Composites Part B 224 (2021) 109220

Contents lists available at ScienceDirect

Composites Part B
journal homepage: www.elsevier.com/locate/compositesb

New development of ultra-high-performance concrete (UHPC)

Jiang Du a, Weina Meng a, *, Kamal H. Khayat b, Yi Bao a, Pengwei Guo a, Zhenghua Lyu a,
Adi Abu-obeidah c, Hani Nassif c, Hao Wang d
a
Department of Civil, Environmental and Ocean Engineering, Stevens Institute of Technology, Hoboken, NJ, USA
b
Department of Civil, Architectural and Environmental Engineering, Missouri University of Science and Technology, Rolla, MO, USA
c
Department of Civil and Environmental Engineering, Rutgers, The State University of New Jersey, Piscataway, NJ, USA
d
Centre for Future Materials, University of Southern Queensland, Queensland, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Ultra-high-performance concrete (UHPC) is a type of cement-based composite for new construction and/or
Autogenous shrinkage restoration of existing structures to extend service life. UHPC features superior workability, mechanical prop­
Design methodologies erties, and durability compared with conventional concrete. However, some challenges limit the wider appli­
Durability
cation of UHPC, such as low workability for large-volume production, high autogenous shrinkage, insufficient
Fresh properties
Mechanical properties
flexural/tensile properties, and unpredictable durability after concrete cracking. Therefore, this paper reviews
Structural applications the state-of-the-art technologies for developing UHPC mixtures with improved properties. This review covers the
Ultra-high-performance concrete (UHPC) following aspects: (1) the existing design methodologies; (2) the typical ingredients (e.g., binders, aggregates,
chemical admixtures, and fibers) for preparation of UHPC and the underlying working principals; (3) the
technologies for improving and controlling key properties (e.g., workability, autogenous shrinkage, compressive
performance, tensile/flexural properties, and durability); and (4) the representative successful applications. This
review is expected to advance the fundamental knowledge of UHPC and promote further research and appli­
cations of UHPC.

1. Introduction micro-defect free cement (MDF) was invented for production of

ultra-high-strength concrete with compressive strength over 200 MPa.
Ultra-high-performance concrete (UHPC) has been developed for However, the extremely high material cost and complicated production
three decades and is considered one of the most promising construction process restrained its applications [15]. Afterwards, dense silica particle
materials for future sustainable and resilient infrastructure [1–5]. Given (DSP) cement was invented in Denmark [16]. The DSP cement was
its low water-to-binder ratio (w/b, 0.15–0.25), high particle packing designed to improve the particle packing density of concrete and was
density (0.825–0.855), high-volume of steel fibers (≥2%, by volume), easier to produce compared with MDF cement. Combining the silica
and proper addition of chemical admixtures, UHPC exhibits good fume, superplasticizer, and pressure curing conditions, the maximum
flowability (mini-slump flow ≥160 mm) and high mechanical properties compressive strength of DSP incorporated concrete can reach up to 345
(28-day compressive strength ≥120 MPa [6] and tensile strength ≥5 MPa [16]. In the mid-1980s, steel fibers were introduced to overcome
MPa [7], under standard curing; 28-day compressive strength ≥150 the brittleness of this concrete. Finally, slurry infiltrated fiber concrete
MPa, under steam curing [8]). Due to the use of high-volume fibers, (SIFCON) was developed, which incorporated up to 20% steel fibers (by
UHPC exhibits strain-hardening behaviors after cracking [9,10]. Addi­ volume). The 28-day flexural strength of SIFCON ranged from 25 MPa to
tionally, UHPC exhibits superior durability because its matrix is nearly 70 MPa [17–19]. However, the SIFCON was not workable, which hin­
impermeable to carbon dioxide, chloride, sulfate, etc., as shown in Fig. 1 dered its wider application.
[11–13]. Thanks to the advancement of superplasticizers in the 1990s, reac­
In the past 30 years, UHPC has developed through four stages [14]. tive powder concrete (RPC) with superior workability was developed by
Before the 1980s, high-strength concrete could only be produced on Richard and Cheyrezy, which is considered the most major milestone in
lab-scale with vacuum mixing and heat curing [14]. In the early 1980s, the UHPC development [20]. The RPC was classified into two grades,

* Corresponding author.
E-mail address: [email protected] (W. Meng).

https://doi.org/10.1016/j.compositesb.2021.109220
Received 12 November 2020; Received in revised form 2 August 2021; Accepted 9 August 2021
Available online 14 August 2021
1359-8368/© 2021 Elsevier Ltd. All rights reserved.
J. Du et al. Composites Part B 224 (2021) 109220

RPC200 (strength below 200 MPa) and RPC800 (strength from 200 MPa Under restrained conditions, there is a high risk of cracking and/or
to 800 MPa). Since then, the term “ultra-high-performance concrete debonding for UHPC structures [53–56]. In addition, while UHPC with
(UHPC)” was introduced [21]. Later, the first commercial UHPC was high-strength steel fiber shows good ductility and resilience, the
developed based on the recipe of RPC [20], then many other UHPCs high-strength steel fibers account for approximately 35% of the total
were commercialized [22–28]. Some representative commercial UHPC costs of UHPC which resulted in the high initial cost of UHPC [57]. Thus,
formulations are summarized in Table A of the Appendix. The features of it is challenged to reduce the steel fiber content while retaining or
these UHPC mixtures include: (1) extremely low water-to-binder ratio further improving flexural/tensile properties. Further, although the
(0.15–0.24), (2) optimized gradation of solid particles for high particle intact UHPC has excellent durability, cracks always exist in concrete
packing density (0.825–0.855), (3) finely grounded quartz sand (dmax < structures under outer service loads. Attention should also be paid to the
0.6 mm), (4) high cement content (800–1100 kg/m3) and silica fume durability when some special raw materials were incorporated, such as
content (150–300 kg/m3), and (5) high steel fiber content (2–5%) [29, waste glass and porous sand, which may introduce expansive gel and
30]. However, the initial material cost is still high. porosity [58]. In recent years, more researchers have focused on the
After 2000, instead of pursuing high strength, the focus of the UHPC development of UHPC using new materials and advanced technologies
development switched to reduce its CO2 emissions and initial materials to resolve the associated issues and expand the adoption of this novel
cost for more eco-friendly and economical UHPC. To this end, various class of cementitious composites, while maintaining its high mechanical
methods have been employed: (1) reduce cement content (<850 kg/m3) properties and durability.
and silica fume content (<200 kg/m3) by using a high volume of sup­ This paper comprehensively reviews the new development of UHPC.
plementary cementitious materials (SCMs) and fillers [31–35]; (2) First, the design methodologies and typical UHPC designs are summa­
reduce binder content (<1200 kg/m3); (3) replace finely grounded rized. Second, the typical ingredients of UHPC are discussed, such as
quartz sand by conventional concrete sand or quartz sand [36–39]; (4) binder materials (e.g., cement, supplementary cementitious materials,
reduce fiber content by hybrid fiber systems [40–42]; (5) use standard and fillers), aggregates (e.g., fine and coarse type), chemical admixture
curing instead of heat curing for lower energy consumption [32,43–45]. (e.g., HRWR, viscosity modifying admixture, shrinkage reducing agent,
The representative mixture designs of the economical and eco-friendly and expansive agent), and reinforcing fibers. The unique properties,
UHPC are summarized in Table B of the Appendix. The performance recommended contents, and underlying mechanisms are elaborated for
of these UHPC mixtures is comparable with the commercial ones [31, each ingredient. Third, the state-of-the-art technologies to improve and
46]. However, the use of different by-products (e.g., fly ash and rice husk control the key properties of UHPC are summarized. The key properties
ash) may lead to inconsistent performance [47]. For example, the include workability, autogenous shrinkage, compressive performance,
chemical composition of fly ash highly depends on coal sources and flexural/tensile properties, and durability. Finally, representative ap­
combustion processes which may vary. Furthermore, many power plants plications of UHPC are introduced.
have changed the combustion process for environmental benefits,
increasing the carbon content in fly ash, which may compromise the 2. Design methodology
properties of UHPC [48]. For this reason, the current application of the
economical UHPC is mainly in small quantities, such as connections and UHPC mixtures are designed to achieve a high particle packing
joints of bridges [49]. Modern UHPC is still struggling to become a density (Fig. 2) that contributes to low porosity, high mechanical
mainstream material for wider applications. While UHPC usually works strengths, and impermeability. Anderson and Andreasen model is the
well on a lab scale, the mixing difficulty significantly increased for most widely used theoretical model to design the UHPC [59] to achieve
large-scale production which might lead to the malfunction of the mixer the maximum packing density of particles involved. However, this
[47]. Thus, the allowable volume for each batch was restricted. For method only considered the particles under dry conditions, which may
example, Berry et al. [50] used a pan mixer with a maximum capacity of not reflect the real particle packing of UHPC because the influence of
0.34 m3, and Mendonca et al. [51] used a pan mixer with a maximum water and other liquids are not considered [60]. Therefore, to obtain the
capacity of 0.45 m3, but the allowable batch size was limited to 0.1 m3 “real” maximum particle packing, the wet particle packing density
which significantly affected the construction efficiency. Besides, due to method was introduced [61]. However, the maximum packing density
the low w/b ratio and high cementitious materials content, the typical does not always result in the best performance of UHPC, thus the
value of autogenous shrinkage of UHPC is reported over 800 με [52]. performance-based method was developed. This section introduces

Fig. 1. Evaluation of durability for UHPC compared with HPC and conventional concrete [11].

2
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 2. Two-dimensional schematic particle packing of conventional concrete and UHPC. The UHPC has a higher particle packing density [20].

some useful methods to design UHPC: model-based methods (dry par­ determine the w/b. Finally, a fixed HRWR content (5% mass of cement)
ticle packing and wet particle packing methods) and performance-based was used to enhance the flowability.
methods. 2
∑
n
RSS = (Pmix (Di+1 i+1
i ) − Ptar (Di )) (2)
2.1. Particle packing model-based methods
i=1

where Pmix is the composed mixture (black dash line); and the Ptar is the
2.1.1. Dry particle packing method target mixture (black solid line) calculated from Eq. (1).
For the dry particle packing method, there are two typical models for However, the dry particle packing method does not consider the
the mixture design: the discrete model [21,62–64] and the continuous effect of water and HRWR on particle packing. In practice, water and
model [65–67]. The continuous model is more preferred in UHPC design HRWR play significant roles in particle packing and the performance of
because it can achieve a denser particle skeleton as illustrated in Fig. 3. UHPC [72].
Fuller [65] and Andersen [66] developed the first continuous model
by introducing target particle size distribution P(D). Considering the 2.1.2. Wet particle packing method
effect of minimum particle size on particle packing, Funk and Dinger When the fine particles are dry, the high particle friction hinders the
[67] developed the Modified Andreasen & Andersen model and intro­ increase of packing density [73]. The presence of water can mitigate the
duced Dmin as presented in Eq. (1). friction force. when fine particles are in the saturated or over-saturated
Dq − Dqmin state [74], the force can be eliminated. In addition, the addition of
P(D) = ( q ) × 100% (1) HRWR can affect the thickness of the water film on particles and thus the
Dmax − Dqmin
packing density of UHPC [72]. Therefore, a higher particle packing
where Dmin is the minimum particle size (μm) in all materials; Dmin is the density will be achieved using the wet particle packing model than that
maximum particle size in all materials; q is the distribution modulus. determined by the dry packing model, as shown in Fig. 5. A high wet
Brouwers and Radix [69] suggested distribution modulus (q) is in the packing density enhances the macro-meso-micro pore structures and the
range of 0–0.25 for concrete with a large amount of cementitious ma­ compressive strength of UHPC [72].
terials [46,70]. For UHPC, the q value was suggested in the range of The wet particle packing model was proposed considering the effect
0.21–0.25 [71]. Yu et al. [46] developed UHPC mixtures by using the of water and HRWR [61]. To obtain the wet particle packing density, the
Modified Andreasen & Andersen model. The distribution modulus q was following procedures should be conducted: (1) set the initial w/b; (2)
set at 0.23. The optimal volume ratio of each raw material was deter­ weigh the water and cementitious materials and mix them; (3) transfer
mined by fitting the composed gradation curve (Fig. 4, dashed line) with the mixture to a cylinder mold and weigh the amount of paste; (4)
the target curve (Fig. 4, solid black line) by using Eq. (2). After con­ calculate the solid concentration (Φ) and void ratio (u) based on Eqs.
firming the proportion of dry materials, the Puntke test was applied to (3)–(5); (5) repeat the above steps at a lower w/b ratio until the

Fig. 3. Two different packing models: (a) binary discrete model, and (b) continuous model [68].

3
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 4. Particle size distributions of representative dry ingredients, target, and composed curves of the designed UHPC [46].

Fig. 5. Particle packing models: (a) dry particle packing model; and (b) wet particle packing model. The wet particle packing model provides a higher packing
density [75].

maximum packing density is achieved. where M and V are mass and volume of paste in the cylindrical mold (the
M mold is of 62 mm diameter × 60 mm height); ρw is the density of water,
Vc = (3)
ρw uw + ρα Rα + ρβ Rβ + ργ Rγ ρα, ρβ and ργ the corresponding solid density of different cementitious
materials; Rα, Rβ and Rγ are the volumetric ratios of different cementi­
u = (V − Vc )/Vc (4) tious materials.
Fig. 6 shows an example of determining the minimum void ratio (u)
φ = Vc /V (5) and the maximum solid concentration (Φ). The optimum w/b can be

Fig. 6. Trends of solid concentration and void ratio of typical mixtures versus w/b. With this relationship, the optimal w/b for the highest wet particle packing can be
determined [60].

4
J. Du et al. Composites Part B 224 (2021) 109220

determined given the highest wet packing density. 3.1. Binder materials
However, the highest particle packing density does not always lead
to the expected performance of UHPC. For instance, high particle 3.1.1. Cement
packing does not ensure the high fire resistance of the UHPC mixture, Depending on the environmental conditions and applications, the
because a relatively high porosity is preferred to release pore pressure in types of cement used in conventional concrete include Type I to V and
a high-temperature environment. white cement [77] which can all be employed to develop the UHPC.
Type III [78] and white cement [14] are the most used types because
2.2. Performance-based method they can provide fast setting and strength development due to the high
C3S content and Blaine fineness. If a high early strength is not manda­
To design UHPC mixtures in compliance with desired performance tory, but a relatively low shrinkage is preferred, Type I cement is
for different applications, performance-based methods were proposed alternatively used as its cost and reactivity are relatively low.
[76]. Meng et al. [31] developed a performance-based method to opti­
mize non-proprietary UHPC mixtures. Fig. 7 illustrates the step-by-step 3.1.2. Supplementary cementitious materials
optimization process. First, the binder combination has three sub-steps: Representative SCMs include silica fume, fly ash, slag, glass powder,
(i) preliminarily select binder combinations by flow characteristics; (ii) and rice husk ash, which have been used in UHPC to reduce its cost and
narrow down the binder combinations considering particle packing, carbon footprint and improve properties. Chemically, SCMs feature a
flowability, and mechanical properties; and (iii) finalize the binder high amount of SiO2 or/and reactive CaO, which can promote cement
combinations according to rheology properties including plastic vis­ hydration through pozzolanic and/or hydraulic reactions, especially
cosity (μp) and yield stress (τ0). Second, the w/b for paste is determined when the w/b is extremely low. Table 1 summarizes the chemical
considering both workability and 28-day compressive strength. Subse­ compositions, physical properties, and typical contents of the most
quently, the sand gradation, binder-to-sand ratio, and steel fiber content common SCMs used in UHPC. Their effects on workability, mechanical
are determined by the modified Andreasen & Andersen model and strength, volume stability, and durability are compared with the
overall performance. The finalized non-proprietary UHPC mixture is “Reference mixture” (see Table C in Appendix).
adjusted to satisfy the requirements of the mini-slump and 28-day
compressive strength. 3.1.2.1. Silica fume. Silica fume is a typical but critical constituent for
The main advantage of the performance-based method is that the UHPC mixtures, with typical usage ranging from 5% to 25% depending
mixture design parameters, such as w/b, fiber content, and binder to on the volume of the binder. Due to the small particle size, the addition
sand ratio, can be directly and accurately determined based on the of silica fume (when <10%) improves the particle packing density of
relationship established between the variables and the target UHPC, thus increasing the workability. However, when the content
performance. exceeds 10%, the workability may be significantly reduced because the
high surface area [79] of silica fume is prone to adsorbing free water and
3. Typical ingredients of UHPC HRWR, which causes particle agglomeration and suppresses the cement
reaction [80,81]. Silica fume has a high SiO2 content (>90%), which
This section introduces the characteristics and working principles of provides a seeding effect and pozzolanic effect, promoting cement hy­
typical ingredients in new development of UHPC, including binder dration [82] and thus refines the microstructure of UHPC. Consequently,
materials, aggregates, reinforcing fibers, and chemical admixtures. the use of silica fume normally increases the mechanical strengths of
UHPC, and the refinement of pores benefits the durability of UHPC.

Fig. 7. Flow chart of a performance-based method for designing UHPC mixtures. The left part is the performance-based procedure, and the right part elucidates the
performance for each step [31].

5
J. Du et al. Composites Part B 224 (2021) 109220

Table 1
The summary of supplementary cementitious materials used in UHPC [31–33,46,83–92].

