Energy 64 (2014) 663e672

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Improvement of the greenhouse climate using a solar air heater with

latent storage energy
Salwa Bouadila*, Sami Kooli, Safa Skouri, Mariem Lazaar, Abdelhamid Farhat
Research and Technology Center of Energy, Thermal Processes Laboratory, Hammam Lif, B.P. 95, 2050 Tunis, Tunisia

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

Article history: The use of latent heat to greenhouse heating in winter days is an important issue. The storage of the
Received 22 May 2013 excess heat in greenhouses for sunny days in cold season is necessary with the increasingly up of the
Received in revised form fossil fuel cost.
23 August 2013
In this research, we used two identical greenhouses. One, uses a latent heat system for heating,
Accepted 19 October 2013
constructed and installed in the CRTEn (Research and Technologies Centre of Energy) in Tunisia. This
Available online 21 November 2013
heating system consists of a new solar air heater collector using a packed bed of spherical capsules with
phase change materials. The solar air heater was inserted inside a greenhouse planted with tomato
Keywords:
Latent heat storage
plants and operated during the winter season. The nighttime recovered heat of this system attained 30%
Packed bed solar air heater of the total requirements of heating.
Control climate greenhouse Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction renewable and fossil-fuel sources for a greenhouse. Solar energy is

one of the important non-conventional sources which has been
The tomato is one of the most popular greenhouse vegetable used for heating greenhouses. Solar systems presented some
products in Tunisia. It is a hot-season plant and grows rapidly at technical problems and were too expensive for commercial use
daily temperatures between 20e30
C and 14e18
C at night. [12]. Bargach et al. [13] used a solar fat-plate collector to improve
Tunisia cultivated 26,000 tones of tomatos under greenhouse in the internal greenhouse microclimate and tested it under two types
2008 [1,2]. of days; a clear day and a cloudy day.
From 2005 to 2008, the greenhouse areas of Tunisia increased Many concepts of stored excess energy inside greenhouses have
from 4600 ha to 8683 ha [1,2]. Various types of heating systems been developed, such as the use of the rock beds [14,15]. Lazaar
used in the conventional greenhouses utilize the fossil fuels. Then, et al. [16] developed a capillary polypropylene exchange system for
the increasing of the price of the fossil fuels causes a higher pro- storing excess solar radiation in greenhouse soil. This system is
duction cost and a decrease the income of farmers [3]. used to attenuate the differences between the diurnal and
One of the major problems encountered in greenhouses is the nocturnal air temperatures under the tunnel greenhouse. In winter,
control of the internal climate. The lack of heating has unfavorable temperatures reached 20
C during the day and 12
C at night.
effects on the precocity of production. Sethi et al. [4] evaluated all Latent heat storage using phase change materials (PCMs) is one
heating technologies for worldwide agricultural greenhouses and of the most efficient ways of storing thermal energy for greenhouse
discussed the representative applications of each technology. The applications [17e20]. Boulard et al. [21] used a CaCl26H2O latent
heat pump systems have been successfully applied for heating, heat storage system in a double-skin polycarbonate greenhouse
cooling and dehumidification greenhouse [3,5,6]. equipped with a forced ventilation system. They found that the
Many heating systems for air-conditioning greenhouse used inside of the greenhouse temperature is 10
C higher than the
renewable energy sources [7e10]. Hepbasli [11] studied the energy outside temperature during typical nights of March and April.
and exergy analyses of three heating methods for greenhouse The aim of this work is to determine the provided heat by a
heating, (i) a solar assisted vertical ground-source heat pump, (ii) a SAHLSC (Solar Air Heater with Latent Storage Collector) inside an IG
wood biomass boiler, and (iii) a natural gas boiler, driven by (Insulated Greenhouse). Experiments were carried out in two
identical greenhouses designed and realized in the CRTEn
* Corresponding author. Tel.: þ216 97772206; fax: þ216 71430934.
(Research and Technology Center of Energy) in Tunisia. We will
E-mail addresses: salwa.bouadila@crten.rnrt.tn, bouadilasalwa@yahoo.fr describe in Section 2, the design of the experimental insulated
(S. Bouadila). greenhouses and the setup. In Section 3, a thermal analysis of the

0360-5442/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.energy.2013.10.066
664 S. Bouadila et al. / Energy 64 (2014) 663e672

Nomenclature l thermal conductivity (W/m K)

r reflectivity
A surface area (m2) s StefaneBoltzmann constant, 5.670108 W m2 K4
Cp specific heat of air at constant pressure (J kg1 K1) s transmissivity
d density (kg m3)
h heat transfer coefficient (W/m2
C) Subscripts
IT total solar radiation (W m2) a air inside greenhouse
l characteristic length of the leaf canopy (m) c cover
LAI leaf area index ca canopy
P(T) saturated water vapor pressure at temperature T (kPa) H heat
Q heat rate (W) i inside greenhouse
ra aerodynamic resistance (s m1) inf infiltration
rs stomatal resistance (sm1) o outside greenhouse
RE rate of air infiltration (m3/s) s soil
T temperature (K) sky sky
U velocity (m s1)
z depth of soil (m) Exposants
A absorbed heat
Greek symbols C convective heat
a absorptivity for solar radiations Cd conductive heat
at absorptivity for thermal radiations L latent heat
ε emissivity R radiative heat
g psychometric constant, 0.0667 k Pa K1

