ACI 224R-01

Control of Cracking in Concrete Structures

Reported by ACI Committee 224

Florian Barth Robert J. Frosch*
Chairman Secretary

Mohamed Abou-Zeid David W. Fowler* Edward G. Nawy*

John H. Allen Grant T. Halvorsen Randall W. Poston*
*
James P. Barlow Will Hansen Royce J. Rhoads
Merle E. Brander* M. Nadim Hassoun Andrew Scanlon
Kathy Carlson Harvey Haynes* Ernest K. Schrader*
David Darwin* Paul Hedli Wimal Suaris*
Fouad H. Fouad* Tony C. Liu Zenon A. Zielinski
*Members of ACI 224 who assisted in revisions to this report.

The principal causes of cracking and recommended crack-control proce- CONTENTS

dures are presented. The current state of knowledge in microcracking and Chapter 1—Introduction, p. 224R-2
fracture of concrete is reviewed. The control of cracking due to drying
shrinkage and crack control in flexural members, overlays, and mass con- Chapter 2—Crack mechanisms in concrete,
crete construction are covered in detail. Long-term effects on cracking are p. 224R-2
considered and crack-control procedures used in construction are pre-
2.1—Introduction
sented. Information is presented to assist in the development of practical 2.2—Compressive microcracking
and effective crack-control programs for concrete structures. Extensive ref- 2.3—Fracture
erences are provided.
Chapter 3—Control of cracking due to drying
shrinkage, p. 224R-11
Keywords: aggregates; anchorage (structural); bridge decks; cement-
3.1—Introduction
aggregate reactions; concrete construction; concrete pavements; concrete
3.2—Cause of cracking due to drying shrinkage
slabs; cooling; corrosion; crack propagation; cracking (fracturing); crack
width and spacing; drying shrinkage; shrinkage-compensating concrete;
3.3—Drying shrinkage
heat of hydration; mass concrete; microcracking; polymer-modified concrete; 3.4—Factors controlling drying shrinkage of concrete
prestressed concrete; reinforced concrete; restraint; shrinkage; temperature; 3.5—Control of shrinkage cracking
tensile stresses; thermal expansion; volume change. 3.6—Shrinkage-compensating concrete

ACI Committee Reports, Guides, Standard Practices, Chapter 4—Control of cracking in flexural
members, p. 224R-17
and Commentaries are intended for guidance in planning, 4.1—Introduction
designing, executing, and inspecting construction. This 4.2—Crack-control equations for reinforced concrete beams
document is intended for the use of individuals who are 4.3—Crack control in two-way slabs and plates
competent to evaluate the significance and limitations of 4.4—Tolerable crack widths versus exposure conditions in
its content and recommendations and who will accept re- reinforced concrete
sponsibility for the application of the material it contains. 4.5—Flexural cracking in prestressed concrete
The American Concrete Institute disclaims any and all re- 4.6—Anchorage-zone cracking in prestressed concrete
sponsibility for the stated principles. The Institute shall 4.7—Crack control in deep beams
not be liable for any loss or damage arising therefrom. 4.8—Tension cracking
Reference to this document shall not be made in con-
tract documents. If items found in this document are de- ACI 224R-01 supersedes ACI 224R-90 and became effective May 16, 2001.
sired by the Architect/Engineer to be a part of the contract Copyright  2001, American Concrete Institute.
All rights reserved including rights of reproduction and use in any form or by any
documents, they shall be restated in mandatory language means, including the making of copies by any photo process, or by electronic or
mechanical device, printed, written, or oral, or recording for sound or visual reproduc-
for incorporation by the Architect/Engineer. tion or for use in any knowledge or retrieval system or device, unless permission in
writing is obtained from the copyright proprietors.

224R-1
224R-2 ACI COMMITTEE REPORT

Chapter 5—Long-term effects on cracking, prestressed members, anchorage zone cracking, and flexural
p. 224R-24 cracking in deep flexural members. Chapter 6 on concrete
5.1—Introduction overlays has been reorganized and revised in modest detail
5.2—Effects of long-term loading to account for updated information on fiber reinforcement
5.3—Environmental effects and on polymer-modified concrete. Chapter 7 on mass
5.4—Aggregate and other effects concrete has been revised to consider structural consequences
5.5—Use of polymers in improving cracking characteristics more extensively.
Chapter 6—Control of cracking in overlays,
p. 224R-25 CHAPTER 2—CRACK MECHANISMS IN
6.1—Introduction CONCRETE
2.1—Introduction
6.2—Fiber-reinforced concrete (FRC) overlays
Cracking plays an important role in concrete’s response to
6.3—Latex- and epoxy-modified concrete overlays
load in both tension and compression. The earliest studies of
6.4—Polymer-impregnated concrete (PIC) systems
the microscopic behavior of concrete involved the response
6.5—Epoxy and other polymer concrete overlays
of concrete to compressive stress. That early work showed
Chapter 7—Control of cracking in mass concrete, that the stress-strain response of concrete is closely associated
p. 224R-28 with the formation of microcracks, that is, cracks that form at
7.1—Introduction coarse-aggregate boundaries (bond cracks) and propagate
7.2—Methods of crack control through the surrounding mortar (mortar cracks) (Hsu, Slate,
7.3—Design Sturman, and Winter 1963; Shah and Winter 1966; Slate and
7.4—Construction Matheus 1967; Shah and Chandra 1970; Shah and Slate
7.5—Operation 1968; Meyers, Slate, and Winter 1969; Darwin and Slate
1970), as shown in Fig. 2.1.
Chapter 8—Control of cracking by proper During early microcracking studies, concrete was considered
construction practices, p. 224R-34 to be made up of two linear, elastic brittle materials; cement
8.1—Introduction
paste and aggregate; and microcracks were considered to be
8.2—Restraint
the major cause of concrete’s nonlinear stress-strain behavior
8.3—Shrinkage
in compression (Hsu, Slate, Sturman, and Winter 1963; Shah
8.4—Settlement
and Winter 1966). This picture began to change in the
8.5—Construction
1970s. Cement paste is a nonlinear softening material, as
8.6—Specifications to minimize drying shrinkage
is the mortar constituent of concrete. The compressive non-
8.7—Conclusion linearity of concrete is highly dependent upon the response
of these two materials (Spooner 1972; Spooner and Dougill
Chapter 9—References, p. 224R-39
9.1—Referenced standards and reports 1975; Spooner, Pomeroy, and Dougill 1976; Maher and Dar-
9.2—Cited references win 1977; Cook and Chindaprasirt 1980; Maher and Darwin
9.3—Other references 1982) and less dependent upon bond and mortar microcracking
than originally thought. Research indicates, however, that a sig-
CHAPTER 1—INTRODUCTION nificant portion of the nonlinear deformation of cement paste
Cracks in concrete structures can indicate major structural and mortar results from the formation of microcracks that
problems and detract from the appearance of monolithic are several orders of magnitude smaller than those observed in
construction. There are many specific causes of cracking. the original studies (Attiogbe and Darwin 1987, 1988). These
This report presents the principal causes of cracking and a smaller microcracks have a surface density that is two to
detailed discussion of crack-control procedures. The report three orders of magnitude higher than the density of bond
consists of eight chapters designed to help the engineer and and mortar microcracks in concrete at the same compres-
the contractor in developing crack-control measures. sive strain, and their discovery represents a significant
This report is an update of previous committee reports step towards understanding the behavior of concrete and
(ACI Committee 224 1972, 1980, 1990). ACI Bibliogra- its constituent materials in compression.
phy No. 9 supplemented the original ACI 224R (1971). The The effect of macroscopic cracks on the performance and
Committee has also prepared reports on the causes, evaluation, failure characteristics of concrete has also received considerable
and repair of cracking, ACI 224.1R; cracking of concrete in di- attention. For many years, concrete has been considered a brittle
rect tension, ACI 224.2R; and joints in concrete construction, material in tension. Many attempts have been made to use
ACI 224.3R. principles of fracture mechanics to model the fracture of
In this revision of the report, Chapter 2 on crack mechanisms concrete containing macroscopic cracks.
has been revised extensively to reflect the interest and attention The field of fracture mechanics was developed by Griffith
given to aspects of fracture mechanics of concrete during the (1920) to explain the failure of brittle materials. Linear elastic
1980s. Chapter 3 on drying shrinkage has been rewritten. fracture mechanics (LEFM) predicts the rapid propagation of a
Chapter 4 has been revised to include updated information microcrack through a homogeneous, isotropic, linear-elastic
on crack-width predictive equations, cracking in partially material. The theory uses the stress-intensity factor K that
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-3

Fig. 2.1—Cracking maps and stress-strain curves for concrete loaded in uniaxial compression
(Shah and Slate 1968).

represents the stress field ahead of a sharp crack in a struc- This chapter is divided into two sections. The first section
tural member which is a function of the crack geometry and on compressive microcracking presents the current knowledge
stress. K is further designated with subscripts, I, II, and III, of the response of concrete and its constituent materials under
depending upon the nature of the deformation at the crack compressive loading and the role played by the various types
tip. For a crack at which the deformation is perpendicular to of microcracks in this process. The second section discusses
the crack plane, K is designated as KI, and failure occurs the applicability of both linear and nonlinear fracture mechanics
when KI reaches a critical value KIc, known as the critical models to concrete. A more comprehensive treatment of the
stress-intensity factor. KIc is a measure of the fracture tough- fracture of concrete can be found in ACI 446.1R.
ness of the material, which is simply a measure of the resis-
tance to crack propagation. Often the region around the crack 2.2—Compressive microcracking
tip undergoes nonlinear deformation, such as yielding in During early microcracking research, a picture devel-
metals, as the crack grows. This region is referred to as the oped that closely linked the formation and propagation of
plastic zone in metals, or more generally as the fracture process microcracks to the load-deformation behavior of concrete.
zone. To properly measure KIc for a material, the test specimen Before loading, volume changes in cement paste cause inter-
should be large enough so that the fracture process zone is facial cracks to form at the mortar-coarse aggregate bound-
small compared with the specimen dimensions. For LEFM ary (Hsu 1963; Slate and Matheus 1967). Under short-term
to be applicable, the value of KIc must be a material property, compressive loads, no additional cracks form until the load
independent of the specimen geometry (as are other material reaches about 30% of the compressive strength of the con-
properties, such as yield strength or compressive strength). crete (Hsu, Slate, Sturman, and Winter 1963). Above this
Initial attempts to measure KIc in concrete were unsuccessful value, additional bond cracks are initiated throughout the
because K Ic depended on the size and geometry of the test matrix. Bond cracking increases until the load reaches about
specimens (Wittmann 1986). As a result of the heterogeneity 70% of the compressive strength, at which time the microc-
inherent in cement paste, mortar, and concrete, these materials racks begin to propagate through the mortar. Mortar crack-
exhibit a significant fracture-process zone and the critical ing continues at an accelerated rate, forming continuous
load is preceded by a substantial amount of slow crack growth. cracks parallel to the direction of compressive load, until the
This precritical crack growth has been studied experimentally concrete is no longer able to sustain the load. The onset of
by several researchers (John and Shah 1986; Swartz and Go mortar cracking is related to the sustained, or long-term,
1984; Bascoul, Kharchi, and Maso 1987; Maji and Shah compressive strength. Derucher (1978) obtained a somewhat
1987; Castro-Montero, Shah, and Miller 1990). This research different picture of the microscopic behavior of concrete
has provided an improved understanding of the fracture process using the scanning electron microscope (SEM). He subjected
zone and has led to the development of more rational fracture dried concrete specimens to eccentric compressive loading
criteria for concrete. within the SEM. He observed that microcracks that exist
224R-4 ACI COMMITTEE REPORT

Fig. 2.2—Stress-strain curves for cement paste, mortar, and concrete; w/c = 0.5 (Martin,
Darwin, and Terry 1991).

before loading are in the form of bond cracks, with exten- Since the early work established the existence of bond and
sions into the surrounding mortar perpendicular to the bond mortar microcracks, it has been popular to attribute most, if
cracks. Under increasing compression, these bond cracks not all, of the nonlinearity of concrete to the formation of
widen but do not propagate at loads as low as 15% of the these microscopic cracks (Hsu, Slate, Sturman, and Winter
strength. At about 20% of ultimate, the bond cracks begin to 1963; Shah and Winter 1966; Testa and Stubbs 1977; Car-
propagate, and at about 30%, they begin to bridge between rasquillo, Slate, and Nixon 1981). A cause and effect rela-
one another. The bridging is almost complete at 45% of the tionship, however, has never been established (Darwin
compressive strength. At 75% of ultimate, mortar cracks 1978). Studies by Spooner (1972), Spooner and Dougill
start to join one another and continue to do so until failure. (1975), Spooner, Pomeroy, and Dougill (1976), and Maher
and Darwin (1982) indicate that the degree of microcracking
In general, microcracking that occurs before loading has little
can be taken as an indication of the level of damage rather
effect on the strength of compressive strength of the concrete.
than as the controlling factor in the concrete’s behavior.
In studies of high-strength concrete, Carrasquillo, Slate,
Experimental work by Spooner (1972), Spooner and Dougill
and Nilson (1981) concluded that it was more appropriate to
(1975), Spooner, Pomeroy, and Dougill (1976), and Martin,
classify cracks as simple (bond or mortar) and combined
Darwin, and Terry (1991) indicates that the nonlinear compres-
(bond and mortar) and that the formation of combined
sive behavior of concrete is strongly influenced by the nonlinear
cracks consisting of more than one mortar crack signaled behavior of cement paste. As illustrated in Fig. 2.2, cement
unstable crack growth. They observed that the higher the paste under compression is not an elastic, brittle material as
concrete strength, the higher the strain (relative to the strain at stated in the past, but a nonlinear material with a relatively high
peak stress) at which this unstable crack growth is observed. strain capacity. The nonlinear behavior of cement paste can be
They observed less total cracking in high-strength concrete tied to damage sustained by the paste, even at very low stresses.
than normal-strength concrete at all stages of loading. Using a cyclic loading procedure, Spooner (1972), Spoon-
Work by Meyers, Slate, and Winter (1969), Shah and er and Dougill (1975), and Spooner, Pomeroy, and Dougill
Chandra (1970), and Ngab, Slate, and Nilson (1981) demon- (1976) demonstrated that both paste and concrete undergo mea-
strated that microcracks increase under sustained and cyclic surable damage at strains (0.0004) at which an increase in bond
loading. Their work indicated that the total amount of micro- and mortar microcracking cannot be detected. The level of
cracking is a function of the total compressive strain in the damage can be detected at low loads by using an energy
concrete and is independent of the method in which the strain method and by a change in the initial modulus of elasticity
is applied. Suaris and Fernando (1987) also showed that the for each load cycle. The process of damage is continuous up
failure of concrete under constant amplitude cyclic loading to failure.
is closely connected with microcrack growth. Sturman, Shah, The physical nature of damage that occurs in cement paste,
and Winter (1965) found that the total degree of microcracking like that in concrete, appears to be related to cracking. This
is decreased and the total strain capacity in compression is point was first made by Spooner, Pomeroy, and Dougill
increased when concrete is subjected to a strain gradient. (1976) based on volumetric strain measurements and then by
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-5

Fig 2.3—Stress-strain curves as influenced by coating aggregates (Darwin and Slate

1970).

Yoshimoto et al. (1972) and Yoshimoto, Ogino, and loads and results in about a 10% reduction in the compressive
Kawakami (1976) who reported the formation of “hair- strength, compared with similar concrete made with aggregate
shaped” and “void-shaped” cracks in paste under flexure and with normal interfacial strength (Fig. 2.3). Darwin and Slate
compressive loading. The relationship between nonlinear also monitored microcracking. In every case, however, the
deformation and cracking in cement paste is now firmly es- average amount of mortar cracking was slightly greater for
tablished by the work of Attiogbe and Darwin (1987, 1988). specimens made with coated aggregate. This small yet
Studies of the stress-strain behavior of concrete under cyclic consistent difference may explain the differences in the
compressive load (Karsan and Jirsa 1969; Shah and Chandra stress-strain curves. Perry and Gillott (1977) used glass
1970) indicated the concrete undergoes rapid deterioration spheres with different degrees of surface roughness as coarse
once the peak stress exceeds 70% of the short-term compres- aggregate. Their results also indicate that reducing the inter-
sive strength of the concrete. In their study of cyclic creep, facial strength of the aggregate decreases the compressive
Neville and Hirst (1978) found that heat is generated even strength by about 10%.
when specimens are cycled below this level. They attributed Work by Carino (1977), using polymer-impregnated
the heat to sliding at the interfacial boundary. The work of
concrete, corroborated these last two studies. Carino
Neville and Hirst, along with the work of Spooner, suggests
found that polymer impregnation did not increase the inter-
that it can be possible that the heat measured is due to some
facial bond strength but did increase the compressive
microscopic sliding within the paste.
strength of concrete. He attributed the increase in strength to
Several studies have attempted to establish the importance
the polymer’s effect on mortar strength, therefore downgrading
of interfacial bond strength on the behavior of concrete in
the importance of interfacial bond.
compression. Two studies seemed to indicate a very large
effect, thus emphasizing the importance of interfacial The importance of mortar in controlling the stress-strain
strength on concrete behavior in compression (Shah and behavior of concrete is illustrated by the finite-element work
Chandra 1970; Nepper-Christensen and Nielsen 1969). of Buyukozturk (1970) and Maher and Darwin (1976, 1977).
These studies used relatively thick, soft coatings on coarse Buyukozturk (1970) used a finite-element representation of
aggregate to reduce the bond strength. Because these soft a physical model of concrete. The model treated mortar (in
coatings isolated the aggregate from the surrounding mortar, compression) and aggregate (in compression and tension) as
the effect was more like inducing a large number of voids in linear elastic materials while allowing cracks to form in the
the concrete matrix. mortar and at mortar aggregate boundaries. Buyukozturk
Two other studies (Darwin and Slate 1970; Perry and simulated the overall crack patterns under uniaxial loading.
Gillott 1977) that did not isolate the coarse aggregate from His finite-element model, however, could not duplicate the
the mortar indicated that interfacial strength plays only a minor full nonlinear behavior of the physical model using the for-
role in controlling the compressive stress-strain behavior of mation of interfacial bond cracks and mortar cracks as the
concrete. Darwin and Slate (1970) used a thin coating of only nonlinear effects. Maher and Darwin (1976, 1977) have
polystyrene on natural coarse aggregate. They found that a shown that a very close representation of the actual stress-
large reduction in interfacial bond strength causes no change strain behavior can be obtained using a nonlinear representation
in the initial stiffness of concrete under short-term compressive for the mortar constituent of the physical model.
224R-6 ACI COMMITTEE REPORT

Tests on mortar showed that nonloaded specimens had about

40% of the crack density of the corresponding cement
paste specimens. As the applied strain was increased,
however, the crack density increased more rapidly in the
mortar, eventually surpassing the value obtained in the cement
paste. While sand particles can reduce crack density due
to volume changes in cement paste, these results indicate
that they act as stress raisers when load is applied. This
increase in crack density under applied loading may explain
the reduction in ultimate strain capacity exhibited in Fig. 2.2
(Martin, Darwin, and Terry 1991) for mortar, compared with
cement paste at the same w/c.
Using analytical procedures, Attiogbe and Darwin (1988)
established that a significant portion of the nonlinear strain
in cement paste and mortar can be attributed to the microcracks
within the cement paste.
Overall, the damage to cement paste in compression seems
Fig 2.4—Crack through calcium silicate-hydrate and calcium
hydroxide in cement paste (Attiogbe and Darwin 1987). to play a dominant role in controlling the primary stress-
strain behavior of concrete under compression. In normal-
weight concrete, aggregate particles act as stress risers,
Maher and Darwin also studied the behavior of the mortar increasing the initial stiffness and decreasing the strength
constituent of concrete under monotonic and cyclic com- of the paste and controlling the compressive strength of the
pression (1982). Degradation in mortar was shown to be a concrete. An understanding of concrete behavior in compres-
continuous process and a function of both total strain and sion, thus, requires an understanding of both the behavior of ce-
load history. The study indicated that residual strain as well ment paste in compression and the interaction of cement
as the change in the initial modulus of elasticity are good paste with aggregate particles.
measures of structural change within the material. Accumu-
lations of residual strain were obtained for values of maxi- 2.3—Fracture
mum strain as low as 0.00027. The work showed that the 2.3.1 Applicability of linear elastic fracture mechanics—
maximum strain alone does not control the degradation of The fracture toughness of a brittle material, which is charac-
mortar in compression and that the total strain range (both terized by a critical stress-intensity factor KIc can be mea-
loading and unloading) adds to the degradation in stiffness sured by using a single-edge notched beam subjected to a
and accumulation of residual strain. Their work concludes as monotonically increasing load. The load is applied so that a
was previously observed (Meyers, Slate, and Winter 1969; constant rate of crack-mouth-opening displacement (CMOD)
Shah and Chandra 1970; Ngab, Slate, and Nilson 1981) that is maintained. If the load-CMOD curve is linear, LEFM can
bond and mortar microcracking in concrete is a function of be used to calculate KIc based on the measured maximum load
the compressive strain in the concrete and is independent of and the length of the crack just before failure (ASTM E 399).
the method in which the strain is applied. Because the maxi- KI c is used in the design of metal structures to prevent brittle
mum strain does not appear to completely control degrada- failure where fatigue crack growth is expected to occur. For
tion, factors other than bond and mortar cracks can dominate LEFM to be applicable, however, the value of KIc should
the degradation of concrete during cyclic loading. be a material property independent of the specimen geometry.
Martin, Darwin, and Terry (1991) studied the behavior of When KI c was calculated for concrete, as described previ-
paste, mortar, and concrete under cyclic and short-term sus- ously, significant effects of the size and geometry of the test
tained compression. They found a great similarity in the be- specimen were observed by many investigators (Kaplan
havior of concrete and its mortar constituent although the 1961; Naus and Lott 1969; Higgins and Bailey 1976). The
bond and mortar microcracking found in concrete were not data presented in Fig. 2.5 (Higgins and Bailey 1976) shows
observed in the mortar specimens. Of the three materials stud- that KI c increases with the specimen depth. Such results led
ied, cement paste has the greatest strain capacity and strength, many to question the applicability of LEFM to concrete.
followed by mortar and concrete (Fig. 2.2). Results obtained from single-edge notched beams were also
To understand the compressive response of the cement analyzed by several investigators to determine if concrete dis-
paste and mortar constituents of concrete, Attiogbe and plays any notch sensitivity. Notch sensitivity can be expressed
Darwin (1987, 1988) used the SEM to study submicro- as the ratio of net stress at the crack tip to the modulus of rup-
scopic cracking under uniaxial compression (Fig. 2.4). Ma- ture of an unnotched specimen. Data on the notch sensitivity
terials with water-cement ratios (w/c) of 0.3, 0.5, and 0.7 of hardened cement paste, mortar, and concrete are shown in
were subjected to monotonic, cyclic, and short-term sustained Fig. 2.6 (Higgins and Bailey 1976; Kesler, Naus, and Lott
loading. Their observations showed that most cracks in 1972; Shah and McGarry 1971; Gjørv, Sorenson, and Arneson
cement paste range in width from 0.2 to 0.7 µm (8 to 28 × 10-5 1977; Hillemeier and Hilsdorf 1977). The specimens showing
in.) and in length from 10 to over 200 µm (4 to over 80 × 10-4 in.). no notch sensitivity are likely the result of deficiencies in the
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-7

Fig. 2.5—Size effect on stress-intensity factor (based on data from Higgins and Bailey 1976).

Fig. 2.6—Effect of relative notch depth on notch sensitivity (based on data from Higgins
and Bailey 1976; Kesler, Naus, and Lott 1972; Shah and McGarry 1971; Gjørv, Soren-
son, and Arneson 1977; Hillemeier and Hilsdorf 1977).

test methods, as explained by Gjørv et al. (1977). The results large concrete structures, such as dams, where the heteroge-
indicate, however, that both mortar and concrete display less neities and the fracture process zone are small compared
notch sensitivity than hardened cement paste. It is widely with the structure dimensions.
accepted today that this lower notch sensitivity for the relatively 2.3.2 Nonlinear fracture models for concrete—The inap-
more heterogeneous materials, particularly concrete, is due to plicability of LEFM to laboratory-size concrete specimens is
the fact that LEFM is inapplicable for laboratory-size the result of the heterogeneity inherent in the concrete. This
specimens of these materials (Gjørv et al. 1977; Wittmann heterogeneity results in a relatively large fracture process
1986). It is also widely accepted (Linsbauer et al. 1989a, zone that results in a substantial amount of crack growth
1989b), however, that LEFM is a valid tool for analyzing (crack extension) preceding the critical (maximum) load and
224R-8 ACI COMMITTEE REPORT

Fig. 2.7—Precritical crack growth (John and Shah 1986).

