One dimensional thermal analysis of silicon carbide ceramic foam

used for solar air receiver

Fengwu Bai
*
Key Laboratory of Solar Thermal Energy and Photovoltaic System of Chinese Academy of Sciences, Institute of Electrical Engineering 6 Beiertiao, Zhongguancun,
P.O. Box 2703, Beijing, 100190, China
a r t i c l e i n f o
Article history:
Received 28 December 2009
Received in revised form
7 July 2010
Accepted 19 August 2010
Available online 22 September 2010
Keywords:
Thermal performance
Heat transfer
Air receiver
Silicon carbide ceramic foam
a b s t r a c t
Using air as heat transfer uid for electricity generation offers some signicant advantages for the
development of Concentrated Solar Power (CSP): high conversion efciency, low environmental impact
and being used in deserts or other areas scarce of water resources. Silicon carbide ceramic foams have the
characteristics of light weight, high strength, large specic surface areas, high porosity and excellent
thermal shock resistance performance which make them particularly t for absorber material in CSP. In
this paper, thermal performance of silicon carbide ceramic foam as solar air receiver is investigated
analytically based on the one dimensional physical model. The analytical results show that the air ow
resistance increases obviously with increasing air outlet temperature, the air ow resistance while the air
outlet temperature is equal to 1000

C is nearly 3 times the one while the air outlet temperature is equal
to 20

C with air velocity range is between one and six meters per second. The results of one dimensional
analysis of owand heat transfer process of ceramic foams suggest that there exists an input solar energy
ux limit for the unpressurized system, which will lead to limit the power capacity and the outlet air
temperature enhancement.
2010 Elsevier Masson SAS. All rights reserved.
1. Introduction
Solar tower is a promising upcoming technology for which more
attention has been paid in recent years by many countries such as
USA, Israel, Spain, Italy, France, Germany, Australia, Korean and
China. A central receiver is stationed on the top of a tower at the
focus of an array of heliostats, which reect concentrated radiation
into it. The main difference between solar thermal electricity and
fossil fuel power plant is the receiver of solar tower power plant.
The boilers used in most fossil fuel power plants have developed for
more than one hundred years. The heat loads on boilers are
controllable because of the controllable fuel input. But for receivers
used in solar thermal power it is difcult to distribute the heat ux
focused by hundreds of heliostats for the Sun movements and
complicated circumstances. Thus, the receiver is one of the key
equipments in whole solar tower power plant and has attracted
many researchers engaged in. In the past four decades, water/
steam, molten salt, liquid metal and air have been used as heat
transfer uid in solar receivers [1]. There is evident difference for
air compared to other three materials. For water/steam, molten salt
or liquid metal get heat inside absorber tubes in two dimensions,
and the heat transfer process between absorber and air takes place
in the whole volume. Metal wire/mesh [2], porous ceramic [3] and
porcupine ceramic tubes [4] have been tested. Until now, the
maximum outlet air temperature is up to 1700

