Structural performance of self-consolidating concrete used ... · enhancing admixture (VEA), also...

Structural performance of self-consolidating concrete used in confined concrete columns

Patrick Paultre, Kamal Khayat, Daniel Cusson,

and Stephan Tremblay NRCC-46894

A version of this document is published in / Une version de ce document se trouve dans : ACI Structural Journal, v.102, no.4, July 2005, pp. 560-568

http://irc.nrc-cnrc.gc.ca/ircpubs

http://irc.nrc-cnrc.gc.ca/ircpubs

1

Structural Performance of Self-Consolidating Concrete

Used in Confined Concrete Columns

by Patrick Paultre, Kamal Khayat, Daniel Cusson, and Stephan Tremblay

Abstract: This paper compares the mechanical performance of highly-confined columns

cast with normal concrete (NC) vibrated into place to ensure proper filling and

consolidation equivalent to that of identical columns cast with self-consolidating concrete

(SCC). The tested columns had nominal concrete compressive strengths of 40 to 80 MPa.

Two confining stirrup configurations representing different degrees of confinement were

used. The confining stirrups had nominal steel yield strengths of 400 to 800 MPa.

A total of 16 columns were tested in this experimental investigation: 11 were cast either

with NC or SCC in reinforced sections; 5 accompanying columns were cast without

reinforcement. Three of the unreinforced columns were tested in uniaxial compression to

determine the overall concrete compressive strength of the large-scale columns, while

two others were cored to determine the distribution of the in situ compressive strength

and modulus of elasticity along column height.

The test results on reinforced columns showed that SCC yielded greater ductility,

although it developed slightly lower ultimate compressive strength than NC. The study

also confirmed that an increase in the stirrup yield strength can generate a high degree

of confinement in well confined concrete columns, provided that stirrup spacing is kept

small. The coring of unreinforced concrete columns demonstrated that the distribution of

in situ properties over the column height is more homogeneous in the case of SCC than

NC, which was also found adequate in general.

Keywords: Self consolidating concrete; vibrated concrete; confinement; strength;

ductility; rheology.

2

INTRODUCTION

Casting concrete in heavily reinforced sections, such as those in columns and beams in

moment-resisting frames in seismic areas and in some repair sections, makes the

placement of concrete quite difficult. Providing proper consolidation can require internal

or external vibration that can be critical in sections with high-density reinforcement.

Ensuring thorough consolidation of critical structures with durability and safety concerns

is essential and can often depend on the competence of the vibrating crew to ensure

adequate consolidation. Using standard vibration techniques with conventional concrete

that is not fluid enough may lead to some surface and structural defects resulting from

lack of proper bond development between the concrete and the reinforcement as well as

the entrapment of air voids in the concrete. Flowable concrete is normally used to reduce

labor cost and shorten construction time. Such concrete can have slump consistency close

to 200 mm to facilitate placement and consolidation. However, special attention should

be given to vibration consolidation of the plastic concrete in order to avoid segregation

and bleeding, which may further impair structural performance and surface quality.

One way to reduce the intensive labor demand for vibration of highly congested sections

is to use self-consolidating concrete (SCC). Such concrete can spread readily into place

and fill the formwork without any mechanical consolidation and with minimum risk of

separation of the material constituents. Such concrete is proportioned to exhibit a low

yield value and a moderate viscosity to maintain high deformability and filling capacity

of the formwork with minimum segregation and flow blockage.

Self-consolidating concrete has been used in a variety of projects in North America,

including precast/prestressed applications, repair of concrete infrastructure and, to some

extent, construction of reinforced concrete structures. In order for such concrete to have

wider acceptance for casting complex and congested structural elements, particularly in

seismic areas, more information regarding in situ properties of the hardened concrete

should be made available. For seismic construction requiring heavy reinforcement,

3

specifying engineers considering using this new category of high-performance concrete

must have adequate knowledge of the structural performance of elements cast with SCC.

Extensive research has been carried out at University of Sherbrooke since the early

1990’s to develop formulation expertise required to produce SCC with 28-d compressive

strength of 30 to 80 MPa with engineering properties comparable or superior to existing

technology. The work has focused particularly on SCC suitable for filling highly

congested and restricted sections, including repair applications as well as

precast/prestressed applications. The research involved testing the effect of SCC stability

on the homogeneity of in situ properties, including mechanical properties, bond strength

with embedded reinforcement and prestressing strands, microstructure characteristics,

and durability. The determination of the structural performance of highly-congested

columns cast with SCC with nominal concrete strengths of 40 to 80 MPa was also part of

the investigation and included the comparison of the structural response (load versus

deflection, ultimate strength, and ductility) of highly congested columns cast with SCC to

that of similar specimens cast with conventionally-vibrated normal concrete (NC) of

equal strength. Part of those data was reported by Paultre et al. (1996) regarding the

structural behavior of congested columns cast with SCC of compressive strength in the

60 and 80 MPa ranges, as well as control columns cast with adequately consolidated NC

of similar strength. The results indicated that high-strength SCC can provide ductility

equivalent to that of high-strength NC but with lower ultimate confined concrete strength.

