Energy Conversion and Management 106 (2015) 759–765

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier .com/ locate /enconman

Performance investigation of a solar hot water driven adsorptionice-making system

http://dx.doi.org/10.1016/j.enconman.2015.10.0320196-8904/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: jixu@ynnu.edu.cn (X. Ji).

Xu Ji ⇑, Xiangbo Song, Ming Li, Jiaxing Liu, Yunfeng WangSolar Energy Research Institute, Yunnan Normal University, Kunming 650500, Yunnan, China

a r t i c l e i n f o

Article history:Received 29 May 2015Accepted 11 October 2015

Keywords:Solar adsorption refrigerationHot water drivenIce-makingCOPHeat utilization efficiency

a b s t r a c t

A solar hot water driven solid adsorption ice-making system with heat storage was designed and con-structed. The finned-tube absorbent bed in the water tank, which also acted as heat storage unit, washeated by the hot water from the solar vacuum tube collector in desorption process. The water in the tankwas also auxiliary heated by electric heater to keep at the set desorption temperature. Activated carbon–methanol was utilized as the working pairs in the system. Effects of heat source temperature on systemperformance were experimentally investigated under four conditions: maintaining the water tempera-ture in the tank at 94 �C (boiling point), 85 �C, 75 �C, respectively, and heating the water to reach 94 �Cthen naturally cooling the hot water without maintaining heating in desorption process. The experimen-tal results showed that the temperatures around the finned-tube of the adsorbent bed was homogeneous,which was beneficial for desorption of the adsorbate. The maximum daily ice-making capacity of 8.4 kgand the lowest temperature of�8.6 �C were achieved when the hot water temperature was maintained at94 �C. The maximum refrigeration cycle coefficient of performance (COP) of 0.139 was obtained under thecondition of heating the water to reach 94 �C then cooling naturally the hot water without maintainingheating. The system heat utilization efficiency decreased with the increase of heat source temperaturedue to the greater heat loss in desorption process.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

With development of human society, the increasing energyneeds and environmental crisis are attracting great attentionacross the world. The requirements for air conditioning and refrig-eration consume a large quantity of conventional energy resourcesand often lead to power shortage in peak period, especially in sum-mer. Solar energy is a clean energy, and its total quantity is approx-imately infinite [1]. Moreover, the summer, which demands thegreatest cooling capacity, is usually the season with the strongestsolar radiation [2]. The abundant solar radiation resources in sum-mer could meet the great energy consumption demand for refrig-eration in this season. Therefore, development and utilization ofsolar energy in refrigeration field were of great practical sig-nificance. Solar refrigeration technology includes mainly thesolar-driven ejector refrigeration, the solar absorption refrigerationand the solar adsorption refrigeration. All these technologies uti-lize environmental friendly refrigerant and do not employcompressor, thus these systems have low noise in operation. Topromote the system coefficient of performance (COP) is the

common objective for these solar-driven refrigeration technolo-gies. Solar absorption refrigeration technology was developedearly, and was relatively mature now. Its system volume and cool-ing capacity were ordinarily large, thus was suitable for centralcooling in large scale buildings. Lithium bromide was usually usedas refrigerant in solar absorption refrigeration system. However,lithium bromide was easy to crystallize, and lithium bromide solu-tion has strong corrosion to general metals, which would shortenthe system lifetime. In addition, the solar absorption refrigerationsystem could only provide cooling capacity above 0 �C. Solar-driven ejector refrigeration system adopted ejector to replace com-pressor in traditional refrigeration system, and produced coolingcapacity by liquid evaporation. The low system performance wasthe main problem for such system. Steam jet refrigerator withwater as the working fluid could not produce ice too. Currentstudies on ejector refrigeration technology focused on structureoptimization of the ejectors, performance improvement of theworkingmedium, and sustaining performance under unstable solarirradiation. Combination of ejector refrigeration and absorption/adsorption refrigeration was an effective way to promote perfor-mance of the ejector refrigeration system.