3.1.2.2. Rice husk ash. Rice husk ash is usually used to partially and/or development of UHPC [94]. Type C fly ash undergoes both hydraulic
entirely replace silica fume due to their comparable amount of amor­ and pozzolanic reactions, but Type F fly ash mainly experiences a
phous silica. The d50 of rice husk ash ranges from 5 μm to 20 μm, which pozzolanic reaction due to its lack of CaO. Fly ash commonly replaces
is 50–100 times larger than that of silica fume. However, rice husk ash cement with volume ratios of 40%–60% for Type C and 10%–30% for
possesses a higher surface area (i.e., 64,700 m2/kg) than silica fume (i.e., Type F. As listed in Table 1, the particle shape of fly ash is more spherical
18,500 m2/kg) due to its porous structure, making it prone to absorbing than cement, which reduces inter-particle friction, and thus increases
more free water and HRWR [93]. As a result, the workability can be the workability [70]. Additionally, d50 of fly ash ranges from 30 μm to
significantly reduced with the addition of rice husk ash. Additionally, 50 μm, which is coarser than cement and silica fume. Fly ash can miti­
the high water absorption can cause rice husk ash to absorb the mixing gate the self-desiccation-induced shrinkage because more free water is
water during the mixing process thus to gradually release to the UHPC available to retain the relative humidity in the pore solution. Due to the
matrix during the hydration process, which delays the decrease of in­ lower hydration reactivity, the replacement of cement with fly ash re­
ternal humidity of pores in UHPC and hence mitigates the autogenous tards the hydration process [95]. The pozzolanic reaction is slow but can
shrinkage, as shown in Fig. 8(a) [33]. As evidenced in Fig. 8(b), the refine the microstructure in long term [96]. The refinement of pores
addition of rice husk ash also refines the microstructure, which improves benefits the durability of UHPC. However, due to the high Al2O3 content
the durability as well as the mechanical performance of UHPC at longer in fly ash, the calcium in the solution is quickly consumed by aluminate
lifespans. to precipitate ettringite on the surface of fly ash, creating a calcium-sink
effect. The reduction of calcium concentration reduces the formation of
3.1.2.3. Fly ash. Type C and Type F fly ash have been used in calcium-rich hydration products at early ages. Therefore, the use of fly
ash suppresses the hydration process and reduces mechanical strengths

Fig. 8. Comparison of UHPC mixtures with silica fume and with rice husk ash: (a) internal humidity with time [33]; and (b) the pore size distribution [93].

6
J. Du et al. Composites Part B 224 (2021) 109220

at early ages [97,98]. thus promoting cement hydration degree [107,108]. Depending on the
particle size, the fillers are categorized into micro-scale and nano-scale
3.1.2.4. Slag. Depending on the main metal oxides, the slag used in fillers.
UHPC can include iron slag [31], copper slag [99], and barium slag
[100]. As a typical alternative SCMs, slag is usually used to replace 3.1.3.1. Micro-scale fillers. Quartz powder is a typical filler used in the
cement with a replacement ratio of 30%–60%, depending on the volume production of proprietary UHPC. Quartz is inert and does not participate
of binder. The slag particles have angular shapes (higher shape factor), in cement hydration when temperatures are lower than 150 ◦ C [109].
increasing inter-particle friction and thus reducing the workability of However, despite the high cost, the International Agency for Research
UHPC compared with using the same amount of fly ash. The d50 of slag on Cancer (IARC) has reported that it can cause untreatable silicosis
ranges from 30 μm to 50 μm, greater than that of cement, which benefits [110]. Alternative filler is therefore in demand.
maintaining a low shrinkage. Similar to the fly ash, the lower hydration Limestone powder has been proposed as an alternative type of micro
reactivity of slag retards the hydration process [99] and tends to refine filler for UHPC [34,111]. The cement replacement ratio was reported to
the microstructure of UHPC [101]). In contrast to fly ash, the slag has a be up to 50% by volume. The d50 of limestone powder ranges from 10
low Al2O3 content [97]. The addition of slag in UHPC has less retarding μm to 20 μm, similar to quartz powder. Due to its inert property,
effect on the development of mechanical strengths compared with fly spherical shape, and small particle size, the proper addition of limestone
ash [46]. powder enhances the workability of UHPC [112]. The use of limestone
powder also reduces the autogenous shrinkage and accelerates the hy­
3.1.2.5. Glass powder. Glass powder has been used to replace cement or dration reaction [34] due to (1) the filler effect, the limestone powder
silica fume of UHPC at a replacement ratio of 10%–50% (by volume of does not hydrate but provides extra nucleation sites for hydration of
binder), and d50 ranges from 1 μm to 20 μm [92]. Among different glass cement and/or other reactive particles; (2) the dilution effect, the
waste sources, soda-lime glass occupies 81% of total glass production replacement of cement with limestone powder increases the effective
and is the main source of glass powder. The smooth surface of glass water-to-cement ratio which allows more free water for cement hydra­
powder helps reduce inter-particle friction [102]. Additionally, the tion and assists in maintaining the internal relative humidity (IRH) [35].
dense microstructure of glass powder makes it possess a low surface However, if an excessive amount of limestone powder is used, the hy­
water absorption compared with other types of SCMs. Thus, the addition dration peak can be significantly reduced, and the porosity is increased
of glass powder improves workability. In a similar mechanism to the [34]. This is because the overuse of fillers may cause a shortage of
above-mentioned SCMs (i.e., fly ash and slag), the proper use of glass reactive materials, limiting the amount of hydration products generated
powder refines the pore structure of UHPC [103]. [35].
However, when used glass powder is used, a major complication is
the potential of an alkali-silicate reaction (ASR). The Na+ and K+ ions 3.1.3.2. Nano-scale fillers. Inactive nanomaterials, such as nano-CaCO3
released by glass powder may increase the pH value of pore solution to [113] and nano-clay (NC), have been introduced to UHPC as nano-scale
be higher than 13, which breaks the chemical bond of siloxane (e.g., ≡ fillers [114,115]. Due to the very high specific surface area, the addi­
Si–O–Si ≡) of glass and generates ≡ Si-Oˉ ions [104]. Then, ≡Si-Oˉ ions tional nucleation and growth sites provided by an appropriate content of
react with Na+ and K+ and produce expansive ASR gel that may cause nano-scale fillers can effectively promote the hydration reactions and
cracking, as shown in Fig. 9. It was reported that the use of fine particle refine the microstructures of the UHPC matrix [116]. Besides, NCs have
size glass powder (less than 50 μm) alleviated the ASR. This is because: been found to increase the yield stress and apparent viscosity, as shown
(1) the pore size in UHPC may accommodate the ASR gel generated by in Fig. 10(a). This is because nanoparticles of NCs can fill the gaps be­
small size glass powder; and (2) fine particle size glass powder can react tween cement particles or pastes agglomerates and the increased phys­
with calcium hydroxide at early ages to produce C–S–H gel through a ical contact points can lead to the improvement in the rheology of UHPC
pozzolanic reaction, which helps binds alkali ions, densify the micro­ [117]. However, given a fixed dispersion method, an excessive amount
structure, and improves the crack resistance of the matrix [105]. of nano-fillers may cause agglomeration due to inter-particle adhesion
via weak forces [118], and thus introduce defects with increased
3.1.3. Fillers porosity, as shown in Fig. 10(b) [119]. Two main methods have been
Due to the low w/b in UHPC, the hydration degree of cement at 28 used for the uniform dispersion: (1) conduct ultra-sonification: the ul­
days is usually less than 35% [106], meaning that the majority of cement trasonic cavitation generates high shear forces that break particle ag­
particles remain unhydrated. The unhydrated cement can be replaced by glomerates into single dispersed particles [120], and (2) add surfactants:
inert fillers without mitigating the properties of UHPC. The main func­ the surfactants convert the hydrophobic surface of nanomaterials into a
tions of the filler are to maximize the particle packing density and hydrophilic surface for better dispersion in the aqueous phase [121].
provide extra sites for the nucleation and growth of hydration products,

Fig. 9. Illustration of glass particle size on ASR in a cementitious matrix [105]: (a) before any reaction of glass particles, and (b) after chemical reactions of
glass particles.

7
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 10. UHPC with nano-scale fillers: (a) the effect of NC particles on the rheology of cement pastes [122]; and (b) the agglomerated NC particles [123].

3.2. Aggregates sand and masonry sand compared with quartz sand leads to a weaker
interfacial transition zone (ITZ) between sand and UHPC matrix [31,37,
Finely grounded quartz sand with sizes ranging from 150 μm to 600 49]. Optimization of the particle packing should be conducted to
μm is usually used to prepare UHPC. However, the cost of fine quartz minimize the ITZ.
sand is high. Alternative aggregates are therefore proposed for the new Limestone sand has a low cost, homogenous composition, abundant
development of UHPC. Table 2 summarizes the chemical compositions reserves, and worldwide availability [130–132]. Yang et al. [133]
and physical properties of the alternatives, and their effects on different validated the feasibility of using recycled rock dust in place of fine
properties of UHPC. These properties are compared with the “reference quartz sand to prepare UHPC. It was reported that the raw cost per unit
mix” (the mixture design is shown in Table C of the Appendix). volume of UHPC was reduced by 40% without sacrificing the work­
ability and mechanical performance [133].
3.2.1. Fine aggregates Most recently, the lightweight porous sand has been utilized to
River sand is the primary fine aggregate used to replace quartz sand produce UHPC, such as lightweight sand [128,134] and volcanic rock
[31,36,49]. However, the particle size of river sand ranges from 0 mm to sand [126]. The porous structure of this sand allows for internal curing
4.75 mm, with the maximum particle size 5–8 times larger than that of in the UHPC. As shown in Fig. 11, the pre-saturated internal curing
fine quartz sand. Consequently, fully replacing the fine quartz sand with agents (i.e., porous sand) hold the curing water during concrete mixing
river sand can undermine the particle packing density. To improve the and release the water during concrete hydration. This increases the
particle packing, finer and more uniformly sized masonry sand (size cement hydration degree and reduces the self-desiccation effect by
range: 0–2 mm) was incorporated [31]. However, masonry sand is increasing the IRH [135,136]. Internal curing can provide more effective
produced by crushing and grinding coarse aggregates, causing more curing than external moisture curing and help develop UHPC with low
angular particles than river sand which adversely affects the workability shrinkage and high mechanical properties [126–128].
of UHPC [129]. Furthermore, the larger average particle size of river

Table 2
The summary of aggregates used in UHPC [31,36,38,55,124–128].

8
J. Du et al. Composites Part B 224 (2021) 109220

3.3. Chemical admixtures

Chemical admixtures are used to modify the fresh and/or hardened

properties of UHPC [142]. This section reviews the typical chemical
admixtures used for UHPC and the underlying mechanisms.

3.3.1. High-range water reducer

High-range water reducer (HRWR) has been used in UHPC to achieve
a self-consolidating property [114]. Among different types. the poly­
carboxylate ethers (PCE) HRWR shows high efficiency in dispersing
cement particles [143]. As shown in Fig. 14(a), the PCE head with
negative charge is absorbed on the surface of cement particles, and the
steric hindrance between the side chains of the absorbed PCE HRWR
disperse the cement particles [144]. Attributed to the stereochemistry of
the allylether-based HRWR, it is preferred to absorb and disperse silica
fume particles. However, the absorbed HRWR on cementitious particles
hinders the hydration reaction (Fig. 14(b)), which prolongs the setting
time and decreases the mechanical strengths at early ages.

3.3.2. Viscosity modifying admixture

Fig. 11. Schematic illustration of the mechanism of internal curing agents used Viscosity modifying admixture (VMA) is composed of water-soluble
in UHPC [128]. polymers and can be used to improve the rheological properties of UHPC
[146]. The most commonly used VMAs in cement-based materials are
3.2.2. Coarse aggregates cellulose-ether derivatives and polysaccharides from microbial sources
Basic principles of developing UHPC have been established by such as welan gum [116] and diutan gum [147]. The working mecha­
Richard and Cheyrezy [20], in which the coarse aggregates (i.e., size nisms of VMAs involve the combination of physicochemical effects: (1)
greater than 4.75 mm) are generally eliminated. This is attributed to the Water retention: the long chain of VMA polymers can adsorb free water
negative effects of coarse aggregates: (1) the angularity of coarse ag­ and swell to increase the viscosity (Fig. 15(a)). (2) Polymer–polymer
gregates significantly reduces the initial packing density of UHPC [38]; interaction and entanglement: adjacent VMA polymers can develop
(2) the stress concentration at the contact between aggregates leads to attractive forces to form a gel-like network, blocking the motion of water
weakness in the UHPC matrix [137]; and (3) the ITZ between coarse and thus increasing viscosity (Fig. 15(b)). (3) Polymer–particle inter­
aggregate and UHPC matrix is weaker than the ITZ between sand and action: the VMA polymers can absorb cement particles, which increases
UHPC matrix (Fig. 12) [138]. cement particle size, thus strengthening the network and increasing
However, the benefits of using coarse aggregates in the UHPC have viscosity (Fig. 15(c)).
been also reported: (1) decreasing the volume of cementitious paste,
which increases the elastic modulus and reduces the shrinkage of UHPC, 3.3.3. Shrinkage reducing admixture
as shown in Fig. 13(a) [139]; and (2) enhancing the penetration impact Shrinkage reducing admixture (SRA) is composed of organic chem­
resistance [140,141]. Due to the low cost, high elastic modulus, and icals [148] - one hydrophilic (polar) head and one hydrophobic
inert property, basalt and limestone have been adopted [38,39]. How­ (non-polar) tail [149]. The main role of SRA is to reduce the autogenous
ever, the results show that when the maximum particle size of aggregate shrinkage of UHPC by reducing the surface tension in the pore solution
was increased from 3 mm to 16 mm, the 28-day compressive and tensile of the UHPC matrix. As shown in Fig. 16, the non-polar tail of SRA can
strengths were reduced by 10% and 15%, respectively [38], as shown in absorb the particles at the water-solid interface, and the electrostatic
Fig. 13(b). repulsive force between the adjacent polar heads compensates for the
shrinkage-induced force and thus reduces shrinkage [150]. Addition­
ally, SRA can absorb air at the water-air interface, retarding water
evaporation, thus maintaining the IRH and mitigating the
self-desiccation-induced autogenous shrinkage [151].

3.3.4. Expansive agents

Expansive agents are powders that can generate expansive products
during the hardening of UHPC. As shown in Fig. 17, the expansive stress
compensates for the shrinkage stress, and thus reduces the shrinkage of
UHPC. There are three main types of expansive agents: calcium
sulfoaluminate-type [152], CaO-type [153], and MgO-type [154]. The
calcium sulfoaluminate-type expansive agents produce ettringite in
UHPC [155]. However, because ettringite consumes a lot of water, the
workability of UHPC can be significantly reduced, and therefore the
capacity for reducing shrinkage is limited [156]. Also, ettringite is
thermally unstable which is decomposed at 70 ◦ C [157]. Compared with
the calcium sulfoaluminate-type, the CaO-type expansive agents need
less water to form portlandite which is more stable than ettringite [158].
However, the CaO-type expansive agents react fast at early ages, which
prohibits the reduction of shrinkage at later ages. The MgO-type
expansive agents have been proposed to compensate shrinkage at later
Fig. 12. Schematic illustration of the ITZ between coarse aggregate and UHPC ages [156] because the reaction between MgO-type expansive agents
matrix [138]. and to produce the Mg(OH)2 is slow.

9
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 13. Effects of coarse aggregate on the performance of a UHPC: (a) the autogenous shrinkage [139]; and (b) the compressive and tensile strengths [38].

Fig. 14. Effects of PCE HRWR on UHPC: (a) fundamental mechanism of steric repulsion effect [144]; and (b) retarded hydration [145].

Fig. 16. Working mechanism of the SRA: the shrinkage-induced force is

compromised by the electrostatic force between the repellent SRA mod­
ules [150].

3.4. Fibers

Fibers are used in UHPC to improve the tensile properties by

restraining the initiation and propagation of cracks [160–162]. The
characteristics of fibers, including fiber materials (texture), aspect ratio,
shape, and surface conditions, significantly influence the properties of
UHPC, as shown in Table 3.

3.4.1. Effect of fiber materials

Steel fibers are the most commonly used fibers in UHPC [41]. Meng
et al. [41] indicated that as the steel fiber content increased from 0% to
2% (by mass of mixture), the 28-day flexural strength and toughness of
Fig. 15. Schematic illustration of VMAs working mechanisms [146]: (a) water UHPC were increased by 120% and 3360%, respectively. However, steel
retention; (b) polymer-polymer interaction; and (c) polymer-particle
fibers have drawbacks such as (1) high initial cost, (2) corrosion po­
interaction.
tential [172,173], (3) adverse influence on structure surface finishing

10
J. Du et al. Composites Part B 224 (2021) 109220

with 2% steel fibers [166]. Alternatively, synthetic fibers such as poly­

ethylene (PE) fibers have been applied. However, the hydrophobic
surface of synthetic fiber leads to a porous ITZ between fiber and matrix,
which may improve the tensile strain capacity (ductility) by introducing
multi-cracking patterns with delayed growth of localized major cracks
[175] but compromises the tensile/flexural strength [169]. On the
contrary, mineral fibers (e.g., basalt fiber) have similar chemical com­
positions to the UHPC matrix, generating a strong chemical bond and
achieving higher cracking stress [176]. However, the ductility is
compromised because cracks initiated due to the rupture of fibers rather
than being pulled out of the matrix [177], the crack propagates quickly,
which leads to failure of the UHPC.
Along with enhancing tensile/flexural properties, some fibers made
Fig. 17. Working mechanism of expansive agents: the expansive stress com­ with special materials can add beneficial functions [178–183] to the
pensates shrinkage induced stress [159]. UHPC. For instance, due to the excellent electrical conductivity of car­
bon, self-sensing UHPC has been developed by adding carbon nanofibers
[174], and (4) high density (add dead load). [184]. As shown in Fig. 18(a), the tensile (bending) process gradually
To overcome the drawbacks of steel fibers, glass fibers with similar generated cracks in the UHPC matrix, and the cracks affected the
tensile strengths have been used [166]. However, the flexural strength of conductive pathway and increased the electrical impedance. Therefore,
UHPC with 2% glass fibers is reduced by 25% compared to that of UHPC the stress-strain curve of the UHPC was satisfactorily monitored by
measuring the fractional change in impedance. In addition, materials

Table 3
The summary of fibers used in UHPC [41,163–171].