The main objective of the study reported herein was to compare the structural

performance of highly confined concrete columns cast with normal concrete and self-

consolidated concrete. The effects of concrete compressive strength (40 to 80 MPa),

confining stirrup configuration, and stirrup yield strength (400 and 800 MPa) were also

evaluated. The distribution of compressive strength and elastic modulus along the height

of plain unreinforced columns was also determined for specimens cast with either NC or

SCC.

4

RESEARCH SIGNIFICANCE

The construction of heavily reinforced concrete members, such as columns and beams in

moment-resisting frames in seismic areas, makes the placement of concrete quite

difficult. Self-consolidated concrete can be used to facilitate the construction of elements

without mitigating structural performance and durability. Most studies on SCC reported

in the literature deal with mixture proportioning and characterization of fresh- and

hardened- concrete properties with limited information on structural performance. One of

the barriers to the widespread acceptance of SCC is the lack of information regarding

structural properties of sections cast with SCC. This paper presents for the first time – to

the best of the authors’ knowledge – results on the mechanical behavior of normal- and

high-strength self-consolidated confined concrete columns under concentric axial

loading. The results should be of interest to engineers considering the use of such

concrete in various structural applications.

MATERIALS

The approach in this research on the development of high-performance SCC for casting

highly congested columns involved using high paste volume (and low aggregate volume)

to promote high deformability and reduce the risk of blockage and segregation during

concrete placement. The water-to-cementitious materials ratio (W/CM) was selected as a

function of the targeted compressive strength. The cohesiveness and stability of the

plastic concrete were enhanced by incorporating a moderate concentration of viscosity-

enhancing admixture (VEA), also known as viscosity-modifying admixture, in order to

retain some of the free water in suspension, hence increasing the plastic viscosity of the

mixture.

Self-consolidating concrete can be distinguished from conventional concrete not only by

its high fluidity but also by its composition. The most important distinctions are:

5

- use of a high content of powder materials (<80 µm, from 450 to 600 kg/m3) that

necessitates the replacement of up to 50% of the powder content by supplementary

cementitious materials and/or fillers;

- use of large quantities of superplasticizer; and

- use of a viscosity-enhancing admixture in some cases when the water content is not

low enough to promote sufficient viscosity of the paste.

Khayat et al. (1997) proposed a mixture design for high-performance SCC suitable for

casting highly congested structural elements. Such concrete was proportioned with 0.41

W/CM, 186 L/m3 of ternary binder, 300 L/m3 of coarse aggregate (max. size of 10 mm),

and a 0.60 sand/paste ratio, by volume. These values were used in this study for

characterizing the SCC.

A Type 10 Canadian Portland cement (CSA3-A5-M98, similar to standard specification

for Portland cement ASTM C150-99 Type I cement) was used. Powder materials

incorporated in SCC were silica fume, limestone powder, and blast-furnace slag. The

admixtures included a naphthalene superplasticizer, a set retarder, and Welan gum VEA.

Table 1 presents the concrete mixture proportions used for the concrete of the column

specimens tested in this study. Table 2 shows the properties of hardened concrete based

on standard cylinders prepared when casting each column specimen. These mechanical

properties for plain concrete are: compressive strength (f’C); modulus of elasticity (EC);

peak axial strain (εC0); and axial strain when axial stress reduces to 0.5 f’C (εC50U).

EXPERIMENTAL PROCEDURE AND INSTRUMENTATION

This research was completed in two steps. The first step involved the study of the

rheological properties of the SCC, while the second part consisted of validating the use of

SCC for structural applications. In order to perform the second step, concrete columns

were made with SCC and normal concrete.

6

Workability of SCC

The workability of fresh SCC was characterized by using a number of testing methods

described below. These tests provided the means for comparing different mixtures and

predicting the response of fresh concrete for mixture optimization.

Slump-Flow Test: Conventional concrete has a relatively high yield value; however,

under vibration, the yield value is artificially lowered, leading to the spread and

compaction of concrete. SCC has a low yield value (or high deformability), making

vibration unnecessary. The consistency of SCC is evaluated with the slump-flow test,

which consists of determining the mean spread of the concrete at the base of the slump

test after the end of spreading. Such value can be related to the yield stress of the

concrete, and the rate of spread with time can be related to the plastic viscosity (Khayat et

al. 2004). The slump-flow test is used to assess filling ability, or the unrestricted

deformability of SCC.

Filling-Capacity Test: This so-called caisson filling capacity test (Yurugi et al. 1993)

evaluates the ability of concrete to flow through a series of parallel bars with constant

spacing without segregation or blockage, as shown in Figure 1. SCC with relatively low

dynamic segregation resistance or high viscosity and relatively low filling ability will

have limited flow through this highly restricted section. SCC with a filling capacity

greater than 80% can be expected to fill congested formwork.

Funnel Test: A trap is opened in a funnel filled with concrete (Figure 2), and the time

required to empty the funnel is measured. This test is based on a model proposed by

Ozawa et al. (1995) except that the outlet opening of 65x75 mm was modified to 75x75

mm. This test is used to evaluate the flowing ability of the SCC and can assess the

dynamic stability of the concrete. A long flow-out duration can be due to segregation

(aggregates stacking at the opening) or excessive concrete viscosity.

