ACI MATERIALS JOURNAL TECHNICAL PAPER

Title No. 114-M69

Complete Stress-Strain Behavior of Ecological Ultra-High-

Performance Cementitious Composite under Uniaxial
Compression
by Zhiyong Liu, Weiwei Chen, Wenhua Zhang, Yunsheng Zhang, and Henglin Lv
An ecological ultra-high-performance cementitious composite increase in autogenous shrinkage and rapid surface drying
(ECO-UHPCC) was investigated in this paper. The ECO-UHPCC that form premature shrinkage cracks of the UHPCC,12-15
has three characteristics: low cement content (400 to 520 kg/m3 which may reduce its early-age and hardened performance.
[24.97 to 32.46 lb/ft3]), contains nature river sand and high- Therefore, reducing production cost and energy consump-
strength coarse aggregate, and is cured in standard condition. The
tion, simplifying the manufacturing process, and eliminating
complete stress-strain behavior of ECO-UHPCC under uniaxial
the negative effectives caused by the high amount of cemen-
compression was systematically investigated. First, a series of
ECO-UHPCC specimens containing different coarse aggregates titious dosage are the key challenges for the application of
and steel fibers were fabricated. Second, the uniaxial compres- UHPCC in practical engineering.
sive tests were conducted by a high-stiffness, closed-loop, servo- In this paper, an ecological ultra-high-performance cemen-
controlled, material testing machine, and the complete stress- titious composite (ECO-UHPCC) was fabricated by the
strain curves of ECO-UHPCC were obtained. Then, the results, following methods. First, a large amount of portland cement
including complete stress-strain behavior, compressive strength, (≥50%) is replaced by industrial waste, including fly ash,
elastic modulus, Poisson’s ratio, toughness, and fracture pattern of slag, and silica fume. The cement content of ECO-UHPCC
ECO-UHPCC were systematically analyzed. The test results show ranges from 400 to 520 kg/m3 (24.97 to 32.46 lb/ft3). Second,
that ECO-UHPCC exhibits better compressive strength and stiff- natural river sand with a maximum size of 3 mm (0.1182 in.)
ness than high-performance concrete. The compressive strength
substitutes for the costly ultra-fine quartz sand. Third, the
and elastic modulus of ECO-UHPCC with basalt coarse aggregate
standard curing (20°C [68°F] and 100% RH) takes the place
and 2% steel fiber were up to 128.4 MPa (18.62 ksi) and 46.2 GPa
(6700.51 ksi), respectively. Finally, a new model was developed for of rigorous curing regimes. Finally, high-strength coarse
predicting the complete stress-strain behavior of ECO-UHPCC aggregates are used to reduce the cementitious material
under uniaxial compression. This model shows a good correlation dosage. The ECO-UHPCC has an equivalent mechanical
with the experimental results. performance to UHPCC.
Understanding the response of ECO-UHPCC to compres-
Keywords: ecological ultra-high-performance cementitious composite sive loading is important for successful analysis and design of
(ECO-UHPCC); model; stress-strain behavior; uniaxial compression.
structures. Many studies had been conducted on the mechan-
ical properties of UHPCC.2-5,16-21 Accordingly, Graybeal21
INTRODUCTION conducted compressive tests to determine the overall stress-
Ultra-high-performance cementitious composite (UHPCC) strain behavior of ultra-high-performance fiber-reinforced
is an advanced cement based material1,2 that has excep- concrete (UHPFRC). However, because of the limitation of
tional compressive and flexural strength, superior ductility, the test machine, the ascending part of the stress-strain curve
and excellent durability.3-5 It has a great potential for use in was obtained without the descending part. Additionally,
high-rise buildings, large-span space structures, bridge engi- most of the other studies mainly concentrated on the cube or
neering, and military engineering.6,7 cylinder compressive strength of UHPCC.2-5,16-20 Therefore,
To obtain superior mechanical properties, UHPCC is research devoted to investigating the complete stress-strain
composed of a high dosage of cement (generally 900 to behavior of UHPCC in uniaxial compression is very limited.
1000 kg/m3 [56.18 to 62.42 lb/ft3]), very fine powder (such The uniaxial compressive behaviors of the ECO-UHPCC
as crushed quartzite and silica fume), and steel fiber.1 Obvi- were systematically investigated. First, the ECO-UHPCC
ously, these expensive raw materials are responsible for specimens with different coarse aggregates and steel fibers
high energy consumption, which results in significant emis- were fabricated. Second, the uniaxial compressive tests
sions of greenhouse gases.8 In addition, the rigorous curing on ECO-UHPCC were conducted by a special closed-
regimes of 200°C (392°F) autoclave curing or 90°C (194°F) loop, servo-controlled, material testing machine. Third,
heating curing are usually employed in the fabrication of the complete stress-strain behaviors of ECO-UHPCC,
UHPCC,1,9-11 which results in very high energy consumption including compressive strength, elastic modulus, Poisson’s
and restricts this advanced material being used for cast-on-
site engineering. Moreover, for the enhancement homoge- ACI Materials Journal, V. 114, No. 5, September-October 2017.
MS No. M-2016-350.R2, doi: 10.14359/51689899, received November 4, 2016, and
neity, the coarse aggregates are eliminated in UHPCC.1 The reviewed under Institute publication policies. Copyright © 2017, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
elimination of the coarse aggregate means the much higher obtained from the copyright proprietors. Pertinent discussion including author’s
amount of cementitious material dosage. This leads to a sharp closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.

ACI Materials Journal/September-October 2017 783

Table 1—Mixture proportions, kg/m3
Mixture Silica Coarse Steel Maximum size of
No. Water Cement fume Fly ash Slag Sand aggregate fiber HRWRA aggregate, mm
Control mixture CM 164 512 102.4 204.8 204.8 1024 0 0 25.6 —
AS-10 128 400 80 160 160 800 800 0 20 10
Aggregate size (AS) AS-15 128 400 80 160 160 800 800 0 20 15
AS-20 128 400 80 160 160 800 800 0 20 20
AT-G 128 400 80 160 160 800 728 0 20 10
Aggregate type (AT) AT-B 128 400 80 160 160 800 800 0 20 10
AT-I 128 400 80 160 160 800 1335 0 20 10
FC-1 164 512 102.4 204.8 204.8 1024 0 78 25.6 —
FC-2 164 512 102.4 204.8 204.8 1024 0 156 25.6 —
Steel fiber content (FC) FC-3 164 512 102.4 204.8 204.8 1024 0 234 25.6 —
FC-4 164 512 102.4 204.8 204.8 1024 0 312 25.6 —
FC-5 164 512 102.4 204.8 204.8 1024 0 390 25.6 —
FT-S 164 512 102.4 204.8 204.8 1024 0 156 25.6 —
Steel fiber type (FT) FT-D 164 512 102.4 204.8 204.8 1024 0 156 25.6 —
FT-H 164 512 102.4 204.8 204.8 1024 0 156 25.6 —
Optimized mixture OM 128 400 80 160 160 800 800 156 20 10

Notes: 1 kg/m3 = 0.0624 lb/ft3; 1 mm = 0.0394 in.

