Final Report

BITS F540
Research Practice

ROLE OF ULTRA HIGH

PERFORMANCE CONCRETE
(UHPC) FOR BLAST AND DYNAMIC
LOADS

Name: Abhishek Kumar

ID No: 2018H1430031H
Guided By: Dr. Kalyana Rama JS
 2

ABSTRACT/OBJECTIVE

The aim of the project is to study how Ultra High Performance Concrete
(UHPC) can be utilized in the construction of defence and security related
infrastructures. Defence structures such as barracks, offices and other security-
related installations are particularly vulnerable to damage and destruction by
terrorist attacks, bombings, explosives etc. Hence, it is important to study a
construction material that provides better strength and safety to such structures.
 3

INDEX

Introduction 6

Methodology 7

Literature Review 8 -23

Development Of UHPC Mixes Based On Plastic Viscosity Of

Cement Paste 24

Bibliography and References 45

 4

LIST OF TABLES

Sl. No. Name Page No.

1. Density of materials used in design 9
2. Mix proportions of the two UHPFRC samples 10
3. Binder amount and 28 days’ compressive strength of optimised 10
and non – optimised UHPFRC
4. Composition of Ductal 12
5. Composition of Ductal 16
6. Material properties of UHPC and NSC samples 16
7. Test slabs and results of blast events 18
8. Properties of test specimens 19
9. Typical viscosities of some common materials 25
10. Mix combinations 30
11. Bingham parameters for OPC paste 33
12. Bingham parameters for 75% OPC + 25% GGBS paste 33
13. Bingham parameters for 80% OPC + 20% FA paste 33
14. Bingham parameters for ternary 50% OPC + 25% FA + 25% GGBS paste 33
15. Measured plastic viscosity of cement pastes for Case I 33
16. Measured plastic viscosity of cement pastes for Case II 34
17. Measured plastic viscosity of cement pastes for Case III 34
18. Measured plastic viscosity of cement pastes for Case IV 35
19. Packing fractions for different types of packing of particles 39
 5

LIST OF FIGURES

Sl. No. Name Page No.

1. Typical UHPC and UHPFRC tensile stress – crack opening diagram 8
2. Force – displacement curve of UHPC samples for different types and 11
proportions of reinforcement
3. Stress – strain curves for cylindrical tests on (a) UHPC, and (b) NC 13
4. Samples prepared for testing
5. UHPC prisms before test, b) UHPC prisms after test, 13
and c) NC prisms after test
6. UHPC splitting cylinder after test, and b) NC splitting cylinder after test 14
7. Blast response of (a) UHPC-D1 (b) UHPC-D2 (c) UHPC-D3A 17
(d) UHPC-D3B0
8. Blast response of (a) UHPC-D4, and (b) NSC 18
9. Sample pressure-time histories from specimen CRC-2%A-#3HS 20
10. Maximum and residual displacements for slabs at selected blast tests 20
11. Classification of fluids with τ as a function of γ (Rao A, 2010) 27
12. Brookfield Viscometer DV3T 32
13. Regression curve for characteristic compressive strength water to cement ratio 37
 6

INTRODUCTION
High strength concrete (HSC) research began in the 1950s and has seen fast
development ever since. According to the American Concrete Institute, HSC is
defined as concrete having a compressive cylinder strength greater than 40
MPa. In Australia, HSC is defined as concrete with a characteristic of 28-day
cylinder strength greater than 50 MPa, and up to 100 MPa.

In modern construction, HSC is favourable in high-rise buildings since it

facilitates the design of smaller structural-member cross-sections for columns
and walls, leaving more usable floor area. In marine structures, the low
permeability characteristics of HSC reduce the risk of corrosion of steel
reinforcement and improve the durability of concrete structures. HSC has also
been used in design of safer structures because key supporting members are
often designed to resist wind, seismic, and other impact forces. After terrorist
attack on September 11, 2001, the main building of the rebuilt World Trade
Center complex, i.e., the One World Trade Center chose HSC. The supporting
columns were made of steel and concrete ranging in strength from 59 to 97
MPa. Columns on the first 40 floors were made from 82 to 97 MPa concrete and
the upper floors with 59-69 MPa mix designs.

In the late 1970s, when superplasticizer began to be developed

and used to reduce the water/cement or water/binder ratio rather than being
exclusively used to fluidize the usual concrete, it was found that concretes with
a very low water/cement or water/binder ratio showed high compressive
strength and at the same time they also had other improved characteristics such
as higher flexural strength, lower permeability, improved abrasion resistance,
and better durability. It was found that a key factor influencing the concrete
strength is the water/ cement or water/binder ratio.

The term high-performance concrete, was a relatively new term for

concrete, which was normally inclusive of the term “HSC.” High-performance
concrete is normally equipped with better material durability and abrasion
resistance. It is generally defined as a low water/binder concrete which has an
 7

adequate curing with water. In a low water/binder ratio mix, there are more
cement grains and consequently less water per unit volume so that cement
grains are much closer to each other leading to reduced porosity.

In recent decades, concrete technology has made remarkable advances, and

many works in this field had led to the development of ultra-high performance
concretes (UHPC). UHPC is characterized by its material composition: fine-
grained sand, silica fume, steel fibers, special blends of high-strength Portland
cement, and an elimination of coarse aggregate. UHPC is normally
characterized by compressive strengths in excess of 150 MPa. Reinforced by a
small amount of steel fibers, UHPC becomes ductile and can reach direct tensile
strength exceeding 15 MPa and flexural strength exceeding 50 MPa. For the
first time, concrete material could be designed to accommodate tension.
Conventional concrete contains a significant capillary porosity while UHPC has
almost no capillary pores and is highly resistant to the water and ion
penetration, therefore the durability of UHPC is much higher than conventional
concrete. It can be used in harsh environments such as the wearing course of a
bridge deck and pavements against chlorides, alkalis, or deicing salts.

