J Therm Anal Calorim

DOI 10.1007/s10973-013-3417-4

Research on historical bricks from a Baroque Church

Rudolf Podoba • Tiit Kaljuvee • Igor Štubňa •

Ľuboš Podobnı́k • Peter Bačı́k

Received: 23 July 2013 / Accepted: 12 September 2013

Ó Akadémiai Kiadó, Budapest, Hungary 2013

Abstract Baroque bricks were investigated by DTA, TG, Introduction

EGA, TDA, and XRD. The analyses showed that the brick
consisted of dehydroxylated illite, quartz, and calcite. De- Materials research has been applied successfully to study
hydroxylation as a consequence of the former rehydroxy- the historical building ceramics, such as bricks and tiles,
lation was not found probably because of protection of the for the last 50 years. Chemical and mineralogical charac-
bricks by plaster. Between the temperatures 600 and terization of ceramics helps to establish the origin, period,
800 °C, (a) intensive mass loss in TG, (b) endothermic and technique used to make the ceramic pieces. To learn
minimum in DTA, (c) intensive escape of CO2 in EGA, about our history is not only to analyze and preserve the
and (d) contraction of the sample in TDA were observed. ceramic objects but also to investigate and understand the
All these events belong to decomposition of calcite. As knowledge and skills used to produce them [1]. That is
follows from these results, the maximum firing temperature important in renovating of the historical buildings where
was about 700 °C. The bricks have relatively high porosity the damaged material should be replaced with the new one,
*43 % and specific surface area *18.6 m2 g-1. properties of which are close to the properties of the his-
torical material.
Keywords Historical bricks  Thermal analysis  Thermal analyses are widely used for investigation of
Firing the historical ceramics. Most of them exploit TG [2–6] and
XRD [4, 6–10]. Infrared spectroscopy [10, 11], Mössbauer
spectroscopy [12], TDA [9, 13], SEM [9], DTA [4], and
other methods are also used to study the historical ceramics
from different points of view. As was found in [2], the
R. Podoba
main and most often object of thermoanalytical investiga-
Department of Physics, Faculty of Civil Engineering,
Slovak University of Technology, Radlinského 11, tions of historical ceramics is an estimation of the firing
813 68 Bratislava, Slovakia temperature.
The articles cited above deal with ceramic pieces
R. Podoba  I. Štubňa (&)  Ľ. Podobnı́k
obtained from relatively thin-wall ceramic vessels or with
Department of Physic, Faculty of Natural Science, Constantine
the Philosopher University, A. Hlinku 1, 94 901 Nitra, Slovakia small ceramic objects. A degree of conversion which
e-mail: [email protected] characterizes thermal transformation in clay during firing
(e.g., kaolinite ? metakaolinite, decomposition of calcite)
T. Kaljuvee
in such objects depends on the temperature and time,
Laboratory of Inorganic Materials, Faculty of Chemical and
Materials Science, Tallinn Technical University, Ehitajate 5, a ¼ f ðT; tÞ. The estimation of the maximum firing tem-
190 86 Tallinn, Estonia perature is usually based on comparison of the thermoan-
alytical curves, e.g., TG, DTA, and TDA, received from the
P. Bačı́k
historical ceramic samples, with the curves received from
Department of Mineralogy and Petrology, Faculty of Natural
Sciences Comenius University, Mlynská dolina, the laboratory-prepared samples [2]. The thin-wall ceramic
842 15 Bratislava, Slovakia vessels or small objects require short time to reach

