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

net/publication/340838776

Design guidelines and optimization of ultra-high-performance fibre-

reinforced concrete blast protection wall panels

Article in International Journal of Protective Structures · April 2020

DOI: 10.1177/2041419620912751

CITATIONS READS

4 465

4 authors, including:

Mohtady Sherif Hesham Othman

Ryerson University Kinectrics
9 PUBLICATIONS 40 CITATIONS 25 PUBLICATIONS 421 CITATIONS

SEE PROFILE SEE PROFILE

Hesham Marzouk
Ryerson University
169 PUBLICATIONS 2,896 CITATIONS

SEE PROFILE

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

High-Strength Concrete Offshore Design View project

concrete materials and concrete structures View project

All content following this page was uploaded by Hesham Marzouk on 26 May 2020.

The user has requested enhancement of the downloaded file.

912751
research-article2020
PRS0010.1177/2041419620912751International Journal of Protective StructuresSherif et al.

Research Article

International Journal of Protective

Design guidelines and Structures

1­–21
© The Author(s) 2020
optimization of ultra-high- Article reuse guidelines:
sagepub.com/journals-permissions

performance fibre-reinforced
DOI: 10.1177/2041419620912751
https://doi.org/10.1177/2041419620912751
journals.sagepub.com/home/prs

concrete blast protection

wall panels

Mohtady Sherif1 , Hesham Othman1,

Hesham Marzouk1 and Hassan Aoude2

Abstract
Ultra-high-performance fibre-reinforced concrete is the latest generation of structural concrete, having
outstanding fresh and hardened properties; this includes the ease of placement and consolidation with
ultra-high mechanical properties, as well as toughness, volume stability, durability, higher flexural
and tensile strength, and ductility. As more research is being focused on it, the material behaviour and
characteristics are getting more understood, and the research demand for the special applications of the
ultra-high-performance fibre-reinforced concrete is growing higher. One special application that ultra-high-
performance fibre-reinforced concrete is thought to have an outstanding performance at is in the field of
protective structures, specifically against blast loads. This article presents part of a study that is concerned
with the behaviour and response of ultra-high-performance fibre-reinforced concrete wall panels under
blast load. Size and shape optimization techniques were combined in this study to optimize the design of a
200-MPa ultra-high-performance fibre-reinforced concrete under blast loads using finite element modelling.
This design optimization aims to maximize stiffness and minimize the cost while satisfying both design
stresses and construction requirements. The design variable to be optimized for are the thickness ranging
from 100 to 300 mm at 25 mm increments, in addition to the reinforcement ratio of 0%, 0.2%, 1% and 3%,
and aspect ratio of 1, 1.5 and 2; the boundary condition is four edges fixed and restrained. The numerical
simulation has been performed using an explicate finite element software package. The complete behaviour
of an ultra-high-performance fibre-reinforced concrete is defined using the concrete damaged plasticity
model. The concrete constitutive model has been developed considering the contribution of tensile
hardening response, fracture energy and crack-band width approaches to accurately represent the tensile
behaviour and guarantee mesh independence of results. The blast load is applied using the Conventional
Weapons method of the US Army Corps of Engineers that is readily available in the finite element software.
The validity of the numerical model used is verified by comparing numerical results to experimental data.

Keywords
UHP-FRC, optimization, precast concrete, protective panels, blast load, blast walls

1
Ryerson University, Toronto, ON, Canada
2
University of Ottawa, Ottawa, ON, Canada

Corresponding author:
Mohtady Sherif, Ryerson University, Toronto, ON M5B 2K3, Canada.
Email: Mohtady.Sherif@Ryerson.ca
2 International Journal of Protective Structures 00(0)

Introduction
The demand for structural protection for existing civilian structures against blast loading hazard
has grown higher over the last two decades. This raises sets of challenges that include, but not
limited to, constructability issues, construction time and its effect on the working hours of the
structure, the space occupied by the structural protection system applied, construction and mainte-
nance cost, and the effect of fragmentation and debris resulting from the protective structure during
incidence. Thus, the implementation of ultra-high-performance (UHP) materials combined with
accelerated construction techniques and the prefabricated/precast technologies in the protection
system becomes a necessity. From a structural perspective, the protection plan of a structure against
external blast loading hazard has two main components/levels of protection: the first level is the
minimum stand-off zone, which is the minimum distance a hazard can get close to the subject
structure, while the second level of protection is the protective structure assembled on the face of
the structure being protected (Figure 1).
UHP-FRC (fibre-reinforced concrete) is a special type of concrete having outstanding fresh and
hardened properties. These exceptional properties include, but not necessarily limited to, the ease
of placement and consolidation with ultra-high early and long-term mechanical properties, such as
toughness, volume stability, higher flexural and tensile strength, ductility and higher abrasion
resistance. Moreover, the low porosity of the UHP-FRC matrix enhances the durability, through
lower permeability, low chloride ion penetration, and higher freeze and thaw resistance (Graybeal,
2006; Kosmata et al., 2003; Naaman, 2007; Wille and Naaman, 2010). To achieve these properties,
coarse aggregate is eliminated, and only fine particles are used; the grain size distribution should
be optimized to densify the mix and improve rheology. Moreover, a superplasticizer or high-range
water reducer should be used to enhance rheology while maintaining the water to cement (W/C)
ratio as low as 0.2. Fibres, usually steel or synthetic fibres, should be added as a volumetric ratio
up to 2% to improve ductility and achieve higher tensile and flexural strength, and fracture energy.
The high compressive capacity and high flowability, as well as low porosity, are achieved by opti-
mizing the particle packing, water content and the use of chemical and mineral admixtures. All
these properties make it suitable for a wide array of applications (Graybeal, 2006; Naaman, 2007;
Wille and Naaman, 2010) and allows for the construction of sustainable and economic structures
with extraordinarily slim designs, for example, storage halls; thin-wall shell structures; and highly
loaded columns, repair overlay material and ductile structural joints.
On the material behaviour level, the research demand on UHP-FRC is growing higher; the
material behaviour and response, and how to identify it and capture the actual behaviour, as well
as the fracture behaviour, are of specific interest to researchers over the last decade. Moreover, as

Figure 1. The concept of minimum stand-off distance and protective panel.

