Heat Transfer Design Considerations for Refractory Linings,

Part 1

G. Palmer

This paper discusses the prob-

lems associated when undertak-
Thermal conductivity [W/m ⋅ K]

ing heat transfer analysis, specif-

ically, the effect of taking simpli-
fied assumptions and the prob-
lem of large variations in refrac-
tory thermal conductivity values.
The effect of refractory material
conductivity values and air gaps
at the interface in a 1D analysis
is provided. It is shown that us- Density [kg/m2]
ing a 1D heat transfer model to
Fig. 1 Thermal conductivity of refractory versus density comparison from published data
predict temperature profiles in [5, 9]
refractory systems can result in
serious errors. and operational trouble-shooting. It is clear is generally applied to one or both sides by
that the design of refractory linings is be- using some simplified correlations to ac-
coming increasingly important both struc- count for wind and other flow effects.
1 Introduction turally and in energy efficiency terms. Also In many cases, simplifying assumptions are
The heat transfer analysis of multi-layer re- industry “fitness for service” (FFS) guidelines made when undertaking a heat transfer
fractory concretes in both transient and for pressure vessels require that the plant is analysis. Incorrect assumptions can lead to
steady state conditions is particularly impor- safe for personnel and the public. FFS as- poor or the wrong selection of materials
tant for many pyro-processing industries. sessments will include refractory linings and which in turn can lead to excess heat loss,
Knowing the temperature profile through the practice of simply changing the refracto- refractory failure, overheating of vessels,
the different layers is important for design ry design without proper engineering ap- too low shell temperatures or poor lining
proval could leave companies exposed. design with increased capital costs.
Greg Palmer Heat transfer theory is generally well under- Earlier research [1, 2] has found that there
Palmer Technologies Pty. Ltd. stood and it is possible to predict tempera- is a significant difference between data-
Coorparoo DC, Queensland, Australia ture profiles under various conditions with sheet thermal conductivity and published
reasonable accuracy, where accurate ther- data. The effect of higher material thermal
Tony Howes mal property and convective / radiative conductivity and gaps between concrete
School of Engineering boundary conditions are known. Refractory layers are discussed.
The University of Queensland linings are generally composed of multiple This paper discusses the problems associ-
4072 St. Lucia, Australia layers of varying insulating materials and a ated when undertaking heat transfer
dense abrasion resistant hot face layer. For analysis, specifically, the effect of taking
Corresponding author: Greg Palmer steady conditions, when the layer is flat or simplified assumptions and the problem of
E-mail: [email protected] thin (thickness < 5 % of the radius of curva- large variations in refractory thermal con-
ture), conduction heat flow, Q, through each ductivity values. The effect of refractory
Keywords: Heat transfer, refractory linings, layer is well defined by material conductivity values and air gaps
design considerations Q = k /Δx A ΔT at the interface in a 1D analysis is provid-
where k is the average thermal conductivity ed. It is shown that using a 1D heat trans-
Received: 09.11.2009 of the layer material, Δx its thickness, A the fer model to predict temperature profiles
Accepted: 11.11.2009 area for heat transfer and ΔT the tempera- in refractory systems can result in serious
ture difference across the layer. Convection errors.

94 refractories WORLDFORUM 2 (2010) [2]

[W/m ⋅ K]

[W/m ⋅ K]
Temperature [°C] Temperature [°C]
[W/m ⋅ K]

[W/m ⋅ K]

Temperature [°C] Temperature [°C]

Fig. 2 Comparison of data sheet thermal conductivity and predicted thermal conductivity of dense and insulation castables

2 Thermal conductivity The most common method to determine re- conductivity for materials of similar density
measurement and estimation fractory thermal conductivity has been the and composition often had thermal conduc-
Heat transfer analysis is one of the most calorimetry method which as been used for tivities that differed from the published data
commonly used tools in evaluating and de- more than 60 years. It would be safe to say often by as much as 50 %.
signing refractory structures. The analysis is that the hot-wire method is now more fre- A more recent analysis undertaken by
almost always carried out using a 1D steady quently used due to speed and cost. Akiyoshi et al [5] investigated the relation-
state program assuming perfect conductions Thermal conductivity of refractory materials ship of thermal conductivity with volumetric
through composite layers. This requires the has been studied both theoretically [3, 4] bulk density3) and temperature for alumina
use of refractory thermal conductivity and and practically [1, 5, 6, 7]. This research has and fireclay refractories. The investigation
the most commonly used source are manu- found that the hot-wire method is well suit- used the hot-wire technique and the results
facturer’s data sheets. Hence the accuracy of ed for heterogeneous refractory materials were statistically analyzed using the least
reported thermal conductivity data and the but not well suited for non-isotropic materi- squares method. A correlation was develop-
validity of perfect conduction through com- als like insulation fibreboard [8]. In 1988 ed for fireclay and alumina refractories with
posite layers are particularly important for Crowley and Young [1] studied a number of volumetric bulk density in the range of
designers. different thermal conductivity test tech-
There are a number of methods for deter- niques and concluded that there were signif-
mining thermal conductivity of refractory icant differences between static and dynam-
materials, for example, the hot wire method, ic methods. It was also concluded that the
calorimetry and laser flash thermal diffusivi- determination of the thermal conductivity (k)
ty. The aim is not to discuss these techniques using the comparative test1), which relates k
but rather to discuss variations on tempera- to geometric bulk density2), was the most
ture predictions between reported and pre- consistent and agreed closely with the hot-
dicted thermal conductivity. wire data. They also found that the thermal

