Effect of heating time of adsorbercollector on the performance of a solar adsorption refrigerator
 Noureddine Cherrad^{1, 2}Email author,
 Adel Benchabane^{1},
 Lakhdar Sedira^{3} and
 Amar Rouag^{1}
DOI: 10.1186/s4071201700773
© The Author(s). 2017
Received: 9 January 2017
Accepted: 16 February 2017
Published: 27 February 2017
Abstract
This paper presents a numerical study of heat transfer inside the adsorbercollector of a solar adsorption refrigerator using the activated carbon AC35methanol pair. The objective is to estimate the amount of the heat loss through the adsorbercollector, during the solar heating phase, and to determine the effect of heating time on the thermal efficiency of the system. The numerical results showed that the heating time is the most important factor affecting the amount of energy loss. It has shown that the shorter heating time corresponds to the higher efficiency of the adsorbercollector. In addition, a new optimal coefficient of performance, COP _{optm}, is proposed to determine the number of adsorbers to be added to a machine. This latter is considered for consuming an energy equivalent to that received by the adsorbercollector. These additional adsorbers use a heat transfer fluid, coming from the adsorbercollector, instead of direct heating by solar radiation. An application example is presented using experimental results obtained from the literature. It has shown that the number of the additional adsorbers can reach three adsorbers.
Keywords
Adsorber thermal efficiency Heat loss Heat recovery Optimal coefficient of performance Solar adsorption refrigeratorBackground
The fossil energy is the foundation of the global economy; the consumption, the transport, and the delivery of this energy source are harmful to the human being, the animals, and the nature. We know that the combustion of oil rejects a thousand polluting substances in the atmosphere; one of the wellknown direct impact is global warming.
The search for alternative energy sources brought together environmentalists and those who for other reasons do not want to be dependent on oil suppliers. Certainly, other sources of energy may be an alternative to oil, such as solar energy.
Solar power in itself does not present a problem. The problem of its use comes mostly from the investment cost which is relatively expensive compared to fossil energy because the profitability of certain types of solar systems is still weak. The improvements are not intended only for the materials of the solar installation but also for the performance of the system, which includes all energy losses caused by its components.
The ordinary refrigerating machines, even those that are operated by electricity, are indirect consumer devices of petroleum energy, so a system that must be replaced in the future by an equipment belonging to solar power system technology. In the refrigeration machines powered by the electricity, the compressor is the driving force of the thermodynamic cycle of the system, while in refrigeration machines powered by the sun, the phenomenon of desorption by solar heating can be an alternative to the compression phase. This phenomenon takes place in the adsorbercollector of the solar machine; the more the desorption is important, the more the cold production is high.
The estimation of coefficient of heat loss per unit of time and temperature within the adsorbercollector is already developed by Klein 1975 and discussed by Duffie and Beckman 2013. Jing and Exell 1994 used the coefficient of loss developed by Klein 1975 for simulation and sensitivity analysis of an intermittent solarpowered charcoal/methanol refrigerator. Anyanwu et al. 2001 have made a study of an adsorption refrigerator using the charcoal/methanol pair, where the numerical model takes into account loss coefficient. The results of coefficient of performance (COP) are given according to the physical and dimensional characteristics of adsorbercollector. The loss coefficient was also considered by Leite and Daguenet 2000. It was a numerical simulation for instant transfers of heat and mass in each element of the refrigerating adsorption machine during the typical average day of each month. Another use of the coefficient of Klein was in the work of Chekirou et al. 2014, who introduced a modeling of heat and mass transfer in the tubular adsorbent of a refrigerating adsorption machine.
The objective of our study is to analyze the effect of heat loss within the adsorbercollector on the performance of solar adsorption refrigeration machines using AC35methanol as working pair and to determine the criteria affecting the increase of this amount of energy.
