Laboratory study on the influence of casting on properties of ultra-high performance fibre reinforced concrete (UHPFRC) specimens.
Zofka, Adam ; Paliukaite, Migle ; Vaitkus, Audrius 等
Introduction
The high compressive strength as well as its relatively low cost
and versatile construction applications have made concrete one of the
most widely used structural materials of our time. However, its
inadequacies in tension limit and its use leave it susceptible to
unpredicted failures. Recent studies on Ultra-High Performance Concrete
(UHPC) have shown significant improvement in mechanical properties
including compressive and tensile strength (Graybeal 2006; Graybeal et
al. 2003). Research performed on Fibre Reinforced Concrete (FRC) has
also shown higher levels of tensile strength, along with a higher
resistance to cracking due in large part to the inclusion of fibres
(Zollo 1997). The combination of UHPC and FRC resulted in Ultra-High
Performance Fibre Reinforced Concrete (UHPFRC), which consists of a very
dense matrix with high quantities of fibre reinforcement (Graybeal 2007;
Gribniak et al. 2011; Corinaldesi, Moriconi 2012; Islam et al. 2012).
With research efforts across the world currently focused on UHPFRC, it
is vital to find the optimal method to cast UHPFRC samples and elements
in the laboratory and in the pre-cast facility in order to maximize the
use of the fibre reinforcement and to achieve desirable effects.
For the last 30 years, Fibre Reinforced Concrete has been a major
focus for many researchers aiming to maximize mechanical properties.
Research began by improving mechanical properties of the concrete with
the addition of randomly dispersed fibres. Studies on fibre addition
showed a significant improvement in flexural tensile strength of
concrete as well as post-peak tensile softening (ductility) (Barris et
al. 2012). Other studies investigated the effect of different fibre
properties - fibre content, length, material types, spatial orientation,
and shape--on concrete performance (Kang et al. 2011; Vejmelkova et al.
2010; Kim et al. 2011; Hassan et al. 2012; Grinys et al. 2013; Pajak et
al. 2013). The results of these studies have shown that flexural tensile
strength of UHPFRC increases linearly with respect to the fibre volume
ratio from 0% to 5%, and that strength parameters are also linearly
dependent on fibre content (Kang et al. 2011). Fibre types were
investigated at low volume by Yao et al. (2003), who found that
carbon-steel hybrid fibres give concrete the highest strength and
flexural toughness. Modifications of fibres by physically shaping them
to create a better mechanical bond and to enhance ductility have also
been researched. By enhancing fibre parameters, Wille et al. (2010)
found that a deformed fibre volume of only 1.5% can result in a tensile
strength of 13 MPa [1.89 ksi] and a strain-capacity improved from 0. 3%
to 0.6% compared to straight steel fibres.
Orientation and distribution of fibres are considered the biggest
contributors to maximizing tensile strength in UHPFRC (Ferrara et al.
2011; Kang et al. 2011; Dai et al. 2012). There are several variables
that affect both the orientation and distribution of fibres when casting
UHPFRC specimens: amount, size and shape of fibres, workability of the
matrix, method of compaction, as well as specimen size and shape (i.e.
mould shape). Altering any one of these variables can distribute and
orient the fibres differently. For example, an excessive amount of
fibres would cause fibres to interact with each other and disallow
proper distribution. If there is too much workability or if the
compaction process is too long, then the fibres will settle and orient
themselves horizontally. Different casting moulds will distort fibres
differently and, therefore, tensile strengths will differ for the same
UHPFRC material (Chanvillard, Rigaud 2003). There is a direct positive
correlation between the number of fibres parallel to the tensile force
and the strength of the specimen (Lee, Kim 2010), making this the main
goal when casting fibre-reinforced specimens. Even the location within
the mould, at which the matrix is poured in, has an effect of how fibres
are oriented due to the direction of flow. Barnett et al. (2010) found
that fibres tend to align perpendicular to the direction of flow, making
the centre the best location to pour the matrix into moulds. Perfecting
and understanding how to properly distribute and orient fibres during
laboratory casting will allow for a smoother transition when preparing
larger structural elements in pre-cast facilities or directly on a
construction site.