7

Structural Testing

The structural research involved axial compression testing of square 235 235 1400× × mm

column specimens for which the typical geometry and stirrup configurations are shown in

Figure 3. Their structural performances are compared to 7 other test specimens

previously tested by Cusson and Paultre (1994), and Paultre et al. (1996), where also the

concrete mixture proportions for these columns can be found.

The principal variables of the investigated column specimens are: (i) type of concrete

(SCC or NC), (ii) nominal concrete compressive strength (40, 50, 60 or 80 MPa), (iii)

type of stirrup configuration (Type B or D, as illustrated in Figure 3), and (iv) nominal

steel yield strength of stirrups (400 or 800 MPa).

The experimental program of this study included the construction of 9 steel-reinforced

columns made with SCC (f’C = 40-80 MPa) and 2 others made with NC (f’C = 40 MPa).

These columns, together with those previously tested by Cusson and Paultre (1994) and

Paultre et al. (1996), form a complete group of 18 confined concrete columns, enabling

the full comparison of the test variables listed above. In addition, three 50 to 60 MPa

unreinforced concrete columns were cast to determine the column compressive strength

under concentric loading. Two other columns were prepared to evaluate the vertical

distribution of in situ compressive strength and modulus of elasticity by testing core

samples taken from the bottom, middle and top sections of these column specimens. In

general, the distribution of in situ mechanical properties along the height of unreinforced

columns was more homogeneous in SCC columns than NC columns. More details on the

testing of the unreinforced concrete columns can be found in Khayat et al. (2001).

Each column specimen is identified with a label consisting of a number followed by a

series of letters, as presented in Table 3. The number corresponds to the nominal concrete

compressive strength series (number 7 corresponds to the 80 MPa series, number 8

corresponds to 60 MPa series, and number 10 to the 40 to 50 MPa series). The letters B

and D indicate the transverse reinforcement configuration type (as illustrated in Figure 3);

8

the letter X identifies an unreinforced column; SCC indicates that the concrete is self-

consolidating (if not mentioned, vibrated NC was used); and the letter O indicates the use

of stirrups made of ordinary steel (if not mentioned, high-strength steel stirrups were

used).

The details of the steel reinforcement used for the column specimens tested are presented

in Table 3, along with the concrete compressive strengths and elastic moduli based on

150 300× mm cylinders. All column specimens have the same percentage of longitudinal

reinforcement (ρg = 3.6 %), the number of longitudinal bars (nb) of 8 for tie configuration

B and 12 for tie configuration D, and diameter (db) varying from 11.3 to 19.5 mm. The

transverse reinforcement volumetric ratio and diameter were ρh = 4.9 % and dh = 9.5 mm

for tie configuration B, respectively, and ρh = 4.8 % and dh = 7.9 mm for tie configuration

D, respectively. A tie spacing of 50 mm on centers was used in all columns. The

transverse reinforcement ratio varied from 90% to 360% of the requirement

recommended by ACI-318 Code (2002) for seismic construction. The steel properties

were obtained from tensile loading tests performed on at least three samples from each lot

of reinforcing bars. The yield strength of the longitudinal reinforcement (fy) varied from

414 to 495 MPa and that of the transverse reinforcement (fyh) ranged from 410 to 440

MPa for ordinary steel (hot-rolled) and 680 to 891 MPa for high-strength steel (cold-

formed).

During casting of the concrete columns, NC was thoroughly vibrated in the forms, while

SCC was only poured into forms from the top without any mechanical vibration. Filling

the forms took less than a minute for each column made with SCC and an average of 15

minutes for each NC column. Fresh concrete was sampled with standard 150×300 mm

cylinders to determine the concrete compressive strength. Additionally, testing of

standard 100×200 mm cylinders under controlled deformation rates was conducted to

obtain other key concrete properties listed in Table 2 and the complete stress-stain curve

of the unconfined concrete. Cylinder strengths were determined within a day of testing

the corresponding columns.

9

The axial deformation of column specimens were measured using 4 LVDTs (Linear

Variable Differential Transformers) installed at each corner of the column, in the central

part, over a length of 800 mm. Two layers of stirrups, at mid-height, were instrumented

with strain gauges to record the strain in the ties. Each column was capped with thin

sulfur layers to achieve flat, smooth and parallel surfaces for uniform compression

loading. To ensure that the failure would occur in the instrumented region of the tested

specimens, the tapered ends of each test column were further confined with bolted boxes

made from 13 mm thick steel plates. The test specimens were loaded on a rigid hydraulic

press with load-controlled capabilities, having a maximum compressive load capacity of

6700 kN. An average loading rate of 1kN/sec was maintained during the test, which

lasted between 45 and 90 minutes for each column. During testing, a data acquisition

system recorded the load and longitudinal deformations of the column specimens at

regular intervals, as well as the lateral deformation of the instrumented ties. The testing

was terminated when the loading system reached its maximum displacement or when the

column specimen was unable to sustain additional loading.

TEST RESULTS

Workability of SCC

In general, all four SCC mixtures with compressive strengths of 40 to 80 MPa had

adequate filling capacity test values (82-96%), which were greater than the targeted value

of 80%. Higher strength concretes displayed higher V-funnel flow times. For instance,

the 80 MPa SCC had a particularly high flow time, which is indicative of high viscosity.

