ACI STRUCTURAL JOURNAL TECHNICAL PAPER

Title No. 114-S46

Pure Mechanics Crack Model for Shear Stress Transfer in

Cracked Reinforced Concrete
by Paolo M. Calvi, Evan C. Bentz, and Michael P. Collins

An analytical model is presented that is capable of predicting or the main shear crack can either be computed as 11.57 MPa
assessing the response of cracks in reinforced concrete elements (1678 psi) according to the Contact Density Model,10 or as
subjected to monotonic, cyclic, or reversed cyclic in-plane shear 2.76 MPa (387 psi) employing the model of Walraven and
and normal stresses. The model is formulated in terms of global Reinhardt.9
and local equilibrium, compatibility, and stress-strain relation-
Interface shear-stress transfer across a crack is a relatively
ships. The small number of required empirical parameters are
complex process depending on parameters such as loading
determined from the test results of 14 pre-cracked reinforced
concrete panels under uniform shear and uniaxial tensile stresses. history, crack width, crack slip, compressive stress acting
The proposed model is shown to capture the important aspects of across the crack, concrete strength, aggregate size, crack
the response of cracks subjected to complex loading conditions, roughness, and tensile stress in the reinforcement crossing
resulting in improved simulations of crack behavior. In addition, it the crack. Given the many mechanical and geometrical
is shown that the model can be employed to conduct crack-based parameters involved, crack models normally rely on a signif-
structural assessments that estimate the stress state of a struc- icant number of empirical equations with constants obtained
ture based on crack displacement information collected as part of by fitting to available results.
site-monitoring activities. This paper presents a simple rational crack behavior
model for cracks subjected to cyclic and/or reversed cyclic
Keywords: aggregate interlock modeling; crack behavior; cyclic loading;
nonlinear response; reinforced concrete; shear strength; structural shear and axial loads. The model, referred to as the “Pure
assessment. Mechanics Crack Model” (PMCM), is formulated in terms
of equilibrium, compatibility and constitutive consider-
INTRODUCTION ations. The only empirical equation employed is a constitu-
After cracking, reinforced concrete (RC) members carry tive equation, successfully employed over the course of the
a significant portion of the applied shear loads across cracks past three decades.2,14 A significant feature of the proposed
through aggregate interlock action. For example, it has been model is that it can be used to perform crack-based struc-
shown that RC beams not containing shear reinforcement tural assessments (that is, estimating the stress state of
carry up to 76% of the total shear load across existing shear the structure from crack opening, sliding, and orientation)
cracks, with the remaining portion of the shear being resisted without needing empirical parameters and requiring only
by the uncracked portion of the cross section.1 Because of a small number of relatively easily obtainable properties
this, to better model the behavior of RC structures subjected of the structure.
to shear loads, it is necessary to accurately understand and
simulate the response of the cracks.2,3 RESEARCH SIGNIFICANCE
The shear-transfer behavior of cracked concrete inter- The model presented in this paper can be used to predict
faces has been a subject of interest to many researchers over the behavior of cracks under complex loading conditions or
the past few decades.4-15 Based on available experimental to assess the stress state and the reserve capacity of the crack
results, a number of these investigators proposed crack based on measurable crack information.
behavior models to predict the shear response of cracked
interfaces subjected to monotonic and cyclic loading condi- OVERVIEW AND CONCEPTUAL MODEL
tions.9-11 However, all the proposed models show limitations A recent experimental study conducted at the University of
depending on the context they have been used in and none Toronto involved 14 precracked RC panel elements subjected
have been shown to perform consistently better than the to cyclic or reversed cyclic shear and axial tension.4,15 The
others.4 In this regard, it is of concern that the shear stress elements were tested using the Panel Element Tester and
transfer predicted by the different models for particular crack were subjected to uniform tensile stress fx and uniform shear
widths and crack slips can vary by factors of up to five.4 stress v, as shown in Fig. 2. Each element was reinforced in
To illustrate the variability of existing crack models, the two orthogonal directions, x and y, and was crossed by a
consider the crack outlined in Fig. 1, which formed in concrete
with a strength of 38 MPa (5510 psi) and a maximum aggre- ACI Structural Journal, V. 114, No. 2, March-April 2017.
gate size of 14 mm (0.55 in.). If it is assumed that the crack MS No. S-2016-157, doi: 10.14359/51689460, received April 18, 2016, and
reviewed under Institute publication policies. Copyright © 2017, American Concrete
width is 1.25 mm (0.05 in.) and the crack slip is 2.5 Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
mm (0.10 in.), the amount of shear stress transferred across closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.

