G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

Journal of Cultural Heritage xxx (2014) xxx–xxx

Available online at

ScienceDirect

Original article

Formulation andmicrostructuralevaluationoftuffrepairmortar
Lidija Korata,∗, Breda Mirticˇb, Ana Mladenovicˇa, Alenka Mauko Pranjic´ a, Sabina Kramara

ar ti cl e i nf o a bs tr ac t

Article history: The purpose of this study was to develop a compatible repair mortar for two lithotypes of Peracicaˇ tuff (Slovenia): fine-grained and
Received 10 January 2014 coarse-grained. Mineralogical–petrographic characterization of tuff was carried out via optical microscopy and X-ray powder
Accepted 11 November 2014 Available online diffraction, and the relevant physical–mechanical properties of tuff and repair mortars were determined: capillary water absorption,
xxx water absorption at atmospheric pressure, open porosity, compressive strength and resistance to salt crystallization. A repair mortar
was prepared by mixing crushed Peracicaˇ tuff and quartz sand, with different grain size and mass ratios, and ethyl silicate (KSE 500
STE, Remmers) as a binder. The results showed that with an appropriate combination of crushed tuff, quartz sand and ethyl silicate, a
Keywords: compatible repair mortars can beobtained
Tuff
Repair mortar
a
Slovenian National Building and Civil Engineering Institute, Dimicevaˇ 12,1000 Ljubljana, Slovenia b University of Ljubljana, Faculty of
Natural Sciences and Engineering, Department of Geology, Askerˇ cevaˇ 12,1000 Ljubljana, Slovenia

Silica-based mortars
Microstructure for both lithotypes.
Deterioration
© 2014Elsevier Masson SAS. All rights reserved.
E-mail addresses: [email protected] (L.Korat), [email protected] (B.
Mirtic),ˇ [email protected] (A. Mladenovic),ˇ [email protected] (A.M. Pranjic),´
[email protected] (S.Kramar).

http://dx.doi.org/10.1016/j.culher.2014.11.002 1296-2074/© 2014

1. Introduction and research aims Elsevier Masson SAS. All rightsreserved.
formation due to the interaction between air and contaminants [3]. It
In the context of the conservation and restoration of stone should also be taken into consideration that beside the aforementioned
monuments, the use of materials that are compatible with the properties of different binders, some characteristics could influence
original materials is essential. Areas of badly damaged stone may mixing processes for example, adding more binder makes the mortar
occasionally necessitate the replacement of the original stone with easier to work but increases subsequent shrinkage [4]. Depending on
equivalent stone of the same origin or with that of similar properties. the humidity and porosity, the hardener can form a kind of skin on the
The replacement stone should replicate the original material as far mortar [3], while the shape of the filler grains is also an important
as possible in terms of authentic appearance, physical parameter influencing mortar properties[4].
characteristics and durability. In some cases, the conservation– Since the deterioration of tuffs is a significant problem not only in
restoration of stone elements requires the use of appropriate Slovenia but also worldwide [1,10–15], the development of proper
moldable materials and/or mortar [1–9]. Such repair mortars repair mortars would represent a significant contribution to
should be visually, chemically and mechanico-physically conservation–restoration interventions. Peracicaˇ tuff (Oligocene)
compatible with the original stone. In the field of silica-based repair has widely been used as building stone in the Gorenjska region (NW
mortars, such as those based on ethyl silicate, when applied to Slovenia) dating back to Roman times, as evidenced by finds at the
siliceous stone material, the predominant advantage is the use of a archaeological site of Mosnjeˇ [16]. Peracicaˇ tuff has been extracted
binder system very similar to the stone composition [4]. Siliceous from several quarries across NW Slovenia, with extensive use of tuff in
bound mortars are very sensitive to the binder–filler ratio and this region taking place from the 16th until the late 19th century for
additionally demand the strict observation of filler grain size building stairs, corner pillars on buildings, window frames and portals
distribution [4]. Similarly, for the restoration, conservators– [17]. Similar use of tuff stone is also reported in other countries [18],
restorers frequently have used mortars composed of crushed stone where its selection is ascribed to its typical green color and the fact that
bound with lime or hydraulic lime [6]. When applied for repairing extraction and processing are relatively simple [19]. However,
of granite and other siliceous stones, lime-based mortars are less Peracicaˇ tuff is not resistant to environmental factors and is thus often
successful ashigh calcium content,whichcauses salt heavily degraded, with some examples inadequately replaced with
cement mortar [20]. The most common weathering forms affecting tuff
are erosion, flaking, rounding, loss of material, discoloration and
fissuring, as well as the presence of microorganisms, granular
∗ Corresponding author. Tel.: +386 1 2804 538.

