ARTICLE IN PRESS

Solar Energy Materials & Solar Cells 94 (2010) 1038–1048

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells

0927-02

doi:10.1

n Corr

E-m

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

Experimental investigation of effects of supercooling on microencapsulatedphase-change material (MPCM) slurry thermal storage capacities

Shuo Zhang, Jianlei Niu n

Department of Building Service Engineering, The Hong Kong Polytechnic University, Kowloon, Hong Kong, PR China

a r t i c l e i n f o

Article history:

Received 19 August 2009

Accepted 3 February 2010Available online 24 February 2010

Keywords:

Experimental investigation

MPCM slurry

Thermal storage capacity

Supercooling

Effective latent heat

48/$ - see front matter & 2010 Elsevier B.V. A

016/j.solmat.2010.02.022

esponding author, Tel.: +852 27667781; fax:

ail address: bejlniu@polyu.edu.hk (J. Niu).

a b s t r a c t

Ice storage is currently the dominant cooling energy storage method. To more effectively utilize natural,

renewable cooling sources, such as evaporative cooling and sky-radiative cooling, diurnal storage media

operated on a daily basis at the temperate range between 10 and 20 1C are the most desirable. This

paper will present the experimental investigation of microencapsulated paraffin slurry as cooling

storage media for building cooling applications. The water slurry of microencapsulated n-hexadecane

with a melting temperature of 18 1C was cooled to 5 1C and heated to 25 1C cyclically in a storage tank of

230 l, and it was observed that full latent heat storage can only be realized at around 7 1C due to

supercooling, and the effective cooling storage capacity at the cooling temperature range between 5 and

18 1C are obtained, which can be used to realistically estimate cooling storage capacity with various

natural cooling schemes.

& 2010 Elsevier B.V. All rights reserved.

1. Introduction

The wide use of air-conditioning in commercial buildingsincreases the electricity peak demand, and makes it difficult totackle the global warming problem due to CO2 emissionsaccompanying electricity generation. Thermal energy storage asa means of managing energy demand and utilization is nowaccepted in many countries. Cool thermal storage systems havebeen intensively studied for their ability to use stored coolingproduced during off-peak hours to provide the energy necessaryfor daytime applications [1].

Phase change materials (PCMs) have long been used forthermal storage/control material because of the large amount ofheat absorption/release during the phase change processes, withonly small temperature variations. The earliest works basicallyconcentrated on analysing pure substances, and often water (forice stores) was used. With ice-slurries the phase transition fromice to water at 0 1C or vice versa is used to store or release largequantities of heat. However, from the energy efficiency point ofview, ice is not a very attractive heat storage material, becausemost applications do not require temperatures below or even nearthe freezing point of water. With the electricity-driven coolingstorage system, such low storage temperature will lower the COPof the chiller, which goes against the energy saving purpose. Inthis situation, alternative phase change materials with highermelting temperature, which is proper to HVAC(heating, ventilat-ing and air conditioning) applications are most desirable and havebeen investigated by several researchers [2,3]. For example,

ll rights reserved.

+852 27746146.

tetradecane (C14H30) has a melting temperature of 6 1C, hexade-cane (C16H34) 18 1C and octadecane (C18H38) about 28 1C. Bymixing different paraffin, melting points between those of pureparaffin can be obtained [4].

In recent years, a new approach was proposed, in which thePCM was microencapsulated and suspended in a single-phaseheat transfer fluid to form the solid–liquid MPCM slurry [5]. Dueto the microencapsulation of PCM particle by a thin plastic shell,the core material is always separated from the carrier fluid, whichmakes MPCM slurry behave like a liquid, while the latent heateffect associated with phase change significantly increases theheat capacity of the carrier fluid. In addition, the heat transfercoefficient between the fluid and heat transfer surface may begreatly increased because of the latent heat effect and the particleto particle interactions [5,6]. The flow and heat transfercharacteristics of MPCM have been experimentally and theoreti-cally investigated by a number of researchers [7,8]. The experi-mental results showed that the MPCM flow enhanced the localheat transfer coefficient relative to the single heat transfer fluidwith only small pressure drop increased; pressure drops wereeven lower than those for pure water at some conditions [8–10].To understand the mechanism of heat transfer enhancement ofMPCM flow, a theoretical model was developed for laminar MPCMflow in a circular duct with constant temperature or heat flux. Theresults predicted that the heat fluxes were about 2–4 times higherthan for single-phase fluids [11]. Modifications were made byaccounting for the microcapsules crust, initial subcooling and thedegree of phase change range, which brought predictions closer tothe experimental results [12,13].