ACI Structural Journal/March-April 2017 545

major crack oriented 90 degrees to the x-axis. The longitu- the applied loads normally does not induce much change in
dinal bars crossing the crack were placed inside plastic tubes the crack displacements, while it does cause an increase of
to eliminate dowel action. the compressive stress on the crack.
Vertical equilibrium of the free body diagram in Fig. 2(b) Reversing the direction of the applied shear stress also does
dictates that the shear stress, vci, transmitted across the crack not initially have any effect on the crack displacements, until
interface by aggregate interlock action must equal the shear the applied stresses have reached a certain magnitude. The
stress v applied to the element. Horizontal equilibrium of application of this “activation stress” results in the reversal of
the same body requires that the tensile stress in the rein- the crack slip direction and in the partial closure of the crack.
forcement at the crack interface is a function of the applied Note that the crack continues displacing until the aggregate
tensile stress fx and of the normal compressive stress fci that on the opposite side engages. Higher loads induce progres-
develops at the crack location. sive crack sliding and opening, and progressive loading
Figure 3 shows the typical response of a crack subjected to cycles lead to the repetition of the sequence described. In
reversed cyclic shear and cyclic tensile stresses. As the shear addition, cycling the applied loads has the effect of causing
stress v increases, the crack width w increases and the crack progressive deterioration of the crack surface and reductions
slip s begins. Cracks tend to experience a quite stiff initial of both crack stiffness and crack strength.
response both in terms of crack slip and crack width. During
these initial load stages, in addition to the shear stress, the ANALYTICAL MODEL
crack is capable of transferring some direct tensile stress.4,15 Compatibility conditions
Increasing the applied loads causes damage at the crack The structural system under consideration consists of a
interface level in the form of local crushing due to exces- cracked reinforced concrete element such as the one outlined
sive stresses at the contact points. The degradation of the in Fig. 2. Assuming that the single set of reinforcing steel
crack surface normally corresponds to a softening of the bars that cross the crack can be considered to be unbonded
shear stress-crack displacement response, accompanied by over a given effective length Leff, global compatibility
the loss of the capacity to transfer direct tension. Reducing requirements can be formulated expressing the strain in the
steel reinforcement as a direct function of the crack displace-
ments, namely crack width w and crack slip s

w
ε sx = − ε sh (1)
Leff

(L )
2
eff + w − ε sh Leff + s 2 − Leff
ε sx = (2)
Leff
Note that the only difference between Eq. (1) and (2) is
that the latter accounts for second-order effects. For this
reason, Eq. (2) may be more appropriate for analysis of “ulti-
mate strength” cases. The term εsh represents the strain due
to concrete shrinkage and can be taken as 0.3 × 10–3 in the
Fig. 1—Shear-critical reinforced concrete beam. (Adapted absence of more accurate measurements.
from Sherwood et al.1)

Fig. 2—Free-body diagram of structural system under investigation.

546 ACI Structural Journal/March-April 2017

Fig. 3—Typical shear stress-crack slip response.
In addition to restraining the steel deformation to be a
function of the crack displacements, local compatibility
requirements dictate a relationship between crack width
and crack slip. More specifically, the crack width and the
crack slip evolve in conformity with the shape of the steepest
aggregate particles crossing the crack. In the model, it is
assumed that the overall geometry of the crack contact plane
can be defined with sufficient accuracy through the crack
width-crack slip ratio. This effective slip plane, referred to
as the “crack secant slope,” is inclined at α degrees to the
average orientation of the crack; refer to Fig. 4 and Eq. (3).
Note that for the cases shown in Fig. 4, the average crack
orientation is vertical.

