Construction and Building Materials 46 (2013) 211–223

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

A model for the prediction of the punching resistance of steel fibre

reinforced concrete slabs centrically loaded
Bernardo N. Moraes Neto a,c, Joaquim A.O. Barros b,⇑, Guilherme S.S.A. Melo c
a
University of Minho, Dept. Civil Eng., School of Eng., Guimarães, Portugal
b
University of Minho, ISISE, Dept. Civil Eng., School of Eng., Campus de Azurém, Guimarães, Portugal
c
University of Brasília – UnB, Dept. Civil Eng., Campus of Darcy Ribeiro, Brasília, Brazil

h i g h l i g h t s

 An analytical model is developed for the prediction of the punching load of SFRC slabs.
 A new approach is proposed to simulate the fibre reinforcement contribution.
 A data base of 154 punching tests was built to assess the predictive performance of the model.
 By predicting the experimental results of the DB the accuracy of the model was evidenced.
 The predictive performance was also demonstrated by comparison to the previsions of other models.

a r t i c l e i n f o a b s t r a c t

Article history: With the aim of contributing for the development of design guidelines capable of predicting with high
Received 23 February 2013 accuracy the punching resistance of steel fibre reinforced concrete (SFRC) flat slabs, a proposal is pre-
Received in revised form 27 April 2013 sented in the present paper and its predictive performance is assessed by using a database that collects
Accepted 29 April 2013
the experimental results from 154 punching tests. The theoretical fundaments of this proposal are based
Available online 26 May 2013
on the critical shear crack theory proposed by Muttoni and his co-authors. The proposal is capable of pre-
dicting the load versus rotation of the slab, and attends to the punching failure criterion of the slab. The
Keywords:
proposal takes into account the recommendations of the most recent CEB-FIP Model Code for modelling
Reinforced concrete
Flat concrete slab
the post-cracking behaviour of SFRC. By simulating the tests composing the collected database, the good
Punching predictive performance of the developed proposal is demonstrated.
Steel fibre reinforced concrete Ó 2013 Elsevier Ltd. All rights reserved.
Analytical models

1. Introduction The resisting tensile stresses supported by the steel fibres in a

In recent years the use of steel fibres to increase of the punching
resistance, and mainly, to convert brittle punching failure mode
into ductile flexural failure mode of reinforced concrete (RC) flat
slabs has been explored. In fact, available research [1–3] showed
that, if proper mix compositions of steel fibre reinforced concrete
(SFRC) are used, steel fibres can be suitable shear reinforcement
for RC flat slabs, by improving the load carrying capacity and the
energy absorption performance of the column–slab connection.
These benefits are derived from the fibre reinforcement mecha-
nisms provided by fibres bridging the micro-cracks that arrest
the crack propagation, favouring the occurrence of large number
of cracks of small width.
⇑ Corresponding author.
E-mail addresses: bnmn@hotmail.com (B.N. Moraes Neto), barros@civil.umin-
ho.pt (J.A.O. Barros), melog@unb.br (G.S.S.A. Melo).
cracked concrete have also the favourable effect of delaying the
yield initiation of longitudinal and transversal conventional steel
reinforcement, which contributes to increase the ultimate load car-
rying capacity of RC structures or to a partial suppression of con-
ventional reinforcements.
By testing prototypes of real [4,5] or smaller scale [6], the use of
steel fibres has been investigated as, practically, the unique rein-
forcement of the flat slabs for residential and commercial build-
ings. This type of slabs, generally designated by Elevated Steel
Fibre Reinforced Concrete (ESFRC) slabs, is reinforced with a steel
fibre volume percentage, Vf, of about 1%, and it includes a mini-
mum continuity bars, also referred as anti-progressive collapse
bars, placed in the bottom of the slab in the alignment of the col-
umns [7]. In spite of the promising results obtained in these tests,
reliable design models capable of predicting, with high accuracy,
the load carrying capacity, the deformational response and the
failure modes possible to occur in ESFRC slabs are not yet available,
which is a considerable resistance for a comprehensive acceptation

0950-0618/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.conbuildmat.2013.04.034
212 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223

Nomenclature

A0s area of compression reinforcement rcr radius of cracked zone

As area of tension reinforcement rq radius of the load introduction at the perimeter
b width of a isolated slab element rq,eq radius of the load introduction at the perimeter in an
b0 critical perimeter for punching shear equivalent slab of circular geometry
bq,c loaded line for square slabs in circular edge conditions rs radius of circular isolated slab element
bq,q loaded line for square slabs in rectangular edge condi- rs,eq radius of circular isolated slab element in an equivalent
tions slab of circular geometry
c0 distance of the flexural reinforcement to the concrete ry radius of yielded zone
tensile surface t tangential orientation
d internal arm of the slab V shear force
d0 diameter of the aggregates Vexp experimental punching shear strength
df diameter of fibre Vf fibre volume percentage
dg maximum diameter of the Vflex shear force associated with flexural capacity of the slab
dg0 reference diameter of the aggregates VR nominal punching shear strength
e edge of the column’s cross section VR,cd design concrete contribution to punching shear strength
E modulus of elasticity of concrete VR,d design punching shear strength
Es modulus of elasticity of reinforcement VR,fd design fibre contribution to punching shear strength
F 0s internal compressive force of compressive reinforce- VR,sd design shear reinforcement contribution to punching
ment shear strength
fc average compressive strength of concrete in cylinder Vthe theoretical punching shear strength
specimens Vu punching failure load
Fcr internal compressive force of concrete in radial direc- w shear crack opening
tion wu maximum acceptable crack width imposed by design
Fct internal compressive force of concrete in tangential conditions
direction x neutral axis of slab
fct average tensile strength of concrete (Brazilian test) z axis orthogonal to the slab with origin at the bottom
fFts post-cracking strength for serviceability crack opening surface of the slab
fFtu post-cracking strength for ultimate crack opening b efficiency factor of the bending reinforcement for stiff-
fRi residual flexural tensile strength of fibre reinforced con- ness calculation
crete corresponding to CMODi Du angle of a cracked radial segment of slab
Fs internal compressive force of tensile reinforcement e0s compressive steel reinforcement strain
Fsr internal tensile force of reinforcement in radial direction ec concrete strain
Fst internal tensile force of reinforcement in tangential ecu ultimate strain of concrete in compression zone
direction efu ultimate strain of fibre in tensile zone
fsy yield strength of reinforcement es strain of steel reinforcement in tensile zone
h slab thickness esu ultimate strain of steel reinforcement in tensile zone
I0 second moment of area of uncracked concrete cross- et,bot concrete tensile strain at the bottom surface of the slab
section mR, nominal shear stress
I1 second moment of area of cracked concrete cross-sec- mc concrete nominal shear strength
tion q tensile reinforcement ratio
L span of slab q0 compressive reinforcement ratio
lf length of fibre rf,r post-cracking tensile strength of SFRC in radial direction
mcr bending moment at crack initiation rf,t post-cracking tensile strength of SFRC in tangential
mr radial moment per unit width direction
mR resisting bending moment (plastic bending moment) sb average interracial bond strength of fibre matrix
mt tangential moment per unit width v1 curvature in stabilized cracking
r radial orientation vcr curvature at cracking
r0 radius of the critical shear crack vts tension stiffening parameter
r1 radius of the zone in which cracking is stabilized vy yielding curvature
rc radius of a circular column w rotation of slab
rc,eq radius of a circular column in an equivalent slab of cir-
cular geometry

of this structural concept that apparently has several technical and
economic advantages. Due to the brittle character of punching fail-
ure mode, the existence of a design model capable of predicting
correctly the punching resistance and the deformation capacity
of SFRC flat slabs is of paramount importance in this context. Some
analytical models were proposed for the evaluation of the punch-
ing resistance of SFRC slabs, some of them with an eminent empir-
ical nature, but the predictive performance of these models was, in
general, limited to the simulation of a relatively small number of
tests carried out by the authors [8–11]. In the present work a data-
base collecting 154 punching tests with SFRC slabs was developed
to appraise the predictive performance of these models and the
one proposed by the authors of the present work. This model is
based on the Critical Shear Crack Theory (CSCT) proposed by Mut-
toni [12], being possible to determine the punching resistance of
SFRC slab by intersecting a curve corresponding to the load versus
rotation (V–w) of the column–slab connection, with a curve that
defines the failure criterion. This model integrates the contribution
of fibre reinforcement mechanisms using the recommendations of
the most recent CEB-FIP Model Code 2010 [13]. The present paper
 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223 213

