ResearchGate An experimental study of freezing and melting of water inside spherical capsules used in thermal energy storage systems a sas Bedal.tgman ohare Se Seana Vaesy FS botng eat vance view poetAin Shams Enginesring Journal (2012) 3, 33.48 Ain Shams University Ain Shams Engineering Journal www levi com ocataase} “ww selene com MECHANICAL ENGINEERING An experimental study of freezing and melting of water inside spherical capsules used in thermal energy storage systems Reda I. ElGhnam *, Ramdan A. Abdelaziz, Mohamed H. Sakr, Hany E. Abdelrhman Faculty of Engineering, Benha University, Egypt Received 31 May 2011; accepted 23 October 2011 Available online 26 December 2011 KEYWORDS. Abstract This paper reports the results of an experimental study on the heat transfer during fez ing (charging) and melting (discharging) of water inside a spherical capsule of the type often found Spherical capsules inthe beds of thermal (ice) storage systems used forthe building air conditioning systems. Spherical ‘Thermal energy storage capsules of different diameters and materials are tested. The aqueous solution of 35.97% ethylene alyeol is used as the heat transfer fluid (HITF). The major studied parameters are the size and mate- tial of the spherical capsule, the volume flow rate and temperature ofthe heat transfer uid (HTF). ‘The effets ofthese parameters on the time for complete charging/ discharging, the solidiiedjmelted Freezing and melting Abbreviations Act, actual; CTES, cool thermal energy storage; ERR, nergy recovery ratio; HT, heat transfer fluid; PCM, phase change material; rot, rolameter; (6, thermocouple TES, thermal energy storage "Corresponding author. Tel +20 111150706 fax: +20 242158803, Esmail address: reda_{elghnam@yahoo.com (RI. EiGhnany, ramadanamer20Q4@yahoocom (RA. Abécler7), mohamed sakr@d Feng buedu.eg (MAH. Saks), hany_assawy200%4 yahoo.com (HE. AAbieichman), 2090-4479 © 2011 Ain Shams University, Production and hosting by sevier BY. All rights reserved Peer review under responsibility of Ain Shams University o:10.1016).ae}.2011-10.004M RL ElGhnam etal sass fraction, the charging|discharging rate, the energy stored/tegain, andthe energy recovery ratio (ERR) are studied. The experimental results show that the energy recovery ratio is Becoming better ‘when using metalic capsules, increasing the capsule size and reducing the HTF volume flow rates. ‘© 2011 Ain Shams Univesity. Production and hosting by Elsevier BLY. All rights reserved Nomenclature C— speciic heat of ice, K/h K CL Specific eat of wate, klk K dq inside diameter of test capsule, m Fo Fourier number, Fo = 4; 1/12 k, thermal conductivity of ice, kW/m K LH, latent heat of fusion of water, Klik ‘mm melted mass, ke ‘mz mass of PCM encapsulated inside the spherical capsule, ke solidified mass, ke smelted mass fraction solidified mass fraction charging rate, kW discharging rate, kW accumulative energy regained, ki accumulative energy stored, KI T, solid phase temperature of POM, °C Ty, thermocouple temperature reading, °C 1’ dimensionless temperature, T™ = Te/ Ty ' time, s complete charging time, s complete discharging time, s time difference, s ‘Melted volume, m? ¥, solidified volume, m* Vise spherical shell volume, m? Vier actual volume flow rate, Lpm Poy volume flow rate obtained from the rotameter reading, Lpm ¥, volume of the PCM encapsulated inside the spher- ‘cal capsule (equal to 80% of internal volume of Quimax taxitum eneray stored at the complete charging the capsule) m? time, Kd Over maximum energy segained at the complete charge Greek symbols ing time, Kd Pe water density, kglen? om, average tadius of sofidtiquid interface in testcap- py ve densily, kin sule, Pale volume correction factor fy horizontal radius of solid-liquid interface in test 3) thermal diffusivity of ice (= hi/p,C)), ms capsule, m £ dimensionless time, (¢= Fy = jra) fo ie radius of test capsule, m Coanncr dimensionless complete charging time ry vertical radius of solid-liquid interface in test cap- tanya, dimensionless complete discharging, ime sule, 2 thermal diffusivity of ice Tug actual temperature, °C ‘AQne accumulative thermal energy regained difference, T,_ temperature of PCM at the center of capsule, °C uw T, liquid phase temperature of PCM, °C AQ accumulative thermal energy stored difference, KI T, inital temperature of PCM (distilled water),“C Ar” time difference, s Tye phase change temperature, °C 1. Inteoduetion to derive empirical equations deseribing charge and discharge ‘Studying the thermal behavior during phase change in spheri- cal capsules is extremely important for the desiga of efficient storage systems. Some experimental and theoretical studies investigations concerning the cool thermal energy storage (CTES) of encapsulated type are found in literature Eames and Adref{I] conducted an experimental study of the freezing and melting processes for water contained in spherical clements. They reported quantitative data on the movement of the solid-liquid interface position with time, the effect of HTF (coolant) temperature, and the effeet of sphere size on the melt- ing and freezing processes. They also reported the discharge and charge rates and the time required to melt and freeze a spherical ice storage element, Finally, their results were used for an ice storage element. Yoon etal. [2] studied experimentally the freezing phenom- enon of saturated water with the supercooled region in a hor- ‘ontal circular eylinder. From the experiments, it was found that there were three types of freezing patterns. The first was the annular ice layer growing from the eylinder surface at a high cooling rate, the second was the asymmetric ice layer at an intermediate cooling rate, and the last was the instanta- neous ice layer growing over the whole region at a low cooling rate Ismail etal [3] presented the results of a numerical study on the beat transfer during the process of solidification of water inside a spherical capsule under convective boundary condi- tions. The numerical solution was based upon the finiteAn experimental study of freezing and melting difference approach and the moving grid scheme. They also validated the numerical predictions by comparison with exper= {mental results realized by the authoss. The size ofthe spherical capsule, wall material, external bath temperature and initial, temperature of water were investigated and their effects on the solidified mass fraction and the time for complete soliifi- cation were presented and discussed Sakr etal [4] conducted experimental and theoretical study oon freezing and melting in eapsules with different configura tions. They used water as a