Materiales de construcción 36 (2012) 46-57

listas de contenidos ofrecidos en SciVerse ScienceDirect

Materiales de construcción

Página principal de la revista: www.el Sevier. com / localizar / conbui ldmat

fibras de acero Fi de neumáticos de desecho como refuerzo en hormigón: Una caracterización mecánica

G. Centonze ⇑ , M. Leona, MA Aiello

Departamento de Ingeniería de la Innovación, Universidad de Salento, a través de Monteroni, 73100 Lecce, Italia

reflejos

" Se estudiaron las propiedades mecánicas de HRFA ​con fibras recicladas a partir de neumáticos de desecho.

" las propiedades del hormigón fresco y endurecido son reportados.

" Los ensayos de flexión en probetas entalladas se llevaron a cabo según la norma UNI 11039.

" Los ensayos de flexión en losas cuadradas se llevaron a cabo de acuerdo con EFNARC específico de cationes.

" Los resultados obtenidos confirman aplicaciones prometedoras de hormigón armado con RSF.

información del artículo resumen

Historia del artículo: El daño ambiental causado por el manejo inadecuado de neumáticos de desecho aumentado en los últimos años creando un problema relevante que
Recibido el 6 de febrero de 2012 hay que resolver. En el campo de la ingeniería civil resulta posible volver a utilizar la fibra de acero fi y el caucho de los neumáticos de desecho. En
Recibida en forma de 27 revisada de abril de 2012 29 aceptada de
particular, el hormigón obtenido mediante la adición de fibras de acero recicladas muestra una buena mejora mecánica de la matriz frágil, como
abril de 2012 Disponible No en línea del 15 de junio de 2012
consecuencia que parece ser un candidato prometedor para ambas aplicaciones estructurales y no estructurales. En el presente trabajo experimental,
como una continuación de la investigación que ya lleva a cabo en este campo por los autores, las actuaciones de craqueo post de RFRC (fibra
reciclada Hormigón Armado) se evaluaron mediante comprobaciones de los medios en los elementos fl exión y losas. La eficacia de las fibras
palabras clave:
recicladas se evaluó en comparación con los datos experimentales obtenidos para las muestras realizadas con la Federación Internacional (Industrial
llantas de acero fibras
Fiber Reinforced Concrete). Todas las propiedades fresco y endurecido de las mezclas de hormigón se estimaron de forma experimental. El
de residuos de

hormigón
comportamiento post-agrietamiento de la RSFRC, obtenida por pruebas exión fl, fue comparable con la de ISFRC. especímenes RSFRC mostraron

Losa resistencia a la flexión una buena absorción de energía y buena resistencia residual después de la fisuración.

Sin embargo, los problemas tecnológicos relacionados con la producción y el hormigón fibras se mezcla la preparación, debe ser investigado y aún sigue
siendo necesaria una investigación más amplia para validar los funda interesantes.
2012 Elsevier Ltd. Todos los derechos reservados.

1. Introducción
En los últimos años, la eliminación indiscriminada de grandes cantidades de neumáticos de
desecho (WT) en tierra fi LLS ha causado graves daños ambientales en muchos Estados europeos,
debido a su función no biodegradable conocido. Por otro lado, según [1] , Desde julio de 2006, la
eliminación de neumáticos enteros o triturados en con fi namiento se convirtió en ilegal. Como
consecuencia casi todos los gobiernos europeos incorporado esta Directiva a las leyes y los códigos
nacionales, promoviendo el desarrollo de opciones sostenibles para la eliminación, recuperación y
reutilización de los neumáticos. En este sentido, Italia ha sido muy cuidadoso y con visión de futuro:
la reciente / 98 / Ce Directiva Europea 2008 se expande en Europa el '' final de los residuos ''
concepto, ya se introdujo
⇑ Autor correspondiente. Tel .: +39 0832 297386.
por el legislador italiano [2] . En particular la posibilidad de gránulos de caucho mercado y polvos, así
como el acero de fibras sin estar sujeto a los estrictos requisitos de la legislación de residuos,
derivada de las disposiciones Códigos, podría facilitar en gran medida la difusión y uso de los
materiales obtenidos por el tratamiento de neumáticos de desecho.
Los principales mercados para el reciclaje de WT actualmente son la recuperación de energía (como
combustible horno en la industria del cemento) y la recuperación de materias primas. Los procesos
mecánicos y químicos, tales como la trituración de neumáticos, pirólisis y reducción criogénico son utilizados
por la industria del reciclaje de neumáticos.
Recientemente, en el contexto de un creciente interés hacia el reciclado de materiales
innovadores y edificios sostenibles, algunos estudios proponen la utilización de fibras de caucho y
acero fi granulados, recuperado de neumáticos de desecho, en concreto [3-11] . En particular, el
hormigón obtenido mediante la adición de fibras de acero reciclados mostró una buena mejora de la
matriz frágil, especialmente en términos de dureza y

Dirección de correo electrónico: giuseppe.centonze@unisalento.it (G. Centonze).

0950-0618 / $ - see front matter 2012 Elsevier Ltd. Todos los derechos reservados.
http://dx.doi.org/10.1016/j.conbuildmat.2012.04.088
 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57 47

el comportamiento post-fisuración, lo más ampliamente ya experimentada por (hormigón de fibras de On the other hand the use of a planetary mixer resulted particularly advantageous to improve the
acero industrial reforzado) ISFRC. De hecho es bien conocido que la dispersión de fibras de acero Fi homogeneous distribution of fibers within the concrete. In fact an increased fibers percentage was
en concreto permite una notable mejora de las propiedades mecánicas de hormigón, sobre todo en allowed in the concrete mix (0.46% by volume of concrete), that significantly improved the mechanical
términos de resistencia dinámica y la fatiga, esfuerzo cortante y post craqueo resistencia [12-16] . properties of the matrix, guarantying at the same time a good concrete workability. Up to

