Thin Solid Films 517 (2009) 5120–5129
Contents lists available at ScienceDirect
Thin Solid Films
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t s f
High emissivity coatings for high temperature application: Progress and prospect
Xiaodong He a,⁎, Yibin Li a, Lidong Wang a,b, Yue Sun a, Sam Zhang c
a
b
c
Center for Composite Materials, School of Astronautics, Harbin Institute of Technology, P.O. Box 3010, Harbin 150080, PR China
School Materials Science and Engineering, Harbin Institute of Technology, P.O. Box 405, Harbin 150001, PR China
School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore
a r t i c l e
i n f o
Available online 22 April 2009
Keywords:
Emissivity
Coatings
Review
a b s t r a c t
High emissivity coatings are widely used in many cases where heat transfers through electromagnetic
radiation that arises due to the temperature of a body. Extensive theoretical and experimental efforts have
been made to synthesize and investigate high emissivity coatings. The emissivity can be improved through
various or combined mechanisms. The characterization of the emissivity is still a fully open problem. In this
paper, we review the various mechanisms associated with the emissivity enhancement and emissivity
characterization techniques. Based on these literature reviews, the prospect will be presented in the
concluding remarks.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
High emissivity coatings are widely used in many high temperature applications to effectively transfer the heat by radiation [1–4]. For
example, the surface of Ni-, Fe-, Co-based alloys applied in metal
thermal protection systems (MTPS) [5,6] in third generation of reused
space vehicles involves high friction heat from acute friction between
space vehicle surface and atmosphere, which causes obvious increase
of surface temperature up to 1000 °C during hypervelocity flight [7–9].
As a result, the lifetime and performance of MTPS will be seriously
degraded. Therefore, the high emissivity coating is intentionally
deposited on MTPS to decrease the surface temperature by radiation.
When a radiation falls on a body it may be partially reflected,
transmitted, or absorbed, which is associated with reflection (R),
absorption (A), transmission (T), and emission (ε). The principle of
conservation of energy ties together these radiation characteristics as
A + R + T = 1. According to Kirchhoff's law, at equilibrium for a given
wavelength λ and temperature T, the emissivity of any body is equal to
its absorption [10], i.e. ε(ν, T, θ) =A(ν,T,θ), where ε is the emissivity, ν
the frequency, and the angle θ specifies the observer angle. This
indicates that the emissivity of an object can be indirectly obtained by
measuring its absorptivity. For a uniform and isotropic opaque surface in
thermal equilibrium, the transmission is zero, the relationship between
the emissivity and the reflectivity is: ε(ν,T, θ) =A(ν,T,θ) = 1 −R(ν,T,θ).
Many materials can be considered as opaque body. The emissivity here is
characterized to describe the surface radiative property which involves
the transfer of heat by electromagnetic radiation arising due to the
temperature of a body. The emissivity is defined as the ratio of energy
⁎ Correspondence author.
E-mail address: [email protected] (X. He).
0040-6090/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.tsf.2009.03.175
radiated by the material to energy radiated by a black body (A body that
emits the maximum amount of heat for its absolute temperature is
called a black body, meaning that a blackbody completely absorbs all
radiation incident upon it, and at the same time it emits all the energy
that it absorbs with the same absorbing spectrum. That is to say, ε = 1)
at the same temperature, which is described by Eq. (1):
R∞
eðT Þ =
0
eðλÞEλ dλ
σ T4
ð1Þ
Where E is radiant heat of gray body to its surroundings, ε
hemispherical emissivity of the gray body (dimensionless), σ Stefan–
Boltzman constant (5.67 × 10− 8 W/(m2×K4)), T temperature (K). The
emittance is dependent on direction and wavelength, thus the
radiance ability can also be evaluated by the spectral emissivity and
directional emissivity, described as Eqs. (2) and (3) respectively:
eλ ðλ; T Þ =
Eλ;actual emitted ðλ; T Þ
E ðλ; T Þ
= λ
Eλ;black body ðλ; T Þ
Ebλ ðλ; T Þ
R∞
eθ ðθ; /; T Þ =
0
Lλ;actual emitted ðλ; θ; /; T Þdλ
Lðθ; T Þ
R∞
=
L
ð
λ;
T
Þdλ
Lb ðT Þ
0 λ;black body
ð2Þ
ð3Þ
A real object does not radiate as much as a perfect black body, which
radiates less heat than a black body and is called gray body (ε b 1).
Thermally emitted radiance from any surface mainly depends on
two factors. (1) the surface temperature, which is an indication of the
equilibrium thermodynamic state resulting from the energy balance
of the fluxes between the gray body surface and its surroundings; and
(2) the surface emissivity, which is the efficiency of the surface for
transmitting the radiant energy generated in the surface into its
surroundings. The latter depends on the temperature (but the
relationship between emissivity and temperature is not definite,
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
5121
depending on material nature, surface parameters and wavelength),
composition, surface roughness, coating thickness, wavelength, and
physical parameters of the surface. Here we review recent progress
with emphasis on the discussion of the design methodology, followed
by briefing the emissivity characterization. Lastly, the prospects
for research and technology in high emissivity coatings will be
summarized.
2. Design methodology
2.1. Doping
According to Wien's displacement law and Planck's law [11], most
energy of the black body at high temperature is radiated in the
wavelength range of 1–5 μm, however it happens that many materials
have the weak intrinsic absorption within this range. As a result, the
emissivity in this region is quite low since the object's absorptivity and
emissivity spectra are identical in thermal equilibrium state. Extensive
work proves that total emissivity or spectral emissivity could be
effectively enhanced via doping (doping refers to the process of
intentionally introducing impurities into an extremely pure material).
