Vous êtes sur la page 1sur 9

International Journal of Heat and Fluid Flow 38 (2012) 159–167

Contents lists available at SciVerse ScienceDirect

International Journal of Heat and Fluid Flow


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

Numerical investigation of liquid flow with phase change nanoparticles


in microchannels
Awad B.S. Alquaity a, Salem A. Al-Dini a, Evelyn N. Wang b, Bekir S. Yilbas a,⇑
a
Department of Mechanical Engineering, King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia
b
Device Research Laboratory, Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

a r t i c l e i n f o a b s t r a c t

Article history: A numerical solution is introduced to investigate the effect of laminar flow with a suspension of phase
Received 10 February 2012 change material nanoparticles (PCMs) in a microchannel. The nanoparticle suspension consisting of lauric
Received in revised form 20 June 2012 acid nanoparticles in water is introduced into a microchannel of 50 lm height and 35 mm length, where a
Accepted 9 October 2012
constant heat flux is applied to the bottom wall. Mass, momentum and energy equations are solved
Available online 3 November 2012
simultaneously using a fluid with effective thermo-physical properties. The effect of various parameters
including mass flow rate (1  105–4  105 kg/s), heat flux (8000–20,000 W/m2) and particle volume
Keywords:
concentrations (0–10%) on the thermal performance is investigated using effectiveness ratio, perfor-
Microchannel
Phase change material
mance index, and Merit number. The study is extended to include the optimum channel length for
Nanoparticles improved thermal performance. For a given particle concentration, an optimum heat flux to mass flow
Laminar flow rate ratio exists that leads to the maximum effectiveness ratio of 2.75, performance index of 1.37 and
Merit number of 0.64. Such a study facilitates understanding the parametric space to optimize heat trans-
fer in microchannels for applications such as thermal management and energy conversion devices.
Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction 1999; Rao et al., 2007; Wang et al., 2007, 2008; Chen et al., 2008;
Zeng et al., 2009). More recently, interest in the thermal perfor-
The development of next generation microchips, microproces- mance of PCM slurries in microchannels have emerged for applica-
sors and other small scale high heat generating applications is tions in microchannel heat exchangers and cooling of electronic
constrained by the issue of effective heat removal (Sabbah et al., devices (Sabbah et al., 2008; Xing et al., 2005; Kondle et al., 2009;
2008). The cooling capacity required to reach switching speeds Kuravi et al., 2009, 2010). Sabbah et al. (2008) performed a three-
needed for next generation computing devices is of the order of dimensional numerical study on the performance of microchannel
105 W/cm3 which cannot be met by current liquid cooling systems heat sinks using micro-encapsulated PCMs and considered the ther-
including microchannel heat sinks (Xing et al., 2005). A promising mal resistance of the heat sink walls while incorporating tempera-
method to meet the electronic cooling demands for next generation ture dependent physical properties for the PCM slurry. Xing et al.
devices, by enhancing heat storage capacity of the heat transfer (2005) evaluated the performance of liquid flow with PCM particles
fluid, is to introduce PCM particles in the fluid. The phase change in circular microchannels. The conservation equations for the parti-
of the PCM particles in the fluid significantly enhances its heat stor- cle and liquid phase were solved separately while considering the
age capacity and thus increases its ability to absorb high heat fluxes. effects of particle–particle interaction and the particle depletion
Therefore, a quantitative assessment of the heat storage capacity boundary near the wall. A particular Reynolds number and wall
increase of the fluid in a microchannel flow with the presence of heat flux was found to achieve maximum heat transfer enhance-
particles becomes essential. In addition to the heat storage capacity ment with PCM particles. Kondle et al. (2009) numerically studied
increase, it is also necessary to determine the effect of particles on heat transfer characteristics of PCMs in a laminar flow for circular
the pressure drop and entropy generation inside the microchannel. and rectangular microchannels. The carrier fluid and particles were
Significant efforts have focused on using PCMs for improving the modeled using homogeneous model while a specific heat model
thermal performance of the carrier fluid in the past decade (Sabbah was used for the phase change of particles. Kuravi et al. (2009) used
et al., 2008; Xing et al., 2005; Kondle et al., 2009; Kuravi et al., 2009, a similar homogeneous model to study numerically the thermal
2010; Goel et al., 1994; Roy and Avanic, 1997; Yamagishi et al., performance of nano-encapsulated PCM slurry in microchannels.
The temperature and velocity fields were obtained in three dimen-
⇑ Corresponding author. Tel.: +966 3 8604481; fax: +966 3 860 5223. sional domain and the model included the microchannel fin effect
E-mail address: bsyilbas@kfupm.edu.sa (B.S. Yilbas). along with the longitudinal conduction along the microchannel

0142-727X/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ijheatfluidflow.2012.10.001
160 A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167