La resistencia a la formación de grietas post es principalmente debido a la capacidad de las
fibras en el control del crecimiento de grietas y las grietas de apertura, dependiendo principalmente
de fi propiedades Bers, unión entre fibras y hormigón y en la distribución de fibras. Una aplicación
con éxito de fibras de acero fi añadido se refiere a las losas de hormigón, donde las capas
tradicionales de una malla de acero cables pueden ser sustituidos por fibras, garantizando una
respuesta estructural similar o incluso mejorada, con una reducción de los costes y el tiempo de
ejecución [17-21] . Los RSFs (fibras de acero reciclado) parece ser un candidato prometedor para la
realización de losas de hormigón, barreras viales y elementos prefabricados, producidos por lo
general con las tecnologías tradicionales. Haciendo referencia a la pavimento de hormigón, en donde
generalmente se utilizan refuerzo de malla de acero, una sustitución parcial por fibras de acero
reciclado podría ser considerado sobre la base de los prometedores resultados ya obtenidos
utilizando fibras de acero industrial, [22] .
0.46% of recycled steel fibers were used also in the investigation reported in this paper; results have
been compared with those of concrete mixes obtained with a similar percentage of industrial steel
fibers and with results referred to the plain concrete. The concrete workability, compressive and
flexural strength were evaluated and discussed also in comparison to the already mentioned available
results.
The RSF, obtained by the process before mentioned, have different lengths and diameters. To
better understand the properties and the effectiveness of the used fibers, a geometrical
characterization was done in [3] on a sample of 1400 recycled steel fibers. A further geometrical
characterization will be reported here on recycled steel fibers, supplied by a different industry even
utilizing the same shredding process, in order to better appreciate the reliability of results obtained
also in relation to the expected variability of fibers properties.

Como resultado de ello, las ventajas económicas, propiedades físico-mecánicas no son de menor y una

contribución al problema de la contaminación bien conocido relacionado con los neumáticos de desecho

pueden ser obtenidos.

3. Geometrical characterization of recycled fibers
Sobre la base de los resultados reportados en la literatura disponible y (fibras de acero
reciclados de hormigón armado), también de los estudios previos realizados por los autores antes
Geometrical properties of recycled steel fibers have been determined on a sample of 2000 fibers,
mencionados, una investigación experimental dedicado al análisis de las propiedades mecánicas de
extracted randomly after the shredding process.
RSFRC se presenta en este documento. Trabajabilidad, ensayos exión de compresión y FL fueron
ejecutados por diferente mix-diseño, la variación de la relación en volumen de fibras de acero con
The diameter of each fiber was recorded by a micrometer and determined averaging three
adición en el hormigón. Después de ello, la mejora en el comportamiento de la muestra de
measures, namely at the two extremities of the fiber and at the mid length. The fibers diameters
post-formación de grietas, debido a la acción de crack-supresor de las fibras, se evaluó en cuatro
varied between 0.10 mm and 2.00 mm, while the average value is
puntos fl exión pruebas llevadas a cabo de acuerdo con [23] . Finalmente, las pruebas exión fl en
losas, de acuerdo con la directriz EFNARC [24] , Fueron ejecutados para analizar el rendimiento
0.24 mm. In addition, in order to give details on the geometrical properties of the utilized recycled
post-agrietamiento de elementos de hormigón bidimensionales, ampliamente utilizados en
fiber, eight diameter ranges were defined and the number of fiber included in each range was
aplicaciones prácticas. Para evaluar la eficacia de la RSFRC, todas las pruebas se llevaron a cabo
counted; this allowed plotting the percentage of the fiber in each class, Fig. 1 . The range that included
también en fibras de acero fi lisos e industriales de hormigón armado. La comparación entre los datos
the largest number of diameter values was 0.15–0.2 mm (32.75%) followed by the class 0.30–
experimentales proporciona una estimación razonable de la actuación mecánica del material
analizado.

0.35 mm and 0.20–0.25 mm, which presented respectively 27.0% and 24.6%.

Similary, the length of the reinforcement, referred to the distance between the outer ends of the
fibers ( Fig. 2 ), [25] , was recorded for each specimen; the minimum and the maximum value was 3
2. Consideraciones de apertura
mm and 170 mm respectively while the average was

La caracterización mecánica de RSFRC se ha llevado a cabo sobre la base de los resultados

31.4 mm. In this case the recorded measures were divided in fifteen classes, Fig. 3 . The range
obtenidos en estudios previos [3] in order to further validate the experimental founds and get a deeper
including the major number of fiber
insight into the mechanical performance of the RSFRC in the post cracking stage. In addition the
potential use of RSF in concrete slabs is investigated, aiming to evaluate if the addition of fibers could
greatly contribute to overcome or, at least, reduce the problems related to the use of conventional 35%
32,75%
steel mesh in consideration of the promising results already available when ISF are used in
substitution of the traditional reinforcement. The mixes preparation is a crucial aspect, due to the 30%
26,99%
tendency of the recycled steel fibers to bundle within the fresh concrete more than industrial fibers. In
24,66%
25%
fact, recycled steel fibers are generally recovered by a shredding process of tires, followed by an
electromagnetic procedure aiming to separate steel from rubber; the fibers obtained are characterized
Frequency (%)

20%
by an irregular shape and a variable length besides a residual magnetization. These aspects
significantly affect the maximum amount of fibers being added to the concrete mix as well as the
15%
quality of fibers dispersion in concrete. Based on the already performed investigation, when a
traditional concrete mixer (made by revolving drum) was employed the highest amount of fibers that
10%
could be effectively added to the mix was 0.26% by volume of concrete. 7,05%

4,72%
5%
2,59% 1,14%

0,16%
0%
< 0,1 0,1÷0,15 0,15÷0,2 0,2÷0,25 0,25÷0,3 0,3÷0,35 0,35÷0,4 > 0,4

Ranges of diameter (mm)

Fig. 1. Relative frequency of the diameter of the fibers.

48 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57

100

90

80

70

Equivalent fiber length

% Cumulative passing
60

50

Fig. 2. Utilized recycled fiber.

40

30

was 26–30 mm (17.20%), followed by the class 21–25 mm and 31– 35 mm respectively with 15.91%
20
and 14.35%. Referring to the mean values of both length and diameter the aspect ratio of the utilized
Cubic distribution
10
fiber is 131.
Design distribution
0
0,01 0,1 1 10 100

log (d)
4. Experimental campaign

Fig. 4. Granulometric curves.