C [4]. The silicon
carbide ceramic foams fabricated are highly versatile porous
materials which are used primarily in catalysts support, molten
metal ltration, hot gas ltration, and combustion, etc. In recent
years, porous ceramic is one of the most promising absorber
materials and has been investigated by many researchers [5e8]. In
this article, silicon carbide ceramic foam is selected as the absorber
material and one dimensional heat transfer performance has been
analyzed.
2. One dimensional analysis of heat transfer process
inside ceramic foam
Silicon carbide ceramic foam is a prospective absorber material
because of its excellent characteristics used under high tempera-
ture and high concentrated solar ux. The general characteristic of
silicon carbide ceramic foam depends on the relative density, pore
size, diameter of ceramic matrix, mechanical strength, thermal
conductivity and penetrability. As shown in Fig. 1 [9], silicon
* Tel.: 86 10 82547036; fax: 86 10 62587946.
E-mail address: baifw@mail.iee.ac.cn.
Contents lists available at ScienceDirect
International Journal of Thermal Sciences
j ournal homepage: www. el sevi er. com/ l ocat e/ i j t s
1290-0729/$ e see front matter 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ijthermalsci.2010.08.010
International Journal of Thermal Sciences 49 (2010) 2400e2404
carbide ceramic foamused in this article has ne three dimensional
macrostructure with a compact matrix.
2.1. Physical model
For ideal volumetric receiver, the surface temperature of the
outside of the absorber is lower than outlet air temperature and the
radiation losses are smaller. This is the case, when the input
concentrated radiation can get into the inside of the ceramic foam
and the incoming air cools the front surface well. As shown in Fig. 2,
for the large scale air receiver system, secondary concentrator can
efciently average the input concentrated radiative ux. Thus the
concentrated solar radiation on the front surface of the absorber
can be considered as uniform distribution. Because of the porosity
of the foam, the input radiation only can be absorbed in very thin
penetration layer, which make the thickness of the absorber is
much less than the diameter of the absorber. The radial heat
conduction can be neglected comparing to the large temperature
gradient in axial direction. In order to simplify the problem, the
following assumptions are made. (1) Concentrated solar radiative
ux is uniformon the surface of foam. (2) Radial heat conduction of
the solid matrix is neglected. (3) The radiation loss of front surface
of the absorber is neglected (4) Heat insulation of the air receiver is
very well and the heat loss to the environment is neglected. (5) Air
is considered as ideal gas.
For the solid matrix of ceramic foam, the energy conversion
equation is given as
l
s
v
2
T
s
vx
2
hd
_
T
s
T
f
_
S1 3 r
s
c
ps
1 3
vT
s
vs
(1)
where x stands for axis variation, T
s
is solid matrix temperature, l
s
is the thermal conductivity of solid matrix, h is convective heat
transfer coefcient between air and solid matrix, dis heat transfer
area, T
f
is air temperature, S is heat source, r
s
is density of solid
matrix, c
ps
is specic heat at constant temperature of solid matrix, s
is time, 3 is porosity, respectively.
For air, the energy conversion equation is
l
f
v
2
T
f
vx
2
hd
_
T
s
T
f
_
r
f
c
pf
u
f
$3
vT
f
vx
r
f
c
pf
$3
vT
f
vs
(2)
where l
f
is thermal conductivity of air, r
f
is density of air, c
pf
is
specic heat at constant temperature of air andu
f
is mean velocity
of air, respectively.
For a steady-state heat transfer process, merged equations (1)
and (2) yields
l
f
v
2
T
f
vx
2
l
s
v
2
T
s
vx
2
S1 3 r
f
c
pf
u$3
vT
f
vx
(3)
Under further assumption of the heat conduction process is
neglected, equation (3) can be further simplied to yield
r
f
c
pf
u
f
$3
vT
f
vx
S1 3 (4)
which indicates the input solar energy is absolutely removed by air.
For most ceramic foams, the classic DarcyeForchheimer lawcan
be used to describe pressure losses

dp
dx

m
f
K
u
f

C
F
r
f

K
p u
2
f
(5)
Here, p is air pressure, m
f
dynamic viscosity, K specic permeability
and C
F
is inertia coefcient.
For steady ow process, mass owrate is kept as constant. Mass
conservation equation can be written as
d
_
Ar
f
u
f
_
dx
0 (6)
where A is section area air ow.
Air is considered as ideal gas in this article yields state equation
p r
f
RT
f
(7)
where R is gas constant of air.
Viscosity of air signicantly depends on temperature and is
independent of pressure. When temperature of air is lower than
2000 K, Sutherlands formula can be used to derive the dynamic
viscosity of an ideal gas as a function of the temperature
Nomenclature
A section area (m
2
)
B Sutherlands constant
C
F
inertial coefcient
c
p
specic heat at constant pressure(J/kg K)
d
p
cell diameter of foam diameter(m)
d
pg
mean particle diameter of the granular medium(m)
h heat transfer coefcient (W/m
2
K)
K permeability(m
2
)
L length (m)
p air pressure(Pa)
q
in
heat ux density (W/m
2
)
_ m mass ow rate per unit area(kg/(s m
2
))
Nu Nusselt number
S heat source(W/m
3
)
T temperature (

