ADVANCES IN CONCRETE TECHNOLOGY IN THE MIDDLE EAST:

SELF-CONSOLIDATING CONCRETE

Kamal H. Khayat
Editor

PROCEEDINGS OF

THE SECOND INTERNATIONAL CONFERENCE ON

ADVANCES IN CONCRETE TECHNOLOGY IN THE MIDDLE EAST:

SELF-CONSOLIDATING CONCRETE

December 8-10, 2009

Rotana Beach Hotel

Abu Dhabi, UAE

Organizing Committee
Chairman
Kamal H. Khayat
Universit de Sherbrooke, Canada
Secretary
Rabih Fakih
Grey Matters, UAE
Treasurer
Fouad Yazbeck
Readymix Abu Dhabi, UAE
Advisory Committee
Osama Abdulbari
Sodamco, UAE
James Aldred
GHD, UAE
Hussein Basma
PCFC-Civil Engineering Department, UAE
Tarek Fransawi
Sodamco, UAE
Abdel Kader Trabulsi
Al Falah Readymix, UAE

Sponsored by:

Organized by:

ACI International

Grey Matters Group of Companies

ii

PREFACE

The Committee for the Organization of ACI International Conferences is very

pleased to sponsor the Second International Conference on Advances in Concrete
Technology in the Middle East: Self-Consolidating Concrete, which will be held
December 8-10, 2009 in Abu Dhabi, United Arab Emirates.
This conference dedicated to self-consolidating concrete (SCC) in Abu Dhabi is a
follow up of the CANMET/ACI International Conference that took place in Dubai
in November 2008 highlighting recent advances in concrete technology in the
Middle East.
The papers assembled here, submitted by industrial and academic researchers
from acoss the world cover a wide range of topics highlighted at the conference,
among them material selection and mix design, workability and rheology, test
methods and processing, production and quality control, engineering and
durability, structural performance and design, fibre-reinforced SCC, experience
and case studies, specifications, and economic and environmental benefits.
The organizing committee would like to thank all of the authors, as well as the
session chairpersons for their participation. Special thanks are also extended to all
the sponsors for their generous support.
It is our hope that this compendium may become a useful resource for the
concrete construction industry in the Middle East and help promote a better
understanding of the opportunities and challenges presented within the scope of
this conference.

Kamal H. Khayat
Editor and Conference Chair
Universit de Sherbrooke, Canada
November 2009

iii

CONTENTS

Organizing Committee ......................................................................................... iv

Preface................................................................................................................... v
Past, Present, and Future of SCC
Shigeyoshi Nagataki, Toru Kawai, and Hiromi Fujiwara .................................... 1
Guidelines for Design and Construction of Self-Consolidating Concrete for
Precast, Prestressed Concrete Members
Kamal H. Khayat, Soo-Duck Hwang, Gillaume Lemieux, and Wu Jian Long ... 27
Friendlier Self-Compacting Concrete through Smart Dynamic Construction
Bryan Barragan, Joana Roncero, Roberta Magarotto, S. Moro, and
R. Khurana ......................................................................................................... 73
Performance-Based Specifications for Self-Compacting Concretes in the Gulf
Region Determined
Jon C. Knights and Don Wimpenny ................................................................... 97
Future Trends in the Processing of Ready-Mixed High-Strength Self-Compacting
Concrete
Harald Beitzel ................................................................................................... 123
Water in Self-Consolidating Concrete Production Its Importance, How to
Accurately Measure it and the Benefits of Automated Water Control
David Serra ....................................................................................................... 133
Pumping of Self-Consolidating Concrete under Extreme Conditions
Harald Beitzel ................................................................................................... 145
Pumpability Assessment of C90 SCC
Said Ahmeed Jodeh and Gabriel E. Nassar ...................................................... 155
Can Ordinary Vibrated Concrete and Self-Consolidating Concrete be Treated
in the Same Way Concerning Pumpability?
Dimitri Feys, Geert De Schutter, Ronny Verhoeven, and Kamal H. Khayat .... 177

iv

Self-Compacting Concrete: Properties, Development and Code Recommendations

Joost Walraven.................................................................................................. 201
Practical and Successful SCC Application in the Precast Concrete Industry
F. Surico, A. Gasperi, R. Marino, and J. Perazzoli .......................................... 229
Evaluation of SCC Formwork Pressure
Kamal H. Khayat and Ahmed F. Omran........................................................... 243
Al Turki Business Park Self-Consolidating Concrete: From Head to Toe
Redwan Amin Hameed and Narasimhulu Gary ............................................... 261
Laboratory Testing and Field Monitoring of Self-Consolidating Concrete (SCC)
Drilled Shafts
Hani Nassif, Nakin Suksawang, Kagan Aktas, and Husam Najm ................... 279
In-Situ properties of Self-Consolidating Concrete in a Thick Concrete Raft in
Dubai
James Aldred ..................................................................................................... 297
Environmentally Sustainable Self-Consolidating Concrete A Case Study
Abu Saleh, Mamo Kebede, and Nagham Yacoub ............................................. 307
Rheology of Self-Compacting Concrete and Eco-SCC
Olafur Wallevik, Florian V. Mueller, and Bjrn Hjartarson ............................ 339

PAST, PRESENT, AND FUTURE OF SCC

Shigeyoshi Nagataki, Toru Kawai, and Hiromi Fujiwara

Synopsis: In Japan, self-compacting concrete/self-consolidating concrete (SCC)

had been receiving particular scrutiny since 1980s, and in 1990s SCC was put to
practical use and increased its results.
Under this situation, the Japan Society of Civil Engineers (JSCE) set up
Research Subcommittee on Self-Compacting Concrete in April 1994 and
published State of the Art-Report on Self-Compacting Concrete in December
1996. Based on the report, JSCE published Recommendation for practical
application of self-compacting concrete in July 1998. Classified types are
explained and self-compactability is clearly defined.
On the basis of the recommendation, SCC has been frequently applied mainly for
the work where vibrating compaction is difficult. Furthermore, SCC has been
made more useful and more applicable through the technology development of
materials, construction method and specific design of mixture proportions.
However the share of SCC has not always been more than 3% of total produced
amount of concrete. On the other hand, some countries in Europe have
considerably higher shares of SCC than Japan. The reason must be investigated in
detail.
Recently in Japan, some researchers have been continuing their study to establish
the prediction method of self-compactability. By establishing this method, it is
considered the reliability of self-compactability of SCC will be much improved.

Keywords: plastic viscosity, pressure loss, reactive powder concrete, selfcompactability, self-compacting concrete, self-consolidating concrete,
superplasticizer, viscosity- agent, yield value

ACI honorary member Shigeyoshi Nagataki is a Professor of Aichi Institute of

Technology, Emeritus Professor of Tokyo Institute of Technology. He was
President of Japan Concrete Institute from May 2004 to 2006. He received his
Doctor of Engineering degree (1966) from the University of Tokyo, Japan. He is
also honorary member of JCI, JSCE and JSMS.
Toru Kawai is a senior research engineer of Shimizu Corporation. His research
interests include concrete materials and concrete construction methods. He
received his Doctor of Engineering degree (1996) from Tokyo institute of
Technology. He is member of JCI and JSCE.
Hiromi Fujiwara is a professor of Utsunomiya University. His research interests
include fresh properties of concrete and developing new cementitious materials.
He received his Doctor of Engineering degree (1996) from Tokyo Institute of
Technology. He is member of JCI and JSCE.

INTRODUCTION
Recent years have seen large number of concrete structures with increasing height
and overall size in Japan, resulting in an increasing demand for a wider diversity
of types of highperformance concrete. In efforts to meet these needs, highstrength, high-durability and high-fluidity concretes, among others, have been
studied.
In Japan, self-compacting concrete/self-consolidating concrete (referred to as
SCC) had been receiving particular scrutiny since 1980s [1], and SCC was put
to practical use in 1990s. With SCC, the use of a vibrator for compacting at site
is not necessary. The development of SCC was initiated from an increasing
demand for improving the reliability of concreting work in Japan. The
background of SCC development in Japan can be seen from the following
information.
The topography of Japan comprises many precipitous mountains and rugged
valleys; thus it is often the case that concrete placement must be carried out under
extremely difficult circumstances. Further, the increasingly complex shape of
concrete structure is making it more difficult to use a vibrator, while the more
densely arranged reinforcing bars resulting from the increasing high-rise concrete
buildings make consolidation more difficult to carry out.

In addition, the following reasons have been cited for the increasing demand for
SCC:
1) There have not been enough workers to carry out compacting work at
construction sites.
2) Vibrating compaction of concrete is extremely noisy and deleterious to the
health of workers, as well as an annoyance to people in the surrounding
neighborhood.
3) Vibrating compaction of concrete is costly and time-consuming.
SCC not only alleviates these problems, but also improves the efficiency of
construction. In other words, the use of SCC requires only the setting up of forms
and placement of concrete by pumping, and eliminates the previously required
hard, labor-intensive work of compacting and the related setting up of scaffolding.
For these reasons, it had been expected that the use of SCC would become
widespread.
Firstly, this paper introduces classification of SCC and verification of selfcompactability prescribed in Recommendation for practical application of selfcompacting concrete established by Japan Society of Civil Engineers (JSCE).
Secondly, this describes technology transition of materials and mixture
proportions, and shows typical examples of SCC applications. Finally, future
research on the estimation of self-compactability of SCC is described.
PAST AND PRESENT OF SCC
Recommendation for self-compacting concrete by JSCE
Details to publish the recommendation In 1990s, SCC increased its practical
results gradually in Japan, therefore, the Concrete Committee of the Japan Society
of Civil Engineers (JSCE) set up Research Subcommittee on Self-Compacting
Concrete in April 1994 and published State of the Art-Report on SelfCompacting Concrete [2] in December 1996. The subcommittee published
Recommendation for practical application of self-compacting concrete based on
the report and accumulated research results in July 1998 [3]. The article includes a
recommendation, a manual for mixture proportioning, a manual for production
and placement, standard test methods and collected data for practical application.
Classification of SCC
In this recommendation, SCC is classified into three types, i.e., Powder-type,
Viscosity agent-type and Combination-type.

Powder-type: SCC proportioned to provide the required self-compactability not

by using a viscosity agent but primarily by reducing the water-powder ratio (in
effect increasing the powder content) to impart adequate segregation resistance
and using an air-entraining and high-range water-reducing admixture to impart
high deformability.
Viscosity agent-type: SCC proportioned to provide the required selfcompactability by the use of a viscosity agent to impart segregation resistance and
air-entraining and high-range water-reducing admixture to impart high
deformability.
Combination-type: SCC proportioned to provide the required self-compactability
primarily by reducing the water-powder ratio (in effect increasing the powder
content) to impart adequate segregation resistance and using an air-entraining and
high-range water-reducing admixture to impart high deformability. A viscosity
agent is also added to reduce the quality fluctuation of fresh concrete, so as to
facilitate the quality control of concrete.
In this recommendation, it is recommended that a suitable type should be selected
from these three types of SCC in consideration of the type of the structures,
structural conditions, types of materials available and limitations at concrete
production plants.
This classification is applied to Guidelines for Design and Construction of SelfConsolidating Concrete for Precast, Prestressed Concrete Menbers [4].
Classification of self-compactability of SCC
Furthermore, in this recommendation, according to the geometry and
reinforcement conditions of the structure to which SCC is to be applied, the
required self-compactability of SCC is classified into three levels, i.e., Rank 1,
Rank 2 and Rank 3 as follows.
Rank 1: Self-compactability into members or portions having complicated shapes
and/or small cross-sectional areas with a minimum steel clearance in the range of
35 to 60mm.
Rank 2: Self-compactability into reinforced concrete structures or members with a
minimum steel clearance in the range of 60 to 200mm.
Rank 3: Self-compactability into members or portions having large cross sectional
areas and a small amount of reinforcement with a minimum steel clearance of
more than 200mm.
It is recommended that the self-compactability level of general reinforced
concrete structures or members should be Rank 2.

Ranks 1 to 3 are set as the levels of self-compactability of concrete. These ranks

are set based on the dimensions and reinforcement conditions of the structures or
members.
Rank 2 is defined as the capability of concrete with which it can be selfcompacted through minimum bar clearance of 60 to 200mm. This normally
corresponds to a steel content of 350 to 100kg/m3. Rank 1 is a level on which
concrete meets severer requirements than Rank 2, and it can be self-compacted
through the minimum bar clearance is less than 60mm and the steel content is
more than 350kg/m3. Rank 3 is a level on which concrete meets easier conditions
than Rank 2, demonstrating self-compactability where the minimum bar clearance
is more than 200mm and maximum steel content is less than 100kg/m3.
Verification of self-compactability
The designed SCC is required to be verified by suitable methods in regard to the
attainment of the established performance. Verification of self-compactability of
fresh concrete is important for ensuring the performance of structures. The
verification should be carried out by preparing full-scale forms of the structure or
members or model forms representing the portions where the placing is
considered most difficult, placing the proportioned concrete in them according to
the assumed construction plan and confirming the state of filling of the concrete.
Where the concrete is placed under standard construction conditions, verification
by the tests specified in JSCE Standards is permitted. These are passability tests
using a filling tester, shown in Fig.1, according to the Rank of the concrete
established in consideration of the structural conditions. When Rank 2 selfcompactability is required, the concrete is judged as satisfying the requirement, if
its filling height is not less than 300mm by a filling tester with obstacle R 2.
When Rank 1 self-compactability is required, the concrete is judged as satisfying
the requirement, if its filling height is not less than 300mm using obstacle R 1.
The methods for concrete of Rank 3 include testing by a filling tester without any
obstacle.
Methods of proportioning SCC
In this recommendation, the methods of proportioning each type of SCC are
described to satisfy the self-compactability requirement corresponding to the rank
for the structures. Especially, maximum size and unit content of coarse aggregate
are specified. For maximum size of coarse aggregate, it is described that it should
be 20mm or 25mm. And for coarse aggregate content, standard values (unit
absolute volume of coarse aggregate) are little different depending on type of
SCC and required rank of self-compactability as shown in Table 1.

Application of SCC
Technology transition of materials and mixture proportions
The most different feature between SCC and ordinary concrete is that SCC has
sufficient self-compactability in the fresh state as before mentioned. Principal
technologies which have been developed to realize the self-compactability of SCC
in terms of materials and mixture proportions are introduced here.
Superplasticizer: The most remarkable development of material for SCC is the
invention of superplasticizer [5]. Dr. Hattori developed formaldehyde condensates
of beta-naphthalene sulfonates with the primary aim of significantly reducing the
water demand of concrete to produce high-strength concrete [6]. Water reduction
of up to 30 percent was achieved with the use of this superplasticizer called
Mighty 150. This admixture was introduced into the Japanese concrete industry as
a nominal name of high-range water-reducing admixture in the early 1960s.
Since then, it has greatly contributed to production of high-strength concrete.
In Germany, Dr. Aignesberger and his colleagues developed the melamine-based
superplasticizer having nearly the same performance as the beta-naphthalenebased one [7]. These two superplasticizers, however, have one common defect
that the loss in slump is considerably large.
Therefore, these superplasticizers were mainly used for factory production where
the problem of loss in slump during transportation is negligible. Another
approach was to add superplasticizer right before placing concrete at site.
To solve this problem, new technology of slump control with a reactive
polymeric dispersant [8], was studied. As a result, superplasticizer called airentraining and high-range water-reducing admixture was developed. This
superplasticizer is mixed into concrete at the mixing plant even for in-situ casting
because the loss in slump during transportation is relatively small.
In terms of rheological aspect, by adding conventional superplasticizers, SCC
shows a small yield stress but it shows a relatively high plastic viscosity due to a
low water-cement (binder) ratio. This too much high plastic viscosity generally
makes it difficult to achieve the easy placement.
A new type of polycarboxylate-based air-entraining and high-range waterreducing admixture has been developed on the basis of steric hindrance theory in
Japan. This polycarboxylate-based superplasticizer imparts the SCC with high
deformability and adequate plastic viscosity in the period immediately after
initial mixing until placement, even when the water-binder ratio is 30% or lower.
Under the circumstances, airentraining and high-range water-reducing
admixture was specified in JIS A 6204 (Chemical admixtures for concrete) in

1995. Since its introduction, this superplasticizer has dramatically increased in

use for achieving SCC and high-strength concrete.
The share of superplasticizers used for SCC in Japan around up to 1998 is
described in Fig. 2. It is clearly understood that polycarboxylate-based one was
commonly used. In recent years (after 1998 to 2009), almost all the
superplasticizers used for SCC were polycarboxylate one or its modified ones in
Japan.
Here, Fig. 3 [9] shows the share of the superplasticizers used for placed-in-situ
concrete in each country. In hot-climate countries like Vietnam, Thailand, India,
Singapore and Malaysia, beta-naphthalene-based superplasticizer (BNS) is the
major one. The reason may be the higher performance of BNS in higher
temperature and its lower price. According to Fig. 3, the share of the
superplasticizers depends on the area of the world.
Principal materials and mixture proportions SCC contains much higher unit
content of powders compared with ordinary concrete. Here, powder means not
only cement but also mineral admixtures such as ground granulated blast-furnace
slag, flyash, limestone powder and expansive agent.
The addition of components other than cement is mainly for reducing the heat of
hydration of the cementitious materials because SCC, especially powder-type and
combination-type mixtures have high contents of powder. Common unit content
of powders in powder-type and combination-type mixtures is 500-575kg/m3,
while that of viscosity agent-type mixture is 400-500kg/m3.
Common water powder ratios of powder-type and combination-type mixtures are
30-40%, while that of viscosity agent-type mixture is 45-55%.
Application volume
The volume of SCC application in Japan since 1992 is shown in Fig. 4 [10]. In
recent years, the volume of SCC placed in situ has been exceeded by that for
factory product. The share of SCC placed in situ is approximately 0.1-0.2% of
total produced amount of ready-mixed concrete in Japan and that of factory
product is about 2-3%.
On the other hand, Europe has a different trend of share of SCC depending on the
country as shown in Table 2 [11]. Shares of SCC of three countries in Table 1 are
considerably higher than those of Japan, possibly due to the difference in cost and
required performance between Japan and those countries. However, the real
reason must be investigated in detail. Information on the application volume in
North America and Asian countries has not been gained.

Examples of SCC applications

Anchorages at Akashi Kaikyo Bridge [12] (Large volume construction): Akashi
Kaikyo Bridge is a suspension bridge described in Fig. 5. The center span length
is 1990m, which is the longest one in the world. The main towers are
approximately 300m high. In the anchorages, cable anchor frames made with steel
member to fix the cables from the bridge were installed. The anchorages were
enormous volume of massive concrete structures.
Total 290 thousand cubic meters of SCC were placed at two anchorages 1A and
4A. The anchorage was composed of reinforced concrete part and cable anchor
frame part with densely arranged by steel frames. 1900m3 of SCC was placed in
one day.
Under the condition that sufficient number of workers to carry out vibrating
compaction for large volume construction could not be prepared for the project,
SCC was adopted because it needs no worker for compaction work. The SCC was
manufactured in properly assembled mixing plant at the site. This plant is for the
exclusive use of SCC. To control the quality fluctuation of fresh concrete, fine
aggregate was dewatered through the centrifugal rotating apparatus, for ensuring
the surface water content of fine aggregate of 4.5-5.1%.
Powder type mixture proportions applied to the project are shown in Table 3. The
cement for 1A was low heat cement composed of low-heat portland cement
(25%), ground granulated blast-furnace slag (55%) and fly ash (20%). The cement
for 4A was low heat cement composed of moderate-heat portland cement (21%),
ground granulated blast-furnace slag (59%) and fly ash (20%). Unit cement
content was maintained less than 260kg/m3 to satisfy the criteria that ultimate
adiabatic temperature rise should not be larger than 25oC. 150kg/m3 of lime stone
powder was used to obtain a suitable viscosity for the placing without viscosity
agent. Maximum size of coarse aggregate of 40mm was selected to increase unit
weight of concrete larger than 2.3t/m3 as required in design.
To avoid the occurrence of thermal cracks, several measures were conducted.
Fresh concrete was cooled down by the spraying liquefied nitrogen gas onto the
concrete right after mixing. Pipe cooling method after placing concrete was also
carried out.
Requirements for concrete were as follows;
1)

Unit weight : Not less than 2.3t/m3

2)

Specified design strength: 24MPa at the age of 91days.

3)

Air content : 1.5 % - 3.5%

4)

Slump flow: 550 - 650mm at mixing plant and 450 - 600mm at placing spot.

Tunnel Anchorage at Kurushima Ohashi Bridge[13] (Compaction difficulty):

Tunnel anchorage of Kurushima Ohashi Bridge (area of a standard cross section
was approximately 80m2) was installed into the earth at a 40-degree incline as
shown in Fig. 6. Concrete was planned to be placed into the tunnel anchorage
where cable anchor frames and reinforced steel members were complicatedly
installed. Considering that sufficient vibrating compaction could not be easily
achieved because setting up the scaffolding for each lift in the tunnel was costly
and time-consuming, SCC was applied to the project.
Since the concrete was manufactured at an ordinary mixing plant, combinationtype SCC containing viscosity-agent was adopted to reduce the quality fluctuation
of fresh concrete due to the change of surface water content of fine aggregate.
Requirements for concrete were as follows;
1) Unit weight : Not less than 2.3t/m3
2) Specified designed strength: 24MPa at the age of 91days.
3) Air content : 1.5 % - 3.5%
4) Slump flow: 550-650mm at discharging spot.
5) Level of self-compactability : Rank 2 (see Section 1.3)
Mixture proportions applied to the project are shown in Table 4. The cement was
low heat cement composed of moderate-heat portland cement (23%), ground
granulated blast furnace slag(50%) and flyash(27%). Unit cement content was
maintained less than 260kg/m3 to satisfy the criteria that ultimate adiabatic
temperature rise should not be larger than 25oC. Viscosity agent was nonadsorbed glycol type one in the liquid state.
Around 13,000m3 of SCC were used for 2 tunnels. SCC was pumped into the
tunnel and poured from the bottom to the top of the tunnel through two pipes
installed along the tunnel. Quality control tests were conducted at the mixing
plant, at the discharging spot near the entrance of the tunnel and at the placing
spot inside the tunnel.
Viaducts of Railway (Densely arranged reinforcing bars): In the last decade,
seismic design has been changed with strict criteria to endure the supposed great
earthquakes in Japan. As a result, in structure like viaducts of railway, especially
at the intersectional part of column, beam and slab, very densely arranged steel
reinforcing bars as shown in Fig. 7 have become common.
Mass of reinforcement at the intersectional part often exceeds 500kg/m3 and
sometimes reaches 600kg/m3. Minimum clearance between reinforcing bars was
40mm or so, vibrator can not be easily inserted through reinforcing bars. In these

cases, Rank 1 SCC (see Section 1.3) is often utilized instead of ordinary concrete
to attain the required quality. Mixture proportions depend on the requirements.
Coarse aggregate with maximum size of 15mm is preferred and often used to
enhance the self-compactability.
Application to CFT: Concrete-filled tubular (referred to as CFT) structure was
firstly adopted in 1990s in Japan. The development of the high-strength SCC
made it possible to fill the concrete completely into a steel tubular column having
a number of diaphragms inside (see Fig.8). CFT structure has rapidly been
prevailing in Japan with the strength requirements for concrete becoming more
demanding. In this section, an application record of CFT structure with highstrength SCC to a high-rise building [14] is reported.
Building A is a 47-story high-rise condominium located in Tokyo. Columns up to
90m above ground were designed in CFT structure using SCC with a specified
design strength of 100MPa. The cross-section of the main columns was a square
measuring 900 by 900mm. Square columns were strengthened with internal
diaphragms, each having a round opening in the center with a diameter of 200 to
400mm, through which SCC passed. SCC was pumped from the ground (see
Fig.9) to the allowable level at which the lateral pressure was expected to reach
the permissible stress of the welds. After aforementioned SCC hardened, concrete
buckets were used for filling SCC into the remaining upper parts of square
columns. Circular columns with a diameter of 800mm were partially used. Since
circular columns were found strong against the lateral pressure, SCC was placed
into circular columns by pumping from the ground to a height of 90m in a single
lift.
Table 5 shows the mix proportions of the high-strength SCC with water binder
ratio of about 20% and slump flow of 650mm. The binder was 70% ordinary
portland cement, 20% blast-furnace slag and 10% silica fume.
Application to high-strength steel fiber reinforce concrete: In Europe, highperformance steel fiber reinforced cementitious composite referred to as Reactive
Powder Concrete (RPC) was developed [15]. The RPC is a sort of SCC since it
needs no vibrating compaction during placement. Steam curing at 90oC and the
densest packing design enable to produce precast concrete having highperformance and ultra high-strength of around 200MPa. The actual applications
of RPC have been done by 35 projects in the world. Recently some applications
have been done in Japan and related recommendations have been established by
JSCE [16]. In this section, main features as to mixture proportions, properties in
fresh and hardened states, mechanical properties and construction method of the
first application to pre-stressing concrete bridge in Japan is reported.

10

Table 6 shows the mixture proportions. Water-cement ratio is around 23% and
water-binder ratio is 12%. Cement, quartz and silica fume particles are well
balanced to obtain the densest packing described in Fig. 10. This packing design
helps achieve high-strength and high durability. RPC shows good selfcompactability for casting into thin or complicated shaped mold. It takes 8-14
minutes of mixing to get the specified flowability even by the revolving-blade
mixer.
The compressive strength of RPC is 200-240MPa and the tensile strength is
around 9MPa after 90oC heat curing for 48hours. The high tensile strength
combined with enough ductility makes conventional reinforcement unnecessary.
The 15mm long steel fibers inside of the mortar matrix act as reinforcement to
resist tensile stress.
RPC premix, water and a superplasticizer were mixed as a primary mixing. After
checking the flow value, steel fibers were added to the mixture and it was mixed
as a secondary mixing for 7 min. Then, the flow value was checked again (see
Fig. 11). Mixed RPC was placed through a tremie pipe attached to the hopper
outlet to manufacture the precast blocks described in Fig. 12. After placing, the
block was cured by sheets to prevent water evaporation as a primary curing. For
secondary curing, 90oC steam curing was conducted for 48 hours in a house. The
rate of rising and dropping temperature were controlled to15oC/h and 7 to 10oC /h,
respectively to prevent the cracks due to temperature difference in the blocks.
Transverse wet jointing of blocks was conducted in the plant prior to installation.
After blocks were fixed on both sides, RPC was poured from the bottom slab to
the top slab. The joints were heat-cured by electrical heaters and insulation. The
strength was achieved as temperature between 70 and 90oC was maintained at
joints. A general view of Sakata Bridge constructed with PRC is shown in Fig. 13.
FUTURE OF SCC
Future research on SCC
In Japan, recently some researches have been working to estimate the selfcompactability, especially the pressure loss between obstacles when SCC is
flowing in formworks [17].
Background of research Recently in Japan, ordinary concrete can not be applied
to construction of concrete structures on some conditions that densely reinforcing
bars are arranged. Concrete cannot flow through these reinforcing bars. In these
cases, it is difficult to use vibrators and to check state of concrete compaction
directly. Therefore, in case of SCC, a perfect concrete compaction inside

11

formwork is expected. However, the required self-compactability is not always

attained.
Therefore, to study the reason of this lack of self-compactability, it is necessary to
clear the mechanism and to establish the prediction method of selfcompactability.
It is considered that this lack of self-compactability is caused by the pressure loss
between obstacles when SCC is flowing in formworks.
Fig. 14 shows the image of the pressure loss. If concrete is a Newtonian body, the
height difference between two surfaces of concrete in tank A and tank B does not
exist. However, the height difference always exists due to the yield value, because
concrete is a Bingham body.
This height difference is explained by pressure loss of concrete flowing. A
potential energy of concrete is decreased with flowing, because the energy is
spent for friction loss between concrete and surface of formwork, contact loss
with coarse aggregate each other when concrete passes through steel bars, and
transformation loss of concrete with flowing. This energy loss appears as a height
difference between concrete surfaces, and it is called pressure loss.
Estimated result for the pressure loss
To investigate a mechanism of pressure loss, it is used the visual testing method
by using model fresh concrete for the pressure loss test; model concrete is
skeleton concrete composed of transparent liquid of Bingham body and artificial
light weight coarse aggregate, and behavior of coarse aggregate is observed. By
using this method and image analysis, continuous change of coarse aggregate
distribution and locus of each coarse aggregate particle can be observed.
From results of these investigations, it is clear that pressure loss has a good
relationship with unit volume of coarse aggregate and clearance between steel
bars (see Fig. 15). As unit volume of coarse aggregate increases and/or as
clearance between steel bars decreases, the pressure loss increases.
In a process of increase pressure loss, flowing velocity of coarse aggregate
particles around steel bars is getting slow or stopping, then only the model liquid
flows through coarse aggregate particles. Then, number of coarse aggregate
particles around steel bars increases. At last, in some cases, around steel bars,
coarse aggregate particles make blockage and this blockage stops the whole
concrete flow.
Mechanism of the pressure loss
Mechanism of causing the pressure loss is considered as follows.

12

[Step 1] Flowing velocity of coarse aggregate in flowing SCC decreases in

upstream area of steel bars
When SCC flows through the interspaces between vertical steel bars in the
formwork, the flowing line of SCC changes from being linear to curved around
the steel bars, and flow of coarse aggregate around the steel bars is obstructed and
the flowing velocity of coarse aggregate is decreased.
[Step 2] Volume of coarse aggregate increases in front of the obstacle
Volume percentage of coarse aggregate particles in SCC increases in upstream
area of steel bars due to the phenomenon described in [Step 1].
[Step 3] The stress transfer mechanism changes in SCC
In SCC, it is considered that the sheer stress and shearing stress act on the
surfaces of the coarse aggregate particles, as shown in Fig.16. In the case of SCC
having low volume of coarse aggregate particles, shearing stress has no influence
on sheer stress.
However, as the volume ratio of coarse aggregate particles to mortar increases,
coarse aggregate particles come into contact with each other easily. As a result, it
is considered that the mechanism of the stress transfer model changes and the
relationship betweenandbecomes linear, as shown in Fig.17.
[Step 4] Sheer stress at the surface of coarse aggregate particles increases
and then yield value of SCC increases partially around steel bars
The shearing stress increases with increasing sheer stress. Thus, the deformation
resistance force of SCC also increases. This means that the yield value of SCC
increases.
[Step 5] The pressure loss increases
When SCC flows through the interspaces between the steel bars, a balance of
dynamic is shown in Fig. 18 and Equation (1). This equation shows that pressure
loss increases as the yield value of SCC increases.
P

2D
D L

Where,
c :

Yield value of SCC

P :

Pressure difference between the steel bars

D:

Diameter of the steel bars

13

(1)

L:

Clearance between steel bars.

Estimated results and needed research in future

Fig. 19 shows the comparative results of measured pressure loss and estimated
one based on the pressure loss mechanism. From this figure, the relationship
between measured pressure loss and estimated one depends on the dosage of
viscosity agent. The pressure loss is influenced by both rheological properties
such as yield value and plastic viscosity. Therefore this difference occurs from the
reason why this model considers the effect of only yield value but does not
consider viscosity.
Therefore it is necessary to improve the model and establish the accurate
predicting method for the pressure loss in the near future.
REFERENCES
1. Nagataki, S. and Fujiwara, H., Self-Compacting Property of Highly-Flowable
Concrete, Proceedings, Second CANMET/ACI International Symposium on
Advances in Concrete Technology, ACI SP-154, pp.301-314, 1995
2. Japan Society of Civil Engineers, State of the Art-Report on Self-Compacting
Concrete, Concrete Engineering Series, No. 15, 1996 (in Japanese).
3. Japan Society of Civil Engineers, Recommendation for practical application of
self-compacting concrete, Concrete Library, No.93, 1998 (in Japanese).
4. Khayat, K. H., et al, Guidelines for Design and Construction of SelfConsolidating Concrete for Precast, Prestressed Concrete Members,
Supplemental Papers, Ninth ACI International Conference on Superplasticizers
and Other Chemical Admixtures in Concrete and Tenth ACI International
Conference on Recent Advances in Concrete Technology and Sustainability
Issues, pp.441-474, 2009.
5. Kawai, T., Chemical Admixtures for Highly Flowable Concretes,
Proceedings of the Intl Workshop, Rational Design of Concrete Structures under
Severe Conditions, pp.291-302, 1995.
6. Hattori, K., Experiences with Mighty Superplasticizers in Japan, SP-62,
pp.37-66, 1979.
7. Aignesberger, A. and Kern, A., Use of Melamine-Based Superplasticizer as a
Water Reducer, SP-68, pp.61-80, 1981.

14

8. Izumi, T., Slump Control with Reactive Polymeric Dispersant, Proceedings

of Third International Conference on Superplasticizers and Other Chemical
Admixtures in Concrete, ACI SP119-1, pp.243-264, 1989.
9. Yamada, K. and Uomoto, T., Cross-National Comparison of Concrete
Materials and Message for Future, -Results of International Questionnaire Survey, Keynote Lecture 2, 4th International Conference on Construction Materials:
Performance, Innovations and Structural Implications, pp.71-80, 2009.
10. Present and future of SCC, Journal of Concrete Technology, Vol.27, No.2.,
2008 (in Japanese).
11. Ouchi, M., International Information:Tangible difference between the
definition and the real state of SCC- 5th International RILEM Symposium on SelfCompacting Concrete-, Vol.46, No.2, pp.89-91, 2008 (in Japanese).
12. Yasuda, M., et al, Construction of Anchorage with Highly Workable Concrete
Capable of Casting 1900 Cubic Meter Per Day/ Akashi Kaikyo Bridge, Cement
& Concrete, No.558, pp.60-64, 1993 (in Japanese).
13. Okada, R., et al, Application of Self-Compacting Concrete to Tunnel
Anchorage of Kurushima Ohashi Bridge, Civil Engineering, Vol.51, No.12,
pp.44-51, 1996 (in Japanese).
14. Jinnai, H., et al, Construction Records of High-rise Buildings: Applications
of High Strength CFT Structure, Technical report, Taisei Corporation, 2005 (in
Japanese).
15. Richard, P., et al, Reactive Powder Concretes with High Ductility and 200800MPa Compressive Strength, ACI SP-144, American Concrete Institute,
pp.507-517, 1994.
16. Japan Society of Civil Engineers, Recommendations for Design and
Construction of Ultra High Strength Fiber Reinforced Concrete Structures
Draft, 2004 (in Japanese).
17. Maruoka, M., et al, Experimental investigation of estimating method of
pressure loss of self-compacting concrete pass through the gaps between re-bars,
JSCE Journal of materials, concrete structures and pavement, No.795/V-68,
pp.111-126, 2005 (in Japanese).

15

Table 1 - Standard values of unit absolute volume of coarse aggregate

Unit absolute volume of coarse aggregate (m3/m3)
Self-compactability
Powder-type

Viscosity agent-type

Combination-type

Rank 1

0.28-0.30

0.28-0.31

0.28-0.30

Rank 2

0.30-0.33

0.30-0.33

0.30-0.33

Rank 3

0.32-0.35

0.30-0.36

0.30-0.35

Table 2 - Share of SCC in Europe

Share
Country
Denmark
Sweden
Netherland

Placed in situ
28%(2006)
24%(2005)
5%(2003)
No information

Factory
product
50%(2005)
50%(2003)
70%(2005)

Table 3 - Mixture proportions (powder-type)

Slump
Gmax
flow
(mm)
(mm)
4501A
40
600
4504A
40
600
LP:Lime stone powder

Unit content (kg/m3)

W/C
(%)

s/a
(%)

Air
(%)

LP

SP

55.8

45

2.5

145

260

150

769

965

6.355

55.8

36

2.5

145

260

150

609

1121

7.8

16

Table 4 - Mix proportions (combination-type )

Slump
Gmax W/C
flow
(mm) (%)
(mm)
600-700
20
62.5
LP: Lime stone powder

Fc
(MPa)
100

Slump
flow
(mm)
600700

Unit content (kg/m3)

s/a
(%)

Air
(%)

LP

Va

SP

49.6

2.5

162.5

260

210

823

880

4.06

7.8

Table 5 - Mixture proportions

Unit content (kg/m3)
Gmax
W/B Air
(mm
(%) (%) W
B
S
G
SP
)
15 77
86
20
20
2.0
634
13.6
5
5
2

B: Ordinary Portland cement 70%, GGBF Slag 20%, Silica fume 10%

Table 6 - Typical mixture proportions

Comp.
strength
(MPa)

Flow
value
(mm)

200-240

240260

Steel fiber
Dia.
(mm)

Length
(mm)

Vol.
(%)

0.2

15

180

17

Unit content (kg/m3)

Grain
Steel
C
(quartz,
SP
fiber
sand etc.)
774

1523

157

22

Fig. 1 - Shapes and dimensions of apparatus for self-compactability test

Fig. 2 - Share of superplasticizers in Japan(up to 1998)

18

PC-based

BNS-based

MM-based

Japan
Korea
China
Vietnam
Thailand
India
Singapore
Malaysia
NewZealand
Australia
USA
Canada
Germany

Share

Ratio

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%

RMC SP
Fig. 3 - Share of superplasticizers in each country

Fig. 4 - Volume of SCC application in Japan

19

Fig. 5 - Outline of Akashi Kaikyo Ohashi

Fig. 6 - Placement of SCC at tunnel anchorage

20

Fig. 7 - Intersectional part of column and beam (densely arranged

reinforcing bars)

Fig. 8 - Diagrams in the CFT

Fig. 9 - Placing with a concrete pump

21

Porosity

Densest
packing

Quartz
Cement

Cement

Silicafume
Fig. 10 - The densest packing design

Fig. 11 - Measurement of flow value

Fig. 12 - Precast block

22

Fig. 13 - General view of completed bridge (Sakata Mirai Bridge)

Tank A partition
gate

Tank A

Tank A

Tank B

concrete

Tank B

Tank B

ha

Steel bars

hb

concrete
concrete

Figure 14 - Image of the pressure loss

Pressure Loss P (Pa)

2000
1500

Clearance: L
30mm
35mm
40mm
45mm

Blockage

1000
500
0
0.26
0.28
0.30
0.32
0.34
Unit volume of coarse aggregate Xv (m3/m3)

Fig. 15 - Relationship among unit volume of coarse aggregate Xv, clearance

L, and pressure loss

23

Shearing stress:
Coarse
aggregate

Sheer stress:

Concrete flow

Steel bar

Shearing Stress :t

Fig. 16 - Concept of sheer and shearing stress

By increasing of volume
of coarse aggregate

Sheer Stress :s

Fig. 17 - Concept of change in stress transfer mechanism

D+L
P

Steel bar

SCC

Concept of the dynamic balance

Yield stress
tc=

P(D+L)
2D

Distribution of shearing stress

Fig. 18 - Concept of dynamic balance between steel bars

24

Estimated Pressure Loss (Pa)

250

Dosage of
viscosity agent
Vis.=12.5%
Vis.=15.0%
Vis.=17.5%

200
150
100
50
0

50 100 150 200 250

Measured Pressure Loss (Pa)

Fig.19 - Relationship between measured and estimated pressure loss

25

26

GUIDELINES FOR DESIGN AND CONSTRUCTION OF SELFCONSOLIDATING CONCRETE FOR PRECAST, PRESTRESSED
CONCRETE MEMBERS
Kamal H. Khayat, Soo-Duck Hwang, Gillaume Lemieux, and Wu Jian Long

Synopsis: Self-consolidating concrete (SCC) is a specially proportioned hydraulic

cement concrete that enables the fresh concrete to flow easily into the forms and
around the reinforcement and prestressing steel without segregation. Use of this
type of high-performance concrete for the manufacture of precast, prestressed
bridge elements provides the benefits of increased rate of production and safety,
reduced labor needs, and lower noise levels at manufacturing plants. In spite of its
benefits and widespread use in Japan and Europe, the use of self-consolidating
concrete in the United States has been limited because of concerns about certain
design and construction issues that are perceived to influence the structural
integrity of the bridge system. These issues include workability, strength
development, creep and shrinkage properties, durability, and other factors that
influence constuctibility and performance.
An extensive laboratory evaluation was carried out to develop guidelines for the
use of SCC in precast, prestressed concrete bridge elements, including
information on material selection, mixture proportioning, and workability
evaluation. Key engineering properties, durability characteristics, and structural
performance of SCC were also investigated.
Guidelines for the use of self-consolidating concrete in precast, prestressed
concrete bridge elements were developed and relevant changes to AASHTO
LRFD bridge design and construction specifications were proposed. These
guidelines should provide highway agencies with the information necessary for
considering concrete mixtures that are expected to expedite construction and yield
economic and other benefits.

Keywords: creep; guidelines; mechanical properties; precast; prestressed

concrete; self-consolidating concrete; shrinkage; workability.

27

Kamal Henri Khayat is Professor of Civil Engineering and Director of the

Integrated Research Laboratory in the Valorization of Materials and Innovative
and Durable Structures at the Universit de Sherbrooke in Canada. He is Fellow
of the American Concrete Institute and serves on several technical committees of
ACI, RILEM, and CSA, including Chairman of ACI 237 on SCC and RILEM TC
on Mechanical Properties of SCC.
Soo-Duck Hwang is a Research Assistant at the Universit de Sherbrooke,
Sherbrooke, Quebec, Canada. His research interests include workability, transport
properties, and visco-elastic properties of self-consolidating concrete used in
repair applications and precast/prestressed elements.
Gillaume Lemieux is an engineer at Euclid Canada. He received his M.S. in Civil
Engineering from Universit de Sherbrooke. His research interests include
workability and mix design of self-consolidating concrete.
Wu Jian Long is R&D Manager at Haozhuoyinghua Technology Development
Co. Ltd, Beijing, China. He received his Ph.D. in Civil Engineering from
Universit de Sherbrooke. His research interests include workability, mechanical
properties, and visco-elastic properties of self-consolidating concrete.

INTRODUCTION
The competitive situation in the precast concrete construction market is
significantly affected by price and cost factors as well as by productivity and
quality. This environment is characterized by ever-shorter construction times,
rising labor costs, as well as greater demand for high workability, strength, and
durability. Technological developments and methods of production that can lead
to improved concrete quality and savings are therefore becoming increasingly
important. Self-consolidating concrete (SCC) represents a significant
advancement in concrete technology that provides great potential for efficiency
and economy in concrete construction.

28

SCC is a highly workable concrete that can flow through densely reinforced or
geometrically complex structural elements under its own weight without
mechanical consolidation and adequately fill the formwork with minimum risk of
segregation. The flowability of SCC is higher than that of normal highperformance concrete typically used in precast, prestressed concrete plants. This
characteristic of SCC, coupled with the absence of the noise associated with
vibration, make SCC a desirable material for fabricating prestressed bridge
elements. The use of SCC in the precast, prestressed applications can therefore
enable complex precast concrete members to be prefabricated with greater ease,
speed, economy, and higher surface quality. This can be achieved even in tightlyspaced areas or congested reinforcement, such as columns, cap beams and
superstructure elements and lead to providing uniform and aesthetically pleasing
surfaces. The quality control and quality assurance measures used for producing
SCC will also help achieve structures with the desired durability and service life.
Properly designed SCC is expected to provide similar properties as the
conventional counterparts except for the high workability. However changes in
mix design and fluidity of SCC can result in SCC with hardened properties and
performance that are different from that commonly expected from conventional
concrete. Proper selection of material constituents and proper proportioning are
necessary for achieving the desired workability and performance of SCC. The
factors that significantly influence the design, constructability, and performance
of precast, prestressed bridge elements with SCC need to be researched. It is also
necessary to develop guidelines for the use of SCC in bridge elements and to
recommend changes to AASHTO LRFD bridge specifications. These guidelines
will provide highway agencies with the information necessary for considering
concrete mixtures that are expected to expedite construction and yield economic
and other benefits that are associated with SCC (surface finish, labor cost, etc.).
For successful design of SCC, some factors require greater attention than that
required for conventional concrete, including type and size and grading of coarse
aggregate, composition and content of binder, and w/cm. Proper selection of
material constituents is necessary for workability and performance of the
hardened concrete.
A number of test methods have been used to characterize workability of SCC. No
single test method has been found to fully characterize all relevant workability
aspects of SCC. Selection of proper combined test methods can facilitate
workability testing protocol and provide means for quality control of field
applications.
Knowledge of the compressive strength, elastic modulus, and flexural strength of
concrete is required for estimating camber of prestressed members at the release
of the prestressing load, and for determining elastic deflections caused by dead

29

and live loads, axial shortening and elongation, and prestress losses. Literature
review showed that a loss of up to 20% in the modulus of elasticity could be
obtained compared to the high-performance concrete of normal consistency
because of the lower coarse aggregate volume of SCC [Holschemacher and Klug,
2002]. However, under air-drying conditions, the elastic modulus of SCC can be
higher than that of normal concrete at long term.
Typically, SCC mixtures are proportioned with higher binder content and lower
coarse aggregate volume and maximum size which increase the risk of thermal,
autogenous and drying shrinkage, and creep leading to loss of prestress and
excessive deflections and elastic shortening. Therefore, creep and drying
shrinkage characteristics of SCC need to be considered in the design of precast,
prestressed bridge elements to account for losses in prestress and avoid cracking
of concrete.
According to the literature survey, there seems to be some discrepancy regarding
the visco-elastic properties of SCC because of differences in mix design (w/cm),
type and content of coarse aggregates, type of chemical admixture, and testing
exposure. It is reported that the creep potential of SCC appears to be slightly
higher than that of conventional concrete made with the same raw materials and
having the same 28-day compressive strength [Attiogbe et al., 2002; Pons et al.,
2003; Byun et al., 1998]. Depending on the selected binder, w/cm, and ambient
temperature at the precasting plant, the use of new generation HRWRA may
eliminate the need to use radiant heat or steam curing.
SCC used in precast, prestressed applications is typically proportioned with a low
w/cm (0.32 to 0.36) to enhance stability of the plastic concrete. Relatively low
w/cm values, coupled with high content of binder lead to greater degree of
autogeneous shrinkage than conventional concrete. Such type of shrinkage also
increases with the fineness of the binder and fillers in use. Therefore, drying
shrinkage, autogeneous shrinkage, and thermal contraction have to be managed in
the mix design process and in the structural detailing of the prestressed element.
Studies have shown that the scatter between measured and predicted drying
shrinkage values is greater in the case of SCC than that for normal concrete.
Experimental shrinkage strains for SCC were found to be larger than those
estimated by various prediction models [Byun et al., 1998]. Also, comparison of
experimental creep data to those obtained from major creep-prediction models
indicated differences. Work is required to compare creep and shrinkage data of
SCC mixtures made with representative mix designs and the material constituents
available in the United States to those obtained from prediction models.
The stability of SCC is a key property in ensuring uniform mechanical properties
and adequate performance of precast, prestressed bridge girders. Properly

30

designed SCC mixtures can exhibit uniform distribution of in-situ compressive

strength. Bond strength and its uniformity along the height of cast girders can be
influenced by flow properties of the SCC, grading of the aggregate, and content of
fines. Some studies have found that bond strength of SCC to reinforcement can be
lower than that of normal concrete [Koning et al., 2001; Hegger et al., 2003].
Other studies, however, have shown that for a given compressive strength,
reinforced concrete members made with SCC can develop higher bond strength
than in the case of normal concrete [Dehn et al., 2000; Chan et al., 2003]. Bond
strength that can be developed between SCC and prestressed strands and its
uniformity along the height of cast wall elements were investigated in this project.
The structural design concerns related to the use of SCC for constructing
prestressed girders include the likely lower modulus and greater shrinkage of SCC
and the possible larger prestress losses, and the reduced shear resistances resulting
from the use of a smaller maximum aggregate size or a smaller volume of coarse
aggregate.
PROJECT OBJECTIVES AND SCOPE
The aim of the proposed research carried out as a USA National Cooperative
Highway Research Program (NCHRP) Project 18-12 (Self-Consolidating
Concrete in Precast, Prestressed Concrete Bridge Elements) carried out at the
Universit de Sherbrooke in collaboration with McGill University in Canada
[NCHRP Report 628, 2009] is to develop guidelines for use of SCC with precast,
prestressed bridge elements, including recommended changes to the Load and
Resistance Factor Design (LRFD) Bridge Design Specifications of the American
Association of State Highway and Transportation Officials (AASHTO), hereafter
referred to as the AASHTO LRFD Specifications. Such guidelines will provide
highway agencies with the information necessary for considering SCC mixtures
that are expected to produce a uniform product, expedite construction, and yield
economic and other benefits. Accomplishing this objective will require a research
effort to:
develop material properties and performance criteria for SCC used for precast,
prestressed concrete bridge elements.
evaluate key engineering properties, durability characteristics, and structural
performance of such concrete.
propose relevant changes to AASHTO LRFD Bridge Design and Construction
Specifications.
Specifically, this project aimed at:

31

developing SCC mixtures that can be produced consistently in the field;

identifying test methods for use in SCC mix design;
identifying test methods for use for quality control in precasting plants;
developing specifications and criteria for SCC mixtures for precast, prestressed
concrete bridge elements;
determining the influence of mix parameters, such as raw materials, mixture
proportioning, mixing, production, placement, and characteristics of the cast
element;
comparing performance of precast, prestressed concrete elements made with
SCC with those made with conventional high-performance concrete;
preparing guidelines for testing, proportioning, and casting SCC bridge
elements;
investigating applicability of current models recommended by AASHTO
LRFD Bridge Design and Construction Specifications and suggest revisions
whenever applicable.
The paper presented here summarizes the guidelines for the use of SCC in precast,
prestressed bridge girders. The guidelines include information on the selection of
concrete constituents and proportioning of concrete mixtures, workability
characteristics, testing methods, mechanical properties, visco-elastic properties,
production and control issues, and durability of SCC.
SELECTION OF CONSTITUENT MATERIALS
The production of SCC requires uniform quality of all constituent materials, and it
is therefore necessary that these materials meet standard specifications. A choice
of suitable constituent materials is vital to the optimization of SCC mix design for
different applications. Constituent material qualification for SCC designated for
precast, prestressed concrete bridge elements generally follows the requirements
of AASHTO LRFD Bridge Design [2004] and Construction [1998]
Specifications. It is important to continually check for any change in materials or
proportions that will affect surface appearance, strength, or other characteristics of
SCC that may affect its overall performance.
Cement and Cementitious Materials
One must ensure that material additions do not adversely affect the desired
architectural appearance, where appearance is a design requirement.
Cement and blended cement

32

All cements which conform to the AASHTO M 85 or ASTM C 150 standard

specifications can be used for the production of SCC. The correct choice of
cement type is normally dictated by the specific requirements of each application
or by the availability. For SCC applications where visual appearance is important,
adequate cement content and uniform w/cm should be adopted to minimize the
color variation. Therefore, the cement should be from the same mill and of the
same type, brand, and color.
Selection of the type of cement will depend on the overall requirements for the
concrete, such as compressive strength at early and ultimate ages, mechanical
properties, durability, and color considerations in architectural applications where color
and color uniformity are important. Blended hydraulic cements which conform to the
AASHTO M 240 or ASTM C 595M can also be used. Unless otherwise specified,
Types I, II, or III cement; Types IA, IIA, or IIIA air-entrained cement; or Types IP
(portland-pozzolan cement) or IS (portland blast-furnace slag cement) blended
hydraulic cements can be used for the construction of precast, prestressed concrete
elements. Types I, II or III cements can be used with some replacements by
supplementary cementitious materials and other hydraulic binders. In general, fly ash
and slag replacement values shall not exceed 20% and 40%, respectively, to ensure
high-early strength for satisfactory release of strands.
The total content of cementitious materials used in prestressed concrete for a 28day design compressive strength of 4,000 to 8,000 psi (28 to 55 MPa) can vary
from 600 to 1,000 lb/yd3 (356 to 593 kg/m3) [PCI Bridge Design Manual, 1997].
The AASHTO LRFD Bridge Design Specifications [2004] suggest that the sum
of portland cement and other cementitious materials should not exceed 800 lb/yd3
(475 kg/m3), except for Class P concrete where the total cementitious materials
should not exceed 1,000 lb/yd3 (593 kg/m3). These values for SCC designated for
precast, prestressed applications shall range between 650 and 800 lb/yd3 (386 and
475 kg/m3) [ACI 237R-07].
Fly ash
Pozzolans and slag meeting ASTM C 618, C 989, or C 1240 are supplementary
cementitious material and may be added to portland cements during mixing to
produce SCC with improved workability, increased strength, reduced
permeability and efflorescence, and improved durability. In general, Class F fly
ash has been shown to be effective in SCC providing increased cohesion and
robustness to changes in water content [European Guidelines, 2005]. Fly ash
should conform to the AASHTO M 295 or ASTM C 618 [AASHTO LRFD
Bridge Design [2004] and Construction [1998] Specifications]. In general, the
content of cement replaced with fly ash is 18% to 22% by mass [Florida DOT,

33

2004]. In some cases, higher level of fly ash replacement may reduce the ability
of SCC to flow. The replacement rate of fly ash also affects strength and
durability. Contribution of fly ash delays the hydration process and strength
development. Fly ash can also affect air entrainment since the carbon present in
fly ash can absorb air-entraining admixture and adversely affect the ability to
entrain air. Therefore, state specific limits on LOI indicative of the carbon
content. Fly ash shall not be used with Type IP or IS cements.
Silica fume
Silica fume conforming to AASHTO M 307 or ASTM C 1240 can be used as
supplementary cementitious material in the proportioning of SCC for improved
strength and durability. Silica fume also improves resistance to segregation and
bleeding. Special care should be taken to select the proper silica fume content. In
some cases, high level of silica fume addition besides increasing cost can cause
rapid surface crusting that leads to cold joints or surface defects if delays occur in
concrete delivery or surface finish. According to Florida DOT [2004], the
quantity of cement replacement with silica fume should be 7% to 9% by mass of
cementitious materials.
Ground granulated blast- furnace slag
Ground granulated blast-furnace slag (GGBFS) meeting AASHTO M 302 or
ASTM C 989 may be used as supplementary cementitious materials. GGBFS
provides reactive fines and due to large replacement rate usually about 40%
enables a low heat of hydration. Cement replacement by GGBFS is based on the
severity of the environment to which the concrete is exposed. The level of
GGBFS addition is 25% to 70% for slightly and moderately aggressive
environments, and 50% to 70% by mass when used in extremely aggressive
environments. A high proportion of GGBFS exceeding 40% may however affect
stability of SCC resulting in reduced robustness with problems of consistency
control while delayed setting can increase the risk of static segregation. When
used in combination with silica fume and/or metakaolin, GGBFS content should
be limited to 50% to 55% of the total cementitious content, by mass of binder
[Florida DOT, 2004]. However, in precast prestressed members, the amount of
slag is usually 40%. GGBFS shall not be used with Type IP or Type IS cements.
Fillers
The particle-size distribution, shape, and water absorption of fillers may affect the
water demand/sensitivity and suitability for use in the production of SCC. Calcium
carbonate-based mineral fillers can enhance workability and surface finish. The fraction
below 0.005 in. (0.125 mm) shall be of most benefit to SCC flow properties. Contents
of fillers should be evaluated to ensure adequate performance of concrete, including
strength development and durability.