The earliest experiment on adsorption refrigeration was con-ducted in 1848. When ammonia was adsorbed by silver chloride,

Nomenclature

ASC surface area of solar vacuum tube collector (m2)Cwater specific heat of water J/(kg K)COPcycle coefficient of performance of the refrigeration cycleMwater1 the mass of water in solar vacuum tube water heating

system (kg)Mwater2 the mass of water in water tank (kg)Mwater_los mass of the evaporated water (kg)I(t) solar radiation intensity (J/m2 S)qwater_los evaporating latent heat of water (J/kg)Qcc sensible heat given off by the refrigerant liquid (J)Qelectric heat supplied by the electric heater (J)Qe_ice refrigeration quantity for ice in ice box from initial tem-

perature (in liquid phase) to final temperature, includ-ing sensible heat and latent heat (J)

Qe_water refrigeration quantity for remaining water in ice boxfrom initial temperature to final temperature (J)

Qe_M sensible heat of the metal evaporator from ambienttemperature to final temperature (J)

Qe_L cooling quantity loss of the ice box (J)Qg heat for regeneration of adsorbent bed (J)QI total solar radiation incident to the collector (J)Qref refrigeration quantity (J)Qu all heat to the adsorber (J)Qwater1 sensible heat to heat the water in solar vacuum tube

water heating system (J)Qwater2 heat obtained by the water in water tank (J)Qwater_eva latent heat of the evaporated water into ambientTa ambient temperature (�C)Tc condensing temperature (�C)Te evaporating temperature (�C)Tg generating temperature (�C)gSC collection efficiency of solar water heating system (%)gsystem system heat utilization efficiency (%)

760 X. Ji et al. / Energy Conversion and Management 106 (2015) 759–765

the cooling effect was produced. In 1920s, Hulse proposed a refrig-eration system with silica gel–sulfur dioxide as working pairs [3].In 1960, Plank and Kuprianoff utilized activated carbon–methanolas refrigeration working pairs [4]. In 1978, Tchernev in AmericanZeolite Power Company developed the first solar-powered inter-mittence adsorption refrigeration equipment with zeolite–wateras the working pairs [5]. Critoph [6] and Meunier [7] comparedthe performance of activated carbon–methanol (water) and molec-ular sieve–water (methanol) in adsorption refrigeration systems,and concluded that the activated carbon–methanol had a highercooling efficiency when the desorption temperature was below150 �C. Wang used different silica gel samples as adsorbates, andfound solid particulate pollution deteriorated greatly the adsorp-tion capacity [8]. Jiang developed successfully CaCl2–NaBr–NH3

working pairs for two-stage chemisorption freezing cycle drivenby low temperature heat source [9].

In 1997, Wang [10] prepared an activated carbon fiber adsor-bent, with cooling capacity 2–3 times higher than that of the reg-ular activated carbon. The adsorption/desorption time was shortento 1/10 of that with the regular activated carbon. Since then, devel-oping novel composite adsorbent with high thermal conductivitybecame a research focus. In 2012, Wang [11] demonstrated a con-solidated composite activated carbon with higher thermal conduc-tivity and thermal diffusivity. In 2014, Jiang [12] investigated aconsolidated composite CaCl2 with a matrix of expanded naturalgraphite treated with sulfuric acid. The composite had very per-spective heat and mass transfer performance. Compared with zeo-lite molecular sieve, the foamed aluminum composite resulted inshorter cycle period and higher COP [13]. By utilizing calcium chlo-ride/expandable graphite composite adsorbent, the average heattransfer coefficient of the system in process of cooling and heatingincreased by 265% and 300% respectively [14]. Lu [15] did experi-mental research on adsorption chillers using micro-porous silicagel–water and compound adsorbent–methanol. Li [16] studiedthe adsorption performance of composite adsorbent of CaCl2 andexpanded graphite with ammonia as adsorbate.