Tangential
reinforcement Edge of the slab

rq
rs Tangential crack

rc
Radial segment
Radial
reinforcement

V V Radial crack

(a)
VR rs
rq
Compression
r0
ψ

d
Punching failure surface
Tension Compression strut
Centre of rotation
rc (Root of shear crack)

(b)
Fig. 1. Column–slab connection: (a) Assumed crack pattern and reinforcement arrangement and (b) slab configuration at ultimate condition.

describes the proposed formulation and compares its predictive
performance using the aforementioned database.
2. Load versus rotation approach
2.1. Refined formulation
The load versus rotation proposed in the present work is based
on the recommendations of Muttoni [12] that are applied to a col-
umn–slab connection assuming axisymmetric conditions for this
structural component (Fig. 1). The crack pattern of the slab at ulti-
mate loading conditions can be assumed as divided into radial seg-
ments (Fig. 1a). Each radial segment is delimited by a tangential
crack formed close to the column, by two radial cracks, and by
the edge considered as a free boarder of the slab.
According to some authors [14–16], it is admissible to assume
that these radial segments rotate as free bodies in turn of the point
localized at the bottom of the punching failure surface (Fig. 1b).
Therefore, it is accepted that a radial segment has constant rotation
w between the critical shear crack and its edge. According to Mut-
toni and Schwartz [17], in the column–slab connection the shear
force V is transmitted through a compression strut formed at the
external region to the punching failure surface (Fig. 1b). In
Fig. 1b the variables rc, r0, rq and rs represent, respectively, the ra-
dius of column’s cross section, the distance to the axis of the col-
umn of the punching failure surface (r0 = rc + d/2, where d is the
internal arm of the longitudinal reinforcement), the distance of
the circumferential loading line and the radius of the slab.
The stresses and the corresponding internal forces formed in
the radial segment are indicated in Fig. 2, where Fcr and Fsr repre-
sent the resultant force in the concrete in compression and in the
reinforcement in tension, both in the radial direction, while Fct
and Fst are the resultant force in the concrete in compression and
in the reinforcement in tension, both in the tangential direction.
The stress components rf,r and rf,t represent the post-cracking ten-
sile strength of SFRC in radial and tangential direction, respec-
tively. Finally, V and VR represent the acting and the resisting
shear force, respectively. The force components Fcr and Fsr, and
the stresses rf,r generate the radial bending moment mr, while
force components Fct and Fst, and the stresses rf,t produce the tan-
gential bending moment mt (Fig. 2a and b). For the evaluation of
the mr, the vertical component of the resisting force due to dowel
effect provided by the longitudinal reinforcement was neglected,
as well as the vertical component of the resultant stresses due to
fibre reinforcement mechanisms. The representation shown in
Fig. 2c corresponds to a column–slab connection assumed in an
axisymmetric conditions. When the slab has square geometry
(Fig. 3a), it can be converted into an equivalent circular slab
(Fig. 3b) by adopting the following transformations [18]:
2e
rc;eq ¼ ð1Þ
p

bq;c for circular edge conditions
r q;eq ¼ ð2Þ
bq;q for rectangular edge conditions
2 h pffiffiffi i
rs;eq ¼  2  ð 2  1Þ  ðL  eÞ þ e ð3Þ
p
where e is the edge of the cross section of the column, and L is the
edge of the slab, both assumed of square geometry.
According to Guandalini [19] and Muttoni [12], the load versus
rotation of a slab in axisymmetric structural conditions can be di-
214 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223

(a) (b) (c)

Fig. 2. Internal forces acting in a radial segment: (a) Stresses and resultant forces, (b) bending moments, and (c) free body diagram.

Rectangular Circular Circular

edge condition edge condition edge condition

r q,e
b q,

q
c

bq,q rs,eq
L

e rc,eq

L
(a) (b)
Fig. 3. Transformation of a square slab (a) into an equivalent slab of circular geometry (b).

m m

mR mR
EI 1
(4L) 1 (2L)
χ ts

m cr m cr

EI 0
1

χ r
χ cr χ1 χy r0 ry r1 rcr rs

(a) (b)
Fig. 4. (a) 2L and 4L moment curvature (m–v) diagrams and (b) m–r diagram corresponding to 4L m–v diagram.

rectly obtained from the quadrilinear (4L) or bilinear (2L) mo- v1 and vy is installed, respectively. In these variables, the sub-
ment–curvature diagram (m–v) represented in Fig. 4a. In this fig- scripts, cr, 1, and y represent the crack initiation, the stabilization
ure, the mcr and mR represents, respectively, the bending moment of the cracking process and the yield initiation of the longitudinal
at crack initiation and the resisting bending moment (plastic bend- tensile reinforcement, respectively.
ing moment) of the slab’s cross section. The EI0 and EI1 represent The equilibrium of bending moments in O0 (Fig. 2c) yields:
the flexural stiffness of the slab’s cross section before and after
X
crack initiation, respectively. The rcr, r1 and ry is the distance from M¼0 ð4Þ
the axis of the column of the cross section where the curvature vcr, O0
 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223 215

Z rs 2
ðrq  rc Þ fct  h
VðwÞ  ¼ mr ðwÞ  r0 þ mt ðrÞ  dr ð5Þ mcr ¼ ð11Þ
2p r0 6
where mt is dependent of the distance r, which has justified the where fct is the concrete tensile strength that can be estimated
transformation of the 4L m–v diagram (Fig. 4a) into the m–r dia- according to the recommendations of CEB-FIP Model Code 2010.
gram depicted in Fig. 4b. Adopting the following assumption: The rcr(w), r1(w) and ry(w) in Eq. (10) is the position, from the
axis of the column, of the section of the slab that develops the cur-
w
vðwÞ ¼ ð6Þ vature vcr(w), v1(w) and vy(w), respectively. These curvatures can
r be determined from the following relationships:
the transformation of m(w) into m(r) is executed by performing a
mcr w
simple procedure of changing the variable: vcr ¼ ) r cr ðwÞ ¼ 6 rs ð12Þ
8 E  I0 vcr
>
> E  I0  vðwÞ if vðwÞ 6 vcr
>
>
> vcr < vðwÞ 6 v1 mcr w
< mcr
> if
v1 ¼  vts ) r 1 ðwÞ ¼ 6 rs ð13Þ
mðwÞ ¼ E  I1  ½vðwÞ þ vts  if v1 < vðwÞ 6 vy ð7Þ E  I1 v1
>
>
> mR
> if vðwÞ > vy
>
>
: mR w
vy ¼  vts ) r y ðwÞ ¼ 6 rs ð14Þ
E  I1 vy
8
>
> mR if r0 < r 6 ry For the evaluation of the uncracked flexural stiffness, EI0, the
>
> w 
>
< E  I1  r þ vts
> if ry < r 6 r1 contribution of the reinforcement is neglected:
mðrÞ ¼ mcr if r 1 < r 6 rcr ð8Þ 3
>
> h
> E  I0  w
> if r cr < r 6 r s E  I0 ¼ E  ð15Þ
>
> r 12
:
The evaluation of the flexural stiffness of SFRC cracked cross
Adopting the diagram of Fig. 4b and considering Eq. (8), the section, EI1, was executed following the procedures adopted for
integral of Eq. (5) can be decomposed into: RC members [18], and assuming a stabilized cracking phase:
Z rs Z ry Z r1 Z r1  x  x 
w 3
E  I 1 ¼ q  b  Es  d  1   1 ð16Þ
mt ðrÞ  dr ¼ mR  dr þ E  I1   dr þ E  I1  vts d 3d
r0 r0 ry r ry
Z rcr Z rs The contribution of fibre reinforcement for the EI1 is only indi-
w
 dr þ mcr  dr þ E  I0   dr ð9Þ rectly taken in the evaluation of the neutral axis, x, Fig. 5 (g and k
r1 r cr r
parameters are evaluated according to [13], and fFtu is introduced
Introducing Eq. (9) into (5) and solving the integrals yields: in next section). In Eq. (16) b is a factor intending to take into ac-
count the arrangement of the reinforcement, since the deduction
2p
VðwÞ ¼  ½mr ðwÞ  r 0 þ E  I0  w  hlnðr s Þ  lnðrcr ðwÞÞi of Eq. (10) was supported on the principle of axisymmetric struc-
ðr q  r c Þ
tural conditions, but the majority of the built and tested RC flat
þ mcr  hr cr ðwÞ  r 1 ðwÞi þ E  I1  w  hlnðr 1 ðwÞÞ  lnðry ðwÞÞi slabs have orthogonal arrangement of the reinforcement [19].