34

OTHER SUPPLEMENTARY CEMENTITIOUS ADDITIONS

Metakaolin, natural pozzolan, ground glass, air-cooled slag and other fine fillers
have also been used or considered as additions for SCC, but their effects need to
be carefully evaluated for both short and long term effects on the fresh and
hardened concrete.
Aggregate Characteristics
A well-graded combined aggregate with sufficient intermediate sizes is highly desirable
for improved stability. Also, if the proper particle shape and texture are selected,
combined aggregate grading can lead to large reductions in water, paste, and cement
contents leading to improved hardened concrete properties. The moisture content,
water absorption, grading and variations in fines content of all aggregates should
be closely and continuously monitored and must be taken into account in order to
produce SCC of constant quality. Changing the source of supply for aggregates is
likely to make a significant change to the concrete properties and should be
carefully and fully evaluated [European Guidelines, 2005]. Gravel, crushed stone,
or combinations can be used as coarse aggregate. In the case of fine aggregate,
natural sand or manufactured sand can be used. Coarse and fine aggregates should
conform to the grain-size distribution recommendations of the project
specifications.
Coarse aggregate
Unless otherwise specified in the contract documents, the recommendation is to
use normal-density coarse aggregate meeting the requirements of AASHTO M 80
or ASTM C 33. The use of continuously graded aggregates is recommended. The
nominal maximum-size of coarse aggregate (MSA) should be selected based on
mix requirements and minimum clear spacing between the reinforcing steel and
prestressing strands, clear cover the reinforcement steel, and thickness of the
member. The recommendations given in the PCI Bridge Design Manual [1997]
apply. Slightly gap-graded aggregates may lead to greater flowability than
continuously graded aggregate. Gap-graded aggregate can, however, increase the risk
of bleeding and segregation, and proper measures are needed to ensure adequate static
stability of the concrete.
In the design of SCC, typically the MSA values are smaller than those of
conventional vibrated concrete. The MSA is generally limited to to in. (12.5
to 19 mm). In the placement of SCC in highly congested and restricted section,
MSA value of 3/8 in. (9.5 mm) can be used.
If aggregates susceptible to alkali-aggregate reactivity are used, special precautions
must be observed. These include the use of low-alkali cement, blended cements, or
pozzolans and GGBFS.

35

Fine aggregate
For normal weight concrete, fine aggregates conforming to AASHTO M 6 are
appropriate for the production of SCC. Fine aggregate component should be wellgraded concrete sand. It may be beneficial to blend natural and manufactured sand
to improve plastic properties of SCC. Common concrete sand, including crushed
or rounded sand, siliceous or calcareous sands, can be used in SCC. Particle size
fractions of less than 0.005 in. (0.125 mm) should be considered as powder
material in proportioning SCC. Such fine content can have marked effect on
rheology. Fine aggregates for SCC should conform to the gradation requirements
of AASHTO M 6 or ASTM C 33, as presented in Table 1.
Chemical Admixtures
Chemical admixtures are used in precast, prestressed concrete to reduce water
content, improve filling ability and stability, provide air entrainment, accelerate
strength development, enhance workability retention, and retard setting time.
Because chemical admixtures can produce different results with different binders,
and at different temperatures, the selection of the admixtures should be based on
the plant materials and conditions that will be utilized in production. For
prestressed concrete, chloride-ion content in chemical admixtures should be
limited to 0.1%, by mass of the admixture [AASHTO LRFD Bridge Design
Specifications, 2004]. Incompatibility of admixtures with binders can lead to
improper air void system and delayed or accelerated setting time. Therefore,
before the start of the project, concrete with the job materials, including the
admixtures, should be tested to ensure compatibility. Such testing should be
repeated whenever there is a change in the binder and admixtures.
High-range water-reducing admixtures
High-range water-reducing admixtures (HRWRA) shall conform to the
requirements of ASTM C 494 Type F (water-reducing, high range) or G (waterreducing, high range, and retarding) or ASTM C 1017. The admixture should
enable the required water reduction and fluidity during transport and placement.
The use of Type F or G HRWRA is essential to achieve SCC fluidity. Such
HRWRA can be used in combination with regular water-reducing admixtures or
mid-range water-reducing admixtures. There are mid-range water-reducing
admixtures that may be classified under ASTM C 494 as Type A or F depending
on dosage rate. The required consistency retention will depend on the application.
Precast concrete is likely to require a shorter retention time than cast-in-place
concrete.

36

Viscosity-modifying admixtures
The use of a viscosity-modifying admixture (VMA) for SCC proportioned with
w/cm higher than 0.40 is recommended to ensure stability of the fresh concrete.
VMA should not be added to SCC as a means for improving a poor mix design or
poor selection of materials. High dosage of VMA may lead to increased HRWRA
demand and in some cases, some delay in setting, and development of early-age
mechanical properties.
VMAs are used in SCC to enhance segregation resistance and to enhance
robustness by minimizing the effect of variations in aggregate moisture content,
temperature, etc. This can make the SCC less sensitive to small variations in the
proportioning and characteristics of material constituents. There are currently no
ASTM specifications for VMA. Producers should confirm by trial mixtures that
VMA does not adversely affect the hardened concrete properties.
Air-entraining admixtures
Air-entraining admixtures shall conform to the requirements of AASHTO M 154
or ASTM C 260. Air-entraining admixtures are used in concrete primarily to
increase the resistance of the concrete to freeze-thaw damage. Proper selection of
air-entraining admixture that can stabilize small bubbles and properly formulated
HRWRA that does not cause a large number of coarse air bubbles are needed to
design the SCC with adequate air-void system.
In some cases, high dosage of HRWRA coupled with the high fluidity of the
mixture can make it difficult to ensure the entrainment of a fine, stable air-void
system in the concrete. HRWRA can also entrain coarse air bubbles.
Compatibility evaluation between the air-entraining admixture and HRWRA is
therefore needed to achieve the targeted air-void characteristics.
Set-retarding and set-accelerating admixtures
An ASTM C 494 Type D set-retarding admixture may be used during hot weather
concreting or when a delay in setting is required, subject to acceptance by the Engineer.
Some water-reducing admixtures at high dosage rates can act as retarding admixtures.
They should be used with caution. Set-accelerating admixture (Type C) shall be used to
decrease setting time and increase the development of early-age mechanical properties.
The admixture is particularly beneficial in precast concrete construction to facilitate
early form removal and release of prestressing [PCI Bridge Design Manual, 1997]. In
the absence of accelerated radiant heat or steam curing, the use of set-accelerating
admixture in SCC may be beneficial in precast applications when using HRWRA,
especially the poly-naphthalene- or melamine-based products.
Shrinkage-reducing admixtures

37

If a shrinkage-reducing admixture is specified in the contract documents,

verification of the air-void system, including air content in hardened concrete,
spacing factor, and specific surface, is recommended. It could be difficult to
entrain air and large dosages of air entraining admixture are needed when a
shrinkage reducing admixture is used.
Other admixtures
Corrosion-inhibiting admixtures can be incorporated to protect the reinforcement
from corrosion. Producers should confirm by trial mixtures that the addition of
any admixture does not adversely affect the hardened concrete properties.
Coloring pigments used in SCC shall conform to the requirements of ASTM C
979. All coloring admixtures required for a project shall be ordered in one lot and
shall be finely ground natural or synthetic mineral oxide or an organic
phothalocyanine dye with a history of satisfactory color stability in concrete
[European Guidelines, 2005]. The use of corrosion-inhibiting admixtures may
hinder the efficiency of other admixtures and cause non-uniformity in color of the
concrete surface (darkening and mottling). There are currently no ASTM
specifications for corrosion-inhibiting admixtures.
Fibers
Synthetic and steel fibers (hybrid fiber) can be used. The dosage rates of the fiber
in SCC ranges between 0.25% and 0.50%, by volume, depending on the type of
applications. The dosage of fibers should be determined given the workability
requirements of the mixtures, which should take into considerations element
characteristics and placement conditions. Changes in mixture proportioning may
be needed to secure good passing ability and filling capacity of the fiberreinforced SCC. The incorporation of synthetic fiber is recommended to reduce
the risk of cracking due to restrained or plastic shrinkage. The dosage of synthetic
fiber should not exceed the 0.50%, by volume, when casting complex and narrow
sections or highly-reinforced structures blocking risk increased.

SELECTION OF WORKABILITY TEST METHODS

Workability describes the ease with which concrete can be mixed, placed, consolidated,
and finished. It describes the filling properties of fresh concrete in relation to the
behavior of concrete in the production process. Workability of SCC is described in
terms of filling ability, passing ability, and stability (resistance to segregation) and is
characterized by data that relates to specific testing methods [ACI Committee 237,
2007]. Various test methods have been used to assess the workability characteristics of

38

SCC. In general, test methods include the components required for evaluating
simultaneously the filling ability, passing ability, and static stability. Table 2
summarizes some of the main test methods that proposed for the evaluation of
workability of SCC. The filling capacity combines the filling and passing abilities
of SCC and can be tested using the caisson filling capacity [Yurugi et al., 1993].
Filling Ability
The ability of SCC to flow into and fill completely all spaces within the
formwork, under its own weight, is of great importance to SCC casting, distance
between filling points, etc. [ACI Committee 237, 2007]. Slump flow test [ASTM
C 1611] is used to assess the horizontal free flow of SCC in the absence of
obstruction. The test method is based on the test method for determining the
slump of a normal concrete. The diameter of the concrete circle is a measure of
the flowability of the SCC. In general, slump flow varies between 23.5 and 29 in.
(600 and 735 mm) for SCC used in precast, prestressed applications [NCHRP
Report 628, 2009]. When slump flow test is performed, the time needed for the
concrete to spread 20 in. (500 mm) is also noted. This test is called T-50 flow
time. Advantages and precautions of slump flow and T-50 flow test methods are
presented in Table 3.
Passing Ability
The passing ability tests evaluate the ability of concrete to pass among various
obstacles and narrow spacing in the formwork without blockage that can rise from
local aggregate segregation in the vicinity of the obstacles that give rise to
interlocking and blockage of the flow in the absence of any mechanical vibration
[ACI Committee 237, 2007].
The J-Ring test [ASTM C 1621] can be used to assess the restricted deformability
of SCC through closely spaced obstacles [Bartos, 1998]. In general, the maximum
difference between slump flow and J-Ring flow varies between 2 to 3 in. (50 to 75
mm) depending on the filling ability (slump flow) of the mixture. A difference
between slump flow and J-Ring flow less than 1 in. (25 mm) indicates good
passing ability and no visible blocking of the concrete. Difference greater than 2
or 3 in. (50 or 75 mm), depending on the slump flow value, reflects blocking of
the concrete.
In the L-box test, the vertical part of the box is filled with concrete and left at rest
for one minute. The gate separating the vertical and horizontal compartments is
then lifted, and the concrete flows out through closely spaced reinforcing bars at
the bottom. The time for the leading edge of the concrete to reach the end of long
horizontal section is noted. The heights of concrete remaining in the vertical
section and at the leaving edge are determined. The blocking ratio (h2/h1) is
calculated to evaluate the self-leveling characteristic of the concrete. A blocking

39

ratio of 0.5 and higher is indicative of adequate passing ability. Higher values are
necessary in densely reinforced and thin sections.
The V-funnel apparatus consists of a V-shaped funnel with an opening of (2.55
3.0 in.) 65 75 mm at its bottom. The funnel is filled with concrete, then after
one minute, the gate is opened and the time taken for concrete to flow through the
apparatus is measured. In the case of structural applications, the V-funnel flow
time lower than 8 seconds indicates good passing ability [Hwang, 2006].
Advantages and precautions of the slump flow and J-Ring flow test, L-box, and
V-funnel methods are presented in Table 4
Filling Capacity
The property to completely fill intricate formwork or formwork containing closely
spaced obstacles is critical for SCC to achieve adequate in-situ performance. SCC
with high filling and passing abilities can achieve good filling capacity and spread
into a predetermined section to fill the formwork under the sole action of gravity
without segregation and blockage [ACI Committee 237, 2007].
Filling capacity test provides a small-scale model of a highly congested section
and is suitable to evaluate the filling capacity and its self-consolidating
characteristics [Ozawa et al., 1992; Yurugi et al., 1993]. For the caisson test, the
maximum size aggregate (MSA) is limited to in. (19 mm). In general, a filling
capacity higher than 70% is recommended for SCC used in precast, prestressed
applications. Advantages and precautions of the caisson filling capacity test are
presented in Table 5.
Static Stability
Static stability refers to the resistance of concrete to bleeding, segregation, and
surface settlement after casting while the concrete is still in a plastic state [ACI
Committee 237, 2007]. Surface settlement test method can be used to evaluate the
surface settlement of SCC from a plastic state until the time of hardening [Manai,
1995]. In general, a maximum surface settlement lower than 0.5% or a rate of
settlement after 30 minutes lower than 0.27% per hour is recommended for SCC
used in precast, prestressed bridge elements. Surface settlement test enables the
quantification of the effect of mixture proportioning on static stability. The
settlement is monitored until achieving a constant value.
The static stability of SCC can also be determined using column segregation test
[ASTM C 1610]. The coefficient of variation of the aggregate among the column
section can be taken as a segregation index (Iseg) [Assaad et al., 2004]. Another
index consisting of the percent static segregation (S) can be obtained by
measuring the difference between aggregate mass at the top and bottom sections
of the column. Column segregation test consists of casting concrete in a column

40

divided into four sections along the concrete sample. From each section, the
concrete is weighed and washed out. Then, the coarse aggregate content is
determined for each section. In general, a segregation index (Iseg) lower than 5%
or a percent of static segregation (S) lower than 15% is recommended for SCC
used in precast, prestressed bridge elements.
The visual stability index (VSI) involves visual examination of SCC prior to
placement and after performance of the slump flow test. It is used to evaluate the
relative stability of batches of the same or similar SCC mixtures. The VSI
procedure assigns a numerical rating of 0 to 3, in 0.5 increments. The VSI test is
most applicable to SCC mixtures that tend to bleed [Daczko and Kurtz, 2001].
The test can be considered as a static stability index when it is observed in a
wheelbarrow or mixer following some period of rest time (static condition). VSI
value of 0 to 1 is recommended for SCC for precast, prestressed concrete bridge
elements. Advantages and precautions of surface settlement and column
segregation tests are presented in Table 6.
Dynamic Stability
Adequate resistance of concrete to separation of constituents upon placement and
spread into the formwork is required for SCC when flowing through closely
spaced obstacles and narrow spaces to avoid segregation, aggregate interlock, and
blockage [ACI Committee 237, 2007]. The caisson test measures the filling
capacity indicative of the filling and passing abilities; therefore, it is a good
indicator of the dynamic stability. Concrete with high filling ability
(deformability) and good passing ability can achieve adequate filling capacity in
restricted and congested sections that are typical precast, prestressed applications.
Adequate combination of filling and passing ability tests can be used to evaluate
the filling capacity of the concrete which is indicative of the dynamic stability.
Rheology
Generally, two key parameters are determined when a rheology measurement test
is performed: the yield stress, 0 and plastic viscosity, p. Below the yield stress
value, the mixture does not undergo any deformation and behaves as an elastic
material. In SCC, the yield stress should be maintained low enough to ensure
good deformability. The plastic viscosity of concrete affects its ease of placement
and speed of flow. In practice, good balance between yield stress and plastic
viscosity should be achieved to ensure both good deformability, ease of
placement, and flow rate of SCC.
Rheological parameters of concrete can be determined using a concrete
rheometer. In general, the test involves recording the shear stress response to
maintain a given rate of shear at different shear rate values. A linear regression of
the data is usually used to determine the rheological parameters (0 and p)

41

according to the Bingham model. Advantages and limitations of rheometer are

presented in Table 7.
MIX DESIGN
The mix design is chosen to satisfy all performance criteria for the concrete in
both the fresh and hardened states. As in the case of conventional vibrated
concrete, the w/cm is one of the fundamental keys governing strength and
durability of SCC. The w/cm of the concrete shall not exceed 0.45 by weight
[AASHTO LRFD Bridge Design Specifications, 2004]. Satisfactory performance
of the proposed mix design shall be verified by laboratory tests on trial batches.
For mix design approval, a minimum of three test cylinders are taken from a trial
batch. The average compressive strength shall be at least 1,200 psi (8.3 MPa)
greater than the specified compressive strength when the strength equal to or less
than 5,000 psi (34.5 MPa). The average strength shall be at least 700 psi (4.8
MPa) greater than 110% of the specified strengths over 5,000 psi (34.5 MPa)
[ACI 318R-05].
Mix Design Principles
Mix design of SCC is vital for the performance of the material, both in the plastic
and hardened states. Compared to conventional vibrated concrete, in designing
SCC, there are a number of factors that should be taken into consideration to a
greater degree than conventional concrete:
properties of locally available raw materials, including mineral, geometric, and
physical properties of aggregates and cementitious materials;
need for a higher level of quality control, greater awareness of aggregate
gradation, and better control of mix water and aggregate moisture;
choice of chemical admixtures and their compatibilities with the selected
binder;
consideration of placement technique, configuration of cast element, and
environmental conditions.
Any mix design approach should consider both the fresh and hardened properties
of the SCC, and include the characteristics of cementitious materials and fillers,
the water content or w/cm, the volume of coarse aggregate, the sand-to-aggregate
ratio (S/A), as well as the air content. The selection of the type and combinations
of chemical admixtures is part of the mix design process and depends closely on
the flow characteristics that are required. For the production of SCC for precast,
prestressed concrete bridge elements, the most relevant hardened properties that
affect material selection and mix design include early and ultimate compressive

42

strengths, flexural strength, elastic modulus, bond-to-reinforcement, creep,

shrinkage, frost durability, impermeability, and resistance to corrosion.
SCC used for structural precast, prestressed applications is typically characterized
by relatively low water content, high concentration of ultra-fine particles (i.e. 80
m), and use of an efficient HRWRA (typically polycarboxylate-based, although
other types are also used). SCC made with polycarboxylate-based HRWRA can
usually exhibit short setting time, high-early strength development, and reduce
tendency to segregation.
In principle, there are three approaches that can be used for the production of
SCC:
increasing the ultra-fines content by using fly ash, blast-furnace slag, limestone
filler (powder type), and in some cases low content of silica fume;
use of suitable viscosity-modifying admixture (VMA) (viscosity agent type);
combination of the above approaches (combination type) where low
concentration of VMA is used in SCC of high fines/powder content.
Minimum free water content
This approach entails the use of high content of ultra-fine materials and low water
content to enhance the workability of the SCC. Such concrete typically has a w/cm
of 0.30 to 0.35 with a content of ultra-fines 80 m (approximately No. 200
sieve) of 845 to 1,110 lb/yd3 (500 to 600 kg/m3) [Okamura, 1997; Ozawa et al.,
1992]. The replacement of part of the cement with a less reactive powder is
necessary to limit the heat of hydration, and minimize volumetric changes. In
general, this approach can result in SCC mixtures with low yield value and
moderate-to-high viscosity. The concrete requires a relatively high dosage of
HRWRA.
Moderate water content and medium concentration of VMA
In this approach, the w/cm can be maintained at the level necessary to satisfy
strength and durability requirements (for example w/cm of 0.40). A moderate
dosage of VMA is then incorporated to secure the required stability. When
incorporated in mixtures with relatively high paste content (exceeding 35% and
sand-to-cement ratio of 0.60 to 0.66, by volume), the use of suitable VMAHRWRA combination can ensure high deformability and adequate stability also
leading to greater filling capacity and better in-situ homogeneity than mixtures
made with low w/cm and no VMA [Khayat, 1999]. The incorporation of VMA
becomes imperative when the powder content is reduced to levels comparable
with those of conventional concrete or high-performance concrete.
Low water content and low concentration of VMA

43

This approach involves the combination of a high content of powder materials and
low dosage of VMA. Such mixtures are typically more robust than those
proportioned with high powder content and low w/cm. Robust mixture can react
less sensitively to fluctuations in the mixture composition, characteristics of the
raw materials, water content, and concrete temperature.
Cementitious Materials Content and Water-Cementitious Material Ratio
The concrete supplier shall determine the cementitious materials content and
w/cm required to satisfy the specified concrete category. In general, the
cementitious materials content recommended for SCC ranges between 650 and
800 lb/yd3 (386 and 475 kg/m3) [ACI 237R-07]. The w/cm ranges from 0.32 to
0.45. Special care should be taken to select the binder composition of the SCC
made with low w/cm to limit the compressive strength to the target value.
Otherwise, high strength and stiffness could lead to cracking given the high
degree of restrained shrinkage that can take place.
Nominal Size of Coarse Aggregate
Select MSA based on mix requirements and minimum clear spacing between the
reinforcing steel, clear cover to reinforcing steel, and thickness of the member.
Use coarse aggregate with MSA of to in. (12.5 to 19 mm), unless otherwise
specified in the contract documents. Coarse aggregate of 3/8 in. (9.5 mm) MSA
shall be used for casting highly reinforced and restricted sections.
Air-Entrainment and Air-Void Stability
Generally, SCC made with polynaphthalene sulfonate (PNS)-based HRWRA can
exhibit a relatively stable air-void system. The use of polycarboxylate ether
(PCE)-based HRWRA can lead in some cases to entrapment of large air bubbles,
especially if the SCC is subjected to prolonged mixing or agitation after the
introduction of such HRWRA. The introduction of shrinkage reducing admixture
(SRA) may also have significant effect on the air-void system since it makes it
more difficult to properly air-entrain the concrete.
Mixture Robustness
During the mixture qualification process, it is recommended to investigate the
robustness of the particular design of SCC to fluctuations in the characteristics of
concrete constituents. In addition, it is desirable to investigate the effect of slump
flow variation on stability for a particular mix design and set of materials. A well
designed and robust SCC can typically accept a change of 8.5 to 17 lb/yd3 (5 to 10
L/m3) in water content without falling outside the specified classes of
performance [European Guidelines, 2005].

44

SCC mixtures are more sensitive to the variations in the properties and conditions
of constituent materials and quantity fluctuations during production. Fluctuations
in raw materials, such as gradations and moisture contents of aggregates and
batching fluctuation can have dramatic influence on the flowability and the
stability of the concrete. Well designed SCC can give acceptable tolerance to
daily fluctuations in ingredients characteristics and environmental changes, such
as temperature. This tolerance is usually termed robustness and is controlled by
good practice in the selection and proportioning of ingredients, storage and
handling of basic constituents, and by appropriate content of the fine powders
and/or by use of VMA [European Guidelines, 2005].
Trial Batches
SCC mix design shall require minimum of four trial batches for varying
cementitious materials or w/cm to establish the proportions that can achieve
workability ranges and robustness: two water contents above and two below the
target value. The following information shall be included in the trial batch data:
source of all materials; specific gravity and gradation results for sand and coarse
aggregate; design slump flow range; target air content and design strength; details
of mixture proportioning, including admixture dosage rates for design slump flow
range; SCC trial mixture test results for Q/C testing; and mixer used for the mix
design, mixing sequence, charging sequence, and mixing time.
Recommended Range of Workability Characteristics
The use of proven combinations of test methods and performance-based
specifications is necessary to reduce time and effort required for the development
and quality control of high-performance SCC. A set of performance-based
specifications of SCC is summarized in Table 8. Such specifications also include
test methods recommended for material selection and mix design that can be
performed when developing the concrete mixture as well as quality control (Q/C)
test methods that can be performed for concrete acceptance at the precasting
plant.
Specific requirements for SCC in the fresh state may change depending on the
type of application and especially on:
confinement conditions related to the element geometry, reinforcement density,
cover, etc.;
placing equipment (e.g. bucket, pump, direct from truck-mixer, skip, tremie);
placing method (e.g. number and position of delivery points);
finishing method.

45

Table 9 presents performance-based specifications for workability of precast,

prestressed SCC.
Quality Confirmation of SCC
Regardless of the mix design approach, laboratory trials should be used to verify
properties of the initial mixture composition with respect to the specified
characteristics and classes. If necessary, adjustments to the mixture composition
should be made. Once all requirements are fulfilled, the mixture should be tested
at full scale in the concrete plant to verify fresh and hardened concrete properties.
In case that satisfactory performance is not obtained, consideration should be
given to a fundamental redesign of the mixture. Depending on the apparent
problem, the following courses of action might be appropriate [European
Guidelines, 2005]:
Adjust the cement to powder ratio and the water to powder ratio and test the
flow and other properties of the mixture.
Try different types of cementitious materials (if available).
Adjust the proportions of the fine aggregate and the dosage of HRWRA.
Adjust the proportion or grading of the coarse aggregate.
Consider using a VMA to enhance the robustness of the mixture.
Shaded zones indicate suggested workability characteristics. All SCC mixtures
must meet requirements for static stability.
Given the same raw material sources and the same 28-day compressive design
strength, the engineering properties of SCC should be similar to those of
conventional high-performance concrete. For mix design qualification of
hardened properties, modulus of elasticity, shrinkage, and creep testing should be
performed.
Mock-ups are recommended to confirm the production methods and to test the
resulting mixture characteristics. If there is any sign of deficiency that impairs the
concrete performance, such as segregation, sedimentation, cold joints, or any
other visual defects, perform the saw-cut of the mock up products to verify the
aggregate distribution along the saw-cut area. The lower w/cm of SCC will
normally provide a higher 28-day compressive strength than conventional
concrete with normal consistency used in similar applications. The actual strength
attained should be used as the basis for the engineering properties.
EARLY-AGE AND HARDENED PROPERTIES
The quality of SCC in terms of strength and durability is expected to be equal to
or better than that of a similar specified conventional concrete mixture.

46

Setting
Typically, SCC used in precast, prestressed applications proportioned with low
w/cm requires high dosage of HRWRA. The setting time increases with the
increase in HRWRA dosage. Set-accelerating admixtures or heat (steam or
radiant) curing may be needed to decrease the setting time and increase the early
strength development. SCC made with Type I/II cement is shown to have lower
HRWRA demand than that with Type III cement with 20% Class F fly ash. The
latter concrete can then exhibit longer setting time. The use of VMA increases the
HRWRA demand and may lead to some set retardation.
The initial and final setting times can be as low as 4 to 6 hours and 5 to 7 hours,
respectively. These values depend on the materials in use, including HRWRA
type and dosage, binder composition, as well as temperature. Greater setting times
can be obtained when using naphthalene- or melamine- based HRWRA. The
difference between initial and final setting time (ASTM 403-05) can range
between 1 to 3 hours for SCC used in precast prestressed applications
proportioned with w/cm of 0.34 and 0.40, and Type I/II cement or Type III
cement with 20% of fly ash replacement [NCHRP Report 628, 2009]. Setting time
of SCC can be determined by using AASHTO T 197.
Temperature Development
In general, SCC proportioned with high cement content or with Type III cement
can lead to considerable temperature rise. SCC made with Type III cement with
20% Class F fly ash can develop comparable heat rise as that of SCC made with
Type I/II cement. SCC proportioned with the 0.34 w/cm has longer time to attain
maximum temperature than SCC made with 0.40 w/cm. This is mainly due to
higher HRWRA concentration of the former concretes. For a given w/cm, the use
of VMA delays cement hydration, thus extending time to attain peak temperature.
Typical temperature development of SCC proportioned with w/cm of 0.34 and
0.40, cement content of 742 and 843 lb/yd3 (440 to 500 kg/m3), and Type I/II and
Type III cement with 20% of fly ash lie in the range of 115 to 125oF (46 to 52oC)
after 48 hours under semi-adiabatic condition. The maximum temperature can
range from 126 to 145F (52.2 to 62.8C). The time to reach maximum
temperature is in the range of 17 to 28 hours [NCHRP Report 628, 2009]. Type
III cement with supplementary cementitious materials (for example 20% of fly
ash or 30% slag) is shown to attain greater release strength than SCC made with
Type I/II cement. Initial curing with heat (steam or radiant) may then be
necessary, especially for SCC proportioned with relatively low w/cm by reason of
high dosage of HRWRA demand causing retardation. The use of finely ground
limestone filler can also enhance compressive strength development at early age.
Finely ground fillers and supplementary cementitious materials can lead to a

47

denser hardened cement matrix and a denser interfacial transition zone with
aggregate and embedded reinforcement. This can lead to greater strength and
durability.
Release Compressive Strength
For precast applications, SCC mixtures are typically proportioned with 0.32 to
0.36 w/cm [ACI Committee 237, 2004]. The upper range may be increased to 0.40
depending on the concrete temperature and mixture compositions. Relatively low
w/cm can lead to higher compressive strength compared to conventional slump
concrete. The minimum specified compressive strength for prestressed concrete
bridge elements and decks is 4,000 psi (27.6 MPa) [AASHTO, 1998]. Typically,
compressive strength at release of the prestressing strands of structural AASHTO
type girders is on the order of 5,000 psi (34.5 MPa) after 18 hours of casting. The
typical 56-day compressive strength is set at 8,000 to 10,000 psi (55 to 69 MPa).
SCC made with polycarboxylate-based HRWRA can develop higher early
compressive strength and ultimate strength than similar SCC made with
naphthalene- or melamine-based HRWRA. The use of VMA can increase the
HRWRA demand and could lead to reduction in early strength development.
Release strength should be achieved within 18 hours after the concrete is cast into
place. The targeted release strength is selected so that the strength of the concrete
in the prestressed beam does not exceed 60% of the design concrete compressive
strength at the time of release (before any losses due to creep and shrinkage) [PCI
Bridge Design Manual, 1997]. This value is limited to 55% in the case of posttensioned members. Maturity testing can be considered as an effective way to
monitor strength development at early age whether accelerated heating is used
or not. In order to estimate f c ' , ACI 209 and CEB-FIP MC90 models can be used:
ACI 209

fc'

t
t

A Bt

fc'

28 d

'

( f c ) t = compressive strength of concrete at a given time t (in psi);

'

( f c ) 28d = 28-day compressive strength of concrete; t = age of concrete (in

days);
16 non AEA SCC + 4 AEA SCC +2 HPC:
Type I/II cement (moist-cured):
A = 1.52; B = 0.92; R2 = 0.95
Type III + 20% FA (moist-cured):

48

A = 1.64; B = 0.91; R2 = 0.90

16 non air-entrained SCC:
Type I/II cement (moist-cured):
A = 1.70; B = 0.90; R2 = 0.97
Type III + 20% FA (moist-cured):
A = 2.15; B = 0.89; R2 = 0.95
CEB-FIP MC90
f cm (t )

exp s 1

28
t / t1

1/ 2

f cm

f cm (t ) = mean compressive strength at t days (in psi);

f cm = mean 28-day compressive strength;

s = coefficient depending on cement type (0.20 for high early-strength

cement, 0.25 for normal-hardening cement, and 0.38 for slow-hardening
cement);
t1 = 1 day.
16 non AEA SCC + 4 AEA SCC + 2 HPC:
s = 0.19 Type I/II cement; R2 = 0.95
s = 0.20 Type III + 20% FA; R2 = 0.92
16 non air-entrained SCC:
s = 0.20 Type I/II cement; R2 = 0.95

49

s = 0.23 Type III + 20% FA; R2 = 0.93

Flexural Strength
For precast and structural civil engineering applications, SCC mixtures are
typically proportioned with relatively low w/cm of 0.32 to 0.36 and with
supplementary cementitious materials and fillers and are expected to achieve
higher flexural strength and flexural-to-compressive ratio than conventional
slump concrete [ACI Committee 237, 2004]. The flexural strength can be
determined by testing in accordance with ASTM C 293 and C 78-02 or can be
estimated from the compressive strength. For SCC used for precast, prestressed
applications, the flexural strength can be estimated with the AASHTO 2007
model, given by:

f r 0.97 f c

'

'

f c = specified compressive strength of concrete (MPa)

Modulus of Elasticity
In applications where the modulus of elasticity (MOE) is an important design
parameter, the MOE should be determined and considered in the design of the
prestressed concrete member. The MOE is used to calculate camber of prestressed
members at the release of the prestressing load, elastic deflections, axial
shortening and elongation, and prestress losses. The MOE is related to the
compressive strength of the concrete, type and content of aggregate, as well as
unit weight of the concrete. The modulus of elasticity is related to compressive
strength and unit weight of the concrete, aggregate type and content, and testing
parameter, including loading rate, moisture and temperature conditions of the test
specimen, as well as specimen size and shape. The content and MOE of the
aggregate have the largest influence on the MOE of the concrete. Selecting an
aggregate with high modulus of elasticity will increase the modulus of elasticity
of the concrete. Increase in sand-to-coarse aggregate ratio can decrease the
modulus of elasticity of the concrete.
In the absence of measured data, it is recommended to use the equation proposed
by the AASHTO LRFD Bridge Design Specifications [2004] to estimate the
elastic modulus of concrete having a unit weight of 2,427 to 4,214 lb/yd3 (1,440
and 2,500 kg/m) and specified compressive strength of up to 15,230 psi (105
MPa). For an accurate prediction, determine the MOE in conformance with
ASTM C 469 using the job-specific materials. The modulus of elasticity for SCC
used for precast, prestressed applications can be estimated using the AASHTO
2007 equation:

50

Ec

0.043

1.5
c

fc

'

= unit weight of concrete (kg/m);

'

f c = specified compressive strength of concrete (MPa)

In some cases, SCC mixtures can develop modulus of elasticity that can be up to
20% lower than typical values found in high-performance concrete of normal
consistency, which is mainly due to the lower coarse aggregate volume, increase
in paste content, and higher content of ultra-fine materials. At equivalent strength,
SCC made with different cement types should develop similar modulus of
elasticity when cured and tested under identical conditions.
Creep
Length changes of prestressed members due to time-dependent deformation,
creep, and shrinkage play a crucial role in the design of concrete structures and on
structural behavior, especially at long term. Incorrect or inaccurate design for
creep and shrinkage may have important undesirable consequence on stability and
performance of the structure.
Creep behavior is related to the compressive strength of the matrix, coarse
aggregate type, relative content of aggregate, as well as magnitude of applied load
and age of loading. Creep takes place in the cement paste and is influenced by
the capillary porosity of the paste. Cement type and w/cm can affect creep.
High early-strength cement can lead to lower creep. The presence of aggregate
restrains creep deformation in the paste. Therefore, an increase in the volume and
elastic modulus of the aggregate can lower creep.
Due to the higher volume of cement paste and fines and smaller MSA of SCC,
creep potential of SCC can be higher than conventional concrete made with the
same raw materials and having the same 28-day design compression strength.
Perform creep testing in accordance with ASTM C 512 using the job-specific
materials. The age when the load is applied affects creep values. For SCC used in
precast, prestressed elements load should be applied at an early age corresponding
to prestress release time. In the absence of measured data, the modified AASHTO
2007 prediction model can be used to predict the creep of SCC.
AASHTO 2007

t, ti

1.9kvs khc k f ktd ti

0.118

in which:

kvs

1.45 0.0051(V / S ) 0.0

51

khc

1.56 0.008H

kf

35
, ktd
7 f ci '

t
61 0.58 f ci ' t

where:
H=

relative humidity (%). In the absence of better information, H may be

taken from Figure 5.4.2.3.3-1.

k vs = factor for the effect of the volume-to-surface ratio of the component

kf =

factor for the effect of concrete strength

k hc = humidity factor for creep

ktd = time development factor

t=

maturity of concrete (day). Defined as age of concrete between time of

loading for creep calculations, or end of curing for shrinkage calculations,
and time being considered for analysis of creep or shrinkage effects

ti =

age of concrete when load is initially applied (day)

V / S = volume-to-surface ratio (mm)

f ci' =

A=

specified compressive strength of concrete at time of prestressing for

pretensioned members and at time of initial loading for nonprestressed
members. If concrete age at time of initial loading is unknown at design
time, f ci' may be taken as 0.80 f ci' (MPa)
factor for the effect of cement type: 1.19 for Type I/II cement and 1.35 for
Type III + 20% FA binder which may be used for P(SCC)

Autogenous Shrinkage
Autogenous shrinkage corresponds to the macroscopic volume reduction due to
cement hydration (chemical shrinkage) as well as self-desiccation of the cement
paste. The volume of the hydration products is less than the original volume of
unhydrated cement and water. Such reduction in volume can lead to tensile
stresses in the cement paste and microcracking. The reduction of relative humidity
in capillary pores due to cement hydration can also result in negative pressure in
the capillary pores, leading to the formation of meniscus and the development of
tensile stresses in the cement paste.
In the case of concrete proportioned with high w/cm (higher than 0.40),
autogenous shrinkage is low given the ample presence of water in capillary pores.

52

SCC and conventional concrete used in precast applications proportioned with

relatively low w/cm (0.32 to 0.36) and high content of cement and supplementary
cementitious materials could exhibit high autogeneous shrinkage. This is
especially the case when capillary porosity is refined when using silica fume.
Cement type has a considerable effect in the development of autogenous
shrinkage. Higher surface area of the cement can activate the reactivity of the
binder, hence increasing the degree of autogenous shrinkage.
Drying Shrinkage
Drying shrinkage must be taken into consideration to avoid cracking and
excessive deflection resulting from time-dependent concrete deformation and loss
of prestress. The increased volume of paste in SCC and reduction in aggregate
content and size can increase the potential for drying shrinkage. The presence of
aggregate restrains shrinkage of the cement paste, therefore the increase in
aggregate volume reduces drying shrinkage. A decrease in the MSA can
necessitate higher paste volume, thus leading to higher shrinkage. Drying
shrinkage increases with the increase in powder material content, which is
particularly high in SCC.
The use of fly ash in normal proportions does not significantly influence drying
shrinkage of concrete. The use of limestone powder with Blaine fineness greater
than that of portland cement can reduce drying shrinkage of SCC. This can be
explained by the denser matrix obtained when fine limestone powder is used
[Holshmacher and Klug, 2002].
The effect of HRWRA and VMA on shrinkage of SCC is reported to be
beneficial. Indeed, the use of HRWRA reduces the surface tension of the water,
thus decreasing the capillary tension of pore water [Ulm et al., 1999; Acker, 1988;
Acker and Bazant, 1998; Neville, 1981; Wittman, 1976; Neville and Meyers,
1964]. However, the air content may increase when using polycarboxylate-based
HRWRA, which could lead to greater shrinkage. In prestressed applications,
shrinkage should be considered in the mix design and taken into consideration in
the structural design of the member. Proportion SCC with relatively low binder
content and w/cm to reduce drying shrinkage.
Drying shrinkage can be evaluated in accordance with ASTM C 157 (AASHTO T
160). In the absence of measured data, the modified AASHTO 2004 or CEB-FIP
90 shrinkage models can be used to estimate drying shrinkage of SCC, as
indicated below. For steam cured concretes devoid of shrinkage-prone aggregates,
the strain due to shrinkage, sh , at time, t, may be taken as:
AASHTO 2004

53

ks kh

sh

t
55 t

0.56 10

A (steam-cured)

t
26e

ks

0.0142(V / S )

1064 3.70(V / S )
923

t=

t
45 t
drying time (day)

ks =

size factor

kh =

humidity factor

V / S = volume-to-surface ratio

A=

cement factor: 0.918 for Type I/II cement and 1.065 for Type III + 20
percent
FA
binder
which
may be
used
for
P(SCC)

CEB-FIP 90
cso

( f cm )(

RH

(t tc )

2 Ac
350
100
s

( f cm ) [160 10

sc

(9 0.1 f cm )] 10

t tc
6

RH

1.55

ARH

= drying shrinkage (mm/mm)

cso

sc

drying shrinkage obtained from RH-shrinkage chart

= cement type factor

RH

= relative humidity factor

f cm = mean 28-day compressive strength (MPa)

Ac =

cross-sectional area (mm2)

perimeter (mm)

tc =

age at which drying commenced (day)

t=

age of concrete (day).

54

ARH

1 ( RH /100)3

Durability and Air-Void System

It is essential to proportion SCC with adequate stability to ensure high
performance of the hardened concrete, including durability. The durability of a
concrete structure is closely associated to the permeability of the surface layer
and curing. The most significant durability characteristics affecting the durability
of SCC used in precast, prestressed elements production include: w/cm, cement
content, degree of consolidation, curing, cover over the reinforcement, and
reactivity of aggregate-cement combinations.
Bridge structures constructed in environments prone to freezing and thawing may
become critically saturated, thus necessitating air-entrainment when exposed to
cycles of freezing and thawing. In some cases, the bridge deck can shelter some of
the precast beam elements, thus reducing the rate of saturation and exposure to
any deicing salt applied on the bridge deck. In most cases, bridge girders can be
considered to be subjected to moderate exposure conditions that correspond to
situations where deicing salts are not used or where the concrete is only
occasionally exposed to moisture prior to freezing and do not get critically
saturated. Therefore, under these conditions, prestressed bridge girders require
sufficient air content for moderate exposure conditions. For example, under these
conditions, SCC proportioned with in. (12.5 mm) nominal MSA would then
require 4% to 7% air volume in the fresh concrete to provide adequate frost
resistance. ASTM C 457 can be used to test the air-void parameters of the
concrete, and ASTM C 666, Procedure A (AASHTO T 161, method A) is used to
test resistance to freezing and thawing.
Segregation and bleeding have significant negative effect on permeability and
quality of the interfacial zone between cement paste and aggregate, embedded
reinforcement, and existing surface, and hence on durability of the concrete.
Higher air content (6% to 9%) may be necessary in most serve frost environment,
especially when using polycarboxylate-based HRWRA which could result in
some entrapment of relatively large air voids. Coalescence of small air bubbles
during agitation can occur when high air contents are obtained and when concrete
is retempered with water. The dosage of air-entraining admixture (AEA) in SCC
prepared with polycarboxylate-based HRWRA can be quite low compared to
values used for conventional concrete of normal consistency. Still, it is critical to
incorporate an AEA in concrete subjected to moderate frost exposure conditions
to secure stable and closely spaced air bubbles (adequate spacing factor). In
general, for mixtures made with a relatively low content of cementitious materials
and a high w/cm, the air-void stability increases when a VMA is incorporated
[Khayat, 1995].

55

Bond to Prestressing Strands

Ensuring proper stability of SCC is essential to ensure homogenous in-situ
properties, including bond to embedded reinforcement, which is critical for
structural performance of precast, prestressed applications [Moustafa, 1974 and
Logan, 1997]. Bond between the strand and concrete is affected by the
position of the embedded reinforcements and quality of the cast concrete. Bond
to prestressed tendons can be influenced by the flow properties of the SCC,
grading of the aggregate and content of fines in the matrix [Holschemacher and
Klug, 2002].
In general, adequate concrete cover is necessary to properly transfer bond
between prestressed tendons and concrete. Despite the high fluidity of SCC, high
static stability of the SCC after placement can lead to more homogenous in-situ
properties and denser matrix at the interface between the cement paste and
reinforcement, thus enhancing bond strength compared to normal conventional
concrete subjected to mechanical vibration. On the other hand, bond can be
significantly affected by excessive segregation found in poorly designed SCC. As
indicated in Table 10, in order to secure adequate static stability, the SCC should
have maximum surface settlement of 0.5%, column segregation index of 5%, or
percent static segregation of 15%.
A surface settlement of 0.5% corresponds to 1.4 modification factor of
prestressing strands [Khayat et al., 1997; Petrov et al., 2001]. It is important to
note that selection of highly viscous SCC can result in some lack of consolidation
of the concrete, which can in turn affect bond stresses between the concrete and
prestressing strand.
PRODUCTION AND CONTROL
General
The need for adequate quality control is much more critical with SCC than in the
case of conventional concrete. In order to maintain a given workability, it is
essential to maintain constant quality of all concrete constituent during SCC
production. Successful production of SCC requires greater competence and proper
control of materials and equipments used for production.
SCC intended for use in precast plants should meet the technical requirements of
the fresh concrete. The mixture needs to be tested to ensure that required
properties are achieved given the performance specifications, casting conditions,
and geometry of the cast element. Before selecting the raw material and finalizing
the mix design, several factors should be known, including the size and shape of
elements to be cast.