Several mechanisms for doping enhanced emissivity in the wavelength range of 1–5 μm were proposed for different materials, namely,
free carrier absorption mechanism (the phenomenon whereby an
electron within a band absorbs radiation by transferring from a lowenergy level to an empty high-energy level) [12], d–d transitions in
octahedral coordination [13,14], distortion of the crystal lattice [15].
For semiconductor material, free carrier absorption mechanism
generally dominates the emissivity enhancement in the wavelength
region of 1–5 μm. Fig. 1 shows the absorption coefficient as a function
of wavelength for three different SiC specimens with different
impurity (dopant) concentrations [16]. It is obvious that increase in
impurity concentration strengthens the absorption in the wavelength
range of 1–5 μm. As higher impurity concentration causes more free
electrons, more electrons within a band absorb more radiation by
transferring from a low-energy level to an empty high-energy level,
leading to more absorption. As such, the emissivity is accordingly
enhanced within this wavelength range as the impurity concentration
increases since the object's absorptivity and emissivity spectra are
identical in thermal equilibrium state.
In glassy oxide coatings, different excitations are responsible for
the absorption of electromagnetic radiation [17]. It is suggested that
suitable dopants with transition metal oxides diluted in silicate
glasses may yield emissivity as high as 0.9 in the whole temperature
range. For example, the d–d transitions of Cr3+ in octahedral
coordination play an important role on the absorption [13,14]. The
Fig. 2. The emissivity spectra at room temperature for 0.1%, 0.25% and 0.5% Cr2O3 doped
compounds [17].
emissivity spectra at room temperature for 0.1%, 0.25% and 0.5% Cr2O3
doped compounds as shown in Fig. 2 indicates that the concentration
of Cr2O3 dramatically enhanced the normal spectral emissivity
between 1 and 5 μm (i.e. 2000–10,000 cm− 1). This enhancement is
related to formation or precipitation of new phases on the grain
boundary due to substitution, which will change the grain boundary
area. But the mechanism still needs to be further digested.
Another consequence caused by dopant is the crystal distortion due
to different ion radii, which also has the contribution to radiation. Pratt
et al. incorporated NiO into MgO–Al2O3–SiO2 system glasses heated at
1020 °C for 2 h and the whole-band normal direction emissivity
changes from 0.82 to 0.90 by doping NiO [18]. When the Ni2+ is
incorporated into cordierite structure and substituted with Mg2+, the
order degree of Al/Si in crystal structure of cordierite is decreased due
to different radii of Ni2+ and Mg2+, leading to the crystal structure
distortion. This in turn decreases the symmetry of lattice vibration,
thus enhances the effects of anharmonic vibration of polar lattice, the
coupled action of phonon, and phonon combination radiation. Therefore, the infrared radiance of this material is enhanced by NiO doping.
Moreover, the dopant (rare-earth oxides and transition elements)
could tune the spectral emissivity thus change the emissive power
[19,20]. Incorporation of 2–4 wt.% Co3O4 or NiO into MgO host (a low
infrared emissivity) produced “matched emitters”, which means its
emissive power spectrum is matched very efficiently to the response
of GaSb photovoltaic cells that convert the infrared radiation into
electricity. Fig. 3 shows the continuous, strong radiant emissions
which is up to 0.9 in the optimal energy range between 1 and 2 μm and
minimal radiation at higher wavelengths [21,22]. The total emissive
power between 1 and 9 μm is calculated by integrating the area under
each curve. Interestingly, both the blackbody emitter and the NiOdoped MgO emitter have the same total emissive power of 144,000 W
cm− 2. However, it is noticeable that the NiO-doped MgO emitter
emits most of power at wavelengths less than 2 μm, instead of 1–5 μm
in theoretical blackbody. The phenomenon results from intra-atomic
electronic transitions determined by the electronic configuration of
the dopant ions and interactions with the coordinating crystal field of
the host oxide [23].
2.2. Surface roughness
Fig. 1. Absorption coefficient as a function of wavelength of SiC. The three curves are for
SiC specimens with different impurity concentrations [16].
The correlation between emissivity and surface roughness has
been explored both theoretically [24,25] and experimentally[26,27].
Wen et al. [24] modeled the effect of surface roughness on emissivity
of aluminum alloys. Normally the surface involves two broad
categories: optically smooth (ideal) and rough (real) surfaces. For
5122
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
Fig. 3. Theoretical blackbody emissive power spectrum at 1010 °C compared with the emissive power spectrum for an experimental 2 wt.% NiO-doped MgO emitter at 1404 °C. [22]
optically smooth surfaces, spectral emissivity is determined by
combining Fresnel's equation and Kirchhoff's law yielding
eλ =
4n2
ðn + 1Þ2 + k2
ð4Þ
Where n is the index of refraction and k the extinction coefficient.
Fig. 4(a) compares spectral emissivity values using three different
methods. It indicates that the experimental result shows the same
trend as calculated one. Namely, the emissivity decreases as
wavelength increases in the infrared range for most metallic surfaces.