Adapted from FEMA-426 (2003).
Sherif et al. 3

the material behaviour characteristics are getting more understood as more research is being
focused on it, some design codes and standards (i.e. German and Japanese codes) have started to
add new sections to address the different mechanical properties and limitations of this new mate-
rial. It has been found out experimentally at Ryerson University that the fracture energy of UHP-
FRC is about 100 times that of ordinary concrete (Wahba and Marzouk, 2012). The high-energy
absorption capacity of UHP-FRC, as well as its resistance to scabbing, spalling and fragmentation
(Li et al., 2015b; Millard et al., 2010; Riisgaard et al., 2006; Yi et al., 2012; Yu, 2015), enables the
construction of shield plates that play an essential part in protecting existing strategic buildings
against extreme loading conditions caused by blast, shock or impact loads (Cavill et al., 2006; Ellis
et al., 2014; Lee et al., 2016; Li et al., 2015a; Othman, 2016; Othman and Marzouk, 2016;
Rebentrost and White, 2011; Schleyer et al., 2010).
UHP-FRC is still a relatively expensive material; thus, it is mostly used in limited applications,
that is, in bridge deck joints and in other limited applications such as high-rise buildings, marine
and offshore structures. It is anticipated that UHP-FRC will find an increasing market as a thin
protective shield for structures in specific zones where superior properties are needed (Habel and
Gauvreau, 2008). A strong engineering evidence on low-speed impact tests done by Othman and
Marzouk (2016) on UHP-FRC slabs showed that UHP-FRC has more ductility and energy absorp-
tion due to the presence of fibres that bridge the cracks and thus also produces a more ductile
failure mode and no fragmentations than the panels of the same size and geometry made from
high-strength concrete (HSC) and normal-strength concrete (NSC). Moreover, the experimental
research conducted over the past decade (Cavill et al., 2006; Mao et al., 2015; Ngo et al., 2013;
Rebentrost, 2009; Slotz et al., 2014; Yi et al., 2012; Yu, 2015) showed that UHP-FRC has an out-
standing performance in the field of defensive structures and protective shields, specifically against
blast loads compared to HSC and NSC.
The work presented herein is part of a more extensive research programme that aims to provide
high-performance and cost-efficient protection for new structures or as an upgrade to the existing
structure against blast loading hazards through the implementation of the accelerated construction
technology, with UHP materials. Moreover, it is to provide the designers of protective structures
with a preliminary design guideline for the selection of the optimum UHP-FRC blast wall panel for
a specific blast environment and scaled distance. The current study aims to present a better under-
standing of the influence of the design parameters on the behaviour of UHP-FRC wall panels under
blast loading, to reach the optimum design dimensions for the blast wall panels and also to set the
foundation for further investigation of the response of different UHP-FRC wall panel assemblies.
The set objectives to fulfil the research aim are reaching the optimum panel dimensions and
reinforcement ratio that resulted in the least possible panel displacement while reducing the overall
material volume to be used, thus reducing the cost per panel. The panel dimensions targeted in this
study are the planer dimensions in terms of length (L) and width (W), presented in a more general
format as the aspect ratio (L/W). The other dimension variable being targeted is the thickness (t),
expressed in a more general form as the surface area (A) to thickness ratio, A/t.
The objectives of this research are achieved through conducting an optimization study using a
calibrated and experimentally verified finite element (FE) numerical simulation. The numerical
simulation is performed using the FE software package ABAQUS/Explicit (Simula, 2016), with a
concrete material model that considers the contribution of tensile hardening response, fracture
energy and crack-band width approaches to accurately represent the tensile behaviour and guaran-
tee mesh independence of results. The complete behaviour of UHP-FRC is defined using the con-
crete damage plasticity (CDP) model. The blast load is applied using the Conventional Weapons
(ConWep) method of the US Army Corps of Engineers (USACE, 2017) that is built into the Fe
software (Simula, 2016). The numerical model used is calibrated and verified for drop-impact load
4 International Journal of Protective Structures 00(0)

Figure 2. Schematic diagram of the difference between the tensile strain hardening response of UHP-
FRC and traditional strain-softening concrete: (a) UHP-FRC and (b) NSC or HSC.

against experimental test data (Othman and Marzouk, 2018) and validated for blast loading against
the available experimental data from research conducted by Yi et al. (2012).

Material properties of UHP-FRC

Generally, the post-peak tensile response of concrete has a significant effect on the accuracy of the
numerical results of FE modelling when compared to the post-peak compressive response (Grassl
and Jirásek, 2006). The tensile post-peak regime of UHP-FRC is different from NSC and HSC.
Figure 2 illustrates the difference between the tensile response of traditional strain-softening con-
crete and UHP-FRC (Wille and Naaman, 2010). The uniaxial tensile response of UHP-FRC is
characterized by linear elastic up to matrix tensile strength, followed by strain hardening behaviour
associated with several micro-cracks distributed in the entire volume of the bulk material until the
tensile strength is reached and a macro-crack is localized. Across this macro-crack, the force is still
transferred by fibres, and the stress decreases until full pull-out of fibres. The fracture energy (GF)
is another critical parameter defining the cracking and post-cracking behaviour of concrete, and it
is significant for any accurate FE analysis (Marzouk and Chen, 1993).

Analysis methods for structural elements subject to blast loading

There are two categories of nonlinear dynamic analysis methods: the analytical macro-models with
single degree of freedom (SDOF) or multiple degree of freedom (multi-MDOF), and the finite ele-
ment models (FEM), with the latter being more complex and requiring a large number of input
parameters, and good experience and judgement to obtain valid and reliable results.
In contrast to the FEM, the macro-SDOF and MDOF models use the fundamental mode of
vibration to predict the structural response and produce reasonably accurate results (El-Dakhakhni
et al., 2009; Razaqpur et al., 2009a, 2009b). Moreover, they are much more straightforward and
require fewer input data compared to FEM (El-Dakhakhni et al., 2009). Furthermore, SDOF and
MDOF could be used to check the overall accuracy of the FEM. A 2009 study (El-Dakhakhni et al.,
2009) stated that the MDOF method is more general than SDOF and can consider more modes of
vibrations than SDOF. Even though FEM is highly complex than SDOF or MDOF and requires
more input parameters, it is more accurate in capturing the response (Li et al., 2005). Moreover, in
terms of damage patterns and energy dissipation and failure modes, the FEM analysis has the upper
Sherif et al. 5

