gas thermochemical storage process for solar air-conditioning

Energy 41 (2012) 261e270

Contents lists available at ScienceDirect

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

Experimental investigation of a solid/gas thermochemical storage process for solar air-conditioning Driss Stitou a, *, Nathalie Mazet a, Sylvain Mauran b a b

PROMES Laboratory (PROcess, Material and Solar Energy), CNRS-UPR8521, Rambla de la Thermodynamique, Tecnosud, 66100 Perpignan, France University of Perpignan UPVD, 52 avenue Paul Alduy, 66860 Perpignan, France

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 September 2010 Received in revised form 16 June 2011 Accepted 16 July 2011 Available online 3 September 2011

This paper focuses on the experimental performances of a solar air-conditioning pilot plant for housing, which is running in PROMES laboratory (Perpignan-Odeillo, France) since 2006. This pilot of daily cooling capacity of 20 kWh consists of a solid/gas thermochemical sorption process which is powered at 60 e70
C by 20 m2 of flat plate solar collectors. The thermochemical sorption process is based on the coupling of a liquid/gas phase change of a refrigerant (NH3) and a reversible chemical reaction between a reactive solid (BaCl2) and this refrigerant. Its functioning mode is intrinsically discontinuous and cyclic. It is relevant for the storage or transformation of solar energy. An analysis of 2-years experimental working of the prototype leads to an averaged yearly efficiency of solar collectors and a process COP ranging respectively from 40 to 50% and 30e40%. This prototype enables thus a daily cooling productivity at 4
C of about 0.8e1.2 kWh of cold per m2 of flate plate solar collector and leads to a global solar COP ranging from 15 to 23%. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Solar air-conditioning Solid/gas sorption Thermochemical reactor

1. Introduction Nowadays, peaks in electricity demand occur more frequently during the summer period in most developed countries, because of the increasing use of air-conditioning. The reasons lie in higher thermal comfort expectations, in lower initial costs for airconditioning equipments and in the heat island effect in urban areas, which leads to microclimatic changes. The International Energy Agency has gathered frightening data on energy consumption trends. During the last four decades (1973e2008) primary energy consumption has grown by 80% and CO2 emissions by 88%, with an average annual increase of 2.5% [1]. Among building energy services, HVAC systems are the most energy consuming devices, accounting for about 10e20% of final energy use in developed countries. Moreover, final energy consumption in buildings in the European Union represents about 40% of the total energy consumption. In the EU residential sector, about 70% of the total final energy consumption is used for space conditioning, 15% for domestic hot water and 15% for electricity [2]. This energy consumption means that the building sector is responsible for about 20% of the total CO2 emissions [3].

* Corresponding author. Tel.: þ33 468682233; fax: þ33 468682213. E-mail address: [email protected] (D. Stitou). 0360-5442/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.energy.2011.07.029

Solar assisted cooling appears to be a promising alternative to the conventional electrical driven air-conditioning from an environmental point of view, since it results in lower CO2 emissions and avoids CFCs and HCFCs uses. Considering the problem of electricity peaks during summer caused by electrically driven air-conditioning systems, and the close coincidence with the maximum solar irradiation, solar assisted refrigeration may be an interesting option to handle successfully the issue of reducing electricity demand peak due to air-conditioning. Thus, the use of solar thermal energy for cooling has the advantage of synchronization between solar irradiation and air-conditioning demand. Sorption refrigeration technologies such as thermochemical reaction, adsorption, absorption and desiccant cooling processes, are the prevailing options for the utilization of solar thermal energy in air-conditioning. These systems are based on the thermal effects of reversible physico-chemical processes, such as the liquid/gas absorption or solid/gas chemical reactions or solid/gas adsorption. Sorption refrigeration uses physical or chemical attraction between a pair of substances to produce a refrigeration effect. It uses a sorbent material that has an ability to attract and absorb an active gas. The sorbent can either be in liquid phase (in liquid/gas absorption technology) or in solid form (in physical adsorption and thermochemical sorption technologies). The commercial application of solar energy for air-conditioning systems is fairly recent. A comprehensive review of these cooling technologies is given in several references [4e8].

262

D. Stitou et al. / Energy 41 (2012) 261e270

Nomenclature a COP E K G L p P PCM Q R S SCP T DH DS DG

n h

activity coefficient (dimensionless) coefficient of performance (dimensionless) energy (Wh) equilibrium constant of the reaction gas liquid pressure (Pa) power (W) Phase change material thermal energy (J) perfect gas constant (8.314) (J mol1 K1) solid salt compound specific cooling power (W/m3) temperature (K) enthalpy variation (J mol1) entropy variation (J mol1 K1) Gibbs free energy (J mol1) stoechiometric coefficient (dimensionless) efficiency (dimensionless)

Subscripts or superscripts amb_day diurnal outdoor environment amb_night nocturnal outdoor environment i reaction, evaporation or condensation v vaporisation

