Documente Academic
Documente Profesional
Documente Cultură
h i g h l i g h t s
Concrete of similar compressive strengths has different UPV depending on elastic modulus.
UPV can measure expansion of crack width in concrete by cooling.
UPV can decreases sharply during heating when thermal strain exceeds static fracture strain.
Proposed method is anticipated to allow integrity evaluation of concrete during heating.
a r t i c l e i n f o a b s t r a c t
Article history: As the mechanical properties of concrete subjected to high-temperature heating decrease due to contin-
Received 29 November 2017 uous degradation, it is necessary to evaluate the integrity of concrete in real time. Hence, the degradation
Received in revised form 6 March 2018 monitoring (continuous integrity evaluation) of concrete subjected to high-temperature heating was
Accepted 21 March 2018
examined by employing ultrasonic pulse velocity. In this study, the mechanical properties and the ultra-
sonic pulse velocity in concrete subjected to high-temperature heating were evaluated for ordinary-
strength to ultra-high-strength concrete. To measure the ultrasonic pulse velocity in concrete during
Keywords:
high-temperature heating, transducers were contacted to the top and bottom surfaces of the concrete
Ultrasonic pulse velocity
Concrete
specimens using SUS bars. This enabled monitoring of concrete degradation due to heating. In a range
Cracks of 200–300 °C, it was confirmed that the thermal strain coincides with the static fracture strain at room
Degradation evaluation temperature. The cracks generated during heating and the expansion of the cracks after cooling were
High temperature clearly confirmed by continuous measurement of the ultrasonic pulse velocity. The cracks generated at
Heating condition temperatures up to 300 °C had little effect on the decrease in compressive strength. However, it was con-
firmed that the elastic modulus continuously decreased, because the cracks generated during heating and
expanded after cooling led to strain increase at the peak stresses. Therefore, the evaluation method pro-
posed in this study is anticipated to allow the integrity evaluation of concrete at high temperature
heating.
Ó 2018 Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.conbuildmat.2018.03.178
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
512 E. Hwang et al. / Construction and Building Materials 171 (2018) 511–520
erties and the integrity of concrete exposed to fire condition. In fck Heating method Heating Test item
recent years, ultrasonic pulse characteristics such as propagation temp.
time and amplitude have been used to evaluate the degradation (°C)
of concrete. Shah [11,12] studied the harmonic wave generation 30 Heating to target temp. 100 Stress–strain
on loading in concrete specimens of standard size and of 150 mm at 1 °C/min 200 Residual compressive
70 300 strength
cubic size. Moreover, [13] investigated cracks in concrete struc- 400 Residual elastic modulus
tures using the interaction of ultrasonic pulse. In these studies, 110 500 Thermal strain
the ultrasonic pulse velocity and the mechanical properties of con- 600 Water vapor pressure
crete were evaluated after concrete degradation has already pro- 700 Permeability
Ultrasonic pulse velocity
gressed. However, since concrete is being continuously degraded
(before, during, after
during high-temperature heating, it is necessary to evaluate the heating)
integrity of concrete during the heating. Also, in order to clarify Temp. maintaining after 100 Ultrasonic pulse velocity
the correlation between the ultrasonic pulse velocity and mechan- heating to target temp. 200 during heating
ical properties of concrete, it is necessary to evaluate it both during at 1 °C/min 300
Table 1 presents the experimental plan of this study. The con- 2.3. Preparation of specimens
crete specimens with design compressive strengths of 30, 70, and
110 MPa were heated to target temperatures of 100, 200, 300, Concrete cylindrical specimens of a 100-mm diameter and
400, 500, 600, and 700 °C at a heating rate of 1 °C/min. The 200 mm height were fabricated for high-temperature heating in
stress–strain relationship, residual compressive strength, residual accordance with ASTM C192-Making and Curing Concrete Test
elastic modulus, thermal strain, water vapor pressure, permeabil- Specimens in the Laboratory. The specimens were cured in water
ity of the concrete specimens, and the ultrasonic pulse velocity in for seven days after demolding, and subsequently, cured in a
Table 2
Mix proportion of concrete.
fck W/B (%) Slump-flow (mm) Air (%) S/a2 (%) Unit weight3 (kg/m3)
W C SF BFS S G
30 55.0 1801 4 45 185 336 0 0 797 956
70 33.0 650 ± 50 2 165 475 25 0 755 905
110 19.0 160 589 126 126 618 741
1
Slump (mm).