Structural testing

The experimental results are presented in Table 4 and illustrated in Figures 4 to 11. They

also include the test results from 7 additional columns tested by Cusson and Paultre

(1994) and Paultre et al. (1996) to allow full comparison between column specimens.

Figures 4 and 5 show the axial load/axial strain relationship for all columns tested, as

10

recorded during testing. Figures 6 to 11 show the results expressed in terms of load

applied on the concrete (PC) normalized with respect to the axial capacity of the

unconfined concrete cross section before spalling (POC = 0.85 f’C AC) or the axial capacity

of the confined concrete core section after spalling (POCC = 0.85 f’C ACC). The axial load

applied on the concrete was determined for each column specimen by subtracting the

load sustained by the longitudinal bars from the total load recorded during testing. The

computation of the axial load sustained by the longitudinal bars was based on their total

cross-sectional area (Ast), on the steel stress-strain curves obtained from tensile loading

tests, and the assumption of strain compatibility between steel and concrete. These curves

were constructed using the techniques presented in Cusson and Paultre (1994).

In Table 4, several parameters were calculated from the resulting load-displacement

curves obtained during the column axial load tests. Measured strength parameters (Pmax,

PC1, and PC2) are the maximum axial load carried by the column, the axial load carried by

the concrete prior to cover spalling (1st peak load), and the maximum axial load carried

by confined concrete (2nd peak load), respectively. The theoretical strength parameters

(PO, POC, and POCC) are the axial capacities of the column cross-section, of the concrete

cross-section before spalling (AC), and of the concrete cross-section after spalling (ACC),

respectively. The ductility parameters (εCO, εC1, εC2 and εC50C) are the axial strain in plain

concrete corresponding to f’c, the axial strain in the column concrete at the first peak load,

the axial strain in the column concrete at the second peak load, and the axial strain in

confined concrete when the load reduces to 0.5 PC2, respectively.

ANALYSIS OF RESULTS

Homogeneity of Strength and Elastic Modulus along Column Height

The coring campaign carried out on the unreinforced columns 10X-2 and 10XSCC-2

allowed the measurement of in situ mechanical properties of both types of concrete along

column height. The 50 MPa NC showed a linear decrease in compressive strength and

elastic modulus over the column height, with greater values obtained at the bottom (f’c =

11

53 MPa and Ec = 37 GPa) than at the top (f’c = 46 MPa and Ec = 36 GPa). On the other

hand, the 50 MPa SCC column evidenced a uniform distribution of in situ properties,

with f’c = 45-47 MPa and Ec = 29 GPa. Such heterogeneity in the vertical distribution of

compressive strength for NC may be attributed to the phenomenon of water migration

upwards under the own weight of concrete, leading to consolidation at the bottom and a

local increase in water content near the top of the column. More results on this part of the

study are presented in Khayat et al. (2001).

Effect of Type of Concrete on Column Strength and Ductility

Strength

The majority of column specimens listed in Table 4 carried a maximum axial load, Pmax,

higher than the corresponding theoretical capacity (Po). The highest ratios of Pmax/PO

were found in the 40 MPa series with values close to 1.5, regardless of concrete type.

However, as nominal concrete strength increased, columns made with SCC started to

show lower ratios than corresponding columns made with NC. For instance, in the 80

MPa series, Pmax/PO values of 1.17 and 1.00 were found for NC and SCC, respectively.

Similar observations can be made when other strength ratios (PC1/POC and PC2/POCC) are

considered.

Ductility

Although the use of SCC, as opposed to NC, did not result in higher peak loads, it

resulted in improved ductility, especially for the 40 MPa SCC columns. The ratio of the

strain at the peak stress of confined concrete to the strain at the peak stress of

corresponding unconfined concrete (εc2/εco), which is an indicator of ductility, varied

from 5.71 to 13.73 for NC columns with high-yield strength ties and from 6.36 to 14.15

for SCC columns with high-yield strength ties. The highest increase in ductility from the

use of SCC over NC was observed in columns with lower nominal concrete strength.

12

Effect of Tie Yield Strength on Column Strength and Ductility

Strength

All columns made with high-yield strength ties (800 MPa) had maximum axial load

(Pmax) greater than the corresponding theoretical capacity (PO = 0.85 f’C AC + fy Ast).

Higher Pmax/PO values were found for columns made with high-yield strength ties when

compared to columns made with normal-strength ties (400 MPa), regardless of concrete

type. However, the relative increase in strength obtained by the use of high-yield strength

ties decreased as the nominal concrete strength increased. This indicates that confinement

is less effective when nominal concrete strength increases. For instance, in the 40 MPa

series, an increase of Pmax/PO from 1.18 to 1.50 was found when high-yield strength ties

were used instead of normal-strength ties, while Pmax/PO increased from 1.02 to 1.17 in

the 60 MPa series. Similar observations can be made when other strength ratios (PC1/POC

and PC2/POCC) are considered.

Most columns made with normal-strength ties reached a maximum load on concrete prior

to spalling (PC1) close to their corresponding theoretical concrete capacity (POC). Values

of PC1/POC varied from 0.85 to 1.03 for the 60- and 80 MPa series, and from 1.10 to 1.13

for the 40 MPa series. Ratios lower than 1.00 are due to premature spalling of the

concrete cover, which is frequently observed in high-strength concrete columns (Cusson

and Paultre, 1994). Unreinforced 50 MPa columns displayed similar ratios, which varied

from 0.96 to 1.07. This indicates that the columns made with normal-strength ties tested

in this study can be considered as columns with low or negligible confinement in terms of

strength increase.