ratio, toughness, and fracture pattern, were systematically Density, porosity, and water absorption ratio index of the
discussed. Finally, a new model was developed for predicting coarse aggregate were tested according to ASTM C127.24
the complete stress-strain behavior of ECO-UHPCC. The compressive strength of the coarse aggregate was tested
according to GB/T14685-20025 of China. The test samples
RESEARCH SIGNIFICANCE were cylinders with a size of ϕ50 x 50 mm (1.97 x 1.97 in.).
The complete stress-strain relationships of ECO-UHPCC Three different steel fibers, including straight smooth
under uniaxial compression were systematically investi- steel fiber, hooked-end steel fiber, and dumbbell-shaped
gated. The results, including complete stress-strain behavior, steel fiber, were used in this study. These steel fibers have
compressive strength, elastic modulus, Poisson’s ratio, different diameters and lengths, but the same aspect ratio
toughness, and fracture pattern of ECO-UHPCC, were (l/d) of 70.
analyzed. An analytical model was developed for predicting
the complete stress-strain behavior of ECO-UHPCC under Mixture design
uniaxial compression. First, an ECO-UHPCC control mixture (CM) was
designed for comparing with the ECO-UHPCC containing
MATERIALS AND METHODS different coarse aggregates and steel fibers. The water-
Raw materials binder ratio and sand-binder ratio of the mixture were 0.16
Four types of cementitious materials were used in this and 1.0, respectively. For cementitious binder, 50% of port-
study: portland cement (PC) with a 28-day compressive land cement was replaced by ternary mineral admixtures
strength of 68.9 MPa (9.99 ksi), silica fume (SF), fly ash composited of 10% silica fume, 20% fly ash, and 20% slag.
(FA), and slag (SL). Natural river sand with the maximum A polycarboxylic type high-range water-reducing admixture
size of 3 mm (0.118 in.) was used to replace the ultra-fine (HRWRA) with water-reducing ratio of 40% was adopted
quartz sand, which is a necessary component of UHPCC to control workability. The dosage of HRWRA was kept at
reported by published literature.6,22 Three different types 2.5% weight of total binder. The mixture proportions of the
of coarse aggregate (CA) were used: crushed basalt, CM are shown in Table 1.
crushed granitic, and crushed iron ore. These aggregates Secondly, the AS-series and AT-series were designed based
were selected because they present significant differences on CM. The AS-series were used to study the effect of aggre-
in strength, surface texture, porosity and absorption, and gate size on the influence of mechanical properties. AS-10,
bond strength. The crushed basalt has an irregular shape AS-15, and AS-20 present the mixture containing coarse
and rough texture. The crushed granitic presents irregular aggregate with the maximum size of 10, 15, and 20 mm
shape, rough texture, and low absorption. The crushed iron (0.394, 0.591, and 0.787 in.), respectively. The AT-series was
ore has a rough surface texture and the highest density. The designed to investigate the aggregate type. AT-G, AT-B, and
same particle size distribution was adopted with a maximum AT-I present that the ECO-UHPCC contains granitic, basalt,
aggregate size of 10 mm (0.394 in.). The grading of coarse and iron ore coarse aggregate, respectively. The volume
aggregates satisfied the requirements of ASTM C33/C33M.23 fraction of coarse aggregate must remain a constant value

784 ACI Materials Journal/September-October 2017

of 27.5%. Because the steel fiber has a significant enhance- of two rigid circular rings and two linear variable differen-
ment on compressive strength of UHPCC,3,5 the steel fiber tial transformers (LVDTs) connected to a digital transducer.
was not used in these six mixtures to isolate the effect of However, this method is only able to capture the stress-strain
coarse aggregate. values up to the first crack strength of the specimens. It is
Third, the FC-series and FT-series were also designed unable to capture the specimen strain behavior of post-
based on CM. FC-1 to FC-5 were designed to investigate the cracking because the contact points of the LVDTs were
influence of steel fiber content (FC-series) on the compres- disturbed and resulted in erratic readings. In this paper, a test
sion behavior of ECO-UHPCC. For these five mixtures, the method proposed by Mansur et al.28 was used to overcome
straight smooth steel fibers with volume fractions of 1%, the shortage of the conventional method. In this method, four
2%, 3%, 4%, and 5% were used. The FT-S, FT-H, and FT-D strain gauges were fitted over the specimens. Two were used
were designed to study the effect of steel fiber type (FT) on to measure axial strain and the other two were used to measure
compressive properties, which present the ECO-UHPCC transverse strain in the elastic stage of the test. In addition,
containing straight smooth steel fiber, hooked-end steel two hardened rigid platens were specially fabricated, and the
fiber, and dumbbell-shaped steel fiber. The volume fraction two LVDTs were secured on it. The LVDTs were used to
of those three mixtures is 2%. capture the post-cracking behavior of specimen. During the
Finally, on the basis of investigating influence of coarse test, the ECO-UHPCC specimens were sandwiched between
aggregate and steel fiber, an optimized mixture (OM) of the two hardened rigid platens. The cylinder compression
ECO-UHPCC containing 2% steel fiber as well as the basalt tests were conducted according to ASTM C39/C39M.30 The
coarse aggregate with maximum size of 10 mm (0.394 in.) was loading rate was 0.001 mm/s (3.94 × 10–5 in./s). The tests
proposed. All the mixtures proportions are shown in Table 1. results were continuously recorded together with the corre-
sponding loads; the data acquisition system took four read-
Specimen preparation ings per second. Three specimens of each batch were tested.
The ECO-UHPCC was mixed by a 56 L (1.98 ft3) labo- To obtain the complete stress-strain curve of the
ratory pan mixer. First, the cementitious materials (portland ECO-UHPCC, a correction was made according to the
cement, silica fume, fly ash, and slag) and river sand were method introduced by Mansur et al.28 In this correction
put into the mixer and dry-mixed for 1 minute. Second, method, ∆tp is defined as the deformation measured by trans-
the mixture of water and HRWRA was added and mixed ducers placed between the top and bottom platens, ∆tp is the
for 3 minutes. Third, the coarse aggregates were poured deformation measured by strain gauge over a gauge length
into cementitious mixture and mixed for another minute. Lg, ∆m is the deformation due to flexibility of the machine, ∆e
Finally, the steel fiber was sprinkled slowly into mixture is the deformation due to end-zone effect, and ∆c is the actual
and mixed for another 3 minutes to make steel fiber homog- axial deformation of the specimen. It follows that
enously distributed through the fresh mixture. Then, the
fresh ECO-UHPCC mixture was cast into steel molds and ∆tp = ∆c + ∆a (1)
compacted by a vibrating table. The specimens were made in
a cylindrical shape with a diameter of 75 mm (2.953 in.) and where ∆a = ∆m + ∆e.
height of 150 mm (5.906 in.), which satisfies the require- Rearranging Eq. (1), and assuming that ∆a is a function of
ments of ASTM C192.26 The specimens were demolded the applied load gives
after 24 hours and then cured in the standard condition of
20°C (68°F) and 100% relative humidity (RH) until 56 days. ∆ co  σ σ
To obtain accurate test results, specimen ends must be ∆ a = ∆ tp − L= − L (2)
Lg  Etp Ec 
kept smooth and parallel to each other. Therefore, all the
cylinder specimens were grinded by a special grinding
machine before the compressive tests. During the grinding, where L is the distance between the hardened rigid platens;
ECO-UHPCC specimen was sandwiched between parallel σ is the applied stress; and Etp and Ec are the initial tangent
grinding heads first, then the engine was started. The spec- moduli of the concrete based on the stress-strain curves
imen would be automatically grinded smooth, and the two derived from the transducer and strain gauge readings,
ends were kept parallel. respectively.
Equations (1) and (2) give
Testing methods
A high-stiffness, closed-loop, servo-controlled, material  σ σ
testing machine was used to conduct the compressive test. ∆ c = ∆ tp −  − L (3)
 Etp Ec 
It is well known that the longitudinal stiffness of the testing
machine is an important factor in compressive strength
testing, which influences the post-peak response.27,28 The Dividing Eq. (3) by L, the actual concrete strain εc at any
rigid frame of this machine can avoid explosive failures stress level σ can be obtained as follows
of specimens and obtain the full stress-strain curves of
ECO-UHPCC.  1 1
ε c = ε tp −  − σ (4)
According to ASTM C469,29 the strain of the compres-  Etp Ec 
sive test specimen was obtained by the device that consists