56

Laboratory trials should be used to validate the material selection and verify the
properties of the mix design to achieve the targeted properties. Once the optimum
properties are achieved, proper quality control for material properties should be
observed to eliminate fluctuations in fresh and hardened properties of the
concrete. Any changes in raw materials properties should be immediately
identified to allow necessary adjustments of the mix to meet the specified
properties.
Control of Raw Materials
Depending on the mix design, SCC may be less robust than conventional
concrete. SCC may therefore undergo greater changes in workability given small
variations in the physical properties of its constituents, especially in the moisture
content of the sand, fine particle content in sand, as well as grading and shape of
the sand and coarse aggregate. This would necessitate frequent controls to check
for any changes in material properties that could affect the performance of SCC.
The maximum deviation of the sand moisture should not exceed 0.2% in order
to minimize the variations in fresh properties of SCC. The water content of sand
should be determined just before production of SCC.
Changes in coarse aggregate physical characteristics (shape, texture, gradation)
can affect workability. Inspection at the storage location should be conducted on
coarse aggregate to characterize their physical characteristics for every aggregate
delivery.
The moisture content, water absorption, aggregate gradation, and variations in
fines content of the aggregate should be continuously monitored and must be
taken into account to produce SCC with constant characteristics. Changing the
source of supply for aggregate is likely having significant change to the concrete
properties and should be carefully and fully evaluated [European Guidelines,
2005].
It is preferable to control the moisture of sand before every batch of SCC. The
moisture content in coarse aggregate must be also taken into account and should
be determined at least twice a day, at the beginning of the first and second
production shifts.
Mixing Process and Sequence
The mixing process should be properly determined given the conditions at hand.
For example, the batch volume should be determined in consideration of the type
of SCC (consistency level), efficiency of the mixer to produce a well-dispersed
and homogeneous mix, and transportation rate from the plant to the casting site.

57

Just prior to mixing of the first batch, the mixer should be pre-wetted or
buttered with SCC of approximately similar consistency. Suitable mixing
sequence should be determined given the mixing and storage equipments
available at the plant. Cement particles should be wetted before contact with
HRWRA. Dry mixing before water introduction is not recommended, because it
may lead to build-up of fine materials in the mixer. All batching water should be
added at the same time. For example, it has been shown that the introduction of
VMA at the end of the mixing sequence and of air-entraining admixture at the
beginning can provide good performance [Khayat 1995; Khayat and Assaad,
2002]. The addition sequence of VMA should be evaluated given the mix design
and admixture in use.
The batch volume is typically limited to 80% to 90% of the maximum capacity of
the mixer to allow efficient mixing energy [JSCE, 1999]. When the mixer is
alternatively used for mixing normal concrete and SCC, testing should be
performed to verify that this does not result in any adverse effect on SCC
properties.
Mixing equipment and mixing sequence should be validated by testing
consistency and self-consolidation properties for a given mix design. Necessary
adjustments to time and speed of mixing should be carried out until consistent and
compliant results are obtained.
Transport
SCC should be delivered in a continuous and timely manner to ensure continuous
placement of precast members with the workability-retention period of the
mixture. This is necessary to avoid lift lines and other surface defects. Transport
method shall be confirmed in order to provide SCC at the casting location that is
sufficiently homogeneous to allow successful placement in the precast element
and to achieve the targeted properties. Mixer trucks have proven to be the best
method of delivery of SCC when transporting over rough terrain or long transport
distance [PCI Interim Guidelines, 2003].
Site Acceptance of Plastic Concrete
The producer should determine the frequency of performing quality control
testing based on available experience mixtures [PCI Interim Guidelines, 2003].
The quality control tests should include visual inspection of every batch of the
concrete and any specific tests and compliance parameters. For example, the
slump flow and VSI tests can be adopted. The T-50 can also be run at least once
on new mixtures and used to check the performance in the event of mix
performance problems.

58

Placement Techniques and Casting Considerations

Placement techniques of SCC can have a significant impact on the required
fluidity level and flowing performance of the concrete. For example, in the case
of higher energy involved during placement, lower fluidity level for the SCC will
be required to achieve a given flow and filling performances. Placement
techniques should be selected based on the total volume of the concrete to be
discharged, the transportation rate, and whether the placement process is
continuous or discontinuous. In the case of placement technique involving higher
energy, extra care should be taken with regard to stability characteristics.
Prior to the production process, full-size mock-ups should be cast for final
approval. Placement method should be selected given the production capacity and
transport rate to the casting point.
Relative energy involved during each placement technique is summarized in
Table 11 [PCI Interim Guidelines, 2003]. In a placement case that will require
multiple batches, mixing facilities are required to ensure that concrete will be
available within a short time frame with proper workability characteristics as
specified in performance specifications.
Placement of SCC in horizontal elements can be done by starting at one end of the
mold, with the discharge as close to the form surfaces as possible. It is
recommended to discharge the SCC in the direction of desired flow to maximize
the travel distance. The recommended limit of the flowing distance is to be
limited to 10 and 33 ft (3 and 10 m), depending on the geometry of the element
[RILEM, 2000].
As in the case of conventional concrete, the free fall distance should be controlled
to avoid concrete segregation. For example, based on the Norwegian experience,
the free-fall distance should be limited to 6.5 ft (2 m) when casting wall and beam
elements. Free-fall distance should be fixed given the element depth to be cast and
static and dynamic stability of the concrete.
Temperature Control
The mix design should be tailored to achieve the targeted properties specified in
the performance specifications. When the use of steam curing is required to
achieve the targeted early-age strength, the temperature of the concrete should not
exceed 160oF (71C) [AASHTO LRFD Bridge Construction Specifications,
1998]. Furthermore, according to the AASHTO LRFD Bridge Construction
Specifications [1998], the temperature within the curing chamber shall increase at
a rate not exceeding 72F (22C) per hour.

59

Formwork Considerations and Lateral Pressure

Formwork for SCC can be constructed of different materials, including wood,
steel, plastic, fiberglass, or combination of these materials. Formwork made with
wood often leads to less pores and bubble formation than smooth formwork.
Because of the high fluidity of SCC compared to conventional concrete,
formwork should be rigid enough to accommodate variations in product
dimensions and form, and to withstand lateral form pressure exerted by the plastic
concrete.
Given the high fluidity of SCC compared to conventional concrete, extra care
should be taken to avoid any leakage. Formwork joints should be adequately
sealed. Vegetable oil has been shown to be a good release agent as it reduces the
amount of pores on the concrete surface [Brite-EuRam Report, 1998]. Before
applying the release agent, the wood of the formwork should be dry to ensure
good release performance and avoid appearance of air-bubbles at the formed
surface of the cast element.
Depending on the casting rate and thixotropy of SCC, lateral pressure can be
lower than the theoretical hydrostatic pressure. This is especially the case when
the casting rate exceeds 10 ft/hr (3 m/hr). Lateral pressure can be 50% to 80% of
the calculated pressure for conventional vibrated bridge concrete with a slump
consistency of 4 in. (100 mm) [RILEM, 2000]. Experience has shown that for a
given casting rate, concrete with higher level of thixotropy can develop lower
lateral pressure, faster decay in lateral pressure, and shorter time to pressure
cancellation [Assaad, 2004]. SCC cast at 16 ft/hr (5m/hr) is shown to develop
maximum initial lateral pressure of 90% of hydrostatic pressure. In general,
sections measuring up to 10 ft (3 m) in height should be designed for full
hydrostatic pressure.
Lateral pressure developed by SCC cast from the top of the formwork is lower
than in the case when the concrete is pumped from the bottom. SCC pumped from
the bottom should be designed for full hydrostatic pressure.
Finishing
Finishing of SCC is easier and faster than for conventional concrete. Finishing
practices employed with conventional concrete can be employed with SCC.
However, finishing operations should be delayed slightly more than for
conventional superplasticized concrete [PCI Interim Guidelines, 2003]. Given the
concrete properties and ambient conditions, some surfaces may require only
nominal screeding and floating, while other surfaces may require mild vibratory
screeding [ACI Committee, 237; PCI Interim Guidelines, 2003].

60

Surface drying during finishing should be prevented. Fog misting to increase the
relative humidity would minimize rate of evaporation and reduce the risk of
plastic shrinkage. Because of the relatively higher content of fines and eventual
presence of VMA, SCC mixtures develop little or no bleed water compared to
conventional concrete. It is important to begin the finishing of the surface with
light vibrating screeds, or other manual equipment, as soon as the correct level of
the concrete in the formwork has been reached.
SCC exposed surfaces may dry faster than those of normal superplasticized
concrete. This can happen when casting at hot temperature or windy conditions.
Also, depending on the SCC mixture proportioning, stiffening can increase
rapidly in the period 10-40 minutes after casting. Setting time of the SCC mixture
should be adjusted to allow necessary time to carry out the placement process.
Curing
Membrane curing, matting, foils, or appropriate materials should be left in place
for at least 4 days for cast in-place concrete elements [Swedish Concrete
Association, 2002]. This measure should be applied to SCC with low w/cm and
that made with high fines content or VMA. During hot or windy weather
conditions, moisture should be added by watering or by protecting the surface
with wetted membranes for proper curing.
AASHTO LRFD Bridge Construction Specifications [1998] recommend that for
concrete cured other than by steam or radiant heat methods, whenever there is a
probability of air temperature below 36F (2C) during the curing period, the
concrete shall be maintained at a temperature of not less than 45F (7C) for the
first 6 days after placement. This period must be extended if pozzolans are used as
partial replacement of cement. If the compressive strength of 65% of the specified
28-day design strength is achieved in 6 days, an extended period of controlled
temperature may be waived [AASHTO LRFD Bridge Construction
Specifications, 1998].
Due to the specification in terms of early-age compressive strength, steam curing
or radiant heat curing can be used for precast concrete members. The initial
application of steam or heat shall be from 2 to 4 hours after the initial placement
of concrete to allow the initial set of the concrete to take place [AASHTO LRFD
Bridge Construction Specifications, 1998]. In the case of concrete incorporating a
set retarder, the waiting period shall be increased to between 4 and 6 hours after
casting. During the waiting period, the temperature within the curing chamber
shall not be less than 10C.

61

ACKNOWLEDGMENTS
The authors wish to thank the National Academies of the United States for the
support on NCHRP project 18-12, Report 628. The kindness and generosity of
Dr. Amir N. Hanna, Staff Officer at the Transportation Research Board, is greatly
appreciated.
REFERENCES
Acker, P., and Bazant, Z.P., Measurement of Time-Dependent Strains of
Concrete, RILEM Draft Recommendations. Material and Structures, RILEM,
Vol. 31, No. 8 (1998) pp. 507-512.
Acker, P., Mechanical Behavior of Concrete: Physico-Chemical Approach,
Etudes et Recherches des LPC, 152, Laboratoire Central des Ponts et Chausses,
(1988), 121 p. (In French)
American Association of State Highway Transportation officials (AASHTO),
AASHTO LRFD Bridge Construction Specifications, 1st edition, (1998).
American Association of State Highway Transportation officials (AASHTO),
AASHTO LRFD Bridge Design Specifications, 3rd edition, (2004).
American Association of State Highway Transportation officials (AASHTO),
AASHTO LRFD Bridge Design Specifications, 4th edition, (2007).
American Concrete Institute (ACI) Committee 237, Provisional Guidelines on
Self-Consolidating Concrete, (2007).
American Concrete Institute (ACI) Committee 318, Building Code Requirements
for Structural Concrete (ACI 318-05) and Commentary (ACI 318R-05), (2005).
Assaad, J., Formwork Pressure of Self-Consolidating Concrete Influence of
Thixotropy, Ph.D. Thesis, Department of Civil Engineering, Universit de
Sherbrooke, (2004) 453 p.
Assaad, J., Khayat, K.H., and Daczko, J., Evaluation of Static Stability of SelfConsolidating Concrete. ACI Materials Journal, Vol. 101, No. 3 (2004) pp. 207215.
Attiogbe, E.K., See, H.T., and Daczko, J.A., Engineering Properties of SelfConsolidating Concrete Proceedings, 1st North American Conference on the
Design and Use of SCC, Chicago, Illinois, USA (2002) pp. 371-376.
Bartos, P.J.M., An Appraisal of the Orimet Test as a Method for On-Site
Assessment of Fresh SCC Concrete Proceedings, International Workshop on
Self-Compacting Concrete, Japan (1998) pp. 121-135.

62

Brite-EuRam Project on SCC-BRPR-CT96-0366, Rational Production and

Improved Working Environmental Through Using Self-Compacting Concrete,
(1998).
Bury, M.A., and Bhler, E., Methods and Techniques for Placing SelfConsolidating Concrete An Overview of Field Experiences in North American
Applications, Proc., 1st North American Conference on the Design and Use of
Self-Consolidating Concrete, ACBM, Chicago, Illinois, Nov. 12-13, (2002) pp.
281-286.
Byun, K.J., Kim, J.K., and Song, H.W., Self-Compacting Concrete in Korea
Proceedings, International Workshop on Self-Compacting Concrete, Kochi, Japan
(1998) pp. 23-33.
Chan, Y.-W., Chen Y.-S., and Liu, Y.-S., Development of Bond Strength of
Reinforcement Steel in Self-Consolidating Concrete. ACI Materials Journal,
Vol. 100, No. 4 (2003) pp. 490-498.
Daczko, J., and Kurtz, M., Development of High Volume Coarse Aggregate
Self-Compacting Concrete Proceedings, 2nd International Symposium on SelfCompacting Concrete, Tokyo, Japan (2001) pp. 403-412.
Dehn, F., Holschemacher, K., and Weibe, D., Self-Compacting Concrete (SCC)
Time Development of the Material Properties and the Bond Behavior, Leipzig
Annual Civil Engineering Report No. 5, Universitt Leipzig (2000) pp. 115-124.
European Guidelines for Self-Compacting Concrete, Specification, Production
and Use (2005), 63 p.
Florida DOT, Standard Specifications for Road and Bridge Construction, Florida
State (2004).
Hegger, J., Gortz, S.G., Kommer, B., Tiggs, C., and Dross, C., Prestressed
Precast Beams Made of Self-Compacting Concrete, Betonwerk + FertigteilTechnik (Concrete Plant + Precast Technology), (2003) pp. 40-46.
Holschemacher, K., and Klug, Y., A Database for the Evaluation of Hardened
Properties of SCC, Leipzig Annual Civil Engineering Report No. 7, (2002) pp.
123-134.
Hwang, S.D., Khayat, H.K., and Bonneau, O., Performance-Based Specifications
of Self-Consolidating Concrete Used in Structural Applications. ACI Materials
Journal, Vol. 103, No. 2 (2006) pp. 121-129.
Japan Society of Civil Engineers (JSCE), Recommendations for Self-Compacting
Concrete, Tokyo, (1999) 77 p.

63

Khayat, K.H., Frost Durability of Concrete Containing Viscosity-Modifying

Admixtures. ACI Materials Journal, Vol. 92, No. 6 (1995) pp. 625-633.
Khayat, K.H., Use of Viscosity-Modifying Admixture to Reduce Top-Bar Effect
of Anchored Bars Cast with Fluid Concrete. ACI Materials Journal, Vol. 95, No.
2 (1998) pp. 158-167.
Khayat, K.H., Workability, Testing, and Performance of Self-Consolidating
Concrete. ACI Materials Journal, Vol. 96, No. 3 (1999) pp. 346-353.
Khayat, K.H., and Assaad, J., Air-Void Stability of Self-Consolidating
Concrete. ACI Materials Journal, Vol. 99, No.4 (2002) pp. 408-416.
Khayat, K.H., Manai, K., and Trudel, A., In-Situ Mechanical Properties of Wall
Elements Cast Using Self-Consolidating Concrete. ACI Materials Journal, Vol.
94, No. 6 (1997) pp. 491-500.
Koning, G., Holschemacher, K., Dehn, F., and Weibe, D., Self-Compacting
Concrete-Time Development of Materials Properties and Bond Behaviour
Proceedings, 2nd International Symposium on Self-Compacting Concrete, October
23-25, Tokyo (2001) pp. 507-516.
Logan, D., Acceptance Criteria for Bond Quality of Strand for Pretensioned
Prestressed Concrete Applications, PCI Journal, Vol. 42, No. 2, (1997) pp. 5290.
Manai, K., Evaluation of the Effect of Chemical and Mineral Admixtures on the
Workability, Stability, and Performance of Self-Compacting Concrete, Masters
Thesis, University of Sherbrooke, Quebec, Canada (1995), 182 p. (in French)
Moustafa, S., Pull-out Strength of Strand and Lifting Loops, Concrete Technology
Associates Technical Bulletin, 74-B5, (1974).
National Cooperative Highway Research Program (NCHRP), Self-Consolidating
Concrete for Precast, Prestressed Concrete Bridge Elements, NCHRP Report
628, Transportation Research Board of The National Academies, (2009).
Neville, A.M., and Meyers, B.L., Creep of Concrete: Influencing Factors and
Prediction, Symposium on Creep of Concrete. SP-9, ACI, Detroit (1964) pp. 133.
Neville, A.M., Properties of Concrete, 3rd Edition, Pittman, London (1981), 779 p.
Okamura, H., Self-Compacting High-Performance
International, Vol.19, No. 7 (1997) pp. 50-54.

Concrete.

Concrete

Ozawa, K., Tangtermsirikul, S., and Maekawa, K., Role of Powder Materials on
the Filling Capacity of Fresh Concrete Supplementary Papers, 4th

64

CANMET/ACI International Symposium on Fly Ash, Silica Fume, Slag, and

Natural Pozzolans in Concrete, Istanbul (1992) pp. 121-137.
Petrov, N., Khayat, K.H., and Tagnit-Hamou, A., Effect of Stability of SelfConsolidating Concrete on the Distribution of Steel Corrosion Characteristics
Along Experimental Wall Elements Proceedings, 2nd International Symposium
on Self-Consolidating Concrete, Tokyo, Japan (2001) pp.441-450.
Pons, G., Proust, E., and Assi, S., Creep and Shrinkage of Self-Compacting
Concrete: A Different Behaviour Compared with Vibrated Concrete
Proceedings, 3rd International RILEM Symposium on Self-Compacting Concrete,
August 17-20, Reykjavik, Iceland (2003) pp. 645-654.
Precast/Prestressed Concrete Institute, Interim Guidelines for the Use of SelfConsolidating Concrete in Precast/Prestressed Institute Member Plants, (2003),
148 p.
Precast/Prestressed Concrete Institute, Precast Prestressed Concrete Bridge
Design Manual, 1st Edition (1997).
State-of-the-Art Report of RILEM Technical Committee 174-SCC, SelfCompacting Concrete, Report 23, Edited by A. Skarendahl, O. Petersson, (2000),
154 p.
Swedish Concrete Association, Self-Compacting Concrete, Recommendations for
Use, Concrete report No. 10(E), (2002), 84 p.
Ulm, F.-J., Le Maou, F., and Boulay, C., Creep and Shrinkage Coupling: New
Review of Some Evidence Creep and Shrinkage of Concrete, Edited by Ulm, F.J., Prat M., Calgaro, J.-A., Carol, I., Special issue of Revue Franaise de Gnie
Civil, HERMES Science Publications (1999) pp. 21-37.
Wittman, F.H., On the Action of Capillary Pressure in Fresh Concrete. Cement
and Concrete Research, Vol. 6, No. 11 (1976) pp. 49-56.
Yurugi, M., Sakata, N., Iwai, M., and Sakai, G., Mix Proportion of Highly
Workable Concrete Concrete 2000, Dundee (1993) pp. 579-589.

65

Table 1 Grading requirements for fine aggregates

Sieve
3/8 in. (9.5 mm)
No. 4 (4.75 mm)
No. 8 (2.36 mm)
No. 16 (1.18 mm)
No. 30 (600 m)
No. 50 (300 m)
No. 100 (150 m)

Percent passing
(AASHTO M 6)
100
95 to 100
80 to 100
50 to 85
25 to 60
10 to 30
2 to 10

Percent passing
(ASTM C 33)
100
95 to 100
80 to 100
50 to 85
25 to 60
5 to 30
0 to 10

Table 2 Key workability characteristics of SCC

Test methods

Applicable standard

Filling
ability

Slump flow and T-50

ASTM C 1611

Passing
ability

L-box
J-Ring
V-funnel

ASTM C 1621

Filling
capacity

Combining filling and passing abilities

Static
stability

Surface settlement
Column segregation
Visual Stability Index

66

ASTM C 1610
ASTM C 1611

Table 3 Advantages and precautions of slump flow test and T-50

Slump flow
Advantages
o Simple
o Reproducible
o Results correlate to yield
stress
o Low sensitivity to water
content
o Can be performed by a single
operator
o Results correlate to plastic
viscosity
o Can be performed
simultaneously with slump
flow using a second operator

Precautions
o Roughness and moisture of base plate
affect results
o Large base plate is required to perform
test
o Must be performed on level surfaces

T-50
o Sensitive to roughness and moisture of
base plate
o Poor single- and multi-operator
repeatability
o High error for low viscosity mixtures

Table 4 Advantages and precautions of J-Ring, L-box, and V-funnel flow

tests
J-Ring
Advantages
o Simple
o Good repeatability
o Can be performed by a single
operator
o Material segregation can be
visually detected
o
o
o

o
o

Precautions
o Roughness and moisture of base
plate affect results
o Large base plate is required to
perform the test
o Must be performed on level
surfaces
L-box
Good repeatability
o Must be performed on level
Can be performed by a single
surfaces
operator
Flow time correlates to plastic
viscosity
V-funnel
Can be performed by a single
o Poor repeatability
operator
o Risk of flow interruption in high
Flow time correlates to plastic
viscosity mixtures
viscosity

67

Table 5 Advantages and precautions of filling capacity test

Filling capacity
Advantages

Precautions

o Good repeatability
o Good indicator of filling capacity, which
combines filling ability and passing ability
of SCC
o Visual appreciation of filling capacity
through congested sections

Difficult to perform by
single operator

Requires some
calculation to evaluate
filling capacity

Table 6 Advantages and precautions of surface settlement and column

segregation tests

o
o
o

o
o

Surface settlement test

Advantages
Precautions
Easy to perform in laboratory
o Requires a dial gage
Good repeatability
o Difficult to perform by a
Maximum settlement can be
single operator
estimated from rate of settlement
o Requires large amount of
between 25 and 30 min
concrete
Column segregation test
Good correlations between column
o Requires electronic balance
Iseg and S
o Requires large amount of
concrete
o Difficult to perform by a
single operator
o Repeatability lower than
settlement
Visual stability index
Simple
o Depends on operator
Can be performed by a single operator
experience
o SCC with low VSI may still
exhibit some lack of stability

68

Table 7 Advantages and limitations of rheometer testing

Rheometer
o
o
o
o

Advantages
Easy to perform in laboratory
Good repeatability, especially for
plastic viscosity
Provides fundamental flow
properties of SCC
Enables evaluation of structural
build-up of SCC at rest

Limitations
o Expensive apparatus, though
portable and more affordable
models are available
o Requires qualified personnel to
operate and interpret data

Property

Test method

Target values

Design

Q/C

Table 8 Recommended workability characteristics for mix design and Q/C

testing at precasting plant

Filling
ability

Slump flow
T-50 (ASTM C 1611)
J-Ring flow
(ASTM C 1621)
L-bx blocking ratio
(h2/h1)
Filling capacity
Slump flow and
J-Ring tests
Slump flow and
L-box tests

23.5 29 in. (600 735 mm)

1.5 6 sec

21.5 26 in. (545 660 mm)

0.5 1.0

70% 100%

Passing
ability

Filling
capacity

Static
stability

Surface settlement

Column segregation
(ASTM C 1610)
VSI (ASTM C 1611)
Air
volume

AASHTO T 152

Rate of settlement, 25 30 min (value can

decrease to 10 15 min)
MSA of 3/8 and in. (9.5 and 12.5 mm)
0.27 %/h
(Max. settlement 0.5%)
MSA of in. (19 mm) 0.12%/h
(Max. settlement of 0.3%)
Column segregation index (C.O.V.) 5%
Percent static segregation (S) 15%
0 1 (0 for deep elements)
4% 7% depending on exposure
conditions, MSA, and type of HRWRA.
Ensure stable and uniform distribution of
small air voids.

69

Table 9 Workability values of SCC used in precast/prestressed applications

J-Ring
(Slump
flow
J-Ring
flow)
(ASTM
C1621)

Slump flow
(ASTM
C1611/C16
11M-05 )

Element characteristics

Short
Moderate

Length

Long
Thin
Moderate

Thickness

Thick
Low
Medium
High

Coarse
agg.
content

70

90%

Deep

75%-90%

Depth

Caisson
filling
capacity

70%-75%

Moderate

0.7

Shallow

0.6-0.7

Congested

Element
shape
intricacy

0.5-0.6

Moderate

2 in.

Small

Reinforcement
density

2-3 in.

High

3- 4 in.

Medium

27.5-29 in.

Low

25-27.5 in.

23.5-25 in.

Relative
values

L-box
blocking
ratio (h2/h1)

Table 10 Recommendations to secure homogenous in-situ properties of

SCC
Material properties
Static stability

Viscosity

Mechanical properties

Recommended values
Maximum surface settlement 0.5%
Column segregation index (Iseg) 5%
Percent static segregation (S) 15%
Plastic viscosity 0.0725 psi.s (500 Pa.s)
(Modified Tattersall two-point rheometer with
vane device)
Core-to-cylinder compressive strength 90%
(similar curing conditions)
Bond strength modification factor 1.4

Table 11 Summary of different placement techniques for SCC [Bury and

Buhler, 2002]
Placement technique
Truck discharge
Pumping
Crane and bucket
Auger (Tuckerbuilt)
discharge

Discharge
rate

Discharge
type

High
Medium/
High
High
Low/
Medium

Continuous

Single
discharge
volume
High

Continuous

Medium

Discontinuous

Low

Continuous

Medium

71

Relative
energy
delivered
High
High/
Medium
Medium
Low/
Medium

72

FRIENDLIER SELF-COMPACTING CONCRETE THROUGH

SMART DYNAMIC CONSTRUCTION
B. Barragan, J. Roncero R. Magarotto, S. Moro, and R. Khurana

Synopsis: The use of Self-Compacting Concrete (SCC) in the precast industry

has increased significantly from its introduction in the market, in the mid 1990s.
It is estimated that, in Europe, more than 50% of the concrete used for structural
precast elements is SCC. However, the situation is quite different in the readymixed concrete industry where less than 1% of the concrete produced is SCC. The
reasons of such scenario are mostly related to the higher cost of the material,
which, very often, is supplied with excessive mechanical performance due to the
higher cementiticious paste content necessary to achieve self-compactability
requirements.
It has been demonstrated that the introduction of the Smart Dynamic Concrete
(SDC) concept, associated to the production of SCC with low fine contents by
making use of an innovative viscosity modifying admixture (VMA), can allow the
ready-mix concrete industry to increase the energy efficiency, achieve a higher
concrete durability and accomplish more rational and economical construction
processes, reducing CO2 emissions, and also increasing productivity.
This work presents and discusses the optimization of normal strength SCC mixes
(25-35 MPa) with low content of fines through the application of the SDC
concept, and corresponding field test results from experiences in different readymix and precast concrete plants. The analysis also includes the evaluation of the
material robustness.

Keywords: viscosity-modifying admixture, self-compacting concrete, smart

dynamic concrete, rheology

73

Bryan E. Barragan holds a Doctoral Degree in Civil Engineering from the

Universitat Politcnica de Catalunya (Barcelona, 2003). Since 2008 he is a
concrete technology research specialist of the Technology and Development team
of BASF Construction Chemicals- BU Admixture Systems Europe, Treviso, Italy.
Has co-authored more than 100 technical publications in different fields of
concrete technology, largely on fiber-reinforced concretes and self-compacting
concretes.
Joana Roncero is graduated in Chemistry in 1994 in University of Barcelona and
received Ph.D degree in 2000 in the Civil Engineering School of the Technical
University of Catalunya in Spain. She has been working in concrete technology
since 1995. Since 2004 she is head of the Technology and Development Center in
Spain of the European Business Unit Admixture Systems of BASF Construction
Chemicals. Her research interests include admixture development and utilization
concrete technology self-compacting concrete and high performance concretes.
Roberta Magarotto is graduated in Industrial Chemistry at Venice University
(Italy). Since 1993, she has been working in the field of concrete admixtures
focusing in the development of new polymeric superplasticizers and the study of
cement-admixtures interaction. From 2002 to 2007 she had been head of the
Admixture Department of R&D at BASF Construction Chemicals Italy. Since
2008, she has been head of the Technology and Development of the European
Business Unit Admixture Systems of BASF Construction Chemicals (E-EBA/D)
Sandro Moro is graduated in Sciences and Technology of Materials at Venice
University (Italy). Since 2001 he is Head of the Technological Laboratory at
BASF Construction Chemicals Italia Spa, where he has been working in the field
of concrete admixtures development and of concrete technology development.
Rabinder S. Khurana obtained his Post-graduated Diploma in Reinforced
Concretes Structures from Milan Polytechnic Italy. He has more than 30 years of
experience in concrete technology and admixtures application world wide.

INTRODUCTION
Self-consolidating concrete (SCC) is undoubtedly the major advance of
concrete technology in the last two decades. Mainly due its flowability, SCC
differs from conventional vibrated concrete essentially when in fresh state.
Guidelines and Standards
Currently, a number of guidelines are available for the proportioning,
characterization, control and application of SCC [e.g. 1-7] which are accompanied

74

by national standard test methods [e.g. 8-13] and, in some countries, like Spain,
by code specifications [14].
The recommendations by the Japanese Society of Civil Engineers [1] define three
types of SCCs according to the use of fines (type P), viscosity modifying agents
(type V) or the combined used of both (type C). In type P, self-compactability is
obtained by minimizing the ratio water/fines through the use of superplasticizers
to provide appropriate resistance to segregation and bleeding. In the case of type
V, the use of Viscosity Modifying Admixtures (VMA) provides enough stability
to the concrete towards bleeding and segregation, since these admixtures fix the
water providing the appropriate rheology to the cement paste. Type C consists of
the combination of the other two types.
On the other hand, the European Guidelines [4] define different types of SCCs
according to its fresh properties, regardless the composition of the concrete.
Along these lines, there are three levels of self-compactability according to the
value of slump flow:
SF1: from 550 to 650 mm (21.7 to 25.6 in.)
SF2: from 660 to 750 mm (26.0 to 29.5 in.)
SF3: from 760 to 850 mm (29.9 to 33.5 in.);
two levels of viscosity according to the value of T50 or the time to pass through
the V-funnel, Tv:
VS1/VF1: T50 2 s or Tv 8 s
VS2/VF2: T50 > 2 s or 9 s Tv 25 s
and two levels of passing ability according to the passing ratio in the L-box:
PA1: 0.80 with 2 bars
PA2: 0.80 with 3 bars.
In spite of the fact that these two guidelines [1,4] involve the use of a wide range
of SCC both considering its composition, as well as its fresh properties, the reality
of the market indicates that SCC is mainly characterized by high content of fines,
usually in the range from 450 to 600 kg/m3 (28,1 to 37,5 lb/ft3), along with a low
volume of coarse aggregates. This high content of fines, which provides high
volume of cementitious paste (i.e., within 35-40% by volume), is needed in order
to provide flowability, passing-ability, and enough cohesion able to prevent
bleeding and/or segregation.
SCC rheology
Though historically with only academic interest, rheological studies using
concrete viscometers have become more important with the appearance of SCC,

75

up to the point that the material itself can be defined and characterized in terms of
its rheological parameters, such as the yield stress ( 0) and plastic viscosity ()
[15]. The rheological studies have permitted a deeper comprehension of the
material behavior, allowing in this way a more rational mixture proportioning.
SCC is characterized by having very low yield stress ( 0 ) and enough plastic
viscosity (), which redounds in the material flowability, passingability, and
segregation resistance during transport, casting and hardening; the engineering
properties of the material. In this manner, SCC requires an adequate combination
of 0 and to allow for mobility without segregation. Generally, very viscous
mixes require a 0 value close to zero, while a higher value of such parameter can
be convenient in the case of low viscosity mixes. A combination of extremely low
0 and could imply risks of segregation [15].
The composition of SCCs varies significantly from one country to another.
According to [16], the content of fines in Sweden and The Netherlands is usually
higher than 550 kg/m3 (34,3 lb/ft3) with yield values and plastic viscosity values
ranging from 0 to 30 Pa and from 50 and 120 Pas, respectively. These ranges
agree with those shown in a recent study by [16]. In Denmark, Norway and
Iceland, the content of fines is usually lower than 450 kg/m3 (28,1 lb/ft3) with
yield values and plastic viscosity ranging from 10 to 60 Pa and from 20 to 45 Pas,
respectively. Along these lines, the concrete with higher content of fines is
characterized by a higher plastic viscosity and is suitable for heavily reinforced
structures, whereas the concrete with lower content of fines show higher yield
values being more suitable for lightly reinforced structures.
Even when the use of rheometers or viscometers leads to a better characterization
of conventional concrete and SCC [17], their use is usually more expensive and
difficult in field applications. Progress is being made to address this issue and
some portable equipment has been recently developed [18,19]. In this way, the
relationships between rheological parameters and results from engineering type
tests turn to be of great practical importance. In this regard, it is shown in [20] that
SCC prepared with similar types of component materials show direct relationships
between the slump flow diameter and the yield stress and also between the plastic
viscosity and the flow times (T50 or TV), even when the concrete temperature, the
mixing energy, the environmental conditions or the time after mixing was varied,
Fig. 1.
The actual situation- The use of SCC in the precast industry has increased
significantly from its introduction in the market, in the mid 1990s. It is estimated
that, in Europe, more than 50% of the concrete used for structural precast
elements is SCC. However, the situation is quite different in the ready-mixed
concrete industry where less than 1% of the concrete produced is SCC. The
reasons of such scenario are mostly related to the higher cost of the material,

76

which, very often, is supplied with excessive mechanical performance due to the
higher cementiticious paste content necessary to achieve self-compactability
requirements. On the other hand, the use of additional fines (fillers) can also
imply new silos and more quality control testing that, along with the elevated
cement content, and the higher sensitiveness of SCC to variations of the mix
proportions, can consequently lead to an increase in production costs.
Evidently, it is technically difficult to prepare SCC with low content of fines
(cement + filler), even using currently available VMAs. On the other hand, the
characteristics of locally-available materials (i.e., cement, aggregates and filler)
for SCC production are not always the most appropriate, and consequently, a lot
of laboratory work is needed in order to adjust the mixture proportions with the
aim of obtaining stable concrete. Also, variations in the moisture content of the
aggregates, especially of the sand, can significantly influence the stability of the
SCC, implying continuous adjustments in the mixture proportions.
Consequently, the use of SCC as a common every-day concrete is nowadays
difficult, especially in the case of the ready-mixed concrete industry. Note that,
the required mechanical performance in most of the applications is significantly
lower than that provided by standard SCC (i.e., with high amount of fines) and,
therefore, SCC is supplied, very often, with suboptimal performance.
An innovative alternative
The above reality justifies the development of an innovative VMA, which is
allowing the design of SCC with the required stability (bleeding and segregation
resistance) but without incorporating extra fines and, consequently, having a total
content of fines (cement + filler) of about 380 kg/m3. The use of this new family
of VMAs, along with superplasticizers especially designed for SCC, leads to the
development of smart dynamic concrete [21-23], in other words, to an
improvement in the cost-effectiveness of SCC that can contribute expanding its
use in the ready-mixed concrete industry, as well as consolidate its use in the
precast concrete industry.
This study presents the results of industrial-scale testing in different European
countries, aimed at achieving the following objectives: proportioning SCC with
normal cement contents for everyday concrete, study the possible upgrade of
conventional concretes of high workability into concretes with self-compacting
characteristics, and the possibilities of replacing/reducing the filler content of
reference SCC maintaining the compressive strength and its evolution.
Results demonstrate the usefulness of the new VMA to overcome the common
drawbacks of current type of admixtures, and its capacity to increase the mix
robustness and decrease the necessity of filler materials, if desirable. Such
powerful capacities could certainly help self-compacting concrete increase its

77

market share in the ready-mix sector and further consolidate its use in precast
applications, leading to more productive, environmentally friendly, optimized and
smart construction processes.
MECHANISM OF ACTION
The preparation of SCC with low fines content requires the optimum control of
the rheological parameters providing the appropriate balance between the high
fluidity needed and the stability required (bleeding and segregation). In this
respect, a VMA able of assuring enough plastic viscosity without sacrificing the
slump flow of the concrete, that is, without modifying significantly the yield
value, is essential.
Some state of the art VMAs belonging to different chemical families (natural
gums, starches and synthetic polymers) modify the rheology of concrete mainly
by increasing the yield value without providing significant improvement of the
plastic viscosity as can be seen in Fig. 2. Note that these tests were done in
mortars from concretes that were sieved at 2 mm and tested with a Viskomat NT
rheometer from Schleibinger Gerte. Note, also, that the dosage of each VMA
was adjusted in order to obtain a slump flow of about 65 cm in each concrete.
Moreover, in other cases, the VMA does not affect the yield value but provides a
slight increase in the plastic viscosity. Note, additionally, that in other cases (see
Natural Gum A and B), both the yield value and the plastic viscosity are not
affected when compared to the reference mortar. Nevertheless, as can be seen in
the figure, the new and innovative VMA is able to increase significantly the
plastic viscosity of the mortar without influencing the yield value and,
consequently is able to provide stable concrete without affecting significantly its
slump-flow.
This innovative chemical admixture consists in a new ionic polymer of high
molecular weight whose molecules interact between them apart from interacting
with the water molecules. This leads to a tridimensional structure that embraces
both water molecules and fine solid particles (cement and the fines belonging to
the sand). This tridimensional network made up, not only by VMA-water
interactions, but also by VMA-VMA and VMA-solid particles interactions as
shown in Fig. 3, leads to a strong interconnected molecular network able to
provide, at the macroscopical level, enough stability in the concrete.
The use of this innovative VMA provides the appropriate stability in the concrete
towards segregation and/or bleeding in concretes with paste volume significantly
lower than those used usual for SCC. Along these lines, the rheological behavior
of this type of concretes depends on the optimized mixture proportions of the
concrete, as well as the use of chemical admixtures of last generation.

78

Accordingly, high-performance polycarboxylate-based superplasticizers are

needed to modify and control the yield value (0) of the concrete, whereas, the
plastic viscosity () is optimized through the use of the new VMA.
MIXTURE PROPORTIONING
The most important characteristic of smart dynamic concrete lies in its
composition. In this way, the amount of cement and water will be fixed by the
mechanical and durability requirements, since its composition is quite similar to
that of conventional vibrated concrete of similar strength class. The selfcompacting ability is obtained by means of using the latest generation
superplasticizers with high-water reducing ability, the new and innovative VMA,
as well as the optimization of the amount and quality of the mortar. Therefore, in
comparison to conventional vibrated concrete, SCC with low fines content has
higher proportion of sand ( < 4 mm, 0.16 in.). This is needed in order to ensure
the appropriate transport of the coarser aggregates during placing and decreasing
their risk of segregation.
EXPERIMENTAL PROGRAM
The experimental program involves a wide industrial-scale test program aimed to
study the possibilities of the new VMA to provide sufficient stability yet not
compromising the flowability, while keeping the fine contents at a minimum.
These industrial tests were carried out in several European countries (Germany,
Italy, Spain and Turkey) focusing on obtaining concrete with self-compacting
properties but with reduced fines content in comparison to standard SCC.
The following sections present the results of the experimental program aiming to
achieve the following three specific objectives: proportioning SCCs with normal
cement contents for everyday concrete, study the possible upgrade of
conventional concretes of high workability into robust concretes with selfcompacting characteristics, and the possibilities of replacing/reducing the filler
content of reference SCC maintaining the compressive strength and its evolution.
The material characterization in fresh state included the slump-flow test, and the
determination of the unitary weight, in some cases. In more qualitative terms, the
general material appearance, the stability after the slump-flow test (bleeding or
paste halo), and stickiness, was evaluated.

79

Decrease of the cement content to obtain conventional compressive strengths

but maintaining the self-compacting properties
The main objective of these industrial tests consists in minimizing the cement
content of a reference SCC with the objective of obtaining compressive strengths
in the range of 25-35 MPa (3600-5000 psi). Industrial tests in two different
Spanish plants were performed (plant A and B).
In plant A, the reference SCC has a composition based on 412 kg/m 3 (25.7 lb/ft3)
of cement without filler, and crushed limestone aggregates, as can be seen in
Table 1 (Reference-A). As described, the objective of the tests consists in
obtaining a cost-efficient mixture with self-compacting properties and lowmedium compressive strengths.
In this experiences, the use of the innovative VMA has permitted decreasing the
cement content to 310 kg/m3 (19.4 lb/ft3), still maintaining the stability of the
concrete towards bleeding and segregation. The composition and properties of
this concrete (SDC-A) is also shown in Table 1. Note, also, that some fly ash
belonging to the cement appears in the surface of the reference concrete.
However, the presence of fly ash in the surface of the concrete was avoided in the
concrete with the new VMA, denoting high stability. It is important also to
highlight the appearance of the concrete with lower fines content (SDC-A), which
is less sticky and easy to handle than the reference SCC with higher fines content.
The appearance of the concretes described in Table 1 is shown in Fig. 4.
A similar experience was done in plant B where, as in the previous case, the
cement content was decreased in comparison to a reference SCC with the aim of
obtaining SCC with compressive strength in the range of 25-30 MPa (3600-4350
psi). The dosage of the reference concrete, with 400 kg/m3 (25.0 lb/ft3) of cement,
is shown in Table 2.
In this case, the cement content was decreased 50 kg/m3 (3.1 lb/ft3) through the
incorporation of the new VMA, maintaining the stability of the concrete as shown
in Table 2 (SDC-B). The appearance of both concretes is shown in Fig. 5. As in
the previous case of plant A, this concrete has a less sticky appearance in
comparison to the reference SCC (Reference-B).
Upgrading of conventionally vibrated concretes to higher performance level
with self-compacting characteristics
The objective of these field tests consist in obtaining concretes with selfcompacting properties but with the same ease of production and robustness as
conventional vibrated concrete, at minimum extra cost. The results of tests
performed in three different ready-mixed concrete plants (C, D and E) are
presented.

80

In plant C from Italy, the reference concrete has a consistency class S4 and
exposure class XC2 (characteristic compressive strength of 30 MPa or 4350 psi)
according to EN 206-1. The durability class fixes the amount of cement and the
water/cement needed. The mixture proportions of the reference concrete are
shown in Table 3 (Reference-C). Accordingly, 800 m3 (1046 yd3) of concrete with
the composition shown in Table 3 (SDC-C) were prepared in series of 10 m3
(13 yd3). The mixing was done directly in the truck since no mixer was available
in the plant. The chemical admixtures (superplasticizer and the new VMA) were
incorporated along with the mixing water. The concrete was transported one hour
before placing (Fig. 6), and the temperature at the job site was of 30C (86F).
In plant D from Turkey, the reference concrete was a conventional concrete with
characteristic compressive strength of 25 MPa (3600 psi) and S3 consistency that
had to be vibrated. The superplasticizer used was naphthalene-based. The
composition of this reference concrete, as well as the composition of the concrete
with the new VMA, is shown in Table 4. In this case, 8 m3 (10 yd3) of concrete
SDC-D with self-compacting characteristics were prepared in batches of 2 m3 (2.6
yd3) that were pre-mixed before discharging in the truck. The concrete, which
shows a slump flow of 71 cm (27.9 in.), was transported throughout 90 minutes at
38C (100F). After this period, the slump flow was measured again and was of 60
cm (23.6 in.).
Finally, in plant E from Germany, the corresponding reference mixture
proportioning was optimized and upgraded to self-compacting concrete. The
reference concrete at this plant had a consistency of type F5 according to DIN
18218 with a composition shown in Table 5, and need vibration. In this case,
24 m3 of concrete SDC-E, as described in Table 5, was prepared. The concrete
was transported for 30 minutes before being placed (Fig. 7).
In these cases, the use of the new VMA allowed transforming conventional
vibrated concretes of S3, S4, and F5 classes into truck-mixed medium-range selfcompacting concretes, capable of maintaining its self-compactability during
realistic periods of time, even at elevated temperatures. The robust SCCs achieved
also maintained the reference compressive strength.
Replacement of the filler used in a reference self-compacting concrete
maintaining the strength development
The objective of this test program was to eliminate completely the filler used in
the mixture proportioning of a reference SCC and to propose an alternative
composition able to provide similar strength development at plant F, in Spain.
This would permit to free a silo that is currently used to store filler and,
consequently, to improve the logistics at the plant. The sand available was natural
siliceous sand with about 1% of fine content ( < 0.125 mm or 0.005 in.). The

81

composition of the reference SCC is shown in Table 6. The cement and filler
content were of 330 and 170 kg/m3 (20.6 and 10.6 lb/ft3), respectively.
The use of the innovative VMA permitted to prepare a new mixture proportioning
(SDC-F) with similar strength development using 400 kg/m3 (24.9 lb/ft3) of
cement. The fines content (cement + filler) was reduced in 100 kg/m3 (6.2 lb/ft3)
compared to the reference SCC (Reference-F). Both concretes had similar
appearance (Fig. 8).
CONCLUSIONS
From the range of materials and conditions studied, the following conclusions can
be drawn from the results obtained:
The new and innovative viscosity modifying admixture tested in this study
permits obtaining concretes with self-compacting properties but with fines content
(cement + filler) lower than 380 kg/m3. This can contribute to extend the use of
SCC in the ready-mixed concrete industry as an everyday concrete and to extend
its use in the precast concrete sector.
Such VMA provides enough cohesion and stability in the concretes by increasing
its plastic viscosity without modifying significantly the yield value.
The concretes incorporating the new viscosity modifying admixture show
appropriate robustness toward changes in the composition, especially water
content, as well as variations in the characteristics of the materials. Allows a)
proportioning SCCs with normal cement contents for everyday concrete, b)
upgrading conventional concretes into robust concretes with self-compacting
characteristics, and c) replacing/reducing the filler content of SCC maintaining
the compressive strength and its evolution.
The achieved smart dynamic concretes with medium-range self-compactability
can be truck-mixed and are capable of maintaining its self-compacting
characteristics during realistic periods of time, even at elevated temperatures.
REFERENCES
1. Japan Society of Civil Engineers (JSCE), Recommendation for SelfCompacting Concrete, Tokyo, Japan, August 1999.
2. Swedish Concrete Association, Report No. 10 (E) Self-Compacting ConcreteRecommendations for Use, Stockholm, Sweeden, August 2002.

