Theoretical and Applied Fracture Mechanics 71 (2014) 21–30

Contents lists available at ScienceDirect

Theoretical and Applied Fracture Mechanics

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

A comparison among some recent energy- and stress-based criteria

for the fracture assessment of sharp V-notched components under
Mode I loading
P. Lazzarin, A. Campagnolo, F. Berto ⇑
University of Padova, Department of Management and Engineering, Italy

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

Article history: Fracture assessment of components weakened by sharp V-notches is typically carried out by means of the
Available online 18 March 2014 Notch Stress Intensity Factors (NSIFs), which quantify the asymptotic linear elastic stress field in the
vicinity of the notch tip. Since units of the NSIFs do depend on the V-notch opening angle, a direct com-
Keywords: parison in terms of the NSIFs needs the constancy of the opening angle. A criterion based on the strain
Strain energy density energy density (SED) averaged over a material-dependent control volume surrounding the notch tip
Sharp V-notch has been proposed by the authors. When the control volume is small enough to make negligible the influ-
Brittle fracture
ence of higher order terms of Williams’ solution, the SED can be theoretically linked to the NSIFs.
Mode I loading
Notch Stress Intensity Factor
Considering Mode I loading conditions, the expression of the critical Mode I NSIF at failure as a function
of the V-notch opening angle as derived according to the averaged SED criterion is compared with those
given by two different versions of the Finite Fracture Mechanics (FFM) criterion, the former due to
Leguillon, the latter due to Carpinteri et alii.
Finally, the considered criteria are applied to components weakened by sharp V-notches under Mode I
loading conditions in order to investigate the predictive capability of each approach. A number of exper-
imental data taken from the literature are used for comparison.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Criteria able to assess the brittle failure of sharply notched com-
ponents, with the notch tip radius being zero or close to zero, have
A key issue in fracture mechanics is associated with the formu- received much interest in the last decades. To state a sufficiently
lation of a sufficiently simple and accurate criterion for crack simple and reliable condition to predict failure initiation in cases
nucleation and propagation in structural elements under static involving points of stress singularity remains a topic of active
and fatigue loading. For what concern structural elements weak- research. At such points local stresses tend to infinity under linear
ened by cracks, the fracture initiation is generally determined by elastic hypotheses so that, in analogy to the crack case, it is neces-
a critical value of the Mode I Stress Intensity Factor KI, associated sary to introduce a stress field parameter, the Notch Stress Inten-
with the opening crack mode loading: under plane strain condi- sity Factor (NSIF), which depends on the loading mode, the notch
tions the critical value to failure is the material fracture toughness opening angle and the geometry of the component. In the linear
KIc. This fracture criterion has been first formalized by Irwin [1]. For elastic Notch Mechanics, NSIFs are generally used to assess the
cracked elements subjected to Mode I loading, a well-known link static failure of brittle components weakened by sharp V-notches
exists between the Irwin criterion and the Griffith criterion [2]. [3–11]. Among these references it is notable the investigation
The latter assumes that failure occurs when the strain energy re- carried out by Leicester [3] who emphasized the scale effect in
lease rate G, defined as the derivative of the potential energy with wood structures with V-notches in the ambit of a project financed
respect to the crack surface area, reaches the critical value Gc, a by the Australian Forest Production Laboratory. Dealing with the
material dependent parameter, which represents the energy fatigue crack initiation conditions of components with V-notches
needed to create a unit crack surface area. other references are also available [12–15]. When fatigue life is
mainly consumed as microcrack initiation and propagation inside
the zone governed by the V-notch singularity, NSIFs can be used
⇑ Corresponding author. Tel.: +39 0444998754. also for total fatigue life assessments, as documented for welded
E-mail address: [email protected] (F. Berto). joints [16–18].

http://dx.doi.org/10.1016/j.tafmec.2014.03.001
0167-8442/Ó 2014 Elsevier Ltd. All rights reserved.
22 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30