4.1. Proprieties of fresh concrete and compression strength: specimens and testing procedure

two mixes reinforced with different ratios of recovered fibers (RSF1a and RSF2a) and a mix

The mechanical properties of fresh and hardened concrete were experimentally evaluated. In reinforced with industrial steel fibers (ISFa) for comparison purposes. The reported mixes are very

particular, slump tests were performed to estimate the influence of fibers on workability according to similar to those analyzed in [3] , in agreement with the aim of this research of enlarging the database

standard [26] . As reported in [27] , slump test is not appropriate for quantitative measures of to further validate obtained results. The industrial steel fibers were 0.6 mm 40 mm (diameter x length)

workability when referring to SFRC; however this test may be used as a quality control test to monitor with an aspect ratio L/D = 67. The slight difference between the analyzed mixes in terms of quantity of

the SFRC consistency from batch to batch. Compressive test were also performed, according to [28] , both aggregate I and II and sand is due to the design chosen to fit in all cases the same granulometric

for all utilized mixtures. Portland cement, water, limestone aggregates and locally available sand design cubic curve ( Fig. 4 ). In this regard, in order to achieve a better slip between the grains with a

were used for the mixture; a superplasticizer was also added to improve the fresh concrete consequent higher workability of the concrete the design was made considering a cubic curve instead

workability. Specimens were realized using a planetary vertical concrete mixer with four different of the Fuller curve. In addition when RSF are considered a higher percentage of superplasticizer was

mix-design ( Table 1 ): one control mix without fibers (CMa),

20%

17,20%
18%

15,91%
16%
14,35%
13,32%
14%

12%
Frequency (%)

11,09%

10%

8%
7,15%
6,42%

6%
4,97%

4%
2,75% 2,02% 1,40%
1,35% 1,30%
2%
0,62%
0,16%
0%
< 5 6÷10 11÷15 16÷20 21÷25 26÷30 31÷35 36÷40 41÷45 46÷50 51÷55 56÷60 61÷65 66÷70 >70

Ranges of length (mm)

Fig. 3. Relative frequency of the length of the fibers.

Table 1
Mix-design.

Mix Portland CEM 32.5R II-A/LL Water Aggregate I (8–20 mm) Aggregate II (4–10 mm) Sand Fibres Plast.

[kg/m 3] [l/m 3] [kg/m 3] [kg/m 3] [kg/m 3] [%v] [kg/m 3] [%] [kg/m 3]

CMa 350 188 511.40 304.40 941.30 – – 0.64 2.20

RSF1a 350 188 509.50 303.30 937.90 0.23 15.90 1.17 4.10
RSF2a 350 188 507.80 302.30 934.70 0.46 31.75 1.29 4.56
ISFa 350 188 508.10 302.50 935.30 0.46 34.80 0.60 2.21
 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57
added to foster the dispersion of the fibers in the concrete mix limiting the technologic problem of their
aptitude to tangle.
Recovered fibers were added by hand to the mixes as the last component with the plasticizer; the
specimens were cast and cured according to the requirements of standards [28] . The mixer used was
a batch mixer with a vertical axis, a fixed tank and four rotating arms. At the end, the mix showed an
almost homogeneous fiber distribution as a consequence of an adequate friction generated during the
mixing process.
The specimens for compressive tests were cast using iron moulds with a 150 mm side, left to
cure for 24 h, then demoulded and placed in a curing room at approximatively 95% RH and 20 C until
testing. All the tests were performed after 28 days under ordinary laboratory conditions.
Compressive tests were executed, according to [28] , on three
cubic specimens for each mixture, by an oil-pressure machine (nominal capacity 3000 kN) after the
measurement of the geometric dimension of each specimen. Moreover, two slump tests were
executed on the fresh concrete for each mixture.
4.2. Flexural tests: specimens geometry and test set-up
It is well known that the addition of steel fibers results in beneficial enhancement of the tensile
properties of concrete: steel fibers modify the brittle concrete into a more ductile material, able to hold
up large deformations before losing its bearing capacity [12,14,29] .
The standard test [23] for determining the nominal flexural properties is usually done on a
notched beam to ensure a better cracking control, due to the a priori knowledge of the weakest cross
section.
The specimens for flexural tests were cast in timber moulds (150 mm 150 mm 600 mm) with a
steel plate fixed in the middle to make the notch. On the day after the casting the specimens were
accurately demoulded and placed in a curing room (95% RH and 20 C) until the day of testing.
Sixteen tests (four for each
mix, Table 1 ) were performed after a minimum of 28 days under ordinary laboratory conditions and
the results averaged.
According to the standard, the bending tests were carried out with a four-point loading beam
scheme, to evaluate the toughness of concrete specimens for each mix-design ( Fig. 5 ). In addition
three full-bridge resistive clip gauges were used for experimental measurements: one clip gauge was
fixed across the notch mouth and the others were glued on the two opposite faces of the beam at the
height of notch tip, ( Fig. 5 ). The former gauge monitored the CMOD (Crack Mouth Opening
Displacement), which is the control parameter for the test, while the other two instruments recorded
the CTOD (Crack Tip Opening Displacement). The
Fig. 5. Bending test set-up [23] .
49
specimens were tested with a 300 kN servo-hydraulic machine under CMOD control at a speed of 0.5
l m/min.
4.3. Flexural tests on slabs
4.3.1. Preliminary remarks
As the addition of fibers may drastically reduce the brittleness of concrete and improve its
mechanical properties and energyabsorption capability, the use of RSF in concrete slabs could
constitute one of the most interesting applications, on the basis of the results already found for
concrete slabs reinforced with ISF used as structural ceiling, pedestrian bridges and industrial floors.
In fact, the replacement of the traditional reinforcement with fibers in concrete slabs has been
addressed in some available research work
[17,19,30] . Ding and Kusterle [19] compared the mechanical behavior at early ages of steel fiber
reinforced concrete slabs and traditional steel mesh reinforced slabs, because in some applications it
is very important to reach high early strength. They concluded that the presence of the steel fibers
increased the ductility and changed the failure mode from punching shear to flexure. These relevant
results confirm the high potentiality of steel fibers in sustaining shear stresses.
In [17] sprayed concrete slabs were tested varying the type of reinforcements: steel mesh, steel
fibers and glass fabric. The authors tested also slabs realized with plain concrete as reference. Their
experimental results evidenced as the steel mesh solution furnished the best mechanical performance
while the glass fabric and steel fiber reinforcement exhibited comparable results.
Generally it can be evidenced that the presence of a distributed reinforcement within the slabs is
particularly advantageous when a complex state of stress may be expected. Further advantages
when using fibers instead of traditional steel reinforcement is the easier execution, which do not
require specialized operators and the higher quality due to the limitation of possible unforeseen
movement of traditional reinforcement during the casting operation.
4.3.2. Specimens and testing procedure
In order to analyze the RSFRC post-cracking performance, flexural tests on four concrete slabs
reinforced with 0.46% by volume of concrete of RSF were carried out, according to EFNARC
specification [24] . For comparison purposes, three samples reinforced with industrially produced
fibers were also considered.
The mix-design used, reported in Table 2 , was quite similar to
that realized for compressive and flexural tests, described in the previous sections. Globally seven
specimens were cast: four slabs for RSFRC mixture and three slabs for ISFRC mixture. Industrial
1: CTOD gauge 2:
CMOD gauge
50 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57

Table 2
Mix-design for slabs preparation.

