Numerical Investigation of Phase Change Materials for Thermal Management Systems

Author(s): Taha Aldoss, David J. Ewing, Yan Zhao and Lin Ma

Source: SAE International Journal of Materials and Manufacturing , Vol. 2, No. 1 (2009), pp.
85-91
Published by: SAE International
Stable URL: https://www.jstor.org/stable/10.2307/26282737

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/10.2307/26282737?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

SAE International is collaborating with JSTOR to digitize, preserve and extend access to SAE
International Journal of Materials and Manufacturing

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
 2009-01-0171

Numerical Investigation of Phase Change Materials for Thermal

Management Systems

Taha Aldoss
Jordan University of Science & Technology

David J. Ewing, Yan Zhao and Lin Ma

Clemson University

Copyright © 2009 SAE International

ABSTRACT Q = heat added to system

CP = specific heat at constant pressure
Phase change materials (PCMs) are extensively used in CPeff = combined effective specific heat of
many engineering areas for thermal management conductor and PCM
purposes. This paper investigated the application of (UCP)MF = (UCP) of the metal foam
PCMs for vehicular systems, especially for the thermal (UCP)PCM = (UCP) of the PCM
protection of vehicle lighting systems based on light k = thermal conductivity
emitting diodes (LEDs). Lighting systems based on kMF = thermal conductivity of the metal foam
LEDs offer many advantages, however, also pose a ks = thermal conductivity of solid metal foam
smaller margin of error for thermal management. This kf = thermal conductivity of PCM at its
paper analyzed the combined use of PCMs with metal respective solid or liquid phase
foam for cooling systems. The cooling performance was kPCM = thermal conductivity of the PCM
studied numerically under different porosity values of the keff = combined effective thermal conductivity of
metal foam, and different boundary conditions. The conductor and PCM
cooling performance was also compared to a solid metal r = area ratio of the solid fiber to void area
sink system (SMS) and was found to offer several href = reference enthalpy
distinct cooling characteristics. hsens = sensible enthalpy
Tref = reference temperature
NOMENCLATURE L = latent heat
Lf = half length of fiber of metal foam
b = half thickness of bump in hexagonal Sh = source term to correct the latent heat
structure of metal foam variation during the melting process
CMS = cooling management system 'H = heat of fusion
SMS = solid management system Htot = total enthalpy of material
T = temperature 't = time step
TM = melting temperature of PCM  FC = forced convection condition
TS = solidus temperature of PCM NC = natural convection condition
TL = liquidus temperature of PCM AD = Adiabatic condition
D = Diameter of CMS and SMS unit
H = height of CMS and SMS unit
h = convective heat transfer

SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1 85

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
GREEK SYMBOLS operational time of the system where the operational
time is defined as the amount of time it takes for the
U = density system to reach its maximum operational temperature.
 UMF = density of the metal foam This operational time gain depends heavily on the PCM
 UPCM = density of the PCM mass ratio and the PCM thermophysical properties k, ,
 Ueff = effective density H, TM, and CP. The configuration of the unit containing
the PCM, such as the thermal properties and the
 H volume fraction of the PCM
distribution of a thermal conductivity enhancer that
E = liquid fraction encapsulates the PCM, also affects the system’s
performance6,7.
INTRODUCTION
It is known that the PCM’s conductivity plays a very
Phase change materials (PCMs) are extensively used in important role in the cooling system’s performance7.
many engineering applications. They are used in energy Higher conductivity is usually preferred, because it aids
storage systems1, electronics cooling systems2, building in heat distribution, more uniform PCM melting process,
materials for energy storage improvement3, heat and overall effectiveness of the CMS. Consequently, a
exchangers4, and many other similar applications. This higher conductivity of the PCM would increase the heat
paper investigated the applications of PMCs for the transfer rate and prolong the operational time of the
thermal management in vehicular systems, especially lighting system. However, many PCMs have relatively
the thermal management of vehicle lighting systems low conductivity. For example, organic PCMs exhibit
based on light emitting diodes (LEDs). Lighting systems very low relatively conductivity, even though they, in
based on LEDs offer many advantages, including comparison with metallic PCMs, are less corrosive, more
improved energy efficiency and potential for weight readily available, and less costly. In order to improve the
reduction. However, such lighting systems also pose a overall effective conductivity of the CMS to leverage the
smaller margin of error for thermal management advantages of such organic PCMs, many researchers
because LEDs can be permanently damaged if their suggest filling the PCM into a honeycomb structure or
operation temperature exceeds a critical temperature. 8, 9
mixing it with metal foam conductors . However, with a
Furthermore, the output power of LEDs varies fixed size for the CMS unit, increasing the effective
significantly with the operation temperature. Therefore, a conductivity of the system will decrease the effective
better thermal management technique is desired for mass of the PCM material. This, in turn, reduces the
such light systems. heat storage potential of the system, thereby, reducing
the operational time of the lighting system. Hence, this
A cooling management system (CMS) based on PCMs paper analyzed the optimal volume fraction of the PCM
appear very attractive for this purpose because a PCM for attaining the maximum operational time at a fixed
essentiallly behaves as a thermal switch. When the size for the CMS unit.
operation temperature of the lighting systems begins to
exceed that of the melting temperature of the PCM, the PROBLEM FORMULATION
temperature of the system stops increasing until the
PCM is completely melted. Therefore, CMS based on During the phase change process (melting or
PCMs show good promise to tightly control the solidification), the PCM encapsulated in a porous
temperature of LEDs. But as described above, the material, in this case, metal foam (MF), can exist in three
operation time of the CMS is fundamentally limited by states: solid, liquid, and a two phase mixture.
the melting time of the PCM, which was a key parameter Additionally, the thermal properties of a PCM-MF matrix
to be investigated in this study. The investigation was are different from the constituent properties. To simplify
focused on maximizing the operation time of the CMS the mathematical model, the PCM–MF combination can
while still operating under specified sizing and other be treated as a body of uniform equivalent physical and
restraints.
thermal properties—principally CP, U and k of the PCM
and MF. The effective properties of the mixture are
More specifically, PCMs are characterized by their TM,
calculated based on the volume ratio, Hof the PCM
H, and their TS and TL (Other properties may need to
material in the mixture, as follows:
be considered for specific applications, e.g., , chemical
stability, safety, and flammability5). Once the operating
temperature of the system being cooled reaches the TL (CP )eff [(1- H )( U CP ) MF  ( U CP ) PCM ] / Ueff (1)
of the PCM, the PCM begins to melt and stores the
thermal energy that the system releases during this
process. The amount of thermal energy that the PCM where Ueff =(1- H ) U MF  H U PCM (2)
can store depends on its H and amount. During the
melting process, the PCM and, therefore, the system and as derived in10
temperature remain nearly constant. The duration of the
melting process delays the system from reaching its
maximum operational temperature, which prolongs the

86 SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
 1 T
§ § r b Lf ·· hsens href  ³ c p dT (8)
¨ ¨ ¸¸ Tref
¨ ¨ k f  1 3 1  b L f ks  k f ¸¸
¨ ¨ ¸¸
¨§ 2 ·¨ 1 r b Lf ¸¸ (3)
The latent heat content of the PCM (a mixture of melted
¨¨ ¸¸ liquid and unmelted solid), L, can now be written in
k eff ¸¨ 
¨ © 3 ¹ ¨ k f  2 3 b L f ks  k f ¸¸ terms of L and E as follows:
¨ ¨ ¸¸
¨ ¨ 3 2  b Lf ¸¸
¨ ¨ ¸¸ 'H EL (9)
¨ ¨ k f  4r 3 3 b L f k s  k f ¸¸
© © ¹¹
Obviously, the latent heat content, L, varies between
zero (for a pure solid) and L (for a liquid).
where b/Lf is given by the following expression
For solidification/melting problems, the energy equation
can be written as12
2 § § 4 ··
r  r 2  1  H ¨ 2  r ¨1  ¸¸ w U C PT wT ’ ˜ k ’T  S h (10)
b 3 © © 3 ¹¹
(4)
Lf 2§ § 4 ·· Where Sh is the source term to account for the atent heat
¨ 2  r ¨1  ¸¸
3© © 3 ¹¹ variation during the melting process. It is represented
by:
Or, in a more simplified yet reasonably accurate form,
keff can be calculated as follows11: Sh  U w 'H wT (11)