Fig. 2.8—Normalized peak stress versus crack width in unaxial tension (Gopalratnam and
Shah 1986).

is responsible for the strong dependence of KIc on the size the test specimen. Laboratory-size specimens satisfy this cri-
and geometry of test specimens. Precritical crack growth terion for metals. For concrete, Bažant (1979) stated that the
(crack extension) for a notched beam test is shown in Fig. 2.7, fracture process zone has a negligible effect if the cross-
where the crack growth ahead of the notch was continuously sectional dimensions of a member is at least 100 times the
monitored using a specially developed brittle crack gage (John maximum aggregate size, which would lead to prohibitive
and Shah 1986). size requirements. For instance, concrete with 19 mm (3/4 in.)
The fracture process zone in concrete is substantially dif- aggregates would require a beam with a depth of at least of 2 m
ferent from the plastic zone in metals. For metals, the plastic (6 ft). In view of these specimen size requirements, when
zone is defined as a zone where the material has yielded LEFM is not applicable for many of the fracture tests that
ahead of the crack. LEFM is based on the assumption that the have been conducted on concrete. Therefore, if laboratory-size
plastic zone is substantially smaller than the dimensions of specimens are used to evaluate the fracture toughness of
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-9

(a)
(a)

(b)
Fig 2.9—(a) Fictitious crack model; and (b) crack band
model.

concrete, it is imperative that the effect of the process zone (b)

is considered. Fig. 2.10—(a) Effective Griffith crack; and (b) typical plot
Figure 2.8 shows the results of a uniaxial tensile test of load versus CMOD (Jenq and Shah 1987).
conducted by Gopalaratnam and Shah (1986). The average
(surface) crack opening displacements during this test
were measured microscopically. The peak of the curve, in a finite element model that essentially matches the results of
shown in Fig. 2.8 at zero displacement, is assumed to be the fictitious crack model. In the crack band model, the crack
equal to the tensile strength of the concrete, and the area un- is represented by an equivalent change in material properties
der the curve is considered to be the fracture energy of the within an element. In Fig. 2.9(b), the figure on the left-hand side
concrete Gf . is analogous to the figure on the left-hand side of Fig. 2.9(a),
Hillerborg, Modeer, and Petersson (1976) developed the showing a variation in stress along the crack front as a function
fictitious crack model, which has been used for finite ele- of location. The right-hand portion of Fig. 2.9(b) shows the
ment analysis of concrete fracture. Figure 2.9(a) illustrates stress-strain curve that defines the behavior of an element as
the basic concept of the approach. For a beam in flexure, the the crack grows. The rising portion of the stress-strain curve
left-hand portion of Fig. 2.9(a) shows the variation in stress is used to simulate a slowly opening crack. The product of the
along the crack path, reaching a peak at the fictitious crack strain ε f shown in Fig. 2.9(b) and the width of the finite ele-
tip, where the stress is equal to f t ′ (the tensile strength of the ment Wc is equal the crack opening displacement δ c shown in
concrete), and the CTOD is zero. Moving to the left of the Fig. 2.9(a). When used in conjunction with the two material
peak, the stress drops as the crack opens, with the real crack properties used for the fictitious crack model, Gf and f t ′ , the
ending at the point where the stress across the crack has two procedures produce nearly identical results (Leibengood,
dropped to zero. To the right, the stress drops in advance of Darwin, and Dodds 1986).
the crack. The material between the real and fictitious crack 2.3.3 Nonlinear fracture models based on adaptation of
tip transmits tensile stress as defined by a (softening) stress- LEFM—Several investigators have proposed the use of an
crack opening displacement curve, such as Fig. 2.8 and the effective crack length ae to account for the fracture process
right-hand portion of Fig. 2.9(a). If the shape of this soften- zone (Catalano and Ingraffea 1982; Nallathambi and Karih-
ing curve is assumed to be fixed, then the fracture of the con- aloo 1986; Refai and Swartz 1987). The effective crack
crete is completely characterized by f t ′ and Gf. length is obtained from the reduction in stiffness at the peak
Bažant and Oh (1983) developed a crack band model to load in a three-point bend test. The effective crack depends on
account for the fracture process zone in concrete in a the maximum grain size of the aggregate and on the geometry
smeared manner through the introduction of a strain-softening of the specimen. The term ae is obtained by comparing the
constitutive relation. In this model, the crack front has a width compliance of the test specimen with compliances obtained
of Wc that is equal to the width of a single finite element (Fig. from a series of prenotched beams. When KIc is calculated
2.9[b]). The crack band model is designed to produce a response using the effective crack length, a size-independent value is
224R-10 ACI COMMITTEE REPORT

P = maximum load (that is, failure load);

b = thickness;
d = characteristic dimension of the specimen or structure;
ft ′ = direct tensile strength; and
B, do = empirical constants, do being a certain multiple of the
maximum size of inhomogeneities in the material da .
The value of B and the ratio of do /da depends only on the
shape of the structure, not on its size. Figure 2.11 shows
the relationship between nominal stress at failure and size.
If the structure is very small, the second term in parenthe-
ses, d /do of Eq. (2-1), is negligible compared with 1, and
σΝ = Β f t ′ is the failure condition that represents the strength
criterion and corresponds to the horizontal line in Fig. 2.11.
If the structure is very large, 1 is negligible compared with d/do
and σΝ = constant / d . This is the typical size effect in LEFM;
it corresponds to the inclined straight line in Fig. 2.11.
According to Eq. (2-1), the size effect in blunt fracture
Fig. 2.11—Size-effect law (Bažant, Kim, and Pfeiffer 1986). represents a gradual transition from the strength criterion to
the energy criterion of LEFM.
obtained. Refai and Swartz (1987) developed empirical The size-effect law has been used by Bažant and Sun
equations that relate effective crack length with specimen (1987); Bažant and Sener (1988); and Bažant, Sener, and
geometry and material properties. Pratt (1988) to predict the size effects for shear, torsion, and
bond pullout testing of concrete.
Jenq and Shah (1987) proposed a method to determine the
2.3.5 Effect of material properties on fracture—Certain
effective crack length, which is then used to calculate a crit-
s and a critical crack tip opening material properties, especially w/cm, play an important role
ical stress-intensity factor KIc
in controlling the compressive strength and durability of
displacement (CTOD). Figure 2.10 illustrates the effective
concrete. The effect of these material properties on the
crack-length concept. The effective crack length concept it-
fracture of concrete are not certain; however, some studies
self is the sum of a measurable crack, visible on the side of a
have specifically addressed this question. Early work by
specimen, plus the additional crack length represented by the
Naus and Lott (1969) indicated that the fracture toughness of
fracture process zone. The effective crack length is evaluated
cement paste and mortar increases with decreasing w/cm, but
using the unloading compliance measurement Cu of the
w/cm has little effect on the fracture toughness of concrete.
load-CMOD curve at the point of maximum load, as shown
Naus and Lott found that KIc increases with age and decreases
in Fig. 2.10(b). Jeng and Shah found that the effective crack
with increasing air content for paste, mortar, and concrete. The
length calculated from compliance measurements is the
fracture toughness of mortar increases with increasing sand
same as that obtained using LEFM and assuming that CTOD
content, and the fracture toughness of concrete increases
has a critical value, which was found to be independent of the
with an increase in the maximum size of the coarse aggre-
size and geometry of the beams tested and may be considered
gate. Gettu, Bažant, and Karr (1990), in a study of the frac-
to be a valid fracture parameter.
ture properties of high-strength concrete, made a number of
2.3.4 Size effect of fracture—The effect of structural size observations that match those obtained in the earlier work.
on the fracture of concrete is perhaps the most compelling They observed that the fracture toughness and fracture energy
reason for using fracture mechanics (ACI 446.1R). obtained with high-strength concrete is not much higher than
For blunt fracture (as occurs in a crack with a diffuse fracture that for lower-strength concrete, and any increase that occurs
process zone in materials such as concrete), the total potential- is at a rate less than in proportion to the square root of
energy release caused by fracture in a given structure depends compressive strength. The work by Gettu, Bažant, and
on the length of the fracture and the area traversed by the frac- Karr (1990) was carried out with mixtures that maintained
ture process zone so that the size of the fracture process zone is a constant maximum-size aggregate. When the results of
constant and independent of the size of the structure. Dimen- their work are combined with the typical procedure of using
sional analysis then shows that the structural size effect smaller maximum-size aggregate for high-strength concrete,
for geometrically similar specimens or structures is governed it becomes clear that improvements in compressive strength,
by the simple relation (Bažant, Kim, and Pfeiffer 1986) obtained with the use of increased cement contents, mineral
admixtures, high-range water-reducers, and with the ac-
Bf t′ companying reduction in total aggregate volume, will not
σ N = -----------------------------
- (2-1) increase fracture toughness. The result is that structural
( 1 + d ⁄ do )
members made with high-strength concrete will exhibit a
lower-than-expected capacity when the member strength
where depends on the concrete tensile strength, and the design is
σΝ = P/bd = nominal stress at failure; based on f c′ . Specific examples are flexural cracking,
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-11

shear strength, and bond strength between concrete and

reinforcing steel. The impact of using high-strength concrete on
these load-carrying mechanisms needs additional study.

CHAPTER 3—CONTROL OF CRACKING DUE TO

DRYING SHRINKAGE
3.1—Introduction
Drying shrinkage of concrete is the reduction in volume
caused by the loss of water. Drying shrinkage can be defined
as the time-dependent linear strain at constant temperature
measured on an unloaded specimen that is allowed to dry.
From a structural point of view, there is no need to separate
drying shrinkage from other kinds of phenomena, such as
carbonation shrinkage and autogenous shrinkage. A typical
value for the final shrinkage strain of concrete in structures
is 600 × 10-6. Because the concrete tensile-strain capacity
can be 150 × 10-6 or less, cracking will result if the shrinkage
is restrained in a concrete member. There is a high degree of
uncertainty in predicting shrinkage of concrete structures,
however, because this property varies considerably with
many parameters, including concrete composition, source of
aggregate, ambient relative humidity, specimen geometry,
and more specifically, the ratio of the exposed surface to the
volume of the structural element. Further, the slow development
of shrinkage over time makes it difficult to obtain an accurate
prediction for a given concrete from short-term laboratory
measurements. As a result, a coefficient variation of 20% or Fig. 3.1—Cracking of concrete due to drying shrinkage.
more can be expected in predicting long-term shrinkage.
Before true moisture equilibrium has been reached within concrete. The combination of shrinkage and restraint devel-
a member cross section, internal shrinkage restraint occurs ops tensile stresses within the concrete. Due to the inherent low
because of moisture gradients. Consequently, self-equilibrating tensile strength of concrete, cracking will often occur (Fig. 3.1).
internal stresses are present with tension on the surface and Additional restraint arises from nonuniform shrinkage.
compression in the interior. This stress condition can cause Because drying occurs nonuniformly from the surface towards
cracking if not relieved by creep. the concrete core, shrinkage will create internal tensile stresses
Shrinkage and creep are often responsible for excessive near the surface and compression in the core. Differential
deflections and curvature, losses in prestress, and redistribu- shrinkage can result in warping and surface cracks. The surface
tion of internal stresses and reactions in statically indetermi- cracks can, with time, penetrate deeper into the concrete
nate members. If not controlled, drying shrinkage can lead to member as the interior portion is subject to additional
serviceability problems, such as excessive deflections, and shrinkage.
durability problems, such as freeze-thaw deterioration and As illustrated in Fig. 3.2, the tensile stress induced by
corrosion at cracks. restraining drying shrinkage is reduced with time due to
Good design and construction practices can minimize creep or stress relaxation. Cracks develop only when the net
the amount of cracking and eliminate or control the visible tensile stress reaches the tensile strength of concrete. The creep
large cracks by minimizing the restraint using adequate relief decreases with age, however, so that the cracking ten-
reinforcement and contraction joints. Further information dency becomes greater with increased time.
can be found in ACI 209R. Cracking due to drying shrinkage
can never be eliminated in most structures. This chapter cov- 3.3—Drying shrinkage
ers cracking of hardened concrete due to drying shrinkage, When concrete dries, it contracts or shrinks. When it is
factors influencing shrinkage, control of cracking, and the wetted, it expands. The expansion does not occur to the same
use of expansive cements to minimize cracking. Construc- extent as shrinkage. These volume changes, along with
tion practices and specifications to minimize drying shrink- changes in moisture content, are an inherent characteristic of
age are covered in Chapter 8. hydraulic-cement concrete. The change in moisture content
of cement paste causes concrete to shrink or swell. Aggre-
3.2—Cause of cracking due to drying shrinkage gate reduces the unit volume of cement paste and provides an
The contraction (due to drying shrinkage) of a concrete internal restraint that significantly reduces the magnitude of
component within a structure is always subject to some these volume changes in concrete.
degree of restraint from either the foundation, another In addition to drying shrinkage, the cement paste is also
part of the structure, or the reinforcing steel embedded in the subject to carbonation shrinkage. Shrinkage results from the
224R-12 ACI COMMITTEE REPORT

Fig. 3.2—Effect of creep on tensile stress.

Fig. 3.3—Relations between shrinkage and time for concretes stored at different relative
humidities. Time reckoned since end of wet curing at 28 days (Troxell, Raphael, and Davis
1958).

effects of carbon dioxide on the chemical changes of calcium- the shrinkage of the concrete in the structure. This effect
silicate hydrate and crystalline-hydration products and the results from the greater surface area to volume ratio in
drying of the pores by removing absorbed water. Calcium smaller specimens. Shrinkage due to carbonation is discussed in
hydroxide will form calcium carbonate by reacting with detail by Verbeck (1958).
atmospheric carbon dioxide. Because carbon dioxide does
not penetrate more than about 12 mm (0.5 in.) into the surface 3.4—Factors controlling drying shrinkage
of concrete
of high-quality concrete with low porosity, carbonation The major factors controlling ultimate drying shrinkage of
shrinkage is of minor importance in the overall shrinkage concrete include relative humidity, aggregate type and con-
of most concrete structures. Carbonation does, however, play tent (or paste content), water content, and w/cm. The rate of
an important role in the shrinkage of small laboratory test moisture loss and shrinkage of a given concrete is influenced
specimens and structures constructed with low-quality, by the size of the concrete member, the relative humidity,
porous concrete, particularly when subjected to long-term distance from the exposed surface, and drying time.
exposure to drying. The amount of carbonation shrinkage 3.4.1 Relative humidity and drying time—Relative humidity
observed on a small laboratory specimen can be greater than has a major influence on ultimate shrinkage and the rate of
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-13

Fig. 3.4—Effect of relative aggregate content and modulus ratio on drying shrinkage of
concrete (Hansen and Almudaiheem 1987).

Table 3.1—Effect of aggregate type on concrete

shrinkage (after Carlson [1938])
Aggregate Specific gravity Absorption 1-year shrinkage, %
Sandstone 2.47 5.0 0.116
Slate 2.75 1.3 0.068
Granite 2.67 0.8 0.047
Limestone 2.74 0.2 0.041
Quartz 2.66 0.3 0.032

shrinkage. Results by Troxell, Raphael, and Davis (1958)

showed that the lower the relative humidity, the greater the
ultimate shrinkage and rate of shrinkage (Fig. 3.3). Figure 3.3
also illustrates that expansion occurs if concrete is exposed to a
continuous supply of water; this process is known as
swelling. Swelling is small compared with shrinkage in Fig 3.5—Typical effect of water content of concrete on drying
ordinary concrete and occurs only when the relative humidity shrinkage (USBR 1981).
is maintained above 94% (Lorman 1940). Swelling can, how-
ever, be significant in lightweight concrete (Neville and (Hansen and Almudaiheem 1987). This is also illustrated in
Brooks 1985). Figure 3.3 also shows that drying is a slow Fig. 3.4, where an elastic modulus ratio between 1 and 2
process. It can take many years before ultimate shrinkage indicates an aggregate stiffness that is much smaller than
is reached because the loss of water from hardened concrete is that of normalweight aggregate.
diffusion controlled. Pickett (1956) and Hansen and Almudaiheem (1987)
3.4.2 Influence of quantity and type of aggregate on developed constitutive models for predicting the influence of
shrinkage—Concrete shrinkage is due primarily to shrinkage of relative aggregate content and modulus ratio on ultimate
the hardened cement paste. The presence of aggregate in con- concrete shrinkage. The latter model clearly explains why
crete reduces the total shrinkage by providing elastic re- lightweight concrete for the same relative aggregate content
straint to paste shrinkage. Concrete shrinkage, however, is exhibits considerably more shrinkage than ordinary concrete.
not solely related to the relative aggregate content; there is This is also illustrated in Fig. 3.4 when the modulus ratio
another effect due to the ratio of elastic modulus of aggregate is between one and two because the aggregate stiffness is
to that of the hydrated paste. When using high-quality aggre- much smaller than that of normalweight aggregate.
gates, which are characterized mainly by low absorption The influence of aggregate-absorption capacity on concrete
capacity, this ratio is typically between four and seven shrinkage was investigated by Carlson (1938) and is illustrated
224R-14 ACI COMMITTEE REPORT

3.4.3 Paste content and w/cm—Consistency, as measured

by the slump test, is an important parameter in proportioning
concrete. The amount of mixing water needed to achieve a
given slump is dependent on the maximum aggregate size
used because the maximum size influences the total aggregate
surface area that needs to be covered with cement paste.
Decreasing maximum aggregate size increases the total
surface area to be covered with paste. Therefore, more water
and cement are needed to achieve a given slump. For the
same w/cm, concrete shrinkage increases with increasing
water content because the paste volume increases; this
agrees with the predictions in Fig. 3.4 and results obtained by
the U.S. Bureau of Reclamation (1975) shown in Fig. 3.5.
For a constant w/cm, there is an approximately linear rela-
tionship between water content (paste content as well) and
concrete shrinkage within the range of water contents listed.
Fig. 3.6—Effect of temperature of fresh concrete on its Temperature also has an influence on the water requirements
water requirement (USBR 1981). of the fresh concrete for same slump (Fig. 3.6). A reduction
in water content, which reduces the paste content, will re-
duce the ultimate drying shrinkage of concrete. Therefore,
the water content (and paste content) of a concrete mix-
ture should be kept to a minimum to minimize potential dry-
ing shrinkage and the cracking tendency of the concrete.
Figure 3.7 illustrates that concrete shrinkage increases
with w/cm for a given aggregate content. This effect is more
pronounced with lower aggregate contents (Odman 1968).
3.4.4 Influence of member size—The size and shape of a
concrete member and the porosity of the cement paste influ-
ences the drying rate of concrete and, therefore, influences
the shrinkage rate. The shape affects the ratio of the surface
area to volume of the member, and a higher ratio results in a
higher drying rate. For a given concrete, the observed shrinkage
at a given time decreases with an increase in the size of the
specimen. This effect is illustrated in Fig. 3.8 (Bryant and
Vadhanavikkit 1987) in which long-term shrinkage results
Fig. 3.7—Influence of w/c and aggregate content on shrinkage were obtained on concrete prisms up to 400 mm (8 in.) thick.
(Odman 1968). Ultimate shrinkage may not be reached for structural members
during the intended service life.
in Table 3.1; the concrete had identical cements and w/cms. The Another consequence of moisture diffusion is that a mois-
ture gradient develops from the surface to the interior. For a
absorption of an aggregate, which is a measure of porosity, in-
specimen that has moisture evaporation from all surfaces,
fluences its modulus or compressibility. A low elastic modu-
shrinkage strain is greatest at the surface where moisture
lus is usually associated with high absorption. content is lowest, and shrinkage strain decreases toward the
Quartz, limestone, dolomite, granite, feldspar, and some center where moisture content is highest. Nonuniform self-
basalts can be classified as higher-modulus aggregates, equilibrating internal stresses develop. Tensile stresses occur
which result in lower shrinkage properties of concrete. High- at and near the surfaces and compressive stresses develop at
shrinkage concrete often contains sandstone, slate, horn- and near core, as shown in Fig. 3.9.
blende, and some types of basalts. Because the rigidity of Warping occurs if drying takes place in an unsymmetrical
certain aggregates, such as granite, limestone, or dolomite, manner, either due to drying from one side or due to a non-
can vary over a wide range, their effectiveness in restraining symmetrical structure. In slabs-on-grade, the warping mech-
drying shrinkage varies. anism is a primary cause of cracking. Moisture evaporates
from the top surface only, which causes higher shrinkage at
Although compressibility is the most important property
the top. The concrete near the top surface is partially re-
of aggregate governing concrete shrinkage, the aggregate strained from shrinking because it is attached to concrete
itself can shrink during drying. This is true for sandstone lower in the slab that is more moist and does not shrink as
and other aggregates of high-absorption capacity. In general, much as the top surface. This restraint produces tensile
aggregate with a high modulus of elasticity and low absorption stresses at and near the top surface, which results in the slab
will produce a concrete with low ultimate shrinkage. warping or curling, and the free edges of the slab can lift off
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-15

Fig. 3.8—Influence of specimen size on shrinkage (Bryant and Vadhanavikkit 1987).

Fig 3.9—Internal restraint of shrinkage.

the ground. If the edges of the slab are restrained from move- concerned, prolonged moist curing may not be beneficial. A
ment, such as footings, and the slab is not allowed to warp, general recommendation is to continue moist curing for at
then the top surface has higher tensile stresses. Cracking can least 7 days. (For further information, refer to ACI 309.)
result if the tensile stresses from restrained shrinkage exceed Sealed curing is curing without loss or addition of water.
the tensile strength of the concrete. Cracking may also result It eliminates other kinds of shrinkage so that all the resulting
near the edge of the slab when a vertical load is applied on shrinkage will be autogenous. Autogenous shrinkage is a
the warped cantilever. result of the fact that the products of hydration occupy a
3.4.5 Effect of curing on shrinkage—Carlson (1938) reported smaller volume than the original volume of cement and water.
that the duration of moist curing of concrete does not have Self-dessication is a problem in low w/c concretes under sealed
much effect on ultimate drying shrinkage. Test results from conditions in which the pores dry out and hydration slows
the California Department of Transportation (1963) show down. Autogenous shrinkage strain is typically about 40 to
that substantially the same shrinkage occurred in concrete 100 × 10-6 (Davis 1940). Houk, Paxton, and Houghton (1969)
that was moist-cured for 7, 14, and 28 days before drying found that autogenous shrinkage increases with increasing
started. As far as the cracking tendency of the concrete is temperature, cement content, and cement fineness.
224R-16 ACI COMMITTEE REPORT

Fig. 3.11—Length-change characteristics for shrinkage-

compensating and portland cement concrete (relative
humidity = 50%).

Fig. 3.10—Basic concept of shrinkage-compensating concrete. maximum-size aggregate. A lower water content can be
achieved by using a well-graded aggregate, stiffer consistency,
and lower initial temperature of the concrete.
3.4.6 Effect of admixtures—The effect of admixtures on
concrete shrinkage is unclear. As an example, early-age Concrete can withstand higher tensile strains if the stress
shrinkage appears to increase by about 100% in the presence is slowly applied; therefore, it is desirable to prevent rapid
of calcium chloride, whereas later-age shrinkage is increased drying of concrete. Prevention of rapid drying can be attained
by about 40% compared with control specimens (ACI 212.3R). by using curing compounds, even after water curing.
Air-entrainment does not seem to increase shrinkage by 3.5.2 Reinforcement—Properly placed reinforcement,
more than 10% for air contents up to about 5% (Carlson 1938). used in adequate amounts, will reduce the number and
Results by Ghosh and Malhotra (1979), Brooks, Wain- widths of cracks, reducing unsightly cracking. By distribut-
wright, and Neville (1979), and Feldman and Swenson ing the shrinkage strains along the reinforcement through
(1975) indicated that the use of high-range water-reducing bond stresses, the cracks are distributed so that a larger num-
admixtures increases shrinkage. According to Ytterberg (1987), ber of narrow cracks occur instead of a few wide cracks.
high-range water-reducing admixtures do not necessarily Although the use of reinforcement to control cracking in
reduce shrinkage in proportion to their ability to reduce a relatively thin concrete section is practical, it is not needed
water content. in massive structures, such as dams, due to the low drying
shrinkage of these mass concrete structures. The minimum
3.5—Control of shrinkage cracking amount and spacing of reinforcement to be used in structural
Concrete tends to shrink due to drying whenever its sur- floors, roof slabs, and walls for control of temperature and
faces are exposed to air of low relative humidity or high shrinkage cracking is given in ACI 318 or in ACI 350R. The
winds. Because various kinds of restraint prevent the con- minimum-reinforcement percentage, which is between 0.18
crete from contracting freely, cracking should be expected, and 0.20%, does not normally control cracks to within gen-
unless the ambient relative humidity is kept near 100%. The con- erally acceptable design limits. To control cracks to a more
trol of cracking consists of reducing the cracking tendency to a acceptable level, the percentage requirement needs to exceed
minimum, using adequate and properly positioned reinforce- about 0.60%.
ment, and using contraction joints. The CEB-FIP Model 3.5.3 Joints—The use of joints is the an effective method
Code (1990) gives quantitative recommendations on the of preventing the formation of unsightly cracking. If a
control of cracking due to shrinkage by listing various coef- sizeable length or expanse of concrete, such as walls,
ficients to determine the shrinkage levels that can be expected. slabs, or pavements, is not provided with adequate joints to
Control of cracking by correct construction practices is accommodate shrinkage, the concrete will make its own
covered in Chapter 8. joints by cracking.
Cracking can also be minimized by using expansive cements Contraction joints in walls are made, for example, by
to produce shrinkage-compensating concrete. This is discussed fastening wood or rubber strips to the form, which leave
in Section 3.6. narrow vertical grooves in the concrete on both faces of the
3.5.1 Reduction of cracking tendency—Most measures wall. Cracking of the wall due to shrinkage should occur at
that can be taken to reduce concrete shrinkage will also reduce the grooves, relieving the stress in the wall and preventing
the cracking tendency. Drying shrinkage can be reduced by the formation of unsightly cracks between the joints. These
using less water in the mixture and the largest practical grooves should be sealed to prevent moisture penetration.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-17