Another important way for improving performance of solaradsorption refrigeration system was to enhance heat and masstransfer in the adsorbent bed. Hong [17] built a fin-tube typeadsorption chiller and found that the fin thickness and the hotwater temperature were the dominant parameters for COP andSCP. Leite [18] designed a copper tube solar powered adsorptionrefrigeration system using the transparent honeycomb material

kneading board. The daily ice-making capacity was 7–10kg/(m2 d). Hildrand [19] developed a metal tube adsorbent bedand the solar cooling COP was between 0.10 and 0.25. Bao [20] pre-sented a small refrigerator based on resorption technique to utiliz-ing low grade thermal energy. CPC collectors were also adopted toimprove the system performance [21,22]. Hassan [23] and Jribi[24] simulated and improved the adsorption refrigeration cycle.Lu [25] developed a novel solar silica gel–water adsorption air con-ditioning and analyzed its performance. Hamdeh [26] optimizedsolar adsorption refrigeration system by experimental and statisti-cal techniques. However, the flat plate collector, which was usuallyadopted as heat collecting unit of solar solid adsorption refrigera-tion system, inevitably resulted in the inhomogeneous heating onthe adsorbent bed.

In this paper, the proposed solar adsorption refrigeration sys-tem employed activated carbon–methanol as working fluid, whichwas noncorrosive and had stable performance under 0 �C. The heatcollected by vacuum tube was stored in water bath, and providedstable energy input for the adsorbent bed at the desorption stage.The adsorbent bed consisted of large-diameter aluminum-alloyfinned tubes with excellent heat and mass transfer performanceand was kept in water bath. Thus temperature around the adsor-bent bed was homogeneous, beneficial to sorption/desorption ofrefrigerant. The performance experiments of the solar hot waterdriven adsorption ice-making system were conducted.

2. System descriptions

The schematic diagram of a solar hot water driven solid adsorp-tion ice-making system with heat storage was shown in Fig. 1. Thesystem consisted of solar collector, adsorbent bed, condenser,evaporator, vacuum valves, vacuum pressure gauge, circulatingwater pumps and electric heater. Solar radiation projected on thesolar collector in the daytime and the water temperature in solarwater heating system rose. The adsorbent bed in water bath washeated by the hot water, which was pumped from solar waterheating system to the water tank at the beginning of desorptionprocess. With increasing of the adsorbent bed temperature, therefrigerant was gradually desorbed from the adsorbent in theadsorbent bed. When the vapor pressure of refrigerant in theadsorbent bed reached the condensing pressure, the vacuum valvewas opened. Then refrigerant vapor was condensed and flowedinto the evaporator. The electric heater was also utilized to

1-adsorbent bed 2-vacuum pressure gauge 3-vacuum valves 4-condenser 5-evaporator

6-ball valve 7-brackets 8-ice box

Fig. 1. System schematic diagram.

Fig. 2. System prototype.

Fig. 3. Schematic diagram of adsorbent bed.

X. Ji et al. / Energy Conversion and Management 106 (2015) 759–765 761

auxiliary heat and keep the water bath at the set desorption tem-perature. The hot water in the water tank was pumped intoanother heat storage water tank at the end of desorption process,which could be reheat by the solar water heating system and uti-lized in next circulation. Fig. 2 showed the system prototype.

Fig. 3 was the schematic diagram of adsorbent bed and watertank. The water tank was made from stainless steel and its outersize was 1520 � 780 � 800 mm. The water tank was packed withan insulating layer of 60 mm thickness. 32 Large-diameter alu-minum alloy finned tubes were designed to enhance heat and masstransfer. The inner space formed by inner finned tube acted as

mass transfer channel. The gap between the outer and inner tubewas filled with 1 kg activated carbon per tube. The mass transferchannel and the absorbent were separated by mesh [27]. The outer,inner diameters and length of these tubes were 90 mm, 40 mm and50 mm respectively (structure parameters of the finned tube wasshown in Table 1). These tubes were connected by a mass transfertube to form the adsorbent bed, which was soaked in 200–220 kgwater in the water tank.