123
 R. Podoba et al.

temperature homogeneity in their whole volume (if we 0.2

suppose homogeneous temperature field in the space of the

dV/d[log(diameter)]/cm3/g
0.16
fired body); therefore, the location of the sample for ther-
mal analysis does not play any role. A different situation is
0.12
in a large ceramic body such as a brick. Layers which are
on the surface are transformed sooner than the layers inside 0.08
the brick. Experiments on large cylindrical samples showed
clear dependence of the degree of conversion on the 0.04
location of the sample taken for TG, a ¼ f ðT; t; rÞ, where
r is a distance of the sample from the axis of rotation [14]. 0
0.001 0.01 0.1 1 10 100 1000
A detailed research was done on bricks taken from eight Pore diameter/micrometer
medieval churches in [15], where a composition of the fired
and green bricks as well as the firing conditions were Fig. 1 Distribution of pores
studied. The firing temperature was estimated as
700–950 °C for the bricks from different churches. (ka1 = 0.15406 nm), accelerating voltage 40 kV and
In this paper, the study is focused on the analysis of the beam current 40 mA. Data were obtained by the Bruker
historical bricks. Our goal was to approximately find their LynxEye detector.
original composition and firing temperature. Before industrial revolution, bricks were fired in field
kilns using wood as fuel. The kilns were rebuilt with each
firing. Field kiln was probably rectangular stacks of brick
Experimental with an interior chamber for the fire and openings at the top
for venting the hot gas. It is natural that physical properties
The samples were bricks of the approximate size of the bricks varied depending on their distance from the
6.5 9 15 9 29.5 cm taken from a Baroque part of the fire. To obtain reliable results, we should analyze a lot of
church located in Pác, Trnava County, Slovakia. The bricks taken from different parts of the building. We had
church, originally built in Romanesque style, was enlarged only three bricks at disposal which were drawn from the
in the Baroque style in 1764, i.e., the age of the samples is masonry and were stored in the open air for 2 years. These
*250 years. The bricks were a part of the masonry that bricks had almost identical properties.
was covered with plaster.
The brick samples were light red of color. This color
gives hematite (Fe2O3) which is one of the most intense Results and discussion
coloring materials in pottery. The presence of only 1 % of
hematite was enough to give a reddish color under the Some physical properties of the brick sample were mea-
oxidizing firing conditions [16, 17]. sured at room temperature, namely thermophysical quan-
Porous area was measured with Sorptometer Kelvin and tities and porosity.
the distribution of the pores was determined using mercury The results of porometry (sorptometer Kelvin 1042) of
porosimeter Quantachrome PoreMaster on four samples samples indicated that the micropore volume was
from each brick. The dimensions of the samples were 1.86 mm3 g-1, microporous specific area was 5.27 m2 g-1,
*1 9 1 9 1 cm. and non-microporous area was 13.26 m2 g-1. A distribution
DTA and TG were performed on the analyzer Derivato- of the pores is visible in Fig. 1. The volume density of the
graph 1000 MOM Budapest [18] using powder sample of brick determined from the mass and dimensions of the
1.2 g and heating rate 5 °C min-1. TDA was done on the sample is 1.49 g cm-3 and the total porosity estimated from
horizontal alumina dilatometer with the samples densities of the mineral constituents, sample volume and its
8 9 8 9 40 mm cut from the brick and heating rate mass, was *43 %.
5 °C min-1. The results of thermal analyses, the DTA and TG curves
The evolved gas analysis (EGA) was done on a powder of the old brick material are presented in Fig. 2. The first
sample (50 mg) using Setaram LabSys 1600 thermoana- endothermic minimum on the DTA curve, which belongs
lyzer coupled with Pfeiffer Omnistar Mass Spectrometer to the interval of 20–300 °C, is typical for porous building
at a heating rate of 5 °C min-1 up to 1050 °C, in the clay ceramics, which does not contain glassy phase, i.e.,
atmosphere 79 % of Ar ? 21 % of O2 with flow rate of fired at relatively low temperatures, that does not allow
60 mL min-1. complete sintering. The endothermic minimum is related to
The powder XRD analysis was performed on the dif- liberation of the physically bound water absorbed on the
fractometer Bruker D8 Advance with Cu anticathode surfaces of the crystals of different minerals. It is known