For real surfaces, the surfaces can be divided into three regions
based on optical roughness (σ/λ, the ratio between surface roughness
to wavelength): the specular region (0 b σ/λ b 0.2), intermediate
region (0.2 b σ/λ b 1) and the geometric region (σ/λ N 1). In the
specular region, it assumes that the reflection of incident radiation
is specular, and the diffraction theory is used to predict the effects of
surface roughness on emissivity [28,29]. For a Gaussian distribution of
surface heights, the relationship between reflectance and optical
roughness follows an exponential decay function of σ/λ, which is
identified both theoretically [30,31] and experimentally [32,33],
4πσ 2
ρr = ρp exp −
λ
ð5Þ
where ρr and ρp are the reflectances for a rough surface and a polished
surface, respectively. Eq. (5) reveals that the larger α could result in
lower reflectance, i.e. higher emittance, which is verified in aluminum-coated ground glass as shown in Fig. 4(b) [29]. In the
intermediate region, the directional reflectance ρ λ′ (θ i) can be
determined by integrating the bidirectional reflectance function, ρλ″
(θi, θs) over the respective hemisphere for a given angle of incidence
[34] as indicated in Fig. 4(c). According to Kirchhoff's law and energy
conservation, the directional emissivity can be calculated by Eq. (6)
ignoring the diffraction effects. Surfaces with repeatable grooved
finish, like V-shaped grooves [40], circular grooves [41], and pyramidal
grooves [42], are commonly used to model the emissivity enhancement. The slope of the grooved surface plays an important role in
emissivity enhancement. For example, the inter-reflection caused by a
60° V-groove in Fig. 4(c) and a small grove angle (also the slope) in
Fig. 4(d) could significantly enhance the normal emissivity. Overall,
modeling results lead to a conclusion that the emissivity is dependent
on surface roughness for σ/λ b 1 and on surface slope for σ/λ N 1.
In experiment, the emissivity enhancement due to roughness was
verified in opaque materials, such as steel [26] and Pr2NiO4 coatings
[27]. Fig. 5 compares normal spectral emissivity of rough Pr2NiO4
coating and single crystal Pr2NiO4 (considered as smooth surface) in
the (a–b) plane at 1000 K, in the working temperature range of the
infrared radiator, from far to midinfrared. The spectral response of the
coating is 30% to 40% higher than that of the single crystal, which is
largely dominated by the infrared absorptions produced by localized
polarons. Clearly, the enhancement of the intrinsic absorption is
obtained by increasing the optical path of the light between each
small cavity of the surface. Furthermore, the effect produced by the
rough Pr2NiO4 + d coating is also clearly demonstrated in Fig. 5. The
low normal spectral emissivity of the original semitransparent
ceramic in the midinfrared range (from 2500 to 5000 cm− 1) is
completely occulted after the deposit.
Although the previous achievements provide valuable insight into
the emissivity principles concerning the relationship between
emissivity and surface roughness, the influences of surface roughness
on emissivity remain illusive, and sometimes current emissivity
models are consistent with experimental emissivity values. A
universal emissivity model concerning roughness effect needs to be
proposed in order to completely agree with the practically measured
emissivity.
2.3. Surface texture
e λVðθi Þ = 1 − ρ λVðθi Þ
ð6Þ
Buckius et al. [35–39] theoretically confirmed that increased
roughness increases directional emissivity in the intermediate region.
In the geometric region, geometric optics is used to predict emissivity,
The spectral radiative properties were completely dependent on
the competition between the absorption coefficient (following the
wavelength) and the radiative scattering. The surface texture (in
materials science, texture is the distribution of crystallographic
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
5123
Fig. 4. Classification of emissivity models based on surface roughness [24].
orientations of a sample) often forming during coating deposition, has
an important effect on radiative scattering. For example, a surface
texture with high aspect ratio feature (height to width ratioN5: 1) can
serve as a collection of a lot of individual black-bodies, which exhibit
an emissivity of around 0.85 [43]. The theoretical and experimental
results of IR spectral emittance of nonsphere-shaped particles indicate
that coating composing of oriented rhenium whiskers with high
aspect ratio (Height/Dimension = 5) provides the high emittance
value, compared to that of randomly oriented whisker coating [44,45].
However, the texture effect on emissivity is still in argument. Recently,
Rozenbaum et al. [46] found that unlike the other previously studied
ceramics, the highly textured alumina ceramic has a much lower
emissivity than the single crystal between 1300 cm− 1 and 1800 cm− 1
(shown in Fig. 6) and the spectral emissivity of this sample was
radically different from the emissivity spectra of lowly textured
alumina. In this range, due to the weak absorption coefficient, the
radiation mean path inside the sample was higher than the thickness
of the single crystal. For longer wavelengths, the emissivity increases
to become higher than those of the single crystal and tends to those of
the lowly textured sample in the transparent zone. This is in good
agreement with the previous reports. So it is concluded that we
should consider wavelength range when we discuss the texture effect.
Fig. 5. Normal spectral emissivity at T = 1000 K and from 400 to 5000 cm− 1 of
cordierite–mullite substrate (dashed line), single crystal of Pr2NiO4 + d (dotted line),
and rough Pr2NiO4 + d coating on the cordierite–mullite substrate (solid line) [27].
Fig. 6. Normal spectral emissivity of alumina ceramics (thickness = 1 mm, T = 1, 350 K)
with two different textures [46].
5124
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
Fig. 7. (a) A SEM view of a 3D tungsten photonic crystal. Within each layer, the 1D rod
width is 0.5 μm and the rod-to-rod spacing is 1.5 μm. (b) A computed absorption spectra
for an eight-layer 3D tungsten photonic-crystal sample. [51].