climate greenhouse is carried out. In Section 4, we will report the shown in Fig. 2. Eleven K-type thermocouples placed in the IG as
experimental results. The main remarks of this work will be re- follows: four sensors to measure the horizontal distributions of
ported in the conclusion. temperature; three sensors for the vertical temperature distri-
butions, four sensors for measuring the temperatures at the
2. Materials and methods top and the bottom of the plexiglass and sandwich panels. The
inside greenhouse temperature and relative humidity sensor
In this study, the two identical greenhouses were oriented to (HMP155A, Campbell Scientific Inc) was inserted in the center at
EasteWest, shown in Fig. 1(a). The first is an Insulated Greenhouse 1.5 m of level from the ground. The HMP155A sensor was put
(IG); the second used a solar air heater with latent storage collector into the 14-plate radiation shield houses to protect from solar
(SAHLSC) and was named IGLHS (Insulated Greenhouse with Latent radiation. Its louvered construction allows air to pass freely
Heat System). through the shield, thereby keeping the probe near ambient
temperature [22]. Temperatures under the soil inside the IG at 0,
2.1. Experimental greenhouse design and site description 0.25 and 0.5 m depth were measured using three PT-107 sensors.
The average temperature of the canopy was measured using an
The experimental small chapel-shaped greenhouse occupies a infrared temperature sensor IR120. The IR-SS Solar Shield is used
floor area equal to 14.8 m2 (3.7 m  4 m) and 3 m of height Fig. 1(c). to protect the IR120 sensor from direct solar radiation [22].
The greenhouse structure is a galvanized steel fixed to the ground The global solar irradiation in the horizontal plane was
with stones and concrete (Fig. 1(d)). The greenhouse wall and roof measured with a Kipp and Zonen pyranometer. It’s situated at 1.5 m
oriented to the south are covered by plexiglass with 3 mm of above the ground in the center of the IG. Ambient temperature and
thickness. Sidewalls and the northern roof are built by sandwich relative humidity are measured by a HMP155A sensor situated at a
panels with 0.4 m and 0.6 m of thickness, respectively. The slopes of height of 1.5 m above the greenhouse.
the southern wall are equal to 30
and of the southern roof is 33
. All climatic and measured parameters are sample recorded
The greenhouse was equipped with a centrifuge fan controlled by a every 10 min using a CR5000 data logger (Campbell Scientific Inc).
differential thermostat. The external global solar radiation, wind speed and direction are
A tomato crop (local variety) was planted in both greenhouses provided by the NRG weather station in the Research and Tech-
Fig. 1(b), organized of 7 plants in the row and arranged in 4 nology Center of Energy in Tunisia. This station is equipped with an
rows with 40 cm between the rows and 35 cm between the acquisition system that record, every 10 min the average, the
plants. maximum, the minimum and the standard deviation values for
The two insulated greenhouses are located in the Research each sensor. The weather station provided a wind speed mea-
and Technology Center of Energy in Borj Cédria, on the Medi- surement at 20 m. The wind velocity at the altitude of 2 m above
terranean coast of North Africa, near the city of Tunis in Tunisia, the greenhouse was corrected by using the power law equation
with the following coordinates: Latitude 36
430 N and Longitude [23]:
10
250 .
b
ðU1 =U2 Þ ¼ ðh1 =h2 Þ (1)
2.2. Measuring equipments inside and outside the greenhouse
where U1 andU2 are the wind speed at height h1 and h2 respectively.
A schematic view of the localization of all measuring in- b is the power law exponent and the subscripts (1 and 2) corre-
struments inserted inside and outside of the greenhouse is spond to different heights.
 S. Bouadila et al. / Energy 64 (2014) 663e672 665

Fig. 1. (a) External, (b) internal views of greenhouse, (c) a schematic section and dimensions of greenhouse (d) uncovered greenhouse.