Nomenclature

Aflow cross-sectional area of the microchannel (m2) Ste Stefan number


C volume concentration of nanoparticles T temperature (K)
cp specific heat (J kg1 K1) Tref reference temperature taken as 298 K
cp,S specific heat of particles in solid state (J kg1 K1) Tsolidus lower melting temperature (K)
cp,L specific heat of particles in liquid state (J kg1 K1) Tliquidus upper melting temperature (K)
Dh hydraulic diameter of the microchannel (m) U x component of velocity (m s1)
I_ rate of irreversibility (W) V velocity (m s1)
K thermal conductivity (W m1 K1)
L latent heat of melting of particles (J kg1) Greek symbols
m_ mass flow rate (kg s1) l dynamic viscosity (N s m2)
P pressure (Pa) P density (kg m3)
Prb Prandtl number, Prb ¼ cpb lb =kb
q00 heat flux (W m2) Subscripts
Q heat transfer rate (W) B bulk fluid
Reb Reynolds number, Reb ¼ qb v Dh =lb F carrier fluid
S000
gen volumetric entropy generation (W m3 K1) p particle
S000
gen;av g averaged volumetric entropy generation rate (W m3 -
K1)

length. Kuravi et al. (2010) investigated the heat transfer perfor- 2. Model framework
mance of water-based microencapsulated PCM slurry in manifold
microchannels both experimentally and numerically. Their findings A schematic of the microchannel incorporated in the model
revealed that the slurry performance was poorer as compared to study is shown in Fig. 1. In this case, a microchannel of constant
single fluid. They also performed a parametric study with nano- height (50 lm, H) and length (35,000 lm, L) is defined in the FLU-
encapsulated PCM slurry flow. The results of the parametric study ENT™ simulations. For 3D simulations, width of the microchannel
showed that using narrower channels and PCM particles with is considered to be 2 mm, which is 40 times the microchannel
higher thermal conductivity improved the thermal performance height. The carrier fluid with nanosized particles is assumed to en-
of PCM slurry as compared to a single fluid. The laminar hydrody- ter into the microchannel at a temperature just below the melting
namic and heat transfer characteristics of suspension flow with temperature of the particles. A constant heat flux is applied at the
micro-nano-size phase-change material (PCM) particles in a micro- bottom wall, which heats the carrier fluid and particles. After tra-
channel were investigated by Hao and Tao (2006). They indicated versing a certain length of the microchannel, the particles undergo
that the heat transfer enhancement took place in wall region due phase change. The phase change of the particles plays an important
to the presence of PCM particles. Liquid–liquid two-phase flow in role in decreasing the temperature rise of the fluid as compared to
pore array microstructured devices for scaling-up of nanoparticle the case with no phase change particles in the flow system and
preparation was examined by Li et al. (2009). The findings revealed thereby increases the thermal storage capacity of the fluid.
that the particle size was decreased with the increase of the droplet In order to formulate the flow and heat transfer problem, the
size in both the drop flow region and the disk flow region whereas it following assumptions were made:
had a reverse trend in the transition region. Although significant
studies using homogeneous model have been performed in the  The flow of the bulk fluid inside the microchannel is steady and
past, none of them have presented the entropy generation due to laminar.
the addition of PCM particles. For this purpose, we have defined  The fluid is Newtonian up to particle volume concentrations of
the Merit number to incorporate the thermodynamic irreversibility 10% (Kuravi et al., 2009).
in the flow system. Moreover, in this paper, we have performed a  The shell encapsulating nanoparticles will be thin, so its effect
comprehensive study of the influence of various parameters, has been neglected (Kuravi et al., 2009).
including mass flow rate, heat flux, and particle volume concentra-  The particles and carrier fluid are assumed to have the same
tions, on the thermal performance of the slurry. We use water as the temperature and velocity in the microchannel (Kuravi et al.,
carrier fluid and lauric acid as the PCM particles with different vol- 2009).
ume concentrations ranging from 0% to 10% in the analysis. Homo-  The distribution of particles inside the microchannel is homoge-
geneous model was used to simulate temperature and flow fields. neous (Kuravi et al., 2009).
The thermo-physical properties of the PCM particles are assumed  The PCM particle melts instantaneously once the melting tem-
to be constant during the simulations and are given in Table 1. perature is reached (Kuravi et al., 2009).
The thermo-physical properties of carrier fluid are assumed to be
temperature dependent. 2.1. Homogeneous model

Table 1 The governing equations of mass, momentum and energy are


Thermophysical properties of PCM particles. solved using the appropriate effective thermo-physical properties
Fluid Density Specific heat Latent heat Thermal of the bulk fluid. The equations governing laminar flow for the bulk
(kg/m3) (kJ/kg K) (kJ/kg) conductivity fluid are shown below:
(W/m K) Continuity equation:
Particle (solid) 1007 1.76 211 0.147
Particle (liquid) 862 2.27 – 0.147 v ¼0
r~ ð1Þ
A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167 161

Fig. 1. Schematic of microchannel used in the FLUENT simulations where heat flux is applied to the bottom wall, and H is the microchannel height, L is the microchannel
length. The side walls are assumed to be adiabatic.

