applied

sciences
Article
Prediction Model for the Anisotropic Thermal
Conductivity of a 2.5-D Braided Ceramic Matrix
Composite with Thin-Wall Structure
Zecan Tu 1 , Junkui Mao 1,2, *, Xingsi Han 1 and Zhenzong He 1
1 Jiangsu Province Key Laboratory of Aerospace Power System, College of Energy and Power Engineering,
Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China; tzc@nuaa.edu.cn (Z.T.);
xshan@nuaa.edu.cn (X.H.); hezhenzong@nuaa.edu.cn (Z.H.)
2 Collaborative Innovation Center for Advanced Aero-Engine, Beijing 100191, China
* Correspondence: mjkpe@nuaa.edu.cn




Received: 31 January 2018; Accepted: 19 February 2019; Published: 1 March 2019


Abstract: The thickness of the hot component in a turbine engine is usually small. Therefore,
the traditional prediction model of anisotropic thermal conductivity (ATC) based on the periodic
hypothesis may be improper for use in the thermal analysis of ceramic matrix composite (CMC)
components with a thin-wall structure. Thus, the prediction model for the ATC of a 2.5-D braided
CMC was investigated, taking into account the actual thickness of the CMC thin-wall structure.
An RVE (Representative Volume Element) model with a periodic boundary and a full-size model with
the actual thickness were built to study the temperature field, the heat flux field, and the effective
thermal conductivity of the CMC. A validation experiment was carried out to verify the accuracy
of the two prediction models. The effect of the composite’s thickness on the ATC and the critical
thickness suitable for the RVE model were also studied. The results showed that in the thermal
analysis of the thin-wall structure, the RVE model had a large deviation in the estimation of the
effective thermal conductivity in the thickness direction. The relative error between the numerical
data based on the RVE model and the experimental data reached 10.93%, while the relative error
was only 3.53% for the full-size model. Additionally, with increasing thickness, the effective thermal
conductivities, based on the RVE model and the full-size model, were close to each other. For the
critical thickness for the RVE model, which would be suitable for the prediction of the ATC, if the
material’s thermal properties such as the absolute value ratio and the level of anisotropy changed,
the corresponding critical thickness was also different. For the ATC of the SiC/SiC composites used
in this study, the critical thickness was found to be 18.4 mm, nearly 31 times larger than the RVE
model’s thickness.

Keywords: ceramic matrix composite; anisotropic thermal conductivity; 2.5-D braided composite;
representative volume element; thin-wall structure

1. Introduction
The inlet temperatures of gas turbine engines are becoming higher in order to obtain better
performances. Those high inlet temperatures could be far beyond the temperature limits of current
superalloy materials [1]. More coolant and more complex cooling configurations need to be applied,
but the performance of the turbine engines could be weakened. Consequently, highly heat-resistant
materials, such as ceramic matrix composite (CMC), have attracted a great deal of attentions in recent
years [2–4].
For the application of CMCs for the components with high temperatures, thermal analysis is an
important research topic. Michael et al. [5] studied the temperature distribution of a CMC turbine vane

Appl. Sci. 2019, 9, 875; doi:10.3390/app9050875 www.mdpi.com/journal/applsci

Appl. Sci. 2019, 9, 875 2 of 16

with a film-cooling configuration, both experimentally and numerically, for a gas turbine environment.
Their results encouraged the potential applications of CMC materials for turbine airfoil. Heidmann
et al. [6] performed a numerical study of the heat transfer process of a nickel-based alloy vane case
and a silicon nitride ceramic vane case in which the thermal conductivities were different. The results
showed that the thermal conductivity had a significant effect on the internal heat transfer, which
further affected the vane’s temperature field.
However, as a fiber reinforced composite, the thermal conductivity of the CMC material is
anisotropic [7,8] due to the significant differences between the thermal conductivities in the fiber’s
axial direction and the radial direction and the differences between the thermal conductivities of the
fibers and the matrix. Lebel et al. [9] investigated the thermal and mechanical performances of the
CMC materials under the cyclic thermal stresses of a gas turbine combustion chamber. The anisotropic
thermal conductivity (ATC) was used in the analysis, and the in-plane and the through-thickness
thermal conductivities were set as 3 W/(m·K) and 1.8 W/(m·K), respectively. Tu and Mao et al. [10]
studied the thermal analysis method of a CMC turbine vane. In the simulation, the effective ATCs
were used to reflect the CMC’s anisotropy. The results showed that the directions of the ATCs changed
at different locations on the CMC turbine vane, and as a result, the braided direction was changed
due to the warped surface. This variation of the ATC affected the heat transfer in the solid region of
the CMC turbine vane. Liu et al. [11] carried out a multi-scale thermodynamic analysis for a CMC
turbine vane. The in-plane and the through-thickness thermal conductivities were 12.1 W/(m·K) and
8.83 W/(m·K), respectively. The variation of the material’s ATC local coordinates due to the vane’s
curved surface was also considered.
The research mentioned above indicates that in the thermal analysis of a CMC component with a
high temperature, the ATC is an important parameter. However, the ATC varies for different CMC
materials, and the internal structure and the percentages of the components with different thermal
properties have significant influences on the ATC. For example, Kiani et al. [12] studied the thermal
conductivity of triaxial braided composites with fibers that changed directions. The results showed that
when the direction of the fibers changed from ±25◦ to ±65◦ , the thermal conductivity in the thickness
direction changed from 4.2 W/(m·K) to 5.15 W/(m·K). In the work of Jiang et al. [13], the thermal
conductivities of 3-D braided composites were investigated. The results showed that the volume
fraction of the fibers had a considerable effect on the thermal conductivity. When the volume fraction
of the fibers increased from 0.25 to 0.55, the thermal conductivity in the braided direction changed
from 0.68 W/(m·K) to 1.3 W/(m·K).
Thus, the prediction method of the braided CMC’s ATC is an important aspect in the application
of the CMC, which has been widely studied. The homogenization method [14], the thermoelectric
analogy model [15], and the mathematical model [16] have been developed for many years, and they
have different advantages depending on the purpose. The Representative Volume Element (RVE)
method was developed based on the homogenization method, and it has attracted much attention
due to the fact that it could introduce many information of the micro/meso-structures. This model
was firstly studied by Hill [17]. A simple way to apply the RVE model is to use the minimal structural
repeating unit with a periodicity hypothesis. For example, Siddiqui et al. [18] investigated the effective
thermal conductivity and thermal resistance of a 2-D woven fabric using the finite element method.
An RVE model including two warp yarns and two weft yarns was applied based on the scanning
electron microscope, and the geometrical size of the RVE was 0.862 mm × 0.862 mm × 0.5 mm.
Ai et al. [19] studied the thermal conductivity of 3-D woven C/C composites at high temperature
numerically and experimentally. The multi-scale method was applied, including the microscale model
and the mesoscale model, in which the mesoscale model was an RVE model including two warp yarns,
two weft yarns, and one yarn in the thickness direction. The geometrical size of the RVE was 1.96 mm
× 1.96 mm × 0.76 mm, and the thickness of the experimental sample was 5 mm. Fang et al. [20] built
an RVE model with a geometrical size of 2.724 mm × 2.724 mm × 5.843 mm to predict the thermal
 Appl. Sci. 2019, 9, 875 3 of 16