2.3. Experimental latent heat system and measurements and technology aspects were taken into consideration in the
development of the IGLHS system.
The research work presented in this paper has been preceeded The experimental greenhouse with latent heat system (IGLHS) is
by an experimental evaluation of the thermal performance of a shown in Fig. 3(a). During the sunshine period, the charging pro-
new solar air heater collector using a packed bed of spherical cess, a fraction of the total solar radiation received inside the
capsules with a latent heat storage system [24]. A new SAHLSC greenhouse is absorbed by the black packed bed absorber of the
(Fig. 3(b)) was used as a means to heat the interior environment of SAHLSC. The absorbed thermal energy is stored as sensible and
the greenhouses during the nighttime. Fig. 3(a) show the SAHLSC latent heat forms into the collector. In the sunset and during the
put on the greenhouse. The length, the width and the total volume night, the greenhouse air temperature drop causes by a radiation
of the collector are 2 m, 1 m and 0.28 m3, respectively. A 0.004 m heat exchange in the IGLHS. In this time a fan blows air across the
thick transparent glass cover placed at 0.015 m above a packed bed PCM capsules and extracted the stored heat into the IGLHS, so the
absorber. A 0.05 m thick polyurethane insulation, with heat con- discharging process is done.
ductivity of 0.028 W m1 K1, is placed in the bottom and the edge
of the collector. The packed bed absorber is the most important 2.5. Physical characteristics
component of the solar air heater collector. It is formed of spherical
capsules with a black coating and fixed with steel matrix. Capsules Physical characteristics of the greenhouse glass cover, canopy
with 0.077 m of outer diameter, confined with a PCM, they are and soil are listed in Table 2. The reflectivity and transmissivity
blow molded from a blend of polyolefin with a thickness of spectra of the canopy and cover layer recorded in the wavelength
0.002 m. range (380e900 nm) plotted in Figs. 4 and 5 respectively. The
Temperatures along the SAHLSC were measured using 8 K-type reflectivity and transmissivity spectra realized using LAMBDA950
thermocouples. Four sensors were inserted inside capsules along UVeviseNIR spectrometer equipped with an integrating sphere.
the absorber packed bed. Also four thermocouples are used to
measuring the absorber surface of one capsule, the glass, the inlet 2.6. Uncertainty analysis
and the outlet air temperatures of the SAHLSC.
The thermo-physical properties of the capsule, air and soil, Uncertainty analysis is needed to prove the accuracy of the ex-
which were used, are given in Table 1. periments. The independent parameters measured in the experi-
ments reported here: are temperature, air velocity, and solar
2.4. Experimental setup and methodology radiations. To carry out these experiments, the sensitiveness of data
acquisition system is about 0.001
C, the measurement error is
An experimental setup of two identical insulated greenhouses, 0.002
C, the sensitiveness of the thermocouple is 0.01
C, the
one of them used a heating system, designed and constructed to sensitiveness of the IR120 is 0.0004
C and sensitiveness of the
investigate a comparative study. The insulated greenhouse design PT107 is 0.01
C. The HMP155A errors are 0.02
C of
666 S. Bouadila et al. / Energy 64 (2014) 663e672

Table 1
Thermal properties of PCM, air and soil.

Material Melting Heat of Specific Density Thermal

Point fusion heat (kg/m3) conductivity
(
C) (kJ kg1) (kJ/kg
C) (W/m
C)

Capsule 27 192.6 Liquid Solid Liquid Solid Liquid Solid

(AC27) 2.22 1.42 1710 1530 0.58 1.05
Air e e 1.0048 1.137 2.49  102
(at 25
C)
Soil e e 1480 1620 2.5

- Radiation heat exchanges between the greenhouse walls and

roofs are neglected.
- Conduction heat exchange between air and the sandwich panel
walls and roof of the greenhouse is neglected.
- Absorbtivity and heat capacity of the enclosed air is neglected.

3.1. Energy balance of greenhouse cover

Fig. 2. A schematic view of the localization sensors. Energy balance of the cover of greenhouse includes: solar ra-
diations absorbed by the plexiglass of the greenhouse QcA , solar
radiations reflected by the canopy and absorbed by cover Qcac A ,
temperature and 2% of humidity. An anemometer with
0.01 m s1 accuracy, and Kipp and Zonen pyranometer with 3% solar radiations reflected by the soil surface and absorbed by cover
A , convective heat transfer from external cover of the green-
Qsc
measurements uncertainties are used. The sensitiveness was ob-
house to ambient air QcoC , convective heat transfer between in-
tained from catalogs of the instruments.
ternal cover and inside air Qci C , thermal radiation exchange
The calculated uncertainties of the dependent parameters were
R
between cover and sky Qcsky, thermal radiation emitted by soil
estimated by Eq. (2). The result R is a given function in terms of the
surface and absorbed by cover QscR , thermal radiation emitted by
independent variables. Let wR be the uncertainty in the result
R
canopy and absorbed by cover Qcac.
andw1, w2,.,wn be the uncertainties in the independent variables.
The result R is a given function of the independent varia- The energy balance of the cover is given by:
blesx1,x2,.,xn. If the uncertainties in the independent variables are
all given with the same odds, then the uncertainty in the result QcA þ Qcac
A A
þ Qsc C
 Qco C
 Qci R
þ Qcsky R
þ Qcac R
þ Qsc ¼0 (3)
having these odds is calculated by Ref. [25]:
The convective heat transfer between internal cover and air
" 2  2  2 #1=2 inside the greenhouse is:
vR vR vR
wR ¼ w1 þ w2 þ. þ wn (2)
vx1 vx2 vxn C
Qci ¼ IT Ac ac þ IT sc ac rca Aca þ IT sc ac rs As þ hCco Ac ðTo  Tc Þ
 
þ sεc Ac Tsky4
 Tc4 þ As sat;c εs Ts4 þ Aca sat;c εca Tca4

3. Thermal analysis greenhouse climate

(4)
In order to analyze the greenhouse climate, the energy balance Different relations have been used to calculate the external heat
equations for different components (cover, canopy, soil and inside loss coefficient at the outside surface of greenhouse covers [26e28].
air) of the greenhouse system with following general assumptions Mac Adams et al. [27] suggested a relation, which used in this study:
are made:
hCco ¼ 5:7 þ 3:8Uo (5)
- The storage capacity of the walls and roof material (sandwich
panels) is neglected. where Uo is the wind speed outside the greenhouse.

