Materials and Structures / Mat6riaux et Constructions, Vol.

36, August-September 2003, pp 439-447

Influence of confinement on high strength concrete

behavior

P. Sema Ros, S. A. Yazzar and A. Coca Calvo

Universidad Politecnica de Valencia, Spain

ABSTRACT R SUMs
An experimental program has been made in order to Un programme expOrimental a OtOdOvelopp? pour Omdier le
study the confined concrete behavior when its strength comportement ctu b&on confind avec une teneur en armature
changes ti'om traditional values to high strength values and transversale de 0% h 4%. L'&ude analyse des bdtons de
with confinement levels ranging from 0% to values higher rdsistance entre 25 MPa et 100 MPa. D ~ttres variables ont dtd
than 4%. The specimen shape and size have also been la forme et la dimension des @rouw~ttes.
included as variables. Avec les r~sultats expdrimentat~v obtenus on a ,~alis~ une
With the data experimentally achieved, the parameters analyse statistique pour d~finir les parambtres de la courbe
that define the stress-strain curve for concrete were r - ddformation avec une bonne corrdlation.
adjusted using a statistical methodology that gives us On a proposd un module de courbe Contrainte- d~]brmation
suitable approximation levels. qui montre le comportement du b&on h hautes pet?jbrmances
A stress-strain curve model is proposed, which lets us conjinO jusqu'3 des niveaux de dOformation trbs OlevOs, et qui
know precisely the confined concrete behavior up to high permet l'analyse de la ductilitd cht matdriau.
strain levels and analyze the material ductility. Les performances obtenues grace au confinement ont dte;
The achieved improvement, thanks to the confinement, quantifides d'apr~s les param&res qui ddfinissent le
was quantified regarding the parameters that define the cornportement du b~tom en partic~dier de sa rd~ismnce, de la
concrete behavior, particularly the maximum strength, the ddjbrmation au pic et sa ductilitd.
strain at the peak and the ductility.

i~iiiiiiii,

1. INTRODUCTION 2. OBJECTIVES

High strength concrete offers great advantages in The main objective of this study is to determine the evolution
durability and building economy. However, the fragile in ductility of a confined concrete when its strength changes
behavior of this material is a drawback, particularly in high- from common values (25 MPa) to high ones (100 MPa).
risk seismic areas, since its ductility decreases as its high Ductility has been determined from the stress-strain
strength increases. Thus, high strength concrete is curve analysis of specimens with transversal reinforcement
significantly more fragile than usual strength concrete. The and tested in compression. At the same time, an analytical
confinement of concrete allows an improvement in the model was adapted for the confined concrete curve,
compressive strength of columns in seismic active areas depending on the parameters of this study, so that it could
because its fragility is reduced. be applied to high strain values.
Over the past few years, several studies on confined high Besides the concrete compressive strength, the following
strength concrete have been made. They have tried to variables have also been studied:
determine the influence of different variables on concrete 9 Confinement degree (transversal reinforcement volume
ductility comparing them to non-confined concrete behavior. ratio). It is defined as the ratio between the transversal
Meanwhile, analytical models have been researched in reinforcement steel volume and the confined concrete volume.
an attempt to find models that describe the stress-strain 9 Specimen shape: cylindrical and square-section
curve of confined concrete. Special emphasis has been laid prismatic specimens were used. In both cases slenderness
on the post-peak downward leg of the curve. was kept constant and equal to 2.