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002
G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

2 L.Korat et al. / Journal of Cultural Heritage xxx (2014) xxx–xxx
disintegration and fragmentation [18–21]. Sensitivity to weathering is 50 m, with data collected at 40 kV and a current of 30 mA, in the range
oftenrelatedto the material’s high clay minerals content [20]. from 2 to . The crystalline standard used to prepare the binary
The main aim of this research was to design, prepare and select mixtures was alumina powder (676a–NIST). Tuff samples and
appropriate repair mortars for Peracicaˇ tuff and to evaluate their standards were mixed in a ratio of 50/50. All Rietveld refinements were
mechanico-physical characteristics. The devised repair mortars are done using the X’Pert HighScore Plus program from Panalytical. To
based on ethyl silicate with an aggregate of crushed tuff and quartz sand better assess chemical properties and to complement the existing data,
fractions. Also, the aim of this study was to determine the both lithotypes of tuffs were examined via the backscattered electrons
mineralogical–petrographic characteristics and mechanico-physical (BSE) image mode of low vacuum Scanning Electron Microscopy
properties of two lithotypes of Peracicaˇ tuff. As a variety of different (SEM) and energy-dispersive X-ray (EDS) technique, using JEOL
potential mixtures are available, it is of vital importance to understand 5500 LV equipment.
the precise role of each individual component on the overall properties Tests for mechanico-physical characteristics (Table 2) were
ofeachmortar formulation. carried out on cubed tuff samples (edge length 50 ± 5 mm for all tests
with the exception of open porosity and tests for resistance to salt
2. Experimental
crystallization, for which a length of 40 ± 1 mm was employed) and
2.1. Materials on cylindrical repair mortar samples (diameter 30 ± 5 mm and
height 50± 5mm).
2.1.1. Tuff
Two types of Peracicaˇ tuff from the Bogataj (Fig. 1c) and Crnivecˇ Water absorption coefficient by capillarity of tuff samples and
(Fig. 1a) quarries in Peracicaˇ (NW Slovenia) were selected for mortar mixtures was conducted as specified in standard EN
investigation: fine-grained (labeled as A, Fig. 1b) and coarse-grained 1925:2000 [25]. Analyses were performed on three cubes for tuff
(labeled as B, Fig. 1d). Both quarries are located southeast of the town of samples and on three cylinders for mortar mixtures samples,
Radovljica. respectively. In case of tuff, the XY direction of the polished surface
was the one in contact with the water, perpendicularly with the
2.1.2. Mortar formulation anisotropy planes. Water absorption at atmospheric pressure of
In order to develop an appropriate repair mortar, mixtures of tuff samples and mortar mixtures was determined according to
crushed tuff (the two types of Peracicaˇ tuff previously described, fine- standard EN 13755:2008 [26], where experiments were performed
and coarse-grained tuff) and quartz sand (extracted from a gravel pit at with three cubes/cylinders for each sample. The tests for open
the Mura River, Slovenia) with different grain sizes, mineralogical– porosity (evacuation vessel Edwards XDS5 with pressure 2.0 ± 0.7
petrographic characteristics and mass ratios were prepared. Crushed kPa) of tuff samples and mortar mixtures were performed on three
tuff and quartz sand were milled (using a Siebtechnik Screening disc cubes/cylinders according to standard EN 1936:2007 [27]. The
mill TS 250) and screened into various gradations (using Retsch test compressive strength of tuff samples and mortar mixtures was