ACI Materials Journal/September-October 2017 785

Fig. 1—Stress-strain curve of ECO-UHPCC. (Note: 1 MPa = 0.145 ksi.)
where εtp is the corresponding strain measured by trans- and εt2 and εt1 are the transverse strains at midheight of the
ducers placed between the plates. specimen produced by stresses S2 and S1, respectively.
On the basis of the aforementioned correction method,
the real complete stress-strain curve of ECO-UHPCC spec- RESULTS AND DISCUSSION
imens can be obtained. According to ASTM C469,29 the Complete stress-strain curve and failure mode
elastic modulus Ec and Poisson coefficient ν are calculated Uniaxial compressive stress-strain curves of the
using Eq. (5) and (6), respectively. ECO-UHPCC with steel fiber or coarse aggregate are
displayed in Fig. 1. Overall, the stress-strain curves consist
S 2 − S1 of an ascending part and a descending part. However, there
Ec = (5)
ε 2 − 0.000050 is no descending part of those ECO-UHPCC without steel
fiber (CM, AT-series, and AS-series). As mentioned previ-
ously, although the special compressive test machine with
ε t 2 − ε t1 high stiffness was used and the strain control facility was
ν= (6)
ε 2 − 0.000050 adopted in this study, the ECO-UHPCC without steel fiber
showed brittle failure, and the descending part of the stress-
where S2 is the stress corresponding to 40% of the peak stain curve was hard to obtain.
stress; S1 is the stress corresponding to a longitudinal strain During the compressive test, the ECO-UHPCC specimens
of 0.00005; ε2 is the longitudinal strain at the stress level S2; without steel fiber behaved elastically until peak strength

786 ACI Materials Journal/September-October 2017

Fig. 2—Typical failure modes of cylinder specimens under compression.
and then followed with a sudden failure. With the forma- Figure 2 shows the failure modes of the specimens. In
tion of the first crack when lateral deformation exceeded the case of unreinforced ECO-UHPCC (CM, AT-series, and
its tensile capacity, the ECO-UHPCC specimens lost their AS-series), the specimens exhibited brittle failure under
total strength and failed in an abrupt, explosive manner. In uniaxial compression. The specimen showed either a cone
contrast, ECO-UHPCC specimens with steel fiber behave or a cone-split failure mode, just as shown in Fig. 2(a) and
elastically up to approximately 90 to 95% of their compres- Fig. 2(b). By contrast, steel fiber-reinforced ECO-UHPCC
sive strength, followed by strain hardening behavior up (OM, FC-series, and FT-series) showed a ductile failure
to peak strength. Following the peak strength, a progres- mode. Beyond the peak load, as the cracks extended, the
sive strain softening occurs in which the presence of fiber bridging effect of fibers were activated and provided lateral
governs the softening stage. constraint to the specimen. The cracks along the specimen