hand. Therefore, the FEM was implemented in the current study and throughout the overall research
programme.
In this study, numerical simulation has been performed using the FE software package ABAQUS/
Explicit (Simula, 2016), with a concrete material model which considers the contribution of tensile
hardening response, fracture energy and crack-band width approaches to accurately represent the
tensile behaviour and guarantee mesh independence of results. The behaviour of UHP-FRC is
defined using the concrete damaged plasticity model.
Some researchers conducted FEM analysis for different types of UHP-FRC (Mao et al., 2014),
modelled small-scale slabs with different fibre content, the fibres were not modelled explicitly but
incorporated within the stress–strain behaviour, and the shape of the stress–strain curve used was
that for a 2% fibre volume and did not consider the actual shape of the curve for 4% and 6% volu-
metric ratios. Therefore, the results of the modelling showed compliance with the experimental
results for the slabs with 2% fibre volume, but this was not the case for slabs with higher volume.
Thus, Mao et al. (2014) concluded that the exact stress–strain curve and curve shape for the speci-
fied fibre volumetric ratio shall be used; moreover, they recommended that fibres should be mod-
elled explicitly and their pull-out properties shall be accounted for and that this would require full
material characterization tests. Mao et al. (2014) used the experimental findings and results of blast
load tests on UHP-FRC one-way slab panels conducted by the same researchers to calibrate a
numerical FEM on LS-DYNA software, the model took into account the strain rate effect of UHP-
FRC and ignored that of steel reinforcement, and the mesh size convergence and sensitivity analysis
was carried out to the element size selected. Compared to test data, the model showed a reasonable
accuracy in predicting the response and damage, with the damage presented by the model being
slightly higher than that of the experimental results. Mao et al. (2014) concluded that their FEM
model could be used to reasonably analyse UHP-FRC members under blast loads.
Ellis et al. (2014) conducted a multi-scale FEM and used the experimental results of the same
study to validate the model by comparing the critical specific impulse, fracture patterns and the
displacement at the centre of the panel. Two length scales were used in the modelling: a multiple
length scale and a structural length scale. A hand-shaking scheme was employed to bridge the two
length scales. The validated model was used to conduct a parametric study to determine the effect
of quasi-static tensile strength and dissipated energy on the critical impulse, and it is found that as
the former parameters increase, the impulse increases, and it is suggested that as the material
designs that increase the dissipated energy is preferred than those that increase the quasi-static
tensile strength, improving the factors that increase the energy dissipation is more critical to
improving the resistance. It is also concluded that multi-scale modelling would, if calibrated well,
facilitate the analysis and design of UHP-FRC slabs better than single length scale modelling.
Finally, Ågårdh (1997) presented an FEM of FRC slabs under blast loading using the ‘Winfrith’
material model in the FEM software LS-DYNA and the ‘constitutive model for brittle cracking’ in
ABAQUS/Explicit. The analysis is validated against the experimental results of shock tube tests
conducted on 12 FRC slabs in four groups. The slabs were of 1.2 m × 1.2 m × 0.06 m, with the
volumetric ratio of steel fibres being 0.7% and orthogonally placed steel reinforcement resulting in
a reinforcement ratio of 0.97%. The concrete for slabs is of a mean fc = 86.1 MPa and the nominal
yield stress Fy = 400 MPa. The study concluded that the FEM analysis of FRC slabs produced rea-
sonably accurate results for displacement and that the strain rate effect is not significant.

Material constitutive models for FEM

Exact explicit modelling of UHP-FRC is highly complex, as the material itself is a composite mate-
rial, and the random distribution of fibres adds to the complexity. Accounting for the contribution of
6 International Journal of Protective Structures 00(0)

fibres could be done following one of the two approaches, either explicitly by modelling the fibres
themselves in a multi-scale modelling approach (Cavill et al., 2006; Millard et al., 2010) or by using
a more straightforward approach by considering the effect of the fibres to be part of the material
properties of the concrete matrix that is modelled as a homogeneous model (Cavill et al., 2006;
Razaqpur et al., 2009b), The fibre effect is considered in this study to be a material property and is
reflected in the fracture energy and the tensile stress–displacement behaviour where the fibres are
accounted for in tensile strength value and post-peak capacity (i.e. fracture energy).
Over the previous decade, several constitutive modules for UHP-FRC have been developed;
some of these models are presented by the Japan Society of Civil Engineers (JSCE) and the French
Association of Civil Engineering (AFGC), and both models showed good results when compared
to experimental results as tested by Yoo et al. (2016). Moreover, the models presented by Gowripalan
and Gilbert (2000) for compression and tension are adopted by the Australian code. Any of the pre-
mentioned models can be used as input for the constitutive model in the FE software. Another
approach is by obtaining the properties of the UHP-FRC experimentally and generating a best fit
approximate plot and using it as material input; this approach is used to obtain the material model
used in the FEM in the current study. Figure 3 presents the steel reinforcement material model,
while the UHP-FRC is presented in Figure 4 and is used in this study.

Figure 3. Adapted steel stress–strain model: (a) overall stress–strain behaviour and (b) plastic regime.

The UHP-FRC modelled in this study had a compressive strength of 200 MPa, tensile strength
of 15 MPa, elastic modulus of 50,000 MPa and Poisson’s ratio of 0.19 (Yi et al., 2012). The steel
reinforcement had a yield strength of 400 MPa, ultimate tensile strength of 650 MPa, and elastic
modulus of 200,000 MPa.
The fibres considered in this study are smooth steel needles of a diameter to length ratio of
0.2/13 mm, with a tensile strength of 2600 MPa, an elastic modulus of 205 GPa and a density of
7850 kg/m3. The fracture energy is the area under the stress–crack opening curve, where the maxi-
mum crack opening is limited to half the fibre length. In the literature, the reported fracture energy
values of UHP-FRC are ranging from 14,000 to 40,000 N/m (Tran et al., 2016; Voit and Kirnbauer,
2014; Wille and Naaman, 2010; Xu and Wille, 2015) in comparison with 100 and 160 N/m for NSC
and HSC, respectively (Marzouk and Chen, 1995). It is evident from this comparison that the frac-
ture energy of UHP-FRC is several orders of magnitudes of NSC and HSC, making it a promising
candidate for blast- and impact-resistant structures. In the current study, the used UHP-FRC has
fracture energy (GF) of 29,800 N/m. More details about the fracture energy can be found in Ågårdh
(1997) and Aoude et al. (2013).
Several materials, including concrete-like materials, exhibit different responses when subject to
higher rates of loading compared to their quasi-static response. NSC and HSC would experience
an increase in compressive and tensile strength, modulus of elasticity and fracture energy. This
Sherif et al. 7

Figure 4. Adapted UHP-FRC uniaxial relationships for concrete damage plasticity model: (a) uniaxial
compression parameters (left: stress–strain; right: damage) and (b) uniaxial tensile parameters (left:
stress–strain/crack width; right: damage).