123
Research on historical bricks from a Baroque Church

1 Table 1 The results of EGA

1
Gas Dt/°C Tpeak1/°C Tpeak2/°C Tpeak3/°C
Temperature difference/°C

0 0
0 200 400 600 800 1000

Relative mass loss/%

–1 CO2 160–763 225 325 731a
–1
–2 CO 651–751 732 – –
–2 H2O 53–150 95 – –
–3
a
–4
The highest maximum
–3
Exo
–5
–4
Endo
–6 Q
–5 –7
Temperature/°C As received

Fig. 2 DTA (black line) and TG (gray line) of the Baroque brick
Q
Q
Q Q Q Q
I I F F CF F Q
that a porosity of the green porcelain body increases 4–5 %
0 10 20 30 40 50 60
during dehydroxylation [19] and the specific surface area
of kaolin body increases *25 % during dehydroxylation 2θ /grad
[20]. A similar behavior of illite, which also contains Fig. 3 XRD pattern of the Baroque brick. I illite, Q quartz, F feldspar,
octahedral sheets with OH- groups in the lattice, can be C calcite
expected. Therefore, the dehydroxylated ceramic body can
absorb a significant amount of water.
The small endothermic minimum at *570 °C is con- emission of water takes place only at the lower tempera-
nected with a ? b transformation of quartz. It is caused by tures attributed to the physically bound water. The slow
the significant amount of quartz in the sample. The third emission of CO2 starts at 170 °C and has two small max-
endothermic minimum belongs to decomposition of calcite ima at 225 and 325 °C. An intensive emission of CO2
which is completed at *830 °C. Finally, we observe a together with small amount of CO starts at *650 °C and
typical DTA curve for illite above 800 °C [21–23], so we both reach to the maximum at 731 °C (Table 1). We have
can conclude that the raw material consists mostly of illite. to remark that the temperatures presented in Table 1
TG curve shows a significant mass loss during heating resulted from MS analysis are lower than those resulted
from 20 to *150 °C which can be explained by the lib- from DTA and TG caused by the big differences in the
eration of the physically bound water from pores. The sample masses: 50 mg of powder for EGA and compact of
second step of the mass loss has sharp termination at 4.2 g for DTA and TG. Table 1
*840 °C which is a typical feature of the TG curve of the Therefore, according to the results obtained with EGA,
decomposition of calcite [24]. Between these two signifi- we did not observe dehydroxylation as a consequence of
cant mass losses, a gradual mass loss is presented. the former rehydroxylation. The rehydroxylation was
If historical ceramic body is in contact with environ- probably precluded with a plaster on the wall. On the other
mental moisture, a slow progressive rehydroxylation takes side, the results of EGA produce a question about emission
place when hydroxyl groups are incorporated into dehy- of CO2 at the temperatures below 600 °C. It can originate
droxylated clay [25–28]. Therefore, during heating of the (or could be originated) from burning of simple organisms
rehydroxylated ceramics over 350 °C, the emission of the and also from emission of absorbed CO2. It is possible
constituent water is another source of the observed mass because of the two-year storage of the bricks in a place not
loss [2, 3, 8]. According to the calculation presented in protected from weather influence. This time was long
[27], a mass gain (as a consequence of rehydroxylation) of enough for penetration of the sources of CO2 inward the
our brick could be *1.1 % during 250 years which passed brick, but too short for significant rehydroxylation.
from a production of the brick till now. Decomposition of calcite runs between 600 and 800 °C
To make a matter more clear, we used mass spectrom- [24] according to reaction CaCO3 ? CaO ? CO2. If the
etry (MS) to find a composition of the gases evolved at sample, which contains illite and calcite, is heated above
thermal treatment of the sample studied (Table 1). In 800 °C and after that above 900 °C gehlenite and anorthite
Table 1 Dt is a temperature interval in which the emission will be formed, respectively [29, 30]. But the results of
of gaseous compounds was registered and Tpeak is the XRD analysis (Fig. 3) showed neither gehlenite nor anor-
temperature at which the rate of the emissions reaches to thite formation in the samples; therefore, we can assume
maximum value. These results showed surprisingly that the that the firing temperature was not higher than 800 °C.