phase change material (PCM). ‘The PCM was encapsulated in five different copper capsules (sphere, eylinder, pyramid, cone, and cuboids) having the same internal volume. The effect of geometrical configuration on the mm of the freezing and melting processes was investigated. The spherical capsule showed the best thermal en- ergy storage performance among the ive test configurations, Tan etal. [5] reported an experimental and computational investigation directed al understanding the role of buoyancy: driven convection during constrained melting of phase change ‘material (PCM) inside a spherical capsule, The melting phase front and melting fraction of the PCM were analyzed and com= pared with numerical solution obtained from the CFD code Fluent. They observed expedited phase change in the top Te tion of the sphere and a wavy surface at the bottom of the PCM after a short period of symmetric melting due to promi- renee of diffusion, Cheralathan etal. [6] investigated the transient behavior of | pphase change material based cool thermal energy storage (CTES). The system discussed was comprised of a cylindrical storage tank filled with encapsulated phase change materials (PCMs) in spherical container integrated with an ethylene gly ol chiller plant. A simulation program was developed to eval- uate the temperature history of the heat transfer fuid (HF) and the phase change material at any axial location during, the charging period. The model was also used to investigate the effect of porosity, Stanton number, Stelan number and Peclet_ number on CTES system performance. The results showed that increase in porosity contributes to a higher rate of eneray storage. However, for a given geometry and heat transfer coefficient, the mass of PCM charged in the unit ‘was found to decrease as the increase in porosity Kousksou et al. [7] investigated numerically the influences of the position of the storage tank and the flow pattern inside the tank for a cylindrical tank randomly filled with PCM encapsulated in spherical capsules. They used two-dimensional porous-medium modeling in two configurations of storage: vertical configuration where the stream lines of natural convee- tion are in the same direction as the Forced convection and the horizontal configuration where the natural convection is pre ponderant on that forced coavection They reported that the optimum running of the charge mode is obtained in the case of a vertical position where the motions due to the natural con- vveetion are in the same direction as the forced convection. Ismail and Henriquez [8] presented » numerical study of the solidification of PCM (distilled water) enclosed in a spherical shell. The mathematical model was based on pure conduction in the PCM subject to boundary conditions of constant temper- ature or convection leat transfer on the external surface of the spherical shel. The model was then used to predict the effect of the size ofthe spherical shell shell thickness, shell material, ini- tial PCM temperature and the external wall temperature on the solidified mass fraction and the time for complete solidification, character Kalaiselvam et al. [9] conducted analytical solutions for solidification and melting process of different PCM encapsu- lated inside cylindrical enclosures. Their study aimed to find the interface locations at various time steps. Transient inter face positions and complete phase change time were predicted by two models for solidification and by three models for melt- ing. Their analytical model was used to study the phase change behavior and heat transfer characteristics inside PCM. Pres- fence of heat generation increased the total solidification time of the cylinder, though it accelerated melting. Total solidiica- tion time depended on Stefan Number and heat generation parameter, whereas complete melting time depended on equiv= alent thermal conductivity Similar investigations were reported by many other authors, eg, Khodadadi, and Zhang {10}, Wu et al. [11], MacPhee and Dincer [12], Erek and Dincer [13], MacPhee and Dincer [14], Ananthansrayanan el al. [15], Beasley and Ramanarayanan [16], Sozen etal. [17] and Ismail and Stuginsky (18) 2. Description of experimental test rig ‘The main target of the current study is to investigate experi- ‘mentally the effect of the size and material ofthe spherial eap- sule (storage element), the volume flow rate and temperature of heat transfer fluid (HTF) on the time for complete charg- ingidischarging, the solidifiedjmelted mass fraction, the per~ centage of energy stored/regained, and the energy recovery ratio (ERR). To accomplish the above needs, a test rig is de signed and manufactured to carry out the heat transfer exper iments using the stitable instrumentation, The test rig used in the present study is shown schematically in Fig. 1a, It consists of charging tank, discharging tank, spherical capsule (test section), refrigeration unit, heating, section, heat transfer fuid circulation unit (centrifugal pump, piping system, gate valves), and suitable measuring instrumen- tation, The test rig components will be described in details through the following sections, The heat transfer fuid (HTF) is selected to be aqueous ethylene glycol solution which has 4 concentration of 35% by weight, The chosen concentration has a freezing point of ~19 °C, [19]. 21. Charging and discharging tanks ‘The charging and discharging tanks have the same dimensions. ‘They are galvanized steel tanks having dimensions of 350% 350 x 450 mm, and painted internally by chemical coat. ing (epoxy). To reduce the heat gain; the tanks are insulated by a 50 mm thickness of injected expanded polyurethane ther- mal insulation, Each tank is illed with a 0.038 m* of the HTP. ‘The two tanks are provided with inlet and outlet pipes includ- ing manual gate valves for the HTF circulation. Also, they are provided with a movable insulated door at their tops, 22, Spherical capsule (test section) In the present study the spherical capsules represent the ther ‘mal storage test clements which made from different materials (copper, brass, stainless stecl, plass, and plastic) and having different internal diameters (0.042, 0.07, 0.092, 0.10, and 0.11) with the same thickness (0.001 m) are employed to carry36 RL ElGhnam etal 5. Condese (ered arcoled 8) Figure ta ‘out the experiments. ‘The capsule materials are selected to introduce a wide range of thermal conductivity. Distilled water is used as the phase change material (PCM). Each one of the spherical capsules filed by 80% of its inner volume with dis tilled water as a PCM to avoid the thermal expansion during the solidification