2
S/a: the ratio of the fine aggregate volume to the total aggregate volume.
3
W: Water, C: Cement, SF: Silica fume, BFS: Blast furnace slag powder, S: Fine aggregate, G: Coarse aggregate.
Table 3 data logger. The installed electric furnace was used to heat the con-
Physical properties of used material. crete specimens. A heat transfer heating method, wherein the heat
Material Physical properties is transferred by heating the force jigs located at the top and bot-
Cement OPC (density: 3.15 g/cm3, specific surface area: 3200
tom of the furnace, was employed. The temperature of the electric
cm2/g) furnace was controlled through the temperature data of the spec-
Blast furnace slag Density: 2.97 g/cm3, specific surface area: 6,000–7000 imen generated by a thermocouple. A heating rate of 1 °C/min was
powder cm2/g employed [14] to maintain the temperature difference between
Silica fume Density: 2.57 g/cm3, specific surface area: 200,000 cm2/g
the interior and outer surface of the specimen within 5 °C, as
Fine aggregate Sea sand (density: 2.65 g/cm3, absorption: 1.00%)
Coarse aggregate Crushed granitic aggregate (size: 20 mm, density: 2.62 shown in Fig. 3 [2,15].
g/cm3, absorption: 0.8%)
Super plasticizer Polycarboxylic-based super plasticizer
2.4.2. Evaluation of thermal strain and mechanical properties
The thermal strain of the concrete specimens during high-
temperature heating was calculated using Eq. (1), after measuring
constant temperature and humidity chamber at a temperature in a the strain of the heating jig using the upper LVDT, and the strains
range of 20 ± 3 °C and relative humidity (RH) in a range of 50 ± 5% of the jig and specimen simultaneously using the lower LVDT, as
up to the material age of 300 days. In addition, since the top part of shown in Fig. 2(b).
the mold for manufacturing concrete specimens is open, the sur-
face of poured concrete is unleveled. Therefore, a concrete grinder DLC ¼ DL2 DL1 ð1Þ
must be used to even out the top and bottom surfaces of the con-
crete specimen before conducting the heating experiment. where
shaped refractory products-determination of permeability to gases surface connecting the SUS bar and the transducers was continu-
[17]. Permeability was calculated using Eq. (2). ously cooled to maintain a temperature of 60 °C or lower.
Furthermore, to investigate changes in the ultrasonic pulse
2P2 hc . transmitted through the SUS bar as a result of heating, the ultra-
K¼ ð2Þ
P21 P22 A sonic pulse velocity was measured in the SUS bar subjected to
high-temperature heating, as shown in Fig. 6. The ultrasonic pulse
where velocity of the SUS bar remained unchanged for a heating temper-
ature range up to approximately 380 °C. Moreover, the ultrasonic
K : Permeability (cm/s) pulse velocity of the SUS bar cooled after heating at a temperature
P1: Loading pressure (MPa) of 700 °C was confirmed to match the ultrasonic pulse velocity of
P2: Atmospheric pressure (MPa) the SUS bar before the heating. Hence, the high temperature heat-
h: Thickness of the concrete specimen (cm) ing did not affect the ultrasonic pulse transmitted through the SUS
. : Quantity of airflow (cm3/s) bar.
A: Area of airflow (cm2) The ultrasonic pulse velocity was measured using a PUNDIT
c : Weight of unit volume of gas (air: 1.205 106 kgf/cm3) comprised of a pair of transducers with a nominal frequency of
54 kHz and a precision of 0.1 ls. The ultrasonic pulse velocity
2.4.4. Evaluation of ultrasonic pulse velocity was calculated using Eq. (3).
Figs. 4 and 5 illustrate the method for measuring the ultrasonic L
pulse velocity. The ultrasonic pulse velocity in a concrete specimen V¼ ð3Þ
T
was measured before, during, and after heating, when the speci-
men was cooled. Before and after heating, the ultrasonic pulse where
velocity was measured in accordance with ASTM C597-Standard
Test Method for Pulse Velocity through Concrete. V : Ultrasonic wave velocity (km/s)
Since there are no commercial devices to measure the ultra- T: Ultrasonic wave propagation time (ms)
sonic pulse velocity in concrete subjected to high-temperature L: Length of the concrete specimen (mm)
heating, a method was devised to measure the ultrasonic pulse
velocity from outside the heating furnace using the existing PUN- The time required for the ultrasonic wave to propagate through
DIT equipment. To transmit ultrasonic pulses into the concrete a concrete specimen varies depending on the internal state of the
specimen located inside the heating furnace, the transducers and concrete. When the concrete integrity is maintained, the path
concrete were connected using bar-type SUS-316 of a £25 diame- through which the ultrasonic pulse reaches the opposite side of
ter. SUS-316 is a material with high strength and good corrosion the specimen is shorter, and thus, the propagation time is less.