behaviour is widely acknowledged as the strain rate effect. The strain rate effect depends mainly
on two sets of rate-dependent phenomena: intrinsic phenomena, including creep, pore water and
capillary behaviour, and temperature-related fracture behaviour; and the apparent phenomena,
including the confinement and inertia effects. The fact is that these phenomena can take place
either simultaneously or individually under higher rates of loading, and they differ from one con-
crete type to another (Cusatis, 2011; Ožbolt et al., 2006, 2011; Ožbolt and Sharma, 2012; Smith
et al., 2014).
To account for the strain rate effect in any analytical or numerical simulation for NSC and HSC, a
dynamic increase factor (DIF) is commonly applied to the stress–strain curve at different strain rates;
the DIF is a relationship between the dynamic value of a property and the quasi-static value of the
same property. A widely accepted approach to obtain the DIF is through Malvar and Ross’ (1998)
equations. Several researchers, however (Cusatis, 2011; Ožbolt et al., 2006, 2011; Ožbolt and Sharma,
2012), have disputed the accuracy of the DIF approach as it lacks the generality and does not include
the contribution intrinsically and the apparent rate-dependent phenomena separately.
Under dynamic loading, the intrinsic phenomena are different for every concrete type since it
depends on the microstructure and voids ratio within the matrix, and these vary from one concrete
type to the other. Moreover, the presence of pore water and its behaviour under dynamic load is
believed to be an essential intrinsic phenomenon, and since it depends on the degree of saturation
and relative humidity as well as the capillary and voids ratio, it is not the same for every concrete
type, and thus its effect should be considered separately in the analysis (Smith et al., 2014). In the
case of UHP-FRC, it can be said that pore water and voids are almost negligible (Graybeal, 2006;
Smith et al., 2014); thus, in contrast to NSC and HSC, the intrinsic rate-dependent phenomena for
UHP-FRC are almost negligible.
Several researchers (Ågårdh, 1997; Cotsovos and Pavlovic, 2008; Georgin and Reynouard,
2003; Li et al., 2015b; Orbovic et al., 2015; Othman and Marzouk, 2018; Smith et al., 2014) stated
8 International Journal of Protective Structures 00(0)

Figure 5. Yield surface and flow rule of the CDP model (Ngo et al., 2013; Wahba et al., 2012): (a) the
deviatoric plane and (b) the meridian plane.

that the strain rate effect of concrete-like material is dominated by the inertia and confinement
effects, especially at higher levels of loading rate. The characteristics of cracking patterns and
crack propagation for concrete materials are enhanced under higher rates of loading (Li et al.,
2015b; Smith et al., 2014). The presence of fibres in FRC including UHP-FRC tends to enhance
the cracking pattern and crack propagation under quasi-static loading, and thus it lessens the effect
of the apparent phenomena of strength enhancement at higher strain rates.
Thus, the DIF in its current form can only be reasonably applied to concrete ranging in strength
from 35 to 70 MPa, and DIF should be reduced for higher concrete strength (Kong et al., 2017;
Smith et al., 2014) and it will tend to overestimate the strength of the confined concrete (Ågårdh,
1997; Cotsovos and Pavlovic, 2008; Georgin and Reynouard, 2003; Orbovic et al., 2015), FRC (Li
et al., 2015b; Musmar, 2013; Zhou and Hao, 2008) or concrete with different properties and micro-
structure than the traditional concrete (Bhusari and Gumaste, 2017; Li et al., 2015b; Smith et al.,
2014). Thus, including the strain rate effect in the modelling of UHP-FRC under blast loading
would result in overestimating the capacity and the response (Li et al., 2015b). Several researchers
including the authors of this study conducted analytical modelling of UHP-FRC under blast load-
ing or higher loading rates neglecting the DIF or using it with the value of 1, which yielded reason-
able results compared to the experimental findings (Li et al., 2015b; Smith et al., 2014).

Recommended damage plasticity model for UHP-FRC

The understanding of the material failure mechanisms is of critical importance when it comes
to the accuracy and validity of FEM. Moreover, the implementation of the applicable failure
mechanism and the material failure model in a code to be used in the FEM analysis is the most
complicated part (Ågårdh, 1997). Most of the commercially available software packages with
built-in material models need validation and optimization against the experimental findings
since these models are based on ‘non-measurable’ parameters that often need calibration
(Ågårdh, 1997).
The concrete damage plasticity model presented in ABAQUS software and most widely used in
modelling concrete is a modified Drucker-Prager yield model. In the deviatoric plane (Figure 5),
the concrete damage plasticity model does not follow a circular shape but rather is modified by a
shape parameter Kc, which is a ratio of the second stress invariant for tension and compression at
the same hydrostatic stress. For the concrete damage plasticity model, the default value of Kc is 2/3
Sherif et al. 9

Table 1. Sensitivity analysis results (Othman and Marzouk, 2017).

Material parameter Significance Material parameter Significance

Tensile strength (ft)  Flow eccentricity (ε) 
Fracture energy (GF)  Shape parameters (Kc) 
Dilation angle (ψ)  Ratio (σbo/σco) 
Poisson ratio (υ)  Damage (dc, dt)) 

Figure 6. Test setup for the blast loading by Yi et al. (2012).

compared to Kc of 1 for Drucker-Prager of NSC. A UHP-FRC FEM model was calibrated under
impact load such that only the values for material parameters with a significant effect were consid-
ered; based on a sensitivity analysis conducted in a previous study by Othman and Marzouk (2017),
these parameters included fracture energy (GF), uniaxial tensile strength (ft) and dilation angle (ψ)
as shown in Figure 5. Other CDP parameters with marginal effect, including є, σ bo / σco and Kc,
are set to the default values of 0.1, 1.16 and 0.67, respectively. It was observed in the calibration
process that the influences of material parameters on the impact force and reaction results are gen-
erally limited. Therefore, only the results of midpoint displacement, midpoint pressure and accel-
eration were considered (Othman and Marzouk, 2014). The calibrated FEM model is validated
against blast loading. The exact experimental test setup and loading and boundary conditions are
modelled in the ABAQUS FEM for UHP-FRC (Table 1).

Blast plate testing validation for the FEM

The validation of the UHP-FRC FEM model under blast load is performed, such that only the val-
ues for material parameters with a significant effect were considered. Only the results of midpoint
displacement and midpoint reflected pressure were considered in the validation (Othman and
Marzouk, 2014). The experimental blast test used for validation was carried out by Yi et al. (2012),
and the model and the experimental sample were a 1000 × 1000 × 150 mm3 unreinforced UHP-
FRC slab of compressive strength of 210 MPa and Young’s modulus of 50,000 MPa; the experi-
mental slab specimen and the modelled slab were unreinforced. The slab is loaded with an
equivalent charge of 15 kg of TNT at a stand-off distance of 1.5 m (Figure 6). The load was applied
in ABAQUS/Explicit using the built-in ConWep module, and the same load and stand-off distance
were modelled in the software. It can be observed that the mid-point displacement and pressure for
10 International Journal of Protective Structures 00(0)

Figure 7. Displacement for UHP-FRC slab.

