You are on page 1of 8

Fracture Mechanics of Concrete and Concrete Structures Assessment, Durability, Monitoring and Retrofitting of Concrete Structures- B. H. Oh, et al.

(eds)
2010 Korea Concrete Institute, Seoul, ISBN 978-89-5708-181-5

A local bond stress-slip model for reinforcing bars in self-compacting


concrete
P. Desnerck, G. De Schutter & L. Taerwe
Ghent University, Ghent, Belgium

ABSTRACT: The force transfer in reinforced concrete is provided by the concrete-to-steel bond. This phenomenon has widely been studied for conventional vibrated concrete (CVC). For self-compacting concrete
(SCC) however less test results are available. To fill in this lack and to develop adapted standards for predicting the bond of reinforcement in SCC, an experimental program has been set up. The bond strength of reinforcement bars with different diameters has been tested by means of beam-test specimens. During testing
the bond stress-slip response was recorded. From the test results it can be seen that the maximum bond
strength of SCC is slightly higher than for CVC when small bar diameters are studied. For larger bar diameters the difference becomes smaller. Comparison of the test data with bond models indicated that the bond
clauses underestimated the bond strength for SCC as well as for CVC. Therefore an adjustment of the bond
model has been made.
1 INTRODUCTION
Reinforced concrete is far from homogeneous. It is
built up of steel and concrete and the concrete itself is not homogeneous neither. Reinforced concrete elements are basically designed so that the
concrete can carry the compressive stresses and
the steel can resist the tensile stresses. Therefore a
good force transfer between the two materials is
necessary which can only be achieved by an interaction between both materials, which is provided
by bond between the reinforcement bars and the
concrete.
The bond has an important influence on the behaviour of reinforced elements in the cracked stage.
Crack widths and deflections are influenced by the
distribution of bond stresses along the reinforcement
bars and by the slip between the bar and the surrounding concrete.
Due to the importance of the interaction between
steel and concrete a lot of research has been done
in the past. In all these projects the main focus
was on the reinforcement bar, its geometrical
characteristics and how these characteristics influence the bond strength. With the appearance of
new concrete types, such as steel-fibre reinforced
concrete and high strength concrete, questions
arose about the bond strength achieved with these
concrete types, and the main focus of the research
on bond shifted to the concrete and its composition (Martin 2002).

The same questions can be formulated for selfcompacting concrete. A concrete type which, in
fresh state, has the ability to flow under its own
weight, fill the required space or formwork completely and produce a dense and adequately homogeneous material without a need for compaction (De
Schutter et al. 2007). The advantages are clear: no
need for vibration of the concrete, a higher quality of
the finished element, reduced construction times,
. The self-compactability is achieved by adding a
superplasticizer to the mixture and by reducing the
amount of coarse aggregates. Although selfcompacting concrete (SCC) is a relatively new material, already a lot of research has been done on the
durability and the workability of the concrete type
(De Schutter et al. 2008). Less studies have been focussing on the mechanical properties and more in
particular the bond aspects.
Some programs have been carried out to determine the force transfer between concrete and reinforcement in self-compacting concrete. These studies show that the bond strength of steel in SCC is not
lower than for conventional vibrated concrete
(CVC), and may be even higher in some cases
(Almeida et al. 2008, Chan et al. 2003, Zhu et al.
2004, Dehn et al. 2004). Nevertheless there is a great
scatter in the results.
To get a better insight in the difference in bond
strength between conventional vibrated concrete and
self-compacting concrete and to develop modified
models describing the bond stress-slip behaviour,
this research program has been set up.

Materials (kg/m)
SCC1
SCC2
CVC1

w I52,5
w
w
CEM
N
360
360
300
h + ( D h ) = e
e853

w
Sande0/4 mm
853
c
s
n 640 (3)
h

h t
Gravel 2/8 mm
462
c 263 s263
Gravel 8/16 mm
434
434
762
Limestone filler
240
300
the sorption/desorption
where
we/h is the slope of
Water
165
165
165
isotherm
(also
called
moisture
capacity).
The
Superplasticizer
3.6
3.0
governing
equation
(Equation
3)
must
be
completed
fccub (N/mm)
71.7
62.1
58.4
by
appropriate boundary and63.7
initial conditions.
fc (N/mm)
57.5
51.8
fct,flThe
(N/mm)
7.2 amount
6.8 of evaporable
6.2
relation between the
(N/mm)
5.0
4.4
4.1
f
ct,sp
water and relative humidity is called adsorption

&+

&+ &

isotherm if measured with increasing relativity


The self-compacting
concrete
SCC2inhas
compahumidity
and desorption
isotherm
theaopposite
rable
compressivetheir
anddifference
tensile strength
as 1994),
the concase. Neglecting
(Xi et al.
in
ventional
vibrated
concrete
CVC1,
as
was
intended.
the following, sorption isotherm will be used with
The
first self-compacting
concrete
SCC1,conditions.
with the
reference
to both sorption and
desorption
same
W/C
ratio,
has
a
significantly
higher
strength.
By the way, if the hysteresis of the moisture
Besideswould
the concrete
steel two
bar diameter
isotherm
be takentype,
into the
account,
different
has
been
varied.
In
this
research
program,
5 different
relation, evaporable water vs relative humidity,
must
nominal
diameters toofthe
thesign
embedded
reinforcement
be used according
of the variation
of the
bars
were chosen:
12, 20,
32 andof40 the
mm. sorption
relativity
humidity.
The25,shape
isotherm for HPC is influenced by many parameters,
especially
thosedetails
that influence extent and rate of the
2.2
Specimen
chemical reactions and, in turn, determine pore
To
test the
of reinforcing
bars in the
structure
andbond
porestrength
size distribution
(water-to-cement
different
concrete
types, composition,
the standard SF
beam-test
ratio, cement
chemical
content,
geometry,
as
described
in
the
RILEM
recommendacuring time and method, temperature, mix additives,
tion,
The specimen
depend
etc.). isInused.
the literature
variousdimensions
formulations
can on
be
the
bar
size.
Three
types
of
specimen
are
used
found to describe the sorption isotherm of normal
depening
on the
bar 1994).
diameter.
A type in
I specimen
is
concrete (Xi
et al.
However,
the present
used
for
tested
reinforcing
bars
with
a
diameter
paper the semi-empirical expression proposed by
smaller
16 mm.(1997)
A typeisII adopted
is used for
bars be-it
Norling than
Mjornell
because
tween 16 mm en 32 mm, and a third type is used for
Proceedings of FraMCoS-7, May 23-28, 2010

