Thermo-Mechanical Response of Concrete

Guruprasad Y.K. Ananth RAMASWAMY

Ph.D. Student Professor
Indian Institute of Science Indian Institute of Science
Bangalore, India Bangalore, India
ananth@civil.iisc.ernet.in

Summary
Consequences of long term exposure to high temperatures (e.g. fires) i.e., thermal induced damages can
result in substantial degradation of concrete and steel, which are the primary materials of construction
for all major structures used in both strategic (e.g., nuclear reactors) and common applications
(buildings, bridges and other urban structures). Response of real buildings to fire has highlighted that
structural resistance to fire depends strongly on the level of redundancy.
The present study experimentally evaluates the performance of control (standard cylinder
specimen), damaged (mechanical loading after thermal exposure) and repaired / retrofitted normal plain
concrete cylinders using different repair schemes such as use of carbon fiber reinforced polymer (CFRP)
wraps. Sustainability of the repair/ retrofit scheme against additional thermally induced damage has
been examined using thermally stable insulating materials such as geo-polymers applied on CFRP, to
assess the combined benefit in mechanical strength due to thermal insulation and confinement. Effect of
temperature on flexural response of concrete and the fracture properties are also discussed. The reasons
for the degradation of the strength of concrete is attributed to drying of concrete and cracking of the
mortar(matrix) at the interface of aggregates in part and to some extent to the melting in the coarse
aggregates at very high temperatures. Computational models based on finite element method (ABAQUS)
have been used to assess the contribution from different elements in the various configurations.
Keywords: Structural Concrete Repair, SCC with fibres, Carbon and glass fibre reinforced polymer,
high temperature, Geo-polymer

1. Introduction
Consequences of long term exposure to high temperatures (e.g. fires) can also result in
substantial degradation of concrete. Micro-textural (petro-graphical) and chemical exchange /
diffusion studies are not widely used in concrete research. The major studies using the petrographic
studies were done on old structures from which only limited samples are available. The major areas
of research with petrographic application are on various aspects of cement alterations by time
example thaumasite sulphate attack [1].
The research in concrete is highly concentrated on the quality development of the cement
however very little research attention given to the material property of coarse and fine aggregates.
Recently, [2] presented the constitutive relationships developed for normal-strength concrete and
high-strength concrete subjected to fire to provide efficient modelling and specify the fire-
performance criteria for concrete structures exposed to fire. A recent study [3] reported the strength
and stiffness properties of concrete deteriorate with an increase in temperature as encountered
during exposure to fire.
Exposure of the concrete to high temperatures, such as 0 to 400 degrees Celsius leads to
release of the pore water which transforms in to water vapour. The vapour imposes a pressure on
the void channels so created in the system. This dehydration process results in inducing brittleness
in the concrete also resulting in eventual failure. Studies [4-7] have considered the complex
phenomena of moisture transport under thermal loading conditions using a combined numerical
model for the mechanical, fluid and thermal effects. Majorana et al [7] have proposed a finite
element model for creep coupled with damage and related cross effects due to hygro-thermal
behaviour, while Gawin et al [5] consider the concrete to be treated as a partially saturated porous
material permitting the consideration of a hydration-dehydration, evaporation-condensation,
adsorption-desorption phenomena and nonlinearities due to temperature and pressure. The
brittleness developed in the material due to dehydration impacts the flexural and fatigue properties
of concrete.
Development of repair / rehabilitation or retrofit techniques in existing structures offers
engineer / planner additional possibilities in deployment of funds in an effective manner. The initial
development in retrofitting RC beams having damage in the tension region using FRP plates
(Carbon, Glass and Aramid fiber plates) was proposed by Meier[8]. A number of other studies [9-
10] have experimentally or analytically shown the benefits of FRP as a strengthening system. These
studies conclude that the surface preparation of the concrete, the selection of the adhesive, amount
of tension steel and its effective cover in the concrete beam were some of the important factors
affecting the overall performance of the retrofit. Carbon and glass fibre based polymer sheets were
found to be more suitable in FRP plates used in retrofit. Significant enhancement in strength and
ductility has been reported in a study [11] with the use of FRP tubes and the filament winding
(GFRP). The use of FRP plates in repair and strengthening of RC member against failure in shear
have also been reported [12-13]. The studies have reported equations for effective FRP strain as
function of axial rigidity of FRP based on several experimental works. Based on the effective strain
of FRP, FRP contribution to shear capacity was calculated. A limiting value of FRP contribution to
shear was given (axial rigidity equal to 0.4 GPa) beyond which the effectiveness of strengthening
ceased to be positive.
Sustainability of the repair/ retrofit scheme against additional damage has been examined using
thermally stable insulating materials. Recently, a study [2] presents constitutive relationships
developed for normal-strength concrete and high-strength concrete subjected to fire to provide
efficient modelling and specify the fire-performance criteria for concrete structures exposed to fire.
Some studies using geo-polymers as a thermal insulation along with mechanical retrofit with CFRP
has been examined in this study. Similar to ceramics that can sustain high temperatures,
development of the proper composition of geo-polymers for thermal stability is explored, as this
holds the promise of being used in combination with CFRP to offer both mechanical and thermal
retrofit. Further to get a better understanding of material behaviour at different length scales due to
damage, modern tools like scanning electron microscopy, micro-indentation and thin section studies
have been employed to characterize concrete at different scales of length.