The two main components of a solar cooling plant are the solar collectors and the sorption cooling process. The overall system efficiency depends on the coupling between these two components. The research in this direction attempts to match optimally solar thermal technologies with thermally driven cooling equipments [9]. Solar coolers and refrigerators have been developed for more than 40 years using liquid absorption [10,11], solid adsorption [12e15] or thermochemical reaction processes [16e22]. Most of solar cooling systems that are used are of liquid/gas absorption type, while solid/gas sorption processes are still at the development stage. The currently marketed absorption machines generally use a saline solution of LiBreH2O and are very widespread in airconditioning market. Solid/gas sorption processes have distinctive advantages compared to absorption technology. As the sorbent material is solid, the main difference is that two or more reactors or adsorbers are necessary in order to provide continuous cooling production. Solid/gas sorption systems allow for somewhat lower driving temperatures but have a lower COP and (SCP) specific cooling power compared to absorption systems under the same conditions. Nevertheless, the use of solid/gas sorption cooling technology seems to be preferable for small cooling capacity chillers [23,24]. Heat-driven cooling processes seem to have an excellent potential in the air-conditioning market, albeit that so far they are not really a competitive alternative to the conventional vapour compression chillers. The problems for all available solar sorption systems for air-conditioning are mainly the high equipment and maintenance cost, the big size and also the need of an auxiliary energy system. A lot of efforts have been taken regarding solar sorption cooling for but there is still a development demand to optimize and reduce the costs of these systems and to carry out experimental investigations on such processes [5]. The present study focuses on experimental investigations carried out on a solar powered solid/gas reaction-based

r reaction o reference or standard value cond condenser evap evaporator dec decomposition reaction syn synthesis reaction coll collected, collectors reac, react reactor or reaction sol solar cold cold reac-dec reactor in decomposition phase reac-syn reactor in synthesis phase hotPCM, HotStor relative to the hot PCM storage coll_out at outlet of the solar collector react_in at inlet of the reactor react_out at outlet of the reactor gaseous ammonia NH3G liquid ammonia NH3L cond_in at inlet of the condenser cond_out at outlet of the condenser ground_in at inlet of the ground cooling loop ground_out at outlet of the ground cooling loop dist chilled water distribution cold_PCM_top at the top inlet/outlet of the cold PCM storage cold_PCM_bot at the bottoming inlet/outlet of the cold PCM storage

thermochemical pilot plant using flat plate collectors for residential air-conditioning. The purpose of this work is to quantify and analyze over two years of operation the energy performances of this thermochemical plant representative for air-conditioning in housing in order to identify design and control improvements. In addition, the authors wish to emphasize here that the prototype of thermochemical process presented in this paper, is a completely new prototype and in our knowledge, no other similar prototype for solar air-conditioning applications has been already experimented.

2. Principle of solid-gas thermochemical sorption process The principle of thermochemical reaction processes is based on the thermal effect of a reversible reaction between a solid and a reactive gas

S1 þ v:G % S2 þ v:ðDHr Þ By coupling such a solid-gas reaction with a liquid/gas phase change of the same working gas, a sorption cooling production process can be designed

GðliqÞ þ DHv %GðgasÞ The two processes (i.e. phase change or chemical reaction) are monovariant and their equilibrium conditions follow the wellknown ClausiuseClapeyron relation. The ClausiuseClapeyron relation is obtained by stating that for the considered transformation, the free Gibbs energy is equal to zero at the thermodynamic equilibrium:

DG ¼ DG
þ RTlnK ¼ DH
 T:DS
þ RTlnK ¼ 0

(1)

where K the equilibrium constant that corresponds for the considered solid/gas reaction to:

D. Stitou et al. / Energy 41 (2012) 261e270

K ¼

aðS1 Þ$pn aðS2 Þ

(2)

Then the ClausiuseClapeyron relation can then be derived by assuming the activities of solid are equal to 1 and the reactive gas behaves as a perfect gas. Then, the thermodynamic equilibrium conditions for each process are determined by only one parameter (p or T) :

 ln

p po



DHio

¼ 

R:T

þ

DSoi R

(3)

where DHoi is the standard enthalpy of the transformation process, either solid-gas reaction DHr or liquidevapour phase change DHv. Similarly DSoi is the corresponding standard transformation entropy and po is a reference pressure of 1 Pa. The implementation of a chemical solid/gas sorption unit implies then the linking of two elements: a fixed-bed reactor that allows the reversible chemical reaction to take place and two-phase heat exchangers in which evaporation or condensation of the reactive gas occurs. The simplest sorption machine consists of a solid/gas reactor coupled with an evaporator, which can also play the role of a condenser [16]. Fig. 1 describes such an implementation in its simplest form. Following the right direction of the reaction, the solid S1 reacts with the gas G produced by the evaporator and forms the solid S2 (synthesis of salt S2). This synthesis reaction is exothermic and produces the heat of reaction DHr that is released to the environment at the ambient temperature. In the left direction, the decomposition reaction, or thermal dissociation, is endothermic and requires a heat input DHr to proceed: salt S2 is then dissociated into the initial salt S1 and the gas G. The gas G flows into the condenser which is coldest part of the system, and condenses into liquid.

