Characterization of Ultra High Performance

Fiber Reinforced Concrete (UHPFRC) Tensile

Behaviour

Nicola Generosi(&), Jacopo Donnini, and Valeria Corinaldesi

Università Politecnica delle Marche, Ancona, Italy

[email protected]

Abstract. Ultra-high performance fiber reinforced concrete (UHPFRC) is

considered a promising material for many applications where high compressive
and tensile strength, small thickness and high energy absorption capacity are
required. However, although these materials were introduced in the mid-1990s,
a comprehensive investigation regarding its tensile characterization is still par-
ticularly challenging. Different tensile test setups have been used by many
researchers, in order to obtain reliable results, but today there are currently no
testing standards available that define test conditions, specimen geometry and
analytical procedures to fully characterize the tensile properties of UHPFRC.
In this study, the tensile behaviour of UHPFRC has been investigated with
direct tensile tests on dog-bone specimens, using a gripping system with rotating
boundary conditions. Tests have been performed in displacement control.
Digital Image Correlation (DIC) has been used to measure displacements,
deformations, number and width of cracks in experimental testing.
The effect of hooked steel fibers with diameter of 0.38 mm and length of
30 mm on the tensile behaviour of UHPFRC has been investigated, varying the
amount of fibers from 0% up to 2.6% by volume. The fiber volume fraction
greatly influences the tensile strength of the material, strain at peak strength,
energy absorption capacity and post-cracking behaviour.

Keywords: Digital image correlation
Fiber reinforced concrete
Steel fibers

UHPFRC

1 Introduction

The interest of the scientific community on the development of Ultra-High-Performance

Fiber Reinforced Concrete (UHPFRC) has been significantly increased in recent years.
UHPFRCs are a new generation of cementitious composites with excellent mechanical
properties. To achieve the ultra-high compressive strength, fine powders, such as quartz
flour and silica fume are normally used in mixture to ensure a highly dense matrix [1–4].
The benefits of fiber reinforcement in improving the fracture toughness, impact
resistance, fatigue endurance and energy absorption capacity of concrete are well
known. Fibers resist the nucleation of cracks by acting as stress-transfer bridges, and
once cracks nucleate, fibers abate their propagation by providing crack tip plasticity
and increased fracture toughness [5].

© RILEM 2021
P. Serna et al. (Eds.): BEFIB 2020, RILEM Bookseries 30, pp. 1068–1078, 2021.
https://doi.org/10.1007/978-3-030-58482-5_94
 Characterization of Ultra High Performance Fiber Reinforced Concrete (UHPFRC) 1069

UHPFRC can be classified in tension as “strain-softening” or “strain-hardening”. In

the case of higher fiber volume fraction (for example Vf > 2%), UHPFRC specimens
subjected to uniaxial tensile loads show a strain hardening behaviour after the formation
of the first crack. As a result, the tensile strength of the composite is higher than that at
first cracking [6]. Figure 1 illustrates a simplified response of strain-hardening Fiber
Reinforced Concrete (FRC). This idealized modelling approach distinguishes the tensile
behaviour into three parts: Part I (Elastic behaviour up to cracking strength cc); Part II
(Strain hardening behaviour with multiple cracking); Part III (Softening behaviour).

Fig. 1. Idealized response of strain-hardening FRC in tension

There are already few national recommendations on the Design and Construction
with UHPFRC: the French AFGC-SETRA [7] and the Japanese JSCE [8] both provide
recommendations on how to perform uniaxial tensile tests on UHPFRC materials. The
Fib Model Code 2010 [9], which serves as a basis for the future code for concrete
structures, also provides important information on the mechanical properties of Fiber
Reinforced Concrete (FRC) elements. The Italian CSLP (Consiglio Superiore dei Lavori
Pubblici) recently published a guideline about qualification, technical assessment cer-
tification and acceptance control of FRC. It also specifies how to test the flexural and
direct tensile strength of FRC. Further details on test methods can be found in [10].
UHPFRC material is under detailed exploration: many experts from all over the
world presented their research results regarding its tensile characterization. Wille et al.
[11] experimentally investigated the direct tensile behaviour of nine dog-bone different
UHPFRC specimens (by varying fiber type and fiber volume fraction). The analysed
results showed, as expected, a strong dependency on fiber volume fraction. In Yoo D.
et al. [12] the effects of steel fiber type on the tensile performance of UHPFRC were
investigated. Four different steel fiber types were used: S-straight, T-twisted, H-hooked,
and HH-half-hooked. The order of effectiveness in enhancing the tensile performance
of dog-bone specimens was S-fibers > T-fibers > HH-fiber > H-fiber.
Direct tensile tests are challenging to perform, since it is difficult to achieve
homogeneously distributed stress throughout the specimen cross section and to control
a stable load versus displacement/crack opening response [13].
Furthermore, specimen’s geometry and dimension, together with the gripping
methods used during the test, can significantly influence the test results. Different
1070 N. Generosi et al.

