Chemical Engineering Research and Design 1 4 3 ( 2 0 1 9 ) 1–3

Contents lists available at ScienceDirect

Chemical Engineering Research and Design

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

Heat transfer phenomena of glassy carbon foams

Anna Gancarczyk ∗ , Wojciech Macek, Andrzej Kołodziej

Institute of Chemical Engineering, Polish Academy of Sciences, Bałtycka 5, 44-100, Gliwice, Poland

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

Article history: The aim of the study was to determine and describe the relationship between the morpho-
Received 24 August 2018 logical parameters of glassy carbon foams and their transport characteristics. The study
Received in revised form 27 was carried out on reticulated vitreous carbon (RVC) foams of different pore density. Based
December 2018 solely on foam morphology, using the theoretical models for laminar flow, the heat transfer
Accepted 4 January 2019 properties were determined with satisfactory accuracy.
Available online 14 January 2019 © 2019 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Keywords:
Heat transfer
Glassy carbon foams
RVC foams

1. Introduction
Nomenclature
Among miscellaneous carbon materials, reticulated vitreous carbon
(RVC) foams appear to be highly interesting (Inagaki and Kang, 2014). cp Gas specific heat, J/(kgK)
The most important method of their manufacture is template carbon- D Characteristic crosswise diameter, m
isation. The first reported method of carbon foam production was the dh Hydraulic diameter, =4␧/Sv , m
n  
carbonisation thermosetting of organic polymer foams through simple   Nui,exp −Nui,calc 
heat treatment (Ford 1964 (filed, 1960)). This method, improved over ey Average relative error,= 1
n  Nui,exp 
the years and still used today, allows control of the pore structure, thus i=1
porosity and density, by choice of the polymer foam template of the L Channel length, m
desired structure. The template, e.g. polyurethane foam, can be used L* Dimensionless length for the thermal entrance
after impregnation with a resin (e.g. phenol, furfuryl or epoxy resin) region, = L/(DRePr)
(Inagaki et al., 2015). Nu Nusselt number, =␣D/␭
RVC foams have many useful features, such as high thermal and Pr Prandtl number, =␩cp /␭
electric conductivity, low density and thermal expansion, very good
Re Reynolds number, = wD␳/(␩␧)
resistance to many corrosive media, high operating temperature up to
w Superficial velocity, m/s
3500 ◦ C in an inert environment (www.ergaerospace.com), extremely
␣ Heat transfer coefficient, W/(m2 K)
high porosity and large specific surface area. Therefore, RVC foams
have been studied intensively for many years as three-dimensional
␧ Porosity
electrodes in electrochemical processes and batteries (Friedrich et al., ␭ Thermal conductivity of the fluid, W/(mK)
2004; Inagaki et al., 2015; Walsh et al., 2016), as adsorbers in applica- ␩ Dynamic viscosity, Pas
tions such as radioactive 137 Cs, and as supports for photocatalytic TiO2 ␳ Density, kg/m3
particles (Inagaki et al., 2015). They can be used as energy absorbers ␴st 
Standard deviation,=
 
or high temperature filters for molten metals before casting. These
 1 n  Nui,exp −Nui,calc  2
foams are also considered for different medical applications, such as  n−1  Nu  − ey · 100%
i,exp
constructs for tissue engineering and for ex vivo cell expansion (Pec
i=1
et al., 2010). Moreover, the surface of carbon foams can be modified
with a variety of coatings. For example, metal coatings such as Cu can

significantly improve the mechanical and thermal properties of carbon

foams (Almajali et al., 2010). Most of the processes require deep knowl-
∗ edge of morphological features describing the foams’ skeletons, as well
Corresponding author.
E-mail address: [email protected] (A. Gancarczyk). as porosity and specific surface area. Knowledge of the transport and
https://doi.org/10.1016/j.cherd.2019.01.008
0263-8762/© 2019 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
2 Chemical Engineering Research and Design 1 4 3 ( 2 0 1 9 ) 1–3

Fig. 1 – (a) SEM micrographs of RVC 100 sample; (b) experimental set-up for heat transfer studies: 1 – blower; 2 – flowmeter,
3 – thermocouples, 4 – electric power generation system, 5 – RVC sample 6 – reactor.

friction characteristics is also crucial for correct process design and

Table 1 – Morphological parameters of the foams
implementation.
studied.
Therefore, the aim of the study was to determine and describe
the heat transport properties of RVC foams. It had already found that Parameter RVC 30 RVC 80 RVC 100
the laminar flow developing in the short capillary channel was a good
cell diameter, dc [mm] 3.08 0.58 0.56
approximation of the flow and transport phenomena occurring in solid window diameter, dw [mm] 1.16 0.23 0.21
foams (Gancarczyk et al., 2017). Therefore, we have attempted here to strut diameter, ds [mm] 0.307 0.054 0.047
describe the heat transport intensity based on the experimental results porosity, ␧ [%] 96 96 96
and theoretical solutions developed for laminar flow (Churchill and specific surface area, Sv [m2 /m3 ] 1223 3557 4154
Ozoe, 1973).