One of the major problems associated with MPCM slurryas cooling storage medium is supercooling. In the previous

ARTICLE IN PRESS

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–1048 1039

experimental investigations, the empirical evidence is that thetwo TES system operating modes, i.e., charge and discharge, arenot symmetrical because of the supercooling phenomenon [14].Most of the crystallization processes of PCM are dominated by theheterogeneous nucleation mechanism, the phase-transitionbehavior of the PCM being complicated and very sensitive tosmall amounts of impurities [15]. At the fusion temperature, therate of nucleation is generally very low. To achieve a reasonablerate of nucleation, the solution has to be supercooled and energyshould be discharged at much lower temperature than the fusiontemperature. Supercooling of more than a few degrees willinterfere with proper heat extraction from the store, and 5–10 1Csupercooling can prevent it entirely [16]. The phenomenon thatthe latent heat can only be released at a supercooled, lowertemperature is disadvantageous for energy storage applications[17]. It is believed that organic materials typically undergocongruent melting, which means melting and freezing repeatedlywithout phase segregation and consequent degradation of theirlatent heat of fusion, and they crystallize with little or nosupercooling and usually non-corrosiveness. By contrast, mostsalt hydrates have poor nucleating properties resulting in super-cooling of the liquid salt hydrate prior to freezing [16].

For MPCM slurry to be successful in transferring heat in a costeffective manner, the supercooling phenomenon should be under-stood and controlled effectively, which has been studied using DSCand other methods in recent years [18–20]. Yamagishi et al. [8]studied the melting and crystallization processes of microencapsu-lated n-tetradecane and n-dodecane as a function of capsule size.The results indicate that the degree of supercooling or differencebetween melting and crystallization temperature points increaseswith decreasing particle diameter, specifically for microcapsules of100 mm in diameter or less. This effect limits the application ofmicroencapsulated n-alkanes. Yamagishi et al. [8] selected1-tetradecanol as a nucleating agent for prevention of supercoolingof microencapsulated n-tetradecane. In Zhang et al.’s [17] research,the DSC cooling curves of mC18 are mainly affected by the averagediameters. Adding 10.0 wt% of 1-octadecanol inside the microcap-sules as a nucleating agent decreases the degree of supercooling ofmicroencapsulated n-octadecane from 26 1C to approximately 12 1Cat a heating and cooling rate of 10.0 1C/min. Alvarado. et al. [18]presented an experimental study on characterization of super-cooling suppression of MPCM slurry. They studied the keyparameters influencing MPCM slurry supercooling, and indicatedthat the initiation of freezing point is inversely proportional tocooling rate. Choi et al. [20] experimentally studied the heat transfercharacteristics of a low temperature latent heat storage system withcircular finned and unfinned tubes using sodium acetate trihydrateas the phase change material. They pointed out that, in the heatrecovery stage, supercooling of PCM in the finned-tube system islarger than that in the unfinned-tube system. Moreover, the heat-transfer coefficient in the thick finned-tube system is approximatelytwo times higher than that in the unfinned-tube system. The

Table 1Physical properties of MPCM slurry and its component.

Density Sp

(kg m�3) he

(J

Hexadecane (solid) 780 18

(liquid) 770 22

Urea–formaldehyde 1490 16

Water (at 20 1C) 998 41

MPCM particle (solid) 829 17

(liquid) 819 21

MPCM Slurry (mass fraction) F=0.2 976 37

thermal performance for three different tube systems was found tobe strongly affected by the inlet temperature but not by the flowrate of the heat transfer fluid.