2. Details of the study

2.1 Experimental Program
In the present study plain concrete cylinders of normal strength corresponding to 25MPa and
35MPa cylinder compressive strength at 28 days were prepared and used. The experimental
program comprised of evaluating the mechanical properties of concrete for these two strengths
under different mechanical and thermal followed by mechanical loading scenarios. Both circular
and rectangular columns, that are exposed to monotonic and cyclic loads, different thermal
excursions comprising of temperature and exposure time, followed by cooling and applying
mechanical loads have been considered in the study.
The first concrete mix of 25MPa cylinder compressive strength was made such that for one cubic
metre of concrete by volume, had 272.7 kg of cement, 900 kg of fine aggregate and 1227.3kg of
coarse aggregate. The water cement ratio was maintained at 0.65 for this concrete mix. The second
concrete mix of 35MPa cylinder compressive strength was made such that for one cubic metre of
concrete by volume, had 460.4 kg of cement, 607.7 kg of fine aggregate and 1331.9kg of coarse
aggregate. The water cement ratio was maintained at 0.5 for this concrete mix . The properties of the
CFRP fabric and epoxy employed in the present study are given in Table 1.

Table 1: Properties of CFRP and the epoxy resin after 7 days of curing (drying)

Properties of CFRP Properties of Epoxy

Tensile Strength 600MPa 350 – 400 Kg / cm2
Compressive Strength 420MPa 800 – 1200 Kg / cm2
Young’s Modulus 50GPa 2000 – 2200 Kg / cm2

As a control reference, cylinder specimens (150mm diameter and 300mm length)

corresponding to the two strengths were monotonically loaded in uni-axial compression to failure.
Electrical resistance gauges were used to measure longitudinal and circumferential strains. In
addition a number of plain concrete cylinder specimens were placed in a high temperature electrical
oven / furnace at temperatures ranging from 450 degree Celsius to 715 degree Celsius for durations
ranging from two to four hours. The specimen were thereafter removed from the furnace and cooled
for over a day prior to testing them under uni-axial compressive loads to failure. In order to assess
the repair performance, companion cylinders that were subject to the heating and cooling process
were repaired with CFRP wraps to obtain strength enhancements due to the confinement introduced
by the wrap.
A similar study was undertaken in prism specimen exposed to various temperature exposure
combinations and then subject to flexural loads, both monotonically and under cyclic conditions.
Repair scenarios applied to similarly heated, cooled, repaired and tested under flexural loads have
also been studied to assess efficacy of the repair..
As epoxy materials used in binding the CFRP to concrete cannot sustain high temperature,
geo-polymer based mix composition was developed and applied on the damaged concrete cylinders
having a CFRP wrap so as to provide both thermal insulation (geo-polymer) and mechanical
strength (CFRP wrap) to damaged concrete so that the repair can sustain subsequent exposure to
high temperature.