Appl. Sci. 2019, 9, x FOR PEER REVIEW 3 of 18

conductivities of three-dimensional four-directional braided composites. Similar works can also be
foundAllin
ofother references
the studies [21–23]. above used an RVE model where the geometrical structure was
mentioned
based All of the
on the studies mentioned
periodicity hypothesis above
and theused global an thickness
RVE model where
of the RVE the geometrical
models was in structure
the rangewas of
based on the periodicity hypothesis and the global thickness of the
0.5 mm to 3 mm. However, the thickness of the hot components, such as the turbine vane, was small RVE models was in the range
of 0.5
(2–3 mmTherefore,
mm). to 3 mm.for However,
the thermalthe analysis
thicknessofofa CMC the hot components,
component and such as the turbine
the prediction of thevane,
thermalwas
small (2–3 mm).
conductivity of theTherefore, for the thermal
corresponding material,analysis
the RVE of amodel
CMC may component and the The
be improper. prediction of the
periodicity
thermal conductivity of the corresponding material,
hypothesis, which is the foundation of the RVE model, is unsatisfied. the RVE model may be improper. The periodicity
hypothesis, which is the
Some researchers foundation
tried to build aof the RVE
model with model,
a full is unsatisfied.
size in the thickness direction. For example,
Some researchers tried to build a model
Dong et al. [24] numerically and experimentally investigated thewith a full size in thermal
the thickness direction.
conductivity For example,
of 2.5-D angle-
interlock woven composites. A full-size model with a geometrical size of 8.8 mm × 8.46 mmof× 2.5-D
Dong et al. [24] numerically and experimentally investigated the thermal conductivity 6.26
angle-interlock
mm was built. In woven composites.
this model, A full-size
the thickness of themodel with a geometrical
fiber bundles was nearly 0.3 size of 8.8
mm. The mm × 8.46
results mm ×
showed
6.26the
that mm was built.
thermal In this model,
conductivity the thickness
in the thickness of thebased
direction fiber bundles was nearly
on the full-size model 0.3agreed
mm. The wellresults
with
showed that the thermal conductivity in the thickness direction based
the experimental data, and the relative error was about 4.1%. Although the full-size model could on the full-size model agreed
well with
describe thethe experimental
geometrical data, and the
characteristics morerelative error was
accurately, about 4.1%.
it needed much Although
more meshes the full-size
than the model
RVE
could describe the geometrical characteristics more accurately, it
model, due to the big differences between the sizes of the microstructure and the macroscopicneeded much more meshes than the
RVE model, This
components. due to the big
limited thedifferences
application between the sizesmodel
of the full-size of theinmicrostructure and theof
the thermal analysis macroscopic
the CMC
components. This limited the application of the full-size model in
components with large sizes. Therefore, there is a critical thickness for the application of the the thermal analysis of the RVE
CMC
model or the full-size model, which need to be studied considering both the accuracy and RVE
components with large sizes. Therefore, there is a critical thickness for the application of the the
model or the full-size model, which need to be studied considering both the accuracy and the efficiency.
efficiency.
Accordingly,the
Accordingly, the present
present workwork
aimsaims to investigate
to investigate the ATC theprediction
ATC predictionmodel of model of the
the 2.5-D 2.5-D
braided
braidedmatrix
ceramic ceramiccomposites,
matrix composites,taking taking into account
into account the actual
the actual thickness
thickness of of
thethehigh-temperature
high-temperature
components. An RVE model with a periodic boundary
components. An RVE model with a periodic boundary and a full-size model based on and a full-size model based onthe
theactual
actual
thickness were built to study the effect of the composite thickness on the estimation accuracy ofofthe
thickness were built to study the effect of the composite thickness on the estimation accuracy the
effectivethermal
effective thermalconductivity.
conductivity.The Thevalidation
validationexperiment
experimentwas wascarried
carriedoutouttotoverify
verifythetheestimation
estimation
method of the effective thermal conductivity. Additionally, the
method of the effective thermal conductivity. Additionally, the influence of the thickness on influence of the thickness onthethe
effective thermal conductivity and the critical thickness that was suitable
effective thermal conductivity and the critical thickness that was suitable for the RVE model were for the RVE model were also
studied
also under
studied different
under operating
different operatingconditions
conditions of the CMC’s
of the CMC’sATC. ATC.

2.2.Research
ResearchModels
Models
Figure1a1ashows
Figure showsa aphotograph
photographofofthethe2.5-D
2.5-Dbraided
braidedCMC CMC(SiC/SiC).
(SiC/SiC).According
Accordingtotothe
theinformation
information
suppliedby
supplied bythe
thematerial
materialsupplier
supplierSAFEI
SAFEIgroup,
group,the
theaxial
axialand
andthetheradial
radialthermal
thermalconductivities
conductivitiesofofthe
the
SiC fiber bundle are 9.66 W/(m · K) and 1.48 W/(m · K), and the thermal conductivity
SiC fiber bundle are 9.66 W/(mK) and 1.48 W/(mK), and the thermal conductivity of the ceramic of the ceramic
matrixisis6.5
matrix W/(m·K).The
6.5W/(mK). Thesample
sampleused
usedininthe
the numerical
numerical simulation
simulation andand the
the experiment
experimentare areshown
shownin
Figure 1. The size of this sample was 10 mm × 10 mm × 2.8 mm (length × width
in Figure 1. The size of this sample was 10 mm × 10 mm × 2.8 mm (length × width × height). × height).

(a) Photograph of the 2.5-D braided CMC sample (b) A schematic diagram of the braided structure

Figure
Figure 1. (a)
1. (a) A photograph
A photograph andand
(b) (b) a schematic
a schematic diagram
diagram of 2.5-D
of the the 2.5-D braided
braided ceramic
ceramic matrix
matrix composite
(CMC). composite (CMC).

Figure1b1bshows
Figure showsa aschematic
schematicdiagram
diagramofofthe
theideal
idealstructure
structureofofthe
the2.5-D
2.5-Dbraided
braidedfiber
fiberbundles.
bundles.
AAminimal
minimalperiodic
periodicunit
unitcell
cellwas
wascreated
createdasasthe
theRVE
RVEmodel,
model,which
whichwaswasapplied
appliedininthe
thethermal
thermalanalysis
analysis
ofofthe
the2.5-D
2.5-Dbraided
braidedCMC.
CMC.Similar
Similartotothe
theRVE
RVEmodel
modelininReference
Reference[25],
[25],ananRVE
RVEmodel
modelwaswasbuilt
builtofof
the 2.5-D braided CMC, considering the geometry characteristics of the 2.5-D braided CMC shown
in Figure 1a. For the model, the following hypotheses are considered:
1. The sections of the axial yarns and the braided yarns are both hexagons.
 Appl. Sci. 2019, 9, 875 4 of 16

the 2.5-D braided CMC, considering the geometry characteristics of the 2.5-D braided CMC shown in
Figure 1a. For the model, the following hypotheses are considered:
Appl. Appl.
Sci. 2019, 9, x FOR
Sci. 2019, 9, x PEER REVIEW
FOR PEER REVIEW 4 of 18
4 of 18
1. The sections of the axial yarns and the braided yarns are both hexagons.
2.2. 2.
There
Thereare
areno
There nocracks
are ininthe
cracks
no cracks inmodels.
the Instead,
models.
the they
Instead,
models. they
Instead, are
arecompletely
they completely
are continuous.
continuous.
completely continuous.
Figure
Figure 22shows
Figure shows the
2 shows theRVE
the model
RVERVE of
model
model oftheof2.5-D
the braided
the 2.5-D
2.5-D CMC.
braided
braided CMC. The
CMC.The size
The of
of the
size size ofmodel
the was
the model
model was was5.72 mm
5.72 5.72
mm mm×× ×
3.34 mm
3.343.34
mmmm××0.6 ×mm
0.6 0.6
mmmm (length ××width
(length
(length ××
× width
width height), asas
× height),
height), shown
as ininFigure
shown
shown 2a.
in Figure
Figure 2a.According
2a. totothe
According
According to data
the the
data ofofthe
data of the
the
scanning
scanning electron
scanning microscope
electron
electron test,
microscope
microscope the
theaverage
test,test, size
the average
average of
ofthe
sizesize ofsections
the of
of the
the sections
sections offiber
the bundles
the fiber
fiber was
bundles
bundles was was1.5
1.5 mm ×× ×
1.5 mm
mm
0.25 mm
0.250.25
mmmm(width ×
(width
(width height), as
× height),
× height), shown in
as shown
as shown Figure 2b.
in Figure
in Figure The angle
2b. angle
2b. The of
The angle the braided
of theofbraided
the braidedyarns was
yarnsyarns
was 45was◦
45°. The parts
45°.parts
. The The ofparts
ofthe
the
ofmatrix
matrix andand
the matrix thethe
and fiber bundles
thebundles
fiber fiber were
bundles
were given
were
given in
given
in Figures 2c,d.2c,d.
in Figures
Figure 2c,d.

(a) The
(a)RVE model
The RVE model (b) The
(b)section of theoffiber
The section bundles
the fiber bundles

(c) The
(c)matrix
The matrix (d) The
(d)fiber bundles
The fiber bundles
Figure
Figure2.2.AAschematic
Figure diagram
2.schematic
A schematic
diagramofofthe
diagram ofRepresentative
the the Volume
Representative
Representative Element
Volume
Volume (RVE)
Element
Element model.
(RVE)
(RVE) model.
model.

Additionally,
Additionally,
Additionally,aafull-size
full-sizemodel
model
a full-size was
modelwas built
was to to
built
builtcompare
compare
to with
compare thethe
with
with RVE model.
theRVE
RVE Figure
model.
model. 3 shows
Figure the the
shows
3 shows
full-size model,
the full-size
full-size which
model, was
which built
was based
built on
based theonactual
the size
actual of the
size sample
of the shown
sample
model, which was built based on the actual size of the sample shown in Figure 1a. in
shown Figure
in 1a.
FigureThe1a.The
geometry
Thegeometrysize was
geometry 11.4
sizewas
size mm
was11.4 ×mm
11.4mm10 × ×1010
mm × mm
mm2.8×mm × 2.8
2.8 (length
mm mm × width
(length
(length ×× width
height),
× width × for which
height),
× height), the
for
for which length
which thewas
the length set set
length
was
as 11.4
was as mm
set11.4 to preserve
as 11.4
mm mm the periodicity
to preserve
to preserve the the of side
periodicity
periodicity surfaces.
of side
of side The distances
surfaces.
surfaces. TheThe between
distances
distances the two
between
between adjacent
thethe
twotwo axialaxial
adjacent
adjacent
yarns
axial on
yarns the
yarns z-axis
onon the
the and and
z-axis x-axis
z-axis werewere
x-axis
andx-axis 0.6
weremm and
0.6mm
0.6 1.4
mmand andmm, respectively.
1.4mm,
1.4 mm,respectively.The The
respectively. distance between
Thedistance
distance the two
between
between thetwo
the
adjacent braided
twoadjacent
adjacent yarns
braided on the
yarns ony-axis was was
y-axis
they-axis 0.2
was mm. The
mm.angle
0.2mm. of the
Theangle
angle ofbraided yarns
thebraided
braided was was
yarns 45°.
was45°. ◦
45 .
braided yarns on the 0.2 The of the yarns

(a) The
(a)full-size model
The full-size model (b) The
(b)sections of theoffull-size
The sections model
the full-size model

Figure 3. Cont.
 Appl.Sci.
Appl. Sci.2019,
2019,9,9,875
x FOR PEER REVIEW 5 of
of 16
18

(c) The matrix (d) The fiber bundles

Figure3.3.Schematic
Figure Schematicdiagrams
diagramsof
ofthe
thefull-size
full-sizemodel.
model.