The research of UHPC began in the 1980s when Danish researcher Hans
Hendrik Bache developed Compact Reinforced Concrete in which fine
aggregates were used synergistically with fiber content. Bache’s idea was taken
by the French contractor Bouygues, who later cooperated with Lafarge to
develop a new mix “Reactive Powder Concrete” (RPC). The concept of RPC
continues to exist in the form of commercialized UHPCs such as “Ductal.” The
earliest application of UHPC was in Cattenom, France, where UHPC beams
were used to replace the original steel beams in the cooling towers of a power
station. In this case, UHPC was chosen not because of its high strength but
rather because of its outstanding durability that can make the structure with a
long service life in a highly corrosive environment.

Ultra High Performance Concrete (UHPC) is a type of concrete consisting

of fine — grained sand, silica fumes, steel fibers, special blends of high
strength Portland cement, and very low quantities of coarse aggregate. The low
water — binder ratio and the addition of steel fibers allows it to achieve a
compression strength of 150MPa flexural strength of 150MPa tensile strength
of 15 MPa. It is also more durable and possesses better ductility than high
 8

strength and normal strength concrete. Due to this UHPC can be used in structures
designed to be protected against blast impact earthquake and explosive etc.
Conventional RC system can resist explosions and bomb blast occurring
beyond a certain distance, but for closely occurring blast loads, RC structures tend
to Spall resulting in high velocity debris and ultimately leading to progressive
collapse. UHPC structures can resist such loads better, thus are useful in
construction of defence and security related infrastructure that are subjected to
terrorist a t t a c k s , b l a s t s e t c n e e d t o b e mo r e d u r a b l e a n d
r e s is t i v e t o s u c h l o a d s .
 9

METHODOLGY
1) Collection of research papers, articles and preparation of review report
based on existing literature.

2) Study of (a) the material constituents used along with their physical and
chemical properties (b) mix proportioning of UHPC as per existing
literature (c) ductility of UHPC and methods to improve it.

3) Role of various constituent raw materials used in addressing the failure

and fragmentation of UHPC for different applications.

4) Exploring the role of UHPC as a construction material for compression

members like columns and walls for lateral load and blast load resistance.

5) Preparing a draft proposal to meet the needs of current Indian scenario

with UHPC as a possible construction material.

6) Develop a report for lab development as a part of one of the labs.

 10

LITERATURE REVIEW

1) Yuliarti Kusumawardaningsiha, Ekkehard Fehlinga, Mohammed

Ismaila, Attitou Amen Mohamed Aboubakra (2015), Tensile strength
behavior of UHPC and UHPFRC

This paper aimed to study the postcracking tensile behavior of UHPC and
UHPFRC. For this purpose, a stress – crack opening behavior test of UHPC and
UHPFCR, was conducted in Kassel University on a type of specimen called
M3Q_210. It was developed by and made in the Official Material Testing
Institute for Construction Industry - AMPA of Kassel University.
In the UHPFRC, steel fiber having length of 9 mm and diameter of 0.10 mm,
was added with a volume of 2 %.
After the test, the compressive strength of the UHPC and UHPFRC after 28
days was ± 180MPa.
This paper concluded that
UHPFRC with 2 % fibres had higher maximum tensile strength and lower crack
opening length than UPHC without fibres. Thus, the tensile strength value was
improved by the addition of fibers.

Fig.1. Typical UHPC and UHPFRC tensile stress – crack opening diagram
 11

2) R. Yu, P. Spiesz, H.J.H. Brouwers (2014), Mix design and properties

assessment of Ultra-High Performance Fibre Reinforced Concrete
(UHPFRC)

This paper investigated the mix design and properties assessment of Ultra-High
Performance Fibre Reinforced Concrete (UHPFRC). The design of the concrete
mixtures is based on the aim to achieve a densely compacted cementitious
matrix, employing the modified Andersen & Andersen particle packing
model. The workability, air content, porosity, flexural and compressive
strengths of the designed UHPFRC were measured and analyzed.

Table 1. Density of materials used in design

In this study, the maximum compressive and flexural strengths at 28 days of the
UHPFRC (with steel fibre 2.5 vol.%) were obtained as 150 MPa and 30 MPa,
respectively. The paper concluded that UHPFRC design with a relatively low
binder amount (650 kg/m3) was possible by utilizing the considered packing
model. After 28 day of curing, it was found that there there was a large amount
of unhydrated cement in the UHPFRC matrix, which could be further replaced
by fillers to improve the workability and cost efficiency of UHPFRC. The
addition of steel fibres was found to decrease the slump flow of UHPFRC
(hence, its workability) and increase its air content in the fresh state and
porosity in the hardened state.
 12

Table 2. Mix proportions of the two UHPFRC samples

Table 3. Binder amount and 28 days’ compressive strength of optimised

and non – optimised UHPFRC
 13

3) N. Cauberg & J. Piérard, O. Remy (2008), Ultra High Performance

Concrete: Mix design and practical applications

This paper focused on following aspects of UHPC such as its mix design of
UHPC, the shrinkage at early age, the fiber reinforcement and its flexural
behavior and optimized the choice of admixtures and micro fillers ; and
aggregate grading, considering two applications for UHPC, cladding panels and
overlays. It obtained a compressive strength between 125 and 180 N/mm2 to be
used for thin cladding panels and overlays respectively.
This paper concluded that the integration of fiber mixes greatly increased the
flexural toughness, which allowed for the production of elements without any
other structural reinforcement, as for instance thin cladding panels with large
spans.