tensile test setups have been proposed by researcher to evaluate the direct tensile
behaviour of cementitious materials [11, 14, 15].
The objective of this research work is to investigate the mechanical properties of
UHPFRC reinforced with different amount of hooked steel fibers, from 0.6% up to
2.6% by volume. First, the material has been characterized by means of compression
and flexural tests. Then, direct tensile tests have been performed on dog-bone shape
specimens to fully characterize their tensile properties and to evaluate the effect of
different amounts of hooked steel fibers both on the tensile strength, deformation
capacity and softening or hardening behaviour.

2 Materials and Methods

A Commercial Portland-limestone blended cement (CEM I 52.5 R), in compliance with

EN-197/1 [16] was used. The Blaine fineness of cement was 0.48 m2/g and its specific
gravity was 3.15 g/cm3. As aggregate two different types of quartz sand with particle size
0–0.6 mm and 0.6–1.0 mm were suitably combined. Silica fume was added at a dosage of
125 kg/m3. In addition, an acrylic-based water-reducing admixture (WRA) was added in
powder at dosage of 1.1% by weight of cement with a water to cement ratio (w/c) of 0.20.
Hooked steel fibers (Fig. 2), 30-mm long with aspect ratio equal to 80, were added at
increasing dosages: 50, 100, 150 and 200 kg/m3. The corresponding fibers volume frac-
tions are 0.6%, 1.25%, 1.9% and 2.6%. The mixture proportions are reported in Table 1.

Fig. 2. Hooked steel fibers (a) and dog-bone specimens manufacturing (b)

Table 1. UHPFRC mixtures (kg/m3)

Specimen CEM I 52.5R Water Sand 0/0.6 Sand 0.6/1 Silica fume WRA Fibers
UHPC 1000 200 400 600 125 11 –
UHPFRC-0.6 50
UHPFRC-1.25 100
UHPFRC-1.9 150
UHPFRC-2.6 200
 Characterization of Ultra High Performance Fiber Reinforced Concrete (UHPFRC) 1071

3 Experimental Results
3.1 Compression and Flexural Tests (4  4  16 cm)
Three prismatic specimens were prepared for each concrete mixture, 4  4  16 cm in
size, by casting them in steel forms. After 2 days they were demoulded and kept up to 28
days in climatic chamber at 20 °C and 50% R.H. to test their behaviour in the worst
curing conditions. They were tested in bending and then in compression, according to the
procedure described in EN 1015-11 [17]. Even if this type of test method is specific for
mortars and it is not suitable for concrete samples, it was adopted in this study to obtain
preliminary information on the mechanical properties of the specimens by varying the
fibers dosage. Results obtained for the different UHPFRC mixtures are listed in Table 2,
while load-midspan displacement curves of flexural tests are reported in Fig. 3.

Table 2. Experimental results of compression and flexural tests on specimens with dimensions
of 40  40  160 mm3
Specimen Compression Increase of Flexural Increase of Flexural
strength compression strength flexural toughness Uf
(MPa) strength (MPa) strength (kNmm)
UHPC Average 98.5 – 13.8 – 1.14
CoV(%) 5 5 11
UHPFRC- Average 121.4 +23.3% 17.2 +24.8% 9.34
0.6 CoV(%) 5 7 3
UHPFRC- Average 124.0 +25.8% 27.2 +97.9% 16.23
1.25 CoV(%) 5 9 8
UHPFRC- Average 138.8 +40.9% 31.7 +130.4% 20.15
1.9 CoV(%) 4 5 5
UHPFRC- Average 143.4 +45.6% 43.6 +217.1% 28.63
2.6 CoV(%) 4 4 3

It can be observed that the reference UHPC cured in dry environment showed an
average compression strength of 98.5 MPa (a quite low value for a mixture with a w/c
of 0.20, due to the dry curing conditions).
The addition of hooked steel fibers was able to gradually increase both the com-
pression and the flexural strength of the concrete mixtures. Concerning the compressive
strength, the reason for this result can be attributed to the confinement effect of the fiber
reinforcement. The maximum increase was found for the mixture with 200 kg/m3 of
steel fibers (equal to about 2.6% by volume): compressive strength raised of about 45%
while flexural strength showed an increase higher than 200%.
Flexural toughness has been calculated by integrating the area under the load-
displacement curves of flexural tests, up to a midspan displacement equal to 2 mm. The
addition of steel fibers allows to considerably increase the flexural toughness of
UHPFRC specimens. This value raises by increasing the amount of fibers, up to a value
of about 28 kNmm for specimens reinforced with 2.6% of fibers. With respect to the
reference UHPC mixture without fibers, the UHPFRCs were able to absorb up to 25-
time higher energy amount with the highest dosage of fibers.
1072 N. Generosi et al.