In desorption process, the desorbed refrigerant vapor flowedfrom the mass transfer pipe into a condenser. In the copper finnedtube condenser, the refrigerant (methanol) vapor flowed throughfour pipes and was condensed into liquid. The outer and innerdiameter of each pipe was 32.5 mm and 15.5 mm, respectively,

Table 1Structure parameters of finned tube.

Structure parameters(mm)

Outer/inner tube diameter of concentric finnedtube

90/40

Length of finned tube 500Thickness of finned tube 1.5Length of the long/short fin 25/10Thickness of fin 1.5Seam width of mass transfer 6

762 X. Ji et al. / Energy Conversion and Management 106 (2015) 759–765

and the length was 220 mm. The condensing tube was soaked incold water. Subsequently, the liquid methanol flowed into theevaporator by gravity. In evaporator, liquid methanol evaporatedand absorbed heat from the refrigeration room (ice box). The evap-orator consisted of three aluminum alloy rectangular boxes(500 mm � 100 mm � 100 mm) in the system, and the effectiveevaporation area was 0.66 m2. The ice was produced in the ice box.

3. System performance analysis

3.1. Performance of solar vacuum tube water heating system

The collection efficiency was defined as the ratio of the sensibleheat to heat the water in the solar vacuum tube collector and thetotal solar radiation incident to the collector, which was given by

gSC ¼ Qwater1

QIð1Þ

where Qwater1 was the sensible heat to heat the water in the solarvacuum tube collector; QI was the total solar radiation incident tothe collector. Qwater1 and QI could be expressed as (2) and (3),respectively.

Qwater1 ¼Z T2

T1

Mwater1Cwater dT ð2Þ

In Eq. (2), Mwater1 was the mass of water in the solar vacuumtube collector, Cwater was the specific heat of water.

QI ¼Z t2

t1

IðtÞASC dt ð3Þ

In Eq. (3), I(t) was the solar radiation intensity, ASC was the sur-face area of the solar vacuum tube collector.

3.2. Cooling performance of adsorption refrigeration system

The system COPref of adsorption refrigeration system could becalculated by the second law of thermodynamics as follow [28],

COPref ¼ TeðTg � TaÞTgðTa � TcÞ ð4Þ

In Eq. (4), Tg is generating temperature, Ta was ambient temper-ature, Tg was evaporating temperature and Tc was condensingtemperature.

The cooling performance of adsorption refrigeration systemcould also be described by the refrigeration cycle COP that wasthe ratio of the effective cooling capacity and the heat for theregeneration of adsorption bed,

COPcycle ¼ Q ref � Q cc

Qgð5Þ

In Eq. (5), Qref was the refrigeration quantity, Qcc was the sensi-ble heat given off by the liquid refrigerant from condensing tem-perature Tc to evaporation temperature Te, Qg was the heat forthe regeneration of adsorbent bed.

For convenient calculation,

Q ref ¼ Qe ice þ Qe water þ Qe M þ Qe L ð6Þwhere Qe_ice was the refrigeration quantity for ice in ice box frominitial temperature (in liquid phase) to final temperature, includingsensible heat and latent heat; Qe_water was the refrigeration quantityfor remaining water in ice box from initial temperature to final tem-perature; Qe_M was sensible heat of the metal evaporator fromambient temperature to final temperature; Qe_L was the coolingquantity loss from the ice box.

Qg ¼ Qu � Qwater2 � Qwater eva

¼ QI þ Q electric �R T3T1

Mwater2Cwater dT �Mwater losqwater lat

ð7Þ

In Eq. (7), Qu was all heat to the system, Qwater2 was the heatobtained by the water in water tank, Qwater_eva was latent heat ofthe evaporated water into ambient, Qelectric was the heat suppliedby the electric heater (as shown in Eq. (8)), Mwater2 was mass ofthe water in water tank, Mwater_los was mass of the evaporatedwater, qwater_lat was the evaporating latent heat of water.