Fig. 18. Mechanisms of multifunctionality of UHPC: (a) self-sensing property due to the use of carbon fibers [184]; and (b) fire-resistance property due to the use of
polypropylene fibers [185].

11
J. Du et al. Composites Part B 224 (2021) 109220

with low melting temperatures such as polypropylene fibers have been straight fibers. The flexural strengths of UHPC with corrugated fibers
used to develop fire-resistance UHPC [185]. When concrete is exposed to and hooked fibers were 5% and 20% greater than that of UHPC with
elevated temperatures, the melted fibers create channels to alleviate straight fibers. The bond strength between the UHPC matrix and twisted
internal vapor pressure and prevent explosive spalling, as shown in triangular fibers was 40% higher than that of straight circular fibers. As
Fig. 18(b). a result, the tensile strength of UHPC with twisted triangular fibers was
35% greater than that of UHPC with straight circular fibers.
3.4.2. Effect of fiber aspect ratio
The aspect ratio is the ratio between fiber length (Lf) and fiber 3.4.4. Effect of fiber surface condition
diameter (Df) [186]. In most cases, steel fibers with a higher aspect ratio Other than using deforming the fibers, applying treatment to
provide better flexural performance (e.g., flexural strength, deflection roughen the fiber surface can also improve the fiber-matrix bond
capacity, and toughness) [57,187]. This is because the reinforcing fibers [195–199]. Several treatment methods have been developed, including
with a higher aspect ratio have a larger effective fiber-matrix bond area, using chemical solutions immersion [194], sandpaper [200], and cold
and thus provide better post-cracking behavior [188]. Yoo et al. found gas plasma [201]. Based on the quantitative evaluation by atomic force
that when the volume of steel fibers was fixed at 2%, an increase in the microscope images, after 6 h of immersion in an ethyl­
fiber aspect ratio from 65 to 97.5 increased the flexural strength and enediaminetetraacetic acid solution, the roughness of steel fibers was
toughness of UHPC by 40% and 105%, respectively [57]. Moreover, the increased by 290%, as shown in Fig. 20(a) and (b) [194]. Therefore, the
number of cracks in UHPC with steel fiber (Lf/Df = 97.5) increased by tensile strength and strain capacity of UHPC with the roughened steel
125%, and the average crack spacing was reduced by 45%. fibers were increased by 15% and 16%, compared with those of UHPC
with untreated steel fibers, as shown in Fig. 20(c).
3.4.3. Effect of fiber shape
Many shapes of metallic fibers include straight, hooked, corrugated, 4. Improve and control the key properties of UHPC
and twisted [40,41,163] have been used in the UHPC. In particular, the
deformed fibers show better performance in enhancing the tensile/­ This section addresses the solutions to the challenges related to the
flexural properties of UHPC [189,190]. Fig. 19 summarized the load-slip workability, autogenous shrinkage, mechanical property, and durability
curves of different shaped steel fibers [191]. For straight fibers, the of UHPC, as identified in Section 1.
debonding and friction at the fiber-matrix interface govern the pull-fiber
behaviors [192]. For deformed fibers, additional resistance can be pro­
vided by the mechanical anchorage/interlock [40]. After the elastic 4.1. Control workability
stage (before A1, A2, and A3), different shapes caused different load-slip
responses. For the straight fiber, the peak load (B1) was followed by To facilitate the construction and ensure homogeneity and dense
debonding (B1–C1). Then, the fiber was pulled out from the matrix microstructure of UHPC, the workability must be well controlled. If the
under friction (C1–S) [192]. For the hooked fiber, a steady region workability is too high, high-density materials (e.g., steel fibers) will
(C3-D3) with a constant load was observed after the pull-out load segregate [202] and low-density materials (e.g., synthetic fibers and
dropped from the peak, which can be attributed to the mechanical lightweight sand) will float upward; if the workability is too low, UHPC
anchorage/interlock provided by the hooked end [193]. Finally, the fi­ cannot be self-consolidated, and the dense microstructure required for
bers were pulled out from the matrix under friction (D3-S). For the UHPC cannot be achieved. The methods for adjusting the workability
corrugated fiber, multiple steady regions were observed (C2 and D2), are introduced below.
because the curved shape led to straightening processes that can carry
sustained loads [191]. 4.1.1. Improve flowability by regulating hydration
Researchers have quantitatively evaluated the effect of fiber shape As mentioned in the introduction, the flowability of UHPC de­
on fiber-matrix bond strength [40,194]. The bond strengths of corru­ preciates quickly during large-scale field construction, especially at
gated and hooked fibers were 200% and 600% higher than that of elevated temperatures. This problem is mainly attributed to the initial
hydration of cementitious materials, in which the tricalcium aluminate

Fig. 19. Comparison of the typical pull-out load-slip curves of different shaped steel fibers.

12
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 20. Effect of surface roughness of steel fibers and tensile properties [194]: (a) atomic force microscope image of untreated steel fibers surface; (b) image of
roughened steel fibers surface (6-h immersion); and (c) results of tensile properties of UHPC with different roughness of fibers.

(C3A) and gypsum in cement react with water and generate a large 4.1.2. Control rheology by adjusting chemical admixtures
amount of ettringite that increases the connections and interparticle The rheological properties of the suspension must be controlled to
frictions [203]. Worse still, higher temperatures can increase the prevent inhomogeneity (Fig. 23) [10]. Meng et al. used VMA for the
dissolution of C3A and gypsum in water, accelerating the initial hydra­ rheology control [10,206]. It was found that for UHPC with 2% steel
tion and exacerbating the reduction of workability. Therefore, delay or fibers, the plastic viscosity of the suspending mortar should be adjusted
reduce the rate of hydration is key to improve the flowability of UHPC. to around 52 Pa s (optimum value) to obtain the best fiber uniformity.
However, the optimum plastic viscosity is dependent on the type and
4.1.1.1. Physical control of early-age hydration. Because the dissolution content of fibers. For example, for UHPC with 1% and 3% steel fibers,
rate of C3A, gypsum, and other ions for the hydration reaction signifi­ the optimal plastic viscosity was 36 Pa s and 66 Pa s, respectively [206].
cantly reduces with a decrease in temperature [203], one of the main
methods to slow down the early-age C3A reaction is to reduce the ma­
terial temperature. To this end, some researchers partially replaced the 4.2. Reduce autogenous shrinkage
mixing water with crushed ice. Xie et al. [204] found that when 50% of
fresh water was replaced with crushed ice, the temperature of cement The autogenous shrinkage of cementitious materials is attributed to
paste was reduced from 25 ◦ C to 1.2 ◦ C (Fig. 21(a)) and as shown in the internal self-desiccation and chemical shrinkage during the hard­
Fig. 21(b), when the mixing temperature was reduced from 30 ◦ C to 10 ening [207]. The typical value for autogenous shrinkage of UHPC is
◦
C, the rate of hydration heat was significantly reduced in the first 50 larger than 800 με [52], which has a high cracking potential [53–56].
min, and thus benefits the workability. Reduction of autogenous shrinkage, especially at the early age, is
necessary. Fundamentally, four mechanisms control the development of
autogenous shrinkage: (1) degree of hydration, (2) reduction rate of
4.1.1.2. Chemical control of early-age hydration. Another method is to
IRH, (3) restraint of volume change, and (4) internal forces for
use high volume SCMs and/or inert fillers to replace the cement and/or
compensating shrinkage-induced tension. Based on different mecha­
silica fume in UHPC [9–11]. The effect of SCMs and inert fillers on hy­
nisms, autogenous shrinkage mitigation strategies have been summa­
dration reactions is introduced in Section 3.1. Yang et al. [205] found
rized in Table 4 and elaborated in the following.
that, as the replacement ratio of slag was increased from 0% to 50% (by
mass of cement), the mini-slump spread of UHPC was increased from
4.2.1. Reduce chemical shrinkage by regulating hydration reaction
260 mm to 310 mm (by 20%) and the dormant period was extended
Chemical shrinkage due to cement hydration is the main contributor
from 2.6 h to 8 h, as shown in Fig. 22. The underlying mechanism is that
to autogenous shrinkage of UHPC [215]. Compared with cement, SCMs
the replacement of cement by slag significantly reduces the reactive C3A
have lower calcium content and hydraulic reactivity, which hinder the
content in the UHPC.
hydration process of UHPC and reduce autogenous shrinkage. The

Fig. 21. Physical control: (a) effect of ice replacement on temperature reduction [204]; (b) hydration heat flow of cement paste (w/b = 0.3) at different temperatures
(preliminary results).

13
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 22. Chemical control using SCMs [205]: (a) the mini-slump spread; and (b) hydration heat.

Fig. 23. Inhomogeneity of UHPC with a low viscosity [10]: (a) segregation of steel fibers; and (b) floating of lightweight sand.

fume was fully replaced by fly ash or slag, the 28-day autogenous
Table 4
shrinkages of UHPC were reduced by approximately 15%. It is note­
Summary of strategies to mitigate autogenous shrinkage.
worthy that the ability of rice husk ash to mitigate the autogenous
Mechanism Mitigation Recommended Mitigation Reference shrinkage is stronger than other alternatives due to the porous structure
method content (%) rate (%)
which provides an internal curing effect to reduce the drop of IRH in
Degree of Fly ash 20–40a 18–24 [31,208] UHPC [33].
cement Slag 30–60a 15–20 [31,208]
hydration Rice husk ash 100b 45–75 [33,209]
Limestone 20–60c 10–30 [35,210]
4.2.2. Reduce self-desiccation by increasing internal relative humidity
Reduction rate Superabsorbent ≤0.5c 80 [211] Self-desiccation occurs due to the reduction of IRH. Internal curing
of internal polymer agents can be used to maintain the IRH by gradually releasing internal
relative Cellulose fibers 0.06d 35 [168] curing water [216]. Different types of internal curing agents have been
humidity Lightweight 25–50e 35 [128]
applied to UHPC, such as superabsorbent polymer (SAP) [217], porous
sand
Porous pumice 30e 60 [127] fine aggerate [127,128], and cellulose fibers [168].
Restraint Aggregates 30 40 [139]
degree for Reinforced fiber 1–3 20–55 [40] 4.2.2.1. Superabsorbent polymer. SAP is a type of polymer with a cross-
volume
linked surface that can absorb water by an osmotic pressure effect
change
Introduction of Shrinkage ≤3c 45–60 [212, [218–221]. Liu et al. [222] found that, as the SAP content increased to
internal reducing 213] 0.6% (by mass of binders), the 7-day IRH was increased from 76% to
forces to admixture 90%, and the 7-day autogenous shrinkage was reduced from 610 με to
compensate Expansive agent – 15–55 [212, 189 με (by 70%), as shown in Fig. 24.
shrinkage- 214]
induced
However, as the SAP content increased to 0.6%, the 91-day
tension compressive strength of UHPC was reduced from 121 MPa to 111 MPa
(by 8%). The negative effect of the SAP addition on strength develop­
Note: a by volume of binder, b by volume of silica fume, c by mass of cement,
ment was also reported by other researchers [211,223]. The strength
d by volume of mixture, e by volume of sand, - stands for the recommended value
is uncertain which is highly related with the type of EA, SRA stands for shrinkage
reduction is mainly due to the formation of voids after the water
reducing agent, and SAP stands for superabsorbent polymer. desorption from SAP, as shown in Fig. 25 [211]. The voids can increase
the porosity of the UHPC matrix and become weak zones.
replacement of cement and/or silica fume with less reactive SCMs or
inert fillers can regulate the hydration reaction and reduce the autoge­ 4.2.2.2. Porous sand. Porous sand such as coral sand, expanded shale
nous shrinkage. For instance, Kang et al. [35] found that, as the lime­ sand, and recycled concrete sand can also provide internal curing and
stone replacement ratio was increased from 0 to 50% (by mass of has advantages over SAP [218], such as high stability and maintained
cement), the 28-day autogenous shrinkage of UHPC was reduced from mechanical strength. The absorption capacity of porous sand is between
818 με to 550 με (by 32%). Ghafari et al. [208] found that, as the silica 10% and 20% [127,128]. Meng and Khayat [128] found that as the

14
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 24. Experimental results of the effect of SAP on (a) IRH; and (b) autogenous shrinkage [222].

Fig. 27. Effect of the LWS content on the porosity of UHPC after hydrating for
28 days [128].

4.2.2.3. Cellulose fiber. Cellulose fibers have been used as internal

Fig. 25. Detail of SAP voids in UHPC matrix [211]. Note: green color indicates
curing agents due to the meso-porous structure and high water ab­
porosity caused by SAP, and pink color indicates CH.
sorption rate [225]. As shown in Fig. 28(a), the slow water release rate
of cellulose fiber at high relative humidity levels (>70%) is advanta­
concentration of pre-saturated porous sand (i.e., expanded shale sand) geous for UHPC because of longer internal curing, and thus benefits the
increased from 0% to 75% (by mass of river sand), the 7-day IRH was autogenous shrinkage reduction, as shown in Fig. 28(b).
increased from 83% to 92%, and the 28-day autogenous shrinkage was
reduced from 490 με to 197 με (by 60%), as shown in Fig. 26. 4.2.3. Restrain volume change
As shown in Fig. 27, proper incorporation (≤25%) of pre-saturated
Shrinkage can be reduced by providing internal restraints and/or
porous sand promotes cement hydration that produces hydration reinforcement such as fibers and aggregates [156]. The mechanism and
products to fill the pores, which reduces porosity, densifies the micro­
effects of aggregates on autogenous shrinkage reduction are elucidated
structure, and strengthens the ITZ. However, excessive porous sand in Section 3.2. Reinforcing fibers such as steel fibers can help maintain
increased porosity and reduced the strength of UHPC [10]. A vacuum
the volume, resist shrinkage stress, and reduce autogenous shrinkage of
saturated method has been proposed to retain the IRH for a longer time, UHPC [156]. The effectiveness of the fibers is dependent on fiber
which may help minimize the use of porous sand [224].

Fig. 26. Results of lightweight sand content on (a) internal relative humidity; (b) autogenous shrinkage. The Figures are adopted from Ref. [128].

15
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 28. Effects of cellulose fibers on autogenous shrinkage of UHPC [226]: (a) the water release rate; and (b) the autogenous shrinkage of UHPC with cellu­
lose fibers.

characteristics. For example, as the amount of straight steel fiber was solution, which retards the dissolution of C3S [229]. To mitigate the
increased from 0% to 3% (by volume), the 72-h autogenous shrinkage negative effects, Soliman and Nehdi [230] suggested combining the use
was decreased by 50%, as shown in Fig. 29(a) [40]. Given the me­ of wollastonite microfiber.
chanical interlock effect, the incorporation of deformed fibers can
further improve the transferring of shrinkage-induced stress and reduce 4.3. Improve compressive performance
the autogenous shrinkage. As shown in Fig. 29(b), when straight steel
fibers were changed to hooked steel fibers, the 72-h autogenous Generally, UHPC has excellent compressive strength due to the low
shrinkage was further decreased by 25%. porosity and refined microstructure [231]. In case further improvement
of the compressive strength is needed, efforts can be focused on (1)
4.2.4. Introduce internal forces reducing the porosity, and (2) densifying the hydration products.

4.2.4.1. Induce expansive force. Using expansive agents to induce an 4.3.1. Reduce porosity
expansive force can compensate for the shrinkage in UHPC (Fig. 30(a))
[156]. The mechanism and different types of expansive agents are 4.3.1.1. Improve particle packing density. Improving particle packing
introduced in Section 3.3.4. However, the 28-day compressive strength density leads to smaller voids in the matrix and creates more contact
of UHPC can be slightly reduced by introducing the expansive agent, as points among particles, thus improving the compressive strength of
illustrated in Fig. 30(b). The expansive products (e.g., ettringite and UHPC [232]. Reducing w/b and optimizing the gradation of particles
portlandite) densify the microstructure of UHPC at early ages, but the are the typical ways to increase the particle packing. As shown in Fig. 32
water consumed by expansive agents reduces the hydration degree of (a), when the cement content increased from 30% to 100% (by mass of
cement. binder), the packing density and compressive strength of UHPC
increased and then decreased. The maximum packing density and
4.2.4.2. Induce repulsive force. During cement hydration, the capillary compressive strength were simultaneously reached when the cement
stress is increased due to the surface tension of pore solution [156], thus content was 80% [233]. Meanwhile, as shown in Fig. 32(b), when the
causing shrinkage. SRA reduces the surface tension by introducing binder systems were fixed, use glass sand with optimized size also
repulsive forces between particles, as detailed in Section 3.3.2. Su et al. benefits the particle packing, and thus the compressive strength of UHPC
[212] found that as the SRA content was increased from 0 to 2% (by [91].
mass of cement), the 7-day autogenous shrinkage of UHPC was reduced
by 45% (Fig. 31(a)). However, as shown in Fig. 31(b), the use of SRA 4.3.1.2. Promote hydration reaction. The promotion of cement hydra­
compromised the mechanical strength of UHPC. This can be attributed tion generates more hydration products, which refine the microstructure
to: (1) the SRA is adsorbed on the water-solid (air) interface and gen­ of UHPC. According to the Power’s model [234], the ultimate hydration
erates “organic molecular film” that hinders hydration reaction [213, degree of a typical UHPC with w/b between 0.20 and 0.22 ranges from
228]; and (2) the admixture reduces the alkali contents in the pore 47% to 52%. The unhydrated cement particles can be replaced by fillers

Fig. 29. Results of autogenous shrinkage of UHPC [40]: (a) effect of steel fiber content; and (b) effect of steel fiber shape.

16
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 30. Results of expansive agent (EA) on (a) autogenous shrinkage and (b) compressive strength of UHPC. The figures are adopted from Ref. [227].

Fig. 31. Results of SRA on (a) autogenous shrinkage, and (b) compressive strength of UHPC (by mass of cement, %) (w/b = 0.2). The figures are adapted
from Ref. [212].

