Acta Polytechnica CTU Proceedings


doi:10.14311/APP.2016.3.0056
Acta Polytechnica CTU Proceedings 3:56–59, 2016 © Czech Technical University in Prague, 2016

available online at http://ojs.cvut.cz/ojs/index.php/app

DETERMINATION OF THERMAL RESPONSE OF CARRARA
AND SNEZNIKOVSKY MARBLE USED AS BUILDING MATERIAL

Veronika Petráňováa, ∗, Jaroslav Valacha, Alberto Vianib,
Marta Peréz Estébanezb

a Institute of Theoretical and Applied Mechanics AS CR, v.v.i., Prosecka 76, Prague, Czech Republic
b Institute of Theoretical and Applied Mechanics AS CR, v.v.i., Centre of Excellence TelÄŊ, Batelovska 485,

TelÄŊ, Czech Republic
∗ corresponding author: petranova@itam.cas.cz

Abstract. Physical weathering of marble, widely used as a cladding material on buildings, is one of
the most common damaging mechanism caused by anisotropic thermal expansion of calcite grains. The
extent of marble deterioration depends mainly on stone fabric and texture. Dry cuboids of Carrara
marble and marble from Dolni Morava quarry were subjected to microscopic analysis and thermal
cycling, to determine the thermal expansion related to stone fabric and predominant lattice orientation
of grains (i.e. texture).

Keywords: marble, anisotropic thermal expansion, thermal dilatation.

1. Introduction
Marble has been predetermined as building and dec-
orative natural stone since antiquity because of its
good workability and unique visual appearance. These
properties are related to the composition of carbonate
sediments and to the nature and intensity of the recrys-
tallization process under the regional metamorphosis
conditions. Marble is considerably vulnerable to both
chemical and physical weathering by reason that this
type of stone predominantly consists of carbonate min-
erals calcite - CaCO3 and dolomite - CaMg(CO3)2.
Periodic temperature variations, together with chang-
ing in moisture can cause serious damage to the stone
structure, finally leading to the complete stone decay.
The enormous sensitivity of marble to variations in
temperature originates in the thermal response of cal-
cite crystals. While the dolomite crystals subjected to
thermal cycling expand only, the calcite crystals show
anisotropic thermal dilatation introduced by thermal
cycles, i.e., it expands along c - axis during heating
and contracts in other directions. However, cooling of
calcite crystals leads to the opposite effect, i.e., con-
traction along the c - axis and expansion in the other
directions as is illustrated in Figure 1. Both mecha-
nisms are considered as the major cause of creation
and propagation of cracks along the grain boundaries
in the stone. Thermal expansion can be expressed
using the thermal expansion coefficient α, describing
the length change per unit of temperature. The coef-
ficient value depends on the considered temperature
interval. To quantify the extent of stone elongation,
the residual strain ε is more suitable [1].
The preferred lattice orientation of calcite grains

i.e. texture is one of the most important stone char-
acteristics, essentially influencing thermal response
in different spatial directions. Marble with strong

texture is generally more vulnerable to thermal weath-
ering, which can be enhanced by significant differences
between the α values along the c- and a-axis of individ-
ual calcite grains [2]. Thermal expansion of marble is
measurable at low temperatures enabling simulation
the climate conditions in the central Europe. Never-
theless, experimentally obtained thermal dilatation
values often deviated from calculation based on theo-
retical models considering only the texture variations
in a marble. Another important parameters are grain
shapes, grain sizes and the type of grain boundaries
because of the strong influence on the rate of stone
decay. A certain directional relationship between the
shape preferred orientation (SPO) of calcite grains
and their lattice preferred orientation (LPO) can ex-
tensively intensify the thermal response of marble. For
example, marble with LPO parallel to SPO exhibits
a strong thermomechanical response because of the
large concentration of maximum principal stress in the
direction of shape elongation. Thermal stresses dur-
ing heating can initiate the formation of new cracks
or widening of preexisting cracks or pores. Micro-
cracking formation can also be influenced by grain
size since the level of marble decay is also connected
with the number of adjacent grains and with the grain
boundaries [3–5]. However, thermal response and con-
sequently the rate of physical deterioration of marble
depends also on other stone characteristics, including
the ratio of calcite and dolomite, moisture content,
porosity and pore size distribution of the investigated
marble [6]. Depending on all the mentioned stone
characteristics, the thermal response of marble after
cooling down to room temperature can be determined
as isotropic or anisotropic, with or without residual
strain [1].