Sawed joints are commonly used in pavements and slabs-on- zone of a reinforced beam starts at stress levels as low as
grade. Joint location depends on the particulars of place- 20 MPa (3000 psi) in the reinforcement. Crack control is
ment. Each element should be studied individually to deter- also important to aesthetics of exposed concrete surfaces.
mine where the joints should be placed. ACI 224.3R The role of cracks in the corrosion of reinforcing steel is
discusses the use of joints in concrete construction. Guidance controversial (ACI 222R). One viewpoint is that cracks re-
on joint sealants and contraction joint location in slabs is avail- duce the service life of structures by permitting more rapid
able in ACI 504R and ACI 302.1R. penetration of carbonation and allow chloride ions, moisture,
and oxygen to reach the reinforcing steel. Another point of
3.6—Shrinkage-compensating concrete view is that while cracks accelerate the onset of corrosion,
Shrinkage-compensating concrete made with expansive the corrosion is localized. With time, chlorides and water
cements can be used to minimize or eliminate shrinkage penetrate uncracked concrete and initiate more widespread
cracking. The properties and use of expansive cement con- corrosion. Consequently, after a few years of service, there
crete are summarized in ACI 223, ACI 223 (1970), ACI SP- is little difference between the amount of corrosion in
38, and ACI SP-64. Of the several expansive cements pro- cracked and uncracked concrete. More important parameters
duced in the past, Type K shrinkage-compensating cement for corrosion protection are concrete cover and concrete quality.
(ASTM C 845) is currently the only one available in the This chapter is concerned primarily with cracks caused by
United States. Several component materials are available to flexural and tensile stresses, but temperature, shrinkage,
produce shrinkage-compensating concrete. shear, and torsion can also lead to cracking. Cracking in certain
In reinforced shrinkage-compensating concrete, the expan- specialized structures, such as reinforced concrete tanks, bins,
sion of the cement paste during the first few days of hydration silos, and environmental structures is not covered in this re-
will develop a low level of prestress, inducing tensile stresses in port. Cracking of concrete in these structures is described by
the steel and compressive stresses in the concrete. The level of Yerlici (1975), and in ACI 313 and ACI 350R.
compressive stresses developed in the shrinkage-compensating Extensive research studies on the cracking behavior of
concrete ranges from 0.2 to 0.7 MPa (25 to 100 psi). Normal beams have been conducted over the last 50 years. Most of
shrinkage occurs when water starts to evaporate from the the work conducted before 1970 was reviewed by ACI
concrete. The contraction of the concrete will result in a Committee 224 (1971) in ACI Bibliography No. 9. Additional
reduction or elimination of its precompression. The initial work is referenced in this chapter. Leonhardt (1977 and 1988)
expansion of the concrete reduces the magnitude of any presents an extensive review of cracking in reinforced- and
tensile stress that develops due to restrained shrinkage. This prestressed-concrete structures. The CEB-FIP Model Code for
basic concept of using expansive cement to produce a Concrete Structures (1990) gives the European approach to
shrinkage-compensating concrete is illustrated in Fig. 3.10. crack width evaluation and permissible crack widths.
To allow for adequate expansion, special details may be The basis for codes of practice, both in the U.S. and Europe,
needed at joints. to limit service-load cracking is rooted in equations to predict
A typical length-change history of a shrinkage-compensating crack widths. Several of the most important crack-prediction
concrete is compared to that of a portland cement concrete in equations are reviewed in this report. The trend in reinforced-
Fig. 3.11. The amount of reinforcing steel normally used in and prestressed concrete design to ensure acceptable cracking
reinforced concrete made with portland cements is usually more at service loads is to provide proper detailing, such as provi-
than adequate to provide the elastic restraint needed for sion of minimum reinforcement and proper selection of bar
shrinkage-compensating concrete. To take full advantage diameters, bar spacing, and reduction of restraint rather than
of the expansive potential of shrinkage-compensating concrete trying to make use of a sophisticated crack calculation
in minimizing or preventing shrinkage cracking of exposed (Schlaich, Schafer, and Jennewien 1987; Halvorsen 1987).
concrete surfaces, it is important that positive and uninterrupted Fiber-reinforced polymer (FRP) bars have been used as a
water curing (wet covering or ponding) be started immediately reinforcing material (Nawy and Neuwerth 1977, Dolan
after final finishing. For slabs on well-saturated subgrades, 1990). Experience is limited, however, and crack control in
curing by sprayed-on membranes or moisture-proof covers structures reinforced with these materials is not addressed in
has been successfully used. Inadequate curing of shrinkage- this report.
compensating concrete can result in an insufficient expansion
to elongate the steel and subsequent cracking from drying 4.2—Crack-control equations for reinforced
shrinkage. Specific recommendations and information on concrete beams
the use of shrinkage-compensating concrete are contained A number of equations have been proposed for predicting
in ACI 223R. crack widths in flexural members; most of them were re-
viewed in the original version of this committee report (ACI
Committee 224 1972) and in key publications listed in the
references. Crack control is provided by calculating the
CHAPTER 4—CONTROL OF CRACKING IN
FLEXURAL MEMBERS probable crack width and proportioning structural elements
4.1—Introduction so that the computed width is less than some predefined value.
The control of cracking can be as important as the control Most equations predict the probable maximum crack width,
of deflection in flexural members. Cracking in the tension which usually means that about 90% of the crack widths in
224R-18 ACI COMMITTEE REPORT

the member are below the calculated value. Research, how- dc = thickness of cover from the extreme tension fiber to
ever, has shown that isolated cracks in beams in excess of the closest bar, in.
twice the computed maximum can occur (Holmberg and When the strain εs in the steel reinforcement is used instead
Lindgren 1970) although generally, the coefficient of varia- of stress fs, Eq. (4-2) becomes
tion of crack width is about 40% (Leonhardt 1977). There is
evidence that this range in crack width variability can increase
with the size of the member (ACI Committee 224 1972). w = 2.2 βε s 3 d c A (4-2b)
Crack-control equations are presented in the sections that
follow. Eq. (4-3) is valid in any system of units.
4.2.1 ACI approach through ACI 318-95—Requirements The cracking behavior in thick one-way slabs (span-depth
for flexural crack control in beams and thick one-way slabs ratio 15 to 20) is similar to that in shallow beams. For one-
(span-depth ratio in the range of 15 to 20) are based on the way slabs with a clear concrete cover in excess of 25.4 mm
statistical analysis (Gergely and Lutz 1968) of maximum (1 in.), Eq. (4-2) can be properly applied if β = 1.25 to 1.35
crack-width data from a number of sources. Based on the is used.
analysis, the following general conclusions were reached: ACI 318-95 Section 10.6 uses Eq. (4-2) with β = 1.2 in the
• The reinforcing steel stress is the most important variable; following form
• The thickness of the concrete cover is an important
variable but not the only geometric consideration;
z = fs 3 dc A (4-3)
• The area of concrete surrounding each reinforcing bar
is also an important geometric variable;
• The bar diameter is not a major variable; and and permits the calculation of z with fs equal to 60% of the
• The ratio of crack width at the surface to that at the specified yield strength fy in lieu of exact calculation.
reinforcement level is proportional to the ratio of the In ACI 318-95 and earlier code versions, the maximum al-
nominal strain at the surface and the reinforcement lowable z = 175 kips per in. for interior exposure corre-
strain. sponds to a probable crack width of 0.41 mm (0.016 in.).
The equations that were considered to best predict the This level of crack width may be excessive for aesthetic
probable maximum bottom and side crack widths are concerns.
ACI 318 has allowed a value of z = 145 kips per in. for ex-
–3
terior exposure based on a crack width value of 0.33 mm
w b = 0.091 3 t b A β ( f s – 5 ) × 10 (4-1a) (0.013 in.). While application of Eq. (4-2a) ((Eq. 10-4) of
ACI 318-95) to beams gives adequate crack-control values,
its application to one-way slabs with standard 20 mm (3/4 in.)
0.091 3 t b A –3
- ( f s – 5 ) × 10
w s = ------------------------- (4-1b) cover and reinforced with steel of 60 ksi (400 MPa) or lower
1 + ts ⁄ h1 yield strength results in large reinforcement spacings. The
provisions of Section 7.6.5 of ACI 318-95, however, directly
where limit the spacing of such reinforcement in one-way slabs.
wb = most probable maximum crack width at bottom of ACI 340R contains design aids for the application of
beam, in.; Eq. (4-3).
ws = most probable maximum crack width at level of 4.2.2 ACI 318-99 approach—ACI Committee 318 now
reinforcement, in.; believes that it can be misleading to purport to effectively
fs = reinforcing steel stress, ksi; calculate crack widths, given the inherent variability in
A = area of concrete symmetric with reinforcing steel cracking. The three important parameters in flexural crack-
divided by number of bars, in.2; ing are steel stress, cover, and bar spacing. Steel stress is the
tb = bottom cover to center of bar, in.; most important parameter.
ts = side cover to center of bar, in.; A reevaluation of cracking data (Frosch 1999) provided a
β = ratio of distance between neutral axis and tension new equation based on the physical phenomenon for the
face to distance between neutral axis and reinforc- determination of the flexural crack widths of reinforced
ing steel about 1.20 in beams; and concrete members. This study showed that previous crack
width equations are valid for a relatively narrow range of
h1 = distance from neutral axis to the reinforcing steel,
covers (up to 63 mm [2.5 in.]).
in.
ACI 318-99, Section 10.6, does not make a distinction
Simplification of Eq. (4-1a) yielded the following equation
between interior and exterior exposure. It requires that for
crack control in beams and one-way slabs, the spacing of
–3
w = 0.076 β f s 3 d c A × 10 (4-2a) reinforcement closest to a surface in tension shall not exceed
that given by
where
w = most probable maximum crack width, in.; and s ( in. ) = [ ( 540 ⁄ f s ) – 2.5c c ] (4-4a)
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-19

Table 4.1—Guide to reasonable* crack widths, w k ≤ w lim (4-6)

reinforced concrete under service loads
Crack width
where wlim = nominal limit value of the crack width speci-
Exposure condition in. mm
fied for cases with expected functional consequences of
Dry air or protective membrane 0.016 0.41
cracking (such as conditions stipulated in Table 4.1). In the
Humidity, moist air, soil 0.012 0.30
absence of specific requirements, such as water tightness or
Deicing chemicals 0.007 0.18
specific exposure classes as tabulated in the CEB Code, a
Seawater and seawater spray, wetting and drying 0.006 0.15
limiting value of wlim equal to 0.30 mm (0.012 in.) is satis-
Water-retaining structures† 0.004 0.10
factory with respect to appearance and ductility.
* It
should be expected that a portion of the cracks in the structure will exceed these The length ls,max in Eq. (4-5) can be defined as
values. With time, a significant portion can exceed these values. These are general
guidelines for design to be used in conjunction with sound engineering judgement.
†Exclusing nonpressure pipes.
( σ s2 – σ s1 )
l s ,max = 2 ⋅ -------------------------
- ⋅ φs (4-7a)
( 4 τ bk )

but not greater than 12(36/fs) or 12 in., where

where
fs = calculated stress in reinforcement at service load
σs2 = reinforcement stress at the crack location, MPa;
(ksi) = unfactored moment divided by the product
σs1 = reinforcement stress at point of zero slip, MPa;
of steel area and internal moment arm. Alternatively,
φs = reinforcing bar diameter or equivalent diameter of
fs can be taken as 0.60;
bundled bars, mm;
cc = clear cover from the nearest surface in tension to the
τbk = lower fractile value of the average bond stress, MPa
fy

flexural tension reinforcement, in.; and

= 1.8 fctm(t); and
s = center-to-center spacing of flexural tension reinforce- fctm(t) = the mean value of the concrete tensile strength at
ment nearest to the surface of the extreme tension the time that the crack forms.
face, in.
For stabilized cracking, the expression can be simplified
The SI expression for the reinforcement spacing in Eq. (4-4a) as follows
( fs in MPa) is
φs
s ( mm ) = [ ( 95 ,000 ⁄ 540f s ) – 2.5c c ] (4-4b) l s ,max = ----------------- (4-7b)
3.6 ρ s ,ef

but not to exceed 300(252/fs) mm.

For single-crack formation, Eq. (4-6) is expressed as
4.2.3 CEB-FIP and Eurocode EC2 recommendations—
Other organizations around the world have developed proce-
φs
dures for predicting crack widths in structural concrete rang- l s ,max = σ s2 -------------------------------------
- (4-8)
ing from conventionally reinforced through partially and 2 τ bk ( 1 + n ρ s ,ef )
fully prestressed. ACI 318 procedures only deal with con-
ventionally reinforced concrete. Crack-control recommen- The term can be assumed equal to 1.0 for simple calculation,
dations proposed in the European Model Code for Concrete n being the modular ratio Es/Ec, where
Structures (CEB-FIP 1990; Euro EC2 1997) apply to pre- ρs,ef = effective reinforcement ratio, As /Ac,ef ;
stressed as well as reinforced concrete with modifications As = area of tension reinforcement, mm2; and
and can be summarized in the following sections. Ac,ef = effective concrete area in tension, mm2.
[Link] CEB-FIP 1990 provisions—The characteristic The effective area of concrete in tension can be calculated
crack width wk in beams is expressed as follows in terms of as
the length ls,max over which slip occurs between the steel
reinforcement and the concrete (approximating crack A c,ef = b [ 2.5 ( h – d ) ] (4-9)
spacing in stabilized cracking)
where
w k = l s ,max ( ε sm – ε cm – ε cs ) (4-5) b = beam width at the tension side;
h = total section depth; and
where d = effective depth to the centroid of the tensile reinforce-
εsm = average reinforcement strain within segment length, ment.
ls,max; For stabilized cracking, the average width of the crack can
εcm = average concrete strain within segment length, ls,max; be estimated on the basis of the average crack spacing such
and that
εcs = strain of concrete due to shrinkage.
The characteristic crack width wk cannot exceed the limit- 2
S rm = --- l s,max (4-10)
ing crack with wlim , namely 3
224R-20 ACI COMMITTEE REPORT

Table 4.2—Maximum bar diameter for high bond bars σs = stress in the tension reinforcement computed on the
Steel stress, MPa Maximum bar size, mm basis of a cracked section, MPa;
160 32 σsr = stress in the tension reinforcement computed on the
200 25 basis of a cracked section under loading conditions
240 20 that cause the first crack, MPa;
280 16 β1 = coefficient accounting for bar bond characteristics
320 12 = 1.0 for deformed bars and 0.5 for plain bars;
360 10 β2 = coefficient accounting for load duration
400 8 = 1.0 for single short-term loading and 0.5 for sus-
450 6 tained or cyclic loading; and
Es = Modulus of elasticity of the reinforcement, MPa.
Table 4.3—Maximum bar spacing for high bond bars The average stabilized mean crack spacing srm is evaluat-
ed from the following expression
Maximum bar spacing, mm
Steel stress, MPa Pure flexure Pure tension
160 300 200
s rm = 50 + 0.25k 1 k 2 d b ⁄ ρt , mm (4-13)
200 250 150
240 200 125 where
280 150 75 db = bar diameter, mm;
320 100 — ρt = effective reinforcement ratio = As /Act ; the effective
360 50 — concrete area in tension Act is generally the concrete
area surrounding the tension reinforcement of depth
where Srm is the mean crack spacing value (mm) in the beam. equal to 2.5 times the distance from the tensile face
of the concrete section to the centroid of the reinforce-
[Link] Eurocode EC2 provisions—The Eurocode
ment. For slabs where the depth of the tension zone
EC2 requires that cracking should be limited to a level
may be small, the height of the effective area should
that does not impair the proper functioning of the structure
not be taken greater than [(c – db)/ 3], where c = clear
or cause its appearance to be unacceptable (Euro EC2 1997;
cover to the reinforcement, mm;
Beckett and Alexandrou 1997; Nawy 2001). It limits the
maximum design crack width to 0.30 mm (0.012 in.) for sus- k1 = 0.8 for deformed bars and 1.6 for plain bars; and
tained load under normal environmental conditions. This k2 = 0.5 for bending and 1.0 for pure tension.
ceiling is expected to be satisfactory with respect to ap- In cases of eccentric tension or for local areas, an average
pearance and durability. Stricter requirements are stipulated value of k2 = (ε1 + ε2 ) / 2ε1 can be used, where ε1 is the
for more severe environmental conditions. greater and ε2 the lesser tensile strain at the section bound-
The code stipulates that the design crack width be evaluated aries, determined on the basis of cracked section.
from the following expression In the absence of rigorous computations as described thus
far, choice of minimum area of reinforcement As for crack
control is stipulated such that
w k = β s rm ε sm (4-11)
A s = k c kf ct, eff A ct ⁄ σ s (4-14)
where
wk = design crack width; where
srm = average stabilized crack spacing; As = reinforcement area within the tensile zone, mm;
εsm = mean strain under relevant combination of loads Act = effective area of concrete in tension, mm;
and allowing for the effect such as tension stiffen- σs = maximum stress permitted in the reinforcement af-
ing or shrinkage; and ter the formation of the crack. The yield strength
β = coefficient relating the average crack width to the may be taken in lieu of σs, although lower values
design value may be needed to satisfy crack width limits;
= 1.7 for load-induced cracking and for restraint fct,eff = tensile strength of the concrete effective at the for-
cracking in sections with minimum dimension in mation of the first crack. A value of 3 MPa (435 psi)
excess of 800 mm (32 in.). can be used;
The strain εsm in the section is obtained from the following kc = coefficient representing the nature of stress distri-
expression: bution,
= 1.0 for direct tension and 0.4 for bending; and
2 k = coefficient accounting for nonuniform stresses due
ε sm = σ s ⁄ E s [ 1 – β 1 β 2 ( σ sr ⁄ σ s ) ] (4-12) to restraint resulting from intrinsic or extrinsic
deformation. It varies between 0.5 and 1.0 (N/ mm2 =
where 1 MPa).
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-21

The EC2 Code also stipulates that for cracks dominantly crack-control check should be made. Subscripts 1 and 2 per-
caused principally by flexure, their widths will not usual- tain to the directions of reinforcement.
ly exceed the standard 0.30 mm (0.012 in.) if the size and For simply supported slabs, the value of k should be mul-
spacing of the reinforcing bars are within the range of values tiplied by 1.5. Interpolated k values apply for partial restraint
in Tables 4.2 and 4.3 for bar size and spacing (Euro EC2 at the boundaries. For zones of flat plates where transverse
1997; Beckett and Alexandrou 1997; Nawy 2001). For severe steel is not used or when its spacing s2 exceeds 305 mm (12 in.),
exposure conditions, such as those listed in Table 4.1, crack use s2 = 305 mm (12 in.) in the equation.
width computations become mandatory. If strain is used instead of stress, Eq. (4-15) becomes

4.3—Crack control in two-way slabs and plates

Crack-control equations for beams underestimate the w = k 1 βε I (4-16)
crack widths developed in two-way slabs and plates (Nawy
and Blair 1971) and do not indicate to the designer how to where values of k1 = 29 × 103 times the k values previously
space the reinforcement. The cracking widths in two-way listed. Nawy (1972) and ACI 340.1R contain design aids for
slabs and plates are controlled primarily by the steel stress applying these recommendations.
level and the spacing of the reinforcement in the two perpen-
Tam and Scanlon (1986) present a model for determining
dicular directions. In addition, the clear concrete cover in
deflection of two-way slabs subjected to transverse loads.
two-way slabs and plates is nearly constant (20 mm [3/4 in.]
Their model accounts for the net effect on deflection of both
for most interior structural slabs), whereas it is a major vari-
restraint cracking and flexural cracking.
able in the crack-control equations for beams.
Analysis of data on cracking in two-way slabs and plates
(Nawy and Blair 1971) has provided the following equation 4.4—Tolerable crack widths versus exposure
conditions in reinforced concrete
for predicting the maximum crack width
Table 4.1 presents a general guide for what could be
considered reasonable crack widths at the tensile face of
w = k βfs I (4-15) reinforced concrete structures for typical conditions.
These reasonable crack width values are intended to serve
where the terms inside the radical are collectively termed the only as a guide for proportioning reinforcement during
grid index: design. They are to be used as a general guideline along
with sound engineering judgment.
d b1 s 2 s1 s 2 dc 8 The table is based primarily on Nawy (1968), who com-
I = ------------ = --------------
- --- piled information from several sources. It is important to
ρ t1 d b1 π
note that these crack width values are not always a reliable
indication of the corrosion and deterioration to be expected.
k = fracture coefficient with a value k = 2.8 × 10-5 for In particular, a larger cover, even if it leads to a larger surface
uniformly loaded restrained two-way action square crack width, may be preferable for corrosion control in cer-
slabs and plates. For concentrated loads or reactions tain environments; therefore, the designer should exercise
or when the ratio of short to long span is less than engineering judgment on the extent of crack control to be
0.75 but larger than 0.5, a value of k = 2.1 × 10-5 is used. When used in conjunction with the recommendations
applicable. For span aspect ratios less than 0.5, k = presented in Sections 4.2.1 and 4.2.3 to limit crack width, it
1.6 × 10-5; should be expected that a portion of the cracks in the struc-
β = 1.25 (chosen to simplify calculations, although it ture would exceed these values by a significant amount. It is
varies between 1.20 and 1.35); also noted that time effects, such as creep, will cause an in-
fs = actual average service-load stress level or 40% of crease in crack widths that should be taken into account by
the specified yield strength fy, ksi; the designer.
db1 = diameter of the reinforcement in Direction 1 closest Another opinion regarding crack control suggests that in
to the concrete outer fibers, in.; the long term there is no link between the level of flexural
s1 = spacing of the reinforcement in Direction 1, in.; cracking and corrosion (Beeby 1983). This suggests that in-
s2 = spacing of the reinforcement in perpendicular Di- dependent of exposure conditions, the acceptable level of
rection 2, in.; cracking is primarily an aesthetic issue. Therefore, in cases
ρt1 = active steel ratio, that is, the area of steel As per ft such as liquid-containing structures where the presence of
width/ [12db1+ 2c1], where c1 is clear concrete cover moisture is constant or leakage is of concern should more
measured from the tensile face of concrete to the restrictive (smaller) crack widths be required. Based on
nearest edge of the reinforcing bar in Direction 1; information in Halvorsen (1987), a case could be made that
and crack widths ranging from 0.15 to 0.3 mm (0.006 to 0.012 in.)
w = crack width at face of concrete caused by flexure, in. could be considered unacceptable for aesthetic reasons as they
Direction 1 refers to the direction of reinforcement closest to are visible to the naked eye, hence generating a sense of
the outer concrete fibers; this is the direction for which insecurity or structural failure.
224R-22 ACI COMMITTEE REPORT

4.5—Flexural cracking in prestressed concrete Equation (4-17) had better application for most data exam-
Partially prestressed members, in which cracks can appear ined; however, Eq. (4-18) shows better accuracy for wide
under working loads, are used extensively. Cracks form in beams with large spacing. These equations predict the average
these members when the tensile stress exceeds the modulus of the maximum crack widths. The scatter is considerable.
of rupture of the concrete (6 to 9 f c′ psi under short-term The maximum crack width (in in.) at the steel-reinforcement
conditions). The control of these cracks is necessary prima- level closest to the tensile face of the concrete, accounting for
rily for aesthetic reasons, as they are visible to the naked eye, the stress in the reinforcement in pretensioned and post-
hence generating a sense of structural insecurity. The resid- tensioned, fully and partially prestressed members can be
ual crack width, after removal of the major portion of the live evaluated from the following simplified expressions (Nawy
load, is small (about 0.03 to 0.09 mm [0.001 in. to 0.003 in.]) and Huang 1977; Nawy 1989a):
and therefore, crack control is usually not necessary if the Pretensioned beams
live load is transient.
There have been studies concerning the calculation of –5 A
crack widths in prestressed concrete members (Meier and w max = 5.85 × 10 ------t ( ∆ f s ) (4-19)
ΣO
Gergely 1981; Suzuki and Yoshiteru 1984; Suri and Dilger
1986; Nawy 1989a). The complexity of the crack width cal-
culations is increased over reinforced concrete members by Post-tensioned unbonded beams
the number of variables that should be considered.
4.5.1 Crack-prediction equations—One approach to –5 A
w max = 6.51 × 10 ------t ( ∆ f s ) (4-20)
crack prediction for bonded prestressed beams has two ΣO
steps. First, the decompression moment is calculated, at
which the stress in the concrete at the prestressing steel level The maximum crack width at the tensile face of the con-
is zero. Then the member is treated as a reinforced concrete crete can be obtained by multiplying the values obtained
member and the increase in stress in the steel is calculated from Eq. (4-19) and (4-20) by a factor R i where
for the additional loading. The expressions given for crack Ri = ratio h2 / h1;
prediction in nonprestressed beams can be used to estimate h1 = distance from the neutral axis to the centroid of the
the cracks for the load increase above the decompression reinforcement, in.;
moment. A multiplication factor of about 1.5 is needed h2 = distance from the neutral axis to the concrete tensile
when strands, rather than deformed bars, are used nearest to face;
the beam surface in the prestressed member to account for ∆fs = the net stress in the prestressed tendon or the mag-
the differences in bond properties. This approach is compli- nitude of the tensile stress in the conventional rein-
cated if most of the parameters affecting cracking are con- forcement at any load level in which the
sidered (Nilson 1987). An approximate method using the decompression load (decompression here means fc = 0
nominal-concrete-stress approach was presented by Meier at the level of the reinforcing steel) is taken as the
and Gergely (1982). They proposed the following equations reference point, ksi = ( fnt – fd)
for prediction of maximum flexural crack width fnt = stress in the prestressing steel at any load beyond
the decompression load, ksi;
f ct fd = stress in the prestressing steel corresponding to the
w max = C 1 -----d (4-17)
Ec c decompression load, ksi;
∑O = sum of reinforcing elements’ circumferences, in.;
and
f ct 3
w max = C 2 -----d A (4-18) At = the effective concrete area in uniform tension, in.2,
Ec c as defined by ACI 318.
Recent work by Nawy on cracking in high strength pre-
where stressed beams of compressive strength f c′ in excess of 85
C1,C2 = bond coefficients that depend on the type of steel MPa (12,000 psi), showed that the factor in Eq. (4.19), (4.20)
nearest the tension face; becomes 2.75 × 10 –5 in U.S. customary units and 4.0 × 10 –5
fct = nominal tensile stress at the tensile face; in SI units (Nawy, 2000).
Ec = modulus of elasticity of concrete; The CEB Model Code has the same equation for predicting
dc = minimum concrete cover to centroid of steel at the the crack width in prestressed members as in nonprestressed
tensile face; and members (Section 4.2.2). The increase in steel strain is calcu-
A = effective concrete area per bar as defined in ACI lated from the decompression stage. Other equations have
318. been proposed (Abeles 1956; Bennett and Dave 1969; Holm-
Equation (4-17) is dimensionally correct and the coeffi- berg and Lindgren 1970; Rao, Gandotra, and Ramazwamy
cient C1 is dimensionless. In in.-lb units, C1 = 12 and C2 = 8.4 1976; Bate 1958; Bennett and Chandrasekhar 1971; Hutton
for reinforcing bars, and C1 = 16 and C2 = 12 for strands. In SI and Loov 1966; Krishna, Basavarajuiah, and Ahamed 1973;
units, if A is specified in mm2, C1 = 1.39 and C2 = 0.97 for Stevens 1969; Suri and Dilger 1986; Suzuki and Yoshiteru
reinforcing bars, and C1 = 1.85 and C2 = 1.39 for strands. 1984; Harajli and Naaman 1989).
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-23