263

A large number of solid-gas working pairs have already been used in refrigeration applications [25]. Among them, some working pairs are suitable for solar cooling applications such as calcium chloride/ammonia [26] or calcium chloride/monomethylamine [27], strontium chloride/ammonia [28]. These ammoniated salts have relatively low dissociation temperatures, below 100
C, which make them usable for solar cooling applications using conventional flate plate solar collector. The reactive salt chosen for the pilot plant described hereafter is the barium chloride, BaCl2, which reacts with 8 mol of ammonia to form BaCl2.8NH3 [29]:

BaCl2 þ 8 NH3 % BaCl2 :8NH3 þ 8 DHr

(4)

This salt has been successfully used for ice-making and airconditioning processes [19,30e32]. Its main advantages lie in the low decomposition temperature of the salt, which is ranging from 50 to 70
C for the high-pressure phase depending on the condensing pressure (Fig. 1) defined by the relation (3): for example a condensation temperature of 20
C will impose pressure of 8.5 bar, which in return will require a minimum decomposition temperature of 48
C. A condensing temperature of 50
C will require that the decomposition reaction occurs at a minimum temperature of 70
C. In addition, this salt presents a high stoechiometry coefficient n that minimizes the size of the reactor [33]. The vapour pressureetemperature relation for the solid-gas reaction equilibrium and the liquidevapour ammonia equilibrium curve are plotted from the ClausiuseClapeyron equations (3). Fig. 2 shows experimental data related to this ammoniated salt from various authors [34e38]. Enthalpy and entropy variations of the chemical reaction have been obtained by averaging these values. These averaged values are valid up to a pressure of 30 bar, which corresponds to a dissociation temperature of 80
C: - for the reaction equilibrium: the heat of reaction DHr ¼ 38250 J mol1 of gas and the entropy of reaction DSr ¼ 232.4 J mol1 K1, lead to :

4600:4 ln ðpÞ ¼  þ 27:95 T

(5)

20

16 14 Pressure (bar)

NH3

Boumaraf 1989 Moutaabbid 1986 Gillespie 1931 Gillespie and Lurie (1931) Averaged

18

BaCl2, 0/8 NH3

12 10 8 6 4 2 0 -30

-20

-10

0

10

20

30

40

50

60

70

80

Temperature (°C)

Fig. 1. Schematic operating phases of a solid-gas sorption system.

Fig. 2. Equilibrium P-T curve for the considered solid/gas reaction obtained from various experimental data.

264

D. Stitou et al. / Energy 41 (2012) 261e270

- for the ammonia phase change equilibrium: DHv ¼ 23366 J mol1 and DSv ¼ 193.3 J mol1 K1, thus:

2810:3 ln ðpÞ ¼  þ 23:25 T

(6)

The consumption of the reactive medium that constitutes a porous medium gives rise to general problems inherent in transport phenomena (heat and mass transfer). The heat of reaction from the synthesis phase (or consumed during the decomposition phase) has to be removed (or delivered) by a heat transfer fluid. Equally, the diffusion of the reactive gas, through the reacting porous medium must not limit the progress of the chemical reaction. This kind of limitation can notably happen in synthesis reactions during which an increase of the volume of the reactive medium is observed, thereby diminishing the gas permeability of the reacting medium. Thus, in order to enhance heat and mass transfer in the porous medium, it is necessary to find an inert binder in relation to the reactants. For these thermochemical processes, the solid reactant is mixed with (ENG) expanded natural graphite, and recompressed together in order to form consolidated reactive blocs. Graphite is an inert porous binder that has a high intrinsic thermal conductivity, an exfoliated structure enabling a high porosity and mechanical elasticity. AS a result, the ENG enhances efficiently heat and mass transfers inside the consolidated reactive blocs without taking part to the reaction. It reduces the effect of volumetric variations of the salt occurring with the chemical reaction [16,40,41]. This innovative implementation was first developed and patented by PROMES laboratory in 1983 to avoid power limitation of thermochemical reactors [39]. Expanded graphite is now widely used as a porous additive in composite reactive medium due to its high thermal conductivity and gas permeability [42].

3. Pilot plant description 3.1. Operating mode This solar sorption system operates discontinuously according two different phases: the diurnal period during which the system is regenerated, and the nocturnal period where the cooling occurs. Fig. 3 presents the pressure vs. temperature operating conditions of the two main steps of this basic sorption cycle at high and low pressures.