resistance under high-temperature conditions. Furthermore, to When the concrete specimen is subjected to a high temperature,
prevent damaging of the transducers due to the heat, the contact the path of ultrasonic waves becomes longer, due to the cracks
or pores generated inside and outside the concrete, as shown in
Fig. 1. Therefore, the ultrasonic pulse velocity decreases. The ultra-
sonic pulse velocity depends on the dynamic elastic modulus, Pois-
son’s ratio, and density of the propagated medium. Malhotra and
Carino [18] reported that the elasticity of concrete significantly
declines and the ultrasonic pulse velocity of concrete decreases,
if the heating temperature is high.
where
Fig. 6. Ultrasonic pulse velocity of SUS bar during heating and cooled after heating. F c : Ratio of comp. strength of concrete
V p : Ratio of ultrasonic pulse velocity
Fig. 9 shows the residual elastic modulus ratio of concrete For a 30 MPa specimen, the estimated compressive strength
cooled after heating. Unlike the residual compressive strength, was lower than the measured compressive strength. For a 70
the residual elastic modulus tended to decrease with an increase MPa specimen, the estimated and measured compressive strengths
in the heating temperature, regardless of the compressive strength were similar. Finally, for a 110 MPa specimen, the estimated com-
of concrete. The residual elastic modulus decreased sharply above pressive strength was higher than the measured compressive
300 °C. strength. It is considered that because the internal structure of con-
The first effect of heating on concrete is the evaporation of free crete changes depending on the amount of binder and aggregate of
water within at 100–200 °C. During the time it takes to reach 250 concrete, heating and cooling conditions.
°C, unevaporated water and water of hydration are lost [19,20]. Among them, the phenomenon that the ultrasonic pulse veloc-
This can be confirmed in the DTA/TGA and XRD results of cement ity is different in a similar range of residual compressive strength
paste [20,21] Moreover, concrete is not particularly affected phys- of concrete is confirmed, as shown in Fig. 11. Typically, 30 MPa
ically at temperatures below 300 °C. concrete exposed at room temperature and 200 °C, and 70 and
Previous studies reported drastic chemical and physical 110 MPa concrete exposed at 500 °C show similar residual com-
changes occurring in concrete at temperatures above 300 °C, which pressive strength ranges, but the stress–strain curve relationships
516 E. Hwang et al. / Construction and Building Materials 171 (2018) 511–520
Fig. 8. Residual compressive strength of concrete cooled after heating. Fig. 10. Relationship between measured and estimated values of compressive
strength of concrete by ultrasonic pulse velocity.
Fig. 13. Ultrasonic pulse velocity of concrete during heating and cooled.
Fig. 12. Strain behavior of concrete cooled after heating (70 MPa).
Moreover, as mentioned previously, the residual compressive ultrasonic pulse velocity during the high temperature heating.
strength of concrete decreases after a 300 °C temperature; how- Instead, the ultrasonic pulse velocity reduced after reaching the
ever, the residual elastic modulus of concrete decreases starting static fracture strain.
at 100 °C. After a 300 °C temperature, the residual compressive The thermal strain of concrete increases with an increase in the
strength decreases sharply and the thermal strain becomes larger heating temperature and reaches the static fracture strain of con-
than the static fracture strain (0.0025–0.0030), as shown in crete at a 300 °C temperature. When the thermal strain is smaller
Fig. 15. In addition, the decrease in the elastic modulus is due to than the static fracture strain, it is confirmed that the water vapor
the plastic strain stemming from to the thermal strain in the con- pressure of concrete increases, while the increase in the permeabil-
crete after 100 °C, and it can be considered that the degradation of ity is insignificant. Though the degradation of concrete occurs due
concrete begins to occur. The degradation of concrete occurs due to to the thermal strain in the range from room temperature to
the plastic strain; however, the effect on the residual compressive 300 °C, it is still in the range of plastic strain. In the range above
strength is negligible. The degradation of concrete that occurred 300 °C, the thermal strain is more than the static fracture strain
before reaching the static fracture strain was not reflected in the of the concrete, the water vapor pressure is discharged, and the
permeability is approximately 100 times higher than that at the
room temperature. This is because the thermal strain becomes
larger than the static fracture strain at a 300 °C or more, resulting
in a rapid increase in the development of micro-cracks in concrete.