sieves), characterized by an average particle diameter lower than 63 m, determined according to the standard EN 1926:2007 [28] under the
another gradation ranging from 63 to 125 m, 125 to 250 m and from 250 ALFA 2–3000, digimax c–20 with the loading speed (1 ± 0.5)
to 500 m, respectively. Ethyl silicate (KSE 500 STE, Remmers, MPa/s and on a set of three cubes/cylinders for each sample. The
tetraethoxysilane, which is based on an ethyl ester of silicic acid, more tests for resistance to salt crystallization of tuff samples and mortar
data is written in technical guidelines [22,23]) was used as a synthetic mixtures were performed according to standard EN 12370:2000
binder between the grains of tuff and quartz. The reaction of the ester of [29]. The 14% solution of sodium sulphate decahydrate (Na2SO4 ×
silicic acid (Si (EtOH)4) with water (H2O) results in the formation of 10H2O) was used for salt crystallization tests and a total of 15 cycles
silica gel (SiO2, aq.). A by-product of the reaction is alcohol (EtOH, were done on a set ofthree cubes/cylinders for each sample.
“ethanol”), which evaporates [24]. The majority of the polymerization
oftetraethoxysilane is completedwithin 30 days.
The mixtures (a more detailed review can be found in [19]), with the
same gradation, but with different mass ratios of the aforementioned 3. Results anddiscussion
tuffs and quartz sand were prepared. The quantity of added KSE 500
STE binder varied within the prepared mixtures. The mixtures were 3.1. Mineralogical–petrographic characteristics ofPeracicaˇ
placed in larger plastic sheaths (50 mm tubes with a diameter of 30 mm) tuff
with openings (holes) on the side, which were used in order to prevent
air mixing with the mortar samples. On the basis of the visual criteria The fine-grained tuff is a homogenous light green in color,
including sample shrinkage, cracking, texture and color, six mixtures corresponding to pale green (5G 7/2) on the Rock Color Chart. A
were selected as the most appropriate for further investigation. These slight grain orientation can be observed in places. This welded
six mixtures (with a mass ratio aggregate to binder of 3/1) are presented vitric tuff consists mostly of grains smaller than 1 mm, together
in this paper.Theirs compositions are given in Table 1. with pumice embedded within the matrix. Thin section
examination revealed grains of quartz to be predominant. Most of
2.2. Methods these quartz grains are xenomorphic, with occasional idiomorphic
examples. The average quartz grain size is 400 m. Grains of
The colors of the two lithotypes of Peracicaˇ tuff were determined plagioclase (Fig. 2c) can also be found, with an average grain size of
according to the Rock Color Chart. Polished thin sections of tuff 300 m. Among the femic minerals individual grains of biotite are
samples were studied in a polarizing light in an optical microscope present, although in some spots biotite is replaced by chlorite
using an AXIO Imager A2m microscope (Carl ZEISS). The mineral (chlorites group of phyllosilicate minerals) (Fig. 2a), which may be
composition of the tuff samples was then determined via X-ray formed diagenetically by the alteration of detrital ferromagnesian
diffraction in a Philips PW3710 X-ray diffractometer equipped with minerals [30]. The husks of chlorite are located in a group of
Cu K1 radiation with = 0.115406 nm and a secondary graphite individual zones and exhibit the same orientation in all thin
monochromator. The samples were milled to a particle size of less than sections. It was formed diagenetically by the alteration of detrital