w
tan (α ) = (3)
s
In a similar approach to that of Walraven and Reinhardt,9
it is assumed that at the time a crack opens, the crack surface
consists of a series of spherical aggregate particles. To
further simplify the problem, it is assumed that the crack
plane intersects the most critical of these particles at their
middepth. The crack slip can then be computed from geom-
etry alone as

a g − a g 2 − 4 ⋅ w2
s= (4)
2
where ag represents the maximum aggregate size.
Note that Eq. (4) is only valid at the initial stages of the
life of a crack. As local crushing phenomena begin to occur,
the crack slip can no longer be computed just from geom-
etry, as will be discussed later.
Fig. 4—Free-body diagram of aggregate at contact point
Equilibrium conditions (note that steel inclination is intentionally exaggerated).
As already mentioned, vertical equilibrium of the free body
shown in Fig. 2(b) requires that the shear stress v applied to the right face of the element and the crack plane, these two
the right face of the panel must be exactly balanced by shear global equilibrium equations for the element can be formu-
stresses on the crack interface, vci. Similarly, horizontal lated as
equilibrium requires that the tensile force in the longitudinal
reinforcement at the crack must be in equilibrium with the fci = ρx ∙ fsx,cr – n ∙ v (5)
applied tensile force fx and the compressive stress fci that
develops on the crack plane. Assuming a unit area for both
vci = v (6)

ACI Structural Journal/March-April 2017 547

 Note that the sign convention adopted in the derivation of
Eq. (7) implies that the stress fci assumes positive values if
compressive and negative values if tensile on the crack.
It can be seen in Fig. 5(a) that fci-vci ratios estimated via
Eq. (7) with a friction coefficient value of 0.8 are in good
agreement with the experimental results from all load levels
of the 14 panels tested by Calvi et al.15

Unloading phase
The crack system subjected to a given combination of
shear and tensile stresses displaces until an equilibrium posi-
tion has been reached. The problem at hand is now to deter-
mine the crack response under decreasing applied loads. The
local stress state at a contact point evolves from the scenario
depicted in Fig. 4(a) to that outlined in Fig. 4(b) and 4(c).
Over the course of the unloading phase, two possible
scenarios can occur:
1. The displacement configuration remains essentially
unchanged; or
2. Crack width and crack slip tend to decrease somewhat
proportionally to the reduction in applied load magnitude.
Crack width and crack slip are insensitive to stress reduc-
tions if

vci
≥ tan α (9)
f ci
If vci/fci < tanα, crack displacements can be either affected
or unaffected by a reduction in applied loads. More specifi-
cally, the crack motion tends to reverse if

vci sin α − µ ⋅ cos α

≤ (10)
f ci cos α + µ ⋅ sin α
Fig. 5—Evolution of crack stresses as function of crack
geometry. It is shown in Fig. 5(b) that vci-fci ratios estimated via
Eq. (10) are in reasonably good agreement with the exper-
where ρx is the longitudinal reinforcement ratio; fsx,cr is the imental results.15 Note that the evolution of crack displace-
tensile stress in the longitudinal reinforcement at the crack ments under decreasing loading is essentially a function of
interface; and n is the tension-to-shear loading ratio (fx/v). crack surface geometry and the friction coefficient. Thus
when the shear has been reduced to zero, Eq. (10) reduces to
Loading phase
As shown in Fig. 4(a), the aggregate “contact point” on the μ ∙ cosα – sinα < 0 (11)
“contact plane,” which is inclined at angle α to the vertical,
must transmit a vertical force equal to vci and a horizontal In the case when tanα is greater than μ, the tangential
force equal to fci. These two forces result in a downwards component of the force induced in the system by the steel
tangential force on the contact plane of vcicosα minus fcisinα reinforcement is sufficient to bring the crack back toward its
and a compressive force normal to the plane of vcisinα plus initial configuration, while no change in crack displacement
fcicosα. For equilibrium, the tangential force on the plane occurs in cases where tanα ≤ μ.
must not exceed the coefficient of friction, µ, times the
normal force on the plane. This condition will be satisfied if Loading reversal phase
vci is less than the value given by Eq. (7). As noted, after significant slip has occurred, cracks do
not exhibit self-centering properties. Thus crack slip is
f ci ⋅ (sin α + µ ⋅ cos α ) typically only reversed after the application of a significant
vci = (7)
(cos α − µ ⋅ sin α ) shear stress in the reverse direction. The shear stress which
induces reversed motion is referred to as the activation shear
Combining Eq. (5), (6), and (7), equilibrium conditions
stress, vact. Once the shear stress on the crack vci equals the
can be expressed in a sometimes more convenient way as
activation stress vact, reverse motion initiates and the crack
configuration changes as shown in Fig. 4(b) and 4(c).
ρx ⋅ f sx , cr ⋅ (sin α + µ ⋅ cos α ) The activation shear stress can be computed looking at
vci = (8)
(cos α − µ ⋅ sin α + n ⋅ sin α + n ⋅ µ ⋅ cos α ) equilibrium requirements, bearing in mind that the activa-