2.4. Surface periodicity
A three-dimensional (3D) complete photonic band gap can be used
to suppress radiation below the electronic band gap [47–50]. Meanwhile, emission can be enhanced at a narrow band near a photonic
band edge or a narrow allowed band. If both effects are combined, a
nearly ideal radiation spectrum can be obtained. Photonic crystals
refer to the periodic optical (nano)structures that are designed to
affect the motion of photons in a similar way that periodicity of a
semiconductor crystal affects the motion of electrons. Photonic
crystals are composed of periodic dielectric or metallo-dielectric
(nano)structures that affect the propagation of electromagnetic waves
(EM) in the same way as the periodic potential in a semiconductor
crystal affects the electron motion by defining allowed and forbidden
electronic energy bands. Since this basic physical phenomenon is
based on diffraction, the periodicity of the photonic crystal structure
has to be of the same length-scale as half the wavelength of the EM
waves i.e. ∼ 200 nm (blue) to 350 nm (red) for photonic crystals
operating in the visible part of the spectrum — the repeating regions of
high and low dielectric constants have to be of this dimension. This
makes the fabrication of optical photonic crystals cumbersome and
complex.
Interestingly, Lin et al. realized a 3D tungsten photonic-crystal
sample (shown in Fig. 7(a)) using a modified silicon process [51],
which can be thermally excited and exhibits emission at a narrow
band [52,53]. In theory, the computed absorption spectrum for an
eight-layer photonic-crystal sample in Fig. 7(b) indicates that the
absorptance is low for λ N 3 μm (the photonic band gap) and increases
slightly at λ ∼ 3–4 μm due to a higher tungsten material absorption at
these wavelengths [54]. This band gap is a complete band gap, capable
of trapping light fully in all three dimensions and for both
polarizations [47]. Beyond the photonic band gap regime, there are
two strong absorptions: one is located at λ = 2.5 μm with an
absorptance of ∼40% and the other located at λ = 1.5–1.9 μm with
an even stronger absorptance of ∼80%. As discussed above, Kirchhoff's
law tells us that at equilibrium for a given wavelength λ and
temperature T, the emittance of any body is equal to its absorptance
[10]. Thus, a strong narrow absorption is suggestive of a narrow band
emission from a photonic-crystal sample.
In experiment, the measured power-density spectrum of the
sample in Fig. 8 is compared with that computed for a blackbody
cavity radiator at T = 750 K. The blackbody radiation follows the
Planck radiation. In the bandgap, the emission is suppressed and
below the blackbody emission. However, photonic-crystal emission in
the peak regime is found to exceed that of the blackbody by ∼300%.
Note the photonic band gap emission exhibits a narrow FWHM of
Δλ ∼ 1.8 μm while black body shows a wide FWHM of Δλ ∼ 4.5 μm. The
absorption rate of 3D metallic photonic crystal is found to be
suppressed in the photonic bandgap regime (λ ∼ 8–16 μm) and
strongly enhanced at the photonic band edge (λ ∼ 6 μm) The peak
absorption in tungsten photonic crystal is an effective absorption that
originates from the intrinsic tungsten absorption and is subsequently
enhanced by the photonic band edge effect. The enhancement is
attributed to a longer photon-matter interaction length as compared
to the metallic skin depth, the slower group velocity of light at the
band edge, and the finite metallic absorption constant. It should be
stressed that the photonic-crystal by no means exceeds the blackbody emission in the entire wavelength spectrum, rather the
exceeding happens at the neighborhood of the band-edge only [55].
Although the photonic crystal has been rapidly developed, the coating
or thin film of photonic crystal has not been realized yet.
2.5. Coating thickness
As mentioned above, when a radiation falls on a body, it may be
partially reflected, transmitted, or absorbed. As such, light penetration
depth (defined as the inverse of the absorption coefficient (α)) is
closely related to the emissivity of coatings. For the light with the
wavelength λ, the depth of light penetration, dp, is associated with the
extinction coefficient (k), given by
dp =
1
λ
=
α
4πk
ð7Þ
Eq. (7) indicates that dp is reversely proportional to k. Thus, k is
expected to be as low as possible in order to get high emissivity of
coatings. As a result, the coating is expected to be as thick as possible,
at least thicker than dp (otherwise, the transmittance will occur). But
in practice it is impossible to fabricate unlimitedly thick coating due to
production cost or fabrication limitation. In the past, models and
Fig. 8. Comparison of emission power density between the photonic-crystal sample and
a blackbody cavity radiator at T = 750 K [52].
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
experiments concerning the relationship between emissivity and
coating thickness were explored in details [56]. A typical relationship
between them in particle embedded in Al2O3 coating is shown in
Fig. 9, revealing that initially the emissivity increases rapidly as
thickness increases and becomes steady above thickness of 40 μm.
Similar relationship was also found in other ceramic coatings, such as
SiC, Si3N4 and A12O3 [5], anodic film on aluminum [57], Si3N4 and SiO2
on silicon substrates [58]. Therefore, it is concluded that the coating
has a critical thickness over and above which emissivity almost does
not change.
When the coating has more than two layers (called multilayer
coating), the optical interference will happen. If the thickness of
multilayered coatings falls into the same order of magnitude as the
incident radiation, a considerable amount of optical interference
within the multilayer structure tends to occur [59]. The net surface
reflectivity and absorption often change significantly even though
there is any thickness change in any layer. The resultant temperature
profiles and quality of the film are therefore affected by these changes
in optical properties. This was evident in SiO2/Si system, as shown in
Fig. 10 [60]. The relationship between emissivity and thickness shows
a converted “V” shape. The emissivity peaks at the SiO2 thickness of
around 240 nm for a given wavelength. When the coating is ultra thin,
approaching zero, smaller than the wavelength of the radiation light,
the increase of emissivity is ascribed to the coating thickness. The
reflections from the second interface are in phase with the one from
the first interface, resulting in a high reflection and low emissivity. As
the coating thickness increases, when the phase difference of reflected
lights from surface and interface is equal to the radiation light
wavelength, extinction happens where the emissivity reaches the
maximum value. As the thickness further increases, this phase
difference deviates from the wavelength and causes the emissivity
decreasing. However, this critical thickness (from several tens to
100 μm) where interference of radiation light occurs depends on the
coating composition [61].