82

3. Precast/Prestressed Concrete Institute (PCI), Interim Guidelines for the Use of

Self-Consolidating Concrete in Precast/Prestressed Concrete Institute Member
Plants, Chicago, USA, April 2003.
4. European Project Group, The European Guidelines for Self-Compacting
Concrete. Specification, Production and Use (joint work by EFNARC, BIBM,
CEMBUREAU, EFCA, ERMCO), May 2005.
5. The Concrete Society, Technical Report No. 62 Self-Compacting Concrete- a
Review, Surrey, UK, 2005.
6. American Concrete Institute (ACI), 237R-07 Self-Consolidating Concrete,
Farmington Hills, USA, April 2007.
7. Spanish Scientific-Technical Association for Structural Concrete (ACHE),
Self-Compacting Concrete- Design and Application, Spain, September 2008. (in
Spanish)
8. ASTM C 1611/ C 1611M- Test Method for Slump Flow of Self-Consolidating
Concrete. USA 2005.
9. ASTM C 1621/ C 1621M- Test Method for Passing Ability of SelfConsolidating Concrete by J-Ring. USA 2006.
10. ASTM C 1610/C 1610M- Test Method for Static Segregation of SelfConsolidating Concrete Using Column Technique. USA 2006.
11. UNE 83363. Self-Compacting Concrete. Characterization of flowability by
the L-box method, Spain 2007. (in Spanish)
12. UNE 83364. Self-Compacting Concrete. Determination of flow time by the Vfunnel test, Spain 2007. (in Spanish)
13. UNI 11044- Test for fresh Self-Compacting Concrete. Determination of
confined flow by the U-box test, Italy 2003. (in Italian)
14. Spanish Structural Concrete Code (EHE). Ministry of Public Works, Madrid,
1999. (in Spanish)
15. Wallevik, O.H., Rheology - A Scientific Approach to Develop SelfCompacting Concrete, Proceedings of the 3rd International RILEM Symposium

83

on Self-Compacting Concrete, RILEM PRO33 Reykjavik, Iceland, 2003, pp. 23

31.
16. Wallevik, O.H., Why is Self-Compacting Concrete Different from Country to
Country, Plenary Session 2nd North Am Conf. on the Design and Use of SelfConsolidating Concrete (SCC) and Fourth Int RILEM Symp on Self-Compacting
Concrete, Chicago, USA, 2005
17. Ferraris C., de Larrard F., Martys N. Fresh Concrete Rheology: Recent
Developments, in Skalny J, Mindess S (eds) Materials Science of Concrete V,
American Ceramic Society, USA, p 215-241, 1998
18. Koehler E.P., Fowler D.W. A portable rheometer for Self-Consolidating
Concrete, in Shah SP (ed), 2nd North Am Conf. on the Design and Use of SelfConsolidating Concrete (SCC) and Fourth Int RILEM Symp on Self-Compacting
Concrete, Hanley Wood Pub, Addison, IL, USA, p 693-699, 2005
19. Koehler E.P., Fowler D.W., Ferraris C.F., Amziane S. A New, Portable
Rheometer for Fresh Self-Consolidating Concrete, in Shi C, Khayat KH (eds)
ACI SP-233-7, p 97-115, 2006
20. Zerbino, R., Agull, L., Barragan, B., Garcia, T. and Gettu, R. Workability
tests and rheological parameters in self-compacting concrete, Materials and
Structures, V.42, I.7, pp. 947-960, 2009
21. Roncero, J., Corradi, M. and Khurana, R.S, New admixture-system for lowfines self-compacting concrete. Proc. 5th Intnl. RILEM Symp. on SelfCompacting Concrete - SCC2007, Ghent, Belgium, Vol. 2, pp. 875-880, 2007.
22. Khurana, R.S., Magarotto, R., Moro, S. and Roncero, J. Smart Dynamic
Concrete: an innovative approach for the construction industry, Role for
Concrete in Global Development (Proc. of 7th International Congress Concrete:
Construction's Sustainable Option, July 2008), Ed. R. K. Dhir, P.C. Hewlett, L.
Csetenyi, M. D. Newlands, IHS, bepress, pp. 725-734, 2008
23. Roncero, J., Corradi, M., Khurana, R.S., Magarotto, R. and Moro, S. Smart
Dynamic Concrete: New approach for the Ready-Mixed Industry, SCC 2008:
Challenges and Barriers to Application, 3rd North-American Conf. On the Design
and Use of Self-Consolidating Concrete, Chicago,USA, 2008

84

Table 1 - Composition and properties of reference SCC and smart dynamic

concrete with lower fines content in plant A
Composition in kg/m3 (lb/ft3)
0-2 mm sand
0-4 mm sand
4-12 mm gravel
CEM II/B-M 42.5R*
Water
Superplasticizer (Glenium
type)
New VMA
Properties
Slump flow

Reference-A
330 (20,6)
760 (47,4)
700 (43,7)
412 (25,7)
175 (10,9)

SDC-A
350 (21,8)
840 (52,4)
700 (43,7)
310 (19,4)
190 (11,9)

7,6 (0,5)

5,5 (0,34)

---

0,5 (0,03)

63 cm (24,8 in.)
57 cm (22,4 in.)
Slight bleeding.
Good, without
Presence of fly ash in
Appearance
bleeding or
the surface of the
segregation
concrete
Compressive strength, 7 days
39,9 MPa (5790 psi)
25,6 MPa (3710 psi)
* Composition of Portland cement CEM II/B-M according to
European Standard EN-197: 65-79% of clinker + 21-35% of blends of
slag, silica fume, natural pozzolan, fly ash or limestone + 0-5%
minority components

85

Table 2 - Composition and properties of reference SCC and smart dynamic

concrete with lower fines content in plant B
Composition in kg/m3 (lb/ft3)
0-2 mm sand
0-4 mm sand
4-10 mm gravel
CEM II/B-M (V-S-LL)*
Water
Superplasticizer (Glenium
type)
New VMA
Properties
Slump flow
Appearance

Reference-B
558 (34,8)
596 (37,2)
604 (37,7)
400 (25,0)
260 (16,2)

SDC-B
558 (34,8)
646 (40,3)
604 (37,7)
350 (21,8)
255 (15,9)

7,0 (0,44)

7,0 (0,44)

----

1,0 (0,06)

77 cm ( 30,3 in.)
Good, without
bleeding or
segregation

73 cm (28,7 in.)
Good, without
bleeding or
segregation

Compressive strength,
7 days
38,0 MPa (5510 psi)
22,0 MPa (3190 psi)
28 days
47,1 MPa (6830 psi)
33,2 MPa (4820 psi)
* Composition of Portland cement CEM II/B-M (V-S-LL) according to
European Standard EN-197: 65-79% of clinker + 21-35% of blends of
fly ash, slag and limestone + 0-5% minority components

86

Table 3 - Composition and properties of reference vibrated concrete and

smart dynamic concrete with lower fines content in plant C
Composition in kg/m3 (lb/ft3)
0-4 mm natural siliceous sand
8-16 mm gravel
16-25 mm gravel
Fly ash
CEM II/A-LL 32.5R*
Water
Superplasticizer (Glenium type)
New VMA
Properties

Reference-C
992 (61,9)
396 (24,7)
596 (37,2)
30 (1,9)
280 (17,5)
165 (10,3)
2,6 (0,16)
---

SDC-C
1090 (68,0)
820 (51,2)
--30 (1,9)
300 (18,7)
180 (11,2)
3,2 (0,20)
0,4 (0,02)

S4 20 cm (7,9 in.)
68 cm (26,7 in.)
slump
T50
-->2 s
Compressive strength, 1 day
NA
10,5 MPa (1520 psi)
7 days
NA
26,4 MPa (3830 psi)
28 days
NA
34,1 MPa (4950 psi)
* Composition of Portland cement CEM II/A-LL according to
European Standard EN-197: 80-94% of clinker + 6-20% of
limestone + 0-5% minority components
Slump Flow

87

Table 4 - Composition and properties of Reference vibrated concrete and

smart dynamic concrete with lower fines content in plant D
Composition in kg/m3 (lb/ft3)
Reference-D
SDC-D
0-2 mm natural siliceous sand
266 (16,6)
371 (23,2)
0-5 mm crushed limestone sand
578 (36,1)
658 (41,1)
5-12 mm gravel
497 (31,0)
371 (23,2)
12-22 mm gravel
478 (29,8)
430 (26,8)
CEM II/B-S 42.5 *
330 (20,6)
330 (20,6)
Water
185 (11,5)
185 (11,5)
Naphthalene based superplasticizer
4,3 (0,27)
--Superplasticizer (Glenium type)
--4,0 (0,25)
New VMA
--0,7 (0,04)
Properties
Slump Flow
S3 (not determined)
71 cm (28,0 in.)
Compressive strength, 7 days
19,9 MPa (2890 psi) 20,5 MPa (2970 psi)
28 days
28,9 MPa (4190 psi) 29,7 MPa (4310 psi)
* Composition of Portland cement CEM II/B-S according to European Standard
EN-197: 65-79% of clinker + 21-35% of slag + 0-5% minority components
Table 5 - Composition and properties of Reference vibrated concrete and
smart dynamic concrete with lower fines content in plant E
Composition in kg/m3 (lb/ft3)
0-2 mm crushed limestone sand
2-12 mm gravel
8-16 mm gravel
Fly ash
CEM III/A 42.5 N*
Water
Superplasticizer (Glenium type)
New VMA

Reference-E
753 (47,0)
368 (23,0)
630 (39,3)
120 (7,5)
280 (17,5)
175 (10,9)
3,5 (0,22)
---

Properties
Slump flow
T50
Compressive strength,

SDC-E
774 (48,3)
172 (10,7)
774 (48,31)
80 (5,0)
300 (18,7)
185 (11,5)
4,5 (0,28)
0,6 (0,04)

--69 cm (27,2 in.)

--2s
7 days
27,5 MPa (4000 psi) 29,2 MPa (4230 psi)
28 days
34,8 MPa (5050 psi) 35,9 MPa (5210 psi)
* Composition of Portland cement CEM III/A according to European Standard
EN-197: 35-64% of clinker + 36-65% of slag + 0-5% minority components

88

Table 6 - Composition and properties of Reference SCC and smart dynamic

concrete with lower fines content in plant F
Composition in kg/m3 (lb/ft3)
Reference-F
SDC-F
0-6 mm natural siliceous sand
1015 (63,4)
1150 (71,8)
6-12 mm crushed limestone gravel
680 (42,4)
620 (38,7)
Limestone filler
170 (10,6)
--CEM II/A-V 42.5R *
330 (20,6)
400 (25,0)
Water
196 (12,2)
225 (14,0)
Superplasticizer (Glenium type)
7 (0,44)
8,5 (0,53)
New VMA
0,40 (0,02)
Properties
Slump flow
63 cm (24,8 in.)
62 cm (24,4 in.)
Fresh density
2322 kg/m3 (145 lb/ft3) 2244 kg/m3 (140 lb/ft3)
Compressive strength, 1 day
15,5 MPa (2250 psi) 13,6 MPa (1970 psi)
7 days
29,0 MPa (4210 psi) 29,2 MPa (4230 psi)
28 days
33,8 MPa (4900 psi) 33,2 MPa (4820 psi)
* Composition of Portland cement CEM II/A-V according to European Standard
EN-197: 80-94% of clinker + 6-20% of fly ash + 0-5% minority components

89

Slump flow (mm)

900
SCC-12

800

SCC-20

700
600
500
400
300
0

40

80

120

Yield stress (Pa)

Flow time T50 (s)

6
SCC-12
SCC-20

5
4
3
2
1
0
0

40

80

120

Plastic viscosity (Pa.s)

Flow time TV (s)

40

SCC-12
SCC-20

30
20
10
0
0

40

80

120

160

Plastic viscosity (Pa.s)

Fig. 1 - Relationship between engineering measurements and rheological

parameters

90

12

Starch A
Synthetic Polymer B
Starch B
Natural Gum D
NEW VMA

Torque (Nmm)

8
Synthetic Polymer A
Natural Gum C
REFERENCE
Natural Gum B
Natural Gum A

0
0

40

80
Speed (rpm)

120

160

Fig. 2 - Influence of several commercial VMAs on the rheology of mortar

compared to a reference mortar without VMA and to the new VMA. Note
that the mortars tested in the rheometer were obtained by sieving the
concretes.

91

Fig. 3 - Mechanism of action of the innovative VMA called

RheoMATRIX that provides a tridimensional network made up
by admixture-admixture, admixture-water and admixture-solid
particles interactions

92

a) Reference-A

b) SDC-A

Fig. 4 - Appearance of (a) reference concrete and (b) smart dynamic concrete
in plant A

93

a) Reference-B

b) SDC-B

Fig. 5 - Appearance of (a) reference concrete and (b) smart dynamic

concrete in plant B

94

Fig. 6 - Placing of smart dynamic concrete prepared in plant C (SDC-C)

Fig. 7 - Placing of smart dynamic concrete prepared in plant E (SDC-E)

95

a) Reference-F

b) SDC-F

Fig. 8 - Appearance of (a) reference concrete and (b) smart dynamic

concrete in plant F

96

PERFORMANCE SPECIFICIATIONS OF SELF-COMPACTING

CONCRETE FOR THE GULF REGION
Jon Knights and Don Wimpenny

Synopsis: Self compacting concrete (SCC) is differentiated from conventionally

placed concrete by its ability to flow and level without vibration or segregation,
however, it is prudent to apply broad limits performance specifications for SCC in
the fresh state and rely on trials to determine placing and compaction performance
using established test methods. Prescriptive specifications are historically
employed for specifying durability. However, in extreme environments or long
service lives, performance based specifications may be more appropriate. The
differences between the two specification routes are discussed. Modelling of
deterioration processes is a powerful tool to establish performance requirements
for SCC properties. Probabilistic modelling and reliability analysis can also be
used to arrive at performance criteria for specification of durable SCC in a robust
and defensible manner. Probabilistic modelling examples showed that prescriptive
limits were sufficient for concrete in relatively benign conditions, but were
inadequate in more aggressive environments of the Gulf Region. Performance
based specifications for durability provides a greater level of comfort to
stakeholders that the structure will achieve its intended design life, reducing the
risk of maintenance and remedial works in later life. It may also increase
efficiencies in construction compared with the prescriptive approach, particularly
in less severe exposure conditions.

Keywords: Durability modelling; Performance specification; Self-compacting

concrete

97

Jon Knights. Principal Materials Engineer, Halcrow Group Ltd. Extensive

experience in design, specification, testing and application of concrete and
materials technology. Development of in-house analytical models to estimate
service life of concrete structures in aggressive environments. Durability audits
and planning strategies. Forensic investigation and expert witness services.
Thermal modelling and design
Don Wimpenny. Associate Director, Halcrow Group Ltd. Head of the Materials
& Asset Engineering Team at Halcrow, Australasia. He has a background in
cement research, civil engineering contracting, concrete supply and construction
materials testing and investigation. Interests include the specification and testing
of concreting materials, the diagnosis of construction material failures and the
design of remedial works.

INTRODUCTION
General
Self- compacting concrete (SCC) has been gaining in popularity since its
development in Japan in the late 1980s. Much work has been undertaken to
establish mixture proportions methods to quantify and optimise its properties.
SCC is differentiated by its properties in the fresh state and has found many insitu
and precast applications due to speed of placement, improved form filling,
enhanced surface finish and noise and vibration reduction.
Production of SCC for the Gulf Region favours the use of performance-based
rather than prescriptive specifications. This paper compares prescriptive and
performance based approaches and discusses the use of performance-based
specifications for the fresh and hardened properties of SCC, particularly with
regard to durability design of reinforced concrete structures in extreme
environments.
Tools, such as deterministic and probabilistic deterioration modelling and
reliability analysis, which provide the designer and specifier with a robust process
for determining performance limits are introduced. An example of the techniques
and process of modelling and performance design is included.

98

Characteristics properties of SCC

SCC is characterised by flowing concrete that self-levels without segregation.
However, the mixture proportions process that has developed for successful selfcompacting concretes results in broad changes to mixture proportions
characteristics, compared with conventional concretes, reducing yield strength
and plastic viscosity but maintaining cohesion. The main changes in design of
SCC mixes are as follows:
Lower coarse aggregate higher fine aggregate fractions
Increased paste volume fraction
Lower water/cement ratio
Increased superplasticiser dosages
Use of viscosity modifying agents (additions or admixtures)
The resultant mixtures have high flow characteristics that need to be quantified in
different ways to conventional concrete. Tests have been developed to evaluate
flowability, filling ability, passing ability (to reinforcement) and resistance to
segregation. These tests may be used during the mixture development phase and
for compliance testing during construction.
The principal differences in the hardened properties, such as increased shrinkage,
are due mainly to the increased paste volume fraction. The benefits of improved
compaction are not quantified clearly in literature, so despite the fact that
differences do exist, SCC is normally treated in current guidance the same as that
of conventional concrete when it comes to mechanical design parameters. Table 1
briefly summarises the differences in mechanical parameters between
conventional concrete and SCC.
SPECIFICATION FOR SCC IN THE FRESH STATE
Prescriptive limits
It is often difficult to quantify the outcome of fresh properties in a completed
structure without intrusive and destructive investigation. In everyday situations, it
is simpler to rely on experience to assign fresh property requirements based on
prescriptive guidance. For example(2)
Slump Flow (SF1)
(550-650mm) appropriate for unreinforced and slightly reinforced
structures(e.g. housing slabs)

99

Slump Flow (SF3)

(760-850mm) vertical applications in very congested structures, complex
shapes, filling under formwork.
Passing ability (PA1) (L-box)
structures with a gap of 80 to 100mm (e.g. housing structures)
Passing ability (PA2)
structures with a gap of 60 to 80mm (e.g. civil engineering structures)
Specifying the performance of the fresh concrete mixture can be onerous, as it
potentially places liability on the specifier should the mixture as specified fail to
perform by filling and passing the formwork in a satisfactory manner. It is
therefore usual to include a caveat that the contractor shall confirm the suitability
of the performance limits by trials.
Summary of guidance
As the fresh properties of SCC are arguably the most important differentiating
factor between conventional concretes, a number of documents have been
produced over the years to provide guidance on the testing and evaluation for the
production and conformity. A summary of available guidance (not exhaustive) on
the practical aspects of design, testing, producing and placing, finishing and
curing SCC is presented:
Specifications and Guidance for Self-Compacting Concrete. EFNARC(1)
European Guidelines for Self-Compacting Concrete. EFCA/EFNARC(2)
Guide to the Design of Concrete in the Arabian Peninsula. Concrete Society
(UK)(3)
Performance Based Design for Self-Compacting Structural High Strength
Concrete. ACI(4)
Self-Consolidating Concrete. ACI Manual 237R(5)
Self-Compacting Concrete Recommendations for Use. SCA(6)
Performance-based specification for fresh SCC
Fresh properties of SCC can be specified by performance, although it is important
to understand the requirements of fresh concrete in relation to the application. Of
particular importance in the Gulf Region is the understanding of the trade-off
between flowability, workability retention, setting time, bleed and segregation.
For example in hot environments, workability (flow) is often reduced rapidly by
the evaporation of free water and onset of initial hydration. Careful mixture
proportions is required to ensure that SCC does not segregate when batched

100

initially (especially if it is to be transported without agitation), but can retain

flowability for a sufficient duration to ensure adequate filling and compaction in
the form. The above statement, in effect, can be used as a performance-based
specification clause.
Performance requirements for fresh SCC are most applicable at mixture
development stages based on laboratory and field testing, after all, these are
practical-based tests that only indicate the rheology of a SCC mixture.
Ultimately, it is far safer for the specifier to supply minimum and/or maximum
limits (with tolerances) for flow, viscosity and passing ability.
Thus, despite the large amount of guidance available on mixture proportions for
fresh properties of SCC, it is recommended that, when approaching the
performance-based specification route, the concrete supplier or contractor is
instructed in specification clauses to demonstrate the required performance of the
mixture with site trials. This may involve visual evaluation during placement and
even invasive investigation of the trial samples, such as coring or cutting, to
establish mixture uniformity and degree of compaction.
DURABILITY DESIGN APPROACHES
Comparisons of durability characteristics between SCC and conventional concrete
are of interest, particularly in the aggressive environments experienced in the Gulf
Region, and with particular regard to chloride ingress and reinforcement
corrosion. Owners, designers and specifiers must be certain that the benefits of
SCC not only include those attributed to placement, compaction and finish, but
also the long term resistance to aggressive media throughout a structures design
life.
Design for durability
Regardless of concrete type, the required or intended service life of a structure
is a major factor in the design for durability, and may dictate how the overall
design of a structure is approached. The longer the expected life, the greater the
severity of the event the structure must generally be designed to withstand, and
the greater the durability (and hence quality) of concrete required.
Designing for durability is also dependent on the loadings imposed.
Environmental loads are the most relevant to durability design and the
mechanisms and processes of deterioration to which different parts of a structure
will be subjected are key considerations. The most relevant processes in the Gulf

101

that require careful consideration are chloride ingress, sulfate attack and salt
crystallisation resulting in surface scaling and weathering.
Durability design needs to be part of an integrated and structured process that is
typically driven by the clients requirements for in-service performance (often
with zero-maintenance requirements throughout the design life), which takes
account of both structure-specific operational and environmental loads. Fig. 1
illustrates an example of an integrated durability design process. The design
process should be iterative, selecting and optimising for appropriate available
materials, performance characteristics and structural function.
Specification routes for durability design
Two principal routes for undertaking durability design are presented in Fig. 1,
either a prescriptive or performance-based approach, or more commonly a
combination (hybrid) of the two:
The prescriptive (or deemed-to-satisfy) approach typically relies on
current national standards and guidance to arrive at specific limiting
proportions to concrete mixture constituents, proportions and cover for a
given service life and exposure, commonly stated in published tables (e.g.
BS EN 206(7), BS8500(8), BS 6349(9), ACI 318(10), Concrete Society CS
163(11)). The requirements in published tables are defined on the basis of
practical experience and engineering judgement.
The performance approach aims to achieve parameters that demonstrate
the suitability of the concrete and its constituents for a given environment
and for an explicitly given design life. The durability performance
parameters are often derived by performance testing and track record,
often accompanied by durability modelling. In its purest form,
performance-based designs are not restricted by limiting the proportions of
concrete constituents, unless required for performance relating to fresh
properties.
Typically, current practice is a combination of mainly prescriptive elements
combined with certain criteria for which a performance based approach is applied
(such as compressive strength, cover and consistence (i.e. workability or flow)).
Regardless of whether prescriptive or performance approaches are implemented,
the design method should provide a structured, cohesive and defensible approach
to the design for concrete durability for SCC.

102

Overview of prescriptive design standards

EN 206-1 - promotes the concept of exposure classes for concrete in different
environments introduced in EC2.
BS 8500 Parts 1 and 2 - include greater detail on the specification of limiting
proportions and cover values for concrete in different UK exposure
conditions.
BS 6349 - design of marine structures in the UK, also includes durability and
cover requirements.
The UK Concrete Society- CS 163 - guidance on durability design and
construction guides for use in the Gulf Region.
North American guidance on durable concrete derives from American
Concrete Institute documents ACI 318-08 on building code requirements for
structural concrete and ACI 201.2R Guide to durable concrete. Information
in relation to durability and design lives is scant
UFGS-03 31 29(12) (N. America) specifies minimum compressive strengths in
areas exposed chloride-laden environments and high abrasion. Limiting
proportions are also given based on exposure condition.
AS 3600(13) and AS 4997(14) two key Australian standards that give
prescriptive limits on concrete in different environments.
Of particular relevance to the Gulf Region, the Concrete Society Report CS 163
provides tables that set definitions for exposure classes applicable to the Gulf
Region, and prescriptive limits on concrete quality and cover are given for
structures for up to a design life of up to 30 years. However, beyond this time, or
for concrete in extreme environments, the guide classes these as special
structures and suggests seeking expert advice on additional measures for
protection, a durability plan and deterioration modelling, possibly resulting in
performance-based designs.
Table 2 summarises a comparison of concrete quality, cover and design life for
concrete in aggressive cyclic wetting and drying environments across the
Standards, Codes and Guidance discussed above. It can be seen that there are a
wide range of cover values given from different sources for the same notional
exposure environment, although limiting proportions (cement content and
water/cement ratio) are all similar.
The UK standards, in comparison with Middle East and Australian values indicate
a large disparity. Concrete Society and Halcrow modelling indicates similar
cover for 30 years, whereas similar or lower cover values are given for 100 years
design lives in UK standards. This is confirmation that UK standards are not

103

suitable for use in specifying durability in the Gulf environment, and should be
avoided. Furthermore, chloride ingress modelling indicates that, on a prescriptive
basis, high covers would be needed for long design lives in extreme
environments.
Performance based durability design
From benign to severe exposure conditions, for structures with normal service
life, appropriate detailing, concrete quality and minimum cover from prescriptive
guidance is likely to be sufficient. However, the importance of maintaining
durable structures in extreme conditions or with extended service lives may
require the designer to consider alternative methods to enable durability of the
structure to be determined on a more realistic basis.
Performance based specification for durability can provide a greater level of
comfort to all stakeholders, that the structure will achieve its intended design life,
thus reducing the risk of costly maintenance and remedial works in later life.
The definition of a performance-based approach is simply that concrete materials
shall be designed and specified on the basis of their demonstrated ability to
adequately perform in a given environment for a specified time period. The
design life should not be less than that required by the client but may be longer to
account for uncertainties, or to increase the probability of achieving the required
service life.
The concept of design life requires the end of the design life to be identified. The
deemed to satisfy approach of prescriptive designs does not ascertain what
the failure of a structure (i.e. its limit state) actually corresponds to, whether it is
intervention by repairs or complete collapse.
The end of a design life in performance based specification should usually be
characterised by some level of damage caused by the deterioration process (e.g.
corrosion initiation, cracking or spalling of concrete). This is generally
understood as the Serviceability Limit State (SLS), indicating service failure or a
time of major intervention, as opposed to the Ultimate Limit State (ULS) which
implies failure or collapse of the structure.
The following guidance is available for performance-based durability design.
The FIB Model Code(15) attempts to establish a durability design approach
similar to that used for structural design and outlined in Eurocodes (EC2)(16),

104

which means the establishment of quantifiable models for both environmental

load and resistance.
Concrete Society Technical Reports 61-63(17): provide general information on
the background to the problem of reinforcement corrosion in concrete and
methods of modelling deterioration. Information is also given on additional
design measures to improve or enhance concrete durability.
Performance Indicators for concrete durability
To enable a design to be successful, performance testing must include durability
test parameters or Performance Indicators that are deemed relevant to the
deterioration process they are intended to assess.
Some indicators may not be used specifically as model input parameters but may
be required to assess the quality of the concrete and its variation with time, such
as compressive strength. Others are related to the deterioration process, in the case
of corrosion induced by chloride ingress or carbonation, such as porosity,
apparent (or effective) chloride diffusion coefficient, permeability to gas, and
permeability to water.
Performance Indicators for concrete durability are generally selected from
parameters representative of certain concrete properties. They may also be used as
indicators during the life time of the structure to help refine modelling and for
residual life studies, if monitored regularly.
Table 3 presents a summary of some tests that may be used as part of a
performance-based specification, either as input parameters for deterioration
modelling, or as stand-alone performance characteristics in their own right.
DURABILITY MODELLING AND RELIABILITY ANALYSIS
AN INTRODUCTION
The concept of durability modelling attempts to establish whether the
environmental load on the structure at a given location or exposure condition
exceeds the resistance of the structure under assessment, and if so, gives an
estimate of when it may be exceeded to a defined limit state. Performance-based
designs may or may not include the use of modelled predictions of deterioration
mechanisms as part of the design. Modelling of degradation mechanisms allows a
level of robustness for the quantification of a performance-based design,
particularly if performance test parameters fit well with real-life data. Most
significantly, performance testing and design may provide an additional degree of
substantiation for designs for prestige or strategically critical structures.

105

Modelling of deterioration mechanisms (principally chloride or carbonation

ingress in relation to concrete) can be divided in to two main types; deterministic
and probabilistic models. Deterministic models require the use of single input
parameters, such as diffusion coefficients, that are justifiable for use within the
context of the load and resistance factors being investigated. The output of the
model is a single value that can be assessed against the design requirements. Fig.
2 illustrates the output of deterministic modelling for the ingress of chlorides.
The Figure shows the input parameters used and the output showing the years to
initiation of corrosion and subsequently cracking due to corrosion propagation.
By simply changing the input parameters it is possible to estimate the concrete
type and quality required to produce durable concrete over the desired design life.
Probabilistic modelling on the other hand requires the input variables as
probability distribution functions not single values, the resulting output is a
probability of failure at given load and resistance conditions. The advantage of
using the probabilistic approach is that it accounts statistically for the uncertainty
in the values of input variables and model uncertainty.
Probabilistic modelling may be enhanced by the use of reliability analysis. The
reliability index is the number of standard deviations from the mean of a normal
distribution outside which the area under the curve represents the probability of
failure(18). This can be demonstrated for a simple case for linear (normal)
distributions:
If R is the resistance or capacity of the concrete against the load S, the limit state
function Z can be formulated for the assessment of reliability(19):
Z=R S

(1)

where:
Z>0 the concrete is in a desired state
Z=0 the concrete has reached the limit state
Z<0 the concrete is in the adverse state
Fig. 3 illustrates the input distributions for load, S, and resistance, R. Random
sampling of these input distributions is undertaken, often by using techniques
such as Monte Carlo or Latin Hypercube techniques. The output of the model is a
service life distribution, Z, which than then be assessed statistically to give
estimates of the service period (tp) for a given limit state failure probability.

106

If R and S have a normal distribution, Z will also be linear, the reliability index
and the probability of failure (Pf) can be calculated using the standard normal
distribution(19).

Pf

PR

( R)
2

( R)

(S )
2

(S )

(2)

where: (.) = standard normal distribution function; (R,S) = mean of the R and
S distributions; and (R,S) = standard deviation of the R and S distributions
The value of (- ) can be found in tables for the standard normal distribution.
The reliability of a structure can be expressed as the failure probability (- t), or
directly as .
The value of (- t) or is given in Codes of Practice. The current Eurocode EC2
gives the serviceability limit state (SLS) for a 100 year design life as between
1.5 1.8, which corresponds to a (Pf) of 6.7 x 10-2 to 3.6 x 10-2, that is, between 7
and 4% failure probability.
EXAMPLE: DETERMINING DURABILITY PERFORMANCE
PARAMETERS FOR MARINE WORKS USING SCC
General
Performance based designs are currently being used in the Gulf for major
structures to determine the durability requirements using parameters and
distributions appropriate to the environment. The following example is based on
a case study of using SCC in the UK, but modified and compared with the Gulf
Region environment using relevant input data.
Background
The construction and maintenance of coastal defence structures, such as
breakwaters and harbour walls, represents a considerable challenge to provide
long-serving, durable structures that are sympathetic to the character of the
surroundings. SCC concrete with appropriate use of pigments and patterned
formwork provided a solution to this challenge.
In order to develop patterned pigmented SCC, laboratory and field work was
undertaken to match the colour and texture with the local stonework. In addition

107

to these characteristics, the concrete must have adequate resistance to chloride

ingress, abrasion and impact, a low water content to control efflorescence and
appropriate workability for compaction without segregation. A SCC mixture
containing silica fume, ground granulated blastfurnace slag (ggbs) and locally
available crushed stone was considered likely to provide a durable and sustainable
solution.
Approach to modelling and performance based design
Whether modelling on a deterministic or probabilistic basis, the general approach
for predictive modelling and performance design can be summarised as follows,
and was used as the basic approach to the design of the SCC seawall.
establish design service life and serviceability limit state
establish the cases to be modelled and define the input parameters based on
the assumptions, including the exposure environments and their potential
severity
undertake predictive modelling to assess durability for different design
options
use appropriate statistical treatments to define service life at the desired limit
state
use of input parameters from modelling as performance-related specification
requirements for durability design.
The intended service life of the structure was 100 years, with a serviceability limit
state defined as the initiation of reinforcement corrosion (defined by exceeding
the chloride threshold of the concrete), with a specified failure probability of 5%
(i.e. = 1.64).
Input parameters and assumptions
Table 4 summarises the input parameters and distributions used in the modelling
and assumed for UK and Gulf Region conditions, respectively. The functions
used in the probabilistic model for chloride ingress are based on empirical test
data taken from the Brite Euram Duracrete Project(19) and project related data,
where applicable.
Three scenarios have been modelled to demonstrate the differences between
performance requirements for the UK and Gulf region environments in
splash/tidal marine exposure zones. For simplistic comparison, each scenario is
varied on the amount of cover to reinforcement required to achieve the specified

108

5% failure probability. Each scenario assumes the same concrete quality, only
environmental conditions were changed.
Scenario 1: UK conditions, 13 C mean ambient temperature, 50mm cover.
Scenario 2: Gulf conditions, 33 C mean ambient temperature, reduced longterm hydration, 50mm cover.
Scenario 3: Gulf conditions, 33 C mean ambient temperature, reduced longterm hydration, 110mm cover.
The concrete under evaluation is considered to be a high quality triple blend
concrete containing both GGBS and silica fume, with a chloride diffusion
coefficient of 1.36 x 10-12 m2/second measured at 28 days from casting.
Predictive modelling
Predictive modelling was undertaken using a chloride ingress model developed
in-house at Halcrow. Established techniques based on Ficks second law of
diffusion and empirical relationships (TRL(20) and DETR(21)) were adapted to
predict chloride ingress and corrosion. The model enables the user to study the
effect of a number of boundary conditions on concrete mixtures and formulate the
most suitable basic mixture proportions that are durable over its specified service
life.
The model has been developed to provide the capability of running probabilistic
scenarios using @Risk, a MS Excel add-in allowing random sampling of input
distributions. The three scenarios were run for 5000 sampling iterations to derive
output for further analysis.
Output of probabilistic modelling
The output of the modelling produced probability distribution functions for the
amount of chloride ingress to the depth of the reinforcement cover. The three
scenarios were analysis statistically and expressed as failure probability with time
and are presented in Fig. 4.
Fig. 4 illustrates that in UK conditions, the concrete is capable of achieving a 100
year design life at below the 5% probability of failure (i.e. for onset of corrosion)
at 50mm mean cover. This represents a saving of 10mm cover when compared
with the requirements of BS 6349, which may be further reduced if the limit state
was relaxed (e.g. to onset of cracking due to corrosion).

109

However, if the same concrete, with the same cover, were to be used in a similar
marine environment in the Gulf Region, modelling estimates that the specified
failure probability would be exceeded after only 22 years, corresponding to a 92%
chance of failure at 100 years. By changing cover depth only, it is estimated that
an increase in mean cover to 110mm for the same concrete would be necessary to
achieve the required failure probability.
Therefore in basic form, a performance specification can be derived from the
modelling to specify that; for concrete to perform with a 5% probability of
failure (to onset of corrosion), a mean cover of 110mm or greater is required, and
the chloride diffusion coefficient of the concrete should not be greater the 1.36 x
10-12 m2/sec at 28 days.
Additional analysis could be undertaken to value engineer the durability of the
structure by improving the mixture further and reducing the diffusion coefficient,
or by considering additional protective measures to reduce cover (e.g. use of
corrosion inhibitors, controlled permeability formwork or stainless steel
reinforcement).
The above requirements can then be added to the specification to provide the
contractor/ concrete producer with a set of requirements (including for SCC in the
fresh state) that are to be demonstrated during mixture proportioning and
development. This defensible approach gives greater confidence to clients and
owners that appropriate concrete for their structure has been identified by
demonstrating its performance, rather than rely on prescribed values in guidance
that not only are inappropriate for any demanding structure requirements
proposed, but also potentially producing savings in less severe environments.
CONCLUSIONS
Self compacting concrete (SCC) is differentiated from conventionally placed
concrete by its ability to flow and level without additional vibration, without
segregation. It is generally regarded to have similar properties to conventional
concrete in the hardened state.
Performance specifications are applicable to SCC in both fresh and hardened
states. When performance specifications are applied for SCC in the fresh state, it
is prudent to apply broad limits and rely on trials to determine placing and
compaction performance using established test methods.
Prescriptive or deemed to satisfy specifications are historically employed for
specifying durability. However, in extreme environments and for structures

110

requiring long service lives, performance based specifications may be more

appropriate. This relies initially on the recognition of client/owner requirements
for a structure and an understanding of the deterioration mechanisms applicable to
the exposure environment. Limit states for serviceability are then defined, and
concrete properties appropriate to the application selected for further assessment.
Modelling of deterioration processes, particularly chloride ingress and
reinforcement corrosion is a powerful tool to establish performance requirements
for SCC properties in extreme environments. Probabilistic modelling and
reliability analysis can also be used to evaluate concrete properties, in a robust
and defensible manner, against the defined Serviceability Limit State (SLS), to
arrive at performance criteria for specification of durable SCC.
An example of probabilistic modelling demonstrated three scenarios in different
environments for the same concrete, at a 5% probability of failure to onset of
corrosion (i.e. SLS) in marine environments for a 100 year design life.
Modelling in relatively benign UK conditions showed that a mean cover of 50mm
was sufficient for high quality concrete in these conditions, but was inadequate in
the more aggressive environments of the Gulf Region (i.e. 22 years to (SLS)
failure). For the same concrete, mean cover was increased significantly to
110mm to achieve the required serviceability. Further modifications to optimise
performance may be necessary.
Performance based specification for durability can provide a greater level of
comfort to all stakeholders that the structure will achieve its intended design life,
thus reducing the risk of costly maintenance and remedial works in later life. It
may also increase efficiencies in construction compared with the prescriptive
approach, particularly in less severe exposure conditions.
REFERENCES
1. Specifications and Guidance for Self-Compacting Concrete. EFNARC. 2002.
2. European Guidelines for Self-Compacting Concrete. EFCA/EFNARC. May
2005
3. Guide to the Design of Concrete in the Arabian Peninsula. Concrete Society
(UK) CS 163. 2008
4. Performance Based Design for Self-Compacting Structural High Strength
Concrete. ACI SP228-2. 2005.

111

5. Self-Consolidating Concrete. ACI Manual 237R. 2007.

6. Self-Compacting Concrete Recommendations for Use. SCA. 2002.
7. EN 206-1: 2000. Concrete. Specification, performance, production and
conformity
8. BS 8500-1: 2006, Concrete. Complementary British Standard to BS EN 206-1.
Method of specifying and guidance for the specifier
9. BS 6349-1: 2000. Maritime structures. Code of practice for general criteria
10. ACI 318-08. Building Code Requirements for Structural Concrete and
Commentary
11. Concrete Society (2008). CS163-Guide to the design of concrete structures in
the Arabian Peninsula
12. UFGS 03-31-29 (2007). Marine concrete specification
13. AS 3600-2001. Concrete structures
14. AS 4997-2005. Guidelines for design of maritime structures
15. FIB Bulletin 34: 2006. Model code for service life design ISBN 978-288394-074-1
16. EN 1992-1-1: 2004. Eurocode 2: Design of concrete structures. General rules
and rules for buildings
17. Concrete Society (2004). Technical Report 61-Enhancing Reinforced
Concrete Durability: Guidance on selecting measures for minimising the risk of
corrosion of reinforcement in concrete Parts 1-3
18. Bamforth PB. 2000. A new approach to durability design using risk analysis.
Durability by Intent Workshop. Building Research Establishment.
19. BRITE EURAM III. DURACRETE General Guidance for Durability
Design and Redesign. Contract BRPR-CT95-0132. Project BE95-1347. 2000.

112

20. TRL Contractor Report 359. 1997. An international review of chloride

ingress into structural concrete. by Bamforth PB, Price WF, Emerson M.
Transport Research Laboratory.
21. DETR 1999. Guidance on the Selection of Measures for Enhancing
Reinforced Concrete Durability. Partners in Technology Programme Contract CI
39/3/376 (cc967). Prepared by Taywood Engineering Ltd.
22. Bamforth P B 1997, The derivation of input data for modelling chloride
ingress from 8 year UK coastal exposure trials. Paper submitted for the 4th
CANMET/International Conference on Durability of Concrete, 17-22 August
1997.

113

Table 1 - Comparison of hardened properties between SCC and conventional

concrete.
Property

Change when SCC

compared with
conventional concrete

Remarks

SCC comparable or
slightly higher

Due to lower w/c ratio

and better compaction

Tensile strength

SCC comparable

Elastic modulus

SCC slightly lower

Due to higher paste

fraction/ lower coarse
aggregate fraction

Creep

SCC slightly higher

Due to higher paste

fraction/ lower coarse
aggregate fraction

Shrinkage

SCC slightly higher

Due to higher paste

fraction although offset
by lower water content

Thermal expansion
coefficient

Broadly similar

Dependent on aggregate
type

Reinforcement bond

SCC comparable or
slightly higher

Due to higher (more

uniform) compaction

SCC comparable

Possible. Due to higher

(more uniform)
compaction in the cover
zone

Compressive strength

Durability

114

Table 2 - Comparison of prescriptive durability limits for marine cyclic wetting and drying exposure classes (i.e.
splash/tidal zone)
Code/Standard

Country/
Region

Minimum
Strength
(MPa)

Maximum
water/cement
ratio

BS8500 - 1

UK

50

0.35

60% GGBS,
40% PC

380

45

100

BS 6349

UK

50

0.40

60% GGBS,
40% PC

370

60

100

UK
Concrete
Soc. CS 163

ME

40

0.40

Triple blend

360

65

30

UK
Concrete
Soc. CS 163

ME

60

0.35

60% GGBS,
40% PC

400

65

30

ACI 318

USA

34.5

0.40

Not given

Not given

64

Not given

UFGS

USA

42

0.40

Not given

400

75

Not given

Australia

50

0.40

Not given

400

65

25

CorrPred
(Halcrow Grp)

ME

60

0.35

60% GGBS,
40% PC

400

60

30

CorrPred
(Halcrow Grp)

ME

60

0.35

60% GGBS,
40% PC

380

90

100

AS 4997

115

Cement
Type

Minimum
Minimum
cementitious cover (mm)
content
(kg/m3)

Design
Life
(Years)

Table 3 - Example of some tests used as performance indicators

Performance test

Test Type

Standard / Reference
(informative)

Precision

Porosity

Hydrostatic weighing

RILEM CPC 11.3

1.5 %

Water absorption

BS 1881-122

Chloride diffusion

Accelerated migration

Steady-state (Nordtest NTB335)

15%

Chloride diffusion

Accelerated migration

Non-steady-state (Nordtest
NTB492)

15%

Gas permeability

Oxygen / CO2

Concrete Society TR 31

30%

CO2 diffusion

Accelerated diffusion

DURACRETE

Water permeability

Water

BS EN 12390-9

Cover

Covermeter

BS 1881-204

Temperature

Temp rise and difference

CIRIA C660

Thermal expansion

Isothermal calorimeter

CIRIA C660

Surface absorption

ISAT

BS 1881-208

Electrical resistivity

Two/four probe

Duracrete Andrade (2001)

116

10%

Table 4 - Input variables and distributions used in probabilistic modelling of chloride ingress and cover values required to
achieve 100 years service life at around 5% probability to initiation of corrosion
Variable/ Input
Parameter
Cement content
(kg/m3)
Water cement ratio
Background
chloride
Mean temperature
Surface chloride
Age factor

Remarks
Total cementitious content (50%
GGBS, 8% silica fume)
Free water cement ratio
Initial chloride content of concrete
(% by mass cement)
Mean ambient over 1 year ( C)
By mass cement (%)
Age factor for chloride diffusion
coefficient

Chloride diffusion
Apparent value at 28 days age, with
coefficient
variation coefficient of 30%
( 10-12 m2/s)
Threshold to initiation (by mass of
Chloride threshold
cement (%))
Mean cover (mm)

Characteristic
Distribution
Value
Type
UK Conditions

Characteristic Distribution
Value
Type
Gulf Region Conditions

400

Deterministic

400

Deterministic

0.38

Deterministic

0.38

Deterministic

0.1

Deterministic

0.1

Deterministic

13
4.5

Deterministic
Pert(2.5, 4.5, 6.5)

33
6.0

0.58

Pert(0.4, 0.6, 0.7)

0.45

Deterministic
Pert(3, 6, 9)
Pert(0.2, 0.45,
0.65)

1.36

N(1.36, 0.41)

1.36

N(1.36, 0.41)

0.4

LN(0.4, 0.15)

0.4

LN(0.4, 0.15)

50

Pert(35, 50, 65)

115

Failure probability from modelling (Pf)

Reliability Index (Beta (- ))

4.1%
1.75

117

Pert (100, 115,

130)
5.1%
1.63

Design Criteria (Client Requirements)

Environmental Loading /
Deterioration
Mechanisms

Operational Loading /
Functional requirements

Structure Service Life

Durability Design Routes

Performance

Prescriptive

H yb rid

Limiting proportions: (minimum

cement content, maximum
water/cement ratio, chloride
content, alkali content)
Cement type
Aggregate reactivity

Flowability/Viscosity
Chloride diffusion coefficient
CO2 diffusion coefficient
Water / air permeability
Porosity
Compressive strength
Minimum/ nominal cover
Tensile / flexural strength
Elastic modulus
Coefficient of thermal expansion
Peak temperature rise
AAR reactivity
Abrasion resistance

Iterative process
Materials Selection
Cement types
Aggregates
Additives
Admixtures

Iterative process
Design Capability Review
Design Compatibility between Structure and Durability
Buildability
Health and Safety
Cost

Further Activities (indirectly design related)

Construction Phase

Operations Phase

Quality control and assurance / Workmanship /

Inspection and Testing
Health and safety
Durability Review and Sign-off

Monitoring, Maintenance and Management

(Performance Monitoring / Asset Management)

Fig. 1 - Routes and criteria for integrated durability design

118

Distribution of R
R,S

(t)

R ( t)

Distribution of S

(t )

S (t )
Time
Failure probability

Service period

( tp )

Service life distribution

Mean service life

Fig. 2 - Schematic of Service period (tp) and lifetime design

119

Modelling for Chloride Ingress into Concrete

Time to corrosion initiation and cracking
Analysis Type: DETERMINISTIC
Project Name:
Concrete Details

Diffusion Coefficient Details

Strength grade

60

Concrete density
Total cementitious content
Binder type (pc, ggbs, pfa, sf )
Percentage Binder
Water/binder (w/b)
Background Chlorides

MPa

Diffusion coefficient estimated

2350

kg/m3

Diff Coeff (at 20C)

400
GGBS
60
0.35
0.20

kg/m3

Age of measured value (yrs)

Age Dependant Diffusion
Switch off age dependency at given date
Turn off age dependancy at (yrs)

% wt of binder

Est by w/c
m2/s

1.50E-13
Y
Y
20

% wt of binder

Estimated Values (Temp adjusted)

Reinforcement Details
Reinforcement Type

Carbon Steel

Bar diameter

25

mm

Ambient Temperature

33

Temperature affected Dca

Dca at 35 days

1.38E-11

Dca at 20 years

3.28E-13

Dca at 20 years
Age Factor

3.28E-13
0.70

35
m2/s

Exposure Details

Temperature affected threshold

Surface Chloride Level (% wt cem)
Exposure Condition

Y
8.1
3. Cyclic wet/dry

Carbonation factor considered?

Relative Humidity (%)

80

Coating used?

C
Surface chloride level

7.00

% cement

Adjusted surface chloride level

8.05

% cement

Temp & Binder adj threshold (Ct)

0.09

% cement

Time to Serviceability Limit State (Years) (Deterministic)

No

Additional Protective Measures

Controlled permeability formwork?
Silane Impregnation

N
N

Minimum
Cover Depth
(mm)

Chloride
threshold >

Time to
initiation

0.09 % by mass cement

plus cracking

60

28

Fig. 3 - Example of deterministic modelling of chloride ingress (Halcrow Group Ltd)

120

51

Failure Probability (to initiation of corrosion) %

0.9
UK conditions, 50mm cover
Gulf conditions, 50mm cover
Gulf conditions, 110mm cover

0.8

0.7

0.6

0.5
In-service performance
achieved

0.4

0.3

Serviceability Limit State

(Pf=0.05)

0.2

0.1

0
0

20

40

60

80

100

120

140

160

180

200

Service Life (years)

Fig. 4 - Failure probabilities (SLS=5% probability to initiation of corrosion) for the same concrete in three modelling scan

121

122

FUTURE TRENDS IN THE PROCESSING OF READY-MIXED

HIGH-STRENGTH SELF-COMPACTING CONCRETE

Harald Beitzel

Synopsis: Concrete is chiefly made in semi-automated or fully-automated

batching and mixing plants. An important part of the process of highperformance concrete making is the actual manufacture of the fresh concrete in
the mixer. The increasingly high demands which, in the past, have been made
with regards to the mixing quality attained have established new criteria
performance for the manufacturers and also for the users of concrete mixers. Two
different high-performance self-compacting concrete (HPSCC) compositions and
various concrete mixers with a nominal capacity from 50 to 3.000 litres were
adopted for the purpose of this research. In order to ascertain potential time limits
for the mixing process difference, the mixing times were adapted. Under the
above-mentioned boundary conditions there are found to be significant
differences between the respective mixing systems as regards the effect of mixing
time on the homogeneity of the mixture assessed by the mixing criteria. A further
advantage over conventional mixing systems is the reduction of the mixing time,
if the consistency is reached within the short time range as a current or a power
indication of the drives is decisive. Quick mixing is resulting in a high output. For
this reason, more mixers with additional tools and different speeds are being used
for the HPSCCs mixing.

Keywords: Concrete mixer, Homogeneity, Mixing time, Superplasticizer

123

Harald Beitzel is Professor of Building Process Technology at the University of

Applied Sciences Trier, and Adjunct Professor for Concrete Production and
Casting at the University of Karlsruhe, as well as Head of IBU, Institute of
Building Process and Environmental Technology (Institut fr Bauverfahrens- und
Umwelttechnik), Trier, Germany. He has long-standing managerial experience in
the construction industry and conducted pioneer research in mixing and pumping
of concrete. He is a member and served on various technical committees of
RILEM and VDI.

INTRODUCTION
The concrete-technological aspects for the use of self-compacting concrete (SCC)
are the optimization and increase of compressive strength, porosity, density, the
resistance to the aggressive action of freeze with de-icing salt, creep and
shrinkage. From these concrete-technological requirements result a number of
economic advantages. SCC can be installed even in densely reinforced areas
without the aid of compaction, retaining at the same time the properties of normal
concrete (NC).
OVERVIEW AND PROBLEM
SCC is distinguished through its advantageous rheological and self-dearating
properties. At the same time, SCC has good resistance to segregation. Other
characteristics of correctly processed SCC are high surface density and improved
durability. The processing properties of SCC differ from the processing properties
of NC as defined in DIN 1045 [4]. The main reason for this is the changed
constitutive composition with a high powder content and the use of innovative
superplasticizers which create the extraordinarily good flow properties. Constant
processing properties are requisite for a uniform concrete production process.
Here, the superplasticizers are of great significance (Fig. 1). Only through the
addition of these admixtures does SCC achieve its temporary flow properties. The
time factor thus plays an important role with regard to the workability and quality
assurance of SCC.
The following criteria are used for assessing the mixing times:
mixture quality test results
homogenization test results
processing test results

124

strength test results

The minimum mixing times of conventional concrete mixers must be assumed to
be longer for the production of SCC than for the production of NC.
The composition of SCC, in comparison to NC, contains a markedly higher
powder content. For the homogenization in the mixing process this means that the
movements of the mixing tools - due to a higher energy input - must produce
many shear planes in the mixture. Here, thorough distribution and mixing of the
superplasticizers, in low quantitative proportion compared to the remaining
mixture constituents, is also of special significance.
SCC is the consistent further development of flow concrete in the direction of a
robust, workable and environmentally sound concrete. The SCCs increasing
application can be expected in particular in the construction of bridges, high-rise
buildings, special civil engineering works, tunnel and precast construction. SCC is
particularly predestined for application in architectural and fair-faced concrete, for
densely reinforced structural members and for slender members, members of
complex geometry and for building in a noise-sensitive environment or on
existing buildings.
The production and processing method, as well as the testing of SCC in Germany,
has been regulated since 2003 in the valid guideline of the German Committee for
Reinforced Concrete (DAfStb) [1] in agreement with DIN 1045 [4] for laboratory
and construction site.
At present, the fact that the materials costs for the manufacture of SCC still lie
around 20% above the costs of a comparable NC is a disadvantage. These extra
costs can be compensated for by improving the production and placement process.
By utilizing all processing and operational resources, the final product SCC can
thus be manufactured in a quality-assured and cost-efficient manner.
The influencing variables necessary for describing the mixing process are shown
in Fig. 2. The characteristic parameters, which are independent of each other and
which influence the overall mixing process, can therefore generally be classified
into the following groups:
system-technical influences
operations-technical influence
The movement of the mixing tools introduces horizontal radial and tangential as
well as vertical movements in the mixture. These quasi-laminar and quasiturbulent basic movement processes determine the kinematics within the mixture.
The maximum processing rate in the outermost zone of the mixing space is
assumed to be the critical magnitude on the mixing process.