The amount of heat loss through the adsorbercollector was calculated using the coefficient of loss over the solar heating time and an interval of temperatures between the temperature of adsorption (temperature at start of heating) and temperature of generation (temperature at end of heating). The study is done using a numerical program (Fortran) and validated with experimental results obtained from the literature.
Methods
Basic components
A valve is set at inlet and outlet of adsorbercollector, to control the flow of the refrigerant from the adsorbent towards the condenser during the desorption phase, and from the evaporator towards the adsorbent during the adsorption phase.
Intermittent cycle
First, the adsorbercollector is heated by solar energy, which increases the pressure of the adsorbentadsorbate pair up to the condensing pressure P _{c} necessary for the condenser (processes A–B). In state B of cycle, the valve at the inlet of condenser is open, and the continuing of the heating at constant pressure induces the desorption of refrigerant (adsorbate) from the adsorbent towards the aircooled condenser to be condensed at condensing temperature T _{c} (processes B–C).
The desorbed mass during processes B–C is the difference between the maximal mass of adsorbate in state B and the minimal mass of adsorbate in state C.
The maximum temperature (generating temperature T _{g}) reached by the adsorbent will be in state C. At this value of temperature, the valve is closed and solar irradiance starts to decrease, which induces the drop of temperature and pressure of adsorbercollector (processes C–D).
When the pressure reaches the value P _{e}, which reigns in the evaporator, the valve is open, and the continuing of the cooling of adsorbent at constant pressure induces the evaporation of refrigerant at an evaporating temperature T _{e} from the evaporator towards the adsorbent by the adsorption phenomena (processes D–A). By this way, the evaporated refrigerant extracts heat and generates cold production within the cold chamber.
The adsorbed mass during processes D–A is the difference between the maximal mass of adsorbate in state A and the minimal mass of adsorbate in state D.
This cycle is said to be intermittent because the cold production starts only at sunset.
Adsorbed mass
Where W _{0} is the maximum volume adsorbed, ρ _{l} is the density of the adsorbate in the liquid state, D is the coefficient of affinity, T is the temperature of the adsorbercollector, P _{s} is the saturation pressure of the adsorbate, P is the equilibrium pressure of adsorbentadsorbate pair, and n is the parameter of adjustment of the D–A equation.
Where T _{s1} and T _{s2} are respectively the temperature at start of desorption and temperature at start of adsorption.
During the adsorption, the evaporating pressure is P _{e} = P _{s} (T _{e}), and for the desorption, the condensing pressure is P _{c} = P _{s} (T _{c}). T _{e} and T _{c} are respectively the evaporating temperature and the condensing temperature.
Thermal efficiency of the adsorbercollector
Cp _{l} is the specific heat of adsorbate in liquid state.
L is the latent heat of the adsorbate, r is the particular gas constant of the adsorbate, and α is the thermal expansion coefficient of the adsorbate.
According to Anyanwu et al. 2001, the heat loss from the lateral surfaces of adsorbercollector is assumed to be negligible.
U _{t} is the loss coefficient from the top of the adsorbercollector, and U _{b} is the loss coefficient from the bottom of the adsorbercollector.
T _{p} is the wall temperature of the top of adsorbercollector, T _{am} is the ambient temperature, n _{g} is the number of glass cover of adsorbercollector, E _{p} is the emissivity of the top wall of the adsorbercollector, E _{g} is the emissivity of the glass cover of the adsorbercollector, Ω is the adsorbercollector inclination, and W _{v} is the wind velocity.
Chekirou et al. 2014 took U _{b} as the constant.
Using Eq. (12), we cover the full time of heating process t _{max} on a range of wall adsorbercollector temperatures from Tp _{min} (temperature of adsorbercollector wall at the start of heating) to Tp _{max} (temperature of adsorbercollector wall at the end of heating).
And for Eq. (11), the temperature ambient T _{am} is considered to be equal to T _{a}.
Equation (15) means that Q _{r} = 0 for a heating time t _{h} = t _{max} and Q _{r} = Q _{L} for heating time t _{h} = 0.
Coefficient of performance
The first term of Eq. (18) represents the heat absorbed by the evaporation of the refrigerant (adsorbate) at the evaporation temperature T _{e}. The second term represents the sensible heat necessary to bring the condensate of its condensing temperature T _{c} to evaporating temperature T _{e} (Chekirou et al. 2011).
Results and discussion
Data used in the present numerical study
Symbol  Parameter  Value 

