RELATIONSHIP BETWEEN MECHANICAL CHARACTERISTICS AND.

RELATIONSHIP B E T W E E N M E C H A N I C A L
MK0400056
C H A R A C T E R I S T I C S AND T H E R M A L S H O C K STABILITY
OF REFRACTORIES
T. Volkov-Husovic, R.Jancic, K.Raic
Faculty of Technology and Metallurgy, Dept. Metall. Eng. Karnegijeva 4, 11000 Belgrade,
Serbia and Montenegro,

Abstract: Thermal stability of the refractory material with the content of 60 % A12O3 was investigated. Water
quench test (JUS.B.D8.319.) was applied as experimental method for thermal stability testing. Damage of porous
materials is commonly related to a modification of mechanical strength that is mostly a reduction. This is linked
with* characteristics related to pore space. Mechanical characteristics are considered such as compressive
strength, dynamic modulus of elasticity and resistance parameters resulting from resonance frequency
measurements, as well as ultrasonic velocity.
Key words: alumina based refractories, dynamic Young modulus, fracture resistance parameters, thermal
stability

1. INTRODUCTION
Thermal shock stability dictates refractory performance in many applications. In many instances, a
two -fold approach, i.e. 1) material properties [1-8] or/and 2) heat transfer conditions [8-13] is taken to
characterize thermal shock behavior of the refractories. As an alternative, information on the thermal
shock behavior of refractories can be obtained by experimental means. One test for this purpose,
which is highly popular because of its simplicity, consists of quenching appropriate specimens from an
oven temperature into a medium such as water, liquid metal, oil, or fused salts maintained at a lower
temperature.
Damage of porous materials is commonly related to a modification of mechanical strength that is
mostly a reduction. This is above all linked with characteristics related to pore space. One has to
realize what type of test procedures and also criteria are all available and come into consideration for
any corresponding correlation. On one hand, structure-related characteristics such as pore volume and
pore size distribution can be cited here as well as some discrete values deduced therefrom, but also
water absorpcion, saturation coefficient and diffusivity. Whereas on the other hand, mechanical data
are considered such as compressive strength, dynamic modulus of elasticity resulting from resonance
frequency measurmets, as well as ultrasonic velocity.
Thermal quenching of the refractories leads to the crack nucleation and/or propagation resulting in loss
of strength. Since the formation of the cracks has profound influence on the ultrasonic velocity and the
Young modulus of the material, measuring either of these properties can directly monitor the
development of the thermal shock damage level.

2. EXPERIMENTAL PROCEDURE
The ultrasonic technique can be used to characterize selected properties of refractories because sonic
pulses travel trought a medium at a rate that is directly affected by the material properties, and internal
structure. Comparing similar materials sonic pulses will travel faster trought the densier samples,
although the interpretation of the results can also be affected by the presence of internal flaws, voids or
cracks. The measurment of ultrasonic velocity was done using the equipment OYO model 5210
according to the standard testing ( JUS.D.B8.121.). The transducers were rigidly placed on the two
parallel faces of the cylindrical specimen using vaseline grease as the coupling medium. The ultrasonic
320

Proceedings of 3rd BMC-2003-Ohnd, R. Macedonia

RELATIONSHIP BETWEEN MECHANICAL CHARACTERISTICS AND. .

velosity was then calculated from the spasing of the transducers and the waveform time delay on the
osciloscope.
By determining the bulk density, and ultrasonic velosity it is possible to calculate the dynamic Young
modulus using the equation:

(1)
where:Vp - ultrasonic velocity of longitudinal waves (m/s),Vs - ultrasonic velosity of transverzal
waves (m/s)^ - Poasson ratio and p - density (kN/m3). The value was given at 0.2 and assumed not
to change with the development of thermal shock damage. According to the literarture [14,17] and
previous work [18-26] that will not be grate restriction.

2.1.

MATERIAL

The refractory samples used in the present study were obtained directly from commercial producer.
This type of refractory was used in several previous studies in which a number of physical properties
related to their thermal shock behavior were measured [17-26].
Table I. Thermal and Mechanical Properties of the Samples Studied
Parameter

value

A12O3, (%)

60

Density, p (g/cm3)

3.05

Thermal conductivity, k (W/mK)

1.62

Specific heat, c (kJ/kgK)

1.06

Thermal expansion coefficient, a (%)

0.8

Compressive strength, a p (MPa)

54

Young modulus, E ( GPa)

16.70

Thermal stability (number of quench experiments, N)

18

3. RESULTS AND DISCUSSION

The effect of the thermal quenching on the ultrasonic velocities is given on the figures 1. and 2. From
the figures, as well as from the regression analysis very good correlation was obtained between
degradation of the ultrasonic velocity caused by the thermal shock. These results suggests that
changes in ultrasonic velocities colud be used as nondestructive monitoring the thermal shock damage
in material caused by thermal shock.

Proceedings of 3rd BMC-2003-Ohnd, R. Macedonia

321

RELATIONSHIP BETWEEN MECHANICAL CHARACTERISTICS AND..