3.3. Numerical
NumericalMethodology
Methodology
For
For the
the two numerical models
two numerical modelsmentioned
mentionedabove,
above,i.e.,
i.e.,the
the RVE
RVE model
model and and
thethe full-size
full-size model,
model, the
the thermal analysis was carried out using the Finite Element Method (FEM)
thermal analysis was carried out using the Finite Element Method (FEM) based on the Comsol based on the Comsol
software.
software.InInthe
the simulation,
simulation, the
the thermal
thermal conductivity
conductivity ofof the
the matrix
matrix and
and the
the fiber
fiber bundles
bundles were
were firstly
firstly
calculated, with particular attention to the ATCs of different parts of the braided yarn
calculated, with particular attention to the ATCs of different parts of the braided yarn due to the due to the varied
directions of the fiber
varied directions bundles.
of the Then, theThen,
fiber bundles. meshes
theofmeshes
the twoofmodels
the two were generated,
models and the boundary
were generated, and the
conditions with constant temperatures were applied. Finally, the finite element
boundary conditions with constant temperatures were applied. Finally, the finite element simulation simulation of the
temperature fields was
of the temperature carried
fields wasout. The information
carried about the heat
out. The information about fluxthe
andheat
the thermal
flux andgradient was
the thermal
obtained simultaneously, and this information was used to calculate the effective
gradient was obtained simultaneously, and this information was used to calculate the effective thermal conductivity.
The stepsconductivity.
thermal of the procedureThe are given
steps below.
of the procedure are given below.
3.1. Governing Equations
3.1. Governing Equations
In the solid region, the steady-state energy transport equation in the tensor form is written as [26]
In the solid region, the steady-state energy transport equation in the tensor form is written as
[26] →

ρC p ν ·∇ T = ∇· k ij ∇ T + Q (1)

where Cp is the specific heat capacity atpconstant pressure

(
 C  T =   kij T + Q )
and Q is the source term. The term on the
(1)

left-hand
where Cpside represents
is the the convective
specific heat capacity atenergy transfer
constant due and
pressure to the
Qrotational or translational
is the source term. The termmotion of
on the
→
the solids. The velocity field, ν , is computed from the motion specified for the solid zone.
left-hand side represents the convective energy transfer due to the rotational or translational motion In this paper,
both
of thethis termThe
solids. Q are zero
andvelocity because
field, the solid region
 , is computed is motion
from the stationary. kij is the
specified for thermal
the solidconductivity,
zone. In this
which is anisotropic in this paper.
paper, both this term and Q are zero because the solid region is stationary. kij is the thermal
conductivity, which is anisotropic in this paper.
3.2. Application of the ATC
In the RVE of
3.2. Application model shown in Figure 2, the thermal conductivity of the matrix is isotropic, but the
the ATC
thermal conductivity of the fiber bundles is anisotropic, and the thermal conductivity in the axial
In the RVE model shown in Figure 2, the thermal conductivity of the matrix is isotropic, but the
direction is bigger than that in the radial direction. To represent the ATC of the fiber bundles, a local
thermal conductivity of the fiber bundles is anisotropic, and the thermal conductivity in the axial
Cartesian coordinate system (ζ, η, ν) was introduced, with the ζ axis along the yarn’s axial direction.
direction is bigger than that in the radial direction. To represent the ATC of the fiber bundles, a local
The ζ axis was considered as the principal direction of the thermal conductivity (PDTC) because the
Cartesian coordinate system (ζ, η, ν) was introduced, with the ζ axis along the yarn’s axial direction.
thermal conductivity was largest in this direction.
The ζ axis was considered as the principal direction of the thermal conductivity (PDTC) because the
The axial yarn’s direction was along the Y axis of the global Cartesian coordinate system (X, Y, Z),
thermal conductivity was largest in this direction.
as shown in Figure 4, so the ATC of axial yarns could be represented by three thermal conductivities
The axial yarn’s direction was along the Y axis of the global Cartesian coordinate system (X, Y,
on the X, Y, and Z axes, respectively. However, the directions of the braided yarns were changed
Z), as shown in Figure 4, so the ATC of axial yarns could be represented by three thermal
in different parts, as shown in Figure 4. There was an inclined angle between the local Cartesian
conductivities on the X, Y, and Z axes, respectively. However, the directions of the braided yarns
coordinate (ζ, η, ν) and the global Cartesian coordinate (X, Y, Z), so the ATC applied for (X, Y, Z)
were changed in different parts, as shown in Figure 4. There was an inclined angle between the local
needed to be transferred, and the transformation method was described as follows.
Cartesian coordinate (ζ, η, ν) and the global Cartesian coordinate (X, Y, Z), so the ATC applied for
(X, Y, Z) needed to be transferred, and the transformation method was described as follows.
Appl. Sci.
Appl. Sci. 2019,
2019, 9,
9, 875
x FOR PEER REVIEW 66 of 18
of 16

4.AAschematic
Figure 4.
Figure diagram
schematic of the
diagram principal
of the principaldirection of the of
direction thermal conductivity
the thermal (PDTC).
conductivity
(PDTC).
According to the study in Reference [27], the anisotropic thermal conductivity kij can be expressed
as theAccording
matrix to the study in Reference [27], the anisotropic  thermal conductivity kij can be
k XX k XY k XZ
expressed as the matrix
k ij =  kYX kYY kYZ  (2)
 
k ZX
k k
XX kXY k 
ZY kXZZZ
kij =  kYX kYY kYZ  (2)
Equation (2) can be rewritten as shown below:
 kZX kZY kZZ 
     
Equation (2) can k k k l m n kζ 0 0 l1 l2 l3
 be rewritten asshown  below: 
XX XY XZ 1 1 1
 kYX kYY kYZ  =  l2 m2 n2  0 k η 0   m1 m2 m3  (3)
 
k XX k XY k XZ  l1 m1 n1   k 0 0  l1 l2 l3 
k ZX k ZY k ZZ   l3 m 3 n 3   0 0 kν   n1 n2 n3
 kYX kYY kYZ  = l2 m2 n2  0 k 0   m1 m2 m3  (3)
where  k ZX k ZY k ZZ  l3 m3 n3  0 0 k   n1 n2 n3 

where
   
l1 m 1 n 1 cos γ cos β − cos α sin γ + cos γ sin β sin α sin γ sin α + cos γ sin β cos α
=
 l2 m2 n2   sin γ cos α cos γ cos α+l1 sin
m1 γnsin β sin α − cos γ sin α + sin γ sin β cos α  (4)
   
1 
l3 m 3 n 3 − sin β l βm
cos sin α  cos β cos α
 2 2 n2  =
Equation (4) shows the transform matrixlbetween 
3 m3 n3 the local coordinate system (principal direction
(4)
 cos  cos  −and
of the thermal conductivity) cos the
sin global
+ cos Cartesian
sin  sin  coordinate
sin  sin  +system,
cos  sinwith
 cos the
  inclined PDTC
 sin  cos  cos  cos  + sin  sin  sin  − cos  sin  + sin  sin  cos  
angles α, β and γ.
 
 − sin  cos  sin  cos  cos  
3.3. Mesh and Boundary Conditions
Equation
The top and(4)bottom
showssurfaces
the transform
of the RVEmatrix
modelbetween the local
and the full-size coordinate
model system
on the z-axes (principal
were applied
direction of the thermal conductivity) and the global Cartesian coordinate system, with
with constant temperature boundary conditions. The temperatures on the top and bottom surfacesthe inclined
PDTCset
were angles
as 273α,Kβand
and283
γ. K, respectively. The four side surfaces were applied with periodic boundary
conditions, as shown in Figure 5a,b.
3.3. Mesh and Boundary
To generate Conditions
the mesh, a grid dependency test based on the RVE model was carried out.
The maximum
The top and mesh
bottomsizesurfaces
was changed frommodel
of the RVE 1.09 mm andtothe
0.2full-size
mm, and the corresponding
model numbers
on the z-axes were of
applied
the
withmeshes
constantwere 9490, 17639,
temperature 41295, 71,898,
boundary 110,014,
conditions. Theand 209,159. Figure
temperatures 5c shows
on the thebottom
top and variation of kZ
surfaces
with
werethe increasing
set as 273 K and number
283 K, of the mesh. The
respectively. The results
four side show that when
surfaces the number
were applied withwas equalboundary
periodic to 71,898,
the relative as
conditions, variation
shown of in kFigure
Z was 5a,b.
smaller than 1%.
Finally,
To generatethe number
the mesh, of athe mesh
grid applied for
dependency testthe RVEon
based model was model
the RVE 71,898.was
Thecarried
maximum out. and
The
minimal
maximum mesh
mesh sizes
sizewere
was 0.458
changedmmfrom
and 0.057
1.09 mmmm,toand 0.2 the
mm,meshand growth rate was smaller
the corresponding thanof1.45,
numbers the
as shown in Figure 5a. For the full-size model, the mesh generation strategy was
meshes were 9490, 17639, 41295, 71,898, 110,014, and 209,159. Figure 5c shows the variation of kZ with the same as that of
the RVE model.number
increasing The corresponding
of the mesh.number of theshow
The results meshthatwaswhen
798,492,
the the maximum
number and minimal
was equal mesh
to 71,898, the
sizes were
relative kept at of
variation 0.458 mm smaller
kZ was and 0.057 mm,
than and the mesh growth rate was smaller than 1.45, as shown
1%.
in Figure 5b. the number of the mesh applied for the RVE model was 71,898. The maximum and
Finally,
minimal mesh sizes were 0.458 mm and 0.057 mm, and the mesh growth rate was smaller than 1.45,
3.4. Operating Conditions and Parameters Definition
as shown in Figure 5a. For the full-size model, the mesh generation strategy was the same as that of
the RVE
In the model. The corresponding
simulation, there were twonumber
kinds ofof the meshmodels,
numerical was 798,492,
i.e., thethe
RVEmaximum
model andandtheminimal
full-size
mesh sizes were kept at 0.458 mm and 0.057 mm, and the mesh growth rate
model. The RVE model was a minimal periodic element in the 2.5-D braided CMC, while the full-size was smaller than 1.45, as
shown was
model in Figure
based5b.on the sample’s real size for the 2.5-D braided CMC.
 Appl. Sci. 2019, 9, x FOR PEER REVIEW 7 of 18

