig. 6aec and Table 4 respectively and the
optimum temperature for coffee beans drying. At the water flow
optimum area ratio of (Ad:Ac) 1:1 was accumulated the thermal
rate of 0.20 kg/s-m^2 the tray temperature was raised rapidly to the
energy in form of hot water 6781.3 kJ. The products tray temper-
optimum drying temperature and it could accumulate thermal
ature was in a range of 45 deg. C which was the optimum tem-
energy in the form of hot water and performance equivalent
perature for coffee beans drying.
4269.7 kJ and 38.8, respectively.

Table 4
The solar hot water producing-with solar collector assistance (Ad:Ac) in thermosyphon and force flow.

Flow Type Area Ratio Ambient Temp. Avg. % RH Avg. Solar Intensity Chamber Dryer Plate Dryer Energy
(Ad:Ac) (Deg. Cel.) Avg. (W/m^2) Temp. Avg. (Deg. Cel.) Temp. Avg. (Deg. Cel.) Storage (kJ)

Thermosyphon 1:1 31.6 56.2 581.9 35.0 35.5 4269.7

1:2 31.2 58.4 588.1 35.0 37.2 4520.9
1:4 31.0 62.6 594.3 33.9 35.3 5776.7
Force 1:1 31.1 61.6 637.8 37.9 39.5 6781.3
1:2 30.7 63.7 602.4 36.8 39.9 7032.5
1:4 29.4 65.5 613.0 36.9 43.2 8539.4
 Fig. 6. The solar hot water producing-with solar collector assistance (Ad:Ac) in force flow.

Fig. 7. The solar hot water producing-with solar collector assistance (Ad:Ac) in 120 L storage tank.
962 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Fig. 8. The solar hot water producing-with solar collector assistance (Ad:Ac) in 180 L storage tank.

3.1.2.3. The solar collector assistance (Ad:Ac) in 60, 120, 180 L storage is inappropriate temperature for coffee beans drying. The amount
tank. The experimental results were tested the solar hot water of water influenced the heat transfer from the plate inside chamber
produced with solar greenhouse dryer and with solar collector and stored in form of hot water which need a long period in
assistance (Ad:Ac) in 60, 120, 180 L storage tank. The ratio of area accordance with the greater capacity of water in system.
(Ad:Ac) was tested at different three ratios; 1:1, 1:2 and 1:4. The
flow rate of water circulating in the system 0.20 kg/s-m^2 in each 3.1.3. The moisture dehumidification and the solar hot water
storage tank were shown in Figs. 6aec, 7a-c, 8a-c and Table 5. The storage
capacity of water storage tank (60 L) with area ratio (Ad:Ac) of 1:1 3.1.3.1. Without solar collector assistance. The coffee beans drying
and storage tank (120 L) with area ratio of (Ad:Ac) 1:2, were and solar hot water produced with solar greenhouse dryers were
measured the products tray temperatures in the range of 45 deg. C. investigated together at the same time. The drying condition was
which is the optimal temperature for coffee beans drying. The ca- continuously tested with a thin layer arabica coffee beans drying.
pacity of water storage tank of 180 L in all area ratio (Ad:Ac) was As shown in Fig. 9b the temperature profile of product tray has
measured the products tray temperature at lower 45 deg. C which soared to align with the temperature of hot water storage tank

Table 5
The solar hot water producing-with solar collector assistance (Ad:Ac) in 60, 120, 180 L storage tank.

Tank Area Ratio Ambient Temp. Avg. % RH Avg. Solar Intensity Chamber Dryer Plate Dryer Energy
(Liters) (Ad:Ac) (Deg. Cel.) Avg. (W/m^2) Temp. Avg. (Deg. Cel.) Temp. Avg. (Deg. Cel.) Storage (kJ)

60 1:1 31.1 61.6 637.8 37.9 39.5 6781.3

1:2 30.7 63.7 602.4 36.8 39.9 7032.5
1:4 29.4 65.5 613.0 36.9 43.2 8539.4
120 1:1 31.4 64.7 623.3 34.6 34.7 7534.8
1:2 32.4 59.6 648.2 35.6 36.5 11,051.0
1:4 31.7 64.1 625.4 35.8 39.3 13,060.3
180 1:1 32.7 56.3 584.0 34.8 34.5 9795.2
1:2 32.6 59.3 586.0 34.8 35.6 12,809.2
1:4 32.6 60.0 581.9 34.5 37.0 15,069.6
 S. Deeto et al. / Renewable Energy 115 (2018) 954e968 963