Qelectric ¼ Pt ð8Þwhere P was the power of electric heater, t was the heating time.

The system heat utilization efficiency was written as follow,

gsystem ¼ Qwater2

Quð9Þ

4. Experiments and results

4.1. Experimental method

The experiments were conducted in winter in Kunming, China.The water at 60 �C in solar water heating system was pumped intothe water tank. After a short period, the water temperature inwater tank reached about 45 �C, and could no longer rise continu-ously. Then the electric heaters were employed to continuouslyheat water in the water tank until its temperature reached a setvalue. The electric heaters were also utilized to maintain the waterat set temperatures for a period. Generally, the maximum temper-ature was attained around 14:00.

When the end of desorption process, the hot water in watertank was rapidly pumped out and moved into a heat storage watertank. The cold water at about 15 �C was pumped into the watertank. The ice box was filled with 6–9 kg water. The temperatureof adsorbent bed dropped drastically to about 18 �C. The adsorp-tion process started and would last until the next morning. Thewater temperature in the water tank, adsorbent bed temperature,condenser temperature, evaporator temperature, water tempera-ture in the ice box and ambient temperature were measured bythermocouples of PT100 with an accuracy of ±0.2 �C. The solar mul-tichannel test recorder of TRM-FD2 was used to record the temper-atures. The adsorbent bed pressure was measured by vacuumpressure gauge. The total radiation was measured by pyranometerof TBQ-2 with an accuracy of ±0.2%.

4.2. Temperature distribution on the adsorbent bed

Performance of solar hot water driven adsorption ice-makingsystem could be promoted by improving the homogeneity of tem-perature distribution on the absorbent bed, which was beneficialfor the desorption of refrigerant during the desorption process. Inour experiments, the water bath temperature was controlled underfour conditions: maintaining the water temperature in water tankat 94 �C(boiling point), 85 �C and 75 �C, respectively, and heating

Fig. 4. Temperature variation at different parts of absorbent bed during desorptionprocess under four conditions.

Fig. 5. Variation of temperatures of adsorbent bed, evaporator and water in ice box(maintaining water temperature in water tank at 94 �C).

Fig. 6. Variation of temperatures of adsorbent bed, evaporator and water in ice box(heating the water to reach 94 �C then naturally cooling the hot water withoutmaintaining heating).

X. Ji et al. / Energy Conversion and Management 106 (2015) 759–765 763

the water to reach at 94 �C then cooling naturally the hot waterwithout maintaining heating during the desorption process.

Fig. 4 showed the temperature variation at different parts ofabsorbent bed during desorption process. In figure, the tempera-tures (at the top, the medium and the bottom) around the adsor-bent tube in water bath were uniform under all four coolingconditions, which was different with the temperature distributionaround adsorbent bed irradiated directly by sun. The temperatureon solar irradiation side of the latter was far greater than that onback side. Therefore, the desorbed refrigerant was less than thatin our system, which resulted deservedly into less cooling capacityin the next circulation.

For all four cases, the adsorbent bed temperature increasedrapidly to about 45 �C at initial heating stage, then increased tothe respective set values at slightly flat slope. The reason was thatthe cold water was pumped out and the hot water was pumpedinto the water tank from the solar vacuum tube collecting systemat the initial heating stage. Subsequently, the electric heater wasemployed to auxiliary heat until the water temperature reachedthe set temperatures.

Afterward, the adsorbent bed temperature were kept at 94 �C,85 �C, 75 �C for 5 h, 6 h, 7 h respectively or naturally cool to desorbthe refrigerant. For the naturally cooling case, the final tempera-ture was 82 �C at the end of desorption process. The temperaturefluctuation in curves at the stage was due to the intermittent heat-ing by electric heater, which worked when the water temperaturereached the set lower limit and stopped when it reached the sethigh limit.