In general all preliminary studies and experiments indicatepromising applications of MPCM slurry as a thermal storagemedium. However, experimental investigations on effects ofsupercooling on MPCM slurry thermal storage capacities appearto be limited so far. In our earlier work, Wang and coworkers[21,22]studied the flow and heat transfer behaviors of phasechange material slurries in a horizontal circular tube and theperformance of cooled-ceiling operating with MPCM slurry,neglecting the supercooling for simplification. In this paper, wewill present our experimental investigation of microencapsulatedparaffin slurry as a cooling storage media for building coolingapplications, especially we focus on establishing the effectivecooling storage capacity at the cooling temperature range withnatural cooling sources such as evaporative cooling and nocturnalradiative cooling, which can be used to realistically estimate theavailability for cooling of these natural cooling sources underdifferent climatic conditions.

2. Experimental approach

2.1. Microcapsule and slurry preparation

MPCM slurry is made by microencapsulating phase changematerial with a thin film as shell and dispersing the microencapsu-lated PCM into an aqueous solution as a carrier fluid. Hexadecane(C16H34) was chosen as the core material, whose melting tempera-ture is 18 1C, and latent heat is 234 kJ/kg [23], and amino plastics asshell material, respectively. The core-shell ratio was controlled to beabout 7:1 by weight during the preparing process; the thickness ofthe shell was �0.3 mm. Pure water was chosen as carrier fluid. Theinitial reason is that water is easy to handle and has no chemicaleffect on the phase change material and shell wall, and anotherimportant reason is that the original form of slurry is in aqueousform and easily diluted with pure water to obtain differentconcentrations according to various engineering applications. Thesolid content of the emulsion used in this study is about 23%.

2.2. Slurry properties

Flow and heat transfer characteristics are associated with thefollowing properties: density, thermal conductivity, melting/freezing point and latent heat of fusion of the PCM, as well asthe properties of the carrier fluid, the particle size andconcentration of the slurry. Thermal property model applied byvarious researchers [11,22] to the phase change slurries was usedin the present study to calculate the properties of the slurry,which are given in Table 1[21].

ecific Thermal Latent

at conductivity heat

kg�11C�1) (W m�1

1C�1) (kJ kg�1)

05 0.4 224

21 0.21

75 0.433

83 0.599

89 0.382 196

53 0.203

07 0.551 39.2

ARTICLE IN PRESS

Fig. 1. DSC results of microencapsulated C16H34.

Watermake-up

Chiller

Pump

Cool storage tank

Cooled Ceiling

Chamber

Exhaust air

Fresh air

Pump

Expansiontank

Watermake-up

Drain

Drain

Expansiontank

Fig. 2. Schematic diagram of experimental air-conditioning system.

Table 2Instrument accuracy.

Type of measurement Instrument accuracy

Temperature difference 70.1 1C

Liquid flow 71.0% of reading

Pressure difference 725 Pa

Time 70.20%

Mass 70.20%

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–10481040

2.3. DSC measurements results

The phase transition point and latent heat of microencapsu-lated C16H34 were measured by a differential scanning calorimeter(Perkin Elmer DSC7) with a heating/cooling rate of 5 1C/min.Perkin Elmer software was used to analyze and plot the thermaldata. Fig. 1 shows the DSC results of microencapsulated C16H34

particles. From the DSC results, the key thermal performance ofmicroencapsulated C16H34 particles can be drawn, with themelting temperature Tm=15.7 1C, freezing temperatureTf=12.5 1C, supercooling degree DT=3.3 1C, and latent heat ofmelting DHm=178 kJ/kg. However, it can also be observed thattwo peaks occur with the freezing process, each associated with alatent heat of freezing of 57 kJ/kg and 83 kJ/kg, respectively. Thisindicates that, for the total latent heat of freezing to reach thevalue of 140 kJ/kg, a supercooling to the temperature around 5 1Cwould be required.

2.4. Experimental system description

To investigate the effects of supercooling at realistic operatingconditions in air-conditioning applications, we constructed asmall scale system, which employs cooled-ceiling supplied with

ARTICLE IN PRESS

Fig. 3. Charging process with water as storage medium at 3 mixer speeds (a) 80, (b) 230 and (c) 380 rpm.