ACI Materials Journal/September-October 2017 787

were controlled. Instead of vertical splitting, the specimen Figure 1(b) presents the stress-strain curve of
showed a cone-shear or shear failure mode, as shown in ECO-UHPCC with different maximum sizes of coarse
Fig. 2(c), Fig. 2(d), and Fig. 2(e). Steel fibers prevented the aggregate. The compressive strength of CM, AS-10, AS-15,
microcracks from joining, and thus arrested the sudden loss and AS-20 are 111, 123, 115, and 110 MPa (16.10, 17.84,
of strength.31 16.68, and 15.95 ksi), respectively. The compressive strength
of ECO-UHPCC with basalt coarse aggregate (AS-10 and
Compressive strength AS-15) is higher than that of the control mixture (CM). This
The full stress-strain curve of ECO-UHPCC with different phenomenon may differ from the concept of reactive powder
types of coarse aggregate is shown in Fig. 1(a). It clearly can concrete (RPC). It is well known that the coarse aggregate
be seen that coarse aggregate type has a significant influence must be eliminated in fabrication of RPC for enhancement of
on compressive strength of ECO-UHPCC. The compressive homogeneity.6 However, the tests data in this study proved
strength of AT-B, AT-G, AT-I are 123, 118, and 115 MPa that the proper maximum size of coarse aggregate beneficial
(17.835, 17.11, and 16.676 ksi), respectively. Obviously, the for the compressive strength of ECO-UHPCC. The compres-
ECO-UHPCC with basalt coarse aggregate (AT-B) shows sive strength decreased when the maximum size of coarse
the highest compressive strength. It means that the basalt aggregate was 20 mm (0.788 in.). This phenomenon agrees
aggregate has a more noticeable enhancement on mechanical with the results of self-consolidating concrete conducted by
properties than granite and iron ore coarse aggregate. This Khaleel et al.35 This is due to smaller maximum size of coarse
phenomenon is different from the normal-strength concrete. aggregate that has the larger surface area, which results in a
For normal-strength concrete, the effect of coarse aggregate higher bonding strength in the transition zone around aggre-
type on compressive strength is not significant.32 It is well gate particles when the specimen is under loading. Addi-
known that concrete is a composite, and its properties depend tionally, the increase in strength with smaller diameter may
on the properties of the component phases (paste matrix and also be attributed to a change in the stress repartition inside
aggregates) and the interfacial transition zone (ITZ) between the specimen. Each aggregate, which acts as a rather rigid
them. The compressive strength of normal-strength concrete inclusion inside a deformable matrix, will tend to restrain the
paste matrix is low and the ITZ is weak.33 The cracks exist contraction of the matrix under compressive load. Conse-
in ITZ to a considerable extent before the concrete was quently, it will lead to a development of a locally nonhomog-
subjected to any external load. Under loading, these small or enous state of stresses around each inclusion, which depends
microscopic cracks extend and interconnect until, at ultimate on the aggregate diameter.36
load, the entire internal structure is completely disrupted.32 From the previous discussion, it clearly can be seen that
The aggregate had, in comparison with concrete, relatively the proper coarse aggregate is helpful for fabricating of the
high strength and its potential strength was not fully used. ECO-UHPCC. The basalt coarse aggregate that exhibits
Therefore, the type of the coarse aggregate did not have a good bond, and similar strength to the surrounding
an obvious effect on the compressive strength of normal- ECO-UHPCC mortar matrix, and proper maximum size
strength concrete. However, in ECO-UHPCC, the strength of (10 mm [0.394 in.]), shows the best enhancement on
the paste matrix and ITZ are vastly improved because of the ECO-UHPCC compressive strength.
very low water-binder ratio (0.16) and the use of the mineral The compressive stress-strain curve of ECO-UHPCC with
admixture including silica fume, fly ash, and slag. With the different volume fractions of steel fiber is shown in Fig. 1(c).
development of loading, the cracks may extend through the It clearly can be seen in Fig. 1(c) that the higher amount of
aggregate, which make use of the full strength potential of fiber provided an improvement of the compressive strength.
the coarse aggregate particles. The coarse aggregates act as The highest compressive strength was found to be 138.8 MPa
crack arrestors in ECO-UHPCC. The coarse aggregate type (20.13 ksi) for the specimen with Vf = 5% (FC-5), and this
influences the mechanical properties of ECO-UHPCC. The is 15.6%, 9.4%, 7.1%, 6.2%, and 4.1% higher than those of
stronger of coarse aggregate, the higher the ECO-UHPCC the specimens with Vf = 0 (CM), Vf = 1% (FC-1), Vf = 2%
compressive strength. Therefore, the compressive strength (FC-2), Vf = 3% (FC-3), and Vf = 4% (FC-4), respectively.
of ECO-UHPCC with basalt coarse aggregate (AT-B) is This can be attributed to the effect of the fiber bridging mech-
higher than the ECO-UHPCC with granite coarse aggre- anism. Steel fibers, mainly the ones disposed at a perpendic-
gate (AT-G). Aitcin and Mehta34 also pointed out that in a ular direction with respect to the longitudinal cylinder axis,
high-strength concrete, the hardened cement paste and the bridge longitudinal cracks and the eventual major inclined
transition zone are no longer strength limiting. The miner- crack emerging when the peak load is reached. These addi-
alogy and the strength of coarse aggregates may control the tional resisting mechanisms oppose to damage propagation
ultimate strength of concrete. Although the strength of iron and delay the after-peak loss of load-carrying capacity.
coarse aggregate is higher than the basalt, the compressive To combine the effect of both the volume fraction of steel
strength of AT-I is lower than the AT-B. This because the fiber and their aspect ratio, the reinforcing index (RI = VfLf/ϕf)
strength of iron coarse aggregate (203 MPa [29.44 ksi]) is can be used as the fiber-reinforcing parameter for a given
much higher than the ECO-UHPCC mortar matrix (111 MPa type steel fiber. Vf is the fiber content in volume fraction, Lf
[16.10 ksi]). Under load, this may cause stress concentra- is the fiber length (mm), and ϕf is fiber diameter (mm). As
tion and initiate more microcracks, resulting in a decrease in shown in Fig. 3, the relationship between RI and compres-
compressive strength. sive strength is linear. Based on the regression analysis of the
test data, empirical models were developed for predicting the

788 ACI Materials Journal/September-October 2017

Table 2—Mechanical properties of ECO-UHPCC
Mixture No. Compressive strength, MPa Elastic modulus, GPa Poisson’s ratio Peak strain Ultimate strain Toughness ratio Toughness index
CM 117.1 27.4 0.15 0.004079 0.004302 0.14 1.00
AS-10 123.0 46.9 0.23 0.003018 0.003065 0.11 0.87
AS-15 114.6 47.2 0.26 0.002549 0.002555 0.09 0.66
AS-20 110.0 46.9 0.27 0.002603 0.002603 0.09 0.65
AT-B 123.0 46.9 0.23 0.003018 0.003065 0.11 0.87
AT-G 118.0 59.0 0.25 0.002470 0.002509 0.09 0.70
AT-I 115.0 72.8 0.26 0.002612 0.002673 0.11 0.81
FC-1 123.7 31.0 0.17 0.004290 0.015000* 0.25 1.96
FC-2 126.4 31.8 0.15 0.004370 0.015000* 0.40 3.20
FC-3 127.4 31.5 0.15 0.004373 0.015000 *
0.45 3.56
FC-4 130.0 32.7 0.17 0.004442 0.015000 *
0.50 4.06
FC-5 135.4 30.6 0.16 0.004824 0.015000 *
0.54 4.57
FT-S 126.4 31.8 0.15 0.004370 0.015000* 0.40 3.20
FT-H 121.4 31.7 0.16 0.004072 0.015000 *
0.29 2.20
FT-D 117.8 31.2 0.17 0.004147 0.015000 *
0.22 1.59
OM 128.4 46.2 0.26 0.003431 0.015000 *
0.53 4.27
*
Toughness of ECO-UHPCC specimens were computed up to strain of 0.015, although specimens still had significant resistance left.
Notes: 1 MPa = 0.145 ksi; 1 GPa = 145 ksi.

aspect ratio is the same, the diameter of micro-straight-

smooth steel fiber is much smaller than the hooked-end and
dumbbell steel fiber. Therefore, the contact surface area of
micro-straight-smooth steel fiber is larger than that of the
other two types of steel fiber.
The complete stress-strain curve of the OM mixture is
shown in Fig. 1(e). As shown in Table 2, the compressive
strength of OM is 128.6 MPa (18.65 ksi). This is higher than
FC of 126.4 MPa (18.33 ksi). It means that the addition of
coarse aggregate has the positive effect on mechanical prop-
erties of ECO-UHPCC.

Elastic modulus
The elastic modulus of ECO-UHPCC with different
coarse aggregate is summarized in Table 2. As shown in
Table 2, the elastic modulus of CM, OM, AS-10, AS-15, and
Fig. 3—Effect of RI on compressive strength. (Note: 1 MPa = AS-20 are 27.4, 46.2, 46.9, 47.2, and 46.9 GPa (3973.89,
0.145 ksi.) 6700.51, 6802.03, 6845.54, and 6802.03 ksi), respectively.
Figure 1(a) also shows that the slope of the stress-strain curve
strength properties of steel fiber-reinforced ECO-UHPCC. of ECO-UHPCC with basalt coarse aggregate (OM, AS-10,
The general form of the proposed strength prediction model AS-15, AS-20) is much steeper than the ECO-UHPCC
is given by Eq. (7). without coarse aggregate (CM). It means that the addition
of coarse aggregate can significantly improve the elastic
fcf = fc + 3.61RI (7) modulus of ECO-UHPCC. The greater deformability
observed in ECO-UHPCC without coarse aggregate may
The compressive stress-strain curve of ECO-UHPCC reside in the much greater paste content typically present
with different types of steel fiber is shown in Fig. 1(d). The in ECO-UHPCC mixtures than that with coarse aggregate.
highest compressive strength is 126.4 MPa (18.33 ksi) for Because aggregates are much less deformable than hard-
the specimen with micro-straight smooth steel fiber (FT-S), ened paste, in a hardened composition with higher paste
and this is 4.1% and 7.3% higher than those of the specimens contents, the deformability is obviously higher. Therefore,
with hooked-end (FT-H) and dumbbell steel fiber (FT-D), coarse aggregate is more rigid than the cement paste and
respectively. This phenomenon may be attributed to the fact will deform less under the same applied stress. However, the
that the peak stress is associated with adhesive bond between stress-strain curve slope of the AS-10, AS-15, and AS-20 in
steel fiber and the ECO-UHPCC matrix. Although the fiber