UHPC - Flexural test UHPFRC-0.6 - Flexural test

20 20
18 18
A A
16 16
14 B 14 B
Load (kN)

Load (kN)
12 12
C
10 10 C
8 8
6 6
4 4
2 2
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 3.5 4
Midspan displacement (mm) Midspan displacement (mm)

UHPFRC-1.25 - Flexural test UHPFRC-1.9 - Flexural test

20 20
18 A 18 A
16 16
B
14 B 14
Load (kN)

Load (kN)

12 12 C
C
10 10
8 8
6 6
4 4
2 2
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 3.5 4
Midspan displacement (mm) Midspan displacement (mm)

UHPFRC-2.6 - Flexural test

20
18
16
14
Load (kN)

12
10
8 A
6 B
4
2 C
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Midspan displacement (mm)

Fig. 3. Load-displacement curves of flexural tests on UHPFRC specimens

 Characterization of Ultra High Performance Fiber Reinforced Concrete (UHPFRC) 1073

3.2 Direct Tensile Tests (Dog Bone Specimens)

Dog-bone specimens with a representative cross section of about 30 mm  45 mm
and with a total length of 330 mm were used to experimentally determine the tensile
stress-strain responses of UHPFRC (Fig. 4). Specimens were cast in wood frameworks
(Fig. 2), after 2 days they were demoulded and kept up to 28 days in climatic chamber
at 20 °C and 50% R.H. to test their behaviour in the worst curing conditions. A steel
frame realized with welded steel plates and four steel cylinders (Fig. 4) was used to
grab the specimen and to transfer the tensile force without applying compression at the
specimen ends. In this configuration rotations at the ends of the specimen are allowed.
Tensile tests have been performed on a tensile testing machine with a load capacity of
50 kN in displacement control, with a loading rate of 0.5 mm/min. Digital Image
Correlation (DIC) was used to measure displacements and deformations in the speci-
mens. During the tests, pictures of the frontal surface of the specimens have been
acquired by two digital cameras (model Pixelink® B371F) at 2 frame per second,
collecting about 600 images. The cameras were equipped with a lens having a focal
length of 16 mm and placed about 400 mm away from the specimen, in order to reduce
the perspective errors due to eventual out-plane motions. The specimen was illuminated
using an LED spotlight. A second camera, placed on the right side, was used to monitor
possible motions out of the plane. Tensile strain, e, was measured on a free length,
LDIC, equal to 100 mm as marked in red in Fig. 4b. Results of the tensile tests have
been reported in Table 3 (the average value of 3 specimens for each mixture and
relative Coefficient of Variation (%)), specimens after testing have been shown in
Fig. 6 and the stress-strain curves have been shown in Fig. 5. Tensile strength was
calculated by dividing the tensile load recorded by the testing machine by the speci-
men’s cross section, which has been calculated as the mean value of three sections at
both ends and the centre of the specimen (kept in the central part of the specimen with
total length of 80 mm). The first cracking stress rt was defined as the tensile stress in
the specimen corresponding to the first crack formation. The formation of the first crack
has been determined by visual inspection (with the help of DIC frames) or looking at
the stress-strain curves, in correspondence with a stress decrease in the elastic phase.
The average maximum tensile strength rmax and the corresponding deformation emax,
for each group of specimens were also reported in Table 3.
1074 N. Generosi et al.