þ E  I1  vts  hr 1 ðwÞ  r y ðwÞi þ mR  hr y ðwÞ  r0 i ð10Þ According to Muttoni [12], b = 0.6 yields to satisfactory results.
The evaluation of the position of the neutral axis, x, was made
which is the equation that establishes the relationship between
according to the recommendations of CEB-FIP Model Code 2010
the rotation and the load carrying capacity of the slab. In this equa-
[13], see Fig. 5.
tion mr(w) is the radial bending moment for the rotation w, evalu-
The vts factor in Eq. (8) simulates the post-cracking tensile
ated at r = r0, and the mathematical operator hxi = x if x P 0 and
strength of cracked concrete (tension stiffening effect) that accord-
hxi = 0 if x < 0. To evaluate the plastic bending moment, mR, the rec-
ing to Muttoni [12] it can be determined from the following
ommendations of CEB-FIP Model Code 2010 [13] for the simulation
equation:
of the contribution of fibre reinforcement are adopted in the pres-
ent work (Fig. 5). A detailed description on the evaluation of mR is fct 1 mcr
available elsewhere [18]. The bending moment at crack initiation,
vts ¼  ffi 0:5  ð17Þ
q  b  Es 6  h E  I1
mcr, is obtained from Eq. (11):
This means that the fibre reinforcement was not taken into ac-
count for the evaluation of vts. Enhancements in this respect can be
adopted by using the closed form solution proposed by Taheri et al.
Δ'⋅d

A's = ρ'⋅b⋅d εc ≤εcu η⋅f c [20] for determining the moment–curvature relationship of FRC of
F's
ε's strain softening or strain hardening character, but this approach
λ⋅x
x=k⋅d

increases significantly the complexity of the model aimed to be

proposed in the present work.
d
h

2.2. Simplified formulation

y

Instead of using the quadrilinear (4L) m–v diagram to derive the

εs≤εsu Fs load versus rotation relationship, in the present section the simpler
bilinear (2L) m–v diagram (or m–r) is adopted, since it provides a
A s = ρ⋅b⋅d ε t,bot ≤ε fu fFtu
Δ⋅d

formulation more suitable for design purposes. To derive the load

b
versus rotation relationship supported on the 2L m–r diagram, it
Fig. 5. Adopted approach to evaluate the ultimate bending moment, mR (adapted is necessary to assume the slab decomposed in two regions: elastic
from the CEB-FIP Model Code 2010 [13]). (ry < r0) and elasto-plastic (r0 < ry < rs), as represented in Fig. 6b.
216 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223

m
z mR
rs
rq
r0 ry r (r =r )
mR y s

rc

V⋅Δϕ/2⋅π ry
mR r (r <r <r )
0 y s

mr mt
O' ry r (r <r )
y 0
r
t V⋅Δϕ/2⋅π
r dr
rc r0 rq rs r
(a)
(b)
Fig. 6. Hypotheses adopted to derive the simplified load–rotation formulation: (a) free body diagram of a radial segment and (b) bilinear (2L) m–r relationship.

Establishing the bending moment equilibrium in O0 point, and Combining Eqs. (23) and (26), and assuming ry = 0.35rs that was
considering the elastic hypothesis (ry < r0), it is obtained Eq. (5). determined from experimental evidence [18], the following rela-
In this equation: tionship between load and rotation was derived [18]:
w 
mr ðwÞ  r 0 ¼ E  I1   r0 ð18Þ mR  r s V
r0 w ¼ 0:5   ð27Þ
E  I1 V flex
and
Z rs Z rs Z rs Eq. (24) can get the following configuration
w dr
mt ðrÞ  dr ¼ E  I1   dr ¼ E  I1  w  3=2
r0 r0 r r0 r V
 w¼ D ð28Þ
rs V flex
¼ E  I1  w  ½lnðrÞrrs0 ¼ E  I1  w  ln ð19Þ
r0
where D is the constant relating w and (V/Vflex)3/2, whose value was
Replacing Eqs. (18) and
(19) into
 Eq.
 (5) it is obtained: determined by dimensional analysis [18]. In this analysis the same
2p rs parameters of Eq. (27) were considered, mR, rs, EI1, having resulted
VðwÞ ¼  E  I1  w  1 þ ln for ry
ðr q  r c Þ r0 the following equation [18]:
6 r 0 ðelastic regimeÞ ð20Þ  3=2
mR  rs V
w¼ D  ð29Þ
Analogously, for the elasto-plastic case, r0 < ry 6 rs, the follow- E  I1 V flex
ing equilibrium equation is derived:
where D ¼ 0:65 for regular concrete and D ¼ 1:625 for concrete of
Z rs
ðrq  rc Þ lightweight aggregates. For slabs in axisymmetric structural condi-
VðwÞ  ¼ mR  r 0 þ mt ðrÞ  dr ð21Þ
2p r0
tions, Vflex is obtained from Eq. (26), while for square slabs the yield
line theory leads to: Vflex = 8mR. Eq. (29) is the simplified expression
where for the load versus rotation that is recommended to be used in the
Z rs Z ry Z rs
w design practice.
mt ðrÞ  dr ¼ mR  dr þ E  I1 
 dr
r0 r0 ryr If the contribution of the reinforcement in compression is ne-

rs glected, and concrete crushing is assumed to occur simultaneously
¼ mR  ðr y  r 0 Þ þ E  I1  w  ln ð22Þ with the yield of the longitudinal reinforcement, Eq. (29) gets the
ry
following configuration, whose deduction is described in detail
Replacing Eq. (22) into Eq. (21) it is obtained: elsewhere [18]:

 
2p rs    3=2
VðwÞ ¼  E  I1  w  1 þ ln for r0 < r y rs fFtu ðwu Þ  ½1 þ 4  ðc0 =dÞ V
ðr q  r c Þ ry w¼ D  4  fsy þ  ð30Þ
d  Es q V flex
< rs ðelasto-plastic regimeÞ ð23Þ
where
Based on experimental results [3,21,22], it was verified that the
wu
slab’s rotation is proportional to (V/Vflex)3/2, where V and Vflex is, fFtu ðwu Þ ¼ fFts   ðfFts  0:5  fR3 þ 0:2  fR1 Þ P 0; f Fts ¼ 0:45  fR1
2:5
respectively, the actuating shear force and the load corresponding
ð31Þ
to the flexural failure of the slab:
 3=2
V  0:8
wa ð24Þ lf
V flex fR1 ¼ 7:5  V f  ð32Þ
df
To obtain Vflex the equilibrium Eq. (21) is considered with:
Z rs
 0:7
lf
mR  dr ¼ mR  ðr s  r 0 Þ ð25Þ fR3 ¼ 6:0  V f  ð33Þ
r0 df

and assuming ry = rs, see Fig. 6b, results: In Eq. (30) D ¼ 0:358 for regular concrete and D ¼ 0:894 for con-
rs crete of lightweight aggregates, and q, fsy and Es represents, respec-
V flex ¼ 2  p  mR  for r y ¼ rs ðflexural failure loadÞ ð26Þ tively, the reinforcement ratio, the yield stress and the modulus of
ðrq  rc Þ
 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223
elasticity of the flexural reinforcement. The parameter c0 in Eq. (30)
is the distance of the flexural reinforcement to the concrete tensile
surface (Dd in Fig. 5).
In Eqs. (30) and (31), fFtu(wu) is the post-cracking residual
strength of SFRC at ultimate limit conditions, where wu is the max-
imum acceptable crack width imposed by design conditions.
According to the CEB-FIP Model Code 2010 [13], fFtu(wu) should
be evaluated for wu = 1.5 mm. In Eq. (31) fR1 and fR3 are the residual
flexural tensile strength parameters, whose evaluation is carried
out according to the recommendations of CEB-FIP Model Code
2010, by performing three point notched beam bending tests
[13]. In Eqs. (32) and (33) Vf and lf/df are the fibre volume percent-
age and fibre aspect ratio (quotient between fibre length, lf, and fi-
bre diameter, df).
2.3. Prediction of the residual flexural tensile strength parameters of
SFRC by using a database
As already mentioned, the predictive performance of the pro-
posed models will be assessed by comparing the estimated results
with those available in a database (DB) that collects 154 punching
tests. In these models, the contribution of fibre reinforcement is sim-
ulated by using the concept of residual flexural tensile strength
parameters, fRi, whose values, in the majority of the works compos-
ing the DB, are not available. Therefore, to apply the proposed mod-
els to the tests composing the DB, another database was built by
collecting results (fRi) of the characterization of the post-cracking
flexural behaviour of SFRC according to the recommendations of
CEB-FIP Model Code 2010. Since the fibre volume percentage, Vf,
and fibre aspect ratio, lf/df, are practically the unique common infor-
mation available in the works forming the DB of the punching tests,
the statistical analysis performed with the collected data for the
characterization of the post-cracking behaviour of SFRC was gov-
erned by the criterion of deriving equations for the fRi dependent
on the Vf and lf/df. The authors are aware that this is a quite simple
approach to simulate the fibre reinforcement mechanisms, since
other variables like the fibre–matrix bond strength, fibre inclination
and fibre embedment length influence the values of fRi [23], but this
information is not available in those works. Therefore, a relatively
large scatter of results is naturally expected for the relationships
fRi–(Vf, lf/df), but actually this is the unique possibility of considering
the fibre reinforcement mechanisms according to the CEB-FIP Model
Code 2010 for the prediction of the punching failure load of the slabs
collected in the DB by applying the proposed models. The authors
are doing an effort to increase this database and, therefore, deriving
more reliable fRi–(Vf, lf/df) relationships. Eqs. (32) and (33) were ob-
tained according to the described methodology [18].
In a design context of a SFRC slab, three point notched SFRC
beam bending tests should be executed according to the recom-
mendations of CEB-FIP Model Code 2010 in order to obtain the fRi
of the SFRC, and these values should be directly used in the pro-
posed model for the evaluation of the punching failure load of a
SFRC slab supported on columns.
3. Failure criterion
3.1. Fundamental formulation
In the present approach it is assumed that the punching resis-
tance of a RC slab can be estimated according to the critical section
concept, e.g., the nominal shear stress, mR, is defined as the ratio be-
tween the punching failure load (Vu) and an area considered as rep-
resentative of the punching failure surface (b0d):
Vu
mR ¼ ð34Þ
b0  d
217
where b0 represents the punching critical perimeter at a distance
ad from the external surface of the column (Fig. 7). It is also as-
sumed that the contribution of fibre reinforcement can be inte-
grated in Eq. (34) by using a kf factor:
Vu 1
mR ¼  ð35Þ
b0  d kaf 1
where this factor simulates the influence of the most important
steel fibre reinforcement mechanism, namely: fibre volume per-
centage (Vf), fibre aspect ratio (lf/df) and fibre–matrix bond strength
(sb). The a1 factor aims to simulate the degree of influence of these
fibre characteristics on the punching resistance of SFRC slab. In or-
der to integrate these new aspects in the Critical Shear Crack Theory
a
(CSCT) proposed by Muttoni [12], the kf 1 was included in the equa-
tion of the CSCT, resulting:
Vu
a ¼ mc  f ðw; d0 Þ ð36Þ
b0  d  kf 1
where f(w, d0) is a function dependent on crack width, w, and on the
roughness of the punching failure surface that was assumed corre-
lated to the diameter of the aggregates, d0. If the proposal of Mut-
toni and Schwartz [17] for the determination of w from the
rotation of the slab (wd) is now considered, yields:
Vu mc
a ¼ ð37Þ
b0  d  kf 1 a2 þ a3  w  d
According to Muttoni [12], and in agreement with the results
from Walraven [24], and Vecchio and Collins [25], the contribution
of the aggregate interlock for the concrete shear resistance can be
estimated by multiplying wd by kdg, where kdg = 1/(dg0 + dg), being
dg0 = 16 mm the reference diameter, and dg the maximum diame-
ter of the aggregates. In concrete of lightweight aggregates it is as-
sumed dg = 0, since the fracture progresses through the aggregates
resulting a relatively smooth surface. Therefore, the following
equation governs the punching failure criterion of SFRC slabs:
Vu mc
a ¼ ð38Þ
b0  d  kf 1 a2 þ a3  w  d  kdg
being mc the concrete nominal shear strength [26], which will be
determined in the next section, as well as the ai (i = 1–3) and kf
parameters.
3.2. Proposal
The values of ai parameters that define Eq. (38) were deter-
mined by fitting as much as possible the Vu–w response recorded
in experimental punching tests with RC slabs. According to ACI
318 [26], the parameter b0, which defines the perimeter of the
α⋅d α⋅d
2⋅rc e
(a) (b)
Fig. 7. Punching critical perimeter adopted for the evaluation of the punching
resistance in column of: (a) circular and (b) square cross section.
218 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223