For a better understanding of the V-notch problem, it is useful over a control volume is substantially insensitive to the mesh
to remember some stress-based criteria proposed in the last refinement [58,59] and can easily take into account also coupled
twenty years. Starting from a previous work due to Novozhilov three-dimensional effects [60,61]. Other advantages of the total
[19], Seweryn proposed a brittle fracture criterion based on the strain energy density have been underlined by other researchers.
assumption that crack nucleation or propagation occurs when In [62], extending previous analyses by the present authors
the mean value of normal stress over a specified segment d0 [58,59], the SED approach has been used in combination with
reaches a limit value for the material [7]. Afterwards, this criterion two separate volumes, located over and below the crack or the
has been extended and applied also to structural elements sub- V-notch bisector line, to compute the Stress Intensity Factors (SIFs)
jected to mixed mode loading [20,21] by introducing a non local and NSIFs for isotropic homogeneous and bi-material plates con-
failure function combining normal and shear stress components, taining cracks and notches under Mode I, II and III loading condi-
both normalized with respect to the relevant critical stresses of tions. In [63] a novel contribution of an energy-based method
the material. has been provided for simulating 3D crack evolution on the basis
A group of methodologies is represented by the Theory of of the first-order variation of energy release rate with respect to
Critical Distances (TCD) [22–24], according to which a characteris- crack advance.
tic material length parameter has to be used when performing In the framework of Finite Fracture Mechanics, the criteria by
fracture assessments on any kind of stress risers. The TCD assumes Leguillon [50,51] and Carpinteri et al. [52] require the simulta-
as a fracture parameter the normal stress calculated on a point at a neous verification of two separate conditions, the former based
specified distance from the notch tip (Point-Method) or the same on stresses, the latter on an energy balance. Each condition is nec-
stress averaged over a specified segment (Line-Method). essary but not sufficient to guarantee the fracture. When both con-
By proceeding on parallel tracks, critical distance and line ditions are simultaneously satisfied it is possible to get a sufficient
dimensions for notched components under fatigue limit conditions condition for fracture. The governing idea is that at failure a finite
were correlated in [25,26] to the El-Haddad parameter a0 [27]. The incremental crack (or a finite crack advance) occurs at the notch
idea of stress averaging is the basis of Neuber’s fictitious notch tip.
rounding (FNR) approach [28–30]. Recently, the influence of plane Leguillon [50,51] proposed a failure criterion for components
stress or plane strain conditions on the multiaxiality factor, s, was weakened by sharp V-notches based on a combination of the
highlighted in Refs. [31–34], as well as the essential role played by Griffith criterion (in the incremental form) and the stress criterion,
the V-notch opening angle. in order to determine the incremental crack length and the critical
Worth mentioning is also the Cohesive Zone Model (CZM) value of the Notch Stress Intensity Factor as a function of the mate-
sometimes called ‘fictitious crack model’. First proposed for con- rial properties and the V-notch opening angle. Also Carpinteri et al.
crete, the CZM was later successfully extended to a number of brit- [52], in analogy to Leguillon, proposed a fracture criterion based on
tle or quasi-brittle materials [35–38]. In these works both sharp a combination of the energy criterion (critical strain energy release
and blunt U and V-notches were considered, under different load- rate) and the averaged stress criterion, which enables to determine
ing conditions. the incremental crack length and the critical value of the Notch
Finally, concluding this short review, we mention some criteria Stress Intensity Factor. One should note that the two criteria are
based on the extension of J-integral to notches [39–41] and on the based on the same energy balance calculated considering a finite
modified Maximum Tangential Stress criterion [42–44] able to take incremental crack, the difference being in the stress calculations:
into account singular and non-singular stress terms of Williams’ the first involves a point-wise stress condition, the second consid-
solution. ers an average stress condition.
In the present contribution, two recent failure criteria involving After having briefly summarized the analytical frame in terms
energy-based calculations are discussed in detail with reference to of singular stress fields at V-notch tip, strain energy density aver-
sharp V-notches, that based on the averaged SED [45–49] and that aged in a finite size volume and strain energy release rate, the three
called Finite Fracture Mechanics criterion, according to two differ- published brittle fracture criteria are compared. The analytical
ent versions, the former due to Leguillon [50,51], the latter due to comparison is performed on the basis of the different expressions
Carpinteri et al. [52], as extension to V-notches of the crack case proposed for the critical Notch Stress Intensity Factor.
solution [53]. Extensions to in-plane mixed mode loading have Finally, the considered criteria are applied to components
been reported in the recent literature [54,55]. weakened by sharp V-notches under Mode I loading conditions
The averaged SED criterion was first proposed in [45] and is in order to investigate the predictive capability of each approach.
based on the idea that the critical parameter for the material is The comparison involves a number of experimental data taken
the mean value of strain energy density evaluated over a control from the literature and related to different materials.
volume surrounding the notch tip. The method is reminiscent of
Neuber’s concept of elementary structural volume [28–30] and
Sih’s criterion based on factor S, which gives the strain energy 2. Analytical frame
density multiplied by a convenient distance from the crack or the
V-notch tip [56,57]. It is obvious that any strain energy density ap- With the aim of clarifying the bases of the failure criteria ana-
proach cannot be used at the tip of a sharp V-notch since not only lysed in this work, this section summarizes the analytical frame
do stresses tend toward infinite but so does the strain energy den- giving the expressions of the singular stress fields at V-notch tip,
sity. On the contrary, in a small but finite volume of material close the SED averaged in the control volume and the strain energy re-
to the notch tip the energy always has a finite value and the main lease rate. Only Mode I loading conditions are taken into account,
question is rather that of estimating the size of this characteristic considering an isotropic and homogeneous material under linear
volume as a function of the material and loading conditions elastic conditions.
[48,49].
The major advantage of the averaged SED approach with re- 2.1. Singular stress fields at V-notch tip
spect to the local stress-based criteria is the mesh independency.
While some parameters integrated in the local criteria (e.g. maxi- In the presence of a sharp (zero notch tip radius) V-notch, the
mum principal stress, hydrostatic stress and deviatoric stress) re- stress distributions (Fig. 1a) due to a symmetric loading with
quire the accurate stress field determination, the SED averaged respect to the notch bisector (Mode I) are [64,65]:
 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30 23

2α

A=R c2⋅γ

notch
r
bisector
θθ σ
θ
σrr σ rθ
γ Rc

Fig. 1. Polar coordinate system centred at the notch tip (a) and control volume (area) of radius Rc surrounding the V-notch tip (b).

8 9
< rhh >
> = K  r k1 1 1
Finally, e1 is the parameter that summarizes the dependence from
rrr ¼ 1pffiffiffiffiffiffiffi the notch geometry:
>
: > 2p ð1 þ k1 Þ þ v1 ð1  k1 Þ
rrh ; Z þc  2 2 2
 
28 9 8 93 I1 ðcÞ ¼ r~ ð1Þ ~ ð1Þ þ r
hh þ rrr
~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ
zz  2m rhh rrr þ rhh rzz þ rrr rzz
< ð1 þ k1 Þ cosð1  k1 Þh >
> = < cosð1 þ k1 Þh >
> = c

6 7
 4 ð3  k1 Þ cosð1  k1 Þh þ v1 ð1  k1 Þ  cosð1 þ k1 Þh 5 þ 2ð1 þ mÞr
~ rh ð1Þ2
dh ð5Þ
>
: >
; >
: >
;
ð1  k1 Þ sinð1  k1 Þh sinð1 þ k1 Þh
8 ð1Þ 9
> rhh >
~ I1 ðcÞ
K 1  rk1 1 < ð1Þ = e1 ð2aÞ ¼ ð6Þ
¼ pffiffiffiffiffiffiffi r~ rr ð1Þ 4k1 c
2p > : ð1Þ > ;
r~ rh Values of e1 for plane strain conditions are given in Table 1 as a
function of the V-notch opening angle [48,49]. The averaged value
where the parameter K1 represents the Mode I Notch Stress Inten-
of the strain energy density needs the evaluation of a control radius
sity Factor (NSIF) according to Gross and Mendelson’s definition
value, which depends on the material, and varies under static and
[66], k1 is the Mode I Williams’ eigenvalue [64], v1 is an auxiliary
high cycle fatigue loading conditions [48,49].
parameter dependent on the notch opening angle 2a [65] while
r~ ð1Þ ~ ð1Þ ~ ð1Þ
hh rrr and rrh represent the angular stress functions for Mode I
2.3. Strain energy release rate
loading. Plane stress and plane strain conditions result in rzz = 0
and rzz = m(rhh + rrr), respectively.
Consider a loaded body initially in an equilibrium state: it is
characterized by a potential energy Wp and a kinetic energy
2.2. Strain energy density Wk = 0. After the nucleation of a new crack, or the propagation of
a pre-existing crack, there is a change in the potential and kinetic
According to Beltrami [67], the total strain energy density (SED) energy of the body. A balance between the internal and external
is equal to the total work done by the system and is given by Eq. work gives:
(2). For a three dimensional state of stress:
dW int ¼ dU þ dW k þ dW s ¼ dW ext
1 2 
) dðU  W ext Þ þ dW k þ dW s ¼ 0 ð7Þ
Wðr; hÞ ¼ r þ r222 þ r233  2mðr11 r22 þ r11 r33 þ r22 r33 Þ ð2Þ
2E 11
where r11, r22 and r33 are the principal stresses in a given reference dW p þ dW k þ Gc dS ¼ 0 ð8Þ
system.
where U is the elastic strain energy contained in the body, Wext
In the case of V-notched plate in Mode I, when only the contri-
is the work done by the external forces, and Wp and Wk are the
bution of the first singular term is significant, the SED can be di-
rectly linked to the NSIF, K1, by substituting into Eq. (2) the Table 1
singular stress field given by Eq. (1): Values of e1(2a, m) under plane strain conditions.