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002
G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

L. Korat et al./ Journal of Cultural Heritage xxx (2014) xxx–xxx 3
ferromagnesian minerals [30]. Grains of illite/muscovite, same orientation in all thin sections. Grains of illite/muscovite,
chalcedony (silicification by secondary changes) and carbonate chalcedony (silicification by secondary changes) and carbonate can
can also be observed, as well as grains of pyrite under crossed alsobe seen under thin
polars. The matrix is cryptocrystalline. Identification of individual
minerals in the matrix is near impossible under an optical Table 1
microscope. section (Fig. 2d), with grains of pyrite visible under crossed polars. The

Fig. 1.Two types of Peracicaˇ tuff from the Bogataj (c and d) and Crnivecˇ (a and b) quarries.
The heterogeneous coarse-grained tuff is also mainly light green in matrix is cryptocrystalline.
color, corresponding to light greenish gray (5G 8/1) on the Rock Color According to X-ray powder diffraction results (Fig. 3a and b),
Chart. Smaller grains (1–3 mm) are darker in color and can be divided quartz and plagioclase feldspars are the most abundant phases in both
into two groups: pale green (10G 6/2) and grayish green (5G 5/2). tuff lithotypes, followed by alkali feldspars, illite/muscovite and
Occasional dark fields of smaller inclusions are also present, while analcime. These results are in accordance with those obtained in
open pores, subsequent filling and cracks are macroscopically not [31,32] and are in agreement with similar tuffs examined elsewhere
observed. In some areas a slight grain orientation is noticeable. This [1,33]. Pyrite, chlorite (chlorites group of phyllosilicate minerals) and
welded vitric tuff consists mostly of grains between 2 mm and 3 mm in carbonate levels were beyond the X-ray detection limits; these
size, together with pumice embedded within the matrix. Grains of minerals could only be recognized under the optical microscope. No
quartz are again predominant, most of which are xenomorphic, with significant differences in XRD peaks between the lithotypes were
some idiomorphic. The average quartz grain size is 300 m. Individual observed. As can be seen from Table 3, the major difference between
grains of plagioclase (Fig. 2b) with an average size of 500 m are also tuff lithotypes is in quantity of amorphous phase and in the presence of
observable. Among the femic minerals individual grains of biotite are quartz, where quartz is enhanced in fine-grained tuff and amorphous
present, although in some spots alteration of biotite to chlorite has phase level is smaller than in coarse-grained tuff. In addition, analcime
taken place. The husks of chlorite (chlorites group of phyllosilicate was found in both tuff lithotypes as a consequence of sedimentation
minerals) are located in a group of individual bands and exhibit the processes of volcanic material in aquatic environment [31,34] or
influence from hydrothermalfluid activity [31,33–36].
Composition of the selected mixtures, where t: tuff, q: quartz sand, A: fine-grained tuff, B: coarse-grained tuff and KSE (500 STE): ethyl silicate binder.
Samples Tuff Tuff Tuff Tuff Quartz sand Binder (mass %)
< 63 m (mass %) 63–125 m (mass %) 125–250 m (mass %) 250–500 m (mass %) 63–125 m (mass %)
3t B 6 15 3 6 / 10
4t B 6 6 12 6 / 10
5t B 6 3 15 6 / 10
4t A 6 6 12 6 / 10
1q A 6 6 6 3 9 11
1q B 6 6 6 3 9 11

Table 2
Water absorption, open porosity and compressive strength properties of the two Peracicaˇ tuffs
lithotypes and the developed repair mortars.
Tuff Mortar

A B 3tB 4tB 5tB 4tA 1qB 1qA

Open porosity (wt%) 20.6 0.4 19.7 0.4 39.7 2.8 35.2 1.9 34.0 2.5 34.1 1.3 33.4 1.4 36.8 0.9
± ± ± ± ± ± ± ±
Water absorption coefficient by capillarity 25.5 ± 2.4 28.1 ± 1.5 53.6 ± 8.9 39.8 ± 1.8 36.8 ± 4.1 36.9 ± 3.1 37.5 ± 2.8 45.8 ± 3.5
(gm−2s−0.5)

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002
G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

4 L.Korat et al. / Journal of Cultural Heritage xxx (2014) xxx–xxx
Water absorption at atmospheric 8.3 ± 0.4 8.0 ± 0.0 22.8 ± 3.2 18.1 ± 0.7 17.4 ± 2.4 17.0 ± 1.3 14.5 ± 1.0 17.8 ± 1.4
pressure (%m/m)
Compressive strength (MPa) 85.8 ± 14.8 72.0 ± 10.0 6.4 ± 1.2 8.0 ± 1.4 5.3 ± 0.4 3.7 ± 1.4 6.5 ± 1.8 6.1 ± 1.9

Fig. 2. Microphotograph of the two tuff lithotypes (fine- and coarse-grained), where (a) is fine-grained tuff with a chlorite (Cl) grain and matrix (m), (b) is coarse-grained tuff with a
plagioclase (Pl) grain, (c) is fine-grained tuff with a plagioclase (Pl) grain, and (d) is coarse-grained tuff with a carbonated (C) plagioclase (Pl) grain.

Fig. 3.The results of X-ray powder diffraction, where (a) is fine-grained tuff (A), (b) is coarse-grained tuff (B).
The SEM-EDS analyses were performed in order to complement were reconfirmed also by the X-ray diffraction method. Also X-ray
the data of both lithotypes of Peracicaˇ tuff. The findings in both tuff diffraction reconfirmed the SEM-EDS analyses of higher amounts of
lithotypes revealed and confirmed by distribution of elements quartz in the finegrained tuff. The coarse-grained tuff contains larger
determined by SEM/EDS high amounts of Si, Al, Na or Ca, indicating matrix content and amounts of amorphous phase, a pattern in
plagioclase feldspars minerals, which were reconfirmed by the X-ray accordance with samples analyzed by Kastelic [31]. Also we can
diffraction method. The SEM-EDS analyses have determined the confirm with results from X-ray powder diffraction and in accordance
presence of felsic and mafic plagioclases. Also, both tuff lithotypes, with classification in the QAPF diagram for volcanic rocks andsamples
beside contain high amounts of Si, and Al, also contained K, which analyzed by Kastelic [31], the Peracicaˇ tuffs have a dacite
indicates alkali feldspar minerals. Besides feldspars minerals both tuff composition.
lithotypes contain illite/muscovite, chlorite content and pyrite. 3.2. Openporosity oftuff andmortar
Analcime crystals are well defined in the matrix and observed by SEM-
EDS analyses, higher amounts are present in fine-grained tuff, which Openporosity valuesrangedfrom19.7 to 20.6wt% for tuff and

