See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/286140135

Comparison of the size-independent fracture energy of concrete obtained by

two test methods

Article · January 2013

CITATIONS READS

2 152

3 authors, including:

Héctor Cifuentes
Universidad de Sevilla
62 PUBLICATIONS   357 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Design of advanced cementitious materials based on target performance View project

All content following this page was uploaded by Héctor Cifuentes on 10 July 2017.

The user has requested enhancement of the downloaded file.

 An International Journal for Experimental Mechanics

Measuring the Size-Independent Fracture Energy of

Concrete
H. Cifuentes, M. Alcalde and F. Medina
Continuum Mechanics and Structural Analysis Department, Escuela Técnica Superior de Ingenierı́a, University of Seville, Camino de los
Descubrimientos s/n E-41092 Seville, Spain

ABSTRACT: The fracture energy of concrete is an important parameter in the analysis of the mechanical behaviour of concrete
structures, so it is considered as a material property. However, the most universal test method for measuring the specific fracture energy of
concrete (RILEM work-of-fracture) has been a subject of intense debate among researchers. The values have been found to vary with the size
and shape of the test specimen. In this study, an experimental comparative analysis of the size-independent fracture energy obtained by two
main methods has been carried out. One of these is based on the local fracture energy model of Hu et al. The second is based on the
curtailment of the tail of the P–d curve by Elices et al. Therefore, the relationship between both methods is highlighted. It is shown that both
methods give almost identical results.
KEYWORDS: concrete, local fracture energy, size-independent fracture energy, test methods

Introduction study available in the literature comparing these two

One of the most important parameters in fracture behav-
iour of concrete and cracking analysis is the fracture
energy, representing cracking resistance and fracture
toughness of concrete. For this reason, the fracture energy
is considered as a material property of concrete. The most
commonly used method for measuring the fracture energy
is the work-of-fracture method recommended by RILEM
[1]. The values determined by this method are dependent
on the size and shape of the test specimen, as demonstrated
and analysed by several authors in the last decades [2–7].
However, it is still regarded as a universal method for
determining the fracture energy of concrete [8–10]. More-
over, the reasons for the variability of the RILEM fracture
energy of concrete with the size of the ligament area are a
subject of ongoing investigations [10–15].
Several researches have analysed this size effect on
fracture energy measurements according to RILEM proce-
dure, and they have proposed new methods or modifica-
tions to obtain a size-independent specific fracture energy
of concrete. Of course, a parameter considered to be a
material property must be independent on any geomet-
rical property of the specimen and test method. The two
most popular methods for measuring the size-indepen-
dent fracture energy of concrete (also called true fracture
energy [16]) are based on the local fracture energy model
of Hu and Wittmann [16] and the experimental correc-
tions to avoid energy dissipations proposed by Elices et al.
[17–19]. Although the procedure of each method is dif-
ferent, they are related. Hu and Wittmann stated that the
size effect is caused by a variation of the fracture process
zone when the crack approaches the free boundary sur-
face of the specimen at the end of the test [20]. On the
other hand, among the several sources of experimental
errors analysed by Planas et al. [17], the most significant is
the influence of the non-measured fracture energy of
concrete due to the curtailment of the tail of the P–d
curve. In both cases, it is considered that the size effect of
the fracture energy of concrete is due to the behaviour of
methods.
This study deals with the comparative experimental
analysis on the size-independent fracture energy of con-
crete determined by these two methods. The results
attempt to show the relationship between these methods,
as well as to ensure the existence of a fracture material
property independent of any geometrical parameter and
the test method.
Theoretical Background
Local fracture energy method
Hu and Wittmann [6] argued that the effect of the free
boundary of the specimen is felt in the fracture process
zone (FPZ) ahead of a real crack in a concrete structure. The
energy required to create a fresh crack decreases as the
crack approaches the free boundary [16]. Initially, when
the crack grows from a pre-existing notch, the rate of
decrease is moderate, almost a constant, but it accelerates
as the crack approaches the end of the uncracked ligament.
This change in the local fracture energy (gf) is represented
by a bilinear approximation, as shown in Figure 1. This
bilinear function consists of a horizontal line with the
value of the size-independent fracture energy and a
descending branch that reduces to zero at the back surface
of the specimen. The transition from GF to the rapid
decrease occurs at the transition ligament length, which is
given by the intersection of these two lines and it depends
on the material properties and specimen size and shape
[21].
On the basis of the boundary effect method of Hu and
Wittmann [6], the measured RILEM fracture energy, Gf,
represents the average of the local fracture energy function
over the ligament area (dotted line in Figure 1). The rela-
tionship between all the involved variables is given by
R
Da
8
gf ðxÞdx < h al =D
i
0 GF 1  2ð1a=D Þ ; 1  a=D>al =D
Gf ða; DÞ ¼ ¼
Da : GF 1a=D
; 1  a=D  al =D
2al =D