125

The operational parameters are determined by the flow of operations in the

concrete plant. The filling height implicates the filling level for a constant mixer
diameter. The mixing time begins when the last mixture component is added and
ends at the onset of the discharge process. The feed arrangement is another output
variable for improving the quality of the mixture.
The selection of the concrete-technological influencing variables, in contrast to
the system-related and operational influencing variables, poses considerable
difficulties.
In order to ensure the optimal processing properties of SCC, an optimization of
constitutive materials is required, as well as the adjustment of the mechanical
equipment for its manufacture imperative. At present, e.g. uneconomically long
mixing periods that diminish the effectiveness of concrete mixing plant must be
scheduled for the production of SCC. To what extent the mechanical engineering
typically used in the manufacture of conventional flow concrete can also be used
for the production of SCC and what modifications should be made, cannot
currently be answered conclusively. Technical literature has so far not reported on
the influence of the mechanical engineering parameters for quality-ensured
production of SCC.
As a result, numerous questions in day-to-day practice arise among operators of
concrete production plants. Owing to the special requirements made on SCC in
respect of batching and mixing, there is a lack of, e.g., proof of the fitness for use
of the conventional concrete mixers and commonly used truck mixers for its
processing.
Currently, knowledge is still insufficient with regard to the mixing process, the
required mixing times, the possible processing time and the necessary machine
adjustments.
The following open questions have to be solved for the manufacture of SCC:
What mixing time is mixer-specific?
What batching sequence is mixer-specific?
How can the mixing efficiency be improved?
What rotational speed must be assumed for the individual mixing phases?
How can the energy input be optimized?
Does segregation occur?
How large is the nominal filling to be assumed?
The mixing processes in concrete technology are stochastic processes. This means

126

that a random elementary particle of the mixture can occupy a specific position in
the mixing space only with a certain probability. A stationary distribution state is
reached after an interminably long mixing time, as the result of the mixture
particle's individual probability of movement. A state described as uniform
random mixture is attained only in case of the same probability of movement.
The assessment of the quality of the mixture is regulated in DIN 459 Part 2 [3]
and in RILEM Final Report TC-150 [2]. In these, the scope of testing for
determining the concrete mixture quality is described for defined experimental
concretes.
The weight of the individual specimens should be selected so high that the
components of the mixture that are available in only small proportions can be
captured through the individual specimens. The maximum particle size should, if
possible, also be included in the probability approach.
Based on a survey conducted by the author and the investigations carried out,
approx. 30% of the manufacturers of concrete mixers have obtained a certificate
in compliance with DIN 459 [3] and the RILEM Final Report TC-150 [2] on the
production of SCC and other concretes. The remaining manufacturers do not have
equivalent, objective proof of the performance capability of their concrete mixers.
To better assess the velocity effect on the homogeneity of the mixture, the
coefficient of variation v is compared as a function of the Froude number Fr.
PRESENTATION AND DISCUSSION OF THE RESULTS
Fig. 3 presents, by way of example, the results of four of the concrete mixers
investigated. The mixture efficiency and the mixture quality for the individual
mixer types are presented by the coefficients of variation as a function of the
mixing time. All of the concrete mixers shown have nominal contents ranging
from 1 m to 3 m and different Froude numbers. Standard mixer No. 1 and
No. 2 and mixers No. 1 and No. 2, which were modified by changing the mixing
speed, are presented in comparison.
The results show that shorter mixing times for producing quality-assured SCC are
possible by changing the mixing speed. This shows that economically favorable
mixing times are possible through appropriate adjustments on the concrete mixer,
also for the production of SCC.
The assessment of the individual coefficient of variation show that the modified
concrete mixers investigated - on the basis of Table 1 - can be classified as highperformance mixers (Fig. 4).
With regard to the processing of SCC, a slump flow of over 700 mm was

127

moreover attained for both modified concrete mixers in the investigations (Fig. 5).
The mean compressive strengths WN after 28 d reached with the modified
concrete mixers, at 63 N/mm and 65 N/mm after a mixing time of 35 seconds
each, can be well classified as a high performance concrete.
SUMMARY AND OUTLOOK
The processing properties of SCC differ from the processing properties of NC as
defined in DIN 1045 [4]. The main reason for this is the changed constitutive
composition with high powder content and the use of innovative superplasticizers
which create the extraordinarily good flow properties. For a reliable concrete
production process, constant processing properties must be ensured. Here, the
superplasticizers are of great significance. Only through the addition of these
admixtures does SCC achieve its temporary flow properties. The time factor thus
plays an important role with regard to the workability and quality assurance of
SCC.
The assessment of the quality of the mixture is regulated in DIN 459 Part 2 [3]
and in RILEM Final Report TC-150 [2]. In these, the scope of testing for
determining the concrete mixture quality is described for defined experimental
concretes. Only approximately 30% of the manufacturers of concrete mixers have
obtained a certificate in compliance with DIN 459 [3] and the RILEM Final
Report TC-150 [2] on the production of SCC and other concretes.
The fine-grained composition of the SCC means for the homogenization in the
mixing process that the movements of the mixing tools - due to a higher energy
input - must produce many shear planes in the mixture. Here, thorough
distribution and mixing of the superplasticizers, in low quantitative proportion
compared to the remaining mixture constituents, is also of special significance.
With regards to concrete raw materials, as well as process and plant technology,
the following developments are being observed in concrete technology:
Concrete raw materials:
New binder combinations
New superplasticizer combination
New aggregate combination
Process technology:
New mixing systems
New materials preparation technologies

128

Shorter mixing times

Multi-stage mixing
Dry-batching
New truck mixer generation
Plant technology:
Higher batching accuracy
Smaller storage capacity (silos)
A stationary distribution state is reached after an interminable long mixing time,
as the result of the mixture particle's individual probability of movement. A state
described as uniform random mixture is attained only in case of the same
probability of movement.
The classification and assessment of the homogeneity of the mixture follows from
RILEM Final Report TC-150 [2], with the boundary proportions of the mixture
quality specified in Table 1. Accordingly, the classification categories standard
mixers, performance mixers and high-performance mixers are possible.
Quality-assured SCC can be manufactured in high-performance mixers
economically short mixing times.
REFERENCES
1. Deutscher Ausschuss fr Stahlbeton (DAfStb),
Selbstverdichtender Beton' (Beuth Verlag, Berlin, 2003)

'DAfStb-Richtlinie

2. Beitzel, H.; Charonnat, Y.; Beitzel, M., 'RILEM TC-1 50 ECM - Efficiency of
Concrete Mixers, Final Report', in Materials and Structures, Vol. 26, No. 256
(RILEM Publications, Bagneux, 2003)
3. Deutsches Institut fr Normung (DIN), 'DIN 459 Teil 1/2 - Mischer fr Beton
und Mrtel' (Beuth Verlag, Berlin, 1995/96)
4. Deutsches Institut fr Normung (DIN), 'DIN EN 206-1 / DIN 1045-2 Festlegung, Eigenschaften, Herstellung und Konformitt' (Beuth Verlag, Berlin,
2001)

129

0,8

WP:
P:
SP:
HSP:

0.8
0.7
0,6
0.6
0,5
0.5
0,4
0.4
0,3
0.3
0,2
0.2
0,1
0.1
0
0.0

w/c ratio

0,7

WP

1940

WP

1950

1960

SP

1970

1980

1990

Without plasticizer
Plasticizer
Superplasticizer
High performance
superplasticizer

HSP

today (2007)

Fig. 1 Overview of the historical progress made in the development of

plasticizers

Mixing chamber: Geometry:

Mixer:

Geometry:

Mixer system influences

Process:
External forces:

Mixing quality

Mixing chamber:

Operational influences

Process :

Diameter
Height
Form
Breadth
Height
Angle of cut
Cutting angle
Form
Number
Working rate
Acceleration of gravity

Filling height
Mixing time
Order of feed
Filling time
Filling variation
Humidity variation
Servicing
Discharge time
Height of fall

Fig. 2 Effects of process engineering on the mixture quality

130

Average coefficient of variation

Coefficient of variation v [%]
v a [%]

10
9
8
7
6
5
4
3
2
1
0
30

M1 Fr 0.60

M1Fr 0.30

M2 Fr 0.60

M2 Fr 0.30

40

50

60

70

Mixing time t [s]

variation vv [%]
of variation
Coefficientof
[%]
Coefficient

Fig. 3 Coefficient of variation as a function of the mixing time with

different Froude numbers
12

Limit HP M

M1
M2

10
8
6

Limit HP M

4
2
0
w

w/f

< 0.25

2/16

Mixture ingredients

Fig. 4 Classification according to RILEM Final Report TC-150 [2]

76
74

M2: Fr = 0.6

74

72

Flow value [cm]

Flow value [cm]

76

M2: Fr = 0.3

70
68
66
64
62

72
70
68
66
64
62

60

60
30

35

40

45
50
55
Mixing time t [s]

60

65

30

35

40

45
50
55
Mixing time t [s]

60

65

Fig. 5 Influence of the mixing time and the processing speed on the slump

131

132

WATER IN SELF-CONSOLIDATING CONCRETE PRODUCTION ITS

IMPORTANCE, HOW TO ACCURATELY MEASURE IT, AND THE
BENEFITS OF AUTOMATED WATER CONTROL
David Serra

Synopsis: Self-consolidating concretes (SCC) are produced using admixtures,

particularly high-range water-reducers. These allow a reduction in the volume of
water required, with typical water content reductions in the range of 20-30%.
The aggregate proportioning, particularly of the fine aggregates, is important to
achieve consistent quality SCC. Typically, for an SCC, aggregate mix proportions
include a higher proportion of fines, with a view to retaining higher mix
cohesiveness.
There are other variables that require control to produce a quality SCC, but the
two direct influences of water are among the most important. This paper aims to
identify the influence of water in an SCC concrete and to define the available
methods to accurately quantify the water contents in different parts of the concrete
batching process. The paper will also briefly approach beneficial side-effects of
accurate moisture measurement and control.
The most direct effects of water in concrete production are usually reflected in a
change in water/cement ratio. However there are many other equally important
effects resulting from water content changes.
This paper also outlines over the experience gained in over 2 decades in moisture
measurement in aggregate hoppers and conveyor belts to ensure the correct
aggregate dry weight proportioning. It also considers measurement of moisture in
mixers and its associated benefits in SCC production.

Keywords: aggregate, microwave moisture sensor, water control

133

David Serra is International Sales Manager at Hydronix, specialists in the

measurement of moisture in aggregates and fresh concrete. He graduated in Civil
Engineering and continued his studies with a Concrete Structures MSc from
Imperial College London. David began his career as a Structural Engineer before
moving into technical business development. He has over 10 years experience
within the construction industry, having advised clients in over 30 countries.

INTRODUCTION
With the growing popularity of specialist concretes such as self-consolidating
concrete, the requirements for information about the raw materials used has
increased considerably for manufacturers. The processes used to produce these
concretes have also had to be considered to ensure they allow the correct
proportioning, mixing and placing on site. The National Cooperative Highway
Research Programme in the USA Report 628 states that the need for adequate
quality control is much more critical with SCC than in the case of conventional
concrete. The same report goes on to say that successful production of SCC
requires greater competence and proper control of materials and equipments used
for production.
THE IMPORTANCE OF WATER
The control of the water content in the production of any concrete is essential but
as will be proven in this paper it is even more crucial in SCC production. Its direct
effect is measurable in terms of strength, fresh properties and durability. These
effects can be broken down in more detail into the following main characteristics
of concretes:
w/c and w/b a change in water content of the concrete mix will directly
affect the w/c or w/b (water/binder)
a/c changes in the water content of the aggregates will, if left unchecked,
also lead to variations in the weights of dry aggregates actually being loaded
into the concrete mix
w/p or water/powder ratio, arguably more applicable to self-consolidating
concretes, is the relationship between water and all cementitious and filler
materials in the concrete mix.
Flowability, Viscosity and Passing ability variation of water contents can
directly and indirectly (by changes in aggregate dry weight proportioning for
example) lead to changes in the concretes flowability

134

Stability unless the mix design is adhered to, for which accurate water
control is required, mix segregation, bleeding and surface settlement can
occur accompanied by their effects on appearance, durability and in the
extreme structural integrity
SOURCES OF VARIATION
Water content changes can originate from different parts of the concrete
production process with differing effects:
1.

natural moisture content variation in aggregates.

2.

accuracy of water feeding system into the concrete mixer, not usually an
issue in modern concrete plants

3.

uncovered material transport or storage systems allow increments of water

content by rainfall or their reduction by evaporation

Clearly the most significant source of moisture variation is the natural aggregate
moisture content.
THE EFFECT OF WATER VARIATION
Variation of water affects the production of concrete in many ways. For a greater
understanding of these effects it is helpful to consider the influence on the
individual constituents and concrete properties that are affected by it. To produce
a consistent SCC, the producer requires consistent materials and processes.
Knowledge of the moisture will therefore help him achieve his mix design.
Moisture will have an effect on:
Dry aggregate weights
A variation of 1% moisture content in a dry aggregate by weight results in a
change in 10 kg of aggregate loaded into the mixer for every 1000 kg of dry
aggregate weighed. Therefore the greater the variation of moisture, and the
greater the weight of aggregate used, the more serious this condition becomes.
As can be seen from Table 1 above, the range of moistures which different size
aggregates can hold as free water, increases with the fineness of the aggregate.
This occurs due to the increased aggregate surface area by weight. Due to
proportionally high use of fine aggregates in the production of SCCs (usually 4855% of total aggregate weight, source: EGSCC), the potential effects of moisture
variation increases in relation to ordinary concretes.

135

The volume of water dispensed into the mixer

Associated to the effect of changing dry weights of aggregates, the volume of
water fed into the mix will change with aggregate moisture variations. Assuming
a density of 2400 kg/m3, a 1% change will result in 24 litres of water change,
which will have considerable effects on a SCCs performance.
Further, water contents may change between the aggregate hoppers/bins and the
mixer when using inadequate material transport systems due to rain, humidity or
evaporation.
Concrete strength
Variations in water contents by weight will directly lead to changes in the w/c or
w/b (water/binder ratio) of the concrete being produced. This in turn will usually
affect the compressive and tensile strengths of the concrete.
Concrete durability, cracking and creep
Creep is related to a concrete mix w/c and therefore moisture can be indirectly
detrimental to durability. Autogenous shrinkage may also occur from varying
water contents as it is caused by the internal consumption of water during
hydration.
Issues associated with an SCCs stability (see below) may also affect the finished
surfaces permeability having an effect on durability.
Consistence
Changes in water content affect the flowability of an SCC in various ways. The
free water content itself will directly affect the flowability of the concrete, but
there are other indirect effects.
One of these indirect effects of water on flowability is the change in aggregate
grading (see effects of dry aggregates above). A change in aggregate grading will
lead to a change in the flowability characteristics of a concrete. The relationship
between fine aggregate and the total weight of aggregate (S/A) was also found to
have a significant effect by Su et al.
Another indirect effect is also caused by the change in dry aggregate grading
variations but relates to the aggregate surface area changes and how these interact
with the high-range water reducers used.
The use of viscosity modifying agents has mitigated to some extent the issues
occurring from moisture variations. Even so, a major concrete admixture
chemicals producer reported typical safety margins of 15 litres of water when
producing SS3 concretes and 7-10 litres of water when producing SCC concretes
in precast plants.

136

Mix stability
Maintaining the proportions envisaged and tested in the mix design is important to
avoid bleeding, segregation and plastic settlement. Tight water control would help
control the occurrence of these issues.
COST BENEFIT OF MOISTURE MEASUREMENT
As well as improvements in quality, the use of moisture control also allows
improved profitability for a concrete production operation.
By avoiding over- or under-cementing the concrete producer will reduce cement
costs and increase production yield. Although more difficult to quantify, the
reduced number of sub-standard batches increases profitability. The improvement
is such that the general consensus from Hydronix customers is that the
equipments return on investment period is of between 3 to 6 months.
OPTIONS FOR MOISTURE MEASUREMENT IN CONCRETE
PRODUCTION
Today there are products available in the market that allow the measurement of
the moisture content of aggregates and fresh concretes with a high degree of
accuracy. The microwave technique has emerged as the most suitable for
measurement of moisture in concrete production, mainly due to its accuracy
(usually in the range of +-0.2% to +-0.4% depending on the quality of the
calibration), its lack of influence by dust or colour, and the competitive costbenefit relationship of this type of equipment.
Measuring moisture in aggregate bins, hoppers and conveyor belts
Aggregate moisture measurement is usually carried out using a digital microwave
moisture sensor at the mouth of the aggregate hopper in question. Once the
sensor has been set up correctly, moisture readings can be taken as an average for
each batch, and the plant control system can adjust the correct dry weight of each
aggregate in real time.
Digital microwave sensors may also be placed on a feeder belt under the hopper
as shown below in Figure 5.
Measuring moisture in mixers
Measuring the moisture of fresh concrete in a mixer allows the concrete producer
to have full control over the final moisture content before the mix is discharged,
ensuring that variations due to aggregate wetting/drying during transport from the
hopper to the mixer, or variable water pressure, are eliminated.

137

The digital microwave technique is also being used to accurately measure the
moisture content of materials while they are inside the mixer. The measurements
from mixer floor moisture sensors and in-mix sensors such as the Orbiter are used
to calculate the correct volume of water required to achieve constant water
contents inside the mixer, batch after batch.
HOMOGENEITY CONTROL USING MOISTURE
Using moisture as an indicator of the degree of homogeneity achieved by the
concrete mixer at any stage of mixing is also possible using microwave mixer
sensors. Based on the use of water as an indicator of material dispersion, when
materials are loaded onto the mixer the sensors see variable moisture contents
over time as usually aggregates will be wet, cementitious materials will be dry. As
these materials mix with water and admixtures, gradually a flatter trace is
achieved as the water disperses evenly over the entire mix. A flat trace output
from the mixer sensor indicates that homogeneity has been reached (see Figure 7).
Using homogeneity control also allows the optimization of mix cycles. Once the
producer can evaluate whether the mix is homogenous or not, he can then adjust
the mixing time to ensure that only the time required to homogenize the mix is
used. This in turn has large benefits in terms of reduced power consumption,
reduced wear of parts, and increased production output.
CONCLUSION
The requirement for adequate quality control is more critical with SCC than in the
case of a conventional concrete. A simple and cost-effective improvement to
reduce fluctuations in the fresh and hardened concrete is to install accurate, digital
moisture measurement equipment in the aggregate hoppers and inside mixers.
According to the NCHRP Report 628, the moisture content,aggregate gradation
and variations in fines content of the aggregate should be continuously
monitored and must be taken into account to produce SCC with constant
characteristics. It goes on to say that it is preferable to control the moisture of
sand before every batch of SCC. The moisture content in coarse aggregate must
also be taken into account.
As well as the moisture itself being important to SCC production, it is also
straightforward to use the same equipment to control mix homogeneity before it is
discharged from the mixer.
The use of digital microwave moisture sensors allows high degrees of accuracy,
repeatable results and simple integration to the control system.

138

Finally, not only is there a high degree of improvement in quality derived from
the use of moisture measurement in SCC production but the increase in
profitability is such that the equipments return on investment is made in a matter
of months.

REFERENCES
National Cooperative Highway Research Program (NCHRP), Self-Consolidating
Concrete for Precast, Prestressed Concrete Bridge Elements, NCHRP Report
628, Transportation Research Board of The National Academies, (2009).
Su J.K. et al.: Effect of Sand Ratio on the Elastic Modulus of Self-Compacting
Concrete, Journal of Marine Science and Technology, Vol. 10, No. 1, pp. 8-13
(2002).

139

Table 1 - Typical moisture ranges for concrete aggregates

Size

Moisture %
Range
0 to 16

Fine Sand

Fig. 1

Course Sand

0 to 12

8mm

0 to 10

10mm

0 to 4

12mm

0 to 3

20mm

0 to 2

Filling height and slump flow vs. S/A ratio (source Su J.K. et al)

140

Fig. 2

Slump versus free water content of concrete (BRE,1997)

Fig. 3

Typical moisture sensor

141

Fig. 4

Fig. 5

Sensor installed under a bin

Typical installation positions for a Hydro-Probe II

142

Fig. 6

Fig. 7

Sensor installed in floor of mixer

A typical mix cycle as seen by a Hydronix mixer sensor

(Hydronix data)

143

144

PUMPING OF SELF-COMPACTING CONCRETE UNDER EXTREME

CONDITIONS
Harald Beitzel

Synopsis: However, pumping time, conveying pressure, conveying height,

conveying length as well as temperature development inside the mixer and the
influence of the ambient temperature respectively on the fresh concrete and at the
conveying lines is not yet clearly examined. Little is known of the concrete
compositions quality assurance during the casting process without segregation
and if its quality is affected under extreme conditions. It is not yet clear, if the
fresh concretes rheological behaviour, while being pumped, enables the pumping
cylinders to conduct the induction process without chocking the delivery pipe.
Furthermore, the safety against segregation during the pumping over long
distances, the fresh concretes temperature development as well as the
deformation of the delivery pipes is not yet clearly examined. Finally, both the
assumption of the pump and hydraulic pressure by vertical and horizontal
conveyance, and the expected abrasion rates during concrete conveyance over
long distances are of interest. For answering these questions, experimental
investigations have been conducted and the results will be evaluated from a
machine-technical point of view. The pumping of SCC is possible with
conventional concrete pumps without any problems. As long as the high
performance flow agent retains its effect, the essential pump pressures are
comparable to the pumping pressures used for standard concrete.

Keywords: concrete pump, concrete temperature, delivery height, delivery

length, delivery pressure

145

Harald Beitzel is Professor of Building Process Technology at the University of

Applied Sciences Trier, and Adjunct Professor for Concrete Production and
Casting at the University of Karlsruhe, as well as Head of IBU, Institute of
Building Process and Environmental Technology (Institut fr Bauverfahrens- und
Umwelttechnik), Trier, Germany. He has long-standing managerial experience in
the construction industry and conducted pioneer research in mixing and pumping
of concrete. He is a member and served on various technical committees of
RILEM and VDI.

INTRODUCTION AND PROBLEM DEFINITION

In addition to optimising the given material, adjusting the essential machinery
used to produce self-compacting concrete (SCC) is indispensable in ensuring its
ideal processing quality. If the conventional machine technology for the
production of normal (flow) concrete can also be used for SCC and which kinds
of modifications have to be made in this case cannot be answered yet. The
purpose of this paper is to analyse the influence of machine-technical parameters
on the production of SCC.
There are two different types of concrete pumps: the stationary concrete pump
and the truck-mounted concrete pump. At the same time there are two possible
mechanisms, which can be used to convey concrete:
Piston pump
Rotor pump
At present the piston pumps are operated using oil hydraulics. The delivery
pistons continue to convey the concrete using a stroke length of 8002100 mm.
Two pistons are used for the latest building constructions. Due to the specific
interplay of pressure and the suction pistons, the concrete is kept in a state of
permanent movement. As a result of the delivery cylinders pressure demands the
hydraulic oil can be loaded rod-sided or piston-sided. Due to the greater delivery
pressure, the piston sided concrete pumps generally generate good pumping
results in practice. In models using high pressure pumps, modifications to the
concrete delivery cylinders resulted in previously unimaginable lengths and
heights.
The individual pressure transmissions have been shortened for rotor pumps. A
pipe links the hopper with the barrel-shaped rotor case. In the rotor case itself, a
rotor-hose is connected in a semi-circle leading upwards. The fresh concrete is
continuously conveyed onwards due to the compression of the rotor-hose and the
rolls, which are fastened to a hydraulically operated rotor. As a result, a vacuum

146

pump in the middle of the rotor case produces low pressure, which serves to
restore the original cross-section of the rotor-hose. This system is distinctive,
because neither valves nor reductions are required. Further, at the point where the
pressure travels from one hub to another, the concrete columns hold is
considerably shorter when compared to other piston pumps. However, the
relatively short delivery pressure is a disadvantage.
The described conveyance methods for concrete pumps are depicted in Fig. 1. On
the left the piston pump is visible and on the right the rotor pump is visible.
At present there is insufficient knowledge of how SCC is transmitted with
concrete pumps. The efficient transportation of SCC with concrete pumps raises
many technical questions. Some publications assume a higher concrete supply
pressure compared to normal concrete whereas other publications maintain the
opposite. There also is no comparative data available with regards to the possible
hoisting height and transport distances, as well as the temperature development
within the concrete amongst others.
To pump SCC over long distances, several questions are of fundamental
importance. Little is known about whether the quality of the concretes
composition is maintained during a casting process without segregation and if its
quality is affected under extreme conditions. It is not yet clear whether or not the
fresh concretes rheological behaviour, while being pumped, will block the
delivery pipe of the pumping cylinders during the induction process. Furthermore,
the danger of segregation over long pumping distances, the fresh concretes
temperature development and the deformation of the delivery pipes has not been
closely examined yet. Finally, both the theoretical pump and hydraulic pressures
during vertical and horizontal transmissions and the expected abrasion rates
during the transmission of concrete over long distances are of interest.
Experimental investigations were conducted in order to answer these questions
and the results have been evaluated from a machine-technical point of view.
EXPERIMENTAL INVESTIGATIONS AND RESULTS
Composition of the SCC
When preparing a concrete mixture to pump, the following criteria must be
considered for the concrete strength classes and for its workability:
Aggregate configuration
Cement content
Cement paste content

147

Grain content
Consistency
When composing the grading curve of the sieve test for concrete aggregates, the
existing debris pores must be considered. At first they contain air and then
become more or less filled with the developing cement paste and/or fine mortar
during the preparation of the fresh concrete.
Further important aspects regarding the SCCs composition are:
The specified volumes of the sands fine material content shouldnt be sharpedged in cement depleted mixtures.
The granulometric composition should be in a convenient range (A/B) and in
the range of sand near the grading curve B according DIN 4226.
The cement content 270 kg/m, improved 300 kg/m, should be cement with a
good ability to retain water.
The mortar content should be determined in subject to the top sized aggregate:
standard values:

top sized aggregate 32 mm: 500550 dm/m,

top sized aggregate 16 mm: 550600 dm/m.

The omitted-size fraction is adequate at the lower values.
The content of ultrafine mortar must be large enough, that after applying the
manometric pressure a saturated mixture can arise.
The concrete consistency should be located in the upper range of F2 according
DIN EN 206.
The composition of the SCC for the purpose of investigating the mixing and
pumping procedure mentioned above is based on the components (left) and the
grading curve (right) presented in Fig. 2.
After the mixing process was completed, the fresh concrete, with particular focus
on the slump flow, was tested using the SCC criteria. One can see in Fig. 3 that
concrete can be pumped when the SCC criteria are fulfilled and a slump flow of
700 mm is achieved. Otherwise, the mixing process has to be repeated.
Experimental setup
The concrete was made in a single-shaft mixer. Fig. 4 shows the concrete mixers
arrangement on the mixer platform at IBUs pilot plant station.
The concrete pumps machine loads were tested under extreme conditions (height,
width and temperature) at IBUs concrete pump test plant. It was of critical
significance that the delivery pressure was regulated to guarantee the concretes

148

specific technological characteristics. The experimental setup that was used

during the experiments is depicted in Fig. 5. It pictures the tested concrete pumps
as well as the applied measuring equipment. The output of the concrete pump was
91 m/h.
During the experiments, particular focus was placed on the delivery and hydraulic
pressure of the concrete pump as well as on the valve load, the pipe deformation
over long delivery distances, the delivery heights and the temperature. At the
same time the resulting changes in the fresh concrete were being monitored.
The analysis of the pumping procedure was carried out over a test distance of up
to 90 m. This partially involved pumping the concrete through a ring circuit, with
rising and falling sections.
Fig. 6 depicts a model of concrete delivery pipes, which was used during the
experiments. In the picture the concrete pump system, which is made up of the
concrete pump and piping, is depicted alongside relevant strain gauges and
pressure sensors.
Results
In Fig. 7, the experimental results over a specific delivery time are depicted. The
data is comprised of the delivery pressure, the slump flow (Fig. 7, left), as well as
the concrete temperature (Fig. 7, right).
From Fig. 7 one can see that over a delivery period of 90 min an increase in
SCCs temperature is to be expected at greater delivery lengths and heights (Fig.
7, right). At the same time, it could be concluded that an increase in the delivery
pressure and a corresponding decrease in the slump flow is dependent on the
delivery time (Fig. 7, left).
CONCLUSIONS
The purpose of the experiments was to find determine the following for concrete
pumps:
The expected conveying pressures compared to conventional concrete.
If it is possible to continue pumping after a certain time-out.
When will SCC start losing its properties when continuously pumped.
It is possible to pump SCC with conventional concrete pumps without any
difficulties. As long as the high performance flow agent retains its effect, the
essential pump pressures are comparable to the pumping pressures used for
standard concrete.

149

Due to the fine material content and good flow ability of SCC, it can be pumped
over extreme distances, heights and heights of fall. The experiment did however
demonstrate that over greater delivery lengths, the delivery pressure did increase.

150

High pressure oil

Low pressure oil
Compressed concrete
Lighting concrete

1 Rod-sided loaded piston

2 Piston-sided loaded piston

50

100

40

80

Passing ratio [%]

Mixture composition [Vol.-%]

Fig. 1 Piston pump (left) and rotor pump (right)

30
20
10

A
B
C

60

40
20
0

0
0/4

4/8

8/16 Cement PFA Water

SP

0.125 0.25 0.5

Mesh size [mm]

Mixture components

Fig. 2 Mixture proportions (left) and grading curve (right) of the SCC

Cement

Water

Additive

Aggregate

SP
Mixing

Slump flow
700 mm

No

Yes
Concrete pump

Fig. 3 Processing sequences for the mixing of SCC

151

16

Single-shaft mixer
Mixer platform

Fig. 4 Experimental mixer setup, with mixer platform used for SCC

Stationary concrete pump

Measuring equipment

Fig. 5 Experimental setup of a stationary concrete pump, with its

measuring equipment

152

Length of conveying line

Conveying line

90 m
125 mm

Pump
Strain gauges
Flexible hose

Pressure sensors

800

80

700

60

600

40

500
Delivery pressure

20

400

Slump flow
0

300
0

30

60

90

Slump flow sf [mm]

100

Temperature
[%]
change tt [%]
Temperaturechange

change
Pressure
p
p [%]
change
Pressure

Fig. 6 Schematic experimental setup of a concrete pump system, with its

piping and the concrete pump as well as the relevant arrangement of strain
gauges and pressure sensors
100
80
60
40
20
Temperature
0
0

30

60

Time t [min]

Time t [min]

Fig. 7 Change of supply pressure, slump flow (left) and concrete

temperature (right)

153

90

154

PUMPABILITY ASSESSMENT OF C90 SCC

Said Ahmmed Jodeh and Gabriel E. Nassar

Synopsis: The objective of the investigation reported in this paper was to extend
the pumpability of a high-performance concrete with a targeted compressive
strength of 90 MPa to meet stringent requirements of the Dubai Tower project in
Doha, Qatar measuring 250 m in height requiring a pumping rate of 30m3/hr. This
objective was achieved via an extensive mixture optimization and the proper
selection of a combination of polycarboxylate-based superplasticizer and
viscosity-modifying admixture. The project also aimed at improving the
understanding of the behavior of concrete as a fluid under pressure by monitoring
several parameters, including workability, hydraulic pressure during pumping,
concrete pressure, and friction coefficient to extrapolate the various pumping
requirements to achieve the targeted height of 250 m. The investigation aimed
also at correlating workability characteristics of the concrete (T50 and V-funnel
flow) to the pumpability range of the concrete and hence the power required by
the selected pump (Putzmeister BSA1400HP) to meet the project needs.
Two mixtures were developed and tested. Mix I refers to the initial C90 mixture,
and Mix II refers the modified pumpable mixture. Key differences between them
consisted of reducing the maximum size of aggregate from 20 to 10 mm,
increasing the content of cementitious materials, enhancing the aggregate
size distribution, replacing a standard polycarboxylate-based superplasticizer by a
blend of polycarboxylate-based superplasticizer and viscosity-modifying
admixture, and finally increasing the W/C from 0.25 to 0.27.
The investigation showed that the pressure required to pump 30 m/hr of Mix I to
a height of 250 m exceeds the pumps maximum pressure as well as the pipelines
maximum allowable pressure. Therefore, while Mix I can easily meet the
physical requirements of the project (strength, durability, etc.), it failed to meet
the delivery/placement requirements. The proposed modifications of Mix I to Mix
II and the addition of the new combined admixture resulted in more than 80%
reduction of the concretes friction coefficient (), while maintaining the physical
properties of hardened concrete. Field trials proved that Mix II could be pumped
at 250 m with the chosen piston pump functioning at a safe value of 75% of the
capacity of the pump.
Keywords: High-strength concrete, polycarboxylate-based superplasticizer,
pumping, viscosity-modifying admixture, self-consolidating concrete.

155

Said Ahmmed Jodeh, Technical Manager of Al Wataniya Concrete in Qatar. He

holds a Bachelor Degree in civil engineering from An Najah National University
in Palestine. He is an ACI Concrete Construction Special Inspector and holds
many certificates from ACI, ASTM, PCS, and QCS.
Gabriel E. Nassar, Resident Manager of Sodamco Qatar. He has more than 15
years experience in the Ready Mix Concrete Operation, including Technical
manager and General Manager position at BCL Ready Mix (Beirut, Lebanon). He
holds of a Bachelor Degree in civil engineering from RMIT Melbourne Australia.
He is a member of the Institute of Engineers of Australia and the ACI Lebanon
Chapter.
INTRODUCTION
Self-consolidating concrete (SCC) is a highly flowable material that can flow and
consolidate under its own weight. At the same time, SCC should be designed to
be cohesive to fill the formwork of almost any size and shape without segregation
or bleeding. SCC is a generic term for mix design; however, the performance of
SCC differs with the mixture proportioning and admixtures types and contents.
The fluidity of SCC ensures high level of workability, while the rapid rate of
placement provides an enhanced surface finish. SCC can reach a high overnight
compressive strength of 30 to 40 MPa enabling an easier and more reliable
demoulding. Long-term compressive strength can exceed the 100-MPa barrier.
Production and quality control of SCC is quite demanding. Compared to
traditional concrete, it is more difficult to control the flowing properties and
stability at the same time in case of SCC. In some cases, it could be difficult to
obtain a high flowability without segregation. Therefore, it is important to
optimize the entire concrete composition from the solid characteristics to the paste
composition and admixtures in use.
High-rise building construction constructed using stationary pumps with static
pipelines and stationary placing booms shows an increasing importance,
especially when high-strength concrete is designed to withstand heavy loading,
and self-consolidate in areas of congested steel reinforcement. In high-rise
construction, the concrete shall also be pumped to significant heights, which
arises the following questions: can the concrete be pumped to the desired height,
and how to calculate the flow resistance in the delivery line? This paper gives an
overview of how to assess pumpability and pumping performance as well as how
to optimize concrete mixture proportioning to enhance pumpability.

156

BACKGROUND
To build as fast as possible, concrete is being pumped through pipes to make it
easily reach its final destination. In this pumping stage, some problems could
occur which cause delays and issues difficult to overcome. Sometimes, the
concrete is not suitable for pumping or it was designed to be suitable, but due to
setting or loss of workability, the concrete was no more pumpable at the time of
pumping. This could cause blocking in the pipeline. Therefore, proper quality
management of the concrete mixture must be carried out to anticipate and resolve
these issues. The major idea of prior quality management is the application of the
concrete composition and other properties of the concrete to predict long-term
properties. For high-rise pumping of high-strength concrete, prior quality
management was used to assess the properties of conventional slump concrete
that can influence pumpability. The expected pumping pressure and pump
performance were calculated by determining the average friction factor and the
required motor power for a certain pumping height.
Basic knowledge about pumping concrete
The pumping process can be executed with a variety of pumps and pipes.
Differences in pumps and pipes can influence the pumpability of the concrete due
to variations in pumping conditions. In the first part of this section, the equipment
used to pump concrete will be discussed. This will be followed by a discussion of
the concrete behavior inside the pipes during the pumping.
The most commonly used pump to deliver concrete in high-rise buildings is the
piston pump for which the working principle consists of pushing a piston back
and forward to deliver concrete. While going backward, the piston creates a
vacuum which pulls the concrete in the piston, and during forward movement it
pushes the concrete through the pipe.
When selecting the pump, the important factors to consider are the maximum
pressure, the length of a stroke, and the duration of a stroke. The last two factors
are important because they determine the flow rate of the concrete leaving the
hopper. The maximum pressure is important because it determines the friction
that could be overcome with the pumping equipment. The pressure applied for
pumping can vary from 40 to 300 bars.
The most significant factor for the pipe is its diameter. Commonly, a pipe
diameter of 100 to 200 mm is used. The difference in diameter is important
because it is one of the factors influencing the velocity of the concrete in the pipe.
Smaller pipe diameter results in higher velocities, hence resulting in an increase in
friction leading to greater pumping difficulties.

157

In order to better understand the pumping process, the behavior of concrete during
the pumping process is explained. Although direct observation is impossible, the
flow of concrete in straight pipes in stationary (steady-state) conditions can be
described by two parts. a plug and a lubricating layer (Figure 1), as can be seen in
centre part, which is called the plug flow, consists of aggregate, sand, cement, and
water. It is important to notice that the velocity over the full width of the plug is
equal. This means that there is no relative velocity within the plug, and that the
forces acting in this plug are small. The outer part, called the lubricating layer, has
a thickness of 1 to 2,5 mm and consists of water, cement and fine sand particles.
The velocity profile in the lubrication layer drops from the value of the flow
velocity of the plug to zero at the pipe wall. This means that in this layer, some
particles are moving faster than other particles, and that several forces are acting
within this layer. This behavior of concrete is only correct in a stationary (steadystate) flow in straight pipes. The flow pattern in non-straight sections, such as
bends, reducers, or bad connections, will be turbulent inducing changes in the
flow pattern and the velocity profile. These changes result in a higher sensitivity
for blocking (Browne & Bamforth, 1977 and Neville, 1997).
Describing pumpability
The main characteristic of concrete for determining its pumpability is workability.
According to Neville (1997), the definition of workability is that property of
freshly mixed concrete or mortar which determines the ease and homogeneity
with which it can be mixed, placed, consolidated, and finished. This means that
the workability is not a fundamental property, but it is related to the method of
placing, consolidating, and finishing. This makes it difficult to measure the
workability of concrete. But, if a concrete has adequate workability, it should also
exhibit good pumpability.
Focusing on the definition of the workability again, it becomes clear that it is a
composite property. The two main components are the consistency and
cohesiveness of the concrete, which are described by the fluidity and segregation
resistance. A good workability is obtained if both the consistency and
cohesiveness are sufficient, leading to a good pumpability as well. Consistency
and cohesiveness of the concrete are influenced by constituting materials in use.
The most important ingredients which influence the consistency and cohesiveness
are the fines and aggregates, cement content, water content, and the use of
admixtures and mineral additives. For better understanding of pumpability,
consistency and cohesiveness are discussed further below (Mehta & Monteiro,
2006, Neville, 1997).

158

COHESIVENESS
The cohesiveness of the concrete is the resistance to bleeding and segregation,
which is also named the stability of the mixture. This property is important
because it describes the homogeneity of the concrete. Bleed water needs to
transfer shear forces to the other materials in concrete. If the cohesiveness is not
sufficient, these forces could not be transferred to the other materials, and
bleeding would occur. The process of bleeding is illustrated in Figure 2. Due to
particle interlock, high friction occurs, and because of a lack of cohesiveness the
forces on the water could not be transferred to the other materials. This results in
water migration between the aggregate particles, resulting in an unsaturated state.
The bleeding of concrete is a problem because concrete in the unsaturated state
will result in high friction flow. As can be seen in Figure 3, the flow resistance
will increase when a concrete is in the unsaturated state. This flow resistance will
certainly be larger than the maximum pressure that the pump could deliver with
blocking as a consequence.
The other property that is important for pumpability is the cohesiveness that
affects the resistance to segregation. Segregation means that the aggregate
particles sink in the mixture. During pumping, aggregates can move further ahead
of the concrete mass in the pipe, relative to the mortar. If the cohesiveness is not
sufficient, aggregate particles could move to the front of the concrete causing a
plug in front of the concrete. This plug will cause high friction in the pipe and
may lead to blocking. To prevent segregation or separation of coarse aggregate
particles from the mortar, the pumping pressure may have to be lower to a value
lower.
The cohesiveness of concrete is not often measured. In most cases, it is estimated
by the normal bleeding test or visual inspection of the slump test. The normal
bleeding test is a time consuming test, as it takes several hours, and the visual
inspection is very subjective as it is dependent on the experience of the operator.
Another test which can be used is the pressure bleeding test (Mehta & Monteiro,
2006, Neville, 1997).
CONSISTENCY
The consistency is defined as the ability to flow, which means the effort needed to
make concrete flow. A concrete with good consistency has low viscosity. The
consistency is an important property for pumpability because it is related to the
friction in the pumpline. Consistency is evaluated by the slump test. According to
Neville (1997), the slump needs to be between 50 and 150 mm for the concrete to
be pumpable. In case the slump is lower than 50 mm, the concrete would be too
stiff resulting in high friction and more difficulties in pumping. A higher pressure

159

is needed to push the concrete through the pipes which will increase the risk for
bleeding or segregation.
To create a mixture composition which will result in good consistency, special
attention must be paid to the grading curve of the aggregate. When there will be a
high amount of coarse aggregates, the aggregates can increase the friction against
the pipe wall if the consistency will is too low. When the mixture has too many
fines, flow friction will also increase. As mentioned previously, the fine sand
particles are part of the lubricating layer in the pipe (Anderson, 1977, (Neville,
1997).
TEST METHODS
This section offers an overview of test methods and their requirements that can be
carried out to ensure successful pumping. The slump flow test is easy to carry out
and can provide an indication of filling ability of SCC. The requirements on final
slump flow and T50 values should be different for mixtures with different
maximum aggregate size and aggregate shape and when using different
admixtures. It is difficult to assess the segregation/settlement tendency of the
concrete using this test. To perform a slump flow test, the fresh concrete is poured
into a mould in the shape of a frustum of a cone. When the cone is lifted upwards,
the distance the concrete has spread provides a measure of the consistency of the
concrete.
The V-funnel flow time can b used to estimate the apparent viscosity of SCC.
However, many factors play a key role in influencing the V-funnel test results,
including the amount, shape and size distribution of the aggregate and the
viscosity and volume of the cement paste. A concrete rheometer can be used to
measure the yield stress and plastic viscosity of the concrete. However, for the
case study discussed in this paper, a concrete rheometer was not used available for
use.
The average friction factor, the expected pressure, and required motor power for a
given pumping height can be assessed by means of measuring the pumping
pressure and calculating the performance of the concrete pump.
TEST RESULTS
Concrete composition
The mixture composition for the C90 concrete that was used for the main core
wall of the high rise structure was optimized to ensure proper pumpability.
Mixture optimization necessitated changes in the water-to-binder ratio, sand-total
aggregate ratio, aggregate maximum size, and the type of superplasticizer. The
results of the mixture optimization are presented in Table 1 and discussed below.

160

The concrete compositions and properties are given in Table 2. The variation in
slump flow and V-Funnel flow time results over 150 minutes of the two tested
mixtures are plotted in Figures 4 and 5, respectively.
INFLUENCE OF SCC ADMIXTURE
The polycarboxylate-based superplasticizer that was incorporated was a blend of
superplasticizer and viscosity-modifying admixture specifically designed for use
in SCC and high-performance concrete. This specialty admixture helps achieve
very high concrete flow while maintaining a adequate segregation resistance
(Figure 6). The admixture also makes the concrete more tolerant to variations in
the water content of the concrete mixture in such way that plastic viscosity is
maintained and segregation is prevented. By making the concrete more cohesive,
the separation of the lubricating mortar fraction from the mixture is prevented
during pumping. SCC made with the special SCC admixture was shown to
maintain consistency for extended periods of time, especially under extreme
conditions where the concrete has to be pumped for long distance.
Pump trial measurements and flow rate calculations
Data Sheet for two pump trials of Mix II and Mix I are summarized in Tables 4
and 5, respectively. Based on the pump trial the following calculations can be
executed. The total concrete volume per stroke was calculated from the concrete
pump delivery piston dimensions and is equal to 0.054 m3. The number of strokes
per hour is calculated as 11 x 60 = 660 Stroke/h for Mix I and 19 60 =1140
Stroke/h for Mix II. Assuming a 90% filling ratio, the concrete flow rate can be
calculated from the following equation:
Q = V S F

(1)

where
V : volume of delivery line (m3)
S : number of strokes per hour
F : filling ratio
This results indicate that the flow rates for Mix I and Mix II were 32.1 m3/h and
55.4 m3/h, respectively.
Concrete pressure
The surface ratio ( i) between the delivery piston (green in the illustration in
Figure 7) and the hydraulic piston (red) determines the hydraulic transmission
ratio. The output Q and concrete pressure Pc depend on this ratio.

161

With a rod side pressurization of the hydraulic system, the transmission is more
efficient compared to the piston side pressurization. This results in a higher
delivery performance but a lower concrete pressure, while for piston side
pressurization, a higher concrete pressure is attained but lower delivery
performance can be achieved.
According to the following physical formula:
Phy A1/2

Phy
i
the concrete pressure Pc can be calculated from the hydraulic pressure Phy, as
follows:
Phy A1 / 2

Pc A3

Phy A1 / 2

Phy

A3
i
The hydraulic pressure can be read from the pump display and control panel.
There is always a system operation pressure of approximately 30 bar, which has
to be deducted from the pressure reading of the manometer at the pump before
calculating the concrete pressure.
Pc

Phy

(2)

A3

30

(3)

(4)

where
Pc : concrete pressure (bar)
i: hydraulic transmission ratio
This results in a concrete pressure of 155 bar for Mix I and only 92 bar for Mix II.
The pressure calculation is based on two positions of the pumping system. The
first position is the current high (Y1+Y2 = 113 m), while the second position is
the target high (Y1+Y2 = 250 m) as can be seen in Figure 8.
In position 1, the friction coefficient is calculated based on site trial data for both
concrete mixtures (Mix I and Mix II), while for position 2, the concrete pressure
is estimated based on the calculated friction coefficient for both concrete
mixtures, as illustrated below.

162

According to the Energy Equation (Bernoullis law) for a fluid, the total energy
can be summarized as the elevation energy, velocity energy, and pressure energy.
The Energy Equation can then be expressed as:

p1 + v12 / 2 + g h1 = p2 + v22 / 2 + g h2 + ploss

(5)

where
p = pressure in fluid (Pa (N/m2), psi (lb/ft2))
ploss = pressure loss (Pa (N/m2), psi (lb/ft2))
= density of the fluid (kg/m3, slugs/ft3)
v = flow velocity (m/s, ft/s)
g = acceleration of gravity (m/s2, ft/s2)

h = elevation (m, ft)

1 : refers to energy at point 1 (Figure 9)
2 : refers to energy at point 2 (Figure 9)

The concrete velocity in both positions is equal: v1 = v2 and with h1 = 0, h2=

Y1+Y2, p2=0 (zero gage pressure), Equation 5 can then be transformed to:
(6)

ploss = p1 - g h2

The pressure loss is divided in major losses due to friction and minor losses due
to changes of velocity in bends, valves and similar situations. The minor losses
are considered as negligible.
The pressure loss in pipes depends on the flow velocity, pipe length, pipe
diameter, and a friction factor based on the roughness of the pipe, and whether the
flow is turbulent or laminar, expressed by the Reynolds Number of the flow. The
pressure loss in a tube due to friction, called the major loss, can be expressed as:

163

(7)

ploss = (l / dh) ( v2 / 2)
where
ploss = pressure loss (Pa, N/m2)
= friction coefficient
l = length of duct or pipe (m)
dh = hydraulic diameter (m)

Equation (7) is also called the Darcy-Weisbach Equation, valid for fully
developed, steady incompressible flow. By applying the above equations in both
positions, using different concrete compositions, the results are summarized
below.
The field trials carried out on Mix I and Mix II with a total pipe length of 195 m, a
vertical length of 113 m, a pipe diameter of 125 mm using a Putzmeister
BSA14000HP concrete pump can be summarized below.
T500
Mix 1

9.2 s

Mix 2

3.4 s

Variance

-62%

Flow
735
mm
720
mm
-2%

VFunnel
30.2 s
6.4 s
-79%

164

Hydraulic
Pressure
300-310
bar
180-200
bar

Strokes /
min

Pump
output (Q)

11

32.1m3/hr

19

55.4m3/hr

-38%

+73%

+73%

The field trial results are extrapolated below for a pipe vertical length of 250
m.
Mix I
Pump output
Concrete
(Q)
Pressure
Values from
trials at
Height=113m
Calculated
values for
Height=250m
Hydraulic
Pressure
required to
pump 30m3/hr
at
Height=250m

32.1m3/hr

Mix II
Pump output
Concrete
(Q)
Pressure
55.4m3/hr

92 bar

250 bar*

30m3/hr
(min.
requirement)

91.5 bar

30m3/hr
(min.
requirement)

155 bar

480 bar
(37% more than pump
maximum capacity)

195 bar
(75% of pump maximum
capacity)

Mix I concrete pressure exceeds safe pipeline operating pressure of 200 bar.