keff =(1- H )kMF  H k PCM (5) And H is calculated utilizing the lever rule from
Equation (6).
This analysis performed in the paper was based on Now, temperature can be solved for by the interaction
Equation (5). Also, instead of tracking the liquid-solid between the energy equation and the liquid fraction
front explicitly, the enthalpy-porosity formulation can be equation.
used in this type of application. The two phase zone is
treated as a porous zone with porosity equal to E, the THE PHYSICAL PROBLEM AND THE PCM
liquid fraction, which is defined in the following Equation
MATERIAL PROPERTIES
(6):
The physical problem was defined according to the
E 0 if T  Ts design constraints of an LED light system. The heat
E 1 if T ! Tl (6) input from the vehicle’s lighting system was given to be
a total of 73.5 W. The maximum permissible
T  Ts temperature of the lighting system was not to exceed
E if T  Ts 90oC and the ambient temperature was given to be
Tl  Ts
27oC. The problem is assumed to be conduction-
dominated within the CMS unit, thus the internal natural
With this definition (also referred to as the lever rule12), convective heat transfer effect of the PCM can be
an enthalpy-porosity technique can be used for modeling neglected.
the melting process9. The two phase zone is a region in
which E of the PCM lies between 0 and 1, with 1 The geometry of the CMS consisted of a cylindrical
corresponding to the PCM being fully melted and 0 container of aluminum material. The container contains
corresponding to the PCM being fully solid. The two a mixture of a PCM suspended in an aluminum metal
phase zone is modeled as a “pseudo" porous medium in foam. Figure 1 shows the sketch of the CMS unit. The
which the porosity decreases from 1 to 0 as the material diameter (D) and height (H) of the domain were 10 cm
melts. When the material has fully melted in the cell, the and 10 cm, respectively.
porosity becomes 0.
Three operating conditions were considered in this
The enthalpy of the material is computed as the sum of investigation, by assuming three different boundary
hsens, and H: conditions. In the first case the unit is assumed to be
fitted outside the engine-hood, exposed to an ambient
H tot hsens  'H (7) temperature of 27oC, and air velocity of 40 kilometers
per hour (the average speed of the vehicle). This gave
an average convective heat transfer rate of 14 W/m2-K.
where

SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1 87

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
This case was considered to be the forced convection Table 1, and are assumed to remain constant over the
(FC) condition. entire temperature range encountered in the operation.

NUMERICAL ANALYSIS
Line of Al
The commercially available software programs Fluent
symmetry container
and ANSYS were used to perform a control-volume-
Bottom Middle Top based technique that converts the governing equations
into algebraic equations to be solved. This control-
volume technique consisted of integrating the governing
D equations about each control-volume, yielding discrete
Q equations that conserve each quantity on a control-
volume basis13.
PCM
H
suspended in A point implicit (Gauss-Seidel) linear equation solver
metal foam was used in conjunction with an algebraic multigrid
Figure 1: Problem schematic. (AMG) method to solve the resultant scalar system of
equations for the dependent variable in each cell.

For transient simulations, the governing equations were

In the second case the unit is assumed to be installed discretized in both space and time. Temporal
outside the engine-hood, but with the vehicle at rest. In discretization involved the integration of every term in
this case, the temperature that the unit was exposed to the differential equations over a time step t.
is the ambient temperature of 27oC. Natural convection
would be present in this case, therefore, a value of h = 5 A fully implicit integration scheme was used for
W/m2-K was then determined. This case was considered integration of the unsteady term (i.e., using “the future
to be the natural convection (NC) condition. time level”). The advantage of the fully implicit scheme is
that it is unconditionally stable with respect to the time
Finally, the third case assumed the unit was to be step size. A grid-independence study was carried out to
installed under the engine-hood. The temperature of the test the sensitivity of the solution to the grid size.
environment under the hood (engine temperature) would
be extremely high, much higher than the maximum RESULTS AND DISCUSSION
allowable temperature of the protected lighting system.
Thus, the unit must be completely isolated. This is It is very important to select a PCM with a TM that is
considered to be the worst operating condition possible, close to the maximum allowable temperature of the
since the unit is operating without any external cooling. protected system to attain its best performance14.
This case was considered to be the adiabatic (AD) According to our survey, the PCMs listed in Table 1 are
condition. among the most suitable organic PCMs which satisfy the
above condition for the thermal management of the LED
Table 1: Properties of some PCM materials and Al Metal lighting system.
Foam
Use of a different PCM involves a tradeoff among
U CP 'H TM k
various factors, e.g., the cooling effectiveness, the
Material 3 o
[kg/m ] [J/kg-K] [J/kg] [ C] [W/mͼK]
operational time, and the thermal protection. For
MF:
Al
2700 963 - - 180 example, as can be seen, ClimSel C70 clearly has the
PCM: highest 'H, k, and (U CP). However, its TM is too low
Climsel 1700 3600 280800 70 0.70 when compared with the target protection temperature
C 70 (90 0C), which will be overprotective in practice.
PCM: Thermasorb-175 and RT80 have a much closer TM to
Therma- 930 2000 200000 79 0.25 that of the failure temperature of the lighting system.
sorb-175 However, these two (along with RT80) have much lower
PCM:
920 2400 175000 81 0.20 values for (U CP) and 'H.
RT80
PCM: In this study, all four PCMs were tested over a range of
Triacont- 810 2050 251000 65 0.23
H'sin order to see which PCM can give the longest
ane
operational time for the lighting system. The results of
this investigation were plotted in Figure 2. Figure 2
clearly shows that all four PCM-MF mixtures yield about
The thermophysical properties of the PCM and the the same duration of operational time at a volume
aluminum metal foam used in this study are listed in
fraction of H = 0.75.