Table 3
The results of X-ray powder diffraction, crystalline and amorphous composition of two tuff lithotypes (A: fine-grained tuff, B: coarse-grained tuff) by Rietveld method.
Quartz/% Plagioclase feldspars/% Illite/muscovite/% Analcime/% Alkali feldspars/% Amorphous/%

Tuff A 41.6 32.8 7.0 5.2 2.9 10.5

Tuff B 34.4 31.8 5.5 1.4 5.3 21.6

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002
G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

L. Korat et al./ Journal of Cultural Heritage xxx (2014) xxx–xxx 5
33.4 to 39.7 wt% for mortars, respectively. As can be seen from [38], i.e. samples with a higher quantity of the fine-grained fraction
Table 2, the open porosity of the fine-grained (labeled as A) tuff is will absorb water faster.
only slightly higher than that of the coarse-grained stone (labeled as
B). The obtained values are quite low compared to those of similar
tuff [1,18,33,37], where porosity levels reached 50%, but are
comparable with those recorded by Török et al. [11] and Topal and
Sözmen[22].
All repair mortar samples exhibited greater open porosity than
both types of tuff, with the highest value recorded for sample 3tB
and the lowest for sample 1qB. Both samples shared the similar
grain size, where mortar 3tB had larger wt% aggregate compounds,
represented by an average particle diameter higher than 250 m. The
fine-grained tuff exhibited higher porosity than that of the coarse-
grained tuff, where sample 1qA presented higher porosity than
sample 1qB. However, this situation is reversed in the case of
samples 4tB and 4tA. The reduced porosity of samples containing a
higher quantity of larger fractions (> 125 m) in the aggregate
suggests that the addition of coarser material decreased mortar
porosity.
Fig. 4. Results of water absorption coefficient by capillarity tests for the investigated
samples.

3.3. Water absorption coefficientby capillarity of tuff and mortar

The results of the water absorption by capillary rise tests for all
tuff and repair mortar samples are reported in Table 2, with the
3.4. Water absorption at atmospheric pressure oftuff andmortar
respective capillary absorption curves presented in Fig. 4. Values
of the capillary absorption coefficient were quite similar in both of
The results of the water absorption tests at atmospheric pressure are
the investigated tuff lithotypes, varying from 25.5gm−2s−0.5 for
shown in Table 2. Analysis of this table shows that water absorption at
fine-grained to 28.1gm−2s−0.5 for coarse-grained tuff.
atmospheric pressure of the investigated tuff samples varied from 8.3
wt% for fine-grained to 8.0wt% for coarse-grained tuff.
Values of the capillary absorption coefficient for the The obtained water absorption data positively correlate with both
investigated repair mortars ranged from 36.9 to 53.6gm−2s−0.5 the open porosity and coefficient of capillarity results, since the higher
(Table 2). The the porosity and coefficient values the greater the water absorption.
The water absorption at atmospheric pressure of the investigated repair
mortars ranged from 14.5 to 22.8 wt%. Among the tested repair
mortars, sample 3tB had the highest absorption and sample 1qB the
highest level of water absorption by capillary rise was exhibited by lowest. Repair mortar sample 1qB exhibited the lowest water
sample 3tB and the lowest by samples 5tB and 4tA. This pattern can absorption value, similar to that of coarse-grained tuff. The
be explained by sample microstructure, with 3tB comprising more investigated fine-grained tuff presented higher values of water
grains smaller than 125 m. The presence of smaller grains thus absorption, resulting in sample 1qA exhibiting greater absorption than
increases capillary absorption in the repair mortars, especially sample 1qB, in which coarse-grained tuff was used in the aggregate.
those composedofgrains smaller than 125m. The opposite situation was again observed in the case of samples 4tB
As can be seen from the above results, the recorded open and 4tA. Nevertheless, it is clearly evident that a higher quantity of the
porosity values positively correlate with those for capillary rise, coarse-grained fraction reduces water absorption, as finer material
with the higher the coefficient the greater the porosity. It is well produces more channels for water penetration.
known that finer pores lead to a higher coefficient of capillarity

Fig. 5.Tuff samples in the process of salt crystallisation (length, width, height: 40 × 40 × 40 mm).

