Can fracture mechanics predict damage due disaster of structures

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
_______________________________________________________________________________________
Volume: 03 Special Issue: 16 | ICPECDM-2014 | Dec-2014, Available @ http://www.ijret.org 104
CAN FRACTURE MECHANICS PREDICT DAMAGE DUE DISASTER
OF STRUCTURES?
Nagesh H.E1
, G. Appa Rao2
1
PhD Scholar, Structural Division, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai-
600036, India
2
Professor, Structural Division, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai-
600036, India
Abstract
This paper discusses shortcomings of existing code provisions for design of reinforced concrete structures, and need for
application of fracture mechanics in design of such structures. For a structure to perform satisfactorily in case of disastrous
incidents, care must be exercised at the time of design of structure itself. Fracture mechanics approach for design of structures
ensures rational design maintaining safety with reasonable safety margin. Currently, several national and international codes of
practice do not recognize the size effect on strength and serviceability of structures. In the design of structures with concrete, the
size effect needs to be incorporated to avoid sudden failures. Generally laboratory based tests have limitation on size of members,
which are relatively very small compared to actual large structures whose strength cannot be extrapolated. The incorporation of
effect of size and other fracture parameters can yield improvement in the design of members with better prediction of safety
margin.
Keywords: Fracture mechanics, Size effect, Minimum reinforcement, Ductility, Shear strength.
-------------------------------------------------------------------***-------------------------------------------------------------------
1. HOW SIGNIFICANT SIZE EFFECT IN
CONCRETE STRUCTURES?
According to the strength of materials approach, large size
and small size beams fail at same stress levels, if the given
material properties. But in practical the large size beams fail
at lower stress levels than small size beams. In designer
point of view, the most common examples for size effect
are, in bending test of concrete beam the strength decreases
with increase in beam depth. Shear strength of geometrically
similar concrete beams also found to be decreasing as the
size increases. Size effect in shear is more predominant than
in flexure. Overall it is noticed from experiments that as the
size increases there will be transition from ductile to brittle
failure, if all other properties maintained same. This change
in material properties with volume is known as size effect.
The main causes for size effect is (i) Statistical due to
material strength randomness, according to Weibull
statistical theory i.e. as the size increases, volume also
increases along with percentage of voids also increases
resulting in strength reduction because of material
heterogeneity. (ii) Energetic due to energy release which
causes stress redistribution, when a large crack or a large
fracture process zone (FPZ) containing damaged material
develops before the maximum load is reached and
discontinuities in flow of stress in case of cracks.
Quasibrittle materials like concrete exhibit two types size
effects that can be studied according to cohesive fracture
mechanics or (NLEFM), The size effect in failures occurring
(i) at macro-crack initiation, which is typical of plain
concrete. (ii) After large stable crack growth, which is
typical in reinforced concrete, especially shear in beams.
The latter is due to energy release associated with stress
redistribution caused by a large crack.
Fig 1 Size Effect according to strength criteria, linear and
nonlinear fracture mechanics
According to strength theory, the failure criterion is
expressed in terms of stresses or strains which is calculated
according to theories of elasticity and plasticity. The failure
condition is indicated by the condition σn = ft, where ft is the
direct tensile strength of concrete. σn is the nominal stress at
0
1
2
3
4
5
6
7
8
9
0 2 4 6 8 10
Nominal
stress
at
failure,
σn
Log (Size)
STRENGTH
LEF
M
NLE
FM
Most
existing
tests 1
2