88 SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
 As expected, the FC boundary conditions corresponded
to the longest melting time (~ 12,000 seconds in this
case), which defines the operational time of the CMS.
Under the AD boundary condition, heat transfer out of
the system is zero. Thus, the time needed to melt the
PCM was the shortest. It is desirable to prolong the
operational time, and one possible strategy involves
using a different geometry of the CMS.

Figures 4, 5, and 6 show the transient temperature

history of the CMS at different locations under the three
boundary conditions, the FC, NC and AD cases,
respectively. These figures illustrate the rise of the
temperature as a function of operating time at the
bottom, the middle, and the top sections of the CMS. As
expected, the bottom section of the unit will reach TM
Figure 2: Maximum operational time achieved for various H's
of four different PCM’s. and the failure temperature first, because it is in direct
contact with the heat source (i.e., the LED array).
Therefore, the temperature at the bottom section
Figure 2 also indicates that ClimSel C70 yielded the essentially defines the operational time of the CMS.
longest operational time in comparison with the rest of
PCMs. Therefore, ClimSel C70 was chosen for the rest
of the study to be discussed below.

To illustrate the operation of the CMS more closely, we

calculated the liquid fraction (E) at different operational
times. Time zero was defined as the time when the
LEDs were turned on to generate a heat flow (Q) of 73.5
W. Figure 3 shows the variation of Eversus the
operational time of the PCM-MF mixture at the three
types of boundary conditions as mentioned above: the
forced convection, FC, the natural convection, NC, and
the adiabatic, AD. Figure 3 shows that under all
boundary conditions, the PCM was fully melted after a
certain duration of operation under the given conditions
(heat load, driving velocity, geometry of the CMS, etc.).
Figure 4: Temperature at different locations within the CMS
under the FC boundary condition.

Figure 5: Temperature at different locations within the CMS

Figure 3: The time evolution of E of the PCM. under the NC boundary condition.

SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1 89

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
 Figure 6: Temperature at different locations within the Figure 8: Operational time comparison between SMS and
CMS under the AD boundary condition. CMS at all three boundary conditions.

Figure 7 summarizes the time at which the critical on a solid metal sink system (SMS). Figure 8 compares
temperature was reached under different boundary the history of temperature rise at the bottom section
conditions at different locations. As can be seen, under using the PCM-based method (the CMS case) and the
the FC condition, the most favorable cooling condition, cooling method based on a solid unit of aluminum (the
the bottom section reached the critical temperature after SMS case). In this comparison, both the CMS and the
13,490 seconds (3.7 hours) of operation; under the NC SMS cases were assumed to have the same geometry
condition, the bottom section reached the critical as defined in Figure 1; and both cases were assumed to
temperature after 11240 seconds (3.1 hours) of operate under the same conditions as defined
operation; and finally, under the AD boundary condition, previously. As Figure 8 shows, with the SMS system,
after 10790 seconds (2.99 hours) of operation. temperature increased monotonically until it reached an
asymptotic value (corresponding to the steady state
value). In contrast, with the CMS system, temperature
first increased to the melting temperature of the PCM,
then remained almost constant around the melting
temperature, and finally started increasing again after
the PCM was completely melted. This difference clearly
elucidates the functionality of the PCM system as a
thermal switch to tightly control the temperature.