1400

3500

3000^

1000

2500l
Q . 2000
>"

(A
>

1500

800
600
400

1000

200

500-

10

IS

20

10

15

20

Figure 1. UIstrasonic velocity (Vp) of longitudinal

vawes versus number of cycles of teh quench
experiment (N)

Figure 2. UIstrasonic velocity (Vs) of transverzal

vawes versus number of cycles of quench experiment
(N)

The effect of the thermal shock fatique on the Young modulus is shown on the figure 3. The good
agreement between decrising the Young modulus and number of cycles suggests that thermal shock
damage could also be quantitatively evaluated nondestructively by measuring the dynamic Young
modulus. Unfortunatly, coefficient of correlation is in this case lower than it was in comparison the
ultrasonic velocity with the number of cycles (Table II).
The calculated values of dynamic Young modulus must be considered as an aproximation, but there
are certain ocasions where experimental E measuments are not feasible, and a calculated value at least
yelds a usable number .The results from literature [16] indicates thet the percent difference between
experimental and calculated values can range between 5% to 10%. With the simplicity of the
method it should be always be possible to generate a usable modulus of elasticity value for a refractory
product, although it must be remembered that there may be a significant error in the values obtained.

20
16
14 i

6
64
2
0

10

15

20

Figure 3. Dynamic Young modulus as function of

number of cycles of the quench experiment (N)

Figure 4. Compressive strength (o, MPa) degradation

as function of number of cycles of the quench
experiment (N)

As the very good results were obtained monitoring changes in ultrasonic velocites, that was used for
the correlation of strength degradation caused by thermal shock. Expression for the strength
degradation was used [14], described as:

(2.)
322

Proceedings of 3rd BMC-2003-Ohrid, R. Macedonia

RELATIONSHIP BETWEEN MECHANICAL CHARACTERISTICS AND. .

where o~0 - is compressive strength of the material before exposure material to the theraml shock
testing and n - material constant ( n = 0.488, ref. 14.). High values of coefficient of correlation
suggests that degradation in strength during thermal stability testing is in strong correlation with the
water quench test experiment results.
Thermal shock stability of the refractories could be presented using resistance parameters [1-4]
Fracture and damage resistance parameters were calculated from thermal and mechanic properties
listed in Table I., and usual equations were used for fracture resistance parameters (R = a ( 1 - v) / E
a and R = k R) as well as for the damage resistance parameters (R'" = E/a^(l-v) and R"" = y R'").
Degradation of resistance parameters due to the degradation of strength, were also presented in figure
5.

25
20
15
CE

10
5
0
)

10

15

20

Fig.5. Resistance parameters behavior as function of number of cycles of the quench experiment (N), A - first
fracture resistance parameter, B - second fracture resistance parameter, C - first damage resistance parameter, D
- second damage resistance parameter.

From the figure 5. excellent correlation between fracture resistance parameters degradation with the
number of cycles of the quench experiment was oserved (Table II). Higher values of fracture
resistance parameters are expected for the higer value of number of cycles of the water quench test.
Damage resistance parameters versus nuber of cycles show the opposite corretaltion. Higher values of
damage resistance parameters are expected for the lower values of number of quench experiment.

Proceedings of 3 rd BMC-2003-Ohrid, R. Macedonia

323

RELATIONSHIP BETWEEN MECHANICAL CHARACTERISTICS AND.

Table II: Regression statistics for the observed correlations
Regression
Statistics
Multiple R
R Square
Adjusted R
Square
Standard Error

Vp-N

Vs-N

Edyn-N

a-N

R-N

R'-N

R'"-N

R""-N

0.922
0.851
0.801

0.987
0.970
0.966

0.872
0.760
0.680

0.946
0.895
0.860

0.948
0.898
0.865

0.948
0.898
0.865

0.986
0.973
0.964

0.987
0.973
0.963

333.9

42.16

3.670

3.18

1.88

3.06

0.959

0.014

5. CONCLUSION
.Monitoring of thermal shock damage development in refractories by nondestructive method is of
much industrial significance in relation to productivity as well as safety. Sonic measurment is very
usefull tool to solve this problem since the advantages are:
Type of instrument is small, portable and operates on either line or self-contained battery
power.
Results of the method could be obtain very fast, and as they are related to damage, the damage
level in material could be predicted with very good accourance.
There is a great diversity of more potencial applications, limited only by imagination of the
user.
The non-destructive evaluation of refractory products by sonic measurments has proven to be a very
useful characterization tool. Thereby it is posible to generate qualitative and/or quantitative results,
several examples are presented in this paper. Physival properties that are commonly correlated with
ultrasonic measurments are apparent porosity, bulk density, permeability, modulus of rupture and cold
crusging strength. Calculable material properties based on ultrasonic velocity data, include modulus of
elasticity , such as resistance parameters were also cosidered for analysis.
Results presented in this study suggests that the resuts of the sonic measurments could be used for
prediction of damage level in specimen due to the thermal shock. Monitoring the damage level could
be obtained using the decrising of the ultrasonic velosity. Also, as the results in the Chapter 4. suggests
the decrining of the Young modulus could be also used for the same purpose. The lower values of the
coefficient of correlation with the water quench results in that case suggests more acurate analysis,
possibly combined with other correlation to get more accurate prediction. Strength degradation in
material due to thermal schoking also could be evaluated, with very strong correlation with the number
of cycles of the quench experiment (N). Prediction of strength degradation could be very useful
parameter to know for life time prediction of material.
Fracture resistance parameters degradation caused by quench experiments was also observed with
excellent correlation with the number of cycles of water quench test. Damage resistance parameters
showed the opposite bahavior, and growth of the parameters caused by quench experiments was
observed.