Cp _{d}  Specific heat of adsorbent  920 J kg^{−1} K^{−1} 
Cp _{t}  Specific heat of tubes (copper) containing the adsorbent  380 J kg^{−1} K^{−1} 
E _{g}  Emissivity of glass cover  0.88^{a} 
E _{P}  Emissivity of the top wall of adsorbercollector  0.1^{a} 
n _{g}  Number of glass cover  1^{a} 
r  Particular gas constant  259.34 J kg^{−1} K^{−1} 
S  Adsorbercollector surface  0.73 m^{2} 
t _{max}  Maximum time of heating process A–C (Fig. 2)  12 h 
U _{b}  Loss coefficient from the bottom of the adsorbercollector  0.9 W K^{−1a} 
W _{v}  Wind velocity  1 m s^{−1} 
α  Thermal expansion coefficient of adsorbate  0.00126 K^{−1b} 
Ω  Adsorbercollector inclination  34 °C^{a} 
σ  StefanBoltzmann constant  5.670367E8 W m^{−2} K^{−4} 
Values of DubininAstakhov equation parameters
Symbol  Parameter  Value 

D  Coefficient of affinity  5.02E07^{a} 
n  Parameter of adjustment of DA equation  2.15^{b} 
W _{0}  Maximum adsorbed volume  0.000425 m^{3} (of adsorbate) kg^{−1}(of adsorbent)^{b} 
Thermophysical properties of methanol along the saturation line (Bejan and Kraus 2003)
T [K]  P _{s} [MPa]  ρ _{l} [kg m^{−3}]  Cp _{l} [kJ kg^{−1} K^{−1}]  L [kJ kg^{−1}] 

200  6.1 × 10^{−6}  880.28  2.2141  1289.99 
250  0.00081  831.52  2.3121  1234.79 
300  0.0187  784.51  2.5461  1166.17 
325  0.0603  760.74  2.7223  1124.564 
350  0.1617  735.84  2.9362  1075.936 
375  0.3748  708.86  3.1891  1017.33 
400  0.7737  678.59  3.4912  944.57 
425  1.4561  643.38  3.8713  852.59 
450  2.5433  600.49  4.4067  739.19 
475  4.1688  544.35  5.3805  610.48 
500  6.5250  451.53  9.9683  391.09 
For each reference of validation, i.e., Douss and Meunier 1988 and Lemmini and Errougani 2007, the mass m _{d} of the adsorbent (activated carbon AC35) has been taken the same as that considered by the authors. Lemmini and Errougani 2007 have given the value of the adsorbent mass, which is equal to 14.5 kg, and for the machine studied by Douss and Meunier 1988, it is estimated by 12.5 kg.
The same for the mass m _{t} of metallic tubes containing the adsorbent, which are estimated to be equal to 45.68 and 26.024 kg for Douss and Meunier 1988 and Lemmini and Errougani 2007 respectively.
The maximum time t _{max} of heating processes A–C (Fig. 2) is considered for all day, so it is an average of 12 h.
Validation of numerical model
First, we have validated our numerical model by the results given by Douss and Meunier 1988. The authors have studied an experimental unit of an adsorption refrigeration machine that operates by electrical heating instead of the solar heating; therefore, heat loss Q _{L} cannot be estimated by Eq. (13), which is destined to solar adsorption adsorbercollectors. Douss and Meunier 1988 have given this amount of heat; it is equal to 365 kJ. For the temperatures’ data, we used the data of working fluid (WF) by Douss and Meunier 1988.
The average relative errors between the results compared in the Figs. 3, 4, and 5 are respectively 4.77, 5.22, and 9.96%.
The validation with the work of Douss and Meunier 1988 was made for the four calculated heats, i.e., Q _{1}, Q _{2}, Q _{3}, and Q _{des}, while Q _{L} is given by the author. Our numerical model takes into account the prediction of this quantity (Q _{L}), that is why we are going to carry out a second validation with Lemmini and Errougani 2007 to validate all the total energy Q _{ge}, ensuring by this way the validation of Eq. (13).
Numerical results of COP variation compared with those calculated during the tested days by Lemmini and Errougani 2007
Day  COP^{a}  COP^{b}  Relative error [%] 