ACI Materials Journal/September-October 2017 789

Fig. 1(a) are very similar, which indicates that coarse aggre- of this study. The addition of coarse aggregate results in a
gate size has little influence on the elastic modulus. higher Poisson’s ratio of concrete.
Figure 1(b) shows that the ascending part of the AT-I In fact, the addition of steel fiber did not show obvious
stress-strain curve almost overlaps those of AT-B and AT-G. influence on the Poisson’s ratio of ECO-UHPCC. This
As shown in Table 2, the elastic modulus of ECO-UHPCC phenomenon is in good agreement with the results given
with iron, basalt, and granite are 72.8 46.9, and 59.0 GPa by Thomas et al.42 They reported that the variation in the
(10,558.38, 6802.03, and 8556.93 ksi), respectively. This Poisson’s ratio of the concrete due to the addition of steel
indicates that coarse aggregate type has a significant influ- fibers was marginal. They explained that Poisson’s ratio of
ence on elastic modulus of ECO-UHPCC. This influence concrete is computed based on the observations at initial
was attributed to the highly dense paste structure and paste- stages of loading, where the fibers do not play a significant
aggregate bond, which cause the concrete to behave like a role in the load sustenance.
composite material.37 Therefore, aggregate characteristics
could be important in determining the elastic properties of Toughness
ECO-UHPCC. The more rigid of the coarse aggregate in an The deformation capacity and energy dissipation capacity
ECO-UHPCC mixture, the higher elastic modulus would be. of ECO-UHPCC were also systematically investigated in
The importance of coarse aggregate quality on elastic prop- this study. Toughness ratio (TR) and toughness index (TI),
erties of concrete was also pointed by Baalbaki et al.38 and defined as Eq. (8) and Eq. (9), respectively,39 are used to
Beshr et al.,37 who studied high-performance concrete and quantify the toughness of different ECO-UHPCC.
high-strength concrete, respectively.
The elastic modulus of ECO-UHPCC with different ED
TR = (8)
volume fraction of steel fiber is shown in Table 2. The elastic 0.015 f cf′
modulus of FC-1, FC-2, FC-3, FC-4, and FC-5 are 31.0,
31.8, 31.5, 32.7, and 30.6 GPa (4496.01, 4612.04, 4568.53,
4742.57, and 4438.00 ksi), respectively. This indicates that ED
TI = (9)
the volume fraction of steel fiber has little influence on the EDc
elastic modulus of ECO-UHPCC. This situation agrees with
the study conducted by Yu-Chen et al.39 They reported that where ED is the energy absorption capacity of ECO-UHPCC
little correlation was found between the modulus of elas- or the area under the stress-strain curve; fcf′ is the peak
ticity and the reinforcing index, as expected.39 This situation compressive stress of ECO-UHPCC. EDc is energy absorp-
may be caused by the fact that steel fiber fraction employed tion capacity of the control specimen. Both ED and EDc are
in this investigation is very low (Vf ≤ 5%); the value of the defined up to a strain of 0.015, though the specimens still
modulus of elasticity was hardly affected. Figure 1(d) shows had significant resistance left. The peak strain (the strain
that the different type of steel fiber also has little influence at the peak compressive stress), ultimate strain, toughness
on the elastic modulus of ECO-UHPCC. This because the ratio (TR), and toughness index (TI) of ECO-UHPCC are
steel fiber volume fractions of the three batches are the same. summarized in Table 2.
As shown in Table 2 and Fig. 1, the highest peak strain was
Poisson’s ratio found to be 0.004824 for the specimen with Vf = 5% (FC-5),
The Poisson’s ratios of ECO-UHPCC are shown in and this is 18.3%, 12.4%, 10.3%, 10.3%, and 8.5% higher than
Table 2. As shown in Table 2, the Poisson’s ratio ranges from those of the specimens with Vf = 0 (CM), Vf = 1% (FC-1), Vf =
0.15 to 0.27. These data agree with that reported by Peren- 2% (FC-2), Vf = 3% (FC-3), and Vf = 4% (FC-4), respectively.
chio and Klieger.40 They reported values for Poisson’s ratio However, the peak strain of ECO-UHPCC without steel fiber
of normalweight high-strength concrete with compressive are much smaller than that with steel fiber. The peak strain of
strengths ranging from 55 to 80 MPa (7.98 to 11.60 ksi) to AT-series and AS-series range from 0.002500 to 0.003000.
be between 0.20 and 0.28. However, Mehta41 pointed out As discussed in the “Complete stress-strain curve and failure
that the values of concrete Poisson’s ratio generally vary mode” section, the specimen of ECO-UHPCC without steel
between 0.15 and 0.20. fiber showed brittle failure during the compressive tests and
The Poisson’s ratio of ECO-UHPCC with coarse aggre- there was no descending part of the stress-train curve. The
gate (OM, AT-series, AS-series) ranges from 0.23 to 0.27. addition of steel fiber significantly enhanced the deforma-
This is much higher than those ECO-UHPCC without coarse tion capacity of the ECO-UHPCC.
aggregate (CM, FC-series, FT-series) whose Poisson’s ratio The toughness index of AT-series and AS-series range
ranges from 0.15 to 0.17. Metha42 pointed out that the Pois- from 0.65 to 0.87, which is less than 1.0 of the control
son’s ratio does not appear to be in a consistent relationship mixture. This means that the energy dissipation capacity of
with concrete characteristics such as water-cement ratio the ECO-UHPCC with coarse aggregate is weaker than that
(w/c), curing age, and aggregate gradation. Poisson’s ratio of the control mixture. Additionally, the maximum size of
is generally lower in high-strength concrete, and higher for coarse aggregate influences the toughness index. The larger
saturated concrete and for dynamically loaded concrete. the coarse aggregate, the smaller the toughness index. This
However, there is a significant relationship between coarse may be attributed to the fact that the addition of coarse aggre-
aggregate and ECO-UHPCC Poisson’s ratio from the data gate constrains the deformation ability of the ECO-UHPCC,
as discussed previously.