Fig. 3. (a) External view of IGLHS and (b) SAHLSC inside greenhouse.
 S. Bouadila et al. / Energy 64 (2014) 663e672 667

Table 2 The aerodynamic resistance, ra (s m1) of the canopy, mainly

Cover, canopy and soil characteristic. depends on the aerodynamic regime that prevails in the green-
Cover Canopy Soil house and the leaf length. It is expressed as [31]:
Reflectivity for solar radiations 0.10 0.1 0.2
Reflectivity for thermal radiations 0.10 0.55 0.16 l0:2
ca
ra ¼ 220 (11)
Emissivity 0.9 0.05 0.9 Ui0:8
Transmissivity for solar radiations 0.85 0 0
Transmissivity for thermal radiations 0.88 0.4 0
where lca is the characteristic length of the leaf and Ui is the mean
interior air speeds.
The appropriate approximation to sky temperature is given by The stomatal resistance rs (s m1) of the canopy is derived from
Swinbak [29]: a simple empirical relationship, Eq. (12), which depends on the
global radiation [31e33]:
Tsky ¼ 0:0552To1:5 (6)  
1
rs ¼ 200 1 þ (12)
expð0:05ðsc IT  50Þ
3.2. Energy balance of canopy surface

3.3. Energy balance of soil surface

The energy balance of canopy surface inside the greenhouse
includes: shortwave radiations absorbed by the canopy Qca A , ther-
R The energy balance of soil surface inside the greenhouse in-
mal radiations emitted by cover and absorbed by canopy Qcca ,
R , convective heat transfer cludes: absorbed solar radiations by soil QsA , conductive heat
thermal radiations emitted by canopy Qca
C , latent heat transfer between the transfer into soil surface QsCd , thermal heat radiations emitted by
from canopy to inside air Qcai
L . soil surface QsR , thermal radiations emitted by greenhouse cover
canopy and the air inside greenhouse Qca R and convective heat transfer from the soil
and absorbed by soil Qcs
Energy balance for canopy in the greenhouse is given below: C .
surface to the inside air Qsi
A R R C L
Qca þ Qcca  Qca  Qcai  Qca ¼ 0 (7)
QsA  Qsi
C
 QsCd  QsR þ Qcs
R
¼ 0 (13)
The convective heat transfer between canopy and the air inside
The convective heat transfer from the soil surface to air inside
the greenhouse is:
the greenhouse is:
C A R R L
Qcai ¼ Qca þ Qcca  Qca  Qca (8) C
Qsi ¼ QsA  QsCd  QsR þ Qcs
R
(14)
C
Qcai ¼ sc IT aca Aca þ εc sTc4 Ac at;ca  εca sTca
4
Aca
  ls
*
C
Qsi ¼ IT sc as As  As ðTs  Tz Þ  εs sTs4 As þ εc sTc4 Ac as (15)
da Cpa :LAI P ðTca Þ  PðTi Þ z
 Aca (9)
g ra þ rs
3.4. Energy balance of inside air of greenhouse
The saturated water vapor pressure P * ðTca Þ at the temperature
of the canopy is calculated by Tetens equation Eq. (10), the
In the energy balance inside air of greenhouse, only the
following empirical formula relating accurately saturated water
convective heat transfer is considered. It is assumed that no solar
vapor pressure according to temperature ranging between 0 and
radiation is absorbed by the air. The balance includes: convective
60
C [30,31]: C ,
heat transfer from inside air to the cover of the greenhouse Qic
  (Eq. (4)), convective heat transfer from inside air to canopy surface
17:27Tca
P * ðTca Þ ¼ 0:6108 exp (10) C , (Eq. (9)), convective heat transfer from inside air to soil sur-
Qica
Tca þ 237:3 C , (Eq. (15)), heat losses due to ventilation or infiltration Q C
face Qis inf
and heat input into greenhouse QH.
The sensible heat balance of the air is described by the following
equation:

C C C C
Qci þ Qcai þ Qsi  Qinf þ QH ¼ 0 (16)

The heat transfer losses due to ventilation and infiltration is:

C
Qinf ¼ da Cpa REðTi  To Þ (17)

QH is the heat gain from the solar air heater with latent storage
collector into the IGLHS and QH ¼ 0 into the IG.

4. Results and discussions

In the IG and IGLHS, tomato plants were transplanted on the

23rd of January and the harvesting started on the 15th of April
2013. During this period, an experimental study has been carried
out for the growing period. Measurements are taken on the
selected days from 23rd to 27th of February 2013 in winter
Fig. 4. Reflectivity spectra of canopy and cover. season.
668 S. Bouadila et al. / Energy 64 (2014) 663e672

Fig. 7. Ambient humidity and wind velocity as a function of days (23rde27th February,
Fig. 5. Transmissivity spectra of canopy and cover. 2013).