2.6. Nanocomposite coatings
For normal incident radiation, reflection coefficient R is given by
the Fresnel's formula:
R=
ðn− 1Þ2 + k2
ðn + 1Þ2 + k2
ð8Þ
where n is the refractive index, k the extinction coefficient. According
to Eq. (4), the extinction coefficient k becomes much larger than 1
Fig. 9. Comparison of theoretical and experimental data of emissivity as function of
oxide coating thickness [55].
5125
Fig. 10. Emissivity (λ = 1.53 μm) versus oxide thickness at 925 °C [60].
when spectral emissivity ε is approaching the maximum value of 1.
Eq. (8) states the reflection coefficient tends to reach a maximum
value of 1. As a result, there is an abrupt decrease of ε while R has a
sudden increase as wavelength increases, which is clearly demonstrated in Fig. 11 for boron nitride [62]. Note that a peak of R in Fig. 11
(a) accompanies with a sudden drop of emissivity and Fig. 11(b). This
concluded that single component material can hardly show high
emissivity on the whole frequency range.
Besides pursuit of high emissivity, the other properties such as
mechanical properties, thermal shock resistance and oxidation
resistance, should be also taken into account when high emissivity
coatings are used at high temperature (N1000 °C) like MTPS. However,
single component coating could rarely possess these comprehensive
properties. Designing nanocomposite coatings may help a lot. A
nanocomposite coating comprises of at least two phases: a crystalline
phase and an amorphous phase, or two different crystalline phases, in
which at least one is capable of absorbing and re-radiating thermal
energy (emissivity agents), such as BN, SiC, SiB6, traditional metal
oxides, refractory metals carbides [63]. The other phase is like binder
such as colloidal silicon dioxide, which can provide high strength,
thermal expansion characteristics similar to their intended substrates,
and adequate bond strength with the substrate [64]. These composite
coatings represent a new class of materials, whose mechanical or
emittance properties are not subjected to volume mixture rules, but
depends on synergetic interactions of the composite constituents.
Recently, our group deposited SiC/SiO2 composite coating on the
pre-oxidized 316 SS substrate via EB-PVD technology [65,66]. It is
interesting to find that nanocrytalline SiC deposited at high temperature was formed, as shown in Fig. 12, giving rise to the nanohardness
as high as 50 GPa. At the same time, it is well known that SiC exhibits
very good emittance property. Combination of emittance and
mechanical properties makes SiC promising applied in high temperature environment or space, like MTPS [67]. In SiC/SiO2 system, SiO2
can act as an additive oxidation insulator because of low thermal
conductivity and high oxidation resistance. The spectral directional
emissivity of SiC/SiO2 composite thin film almost keeps a constant of
around 0.65 at a wide wave-number range from 400 to 4000 cm− 1, as
shown in Fig. 13. Unfortunately, the other properties like oxidation
resistance, shock resistance, have not been investigated. But it can be
predicted that the nanocomposite coatings are a promising branch of
materials that could potentially possess both high emissivity and
unique mechanical or thermal shock resistance properties.
Another composite coating structure, i.e. multilayer, is proposed in
order to meet comprehensive requirements, such as oxidation,
emittance, and low catalysis characteristics in hypersonic structures.
In the multilayered system, normally five different layers are realized
(schematically shown in Fig. 14) [68]: (i) low catalysis layer (based on
borosilicate or borophosphate glasses) to minimize aerothermodynamic heating; (ii) the high-emittance layer consisting of various
5126
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
Fig. 11. Reflectance and emittance of boron nitride [62].
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
5127
Fig. 14. Schematic diagram of the multilayer coating [68].
Fig. 12. HRTEM and cross-sectional images of SiC thin film prepared by EB-PVD at
900 °C.
has just begun, and it is expected to attract ever increasing attention in
the near future.
3. Emissivity characterization
constituents, such as SiB6, providing emittance enhancement to the
total system; (iii and iv) the oxygen-resistance layer to block oxygen
diffusion, comprising of a base layer and sealing layer; (v) reaction
barrier layer (like Al2O3, TiAl3) to prevent reaction between the
substrate alloy and the other coating layers. It is recognized that
thermal expansion coefficient differences between layers are inherent
in this multilayer design. Accordingly, we can accommodate thermal
strains by adjusting the layer thickness, and its self-healing characteristics can help avoid the cracking that may occur.
Bird et al. realized this multilayer design by sol–gel method [68].
The emittance layer of the coating imparted an emittance in excess of
0.8 during exposure at 980 °C for up to 100 h, which was at least 15%
greater than that for the coated alloy without the emittance layer as
shown in Fig. 15. Likewise, the low catalysis outer layer of the coating
reduced the catalytic efficiency for recombination of atomic oxygen
and nitrogen in hypersonic flowing air. The catalytic efficiency of
coated Inconel 617 was reduced by a factor of four, compared to the
uncoated condition, for a 1-h exposure at 980 °C.