Appl. Sci.Sci.
Appl. 2019, 9, 875
2019, 9, x FOR PEER REVIEW 77 of
of 18
16

(a) RVE model (b) Full-size model

(a) RVE model (b) Full-size model

(c) Grid dependency test

Figure 5. The mesh and boundary conditions.

3.4. Operating Conditions and Parameters Definition

(c) Grid dependency test
In the simulation, there were two kinds of numerical models, i.e., the RVE model and the full-
size model. The RVE model Figure Figure
was a 5. 5. The
minimal
The mesh andboundary
periodic
mesh and boundary
element conditions.
in the 2.5-D braided CMC, while the
conditions.
full-size model was based on the sample’s real size for the 2.5-D braided CMC.
3.4. Operating Conditions
Simultaneously,
Simultaneously, and Parameters
toto study
study the Definition
the influence
influence ofof the
the sample
sample thickness
thickness on on the
the Effective Thermal
Effective Thermal
Conductivity (ETC) and
(ETC) and the the critical
critical thickness that was suitable for the RVE model, full-size models
Conductivity
In the simulation, there were thickness
two kindsthat was suitable
of numerical for the
models, i.e.,RVE
the model, full-size
RVE model and models
the full-
with
withdifferent
size different
model. Thethickness
thickness of
RVE modelof the z-axis
the was
z-axis werebuilt.
were
a minimal built. The
The
periodic thicknesses
thicknesses
element inLZthe Lwere were set2.8
as mm,
set braided
Z 2.5-D as 2.8 mm,
CMC, 4.0 mm,
4.0 while
mm, 7.6
the
7.6mm,
mm, 11.2
11.2 mm,
mm, 14.8
14.8 mm,
mm, and
and 18.4
18.4 mm,
mm, asas shown
shown inin Figure
Figure 6.6. The
The
full-size model was based on the sample’s real size for the 2.5-D braided CMC. boundary
boundary conditions
conditions and
and the
the mesh
mesh
generation
generation strategies
strategieswere
Simultaneously, tothe
were thesame
study same for
the the
themodel
forinfluence
model shown
of the in
shown inFigure
Figure
sample 5b.
5b.The
thicknessThecorresponding
corresponding
on the Effective numbers
numbers of
Thermal
the meshes were
ofConductivity
the meshes 798,492,
were
(ETC) and 1,297,051,
798,492, critical2,785,965,
the1,297,051, 4,264,916,
2,785,965,
thickness was 5,465,846,
4,264,916,
that forand
5,465,846,
suitable 6,760,725,
and
the respectively.
6,760,725,
RVE model,respectively.
full-size models
with different thickness of the z-axis were built. The thicknesses LZ were set as 2.8 mm, 4.0 mm, 7.6
mm, 11.2 mm, 14.8 mm, and 18.4 mm, as shown in Figure 6. The boundary conditions and the mesh
generation strategies were the same for the model shown in Figure 5b. The corresponding numbers
of the meshes were 798,492, 1,297,051, 2,785,965, 4,264,916, 5,465,846, and 6,760,725, respectively.

Figure6.6.Full-size
Figure Full-size models
models with different thicknesses.
thicknesses.

Additionally,
Additionally,totomake
makethe
theresults
resultsand
andconclusions
conclusions more
more useful, in the
useful, in theinvestigation
investigationofofthe
thethermal
thermal
conductivities with increasing thickness, the absolute value ratio, n (as shown in Equation
conductivities with increasing thickness, the absolute value ratio, n (as shown in Equation (5)), (5)), and the
and
level of anisotropy,
the level r (asr shown
of anisotropy, in Equation
(asFigure
shown (6)), ofof
the thermal conductivities thefiber
fiberbundles
bundles
6.in Equation
Full-size (6)),
models the
with thermal
different conductivities of the
thicknesses.
were
werechanged.
changed.
kζ kη k νuseful, in the investigation of the thermal
Additionally, to make the resultsn and= conclusions
= more
= (5)
9.66 1.48 1.48
conductivities with increasing thickness, the absolute value ratio, n (as shown in Equation (5)), and
the level of anisotropy, r (as shown in Equationk(6)),
ζ ofk ζthe thermal conductivities of the fiber bundles
r= = (6)
were changed. kη kν
Appl. Sci. 2019, 9, 875 8 of 16

where kζ , kη , and kν are the thermal conductivities along the ζ, η, and ν directions and their values are
9.66 W/(m·K), 1.48 W/(m·K), and 1.48 W/(m·K), respectively, of the sample in Figure 1.
The operating conditions of the absolute value ratio n and the level of anisotropy r were set
according to the data in Reference [28], as shown in Table 1. The ATC of the sample shown in Figure 1a
was set as the basic Case 1. For Cases 2–5, the values of kζ , kη , and kν were increased at the same
time to improve the absolute value ratio n and the level of anisotropy r was kept constant. For Cases
6–9, the kη and kν were kept constant and the kζ was increased to improve the level of anisotropy r.
Correspondingly, the absolute value n did not exist. The corresponding kζ , kη , and kν for each case are
also given in Table 1.

Table 1. A summary of the cases studied.

Case kζ /kη /kν W/(m·K) n r

1 9.66/1.48/1.48 1 6.53
2 19.32/2.96/2.96 2 6.53
3 38.64/5.92/5.92 4 6.53
4 57.96/8.88/8.88 6 6.53
5 96.6/14.8/14.8 10 6.53
6 14.8/1.48/1.48 / 10
7 19.24/1.48/1.48 / 13
8 23.68/1.48/1.48 / 16
9 29.6/1.48/1.48 / 20

In the analysis of the results, the parameter Ff that represents the volume fraction of the fibers is
defined as follows:
Vf
Ff = (7)
Vt
where Vf is the volume of fibers and Vt is the total volume of the model.
Additionally, although the constant temperature boundary conditions in different models were
the same, it should be noted that the thickness changed, so the temperature and the heat flux in
different models varied. To compare the temperature and the heat flux in different models fairly,
the relative temperature fluctuations δT and the relative heat flux fluctuations δhf are defined:

Ti − Tavg
δT = (8)
10 [K]

h f i − h f avg
δh f = (9)
h f avg
where Ti is the local temperature, Tavg is the average temperature, and 10 K is the difference between
the two constant temperature boundary conditions. The hfi term is the local heat flux, and hfavg is the
average heat flux.
When the relative deviation between the ETCs based on the RVE model and the full-size model
was smaller than 5%, the corresponding thickness of the full-size model was defined as the “critical
thickness”, LZ . For different ATC cases in Table 1, the critical thickness LZ which was suitable for the
RVE model was investigated in this paper.

4. Results and Discussions

4.1. Comparison between the RVE Model and the Full-Size Model
To analyze the difference between the results based on the RVE model and the full-size model,
the temperature fields, the heat flux fields, and the effective thermal conductivity of the two models
were analyzed. In the simulation, the ATC was set as the basic Case 1, and the thickness of the
Appl. Sci. 2019, 9, 875 9 of 16

full-size model was the sample’s actual size (2.8 mm). Additionally, tests of the sample’s thermal
conductivity through the thickness were carried out to validate the accuracy of the RVE model and the
full-size model.

4.1.1. Temperature Field

Figure 7 shows the temperature fields of the RVE model and the full-size model. The temperature
distribution of the RVE’s surface is shown in Figure 7a. In the figure, regions A and C are the matrix
and region B is the fiber bundles. It can be seen that the temperature distributions of the three regions
had significant differences. In the regions between the fiber bundles and the matrix (A and B; B and C),
the directions of the isotherms had considerable variations. The reason for this was that in the process
of heat transfer from the high-temperature surface to the low-temperature surface (via the red arrows
shown in Figure 7a), there were different regions with different thermal conductivities, which led to
heat conduction along multiple walls. In the fiber bundles, the horizontal yarns and the diagonal yarns
had different directions, which led to different ATCs, so the transmission paths of the heat in the fiber
bundles also varied.
Figure 7b shows the temperature distribution on the surface of the full-size model.
The temperature field was also heterogeneous. There were significant thermal gradients in the region
between the fiber bundles and the matrix, which was similar to the RVE model. The reason for this
was
Appl. also that9,the
Sci. 2019, transmission
x FOR PEER REVIEWpath of the heat changed due to the difference of the fibers’ and the
10 of 18
matrix’s thermal conductivities.