Fig. 2. Force – displacement curve of UHPC samples for different types and
proportions of reinforcement
 14

4) Mohamadreza Shafieifar, Mahsa Farzad, Atorod Azizinamini (2017),

Experimental and numerical study on mechanical properties of Ultra
High Performance Concrete (UHPC)

This paper aimed to study the tensile and compressive behavior of UHPC
through experiments and a comparison is made with Normal Strength Concrete
(NC). The experimental tests including a cylinder and cube compressive test,
flexural, briquette and splitting tension tests to evaluate the ultimate capacity of
the material in compression and tension and its modulus of elasticity.
The materials used in the study were premix i.e. Ductal (containing
cementitious, aggregate, and filler materials), liquids in the form of water and
high-range water-reducing admixture (HRWA). The W/C ratio of all batches
was 0.15. Straight steel fibers of tensile strength 2600 MPa and of concentration
2% by volume of fibers were included in the mix.
For testing, a universal testing machine (UTM) with a maximum capacity of
500 Kips and a 50-kip capacity MTS testing machine was used to apply a load
to the specimens and the displacement was measured using a potentiometer. In
this paper, three test methods were used to determine the tensile strength of
concrete including flexural test, direct tensile test, and splitting tension.

Table 4. Composition of Ductal

 15

Fig. 3. Stress – strain curves for cylindrical tests on (a) UHPC, and (b) NC

Fig. 4. Samples prepared for testing

 16

Fig. 5. UHPC prisms before test, b) UHPC prisms after test, and c) NC
prisms after test

Fig. 6. UHPC splitting cylinder after test, and b) NC splitting cylinder after
test
 17

This paper found that the material properties of UHPC were superior to that of
regular concrete in both compression and tension. The compressive strength of
commercial UHPC (using Ductal) was three to four times greater than normal
strength concrete. Higher modulus of elasticity (approximately two times) of
UHPC specimens was obtained compared with the NC. For UHPC specimens, a
higher tensile strength and ductility of the material compared to regular concrete
was observed (two to four times greater).
 18

5) Jun Li a, Chengqing Wua, Hong Hao, (2015), An experimental and

numerical study of reinforced ultra-high performance concrete slabs
under blast loads, Materials & Design 82 (2015) 64–76

In this paper, the mechanical strength and material ductility of UHPC were
studied by conducting uniaxial compression tests on cylindrical UHPC samples
and the stress strain relationship was obtained. Field blast tests were designed
and carried out on UHPC slabs with varying reinforcements. With a normal
strength concrete slab with conventional reinforcement being tested as control
sample. In these blast scenarios, various damage modes including flexural
damage, combined shear and flexural damage were observed. Impacts from
reinforcement strength (ranging from 300 MPa to 1750 MPa) and scaled
standoff distance (ranging from 0.41 m/kg1/3 to 3.05 m/kg1/3) were
investigated. Slabs Maximum deflection and residual deflection at midspan of
the slabs were measured using Linear Variable Differential Transformer
(LVDT).

Table 5. Composition of Ductal

Table 6. Material properties of UHPC and NSC samples

 19

Test samples included UHPC slabs with various reinforcing ratios and different
types of reinforcing steel. One additional normal strength slab which had
pressure transducers attached to the surface was tested to determine the pressure
loading distribution.
The type of UHPC used in construction was Ductal and was identical for all
four UHPC slabs. In the present study, 2% of steel fibres by was adopted.
In the experiments, each of the four UHPC slabs was referenced with a unique
identifier beginning with ‘D’, followed by a number from 1 to 4. For UHPC-D3
slab, it is designed for two explosives, thus its name is followed by an ending of
‘A’ and ‘B’ to identify the blast scenarios.

Fig. 7. Blast response of (a) UHPC-D1 (b) UHPC-D2 (c) UHPC-D3A (d)
UHPC-D3B
 20

Fig. 8. Blast response of (a) UHPC-D4, and (b) NSC

Table 7. Test slabs and results of blast events

 21

6) Christian Melançon, Sarah De Carufel and Hassan Aoude,

(2016),Blast Behavior of One-Way Panel Components Constructed
with UHPC, First International Interactive Symposium on UHPC –
2016

This paper summarizes the results from three one-way slabs which were tested
under gradually increasing shockwave loads using a high-capacity shock-tube at
the University of Ottawa.
The series includes one control slab built with conventional concrete and two
companion slabs built with ultra-high performance concrete (UHPC).
Results in terms of blast resistance, control of displacements, and damage
tolerance are used to study the effects of the design parameters on the
performance of the panels.
Overall, the results demonstrate significant benefits associated with the use of
UHPC in reinforced concrete slabs tested under extreme blast pressures.
The control slab was cast using plain self-consolidating concrete (SCC) with a
specified strength of 50 MPa (5.8 ksi). The SCC mix properties include a
maximum aggregate size of 10 mm (0.4 in), a sand-to-aggregate ratio of 0.55
and a water-cement ratio of approximately 0.42.
The UHPC specimens were cast using compact reinforced composite (CRC)
with a specified strength of 140 MPa (20 ksi).

Table 8. Properties of test specimens

Each specimen was subjected to gradually increasing blast loads until failure.
Maximum midheight displacements (δmax) and residual mid-height
displacements (δresidual) for each blast test were recorded.
 22

Fig. 9. Sample pressure-time histories from specimen CRC-2%A-#3HS

Fig. 10. Maximum and residual displacements for slabs at selected blast
tests

Using UHPC in slabs improves blast performance by reducing maximum and

residual displacements under equivalent blast loads.
Using UHPC enhances damage tolerance, by improving the control of tensile
racking and minimizing secondary fragmentation in slabs subjected to blast
loads.
The combined use of UHPC and high-strength steel reinforcement leads to
further enhancements in the blast performance of slabs and results in reduced
displacements at equivalent blasts and increased blast capacity.
 23

Using UHPC in slabs improves blast performance by reducing maximum and

residual displacements under equivalent blast loads.
Using UHPC enhances damage tolerance, by improving the control of tensile
racking and minimizing secondary fragmentation in slabs subjected to blast
loads.
The combined use of UHPC and high-strength steel reinforcement leads to
further enhancements in the blast performance of slabs and results in reduced
displacements at equivalent blasts and increased blast capacity.
 24

DEVELOPMENT OF UHPC MIXES BASED

ON PLASTIC VISCOSITY OF CEMENT
PASTE
A mix design method for UHPC based on the desired target plastic viscosity and
compressive strength of the mix was developed, based on a similar mix design
procedure for SCC proposed by M.S. Abo Dhaheer, M.M. Al-Rubaye, W.S.
Alyhya, B.L. Karihaloo* and S. Kulasegaram.