Aalami and Barth (1989) discuss the mitigation of restraint for which the stirrups are to be provided. Because experi-
cracking in buildings constructed with unbonded tendons. mental evidence shows that higher stresses can result than
Nonprestressed deformed bars can be used to reduce the those indicated by these analyses (Zielinski and Rowe 1960),
width of the cracks to acceptable levels. and because the consequences of under-reinforcement can be
4.5.2 Crack widths—Some authors state that corrosion is a serious, it is advisable to provide more steel than required
greater problem in prestressed-concrete members because of by this type of analysis. More recently, designs have been
the smaller area of steel used and because of the possible conse- based on cracked section analyses. A design procedure for
quences of corrosion on highly stressed steel. Research (Beeby post-tensioned members using a cracked section analysis
1978a, 1978b) indicates that there is no general relationship (Gergely and Sozen 1967) has found acceptance with many
between cracking and corrosion in most circumstances. Poston, designers. For pretensioned members, an empirical equation
Carrasquillo, and Breen (1987), however, cites contradictory has proven to be quite useful (Marshall and Mattock 1962).
laboratory test results on prestressed and nonprestressed Stone and Breen (1984b) present a design procedure for
exposure specimens in which chloride-ion concentration at post-tensioned beam anchorage zones. A general equation is
the level of reinforcement due to penetration of chlorides from given for predicting the cracking load in beams without sup-
external sources was proportional to crack width. Poston and plemental anchorage zone reinforcement along with provi-
Schupack (1990), present results from a field investigation of sions for designing supplementary reinforcement and
pretensioned beams in an aggressive chloride environment in calculating the effect it will have on cracking and ultimate
which brittle wire failure of a seven-wire strand occurred at a load.
flexural crack, apparently due to corrosion with significant Design recommendations for controlling cracking in an-
pitting observed on the other wires at the crack location. The chorage zones of flexural members with closely spaced an-
surface crack widths were 0.13 mm (0.005 in.) or less. The chors, such as in slabs and bridge decks, are provided by
prestressing strand was generally bright on either side of its Burgess, Breen, and Poston (1989) and Sanders, Breen, and
crack with no significant sign of corrosion distress. Duncan (1987).
As discussed by Halvorsen (1987), provisions for sur- Spalling cracks form between anchorages and propa-
face crack-width control as a means of protecting against gate parallel to the prestressing forces and can cause grad-
corrosion should be strongly tied to provisions for high- ual failure, especially when the force acts near and
quality concrete and plenty of cover. The importance of parallel to a free edge. Because analyses show that the
having high-quality (low w/cm) concrete with sufficient spalling stresses in an uncracked member occur primarily
cover to provide long-term protection of steel elements, near the end face, it is important to place the first stirrup
both prestressed and nonprestressed, cannot be overem- near the end surface and to distribute the stirrups over a
phasized. The design should provide more stringent crack distance equal to at least the depth of the member to fully
control than reinforcement spacing stipulated in ACI 318, account for both spalling and bursting stresses. In lieu of
for prestressed-concrete members, and particularly those normal orthogonal reinforcement to control cracking,
subjected to aggressive environments, by providing addi- Stone and Breen (1984a, 1984b) showed the very benefi-
tional mild steel reinforcement, reducing the allowable cial effect of using spiral reinforcement or active rein-
extreme fiber tension stresses under service loads to a val- forcement in the form of transverse prestressing to control
ue below 6 f c′ psi, perhaps as low as 2 f c′ psi, or both, cracking in anchorage zones where the prestressing forces
and to minimize the potential for flexural cracking. are large.

4.6—Anchorage-zone cracking in prestressed 4.7—Crack control in deep beams

concrete Major changes in reinforced concrete design in the
Longitudinal cracks frequently occur in the anchorage past two decades, namely the widespread adoption of
zones of prestressed concrete members due to transverse ten- strength design, have resulted in some structures with
sile stresses set up by the concentrated forces (Gergely 1969; high service-load-reinforcement stresses. Several cases
Zielinski and Rowe 1960; Stone and Breen 1984a). Such have been reported (Frantz and Breen 1980a, 1980b)
cracks can lead to (or in certain cases are equivalent to) the where wide cracks have developed on the side faces of
failure of the member. Transverse reinforcement (stirrups), beams between main flexural reinforcement and the neutral
active reinforcement in the form of lateral prestressing, or axis. Although the measured crack widths at the main rein-
both, should be designed to restrict these cracks. forcement level were within acceptable code limits, the side-
Two types of cracks can develop: spalling cracks that begin face crack widths near middepth were as much as three
at top and bottom beam ends outside the end anchorage zones times as wide.
and propagate parallel to the prestressing force, and bursting Based on an experimental and analytical investigation of
cracks that develop along the line of the force or forces but cracking in deep beams (in the sense of separation of tension
away from the end face. and compression force resultants, not span-depth ratio),
For many years, stirrups were designed to take the entire Frantz and Breen developed recommendations for side-face
calculated tensile force based on the analysis of the uncracked crack control in beams in which the depth d exceeds 915 mm
section. Classical and finite-element analyses (Stone and (36 in.). Modifications of these recommendations have been
Breen 1984a; Nawy 1989b) show similar stress distributions included in ACI 318 since 1989. Section 10.6.7 of ACI 318
224R-24 ACI COMMITTEE REPORT

requires skin reinforcement to be uniformly distributed Bennett and Dave 1969; Holmberg and Lindgren 1970; Ill-
along both faces of the member for a distance d/2 nearest the ston and Stevens 1972; Holmberg 1973). While there is a
flexural tension reinforcement. large scatter in the data, information obtained from sustained
loading tests of up to 2 years (Illston and Stevens 1972) and
4.8—Tension cracking fatigue tests with up to 1 million cycles (Bennett and Dave
The cracking behavior of reinforced concrete mem- 1969; Holmberg 1973; Rehm and Eligehausen 1977) indi-
bers in axial tension is similar to that of flexural mem- cate that a doubling of crack width with time can be expected.
bers, except that the maximum crack width is larger than Under most conditions, the spacing of cracks does not change
that predicted by the expressions for flexural members with time at constant levels of stress (Abeles, Brown, and
(Broms 1965a,b). The lack of strain gradient and result- Morrow 1968; Illston and Stevens 1972; Holmberg 1973).
ant restraint imposed by the compression zone of flexur- An exception to this occurs at low loads or in beams with
al members is probably the reason for the larger tensile high percentages of reinforcement, in which case the total
crack width. number and width of cracks increase substantially after the
Data are limited, but it appears that the maximum ten- loading has begun (Brendel and Ruhle 1964; Abeles, Brown,
sile crack width can be about expressed in a form similar and Morrow 1968; Holmberg 1973). The largest percent-
to that used for flexural crack width age increase in crack width is then expected in flexural
members subjected to low levels of load because the
–3 cracks take more time to develop.
w = 0.10f s 3 d c A × 10 (4-21)
For both prestressed and reinforced concrete flexural
members, long-term loading and repetitive loading give
where crack width is measured in in. about the same crack widths and spacing (Rehm and Elige-
A more complicated procedure for predicting crack width hausen 1977). The rate of crack width development, however,
in tension members has been developed that incorporates is considerably faster under repetitive loading (Bennett and
both slip and bond stress (Yang and Chen 1988). Although Dave 1969; Holmberg 1973; Rehm and Eligehausen 1977;
the crack width prediction equation appears to show good Stevens 1969).
agreement with available test data, the procedure is too com- As discussed in Chapter 4, crack width is a function of
plicated for design purposes. A similar approach was also cover. For short-term static and fatigue loading, surface
developed for predicting crack widths in concrete tension crack width is approximately proportional to the steel strain
members reinforced with welded-wire fabric (Lee et al. (Illston and Stevens 1972; Holmberg 1973; Stevens 1969).
1987). A more complete discussion of concrete cracking in di- Crack widths increase under sustained loading at a decreas-
rect tension is provided in ACI 224.2R. ing rate. The rate of growth in crack width, however, is faster
than the average observed surface strain at the level of the
CHAPTER 5—LONG-TERM EFFECTS ON steel. For long-term loading, crack width is proportional
CRACKING
5.1—Introduction to the steel strain (including the effects of creep), plus the
Cracking in concrete is affected by the long-term conditions strain induced in the concrete due to shrinkage (Illston
to which the concrete element is subjected. In most cases, and Stevens 1972).
long-term exposure and long-term loading extend the magni- Under initial loads, cracks intercepting reinforcement are
tude of cracks, principally their width, in both reinforced and restricted by the bond between the steel and the concrete (Ill-
plain concrete. The discussion in this chapter summarizes the ston and Stevens 1972; Broms 1965b), and the width of sur-
major long-term factors that affect the crack-control perfor- face cracks does not provide a good indication of the
mance of reinforced and prestressed concrete. exposure of the reinforcing steel to corrosive conditions.
Over a period of time, however, the adhesion bond between
5.2—Effects of long-term loading the steel and the concrete undergoes breakdown. After about
As discussed in Chapter 2, both sustained and cyclic load- 2 years, the crack width at the reinforcement is approximate-
ing increase the amount of microcracking. Microcracking ly equal to the crack width at the surface (Illston and Stevens
appears to be a function of the total strain and is largely in- 1972). At this stage, cracks in flexural members are triangu-
dependent of the method by which the strain is induced. lar in shape, increasing in width from the neutral axis to the
Microcracks formed at service load levels do not seem to soffit and are approximately uniform across the width of the
have a great effect on the strength or serviceability of rein- beam.
forced and prestressed concrete.
The effect of sustained or repetitive loading on macro- 5.3—Environmental effects
scopic cracking, however, can be an important consideration The long-term effects of an adverse environment in both
in the serviceability of reinforced concrete members, espe- producing and in enlarging concrete cracks (Mather 1957,
cially in terms of corrosion of reinforcing steel and appear- 1968) can be damaging to both concrete and reinforcement.
ance. The increase in crack width due to long-term or If concrete is not resistant to freezing and thawing when crit-
repetitive loading can vary between 100 and 200% over sev- ically saturated, cracks will develop due to bursting effects
eral years (Bate 1963; Brendel and Ruhle 1964; Lutz, Shar- of the freezing water. The lack of such resistance can be due
ma and Gergely 1968; Abeles, Brown, and Morrow 1968; to the following reasons: the use of resistant-resistant coarse
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-25

aggregate, inadequate air-void system, or failure to protect to be subjected to deicers or similar chemicals should be air
the concrete from freezing before curing. Critical saturation entrained and properly proportioned and cured to produce
in nonfrost-resistant concrete can occur by the presence of low permeability.
preexisting cracks that allow entry of water. The initiation of
D-cracking near joints or other cracks in pavements is a good 5.5—Use of polymers in improving cracking
example. In more extreme cases, it is not uncommon for characteristics
cracks caused either by thermal stress or shrinkage of the Extensive work is available on the use of polymers in
richer topping mixture in the roadway deck of dams and nav- modifying the characteristics of concrete (Brookhaven Na-
igation locks to cause spalling due to the freezing of water in tional Laboratory 1968; ACI SP-40; ACI SP-58; ACI 548R).
the cracks themselves independent of the frost resistance of Polymer-portland cement concretes have a large deforma-
the concrete. On the other hand, pre-existing cracks can also tion capacity, high tensile and compressive strengths, and
function to allow concrete to dry below critical saturation be- negligible permeability. The tensile splitting strength can be
fore freezing when this might not occur in the absence of as high as 10.7 MPa (1550 psi) (Nawy, Ukadike, and Sauer
such cracks. The role of cracks as they affect frost resistance 1977). Polymer impregnation, though rarely used today, is
will vary with the environmental conditions, such as, typical another method of introducing beneficial polymer systems
time of drying after wetting before freezing, crack width, and into concrete. These materials are discussed in greater detail
ability of cracks to drain. in Chapter 6.
Concrete durability is better when the aggregate used is
durable under freezing and thawing conditions and the CHAPTER 6—CONTROL OF CRACKING IN
OVERLAYS
strength of the concrete is appropriate (ACI 201.2R). Field 6.1—Introduction
exposure tests of reinforced concrete beams (Roshore 1967) An overlay can be constructed by placing mortar or con-
subjected to freezing and thawing in an ocean-side environ- crete over a concrete surface. The use of overlays has rapidly
ment indicate that the use of air-entrained concrete made the increased since the early 1970s. They are now commonly
beams more resistant to weathering than the use of non-air- used for rehabilitation of deteriorated bridge decks; strength-
entrained concrete. Beams with modern deformed bars were ening or renovating pavements, warehouse floors, walkways
more durable than those using bars with old-style deforma- and other concrete flatwork; and in new two-course con-
tions. Maximum crack widths did not increase with time struction.
when the steel stress was less than 30 ksi (210 MPa), but did Overlays can be divided into three groups. The first
increase substantially (50 to 100%) over a 9-year period group is when portland cement is used. These overlays
when the steel stress was 30 ksi (210 MPa) or more. can be low-slump dense concrete (LSDC), polymer-mod-
ified concrete (also called latex-modified concrete
5.4—Aggregate and other effects [LMC]), and fiber-reinforced concrete (FRC). These
Concrete can crack as the result of expansive reactions be- overlays may also contain silica fume, fly ash, or granulat-
tween aggregate and alkalis present in the cement hydration, ed blast-furnace slag. The second group includes polymer
admixtures, or external sources, such as curing water, and epoxy mortars or concretes. The third group includes
groundwater, and alkaline solutions stored or used in the fin- polymer-impregnated concrete (PIC), which has not become
ished structure. generally effective, economical, or practical. In a PIC sys-
Possible solutions to these problems include limitations on tem, hardened concrete is impregnated with a low molecular
reactive constituents in the aggregate, limitations on the al- weight monomer that fills small cracks and voids to a shallow
kali content of cement, and the addition of a satisfactory poz- depth (about 5 mm [1/4 in.]) beneath the surface. The mono-
zolanic material or a combination of these. The potential for mer is then polymerized and a relatively impervious surface
some expansive reactions, such as alkali-carbonate, is not re- layer results.
duced by pozzolanic admixtures. ACI 201.2R and Woods If the base slab is relatively crack free, or if the overlay is
(1968) give details on identification and evaluation of aggre- sufficiently thick and strong to resist the extension of cracks
gate reactivity. ACI 221.1R gives guidelines on the alkali- in the original slab, a well-bonded layer with matched joints
aggregate reaction and selectivity process for mixture pro- is generally the best approach. If the overlay has sufficient
portioning and durability. thickness, a totally unbonded overlay is generally best where
Based on ACI 201.2R, ACI 212.3R, ACI 222R, and severe cracking is present or where it can later develop in the
Mather (1957, 1968), the hazard of using calcium chloride, base slab. Systems that are essentially unbonded have been
which may initiate corrosion, warrants a recommendation constructed satisfactorily where the overlay is placed over an
against its use when crack control is a major factor affecting asphalt layer. The asphalt itself acts as a debonding layer if
long-term performance and durability of a structural system. it has a reasonably smooth surface without potholes. This
Also, the use of calcium chloride in reinforced structures ex- type of construction lends particularly well to deteriorated
posed to moist environments should be avoided regardless of airfield slabs that have been resurfaced with asphaltic con-
the presence or absence of water-soluble salts in adjacent crete but require additional rigid pavement to take the in-
waters and soil. creased loads of heavy aircraft. Another technique that has
Detrimental conditions can also result from the application been used when the material to be overlaid is reasonably
of deicing salts to the surface of hardened concrete. Concrete smooth consists of placing the overlay over a polyethylene
224R-26 ACI COMMITTEE REPORT

sheet. On irregular, spalled, or potholed surfaces, a thin leveling overlay over the joint, and quite possibly, some debonding
and debonding layer of asphalt is desirable under the poly- adjacent to the joint. The preferred method is to form
ethylene sheet. the joint with a compressible material and place the
The main causes of cracking in overlays are: overlay against it. After curing, the compressible mate-
• Plastic shrinkage caused by excessive evaporation due rial can be removed and replaced with the final joint
to environmental conditions while the concrete is in its material;
fresh or plastic state; • In new two-course construction of bridge decks, the
• Differential drying shrinkage between material in the overlay should be placed after removing the deck forms
layer and the substrate concrete; and shoring from the base concrete so that stresses
• Differential thermal stresses between the overlay and caused by the weight of the overlay are carried by the
the substrate concrete. This can be caused by a different underlying concrete. If placed before the forms are
temperature in the layer as compared to the substrate removed, the overlay will have to carry a portion of its
and can also be caused or aggravated by different coef- own weight and can crack in negative moment regions;
ficients of thermal expansion and elastic properties; • Overlays should be placed only when the ambient
• Reflective cracking from cracks in the substrate; weather conditions are favorable, as defined in ACI 308
• Edge and corner curling stresses that can lead to delam- or when appropriate actions are taken for hot-weather
inations and other cracking; and (ACI 305R) or cold-weather concreting (ACI 306R).
• Poor construction practices. Evaporation rates of about 1 kg/m3/h (0.2 [lb/ft3]/h),
Long-term observations (Schrader and Munch 1976; Bish- as measured from a free water surface, can cause plastic
ara 1979; Shah and Skarendahl 1986) of many overlays have shrinkage cracking that can increase the extent of cracking
shown that cracking due to differential shrinkage is the most and increase the probability of delamination. Curing
common problem. These cracks are also more likely to in- procedures, such as wet mats and fog spraying, can be
crease or widen with time. Another problem, delamination of required. For large construction projects, such as
the overlay, has been found to occur only at cracks in the pavement overlays, the evaporation rate should be
overlay or at boundaries, normally at very early ages. These monitored to determine when more stringent curing
delaminations will spread with time. procedures should be used; and
To reduce the incidence of cracking in rigid concrete over- • Mechanical shear reinforcement is effective in reducing
lays, the following procedures are recommended: cracking in overlays placed during periods of high
• The surface of the underlying concrete should be thor- evaporation rates.
oughly prepared to ensure adequate bonding of the
overlay. This can be accomplished by mechanical 6.2—Fiber-reinforced concrete (FRC) overlays
methods, such as shotblasting, scabbling, hand chip- When properly proportioned, mixed, and placed, a crack-
ping, or sandblasting, and hydraulically by high-pres- resistant topping layer of FRC can be the solution to certain
sure waterblasting (hydrodemolition). Scarifying field problems. Fibrous concrete overlays of highways, air-
methods that impact the surface can cause cracking in fields, warehouse floors, and walkways have been used since
the substrate that can result in delamination. Procedures the mid-1970s. Fibers are usually steel or polypropylene
for each project should be selected considering the con- with lengths between 10 and 70 mm (1/2 and 2-3/4 in.). The
dition of the concrete, the availability of equipment, effects of fibrous concrete on cracking in an overlay depend
and the environmental conditions. The end result largely on the field conditions in each situation. (Schrader
should be a clean, sound concrete surface; and Munch 1976; Shah and Skarendahl 1986; Shah and Bat-
• All equipment used for mixing, placing, and finishing son 1987; ACI 544.2R; ACI 544.3R; ACI 544.4R)
should be designed for the type of overlay being used The basic concept of FRC—that fibers arrest the growth of
and should be accurately calibrated and in good work- microcracks in concrete—is applicable to steel, synthetic
ing order. Both the contractor and inspecting personnel (such as polypropylene), and mineral (such as glass) fibers.
should be trained in the proper construction techniques Steel fibers have a significant effect on the toughness of the
of the particular overlay system; concrete. Synthetic resin fibers have a lower modulus of
• Material quantities, including total water content, w/cm, elasticity and a poorer bond compared with steel fibers; they
and amount of polymer, should be closely monitored do not corrode but can reduce bleeding and plastic shrinkage
and recorded; cracking. Glass fibers are used primarily in cladding panels
• Traffic control should be evaluated for highway appli- and other precast products that are formed by spraying
cations. The maintenance of traffic during reconstruc- chopped glass fibers and mortar slurry into forms at a precast
tion causes deflections, vibrations, or both in bridge plant. Glass fibers do not mix well in conventional concrete
decks. Consideration should be given to placing over- mixers. There are significant long-term durability problems
lays when traffic is low, when vehicle speed is associated with glass fibers (Hoff 1987; Shah, Ludirja, and
restricted, or both; Daniel 1987).
• Contraction joints in the deck should not be overlaid 6.2.1 Steel fiber concrete bond to underlying concrete—In
unless a joint or saw cut is immediately provided. initial studies of FRC, it was believed that a partially bonded
Delayed saw cutting will usually result in a crack in the layer was the ideal system. The term partially bonded means
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-27

that no deliberate attempt is made to improve the bonding 6.3—Latex- and epoxy-modified concrete overlays
between the topping layer and the underlying material Bonded overlays of styrene-butadiene latex-modified
through bonding agents, fasteners, and polyethylene sheets. mortar and concrete with a minimum thickness of 20 to 40 mm
The surface to be overlaid is cleaned of all loose material, (3/4 to 1-1/2 in.), respectively, have been used in the renovation
usually by hosing, and left in a damp condition. Evaluations of bridge decks and in new two-course construction to effec-
of partially bonded projects have indicated that this is the tively resist the penetration of chloride ions from deicing
least-desirable technique to use. Over a period of years, salts and prevent the subsequent corrosion of the reinforcing
many partially bonded FRC overlays have shown noticeable steel and spalling of the concrete deck (Bishara and Tantay-
amounts of reflective cracking and edge curling. Curled edges anondkul 1974; Clear 1974). Overlays containing water-
are typical in thin overlays (less than about 75 mm [3 in.]), and dispensable epoxy modifiers have also been used successfully,
they can result in cracks. but on a much more limited basis. Latex- and epoxy-modified
6.2.2 Fiber size and volume—The theory of FRC is overlays are discussed in ACI 548R and ACI 548.1R.
based on a crack-arresting mechanism that depends on Inspection of a large number of bridge decks overlaid with
many parameters (Shah and Naaman 1976; Shah and Batson latex-modified concrete (Bishara 1979) revealed fine, ran-
1987). Some of the parameters that influence the reinforcing dom, shrinkage cracks in some projects. This type of crack-
effect of fibers include the fiber’s mechanical properties, as- ing is not as extensive in new two-course construction. The
pect ratio (ratio of fiber length to fiber diameter), and the vol- random shrinkage cracks deserve special comment. At times
ume fraction of fibers (ratio of volume of fibers to volume of they can be attributed to poor control or construction practices,
concrete). Increasing the aspect ratio or the volume fraction such as the use of concrete with a high water content.
of fibers can enhance the crack-arresting mechanism, pro- Placement of an overlay in hot weather without adequate
vided that fibers are uniformly distributed. If the number of protection against early drying is also a cause of plastic
fibers crossing a crack is relatively small, then the crack- shrinkage cracking.
arresting mechanism is limited.
On occasion, random pattern cracks have appeared even
6.2.3 Fiber type and shape—Because their resistance to when the mixture proportions and construction methods
pullout is greater, deformed steel fibers have a significant followed good practice. Transverse cracks, spaced 3 to 4 ft
advantage over smooth ones with regard to both precrack-
(0.9 to 1.2 m) apart, have also been noticed in some bridge
ing and postcracking behavior.
decks. The cracking can be due to unique conditions that
6.2.4 Fibers in open cracks—There has been considerable cause thermal contraction of the surface while the sub-
discussion about the condition and effectiveness of steel fi- strate and bottom portion of the LMC layer does not ex-
bers that cross a crack. At the time of cracking, fibers lose perience similar thermal contraction. This shock usually
their adhesion to the concrete but continue to provide a me- occurs during the first night after placement when the
chanical resistance to pullout. This postcracking strength is
overlay has rigidity but has not yet developed an apprecia-
one of the most important characteristics of FRC, and it can
ble tensile strength. Tight, random pattern and transverse
be significant for deformed fibers. The concern is that after
cracks have caused concern from the standpoint of aes-
cracking, steel fibers will oxidize and provide no long-term
thetics, but they have not been a cause of overlay failure.
benefit. Investigations (Schrader and Munch 1976; ACI
Typically, such cracking is shallow (2 to 10 mm [1/16 to
544.2R; ACI 544.3R), however, have shown that if the crack
3/8 in.]) and stable. A safe, conservative, and recommend-
widths are small (0.03 to 0.08 mm [0.001 to 0.003 in.]), the
ed approach is to treat these cracks with a penetrating high
fibers will not corrode, even after years of exposure (Schrad-
molecular weight methacrylate or low-viscosity epoxy or ure-
er and Munch 1976; Schupack 1985).
thane, which can be broomed on the surface after the curing
6.2.5 Mixture proportioning considerations—Even with a
and drying period but before traffic is allowed on the
high-range water-reducer, the water requirement for fibrous
overlay. The penetrant will generally fill and seal the sur-
concrete is higher than that of the same mixture without
face cracks.
fibers due to reduced slump that accompanies the pres-
ence of fibers. The higher water demand of FRC tends to Finishing and texturing should be done rapidly behind the
cause shrinkage cracks. Through the use of normal- or high- placement operation and before the polymer in the latex be-
range water-reducing admixtures, the mixture water can be gins to dry or coalesce at the surface. Otherwise, tearing,
held to reasonable levels (Walker and Lankard 1977). Ad- scarring, and possible cracking can result. If, for example, a
mixtures should be used to adjust mixture proportioning for rake is used to groove a surface after it has begun to dry, tears
bonded overlays so that the w/cm and cement content ap- about 13 mm (1/2 in.) long and 3 mm (1/8 in.) deep can oc-
proach the same values as used in the underlying material. If cur. These will be oriented at right angles to the direction of
possible, the overlay should have aggregates of similar phys- raking. Texturing can be provided after the concrete has
ical properties, unless the original aggregates are unsuitable. hardened using cutting wheels.
6.2.6 Overlays over joints—Different methods of placing
unjointed overlays over joints in the underlying concrete 6.4—Polymer-impregnated concrete (PIC) systems
have been tried; most have been unsuccessful (ACI 544.4R). Surface impregnation and polymerization of concrete in
As with conventional concrete overlays, if joints exist in a place has been used in a number of field projects (Schrader
base slab, they should be maintained through the overlay. et al. 1978). Practical difficulties were experienced in early
224R-28 ACI COMMITTEE REPORT