the material near the end of the test. However, there is no (1)

54  2012 Wiley Publishing Ltd j Strain (2013) 49, 54–59

doi: 10.1111/str.12012
 H. Cifuentes, M. Alcalde and F. Medina : Size-independent fracture energy methods
Figure 1: Local fracture energy model of Hu and Wittmann
where, D is the total depth of the specimens and a is the
initial notch depth.
To obtain the values of GF and al for each mix of con-
crete, the size-dependent RILEM fracture energy of at least
four specimens with single size and a full range variation of
the notch depth must be previously determined [21]. With
application of equation (1) for each different notch depth
specimen, an overdetermined system of equations must be
solved by a least squares method to obtain the best esti-
mates of GF and al. Hu et al. [20] showed that although the
measured values Gf depend on D and a, the above proce-
dure indeed leads to a GF value that is essentially inde-
pendent of the specimen size and relative notch depth.
Based on the BE method of Hu and Wittmann [6], Abdalla
and Karihaloo [22, 23] have proposed and validated a
simplified method (SBE) that allows the estimation of the
size-independent fracture energy by testing specimens of a
single size but with the smallest and the largest notch to
depth ratios only and solving the system of two equations (1)
in two unknowns GF and al. This simplified method elimi-
nates the need for the least squares method for the solution
of an overdetermined system of simultaneous equations and
the time-consuming testing of a large number of specimens
with different D and a = a/D. They have demonstrated that
the necessary values of the smallest relative notch depth
must be a £ 0.1 for three-point bending and £ 0.2 for wedge
splitting tests. The values of the deepest starter notch must
be a ‡ 0.5 for both the TPB and WS tests [23].
Method proposed by Elices et al.
Elices, Planas and Guinea [17–19] identified several sources
of energy dissipation that may influence the measurement
of GF. They found that most of these sources are due to
experimental errors which affect the measured value of the
work-of-fracture during the test. They classified these
sources in three main groups [17], (i) due to the testing
equipment and experimental setup, (ii) to the energy dis-
sipation in the specimen bulk and (iii) the non-measured
energy corresponding to the unrecorded tail of the P–d
curve at the end of the test. The former can be avoided with
a proper calibration of the equipment and design of sup-
ports and loading system. The second is due to crushing at
supports and under the loading point, and to high tensile
 2012 Wiley Publishing Ltd j Strain (2013) 49, 54–59
doi: 10.1111/str.12012
stresses. The crushing dissipation can be corrected by
adjusting the initial stiffness of the P–d curve and measur-
ing the midspan deflection at the lower part of the beam.
The dissipation in the specimen bulk, Wdb, due to high
tensile stresses cannot be avoided; however, they represent
less than the 2% [17]. The last energy dissipation corre-
sponds to the curtailment of the tail of the P–d at the end of
the test and is the most significant affecting the size
dependency of the measured fracture energy.
To estimate this non-measured energy when the test is
interrupted (Wnm in Figure 3) at very low loads, it is necessary
to model the beam behaviour when the cohesive crack closely
approaches to the free surface [17]. For cohesive materials and
specimens where weight is compensated, the last phase of a
stable three-point bend test can be modelled following the
rigid-body kinematics (Figure 2) used by Petersson [24].
Taking into account the geometric relationships given by
the rigid-body model (Figure 2), the load can be approxi-
mated by a hyperbolic function of the midspan deflection:
bSM1
P¼ (2)
4d2
where, M1 is the static moment of the cohesive law of
concrete.
The non-measured work-of-fracture can be estimated by
integration of equation (2) in a simple way, avoiding the
determination of M1, with the adjustment of the tail of the
curve as follows:
Z1
2A
Wnm ¼ Pdd ¼ (3)
du
du
where,A is the experimental coefficient of adjustment of
the P’–d tail and du is the last recorded midspan deflection
of the specimen at the end of the test (Figure 3).
Figure 2: Rigid-body model of the behaviour of the specimen at
the end of the test
Figure 3: P–d curve in a three-point bend test and the measured
(Wf) and non-measured (Wnm) fracture energy
55
Size-independent fracture energy methods : H. Cifuentes, M. Alcalde and F. Medina