1) Using known standard materials and comparing temperature differentials across the standards
2) Determined by sample measurement and weight
3) Immersion technique Fig. 3 Thermovision of furnace roof showing shell
4) P-Thermal is a1D transient heat transfer program developed by the authors temperatures

refractories WORLDFORUM 2 (2010) [2] 95

 show the difference between thermal con-
ductivity values under perfect conductions
for a one refractory layer system.
The shell temperature was compared to a 1D
P-Thermal4) [12] heat transfer model using
manufacturer’s data sheet thermal conduc-
tivity and the refractory thermal conductivity
Temperature [°C]

predicted by equation 1 for a single insula-

tion layer.
Fig. 3 shows an infrared image of a furnace
roof with 150 mm thick insulation refractory
under a 6 mm carbon steel shell. The tempe-
ratures were taken shortly after start up and
the refractory lining had been completely re-
placed. The shell temperature varies from
Time [h]
149 °C (133 °C near the edge) to 160 °C.
Fig. 4 P-thermal analysis of furnace roof using thermal conductivity calculated by equa- The 1D heat transfer model using the pre-
tion 1 dicted thermal conductivity data (equation
1) is shown in Fig. 4. The 1D model for the
550 kg/m3 and 3140 kg/m3 with porosity in tivity using equation 1, for four different re- one layer system is 156 °C which validates
the range of 15 to 81 %. fractory castables. It can be seen there is a the refractory thermal conductivity values
The relation for k in a range of refracto- significant difference between predicted and predicted by equation 1 (correlated to the
ries is shown in equation 1.This equation datasheet thermal conductivity values, in hot-wire method).
was reported to be valid for materials for some cases the error is greater than 50 %. The 1D heat transfer model using the manu-
25 °C ≤ T ≤ 1200 °C, 550 kg/m3 ≤ ρ v ≤ This is in line with data published by Crow- facturer’s data sheet thermal conductivity
3140 kg/m3 and 36 mass-% ≤ Al2O3 ≤ 94 ley and Young [1]. Published literature has predicts a steady state shell temperature of
mass-%. found that the hot-wire method is an accu- 121 °C. This is significantly less than what
The work published by Crowley and Young rate and reliable [10, 11] method for deter- has been measured from actual field data. It
[1] presents data suitable for comparison mining thermal conductivity of refractory is concluded that a 1D heat transfer model
purposes. The field data published by material. Thus it is concluded that correla- using the manufacturer’s data sheet thermal
Crowley and Young [9] was also compared tions based on the hot-wire test can be used conductivity is likely to under-predict tempe-
against the Akiyoshi et al. [5] correlation to accurately predict refractory thermal con- rature profiles in refractories systems by 20
and is plotted in Fig. 1. This shows an ex- ductivity. It is also concluded that manufac- to 30 %. If the refractory thermal conductiv-
cellent correlation between refractory ther- turer’s datasheet thermal conductivity values ity values measured by the hot-wire tech-
mal conductivity measured by Crowley and in some cases can be very inaccurate for un- nique or calculated by equation 1 is used in
Young and Akiyoshi et al. known reasons. a 1D perfect model then the shell tempera-
Fig. 2 shows the difference between the ture can be predicted to within 5 % for a
manufacturer’s data sheet thermal conduc- 3 Analysis for a one layer system one layer system. It is concluded that corre-
tivity values and predicted thermal conduc- An example for a furnace roof is used to lations based on the hot-wire test can be

(1)

Fig. 5 Thermovision of reactor shell showing a hot spot and surrounding shell temperature (~180 °C)

96 refractories WORLDFORUM 2 (2010) [2]

 used to accurately predict refractory thermal
conductivity in a one layer refractory system
using a 1D heat transfer model.

4 Analysis for a two layer system

with and without air gaps
Due to the fact that small air gaps exist be-
Temperature [°C]

tween refractory composite layers a 1D mod-

el was developed which included an air gap
at the interface between the two concrete
layers in order to study the effect of data-
sheet thermal conductivity on lining temper-
ature profiles.
Previous research has shown that two re-
fractory conditions are existing, the first is
Time [h]
the green state when the concrete is first
cast [13]. In this condition the bonding be-
Fig. 6 1D heat transfer temperature profile using k from equation 1 with no air gap
tween the two concrete layers is most likely
to be well bonded. The second and most im-
portant condition for designers is the fired
state when the concrete has dried and gaps
can exist at the interface.
The heat transfer for a two layer system
(shell, insulation and hot-face) was evaluat-
ed by analyzing a reactor vessel. The original
refractory lining is composed of 150 mm
Temperature [°C]