Momentum equation: For T solidus < T p < T liquidus :

v Þ ¼ rp þ lb r2 v
v~
r  ðqb~ ð2Þ   
c þc L
c q p;S 2 p;L þ T þ ð1  cÞðqcp Þf
liquidus T solidus p
Energy equation: cpb ¼ ð8Þ
qb
v cpb TÞ ¼ r  ðkb rTÞ
r  ðqb~ ð3Þ For T P > T liquidus :

2.2. Bulk fluid thermo-physical properties cðqcp;L Þp þ ð1  cÞðqcp Þf


cpb ¼ ð9Þ
qb
2.2.1. Viscosity
The mass flow rate of the bulk fluid at the inlet was varied
The introduction of 50 nm diameter of PCM nanoparticles into
between 105 kg/s and 4  105 kg/s. The heat flux at the bottom
the carrier fluid increases its viscosity. The bulk viscosity is calcu-
wall of the microchannel was varied between 8000 W/m2 and
lated as (Vand, 1945):
20,000 W/m2. The temperature of the particles and carrier fluid is
315 K at the inlet, which is less than the melting temperature of
lb ¼ ð1  c  1:16c2 Þ2:5 lf ð4Þ the PCM particles. At the microchannel outlet, pressure outlet
The above correlation for bulk viscosity was found to agree well boundary condition is considered. The pressure outlet boundary
with experimental data presented in (Fang et al., 2009) for nanopar- condition assumes an absolute pressure of 1 atm at the outlet
ticle volume concentrations below 11%. and zero diffusion fluxes in the direction normal to the exit plane
for all flow variables (i.e., the velocity and temperature) except
2.2.2. Thermal conductivity pressure.
The Maxwell model (Maxwell, 1954) is used for calculating the
bulk thermal conductivity, which is:
3. Numerical solution
2 þ kp =kf þ 2cðkp =kf  1Þ
kb ¼ kf ð5Þ
2 þ kp =kf  cðkp =kf  1Þ The control volume approach is used for the discretization of
governing equations using FLUENT 12.1.2 C.F.D. code (FLUENT,
Recently a benchmark experimental study was carried out to
2010). All variables are computed at each grid point except the
measure the effective thermal conductivity of nanofluids. The
velocities, which are determined midway between the grid points.
study concluded that the effective medium theory developed for
A staggered grid arrangement is used in the present study, which
dispersed particles by Maxwell (1954) is in good agreement with
links the pressure through the continuity equation and is known
the experimental data (Buongiorno et al., 2009).
as SIMPLE algorithm (Patankar, 1980). The pressure relationship
between continuity and momentum is established by transforming
2.2.3. Density
the continuity equation into a Poisson equation for pressure. The
The density of the bulk fluid can be calculated using mass bal-
convergence criterion for the scaled residuals is set to 106 for con-
ance as (Sabbah et al., 2008):
tinuity and 109 for energy equation. Further, surface monitor of
temperature at the outlet was used to ensure convergence. Second
qb ¼ cqp þ ð1  cÞqf ð6Þ
order upwind schemes were used for discretization of momentum
and energy equations.
2.2.4. Specific heat
In order to account for the phase change of the particles, a spe- 3.1. Grid
cific heat model is used (Kondle et al., 2009). In the specific heat
model, the particle’s phase change is modeled by varying the A numerical mesh generator was used to create the geometry
specific heat capacity of the particles across the solidus and the liq- and mesh the domain using hexahedral elements. A fine mesh
uidus temperatures. Since the melting temperature of lauric acid is was created near the walls to capture the large gradients normal
317.2 K, the melting range of PCM particles is assumed to 316.7– to the flow direction. Three different grid resolutions, 20 
317.7 K. The specific heat of the bulk fluid can be calculated using 14,000 (Grid 1), 35  18,000 (Grid 2) and 50  35,000 (Grid 3) were
the energy balance (Sabbah et al., 2008): used to obtain a grid-independent solution. The maximum differ-
For T p < T solidus : ence in Nusselt between the three grids was 0.14. Therefore, Grid
2 was used for further simulations. After the grid independence
cðqcp;S Þp þ ð1  cÞðqcp Þf test, Grid 2 was also validated against analytical Nusselt number
cpb ¼ ð7Þ and fanning friction factor results for 2D parallel plate subjected
qb
to constant heat flux boundary condition at the top and bottom
162 A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167

Table 2
Comparison between constant properties and temperature dependent properties.

Constant Temperature Percentage


properties dependant properties difference (%)
DT (K) 33.69 33.7 0.04
DP (Pa) 10621.7 8193.8 29.6

walls (Kandlikar, 2006). The difference in Nusselt numbers was of


the order of 2%, while the numerical value of friction factor
matched exactly with that of the analytical value.