Moreover, a close relationship exists with the phenomenon,
wherein the ultrasonic pulse velocity of concrete rapidly decreases
after 300 °C.
In addition, the cross sections of the concrete specimens cooled
after heating up to 700 °C were checked as showed in Fig. 16.
Cracks were generated in the interfacial transition zone (ITZ) of
the aggregate and paste, and inside the aggregate. Crack bridges
were also observed between the cracks. The number of internal
cracks formed may vary according to the compressive strength of
concrete. However, concrete specimens manufactured with the
same materials are expected to show identical crack formations
from high temperature heating. It is considered that the mechani-
cal properties of concrete degraded and the ultrasonic pulse veloc-
ity decreased because of the internal cracks developed by the high
temperature heating.
Although the ultrasonic pulse velocity decreased in the concrete
specimens subjected to high temperature, due to the micro-cracks,
it did not decrease during heating up to 300 °C. It was the case
because the width of micro-cracks was reduced via the thermal
expansion due to heating. The ultrasonic pulse velocity measured
in the concrete specimens cooled after heating significantly
decreased, because the width of micro-cracks in the concrete spec-
imens increased due to cooling. Furthermore, a higher heating
temperature led to a higher reduction in the ultrasonic pulse veloc-
ity after cooling.
It was confirmed that this difference in the ultrasonic pulse
velocity was different from the results of residual and high-
temperature compressive strength of concrete, as shown in Figs. 8
and 14. It was found that, although the width of the micro-cracks
in concrete was increased by cooling, the residual and high-
temperature compressive strengths of concrete remained similar
[22]. However, the ultrasonic pulse velocity, significantly varied
due to a high temperature heating and cooling after heating.
Because the influence of micro-cracks generated by the heating
on the residual and high-temperature compressive strength of con-
crete was insignificant, it was making it possible to measure the
changes in micro-cracks of the concrete subjected to the tempera-
ture using the ultrasonic pulse velocity.
Also, the ultrasonic pulse velocity in concrete, while the heating
temperature is maintained after heating, was examined, as shown
in Fig. 17. While the heating temperature was maintained, the
ultrasonic pulse velocity in concrete decreased, except for the case
when temperature was 100 °C. It was confirmed that the concrete
degradation continued while the temperature was maintained,
even though the decrease in the ultrasonic pulse velocity was
Fig. 15. Thermal strain, water vapor pressure and permeability of concrete with insignificant, compared to the high-temperature heating
temperature. conditions.
E. Hwang et al. / Construction and Building Materials 171 (2018) 511–520 519
Fig. 16. Cross section of concrete cooled after heating (700 °C).
4. Although the accuracy of estimating the compressive strength [6] M.H. Kim, S.J. Choi, S.P. Kang, J.H. Kim, J.H. Jang, A study on the application of
non-destructive testing equation for the estimation of compressive strength of
via the ultrasonic pulse velocity is low, it is possible to diagnose
high strength concrete, J. Korea Inst. Build. Constr. 2 (3) (2002) 123–130.
presence of defects in concrete through a relative reduction in [7] N. Gucunski, R. Feldmann, F. Romero, S. Kruschwitz, A. Abu-Hawash, M. Dunn,
the ultrasonic pulse velocity, continuously measured in a con- Multimodal condition assessment of bridge decks by NDE and its validation.
crete specimen subjected to high-temperature heating. Hence, In: Proc. 2009 Mid-Continent Transportation Research Symp., Ames, Iowa,
2009, 261(5), 1–18.
the ultrasonic pulse velocity makes possible a continuous eval- [8] H. Yang, Y. Lin, C. Hsiao, J.Y. Liu, Evaluating residual compressive strength of
uation of the degradation of concrete specimens, also the eval- concrete at elevated temperatures using ultrasonic pulse velocity, Fire Saf. J. 44
uation method proposed in this study is anticipated to allow the (1) (2009) 121–130.
