Construction and Building Materials 38 (2013) 785–789

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials

Fibre-reinforced lime-based mortars: A possible resource for ancient

h i g h l i g h t s

" Fibre reinforced hydraulic lime mortars can be used for repairing historic structures.
" Glass and basalt fibres can improve mortar mechanical behaviour in the post-cracking.
" Hydraulic mortars containing 2% of glass fibres have the best mechanical performances.

a r t i c l e i n f o a b s t r a c t

Article history: In order to design binding materials able to historical buildings restoration, physical–mechanical proper-
Received 4 May 2012 ties and microstructure of fibre-reinforced lime-based hydraulic mortars, have been studied, in compar-
Received in revised form 4 September 2012 ison with a reference hydraulic lime-based mortar with no addition. Fibre-reinforced mortars, even
Accepted 22 September 2012
characterised by larger porosity and lower mechanical strength than the reference, pointed up a clear
Available online 26 October 2012
improvement in the post-cracking behaviour and, regardless of nature and concentration of fibres, turned
from brittle materials to ductile materials. It was demonstrated that the addition of as low as 2% of glass
Keywords:
fibres leads to a toughening of the mortar and jointly to an improvement of the flexural load.
Glass fibre
Basalt fibre
Ó 2012 Elsevier Ltd. All rights reserved.
Hydraulic lime
Fibre-reinforced mortar

1. Introduction ters has been reported in recent years [3–5]. Hydraulic lime is also
attractive for its favourable thermohygrometric features (i.e., tran-
The restoration and conservation of buildings and monuments spiration, dehumidifying ability and insulation), which assure
of some hundred years ago making use of Portland cement mor- appropriate microclimatic conditions.
tars, a very common practice until recently, is matter of particular The major problem in the use of lime-based mortars is, how-
concern from either technical or eco-sustainability points of view. ever, connected to their undesirable plastic shrinkage, mainly in
Physical–mechanical features of cement mortars are, in fact, too dry environments, due to fast water evaporation. This limitation
different from those of the original lime-based mortars, in terms, may, however, be overcome by the incorporation of short fibres,
e.g., of brittleness, mechanical strength and thermal expansion able to reduce plastic shrinkage and jointly to improve some prop-
coefficient [1], with consequent negative effects on the durability erties of the product, such as ductility, flexural strength and dura-
of the masonry works. In addition, from the chemical side, dam- bility (in particular freeze–thaw resistance). Fibres, in fact, limiting
ages may arise as a result of the interaction between cement crack opening and distributing the stress to the nearby matrix, are
hydration products and some building stones, as observed, for in- apparently able to suppress strain localization and to prevent
stance, recently with clay bricks [2]. On the other hand, it is well therefore microcracks from developing into macrocracks [6].
known that manufacturing and use of increased amounts of Port- Performance depends logically on dimension, nature and sur-
land cement is a source of environmental alarm, due to exceeding face treatment of fibres. Different types of short fibres have re-
CO2 production and the connected greenhouse effect. cently been used as reinforcement with the aim to arrest plastic
To overcome the above problems, a re-awaking interest in the shrinkage and to reduce microcracks [7–10]. Most literature re-
use of hydraulic lime for the preparation of repair mortars or plas- ports are however focused on two fibres: (a) glass fibre [11], com-
monly used to reinforce cement mortars, concretes, plasters and
⇑ Corresponding author. Tel.: +39 081 7682395; fax: +39 081 7682394. gypsum and (b) basalt fibre [12], a material originated from volca-
E-mail address: [email protected] (B. Liguori). nic rock, commonly used to reinforce a wide class of composites.

0950-0618/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.conbuildmat.2012.09.050
786 F. Iucolano et al. / Construction and Building Materials 38 (2013) 785–789

Table 1
Main technical features of the fibres used.

Fibre SiO2/Al2O3 Strand Treat. Mean length (mm) Mean diameter (lm) E (GPa)
Glass (G) 6.25 Chopped Sizing 10 13 75
Basalt (B) 5.2 Chopped – 4–5 8–10 84

Table 2 Table 3
Composition of the experimental mortars. Chemical analysis (weight%) of lime used in this study.