_______________________________________________________________________________________
failure characterizing the stress state at a certain critical
point in structure. For dimensionality purpose,
where, cn is the coefficient which is independent of
the size, shape and loading. P is the load, d is the
characteristic dimension and b is the thickness. The above
relationship between log (σn) vs. log (size) which is a non-
dimensional size parameter log (d/da), where‘d’ is the
characteristic dimension and ‘da’ is the maximum size of the
aggregate. It can be seen from Fig 1, according to the
strength criteria it is horizontal line, which is independent of
the size of member. In Linear Elastic Fracture Mechanics
(LEFM) failure criterion is expressed in terms of the energy
consumed per unit crack length increment and σn varies
inversely as √d. so the variation of log (σn) vs. log (size) is a
straight line with a slope of -1/2 as shown in Fig. 1. The true
size effect in concrete structures can be represented as a
gradual transition from the horizontal line to the inclined
straight line according to Non Linear Elastic Fracture
Mechanics (NLEFM).
2. PRESENT DESIGN BASIS
Presently all national and international codes of practice are
based on tests conducted on small size specimens at various
laboratories. The codes do not explain or limitations of its
application up to certain size and the same formulae are
being used to design large structures also without knowing
its limitations. The extrapolation should be well anchored in
a theory so that the practical results match well with the
extrapolated values.
Fig 2 Histogram of beam depth (number of test data vs. depth) Courtesy: Bazant and Yu (2005)
Figure 2 shows the histogram of the number of test data
versus beam depth, ‘d’ according to the database compiled
by subcommittee 445F of the American Concrete Institute
(ACI). The size effect is of practical concern mainly for
beam depths ranging from 1.0m to 15.0m. Unfortunately,
86% of all the available test data pertain to beam depths less
than 0.5 m. This highlights the importance of testing larger
size test specimens in laboratories rather than smaller size.
The extrapolation should be anchored to a theory in order to
ensure safe designs.
3. RESEARCH SIGNIFICANCE
Present design codes mainly based on strength and
serviceability aspects. Neglecting tensile strength of
concrete results in conservative designs. Failure of RC
beams according to the conventional strength or yield
criterion exhibit no size dependence, nor should the size of
the beam have any effect on its ductility. The effect of size
of member and ductility in the design of RC members can
be predicted by using fracture mechanics. The fracture
mechanics based research also shows that, there is a need to
test large size specimens in laboratories.
4. BACKGROUND
The first attempt recorded on use of fracture mechanics of
concrete was by Kaplan (1961) attempting the use of linear
elastic fracture mechanics (LEFM). It was confirmed that
the large size beams exhibit strong size effect on nominal
stress at failure, which is size independent on small beams
(Walsh, 1972). The first theoretical model for reinforced
concrete beam in flexure was by Carpinteri (1981, and
1984) adopted the principles of fracture mechanics to solve
the problem of a reinforced concrete beam with an initial
edge crack ‘a’ subjected to bending. According to this
0
50
100
150
200
0.0 0.1 0.3 0.4 0.5 0.6 0.8 0.9 1.0 1.1 1.3 1.4 1.5 1.7 1.8 1.9 2.0 2.2 2.3 2.4 2.5 2.7 2.8 2.9 3.0 3.2 3.3 3.4 3.6 3.7 3.8
Number
of
tests
Depth in (m)
Range of main practical interest for size effect
86% of Test Data