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002
G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

6 L.Korat et al. / Journal of Cultural Heritage xxx (2014) xxx–xxx

Fig. 6.Repair mortar samples in the process of salt crystallisation (height, radius: 40 mm, 15 mm).
3.5. Compressive strength of tuff and mortar positively correlated with both the capillary absorption coefficient
and water absorption tests, with the higher the mortar porosity the
The results achieved after 60 days of compressive strength tests are greater the coefficient and water absorption values. The addition of
shown in Table 2. As can be seen from this table, values ranged from quartz sand and a higher quantity of the coarse-grained fraction
72.0 MPa for coarse-grained tuff to 85.8 MPa for fine-grained tuff. This thus both reduced water absorption.
range likely reflects differencesin sample microstructure, with coarse- In conclusion, on the basis of the data of salt crystallization, the
grained tuff containing a greater volume of matrix (as well as open most compatible among the repair mortars were sample 5tB, which
porosity). contained an aggregate of tuff and binder, and repair mortar 1qB,
The developed repair mortars exhibited compressive strength which contained an aggregate of tuff, quartz sand and binder.
levels between 3.7 and 8.0 MPa. A repair mortar with fine-grained tuff Whereas mortar sample 5tB exhibited reduced absorption of water
exhibited lower compressive strengths than that with coarsegrained by capillary rise and lower compressive strength, mortar 1qB
tuff aggregate. In particular, this is observed at repair mortars 4tB and presented lower levels of both water absorption at atmospheric
4tA. Samples 1qA and 1qB have fairly similar compressive strength pressure and open porosity. In addition, the samples that seem to
valuesdue to their similar compositions of added quartz sandfraction. exhibit the best physical–mechanical properties (such as capillary
water absorption and compressive strength) are repair mortars 1qB
3.6. Salt crystallization of tuff and mortar and4tB.
Tuff deterioration is a major global problem and thus the
The results of the resistance to salt crystallization tests revealed that development of appropriate repair mortars will be beneficial not
the tuff samples underwent visible change no earlier than the 11th only for Slovenia but also for other countries worldwide. The
exposure cycle, when the surface of the stone began to degrade present study can hopefully contribute to the better conservation
(although not yet fully disintegrated; Fig. 5). The tuff samples treatment of tuff-based sculpturesandbuildings.
deteriorated via the loss of material, with the eventual complete
collapse of the specimens after exposure to salt solution occurring after Acknowledgements
15 cycles. During this process, all samples lost their green color. The
fine-grained tuff samples fissured along their edges and the external This research was supported financially by the Slovenian
surfaces were subjected to scaling. Due to the complete failure of the Research Agency (Grant No. 1000-10-310085). Many thanks go
specimens, it wasimpossible to calculate an accurate weight loss value. to Matej Dolenec (University of Ljubljana, Faculty of Natural
The results of salt crystallization tests for the investigated repair Sciences and Engineering, Department of Geology) for help with
mortars are shown in Fig. 6. During the testing process the repair mortar X-ray diffraction and to Nastja Rogan Smucˇ (University of
samples also gradually deteriorated and by the end of the cycles had Ljubljana, Faculty of Natural Sciences and Engineering,
been completely demolished. The first visible changes were observed DepartmentofGeology) for help with thin sections.
at the 5th cycle, with the loss of stone at the sample edges. Sample 4tA
broke up at the 7th cycle and also at the 10th cycle. Samples 3tB and 1qA References
broke up at the 11th cycle, two samples of mortar 5tB at the 13th cycle,
and1qB and5tB (again) at the 14th cycle (Fig. 5). [1] C.D. Vacchiano, L. Incarnato, P. Scarfato, D. Acierno, Conservation of tuff-
stone with polymeric resins, Constr.Build. Mater. 22 (2008) 855–865.
[2] M.J. Varas, M. Alvarez de Buergo, E. Perez-Monserrat, R. Fort, Decay of the
4. Conclusions restoration render mortar of the church of San Manuel and San Benito, Madrid,
Spain. Results from optical and electron microscopy, Mater. Charact. 56 (2008)
1531–1540.
The main difference between the investigated fine- and [3] O. Rolland, P. Floc’h, G. Martinet, V. Vergès-Belmin, Silica bound mortars for
coarsegrained tuff is microstructural, with the former containing a the repairing of outdoors granite sculptures, in: Proceedings of 9th International
greater amount of quartz and lower amounts of amorphous phase. The Congress on deterioration and conservation of natural stone, Venice, vol. 2,
2000,pp.307–315.
coarse-grained tuff also comprises a lower volume of plagioclase and a
[4] H. Leisen, E. Plehwe-Leisen, S. Warrack, Success and limits for stone repair
larger proportion of matrix material. mortars based on tetra ethyl silicate-conservation of the reliefs at Angkor Wat
The repair mortars containing quartz sand were characterized Temple, Cambodia, in: 12th Triennal Meeting ICOM Committee for
by lower open porosity levels, as were those subject to the addition Conservation, Lyon, 2000, pp. 737–742.