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–1048 1041

MPCM slurry to cool an office room which could be occupied bytwo persons. The system will be used as the experiment platformto observe the cyclic cooling storage (or charging) and release

(or discharging) processes, in complementary to the DSCmeasurements. The schematic diagram of the whole system isgiven in Fig. 2.

ARTICLE IN PRESS

Table 3System heat balance analysis with water.

Q_chiller DU_water Q_t.loss DQ DQ/Q_chiller

(kJ) (kJ) (kJ) (kJ) (%)

80 rpm 7597 7366 53 179 2.4

230 rpm 9731 9027 125 579 5.9

380 rpm 8664 7892 103 669 7.7

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–10481042

The dimension of the chamber is 2.7 m (w)�4 m (l)�3m (h).Four water-type ceiling panels are installed in the chamber toextract the sensible load. Fresh air is cooled and dehumidifiedby a conventional air handing unit (AHU), and then supplied tothe chamber at minimum ventilation rate for the ventilationpurpose.

The cool storage tank is a cylindrical stainless steel tank, withan insulation layer of 40 cm glass wool clad with a stainless steelsheet. The diameter of the tank is 0.6 m, the height is 0.8 m, andthe internal volume is 0.23 m3. Two coil heat exchangers are set inthe tank, both are f20�15 m (l), made of titanium alloy. There isa variable-speed stirrer in the tank to enhance the heat transfer,controlled by an electronic controller.

During the charging, the chiller runs and supplies chilled waterto cool the MPCM slurry in the cool storage tank via a heatexchanger. The compressor of the chiller operates with an on/offcontrol, with a set-point for the supply chilled water temperatureof 5 1C. The cooling was stored mostly as the form of latent heat inthe MPCM particles, and a small portion as sensible heat in theMPCM particles and water.

During the discharging, water is cooled by the cool storagetank via the other heat exchanger, and then circulated through thecooled ceiling panels in the chamber. A by-pass pipe and a three-way modulating control valve are installed to vary the flow rate ofthe chilled water through the ceiling panel. A variable-speedmixer in the storage tank runs to enhance the heat transfer of thetwo heat exchangers during both the charging and dischargingprocess.

There are 8 T-type thermocouples, 4 bourdon manometers,2 turbine flowmeters, and 4 W transducers installed to measurethe system operation parameter. All the data were collected andtransferred by real-time data acquisition system (DAS), and thenstored in a computer. The instrument accuracies are shown inTable 2, in compliance with ASHRAE 94-77 [23].

3. Results and discussion

3.1. Heat loss of the thermal storage tank and heat balance analysis

Pure water is generally used in water-storage and ice-storagesystem, which is one of the most conventional thermal storagesystems, and has been used in many practical commercial andindustrial applications. In this research project, pure water is usedfor the calibration of the heat balance of the new system usingMPCM slurry. Pure water in the tank was first cooled to 8.8 1C, andthen kept in the laboratory for 18 h, where the room temperatureis around 20 1C, the end water temperature raised to 10.8 1C. Theheat loss of the storage tank is given by

Qt:loss ¼mwater CðTend�TinitialÞ ð1Þ

where mwater is the mass of pure water in the storage tank, Tend

and Tinitial are the end and initial water temperatures in thestorage tank, C the specific heat capacity of pure water.

Then the total thermal resistance of the storage tank is given by

Rtank ¼Qt:loss

tðTlib:average�Twater averageÞð2Þ

where Rtank is the total thermal resistance of the storage tank, t thetest period, Tlib.average is the average laboratory temperature, andTwater.average the average water temperature. The calculation resultsare Qt.loss=1709 kJ over the 18 h test period, and Rtank=2.69 W/1C. Inthis way it was ascertained that the heat losses to the surroundingsare small, a maximum 3% of the total cooling storage.

Based on the calculation about the heat loss of the thermalstorage tank, the heat balance is checked when running withwater. Theoretically, the cooling supplied by chiller Qchiller will betotally stored in water as sensible heat storage DUwater. Consider-ing the heat loss of the tank, the difference between the amount ofheat coming into and leaving the thermal storage tank DQ isdefined as

DQ ¼Qchiller�DUwater�Qt:loss ð3Þ

The mixer in the tank runs at 3 different speeds (80, 230 and380 rpm). The charging process results with water as the mediumare shown in Fig. 3.