No Portland CEM 32.5R II-A/LL Water Aggregate I (8–20 mm) Aggregate II (4–10 mm) Sand Fibres Plast.

[kg/m 3] [l/m 3] [kg/m 3] [kg/m 3] [kg/m 3] [%v] [kg/m 3] [%] [kg/m 3]

RSF 350 188 508.00 302.40 935.10 0.46 31.75 0.69 2.42
ISF 350 188 508.00 302.40 935.10 0.46 34.80 0.63 2.20

Steel fibers added in concrete mix were those before mentioned (diameter equal to 0.6 mm and 40 Table 3
mm length). The specimens were cast using appropriate moulds, wet inside with a release agent and Results of slump and compressive tests (UNI 12350, 2003; UNI 12390, 2003).

then placed on a vibration table to obtain a more homogeneous fibers distribution. On the day after
Mix Slump (mm) Class of consistency Compression strength
the casting, the specimens were accurately demoulded and then placed in a moist room, at
R c ( MPa) COV. (%)
approximately 20 C and 95% HR, until testing. All tests were performed after 28 days under ordinary
laboratory conditions by a servo-hydraulic machine with a nominal capacity of 50 tons. CMa 215 S5 31.63 0.99
RSF1a 220 S5 39.00 0.65
RSF2a 210 S5 39.68 1.61
ISFa 205 S4 29.01 2.85

The slabs were square with dimensions 600 mm 600 mm and thickness 100 mm; they were
of workability obtained for plain concrete, providing an adequate percentage of plasticizer. On the
simply supported on its four edges by a rigid steel frame with a span of 500 mm and tested under a
other hand when adding fibers maintaining the same quantity of plasticizer (see mixes CMa and ISFa)
centre point load through a contact surface of 100 mm 100 mm, as shown in Fig. 6 . All the tests were
the workability reduces. This observation is confirmed in other experimental works [32,33] , where a
executed under displacement control, at a gradient of 1.5 mm/min.
negative effect on the workability of the concrete mix was found with increasing the fiber percentage.

Three displacement transducers were fixed on each specimen ( Fig. 6 ). The first transducer was
applied at the middle of the slab to evaluate the mid-deflection. The two others measured the
The results of the compression tests are analyzed in term of compression strength ( R c) and
displacement on two opposite sides at a distance of 125 mm from the middle of the slab. To avoid the
corresponding coefficient of variation (COV), Table 3 . By the data, it can be observed that
failure of the specimen and possible damages of the test equipment, all the tests were stopped when
compressive strength for RSF reinforced specimens is higher than that found for the control mix.
a mid-displacement of about 25–30 mm was reached.
Similar consideration cannot be repeated in the case of ISF reinforced concrete, namely within the
experimental error the obtained compressive strength of control and reinforced mix is similar.

5. Experimental results and discussion

5.1. Workability and compressive strength
The experimental results of both slump and compression tests are listed in Table 3 . According to [31]
, the class of consistency for each average slump value was identified: it can be noted that an
opportune mix design of FRC is able to guarantee the same class
The good workability and the increase of compressive strength, due to the presence of RSF,
confirm results of previous studies performed by authors. It can be argued that the random
geometrical irregularity of the RSF contributes in average to limit the tensile transverse strain during
loading, delaying cracks propagation and, finally, increasing the compressive strength. On the basis
of the all results obtained by the authors, a compressive strength

Fig. 6. Apparatus for tests on slabs – particulars of the equipment.

 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57 51

Fig. 7. Compressive failure mode: (a) RSF-2 and (b) ISF.

Table 4
Results of the flexural tests.

No. CTOD 0 ( mm) f If ( MPa) f eq(0-0,6) ( MPa) f eq(0,6-3) ( MPa) D0 D1

CMa Mean value 17.1 3.35

St. dev. 1.40 0.37
COV. 8.2% 10.9%

RSF1a Mean value 17.1 3.44 2.59 1.99 0.77 0.77

St. dev. / 0.59 0.21 0.47 0.18 0.18
COV. / 17.2% 7.9% 23.6% 23.4% 23.8%

RSF2a Mean value 17.1 2.84 3.58 2.61 1.26 0.73

St. dev. / 0.29 0.81 0.59 0.20 0.08
COV. / 10.4% 22.8% 22.5% 15.6% 11.6%

ISFa Mean value 17.1 3.00 2.41 2.79 0.78 1.10

St. dev. / 0.41 1.37 1.84 0.37 0.18
COV. / 13.8% 56.7% 65.8% 47.4% 16.7%

equal to 1.21 f c is found when a 0.46 percentage by volume of RSF is added to the concrete mix, of the FRC specimens with respect to unreinforced ones; however it should be underlined that a more
being f c the average compressive strength of the plain concrete. accurate analysis should take into account the variation of the concrete mixes utilized.