To summarize, nanocomposite coatings with multiple components
or multilayered structure may be most attractive to reach high
emissivity value within the whole frequency range as well as excellent
mechanical or thermal properties. The research of multilayer design
Fig. 13. The emissivity and reflectivity of SiC/SiO2 composite thin film as a function of
wavenumber [9].
In a classical definition, emissivity measures the ability of a
material to radiate the heat through electromagnetic wavelength.
Emissivity measurements are difficult or tedious in practice because it
is a function of several variables and an extrinsic parameter of a
material. Although many methods have been proposed and developed
so far, a standard procedure or a universally accepted methodology
does not exist yet. Qualitatively, the emissivity can be estimated by
four developed basic methods: Calorimetric method [69–71]; optical
reflectivity method [72], multispectral radiation thermometry [73]
and radiation energy method [74]. Calorimetric method involves
measuring the heat power delivered to the test sample as well as the
temperatures of the emitting surface (the sample) and the surroundings in a vacuum over time [70,71]. An energy balance is then used to
Fig. 15. Emittance (a) and weight-change (b) as a function of exposure in air at 1000 °C
for coated Inconel 617 with emittance coating embedded within the oxidation coating
[68].
5128
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
determine the emissivity. The sample must be well insulated on the
side and back, and parasitic heat losses must be accounted for to
accurately determine the heat transfer through the sample surface
[75]. Reflectivity method involve illuminating a sample with infrared
energy and measuring the percentage of energy reflected from the
surface [4,76]. The absorptance is calculated from the reflectance and
used to calculate normal emittance by Kirchhoff's law and the Stefan–
Boltzmann equation. Multispectral radiation thermometry (MRT) [73]
employs radiance measurements at three or more wavelengths and an
emissivity model [77,78]. This method could measure the spectral
emissivity and temperature simultaneously based on the relationship
between temperature and spectral emissivity as follows:
pffiffiffi
eλ = exp a0 λ
ð9Þ
This method provides many advantages, such as sample-preparation fast test free, on site test, no limitation for temperature, but its
accuracy is lower and there is no common model to be applicable of all
the materials. The most widely method used in the assessment of
emissivity of coating is energy method which is based on the
measurement of radiation power of sample surface over the radiation
power of black body at the same temperature (Stefan–Boltzmann
equation). The key problem is to select the suitable black body which
determines the measurement accuracy. More details for emissivity
measurement via energy method could be found in references [74,79–
83]. Fourier spectroscopy based energy method for emissivity test has
been rapidly developed recently, which prides more accuracy (less
than 3%) of emissivity measurement. The temperature range of
measurement is from − 20 to 2000 °C and the wavelength ranges from
visible light to above 25 um.
4. Prospect
Absorption coefficient α and scattering coefficient S are important
factors to be considered when designing a high emissivity coating;
increasing α and decreasing S can enhance the emissivity. In principle,
all emissivity enhancement mechanisms can be applied to the
enhancement of ceramic coatings, the real challenge lies in the
implementation. However, for real engineering applications, pursuing
high emissivity is not enough, while a combination of high emissivity
and other comprehensive properties (called multifunctional coatings)
will be pursued more by the academics driven by the industrial needs.
In-situ synthesis of nanocrystalline or nanocomposite coatings (such as
nc-SiC/α-SiO2 or nc-SiC/α-C) or designing multilayered structural
layers may hold the key (at least, two of the keys) to this multifunctional
coating. Another promising branch called variable emittance coating has
just come up [84]. Small spacecraft, including micro and nanosatellites
will require an alternative means to achieve thermal control due to their
small power and mass budgets. Variable emittance coatings will be used
as spacecraft radiators for thermal balance.
In respect to the emissivity measurements, the lack of a universally
accepted emissivity measurement methodology and instrumentation
for any shape or different size samples has effectively hindered the
progress of the research. It is hoped that in a near future, a universally
accepted methodology will be born, and that this will accompany the
birth of an emissivity measurement apparatus for coatings, a standalone machine. To catalyze the birth of the universal emissivity
measurement, regular international seminars should be organized
especially on this topic. Fourier spectroscopy based energy method
may become the promising technique and needs further development.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
References
[1] B.V. Cockeram, D.P. Measures, A.J. Mueller, Thin Solid Films 355–356 (1999) 17.
[2] S. Kawakita, M. Imaizumi, T. Sumita, K. Kushiya, T. Ohshima, M. Yamaguchi, S.
Matsuda, S. Yoda, T. Kamiya, K. Kurokawa, L.L. Kazmerski, B. McNelis, M. Yamaguchi, C.
[58]
[59]
[60]
Wronski, W.C. Sinke, Proceedings of 3rd World Conference on Photovoltaic Energy
Conversion, Osaka, Japan, May 11–18, 2003, p. 8PB511.
T. Sumita, M. Imaizumi, S. Kawakita, S. Matsuda, S. Kuwajima, in: J.L. Wert (Ed.), IEEE
Radiation Effects Data Workshop Record, Monterey, U.S.A., July 21–25, 2003, P11, 2003.
T. Sumita, M. Imaizumi, S. Kawakita, S. Matsuda, S. Kuwajima, T. Ohshima, T. Kamiya, in:
J. Black (Ed.), IEEE Radiation Effects Data Workshop Record, Atlanta, U.S.A., July 19–23,
2004, P130, 2004.
M. FaIz, G. Leonhardt, Surf. Coat. Technol. 61 (1993) 97.
M.G. Hocking, V. Vasantasree, P.S. Sidky, Metallic Ceramic Coatings: Production,
High Temperature Properties Applications, Longman, London, 1989 P670.