548 ACI Structural Journal/March-April 2017

tion of reversed motion can either be due to sliding along the
aggregate surface or, more rarely, to loss of contact between
the crack faces.
The shear stress which activates sliding of the system in
Fig. 4 is computed as

f ci ⋅ ( µ ⋅ cos α − sin α )
vact ,1 = (12)
(cos α + µ ⋅ sin α )
The shear stress that produces loss of contact between the
opposing faces of the crack can be computed as

ρsx ⋅ f sx ⋅ cos α
vact , 2 = (13)
(sin α + n cos α )
The activation stress is then taken as the value whose
absolute value is the smaller of that given by Eq. (12)
and (13).

Constitutive relations
Constitutive relationships are required to link stresses
and strains for both the steel reinforcement and the aggre-
gate-paste matrix at the crack location. Note that the
compatibility and equilibrium equations introduced so far
are not sufficient in number to make the problem at hand
statically determinant.
In regards to the steel reinforcement crossing the crack
plane, the stress fsx,cr is assumed to depend only on the axial
strain in the reinforcement εsx,cr. In addition, it is assumed
that the dowel action of the reinforcement can be neglected.
Consequently, axial stress and axial strain can be related
through the usual bilinear uniaxial stress-strain relationships
such as

fsx,cr = Es ∙ εsx,cr ≤ fsy (14)

where Es is the elastic modulus of steel; and fsy is the

yield stress.
The formulation of a constitutive model for the aggre-
gate-paste matrix in contact at the crack interface is more
complex, due to its dependency upon parameters such as the
shear across the crack vci, the crack width w, the crack slip
s, and the orthogonal stress fci that arises on the crack. As
discussed in the introduction, aggregate interlock mecha-
nisms have been studied by a number of investigators. In the Fig. 6—Relationships between shear stress vci and orthog-
mid-1980s, relying on experiments of Walraven and Rein- onal stress fci. (Note: 1 MPa = 145 psi.)
hardt,9 Walraven,16 and Vecchio and Collins2 derived the where fc′ is the concrete compressive strength (MPa); w is
following empirical expression for the shear stress that can the crack width (mm); and ag is the maximum aggregate
be transmitted across a crack size (mm).
Although empirical in nature, Eq. (15) and (16) have
f ci 2 been shown capable of good performances when used in
vci = 0.18 ⋅ vci , max + 1.64 ⋅ f ci − 0.82 ⋅ (15)
vci , max early Modified Compression Field Theory (MCFT) calcula-
tion routines2 as well as in other occasions14 and have been
where vci,max is given by successfully used to estimate the relationship between shear
and orthogonal stresses on the crack over the course of a
fc′ recent experimental campaign15; refer to example compar-
vci , max = (16) isons in Fig. 6.
24 ⋅ w
0.31 + As shown by Calvi4 and Calvi et al.,15 reasonably wide
(
ag + 16 ) cracks that are subjected to shear can also transfer signifi-

ACI Structural Journal/March-April 2017 549

 Fig. 8—Crack slip increment as function of energy absorbed
by crack system. (Note: 1 N = 0.2248 lbf; 1 mm = 0.03937 in.;
1 mm2 = 0.00155 in.2.)
increment proportional to the energy per unit area absorbed
Fig. 7—Evolution of shear stress, crack slip, and energy by the crack system (intended as the area under the shear
absorbed by crack system (PC9). Note that energy term was stress-crack slip curve). The crack slip results were therefore
scaled down by factor of 10. (Note: 1 MPa = 145 psi; 1 cm = computed in the following fashion
0.3937 in.; 1 mm = 0.03937 in.)
s = s(w, μ, fci, vci) + Δs(En) (17)
cant direct tensile stresses, especially under reversed cyclic
loadings. In fact, Eq. (15) was only structured to simulate the where En is the energy term calculated as the area under the
relation between compressive stress fci and shear stress vci, shear stress-crack slip curve, as
with little or no care for the portion of the curve falling on
the negative side (with reference to the fci axis) of the graph. En = ∫vci(s) ∙ ds (18)
However, it will be shown that, if employed in combination
with the other equations forming the crack behavior model Note that the first term on the right side of Eq. (17)
being presented, Eq. (15) can be used in the crack-response represents the crack slip that a crack would experience under
calculation routine to capture the full tension to compression monotonic loading conditions. This term should be calcu-
evolution of the orthogonal stress fci. lated as the maximum of Eq. (4) and