projects and it has not become a popular procedure for treat- alkali-aggregate reaction. Surface cracking can appear as
ment of slabs. pattern cracking and result from a decrease in volume of the
material near the surface or increase in volume below the
6.5—Epoxy and other polymer concrete overlays surface. Random cracking from material-related causes can
Epoxy and other polymer concretes and mortars are dis- pass through a massive concrete element, and the crack
cussed in ACI 548R and ACI 548.1R. These materials use a widths can vary from hairline to wide. Structural cracking
monomer or an epoxy as the binder, aggregate as the filler, can result from individual loads or load combinations, such
and no water. Occasionally, portland cement or fly ash is as gravity, liquid pressure, and severe impact. Structural
added as a mineral filler. Overlays made with these materials cracks are of any width but generally align in a structurally
are normally thin and do not use coarse aggregate. Typical plausible direction. The formation of a single, wide crack
applications using smooth surfaces are in food processing usually indicates the existence, before the crack’s formation,
and sanitary or clean rooms or where a floor requires chem- of principal tensile stress perpendicular to the crack. Struc-
ical resistance without a significant increase in thickness. tural distress is also noticeable near changes in geometry.
Textured surface applications include bridge decks, parking
garages, and stadium walkways. Because the reactions that 7.2—Methods of crack control
harden these materials are normally highly exothermic, they Given the probable temperatures and strains, the designer
cannot be used in thick placements or in hot weather without should determine what measures are most practical to pre-
thermal stress problems. vent cracking. Some of the conditions that facilitate crack
Polymers have significantly higher coefficients of thermal prevention are:
expansion than concrete, even when aggregate fillers are • Concrete with large tensile-strain capacity;
used. Changes in temperature create normal and shear stress- • Low cement content (permitted by low design stresses);
es at the interface of the overlay and base slab, which may • Cement of low heat generation or use of pozzolans;
result in cracking or delamination. To reduce cracking in PC • Cast small concrete segments or blocks;
overlays, thin overlays with low elastic modulus polymers • Low placement temperature;
should be used. • Slow rate of construction when no artificial cooling is
Polymer and epoxy concrete overlays can achieve excel- used;
lent bond to dry surfaces. Subsurface preparation techniques • Artificial cooling by an internal network of cold water
that use water should be avoided. These overlays are vapor pipes;
tight and should be carefully evaluated before applying, if • Insulate concrete surfaces;
transmission of water vapor through the overlay is desirable. • Low degree of restraint, as with yielding foundation, or
It is important to evaluate the moisture content of concrete in portions of the structure well removed from restrain-
to be overlaid. This is done by taping a piece of polyethylene ing foundation; and
(mat test) to the concrete. If moisture collects on the underside
• Absence of stress raisers, such as galleries.
within the time frame that polymer or epoxy needs to cure, then
There are two measures that can minimize cracking. The
the concrete should be allowed to continue to dry.
first is to modify the materials and mixture proportions to
produce concrete with the best cracking resistance or the
CHAPTER 7—CONTROL OF CRACKING IN MASS
CONCRETE greatest tensile-strain capacity. This can require careful
7.1—Introduction aggregate selection, using minimum cement content for
7.1.1 Definition of mass concrete—Mass concrete is de- interior concrete, restricting the maximum aggregate size.
fined by ACI 116R as “any volume of concrete with dimen- The attempt made to produce a concrete with a large tensile-
sions large enough to require that measures be taken to cope strain capacity can limit the maximum aggregate size to a
with generation of heat from hydration of the cement and value somewhat below that which might be the most eco-
attendant volume change to minimize cracking.” Mass con- nomical. Where several sources of aggregate are economi-
crete structures consist of concrete dams, powerplants, cal, preference should be given to that which provides the
bridge piers, and other large structural elements. Common to best resistance to cracking. Usually, this will be a crushed
many mass concrete structures, water is directly in contact with material of low thermal expansion and low modulus of
the concrete, creating a high humidity or moist environment. elasticity.
By design, mass concrete water structures have little or no The second measure to prevent cracking is to control the
reinforcing steel. They should resist the loads through shape, factors that produce tensile strain. This may mean precooling,
size, and strength. Consequently, control of cracking is es- postcooling, insulating, or possibly heating the exposed surfac-
sential. Cracking can occur at any time during the life of the es of the concrete and designing to minimize strains around
structure, and engineers should design the structure’s shape galleries and other openings.
and materials to adjust to changing conditions (ACI 207.1R; From these considerations, it is apparent that the degree of
ACI 207.2R; ACI 207.4R; ACI 207.5R). crack control necessary can vary from nothing at all for a
7.1.2 Types of cracks—Cracking in mass concrete struc- dam near the equator or with favorable aggregates to very
tures can result from material, structural causes, or both. Ma- costly measures in a location where temperature variations
terial-induced cracks result from drying shrinkage, a severe are great and where the only economical aggregates have
nonlinear thermal gradient initiated by heat of hydration and high elastic moduli and high thermal expansion.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-29

In the case of a dam, the height affects the need for crack time period is related to the expected time until loading. For
control. If the dam is very high, the design stresses can be example, if full loading is not until 1 year, a lower early-
high and a higher cement content is needed to attain the strength concrete can be prescribed consistent with the con-
strength needed for the required safety margin. This results struction sequence anticipated. Because the cylinders made
in more heat generation and a consequent tendency toward at the site are made from wet-screened concrete, the mea-
higher internal temperatures. Also, the higher dam will have sured strength is corrected to an equivalent mass concrete by
greater horizontal dimensions that cause greater restraint and applying a reduction factor of about 0.80 for typical condi-
slow the rate of heat dissipation, resulting in the need for tions or else developing a factor by comparing wet-screened
closer temperature control. cylinder strengths to those obtained from 12 or 24 in. (300 or
Recommended practice calls for both precooling and post- 600 mm) cylinders. Specific data on appropriate reduction fac-
cooling and for the application of thermal insulation to ex- tors with varying parameters can be found in USBR (1981).
posed surfaces during cold weather. The insulation is left in For interior concrete, the lowest practical strength should
place long enough to permit the concrete temperatures at the be specified to reduce the cement content. This will reduce
surface to slowly approach ambient or until additional con- the heat of hydration and the consequent thermal gradient,
crete is placed on or against the surface being protected. decreasing the likelihood of cracking. More than the neces-
After the anticipated temperature history is established, sary amount of cement is detrimental rather than advanta-
the determination of probable tensile stress is the next step. geous. A compromise between early strength and heat of
This can be accomplished using finite-element computer hydration can often be obtained by replacing a significant
programs (Wilson 1968; Polivka and Wilson 1976; Liu, part of the cement with pozzolan. A one-for-one replacement
Campbell, and Bombieh 1979) that are formulated from a of portland cement with pozzolan can reduce the heat of
heat-transfer viewpoint. Even with the finite-element ap- hydration by one-half of that generated by the same
proach, analysis is laborious because of the time-dependent amount of cement.
variables. Problems include that the data must be generated 7.3.1 Safety against sliding—Uncracked concrete pro-
to establish these time-dependent properties for the con- vides a large factor of safety against [Link] horizontal
cretes to be used. The analysis should include many time lift joints, however, can impair safety. Therefore, the speci-
steps to account properly for creep (or relaxation) and for the fications should require care in the preparation of lift surfaces
different or changing properties of each lift of concrete. and in the placement and compaction of concrete onto them. In
On the other hand, stresses near a boundary due to brief the case of a dam, the lift surfaces should slope slightly
thermal shocks can be computed quite readily (ACI 207.2R) upward toward the downstream edge to improve drainage
because the concrete can be assumed to be fully restrained in during lift surface preparation as well as shear resistance
such cases. If there is full restraint, the strain is zero, and the during operation of the dam. It is not necessary to use a
stress is obtained by multiplying the elastic modulus by the mortar layer on lift surfaces before placing the next lift.
strain that would have occurred if there were no restraint. These conditions are appropriate for conventionally placed
The strain is the temperature drop multiplied by the coeffi- concrete. In roller-compacted concrete construction, a layer
cient of thermal expansion. This situation is important be- of mortar is placed between lifts near the upstream face as a
cause the control of boundary strain is sufficient to control protection against seepage.
cracking in many cases. Internal strains develop slowly 7.3.2 Economy—Many factors that affect economy affect
enough to be tolerable, even if they are large. crack resistance. For example, the least-expensive aggregate
can have undesirable thermal properties and require expen-
7.3—Design sive temperature control to prevent cracking. The aggregate
Design of mass concrete structures means determining the that imparts concrete with the highest tensile-strain capacity
appropriate loads and load combinations together with a can increase the water requirement and the cement require-
reliable method of structural analysis to arrive at an econom- ment, offsetting the benefits of high strain capacity. Similar-
ical geometric shape meeting safety criteria. Loads and com- ly, lesser-quality aggregates can have a lower stiffness than
binations will vary for the different structures, such as the surrounding matrix and not carry their proportionate
hydrostatic pressure, gravity, ice, silt, temperature, earthquake, share of the load.
superstructure weight, differential foundation conditions, and [Link] Selection of aggregates—The aggregates chosen
impact. should be those that make good concrete with the lowest
Material tests are conducted on cylindrical specimens that overall cost. If natural aggregate near the site has unfavor-
are not completely representative of the structure because able properties for crack prevention, crushing to increase
aggregate larger than 40 mm (1.5 in.) is removed from the crack resistance can be an economical expedient because of
mixture by wet screening before 300 mm (6 in.) diameter the consequent saving in temperature control. When crush-
cylinders are prepared (USBR 1981). The maximum stress is ing is either advantageous or necessary, rock that has the
the design stress that is based upon structural requirements. most favorable properties should be chosen. The rock should
It is considered good practice to use a safety factor as high as have a low coefficient of thermal expansion and a low mod-
three or four, meaning that the strength should be three or ulus of elasticity, and it should produce particles with good
four times the expected maximum stress. Strength tests at 2, shape and surface texture. Rock elastic moduli higher than
7, 14, 28, 90, 180 days or 1 year are often used. The selected the matrix will concentrate load on the rock, producing an
224R-30 ACI COMMITTEE REPORT

unequal stress distribution in compression. Stronger rock Table 7.1—Elastic and thermal properties of
will also act as a crack arrestor to cracking of the matrix. mass concrete
These factors are important considerations in increasing the Elastic properties
resistance of the concrete to cracking. Static modulus of elasticity E at indicated age of test, days
Poisson’s
[Link] Aggregate size—The maximum-size aggregate 1 3 7 28 90 ratio
should be determined by the ability to place concrete proper- 46.4 GPa 141 GPa 180 GPa 281 GPa 352 GPa
0.15 to
ly in the structure. Up to a 150 mm (6 in.) diameter aggregate (0.66 × 106 (2.00 × 106 (2.56 × 106 (4.00 × 106 (5.00 × 106 0.25
psi) psi) psi) psi) psi)
can be used, except when concrete should resist high-velocity
Thermal properties
water flow and when structural concrete is needed. Larger
Coefficient of
aggregates permit the use of less water and cement for a given linear expansion Conductivity Diffusivity Specific heat
volume with a reduction in the amount of temperature control 2.00 to 3.31 0.0037 to 0.0062
7 to 15 × 10-6/C 0.22 cal/g·C
required to achieve a particular level of crack resistance. The kg·cal/m·h·C m2/h
advantages of larger aggregates should be weighed against (1.35 to 2.24 (0.040 to 0.067
(4 to 8 × 10-6/F) (0.22 BTU/lb·F)
the need for larger equipment and the potential for workability BTU/ft·h·F) ft2/h)
and segregation problems. The trend in recent years is to limit
maximum-size aggregate to 75 mm (3 in.) for conventional
mass concrete and 50 mm (2 in.) for roller-compacted concrete. An alkali-silica reaction can result when alkali in cement
[Link] Water content—A reduction in the water content (or from other sources) reacts with certain aggregates, such
of concrete permits a corresponding reduction in the cement as argillites, graywakes, phyllites, quartzites, schists,
content. Concrete with less water and cement is superior be- strained quartz, opal, chalcedony, and rhyolites (ACI 221R).
cause it undergoes less temperature change due to hydration Cracking usually appears first on the surface as pattern
effects, less drying shrinkage, and is more durable and crack cracking (ACI 201.1R) within a few years after construction.
resistant. Minimum water content can be achieved by speci- Concrete durability is closely related to exposure condi-
fying adequately powered vibrators that permit the use of tions. In tropical climates, for example, there may be no
low-slump concrete by using a water-reducing admixture deteriorating influences acting on the concrete except cav-
when appropriate and by producing and placing the concrete itation in areas subjected to high-velocity water flow. For the
at lower temperatures. main structure areas not subject to high-velocity water flow,
[Link] Pozzolan—In most cases, good pozzolans, such concrete that has the required strength can be expected to last
as fly ash, are available and can be used to replace a portion well beyond the design life of the structure. In those cases,
of the cement. This can result in a considerable cost saving, cement content should be kept low to minimize heat genera-
and more importantly for mass construction, can reduce the tion and cracking.
heat generation and improve the resistance against cracking. In a severe climate, with repeated freezing and thawing in
Another advantage of using pozzolan is that when used in winter, the w/cm of surface concrete should be kept lower
adequate amounts it can reduce the expansion due to reactive than that necessary for strength alone. Higher-strength con-
aggregates. The appropriate amount of pozzolan for a reac- cretes (28 MPa [4000 psi] at 1 year) are often specified for
tive aggregate should be based on test data obtained with the durability. Air entrainment should be mandatory. For any
specific aggregate, pozzolan, and cement; it can be in the concrete that might be subjected to both freezing and water
range of 25 to 35%. Pozzolans tend to reduce early strength pressure, the w/cm should be about 0.4. The effect of the
gain that could delay form stripping (ACI 221R). higher-quality concrete on thermally induced cracking will
7.3.3 Durability—Durability of portland cement concrete be minimized by keeping the thickness of the boundary be-
is defined as its ability to resist weathering action, chemical tween layers to a minimum, probably 0.6 m (2 ft) or less.
attack, abrasion, or any other process of deterioration 7.3.4 Material properties—The concrete’s properties af-
(ACI 201.2R). The most common weathering action is fect crack control. Concretes differ widely in the amount of
multiple cycles of freezing and thawing of water near the tensile strain they can withstand before cracking. For strain
surface that fracture and deteriorate the concrete over time. that is applied rapidly, the two factors that govern the strain
There are chemicals in the environment that lessen the life capacity are the modulus of elasticity and tensile strength.
of the concrete. Naturally occurring sulfate can combine For strain that is applied slowly, the creep (or relaxation) of
with calcium and alumina in hardened concrete to form the concrete is important.
products that can cause an increase in solid volume accom- When loading is applied rapidly, many tests on lean con-
panied by cracking and deterioration or a substantial reduc- cretes, such as those used for the interior of large dams, have
tion in strength of the paste matrix. Abrasion on dams and shown that tensile failure occurs without much plastic strain.
bridge abutments from abrasive materials carried by flow- For these lean concretes, the tensile strain that the concrete
ing water or from high-velocity flow causing cavitation can withstand is approximately equal to the tensile strength
can be mitigated with the proper mixture, hydraulic design, or divided by the modulus of elasticity of the concrete in ten-
both. sion. It is accurate to assume that the tensile-strain capacity
Mass concrete structures, normally unreinforced, can have an is inversely proportional to the modulus of elasticity of the
outer layer of temperature reinforcement that requires corrosion concrete. The modulus of elasticity of the aggregate is important
protection similar to that of reinforced structural concrete. because of its large effect on the deformability of the concrete.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-31

Tensile strength is also important, and for this reason, (7 to 15 × 10 –6 C). Therefore, in the extreme case where a
crushed aggregates are apt to be superior to natural aggregates concrete has a low tensile strength, a high modulus of
for crack prevention. elasticity, a high coefficient of thermal expansion, and it
[Link] Modulus of elasticity—This subject is discussed in is fully restrained, the concrete can crack when there is a
detail in ACI 207.1R. Table 7.1 shows values of the modulus of rapid drop in temperature of only 6 F (3 C). On the other
elasticity of a particular concrete after various ages of curing. hand, some concretes can withstand a quick drop in temperature
[Link] Crack resistance—The tensile strain that con- of as much as 20 F (10 C), even when fully restrained. More
crete can withstand varies greatly with the composition of data on the thermal expansion of concrete can be found in
the concrete and the strain rate. When strain is applied ACI 207.1R and ACI 207.2R.
slowly, the strain capacity is far greater than when the strain is Tensile stress in mass concrete results mainly from the re-
applied. Therefore, concrete in the interior of a large mass that straint of thermal contraction, and to a far lesser degree, from
must cool slowly can undergo a large strain before cracking. autogenous shrinkage. Drying shrinkage is important be-
If concrete contains rough-textured aggregate with small cause it can cause shallow cracks to occur at the surfaces.
maximum size, the strain capacity will be high. There is an Temperature change is a principal contributor to tensile
optimum, however, with respect to aggregate size. Concrete strain in mass concrete. Predicting probable strain requires
with smaller aggregates requires more cement for a given the expected temperature, which can be determined if the
strength. This results in more heat, a higher maximum tem- adiabatic temperature curve for the concrete is known, as
perature, and greater subsequent strain due to cooling. well as the thermal diffusitivity, boundary temperatures, and
Therefore, these effects can offset the strain increase devel- dimensions. The temperature rise that would occur if there
oped from the use of smaller aggregates. were no heat loss is defined as adiabatic temperature rise.
[Link] Tensile-strain capacity—Tensile-strain capacity The finite-element method can be used for predicting the
tests are performed on plain concrete beams under third- temperature distribution. The main problem is choosing the
point flexural loading. Relatively large beams, ranging from correct boundary temperatures, which often depend upon the
300 × 300 mm (12 × 12 in.) to 600 × 600 mm (24 × 24 in.) in ambient air or water temperatures. It is satisfactory to use air
cross section and 160 to 3250 mm (64 to 130 in.) long are of- temperatures provided by the National Weather Service. For
ten used (Houghton 1976). Strain capacity is determined dams, measured reservoir temperatures are used. Refer to
from these tests under both rapid and slow loading to simulate ACI 207.1R for information on other methods for predicting
both rapid and slow temperature changes in the concrete. temperatures in massive concrete.
Expressed as extreme fiber stresses, the loading rates are The heat-producing characteristics of cement play an im-
generally 0.3 MPa/min (40 psi/min) for rapid loading and portant role in the temperature rise. ASTM C 150 Type IV,
0.17 MPa/week (25 psi/week) for slow loading. low-heat cement is recommended; however, it usually is not
For rapid loading, the strain can be measured using either available. Type II moderate-heat cement is also recommend-
surface or embedded strain gages or meters (Houk, Borge, ed, and it usually is available. Pozzolans can be used to re-
and Houghton 1969; Houghton 1976). Embedded meters are place a portion of the cement to reduce the peak temperature
best for long-term tests. The strain can also be determined due to the heat of hydration (ACI 207.2R). In some cases,
from deflection measurements. The concrete test beam used 35 to 50% of the cement can be replaced by an equal vol-
for determining the strain capacity should be protected by ume of a suitable pozzolan and still produce the specified
wrapping with an impermeable material during the test to strength at 90 days or 1 year. Some of the more common poz-
prevent loss of moisture. Testing should be conducted at a zolans used in massive concrete include calcined clays, di-
constant temperature for maximum accuracy in measure- atomaceous earth, volcanic tuffs and pumicites, and fly ash.
ment. Detailed test procedures can be found in Houk, Borge, The actual type of pozzolan to be used and its appropriate re-
and Houghton (1969) and McDonald, Bombieh, and Sulli- placement percentage are determined by test, cost, and avail-
van (1972). In the preliminary studies of temperature and ability.
construction control plans for mass concrete projects, approxi- [Link] Adiabatic temperature rise—ACI 207.1R describes
mate methods for estimating tensile-strain capacity under test methods and provides data on adiabatic temperature rise
rapid and slow loadings given by Houghton (1976) and Liu of concretes with a single cement content but with different
and McDonald (1978) can be used. types of portland cement. Figure 7.1 gives typical adiabatic
[Link] Thermal properties—Thermal diffusitivity and curves for Type II cement and various quantities of cement
thermal expansion are important in the control of cracking and pozzolan. The data show the effect of pozzolan replace-
due to temperature change; their determination is detailed in ment of cement on temperature reduction.
ACI 207.1R. The approximate range of thermal properties is
shown in Table 7.1. 7.4—Construction
The coefficient of thermal expansion is an important prop- 7.4.1 Basic considerations for construction—Mass con-
erty of concrete. The amount of strain that a temperature crete structures are required to be safe, economical, durable,
change will produce is directly proportional to the coefficient and aesthetically pleasing. Each of these requirements influ-
of thermal expansion of the concrete. The average coefficient ences the crack resistance. The cost will depend on features
of thermal expansion of mass concrete is about 5 × 10 –6 F such as the best choice of aggregates, adequate but not exces-
(9 × 10 –6 C), but the coefficient can range from 4 to 8 × 10 –6 F sive temperature control, and low cement content. Durability
224R-32 ACI COMMITTEE REPORT

Fig. 7.1—Typical adiabatic temperature curves for mass concrete (Houghton 1969).