180

180

100 30

12

c )h

60

130

50
650
800

28

26

100

20

10

10

10

180

10

60

185

50

where the first term (gel isotherm) represents the


physically bound (adsorbed) water and the second
term (capillary isotherm) represents the capillary
water. This expression is valid only for low content
of SF. The coefficient G1 represents the amount of
water per unit volume held in the gel pores at 100%
relative humidity, and it can be expressed (Norling
Mjornell 1997) as
120

vapor, and adsorbed water) and the non-evaporable


(chemically
2.1
Materials bound) water wn (Mills 1966,
Pantazopoulo & Mills 1995). It is reasonable to
Three
havewater
been used:
2 powderassumetypes
that oftheconcrete
evaporable
is a function
of
type
self-compacting
concretes
and
one
conventional
relative humidity, h, degree of hydration, c, and
vibrated
details
i.e. we=the
we(h,mixing
c,s)
degree ofconcrete.
silica fumeMore
reaction,
s, about
procedure
and
the
used
materials
can
be
in
= age-dependent sorption/desorption found
isotherm
(Desnerck
2008).
The
mix
proportions
and
concrete
(Norling Mjonell 1997). Under this assumption and
strengths
are summarized
by substituting
Equationin1Table
into 1.Equation 2 one
obtains
Table 1. Mix design for SCC and CVC mixes.

10(g
K1 ( c , s )e 1 c

plastic tubes
150

50

TheThe
common
way to testcoefficient
the bond strength
proportionality
D(h,T) of
is rebars
called
in
concrete
is
by
means
of
pull-out
tests
moisture permeability and it is a nonlinear (RILEM
function
1973).
The behaviour
types of specimens
is
of the relative
humidityofhthese
and temperature
T (Baant
quite
different
from
that
in
reinforced
elements
sub& Najjar 1972). The moisture mass balance requires
jected
bending. in time of the water mass per unit
that theto variation
The
beam
test specimen
suggested
RILEM
volume of concrete
(water content
w) be by
equal
to the
recommendation
part 1flux
(RILEM
1973) is more
divergence of theRC6
moisture
J
suitable to evaluate the bond strength of reinforced
elements
subjected to bending. The specimen, con(2)
w = of 2J half-beams, is loaded on top introducing
sisting
t
bending moments in the beam. In this way a more
realistic
stresscontent
distribution
and around
the sum
bar
The water
w caninside
be expressed
as the
is
created.
of the evaporable water w (capillary water, water

explicitly
for than
the 32
evolution
hydration
bars
equal accounts
to or larger
mm. Anofexample
is
reaction
content. (Fig.
This1).sorption isotherm
given
for aand
typeSF
I specimen
reads
The prescribed bond length is 10 times the bar diameter . However this leads to yielding, and in
some cases even rupture,
bar
of the reinforcement

before reaching the ultimate


bond 1strength. There
we (h,most
c , s ) = G ( c , s )1
+
a bond
fore
of the
specimens 10are
cast
with
1

(g
c )h

length of 5 times .
e 1 c

(4)

160

(1)

180

D ( h , T ) h
2J =EXPERIMENTAL
PROGRAM

6 x 50
335

100

Figure 1. Size and


of beam test specimen type I
c reinforcement
s
G1 ( c , s ) in
= kmm).
(5)
(dimensions
vg c c + k vg s s

k vg are material parameters. From the


where
k vg and
2.3
Testing
procedure
c

maximum amount of water per unit volume that can

During
the tests,
the specimens
weregelloaded
fill all pores
(both capillary
pores and
pores),atonea
constant
rate
corresponding
to
an
increase
in steel
can calculate K1 as one obtains
stress of 30 N/mm per minute.

K ( c s ) =

g c c h

10

1
1
Frame
anchored on test floor

s + 0.22 s G

0.188

10

g c c h

(6)

The material parameters kcvg and ksvg and g1 can


be calibrated by fitting experimental data relevant to
free (evaporable) water content in concrete at
various ages (Di Luzio & Cusatis 2009b).
Hydraulic jack

2.2 Temperature evolution


Note that, at early age, since the chemical reactions
associated with cement hydration and SF reaction
are exothermic, the temperature field is not uniform
for non-adiabatic systems even if the environmental
temperature is constant. Heat conduction can be
described in concrete, at least for temperature not
exceeding 100C (Baant & Kaplan 1996), by
Fouriers law, which reads
Load cell

Spreader steel profile

Steel collar with LVDT's

Beam Test Specimen

Support

= T

(7)

Figure 2. Test set-up for beam test specimen.

where q is the heat flux, T is the absolute


temperature,
and istypes,
the heat
this
For all specimen
the conductivity;
actuator wasinposi-

tioned in the centre of the specimen and the total

load P was transferred by means of a spreader steel


profile to each half-beam (Fig. 2). A load cell measured the load applied to the specimen during the test.
The slip of the bar, at its free end, was recorded
using 3 linear variable differential transducers (LVDT)
on both sides of the specimen.
Loading continued until the slip at one end of the
specimen reached 3 mm. For the half-beam with
3 mm slip the bar was fixed in a clamping device so
that the test could be continued without further slip
at this side of the specimen. Loading continued until
the slip at the second half of the specimen exceeded
3 mm as well.
2.4 Test results
From the obtained test results, values of the bond
stress along the surface of the bonded reinforcing
bar can be derived. The formulas to calculate the total force acting in the reinforcing bar depends on the
geometry of the specimen used.
The mean bond stress can be calculated by assuming the force Fs in the reinforcing bar to be transferred to the concrete in the cylindrical zone of the
embedment length ld:

P
As

32.0
28.0
24.0

(1)

by writing ld as k. and s the tensile stress in the


reinforcing bar. This stress is, as mentioned earlier, a
function of the applied total load P and the geometry
of the specimen.
s = .