during the first 10 h (7:00 a.m. to 17:00 p.m. on the same day). The water storage tank were accumulated the thermal energy similarly
temperature started to decline steadily until the period of 18 h at a and equally at certain period. As shown in Figs. 10a and 11a the
temperature of products tray and the temperature of the hot water thermal energy accumulations get positive over on sunlight and
storage tank equalization. Fig. 9a illustrates that the thermal energy during the night (without sunlight) energy was used to get nega-
accumulation over time on sunlight get positive and during the tive. However, thermal energy accumulation was restarted during
night (without sunlight) energy was used to get negative. Thermal the sunlight next day. As shown in Figs. 10c, 11c and 13 the coffee
energy accumulation was started again during the sunlight next bean drying and solar hot water produced with solar greenhouse
day. Using the coffee beans drying and hot water producing with dryer at the same time in cases of the natural and force flow of
solar greenhouse dryers in the same time, Fig. 9c cause to reduce water circulation. The natural and force flow of water circulation
the initial moisture content of coffee beans of 55 to a final moisture could reduced the initial moisture content of arabica coffee beans
content below 12% wet basis (w.b.) within 30 h. The moisture from 55 to below 12% wet basis (w.b.) within 26 and 12 h,
content of the arabica coffee beans dried shell in these conditions respectively.
was optimized for a storage standard [20].
3.1.3.3. Solar collector assistance (Ad:Ac) in 60 and 120 L storage
3.1.3.2. Solar collector assistance (Ad:Ac) in thermosyphon and force tank. The coffee bean drying and solar hot water producing with
flow. The coffee beans drying and solar hot water producing with solar greenhouse dryer was tested together at the same time. The
solar greenhouse dryers was operated together. The thin layer thin layer arabica coffee bean drying was tested continuously and
arabica coffee bean drying was continuously tested and the controlled the flow rate of the water circulation in the system
experimental results were shown in Figs. 10b and 11b. The water 0.20 kg/s-m^2. The results were demonstrated that the capacity of
circulation of natural flow or thermosyphon flow was displayed water storage tank (60 L) decreased with temperature of products
that the temperature profile of products tray increased with the tray in the same trend temperature of the hot water storage tank.
temperature of the hot water storage tank. However, the accumu- As a result the accumulated thermal energy was faster than the
lated thermal energy was slower than the water circulation by force capacity of water storage tank (120 L) (Figs. 11b and 12b). During
flow. Temperature of products tray and temperature of the hot the first 10 h, were shown the temperature of products tray and hot

Fig. 9. The moisture and the solar hot water producing-without solar collector assistance (Ad:Ac) in 60 L storage tank, force flow.
964 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Fig. 10. The moisture and the solar hot water producing-with solar collector assistance (Ad:Ac) in 60 L storage tank, thermosyphon flow.

water storage tank accumulated the thermal energy similarly and in the form of hot water and rotated again at a time of no light. The
equally as certain period. As shown in Figs. 11a and 12a the thermal moisture content of coffee bean was reduced to lower than 12% wet
energy accumulated over sunlight get positive and during the night basis (w.b.). These suitable models were found a thin layer coffee
(without sunlight) energy was used to get negative. Moreover, the bean drying as effective moisture diffusivity coefficient (Tables 6
thermal energy accumulation restarted again during in the sunlight and 7). The effect of flow types, area ratio and tank volume
next day. Figs. 11c, 12c and 14 were shown the coffee bean drying impacted to the moisture content decreasing and the effective
and solar hot water producing with solar greenhouse dryer at the moisture diffusivity coefficient. These conditions related to math-
same time (the capacity of the water storage tank with 60 and ematical model to explain and consider for predicting the drying
120 L). The thin layer arabica coffee bean drying reduced the initial characteristic of coffee bean drying.
moisture content continuously from 55 to below 12% wet basis
(w.b.) within 12 and 16 h, respectively.
4. Conclusions

3.2. Mathematical model of coffee bean drying The thin layer coffee bean drying and solar hot water producing
with solar greenhouse dryer were installed simultaneously with
The thin layer coffee bean drying and solar hot water producing solar collector assistance in the area ratio of product dehumidifi-
with solar greenhouse dryer were investigated in four cases: I. cation to solar hot water producing (Ad:Ac) 1:1 and the force flow
storage tank 60L, force flow, without-assist. solar collector II. stor- pattern of water circulation 0.20 kg/s-m^2. The temperature of
age tank 60L, thermosyphon flow, assist. solar collector III. storage products tray (45 deg. C) was controlled in a capacity of the water
tank 60L, force flow, assist. solar collector IV. storage tank 120L, storage tank (60 L) and stored the thermal energy in the form of hot
force flow, assist. solar collector. The thermal accumulated energy water and rotated at a time of no sunlight. Therefore, the moisture
 Fig. 11. The moisture of the solar hot water producing-with solar collector assistance (Ad:Ac) in 60 L storage tank, force flow.