4.1. Climatic conditions

attained 40
C and the relative humidity in these days is ranging
between 35 and 40%.
The external climate conditions for the 5-days period are shown
The 24th of February 2013 is characterized by several solar ra-
in Fig. 6 and Fig. 7. The maximum of the global solar radiation in-
diation fluctuations with maximum ambient temperature about
tensity in the horizontal plane was ranging between 300 and
11
C and wind velocity varied from 4 to 8 m/s. We noted that the
650 W m2, and outdoor temperature varied from 4 to 22
C
temperature inside the IG in this day attained 22
C and the diurnal
(Fig. 6). The wind velocity in the region of the greenhouses and the
relative humidity, is around 70%. At night, the temperature into the
relative humidity of ambient air as a function of days are presented
IG attains the ambient temperature value and the relative humidity
in Fig. 7. The first two days of this period are very windy, where air
inside the IG exceeded 90%.
speed varied between 3 and 8 m s1. The average speed of other
In this 5-days period it is seen that the difference between the
days is 2 m s1 and the mean value of relative ambient humidity is
greenhouse air temperature and the ambient air temperature
around 70%.
ranged between 10 and 20
C end is higher at 13 h (local time)
(Fig. 8).
4.2. Insulated greenhouse climate analysis
4.3. Performance of the SAHLSC inside IGLHS
The design of the IG maximizes solar radiation input while
reducing heat losses from the insulated northern wall covered by
The major environmental factors that affect the growth and the
sandwich panels.
precocity of production of greenhouse plants are carbon dioxide,
The daily variation of the greenhouse environmental tempera-
light, humidity and temperature. The tomato plants have two op-
ture and the relative humidity are presented in Fig. 8. The air
timum temperatures, one during the day, that varied between 20
temperature inside the greenhouses followed the same trend as the
and 30
C, and the second is the nocturnal temperature that is the
ambient temperature (Fig. 6). The 23rd and 27th of February 2013 is
most crucial one, it varied between 14 and 18
C. Fig. 8 gives the
characterized by clear sky with maximum ambient temperature
inside IG temperatures which exceed the daily tomato optimum
equal to 22
C. We found that the diurnal temperature into IG

Fig. 6. Global solar radiation and ambient temperature as a function of days (23rde Fig. 8. Humidity and air temperature inside IG as a function of days (23rde27th
27th February, 2013). February, 2013).
 S. Bouadila et al. / Energy 64 (2014) 663e672 669

Fig. 11. Inlet, outlet and PCM temperature of the SAHLSC as a function of days (23rde
Fig. 9. Air temperature inside IG and IGLHS as a function of days (26th and 27th 27th February, 2013).
February, 2013).

maintained to 10e17% lower than the IG relative humidity. The

temperature under severe winter weather. Consequently, it is SAHLSC create a passive dehumidification process at nighttime due
necessary to improve the air conditioning of this greenhouse. A to the increase of the air temperature inside the IGLHS.
new solar air heater collector with a latent heat storage system is Fig. 11 show the inlet, the outlet SAHLSC temperature and the
integrated inside the IGLHS. This collector is used to store the PCM temperature in the 5-days period. Under clear sky conditions
excess of the transmitted solar irradiation in the packed bed (23rd and 27th February, 2013) and during the charging period,
absorber and to provide it at night. The discharging process of the from 7 h to 17 h (local time), the temperature into the solar collector
SAHLSC starts at 21 h (local time). A fan is used to blow the air at a increases gradually and attained 54
C at 14 h. The PCM tempera-
fixed speed equal to 1 m/s. ture inside the capsule situated in the middle of the collector
The air temperature variation inside both IG and IGLHS are attained 39
C at 15:30 h (23rd February, 2013), the PCM is
presented in Fig. 9. On 27th of February 2013, the diurnal variation completely charged. In fact, from 21 h to 7 h, corresponds to the
of the IG inside temperature exceed the IGLHS’s inside temperature discharge process of the PCM (solidification). The variation be-
with a difference varied between 1 and 3.6
C from 10 h to 16:30 h tween the outlet and the inlet temperatures of the SAHLSC, is about
(local time). This is due to the fraction of solar radiations stored in 5e7
C and the PCM temperature is 10e14
C higher than the inlet
the packed bed absorber. From 21 h (local time), the analysis of the temperature. The outlet temperature remains approximately un-
curves illustrates a clear effect of heating of the solar air heater at changed around 20
C all the night. The SAHLSC can be considered
night. The SAHLSC maintain the air temperature inside the IGLHS as a heat source, the transferred heat from the collector to the
constant, around 12
C, along the night. greenhouse components (inside air, soil, plants, and cover) is by
Fig. 10 present the effect of the SAHLSC on relative humidity convection and radiation.
inside the greenhouse. The IGLHS air relative humidity is Fig. 12 show the temperatures variations of four sensors used for
measuring the horizontal temperature distribution at a height of 2,
1.5, 0.8 and 0.1 m from the ground. During the charging period on