This study focuses on comprehending the effects of the UHPFRC
cylinder casting on fibre distribution and orientation by employing
three distinct approaches: density measurements, engineering fracture
testing and X-ray CT.
1. Materials and methods
1.1. Materials
In this study, samples were prepared from a special UHPFRC
manufactured by Lafarge. This material consists of premix, water, steel
fibres, polypropylene fibres and a high-range water-reducing admixture
(HRWA). The premix consists of Portland cement, silica fume, ground
quartz and fine sand. The hybrid high-carbon fibres are ranging in
length from 13 to 15 mm [.511 to .590 in.] with a diameter of 0.2 mm
[.008 in.] and have concentration of 2% by volume within the UHPFRC. A
breakdown of other components is shown in Table 1, which follows the
mixture recipe recommended by the producer.
The total of six 150 mm [5.90 in.] UHPFRC cylinders were prepared
from the mix shown in Table 1, containing a water to cement ratio of
0.18. The mixing procedure was based on the producer recommendations,
with some minor adjustments due to the conditions in the laboratory.
First, all components of the mix were weighed and placed in the mixer
pan and mixed for two minutes. Water, together with half the amount of
HRWA, was added slowly over four minutes, followed by one minute of
mixing. The second half of HRWA was then added over 30 seconds, followed
by another minute of mixing. The accelerator was then added over the
next minute, followed by another minute of mixing, or until the material
became a thick paste. Fibres were then added to the mix slowly over two
minutes, and then mixed for a minute, or until the fibres were well
dispersed. The total time for this process never exceeded 35 minutes.
Once the fibres were visually found to be well dispersed, the UHPFRC mix
was removed from the mixers and tested for flow. The ASTM C1437 (2013)
was used to check the flow of each batch, and it was found that each
batch exceeded the capacity of the flow table.
Specimens were then cast using a vibratory table, which vibrated at
60 Hz with adjustable amplitudes. Since the used mix had a high flow,
the vibrations were kept to a minimum to avoid separation of steel
fibres. The cylinders were next cast in three to four lifts in 150 mm
[5.90 in.] plastic moulds. The casting for all specimens was completed
within 30 minutes from its removal from the mixer.
Subsequent to completion of casting, specimens were covered in a
heavy plastic sheet and left to set for 48 hours before being removed
from the mould. They were then placed in a steam cure box. The curing
process began at room temperature and ambient relative humidity (RH),
and was brought to the target levels of 90[degrees]C [194[degrees]F] and
95% RH typically within 90 minutes. The total curing time was 48 hours,
after which specimens were removed from the steam cure box and placed
back in laboratory conditions to cool down.
1.2. Sample preparation
Before cutting the test samples from the cylinders, a testing
matrix was constructed to randomize sample locations (zones) for the
further investigation. The cylinders were cut into three zones (top,
middle, bottom) to allow for analysis on the fibre distribution and
orientation, and to differentiate fracture properties as a function of
cylinder height. The slice thicknesses required for fracture tests are
25 mm [.984 in.] and 37.5 mm [1.48 in.], which lead to one 25 mm [.984
in.] and two 37.5 mm [1.48 in.] slices per each cylinder (Fig. 1). Table
2 shows the detailed matrix with randomized locations for all tests.
Since two test samples are produced from each 25 mm [.984 in.] slice,
there were four 37.5 mm [1.48 in.] samples and four 25 mm [.984 in.]
samples created for each zone from all six cylinders. It should be noted
that half the specimens were heat treated after the cutting in order to
introduce additional factor to the analysis of the fracture energy.