1359-5997/03 9 RILEM 439

Serna Ros, Yazzar, Coca Catvo

9 Specimen size: Specimens of 100, 150, and 200 mm side Casinos". Maximum size 12 ram.
or diameter were used. - Compacted microsilica, "Ferroatlfintica |
9 Last generation fluidizer additive (polycarboxylic
ether), with 20% of solid waste "Bettor |
3, E X P E R I M E N T A L P R O G R A M Table 2 summarizes the dosage of the concrete used.
A "B 500 S" steel of 8ram in diameter was used for the
Four different concrete dosages were made with a nominal confining of 400mm high specimens; the other specimens
compressive strength of 25, 50, 75 and 100 MPa were confined using 6mm diameter "B 500 S" steel. Every
respectively. Two mixes were made fbr each strength level. specimen has a longitudinal assembling frame made of four
Concrete was made in a horizontal axis mixer and the bars unitbrmly distribmed on its perimeter (Fig. 2). These
volume of each mix was of 280 liters. bars have the same diameter as the stirrups.
Table 1 shows a smnmary of the specimens made in each The specimens were kept in the laboratory for 24 hours,
mix. demoulded and then kept in a chamber with a relative
humidity of 95% and a
Table 1 - Characteristics of the specimens made in each mix temperature of 20~ ~ 2, until
....... Specimen ,Size Volumetric ratio of they reached the testing age (28
Concrete the transverse
pecimen No days).
Strength Diameter/Side Height reinforcement (9s)
Shape Specimen capping was done
[MPa] [mml [mml
1%! with sulfur 24 hours betbre the
100 200 0-2.06-3.43-4.12 4 test. The specimens were left two
O
ylindrical 150 300 0 - 0.95 - 1.90 - 2.54 4 hours in the open air in order to
i
200 400 0 - 0.95 - 1.90 - 2.53 4 avoid sharp variations in the
p...
i 100 200 0 - 2.53 - 3.38 - 4.22 4 temperature of the sulfhr layer.
C~
tt~ 'rismatic 150 300 0 - 1.04 - 2.08 - 2.78 4 Afterwards they were kept in a
i
tt3 200 400 0 - 1.02- 2.03- 2.71 4 damp chamber until the testing
resting 150 . 30o - .... : ................ , 4
time.
The tests on the specimens
The total number of specimens was 224. Fig. 1 shows made up of 25 and 50 MPa concrete were carried out with a
one o f the cylindrical specimen series ah'eady made.

Fig. 2 - Reinforcement layout.

Fig. 1 - Cylindrical specimens made up in each mix.

The materials used for manufacturing the concrete were: 250 Ton capacity press. In order to test the specimens made
9 Cement: Type CEM I 42.5 R/SR, sulfate resistant in up of 75 and 100 MPa concrete, a 500 Ton capacity press,
compliance with the Spanish Code, supplied by "Lafarge specially designed tbr this experiment, was used.
Asland" in Sagunto. Tests were carried out at a constant straining speed and the
9 Sand: Washed rolling limestone, supplied by "Aridos stress-strain curve for each specimen was obtained by means of
Casinos S i . " a computerized system of data research (Figs. 3 and 4).
9 Crushed coarse aggregate, also supplied by "Aridos The tests were caxfied out up to strain values over 4%, or in
the case of the confined specimens,
Table 2 - Dosages studied until three or more stirrups failed. Figs.
I

Strength Cement Sand [ Coarse Water Additive I S. F, 5 and 6 show the two confined
(kglm3) aggregate (kg/m3) (kg/m3) I (kg/m3) w/c
(Mea) (kg/m3) (kg]m3) spechnens series (cylindrical and
prismatic) after the test. The cylindrical
25 300 903 t030 2t0.0 --- 0.700
series have a different size and the
50 450 903 1030 180.0 4.5 --- 0.408 same confinement degree. The
75 405 903 1030 155.0 ! 9.0 45 0.360 prismatic ones are of the same size
wi'd~ a different confinement degree.
100 405 903 1030 122.4 15.8 45 0.300

440
A4aterials and Structures / Matdriaux et Constructions, Vol. 36, August-September 2003

Fig. 7 shows the results obtained for cylindrical

specimens of 150 x 300 ram, which were prepared with
tbur dil~rent types of concrete, and depicts four different
confinement degrees lbr each strength level.
It can be observed that when the confinement level rises,
the stress and strain at the peak of the curve increase and
the curves flatten. On the other hand, when the strength
increases the strain at the peak of the confined specimens is
lower and the curves are sharper. In this figure one can also
observe the moment when some of the stirrups fail.
Analyses of the achieved results lead to think that the
specimen size and shape do not influence the confined concrete
strength. This fhct is also noticed in the equations fbund.