In the available literature the influence of the presence of the fibers on the compressive strength
of the FRC does not show a unique trend. In fact, some experimental works on the mechanical
performance of industrial FRC [12,14,25] underlined that the compressive strength of SFRC is less
affected by the presence of fibers. On the contrary in [10,32,33] analyzing the effect of the steel fibers
on the mechanical properties of FRC, the compressive strength of the material seems to increase with
both the percentage of short fibers in concrete mix. The increment in compressive strength with
respect to plain concrete varies in the range 3.5–18.7% for the specimens tested in [33] , with the
highest value corresponding to specimens having an aspect ratio of fibers equal to 65 and 80 and a
1% fiber content by volume of concrete. The percentage increment of compressive strength falls in
the range 3–26% according to [32] , in this case the highest values were obtained when a fiber content
between 1% and 2% by volume of concrete and a aspect ratio of 40 were utilized. Finally in [10] a
maximum increment in compression strength of 24% was observed for ISFRC specimens with a fiber
content of 6% by mass and a fibers aspect ratio of 48.
Besides in [34] the incorporation of waste fibers (metallic and polymeric) into cement matrix to
improve the mechanical properties of cement based materials was experimentally investigated. The
results on the compression strength of concrete reinforced with waste metallic fibers show two
different behaviors: the addition of short fibers (<30 mm) causes a slight decrease of the concrete
compressive strength while the addition of long fibers (approximately 60 mm) did not affect the
compressive strength of the materials. The authors basically conclude that the presence of fiber have
no significant effect on the compressive strength of FRC as the observed reduction (in the cases of
short fibers) could be explained by a decrease of the coarse aggregates in the mix.
The increase of compressive strength observed in this work is similar to that reported in the
literature for concrete reinforced with industrial steel fiber. However a wider database is needed to
asses a more reliable trend specially referring to waste fibers.

In few experimental works the mechanical proprieties of RSFRC are analyzed. In [10] the flexural
behavior of concrete reinforced with recycled fiber recovered from waste tires was studied with
special care to the effect of different types of fibers. In particular, the authors tested inter alia two
groups of specimens obtained by using pyrolysed and shredded recycled fibers. Their available data
in terms of compression strength generally evidences a reduction
In Fig. 7 two tested specimens are shown: the former reinforced with 0.46% RSF ( Fig. 7 a), the
latter with 0.46% ISF ( Fig. 7 b). It can be noted that the failure modes are similar and consistent with
that indicated in [28] as satisfactory.
5.2. Post-cracking behavior of RFRC
The recorded experimental results of the flexural tests are elaborated in terms of the following
parameters [23] :
52 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57

CMa-1 RSF1a-1
CMa-2 RSF1a-2
20 CMa-3 20 RSF1a-3
CMa-4
18 18

16 16
Load (kN)

Load (kN)
14 14

12 12

10 10

68 68

24 24

0 0

0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 500 1000 1500 2000 2500 3000 3500 4000 4500

CTOD m ( µ m) CTOD m ( µ m)

RSF2a-1 ISFa-1
RSF2a-2 ISFa-2
20 RSF2a-3 20 ISFa-3
RSF2a-4 ISFa-4
18 18

16 16
Load (kN)

Load (kN)

14 14

12 12

10 10

68 68

24 24

0 0

0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 500 1000 1500 2000 2500 3000 3500 4000 4500

CTOD m ( µ m) CTOD m ( µ m)

Fig. 8. Load-CTOD curves for each tested mixture.

Basic Crack Tip Opening Displacement, CTOD 0: the value of the CTOD m corresponding to the Equivalent Strength, f eq(0.6–3): the equivalent strength calculated in the range of CTOD net between
conventional onset of the cracking process in the plain concrete (first crack), where CTOD m represents 0.6 and 3 mm. Ductility Index, D 0: ductility index evaluated in the range of CTOD net between 0 and
the mean value of the two CTOD values measured on the two opposite faces of the tested 0.6 mm: it is given by the ratio between the equivalent strength f eq(0–0.6) and the first crack
specimen ( Fig. 5 , section A–A 0). strength f If.

First Crack Strength, f If: the conventional flexural strength of the SFRC calculated on the basis of Ductility Index, D 1: ductility index evaluated in the range of CTOD net between 0.6 and 3 mm: it is
the load value corresponding to the CTOD 0. given by the ratio between the equivalent strength f eq(0.6–3) and f eq(0–0.6).

Equivalent Strength, f eq(0–0.6): the equivalent strength calculated in the range of CTOD net between
0–0.6 mm, where CTOD net represent the mean of the two CTOD values measured on the two The average values for all mentioned parameters together with the correspondent standard
opposite faces of the specimen starting from CTOD 0 value. deviation (St. Dev.) and the coefficient of variation (COV) are reported in Table 4 for all analyzed
concrete

Fig. 9. Comparison between the broken sections: (a) ISFa-3; (b) ISFa-4 and (c) RSF2a-3.
 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57 53

MC-2

MC-3
18
16 MC-4

CMa-1 16
14

Load (kN)
Load (kN)
CMa-2 14
12
CMa-3 12
10 CMa-4 RSF1a-3
10
RSF1a-2
68
68 RSF1a-1
RSF1-c
24 24
RSF1-b

0 0 RSF1-a
0 100 200 300 400 500 60 0 0 500 1000 1500 2000 2500 3000 350 0

CTOD m ( mm) CTOD m ( mm)

RSF2-1 RSF2-2

RSF2-3 RSF2-4
20 20
RSF2a-1 RSF2a-2
18 18
RSF2a-3 RSF2a-4
16 16
Load (kN)

Load (kN)
14 14

12 12

10 10

68 68
ISFa-3 ISFa-4
24 24 ISFa-1 ISFa-2
ISF-3 ISF-4
0 0 ISF-1 ISF-2
0 1000 2000 3000 400 0 0 1000 2000 3000 4000 5000

CTOD m ( mm) CTOD m ( mm)

Fig. 10. Comparison with experimental data reported in [3] .

mixes. Comparing the parameters of the FRC specimens respect to these of plain concrete it is
Table 5
possible to note the effect of improvement of the toughness of the material due to the presence of
Average value of the parameter K and Z.
fibers
Specimens K Z
[3,12,14] . In fact due to the well known fragile behavior of the plain concrete the two equivalent
strengths ( f eq(0–0.6) and f eq(0.6–3)) as well as the ductility indexes ( D 0 and D 1) was not calculated. In the CMa Average 6.65 1
Dev. st. 1.11
case of reinforced concrete, the presence of fibers prevents and controls the cracks propagation; as a
consequence the flexural behavior is characterized by a residual strength in the post-cracking stage. RSF1a Average 72.10 10.84
Dev. st. 39.84 5.99
However, from the analysis of both numerical data ( Table 4 ) and curves load–CTOD m for all tested
specimens ( Fig. 8 ) a great dispersion of results can also be observed. The scatter is more marked in RSF2a Average 201.93 30.37
Dev. st. 65.26 9.81
the case of specimens reinforced with commercial fibers in which two specimens (ISFa-1 and ISFa-2)
show a hardening behavior while the others present a softening behavior (ISFa-2 and ISFa-3). ISFa Average 197.87 29.76
Dev. st. 90.98 13.68
Perhaps, as the toughness behavior of the FRC materials is due to the bridge effect of the fiber within
the notched section, a different distribution of the fibers in that section could be the cause of the
registered scatter. To support this observation, in
50