1 2ðk1 1Þ 2 h ð1Þ2 2 2

  2
i 2a (°) k e1 (m = 0.2) e1 (m = 0.3) e1 (m = 0.4)
W 1 ðr; hÞ ¼ r K1 r ~ ð1Þ
~ hh þ r ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ ~ ð1Þ þ 2ð1 þ mÞr
rr þ rzz  2m rhh rrr þ rhh rzz þ rrr rzz
~ ð1Þ
rh
2E 0 0.5000 0.1623 0.1345 0.1003
ð3Þ 15 0.5002 0.1687 0.1400 0.1046
20 0.5004 0.1706 0.1417 0.1061
The strain energy density, averaged in a circular sector of radius 30 0.5014 0.1739 0.1448 0.1091
Rc surrounding the notch tip (Fig. 1b), is given by the ratio between 40 0.5035 0.1764 0.1475 0.1119
the elastic strain energy E(Rc) and the area of the circular sector 45 0.5050 0.1772 0.1485 0.1131
60 0.5122 0.1780 0.1504 0.1162
A(Rc):
75 0.5247 0.1806 0.1498 0.1179
R R Rc R þc 80 0.5304 0.1739 0.1490 0.1179
EðRc Þ AR
W dA 0 c
W 1 ðr; hÞ rd rdh 90 0.5445 0.1693 0.1462 0.1174
W¼ ¼ ¼ R Rc R þc 100 0.5628 0.1631 0.1421 0.1158
AðRc Þ A
dA rd rdh 0 c 105 0.5739 0.1593 0.1394 0.1145
1 I1 ðcÞ 2 2k1 120 0.6157 0.1459 0.1296 0.1090
E 4k1
K 1 Rc 1
¼ ¼ e1 K 21 R2ðk
c
1 1Þ
ð4Þ 135 0.6736 0.1298 0.1172 0.1010
R2c c E 140 0.6972 0.1241 0.1127 0.0979
150 0.7520 0.1122 0.1030 0.0909
where I1 is the integral of the angular stress functions, which de- 160 0.8187 0.1002 0.0930 0.0834
170 0.9000 0.0882 0.0828 0.0758
pends on the notch opening angle, 2a = 2p  2c, and the Poisson’s
180 1.0000 0.0771 0.0730 0.0673
ratio m. I1 changes under plane stress or plane strain conditions.
24 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30

potential and kinetic energy, respectively. Ws represents indeed the The radius Rc can be easily determined by using a set of exper-
energy available for crack formation, which can be expressed as a imental data giving the critical value of the NSIF from a specific
function of the new crack surface dS and fracture energy per unit V-notch angle [45]. If the notch opening angle is zero, one can
surface Gc. use the fracture toughness KIc, as made by Yosibash et al. [46],
A condition for fracture can be derived from Eq. (8), being the and the final expression is:
kinetic energy always positive (dWk P 0):  2
ð1 þ mÞ  ð5  8mÞ K Ic
dW p dW p Rc ¼  ð15Þ
þ Gc 6 0 ) G ¼  P Gc ð9Þ 4p rc
dS dS
Substitution of Eq. (15) into Eq. (14) gives the NSIF at failure K1c
The inequality obtained can be taken as a fracture criterion, in
in the form:
which G represents the Strain Energy Release Rate (SERR). When
" ð1k1 Þ #
the equality is verified the change in kinetic energy is negligible
ð1 þ mÞ  ð5  8mÞ 1 2ð1k Þ
(dWk ? 0) and the crack propagation is stable; instead if G exceeds K 1c ¼  pffiffiffiffiffiffiffiffi  K Ic 1  r2k
c
1 1
ð16Þ
4p 2e1
the critical value Gc then the crack propagation is unstable because
the extra energy is transformed into kinetic energy of the crack
itself (dWk – 0).
3.2. Leguillon’s criterion
For two-dimensional problems a relationship between the
strain energy release rate (G) and the Mode I Stress Intensity Factor
Leguillon [50,51] proposed a fracture criterion for components
(KI) has been demonstrated by Irwin [1]:
8 weakened by sharp V-notches based on the Finite Fracture
< G ¼ K 2I plane stress Mechanics concept: at failure an incremental crack of length lc
E
ð10Þ nucleates at the notch tip.
: K 2I
G ¼ ð1  m2 Þ  E
plane strain According to Leguillon’s criterion, two necessary conditions can
be stated on stress and energy, but neither the one nor the other,
Eq. (9) represents the differential form of the Griffith fracture crite-
taken separately, are sufficient. Only when they are simultaneously
rion [2], obtained by considering an infinitesimal increment of the
satisfied it is possible to get a condition sufficient for fracture.
crack surface dS ? 0. Considering indeed a finite increment of the
According to the stress criterion the V-notched element failure
crack surface dS, Leguillon [50,51] obtained the incremental form
occurs when the singular stress component normal to the fracture
of the fracture criterion:
direction 
hc exceeds the material tensile strength rc at a distance lc
dW p dW p from the notch tip, that is at the tip of the incremental crack.
þ Gc 6 0 ) G ¼  P Gc ð11Þ
dS dS According to the energy criterion, instead, the component fail-
This is the governing idea of the Finite Fracture Mechanics ure occurs when the strain energy release rate G, that is the ratio
(FFM). between the potential energy variation in correspondence of the
crack nucleation (dWp) and the new crack surface created (dS),
exceeds the critical material value Gc.
3. Failure criteria for sharp V-notches
Therefore the simultaneous verification of the conditions for-
malized in Eqs. (17) and (18) provides a general criterion for the
In the literature the problem of the fracture assessment of com-
fracture of components weakened by sharp V-notches.
ponents weakened by V-notches is typically treated with stress-
based or energy-based approaches. In the present contribution, Stress criterion : rh ðlc ; hc Þ ¼ k1  lkc1 1  sh ðhc Þ P rc ð17Þ
three different criteria are considered, in order of publication they
are: the strain energy density criterion [45], the Leguillon’s crite- 2 2k
dW p k1  Kð2a; hc Þ  lc 1  d
rion [50,51] and the Carpinteri’s criterion [52]. Energy criterion : G¼ ¼ P Gc ð18Þ
dS lc d
3.1. Strain energy density (SED) criterion In Eqs. (17) and (18) lc is the length of the incremental crack nucle-
ated at the notch tip (see Fig. 2), k1 is the Mode I Williams’ eigen-
According to Lazzarin–Zambardi [45] the brittle fracture of the value tied to the V-notch angle 2a, sh is a function of the angular
material occurs when the average value of the strain energy den- coordinate  h, d is the thickness of the component. Finally, K(2a,
sity, calculated on a control volume of radius Rc surrounding the 
hc ) is a scale factor dependent on the local geometry (2a) and the
notch tip (Fig. 1b), is equal to the critical value Wc (Eq. (12)). direction of fracture (hc ).
According to Beltrami’s hypothesis, we have:

r2c
Wc ¼ ð12Þ
2E
where rc is the ultimate tensile strength and E the Young’s
modulus.
Under Mode I loading, by imposing W ¼ W c , and combining lc
Eqs. (4) and (12) one obtains:
2α
e1 K 21c r2
 2ð1k Þ
¼ c ð13Þ θc
E Rc 1 2E
r
The critical value of the Notch Stress Intensity Factor at failure,
θ
K1c, becomes:
1
K 1c ¼ pffiffiffiffiffiffiffiffi  rc  Rcð1k1 Þ ð14Þ
2e1 Fig. 2. Leguillon’s coordinate system, with incremental crack of length lc.
 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30 25

The parameter k1 represents the NSIF according to Leguillon’s Table 2

definition, which is linked to Gross and Mendelson’s definition Values of the function c(2a).