Fig. 3 - Computerized system of data research. 4. P R O P O S E D M O D E L

Once the tests were finished, an analytic model that fitted

the experimentally achieved stress-strain curve was proposed.
The proposed model consists in the following
modification in Sargin's curve [34]:

For ~ < ~ o

o" = ~ o l + ( K : , - 2 ) e j

For ~ > e o
Kbej+(D-l)~ 13
Fig 4 - Data research system in the specimen. cy 0 I + ( K b _ 2 ) e ~ + D e l3

where:
K b _ E0E o

E
El=-
g0
where K b is greater than 1 and g 0 and ~ 0 are the
experimentally obtained stress and strain values at the peak.
E 0 is the elasticity modulus at the origin.
The equation that defines the upward leg of the curve is
the same as that proposed by Sargin [34] (1971), since it
fits well our tests results. Nevertheless, a modification of
Fig. 5 - Cylindrical specimens. "the equation that defines the downward leg of the curve was
set out. Both the numerator and the denominator of the
polynomial were increased in one degree in order to
improve the adjustment of this leg in high strength
concrete. This modification was necessary due to the high
concavity featured by this material.
For each test, a theoretical curve was determined with the
0 and e 0 experimental values. The non-dimensional D
parameter, which defines the downward leg of the concrete
stress-strain curve, was fixed in such a way that the
theoretical curve would cross the experimental point that
corresponds to 50 % of the maximum stress.
E o is obtained fi'om the tbrmula E o = 10150"~ ~
(MPa.) which had been tested befbre for concrete with
similar characteristics [6].
Fig 6 - Prismatic specimens.

441
Sema Ros, Ya~zzar, Coca Calvo

Cylinddccl S peamen 15x30 - 25 MPa Cylincticd S p e a r r e n 15x 30 - 50 MP a

i 2G, 2Q

o5
~6
?6

W s4
c~

s
co
48
48
=I .90

24

2 r~ g,95
$.....
o.5:x I ;N ~ 2L 2 o 3oc 4?0 45O ~
hO 58 ~'0 2{1~} 2~1 5'( 350 4s 45}
St~oin [%] StreWn ~'~]

Cylir'K~'iod S p e d n " e n 15x30 - 75 MPa Cytinc~iccl S I s e d m e n 1 5 x 3 0 - 1 0 0 M P a

2

ib

84. ~a

72

48

3(. ' 35
24
~-o,oo ~.. \ \
2~
12 r~ 0 , 9 5 \
o r O.90' ~ ' ~ r -:0.95
\
0 (xi rs) r fx? 50 2 [;0 2 ,} 3 [,8 5 ~,0 4o? 45;~
(~;0 I 0 2( 25'.! :} I;(3 35i 4Ll ![; ~X;
Stran [%]

Fig. 7 - Experimental stress-strain curves for [bur different strengths and four confinement levels.

Fig. 8 shows one of the 192 graphs that were plotted in 9 f~,,,: Concrete strength, calculated as the average strength
order to determine D, where both the experimental and the of all the non reinforced specimens in the same mix (MPa).
proposed theoretical model are depicted. It can also be 9 P~: Confinement degree: transversal reinforcement
observed that the theoretical model adjusts perfectly to the volume ratio (%).
experimental curve. 9 H: Specimen dimensions. The specimen height was taken
as the variable (m).
~00 -
9 @: Specimen shape. Cylindrical specimens were
90-
considered as "0" and prismatic ones as "1"
80- The relationship among these variables can be very
70- complex. For this reason, they were analyzed linearly up to
60- degree 3 (ps2; ofcrn2;H3; etc.), non-linearly (9~1'~; H1/3;e/Cm;etc.)
}~- and in combination among themselves (p~: * H; H / f~.,, etc.).
40-
The correlation coefficient (R2), which indicates the
30-
variance percentage of the mathematical model in relation with
20.
the experimental curve, was taken as reference for each of the
16-
statistical analyses. Thus, if a new variable is added to the
o,oo o3Q 020 o'9o ~.io ~,~o Go ~io z4o 25o'"'""";"o~ model and R 2 does not increase more than 0.5 %, this variable
S#d n [%)
is considered to have no influence on the parameter studied.
As a result of the analysis, a mathematical expression
Fig. 8 - Experimental stress-strain curve plotted together with
the proposed mathematical model. was obtained tbr each curve parameter (Table 3). These