Lime/sand Water/solida Glass fibre Basalt fibre Latex SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 I.L.
ratio ratio (%) (%) (%)
9.75 1.98 0.16 65.11 2.29 0.79 0.24 0.30 19.32
G1 1:3 0.18 1 – 1.5
G2 1:3 0.18 2 – 1.5
B1 1:3 0.18 – 1 1.5
B2 1:3 0.18 – 2 1.5 Apparent and real density, open porosity and pore size distribution were eval-
REF 1:3 0.22 – – – uated by mercury intrusion porosimetry using a Micromeritics (Autopore III)
a
porosimeter.
Solid is intended as lime + sand; percentages (w/w) refer to the solid fraction of
Mechanical characterization of the mortars was carried out according to Euro-
the mixture.
pean Standards [17]. Three-point flexural tests were performed using an Instron
5566 compression machine, with a 5 kN load cell. Processing flexural test curves al-
The latter is a glassy material, too. It is produced by melting of a lowed Young module and flexural toughness to be calculated. Compression strength
basalt rock at about 1450 °C, followed by a rapid extrusion. The tests were carried out on the two fragments of each compact, resulting from the
above flexural test, using an Instron 8501 machine, equipped with a 50 kN load cell.
continuous glass thread is then chopped into fibres of various The loading rate was 0.6 mm/min in each test. All the measurements were per-
length. The main features of basalt fibres are: high strength and formed in triplicate.
thermal stability and resistance to high temperature, corrosion, Microstructure of the fracture surfaces of the hardened compacts was investi-
and both acid and basic attack. gated by scanning electron microscopy (SEM, Cambridge S440).
This paper aims at evaluating the physical and mechanical ef-
fects of the inclusion of the two above fibres in hydraulic lime-
3. Results and discussion
based mortars for historical buildings restoration. A preliminary
report on this investigation has been presented in an Italian Meet-
Table 3 reports the chemical analysis of lime, whereas Fig. 1a
ing and published in Italian on a local technical journal [13].
shows the X-ray diffraction pattern of the same material. The
bumps appearing in the diffraction pattern of Fig. 1a are likely re-
2. Materials and methods lated to the possible presence of amorphous phases, considering
that the material is of natural origin. DTA and TG profiles, not re-
Mortars were prepared using a commercial, natural hydraulic lime, supplied by ported here for the sake of brevity, show, as expected, two endo-
MGN company (Schio, Vicenza, Italy). Lime origin is unknown but it has likely been
thermic effects at about 441 °C and 756 °C, connected to Ca(OH)2
obtained by ignition of a local marly limestone. According to the European Stan-
dards [14,15], the binder belongs to the class designated as ‘‘NHL 3,5’’ having the
dehydration and CaCO3 decarbonation. Combining all the above
following specifications: SO3 6 3% and free CaO P 9%. experimental data, points out that the investigated NHL is consti-
The investigated hydraulic lime, from now onwards designated as NHL, was tuted essentially by Ca(OH)2, CaCO3 and calcium silicates (CH, C
subjected to preliminary chemical, X-ray and thermal analyses. and CS, respectively in Fig. 1a). The high CaO content of NHL
Chemical analysis was performed as follows. A weighted sample was first cal-
(Table 3) demonstrates the prevalence of high-basicity CS, which
cined, then subjected to digestion, under microwave-induced heating (Perkin–
Elmer Multiwave 3000 oven), in a standard solution prepared by mixing 1 ml of suggests either the raw material (marly limestone) to have been
HCl, 1 ml of HNO3 and 4 ml of HF. After addition of H3BO3 (24 ml) to attain fluoride treated at high temperature or the final product to have been en-