Table 1: Geometrical properties of the notched specimens

Specimen D (mm) b (mm) S (mm) L (mm) a (mm) S/D a = a/D

TFE005 120 60 480 540 6 4 0.05

TFE05 120 60 480 540 60 4 0.5
SWC05 120 60 480 540 60 4 0.5

Once the non-measured energy has been estimated,
the size-independent fracture energy of concrete can be
obtained as [17]
Rdu
Pdd  Wdb þ 2A=du
0
GF ¼ (4)
bðD  a0 Þ
Experimental Procedure
To compare the measured values of the size-independent
fracture energy given by application of the two methods
described above, an experimental campaign was carried
out. Prismatic notched specimens were subjected to three-
point bend tests according to the experimental require-
ments of each method. Three kinds of notched beams were
manufactured. The TFE05 and TFE005 specimens were used
to obtain the size-independent fracture energy of concrete
by means of the simplified boundary effect method [22]. In
contrast, SWC05 specimens were used to the adjustment
of the tail of the P–d curve method. Table 1 shows the
geometrical dimensions of all specimens as displayed in
Figure 4.
Four samples were tested for each type of specimen, and
the main values and the coefficient of variation for every
analysed parameter were obtained. The initial tangent
correction due to crushing of concrete at supports was
made for all specimens. The TFE05 and TFE005 specimens
were tested according to the RILEM procedure, and the
self-weight was not compensated. Figure 5 shows a picture
of a three-point bend test of these specimens (left) and its
instrumentation (right).
The SWC05 specimens were subjected to three-point
bend tests with the indicated self-weight compensation
(Figure 6). In these specimens, the ultimate displacement
at which the test was stopped was about 3.5 mL.
All tests were driven in a closed-loop servo-hydraulic
dynamic testing machine (± 50 kN).The rate of loading was
controlled by a crack mouth opening displacement
(CMOD) gauge at a very low rate (0.001 mm s)1) so that the
fracture occurred in a stable manner. The CMOD displace-
ment was measured with a clip gauge transducer, and a
LVDT linear displacement transducer was used to measure
the vertical displacement at midpoint (Figure 5 right). A
reference frame was used to fix the LVDT transducer to the
bottom of the specimen, and only the vertical displace-
ment by deformations of the specimen was measured. The
load–CMOD and load–displacement (P–d) curves for all
specimens have been recorded.
All specimens have been manufactured with a single
mixed concrete. Concrete mix proportions by weight of
sand/gravel/cement/water were 1.4/3.5/1/0.4. The sand
and gravel were siliceous aggregates with a maximum
size of 8 mL mixed according to the Fuller method. Initial
notches were made by cutting with a diamond saw blade
3 mL in thickness. There was a strict control of the speci-
men-making process to minimise scatter in test results.
Compressive tests were carried out on 150 mm cubes to
determine the compressive strength of the concrete, fc, in
accordance with UNE EN12930-3. Brazilian tests were also
carried out on cylindrical specimens of 150 · 300 mm
according to UNE EN12930-6 to determine the split tensile
strength of the concrete, fst. The modulus of rupture, ff, of the
concrete mixes was also determined from four-point bend-
ing test on prismatic specimens of 150 · 150 · 600 mm
according to UNE EN12930-5. The static elastic modulus of
the concrete, Ec, was estimated from the P–CMOD curve as
indicated in [25]. Table 2 shows the mechanical properties of
concrete.

