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

net/publication/342277218

Fracture Energy properties of Ultra High Performance Fiber Reinforced Concrete

Conference Paper · June 2020

CITATION READS
1 133

3 authors:

Kassem M. Wahba Hesham Marzouk

6 PUBLICATIONS   86 CITATIONS   
Ryerson University
168 PUBLICATIONS   2,082 CITATIONS   
SEE PROFILE
SEE PROFILE

N. Dawood
Memorial University of Newfoundland
25 PUBLICATIONS   106 CITATIONS   

SEE PROFILE

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

Ultra High performance Concrete View project

Structural Health Monitoring on concrete Bridges View project

All content following this page was uploaded by Hesham Marzouk on 18 June 2020.

The user has requested enhancement of the downloaded file.

 3rd International Structural Specialty Conference
3ième conférence internationale spécialisée sur le génie des structures

Edmonton, Alberta
June 6-9, 2012 / 6 au 9 juin 2012

Fracture Energy properties of Ultra High Performance Fiber

Reinforced Concrete

K. Wahba1, H. Marzouk2, N. Dawood2

1
M.A.Sc Candidate at Ryerson University, Toronto, Ontario, Canada
2
Faculty of Engineering, Architecture and Science Ryerson University, Toronto, Ontario, Canada

Abstract: This paper discusses the results of an experimental investigation on the fracture
properties of Ultra-High Performance Fiber Reinforced Concrete (UHPFC), using fiber optics brag
grating, FBG, strain gauges rather than the conventional mechanical accelerometer. FBG
sensors are one of many fiber optic sensor technologies that are currently being used in structural
health monitoring systems. The sensors operate by detecting a shift in the wavelength of the
reflected maximum due to applied strain. Structurally, each conventional strain gauge has two
wires serving as input and output ports respectively, unlike fiber Bragg grating based sensors
where several sensors can be multiplexed onto the same optical fiber. Consequently, FBG
sensors could allow for a larger and a more accurate measurement for the fracture and cracking
response of concrete members. The fracture energy, GF, is defined as the area under the stress-
strain curve per unit fractured surface area, is experimentally determined for the UHPFC.
Fracture energy is important in determining design parameters such as shear strength and tensile
strength, obtained by the work-of-fracture method, are used in the development of models useful
in engineering practice.

1. Introduction

Ultra High Performance Fiber Reinforced Concrete, referred to as UHPFC henceforth, is the
name given to a class of materials that exhibits properties superior to those of conventional
concrete. Numerous researches have been conducted to determine the structural behavior of
concrete. Many authors suggest the fracture characteristics of concrete are of upmost importance.
Consequently, the fracture energy, GF, tensile strength, ft and the stress-CMOD relationship
completely describes the fracture characteristics of concrete (Hillerborg, Modeer, & Petersson,
1976). Tensile strength of concrete is an essential property, though it is not directly used in
design calculations for normal and high strength concrete. Nevertheless, not only is the tensile
strength is important also the tensile fracture behavior is of significance value for fiber reinforced
concrete. One way of quantifying fractural behavior of concrete is by means of fracture energy.

The fracture energy, GF, is defined as the area under the load-CMOD curve per unit fractured
surface area (Petersson, 1980). Researchers have advised, unlike other materials, energy
absorbed by concrete members in tension is associated with the descending portion of the stress-
displacement curve. A significant fractural parameter of concrete is the characteristic length, lch,
though it has no physical meaning it is a representation of material properties such as modulus of
elasticity Ec; tensile strength ft’ and fracture energy, GF, (Marzouk & Chen, 1995). A high value of
lch represents concrete that is less brittle. The brittleness behavior of concrete, made it almost
impossible to capture the fracture energy of concrete due to the sudden failure under flexural
stresses. However, when using UHPFC it is expected, due to the fiber matrix, that an inelastic
strain-hardening region exists. The region between the ends of linear elastic range and the peak

STR-1108-1
load denoted, region II, in Figure 1 is a result of multiple micro cracking. The linear elastic region,
region I, represents the micro cracking stage. The softening stage, region III, corresponds to the
single failure crack opening and is mainly controlled by the fiber pulling-out process ( Martin &
Stanton, 2007).