Table 1 [27], where a penalty (PEN) is assigned to each range of k parame-

Modified version of the Demerit Points Classification (DPC) [27]. ter according to Table 1, and the total of penalties determines the
k = mexp/mthe Classification Penalty (PEN) performance of the proposal. The theoretical results were obtained
<0.50 Extremely dangerous 10 from the following equation:
[0.50–0.85] Dangerous 5
1
[0.85–1.15] Appropriate safety 0 tthe ¼ ð43Þ
[1.15–2.00] Conservative 1 1:33 þ 1:82  w  d  kdg
P2.00 Extremely conservative 2
while the experimental results were determined from the following
equation:

punching failure surface, is localized at a distance d/2 from the Vu

texp ¼ pffiffiffiffi   ð44Þ
external surface of the column. This parameter and mc are obtained l 1=3
b0  d  fc  V f  df
from the following equations: f

b0 ¼ 4  e þ p  d ðfor column of square cross sectionÞ ð39Þ
pffiffiffiffi
mc ¼ 0:33  fc ð40Þ
being e the edge of the column’s cross section. For the reasons indi-
cated at the end of the previous chapter, the kf factor was assumed
dependent on the Vf and lf/df fibre parameters:
lf
kf ¼ V f  ð41Þ
df
To determine the ai parameters, the value of a1 was varied
between 0 and 1 with small increments of a1, and for each a1,
the values of a2 and a3 that best fit the experimental results
(smallest R2) were obtained. A preliminary parametric study in
this context has indicated that values of a1 outside of the afore-
mentioned interval conduct to worst predictions of the experi-
mental results. The following optimized ai parameters were
those corresponding to the smallest R2 of the performed analy-
sis: a1 = 1/3, a2 = 1.33, a3 = 1.82. Replacing these values in Eq.
(38) results:
Vu 1 pffiffiffiffi 1=3 ¼
 ¼ ð42Þ
pffiffiffiffi  lf 1=3 1:33 þ 1:82  w  d  kdg
b0  d  fc  V f  d
f > b0 ¼ 4  e þ p  d ðfor column of square cross sectionÞ
3.3. Assessment of the predictive performance of the proposal
In this section the predictive performance of Eq. (42), as repre-
sentative of the punching failure criterion of SFRC flat slabs, is as-
sessed. For this purpose, the k = mexp/mthe parameter, comparing
the experimental (mexp) and the theoretical (mthe) values, is deter-
mined. The values of k are evaluated according to a modified ver-
sion of the Demerit Points Classification (DPC) proposed by Collins
Fig. 8a shows the predictive performance of the proposal,
while Figs. 8b and c present the safe (k P 1) and unsafe (k < 1)
results (in percentage) and a ‘‘box and whiskers’’ plot of the k
parameter, respectively. The box plot diagram graphically depicts
the statistical five-number summary, consisting of the minimum
and maximum values, and the lower (Q1), median (Q2) and
upper (Q3) quartiles. The obtained results are included in Table 2,
where the number of samples, the penalty for each range of k
parameter, and the total of penalties (Total PEN) are indicated,
as well as the average (AVG), the standard deviation (STD) and
the coefficient of variation (COV). The number of samples was
limited to 40 because, in the majority of the slabs composing
the DB the ultimate deflection was not provided, whose value
if fundamental for the evaluation of the ultimate rotation of
the slab to be introduced in Eq. (42). Fig. 8 shows that R2 of
the mexp–mthe is quite small. However, Table 2 reveals that the
average of k is close to the unit value, and its COV is relatively
low. Therefore, the assumptions subjacent to this proposal for
representing the punching failure criterion of SFRC flat slabs
are acceptable:
Vu 1
b0  d  fc  kf 1:33 þ 1:82  w  d  kdg
8
>
>
>
< k ¼ V  lf
f f d
f
ð45Þ
>
> k ¼ 1
>
> dg dg0 þdg
:
This equation, in spite of describing satisfactorily the punching
failure criterion for SFRC flat slabs, has the inconvenient of do not
being generalized for the cases of plain concrete (Vf = 0). To over-
come this deficiency, the parameters kf and lf are introduced into
Eq. (45) resulting:

100%

1.2
1.5
80%
1 57.50
60%
νexp (√MPa)

0.8
λ

0.6
40%
0.4 0.5
20% 42.50
0.2 νexp = 1.04νthe
2
R = 0.28
0 0% 0
0 0.2 0.4 0.6 0.8 1 1.2

νthe (√MPa) λ<1 λ≥1

(a) (b) (c)

Fig. 8. Analysis of the results: (a) predictive performance, (b) safe (k P 1) and unsafe (k < 1) percentage of slabs, and (c) dispersion of the results.
 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223 219

Table 2 Vu 1
Prediction of mexp: classification of the proposals according to the modified version of pffiffiffiffi h i¼ ð46Þ
b0  d 
1=3
f c  kf þ kf 1:33 þ 20  lf  w  d  kdg
the DPC.

k = mexp/mthe N° samples PEN where

<0.50 0 0 
[0.50–0.85] 3 15
0 if V f – 0
kf ¼ ð47Þ
[0.85–1.15] 28 0 1 if V f ¼ 0
[1.15–2.00] 9 9
P2.00 0 0 
1=11 if V f –0
Total PEN 40 24 lf ¼ ð48Þ
1 if V f ¼ 0
Statistical resume
Average (AVG) 1.04 When Vf = 0, Eq. (46) becomes the one proposed by Muttoni
STD 0.17 [12] for RC slabs:
COV (%) 15.97
Vu 1 3=4
pffiffiffiffi ¼ ¼ ð49Þ
b0  d  fc 1:33 þ 20  w  d  kdg 1 þ 15  w  d  kdg
2
Taking experimental results found in the bibliography, Fig. 9
evidences that Eq. (46) assures a satisfactory predictive perfor-
1.6
mance for the punching failure criterion. In this figure the m symbol
of the ordinate axis represents the left term of Eq. (46).
ν (√MPa)

1.2

0.8 4. Appraisal of the predictive performance of the proposals

0.4 4.1. Database (DB)

0 A database (DB) composed by 154 slabs submitted to punching

0 0.1 0.2 0.3 0.4 0.5 0.6
test configuration was built, 137 of them were reinforced with lon-
ψ· d·kdg (rad) gitudinal steel bars/grids in order to avoid the occurrence of flex-
ural failure mode. None of these slabs has conventional shear/
Fig. 9. Assessment of the predictive performance of the proposed punching failure
criterion (Eq. (46)). [21] [22] [3] [33] [34] [28] [35] [36] [31] punching reinforcement. However, 105 slabs composing the DB
were made by SFRC. In terms of concrete average compressive
strength, fcm, the DB is composed of slabs with fcm in the range of
14–93 MPa, so a quite high interval exists for a parameter that

1.2 1 0.8

0.8
0.9 0.6
ν (√MPa)

ν (√MPa)

ν (√MPa)

0.6
0.6 0.4
0.4

0.3 0.2
0.2

0 0 0
0 0.01 0.02 0.03 0.04 0 0.01 0.02 0.03 0.04 0 0.02 0.04 0.06 0.08
ψ (rad) ψ (rad) ψ (rad)

Exp Refined Simple Balanced Exp Refined Simple Balanced Exp Refined Simple Balanced

(a) (b) (c)

Fig. 10. Experimental versus theoretical results in terms of the relationship between normalized load and slab’s rotation for slab: (a) L6 [22], (b) A1 [3], and (c) P11F31 [28].