Ductility

Columns made with normal-strength ties exhibited the smallest level of confinement.

Values of εc2/εco varied from 1.47 in the 80 MPa series to 5.77 in the 40 MPa series.

These values are rather small when compared to those measured on columns made with

high-yield strength ties, which varied from 5.71 in the 80 MPa series to 14.15 in the 60

MPa series. Columns made with normal-strength ties cannot provide high levels of

13

ductility, since once the low yield strength of the steel is reached, additional lateral

expansion of the concrete core cannot be further restrained, resulting in rapid tie failure,

sudden loss of confinement, and buckling of the longitudinal bars. Specimens 7DSCC0

and 7BSCC0, illustrated in Figure 12, displayed such brittle behavior with the typical

failure plane occurring at about 35° to 40° from the vertical axis. An earlier study

(Cusson and Paultre, 1994) found that failure by crushing of concrete or with a failure

plane of 45° was an indication of the achievement of optimal confinement in the column.

In general, it can be concluded that high-yield strength ties provided the columns with

superior confinement and, as a result, the columns displayed smooth, progressive failure

by concrete crushing. This behavior was more distinct for SCC in the 40 MPa series.

Effect of Tie Configuration on Column Strength and Ductility

Strength

The strength parameters in Table 4 revealed that columns made with tie configuration B

performed just as well as columns made with tie configuration D, irrespective of concrete

strength (40 to 80 MPa), concrete type (SCC vs. NC), and tie yield strength (400 to 800

MPa). For instance, when the 9 columns made with tie configuration B are compared to

the 9 columns made with tie configuration D, the average PC1/POC value increased by 1%,

and the average PC2/POCC value increased by 2% when configuration D is used over

configuration B. It is interesting to note that the smaller clear spacing of configuration D

(3 individual ties per set and 12 longitudinal bars) did not result in earlier spalling of the

concrete cover when compared to configuration B (2 individual ties per set and 8

longitudinal bars).

Ductility

Similarly, the ductility parameters indicated that columns of configuration B performed

just as well as columns of configuration D, irrespective of concrete strength, concrete

type, and tie yield strength. For instance, the average εC1/εCO value increased by 5%, and

the average εC50C/εCO value increased by 3% when configuration D is used over

configuration B. Although the ductilities provided by configurations B and D are similar,

14

Cusson and Paultre (1994) demonstrated that these 2 tie configurations were far superior

in providing ductility to a tie configuration using only 1 square tie per set with 4

longitudinal corner bars (of similar ρh, ρg and s).

It is therefore suggested that tie configuration B, with a center-to-center spacing of 50

mm, should be given preference in the design of highly-confined concrete columns,

because it can provide excellent strength and ductility. Moreover, the construction of

columns with such tie configuration is less labor intensive than columns with tie

configuration D for the same volume of steel reinforcement.

DISCUSSION

It is believed that the lower ultimate strengths and higher ductilities measured for SCC

columns compared to NC columns are probably due to a lower SCC modulus of

elasticity. SCC contains a smaller volume of coarse aggregate than NC, resulting in lower

elastic modulus and larger transverse expansion under compressive load. As a matter of

fact, this tendency was demonstrated by comparing the less ductile high-strength NC

columns in the 80 MPa series to the more ductile normal-strength SCC in the 40 MPa

series. It was found that the difference in ductility between NC and SCC columns

decreased as nominal concrete strength increased, probably due to the higher strength and

stiffness of the paste itself.

It is therefore suggested that ductility can be achieved, not only by using closely-spaced

ties of high-yield strength, but also by using a concrete such as SCC with a high paste

content (or low aggregate content). This could also be due to the fact that SCC can

develop better bond to the reinforcement, and hence improve ductility, than NC with

mechanical consolidation. More research is needed to confirm this statement.

15

CONCLUSIONS

The mechanical performance of highly confined columns cast with normal concrete (NC)

vibrated into place was compared to that of identical columns cast with self-consolidating

concrete (SCC). The following conclusions were obtained:

• The use of SCC reduced significantly the time of casting highly congested column

sections. The optimized mixtures allowed adequate placement of concrete into

heavily reinforced columns and achieved high levels of filling capacity, regardless of

tie configuration and design strength.

• Distribution of in situ properties along column height was found to be more

homogeneous for SCC columns than NC columns.

• Columns made with SCC offered slightly lower axial load carrying capacities but

were more ductile than equally confined columns made with NC. The lower modulus

of elasticity of SCC resulting in larger lateral expansion under a given compressive

load, may be responsible for this difference between NC and SCC.

• Columns made with high-yield strength ties provided significantly better confinement

than similar columns made with normal-strength ties, regardless of concrete type.

Although columns made with normal-strength ties displayed some ductility, they did

not offer axial capacities higher than the expected theoretical values.

• Confinement was more effective in normal-strength concrete than in high-strength

concrete, as the highest ductility (promoted by the use of either SCC concrete or high-

yield strength ties) was obtained in columns with lower nominal concrete strength.