To prepare the cylinders for cutting, 10 mm [.394 in.] slices from
both ends of the cylinders were cut off using a diamond wet saw with a
special clamping fixture to hold specimens and to ensure parallel cut
planes. Specimens were then cut from the top down as prescribed by the
test matrix shown in Table 2. Once the slices were obtained from the
cylinders, fine cutting was done to geometrically prepare each sample
for their appropriate testing.
[FIGURE 1 OMITTED]
Three separate experimental approaches were undertaken in order to
investigate the influence of sample preparation and zone location on
fibre orientation. The first approach evaluated the density of the
UHPFRC at different locations within the cylinder. The densities were
calculated by measuring masses and dimensions of all samples from
different locations. The second approach observed the differences in
fracture energy at different locations of the casted specimens, both in
tension and compression. Fracture energies were determined from the
Disc-Compact Tension Test (DCT) for tension (37.5 mm [1.48 in.]
specimens) and the Semi-Circular Bending Test (SCB) for compression (25
mm [.984 in.] specimens). Lastly, X-ray tomography scans were obtained
to qualitatively and quantitatively capture the differences in the
orientation and distribution of fibres within three different locations,
i.e. top, middle and bottom.
As mentioned before, half the specimens designated for each test
were heat treated in the ignition oven. This treatment created another
factor in studying fracture properties of the UHPFRC specimens and in
particular allowed to measure their residual properties. Specimens
assigned for the heat treatment were heated from room temperature to the
target temperature of 600[degrees]C [1112[degrees]F] at the heating rate
of 5[degrees]C [41[degrees]F]/minute. Once the specimen reached the
target temperature, they were held at that temperature for 6 hours.
After that period, specimens were removed from the oven and cooled at
room temperature before testing.
1.3. Fracture testing
Fracture testing for both SCB and DCT were performed in a
servo-hydraulic 100-kN load frame with a closed-loop computerized data
acquisition system. All tests were performed at ambient temperature and
within 3 weeks after specimen preparation.
The DCT test was originally developed for testing fracture
properties of metals but has never been used to test UHPFRC. In this
study, the fracture energy was determined using a 37.5 mm [1.48 in.]
thick specimens as opposed to 50 mm [1.968 in.] asphalt specimens.
Thinner samples are sufficient for testing UHPFRC since such a concrete
has significantly finer aggregate structure as compared to the asphalt
concrete. DCT specimens were cut to a disc-like geometry with a
pre-crack and two pull-points as shown in Figure 2.
[FIGURE 2 OMITTED]
The SCB is another test used to evaluate fracture properties of
different materials but similar to the DCT test, was yet to be used on
the UHPFRC specimens. In this study, slices were cut to thickness of 25
mm [.984 in.], and then cut diametrically to create two semicircular
specimens. A pre-crack notch was then created in the middle of the
specimen similar to the geometry shown in Figure 3.
Before testing, each specimen was accurately measured to properly
calculate the fracture area once the testing was finalized. At the
beginning of a test, a crack mouth opening displacement (CMOD) sensor
was placed on the pre-crack and 0.35 kN seating load was applied to the
specimen. Once the load reached 0.35 kN, the actual test started in the
closed-loop mode, i.e. the load is constantly adjusted in order to
maintain a constant CMOD rate. The CMOD rates for DCT and SCB were 1 mm
[.039 in.]/min and 0.3 mm [.012 in.]/min, respectively. This rate
continued until the load fell below 0.5 kN in the softening region of
the fracture curve or if the CMOD opened beyond the working range of the
gauge (approximately 7 mm [.275 in.]). The fracture energy [G.sub.f] was
then computed for both DCT and SCB tests using Eqn:
[G.sub.f] = [integral] [P/[A.sub.lig]] du,
where: [G.sub.f]--fracture energy (J/[m.sup.2]); P--load (kN);
u--CMOD opening (m); [A.sub.lig]--area of ligament ([m.sup.2]).