Fig. 32. The relationship between initial packing density and compressive strength: (a) packing density variation by adjusting binder combinations [233]; and (b)
packing density variation by adjusting sand gradation [91]. Note: GS-225 stands glass sand with the d50 of 225 μm.

and/or SCMs for sustainability purposes, and to promote hydration re­ successfully used porous sand as an internal curing agent to prepare
actions that further refine the microstructure and improve the UHPC with reduced autogenous shrinkage and improved compressive
compressive strength, as shown in Fig. 33. As elaborated in Section strength.
3.1.3, fillers provide extra nucleation sites for precipitation of cement
hydration products and thus promote the hydration reaction. However, 4.3.2. Densify hydration products
when the addition of nanomaterials exceeds a certain amount, the Hydration products can be densified by applying heat curing or/and
compressive strengths are reduced due to the agglomeration of nano­ promoting pozzolanic reaction to strengthen UHPC matrix and ITZ, thus
materials forming weak zones [235]. Other nanomaterials follow a improving the compressive strength of UHPC. Fig. 34 shows the
similar trend, as reported by other researchers [236–238]. compressive strength and proportion of different types of C–S–H. As the
As elaborated in Section 3.2.1, applying internal curing effectively fraction of ultra-high-density C–S–H increased from 15.4% to 100%, the
promotes cement hydration in UHPC [240]. Meng et al. [128] compressive strength increased from 150 MPa to 180 MPa (by 20%).

17
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 33. Results of filler effect on compressive performance of UHPC: (a) micro-scale filler [34]; and (b) nano-scale filler [235,239].

Fig. 34. Results of heat curing on compressive strength and fractions of different types of C–S–H in UHPC [241].

4.4. Improve flexural/tensile properties

By using high-volume reinforcing fibers (≥2%, by volume of

mixture), UHPC can achieve strain-hardening behavior. However, other
than the high cost, a high fiber content can cause handling issues (e.g.,
hard to mix and low workability due to the fiber agglomeration) [57,
242,243]. To minimize the fiber content while retaining or improving
flexural/tensile properties, innovative methods have been proposed: (1)
improving the fiber dispersion and fiber orientation; and (2) incorpo­
rating multi-scale reinforcements.

4.4.1. Improve fiber dispersion and fiber orientation

4.4.1.1. Improve fiber dispersion. Fiber dispersion represents the layout

of the reinforcing fibers in UHPC and is quantified by the deviation of
the number of fibers per unit area from the total number of fibers in the Fig. 35. Relationship between fiber dispersion coefficient and flexural strength
whole cross-section. Fig. 35 shows that uniformly dispersed fibers lead of UHPC [246].
to a higher flexural strength of UHPC beams [244] because an even fiber
dispersion can improve the encapsulation effect of the UHPC mortar on Fig. 36(b). The slight reduction of flexural strength was due to an in­
the fiber surface and increase the fiber-matrix interfacial properties crease of entrapped air in the UHPC matrix due to the excessive addition
[245]. of VMA.
Fiber dispersion can be controlled by the viscosity of the UHPC
suspending mortar [10]. According to Ref. [206], overall, the coefficient 4.4.1.2. Improve fiber orientation. The optimal fiber orientation aligns
of fiber dispersion (α) increased with the plastic viscosity of UHPC the reinforcing fibers along the direction of tensile stress [247]. For the
mortar, as shown in Fig. 36(a). Meanwhile, the flexural strength and UHPC cast with the conventional method, the fibers are randomly ori­
toughness increased with the coefficient of fiber dispersion, as shown in ented [248]. According to single-fiber pullout test results, when the

18
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 36. Results of rheology control for UHPC: (a) the plastic viscosity of UHPC suspension mortar versus fiber dispersion; and (b) fiber dispersion versus the flexural
strength of UHPC [206].

angle of reinforcing fibers is over 30◦ off the tensile direction, the energy mm to 280 mm (Fig. 39(c)).
dissipation capacity and bond strength can be significantly reduced
[249–251]. When more fibers are orientated to the tensile direction, 4.4.2. Incorporate multi-scale/physics reinforcements
higher flexural properties of UHPC can be achieved, as depicted in
Fig. 37 [10]. 4.4.2.1. Use hybrid fibers. The flexural/tensile properties of UHPC are
The casting methods show a significant effect on the fiber orientation dependent on the material, aspect ratio, shape, and surface condition of
[160,252,253]. The best practice for casting a UHPC beam is to induce fibers. Usually, a single type of fibers are used to prepare UHPC,
self-orientation of fibers by allowing the UHPC to flow from one side to designated as mono fiber. Recently, multiple types of fibers have been
the other side of a beam. Due to the high viscosity of UHPC, the velocity used to improve the flexural/tensile properties of UHPC, designated as
gradient of the shear flow of fresh UHPC drives the fibers to be oriented hybrid fibers.
along the flow direction [10]. It is noteworthy that the fiber orientation
is dominated by the size of the specimen [253,254]. Huang et al. [254] (1) Hybrid fibers with different materials
indicated that the free rotation of fibers was impeded near the formwork
boundaries so that the fiber orientation near the formwork is better than Hybrid fibers with different materials have been proposed to
that far from the formwork walls (i.e., wall effect), as shown in Fig. 38. improve multiple properties. For example, basalt fibers have been used
The thickness of the wall-effect zone is dependent on the dimension of to increase the first cracking strength due to their strong chemical bond
the molds [254], thinner specimens generally have better fiber orien­ with cement, although this can hinder the post-cracking behaviors and
tation when the same casting method is employed. ductility of UHPC. However, steel has excellent ductility, the hybrid of
Besides, Song et al. [253] found the fiber orientation in UHPC was steel and basalt fibers can result in high first cracking strength as well as
also improved by increasing the length of the casting mold. The casting good ductility. Kang et al. [42] reported that the first cracking stress of
mold was divided into three zones: (1) adjusting zone: fibers start to UHPC with 1.0% steel fiber + 0.5% basalt fiber was increased by 40%
re-orient; (2) optimized zone: most of the fibers align to the flow di­ compared with that of UHPC with 1.5% mono steel fibers [171]. Poly­
rection; and (3) re-disorder zone: fibers were disordered due to the ethylene fibers can be incorporated to further improve the ductility
reduction of flow velocity. As low Saver (the average area of fibers) [171] because their hydrophobic surfaces can create a porous
represents a good fiber orientation (Fig. 39(a)), the Saver value first de­ fiber-matrix interfacial zone and weak bond. Yoo and Kim [171] re­
creases with the increase of flow distance and increases at the other end ported that the strain capacity of UHPC with 1.0% steel fiber + 1.0% PE
of the mold (Fig. 39(b)). When the casting length increased from 400 fiber was increased by 45% compared to that of UHPC with 2% mono
mm to 1000 mm, the length of the optimized zone was increased from 0 steel fiber.

(2) Hybrid fibers with different aspect ratios/scales

Reinforcing fibers with different aspect ratios have been combined to

improve the flexural properties of UHPC through multi-scale bridge ef­
fects for macro-, meso-, and micro-scale cracks [186,255], as illustrated
in Fig. 40. For example, Meng and Khayat first proposed to use graphite
nanoplatelets and carbon nanofibers to significantly improve the flex­
ural properties of UHPC [106,174]. The effect of the combination of
straight steel fiber with different aspect ratios, e.g., lf/df = 65, 97.5, and
100, on the flexural strength was studied, and the results indicated that
the combination of 1% long steel fibers (lf/df = 100) with 1% medium
steel fibers (lf/df = 97.5) increased the 28-day flexural strength and
deformation by 10% and 25%, respectively, compared with UHPC with
2% mono long steel fibers [256].

(3) Hybrid fibers with different shapes

Reinforcing fibers with different shapes can be combined to improve

Fig. 37. Relationship between the fiber orientation and the flexural strength of
UHPC [246]. flexural properties. Deformed fibers can increase the mechanical

19
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 38. The illustration of the mechanism of the wall effect [254].

Fig. 39. Fiber orientation in UHPC: (a) effect of fiber orientation on Saver; (b) variation of Saver with flow distance; and (c) the optimized zone in different casting
lengths. Note: A low Saver represents a good fiber orientation. The length of the optimized zone increases with the increase of the casting length [253].

interlock between reinforcing fibers and the UHPC matrix, thus

improving the flexural/tensile properties of UHPC, as elucidated in
Section 3.4.3 [163]. However, in some cases, the use of high-volume
deformed fibers decreases the tensile/flexural properties of UHPC [57,
257]. This is because deformed fibers in the brittle matrix might
generate split cracks during the pull-out process, which creates weak
zones and deteriorates the reinforced effect of adjacent fiber (Fig. 41).
Further, the cost of UHPC using deformed fibers further increases due to
the additional manufacturing process of deformed fibers [57]. There­
fore, it is recommended to combine deformed fibers with straight fibers.
Meng and Khayat [41] found that when the steel fiber content was fixed
at 2%, hybrid 1% straight steel fiber with 1% hooked steel fiber achieved
the optimal flexural performance. The 28-d flexural strengths were
improved by 25% and 30%, and the toughness was increased by 30%
and 35%, respectively, compared with that of UHPC with 2% mono
straight steel fiber and UHPC with 2% mono hooked steel fiber.

4.4.2.2. Use fiber-reinforced polymer meshes with multi-directional rein­

forcement. Fiber-reinforced polymer (FRP) meshes can be used to
Fig. 40. The mechanism of hybrid fiber reinforcement (with different fiber strengthen UHPC elements to achieve high flexural strength and
aspect ratios) on flexural performance of UHPC. toughness, corrosion resistance, and lightweight property [258,259].
Meng et al. [246,260] studied flexural behaviors of UHPC with steel

20
J. Du et al. Composites Part B 224 (2021) 109220

chemical ions (e.g., Cl− and SO2−4 ) and CO2 [91,264]. However, dura­
bility issues exist, which can be attributed to: (1) inappropriate use of
porous materials (e.g., SAP and porous sand); (2) crack development
under service loads; and (3) alkali-aggregate reaction when recycled
materials are used such as recycled concrete aggregate and glass. The
solutions for these potential problems are addressed in the following.

4.5.1. Optimize content of porous materials for dense microstructure

As mentioned in Section 4.2, the successful application of porous
sand as an internal curing agent in UHPC effectively decreased the
autogenous shrinkage of UHPC. Fan et al. [58] studied the effects of
Fig. 41. The use of mono-deformed fiber may cause split cracks and reduce the
porous sand on corrosion of steel reinforcing bars embedded in UHPC
flexural performance of UHPC [57].
and the resistance of the matrix [128]. The results showed that excessive
use of porous sand could compromise the corrosion resistance of steel
fibers and FRP meshes, including glass FRP (GFRP) and carbon FRP bars and lead to the shortcut of current by the chopped steel fibers in
(CFRP, shown in Fig. 42) meshes. The steel fibers restrain the micro­ UHPC. This is because, as the porous sand content exceeds the optimum
cracks, and the FRP mesh provides two-directional reinforcement to value, the total porosity of UHPC was significantly increased. To miti­
enhance the UHPC panel on a larger scale. The results showed that the gate the adverse effect, the replacement ratio of the porous sand should
combination with 2% steel fiber and a single-layer CFRP increased the be limited to the optimum value. When below the value, the introduc­
peak load and energy dissipation by 35% and 81%, compared with the tion of pre-saturated porous sand promotes the hydration reaction, and
UHPC panel without FRP [246]. the hydration products fill the pores that help compensate for the
Fig. 43 illustrates a nacre-inspired cementitious-FRP composite porosity introduced by the sand, as shown in Fig. 44. Similarly, the
structure that was reinforced by polypropylene fibers as well as vinyl- addition of other porous materials in UHPC should be carefully designed
coated polyester FRP [261]. The FRP meshes were placed between the and optimized to maintain or improve the microstructure.
layers of high-strength engineered cementitious composites. This nacre
structure can boost the flexural performance of cementitious-FRP com­ 4.5.2. Control crack width for corrosion resistance
posites by isolating the accumulation of damage within each layer, The superior impermeability of intact UHPC can effectively protect
which inherently distributes damage instead of allowing damage to steel reinforcement inside UHPC. However, cracking is inevitable for
localize at a single weak point to improve the ductility [261–263]. Based concrete structures. Cracks provide short paths that facilitate the
on the nacre-inspired composite design schemes, the modulus of migration of aggressive ions (e.g., chloride ions and sulfate ions) to
rupture, inelastic flexural toughness, and tensile capacity increased by contact the steel. Most often, cracks are accompanied by debonding
65%, 520%, and 345%, compared with that of the engineered cemen­ between steel and UHPC, which can damage the passive film on the steel
titious composite without FRP meshes [261]. surface and further accelerate corrosion [275]. In this sense, the
self-healing of cracks in UHPC is significant for long-term durability.
Self-healing of UHPC is attributed to the low degree of cement hydration
4.5. Improve durability
[276]. Once cracked, unhydrated particles in UHPC are exposed to
moisture and air, producing hydration products from hydraulic and/or
UHPC exhibits superior durability due to its dense matrix. Table 5
pozzolanic reactions and precipitating calcium carbonate and C–S–H, as
lists the technical specifications, parameters, typical values, and test
shown in Fig. 45. It is noteworthy that the crack width is critical for
methods for evaluating the durability of UHPC. The impermeability of
self-healing because wide cracks cannot be filled up, and typically it is
UHPC has been proven by the water adsorption and sorptivity tests,
preferred to control the crack width within 100 μm [277]. The usage of
which justify the high resistance with the intrusion of detrimental

Fig. 42. Flexural behaviors of UHPC with FRP meshes [246]. Reference stands for UHPC without FRP (only 2% steel fiber); U1G stands for 2% steel fiber + single
layer GFRP; U2G stands for 2% steel fiber + dual layer GFRP; and U1C-2% steel fiber + single layer CFRP.

21
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 43. The mechanism of nacre-inspired cementitious composite structure [261].

Table 5
Test methods, specification, and suggested value of UHPC durability.
Test method Specification Parameters Suggested Reference
value

Water ASTM C642 Permeability <0.005 [265]

adsorption
Water ASTM C1585 Sorptivity <0.044 kg/ [266,
sorptivity coefficient m2/h0.5 267]
Alkali- ASTM C1567 Expansive <0.1% [36,91]
aggregate percentage
reaction
Sulfate ASTM C1012 Length change ≈0 [268,
resistance 269]
Fire resistance ASTM E119 Residual fc’ <20% [170,
270]
Rapid chloride ASTM C1202 Charge Passed <100 [271,
permeability coulombs 272]
Freezing- ASTM C666 Durability >95% [36,273]
thawing factor
resistance Mass loss <0.6%
Carbonation ISO Carbonation <3 mm [268, Fig. 45. Schematic illustration of the self-healing phenomenon and the
1920–12:2015 depth 274]
mechanisms in UHPC.
Abrasion ASTM C944 Wearing depth <1 mm [36]
resistance
promising to improve the crack resistance and control the crack width.

4.5.3. Limit particle size of reactive materials for ASR resistance

Reactive aggregates and powders such as waste glass and recycled
concrete aggregate have been used to develop cost-effective UHPC
mixtures. Under a high pH (>13), the siloxane bond (≡ Si–O–Si ≡) of
waste glass is broken, and ≡Si-Oˉ ions are generated, which can react
with alkali ions (K+, Na+) to produce ASR gel [105]. An osmotic pres­
sure gradient between the ASR gel and concrete pore solution in con­
crete drives water and other ions to be absorbed into ASR, which leads to
the expansion of ASR gel [278]. As elucidated in Section 3.1, to mitigate
the cracking potential of internal ASR, the maximum diameter of the
glass powder particles needs to be smaller than the average pore size of
UHPC, so the ASR products by glass powder cannot crack the UHPC
matrix [279].

5. Representative applications of UHPC

Fig. 44. EDS mapping of interfaces between LWS and matrix [128]. Note: the
blue color indicates Si, and the yellow color indicates Ca. In recent years, UHPC has been attracting increasing interest from
the construction industry and it has been applied in numerous structural
synthetic fibers in UHPC is promising to control crack width. With the projects. This section reviews representative structural applications of
introduction of synthetic fibers, the cracking behaviors of UHPC are UHPC.
more prone to multi-micro-cracking patterns rather than localized major
cracking patterns. Thus, the number of cracks is increased, and the width 5.1. Buildings
of each crack is reduced. In addition, any methods that can reduce
autogenous shrinkage and improve the tensile strength of UHPC are The superior mechanical properties of UHPC enable the design and

22
J. Du et al. Composites Part B 224 (2021) 109220

fabrication of slender, light, and aesthetically-appealing building com­ (Fig. 47(a)) in the United States was built in Wapello County, Iowa,
ponents, as shown in Fig. 46. For example, the Museum of European and United States. The 33.5-m-long I-shaped prestressed bridge girders were
Mediterranean Civilizations, which is the first building in the world that fabricated using UHPC. The superior compressive strength of UHPC
extensively uses UHPC, was constructed in Marseille, France in 2013. helped reduce the amount of prestress tendons and the girder depth
The tree-like façade, brackets and decks of the perimeter footbridges, [286]. Based on the success of the project, the shape of UHPC girder was
façade and roof lattices, and the protective covers to the prestressing further optimized, and Pi-shaped girders were designed, as shown in
anchorage points were all made using UHPC [280]. In 2014, Foundation Fig. 47(b). With the optimized design, the Jakway Park Bridge showed
Louis Vuitton pour la Creation was constructed in Paris, France, which excellent performance after it was open to traffic [287]. The successful
features high geometric complexity. The claddings were composed of development of Pi-shaped UHPC girder greatly expanded the applica­
19,000 unique and prefabricated UHPC panels [281]. UHPC was also tion of UHPC in bridges. However, the high upfront cost of proprietary
used to construct the roof of the Jean Bouin Stadium in Paris [282], the UHPC used in the two projects prevented wider acceptance of UHPC in
roof of the Olympic Museum in Lausanne [283], and the cladding for the other states. Therefore, cost-effective locally-available materials have
Qatar National Museum [284]. Besides, a 4-m-high freeform pillar was been utilized to develop non-proprietary cost-effective UHPC. However,
fabricated using UHPC in Provence, France [285]. the feasibility of using non-proprietary UHPC as structural elements of
bridges needs to be further investigated.
In addition to the cast-in-place bridge girders, precast UHPC ele­
5.2. Bridges
ments have been widely used in Accelerated Bridge Construction (ABC)
projects. In 2011, the first precast UHPC waffle bridge deck panels were
Compared with buildings, bridges are generally subjected to higher
installed in the Little Cedar Creek Bridge in Wapello County, Iowa [288],
live loads, so UHPC has been proven a promising material for bridges, as
as shown in Fig. 47(c). After bridge construction, no fatigue damage was
shown in Fig. 47. In 2006, the first UHPC bridge – Mars Hill bridge

Fig. 46. Representative applications of UHPC in buildings: (a) Museum of European and Mediterranean Civilizations [280]; (b) Foundation Louis Vuitton pour la
Creation [281]; (c) Jean Bouin Stadium [282]; (d) Olympic Museum in Lausanne [283]; (e) Qatar National Museum [284]; and (f) a 4-m-high pillar fabricated via 3D
printing [285].