1. INTRODUCTION TO RHEOLOGY OF SELF-COMPACTING

CONCRETE

Rheology is the branch of physics that deals with the deformation and flow
behavior of matter. It mainly deals with the non-Newtonian flow of liquids and
the plastic flow of solids. It was started way back in 1920s to describe the flow
of liquids and the deformation of solids.

• Rheology is important to measure the properties of fluids, to predict the

flow behaviour of liquids and also to model it. It deals with materials whose
deformation is in between liquids and solids

• Important to understand the relation between structure and flow properties

of materials. (H.A.Barnes, J.F., & K., 1999).

It is important because of the scope it offers for characterizing fresh cement

paste, grout, mortar and concrete, and for understanding how they perform in
practical applications. Without satisfactory fresh properties it is unlikely that the
desirable properties of the hardened materials can be achieved. So, to describe
the flow behavior of concrete, rheological study using parameters like plastic
viscosity, yield stress etc. is much better than workability which is calculated by
conducting slump flow test, compaction factor tests etc.
 25

1.1. NEWTONIAN AND NON-NEWTONIAN FLUIDS (Rao A, 2010)

Any fluid that obeys the Newton’s law of viscosity is called a Newtonian fluid.
All other fluids which do not comply with it are termed as non-Newtonian fluids.

In a Newtonian fluid, there is a single coefficient of viscosity for a given

temperature. In a Newtonian fluid, viscosity is therefore a function of temperature
alone and is unaffected by the shear rate or strain rate.

According to the Newton’s law of viscosity, the relationship between viscous

stress and shear rate for an incompressible, isotropic Newtonian fluid is given by

𝑑𝑢
𝜏=µ
𝑑𝑦

where
τ is the shear stress in the fluid
µ is a scalar constant of proportionality, the shear viscosity of the fluid
du/dy is the derivative of the velocity component that is parallel to the direction
of shear, relative to displacement in the perpendicular direction. (Rao A, 2010)
There is no such fluid whose behavior perfectly resembles the Newtonian fluid
but many common liquids and gases such as water and air can be considered as
Newtonian under ordinary conditions (Table 1.1)

Table 1.1 Typical viscosities of some common materials (Tennet R M, 2004)

Approximate
Material
Viscosity(Pas)
Air 10-5
Acetone(C3H6O) 10-4
Water(H2O) 10-3
Olive oil 10-1
Glycerol(C3H6O3) 10+0
 26

Molten Polymers 10+3

Bitumen 10+8

In a non-Newtonian fluid, viscosity i.e. resistance to gradual deformation is

dependent on the shear rate to which the fluid is subjected to. Few examples of
non-Newtonian fluids are ketchup, toothpaste, blood etc.

Flow curves which indicate the relationship between shear rate and shear stress
in a material are used to fine the yield stress and plastic viscosity of the material.
Based on the shape of the curve, they can be classified into the following: (Figure
11)

1.1.1. Newtonian behavior: Viscosity remains constant. Shear stress is

dependent on the shear rate that is applied.
1.1.2. Shear thinning behavior: Viscosity decreases as shear rate is increased.
The fluids that exhibit this type of behavior are called shear thinning fluids
or pseudo plastic fluids. Blood is a good example for this behavior
1.1.3. Shear thickening behavior: Viscosity increases as shear rate is increased.
The fluids that exhibit this type of behavior are called Shear thickening
fluids or dilatant fluids. Corn starch dissolved in water is a good example
for this behavior
1.1.4. Bingham Plastic: Viscosity appears to be infinite until a certain shear stress
is achieved. Once this value called yield stress is achieved, they began to
exhibit a linear relationship between shear stress and shear rate. These
fluids are called Bingham Plastic fluids. Clay suspensions, toothpaste,
chocolate etc. are good example for this behavior
 27

Bingham
Pseudoplastic

Shear Stress (τ)

[Pa]

Shear rate (γ) [s-1]

Figure 11. Classification of fluids with τ as a function of γ (Rao A, 2010)

Based on the way viscosity is affected with the passage of time, two
classifications can be made.

1.1.5. Thixotropic behavior: It is a time dependent shear thinning property. These

fluids take some time to attain equilibrium viscosity when subjected to a
change in shear rate. Yoghurt is one such example of this behavior
1.1.6. Rheopectic or anti-thixotropic behavior: It is a time dependent shear
thickening property. In these fluids, apparent viscosity increases with the
duration of the stress. Examples of such a fluid are printer ink. (Rao A,
2010)
1.2. MODELS TO DESCRIBE THE FLOW BEHAVIOR

Some of the fundamental models that are generally used to describe the flow
behavior of materials are stated here. These models hold good only for a small
shear range and thus cannot be used for large shear ranges.

1.2.1. Newtonian model:

 28

A simple relationship exists between shear rate and shear stress. Here, viscosity
of the material remains constant and does not change with the alteration of shear
rate. Newtonian liquids such as water, oil etc. satisfy this model
𝝉 = µ𝜸
1.2.2. Power law:
This equation presented here is used to describe the behavior of non-Newtonian
materials.
𝝉 = µ𝜸𝒏

Here, n is referred to as the power law index of the material.

If the viscosity decreases with the increase in shear rate i.e. for a shear thinning
material, the value of n will be less than one

If the viscosity increases with the increase in shear rate i.e. for a shear thickening
material, the value of n will be more than one

1.2.3. Bingham model:

The materials that require a minimum shear stress so as to start deformation and
then behave as a Newtonian fluid are best described by this model.
𝜏 = 𝜏˳ + µ𝛾
This minimum shear stress required for the material to flow is generally called as
yield stress or Bingham yield stress.
1.2.4. Herschel Bulkley Model:
This model constitutes elements from all the above three models.
𝜏 = 𝜏˳ + 𝑘𝛾 𝑛
For a pure Newtonian material, limiting stress = 0 and n = 1
For a power law fluid, limiting stress = 0 and n = power law index
For a Bingham fluid, limiting stress = yield stress and n = 1 (Instruments, B. 1999)

It is also defined as the science of deformation and flow by (Barnes H A, et al.