process, Fifteen calibrated copper-constantan, (T-type) thermocouples are employed to measure the tempera ture distribution, Thirteen of them are distributed on the hor- izontal and vertical axes of the spherical capsule at specified distances, as shown in Fig. 1b, and the other two thermocou- ples are used to measure the HTTF temperature around the spherical capsule, 2.3. Refrigeration unit and electric heating section The reftigeration unit is a simple vapor-compression cycle, operated with R-22. It includes a hermetic compressor @hp), ar cooled condenser (forced type), filter, drier, thermo- static expansion valve, and an evaporator (tube-in-tube type) The main purpose of the refrigeration unit is to cool the HT Line disgram of the test rg 10 a specified temperature to achieve the requirements of the charging process. The main purpose of the electric heating section is to heat the HTF to a specified temperature, to achieve the require ‘ments of the discharging process. The heating section is a cop- per pipe with an outer diameter of 0.0218 m, It is externally ‘wrapped with an electric resistance (200 W heating capacity), then electrically and thermally insulated. The electric resistance is connected to AC voltage regulator (0-220 V), to enable elec- tric power control ‘A digital temperature controller (electronic thermostat) is employed to control the required temperature inside the dis. charging tank and charging tank with a set-point differential of #1°C, 24. HITE circulation unit and piping system A centrifugal pump is used to circulate the HTF through the piping system to carry out the charging and discharging exper- iments. The piping system and the manual gate valves areAn experimental study of freezing and melting 7 se ape ~ ruse ose cece (lea wae < Treenacouples postions re th Test theemocouples, deta section showing distribution of ‘Table 1_Experimental parameters HTP parameters “Temperature (uring charge process) Temperature (Gucing discharge proces) Volume flow rate 10, and -12 6,8, 10,12, and 14°C 1,3,4,5, and 7 Lpm Spherical capsule parameters Size (internal diameter) Material 0.042, 007, 0.092, 0.11, und 0.126 (Copper, bras, suiless ste, las, and plastic arranged to enable the same pump to cireulate the heat trans- fer fluid through the charging and discharging cycles. The dis- charge pipe connected to the centrifugal pump is branched into two parallel lines; one of them to the heating seetion and the other to the evaporator of the refrigeration unit. The two branches are integrated with manual gate valves at inlet and ooulet ofthe healing section and the evaporator. The common pipe line afer the heating section and the evaporator is sepa- rated into two lines; one of them is directed to the rolameter and then to the charging or discharging tank and the other is bypassed to the suetion line of the pump to achieve volume flow rate control 2.5. Measuring instruments Suitable measuring instruments are used for monitoring, temperatures and ITTF volume flow rates, A data acquisition card (National Instruments, NI USB-6210, 16-inpuls, resolu- tion of 16-bit and scanning rate of 250 X8/s) and a laptop are used (0 monitor the temperatures through fiften calibrated, copper-constantan (T-type) thermocouples. The ITTF volume flow rate is monitored by using a calibrated rotameter. To con- tol a certain low rate as an experimental parameter; a manual, gate valve and a bypass pipe fited with another manual gate Valve are installed before the rotameter. A digital temperature controller (electronic thermostat) is used to control the required temperature inside the charging or discharging tank during the charging and discharging experiments, respectively. The speci- fications of the digital temperature controller are (ELIWELL. IC 901, 0.5% accuracy, and 1 °C set-point differential). A dig- ital AC vollage regulator is used to adjust the input power Lo the heating seetion. 3. Experimental procedures ‘A series of charging and discharging experiments are per- formed under various parameters. ‘Tae varied parameters in the present study aze listed in Table 1. The charging and discharging experiments are conducted, respectively and their experimental procedures arc discussed in the following two sections. The manval gate valves incorporated with the piping sys- tem are positioned for the charging mode, and the relrigera tion unit operates to cool the circulating HTF. The digital temperature controllers set at one of the five test temperatures ‘mentioned in Table 1. The PCM temperature is maintained at 25°C as an initial PCM temperature for all charging experi ments. Once the adjusted HTF temperature reaches, the volume flow rate of the HTF is set at one of the five volume flow rates mentioned in Table 1. Measurements of the PCM. temperatures inside the test capsule and the HTF temperatures around it are scanned and recorded every two seconds by the data acquisition system. The experiment is terminated when the temperature of the PCM inside the test capsule starts to be equal t0 that of the HTF. This indicates that the water is frozen completely and the ice is being sensibly sub-cooled. ‘The reffigeration unit and the pump are switched off, and the frozen test capsule is kept inside the charging tank to main- lain its temperature until finishing the preparation of the dis charging experiment (which takes few minutes) Prior to discharging experiments, the manual gate valves on the piping system are positioned for the discharging mode. The digital thermostat is set at one of the five test temperatures mentioned in Table I. The pump is switched on and the reig- eration unit operates only during the preparation to cool the circulating HTF. Once the adjusted HTP temperature is reached, both of the refrigeration unit and the pump are switched off. The next step is to close the manual gate valves, of the evaporator and open those of the healing section, The pamp and the heating section are switehed on, The heating section is controlled by the digital thermostat. The PCM tem= perature of the frozen spherical capsule is fixed at —8 °C us an initial PCM temperature for all discharging experiments, The volume flow rate of the HTF is set at one of the five volume ‘The next step is to run the ata acquisition software and transmit rapidly the spherical ‘capsule from the charging tank to the discharging one, Measurements of the PCM. temperatures inside the spherical flow rates mentioned in Table38 RL ElGhnam etal Figure 2a solidified volume of the PCM. capsule and the HTF temperatures around it are scanned and recorded every 2 by the data acquisition system. The experi- ‘ment is