3.2. Temperature dependence

As indicated earlier, temperature dependent thermo-physical


properties are used for the carrier fluid. Table 2 shows a compari-
son of the temperature rise and pressure drop in case of using con-
stant thermo-physical properties and temperature dependent Fig. 3. Comparison of Nusselt number obtained using homogeneous model with
thermo-physical properties. The temperature rise shows a negligi- experimental data (Chen et al., 2008).
ble difference in both cases, but the use of constant properties
leads to pressure drop being over-estimated by about 30%. The
high pressure drop caused by using constant thermo-physical It can be observed that the wall temperature predicted by the
properties is due to the viscosity of water, which is highly depen- current model compares well with experimental data.
dent on temperature. The significant over-estimation of pressure Fig. 3 compares the Nusselt number for two Reynolds number
drop motivates performing simulations using temperature depen- and Stefan number combinations, i.e., Reynolds numbers of 691
dent properties for the carrier fluid despite the longer computation and 1418, and Stefan numbers of 1.09 and 1.38. It can be observed
time. that the Nusselt number predicted by the current homogeneous
model compares well with the experimental data.
3.3. Model validation The non-dimensional length used in Fig. 3 is defined as:

2x
Due to the lack of comparable experimental data for PCM slurry xþ ¼ ð11Þ
Dh Reb Prb
flow in microchannels, the homogeneous model used in this study
is validated through comparing the predictions with the experi- where Reb is Reynolds number and Prb is the Prandtl number of the
mental data obtained for PCM slurry flow in a 3.14 mm diameter bulk fluid.
circular tube (Goel et al., 1994) and in a 4 mm diameter circular
tube (Wang et al., 2008). Fig. 2 shows the comparison of wall tem-
4. Results and discussion
perature for Stefan number of 3 predicted using homogeneous
model with the experimental data for 10% particle volume concen-
We investigated the effect of particle volume concentration,
tration. Stefan number is defined as the ratio of sensible heat
mass flow rate and heat flux on the thermal performance of bulk
capacity of the slurry to its latent heat capacity and is given as
fluid inside the microchannel. The effectiveness ratio (Xing et al.,
(Goel et al., 1994):
2005), performance index (Xing et al., 2005), and Merit number
cpb q00 Dh qb are used, where the heat flux applied to different volume concen-
Ste ¼ ð10Þ
2kb cLqp trations of PCM fluids is the same, while the heat flux applied to
water is varied to maintain the same temperature rise as that of
PCM fluid. This approach helps compare the heat transfer increase
in the PCM slurry for the same temperature rise as that of water.
Fig. 4 shows temperature contours along the channel length for
the case with no PCM particles and with 10% particle volume con-
centration. The temperature contours shown in the figures corre-
spond to the case of mass flow rate of 4  105 kg/s and heat flux
of 16,000 W/m2. It can be observed from the figures that increasing
the concentration of PCM particles contributes to temperature de-
crease in the channel towards the channel exit by 4.12 K (61%
reduction in temperature), which is attributed to the latent heat
of fusion associated with the phase change of the PCM particles.

4.1. Effectiveness ratio

The effectiveness ratio is defined as the ratio of heat transfer


rate of bulk fluid to the heat transfer rate of carrier fluid for the
same temperature rise from inlet to exit of the microchannel.
Therefore, the effectiveness ratio gives an indication of the increase
in heat transfer of the PCM slurry for the same temperature rise as
Fig. 2. Comparison of wall temperature obtained using homogeneous model with that of the carrier fluid. For applications having restriction on tem-
experimental data (Goel et al., 1994). perature rise of the heat transfer fluid, effectiveness ratio can be
A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167 163

Fig. 4. Temperature contour of carrier fluid along the channel length.

used to evaluate the enhancement in heat transfer which can be


obtained by using PCM particles in the carrier fluid. The effective-
ness ratio is defined as (Xing et al., 2005):