56

http://dx.doi.org/10.14311/APP.2016.3.0056
http://ojs.cvut.cz/ojs/index.php/app


vol. 3/2016 Thermal Response of Carrara and Sneznikovsky Marbles

Figure 1. Anisotropic thermal expansion of calcite crystals. Adopted from [1].

2. Materials and Methods
Testing of the thermal response was performed on two
structurally different marble types. The first one from
the Carrara area in northern Italy was chosen for its
simple fabric. The second one - Sneznikovsky marble
was taken from Dolni Morava quarry in the Czech
Republic. The texture as well as the shape and size
of grains was investigated by optical microscopy and
electron back-scattered diffraction (EBSD) method in
a scanning electron microscope (SEM, FEI Quanta
450, FEI corp., USA). For the SEM-EBSD investi-
gations, slide specimens of marble were coated with
a thin layer of carbon to protect against specimen
charging. To evaluate the dependence of dilatation
on the sample’s microstructure, two pairs of perpen-
dicular cuboids A-a, B-a, C-1 and C-4 (6 x 6 x 20
mm) were prepared from the original marble blocks.
The orientation of the cuboid samples is depicted in
Figure 2. Thermal expansion measurements were per-
formed using a vertical dilatometer (L75 PT Linseis
corp., Germany). The specimen elongation was mea-
sured after each 1 °C step. The cuboid samples were
subjected to heating and cooling cycles, ranging in
temperature from -40 to 100 °C. The cycles were re-
peated 7 times for specimens A-a, B-a and C-1 and
5 times for specimen C-4. The temperature interval
was gradually increased for the A-a, B-a and C-1 spec-
imens. The interval for specimen C-4 remained stable
in all 5 cycles (from -40 to 100 °C). The values for the
thermal expansion coefficient α values were calculated
using equation (1). The length change in 4L was
determined within a temperature interval 4T, from
18 °C to the maximum temperature in the considered
cycle, i.e. for cycle 1, the maximum temperature was
33 °C, whereas for cycle 7, it was 100 °C. To determine
α, only a positive temperature range was selected,
because the results from the range below the freezing
point were not reliable.

α =
4L
L4T

(1)

Figure 2. Orientation of cuboid specimens, A-a and
B-a is Sneznikovsky marble and C-1 and C-4 is Carrara
marble.

The symbols and their physical meanings are listed
at the end of the paper.

3. Experimental Results
Based on optical and SEM-EBSD observations, the
Carrara marble exhibits a weak texture in both direc-
tions (x-y and x-z). Specimens C-1 and C-4 contain
calcite grains with a nearly uniform size of around
0.3 mm with straight polygonal grain boundaries and
no shape preferred orientation. Contrary to Carrara,
Sneznikovsky marble exhibits cracking, predominantly
on planes parallel to the x-y plane. The microfabric of

57



V. Petráňová, J. Valach, A. Viani, M. Peréz Estébanez Acta Polytechnica CTU Proceedings

Figure 3. a - microfabric of investigated samples, b - orientation of calcite grains in B-a sample.

Figure 4. Dependence of relative extension on the temperature.

the investigated marble types is illustrated in Figure
3. The size of the calcite grains in the A-a and B-a
specimens is not uniform; grains in sizes up to 4 mm
with sutured grain boundaries can be found. As in
the Carrara specimen, both perpendicular planes of
Sneznikovsky marble contain grains with no shape
preferred orientation.
The elongation of the cuboid specimens was ex-

pressed as relative extension, i.e. percentage difference
between 4T for each temperature step and original
length of the specimen. The highest relative extension
of 0.18% was observed in C-4 at 100 °C in the first
cycle. In the other cycles (2-5) the value was equal to
0.16%. Specimen C-1 showed an increasing trend over
the temperature and reached its maximum relative ex-
tension of 0.14% at 100 °C. Regarding the coefficient
α, specimen C-4 also showed the highest average value
of 12.5 · 10−6K−1. Specimen C-1 achieved the most
significant difference in α, varying from 3.9 ·10−6K−1
to 11.3 · 10−6K−1. The results are illustrated in the
Figures 4 and 5. The first cycle of the measurements
performed on the specimen C-1 was excluded from the
results. Values obtained from the first cycle performed
in C-1 specimen weren’t included to the results.