Figure 8. Reflected pressure on the top surface of the UHP-FRC slab.

Figure 9. Damage pattern (left) and plot of crack pattern for experimental UHP-FRC slab (left): (a)
model damage pattern, (b) RPC1 (Yi et al., 2012) and (c) RPC2 (Yi et al., 2012).

the FEM model fall within an acceptable range of the experimental results. The results of the vali-
dation for UHP-FRC are shown in Figures 7 and 8.
The exact experimental test setup and loading and boundary conditions are modelled in the FEM
for UHP-FRC. The CDP parameters used in the model validation were the same parameters recom-
mended by Othman and Marzouk (2017). The results of the validation for UHP-FRC are shown in
Figures 7 through 9. It can be observed that the mid-point displacement and pressure for the FEM
model fall within an acceptable range of the experimental results presented in Yi et al. (2012).
Sherif et al. 11

Figure 10. Blast environment.

Adapted from FEMA-426 (2003).

Blast load environment

The loading scheme presented by the Federal Emergency Management Agency (FEMA-426,
2003), as illustrated in Figure 10 for the unconfined explosion caused by different vehicles, could
be very useful for defining the design load based on the size of the vehicle and stand-off distance
of the vehicle from the structure which is, in this case, the closest distance the vehicle causing the
threat is expected to reach with respect to the targeted structure, as illustrated in Figure 1. The
stand-off distance should be known or predefined in the design phase as the first level of protection
by defining the zone at which trucks are not allowed to cross. The explosive environment chart
(Figure 10) presents five ranges of scaled distance up to Z = 2.37 m/kg1/3 which is a threshold that
would result in structural damage to walls and columns; this also means that any blast scenario that
falls below that range of scaled distance would result in a higher effect. The design optimization of
UHP-FRC protective panel is performed considering blast loads up to 420 kg equivalent charge
mass of TNT at a stand-off distance of 6 m; the load was chosen such that it gives a scaled distance
value of Z = 0.8 m/kg1/3 which is higher than the threshold that would result in a structural damage
to walls and columns which has a scaled distance of Z = 2.37 m/kg1/3, as shown in Figure 10.

Structural optimization of UHP-FRC blast wall panels

A manual size optimization technique is implemented in the current design optimization study
using the experimentally validated FEM model. Material optimization is achieved by choosing
UHP-FRC over the NSC or HSC, while other techniques of optimization are out of the scope of the
current study, such as shape optimization and topology optimization, as this study is focused on
protective wall panels that are to be used on the exterior of the structure against unconfined blast
loading; the panels are assumed to be rectangular, prismatic and of a flat surface.
The optimization design variables are the thickness, steel reinforcement ratio and length to
width aspect ratio (i.e. length/width ratio). The design optimization process is conducted for a
specific blast load environment that represents a scaled distance of Z = 0.8 m/kg1/3; a blast load up
12 International Journal of Protective Structures 00(0)

Figure 11. Flowchart of the optimization process (Othman et al., 2019).

to 420 kg equivalent charge mass of TNT at a stand-off distances 6 m is used as the design load. The
optimization design flow is illustrated in Figure 11.

Optimization objective function

Two main objective functions are targeted in this study. The first objective function is minimizing
the space between the protective wall panel and the protected exterior of the structure and is
achieved through minimizing the mid-point displacement while maintaining a decent amount of
damage as energy dissipation mean. The second objective function of minimizing the cost will be
achieved by minimizing the concrete volume (i.e. minimizing material usage) and using the mini-
mum steel reinforcement ratio.

Optimization constraints
The optimization process followed is a constrained optimization; constraints taken into considera-
tion for the design are constructability requirements as well as the minimum thickness and rein-
forcement of the requirements of the Canadian concrete design code (CSA A23.3, 2004), the
Canadian blast design requirements (CSA S850-12, 2017) and US Department of Defence report
(TM 5-1300, 1990). The minimum and maximum constraints applied to the thickness were from
100 to 300 mm at 25 mm increments, The second design variable, the reinforcement ratio, is con-
strained by investigating the reinforcement ratios of 0%, 0.2%, 1% and 2%, and the third design
variable, the length to width (L/W) aspect ratio, is constrained by the lower limit of L/W = 1 and
the upper limit of L/W = 2 and the step size of 0.25. The surface area of the panel of 4 × 106 mm 2
was kept constant (Figure 12).
Sherif et al. 13

Figure 12. Schematic illustration of the tested aspect ratio (L/W) geometry and dimensions: (a) L/W = 1,
(b) L/W = 1.25, (c) L/W = 1.5 and (d) L/W = 2.

Model geometry and boundary conditions

The blast wall panels were modelled as being rigidly fixed on all four edges using steel channels
that are rigid and non-deformable. Such supporting system was used as the effect of the boundary
condition is not part of this study. The load was applied on the blast panels using the ABAQUS
built in ConWep module; the charge was defined at a reference point located at the specified stand-
off distance from the centre of the panels.

Analysis results
The optimization process started with optimizing the thickness through running the analysis for
wall panels having an aspect ratio of L/W = 1.00 and 1.50 as presented in Tables 2 and 3, respec-
tively, this yielded an optimum range of 150–225 mm for L/W = 1.00, as shown in Figure 13, and
the range narrowed down to 150–200 mm for L/W = 1.50, as shown in Figure 14. As for the other
two groups having an aspect ratio of L/W = 1.25 and L/W = 1.75, the analysis was carried out for
the panels within the pre-determined optimum range of 150–200 mm, as presented in Tables 4
and 5, respectively.
Analysis of panels with aspect ratio L/W = 1.00 revealed that the thickness had a more signifi-
cant influence on improving the response and midspan displacement. Table 2 shows the maximum
displacement experienced by each wall panel. Moreover, the reinforcement ratios of 0.2% and 1%
were not significant in improving the performance nor in reducing the midspan displacement. The
optimum thickness range reached in this group was 150–225 mm.
14 International Journal of Protective Structures 00(0)

Table 2. Maximum displacement for slabs of aspect ratio L/W = 1.0.