• Tie configuration B (with s = 50 mm) should be given preference in the design of

highly-confined concrete columns, because it can provide excellent strength and

ductility (like configuration D), and the construction of columns with tie

configuration B is less labor intensive than with configuration D for the same volume

of steel reinforcement.

16

ACKNOWLEDGMENTS

The financial assistance provided by the Natural Sciences and Engineering Research

Council of Canada (NSERC) and by the Fonds de recherche sur la Nature et les

technologies du Québec (NATEQ) is gratefully acknowledged.

AFFILIATION OF AUTHORS

Patrick Paultre, ACI member, holds a Canada Research Chair in Seismic Engineering at

the University of Sherbrooke (Sherbrooke, Quebec, Canada). He is a member of ACI

Committee 441, Reinforced Concrete Columns. He is the technical secretary of the

Canadian Standards Association Committee A23.3, Design of Concrete Structures.

Kamal. H. Khayat, ACI member, is Professor of Civil Engineering at the University of

Sherbrooke. He sits on ACI committees 234, Silica Fume in Concrete; 236, Materials

Science of Concrete; 552, Geotechnical Cement Grouting; and he his the Secretary of

237, SCC. His research interests include rheology, self-consolidating concrete, and

repair.

Daniel Cusson, ACI member, is a Research Officer at the National Research Council

(Ottawa, Ontario, Canada). He is a member of ACI Committee 363, High-Strength

Concrete. His research interests include early-age behavior of high-performance

concrete structures, and in situ performance and durability of rehabilitated concrete

bridges.

Stephan Tremblay is a structural engineer with Neal Smith Engineering, Inc. in North

Carolina. He received his M.Sc. from the University of Sherbrooke. His research

interests include the study of self-consolidating concrete and its structural use in highly

reinforced members.

17

CONVERSION FACTORS

1 MPa = 145 psi

1 mm = 0.0394 in

1 kN = 0.2248 kip

NOTATION

AC cross-sectional area of concrete

ACC cross-sectional area of concrete core bounded by centerline of outer tie

Ag gross cross-sectional area of column

Ast cross sectional area of longitudinal steel reinforcement

db longitudinal steel bar diameter

dh transverse steel bar diameter

EC modulus of elasticity of plain concrete

fC stress in concrete

fCC maximum stress in confined concrete

f'C compressive strength of plain concrete measured on 150x300 mm cylinders

fy yield strength of longitudinal reinforcement steel

fyh yield strength of transverse reinforcement steel

nb number of longitudinal steel bars in column cross-section

P axial load carried by column

PC axial load carried by concrete

PC1 axial load carried by concrete at onset of cover spalling

PC2 maximum axial load carried by confined concrete

Pmax maximum axial load carried by column

PO axial capacity of column cross section

POC axial capacity of concrete cross section

POCC axial capacity of concrete core bounded by centerline of outer tie

s center-to-center spacing between sets of ties

εC axial strain in concrete

18

εCO axial strain in plain concrete corresponding to f'C

εC1 axial strain in column concrete corresponding to PC1

εC2 axial strain in column concrete corresponding to PC2

εC50C axial strain in confined concrete when stress reduces to 0.5 fCC

εC50U axial strain in unconfined concrete when stress reduces to 0.5 f'C

ρg volumetric ratio of longitudinal reinforcement in column cross section

ρh volumetric ratio of transverse reinforcement in concrete core

REFERENCES

ACI Committee 318, “Building Code Requirements for Reinforced Concrete, ACI 318-

02, and Commentary, ACI 318R-02,” American Concrete Institute, Detroit, MI, 2002,

443 pp.

American Society for Testing and Materials, “Standard Specification for Portland

Cement,” ASTM C150-99, ASTM, Philadelphia, 1999, 5 pp.

Canadian Standards Association, “Portland Cements,” CSA A5-M98, CSA, Rexdale,

Canada, 1998.

Cusson, D. and Paultre, P, “High-Strength Concrete Columns Confined by Rectangular

Ties,” ASCE Journal of Structural Engineering, Vol. 120, No. 3, 1994, pp. 783-804.

Khayat, K.H., Assaad, J., and Daczko, J., “Comparison of Field-Oriented Test Methods

to Assess Dynamic Stability of Self-Consolidating Concrete,” ACI Materials Journal,

Vol. 101, No. 2, 2004, pp. 168-176.

Khayat, K.H., Manai, K., and Trudel, A. “In Situ Mechanical Properties of Wall

Elements Cast Using Self-Consolidating Concrete,” ACI Materials Journal, Vol. 94,

No. 6, 1997, pp. 491-500.

Khayat, K.H., Paultre, P., and Tremblay, S., “Structural Performance and In Situ

Properties of Self-Consolidating Concrete,” ACI Materials Journal, Vol. 98, No. 5,

2001, pp. 371-378.

Ozawa, K., Sakata, N., and Okamura, H., “Evaluation of Self-Compactibility of Fresh

Concrete Using the Funnel Test,” JSCE Concrete Library, Vol. 25, 1995, pp. 59-75.

19

Paultre, P., Khayat, K.H., Langlois, Y., Trudel, A., and Cusson, D., “Structural

Performance of Some Special Concretes,” 4th Int. Symposium on Utilization of High-

Strength/High-Performance Concrete, Paris, 1996, pp. 787-796.