C, K)
u velocity (m/s)
x axis variation
Greek symbols
3 porosity
d heat transfer area(m
2
)
l thermal conductivity (W/m K)
m Dynamic viscosity (Pa s)
r density (kg/m
3
)
s time(s)
Subscripts
f uid
i inlet
o outlet
r reference
s solid
Table 1
Parameters of silicon carbide ceramic foam.
NO Porosity 3 (%) Cell diameter
dp (mm)
FOAM1 90 2.0
FOAM 2 85 2.5
FOAM 3 75 3.0
F. Bai / International Journal of Thermal Sciences 49 (2010) 2400e2404 2401
m
f
m
r
_
T
f
T
r
_
1:5
T
r
B
T
f
B
(8)
where m
r
is reference viscosity in (Pa s) at reference temperature T
r
,
and B is Sutherlands constant B 110:4 for air. The unit of
temperature is in Kelvin K.
Dene air mass ow rate per unit area as
_ m r
f
u
f
(9)
Following equation can be obtained by integration of
equation (4).
T
f

S1 3
3 c
pf
_ m
x T
fi
(10)
where T
fi
is the inlet air temperature.
Combining equations (7), (9) into equation (5), pressure drop
equation is written as
p
dp
dx

_ mR
K
m
f
T
f

C
F
_ m
2
R

K
p T
f
(11)
The following non-dimensional variables are dened.
q
T
f
T
fi
T
fo
T
fi
; X
X
L
(12)
where T
fo
is the outlet temperature of air and L is the length of foam
along the ow direction.
Equations (10) and (11) transform to
q X (13)

p
L
dp
dX

_ mR
K
m
r
T
r
B
T
1:5
r
__
T
fo
T
fi
_
X T
fi
_
2:5
_
T
fo
T
fi
_
X T
fi
B

C
F
_ m
2
R

K
p
__
T
fo
T
fi
_
X T
i
_
(14)
By analytical integration, simple expression is obtained as
_
p
2
i
p
2
o
_
2
L
_
a
1
_ ma
2
_ m
2
_
(15)
Where p
i
, p
o
is the air pressure of inlet and outlet, respectively. The
constants a
1
, a
2
only depend on the air temperature of inlet and
outlet as followings.
a
1

R
K
2m
r
T
r
B
T
1:5
r
_
T
fo
T
fi
_
_

_
1
5
_
T
2:5
fo
T
2:5
fi
_

B
3
_
T
1:5
fo
T
1:5
fi
_
B
2
_

T
fo
_

T
fi
_
_
B
2:5
arctg

B
p
_

T
fo
_

T
fi
_
_
B

T
fo
T
fi
_
_

_ 16
and
a
2

C
F
R

K
p
_
T
fo
T
fi
_
2
(17)
On the basis of the assumption of the uniform input radiative
heat ux on the foam surface, mass ow rate can be expressed as
_ m
q
in
c
pf
3
_
T
fo
T
fi
_ (18)
where q
in
is the mean heat ux density.
According to equation (15), air ow pressure drop DP across the
foam absorber is
DP p
i
p
o
p
i

p
2
i
2L
_
a
1
_ ma
2
_ m
2
_
_
(19)
For the inlet air is obtained directly from atmosphere, the inlet
pressure is at most 1 bar, so the equation (19) has to satisfy the
following limit.
Fig. 1. (a) Image of a SiC-foam sample. (b) Local region zooms in.
Foam
Fan
Air
x
Secondary concentrator
Fig. 2. One dimensional heat transfer.
F. Bai / International Journal of Thermal Sciences 49 (2010) 2400e2404 2402
p
2
i
2L
_
a
1
q
in
c
p
3T
o
T
i

a
2
q
2
in
_
c
p
3T
o
T
i

2
_
! 0 (20)
Where p
i
1:01325 10
5
Pa is the atmospheric pressure.
The outlet air temperature is a very important parameter for
designers which is restricted by the allowable temperature of the
absorber material. Thus, for the given outlet air temperature, the
input radiative heat ux has to satisfy the following limit.
q
in