complexation, the solution was analysed by ICP atomic emission spectrophotome- riched in CS by adding some Portland clinker.
try (ICP-OES, Perkin–Elmer Optima 2100 DV). Fig. 1b reports the X-ray diffraction pattern of the siliceous
X-ray diffraction (XRD) patterns for phase identification were obtained using a
Philips PW 1730 diffractometer (rad. Cu Ka1).
aggregate. It is mostly a quartz sand, with minor amounts (some
Simultaneous thermogravimetric (TG) and differential thermal analyses (DTA) units percent in total) of a feldspar (possibly albite, A in the figure)
were carried out in the temperature range 25–1000 °C, using a Netzsch STA409 and calcite, detected by TG and DTA, whose traces are not reported
PCLuxx apparatus (alumina crucibles; N2 gas flow; heating rate: 10 °C/min). here. The grain size, measured by sieving, was found to be <1 mm,
A siliceous fine aggregate (sand), supplied by Gras Calce company (Trezzo sul-
with almost 90% passing through a 0.5 mm sieve.
l’Adda, Milan, Italy) was used, too. Such material was subjected to XRD and grain
size analyses. Particle size distribution was obtained by mechanical sieving, accord- The X-ray diffractograms in Fig. 1c and d refer to fibres G and B,
ing to European Standards [16]. respectively (Table 1). Both materials proved to be completely
Two different types of fibres were added to the mortars: (a) a glass fibre, type E, amorphous. The large shoulder between 20° and 35°2H in Fig. 1c
earth-alkaline network modifier, supplied by Mapei, Milan, Italy, pre-treated with is normally interpreted in terms of a random network.
an acrylic sizing (referred to as G throughout the text) and (b) a basalt fibre, sup-
plied by Basaltex, Wevelgem, Belgium (referred to as B). The main technical features
Table 4 summarizes the physical parameters measured for all
of both fibres, made available by the suppliers, are reported in Table 1. Fine pow- the manufactured mortars. The values of real density of the four fi-
ders, obtained by grinding both fibres, were analysed by XRD. Fibres were employed bre-reinforced mortars, averaging 2.63 g/cm3, are not significantly
as chopped strands. Mortar mixtures, whose composition is shown in Table 2, were different. This amount is higher than the corresponding value for
prepared adding to the solids suitable amounts of water to obtain a normal consis-
the REF mortar (2.47 g/cm3), likely due to the higher water/solid ra-
tency and a good workability. Binder to aggregate ratio (w/w) was fixed at 1:3. G or
B fibres were added at two different rates (1% or 2%). In order to improve workabil- tio chosen for the preparation of the latter (Table 2). The apparent
ity and placement of the mortars after fibres addition, 1.5% (w/w) of a polymeric la- density and logically the open porosity show, on the contrary, sub-
tex (Mapei) was added to each mixture. A reference mortar (REF in Table 2), stantial differences. As expected, mortars with included fibres are
containing no fibres and no latex was also prepared and tested. remarkably more porous than the REF mortar (porosity: 45–52%
According to standard procedures, all the mortars were cured for 28 days in a
climatic chamber (MSL, mod. Humichamber EC 125) in the following conditions:
vs. 36%; apparent density: 1.22–1.45 g/cm3 vs. 1.59 g/cm3). This is
temperature, 20 °C; relative humidity, 95% during the first 7 days and 65% during due, in general, to the interruption of the continuity of material
the residual 21 days [17]. microstructure, as a result of the inclusion of heterogeneities in a
 F. Iucolano et al. / Construction and Building Materials 38 (2013) 785–789 787