1989). (Tattersall G H, 1991) reported that the flow properties of concrete can be
 29

represented by Bingham model, which has two parameters i.e. yield stress and
plastic viscosity. These parameters are evaluated by conducting experiments
using a viscometer but not based on the physical properties of concrete. A study
done by (Grzeszczyk, S., & Lipowski, G, 1997) on the rheological behavior of
cement paste when high calcium fly ash is used as a SCM revealed that with the
increase in fly ash content, yield stress and plastic viscosity of the paste increases.
It is also observed that the fine fraction content was a better parameter than the
specific surface area to describe the fluidity of the paste. Their study was a clear
indication that the chemical composition of the materials used will be a parameter
to assess the rheological behavior of the cement paste. (Yahia et al. 1999)
reported that use of silica fume along with superplasticizer has reduced the
viscosity of paste by 30% compared to paste containing cement and
superplasticizer only. (Kurita et al. 1999) observed that use of fly ash improves
rheological behavior and reduces the possibility of cracking of concrete because
less amount of heat is generated during the process of hydration. For SCC, which
has the ability to flow into every corner of framework, to avoid segregation, it
should possess low yield stress to achieve better flow ability and sufficient plastic
viscosity to maintain its stability. (Billberg 1999) has found that using viscosity
modifying agents or increasing the solid fraction of the cement paste, viscosity of
the paste can be increased which will satisfy the requirements of SCC mix.
(Nehdi.M and M.A.Rahman, 2004) in their paper highlighted that it is too
difficult to compute shear stress at zero shear rate i.e. yield stress. Thus to
overcome these problems, rheological models serves as a good statistical
approximation. (Shienn, 2007) has concluded that the rheology of cement paste
is affected by the properties of cement, SCMs, chemical admixtures, water
content in the paste and on the reaction between cement and admixtures used.
(Mukhopadhyay and Jang, 2009) focused on the importance of proper
compatibility between the cement and chemical admixtures used in the mix,
which is a function of rheology of the cement paste to some extent. (Cordeiro G.
 30

C. and de Alvarenga 2016) conducted studies to observe the rheological behavior

of cement paste when rice husk ash (RHA) is used as a SCM. It is observed that
with the addition of RHA, there was an improvement in the rheological and
mechanical properties of concrete. In the same study, mechanical and rheological
properties of concrete are also studied when a percentage of natural sand (fine
aggregates) is replaced by crushed fine aggregates (CGA). It is observed that the
compressive strength of the concrete mix has seen a significant increase with 50%
replacement of CGA. Also, with the increase in CGA content there is a reduction
in segregation of the material.

1.3. MEASURING PLASTIC VISCOSITY FOR VARIOUS CEMENT

PASTE COMPOSITIONS
Rheological studies are done to measure the yield stress and most importantly
plastic viscosity of cement paste with different compositions.
In the initial part of the study, a water to binder ratio varying from 0.35 to 0.55 is
chosen with superplasticizer to cementitious ratio of 0.0075 to study the influence
of water to binder ratio on the yield stress and plastic viscosity of the cement
pastes. The compositions are given in Table 1.2.

Table 1.2 Mix combinations

Superplasticizer to
Mix combination Water to binder ratio
cementitious ratio
0.35, 0.4, 0.45, 0.5 &
100% OPC 0.55

75% OPC+25% 0.35, 0.4, 0.45, 0.5 &

0.55 0.0075
GGBS

80% OPC+20% Fly 0.35, 0.4, 0.45, 0.5 &

Ash 0.55
 31

50% OPC+25% Fly 0.35, 0.4, 0.45, 0.5 &

Ash+25% GGBS 0.55

In continuation to the initial study, four different cases are chosen with varying
water to binder ratio and superplasticizer to cementitious ratio as given below.
Case I: Water to binder ratio-0.57 and Superplasticizer to cementitious ratio-
0.0065
Case II: Water to binder ratio-0.57 and Superplasticizer to cementitious ratio-
0.0075
Case III: Water to binder ratio-0.5 and Superplasticizer to cementitious ratio-0.01
Case IV: Water to binder ratio-0.5 and Superplasticizer to cementitious ratio-
0.0125

The following Step by step procedure is adopted for conducting the experiment:

1. Required weights of cement and other supplementary cementitious materials,

are taken into a glass beaker

2. Based on the water to binder content ratio, required amount of water is added

using a syringe
3. Based on the superplasticizer to cementitious content ratio, required amount

of superplasticizer is added using a syringe

4. The contents of the beaker are stirred thoroughly and allowed to settle for

about 10 minutes
5. The sample paste is subjected to viscosity test in a Brookfield Viscometer

DV3T at a room temperature for a chosen shear rate. The spindle used for
testing is SC4-21
6. A graph is then plotted between shear stress and shear rate for different shear

rates at different times of the test. From the graph, yield stress and plastic
viscosity values are obtained.
 32

Brookfield rotational viscometer DV3T as shown in Figure 1 is used for the entire
testing sequence. The viscometer was equipped with a Vane geometry spindle of
2.53 mm height and 0.64 radius. The rheological properties of cement pastes are
measured with varying shear rates for different intervals of time at a constant
room temperature.