Yurugi, M., Sakata, N. Iwai, M., and Sakai, G., “Mix Proportions for Highly Workable

Concrete,” Proceedings of Concrete 2000, Dundee, 1993, pp. 579-589.

20

LIST OF FIGURES

Figure 1: Filling capacity test

Figure 2: Funnel test

Figure 3: Dimensions and geometry of columns tested

Figure 4: Total load versus axial strain curves for columns with tie configuration B

Figure 5: Total load versus axial strain curves for columns with tie configuration D

Figure 6: Normalized concrete load versus axial strain for the 40 MPa columns with tie

configuration B


configuration D


configuration B


configuration D


configuration B


configuration D

Figure 12: Specimens 7DSCCO and 7BSCCO after testing

21

Table 1: Mixture Proportions of Concrete

NC SCC Nominal Strength Unit 40 MPa 50 MPa 40 MPa 50 MPa 60 MPa 80 MPa Water kg/m3 180 178 239 234 228 189 Cement Type 10 kg/m3 360 357 386 388 307 – Cement Type 10-SF kg/m3 – – – – – 476 Silica fume kg/m3 – – 18 18 18 (33.5)* Limestone filler kg/m3 – – 179 180 – – Blast furnace slag kg/m3 – – – – 235 134 Total binder content kg/m3 360 357 583 586 561 610 Sand kg/m3 960 951 703 707 705 665 Gravel, 5-10 mm kg/m3 900 891 828 833 877 949 Superplasticizer L/m3 1.4 1.2 5.0 4.8 4.5 7.0 VEA (Welan gum) kg/m3 – – 0.437 0.440 0.422 0.068 Set-retarding agent L/m3 0.5 0.5 0.5 0.5 0.5 0.5 W/CM – 0.50 0.50 0.42 0.41 0.41 0.31 Slump mm 110 125 – 265 – 265 Slump flow mm – – 630 640 660 635 Filling capacity % – – 93 83 96 82 V-funnel flow time s – – 1.5 1.6 5.5 8.7 Fresh air volume % 1.8 3.2 0.6 0.6 1.2 0.9 Unit weight kg/m3 2,400 2,375 2,360 2,365 2,375 2,433 * Quantity included in cement

22

Table 2: Hardened Properties of Plain Concrete

Specimens f’C , Age EC εCO εC50U εC50U /εCO (MPa)* (GPa)** 10 B 43.6, 31d 37.0 0.00312 0.0061 1.95 10 BSCC 39.1, 38d 30.3 0.00307 0.0059 1.91 10 BSCCO 41.0, 25d 31.7 0.00380 0.0062 1.62 10 D 43.6, 31d 37.0 0.00312 0.0061 1.95 10 DSCC 39.1, 38d 30.3 0.00307 0.0059 1.91 10 DSCCO 41.0, 25d 31.7 0.00380 0.0062 1.62 10 X 50.3, 32d 35.7 0.00257 0.0051 1.99 10 XSCC 49.4, 32d 28.7 0.00355 0.0048 1.36 8 BSCC 61.6, 59d 38.6 0.00294 0.0043 1.47 8 BSCCO 61.6, 59d 38.6 0.00294 0.0043 1.47 8 DSCCO 61.6, 59d 38.6 0.00294 0.0043 1.47 8 XSCC 61.6, 59d 38.6 0.00294 0.0043 1.47 7 BSCCO 87.2, 47d 36.9 0.00301 0.0048 1.58 7 DSCCO 87.2, 47d 36.9 0.00301 0.0048 1.58

* Mean of four 150x300 mm cylinders ** Mean of two 100x200 mm cylinders

23

Table 3: Details of Test Column Specimens

Longitudinal Reinforcement Lateral Reinforcement Concrete

Column nb and sizea ρg (%)

fy (MPa)

dh (mm)

S (mm)

ρh (%)

fyh (MPa)

f’c (MPa)

Ec (GPa)

10 B 4No 20 + 4No 15 3.6 418, 445 9.5 50 4.9 820 43.6 37.0 10 BSCC 4No 20 + 4No 15 3.6 418, 445 9.5 50 4.9 820 39.1 30.3 10 BSCCO 4No 20 + 4No 15 3.6 418, 445 9.5 50 4.9 410 41.0 31.7 10 D 4No 20 + 8No 10 3.6 442, 495 7.9 50 4.8 801 43.6 37.0 10 DSCC 4No 20 + 8No 10 3.6 442, 495 7.9 50 4.8 801 39.1 30.3 10 DSCCO 4No 20 + 8No 10 3.6 442, 495 7.9 50 4.8 440 41.0 31.7 10 X Plain column tested in compression – – – – 50.3 35.7 10 X-2 Plain column cored – – – – 50.3 35.7 10 XSCC Plain column tested in compression – – – – 49.4 28.7 10 XSCC-2 Plain column cored – – – – 49.4 28.7 8 B † 4No 20 + 4No 15 3.6 482, 436 9.5 50 4.9 715 52.6 35.5 8 BSCC 4No 20 + 4No 15 3.6 418, 445 9.5 50 4.9 820 60.0 34.7 8 BSCCO 4No 20 + 4No 15 3.6 418, 445 9.5 50 4.9 411 60.0 34.7 8 D † 4No 20 + 8No 10 3.6 482, 467 7.9 50 4.8 680 55.6 35.5 8 DSCC ‡ 4No 20 + 8No 10 3.6 414, 444 7.9 50 4.8 769 60.0 34.7 8 DSCCO 4No 20 + 8No 10 3.6 442, 495 7.9 50 4.8 440 61.6 38.6 8 XSCC Plain column tested in compression