Figure 4: Geometrical shape of notched specimens used for three-
point bend tests
Results
Simplified boundary effect method
Table 3 shows the average values obtained from the three-
point bend tests of TFE005 and TFE05 specimens according
to the RILEM procedure [1], such as the maximum load
obtained from tests (Pmax), the self-weight of specimens
(m), the vertical displacement at the end of the test (du), the
maximum load corrected according to the self-weight of
the specimen (P’max), the work-of-fracture measured from

56  2012 Wiley Publishing Ltd j Strain (2013) 49, 54–59

doi: 10.1111/str.12012
 H. Cifuentes, M. Alcalde and F. Medina : Size-independent fracture energy methods

Figure 5: Three-point bend test of the TFE05 specimens (left) and its instrumentation (right)

Table 4: Determination of size-independent fracture energy of

concrete in SWC05 specimens

Specimen SWC05-1 SWC05-2 SWC05-3 SWC05-4

Pmax (N) 1444 1585 1410 1547

m (kg) 9.4 9.8 9.7 9.7
d0 (mm) 2.1 2.1 2.1 2.1
du (mm) 3.5 3.4 3.6 3.5
A (Nmm2) 84.4 172.3 98.1 113.8
Wf (Nmm) 518.6 484.4 390.3 430.3
Wnm (Nmm) 48.3 99.9 56.6 65.5
)Wdb (Nmm) )7.3 )6.8 )5.5 )6.0
Alig (mm2) 3600 3600 3600 3600
GF (N/m) 155.4 160.4 122.6 136.1

Figure 6: Three-point bend test of SWC05 specimens with weight

compensation

specific fracture energy with the size of the uncracked lig-

Table 2: Mechanical properties of concrete ament area (D – a0). Applying the simplified method of
Abdalla and Karihaloo [22] to these values, the size-inde-
Compressive strength, fc (MPa) 36.9 ± 6% pendent fracture energy of concrete (GF1) and the transi-
Splitting tensile strength, fst (MPa) 3.1 ± 13% tion length (al1) were obtained:
Modulus of rupture, ff (MPa) 4.6 ± 1% GF1 = 144.2 (N/m)
Young’s modulus, Ec (GPa) 28.2 ± 11% al1 = 16.4 mm.

Adjustment of the tail method

Table 3: Determination of RILEM fracture energy in TFE005 and
Table 4 shows the SWC05 specimen results for determining
TFE05 specimens
GF according to the adjustment of the tail method where d0
is the displacement considered as the initial point for
Specimen TFE005 % TFE05 %
adjusting the tail of the curve, A is the constant of adjust-
Pmax (N) 4733 ± 4 1332 ± 4 ment, Wnm is the non-measured work-of-fracture, Wdb is
m (kg) 9.6 ± 2 9.6 ± 1 the energy dissipated in the bulk of the specimens and GF is
du (mm) 1.41 ± 4 1.35 ± 1 the size-independent specific fracture energy. The values of
P’max (N) 4775 ± 4 1358 ± 4 Wdb were considered to represent the 1.4% of the measured
Wf (Nmm) 797.7 ± 12 333.5 ± 5 work-of-fracture energy, as it was obtained by Elices et al.
WfT (Nmm) 915.4 ± 11 448.3 ± 4 [19] for specimens with a total depth of 120 mm. The other
Alig (mm2) 6840 3600 sources of energy dissipation were minimised by the proper
Gf (N/m) 133.8 ± 11 124.5 ± 4 calibration of the equipment and design of the supports
and loading system.
The specific fracture energy (GF) determined with this
the P–d curve (Wf), the total work-of-fracture considering method is directly the size-independent fracture energy of
the self-weight correction (WfT), the ligament area (Alig) concrete. So, the value of the size-independent fracture
and the RILEM specific fracture energy (Gf). energy obtained with this method (GF2) is the average value
The comparison of the Gf values obtained for TFE005 and for all specimens SWC05:
TFE05 specimens reveals the dependency of the RILEM GF2 = 143.6 (N/m) ± 12%