23
J. Du et al. Composites Part B 224 (2021) 109220

Fig. 47. Representative applications of UHPC in bridges: (a) I-shaped girder [286]; (b) Pi-shaped girder [287]; (c) precast waffle bridge deck panels [288]; (d) joint
connection [289]; (e) bridge deck overlay [291]; and (f) pier jacket [293].

found, and the displacements of the bridge deck were considerably would flow freely under gravity due to the slope, and the formworks
smaller than the allowable limits. must be used, which further increases the construction cost. Design a
Another application of UHPC is a filed-cast joint connection for thixotropic UHPC can help mitigating or even eliminating the use of
bridge decks, which is commonly implemented in the ABC projects. The formworks [292]. Further, to improve the seismic behavior, spalling
typical connection requires complicated reinforcement layouts which is resistance, and corrosion resistance of the concrete bridge pier, the
time-consuming. The use of UHPC simplifies the on-site assembling UHPC was also applied as a pier jacket. In 2014, one of the UHPC pier
process for reinforcement and shortens the construction time [289]. In jacket projects was constructed in Canada (The Mission Bridge) [293], as
2009, two field-cast UHPC connection projects (Route 31 Bridge in shown in Fig. 47(f).
Lyons and Route 23 Bridge in Oneonta) were successfully implemented,
as shown in Fig. 47(d) [290].
Most currently, in the U.S., UHPC has been applied for the repair and 5.3. Other applications
rehabilitation of deteriorated bridge decks and columns/piers. In 2017,
the first UHPC bridge deck overlay was constructed in Buchanan In addition to buildings and bridges, it is promising to apply UHPC to
County, Iowa States (Mud Creek Bridge) [291], as shown in Fig. 47(e). other structures, such as tunnels, wind turbine towers, and nuclear
The field test showed satisfactory bonding properties between the UHPC power plants. For tunnel application, UHPC can develop more efficient
overlay and substrates, as well as commendable crack resistance of the tunnel systems with larger useable spaces by reducing the thickness of
overlay [292]. However, given the slope of the bridge deck, when tunnel elements [294]. For wind turbine towers, UHPC components
applied as an overlay, the UHPC mixtures should be designed with allow for the creation of taller and more slender wind turbine towers,
thixotropic property for the slope casting; otherwise, the UHPC mixture thus increasing the efficiency of energy generation [295]. For nuclear
power plants, UHPC possesses better radiation shielding property and

24
J. Du et al. Composites Part B 224 (2021) 109220

higher blast tolerance which can better guarantee the security of critical (6) Three potential durability issues and corresponding solutions are
infrastructure [296]. proposed. The content of porous materials needs to be optimized
to prevent affecting the porosity of UHPC. The crack width needs
6. Conclusions and further research to be controlled within 0.1 mm to allow sufficient self-healing.
The particle size of reactive aggregates needs to be restricted to
6.1. Conclusions prevent ASR cracking.

A critical review is carried out on new development of UHPC mix­ 6.2. Future research trend
tures. The mixture design methodologies and typical ingredients of
UHPC are comprehensively introduced. Besides, as an effort to resolve Based on the above review, some future research opportunities have
some challenges associated with the UHPC, some state-of-art solutions been identified as follows:
are summarized. Finally, some representative applications are intro­
duced. Based on that, the following conclusions can be drawn: (1) The unit weight of UHPC is high due to the use of steel compo­
nents. Future studies should be conducted to further reduce the
(1) The main mixture design methodologies for UHPC include the unit weight of UHPC for lightweight structures and repair and
dry particle packing method, wet particle packing method, and rehabilitation applications.
performance-based method. The principle of the dry particle (2) Some specific nanomaterials can provide UHPC with function­
packing method and wet particle packing method is to maximize ality, such as self-sensing property, self-cleaning property, and
the initial packing density of UHPC. The performance-based electromagnetic interference shielding property. More research
method is to directly link the design variables with the target can be done in the area.
properties of UHPC. (3) SCMs such as slag, fly ash, and silica fume are important com­
(2) The typical ingredients of UHPC include: (1) binder materials: ponents in UHPC mixture design. However, due to the switching
SCMs and fillers (2) aggregates: fine aggregates and coarse ag­ from coal to natural gas power plants, the typical SCMs are
gregates; (3) chemical admixtures; (4) reinforced fibers. The scarcer. As a result, research on alternative SCMs is highly in
unique properties, recommended contents, effects on UHPC demand.
properties, and the working principles of the ingredients have (4) UHPC usually contains high volumes of binders which can be
been comprehensively reviewed. used to sequester recycled or atmospheric CO2 while improving
(3) The workability of UHPC should be well controlled for con­ the mechanical properties and durability of the UHPC. The
structability. The physical control and chemical control are technologies of CO2 sequestration by the UHPC are of interest to
elucidated to effectively improve the flowability of UHPC to be developed.
guarantee the self-consolidating property. The adjustment of
viscosity modified agent content is proposed to control the Declaration of competing interest
rheology of UHPC suspending mortar to prevent fiber segregation
and floating of lightweight raw materials. The authors declare that they have no known competing financial
(4) Four effective methods to reduce the autogenous shrinkage of interests or personal relationships that could have appeared to influence
UHPC are discussed, including control degree of cement hydra­ the work reported in this paper.
tion, reduction rate of internal relative humidity, restraint degree
for volume change, and introduction of internal forces to Acknowledgements
compensate shrinkage-induced tension.
(5) Two methods to improve the flexural/tensile properties are This paper is supported by National Science Foundation under
summarized, including fiber dispersion and fiber orientation award: CMMI 2046407 and New Jersey Department of Transportation:
improvement and incorporation of multiscale/physics Task Order 349 – Bridge Resource Program (2017-2020), contract ID
reinforcements number: 17-60139, federal project number: D00S237. The authors
thank Mr. Leo Boyer for editing the language of the manuscript.

Appendix
Table A
Commercial UHPC mixture designs

Designation formulation (kg/m3) Mini slump (mm) 28-day fc (MPa)

′
Reference

Code Premix Cement SF QP Sand AC HRWR Water Steel fiber

Ductal® – 712 231 211 1020 30 31 109 156 180 ± 20 149 [22,23]
BSI® – 1114 169 – 1072 – 40 209 234 640* 175 [24]
CEMTEC® – 1050 268 – 514 – 44 180 470 – 205☆ [25]
CRC® 930 – – – 1300 – – 150 225 – 150 [26]
BCV® 2136 – – – – – 21.5 – 156 750* 140 [27]
Cor-Tuf® – 758 497 295 733 – 13 158 140 – >200⊙ [28]

Note: Premix stands for the combination of all cementitious materials (proportion is unknown), QP stands for quartz powder; AC stands for accelerator; fc stands for
′

compressive strength, * stands for slump flow results, ☆ stands for heat curing, ⊙ stands for the special curing method (moisture curing at 22 ◦ C and 100% for 8 days,
then cured in water bath at 85 ◦ C for 4 days, finally dried in an oven for 2 days at 85 ◦ C for a cumulative age of 13 days).

25
J. Du et al. Composites Part B 224 (2021) 109220

Table B1
UHPC mixture designs (using high-volume SCMs)

Designation formulation (kg/m3) Mini slump (mm) 28-day fc (MPa)

′
Reference

Code Cement SF QP FA Slag GP RHA NP Sand HRWR Water Steel fiber

G50SF5 548 42 – – 535 – – – 998 16.0 167 156 280 125 [31]
G50 593 – – – 546 – – – 993 12.5 182 156 285 124
FAC40SF5 663 42 – 367 – – – – 1011 12.0 171 156 285 124
FAC60 486 – – 556 – – – – 1019 5.5 188 156 285 120
80C20 GP 623 216 – – – 390 – – 935 13.0 188 158 217 170 [32]
U1-RHA 781 – 208 – – – 155 – 1029 23.4 202 78 280 170 [33]
U2-RHA 579 – 208 – 183 – 155 – 1029 22.9 203 78 315 185
M2 900 110 – – – – 110 1036 40.0 163 157 184 138 [297]
M5 630 220 – – – – 270 994 40.0 163 157 184 131
Note: SF stands for silica fume, QP stands for quartz powder; FA stand for fly ash, GP stands for glass powder, RHA stands for rice husk ash, NP stands for natural
pozzolan, HRWR = high-range water reducer. All UHPC mixtures are under standard curing, fc stands for compressive strength.
′

Table B2
UHPC mixture designs (use high-volume fillers)

Designation formulation (kg/m3) Mini slump (mm) 28-day fc (MPa)

′
Reference

Code Cement SF LSP QP NS NC Sand HRWR Water Steel fiber

LP25C 595 198 198 278 – – 873 7.1 191 162 – 160 [35]
LP50C 397 198 397 278 – – 873 4.7 191 162 – 130
CSFLS34 826 192 486 – – – 444 10.5 201 – 360 148 [34]
NC3.2-2 829 216 – – – 34.5 923 NG 177 156 220–250 115 [113]
NS1.0-2 852 216 – – 10.8 – 923 NG 177 156 220–250 114
NS1 900 135 – – 10.4 – 1125 22.5 186 78 – 132 [235]

Note: SF stands for silica fume, LSP stands for limestone powder, QP stands for quartz powder, NS stands for nano-silica, NC stands for nano-CaCO3, fc stands for
′

compressive strength.

Table B3
UHPC mixture designs (low powder content by incorporating coarse aggregates)

Designation formulation (kg/m3) Mini slump (mm) 28-day fc (MPa)

′
Reference

Code Cement SF LSP Sand Coarse aggregates HRWR Water Steel fiber

UHPC1 675 45 180 864 576 (Max size = 3 mm) 10.8 180 156 – 142 [38]
UHPC2 675 45 180 588 895 (Max size = 8 mm) 10.8 180 156 – 139
UHPC3 675 45 180 304 1189 (Max size = 16 mm) 10.8 180 156 – 136
1–1 450 50 – 797 1195 (Max size = 19 mm) 18.0 90 – 200* 131 [39]
1–2 630 70 – 715 1073 (Max size = 19 mm) 18.0 126 – 270* 135
1–3 810 90 – 616 923 (Max size = 19 mm) 18.0 162 – 550* 138

Note: SF stands for silica fume, LSP stands for limestone powder, HRWR stands for high-range water reducer, fc stands for compressive strength.
′

Table B4
UHPC mixture designs (hybrid fibers)

Designation formulation 28-day fc (MPa)

′
28-day ff (MPa)
′
28-day ft (MPa)
′
Toughness Reference
T150 (J)
Code SF BF PVA GF WF CF

MGF 1.5 – – – 1% – – 152 11.2 – [166]

S1.0-B0.5 1% 0.5% – – – – 128 – 14.7 – [42]
S1.0-PVA0.5 1% – 0.5% – – – 143 – 11.8 – [42]
PVA0.5S1.5 1.5% – 0.5% – – – 160 23.4 – 58.5 [41]
Basalt 1 – 1% – – – – 139 – – – [169]
Wollastonite – – – – 1% – 145 – – – [169]
UHPC-J10 – – – – – 10* 118 13.8 – – [298]
Note: SF stands for steel fiber, BF stands for basalt fiber, PVA stands for polyvinyl alcohol fiber, GF stands for glass fiber, WF stands for wollastonite fiber, CF stands for
cellulose fiber, fc stands for compressive strength, ff stands for flexural strength, * stands for kg/m3, ft stands for tensile strength.
′ ′ ′

Table C
Proportioning of the Reference UHPC mixture

Designation formulation (kg/m3) Mini slump 28-day fc

′
28-day ff
′
Toughness Autogenous Reference
(mm) (MPa) (MPa) T150 (J) shrinkage (μm/m)
Code Cement SF QS FS HRWR Water Steel
fiber

Reference 712 231 1020 211 6.5 164 156 275 135 19.7 40.4 730 [31]
Note: Reference UHPC mixture is under standard curing.