Kastolite 2300LI and 100 mm 1800 grade

hot-face. The reactor operates at a tempera-
ture of 1250 °C.
Fig. 5 shows the shell temperatures for a re-
actor which had a hotspot in one area. The
shell temperature away from the hotspot
was approximately 180 °C. After the hotspot
was repaired the shell temperature was mo-
Time [h] nitored using welded thermocouple wire.
The temperature varied from 160 °C to
Fig. 7 1D heat transfer temperature profile using k from the manufacturer’s data sheet 180 °C. The temperature profile was calcu-
lated using a 1D heat transfer model (with-
out an air gap) using both Akiyoshi (eqn 1)
and manufacturer’s data sheet k values. The
temperature profile using k calculated from
equation 1 is shown in Fig. 6. This predicts a
shell temperature of 209 °C.
Repeating the analysis using the manufac-
turer’s data sheet k values has a tempera-
Temperature [°C]

ture profile as shown in Fig.7. This predicts a

shell temperature of 195 °C which is higher
than the measured shell temperature by ap-
proximately 15 °C.
The analysis is repeated again using the
Akiyoshi k values (equation 1) with an air
gap. The temperature profile with air gap of
~10 W/m2 K at the interface is shown in
Time [h]
Fig. 8. In this case the predicted shell tem-
perature is 181 °C which is within the meas-
Fig. 8 1D heat transfer temperature profile using k from equation 1 with an air gap
ured shell temperature range.

refractories WORLDFORUM 2 (2010) [2] 97

It is concluded that the effect of a gap in References [8] Wulf, R.; Barth, G.; Gross, U.: Intercomparison of
the order 1 mm at the interface between [1] Crowley, M.S.; Young, J. S.: Thermal conducti- insulation thermal conductivities measured by
the concrete layers is important and vity of monolithic refractories. Ceram. Bull. 67 various methods. Int. J. Thermophys. 28 (2007)
should be taken into consideration when (1988) [7] 1679–1692
designing refractory linings. [2] Hemrick, J.G.; Charles, W.K. jr.; Andrew, A.W.; [9] Crowley, M.S.; Young, J.S.: Evaluation of thermal
Mattison, K.F.: Thermal conductivity of alumi- film coefficients in furnace walls. Proc. Int. Forum
5 Conclusions na measured with three techniques. J. Testing on Advances in Refractories Technology, Cincin-
It is concluded that the current design and Evaluation 31 (2003) [4] nati, 1988. Am. Ceram. Trans. 42-4 (1988) 547–
procedure of using a 1D heat transfer [3] Allen, A.W.; Heat transfer mechanisms in re- 556
model to predict temperature profiles in fractory materials. Refractory Applications [10] Nunes dos Santos, W.: Advances on the hot wire
refractory systems can result in serious er- and News 10 (2005) [1] technique. J. Eur. Ceram. Soc. 28 (2008) 15–20
rors and the use of simplified forced con- [4] de Sylos Cintra jr., J.; Nunes dos Santos, W.: [11] Aksel'rod, E.I.; Vishnevskii, I.I.: Use of the hot
vection coefficient equations can lead to Numerical analysis of sample dimensions in wire method for measuring the thermal con-
errors of 25 % or more. The analysis and hot wire thermal conductivity measurements. ductivity of lightweight, fiber, and powder re-
historical data shows that manufacturer’s J. Eur. Ceram. Soc. 20 (2000) 1871–1875 fractory materials. Thermal Engng. (1984) [4]
published thermal conductivity values can [5] Akiyoshi, M.M.; da Silva, A.P.; da Silva, M.G; 49–53
be very inaccurate for unknown reasons. Pandolfelli, V.C.: Impact of thermal conducti- [12] Palmer, G.B.; Howse, T.: Calculation of transient
The error in the manufacturer’s thermal vity on refractories. Am. Ceram. Soc. Bull. 81 and steady state heat transfer in refractory li-
conductivity can be by as much as 50 %. (2002) [3] nings under various conditions. Refractory Ap-
Research by others has shown that corre- [6] Akiyoshi, M.M.; Pereira, R.; da Silva, A.P.; Pan- plication and News 12 (2007) [1]
lations based on the hot-wire test can ac- dolfelli, V.C.: Effect of alumina content, poro- [13] Palmer, G.B.; Ellis, R.E.: Behaviour of monolithic
curately predict refractory thermal con- sity and temperature on the thermal conduc- refractory and its importance in design of re-
ductivity. It has been found that using tivity of refractories. Am. Ceram. Soc. Bull. 82 fractory structures. Refractory Application and
hot-wire thermal conductivity data with (2003) [6] News 12 (2007) [6]
an interface air gap is an accurate method [7] Hemrick, J.G.; Loveland, E.: Technique develop-
for predicting temperature profiles in re- ment for large sample thermal conductivity me-
fractory systems. asurement. Proc. UNITECR 2005, pp. 846

98 refractories WORLDFORUM 2 (2010) [2]