[66] by means of Eq. (19). 2a (°) c(2a) 2a (°) c(2a)

pffiffiffiffiffiffiffi 0 1.00 95 1.10
limr!0 rh  2p  r1k1 K1 5 1.00 100 1.11
k1 ¼ limrh  r1k1 ¼ pffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffi ð19Þ
r!0 2p 2p 10 1.00 105 1.12
15 1.00 110 1.14
In the following the lowercase letter will be used for the NSIFs 20 1.00 115 1.15
defined according to Leguillon, capital letter instead for the NSIFs 25 1.01 120 1.16
defined according to Gross and Mendelson. 30 1.01 125 1.17
35 1.01 130 1.18
The conditions (17) and (18) must be simultaneously satisfied.
40 1.02 135 1.19
Solving the system in two unknowns, lc and k1, it is possible to 45 1.02 140 1.20
determine the length of the incremental crack: 50 1.03 145 1.20
55 1.03 150 1.19
2
Gc  sh ðhc Þ 60 1.04 155 1.19
lc ¼  ð20Þ 65 1.04 160 1.17
Kð2a; hc Þ  r2c 70 1.05 165 1.14
75 1.06 170 1.11
As soon as the incremental crack length lc is introduced into Eq. 80 1.07 175 1.06
(17) or Eq. (18), it is possible to give the failure criterion in terms of 85 1.08 180 1.00
the classic Irwin’s expression, KI P KIc. Since the fracture direction 90 1.09
for a V-notched component in a homogeneous and isotropic mate-
rial subjected to a symmetric load (Mode 1) is known a priori,

hc ¼ p  a, the critical value of the Notch Stress Intensity Factor
k1c can be defined as a function of the material mechanical proper- material ultimate tensile strength rc, can be obtained substituting
ties (rc and Gc) and of the notch opening angle 2a, through K(2a) expressions (22) and (23) into Eq. (21). As a result:
and k1. Normalizing the angular function in such a way that
sh ð
hc Þ ¼ 1, one obtains: k1c ¼ cð2aÞ  kIc
2ð1k1 Þ
 r2k 1 1
ð24Þ
c
 1k1
Gc A slightly different formulation of the criterion can be obtained
k1 P  rc2k1 1 ¼ k1c ð21Þ
Kð2aÞ by using Gross and Mendelson’s definition for the critical NSIF K1c
and the fracture toughness KIc according to Eq. (19).
Under plane strain conditions Gc can be linked to the critical va-
lue kIc in the following form: 1 2ð1k1 Þ
K 1c ¼ ð2pÞðk1 2Þ  cð2aÞ  K Ic  r2k
c
1 1
ð25Þ
2
2 1m
Gc ¼ 2p  kIc  ð22Þ
E
3.3. Carpinteri’s formulation
The value of the scale factor K for the crack case (2a = 0, k1 = 0.5)
can be obtained from Eqs. (21) and (22) considering that k1c = kIc. In Similarly to Leguillon, Carpinteri et al. [52] proposed a fracture
order to simplify the involved expressions, the dependence of k1c criterion for components weakened by sharp V-notches based on
on the notch opening angle can be given defining a universal func- Finite Fracture Mechanics concept: at failure a crack of length
tion c(2a), calculated using an integration procedure described in DSE nucleates at the notch tip. Also in this formulation, it is possible
[11]. Doing so, the general expression for the scale factor K(2a) is: to obtain a sufficient condition for the fracture by satisfying simul-
taneously a stress criterion and an energy criterion.
1 1  m2 1
Kð2aÞ ¼ Kð2a ¼ 0Þ  1
¼ 2p   1
ð23Þ According to the averaged stress criterion the V-notched
cð2aÞ 1k1 E cð2aÞ1k1 element failure occurs when the notch singular stress component
normal to the crack faces, averaged over the crack length DSE,
The universal function c(2a) is plotted in Fig. 3 and its values
exceeds the material tensile strength rc.
listed in Table 2 [11].
According to the energy criterion, instead, the component fail-
A simpler expression of the critical NSIF k1c, as a function of the
ure occurs when the strain energy released at the nucleation of a
‘fracture toughness’ kIc (defined according to Leguillon) and the
crack of length DSE exceeds the critical value of the material, which
depends on Gc. A simplified formulation can be obtained by
1.22 expressing the strain energy release rate G as a function of the
Stress Intensity Factor KI according to Irwin’s relation valid under
1.18 plane strain conditions.
Therefore the simultaneous verification of the conditions for-
1.14 malized in Eqs. (26) and (27) provides a general criterion for the
fracture of components weakened by sharp V-notches.
γ (2α)

1.10 Z DSE
Averaged stress criterion : ry ðxÞ dx
1.06 0
Z DSE
K I
1.02 ¼ dx P rc  DSE ð26Þ
0 ð2pxÞ1k1
0.98 Z DSE Z DSE
0 20 40 60 80 100 120 140 160 180 dW p
Notch opening angle 2α, [°] Energy criterion :  da ¼ GðaÞ da P Gc  DSE
0 da 0

Fig. 3. Leguillon’s universal function c(2a). ð27Þ

26 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30

Table 3
Values of the functions w(2a) and n(2a).

2a (°) w(2a) n(2a)