In order to calculate the total fracture energy, the load-CMOD
curves were extrapolated to obtain energy beyond the end of the test,
i.e. between last recorded load value and zero force. After preliminary
trials, it was determined that an exponential equation provides the best
extrapolation fit.
[FIGURE 3 OMITTED]
2. Results and discussion
To investigate the impact of fibre orientation in the UHPFRC
specimens as the function of zone location within a cylinder, three
aforementioned approaches were employed. The results from each approach
are presented in next three paragraphs.
2.1. Density distribution
It was hypothesized that distribution of fibres along the height of
the cast cylinders is not uniform, which results in varying quantity of
fibres and/or differing fibre orientations between cylinder locations.
The bulk density approach seemed to be very practical as it calculated
only the density at different locations of the specimens. The density
results are shown in Figure 4. Although there is a linear trend showing
a decrease in average densities with height, statistical analysis of the
means showed the location factor was insignificant. Since the
statistical analysis showed no significance, the density calculations
may not have captured the potential differences in fibre amounts along
the cylinder height. While steel fibres were only 2% by volume of the
mix, the differences in mass between sections is likely difficult to
quantify.
[FIGURE 4 OMITTED]
2.2. Fracture energy
2.2.1. Disc-Compact Tension Test (DCT)
All specimens from bottom zones and the majority of middle zone
specimens failed to test properly in the DCT configuration by initiating
and propagating cracks beyond the pre-crack. Figure 5 a shows a DCT
specimen from the top cylinder zone that cracked properly with the crack
initiation at the tip of the pre-crack. Alternatively, a bottom specimen
in Figure 5b failed to test properly in the DCT since the crack
initiated beyond the pre-crack. The energies of the failing specimens
were not taken into account for analysis. It was hypothesized at this
point that fibres settled during casting and oriented in a non-random
pattern creating weak paths in the specimens cut from the bottom as well
as some middle zones. The ability to test the same specimen at different
locations is a significant advantage to using the DCT experiment setup
as opposed to traditional tensile testing setups.
[FIGURE 5 OMITTED]
The summary of the DCT results from the top locations is shown in
Figure 6a. It is evident that the heat treatment played a statistically
significant role in fracture energy. The main difference between these
specimens was the post-peak region: non-heat treated specimen exhibited
much slower softening behaviour, which resulted in double the total
fracture energy. The prolonged exposure to severe heat made the UHPCFR
more brittle, including the fibres, which were audibly breaking during
the post-peak period. These results are more detrimental than other
studies conducted previously on the tensile strength of heat-treated
UHPFRC specimens. For example, in the study by Vejmelkova et al (2010)
it was found that high-density glass fibre reinforced cement composite
delivers 33% of its nominal tensile strength after heating specimens to
600[degrees]C [1112[degrees]F]. The results presented herein suggest
that this UHPCFR delivers on average only 36% of its nominal tensile
fracture strength after extreme heating treatment. This difference
suggests that heating the specimens faster and for a longer duration can
significantly weaken the material. It also suggests that the DCT test
may be more sensitive in analysing concrete for tensile strength as
compared to the direct tensile test.
[FIGURE 6 OMITTED]
2.2.2. Semi-Circular Bending Test (SCB)
Similar to DCT results, all specimens tested from zones at the
bottom of cylinders failed to test properly in the SCB configuration.
Most of these failures occurred by crack initiation above one of the
roller supports rather than at the tip of the pre-crack. Figure 7a shows
a properly tested SCB specimen with a crack propagating up from the
pre-crack, whereas Figure 7b shows an improper test where an abrupt
fracture occurred at the left roller. It should also be noted that the
visual observation of all samples similar to the one shown in Figure 7b
resulted in discovery that most of steel fibres were aligned with the
fracture path, which lead to the crack initiation in the weak spot that
was lacking the bridging reinforcement of steel fibres.