2. Materials and methods

2.1. Characterisation of materials

The foams studied were reticulated vitreous carbon (RVC)

foams with pore densities of 30, 80 and 100 PPI (pores per
inch), respectively, according to the producer’s specification
(ERG Materials and Aerospace Corp.). An example of RVC
100 foam is shown in Fig. 1a. The morphological param-
eters were determined using the X-ray micro-tomography
(SkyScan 1172) equipped with reconstruction software (NRe-
con, SkyScan software). The low atomic mass of carbon caused Fig. 2 – Nu number vs Re number for tested RVC foams.
some difficulties in obtaining high resolution images, so a
40 keV X-ray beam was used without any filters (details can et al., 2016)). Therefore, micro-CT images were binarised using
be found in (Leszczyński et al., 2016)). iMorph software (Brun et al., 2008) and supported by the ero-
Additionally, the optical microscopy method was applied to sion and dilatation procedures that allow determination of
identify cells, windows and struts, and to measure their diam- hydrodynamic porosity (void fraction available for fluid flow)
eters. The measurements were taken for at least one hundred and specific surface area available, e.g. in regard to active cat-
cells, windows and struts (see Fig. 1a) for each foam tested. alytic layer deposition. The results, as well as dimensions of
cells, windows, and struts obtained using optical microscopy,
2.2. Transport coefficients are presented in Table 1.

The heat transfer experiments were conducted using a reactor 3.2. Heat transfer
of rectangular cross section (Fig. 1b). The foam samples were
The heat transfer results are presented in Fig. 2 as the Nusselt
matched to the inner reactor cross-section (30 x 45 mm) and
number vs. Reynolds number. Because the foams’ structures
the sample thickness was either 5 mm or 25 mm. The carbon
differ significantly, the dimensionless numbers are conse-
foams are good electrical conductors, so were heated using an
quently defined with the hydraulic diameter, dh .
electric current (up to 5.5 A) flowing directly through the foam
For this relationship, when both dimensionless numbers
skeleton. The temperatures of the foam surface and flowing
are defined with the same specific diameter (D = dh ), the results
gas (air, about 22 ◦ C) were measured using 12 small thermo-
obtained for RVC 30, 80 and 100 foams are very close. Based on
couples pasted to the foam using special heat-conducting glue
the experimental results, a correlation equation for RVC 30, 80
(for details see (Iwaniszyn et al., 2017)). The air stream flowing
and 100 foams was developed, obtaining:
from the blower via gas rotameters to the reactor, was changed
in the range w = 0.3−15.3 m/s.
−0.672
Nudh = 3.25 + 0.12 · L* (1)
3. Results
where L∗ =L/(DRePr) and assuming model channel diameter
3.1. Characterisation of materials equal to the hydraulic one, D = dh and length equal to the strut
diameter, L = ds (cf. Fig. 1).
Micro-CT revealed some information that can be assessed as The correlation equation (1) describes the experimental Nu
irrelevant from our point of view (e.g. the inner porosity of number with average error ey = 11.7% and standard deviation
the foam skeleton) or even false (e.g. artefacts (Leszczyński ␴st = 9% (Fig. 3).
 Chemical Engineering Research and Design 1 4 3 ( 2 0 1 9 ) 1–3 3

the laminar flow developing in short capillary channels is a

good approximation of the flow and heat transport occur-
ring in the RVC foams. However, theoretical models require
some correction, based solely on the foams’ morphological
parameters.

Acknowledgements

This work was supported by the Polish National Sci-

ence Centre (Projects No. DEC-2011/03/B/ST8/05455, DEC-
2016/21/B/ST8/00496 and Dec-2016/23/B/ST8/02024).
Fig. 3 – Nu number vs dimensionless channel length L*
(calculated for D = dh and L = ds ) for tested RVC foams.
References

Almajali, M., Lafdi, K., Prodhomme, P.H., Ochoa, O., 2010.