C. Bartuli, T. Valente, M. Tului, Surf. Coat. Technol. 155 (2002) 260.
A. Ordine, C.A. Achete, O.R. Mattosa, I.C.P. Margarit, S.S. Camargo Jr., T. Hirsch, Surf.
Coat. Technol. 583 (2000) 133.
J. Yi, X.D. He, Y. Sun, Y. Li, Appl. Surf. Sci. 253 (2007) 4361.
M. Auslender, S. Hava, Infrared Phys. Technol. 36 (1995) 1077.
G.B. Rybicki, A.P. Lightman, Radiative Processes in Astrophysics, John Wiley & Sons,
New York, 1979 22 pp.
N.M. Ravindra, W. Chen, F.M. Tong, Arun Nanda, in: N.M. Ravindra, R.K. Singh
(Eds.), Transient Thermal Processing Techniques in Electronic Materials, TMS Soc.,
Warrendale, PA, Aug. 1996, pp. 159–164.
C.R. Bamford, Phys. Chem. Glasses 3 (1962) 189.
J.A. Duffy, J. Am. Ceram. Soc. 60 (1977) 440.
B.M. Wang, D.Z. Su, G.Y. Zhang, China J. Infrared Res. 2 (1983) 55 (in Chinese).
W.G. Spitzer, D. Kleinman, D. Walsh, Phys. Rev. V113 (1959) 127.
F. Seronde, P. Echegut, J.P. Coutures, F. Gervais, Mater. Sci. Eng. B8 (1991) 315.
R.H. Pratt, Radiat. Phys. Chem. 70 (4–5) (2004) 595.
M.G. Krishna, M. Rajendran, D.R. Pyke, A.K. Bhattacharya, Sol. Energy Mater. Sol.
Cells 59 (1999) 337.
A. Lowe, D.L. Chubb, S.C. Farmer, B.S. Good, Appl. Phys. Lett. 64 (1994) 3551.
L.G. Ferguson, Fatih Dogan, Mater. Sci. Eng. B83 (2001) 35.
L.G. Ferguson, F. Dogan, J. Mater. Sci. 36 (2001) 137.
K. Nassau, The Physics and Chemistry of Color — The Fifteen Causes of Color, Wiley,
New York, 1983 104 pp.
C.D. Wen, I. Mudawar, Int. J. Heat Mass Transfer 49 (2006) 4279.
F. Ghmari, T. Ghbara, J. Appl. Phys. 96 (2004) 2656–2664.
J.F. Ony Sacadura, Int. J. Heat Mass Transfer 15 (1971) 1451.
B. Rousseau, M. Chabin, P. Echegut, A. Sin, F. Weiss, P. Odier, Appl. Phys. Letts. 79
(2001) 3633.
R. Siegel, J.R. Howell, Thermal Radiat. Heat Trans. McGraw-Hill, New York, NY, 1981.
M.J. Haugh, Radiation thermometry in the aluminum industry, in: D.P. DeWitt, G.D.
Nutter (Eds.), Theory and Practice of Radiation Thermometry, John Wiley,
New York, NY, 1988.
J. De Santo, G. Brown, in: E. Wolf (Ed.), Progress in Optics, Cambridge University
Press, Cambridge, 1991.
R. Carminati, J.J. Greffet, A. Sentenac, in: J.S. Lee (Ed.), Proceedings of the 11th
International Heat Transfer Conference, vol. 7, Korean Society of Mechanical
Engineers, Kyongju, 1998, p. 427.
R. Petit, Electromagnetic Theory of Gratings, Springer, Berlin, 1980.
P. Lalanne, G.M. Morris, J. Opt. Soc. Am. A 13 (1996) 779.
M.Q. Brewster, Thermal Radiative Transfer and Properties, John Wiley, New York,
NY, 1992.
R.A. Dimenna, R.O. Buckius, J. Thermophys. Heat Transf. 3 (1994) 393.
R.A. Dimenna, R.O. Buckius, Trans. ASME 116 (1994) 639.
K. Tang, R.A. Dimenna, R.O. Buckius, Int. J. Heat Mass Transfer 40 (1997) 49.
K. Tang, R.A. Dimenna, R.O. Buckius, Int. J. Heat Mass Transfer 41 (1997) 2037.
K. Tang, R.A. Dimenna, R.O. Buckius, Int. J. Heat Mass Transfer 44 (2001) 4059.
G.J. Dail, M.G. Fuhrman, D.P. DeWitt, Proceeding of 4th Int. Aluminum Extrusion
Technology Seminar, Chicago, IL, vol. 2, April, 11–14, 1988, p. 281.
K. Kanayama, Heat Transfer-Jpn. Res. 1 (1972) 11.
A. Abdulkadir, R.C. Birkebak, Rev. Sci. Instrum. 45 (1974) 1356.
W. Konz, J. Hildenbrand, M. Bauersfeld, S. Hartwig, A. Lambrecht, V. Lehmann, J.
Wöllenstein, Proceedings Microsystems for the New millennium, SPIE, vol. 5836,
2005, doi:10.1117/12.608712.
J.R. Aronson, A.G. Emslie, F.E. Ruccia, C.R. Smallman, E.M. Smith, P.F. Strong, Appl.
Opt. 18 (1979) 2622.
W.L. Xu, S.C. Shen, Appl. Opt. 36 (1997) 1644.
D. Rozenbaum1, De Sousa Meneses P. Echegut, Int. J. of Thermophys. 30 (2009)
580.