During the day time, the reactor is heated by solar energy and desorbs the gas that flows to the condenser and condenses at the diurnal outdoor temperature Tamb_day, which is greater than the nocturnal outdoor temperature Tamb_night. The high operating pressure of this step is imposed by the condensation temperature of the reactive gas. During the night time, as the reactor is cooled down, it reabsorbs the ammonia gas and induces a pressure decrease in the reactor linked to the evaporator. This enables the boiling of the liquid ammonia in the evaporator and produces the cooling effect. The ammonia gas produced in the evaporator is then absorbed by the reactor, which releases the heat of absorption at the nocturnal ambient temperature. As the cooling effect is produced during the night, a cold storage is required for the subsequent utilization of cold. 3.2. Components of the pilot plant The pilot plant was designed to provide a daily cooling capacity of about 20 kWh. By considering a specific refreshing demand of 50 W per m2 of floor, the cooling energy produced by the thermochemical unit corresponds to air-conditioning needs of a conference room of 100m2 during approximately 4 . Figs. 4 and 5 present some views of the pilot plant and a schematic diagram of the installation. The pilot plant consists of four subsystems (see Fig. 4): -

NH

-

BaCl , 0/8

22

-

Daytime : Process regeneration

20

solar heating loop, thermochemical sorption unit, ground cooling loop, chilled water production loop. distribution loop of chilled water

The solar heating loop is composed of 21.6 m2 of flat plate collectors manufactured by HelioAkmi (ST2000 model). These collectors are characterized according to the European standard certification EN-12975 by an optical efficiency ho of 0.72, and heat loss coefficients of first order a1 of 4.804 W/m2.K and second order a2 of 0.04 W/m2.K2. The collectors are tilted at 30
and south oriented. The solar collector loop is connected during the day to the thermochemical reactor and a hot thermal storage, which is filled with 360 kg of a phase change material (wax RT80 from Rubitherm GmbH) that melts at 75e80
C. This hot PCM storage enables:

-

24

a a a a a

the storage of the excess solar heat if the decomposition reaction taking place in the reactor is completed before the end of the afternoon, the partial storage of the sensible heat released by the reactor cooling after the end of the decomposition phase, the heating of the reactor at the beginning of the day or when the solar heat is not sufficient.

16 14 12

NH

(G)

Tcoll

Tamb_day

Pressure (bar)

18

Cond

Dec

10

6

Evap

NH

(G)

Tcold

4

Tamb_night

8

2

Synt

Nightime : Cold Production

0 -5

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

Temperature (°C)

Fig. 3. Typical thermodynamic operating conditions of the solar solid/gas sorption process.

The thermochemical sorption unit consists of one reactor made of a set of 19 tubes filled with a composite reactive porous medium (Fig. 5). The reactive composite consists of a compressed mixture of 140 kg of anhydrous BaCl2 and 35 kg of expanded natural graphite, which is an inert binder that enhances heat and mass transfer. The reactor is either connected to a condenser during the day or to the evaporator during the night through the valves V8 and V7, respectively. The condensed ammonia is stored in a 100 tank in order to be used during the night. A level gauge is located in this tank to measure the amount of ammonia released or absorbed by the reactor. The ground cooling loop enables alternatively the cooling of the condenser during the day and the cooling of the reactor during the

D. Stitou et al. / Energy 41 (2012) 261e270

265

Fig. 4. View of the solar sorption pilot plant for air-conditioning, flate plate solar collectors and the thermochemical reactor design.

night thanks to a set of two 3-way valves (V4 and V5). This cooling loop is made of a horizontal in-ground plate heat exchanger of 16m2, which is buried at 2 m depth. The chilled water loop is connected to the evaporator. The chilled water is produced at about 0e5
C during the night phase. This cooling capacity is then stored into a second PCM storage, which is filled with 340 kg of a wax that solidify at 5
C (wax RT5 from Rubitherm GmbH). During the day, fresh water at 13
C is distributed from the PCM storage via a mixing valve V6 to the fan-coil units located in a conference room. 3.3. Operating and control of the solar process The entire pilot plant is instrumented and its operation is controlled under Labview software. the instrumentation consists of  Temperature measurements by thermocouples (T-type, accuracy class 1, 0.5
C) located at the inlet and outlet of each main components, and 100U platinum RTD (4-wire, accuracy class A,0.2
C) for remote components such as collectors,  Pressure measurements of the reactor, evaporator and condenser by piezoresistive pressure transducers (Keller manufacturer, 0/25 bar, accuracy class 0.5, 0.125 bar) with a signal conditioning in 4e20 mA,  Mass flow rate measurements of the heat transfer fluids (solar collector fluid, outlet reactor, ground loop, chilled water at outlet of the cold storage) by turbine-type flow meters (Kobold manufacturer, 0e20 /mn, accuracy class 1, 0.2 /mn)