123
 R. Podoba et al.

0.8 800

0.7 700

Firing temperature/°C
Relative expansion/%

0.6 600

0.5 500

0.4 400

0.3 300

0.2 200

0.1 100

0 0
0 200 400 600 800 1000 0 20 40 60 80 100 120 140 160 180
Temperature/°C Firing time/h

Fig. 4 Thermodilatometry of the Baroque brick Fig. 5 The assumed firing schedule for 7-day firing

We considered the possibility that the calcite found in thermodilatometric results that the maximum firing tem-
the bricks could be a secondary calcite that emerged as the perature of the bricks was lower than 800 °C. The TDA
product of the calcite decomposition according to reactions curve can be used for the estimation of the firing temper-
CaO ? H2O ? Ca(OH)2 and Ca(OH)2 ? CO2 ? CaCO3. ature of ceramic goods [32–34]. The method is based on
These reactions run at the natural temperature of the the fact that when the partially fired clay body undergoes
masonry, and for them to take place a contact with air an increasing temperature, shrinkage occurs above a certain
moisture and CO2 is necessary. However, the bricks in the temperature. The point on the dilatometric curve where
masonry were protected with plaster, and we did not find contraction starts is considered to be the original firing
the presence of any remnant of CaO in the bricks. The temperature. According to this rule, the temperature
absence of a clear mass loss step between 400 and 600 °C *700 °C, which corresponds to the beginning of the
in the TG curve, which could be ascribed to H2O from contraction, is the firing temperature of the studied bricks.
rehydroxylated clay, implies that no significant amount of As written above, the estimation of the firing temperature
H2O penetrated into the bricks. Consequently, calcite in the is based on the evaluation of the degree of conversion.
bricks can be considered original. Irreversible phase transformations, which were taken into
The results of XRD are presented in Fig. 3. The Baroque account, were dehydroxylation of phyllosicates and
brick dominantly contains quartz, along with small pro- decomposition of calcite. As mentioned above, the degree of
portions of felsdspar, calcite, and dehydroxylated illite/ conversion depends on the temperature and time that
muscovite. The presence of feldspar in the bricks indicates determine the amount of the dehydroxylated/rehydroxylated
that the bricks were fired below 900 °C [16]. Similarly, the phyllosicates and calcite after firing; and, consequently, the
presence of calcite in brick indicates that the maximum results obtained by the used method (TG, DTA, TDA, EGA,
firing temperature was even less than *800 °C which is and XRD). Therefore, if we want to estimate the initial firing
the upper limiting temperature of calcite decomposition for temperature of the studied samples, we should know an
calcite from south-western Slovakia [24]. approximate firing schedule. If the firing of bricks at higher
The results of thermodilatometric analysis (TDA) are temperatures is carried out with sufficient time, these reac-
depicted in Fig. 4. The TDA showed the presence of sig- tions described above can be completed throughout the
nificant amount of quartz, which corresponds to small DTA whole brick’s volume. After analysis of the possible firing
endothermic peak at *570 °C. Above 700 °C, TDA curve schedule (Fig. 5), we can conclude that there was enough
begins to bend down, i.e., contraction of the sample takes time to complete dehydroxylation in the whole brick’s vol-
place. It can be ascribed to decomposition of calcite [24]. ume at temperatures higher than 600 °C during at least 25 h.
After completion of this decomposition, the sample con- We can say that the minimum firing temperature was
tinues to expand up to the final experimental temperature of *600 °C. On the other side, the firing conditions were not
1,000 °C. If a greater amount of the glassy phase would be sufficient for the total decomposition of calcite. The upper
present in the brick, it would be indicated on TDA curve limit of the firing temperature cannot be estimated from the
via typical contraction caused by the pressing force of the decomposition of calcite only, which expresses itself in
dilatometer’s push-rod [31]. This contraction appears Fig. 1 by mass loss and the endothermic minimum and by a
above the transformation temperature of the glassy phase. contraction in Fig. 4. Besides that, we should take into
Since the glassy phase can arise during firing at the tem- consideration the technical limit of the kiln for firing the
peratures higher than 800 °C, we suppose from the bricks of sufficient quality for a small church not higher than