Literature
[1] D.P.H. Hasselman, " Unified Theory of Thermal Shock Fracture Initation and Crack Propagation
in Brittle Ceramics ", J.Am.Ceram.Soc. ,Vol. 52 , No.l, 600 -604 (1969)
[2] D.P.H. Hasselman, " Elastic Energy at Fracture and Surface Energy as Design Criteria for Thermal
Shock ", J. Am.Ceram. Soc. Vol 46, No 11, 535-540 ( 1963)
[3] J.Nakayama J.Am.Ceram.Soc, " Direct Measurments of Fracture Energies of Brittle Heterogenius
Materials ", J.Am.Ceram Soc, Vol 48. No 11., 583 - 587 ( 1965)
[4] J.Nakayama, M.Ishizuka, " Experimental Evidence for Thermal Shock Damage Resistance ",
Amer.Cer.Soc.Bull, Vol. 45 , No.7 666-669 (1965)
324

Proceedings of 3rd BMC-2003-Ohrid, R Macedonia

RELATIONSHIP BETWEEN MECHANICAL CHARACTERISTICS AND .

[5] W.D.Kingery, " Factors Affecting Thermal Stress Resistance of Ceramic Materials ",
J.Am.Ceram.Soc, Vol 38, NNo 1, (1955) 3-15
[6] S S.Manson, R.W.Smith, " Theory of Thermal Shock Resistance of Brittle Materials Based on
Weibull >s Statistical Theory of Strength ", J.Am..Ceram.Soc.l Vol 38, No 1, (1955) 18-27
[7] H.Hencke, J.R.Thomas, JR and D.P.H.Hasselman, " Role of the Material Properties in the
Thermal-Stress Fracture of Brittle Ceramics Subjected to the Conductive Heat Transfer",
J.Am.Ceram.Soc. Vol 67, No 6, (1984) 393-398
[8] R.W.Davidge, G.Tappin, " Thermal Shock and Fracture in Ceramics" , Trans. Br.Ceram.Soc. Vol
66, No 8, (1967)405-22
[9] S.K.Nyiogi, A.C.Das, " Prediction of the Thermal Shock Behaviour of Castable Refractories by
Sonic Measurments", Interceram, Vol 43, No 6, (1994), 453-457
[10] J.H.Ainworth and R.H.Herron, " Thermal Shock Damage Resistance of Refractories", Ceramic
Bulletin, Vol.53, No.7, (1974) 533-538
[11] C.E.Semler, " Nondestructive Ultrasonic Evaluation of Refractories", Interceram, Nr 5, (1981)
485-488
[12] T.Volkov-Husovic, R.Jancic, M.Cvetkovic, D.Mitrakovic, Z.Popovic, " Thermal Shock Behavior
of Alumina Based Refractoories : Fracture Resistance Parameters and Water Quench Test ",
Matterials Letters, 38 (1999) 372-378
[13] T.D.Volkov-Husovic, Z V.Popovic, " Resistance parameters and water quench test as criteria of
thermal shock behaviour of alumina reafractories", Material Science and Technology, Vol 15, No 10
(1999) 1216-1219
[14] T.Volkov-Husovic, RMJancic," Thermal Shock Behavior of Alumina Based Refractories:
Comparison between Damage Resistance Parameters and Water Quench test", Industrial Ceramics,
Vol 20, No 3, (2000) s 94-97
[15] T.D.Volkov-Husovic, R.M.Jancic, V.Radojevic. Z. Popovic, " Prediction of the thermal shock
behavior of alumina based refractonesC temperature difference, damage resistance parameters and
water quench test " EUROMAT 2001, Rimini, Italy, 10-14 June 2001, Abstracts and Papers CD p 150
[16] T.D.Volkov-Husovic, R.M.Jancic, " Thermal Shock Behavior of Alumina Based Refractories",
Industrial Ceramics, Vol. 21, No 1. (2002) 159-164
[17] T.D.Volkov-Husovic, "Material Propertiies and Thermal Shock Behavior of Alumina Based
Refractories", Silicates Industriales, May-June (2001) 66 (5-6) 55-61

Proceedings of 3rd BMC-2003-Ohrid, R Macedonia

325