that the variation in time of the water mas

Table 2. Test results volume


for beam tests
with bond(water
length of
5.
of concrete
content
w) be eq
M divergence
DEV of the
R moisture
DEVflux J
[N/mm] [N/mm]
[N/mm] [N/mm]
SCC1-12
18.13
0.99
27.82
3.17
SCC1-20
14.94 w =0.77
24.07
1.84
J
SCC1-25
12.80 t 0.81
19.39
1.27
SCC1-32
11.24
0.59
20.49
1.07
SCC1-40
9.71
0.55
19.86
0.93
The1.47
water content
w can
be expressed a
SCC2-12
15.77
25.70
2.93
of
the
evaporable
water
w
e (capillary wa
SCC2-20
13.31
0.25
21.54
1.56
and adsorbed
SCC2-25
12.10vapor,0.29
18.60 water)
2.03and the non-e
bound)
water
wn (Mil
SCC2-32
10.65(chemically
0.28
19.77
0.85
SCC2-40
8.81 Pantazopoulo
0.29
0.44
&17.48
Mills 1995).
It is reas
CVC1-12
13.45assume
0.73that the 19.88
0.75water is a fu
evaporable
CVC1-20
12.96relative
0.53humidity,
19.46
0.82 of hydration
h, degree
CVC1-25
11.14
0.94
16.28
1.50
of silica fume
CVC1-32
9.67 degree0.28
18.10 reaction,
1.00 s, i.e. we=w
=
age-dependent
sorption/desorption
CVC1-40
8.13
1.22
16.61
2.09

(2)

The factor can be determined from the specimens dimensions and has a value of 1.25 for specimen type I (diameter 12 mm), 1.50 for specimen
type II (diameters 20 and 25 mm) and 1.75 for
specimen type III (diameters 32 and 40 mm).
Two values are of major interest: the ultimate
bond strength R and the so-called characteristic
bond strength M. The ultimate bond strength is defined as the bond stress corresponding to the ultimate load recorded during testing. The characteristic
bond strength is calculated as the mean value of the
bond stresses corresponding to a slip of 0.01 mm,
0.10 mm and 1.00 mm. Both values differ for the
two halves of the specimen.
2.4.1 Influence of the concrete type
The main goal of the study is to compare bond
strengths for self-compacting concrete with those for
conventional vibrated concrete. In figure 3 one of

Bond Strength [N/mm]

Fs

d =
= s
l d . . 4.k

D (h, T )slip
h curves (mean of 4
the recorded bondJ =stress
measurements) is plotted for bar diameter 20 mm
and different concrete
In coefficient
table 2 the D(h,T)
Thecompositions.
proportionality
values of the characteristic
bond
strength
andit the
moisture permeability and
is aul-nonlinea
timate bond strength
arerelative
given for
all specimen
of the
humidity
h and types
temperature
as well as the standard
deviation
(DEV).
& Najjar 1972). The moisture mass balanc

20.0

(Norling Mjonell 1997). Under this assum


by substituting Equation 1 into Equati
Diam. 20 mm
obtains

w h
e + ( D h) = we
h

h t

w
&c + e &s + w

c
s

where we/h is the slope of the sorption/


isotherm (also called moisture capac
governing equation (Equation 3) must be
12.0
by appropriate boundary and initial conditi
The relation between the amount of e
8.0
water and relative humidity
is called
SCC1
SCC2
isotherm
if
measured
with
increasing
4.0
humidity and desorptionCVC1
isotherm in th
case. Neglecting their difference (Xi et al.
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
the following,
sorption
isotherm
will be
Slip [mm]
reference to both sorption and desorption c
Figure 3. Bond stress slip diagram for bar diameters 20 mm.
By the way, if the hysteresis of the
isotherm
would
be taken intoforaccount,
two
Comparing therelation,
differentevaporable
types of concrete
the humi
water
vs
relative
same bar diameter,
CVC1
and SCC2
(which
have
be used
according
to the
sign
of
the varia
almost the same compressive
strength)
have
comparelativity
humidity.
The
shape
of the
rable values for the
characteristic
bond
stress

M, exisotherm
for HPC
is influenced
by many p
cept for bar diameters
12 mm
for which
a significant
especially
those
that
influence
extent
and
difference between
the 2 concretes
isand,
noticed.
The determ
chemical
reactions
in
turn,
difference for thestructure
ultimateand
bond
R is some- (waterporestress
size
distribution
what larger. For ratio,
all tests
on
SCC2,

is
above
the
R
cement chemical
composition,
SF
ultimate bond stress
of
CVC1.
curing
time
and
method,
temperature,
mix
When the bond
stress-slip
ofvarious
the differetc.).
In the relations
literature
formulatio
ent concrete typesfound
are plotted
for
tests
on
specimen
to
describe
the
sorption
isotherm
with a reinforcingconcrete
bar of the
same
diameter,
it
can
be in th
(XiofetSCC1
al. 1994).
However,
seen that the bond
strength
is
larger
than
paper
the (as
semi-empirical
expression
pro
those of SCC2 and
CVC1Mjornell
was expected
dueadopted
to
Norling
(1997)
is
b
the higher compressive strength) at all stress levels,
16.0

Proceedings of FraMCoS-7, May 23-28, 2010

J = D ( h, Tin)ah steeper curve. For bar diameters(1)


resulting
of
40 mm the curves for SCC2 and CVC1 are almost
identical
for small slip values,
whileD(h,T)
the bond
stress
The proportionality
coefficient
is called
level
for
SCC1
for
the
same
slip
is
higher.
For
all
moisture permeability and it is a nonlinear function
other
the bond
for the
SCC2
of the diameters,
relative humidity
h andstresses
temperature
T (Baant
specimens
are
higher
for
the
same
slip
as
recorded
& Najjar 1972). The moisture mass balance requires
for
specimens.
In the
some
casesmass
the per
stresses
thatthe
theCVC1
variation
in time of
water
unit
even
approach
the
values
for
SCC1.
volume of concrete (water content w) be equal to the
divergence of the moisture flux J
2.4.2 Influence of the bar diameter
When
the results of all test are compared, it can be
(2)a
w that
= an
J increase in the bar diameter results in
seen
t
decrease of M and R. As the concrete strength influences
the content
bond properties
of the concrete,
the
The water
w can be expressed
as the sum
bond
stress
is
normalised
by
the
root
of
the
compresof the evaporable water we (capillary water, water
sive
strength:
vapor,
and adsorbed water) and the non-evaporable

