Available online at www.sciencedirect.

com

Solar Energy 83 (2009) 1245–1252

www.elsevier.com/locate/solener

Phase change characteristic study of spherical PCMs

in solar energy storage
M. Veerappan a, S. Kalaiselvam a,*, S. Iniyan a, Ranko Goic b
a
Refrigeration and Air Conditioning Division, Department of Mechanical Engineering, Anna University, Chennai, Tamil Nadu 600025, India
b
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Croatia

Received 4 May 2007; received in revised form 14 February 2009; accepted 19 February 2009
Available online 7 April 2009

Communicated by: Associate editor Halime Paksoy

Abstract

This paper investigates the phase change behavior of 65 mol% capric acid and 35 mol% lauric acid, calcium chloride hexahydrate, n-
octadecane, n-hexadecane, and n-eicosane inside spherical enclosures to identify a suitable heat storage material. Analytical models are
developed for solidification and melting of sphere with conduction, natural convection, and heat generation. Both the models are val-
idated with previous experimental studies. Good agreement was found between the analytical predictions and experimental study and the
deviations were lesser than 20%. Heat flux release at the wall, cumulative energy release to the external fluid, are revealed for the best
PCM. The influence of the size of encapsulation, initial temperature of the PCM, the external fluid temperature on solidified and molten
mass fraction, and the total phase change time are also investigated.
Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: Phase change material; Thermal energy storage; Analytical model; Solidification; Melting

1. Introduction these systems have potential applications in active and pas-

sive solar heating, water heating, air conditioning, etc., and
Thermal energy storage (TES) is used to assist in the are regarded as an economical and safe energy storage
effective utilization of thermal energy in wide number of technology. Solar energy can effectively be stored during
industrial and private applications. TES is a method for the day and thereafter it can be used at the later stage. In
adjusting the time-discrepancy between power supply and solar TES, it is very difficult to access the performance of
demand. There are three ways to store thermal energy, a PCM because of the complexity of the phenomenon
namely sensible heat, latent heat, and reversible reaction involved. This is due to non linearity of the solid–liquid
storage (thermo chemical heat). Latent heat stores may front, convective effects of the liquid PCM, and geometry
be advantageous to water stores (Fath, 1998). Excess ther- of the containment. It is also necessary to include viscous
mal energy is stored in a material as latent heat by melting effects in liquid PCM. Hence a model has been developed
the material. The stored thermal energy is utilized later in to find the transient positions of the moving interface
necessary timing by freezing the material, again. Solar and to interpret the heat transport inside PCM. Since this
energy is a renewable energy and hence the need of fossil model included heat generation it can be applied to nuclear
fuels, green house gas emissions can be reduced. Thus, fuel freezing, laser melting, microwave thawing, and spray
drying where inclusion of heat generation parameter is very
* significant (Jiji and Gaye, 2006). Heat generated within the
Corresponding author. Tel./fax: +91 44 2220 3262.
E-mail addresses: [email protected] (M. Veerappan), kalai@
PCM is very significant and has a dominant influence in the
annauniv.edu (S. Kalaiselvam), [email protected] (S. Iniyan), heat transfer of the phase change process. Since the model
[email protected] (R. Goic). includes natural convection effects, this can also be directly

0038-092X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.solener.2009.02.006
1246 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252

Nomenclature

c specific heat, J/kg K Greek symbols

E dimensionless energy release a thermal diffusivity, m2/s
Fo Fourier number b dimensionless heat generation parameter
H latent heat of fusion, kJ/kg c the mass ratio of liquid in the mixture to the to-
k thermal conductivity, W/mK tal mass of mixture
Pr Prandtl number h dimensionless temperature
q dimensionless heat flux n spatial variable, m
q000 heat generation, W/m3 q density, kg/m3
R radius of spherical capsule, m r position of the interface in spatial variable, m
RaL Rayleigh number based on gap width
ri radii of moving front, m Subscripts
s dimensionless location of the solid liquid inter- 0 external fluid
face eff effective
Ste Stefan number i initial
T temperature, °C L liquid phase
t time, s L  S fusion
W dimensionless velocity S solid phase
x dimensionless radial co-ordinate