will depend on the quality of the concrete, the exposure con- precast panels is practical; these are left in place as a perma-
ditions, and the freedom from deteriorating chemical reac- nent part of the structure. The panels should be of good quality
tions. Aesthetics will come from good workmanship, for durability and lightweight to provide thermal insulation.
freedom from cracks and strains, and absence of leakage and Because most cracks originate at boundaries, this method
leaching. It is important to have a comprehensive materials improves crack control. More information on the use of precast
test program to establish necessary control before preparing panels for protecting mass concrete can be found in ACI 347R.
construction controls and specifications. [Link] Artificial cooling—The overall program for
7.4.2 Thermal effects—Temperature-induced cracking in cooling concrete, including important field-control criteria,
a large mass of concrete can be minimized if proper mea- should be determined during the design stage. Precooling
sures are taken to reduce the amount and rate of temperature concrete before placement is accomplished by a variety of
change. Measures commonly used include precooling; post- methods, including evaporative cooling of the stockpiled ag-
cooling, or a combination of the two; and thermal insulation gregates using sprinklers, cooling all ingredients of the mix-
to protect exposed surfaces. The degree of temperature con- ture by liquid nitrogen, or using small ice particles as a
trol necessary to control cracking varies with factors such as replacement of part of the mixing water. When these precau-
location, height, and thickness of the structure, character of tions and natural cooling are ineffective in reducing crack-
the aggregate, properties of the concrete, and the external re- ing, postcooling is prescribed, especially for large concrete
straints. Although a large amount of the data for this chapter dams. Postcooling concrete is accomplished by circulating
has been obtained by experience from using mass concrete in cool liquids (usually water) through pipes embedded in the
dams, it applies equally well in mass concrete used in other concrete. The water source from a nearby lake or river can be
structures, such as steam powerplants, powerhouses, bridge directly circulated or refrigerated as needed.
and building foundations, and navigation locks. Tremie con- Studies made during the design stage will establish items
crete is covered in ACI 304R. Roller-compacted concrete is such as lift height, pipe spacing, water temperature, flow
discussed in ACI 207.5R. rate, acceptable rate of temperature drop (for both rapid and
Precooling the concrete during its production and post- slow drops), and approximate duration of cooling.
cooling it with embedded pipe systems after it is placed are Steep cooling gradients, which can result in cracking the
effective measures. Details on pipe cooling are given in mass, should be avoided. This is particularly true in smaller
Section [Link]. masses where circulation of cooling water should be stopped
Another promising method of crack control is placing when the maximum temperature has been reached and be-
crack-resistant concrete at boundaries (sides and top of lifts). gins to drop. A vulnerable location in pipe cooling systems
Even though the more crack-resistant concrete can be too is centered at the cooling coils where sharp gradients and
costly to be used throughout the structure, it can be used to cracking can be induced if termination of cooling water
limited extent without serious effect on cost. Casting thin circulation is not timely. A primary advantage in artificial
layers of concrete next to the forms is difficult, but the use of cooling is the uniform removal of the heat of hydration and
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-33

the precise control of the rate. In mass concrete, a maximum sufficiently dissipated or slow concrete placement to a rate
rate of cooling of 0.6 C (1 F) per day or less is recommended that interferes with the overall construction schedule.
but should be monitored for suitability during construction. Cooling should continue until the prescribed temperature
Thermal couples should be used in sufficient numbers to is reached. This temperature is dependent on the type of
permit adequate monitoring and control of the internal con- structure and loading. The prescribed temperature can be the
crete temperatures. They should be located to record repre- mean annual temperature or other temperature as a compromise
sentative temperatures. For example, a grid that records for the seasonal variations.
thermal gradients in a concrete dam would consist of three to Some mass concrete structures can have vertical contrac-
five (depending on structure size) transverse planes, each tion joints to facilitate construction; these joints will open as
containing five to 10 thermal couples. Spacing of the thermal the concrete cools. In some dams (especially arch type),
couples should be closer near the dam faces. these contraction joints are later grouted to restore monolithic
Construction drawings should show basic pipe layout, in- behavior.
cluding minimum spacing, the layout at faces, transverse After cooling is completed and the pipe is no longer needed,
construction joints, interior openings and in sloping, partial, it should be grouted full with a cementitious mixture. Grout
and isolated concrete lifts, pipe diameter, risers, and pipe techniques used for post-tensioned concrete should be used.
connections (Houghton 1969). In most of the dam, a uniform [Link] Natural cooling—Thermal insulation on ex-
spacing can be maintained for the cooling pipe, but isolated posed surfaces during cold weather can protect concrete
areas exist in all dams that tend to have a concentration of from cracking if enough insulation is used for a proper length
pipes. These concentrations tend to occur at the downstream of time. If the insulation is sufficient to allow slow cooling,
face of the dam where inlets and outlets to cooling pipes are the tensile strain need never exceed the cracking limit. The
located, adjacent to openings in the dam, and at isolated and concrete can relax at the same rate as the tensile stress tends
sloping lifts of concrete. Proper planning will alleviate many to develop until stable temperatures are reached. If the con-
of the undesirable conditions that can result from these con- crete has a very slow relaxation rate (or creep rate), however,
centrations. For example, it should be determined to what the amount of insulation and the long protection time re-
extent the cost-saving procedure of concentrating cooling quired can make this measure difficult.
pipe inlets and outlets near contraction joints can be permitted In extreme environments where large amounts of insula-
at the face of the dam. Also, it should be decided if cooling tion will be required during severely cold months, it may be
pipes to isolated areas in the foundation and at openings, necessary to remove the insulation in stages as the warmer
such as galleries, can extend from the downstream face of the months approach. Temperatures within the concrete just be-
dam or if a vertical riser should be used. low the insulation should be allowed to slowly approach the
For ease of installation, the pipe used for postcooling environmental temperature to prevent thermal shock, which
should be thinwall tubing. Aluminum tubing is lightweight could induce cracking at or below the surface with possible
and easy to handle; however, breakdown from corrosion is a subsequent deeper penetrations into the mass. Avoid using
potential problem if cooling activities are carried on over a too much insulation or leaving it in place too long, which
period of several months. In this case, steel tubing is pre- could result in stopping the desired cooling of the interior
ferred. Surface connections to the cooling pipe should be re- mass and possibly cause the interior temperature to begin to
movable to a depth of 100 to 150 mm (4 to 6 in.) so that holes increase again.
can be reamed and dry-packed when connections are removed. Insulation can be obtained in a variety of forms and materials.
Coils should be pressure tested for leaks at the maximum It can be either semirigid panel-type material, or foamed spray-
pressure they will receive from the cooling system before on material that becomes semirigid in place. The semirigid
placing concrete. Pressure should also be maintained during panels are usually installed on the inside face of the forms.
concrete placement to prevent crushing and permit early de- Temporary anchors embedded in the newly placed lift of
tection of damage, should it occur. concrete retain the insulation on the concrete surface when
Sight-flow indicators should be installed at the end of each the forms are lifted. The insulation can be easily removed from
embedded pipe coil to permit ready observance of cooling the surface. Roll-out application is particularly applicable for
water flow. In addition to regular observance of flows, water use on horizontal lift joints. It is easy to install and remove
temperatures, and pressures, concrete temperatures should and can be reused many times. Spray-on insulation can be
be observed and recorded at least once daily while the lift is used on either horizontal or vertical surfaces. This type of in-
being cooled. Thermometers are placed at the inlet and outlet sulation is particularly useful for increasing the thickness
of the cooling water system to record the temperature and and effectiveness of insulation already in place and for insu-
change throughout the placement. lating forms. Insulation that permits transmission of light
The refrigeration plant for cooling water can be located rays should not be used because a temperature rise occurs be-
centrally, or several complete smaller portable plants can be tween the insulation and the concrete when the insulation is sub-
used to permit moving the refrigeration system as the dam jected to direct sunlight. Precast panels made of low-
progresses upward. Sufficient standby components, equal in conductance lightweight concrete or regular-weight concrete
capacity to the largest individual refrigeration units, should cast with laminated or sandwich layers of low-conductance
be provided. Insufficient plant size can cause premature cellular concrete are also acceptable for insulation. These pan-
shutdown of cooling before the heat of hydration has been els would then serve as both forms and face concrete.
224R-34 ACI COMMITTEE REPORT

Table 7.2(a)—Sustained modulus stress. Measurement is usually made by means of embedded

Sustained modulus strain meters, although any reliable method of measuring
Es at time of loading in days, GPa (psi) strain can be employed. Butyl rubber is satisfactory for seal-
Time after loading, d 1 3 7 28 ing the specimens, but neoprene should be avoided because
0
47.6 134 183 303 it allows some moisture to escape. Specimens should be
(680,000) (1,920,000) (2,610,000) (4,330,000)
loaded at the same ages as specified for the modulus of
46.2 123 172 263
1
(660,000) (1,760,000) (2,460,000) (3,760,000) elasticity tests, but loading at the early age of 1 day is not al-
44.8 113 151 234 ways practical. Again, the specimens should be large enough
3
(640,000) (1,620,000) (2,150,000) (3,334,000) to permit concrete like that used in the structure. Frequently,
7
44.1 95 139 210 280 × 560 mm (9 × 18 in.) cylinders with 75 mm (3 in.)
(630,000) (1,350,000) (1,980,000) (2,990,000)
maximum-sized aggregate, or 150 × 400 mm (6 × 16 in.)
cylinders with 40 mm (1-1/2 in.) aggregate are used. Useful
Table 7.2(b)—Stress coefficients coefficients for converting concrete creep with smaller-size
Tensile stress coefficient for condition of full restraint and decreasing aggregates to concrete creep for larger aggregate sizes are
temperature for age of concrete at time of loading in days, MPa/C (psi/F) found in ACI SP-9. Figure 7.2 shows typical creep data
Time after loading d 1 3 7 28 obtained from laboratory investigations (McCoy, Thorton, and
0 0.047 (3.7) 0.133 (11.0) 0.1181 (14) 0.300 (24) Allgood 1964). Table 7.2 indicates the values for the sus-
1 0.046 (3.6) 0.122 (9.7) 0.1170 (14) 0.260 (21) tained modulus of elasticity, which are used to develop ten-
3 0.045 (3.5) 0.112 (8.9) 0.1150 (12) 0.231 (18) sile stress coefficients for a temperature drop with full
7 0.044 (3.5) 0.094 (7.4) 0.1138 (11) 0.208 (16) restraint. ACI 209R presents a discussion of the sustained
elastic modulus. A high rate of creep for concrete is helpful
Cooling mass concrete structures can be permitted without when the tensile strain is applied gradually. Because the tensile
insulation in relatively thin sections of about 7 m (20 ft) in strength of concrete is nearly independent of prior loading,
moderate climates where the natural dissipation of heat is creep tends to increase the strain capacity. In the case of Dwor-
gradual and the thermal gradient is not steep. shak Dam, for example, the strain to failure was almost three
7.4.3 Autogenous volume change—Autogenous volume times as great for strain applied over 2 months, compared with
change is the expansion or contraction of the concrete due to quickly applied strain (Houk, Paxton, and Houghton 1970).
causes other than changes in temperature, moisture, or stress; The concrete creep under sustained stress is affected by
therefore, it is a self-induced expansion or contraction. Ex- the stiffness of the aggregate. When the modulus is high, the
pansion can be helpful in preventing cracks, but contraction creep is low and vice versa. The importance of aggregate ri-
increases the tendency to crack. Autogenous volume change gidity on concrete creep can be illustrated by two examples.
is usually measured by strain meters embedded in concrete First, assume that the aggregate and the cement paste have
cylinders that are carefully sealed to ensure that there is no the same modulus of elasticity. When compressive stress is
loss of moisture, and the concrete kept at constant tempera- applied, the stress and corresponding strain will be the same
ture. Measurements begin as soon as the specimens are hard- in the aggregate as in the cement paste. The aggregate does
ened and sealed, and are continued periodically for months. not creep under moderate stress but the paste does, and the
paste that is between aggregate particles relaxes and transfers
7.5—Operation load to the aggregate to maintain equilibrium. This imposes
After construction, the structure is exposed to the annual cy- an elastic strain on the aggregate that accounts for a large
cle of load combinations for which it is designed. During the ini- part of the concrete creep. The amount of elastic strain is di-
tial operation period, which may last for a few months or even rectly related to the modulus of elasticity of the aggregate;
years, the structure and the immediate foundation will undergo the more rigid the aggregate, the lower the creep. Next,
some elastic movement superimposed on a permanent set. Dur- assume that the aggregate has a much higher modulus
ing the succeeding years, small permanent deformations occur than the cement paste. When compressive stress is applied,
in the structure, foundation, or both, that are categorized as the average stress in the aggregate will be higher than that in
creep. Creep can initiate unanticipated cracking by causing a the paste and the paste will creep less than it did when the
load shift to structural areas not intended to deform or carry moduli were equal. The elastic strain in the aggregate due to
more of the loads. These conditions can be somewhat anticipat- the creep of the paste will then be less than it was when the
ed with analyses that include concrete creep and designing the moduli were equal. Therefore, an increase in the rigidity of the
structure to accommodate these movements. aggregate acts in two ways to reduce the creep of the concrete.
7.5.1 Concrete creep—Creep is the continued deforma-
tion of concrete during sustained stress. ASTM C 512 details CHAPTER 8—CONTROL OF CRACKING BY
a standard test for determining concrete creep in compres- PROPER CONSTRUCTION PRACTICES
sion. Concrete creep in tension is difficult to measure; therefore, 8.1—Introduction
creep properties in tension and compression are assumed to Construction practices discussed in this chapter include
be the same. specifications, materials, designs, mixture considerations,
Concrete creep is measured on carefully sealed specimens and on-the-job construction performance. Before discussing
stored at constant temperature and loaded by a constant the control of cracking by proper construction practices, it is
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-35

Fig. 7.2—Typical concrete creep curves for mass concrete.

worthwhile to mention the basic cause of cracking related to concrete is not strong enough to withstand the tensile stresses
volume change of concrete—restraint. If all parts of the con- developed.
crete in a concrete structure are free to move as the concrete A wall or parapet anchored along its base to the founda-
expands or contracts due to volume change, there will be no tion, or to lower structural elements less responsive to vol-
cracking. ume changes, can be restrained from shrinking when its
Obviously, however, all parts of concrete structures are upper portions shorten due to drying or cooling (ACI
not free, and inherently, cannot be free to respond in the 224.3R). Cracking is usually inevitable. Due consideration
same degree to volume changes. Consequently, differential should also be given for temperature drops in wall construc-
strains develop and tensile stresses are induced. When these tion. Contraction joints (or at least grooves of a depth not less
that 10% of the wall thickness on both sides) in which the
differential stresses exceed the capability of the concrete to
cracks can occur and be hidden should be provided at inter-
withstand them, cracking occurs. New concrete should be
vals ranging from one (for high walls) to three (for low
protected for as long as possible from the loss of moisture or
walls) times the height of the wall.
drop in temperature. These considerations can result in
The interior and exterior concrete, particularly in larger,
stresses capable of causing cracks at an early age but that
thicker sections, can be subjected to changes in temperature
might be sustained at greater maturity. Concrete should have or moisture content at different rates and different degrees.
a high tensile-strain capacity, which is influenced by the ag- When this happens, the interior concrete restrains the exteri-
gregate. Therefore, a low modulus of elasticity in tension is or concrete from shrinking, and tensile stresses develop that
desirable. can cause the surface to crack. This occurs when the surface
cools while the interior is still warm from the heat of hydra-
8.2—Restraint tion or when the surface concrete dries faster than the interior
Restraint exists in many circumstances under which the concrete. It is often feasible to protect the surface at early
structure and its concrete elements should perform. Typical ages so that stress-inducing differential strains cannot develop
examples illustrate how restraint can cause cracking if the before the concrete is strong enough to withstand the strain
224R-36 ACI COMMITTEE REPORT

without cracking. Reinforcement provided for both thermal surface dryness; mixture proportions, especially mixing wa-
contraction and drying shrinkage can only partially restrain ter; the character and degree of restraint; and the extensibili-
the contraction of surface concrete, but more and narrower ty (tensile-strain capacity) of the concrete. The extensibility
cracks can result. represents how much the concrete can be strained without
Restraint can occur at changes in sections because of dif- exceeding its tensile strength and is the sum of creep plus
ferences in temperature or drying shrinkage in the two sec- elastic-strain capacity. The latter is largely related to the
tions. If feasible, a contraction joint can be used to relieve the composition of the aggregate and can vary widely. Typical-
restraint. Restraint of flatwork results from anchoring slab ly, some concretes of highly quartzitic gravels have a low
reinforcements in perimeter concrete footings. When a slab strain capacity and a high modulus of elasticity. Concretes
is free to shrink from all sides toward its center, there is min- with a low strain capacity are much more sensitive to shrink-
imal cracking. Contraction joints and perimeter supports age due to drying (and to a decrease in temperature) and will
should be designed accordingly. be subjected to a greater amount of cracking.
Walls, slabs, and tunnel linings placed against the irregular A prime objective of crack-control procedures is to keep
surface of a rock excavation are restrained from moving the concrete wet as long as possible so that the concrete will
when the surface expands or contracts in response to changes have time to develop more strength to resist cracking forces.
in temperature or moisture content. Closely spaced contraction As described in Chapter 3 and Carlson (1938), there are
joints or deep grooves should be provided to hide the cracks some cases where prolonged moist curing is not beneficial.
that would disfigure such surfaces. In tunnel linings, the con- The importance of moist curing will vary with the weather
traction in the first few weeks is primarily thermal, and the and the time of year. Cold concrete (below 10 C [50 F]) dries
use of cold concrete 10 C (50 F) can reduce cracking materi- very slowly, provided the relative humidity is above 40%. At
ally. By the time drying is significant, the concrete lining has some depth, concrete loses moisture slowly. Where surface
a greater tensile strain capacity and therefore is able to resist drying can be rapid, more care should be devoted to uninter-
shrinkage cracking. Circumferential cracks in tunnel linings rupted curing to improve surface strength. Cracking will be
and other cast-in-place concrete conduits and pipe lines, further reduced by creep if the surface is prevented from drying
however, can be greatly reduced in number and width if a quickly at the end of the curing period. To accomplish slow
bulkhead is used to prevent air movement through the tunnel, drying, wet curing should remain for several days without
and shallow ponds of water are placed in the invert as soon wetting after the specified curing period (preferably 7 to
as possible after placing the lining. They should remain there 10 days) until the cover and the concrete under it appear
until the tunnel is put into service (USBR 1981). The con- to be dry. If these procedures are required to be used, they
crete can become much stronger in the humid environment should be included in project contract documents.
and will be better able to resist shrinkage-induced tensile
8.3.3 Plastic shrinkage—Objectionable plastic shrinkage
stresses if allowed to dry in service. If the tunnel carries wa-
cracks commonly occur in the surfaces of floors and slabs
ter, there will be no further drying shrinkage.
when job conditions are so dry that moisture is removed
The previous examples indicate that many crack-control
from the concrete surface faster than it is replaced by bleed-
procedures should be considered during design. While prop-
water. These cracks occur before the start of curing and can
er construction performance can contribute a great deal (as
occur either before or after final finishing. As the moisture is
will be discussed), the contractor cannot be expected to use
removed, the surface concrete contracts, resulting in tensile
the best procedures unless these procedures are included in
stresses in the weak, stiffening, plastic concrete that cause
the contract documents.
short random cracks in the surface. These cracks are usually
wide at the surface but only a few millimeters (inches) in
8.3—Shrinkage
depth, although full-depth cracks can occur in pavements.
8.3.1 Effect of water content—As shown in Fig 3.5, concrete
drying shrinkage is proportional to mixture water content. The The cracks range from a few millimeters (inches) to a few
use of the lowest practical slump adds in the control of crack- meters (feet) in length and are a few millimeters (inches) to
ing. Of major importance is the selection of mixture propor- 2/3 m (2 ft) apart.
tions that require the least amount of water for the desired Sometimes plastic shrinkage cracks appear early enough
concrete strength and durability. This means avoiding over- to be removed in later floating or troweling operations.
sanded mixtures, using the largest maximum aggregate size When this occurs, it is advisable to postpone these operations
practical, and using aggregate with the most favorable shape as long as possible to get their maximum benefit without the
and grading conducive to workability. Also use a well-grad- recurrence of cracking.
ed sand with a minimum of fines passing the 100 mesh and In other cases, early floating can destroy the growing ten-
free of clay, such that its sand equivalent value is not less sion by reworking the surface mortar and prevent plastic
than 80% (AASHTO T 176). Shrinkage tends to be propor- cracking. At the first appearance of cracking while the
tional to the total paste volume in the mixture. concrete is still responsive, a vigorous effort should be
8.3.2 Surface drying—Surface drying will occur except made to close the cracks by tamping or beating with a float.
when the surface is submerged or below grade. Drying If firmly closed, they are unlikely to reappear. They can,
shrinkage strains of up to 600 × 10–6 or more are likely. The however, reappear if they are merely troweled over. In any
amount of shrinkage cracking depends on factors such as event, curing should be started as soon as possible.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-37

Plastic shrinkage cracking is most likely when environ- differences between the interior and the exterior concrete.
mental conditions, concrete temperature, and mixture propor- Primarily, these controls lower the interior temperature rise
tions combine to cause a rapid loss of available surface caused by the heat of hydration by using minimum cement
moisture. These conditions can develop in either hot or cold content, pozzolans to replace a portion of the cement, water-
weather when low humidity and high wind speed combine reducing admixtures, air-entrainment, large aggregate sizes,
with a warm concrete surface temperature. Project specifica- low slump, and chipped ice for mixing water to maintain the
tions should stipulate that effective moisture-control precau- temperature of the fresh concrete as close to 10 C (50 F) as
tions should be taken to prevent a serious loss of surface possible. At no time should forms be removed to expose
moisture under such conditions. Principal among these pre- warm surfaces to low temperatures. As mentioned in Section
cautions are the use of fog (not spray) nozzles to maintain a 8.3.2, the extensibility or strain the concrete will withstand
sheen of moisture on the surface between the finishing oper- before tensile failure is a function of the aggregate and
ations. Plastic sheeting can be rolled on and off before and should be evaluated, especially on larger projects. What ap-
after floating, preferably exposing only the area being plies to one project will not necessarily apply to another.
worked on at that time. Less effective but helpful precautions
include sprayed monomolecular films that inhibit evapora- 8.4—Settlement
tion. Windbreaks are helpful; therefore, it is desirable to Settlement or subsidence cracks develop while concrete is
schedule flatwork after the walls are erected (ACI 305R; in the plastic stage after the initial consolidation. Settlement
ACI 302.1R). cracks are the natural result of heavy solids settling in a liq-
Plastic shrinkage cracking is associated with the rate loss uid medium. Settlement cracks occur along rigidly supported
of surface moisture relative to the replenishment rate of elements, such as horizontal reinforcement, form ties, or
moisture by bleeding. As a result, the tolerable rate of mois- embedments. Sometimes concrete will adhere to the forms.
ture loss is significantly affected by the rate of bleeding. Fur- A crack will appear at these locations if the forms are hot at
ther, the rate of moisture loss from the surface of concrete is the top or are partially absorbent. Cracks often appear in hor-
driven by a combination of factors that include air tempera- izontal construction joints and in bridge deck slabs over re-
ture, concrete temperature, relative humidity of the air above inforcing or form ties with only a few inches of cover (about
the concrete, degree of saturation of the surface of the con- 75 to 125 mm [3 to 5 in.]). Settlement cracks in bridge decks
crete (amount of bleedwater), and the wind velocity at the can be reduced by increasing the concrete cover along with
concrete surface. mixture proportioning that minimizes bleeding and settle-
The rate of evaporation can be greater during cold-weather- ment. Properly executed late revibration can be used to close
concreting conditions than during hot weather, particularly settlement cracks and improve the quality and appearance of
when the concrete has been heated, the air is cold and dry, the concrete in the upper portion of such placements, even
and the wind speed is high. The reason for this is that the though settlement has taken place and slump has been lost
thermal driving force for surface evaporation is the differ- (ACI 309R). Use of a low-slump concrete is also recom-
ence between air and concrete temperature. When the con- mended to help prevent settlement cracks in bridge decks
crete is warmer than the air, evaporation is favored; this is and slabs.
dependent on humidity and wind speed as well.
Other helpful practices that can counteract the excessive 8.5—Construction
loss of surface moisture are using a well-dampened sub- A great deal can be done during construction to minimize
grade, cooling the aggregates by dampening and shading cracking. The individual project specifications should be
them, and using cold mixing water or chipped ice as mixing specific with regard to actions that should be taken to cope
water to lower the fresh concrete temperature. with extremes of nature and to make enforceable requirements.
8.3.4 Surface cooling—Surface cooling will typically 8.5.1 Concrete aggregates—The aggregates should be se-
shrink the surface of unrestrained concrete about 10 × 10–6 C lected to make concrete of high-strain capacity, if reasonably
(5.5 × 10–6 F) as temperature decreases. This amounts to 9 mm available. It is important that fine and coarse aggregates be
in a 30 m length with a drop of 30 C (1/3 in. in 100 ft length clean and free of unnecessary fine material, particularly
with a drop of 50 F). The amount of shrinkage deformation clays. The fine aggregate should have a sand equivalent val-
is reduced by restraint and creep but tensile stresses result. ue greater than 80% and should be verified frequently
The earlier the age and the slower the rate at which cooling (AASHTO T 176). The sand should have sufficient time in
or drying occur, the lower the tensile stresses will be. The in- storage for the moisture content to stabilize at a level of less
fluence of creep decreases the effective modulus of concrete than 7% on an oven-dry basis.
loaded at early ages and allows more extensibility. 8.5.2 Shrinkage-compensating cement—Shrinkage-compen-
In ordinary concrete work, winter protection required for sating cement can be used to compensate for shrinkage in
the development of adequate strength will prevent the most restrained elements or elements with the minimum shrink-
critical effects of cooling. Contraction joints for control of age reinforcement required by ACI 318. The principal prop-
shrinkage cracking will also control cracking due to drops in erty of these cements is that the expansion induced in the
surface temperature. In addition to Chapter 7 of this report, concrete while setting and hardening is designed to offset the
Chapters 4 and 5 of ACI 207.1R discuss temperature con- normal drying shrinkage. With correct usage (particularly
trols for mass concrete to minimize the early temperature with early and ample water curing required for maximum
224R-38 ACI COMMITTEE REPORT