(a) Surface of the RVE model (b) Surface of the full-size model

(c) Middle section of the RVE model (d) Middle section of the full-size model

Figure 7. The
Figure 7. The temperature
temperature fields.
fields.

Figure
Figure 7c,d
7c,d shows
shows the
the distributions
distributions of of the
the temperature
temperature fluctuation
fluctuation δδTT in
in the
the middle
middle sections
sections of
of
the z-axes
the z-axes of the two models. The results show that δTT for these sections was obvious in both the
of the two models. The results show that δ for these sections was obvious in both the fiber
fiber
bundles
bundles and
and the
the matrix.
matrix. Additionally,
Additionally,ininthe
themiddle
middlesection
sectionof
ofthe
thefull-size
full-sizemodel,
model,δδTT of
of the
the part
part with
with
the same size of the RVE model was found. The distributions of δT in those two models had significant
differences, and δT in the RVE model had more dramatic changes compared to δT in the full-size
model. For the RVE model, the maximum δT in the middle section was 9.96%, while for the full-size
model, the maximum δT in the middle section was 6.39%.
Appl. Sci. 2019, 9, 875 10 of 16

the same size of the RVE model was found. The distributions of δT in those two models had significant
differences, and δT in the RVE model had more dramatic changes compared to δT in the full-size model.
For the RVE model, the maximum δT in the middle section was 9.96%, while for the full-size model,
the maximum δT in the middle section was 6.39%.

4.1.2. Heat Flux Field

To analyze the influence of ATC on the heat transfer in the 2.5-D braided CMC, Figure 8 shows
the heat flux distributions of the RVE model and the full-size model. From Figure 8a,b, it can be seen
that the heat flux fields were heterogeneous and that the heat flux of the diagonal braided yarns was
larger than the heat flux of the axial yarns on the Y-axis and the horizontal braided yarns on the x-axis.
The reason for this was that the axial and radial thermal conductivities of the fibers bundle were
9.66 W/(m·K) and 1.48 W/(m·K), and the thermal conductivity of the matrix was 6.5 W/(m·K). In the
diagonal braided yarns, there was a component of the bigger thermal conductivity (9.66 W/(m·K)) on
Appl. Sci. 2019, 9, x FOR PEER REVIEW 11 of 18
the z-axis, leading to an enhancement of the heat conduction on the z-axis.

(a) Surface of the RVE model (b) Surface of the full-size model

(c) Middle section of the RVE model (d) Middle section of the full-size model

Figure 8. The
The heat
heatflux
fluxfields
fieldsofofthe
theRVE
RVEmodel.
model.

Figure
Figure8c,d
8c,dshows
shows thethe distributions the heat
distributions of the heat flux
flux fluctuation
fluctuationδδhfhffor
forthe
themiddle
middlesections
sections
ofof the
the
ZZaxes
axesofofthe
thetwo
twomodels.
models.The Theresults
resultsshow
showthat
thatthe
theheat
heatflux
fluxfluctuations
fluctuationswere
weresignificant
significantand
andthat
thatthe
the distributions
distributions for those
for those two models
two models were similar.
were similar. For
For the RVEthemodel,
RVE model, the maximum
the maximum δhf wasδ255.58%.
hf was

255.58%.
For For themodel,
the full-size full-size
themodel, the maximum
maximum δhf 280.40%.
δhf reached reached 280.40%.

4.1.3.Effective
4.1.3. EffectiveThermal
Thermal Conductivity
Conductivity
Accordingto
According to the
the results
results mentioned
mentioned above,
above, the
theaverage
averageheat
heatflux
fluxon
onthe
thez-axis
z-axisofofthe RVE
the RVEmodel
model
was 56,958 W/m 2, the difference between the constant temperatures of top and bottom
was 56,958 W/m , the difference between the constant temperatures of top and bottom surfaces surfaces was
was
10K,K,and
10 andthe
thethickness
thicknessofofthe
theRVERVEmodel
modelwaswas0.6
0.6mm.
mm.InInthe
thefull-size
full-sizemodel,
model,the
theaverage
averageheatheatflux
fluxon
on z-axis
the the z-axis was 13,968
was 13,968 W/mW/m
2, the difference between the constant temperatures of the top and
2 , the difference between the constant temperatures of the top and bottom
bottom surfaces was 10 K, and the thickness was 2.8 mm. Therefore, the effective thermal
conductivities on the z-axes kZ of the two models could be calculated based on Fourier’s thermal
conduction equation.
The kZ values of the RVE model and the full-size model were 3.42 W/(mK) and 3.91 W/(mK),
respectively. The ETC kZ based on the RVE model and the ETC kZ based on full-size model had a
significant difference. The reason for this was that the thickness of the full-size model was small (2.8
Appl. Sci. 2019, 9, 875 11 of 16

surfaces was 10 K, and the thickness was 2.8 mm. Therefore, the effective thermal conductivities on
the z-axes kZ of the two models could be calculated based on Fourier’s thermal conduction equation.
The kZ values of the RVE model and the full-size model were 3.42 W/(m·K) and 3.91 W/(m·K),
respectively. The ETC kZ based on the RVE model and the ETC kZ based on full-size model had
a significant difference. The reason for this was that the thickness of the full-size model was small
(2.8 mm) and the periodicity was unsatisfied in the Z axis direction for the full-size model, leading to a
large difference for the periodic RVE model.
Furthermore, the effective thermal conductivities along the x- and y-axes (kX and kY ) could also
be calculated using a similar method. The kX values of the RVE model and full-size model were
5.63 W/(m·K) and 5.62 W/(m·K), respectively, and the kY values of the two models were 5.28 W/(m·K)
and 5.77 W/(m·K), respectively. A comparison between the ATCs based on the RVE model and full-size
model
Appl. Sci.is shown
2019, in Figure
9, x FOR 9.
PEER REVIEW 12 of 18

Figure 9.
Figure The effective
9. The effective thermal
thermal conductivities
conductivities based on the two models.

4.1.4. Validation of the RVE Model and the Full-Size Model

4.1.4. Validation of the RVE Model and the Full-Size Model
To validate the accuracies of the RVE model and the full-size model, the thermal conductivity
To validate the accuracies of the RVE model and the full-size model, the thermal conductivity of
of the z-axis kZ for the sample shown in Figure 1a was tested. A NETZSCH LFA 457 Micro-Flash
the z-axis kZ for the sample shown in Figure 1a was tested. A NETZSCH LFA 457 Micro-Flash
instrument was used to measure the thermal diffusion coefficient, and the specific heat capacity was
instrument was used to measure the thermal diffusion coefficient, and the specific heat capacity was
measured with a DSC 200 F3. The geometrical sizes of the samples were 10 mm × 10 mm × 2.8 mm
measured with a DSC 200 F3. The geometrical sizes of the samples were 10 mm × 10 mm × 2.8 mm
(length × width × height), which satisfied the demands of the test equipment. Finally, the thermal
(length × width × height), which satisfied the demands of the test equipment. Finally, the thermal
conductivity was calculated using the density, the thermal diffusivity, and the specific heat capacity.
conductivity was calculated using the density, the thermal diffusivity, and the specific heat capacity.
In the experiment, the k values at four different positions of the sample were tested.
In the experiment, the kZZ values at four different positions of the sample were tested.3
According to the experimental results, the density of the sample was 1.711 × 10 kg/m3 and the
According to the experimental results, the density of the sample was 1.711 × 103 kg/m3 and the
specific heat capacity was 0.688 × 1033 J/(kg·K). Table 2 shows the experimental data of the thermal
specific heat capacity was 0.688 × 10 J/(kg·K). Table 2 shows the experimental data of the thermal
diffusivity and the corresponding thermal conductivity on the Z axis. The thermal conductivities at
diffusivity and the corresponding thermal conductivity on the Z axis. The thermal conductivities at
four different positions on the sample were tested. The results show that the four different thermal
four different positions on the sample were tested. The results show that the four different thermal
conductivities were close to each other. The average k value was 4.053 W/(m·K).
conductivities were close to each other. The average kZZ value was 4.053 W/(mK).
Table 2. A comparison of the experimental and numerical data.
Table 2. A comparison of the experimental and numerical data.
Experimental Data Numerical Data
Experimental Data Numerical Data
Thermal Conductivity Mean
Mean Standard
Standard RVE
RVE Full-Size
Full-Size
Thermal Diffusivity/m2 /s Thermal conductivity
kZ /W/(m ·K) Value/W/(m·K) Deviation/W/(m·K) Model/W/(m·K) Model/W/(m·K)
Thermal Diffusivity/m2/s Value/W/(m Deviation/W/(m Model/W/(m Model/W/(m
kZ/W/(mK)
3.427 3.460 3.444 3.443 4.036 4.072 4.053 4.052 4.053
K) 0.015
K) 3.42
K) 3.91
K)
3.42 3.46 3.44 3.44 4.03 4.07 4.05 4.05
4.053 0.015 3.42 3.91
7 0 4 3 6 2 3 2
The numerical results in Section 4.1.3 are also given in Table 2. The numerical results of the
kZ valuesnumerical
The based onresults
the RVEin Section 4.1.3the
model and arefull-size
also given in Table
model 2. The
are 3.42 W/(mnumerical
·K) andresults of the·K),
3.91 W/(m kZ
values based on the RVE model and the full-size model are 3.42 W/(mK) and 3.91
respectively, and the relative differences compared to the experimental data are 15.62% and 3.53%, W/(mK),
respectively, andnumerical
respectively. The the relative differences
result based oncompared to model
the full-size the experimental data
is closer to the are 15.62% data.
experimental and 3.53%,
In the
respectively. The numerical result based on the full-size model is closer to the experimental
thermal analysis of thin-walled structures such as a turbine vane, the RVE model would lead to a large data. In
the thermal analysis of thin-walled structures such as a turbine vane, the RVE model would lead to
a large deviation in the estimation of the effective thermal conductivity. The reason for this is that the
main difference between the RVE model and the full-size model is the boundary part shown in Figure
10. The periodic hypothesis of the RVE model could not be satisfied in the CMC’s boundary parts,
and the volume fractions of the fibers and the braided structure are also different, which would lead
Appl. Sci. 2019, 9, 875 12 of 16