applied to study the PCMs in thermal energy storage appli- chloride hexahydrate (Vyshak and Jilani, 2007; Zivkovic
cations. Model considers, the PCM has its initial tempera- and Fujii, 2001), n-octadecane (Pinelli and Piva, 2003), n-
ture greater than melting or freezing temperature. hexadecane (Kenisarin and Mahkamov, 2007), and n-eico-
Conduction in liquid phase is also considered. This shows sane (Jones et al., 2006). All the materials have phase tran-
the versatility of the model developed. sition (Melting point) temperature in between 15 and
Commonly used phase change materials in solar thermal 40 °C. These phase change materials may be used in Ther-
energy storage are fatty acids, inorganics, and organics. mal Enclosures like Telecom Chambers BTS or BT Shel-
Fatty acids have always driven interests among researchers ters, solar domestic hot water heating, passive solar space
as a feasible potential PCM. This is due to low toxicity, heating/cooling systems in air-conditioning. Thermo phys-
melting congruency and better chemical stability (Natalia ical properties of the selected PCMs are given in Table 1.
et al., 2002). But inorganics have greater phase change This paper presents an analytical model based upon
enthalpy. Corrosion, phase separation, lack of thermal sta- quasi steady approximations for solidification and melting
bility, undercooling, and phase segregation are some of the of PCMs in a spherical container under isothermal bound-
problems encountered while using inorganics as PCM ary conditions at the external surface of the spherical shell.
(Hasnain, 1998). Organics are non corrosive, chemically Considering all PCMs to be enclosed within the same con-
and thermally stable. But thermal conductivity of the tainer, the resistance of the outer shell is neglected. Impor-
organics in both the phases is very low (Zalba et al., tant parameters such as size of the spherical capsule,
2003). The present study has chosen five materials as external fluid temperature and the initial PCM temperature
potential PCMs, which can be used as low temperature are investigated and their effects on the solidified mass frac-
thermal energy storage materials. They are 65 mol% capric tion and the time for complete solidification are presented.
acid and 35 mol% lauric acid (Natalia et al., 2002), calcium The present predictions are also validated with previous

Table 1
Thermo physical properties of phase change materials.
Properties Capric/lauric Acid CaCl26H2O n-Octadecane n-Hexadecane n-Eicosane
Melting temperature, °C 18 29 27.5 18 36.4
Latent heat of fusion, kJ/kg 140.8 187 241.36 235 248
Specific heat of liquid, kJ/kg K 2.24 2.20 2.20 2.10 2.40
Specific heat of solid, kJ/kg K 1.97 1.40 1.90 1.95 1.926
Thermal conductivity of liquid PCM, W/mK 0.139 0.53 0.157 0.1562 0.146
Thermal conductivity of solid PCM, W/mK 0.143 1.09 0.39 0.43 0.426
Density of liquid, kg/m3 894.9 1530 814 765 769
Density of solid, kg/m3 900 1710 840 835 910
Dynamic Viscosity, 103 Pa s 1.65 22.5 3.878 3 3.27
 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252 1247

LS T i
experimental measurements. Similar parametric studies are (5) hL (x, Fo = 0) = TT LS T 0
= hi
also carried out for melting. (6) s (Fo = 0) = 1