J.G. Fleming, S.Y. Lin, I. El-Kady, R. Biswas, K.M. Ho, Nature (London) 417 (2002) 52.
S.Y. Lin, J.G. Fleming, Z.Y. Li, I. El-Kady, R. Biswas, K.M. Ho, J. Opt. Soc. Am. B 20 (2003)
1538.
E. Yablonovitch, Phys. Rev. Lett. 58 (1987) 2059.
S. John, Phys. Rev. Lett. 58 (1987) 2486.
S.Y. Lin, J. Moreno, J.G. Fleming, Appl. Phys. Lett. 83 (2003) 380.
S.-Y. Lin, J.G. Fleming, I. El-Kady, Optics Lett. 28 (2003) 1909.
T. Trupke, P. Würfel, M.A. Green, Appl. Phys. Lett., 84 (2004) 1997.
M.A. Ordal, L.L. Long, R.J. Bell, S.E. Bell, R.R. Bell, R.W. Alexander Jr., C.A. Ward, Appl.
Opt. 22 (1983) 1099.
S.Y. Lin, J. Moreno, J.G. Fleming, Appl. Phys. Lett. 84 (2004) 1999.
J.W. Zhou, X.Q. Wang, J.C. Han, Acta Mater. Compositae Sin. 24 (2007) 86.
C.S. Kumar, A.K. Sharma, K.N. Mahendra, S.M. Mayanna, Sol. Energy Mater. Solar
Cells 60 (2000) 51.
P.L.A. Chr. M. van der Meer, L.J. Giling, S.G. Kroon, J. Appl. Phys. 47 (1976) 652.
P.Y. Wong, I.N. Miaouiis, P. Zavracky, Sens. Actuator 19 (1990) 175.
N.M. Ravindra, S. Abedrabbo, W. Chen, Feiming M. Tong, Arun K. Nanda, Anthony C.
Speranza, IEEE Trans. Semicond. Manuf. 11 (1998) 1.
X. He et al. / Thin Solid Films 517 (2009) 5120–5129
[61] M.H. Han, X.G. Liang, Y. Huang, Chin. Sci. Bull. 50 (2005) 588.
[62] Y.S. Touloukian, D.P. DeWitt, Thermal Radiative Properties: Nonmetallic Solids, IFI/
Plenum, New York, 1972 P. 1042, 1047.
[63] Y. Ozaki, R.H. Zee, Mater. Sci. Eng. A, 202 (1995) 134.
[64] H.W. Babel, H.G. Le, Us patent No. 5,296,288 (1994).
[65] X.D. He, J. Yi, Y. Sun, P. Xiao, X.F. Zhao, Appl. Phys. Lett. 91 (2007) 241914 1.
[66] J. Yi, X.D. He, Y. Sun, Y. Li, M.W. Li, Appl. Surf. Sci. 253 (2007) 7100.
[67] A.K. Costa, S.S. Camargo Jr., C.A. Achete, R. Carius, Thin Solid Films 377–378 (2000)
243.
[68] R. Keith Bird, Terryl A. Wallace, J. Spacecr. Rockers 41 (2004) 213.
[69] J. Hameury, B. Hay, J.R. Filtz, Int. J. Thermophys. 28 (2007) 1607.
[70] R.D. Karam, in: P. Zarchan (Ed.), “Thermal Control Hardware,” Satellite Thermal
Control for Systems Engineers, AIAA, Reston, VA, 1998, pp. 160–164.
[71] A. Mychkovsky, and R. Ponnappan, “Calorimetric Measurement of Emissivity in
Space Conditions,” AIAA 2005-961, Jan. 2005.
[72] Dunkle R.V. In progress in International Research on Thermodynamics and
Transport Properties Asme[M], 1962, p. 100.
[73] C.D. Wen, I. Mudawar, J. Commun. Heat Mass Transfer 33 (2006) 1063.
5129
[74] J. Ishii, A. Ono, Meas. Sci. Technol. 12 (2001) 2103.
[75] R.D. Karam, Thermal control hardware, in: P. Zarchan (Ed.), Satellite Thermal
Control for Systems Engineers, AIAA, Reston, VA, 1998, pp. 160–164.
[76] D.A. Jaworske, T.J. Skowronski, in: M. El-Genk (Ed.), Proceedings of the Space
Technology and Applications International Forum (STAIF-2000), AIP Conference
Proceedings, No. 504, American Institute of Physics Press, New York, 2000, p. 791.
[77] P.B. Coates, Metrologia 17 (1981) 103.
[78] C. Wen, I. Mudawar, J. Heat Mass Transfer 48 (2005) 1316.
[79] B.M. Liu, Z.X. Chu, 10th European Conference on Thermophysical Properties, 1986,
p. 22.
[80] S. Clausen, A. Morgenstjerne, O. Rathmann, Appl. Optics 35 (28) (1996) 5683.
[81] A. R Korb, P. Dybwad, W. Wadsworth, J.W. Salisbury, Appl. Optics 35 (10) (1996) 1679.
[82] E. Lindermeir, V. Tank, P. Haschberger, Proceedings of SPIE-The Int. Society for
Optical Eng., vol. 1682, 1992, pp. 354–364.
[83] J.M. Dai, X.D. Liu, X.G. Sun, 7th Int. Workshop on Subsecond Thermophysics, 2004.
[84] D. Farrar, W. Schneider, R. Osiander, J.L. Champion, A.G. Damn, Inter Society
Conference on Thermal Phenomena, 2002, p. 1020.