During the 7 cycles in specimens A-a and B-a rela-
tive elongation also showed an increasing trend, but
the maximum values were more different than in the
Carrara marble. At 100 °C in cycle 7, the relative

Figure 5. Thermal expansion coefficient α of investi-
gated specimens, loaded in the 18-100 °C temperature
interval. The value marked "A" represents an average
value of α determined by [7] on large set of marble
specimens.

extension in specimen A-a was equal to 0.15% while in
B-a it was just 0.10%. The α coefficients determined
for the Sneznikovsky marble were not as variable as
in the Carrara type. In the A-a sample, α reached an
average value of 10 · 10−6K−1 and in the B-a sample
of 5 ·10−6K−1. Based on the relative extension differ-
ence between corresponding cycles, an α variation of
5 ·10−6K−1 was determined, which can be attributed
to the observed anisotropy in the Sneznikovsky mar-

58



vol. 3/2016 Thermal Response of Carrara and Sneznikovsky Marbles

ble, i.e. the x-y orientated plane in specimen A-a was
more fissured.

4. Conclusions
It can be concluded that simultaneous microscopic
and dilatometry investigations can help in elucidation
of thermally induced degradation of marbles. Based
on the results, the thermal response and generation
of damage caused by accumulation of residual strains
related to stone fabric and texture in investigated
marbles will be determined on a larger set of perpen-
dicular specimens, heated and cooled in the positive
range of temperatures (i.e. from 20 to 100 °C). Investi-
gations carried out on both specimen types reveal an
increasing rate of thermal expansion with increasing
temperatures. In presence of a temperature gradient,
this characteristic can contribute to the well known
bowing effect on marble slabs.

List of symbols
α Thermal expansion coefficient [ K−1]
4L Difference between the actual specimen length and

the length at the room temperature [m]
L Length of the specimen at the room temperature [m]
4T Temperature difference between the actual and the

room temperature [K]

Acknowledgements
The research has been supported by Czech Science Foun-
dation (Project no. P105/12/G059).

References
[1] M. Steiger, A. Charola. Weathering and deterioration.

In The Stone in Architecture, chap. 4., Springer-Verlag,
2011. doi:10.1007/978-3-642-14475-2_4

[2] S. Siegesmund, K. Ullemeyer, T. Weiss, E.K. Tschegg.
Physical weathering of marbles caused by anisotropic
thermal expansion. Inter J Earth Sci 89(1):170-182,
2000.

[3] V. Shushakova, E.R. Fuller Jr, S. Siegesmund.
Influence of shape fabric and crystal texture on marble
degradation phenomena:simulations. Env Earth Sci
63(7-8):1587-1601, 2011. doi:10.1007/s12665-010-0744-7

[4] B. Leiss, T. Weiss. Fabric anisotropy and its influence
on physical weathering of different tyes of Carrara
marbles. J Struct Geol 22(11-12):1737-1745, 2000.

[5] U. Akesson, J.E. Lindqvist, B. Schouenborg, B. Grelk.
Relationship between microstructure and bowing
properties of calcite marble claddings. Bull Eng Geol
Env 65(1):73-79, 2006.

[6] E. Cantisani, E. Pecchioni, F. Fratini, C.A. Garzonio, P.
Malesani, G. Molli. Thermal stress in the Apuan marbles:
Relationship between microstructure and petrophysical
properties characteristics. Inter J Rock Mech Min Sci
46(1):128-137, 2009. doi:10.1016/j.ijrmms.2008.06.005

[7] A. Zeisig, S. Siegesmund, T. Weiss. Thermal
expansion and its control on the durability of marbles.
In The Natural Stone, Weathering Phenomena,
Conservation Strategies and Case Studies, chap. 2., The
Geological Society of London, 2002.

59

http://dx.doi.org/10.1007/978-3-642-14475-2_4
http://dx.doi.org/10.1007/s12665-010-0744-7
http://dx.doi.org/10.1016/j.ijrmms.2008.06.005

	Acta Polytechnica CTU Proceedings 3:56–59, 2016
	1 Introduction
	2 Materials and Methods
	3 Experimental Results
	4 Conclusions
	List of symbols
	Acknowledgements
	References