26
J. Du et al. Composites Part B 224 (2021) 109220

References [33] Rößler C, Bui D-D, Ludwig H-M. Rice husk ash as both pozzolanic admixture and
internal curing agent in ultra-high performance concrete. Cement Concr Compos
2014;53:270–8.
[1] De la Varga I, Graybeal BA. Dimensional stability of grout-type materials used as
[34] Huang W, Kazemi-Kamyab H, Sun W, Scrivener K. Effect of cement substitution
connections between prefabricated concrete elements. J Mater Civ Eng 2015;27
by limestone on the hydration and microstructural development of ultra-high
(9). 04014246.
performance concrete (UHPC). Cement Concr Compos 2017;77:86–101.
[2] Wille K, Naaman AE, Parra-Montesinos GJ. Ultra-high performance concrete with
[35] Kang S-H, Jeong Y, Tan KH, Moon J. High-volume use of limestone in ultra-high
compressive strength exceeding 150 MPa (22 ksi): a simpler way. ACI Mater J
performance fiber-reinforced concrete for reducing cement content and
2011;108(1).
autogenous shrinkage. Construct Build Mater 2019;213:292–305.
[3] Habel K, Viviani M, Denarié E, Brühwiler E. Development of the mechanical
[36] Berry M, Snidarich R, Wood C. Development of non-proprietary ultra-high
properties of an ultra-high performance fiber reinforced concrete (UHPFRC).
performance concrete. Montana: Dept. of Transportation. Research Programs;
Cement Concr Res 2006;36(7):1362–70.
2017.
[4] Meng Q, Wu C, Li J, Liu Z, Wu P, Yang Y, Wang Z. Steel/basalt rebar reinforced
[37] Subedi D. Non-proprietary UHPC for anchorage of large diameter column bars in
Ultra-High Performance Concrete components against methane-air explosion
grouted ducts. Doctoral dissertation. 2019.
loads. Compos B Eng 2020;198:108215.
[38] Li P, Yu Q, Brouwers H. Effect of coarse basalt aggregates on the properties of
[5] Li P, Sluijsmans MJ, Brouwers H, Yu Q. Functionally graded ultra-high
Ultra-high Performance Concrete (UHPC). Construct Build Mater 2018;170:
performance cementitious composite with enhanced impact properties. Compos B
649–59.
Eng 2020;183:107680.
[39] Wang C, Yang C, Liu F, Wan C, Pu X. Preparation of ultra-high performance
[6] Perry V. What really is ultra-high performance concrete – towards a global
concrete with common technology and materials. Cement Concr Compos 2012;34
definition. In: Proceedings of the 2nd international conference on UHPC materials
(4):538–44.
and structures. Fuzhou: RILEM Publications; 2018.
[40] Wu Z, Shi C, Khayat KH. Investigation of mechanical properties and shrinkage of
[7] Russell HG, Graybeal BA, Russell HG. Ultra-high performance concrete: a state-of-
ultra-high performance concrete: influence of steel fiber content and shape.
the-art report for the bridge community. United States: Federal Highway
Compos B Eng 2019:107021.
Administration. Office of Infrastructure Research and Development; 2013.
[41] Meng W, Khayat KH. Effect of hybrid fibers on fresh properties, mechanical
[8] ACI 239R-18 ultra-high-performance concrete: an emerging technology report.
properties, and autogenous shrinkage of cost-effective UHPC. J Mater Civ Eng
American Concrete Institute; 2018.
2018;30(4):04018030.
[9] Graybeal BA. Flexural behavior of an ultrahigh-performance concrete I-girder.
[42] Kang ST, Choi JI, Koh KT, Lee KS, Lee BY. Hybrid effects of steel fiber and
J Bridge Eng 2008;13(6):602–10.
microfiber on the tensile behavior of ultra-high performance concrete. Compos
[10] Meng W, Khayat KH. Improving flexural performance of ultra-high-performance
Struct 2016;145:37–42.
concrete by rheology control of suspending mortar. Compos B Eng 2017;117:
[43] Zhang H, Ji T, He B, He L. Performance of ultra-high performance concrete
26–34.
(UHPC) with cement partially replaced by ground granite powder (GGP) under
[11] Vande Voort TL, Suleiman MT, Sritharan S. Design and performance verification
different curing conditions. Construct Build Mater 2019;213:469–82.
of UHPC piles for deep foundations. Iowa State University. Center for
[44] Zhang H, Ji T, Zeng X, Yang Z, Lin X, Liang Y. Mechanical behavior of ultra-high
Transportation Research and Education; 2008.
performance concrete (UHPC) using recycled fine aggregate cured under different
[12] Liu, Z. and Meng, W., Fundamental understanding of carbonation curing and
conditions and the mechanism based on integrated microstructural parameters.
durability of carbonation-cured cement-based composites: a review. J. CO2 Util.,
Construct Build Mater 2018;192:489–507.
44, pp: 101428.
[45] Schachinger I, Hilbig H, Stengel T, Fehling E. Effect of curing temperature at an
[13] Li Y, Mi T, Liu W, Dong Z, Dong B, Tang L, Xing F. Chemical and mineralogical
early age on the long-term strength development of UHPC. In: 2nd international
characteristics of carbonated and uncarbonated cement pastes subjected to high
symposium on ultra high performance concrete. Kassel University Press; 2008.
temperatures. Compos B Eng 2021;216:108861.
[46] Yu R, Spiesz P, Brouwers H. Development of an eco-friendly Ultra-High
[14] Azmee NM, Shafiq N. Ultra-high performance concrete: from fundamental to
Performance Concrete (UHPC) with efficient cement and mineral admixtures
applications. Case Stud. Construct. Mater. 2018;9.
uses. Cement Concr Compos 2015;55:383–94.
[15] Birchall J, Howard A, Kendall K. Flexural strength and porosity of cements.
[47] El-Tawil S, Tai Y-S, Meng B, Hansen W, Liu Z. Commercial production of non-
Nature 1981;289(5796):388–90.
proprietary ultra high performance concrete. Michigan Department of
[16] Bache HH. Densified cement ultra-fine particle-based materials. 1981.
Transportation Research Administration; 2018.
[17] Kar DRL. Properties, applications: slurry infiltrated fiber concrete (SIFCON).
[48] Ramme BW, Tharaniyil MP. Coal combustion products utilization handbook.
Concr Int 1984;6(12):44–7.
2004.
[18] Bache HH. Compact reinforced composite basic principles. 1987.
[49] Qiao P, Zhou Z, Allena S. Developing connections for longitudinal joints between
[19] Lankard DR. Slurry infiltrated fiber concrete (SIFCON): properties and
deck bulb tees-development of UHPC mixes with local materials. Washington
applications. MRS Online Proc Libr 1984;42.
(State): Department of Transportation; 2016.
[20] Richard P, Cheyrezy M. Composition of reactive powder concretes. Cement Concr
[50] Berry M, Scherr R, Matteson K. Feasibility of non-proprietary ultra-high
Res 1995;25(7):1501–11.
performance concrete (UHPC) for use in highway bridges in Montana: phase II
[21] De Larrard F, Sedran T. Optimization of ultra-high-performance concrete by the
field application. Montana Department of Transportation; 2020.
use of a packing model. Cement Concr Res 1994;24(6):997–1009.
[51] Mendonca F, El-Khier MA, Morcous G, Hu J. Feasibility study of development of
[22] Graybeal BA. Compressive behavior of ultra-high-performance fiber-reinforced
ultra-high performance concrete (UHPC) for highway bridge applications in
concrete. ACI Mater J 2007;104(2):146.
Nebraska. Nebraska Department of Transportation; 2020.
[23] Graybeal BA. Material property characterization of ultra-high performance
[52] Yoo D-Y, Park J-J, Kim S-W, Yoon Y-S. Influence of reinforcing bar type on
concrete. United States: Federal Highway Administration. Office of Infrastructure
autogenous shrinkage stress and bond behavior of ultra high performance fiber
Research and Development; 2006.
reinforced concrete. Cement Concr Compos 2014;48:150–61.
[24] Hajar Z, Lecointre D, Simon A, Petitjean J. Design and construction of the world
[53] Valipour M, Khayat KH. Coupled effect of shrinkage-mitigating admixtures and
first ultra-high performance concrete road bridges. In: Proceedings of the Int.
saturated lightweight sand on shrinkage of UHPC for overlay applications.
Symp. On UHPC; 2004. Kassel, Germany.
Construct Build Mater 2018;184:320–9.
[25] Rossi P, Arca A, Parant E, Fakhri P. Bending and compressive behaviours of a new
[54] Cusson D, Hoogeveen T. An experimental approach for the analysis of early-age
cement composite. Cement Concr Res 2005;35(1):27–33.
behaviour of high-performance concrete structures under restrained shrinkage.
[26] Aarup B. CRC–Structural applications of ultra high performance fiber reinforced
Cement Concr Res 2007;37(2):200–9.
concrete. In: Proceedings of the Second international symposium on ultra high
[55] Teng L, Valipour M, Khayat KH. Design and performance of low shrinkage UHPC
performance concrete; March. 2008.
for thin bonded bridge deck overlay. Cement Concr Compos 2021:103953.
[27] Camacho E, López J, Ros PS. Definition of three levels of performance for
[56] Zhu L, Wang J-J, Li X, Zhao G-Y, Huo X-J. Experimental and numerical study on
UHPFRC-VHPFRC with available materials. In: Ultra-high performance concrete
creep and shrinkage effects of ultra high-performance concrete beam. Compos B
and nanotechnology in construction. Proceedings of hipermat 2012. 3rd
Eng 2020;184:107713.
international symposium on UHPC and nanotechnology for high performance
[57] Yoo D-Y, Kim S, Park G-J, Park J-J, Kim S-W. Effects of fiber shape, aspect ratio,
construction materials. Kassel University Press; 2012.
and volume fraction on flexural behavior of ultra-high-performance fiber-
[28] Williams EM, Graham SS, Reed PA, Rushing TS. Laboratory characterization of
reinforced cement composites. Compos Struct 2017;174:375–88.
Cor-Tuf concrete with and without steel fibers. Enginner research and
[58] Fan L, Meng W, Teng L, Khayat KH. Effects of lightweight sand and steel fiber
development center; 2009.
contents on the corrosion performance of steel rebar embedded in UHPC.
[29] Ahmad S. Use of alternative waste materials in producing ultra-high performance
Construct Build Mater 2020;238:117709.
concrete. In: MATEC web of conferences. EDP Sciences; 2017.
[59] Yu R, Spiesz P, Brouwers H. Mix design and properties assessment of ultra-high
[30] Chen Y, Matalkah F, Soroushian P, Weerasiri R, Balachandra A. Optimization of
performance fibre reinforced concrete (UHPFRC). Cement Concr Res 2014;56:
ultra-high performance concrete, quantification of characteristic features. Cogent
29–39.
Eng. 2019;6(1):1558696.
[60] Li L, Kwan A. Packing density of concrete mix under dry and wet conditions.
[31] Meng W, Valipour M, Khayat KH. Optimization and performance of cost-effective
Powder Technol 2014;253:514–21.
ultra-high performance concrete. Mater Struct 2017;50(1):29.
[61] Wong HH, Kwan AK. Packing density of cementitious materials: part
[32] Soliman N, Tagnit-Hamou A. Development of ultra-high-performance concrete
1—measurement using a wet packing method. Mater Struct 2008;41(4):689–701.
using glass powder–Towards ecofriendly concrete. Construct Build Mater 2016;
[62] Stovall T, De Larrard F, Buil M. Linear packing density model of grain mixtures.
125:600–12.
Powder Technol 1986;48(1):1–12.

27
J. Du et al. Composites Part B 224 (2021) 109220

[63] De Larrard F, Sedran T. Mixture-proportioning of high-performance concrete. [97] Wei F, Grutzeck MW, Roy DM. The retarding effects of fly ash upon the hydration
Cement Concr Res 2002;32(11):1699–704. of cement pastes: the first 24 hours. Cement Concr Res 1985;15(1):174–84.
[64] Fennis S, Walraven J, Den Uijl J. Compaction-interaction packing model: [98] Moradian M, Hu Q, Aboustait M, Robertson B, Ley MT, Hanan JC, Xiao X. Direct
regarding the effect of fillers in concrete mixture design. Mater Struct 2013;46(3): in-situ observation of early age void evolution in sustainable cement paste
463–78. containing fly ash or limestone. Compos B Eng 2019;175:107099.
[65] Fuller WB, Thompson SE. The laws of proportioning concrete. 1907. [99] Edwin RS, De Schepper M, Gruyaert E, De Belie N. Effect of secondary copper slag
[66] Andreasen A. Über die Beziehung zwischen Kornabstufung und Zwischenraum in as cementitious material in ultra-high performance mortar. Construct Build Mater
Produkten aus losen Körnern (mit einigen Experimenten). Kolloid Z 1930;50(3): 2016;119:31–44.
217–28. [100] Chen P, Ma B, Tan H, Liu X, Zhang T, Li C, Yang Q, Luo Z. Utilization of barium
[67] Funk JE, Dinger DR. Predictive process control of crowded particulate slag to improve chloride-binding ability of cement-based material. J Clean Prod
suspensions: applied to ceramic manufacturing. Springer Science & Business 2021;283:124612.
Media; 2013. [101] Edwin RS, Gruyaert E, De Belie N. Influence of intensive vacuum mixing and heat
[68] De Larrard F. Concrete mixture proportioning: a scientific approach. CRC Press; treatment on compressive strength and microstructure of reactive powder
1999. concrete incorporating secondary copper slag as supplementary cementitious
[69] Brouwers H, Radix H. Self-compacting concrete: theoretical and experimental material. Construct Build Mater 2017;155:400–12.
study. Cement Concr Res 2005;35(11):2116–36. [102] Taha B, Nounu G. Utilizing waste recycled glass as sand/cement replacement in
[70] Hunger M. An integral design concept for ecological self-compacting concrete. concrete. J Mater Civ Eng 2009;21(12):709–21.
2010. [103] Vaitkevičius V, Šerelis E, Hilbig H. The effect of glass powder on the
[71] Mehdipour I, Khayat KH. Understanding the role of particle packing microstructure of ultra high performance concrete. Construct Build Mater 2014;
characteristics in rheo-physical properties of cementitious suspensions: a 68:102–9.
literature review. Construct Build Mater 2018;161:340–53. [104] Pacheco-Torgal F, Labrincha J, Leonelli C, Palomo A, Chindaprasit P. Handbook
[72] Wang X, Yu R, Song Q, Shui Z, Liu Z, Wu S, Hou D. Optimized design of ultra-high of alkali-activated cements, mortars and concretes. Elsevier; 2014.
performance concrete (UHPC) with a high wet packing density. Cement Concr [105] Guo P, Bao Y, Meng W. Review of using glass in high-performance fiber-
Res 2019;126:105921. reinforced cementitious composites. Cement Concr Compos 2021:104032.
[73] Iveson SM, Litster JD, Hapgood K, Ennis BJ. Nucleation, growth and breakage [106] Korpa A, Kowald T, Trettin R. Phase development in normal and ultra high
phenomena in agitated wet granulation processes: a review. Powder Technol performance cementitious systems by quantitative X-ray analysis and
2001;117(1–2):3–39. thermoanalytical methods. Cement Concr Res 2009;39(2):69–76.
[74] Iveson SM, Wauters PA, Forrest S, Litster JD, Meesters GM, Scarlett B. Growth [107] Fehling E, Schmidt M, Walraven J, Leutbecher T, Frönlich S. Ultra-high
regime map for liquid-bound granules: further development and experimental performance concrete UHPC: fundamentals. Design, examples. Beton-Kalender;
validation. Powder Technol 2001;117(1–2):83–97. 2013.
[75] Cai W. Effect of particle packing on flow property and strength of concrete [108] Poppe A-M, De Schutter G. Cement hydration in the presence of high filler
mortar. 2017. contents. Cement Concr Res 2005;35(12):2290–9.
[76] Li L, Cai Z, Yu K, Zhang YX, Ding Y. Performance-based design of all-grade strain [109] Zanni H, Cheyrezy M, Maret V, Philippot S, Nieto P. Investigation of hydration
hardening cementitious composites with compressive strengths from 40 MPa to and pozzolanic reaction in reactive powder concrete (RPC) using 29Si NMR.
120 MPa. Cement Concr Compos 2019;97:202–17. Cement Concr Res 1996;26(1):93–100.
[77] ASTM C150/C150M-20. Standard specification for portland cement. ASTM [110] De Rosa C. International Programme on Chemical safety (IPCS), concise
International; 2020. international chemical assessment document, 61 (CICAD), hydrogen cyanide and
[78] Taylor HF. Cement chemistry, vol. 2. Thomas Telford London; 1997. cyanides: human health aspects. Indian J Med Res 2005;122(2):180.
[79] Van, V. and Ludwig, H. Proportioning optimization of UHPC containing rice husk [111] Li C. Mechanical and transport properties of recycled aggregate concrete modified
ash and ground granulated blast-furnace slag. in Proceedings of the 3rd with limestone powder. Compos B Eng 2020;197:108189.
international symposium on UHPC and nanotechnology for high performance [112] Burroughs JF, Shannon J, Rushing TS, Yi K, Gutierrez QB, Harrelson DW.
construction materials, Kassel University Press. Potential of finely ground limestone powder to benefit ultra-high performance
[80] Schröfl C, Gruber M, Plank J. Preferential adsorption of polycarboxylate concrete mixtures. Construct Build Mater 2017;141:335–42.
superplasticizers on cement and silica fume in ultra-high performance concrete [113] Wu Z, Shi C, Khayat KH, Xie L. Effect of SCM and nano-particles on static and
(UHPC). Cement Concr Res 2012;42(11):1401–8. dynamic mechanical properties of UHPC. Construct Build Mater 2018;182:
[81] Liao W, Sun X, Kumar A, Sun H, Ma H. Hydration of binary portland cement 118–25.
blends containing silica fume: a decoupling method to estimate degrees of [114] Shi C, Wu Z, Xiao J, Wang D, Huang Z, Fang Z. A review on ultra high
hydration and pozzolanic reaction. Front. Mater. 2019;6:78. performance concrete: Part I. Raw materials and mixture design. Construct Build
[82] Yu R, Spiesz P, Brouwers H. Effect of nano-silica on the hydration and Mater 2015;101:741–51.
microstructure development of Ultra-High Performance Concrete (UHPC) with a [115] Su Y, Li J, Wu C, Wu P, Li Z-X. Influences of nano-particles on dynamic strength of
low binder amount. Construct Build Mater 2014;65:140–50. ultra-high performance concrete. Compos B Eng 2016;91:595–609.
[83] Wu Z, Shi C, Khayat K. Influence of silica fume content on microstructure [116] Teng L, Zhu J, Khayat KH, Liu J. Effect of welan gum and nanoclay on thixotropy
development and bond to steel fiber in ultra-high strength cement-based of UHPC. Cement Concr Res 2020;138:106238.
materials (UHSC). Cement Concr Compos 2016;71:97–109. [117] Qian Y. Characterization of structural rebuilding and shear migration in
[84] Qasim OA. Experimental investigation on autogenous shrinkage of high and ultra- cementitious materials in consideration of thixotropy. Doctoral dissertation.
high strength concrete. IOP Conf Ser Mater Sci Eng 2018;454:012067. 2017.
[85] Van V-T-A, Rößler C, Bui D-D, Ludwig H-M. Mesoporous structure and pozzolanic [118] Gosens I, Post JA, de la Fonteyne LJ, Jansen EH, Geus JW, Cassee FR, de
reactivity of rice husk ash in cementitious system. Construct Build Mater 2013;43: Jong WH. Impact of agglomeration state of nano-and submicron sized gold
208–16. particles on pulmonary inflammation. Part Fibre Toxicol 2010;7(1):1–11.
[86] Huang H, Gao X, Wang H, Ye H. Influence of rice husk ash on strength and [119] Wu Z, Khayat KH, Shi C, Tutikian BF, Chen Q. Mechanisms underlying the
permeability of ultra-high performance concrete. Construct Build Mater 2017; strength enhancement of UHPC modified with nano-SiO2 and nano-CaCO3.
149:621–8. Cement Concr Compos 2021;119:103992.
[87] Alsalman A, Dang CN, Martí-Vargas JR, Micah Hale W. Mixture-proportioning of [120] Konsta-Gdoutos MS, Metaxa ZS, Shah SP. Highly dispersed carbon nanotube
economical UHPC mixtures. J. Build. Eng. 2020;27:100970. reinforced cement based materials. Cement Concr Res 2010;40(7):1052–9.
[88] Staquet S, Espion B. Early age autogenous shrinkage of UHPC incorporating very [121] Peyvandi A, Soroushian P, Abdol N, Balachandra AM. Surface-modified graphite
fine fly ash or metakaolin in replacement of silica fume. In: International nanomaterials for improved reinforcement efficiency in cementitious paste.
symposium on ultra high performance concrete. Kessel Germany; 2004. Carbon 2013;63:175–86.
[89] Liu J, Ou Z, Mo J, Wang Y, Wu H. The effect of SCMs and SAP on the autogenous [122] Ma S, Qian Y, Kawashima S. Experimental and modeling study on the non-linear
shrinkage and hydration process of RPC. Construct Build Mater 2017;155: structural build-up of fresh cement pastes incorporating viscosity modifying
239–49. admixtures. Cement Concr Res 2018;108:1–9.
[90] Randl N, Steiner T, Ofner S, Baumgartner E, Mészöly T. Development of UHPC [123] Hamed N, El-Feky M, Kohail M, Nasr E-SA. Effect of nano-clay de-agglomeration
mixtures from an ecological point of view. Construct Build Mater 2014;67:373–8. on mechanical properties of concrete. Construct Build Mater 2019;205:245–56.
[91] Soliman NA, Tagnit-Hamou A. Using glass sand as an alternative for quartz sand [124] Kang S-H, Jeong Y, Tan KH, Moon J. The use of limestone to replace physical filler
in UHPC. Construct Build Mater 2017;145:243–52. of quartz powder in UHPFRC. Cement Concr Compos 2018;94:238–47.
[92] Soliman N, Tagnit-Hamou A. Partial substitution of silica fume with fine glass [125] Tasong WA, Lynsdale CJ, Cripps JC. Aggregate-cement paste interface: Part I.
powder in UHPC: filling the micro gap. Construct Build Mater 2017;139:374–83. Influence of aggregate geochemistry. Cement Concr Res 1999;29(7):1019–25.
[93] Van Tuan N, Ye G, Van Breugel K, Copuroglu O. Hydration and microstructure of [126] Liu K, Yu R, Shui Z, Li X, Ling X, He W, Yi S, Wu S. Effects of pumice-based porous
ultra high performance concrete incorporating rice husk ash. Cement Concr Res material on hydration characteristics and persistent shrinkage of ultra-high
2011;41(11):1104–11. performance concrete (UHPC). Materials 2019;12(1):11.
[94] ASTM C618-19. Standard specification for coal fly ash and raw or calcined [127] Liu K, Yu R, Shui Z, Li X, Guo C, Yu B, Wu S. Optimization of autogenous
natural pozzolan for use in concrete. ASTM International; 2019. shrinkage and microstructure for Ultra-High Performance Concrete (UHPC) based
[95] Yu J, Li G, Leung CKY. Hydration and physical characteristics of ultrahigh- on appropriate application of porous pumice. Construct Build Mater 2019;214:
volume fly ash-cement systems with low water/binder ratio. Construct Build 369–81.
Mater 2018;161:509–18. [128] Meng W, Khayat KH. Effects of saturated lightweight sand content on key
[96] Zhao S, Sun W. Effect of silica fume and fly ash on pore structures of blended characteristics of ultra-high-performance concrete. Cement Concr Res 2017;101:
pastes at low water to binder ratios. Adv Cement Res 2015;27(9):506–14. 46–54.