Reinforcement Thickness (mm) 100 125 150 175 200 225 250 275 300
ratio
Thickness/Area (m−1) 0.025 0.031 0.038 0.044 0.050 0.056 0.063 0.069 0.075
0 Maximum 69.28 29.12 15.86 11.00 7.09 5.62 4.35 3.04 2.64
displacement (mm)
0.2 66.13 28.15 15.45 10.59 7.78 5.28 4.31 3.07 2.59
1 56.33 27.12 14.23 7.81 6.13 4.56 3.41 2.86 2.44

Table 3. Maximum displacement for slabs of aspect ratio L/W = 1.5.

Reinforcement Thickness (mm) 100 125 150 175 200 225 250 275 300
ratio (ρ, %)
Thickness/Area (m−1) 0.025 0.031 0.038 0.044 0.050 0.056 0.063 0.069 0.075
0 Maximum 45.3 20.9 12.0 8.9 5.9 4.4 3.9 3.1 2.4
displacement (mm)
0.2 42.0 20.2 11.8 8.4 5.5 4.3 3.6 2.6 2.1
1 40.2 19.0 11.3 7.6 5.1 3.8 3.4 2.7 2.3

Figure 13. The ratio of thickness to area versus maximum displacement for an aspect ratio of L/W = 1.00
for different reinforcement ratios.

On the other hand, panels of aspect ratio L/W = 1.50 narrowed down for walls in this group. The
thickness had a more significant influence on improving the response and midspan displacement.
Table 3 shows the maximum displacement experienced by each wall panel. Moreover, it is still
apparent that the reinforcement ration is not significant in improving the performance nor in reduc-
ing the midspan displacement. The optimum thickness range reached in this group was 150–
200 mm as shown in Figure 14 At this point, it was determined that the optimum panel thickness
for the specified scaled distance of Z = 0.8 m/kg1/3 would be around 150 mm, or in a more general
term, the optimum wall panel thickness would be of a thickness to surface area ratio of 0.044 m−1.
Two more rounds of analysis were carried out to obtain the optimum aspect ratio, for aspect
ratios of L/W = 1.25 and 1.75; for each aspect ratio, panels of thickness 150, 175 and 200 mm were
each having reinforcement ratio of 0%, 0.2%, 1% and 2%. Maximum displacement results for the
two sets are presented in Tables 4 and 5 and Figures 15 and 16. It was observed that as the aspect
Sherif et al. 15

Figure 14. The ratio of thickness to area versus maximum displacement for an aspect ratio of L/W = 1.5
for different reinforcement ratios.

Table 4. Maximum displacement for slabs of aspect ratio L/W = 1.25.

Reinforcement Thickness (mm) 150 175 200

ratio (ρ, %)
Thickness/Area (m−1) 0.038 0.044 0.050
0.0 Maximum displacement (mm) 13.9 10.0 6.3
0.2 13.6 9.9 6.2
1.0 11.8 8.2 5.2
2.0 10.2 6.6 4.4

Table 5. Maximum displacements for slabs of aspect ratio L/W = 1.75.

Reinforcement Thickness (mm) 150 175 200

ratio (ρ, %)
Thickness/Area (m−1) 0.038 0.044 0.050
0.0 Maximum displacement (mm) 8.922 6.618 4.467
0.2 8.663 5.957 4.265
1.0 7.576 5.389 4.431

Figure 15. The ratio of thickness to area versus maximum displacement for an aspect ratio of L/W = 1.25
for different reinforcement ratios.
16 International Journal of Protective Structures 00(0)

Figure 16. Thickness/area versus maximum displacement for an aspect ratio of L/W = 1.75 for different
reinforcement ratios.

Figure 17. The effect of aspect ratio L/W at different reinforcement ratios.

Figure 18. The effect of reinforcement ratio at different aspect ratios.

ratio increases, the maximum displacement decreases; moreover, the effect of the reinforcement
ratio on reducing the maximum displacement was almost negligible as shown in Tables 4 and 5.
The optimum range for thickness remained between 150 and 200 mm for a scaled distance of
Z = 0.08 m/kg1/3, or within a thickness to surface area ratio of 0.038–0.05, respectively. The opti-
mum thickness chosen is 150 mm for cost-effectiveness purposes.
Further processing of the simulation data showed that the optimum aspect ratio of the panel is
L/W = 1.75 as shown in Figure 16; moreover, it revealed that the reinforcement is not much signifi-
cant in improving the displacement. Thus, a minimum reinforcement ratio of 0.2% should be suf-
ficient (Figures 17 and 18).
Sherif et al. 17

Figure 19. Effect of reinforcement ratio on the damage the pattern for a 100-mm-thick slab with
L/W = 1: (a) ρ = 0% and (b) ρ = 1%.

Figure 20. Effect of aspect ratio L/W on damage pattern for 175-mm-thick slab and ρ = 1%: (a) L/W = 1,
(b) L/W = 1.5 and (c) L/W = 1.75.

Numerical damage patterns

The damage patterns obtained from the FEM in this study were mainly flexural damage patterns
resembling the flexural cracks along the yield lines as depicted in Figures 19 and 20. The damage
pattern was impacted by the reinforcement as unreinforced slabs tend to exhibit more localized
damage (Figure 19(a)), and as the reinforcement ratio increases, the damage pattern tends to be
more evenly spread over the area and less localized as depicted by Figure 19(b), although the
reinforcement ratio had a negligible effect on the response of the slab. The damage pattern was
consistent with the expected two-way action for aspect ratio of L/W = 1 as depicted by Figure
19(a) and (b) and Figure 20(a) and more towards the one-way action as the aspect ratio gets more
towards L/W = 2 as depicted by Figure 20. The damage pattern obtained complies with the yield
line theory (Kennedy and Goodchild, 2003) and further verifies the validity of the FEM model.