2. Analytical formulation The dimensionless interface energy equation is,

 
@hs ðs; F o Þ @hL ðs; F o Þ ds
2.1. Solidification of sphere Ste  ¼ ; ð6Þ
@x @x dF o
To predict the characteristics behavior of moving
where Ste is the Stefan number defined as, Ste ¼ cðT LSH T 0 Þ
boundary in the spherical capsules, single capsule is consid-
Neglecting the transient terms, and direct integration of
ered for the analysis. Spherical co-ordinate system, with
Eqs. (4) and (5) and application of corresponding initial
PCM has its external surface at the co-ordinate n = R, sud-
and boundary conditions, give the solution to the temper-
denly cooled to the temperature T0 < TLS is introduced,
ature distribution in the solid and liquid phase as,
as shown in Fig. 1. The Sphere is initially at its melting
point temperature Ti. 6  b þ bs3 ðb  6Þs  bs3 b 2
hS ðx; FoÞ ¼ þ þ x; ð7Þ
The governing heat balance equations in both the phases 6ð1  sÞ 6ð1  sÞx 6
are,
 2  b b
@ T S 2 @T S q000 @T S hL ðx; FoÞ ¼ x2  s2 ð8Þ
a þ þ ¼ ; rnR ð1Þ 6 6
2
@n n @n qc @t
 2  Differentiating Eqs. (7) and (8), and substituting into inter-
@ T L 2 @T L q000 @T L face Eq. (6), and applying the initial condition s
a 2
þ þ ¼ ; 0nr ð2Þ
@n n @n qc @t (Fo = 0) = 1, results in
Dimensionless variables are adopted and the governing Z
6 s sðs  1Þ
equation, boundary conditions, initial conditions and the SteF o ¼ ds ð9Þ
b 1 s2 þ ðb6  1Þ
interface energy balance equation are cast in dimensionless
form. The dimensionless quantities are, The above integral has to be evaluated for three cases
n T LS  T S ðn; tÞ b < 6.0, b = 6.0 and b > 6.0. For the selected PCM the b
x ¼ ; hs ¼ ; value is less than 6.0. In this case, total solidification takes
R T LS  T 0
T LS  T L ðn; tÞ r at place without reaching steady state. The unique case of
hL ¼ ; s¼ Fo ¼ 2 & b = 6.0, total solidification will take place at the steady
T LS  T 0 R R
state, but for cylinder it is at b = 4.0, and for slab it is at
R2 q000 b = 2.0.
b¼ ð3Þ
kðT LS  T 0 Þ The transient positions of the interface for b < 6.0,
The dimensionless form of the governing equations is, sffiffiffiffiffiffiffiffiffiffiffi" sffiffiffiffiffiffiffiffiffiffiffi! sffiffiffiffiffiffiffiffiffiffiffi!#
6 6 6 1 b 1 b
SteF o ¼ ðs  1Þ þ  1 tan  tan s
@ 2 hS 2 @hS @hS b b b 6b 6b
þ b¼ ; where sx1 ð4Þ
@x2 x @x @Fo 5:3753 3
þ  lnð6  b þ bs2 Þ ð10Þ
@ 2 hL 2 @hL @hL b b
þ b¼ ; where 0xs ð5Þ
@x2 x @x @Fo
During solidification of pure material, buoyancy forces
Similarly the non dimensional boundary and initial condi- may arise due to temperature gradients in the liquid caus-
tions are, ing natural convection that increases the heat flux at the
solid–liquid interface. The major effects of natural convec-
(1) hS (x = 1, Fo) = 1 tion are reduction in solidification rate, and modification of
(2) hS (x = s, Fo) = 0 the solid microstructure (Hanumanth, 1990). The natural
(3) hL (x = s, Fo) = 0 convective effects are incorporated in the model by substi-
@h ðx¼0;F
(4) L @x Þ ¼ 0 tuting the effective thermal conductive coefficient instead of
thermal conductivity of liquid PCM in the interface energy
equation.
Solid Phase r Dimensionless Dimensionless cumulative energy released by the PCM
Co-Ordinates
1 is given by,
Liquid Phase s
 sffiffiffiffiffiffiffiffiffiffiffi" sffiffiffiffiffiffiffiffiffiffiffi!
Spherical 3 6 6 6 1 b
0 σ R ξ Co- Ordinates E ¼1  s þ b ðs  1Þ þ  1 tan
b b b 6b
Moving
Boundary sffiffiffiffiffiffiffiffiffiffiffi!# 
1 b 5:3753 3
tan s þ  lnð6  b þ bs2 Þ ð11Þ
6b b b
Fig. 1. Solidification domain of the spherically encapsulated PCM.
1248 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252

The dimensionless thermal flux at the external surface of ð6 þ b  3beff s2 þ 2beff s3 Þð1  SteF o beff Þ ¼ 6 ð18Þ
the capsule is
  This analytical solution is derived based on conduction,
@hS  6s  bs þ bs3 b
q ¼   ¼ þ ð12Þ natural convection and heat generation to be the dominant
@x x¼1 6ð1  sÞ 3 mode of heat transfer.