CONCLUSIONS
Based on the field trials to evaluate the pumpability of C90 SCC to enable
concrete delivery to a height of 250 m, the following conclusions can be stated:
1. The pressure required to pump 30m/hr of Mix I to a height of 250m exceeds
the pumps maximum pressure as well as the pipelines maximum allowable
pressure. Therefore, while Mix I easily meets the physical requirements of the
project (strength, durability, etc.) it fails to meet the delivery/placement
requirements.
2. The modification of the mixture proportioning of Mix I to develop Mix II that
included the incorporation of a novel SCC admixture that combines a lowviscosity polycarboxylate-based superplasticizer and a special viscositymodifying admixture has resulted in more than 80% reduction of concretes
friction coefficient (), while maintaining the required physical properties of
hardened concrete.
3. Field trials proved that Mix II can now be pumped successfully at 250m with
the chosen piston pump functioning at a safe value of 75% of its capacity.

165

REFERENCES
1. ACI Committee 304. (1996, January). Placing Concrete by Pumping Methods.
Retrieved September 26, 2007, American Concrete Institute: www.concrete.org.
2. Anderson, W. G. (1977). Analyzing Concrete Mixtures for Pumpability. ACI
Journal 74-42, 447-451.
3. Browne, R.D., Bamforth, P.B. (1977). Tests to Establish Concrete Pumpability.
ACI Journal, 74-19 , 193-203.
4. EuroLightCon. (2000). Stability and Pumpability of Lightweight Aggregate
Concrete Test Methods.
5. Ferraris, C.F. (1999). Measurement of the Rheological Properties of High
Performance Concrete: State of the Art Report. Journal of Research of the
National Institute of Standards and Technology , 461-478.
6. Kaplan, D., Larrard, F.D., and Sedran, T. (2005). Avoidance of Blockages in
Concrete Pumping Process. ACI Materials Journal, 102-M21, 183-191.
7. Khayat, K.H., Assaad, J., and Daczko, J. (2004, March-April). Comparison of
Field-Oriented Test Methods to Assess Dynamic Stability of SCC. ACI Materials
Journal , 168-176.
8. Koehler, E.P., Fowler, D.W. (2003, August). Summary of Concrete Workability
Test Methods. Retrieved August 22, 2007, International Center for Aggregates
Research: http://www.icar.utexas.edu/publications/105/105_1.pdf
9. Larrard, F.D., Hu, C., Sedran, T., Szitkar, J.C., Joly, M., Claux, F., et al.
(1997). A New Rheometer for Soft-to-Fluid Fresh Concrete. ACI Materials
Journal , 234-243.
10. Mehta, P. K., Monteiro, P. J. (2006). Concrete - Microstructures, Properties
and Materials. New York: McGraw-Hill.
11. Neville, A. (1997). Properties of Concrete. New York: John Wiley & Sons,
Inc.

166

12. Pashias, N., Boger, D.V., Summers, J., and Glenister, D.J. (1996). A Fifty
Cent Rheometer for Yield Stress Measurement. Journal of Rheology , 1179-1189.
13. Powers, T., Dahl, L. (1939). The Bleeding of Portland Cement Paste, Mortar
and Concrete Treated as a Special Case of Sedimentation. Chicago: Portland
Cement Association.
14. Zhang, M.Y. (2003). Coal-Water Paste Water-Seeping Properties under
Pressure and its Pumpability. Journal of Engineering for Thermal Energy &
Power, 78-81 (Chinese version).
15. Christian Stirm (2009). Dubai Tower Doha pump trial report.
16. Robert W., Alan T., and Philip J., Introduction to Fluid Mechanics. John
Wiley & Sons, Inc.

167

Table 1 Mixture proportioning and test results

Mixture proportioning
Total binder, (kg/m3)
W/C
Max. Aggregate (mm)
Type of Aggregate
Silica Fume, (kg/m3)
P.F.A (Fly Ash), (kg/m3)
OPC (portland cement),
(kg/m3)
5- 20 mm (kg/m3)
5- 10 mm, (kg/m3)
Sand 1, (kg/m3g)
Sand 2, (kg/m3)
PC
VMA110

Mix I
500
0.25
20
Limestone
40
138
322

Mix II
550
0.27
10
Limestone
55
95
400

528
473
804
0
7.5
0

0
892
503
310
0
8.0

168

Table 2 Mixture proportioning and test results

V-Funnel (Sec)

T500 (Sec)

Flow Slump
(mm)

Test
Air Content (%)
Concrete Temperature (C)
Visual Stability Index

Duration (min)
(Elapsed Time)
60
60
60
5
30
60
90
120
150
5
30
60
90
120
150
5
30
60
90
120
150

Mix I

Mix II

0.8
25
0
750
745
730
680
650
520
4.41
6.16
7.02
7.45
9.02
11.51
20.31
20.54
24.5
31.21
39.13
42.36

0.7
25
1
730
755
755
740
760
755
3.26
2.35
3.29
3.2
3.06
2.5
10.32
7.49
11.54
10.54
9.48
10.24

Table 3 Hardened properties for Mix II

TEST
Compressive strength at 1 day (MPa)
Compressive strength at 7 days (MPa)
Compressive strength at 28 days (MPa)
Compressive strength at 56 days (MPa)
Flexural strength at 28 days (MPa)
Chloride content (%)
Sulfate content (%)
Modulus of elasticity at 7 days

169

STANDARD
BSEN
BSEN
BSEN
BSEN
BSEN
ASTM
ASTM
ASTM

RESULT
39
75
95
104
8
0.01
0.39
42.0

Table 4 Data sheet for pump trial Mix II

Concrete Property

Grade

Conc.
batch
size
(m3)

90/10

27

First
Concret
e mixer
arriving
Time
6:50:00
5:40 PM PM

Concrete
Batching
Time

T500
(s)*

Flow
Slump
*
(mm)

Concrete
temp.* (C)

V
Funnel*
(s)

3.4

720

31.4

6.4

Stationary concrete pump

Machine
Model
BSA 14000 HP-D

Machine
No.

Pipeline
Horizontal length (m)

57.5

Vertical length (m)

103

Placing boom
Type of placing boom
Placing boom
operation position

32m
Horizontal length (m)
Vertical length (m)

24
10

Measurement (Time point: 12:00AM )

Hydraulic pressure in
the middle of stroke
(bar)

Average Strokes/Min
19

Before discharging one mixer

Stroke number for one
After discharging one mixer
mixer
Strokes total for mixer
* Average measured values

170

180 - 200
608656
609211
555

Table 5 Data sheet for pump trial Mix I

Concrete Property
Conc.
batch
Grade
size
(m3)
90/20

19

Concrete
Batching
Time

Concrete
arriving
Time

T500
* (s)

Flow
Slump*
(mm)

Concrete
temp.*
(C)

V
Funnel
* (s)

9.2

735

30.5

30.2

10:32 11:200:00
PM PM

Stationary concrete pump

Machine
Model
BSA 14000 HP-D

Machine No.

Pipeline
Horizontal length (m)

57.5

Vertical length (m)

103

Placing boom
Type of placing boom
Placing boom
operation position

32m
Horizontal length (m)
Vertical length (m)

24
10

Measurement (Time point: 11:30PM )

Average Strokes/Min
11
Before discharging one
mixer

Hydraulic pressure
in the middle of
stroke (bar)

Stroke number for one

mixer
After discharging one mixer
Strokes total for mixer
* Average measured values

171

300 - 310

609784
610184
400

Figure 8: Plug flow (Browne & Bamforth, 1977)

Figure 2: Blocking due to particle interlock (Browne & Bamforth, 1977)

Figure 3: Flow friction (Browne & Bamforth, 1977)

172

Flow (mm0

Flow Slump Test

800
750
700
650
600
550
500
450
400
0

20

40

60

80

100

120

140

160

Elapsed Time (min)

Mix I

Mix II

Figure 9: Test results of slump flow results

V Funnel Test
50

(Sec)

40
30
20
10
0
0

20

40

60

80

100

Elapsed Time (min)

Mix I

Mix II

Figure 5: Test results of V-funnel flow results

173

120

140

160

Figure 6: Slump Flow Test for Mix II

Figure 7: Configuration of the piston pump

174

Figure 8: Technical data sheet of pump

175

Figure 9: Pumping system schematic diagram

176

CAN ORDINARY VIBRATED CONCRETE AND SELFCONSOLIDATING CONCRETE BE TREATED IN THE SAME WAY
CONCERNING PUMPABILITY?

Dimitri Feys, Geert De Schutter, Ronny Verhoeven, and Kamal H. Khayat

Synopsis: Pumping of concrete is one of the most applied construction methods,

due to its easiness of application and the high construction rate. For ordinary
vibrated concrete (OVC), a set of theoretical and practical studies has been
performed and provides guidelines to improve the concrete pumping performance.
With the introduction of Self-Consolidating Concrete (SCC) on the market, the
pumping process did not change, but can the same rules for OVC be applied on
SCC?
This paper compares the results of full-scale pumping tests on SCC with literature
results for OVC. It is shown that the velocity profiles are different for OVC and
SCC, as SCC is also sheared in the pipes. Bends cause additional pressure losses
which appear to be higher in case of SCC, but no final conclusion can be drawn
due to the large scatter on the results. The air content increases and the air bubbles
size decreases in case of SCC, which is completely opposite compared to OVC.
The physical causes, pressure and shearing, causing this difference are explained
and their consequences on the rheological properties of SCC are demonstrated

Keywords: air, pumping, pressure, rheological properties, SCC, velocity profile

177

Dimitri Feys graduated in 2004 as a civil engineer at Ghent University. Since

then, he focused his research on the interactions between rheological properties
and pumping of Self-Compacting Concrete, which formed the basis for the
doctoral degree he obtained at Ghent University in May 2009. Since September
1st 2009, he is affiliated as a post-doctoral fellow with the concrete research
group of the Universit de Sherbrooke, Qubec, Canada, under the lead of Prof.
Kamal H. Khayat.
Geert De Schutter is full professor at Ghent University, Belgium. He is doing
research in the field of concrete technology, at the Magnel Laboratory for
Concrete Research, Department of Structural Engineering. He is laureate of
several national and international awards, among which the important
Vreedenburgh Award in 1998 and the prestigious international RILEM Robert
LHermite Medal in 2001. This year (2009) he is also invited professor at the
University of Cergy-Pontoise, near Paris, France.
Ronny Verhoeven is a full professor at Ghent University. In 1983, he finished his
Ph-D on waterhammer in pipes. Since 1989, he is the head of the hydraulics
laboratory, department of civil engineering, Ghent University. From 1998 until
2002, he was dean of the faculty of engineering and since 2004 until now he is
director of studies at this faculty. He is guest professor at the Catholic University
of Leuven (Belgium) and Honorary professor at the Wroclaw University of
Environmental and Life Sciences (Poland).
Kamal Henri Khayat is Professor of Civil Engineering and Director of the
Integrated Research Laboratory in the Valorization of Materials and Innovative
and Durable Structures at the Universit de Sherbrooke in Canada. He is Fellow
of the American Concrete Institute and serves on several technical committees of
ACI, RILEM, and CSA, including Chairman of ACI 237 on SCC and RILEM TC
on Mechanical Properties of SCC.

INTRODUCTION
Pumping of concrete is one of the most used casting methods, well known for its
ease of application, the possibility to apply it in many different situations
(especially in the ready-mix industry), the high construction speed, etc. [1-2].
Many theoretical and practical studies have been performed on the pumping of
concrete in order to investigate several aspects such as different pumping systems
[1-4], the pressure required to pump [3-7], the configuration of the conveying line

178

[1-2] and the consequences of pumping on the concrete properties [4, 8-10]. With
the introduction of Self-Consolidating Concrete (SCC) on the market, the
practical rules for pumping of concrete are still applied. The question remains
whether the same phenomena influence the pumping parameters and whether
SCC shows similar changes in properties due to pumping as Ordinary Vibrated
Concrete (OVC).
This paper compares the results obtained in a research project on pumping of SCC
executed at Ghent University, verified by similar projects at other institutes, with
results available in literature dealing with pumping of OVC. After a brief
introduction into concrete rheology, the experimental setup is described and the
results for SCC are compared with literature. The velocity profiles of OVC and
SCC flowing in pipes are discussed in detail, followed by a study on the influence
of bends on the pumping pressures, the change in air content and air bubbles size
distribution due to pumping and possible variations in rheological properties.

RHEOLOGY OF FRESH CONCRETE

Steady-state behaviour
The rheological properties of fresh concrete can be divided into two major
classes: the steady-state properties and the transient behaviour [11-12]. Steadystate means that no time effects (transient behaviour) influences the properties
during the measurement or application of the concrete. In steady-state, the
rheological properties of fresh concrete are mostly described by the Bingham
model (eq. 1) [11]. This model states that the yield stress must be exceeded before
any flow starts and that a further increase in shear rate is determined by the plastic
viscosity. In some cases, with SCC and when high shear rates are involved, a
deviation of the Bingham behaviour has been observed and the SCC is said to be
shear thickening [13-15]. In this case, the viscosity increases with increasing
shear rates. A detailed study of shear thickening is out of the scope of this paper,
but it is mentioned as it has an influence on the results obtained for SCC.
0

where

(1)

= shear stress (Pa)

0 = yield stress (Pa)
= plastic viscosity (Pa s)
= shear rate (s-1)

179

From a rheological point of view, the main differences between OVC and SCC
are the lower yield stress in case of SCC, which can be observed by executing the
slump test, and, in general, a higher viscosity for SCC [16].
Transient behaviour
The transient behaviour can be described as the change of the steady-state
rheological properties with time. Theoretically, the transient behaviour can be
divided into three different parts: thixotropy, structural breakdown and loss of
workability [12].
Thixotropy is defined as the reversible breaking of physical connections between
small particles due to shear, and the rebuilding at rest [17]. The most interesting
part of thixotropy in practice is the rebuilding at rest because it can reduce the
formwork pressure exerted by the fresh concrete cast [18]. Thixotropy can be
visualised by monitoring the stress response when a sudden increase and decrease
in shear rate are applied (Fig. 1) [19]. When the shear rate suddenly increases, an
overshoot in the stress is observed and the stress decreases with time (at constant
shear rate) until the equilibrium value is achieved. In case of a sudden decrease in
shear rate, the opposite happens: the stress is too low and rebuilds as a function of
time until the equilibrium value is obtained. At rest, there is no equilibrium value
for the stress, and as a result, the static yield stress (yield stress at rest) increases
linearly with time [20].
As thixotropy is of physical nature, induced by inter-particle forces, structural
breakdown has a chemical nature [11-12]. At the first contact between cement and
water, chemical reaction products are formed which can bond some cement
particles. When the shearing forces are sufficiently high, these connections can be
broken. The effect of structural breakdown due to a sudden increase in shear rate
is similar as for thixotropy, but there is no rebuild when the shear rate decreases.
Loss of workability is regarded as the non-reversible increase in yield stress and
plastic viscosity with time. It can be both physical and chemical of nature, but
finally, the chemical reaction between cement and water transforms the material
from a liquid into a solid.
In practice, the distinction between thixotropy and structural breakdown is not
easy to make. In this paper, the decrease in shear stress at a constant shear rate
will be described as structural breakdown, but it includes both the thixotropic
breakdown and the pure structural breakdown described above. As a consequence,
there exists an equilibrium shear stress at each shear rate and the lower the
number of connections in the concrete, the lower the rheological properties. As a
result, the equilibrium yield stress and plastic viscosity are lower at higher shear
rates.

180

EXPERIMENTALS
Test setup
Full-scale pumping tests were performed inside the laboratory, using a truckmounted concrete piston pump (Fig. 2). This pump has two pistons, containing
theoretically 83.1 liter each, alternately pushing concrete inside the pipes and
pulling concrete from the reservoir. The operator can vary the discharge in ten
steps from 4-5 l/s (step 1) to 40 l/s (step 10). For safety reasons, a maximum
discharge of 20 l/s (step 5) was set during the experiments.
Behind the pump, two different loop circuits were installed. The short circuit
measured 25m and consisted of a straight horizontal section of 12m, a 180 bend
constructed with two 90 bends and a 1m straight pipe, and an inclined part
heading back towards the pump (Fig. 3). At the end of the circuit, a reservoir with
a valve at the bottom was suspended to a rolling bridge, which enabled sampling
and discharge calibration. During tests, the valve was open and the concrete
flowed back inside the reservoir of the pump. The long circuit measured
approximately 100m and was built as an extension of the short circuit with four
straight horizontal sections and 180 bends in between (Fig. 4). The principle is
similar: the circuit formed a loop and the concrete flowed back inside the
reservoir of the pump.
In the main measurement section, which is the only measurement section in the
short circuit and the last straight horizontal section in the long circuit, the pressure
loss was measured with two flush-mounted pressure sensors (Fig. 5). The pressure
loss per unit length was calculated by dividing the difference between the
upstream and downstream pressure by the length between the two sensors. In the
vicinity of the each pressure sensor, three strain gauges were attached to the outer
pipe wall, initially as a back-up for the pressure sensors (in the short circuit) and
finally as independent measuring systems in other sections in the long circuit [3,
6-7]. The deformation of the pipe is related to the local pressure. Close to the
strain gauges, a temperature sensor monitored the temperature evolution.
Discharge was not measured directly, but it was calculated by measuring the time
interval for a certain amount of pumping strokes. As verification, the content of a
full pumping cylinder was pumped inside the suspended reservoir with the valve
closed. The load cell between this reservoir and the rolling bridge registered an
increase in force, which is related to the discharge when density is known.
Comparison of these data with the time measurements shows a very good
agreement between both methods.

181

Testing procedure
As the volume of concrete needed was around 1.5m for the short circuit and
3.5m for the long circuit, the concrete was produced and delivered by a readymix company. In normal circumstances, the concrete arrived at the laboratory at
an age of 45 minutes (relative to the water adding time) and after insertion of the
concrete in the pipes, the first test could start at 1 hour of age. Two different
testing procedures were executed during the investigation. The regular testing
procedure consists of decreasing the discharge stepwise from step 5 to step 1,
maintaining each discharge during 5 full strokes. This tests takes around 4-5
minutes and was repeated each 30 min, until the concrete was removed from the
site. Before this regular procedure, a sample of the concrete was taken and
subjected to a rheometer test with the Tattersall Mk-II rheometer and to different
tests on fresh SCC, as slump flow, V-funnel, density, sieve stability, etc.
The special testing procedure consists of three tasks, repeated for each
discharge step (Fig. 6), after the insertion of the concrete at the lowest discharge
(step 1). The three tasks are as follows:
Maintaining discharge constant until equilibrium in pressure loss is
observed. In practice, the full amount of concrete was pumped through
the circuit. Especially at the low discharges, this task can take several
minutes.
Sampling and discharge calibration (with the reservoir and the load
cell). The concrete sample is used to determine the rheological
properties, slump flow, V-funnel flow time, air content, density and
sieve stability (segregation resistance).
Stepwise decreasing discharge curve, similar as the regular procedure,
but with the discharge from task one as maximum discharge.
Logically, at discharge step 1, there is no decreasing discharge curve.
After the decreasing discharge curve, the discharge is increased by one step and
the tasks are repeated. This procedure takes between 60 and 90 minutes, and due
to some blocking issues during the insertion of the concrete in the long circuit, the
concrete age was 3h 30 min at the end of the test in some cases. As a result, due to
high pressures or high concrete ages, discharge step 5 was cancelled sometimes.
Concrete compositions
In total, 19 concrete mixtures were produced at the mixing plant, of which the
compositions can be found in table 1. Mixtures LM 0, OVC 1, LM 15 and LM 17
were commercial products of the mixing plant, while the other mixtures were
designed by the laboratory. The laboratory mixes contained Ordinary Portland
Cement (CEM I), limestone filler, sand, rounded aggregates, water and a

182

polycarboxylate superplasticizer (SP) with a long workability retention. The

commercial mixes contained the same SP. All mixtures were designed to be selfconsolidating, except mixture OVC 1. Mixtures LM 14 to 17 were pumped in the
long circuit and underwent the special testing procedure. All other mixtures were
pumped in the short circuit and were investigated by means of the regular testing
procedure, except mixture LM 12, which was tested with the special procedure.

RESULTS
Pumping of ordinary vibrated concrete literature review
Saturated versus unsaturated concrete
When pumping ordinary vibrated concrete, two different situations can occur, as
described by Browne and Bamforth [6]. Either the concrete remains saturated or it
becomes unsaturated due to internal water migration in the concrete caused by the
applied pumping pressure. In the case of saturated concrete, the stress transfer is
of the liquid type and the fundamental laws of rheology can be applied [3,6]. As a
consequence, the pressure loss per unit length does not depend on the local
pressure and the pressure decreases linearly over the length of the pipe [3]. In case
the concrete becomes unsaturated, some zones with a shortage of water are being
created, which results in a frictional type of stress transfer instead of a liquid type.
In this case, Coulombs friction law applies and the frictional stress is dependent
on the local stress [56]. As a result, the pressure loss per unit length does depend
on the local pressure and the total pressure decreases exponentially with the
length of the pipe. The case of unsaturated concretes is the least advantageous
one, as shown by Browne and Bamforth [6]. For a certain setup and certain
concrete properties, they calculated the maximum length of the conveying line
and obtained 250m in case the concrete remains saturated and only 1.1m if a part
becomes unsaturated.
The main pumpability issue for OVC is thus avoiding the concrete to become
unsaturated and avoiding any friction type of stress transfer to occur. As a result,
the amount of coarse aggregates is reduced and some tests are developed to test
the forced bleeding of concrete under pressure [13,6]. Blocking during start-up is
also the consequence of concrete becoming unsaturated. Kaplan described
blocking during start-up as the consequence of the coarse aggregates moving
ahead of the concrete in the water/cement mixture pumped in before to grease
the pipes [3]. Furthermore, the pipe walls are covered by a small layer of cement
paste and the rubber connection seals are filled with cement paste. As a result, the
total cement paste in the first meters of concrete pumped is reduced.

183

Flow in straight sections

In the remaining part of the paper, it will be assumed that the concrete remains
saturated in order to describe flow phenomena. As a result, the flow behaviour is
dominated by rheology and the Buckingham-Reiner equation (for Bingham
liquids) should be applicable [21]. Unfortunately, using this equation leads to a
large overestimation of the pressure loss at a certain discharge, or predicts no flow
at all, while it is observed that the concrete flows [3,7]. Several authors have
described the flow of OVC in pipes as a plug flow, where all shearing is taking
place in a lubrication layer near the wall. The concrete apparently segregates in
the pipes. The composition and properties of the lubrication layer are rather
difficult to measure, as the phenomena causing it are not completely understood
yet. The thickness is estimated to be between 1 and 8mm, although even smaller
layer thicknesses were obtained [3]. Measuring its rheological properties was
done by Kaplan [3] and Chapdelaine [7] by using a concrete tribometer in which
slippage between the inner cylinder and the concrete is allowed.
Kaplan mentions that the flow velocity of concrete can be divided into two parts,
the slippage velocity, which is constant across the pipe section (Kaplan simplifies
the lubrication layer by assuming a slippage velocity), and a shearing velocity,
which can be determined according to the rheological properties of the concrete
[3]. In cases of concretes with a rather high yield stress compared to the shear
stress in the pipes, the shearing velocity equals 0 and the flow velocity is only
determined by the slippage velocity.
The pressure loss discharge curve can be divided into two parts, as can be seen
if Fig. 7 [3]. In part 1, there is no shearing velocity and the characteristics of the
relationship are determined by the properties of the lubrication layer, while in part
2, the relationship is determined by both the properties of the lubrication layer and
the concrete, as shearing occurs. Most OVC do not show any shearing of the
concrete and can be situated in part 1 of the curve of Fig. 7 [3,7].
Flow in bends and reductions
The flow in bends and reductions is a complicated flow phenomenon as the
streamlines are changing direction. In literature, several results for the pressure
losses in bends and reductions are available and generally, the principle of the
equivalent length is used [1-2]. This principle transforms the pressure loss over a
bend or over a reduction into a pressure loss over a straight section with a certain
length. When calculating the total pressure needed in a certain conveying line, the
bends and reductions are replaced (imaginary) with straight sections of a length
equal to the equivalent length.
Kaplan and Chapdelaine observed no additional pressure loss over bends and
reductions in their experiments with OVC [3,7]. As a result, only the distance
over the centre line of the bend must be taken into account to calculate the total

184

equivalent length of the circuit. The ACI-committee 304 on pumping of concrete

[1] concluded that bends cause an additional pressure loss, but no values are
mentioned in the reports. Some more practical guidelines for pumping of concrete
provide some general data, stating for example that a bend of 90 is equivalent to
3m of straight section [2].
Anyhow, the results obtained by the different researchers vary largely, indicating
the complexity of the problem. In case streamlines need to change direction,
additional forces are involved in the problem, such as centrifugal forces in cases
of bends, acceleration in case of reductions, inertia, viscous forces, etc.
Depending on the concrete composition, its rheological properties, the
configuration of the conveying line, the velocity profile, etc. different values for
the equivalent length can be obtained.
Influence of pumping on concrete properties
Pumping can change the concrete properties which can, in the worst case, result in
a concrete with insufficient quality for the designed application. Large studies
have been performed dealing with the influence of pumping on the air content and
air bubbles size of OVC [8-9]. As a general conclusion, it can be stated that the air
content decreases and the air bubbles size increases due to pumping. This is very
disadvantageous if the concrete needs a good air void distribution in order to
withstand freeze-thaw cycles.
Other reports mention that the concrete can show a loss of slump due to pumping
and that temperature increases [22]. Especially in hot environments, temperature
can become a dominating factor, as it increases the hydration process and can
reduce the workability time window of the concrete due to enhanced loss of
workability [4,23].
Pumping of self-consolidating concrete experimental results
Saturated versus unsaturated concrete
The problem of internal water migration in SCC is not as important as for OVC,
as SCC is designed to be a non-segregating material with minimal bleeding.
Furthermore, larger contents of fines and sand are imposed by the mixture
composition, leading to a significantly lower chance for frictional stress transfer
to occur. In general, it can be stated that the unsaturated condition occurs rarely
for SCC, except during start-up. Due to the loss of cement paste at the wall and in
the rubber seals, an accumulation of aggregates forms in front of the concrete. In
this case, the stress transfer becomes frictional and the flow finally stops because
the pump cannot deliver the required pressure. Generally, as is stated by Kaplan
for OVC and as is observed during the insertion of SCC in the long circuit, the

185

blocking occurs behind a section where the streamlines needed to change (bends)
[3].
Flow in straight sections
Recalling Fig. 7, the flow of SCC is situated in part 2 of the curve, as the yield
stress of the concrete is sufficiently low to allow shearing of the concrete itself.
Nevertheless, a lubrication layer must be present, as the Buckingham-Reiner
equation predicts too elevated pressures at a certain discharge compared to the
experimental results. The SCC pumped during the experiments showed shear
thickening behaviour, which is reflected in the pressure loss discharge curve. As
can be seen in Fig. 8 the pressure loss discharge curve is not linear but it shows
an upward curvature. When plotting the pressure loss per unit of length as a
function of the tangential (or differential) viscosity at 10s-1, a very clear
relationship is obtained for each discharge step, as can be seen in Fig. 9. As a
result, the viscosity of SCC is, for a certain geometry of the conveying line and a
certain discharge, the determining factor for the pressure losses. Reducing the
viscosity of the concrete will logically lead to a reduction in pressure losses.
Flow in bends
The flow in bends is certainly not negligible in case of SCC. The pressure loss
over a 90 bend was found to be equivalent to a pipe length between 0.5m and
5.5m, with an average of 2m, which corresponds very well to the practical
statements in [2]. A 180 bend has an equivalent length between 3.5m and 11m,
with an average of 6m. Due to the large scatter in the results, no conclusions for
the equivalent length can be taken. Furthermore, for different rheological
properties of the concrete, different aggregate content and aggregate size and a
different bend configuration, different values of the equivalent length will be
obtained.
In this respect, one must realize that in case of pure water flow the pressure losses
in bends show a large scatter in the measurement results.
Influence of pumping on concrete properties
By means of the special testing procedure, the influence of pumping on the
concrete properties was monitored carefully. As a general trend, a decrease in
plastic viscosity, a decrease in V-funnel flow time and an increase in air content
were observed, as can be seen in Table 2. The yield stress and segregation
resistance sometimes increased and sometimes decreased. The physical causes of
these changes will be discussed in the next section.
Air void analysis on hardened samples of pumped concrete shows a decrease in
air bubbles size compared to samples of similar non-pumped concrete.

186

A temperature increase during pumping was observed during all experiments, but
it did not have a significant influence on the concrete properties or pumping
pressures in these experiments, except for some cases. The temperature increase
per unit of time (dT/dt) is linearly related to the pressure loss per unit of length
(dp/dx), but the rate of increase is different for each concrete [24]. Further
investigation is needed to clarify this phenomenon.

COMPARISON BETWEEN ORDINARY VIBRATED CONCRETE AND

SELF-CONSOLIDATING CONCRETE
Velocity profile
As shown in Fig. 7, Kaplan distinguished the concrete flow in pipes into a pure
lubricated flow and a mixed case of lubrication and shearing. The border between
the two types of flow is a function of the yield stress of the concrete and the
pressure loss in the pipes. Independently of the rheological properties of the
material, the shear stress distribution in a circular pipe evolves linearly from zero
at the centreline to the maximum value at the wall, which is determined by the
equilibrium of forces in equation 2.
w

where

p R2

(2)

w = shear stress at the pipe wall (Pa)

p = pressure loss per unit of length (Pa/m)
R = pipe radius (m)

When the value of the shear stress at the boundary of the lubrication layer is lower
than the yield stress of the concrete, there is no shearing flow and the flow can be
categorized in part number 1 of Fig. 7. In the other case, shearing flow occurs. As
a result, a low yield stress of the concrete or a smaller pipe diameter enhances the
shearing flow at equal discharge. SCC shows generally shearing while OVC does
generally not. This has significant consequences on the flow behaviour of
concrete in pipes.
Flow in bends
As shown in the previous sections, the flow in bends is a very complicated flow
phenomenon, which is very difficult to understand. As far as the results are
reliable, the pressure loss in bends, relative to the pressure loss in straight
sections, caused by SCC is higher compared to OVC, which can indicate more
complex flow behaviour for SCC. Further investigation on the influence of
different parameters on the flow in bends must be executed.

187

Air content and air bubbles size

The observations concerning air content and air bubbles size appear to be opposite
for SCC compared to OVC [24-25], although in some cases, no increase in air
content was observed for SCC [22]. The air content and air bubbles size are
changed due to two effects during pumping: pressure and shearing.
As mentioned earlier, the pressure decreases linearly with the length of the pipe.
At high pressures, the air bubbles are compressed and can possibly dissolve in the
water, while at lower pressures, they grow, and possibly, coalescence of air
bubbles can occur resulting in much larger specimens [8-9]. These large air
bubbles are easy to remove and lower air content is obtained. This effect is even
more pronounced if suction occurs in the pipe when the concrete is pumped
downward. If the pressure in the pipes is not kept above the atmospheric pressure,
the air bubbles become (temporarily) even larger than the air bubbles in the cast
concrete and the change of coalescence increases significantly.
Shearing causes the air bubbles to deform in the flow direction, at least if the
shearing forces are sufficiently high compared to the surface tension [26].
Defining the capillary number (Ca) as the ratio of the shearing forces to the
surface tension, air bubbles do not deform at low Ca (< 1), but they align in the
flow direction in case of high Ca. In case of very high capillary numbers, the
bubbles can even break and create smaller bubbles [27].
Applying these two effects on concrete confirms that OVC shows an increase in
air bubbles size and a decrease in air content, due to the effect of the pressure. As
shearing is only concentrated in the lubrication layer, the bubbles are not
deformed during flow. In case of SCC, as the concrete is sheared at sometimes
rather high shear rates, the bubbles deform and break into smaller units,
confirming the decrease in air bubbles size. The increase in air content can be due
to air entrainment during filling of the concrete reservoir of the pump, during the
aspiration by the pulling cylinder, or it can even be the consequence of the larger
number of small air bubbles which do not escape from the concrete.
The consequences of the larger air content and smaller air bubbles size in case of
SCC have not been investigated at this time.
Rheological properties
As stated in the previous section for SCC, the viscosity of the concrete generally
decreases due to pumping, while the yield stress can evolve in two different ways.
According to the capillary number, the viscosity should decrease as the air content
increases, but the yield stress should increase. As a result, the increase in yield
stress (decrease in slump flow) and increase in segregation resistance (decrease in
sieve-stability result) can be explained by the variation in air content. On the other
hand, for SCC LM 12 and 17, no significant decrease in slump flow is observed
and the segregation resistance appeared to decrease. This effect cannot be

188

explained by the increase in air content and as a result, a second theory must be
applied onto the results.
The second theory is the structural breakdown theory introduced when discussing
transient behaviour. Due to structural breakdown, the equilibrium values of both
yield stress and plastic viscosity decrease when a higher shear rate is applied. By
performing the special testing procedure, as equilibrium is awaited, the concrete is
at a new, lower equilibrium stage at each discharge step. As a result, the
rheological properties and the segregation resistance measured on the sampled
concrete, decrease with increasing maximum discharge applied.
Depending on which effect is dominant, the yield stress increases or slightly
decreases. If the SCC has initially a rather high slump flow, it is more likely that
the segregation resistance will decrease due to structural breakdown, while for
SCC with lower initial slump flows, the yield stress will increase due to the
increase in air content. Independently of which theory is dominating, the viscosity
always decreases.
During the experiments performed, the concrete has passed the pump seven or
eight times. In practice, the concrete only passes the pump once and it is expected
that these effects will be less pronounced. On the other hand, it appeared that
these effects had a more significant influence at high pumping velocities, resulting
in the advice to pump SCC at rather low velocities to reduce these effects [24].
Note that the structural breakdown theory and the increase in air content are not
applicable in case of OVC, as the concrete is not sheared in the pipes.

CONCLUSIONS
Based on experimental results for self-consolidating concrete (SCC) and a
literature review for ordinary vibrated concrete (OVC), a comparison between the
flow behaviour of these concrete types in pipes and the consequences of pumping
on the fresh properties has been investigated.
The velocity profiles of SCC and OVC during pumping are different. Both are
composed of a plug flow in the centre of the pipe and a lubrication layer near the
pipe wall, but in case of SCC, a part of the concrete is also sheared. The
properties of the lubrication layer have a large influence on the pumping pressure,
but in case of SCC, a good relationship between the pressure loss and the
viscosity of the concrete has been obtained at each discharge step applied. As a
result, in order to decrease the pumping pressure, the viscosity of SCC must be
reduced.

189

The flow in bends is a very complex phenomenon in which streamlines need to

change direction. For OVC, different results are available in literature stating that
bends do not have any significance or they can have an equivalent length of up to
3m of straight pipes. For SCC, it was measured that the bends cause an additional
pressure loss, and that for a 90 bend, the equivalent length can vary between
0.5m and 5.5m. A large scatter on the results has been observed as the pressure
loss in bends is influenced by many phenomena such as the flow velocity, the
rheological properties of the concrete, the properties of the coarse aggregates,
inertia, the configuration of the bend, etc.
The air content and air bubbles size appear to evolve in opposite ways for SCC
compared to OVC. In case of OVC, generally, a reduction in air content and an
increase in air bubbles size are observed, while the experiments indicated an
increase in air content and a decrease in air bubbles size for SCC. Due to the high
pressure, air bubbles are compressed but gradually grow when approaching the
exit of the conveying line. Near the end, the bubbles can coalesce and are much
easier to remove. Due to shearing, bubbles deform in the flow, and for high shear
rates, the break up into smaller air bubbles. In case of OVC, only pressure
influences the air bubbles size, resulting in larger air bubbles, while for SCC, as
both pressure and shearing occur, the bubbles are broken up into smaller parts and
a decrease in air bubbles size is observed.
In case of SCC, the viscosity always decreases, but the yield stress can both
increase and decrease. The increase in yield stress is the consequence the
domination of the increase in air content, while when structural breakdown
dominates, a decrease in yield stress can be observed. An increase in yield stress
can result in a too stiff concrete due to pumping, while a decrease in yield stress
can lead to segregation. As both effects appear to have an increasing importance
with increasing pumping velocity, it is advised to pump SCC at rather low
velocities. Note that for OVC, both theories are not applicable, resulting in a very
different influence of pumping on OVC and SCC.

REFERENCES
[1]

Guptill N.R. et al. (ACI-Comm 304), Placing concrete by pumping

methods, American Concrete Institute, Farmington Hills (1998).

[2]

Crepas R.A., Pumping concrete, techniques and applications, 3rd

edition, Crepas and Associates, Inc., Elmhurst (1997).

190

[3]

Kaplan D., Pumping of concretes, Ph.D. Dissertation (in French),

Laboratoire Central des Ponts et Chauses, Paris (2001).

[4]

Beitzel H., Beitzel M., Pump application for self-compacting concrete

under extreme conditions, Proc. of the 3rd North-American Conf. on the
Design and Use of Self-Consolidating Concrete, Chicago (2008).

[5]

Ede A.N., The resistance of concrete pumped through pipelines, Mag.

Conc. Res. 9 (1957), 129-140.

[6]

Browne R.D., Bamforth P.B., Tests to establish concrete pumpability,

ACI-Journal 74 (1977), 193-203.

[7]

Chapdelaine F., Fundamental and practical study on the pumping of

concrete, Ph-D-thesis (in French), Universit Laval, Laval (2007).

[8]

Lessard M., Baalbaki M., Atcin P.-C., Effect of pumping on air

characteristics of conventional concrete, Transportation Research Record,
1532 (1996), 9-14.

[9]

Hover K.C., Phares R.J., Impact of concrete placing method on air

content, air-void system parameters, and freeze-thaw durability,
Transportation Research Record, 1532 (1996), 1-8.

[10]

Ouchi M., Sakue J., Self-compactability of fresh concrete in terms of

dispersion and coagulation of particles of cement subject to pumping,
Proc. of the 3rd North-American Conf. on the Design and Use of SelfConsolidating Concrete, Chicago (2008).

[11]

Tattersall G.H., Banfill P.F.G., The rheology of fresh concrete, Pitman,

London (1983).

[12]

Wallevik J.E., Rheological properties of cement paste: Thixotropic

behaviour and structural breakdown, Cem. Conc. Res. 39 (2009), 14-29.

[13]

de Larrard F., Ferraris C.F., Sedran T., Fresh concrete: A HerschelBulkley material, Mat. Struct. 31 (1998), 494-498.

[14]

Heirman G., Vandewalle L., Van Gemert D., An analytical solution of

the Couette inverse problem for shear thickening SCC in a wide-gap

191

concentric cylinder rheometer, J. non-Newt. Fluid Mech. 150 (2008), 93103.

[15]

Feys D., Verhoeven R., De Schutter G., Fresh self compacting concrete: a
shear thickening material, Cem. Conc. Res. 38 (2008), 920-929.

[16]

Wallevik O.H., Rheology a scientific approach to develop selfcompacting concrete, Proc. of the 3rd Int. RILEM Symp. on SCC,
Reykjavik (2003), 23-31.

[17]

Wallevik J.E., Rheology of particle suspensions, Fresh concrete, mortar

and cement paste with various types of lignosulphonates, Ph-D
dissertation, The Norwegian University of Science and Technology,
Trondheim (2003).

[18]

Assaad J., Khayat K.H., Formwork Pressure of Self-Consolidating

Concrete made with Various Binder Types and Contents, ACI Mat. J.
102:4 (2005), 215-223.

[19]

Barnes H.A., Thixotropy a review, J. non-Newt. Fluid Mech. 70

(1997), 1-33.

[21]

Buckingham E., On plastic flow through capillary tubes. Proc. Am. Soc.
Testing Mat. 21 (1921), 11541161.

[22]

Aldred J., Pumping concrete to 601 metres in a single lift, Proc. of the
Ready-mixed concrete congress Beton2008, Istanbul (2008), 456-462.

[23]

Petit J.-Y., Wirquin E., Vanhove Y., Khayat K., Yield stress and viscosity
equations for mortars and self-consolidating concrete, Cem. Conc. Res.
37 (2007), 655-670.

[24]

Feys D., Interactions between rheological properties and pumping of selfcompacting concrete, Ph-D-thesis, Ghent University, Ghent (2009).

[25]

Thrane L.N., Form filling with self-compacting concrete, Ph-D

dissertation, Danish Technological Institute Technical University of
Denmark, Copenhagen (2007).

[26]

Rust A.C., Manga M., Effect of bubble deformation on the viscosity of

dilute suspensions, J. Non-Newt. Fluid Mech. 104 (2002), 53-63.

192

Table 1 - Concrete compositions. Concretes LM 0, OVC 1, LM 15 and LM

17 are the commercial products of the production plant.

Amount (m)
Composition (kg/m)
Gravel 8/16
Gravel 3/8
Sand 0/5
CEM I 52.5 N
Limestone filler
Water
SP 2
Powder content (kg/m)
W/C-ratio (-)
W/P-ratio (-)
SP/C-ratio (%)

LM 0
1.5

6/03/2007
LM 1
1.25

LM 2
1.25

LM 3
1.5

17/04/2007
LM 4
1.25

LM 5
1.25

LM 6
1.5

5/06/2007
LM 7
1.25

LM 8
1.25

434
263
853
360
239
165
11

434
263
853
360
239
165
11

434
263
853
360
239
165
15.22

459
278
901
300
200
165
12.16

434
263
853
360
239
165
20.95

434
263
853
360
239
165
13.33

434
263
853
360
239
165
12.69

434
263
853
360
239
165
14.44

599
0.458
0.275
4.60

599
0.458
0.275
4.60

599
0.458
0.275
6.37

500
0.550
0.330
6.08

599
0.458
0.275
8.77

599
0.458
0.275
5.58

599
0.458
0.275
5.31

599
0.458
0.275
6.04

690

710

710

720

650

680

640
0.551
0.314
1.98

Slump flow at plant (mm)

Remarks

Amount (m)
Composition (kg/m)
Gravel 8/16
Gravel 3/8
Sand 0/5
CEM I 52.5 N
Limestone filler
Water
SP 2
Powder content (kg/m)
W/C-ratio (-)
W/P-ratio (-)
SP/C-ratio (%)
Slump flow at plant (mm)
Remarks

Plant-Mix

OVC 1
1.55

328
0.538
0.521
1.31

18/09/2007
LM 9
1.25

LM 10
1.25

LM 11
1.5

4/12/2007
LM 12
1.5

LM 13
1.5

410
248
805
400
300
165
18.15

434
263
853
360
239
165
11

410
248
805
400
300
165
unknown

434
263
853
360
239
165
unknown

434
263
853
360
239
165
unknown

434
263
853
360
239
160
21.9

700
0.413
0.236
6.05

599
0.458
0.275
4.60

700
0.413
0.236
unknown

599
0.458
0.275
unknown

599
0.458
0.275
unknown

599
0.444
0.267
9.16

581
0.452
0.324
2.47

599
0.458
0.275
unknown

581
0.452
0.324
unknown

700

650

700

675

700

640

650

700

700

target SF

target SF

Plant-Mix

target SF

Plant-Mix
target SF

Plant-Mix Segregating due to heavy target SF

contains FA
rainfall

193

12/02/2008
LM 14
LM 15
3.25
3.25

6/05/2008
LM 16
LM 17
3.25
3.25
434
263
853
360
239
165
unknown

Table 2 - Results of tests on fresh SCC for all mixes subjected to the special
testing procedure.

Age (hour)
Tests on fresh SCC
Slump flow (mm)
V-Funnel (s)
Sieve Stability (%)
Air content (%)

Age (hour)
Tests on fresh SCC
Slump flow (mm)
V-Funnel (s)
Sieve Stability (%)
Air content (%)

Age (hour)
Tests on fresh SCC
Slump flow (mm)
V-Funnel (s)
Sieve Stability (%)
Air content (%)

Age (hour)
Tests on fresh SCC
Slump flow (mm)
V-Funnel (s)
Sieve Stability (%)
Air content (%)

Age (hour)
Tests on fresh SCC
Slump flow (mm)
V-Funnel (s)
Sieve Stability (%)
Air content (%)

Age (hour)
Tests on fresh SCC
Slump flow (mm)
V-Funnel (s)
Sieve Stability (%)
Air content (%)

Q1
1:03

Q2
1:12

SCC LM 12
Q3
1:21

Q4
1:33

Q5
1:42

735
3.16
12.3
2.4

725
3.07
8.1
2.7

755
2.34
12.3
2.8

735
2.23
12.9
2.8

645
2.09
9.4
3.2

Q1
2:30

SCC LM 14
Q2
Q3
2:45
3:00

Q4
3:30

818
5.23
10.9
1.6

758
6.1
11.0
1.8

658
5.82
7.5
1.5

Q1-1
1:30

745
3.65
14.2
1.6

SCC LM 15 - test 1
Q1-2
Q1-3
1:45
2:00
660
3.77
6.6
3.2

Q1-4
2:10

645
5.43
4.2
2.1

625
4.18
7.0
2.4

570
3.42
4.0
4.2

Q2-1
2:50

SCC LM 15 - test 2
Q2-2
Q2-3
Q2-4
3:00
3:10
3:20

Q2-5
3:30

525
3.54
3.4
3.7

543
3.06
4.5
3.9

505
3.29
1.9
4.6

445
3.74
0.3
6.2

Q1
2:35

Q2
2:45

SCC LM 16
Q3
Q 3 bis
3:00
3:10

670
5.24
8.7
1.1

675
4.02
12.7
1

655
4.78
6.9
1.4

Q1
1:20

Q2
1:30

785
3.39
10.5
1.4

780
3.08
1.9

498
3.46
0.8
5.0

Q 3 ter
3:20

Q4
3:30

585
3.72
6.8
1.3

620
3.76
7.8
2.2

535
3.89
5.7
3.9

SCC LM 17
Q3
1:40

Q4
1:50

Q5
2:00

750
2.66
11.7
3.1

765
2.35
15.6
3.9

750
2.22
18.5
4.9

194

Shear stress
Shear rate
shear rate
shear stress
Equilibrium shear stress

Time

Fig. 1 - Stress response (black line) of thixotropic material to sudden increase

and decrease in shear rate (grey line). The dashed line represents the
equilibrium shear stress at the applied shear rate. Figure from Barnes [20]

Fig. 2 - Truck-mounted concrete piston pump

195

Fig. 3 - Short circuit with a length of 25m, the measurement section is located
in the straight horizontal part of the circuit, on the left side of the picture

Fig. 4 - Long circuit (105m)

196

Strain gauges
Temperature sensor

Pre

ssu

re

sen
so

Fig. 5 - Installation of flush-mounted pressure sensor and 3 strain gauges

attached to the outer pipe wall, acting as a back-up.
Discharge number
6

Maintaining constant discharge

to reach equilibrium

Stepwise decreasing discharge

Sampling

Time

Fig. 6 - Special testing procedure, consisting of three phases each time:

keeping the discharge constant during a relatively long period; taking a
concrete sample, combined with a discharge calibration; quickly decreasing
the discharge stepwise, similar as a regular cycle. After the three phases, the
maximal discharge is increased by one step. Note that at the lowest maximal
discharge (number 1), there is no stepwise decreasing curve

197

Fig. 7 - Bi-linear relationship between pressure (pression) and discharge

(dbit), proposed by Kaplan. Figure from D. Kaplan [3]
Pressure loss (kPa/m)
45
SCC
OVC

40
35
30
25
20
15
10
5
0
0

10

15

20

25

Discharge (l/s)

Fig. 8 - Pressure loss discharge curve for SCC and OVC, reflecting the
rheological properties of the concretes. The OVC is a perfectly Bingham
material, while the SCC shows shear thickening behaviour

198

Pressure loss (kPa/m)

100
discharge 5
discharge 3
discharge 1

90
80

y = 1.6093x + 6.9824
2

R = 0.9271

70
60
y = 0.9427x + 1.2378
50

R2 = 0.8899

40
30
y = 0.3234x + 1.4083

20

R2 = 0.6898

10
0
0

10

20

30

40

50

60

70

Viscosity at 10/s (Pa s)

Fig. 9 - The relationship between pressure loss and the differential viscosity
at a shear rate of 10/s, shows, for each discharge, good to very good
correlations

199

200

SELF COMPACTING CONCRETE: PROPERTIES, DEVELOPMENT

AND CODE RECOMMENDATIONS
Joost Walraven

Synopsis: Self compacting concrete (SCC) is a material with a short history and a
large impact. After its invention considerable research has been carried out and
practical applications have been realised. An important question has always been
whether SCC should be considered as a regular concrete or if its needs a separate
set of recommendations. In this paper it is argued why the properties in the
hardened state can be taken from regular codes for structural concrete. Contrary to
this the behaviour in the fresh state cannot be taken from existing
recommendations, since other measuring methods than used for conventional
concrete are required. This refers especially to the concrete consistence. In this
respect the latest rules according to the new Eurocode are discussed. For practical
applications it is important to know the relation between the type of concrete
structure or member considered and the SCC-properties required. As an example
the properties are classified by the slump flow diameter and the funnel time,
resulting in 9 subclasses for the consistence, linked to typical structural
prototypes. In future defined performance design will be an important approach
both for structures and structural materials. Within this scope some special SCCmixtures are treated. Especially self compacting fibre concrete turns out to be an
attractive option. Suggestions for further research and development are given.