Shear stiffness degradation f ci + µ ⋅ v

The main effect of subjecting a crack to reversed cyclic s = w⋅ (19)
v − µ ⋅ f ci
loading conditions is the increased deterioration of the
crack surface. Progressive and recurring contacts between where Eq. (19) was derived by rearranging Eq. (7), and
the opposing sides of the crack result in local crushing noting that sinα/cosα = w/s. The second term in Eq. (17)
phenomena which tend to decrease the crack roughness, represents the crack slip off-set induced by the application
smoothing the crack plane (that is, decrease of the crack of the loads in a reversed cyclic fashion.
secant slope defined earlier). A reduction in crack roughness An empirical relationship between Δs and En was calibrated,
can cause substantial degradation of both the shear stiffness relying on experimental evidence. First, the energy per unit area
and the shear strength of the crack. absorbed by each crack system tested under reversed cyclic
The crack response softening and weakening is mainly loading conditions was computed using Eq. (19). Then, the
a function of the magnitude of the loading cycles and of response of each crack was estimated under monotonic loading
the number of loading cycles that are completed.13,15 For conditions, using the behavior model as introduced so far. At
instance, it has been experimentally observed that consec- each load stage level, the monotonic crack slip value was taken
utive cycles performed at the same magnitude roughly as a reference for the calculation of consecutive increments of
provoke similar increments in crack slip. At the same time, slip associated with progressive loading cycles. The crack slip
it was noted that more intense loads cause higher crack slip off-set was therefore calculated as the difference between the
increments. An example of this is provided in Fig. 7, where experimental values of crack slip, taken at various load levels,
the symbols Δs1, Δs2, and Δs3 are used to represent the crack and the corresponding theoretical crack slip, calculated for
slip increments measured from a crack slip reference value, monotonic loads. The results of this process are summarized
taken as the value of crack slip expected under monotonic in Fig. 8.
loading conditions. The power trend-line equation with appropriate character-
The modeling approach adopted was therefore to quantify istics and a good fit to the data presented in Fig. 8 is
the crack response degradation by introducing a crack slip

550 ACI Structural Journal/March-April 2017

Fig. 9—Crack model solution procedure.

ACI Structural Journal/March-April 2017 551

Fig. 10—Observed and predicted response of Specimen PC15.
Δs = 0.169 ∙ En0.634 (20) CRACK RESPONSE ASSESSMENT
An important feature of the crack behavior model presented
in this paper is its capability to perform crack-based struc-
CRACK RESPONSE PREDICTIONS tural assessments. That is, from crack width, crack slip, and
This section provides some guidance in regards to the crack orientation data collected on site during inspection of
implementation of the crack behavior model introduced a structure, an estimate can be made of the corresponding
and provides one example of response prediction obtained stress state of the structure. When crack width, slip, and
employing the solution procedure presented. orientation are known, the set of equations constituting the
The general solution algorithm to be employed when core of the formulated crack behavior model reduces to the
dealing with crack systems subjected to stress states ranging following four equations
from monotonic to cyclic or reversed cyclic is summarized
in Fig. 9. ρx f sx , cr ( w + µ ⋅ s )
Specimen PC15 of the experimental campaign was a high- vci = (21)
s − µ⋅w + n⋅w + n⋅µ⋅s
strength concrete (fc′ = 103 MPa [14.9 ksi]) element loaded
in reversed cyclic shear and cyclic tension (the loading ratio,
n, was 0.25) and was characterized by a residual crack width fci = ρxfsx,cr – n ∙ vci (22)
of approximately 0.20 mm (0.01 in.) after the release of the
initial tension which caused the crack. Maximum aggre-
gate size, steel yield stress, and the steel effective unbonded fsx,cr = Es ∙ εsx ≤ fsy (23)
length were equal to 14 mm (0.55 in.), 618 MPa (89.6 ksi),
and 700 mm (27.5 in.), respectively.
The predicted crack response is compared with the exper- w
ε sx = − ε sh (24)
imental results in Fig. 10. It can be seen that the crack Leff
behavior model provides a reasonably good representation of
With that formulation, the stress state of an element such
the response observed experimentally in terms of both crack
as that shown in Fig. 2 can be directly calculated from the
stresses and crack displacements. The crack failure mode, which
three crack parameters without performing any iterations
can be defined as a combination of reinforcement yielding and
nor employing any empirical equation or parameter other
concrete crushing, is also well capture by the model and results
than the friction coefficient μ. To this extent, it should be
in a good estimate of the ultimate strength.
recalled that Eq. (21) and Eq. (22) were derived satisfying
equilibrium constraints, Eq. (23) represents the simplest and
most commonly employed constitutive formulation for steel