2.2. Melting of sphere 3. Results and discussion

The problem of melting inside a PCM can be treated as 3.1. Solidification of sphere
a problem of natural convection inside concentric spheres.
Movement of solid phase due to density difference is 3.1.1. Experimental validation
neglected. Natural convection purely depends on Rayleigh Fig. 3 shows the comparative results of the analytical
number in terms of gap width (RaL) (Chu and Lee, 1993). predictions of the present model and the experimental
This problem can be resolved by defining conduction as the results (Eames and Adref, 2002) realized to validate
dominant mode of heat transfer and merging the convec- the model under the same geometrical and operational
tive effects in the thermal conductivity of the liquid (Rai- conditions. Analytically, the transient interface positions
thby and Hollands, 1975). The effective thermal and the complete solidification time are predicted by
conductivity of PCM is obtained from scanlan et al. four models. It can be noted in Fig. 2, that analytical
(1970) correlation. predictions from the model which included conduction,
 0:252 convection and heat generation are closer to experimen-
k eff 0:228 L
¼ 0:202RaL Pr0:029 ð13Þ tal measurements. This shows the significance of consid-
kL ri ering all the three heat transfer mechanisms. The
where keff – effective thermal conductivity, W/m K; kL – deviations between the present predictions and experi-
thermal conductivity of the liquid PCM. mental results are within 15%. During experimentation,
In the dimensionless formulation, the heat generation the interface positions are calculated from the air space
parameter b, Stefan number Ste and dimensionless temper- pressure with time.
ature hL are redefined as,
R2 q000 cðT 0  T LS Þ 3.1.2. Comparisons of PCMs
beff ¼ ; Ste ¼ ; A validated analytical solution is used to study the
k eff ðT 0  T LS Þ H
behavior of PCMs 65 mol% capric acid and 35 mol% lau-
T 0  T L ðn; tÞ ric acid, calcium chloride hexahydrate, n-octadecane, n-
hL ¼ ð14Þ
T 0  T LS hexadecane, and n-eicosane. Transient interface plots for
Dimensionless heat equation, boundary and initial condi- PCMs encapsulated inside a radius of 4 cm with external
tions are, fluid temperature 10 °C are presented in Fig. 4. It can be
inferred that calcium chloride hexahydrate solidifies fas-
@ 2 hL 2 @hL @hL ter, compared with other materials. But n-eicosane has
2
þ  beff ¼ ; sx1 ð15Þ
@x x @x @Fo equally good solidifying characteristics as CaCl26H2O.
hL ðx ¼ 1; F o Þ ¼ 0 It can be noted that the thermal conductivity of calcium
hL ðx ¼ s; F o Þ ¼ 1
1
sðF o ¼ 0Þ ¼ 1 Experimental
Dimensionless Solid - Liquid Interface Position, s

0.9 Conduction
Interface energy equation is,
Conduction + Heat Generation
0.8
@T L ðn ¼ r; tÞ dr Conduction + Natural Convection
k eff ¼ qH ð1  cÞ ð16Þ 0.7 Conduction + Heat Generation + Natural
@n dt Convection
0.6
where c is the mass ratio of liquid in the mixture to the total
mass of mixture. 0.5

Dimensionless interface energy equation is, 0.4


@hL  ds 0.3
Ste  ¼ ð1  SteF o beff Þ ð17Þ Solidification inside Spherical
@x x¼s dF o 0.2
Enclosures
Diam eter of the Sphere = 6.27 cm
Coolant Tem perature = -6.5 C

Transient temperature variations are arrived by solving 0.1

Constant Surface Tem perature

the governing equations and applying the boundary and 0

initial conditions. This solution is substituted in Eq. (17), 0 0.05 0.1 0.15 0.2 0.25
Dimensionless Time Ste Fo
and the transient interface position equation for melting
is obtained as, Fig. 2. Experimental validation (solidification).
 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252 1249

Fig. 3. Comparisons of solidification characteristics of the PCMs. Fig. 5. Effect of diameter of the spherical PCM on total solidification
time.

chloride hexahydrate is higher than the other PCMs, in

both the phases. This shows the strong influence of ther- 3.1.4. Influence of diameter of the encapsulated PCMs
mal conductivity on solidified mass fraction and complete The effect of capsule size on complete solidification time
solidification time of the PCM. Complete solidification of PCM is studied and it is shown in Fig. 5. Initial tempera-
time for CaCl26H2O is 84 min, whereas n-eicosane solid- ture of the PCM is 29 °C (fusion temperature), with thermal
ifies at 112 min. Since Calcium chloride hexahydrate has bath temperature 10 °C. It can be found that the spherical
best solidifying properties, it is used to study the influence PCM with 0.3 m diameter takes 1576 min, whereas a sphere
of the size of the capsule shell, initial temperature of the of diameter 0.15 m takes only 320 min for complete solidifi-
PCM, the coolant temperature on the solidified mass frac- cation. Spherical PCMs with larger diameters take more
tion and the time for the complete solidification. time to completely solidify than smaller diameters, since
heat has to travel only a smaller distance and hence rate of
3.1.3. Transient temperature profiles solidification is higher for smaller capsules. Variations of
The temperature profiles of the PCM in the solidified solidified mass fraction with respect to time for various val-
region for each non-dimensional step Fo with internal heat ues of sphere diameter are given in Fig. 6.
generation is given in Fig. 4. It can be noted that PCMs
with higher Ste number solidifies faster. For a rapid release 3.1.5. Influence of external fluid temperature
of the latent heat stored in the material, spherical geometry The effect of heat transfer fluid (HTF) temperature on
with very small radius is preferred (since Fo a 1/R2). solidified mass fraction and complete solidification time
are signified in Figs. 7 and 8, respectively. The radius of

Fig. 4. Transient temperature profiles of solidification. Fig. 6. Effect of size of the capsule on solidified mass fraction.
1250 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252

Fig. 7. Effect of external fluid temperatures on solidified mass fraction.