Figure 1: Stress-Strain Relation for UHPFC ( Martin & Stanton, 2007)

2. Experimental Procedure

2.1 General

High strength concrete, HSC, as we know it today is a result of ongoing research and invention
dated back to at least 30 years ago. It is widely used all over the world and is not limited to North
America. A new versatile material, UHPFC possesses desirable properties including strength,
ductility, durability and aesthetic design flexibility that are far superior to those of HSC. UHPFC is
specified where reduced weight is important or where architectural considerations require small
load carrying members

Ductal©, a family of UHPFC designed by Lafarge North America provides a unique combination of
superior properties (strength, durability, ductility and aesthetics) due to an integrated fiber matrix
with low porosity from a combination of fine raw materials selected for their optimized size
distribution and chemical characteristics. With improved resistance to abrasion, chemical
resistance, freeze-thaw, carbonation and chloride ion penetration, Ductal is well suited for
construction in aggressive environments. Ductal possesses superior material and structural
properties when compared to normal and high strength concrete. Table 1 summarizes the
general mechanical properties of Ductal Concrete as specified by Lafarge Canada.

Table 1: Mechanical Properties of Ductal Concrete (Lafarge North America)

Density 2500kg/m3
Compressive Strength on
150-180 MPa
75x150 cylinders
Direct Tensile Strength 8 MPa
Young’s Modulus 50 GPa
Poisson Ratio 0.2

STR-1108-2
2.2 Test Specimens Preparation and Testing

The mixing procedure was conducted according to Lafarge North America specification for
producing Ductal concrete. For confidentiality reason, the mixing procedure and proportions
cannot be specified within the paper. Following the mixing, the concrete was then transported, in
plastic buckets, to be casted in wooden molds previously prepared. The molds constructed had
dimension of 200 x 300 x 1000 mm. prior to casting, an artificial notch, a0=40mm, was prepared.
In addition, to measure the strain along the fracture zone, a fiber optic strain gauge, 150 mm, was
installed. Below is the specification of the constructed beams, Figure 2.

Figure 2: Test Beam Specifications

Literature review suggests strain or deflection control loading rate to correctly capture the fracture
energy of concrete (Marzouk & Chen, 1995). MTS testing machine used to apply the load in this
experiment is capable of load, strain and deflection loading rate. As for the strain measurements
a separate acquisition system was used to capture the change in wavelength from the fiber optics
cable, with a gauge length of 150mm, which then can be used to calculate the strain along the
cable. Both acquisition systems we programed to record measurements four times per second.
Figure 3 below shows the experimental test setup. The experiment was conducted at a
displacement rate of 0.0004 mm/s. Based on the results and the behavior observed from FE1, the
loading rate was increased to, 0.008 mm/s, as it became evident a slow rate is not required due
to the ductility of the test specimens due to the presence of the steel fibers.

(f) (a) MTS acquisition system

(c) (b) Fiber optic wavelength

acquisition system
(d)
(c) MTS machine
(a)
(d) Fiber optics cable
(b)
(f) Test specimen

Figure 3: Test setup and instrumentation

STR-1108-3
3. Test Results and Discussion

In general, concrete is a brittle material thus making the capturing of complete load-deformation
graphs a problematic task. Usually, a portion of the descending curve is captured then projected
to produce a complete load-deformation curve. However, when using UHPFRC, the addition of
the steel fiber matrix allows the concrete to behave with a more ductile property. Consequently,
the load-deformation is easily captured. Figure 4 and 5 below; show the load strain relationship of
the test beams FE1, FE2 and FE3. As shown in Figure 4, the load strain increased proportionally
up to approximately 90% of the ultimate load fu. Afterward, the strain increased dramatically while
a minor increase in the load was observed, explained by the elongation of the steel fibers
present.

In all three tested beams, surface cracks were observed at approximately 80-90 % of the ultimate
load. Subsequently, the strain gradually increased corresponding for crack opening, as the load
decreased. However, internal cracks formed far before reaching the 80-90% of the ultimate load
stage. These, internal cracks, would immediately propagate to the surface in the case of normal
or high strength concrete. Nevertheless, one of the main advantages of using UHPFRC is that
cracks are sealed and held tightly by the steel fibers present in the concrete mix.

120000
100000
Average Load, f (N)

80000
60000 FE1

40000 FE2

20000 FE3

0
0 0.005 0.01 0.015
Average Strain, εt

Figure 4: Average Tensile Strain Measurements of Beams FE1, FE2 and

FE3
1.2
1
0.8
Average Normalized Load,

0.6 FE1
0.4 FE2
FE3
f/fu

0.2
0
0 2 4 6 8
Average Normalized Strain, εt/εp

Figure 5: Average Normalized Strain Measurements of Beams FE1, FE2

and FE3

STR-1108-4
3.1 Fracture Energy

Fracture energy is the energy required to form a unit area of crack surface (Hillerborg A. , 1985).
Many researchers suggest that the fracture energy of concrete be treated as a material property.
Fracture energy, GF is considred to be the area under the stress strain-curve. However, when
dealing with concrete, unlike metal, the energy absorbed by conctere is only assoiated with the
area under the decending portion of the stress-strain curve (Marzouk & Chen, 1995). It must be
noted that the effect of the steel fiber must also be included in the fracture energy, as it ultimalty
gives UHPFRC it high compressive strength and tensile strength which are prameters that affect
the fracture energy.