(chemically bound) water wn (Mills 1966,

Pantazopoulo
& Mills 1995). It is reasonable to
R ,n = R
assume that
the
evaporable water is a function of
fc
(3)
relative humidity, h, degree of hydration, c, and
degree of silica fume reaction, s, i.e. we=we(h,c,s)
Figure 4, R,n is sorption/desorption
plotted for all concrete
mixes
= Inage-dependent
isotherm
and
tested
bar
diameters.
(Norling Mjonell 1997). Under this assumption and
differences
in the normalized
ultimate2bond
by The
substituting
Equation
1 into Equation
one
strength
obtains for the conventional vibrated concrete and
the self-compacting concrete is largest for bar diameters of 12 mm. The difference becomes smaller
w
w
w h
e but
e + bar
for
the e results
&
&s + w&nfor self ( Ddiameters,
h ) =
higher
(3)
c +
h

h
t
compacting concrete are chigher ins all cases. There
are no significant differences between the normalized
ultimate
SCC1
and SCC2, exisa 40
thestrength
slope
ofofthe
sorption/desorption
wherein
we/hbond
cept
case
of
mm
reinforcement
bar.
isotherm
(also
called
moisture
capacity).
The
By increasing
the(Equation
bar diameter,
thebeslip
at ultigoverning
equation
3)
must
completed
mate
bond stress
su is increasing
in all
cases. No sigby appropriate
boundary
andnoticed
initial
conditions.
nificant
difference
can
be
between
the reThe
relation
between concrete
the amount
of evaporable
sults
for
self-compacting
and
the
results
for
water
and
relative
humidity
is
called
adsorption
conventional
concrete.
isotherm if vibrated
measured
with increasing relativity

humidity and desorption isotherm in the opposite


case. Neglecting their difference (Xi et al. 1994), in
the following, sorption isotherm will be used with
reference to both sorption and desorption conditions.
By the way, if the hysteresis of the moisture
isotherm would be taken into account, two different
relation, evaporable water vs relative humidity, must
be used according to the sign of the variation of the
relativity humidity. The shape of the sorption
isotherm for HPC is influenced by many parameters,
especially those that influence extent and rate of the
chemical reactions and, in turn, determine pore
structure and pore size distribution (water-to-cement
ratio, cement chemical composition, SF content,
curing time and method, temperature, mix additives,
etc.). In the literature various formulations can be
Figure
bond strength
for different
difound 4.toNormalized
describe ultimate
the sorption
isotherm
of normal
ameters
and
concrete
compositions.
concrete (Xi et al. 1994). However, in the present
paper the semi-empirical expression proposed by
Norling Mjornell (1997) is adopted because it
Proceedings of FraMCoS-7, May 23-28, 2010

explicitly
accounts
2.4.3
Crack
pattern for the evolution of hydration
reaction
andtheSFcrack
content.
This
sorptiononisotherm
After
testing
pattern
appearing
the surreadsof the specimen was recorded. No big differface
ences between the crack pattern for CVC1 and the
self-compacting concretes
no SCC1 and SCC2 were

ticed. In almost all cases the crack 1pattern was


lim
( , ) 1
we (hto, the
+
the rec , zone
s ) = Gin
c s actual
ited
bond
between
1 which

10(g

c c )h so along
inforcing bar and the concrete
e was1 possible,

(4)
the bonding length.

)h
(g crack
patterns
An example of one of 10
the
c
1 c
K1 (inc ,Figure
) e 5.
1 of the
specimen is given
s

where the first term (gel isotherm) represents the


physically bound (adsorbed) water and the second
term (capillary isotherm) represents the capillary
water. This expression is valid only for low content
of SF. The coefficient G1 represents the amount of
water per unit volume held in the gel pores at 100%
relative humidity, and it can be expressed (Norling
Mjornell 1997) as
Figure 5. Crack cpattern forsone of the specimen of CVC1 with
G (12 ,mm
)diameter.
=k c+k s
(5)
bars
1

c s

vg c

vg s

are material parameters. From the


k vg and k vgMODELS
3where
PREDICTION
c

maximum amount of water per unit volume that can


fillliterature
all pores (both
capillary
and gelthe
pores),
one
In
a lot of
modelspores
to predict
ultimate
as
one
obtains
can
calculate
K
1
bond strength, corresponding slip and equations to
describe the bond stress-slip behaviour can be found.

In these models several parameters


g c as
bar
h di such

e
w cover,
s + concrete
s G compressive

ameter, concrete
strength,
c
s

are incorporated.
(6)

K ( c s ) =
All equations have been
by
linear
or

established
g h
c
c

non-linear regression eon test results with varying parameters, but mostly for conventional vibrated concretes
compressive
strengths
beg1 can
Thewith
material
parameters
kcvg andinksvgtheandrange
tween
20 and by
50 fitting
MPa. Few
tests havedata
been
done on
be calibrated
experimental
relevant
to
high
concretes
compressive
strengths
free strength
(evaporable)
waterwithcontent
in concrete
at
above
MPa.
The compressive
strengths of the
various60ages
(Di Luzio
& Cusatis 2009b).
concretes used in this research project are all around
60 MPa or higher for the SCC1 mixture.
10

0.188

0.22

10

2.2 Temperature evolution


Note that, at early age, since the chemical reactions
3.1 Bond stress slip relation
associated with cement hydration and SF reaction
A
for thethe
bond
stress- slip
is rearemodel
exothermic,
temperature
fieldbehaviour
is not uniform
quired
to be able systems
to make even
calculations
of the crack
for non-adiabatic
if the environmental
pattern,
crackiswidths,
.Heat
Therefore
a relationship
temperature
constant.
conduction
can be
has
been proposed
in theatCeb-Fib
Code 1990
described
in concrete,
least forModel
temperature
not
(Fig.
6). It consists
of an increasing
first 1996),
branch up
exceeding
100C (Baant
& Kaplan
by
to
the ultimate
bond reads
stress. This branch is followed
Fouriers
law, which
by a plateau during which slip is increasing for constant
to
q = bond
T stress, after which bond stress starts(7)
decrease for increasing slip values. Finally a constant
residual
strength
is reached
whichabsolute
is due
where
q is bond
the
heat
flux,
T is the
to
pure friction
reinforcing
bar with
the
temperature,
andbetween
is thetheheat
conductivity;
in this
cracked concrete lugs and the surrounding concrete:

J = D ( h, T values
) h
the range of the measured
for larger bar diameters.
The bar diameter The
has also
an influencecoefficient
on the slip D(h,T)
proportionality
corresponding with
the
ultimate
bond
strength
moisture permeability and it as
is can
a nonlinea
be seen on the graphs.
Both models
however
have
of the relative
humidity
h and temperature
fixed values of this
slip value
regardless
the barmass
di- balanc
& Najjar
1972).
The moisture
ameter.
that the variation in time of the water mas

volume of concrete (water content w) be eq


divergence of the moisture flux
J
SCC1

32.0
28.0

Figure 6. Prediction model for the bond stress slip relationship according to MC90 and Huang et al.