Mechanical properties of copper-coated carbon foams. Carbon
48, 1604–1608.
Brun, E., Vicente, J., Topin, F., Occelli, R., 2008. Germany,
DresdeniMorph : A 3D Morphological Tool to Fully Analyze all
kind of Cellular Materials, CELLMET 2008: Proceedings of the
International Symposium on Cellular Metals for Structural
and Functional Applications,2008. iMorph : A 3D
Morphological Tool to Fully Analyze all kind of Cellular
Materials, CELLMET 2008: Proceedings of the International
Symposium on Cellular Metals for Structural and Functional
Applications.
Churchill, S.W., Ozoe, H., 1973. Correlations for laminar forced
convection in flow over an isothermal flat plate and in
developing and fully developed flow in an isothermal tube. J.
Fig. 4 – Comparison of experimental Nusselt numbers and Heat Transfer 95, 416–419.
calculated based on the Churchill and Ozoe (1973) model Ford W. 1964 (filed 1960). Method of making cellular refractory
thermal insulating material. Union Carbide Corp.
(eq. (2)) and its modified form (eq. (3)).
Friedrich, J.M., Ponce-de-León, C., Reade, G.W., Walsh, F.C., 2004.
Reticulated vitreous carbon as an electrode material. J.
As was found for solid foams (Gancarczyk et al., 2018), the
Electroanal. Chem. 561, 203–217.
transport of RVC foams may be modelled using the developing Gancarczyk, A., Piatek, M., Iwaniszyn, M., Jodlowski, P.J.,
laminar flow approach, i.e. laminar flow in a short capillary Lojewska, J., Kowalska, J., Kolodziej, A., 2017. In search of
channel. Therefore, heat transfer may also be described by the governing gas flow mechanism through metal solid foams.
theoretical solution presented by Churchill and Ozoe (1973) for Catalysts 7, 124.
short circular capillary channels: Gancarczyk, A., Iwaniszyn, M., Piatek,˛ M., Sindera, K., Korpyś, M.,
Jodłowski, P.J., Łojewska, J., Kołodziej, A., 2018. Interfacial heat
 −1/2 and momentum transfer relation for porous media. Int. J.
2 (4/) L* Therm. Sci. 132, 42–51.
Nu =  1/4 (2)
Inagaki, M., Kang, F., 2014. Materials Science and Engineering of
1 + (Pr/0.0207)2/3
Carbon: Fundamentals, second ed. Elsevier.
Inagaki, M., Qiu, J.S., Guo, Q.G., 2015. Carbon foam: preparation
However, this equation overestimates the experimental and application. Carbon 87, 128–152.
results obtained for RVC 30, 80 and 100 foams, especially for ˛
Iwaniszyn, M., Piatek, M., Gancarczyk, A., Jodłowski, P.J.,
high L∗ values (i.e. low Reynolds numbers) as presented in Fig. Łojewska, J., Kołodziej, A., 2017. Flow resistance and heat
4. Therefore, the theoretical solution of Churchill and Ozoe transfer in short channels of metallic monoliths: experiments
(1973) (eq. (2)) was corrected based on the experimental results versus CFD. Int. J. Heat Mass Transfer 109, 778–785.
Kolodziej, A., Lojewska, J., Jaroszynski, M., Gancarczyk, A.,
obtained for RVC 30, 80 and 100 foams:
Jodlowski, P., 2012. Heat transfer and flow resistance for
 −1/2 stacked wire gauzes: experiments and modelling. Int. J. Heat
2 (4/) L*
−0.201 Fluid Flow 33, 101–108.
Nu =   · 0.101 Pr · L* (3) Leszczyński, B., Gancarczyk, A., Wróbel, A., Piatek,˛ M., Łojewska,
2/3 1/4
1 + (Pr/0.0207)
˛
J., Kołodziej, A., Pedrys, R., 2016. Global and local thresholding
methods applied to x-ray microtomographic analysis of
for D = dh and L = ds . Eq. (3) describes the experimentally metallic foams. J. Nondestruc. Eval., 35.
obtained Nusselt number with an accuracy of ey = 15.3% (Fig. 4). Pec, M.K., Reyes, R., Sánchez, E., Carballar, D., Delgado, A.,
A similar correction factor was also proposed for wire gauzes Santamaría, J., Arruebo, M., Evora, C., 2010. Reticulated
(Kolodziej et al., 2012). vitreous carbon: a useful material for cell adhesion and tissue
invasion. Eur. Cell. Mater. 20, 282–294.
Walsh, F.C., Arenas, L.F., Ponce de León, C., Reade, G.W., Whyte, I.,
4. Conclusions Mellor, B.G., 2016. The continued development of reticulated
vitreous carbon as a versatile electrode material: structure,
Based on the database created, the relationship between properties and applications. Electrochim. Acta 215, 566–591
the foams’ morphology and their heat transfer characteris- www.ergaerospace.com.
tics was examined on a theoretical basis. It was proved that