terminated when the temperature of the PCM inside the spherical eapsule begins to approach that of the HTE. This indicates that the ice is melted completely and the water is being sensibly super-heated. The electric heater and the pump are then switched of 4. Data reduction ‘Through all the charging experiments, the test sphere i filled by 80% of its internal volume wit PCM (distilled water) and care is taken to ensure that the test capsules immersed in the middle (of the charging tank. Also, the HTF flows upward around the lest capsule, Consequently, a symmetric solidification around the vertical axis of the test capsule is assumed, Fig. 2a shows 4 sketch of the solidified volume of the PCM formed inside the test spherical capsule. This figure clarifies the symmetric solidification around the vertical axis of the test eapsule, but it is clear that the development of ice is different in both the horizontal and vertical axes, So, one may simplify the solidified volume by taking the arithmetic average of the horizontal and vertical radi of the solid-tiquid interface as shown in Fig, 25 According to Fig. 2b, the spherical shell volume is ealeulated according to the following formula: Vana ~ (213), )+ AMP ~ Bag, ) where Vijei is the spherical shell volume m’, r,, is the internal radivs ofthe es spherical easule mri he orizontal rads of solid-quid interface in test eapsle m, ris the vertical radius of solid-liguid interface in lest capsule m, rays, 38 the arithmetic average radius (=) m, and his the vertical dis. tance, measured from the center of the spherical capsule to the free surface of the encapsulated PCM m, 4.1. Solidfied mass m, and melted mass My ‘The solidified mass of the PCM is ealeulated from: m= ole/P Ves land the melled mass of the PCM is ealeulated from: My = PuV st Figure 2 Sketch of the solditied volume of the PCM efter simplification, 4.2. Accumulative stored andregained thermal energy Ou & Qree ‘The accumulative thermal energy stored within the test capsule is derived as follows: ox narfeur—r+ (Burs Bearer and the accumulative thermal energy regained from the melted PCM formed within the test spherical capsule is calculated from following easton (e)een 13} ard 43. Charging rate Qu and discharging rate Qas ng = The charging rate of thermal energy stored within the test cap- sule is calculated from: AQ, On == ‘The discharging rate of thermal energy regained from the test capsule is calculated from; 44. Energy recovery ratio (ERR) Qeafteon: ERR = Qee!tomnes Qu Fmp 4.3. Dimensionless temperature (1 o ") and dimensionless time Dimensionless temperature (1°) is defined as the ratio of the temperature of PCM at the center of spherical capsule to the ‘Phase change temperature of PCM. It is computed from the following equation:An experimental study of freezing and melting » Dimensionless time (c) is expressed as Fourier number (Fo). Conceptually iis the ratio of the heat conduction rate to the rate of thermal energy storage. It is defined as: teFos tt 46. Solidfied mass fraction and melted mass fraction Solidified mass fraction is defined as the ratio ofthe solidified ‘mass to the mass of the PCM encapsulated inside the spherical capsule. Thus, the solidified mass fraction is calculated from the following formula: br Solited mass fracGon =P = BF Melted mass fraction is defined as the ratio of the melted ‘mass to the mass of the PCM encapsulated inside the spherical capsule, Accordingly, the melted mass fraction is computed from the following formula: Melted mass fraction = = 4.7. Percentage of energy stored and percentage of energy regained ‘The percentage of energy stored within a test capsule is defined as the ratio of accumulative energy stored at any charging time to the maximum energy stored at the complete charging time, namely; Qu Zee» 100 Ynergy stored ‘The percentage of energy regained from a test capsule is de- fined as the ratio of accumulative energy regained at any di charging time to the maximum enetgy regained at the ‘complete discharging time, namely; 100 ‘8. Results and discussion ‘Through the present chapter, the experimental results concern- ing the effect of eapsule material, capsule size, HTF tempera- lure, and the volume low rate of HTF on the dimensionless lime for complete charging/discharging, the solidified melted ‘mass fraction, the percentage of energy stored/reguined, and the chargingldischarging rate are analyzed, compared and discussed. 5.1. Charging process results S.l.L. Effect of capsule material ‘The effect of the material of the spherical capsisle on the dimensionless time for complete charging is shown in Fig. 3 ‘The tested materials are copper, brass, stainless steel, glass, igure 3 Effet of the capsule material on the dimensionless time for complete charging. go* § = i ESSE Dimensionless time, Soliiied mass fraction igure 4 Variation of the solidified mass fraction with the dimensionless lime using different capsule materials during the charging process. ‘and plastic, It is indicated that for the same operating condi- tions, the metallic capsules introduced lower dimensionless time for complete charging, especially the copper capsule Which took the lowest dimensionless time. Also, itis interesting, to notice that the difference between the times for complete charging when using either metallic or non metallic capsules is relatively small. This small difference makes the use of the non metallic capsules is more economical. For further explor~ ing, Fig. 4 presents the effect of the capsule material on the ‘variation of the solidified mass fraction with the dimensionless time, «, during the charging process while keeping the rest of parameters unchanged. It is noticed that for al test materials, the solidified mass fraction increases as the dimensionless time increases. Also, for a given dimensionless time, the magnitude ‘of the solidified mass fraction increases by increasing the ther- mal conductivity of the capsule material. For that aspect, the copper capsule introduces the highest solidified mass fraction, ‘while the nonmetallic capsules produce the lowest one. It is also noticed that for all capsule materials the rate of increase of the mass fraction at the beginning is faster than that at the ead of the solidification process. This can be attributed to the fact that the progressive formed ice layer acts as an insu- lating material for the rest of the contained water. Fig. 5 indicates the effect of the capsule material on the variation of the pereentage of energy stored with the dimen- sionless time during the charging process, while keeping the ‘other parameters unchanged. I is realized that the percentage0 