= 1

0 s s1
s1 < s s 2

s s2

= 1 ( 1 3 ).

s3 s 2
=3

(4)
(5)

s 2 < s s3
s3 < s

(6)
(7)

Table 3. Values for the prediction equations according to


Ceb-Fip MC90 (1999) for good bond conditions.
Confined concrete
Unconfined concrete
s1
1.0 mm
0.6 mm
s2
3.0 mm
0.6 mm
s3
distance betw. ribs
1.0 mm

0.4
0.4
1
2.5 f c
2.0 f c
3
0.4 1
0.15 1
Table 4. Values for the prediction equations according to
Huang et al. (1996) for good bond conditions.
High strength concrete Normal strength concrete
0.5 mm
1.5 mm
distance betw. ribs
0.3
0.4. fcm
0.4 1

16.0
12.0

Ceb-Fip MC90
Huang
12 mm
20 mm
25 mm
32 mm
40 mm

8.0
4.0

The parameters in this model have been prescribed in the code for confined and unconfined
normal strength concrete with good or other bond
conditions. Huang et al. (1993) proposed values for
the parameters for normal and high strength concrete
under good bond conditions (Table 3 and 4).

s1
s2
s3

1
3

20.0

The water content w can be expressed a


of the evaporable water we (capillary wa
vapor, and adsorbed water) and the non-e
(chemically bound) water wn (Mil
Pantazopoulo & Mills 1995). It is reas
assume that the evaporable water is a fu
relative humidity, h, degree of hydration
degree of silica fume reaction, s, i.e. we=w
= of measured
age-dependent
Figure 7. Comparison
bond stresssorption/desorption
slip behaviour
(Norling
Mjonell
1997). Under this assum
for SCC1 with prediction models.
by substituting Equation 1 into Equati
obtains

0.6 mm
0.6 mm
1.0 mm
0.4
0.4. fcm
0.4 1

Both models (CEB-Fip confined concrete and


Huang high strength concrete) are compared with
the measured bond stress-slip behaviour of the
specimen (Figs. 7 to 9). It can be seen that the predicted values calculated with the Huang-model for
high strength concrete overestimate the bond
strength in all cases. For the MC90-model, the values are underestimated for small diameters and in

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

Slip [mm]

32.0

SCC2

28.0

24.0
Bond Strength [N/mm]

for

Bond Strength [N/mm]

w h
e + ( D h) = we
h

h t

w
&c + e &s + w

c
s

20.0

where we/h is the slope of the sorption/


isotherm (also called moisture capac
governing equation (Equation 3) must be
by appropriate boundary and initial conditi
The relation between the amount of e
water and relative humidity is called
isotherm if measured with increasing
humidity and desorption isotherm in th
Figure 8. Comparison
of measured
bond stress
behaviour
case.
Neglecting
their slip
difference
(Xi et al.
for SCC2 with prediction models.
the following, sorption isotherm will be
reference to both sorption and desorption c
By the way, if the hysteresis
of the
CVC1
isotherm would be taken into account, two
relation, evaporable water vs relative humi
be used according to the sign of the varia
relativity humidity. The shape of the
isotherm for HPC is influenced by many p
especially those that influence extent and
chemical reactions and, in turn, determ
structure and pore size distribution (waterratio, cement chemical composition, SF
curing time and method, temperature, mix
etc.). In the literature various formulatio
found to describe the sorption isotherm
concrete
(Xibond
et al.
1994).
However, in th
Figure 9. Comparison
of measured
stress
slip behaviour
for CVC1 with prediction
papermodels.
the semi-empirical expression pro
Norling Mjornell (1997) is adopted b
16.0
12.0

Ceb-Fip MC90
Huang
12 mm
20 mm
25 mm
32 mm
40 mm

8.0
4.0

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

Slip [mm]

32.0
28.0
24.0

Bond Strength [N/mm]

s
= 1.
s1

= J

24.0

20.0
16.0
12.0

Ceb-Fip MC90
Huang
12 mm
20 mm
25 mm
32 mm
40 mm

8.0
4.0

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

Slip [mm]

Proceedings of FraMCoS-7, May 23-28, 2010

moisture permeability and it is a nonlinear function


of theMaximum
relative humidity
h and temperature T (Baant
3.2
bond strength
& Najjar 1972). The moisture mass balance requires
To
a better in
prediction
of water
the ultimate
that get
the variation
time of the
mass perbond
unit
strength
more
sophisticated
models
are
necessary.
volume of concrete (water content w) be equal
to the
Some
of the
to predict
divergence
of models
the moisture
flux J the ultimate bond

strength are show in table 5 with the used unit system.


Key parameters in these equations are the ratio
w
(2)
the
= J
of
t concrete cover c to the bar diameter , the ratio of the bar diameter to the bond length ld and the
rootThe
of the
compressive
fc.
water
content w strength
can be expressed
as the sum

of the evaporable water we (capillary water, water


vapor, and adsorbed water) and the non-evaporable
Table 5. Prediction models for bond strength.
(chemically
bound) water wn (Mills 1966,
Author
Equation
Units
Pantazopoulo
& Mills 1995). fIt is reasonable
to
Eligehausen
R = 0.75. c .
SI
assume
that
the
evaporable
water
is
a
function
of
Esfahani*
R = 4.73.[ (c ) +0.5]/[ (c ) +5.5]. f
SI
relative
h, degree
of hydration,
Psi
Harajli humidity,
R = [1.2+3.
(c ) +50.
( l d ) ]. f c, and
Huang**of silicaRfume
=0.45.freaction,
s)
degree
s, i.e. we=we(h,c,SI
cm
(c ) +53. ( l d ) ].isotherm
f
R = [1.22+3.23.
Psi
=Orangun
age-dependent
sorption/desorption
f
SI
MC 90
R =2.5.
(Norling
Mjonell
1997).
Under this assumption and
* Equation is valid for fc > 50 MPa
by
substituting
1 60into
Equation
2 one
** Equation
is valid Equation
for fc between
and 120
MPa
obtains
c

ck

In figures 10 to 12 a comparison is made between


w
thewobtained
andwthe
for the
e &predicted
e & +ones
e h + values
&
w
+
( D h ) =