_______________________________________________________________________________________
elementary model, moment and steel forces applied
remotely from the crack faces are superposed. The
assumption was that concrete behaves as a linear elastic
material and steel as a linear elastic-perfectly plastic. After
that many modifications and refined models are now
available. Bazant in his series of works showed the
significance of size effect in designing civil engineering
structures. Bazant demonstrated in theoretical model that
size effect mainly caused by stress redistribution and
localization of cracking damage associated with the release
of energy stored in the structure. Many theoretical models
and applications of size effect also contributed by Bazant. It
includes crack band model (1983), Bazant size effect law
(1984) and nonlocal softening damage models.
5. SIZE EFFECT ON DUCTILITY
The performance of structures due to loading effects of
earthquake, storms and other natural disasters mainly
depends on ductility of the members. Because ductility
ensures ductile failure which prevents sudden or brittle
failure that leads to sudden collapse. The following are some
of the investigations related to ductility and size effects in
beams.
Hillerborg (1990) made theoretical investigation of
rotational capacity of RC beams by applying fracture
mechanics concepts. It has been concluded that the
rotational capacity is nearly inversely proportional to the
depth of beam. Bigaj and Walraven (1993) investigated size
effect on rotational capacity of beams. They reported that
the members with small dimensions show much higher
plastic rotations than members with larger dimensions of
same characteristics. Carpinteri et al. (2009) developed new
model for the analysis of size-scale effects on the ductility
of RC beams in bending. They concluded that behavior
tends to become more brittle as beam depth increases. This
behavior is very well evidenced by a progressive reduction
of the beam rotation at failure.
Rao et al. (2008) derived ductility number from principles
of dimensional analysis, which is function of size of beam
(h), fracture energy (GF), yield strength of steel (fy),
modulus of elasticity of concrete (Ec), area of steel
reinforcement (As) and gross-area of cross-section
(Ag) . They concluded that the ultimate
flexural strength and ductility of RC beam decreases with
increase in size. The ductility factor increases with increase
in size for certain limit and again it decreases. Carpinteri et
al. (2013) outlined conditions for the structural design of RC
elements exhibiting ductile response. The decrease in one
parameter among depth (h), steel ratio (ρ), and yield
stress , or the increase in concrete tensile properties
or fracture energy (GF), all the other parameters being kept
constant, determines a transition from ductile response to
unstable tensile crack propagation. On the other hand, the
increase in h, ρ, and , or the decrease in ultimate
compressive stress or (Gc) all other parameters being
kept constant, produces a transition towards crushing failure
without steel yielding. In an overview it is noticed that as
size increases brittleness of the structure also increases.
Hence this phenomenon should be accounted while
designing any structures. Underestimation of ductility in
large scale specimens leads to transition from ductile to
brittle behavior, which is not desirable and it has to be
accounted in codes.
Figure 3. shows the plot of plastic rotation (φPL) versus
relative neutral axis position (x/d) with comparison of
Eurocode2 provision and numerical results of Carpinteri and
Corrado (2010). From the observation of plot it is clear that,
Eurocode 2 underestimates the plastic rotational capacity of
small size beam (0.2m depth) and overestimates the
rotational capacity of large size beam (0.6m and 0.8m
depth).
Fig 3 Plastic rotation vs. relative neutral axis
0
10
20
30
40
50
60
70
0 0.1 0.2 0.3 0.4 0.5
φ
PL(mrad)
x/d
0.2 m
0.4 m
0.6 m
0.8 m
EC2

_______________________________________________________________________________________
6. FRACTURE MECHANICS APPLICATIONS
IN DESIGN