RL ElGhnam etal a , 2° 22 § == Esl 8 bam 2 eS 10 rerstimetones 7m | pi dimensionless time using different capsule materials during the charging process. Dimensionless time, € igure 6 Variation of the charging rate with the dimensionless lime using different capsule materials during the charging process. sw i [omar > Boe . Fo ° eae i” a Sm Capsule internal diameter, igure 7 charging Effect of the capsule size on the time For complete of energy increases as dimensionless time increases. It is also noticed that for a given dimensionless time, copper and brass capsules store more energy rather than stainless steel and non metallic capsules. This is due to their better thermal conductivity and accordingly they took relatively short dimen- sionless times for complete charging. Moreover, Fig. 6 shows the variation of the charging rate (energy stored rate) with the dimensionless time during the charging process. The results, Of the different test capsule materials are presented for com= parison. It is shown that at the beginning of the charging Bes] Gy 0 & Ltn Slo, 2 eee Ew So] ane = el Je nm 2» nl es pr on 10] charg aoc | HF erg Es Charging time, in Figure 9 Variation ofthe percentage of energy stored withthe charging time using different capsule sizes. process, the charging rate of the metalic capsules is higher ‘than that of the nonmetallic ones, After that, the charging rate of all capsules is rapidly decreased and their values are coinci- deat. This behavior is related to the decrease in the tempera ture difference between the PCM and HTF 5.1.2. Effect of capsule siz In the present study, the glass capsule is chosen to investigate the effect of capsule size on the characteristics of the charging and the discharging processes. Fig. 7 shows the variation ofthe capsule internal diameter with the charging time. The results, ensure the increase of the charging time with increasing the capsule size. Moreover, Fig. 8 shows the variation of the solid iffed mass fraction with the charging time. The results are presented for comparison. Its noticed that; the solidified mass, fraction increases with increasing of the charging time for all capsule sizes, Also, itis realized that fora given charging time, the solidified mass fraction of smaller capsule size is greater than that of bigger one. Furthermore, Fig, 9 clarifies the variation of the percentage ‘of energy stored within a capsule with the charging time, The data ofall capsule sizes are compared. Similar trend is observed for all capsule sizes. Its noticed that for a given capsule size, increasing the charging time; increases the percentage of energy stored, Also, for a given charging time, increasing the capsule size; increases the percentage of energy stored. On the other hhand, one can observe from Fig. 10 that as the charging time is increased, the curves of charging rate of all capsule sizes seemAn experimental study of freezing and melting 41 080 fs oom [o* wr tocastn Abas #8 Bb xy x Ce etgtingi Charging rate KW Figure 10. Variation of the charging rate with the charging time using different capsule sizes. LITE volume flow rate, pm igure 11 Effect of the HITF volume flow rate on th sionless time for complete charging to be loser together. Also, itis indicated thatthe higher charg- ing rate is associated with the larger capsule size. For all capsule sizes, increasing the charging time; decreases the charging rate, This behavior is attributed to the Fact that as the charging time increases, the temperature difference between the PCM and HTF is decreased, 5.1.3. Effect of HTF volume flow rate ‘The glass capsule is selected to investigute the effect of HTF volume flow rate on the characteristics of the charging and the discharging processes. Fig. 11 shows the variation of the dimensionless time for complete charging with the HTF vol- ume flow rate. It is observed that inereasing the HTF volume flow rate; decreases the dimensionless time for complete charg- ing, The effect of the HTF volume flow rate on the variation of the solidified mass fr ‘onless time during the charging process is shown in Fig. 12. Iti indicated that for all HTF volume flow rates, increasing the dimensionless time increases the solidified mass fraction till reaching the time for, ‘complete solidification (the solidified mass fraction equals to unity). Also, itis observed that for a given dimensionless time, the solidified mass fraction inereases by increasing the HTP) volume flow rate, Accordingly, the higher the HTF volume flow rate, the shorter the dimensionless time for complete charging, ae wer Bu af : == ne = aul ed i ome Dimensionless time, ¢ Figure 12 Variation of the solidified mass fraction with the dimensionless time using different HITP volume flow rates during the charging proces i a! He if = Be i Baw ea} @ am » eae” ft Woe FT siete. ‘ Fie 13 Yarnton othe pcs of nergy sored wth he dimensionless time using different HTF volume flow rates during the charging proces. om ewh eect |S se ¢ soe une Bond un Eel a S om. 1 oa a Serco 8 Eo Figure 4 Varion othe rig rat with the dimensiones time using different HTP volume fow rates during the charging process. To further explore, Fig. 13 depicts the effect of the HTF volume flow rate on the relation between the percentage of energy stored and the dimensionless time during the charging process, Different volume flow rates introduce similar trend, [is shown that for a given HITF volume flow rate increasing. the dimensionless time; increases the percentage of energy stored. Also, for a given dimensionless time, higher HTF vole lume flow rate introduces higher percentage of energy stored. ItLITE temperatore, °C Figure 15. Effect of the HTF temperature on the dimensionless time for complete charging. oftha® 5 0° Soldiied mass fraction 7 os 0 2 ww us Dimensionless time, € Figure 16 Variation of the solidified mass fraction with the dimensionless time using diferent HTF temperatures during the charging process. ccan be observed from Fig. 14 that for a given HITE volume flow rate, as the dimensionless time increases, the charging rate fof energy stored decreases, Also, the curves of charging rate with dimensionless time are coincident. Also, it is observed that at the beginning of charging process, higher charging rate is associated with higher HT volume flow rate. SIA. Effect of HTF temperature ‘Through the present study, the glass capsule is selected to investigate the effect of HTF temperature on the characteris. ties of the charging and the discharging processes. The temperature of HTF is varied from -4°C to =12°C. The HTF volume flow rate is kept at 7 Lpm, the internal capsule diameter is kept