Fig. 12. The moisture and the solar hot water producing-with solar collector assistance (Ad:Ac) in 120 L storage tank, force flow.
966 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Fig. 13. The moisture of the solar hot water producing 60 L storage tank.

Fig. 14. The moisture of the solar hot water producing 60 and 120 L storage tank.

Table 6
Description of coffee bean drying and solar hot water producing with solar greenhouse dryer.

Solar Collector Tank Volume (Liters) Flow Type Drying Time (h) Moisture Diffusivity (m^2/s) R2 RMSE

Without-Assist. 60 Force 30 7.804  10^-11 0.9550 0.2339

Assist. 60 Thermosyphon 26 9.154  10^-11 0.9540 0.2017
Assist. 60 Force 12 9.754  10^-11 0.8470 0.1654
Assist. 120 Force 16 9.304  10^-11 0.8930 0.1706
 S. Deeto et al. / Renewable Energy 115 (2018) 954e968 967

Table 7
The constant of mathematical model of drying.

No. Model Name Conditions Model R2 RMSE

1 Newton I. k ¼ 0.04795 0.9887 0.0373

II. k ¼ 0.06847 0.9777 0.0607
III. k ¼ 0.09890 0.9652 0.1248
IV. k ¼ 0.08487 0.9686 0.0923
2 Page I. k ¼ 0.04614,n ¼ 1.01307 0.9888 0.0372
II. k ¼ 0.05819,n ¼ 1.06001 0.9783 0.0599
III. k ¼ 0.09774,n ¼ 1.00498 0.9652 0.0761
IV. k ¼ 0.09438,n ¼ 0.95803 0.9690 0.0704
3 Modified Page I. k ¼ 0.04801,n ¼ 1.01308 0.9888 0.0392
II. k ¼ 0.06836,n ¼ 1.06002 0.9783 0.0748
III. k ¼ 0.09887,n ¼ 1.00498 0.9653 0.0762
IV. k ¼ 0.08511,n ¼ 0.95803 0.9689 0.0761
4 Henderson and Pebis I. k ¼ 0.04833,a ¼ 1.00615 0.9888 0.0372
II. k ¼ 0.07359,a ¼ 1.06692 0.9810 0.0562
III. k ¼ 0.10530,a ¼ 1.05942 0.9672 0.0740
IV. k ¼ 0.08892,a ¼ 1.04263 0.9698 0.0694
5 Wang and Singh I. a ¼ 0.04060,b ¼ 0.00049 0.9862 0.0412
II. a ¼ 0.05918,b ¼ 0.00103 0.9799 0.0576
III. a ¼ 0.07603,b ¼ 0.00155 0.9619 0.0797
IV. a ¼ 0.06983,b ¼ 0.00137 0.9684 0.0710
6 Logarithmic I. k ¼ 0.04098,a ¼ 1.08316,c ¼ 0.09033 0.9892 0.0364
II. k ¼ 0.09277,a ¼ 1.01351,c ¼ 0.08712 0.9844 0.0509
III. k ¼ 0.14028,a ¼ 1.01705,c ¼ 0.08972 0.9778 0.0610
IV. k ¼ 0.12574,a ¼ 0.99057,c ¼ 0.10916 0.9813 0.0548
7 Two term I. k1 ¼ 0.04833, k2 ¼ 0.04833, a ¼ 0.68122,b ¼ 0.32492 0.9888 0.0372
II. k1 ¼ 0.08396, k2 ¼ 0.05816, a ¼ 1.08522,b ¼ 0.01154 0.9849 0.0642
III. k1 ¼ 0.12657, k2 ¼ 0.04974, a ¼ 1.08164,b ¼ 0.02027 0.9790 0.0804
IV. k1 ¼ 0.03357, k2 ¼ 0.11409, a ¼ 0.03772,b ¼ 1.05772 0.9818 0.0804
8 Two term exponential I. k ¼ 0.04795,a ¼ 0.99996 0.9887 0.0373
II. k ¼ 0.08461,a ¼ 0.59832 0.9779 0.0604
III. k ¼ 0.15007,a ¼ 0.47200 0.9674 0.0737
IV. k ¼ 0.14261,a ¼ 0.42528 0.9719 0.0670
9 Modified Henderson and Pebis I. k ¼ 0.04833,a ¼ 0.33538,g ¼ 0.04833, b ¼ 0.33538, h ¼ 0.04833, c ¼ 0.33538 0.9888 0.0372