02 April  9.00  9.65  7.22 
03 April  8.00  7.86  1.75 
04 April  11.00  10.10  8.18 
05 April  9.00  8.49  5.67 
07 April  6.90  7.54  9.28 
10 April  6.00  5.73  4.50 
11 April  10.20  10.69  4.80 
12 April  8.60  8.55  0.58 
13 April  7.90  7.73  2.15 
14 April  6.00  6.24  4.00 
15 April  9.00  9.65  7.22 
We have calculated the cooling production quantities Q _{ev} obtained by Lemmini and Errougani 2007 for eleven tested days by using Eq. (17), where COP and Q _{ge} are known. After that, we have predicted the operating temperatures of the studied machine.
The advantage of the developed numerical program is the prediction of operating temperatures T _{a}, T _{c}, and T _{g} for given cold production Q _{ev} and evaporating temperature T _{e}. The calculated temperatures will be used to estimate the amount of heat Q _{ge}.
Thermal efficiency and performance coefficient
The obtained results for the amounts of heats estimated, the thermal efficiency of the adsorbercollector, and the coefficient of performance of solar adsorption refrigerator in this subsection are based on the data considered for validation with the results of Lemmini and Errougani 2007.
The center of this curve takes a parabolic form, where it must coincide with the recovered heat quantity Q _{r}(t _{h})/Q _{ge}(t _{max}) = 50% and the quantity of heat received Q _{ge}(t _{h})/Q _{ge}(t _{max}) = 50% at t _{h}/t _{max} = 0.33.
At this time where Q _{u}(t _{h})/Q _{ge}(t _{max}) = Q _{L}(t _{h})/Q _{ge}(t _{max}) = 25% and using Eq. (16), the value of thermal efficiency η is 50%. It is the moment where the three quantities must take the same value which equals to 50% and corresponds to t _{h}/t _{max} = 0.33.
The shorter heating times correspond to higher efficiencies of the adsorbercollector. It can allow to recover up to 75% of total heat received by the adsorbercollector during t _{max} with zero heat loss. This means that improving design of the adsorbercollector is required to ensure the useful amount of solar energy necessary for system functioning during a very short time. So, it is recommended to use a heat transfer fluid instead of direct heating by solar radiation. Thus, we can isolate the upper side of the adsorbercollector that represents the large hole of the heat, and we command the heating time by the temperature and the flow rate of the heat transfer fluid. The fluid will flow in the tubes of a heat exchanger set inside the adsorbercollector. Even if the prior heating of heat transfer fluid by solar energy takes time, we can use a system of energy storage in the form of heat. Thus, we ensure the heating of the adsorbercollector even at night, and the cycle of the machine becomes continuous instead of intermittent cycle with a very good performance.

One additional adsorber needs →Q _{u} (*)