28
J. Du et al. Composites Part B 224 (2021) 109220

[129] Yang S, Millard S, Soutsos M, Barnett S, Le TT. Influence of aggregate and curing [160] Zhang Y, Zhu Y, Qu S, Kumar A, Shao X. Improvement of flexural and tensile
regime on the mechanical properties of ultra-high performance fibre reinforced strength of layered-casting UHPC with aligned steel fibers. Construct Build Mater
concrete (UHPFRC). Construct Build Mater 2009;23(6):2291–8. 2020;251:118893.
[130] Bentz DP, Ferraris CF, Jones SZ, Lootens D, Zunino F. Limestone and silica powder [161] Hung C-C, Lee H-S, Chan SN. Tension-stiffening effect in steel-reinforced UHPC
replacements for cement: early-age performance. Cement Concr Compos 2017;78: composites: constitutive model and effects of steel fibers, loading patterns, and
43–56. rebar sizes. Compos B Eng 2019;158:269–78.
[131] Fanghui H, Qiang W, Mutian L, Yingjun M. Early hydration properties of [162] Xue G, Yilmaz E, Song W, Cao S. Mechanical, flexural and microstructural
composite binder containing limestone powder with different finenesses. J Therm properties of cement-tailings matrix composites: effects of fiber type and dosage.
Anal Calorim 2016;123(2):1141–51. Compos B Eng 2019;172:131–42.
[132] Scrivener K, Martirena F, Bishnoi S, Maity S. Calcined clay limestone cements [163] Wu Z, Shi C, He W, Wu L. Effects of steel fiber content and shape on mechanical
(LC3). Cement Concr Res 2018;114:49–56. properties of ultra high performance concrete. Construct Build Mater 2016;103:
[133] Yang R, Yu R, Shui Z, Gao X, Xiao X, Fan D, Chen Z, Cai J, Li X, He Y. Feasibility 8–14.
analysis of treating recycled rock dust as an environmentally friendly alternative [164] Wu Z, Shi C, Khayat KH. Investigation of mechanical properties and shrinkage of
material in Ultra-High Performance Concrete (UHPC). J Clean Prod 2020;258: ultra-high performance concrete: influence of steel fiber content and shape.
120673. Compos B Eng 2019;174:107021.
[134] Liao W, Kumar A, Khayat K, Ma H. Multifunctional lightweight aggregate [165] You I, Yoo D-Y, Kim S, Kim M-J, Zi G. Electrical and self-sensing properties of
containing phase change material and water for damage mitigation of concrete. ultra-high-performance fiber-reinforced concrete with carbon nanotubes. Sensors
2019. 2017;17(11):2481.
[135] Bentur A, Igarashi S-i, Kovler K. Prevention of autogenous shrinkage in high- [166] Gesoglu M, Güneyisi E, Muhyaddin GF, Asaad DS. Strain hardening ultra-high
strength concrete by internal curing using wet lightweight aggregates. Cement performance fiber reinforced cementitious composites: effect of fiber type and
Concr Res 2001;31(11):1587–91. concentration. Compos B Eng 2016;103:74–83.
[136] Hwang S-D, Khayat KH, Youssef D. Effect of moist curing and use of lightweight [167] Yan L, Kasal B, Huang L. A review of recent research on the use of cellulosic fibres,
sand on characteristics of high-performance concrete. Mater Struct 2013;46(1): their fibre fabric reinforced cementitious, geo-polymer and polymer composites in
35–46. civil engineering. Compos B Eng 2016;92:94–132.
[137] Liu J, Han F, Cui G, Zhang Q, Lv J, Zhang L, Yang Z. Combined effect of coarse [168] Ma R, Guo L, Ye S, Sun W, Liu J. Influence of hybrid fiber reinforcement on
aggregate and fiber on tensile behavior of ultra-high performance concrete. mechanical properties and autogenous shrinkage of an ecological UHPFRCC.
Construct Build Mater 2016;121:310–8. J Mater Civ Eng 2019;31(5):04019032.
[138] Liao K-Y, Chang P-K, Peng Y-N, Yang C-C. A study on characteristics of interfacial [169] Hannawi K, Bian H, Prince-Agbodjan W, Raghavan B. Effect of different types of
transition zone in concrete. Cement Concr Res 2004;34(6):977–89. fibers on the microstructure and the mechanical behavior of ultra-high
[139] Ma J, Orgass M, Dehn F, Schmidt D, Tue N. Comparative investigations on ultra- performance fiber-reinforced concretes. Compos B Eng 2016;86:214–20.
high performance concrete with and without coarse aggregates. In: International [170] Liang X, Wu C, Su Y, Chen Z, Li Z. Development of ultra-high performance
symposium on ultra high performance concrete. Germany: Kassel; 2004. concrete with high fire resistance. Construct Build Mater 2018;179:400–12.
[140] Peng Y, Wu H, Fang Q, Liu J, Gong Z. Impact resistance of basalt aggregated UHP- [171] Yoo D-Y, Kim M-J. High energy absorbent ultra-high-performance concrete with
SFRC/fabric composite panel against small caliber arm. Int J Impact Eng 2016;88: hybrid steel and polyethylene fibers. Construct Build Mater 2019;209:354–63.
201–13. [172] Fan L, Meng W, Teng L, Khayat KH. Effect of steel fibers with galvanized coatings
[141] Fládr J, Bílý P. Specimen size effect on compressive and flexural strength of high- on corrosion of steel bars embedded in UHPC. Compos B Eng 2019;177:107445.
strength fibre-reinforced concrete containing coarse aggregate. Compos B Eng [173] Banthia N, Foy C. Marine curing of steel fiber composites. J Mater Civ Eng 1989;1
2018;138:77–86. (2):86–96.
[142] Senft S, Gallegos S, Manson DP, Gonzales C. Chemical admixtures for concrete. [174] Shah SP, Daniel JI, Ahmad SH, Arockiasamy M, Balaguru P, Ball CG, Ball Jr HP,
CRC Press; 1999. Banthia N, Batson GB, Bentur A. Guide for specifying, proportioning, mixing,
[143] Zhang L, Du W, Wang D, Wang F, Fang K, Yu J, Sheng B. Syntheses of placing, and finishing steel fiber reinforced concrete. ACI Mater J 1993;90(1):
polycarboxylate superplasticizers: microwave induction versus conventional 94–101.
thermal induction. Compos B Eng 2021;207:108560. [175] Zhou Y, Xi B, Yu K, Sui L, Xing F. Mechanical properties of hybrid ultra-high
[144] Ilg M, Plank J. Non-adsorbing small molecules as auxiliary dispersants for performance engineered cementitous composites incorporating steel and
polycarboxylate superplasticizers. Colloid Surface Physicochem Eng Aspect 2020; polyethylene fibers. Materials 2018;11(8):1448.
587:124307. [176] Kang S-T, Choi J-I, Koh K-T, Lee KS, Lee BY. Hybrid effects of steel fiber and
[145] Li P, Yu Q, Brouwers H. Effect of PCE-type superplasticizer on early-age behaviour microfiber on the tensile behavior of ultra-high performance concrete. Compos
of ultra-high performance concrete (UHPC). Construct Build Mater 2017;153: Struct 2016;145:37–42.
740–50. [177] Choi J-I, Lee BY. Bonding properties of basalt fiber and strength reduction
[146] Roussel N. Understanding the rheology of concrete. Elsevier; 2011. according to fiber orientation. Materials 2015;8(10):6719–27.
[147] Van Der Vurst F, Grünewald S, Feys D, Lesage K, Vandewalle L, Vantomme J, De [178] Ahmad S, Rasul M, Adekunle SK, Al-Dulaijan SU, Maslehuddin M, Ali SI.
Schutter G. Effect of the mix design on the robustness of fresh self-compacting Mechanical properties of steel fiber-reinforced UHPC mixtures exposed to
concrete. Cement Concr Compos 2017;82:190–201. elevated temperature: effects of exposure duration and fiber content. Compos B
[148] ZANA R. Introduction to surfactants and surfactant self-assemblies. In: Dynamics Eng 2019;168:291–301.
of surfactant self-assemblies. CRC Press; 2005. p. 18–52. [179] Huang H, Wang R, Gao X. Improvement effect of fiber alignment on resistance to
[149] Bischoff DL, Toepel A. Laboratory testing of portland cement concrete patch elevated temperature of ultra-high performance concrete. Compos B Eng 2019;
material modified to reduce or eliminate shrinkage. Wisconsin Department of 177:107454.
Transportation, Division of Transportation; 2004. [180] Kim MK, Kim DJ, An Y-K. Electro-mechanical self-sensing response of ultra-high-
[150] Zhan PM, He ZH. Application of shrinkage reducing admixture in concrete: a performance fiber-reinforced concrete in tension. Compos B Eng 2018;134:
review. Construct Build Mater 2019;201:676–90. 254–64.
[151] Bentz D, Geiker MR, Hansen KK. Shrinkage-reducing admixtures and early-age [181] Zhang Y, Ju JW, Chen Q, Yan Z, Zhu H, Jiang Z. Characterizing and analyzing the
desiccation in cement pastes and mortars. Cement Concr Res 2001;31(7): residual interfacial behavior of steel fibers embedded into cement-based matrices
1075–85. after exposure to high temperatures. Compos B Eng 2020;191:107933.
[152] Carballosa P, Calvo JG, Revuelta D. Influence of expansive calcium [182] Akbar A, Liew K. Influence of elevated temperature on the microstructure and
sulfoaluminate agent dosage on properties and microstructure of expansive self- mechanical performance of cement composites reinforced with recycled carbon
compacting concretes. Cement Concr Compos 2020;107:103464. fibers. Compos B Eng 2020;198:108245.
[153] Corinaldesi V, Nardinocchi A. Mechanical characterization of Engineered [183] Liu K, Cheng X, Li J, Gao X, Cao Y, Guo X, Zhuang J, Zhang C. Effects of
Cement-based Composites prepared with hybrid fibres and expansive agent. microstructure and pore water on electrical conductivity of cement slurry during
Compos B Eng 2016;98:389–96. early hydration. Compos B Eng 2019;177:107435.
[154] Mo L, Deng M, Tang M. Effects of calcination condition on expansion property of [184] Lee SH, Kim S, Yoo D-Y. Hybrid effects of steel fiber and carbon nanotube on self-
MgO-type expansive agent used in cement-based materials. Cement Concr Res sensing capability of ultra-high-performance concrete. Construct Build Mater
2010;40(3):437–46. 2018;185:530–44.
[155] Yoo D-Y, Park J-J, Kim S-W, Yoon Y-S. Combined effect of expansive and [185] Ozawa M, Parajuli SS, Uchida Y, Zhou B. Preventive effects of polypropylene and
shrinkage-reducing admixtures on the properties of ultra high performance fiber- jute fibers on spalling of UHPC at high temperatures in combination with waste
reinforced concrete. J Compos Mater 2014;48(16):1981–91. porous ceramic fine aggregate as an internal curing material. Construct Build
[156] Yang L, Shi C, Wu Z. Mitigation techniques for autogenous shrinkage of ultra- Mater 2019;206:219–25.
high-performance concrete–A review. Compos B Eng 2019;178:107456. [186] Chun B, Yoo D-Y. Hybrid effect of macro and micro steel fibers on the pullout and
[157] Yan P, Qin X. The effect of expansive agent and possibility of delayed ettringite tensile behaviors of ultra-high-performance concrete. Compos B Eng 2019;162:
formation in shrinkage-compensating massive concrete. Cement Concr Res 2001; 344–60.
31(2):335–7. [187] Yoo D-Y, Banthia N, Kang S-T, Yoon Y-S. Size effect in ultra-high-performance
[158] Higuchi T, Eguchi M, Morioka M, Sakai E. Hydration and properties of expansive concrete beams. Eng Fract Mech 2016;157:86–106.
additive treated high temperature carbonation. Cement Concr Res 2014;64:11–6. [188] Yoo D-Y, Yoon Y-S. Structural performance of ultra-high-performance concrete
[159] Liu F, Shen S-L, Hou D-W, Arulrajah A, Horpibulsuk S. Enhancing behavior of beams with different steel fibers. Eng Struct 2015;102:409–23.
large volume underground concrete structure using expansive agents. Construct [189] Wille K, Kim DJ, Naaman AE. Strain-hardening UHP-FRC with low fiber contents.
Build Mater 2016;114:49–55. Mater Struct 2011;44(3):583–98.