The heat balance calculation results based on Eqs. (1–3) areshown in Table 3.

Based on the test and calculation results, it is ascertained thatthe error of heat balance is below 8% of the total input coolingenergy.

3.2. Charging/discharging process

The pure water in the storage tank was replaced with theMPCM slurries, which were cooled to 5 1C and heated to 25 1Ccyclically at 3 different tank mixer speeds (80, 230 and 380 rpm),and the test repeated 5 times at every mixer speed. One set ofcharging/discharging process results with MPCM slurry as themedium at 3 different mixer speeds (80, 230 and 380 rpm) areshown in Fig. 4 and 5. The sensible thermal storage process andthe latent heat storage process can be observed via the slope ofthe slurry temperature change over time. It can also be observedthat the temperature slope variation at the higher stirring speed ismore obvious, which indicates that proper forced convection heattransfer is necessary in the slurry tank.

3.3. Supercooling and effective latent heat

As mentioned above, latent heat, melting/freezing points andsupercooling degree of MPCM are key parameters that determinethe cooling storage tank size, operating temperature range andenergy efficiency of mechanical cooling storage or coolingavailability of a natural cooling storage. Some studies [18,24]show that the initiation of freezing point is inversely proportionalto cooling rate, with a linear trend, which means the supercoolingdegree is proportional to cooling rate. Due to the accuratemeasuring range of the DSC equipment and the experimentalefficiency, cooling rate from 3 to 15 1C/min [18,25,26] aregenerally used to measure thermal characters of PCM/MPCMs.In our experiments, microencapsulated C16H34 were measured byDSC with heating/cooling rate of 5 1C/min. But when MPCM slurryis used in a practical air-conditioning system, it generally operateswith a much lower heating/cooling rate, and the key thermalcharacters of microencapsulated C16H34 tested in the practical air-conditioning system are expected to be different from the DSCresults. About 0.4 1C/min cooling rate was used in our experimentsystem,

In practice, latent heat storage systems also make use of somesensible heat capacity in the system, so the whole thermal storage

ARTICLE IN PRESS

Fig. 4. Charging process with MPCM slurry as storage medium at 3 mixer speeds (a) 80, (b)230 and (c) 380 rpm.

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–1048 1043

was defined as [23]

DUMPCM ¼mcplðTinitial�TmÞþmMPCM DhþmcpsðTm�TendÞ ð4Þ

where m is the mass capacity of MPCM slurry in the storage tank,mMPCM the mass capacity of MPCM particle, Dh the latent heat offusion per unit mass, cpl and cps are specific heat capacities in

ARTICLE IN PRESS

Fig. 5. Discharging process with MPCM slurry as storage medium at 3 mixer speeds (a) 80, (b) 230 and (c) 380 rpm.

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–10481044

liquid phase and solid phase separately, Tend and Tinitial are the endand initial slurry temperatures in the storage tank, Tm the meltingtemperature of MPCM slurry. But in Eq. (4), it is assumed that

phase change is completed at Tm and no supercooling isconsidered. A different calculation of the storage heat is neededwhen supercooling is considered, as illustrated below.

ARTICLE IN PRESS

Table 4MPCM mass fraction of the MPCM slurry determined by heat balance analysis.

Mixer speed Mass fraction

80 rpm 21.2%71.2%

230 rpm 24.7%70.3%

380 rpm 24.1%70.6%

-20%

0%

20%

40%

60%

80%

100%

5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5

MPCM slurr

hl.e

/ he

x 1

00%

0%

20%

40%

60%

80%

100%

120%

5 6 7 8 9 10 11 12 13

MPCM slurr

hl.e

/ he

x 1

00%

-20%

0%

20%

40%

60%

80%

100%

120%

5 6 7 8 9 10 11 12

MPCM slurr

hl.e

/ he

x 1

00%

Fig. 6. Ratio of freezing latent heat completion at 3 mixer speed

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–1048 1045

In every measurement interval of 5 s, the heat balance ofthe MPCM slurry system charging/discharging process can bedefined as

Qchiller ¼DUslurryþQt:loss ð5Þ

Qceiling ¼DUslurry�Qt:loss ð6Þ

13.5 14.5 15.5 16.5 17.5 18.5 19.5 20.5

y Temperature (°C)

14 15 16 17 18 19 20 21 22 23

y Temperature (°C)

13 14 15 16 17 18 19 20 21

y Temperature (°C)

s during charging process (a) 80, (b) 230 and (c) 380 rpm.