45

40

35

Fig. 9 a comparison between two cracked section of two ISF different specimens is reported: it is 30
Average Z values

evident as the number of the fibers in the two specimens is different and, in addition, their distribution
25
for specimen ISFa-3 seems to be less uniform respect to that of the specimen ISFa-4. This different
20
behavior between similar specimens was obtained also by other author [35] testing under flexural
industrial steel fiber reinforced specimens. It is in the mind of the authors to do and suggest to 15

associate flexural tests with an accurate analysis of the rupture surface in terms of fibers number and 10

fibers distribution in order to furnish acceptance criteria of test results in relation to the concrete
5
homogeneity.
0
RSF (0,23%v) RSF (0,46%v) ISF (0,46%v)

Fig. 11. Z Value for reinforced specimens.

Analyzing the load versus CTOD m curves ( Fig. 8 ) the great effect
of the fibers in the concrete could be clearly observed. In fact, the un-reinforced concrete specimens
(CMa) exhibited the typical brittleness of plain concrete in tension, which makes this material
unsuitable for a reliable structural use without tensile reinforcement. The RSF1a reinforced specimens
(0.23% by volume of
recycled steel fibers) showed a typical softening behavior, with a low residual tensile stress after a
post-peak descending branch. The RSF2a reinforced specimens (0.46% by volume of recycled steel
fibers) had a relevant increase of the residual tensile stress due to
54 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57

(a) (b)
100 100

90 90 ISF_a
ISF_b
80 RSF_a 80
ISF_c
RSF_b
70 70
RSF_c
RSF_d
60 60
Load (kN)

Load (kN)
50 50

40 40

30 30

20 20

10 10

0 0

0 5 10 15 20 25 30 0 5 10 15 20 25 30

Mid deflection (mm) Mid deflection (mm)

Fig. 12. Applied load vs. mid-deflection for tested slabs: (a) RSFRC slabs and (b) ISFRC slabs.

Table 6
Experimental results.

No F max-s ( kN) S Fmax-m ( mm) S Fmax-l ( mm) S Fu-m ( mm) S Fu-l ( mm) T s ( kNmm)

RSF 61.5 3.52 2.01 25.7 12.80 430

(COV. 18%) (COV. 33%) (COV. 29%) (COV. 19%) (COV. 28%) (COV. 35%)

ISF 72.53 3.85 1.97 26.2 13.05 551

(COV. 20%) (COV. 31%) (COV. 27%) (COV. 12%) (COV. 11%) (COV. 35%)

80
the increased fibers percentage. Finally, the curves of ISF reinforced concrete specimens (0.46% by
volume of industrial steel fibers) show, as before mentioned, both softening and hardening behavior RSF
70
with an increase of the post-peak strength in the last case (ISFa-1 and ISFa-4 specimens). In ISF

addition, analyzing the curves referred to SFRC, several drops of load appear, due to the fibers
60

pull-out during the test and, therefore, to the progressive damage of the specimens.

50
Load (KN)

40

From the experimental data it is possible to confirm the satisfactory contribution of the RSF to
improve the brittle behavior of the plain concrete. 30

20
In Fig. 10 the comparison of the experimental results reported in this paper with previously
achieved data [3] in terms of load versus CTOD m curves is shown. The comparison evidenced as the
10
two experimental campaigns furnished the same trend of the load– CTOD m curve for each analyzed
mix design along with a general improvement in the flexural behavior registered for the specimen
0
analyzed in the present work. The optimization of the concrete mixes performed in the design phase
0 5 10 15 20 25 30
could influenced the mechanical proprieties, however to confirm this hypothesis a wider experimental
Mid-deflection (mm)
analysis result necessary in order to reduce the higher experimental error typical of this type of tests [29,35]
. Taking into account that the equivalent strength f eq(0–0.6) as well as f eq(0.6–3) depends on the areas Fig. 13. Applied load vs. mid-deflection: average curve for each mix.

underneath the load-CTOD m

curves, in the range 0–0.6 mm and 0.6–3 mm respectively, different parameters (in the following
called Z and K) were defined in order to furnish a more objective comparison between the two different
experimental campaigns.
Firstly, for each test the ratios between the whole area ( A t) and the area corresponding to the
conventional first crack ( A f) in the load versus CTOD curves were calculated: K ¼ A t
Z ¼ K FRC
K PLAIN. Results are reported in Table 5 . In this way, the experi-
mental results became dimensionless respect to the reference plain concrete; utilized thus more
general information on the RFRC in terms of energy absorption can be given, furnishing an indication
of the material toughness.
Fig. 11 shows the values of Z for concrete reinforced with both recycled and industrial fibers
collecting results of the present and previous work. Firstly it is possible to underline again as
increasing the fiber contents, the toughness of the materials increases as well. Comparing the RSF
(0.46%v) and ISF (0.46%v) specimens appears evident as the mechanical performance are quite

Af
. Then the corre- similar even if the industrial fibers are able to furnish a slightly high toughness respect to the
sponding mean K value was determined averaging the coefficient of similar specimens both for FRC specimens reinforced with recycled fibers. However the scatter of the experimental results is still high
and plain concrete, in the following called K FRC and K PLAIN, respectively. Successively, the ratio even if this level
between K FRC and the correspondent K PLAIN was also evaluated:
 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57 55

Fig. 14. RSFRC specimen at the end of the test.

of repeatability seems to be independent of the type of fibers. An extension of the experimental

campaign is necessary.