expansion), the distance between joints can sometimes be • Reinforcement and embedments should be firmly posi-
tripled without increasing the level of shrinkage cracking. De- tioned with the designated thickness of cover to prevent
tails on the types and the correct usage of shrinkage-compen- corrosion, expansion, and cracking;
sating cements are given in Section 3.6 and ACI 223. • Concrete should not be placed against hot reinforce-
8.5.3 Handling and batching—Aggregates should be han- ment or forms;
dled so as to avoid contamination, segregation, and break- • Formwork support should be both strong and stiff
age. Handling and batching is best done by finish screening enough to be free of early failures and distortion caus-
and rinsing coarse aggregate into their various sizes and ing cracking;
placing them in the appropriate bins at the batch plant. Every • Subgrade and other supports should not settle unevenly,
effort should be made to uniformly batch and mix the con- which may lead to cracking due to overstress in the
crete so that there will be a minimum of variation in slump structure;
and workability, which invariably lead to demands for a • Calcium chloride should not be used if steel reinforcement
greater margin of workability. is present. If acceleration in setting or strength gain is
8.5.4 Cold concrete—In mass concrete structures, reduc- needed, additional cement, hot water, or a nonchloride
ing water and cement contents to a practical minimum and accelerator should be used;
using cold concrete will reduce temperature differentials that • Special care is needed in handling precast units to pre-
cause cracking. Less mixing water reduces drying shrinkage. vent overstress due to handling. Pickup points and rig-
In warm weather, cold concrete reduces slump loss, increases ging should be considered;
pumpability, and improves the response to vibration. Chipped • Avoid the use of unvented salamanders in cold weather
ice can be substituted for all or a part of the batched mixing (ACI 306R) or gasoline-operated equipment where
water. In cold weather, concrete is naturally cold, and every ventilation is not adequate. These will increase the dan-
effort should be made to use it as cold as possible without in- ger of carbonation, causing shrinkage and surface
viting damage from freezing. It is pointless to try protecting cracking;
surfaces, edges, and corners by placing needlessly warm • Contraction joints should not be omitted and grooves
concrete in cold weather. These vulnerable parts require pro- should be of sufficient depth and well within the maxi-
tection by insulation or protective enclosures (ACI 306R). mum permitted spacing. In hot weather or arid environ-
8.5.5 Revibration—When done as late as the formed con- ments, contraction joints should be installed in the fresh
crete will respond to the vibrator, revibration can eliminate concrete as inserts or saw-cut when the concrete is hard
cracks and checks where something rigidly fixed in the enough not to be torn or damaged by the blade;
placement prevents a part of the concrete from settling with • Reactive elements of coarse aggregate should be neu-
the rest of it. Settlement cracks are most apparent in the up- tralized through the use of low-alkali cement, suitable
per part of wall and column placements where revibration pozzolans, or both. Certain cherts and other expansive
can be readily used. Deep revibration corrects cracks caused aggregates can cause cracks and popouts. Project speci-
by differential settlement around blockout and window fications should cover these aggregate properties and
forms and where slabs and walls are placed monolithically the cleanliness of these aggregates; and
(ACI 309R). • Correct amounts of entrained air should be used and
8.5.6 Finishing—Proper flatwork finishing can make a monitored to prevent cracking due to freezing and
difference in many types of cracking (ACI 302.1R). Low- thawing and exposure to deicing salts.
slump concrete should be used. More than a 75 mm (3 in.)
slump is rarely necessary, except in hot weather, where both 8.6—Specifications to minimize drying shrinkage
slump and moisture are lost quite rapidly. Finishing should Actions during construction to obtain the lowest possible
not be done in the presence of surface water. Precautions change in volume of the concrete should be supported by the
(Section 8.3.3) should be taken to prevent plastic shrinkage. specifications. Unless bids are taken on this basis, the con-
Any required marking and grooving should be carefully cut tractors cannot be expected to provide other than ordinary
to the specified depth. Curing should be promptly conducted. materials, mixtures, and procedures. The following items
8.5.7 Curing and protection—Concrete should be brought should be carefully spelled out in the project specifications.
to a level of adequate strength and protected from low tem- 8.6.1 Concrete materials—Cement should be ASTM C
peratures and drying conditions that would otherwise cause 150 Types I, II, V, ASTM C 595 Type IS, or ASTM C 845
cracking. The curing and protection should not be discontin- for expansive hydraulic cement. Use special care if Type III
ued abruptly. If the new concrete is given a few days to grad- cement is used because of its high rate of heat generation and
ually dry or cool, creep can reduce the possibility of cracking strength (and stiffness) development. Recommendation of ACI
when the curing and protection are fully discontinued. 224 should be followed when using shrinkage-compensating
Subsequent application of a curing compound after initial cement. ASTM C 595 Type IP can also be used. Aggre-
curing will slow drying out. gates favorable to low mixing-water content are well graded,
8.5.8 Miscellaneous—Some items normally covered in well shaped (not elongated, flat, or splintery), with minimum
project specifications (or which should be covered where amounts of clay, dirt, and excessive fines. Aggregate should
appropriate) require special attention during construction consist of rock types that will produce low-shrinkage con-
because of their potential effects on cracking. crete. Utilization of pozzolans and chemical admixtures
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-39

should be considered as appropriate. Calcium chloride and loosened bolts long enough to allow the concrete surfaces
other admixtures containing a significant amount of chloride to dry gradually.
should be prohibited. If ponding is used for curing in an arid climate, precau-
8.6.2 Concrete mixtures—For the least amount of shrink- tions such as a properly applied curing compound should
age, the mixture proportions should incorporate those factors be used when ponding is discontinued to avoid quick drying.
that contribute to the lowest paste content. Use the largest Because drying is slow and prolonged, a properly applied
practical maximum size of aggregate (MSA). The lowest curing compound provides good curing for flatwork
practical sand content, slump (water content), and mixture placed on a well-wetted subgrade, and it provides ade-
temperature should also be used. Where possible, limit the quate curing for formed surfaces. In an arid climate, cur-
smaller-size fractions, that is, 4.75 mm to 9.5 or 20 mm (No. ing compounds are not adequate for thinner structural
4 to 3/8 or 3/4 in.), of coarse aggregate to half of the total, sections. When used on formed surfaces, they should be applied
especially if the aggregate is crushed. when the surface is still damp but no longer wet (ACI 308).
8.6.3 Concrete handling and placing—Equipment
(chutes, belts, conveyors, pumps, hoppers, and bucket open- 8.7—Conclusion
ings) should be capable of working effectively with lower It is the responsibility of the engineer to develop effective
slump, larger MSA concrete wherever it is appropriate and designs and clear, specific provisions in project specifica-
feasible to use it. The actions that tend to make a mixture tions. To ensure both the owner’s and the engineer’s satisfac-
more pumpable also tend to make the resulting concrete tion with the results, the engineer should have the owner
more prone to shrinkage and cracking. The use of pumping arrange for inspection by either the owner’s personnel, the
equipment capable of handling mixtures favorable to mini- engineer, or a reliable professional inspection service that
mize cracking should be required. will ensure that the construction is performed on the same
Vibrators should be the largest and most powerful that can basis as it was bid. Without the full and firm intent to confirm
be operated in the placement. Upper lifts of formed concrete the specified character and degree of performance, there is a
could be revibrated as late as the running vibrator will pene- likelihood that undesirable results will be obtained. Without
trate under its own weight. firm inspection, a quality-control-assurance program, and a
8.6.4 Finishing—Finishing should follow the recommen- clear understanding of the project requirements by the
dations of ACI 302.1R to minimize or avoid surface crack- contractor, it is likely that concrete will contain more water
ing. It is particularly important that flatwork joint grooves than it should. With less water content in the concrete, the
have a depth of at least one-quarter of slab thickness. finishing operations can be expedited and the curing process
8.6.5 Forms—Forms should have ample strength to sus- can start earlier. If properly applied, the procedures discussed in
tain strong vibration of low-slump concretes. Exposing this chapter can have an effective influence on producing high-
warm concrete surfaces to fast drying conditions or to low quality concrete with minimal cracking.
temperatures before curing should be avoided during form
removal if drying and thermal shrinkage cracking is to be CHAPTER 9—REFERENCES
prevented (ACI 347R). 9.1—Referenced standards and reports
8.6.6 Contraction joints—Project drawings should include The standards and reports listed as follows were the latest
an adequate system of contraction joints to provide for editions at the time this document was prepared. Because
shrinkage. Formed grooves should be constructed in both these documents are revised frequently, the reader is advised
sides of parapet, retaining, and other walls at the appropriate to contact the proper sponsoring group if it is desired to refer
depth and spacing. ACI 224.3R provides useful information to the latest version.
on contraction joints.
8.6.7 Curing and protection—These procedures should American Association of State Highway and Transportation
ensure the presence of adequate moisture to sustain hydra- Officials (AASHTO)
tion in the surface concrete during the early ages of strength T 176 Plastic Fines In Graded Aggregate and Soils by
development. Rapid drying of the surfaces at the conclusion Use of the Sand Equivalent Test
of the specified curing period should be avoided. Providing
time for adjustment and relaxation of restraint-induced American Concrete Institute (ACI)
stresses minimizes cracking. 116R Cement and Concrete Terminology
The best curing environment is to keep the concrete con- 201.1 Guide for Making a Condition Survey of Concrete
tinuously wet during the curing period. Water curing should in Service
use a wet cover in contact with the concrete surfaces. At the 201.2R Guide to Durable Concrete
end of the wet curing period, at least 7 days, the cover should 207.1R Mass Concrete
be left in place until it and the concrete surface appear to be 207.2R Effect of Restraint, Volume Change, and Rein-
dry, especially in arid weather. forcement on Cracking of Mass Concrete
In less-arid areas and for interiors, the forms will pro- 207.4R Cooling and Insulating Systems for Mass Concrete
vide adequate curing if exposed surfaces are protected 207.5R Roller Compacted Mass Concrete
from drying and left in contact with the concrete for at 209R Prediction of Creep, Shrinkage, and Temperature
least 7 days. Thereafter, the forms should be left on with Effects in Concrete Structures
224R-40 ACI COMMITTEE REPORT

212.3R Chemical Admixtures for Concrete American Association of State Highway and Transportation
221R Guide for Use of Normal Weight and Heavyweight Officials
Aggregates in Concrete 444 N. Capitol Street NW
221.1R State-of-the-Art Report on Alkali-Aggregate Suite 225
Reactivity Washington, DC 20001
222R Corrosion of Metals in Concrete
223 Standard Practice for the Use of Shrinkage- American Concrete Institute
Compensating Concrete PO Box 9094
224.3R Joints in Concrete Construction Farmington Hills, MI 48333-9094
302.1R Guide for Concrete Floor and Slab Construction
ASTM
304R Guide for Measuring, Mixing, Transporting, and
100 Barr Harbor Drive
Placing Concrete
West Conshohocken, PA 19428
305R Hot Weather Concreting
306R Cold Weather Concreting CEB-FIP
308 Standard Practice for Curing Concrete c/o British Cement Association
309R Guide for Consolidation of Concrete Century House
313 Standard Practice for Design and Construction of Telford Avenue
Concrete Silos and Stacking Tubes for Storing Crowthorne, Berkshire RG45 6YS
Granular Materials United Kingdom
318/318R Building Code Requirements for Structural
Concrete and Commentary 9.2—Cited references
340R ACI Design Handbook (SP-17(97)) Aalami, B. O., and Barth, F. G., 1989, “Restraint Cracks
347R Guide to Formwork for Concrete and their Mitigation in Unbonded Post-Tensioned Building
350 Code Requirements For Environmental Structures,” Cracking in Prestressed Concrete Structures,
Engineering Concrete Structures SP-113, G. T. Halvorsen and N. H. Burns, eds., American Con-
crete Institute, Farmington Hills, Mich., pp. 157-202.
446.1R Fracture Mechanics of Concrete: Concepts,
Abeles, P. W., 1956, “Cracks in Prestressed Concrete
Models, and Determination of Material Properties
Beams,” Proceedings, 5th IABSE Congress, International
504R Guide to Sealing Joints for Concrete Structures
Association for Bridge and Structural Engineering, Zurich,
544.2R Measurement of Properties of Fiber-Reinforced pp. 707-720.
Concrete Abeles, P. W.; Brown, E. L., II; and Morrow, J. W., 1968,
544.3R Guide for Specifying, Proportioning, Mixing, “Development and Distribution of Cracks in Rectangular
Placing, and Finishing Steel Fiber Reinforced Prestressed Beams during Static and Fatigue Loading,”
Concrete Journal, Prestressed Concrete Institute, V. 13, No. 5, Oct.,
544.4R Design Considerations for Steel Fiber Reinforced pp. 36-51.
Concrete ACI Committee 223, 1970, “Expansive Cement Con-
548R Polymers in Concrete cretes: Present State of Knowledge,” ACI JOURNAL, Pro-
548.1R Guide for the Use of Polymers in Concrete ceedings V. 67, No. 8, Aug., pp. 583-610.
SP-38 Klein Symposium on Expansive Cement Con- ACI Committee 224, 1971, “Causes, Mechanism, and
cretes Control of Cracking in Concrete,” ACI Bibliography No. 9,
SP-40 Polymers in Concrete American Concrete Institute, Farmington Hills, Mich., 92 pp.
SP-58 Polymers in Concrete ACI Committee 224, 1972, “Control of Cracking in Con-
SP-64 Cedric Wilson Symposium on Expansive Cement crete Structures,” ACI JOURNAL, Proceedings V. 69, No. 12,
Dec., pp. 717-753.
American Society for Testing and Materials (ASTM) ACI Committee 224, 1980, “Control of Cracking in Con-
crete Structures,” Concrete International: Design and Con-
C 150 Specification for Portland Cement
struction, V. 2, No. 10, Oct., pp. 35-76.
C 845 Specification for Expansive Hydraulic Cement
ACI Committee 318, 1999, “Discussion of a Report Pub-
E 399 Test Method for Plain-Strain Fracture Toughness
lished in the November 1998 Issue of Concrete Internation-
of Metallic Materials
al, pp. S1-S90,” Concrete International, V. 21, No. 5., May,
pp. 318.1-318.49.
European Committee for Concrete-International Federation Atimtay, E., and Ferguson, P. M., 1973, “Early Chloride
of Prestressed Concrete (CEB-FIP) Corrosion of Reinforced Concrete—A Test Report,” ACI
CEB-FIP European Model Code for Concrete Structures JOURNAL, Proceedings V. 70, No. 9, Sept., pp. 606-611.
Attiogbe, E. K., and Darwin, D., 1987, “Submicrocracking
The above publications may be obtained from the following in Cement Paste and Mortar,” ACI Materials Journal, V. 84,
organizations: No. 6, Nov.-Dec., pp. 491-500.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-41

Attiogbe, E. K., and Darwin, D., 1988, “Strain Due to Sub- Bishara, A. G., and Tantayanondkul, P., 1974, “Use of La-
microcracking in Cement Paste and Mortar,” ACI Materials tex in Concrete Bridge Decks,” Report No. EES 435,
Journal, V. 85, No. 1, Jan.-Feb., pp. 3-11. (ODOT-12-74) Ohio Department of Transportation, Ohio
Bascoul, A.; Kharchi, F.; and Maso, J. C., 1987, “Concern- State University, Columbus, Ohio.
ing the Measurement of the Fracture Energy of a Micro-Con- Brendel, G., and Ruhle, H., 1964, “Tests on Reinforced Con-
crete According to Crack Growth in Three Point Bending crete Beams Under Long-Term Loads (Dauerstandvrsuche
Tests of Notched Beams,” Proceedings, SEM-RILEM Inter- mit Stahlbetonbalken),” Proceedings, Seventh IABSE Con-
national Conference on Fracture of Concrete and Rock, S. P. gress (Rio de Janeiro, 1964), International Association of
Shah and S. E. Swartz, eds., Houston, Tex. Bridge and Structural Engineering, Zurich, Switzerland,
Bate, S. C. C., 1958, “Relative Merits of Plain and De- pp. 916-922.
formed Wires in Prestressed Concrete Beams under Static Broms, B. B., 1965a, “Technique for Investigation of In-
and Repeated Loading,” Proceedings, Institution of Civil ternal Cracks in Reinforced Concrete Members,” ACI JOUR-
Engineers (London), V. 10, Aug., pp. 473-502. NAL, Proceedings V. 62, No. 1, Jan., pp. 35-44.
Bate, S. C. C., 1963, “A Comparison Between Prestressed Broms, B. B., 1965b, “Crack Width and Crack Spacing in
Concrete and Reinforced Concrete Beams under Repeated Reinforced Concrete Members,” ACI JOURNAL, Proceed-
Loading,” Proceedings, Institution of Civil Engineers ings V. 62, No. 10, Oct., pp. 1237-1256.
(London), V. 24, Mar., pp. 331-358. Brookhaven National Laboratory, 1968, “Concrete Poly-
Bažant, Z. P., 1979, “Material Behavior under Various mer Materials,” BNL Report 50134 (T-509).
Types of Loading,” High Strength Concrete, S. P. Shah, ed., Brooks, J. J., and Neville, A. M., 1977, “A Comparison of
University of Illinois at Chicago, pp. 79-92. Creep, Elasticity and Strength of Concrete in Tension and in
Bažant, Z. P.; Kim, J. K; and Pfeiffer, P., 1986, “Determi- Compression,” Magazine of Concrete Research, London,
nation of Non-linear Fracture Parameters from Size Effect V. 29, No. 100, Sept., pp. 131-141.
Tests,” Journal of Structural Engineering, ASCE, V. 112, Bryant, A. H., and Vadhanavikkit, C., 1987, “Creep
No. 2, Feb., pp. 289-307. Shrinkage—Size, and Age at Loading Effects,” ACI Materi-
Bažant, Z. P., and Oh, B. H., 1983, “Crack Band Theory als Journal, V. 84, No. 2, Mar.-Apr., pp. 119-123.
for Fracture of Concrete,” Materials and Structures, Burgess, J. A.; Breen, J. E.; and Poston, R. W., 1989,
RILEM, V. 16, No. 93, pp. 155-177. “Anchorage Zone Cracking of Post-Tensioned Bridge
Decks with Closely Spaced Anchors,” Cracking in Prestressed
Bažant, Z. P., and Sener, S., 1988, “Size Effect in Pullout
Concrete Structures, SP-113, G. T. Halvorsen and N. H. Burns,
Tests,” ACI Materials Journal, V. 85, No. 5, May, pp. 347-351.
eds., American Concrete Institute, Farmington Hills, Mich.,
Bažant, Z. P.; Sener, S.; and Pratt, P. C., 1988, “Size Effect
pp. 79-108.
Tests of Torsional Failure of Plain and Reinforced Concrete
Buyukozturk, O., 1970, “Stress-Strain Response and Frac-
Beams,” Materials and Structures, V. 21, pp. 425-430.
ture of a Model of Concrete in Biaxial Loading,” PhD thesis,
Bažant, Z. P., and Sun, H. H., 1987, “Size Effect in Diag-
Cornell University, Ithaca, N.Y., June.
onal Shear Failure: Influence of Aggregate Size and Stirrups,”
California Department of Transportation, 1963, Report:
ACI Materials Journal, V. 84, No. 4, Apr., pp. 259-272.
Effect of Curing on Shrinkage of Concrete.
Beckett, D., and Alexandrou, A., 1997, Introduction to Euro- Carino, N. J., 1977, “Effects of Polymer Impregnation on
code 2— Design of Concrete Structures, E&FN Spon, 191 pp. Mortar-Aggregate Bond Strength,” Cement and Concrete
Beeby, A. W., 1978a, “Concrete in the Oceans—Cracking Research, V. 7, No. 4, July, pp. 439-447.
and Corrosion,” Technical Report No. 1, Cement and Concrete Carlson, R. W., 1938, “Drying Shrinkage of Concrete as
Association, London. Affected by Many Factors,” Proceedings, ASTM, V. 38,
Beeby, A. W., 1978b, “Corrosion of Reinforcing Steel in Part II, pp. 419-437.
Concrete and Its Relation to Cracking,” The Structural Engi- Carrasquillo, R. L.; Slate, F. O.; and Nilson, A. H., 1981,
neer, London, V. 56A, No. 3, Mar., pp. 77-81. “Microcracking and Behavior of High Strength Concrete
Beeby, A. W., 1983, “Cracking Cover at Corrosion of Subject to Short-Term Loading,” ACI JOURNAL, Proceedings
Reinforcement,” Concrete International: Design and V. 78, No. 3, May-June, pp. 179-186.
Construction, V. 5, No. 2, Feb., pp. 35-40. Castro-Montero, S.; Shah S. P., and Miller, R. A., 1990,
Bennett, E. W., and Chandrasekhar, C. S., 1971, “Calcula- “Study of Fracture Process in Mortar with Laser Holographic
tion of the Width of Cracks in Class 3 Prestressed Beams,” Measurements,” Proceedings of the NATO Advanced Research
Proceedings, Institution of Civil Engineers, London, V. 49, Workshop on Toughening Mechanisms in Quasi-Brittle
July, pp. 333-346. Materials, Evanston, Ill.
Bennett, E. W., and Dave, N. J., 1969, “Test Performances Catalano, D., and Ingraffea, A. R., 1982, “Concrete Frac-
and Design of Concrete Beams with Limited Prestress,” The ture: A Linear Elastic Fracture Approach,” Department of
Structural Engineer, London, V. 47 No. 12, Dec., pp. 487-496. Structural Engineering Report, School of Civil and Environ-
Bishara, A. G., 1979, “Latex Modified Concrete Bridge mental Engineering, Cornell University, Ithaca, N.Y., 161 pp.
Deck Overlays—Field Performance Analysis,” Report Clear, K. C., 1974, “Time to Corrosion of Reinforcing
No. FHWA/OH/79/004, Federal Highway Administration, Steel in Concrete Slabs,” Transportation Research Record
Washington D.C., Oct., 97 pp. No. 500, Transportation Research Board, pp. 16-24.
224R-42 ACI COMMITTEE REPORT

Cook, D. J., and Chindaprasirt, P., 1980, “Influence of Gopalaratnam, V. S., and Shah, S. P., 1986, “Softening
Loading History Upon the Compressive Properties of Con- Response of Plain Concrete in Direct Tension,” ACI JOUR-
crete,” Magazine of Concrete Research, London, V. 32, NAL, Proceedings V. 82, No. 3, May-June, pp. 310-323.
No. 111, June, pp. 89-100. Griffith, A. A., 1920, “The Phenomena of Rupture and
Darwin, D., 1978, Discussion of “Bond Failure and Inelas- Flow of Solids,” Transactions, Royal Society of London,
tic Response of Concrete,” by Testa and Stubbs, Journal of No. 221A, pp. 163-198.
the Engineering Mechanics Division, ASCE, V. 104, No. EM2, Halvorsen, G. T., 1987, “Code Requirements for Crack
Apr., pp. 507-509. Control,” Concrete and Concrete Construction, SP-104,
Darwin, D., and Slate, F. O., 1970, “Effect of Paste- American Concrete Institute, Farmington Hills, Mich.,
Aggregate Bond Strength on Behavior of Concrete,” pp. 275-322.
Journal of Materials, V. 5, No. 1, Mar., pp. 86-98. Hansen, W., and Almudaiheem, J. A., 1987, “Ultimate
Davis, H. E., 1940, “Autogenous Volume Changes of Drying Shrinkage of Concrete—Influence of Major Parame-
Concrete,” Proceedings, ASTM, V. 40, pp. 1103-1110. ters,” ACI Materials Journal, V. 84, No. 3, May-June,
Derucher, K. N., 1978, “Application of the Scanning Elec- pp. 39-46.
tron Microscope to Fracture Studies of Concrete,” Building Harajli, M. H., and Naaman, A. E., 1989, “Cracking in
and Environment, V. 13, No. 2, 1978, pp. 135-141. Partially Prestressed Beams Under Static and Fatigue
Dolan, C. W., 1990, “Developments in Non-Metallic Loading,” Cracking in Prestressed Concrete Structures,
Prestressing Tendons,” PCI Journal, V. 35, No. 5, Sept.-Oct., SP-113, G. T. Halvorsen and N. H. Burns, eds., American
pp. 80-88. Concrete Institute, Farmington Hills, Mich., pp. 29-56.
Euro EC2 Code, 1997, Design of Concrete Structures, Higgins, D. D., and Bailey, J. E., 1976, “Fracture Mea-
Paris. surements on Cement Paste,” Journal of Materials Science,
Feldman, R. F., and Swenson, E. G., 1975, “Volume V. 11, pp. 1995-2003.
Change on First Drying of Hydrate Portland Cement with Hillemeier, B., and Hilsdorf, H. K., 1977, “Fracture
and without Admixture,” Cement and Concrete Research, V. 5, Mechanics Studies of Concrete Compounds,” Cement and
No. 1, pp. 25-35. Concrete Research, V. 7, No. 5, Sept., pp. 523-536.
Frantz, G. C., and Breen, J. E., 1980a, “Cracking on Side Hillerborg, A.; Modeer, M; and Petersson, P. E., 1976,
Faces of Large Reinforced Concrete Beams,” ACI JOURNAL, “Analysis of Crack Formation and Crack Growth by Means
Proceedings V. 77, No. 5, Sept.-Oct., pp. 307-313. of Fracture Mechanics and Finite Elements,” Cement and
Frantz, G. C., and Breen, J. E., 1980b, “Design Proposal Concrete Research, V. 6, pp. 773-782.
for Side Face Crack Control Reinforcement for Large Rein- Hoff, G. C., 1987, “Durability of Fiber Reinforced Concrete
forced Concrete Beams,” Concrete International: Design in a Severe Marine Environment,” Concrete Durability,
and Construction, V. 2, No. 10, Oct., pp. 29-34. SP-100, American Concrete Institute, Farmington Hills,
Frosch, R. J., 1999, “Another Look at Cracking and Crack Mich., pp. 997-1041.
Control in Reinforced Concrete,” ACI Structural Journal, Holmberg, A., 1973, “Crack Width Prediction and Minimum
V. 96, No. 3, May-June, pp. 437-442. Reinforcement for Crack Control,” Dansk Selskab for
Gergely, P., 1969, “Anchorage Systems in Prestressed Con- Byaningsstatik, Copenhagen, V. 44, No. 2, June, pp. 41-50.
crete Pressure Vessels’ Anchorage Zone Problems,” Report No. Holmberg, A., and Lindgren, S., 1970, “Crack Spacing
ORNL-TM-2378, Oak Ridge National Laboratory, U.S. Atom- and Crack Width Due to Normal Force or Bending Moment,”
ic Energy Commission, Oak Ridge, Tenn., pp. 1-49. Document D2, National Swedish Council for Building
Gergely, P., and Lutz, L. A., 1968, “Maximum Crack Research, Stockholm, 57 pp.
Width in Reinforced Concrete Flexural Members,” Causes, Houk, I. E., Jr.; Borge, O. E.; and Houghton, D. L., 1969,
Mechanism, and Control of Cracking in Concrete, SP-20, “Studies of Autogenous Volume Change in Concrete for
American Concrete Institute, Farmington Hills, Mich., Dworshak Dam,” ACI JOURNAL, Proceedings V. 66, No. 7,
pp. 87-117. pp. 560-568.
Gergely, P., and Sozen, M. A., 1967, “Design of Anchor- Houk, I. E., Jr.; Paxton, J. A.; and Houghton, D. L., 1970,
age Zone Reinforcement in Prestressed Concrete Beams,” “Prediction of Thermal Stress and Strain Capacity of Con-
PCI Journal, V. 13, No. 2, Mar.-Apr., pp. 63-75. crete by Tests on Small Beams,” ACI JOURNAL, Proceedings
Gettu, R.; Bažant, Z. P.; and Karr, M. E.; 1990, “Brittleness V. 67, No. 3, Mar., pp. 253-261.
of High Strength Concrete,” Serviceability and Durability of Houghton, D. L., 1969, “Concrete Volume Change for
Construction Materials, B. A. Suprenant, ed., ASCE, New Dworshak Dam,” Proceedings, ASCE, V. 95, No. 2, Oct.,
York, V. 2, pp. 976-985. pp. 153-166.
Ghosh, R. S., and Malhotra, V., 1979, “Use of Superplas- Houghton, D. L., 1976, “Determining Tensile Strain
ticizers as Water Reducers,” Cement Concrete and Aggre- Capacity of Mass Concrete,” ACI JOURNAL, Proceedings
gates, ASTM, V. 1, No. 2, pp. 56-63. V. 73, No. 12, Dec., pp. 691-700.
Gjørv, O. E.; Sorenson, S. I.; and Arnesen, A., 1977, Hsu, T. T. C., 1963, “Mathematical Analysis of Shrinkage
“Notch Sensitivity and Fracture Toughness of Concrete,” Stresses in a Model of Hardened Concrete,” ACI JOURNAL,
Cement and Concrete Research, V. 7, pp. 333-344. Proceedings V. 60, No. 3, Mar., pp. 371-390.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-43