deviation in the estimation of the effective thermal conductivity. The reason for this is that the main
difference between the RVE model and the full-size model is the boundary part shown in Figure 10.
The periodic hypothesis of the RVE model could not be satisfied in the CMC’s boundary parts, and
the volume fractions of the fibers and the braided structure are also different, which would lead to
the variation of the heat transfer in these parts. In the RVE model and the full-size model, the volume
fractions of the fibers Ff were 54.34% and 39.61%, respectively. Therefore, the full-size model could
Appl. Sci. 2019, 9, x FOR PEER REVIEW 13 of 18
truly reflect
Appl. Sci. 2019, the internal
9, x FOR PEER structure
REVIEW of a CMC thin-wall structure. 13 of 18

Figure 10. A comparison between the boundary part and the periodic internal structure.
Figure 10. A comparison between the boundary part and the periodic internal structure.
4.2. Influence of the Thickness
4.2. Influence of
4.2. Influence of the Thickness
thetheThickness
Although full-size model can obtain more accurate results in comparison to the RVE model,
it Although
needs too the
manyfull-size
Although the full-size model resources
computing
model can obtain
can obtainwhen more the
more accurate results
geometric
accurate results in comparison
size
in comparison
is large. Theto toinfluence
the RVE
the RVE model,
model,
of the
it
it needs
thickness
needs too
toomany
on thecomputing
many ETC
computing resources
and the critical
resources when
thicknessthe geometric
when that
the was size is size
suitable
geometric large.isThe
for the RVE influence
large. model of the
are
The influence thickness
analyzed in
of the
on the
this ETC
section.and the critical thickness that was suitable for the RVE model
thickness on the ETC and the critical thickness that was suitable for the RVE model are analyzed in are analyzed in this section.
Figure
Figure
this section. 11a11a shows
shows thethe variation
variation curve
curve of theof the kZ with
kZ with increasing
increasing the thickness
the thickness LZcan
LZ . It . It can be seen
be seen that
kZ that kZdecreased
firstly
Figure firstly decreased
11a shows andthe
then and then to
tended
variation tended oftothe
be constant.
curve bekconstant.
When LZWhen reached
Z with increasing
LZthe
reached
18.4 mm,18.4
thickness mm,
the Lrelative thevariation
relative
Z. It can be seen
of variation of only
the k0.51%,
Z was only 0.51%, which could be treated as convergent. When theequal
LZ was equal to
that kZ firstly decreased and then tended to be constant. When LZ reached 18.4 mm, the 2.8
the k Z was which could be treated as convergent. When the L Z was to mm
relative
and 2.818.4
mmmm, and the
18.4corresponding
mm, the correspondingk values k values
Zwere were
3.91 3.91
W/(m · W/(mK)
K) and and
3.53 3.53
W/(m W/(mK),
· K), respectively,
respectively, and
variation of the kZ was only 0.51%,Zwhich could be treated as convergent. When the LZ was equal to
the and the relative
relative variation was 10.76%. The results show that the thickness ahad ainfluence
big influence onETC
the
2.8 mm andvariation
18.4 mm, was 10.76%.
the corresponding The results showwere
kZ values that3.91
the thickness
W/(mK) and had 3.53 bigW/(mK), on the
respectively,
k ETC
Z ofthethekrelative
of the full-size
Zfull-size model. was model.
The reason The reason for this was volume
that thefraction
volumeoffraction of the fibers Vf
and variation 10.76%.for Thethis was
results that
showthethat the thickness hadthe fibers
a big Vf increased
influence on the
withincreased
increasing with L Z
increasing
, as shown L in
Z , as shown
Figure in
11b. Figure
For 11b.
example,For example,
when L Z
when
was L
equal Z was
to equal
2.8 mm to 2.8
and mm mm,
18.4 and
ETC kZ of the full-size model. The reason for this was that the volume fraction of the fibers Vf
the 18.4 mm,
corresponding the corresponding
Vf values V f values were 39.61% and 51.92%, respectively. The variation of Vf
increased with increasing LZ, were
as shown 39.61% and 51.92%,
in Figure 11b. Forrespectively.
example, when The variation
LZ was equal of Vto f changed
2.8 mm and the
changed of
percentage thethe
percentage of the components with different thermal properties, which further affected
18.4 mm, the corresponding Vf values were 39.61% and 51.92%, respectively. The variation ofthe
components with different thermal properties, which further affected the ETC of Vf
the ETC
full-size of the full-size model.
model.
changed the percentage of the components with different thermal properties, which further affected
the ETC of the full-size model.

(a) kZ (b) Volume fraction Ff

Figure
Figure 11.11.
TheThe curves
curves of of
thethe
kZkand
Z and thevolume
the volumefraction
fractionwith
withthe
theLLZZ increasing.
(a) kZ (b) Volume fraction Ff
At At
thethe
same time,
same
Figure according
time,
11. The to the
according
curves results
tothe
of the inthe
kZresults
and Section
involume4.1.3,
Section the ETC
4.1.3,
fraction the
with kthe
ETCZ based on the
on RVE
kLZZ increasing.
based model
the RVE was
model
3.42 W/(m
was 3.42·W/(mK),
K), as shown by theby
as shown redthedashed line inline
red dashed Figure 11. It11.
in Figure canIt be
canseen that that
be seen the line based
the line on the
based on
theAtfull-size model tended to converge on the red dashed line of the RVE model.
the same time, according to the results in Section 4.1.3, the ETC kZ based on the RVE model When L Z changed
wasfrom
3.422.8 mm to 18.4
W/(mK), mm, the
as shown byrelative
the red variation
dashed line between the results
in Figure based
11. It can on the
be seen full-size
that the linemodel
basedand
on
theRVE model
full-size varied
model from 12.53%
tended to 3.12%.
to converge on theTheredresults indicate
dashed line ofthat
thewhen
RVEthe thickness
model. WhenLZLwas large,
Z changed
the2.8
from ETC mm kZ to
based
18.4onmm,thethe
RVE modelvariation
relative and the ETC kZ based
between on the full-size
the results based onmodel were close
the full-size to each
model and
other. The reason for this was that when the L Z was large enough, the periodicity was approximately
RVE model varied from 12.53% to 3.12%. The results indicate that when the thickness LZ was large,
Appl. Sci. 2019, 9, 875 13 of 16