 2012 Wiley Publishing Ltd j Strain (2013) 49, 54–59 57

doi: 10.1111/str.12012
Size-independent fracture energy methods : H. Cifuentes, M. Alcalde and F. Medina
Discussion of Results
Comparing GF1 and GF2 results, a great accuracy between
them is observed. Consequently, the two methods used in
this study to measure the size-independent fracture energy
of concrete show the same value, as corresponds to a
property of the material. Although these two methods use
different experimental procedures, they are interrelated, as
had been anticipated by Abdalla and Karihaloo. Both pro-
cedures apply some corrections to the final part of the P–d
diagram in the work-of-fracture test [11]. The local fracture
energy model considers the influence when the crack
approaches the back-face free boundary surface of the spec-
imen towards the end of the test [26]. On the other hand, the
method of Elices et al. [17] consists in determining the non-
measured work-of-fracture by adjusting the tail of the P–d
curve that also corresponds to the final part of the test.
The results presented in this study have demonstrated
the independency of the size-independent fracture energy
of concrete with the test procedure. Moreover, it has also
been shown as the influence of the back free boundary of
the uncracked ligament area on the FPZ plays an important
role in the fracture behaviour of concrete [27]. The size
effect observed in the measured RILEM fracture energy is
mainly related with the size of the uncracked ligament area
as it was demonstrated for one-size specimens with differ-
ent notch depths. For this reason, when geometrically
similar specimens are employed, as described in the RILEM
Size Effect Method [28], the influence of the size of the
uncracked area (not only the size of the specimen) could
also be studied to obtain a value of the size-independent
specific fracture energy of concrete according to the BE
method.
Conclusions
An experimental comparative analysis of the two main
methods used to measure the size-independent fracture
energy of concrete has been carried out. This analysis has
not been carried out before in the literature. The value of
the fracture energy obtained with both methods was
practically the same. Both methods are interrelated, as they
analyse the size effect on the fracture energy at the end of
the three-point bend test. This study has shown a value of
the specific fracture energy that could be considered as a
fracture material property, independent on the test meth-
od. It is, therefore, concluded that either method can be
used to obtain a unique value of the size-independent
fracture energy of concrete.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the financial support
provided to this research by the Spanish Ministry of Science
and Technology under project BIA2010-21399-C02-02.
REFERENCES
1. RILEM TCM-85. (1985) Determination of the fracture energy
of mortar and concrete by means of three-point bend tests on
notched beams. Mater. Struct. 18, 287–290.
58
2. Bazant, Z. P. and Kazemi, M. T. (1991) Size dependence
of concrete fracture energy determined by RILEM work-of-
fracture method. Int. J. Fract. 51, 121–138.
3. Bazant, Z. P. (1996) Analysis of work-of-fracture method for
measuring fracture energy of concrete. J. Mater. Civil Eng. 122,
138–144.
4. Nallathambi, P., Karihaloo, B. L. and Heaton, B. S. (1985)
Various size effect in fracture of concrete. Cem. Concr. Compos.
15, 117–126.
5. Carpinteri, A. and Chiaia, B. (1996) Size effects on concrete
fracture energy: dimensional transition from order to disor-
der. Mater. Struct. 29, 259–266.
6. Hu, X. and Wittmann, F. (1992) Fracture energy and fracture