CONCLUSIONS AND DISCUSSION

The use of PCMs was investigated for the thermal

management of LED lighting systems. The advantages
of the PCMs to tightly control temperature below the
failure temperature of the LEDs were clearly illustrated.
The performance and limitations of the PCM-based
cooling methods for such applications were examined
Figure 7: The operational times for all three boundary
using extensive numerical simulations. The use of metal
conditions at three different locations within the CMS. foam significantly improved the cooling performance,
and the optimal MF fraction was determined. The
Again, one possible approach to prolong the operational cooling performance was also compared to a solid metal
time of the CMS involves varying the geometry of the sink system and was found to offer several distinct
CMS. Enlarging the size of the CMS represents a cooling characteristics.
simple, yet effective method. For example, our
calculations showed that with a unit size of 20 cm Two conclusions can be drawn from this study. First, a
diameter and 20 cm height, an operational time of 18 well-chosen PCM provides better control of temperature
hours can be obtained under the FC boundary condition. when compared with a metallic heat sink. Second, the
Research is underway to optimize the operational time operational time of the PCM-based method depends
under various geometrical constraints. sensitively on several factors, especially the size and
geometry of the cooling unit. Research is underway to
Lastly, we compared the distinct cooling characteristics optimize the design of PCM-based cooling systems
between the PCM-based method and the method based under geometrical constraints.

90 SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms
Current ongoing research involves investigating the 8. Calmidi, V. V., Mahajan, R. L. (1999). The Effective
performance of the PCM-based cooling systems under Thermal Conductivity of High Porosity Fibrous Metal
more realistic boundary conditions, and validating the Foams. ASME Journal of Heat Transfer, 121.
numerical analysis via experimental tests. In reality, the 9. Henze, R. H., Humphrey, J. A. (1981). Enhanced
boundary conditions are characterized by mixed modes Heat Conduction in Phase Change Thermal Energy
of heat transfer, and transient temperatures. Though the Storage Devices. International Journal of Heat and
rigorous is undergoing, many insights can be obtained Mass Transfer, 24.
from the results presented here. For example, the case 10. Joshi, Y. K., Pal, D. (1995). Application of Phase
with an adiabatic boundary condition represents the
Change Materials to Thermal Control of Electronic
worst scenario under a constant under-hood
Modules: A Computational Study. Proceedings of
temperature. When the temperature fluctuates (as long
the International Packaging Conference, 10.
as it does not exceed the temperature studied here), the
operational time will be prolonged. A prolonged 11. Kaviany, M., “Principles of Heat Transfer in Porous
operational time can also be expected when mixed Media”. New York. Pringer-Verlag, 1995.
modes of heat transfer are considered. 12. Brent, A. D., Voller, V. R., Reid, K. J. (1988).
Enthalpy-Porosity Technique for Modeling
REFERENCES Convection-Diffusion Phase Change: Application to
the Melting of a Pure Metal. Numerical Heat
1. Sari, A., Kaygusuz, K. (2001). Thermal Energy Transfer Part A: Applications 13.
Storage System Using Stearic Acid as a Phase 13. Patankar, S., “Numerical Heat Transfer and Fluid
Change Material. Solar Energy, 71(6). Flow”. Washington, D.C., Hemisphere, 1980.
2. Cho, K., Choi, M. (2000). Experimental Study on the 14. Peippo, K., Kauranen, P., Lund, P. D. (1991).
Application of Paraffin Slurry to High Density Multicomponent PCM Wall Optimized for Passive
Electronic Package Cooling. Heat and Mass Solar Heating. Energy and Buildings, 17.
Transfer, 36.
3. Fang, X., Zhang, Z. (2006). A Novel Montmorillonite-
Based Composite Phase Change Material and its
Applications in Thermal Storage Building Materials.
Energy and Buildings, 38.
4. Liu, Z., Wang, Z., Ma, C. (2006). An Experimental
Study on the Heat Transfer Characteristics of a Heat
Pipe Heat Exchanger with Latent Heat Storage. Part
II: Simultaneous Charging/Discharging Modes.
Energy Conversion and Management, 47.
5. Salyer, I. O., Sircar, A. K. (1997). Review of Phase
Change Materials Research for Thermal Energy
Storage in Heating and Cooling Applications of the
University of Daton from 1982 to 1996. International
Journal of Global Energy Issues, 9(3).
6. Henze, R. H., Humphrey, J. A. (1981). Enhanced
Heat Conduction in Phase Change Thermal Energy
Storage Devices. International Journal of Heat and
Mass Transfer.
7. Akhilesh, R., Narasimhan, A., Balaji C. (2005)
Method to Improve Geometry for Heat Transfer
Enhancement in PCM Composite Heat Sinks.
International Journal of Heat and Mass Transfer, 48.

SAE Int. J. Mater. Manuf. | Volume 2 | Issue 1 91

This content downloaded from

196.200.165.13 on Mon, 27 Nov 2023 19:54:17 +00:00
All use subject to https://about.jstor.org/terms