full-size model tended to converge on the red dashed line of the RVE model. When LZ changed from 2.8
mm to 18.4 mm, the relative variation between the results based on the full-size model and RVE model
varied from 12.53% to 3.12%. The results indicate that when the thickness LZ was large, the ETC kZ based
on the RVE model and the ETC kZ based on the full-size model were close to each other. The reason for
this
Appl.was that9,when
Sci. 2019, x FORthePEERLZREVIEW
was large enough, the periodicity was approximately satisfied in the14 z-axis
of 18
direction for the full-size model, leading to a small difference with the periodic RVE model and that the
Frelative
f of these variation
two modelsof only
were 4.45%.
closeFor a full-size
to each other. model with LZF=f of
For example, 2.8the
mm, RVE this relative
model andvariation reached
Ff of the full-size
27.11%.
model with LZ = 18.4 mm were 54.34% and 51.92%, respectively, with a relative variation of only 4.45%.
For a For the operating
full-size model with condition
LZ = 2.8 mm,discussed in thisvariation
this relative section, reached
when LZ27.11%. was bigger than 18.4 mm, the
relative
For variation was smaller
the operating condition thandiscussed
5%, so theinETC thisksection,
Z based on whenthe RVE
LZ was model could
bigger represent
than 18.4 mm, the
result
the basedvariation
relative on the full-size
was smallermodel. The
than 5%,corresponding
so the ETC kZLzbased of 18.4 onmm was determined
the RVE model couldasrepresent
the “critical
the
thickness”,
result basedwhich
on thewas nearly
full-size 31 times
model. Thegreater than theLz
corresponding thickness
of 18.4 mm of thewasRVE model. as the “critical
determined
thickness”, which was nearly 31 times greater than the thickness of the RVE model.
4.3. Results with Different ATCs
4.3. Results with Different ATCs
The results discussed above were based on the 2.5-D braided CMC sample, as shown in Figure
1. In the results
The discussed
engineering above were
application, based on the
the material’s ATCs2.5-D braided
were differentCMC if sample,
the internalas shown in Figure
components and1.
In
thethe engineering
structure wereapplication,
changed. To the make
material’s
the ATCs
present were different
results and ifconclusions
the internalmore components and the
applicable, the
structure
influencewere of thechanged.
thickness Toandmake
thethe present
critical resultsthat
thickness andwereconclusions
suitablemore for theapplicable,
RVE model thewere
influence of
further
the thickness and the critical thickness that were
investigated under different operating conditions of the ATCs. suitable for the RVE model were further investigated
underFiguredifferent operating
12 shows the conditions of the ATCs.kZ based on the RVE model and the full-size model
thermal conductivities
Figure 12 shows the thermal conductivities
with an increasing thickness LZ, when the absolute kZ based
value on ratiothenRVEand model
the level andof the full-size rmodel
anisotropy of the
with an increasing thickness
thermal conductivities of the fiber L Z , when the absolute
bundles were changed. value ratio n and the level of anisotropy r of the
thermal Theconductivities
results in Figure of the
12afiber
show bundles
that whenwerethechanged.
values of kζ, kη, and kν were different, the influence
The results in Figure 12a show that when
of the thickness was also changed. In Case 1 and Case 2, the values kζ , kvalue
of the η , andof kν kwere different,
Z decreased the the
with influence
increaseof
the
of Lthickness
Z, while in was also 3–5,
Cases changed. In Case 1with
kZ increased andtheCase 2, the value
increase of LZof kZ decreased
. Figure 12b gives withthethe increaselaw
influence of Lof
Z,
while
the levelin Cases 3–5, kZ increased
of anisotropy r. It can with
be the
seen increase of LZ .r Figure
that when increased,12b gives
the ktheZ’s influence
variation law of the
curves level
always
of anisotropy r. decreased,
monotonically It can be seen butthat
thewhen
ranger increased, the kZ ’sThe
became smaller. variation
reasoncurves always
for this was monotonically
that the main
decreased, but the range
difference between the RVE became
model smaller.
and theThe reasonmodels
full-size for thiswas wasthe that the main
existence of difference
the boundary between
parts
the RVE model and the full-size models was the existence of the boundary
and that its influence was affected by the relative variations between the thermal conductivities of parts and that its influence
was
the affected by theand
fiber bundles relative variations
the matrix, between
which the thermal
changed with the conductivities
absolute value of the fiber
ratio n bundles
and the and levelthe
of
matrix,
anisotropy which r. changed with the absolute value ratio n and the level of anisotropy r.

(a) The absolute value ratio n (b) The level of anisotropy r

Figure 12.
Figure 12. The
The variations
variations of
of the
the thermal
thermal conductivities
conductivitieskkZZ with different
different ATCs.
ATCs.

Additionally,
Additionally, the
the results in Figure
Figure 12
12 show
showthat
thatwhen
whenthe
theabsolute
absolutevalue
valueratio
ratioand
and
thethe level
level of
of anisotropy
anisotropy changed,
changed, thethelines
linesbased
basedononthethefull-size
full-sizemodels
models always
always tended
tended toto converge on the the
corresponding
corresponding dash
dash lines
lines of
of the
the RVE
RVEmodels.
models. This
This phenomenon
phenomenon waswas explained
explainedininSection
Section4.2.
4.2. In
In this
this
paper,
paper,the
theconvergence
convergencecriteria
criteriawas
wasset
setto
tothat
thatstandard,
standard,and
andthe
the relative
relative variation
variation between
between the
the results
results
based on the RVE model and the full-size model was smaller than 5%. Figure 13 shows the variation
of the “critical thickness” LZ under different ATCs. The results show that when the absolute value
ratio n increased, the “critical thickness” LZ first decreased and then increased. When the absolute
value ratios were 1, 2, 4, 6, and 10, the corresponding critical thicknesses were 11.2 mm, 2.8 mm, 4.0
mm, 7.6 mm, and 11.2 mm, respectively. When the level of anisotropy r increased, the “critical
Appl. Sci. 2019, 9, 875 14 of 16

based on the RVE model and the full-size model was smaller than 5%. Figure 13 shows the variation of
the “critical thickness” LZ under different ATCs. The results show that when the absolute value ratio
n increased, the “critical thickness” LZ first decreased and then increased. When the absolute value
ratios were 1, 2, 4, 6, and 10, the corresponding critical thicknesses were 11.2 mm, 2.8 mm, 4.0 mm,
7.6 mm, and 11.2 mm, respectively. When the level of anisotropy r increased, the “critical thickness”
LzAppl.
decreased almost monotonously. When the levels of anisotropy r were 6.53, 10, 13, 16,15and
Sci. 2019, 9, x FOR PEER REVIEW
20,
of 18
the corresponding critical thicknesses were 11.2 mm, 7.6 mm, 7.6 mm, 4.0 mm, and 2.8 mm, respectively.

(a) The absolute value ratio n (b) The level of anisotropy r

Figure13.
Figure 13.The
Thevariations
variations of
of the critical thickness
thickness with
withdifferent
differentATCs.
ATCs.

InIngeneral,
general,thethestructure
structureofofthe
the boundary
boundary part of the braided
braided CMC,
CMC, which
whichwaswasdifferent
differentforfora a
periodic internal structure, was the main reason
periodic internal structure, was the main reason for for the difference between the results of the full-size
between the results of the full-size
model
modeland andthetheRVE
RVEmodel.
model.This
Thisinfluence
influence depended the value
depended on the value and
andthe
thelevel
levelof
ofanisotropy
anisotropyofofthethe
ATC,
ATC, i.e.,
i.e., thetherelative
relativedifference
differencebetween
betweenthethethermal
thermalconductivities
conductivities of
of the matrix and the
the fibers.
fibers.

5.5.Conclusions
Conclusions
AAnumerical
numericalstudystudyofofthe theanisotropic
anisotropicthermalthermalconductivities,
conductivities,especially
especiallythe thethermal
thermalconductivity
conductivityin
the
inthickness direction,
the thickness direction,of the 2.5-D
of the braided
2.5-D braided CMC CMCmaterial
material was wascarried
carried out.
out.TwoTwo prediction
prediction models,
models,an
an model
RVE RVE model
with awith a periodicity
periodicity hypothesis hypothesis and a full-size
and a full-size model with modelactualwith actual thickness,
thickness, were
were established.
Anestablished.
experimental An experimental
test of the thermal test ofconductivity
the thermal conductivity
in the thickness in thedirection
thicknesswas direction was to
also used also used
validate
to validate
these these two
two prediction prediction
models. The models.
influenceThe influence
of the thicknessof theon thickness
the ATC and on thethe ATC and
critical the critical
thickness that
thickness that was suitable for the RVE model were further investigated
was suitable for the RVE model were further investigated under different operating conditions of the under different operating
conditions
thermal of the of
properties thermal properties
the fibers. The main of the fibers. Thecan
observations main observations as
be summarized can be summarized as
follows:
follows:
(1)(1) The
Thetemperature
temperaturefield fieldandandthethe heat
heat flux
flux field inside the
field inside the 2.5-D
2.5-Dbraided
braidedCMC CMCmaterial
materialwere
wereclearly
clearly
heterogeneous, and these fields were affected by the difference
heterogeneous, and these fields were affected by the difference between the thermal between the thermal conductivities
ofconductivities
the fiber bundles of theand matrix.
fiber bundles For and
example,
matrix.in For
the full-size
example,model,in the the relative
full-size fluctuation
model, of the
the relative
temperature field and the relative fluctuation of the heat flux field
fluctuation of the temperature field and the relative fluctuation of the heat flux field in the in the middle section reached
6.39%
middleand 280.40%,
section respectively.
reached 6.39% and 280.40%, respectively.
(2)(2) InInthe
the thermal analysis of aa thin-walled
thermal analysis of thin-walled structure, such such as as aa turbine
turbinevane,vane,thetheRVERVEmodel
modelwould
would
lead
leadtotoa large
a largedeviation
deviation in the
in estimation
the estimation of theofeffective thermal
the effective conductivity
thermal so thatso
conductivity thethat
periodic
the
hypothesis could not be
periodic hypothesis couldsatisfied.
not beThe relativeThe
satisfied. variation
relativeofvariation
the thermal conductivity
of the based on the
thermal conductivity
RVE
basedmodel
on the compared
RVE model withcompared
the experimental
with the data was 15.62%,
experimental datawhile the relative
was 15.62%, whilevariation was
the relative
variation
only 3.53%was when only the3.53% when
full-size the full-size
model model was applied.
was applied.
(3)(3) When
Whenthe thethickness
thicknessincreased,
increased, the the effective thermal conductivities
conductivitiesbased basedon onthetheRVE
RVEmodel
modelandand
the full-size model were close to each other. For the ATC of the sample
the full-size model were close to each other. For the ATC of the sample used in this study, when used in this study, when
thethickness
the thicknesswas wasbigger
bigger thanthan thethe critical
critical thickness
thickness of of 18.4
18.4mm,mm,the theRVERVEmodel
modelwas wassuitable
suitablefor
for
theprediction
the predictionof ofthe
theATC.
ATC.
(4)(4) When
Whenthe theabsolute
absolutevaluevalueratioratioand andthethe level
level of anisotropy
of anisotropy of the
of the thermal
thermal conductivities
conductivities of fiber
of the the
fiber bundle were changed, the influence of the thickness on the thermal conductivity was
bundle were changed, the influence of the thickness on the thermal conductivity was different
different and the critical thickness for the RVE model changed. When the absolute value ratio
and the critical thickness for the RVE model changed. When the absolute value ratio increased,
increased, the critical thickness firstly decreased and then increased, and the critical thickness
decreased almost monotonously with increasing the level of anisotropy.