C, CS
(a)

CH,CS
CS

CH

C, CH
C C

(b) Q

Fig. 2. SEM micrograph showing the interface between fibre and matrix in the G1.

Q A

Q
50
45 G1 (a)
G2
40
(c) 35
REF

30
Porosity, %

25
20
15
10
5
0
0.001 0.01 0.1 1 10 100
Mean diameter, μm

50

(d) 45 B1 (b)
B2
40
REF
35
30
Porosity, %

25
20
15
10
5
0
0.001 0.01 0.1 1 10 100
Mean diameter, μm
10 20 30 40 50
2θ (°) Fig. 3. Pore size distribution of mortars reinforced with: (a) glass, and (b) basalt
fibres.
Fig. 1. XRD patterns of: (a) natural hydraulic lime, (b) siliceous aggregate, (c) glass
fibres and (d) basalt fibres. C = calcium carbonate, CH = calcium hydroxide,
CS = calcium silicates, Q = quartz, A = albite.
interfacial transition zone between lime paste and fibres (see, e.g.,
Fig. 2), hindering the hydration or carbonation of lime paste.
Table 4 Additional causes of increase of porosity might be some surface
Physical properties of cured mortars. hydrophobicity, due to pre-treatments of fibres (as it has been
Mortar Open porosity Apparent density (g/ Real density (g/ clearly declared by the manufacturer for the G fibre, see Experimen-
(%) cm3) cm3) tal) or excess of gas trapped in the blend.
REF 35.6 1.59 2.47 Fig. 3a and b shows the porosimetric curves of five hardened
G1 48.3 1.37 2.66 mortars examined. The reference mortar (REF) shows a bi-modal
G2 45.2 1.44 2.64 pore size distribution, mainly characterised by small pores of
B1 52.4 1.24 2.61 1–3 lm diameter and a minor presence of pores in the range of
B2 53.6 1.22 2.63
hundreds of lm. This situation is substantially reversed when
fibres are added, whatever the type and amount. Fig. 3 evidences,
in fact, a marked increase of pore diameter in the range
homogeneous body. Specifically, glassy fibres, modifying the mor- 100–200 lm (especially for G1 and B1 mortars), whereas pores
tar workability, promote the formation of microcavities in the with 1–3 lm diameter are significantly less.
788 F. Iucolano et al. / Construction and Building Materials 38 (2013) 785–789

Table 5
Mechanical properties of cured mortars.

Mortar Compressive strength (MPa) Flexural strength (MPa) Flexural toughness (MPa) E (MPa)
7 Days 28 Days 7 Days 28 Days 28 Days 28 Days
REF 4.06 5.55 1.26 2.15 0.01 231.9
G1 1.72 3.12 1.06 1.58 0.05 350.3
G2 2.01 3.62 1.45 2.41 0.08 134.0
B1 1.53 3.52 0.84 2.16 0.04 285.3
B2 0.97 2.22 0.72 1.65 0.05 72.1

G2
2.5
B1
Flexural strengh, MPa

2
B2

1.5
REF G1
1

0.5

0
0 0.01 0.02 0.03 0.04 0.05
Strain, mm/mm

Fig. 4. Flexural behaviour of reinforced mortars.

The rising of total porosity and the increase of large pores in the
fibre-reinforced mortars affect negatively, as expected, the com-
pressive strength of the mortars (Table 5). The measured values
for all the reinforced mortars appear, in fact, lower than the refer-
ence, especially after 7 days, but also after 28 days curing (decre-
ment is 50–75% after 7 days and some 40%, on average after
28 days (Table 5, columns 2 and 3).
Nevertheless the presence of short fibres may result in an
improvement as regards the flexural strength. The fourth and fifth
columns of Table 5 demonstrate, in fact, that the values of the rein-
forced mortars compare better (at both curing times) with those of
the REF sample and in one case are decidedly better (G2 mortar
exhibits values which are some 15% higher than the reference).
Considering that cracking of a hardened lime mortar occurs as a
propagation of defects inside the lime matrix or between the inter-
face matrix–aggregate, it is definitely proved the fibre-bridge ac-
tion, that consists evidently in a break of matrix continuity with
a consequent reduction of defects propagation.
In addition, results obtained from three point flexural tests (see
Fig. 4) show that all the fibre reinforced mortars exhibit a ductile-
like behaviour, compared with the brittle-like behaviour of the
Fig. 5. SEM micrograph (3500) of: (a) G2, and (b) B2 samples (Table 2).
REF. Accordingly, the toughness values (sixth column in Table 5)
are definitely higher than the REF values. Inspecting the slopes of
the initial part of the curves in Fig. 4 points out that fibre addition the strong interaction between fibres and matrix. It is also evident
affects to different extent the elastic modulus, depending on fibre that fracture occurs through a fibre pullout rather than a fibre
nature and content (see also the seventh column in Table 5). B2 breakdown [18].
sample shows, in particular, an elastic modulus decidedly lower
than the other samples, resulting in a more ductile behaviour.
The better result is presented by the G2 samples, with a maximum 4. Conclusions
value of some 2.41 MPa.
Further information about fracture mechanism of the mortars The results obtained confirm that glass and basalt fibres are able
can be obtained from the morphological analysis of matrix–fibre to modify the mechanical behaviour of hydraulic lime-based mor-
interface (Fig. 5a and b). tars. In fact, a significant improvement of toughness and flexural
SEM micrographs of G2 and B2 mortars show a strong adhesion post-cracking behaviour have been observed in reinforced mortars,
between matrix and fibres, with no evident phases separation. This regardless of the nature and amount of fibres. The best results have
proves that the toughness of the fibre-reinforced mortars is due to been obtained adding 2% of glass fibres (G2 sample), either as
 F. Iucolano et al. / Construction and Building Materials 38 (2013) 785–789 789