790 ACI Materials Journal/September-October 2017

 εE
bf c′
α = ae −a (13)

This model showed good agreement with the experi-

mental results of UHPFRC in Graybeal’s21 research. Zhou
et al.43 also used this model to predict the ascending part
of engineering cementitious composites (ECCs). However,
as the compressive strength of UHPFRC was too high,
the descending part of the stress-strain curve could not be
obtained in that study.
In the present study, the analytical expression of the
stress-strain law suggested by CEB-FIP Model Code44
was introduced as the basic model equation to predict the
ascending part of the curves. The shape of the equation has
good correlation with experimental results in this study. The
Fig. 4—Relationship between TR, TI, and volume fraction expression is
of steel fiber.
2
The toughness index values of OM, FC-series, and FT-  ε  ε
A  −  
series range from 1.59 to 4.57, which is much higher than σ  ε0   ε0 
= (ε ≤ ε 0 ) (14)
that of the control mixture. ECO-UHPCC with steel fiber f c′  ε
shows excellent energy dissipation capacity. ECO-UHPCC 1 + ( A − 2)  
 ε0 
reinforced with higher contents of fibers show a more
extended softening branch. Therefore, the area under stress-
where σ and ε are the actual stress and actual strain value,
strain curve of ECO-UHPCC with more steel fiber were
respectively; fc′ and ε0 are the peak stress (compressive
much larger than lesser one. The TI of OM is 4.27, which is
strength) and peak strain, respectively; and A is a parameter
higher than the FC-4 of 4.06. The addition of coarse aggre-
that controls the ascending branch of the curve.
gate is helpful for the toughness of steel fiber-reinforced
The stress-strain curve represented by Eq. (14) has a
ECO-UHPCC. On the basis of tests data of the FC-series and
defect, which underpredicts the decline of the stress-strain
CM, the relationship between TR, RI, and volume fraction
curve. Therefore, a new model must be proposed to satisfy
of steel fiber regressed as Eq. (10), Eq. (11), and the fitting
descending part of the curves. The following conditions
curve are shown in Fig. 4.
should be considered to represent the complete descending
branch of the stress-strain curve
TR = 0.2754Vf0.43144 (R2 = 0.96) (10)
1. σ(1)/fc′ = 1, the stress equals the peak stress for ε0 =
ε(1) = 1;
TI = 2.12312Vf0.47759 (R2 = 0.97) (11)
d σ (ε )
2. = 0, the descending curve has a zero tangent at
It clearly can be seen from Fig. 4 that the relationship d ε f c′
between TR, TI, and volume fraction of steel fibers are both the peak stress; and
power functions. The fitting curve of toughness ratio and σ (ε )
toughness index are very similar to each other. 3. lim =0
ε→∞ f c′
ANALYTICAL MODELS OF COMPRESSIVE Assimilating all these conditions, the formulation of the
STRESS-STRAIN RELATIONSHIP softening branch of the stress-strain curve was proposed
The compressive stress-strain relationship of ECO-UHPCC
is important in the analysis and design of structural members. σ 1
In the last few decades, few analytical models have been = C
(ε ≥ ε 0 ) (15)
f c′  ε
proposed for predicting the behavior of ECO-UHPCC under B ×   +1− B
uniaxial compression. Accordingly, the model proposed by  ε0 
Graybeal21 for the ultra-high-performance fiber-reinforced where B and C are parameters that control the descending
concrete (UHPFRC) is shown as follows branch of the curve. Equations (14) and (15) can be used
to generate the complete stress-strain curve for different
σ = εE(1 – α) (12) A-, B-, and C-values in normalized manner. Those values
are summarized in Table 3. In this table, the symbol “—”
Equation (12) shows the actual stress σ and strain ε values means that the descending branch is not available for the
are related by the modulus of elasticity E and a reduction ECO-UHPCC without steel fiber. The ascending part of the
factor α. α was defined as the decrease in the actual stress stress-strain curve is effected by the A-value. As the A-value
from the linear elastic stress, which is shown as Eq. (13), increases, the slope of the curve increases, meaning that
where a and b are the fitting parameters. the elastic modulus of the ECO-UHPCC increases. There-

ACI Materials Journal/September-October 2017 791

Table 3—Values of parameters A, B, and C for different ECO-UHPCC
Ascending part Descending part
Mixture No. A B C R2
CM 1.00385 — — 0.99721
FC-1 1.07278 19.16287 1.11616 0.92502
FC-2 1.11079 46001.91592 9.3868 × 10–5 0.73140
Fiber content FC-3 1.07738 9043.41488 2.71588 × 10–5 0.99559
FC-4 1.10733 9292.72487 1.95629 × 10 –5
0.96709
FC-5 1.14915 7552.30288 1.93417 × 10 –4
0.84696
FT-H 1.05611 167630.7218 5.8062 × 10–5 0.81959
Fiber type
FT-D 1.07968 71161.1575 2.62591 × 10–4 0.98097
AT-B 1.12343 — — 0.99946
Aggregate type AT-G 1.22873 — — 0.99405
AT-I 1.71888 — — 0.99690
AS-10 1.14139 — — 0.99966
Aggregate size AS-15 1.09602 — — 0.99857
AS-20 1.12447 — — 0.99777
OM 1.17989 3479.43588 4.2996 × 10 –4
0.96784

fore, the A-value is related to the elastic modulus of the Weiwei Chen is a Graduate Student in the Jiangsu Key Laboratory of Envi-
ronmental Impact and Structural Safety in Engineering, China University
ECO-UHPCC. The descending part of the stress-strain curve of Mining and Technology. His research interests include characterization
is effected by the B- and C-values. and modeling of concrete microstructure.
The comparison of proposed model equations with exper-
Wenhua Zhang is an Associate Professor at Nanjing Forestry University,
imental results is shown in Fig. 5. Obviously, there is a good Nanjing, China. He received his PhD in material science from Southeast
correlation between the experimentally measured stress- University in 2012. His research interests include durability of high-
strain curves with the analytical curves. performance concrete.