at 0.1 m and the initial PCM temperature is, ‘maintained at 10°C. Fig. 15 depicts the variation of the HITF temperature with the dimensionless time for complete charging. Is can be seen that the decrease of the HTF temper- ature from 4 °C to ~12°C reduces greatly the dimensionless time for complete charging. Furthermore, the results presented in Fig. 16 ensure those of Fig. 15 in the sense that lowering the HITF temperature; decreases the dimensionless time for complete solidification (the time for solidific to be equal 10 unity), One can observe from Fig. 17 that as the ITF temperature decreases, the percentage of energy stored increases. Conse- ‘mass fraction RL ElGhnam etal ‘Charging proces Energy stored Dimensionless time, € Figure 17 dimensionless time using diferent HTF temperatures during the Variation ofthe percentage of energy stored with the carping process. on Zoo fw © Bele ke 5 aor} x%, oa LOB RAR gan nxxxcxxn ioe es Figure 18 Variation of the charging rate with the dimensionless time using differen proces. HITE temperatures during the charging quently, the lower the HTTF temperature, the shorter the dimensionless time for complete charging. This agrees with the previous results. Moreover, Fig. 18 shows the effect of the HTF temperature on the variation of the charging rate (rate of energy stored) with the dimensionless time. It is seen that the charging rate is rapidly decreased by increasing the dimensionless time, The data for all HTF temperatures have the same trend and are closer together. 5.2. Discharging process results 5.211. Effect of capsule material Fig. 19 presents the effect of the eapsule material on the dimen- sSionless time for complete discharging. The rest of parameters, are kept unchanged. It is shown that the metalic capsules, introduce lower dimensionless time for complete discharging (On the other hand, Fig. 20 shows the variation of the melted ‘mass fraction with the dimensionless time during a discharging process. The data of different test capsule materials are pre- sented for comparison. Itis seen that as the dimensionless time increases, the melted mass fraction ine that the increase in the melted mass Fraction for metallic eap- es. One can observe sules is higher than that of nonmetallic ones. For that aspect, the dimensionless time for complete discharging for metallic capsules is shorter than that of nonmetallic ones. Also, itis ob-‘An experimental study of freezing and meting 43 to] Gate 1 ee ere xe See sie oon gone), Ltn in wn B omste ° = de sm ele g ss Foo § Bo 2 ® & se B oms g Poe. “ DPlastic Glass Salles steel OBrass Copper ~—e ‘ 2 s ‘ s . 7 . Dimension tne, Figure 19 kifct of the capsule material on the dimensionless tie for complete decrying Figure 22. Variation of the discharging rate with the dimension i Netra 2 8 i i i a 3 i a . o a Discharging process Dimensionless tne, we 20 Variation of the melted mass fraction with the dimensionless time using different capsule materials during the discharging process 1 2 yee ie. 8 Bs af ba ae a? [ese ee _ ». . ‘ Discharging process | Pune a fs a 5s «© 9 8 Dimensions in, ie 21_Variaton ofthe perentage of ensrey the dimensionless time using die discharging process, ren capsule mater served that the whole capsule materials have the same trend in the sense that at the beginning of the discharging process, the ‘melted mass fraction increases slowly with time compared to that at the end of the melting process. This can be attributed, to the fact that at the beginning, the formed melted layer be- tween the inner surface of the capsule and the remaining ice is small, Accordingly, the heal transfer by conduction is the dominant, while the Convective heat transfer is not effective. As time goes, the gap of melting layer incre «quently, the heat transfer by convection is enhanced, that lead to increase the melting rate ‘and conse: less time using different capsule materials during the discharging roves, zs 2 ioe] Caran pcen &, 400- ° 3m . 3 aso . zie © remeroT ES HIE volume ftw ate <7 pon ‘apsole internal diameter; Figure 23 discharging, Etfect of the capsule size on the time for complete To further expfore, Fig. 21 depicts the variation of the per~ centage of energy regained with the dimensionless time during the discharging process, The results ofthe differen test capsule ‘materials are presented for comparison. One can observe that the increasing of the dimensionless time increases the percent= age of energy regained. This is true for all capsule material, ‘Also itis realized that the nonmetallic capsules consumes its, stored energy in a relatively longer time rather than the metal- lic ones. Fig. 22 presents the variation of the discharging rate with the dimensionless time during the discharging process. The data of all test capsule materials are compared. The results of the capsule materials have the same trend. Its notived that at the beginning of discharging process, high discharging rate is happened, which can be attributed to the high temperature difference between the HTF and PCM. After that, the dis: charging rate decreases suddenly due to the fast decrease in temperature difference during the sensible heating of ice till reaching the phase change temperature. As mentioned earlier, increasing the melting layer by passing the time enhances the heat transfer by convection. So, the discharging rate increases, ‘eradually tll Gnishing the melting process. Finally; the dis- charge rate decreases as a result of the decrease in the temper- ature difference between the HTF and PCM. Also, it is observed that the discharging rate of nonmetallic capsules, takes longer time than those of metallic ones.RL ElGhnam etal o 8 6 «Le re] ee Re & ee ao le ue ans Jeppesen = x Je east t ° ” HTF volume flow rate, Lpm. Figure 24 Vasaon of the meld mass fraction with the discharging time using different capsule sizes. Figure 27 Effect of the HTF volume flow rate ot 100 oo a oo y xX En a ox En] © . ocscng pre Foo ° = Ba] co ¢ “ah Boe] ote ox 2 eee ae wiqgog 8 ome oS) 16) 18200 280 00 aye Discharging ime, min Figure 25 Variation of the percentage of energy regained with, the discharging time using different capsule sizes. 008s & cow Dicbersing proves] Copan Gas fens gone a* FeoP? eat * vo poeta cots > 1 Wo ao x) 00 G0 a0 Go sto Discharging time, min Figure 26 Variation ofthe discharging rate with the discharging lime using diferent capsule sizes. 