(3)
c
s
n
different
h mixesin a bond
stress versus bar
h t concrete
c
s
diameter diagram. It shows an underestimation of
the ultimate bond strength in all cases, except for
is the(1996)
slope ofwhich
the sorption/desorption
where
we/h
the
Huang
model
is developed for
isotherm
(also
called
moisture
capacity).
high strength concrete. The predicted
valueThe
is
governing
equation
(Equation
3)
must
be
completed
sometimes only 40% of the recorded one. Some
by appropriate
boundary
and initial
conditions. from
models
give bond
strengths
independently
relation between
the amount
of evaporable
the The
bar diameter,
others seems
to predict
the trend
water
and
relative
humidity
is
called
adsorption
quite good but give values that are too low, e.g.
isotherm
with
increasing
relativity
the
models if
by measured
(Orangun et
al. 1977;
Harajli
1994).
humidity
and
desorption
isotherm
in
the
opposite
Therefore the coefficients out of the Oranguncase. Neglecting
difference (Xi
al. 1994),
in
model
have beentheir
re-determined
by et
linear
regresthe
following,
sorption
isotherm
will
be
used
with
sion based on the obtained results.

R [N/mm]

reference to both sorption and desorption conditions.


By the way, if the hysteresis of the moisture
isotherm
would be taken into account, two different
30,0
relation,
evaporable water vs relative humidity,
SCC1 must
Eligehausen
25,0 according to the sign of the variation
be used
of the
Esfahani
relativity
humidity.
The
shape
of
the
sorption
20,0
Harajli
isotherm for HPC is influenced by many parameters,
Huang
15,0
especially
those that influence extent and rateOrangun
of the
MC90 pore
chemical
reactions
and,
in
turn,
determine
10,0
structure and pore size distribution (water-to-cement
ratio,5,0cement chemical composition, SF content,
curing0,0time and method, temperature, mix additives,
etc.). In 5 the literature
can be
15
25various
35 formulations
45
found to describe the [mm]
sorption isotherm of normal
Figure
10. Bond
versus bar
diameter in
comparison
of
concrete
(Xi etstrength
al. 1994).
However,
the present
models
and
test
results
for
SCC1.
paper the semi-empirical expression proposed by
Norling Mjornell (1997) is adopted because it
Proceedings of FraMCoS-7, May 23-28, 2010

20,0

Harajli

15,0

we (h10,0
c s ) =
,

5,0
0,0

G1 ( c , s )1

(g
c
e
)h

(g
c
c

K15 ( c 25s ) e 35
45

[mm]
10

Huang
Orangun
+
MC90
c )h

10

(4)

Figure 11. Bond strength versus bar diameter comparison of


models and test results for SCC2.

where the first term (gel isotherm) represents the


physically bound (adsorbed) water and the second
term30,0(capillary isotherm) represents the capillary
CVC1
water.
content
25,0This expression is valid only for lowEligehausen
Esfahani
of SF. The coefficient G1 represents the amount of
Harajli
water20,0per unit volume held in the gel pores at
100%
Huang
15,0 humidity, and it can be expressed (Norling
relative
Orangun
Mjornell
1997) as
MC90
10,0
5,0
c c+ ks s
G ( c s ) = k vg
c vg s
0,0

(5)

15

25

35

45

material parameters. From the


where kcvg and ksvg are [mm]
Figure
12.
Bond
strength
versus
comparison
of
maximum amount of water bar
perdiameter
unit volume
that can
models and test results for CVC1.
fill all pores (both capillary pores and gel pores), one
onethe
obtains
canDue
calculate
K1 asthat
to the fact
bond length has been kept
constant in these test, the term with ld becomes con
both
h self gfor
stant. The determined coefficients

w
s + s G e c
c weres almost
compacting concretes
identical,
but
the

(6)

coefficients
for
the
conventional
vibrated
concrete
K ( c s ) =

differed.
The new equations
hgiven in Table 6.
g are
c
c

The good correlation ebetween the


predicted and the
measured bond strengths can cbe noticed
in Figure 13
parameters k vg and ksvg and g1 can
(forThe
SCCmaterial
mixtures).
10

0.188

0.22

10

be calibrated by fitting experimental data relevant to


free 6.(evaporable)
watermodels
content
concrete at
Table
Modified prediction
for bondinstrength.
various type
ages (DiEquation
Luzio & Cusatis 2009b). Units
Concrete
SCC
CVC
2.2

R = [1.77+0.49. (c ) ]. f cm

SI

fcm

SI

= [1.87+0.35. (c ) ].
TemperatureR evolution

Note that, 30,0


at early age, since the chemical reactions
associated with cement hydration and SFSCC1reaction
are exothermic, the temperature field is notSCC2uniform
for non-adiabatic
systems even if the environmental
25,0
temperature is constant. Heat conduction can be
described in concrete, at least for temperature not
exceeding 20,0100C (Baant & Kaplan 1996), by
Fouriers law, which reads
R,meas [N/mm]

should be modified to take in to account the effect of


the The
bar diameter.
proportionality coefficient D(h,T) is called

explicitly
accounts for the evolution of hydration
30,0
SCC2
reaction and SF content. This sorption isotherm
Eligehausen
25,0
reads
Esfahani
R [N/mm]

=Out
D (of
h, Tthis
)h it can be concluded that the models
(1)

R [N/mm]

= T

15,0
15,0

where

20,0

25,0
30,0
R,pred [N/mm]

(7)

is the heat flux, T is the absolute

Figure 13. Correlation between measured values and predicted


temperature,
andthe modified
is the heat
conductivity; in this
values
for SCC with
equations.