123
Research on historical bricks from a Baroque Church

15 m. The temperature limit of the kiln was probably not 11. De Benedetto GE, Laviano R, Sabbatini L, Zambonin PG.
higher than *700 °C. Infrared spectroscopy in the mineralogical characterization of
ancient potery. J Cult Heritage. 2002;3:177–86.
12. Feathers JK, Berhane M, May L. Firing analysis of south-eastern
Missouri Indian potery using iron Mössbauer spectroscopy. Ar-
Conclusions chaeometry. 1998;40:59–70.
13. Hanykýř V, Kloužková A, Bouška P, Vokáč M. Aging of his-
Bricks taken from a Baroque masonry of the church in Pác torical ceramics. Acta Geodyn Geomater. 2009;6(153):59–66.
(Trnava County, Slovakia) were investigated by XRD and 14. Ondruška J, Trnı́k A, Vozár L. Degree of conversion of dehy-
droxylation in a large electroceramic body. Int J Thermophys.
different thermal analyses methods like DTA, TG, EGA, 2011;32:729–35.
and TDA. It was found that the bricks contained dehydr- 15. Kristály F, Kelemen E, Rózsa P, Nyilas I, Papp I. Mineralogical
oxylated illitic clay, calcite, feldspar, and quartz. As investigation of medieval brick samples from Békés county (SE
revealed, dehydroxylation was completely finished and no Hungary). Archaeometry. 2012;55:250–66.
16. Velraj G, Sudha R, Hemamalini R. X-ray diffraction and TG-
rehydroxylation was observed probably because of the DTA studies of archeological artifacts recently excavated in
protection of the bricks by plaster. Partial decomposition of Salamankuppam Tamilnadu. Recent Res Sci Technol. 2010;2:
calcite was also determined. The estimated firing temper- 89–93.
ature is *700 °C. The bricks have relatively high porosity 17. Wagner U, Gebhard R, Hausler W, Hutzelmenn T, Reiderer J,
Shimada I, Sosa J, Wagner FE. Reducing firing of an early pot-
*43 % and specific surface area *18.6 m2 g-1. tery making kiln at Batan Grande, Peru: a Mossbauer study.
Hyperfine Interact. 1999;122:163–70.
Acknowledgements This work was supported by grant VEGA 18. Podoba R, Trnı́k A, Podobnı́k Ľ. The upgrating of TGA/DTA
1/0646/12 and APVV-VVCE-0033-07, Ministry of Education of analyzer derivatograph. Épitöanyág. 2012;64:28–9.
Slovakia and from the grant SF0140082s08, Estonian Ministry of 19. Štubňa I, Kozı́k T. Permeability of the electroceramics to gas and its
Education and Research. The authors thank E. Smrčková (Department dependence on the firing temperature. Ceram Int. 1997;23:247–9.
of ceramics, glass and cement; Slovak University of Technology) for 20. Norton FH. Fine ceramics technology and applications. New
analysis of the CaO contents. York: McGraw-Hill Co; 1970.
21. Šašek L. et al. Laboratory methods in silicate science. Praha:
SNTL/Alfa; 1981.
References 22. Czechoslovak state standard CSN 72 1083: Thermal analyses of
ceramic raw materials.
1. Vandiver P. The role of materials research in ceramics and 23. Freund F. The defect structure of metakaolinite. Proc Int Clay
archeology. Annu Rev Mater Res. 2001;31:373–85. Conf. Madrid, 1972;13–25.
2. Drebushchak VA, Mylnikova LN, Drebushchak TN. The mass- 24. Teleki A, Vozár L, Tóth A. Kinetics of the irreversible chemical
loss diagram for the ancient ceramics. J Therm Anal Calorim. process. Cophys—International Physics Workshop. UKF Nitra,