Table 3 - Parameter adjustments of the proposed model

a 2
Parameter Equation
5. R E S U L T S A N D D I S C U S S I O N
Go [MPa] f~m (1 + 0.289 ~ ) 98.29 % !
5.1 P a r a m e t e r s a d j u s t m e n t m e t h o d
The study of the dependence of the proposed model ~o[%] (5.1 9s - 10.11 ~)/fem + 0.731 H 91.0 %
Confined Concrete
parameters (O" 0 , ~ 0 and D) with the analyzed variables
was carried out by means of a multivariable statistical ~01%] 0.2 + 0.00055 f~,,
analysis performed with the STATGRAPHICS Plus r Unconfined Concrete
computer program. The variables analyzed were:
D 1.09 ~ +(2.89-p~)8.35/f~m 98.3%

442
Materia& a n d & r z w m r e s / Matdriaux et Constructions, Vol. 36, August-September 2003

expressions include variables that are either statistically specimen (shape, size and confinement degree), and with the
significant or explain the analyzed property with enough values of dr o. s and D, determined in the equations
approximation.
shown in Table 3, a mathematical model that rePresents the
The equation that defines the maximum stress shows that
confined concrete stress-strain curve can be set.
confinement causes a non-lineal increase in the concrete
The data obtained in the analysis of four cylindrical
strength compared with the strength of non-confined
specimens with different confinement degrees and made of
concrete. It can also be noticed that both shape (fib) and
80 MPa concrete were compared with the curves proposed
height (H) are not significant variables. Thus, the
by the article 3.5 of the CEB - F1P 90 Model Code [5] for
confinement produced by round o1" right-angled stirrups
confined concrete as well as with the curves proposed by
leads to a similar increase in stren~h.
Diniz et at. [21], Mander et al. [27] and Cusson et al. [18].
The peak strain of the confined specimens increases with
The results of this comparison are shown in Fig. 9.
the confinement
Stress al the peak [Mpal Strd n at the peak [%]
degree and it is
lower in the 150 1.2 . ,/
prismatic 2o_ 1
specimens than 120 -~ "- /
in the cylindrical o.8 ..........zp
ogles. Thtls, the x
t.u
9o "

strain at the peak

of a prismatic 0 ....... 0.4 ~ 1 ~ ' r'~
specimen is 0.2 ...........
30 ~ ~"
equivalent to the 0./
one of another
cylindrical with a
0 /''* O 0.2 0,4 0,6 0,8
Mcd~
1 1,2
O 30 60 (20 120 150
transversal
Mc:c~
reinforcement
volume ratio Fig. 9 - Comparison of the experimental strain and stress at the peak with the theoretical model.
lower in a 2%.
It can be observed that the curves proposed in this work
This difference is more evident at lower values of concrete
featm'e the best adjustment to the tests. The model proposed by
strength. The value obtained for the variation coefficient in
Mander et al. [27] overestimates the concrete ductility, what
this case was about 17 %.
leads to an excessively flattened downward leg of the stress-
If we solve the equation tbr a non-confined concrete
strain curve. The curves proposed by Cusson [ 18] and Diniz
cylindrical specimen (p~ = 0 and fib = 0) of 150 mm in diameter
[2t] are also optimistic regarding the ductility estimates.
and 300 mm high, a maximum strain of 0.22 is found
However, the strength values reached with their models are
independently of the strength. This value coincides with the e~l
very' conservative regarding the experimental data.
value proposed by the CEB-FlP 90 Model Code [4] in its The Model Code [5], which limits its validity up to a
article 2.1.4.4. for non-confined concrete. Nevertheless, and in strength point of 85% of the maximum point in the downward
order to consider the influence of the concrete strength in these leg is also flat in comparison to the experimental curves.
cases, an independent adjustment of the results obtained for Figs. 10 to 13 show the comparison between the stress
non-confined concrete specimens has been made. The and strain values at the peak of our proposed model with
following equation, which only depends on the strength, was those calculated from the equations proposed by the
obtained:
atbrementioned authors.
It can be noticed that the strength and the strain at the
0 = 0.2 + 0.00055 f~m
peak predicted by our model are higher than those tbund lbr
The continuity among the proposed expressions for the the CEB - FIP 90 Model Code [5].
determination of the strain at the peak breaks in the cases in The confined concrete strength proposed by Cusson et al.
which the transversal reinforcement stops being effective for [18] adjusts very well to our results, whereas the strength
the improvement of this parameter. This happens in the cases proposed by Mander et aL [27], despite its good approximation,
of high strength concretes with yew little reintbrcement (less features a higher dispersion. The results proposed by Diniz et al.
than the 1%) or in the prismatic specimens with less than the [21 ] lead to sli~tly lower strength values.
2% of transversal reinforcement volume ratio. In these cases, In general terms, it can be observed that the results
the value obtained applying formulates it corresponding tbr present a higher dispersion when the strain at the peak is
concrete without confining is greater, and this is the one that predicted under the maximum load. It is also noticed that
should be considered. Mander et at. [27] results overestimate strain values at the
Parameter D increases with the degree of concrete peak of the curve, while Cusson et at. [18] only do it for
confinement, causing the flattening of the stress-strain low confinement levels. For high confinement levels the
curve and it decreases with its strength. Furthermore, it prediction turns out to be conservative. Diniz et al. [21]
does not depend on the specimen shape or dimension. results fit well with those achieved in this study.