of a coarse-grained fraction. The obtained porosity data also

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002
G Model

CULHER-2910; No. of Pages 7 ARTICLE IN PRESS

L. Korat et al./ Journal of Cultural Heritage xxx (2014) xxx–xxx 7
[5] B. Szenerey-Kiss, Á. Török, Time-dependent changes in the strength of repair [33] T.G. Nijland, R.P.J. Van Hess, L. Bolondi, Evaluation of three Italian tuffs
mortar used in the loss compensation of stone, Environ. Earth Sci. 63 (2011) (Neapolitan Yellow Tuff, ufo Romano and Tufo Etrusco) as compatible
1613–1621. replacement stone for Römer tuff in Dutch built cultural heritage, Geol. Soc.
[6] K. Beck, M. Al-Mukhtar, Formulation and characterization of an appropriate Lond.333 (2010) 119–127.
lime-based mortar for the use with a porous limestone, Environ. Geol. 56 (2008) [34] A. Iijima, in: L.B. Sand, F.A. Mupton (Eds.), Geological occurrences of zeolites
715–727. in marine environments. Natural zeoloites, Pergamon, Oxford, 1987.
[7] Z. Al-Saad, M.A.H. Abdel-Halim, Laboratory evaluation of various types of [35] A. Iijima, A petrochemical aspect of the zeolite formation in volcanoclastic
mortars for the conservation of Qasr al-Bint monument, Petra-Jordan, Eng. rocks, in: Proceedings of the 27th International Geological Congress, vol. 4,
Struct. 23 (2001) 926–933. 1984.
[8] K. Beck, X. Brunetaud, J.D. Mertz, M. Al-Mukhtar, On the use of eggshell lime [36] M. Utada, Zeolitization in the continental margin, with special reference to those
and tuffeau powder to formulate an appropriate mortar for restoration purposes, in the Green tuff region in Japan, Yerbilimleri 14 (1987).
Geol.Soc. Lond. 331 (2010) 137–145 [special publications]. [37] H.T. Ozkahraman, A. Bolatturk, The use of tuff stone cladding in buildings for
[9] C. Bläuer, H. Häberli, A. Löffel, B. Rousset, Repair mortars for the sandstones of energy conservation, Constr. Build. Mater. 20 (2006) 435–440.
the Cathedral of Berne, in: 2nd Historic Mortars Conference HMC2010 and [38] A. Bourgès, Holistic correlation of physical and mechanical properties of
RILEM TC 203-RHM final workshop, 22–24 September, Prague, Czech selected natural stones for assessing durability and weathering in the natural
Republic, 2010. environment, Universität München, 2006 [Dissertation der Fakultät für
[10] T.C. Chen, M.R. Yeung, N. Mori, Effect of water saturation on deterioration of Geowissenschaften der Ludwigs – Maximilians].
welded tuff due to freeze-thaw action, Cold Reg. Sci. Technol. 38 (2004) 127–
136.
[11] A. Török, L.Z. Forgo, T. Vogt, S. Löbens, S. Siegesmund, T. Weiss, The
influence of lithology and pore-size distribution on the durability of acid
volcanic tuffs, Hungary, in: R. Prikryl, B.J. Smith (Eds.), Building stone decay:
from diagnosis to conservation. Geological Society, London, special
publication, 271,2007,pp. 251–260.
[12] T. Topal, B. Sözmen, Deterioration mechanisms of tuffs in Midas monument,
Eng. Geol. 68 (2003) 201–223.
[13] T. Topal, V. Doyuran, Analyses of deterioration of the Cappadocian tuff,
Turkey, Environ. Geol.34 (1) (1998) 5–20.
[14] G. Dell’Agli, C. Ferone, G. Mascolo, O. Marino, A. Vitale, Durability of
tufaceous stones threated with protection and consolidation products, in: V.
Fasina (Ed.), 9th International Congress on deterioration and conservation of