Safe design of structure requires a valid mechanical model
and probabilistic analysis. Fracture mechanics provides a
scientific approach to define the failure of concrete. The
application of fracture mechanics to the design practice is
possible through the size effect. The two main applications
are presented here.
6.1 Minimum Reinforcement
Problems that can be solved using principles of fracture
mechanics is the provision of minimum reinforcement
because lightly reinforced beams are fracture sensitive.
Currently, various national and international codes of
practice provide empirical formulae for evaluation of
minimum reinforcement. Even though they are not rational
but performed satisfactorily for many years in design. These
codes disregard the nonlinear contribution of concrete in
tension and the size effects. They normally incorporated
compressive strength of concrete and yield strength of steel
as main parameters in the empirical formulae. Provisions in
American Concrete Institute (ACI) and Australian Standards
(AS) codes consider the compressive strength of concrete,
but British Standards (BS) and Indian Standards (IS) codes
do not. Australian Standards (AS) and Norway Standards
(NS) code defines minimum flexural reinforcement as a
function of size of member where as other codes do not.
Conventional method of evaluating cracking moment using
modulus of rupture (which is not strictly a material property)
for evaluating minimum reinforcement is empirical in
nature. The behavior of concrete in tension, crack initiation,
and crack propagation and fracture process zone are well
described in current fracture mechanics of concrete
approach. Such provisions ensure sufficient ductility in
flexural members by providing a reasonable margin of
safety between the first cracking and ultimate flexural loads.
The use of fracture mechanics of concrete enables rational
procedure to evaluate the optimum steel reinforcement
which ensures safe designs incorporating the size effect, and
other influencing parameters.
6.2 Evaluation of Shear Strength in Deep Beams
It is observed from experimental results of Kani (1967) that
about 40% of the ultimate shear strength reduction was
observed when the beam depth was increased from 300mm
to 1200 mm in deep beams without stirrups. Rao and
Sundaresan (2012) proposed equation for predicting shear
strength in deep beams incorporating size (depth) of the
beam and other influencing parameters. Bazant and Yu
(2009) showed that an increase of beam depth from 0.3 m to
1 m raises the shear failure probability of beams without
stirrups from (10-6
) to (10-3
) per lifetime. Yu and Bazant
(2011) demonstrated from experimental database and from
numerical studies that size effect on shear strength cannot be
suppressed by the presence of minimum or heavier stirrups.
For beams less than 1m, stirrups can mitigate significantly
but not in case of beams more than 1m. The current standard
ACI 318-2008 and other codes specifies for RC beams with
shear reinforcement, whether minimum or heavier, a size
independent shear strength equation. Hence these code
provisions need to be verified and adjusted to size dependent
shear strength.
Other applications such as brittle failures of concrete
structures (shear and torsion of R.C. beams with or without
stirrups, plain concrete flexure, slab punching, column
failure, bar embedment length, splices, bearing strength,
etc.) can also be studied by using fracture mechanics
principles.
7. CASE STUDIES
7.1 Failure of Tricell of Sleipner Oil Platform
This case study is about the failure of Sleipner an Oil
Platform, Norway in 1991. The tricell is 190m tall, structure
imploded under water head of 67 m, causing the platform to
sink. Investigation team outlines factors caused failure. (i)
The wall failed as a result of a combination of poor finite
element mesh (ii) insufficient anchorage of the
reinforcement in a critical zone. A separate investigation
documented a third contributing factor, the size effect in the
shear failure shown in figure. 4. which reduced the shear
capacity by about 40%. Caissons and shear failure of tricell
are shown in the figure.4.