Keywords:

Self

Compacting

Concrete,

201

properties,

consistence,

codes

Joost Walraven is Professor of Concrete and Concrete Structures at the Delft

University of Technology in The Netherlands. He carried out research on various
topics, with special attention to the development of new types of concrete and
their application. He has been president of the international concrete federation fib
and chairman of the project team for Eurocode 2 Concrete Structures. At this
moment he is chairman of the fib-committees TG-8.6 Ultra High Performance
Fibre Concrete and SAG5 New Model Code for Concrete Structures

INTRODUCTION
Self compacting (or self consolidating) concrete (SCC) was first developed in
Japan, in the early nineties of the previous century, under the stimulating
leadership of Prof. Okamura. The main idea behind self compacting concrete was,
that such a concrete is robust and relatively insensitive to bad workmanship. In
Western Europe the idea was picked up at the end of the last century. The main
drive to develop self compacting concrete was the option to improve the labour
conditions at the building site and in the factory (noise, dust, vibrations). During
recent years self-compacting concrete developed to an important research item. A
large number of research projects were carried out, followed by recommendations
for potential users. Especially for the precast concrete industry self compacting
concrete was a revolutionary step forward. Contrary to that, casting of SCC at the
construction site was regarded with some reservation. The variable conditions at
the construction site, the more complicated control of the mixture composition
and disagreement with regard to the question how the properties should be
measured at the site were retarding factors. In spite of a number of successful
examples, some problems due to unsuitable use of SCC in initial applications
generated further scepticism. Since then substantial effort has been done in
developing more stable and robust mixtures. Moreover a better insight has been
gained with regard to SCC composition and consistence criteria in relation to
types of application. For a better understanding of relations between SCCproperties and SCC-composition, at first a short treatment is given of the question
why basically a concrete can be self-compacting, and what influence can logically
be expected on the properties in the fresh and the hardened state.
PROPERTIES OF SELF COMPACTING CONCRETE
The Japanese way of composing the optimum mixture composition of SCC
consists of a number of steps. At first, in a simple test, the optimum ratio of water
to powder is determined. Then a number of general criteria have to be met, the
most important of which are that the coarse aggregate volume should be 50% of

202

the solid volume of the concrete without air, and that the fine aggregate volume
should be 40% of the mortar volume, where particles finer than 0.09 mm are not
considered as aggregate, but as powder. If the composition of the mixture,
obtained in this way, is mathematically analyzed, it is found that this procedure
leads to a concrete composition with a little bit of excess paste. That means that
there is slightly more paste in the mixture than necessary to fill all the holes
between the particles: this implies that around any particle a very thin
lubricating layer exists, by virtue of which the friction between the particles in
the fluid mixture is greatly reduced in comparison to conventional mixtures, Fig.
1. The optimum thickness of those layers lies between narrow limits. If the
thickness is too small, there is too much friction to achieve self compactibility. If
the thickness is too large, the coarse aggregate particles sink down and
segregation occurs. The rheologic properties of the excess paste layers are
determined by the choice of the superplasticizer. Furthermore, in the fresh state
around the cement and powder particles thin layers of water are formed [1]. In
this way a three phase system (coarse particles, fine particles and powder) with
intermediate layers of paste and water is obtained which minimize the internal
friction in the fresh state.
Midorikawa [2] carried out tests in order to find the optimum thickness of the
excess paste layer. Fig. 2 shows the optimum thickness of the layer for a varying
ratio Vw/Vp (volume of water to volume of powder) for different grading curves.
It is seen, that the thickness of the paste layer, for which the concrete is still self
compacting, increases with decreasing volume of water. Below V w/Vp = 0.8 the
appropriate thickness increases overproportionally. For practical application,
however, this area is not relevant. The optimum ratio in this case is in the range
Vw/Vp = 0.8 0.9. The mean thickness of the excess paste layer is then in the
order of magnitude of 0.05mm.
An important question is, to which extent the lubricant layer influences the
properties of the concrete in the hardened stage. It seems to be obvious, that for
instance the modulus of elasticity of SCC is smaller than that of a conventional
concrete of the same strength, as a result of the effect of the relatively soft
lubricant layers. An evaluation by Holschemacher [3] showed, that the E-modulus
of an SCC is indeed somewhat smaller, Fig. 3. It should, however, not be
forgotten, that also the E-moduli of normal concretes are subjected to scatter,
most of all in dependence of the stiffness of the aggregate used. For practical
applications it can therefore be assumed, that the E-modulus of self-compacting
concrete is not outside the region of scatter of conventional concretes. Of course it
makes sense to carry out suitability tests in the case of special applications, such
as for structures in high speed railway lines. In that case, however, also the creepand shrinkage properties should carefully be investigated.

203

Another aspect important for the behavior in the hardened state is the concrete
tensile strength. When the axial tensile strength of a SCC would substantially
differ from that of the tensile strength of conventional concrete, this should have
large implications for design, since the tensile strength is a governing aspect in the
design for shear, punching, anchorage, crack width control and the minimum
reinforcement. It is obvious to expect that the tensile strength of SCC is higher
than for a conventional concrete, because of the more homogeneous interface
between the aggregate particles and the cement past (no direct contact between
the aggregate particles). An evaluation of test results [3] confirms this, see Fig. 4.
However, also here the results are in the range of scatter of conventional
concretes, so that no complicating exception for SCC has to be made.
Another important aspect is the formwork pressure of self compacting concrete.
Many measurements have been carried out, but the results were often conflicting.
Often the role of the rising speed of the concrete in the formwork was
disregarded. Fig. 5 shows the results of a number of Swedish [4] and Dutch [5]
tests, collected in one diagram. It is visible that the rising speed of the concrete in
the formwork influences the formwork pressure. For the concretes tested, from a
rising speed of about 2m/hour the distribution of the pressure corresponds
approximately to the hydrostatic pressure. This, however, does not implicate, that
for lower rising speeds a reduction of the formwork pressure is a reliable
assumption. According to its rheologic behavior SCC is a Bingham fluid. Such a
fluid is characterized by two parameters: the yield value and the plastic viscosity.
The yield value is a measure for the force, necessary to get the concrete moving.
The plastic viscosity is a measure for the flow rate (toughness) of the mixture.
When the yield value is high and the plastic viscosity is low, it may happen that
the formwork pressure is initially very low, but suddenly increases due to a shock
against the formwork. It is therefore advisable to work always with the
hydrostatic formwork pressure.
PROPERTIES OF SELF COMPACTING CONCRETE AND CODE
RECOMMENDATIONS
Behaviour of SCC in the hardened state
The most recent large scale international building code is the Eurocode. The
Eurocode consist of many parts. The basic code for the design of concrete
structures is EN 1992-1-1 Design of concrete structures: general rules and rules
for buildings. Complementary, the properties and specifications of the material

204

concrete are given in EN206-1 Specification, performance, production and

conformity .
EN 1992-1-1, which treats the properties of hardened concrete, does not contain
any special rule for self compacting concrete. The reason is that in EN 1992-1-1
all properties in the hardened state are referred to the concrete compressive
strength, obtained on cubes of cylinders. Fig. 3 and 4 showed as an example the
concrete tensile strength and the E-modulus as a function of the concrete
compressive strength. It is obvious that in all those cases the results for SCC are
within the range of scatter of other concretes of the same strength. Therefore no
exception is made for SCC in relation to other concretes, although in tendency the
results for SCC may be lower (E-modulus, Fig. 4) or higher (tensile strength, Fig.
3). This might have consequences for cases where the behaviour of the structure is
sensitive to deviations of the characteristics. In EN 1992-1-1, Chapter 3.1.3
Elastic deformation the following statement is made:
(1) The elastic deformations of concrete largely depend on its composition
(especially the aggregates). The values given in this Standard should be
regarded as indicative for general applications. However, they should be
specifically assessed if the structure is likely to be sensitive to deviations
from these general values. If e.g. SCC is used for a bridge in a high speed
railway line, where the functionality depends essentially on the time
dependant deformations, it is advisable to carry out tests on E-modulus,
creep and shrinkage. However, this statement holds true for any concrete,
and not only for SCC.
Behaviour of SCC in the fresh state
In EN206-1 the properties of concrete in the fresh state are dealt with. The
document consists of a general part for conventional types of concrete, but
contains an Annex for self compacting concrete. In the general part, for
conventional concrete, the classification of concrete consistence is given. In
general for classification purposes four test methods can be used, regarding the
determination of the slump, the compaction, the flow diameter and the vebe time.
Fig. 6 shows the principle of the first three methods.
For details about the dimensions of the equipment and the procedures to be
followed, reference is made to Eurocode document EN12350.
The consistence of a concrete can generally be defined by combining the tests
described before. This is shown in Table 1
It is clear that SCC can not be classified in this way. The categories Flowable
and Very flowable are not appropriate for SCC. Most SCC mixtures act as a so-

205

called Bingham Fluid, characterized by two parameters: the yield value and the
plastic viscosity.
Therefore EN206-1 offers an appendix for SCC, giving additional methods to
classify SCC. These are:
EN 12350-8: Slump-flow test for self-compacting concrete (flow diameter test)
EN 12350-9: V-funnel test for self-compacting concrete
EN 12350-10: L-box test for self compacting concrete
EN 12350-11: Sieve segregation test for self compacting concrete
EN 12350-12, J-ring test for self-compacting concrete
Where the flowability of self-compacting concrete is to be determined, it shall
be measured as a slump-flow test in accordance with EN 12350-8
Where the viscosity of self-compacting concrete is to be determined, it shall be
measured by means of:
- t500 test conforming to EN 12350-8
- V-funnels test conforming to EN 12350-9
Where the passing ability of self-compacting concrete is to be determined, it
shall be measured either by means of:
- L-box test conforming to EN 12350-10
- J-Ring test conforming to EN 12350-12
Where the resistance to segregation of self-compacting concrete is to be
determined, it shall be measured by means of:
- Sieve segregation test according to EN 12350-11
For the various tests classes have been defined, given in Tables 2-7.
Some further requirements have been given with regard to the time and procedure
of testing, like:
- where the consistence of self-compacting concrete is to be determined, it shall
be tested at the time of use of the concrete or in the case of ready mix concrete,
at the time of delivery.
- If concrete is delivered in a truck mixer or agitating equipment, the consistence
may be measured using a spot sample obtained from the initial discharge.

206

- The consistence may be specified either by reference to a consistence class as

given previously, or by a target value or limit value. For target values, the
related tolerances are given in Table 8.
Further recommendations are given for the control of production procedures and
of consistence for SCC. A survey is given in Table 9.
Practical use of consistence classes for getting optimum performance in practice
A classification of the consistence of SCC in classes is useful in order to define
certain behavioural characteristics of SCC in the fresh state, linked to types of
applications. In this way optimum use of experience can be made. As an example
the practice of using SCC in the Netherlands is shown. In The Netherlands the
concrete consistence is defined according to the Japanese Method, using the
funnel test and the slump diameter test, Fig. 7. As argued already before, a
Bingham fluid like SCC is characterized in terms of rheology, by two
parameters, being the yield value and the plastic viscosity. As a practical
alternative the slump flow diameter test and the V-funnel tests can be used. Those
tools are very suitable to be used at the building site, because they can be handled
very easily. Both for the slump flow diameter and the V-funnel time three
intervals are defined, as shown in Fig. 8. In this way 9 classes are distinguished.
For any application a most appropriate class exists, see fig. 9. If for instance, selfcompacting concrete is specified for a lightly reinforced wide floor, for practical
reasons a short funnel time is required. If, on the contrary, a column with
congested reinforcement has to be cast, a large slump flow in combination with a
low funnel time (high viscosity) is most appropriate. In fig 9 also other areas are
defined.
APPLICATIONS IN THE PRECAST CONCRETE INDUSTRY
Previously, it was pointed out that self compacting concrete mixtures are sensitive
to variations in composition and environmental influences. For the precast
concrete industry this is not a considerable difficulty, since the processes at the
plant can be very well controlled. The advantages for using SCC in precast
concrete plants are very considerable like,
-

the substantial reduction of the noise level

the absence of vibration

the reduction of dust (quartzite!) in the air due to vibration

the energy saving

the omission of the expensive mechanical vibrators

207

the reduction of wear to the formwork

the use of less robust formwork with simpler connections

the reduction of absence for illness

the possibility to produce elements with high architectural quality

For successful production of SCC it is essential that the basic constituents, like
sand, gravel, fillers and the third generation of superplasticizers, have a constant
quality. This is not always the case. Moreover, not all cement producers supply a
constant quality. So, there should be good agreements between the concrete
producers and the suppliers of the constituents on the quality control. The step
from a traditional concrete production to the production of SCC is not a big one.
Installations with an age of say 5-10 years are generally suitable. Further to the
traditional equipment a high intensity mixing machine and an installation to dose
the fillers are needed.
As a result of the introduction of SCC the formwork is hardly loaded anymore: it
has only a retaining function. So, the wall can be made of other materials than
timber, like polystyrene. Also steel formwork with magnetic couplers is possible.
The time for demoulding and re-installing the formwork has been reduced by
50%. There is no need for the installation of vibration isolators anymore. Rubber
joint sealings can be omitted, since by virtue of SCC no leakage through the joints
occurs anymore.
Fig. 10 shows an example of an architectural balcony element of SCC. The
element does not only show a beautiful shape with very sharp profiles, it has also
a homogeneous white colour. Fig. 11 shows the assembly of a precast prestressed
concrete girder of SCC for the new metro station at the Amsterdam Arena, the
stadium of soccer club Ajax. The girder has a length of 22,5m. The concrete
strength class is C55 (characteristic cylinder compressive strength of 55 MPa
(7850 psi)). The metro station has a length of 350m with 4 tracks. This means that
60 girders had to be produced with a total length of 1.4 km. If the girder would
have been compacted in the traditional way, heavy vibrating machines would
have been necessary. Due to that, the formwork would have had to be replaced
after a relatively small number of casts. By virtue of the use of SCC the life of the
formwork was very long. Another important reason to choose for SCC was the
improvement of the labour conditions in the factory.
Fig. 12 shows a set of foundation piles. The production of such type of piles in the
firm was 70 000 piles a year. For an average length of 15m a total production
length of 1000 km a year is obtained. Until recently the piles were produced with
the so-called shock procedure. This means that the formwork was forced to
repeatedly fall down from a height of 50mm (2 inches), which created a shock

208

effect. By virtue of the change to SCC the necessary casting time was reduced
from 7.5 minutes to 1.5 minutes. Since mechanical compaction was not necessary
anymore 12 further minutes were gained. Taking also into account the advantages
with regard to the reduction of noise and dust, energy consumption and wear, it is
clear that SCC yields considerable advantages. Fig. 13 shows a number of
concrete arches. Every arch has a length of 65 meter and is composed of 5 pieces
of 13m. The cross section has a box-shape, with a foam core. Producing such an
element with conventional concrete does not make sense, since the foam core
would move due to vibration. A production in parts could be an alternative but is
by far too time consuming, and therefore too costly. With SCC perfect elements
could be made. Meanwhile many precast concrete firms have changed their
production to SCC, some even for 100%.
APPLICATIONS OF SCC IN SITU
The introduction of SCC for in-situ applications was slower than in the precast
concrete industry. There are a number of reasons for this:
- In case of failure the consequences for an in-situ application are much more
severe than in the precast concrete industry. In the latter case the unsuitable
elements can be simply rejected, whereas in the first case demolition might
be the ultimate consequence.
- There was often no agreement on the way in which the properties at the
building site have to be controlled.
- Self compacting properties can be more easily reached with higher strength
than with lower concrete strength. In a number of practical applications the
concrete strength was therefore higher than actually necessary, which has
cost consequences. For many applications a concrete strength class C25 is
sufficient. However, especially for the lower strengths classes it is more
difficult to obtain robust and reliable self compacting concretes.
Meanwhile, however, a lot of barriers have, or are being, removed. There is now a
better insight into the required properties of SCC for particular applications, like
previously shown in Fig. 9. Furthermore qualifying test methods have been
evaluated. Finally a new generation of superplasticizers has been introduced.
Nevertheless, a number of convincing examples exist, which proof that SCC, if
applied in an appropriate way, can give excellent results. The first application of
SCC in The Netherlands according to modern principles was such an example,
Fig. 14. In 1998 a large faade was made for the National Theatre in The Hague,
which, for architectural reasons, was provided with fine triangular ribs with a side
length of 8mm. In this case an SCC with relatively high flowability was used

209

(flow diameter 730 mm) and a low viscosity (funnel time 8-9 seconds). Fig. 15
shows the sports stadium at Tor Vergata University (Roma II), under
construction. The architect is Santiago Calatrava, who required self compacting
concrete for all the concrete used. The contractor succeeded in realizing excellent
surface properties.
There are many practical problems where SCC gives a suitable solution. An
example is the retrofitting of the Ketelbridge, a glued segmental bridge in The
Netherlands. At the time of retrofitting in the year 2002 the bridge was 45 years
old. During the years the bridge deck was renovated several times, but the old
deck was often not totally removed. So, finally the bridge deck was 180 mm thick
in stead of 50 mm. Since as well the traffic load had increased, the joints between
the segments opened. Therefore it was decided to increase the load bearing
capacity by external prestressing. A difficulty was the provision of the deviators
inside the box girder. Because the lower flange of the girder had not been
designed for the transport of heavy materials, casting concrete inside the girder
was no realistic option, even regardless of the technological difficulties involved.
As a solution therefore SCC was used. The formwork with the reinforcement was
built up in the interior of the girder (Fig. 16), and the SCC was cast from the
outside through a little window in the upper flange. The concrete strength class
was C35. By a suitable use of the rheological properties an excellent result was
obtained.
Another interesting case for which SCC gave a solution was the provision of the
end walls in elements for a submerged tunnel, Fig. 17. Those end walls had a
temporary character and served only for enabling floating transport and
submerging. After the elements had been coupled under water, the walls were
demolished. In order to facilitate easy demolishing, SCC in a strength class C20
was used. For casting the concrete between the tunnel walls through small
windows in the formwork SCC appeared to the most appropriate solution.
SPECIAL SELF COMPACTING CONCRETES
A remarkable development occurred with regard to the workability of fibre
reinforced concrete. For a very long period it was noted that the addition of fibres
to concrete decreased the workability. However, in his PhD-thesis Grnewald [7]
showed that this is not necessary at all. He proved that self compacting fibre
concretes are very well possible, even up to fibre contents of 140 kg/m3, if the
right combination of fibres and mixture composition is chosen. Fig. 18 shows the
maximum possible fibre content for which mixtures are still self-compacting
(defined as having a flow circle with a diameter of at least 600 mm, a round shape
and a homogeneous fibre distribution). At the vertical axis the fibre content in

210

kg/m3 is given. At the horizontal axes the fibre type (aspect ratio/length) and the
mixture type (with the sand/gravel vol. ratio) are given.
Fig. 19 gives an impression of the excellent flowing properties during casting of a
concrete with 125 kg/m3 fibres. Fig. 19 shows the measurement of the flowability
of an ultra high performance fibre reinforced concrete in a U-shaped formwork.
The concrete had a mean cube strength of about 180 MPa (25000 psi). It
contained 235 kg/m3 steel fibers 20/0.3mm. It was used in a factory to produce
prestressed beams for a bridge. In another paper the topic of optimizing the selfcompacting properties of fibre concrete was treated more in detail [8].
The self-compacting properties of fibre reinforced self compacting concrete can
be of great use for practical applications. An example is the repair of the Huisne
Bridge in France [13], Fig. 21. This bridge should be strengthened in order to be
able to carry a larger traffic load. A solution was found by widening the width of
the webs of the prestressed I-beams with high performance fibre reinforced
concrete. As a result of the large E-modulus of the HPFRC (about 60.000
N/mm2), the normal force, exerted by additional external prestressing, is
predominantly sustained by the high performance concrete strips, in stead of
loading the old concrete, the capacity of which was nearly fully utilized according
to the original design. Casting concrete against the webs of I-beams under the
existing bridge decks is hardly possible with conventional concrete. With the high
performance fibre concrete used, this was no problem, by virtue of the selfcompacting properties. A little window in the deck was sufficient for remote
casting of the concrete under the deck.
Recently it was shown that not only concrete with medium or high strength
concrete can be self-compacting. At TU Delft a research program was carried out
with the aim to fill large underground corridors, used for the storage of nuclear
waste, with self compacting concrete by remote casting. For this application no
strength requirements applied for the concrete. Therefore an economic solution
was strived at: developing a low strength self compacting concrete. The grout,
which was composed with 10% cement, 30% limestone powder, 0,03%
superplasticizer en 59,95% sand, with a water cement ratio of 1.30-1.35, showed a
28 days strength of 19-20 N/mm2. Fig. 22 shows a laboratory test where the
mixture flows through a pipe investigating the suitability for remote casting [14].
Another interesting option is self-compacting lightweight concrete. High
performance lightweight concrete could allow significant savings in reinforcingand prestressing steel and foundations. Self compacting properties would even
increase the attractiveness of such a material. With regard to the production
technology, there is a major difficulty. The lightweight aggregate particles are
porous and therefore influence the mixture composition by sucking water from
the mixture in the fresh state. As self compacting concrete is sensitive to the right

211

composition this causes a major difficulty. A solution was developed by Mller

[9]. He developed a technology which consists of enveloping the sucking
aggregates with a thin cement bonded surface coating. The composition of the
cement pastes used for the enveloping is optimized in such a way as to make it
economically possible to apply a thin layer to the agglomerate in the fresh state.
After that, the storage of the freshly enveloped aggregates preventing them from
sticking together is assured and finally after a rapid setting a high density and
strength of the formed envelope is guaranteed. Fig. 23 shows an enveloped
lightweight aggregate particle of the type Liapor F5, where the thickness of the
cement bonded layer, which in the section is distinctly recognizable by its lighter
colour, is equal on the average to approximately 0.25-0.35 mm.
Compared to non-enveloped aggregates, water absorption by dry materials is
drastically reduced if the dry materials are stored for 30 minutes in water and
under a pressure of 50 bars and if a low or high performance and thus denser
lightweight aggregate is used. The result is that concrete mixtures with enveloped
lightweight aggregate behave with respect to the characteristics and processing of
fresh concrete exactly like mixtures with dense, normal weight additive material.
Further information on this topic is found in [10].
NEEDS FOR FURTHER DEVELOPMENT
The sensitivity of SCC mixtures for minor variations in the mixture composition
should be decreased. This can be done by adding appropriate types and amount of
fillers. Another, not yet fully explored possibility is the use of viscosity agents.
Experiments on mixtures with viscosity agents show that the sensitivity of for
instance variations of the water content on the viscosity can be strongly reduced
by applying an appropriate viscosity agent, see e.g. Grnewald [11]. Especially
the potential of viscosity agents for improving the stability of mixtures with low
and medium strength, suitable for large scale in-situ applications, deserves further
attention.
Also the further development of suitable superplasticizers for SCC is worthwhile,
possibly in combination with viscosity agents. Takada [12] showed in his PhDthesis that there is a strong influence of the type of superplasticizer on the
necessary mixing time and mixing intensity. In this area there is still a need for
further research.
A very important aspect to be regarded is the durability of SCC. There is a
tendency that in the near future structures should not only be designed for safety
(ULS) and serviceability (SLS), but as well and with the same importance - for
service life. This means that increased demands will be raised on the resistance of

212

SCC with regard to chloride ingress, carbonation and frost-thaw cycles. It was
shown by many research projects that SCC is approximately equivalent to
conventional concretes with regard to the majority of its mechanical properties in
the hardened state. However, with regard to the microstructure of hardened SCC
and its significance to durability there are still quite a number of open questions.
In this respect the interface between matrix and aggregates plays an important
role. Furthermore the role of (combinations of) additives (superplasticizers, air
entraining agents, viscosity agents) on the microstructure, including porosity and
permeability should get due attention. In this respect special attention has to be
devoted to the average and low strength concretes used in in-situ structures, if
exposed to more severe environmental conditions.
CONCLUSIONS
1. In spite of its short history, self compacting (or consolidating) concrete has
confirmed itself as a revolutionary step forward in concrete technology.
2. For the application of SCC in situ, it is necessary that SCCs are designed
(tailor-made) for any particular case. General rules are available on the basis of
experience.
3. It can be shown by cost analysis, that SCC in precast concrete plants can be
more economically produced than conventional concretes, in spite of the slightly
higher material price. Cost comparisons should always be made on the basis of
integral costs.
4. There is a considerable future for tailor-made self compacting fiber reinforced
concretes
5. The most important task for research is to develop SCCs with decreased
sensitivity to variations in constituents and environmental influences. This holds
particularly true for in-situ concretes, with medium and low strengths.
6. Further research into the potential role of viscosity agents and their interaction
with superplasticizers is worthwhile
7. Since in the near future service life design (SLD) of concrete structures will be
as important as design for safety and serviceability, increased attention should be
given to the role of the microstructure of the various types of available SCCs and
its role for durability.
8. With regard to its mechanical properties it is not necessary to make a
distinction between SCC and conventional concretes. Therefore also in modern
codes like the Eurocode for Concrete Structures EN 1992-1-1, SCC is treated like
any other regular concrete.

213

REFERENCES
1.

Midorikawa, T., Maruyama, K., Shimomura, T. and Momonoi, K.,

Application of the water layer model to mortar and concrete with various
powders, Proceedings of the Japan Society of Civil Engineers, No. 578/V37, pp. 89-98, 1997 (in Japanese).

2.

Midorikawa, T., Pelova, G.I., Walraven, J.C., Application of the water layer
to self compacting concrete with different size distribution of fine aggregate,
Proceedings of the Second International Symposium on Self-Compacting
Concrete, Tokyo, Japan, 23-25 October 2001, pp. 237-246.

3.

Holschemacher, K., Design relevant properties of self compacting concrete,

Symposium Self Compacting Concrete, Leipzig, Nov. 2001, Proceedings,
pp. 237-246 (in German).

4.

Billberg, P., Form pressure generated by self-compacting concrete, 3rd

International Rilem Symposium Self-compacting concrete, 17-20 August
2003, Reykjavic, Iceland, Proceedings, pp 271-280.

5.

Den Uijl, J.A., Properties of self-compacting concrete, Cement 6, 2002, pp.

88-94 (in Dutch).

6.

www.concretecentre.com.

7.

Grnewald, S., Performance based design of self-compacting reinforced

concrete Dissertation, TU Delft, 4. June 2004.

8.

Grnewald, S., Walraven, J.C., Optimization of the mixture composition of

self-compacting fiber reinforced concrete, Conference SCC 2005, Chicago,
USA, Oct. 30 November 2, 2005.

9.

Mller, H.S., Guse, U.,Concrete Technology Development: important

research results and outlook in the new millennium, Concrete Plant +
Precast Technology, 2000, Nr. 1, pp. 32-45

10. Haist, M., Mechtcherine, V., Beitzel, H., Mller, H.S., Retrofitting of
building structures using pumpable self-compacting lightweight concrete,
Proceedings of the 3rd International RILEM Symposium on SelfCompacting Concrete, pp. 776-795.
11. Grnewald, S., Walraven, J.C., The effect of viscosity agents on the
characteristics of self-compacting concrete, Conference SCC 2005, Chicago,
USA, Oct. 30th Nov. 2nd, 2005.

214

12. Takada, K., Influence of admixtures and mixing efficiency on the properties
of self compacting concrete, PhD-Thesis, TU Delft, May 11th, 2004.
13. Thibault, Th. Strengthening of Huisne Bridge using Ultra-High-Performance
Fibre-Reinforced Concrete, Proceedings fib Congress 2008, Amsterdam, pp.
331-334.
14. Bakker, P., Grout backfill process of nuclear waste disposal gallery, Masters
Thesis, TU Delft, Faculty CiTG, Group of Concrete Structures, June 9th, 2009

215

Table 1. Qualification of consistence of fresh concrete

Consistence
Dry
Earth dry
Semi plastic
Plastic
Very plastic
Flowable
Very
flowable

Compaction
Class
C0
1.46
C1
1.45-1.26
C2
1.25-1.11
C3
1.10-1.04

Slump
Class
mm

Flow diameter
Class
mm

S1
S2
S3
S4
S5

F1
F2
F3
F4
F5
F6

10-40
50-90
100-150
160-210
220

Table 2. Slump flow diameter classes

Class
SF1
SF2
SF3

Flow diameter in mm
(Limit values for individual batches)
550 650
660 750
760 - 850
Table 3. Viscosity classes t500

Class
VS1
VS2
VS3

T500 (seconds)
2
36
>6
Table 4. Viscosity classes V-funnel

Class
VF1
VF2

V-funnel time (seconds)

<9
9 - 25

216

340
350-410
420-480
490-550
560-620
630

Table 5. Passing ability classes L-box

Class
PL1
PL2

L-Box ratio (-)

0,80 with 2 rebars
0,80 with 3 rebars
Table 6. Passing ability J-Ring

Class
PJ1
PJ2

J-ring step (mm)

10 with 12 rebars
10 with 16 rebars

Table 7. Sieve segregation resistance classes

Class
SR1
SR2

Segregated portion (%)

20
15

Table 8. Tolerances for target values of consistence

Performance criterion
Slump flow tolerance in mm
t500 tolerance in seconds
V-funnel time : Target value in seconds
: Tolerance in seconds

217

Tolerance for target values of

consistence
50 mm
1 sec.
< 9 sec
9 sec
3 sec
5 sec

Table 9. Control of production procedures and of consistence for SCC

Type of test

Inspection/test

Purpose

Minimum frequency

Consistence

Visual inspection

For comparison
with normal appearance

Each batch or load

Slump flow test

To assess the
achievement of the
specified values of
consistence and to check
e.g. possible changes of
water content

V-funnel test
L-box test
Sieve segregation
test
J-ring test

To assess the
achievement of the
specified values of
consistence

- If consistence is specified
as EN 206-1:2000, Table
13 for compressive
strength
- In case of self compacting
concrete at least once day
- when testing air content
- In case of doubt
following visual
inspections
- When performing initial
test
- Before using a new
concrete composition
- In case of a change in the
constituents. In case of
doubt following visual
inspections or slump-flow
test.

218

Fig. 1 - Excess paste layers around

Vwater/Vpowder

Thickness of excess paste [m]

Fig. 2 - Relation between thickness of excess

aggregate particles paste layer and water to powder ratio for various particle
grading curves [2]

219

Fig. 3. Modulus of elasticity of self-compacting concrete [3].

Concrete splitting tensile
strength (N/mm2)

Concrete compressive strength (N/mm2)

Fig. 4. Concrete splitting tensile strength as a function of concrete

compressive strength for SCC in comparison with conventional concrete [3].

220

Height
[m]

hydrostati
c

1,
0

0,
8

1,3

0
0

1
0

2
0

1,
6
1,
4
3
0

rising speed in
m/h

10

2
4
0

5
0

6
0

7
0

Pressure [kPa]

Fig. 5. Formwork pressure for different rising speeds for SCC

Slump

Flow diameter

Degree of compaction

Fig. 6. Three determination methods for determining the consistence of fresh

concrete.

221

270
3
0

70

60

flow
cone

r0=100
paste

r1

240

r2
120

60
30

Fig. 7. Japanese tools to measure the rheological properties of SCC in the

fresh state: the slump flow test (left) and the funnel test (right).
Funnel time [sec]
9-25

5-9

3-5
Slump flow
1

2
3
Consistency classes

470-570
4

>

540-660

630-800 [mm]

Self Compacting Concrete

Fig. 8. Extension of conventional consistence classes with SCC, according to a

Dutch proposal.

222

Funnel time (sec)

9-25

High &

Ramps

Slender

Walls

5-9

Floors

3-5

Consistency class

470-570

540-660

630-800

Slump flow (mm)

Fig. 9. Areas of application of SCC in relation to optimum rheological

properties, defined using the criteria funnel time and flow diameter.

Fig. 10. Architectural element of SCC

SCC girder for metro station in Amsterdam.

223

Fig. 11. Large prestressed

Fig. 12. Foundation piles of SCC

Fig. 14. SCC faade in The

Hague, The Netherlands

Fig. 13. Concrete arches made of SCC

Fig. 15. Sports stadium in Roma under

construction (2009) with full use of SCC.

224

Fig. 16. Remote casting of a wall with openings in the interior of a box girder
bridge for creating deviation points for additional external prestressing
tendons, aiming at increasing the bearing capacity (Ketelbridge in The
Netherlands, 2002).

225

Fig. 17 Casting the end wall of an element for a submerged tunnel in SCC.

140
120
100
Fibre
content

80

[kg/m3]

60
Reference
mixture

40
20

Mix 4 (68/39.0)
Mix 3 (68/36.5)
Mix 2 (57/39.0)
Mix 1 (57/36.5)

0
45/30
Fibre type

65/40

80/30

80/60

Fig. 18 Maximum fibre content

for SCC in dependence of fibre type
and mixture composition

226

Fig. 19 Self-compacting
concrete, with 125 kg/m3
fibres, in strength class C115

Fig. 20. Testing the flowability of a high performance fibre reinforced

concrete C200

Fig. 21. Strengthening of the Huisne bridge in France [13]:

- widening of beam webs with ultra high strength fibre concrete (left)
- remote casting of concrete through the deck (right)

227

Fig. 22. Remote casting test for low strength SCC for nuclear waste disposal
[14].

Fig. 23. Lightweight aggregate particle with skin of with cement paste [9].

228

PRACTICAL AND SUCCESFUL SCC APPLICATION IN

THE PRECAST CONCRETE INDUSTRY
F. Surico(1), A. Gasperi(2), R. Marino(3), J. Perazzoli(4)
(1) R&D Admixtures for Concrete, Mapei SpA, (2) Structures Designer, (3)
Concrete technology consultant, (4) Technical director of Area Prefabbricati SpA

INTRODUCTION
Self Compacting Concrete (SCC) has gained wider acceptance since its
development in the 80s in Japan. Today, especially in the precast industry, it is
an option for the production managers who search for production costs savings
and improvements of the quality of the concrete structures at the same time.
SCC utilization in the precast industry is nowadays a consolidated reality also in
Italy, even if there is still room for growth to get to the same share of utilization of
other European countries (1)
Moreover the utilization of Self Compacting Concrete represents a technological
innovation that involves both on the material and casting procedure sides. As a
matter of fact to implement SCC in an industrial process requires a careful study
of the mix formulation to achieve the desired technical requirements, but also a
thorough evaluation of the mixing and pouring processes that are usually designed
and meant for lower (usually S5) consistency class concrete mixes. (2)
This work has also the scope of underlining how, on the one hand SCC
production is easily feasible in a modern precast yard even if not designed for
SCC, on the other to draw attention to some necessary precautions to adopt in
order to have satisfactory results (3).
Technical requirements for the structural elements
Different types of beams were considered for the present case study: TI2 shaped,
TI shaped and rectangular section shaped. The first two types being produced
with prestressed reinforced concrete.
The above mentioned elements were originally designed for 55 N/mm2
characteristic strength and to be cast with fluid concrete with consistency class of
S5, determined by the slump value of the Abrams cone method. The maximum
diameter of the aggregate used was set to 25mm.
The request was to improve the appearance of the elements that were originally
characterized by excess of blow holes on the surface, and to speed up the
production operations at the same time.

229

It is known that the utilization of SCC sharply improves the surface appearance of
the elements bringing in addition a better ability to resist to the aggressive agents
of the environment, and thus to an improved durability of the service life of the
whole structure.
The choice was then to design again the concrete mix to switch from S5
consistency class to SCC. In addition to that an experimental program to improve
early strength by selecting among different admixtures for precast concrete was
decided, in order to gain productivity at the same time.
The study of the mix design was carried out according to the technical
requirements without changing the raw materials already used for normal
production with the only exception of adding a limestone filler and, of course,
selecting the most appropriate admixture combination.
MATERIALS
Limestone Filler: Commercially available limestone filler from Rezzato, Italy was
used. Its particle size distribution is reported in Fig. 1.
Superplasticizers: Two PCE High Range Water Reducing admixtures specifically
designed by MAPEI for precast concrete production (Type HRWR + HAA pr.
3+7 according to EN 934/2) were selected. Their characteristics are listed in Table
1.
Aggregates: Particle size distribution of aggregates is reported in Table 2.
Cement : A commercially available Cem I 52,5R from Italcementi was used.
TEST METHODS
Concrete Tests: The mixes were designed to have Slump Flow value of about 750
mm measured according to prEN 12350-8 (4). V-funnel tests were done according
to prEN 12350-9 (5) and L-Box test according to prEN 12350-10 (6) to better
define the mix design at laboratory scale. The proportions of the selected mixes
are reported in Table 3.
SCC mixes were prepared in a forced action mixer with interchangeable mixing
plastic tubs (Collomix mod. XM3-900). All materials (aggregates, cement and
water) were dried at ambient temperature overnight prior to use.
At industrial scale several mixes were produced in a 1cubic meter forced action
mixer. Different mix designs were tested in the above mentioned elements
evaluating both rheological and aesthetical aspects.

230

RESULTS AND DISCUSSION

Laboratory Concrete Tests: The results of the three selected mixes from lab tests
are reported in Table 4. They show that all the mixes were adequate in terms of
initial Slump Flow with admixture dosages ranging from 3.3 to 5.5 l/m3. SCC1
was clearly indicated as the most suitable for the application at industrial scale,
being the one with the highest flow and the faster V-funnel test even if the dosage
of superplasticizer required was the highest.
SCC2 was instead not considered as a good solution in the proposed mix, because
of the poor behavior in terms of V- funnel and L-box, despite to the fact that the
admixture dosage was the least.
SCC3 was then considered interesting and promising for an industrial scale
evaluation considering the simultaneous presence of good initial flow and faster
strength development in comparison to the one of SCC1.
Industrial production tests: The mixes produced and tested at industrial scale
resulted in a slight modification of the aggregate proportions and in the limestone
filler content.
The limestone filler content was increased from 150 to 200 kg/m3 in order to
improve segregation resistance of the SCC taking into account the load-to-load
variations in the moisture content of the aggregates.
The final recipe of the SCC produced in the Precast yard, as reported in Table 4,
is characterized by the following rheological and compositional indexes: Slump
Flow 700-750 mm Time to reach flow diameter of 500 mm (T50) ab. 4 sec VFunnel ab. 4 sec w/c 0,50 w/p 0,34 Paste volume 420 l/m3 Volume of the
total matrix (< 0,125 mm) 446 l/m3. In Table 5 the results of the fresh and
hardened properties of the industrial SCC tested are reported.
Due to the modification in the mix design, the admixture dosage required to
achieve the target slump flow had increased from 5,5, to 6,0 l/m3 in the case of
Dynamon SP1 and from 4,5 to 5,0l/m3 in the case of Dynamon NRG 1012.
Both mixes were considered excellent in terms of surface finishing in all the
element tested and their filling ability without vibration was adequate to the
reinforcement structure of the elements.
The analysis of the compressive strength results shows that the SCC admixed with
Dynamon NRG 1012 has faster development, resulting in a 27% increase after 16
hours of curing at ambient temperature compared with Dynamon SP1. It can be
noted that the industrial mix with Dynamon SP1 needs 20 hours of curing (4
hours more) to get to the same strength level fo the one mixed with Dynamon
NRG 1012.

231

CONCLUSIONS
The study, was conducted according to the new European Normative on Self
Compacting Concrete (7), has lead to the conclusion that the technical
requirements of the final compressive strength, the filling of the moulds and the
surface finishing of the elements were met by adopting the self compacting
concrete technology. In addition, when using a recently developed admixture
designed for the precast concrete production, it is possible to improve the
compressive strength results at early ages allowing thus a faster mould release
with a positive effect on productivity.
The most recent precast yards are designed to have little or even no transportation
of the fresh SCC and are equipped with moulds on rails that are able to be filled
directly under the concrete mixer. The precast yard involved in the study was
instead not originally designed to produce SCC, and is equipped with a
production line for traditional vibrated concrete. Therefore SCC production and
casting led to a substantial change in the production methods.
Even if the 2 m3 bucket was not possible to eliminate, the casting procedures were
conducted with a modification able to let the SCC flow in the mould without
falling. By means of this precaution it was possible to take the most of the
rheological properties of this highly flowable material.
Picture 2 shows the modified bucket with a flexible hose allowing SCC to be cast
from the bottom of the element without free falling. Moreover, in order to further
allow entrapped air to easily flow away, the pouring was done from one side of
the element without moving the bucket, allowing the SCC to flow from the one
end to the other.
This result was only possible by an accurate selection and proportioning of all the
ingredients (aggregate skeleton, cementitious systems and admixture) in order to
have an highly flowable material with no segregation and optimum strength
development combined with a critical evaluation of the mixing and pouring
processes. The precast concrete producers presently adopting traditional vibrated
concrete can consider to improve the quality of their elements and the overall
productivity by moving to Self Compacting Concrete production after a careful
study of the mix and of the production conditions.

1.

2.

REFERENCES
J. Walraven, Self Conpacting Concrete: Challenge For Designers And
Researchers, SCC Symposium, 2005 Chicago.
K. Juvas, European Experiences with Self Compacting Concrete in precast
Concrete Industry, SCC Symposium, 2008 Chicago.

232

3.

Cotic, Koglot, Zarnic, Bokan Bosiljkov, Experiences For Scc Mixes For
Precast Elements And Field Applications In Slovenia, SCC symposium,
2008 Chicago.

4.

prEN12350-8:Testing fresh concrete, Part.8 Self-compacting_concrete,

Slump-flow_test.

5.

prEN12350-9:Testing fresh concrete, Part. 9 Self-compacting concrete,

Vfunnel test.

6.

prEN12350-10:Testing fresh concrete, Part. 10 Self-compacting concrete,

Lbox test.

7.

prEN206-9:Concrete Part. 9 Additional Rules for Self-compacting Concrete

(SCC)

233

Admixture

Table 1 - PCE Admixtures characteristics

Chemical
Nature

Dynamon SP1

Solid Density
(%)
(g/cm3)

Modified Acrylic Polymer - PCE

30

1.09

Dynamon NRG 1020 Modified Acrylic Polymer - PCE

30

1.07

Dynamon NRG 1012 Modified Acrylic Polymer - PCE

26

1.06

Passing %

Table 2 - Particle size distribution of aggregates

Sieve
Gravel
Sand
Gravel
Opening
6-10
0-4 mm 2-6 mm
(mm)
mm
31,5
100
100
100
20,0
100
100
100
16,0
100
100
100
12,5
100
100
100
10,0
100
100
99
8,0
100
100
55
6,3
100
90
7
4,0
99
19
0.9
2,0
75
0.1
0.5
1,0
50
0.1
0.5
100
90
80
70
60
50
40
30
20
10
0
0,0

0,1

1,0

10,0

Sieve opening in mm
S1

G1

G2

234

G3

Risultante aggregati

100,0

Table 3 -Proportions of SCC

SCC
SCC
Laboratory Production
w/c

0.50

0.50

w/p

0.36

0.33

Cement kg/m3

400

400

Limestone Filler
kg/m3

150

200

Sand 0-4 mm kg/m3

798

877

Gravel 2-6 mm kg/m3 267

197

Gravel 6-10 mm
kg/m3

461

526

235

Dos. l/m

l/m3

Temp. of the mix

Lbox (%)

V funnel (sec)

T50 (sec)

Air (%)

Slump Flow (mm)

SCC
1

Dynamon
SP1

0.90

5.5

201

0.36

2373

26 C

100

2.2

0.5

780

SCC
2

Dynamon
NRG 1020

0.56

3.3

198

0.36

2347

26 C

<80

22

3.5

2.4

740

SCC
3

Dynamon
NRG 1012

0.75

4.5

201

0.36

2377

26 C

95

12

2.8

1.0

770

Admixture

Dos. % vol/pwd

Table 4 - Results of fresh and hardened properties of the lab mixes

Water

a/p

S.G.

kg/m
3

Water
Dos. % Dos.
vol/pwd l/m

l/m3

w/p

Cs (N/mm2) 20C 95%u.r.

16h

18h

20h

24h

7gg

28gg

SCC1

0.90

5.5

201 0.36 17.5 21.4 23.4 29.3 66.3

77.4

SCC2

0.56

3.3

198 0.36 25.8 28.6 29.4 35.1 66.0

75.8

SCC3

0.75

4.5

201 0.36 23.3 26.8 26.8 34.4 66.7

76.4

236

Table 5 Test results

Fresh
measures

Water

Cs (N/mm2) 20C
in water

Dos.
%
vol/
pwd

Dos
.
l/m

l/m3

w/p

Slump
Flow
(mm)

T50
(sec)

16h

20h

7d

28d

Dynamon
SP1

1.00

6.0

203

0.34

730

3.0

23.2

30.5

54.3

62.6

Dynamon
NRG1012

0.85

5.0

201

0.34

780

4.2

29.6

53.9

63.2

Admixture

237

238

Fig. 1 - Modified bucket

239

Fig. 2 SCC flow

240

Fig. 3 Surface finishing examples

241

242

EVALUATION OF SCC FORMWORK PRESSURE

Kamal H. Khayat and Ahmed F. Omran

Synopsis: Self-consolidating concrete (SCC) is a new class of high-performance

concrete that can flow readily under its own weight and consolidate without
mechanical vibrations and with minimum risk of segregation. SCC has been
successfully used in North America in the precast industry; however, the use of
SCC in cast-in-place applications has been hindered by some technical issues,
including formwork pressure exerted by SCC. Lack of information regarding
formwork pressure variations during casting and pressure decay following
placement has prompted contractors and engineers, as recommended by ACI 347
(Guide to Formwork for Concrete), to design for full hydrostatic pressure.
Unfortunately, this leads to cost increase of the formwork system that
compromises profitability due to rapid placements and possible labor savings
associated with SCC.
This paper presents a summary of an extensive research project that aimed at
developing formulation expertise and guidelines to lower lateral pressure of SCC.
The project was sponsored by the National Ready Mix Concrete Foundation and
the Strategic Development Council of the American Concrete Institute. The paper
highlights the development of a pressure column devise that can be used to
predict lateral pressure exerted by SCC as well as empirical test methods to
determine the structural build-up at rest of SCC, which is shown to significantly
affect formwork pressure. The role of material constituents, mix design, concrete
casting rate, formwork geometry, and temperature on SCC form pressure are
highlighted. Proposed design equations to predict formwork pressure of SCC on
column and wall elements are also presented.

Keywords: Form pressure, SCC, structural build-up at rest, thixotropy

243

Kamal Henri Khayat is Professor of Civil Engineering and Director of the

Integrated Research Laboratory in the Valorization of Materials and Innovative
and Durable Structures at the Universit de Sherbrooke in Canada. He is Fellow
of the American Concrete Institute and serves on several technical committees of
ACI, RILEM, and CSA, including Chairman of ACI 237 on SCC and RILEM TC
on Mechanical Properties of SCC.
Ahmed F. Omran is Postdoctoral Fellow at the Department of Civil Engineering
at the Universit de Sherbrooke where he received his Ph.D. His research interests
include rheology and form pressure of self-consolidating concrete.