Number of additional adsorbers (Num _{re}) to be added need →Q _{r} (**)
By Eq. (20), we can see that a high optimal coefficient of performance COP _{optm} corresponds to a heating of shorter time t _{ h }. If t _{ h } = t _{max}, no energy recovery (by Eq. (19) Qr = 0⇒Num _{re} = 0) and COP optm would take the value of the normal COP.
Conclusions
In the present paper, a numerical model was developed to predict the different amounts of heat received by the adsorbercollector of a solarpowered adsorption refrigeration machine using activated carbon AC35/methanol as a working pair.
Four amounts of heat are useful for the operation of system, while the fifth amount is an energy loss. The estimate of this lost energy showed that the time of heating of the adsorbercollector is a very important factor affecting its thermal efficiency.
The shorter heating time allows to cut off the continuation of heat loss into the external environment and gives a good performance.
If we use a system heated by solar energy, the reduction of the heating time is somewhat difficult because it is related to the radiation intensity, which is as function of the time. So, it is useful to use a heat transfer fluid heated by the sun with storage system using the water for example instead the direct heating by solar radiation.
Furthermore, the performed study allows to define a new optimal coefficient of performance (COP _{optm}) more generalized. It takes the value of normal coefficient of performance (COP) for zero recovery of energy loss, and greater value for more energy recovery, keeping the same total energy necessary for the functioning of the system.
The usefulness of this optimal coefficient of performance is to determine the number of additional adsorbers that can be added to the machine for an equivalent consumption to energy received by the adsorbercollector.
The cold production increases with the number of additional adsorbers. Even though the days when solar radiation is low, the decrease in cold production will be better than that without an energy recovery system.
Nomenclature

COP Coefficient of performance

COP _{optm} Optimal coefficient of performance

Cp _{d} Specific heat of adsorbent, J kg^{−1} K^{−1}

Cp _{l} Specific heat of adsorbate in liquid state, J kg^{−1} K^{−1}

Cp _{t} Specific heat of tubes in metal containing the adsorbent, J kg^{−1} K^{−1}

D Coefficient of affinity

E _{g} Emissivity of glass cover of adsorbercollector

E _{P} Emissivity of top wall of adsorbercollector

L Latent heat of adsorbate, J kg^{−1}

m Adsorbate mass, kg

m _{d} Adsorbent mass, kg

m _{t} Mass of metallic tubes containing the adsorbent, kg

n Parameter of adjustment of D–A equation

n _{g} Number of glass cover of adsorbercollector

Num _{re} Number of additional adsorbers

r Particular gas constant of adsorbate, J kg^{−1} K^{−1}

P Equilibrium pressure of adsorbentadsorbate pair, Pa

P _{c} Condensing pressure of adsorbate, Pa

P _{e} Evaporating pressure of adsorbate, Pa

P _{s} Saturation pressure of adsorbate, Pa

Q _{des} Heat necessary for desorption process, J

Q _{ev} Heat load in evaporator (cooling production), J

Q _{ge} Total energy received by adsorbercollector, J

q _{st} Isosteric heat, J kg^{−1}

Q _{1} Energy supplied to heat the adsorbent, J

Q _{2} Energy supplied to heat the tubes in metal containing the adsorbent, J

Q _{3} Energy supplied to heat the adsorbate mass, J

Q _{L} Heat loss by adsorbercollector during the time of heating (processes A–C of Fig. 2), J

Q _{r} Recovered heat from adsorbercollector, J

Q _{u} Useful energy for functioning of system, J

S Adsorbercollector surface, m^{2}

T Temperature, K

T _{a} Adsorption temperature (temperature at end of adsorption), K

T _{am} Ambient temperature, K

T _{c} Condensing temperature of adsorbate, K

T _{e} Evaporating temperature of adsorbate, K

T _{g} Generating temperature (temperature at end of desorption), K

T _{p} Wall temperature of the top of adsorbercollector, K

Tp _{max}Maximal temperature of heated walls of adsorbercollector (including upper and bottom parts), K

Tp _{min}Minimal temperature of heated walls of adsorbercollector (including upper and bottom parts), K

t _{h} Heating time of adsorbent, s

t _{max}Maximum time of heating processes A–C (Fig. 2), s

U _{b} Loss coefficient from the bottom of the adsorbercollector, W K^{−1}

U _{L} Global loss coefficient from the adsorbercollector surfaces, W K^{−1}

U _{t} Loss coefficient from the top of the adsorbercollector, W K^{−1}

T _{s1} Temperature at start of desorption, K

T _{s2} Temperature at start of adsorption, K

W _{v} Wind velocity, m s^{−1}

W _{0} Maximum adsorbed volume, m^{3} (of adsorbate) kg^{−1} (of adsorbent)