After conclusion of desorption process, the water at ambienttemperature of 18 �C was imported into the water tank to coolthe adsorbent bed, the adsorbent bed temperature dropped drasti-cally from 94 �C, 85 �C, 82 �C and 75 �C to 18 �C within 1 h. In thecooling process, the cold water was used to cool the adsorptionbed, so temperature of the adsorption bed decreased relativelyrapid. The lower the adsorption bed temperature was, the betterthe adsorption performance of the adsorbent bed was. The rapidcooling of the adsorbent bed was beneficial for adsorption of therefrigerant, thereby the rapid evaporative cooling was achieved.

Fig. 7. Variation of temperatures of evaporator and water in ice box (maintainingthe water temperature in water tank at 85 �C).

4.3. Temperature variation of evaporator and water in ice box

The system cooling performance was also influenced by heattransfer performance of the evaporator, which could be improvedby enlarging the evaporation area of evaporator. The variation oftemperature of evaporator and water in the ice box with timeunder the four conditions were illustrated respectively in Figs. 5–8.

Fig. 8. Variation of temperatures of evaporator and water in ice box (maintainingthe water temperature in water tank at 75 �C).

Fig. 9. Ice production.

764 X. Ji et al. / Energy Conversion and Management 106 (2015) 759–765

In Fig. 5, with maintaining water temperature in water tank at94 �C, the adsorbent bed temperature dropped rapidly from 94 �Cto 18 �C, then rose slowly from 18 �C to about 23 �C due to theadsorption heat released in adsorption process. The temperatureof evaporator and water in the ice box dropped correspondinglyfrom about 13 �C to 0 �C, then the temperature dropped slowly to�8.8 �C. A large amount of phase change latent heat was releasedin water phase change process, thus the decreasing trend of tem-perature was slow after reaching at 0 �C. The temperature differ-ence between evaporator and ice was small, which indicated thatthe heat transfer performance of aluminum alloy evaporator wasexcellent.

In Fig. 6, under the condition of heating the water to reach at94 �C then naturally cooling the hot water without maintainingheating, the adsorbent bed temperature dropped rapidly from85 �C to 18 �C. The temperature of evaporator and water in theice box also dropped rapidly from 13 �C to 0 �C. Afterward, the tem-perature of evaporator and ice in the ice box dropped slowly. Thelowest ice temperature reached �4.3 �C.

In Figs. 7 and 8, with maintaining the water temperature inwater tank at 85 �C and 75 �C during the desorption process, thelowest temperature in the ice box was about �0.7 �C and �0.4 �Crespectively. Hence, the heat source temperature had a great influ-ence on the cooling performance of adsorption refrigeration sys-tem. The higher the heat source temperature was, the larger thecooling capacity was in a certain temperature range.

Table 2System performance.

No. 1 No. 2 No. 3 No. 4

Heat sourcetemperaturecondition

Maintainat 94 �C

Reach 94 �C, thennature cooling

Maintainat 85 �C

Maintainat 75 �C

Input energy (MJ) 116.8 94.34 71.8 67.98Ice/cooling water

capacity (kg)8.4/0.6 5.5/0.5 0/6 0/6

Lowesttemperature(�C)

�8.6 �4.2 �0.6 �0.4

Direct refrigeratingcapacity (MJ)

4.496 3.277 1.281 1.276

Hot water output(kg/�C)

212/94 215/89 210/85 210/75

COPcycle 0.127 0.139 0.097 0.069gsystem (%) 54.28 62.99 75.31 69.54

4.4. System performance and ice production

Table 2 illustrated the cooling performance of the adsorptionrefrigeration system under four conditions.

The system COP was respectively 0.127, 0.139, 0.097 and 0.069under four heat source temperature conditions. The maximumdaily ice-making capacity of 8.4 kg (as shown in Fig. 9) and thelowest temperature of �8.6 �C were achieved when the water tem-perature in water tank was maintained at 94 �C (boiling point) dur-ing the desorption process. The maximum COP of 0.139 wasobtained under the condition of heating the water to reach at94 �C then naturally cooling the water without maintaining heat-ing. The system daily ice-making capacity increased with increas-ing of heat source temperature, however, the system COP didn’tincrease obviously. The reason was that the heat loss caused byevaporation of water and heat dissipation increased drasticallywith increase of heat source temperature in the system, this partof heat loss could not be effectively utilized by the system.