Hsu, T. T. C.; Slate, F. O.; Sturman, G. M; and Winter, G., Miscellaneous Paper No. SL-79-7, U.S. Army Engineer
1963, “Microcracking of Plain Concrete and the Shape of the Waterways Experiment Station, Vicksburg, Miss., Apr.
Stress Strain Curve,” ACI JOURNAL, Proceedings V. 60, No. 2, Liu, T. C., and McDonald, J. E., 1978, “Prediction of
Feb., pp 209-224. Tensile Strain Capacity of Mass Concrete,” ACI JOURNAL,
Hutton, S. G., and Loov, R. E., 1966, “Flexural Behavior Proceedings V. 75, No. 4, May, pp. 192-197.
of Prestressed, Partially Prestressed, and Reinforced Con- Lorman, W. R., 1940, “The Theory of Concrete Creep,”
crete Beams,” ACI JOURNAL, Proceedings V. 63, No. 12, Proceedings ASTM, V. 40, pp. 1082-1102.
Dec., pp. 1401-1410. Lutz, L. A.; Sharma, N. K.; and Gergely, P., 1968, “Increase
Illston, J. M., and Stevens, R. F., 1972, “Long-Term in Crack Width in Reinforced Concrete Beams under Sustained
Cracking in Reinforced Concrete Beams,” Proceedings, Loading,” ACI JOURNAL, Proceedings V. 64, No. 9, Sept.,
Institution of Civil Engineers, London, Part 2, V. 53, pp. 538-546.
Dec., pp. 445-459. Maher, A., and Darwin, D., 1976, “A Finite Element Model
Jenq, Y. S., and Shah, S. P., 1987, “Mixed Mode Fracture to Study the Microscopic Behavior of Plain Concrete,” CRINC
Parameters of Concrete,” Proceedings, SEM-RILEM Inter- Report SL-76-02, University of Kansas Center for Research,
national Conference on Fracture of Concrete and Rock, S. P. Inc., Lawrence, Kans., Nov., 83 pp.
Shah and S. E. Swartz, eds., Houston, Tex., pp. 359-369. Maher, A., and Darwin, D., 1977, “Microscopic Finite
John, R., and Shah, S. P., 1986, “Fracture of Concrete Element Model of Concrete,” Proceedings, First Interna-
Subject to Impact Loading,” Journal of Cement Concrete tional Conference on Mathematical Modeling (St. Louis,
and Aggregates, ASTM, V. 8, No. 1, pp. 24-32. Aug. 29-Sept. 1, 1977), University of Missouri-Rolla, V. III,
Kaplan, F. M., 1961, “Crack Propagation and the Fracture pp. 1705-1714.
of Concrete,” ACI JOURNAL, Proceedings V. 58, No. 5, Maher, A., and Darwin, D., 1982, “Mortar Constituent of
Nov., pp. 591-610. Concrete in Compression,” ACI JOURNAL, Proceedings V. 79,
Karsan, I. D., and Jirsa, J. O., 1969, “Behavior of Concrete No. 2, Mar.-Apr., pp. 100-109.
under Compressive Loadings,” Journal of the Structural Divi-
Maji, A. K., and Shah, S. P., 1987, “Process Zone and
sion, ASCE, V. 95, ST 12, Dec., pp. 2543-2563.
Acoustic Emission Measurements in Concrete,” Experimental
Kesler, C. E.; Naus, D. J.; and Lott, J. L.; 1972, “Fracture
Mechanics, SEM, Paper No. 3609.
Mechanics, Its Applicability to Concrete,” Proceedings, In-
Marshall, W. T., and Mattock, A. H., 1962, “Control of
ternational Conference on Mechanical Behavior of Materi-
Horizontal Cracking in the Ends of Pretensioned Concrete
als, Kyoto, Aug. 15-20, 1971, The Society of Materials
Girders,” Journal, Prestressed Concrete Institute, V. 7, No. 5,
Science, Japan, V. IV, pp. 113-124.
Aug.-Oct., pp. 56-74.
Krishna, R. N.; Basavarajuiah, B. S.; and Ahamed K. U. C.,
1973, “Flexural Behavior of Pretensioned Concrete Beams with Martin, J. L.; Darwin, D.; and Terry, R. E., 1991, “Cement
Limited Prestress,” Building Science, V. 8, No. 2, June, Paste, Mortar and Concrete in Compression,” SM Report
pp. 179-185. No. 31, University of Kansas Center for Research,
Lee, S. L.; Mansur, M. S.; Tan, K. H.; and Kasiraju, K., Lawrence, Kans., 161 pp.
1987, “Cracking Behavior of Concrete Tension Members Mather, B., 1957, “Factors Affecting Durability of Concrete
Reinforced with Welded Wire Fabric,” ACI Structural Jour- in Coastal Structures,” Technical Memorandum No. 96, Beach
nal, V. 84, No. 6, Nov.-Dec., pp. 481-491. Erosion Board, Washington D.C., June.
Leibengood, L. D.; Darwin, D.; and Dodds, R. H.; 1986, Mather, B., 1968, “Cracking Induced by Environmental
“Parameters Affecting FE Analysis of Concrete Structures,” Effects,” Causes, Mechanism, and Control of Cracking in
Journal of the Structural Division, ASCE, V. 112, No. 2, Concrete, SP-20, American Concrete Institute, Farmington
Feb., pp. 326-341. Hills, Mich., pp. 67-72.
Leonhardt, F., 1977, “Crack Control in Concrete Struc- McCoy, E. E., Jr.; Thorton, H. T.; and Allgood, J. K.,
tures,” IABSE Surveys No. S4/77, International Association 1964, “Concrete Laboratory Studies, Dworshak (Bruce’s
for Bridge and Structural Engineering, Zurich, 26 pp. Eddy) Dam, North Fork Clearwater River near Orofino, Ida-
Leonhardt, F., 1988, “Crack and Crack Control in Concrete ho: Creek Tests,” Miscellaneous Paper No. 6-613, Report 2,
Structures,” PCI Journal, V. 22, No. 4, pp. 124-145. U.S. Army Engineer Waterways Experiment Station, Vicks-
Linsbauer, H. N.; Ingraffea, A. R.; Rossmanith, H. P.; and burg, Miss., Dec.
Wawrzynek, P. A., 1989a, “Simulation of Cracking in Large McDonald, J. E.; Bombich, A. A.; and Sullivan, B. R.,
Arch Dam: Part I,” Journal of the Structural Division, 1972, “Ultimate Strain Capacity and Temperature Rise Studies,
ASCE, V. 115, No. 7, Feb., pp. 1599-1615. Trumbull Pond Dam,” Miscellaneous Paper C-72-20, U.S.
Linsbauer, H. N.; Ingraffea, A. R.; Rossmanith, H. P.; and Army Engineer Waterways Experiment Station, Vicksburg,
Wawrzynek, P. A., 1989b, “Simulation of Cracking in Large Miss., Aug.
Arch Dam: Part II,” Journal of the Structural Division, Meier, S. W., and Gergely, P., 1981, “Flexural Crack
ASCE, V. 115, No. 7, Feb., pp. 1616-1630. Width in Prestressed Concrete Beams,” Journal of the
Liu, T. C.; Campbell, R. L.; and Bombich, A. A.; 1979, Structural Division, ASCE, V. 107, No. ST2, pp. 429-433.
“Verification of Temperature and Thermal Stress Analy- Meyers, B. L.; Slate, F. O.; and Winter, G., 1969, “Relationship
sis Computer Programs for Mass Concrete Structures,” between Time Dependent Deformation and Microcracking
224R-44 ACI COMMITTEE REPORT

of Plain Concrete,” ACI JOURNAL, Proceedings V. 66, No. 1, Concrete,” ACI JOURNAL, Proceedings V. 78, No. 4, July-
Jan., pp. 60-68. Aug., pp. 262-268.
Nallathambi, P., and Karihaloo, B. L., 1986, “Influence of Nilson, A. H., 1987, Design of Prestressed Concrete, 2nd
Slow Crack Growth on the Fracture Toughness of Plain Edition, John Wiley and Sons, New York, 592 pp.
Concrete,” Fracture Toughness and Fracture Energy, F. Odman, S. T. A., 1968, “Effects of Variations in Volume,
H. Wittmann, ed., Elsevier Science Publishers, The Nether- Surface Area Exposed to Drying, and Composition of
lands, pp. 271-280. Concrete on Shrinkage,” RILEM/CEMBUREAU, Inter-
Naus, D. J., and Lott, J. L., 1969, “Fracture Toughness of national Colloquium on the Shrinkage of Hydraulic Concretes,
Portland Cement Concrete,” ACI JOURNAL, Proceedings V. 66, Madrid, 20 pp.
No. 6, June, pp. 481-489. Oh, B. H., and Kang, Y. J., 1987, “New Formulas for
Nawy, E. G., 1968, “Crack Control in Reinforced Con- Maximum Crack Width and Crack Spacing in Reinforced
crete Structures,” ACI JOURNAL, Proceedings V. 65, No. 10, Concrete Flexural Membranes,” ACI Structural Journal,
Oct., pp. 825-836. V. 84, No. 2, Mar.-Apr., pp. 103-112.
Nawy, E. G., 1972, “Crack Control through Reinforcement Perry, C., and Gillott, J. E., 1977, “The Influence of Mortar-
Distribution in Two-Way Acting Slabs and Plates,” ACI Aggregate Bond Strength on the Behavior of Concrete in
JOURNAL , Proceedings V. 69, No. 4, Apr., pp. 217-219. Uniaxial Compression,” Cement and Concrete Research,
Nawy, E. G., and Blair, K. W., 1971, “Further Studies on V. 7, No. 5, Sept., pp. 553-564.
Flexural Crack Control in Structural Slab Systems,” Cracking, Pickett, G., 1956, “Effect of Aggregate on Shrinkage of
Deflection, and Ultimate Load of Concrete Slab Systems, Concrete and Hypothesis Concerning Shrinkage,” ACI
SP-30, American Concrete Institute, Farmington Hills, JOURNAL, Proceedings V. 52, No. 5, Jan., pp. 581-580.
Mich., pp. 1-41. Polivka, R. M., and Wilson, E. L., 1976, “Finite Element
Nawy, E. G., and Huang, P. T., 1977, “Crack and Deflec- Analysis of Nonlinear Heat Transfer Problems,” Report No.
tion Control of Pretensioned Prestressed Beams,” PCI Jour- UC SESM 76-2, Department of Civil Engineering, University
nal, V. 22, No. 3, May-June, pp. 30-47. of California, Berkeley, Calif., June.
Nawy, E. G., and Neuwerth, G. E., 1977, “Behavior of Poston, R. W.; Carrasquillo, R. L.; and Breen, J. E., 1987,
Concrete Slabs, Plates and Beams with Fiber Glass as Main “Durability of Post-Tensioned Bridge Decks,” ACI Materi-
Reinforcement,” Proceedings, ASCE, V. 103, ST2, Feb., als Journal, V. 84, No. 4, July-Aug., pp. 315-326.
pp. 421-440. Poston, R. W., and Schupack, M., 1990, “Durability of a
Nawy, E. G.; Ukadike, M. M.; and Sauer, J. A.; 1977, “High Precast Prestressed Concrete Conveyor Bridge Structure at a
Strength Field Polymer Modified Concretes,” Proceedings, Salt Mine after 17 Years of Service,” Paul Klieger Symposium
ASCE, V. 103, ST12, Dec., pp. 2307-2322. on Performance of Concrete, SP-122, D. Whiting, ed.,
Nawy, E. G., 1989a, “Flexural Cracking Behavior of Par- American Concrete Institute, Farmington Hills, Mich.,
tially Prestressed Pretensioned and Post-Tensioned pp. 363-384.
Beams—State of the Art,” Cracking in Prestressed Concrete Rao, A.; Gandotra, K.; and Ramaswamy, G. S., 1976,
Structures, SP-113, G. T. Halvorsen and N. H. Burns, eds., “Flexural Tests on Beams Prestressed to Different Degrees
American Concrete Institute, Farmington Hills, Mich., pp. 1- of Prestress,” Journal, Institution of Engineers (Calcutta),
28. V. 56, May.
Nawy, E. G., 1989b, “Stresses, Strains, and Bursting Refai, T. M. E., and Swartz, S. E., 1987, “Influence of Size
Cracks in Anchorage Zones of Post-Tensioned Beams,” Effects on Opening Mode Fracture Parameters for Precracked
Cracking in Prestressed Concrete Structures, SP-113, G. T. Concrete Beams in Bending,” Proceedings, SEM-RILEM
Halvorsen and N. H. Burns, eds., American Concrete In- International Conference on Fracture of Concrete and
stitute, Farmington Hills, Mich., pp. 109-132. Rock, S. P. Shah and S. E. Swartz, eds., Houston, Tex.
Nawy, E. G., 2001, Fundamentals of High-Performance Rehm, G., and Eligehausen, R., 1977, “Lapped Splices of
Concrete, John Wiley and Sons, New York, 446 pp. Deformed Bars Under Repeated Loadings (Ubergreifungss-
Nepper-Christensen, P., and Nielsen, T. P. H., 1969, toBe von Rippenstahlen unter nicht ruhender Belastung),”
“Modal Determination of the Effect of Bond between Coarse Beton Und Stahlbetonbau (Berlin), No. 7, pp. 170-174.
Aggregate and Mortar on the Compressive Strength of Roshore, E. C., 1967, “Field Exposure Tests of Reinforced
Concrete,” ACI J OURNAL, Proceedings V. 66, No. 1, Jan., Concrete Beams,” ACI JOURNAL, Proceedings V. 64, No. 5,
pp. 69-72. May, pp. 253-257.
Neville, A. M., and Hirst, G. A., 1978, “Mechanism of Sanders, D. H.; Breen, J. E.; and Duncan III, R. R., 1987,
Cyclic Creep of Concrete,” Douglas McHenry Symposium “Strength and Behavior of Closely Spaced Post-Tensioned
on Concrete and Concrete Structures, SP-55, B. Bresler, ed., Monostrand Anchorages,” PTI Report, Oct., 49 pp.
American Concrete Institute, Farmington Hills, Mich., Schlaich, J.; Schafer, K.; and Jennewein, M.; 1987,
pp. 83-101. “Toward a Consistent Design of Structural Concrete,”
Neville, A. M., and Brooks, J. J., 1985, Concrete Technol- PCI Journal, V. 32, No. 3, May-June, pp. 74-150.
ogy, John Wiley & Sons, Inc., New York. Schrader, E. K. and Munch, A. V., 1976, “Deck Slab
Ngab, A. S.; Slate, F. O.; and Nilson, A. H., 1981, “Micro- Repaired by Fibrous Concrete Overlay,” Proceedings,
cracking and Time-Dependent Strains in High-Strength ASCE, V. 102, No. CO1, Mar., pp. 179-196.
 CONTROL OF CRACKING IN CONCRETE STRUCTURES 224R-45

Schrader, E. K.; Fowler, D. W.; Kaden, R. A.; and Steb- Stone, W. C., and Breen, J. E., 1984b, “Design of Post-
bins, R. J., 1978, “Polymer Impregnation Used in Concrete Tensioned Girder Anchorage Zones,” PCI Journal, V. 29,
Repairs on Cavitation/Erosion Damage,” Polymers in Con- No. 2, Mar.-Apr., pp. 28-61.
crete, SP-58, American Concrete Institute, Farmington Hills, Sturman, G. M.; Shah, S. P.; and Winter, G., 1965, “Effects
Mich., pp. 225-248. of Flexural Strain Gradients on Micro-Cracking and Stress-
Schupack, M., 1985, “Durability of SFRC Exposed to Strain Behavior of Concrete,” ACI JOURNAL, Proceedings
Severe Environments,” Steel Fiber Concrete, S. P. Shah V. 62, No. 7, July, pp. 805-822.
and A. Skarendahl, eds., Elsevier Applied Science Publishers, Suaris, W., and Fernando, V., 1987, “Ultrasonic Pulse At-
London, pp. 479-496 tenuation as a Measure of Damage Growth during Cyclic
Shah, S. P., and Batson, G. B., eds., 1987, Fiber Reinforced Loading of Concrete,” ACI Materials Journal, V. 84, No. 3,
Concrete Properties and Applications, SP-105, American May-June, pp. 185-193.
Concrete Institute, Farmington Hills, Mich., 600 pp. Suri, K. M., and Dilger, W. H., 1986, “Crack Width of
Shah, S. P., and Chandra, S., 1970, “Fracture of Concrete Partially Prestressed Concrete Members,” ACI J OURNAL,
Subjected to Cyclic and Sustained Loading,” ACI JOURNAL, Proceedings V. 83, No. 5, Sept.-Oct., pp. 784-797.
Proceedings V. 67, No. 10, Oct. Suzuki, K., and Yoshiteru, O., 1984, “Calculation of
Shah, S. P.; Ludirja, D.; and Daniel, J. I., 1987, “Toughness Crack Width and Crack Control Design of PPC Members,”
of Glass Fiber Reinforced Concrete Panels Subjected to NATO Advanced Research Workshop on Partial Prestressing,
Accelerated Aging,” PCI Journal, V. 32, No. 5, Sept.-Oct., Paris, 35 pp.
pp. 82-99. Swartz, S. E., and Go, C. G., 1984, “Validity of Compliance
Shah, S. P., and McGarry, F. J., 1971, “Griffith Fracture Calibration to Cracked Concrete Beams in Bending,” Exper-
Criteria and Concrete,” Journal of the Engineering Mechanics imental Mechanics, SEM, V. 24, No. 2, pp. 129-134.
Division, Proceedings, ASCE, V. 47, No. EM 6, Dec., Tam, K. S., and Scanlon, A., 1986, “Deflection of Two-
pp. 1663-1676. Way Slabs and Transverse Loads,” ACI JOURNAL, Proceedings
Shah, S. P., and Naaman, A. E., 1976, “Mechanical Prop- V. 83, No. 5, Sept.-Oct., pp. 737-744.
erties of Glass and Steel Fiber Reinforced Mortar,” ACI Testa, R. B., and Stubbs, N., 1977. “Bond Failure and
JOURNAL, Proceedings, Jan., pp. 50-53. Inelastic Response of Concrete,” Journal of the Engineering
Shah, S. P., and Skarendahl, A., 1986, Steel Fiber Con- Mechanics Division, ASCE, V. 103, No. EM2, Apr. 1977,
crete, Elsevier Applied Science Publishers, London, 526 pp. pp. 296-310.
Shah, S. P., and Slate, F. O., 1968, “Internal Microcrack- Troxell, G. E.; Raphael, J. M.; and Davis, R. E., 1958, “Long-
ing, Mortar-Aggregate Bond, and the Stress-Strain Curve of Time Creep and Shrinkage Tests of Plain and Reinforced
Concrete,” Proceedings, International Conference on the Concrete,” Proceedings, ASTM, V. 58, pp. 1101-1120.
Structure of Concrete (London, Sept. 1965), Cement and USBR, 1981, Concrete Manual, 8th Edition, U.S. Bureau
Concrete Association, London, pp. 82-92. of Reclamation, 627 pp.
Shah, S. P., and Winter, G., 1966, “Inelastic Behavior and Verbeck, G. J., 1958, “Carbonation of Hydrated Portland
Fracture of Concrete,” ACI JOURNAL, Proceedings V. 63, Cement,” Cement and Concrete, STP-205, American Soci-
No. 9, Sept., pp. 925-930. ety for Testing and Materials, West Conshohocken, Penn.,
Slate, F. O., and Matheus, R. E., 1967, “Volume Changes pp. 17-36.
on Setting and Curing of Cement Paste and Concrete from Walker, A. J., and Lankard, D. R., 1977, “Bridge Deck
Zero to Seven Days,” ACI JOURNAL, Proceedings V. 64, No. 1, Rehabilitation with Steel Fibrous Concrete,” Presented at
Jan., pp. 34-39. the Third International Exposition on Concrete Construction
Spooner, D. C., 1972, “The Stress-Strain Relationship for (New Orleans, LA, Jan.), Battelle Columbus Laboratories.
Hardened Cement Pastes in Compression,” Magazine of Wittmann, F. H., 1986, Fracture Toughness and Fracture
Concrete Research, London, V. 24, No. 29, June, pp. 85-92. Energy of Concrete, Elsevier Science Publishers, The
Spooner, D. C., and Dougill, J. W., 1975, “A Quantitative Netherlands.
Assessment of Damage Sustained in Concrete during Wilson, E. L., 1968, “The Determination of Temperatures
Compressive Loading,” Magazine of Concrete Research, within Mass Concrete Structures,” Report No. 68-17, Structural
London, V. 27, No. 92, Sept., pp. 151-160. Engineering Laboratory, University of California, Berkeley,
Spooner, D. C.; Pomeroy, C. D.; and Dougill, J. W.; 1976, Calif., Dec.
“Damage and Energy Dissipation in Cement Pastes in Com- Woods, H., 1968, Durability of Concrete Construction,
pression,” Magazine of Concrete Research, London, V. 28, Monograph No. 4, American Concrete Institute/Iowa State
No. 94, Mar., pp. 21-29. University, Farmington Hills, Mich., 187 pp.
Stevens, R. F., 1969, “Tests on Prestressed Reinforced Yang, S., and Chen, J., 1988, “Bond Slip and Crack Width
Concrete Beams,” Concrete, London, V. 3, No. 11, Nov., Calculations of Tension Members,” ACI Structural Journal,
pp. 457-462. V. 85, No. 4, July-Aug., pp. 414-422.
Stone, W. C., and Breen, J. E., 1984a, “Behavior of Post- Yerlici, V. A., 1975, “Minimum Wall Thickness of Circular
Tensioned Girder Anchorage Zones,” PCI Journal, V. 29, Concrete Tanks,” Publication No. 35-II, International Associa-
No. 1, Jan.-Feb., pp. 64-109. tion for Bridge and Structural Engineering, Zurich, p. 237.
224R-46 ACI COMMITTEE REPORT

Yoshimoto, A.; Ogino, S.; and Kawakami, M., 1972, ACI Committee 224, 1993, “Causes, Evaluation, and
“Microcracking Effect on Flexural Strength of Concrete Repair of Cracks in Concrete Structures (224.1R-93),”
after Repeated Loading,” ACI JOURNAL, Proceedings V. 69, American Concrete Institute, Farmington Hills, Mich., 22 pp.
No. 4, Apr., pp. 232-240. ACI Committee 347, 1969, “Precast Concrete Units Used
Yoshimoto, A.; Ogino, S.; and Kawakami, M., 1976, as Forms for Cast-in-Place Concrete (347.1R-69),” American
“Microscopic Cracks in Cement Matrix and Deformation Concrete Institute, Farmington Hills, Mich.
Behavior of Concrete,” Proceedings, 19th Japan Congress
ACI Committee 544, 1996, “State-of-the-Art Report on
on Materials Research, pp. 126-131.
Fiber-Reinforced Concrete (544.1R-96),” American Concrete
Ytterberg, R. F., 1987, “Shrinkage and Curling of Slabs on
Grade,” Concrete International: Design and Construction, Institute, Farmington Hills, Mich., 66 pp.
V. 9, No. 4, Apr., pp. 22-31. ASTM C 512, “Test Method for Creep of Concrete in
Zielinski, J. L., and Rowe, R. E., 1960, “An Investigation Compression,” American Society of Testing and Materials,
of the Stress Distribution in the Anchorage Zones of Post- West Conshohocken, Penn.
Tensioned Concrete Members,” Technical Report No. 9, ASTM C 595, “Specifications for Blended Hydraulic
Cement and Concrete Association, London, Sept., 32 pp. Cement,” American Society of Testing and Materials,
West Conshohocken, Penn.
9.3—Other references Goyal, B. B., ed., 1982, Designing for Creep and Shrink-
ACI Committee 214, 1977, “Recommended Practice for
age in Concrete Structures, SP-76, American Concrete Insti-
Evaluation of Strength Test Results of Concrete (214-77),”
tute, Farmington Hills, Mich., 484 pp.
American Concrete Institute, Farmington Hills, Mich., 14 pp.
ACI Committee 224, 1990, “Control of Cracking in Designing for Effects of Creep, Shrinkage and Temperature
Concrete Structures (224R-90),” American Concrete Institute, in Concrete Structures, 1971, SP-27, American Concrete
Farmington Hills, Mich., 43 pp. Institute, Farmington Hills, Mich., 430 pp.
ACI Committee 224, 1992, “Cracking of Concrete Members Neville, A. M., ed., 1964, Symposium on Creep of Concrete,
in Direct Tension (224.2R-92),” American Concrete Institute, SP-9, American Concrete Institute, Farmington Hills,
Farmington Hills, Mich., 12 pp. Mich., 160 pp.