II. k ¼ 0.08396,a ¼ 0.56577,g ¼ 0.08396, b ¼ 0.51945, h ¼ 0.05816, c ¼ 0.01154 0.9849 0.0501
III. k ¼ 0.12657,a ¼ 0.41778,g ¼ 0.12657, b ¼ 0.66386,h ¼ 0.04973, c ¼ 0.02027 0.9790 0.0594
IV. k ¼ 0.03357,a ¼ 0.03772,g ¼ 0.11409, b ¼ 0.39742,h ¼ 0.11409, c ¼ 0.66030 0.9818 0.0792
10 Midilli I. k ¼ 0.09008, n ¼ 0.00001, a ¼ 0.83374,b ¼ 0.02407 0.9647 0.1520
II. k ¼ 0.04001, n ¼ 1.33553, a ¼ 1.04853,b ¼ 0.00478 0.9899 0.0410
III. k ¼ 0.03838, n ¼ 1.56771, a ¼ 1.03046,b ¼ 0.00477 0.9896 0.0420
IV. k ¼ 0.04987, n ¼ 1.37111, a ¼ 1.04205,b ¼ 0.00512 0.9883 0.0435
11 Approximation of diffusion I. k ¼ 0.06229, a ¼ 1.40239,b ¼ 0.89600 0.9888 0.0371
II. k ¼ 3.87536, a ¼ 0.13173,b ¼ 0.02028 0.9839 0.0517
III. k ¼ 0.11011, a ¼ 0.99156,b ¼ 0.65847 0.9737 0.0663
IV. k ¼ 0.08488, a ¼ 0.02560,b ¼ 0.99993 0.9686 0.0707
12 Verma et al. I. k ¼ 0.04833, a ¼ 1.00307,g ¼ 0.04832 0.9888 0.0372
II. k ¼ 0.08946, a ¼ 1.04915,g ¼ 0.01687 0.9846 0.0505
III. k ¼ 0.13288, a ¼ 1.04872,g ¼ 0.02171 0.9786 0.0599
IV. k ¼ 0.11597, a ¼ 1.04744,g ¼ 0.02657 0.9818 0.0540
13 Logistic I. k ¼ 0.04963, a0 ¼ 27.38622,a ¼ 26.27754 0.9888 0.0372
II. k ¼ 0.07537, a0 ¼ 311161.9,a ¼ 285487.2 0.9806 0.0567
III. k ¼ 0.10605, a0 ¼ 861149.0,a ¼ 807592.8 0.9672 0.0740
IV. k ¼ 0.08933, a0 ¼ 1928288,a ¼ 1841709 0.9698 0.0694
14 Thomson I. a ¼ 9.06647,b ¼ 18.2340 0.9886 0.0375
II. a ¼ 9.914582,b ¼ 22.4741 0.9817 0.0551
III. a ¼ 5.43324,b ¼ 10.9674 0.9713 0.0694
IV. a ¼ 6.11908,b ¼ 12.3389 0.9720 0.0669

Recommendations
I. 60 L, drying & producing (Without-Assist. solar collector), II. 60 L, Thermosyphon (Assist. solar collector).
III. 60 L, Force Flow (Assist. solar collector), IV. 120 L, Force Flow (Assist. solar collector).

content of coffee beans could be reduced lower 12% wet basis (w.b.) Acknowledgements
drying 12 h with the optimal mathematical model based on Midilli
mathematical model (MR ¼ a exp(-kt^n)-bt) a constant The authors would like to thank Division of Energy Technology,
k ¼ 0.03838, n ¼ 1.56771, a ¼ 1.03046, b ¼ 0.00477, R2 ¼ 0.9896, School of Energy, Environment and Materials, King Mongkut's
RMSE ¼ 0.0420 and the effective moisture diffusivity coefficient University of Technology Thonburi and Division of Science, Faculty
9.754  10^-11 m^2/s. of Science and Agricultural Technology, Rajamangala University of
Technology Lanna, Tak Campus for supporting experimental
apparatus.
968 S. Deeto et al. / Renewable Energy 115 (2018) 954e968

Nomenclature technologies, Renew. Sustain. Energy Rev. 16 (2012) 2652e2670.