 Level measurement of liquid ammonia in the tank by a capacitive probe (Kobold manufacturer, accuracy  1 mm, height probe 400 mm).  Solar irradiance by pyranometer (CMP11 model from Kipp & Zonen manufacturer, spectral range 310e2800 nm, accuracy class 1, 10 W/m2, signal conditioning 4e20 mA). The solar heating loop is controlled by the pyranometer and temperature measurements of the heat transfer fluid of solar collectors, inside the reactor and hot PCM storage. By comparing these temperatures, different control strategies are possible: the reactor can be heated either by the collectors or by the hot PCM storage in the morning, the collector field can supply heat only to the hot storage or both the reactor and the hot storage, the reactor can also heat up the hot storage tank,.All these strategies are implemented into (SCADA) supervisory control and data acquisition software developed under Labview software. This SCADA system monitors and controls the whole solar process by managing the different operating phases (normal operation and security issues). The connection of the reactor with either the condenser or the evaporator is controlled by pressure sensors (in the reactor, evaporator and condenser) and the level gauge in the ammonia tank. When the high level in the liquid ammonia tank is reached, the decomposition phase stops, the reactor is disconnected from the condenser and is cooled down by the ground cooling loop. During the synthesis phase, as the salt in the reactor reabsorbs the ammonia coming from the evaporator, the liquid level in the tank decreases. When the ammonia level reaches the low level

266

D. Stitou et al. / Energy 41 (2012) 261e270

Fig. 5. Schematic description of the solar air-conditioning thermochemical pilot plant.

threshold, the synthesis phase is stopped; the reactor is then disconnected from the evaporator and begins to be heated by the hot PCM storage. The ground cooling loop feeds separately either the reactor or the condenser. The cooling of the condenser starts when the reactor is connected to the condenser during the dissociation reaction, while the cooling of the reactor starts when the pyranometer indicates a low irradiation value at the end of the afternoon, or when the liquid ammonia reaches the high level in the tank. The chilled water loop is actuated when the reactor is connected the evaporator and if the evaporating temperature is lower than the cold storage temperature. The distribution of the fresh water from the cold PCM storage to the fan-coil units is activated on user’s demand during the day.

Some security procedures are foreseen to prevent any overpressure in the reactor during hot summer days. In this case, the solar heating loop is stopped and/or the connection between the reactor and the condenser is constrained in order to decrease rapidly the reactor pressure. 4. Experimental results A typical cycling of the pilot plant over several days in June is shown in Figs. 6 and 7. Fig. 6 shows the evolutions of the outlet temperature of solar collectors, inlet and outlet temperatures of the reactor, hot storage, and ground cooling loop, and the ammonia temperature at the inlet and outlet of the condenser. Temperatures of the chilled water (cold storage e evaporator loop) and

D. Stitou et al. / Energy 41 (2012) 261e270

267

85 80

Tcoll_out

75

Treac_in T_hotPCM

70

Treac_out

Temperature (°C)

65 60 T_NH3G_Cond_in

55 50 45 40

T_ground_in

T_NH3L_Cond_out

35 30 25

T_ground_out

20 2/6 18h

2/6 21h

3/6 0h

3/6 3h

3/6 6h

3/6 9h

3/6 12h

3/6 15h

3/6 18h

3/6 21h

4/6 0h

4/6 3h

4/6 6h

4/6 9h

4/6 12h

Daytime

Fig. 8. Evolutions of ammonia level in the reservoir and pressures of reactor, condenser and evaporator.

Fig. 6. Temperature evolutions of the main components of the solar loop.

distributed chilled water are shown in Fig. 7. During these days, the distribution loop was in operation from 10A.M. to 5P.M. Fig. 8 illustrates the typical pressure swings of the reactor, condenser and evaporator over two days. The evolution of the amount of condensed or evaporated ammonia is also shown. The solar collectors deliver hot water at a maximum temperature of 75e80
C and enable the heating of the reactor, which reaches a maximum temperature of 68
C. At the beginning of the day, the reactor at low temperature is quickly heated up by the hot PCM storage: its temperature increases from 25
C to 45
C, then it is heated up by the solar collectors (Fig. 6). As the reactor is closed, its pressure increases (Fig. 8). The reactor is then connected to the condenser when its pressure becomes greater than the condenser pressure. The gaseous ammonia is desorbed from the reactor, then condensed and stored in the tank: the level of liquid ammonia increases (Fig. 8). At the end of the day, when solar irradiance becomes insufficient, the reactor is closed (disconnected from the condenser) and is cooled down by the ground loop. The reactor pressure decreases as the salt reabsorbs ammonia. When the reactor pressure becomes lower than the evaporator pressure then a connection reactorevaporator is established (Fig. 8). During the night, the reactor is cooled down below 40
C, it absorbs the ammonia gas produced by the evaporator and enables a cooling production at a temperature of about 0
C (Fig. 7). The

liquid ammonia level in the evaporator is controlled and maintained more or less constant by a second level gauge that acts on the valve V9 located between the evaporator and the liquid ammonia tank. The valve V9 is opened when a low level set point is reached and liquid ammonia flows from the ammonia tank to the evaporator due to the pressure difference. When the high level is reached, the valve V9 is closed. This level control induces pressure peaks in the evaporator and a discontinuous decrease of the ammonia level in the reservoir (Fig. 8). Chilled water is then produced at a temperature below 3
C, which is enough to solidify the PCM in the cold storage. During the day after, the evaporator is closed and the cold PCM storage is discharged. The temperature of the evaporator increases during all the day as it is disconnected from the rest. Chilled water at 5
C is mixed with returned fresh water by a mixing valve in order to provide water at 12e13
C to fan-coil units of the conference room. Fig. 9 depicts the evolutions of the thermal powers for different components of the process over three consecutive days. The solar collectors supply a maximum power Pcoll of 10 kW under a global solar irradiation Psol of 18 kW. The decomposition reaction (Preacdec) in the reactor absorbs approximately 70% of this input power and the remaining part is stored in the hot PCM (Pstor). The efficiency of the solar collector is about 55% for these sunny days.