ARTICLE IN PRESS

-40%

-20%

0%

20%

40%

60%

80%

100%

hl.e

/ he

x 1

00%

-20%

0%

20%

40%

60%

80%

100%

120%

5 6.5 8 9.5 11 12.5 14 15.5 17 18.5 20 21.5

5 6.5 8 9.5 11 12.5 14 15.5 17 18.5 20 21.5

hl.e

/ he

x 1

00%

-40%

-20%

0%

20%

40%

60%

80%

100%

120%

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

MPCM slurry Temperature (°C)

MPCM slurry Temperature (°C)

MPCM slurry Temperature (°C)

hl.e

/ he

x 1

00%

Fig. 7. Ratio of freezing latent heat completion at 3 mixer speeds during discharging process (a) 80, (b) 230 and (c) 380 rpm.

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–10481046

where, Qchiller, Qceiling can be calculated as

Q ¼ _mCðTout�TinÞ ð7Þ

Qt.loss can be calculated using Eq. (2), and DUslurry is related to thesensible and latent heat change as follows:

DUslurry ¼DUSþDUL

¼mwatercp:waterðTend�TinÞþmMPCMcp:MPCMðTend�TinÞþDmMPCMhl ð8Þ

hl is Dh as in Eq. (4), which is determined using DSC technique,and then the actual phase change quantity DmMPCM can be obtained

DmMPCM ¼QChiller�Qt:loss�DUS

Dhð8aÞ

ARTICLE IN PRESS

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–1048 1047

Then we can define the effective latent heat hl.e as

hl:e ¼DmMPCM

mMPCMhl ð9Þ

where DmMPCM is the mass of MPCM particles, which havechanged phase, mMPCM the total mass of MPCM particles, and hl

the latent heat of fusion per unit mass.Also, the mass fraction of MPCM particles in the slurry were

double-checked, based on the heat balance for the dischargingprocess

mMPCM ¼QCeilingþQt:loss�DUS

Dhð10Þ

The results are shown Table 4, and agree with the initial massratio when the slurry was prepared.

Based on the methodology above, the ratio of latent heatcompletion, defined as hl.e/hl at each cooling temperature, can beacquired, as shown in Figs. 6 and 7.

Judging from the transition points of the curves in Figs. 6 and 7,the freezing starts at the temperature of 13 1C, and thecorresponding supercooling degree is about 3 1C. On the otherhand, the latent heat conversion is not complete even at the endtemperature of 5 1C. As shown in Table 5, the end effectivefreezing latent heats of the MPCM vary with the stirringconditions of the slurry.

In addition, the functions in the form of polynomial obtainedvia regression analysis, describing the relationship of the ratio oflatent heat completion and MPCM slurry temperature can bewritten as follows:

For the charging process

Table 5Freezing latent heat of MPCM when cooled to the temperature of 5 1C.

Mixer speed hl.e (kJ/kg)

80 rpm 149726

230 rpm 166715

380 rpm 17979

Table 6Regression parameters and error term.

c0 c1 c2 c3

Charging

80 rpm 0.1021 2.9251 �25.751 98.6

230 rpm �0.004 2.5433 �26.335 103.3

380 rpm 0.0681 1.5992 �13.665 36.4

Discharging

80 rpm �0.1029 1.3229 �15.282 70.3

230 rpm 0.0139 �1.5337 17.953 �81.4

380 rpm �0.0235 �2.7172 30.872 �136.1

hl:e

hl¼

0 tZ15

c0þc1Tþc2T2þc3T3þc4T4þc5T5þc6T6 5:5oto15

0:9 tr5:5

where

8><>:

and for the discharging process

hl:e

hl¼

0 tr14

c0þc1Tþc2T2þc3T3þc4T4þc5T5þc6T6 14oto18

1 tZ18

where

8><>:

The results of regression analysis of the charging anddischarging process are shown in Table 6.