[FIGURE 7 OMITTED]
The results from the SCB testing are presented in Figure 8a. As
compared to the DCT results, more SCB specimens from middle locations
tested correctly and were included in the analysis. Similar to the DCT
results though, the post-peak energy as well as peak load for the
heat-treated specimens was significantly lower than for the non-heat
treated specimens as seen in Figures 8b and 7c. An analysis of variance
indicated that heat-treatment was a significant factor in fracture
energy and that nonheated specimens cut from zones in the middle of
cylinders demonstrated higher fracture energy than the specimens from
the top zone.
2.3. X-ray CT
Results from density and fracture energy investigations indicated
that there is a variation between the three zones of the UHPFRC
cylinders. As the final attempt to verify fibre orientation within
different cylinder zones, X-ray CT scans were performed using an Xradia
[C] MicroXCT 400 device. Spatial quantitative analysis was done on the
orientation of fibres with respect to the horizontal and vertical
planes. In order to conduct X-ray scans, 25-mm diameter cylinders were
drilled from the same location of both a top- and bottom-zone DCT
specimen and run through the X-ray device. Multiple 2D images were
stacked together and processed to create 3D representations of internal
fibre structures shown in Figure 9. The qualitative evaluation of fibre
structures in Figure 9 shows that fibres in the bottom-zone specimen (on
the right) are favouring one direction and exhibit less random
orientation than the top-zone sample (on the left). However, such an
evaluation is fairly subjective so more robust procedure was employed in
the next analysis step.
In order to quantify the fibre orientation in space, 3D X-ray
images were analysed using digital image processing software. Raw stacks
of horizontal 2D slices from the X-ray device were manipulated using
combination of thresholding and morphological operations. In the
results, all identified fibres were saved as separate objects and
different geometrical properties were calculated for each object. In
particular, the following two parameters are used in this study: Phi
angle and Theta angle. The directions of both angles are defined in Fig.
10. The Phi angle provides information on the vertical orientation of an
object and can vary between -90[degrees] and +90[degrees] with respect
to the horizontal plane (xy-plane). The Theta angle, on the other hand,
measures the orientation of an object in the horizontal plane and ranges
from 0[degrees] to 360[degrees] starting from the x-axis.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
Figure 11 shows the distribution of fibre angles for both specimens
(top-zone and bottom-zone). Several observations can be made based on
Figure 11:
--The Phi angle distribution for top-zone specimen is more spread,
which indicates a more random fibre orientation in the vertical
direction between -75[degrees] and 75[degrees]. There are also very few
fibres oriented in the horizontal direction (0[degrees]).
--The Phi angle distribution for the bottom-zone specimen shows a
distinct peak close to -15[degrees], which confirms the visual
observation drawn from Figure 9 and suggests that fibres were
consolidated flat and aligned along the horizontal plane.
--The Theta angle distribution for the top-zone specimen shows a
wide spread of values, which again indicates a more random fibre
orientation in this location.
--Finally, the Theta angle distribution for bottom-zone specimen
shows one distinct direction, which agrees with Figure 9 and the Phi
angle observations from this location.
In the summary, quantitative analysis of the 3D Xray scans
confirmed that fibre orientations in the top-zone specimen are more
random and angle ranges in both directions are wider spread. On the
other hand, the bottom-zone specimen comprises unfavourable orientations
in both directions, which is a potential cause for the specimen failures
to test properly in the DCT and SCB fracture experiments.
Conclusions
This study presents the effects of a standard laboratory
preparation on the fibre dispersion and orientation in the Ultra-High
Performance Fibre Reinforced Concrete (UHPFRC). Specimens were prepared
using a special premix combined with steel fibres. As an additional
factor, half of the specimens were heat-treated for 6 hours at
600[degrees]C [1112[degrees]F]. SCB and DCT tests were then run on the
specimens to determine UHPFRC fracture energies in the compressive and
tensile modes. Lastly X-ray CT was used to qualitatively capture the
fibre orientation within different locations of the cast cylinder. The
following conclusions can be made based on observations of this study:
1. Density calculations showed no statistical difference of fibre
dispersion within different locations of the specimens.