Fig. 9. Effect of initial temperature of the PCM on solidified mass
fraction.

Fig. 8. Effect of external fluid temperature on total solidification time.

Fig. 10. Transient thermal heat flux profiles.

the capsule is 4 cm and the temperature of the heat transfer

fluid is varied from 10 to 25 °C, while other parameters 3.1.7. Transient heat flux release at the wall
are kept same. It can be seen that, reducing the working The Fig. 10 shows typical transient heat flux profiles for
fluid temperatures, results in rapid decrease in the complete CaCl26H2O with working fluid temperature 10 °C, and
phase change time. Hence rapid solidification takes place radius 4 cm. Higher heat fluxes are observed at the initial
for lower external fluid temperatures. Higher reduction in time period, and thereafter it decreases rapidly. At the
HTF temperatures leads to a smaller change in solidified beginning, PCM has direct contact with the capsule sur-
mass fraction. This can be inferred from Fig. 7, since the face, but later a thermal resistance is formed in between
solidified mass fraction curves appear to be closer. the liquid PCM and capsule surface. Hence thermal energy
has to travel through this solid layer before reaching the
3.1.6. Influence of initial temperature of PCM external surface. So, a sharp decrease in flux values is
Influence of initial temperature of PCM is studied and it observed. Heat flux value approaches zero after a certain
is shown in Fig. 9. The initial temperature of the PCM is period of time which is called as total solidification time.
varied from 30 to 50 °C. Initial temperature of the PCM Initially, there is a higher heat flux release for lower heat
has very minimal effect on both solidified mass fraction transfer fluid temperatures and decrease in slope for lower
and complete solidification time. Since, the influence is very external fluid temperatures is very high. For higher HTF
marginal, almost for all initial temperatures, complete temperatures, after certain time interval the slope decreases
solidification time is same. slowly with time.
 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252 1251

Fig. 13. Comparison of melting characteristics of PCMs.

Fig. 11. Cumulative energy released by PCM to the external fluid.

3.1.8. Cumulative energy release by the PCM to HTF conduction model because at the beginning the conduction
Cumulative energy released by the PCM to the coolant is the dominant mode of heat transfer. After that, melting
fluid is given in Fig. 11. It can be observed that, the energy is greatly influenced by natural convection phenomena.
liberated for a PCM with lower HTF temperature is high. Pure conduction model over predicts the complete melting
PCM with HTF temperature 25 °C released 80 kJ/kg of time. The deviations between the present predictions and
energy at 50 min, whereas PCM with HTF temperature experimental results are within 20%.
5 °C has liberated 18.4 kJ/kg. Hence PCM with very low
HTF temperature characterizes rapid charging rates. 3.2.2. Comparisons of PCMs
Melting characteristics of the PCMs are compared in
Fig. 13. The radius of the capsule is 4 cm and external fluid
3.2. Melting of sphere
temperature is 45 °C. The melting characteristics of
65 mol% capric acid and 35 mol% lauric acid are better
3.2.1. Experimental validation
than the other phase change materials. But this PCM has
Fig. 12 shows a comparison between the analytical pre-
worst solidifying characteristics. CaCl26H2O has excellent
dictions from the present model and experimental results
solidifying characteristics, whereas melting characteristics
under same operating conditions. The transient interface
are also equally good. Hence calcium chloride hexahydrate
positions and complete solidification time predicted by
is identified as better PCM among the chosen PCMs in
model with conduction, natural convection and heat gener-
terms of rapid charging and discharging rates. Hence this
ation matches with the experimental measurements. Ini-
potential PCM can be used in low temperature thermal
tially the experimental values closely match with the pure
energy storages.

3.2.3. Effect of capsule size

Variations of Molten fraction with time for various diam-
eters are given in Fig. 14. Parametric studies were conducted
for CaCl2.6H2O with warm fluid temperature being 45 °C.
The diameter of the capsule is varied from 4 to 12 cm. As
expected, smaller capsule radius PCMs melt faster than the
larger capsules. PCM with 4 cm diameter melts in 26 min,
whereas PCM with 8 cm diameter melts only at 94 min.