process zone. Mater. Struct. 25, 319–326.
7. Mindess, S. (1984) The effect of specimen size on the fracture
energy of concrete. Cem. Concr. Res. 14, 431–436.
8. Pan, Z. (2011) Fracture properties of geopolymer paste and
concrete. Mag. Concr. Res. 63, 763–771.
9. Şahin, Y. and Köksal, F. (2011) The influences of matrix and
steel fibre tensile strengths on the fracture energy of high-
strength concrete. Constr. Build. Mater. 25, 1801–1806.
10. Vydra, V., Trtı́k, K. and Vodák, F. (2012) Size indepen-
dent fracture energy of concrete. Constr. Build. Mater. 26,
357–361.
11. Zhao, Z., Kwon, S. and Shah, S. (2008) Effect of specimen size
on fracture energy and softening curve of concrete: part
I. Experiments and fracture energy. Cem. Concr. Res. 38, 1049–
1060.
12. Kwon, S., Zhao, Z. and Shah, S. (2008) Effect of specimen size
on fracture energy and softening curve of concrete: part II.
Inverse analysis and softening curve. Cem. Concr. Res. 38,
1061–1069.
13. Muralidhara, S., Raghu Prasad, B. K., Karihaloo, B. L. and
Singh, R. K. (2011) Size-independent fracture energy in plain
concrete beams using tri-linear model. Constr. Build. Mater. 25,
3051–3058.
14. Hu, X. (2010) Size effect on tensile softening relation. Mater.
Struct. 44, 129–138.
15. Yu, Q., Le, J. L., Hoover, C. G. and Bazant, Z. P. (2010) Prob-
lems with hu-duan boundary effect model and its comparison
to size-shape effect law for quasi-brittle fracture. J. Eng. Mech.
136, 40–50.
16. Hu, X. and Wittmann, F. (2000) Size effect on toughness
induced by crack close to free surface. Eng. Fract. Mech. 65,
209–221.
17. Elices, M., Guinea, G. V. and Planas, J. (1992) Measurement of
the fracture energy using three-point bend tests: part
3—Influence of cutting the P -d tail. Mater. Struct. 25, 137–163.
18. Guinea, G. V., Planas, J. and Elices, M. (1992) Measurement of
the fracture energy using three-point bend tests: part
1—Influence of experimental procedures. Mater. Struct. 25,
212–218.
19. Planas, J., Elices, M. and Guinea, G. V. (1992) Measurement of
the fracture energy using three-point bend tests: part
2—Influence of bulk energy dissipation. Mater. Struct. 25, 305–
312.
20. Hu, X. and Duan, K. (2007) Size effect: influence of proximity
of fracture process zone to specimen boundary. Eng. Fract.
Mech. 74, 1093–1100.
21. Hu, X. (2004) Influence of fracture process zone height on
fracture energy of concrete. Cem. Concr. Res. 34, 1321–1330.
22. Karihaloo, B. L., Abdalla, H. M. and Imjai, T. (2003) A simple
method for determining the true specific fracture energy of
concrete. Mag. Concr. Res. 55, 471–481.
 2012 Wiley Publishing Ltd j Strain (2013) 49, 54–59
doi: 10.1111/str.12012
 H. Cifuentes, M. Alcalde and F. Medina : Size-independent fracture energy methods
23. Abdalla, H. M. and Karihaloo, B. L. (2003) Determination of
size-independent specific fracture energy of concrete from
three-point bend and wedge splitting tests. Mag. Concr. Res.
55, 133–141.
24. Petersson, P. E. (1980) Fracture energy of concrete: method of
determination. Cem. Concr. Res. 10, 78–89.
25. TC 187-SOC (2007) Final Report of RILEM Technical Com-
mittee TC 187-SOC: ‘Experimental determination of the
stress-crack opening curve for concrete in tension’.
 2012 Wiley Publishing Ltd j Strain (2013) 49, 54–59
doi: 10.1111/str.12012
View publication stats
26. Duan, K. (2003) Boundary effect on concrete fracture and
non-constant fracture energy distribution. Eng. Fract. Mech.
70, 2257–2268.
27. Hu, X. Z. and Duan, K. (2010) Mechanism behind the size
effect phenomenon. J. Eng. Mech. 136, 60–68.
28. RILEM (1990) Size-effect method for determining fracture energy
and process zone size of concrete. Mater. Struct. 23, 461–465.
59