Qb
eeffectiv eness ¼ ð12Þ
Qf

where

Q b ¼ q00  Length  Width ð13Þ

_ p;f DT f
Q f ¼ mc ð14Þ
For the same carrier fluid and bulk fluid mass flow rates, the
effectiveness ratio simply reduces to a ratio of their average
specific heats. Fig. 5 shows the variation of effectiveness ratio with
heat flux to mass flow rate ratio for different particle volume con-
centrations of PCM particles. Initially, the effectiveness ratio in-
creases with increasing heat flux to mass flow rate ratio of PCM
slurry and keeps increasing until it reaches a peak value. Further
Fig. 5. Variation of effectiveness ratio with ratio of heat flux to mass flow rate of
increase in heat flux to mass flow rate ratio decreases the effective-
PCM slurry, where VF 3, VF 5, VF 7 and VF 10 represents 3%, 5%, 7% and 10% volume
ness ratio. The heat flux to mass flow rate ratio required to reach concentration of the particles, respectively.
the peak effectiveness ratio value increases with increasing volume
concentration of particles. Moreover, for higher volume concentra-
tion of PCM particles, the peak value of effectiveness ratio is higher. tio, increase in heat flux is reflected in an increase in the value of
This trend in the effectiveness ratio can be explained as follows: effectiveness ratio. The effectiveness ratio keeps increasing with
For a given mass flow rate, the effectiveness ratio is the highest increasing heat flux to mass flow rate ratio until enough heat flux
when the ratio of latent heat to sensible heat of the bulk fluid in the is applied to complete the melting of PCM particles within the
microchannel is the largest. The maximum value of latent heat to length of the microchannel. The heat flux to mass flow rate ratio re-
sensible heat ratio occurs when the particles reach the liquidus quired for complete phase change of PCM particles is larger for
temperature (upper melting temperature) at the exit of the micro- higher volume concentration of particles in the carrier fluid. Further
channel. This ensures that sensible heating occurs only near the in- increase in heat flux to mass flow rate ratio leads to an increase in
let of the microchannel, and, thus, the latent heat ratio is the the sensible heating of PCM particles after their phase change, and
maximum. For low heat flux to mass flow rate ratio values, the thus the effectiveness ratio starts to reduce. Consequently, the
phase change of PCM particles is not complete due to the low value effectiveness ratio increases initially, reaches a peak value and,
of heat flux applied. Therefore, at low heat flux to mass flow rate ra- then, decreases for all volume concentration of PCM particles.
164 A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167

The maximum effectiveness ratio of 2.75 is obtained for PCM


particle volume concentration of 10%. The corresponding value of
effectiveness ratio indicates that for the same temperature rise,
the bulk fluid at the given operating condition can store 2.75 times
the heat flux, which is stored by carrier fluid. From Fig. 5 we can
conclude that for a given volume concentration of particles, there
exists a heat flux to mass flow rate ratio at which the effectiveness
ratio is maximum. A similar observation was presented in (Xing
et al., 2005).

4.2. Performance index

The performance index is defined as the ratio of heat transfer


rate to fluid pumping power of bulk fluid to the heat transfer rate
to fluid pumping power of carrier fluid. Since the pumping power
required for heat transfer fluids is an important criterion in thermal
management applications, the performance index enables the de-
signer to evaluate the increase in heat transfer rate of the PCM slur-
ry in comparison to the increase in its pumping power requirement. Fig. 6. Variation of performance index with ratio of heat flux to mass flow rate of
Since the flow field considered is laminar, the pressure drop in the PCM slurry, where VF 3, VF 5, VF 7 and VF 10 represents 3%, 5%, 7% and 10% volume
concentration of the particles, respectively.
channel is associated with the frictional loss, which is directly re-
lated to the pump power requirements. Since the effectiveness ratio
only helps the designer in quantifying the increase in heat transfer
rate ratio is high, the increase in effectiveness ratio due to increase
due to the addition of PCM particles, the performance index gives a
in volume concentration of particles is not significant due to in-
more complete picture of the situation as it incorporates the in-
crease in the sensible heating of the bulk fluid inside the micro-
crease in pumping power requirement in addition to the increase
channel. Therefore, when the increase in effectiveness ratio is
in heat transfer obtained due to the addition of PCM particles. The
unable to offset the increase in pressure drop of the bulk fluid,
performance index is defined as (Xing et al., 2005):
the performance index decreases. This results in bulk fluid with
ðQ=PÞb high particle volume concentrations having a lower performance
Performance index ¼ ð15Þ index as observed in Fig. 6.
ðQ =PÞf
A performance index below unity indicates that the heat trans-
where the pumping power of the bulk fluid and carrier fluid are de- fer rate per fluid pumping power for the bulk fluid is lower than
fined, respectively as: the carrier fluid and, therefore, the bulk fluid should not be used
under these operating conditions. For 3% volume concentration of
Pb ¼ Dpb V b Aflow ð16Þ
particles in the slurry, the performance index of the bulk fluid for
all heat flux to mass flow rate ratios is below unity (Fig. 6), indicat-
Pf ¼ Dpf V f Aflow ð17Þ
ing that the use of bulk fluid entails a lower heat transfer rate to
fluid pumping power ratio, as compared to water. Moreover,
32Lqf V 2f Fig. 6 reveals that the performance index of the bulk fluid contain-
Dpf ¼ ð18Þ
Ref D ing 5%, 7% and 10% volume concentrations exceeds unity only
within a range of heat flux to mass flow rate ratio.
Fig. 6 shows the variation of performance index with heat flux
The maximum performance index of 1.37 is obtained for parti-
to mass flow rate ratio for different particle volume concentrations
cle volume concentration of 10%. The performance index follows a
of PCM slurry. The performance index initially increases with
similar trend as effectiveness ratio for a given particle volume con-
increasing heat flux to mass flow rate ratio, reaches a peak value
centration. From the above discussion, it can be concluded that for
and then decreases. The performance index for each particle vol-
a given volume concentration of particles, there exists a heat flux
ume concentration follows the same trend as effectiveness ratio
to mass flow rate ratio at which the performance index is the max-
with the same value of the optimum heat flux to mass flow rate ra-
imum. A similar observation was also made in (Xing et al., 2005).
tio. However, at high heat flux to mass flow rate ratio, the perfor-
mance index of the bulk fluid with higher volume concentration of
particles is lower than the performance index for lower volume 4.3. Merit number
concentrations. The behavior of bulk fluid with higher particle vol-
ume concentrations having a lower performance index, compared The thermal conductivity of the bulk fluid decreases due to the
to bulk fluid with lower particle volume concentrations, is associ- addition of PCM particles into the carrier fluid. Further, the addi-
ated with increase in temperature of the bulk fluid, which causes tion of PCM particles increases the viscosity of the bulk fluid, there-
sharp decrease in viscosity. A larger temperature rise inside the by, increasing the pressure drop in the microchannel. The decrease
microchannel translates to a lower pressure drop compared to in thermal conductivity and increase in pressure drop contribute to
the case in which a lower temperature rise of bulk fluid inside an increase in the entropy generation rate inside the microchannel.
the microchannel is encountered. The use of bulk fluid with high However, increase in volume concentration of particles increases
particle concentrations suppresses the temperature rise as well the heat flux than can be absorbed by the bulk fluid for a given
as increases the viscosity of the bulk fluid due to the presence of temperature rise. Therefore, Merit number is introduced to calcu-
high particle concentrations. The reduction in temperature rise is late the ratio of the gain versus input and losses due to the addition
due to the phase change of the particles inside the microchannel, of particles. The Merit number takes into account the entropy
which increases the effectiveness ratio with increasing particle vol- generation rate due to fluid friction and heat transfer caused by
ume concentration, however, this also results in an increase in the the addition of PCM particles. It is defined as the ratio of the gain
pressure drop of the bulk fluid. When the heat flux to mass flow in heat transfer due to the use of PCM particles to the sum of heat
A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167 165