Fig. 10. Humidity inside IG and IGLHS as a function of days (26th and 27th February,
2013). Fig. 12. Horizontal distributions temperatures inside IGLHS (27th February, 2013).
670 S. Bouadila et al. / Energy 64 (2014) 663e672

February 2013. During the best winter days, a significant heat

portion can be provided by the SAHLSC at the nighttime inside the
greenhouse. The hourly average heating recovered from SAHLSC all
the night is around 1.1 MJ and the hourly average heating demand
at the nighttime is around 4 MJ. The stored solar heat provided is
26% of the total greenhouse heating requirements. The hourly
heating excess of solar energy in the IGLHS (Fig. 13), from 11 h to
16 h (local time), varied between 5 and 13 MJ. An important portion
of solar heat transmitted to the greenhouse is stored by SAHLSC
which will be exploited later.
Fig. 14 presented the daily heating requirements, the daily
heating excess inside IGLHS and nighttime heating recovered from
SAHLSC, during 23rd - 27th of February 2013. The diurnal and
nocturnal greenhouse temperatures varied according to ambient
conditions.
The stored energy in the solar air heater with thermal storage is
13.7 MJ on 23rd of February, 2013. This value is equivalent to 56% of
the total daytime excess heat. For the SAHLSC system, the contri-
bution of stored energy to the nighttime heating requirements are
Fig. 13. Variation of hourly heating recovered from SAHLSC, heating excess inside
IGLHS and heating requirements on 27th February, 2013.
31%, 27% and 26%, respectively on 23rd, 26th and 27th of February
2013 under best ambient conditions.
On 24th and 25th of February 2013, the recovered stored heat
27th of February 2013, the inside temperature of the greenhouse are 6% and 11% with requirements heat exceeds 65 MJ.
presented a significant effect of stratification. The difference of
temperatures inside the IGLHS between each successive curves at 4.5. Comparison of the results
13 h ranged from 4 to 7
C. The SAHLSC created an important
convective heat transfer inside the greenhouse. The present work is based on thermal performance of green-
The 24th and 25th February, 2013 are characterized by severel house climate under Tunisia environment, comparison between
solar radiation fluctuations, windy and the ambient air temperature previous and current research are analyzed. Although this kind of
which didn’t exceed 11
C. We noted that the temperature into the research is new in Tunisia. These results are compared in Table 3
solar collector is ranging from 27
C to 37
C and that the PCM and summarized as follows. Studies about the application and
temperature is ranging from 27
C to 30
C (fusion temperature of utilization of PCM for space heating greenhouses dates back to
PCM) during the charging period (Fig. 11). In fact, during the dis- 1980, when Kern and Aldrich [34] used 1650 kg of CaCl2$6H2O in
charging process the outlet temperature and PCM temperature aerosol cans each weighing 0.74 kg was used to investigate energy
decrease progressively to be equal to the inlet temperature, the bed storage possibilities both inside and outside a 36 m2 ground area
is discharged after 5 h. greenhouse. Results of this study indicated that, whilst energy
stored by the outside unit was between 105.5 and 158.25 MJ with a
4.4. Energy performance assessment solar energy fraction of 38e43%, these were respectively 21.1 and
31.65 MJ for the internal unit with a somewhat much smaller solar
Eqs. (4),(9),(15)e(17) have been used for calculating re- fraction of 6e8%. Furthermore, it was calculated that while the
quirements, excess and recovered thermal energy inside the IGLHS. external unit released about 80e90% of the energy absorbed, this
Fig. 13 gives the hourly variation of heating requirements, heating was 60e80% for the internal energy storage unit. Paris [35] was
excess inside IGLHS and heating recovered from SAHLSC on 27th of used a eutectic mixture of sodium hydroxide (NaOH) solution and
chromium nitride (Cr2N melting point 15.6
C) in a 445 m2
experimental glasshouse. The material was stored in 21 containers.
The heat storage produced a minimum air temperature of 8
C
inside the greenhouse under extreme winter conditions and saved
about 5000 l of oil. Jaffrin and Cadier [36] described the perfor-
mances of a solar greenhouse equipped with a storage system
using 13.5 tons of CaCl2$6H2O phase change materials encapsu-
lated into flat plastic pouches; in a glass-covered multi-span
greenhouse with 500 m2 ground area. Thermal and crop perfor-
mance of this greenhouse was compared with an air-bubble
double plastic-covered and a conventional greenhouse. The con-
tainers were placed in concrete shelves with five layers in semi-
circular tunnels and air was used to transfer heat. This system
saved 80 and 60% of propane gas in comparison with the con-
ventional and double-covered greenhouses, respectively. Elec-
tricity consumption of the fans was determined to be less than 10%
of the heating load of the greenhouse. Huang et al. [20] has
designed and constructed a storage system with two different
stacking configurations and air baffling integrated with green-
house solar system. Cylindrical storage rods were used as the
Fig. 14. Nighttime heating recovered from SAHLSC, daytime heating excess inside primary storage elements. The result showed that the designed
IGLHS and heating requirements as a function of days (23rde27th February, 2013). latent storage systems demonstrated significantly higher compact
 S. Bouadila et al. / Energy 64 (2014) 663e672 671

Table 3
Summery of the performance of various agricultural greenhouses using phase change materials.