5.3. Flexural behavior of RFRC slabs

In Fig. 12 the curves applied load–mid deflection are drawn for RSFRC slabs ( Fig. 12 a) and
ISFRC slabs ( Fig. 12 b), respectively. It can be noted that all the tested specimens exhibited similar
trend: a linear branch until reaching the peak of load followed by a sudden decay of the load that
evidenced the typical softening behavior of these materials. In both cases (RSFRC and ISFRC slabs)
a relevant scatter was observed especially in the softening braches. This suggests again that the fiber
distribution in the cracked sections governs the post peak behavior. A detailed study of these sections
(in term of number and distribution of the fibers) should represent a key aspect in the future
development. As expected, the performance of ISFRC specimens was slightly better than that of
RSFRC specimens. In fact, for ISFRC slabs, it can be observed a lower slope of the curve at the
post-cracking stage and a higher average value of the maximum load.

Fig. 15. Zoom of the broken section of a RSFRC slab.

These results are confirmed by the experimental data reported in Table 6 and by the comparison
of the average curves for the two tested mixes, shown in Fig. 13 . In particular in Table 6 the average
numerical results in terms of maximum load ( F max-s), middle deflection at F max-s ( S Fmax-m), deflection at be considerate a good candidate to replace the industrial steel fiber.
125 mm from the middle of the slab at F max-s ( S Fmax-l), middle deflection at ultimate load ( S Fu-m), deflection
at 125 mm from the middle of the slab at ultimate load ( S Fu-l), and toughness ( T s) are reported. The S Fmax-l In Fig. 14 RSFRC specimen is shown after testing. It can be noted that the slab undergoes large
deflection before the failure, due to the effective contribution given by the RSF on the cracking
phenomenon.

and S Fu-l values referred to the average values of the two measurements while the T s value was
evaluate as the area under the curve load versus mid deflection.
However a good residual strength and wide deformations have been registered for both mixtures.
The total energy absorption was evaluated as measure of the toughness of the material in
accordance with [24] . The value obtained for ISF and RSF specimens are comparable confirming the
effectiveness of both types of fibers in improving the post-cracking behavior of concrete. Similar
considerations can be repeated referring to the maximum load and all recorded displacements. The
values reported in Table 6 are quite similar especially in terms of displacement, evidencing again the
effectiveness of the recycled steel fiber reinforcement that can
In Fig. 15 fibers at the cracked section of a RSFRC specimen can be observed: the RSF are not
broken but pulled-out evidencing the bridging contribution of fibers. Finally, in Fig. 16 all the
specimens at the end of the test are reported. It is important to underline as the failure mechanisms
for all the tested slabs is quite similar: bending failure occurs by 4, 5 or 6 cracks and the cracks
propagated from the center along medians or towards the corner regions. Similar mode of failure can
be found in [17] . In the literature are available data concerning tests on concrete panels (600 mm 600
mm 100 mm) reinforced with an industrial steel fibers dosage of 30 kg/m 3 ( 0.38% by volume). As
reported in Table 7 , the average values of mid-deflection and maximum load obtained by the authors
are similar to the results obtained by [21] .
56 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57

RSF-1 RSF-2

RSF-3 ISF-1

ISF-2 ISF-3

Fig. 16. Slabs at the end of the test.