Conclusion
The shape and size optimization techniques are combined in the current study by minimizing maxi-
mum displacement and cost. The proposed technique of optimization can be considered as a
18 International Journal of Protective Structures 00(0)

valuable tool for the blast-resistant design. The adopted technique can be used to set a proper value
range of different dimensions for different types of protective panels under different scaled dis-
tance. The current study considered an unconfined blast environment, resembling a scaled distance
of Z = 0.80 kg/m1/3; the panel surface area is kept constant to facilitate the optimization of the main
design variables. From the results obtained in the current study, the following aspects can be
highlighted:

1. The CDP model, as well as the material constitutive model adopted by this study, could be
used for blast load analysis of UHP-FRC slabs.
2. FEM of UHP-FRC slabs under blast loading yielded consistent results when compared
with experimentally available data without considering the strain rate effect.
3. The damage pattern obtained from the FEM analysis complies with the yield line theory
(Kennedy and Goodchild, 2003) for one-way and two-way slabs, and thus further verifies
the validity of the FEM model.
4. For blast wall panels subject to unconfined blast loads of a scaled distance Z = 0.80 kg/m1/3,
the optimum range of thickness is between 150 and 200 mm, or within the thickness to
surface area ratio of 0.038–0.05. The optimum thickness recommended is 150 mm for cost-
effectiveness purposes under the considered loading environment and a panel surface area
of 4 m2. The optimum shape of the blast wall panel should be rectangular with a side length
aspect ratio of L/W = 1.75.
5. The effect of reinforcement ratio up to 2% on reducing the maximum midspan displace-
ment was almost negligible. However, the presence of reinforcement showed improvement
in the damage pattern and better distribution of the damage over the area.
6. As the aspect ratio L/W of the panel increases, the maximum displacement decreases.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publi-
cation of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publica-
tion of this article: This research program was financially supported by the Natural Sciences and Engineering
Research Council of Canada (NSERC) under the NSERC discovery grant number 51-52669.

ORCID iD
Mohtady Sherif https://orcid.org/0000-0003-3372-6185

References
Ågårdh L (1997) Fe-modelling of fibre reinforced concrete slabs subjected to blast load. Supplement au
Journal de Physique I11 7: 723–728.
Aoude H, Belghiti M, Cook WD, et al. (2013) Response of steel fiber-reinforced concrete beams with and
without stirrups. ACI Structural Journal 109: 359–368.
Bhusari JP and Gumaste KS (2017) A state of the art report on ultra high performance concrete: reactive
powder concrete. IOSR Journal of Engineering 7: 1–6.
Cavill B, Rebentrost M and Perry V (2006) ‘Ductal®: An Ultra-High Performance Material for Resistance
to Blasts and Impacts’. In: 1st Specialty Conference on Disaster Mitigation (CSCE), Calgary, Canada,
pp. 1–10.
Sherif et al. 19

Cotsovos DM and Pavlovic MN (2008) Numerical investigation of concrete subjected to high rates of uniaxial
tensile loading. International Journal of Impact Engineering 35: 319–335.
CSA A23.3 (2004) Design of concrete structures.
CSA S850-12 (2017) Design and assessment of buildings subjected to blast loads.
Cusatis G (2011) Strain-rate effects on concrete behavior. International Journal of Impact Engineering 38:
162–170.
El-Dakhakhni WW, Changiz Rezaei SH, Mekky WF, et al. (2009) Response sensitivity of blast-loaded rein-
forced concrete structures to the number of degrees of freedom. Canadian Journal of Civil Engineering
36: 1305–1320.
Ellis BD, Dipaolo BP, McDowell DL, et al. (2014) Experimenta and numerical investigation and multiscale
modeling of ultra-high-Performance concrete panels subject to blast loading. International Journal of
Impact Engineering 69: 95–103.
FEMA-426 (2003) Risk Management series: Reference Manual to Mitigate Potenial Terrorist Attacks
Against Buildings. Available at: https://www.fema.gov/media-library-data/20130726-1455-20490-6222/
fema426.pdf.
Georgin JF and Reynouard JM (2003) Modeling of structures subjected to impact: concrete behaviour under
high strain rate. Cement and Concrete Composites 25: 131–143.
Gowripalan N and Gilbert RI (2000) Design Guidelines for Ductal Prestressed Concrete Beams. Sydney,
NSW, Australia: VSL Australia.
Grassl P and Jirásek M (2006) Plastic model with non-local damage applied to concrete. International Journal
for Numerical and Analytical Methods in Geomechanics 30: 71–90.
Graybeal BA (2006) Material Property Characterization of Ultra-High Performance Concrete (Report no.
FHWA-HRT-06-103). Washington, DC: Federal Highway Administration.
Habel K and Gauvreau P (2008) Response of ultra-high performance fiber reinforced concrete (UHPFRC) to
impact and static loading. Cement and Concrete Composites 30: 938–946.
Kennedy G and Goodchild C (2003) Practical Yield Line Design. Available at: http://wsmurti.lecture.ub.ac.
id/files/2012/10/Perencanaan-Praktis-Garis-Leleh1.pdf
Kong X, Fang Q, Li QM, et al. (2017) Modified K&C model for cratering and scabbing of concrete slabs
under projectile impact. International Journal of Impact Engineering 108: 217–228.
Kosmata S, Kerkhoff B, Hooton R, et al. (2003) Design and Control of Concrete Mixture. Ottawa, ON,
Canada: Cement Association of Canada.
Lee SW, Choi SJ and Kim J-HJ (2016) Analytical assessment of blast damage of 270,000-kL LNG storage
outer tank according to explosive charges. Computers and Concrete 17: 201–214.
Li J, Wu C and Hao H (2015a) An experimental and numerical study of UHPFRC slabs under blast loads.
Materials & Design 82: 64–75.
Li J, Wu C and Hao H (2015b) Investigation of ultra-high performance concrete slab and normal strength
concrete slab under contact explosion. Engineering Structures 102: 395–408.
Li QM, Reid SR, Wen HM, et al. (2005) Local impact effects of hard missiles on concrete targets. International
Journal of Impact Engineering 32: 224–284.
Malvar L and Ross C (1998) Review of strain rate effects for concrete in tension. ACI Materials Journal 95:
735–739.
Mao L, Barnett SJ, Begg D, et al. (2014) Numerical simulation of ultra high performance fibre reinforced
concrete panel subjected to blast loading. International Journal of Impact Engineering 64: 91–100.
Mao L, Barnett SJ, Tyas A, et al. (2015) Response of small scale ultra high performance fibre reinforced
concrete slabs to blast loading. Construction and Building Materials 93: 822–830.
Marzouk H and Chen Z (1993) Finite element analysis of high-strength concrete slabs. ACI Structural Journal
90: 505–513.
Marzouk H and Chen Z (1995) Fracture energy and tension properties of high-strength concrete. Journal of
Materials in Civil Engineering 7: 108–116.
Millard S, Molyneaux T, Barnett S, et al. (2010) Dynamic enhancement of blast-resistant ultra high per-
formance fibre-reinforced concrete under flexural and shear loading. International Journal of Impact
Engineering 37: 405–413.
20 International Journal of Protective Structures 00(0)