Figure 12. Brookfield Viscometer DV3T

For the initial study with varying water to binder ratios, the plastic viscosity of
the cement pastes as shown in Table 1.3 to Table 1.6 followed a decreasing trend
with the increase in w/b ratio due to the reaction between cementitious molecules
and water. The full hydration process will happen at a w/b of 0.38 to 0.42 for a
pure OPC paste which directly indicates that there will be a sudden decrease in
plastic viscosity till 0.4 followed by a gradual decrease. But in the case of paste
with 25% GGBS as a replacement, due to the flaky nature of particles, paste with
GGBS requires more water resulting in a gradual decrease of plastic viscosity.
For 20% fly ash as replacement, reaction of water with fly ash requires more time
resulting in a gradual decrease of plastic viscosity. For ternary mix as, the total
contribution of fly ash and GGBS are 25% and with the increase in water to
cement ratio a decreasing trend of plastic viscosity is observed.
 33

Table 1.3: Bingham parameters for OPC paste

W/b 0.35 0.4 0.45 0.5 0.55

Plastic 0.39 0.35 0.32 0.30 0.22
Viscosity

Table 1.4: Bingham parameters for 75% OPC + 25% GGBS paste

W/b 0.35 0.4 0.45 0.5 0.55

Plastic 0.41 0.37 0.33 0.32 0.23
Viscosity

Table 1.5: Bingham parameters for 80% OPC + 20% FA paste

W/b 0.35 0.4 0.45 0.5 0.55

Plastic Viscosity 0.34 0.30 0.28 0.24 0.19

Table 1.6: Bingham parameters for ternary 50% OPC + 25% FA + 25%
GGBS paste

W/b 0.35 0.4 0.45 0.5 0.55

Plastic Viscosity 0.43 0.39 0.35 0.33 0.24

Similarly, the values of plastic viscosity for Cases I to IV are shown in Table 1.7
to Table 1.10

Table 1.7 Measured plastic viscosity of cement pastes for Case I

Cementitious material combinations Paste plastic viscosity (Pa-s)

100 % - OPC 0.17

75 % - OPC + 25 % - GGBS 0.18

80 % - OPC + 20 % - Fly ash 0.16

 34

50 % - OPC + 25 % - GGBS + 25 % - 0.20

Fly ash

Table 1.8 Measured plastic viscosity of cement pastes for Case II

Cementitious material combinations Paste plastic viscosity (Pa-s)

100 % - OPC 0.165

75 % - OPC + 25 % - GGBS 0.174

80 % - OPC + 20 % - Fly ash 0.15

50 % - OPC + 25 % - GGBS + 25 % - 0.19

Fly ash

Table 1.9 Measured plastic viscosity of cement pastes for Case III

Cementitious material combinations Paste plastic viscosity (Pa-s)

100 % - OPC 0.25

75 % - OPC + 25 % - GGBS 0.26

80 % - OPC + 20 % - Fly ash 0.235

50 % - OPC + 25 % - GGBS + 25 % - 0.275

Fly ash
 35

Table 1.10 Measured plastic viscosity of cement pastes for Case IV

Cementitious material combinations Paste plastic viscosity (Pa-s)

100 % - OPC 0.24

75 % - OPC + 25 % - GGBS 0.25

80 % - OPC + 20 % - Fly ash 0.22

50 % - OPC + 25 % - GGBS + 25 % - 0.26

Fly ash

From Table 1.7 to Table 1.10, it is inferred that with the increase in water to
cement ratio and superplasticizer dosage, the plastic viscosity decreases.

4.4. ROLE OF WATER TO CEMENT RATIO ON COMPRESSIVE

STRENGTH OF CONCRETE

According to Abram’s law of water to cement ratio, the compressive strength of

concrete depends on the water-cement ratio adopted and the strength is inversely
proportional to water to cement ratio (in terms of mass). So, based on this law it
is clear that the strength of UHPC also depends mainly on the water to binder
material. In order to establish a relation between the strength of concrete and the
water to cement ratio, a set of values for water to cement ratio and the resulting
28 day-compressive strength using various mineral admixtures are adopted from
the various literatures. Using this data, regression analysis is performed to obtain
a best fit curve as shown in Figure 1 which is Abram’s type power curve with
R2 = 0.9456. The expression for compressive strength in terms of w/b ratio is
given by

𝑓𝑐𝑢 = 225.69𝑒 −3.827(𝑤/𝑐𝑚)

 36

Where,

f cu is the 28-day cube compressive strength of concrete in MPa.

w
represents the adopted water to cement ratio of the concrete mixture.
cm

1.5. DEVELOPMENT OF PLASTIC VISCOSITY BASED MIX DESIGN

FOR SCC

UHPC is generally modelled as rigid solid particles (aggregates) suspended in

viscous liquid medium (cement paste). Based on this principle many theories
have been proposed:

1.5.1. For low concentrations

At low concentrations, the particles are assumed to be far apart from each other
and the interactions between them can be neglected. Einstein (Struble and Sun,
1994) proposed an equation to find the relative viscosity of one phase with respect
to other which leads to increase in plastic viscosity of the suspension of particles.

r  1  [ ] (2)

Where,

[ ] represents the intrinsic viscosity which is the viscosity of individual particles.

 represents the volume fraction of particles.

The intrinsic viscosity of particles can be calculated from relative viscosity of the
suspension as
 37

Characteristic compressive strength (MPa) Characteristic compressive strength vs. water/cement ratio

300

250
y = 225.69e-3.827x
200 R² = 0.9456
150

100

50

0
0 0.05 0.1 0.15 0.2 0.25
Water/cement ratio

Figure 13. Regression curve for characteristic compressive strength water

to cement ratio

r  1
[ ]  lim (3)
 0 

Shenoy (2013) proposed an equation based on Einstein’s equation (2) and named
it as modified Einstein’s equation in the binomial form as

r  (1  [ ] ) 1 (4)

(Utracki and Wilkie, 2002) modeled the solid particles as rigid spheres having
radius ‘a’ and the particles are enclosed in a closed domain of radius b. Generally
[ ] = 2.5, when the particles are rigid and have a hexagonal random packing.
(Neglecting the particle interaction effects) It is also assumed that the movement
of the particles is very slow and their kinetic energy can be neglected. Simplifying
the above equation (3) using these assumptions leads to
 38

 5
2 
 
25   21  3 625   
 
r  1  2.5 1         ... (5)
 32  m  64  m  128  m  
 

Where,

m represents the maximum packing fraction of the particle.