2. Specimens located at the bottom of the cylinders failed to test
adequately in the SCB, while both bottom and most middle specimens
tested incorrectly in the DCT. These failures trigger the hypothesis
that fiber orientation is changing along the depth of the cast
cylinders.
3. Heat treated specimens in both compression and tension
conditions had significantly smaller fracture energy than non-heated
specimens and the difference was more distinct in tension mode. The
average percent difference between untreated and heated specimens was
found to be on average 88% in tension and 83% in compression.
4. X-ray tomography analysis confirmed the hypothesis that fiber
orientation in top- and bottom-zone specimens differ significantly.
Distributions of Phi and Theta angles showed clearly that top specimens
have more random fiber orientations whereas bottom samples are oriented
flat and favor one specific direction in the horizontal plane. Such an
orientation is a potential cause for the failure to test properly in the
DCT and SCB fracture experiments.
It is concluded that a special attention should be made while
preparing the UHPFRC samples in the laboratory conditions in order to
ensure homogeneity of the material and to determine representative
properties of the actual material used in a structure.
doi: 10.3846/13923730.2014.913680
Acknowledgments
The authors would like to thank the Department of Homeland
Security, specifically the Transportation Security Center of Excellence
(NTSCOE) at the University of Connecticut for partially funding of this
study. The Authors are grateful to Brian Burke, Tyler Swanson, and
Russell Dutta for their help with specimen preparation. The authors
would also like to thank the University of Connecticut Fuel Cell Center
for help with X-ray tomography. Authors are also thankful to Lafarge for
materials provided for this study. The results and opinions presented in
this paper are those of the authors and do not necessarily reflect those
of involved agencies.
References
ASTM C1437. Standard test method for flow of hydraulic cement
mortar. 2013. http://dx.doi.org/10.1520/C1437
Barnett, S.; Lataste, J.; Parry, T.; Millard, S.; Soutsos, M. 2010.
Assessment of fibre orientation in Ultra High Performance Fibre
Reinforced Concrete and its effect on flexural strength, Materials and
Structures 43(7): 1009-1023. http://dx.doi.org/10.1617/s11527-009-9562-3
Barris, C.; Torres, L.; Mias, C.; Vilanova, I. 2012. Design of FRP
reinforced concrete beams for serviceability requirements, Journal of
Civil Engineering and Management 18(6): 843-857.
http://dx.doi.org/10.3846/13923730.2012.720934
Chanvillard, G.; Rigaud, S. 2003. Complete characterization of
tensile properties of DUCTAL UHP-FRC according to the French
recommendations, Fourth International Workshop on High Performance Fiber
Reinforced Cement Composites (HPFRCC4), 16-18 June, 2003, Ann Arbor,
Michigan, USA, 21-34.
Corinaldesi, V.; Moriconi, G. 2012. Mechanical and thermal
evaluation of ultra-high performance fiber reinforced concretes for
engineering applications, Construction and Building Materials 26(1):
289-294. http://dx.doi.org/10.1016/j.conbuildmat.2011.06.023
Dai, Q.; Ng, K.; Zhou, J.; Kreiger, E.; Ahlborn, T. 2012. Damage
investigation of single-edge notched beam tests with normal strength
concrete and ultra-high performance concrete specimens using acoustic
emission techniques, Construction and Building Materials 31: 231-242.
http://dx.doi.org/10.1016/j.conbuildmat.2011.12.080
Ferrara, L.; Ozyurt, N.; Prisco, M. 2011. High mechanical
performance of fibre reinforced cementitious composites: the role of
"casting-flow induced" fibre orientation, Materials and
Structures 44(1): 109-128. http://dx.doi.org/10.1617/s11527-010-9613-9
Graybeal, B. A.; Hartmann, J. L. 2003. Strength and durability of
Ultra High Performance Concrete, in The Concrete Bridge Conference,
2003, Portland Cement Association, Orlando. 20 p.