stone,Venice, June 19–24, 2000, pp. 379–385.
[15] K.J. Arreola, W. Wedekind, S. Siegesmund, Weathering, conservation and
restoration of the Santa Mónica Church in Guadalajara, Mexico, in: 12th
International Congress on the deterioration and conservation of stone, 2012
http://www.arch.columbia.edu/files/gsapp/imceshared/lld2117/StoneConf-
07-Tuesday-Poster-Presentations.pdf
[16] J. Lux, S. Kramar, B. Mirticˇ, Analysis of selected samples of mortars and other
construction materials from object 2 of the Roman villa near Mosnjeˇ ,
Monument Prot.44 (2008) 170–189.
[17] A. Ramovsˇ, Peracˇiskˇ i zeleni kamen. Pomnik mnogih gorenjskih cerkva in
domacijˇ , 2008, pp. 11–12.
[18] K. Polak, Weiberner tuff-deterioration and conservation based on the example
of Collegium Maximum in Torun, in: 11th International congress on
deterioration and conservation of stone,Torun, Poland, 2008.
[19] L. Korat, B. Mirticˇ, A. Mladenovicˇ, A. Mauko, S. Kramar, Development of
replacement materials to repair damage to stone elements made of Peracicˇ a
tuff, J. Prot.
Monuments (2014) [in press].
[20] H. Stück, L.Z. Forgó, J. Rüdrich, S. Siegesmund, A. Török, The behavior of
consolidated volcanic tuffs: weathering mechanisms under simulated
laboratory conditions, Environ. Geol.56 (2008) 699–713.
[21] B. Zatler Zupancˇicˇ, A. Mladenovicˇ, A. Ramovsˇ, J. Vesel, C. Avgustinˇ ,
Andezitni tuf v Sloveniji-kulturna dedisˇcinˇ a na Gorenjskem »Prispevek
skupnega dela slovenskih strokovnjakov s podrocjˇ a kamna, predstavljen na 7.
mednarodnem kongresu o poskodbaˇ h in obnovi kamna, ki je bil v Lizboni od
15. do 18. junija 1992« Informacije Zavoda za raziskavo materiala in
konstrukcij v Ljubljani, 309, 1993,pp.8–10 [34].
[22] http://www.remmers.co.uk/fileadmin/doc/pz/TL 0571 EN.pdf
[23] http://www.remmers.de/html/doc/tm/TM1 0713 DE.pdf
[24] P. Jürgens, Die Konservierung der Reliefs am Angkor Vat; Kambodscha;
Optimierung der Steinergänzungsmörtel im KSE-Modulsystem:
Diplomarbeit, Köln, 2002,pp.61.
[25] Natural stone test methods – Determination of water absorption coefficient by
capillarity. EN 1925:1999, European Committee for Standardization, Brussels,
1999.
[26] Natural stone test methods – Determination of water absorption at atmospheric
pressure. EN 13755:2008, European Committee for Standardization, Brussels,
2008.
[27] Natural stone test methods – Determination of real density and apparent density,
and of total and open porosity. EN 1936:2007, European Committee for
Standardization, Brussels, 2007.
[28] Natural stone test methods – Determination of uniaxial compressive strength.
EN 1926:2007, European Committee for Standardization, Brussels, 2007.
[29] Natural stone test methods – Determination of resistance to salt crystallisation.
EN 12370:2000, European Committee for Standardization, Brussels, 2000.
[30] G.V. Middleton, Encyclopedia of sediments and sedimentary rocks, Springer,
2003.
[31] V. Kastelic, Petrological and mineralogical characteristics of Peracicˇ a tuff,
RMZ Mater. Geoenviron. 55 (3) (2008) 377–388.
[32] P. Kralj, Zeolites in the Smrekovec volcanistic rocks, northern Slovenia,
Geologija 40 (1997) 247–281.

Please cite this article in press as: L. Korat, et al., Formulation and microstructural evaluation of tuff repair mortar, Journal of Cultural Heritage (2014),
http://dx.doi.org/10.1016/j.culher.2014.11.002