Shenoy (2013) based on the above equation (5) derived an expression for dilute
suspensions as

 25 
r  1  2.5 1   (6)
 4a13 

For low concentrations of particles generally, a1 = 1.111

1.5.2. For high concentrations:

As the concentration of suspension increases, the viscosity of the suspension not

only depends on the concentration of particles but also on the size and shape i.e.
packing of the particles. Based on this concept, the plastic viscosity of the
suspension is given as

r  1     B 2  C 3  ... (7)

Where, the values of these constants B & C depend on the size, shape and
distribution of particles in the suspension i.e. on the nature of the suspension.
These values are available in literature (Shenoy, 2013; Batchelor, 1977; Vand,
1948; Saito, 1950).

Thomas Kumar (1999) proposed an exponential replacement for the constant

term C in equation (7) as

r  1  2.5  10.05 2  0.00273e16.6 (8)

 39

This expression is derived from experimental data by performing a regression

analysis and finding out the best fit curve. The range of values of  suggested for
this expression are 0.15 to 0.6.

From the equations (7, and 8), it is clear that there is a limitation in using the
constants B and C as they are restricted to a particular range of volume fraction
of particles. So, in view of these restrictions Krieger and Dougherty (1959) came
up with expression to calculate the viscosity of suspensions based on their
maximum packing fraction concept. It is assumed that in the maximum packing
fraction state, the particles are bound to have minimum voids in them and
maximum possible viscosity. These values are shown in (Table 1.11)

Table 1.11 Packing fractions for different types of packing of particles

Packing type Maximum packing fraction

Cubic 0.524

Random Hexagonal 0.637

Hexagonal 0.74

From the above values, they came up with a generalized equation for calculation
of plastic viscosity based on the maximum packing fraction of particles and the
viscosity of individual particles expressed as

[ ]m
  
r  1   (9)
 m 

Where, m is the maximum packing fraction of particles and is dependent on the

distribution of particles and [ ] is the intrinsic viscosity of particles in the
suspension. These values also depend on the shear rate applied on the suspension.
 40

The value of [ ] decreases with increase in shear rate and m increases with
increase in shear rate indicating that both are inversely proportional to each other.
But practically, the products of both the parameters remain constant and the value
is approximately equal to 1.9.

Frankel et al (1967) proposed an expression for high concentration suspensions

when the volume fraction of particles almost equal to the maximum volume
packing fraction of particles i.e. when   m as

 1

9   / m  3 
r  (10)
8  1   /  13 
  m 

Chong et al (1971) developed an expression for a complete range of volume

fraction of particles i.e. from  0 to   m by applying the Brownian random
distribution of particles based on the regression analysis of experimental data. It
is given as

2
 [ ]m   / m   
r  1     (11)

 2  1   / m  

This Brownian distribution contributes to a slight increase in viscosity of the


suspension and the range of values of concentrations considered for is from 0
m

to 0.7.

1.5.3. Methodology for proportioning UHPC mixes

UHPC mix is considered as a suspension of particles in which the aggregates are

suspended in viscous fluid (cement paste). Firstly, to the suspension of cement
paste, fine aggregate is added and then to this mixture Filler material is added and
 41

then the process continues. Now, following this procedure, a standard expression
is developed as

mix   paste * f1 (1 )* f 2 (2 )....* f n (n ) (12)

Where,

 paste represents the plastic viscosity of cement paste.

1 , 2 .........n , are the volume fractions of the particles (phases) to be added to the
suspension of cement paste.

Applying the Krieger and Dougherty equation (8) to calculate the contribution of
individual particle to the plastic viscosity of the mix is given as

[ ]m
  
fi (i )  1  i  (13)
 m 

From the equations (12, 13), the plastic viscosity of the UHPC mix is given as

𝜙𝐹𝐴 𝜙𝐹
𝜂𝑚𝑖𝑥 = 𝜂𝑝𝑎𝑠𝑡𝑒 ∗ ((1 − ⁄𝜙𝑚) ∗ (1 − ⁄𝜙𝑚)) (14)

By adding fine aggregate to the suspension of cement paste, first the packing
fraction of the suspension or the mixture is assumed to be random hexagonal
packing and then after the addition of filler material (e.g. fly ash) the packing
becomes dense and then the packing is assumed to be hexagonal packing. Based
on these assumptions, the plastic viscosity of UHPC mix is given as

𝜙𝐹𝐴⁄ 𝜙𝐹
𝜂𝑚𝑖𝑥 = 𝜂𝑝𝑎𝑠𝑡𝑒 ∗ ((1 − 0.63 ) ∗ (1 − ⁄0.524)) (15)

Based on the literature available as well as the gaps identified, the following cases
are chosen to study the effect of plastic viscosity on the workability and strength
 42

characteristics of Ultra – High Performance Concrete using Crushed Rock Fines

(CRF) as a fine aggregate with varying water to binder ratio.

Case I: A trial plastic viscosity of 9 Pa s is chosen for water to binder ratio of 0.57
(adopted from Abo Daheer, 2016 without filler and with CRF as a fine aggregate).

Case II: Influence of plastic viscosities 7 Pa s and 11 Pa s on the fresh and

hardened properties for water to binder ratio of 0.57.

Case III: Influence of plastic viscosities 9 Pa s on the fresh and hardened

properties for water to binder ratio of 0.5 with river sand as fine aggregate.

Case IV: Influence of plastic viscosities 9 Pa s and 13 Pa s on the fresh and

hardened properties for water to binder ratio of 0.5.

Specific gravities of materials used for proportioning the materials required for
SCC mix wit

Another input required for the proposed mix design methodology is the plastic
viscosity of the cement paste. So, for the given grade of UHPC to be
proportioned, the water to cement ratio adopted is 0.5 (as per equation. (1)).
Corresponding to 0.5 water to cement ratio, the plastic viscosity of cement pastes
for different cementitious materials are measured using Brookfield viscometer as
explained in the section 4.3. The plastic viscosities of pastes with different
cementitious materials are given in Table 1.7 for cases I and II and Table 1.10 for
case III.