x
0 2.5068 1.000
15 2.5192 0.995
y a 20 2.5232 0.993
30 2.5306 0.990
40 2.5364 0.987
45 2.5384 0.986
60 2.5401 0.983
75 2.5326 0.981
80 2.5275 0.982
2α
90 2.5127 0.983
100 2.4910 0.985
Fig. 4. Carpinteri’s coordinate system with crack of length a.
105 2.4771 0.987
Z Z " #2 120 2.4233 0.994
DSE DSE
K I k1 12 135 2.3486 1.003
K 2I ðaÞ da ¼ wð2aÞ a da P K 2Ic  DSE 140 2.3189 1.006
0 0 ð2pÞ1k1
150 2.2516 1.011
160 2.1739 1.014
where DSE is the length of the crack nucleated at the notch tip
170 2.0863 1.011
(Fig. 4), k1 the Mode I Williams’ eigenvalue [64], x the coordinate 180 1.9869 1.000
along the notch bisector line and a the generic crack length.
The parameter K I represents again the NSIF according to Carp-
interi’s definition, which is slightly different from that given by
The values of the auxiliary function n(2a) are plotted in Fig. 5
Gross and Mendelson. In particular:
pffiffiffiffiffiffiffi and listed in Table 3.
limry  2p  x1k1 K1 By using for the critical NSIF, K1c, the definition of Gross and
K I ¼ limry  ð2pxÞ1k1 ¼ x!0
1
¼ 1
ð28Þ Mendelson [66], see Eq. (28), the final result is:
x!0
ð2pÞk1 2 ð2pÞk1 2 h i
1 2ð1k Þ
In order to apply the energy criterion it is necessary to know the K 1c ¼ ð2pÞk1 2  nð2aÞ  K Ic 1  r2k
c
1 1
ð32Þ
Stress Intensity Factor KI of the crack nucleated at V-notch tip, as a
function of the crack length a. For this purpose it is possible to use
the expression by Tada et al. [68], which provides the SIF KI as a 4. Analytical comparison
function of the crack length a, the notch opening angle 2a and
the NSIF K I . Considering the three different formulations, it is possible to
K I compare the final relationships of the critical NSIF according to
k1 12
K I ðaÞ ¼ wð2aÞ  a ð29Þ Gross and Mendelson’s definition:
ð2pÞ1k1
" ð1k1 Þ #
According to Carpinteri’s criterion the failure of a V-notched ð1 þ mÞ  ð5  8mÞ 1 2ð1k Þ
SED : K 1c ¼  pffiffiffiffiffiffiffiffi  K Ic 1  r2k
c
1 1
component occurs when the stress and energy criteria given by 4p 2e1
Eqs. (26) and (27) are both satisfied. Solving the system, it is
ð33Þ
possible to determine the length of the nucleated crack DSE (Eq.
(30)) and the critical NSIF K Ic (Eq. (31)). h i
1 2ð1k Þ
 2 Leguillon : K 1c ¼ ð2pÞðk1 2Þ  cð2aÞ  K Ic 1  r2k
c
1 1
ð34Þ
2 K Ic
DSE ¼  ð30Þ
k1  w 2 rc h i
1 2ð1k Þ
" # Carpinteri et al: : K 1c ¼ ð2pÞðk1 2Þ  nð2aÞ  K Ic 1  rc2k1 1 ð35Þ
 1k1
4p 2ð1k Þ
K I P kk11   K Ic 1  rc2k1 1 It can be observed that the three criteria give the same propor-
w2
tionality relation:
2ð1k1 Þ
¼ ½nð2aÞ  K Ic  r2k
c
1 1
¼ K Ic ð31Þ
2ð1k1 Þ
K 1c / K Ic  rc2k1 1 ð36Þ
1.10 The difference is given only by the proportionality factor. This
factor depends only on the notch opening angle (2a) in Leguillon’s
and Carpinteri’s criteria. In the SED criterion, instead, the propor-
1.06 tionality factor depends on the opening angle 2a and the Poisson’s
ratio m, as shown in Fig. 6.
The proportionality factors do not differ significantly in their
ξ (2α)

1.02 trends, with the values of Leguillon’s factor being always slightly
greater than those assumed by the other two criteria for the same
notch opening angle (2a), as shown in Fig. 7 and in Table 4. For
0.98 2a = 0° and 180° Leguillon’s and Carpinteri’s criteria give the same
proportionality factor, 1.0 and 2.507, respectively. The factor based
on SED matches that of the other two criteria only for 2a = 0°, due
0.94 to its dependence on the Poisson’s ratio. However, Fig. 7 shows that
0 20 40 60 80 100 120 140 160 180
Notch opening angle 2α, [°] the values of proportionality factors as a function of the notch
opening angle (2a) are very similar for the SED and the Carpinteri’s
Fig. 5. Plot of the function n(2a), according to Carpinteri et al. [52]. criterion, despite the different initial assumptions.
 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30 27

3.20 Table 4
ν = 0.40 Proportionality factors in Eqs. (33)–(35).

2.80 ν = 0.30 2a (°) SED (m = 0.3) Leguillon Carpinteri et al.

Proportionality factor (SED)

k
ν = 0.20 0 0.5000 1.000 1.000 1.000
2.40 15 0.5002 0.981 1.000 0.995
20 0.5004 0.975 1.003 0.994
2.00 30 0.5014 0.965 1.013 0.993
40 0.5035 0.959 1.024 0.993
45 0.5050 0.958 1.029 0.995
1.60
60 0.5122 0.961 1.062 1.005
75 0.5247 0.979 1.109 1.027
1.20 80 0.5304 0.989 1.130 1.038
90 0.5445 1.017 1.183 1.066
100 0.5628 1.056 1.247 1.106
0.80
0 20 40 60 80 100 120 140 160 180 105 0.5739 1.082 1.283 1.130
Notch opening angle 2α, [°] 120 0.6157 1.186 1.434 1.229
135 0.6736 1.345 1.637 1.379
Fig. 6. Variability of proportionality factor in the SED criterion, Eq. (33), as a 140 0.6972 1.416 1.715 1.444
function of the Poisson’s ratio. 150 0.7520 1.591 1.891 1.606
160 0.8187 1.828 2.099 1.821
170 0.9000 2.156 2.316 2.109
3.00 180 1.0000 2.616 2.507 2.507
SED
2.50 Leguillon
Proportionality factors

Carpinteri et al.
The material properties were: Young’s modulus E = 3000 MPa,
2.00
Poisson’s ratio m = 0.4, fracture toughness KIc = 1 MPa m0.5. Further-
1.50
more the true curve r–e of unnotched specimens exhibited a
non-linear behaviour whereas the notched specimens presented
1.00
a brittle behaviour, with a linear trend up to the critical load, fol-
lowed by a sudden failure. Under these circumstances the critical
0.50 stress rc should be substituted by ‘‘the maximum normal stress
existing at the edge at the moment preceding the cracking’’, as
0.00 underlined in Ref. [7] where it is also recommended to use tensile
0 20 40 60 80 100 120 140 160 180 specimens with semicircular notches. Hence in order to apply the
Notch opening angle 2α, [°] analysed criteria, the critical loads obtained experimentally on
the specimens with the maximum notch root radius (q = 2 mm)
Fig. 7. Comparison among proportionality factors given by Eqs. (33)–(35).
have been used to estimate the critical strength rc. The value of
the ‘critical tensile stress’ at the notch root was 136 MPa, in good
5. Experimental validation agreement with other values reported in the literature for PMMA:
for example Carpinteri [5] and Dunn et al. [9,10] gave 130 and
The predictive capability of the three different approaches is as- 124 MPa, respectively.
sessed by using a good number of experimental data taken from The experimental values of the Notch Stress Intensity Factor K1c
the literature. The analysed data are referred to V-notched speci- to failure are plotted for each specimen in Fig. 8 (SEN specimens)
mens made of two materials: the polymer PMMA and the alumin- and Fig. 9 (TPB specimens) as a function of the notch opening angle
ium alloy Duraluminum. The mechanical properties of each tested 2a. In the same figures the theoretical curves of K1c obtained apply-
material as well as the experimental procedures adopted are de- ing the three different formulations are given for comparison.
scribed below. The second set of experimental data is due to Seweryn [7], who
The comparison between the experimental data and the theo- performed traction tests on specimens made of PMMA and
retical estimations based on Eqs. (33)–(35) is performed in terms Duraluminum.
of Notch Stress Intensity Factor K1c to failure as a function of the
V-notch opening angle 2a. When not directly available, the values
100
of the critical NSIF to failure K1c have been obtained by two-
dimensional FEM analyses, by introducing into the models the
NSIF to failure K1c, [MPa m(1-λ)]