Yunsheng Zhang is a Professor in the Jiangsu Key Laboratory for

CONCLUSIONS Construction Materials of Southeast University. He received his PhD in
Based on the results of this study, the following conclu- structure engineering from Southeast University in 2004. His research inter-
ests include geopolymeric cement and its composites, high-performance
sions can be made: concrete made with fly ash or slag, durability, and service life prediction of
1. An ecological ultra-high-performance cementitious high-performance concrete.
composite (ECO-UHPCC) was successfully fabricated. A
Henglin Lv is a Professor at China University of Mining and Technology,
large amount of portland cement (≥50%) was replaced by where he also received his PhD in structure engineering. His research inter-
industrial waste, including fly ash, slag, and silica fume. ests include durability of reinforced concrete.
The cement content of ECO-UHPCC ranged from 400 to
520 kg/m3 (24.97 to 32.46 lb/ft3). Natural river sand and ACKNOWLEDGMENTS
various coarse aggregates were substitutes for the costly The authors gratefully acknowledge the financial support of the National
Natural Science Foundation of China (51778613) and the Fundamental
ultra-fine quartz sand. Research Funds for the Central Universities (2015XKMS011).
2. The ECO-UHPCC has an equivalent mechanical perfor-
mance with UHPCC. The compressive strength and elastic REFERENCES
modulus of ECO-UHPCC containing basalt coarse aggre- 1. Richard, P., and Cheyrezy, M., “Composition of Reactive Powder
gate with a maximum size of 10 mm (0.394 in.) and 2% Concretes,” Cement and Concrete Research, V. 25, No. 7, 1995, pp. 1501-
1511. doi: 10.1016/0008-8846(95)00144-2
steel fiber were up to 128.4 MPa (18.62 ksi) and 46.2 GPa 2. Wille, K.; El-Tawil, S.; and Naaman, A. E., “Properties of Strain
(6700.51 ksi), respectively. Hardening Ultra-High Performance Fiber Reinforced Concrete (UHP-FRC)
3. A new model was developed for predicting the full under Direct Tensile Loading,” Cement and Concrete Composites, V. 48,
2014, pp. 53-66. doi: 10.1016/j.cemconcomp.2013.12.015
stress-strain behavior of ECO-UHPCC under uniaxial 3. Rong, Z., and Sun, W., “Experimental and Numerical Investigation
compression. This model shows a good correlation between on the Dynamic Tensile Behavior of Ultra-High Performance Cement-
the experimental results. Based Composites,” Construction and Building Materials, V. 31, 2012,
pp. 168-173. doi: 10.1016/j.conbuildmat.2011.12.058
4. Zhang, W.; Zhang, Y.; and Zhang, G., “Static, Dynamic Mechanical
AUTHOR BIOS Properties and Microstructure Characteristics of Ultra-High Performance
Zhiyong Liu is an Associate Professor in the State Key Laboratory of Cementitious Composites,” Science and Engineering of Composite Mate-
Geomechanics & Deep Underground Engineering at China University of rials, V. 19, No. 3, 2012, doi: 10.1515/secm-2011-0136
Mining and Technology, Xuzhou, China. He received his PhD in material 5. Yang, S. L.; Millard, S. G.; Soutsos, M. N.; Barnett, S. J.; and Le, T. T.,
Science from Southeast University, Nanjing, China, in 2013. His research “Influence of Aggregate and Curing Regime on the Mechanical Proper-
interests include nondestructive testing, microstructure, and transport ties of Ultra-High Performance Fibre Reinforced Concrete (UHPFRC),”
properties of cement-based materials.

792 ACI Materials Journal/September-October 2017

Fig. 5—Comparison of proposed model equations with experimental results.
Construction and Building Materials, V. 23, No. 6, 2009, pp. 2291-2298. 9. Yazıcı, H.; Yiğiter, H.; Karabulut, A.; and Baradan, B., “Utilization of
doi: 10.1016/j.conbuildmat.2008.11.012 Fly Ash and Ground Granulated Blast Furnace Slag as an Alternative Silica
6. Zhang, W.; Zhang, Y.; and Zhang, G., “Single and Multiple Dynamic Source in Reactive Powder Concrete,” Fuel, V. 87, No. 12, 2008, pp. 2401-
Impacts Behaviour of Ultra-High Performance Cementitious Composite,” 2407. doi: 10.1016/j.fuel.2008.03.005
Journal of Wuhan University of Technology – Materials, V. 26, No. 6, 2011, 10. Cwirzen, A.; Penttala, V.; and Vornanen, C., “Reactive Powder Based
pp. 1227-1234. doi: 10.1007/s11595-011-0395-x Concretes: Mechanical Properties, Durability and Hybrid Use with OPC,”
7. Yunsheng, Z.; Wei, S.; Sifeng, L.; Chujie, J.; and Jianzhong, L., Cement and Concrete Research, V. 38, No. 10, 2008, pp. 1217-1226. doi:
“Preparation of C200 Green Reactive Powder Concrete and its Static- 10.1016/j.cemconres.2008.03.013
Dynamic Behaviors,” Cement and Concrete Composites, V. 30, No. 9, 11. Feylessoufi, A.; Crespin, M.; Dion, P.; Bergaya, F.; Van Damme,
2008, pp. 831-838. doi: 10.1016/j.cemconcomp.2008.06.008 H.; and Richard, P., “Controlled Rate Thermal Treatment of Reactive
8. Lima, C.; Caggiano, A.; Faella, C.; Martinelli, E.; Pepe, M.; and Real- Powder Concretes,” Advanced Cement Based Materials, V. 6, No. 1, 1997,
fonzo, R., “Physical Properties and Mechanical Behaviour of Concrete Made pp. 21-27. doi: 10.1016/S1065-7355(97)90002-X
with Recycled Aggregates and Fly Ash,” Construction and Building Mate- 12. Yoo, D.; Min, K.; Lee, J.; and Yoon, Y., “Shrinkage and Cracking
rials, V. 47, 2013, pp. 547-559. doi: 10.1016/j.conbuildmat.2013.04.051 of Restrained Ultra-High-Performance Fiber-Reinforced Concrete Slabs at