29
J. Du et al. Composites Part B 224 (2021) 109220

[190] Yoo D-Y, Kim J-J, Chun B. Dynamic pullout behavior of half-hooked and twisted [222] Liu J, Farzadnia N, Khayat KH, Shi C. Effects of SAP characteristics on internal
steel fibers in ultra-high-performance concrete containing expansive agents. curing of UHPC matrix. Construct Build Mater 2021;280:122530.
Compos B Eng 2019;167:517–32. [223] Liu J, Farzadnia N, Shi C, Ma X. Shrinkage and strength development of UHSC
[191] Wu Z, Khayat KH, Shi C. How do fiber shape and matrix composition affect fiber incorporating a hybrid system of SAP and SRA. Cement Concr Compos 2019;97:
pullout behavior and flexural properties of UHPC? Cement Concr Compos 2018; 175–89.
90:193–201. [224] Liu J, Shi C, Farzadnia N, Ma X. Effects of pretreated fine lightweight aggregate
[192] Pompo A, Stupak P, Nicolais L, Marchese B. Analysis of steel fibre pull-out from a on shrinkage and pore structure of ultra-high strength concrete. Construct Build
cement matrix using video photography. Cement Concr Compos 1996;18(1):3–8. Mater 2019;204:276–87.
[193] Laranjeira F, Molins C, Aguado A. Predicting the pullout response of inclined [225] Ferrara L, Ferreira SR, della Torre M, Krelani V, de Silva FA, Toledo Filho RD.
hooked steel fibers. Cement Concr Res 2010;40(10):1471–87. Effect of cellulose nanopulp on autogenous and drying shrinkage of cement based
[194] Yoo DY, Jang YS, Chun B, Kim S. Chelate effect on fiber surface morphology and composites. In: Nanotechnology in construction. Springer; 2015. p. 325–30.
its benefits on pullout and tensile behaviors of ultra-high-performance concrete. [226] Hisseine OA, Soliman NA, Tolnai B, Tagnit-Hamou A. Nano-engineered ultra-high
Cement Concr Compos 2021;115:103864. performance concrete for controlled autogenous shrinkage using nanocellulose.
[195] Oh T, You I, Banthia N, Yoo D-Y. Deposition of nanosilica particles on fiber Cement Concr Res 2020;137:106217.
surface for improving interfacial bond and tensile performances of ultra-high- [227] Li S, Cheng S, Mo L, Deng M. Effects of steel slag powder and expansive agent on
performance fiber-reinforced concrete. Compos B Eng 2021:109030. the properties of ultra-high performance concrete (UHPC): based on a case study.
[196] Pi Z, Xiao H, Liu R, Liu M, Li H. Effects of brass coating and nano-SiO2 coating on Materials 2020;13(3):683.
steel fiber–matrix interfacial properties of cement-based composite. Compos B [228] Wehbe Y, Ghahremaninezhad A. Combined effect of shrinkage reducing
Eng 2020;189:107904. admixtures (SRA) and superabsorbent polymers (SAP) on the autogenous
[197] Pi Z, Xiao H, Liu R, Li H. Combination usage of nano-SiO2-coated steel fiber and shrinkage, hydration and properties of cementitious materials. Construct Build
silica fume and its improvement effect on SFRCC. Compos B Eng 2021:109022. Mater 2017;138:151–62.
[198] Zhang W, Zou X, Wei F, Wang H, Zhang G, Huang Y, Zhang Y. Grafting SiO2 [229] Zuo W, Feng P, Zhong P, Tian Q, Gao N, Wang Y, Yu C, Miao C. Effects of novel
nanoparticles on polyvinyl alcohol fibers to enhance the interfacial bonding polymer-type shrinkage-reducing admixture on early age autogenous deformation
strength with cement. Compos B Eng 2019;162:500–7. of cement pastes. Cement Concr Res 2017;100:413–22.
[199] Rezaie AB, Liebscher M, Ranjbarian M, Simon F, Zimmerer C, Drechsler A, [230] Soliman A, Nehdi M. Effects of shrinkage reducing admixture and wollastonite
Frenzel R, Synytska A, Mechtcherine V. Enhancing the interfacial bonding microfiber on early-age behavior of ultra-high performance concrete. Cement
between PE fibers and cementitious matrices through polydopamine surface Concr Compos 2014;46:81–9.
modification. Compos B Eng 2021;217:108817. [231] Graybeal BA. Characterization of the behavior of ultra-high performance
[200] Chun B, Yoo DY, Banthia N. Achieving slip-hardening behavior of sanded straight concrete. Doctoral dissertation. 2005.
steel fibers in ultra-high-performance concrete. Cement Concr Compos 2020: [232] Cheng Y-H, Zhu B-L, Yang S-H, Tong B-Q. Design of concrete mix proportion
103669. based on particle packing voidage and test research on compressive strength and
[201] Yoo D-Y, Oh T, Chun B. Highly ductile ultra-rapid-hardening mortar containing elastic modulus of concrete. Materials 2021;14(3):623.
oxidized polyethylene fibers. Construct Build Mater 2021;277:122317. [233] Van Der Putten J, Dils J, Minne P, Boel V, De Schutter G. Determination of
[202] Ženíšek M, Vlach T, Řepka J, Pavlů T. Segregation of steel fibres of UHPFRC. In: packing profiles for the verification of the compressible packing model in case of
Solid state phenomena. Trans Tech Publ; 2018. UHPC pastes. Mater Struct 2017;50(2):1–14.
[203] Bensted J. Effects of the clinker-gypsum grinding temperature upon early [234] Powers TC. A discussion of cement hydration in relation to the curing of concrete.
hydration of Portland cement. Cement Concr Res 1982;12(3):341–8. In: Highway research board proceedings; 1948.
[204] Xie T, Fang C, Ali MM, Visintin P. Characterizations of autogenous and drying [235] Mostafa SA, Faried AS, Farghali AA, El-Deeb MM, Tawfik TA, Majer S, Abd
shrinkage of ultra-high performance concrete (UHPC): an experimental study. Elrahman M. Influence of nanoparticles from waste materials on mechanical
Cement Concr Compos 2018;91:156–73. properties, durability and microstructure of UHPC. Materials 2020;13(20):4530.
[205] Yang R, Yu R, Shui Z, Gao X, Xiao X, Zhang X, Wang Y, He Y. Low carbon design [236] Meng W, Khayat KH. Effect of graphite nanoplatelets and carbon nanofibers on
of an ultra-high performance concrete (UHPC) incorporating phosphorous slag. rheology, hydration, shrinkage, mechanical properties, and microstructure of
J Clean Prod 2019;240:118157. UHPC. Cement Concr Res 2018;105:64–71.
[206] Teng L, Meng W, Khayat KH. Rheology control of ultra-high-performance [237] Norhasri MM, Hamidah M, Fadzil AM. Inclusion of nano metaclayed as additive in
concrete made with different fiber contents. Cement Concr Res 2020;138:106222. ultra high performance concrete (UHPC). Construct Build Mater 2019;201:590–8.
[207] Huang H, Ye G. Examining the “time-zero” of autogenous shrinkage in high/ultra- [238] Liu C, He X, Deng X, Wu Y, Zheng Z, Liu J, Hui D. Application of nanomaterials in
high performance cement pastes. Cement Concr Res 2017;97:107–14. ultra-high performance concrete: a review. Nanotechnol Rev 2020;9(1):1427–44.
[208] Ghafari E, Ghahari SA, Costa H, Júlio E, Portugal A, Durães L. Effect of [239] Wu Z, Shi C, Khayat KH. Multi-scale investigation of microstructure, fiber pullout
supplementary cementitious materials on autogenous shrinkage of ultra-high behavior, and mechanical properties of ultra-high performance concrete with
performance concrete. Construct Build Mater 2016;127:43–8. nano-CaCO3 particles. Cement Concr Compos 2018;86:255–65.
[209] De Sensale GR, Ribeiro AB, Gonçalves A. Effects of RHA on autogenous shrinkage [240] Aïtcin P-C. High performance concrete. CRC press; 1998.
of Portland cement pastes. Cement Concr Compos 2008;30(10):892–7. [241] Shen P, Lu L, He Y, Wang F, Hu S. The effect of curing regimes on the mechanical
[210] Li P, Brouwers H, Chen W, Yu Q. Optimization and characterization of high- properties, nano-mechanical properties and microstructure of ultra-high
volume limestone powder in sustainable ultra-high performance concrete. performance concrete. Cement Concr Res 2019;118:1–13.
Construct Build Mater 2020;242:118112. [242] Zhang Y, Li X, Zhu Y, Shao X. Experimental study on flexural behavior of
[211] Justs J, Wyrzykowski M, Bajare D, Lura P. Internal curing by superabsorbent damaged reinforced concrete (RC) beam strengthened by toughness-improved
polymers in ultra-high performance concrete. Cement Concr Res 2015;76:82–90. ultra-high performance concrete (UHPC) layer. Compos B Eng 2020;186:107834.
[212] Anshuang S, Ling Q, Shoujie Z, Jiayang Z, Zhaoyu L. Effects of shrinkage reducing [243] Lantsoght EO. How do steel fibers improve the shear capacity of reinforced
agent and expansive admixture on the volume deformation of ultrahigh concrete beams without stirrups? Compos B Eng 2019;175:107079.
performance concrete. Adv. Mater. Sci. Eng. 2017;2017:6384859. [244] Kang ST, Lee BY, Kim J-K, Kim YY. The effect of fibre distribution characteristics
[213] Zhibin Z, Lingling X, Minshu T. Synergistic effect of MgO-based expansive agent on the flexural strength of steel fibre-reinforced ultra high strength concrete.
and shrinkage-reducing admixture on compensating the shrinkage of Construct Build Mater 2011;25(5):2450–7.
cementitious materials, vol. 262. Special Publication; 2009. p. 395–406. [245] Teng L, Huang H, Du J, Khayat K. Prediction of fiber orientation and flexural
[214] Shen P, Lu L, He Y, Wang F, Lu J, Zheng H, Hu S. Investigation on expansion effect performance of UHPC based on suspending mortar rheology and casting method.
of the expansive agents in ultra-high performance concrete. Cement Concr Cement Concr Compos 2021;122:104142.
Compos 2020;105:103425. [246] Meng W, Khayat KH, Bao Y. Flexural behaviors of fiber-reinforced polymer fabric
[215] Tazawa E-i, Miyazawa S, Kasai T. Chemical shrinkage and autogenous shrinkage reinforced ultra-high-performance concrete panels. Cement Concr Compos 2018;
of hydrating cement paste. Cement Concr Res 1995;25(2):288–92. 93:43–53.
[216] Henkensiefken R, Bentz D, Nantung T, Weiss J. Volume change and cracking in [247] Martinie L, Roussel N. Simple tools for fiber orientation prediction in industrial
internally cured mixtures made with saturated lightweight aggregate under practice. Cement Concr Res 2011;41(10):993–1000.
sealed and unsealed conditions. Cement Concr Compos 2009;31(7):427–37. [248] Swamy R. Fibre reinforcement of cement and concrete. Matériaux et Construction
[217] Lyu Z, Guo Y, Chen Z, Shen A, Qin X, Yang J, Zhao M, Wang Z. Research on 1975;8(3):235–54.
shrinkage development and fracture properties of internal curing pavement [249] Banthia N, Trottier J-F. Concrete reinforced with deformed steel fibers, part I:
concrete based on humidity compensation. Construct Build Mater 2019;203: bond-slip mechanisms. Mater. J. 1994;91(5):435–46.
417–31. [250] Armelin HS, Banthia N. Predicting the flexural postcracking performance of steel
[218] Yang L, Shi C, Wu Z. Mitigation techniques for autogenous shrinkage of ultra- fiber reinforced concrete from the pullout of single fibers. Mater. J. 1997;94(1):
high-performance concrete–A review. Compos B Eng 2019:107456. 18–31.
[219] Lyu Z, Shen A, Mo S, Chen Z, He Z, Li D, Qin X. Life-cycle crack resistance and [251] Svec O, Skocek J, Stang H, Olesen JF, Thrane L. Application of the fluid dynamics
micro characteristics of internally cured concrete with superabsorbent polymers. model to the field of fibre reinforced self-compacting concrete. In: Proceedings of
Construct Build Mater 2020;259:119794. the numerical modeling strategies for sustainable concrete structures (NMSSC);
[220] Lyu Z, Shen A, He Z, Wang W, Mo S, Chen Z, Qin X. Absorption characteristics and 2012. p. 1–9.
shrinkage mitigation of superabsorbent polymers in pavement concrete. Int J [252] Song Q, Yu R, Shui Z, Wang X, Rao S, Lin Z. Optimization of fibre orientation and
Pavement Eng 2020:1–15. distribution for a sustainable ultra-high performance fibre reinforced concrete
[221] Sun B, Wu H, Song W, Li Z, Yu J. Design methodology and mechanical properties (UHPFRC): experiments and mechanism analysis. Construct Build Mater 2018;
of Superabsorbent Polymer (SAP) cement-based materials. Construct Build Mater 169:8–19.
2019;204:440–9.

30
J. Du et al. Composites Part B 224 (2021) 109220

[253] Song Q, Yu R, Shui Z, Wang X, Rao S, Lin Z, Wang Z. Key parameters in optimizing [275] Pease B, Geiker M, Stang H, Weiss J. The design of an instrumented rebar for
fibres orientation and distribution for ultra-high performance fibre reinforced assessment of corrosion in cracked reinforced concrete. Mater Struct 2011;44(7):
concrete (UHPFRC). Construct Build Mater 2018;188:17–27. 1259–71.
[254] Huang H, Su A, Gao X, Yang Y. Influence of formwork wall effect on fiber [276] Rajczakowska M, Habermehl-Cwirzen K, Hedlund H, Cwirzen A. Autogenous self-
orientation of UHPC with two casting methods. Construct Build Mater 2019;215: healing: a better solution for concrete. J Mater Civ Eng 2019;31(9):03119001.
310–20. [277] Beglarigale A, Eyice D, Tutkun B, Yazıcı H. Evaluation of enhanced autogenous
[255] Qureshi TS, Panesar DK. Nano reinforced cement paste composite with self-healing ability of UHPC mixtures. Construct Build Mater 2021;280:122524.
functionalized graphene and pristine graphene nanoplatelets. Compos B Eng [278] Deschenes Jr RA. Mitigation and evaluation of alkali-silica reaction (ASR) and
2020;197:108063. freezing and thawing in concrete transportation structures. 2017.
[256] Yoo D-Y, Kim S-W, Park J-J. Comparative flexural behavior of ultra-high- [279] Rachida I, Cyr M, Tagnit-Hamou A. Use of waste glass in cement-based materials.
performance concrete reinforced with hybrid straight steel fibers. Construct Build Déchets, Sciences et Techniques 2010;57:10.
Mater 2017;132:219–29. [280] Mazzacane P, Ricciotti R, Teply F, Tollini E, Corvez D. MUCEM: the builder’s
[257] Wille K, El-Tawil S, Naaman AE. Properties of strain hardening ultra high perspective. Proceedings UHPFRC 2013:3–16.
performance fiber reinforced concrete (UHP-FRC) under direct tensile loading. [281] Aubry S, Bompas P, Vaudeville B, Corvez D, Lagrange T, Mazzacane P, Brizou A.
Cement Concr Compos 2014;48:53–66. A UHPFRC cladding challenge: the fondation Louis Vuitton pour la création"
[258] El-Hacha R, Chen D. Behaviour of hybrid FRP–UHPC beams subjected to static Iceberg. In: 2nd RILEM-fib-AFGC international symposium on ultra-high
flexural loading. Compos B Eng 2012;43(2):582–93. performance fibre-reinforced concrete; 2013.
[259] Ren L, Fang Z, Wang K. Design and behavior of super-long span cable-stayed [282] Mazzacane P, Ricciotti R, Lamoureux G, Corvez D. Roofing of the stade jean Bouin
bridge with CFRP cables and UHPC members. Compos B Eng 2019;164:72–81. in UHPFRC. In: Proceedings of international symposium on ultra-high
[260] Meng W, Khayat KH. Experimental and numerical studies on flexural behavior of performance fibre-reinforced concrete; 2013.
ultrahigh-performance concrete panels reinforced with embedded glass fiber- [283] Muttoni A, Brauen U, Jaquier J-L, Moullet D. A new roof for the olympic museum
reinforced polymer grids. Transport Res Rec 2016;2592(1):38–44. at Lausanne, Switzerland. In: Proceedings of international symposium on ultra-
[261] Soltan DG, Li VC. Nacre-inspired composite design approaches for large-scale high performance fiber-reinforced concrete; 2013.
cementitious members and structures. Cement Concr Compos 2018;88:172–86. [284] Menétrey P. UHPFRC cladding for the Qatar national museum. In: Proceedings of
[262] Cao Y, Liu G, Brouwers H, Yu Q. Enhancing the low-velocity impact resistance of international symposium on ultra-high performance fiber-reinforced concrete;
ultra-high performance concrete by an optimized layered-structure concept. 2013. Marseille, France.
Compos B Eng 2020;200:108221. [285] Gaudillière N, Duballet R, Bouyssou C, Mallet A, Roux P, Zakeri M,
[263] Cao Y, Li P, Brouwers H, Sluijsmans M, Yu Q. Enhancing flexural performance of Dirrenberger J. Large-scale additive manufacturing of ultra-high-performance
ultra-high performance concrete by an optimized layered-structure concept. concrete of integrated formwork for truss-shaped pillars. In: Robotic fabrication
Compos B Eng 2019;171:154–65. in architecture, art and design. Springer; 2018.
[264] Zhang G, Wu C, Hou D, Yang J, Sun D, Zhang X. Effect of environmental pH [286] Graybeal BA. Structural behavior of ultra-high performance concrete prestressed
values on phase composition and microstructure of Portland cement paste under I-girders. United States: Federal Highway Administration; 2006.
sulfate attack. Compos B Eng 2021;216:108862. [287] Graybeal BA. Structural behavior of a 2nd generation ultra-high performance
[265] Tam CM, Tam VW, Ng KM. Assessing drying shrinkage and water permeability of concrete pi-girder. United States: Federal Highway Administration; 2009.
reactive powder concrete produced in Hong Kong. Construct Build Mater 2012;26 [288] Aaleti S, Petersen B, Sritharan S. Design guide for precast UHPC waffle deck panel
(1):79–89. system, including connections. United States: Federal Highway Administration;
[266] Roux N, Andrade C, Sanjuan M. Experimental study of durability of reactive 2013.
powder concretes. J Mater Civ Eng 1996;8(1):1–6. [289] Zhou M, Lu W, Song J, Lee GC. Application of ultra-high performance concrete in
[267] Franke L, Schmidt H, Deckelmann G. Behavior of ultra-high performance concrete bridge engineering. Construct Build Mater 2018;186:1256–67.
with respect to chemical attack. In: Proceedings of the 2nd international [290] Graybeal B. Design and construction of field-cast UHPC connections. United
symposium on ultra-high performance concrete. Germany: Kassel; 2008. States: Federal Highway Administration; 2014.
[268] Piérard J, Dooms B, Cauberg N. Durability evaluation of different types of UHPC. [291] Wibowo H, Sritharan S. Use of ultra-high-performance concrete for bridge deck
In: Proceedings of the RILEM-fib-AFGC international symposium on ultra-high overlays. United States: Federal Highway Administration; 2018.
performance fiber-reinforced concrete; 2013. [292] Haber ZB, Munoz JF, De la Varga I, Graybeal BA. Bond characterization of UHPC
[269] Chen Y, Yu R, Wang X, Chen J, Shui Z. Evaluation and optimization of ultra-high overlays for concrete bridge decks: laboratory and field testing. Construct Build
performance concrete (UHPC) subjected to harsh ocean environment: towards an Mater 2018;190:1056–68.
application of layered double hydroxides (LDHs). Construct Build Mater 2018; [293] Kennedy D, Habel K, Fraser G. Ultra high-performance concrete column jacket
177:51–62. retrofit for the mission bridge. In: 11th Canadian conference on earthquake
[270] Heinz D, Dehn F, Urbonas L. Fire resistance of ultra high performance concrete engineering; 2015.
(UHPC)-Testing of laboratory samples and columns under load. In: International [294] Gamarra J. Potential application of ultra-high performance fiber-reinforced
symposium on ultra high performance concrete. Kassel Germany; 2004. concrete with wet-mix shotcrete system in tunneling. Doctoral dissertation. 2016.
[271] Ahlborn TM, Misson DL, Peuse EJ, Gilbertson CG. Durability and strength [295] Sritharan S, Schmitz GM. Design of tall wind turbine towers utilizing UHPC. In:
characterization of ultra-high performance concrete under variable curing 2nd international symposium on ultra-high performance fibre-reinforced
regimes. In: Fehling E, Schmidt M, Stürwald S, editors. Proc. 2nd Int. Symp. On concrete. Marseille, France: UHPFRC; 2013.
ultra high performance concrete. Germany: Kassel; 2008. [296] Khan MU, Ahmad S, Naqvi AA, Al-Gahtani HJ. Shielding performance of heavy-
[272] Scheydt JC, Muller S. Microstructure of ultra high performance concrete (UHPC) weight ultra-high-performance concrete against nuclear radiation. Prog Nucl
and its impact on durability. In: Proceedings of the 3rd international symposium Energy 2020;130:103550.
on UHPC and nanotechnology for high performance construction materials. [297] Ahmad S, Mohaisen KO, Adekunle SK, Al-Dulaijan SU, Maslehuddin M. Influence
Germany: Kassel; 2012. of admixing natural pozzolan as partial replacement of cement and microsilica in
[273] Smarzewski P, Barnat-Hunek D. Effect of fiber hybridization on durability related UHPC mixtures. Construct Build Mater 2019;198:437–44.
properties of ultra-high performance concrete. Int. J. Concr. Struct. Mater. 2017; [298] Zhang D, Tan KH, Dasari A, Weng Y. Effect of natural fibers on thermal spalling
11(2):315–25. resistance of ultra-high performance concrete. Cement Concr Compos 2020:
[274] Liu S, Sun W, Lin W, Lai J. Preparation and durability of a high performance 103512.
concrete with natural ultra-fine particles. Guisuanyan Xuebao 2003;31:1080–5.

31