Based on these plastic viscosities and considering the target plastic viscosity of
the UHPC to be proportioned as 7, 9 and 11 Pa-s for water to binder ratio of 0.57
and 9 and 13 for water to binder ratio of 0.5. The proportions for different
materials used to prepare an UHPC mix with a grade equal to M 40 are calculated
with the help of a computer program. Different values of parameters t 1 & t2 are
 43

considered as an input starting from 0 to the maximum value with a condition that
the volume fractions of fine aggregate and filler do not exceed 1. The output of
this code generated several combinations of UHPC mix proportions and some of
the best mixes from many combinations generated are chosen based on the
satisfactory requirements of EFNARC guidelines.

The following step-by-step process is followed for the mix design of UHPC:
1. First a trial plastic viscosity value is chosen considering that slump cone
T50 increases with the increase in plastic viscosity.
2. Water to cement ratio is calculated using equation given by
𝑓𝑐𝑢 = 225.69𝑒 −3.827(𝑤/𝑐𝑚)
3. Choose the water content following EFNARC guidelines in the range of
150 to 210 kg/m3.
4. The percentage replacement of cement with GGBS and Fly ash is assumed
to be 25%
5. Plastic viscosity of the paste (ηpaste) for 75% OPC+25% GGBS, 80% OPC
+ 20% Fly ash and 50% OPC+25% GGBS+25% Fly ash are estimated
using Brookfield viscometer.
6. Mass of fine aggregate and filler are calculated based on their volume fractions
using Eq. (6) and (7). Volume fractions of fine aggregate and filler are
estimated using a randomization computer code such that the amount of fine
and coarse aggregate does not exceed the limits as per EFNARC guidelines
(The European Guidelines for Self-Compacting Concrete – EFNARC, 2005).
FA
 FA
FA  (6)
 cem w SP  FA
    0.02  
 cem  w  SP   FA
 44

𝐹
𝜌𝐹
𝜙𝐹 = 𝑤 𝐹𝐴 𝑐𝑒𝑚 𝑆𝑃 𝐹 (7)
( + + + +0.02)+
𝜌𝑤 𝜌𝐹𝐴 𝜌𝑐𝑒𝑚 𝜌𝑆𝑃 𝜌𝐹

7. The total volume of the mix should be equal to 1 m3. If not, suitable corrections
are to be applied for the raw materials to attain a total volume of 1 m3.
8. The measured plastic viscosity of the mix is compared with the assumed
plastic viscosity (step 1). The assumed value of plastic viscosity of mix is in
good agreement with the estimated value if the difference between the two
is within ±5%. If not, choose different volume fractions of solid phase
ingredients i.e. fine and coarse aggregates and repeat the steps 7 and 8.
Where,

fcu = Characteristic compressive strength of concrete (M Pa)

ηmix = Plastic viscosity of mix

ηpaste = Platic viscosity of paste (assumed between 3 − 15 Pa − s)

ΦFA = Mass of fine aggregate
ΦF = Mass of filler material
ρFA = Mass density of fine aggregate
ρF = Mass density of filler
ρw = Mass density of water
ρcem = Mass density of cement
ρSP = Mass density of superplasticizer
w, SP, FA, F, cem = volume fractions of water, superplasticizer, fine
aggregate, filler and cement respectively
 45

BIBLIOGRAPHY AND REFERENCES

1. Material Property Characteristics of Ultra High Performance
Concrete (2006), Publication No. FHWA — HRT— 06 -103, US
Department of Transportation.
2. Caulberg G., Pierard J., Remy O.,(2008),Ultra High Performance
Concrete : Mix Design and Practical Applications, Wlravan and
Stoerhurst (eds).
3. Julie Anne Willey (2013), Use of ultra high performance concrete
to mitigate impact and explosive threats, Curtis Laws Wilson Library,
Missouri University of Science and Technology.
4. Yu, R. Spiesz, P., & Brouwers, H. J. H. (2014). Mix design and
properties assessment of ultra-high performance fibre reinforced concrete
(UHPFRC). Cement and concrete research, 56, 29-39.
5. Yuliarti Kusumawardaningsih, Ekkehard Fehling, Mohammed
Ismail, Attitou Amen Mohamed Aboubakr (2015), Tensile strength
behavior of UHPC and UHPFRC, The 5th International Conference of
Euro Asia Civil Engineering Forum (EACEF-5), Procedia Engineering
125 (2015) 1081 — 1086.
6. Mohamadreza Shafieifar, , Mahsa Farzad, Atorod Azizinamini
(2017), Experimental and numerical study on mechanical properties of
UltraHigh Performance Concrete (UHPC), Construction and Building
Materials, Construction and Building Materials 156 (2017402-411.
7. Jun Li a, Chengqing Wua, Hong Hao , (2015), An experimental
and numerical study of reinforced ultra-high performance concrete slabs
under blast loads, Materials & Design 82 (2015) 64–76
8. Christian Melançon, Sarah De Carufel and Hassan Aoude,
(2016),Blast Behavior of One-Way Panel Components Constructed with
UHPC, First International Interactive Symposium on UHPC – 2016
9. Petr Máca, Radoslav Sovják, Petr Konvalinka (2014), Mix design
of UHPFRC and its response to projectile impact , International Journal
of Impact Engineering 63 (2014) 158e163
10. Jun Lia Chengqing Wua, Hong Haoc and Yu Sua (2015),
Investigation of Ultra-high Performance Concrete Under Static and Blast
Loads,International Journal of Protective Structures – Volume 6 ·
Number 2 · 2015
 46

11. Shih-Ho Chao, Venkatesh Kaka, Guillermo Palacios, Alireza

Nojavan, et. al(2016), Seismic Behavior of Ultra-High-Performance
Fiber-Reinforced Concrete Moment Frame Members,First International
Interactive Symposium on UHPC – 2016