Author Contributions: Conceptualization, Zecan Tu and Junkui Mao; methodology, Zecan Tu; software, Zecan
TuX.X.; validation, Zhenzong He and Xingsi Han; formal analysis, Zecan Tu; investigation, Zecan Tu; resources,
Appl. Sci. 2019, 9, 875 15 of 16

the critical thickness firstly decreased and then increased, and the critical thickness decreased
almost monotonously with increasing the level of anisotropy.

Author Contributions: Conceptualization, Z.T. and J.M.; methodology, Z.T.; software, Z.T., X.X.; validation, Z.H.
and X.H.; formal analysis, Z.T.; investigation, Z.T.; resources, Z.T.; data curation, Z.T.; writing—original draft
preparation, Z.T.; writing—review and editing, X.H.; visualization, Z.H.; supervision, J.M.; project administration,
J.M.; funding acquisition, Z.T., Z.H. and X.H.
Funding: This research was funded by the China Postdoctoral Science Foundation, grant number 2018M642248,
the National Natural Science Foundation of China, grant number 51806103 and the Jiangsu Provincial Natural
Science Foundation of China, grant number BK20160794.
Acknowledgments: Xingsi Han acknowledges the support of the Jiangsu Specially-Appointed Professor Program.
Conflicts of Interest: The authors declare no conflict of interest.

Nomenclature
Cp Specific heat capacity at constant pressure (J/(kg·K))
F Volume fraction
k Thermal conductivity (W/(m·K))
L Thickness (mm)
n Absolute value ratio comparing to the sample’s ATC
r Anisotropy’s level
T Temperature (K)
→
ν Velocity (m/s)
X, Y, Z Global Cartesian coordinates
Greek symbols
α Rotation angles around the x-axis between the PDTC coordinates and the global coordinates
(◦ )
β Rotation angles around the y-axis between the PDTC coordinates and the global coordinates
(◦ )
γ Rotation angles around the z-axis between the PDTC coordinates and the global coordinates (◦ )
ρ Density (kg/m3 )
ζ, η, ν Local Cartesian coordinates
δ Relative fluctuation
Subscripts
f Fiber
ij Coordinates of the mesh nodes

Abbreviations
ATC Anisotropic Thermal Conductivity
CMC Ceramic Matrix Composite
ETC Effective Thermal Conductivity
FEM Finite Element Method
PDTC Principal Direction of Thermal Conductivity
RVE Representative Volume Element

References
1. Fang, C.D. Development Research of Aeroengine; Aviation Industry Press: Beijing, China, 2009.
2. Hiroshi, K. The application of ceramic-matrix composites to the automotive ceramic gas turbine. Compos. Sci.
Technol. 1999, 59, 861–872.
3. Deng, Y.; Li, W.G.; Wang, R.Z. The temperature-dependent fracture models for fiber-reinforced ceramic
matrix composites. Compos. Struct. 2016, 140, 534–539. [CrossRef]
4. Yang, J.S.; Dong, S.M.; Xu, C.Y. Mechanical response and microstructure of 2D carbon fiber reinforced
ceramic matrix composites with SiC and Ti3SiC2 fillers. Ceram. Int. 2016, 42, 3019–3027. [CrossRef]
Appl. Sci. 2019, 9, 875 16 of 16

5. Michael, V.; Anthony, C.; Robinson, R.C.; Thomas, D.J. Ceramic Matrix Composite Vane Subelement Testing in a
Gas Turbine Environment; ASME: New York, NY, USA, 2004.
6. Heidmann, J.D.; Kassab, A.J.; Divo, E.A.; Rodriguez, F.; Steinthorsson, E. Conjugate Heat Transfer Effects on a
Realistic Film-Cooled Turbine Vane; ASME: New York, NY, USA, 2003.
7. Behzad, T.; Sain, M. Measurement and prediction of thermal conductivity for hemp fiber reinforced
composites. Polym. Eng. Sci. 2007, 10, 977–983. [CrossRef]
8. Xu, Y.B.; Yagi, K. Automatic FEM model generation for evaluating thermal conductivity of composite with
random materials arrangement. Comput. Mater. Sci. 2004, 30, 242–250. [CrossRef]
9. Lebel, L.; Turenne, S.; Boukhili, R. An experimental apparatus and procedure for the simulation of thermal
stresses in gas turbine combustion chamber panels made of ceramic matrix composites. J. Eng. Gas
Turbines Power 2017, 139, 091502. [CrossRef]
10. Tu, Z.C.; Mao, J.K.; Jiang, H.; Han, X.; He, Z. Numerical method for the thermal analysis of a ceramic matrix
composite turbine vane considering the spatial variation of the anisotropic thermal conductivity. Appl. Therm.
Eng. 2017, 127, 436–452. [CrossRef]
11. Xin, L.I.U.; Xiuli, S.H.E.N.; Longdong, G.O.N.G.; Peng, L.I. Multi-scale thermodynamic analysis method for
2D SiC/SiC composite turbine guide vanes. Chin. J. Aeronaut. 2018, 31, 117–125.
12. Kiani, S.; Chan, W.S.; Sheikh, A.H. Measurement of thermal conductivity of triaxial braided composites.
J. Compos. Mater. 2008, 42, 12. [CrossRef]
13. Jiang, L.L.; Xu, G.D.; Cheng, S. Predicting the thermal conductivity and temperature distribution in 3D
braided composites. Compos. Struct. 2014, 108, 578–583. [CrossRef]
14. Dasgupta, A.; Agarwal, R.K. Orthotropic thermal conductivity of plain-weave fabric composites using a
homogenization technique. J. Compos. Mater. 1992, 26, 2736–2758. [CrossRef]
15. Ning, Q.G.; Chou, T.W. Closed-form solutions of the in-plane effective thermal conductivities of woven-fabric
composites. Compos. Sci. Technol. 1995, 55, 41–48. [CrossRef]
16. Bhattacharjee, D.; Kothari, V.K. A theoretical model to predict the thermal resistance of plain woven fabrics.
Indian J. Fibre Text. Res. 2005, 30, 252–257.
17. Hill, R. Elastic properties of reinforced solids: Some theoretical principles. J. Mech. Phys. Solids 1963, 11,
357–372. [CrossRef]
18. Siddiqui, M.O.R.; Sun, D.M. Finite element analysis of thermal conductivity and thermal resistance behavior
of woven fabric. Comput. Mater. Sci. 2013, 75, 45–51. [CrossRef]
19. Ai, S.G.; He, R.J.; Pei, Y.M. A numerical study on the thermal conductivity of 3D woven C/C composites at
high temperature. Appl. Compos. Mater. 2015, 22, 823–835.
20. Fang, W.Z.; Chen, L.; Gou, J.J.; Tao, W.Q. Predictions of effective thermal conductivities for three-dimensional
four-directional braided composite using the lattice Boltzmann method. Int. J. Heat Mass Transfer 2016, 92,
120–130. [CrossRef]
21. Islam, M.R.; Pramila, A. Thermal conductivity of fiber reinforced composites by FEM. J. Compos. Mater. 1999,
33, 1699–1715. [CrossRef]
22. Klett, J.W.; Ervin, V.J.; Edie, D.D. Finite-element modeling of heat transfer in carbon/carbon composites.
Compos. Sci. Technol. 1999, 59, 593–607. [CrossRef]
23. Car, E.; Zalamea, F.; Oller, S.; Miquel, J.; Qnate, E. Numerical simulation of composite materials: Two
procedures. Int. J. Solid Struct. 2002, 39, 1967–1986. [CrossRef]
24. Dong, K.; Liu, K.; Pan, L.J.; Gu, B.H.; Sun, B.Z. Experimental and numerical investigation on the thermal
conduction properties of 2.5D angle-interlock woven composites. Compos. Struct. 2016, 154, 319–333. [CrossRef]
25. Guan, T.R. Micro Geometry and Mechanical Model and Experimental Study of 2.5D Braided Quartz/SiO2 Ceramic
Matrix Composites; Nanjing University of Aeronautics and Astronautics: Nanjing, China, 2012.
26. COMSOL. COMSOL Multiphysics Help—Theory for the Heat Transfer, Release 4.3, Sweden. 2011.
27. Ree, Y.; Lin, J.R.; Zhu, J.G. Coordinate transformation uncertainty analysis in Larger-Scale metrology.
IEEE Trans. Instrum. Meas. 2015, 64, 2380–2388.
28. He, F. Carbon Fiber and Graphite Fiber; Chemical Industry Press: Beijing, China, 2010.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).