The system heat utilization efficiencies were respectively0.5428, 0.6299, 0.7531 and 0.6954 under four heat source temper-ature conditions. It could be found that the higher the heat sourcetemperature was, and the longer the heating time was, the lowerthe heat utilization efficiency was. It was due to the greater heatloss caused by the evaporation of water and heat dissipation. Whenthe water temperature in water tank was maintained respectivelyat 85 �C and 75 �C, there was no ice-making phenomenon in the icebox. However, the cooling effect was also produced in ice box,which indicated that the system also had a good cooling perfor-mance at a low heat source temperature.

5. Conclusions

A solar hot water driven solid adsorption ice-making systemwith heat storage was designed and built in this paper. Perfor-mance experiments of the system prototype with activated car-bon–methanol as working pairs were conducted under four heatsource temperature conditions. Experimental results showed thatthe temperature around the finned-tube of adsorbent bed washomogeneous in desorption process, which was beneficial for des-orption of the refrigerant. The rapid cooling of adsorbent bed at thebeginning of adsorption process was beneficial for adsorption ofthe refrigerant by adsorbent. The maximum daily ice-makingcapacity of 8.4 kg and the lowest temperature of �8.6 �C wereachieved when the water temperature in water tank was main-tained at 94 �C. The maximum refrigeration cycle COP of 0.139

X. Ji et al. / Energy Conversion and Management 106 (2015) 759–765 765

was obtained under the condition of heating the water to reach at94 �C then naturally cooling the hot water without maintainingheating. There was still a certain improvement space for systemCOP at boiling point temperature by reducing the water evapora-tion. When the water temperature in water tank was maintainedat 85 �C and 75 �C, there was no ice-making phenomenon, how-ever, the cooling effect was also produced in ice box. Heat utiliza-tion efficiencies of the system were respectively 0.5428, 0.6299,0.7531 and 0.6954 under four heat source temperature conditions.The system heat utilization efficiency decreased with increase ofheat source temperature due to the greater heat loss in desorptionprocess.

Acknowledgement

The present study was supported by National Natural ScienceFoundation, China (Grant No. 51366014).

References

[1] Shu Xu. Experiment on new adsorption bed about adsorption refrigerationdriven by solar. Energy Proc 2012;14:1542–7.

[2] Liu Yue. A study on the status and prospects of solar photovoltaic fridge. Chin JRefrig Technol 2012;32(1):64–7.

[3] Hulse GE. Refroidissement d’un wagon frigorifique marchandises par unsystem adsorption utilisant le gel de silice. Revue Gen Froid 1929;10:281–9.

[4] Plank R, Kuprianoff J. In die Kieinltemaschine. Berlin: Springer-Verlag; 1960.[5] Tchernev DI. Solar energy application of natural zeolites, in natural zeolite:

occurrence, properties and use. Oxford: Pergamon Press; 1978.[6] Critoph RE, Vogel R. Possible adsorption pairs for use in solar cooling. Ambient

Energy 1986;7(4):183–90.[7] Meunier F. Theoretical performance of solid adsorbent cascading cycles using

the zeolite–water and active carbon–methanol pairs: four cases studies. HeatRecovery Syst 1988;6(6):491–8.

[8] Wang Dechang, Zhang Jipeng, Yang Qirong, Li Na, Sumathy K. Study ofadsorption characteristics in silica gel–water adsorption refrigeration. ApplEnergy 2014;113:734–41.

[9] Jiang L, Wang LW, Luo WL, Wang RZ. Experimental study on working pairs fortwo-stage chemisorption freezing cycle. Renewable Energy 2015;74:287–97.