experimental fracture loads. The finite element code ANSYS,

version 14.5, has been used. In particular, the critical NSIF to failure
10
K1c coincides with the Notch Stress Intensity
pffiffiffiffiffiffiffi Factor K1 calculated
along the notch bisector line (K 1 ¼ rh 2pr1k1 with r ! 0) when
the final load to failure is applied to the FE model.
The first set of experimental data [37] was obtained from three- 1
Experimental data
point bending and traction tests performed on PMMA V-notched SED
specimens. Leguillon
The samples geometry was characterized by a length Carpinteri et al.
l = 196 mm for the traction specimens (SEN) and l = 112 mm for 0
0 20 40 60 80 100 120 140 160 180
the three-point-bending specimens (TPB), width D = 28 mm, notch
depth a = 14 mm and thickness t = 14 mm. Four values of the notch Notch opening angle 2α, [°]
opening angle 2a = 60°, 90°, 120°, 150° in the case of SEN speci- Fig. 8. Plots of the NSIF to failure K1c (Log-scale) and comparison with the
mens and six values 2a = 15°, 30°, 60°, 90°, 120°, 150° in the case experimental data (mean values) from SEN specimens made of PMMA tested by
of TPB specimens have been considered. Gomez and Elices [37].
28 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30

100 1000
NSIF to failure K1c, [MPa m (1-λ)]

NSIF to failure K1c, [MPa m (1-λ)]

100
10

10

Experimental data a = 10 mm
1 Experimental data
Experimental data a = 20 mm
SED 1
SED
Leguillon
Leguillon
Carpinteri et al.
Carpinteri et al.
0 0
0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 200
Notch opening angle, 2α [°] Notch opening angle 2α, [°]
Fig. 9. Plots of the NSIF to failure K1c (Log-scale) and comparison with the Fig. 12. Plots of the NSIF to failure K1c (Log-scale) and comparison with the
experimental data (mean values) from TPB specimens made of PMMA tested by experimental data from PMMA specimens tested by Carpinteri [5].
Gomez and Elices [37].

1000 MPa m0.5 and tensile strength rc = 104.90 MPa. In the case of
Duraluminum, instead, the properties were: Young’s modulus
NSIF to failure K1c [MPa m (1-λ)]

E = 70,000 MPa, Poisson’s ratio m = 0.3, fracture toughness

100 KIc = 50.56 MPa m0.5 and tensile strength rc = 705.27 MPa.
The experimental values of the Notch Stress Intensity Factor K1c
to failure are reported for each specimen in Fig. 10 for PMMA and
10 in Fig. 11 for Duraluminum. In the same figures the theoretical
curves of K1c as a function of the notch opening angle 2a are
Experimental data
reported for comparison.
SED
1 The third set of experimental data [5] was obtained from three-
Leguillon point bending tests performed on PMMA V-notched specimens.
Carpinteri et al. The specimens geometry was characterized by a length
0
0 20 40 60 80 100 120 140 160 180 200
l = 190 mm, width b = 50 mm and thickness t = 50 mm. Two values
Notch opening angle, 2α [°]
of the notch depth a = 10, 20 mm and six values of the notch open-
ing angle 2a = 0°, 45°, 90°, 120°, 150°, 180° have been considered. A
Fig. 10. Plots of the NSIF to failure K1c (Log-scale) and comparison with the total number of 36 specimens were tested, three specimens for
experimental data (mean values) from PMMA specimens tested by Seweryn [7]. each geometry.
The material properties were: Young’s modulus E = 3000 MPa,
Poisson’s ratio m = 0.3, fracture toughness KIc = 1.89 MPa m0.5 and
2500 tensile strength rc = 130.30 MPa.
The experimental values of the Notch Stress Intensity Factor K1c
NSIF to failure K1c, [MPa m (1-λ)]

to failure are shown for each specimen in Fig. 12 as a function of

the notch opening angle 2a and the notch depth a.
The plots of Figs. 8–12 shown that the agreement between the
theoretical estimates and the experimental values of Notch Stress
250 Intensity Factor K1c to failure is good in all cases. For notch opening
angles greater than 120°, the minimum deviation seems to be
Experimental data
slightly reduced in the case of the averaged SED and the Carpinte-
SED
ri’s criteria, which are substantially coincident although based on
Leguillon
different assumptions. Furthermore it can be observed that theo-
Carpinteri et al.
retical assessments based on the averaged SED and the Carpinteri’s
25 criteria are always conservative with respect to those obtained by
0 20 40 60 80 100 120 140 160 180 200
Notch opening angle 2α, [°] applying the Leguillon’s criterion.

Fig. 11. Plots of the NSIF to failure K1c (Log-scale) and comparison with the
experimental data (mean values) from Duraluminum specimens tested by Seweryn 6. Conclusions
[7].
A comparison between different fracture criteria for brittle
The specimens geometry was characterized by a length components weakened by sharp V-notches has been presented in
l = 192 mm, width W = 109 mm and a notch depth a = 27 mm. the paper considering Mode I loading conditions. The comparison
The thickness of PMMA specimens was t = 4 mm, whereas the has considered the averaged SED criterion, and two different for-
thickness of Duraluminum specimens was t = 5 mm. Nine values mulations of the Finite Fracture Mechanics theory, according to
of the notch opening angle, 2a = 0°, 20°, 40°, 60°, 80°, 120°, 140°, Leguillon’s or to Carpinteri et al.
160°, 180° have been considered. Three PMMA and two The analytical comparison has been performed on the basis
Duraluminum specimens were tested for each opening angle. of the different expressions for the critical value of the NSIF K1c.
The mechanical properties of PMMA were: Young’s modulus The same proportionality relation exists between K1c and two
E = 3000 MPa, Poisson’s ratio m = 0.3, fracture toughness KIc = 1.86 fundamental mechanical properties of the material: the fracture
 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30 29