Research Greenhouse location Greenhouse PCM Mass of Remarks

area (m2) PCM (kg)

Kern and Aldrich e 36 CaCl2$6H2O 1650 The stored energy by the outside unit
(1979) [34] was between 105 and 158 MJ with a
solar energy fraction of 38e43%, and
respectively 21.1 and 31.65 MJ for the
internal unit with a solar fraction of 6e8%.
Paris (1981) [35] Antibes (43.34
N 07.06
E), 445 NaOH þ Cr2N 21 containers The heat storage produced a minimum air
France temperature of 8
C inside the greenhouse
under extreme winter conditions and
saved 5000 l of oil.
Jafrin and Cadier Nice (43.65
N 500 CaCl2$6H2O 13,500 The solar storage system achieved 80% per
(1982) [36] 07.20
E), France cent savings in propane gaz, compared with
the control compartment run at the
same temperature.
Huang Carolina (34.83
N 100 CaCl2$6H2O 1200 PCMs exhibited much larger energy storage
et al. (1986) [20] 7.30
W), USA capacity than rock storage.
Boulard Montfavet, France 176 CaCl2$6H2O þ Chlorides 2100 The stored energy during the daytime
et al. (1990) [21] and nitrates was 0.260 kWh/m2 energy and 0.360 kWh/m2
was released at night. 40% of heating load
was supplied and 30% in energy saving
was achieved.
Ozturk (2005) [18] Adana (37.01
N 35.18
E), 180 Paraffin wax 6000 The rate of heat transferred in the latent
Turkey storage unit ranged from 1.22 to 2.63 kW
and the rate of heat stored in the unit was
in the range of 0.65e2.1 kW.
Benli and Elazig, East Anatolia, Turkey 30 CaCl2.6H2O 300 PCM provided 18e23% of total daily thermal
Durmus (2009) [19] energy requirements of the greenhouse for 3e4 h,
in comparison with the conventional heating device.
Present work Borj Cédria, (36
430 N 10
250 E), 14.8 AC27 300 capsules The nighttime recovered heats attain 30% of the total
(2013) Tunisia heating requirements. The stored heat is equivalent
to 56% of the daytime total excess heat inside
the greenhouse.

storage capacity than the rock or water storage. Ozturk [18] pre- References
sented a seasonal thermal energy storage using 6000 kg of paraffin
wax as a PCM with the latent heat storage technique was [1] The National Observatory of Agriculture (ONAGRI) in Tunisia. http://www.
onagri.tn/STATISTIQUES/ENQUTES%20STRUCTURES/ESEA%202004-2005.
attempted to heat the greenhouse of 180 m2 floor area. During the htm#_Toc125361762.
experimental period, it was found that the average net energy and [2] Agency of Agricultural Investments Promotion. Les cultures maraichères sous
exergy efficiencies were 40.4% and 4.2%, respectively. Benli and abris. www.apia.com.tn/pdf/abri.pdf.
[3] Yang SH, Rhee JY. Utilization and performance evaluation of a surplus air heat
Durmus [19] conducted a study on the performance analysis of a pump system for greenhouse cooling and heating. Appl Energy 2013;105:
latent heat storage system. In their study, a 10 piece flat plate air 244e51.
solar collector was used to heat up a greenhouse located in Turkey. [4] Sethi VP, Sharma SK. Survey and evaluation of heating technologies for
worldwide agricultural greenhouse applications. Sol Energy 2008;82:832e59.
A corrugated plate, reverse corrugated plate, trapeze plate, reverse [5] Aye L, Fuller RJ, Canal A. Evaluation of a heat pump system for greenhouse
trapeze plate and a flat absorption plate in combination with a heating. Int J Therm Sci 2010;49:202e8.
phase change material were used for this work. They reported that [6] Tong Y, Kozai T, Nishioka N, Ohyama K. Greenhouse heating using heat pumps
with a high coefficient of performance (COP). Biosyst Eng 2010;106:405e11.
such system is capable to provide 18e23% of the total daily thermal
[7] Rousse DR, Martin DY, Thériault R, Léveillée F, Boily R. Heat recovery in
energy of a greenhouse for about 3e4 h in comparison to con- greenhouses: a practical solution. Appl Therm Eng 2000;20:687e706.
ventional heating devices. And the difference in temperatures [8] Bakos GC, Fidanidis D, Tsagas NF. Greenhouse heating using geothermal en-
between indoor and outdoor was around 6e9
C. ergy. Geothermics 1999;28:759e65.
[9] Adaro JA, Galimberti PD, Lema AI, Fasulo A, Jorge R, Barral JR. Geothermal
contribution to greenhouse heating. Appl Energy 1999;64:241e9.
5. Conclusion [10] Nayak S, Tiwari GN. Theoretical performance assessment of an integrated
photovoltaic and earth air heat exchanger greenhouse using energy and
exergy analysis methods. Energy Build 2009;41:888e96.
An experimental study was conducted to evaluate the nighttime [11] Hepbasli A. A comparative investigation of various greenhouse heating op-
recovered heat of the SAHLSC in the greenhouse. A new solar air tions using exergy analysis method. Appl Energy 2011;88:4411e23.
[12] Garcia JL, Plaza SSL, Navas LM, Benevente RM, Luna L. Evaluation of the
heater collector using a packed bed of spherical capsules with feasibility of alternative energy sources for greenhouse heating. J Agric Engng
latent heat system was operated and installed inside a greenhouse. Res 1998;69:107e14.
This collector stored solar energy during the daytime and supplied [13] Bargach MN, Tadili R, Dahman AS, Boukallouch M. Survey of thermal per-
formances of a solar system used for the heating of agricultural greenhouses
it for heating at night. As a result of this system, on 23rd of February
in Morocco. Renew Energy 2000;20:415e33.
2013, the amount of the nighttime recovered heat of this system [14] Willits DH, Chandra P, Peet MM. Modeling solar energy collection/storage
attained 30% of the total heating requirements. The stored heat is system for greenhouses. J Agric Engng Res 1985;32:73e9.
[15] Bouhdjar A, Belhamel M, Belkhiri FE, Boulbina A. Performance of sensible heat
equivalent to 56% of the daytime total excess heat inside the
storage in a rock bed used in a tunnel greenhouse. Renew Energy 1996;9:
greenhouse. Under best ambient conditions, the outlet temperature 724e8.
for the SAHLSC, is 5e7
C higher than the inlet temperature and [16] Lazaar M, Kooli S, Hazami M, Farhat A, Belghith A. Use of solar energy for the
remains constant at 20
C, all the night. The SAHLSC create a passive agricultural greenhouses autonomous conditioning. Desalination 2004;168:
169e75.
dehumidification process at night and allow a relative humidity [17] Vadiee A, Martin V. Thermal energy storage strategies for effective closed
from 10 to 17% lower than the IG greenhouse. greenhouse design. Appl Energy 2013;109:337e43.
672 S. Bouadila et al. / Energy 64 (2014) 663e672

