Professional Documents
Culture Documents
h i g h l i g h t s
a r t i c l e i n f o a b s t r a c t
Article history: The effect of freeze-thaw cycles (FTC) on the flexural strength of granite panels is investigated in this
Received 23 May 2017 paper. Specimens from three different types of Australian granite were sampled. Fifteen specimens were
Received in revised form 11 September used as control (unweathered) specimens while another 36 specimens were subjected to accelerated
2017
weathering consisting of 100 FTC over a period of 34 days. A controlled freeze-thaw chamber and a tem-
Accepted 11 September 2017
perature range between 10 °C and +70 °C was used for this purpose. Ultrasonic pulse velocity (UPV) of
the specimens was measured before and after FTC testing. Flexural strength tests were conducted after
the FTC were completed. Results showed that thin granite veneers suffered physical degradation accom-
Keywords:
Ultrasonic pulse velocity
panied by a noticeable reduction in UPV, density and flexural strength following FTC. Correlation between
Granite flexural strength and UPV and between unweathered and weathered UPV is proposed for the three types
Flexural tensile strength of Australian granite used in this paper.
Freeze-thaw cycles Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Over the past few decades, an increasing demand for aestheti-
cally pleasing, structurally adequate thin structural stone veneer
Failure of cladding panels can lead to expensive restoration as cladding panels on buildings, particularly high-rise, has been
works as well as serious risks from falling debris. Numerous inci- observed. Among thin structural cladding veneers, Granite is the
dents of stone cladding failures have been reported around the most commonly used (Pires et al. [2]). Being superior to other thin
world [1]. A notable example is the Amoco building in Chigaco that structural stone veneers in terms of strength and durability, gran-
cost around US $75 million to replace the façade stone work. In ite has received less research attention. In particular, a limited
Australia, stone cladding failure was reported in the Melbourne research is available on the effect of long-term freeze-thaw cycles
Metropolitan Board building, 210 George St building in Sydney (FTC) on the flexural strength of thin granite veneer [3]. It is gener-
and 344 Queen St building in Brisbane, among others. There is a ally accepted that weathering has the propensity to degrade the
notable increase in the number of new high-rise buildings around strength of building stones and thin cladding panels over time
Australia. With the recent safety concerns regarding aluminium and reduce their capacity to perform under seismic and cyclonic
panels cladding, it is expected that the use of thin granite veneers loadings during their service life [3,4].
as a façade solution on high-rise buildings will increase. Although destructive testing methods (such as four point bend-
ing test) are still preferred in conventional engineering practice,
⇑ Corresponding author. recent developments in non-destructive testing (NDT) methods
E-mail addresses: s.noorekhuda@cqu.edu.au (S. Noor-E-Khuda), f.albermani@
such as acoustic emission, rebound hammer, laser scanning and
cqu.edu.au (F. Albermani). ultrasonic pulse velocity (UPV) have facilitated detailed structural
http://dx.doi.org/10.1016/j.conbuildmat.2017.09.049
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
892 S. Noor-E-Khuda et al. / Construction and Building Materials 156 (2017) 891–901
investigation and retrofitting of modern structures [5]. Several Engineering laboratory of the University of Queensland. Details of
researchers adopted NDT techniques in micro level material charac- the experimental program are discussed in this section.
terisation and damage detection [5,6]. The UPV method has been
successfully employed for assessing the mechanical behaviour of
2.1. Material and test specimens’ description
concrete [7], masonry [6] and various thin structural stone veneers
[8,9]. High frequency sound waves above 20 kHz are referred to as
Three common Australian granite types were considered in this
ultrasound. These waves are transmitted through the material
study. These are; Calca, Imperial Black (or Adelaide Black) and
under test and the velocity of the wave passing through is termed
Grandee. Calca and Imperial Black stones are quarried in the Eyre
as the ultrasonic pulse velocity. The measured UPV is a function of
Peninsula and Black Hill regions of South Australia while Grandee
the inherited mineral composition, material properties and defects.
in the Mulliandry district of central New South Wales. These three
Vasconcelos et al. [10] used UPV to assess the strength, stiffness and
stone types are characterised by high compressive strength, hard-
fracture energy of granite. Furthermore, Chen et al. [11] successfully
ness, durability and visible crystalline structure with interlocking
estimated crack depth and direction and determined the thickness
crystals of medium to coarse grain. The three stones are distin-
of damaged surface layers using UPV. This paper refers to work con-
guishable by their appearances as shown in Fig. 1; with Calca hav-
ducted on the structural properties evaluation of various granite
ing pink/red clusters on black appearance, Imperial Black with dark
specimens [9–11] using the ultrasonic NDT method.
black appearance and Grandee with dark grey appearance. The
It is well known that thin structural stone veneers possess low
specimens were prepared from 32 mm granite slabs using laser
tensile strength, usually <10% of its compressive counterpart.
guided saw.