5.2 Analytical model 5.3 Confined concrete ductility

From the compressive strength values of the non-confined Confined concrete ductility can be analyzed taking into
concrete as well as from the characteristics of the confined account several parameters that are used in the analysis of

443
Serna Ros, Yazzar, Coca Calvo

Experi~td P roFosed Mod~

140
140
120
120,
100
100
80
80
60
<o 60
40
40 84

20
20
0 ! ! r r 9 0
0,0 0.5 1,0 1,5 2,0 2,5 3.0 0,0 05 1.@ 1.5 2.El 25 3.0
Strdn{%}
Strdn (%)

Mander [27] Cusscn {~8]

140 140

120 120

1 O0 100

~" 80 8o.
.
~o 60'

40 [ 4.0"

2O
20.

0
0
0,0 0,5 1,0 1.5 2,0 2,5 3,0
0,0 0,5 1.0 15 2r0 2.5 30
Stfdn (%} S trcm [%}

Diniz [21]
140 140 '

120 120 '

IO0 1 O0 '

= 60 60"

40 40"

2O 20"
/

0
0,0 0,5 t0 1.5 2,0 2,5 3.0 0,O 05 1,0 1,5 2,0 2,5 3.0
Strdn(%} St~cln (%}

Fig. 10 - Comparison of the experimental results with the theoretical curves proposed by different authors.
the confined concrete stress-strain curve. In this study, the In this sense, another multivariable analysis has been
following parameters have been proposed: carried out in order to set the atbrementioned ductility
"ass: Strain at 85% of the maximum load in the indexes. The achieved results were as follows (see Table 4).
downward leg. (%). It can be seen from the analysis of the equations that
9 ss0: Strain at 50% of the maximum load in the every parameter that measures ductility increases with the
downward leg (%). confinement degree of the specimens. On the other hand,
9 A85: Total energy absorbed by the specimen at a strain concrete strength does not appear as a variable neither in
of 50% o f the maximum load in the downward leg. This the equations that determine energy (A85 and As0) nor in the
value is associated to the area under the stress-strain curve strain at high strain values (~50). This is due to the fact that
(cal / m3). the drop in the stress-strain curve after the peak, which the
9 ~s5 / ~0: Ratio between the strain at 85% of the proposed model entails, is compensated, at an energetic
maximum load and the strain at the peak. level, by the improvement in strength.