2
800 800 800
Vexp = 1.03Vthe Vexp = 0.98Vthe Vexp = 0.99Vthe
2
R = 0.87 2
R = 0.82
2
R = 0.83
1.5
600 600 600
Vexp (kN)

Vexp (kN)
Vexp (kN)

400 400 400 1

200 200 200 0.5

0 0 0 0
0 200 400 600 800 0 20 0 400 600 800 0 200 400 600 800
e

ed
ed

pl

nc
fin

m
Si

la
Re

Ba

Vthe (kN) Vthe (kN) Vthe (kN)

(a) (b) (c) (d)

Fig. 11. Performance in terms of predicting the punching failure load of the proposals: (a) refined, (b) simple, (c) balanced, and (d) dispersion of the predictions.
220 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223

Table 3 can have a detrimental consequence on this study. Furthermore,

Prediction of Vexp: classification of the proposals according to the modified version of the slabs where the concrete compressive strength has decreased
the DPC.
more than 15% in consequence of the addition of fibres were also
Proposals Refined Simple Balanced neglected, since this decrease reveals that the SFRC mix composi-
k = Vexp/Vthe N° samples PEN N° samples PEN N° samples PEN tion was not properly designed.
<0.50 0 0 0 0 0 0
[0.50–0.85] 2 10 5 25 4 20 4.2. General statistical analysis procedures
[0.85–1.15] 37 0 38 0 39 0
[1.15–2.00] 11 11 7 7 7 7
P2.00 0 0 0 0 0 0 The performance of the proposals for the prediction of the
Total PEN 50 21 50 32 50 27
punching resistance of SFRC slabs is appraised using the collected
data registered in the DB. For each proposal, the obtained values
Proposal Refined Simple Balanced
of Vthe are compared with Vexp, and a k factor corresponding to
Statistical resume the Vexp/Vthe ratio is evaluated. The values of k were classified
Average (AVG) 1.04 1.00 1.01
according to the modified version of the DPC (Table 1).
STD 0.13 0.13 0.13
COV (%) 12.72 13.22 12.74
4.3. Results

has a relevant impact on the punching resistance of concrete slabs.
For the slabs that were flexurally reinforced with steel bars, the
internal arm of this reinforcement (d, Fig. 1) has varied from
13 mm to 180 mm, while the reinforcement ratio (q) is in the
interval 0.4–2.75%. In the SFRC slabs, ‘‘hooked’’, ‘‘twisted’’,
‘‘crimped’’, ‘‘corrugated’’, ‘‘paddle’’ and ‘‘Japanese’’ type of fibres were
used, with an aspect-ratio that varied from 20 to 100, and in a vol-
ume percentage 62%. In some of the SFRC slabs (six specimens),
the SFRC was only applied in a region around the loaded area (that
represents the position of the column), considered the region
where punching failure could occur. In terms of loading conditions,
all the slabs of the DB were submitted to a load distributed in a cer-
tain area of the slab without transferring any bending moments
from the loading device to the slab. In the tests of the DB, the col-
umns were simulated by a RC element monolithically connected to
the slab, or applying steel plates, or even introducing a semi-spher-
ical device in between the piston of the actuator and the tested
slab. The cross section of the columns and steel plates was square
or circular. To avoid results that can compromise the reliability of
this statistical analysis, the slabs with a thickness lower than
80 mm were discarded, since an eventual influence of size effect
The performance of the proposals corresponding to Eqs. (10),
(29), and (30) for the prediction of the load versus rotation, with
Eq. (46) that represents the failure criterion, is presented in
Fig. 10, where experimental and theoretical results are compared.
In this figure and in the following analyses, the proposal corre-
sponding to Eqs. (10), (29), and (30) is designated as Refined, Simple
and Balanced, respectively. The m symbol of the ordinate axis repre-
sents the left term of Eq. (46). Fig. 10 shows that the three ap-
proaches have satisfactory predictive performance in terms of
load versus rotation, and when conjugated with the proposed fail-
ure criteria can estimate with good accuracy the punching failure
load. The performance of the three proposals for the prediction of
the punching failure load of the slabs composing the DB is assessed
in Figs. 11a and c, by comparing Vexp and Vthe. Fig. 11d shows the
minimum and maximum values, and the lower (Q1), median
(Q2) and upper (Q3) quartiles of the k = Vexp/Vthe values predicted
by the three proposals. Table 3 includes the analysis of k according
to the modified version of DPC (see also Table 1). Fig. 11 and Table 3
reveal that the proposals corresponding to Eqs. (10), (29), and (30)
with the failure criteria represented by Eq. (46) predict with good
accuracy the results registered experimentally, with a relatively

Table 4
Performance of several models to predict Vexp: classification of the models according to the modified version of the DPC.

k = Vexp/Vthe <0.50 [0.50–0.85] [0.85–1.15] [1.15–2.00] P2.00 Total PEN Average (AVG) STD COV (%)
MOD1 N° samples 0 21 21 8 0 50 0.92 0.23 25.29
PEN 0 105 0 8 0 113
MOD2 N° samples 0 2 18 29 1 50 1.24 0.26 20.89
PEN 0 10 0 29 2 41
MOD3 N° samples 0 5 18 20 7 50 1.42 0.62 43.38
PEN 0 25 0 20 14 59
MOD4 N° samples 0 0 8 42 0 50 1.32 0.20 15.47
PEN 0 0 0 42 0 42
MOD5 N° samples 0 6 17 27 0 50 1.20 0.29 24.03
PEN 0 30 0 27 0 57
MOD6 N° samples 0 6 37 7 0 50 0.99 0.13 13.26
PEN 0 30 0 7 0 37
MOD7 N° samples 0 20 24 6 0 50 0.92 0.18 19.45
PEN 0 100 0 6 0 106
Refined N° samples 0 2 37 11 0 50 1.04 0.13 12.72
PEN 0 10 0 11 0 21
Simple N° samples 0 5 38 7 0 50 1.00 0.13 13.22
PEN 0 25 0 7 0 32
Balanced N° samples 0 4 39 7 0 50 1.01 0.13 12.74
PEN 0 20 0 7 0 27

MOD1 = Narayanan and Darwish [8]; MOD2 = Shaaban and Gesund [29]; MOD3 = Harajli et al. [9]; MOD4 = Holanda [22]; MOD5 = Choi et al. [30]; MOD6 = Muttoni and Ruiz
[10]; MOD7 = Higashiyama et al. [31].
 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223 221

800 800 800 800

Vexp = 0.93Vthe
2
R = 0.45
600 600 600 600
Vexp (kN)

Vexp (kN)

Vexp (kN)

Vexp (kN)
400 400 400 400

200 200 Vexp = 1.19Vthe 200 Vexp = 1.21Vthe 200 Vexp = 1.31Vthe
2 2 2
R = 0.60 R = 0.04 R = 0.79
0 0 0 0
0 200 400 600 800 0 200 400 600 800 0 200 400 600 800 0 200 400 600 800
Vthe (kN) Vthe (kN) Vthe (kN) Vthe (kN)

(a.1) MOD1 (a.2) MOD2 (a.3) MOD3 (a.4) MOD4

800 800 800 800

Vexp = 0.98Vthe Vexp = 1.03Vthe
2 2
600 R = 0.86 R = 0.87
600 600 600
Vexp (kN)

Vexp (kN)

Vexp (kN)

Vexp (kN)
400 400 400 400

200 Vexp = 1.10Vthe 200 200 Vexp = 0.94Vthe 200

2 2
R = 0.64 R = 0.68
0 0 0 0
0 200 400 600 800 0 200 400 600 800 0 200 400 600 800 0 200 400 600 800
Vthe (kN) Vthe (kN) Vthe (kN) Vthe (kN)
(a.5) MOD5 (a.6) MOD6 (a.7) MOD7 (a.8) Refined

4
800 800
Vexp = 0.98Vthe Vexp = 0.99Vthe
600
2
R = 0.82 600
2
R = 0.83 3
Vexp (kN)

Vexp (kN)

2
λ

400 400

200 200 1

0 0 0
0 200 400 600 800 0 200 400 600 800

ed

ed
e
1

7
3
D2

D5

pl
D

D
D

in

c
Vthe (kN) Vthe (kN)

lan
O

O
O

f
Re

Si
M

Ba
(a.9) Simple (a.10) Balanced (b)
Fig. 12. Performance of several models to predict Vexp: MOD1 = Narayanan and Darwish [8]; MOD2 = Shaaban and Gesund [29]; MOD3 = Harajli et al. [9]; MOD4 = Holanda
[22]; MOD5 = Choi et al. [30]; MOD6 = Muttoni and Ruiz [10]; MOD7 = Higashiyama et al. [31]; Refined; Simple; Balanced.