toughness KIc and the ultimate tensile strength rc. The difference is [27] M.H. El Haddad, T.H. Topper, K.N. Smith, Prediction of non-propagating cracks,
Eng. Fract. Mech. 11 (1979) 573–584.
given by a proportionality factor, which depends only on the
[28] H. Neuber, Zur Theorie der technischen Formzahl, Forsch. Ing. Wes. 7 (1936)
V-notch opening angle in Leguillon and Carpinteri’s criteria, 271–274.
whereas it depends on the opening angle and the Poisson’s ratio [29] H. Neuber, Theory of Notch Stresses, Springer-Verlag, Berlin, 1958.
in the SED approach. The variability of the proportionality factors [30] H. Neuber, Uber die Berücksichtigung der Spannungskonzentration bei
Festigkeitsberechnungen, Konstruktion 20 (1968) 245–251.
has been examined in detail. [31] F. Berto, P. Lazzarin, D. Radaj, Fictitious notch rounding concept applied to
Finally, the three criteria have been applied to components sharp V-notches: evaluation of the microstructural support factor for different
weakened by sharp V-notches under Mode I loading conditions failure hypotheses. Part I: Basic stress equations, Eng. Fract. Mech. 75 (2008)
3060–3072.
in order to investigate the predictive capability of each approach. [32] F. Berto, P. Lazzarin, D. Radaj, Fictitious notch rounding concept applied to
A number of experimental data taken from the literature have been sharp V-notches: evaluation of the microstructural support factor for different
used for the comparison. A good agreement in terms of critical failure hypotheses. Part II: Microstructural support analysis, Eng. Fract. Mech.
76 (2009) 1151–1175.
values of the NSIF K1c has been found in all cases. It was observed [33] F. Berto, P. Lazzarin, Fictitious notch rounding approach of pointed V-notch
that the static strength assessment based on the SED and the under in-plane shear, Theoret. Appl. Fract. Mech. 53 (2010) 127–135.
Carpinteri’s criteria is always conservative with respect to the [34] D. Radaj, P. Lazzarin, F. Berto, Generalised Neuber concept of fictitious notch
rounding, Int. J. Fatigue 51 (2013) 105–115.
theoretical prediction based on the Leguillon’s criterion. [35] M. Elices, G.V. Guinea, J. Gómez, J. Planas, The cohesive zone model:
advantages, limitations and challenges, Eng. Fract. Mech. 69 (2002) 137–163.
[36] J. Planas, M. Elices, G.V. Guinea, F.J. Gómez, D.A. Cendón, I. Arbilla,
References Generalizations and specializations of cohesive crack models, Eng. Fract.
Mech. 70 (2003) 1759–1776.
[37] F.J. Gomez, M. Elices, Fracture of components with V-shaped notches, Eng.
[1] G.R. Irwin, Analysis of stresses and strains near the end of a crack traversing a
Fract. Mech. 70 (2003) 1913–1927.
plate, J. Appl. Mech. 24 (1957) 361–364.
[38] F.J. Gómez, M. Elices, Fracture loads for ceramic samples with rounded
[2] A.A. Griffith, The phenomena of rupture and flow in solids, Philos. Trans. R. Soc.
notches, Eng. Fract. Mech. 73 (2006) 880–894.
Lond. A 221 (1921) 163–198.
[39] P. Livieri, Use of J-integral to predict static failures in sharp V-notches and
[3] R.H. Leicester, Effect of size on the strength of structures, CSIRO Forest
rounded U-notches, Eng. Fract. Mech. 75 (2008) 1779–1793.
Products Laboratory, Division of Building Research, Paper No. 71, 1973.
[40] E. Barati, J. Aghazadeh Mohandesi, Y. Alizadeh, The effect of notch depth on J-
[4] P.A. Gradin, A fracture criterion for edge-bonded bi-material bodies, J. Compos.
integral and critical fracture load in plates made of functionally graded
Mater. 16 (1982) 448–456.
aluminum–silicone carbide composite with U-notches under bending, Mater.
[5] A. Carpinteri, Stress singularity and generalised fracture toughness at the
Des. 31 (2010) 4686–4692.
vertex of re-entrant corners, Eng. Fract. Mech. 26 (1987) 143–155.
[41] F. Berto, P. Lazzarin, Y. Matvienko, J-integral evaluation for U- and V-blunt
[6] Z. Knels, A criterion of V-notch stability, Int. J. Fract. 48 (1991) 79–83.
notches under Mode I loading and materials obeying a power hardening law,
[7] A. Seweryn, Brittle fracture criterion for structures with sharp notches, Eng.
Int. J. Fract. 146 (2007) 33–51.
Fract. Mech. 47 (1994) 673–681.
[42] M.R. Ayatollahi, M.R.M. Aliha, Mixed mode fracture analysis of polycrystalline
[8] L.S. Nui, C. Chehimi, G. Pluvinage, Stress field near a large blunted tip V-notch
graphite – a modified MTS criterion, Carbon 46 (2008) 1302–1308.
and application of the concept of the critical notch stress intensity factor (NSIF)
[43] H. Saghafi, M.R. Ayatollahi, M. Sistaninia, A modified MTS criterion (MMTS) for
to the fracture toughness of very brittle materials, Eng. Fract. Mech. 49 (1994)
mixed-mode fracture toughness assessment of brittle materials, Mater. Sci.
325–335.
Eng., A 527 (2010) 5624–5630.
[9] M.L. Dunn, W. Suwito, S.J. Cunningham, Fracture initiation at sharp notches:
[44] M.R.M. Aliha, M.R. Ayatollahi, Analysis of fracture initiation angle in some
correlation using critical stress intensities, Int. J. Solids Struct. 34 (1997) 3873–
cracked ceramics using the generalized maximum tangential stress criterion,
3883.
Int. J. Solids Struct. 49 (2012) 1877–1883.
[10] M.L. Dunn, W. Suwito, S.J. Cunningham, C.W. May, Fracture initiation at sharp
[45] P. Lazzarin, R. Zambardi, A finite-volume-energy based approach to predict the
notches under mode I, mode II, and mild mixed mode loading, Int. J. Fract. 84
static and fatigue behavior of components with sharp V-shaped notches, Int. J.
(1997) 367–381.
Fract. 112 (2001) 275–298.
[11] D. Leguillon, Z. Yosibash, Crack onset at a V-notch. Influence of the notch tip
[46] Z. Yosibash, A.R. Bussiba, I. Gilad, Failure criteria for brittle elastic materials,
radius, Int. J. Fract. 122 (2003) 1–21.
Int. J. Fract. 125 (2004) 307–333.
[12] B. Atzori, Fracture mechanics or notch effect for fatigue design, in: Proc. XIII
[47] P. Lazzarin, F. Berto, M. Elices, F.J. Gomez, Brittle failures from U- and V-
National Congress of the Italian Society for Strain Analysis (Edited by AIAS),
notches in mode I and mixed, I + II, mode: a synthesis based on the strain
Bergamo Italy, 1985 (in Italian).
energy density averaged on finite-size volumes, Fatigue Fract. Eng. Mater.
[13] T. Boukharouba, T. Tamine, L. Nui, C. Chehimi, G. Pluvinage, The use of notch
Struct. 32 (2009) 671–684.
stress intensity factor as a fatigue crack initiation parameter, Eng. Fract. Mech.
[48] F. Berto, P. Lazzarin, A review of the volume-based strain energy density
52 (1995) 503–512.
approach applied to V-notches and welded structures, Theoret. Appl. Fract.
[14] Y. Verreman, B. Nie, Early development of fatigue cracking at manual fillet
Mech. 52 (2009) 183–194.
welds, Fatigue Fract. Eng. Mater. Struct. 19 (1996) 669–681.
[49] F. Berto, P. Lazzarin, Recent developments in brittle and quasi-brittle failure
[15] G. Pluvinage, Rupture and fatigue initiated from notches. Application of the
assessment of engineering materials by means of local approaches, Mater. Sci.
notch intensity factor, Rev. Fr. Mecanique (1997) 53–61 (in French).
Eng. R 75 (2014) 1–48.
[16] P. Lazzarin, R. Tovo, A notch intensity factor approach to the stress analysis of
[50] D. Leguillon, A criterion for crack nucleation at a notch in homogeneous
welds, Fatigue Fract. Eng. Mater. Struct. 21 (1998) 1089–1103.
materials, Comptes Rendus de l’Acadèmie des Sciences – Series IIB –
[17] B. Atzori, P. Lazzarin, R. Tovo, From the local stress approach to fracture
Mechanics 329 (2001) 97–102.
mechanics: a comprehensive evaluation of the fatigue strength of welded
[51] D. Leguillon, Strength or toughness? A criterion for crack onset at a notch, Eur.
joints, Fatigue Fract. Eng. Mater. Struct. 22 (1999) 369–381.
J. Mech. A/Solids 21 (2002) 61–72.
[18] B. Atzori, P. Lazzarin, R. Tovo, Stress field parameters to predict the fatigue
[52] A. Carpinteri, P. Cornetti, N. Pugno, A. Sapora, D. Taylor, A finite fracture
strength of notched components, J. Strain Anal. 34 (1999) 437–453.
mechanics approach to structures with sharp V-notches, Eng. Fract. Mech. 75
[19] V. Novozhilov, On a necessary and sufficient criterion for brittle strength, J.
(2008) 1736–1752.
Appl. Math. Mech. (Trans. PMM) 33 (1969) 201–210.
[53] P. Cornetti, N. Pugno, A. Carpinteri, D. Taylor, Finite fracture mechanics: a
[20] A. Seweryn, Z. Mróz, A non-local stress failure condition for structural
coupled stress and energy failure criterion, Eng. Fract. Mech. 73 (2006) 2021–
elements under multiaxial loading, Eng. Fract. Mech. 51 (1995) 955–973.
2033.
[21] A. Seweryn, S. Poskrobko, Z. Mróz, Brittle fracture in plane elements with sharp
[54] Z. Yosibash, E. Priel, D. Leguillon, A failure criterion for brittle elastic materials
notches under mixed-mode loading, J. Eng. Mech. 123 (1997) 535–543.
under mixed-mode loading, Int. J. Fract. 141 (2006) 291–312.
[22] D. Taylor, Geometrical effects in fatigue: a unifying theoretical model, Int. J.
[55] A. Sapora, P. Cornetti, A. Carpinteri, A finite fracture mechanics approach to V-
Fatigue 21 (1999) 413–420.
notched elements subjected to mixed-mode loading, Eng. Fract. Mech. 97
[23] L. Susmel, D. Taylor, The theory of critical distances to predict static strength of
(2013) 216–226.
notched brittle components subjected to mixed-mode loading, Eng. Fract.
[56] G.C. Sih, Strain-energy-density factor applied to mixed mode crack problems,
Mech. 75 (2008) 534–550.
Int. J. Fract. 10 (1974) 305–321.
[24] L. Susmel, D. Taylor, The theory of critical distances to estimate the static
[57] G.C. Sih, J.W. Ho, Sharp notch fracture strength characterized by critical energy
strength of notched samples of Al6082 loaded in combined tension and
density, Theoret. Appl. Fract. Mech. 16 (1991) 179–214.
torsion. Part I: Material cracking behaviour, Eng. Fract. Mech. 77 (2010) 452–
[58] P. Lazzarin, F. Berto, F.J. Gómez, M. Zappalorto, Some advantages derived from
469.
the use of the strain energy density over a control volume in fatigue strength
[25] K. Tanaka, Engineering formulae for fatigue strength reduction due to cracklike
assessments of welded joints, Int. J. Fatigue 30 (2008) 1345–1357.
notches, Int. J. Fract. 22 (1983) 39–46.
[59] P. Lazzarin, F. Berto, M. Zappalorto, Rapid calculations of notch stress intensity
[26] P. Lazzarin, R. Tovo, G. Meneghetti, Fatigue crack initiation and propagation
factors based on averaged strain energy density from coarse meshes:
phases near notches in metals with low notch sensitivity, Int. J. Fatigue 19
theoretical bases and applications, Int. J. Fatigue 32 (2010) 1559–1567.
(1997) 647–657.
30 P. Lazzarin et al. / Theoretical and Applied Fracture Mechanics 71 (2014) 21–30