While more research is focused on the compressive strength of thin
Sample dimensions were selected based on ASTM C880/C880M
structural stone veneers [3,11], the flexural tensile behaviour has
specifications [21] and were approximately L400 W100 t
received fewer attention. The direct tensile and compressive
32 mm, with the exact dimension of each sample being specified
strength of granite specimens were reported in [10] from series
in Table 2. A total of 36 specimens (12 specimens of each stone
of destructive/non-destructive tests. A correlation between the
type, S1–S12, Table 2) were subjected to freeze-thaw cycling
strength and UPV was established [10] which showed that the
before conducting flexural test. Another 15 control specimens (5
direct tensile strength increased exponentially while the compres-
sive strength increased linearly with UPV. A number of studies
have reported on the influence of weathering due to salt crystalli-
sation [8], wetting/flooding [6,12] and heating [13] on the mechan-
ical properties of thin structural stone veneers. Franzoni et al. [13]
tested several stones including limestone and sandstone under
cycling heating between 100 400 °C, showed that the mechanical
properties of limestone degraded with temperature, while the
sandstone specimens showed the opposite characteristics, which
triggered the need to investigate the influence of FTC on the mate-
rial characteristics of granite. While FTC testing procedure has
been standardized, the guidelines provided in ASTM D5312/
D5312M-12 [14] and EN 12371 [15] reflect different climates and
practices and hence vary in-terms of the number of applied cycles,
heating/cooling duration and the required temperature range. In-
fact FTC for different rocks has been simulated using 50, 200 and
1400 FTC cycles in [16,3,17], respectively. Furthermore, various
temperature ranges were adopted; for example [3,18,19,20] used
temperatures range of (40 °C to 40 °C), (40 °C to 180 °C),
(20 °C to 20 °C) and (7 °C to 14 °C), respectively. Most of the
Australian cities experience a wide range of temperature from
above +50 °C in summer to 10 °C in winter. Furthermore, a sharp
temperature difference between daytime and nighttime tempera-
tures is common within the 24hrs cycle. For these reasons, a tem-
perature range of 10 °C to +70 °C is selected for the accelerated
FTC used in this work.
The effects of freeze-thaw cycling on the flexural strength of a
series of granite cladding panels using UPV and four point bending
tests are investigated in this paper. Specimens from three different
types of Australian granite, namely; Calca, Imperial Black and
Grandee were considered in this study. The specimens were par-
tially immersed in water and placed in a controlled freeze-thaw
chamber to simulate accelerated weathering through FTC. Based
on test results, correlations between UPV of weathered and
unweathered specimens and between UPV and flexural strength
are established.
2. Experimental program
The pulse velocity test set-up depends on: the natural reso- Granite Petrographic Description
nance frequency of the used transducers (f), the minimum dimen- Type
sion of the test specimen (t) and the average grain size of the rock Calca Coarse grained granite of light reddish color, high orthoclase
used (d). The thickness (t) of the test specimens was checked feldspar and quartz content, and contains minor weakened
against the UPV test guidelines of ASTM D2845-05 [22] using Eq. cleavage fractures; felsic rock
Imperial Dark black color Monzo gabbro with high plagioclase feldspar
(1) Black content, mafic rock, hard, holocrystalline and with interlocking
t P 5ðv =f Þ P 5d ð1Þ crystals of medium to coarse grain
Grandee Dark colour quartz Monzo gabbro, high plagioclase feldspar and
Since the pulse propagation velocity ðv Þ and the grain size (d) pyroxene, and low quartz content, hard, holocrystalline and
with interlocking crystals of medium grain size, mafic rock
are intrinsic material properties and the coupon’s nominal thick-
ness, t = 32 mm, was set to meet the flexural test requirements; tri-
als were used to select suitable pulse frequencies for each test run.