_______________________________________________________________________________________
Fig 4 Sleipner oil platform caisson cells and failure of tricell
7.2 Failure of Koror-Babeldaob (KB) Bridge
(a)
(b)
(c)
Fig 5 Koror-Babeldaob (KB) Bridge (a) Before collapse (b)
and (c) After collapse

_______________________________________________________________________________________
Koror-Babeldaob (KB) bridge was built in 1977 in the
Republic of Palau. Collapse of bridge was occurred in 26th
September 1996 and investigation showed that erroneous
prediction of deflection due to creep and shrinkage, and
inappropriate remedial prestressing was major cause for
collapse. Consideration of size effect due to brittleness of
concrete would have prevented the collapse of above bridge.
The low safety margin and severity of size effect on large
size members causes catastrophic collapse. Size effect factor
for normal strength concrete can have a maximum value of
2.0, the size effect alone would be sufficient to cause the
collapse even though a safety margin related to error in the
design formulae and material randomness are included.
It is also remarkable to note that size effect must have
contributed to other disasters (e.g. bridges in Northridge.
Loma Prieta and Kobe earthquakes. Schoharie Bridge,
Shelby AF Warehouse, St. Francis Dam).
8. CONCLUSION
Fracture mechanics based designs give a promising safe
design of reinforced concrete structures incorporating size
effect and fracture properties of concrete. In real structures
like bridges, size of members is relatively large, so
brittleness also is high. If this behavior is properly
accounted for in the design, probability of failure in case of
any natural disaster can be avoided. This reveals that
introduction of size effect in the specifications of concrete
design codes is of the utmost practical importance.
REFERENCES
[1] Bazant, Z.P. (1983) “Crack band theory for fracture
of concrete” Materials and structures, 16, 155-177.
[2] Bazant, Z.P. (1984) “Size effect in blunt fracture:
concrete, rock, metal” Journal of Engineering
Mechanics, ASCE. 518-535.
[3] Bazant, Z.P., and Yu, Q. (2004) “Size effect in
concrete specimens and structures: New problems
and progress” Fracture Mechanics of Concrete
Structures-5 (Proc., FraMCoS-5 Vail, Colo.), Vol. 1,
153-162.
[4] Bazant, Z.P., and Yu, Q (2005) “Designing against
size effect on shear strength of reinforced concrete
beams without stirrups: I. formulation” ASCE,
Journal of structural engineering. (131)1877-1885.
[5] Bazant, Z. P., Planas, J. (1998) “Fracture and size
effect in concrete and other quasibrittle materials”
CRC Press.
[6] Bazant, Z.P and Pang, S.D (2005) “Effect of size on
safety factors and strength of quasibrittle structures:
beckoning reform of reliability concepts” The
structural engineering convention (sec 2005) Indian
institute of science, Bangalore, India.
[7] Yu, Q and Bazant, Z.P (2011) “Can stirrups
suppress size effect on shear strength of RC beams?”
Journal of structural engineering ASCE, 607-617.
[8] Bosco, C., and Carpinteri, A. (1992) “Fracture
mechanics evaluation of minimum reinforcement in
concrete structures”, in Application of Fracture
Mechanics to Reinforced Concrete, Elsevier Applied
Science, London, 347-377.
[9] Carpinteri, A., Corrado, M., and Ventura, G.
(2013) “Failure mode scaling transitions in RC
beams in flexure: tensile, shearing, and crushing”,
FraMCoS-8, Spain.
[10] Cadamuro, E., Corrado, M., and Carpinteri, A.
(2010) “Size-scale effects on minimum flexural
reinforcement in RC beams: application of the
cohesive crack model” Fracture Mechanics of
Concrete and Concrete Structures - Recent Advances
in Fracture Mechanics of Concrete, Seoul.
[11] Carpinteri, A. (1984) “Stability of fracturing process
in RC beams”, Journal of Structural Engineering.
(ASCE), 110, 544-558.
[12] Carpinteri, A. (1999) “Minimum reinforcement in
concrete members”, ESIS-Technical Committee 9, 1-
202.
[13] Carpinteri, A., Corrado, M., Mancini, G., and
Paggi, M. (2009) “Size-scale effects on plastic
rotational capacity of reinforced concrete beams”
ACI structural journal, 106, S83.
[14] Carpinteri, A., Corrado, M., Paggi, M., and
Mancini, G. (2009) “New model for the analysis of
Size-scale effects on the ductility of reinforced
concrete Elements in Bending” Journal of
engineering mechanics, ASCE, 135, 221-229.
[15] Carpinteri, A. (1981) “A fracture mechanics model
for reinforced concrete collapse”, IABSE Colloquium
on Advanced Mechanics of Reinforced Concrete,
Delft, 17-30.
[16] CEB-FIP Model Code 2010, Vol 1, International
Federation for Structural Concrete (FIB),
Switzerland.
[17] El-Khatieb, W.M.H.M., and Al-Ani, H. (2011)
“Steel reinforcement ratio dependency of plastic
rotational capacity of reinforced concrete beams”
Jordan Journal, 5(4), 480-492.
[18] Hillerborg, A. (1990) “Fracture mechanics concepts
applied to moment capacity and rotational capacity of
reinforced concrete beams” Engineering Fracture
Mechanics, vol.35, No 1/2/3, 233-240.
[19] Kani, G.N.J (1967) “How safe are our large
reinforced concrete beams?” ACI J. Proc. 64(3): 128–
141
[20] Kaplan, M.F. (1961) “Crack Propagation and the
fracture of concrete.” ACI Journal 58 (11).
[21] Rao, G.A., Aravind, J., and Eligehausen, R. (2007)
“Evaluation of minimum flexural reinforcement in
RC beams using fictitious crack approach”, JoSE,
34(4), 277-283.
[22] Rao, G.A., Vijayanand, I., and Eligehausen, R.
(2008) “Studies on ductility and evaluation of
minimum flexural reinforcement in RC beams”
Materials and Structures, 41, 759–771.
[23] Rao, G.A., and Sundaresan (2012) “Evaluation of
size effect on shear strength of reinforced concrete
deep beams using refined strut-and-tie model”
Sadhana, Vol. 37, Part 1, pp. 89–105.
[24] Walsh, P F. (1972) “Fracture of plain concrete.”
International Concrete Journal, 46(11), 469– 476.

Can fracture mechanics predict damage due disaster of structures

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to Can fracture mechanics predict damage due disaster of structures (20)

More from eSAT Publishing House (20)

Recently uploaded (20)

Can fracture mechanics predict damage due disaster of structures