small dispersion of results (the COV varied between 12% and 13%), (37 + 11), to which correspond the lowest penalty values, being
being the Refined model the one that conducts to the higher R2, consequently the more reliable model.
smallest number of total penalties, and largest number of samples
with proper safety margin.
6. Conclusions

In the present work three proposals were described for the pre-
5. Comparison to the predictive performance of other models diction of the punching failure of steel fibre reinforced concrete
(SFRC) flat slabs submitted to centrically loading conditions. The
In this section the predictive performance of the developed proposals are based on the critical shear crack theory, and only di-
model is compared to the one of the formulations proposed by: verge on the level of sophistication adopted to define the load ver-
Narayanan and Darwish [8], Shaaban and Gesund [29], Harajli sus rotation for the slabs. The punching failure criterion adopted in
et al. [9], Holanda [22], Choi et al. [30], Muttoni and Ruiz [10] these proposals is based on the experimental results collected in a
and Higashiyama et al. [31]. A detailed description of these formu- database and on the recommendations of ACI 318. The database is
lations can be found elsewhere [32], but a resume is in the appen- composed of 154 experimental punching tests with flat slabs. This
dix. Like in the previous sections, the predictive performance is database was also used to appraise the predictive performance of
appraised in terms of the k = Vexp/Vthe factor and considering the developed proposals, by performing a statistical analysis of the
modified version of DPC (Table 1). The obtained results are indi- k = Vexp/Vthe, where Vexp and Vthe is the punching failure load regis-
cated in Table 4 and represented in Fig. 12. Each of the models in tered experimentally and predicted analytically, respectively. The
comparison is designated by MODi (i = 1–7) and its corresponding predictive performance of the model was also assessed by applying
reference is indicated in the footnote of Table 4 and in the caption a modified version of the Demerit Points Classification. All the
of Fig. 12. From the obtained results it can be concluded that the three versions of the proposal predict satisfactorily the load versus
model proposed in the present work, regardless of the three levels rotation of the slabs of the database, and estimates k values in the
sophistication, together with the model of Muttoni and Ruiz [10], interval 1.0–1.04 with a coefficient of variation less than 13%,
are those that assure the highest performance for the prediction which is relatively small considering the complexity of the punch-
of the punching failure load of SFRC flat slabs. In terms of the mod- ing phenomenon. By comparing the performance of the developed
ified version of DPC, Table 4 shows that the proposed Refined model proposal to the another seven models for the prediction of the Vexp,
provides the lowest total penalties, with the highest number of it was also verified that the more refined version of the proposed
predictions of k in the intervals [0.85–1.15[ and [1.15–2.0] model assured the highest predictive performance with the largest
222 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223

number of predictions in the intervals considered proper for safety fcf0 ¼ 1:9  V f  ðlf =df Þ  b þ fc ½MPa ð12Þ
design.
ð2  e1 þ 2  e2 þ 4  cot u  dÞ  ðd  cu Þ
AT ¼ ð13Þ
Acknowledgements sin u

The study presented in this paper is a part of the research pro- AC ¼ ð2  e1 þ 2  e2 þ 4  cot u  cu Þ  cu ð14Þ
ject titled ‘‘SlabSys-HFRC – Flat slabs for multi-storey buildings using
8
hybrid reinforced self-compacting concrete: an innovative structural <1
> for hooked or crimped steel fibres
system’’, with reference number of PTDC/ECM/120394/2010. The b ¼ 2=3 for plain or round steel fibres with normal concrete
first author acknowledges the support provided by the CAPES >
:
3=4 for plain or round steel fibres with lightweight concrete
and CNPq grant, and the grant provided by the project SlabSys.
where fcf0 is the compressive strength of FRC, fc is the compressive
Appendix A:. Theoretical models strength of an equivalent concrete without fibres, b accounts for
the effect of fibre shape, u = 30° is the average value of the angle
1. Narayanan and Darwish [8] formed by the punching failure crack with the slab’s plan, aecof is

lf the compressive strain at the extreme compression fibre of the
V R ¼ 0:24  fsp þ 16  q þ 0:41  sb  V f  af   ns  ub cross section, and ecof is the compressive strain corresponding to
df
the compressive strength of FRC, cu is the neutral axis position, e1
 d ½MPa; mm ð1Þ
and e2 are the edges of the column’s cross section and AT and AC
 are, respectively, the failure surfaces of tension and compression
lf zones.
ub ¼ 1  0:55  V f  af   ð4  e þ 3  p  hÞ ½mm ð2Þ
df
8 6. Muttoni and Ruiz [10]
< 0:5
> for round fibres
V R ¼ V R;c þ V R;f ð15Þ
af ¼ 0:75 for crimped fibres ð3Þ
>
: 
1:0 for duoform fibres wd
V R;f ¼ rtf ðwÞ  Ap ¼ rtf  Ap ð16Þ
where fsp is the indirect cylinder tensile strength of fibre reinforced 6
concrete (FRC), sb = 4.15 MPa is the average fibre–matrix interfacial   2
bond stress, af is a factor depending of the fibre geometry, 1 w 2w lf
rtf ðwÞ ¼  arctan  1   V f  sb ð17Þ
ns ¼ ð1:6  0:002  hÞ is an empirical factor depending of the slab’s p a 1  lf lf df
thickness, h, and ub is the critical perimeter.
df
2. Shaaban and Gesund [29]
a1 ¼ ð18Þ
3:5  lf
pffiffiffiffi
V R ¼ 0:6  ð0:025  W f þ 0:567Þ  b0  d  fc ½MPa; mm ð4Þ 8
>
> 0:8  fc0:5 for hooked ends fibres
>
< 0:6  f 0:5
b0 ¼ 4  ðe þ dÞ ð5Þ c for crimped fibres
sb ¼ ð19Þ
> 0:4  fc0:5
> for straight fibres
>
:
7850  V f 7850  V f
Wf ¼ ¼ ¼ 3:27  V f ð6Þ
wc 2400
where VR,c is the concrete contribution to the punching
where Wf is the weight fibre percentage, and wc is the specific shear resistance, rtf is the post-cracking tensile strength of FRC
weight of plain concrete. for a crack width w, and Ap is the horizontal projection of failure
surface.
3. Harajli et al. [9]
pffiffiffiffi 7. Higashiyama et al. [31]
V R;f ¼ ð0:033 þ 0:075  V f Þ  b0  d  fc ½MPa; mm ð7Þ
V R ¼ bd  bp  br  ðfpcd þ mb Þ  up  d ð20Þ
4. Holanda [22]
pffiffiffiffi
V R ¼ 0:0035  d fpcd ¼ 0:2  fc < 1:2 MPa ð21Þ
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

q  fsy q  fsy pffiffiffiffi
 e  d  fc   170   ð0:15  V f þ 0:51Þ  fc ½MPa; cm;% rffiffiffiffiffiffiffiffiffiffiffiffi
fc fc 4 1000
bd ¼ < 1:5 ð22Þ
ð8Þ d
5. Choi et al. [30] p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
bp ¼ 3
100  q < 1:5 ð23Þ
rffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4 400
pffiffiffiffi
pffiffiffiffi  a2