Furthermore, since veneer claddings are commercially produced in Fig. 3 (a) shows the FTC during one day period while Fig. 3(b)
the form of thin slabs, it is expected that freeze-thaw cycling is shows FTC over the duration of the test (34 days). Each of the pos-
likely to have more impact on their mechanical properties. This itive/negative temperature peaks were maintained for at least half
negates the use of thicker specimens in the experimental program. an hour throughout the freeze-thaw process. The trays were toped
Moreover, similar sized granite specimens used in direct tensile up with water on a daily basis to compensate for evaporation
and UPV testing were reported to give acceptable results [10]. Pet- losses and to maintain the polished face of the specimens under
rographic description of the three types of granite used in this water throughout the test. This is usually done during the heating
study is summarized in Table 1. cycle and when the temperature within the chamber is close to
ambient air temperature. The use of 100 FTC deemed adequate
2.2. Accelerated Freeze-Thaw cycles (FTC) to cause notable degradation in the mechanical properties of build-
ing stones [3,16,19]. A calibrated ohm reader placed inside the
FTC were performed considering the ASTM D5312/D5312M-12 freeze-thaw chamber was used to continuously record the temper-
and EN 12371 [14,15] standard specifications. All the 36 granite ature inside the chamber throughout the test.
specimens were subjected to 100 freeze-thaw cycles over a period Fig. 2(e) shows a granite specimen after the completion of FTC
of 34 days. test. It is apparent that the specimen underwent physical degrada-
The accelerated FTC were conducted using a large freeze-thaw tion with salt crystallization (efflorescence) as white powdery
chamber shown in Fig. 2(a). A control panel next to the chamber deposits could be seen all over the specimen. Moreover, visible fine
is used to set the temperature range (10 °C to +70 °C) and freez- cracks on the surface of the specimens were observed as evidence
ing/heating durations. In each of the four trays nine specimens of differential thermal expansion and contraction and frost action
were placed; three each of Calca, Imperial Black and Grandee, during FTC.
shown in Fig. 2(b), (c) and (d), respectively. Water is poured in
the tray such that the bottom face of the specimens (the polished 2.3. Ultrasonic pulse velocity (UPV) testing
face) is immersed in water. The trays were designed to support
the specimens above the base of the tray to ensure adequate satu- UPV measurement of all granite specimens before and after the
ration of the polished face of the specimens. FTC testing was conducted using a transient wave test method. The
The specimens were subjected to eight hours freeze-thaw test instrumentation and UPV measurements along the width and
cycles. Each cycle comprising five hours of freezing and three hours length of the specimen are shown in Fig. 4. The test setup com-
of heating between 10 °C and +70 °C. Although the temperature prised of a Tektronix TDS210 LCD oscilloscope, a Panametrics
range is different from ASTM D5312/D5312M-12 [14] they were 5077 PR pulser-receiver and two contact transducers. The UPV
considered an acceptable range that reflected Australian climate. was obtained by direct transmission, by placing the transducers
Table 2
Specimens used in the UPV and flexural testing program.
Time (Hours)
0 2 4 6 8 10 12 14 16 18 20 22 24
80.00
70.00 30mins
Measured
peak
60.00
8hours cycle
Te mpe rature o C
50.00 Simulated
40.00
30.00
5hours
20.00 freezing
10.00
0.00
-10.00 3hours
heating 30mins peak
-20.00
31/05 0:00 31/05 9:36 31/05 19:12
Time (Date/Hours/Mins)
(a)
100.00
80.00
60.00
Te mpe rature o C
40.00
20.00
0.00
-20.00
-40.00
8/05 15/05 22/05 29/05 5/06 12/06
Time (Date)
(b)
Fig. 3. Temperature cycles in the freeze/thaw chamber (a) Cycles in one day; (b)
Cycles over the testing period.
Ge l O scilloscope
Pulser/Receiver
Transducers 80mm 160mm 80mm
Dwave 3WLb
ft ¼ 2
ð2bÞ
4l t 2
where, W is the maximum applied load on the specimen, b and t are
the width and thickness dimensions, respectively and l is the actual
length of the inclined failure plane across the tension face. The
actual dimensions of the specimens (Table 2) rather than the nom-
inal dimensions were used in calculating the flexural strength (Eq.
(2b)). When the failure plane is perpendicular to the span, Eq.
(b) (2b) gives the same results as Eq. (2a). However, for inclined failure
plane, the flexural strength from Eq. (2b) is up to 4% lower than that
from Eq. (2a).