Eqs. (11) and (12) describe the charging and dischargingprocess in realistic building cooling application conditions. Inpractical air-conditioning utilizations, the slurry would be cooledto 7–8 1C, to optimize the chiller and whole air-conditioningsystem operation efficiency and cost. Based on this operationcondition, the latent heat transition is completed by about 80%according to Fig. 6 and Eqs. (11) and (12). However, to utilizenatural cooling sources such as evaporative cooling for buildingcooling, higher storage temperature around 17 1C would bedesired. In this respect, the MPCM experimented in this study,due to the supercooling occurred, would offer very limited coolingstorage capacity at this temperature, but will work effectively atthe climates when natural cooling such as nocturnal sky radiativecooling can cool the slurry to around 7 1C. This analysisdemonstrates that developing PCM working at the temperatureof around 17 1C with small supercooling can greatly maximize theopportunities of using natural cooling sources for building coolingapplications.

4. Conclusions

Melting and crystallization behaviors of MPCM slurry runningin a thermal storage test system are investigated experimentally.Supercooling and effective latent heat of MPCM slurry withdifferent experimental conditions are also investigated. Theexperiment results show that, in a practical air conditioningsystem with thermal storage, the latent heat transition is notentirely complete. For the specific MPCM investigated, the

T ¼t�15

5:5�15ð11Þ

T ¼t�14

18�14ð12Þ

utilization ratio of the latent heat is related to the endtemperature, and is around 80% when cooled to around 8 1C.

Because of the supercooling phenomena, supercooling isnecessary to utilize the latent storage capacity, or only the partialstorage capacity can be used at a limited cooling temperature.With electric-driven cooling storage, this supercooling will lowerthe COP of the cooling storage process. For passive coolingschemes, this would mean reduced utilization hours of naturalcooling sources such as evaporative cooling and reduced per-volume storage capacities.

c4 c5 c6 R2

27 �160.97 118.97 �33.045 0.9126

9 �167.48 123.5 �34.682 0.8961

83 �16.796 �25.188 18.586 0.9941

09 �142.06 131.97 �45.416 0.7566

68 173.81 �174.04 66.4 0.9866

8 280.61 �270.71 99.154 0.9656

ARTICLE IN PRESS

S. Zhang, J. Niu / Solar Energy Materials & Solar Cells 94 (2010) 1038–10481048

Acknowledgments

The authors would like to express thanks to Prof. Yi Li andDr. Qingwen Song of ITC, the Hong Kong Polytechnic Universityfor the MPCM slurry preparation. The authors are also grateful toMr. Yu Han for experimental assistance and Mr. Chak-kit Chengfor experimental system construction assistance.

References

[1] I. Dinc-er, M. Rosen, in: Thermal Energy Storage: Systems and Applications,Wiley, Chicheste, New York, 2002.

[2] B. Zalba, et al., Review on thermal energy storage with phase change:materials, heat transfer analysis and applications, Applied Thermal Engineer-ing 23 (3) (2003) 251–283.

[3] M.M. Farid, et al., A review on phase change energy storage: materialsand applications, Energy Conversion and Management 45 (9–10) (2004)1597–1615.

[4] S. Gschwander, P. Schossig, H.M. Henning, Micro-encapsulated paraffin inphase-change slurries, Solar Energy Materials and Solar Cells 89 (2–3) (2005)307–315.

[5] K.E. Kasza, M.M. Chen, Improvement of the performance of solar energy orwaste heat utilization systems by using PHASE-CHANGE slurry as anenhanced heat-transfer storage fluid, Journal of Solar Energy Engineering,Transactions of the ASME 107 (3) (1985) 229–236.

[6] C.W. Sohn, M.M. Chen, Microconvective thermal conductivity in dispersetwo-phase mixtures as observed in a low velocity Couette flow experiment,Journal of Heat Transfer. Transactions of the ASME 103 (1) (1981) 47–51.