[9] A. Benedetti, On the ultrasonic pulse propagation into fire damaged concrete,
integrity evaluation of concrete at high temperature heating. In Struct. J. 95 (3) (1998) 259–271.
addition, in the future, it will be necessary to improve the [10] T.T. Lie, T.J. Rowe, T.D. Lin, Residual strength of fire-exposed reinforced
equipment to evaluate the integrity of concrete exposed to high concrete columns, Spec. Publ. 92 (1986) 153–174.
[11] A.A. Shah, Y. Ribakov, Non-linear ultrasonic evaluation of damaged concrete
temperature. based on higher order harmonic generation, Mater. Des. 30 (10) (2009) 4095–
4102.
[12] A.A. Shah, Y. Ribakov, S. Hirose, Nondestructive evaluation of damaged
Conflict of interests concrete using nonlinear ultrasonics, Mater. Des. 30 (3) (2009) 775–782.
[13] Lotfollah Pahlavana, Fengqiao Zhang, Gerrit Blacquière, Yuguang Yang, Dick
There are no conflicts of interest to declare. Hordijkb, Interaction of ultrasonic waves with partially-closed cracks in
concrete structures, Constr. Build. Mater. 167 (2018) 899–906.
[14] T.C. Rilem, 129-MHT, Test methods for mechanical properties of concrete at
Acknowledgement high temperatures: Part 3 - Compressive strength for service and accident
conditions, Mater. Struct. 28 (1995) 410–414.
[15] M.H. Yoon, G.Y. Kim, T.G. Lee, G.C. Choe, E.C. Hwang, J.S. Nam, Creep behavior
This work (NRF-2015R1A2A2A01007705) was supported by of high-strength concrete subjected to elevated temperatures, Materials 10 (7)
Mid-career Researcher Program through NRF grant funded by the (2017) 781.
MEST. [16] G.Y. Kim, G.C. Choe, J.H. Lee, S.H. Lee, T.G. Lee, Evaluation on spalling properties
of ultra high strength concrete with melting and vaporization of fiber, J. Korea
Concr. Inst. 24 (2) (2012) 173–183.
References [17] ISO standard 8841. Dense, shaped refractory products-Determination of
permeability to gases; 1991.
[18] V.M. Malhotra, N.J. Carino, Handbook on Nondestructive Testing of Concrete,
[1] G.C. Choe, G.Y. Kim, N. Gucunski, S.H. Lee, Evaluation of the mechanical 2nd ed., CRC Press, 2003.
properties of 200MPa ultra-high-strength concrete at elevated temperatures
[19] W.M. Lin, T.D. Lin, L.J. Powers-Couche, Microstructures of fire-damaged
and residual strength of column, Constr. Build. Mater. 86 (2015) 159–168. concrete, Mater. J. 93 (3) (1996) 199–205.
[2] G.Y. Kim, Y.S. Kim, T.G. Lee, Mechanical properties of high-strength concrete
[20] S.K. Handoo, S. Agarwal, S.K. Agarwal, Physicochemical, mineralogical, and
subjected to high temperature by stressed test, Trans. Nonferrous Met. Soc. morphological characteristics of concrete exposed to elevated temperatures,
China 19 (2009) 128–133. Cem. Concr. Res. 32 (7) (2002) 1009–1018.
[3] M.H. Yoon, G.Y. Kim, G.C. Choe, Y.W. Lee, T.G. Lee, Effect of coarse aggregate
[21] O. Arioz, Effects of elevated temperatures on properties of concrete, Fire Saf. J.
type and loading level on the high temperature properties of concrete, Constr. 42 (8) (2007) 516–522.
Build. Mater. 78 (2015) 26–33.
[22] G.Y. Kim, S.H. Jung, T.G. Lee, Y.S. Kim, J.S. Nam, Compressive behavior of
[4] M. Ozawa, H. Morimoto, Effects of various fibers on high-temperature spalling concrete with loading and heating, J. Korea Inst. Struct. Maint. Insp. 14 (4)
in high-performance concrete, Constr. Build. Mater. 71 (2014) 83–92. (2010) 119–125.
[5] S.U. Hong, Y.S. Cho, A study on the estimation of the compressive strength of [23] Architectural Institute of Japan, Guide book for fire-resistive performance of
concrete structures using ultrasonic pulse velocity method and rebound structural materials, 2009;40–54.
hardness method, Archit. Inst. Korea 27 (1) (2011) 19–26.