30

25

Temperature (°C)

20

15

10 T_dist

5 T_Evap

0

Tcold_PCM_top Tcold_PCM_bot

-5 2/6 18h

3/6 0h

3/6 6h

3/6 12h

3/6 18h

4/6 0h

4/6 6h

4/6 12h

4/6 18h

5/6 0h

5/6 6h

5/6 12h

5/6 18h

6/6 0h

6/6 6h

6/6 12h

Daytime

Fig. 7. Temperature evolutions of others components (cold part).

Fig. 9. Evolution of the thermal powers involved in the thermochemical solar process.

268

D. Stitou et al. / Energy 41 (2012) 261e270 50

Ecoll→stor

Ereac_cooling

Estor→ reac Ereac→ stor

Process Coefficient of Performance (%)

Ecoll→reac

REACTOR

Ecoll

E reac_heating

Ecoll_loss

Hot PCM Storage

Esol

COLLECTOR

Ereac_loss

45

summer 2008

40 35 30 25 20

summer 2007

15 10 5 0 0

Estor_loss

500

1500

2000

2500

3000

3500

4000

Daily heat output of the collectors (Wh/m²)

Fig. 10. Schematic diagram of the thermal flux involved in the solar loop.

Fig. 12. Evolution of the daily process COP.

During the night, the evaporator enables a cooling power of about 2 kW during 10 . The heat of synthesis reaction is released into the ground with a power about 3 kW. The diagram of Fig. 10 describes the energetic flux involved between the main components of the solar heating loop. According to the operating configurations, a part of the daily collected solar energy Ecoll is stored into the hot storage (Ecoll/stor) and the other part is used to heat to the reactor (Ecoll/reac). During the morning or cloudy passages, the hot storage provides heat to the reactor (Estor/reac). At the end of the day when the reactor is cooled down, a fraction of the heat removed is recovered and stored in the hot storage (Ereac/stor). Some thermal losses of the storage Estor_loss and the reactor Ereac_loss occur also during all the day and night. The pilot plant is running since 2006. Figs. 11e13 present the performances of the thermochemical process over 2007 and 2008 summers. Each point represents a daily performance. The daily solar collector efficiency is defined as the ratio of the total thermal energy collected Ecoll that is transmitted to the process via the heat transfer fluid, to the total incident solar irradiance Esol:

hcoll ¼

1000

Ecoll Esol

(7)

Comparing the two experimental seasons in Fig. 11, a decrease in the collector efficiency is observed. During summer 2007, the daily collector efficiency is about 50e60% while in summer 2008 the solar collector efficiency is lower, ranging from 40 to 50% due to a bad aging of the solar collectors. Fig. 12 depicts the evolution of the coefficient of performance of the process COPprocess, which is defined as the ratio of the daily

cooling energy Ecold produced by thermochemical unit to the total heat solar collected during the day Ecoll :

COPprocess ¼

Ereact heating Ecold Ecold ¼ $ Ecoll Ereac heating Ecoll

(8)

This definition of the process COP takes into account the efficiency of the heat storage and that of the thermochemical unit. The daily thermal energy Ereac_heating that is supplied to the reactor for its regeneration is equal to the part of the collected solar energy Ecoll/reac that feeds directly the reactor plus the thermal energy EStor/reac that is discharged from the hot PCM storage to the reactor at the beginning of the day or when the collected solar energy is not sufficient. Fig. 13 shows the evolution of the daily heat supplied to the reactor as a function of the daily collected solar energy. Overall, for a same amount of collected energy, the amount of heat that is supplied to the reactor is greater for 2008 summer. Even if the daily heat collected during summer 2007 is mostly greater than in 2008 (Fig. 11), the daily heating energy of the reactor is somewhat lower during summer 2007. Therefore, daily COPprocess were generally lower, ranging from 15 to 40% for summer 2007. This evolution in performance of this solar process is the result of a better control and management of the solar heating loop and of the different components of the sorption pilot plant that have been implemented for the summer 2008. This control mainly consisted

70 summer 2007 Solar collector efficiency (%)

60 50 40 summer 2008

30 20 10 0 0

1000

2000

3000

4000

5000

6000

7000

Daily irradiation (Wh/m²)

Fig. 11. Evolution of the daily solar collector efficiency over the summer periods of 2007 and 2008. Each point corresponds to a daily experimental value.

Fig. 13. Evolution of the daily heat amount supplied to the reactor during the diurnal phase of regeneration.