444
Materials and Structures /Mat6riaux et Constructions, Vol. 36, August-September 2003

9 The increase in strength reduces strain under a

Table 4 - Regressions for the ductility indexes maximum concrete load and causes a more fragile post-
peak behavior.
Parameter Equ ation R2 9 The increase in concrete confinement raises the
confined concrete strength and the strain under maximum
~ss [%] 15.35 ps/f~ - 0.07 ~P/H 85.4 % load is fbund to be higher. Apart from that, the curve in the
peak flattens and the decreasing branch slope becomes less
~so [%] 0.69 Ps 87.0 %
[ l
accentuated, what means that the concrete is more ductile.
~s.deo 1.69 4 D s - o.o05p~f~m 95.3 % 9 A minimum confinement level is needed in order to
make the stirrups effective. The spacing among stirrups
A~s [Cal/m 3] 5.24 p~ - 1.39 qb / H 85.8 % should be smaller than the specimen diameter (or width).
Otherwise, it should be considered as non-confined concrete.
Aso [Cal/m3] 11.51 ps -3.1 ~ t H 85.6 %
9 The specimen shape has an influence on the energy
dissipated during the specimen crushing and on the strain at
This efl~ct of post peak drop is detected by the parameter the peak. These variables are lower for prismatic
ess, which is the only parameter sensitive to concrete frailty specimens, which implies that these specimens have a
when its strength increases. The shape and size of the lower ductility. On the other hand, strength is not
specimens also affect the concrete ductility. In this way, the influenced by the specimen shape or size.
bigger the prismatic specimens the more fragile they are. 9 It has been proposed a theoretical behavior model that,

Strength of o o n f i n e d c o n c r e t e [ M P a ]
Strain at t h e peak [%]
t 4O
2.0
~120 Z
1.8
100 ,~ 1.6

1,4
8o
to 1,2
1.0
9 wD
0,8
4O
0.6
20 0.4 B Dm -... 9 9w

0.2 ~1
0
0.2 0,4 0.6 0,8 1,0 1,2 1.4 1.6 t.8 2,0
0 20 40 60 80 1O0 120 140
P r o p o s e d model P ropos ed rrodel

Fig. 11 - Regression between the results obtained with the proposed model and The CEB FIP 90 Model Code [4].
Strength of confined concrete~MPa] Strain at the peak [%]
140 2,0

1,8
120-
1,6
5 1130.
1,4
4~ 80, 12
8 8
m 1,0
N 60,
.N
c 0,8
40. ~5 r.~'../
0,6
20- 0,4

O ~ ; . ,,,,,,, ., 0,2 , J , i i i ,

O 20 40 60 80 100 120 t40 0.2 0,4 0,6 0.8 1.0 1,2 1.4 L6 1,8 2.0

P ropos ed model P topos ed model

Fig. t2 - Regression between the results obtained with the proposed model and Diniz's results [21].

contrary to other studied models, detects with precision,

both the behavior at maximum load and the post-peak
6. CONCLUSIONS behavior of high strength concrete.
9 The model proposed by CEB - FIP 90 Model Code [5]
The conclusions drawn after analyzing the proposed leads to conservative results when determining the concrete
stress-strain curves for confined concrete are: strength according to the confinement degree. It also
underestimates the strain at the peak. Concerning the strain

445
Serna Ros, Yazzar, Coca Calvo

S trengfh of ozntined concfete[MP a}

Stran at the peak [%]
140 2,0 ' ' " l ~
~" 1,8"
120.
1.6"
. ~ I a
100, G u
1,4'
I ,,m
0 t a#
."~ L0
~0.8"

2(?.
06. , N ~
0,4
Q (~,2 ~ , ...... , ....... , ii ii . ii i . ....
0 20 40 03 8 1GO 120 140 0.2 0,4 0.6 0.8 1.0 i,2 1,4 1,6 1,g 2,0
O
Propos c:~l rr'od~ Propos ed model

Fig. 13 - Regression between the results obtained with the proposed model and Mander's results [27].

Strengfh of confined concrete[MP aj Strain at the peak [%1

~40

D, 120

w
100 ~,'~~
S
S
8 6o LV
Y
O 40

20 84

20 40 643 80 ]CO 120 140 0,2 0.4 0,6 0,8 1,0 1,2 i,4 ),6 ],g 2,0

Proposed model Proposed model

Fig 14 - Regression between the results obtained with the proposed model and Cusson's results [ 18].