Fig. 6 shows the UPV values for the Calca (Fig. 6a), Imperial
Black (Fig. 6b) and Grandee (Fig. 6c) specimens before FTC
(unweathered state). Three individual readings along the width
of each specimen are shown in Fig. 6.
The results shown in Fig. 6 are consistent and fall within Class 5
of UPV classification [23] (Table 3) for the three types of granite
used in this study. The average UPV of unweathered Imperial Black,
(c) Grandee and Calca specimens were 6389 m/s, 5704 m/s and
5166 m/s, respectively.
Fig. 4. Ultrasound testing of granite specimens (a) Test arrangement; (b) Measure-
Fig. 7 shows the average UPV values of the unweathered granite
ment along width; (c) Measurement along length.
specimens measured along the width and the length of the speci-
men. It was found that the UPV measurements along the length
of specimen were generally higher than those along the width.
used. The load-deflection response was recorded from the load cell The variation in UPV values along the length and width directions
and the crosshead displacement data. Length and orientation of the corresponds to planar anisotropy of the granite material [22,23].
failure plane and the average thickness at the failure plane were The difference is more consistent for Imperial Black and Grandee
measured at the completion of each test. than that for Calca.
896 S. Noor-E-Khuda et al. / Construction and Building Materials 156 (2017) 891–901
5500 Table 3
Classifications of UPV [23].
5400 Class UPV (m/s) Definition
5000
Table 4 presents the average UPV of the unweathered and
4900
weathered specimens measured along the length and the width
Reading 1
4800 of the specimen. Also shown in this table is the percentage reduc-
Reading 2 tion of pulse velocity due to weathering. The results show that
4700 Imperial Black was the worst affected by FTC with an average
Reading 3
reduction in UPV of 17% and 21% in the width and length direc-
4600
0 1 2 3 4 5 6 7 8 9 10 11 12 tions, respectively. Percent reduction of UPV for the Calca and
Specimen Grandee specimens were 12–13% and 8–12%, respectively.
(a)
3.2. Flexural strength
6550
Figs. 9 (Calca), 10 (Imperial Black) and 11 (Grandee) show the
6500 load-deflection curves obtained from the 4-point bending test of
all 36 weathered and 15 unweathered (control) specimens. The
6450 average load resisted by Calca, Imperial Black, and Grandee control
specimens were 4.36 ± 0.5 kN, 6.45 ± 0.58 kN and 6.36 ± 0.21 kN,
and the same of the weathered specimens were 3.38 ± 0.54 kN,
UPV [m/s]
6400
5.14 ± 0.38 kN and 4.49 ± 0.4 kN, respectively. A notable decrease
in the load capacity of all the weathered granite specimens is
6350 observed that indicates degradation of the structural properties
due to FTC. It is evident that Imperial Black is the strongest among
6300 Reading 1 the three granite types considered in this paper. This is in-line with
Reading 2 the FTC test observations discussed in Section 2.3.
6250 As expected, the obtained load-deflection curves confirm the
Reading 3
brittle nature of granite material; the load increased linearly until
6200 the ultimate capacity followed by a sudden sharp drop correspond-
0 1 2 3 4 5 6 7 8 9 10 11 12 ing to failure. These plots will be particularly useful in modelling
Specimen and validating the flexural response of weathered and unweath-
(b) ered granite, and veneer walls comprising of granite claddings.
Fig. 12 shows the average flexural tensile strength, ft, of the
6200 unweathered and weathered granite specimens. The results shown
in Fig. 12 demonstrate that ft of the unweathered (control set)
6000 Calca, Imperial Black and Grandee granite specimens is
10.25 MPa, 16.77 MPa and 15.98 MPa; respectively, and that of
the same weathered specimens is 7.74 MPa, 11.98 MPa and
5800 13.32 MPa, respectively, corresponding to a reduction in flexural
UPV [m/s]
5600 5600
5500 5400
5400 5200
5300 5000
UPV [m/s]
UPV [m/s]
5200 4800
5100 4600
5000 4400
4900 4200
Unweathered
4800 Width 4000
Weathered
Length
4700 3800
0 1 2 3 4 5 6 7 8 9 10 11 12 0 2 4 6 8 10 12
Specimen Specimen
(a) (a)
7000 6800
6800 6300
6600
UPV [m/s] 5800
UPV [m/s]
6400 5300
6200 4800
6200
5800
6000
UPV [m/s]
5300
UPV [m/s]
5800
5600 4800
5400
4300
5200 Width Unweathered
Length Weathered
5000 3800
0 1 2 3 4 5 6 7 8 9 10 11 12 0 1 2 3 4 5 6 7 8 9 10 11 12
Specimen Specimen
(c) (c)
Fig. 7. Variation of UPV along the length and width of unweathered specimens (a) Fig. 8. Variation of UPV due to weathering (a) Calca; (b) Imperial Black; (c)
Calca; (b) Imperial Black; (c) Grandee. Grandee.