D. Stitou et al. / Energy 41 (2012) 261e270

in the replacement of a defective valve V2 of the hot PCM storage, improvement of the insulations of the reactor and hot and cold PCM storage (insulation increased from 5 to 10 cm thickness), improvement of the Labview control strategies such as the increase of the threshold and hysteresis values of irradiation required for starting/stopping the solar loop. In the end, the total thermal energy that is collected is efficiently used by the reactor, and enables a higher cooling production. These improvements have counter-balanced the decrease in the solar collector efficiency. The daily process COP for summer 2008 is improved and ranges roughly from 25 to 45%. Finally, the daily solar coefficient of performance of the whole solar process COPsol, which is defined below varies from 10 to 23%

COPsol ¼

Ecold Esol

(9)

The daily cooling productivity of such thermochemical process ranges from about 800 to 1200 Wh of cold produced at 4
C per m2 of solar collector 4.1. Conclusions Solar cooling can become a promising solution for the clean and sustainable air-conditioning of buildings. As the demand for airconditioning is increasing and, considering the fact that soft technologies like passive cooling are very difficult to apply in existing buildings, the sorption cooling technology discussed in this paper may provide a competitive alternative to conventional airconditioning systems. Solar refrigeration is a technology that has a great variety of methods of producing low temperatures; however, very few have demonstrated a technical and economical viability. The experimental study presented in this paper shows that is possible to develop thermochemical solid-gas sorption process for air-conditioning using only standard flat plate solar collectors operating at 70
C. Our results shows that an optimal management and control of the different components of the sorption pilot plant, such as the coupling of the solar collector loop with the hot PCM storage and the reactor or the connection of the reactor with the condenser and evaporator, can result in a substantial increase in the daily cooling production. The yearly solar COP of such thermochemical sorption process is around 18%, which is not far from those already obtained (around 22%) by other more efficient sorption process, e.g. liquid/gas absorption machines. Acknowledgments The authors wish to acknowledge the French Agency ADEME and the French research program ANR-PREBAT “ORASOL” 2006e2010 for their financial support to this project. References [1] International Energy Agency. Key World Energy Statistics; 2010. [2] Pérez-Lombard L, Ortiz J, Pout C. A review on buildings energy consumption information. Energy and Buildings 2008;40(3):394e8. [3] Chwieduk D. Towards sustainable-energy buildings. Applied Energy 2003; 76(1e3):211e7. [4] Srikhirin P, Aphornratana S, Chungpaibulpatana S. A review of absorption refrigeration technologies. Renewable and Sustainable Energy Reviews 2001; 5:343e72. [5] Ziegler F. Sorption heat pumping technologies: Comparisons and challenges. International Journal of Refrigeration 2009;32(4):566e76. [6] Dieng AO, Wang RZ. Literature review on solar adsorption technologies for ice-making and air-conditioning purposes and recent developments in solar technology. Renewable and Sustainable Energy Reviews 2001;5: 313e42.