5.2.2. Bffeet of capsule size Its indicated from Fig. 23 that increasing the capsule size in- creases the time for complete discharging. Moreover, Fig. 24 indicates the effect of the capsule size on the variation of the ‘melted mass fraction with the discharging time. This figure reveals that for a given discharging time, the melted mass frac- tion is decreased by increasing the capsule size. Also, itis no- ticed that for a given capsule size, the melted mass fraction is, increased by increasing the discharging time. The effect of the capsule size on the relation between the percentage of energy regained and the discharging time is sonless time for eomplet discharging, Dimensionless ime, igure 28 Variation of the melted mass fraction with the dimensionless time using different HITF volume flow rates during the discharging proves. shown in Fig, 25. It is noticed that for a given capsule size, the melted mass fraction is increased by increasing the dis- charging time. Also, it is observed that for a given discharging time, the larger the capsule size, the lower the percentage energy repained. Accordingly, the larger the capsule size, the longer the time for complete discharging. Fig. 26 reveals the relation between the discharge rate and the discharging time The data for all capsule sizes are presented for comparison, It is observed that the data forall capsule sizes have the same ‘end in the sense as discussed earlier (Section 5.2.1). Also, itis, shown that the largest capsule size introduces the longest dis. charging time. 5.23. Effect of HTF volume flow rate Icis realized from Fig. 27 that inereasing the ITTF volume flow rate, decreases the time for complete discharging. Further- ‘more, Fig. 28 reveals the variation of the melted mass fraction, with the dimensionless time during the discharging process ‘The data of all tested HTF volume flow rates are compared, All the date introduced the same trend. It is shown that increasing the dimensionless time, inereases the melted mass Fraction. Ibis also observed that for a given dimensionless time, the melted mass fraction is increased by increasing the HTF volume flow rate The tween the percentage of energy regained and the dimensionless time during the discharging process is shown in Fig. 29. It is ject of the HTF volume flow rate on the relation be-An experimental study of freezing and melting Dimensionless time, ¢ Figure 29. Variation of the percentage of energy regained with the dimensionless time using different HTF volume flow raves uring a discharging process 2 een # ono B ons 5 in] gaa * amo Figure 30 Variation of the discharging rue with the dimension- les time using diferent HIT volume flow rales dunn hc discharging process revealed that for a given dimensionless time, increasing the HIF volume flow rate increases the percentage of energy re ‘gained. Also, for a given HTT volume flow rate, the percentage of energy regained is inereased by increusing the dimensionless time. Fig. 30 reveals the variation of the discharging rate with ‘dimensionless time during the discharging process, The results of the HTF volume low rates are compared. These re sults ensure that the discharging rate is decreased by increasing. the dimensionless time. Also, it is noticed that the results of all the HTP volume flow rates have the same trend, The for this behavior is discussed in Section 5.2.1 5.24. Effect of HTF temperature Fig. 31 presents the effect of the HTF temperature on the dimensionless time for complete discharging. It is indicated that decreasing the temperature of the HTF, increases the dimensionless time for complete discharging. On the other side, Fig. 22 depicts the relation between the melted mass Fraction and the dimensionless time during the discharging process. The data for all tested HTF temperatures are com= pared. It is shown that for a given HTF temperature, the ‘melted mass fraction is increased by increasing the dimension- less time. The results of all est HTF temperatures have the same trend. Also, itis observed that for a given dimensionless time, the melted mass fraction is increased by inereasing the ITF temperature 45 fe HITE temperstore, °C Figure 31 Effect of the HTE temperature on the dimensionless time for complete discharging Dimensions time, ¢ Figure 32° Variation of the melted mass fraction with the dimensionless time using different ITTF temperatures during the discharging process ER. rod Ee) Se [sre Bee br: “al ane 8 cen Dimensionless ime, € Figure 33 Variation of the percentage of energy regained with the dimensionless time using different HITF temperatures during Moreover, Fig. 33 reveals that the percentage of energy re- ‘gxined is increased by increasing the HTF temperature for a given dimensionless time, Also, for a given HTP temperature, the percentage of energy regained is increased by increasing the discharging time, One can observe from Fig. 34 that the dis- charging rate using a given HTF temperature is decreased by increasing the dimensionless time. In addition, for a given dimensionless time, the discharging rate is higher for high HTF temperatures. The reason for this behavior is related to the progressive convection effect by passing the time as men~ tioned earlier46 iL ElGhnam et al tase =F aa Came ce =e 195 | EEE nasose & coe: intrettapentse | os 6 A Eoowp © oa Hope 0° om dosfo 2 4 eS 50 ors E oa. aeoo lg xx* an wor] Bg SESH Boe xy on owe eas 7b. ata; hs Dead o8oT sean enn wean Dimensones tne, Intel dante st wheel oe, Figure 34. Variation of the discharging rate with the dimension Figure 36 Effet ofthe capsule size on ERR. les ime using different HTP temperatures during the discharging proces. Ears rm : ‘aca aha ox rvunpersu io" oy oss on oan a oes er [MTF volume ow ae, Lom Figure 38 Effect of the capsule material on ERR, 5.3. Energy recovery ratio ‘The energy recovery ratio (ERR) is defined as the ratio of the average energy regained rate to the average energy stored rate under the same operational conditions. The energy slorage performance is measured by the energy recovery ratio, the higher the energy recovery ratio, the better the eneray storage performance. Fig. 35 reveals the effect of capsule material on the ERR while the rest of parameters are unchanged. The data, for all test capsule materials are compared. It is noticed that, the copper capsule introduce the highest ERR. Also, the plas: tic eapsule introduce an acceptable ERR compared (0 those of higher thermal conductivities. Accordingly, the use of nonme- tallic capsules is encouraging for its economical aspects (cheap and easy to fabricate). On other side, Fig, 36 shows the effect, ‘of the eapstle size (capsule internal diameter) on the parameter ERR, Its indicated that the higher the capsule size, the higher ERR. Furthermore, Fig. 37 depicts the effect of HTF volume flow rate on the parameter ERR. It is observed that the lowest, HTF volume flow