1
VR ¼  0:263  fc  0:263  fc þ a   fcf0  AC br ¼ 1 þ ð24Þ
d 3 1 þ 0:25  ðu=dÞ
pffiffiffiffi
þ 0:105  ðlf =df Þ  fc  b  AT  cos u ð9Þ
mb ¼ 0:41  sb  F with sb ¼ 4:15 MPa ð25Þ
a  ecof ¼ 0:00196 ð10Þ
up ¼ ðu þ p  dÞ  ð1  K  FÞ with K ¼ 0:32 ð26Þ
 !
lf fcf0 lf
ecof ¼ 0:00079  V f  þ 0:0041  ð11Þ F¼  V f  af ; with V f in percentage ð27Þ
df fc df
 B.N. Moraes Neto et al. / Construction and Building Materials 46 (2013) 211–223
8
> 0:5 for round fibres
>
> Department of Civil Engineering, Structural Engineering report no. 174.
< 0:75 for crimped fibres
af ¼ ð28Þ
> 1:0
> for duoform fibres
>
: shear reinforcement. IABSE Colloq 1991;62:703–8. Zurich, Switzerland.
where u is the perimeter of the loaded area, up is the perimeter of
the critical section located at a distance of d/2 from the contour of
the loaded area.
References
[1] Tan K-H, Paramasivam P. Punching shear strength of steel fiber reinforced
concrete slabs. J Mater Civ Eng 1994;6(2):240–53.
[2] Prisco M, Felicetti R. Some results on punching shear in plain and fibre-
reinforced micro-concrete slabs. Mag Concr Res 1997;49(180):201–19.
[3] Nguyen-Minh L, Rovňák M, Tran-quoc T, Nguyen-Kim K. Punching shear
resistance of steel fiber reinforced concrete flat slabs. Punching shear
resistance of steel fiber reinforced concrete flat slabs. In: Proceedings of the
twelfth east Asia–Pacific conference on structural engineering and
construction (EASEC-12). Hong Kong Special Administrative Region, China
(with CD-rom); 24–26th January 2011. p. 535–6.
[4] Destrée X. Free suspended elevated slabs of steel fibre reinforced concrete: full
scale test results and design. In: Gettu R, editor. 7th International symposium
of fiber-reinforced concrete: design and applications BEFIB 2008; September
2008. p. 941–50.
[5] Destrée X, Mandl J. Steel fibre only reinforced concrete in free suspended
elevated slabs: case studies, design assisted by testing route, comparison to the
latest SFRC standard documents. In: Walraven, Stoelhorst, editors. Tailor made
concrete structures. London: Taylor & Francis Group; 2008. p. 437–43.
[6] Barros JAO, Salehian H, Pires NMMA, Gonçalves DMF. Design and testing
elevated steel fibre reinforced self-compacting concrete slabs. BEFIB2012-fibre
reinforced concrete 2012.
[7] Sasani M, Sagiroglu S. Progressive collapse of reinforced concrete structures: a
multihazard perspective. ACI Struct J 2008;105(1):96–105.
[8] Narayanan R, Darwish IYS. Punching shear tests on steel fibre reinforced
microconcrete slabs. Mag Concr Res 1987;39(138):42–50.
[9] Harajli MH, Maalouf D, Khatib H. Effect of fibers on the punching shear
strength of slab–column connections. Cement Concr Compos
1995;17(2):161–70.
[10] Muttoni A, Ruiz MF. The critical shear crack theory as mechanical model for
punching shear design and its application to code provisions. Fédération
Internationale du Béton, Bulletin 2010;57:31–60. Lausanne, Switzerland.
[11] Michels J, Waldmann D, Maas S, Zürbes A. Steel fibers as only reinforcement
for flat slab construction – experimental investigation and design. Constr Build
Mater 2012;26:145–55.
[12] Muttoni A. Punching shear strength of reinforced concrete slabs without
transverse reinforcement. ACI Struct J 2008;105(4):440–50.
[13] CEB-FIP. Model code 2010: final draft. Model code prepared by Special Activity
Group 5. Lausanne; September 2011.
[14] Kinnunen S, Nylander H. Punching of concrete slabs without shear
reinforcement, vol. 158. Stockholm, Sweden: Transactions of the Royal
Institute of Technology; 1960. 112p.
[15] Shehata IAEM. Punching of prestressed and non-prestressed reinforced
concrete flat slabs. PhD thesis. London: The Polytechnic of Central London;
1985.
223
[16] Alexander SDB, Simmonds SH. Bond model for strength of slab–column joints.
University of Alberta; 1991.
[17] Muttoni A, Schwartz J. Behaviour of beams and punching in slabs without
[18] Moraes Neto BN. Punching behaviour of steel fibre reinforced concrete slabs
submitted to symmetric loading. PhD in Civil Engineering, Department of Civil
and Environmental Engineering, University of Brasília, Brasília, DF; January
2013 (in Portuguese).
[19] Guandalini S. Poinçonnement symétrique des dalles en béton armé. PhD thesis
3380. Ecole Polytechnique Fédérale de Lausanne, Switzerland; 2005 (in
French).
[20] Taheri M, Barros JAO, Salehian HR. A parametric study on the use of strain
softening/hardening FRC for RC elements failing in bending. ASCE Mater Civ
Eng J 2012;24(3):259–74.
[21] Azevedo AP. Resistance and ductility of column–slab connections in slabs of
high strength steel fibre reinforced concrete and including steel studs. MSc
thesis. EESC-USP, Brazil; 1999 (in Portuguese).
[22] Holanda KMA. Analysis of resistant mechanisms and similarities of the
addition effect of steel fibers on strength and ductility to both punching
shear of flat and the shear of concrete beams. PhD thesis. São Carlos, Brazil;
2002 (in Portuguese).
[23] Cunha VMCF, Barros JAO, Sena-Cruz JM. Pullout behaviour of steel fibres in
self-compacting concrete. ASCE Mater Civ Eng J 2010;22(1):1–9.
[24] Walraven JC. Fundamental analysis of aggregate interlock. J Struct Eng ASCE
1981;107(11):2245–70.
[25] Vecchio FJ, Collins MP. The modified compression-field theory for reinforced
concrete elements subjected to shear. ACI J Proc 1986;83(2):219–31.
[26] ACI 318. Building code requirements for structural concrete. Farmington Hills,
Michigan: American Concrete Institute; 2008.
[27] Collins MP. Evaluation of shear design procedures for concrete structures. A
report prepared for the CSA technical committee on reinforced concrete
design; 2001.
[28] Alexander S, Simmonds S. Punching shear tests of concrete slabs–column
joints containing fiber reinforcement. ACI Struct J 1992;89(4):425–32.
[29] Shaaban AM, Gesund H. Punching shear strength of steel fiber reinforced
concrete flat plates. ACI Struct J 1994;91(4):406–14.
[30] Choi K-K, Taha MMR, Park H-G, Maji AK. Punching shear strength of interior
concrete slab–column connections reinforced with steel fibers. Cement Concr
Compos 2007;29:409–20.
[31] Higashiyama H, Ota A, Mizukoshi M. Design equation for punching shear
capacity of SFRC slabs. Int J Concr Struct Mater 2011;5(1):35–42.
[32] Moraes-Neto BN, Barros JAO, Melo GSSA. The predictive performance of design
models for the punching resistance of SFRC slabs in inner column loading
conditions. In: 8th RILEM international symposium on fibre reinforced
concrete: challenges and opportunities (BEFIB 2012); September 2012.
[33] Cheng M-Y, Parra-Montesinos GJ. Evaluation of steel fiber reinforcement for
punching shear resistance in slab–column connections – Part I: monotonically
increased load. ACI Struct J 2010;107(1):101–9.
[34] Mcharg PJ. Effect of fibre-reinforced concrete on the performance of slab–
column specimens. Master thesis. Montreal-Canada: McGill University; 1997.
[35] Theodorakopoulos DD, Swamy N. Contribution of steel fibers to the strength
characteristics of lightweight concrete slab–column connections falling in
punching shear. ACI Struct J 1993;90(4):342–55.
[36] Özden S, Ersoy U, Ozturan T. Punching shear tests of normal- and high-
strength concrete flat plates. Can J Civ Eng 2006;33:1389–400.