Graybeal, B. A. 2007. Comparative behavior of ultra-high
performance fiber-reinforced concrete, ACI Material Journal 104(2):
146-15.
Graybeal, B. A. 2006. Material property characterization of
ultra-high performance concrete, Report No. FHWA-HRT-06-103, Federal
Highway Administration, Washington, D. C. 186 p.
Gribniak, V.; Kaklauskas, G.; Bacinskas, D.; Sung, W.-P.; Sokolov,
A.; Ulbinas, D. 2011. Investigation of shrinkage of concrete mixtures
used for bridge construction in Lithuania, The Baltic Journal of Road
and Bridge Engineering 6(2): 77-83.
http://dx.doi.org/10.3846/bjrbe.2011.10
Grinys, A.; Sivilevicius, H.; Pupeikis, D.; Ivanauskas, E. 2013.
Fracture of concrete containing crumb rubber, Journal of Civil
Engineering and Management 19(3): 447-455.
http://dx.doi.org/10.3846/13923730.2013.782335
Hassan, A.; Jones, S.; Mahmud, G. 2012. Experimental test methods
to determine the uniaxial tensile and compressive behaviour of Ultra
High Performance Fibre Reinforced Concrete (UHPFRC), Construction and
Building Materials 37: 874-882.
http://dx.doi.org/10.1016/j.conbuildmat.2012.04.030
Islam, N.; Zain, M. F. M.; Jamil, M. 2012. Prediction of strength
and slump of rice husk ash incorporated high-performance concrete,
Journal of Civil Engineering and Management 18(3): 310-317.
http://dx.doi.org/10.3846/13923730.2012.698890
Kang, S.; Lee, B.; Kim, J.; Kim, Y. 2011. The effect of fibre
distribution characteristics on the flexural strength of steel
fibre-reinforced ultra-high strength concrete, Construction and Building
Materials 25(5): 2450-2457.
http://dx.doi.org/10.1016/j.conbuildmat.2010.11.057
Kim, D.; Park, S.; Ryu, G.; Koh, K. 2011. Comparative flexural
behaviour of hybrid ultra-high performance fiber reinforced concrete
with different macro fibers, Construction and Building Materials 25(11):
4144-4155. http://dx.doi.org/10.1016/j.conbuildmat.2011.04.051
Lee, C.; Kim, H. 2010. Orientation factor and number of fibers at
failure plane in ring-type steel fiber reinforced concrete, Cement and
Concrete Research 40(5): 810-819.
http://dx.doi.org/10.1016/j.cemconres.2009.11.009
Pajak, M.; Ponikiewski, T.; 2013. Flexural behavior of
self-compacting concrete reinforced with different types of steel
fibers, Construction and Building Materials 47: 397-408.
http://dx.doi.org/10.1016/j.conbuildmat.2013.05.072
Vejmelkova, E.; Konvalinka, P.; Padevet, P.; Cerny, R. 2010.
Thermophysical and mechanical properties of fiber reinforced composite
material subjected to high temperatures, Journal of Civil Engineering
and Management 16(3): 395-400. http://dx.doi.org/10.3846/jcem.2010.45
Wille, K.; Kim, D.; Naaman, A. E. 2010. Strain-hardening UHP-FRC
with low fiber contents, RILEM-Materials and Structures.
http://dx.doi.org/10.1614/S11527-101-9650-4
Yao, W; Li, J.; Wu, K. 2003. Mechanical properties of hybrid
fiber-reinforced concrete at low fiber volume fraction, Cement and
Concrete Research 33(1): 27-30.
http://dx.doi.org/10.1016/S0008-8846(02)00913-4
Zollo, F. 1997. Fiber-reinforced concrete: an overview after 30
years of development, Cement Concrete Composites 19(2): 107-122.