(a) (b)

Fig. 4. Specimens dimension in mm (a), and direct tensile test setup (b)

Table 3. Experimental results of direct tensile tests on dog-bone specimens

Specimen rt (MPa) et (%) rmax (MPa) emax (%)
UHPC Average 5.65 0.037 5.65 0.037
CoV(%) 17 28 17 28
UHPFRC-0.6 Average 3.74 0.029 4.49 0.07
CoV(%) 17 31 14 19
UHPFRC-1.25 Average 4.90 0.035 5.40 0.49
CoV(%) 5 17 6 45
UHPFRC-1.9 Average 3.93 0.036 5.88 0.56
CoV(%) 14 25 17 5
UHPFRC-2.6 Average 6.14 0.062 8.11 0.87
CoV(%) 14 34 10 40

In uniaxial tensile tests, specimens without fibers (UHPC) showed a linear elastic
behaviour up to failure, with the formation of a single pass-through crack (Fig. 5 and
6). Failure was always brittle after the formation of one unique crack. The addition of
hooked steel fibers, even at the lowest dosage (0.6% by volume), is able to avoid brittle
failure and it allows the specimen to undergo plastic deformation after the formation of
the first crack. However, with fibers dosages of 50 and 100 kg/m3, the formation of the
first crack is reached at lower tensile stresses with respect to the plain UHPC specimen.
This is probably due to a bad dispersion and misalignment of the fibers within the
cross-section, thus reducing the resistant section: it might have constituted a discon-
tinuity, thus anticipating the formation of the first crack. Only by exceeding a critical
threshold on the fibers dosage (about 200 kg/m3) the addition of steel fibers is able to
increase the tensile strength at first cracking.
Characterization of Ultra High Performance Fiber Reinforced Concrete (UHPFRC) 1075

UHPC UHPFRC-0.6
10 10

A A
8 8
B B
6 6
σ (MPa)

σ (MPa)
C C

4 4

2 2

0 0
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
ε (%) ε (%)

UHPFRC-1.25 UHPFRC-1.9
10 10 A
A
8 8 B (failed)
B
C
6 C 6
σ (MPa)

σ (MPa)

4 4

2 2

0 0
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
ε (%) ε (%)

UHPFRC-2.6
10

6
σ (MPa)

4 A

B
2
C
0
0 0.5 1 1.5 2 2.5 3
ε (%)

Fig. 5. Stress-strain curves of direct tensile tests on dog-bone UHPFRC specimens

1076 N. Generosi et al.

It can be noticed that the post-cracking behaviour changes by varying the amount of
steel fibers. For lower dosages (50 and 100 kg/m3) a softening behaviour is observed
while with higher dosages (150 and 200 kg/m3) the post-cracking behaviour changes
and significant stress-hardening branches can be noticed in the stress-strain curves
(Fig. 5). This change from softening to hardening behaviour in direct tensile tests,
when a critical threshold in fibers dosage is exceeded, is confirmed by some studies
present in literature [11, 18].
Failure modes observed in UHPFRC specimens are different depending on the
amount of steel fibers, as shown in Fig. 6. UHPFRC specimens remain intact due to the
presence of the steel fibers while those without fibers failed immediately after the
formation of a unique pass-through crack. Specimens with lower fibers dosages
(UHPFRC-0.6 and UHPFRC-1.25) showed the formation of a big crack which grows
at increasing load. Only specimens with higher fibers dosages (UHPFRC-1.9 and
UHPFRC-2.6) showed the formation of multiple cracks. However, once the maximum
load is reached, only one of them continues to grow in the softening branch.

Fig. 6. Failure modes of dog-bone UHPFRC specimens under tensile loads

4 Conclusions

The objective of this study was to evaluate the tensile properties of UHPFRC mixtures,
reinforced with different amount of 30-mm long hooked steel fibers, varying from 0.6%
up to 2.6% by volume. Direct tensile tests have been carried out on dog-bone shaped
specimens, with total length of 330 mm and cross section of 45  30 mm2. Com-
pression and flexural tests have been performed on 40  40  160 mm3 specimens.
Experimental results allowed to compare bending and tensile properties for UHPFRC
specimens with different fiber dosages.
 Characterization of Ultra High Performance Fiber Reinforced Concrete (UHPFRC) 1077

The flowing concluding remarks can be drawn:

• The addition of hooked steel fibers allowed to significantly increase the compres-
sion and flexural strength of UHPFRC mixtures (up to 45% and more than 200%,
respectively, at the highest dosage of 2.6%).
• Flexural toughness has been greatly raised by increasing the steel fiber volume: with
respect to the reference UHPC mixture without fibers, UHPFRCs were able to
absorb up to 25-time higher energy amount with the highest dosage of fibers.
• The addition of steel fibers significantly modified the behaviour of UHPFRC
specimens subject to uniaxial tension, by avoiding their brittle failure even at low
dosages (0.6% by volume).
• The post-cracking behaviour in direct tensile tests was strongly influenced by the
amount of steel fibers, passing from softening to hardening if high amount of fibers
was used (higher than 1.9% by volume).
• This study confirms that the post-cracking behaviour in uniaxial tensile tests is
different from the post-cracking behaviour in bending tests. Once cracked, the
UHPFRC is still able to increase its flexural strength even at low fibers dosage while
its tensile strength can be increased only when high fibers volume is used.