at 85% of the load, this code predicts values with enough [5] 'High performance concrete. Recommended extensions to
approximation, what leads to good results as high strength the Model Code 90 Research needs', Bulletin dTnformation
concrete ductility is estimated. CEB n~ (1995).
[6] Serna Ros, P., Yazzar, S. et Coca, A., 'Influence de la
9 Every parameter that measures the concrete ductility
granulomdtrie des granulats dans les caract6ristiques des
shows that this one increases according to the confinement b~tons h haute r6sistance', Hormigdn y Acero 213 (S6ville,
degree. Nevertheless, only the strain at 85% of the 1999) (in Spanish).
maximum load (eas) detects the strength increase eflbct in [7] Serna Ros, P., Yazzar, S. et Coca, A., 'Influence du
the material frailness. confinement sur la ductilit6 du b6ton/l haute r6sistance', l ~r
9 Toughness is hardly affected by the strength increase, Congr6s de l'Association Scientifique Technique du Bdton
as the decrease in strain is compensated by a higher stress at Structurel ACHE (S6ville, 1999) (in Spanish).
the peak, although it reduces the ratio between toughness [8] Yazzar, S., 'Etude de la ductilit~ d'616ments comprim6s
fabriqu6s avec b6ton ~ haute r6sistance soumis g charges
and strength.
cycliques de flexion', Thbse doctorale, Universit8
PolytechnJque de Valence (Valencia, 2000) (in Spanish).
[9] Recueil des normes UNE, 'Essais pour bdtons, mortiers et
REFERENCES ses composants', ToNe n~'5. S6rie 83 (in Spanish).
[10] Toralles, B., Agull6, L. et Gettu, R., 'Proc6s pour
[l] 'Code espagnot du b6ton structurel E.H.E.', (Madrid, 1999) l'optimisation des b6tons g haute rSsistance', tlormigdn y
(in Spanish). Acero 2 (1998) 19-30 (in Spanish).
[2] 'R6glement Argentin pour Constructions Civiles 'CIRSOC Ill] Alaejos Gutierrez, P. and Fern/mdez Cfinovas, M.,
201", (Buenos Aires, 1982) (in Spanish). 'Caractdrisation m~canique d'un b6ton /~ haute r6sistance',
[3] 'Commentary on building code requirements for reintbrced Hormigdn y Acero 191 (! 994) (in Spanish).
concrete (ACI 318-95)', ACt Committee, American [12J Ahmad, S.H. and Mallare, M.P., 'A comparative study of
Concrete Institute, (Detroit, 1995). models for confinement of concrete by circular spirals',
[4] 'Code Mod61e CEB - FIP 1990 pour b6ton structurel'. Magazine of Concrete Research 46 (166) (1994) 484-490.
G.E.H.O.C.E.B. (Madrid, 1995) (in Spanish).