[10] Wang RZ, Jia JP, Zhou YH. Study on a new solid adsorption refrigeration pair:active carbon fiber–methanol. ASME J Sol Energy Eng 1997;19(3):214–8.

[11] Wang LW, Metcalf SJ, Critoph RE, Thorpe R, Tamainot-Telto Z. Development ofthermal conductive consolidated activated carbon for adsorption refrigeration.Carbon 2012;50(3):977–86.

[12] Jiang L, Wang LW, Wang RZ. Investigation on thermal conductive consolidatedcomposite CaCl2 for adsorption refrigeration. Int J Therm Sci 2014;81:68–75.

[13] Hu Peng, Yao Juanjuan, Chen Zeshao. The non-balance adsorption refrigerationcharacteristic analyzes on zeolite molecular sieve/foamed aluminumcomposite adsorption material–water. J Eng Thermophys 2009;1(30):13–6.

[14] Li Suling, Xia Zaizhong, Wu Jingyi. Application of new composite adsorbent foradsorption refrigeration unit. J Refrig 2009;8:20–4.

[15] Lu Zisheng, Wang Ruzhu, Xia Zaizhong, Gong Lixia. Experimental investigationadsorption chillers using micro-porous silica gel–water and compoundadsorbent–methanol. Energy Convers Manage 2013;65:430–7.

[16] Li SL, Wu JY, Xia ZZ, Wang RZ. Study on the adsorption performance ofcomposite adsorbent of CaCl2 and expanded graphite with ammonia asadsorbate. Energy Convers Manage 2009;50(4):1011–7.

[17] Hong SW, Ahn SH, Kwon OK, Chung JD. Optimization of a fin-tube typeadsorption chiller by design of experiment. Int J Refrig 2015;49:49–56.

[18] Leite Antonio Pralon Ferreira, Daguenety Michel. Performance of a new solidadsorption ice maker with solar energy regeneration. Energy Convers Manage2000;41:1625–47.

[19] Hildbrand Catherine, Dind Philippe, Pons Michel, et al. A new solar poweredadsorption refrigerator with high performance. Sol Energy 2004;77:311–8.

[20] Bao HS, Wang RZ, Wang LW. A resorption refrigerator driven by low gradethermal energy. Energy Convers Manage 2011;52(6):2339–44.

[21] Umair Muhammad, Akisawa Atsushi, Ueda Yuki. Performance evaluation of asolar adsorption refrigeration system with a wing type compound parabolicconcentrator. Energies 2014;7(3):1448–66.

[22] González Manuel I, Rodríguez Luis R. Solar powered adsorption refrigeratorwith CPC collection system: collector design and experimental test. EnergyConvers Manage 2007;48(9):2587–94.

[23] Hassan HZ, Mohamad AA, Bennacer R. Simulation of an adsorption solarcooling system. Energy 2011;36:530–7.

[24] Jribi Skander, Saha Bidyut Baran, Koyama Shigeru, Bentaher Hatem. Modelingand simulation of an activated carbon–CO2 four bed based adsorption coolingsystem. Energy Convers Manage 2014;78:985–91.

[25] Lu ZS, Wang RZ, Xia ZZ. An analysis of the performance of a novel solar silicagel–water adsorption air conditioning. Appl Therm Eng 2011;31:3636–42.

[26] Hamdeh Nidal H Abu, Al-Muhtaseb Mu’Taz A. Optimization of solar adsorptionrefrigeration system using experimental and statistical techniques. EnergyConvers Manage 2010;51:1610–5.

[27] Ji Xu, Li Ming, Fan Jieqing, et al. Structure optimization and performanceexperiments of a solar-powered finned-tube adsorption refrigeration system.Appl Energy 2014;113:1293–300.

[28] Al-Muhtaseb M. Design of solar adsorption refrigeration unit. Master thesis inMechanical Engineering Department. Jordan: Jordan University of Science andTechnology; 2008.