than other thin structural stone veneers reported in the literature (Table 4) and flexural strength obtained in the experimental
[9,16,18]. The mean density of unweathered Calca, Imperial Black program.
and Grandee specimens is 2644.97 kg/m3, 2980.58 kg/m3 and Furthermore, loss in density of the granite specimens due to FTC
2912.94 kg/m3, respectively. This explains the high UPV values is apparent in Fig. 13; Calca, Imperial Black and Grandee specimens
898 S. Noor-E-Khuda et al. / Construction and Building Materials 156 (2017) 891–901
Table 4
Average UPV of unweathered and weathered specimens and % loss.
6 9
C1 C2 C3 C1 C2 C3
C4 C5 8 C4 C5
5
7
4 6
Load (kN)
Load (kN)
5
3
4
2
3
2
1
1
0
0 0.1 0.2 0.3 0.4 0.5 0
Deflection (mm) 0.00000 0.10000 0.20000 0.30000 0.40000 0.50000 0.60000
Deflection (mm)
(a)
(a)
6 7
S1 S2 S3 S4 S1 S2 S3 S4
S5 S6 S7 S8 S6 S7 S8 S9
5 6
S9 S10 S11 S12 S10 S11 S12
5
4
Load (kN)
Load (kN)
4
3
3
2
2
1
1
0
0 0.1 0.2 0.3 0.4 0.5 0
0 0.1 0.2 0.3 0.4 0.5 0.6
Deflection (mm)
(b) Deflection (mm)
(b)
Fig. 9. Load vs deflection plot of Calca granite (a) Control unweathered specimens;
(b) Weathered specimens. Fig. 10. Load vs deflection plot of Imperial Black granite (a) Control unweathered
specimens; (b) Weathered specimens.
8 3100 7000
Unweathered
C1 C2 C3 Weathered
3000
7 C4 C5 UPV 6500
2900
De nsitymean (kg/m3 )
6 6000
UPVmean [m/s]
2800
5 2700 5500
Load (kN)
2600
4 5000
2500
3 4500
2400
2 2300 4000
Calca Imperial Grandee
Black
1
Fig. 13. Density of weathered and unweathered specimens.
0
0 0.1 0.2 0.3 0.4 0.5 UPV max UPV min
Deflection (mm) Anisotropy% ¼ 100 ð3Þ
UPV max
(a) where, UPVmax and UPVmin represents the maximum and the mini-
mum recorded UPV values irrespective of the measurement direc-
8
tion. Planar anisotropy of granite can be attributed to the
S1 S2 S3 S4
alignment of microcracks with the rift plane [10], resulting in low
7 S5 S6 S7 S10 to moderate ansiotropic behaviour. From Fig. 14, it is obvious that
S11 S12 the Calca specimens showed higher anisotropy among the three
6 granite types both in the unweathered and weathered state. Aniso-
tropy of Imperial Black and Grandee specimens were comparatively
low, within 4%. Notable changes to anisotropy of the weathered
5
Calca and Imperial Black specimens was observed, which indicates
Load (kN)
20 Unweathered
Fle xural Strength, ft (MPa)
18 Weathered 7
Unweathered
16 6 Weathered
14
Anisotropy (%)
5
12
10 4
8 3
6
2
4
2 1
0 0
Calca Imperial Grandee Calca Imperial Grandee
Black Black
Fig. 12. Flexural strength of granite specimens (unweathered and weathered). Fig. 14. Anisotropy of granite specimens.
900 S. Noor-E-Khuda et al. / Construction and Building Materials 156 (2017) 891–901
6500 19
UPVweathered = 0.7438UPVunweathered + 778.04
6000 R² = 0.67 17
5000
4500 13
4000 11
3500
9 ft = 0.0041UPVmean - 9.6345
3000 R² = 0.72
7
2500
2000 5
4500 4700 4900 5100 5300 5500 5700 5900 6100 6300 6500 4000 4500 5000 5500 6000 6500 7000
UPVunweathred [m/s] UPVmean [m/s]
(a)
Fig. 16. Correlation between UPV and flexural strength.