"   2 #  2
2
kb @T @T lb @u
S000
gen ¼ 2 þ þ ð22Þ
T @x @y T @y

In Fig. 7 the Merit number is plotted against heat flux to mass


flow rate ratio for different particle volume concentrations. The
Merit number follows the same trend as effectiveness ratio with
the same value of optimum heat flux for a given particle volume
concentration. The Merit number of the bulk fluid with 10% volume
concentration of particles is the highest for all heat flux to mass
flow rate ratios considered in the simulations and it shows that
the increase in irreversibility due to the addition of PCM particles
is being offset by the gain in heat transfer for all heat flux to mass
flow rate ratios considered. While the performance index compares
the heat flux to pumping power ratio for PCM slurry with the heat
flux to pumping power ratio for carrier fluid only, the Merit num-
ber provides a comparison between the gain in heat transfer and
irreversibility due to different volume concentrations of PCM par-
ticles. The irreversibility takes into account the decrease in thermal
Fig. 7. Variation of Merit number with ratio of heat flux to mass flow rate of PCM conductivity as well as the increase in fluid friction caused by the
slurry, where VF 3, VF 5, VF 7 and VF 10 represents 3%, 5%, 7% and 10% volume addition of PCM particles to the carrier fluid. Therefore, the Perfor-
concentration of the particles, respectively.
mance Index includes the effect of increase in pressure drop only,
while the Merit number takes into account the effect of increase in
pressure drop as well as decrease in thermal conductivity of the
transferred at the bottom wall of the microchannel and the irre- bulk fluid. The highest Merit number is 0.64 and it can be observed
versibility. The Merit number is defined as: from Fig. 7 that for a given volume concentration of particles, there
exists a heat flux to mass flow rate ratio at which the Merit number
Q gain is maximum.
Merit number ¼ ð19Þ
Q b þ I_
where 4.4. Optimum length
Q gain ¼ Q b  Q f ð20Þ
As mentioned earlier, the enhancement in heat storage capacity
due to the addition of PCM particles is significant when the
I_ ¼ S000
gen;av g  v olume  T ref ð21Þ
completion of phase change of PCM particles coincides with the exit

Fig. 8. Optimum length of the microchannel.