446
Materials' and Structures' / Matdriaux et Constructions, Vol. 36;, August-September 2003

l13] Ahmad, S.tl and Shah, S.P., 'Stress-strain curves of concrete [28] Mander, J.B, Priestley, M.J.N. and Park R, 'Theoretical
confined by spiral reinforcement', ACljournal 79 (6) (1982) stress-strain model for confined concrete', Journal of
484-490. Structural Engineering, ASCE 114 (8) (1988) 1804- t 826.
[14] Balmer, C.G., 'Shearing strength of concrete under high [29] Nagashima, T., Sugano, S., Kimura, H. and Ichikawa, A,
triaxial stress-computation of Moorhr's envelope as a curve', 'Monotonic axial compression test on ultra high strength
Structural research report SP-23 (Denver, Colorado, 1949). concrete tied columns', Proceedings of 10th World Conference
[15] Bjerkeli, L., 'Deformation properties and ductility of high of Earthquake Engineering (Madrid, 1992) 2983-2988.
strength concrete', in 'Utilization of ItSC Symposium', SP - [30] Polat, M.B., 'Behavior of normal and high strength concrete
121 (ACI, Detroit, 1990) 215-238. under axial compression', M.A.Sc. Thesis, Department of
[16] Consid6re, A., 'Experimental Researches on Reinforced Civil Engineering (University of Toronto, 1992) 175.
Concrete' (McGraw Publishing Co., 1899). [31] Razvi, S.R. mad Saatcioglu, M., 'Strength and deformability
[17] Cusson, D. and Paultrc, P., 'High Strength concrete columns of confined high strength concrete columns', ACI Structural
confined by rectangular ties', Journal of Structural Journal 91 (6) (1994) 678-687.
Engineering. ASCE 120 (3) (1994) 783-804. [32] Roy, H.E. and Sozen, M.A., 'Ductility of concrete', Flexurat
[18] Cusson, D. and Paultre, P., 'Stress-strain model for confined mechanics of reinJbrcedconcrete, A. C.I. SP- 12, (1964) 213-224.
High-Strength concrete', Journal of Structural Engineering, [33] Saatcioglu, M. and Razvi, S.R., 'Strength and ductility of
ASCE 121 (3) (1995) 468-477. confined concrete', Journal of Structural Engineering, ASCE
[19] Cusson, D., de Larrard, F., Boulay, C. and Paultre, P., 'Strain 118 (6) (I 992) 1590-1607.
localization in confined high strength concrete columns', [34] Sargin, M., 'Stress-strain relationships tbr concrete and the
Journal of Structural Engineering, ASCE 122 (9) (1996) analysis of structural concrete sections', Solid Mechanics
1055-1061. Division, University of Waterloo, Study 4 (1971) 177.
[20] Chin, M.S., Mansur, M.A. and Wee, T.H., 'Effects of shape, [35] Scott, B.D., Park, R. and Priestly, M.J.N., 'Stress-strain
size and casting direction of specimens on stress-strain behavior of concrete confined by overlapping hoops at low
curves of high strength concrete', ACI Materials Journal 94 and high strain rates', A CI Journal, 79 (I) (1982) 13-27.
(3) (1997) 209-219. [36] Sheikh, S.A., 'A comparative study of confinement models'
[21] Diniz, S.M.C. and Franggopol, D.M., 'Strength and ductility ACIJourna179 (4) (1982) 296-306.
simulation of high-strength concrete columns', Journal of [37] Sheikh, S.A. and Toklucu, M.T., 'Reinforced concrete
Structural Engineering, ASCE 123 (10) (1997) 1365-1374. columns confined by circular spirals and hoops', ACI
[22] El-Dash, K.M. and Ahmad, S.H., 'A model for stress-strain Structural.lourna190 (5) (1993) 542-553.
relationship of rectangular confined normal and high-strength [38] Sheikh, S.A. and Uzmneri, S.M., 'Analytical model tbr
concrete columns', MateJ: Struct. 27 (1994) 572-579. concrete confinement in tied columns', Journal of Structural
[23] El-Dash, K.M. and Ahmad, S.H., 'A model for stress-strain Division, ASCE 108 (12) (1982) 2703-2722.
relationship of spirally confined normal and high-strength [39] Soliman, M.T. and Yu, C.W., 'The flexural strain relationship of
concrete columns', Magazine of Concrete Research 47 (171) concrete confined by rectangular transverse reinforcement',
(1995) 177-184. Magazine of Concrete Research 19 (61) (1967) 223-238.
[24] Fafitis, A. and Shah, S.P., 'Predictions of ultimate behavior [40] Vallenas, J., Bertero, B. and Popov, E., 'Concrete confined by
of confined columns subjected to large detbrmations', A.C.I. rectangular hoops and subjected to axial loads', Journal of
Journal 82 (4) (1985) 423 -433. Structural Engineering Repot~ N ~ UCB/EERC - 77/13,
[25] Hognestad, E.; 'Concrete stress distribution in ultimate stress Earthquake Engineering Research Center, Berkeley (1977) 114.
design', A.C.L Journal 52 (3) (1995) 455-480. [41 ] 'Intdgration de contraintes dans des sections de bdton soumises ~t
[26] Kent, D.C. and Park, R, 'Flexural members with confined flexo-compression ddvi6e', Revue Mdthodes Numdriques pour
concrete', Journal of Structural Division, ASCE 27 (7) Calcul et Dessin en Gdnie (Revista Me'todos Numdricos para
(1971) 1969-1990. Cc~lculoy Diseho en ingenie~4a)(1999) (in Spanish).
[27] Mander, J.B., Pfiestley, M.J.N. and Park, R., 'Observed
stress-strain model for confined concrete', Journal of
structural Engineering, ASCE 114 (8) (1988) 1827-1849.

Paper received." June 1, 200l; Paper accepted." May 17, 2002

44'7