2.2 FE Modelling
ABAQUS 6.12 [14] finite element software has been used to computationally simulate both
the effects of the heating and cooling process leading to concrete damage and the subsequent
mechanical loading without and with CFRP wrap. Concrete is modelled as a 3D deformable solid
with homogeneous as section type. A coupled temperature displacement elements (C3D8T) 8-node
thermally coupled brick elements with, tri-linear displacement and temperature for cases with
thermal exposure is used to model concrete, while CFRP is modelled as a 3D deformable planar
shell with membrane as section type, namely, an 8-node quadrilateral membrane element (M3D8) is
used for the CFRP. The CFRP is modelled as a complete wrap without overlap. The concrete
material properties/parameters that is entered in the finite element analysis are: Density, elasticity,
Poisson’s ratio, thermal conductivity, thermal expansion, specific heat, concrete damaged plasticity
parameters. For concrete with thermal exposure, the data for material properties were entered as
temperature dependent. According to the material behaviour, the elasticity (E) of concrete decreases
with increase in temperature [3], during the heating phase. When allowed to cool up to ambient
temperature (Tambient), the value of the elasticity at the end of the cooling phase attains a value
slightly less than the value which was at the end of the heating phase [3]. The elasticity was entered
as temperature dependent data in the analysis. The thermal conductivity, thermal expansion and
specific heat of concrete are considered as temperature dependent data for the finite element
analysis are taken from literature [3]. ABAQUS 6.12 employs a concrete damage plasticity model
in order to capture cracking and crushing of concrete. The parameters used to construct the failure
surface in this model are the dilatation angle, an eccentricity parameter and a parameter that
quantifies the shape of the failure surface in the deviatoric plane.
The interaction condition between the concrete and CFRP is entered in the finite element
analysis is in terms of contact properties as tangential and normal behaviour. The tangential contact
behaviour is entered with the friction formulation: rough. The normal contact behaviour is entered
with the pressure over closure criterion as: hard contact with the option of allowing separation after
contact. For concrete exposed to high temperature in the interaction condition sink temperature
(ambient temperature) is entered as interaction between the heat analysis and the mechanical
analysis. The thermal load is applied on the structural component in the finite element analysis as
surface heat flux.

3. Results and Discussion

The comparison of cylinder strength degradation with increased temperature and exposure
time is seen in Table 2 for both concrete mixes. The effectiveness of even one layer of CFRP is also
evident from the results shown in Table 2. While on the control specimen one layer results in an
improvement of 40% on the control specimen peak strength, the increase in case of specimen
exposed to 55 deg Celsius and then repaired with a layer of CFRP provides enhancements on the
unrepaired value to the extent of providing the control concrete strength.

Table 2: Comparison of Thermally Damaged Concrete Cylinder Strengths without and with
one layer of CFRP

Ultimate Uni-axial
compressive strength at
Classification of Concrete Cylinders failure
Without With CFRP
CFRP (one layer)
Control concrete cylinder 25MPa 25.5 MPa 35MPa
Concrete cylinder 25MPa exposed to 425deg C for 4hrs 19.3 MPa 31.6MPa
Concrete cylinder 25MPa exposed to 550deg C for 4hrs 12.4 MPa 21.8MPa
Concrete cylinder 25MPa exposed to 715deg C for 2hrs 6 MPa 14.6MPa
Concrete cylinder 25MPa exposed to 715deg C for 4hrs 4.6 MPa 11.4MPa
Control concrete cylinder 35MPa 35 MPa 45MPa
Concrete cylinder exposed to 425deg C for 2 hours 28.8 MPa -
Concrete 35MPa exposed to 425deg C for 4hrs 21.0 MPa -
Concrete cylinder 35MPa exposed to 550deg C for 4hrs 12.0 MPa 25MPa
Concrete cylinder 35MPa exposed to 715deg C for 2hrs 7.0 MPa 14.9MPa
Concrete cylinder 35MPa exposed to 715deg C for 4hrs 4.9 MPa 9.7MPa

Micro scale investigations (Scanning electron microscopy of thin sections, Electron Probe
Micro Analysis, micro/ nano indentation methods) is used to develop insights into behavior at these
length scales. The reasons for the degradation of the strength of concrete is attributed to drying of
concrete and cracking of the mortar(matrix) at the interface of aggregates in part and to some
extent to the melting in the coarse aggregates at very high temperatures.
Figure 1 shows a comparison of the ABAQUS based simulation of the concrete cylinder
exposed to 425 Celsius for four hours, cooled wrapped with a layer of CFRP and then loaded to
failure along with the corresponding experimental result. The comparison is quite good indicating
that the modelling parameters are adequately capturing the multi-physics features of this coupled
problem.
 Figure 1: Comparison of Experimental and Finite Element Simulation of 25MPa Concrete
Cylinder exposed to 425 degree Celsius temperature for four hours, cooled, wrapped with one
layer of CFRP and tested to failure [blue line FEM, red line Experiment]