[60] A. Kotousov, F. Berto, P. Lazzarin, F. Pegorin, Three dimensional finite element [64] M.L. Williams, Stress singularities resulting from various boundary
mixed fracture mode under anti-plane loading of a crack, Theoret. Appl. Fract. conditions in angular corners of plates in tension, J. Appl. Mech. 19 (1952)
Mech. 62 (2012) 326–333. 526–528.
[61] F. Berto, P. Lazzarin, A. Kotousov, L.P. Pook, Induced out-of-plane mode at the [65] P. Lazzarin, R. Tovo, A unified approach to the evaluation of linear elastic
tip of blunt lateral notches and holes under in-plane shear loading, Fatigue stress fields in the neighborhood of cracks and notches, Int. J. Fract. 78 (1996)
Fract. Eng. Mater. Struct. 35 (2012) 538–555. 3–19.
[62] M. Treifi, S.O. Oyadiji, Strain energy approach to compute stress intensity [66] R. Gross, A. Mendelson, Plane elastostatic analysis of V-notched plates, Int. J.
factors for isotropic homogeneous and bi-material V-notches, Int. J. Solids Fract. Mech. 8 (1972) 267–276.
Struct. 50 (2013) 2196–2212. [67] E. Beltrami, Sulle condizioni di resistenza dei corpi elastici, Il Nuovo Cimento
[63] B.R. Davis, P.A. Wawrzynek, A.R. Ingraffea, 3-D simulation of arbitrary crack 18 (1885) 145–155 (in Italian).
growth using an energy-based formulation – Part I: Planar growth, Eng. Fract. [68] H. Tada, P. Paris, G. Irwin, The stress analysis of cracks, Handbook, St. Louis,
Mech. 115 (2014) 204–220. MO, USA, Paris Productions Incorporated, 1985.