8000
7000 rc (Eq. (7)). A range of UPV between 3000 m/s to 6500 m/s was
selected to represent various types of old and quarried granites,
6000
reported in the literature [4,9–11].
UPVLength [m/s]
5000
rt ¼ 0:701eð0:00052UPVÞ ð6Þ
4000 y = 0.9876x + 177.12
R² = 0.802 rc ¼ 0:0407UPV 36:31 ð7Þ
3000
Fig. 17(a) shows the predicted tensile strength (ft and rt ) using
2000 Eqs. (5) and (6) against the corresponding UPV values. As seen in
Fig. 17a, rt is less than ft for UPV below 6000 m/s. This is expected
1000 assuming evenly distributed defect density and uniform stress dis-
tribution under direct tension. However, for higher UPV values, Eq.
0 (6) seems to over-predict the direct tensile strength compared to
4000 4500 5000 5500 6000 6500 7000 flexural strength.
UPVWidth [m/s] Fig. 17(b) shows the ratio of flexural strength (Eq. (5)) to com-
(b) pressive strength (Eq. (7)) of granite plotted against the corre-
sponding UPV values. As expected, the flexural tensile strength of
Fig. 15. Linear correlation of UPV; (a) UPV in unweathered and weathered granite is under 10% of the corresponding compressive strength.
conditions; (b) UPV along the length and width of the specimens. Based on Fig. 17a and b, the compressive strength for Calca, Impe-
rial Black and Grandee specimens is calculated as 156 MPa,
185 MPa and 171 MPa, respectively. This is in reasonable agree-
Linear correlation between UPV values measured along the
ment with results obtained from compression test of
length and width dimension of the granite specimens is shown
50 50 50 mm cubes of Calca and Imperial Black granite which
in Fig. 15(b) and given by Eq. (4b) with R2 = 0.802. From practical
was reported as 148 and 192 MPa [1].
point of view, the developed relation can be particularly useful
where there is a limited access to take full measurements of
installed granite panels. 4. Conclusion
25 Acknowledgements
15 References
[1] R. Blackwell, No Stone Unturned: Towards the Design of Thin Structural Stone
Veneers [PhD Thesis], University of Queensland, 2005.
10 [2] V. Pires, L.G. Rosa, A. Dionisio, Implications of exposure to high temperatures
for stone cladding requirements of three Portuguese granites regarding the use
of dowel-hole anchoring systems, Constr. Build. Mater. 64 (2014) 440–450.
5 ft [3] X. Tan, W. Chen, J. Yang, J. Cao, Laboratory investigations on the mechanical
properties degradation of granite under freeze-thaw cycles, Cold Reg. Sci.
σt Technol. 68 (2011) 130–138.
[4] M. Uranjek, V. Bokan-Bosiljkov, Influence of freeze-thaw cycles on mechanical
properties of historical brick masonry, Constr. Build. Mater. 84 (2015) 416–
0
428.
2000 3000 4000 5000 6000 7000 [5] I.S. Colombo, I. Main, M. Forde, Assessing damage of reinforced concrete beam
UPV [m/s] using ‘‘b-value” analysis of acoustic emission signals, J. Mater. Civil Eng.
(ASCE). 15 (2003) 280–286.
(a) [6] G. Cardani, L. Cantini, S. Munda, L. Zanzi, L. Binda, Non invasive measurements
of moisture in full-scale stone and brick masonry models after simulated
8 flooding: effectiveness of GPR, in: Nondestructive Testing of Materials and
Structures, Springer, 2013, pp. 1143–1149.
[7] G. Trtnik, F. Kavčič, G. Turk, Prediction of concrete strength using ultrasonic
7 pulse velocity and artificial neural networks, Ultrasonics 49 (2009) 53–60.
[8] G. Barbera, G. Barone, P. Mazzoleni, A. Scandurra, Laboratory measurement of
6 ultrasound velocity during accelerated aging tests: Implication for the
determination of limestone durability, Constr. Build. Mater. 36 (2012) 977–
983.
5 [9] C. Cerrillo, A. Jiménez, M. Rufo, J. Paniagua, F. Pachón, New contributions to
ft /σc (% )