A
2
1
4A
2
2

p
2
i
A
2

_

A
1
2A
2
(21)
Where A
1
2La
1
=c
p
T
o
T
i
3, A
2
2La
2
=c
2
p
T
o
T
i

2
3
2
.
Based on above analysis, it is clear that there exists an input
solar energy ux limit for the unpressurized system, which will
lead to limit the power capacity and the outlet air temperature
enhancement. In volumetric air receiver system, the enhancement
of the inlet air pressure is very important to improve system
performance including increasing the outlet air temperature.
2.2. System analysis
For open system, inlet air pressure equals to atmospheric
pressure, and maximum of the system pressure drop is
atmospheric pressure. So in the case of outlet air pressure equals to
zero, outlet air temperature is an important parameter to deter-
mine the limit input solar ux.
Based on the formula of reference [10],
1
K
E
1
1 3
2
3
3
d
2
pg
C
F

K
p E
2
1 3
3
3
d
pg
(22)
where E
1
and E
2
are the constants 150 and 1.75, respectively.
And d
pg
is the mean particle diameter of the granular medium
which can be calculated by following equation.
d
pg

1:5
2:3

4
3p
1 3
_
1

4
3p
1 3
_ d
p
(23)
where d
p
is the cell diameter of foam.
2.2.1. The air outlet temperature effects on the ow resistance
Foam thickness is set as 100 mm, three kinds of silicon carbide
ceramic foam absorber ow resistance at different outlet air
temperature conditions variation with the inlet air velocity changes
Fig. 3. Flow resistance variation with air temperature of FOAM1.
Fig. 4. Flow resistance variation with air temperature of FOAM2.
Fig. 5. Flow resistance variation with air temperature of FOAM3.
0.00 0.02 0.04 0.06 0.08 0.10
1x10
7
2x10
7
3x10
7
4x10
7
5x10
7
6x10
7
q
n
i
m
/
W
(
2
)
L(m)
FOAM1
FOAM2
FOAM3
Fig. 6. Input solar ux limit vs absorber thickness while outlet air temperature is set as
800

C.
F. Bai / International Journal of Thermal Sciences 49 (2010) 2400e2404 2403
are shown in Figs. 3, 4, 5, respectively. The parameters of above
silicon carbide ceramic foam are listed as Table 1. The analytical
results show that the air ow resistance increases obviously with
increasing air outlet temperature, the air ow resistance while the
air outlet temperature is equal to 1000

C is nearly 3 times the one
while the air outlet temperature is equal to 20

C. It is fully
demonstrated that the process of temperature change has a signif-
icant effect on the ow resistance.
Usually, the fan of the volumetric receiver system provides
pressure head to drive the air through the absorber. The air
pressure drop of the absorber is same and it is not easy to
separately adjust the pressure head for every regions of absorber
with different air temperature. If the input radiation is not
uniform, that will lead to the uneven temperature distribution of
the solid matrix of absorber and air, which directly cause uneven
air ow resistance distribution. More air will pass through the
absorber in low air temperature regions than in the high air
temperature regions.
In other words, the air velocity in lowair temperature regions is
higher and in low air temperature regions. Thus, based on the
equation (5), the pressure drop increases in low air regions and
decreases in higher air temperatures regions because of the
velocity variation.
Considering the effect of heat transfer between air and absorber,
for high air temperature regions the lower air velocity also gives
rise to heat transfer deterioration which will make absorber
temperature and ow resistance further increased. So the air ow
rate continues to decline and form a vicious cycle in which further
deterioration of heat transfer process. Therefore, thermal diffusion
properties of the absorber material should by improved so that heat
in high temperature regions can be faster transferred to the lower
temperature regions. In addition, the uniformity treatment of the
input radiation ux density is also an effective measure to improve
ow resistance to prevent the heat transfer deterioration.
2.2.2. Input solar ux limit
As shown in Figs. 6, 7, for same absorber thickness, higher outlet
air temperature means higher input radiative ux limit. So by
enlarging air temperature difference of outlet and inlet more solar
energy can be removed. But for the cases of same outlet air
temperature, thicker absorbers can absorb less input radiative ux,
which mainly because increasing absorber thickness leads to air
owresistance increasing rapidly, which decrease the heat removal
ability of ceramic foam. Therefore, for the systemwith air entrance
is atmospheric pressure, outlet air temperature and absorber
thickness should be designed on the basis of the limit of input solar
radiation.
3. Conclusions
(1) The analytical results show that the air ow resistance
increases obviously with increasing air outlet temperature, the
air ow resistance while the air outlet temperature is equal to
1000