166 A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167

Table 3 flow rate ratio for a given volume concentration of particles at


Constant C2 for different volume concentrations of PCM which the effectiveness ratio, the performance index, and the Merit
particles.
number are the maximum. The effectiveness ratio is greater than 1
Volume concentration C2 for all cases and it reaches a maximum value of 2.75. This indicates
of PCM particles (%) that for the same temperature rise, the bulk fluid can store 175%
3 1.083 more heat as compared to carrier fluid without particles. The per-
5 0.814 formance index is lower than 1 for all heat flux to mass flow rate
7 0.7
10 0.613
ratios for the PCM slurry of 3% particle volume concentration.
The maximum value of the performance index is 1.37, which re-
flects a 37% increase in heat transfer rate to fluid pumping power
of the microchannel. The completion of phase change at the exit ratio of the bulk fluid as compared to carrier fluid. The Merit num-
ensures that the PCM slurry flow undergoes sensible heating only ber follows the same trend as the effectiveness ratio, which indi-
near the inlet and thus maximum increase in heat storage capacity cates that the increase in irreversibility due to addition of PCM
is obtained. The length required for the phase change of PCM parti- particles is offset by the gain in heat transfer. The effectiveness ra-
cles, defined as optimum length (Lopt), however, depends on the heat tio, performance index and the Merit number have the same corre-
flux applied, the mass flow rate of the PCM slurry as well as the vol- sponding value of optimum heat flux to mass flow rate ratio for a
ume concentration of PCM particles in the slurry. In this study, the given particle concentration. The ratios calculated in this study,
optimum length is obtained by calculating the distance from the in- namely, effectiveness ratio, performance index and Merit number
let of the microchannel to the axial location at which phase change of guide the designer regarding the conditions to be used in order
PCM particles is completed for different heat fluxes, mass flow rates to obtain the maximum benefit by the addition of PCM particles.
and volume concentrations of particles. Using microchannels of Moreover, Lopt/Dh guides the designer in selecting the dimensions
optimum length will result in peak values of the effectiveness ratio, of the microchannel for the operating conditions used in this study.
performance index and Merit number for given volume concentra-
tion of particles. Fig. 8 shows the variation of Lopt/Dh with the differ-
ent parameters, where Dh is the hydraulic diameter of the Acknowledgements
microchannel. From Fig. 8, it can be clearly observed that increasing
the mass flow rate of the PCM slurry, while keeping volume concen- The authors acknowledge the support of Center of Excellence for
tration of PCM particles and heat flux constant increases the Scientific Research Collaboration with MIT and King Fahd University
optimum length required for phase change. Increase in volume con- of Petroleum and Minerals. Dhahran, Saudi Arabia for this work.
centration of PCM particles while keeping other parameters con-
stant also leads to increase in the optimum length. Increase in heat References
flux applied to the PCM slurry always decreases the optimum length.
Fig. 8 provides detailed design information for different heat fluxes, Buongiorno, J., Venerus, D.C., Prabhat, N., McKrell, T., Townsend, J., Christianson, R.,
Tolmachev, Y.V., Keblinski, P., Hu, L.-W., Alvarado, J.L., Bang, I.C., Bishnoi, S.W.,
mass flow rates and volume concentrations of PCM slurry for the Bonetti, M., Botz, F., Cecere, A., Chang, Y., Chen, G., Chung, S.J., Chyu, M.K., Das,
operating conditions used in this study. In general, the ratio Lopt/Dh S.K., Di Paola, R., Ding, Y., Dubois, F., Dzido, G., Eapen, J., Escher, W.,
can be written as: Funfschilling, D., Galand, Q., Gao, J., Gharagozloo, P.E., Goodson, K.E.,
Gutierrez, J.G., Hong, H., Horton, M., Hwang, K.S., Iorio, C.S., Jang, S.P.,
Lopt m_ Jarzebski, A.B., Jiang, Y., Jin, L., Kabelac, S., Kamath, A., Kedzierski, M.A., Kieng,
¼ B1  00  ðB2 þ B3 Þ ð23Þ L.G., Kim, C., Kim, J.-H., Kim, S., Lee, S.H., Leong, K.C., Manna, I., Michel, B., Ni, R.,
Dh q Patel, H.E., Philip, J., Poulikakos, D., Reynaud, C., Savino, R., Savino, R., Singh, P.K.,
Song, P., Sundararajan, T., Timofeeva, E., Tritcak, T., Turanov, A.N., Van
where B1 is a constant which depends on the dimensions of the Vaerenbergh, S., Wen, D., Witharana, S., Yang, C., Yeh, W.-H., Zhao, X.-Z.,
microchannel and constant B2 depends on the particle volume con- Zhou, S.-Q., 2009. A benchmark study on the thermal conductivity of nanofluids.
centration, thermo-physical properties of the carrier fluid and PCM J. Appl. Phys. 106, 94312–94326.
Chen, B., Wang, X., Zeng, R., Zhang, Y., Wang, X., Niu, J., Li, Y., Di, H., 2008. An
particle and the slurry inlet temperature. The constant B3 depends
experimental study of convective heat transfer with microencapsulated phase
on the melting range of PCM particle and the thermo-physical prop- change material suspension: laminar flow in a circular tube under constant heat
erties of the PCM particle and carrier fluid. In the above equation, m_ flux. Exp. Therm. Fluid Sci. 32, 1638–1646.
Fang, Y., Kuang, S., Gao, X., Zhang, Z., 2009. Preparation of nanoencapsulated phase
is the slurry mass flow rate and q00 is the heat flux applied to differ-
change material as latent functionally thermal fluid. J. Phys. D: Appl. Phys. 42,
ent volume concentrations of PCM slurry. For the conditions used in 35407–35413.
this study, the above equation (Eq. (23)) can be written in terms of FLUENT, Product Documentation, Fluent 12.1., ANSYS Inc.
heat flux, slurry mass flow rate and particle volume concentration Goel, M., Roy, S.K., Sengupta, S., 1994. Laminar forced convection heat transfer in
microcapsulated phase change material suspensions. Int. J. Heat Mass Transfer
as: 37, 593–604.
Hao, Y.L., Tao, Y.-X., 2006. A numerical model for phase-change suspension flow in
Lopt m_
¼ C 1  00  ðC 2  c þ C 3 Þ ð24Þ microchannels. Numer. Heat Transfer, Part A (Applications) 46, 55–77.
Dh q Kandlikar, S.G., 2006. Heat Transfer and Fluid Flow in Minichannels and
Microchannels. Elsevier Science.
where C1, C2 and C3 are constants and c is the volume concentration Kondle, S., Alvarado, J.L., Marsh, C., Kessler, D., Stynoski, P., 2009. Laminar flow
of PCM particles. For all volume concentration of PCM particles, forced convection heat transfer behavior of a phase change material fluid in
microchannels. In: Proceedings of the ASME 2009 International Mechanical
C1 = 5  1010 and C3 = 0.4. The value of constant C2 is different for Engineering Congress and Exposition, Lake Buena Vista, Florida, USA, November
different volume concentration of particles and is given in Table 3. 13–19.
The above Eq. (24) gives the value of Lopt/Dh within 3% error. Kuravi, S., Kota, K.M., Du, J., Chow, L.C., 2009. Numerical investigation of flow and
heat transfer performance of nano-encapsulated phase change material slurry
in microchannels. J. Heat Transfer 131, 62901–62907.
5. Conclusions Kuravi, S., Du, J., Chow, L.C., 2010. Encapsulated phase change material slurry flow
in manifold microchannels. J. Thermophys. Heat Transfer 24, 364–373.
Li, S., Xu, J., Wang, Y., Luo, G., 2009. Liquid–liquid two-phase flow in pore array
We investigated microchannel flow for water with the presence microstructured devices for scaling-up of nanoparticle preparation. AIChE J. 55,
of nanosized lauric acid PCM particles using the homogeneous fluid 3041–3051.
model. A parametric study is carried out through varying the mass Maxwell, J.C., 1954. A Treatise on Electricity and Magnetism, third ed. Dover, New
York, vol. 1, pp. 440–441.
flow rate, heat flux and particle volume concentration of the bulk Patankar, S.V., 1980. Numerical Heat Transfer and Fluid Flow. Hemisphere
fluid. It is found that there exists an optimum heat flux to mass Publishing Company, Washington, DC.
A.B.S. Alquaity et al. / International Journal of Heat and Fluid Flow 38 (2012) 159–167 167