[3] Serm Janjai, Poolsak Intawee, Jinda Kaewkiewa, Chanoke Sritus,
Vathsana Khamvongsa, A large-scale solar greenhouse dryer using poly-
A surface area (m^2) carbonate cover: modeling and testing in a tropical environment of Lao
Cp specific heat (kJ/kg K) People's Democratic Republic, Renew. Energy 36 (2011) 1053e1062.
d dry matter weight (kg) [4] E.A. Arinze, G.J. Schoenau, S. Sokhansanj, Design and experimental evaluation
of a solar dryer for commercial high-quality hay production, Renew. Energy
hL latent heat of evaporization (kJ/kg) 16 (1999) 639e642.
I solar radiation (W/m^2) [5] Marco Intelvi, Apolinar Picado, Joaquin Martinez, Contact drying simulation of
m_ mass flow rate (kg/s) particulate materials: a comprehensive approach, World Acad. Sci. Eng.
Technol. Int. J. Chem. Mol. Nucl. Mater. Metall. Eng. 5 (No 11) (2011).
M moisture content (%) [6] ASHRAE, Methods of Testing to Determine the Thermal Performance of Solar
Me equilibrium moisture content (%) Collectors. New York (USA), American Society of Heating, Refrigeration and
Mi initial moisture content (%) Air Conditioning Engineers, 1977.
[7] J.A. Duffie, W.A. Beckman, Solar Engineering of Thermal Processes, 2 nd ed.,
Mf final moisture content (%) Wiley-Interscience, New York, 1991.
Q amount of heat energy (kJ) [8] B.M.A. Amer, M.A. Hossain, K. Gottschalk, Design and performance evaluation of a
t time (h) new hybrid solar dryer for banana, Energy Convers. Manag. 51 (2010) 813e820.
[9] Ahmad Fudholi, Kamaruzzaman Sopian, Mohammad H. Yazdi, Mohd Hafidz
T temperature (K) Ruslan, Mohamed Gabbasa, Hussein A. Kazem, Performance analysis of solar
w initial weight (kg) drying system for red chili, Sol. Energy 99 (2014) 47e54.
W mass of water evaporation (kg) [10] Alireza Mohajer, Omid Nematollahi, Mahmood Mastani Joybari, Seyed Ahmad
Hashemi, Mohammad Reza Assari, Experimental investigation of a hybrid
h efficiency (decimal)
solar drier and water heater system, Energy Convers. Manag. 76 (2013)
935e944.
Subscripts [11] I. Attar, N. Naili, N. Khalifa, M. Hazami, A. Farhat, Parametric and numerical
study of a solar system for heating a greenhouse equipped with a buried
a air
exchanger, Energy Convers. Manag. 70 (2013) 163e173.
c collector [12] AOAC, Official Methods of Analysis of the Association of Official Analytical
d drying Chemists, fourteenth ed., Association of Official Analytical Chemists, Arling-
f fan ton, (VA), 1984.
[13] T.Y. Tunde-Akintunde, Mathematical modeling of sun and solar drying of chilli
g global pepper, Renew. Energy 36 (2011) 2139e2145.
i inlet [14] G.P. Sharma, R.C. Verma, Pankaj Pathare, Mathematical modeling of infrared
o outlet radiation thin layer drying of onion slices, J. Food Eng. 71 (2005) 282e286.
[15] C. Ertekin, O. Yaldiz, Drying of eggplant and selection of a suitable thin layer
p pump drying model, J. Food Eng. 63 (2004) 349e359.
s solar [16] L.M. Diamente, P.A. Munro, Mathematical modeling of the thin layer solar
t total drying of sweet potato slices, Sol. Energy 51 (4) (1993) 271e276.
[17] O. Yaldız, C. Ertekin, Thin layer solar drying of some different vegetables, Dry.
w water Technol. 19 (3) (2001) 583e596.
[18] O. Yaldız, C. Ertekin, H.I. Uzun, Mathematical modeling of thin layer solar
References drying of Sultana grapes, Energy 26 (5) (2001) 457e465.
[19] M.A. Sfredo, J.R.D. Finzer, J.R. Limaverde, Heat and mass transfer in coffee
fruits drying, J. Food Eng. 70 (2005) 15e25.
[1] C.T. Reh, A. Gerber, J. Prodolliet, G. Vuataz, Water content determination in
[20] Edward Winston, Jacques Op de Laak, Tony Marsh, Herbert Lempke,
green coffee e method comparison to study specificity and accuracy, Food
Keith Chapman, Arabica Coffee Manual for Lao-PDR, FAO Regional Office for
Chem. 96 (2006) 423e430.
Asia and the Pacific, Bangkok, Thailand, 2005.
[2] S. VijayaVenkataRamana, S. Iniyanb, Ranko Goicc, A review of solar drying