3.2.4. Effect of warm fluid temperature

The effect of hot working fluid temperature on the mol-
ten fraction is represented in Fig. 15. The radius of the cap-
sule is 4 cm. It can be noted that, higher working fluid
temperatures results in rapid decrease in the total melting
Fig. 12. Experimental validation (melting). time of PCM. Hence rapid melting takes place for higher
1252 M. Veerappan et al. / Solar Energy 83 (2009) 1245–1252

application, local government regulations, toxicity levels,

corrosion and the cost of the PCM.
This work also investigates phase change characteristics
of five PCMs, and the results show that CaCl26H2O has
excellent solidifying and melting characteristics for solar
latent heat storage applications where rapid charging and
discharging properties are significant. The parametric study
concludes that, lower values of initial temperature did not
make significant contribution on the solidified mass frac-
tion of PCM. It as well infers that for solidification, ini-
tially there is a higher heat flux release for lower coolant
fluid temperatures. Thereafter, rapid decrease in heat flux
values is observed.

References

Fig. 14. Effect of size of the spherical capsule on molten fraction. Chu, H.S., Lee, T.S., 1993. Transient natural convection heat transfer
between concentric spheres. International Journal of Heat and Mass
Transfer 13, 3159–3170.
Eames, I.W., Adref, K.T., 2002. Freezing and melting of water in spherical
enclosures of the type used in thermal (ice) storage systems. Applied
Thermal Engineering 22, 733–745.
Fath, H.E.S., 1998. Technical assessment of solar thermal energy storage
technologies. Renewable Energy 14, 35–40.
Hanumanth, G.S., 1990. Solidification in the presence of natural convec-
tion. International Communications in Heat and Mass Transfer 17,
283–292.
Hasnain, S.M., 1998. Review on sustainable thermal energy storage
technologies. Part I: Heat storage materials and techniques. Energy
Conversion and Management 39, 1127–1138.
Jiji, L.M., Gaye, S., 2006. Analysis of solidification and melting of PCM
with energy generation. Applied Thermal Engineering 26, 568–575.
Jones, B.J., Sun, D., Krishnan, S., Garimella, S.V., 2006. Experimental
and numerical study of melting in a cylinder. International Journal of
Heat and Mass Transfer 49, 2724–2738.
Kenisarin, M., Mahkamov, K., 2007. Solar energy storage using phase
change materials. Renewable and Sustainable Energy Reviews 11,
1913–1965.
Natalia, M., Dimaano, R., Watanabe, T., 2002. The capric–lauric acid and
Fig. 15. Effect of external fluid temperature on molten fraction. pentadecane combination as phase change material for cooling
applications. Applied Thermal Engineering 22, 365–377.
Pinelli, M., Piva, S., 2003. Solid/liquid phase change in presence of natural
external fluid temperatures. Higher HTF temperatures lead convection: a thermal energy storage case study. Journal of Energy
to only a smaller change in molten fraction. Resources and Technology 125, 190–198.
Raithby, G.D., Hollands, K.G.T., 1975. A general method of obtaining
approximate solutions to laminar and turbulent free convection
4. Conclusions problems. Advanced Heat Transfer 11, 265–315.
Scanlan, J.A., Bishop, E.H., Powe, R.E., 1970. Natural convection heat
transfer between concentric spheres. International Journal of Heat and
Analytical models for finding the transient interface Mass Transfer 13, 1857–1872.
positions and complete phase change time for solidification Vyshak, N.R., Jilani, G., 2007. Numerical analysis of latent heat thermal
and melting of spheres are developed. The analytical mod- energy storage system. Energy Conversion and Management 48, 2161–
els developed are validated with the previous experimental 2168.
Zalba, B., Marin, J.M., Cabeza, L.F., Mehling, H., 2003. Review on
results and the deviations are found within 15% for solidi- thermal energy storage with phase change: materials, heat transfer
fication and 20% for melting. This model provides a tool analysis and applications. Applied Thermal Engineering 23, 251–283.
for the selection of a suitable PCM for solar energy storage Zivkovic, B., Fujii, I., 2001. An analysis of isothermal phase change of
system. However, for final selection, it is also essential to phase change material within rectangular and cylindrical containers.
consider other factors such as temperature level of the Solar Energy 70, 51–61.