Rao, Y., Dammel, F., Stephan, P., Lin, G., 2007. Convective heat transfer Wang, X., Niu, J., Li, Y., Zhang, Y., Wang, X., Chen, B., Zeng, R., Song, Q., 2008. Heat
characteristics of microencapsulated phase change material suspensions in transfer of microencapsulated PCM slurry flow in a circular tube. AIChE J. 54,
minichannels. Heat Mass Transfer 44, 175–186. 1110–1120.
Roy, S.K., Avanic, B.L., 1997. Laminar forced convection heat transfer with phase Xing, K.Q., Tao, Y.X., Hao, Y.L., 2005. Performance evaluation of liquid flow with PCM
change material emulsions. Int. Commun. Heat Mass Transfer 24, 653–662. particles in microchannels. J. Heat Transfer 127, 931–940.
Sabbah, R., Farid, M.M., Al-Hallaj, S., 2008. Micro-channel heat sink with slurry of Yamagishi, Y., Takeuchi, H., Pyatenko, A.T., Kayukawa, N., 1999. Characteristics of
water with micro-encapsulated phase change material: 3D-numerical study. microencapsulated PCM slurry as a heat transfer fluid. AIChE J. 45, 696–707.
Appl. Therm. Eng. 29, 445–454. Zeng, R., Wang, X., Chen, B., Zhang, Y., Niu, J., Wang, X., Di, H., 2009. Heat transfer
Vand, V., 1945. Theory of viscosity of concentrated suspensions. Nature (London) characteristics of microencapsulated phase change material slurry in laminar
155, 364–365. flow under constant heat flux. Appl. Energy 86, 2661–2670.
Wang, X., Niu, J., Li, Y., Wang, X., Chen, B., Zeng, R., Song, Q., Zhang, Y., 2007. Flow
and heat transfer behaviors of phase change material slurries in a horizontal
circular tube. Int. J. Heat Mass Transfer 50, 2480–2491.