269

[7] Florides GA, Tassou SA, Kalogirou SA, Wrobel LC. Review of solar and low energy cooling technologies for buildings. Renewable and Sustainable Energy Reviews 2002a;6:557e72. [8] Grossman G. Solar-powered systems for cooling, dehumidification and air conditioning. Solar Energy 2002;72:53e62. [9] Ziegler Felix. State of the art in sorption heat pumping and cooling technologies. Applied Thermal Engineering 1998;18(9e10):755e64. [10] Ziegler F. State of the art in sorption heat pumping and cooling technologies. International Journal of Refrigeration 2002;25(4):450e9. [11] De Francisco A, Illanes R, Torres JL, Castillo M, De Blas M, Prieto E, et al. Development and testing of a prototype of low-power water/ammonia absorption equipment for solar energy applications. Renewable Energy 2002; 25:537e44. [12] Sumathy K, Yeung KH, Yong L. Technology development in the solar adsorption refrigeration systems. Progress in Energy and Combustion Science 2003;29:301e27. [13] Boubakri A, Guilleminot JJ, Meunier F. Adsorptive solar powered ice maker: experiments and model. Solar Energy 2000;69:249e63. [14] Hildbrand C, Dind P, Pons M, Buchter F. A new solar powered adsorption refrigerator with high performance. Solar Energy 2004;77(3):311e8. [15] Li M, Wang RZ, Xu YX, Wu JY, Dieng AO. Experimental study on dynamic performance analysis of a flat-plate solar solid-adsorption refrigeration for ice maker. Renewable Energy 2002;27:211e21. [16] Spinner B. Ammonia-based thermochemical transformers. Heat Recovery Systems and CHP 1993;13(4):301e7. [17] Mauran S, Lahmidi H, Goetz V. Solar heating and cooling by a thermochemical process. First experiments of a prototype storing 60 kW h by a solid/gas reaction. Solar Energy 2008;82(7):623e36. [18] Le Pierres N, Stitou D, Mazet N. New deep-freezing process using renewable low-grade heat: from the conceptual design to experimental results. Energy 2007;32(4):600e8. [19] Le Pierrès N, Mazet N, Stitou D. Experimental results of a solar powered cooling system at low temperature. International Journal of Refrigeration 2007;30(6):1050e8; Le Pierres N, Stitou D, Mazet N. Modelling and performances of a deepfreezing process using low-grade solar heat. Energy 2007;32(2):154e64. [20] Stitou D, Mazet N, Bonnissel M. Performance of a high temperature hydrate solid/gas sorption heat pump used as topping cycle for cascaded sorption chillers. Energy 2004;29(2):267e85. [21] Dutour S, Mazet N, Joly JL, Platel V. Modeling of heat and mass transfer coupling with gasesolid reaction in a sorption heat pump cooled by a two-phase closed thermosiphon. Chemical Engineering Science 2005;60(15):4093e104. [22] Sorin M, Spinner B, Stitou D. Thermodynamic techniques for the conceptual design of thermochemical refrigerators using two salt materials. Chemical Engineering Science 2002;57:4243e51. [23] Li ZF, Sumathy K. Experimental studies on a solar powered air conditioning system with partitioned hot water storage tank. Solar Energy 2001;71(5): 285e97. [24] Wang RZ, Oliveira RG. Adsorption refrigeration-An efficient way to make good use of waste heat and solar energy. Progress Energy Combustion 2006;32(4): 424e58. [25] Wang LW, Wang RZ, Oliveira RG. A review on adsorption working pairs for refrigeration. Renewable and Sustainable Energy Reviews 2009;13(3): 518e34. [26] Iloeje OC. Design, construction and test run of CaCl2eNH3 solar refrigerator. Solar Energy 1985;35(5):447e55. [27] Erhard A, Spindler K, Hahne E. Test and simulation of a solar powered solid sorption cooling machine. International Journal of Refrigeration 1998;21(2): 133e41. [28] Erhard A, Hahne E. Test and simulation of a solar-powered absorption cooling machine. Solar Energy 1997;59(4e6):155e62. [29] Rivera C, Pilatowsky I, Méndez E, Rivera W. Experimental study of a thermochemical refrigerator using the barium chlorideeammonia reaction. International Journal of Hydrogen Energy 2007;32(15):3154e8. [30] Le Pierres N, Stitou D, Mazet N. Design of a thermochemical processor deep freezing using solar low-grade heat. Chemical Engineering Processing 2008; 47:484e9. [31] Le Pierrès N, Stitou D, Mazet N, Spinner B. Ideal Process conceptual design based on the exergetic analysis. New couplings of thermochemical dipoles for cold production from low grade heat. International Journal Energy Environment Economy 2003;12(3):105e15. [32] Dueñas C, Pilatowsky I, Romero RJ, Oskam A, Finck A. Dynamic study of the thermal behaviour of solar thermochemical refrigerator: barium chlorideammonia for ice production. Solar Energy Materials and Solar Cells 2001; 70(3):401e13. [33] Crozat G, Mauran S. Intégration de la pompe à chaleur chimique dans des systèmes de chauffage de l’habitat ou de stockage d’énergie. Revue Générale de Thermique 1987;311:605e11. [34] Boumaraf L, Etude des réactions chimiques dans les ammoniacates solides. Application du BaCl2/NH3 à la réfrigération solaire, thesis, Institut Polytechnique de Grenoble, 1989. [35] Dufour L-C, de Hartoulari R, Moutaabbid M. Solid-Ammonia systems and solar refrigeration: a kinetic study of the baryum chloride/ammonia system. Reactivity of Solids 1988;5:205e18.

270

D. Stitou et al. / Energy 41 (2012) 261e270

[36] Moutaabbid M, Pompes à chaleur chimiques et réfrigération solaire. Etude des systèmes solide-gaz: BaCl2-NH3-BaCl2.8NH3 et Zéolithe 13X-Fréon R22, thesis, Université de Bourgogne, 11 juillet 1986. [37] Gillespie LJ, Lurie E. Vapor pressures and latent heats for the system: BaCl2.8NH3BaCl2-NH3. Journal of American Chemical Society 1931;53(8):2978e83. [38] Biltz W, Hansen W. Über Ammoniakate der Alkalimetallhalogenide, Z. anorg. u. allgem. Chemistry 1923;127:1e33. [39] Coste C., Crozat G., Mauran S. Procédé de mise en oeuvre et réaction solidegaz. Patent FR 8 309 885, US patent 4 595 774; 1983.

[40] Mauran S, Prades P, L’Haridon F. Heat and mass transfer in consolidated reacting beds for thermochemical systems. Heat Recovery Systems and CHP 1993;13(4):315e9. [41] Lu HB, Mazet N, Coudevylle O, Mauran S. Comparison of a general model with a simplified approach for the transformation of solid-gas media used in chemical heat transformers. Chemical Engineering Science 1997;52(2): 311e27. [42] Wang SG, Wang RZ, Li XR. Research and development of consolidated adsorbent for adsorption systems. Renewable Energy 2005;30:1425e41.