References
1. Yu, R., Spiesz, P., Brouwers, H.J.H.: Mix design and properties assessment of Ultra-High
Performance Fibre Reinforced Concrete (UHPFRC). Cem. Concr. Res. 56, 29–39 (2014).
https://doi.org/10.1016/j.cemconres.2013.11.002
2. Yoo, D.-Y., Yoon, Y.-S.: A review on structural behavior, design, and application of ultra-
high-performance fiber-reinforced concrete. Int. J. Concr. Struct. Mater. 10(2), 125–142
(2016)
3. Song, Q., Yu, R., Shui, Z., Wang, X., Rao, S., Lin, Z.: Optimization of fibre orientation and
distribution for a sustainable Ultra-High Performance Fibre Reinforced Concrete
(UHPFRC): Experiments and mechanism analysis. Constr. Build. Mater. 169, 8–19
(2018). https://doi.org/10.1016/j.conbuildmat.2018.02.130
4. Corinaldesi, V., Donnini, J., Nardinocchi, A.: The influence of calcium oxide addition on
properties of fiber reinforced cement-based composites. J. Build. Eng. 4, 14–20 (2015)
5. Banthia, N., Nandakumar, N.: Crack growth resistance of hybrid fiber reinforced cement
composites. Cem. Concr. Compos. 25, 3–9 (2003). https://doi.org/10.1016/S0958-9465(01)
00043-9
6. Fantilli, A.P., Mihashi, H., Vallini, P.: Multiple cracking and strain hardening in fiber-
reinforced concrete under uniaxial tension. Cem. Concr. Res. 39, 1217–1229 (2009)
7. Bffup, A.: Ultra High Performance Fiber-Reinforced Concretes: Interim Recommendations:
Scientific and Technical Committee. Assoc Française Genie Civ (2002)
8. Japan Society of Civil Engineers. Recommendations for Design and Construction of High
Performance Fiber Reinforced Cement Composites with Multiple Fine Cracks (HPFRCC).
Concr. Eng. Ser. 82: Testing Method 6–10 (2008)
9. fib Model Code for Concrete Structures 2010 (2013). https://doi.org/10.1002/
9783433604090
10. Delle, C., In, S., Armato, C.: Consiglio Superiore dei Lavori Pubblici Servizio Tecnico
Centrale. Cemento 93, 62–00146 (2003)
1078 N. Generosi et al.

11. Wille, K., El-Tawil, S., Naaman, A.E.: Properties of strain hardening ultra high performance
fiber reinforced concrete (UHP-FRC) under direct tensile loading. Cem. Concr. Compos. 48,
53–66 (2014). https://doi.org/10.1016/j.cemconcomp.2013.12.015
12. Yoo, D.Y., Kim, S., Kim, J.J., Chun, B.: An experimental study on pullout and tensile
behavior of ultra-high-performance concrete reinforced with various steel fibers. Constr.
Build. Mater. 206, 46–61 (2019). https://doi.org/10.1016/j.conbuildmat.2019.02.058
13. Wille, K., Kim, D.J., Naaman, A.E.: Strain-hardening UHP-FRC with low fiber contents.
Mater. Struct. Constr. 44, 583–598 (2011). https://doi.org/10.1617/s11527-010-9650-4
14. Kamal, A., Kunieda, M., Ueda, N., Nakamura, H.: Evaluation of crack opening performance
of a repair material with strain hardening behavior. Cem. Concr. Compos. 30, 863–871
(2008). https://doi.org/10.1016/j.cemconcomp.2008.08.003
15. Jun, P., Mechtcherine, V.: Behaviour of strain-hardening cement-based composites (SHCC)
under monotonic and cyclic tensile loading: Part 1 - Experimental investigations. Cem.
Concr. Compos. 32, 801–809 (2010). https://doi.org/10.1016/j.cemconcomp.2010.07.019
16. EN 197-1. Cement - Part 1: Composition, specifications and conformity criteria for common
cements (2011)
17. EN 1015-11. Metodi di prova per malte per opere murarie Parte 11 : Determinazione della
resistenza a flessione e a compressione della malta indurita (2007)
18. Naaman, A.E., Reinhardt, H.W.: Proposed classification of HPFRC composites based on
their tensile response. Mater. Struct. Constr. 39, 547–555 (2006). https://doi.org/10.1617/
s11527-006-9103-2