http://dx.doi.org/10.1016/S0958-9465(96)00046-7
Adam ZOFKA (a), Migle PALIUKAITE (b), Audrius VAITKUS (b), Dominika
MALISZEWSKA (a), Ramandeep JOSEN (c), Alexander BERNIER (d)
(a) Road and Bridge Research Institute (IBDiM), ul. Instytutowa 1,
03-302 Warsaw, Poland
(b) Road Research Institute, Vilnius Gediminas Technical
University, Linkmen? g. 28, 08217 Vilnius, Lithuania
(c) Fay, Spofford & Thorndike, Inc., 5 Burlington Woods Dr,
Burlington, MA 01803, USA
(d) Stantec Consulting, 261 Fifth Ave. 23rd Floor, New York, 10016,
USA
Received 07 Nov 2013; accepted 02 Apr 2014
Corresponding author: Adam Zofka E-mail: azofka@ibdim.edu.pl
Adam ZOFKA. PhD, Eng, Professor at the Road and Bridge Research
Institute (IBDiM) in Poland. He received his PhD degree at the
University of Minnesota in 2007 and then worked as an Assistant
Professor at the University of Connecticut (2007-2012). He authored and
co-authored close to 100 scientific publications. His research interest
includes pavement technology, pavement maintenance and pavement
evaluation.
Migle PALIUKAITE. PhD student, Junior Scientist at Road Research
Institute in Vilnius, Lithuania. She received her Master's degree
at Road Department at Vilnius Gediminas Technical University in 2010.
Research interest: bituminous binders, physical-chemical analysis of
bitumen, rheological characterisation of bitumen, durability assessment
of bituminous binders.
Audrius VAITKUS. PhD, Director of Road Research Institute in
Vilnius, Lithuania. He received his PhD degree at Vilnius Gediminas
Technical University in 2007. Research interest: pavement structure
design, performance-based characteristics of materials, pavement
structure performance, pavement surface characteristics.
Dominika MALISZEWSKA. MSc, Eng, Pavement Technology Division at the
Road and Bridge Research Institute (IBDiM) in Poland. She received her
Master's degree at Civil Engineering Department at Warsaw
University of Technology. Research interest: bituminous mixtures,
pavement design, research and technology.
Ramandeep JOSEN. MSc, Eng, Pavement Management Engineer at Fay,
Spofford & Thorndike, Inc. in Massachusetts, USA. He received his
Master's degree at the Civil and Environmental Engineering
Department at the University of Connecticut in 2012. Research interest:
pavement management and evaluation, artificial intelligence, mechanical
characterization of cementitious materials.
Alexander BERNIER. MSc, Eng, Civil Designer at Stantec Consulting
in New York, USA. He received his Master's degree at the Civil and
Environmental Engineering Department at the University of Connecticut in
2012. Research interest: airfield pavement design and evaluation,
computer-aided design, mechanical characterization of cementitious
materials.
Table 1. Composition of UHPFRC mix
Material Amount kg/[m.sup.3] % by weight
(lbs/yd3)
Premix 2198 (3705) 86.5
Super-plasticizer (HRWA) 30.0 (50.6) 1.18
Accelerator 30.0 (50.6) 1.18
Steel fibres 151 (253) 5.91
Polypropylene fibres 5.02 (8.46) 0.20
Water 129 (217) 5.03
Table 2.Test matrix for cylinders: 1 = one non-heat
treated sample, 1 * = one heat treated sample
Cylinder Location 25 mm 37.5 mm
[.984 in.] [1.48 in.]
slices slices
1 Top 1, 1 *
Middle 1
Bottom 1 *
2 Top 1, 1 *
Middle 1 *
Bottom 1
3 Top 1
Middle 1 *
Bottom 1, 1 *
4 Top 1 *
Middle 1
Bottom 1, 1 *
5 Top 1 *
Middle 1, 1 *
Bottom 1