Also the toughness values are reported in Table 7 : it can be observed that the scatter between
Table 7
RSFRC and ISFRC is not relevant; on the other hand the obtained values are lower than those
Experimental comparison.
reported by Minelli and Plizzari [20] . The interesting results achieved further confirm the promising
F max-s ( kN) S Fu-m ( mm) T s ( kNmm) application of RSF in concrete, even if a wider experimental investigation is suggested to get a
Obtained RSF 61.5 25–30 430 deeper knowledge of the mechanical behavior of RSF reinforced slabs.
Results ISF 72 551
Minelli et Plizzari [20] 50 30 847
Lambrechts [21] 80 30 –
 G. Centonze et al. / Construction and Building Materials 36 (2012) 46–57
6. Conclusions
In the present work the mechanical behavior of waste steel fiber reinforced concrete was
investigated in terms of workability, compressive and post-cracking behavior of hardened materials
and flexural properties of slabs. The experimental campaign was planned based on a previous works
of authors in order to gain a more in depth knowledge on the performance of RFRC.
The obtained results allow drawing the following concluding remarks:
an ‘‘ad hoc’’ design of the concrete mix could avoid the wellknow negative effect on the
workability of the concrete matrix due to the presence of the fibers; referring to the present work
a modification in the quantity of plasticizer appears enough to reach for the FRC matrix the same
slump class of the plain concrete;
as far as the contribution of fibers to compressive strength of concrete the addition of RSF, at
least referring to the performed tests, increases the compression strength with respect to plain
concrete. This result has been found in some literature works regarding the industrial steel fibers.
On the other hand the available literature on recycled steel fiber does not confirm the result found
here even if an accurate comparison is difficult because of the variability of the parameters
involved; the post-cracking mechanical behavior of the RSFRC underlined as the brittle behavior
of the concrete matrix could be avoided adding recycled steel fibers. Based on the experimental
data obtained by the authors on concrete specimens reinforced with recycled steel fiber it is
possible to note as doubling the RSF content the toughness of the materials increases more than
100%;
the positive effect of the recycled steel fiber on the toughness of the concrete matrix is similar to
that found for industrial steel fiber: the ductile indexes in the cases of specimens realized with
RSF ranging in 0.73–1.26 while the variation is 0.78–1.10 for that realized with ISF;
both type of materials (RSFRC and ISFRC) evidenced a significant scatter in the post-cracking
stage of the flexural tests often confirmed in the available literature: the different distribution of
the fibers in the broken section governs the behavior of the specimens. It is in the mind of the
authors to do and suggest to associate flexural tests with an accurate analysis if the rupture
surface in terms of fibers number and fiber distribution in order to furnish acceptance criteria of
test results in relation to the concrete homogeneity;
all tested slabs showed similar failure mechanisms: 4–6 flexural cracks that propagate from the
center along medians or towards the corner region. The experimental results in terms of load and
displacements for both RSFRC and ISFRC slabs are comparable confirming, again, the
effectiveness of the recycled fibers. In particular, the maximum load recorded for RSF slabs is
around 18% lower than that of ISF slabs, while the difference in displacements are smaller
ranging in 2% to +9%. However the high coefficient of variation recorded in these tests (12–33%)
evidences the need of a wider experimental analysis aiming to validate the presented
experimental campaign.
The interesting results obtained confirm the promising application of concrete reinforced with
RSF; however, further research work is still necessary, extending also the investigations to medium
and full-scale structural elements, representative of possible practical situations. Finally, theoretical
models are needed to furnish reliable design guidelines for concrete reinforced with steel fibers
recycled from waste tires (RSFRC).
57
Acknowledgements
The authors would like to acknowledge Irigom s.r.l. for supplying steel fibers recovered from
waste tyres and ItalcementiS.p.A. for the important support in the experimental work.
References
[1] European Commission. Council directive 1999/31/EC of 26 April 1999 on the
landfill of waste. Off J Eur Commun 1999;L182:1–19. [2] Decreto Legislativo 3 aprile 2006, no. 152 Norme in
materia ambientale. [3] Aiello MA, Leuzzi F, Centonze G, Maffezzoli A. Use of steel fibres recovered
fromwaste tyres as reinforcement in concrete: pull-out behavior, compressive and flexural strength. Waste
Manage 2009;29:1960–70. [4] Aiello MA, Leuzzi F. Waste tyre rubberized concrete: properties at fresh and
hardened state. Waste Manage 2010;30:1696–704. [5] Bignozzi MC, Sandrolini F. Tyre rubber waste recycling
in self-compacting
concrete. Cem Conc Res 2006;36:735–9.
[6] Fattuhi NI, Clark LA. Cement based materials containing shredded scrap truck
tyre rubber. Constr Build Mater 1996;10:229–36. [7] Neocleous K, Tlemat H, Pilakoutas K. Design issues for
concrete reinforced with
steel fibers. J Mater Civil Eng 2006;18:677–85. [8] Topçu IB, Avcular N. Analysis of rubberized concrete as a
composite material.
Cem Concr Res 1997;27:1135–9.
[9] Tlemat H, Pilakoutas K, Neocleous K. Design issues for concrete reinforced with
steel fibres recovered from waste tyres. J Mater Civil Eng 2006;18:677–85. [10] Tlemat H, Pilakoutas K,
Neocleous K. Stress strain characteristic of SFRC using
recycled fibre. Mater Struct 2006;39:365–77. [11] Achilleos C, Hadjimitsis D, Neocleous K, Pilakoutas K,
Neophytou PO, Kallis S.
Proportioning of steel fibre reinforced concrete mixes for pavement construction and their impact on
environment and cost. Sustainability 2011;3(7):965–83. http://dx.doi.org/10.3390/su3070965.
[12] Buratti N, Mazzotti C, Savoia M. Post-cracking behaviour of steel and macro-
synthetic fibre-reinforced concretes. Constr Build Mater 2011;25(5):2713–22. [13] di Prisco M, Felicetti R. On
fatigue of plain and fibre-reinforced concrete
ground slabs. In: 6th International RILEM symposium on fibre reinforced concretes, 2004. p. 1195–1206.
[14] Olivito RS, Zuccarello FA. An experimental study on the tensile strength of
steel fiber reinforced concrete. Compos Part B: Eng 2010;41(3):246–55. [15] Xu Z, Haoa H, Lib HN.
Experimental study of dynamic compressive properties
of fibre reinforced concrete material with different fibres. Mater Des 2012;33:42–55.
[16] Xu Z, Haoa H, Li HN. Mesoscale modelling of fibre reinforced concrete material
under compressive impact loading. Constr Build Mater 2012;26(1):274–88. [17] Colombo M, di Prisco M,
Mazzoleni L. Sprayed tunnel linings: a comparison
between several reinforcement solutions. Mater Struct 2009;42:1295–311. [18] Bernard ES. Correlations in the
behaviour of fibre reinforced shotcrete beam
and panel specimens. Mater Struct 2002;35:156–64. [19] Ding Y, Kusterle W. Comparative study of steel
fibre-reinforced concrete and
steel mesh-reinforced concrete at early ages in panel tests. Cem Concr Res 1999;29:1827–34.
[20] Minelli F, Plizzari GA. Round panel tests for characterizing steel fiber reinforced
concrete, Fib symposium Keep concrete attractive. Budapest 2005;1:310–5. [21] Lambrechts AN. The variation
of steel fibre concrete characteristics. Study on
toughness results 2002–2003, Fiber reinforced concrete from theory to practice.
In: International workshop on advances in fiber reinforced concrete, Bergamo (Italy); September
24–25, 2004. [22] Belletti B, Cerioni R, Meda A, Plizzari G. Design aspects on steel fiber-reinforced
concrete pavements. J Mater Civil Eng 2008;20(9):599–607. [23] UNI 11039. Steel fibre reinforced concrete.
Italy; 2003. [24] European Federation of Producers and Applicators of Specialist Products for
Structures. European specification for sprayed concrete (EFNARC); 1996. [25] CNR-DT 204/2006. Guide for
the design and construction of fiber-reinforced
concrete structures. Italy: Italian National Research Council; 2007. [26] UNI EN 12390. Testing fresh concrete.
Italy; 2003. [27] ACI 544.2R-89. Measurement of properties of fiber reinforced concrete; 1989. [28] UNI EN 12350.
Testing hardened concrete. Italy; 2003. [29] di Prisco M, Plizzari G, Vanderwalle L. Fibre reinforcement concrete: new
design prospective. Mater Struct 2009;42:1261–81. [30] Cengiz O, Turanli L. Comparative evaluation of steel
mesh, steel fibre and high-
performance polypropylene fibre reinforced shotcrete in panel test. Cem Concr Res 2004;34:1357–64.
[31] UNI EN 206-1. Concrete – Part 1: Specification, performance, production and
conformity. Italy; 2006.
[32] Mohammadi Y, SinghSP, KaushikSK. Propertiesof steel fibrousconcretecontaining
mixed fibres in fresh and hardened state. Constr Build Mater 2008;22(5):956–65. [33] Yazici S, Inan G, Tabak
V. Effect of aspect ratio and volume fraction of steel fiber
on the mechanical properties of SFRC. Constr Build Mater 2007;21:1250–3. [34] Meddah MS, Bencheikh M.
Proprieties of concrete reinforcement with
different kinds of industrial waste fibre materials. Constr Build Mater 2009;23(10):3196–205.
[35] Vasanelli E. Durability of concrete beams reinforced with short fibers and
traditional steel bars. PhD thesis, Italy: Dep. of Engineering for Innovation, University of Salento; 2011.