rate introduces the highest value of ERR. 6. Comparisons with the available published data Fig 38 shows a comparison between the present results and the ‘numerical results of Ismail etal [3] for the variation of solidified ‘mass fraction with the charging time, The comparison ensures reasonable agreement between the present results and those of [3]. The slight discrepancy between them i related to the differ- tence between the present operating conditions and those of 3) Figure 37 Effect of the HF volume flow rate o9 ERR, TE Rarer lowest i ge i ge |e Fuf ° esa Not . i EE) ‘Chugingume. igure 38 Comparison between the present experimental study and the results of Ismail et al. [3 7. Conclusions Based on the present results, the following significant features can be drawn 1. The time for complete charging is decreased by using metallic capsules, smaller capsule sizes, lower HTF tem- peratures and higher HTF volume flow rates, 2. The solidified mass fraction is increased by employing metallic capsules, using smaller capsule sizes, higher TITF volume flow rates and lower HITF temperaturesAn experimental study of freezing and melting a 3. The percentage of the energy stored is increased by using metallic capsules, bigger capsule sizes, higher HTF vol- lume rates and lower HTF temperatures, 4. The time for complete discharging is increased by using nonmetallic capsules, bigger capsule sizes, lower HTF temperatures and lower volume flow rates. 5. The melted mass fraction and the percentage of the energy regained are increased by employing metallic capsules, using smaller capsule sizes, higher HTF vol- ume flow rates and higher HTF temperatures. 6. The energy recovery ratio is becoming better, when using metallic capsules, increasing the capsule size and reducing the HTF volume flow rates. 7. The thermal conductivity of the capsule material has a relatively small effect oa the time for complete charging, time for complete discharging, solidified mass fraction, ‘melted mass fraction and the energy recovery ratio. 8, Although shorter time is consumed for the charging pro- cess using the metallic capsules, significant longer time is taken for the discharging process when using the nonme- rallic capsules, which lead to select the later type to enhance the thermal energy storage system. 9. Itis economical to use nonmetallic capsules (cheap and easy to fabricate), as the change in time for complete charging and discharging as well as ERR is small. 10. The gained results of the present research are confident as reasonable agreement is noticed when compared with the results of one of the literature. References [0] Eames W, Adret KT. Freezing and melting of water in spherical enclosures ofthe type used in thermal (le) storage systems. Appl ‘Therm Eng 20022273345, [2] Jung In Yoon, Choon Geun Moon, Eunpil Kim, Young Sook Son, Jae Dol Kim, Toyofumi Kato. Experimental study on Irevzing of water with supercooled repion ina horizontal einer, Appl Therm Eng 2001;21:657-68 [3] Ismail KAR, Hensiquez JR da Sve TM. A parametiestudy onice {formation insdea spherical capsule. IntJ Therm Sei2008:42:881-7, [a] Seke MH, Abde'-Aviz RM, Ghorab AB, Experimental and ‘theoretical study on freezing and melting in capsules Yor thermal storage. ERJ Soubra Faculty Eng 20084865. [5] Tan FL, Hosseinizadeh SF, Khodadadi IM, Fan Livy, Experi= mental and computational sudy of constrained melting of phase change materials (PCM) insie a spherical capsule, Int J Heat Mass Trans 2008:52:464-72, (6] Cheralathan M, Veira) R. Renganarayanan S. Heut transfer and parametric study of an encapsulated phase change material based cool thermal energy storage system. 3 Zhejiang Univ Sci A 2006; 7¢11:1886-95, [7] Kousksou T, Bodecorrets JP, Dumas JP, Mimet A. Dynamic modeling ofthe storage of an encapsulated ice tank. Appl Therm Eng 2008;25:1S34488, [8] Ismail KAR, Henrique JR. Solidification of PCM inside a spherical eapsule. Energy Convers Manage 2000;4:173-87. [9] Kalaiselvam 8, Veerappan M, Arul Aaron A. Iniyan S. Exper- ‘mental and analytical investigation of solidifieation and melting characteristics of PCMS inside eylindrieal encapsulation. Int J “Therm Sei 2008:47:858-74, (10) Khodadadi JM, Zhang Y. Effects of buoyancy-driven convection ‘on melting within spherical containers, Tat J Heat Mass Trans 2001; 44:1605-8, (11] Wo T, Liaw HC, Chen ¥Z. Thermal effect of surface tension on the inward solidifcaton of spheres. Int J Heat Mass Trans 12002:45:2055-65, [12] MacPce David, Dincer Ibrahim. Thermal modeling of 2 packed bed thermal energy storage system during charging. Appl Therm Eng 2009:29:695-705, 103] Erek Aytine, Dincer Ibrahim. Numerical heat transfer analysis of encapsulated ice thermal energy storage system with variable heat transfer coelicent in downstream. Int J Heal Mass Trans 2008;52:851. [14] MacPhee David, incor Thrahim. Performance assessment of some ice TES systems, Int 1 Therm Sei 2009x0412. [15] Ananthanarayanan V, Shai ¥, Mobley CE, Rapp RA. Modeling of fixed bed heat socape units uiizng phase-change materi ‘Metall Trans B 19871833946, (16 Beasley DE, Ramanarayanan C. Thermal response of a packed bed of spheres containing a phase change material. Int J Energy Res 1989;13.253-65, (07) Sozen M, VafaiK, Kennedy LA. Taermal charging and ischarging of seasible and latent heat storage packed beds ATAA J Thermophys 1991;5(4)623-5, {18} Ismail KAR, Stuginsky R. A parametcic study on possible xed ‘hod models for PCM and sensible heat storage. Appl Therm Eng 1999:19(9;757-8. (19) Properties of Working Fluids-Brines. M. CONDE Engineering, Zach; 2002 De. Reda 1, Highnam is currently Associate Professor of Mechanical Engineering (Power partment) at Shouba Faculty of engincer- jing, Benha University, Egypt. He has pub- lished about 19 papers im referred national and international journals and conference proceedings. His area of research is heat transfer, heat pipes and combustion, Dr. Ramdan A. Abdelaziz ie curently Profee- sor of Mechanical Engineering (Power ‘epactment) at Shoubea Faculty of engineer ing, Benha University, Exypt. His area of resarch is heat transfer, sat pipes and refrigeration and air condioning Dr. Mohamed H. Sake is curently Professor fof Mechanical Engineering (Power depart tent) at Shoubra Faculty of engineering, Benka University, Eaypt. His area of researc is heat transfer, heat pipes and refrigeration and sie conditioningRL ElGhnam etal Eng. Hany E, Abdelrhman is currently Asso Gate Lecturer of Mechanical Enginosring (ower department) at Shoubra Faculty of ‘engineering, enha. University, Laypt. His trea of reseazeh is heat transfer, het pipes and refigeration and sir conditioning.