552 ACI Structural Journal/March-April 2017

Fig. 11—Observed and assessed response of Specimen PC4.
reinforcement, and Eq. (24) was derived satisfying basic the structure over the course of its lifetime (including that
compatibility requirements. In addition, note that the stress shown in Fig. 1).
state of the system can be estimated without the need to
know the mechanical characteristics of the concrete. CONCLUSIONS
Specimen PC4 of the experimental campaign was loaded The crack behavior model presented in this paper is
in shear and tension (n = 1.0) and was characterized by a capable of predicting the behavior of cracks in RC elements
residual crack width of approximately 0.25 mm (0.01 in.). subjected to complex stress states.
Crack width and crack slip, monitored throughout the whole The crack response is accurately simulated in terms of
test, evolved as shown in Fig. 11(a). Those crack displace- all key parameters, including the crack normal stress fci
ment measurements were used in combination with Eq. (21) (neglected in some of the existing models, including the
to (24) to estimate the stress state over the course of the test. MCFT) along with its orientation and its transition from
The comparison of the calculated (employing a friction tension to compression over the course of the loading stages.
coefficient μ of 0.8 and the experimental effective length The model is entirely formulated in terms of equilibrium,
of 700 mm [27.6 in.]) and observed response of specimen compatibility, and constitutive equations and is structured
PC4 is provided in Fig. 11(b), 11(c), and 11(d). to simulate cracks subjected to in-plane monotonic, cyclic,
The results shown in Fig. 11 demonstrate that the crack and reversed cyclic shear and axial loads. The model inputs
model has the capability of associating the correct values of are the basic properties of the structure: concrete strength,
shear stress vci and orthogonal stress fci to the corresponding loading ratio, reinforcement ratio, steel yielding stress,
crack displacements. The stress state of the system is indeed concrete friction coefficient, and maximum aggregate size.
estimated with a high level of accuracy at all load levels. The The crack model can also be adapted and used to conduct
divergence between calculated and observed response that structural assessment that is estimating the stress state of a
occurs at the late load stages is likely due to the additional structure based on crack displacement information collected
strength provided by the steel reinforcement through dowel over the course of site monitoring activities. In this regard,
action and second order effects, which are not currently the only required input data are the steel yielding stress, the
incorporated in the model. reinforcement ratio, the effective embedment length, and the
Provided that cracks are plastic systems, a crack displace- concrete friction coefficient.
ment based assessment approach such as the one presented At the present stage, the formulation presented is
does not have the capability of capturing load reduc- specific for cracks crossed by a single set of steel rein-
tion events. It is nonetheless appealing that, given a crack forcement oriented perpendicularly to the crack plane
displaced configuration, the model provides a means of and the transition from crack displacements to strains in
accurately estimating the peak stresses experienced by the steel reinforcement is performed by using a quantity