X _{max} Maximal adsorbate mass concentration in 1 kg of adsorbent, kg kg^{−1}

X _{min} Maximal adsorbate mass concentration in 1 kg of adsorbent, kg kg^{−1}
^{Greek symbols}

α Thermal expansion coefficient of adsorbate, K^{−1}

ΔT _{am} Variation of ambient temperature, K

ΔT _{p} Variation of upper wall temperature of adsorbercollector, K

η Thermal efficiency of adsorbercollector

\( \varOmega \) Adsorbercollector inclination, °

ρ _{l} Density of the adsorbate in the liquid state, kg m^{−3}

σ StefanBoltzmann constant, W m^{−2} K^{−4}
Declarations
Authors’ contributions
All authors read and approved the final manuscript.
Competing interests
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Authors’ Affiliations
References
 Anyanwu, E. E., Oteh, U. U., & Ogueke, N. V. (2001). Simulation of a solid adsorption solar refrigerator using activated carbon/methanol adsorbent/refrigerant pair. Energy Conversion and Management, 42, 899–915.View ArticleGoogle Scholar
 Bejan, A., & Kraus, A. D. (2003). Heat transfer handbook. New York: Wiley.Google Scholar
 Chekirou, W. (2008). Etude et analyse d’une machine frigorifique solaire à adsorption. Alegria: University of Constantine. Doctorat Thesis.Google Scholar
 Chekirou, W., Boussehain, R., Feidt, M., Karaali, A., & Boukheit, N. (2011). Numerical results on operating parameters influence for a heat recovery adsorption machine. Energy Procedia, 6, 202–216.View ArticleGoogle Scholar
 Chekirou, W., Chikouche, A., Boukheit, N., Karaali, A., & Phalippou, S. (2014). Dynamic modelling and simulation of the tubular adsorber of a solid adsorption machine powered by solar energy. International Journal of Refrigeration, 39, 137–151.View ArticleGoogle Scholar
 Critoph, E. (1988). Performance limitations of adsorption cycles for solar cooling. Solar Energy, 41(1), 21–31.View ArticleGoogle Scholar
 Douss, N., & Meunier, F. (1988). Effect of operating temperatures on the coefficient of performance of active carbonmethanol systems. Heat Recovery Systems & CHP, 8(5), 383–392.View ArticleGoogle Scholar
 Duffie, J. A., Beckman, W. (2013). Solar engineering of thermal Processes (4th ed.). New York: Wiley.
 Goodwin, R. D. (1987). Methanol thermodynamic properties from 176 to 673 K at pressures to 700 bar. Journal of Physical Chemistry, 16(4), 799–892.MathSciNetGoogle Scholar
 Jing, H., & Exell, R. H. B. (1994). Simulation and sensitivity analysis of an intermittent solarpowered charcoal/methanol refrigerator. Renewable Energy, 4(1), 133–149.View ArticleGoogle Scholar
 Klein, S. A. (1975). Calculation of fiatplate collector loss coefficients. Solar Energy, 17, 79–80.View ArticleGoogle Scholar
 Leite, A. P. F., & Daguenet, M. (2000). Performance of a new solid adsorption ice maker with solar energy regeneration. Energy Conversion and Management, 41, 1625–1647.View ArticleGoogle Scholar
 Lemmini, F., & Errougani, A. (2007). Experimentation of a solar adsorption refrigerator in Morocco. Renewable Energy, 32, 2629–2641.View ArticleGoogle Scholar
 Pons, M., & Grenier, P. H. (1986). A phenomenological adsorption equilibrium law extracted from experimental and theoretical considerations applied to the activated carbon + methanol pair. Carbon, 24(5), 615–625.View ArticleGoogle Scholar
 Rouag, A., Benchabane, A., Labed, A., & Boultif, N. (2016). Thermal design of air cooled condenser of a solar adsorption refrigerator. Journal of Applied Engineering Science & Technology, 2(1), 23–29.Google Scholar