As can be seen in Figure 7, the slip corresponding to


the moment of maximum bond strength is influenced
by the reinforcing bar diameter. For smaller diameters the maximum bond strength is larger but the slip
is smaller.
In the bond model out of MC90, no influence of
the bar diameter on the value of s1 is noticed, although it is clearly present in the experimental results. Most authors suggest a fixed value for s1,
while others suggest values that depend on the clear
rib spacing c0. Some of the values for confined reinforcing bars are summarized in Table 7.
Table 7. Prediction models for bond strength.
Author
Equation
Harajli
s1 = 0.15 c0
Huang
s1 = 1.0 mm (normal strength concrete)
s1 = 0.6 mm (high strength concrete)
MC 90
s1 = 1.0 mm
Oh
s1 = 1.04 mm

Units
SI
SI
SI
SI
SI

SCC1

2,5

SCC2
CVC1

s1 [mm]

2,0

Harajli

1,5

Huang (HSC)
Oh

1,0

MC90

0,5
0,0
5,000

10,000

15,000 20,000
c 0 [mm]

(8)

Theequation
proportionality
coefficient
The new proposed
is plotted
in Fig- D(h,T)
moisture
permeability
and
it
ure 14 together with the obtained test results. is a nonlinea
3,0
2,5
2,0
1,5
1,0

of the relative humidity h and temperature


& Najjar 1972). The moisture mass balanc
that the variation in time ofSCC1the water mas
SCC2
volume of concrete (water content
w) be eq
divergence of the moisture CVC1
flux
J
Prop. model
= J

w
t

0,5

The water content w can be expressed a


of the evaporable water we (capillary wa
5,000
10,000vapor,
15,000and20,000
25,000water) and the non-e
adsorbed
c [mm]
(chemically bound) water wn (Mil
Figure 14. Clear rib spacing versus slip s1 Comparison of test
Pantazopoulo & Mills 1995). It is reas
results with new proposed model.
assume that the evaporable water is a fu
h, degree(8)ofcan
hydration
The equations relative
of Tablehumidity,
6 and the equation
,
i.e. we=w
degree
of
silica
fume
reaction,

s
now be used in the MC90 model (equation 4 till 7)
=
age-dependent
sorption/desorption
to replace the expressions for 1 and s1.
(Norling Mjonell 1997). Under this assum
by substituting Equation 1 into Equati
obtains
4 CONCLUSIONS
0,0

A comparison between the predicted values for s1


and the experimental determined values is given in
Figure 13.
None of the models gives a good prediction of the
slip corresponding with the maximum bond strength.
The clear rib spacing, which is increasing for increasing bar diameters, is influencing the slip values.
No big differences are noticed between selfcompacting and conventional concrete (except for
bars diameter 40 mm), as well as no significant difference can be seen between the values for SCC1
compared to the values of SCC2 and CVC1. So it
can be concluded that the concrete type (SCC or
CVC) and the compressive strength of the concrete
does not have an influence on s1.
3,0

= D ( h , T ) h
s1 = c 0 .(0.0035c 0 + J0.006)

s1 [mm]

3.3 Slip corresponding to maximum bond strength

25,000

Figure 13. Clear rib spacing versus slip s1 comparison of


models and test results.

A regression analysis has been performed to determine a good relationship between the clear rib
spacing c0 of the tested reinforcing bar and the observed slip corresponding to the maximum bond
strength:

w concluw results,
Based on the obtained
the following
h
e & + we & + w
+ ( D h ) =
e
c
s
sions can be made: h t
h

c
s
a. The bond strength of self-compacting concrete
is as high as the bond strength for conventional vithe slope are
of the
sorption/
wherelarge
we/h
brated concrete when
barisdiameters
studisotherm
(also
called
moisture
ied. For smaller bar diameters, the bond strength of capac
governing
equation
(Equation
3) must be
SCC is slightly higher,
with
the largest
difference
by
appropriate
boundary
and
initial
conditi
occuring for the smallest bar diameters.
Theto relation
amount of e
b. For equal water
cement between
ratio the the
compreswater
and
relative
humidity
is called
sive strength of the powder-type self-compacting
isotherm
if
measured
with
increasing
concrete is higher (due to the limestone filler conhumidity
and
desorption
isotherm
tent), and so are the maximum and characteristic in th
bond strengths. case. Neglecting their difference (Xi et al.
the following,tosorption
isotherm
c. The slip corresponding
the maximum
bondwill be
reference
to
both
sorption
and
desorption c
strength is increasing for increasing bar diameters.

By the way, if the hysteresis of the


isotherm
would bethetaken
into account,
two
Considering the
bond models,
following
can
relation,
evaporable
water
vs
relative
humi
be concluded:
be used according
sign of
the varia
a. The bond stress-slip
model outtooftheMC90
does
relativity
humidity.
The
shape
not implement the influence of the bar diameter, re- of the
isotherm
for HPC is influenced
by many p
sulting in a model
that underestimates
the ultimate
especially
those
that
influence
extent
bond strength for small bar diameters and ap- and
chemical
reactions
in turn, determ
proaches the values
for R for
large barand,
diameters.
structure
and
pore
size
distribution
b. The bond stress slip model does not take into (waterratio,ofcement
composition,
SF
account the influence
the bar chemical
diameter on
the slip
curing
time
and
method,
temperature,
mix
corresponding with the ultimate bond strength.
In models
the literature
various formulatio
c. Almost all etc.).
existing
for predicting
the
found
to
describe
the
sorption
isotherm
ultimate bond strength underestimate the actual
(XiCVC
et al. 1994). However, in th
value for SCC as concrete
well as for
paper the semi-empirical expression pro
Norling Mjornell (1997) is adopted b
Proceedings of FraMCoS-7, May 23-28, 2010

=d. DAlmost
( h, T )hall models neglect the influence of the
(1)

clear rib spacing on the slip corresponding with the


ultimate
bond strength. coefficient D(h,T) is called
The proportionality
e.
A
modification
to the models
to
moisture permeabilitycan
andbeit made
is a nonlinear
function
get
a better
prediction
bondTstrength,
of the
relative
humidityofhthe
andultimate
temperature
(Baant
as
discussed.
& Najjar 1972). The moisture mass balance requires

that the variation in time of the water mass per unit


volume of concrete (water content w) be equal to the
REFERENCES
divergence of the moisture flux J
CEB-FIP. 1990. Model Code 1990. Comit Euro-International
w
(2)
du
=Bton,
J Lausane, France
t Y. & Chen Y. & Liu Y. 2003. Development of bond
Chan,
strength of reinforcement steel in self-consolidating concrete,
ACI Structural
(4)
The water
contentJournal,
w can 100
be expressed
as the sum
De
Schutter,
G. and Boel,
V. w
(eds)(capillary
2007, Proceedings
of the
water,
water
of 5th
theInternational
evaporable
water
e
RILEM Symposium on Self-Compacting
vapor,
and adsorbed water) and the non-evaporable
Concrete, RILEM Proceedings PRO54, RILEM Publica(chemically
bound) water wn (Mills 1966,
tions
Pantazopoulo
MillsP. &1995).
It P.is&reasonable
to
De
Schutter, G. &&Bartos,
Donome
Gibbs, J. 2008.
Self-Compacting
Caithness,
Publishingof
assume
that the Concrete,
evaporable
waterWhittles
is a function
Dehn,
& Holschemacher,
K. &of Weie
D. 2000.
and
relativeF. humidity,
h, degree
hydration,
c, SelfCompacting
Concrete
(SCC)
Time
Development
of
the Madegree
of
silica
fume
reaction,