[18] Öztürk HH. Experimental evaluation of energy and exergy efficiency of a [28] Kindelan M. Dynamic modeling of greenhouse environment. Trans ASAE
seasonal latent heat storage system for greenhouse heating. Energ Convers 1980:1232e9.
Manag 2005;46:1523e42. [29] Duffie JA, Beckman WA. Solar engineering of thermal processes. 2nd ed. New
[19] Benli H, Durmuş A. Performance analysis of a latent heat storage system with York: USA: John Wiley and Sons; 1991.
phase change material for new designed solar collectors in greenhouse [30] Tetens VO. Uber einige meteorologische Begriffe. Zeitschrift fur Geophysik
heating. Sol Energy 2009;83:2109e19. 1930;6:297e309.
[20] Huang BK, Toksoy M, Cengel YA. Transient response of latent heat storage in [31] Boulard T, Wang S. Greenhouse crop transpiration simulation from external
greenhouse solar system. Sol Energy 1986;37:279e92. climate conditions. Agr Forest Meteorol 2000;100:25e34.
[21] Boulard T, Razafinjohany E, Baille A, Jaffrin A, Fabre B. Performance of a [32] Boulard T, Baille A, Mermier M, Vilette F. Mesures et modélisation de la
greenhouse heating system with a phase change material. Agr For Meteorolo résistance stomatique foliaire et de la transpiration d’un couvert de tomate de
1990;52:303e18. serre. Agronomie 1991;11:259e74.
[22] http://www.campbellsci.com. [33] Boulard T, Mermier M, Fargues J, Smits N, Rougier M, Roy JC. Tomato leaf
[23] Manwell JF, McGowan JG, Rogers AL. Wind energy explained: theory, design boundary layer climate: implications for microbiological whitefly control in
and application. 2nd ed. New York: USA: John Wiley and Sons; 2009. greenhouses. Agr Forest Meteorol 2002;110:159e76.
[24] Bouadila S, Kooli S, Lazaar M, Skouri S, Farhat A. Performance of a new solar [34] Kern M, Aldrich RA. Phase change energy storage in a greenhouse solar
air heater with packed-bed latent storage energy for nocturnal use. Appl heating system. ASAE Paper No. 79e4028. St. Joseph. MI: Am Soc Agric Engng;
Energy 2013;110:267e75. 1979.
[25] Holman JP. Experimental methods for engineers. 6th ed. Singapore: McGraw- [35] Paris V. Eutectic mixture for greenhouse heating. In: Proceedings of second
Hill Book Co.; 1994. conference on solar greenhouses. Perpignan 1981. pp. 240e6.
[26] Papadakis G, Frangoudakis A, Kyritsis S. Mixed, forced and free convection [36] La Baronne Solar Greenhouse, Jaffrin A, Cadier P. Latent heat applied to hor-
heat transfer at the greenhouse cover. J Agric Engng Res 1992;51:191e205. ticulture. Sol Energy 1982;38(4):313e21.
[27] Mac Adams WH. Heat transmission. 3rd ed. Mac Graw-Ifill; 1954.