The maximum effective strain, εmax, for HSC is reported to be around 16 times εp, strain at ultimate
load (Marzouk & Chen, 1995). However, when using UHPFRC it is observed that εmax is 4-6
times greater than εp. The fracture energy can then be claculated using the load-deformation
response using Eq1 (Japan Concrete Institute Standard, 2003). Table 2 summarizes the
mechanical properties of the test specimens.

[1] GF= (0.75W0+W1)/Alig

Where,

GF = fracture energy (N/mm)

W1= 0.75(Sm1/L +2m2)*g*CMODc

W0 = area below CMOD curve up to rupture of specimen (N.mm)

W1 = work done by deadweight of specimen and loading jig (N.mm)

Alig = area of broken ligament (b*(h-a0) (mm2)

m1 = mass of specimen (kg)

S = loading span (mm)

L = total length of specimen (mm)

m2 = mass of jig not attached to testing machine but placed on specimen until rupture
(kg)

g = gravitational acceleration (9.807 m/s2)

CMODc = crack mouth opening displacement at the time of rupture (mm)

Table 2: Mechanical Properties of UHPFRC

Specimen fc’ fu εp GF Et fr’

Name (MGa) (kN) x 10-6 (N/m) (GPa) (MPa)
FE1 162 70.5 1300 260.58 34 6.11
FE2 163 100.25 3500 1483.94 18 8.68
FE3 137 97.5 1700 1585.24 57 8.44

STR-1108-5
In general, fracture energy increases with the increases of the maximum coarse aggregate size.
Howeve, such an observation can not be applied to UHPFRC since there is no coarse aggregate
present. One,could make the argument of the fiber content having a signicicant affect on the
fracture energy. The fber content has a significant effect on the load-CMOD graph; increasing the
CMOD values significntly to those of normal or high strength concrete. The fracture energy of, GF,
of normal and high strength concrete, based on a directr tension test, are 110 N/m and 160 N/m
respectively (Marzouk & Chen, 1995). This effect of the fibers is reflected in the high fracture
energy values, Table 2. Its observed that UHPFRC yield fracture energy values that are much
greater than those of normal and high strengh concrete. It is clear however that FE1 fracture
energy results are imprecise and are not an accurate representation of the mechanical properties
of UHPFRC.

3.2 Characteristic length

Characteristic length, lch, is a material property representing the size of the fracture zone. It
expresses the fracture properties of concrete such as the modulus of elasticity, Ec; fracture
energy, GF; and tensile strength, ft, where ft is the direct tensile strength of concrete, as
determined by the direct tension test (Hillerborg A. , 1985) and is represented by Eq 2. While, lch,
has no physical meaning; it act as a representation of the brittleness of concrete, the higher the
lch value the less brittle the concrete.

[2] lch=Ec GF/ ft2 (mm)

Given the difficulties associated with fracture energy determination of concrete, researches have
proposed empirical equations to predict the characteristic length using the compressive strength
of concrete, Eq 3 (Hilsdorf & Brameshuber, 1991) and Eq 4 (Barr, Lydon , & Zhou, 1995).

[3] lch = 600(fc’)-0.3 (mm)

[4] lch = -3.84fc’+580 (mm)

Table 3, below shows a comparison between chractristic length, derived from fracture energy Eq
2, and avilable models, Eq 3 and 4. In general it is understandable that the brittleness of concrete
increases with the increase of compressive strength. Available models, Eq 3 and 4, follow that
thearoy. The models available do not accurtly predict the chractristic length of concrete when
compared to vlaues drived from Eq 2. However the models are conservitive, yeilding values
resulting in a higher brittleness assumption. Needless to say, UHPFRC does not follow that trent.
The significantly increased strength suggest a very brittle material, however the addition of the
fiber matrix results in otherwise a very ductile material. Consequenlty, resulting in inaccurate lch
values. This behaviour is also demenostrated in the fracture behaviour of the test beams, Figure
6. Typically, in conventional concrete with no reinforcement, a crack results in a sudden failure.
However, in UHPFRC, as seen in Figure 6, the test beam remian intact even after crack
propogation to the surface.