ACI Structural Journal/March-April 2017 553

referred to as effective reinforcement length, which is 3. Vecchio, F. J., “Disturbed Stress Field Model for Reinforced Concrete:
Formulation,” Journal of Structural Engineering, ASCE, V. 126, No. 9,
normally an unknown quantity. 2000, pp. 1070-1077. doi: 10.1061/(ASCE)0733-9445(2000)126:9(1070)
Research is underway to generalize the crack behavior 4. Calvi, P., M., “A Theory for the Shear Behaviour of Cracks Providing
model presented both in terms of crack and steel reinforce- the Basis for the Assessment of Cracked Reinforced Concrete Structures,”
PhD dissertation, University of Toronto, Toronto, ON, Canada, 2015,
ment orientation. In addition, universal expressions for 346 pp.
the calculation of the reinforcement effective length are 5. Mattock, A. H., and Hawkins, N. M., “Shear Transfer in Reinforced
currently under development. Concrete—Recent Research,” PCI Journal, V. 17, No. 2, 1972, pp. 55-75.
doi: 10.15554/pcij.03011972.55.75
The model will also be further validated versus available 6. Laible, J. P., “An Experimental Investigation of Interface Shear
experimental evidence, particularly to test the reliability of Transfer and Applications in the Dynamic Analysis of Nuclear Containment
the degradation rule proposed with respect to scenarios in Vessels,” PhD thesis, Cornell University, Ithaca, NY, 1973, 686 pp.
7. Paulay, T., and Loeber, P. J., “Shear Transfer by Aggregate Interlock,”
which the cracks are subjected to large number of loading Shear in Reinforced Concrete, SP-42, American Concrete Institute, Farm-
cycles. ington Hills, MI, 1974, pp. 1-15.
8. Bažant, Z. P., and Gambarova, P., “Rough Cracks in Reinforced
Concrete,” Journal of the Structural Division, ASCE, V. 106, No. 4, 1979,
AUTHOR BIOS pp. 819-842.
Paolo M. Calvi is an Assistant Professor at the University of Washington,
9. Walraven, J. C., and Reinhardt, H. W., “Theory and Experiments
Seattle, WA. He received his bachelor’s and master’s degrees from the
on the Mechanical Behavior of Cracks in Plain and Reinforced Concrete
University of Pavia, Pavia, Italy, in 2008 and 2010, respectively, and his
Subjected to Shear Loading,” Heron, V. 26, No. 1A, 1981, 68 pp.
PhD from the University of Toronto, Toronto, ON, Canada, in 2015. His
10. Li, B.; Maekawa, K.; and Okamura, H., “Contact Density Model for
research interests include testing and analysis of large-scale reinforced
Stress Transfer Across Cracks in Concrete,” Journal of the Faculty of Engi-
concrete and prestressed concrete structural components.
neering, University of Tokyo, V. 40, No. 1, 1989, pp. 9-52.
11. Bujadham, B., and Maekawa, K., “The Universal Model for Stress
Evan C. Bentz, FACI, is an Associate Professor of civil engineering at the
Transfer Across Cracks in Concrete,” Proceedings of JSCE, V. 17, No. 451,
University of Toronto. He received his bachelor’s degree from the Univer-
1992, pp. 277-287.
sity of Waterloo, Waterloo, ON, Canada, in 1994, and his PhD from the
12. Nagle, T. J., and Kuchma, D. A., “Shear Transfer Resistance in High-
University of Toronto in 2000. He is Chair of ACI Committee 365, Service
Strength Concrete Girders,” Magazine of Concrete Research, V. 59, No. 8,
Life Prediction, and is a member of Joint ACI-ASCE Committee 445, Shear
2007, pp. 611-620. doi: 10.1680/macr.2007.59.8.611
and Torsion.
13. Gebreyouhannes, E.; Kishi, T.; and Maekawa, K., “Shear Fatigue
Response of Cracked Concrete Interface,” Journal of Advanced Concrete
ACI Honorary Member Michael P. Collins is a Professor of structural
Technology, V. 6, No. 2, 2008, pp. 365-376. doi: 10.3151/jact.6.365
engineering at the University of Toronto. He is a member and past Chair of
14. Sagaseta, J., and Vollum, R. L., “Influence of Aggregate Fracture on
Joint ACI-ASCE Committee 445, Shear and Torsion. His research interests
Shear Transfer through Cracks in Reinforced Concrete,” Magazine of Rein-
include developing rational but simple shear design procedures for both
forced Concrete Research, V. 63, No. 2, 2011, pp. 119-137. doi: 10.1680/
reinforced and prestressed concrete structures.
macr.9.00191
15. Calvi, P. M.; Bentz, E. C.; and Collins, M. P., “Reversed Cyclic
REFERENCES Experiments on Shear Stress Transfer across Cracks in Reinforced Concrete
1. Sherwood, E. G.; Bentz, E. C.; and Collins, M. P., “Effect of Aggre- Elements,” ACI Structural Journal, V. 113, No. 4, July-Aug. 2016,
gate Size on Beam-Shear Strength of Thick Slabs,” ACI Structural Journal, pp. 851-859. doi: 10.14359/51688926
V. 104, No. 2, Mar.-Apr. 2007, pp. 180-190. 16. Walraven, J. C., “Fundamental Analysis of Aggregate Interlock,”
2. Vecchio, F. J., and Collins, M. P., “The Modified Compression Field Proceedings, ASCE, V. 107, No. 11, 1981, pp. 2245-2270.
Theory for Reinforced Concrete Elements Subjected to Shear,” ACI Journal
Proceedings, V. 83, No. 2, Mar.-Apr. 1986, pp. 219-231.

554 ACI Structural Journal/March-April 2017