[7] A.F. Regin, S.C. Solanki, J.S. Saini, Heat transfer characteristics of thermalenergy storage system using PCM capsules: a review, Renewable andSustainable Energy Reviews 12 (9) (2008) 2438–2458.

[8] Y. Yamagishi, et al., An evaluation of microencapsulated PCM for use in coldenergy transportation medium, in: Energy Conversion Engineering Con-ference, 1996. IECEC 96. Proceedings of the 31st Intersociety, Washington,DC, USA, 1996.

[9] Y. Yamagishi, et al., Characteristics of microencapsulated PCM slurry as aheat-transfer fluid, AIChE Journal 45 (4) (1999) 696–707.

[10] H. Inaba, M.-K. Kim, A. Horibe, Melting heat transfer characteristics ofmicroencapsulated phase change material slurries with plural microcapsuleshaving different diameters, Journal of Heat Transfer 126 (4) (2004)558–565.

[11] P. Charunyakorn, S. Sengupta, S.K. Roy, Forced convection heat transfer inmicroencapsulated phase change material slurries: flow in circular ducts,International Journal of Heat and Mass Transfer 34 (3) (1991) 819–833.

[12] Y. Zhang, A. Faghri, Analysis of forced convection heat transfer inmicroencapsulated phase change material suspensions, Journal of Thermo-physics and Heat Transfer 9 (4) (1995) 727–732.

[13] X.X. Hu, Y.P. Zhang, Novel insight and numerical analysis of convective heattransfer enhancement with microencapsulated phase change materialslurries: laminar flow in a circular tube with constant heat flux, InternationalJournal of Heat and Mass Transfer 45 (15) (2002) 3163–3172.

[14] H.W. Ryu, et al., Heat transfer characteristics of cool-thermal storage systems,Energy 16 (4) (1991) 727–737.

[15] M.J. Oliver, P.D. Calvert, Homogeneous nucleation of n-alkanes measured bydifferential scanning calorimetry, Journal of Crystal Growth 30 (3) (1975)343–351.

[16] A. Sharma, et al., Review on thermal energy storage with phase changematerials and applications, Renewable and Sustainable Energy Reviews 13(2) (2009) 318–345.

[17] X.-x Zhang, et al., Crystallization and prevention of supercooling ofmicroencapsulated n-alkanes, Journal of Colloid and Interface Science 281(2) (2005) 299–306.

[18] J. Alvarado, et al., Characterization of supercooling suppression of micro-encapsulated phase change material by using DSC, Journal of ThermalAnalysis and Calorimetry 86 (2) (2006) 505–509.

[19] Y.F. Fan, et al., Super-cooling prevention of microencapsulated phase changematerial, Thermochimica Acta 413 (1–2) (2004) 1–6.

[20] J.C. Choi, S.D. Kim, G.Y. Han, Heat transfer characteristics in low-temperaturelatent heat storage systems using salt-hydrates at heat recovery stage, SolarEnergy Materials and Solar Cells 40 (1) (1996) 71–87.

[21] X. Wang, J. Niu, A.H. van Paassen, Performance of cooled-ceiling operatingwith MPCM slurry, Energy Conversion and Management 50 (3) (2009)583–591.

[22] X. Wang, J. Niu, et al., Flow and heat transfer behaviors of phase changematerial slurries in a horizontal circular tube, International Journal of Heatand Mass Transfer 50 (13–14) (2007) 2480–2491.

[23] G.A. Lane, in: Solar Heat Storage: Latent Heat Materials; V1: Background andScientific Principles, CRC Press,Inc., Boca Raton, Florida, 1983.

[24] Y.P. Zhang, H.P. Hu, et al., in: Heat Storage: Phase Rule and Equilibrium (inChinese), 1st ed., University of Science and Technology of China Press, Heifei,1996 484p.

[25] B. He, F. Setterwall, Technical grade paraffin waxes as phase change materialsfor cool thermal storage and cool storage systems capital cost estimation,Energy Conversion and Management 43 (13) (2002) 1709–1723.

[26] G.J. Suppes, M.J. Goff, S. Lopes, Latent heat characteristics of fatty acidderivatives pursuant phase change material applications, Chemical Engineer-ing Science 58 (9) (2003) 1751–1763.