ACI Materials Journal/September-October 2017 793

Early Age,” Construction and Building Materials, V. 73, 2014, pp. 357-365. 28. Mansur, M. A.; Wee, T. H.; and Chin, M. S., “Derivation of the
doi: 10.1016/j.conbuildmat.2014.09.097 Complete Stress-Strain Curves for Concrete in Compression,” Magazine
13. Yoo, D.; Park, J.; Kim, S.; and Yoon, Y., “Early Age Setting, of Concrete Research, V. 47, No. 173, 1995, pp. 285-290. doi: 10.1680/
Shrinkage and Tensile Characteristics of Ultra High Performance Fiber macr.1995.47.173.285
Reinforced Concrete,” Construction and Building Materials, V. 41, 2013, 29. ASTM C469/C469M-10, “Standard Test Method for Static Modulus
pp. 427-438. doi: 10.1016/j.conbuildmat.2012.12.015 of Elasticity and Poisson’s Ratio of Concrete in Compression,” ASTM
14. Soliman, A. M., and Nehdi, M. L., “Effects of Shrinkage Reducing International, West Conshohocken, PA, 2010, 5 pp.
Admixture and Wollastonite Microfiber on Early-Age Behavior of Ultra- 30. ASTM C39/C39M-11, “Standard Test Method for Compressive
High Performance Concrete,” Cement and Concrete Composites, V. 46, Strength of Cylindrical Concrete Specimens,” ASTM International, West
2014, pp. 81-89. doi: 10.1016/j.cemconcomp.2013.11.008 Conshohocken, PA, 2011, 7 pp.
15. Yoo, D.; Park, J.; Kim, S.; and Yoon, Y., “Influence of Reinforcing 31. Kayali, O.; Haque, M. N.; and Zhu, B., “Some Characteristics of
Bar Type on Autogenous Shrinkage Stress and Bond Behavior of Ultra High High Strength Fiber Reinforced Lightweight Aggregate Concrete,” Cement
Performance Fiber Reinforced Concrete,” Cement and Concrete Compos- and Concrete Composites, V. 25, No. 2, 2003, pp. 207-213. doi: 10.1016/
ites, V. 48, 2014, pp. 150-161. doi: 10.1016/j.cemconcomp.2013.11.014 S0958-9465(02)00016-1
16. Wille, K.; Naaman, A. E.; El-Tawil, S.; and Parra-Montesinos, 32. Wu, K.; Chen, B.; Yao, W.; and Zhang, D., “Effect of Coarse Aggre-
G. J., “Ultra-High Performance Concrete and Fiber Reinforced Concrete: gate Type on Mechanical Properties of High-Performance Concrete,”
Achieving Strength and Ductility without Heat Curing,” Materials and Cement and Concrete Research, V. 31, No. 10, 2001, pp. 1421-1425. doi:
Structures, V. 45, No. 3, 2012, pp. 309-324. doi: 10.1617/s11527-011-9767-0 10.1016/S0008-8846(01)00588-9
17. Habel, K.; Viviani, M.; Denarié, E.; and Brühwiler, E., “Develop- 33. Beshr, H.; Almusallam, A. A.; and Maslehuddin, M., “Effect of
ment of the Mechanical Properties of an Ultra-High Performance Fiber Coarse Aggregate Quality on the Mechanical Properties of High Strength
Reinforced Concrete (UHPFRC),” Cement and Concrete Research, V. 36, Concrete,” Construction and Building Materials, V. 17, No. 2, 2003,
No. 7, 2006, pp. 1362-1370. doi: 10.1016/j.cemconres.2006.03.009 pp. 97-103. doi: 10.1016/S0950-0618(02)00097-1
18. Aldahdooh, M. A. A.; Muhamad Bunnori, N.; and Megat Johari, 34. Aitcin, P. C., and Mehta, P. K., “Effect of Coarse Aggregate Charac-
M. A., “Development of Green Ultra-High Performance Fiber Rein- teristics on Mechanical Properties of High-Strength Concrete,” ACI Mate-
forced Concrete Containing Ultrafine Palm Oil Fuel Ash,” Construc- rials Journal, V. 87, No. 2, Mar.-Apr. 1990, pp. 103-107.
tion and Building Materials, V. 48, 2013, pp. 379-389. doi: 10.1016/j. 35. Khaleel, O. R.; Al-Mishhadani, S. A.; and Abdul Razak, H., “The
conbuildmat.2013.07.007 Effect of Coarse Aggregate on Fresh and Hardened Properties of Self-
19. Habel, K., and Gauvreau, P., “Response of Ultra-High Performance Compacting Concrete (SCC),” Procedia Engineering, V. 14, 2011,
Fiber Reinforced Concrete (UHPFRC) to Impact and Static Loading,” pp. 805-813. doi: 10.1016/j.proeng.2011.07.102
Cement and Concrete Composites, V. 30, No. 10, 2008, pp. 938-946. doi: 36. Szczesniak, M.; Rougelot, T.; Burlion, N.; and Shao, J. F., “Compres-
10.1016/j.cemconcomp.2008.09.001 sive Strength of Cement-Based Composites: Roles of Aggregate Diameter
20. Lai, J., and Sun, W., “Dynamic Behaviour and Visco-elastic Damage and Water Saturation Degree,” Cement and Concrete Composites, V. 37,
Model of Ultra-High Performance Cementitious Composite,” Cement and 2013, pp. 249-258. doi: 10.1016/j.cemconcomp.2012.08.001
Concrete Research, V. 39, No. 11, 2009, pp. 1044-1051. doi: 10.1016/j. 37. Beshr, H.; Almusallam, A. A.; and Maslehuddin, M., “Effect of
cemconres.2009.07.012 Coarse Aggregate Quality on the Mechanical Properties of High Strength
21. Graybeal, B. A., “Compressive Behavior of Ultra-High-Perfor- Concrete,” Construction and Building Materials, V. 17, No. 2, 2003,
mance Fiber-Reinforced Concrete,” ACI Materials Journal, V. 104, No. 2, pp. 97-103. doi: 10.1016/S0950-0618(02)00097-1
Mar.-Apr. 2007, pp. 146-152. 38. Baalbaki, W.; Benmokrane, B.; Chaallal, O.; and Aitcin, P., “Influence
22. Zanni, H.; Cheyrezy, M.; Maret, V.; Philippot, S.; and Nieto, P., of Coarse Aggregate on Elastic Properties of High-Performance Concrete,”
“Investigation of Hydration and Pozzolanic Reaction in Reactive Powder ACI Materials Journal, V. 88, No. 5, Sept.-Oct. 1991, pp. 499-503.
Concrete (RPC) Using 29Si NMR,” Cement and Concrete Research, V. 26, 39. Ou, Y.; Tsai, M.; Liu, K.; and Chang, K., “Compressive Behavior of
No. 1, 1996, pp. 93-100. doi: 10.1016/0008-8846(95)00197-2 Steel-Fiber-Reinforced Concrete with a High Reinforcing Index,” Journal
23. ASTM C33-03, “Standard Specification for Concrete Aggregates,” of Materials in Civil Engineering, ASCE, V. 24, No. 2, 2012, pp. 207-215.
ASTM International, West Conshohocken, PA, 2003, 11 pp. doi: 10.1061/(ASCE)MT.1943-5533.0000372
24. ASTM C127-01, “Standard Test Method for Density, Relative 40. Perenchio, W. F., and Klieger, P., Some Physical Properties of High-
Density (Specific Gravity), and Absorption of Coarse Aggregate,” ASTM Strength Concrete, Portland Cement Association, Skokie, IL, 1978, 56 pp.
International, West Conshohocken, PA, 2001, 6 pp. 41. Mehta, P. K., and Monteiro, P. J., Concrete: Microstructure, Proper-
25. GB/T14685-2001, “Pebble and Crushed Stone for Building,” China ties, and Materials, McGraw-Hill, New York, 2006, 93 pp.
Standards, 1992. 42. Thomas, J., and Ramaswamy, A., “Mechanical Properties
26. ASTM C192/C192M-06, “Standard Practice for Making and Curing of Steel Fiber-Reinforced Concrete,” Journal of Materials in Civil
Concrete Test Specimens in the Laboratory,” ASTM International, West Engineering, ASCE, V. 19, No. 5, 2007, pp. 385-392. doi: 10.1061/
Conshohocken, PA, 2006, 8 pp. (ASCE)0899-1561(2007)19:5(385)
27. Hassan, A. M. T.; Jones, S. W.; and Mahmud, G. H., “Experi- 43. Zhou, J.; Pan, J.; and Leung, C. K. Y., “Mechanical Behavior
mental Test Methods to Determine the Uniaxial Tensile and Compres- of Fiber-Reinforced Engineered Cementitious Composites in Uniaxial
sive Behaviour of Ultra High Performance Fibre Reinforced Concrete Compression,” Journal of Materials in Civil Engineering, ASCE, V. 27,
(UHPFRC),” Construction and Building Materials, V. 37, 2012, No. 1, 2015, 04014111, doi: 10.1061/(ASCE)MT.1943-5533.0001034
pp. 874-882. doi: 10.1016/j.conbuildmat.2012.04.030 44. CEB-FIP Model Code, “Design of Concrete Structures,” British
Standards Institution, London, UK, 1993, 2 pp.

794 ACI Materials Journal/September-October 2017