Figure 6: Crack Development During Testing of FE1

STR-1108-6
 Table 3: Characteristic length of Concrete Ranging from 40-165 MPa

compressive strength lch, Eq2 lch, Eq3 lch, Eq4

(MPa) (mm) (mm) (mm)
40 500 198.39 426.4
55 742 180.31 368.8
57.8 532 177.65 358.04
58.7 489 176.83 354.59
61 649 174.80 345.76
63 503 173.12 338.08
74 478 164.96 295.84
75 394 164.29 292
137 231 137.13 53.92
162 647 130.40 -42.08
163 693 130.16 -45.92

4. Applications of Fracture energy

While the fracture energy is not directly used in the design of concrete structures, significant
studies have been conducted to incorporate the fracture energy and behavior. Though it is not a
representation of any physical property, lch, represents the brittleness of concrete. Numerous
researches have gone into incorporating the characteristic length into design equation. For
concrete beams without stirrups a model was developed by the Rilem Technical Committee
represented by Figure 7, where fv is the shear strength, ft’ is the direct tensile strength, d is the
depth of the beam and ρ is the flexural reinforcement and lch is the characteristic length (Hillerborg
A. , 1985).

Figure 7: Theoretical Shear Strength of Reinforced Concrete

Beams Without Shear Reinforcement (Hillerborg A. , 1985).

STR-1108-7
A later model was developed for shear strength of high strength concrete slabs represented by
Eq 5 where c is the length of a side of a square column, d is the effective depth of slab, h is the
total slab thickness, ft is the uniaxial tensile strength of concrete and lch is the characteristic
length (Marzouk, Emam, & Hilal, 1998)

.
[5] 𝑣 0.88 𝑓 ρ

In design equation it is generally assumed that the shear strength is proportional to the tensile
strength of concrete. However, many researchers argue, the use of the square root of the
compressive strength to represent the tensile strength of concrete is not adequate when using
HSC, HPC or UHPFRC. Hence, a major advantage is to incorporate the fracture energy within
design equations.

5. Conclusion

UHPFRC, exhibits superior qualities compared to normal and high strength concrete. The
introduction of fiber reinforcement proves to increase, significantly, the ductility of the mix;
resulting in a concrete material that behaves rather different than conventional concrete. High
strength concrete has a more brittle behavior to that of normal strength concrete, demonstrated
by the sharp descends in the stress-strain curve. UHPFRC exhibits a far more ductile behavior
when compared to normal and high strength concrete. This ductile behavior and the increased
strength ultimately result in concrete with significantly higher fracture energy, consequently a
higher characteristic length.

Acknowledgments

This research project was conducted at Ryerson University as a partial fulfillment of the first
author’s M.A.Sc degree. The author would like to acknowledge the Natural Sciences and
Engineering Research Council for financial support of the research. The support of Lafarge North
America, providing Ductal© concrete dry materials, is greatly appreciated. In addition the support
of Dr. Marzouk, Dr. Dawood, Mr. Hughi, Mr. Sucic and the technical staff at Ryerson University
are greatly appreciated.

References

Martin, J., & Stanton, J. (2007). Experimental Testing to Determine Concrete Fracture Energy
Using Simple Laboratory Test Setup. ACI Materials Journal, 575-584.

Barr, B., Lydon , F., & Zhou, P. (1995). Fracture properties of high strength concrete with varyunf
silica fumes content and aggregates. Cement and Concrete Research, 543-553.

Hillerborg, A. (1985). The Theoretical Basis of a Method to Determine the Fracture Energy GF of
Concrete. Lund: BORDAS-GAUTHIER-VILLARS.

Hillerborg, A., Modeer, M., & Petersson, P. (1976). Analysis of crack formation and crack growth
in concrete by means of fracture mechanics and finite elements . Cement and Concrete
Research, 773-782.

Hilsdorf, H., & Brameshuber, W. (1991). Code-type formulation of fracture mechanics concepts
for concrete. International Journal of Fracture, 61-72.

Japan Concrete Institute Standard. (2003). Method of test for fracture energy of concrete by used
of notched beam JCI-S-001-2003. Japan Concrete Institute Standard.

STR-1108-8
 Lafarge North America. (n.d.). Retrieved May 18, 2011, from Ductal: http://www.ductal-
lafarge.com

Marzouk, H., & Chen, Z. (1995). Fracture Energy and Tension Properties of High Strength
Concrete. Materials in Civil Engineering , 108-116.

Marzouk, H., Emam, M., & Hilal, M. (1998). Sensitivity of shear strength to fracture energy of high
stregnth concrete slabs. Canadian journel of civil Engineering, 40-50.

Petersson, P. (1980). Fracture Energy of Concrete: Method of Determination. Cement and

Concrete Research, 79-89.

Rao, G., & Prasad, B. (2002). Fracture energy and softening behaviour of high-strength concrete.
Cement and Concrete Research, 247-252.

STR-1108-9

View publication stats