and extremities cored after testing – – – – 61.6 38.6

7 B † 4No 20 + 4No 15 3.6 482, 486 9.5 50 4.9 715 75.9 40.1 7 BSCC ‡ 4No 20 + 4No 15 3.6 414, 465 9.5 50 4.9 891 80.5 38.7 7 BSCCO 4No 20 + 4No 15 3.6 442, 430 9.5 50 4.9 410 87.2 36.9 7 D † 4No 20 + 8No 10 3.6 482, 467 7.9 50 4.8 680 67.9 40.5 7 DSCC ‡ 4No 20 + 8No 10 3.6 414, 444 7.9 50 4.8 769 80.5 38.7 7 DSCCO 4No 20 + 8No 10 3.6 442, 495 7.9 50 4.8 440 87.2 36.9

a No. 10: db = 11.3 mm; No. 15: db = 16.0 mm; No. 20: db = 19.5 mm † Columns tested by Cusson and Paultre (1994) ‡ Columns tested by Paultre et al. (1996)

24

Table 4: Structural Test Results

Columns Axial Load * Axial Strain Pmax Pmax /

PO PC1 PC1 /

POC PC2 PC2/

POCC εC1 εC1 /

εCO εC2 εC2 /

εCO εC50C εC50C /

εCO (kN) (kN) (kN) 40 MPa series: 10 B 4,249 1.51 2,433 1.23 3,145 2.36 0.0031 0.98 0.0287 9.19 0.0429 13.75 10 BSCC 3,939 1.50 2,276 1.29 2,739 2.29 0.0034 1.12 0.0356 11.58 0.0686 22.34 10 BSCCO 3,319 1.22 2,043 1.10 2,191 1.75 0.0029 0.76 0.0219 5.77 0.0424 11.15 10 D 4,425 1.53 2,441 1.24 3,231 2.42 0.0026 0.85 0.0274 8.77 0.0449 14.39 10 DSCC 4,063 1.50 2,195 1.24 2,766 2.31 0.0034 1.09 0.0385 12.53 0.0592 19.28 10 DSCCO 3,291 1.18 2,103 1.13 2,173 1.73 0.0034 0.88 0.0166 4.36 0.0366 9.28 50 MPa series: 10 X 2,441 1.07 2,441 1.07 – – 0.0024 0.96 – – – – 10 XSCC 2,153 0.96 2,153 0.96 – – 0.0029 0.81 – – – – 60 MPa series: 8 B 4,530 1.37 2,213 0.93 3,219 2.00 0.0034 1.45 0.0321 13.73 0.0459 19.63 8 BSCC 4,067 1.14 2,352 0.87 2,918 1.59 0.0027 0.93 0.0349 11.88 0.0501 12.62 8 BSCCO 3,235 0.91 2,312 0.85 2,322 1.26 0.0027 0.92 0.0103 3.50 0.0282 9.61 8 D 4,532 1.31 2,338 0.93 3,271 1.92 0.0032 1.43 0.0287 12.68 0.0455 20.15 8 DSCC 4,156 1.17 2,372 0.87 2,994 1.63 0.0029 1.75 0.0284 14.15 0.0550 24.47 8 DSCCO 3,803 1.02 2,877 1.03 2,787 1.48 0.0025 0.86 0.0065 2.22 0.0293 9.97 8 XSCC 2,943 1.02 2,943 1.02 – – 0.0020 0.67 – – – – 80 MPa series: 7 B 4,954 1.14 3,443 1.00 3,849 1.66 0.0030 1.10 0.0156 5.71 0.0313 11.43 7 BSCC 4,507 1.00 3,634 1.00 3,556 1.44 0.0030 1.47 0.0172 8.02 0.034 13.39 7 BSCCO 4,552 0.94 3,678 0.93 3,382 1.27 0.0029 0.97 0.0067 2.24 0.0203 6.74 7 D 4,701 1.17 3,058 1.00 3,615 1.74 0.0028 1.07 0.0155 5.95 0.0279 10.70 7 DSCC 4,471 1.00 3,237 0.89 3,502 1.42 0.0024 1.18 0.0133 6.36 0.037 14.57 7 DSCCO 4,789 0.98 3,863 0.98 3,547 1.33 0.0032 1.06 0.0044 1.47 0.0180 1.58

* PO = 0.85 f’C AC + fy Ast, POC = 0.85 f’C AC, POCC = 0.85 f’C ACC

25

Figure 1 – Paultre, Khayat, Cusson and Tremblay

26


27

Figure 3– Paultre, Khayat, Cusson and Tremblay

28


29


30


31


32


33


34


35


36


Structural performance of self-consolidating concrete used ... · enhancing admixture (VEA), also...

Documents

Transcript of Structural performance of self-consolidating concrete used ... · enhancing admixture (VEA), also...