Musmar M (2013) Tensile strength of steel fiber reinforced concrete. Contemporary Engineering Sciences
6: 225–237.
Naaman A (2007) High performance fiber reinforced cement composites: classification and applications. In:
CBM-CI international workshop, Karachi, Pakistan, pp. 389–401. Available at: http://citeseerx.ist.psu.
edu/viewdoc/download?doi=10.1.1.615.3856&rep=rep1&type=pdf
Ngo T, Mendis P and Whittaker A (2013) A rate dependent stress-strain relationship model for normal, high
and ultra-high strength concrete. International Journal of Protective Structures 4: 451–466.
Orbovic N, Sagals G and Blahoianu A (2015) Influence of transverse reinforcement on perforation resistance
of reinforced concrete slabs under hard missile impact. Nuclear Engineering and Design 295: 716–729.
Othman H (2016) Performance of UHPFRC plates under impact loads. PhD Dissertation, Ryerson University,
Toronto, ON, Canada.
Othman H and Marzouk H (2014) Numerical investigation of reinforced concrete slabs under impact load-
ing. In: Proceedings of the 10th fib International PhD Symposium in Civil Engineering (eds J Bastien,
N Rouleau, M Fiset, et al.), Québec, Canada, pp. 263–270. International Federation for Structural
Concrete (fib).
Othman H and Marzouk H (2016) Impact response of ultra-high-performance reinforced concrete plates. ACI
Structural Journal 113: 1325–1334.
Othman H and Marzouk H (2017) Finite-element analysis of reinforced concrete plates subjected to repeated
impact loads. Journal of Structural Engineering 143: 1–16.
Othman H and Marzouk H (2018) Applicability of damage plasticity constitutive model for ultra-high perfor-
mance fibre-reinforced concrete under impact loads. International Journal of Impact Engineering 114:
20–31.
Othman H, Sabrah T and Marzouk H (2019) Conceptual design of ultra-high performance fiber reinforced
concrete nuclear waste container. Nuclear Engineering and Technology 51: 588–599.
Ožbolt J and Sharma A (2012) Numerical simulation of dynamic fracture of concrete through uniaxial tension
and L-specimen. Engineering Fracture Mechanics 85: 88–102.
Ožbolt J, Rah KK and Meštrović D (2006) Influence of loading rate on concrete cone failure. International
Journal of Fracture 139: 239–252.
Ožbolt J, Sharma A and Reinhardt HW (2011) Dynamic fracture of concrete – compact tension specimen.
International Journal of Solids and Structures 48: 1534–1543.
Razaqpur AG, Contestabile E and Tolba A (2009a) Experimental study of the strength and deformations of
carbon fibre reinforced polymer (CFRP) retrofitted reinforced concrete slabs under blast load. Canadian
Journal of Civil Engineering 36: 1366–1377.
Razaqpur AG, Mekky W and Foo S (2009b) Fundamental concepts in blast resistance evaluation of struc-
tures. Canadian Journal of Civil Engineering 36: 1292–1304.
Rebentrost M (2009) Investigation of UHPFRC slabs under blast loads: previous research. In: Resplendino J
and Toulemonde F (eds) Designing and Building with UHPFRC. Hoboken, NJ: Wiley, pp. 1–8.
Rebentrost M and White G (2011) Investigation of UHPFRC slabs under blast loads. In: Toutlemonde F and
Resplendino J (eds) Proceedings of the designing and Building with UHPFRC: State-of-the-art and
development. Marseille: Wiley, pp. 363–376.
Riisgaard B, Gupta A, Mendis P, et al. (2006) Enhancing the performance under close-in detonations with
polymer reinforced CRC. Electronic Journal of Structural Engineering 6: 75–79.
Schleyer GK, Barnett SJ, Millard SG, et al. (2010) Modelling the response of UHPFRC panels to explosive
loading. Structures under Shock and Impact XI 113: 173–184.
Simula (2016) ABAQUS 6.14 User’s Manual. Available at: http://wufengyun.com/v6.14/books/usb/default.
htm?startat=pt05ch22s02abm03.html
Slotz A, Fischer K, Roller C, et al. (2014) Dynamic bearing capacity of ductile concrete plates under blast
loading. International Journal of Impact Engineering 69: 25–38.
Smith J, Cusatis G, Pelessone D, et al. (2014) Discrete modeling of ultra-high-performance concrete with
application to projectile penetration. International Journal of Impact Engineering 65: 13–32.
TM 5-1300 (1990) Structures to resist the effects of blast loads (NAVFAC P-397/ AFE 88-22).
Sherif et al. 21

Tran NT, Tran TK, Jeon JK, et al. (2016) Fracture energy of ultra-high-performance fiber-reinforced concrete
at high strain rates. Cement and Concrete Research 79: 169–184.
US Army Corps of Engineers (USACE) (2017) ConWep – Protective Design Center (PDC). Available at:
https://www.nwo.usace.army.mil/About/Centers-of-Expertise/Protective-Design-Center/PDC-Software/
Voit K and Kirnbauer J (2014) Tensile characteristics and fracture energy of fiber reinforced and non-rein-
forced ultra high performance concrete (UHPC). International Journal of Fracture 188: 147–157.
Wahba K and Marzouk H (2012) The Use of FBG Sensor to Determine The Fracture Energy Properties of
UHPFRC. In: CSHM-4, Berlin, Germany.
Wahba K, Marzouk H and Dawood N (2012) Fracture energy properties of ultra high performance fibre rein-
forced concrete. In: Annual conference of the Canadian society for civil engineering, Edmonton, AB,
Canada, 6–9 June.
Wille K and Naaman AE (2010) Fracture energy of UHP-FRC under direct tensile loading. In: Proceedings
of fracture mechanics of concrete and concrete structures-recent advances fracture mechanics of con-
crete-7, Seoul, South Korea, 23–28 May.
Xu M and Wille K (2015) Fracture energy of UHP-FRC under direct tensile loading applied at low strain
rates. Composites Part B: Engineering 80: 116–125.
Yi N-H, Kim J-HJ, Han T-S, et al. (2012) Blast-resistant characteristics of ultra-high strength concrete and
reactive powder concrete. Construction and Building Materials 28: 694–707.
Yoo D-YY, Banthia N and Yoon Y (2016) Flexural behavior of ultra-high-performance fiber-reinforced con-
crete beams reinforced with GFRP and steel rebars. Engineering Structures 111: 246–262.
Yu R (2015) Development of sustainable Protective Ultra-High Performance Fibre Reinforced Concrete
(UHPFRC): Design, Assessment and Modeling. Eindhoven: Technische Universiteit Eindhoven.
Zhou XQ and Hao H (2008) Modelling of compressive behaviour of concrete-like materials at high strain rate.
International Journal of Solids and Structures 45: 4648–4661.

View publication stats