Results from studies on cylinders having both CFRP wrap and geo-polymer layer indicating
strength enhancements when exposed to high temperature again are in progress. Results from
flexural studies on control and heat treated specimen will be presented. Strength enhancements with
CFRP on notch prism specimen result in strength enhancements of the order of the undamaged
specimen.

4. Closing Remarks
The study shows that CFRP is an effective repair material for concrete damage induced by
mechanical or thermal loads due to the confinement provided by it as a wrap. The use of geo-
polymer as a thermal insulation over the CFRP wrap promises to be an effective repair of damaged
concrete against subsequent thermal damage. The computational modelling of the experiments
using ABAQUS is quite satisfactory in reproducing the test results.

5. Acknowledgements
The authors would like to gratefully acknowledge the financial support received from the
Board for Research in Nuclear Sciences, Government of India under grant No. 2012/36/42/BRNS
and the funds received from CISTUP, IISc.

6. References
[1] Eden, M.A. “The Laboratory Investigation of concrete Affected by TSA in U.K.”. Cement and
Concrete Composites, 25, 2003, 847-850.
[2] Aslani, F. and Bastami, M. (2011), “Constitutive relationships for normal- and high-strength
concrete at elevated temperatures:” ACI Materials Journal pp 355-365.
[3] Khaliq, W. and Kodur, V.K.R. (2011) Effect of High Temperature on tensile strength of
different types of high-strength concrete, ACI Materials Journal pp. 394-402
[4] Gawin, D. and Schrefler, B.A. (1996) “Thermo-hydro-mechanical analysis of partially saturated
porous materials”, Engineering computations, 13(7), 113-143.
[5] Gawin, D. Majorana, C.E., Pesavento,F. and Schrefler, B.A. (2001) “Modelling of thermo-
hygro-mechanical behavior of high performance concrete in high temperature environments”,
Fracture Mechanics of Concrete Structures, de Borst et. al (ed.), pg. 199-206, Swets and
 Zeitlinger, France.
[6] Nechnech, W., Reynourd, J.M., and Meftah, F. (2001) “On the modeling of thermo-mechanical
concrete the finite element analysis of structures submitted to elevated temperatures”, Fracture
Mechanics of Concrete Structures, de Borst et. al (ed.), pg. 271-278, Swets and Zeitlinger,
France.
[7] Majorana,C., Salomoni V., Schrefler, B. (1998) “Hygrothermal and Mechanical model of
concrete at high temperature”, Materials and structures, 31, 378-386.
[8] Meier, U. Bridge “Repair with high performance Composite Materials”, Material und Technik,
v4, 125128, in German, 1987.
[9] Swamy, R.N., Jones, R. and Bloxham, J.W. “Structural Behavior of Beams Strengthened with
Epoxy Bonded Plates”, The Structural Engineer, 65A(2), 59-68, 1987.
[10] Triantafillou, T., and Plevris, N. “Strengthening of Reinforced Concrete Beams Using Epoxy
Bonded FRP Composites”, Materials and Structures, RILEM, v25, 201-211, 1992.
[11] Nanni, A., Bakis, C.E. and Boothby, T.E. “Test Methods for FRP-Concrete Systems Subjected
to Mechanical Loads: State of The Art Review”, Journal of Reinforced Plastics and
Composites, v14, 524-558, 1995.
[12] Norris, T., Saadatmanesh, H. and Ehsani, M.R. “Shear and Flexure Strengthening of RC
Beams with Carbon Fiber Sheets”, ASCE Journal of Structural Engg., 123(7), 903-911, 1996.
[13] Triantafillou, T.C., ``Shear Strengthening of reinforced Concrete Beams Using Epoxy-Bonded
Composites,'' ACI Structural Journal V.95,No. 2, Mar-Apr 1998,pp. 107-115.
[14] ABAQUS “User’s Manual”, Simulia V6, Dassault Systems, 2013, U.S.A.