2011;104:459–66. Nitra 2008;99–107.
3. Drebushchak VA, Mylnikova LN, Molodin VI. Thermogravi- 25. Nunez R, Capel J, Reyes E, Delgado A. Hydrolysis of ceramic
metric investigation of ancient ceramics. J Therm Anal Calorim. materials: neoformation or rehydroxylation of clay minerals
2007;90:73–9. oxygen stable isotope analysis. Clay Miner. 2002;37:345–9.
4. Cardiano P, Ioppolo S, De Stefano C, Pettignano A, Sergi S, 26. Hall C, Hamilton A, Wilson MA. The influence of temperature on
Piraino P. Study and characterization of the ancient bricks of rehydroxylation (RHX) kinetics in archeological pottery.
monastery of ‘‘San Filippo di Fragalà’’ in Frazzano (Sicily). Anal J Archeol Sci. 2013;40:305–12.
Chim Acta. 2004;519:103–11. 27. Wilson MA, Carter MA, Hall C, Hoff WD, Ince C, Savage SD, McKay
5. Branda F, Luciani G, Costantini A, Piccioli C. Interpretation of B, Betts IM. Dating fired-clay ceramics using long-term power law
the thermogravimetric curves of ancient pozzolanic concretes. rehydroxylation kinetics. Proc Royal Soc. 2108;A465:2407–15.
Archaeometry. 2001;43:447–53. 28. Hamilton A, Hall C. A review of redehydroxylation in fired. clay
6. Ion RM, Dumitriu I, Fierascu RC, Ion ML, Pop SF, Radovici C, ceramics. J Am Ceram Soc. 2012;95:2673–8.
Bonghez RI, Niculescu VIR. Thermal and mineralogical inves- 29. Rodriguez-Navarro C, Ruiz-Agudo E, Luque A, Rodriguez-
tigations of historical ceramics. J Therm Anal Calorim. Navarro AB, Ortega-Huertas M. Thermal decomposition of cal-
2011;104:487–93. cite: mechanisms of formation and textural evolution of CaO
7. Sánchez Ramos S, Bosch Reig F, Gimeno Adelantado JV, Yusá nanocrystals. Am Mineral. 2009;94:578–93.
Marco DJ, Doménech Carbó A. Application of XRF, XRD, 30. Cultrone G, Rodriguez-Navarro C, Sebastian E, Cazalla O, De la
thermal analysis, and voltammetric techniques to the study of Torre MJ. Carbonate and silicate phase reactions during ceramic
ancient ceramics. Anal Bioanal Chem. 2002;373:893–900. firing. Eur J Mineral. 2001;13:621–34.
8. Drebushchak VA, Mylnikova LN, Drebushchak TN, Boldyrev 31. Ventureli C, Paganelli M. Sintering behavior of clays for the
VV. The investigation of ancient pottery–application of thermal production of ceramics. Cfi/Ber. DKG. 2007;N5:E1–3.
analysis. J Therm Anal Calorim. 2005;82:617–26. 32. Tite MS. Determination of the firing temperature of ancient
9. Garcia Gimenez R, Vigil de la Villa R, Petit Domınguez MDMI, ceramics by measurement of thermal expansion: a reassessment.
Rucandio MI. Application of chemical, physical and chemo- Archaeometry. 1969;11:131–43.
metric analytical techniques to the study of ancient ceramic oil 33. Roberts JP. Determination of the firing temperature of ancient
lamps. Talanta. 2006;68:1236–46. ceramics by measurement of thermal expansion. Archaeometry.
10. Zindovič ND, Damjanovič LS, Holcajtner-Antunovič ID, Mioč 1963;6:21–3.
UB, Bajuk-Bogdanovič D. Ispitivanje srednjevekovne keramike 34. Petrovič S, Jovanov V, Vujovič S, Ranogajec J. Historical
Ras fizičko-hemijskim metodama. Hemijska industrija. 2008;62: materials from the medieval fortress Bač. Process Appl Ceram.
138–42. 2007;1:75–80.

123