s, i.e. we=we(h,c,s)
Properties and the Bond Behaviour, LACER, 5
= terial
age-dependent
sorption/desorption
isotherm
Desnerck,
P. & De Schutter,
G. & Taerwe, L. 2008.
Bond
(Norling
Mjonell
1997).
Under
this
assumption
and
Strength of Reinforcing Bars in Self-Compacting Concrete:
by Experimental
substitutingDetermination,
Equation 13rdinto
2 one
NorthEquation
American Conference on the Design and Use of Self-Consolidating Concrete
obtains
(SCC2008), Chicago, USA
Harajli, MH. 1994. Development/splice strength of reinforcing
w fiber reinforced
bars
we embedded
and
ACI
h + ( Dinplain
e we concrete,
wn
) =
Structural
h
(3)
c
s
Journal,
91
(5),
511-520
h

h t

&+

&+ &

where we/h is the slope of the sorption/desorption


isotherm (also called moisture capacity). The
governing equation (Equation 3) must be completed
by appropriate boundary and initial conditions.
The relation between the amount of evaporable
water and relative humidity is called adsorption
isotherm if measured with increasing relativity
humidity and desorption isotherm in the opposite
case. Neglecting their difference (Xi et al. 1994), in
the following, sorption isotherm will be used with
reference to both sorption and desorption conditions.
By the way, if the hysteresis of the moisture
isotherm would be taken into account, two different
relation, evaporable water vs relative humidity, must
be used according to the sign of the variation of the
relativity humidity. The shape of the sorption
isotherm for HPC is influenced by many parameters,
especially those that influence extent and rate of the
chemical reactions and, in turn, determine pore
structure and pore size distribution (water-to-cement
ratio, cement chemical composition, SF content,
curing time and method, temperature, mix additives,
etc.). In the literature various formulations can be
found to describe the sorption isotherm of normal
concrete (Xi et al. 1994). However, in the present
paper the semi-empirical expression proposed by
Norling Mjornell (1997) is adopted because it
Proceedings of FraMCoS-7, May 23-28, 2010

Harajli,
MH. accounts
2007. Numerical
bondevolution
analysis using
explicitly
for the
of experimenhydration
tally
derived
local
bond
laws:
A
powerful
method
for
reaction and SF content. This sorption isotherm
evaluating
the
bond
strength
of
steel
bars.
Journal
of
Strucreads
tural Engineering. 133 (5). 695-705
Huang, Z. & Engstrm, B. & Magnusson, J. 1996. Experimental investigation of the bond
of
and anchorage behaviour

deformed bars in high strength concrete.


Report
94:4,
1

(h, , University
) 1
weChalmers
+
Chalmers

c s ) = G1 ( c ,of sTechnology,

h (Pull-Out10(g
Martin, H. 2002. Bond Performance
of 1Ribbed

c Bars
c )and
e Composition

Consis(4)
Tests) Influence of Concrete
tency, Proceedings of the 3rd International
)h Symposium:

10(g
Bond in Concrete

from
Research
c
1 cto Standards,

K1 ( c , s ) e
1289-295,
Budapest

Menezes de Almeida Filho, F., et al. 2008. Bond-slipBehavior


of Self-Compacting Concrete and Vibrated Concrete using
Pull-out
and represents
Structures, 41,.
where
theand
firstBeam
termTests,
(gelMaterials
isotherm)
the
1073-1089.
physically
bound
(adsorbed)
water
and
the
second
Orangun, CO. & Jirsa, JO & Breen, JE. 1977. A reevaluation
term
(capillary
isotherm) length
represents
the ACI
capillary
of test
data on development
and splices.
Jourwater.
This 114-122.
expression is valid only for low content
nal. 74(3),
Oh,
B.H.The
& Kim,
S.H. 2007.GRealistic
Modelsthe
for amount
Local Bond
of
of SF.
coefficient
1 represents
Stress-slip
of
Reinforced
Concrete
under
Repeated
Loadwater
per
unit
volume
held
in
the
gel
pores
at
100%
ing. Journal of Structural Engineering. 133(2), 216-224.
relative
humidity, and it can be expressed (Norling
RILEM. 1973. Technical recommendations for the Testing and
Mjornell
1997) as Materials: RC6, Bond Test for ReinUse of Construction
forcing Steel: 2. Pull-out Test, Materials and Structures.
RILEM. 1973. Technical
recommendations
for the Testing and
c
s
( , ) = k c + k s
G1Use
(5)
c ofs Construction
vg c Materials:
vg s RC6, Bond Test for Reinforcing Steel: 1. Beam Test, Materials and Structures,
96-105.
c
and ksM.
material
parameters.
the
where
vg are
Zhu,
W.k&vgSonebi,
& Bartos,
P.J.M.
2004. BondFrom
and interfacial
properties
of
reinforcement
in
self-compacting
conmaximum amount of water per unit volume that can
Materials
andcapillary
structures,pores
37, 442-448
fillcrete,
all pores
(both
and gel pores), one

can calculate K1 as one obtains


w

K ( c s ) =
,

1
1

s + 0.22 s G

0.188

10

10

g c c h

g c c h
1

(6)

The material parameters kcvg and ksvg and g1 can


be calibrated by fitting experimental data relevant to
free (evaporable) water content in concrete at
various ages (Di Luzio & Cusatis 2009b).
2.2 Temperature evolution
Note that, at early age, since the chemical reactions
associated with cement hydration and SF reaction
are exothermic, the temperature field is not uniform
for non-adiabatic systems even if the environmental
temperature is constant. Heat conduction can be
described in concrete, at least for temperature not
exceeding 100C (Baant & Kaplan 1996), by
Fouriers law, which reads
q

= T

(7)

where q is the heat flux, T is the absolute


temperature, and is the heat conductivity; in this