regards the maximum strength (about 2.41 MPa) or the toughness [6] Pascal S, Alliche A, Pilvin P. Mechanical behaviour of polymer modified
mortars. Mater Sci Eng A 2004;380:1–8.
(0.08 MPa).
[7] Lanas J, Alvarez JI. Masonry repair lime-based mortars: Factors affecting the
The examined mortars can therefore be proposed as materials mechanical behaviour. Cem Concr Res 2003;33:1867–76.
for repairing historic structures, as they combine a good natural [8] Lanas J, Perez Bernal JL, Bello MA, Alvarez Galindo JI. Mechanical properties of
compatibility with ancient constructions, a good freeze–thaw resis- natural lime-based mortars. Cem Concr Res 2004;34:2191–201.
[9] Betterman LR, Ouyang C, Shah SP. Fiber-matrix interaction in microfiber-
tance and a reasonable mechanical behaviour in the post-cracking reinforced mortar. Adv Cem Based Mater 1995;2:53–61.
step. [10] Pigeon M, Pleau R, Azzabi M, Banthia N. Durability of microfibre-reinforced
mortars. Cem Concr Res 1996;4:601–9.
[11] Proctor BA. Glass fibre reinforced cement. Phys Technol 1975;6:28–32.
Acknowledgement [12] Sim J, Park C, Moon DY. Characteristics of basalt fiber as a strengthening
material for concrete structures. Compos Part B – Eng 2005;36:504–12.
G. Albano is gratefully acknowledged for assistance in carrying [13] Iucolano F, Liguori B, Colella C. Malte di calce fibro-rinforzate di potenziale
impiego nel ripristino del patrimonio edilizio (Fibre-reinforced lime-based
out some of the experiments. mortars potentially useful in the restoration of the building heritage). In:
Mascolo G, editor. Proceedings IV Meeting Monitoring and Saving Cultural
References Heritage. Italy: University of Cassino; 2010. p. 249–56 [ENCO Journal 2010;
XV(50):10–12.
[14] European Standards UNI EN 459-1: building lime – definitions, specifications
[1] Rota Rossi-Doria P. Mortars for restoration: basic requirements and quality
and conformity criteria. Ente Nazionale Italiano di Unificazione, Milan, Italy;
control. Mater Struct 1986;19(6):445–8.
2001.
[2] Moropoulou A, Cakmak AS, Biscontin G, Bakolas A, Zendri E. Advanced
[15] European Standards UNI-EN 459-2: building lime – test methods. Ente
Byzantine cement based composites resisting earthquake stresses: the
Nazionale Italiano di Unificazione, Milan, Italy; 2001.
crushed brick/lime mortars of Justinian’s Hagia Sophia. Constr Build Mater
[16] European Standards UNI EN 933-1: tests for geometrical properties of
2002;16:543–52.
aggregates – determination of particle size distribution – sieving method.
[3] Moropoulou A, Bakolas A, Bisbikou K. Investigation of the technology of
Ente Nazionale Italiano di Unificazione, Milan, Italy; 2009.
historic mortars. J Cult Heritage 2000;1:45–58.
[17] European Standards UNI EN 1015-11: methods of test for mortar for masonry
[4] Biscontin G, Pellizon Birelli M, Zendri E. Characterization of binders employed
– determination of flexural and compressive strength of hardened mortar. UNI
in the manufacture of Venetian historical mortars. J Cult Heritage
EN Ente Nazionale Italiano di Unificazione, Milan, Italy; 2007.
2002;3:31–7.
[18] Silva FJ, Thaumaturgo C. Fibre reinforcement and fracture response in
[5] Fassina V, Favaro M, Naccari A, Pigo M. Evaluation of compatibility and
geopolymeric mortars. Fatigue Fract Eng Mater Struct 2003;26:167–72.
durability of a hydraulic lime-based plaster applied on brick wall masonry of
historical buildings affected by rising damp phenomena. J Cult Heritage
2002;3:45–51.