C is nearly 3 times the one while the air outlet temper-
ature is equal to 20

C with air velocity range is between one
and six meters per second.
(2) The results of one dimensional analysis of ow and heat
transfer process of ceramic foams suggest that there exists an
input solar energy ux limit for the unpressurized system,
which will lead to limit the power capacity and the outlet air
temperature enhancement.
(3) In volumetric air receiver system, an enhanced inlet air pres-
sure is very important to improve system performance,
including the outlet air temperature. A closed pressurized
system will be a promising technical choice in large scale solar
air receivers.
Acknowledgments
The author wishes to thank the National Natural Scientic
Foundation of China (Grant No.50806072) for nancial support.
References
[1] Manuel Romero, Reiner Buck, James E. Pacheco, An update on solar central
receiver systems, projects, and technologies, Journal of Solar Energy Engi-
neering, Transactions of the ASME 124 (2002) 98e108.
[2] Meinecke Wolfgang, Michael Kiera, Peter Wehowsky, 30MWe PHOEBUS
feasibility study: results of system engineering, Solar Energy Materials 24
(1991) 95e107.
[3] James M. Chavez, Cristina Chaza, Testing of a porous ceramic absorber for
a volumetric air receiver, Solar Energy Materials 24 (2) (1991) 172e181.
[4] J. Karni, A. Kribus, R. Rubin, P. Doron, The Porcupine: a novel high-ux
absorber for volumetric solar receivers, Journal of Solar Energy Engineering,
Transactions of the ASME 120 (1998) 85e95.
[5] R. Pitz-paal, B. Hoffschmidt, M. Bohmer, M. Becker, Experimental and
numerical evaluation of the performance and ow stablility of different types
of open volumetric absorbers under non-homogeneous irradiation, Solar
Energy 60 (3/4) (1997) 135e150.
[6] Thomas Fend, Bernhard Hoffschmidt, Robert Pitz-Paal, Oliver Reutter,
Peter Rietbrock, Porous materials as open volumetric solar receivers: experi-
mental determination of thermophysical and heat transfer properties, Energy
29 (2004) 823e833.
[7] Thomas Fend, Robert Pitz-Paal, Oliver Reutter, J. org Bauer,
Bernhard Hoffschmidt, Two novel high-porosity materials as volumetric
receivers for concentrated solar radiation, Solar Energy Materials &Solar Cells 84
(2004) 291e304.
[8] M. Becker, Th. Fend, B. Hoffschmidt, R. Pitz-Paal, O. Reutter, V. Stamatov,
M. Steven, D. Trimis, Theoretical and numerical investigation of ow stability
in porous materials applied as volumetric solar receivers, Solar Energy 80
(2006) 1241e1248.
[9] Hongtao Zhang, Jinsong Zhang, Hongyan Zhang, Computation of radar
absorbing silicon carbide foams and their silica matrix composites, Compu-
tational Materials Science 38 (2007) 857e864.
[10] Lacroix Maxime, Patrick Nguyen, Daniel Schweich, Cuong Pham Huu,
Sabine Savin-Poncet, David Edouard, Pressure drop measurements and
modeling on SiC foams, Chemical Engineering Science 62 (2007) 3259e3267.
0.00 0.02 0.04 0.06 0.08 0.10
1x10
7
2x10
7
3x10
7
4x10
7
5x10
7
6x10
7
7x10
7
8x10
7
L(m)
FOAM1
FOAM2
FOAM3
q
n
i
m
/
W
(
2
)
Fig. 7. Input solar ux limit vs absorber thickness while outlet air temperature is set
as 1000

C.
F. Bai / International Journal of Thermal Sciences 49 (2010) 2400e2404 2404