PORTABLE DEVICE TO MEASURE SCC FORMWORK PRESSURE

A portable pressure device, referred to as UofS2 pressure column, was developed
following a former device (UofS1 pressure column) [1] that was successfully used
by the researchers to evaluate lateral pressure exerted by plastic concrete. The
UofS2 pressure column shown in Fig. 1 has a circular in cross-section measuring
0.2 m in internal diameter and 0.7 m in total height with a wall thickness of 10
mm. The tube is initially filled to a height of 0.5 m with SCC at the required rate
from the top without any vibration. The top of the pressure column is then firmly
closed, and air pressure is gradually introduced from the top to simulate pressure
increase up to 13 m at a given placement rate. A pressure sensor is set flush with
the fresh concrete surface at 63 mm from the base of the column to record the
exerted lateral pressure during casting and to monitor the pressure decomposition
during the plastic stage of the concrete. Another transducer is fixed above the
concrete surface at 625 mm from the column base to determine the net overhead
pressure inside the column.
The sensor is AB-high-performance pressure transducer supplied by Honeywell
and is extremely accurate down to 0.25% over a wide compensated temperature
range. It works using semiconductor gages on bending beams isolated by stainless
steel media and has a capacity of 1380 kPa (200 psi). The sensor is 19 mm in
diameter and can operate over a temperature range varying from -54 to +93C.
The sensors are connected to data acquisition system to monitor pressure variation
at 90-s intervals. The sensor is excited using 5 V dc current. Numerical dial-gauge
(manometer) was set on a controlling chamber attached to the UofS2 pressure
column to manage the air-pressure flow on the top surface of the concrete to
simulate further concrete heights.
Typical examples of the variations of lateral pressure envelops with time are plotted
in Figs. 3 to 5. The tested SCC had initial slump flow of 660 15 mm, paste

244

volume of 370 l/m3, w/cm of 0.42, and polycaroxylate-based high-range waterreducing agent (HRWRA) and viscosity-modifying agent (VMA) dosages of 3.6
and 2.8 l/m3, respectively. The casting rate was set at 10 m/hr. The column was then
sealed, and air pressure was applied in steps equivalent to one-meter of concrete
head at the same casting rate of 10 m/hr until a concrete head corresponding to 13
m was reached. The corresponding lateral pressure exerted on the column wall was
recorded using the sensor at the base of the UofS2 pressure column, as shown in
Fig. 3. Between each consecutive increment, a decrease in lateral pressure profile
was obtained reflecting the restructuring (or structural build-up at rest) of the
concrete. This was more apparent at the end of casting where pressure decay is
observed. The pressure device is usually demolded before concrete hardening. The
maximum lateral pressure (Pmax) recorded at each increase of simulated concrete
height is plotted in Fig. 4 indicating that the SCC developed 40% lower relative
lateral pressure (K0) than the equivalent hydrostatic pressure at concrete height of
13 m. K0 refers to the ratio between Pmax and equivalent hydrostatic pressure (Phyd)
and is plotted against concrete height in Fig. 5.
In order to determine the repeatability of the UofS2 pressure column, an SCC
mixture was prepared and tested four times. The results of the relative errors of K0
at various heights are given in Table 1 and indicate high precision of pressure
measurements.
The lateral pressure characteristics of three different SCC mixtures determined
using the UofS2 device were also compared to measurements obtained from a
PVC column measuring 0.2 m in diameter and 3 m in height. Good agreement
was obtained between both systems in terms of initial lateral pressure and
pressure drop in time [2]. Further validation of the UofS2 pressure device clearly
demonstrated that the device is capable of adequately reflecting the effect of
placement rate, initial slump flow, and mix design (including paste volume, w/cm,
maximum size aggregate (MSA), and VMA dosage) on lateral pressure
characteristics [2].
EMPIRICAL TESTS TO EVALUATE STRUCTURAL
BUILD-UP AT REST
A total of six field-oriented test methods were developed and used to evaluate the
structural build-up at rest of concrete-equivalent mortar (CEM) and SCC mixtures.
The portable vane, inclined plane, and undisturbed spread test methods showed good
repeatability and low relative error [2]. The response obtained with the portable vane
and inclined plane were compared to rheological measurements obtained using a
modified Tattersall-Type MK-III concrete rheometer [3] using 22 SCC mixtures
prepared with various mix designs and material constituents. The comparison clearly

245

indicates that the empirical methods can capture the structural build-up of SCC at rest
in terms of static yield stress and its rate of gain in time [2].
Portable vane test
The portable vane test is inspired from a field test for in-situ measurement of
shear strength of soil (in particular clay soils). Four-blade vanes of different sizes
(Table 2) were manufactured from stainless steel to enable the use of one torquemeter to capture shear strength of the plastic concrete after various times of rest
(Figs. 6 and 7). The largest vane is used for the weakest structure, i.e., shortest
resting time, and vice versa. A torque-meter measuring with high precision was
employed to capture the torque values.
Immediately after mixing, the four vanes are centered vertically in the containers.
The containers are filled with SCC to a given height (h) indicated in Table 2. The
rested materials are covered. After 15, 30, 45, and 60 minutes of rest, the torquemeter is attached to the axis of the vane and turned slowly (10 to 15 s for a quarter
turn). The maximum torque needed to break down the structure is then noted. The
torque values are converted to static shear stress (0rest) using Eqs. 1 and 2 [3] as
follows:
0rest

T
G

G 2 r2 h

(1)

1
r
3

(2)

where: T = measured torque (N.m), and the other terms are defined in Fig. 7.
Variations of static yield stress (0rest) determined using the portable vane test
(PV0rest) with resting time for typical SCC mixtures designed with different
thixotropic characteristics are shown in Fig. 8. The 0rest obtained from the
portable vane at 15 min (PV0rest@15min) corresponding to the initial response (Ri)
is used as a structural build-up index. Similarly, the rate of change of static yield
stress with time [PV0rest(t)] or [R(t)] and the coupled effect of RiR(t) can also be
for structural build-up indices.
Inclined plane test
The inclined plane method [4] involves casting concrete in a cylindrical mould
onto a horizontal plate of a given roughness, then lifting the plate to initiate flow
of the material, as shown in Figs. 9 and 10. The corresponding angle the
necessary to initiate flow is used to determine the static yield stress, 0rest (Pa), as
follows:
0rest = .g.h.sin

246

(3)

where is the density of the sample (in g/cm3), g is the gravitation constant (=
9.81 m/s2), h is the characteristic height (in mm) of the slumped sample, and is
the critical angle of the plane (in degree) when the sample starts to flow. The
characteristic height (h) is determined by calculating the mean value of five
heights of the slumped sample near the center of the spread. Four tests are
performed after different periods of rest (after 15, 30, 45, and 60 min) to evaluate
the rate of increase in 0rest at rest.
Variations of 0rest obtained with the inclined plane test (IP0rest) with resting time
for typical SCC mixtures of different structural build-up, or thixotropic, properties
are shown in Fig. 11. Similar to the portable vane test, three structural build-up
indices can be obtained using the inclined plane method: (Ri), [R(t)], and
[RiR(t)].

Undisturbed spread test

The undisturbed spread test involves leaving the samples of the plastic material at
rest in four cylinders (Fig. 12) for different durations of time, typically 15, 30, 45,
and 60 min. At each time of testing, the undisturbed spread (US) of the rested
sample is measured as the average of two perpendicular diameters. The
decreasing rate of concrete spread is used for the structural build-up index [R(t)],
as shown in Fig. 13.
A summary of the structural build-up indices determined from the selected
empirical test methods are presented in Table 3.
PREDICTION MODEL FOR RELATIVE LATERAL PRESSURE
A comprehensive testing program was undertaken to evaluate key mixture
parameters affecting formwork pressure exerted by SCC. The investigated
parameters are given in Table 4 and included mix design, concrete constituents,
concrete temperature, casting characteristics, and minimum formwork dimension.
The UofS2 pressure column and the empirical test methods were employed to
evaluate the lateral pressure characteristics and relate them to relevant SCC
rheological properties.
Analytical models for relative lateral pressure prediction
The results obtained from approximately 795 data points were used to establish
analytical models to predict form pressure of SCC. In total, seven analytical
models enabling the prediction of relative lateral pressure (K0) including the

247

thixotropic indices shown in Table 3 determined from the empirical test methods
are discussed. The first model was derived to predict K0 as function of concrete
height (H in m), casting rate (R in m/hr), concrete temperature (T in C), and
structural build-up (Eq. 4). The structural build-up is expressed in terms of static
yield stress at rest after 15 min of rest (referred as initial value) determined using
the portable vane test (PV0rest@15min in Pa). The influence of maximum size
aggregate (MSA) is incorporated in the model (fMSA) when using MSA other than
14 mm. The effect of different waiting times between successive lifts (fWT) is also
taken into consideration. PV0rest@15min measurements in Eq. 4 are measured at 22
2 C. Thus, a separate factor expressing the influence of concrete temperature is
introduced.
K0 = [112.5 - 3.8H + 0.6R - 0.6T +
0.01Dmin - 0.021 PV0 rest@15 min] fMSA fWT

(4)

where
fMSA is a correction factor for MSA other than 14 mm.
For relatively low thixotropic SCC [PV0rest@15 min 700 Pa] and
H<4m

fMSA = 1

H = 4 -12 m

fMSA = 1

fMSA = 1 +

1.26 H - 5.04
100

when MSA = 20 mm

. when MSA = 10 mm

For high thixotropic SCC [PV0rest@15 min > 700 Pa], fMSA = 1
fWT is a correction factor reflecting the effect of waiting time (WT)
between successive lifts and ranges between 0.85 and 1.0 for a waiting
time of 30 min, depending on the thixotropy of the concrete.
When the structural build-up varies with the temperature of the SCC at casting, a
modified prediction model for K0 values is shown in Eq. 4'. The fMSA and fWT
factors are the same as those defined for Eq. 4.
K0 = [98 - 3.82H + 0.63R + 0.011Dmin - 0.021 PV 0rest@15min] fMSA fWT

(4')

Correlations between measured and predicted K0 values are shown in Figs. 14 and
15. The predicting K0 from the models given in Eqs. 4 and 4' gives 1:1
relationship with coefficient of correlation (R2) of 1, which indicates excellent
agreement between both prediction models for K0.
Other two predicting models using the structural build-up indices of the portable
vane test were developed. The time-dependant change of static yield stress, i.e.,
rate, (PV0rest(t) in Pa/min) and the couple effect of initial static yield stress and its

248

rate of change with time [PV0rest@15minPV0rest(t) in Pa2/min] are expressed as

structural build-up indices. The two models compromising the two indices are
shown in Eqs. 5 and 6, respectively.
K0 = [109.5 - 3.9H + 0.7R - 0.6T + 0.003Dmin - 0.29 PV0rest(t)] fMSA fWT

(5)

K0 = [106 - 4H + 0.6R - 0.63T + 0.01Dmin - 0.00015 PV0rest@15minPV0rest(t)] fMSA fWT

(6)

Similar predicting models for K0 derived using the three structural build-up
indices of the inclined plane test are shown in Eqs. 7 to 9.
K0 = [112 - 3.83H + 0.6R - 0.6T + 0.01Dmin - 0.023 IP0rest@15min] fMSA fWT

(7)

K0 = [104.2 - 3.9H + 0.6R - 0.6T + 0.0036Dmin 1.22 IP0rest(t)] fMSA fWT

(8)

K0 = [104.7 - 4H + 0.6R - 0.63T + 0.019Dmin - 0.0007 IP0rest@15minIP0rest(t)] fMSA fWT

(9)

The decreasing rate of concrete spread [US(t)] obtained from the undisturbed
spread test is used for the structural build-up index in the predicting equation of
K0 (Eq.10).
K0 = [106 - 3.95H + 0.6R - 0.62T + 0.016Dmin 1.762 US(t)] fMSA fWT

(10)

It is imperative noting down that the prediction models (Eqs. 4 to 10) are valid for
the ranges of tested parameters shown in Table 4 in terms of mix design, casting
conditions, and formwork geometry. These limitations are as follows:
H= 1 13 m

PV0rest@15min= 0 2000 Pa

R= 2 30 m/hr

PV0rest(t) = 0 125 Pa/min

T= 12 32 C

IP0rest@15min= 0 2000 Pa

Dmin= 200 400 mm

IP0rest(t) = 0 30 Pa/min
US(t) = 0 22 mm/min

The predicting models shown in Eqs. 4 to 10 can be used to create abacuses to

ease their practical application. Samples of the abacuses are shown in Figs. 16 to
19, where K0 at a given concrete height can be estimated given one of the
structural build-up indices resulted from the various empirical test devices. The
relationships shown in Figs. 16 to 19 were constructed for constant casting rate,
concrete temperature, minimum formwork dimension, and waiting time between
successive lifts. The values of the fixed parameters are indicated in each figure.
CONCLUSIONS
A portable device (UofS2 pressure column) is developed to monitor lateral
pressure exerted by SCC. The pressure column is filled with 0.5 m of concrete

249

subjected to overhead air pressure to simulate free concrete head of up to 13 m in

height. Empirical field test methods (portable vane, inclined plane, and
undisturbed spread) are proposed to evaluate the structural build-up of SCC at
rest. The initial static yield stress after a rest period of 15 min, rate of change in
static yield stress with rest time, and the couple effect of these parameters can be
considered to describe the degree of structural build-up of the concrete. These
indices are shown to correlate well to the lateral formwork pressure determined
from the UofS2 device when using SCC. The analytical models (Eqs. 4 and 10)
expressing the major influencing parameters on formwork pressure are
established to predict formwork pressure exerted by SCC. Abacuses to ease the
practical application of the predicting models (Eqs. 4 and 10) are created.
REFERENCES
[1]

[2]

[3]

[4]
[5]

Khayat, K.H., and Assaad, J., Measurement Systems for Determining

Formwork Pressure of Highly Flowable Concrete, Materials and
Structures, 41, 2008, pp. 37-46.
Khayat, K.H., Omran, A., Neji, S., Billberg, P., and Yahia, A., Test
Methods to Evaluate Form Pressure of SCC,'' Proceedings of the
international conference of ACBM on SCC (SCC 2008), Chicago, USA,
Nov. 2008.
Yahia, A., Khayat, K.H., Modification of the Concrete Rheometer to
Determine Rheological Parameters of SCC Vane Device, Proc. 2nd Int.
RILEM Symp. on Advances in Concrete through Science and Engineering,
Sept. 2006, pp. 375-380.
Coussot, P., Boyer, S., Determination of Yield Stress Fluid Behaviour
from Inclined Plane Test, Rheol Acta, 34, 1995, pp. 534-543.
NCSS: statistical & power analysis software, http://www.ncss.com/.

250

Table 1- Relative error in predicting relative lateral pressure value (K0)

Concrete height, H (m)

Relative error (%)

0.7

2.4

2.3

12

4.0

Table 2 - Vane dimensions

Time at rest
(min)
Vane # 1 (largest)
Vane # 2
Vane # 3
Vane # 4 (smallest)

15
30
45
60

Vane dimensions (mm)

R
37.5
37.0
37.5
37.5

H
250
200
149
100

h
Vary according
to the filling
height of up to
total height (H)

Table 3- Various thixotropic indices obtained from the empirical test

methods
Response

Portable Inclined Undisturbed

vane (PV) plane (IP) spread (US)

Initial value at 15 min resting time

(0rest)
Ri, (Pa)

(0rest)
Ri, (Pa)

R(t),
(Pa/min)
RiR(t),
(Pa2/min)

R(t),
R(t), (mm/min)
(Pa/min)
RiR(t), -(Pa2/min)

Rate of change in the response with

time (slope of the regression line)
Couple effect of initial value and
rate (slope)

251

(average spread)
--

Table 4 - Investigated parameters affecting SCC formwork pressure

Parameter

Range

Initial slump flow

Dosage of VMA
Volume of coarse aggregate
Mix design
(Vca)
and
Paste volume (Vp)
material
Sand-to-total aggregate ratio
constituents
(S/A)
Maximum size aggregate
(MSA)
Concrete height (H)
Casting rate (R)
Concrete temperature (T)
Waiting time between consecutive lifts
(WT)
Minimum formwork dimension (Dmin)

252

Can be replaced by one of

structural build-up indices
determined using concrete
rheometer or empirical test
methods

10, 14, and 20 mm

1 - 13 m
2, 5, 10, 17, 24, and 30 m/hr
12, 22, and 32 2C
Continuous
WT of 30 min at middle of
casting
Two WTs of 30 min each at
middle of casting
200, 250, 300, and 350 mm

63.5 mm
Sensor
19 mm

577 mm

704 mm
500 mm

Sensor
19 mm
63.5 mm
190 mm

Fig. 1 - Schematic of UofS2

portable pressure column to
evaluate formwork pressure

Fig. 2 - Photo of UofS2 portable

pressure column to evaluate
formwork pressure

Fig. 3 - Variations of lateral pressure with time for typical SCC mixtures

253

Fig. 4 - Decrease in maximum lateral Fig. 5 - Variations of relative lateral

pressure of SCC with height
pressure (K0) of SCC

Fig. 6 - Four buckets and four vanes

used in the portable vane test

254

Fig. 7 - Schematic for the portable

vane test

PV o rest (Pa)

4000
3500
3000
2500
2000
1500
1000
500
0
0

15

30

45

60

75

90

105

Resting time (min)

Fig. 8 - Variations of static yield stress at rest with time obtained
with portable vane test

SCC concrete specimen

Sand paper covered
inclined plans

ISOMETRIC VIEW
10 mm

Bottom plate
Inclined plane

Angle
fixing rod
5 mm

FRONT VIEW
610 mm

400
400
mm mm

Fig. 9 - Inclined plane tests at

different rest times

PVC Plate 10 mm thick

C top & lower plate
Surface covered by sand
paper (No. 600) to avoid
slipping and sliding

PLAN

Fig. 10 - Schematic for inclined plane

test

255

Fig. 11 - Variations of static yield stress at rest with time obtained

with inclined plane test

b) Lifting rested cylinder to

measure spread

a) Cylinders filled with SCC

D2
D1

c) Slumped samples after

different resting times
Fig. 12- Undisturbed spread test

256

Fig. 13 - Variations of flow diameter obtained using the undisturbed spread

test (US)

Figure 14 - Measured to predicted

K0 from analytical model Eq. 4

Figure 15 - Measured to predicted K0

from analytical model Eq. 4'

257

Figure 16 - Correlation between K0 and static yield stress Figure 17 - Correlation between K0 and couple effect of static yield
after 15 min resting time obtained using portable vane test stress after 15 min resting time and its rate obtained using portable
for SCC cast, R=10 m/hr, resulted from Eqs. 4 or 4'
vane test for SCC, R=10 m/hr, resulted from Eq 6

258

Figure 18 - Correlation between K0 and static yield stress

after 15 min resting time obtained using inclined plane
test for SCC, R=2 m/hr, resulted from Eq 7

Figure 19 - Correlation between K0 and time-change of

undisturbed spread obtained from undisturbed spread test
[US(t)] for SCC, R=2 m/hr, resulted from Eq 10

259

- 260 -

ALTURKI BUSINESS PARK

SELF CONSOLIDATING CONCRETE: FROM HEAD TO TOE
Redwan A. Hameed and Narasimhulu Gary
Saudi Readymix Concrete Company

INTRODUCTION
Self consolidating concrete (SCC) is considered as a revolution in concrete
technology. Though its advantages are many, its usage is still limited worldwide.
The two main reasons behind this slow growth and acceptance are: the higher
cost of SCC compared to conventional concrete and the lack of knowledge about
SCC in construction industry. This is true worldwide and the market in Saudi
Arabia is no exception.
In 2003 Saudi Readymix Concrete Company took a responsibility to raise the
awareness about SCC in Saudi Arabia by sponsoring and conducting many public
and private seminars about SCC. Since then Saudi Readymix managed to pour
SCC in several projects throughout the kingdom. However, Alturki Business Park
(ABP) was a unique project and, to the best of our knowledge, the only building
in the Gulf Area that was constructed completely using SCC.
SCC history and technology, its properties, advantages and testing methods are
well documented in many other papers, guidelines and technical reports. It is not
within the scope of this paper to talk about these points. The main objective of
this paper is to share the experience and knowledge gained during the
construction of ABP and to present the challenges encountered and what can be
expected on projects where SCC is specified.
The project
Al Turki business Park project (Figure 1) is located in Al Khobar city in the
Eastern Province of Saudi Arabia. The project consists of one office tower and 10
office lofts brought together by spacious surroundings, harmonized architectural
treatment, serene landscaping, and unified image. The development provides
34,000 m2 of leasable space with a project site area of 8,600 m2. The lofts are
three-floor buildings. The tower is a 16 floor building with two underground
floors providing parking for 300 cars connecting directly to the building lobby.
The design is based on a 9-meter structural grid with a variable cantilever to form
a cylindrical building with a 38-meter diameter. The structure is a multiframe
system. The slabs are designed as post-tensioned slabs with drop panel at column
locations. Vertical elements were founded on a thick reinforced concrete mat
foundation supported on soil. Vertical loads transmission is done by concrete

- 261 -

columns and concrete walls, lateral stability is given by concrete shear walls
around the staircases and elevators. The building is fully modern and will
incorporate all of the latest features of intelligent building design, sophisticated
communications, security, computer and other low voltage systems.
The project was designed by Pei Partnership Architects; a New York based design
office, and built by a local Saudi contractor (Alsaad). A local engineering office
(THA) was hired to manage the project while a Lebanese engineering firm (SKP)
was appointed as the project consultant.
Construction commenced on June 2008 and expected to be completed on June
2010.
Why SCC?
Though it was not necessary to use SCC for the whole structure, the decision to
build ABP using only SCC achieved three objectives.
1. Technical advantages: SCC generally provides better quality concrete for the
structure. This was observed in ABP project.
Some elements were highly congested with steel reinforcement (Figure 2) and
SCC was the right choice to prevent segregation and honeycombing. SCC
passed though congested steel and filled the forms without any problem.
Moreover, the use of SCC assured a uniform concrete consolidation.
Compacting conventional concrete using poker vibrators usually results in non
uniform compaction where one can find well compacted areas, partially
compacted areas and areas that were not compacted at all. Placing and
finishing SCC was much easier than normal concrete. It was possible to pour
concrete for a 7 m high wall continuously in a very short time without any
placing difficulties and concrete surface was very smooth.
The surface finish was excellent and almost blemish free and there was no
need to do any surface repair (Figure 3). The contractor acknowledged that
this was one of the best surface finishes they had ever produced.
2. Raising the construction industry awareness. This was done by giving a real
full scale live example about the constructability and many advantages of
SCC. As a matter of fact, many interested engineers, consultants, students and
university teachers visited the site during construction to see SCC being
placed and learn from this experience. Technical leaflets will be published
about the project and distributed to potential clients, engineers, designers and
consultants.

- 262 -

3. Giving Saudi Readymix the chance to remain a leader in readymix industry,

gain more experience in SCC technology, and stay ahead of other competitors
in this field.
MATERIALS
It was very necessary to select raw materials with a good and consistent quality
for this project. Selection of these materials and laboratory trial mixes were
completed from an early stage before starting the construction.
Local ordinary portland cement (type I) conforming to ASTM C-150 was used in
all mixes. Superpozz , an ultra fine fly ash, with a mean particle size of 5 micron
and a surface area of 13000 cm2/g was used as a supplementary cementitious
material and filler. Dosage of Superpozz was initially 10% by weight of cement.
In a later stage silica fume conforming to ASTM C-618 was used instead of
Superpozz . The initial dosage of silica fume was 7% by weight of cement.
Crushed lime stone aggregates of 20 mm, 10 mm and 5 mm sizes were used. The
fine aggregate was local fine dune sand. Typical gradation results are shown in
figure 4.
The high-range water reducer was a polycarboxylate based admixture.
Project specifications
- Compressive strength for vertical elements: 60 MPa
- Compressive strength for other elements: 45 MPa
- Max concrete temperature at the point of discharge: 32o C
- Workability - Acceptance criteria (see Table 1)
The mixture
During the bidding stage, SCC mixes were designed and we started running lab
trials and adjust mixes proportion until we reached to what we thought are the
final mixes. However, when construction activities began and actual concrete
supply and pouring started, we had to do more adjustments and changes to come
up with as robust mixes as possible to suite the encountered challenges.
The main adjustments which took place included increasing paste content,
decreasing w/b ratio, optimizing mix gradations, optimizing the HRWR dosage,
replacing superpozz with silica fume and changing the dosage of these additives.
Table A1 (in the Appendix) shows the typical mixes used in the project.
Testing and quality control
Due to its nature and sensitivity, SCC required a stricter quality control program
than conventional concrete. This was taken into consideration from day one and a
testing and inspection plan was put in place.

- 263 -

The following tests were performed on raw materials throughout the project
duration:
- Tests on aggregates: gradation, moisture content and absorption, specific
gravity, flakiness and elongation, abrasion as well as chemical analysis.
-

Our main concern was the variation in aggregates quality.

Tests on cement: compressive strength and setting time to assure the
consistency of cement quality from the same source and from different
sources.

Tests on superplasticizer: Both specific gravity and TDS were determined

for each batch of admixtures received. Batches not conforming to the
manufacturers specifications were rejected.

Marsh cone test was conducted to determine the optimum admixture

dosage to be used in the mixes.

The performance of several polycarboxylate superplasticizers was

evaluated and the one which gave us the best performance was selected.

The following tests were performed on fresh concrete:

- Slump flow test (spread and T50), J-ring test, L-box test and the V-funnel
test. Typical results of different mixes are presented in Table A1. Only
mixes that satisfied the acceptance criteria were used.
-

Concrete temperature and air content were measured.

Resistance to segregation was monitored by checking the visual stability

index (VSI) (see Figure A-1 in the appendix). Though there was no
specification in this regard. We were accepting mixtures with VSI of only
1 or 0 (Figure 5).

Fresh samples were collected in standard 150 x 300 mm cylindrical molds

to determine the compressive strength of concrete.

Pumpability of SCC
Pumping high strength, SCC in high rise buildings is a challenge. Despite the fact
that SCC is flowable, pumping this type of concrete requires more pressure1 and
more care and attention. SCC is very sensitive and may exhibit segregation and
slump flow loss due to the change in pumping pressure or discharge rate.

- 264 -

In order to get information about pumpability of the designed mixes, Saudi

Readymix set up a pumping simulation system consisting of a Putzmeister 2110
HPD stationary pump and a 230 m long pipeline including 14 U-shape bends and
4 meters rise pipe (Figure 6). Pumping trials were conducted in cooperation with a
Putzmeister team.
Different mixes were tested in the laboratory with a special rheometer developed
by Mr. K. Kasten from Putzmeister (Figure 7). The results obtained from the
rheometer were checked against the results obtained later from the pumping
simulation. A good correlation was observed (Figure 8). Based on these results it
was determined whether the mixes were pumpable or not.
Pump hydraulic pressure and concrete pressure in the pipeline were measured at
two points. One concrete pressure sensor was installed directly after the hopper
(Figure 9); the other sensor was installed after 120m.
During pouring pumping pressure vs. discharge rate was monitored and recorded.
The pressure was quite low even when concrete was pumped to the 14th floor
(Figure 10). These results were very close to that obtained in the lab from the
rheometer. This indicated that the mix design and rheological properties of the
mixtures were excellent.
Challenges and obstacles
Production and supply of SCC is not easy. Producing 0.05 m3 of SCC in a lab trial
under controlled conditions using uniform and consistent raw materials is very
easy. Unfortunately, this is not the case when it comes to actual production and
supply. One should be prepared for surprises, challenges and obstacles. The main
challenges encountered in this project were:
- Lack of awareness and knowledge about SCC among most parties
involved in the project. They started to deal with SCC with skepticism and
considered it guilty until proven otherwise. The benefits of using SCC
were not so sure and clear to them. With time and work progress, the
knowledge increased and some benefits started to be realized.
-

Resistance to change. This was expected and it was a direct result of the
previous point.

Inconsistency of raw materials. SCC is very sensitive to change in raw

materials. Through the course of the project the quality of the aggregates
and cement kept changing and the performance of SCC was affected
accordingly. It was a challenge to produce a consistent SCC with the
required rheological properties. In many cases, it was hard to control the
mixtures quality and each batch had to be adjusted. Admixtures dosage

- 265 -

and mixtures ingredients had to be changed and tweaked with the variation
of aggregates quality and moisture condition. There was a shortage in
good quality aggregates and it was not possible to maintain the same
quality of raw materials. The same problem was observed with cement
whose quality was oscillating and difficult to predict.
-

Loss of flow. This was confronted several times and in some cases it was
necessary to use vibrators to compact the concrete. One reason behind this
drop in flow was pumping. We knew that pumping is affecting the
rheological properties of SCC and that was causing the drop in flow from
700 50 mm before pumping to about 575 25 mm after pumping.
However, the exact scientific explanation was not clear to us. Recently, we
understood that pumping affects the viscosity and the yield stress, and
hence flowability, of SCC concrete due to structural breakdown and an
increase in the air content. Both factors decrease the viscosity. However,
yield stress increases due to structural break down and increases due to an
increase in air content1. In ABP project we measured the air content and
found it to increase from about 1.6% before pumping to about 2.5% after
pumping.

Loss of flow was also caused by the delay of trucks on site and due to high
temperature and drying of concrete.

Though it was not observed in lab trials and on station during production,
in some pours, bleeding water was noticed clearly on the top of concrete
after pouring. In one case, a paste layer of about 2.5 cm thick was formed
on the top of the slab and had to be removed. In those cases, the mix was
adjusted immediately. In other cases superpozz caused an unpleasant
stained surface on some columns due to the floatation of superpozz to the
surface with the bleeding water. As a result, we stopped using supepozz
and replaced it with silica fume. This adjustment gave more robust mixes
and the stains were never noticed again.

At early stage of the project and upon striking of the forms, pour lines
were observed in the soffit of one slab. However, further inspection and
investigation proved that these pour lines were superficial and did not
have any adverse results on the structure.

- 266 -

Occasionally, blow holes were noticed on the surfaces. This was attributed
to air migration or air popping where air migrates to the surface. In order
to minimize and eliminate this undesirable occurrence the viscosity and
flow of the mix were modified.

Benefits
The benefits obtained from using SCC for the project may be summarized as
follows:
- More understanding of SCC behavior, production, transporting and
placing. The project provided important learning topics.
-

Increased awareness and knowledge of all parties involved in the project

about SCC.

Uniform concrete consolidation.

Reduction in manpower and equipment

Reduction in noise, since poker vibrators were not needed.

Excellent, smooth surface finish

Easier placing and finishing of concrete.

Accelerated project schedule

Reduction in equipment wear and tear.

CONCLUDING REMARKS AND RECOMMENDATIONS

As a ready-mixed concrete supplier, we learned a lot from this project. The main
learning points are:
- Education and awareness about SCC is very essential. It is important to
train and educate all personnel involved in the design, production, quality
control, and placing of SCC before starting the project.
-

Pre-pour meetings are important to plan for successful pouring and to

assign the roles. In ABP project case, a meeting between technical and
production team would be held a day before the pour to put a detailed plan

- 267 -

and to assign tasks for the team members in order to make sure that the job
would go smoothly.
-

Conduct necessary trials and mock ups ahead of time and get good
knowledge about the behavior of SCC using different raw materials.

Modify and adjust and optimize the mix as necessary during the course of
the project.

SCC is sensitive to variation. Try to keep the raw materials uniform and
consistent and from a single source, if possible.

Select the most proper and suitable admixture. Different admixtures will
perform differently.

Working with SCC requires stricter and closer quality control.

Coordination between plant and site is essential.

Interval between trucks should be planned carefully.

Rate of placing SCC should not be fast to allow for entrapped air to escape

Pump SCC using low pressure and continuous constant flow.

When working with SCC, expect and be ready to encounter different

challenges.

Since SCC does not bleed, plastic shrinkage cracks are more likely to
occur during hot weather concreting and necessary preventive precautions
need to be taken.

There is still a lot of work to be done to optimize SCC mixes and reduce
its cost.

Both academia and industry in local markets should cooperate to reach to

a better understanding and utilization of SCC technology.

Finally, remember the six Ps Proper Prior Planning Prevents Poor

Performance

- 268 -

REFERENCES
1. Feys, D., De Schutter, G. and Verhoeven, R., Rheology and Pumping of SelfCompacting Concrete, Proceedings of the tenth ACI international conference on
recent advances in concrete technology and sustainability issues, Seville, Spain,
2009.
2. Kasten, K., Pump Trial for SCC at SRMCC (unpublished report),
Putzmeister, Germany, 2009.

- 269 -

Table 1 - Workability acceptance criteria

Test

Unit

Limits

Slump flow

Mm

Min
650

max
750

T50

Sec

J-Ring(flow
difference)
L-Box

Mm

50

h2/h1

0.8

1.0

V-funnel

Sec

12

Test

Unit

Limits

Slump flow

mm

Min
650

max
750

T50

Sec

(flow mm

50

h2/h1

0.8

1.0

Sec

12

J-Ring
difference)
L-Box
V-funnel

- 270 -

1A -Basements

1C - Almost completed

1B -First Floors

1D Imaginary view of tower upon completion

Figure 1 - ABP project at different stages

- 271 -

Figure 2 - Congested reinforcement in shear walls

Figure 3 - Excellent surface finish was obtained using SCC

- 272 -

100
90
80

Percentage passinge

70
60
20 mm

50

10 mm
40

5 mm
Fine sand

30
20
10
0

0.075

0.75

7.5

Sieve size (mm)

Figure 4 -Typical gradation of aggregates

Figure 5- SCC slump flow

- 273 -

Figure 6- Saudi Readymix pumping simulation setup

Figure 7- Rheological testing of concrete in the lab

- 274 -

Figure 8- Comparison between rheometer and pump trial results

Figure 9- Concrete pressure sensor after hopper

- 275 -

90
80

Pumping presure (bars)

70
60
50
40
30
20
Actual

10

Linear (Rheometer)
0
0

20

40

60

80

Pumping rate (m3/hr)

Figure 10- Pumping pressure vs. pumping rate

- 276 -

100

APPENDIX

Figure A.1 Visual Stability Index

- 277 -

- 278 -

Laboratory Testing and Field Monitoring Of Self-Consolidating

Concrete (SCC) Drilled Shafts
Hani Nassif, Nakin Suksawang, Kagan Aktas, and Husam Najm

Synopsis: The use of self consolidating concrete (SCC) is becoming increasingly

popular in the United States. Several Departments of Transportation are currently
accepting SCC mix design in some of their projects. However, because SCC is a
relatively new material, its mechanical properties and durability are not fully
understood. The main objective of this paper is to evaluate and compare the
mechanical properties of SCC with normal/conventional and high-performance
concrete (HPC) and to monitor the performance of three concrete drilled shafts
fabricated using SCC. Laboratory results show that SCC has a slightly lower
modulus of elasticity compared to normal/conventional concrete but higher tensile
splitting strength. SCC also has higher drying shrinkage and passes a higher
charge (Coulombs when using the rapid chloride permeability test). Results from
three shafts that were instrumented with temperature sensors and cross-hole sonic
logging (CSL) along the shaft length show that the performance of SCC obtained
from the drilled shaft was satisfactory. Segregation due to high slump flow was
observed in the cylinders during testing which adversely affected the strength of
this mix. Cross Seismic Log (CSL) data showed homogenous concrete in the
shaft with no defects and field tests showed that the mix was within the spread
slump test and J-ring test acceptability criteria. It is recommended that the spread
test be used alone in the field for classifying the SCC fresh concrete property but
it should be used in conjunction with other segregation tests, such as the J-ring or
the L-box tests in order to determine the consistency of the mix.

Keywords: Drilled Shaft, Pozzolans, Self Consolidating Concrete, Silica fume,

Slag

- 279 -

Hani Nassif teaches bridge engineering and prestressed concrete at Rutgers, The
State University of New Jersey. He is a Fellow of the American Concrete
Institute (ACI) and member of its technical activities committee (TAC). His
research area is in structural concrete with emphasis on using SCC and HPC for
infrastructure applications. He holds a BS and MS from the Uuniversity of
Detroit, and Ph.D. from The University of Michigan-Ann Arbor.
Nakin Suksawang is an Assistant Professor at Florida International University,
Miami, Florida. His research area is in HPC, SCC, and cementitious materials.
He holds his BS, MS, and Ph.D. from Rutgers, the State University of New
Jersey.
Kagan Ates holds an MS Degree from Rutgers, The State University of New
Jersey and has performed research work on Restrained Shrinkage of HPC.
Husam Najm is an Associate Professor at Rutgers, The State University of New
Jersey. His area of research is fiber reinforced concrete, SCC, and advanced
materials.

INTRODUCTION
Self-consolidating concrete (SCC) is defined as concrete with high viscosity that
can flow freely without segregation. SCC is able to flow under its own selfweight into corners of formwork and through closely spaced reinforcement with
little or no vibration or compaction. This leads to lower energy cost, lower stress
on the formwork, reduced labor cost, and elimination of the potential human error
in consolidation of the concrete. The concrete becomes more consistent since
there are equal dispersions of the cementitious paste and aggregates. As a result,
both SCC mechanical properties and durability are improved over
normal/conventional concrete, which has increased its popularity in Europe and
Japan (1-5). However, the use of SCC in the United State (US) remains limited.
Part of the reason is because of the limited knowledge and experience in using it,
as well as its high initial cost. Nevertheless, there is an effort led by the Federal
Highway Administration (FHWA) to promote the use of SCC in transportation
structures in the US (6).
Fresh concrete can easily attain high flowability by simply increasing the waterto-binder (w/b) ratio. However, increasing the w/b ratio alone could lead to
concrete segregation and less durability. Thus, in order to successfully develop

- 280 -

SCC, mineral and chemical admixtures, e.g., pozzolans, lime stone filler,
superplasticizer, and viscosity-modifying admixture (VMA), need to be added to
the mix design to prevent segregation and enhance the durability of SCC. In
addition, the absolute volume of coarse aggregates also needs to be limited to
reduce inter-particle friction and allow the SCC to flow under its self-weight
without segregation (7-11). Also a reduction in the volume of coarse aggregate
would need to be balanced by an increase in the volume of cement paste, which
would result in higher material cost and an increase in the capillary pores. One
solution to decrease the paste volume is to use VMA that reduces interparticle
friction and increases flowability, and therefore, the volume of coarse aggregates
could be increased. Despite this solution, concrete producers and owners still
have questions on the use of VMA because there is little information on its longterm effects and the effects of various chemical admixtures on SCC. Therefore,
there is a need for research to study the effects of various chemical admixtures on
SCC and evaluate and compare alternative solutions.
One of the alternative solutions is to use Pozzolans, such as fly ash or dust
powder, to replace the cement content (12-14). The pozzolanic materials do not
only reduce the cement content, but also fill the capillary pores, which make the
concrete denser, thus increasing the concrete durability. Moreover, some of the
Pozzolans, such as fly ash and slag, can also increase the flowability of concrete,
which results in reducing the amount of superplasticizers and lowering production
cost. However, any decrease in the volume of coarse aggregate and any increase
in the volume of cementitious paste will greatly affect the mechanical properties
of SCC. Thus, in order for SCC to become a standard concrete mixture, the
mechanical properties need to be evaluated to ensure a level of comfort for the
bridge designers and owners.
There are different methods to produce SCC and we need to evaluate these
methods for its proper use in construction projects, specifically for drilled shafts.
Before SCC can be incorporated in New Jersey Department of Transportation
(NJDOT) projects, mix designs and specifications need to be developed. These
items should reflect the state-of-art in design of SCC while incorporating factors
specific to New Jersey and the availability of SCC producers and need in various
projects. Factors to be considered in developing specifications include: aggregate
gradation requirements, maximum volume of coarse aggregates, appropriate filler
materials and other admixtures, flow and segregation testing methods, durability,
and material handling and placement requirements.
The main objective of this paper is to provide results from a laboratory performed
to compare the mechanical properties of SCC, HPC and NC and to construct and
monitor the performance of three concrete drilled shafts that was cast using SCC

- 281 -

(15). Three out of five shafts were instrumented with strain and temperature gages
to monitor the behavior of SCC during hydration and in curing phases. Fresh
concrete testing was also performed to evaluate existing testing procedures and
identify those that are reliable and easily implementable under field conditions.
Moreover, concrete cylinder specimens were taken to verify the design strength
requirements. Finally cross-hole sonic logging tests were performed to test the
continuity and integrity of the drilled shaft profile and cross section.
EXPERIMENTAL PROGRAM
This study consisted of a laboratory testing program of SCC, HPC and NC
specimens as well as constructing drilled shafts with SCC on I-280 Interchange
Project located in Newark, New Jersey.
Laboratory Tests
The laboratory tests included compressive, tensile, elastic modulus, drying
shrinkage, and rapid chloride permeability. Table 1 shows the various mixes and
the results of compressive, tensile, and elastic modulus tests.
Figure 1 illustrates the drying shrinkage of normal/conventional concrete, SCC,
and HPC mixes. It is observed that the HPC mixes had the lowest drying
shrinkage, whereas the SCC mixes had the highest drying shrinkage. The reason
for this is because the SCC mixes had higher cement paste and higher capillary
pores that lead to higher drying shrinkage. It is also observed that the addition of
pozzolans to SCC reduced its drying shrinkage. Among all the SCC mixes
evaluated in this study, the mix containing both SF and F has the lowest drying
shrinkage with a 20% reduction in comparison with the SCC1 or regular SCC
mix. Mix SCC2 (or SCC mix containing 20%F) also had a 10% reduction,
whereas, mix SCC3 (or SCC containing 30%SL) did not have a significant
reduction.
Figure 2 shows the rapid chloride permeability test (RCPT) results of all mixes. It
is observed that there is a significant variation in the RCPT charge passing rates
among normal/conventional concrete, SCC, and HPC. The SCC mixes had
significantly higher charge-passed values at all testing days. This can be
attributed to the increase in paste volume that leads to higher capillary voids. The
added pozzolanic materials (with the exception of SL) also had a significant effect
on the RCPT of SCC.
Also, Figure 2 shows that SCC with and SF Pozzolans had about 30% reduction
in the charge-passing rate compared to the regular SCC. The combination of SF

- 282 -

and F seems to have the most influence on the permeability with a 57, 74, and
81% reduction at 28, 56, and 90 days, respectively, over regular SCC. The added
SF also reduces the permeability of SCC containing F by approximately 50%.
The reason for this is that SF is highly reactive and has a high surface area,
allowing it to fill the capillary pores and making the concrete denser and less
permeable. In addition, as expected, both HPC mixes outperform all other mixes
in RCPT test.
Field Instrumentation and Testing
The field work started in November of 2006 and the drilled shaft construction was
completed in mid 2007. The authors worked with the contractor and the NJDOT
resident engineer on coordinating the instrumentation and collection of field
samples during the construction of the drilled shafts. Two SCC Shafts from the
project and a demonstration shaft were instrumented with vibrating wire strain
gages to monitor load effects and temperature. Results from drilled shaft testing
(including the standard cross-holes sonic logging (CSL) performed by the
Contractor) were used to evaluate shaft integrity similar to the other shafts of the
project.
The drilled shaft was instrumented with five Geokon Model 4200 Concrete
Embedment Gages to monitor the temperature and strains during and after
construction. In addition to measuring the concrete strain, the gage is also
equipped with a thermistor to measure the temperature. Table 2 and Figure 3
illustrate the sensor locations inside the drilled shaft. Figure 4 shows the concrete
embedment gage attached to the cage of the drilled shafts. The data acquisition
system was used to collect data at 5-minutes interval from all sensors starting
from the time of pour.
FRESH CONCRETE PROPERTIES
Various fresh concrete tests were performed on the SCC sampled from the truck
(Figure 5). Due to space limitations at the location of the drilled shaft concrete
pours, not all tests were performed for each truck. However, the slump flow test
was performed for every truck arriving at the site.
CONCRETE FIELD SAMPLING AND LABORATORY TESTING
Twenty 4 x 8 inch and three 6 x 12 inch concrete cylinders were taken at the day
of the concrete pour from the second of five trucks that were assigned for the job.
The specimens taken were to be used to test the compressive strength as well as
the modulus of elasticity of the mix at different ages. The 6 x 12 cylinders were
also used to confirm the results from the other cylinders at 28 days.

- 283 -

RESULTS
Fresh Concrete Properties
Table 3 and 4 summarize the results obtained from the spread, j-ring, and l-box
tests for drilled shafts #3 and #1, respectively. For drilled shaft #3, the value of
the test from the first truck was higher compared to the other trucks. According to
the results of the J-Ring test the concrete from the first truck will have minimal to
noticeable blocking and the concrete from Truck 3 will have no visible blocking.
L Box test performed on the concrete from Truck 4 indicated that blocking may
be experienced in the concrete from the 4th truck since h2/h1 was observed to be
less than 0.9.
For drilled shafts #1, the slump flow values from both trucks varied between 19
and 21.5 inches, which is less than the limit of 24 inches specified by NJDOT
Specifications. The JRing test performed on the second truck (measured flow is
21.5 inch.) indicated that blocking should be observed in the concrete from this
truck. The LBox readings correlated well with the J-Ring test results showing
that blocking might be obtained due to the h2/h1 ratio of 0.632, which is lower
than the LBox test limit 0.9.
MECHANICAL PROPERTIES
Figure 6 shows a comparison of mixes obtained from the field including the
demonstration shaft. It is clear that concrete from drilled shafts 1 and 3, both
having lower slump flows, outperformed the concrete from the demonstration
shaft. Figures 7 and 8 compare the tensile splitting and modulus of elasticity of
drilled shafts, respectively. Both shafts have comparable mechanical properties,
with the exception of the modulus of elasticity where shafts 1 was stiffer. This
could be attributed to the lower slump flow in shaft 1.
STRAIN AND TEMPERATURE PROFILE
Figures 9 and 10 show the change in strains and temperature from the time of
pour (December 11, 2006) to 28 days (10:25 am January 8, 2006), respectively.
Sensors 3 and 4 recorded temperature values around 63 0F. This is expected since
these sensors are in the middle of the shaft. Dissipation of the heat of hydration is
therefore much slower than the concrete near the top surface (which is exposed to
ambient temperatures) and the concrete that is close to bottom (which dissipates
heat through contact with the surrounding soil). The initial temperature recorded
by the gages is the temperature of the concrete at the time of arrival of the truck to

- 284 -

the site. This value is around 75 0F for all sensors. Peak temperature of Sensors 1
and 2 are 106 0F and 94 0F, respectively. For Sensors 3 and 4 the values were
126.5 0F and 128 0F, respectively, while the value for Sensor 5 was 111 0F.
Strains in all sensors are in compression indicating that the concrete is shrinking
freely.
Cross-Hole Sonic Logging Evaluation of Drilled Shafts
The project Contractor performed Cross-hole sonic logging evaluation on each
drilled shaft including the demonstration shaft. The results showed no (major) air
pockets or a discontinuity in the integrity of any of the shafts. This means that the
SCC mix used was successful in passing through the dense reinforcement layers
in the 6 feet diameter shafts. Appendix A shows typical results from the analysis
of the CSL of the Drilled Shafts in I-280 Interchange.
RECOMMENDATIONS AND CONCLUSIONS
The following conclusions and recommendations could be made from this study
1. Performance of the SCC obtained from the drilled shafts was found
satisfactory. The only problem that was encountered was during the pour of
the demonstration shaft. The slump flow measured was over 28 inches and
segregation was observed in the cylinders during testing which adversely
affected the strength of this mix. This problem was overcome by reducing the
slump flow requirement to in between 20 inches and 24 inches. Therefore, it
is suggested to keep the slump flow in SCC applications within these limits to
prevent segregation and strength loss.
2. SCC in drilled shaft should not be dropped from higher distances. In this
project, dropping SCC into the drilled shafts from heights more than 6 ft,
resulted in segregation. For higher drops, a tremie should be used to avoid
segregation.
3. ASTM Standard J-Ring Test (C1621/C1621M-06 Standard Test Method for
Passing Ability of Self-Consolidating Concrete by J-Ring) was successfully
used to test the passing ability of SCC. It is suggested that this test be used as
a part of quality control measure along with the regular slump flow test when
a more qualitative result is needed. This is especially true for mixes with
superplasticizer only and high coarse aggregate content where the slump flow
may pass the slum flow test but may segregated when passing through the Jring. Visual rating could also be used in lieu of the J-ring test but required
trained eyes.

- 285 -

4. If properly made, SCC can be used in all transportation structural members as

long as there is no slope. Thus, it is not recommended for casting bridge deck.
More quality controls are needed for cast-in place application to monitor the
segregation. It is recommended that a demo SCC member be made on site and
extract to see the aggregate distribution of the mix before allowing the
contractor use SCC.
ACKNOWLEDGEMENTS
This study was funded by the New Jersey Department of Transportation (NJDOT)
and the Center for Advanced Infrastructure and Transportation (CAIT) at Rutgers,
The State University of New Jersey. The findings expressed in this article are
those of the authors and do not necessarily reflect the view of NJDOT. The
financial support and the technical assistance of NJDOT staff, as well as the
assistance of students Jim Ordija and Amer Mohammed, is gratefully
acknowledged.
REFERENCES
1. Persson, B., A Comparison Between Mechanical Properties of SelfCompacting Concrete and the Corresponding Properties of Normal Concrete,
Cement And Concrete Research, Vol. 31, No.2, 2001, pp. 193-198.
2. Zhu, W., P. Bartos, Permeation Properties of Self-Compacting Concrete,
Cement and Concrete Research, Vol. 33, No. 6, 2003, pp. 921-926.
3. Persson, B., Internal Frost Resistance and Salt Frost Scaling Of SelfCompacting Concrete, Cement And Concrete Research, Vol. 33, No. 3, 2003, pp.
373 379.
4. Okamura, H., Self-Compacting High-Performance Concrete, Concrete
International, Vol. 19, No. 7, 1997, pp.50-54.
5. Trandifilou, L.N., Implementation of High-Performance Materials: When Will
They Become Standard, Proceeding of the 6th International Bridge Engineering
Conference, TRB, 2005, pp. 33-48 (on CD).
6. Su, N., K.C. Hsu, H.W. Chai, A Simple Mix Design Method for SelfCompacting Concrete, Cement And Concrete Research, Vol. 31, No. 12, 2001,
pp. 1799-1807.

- 286 -

7. Saak, A.W., H.M. Jennings, S.P. Shah, New Methodology for Designing SelfCompacting Concrete, ACI Materials Journal, 98 (6), 2001, pp. 429 439.
8. Sedran, T., F. De Larrard, Optimizing of Self-Compacting Concrete,
Proceedings Of The 1st RILEM Symposium on Self-Compacting Concrete, 1999,
pp. 321 332.
9. Edamatsu, Y., N. Nishida, A Rational Mix-Design Method for SelfCompacting Concrete Considering Interaction Between Coarse Aggregates And
Mortar Particles, Proceedings of The 1st RILEM Symposium on Self-Compacting
Concrete, 1999, pp. 309 320.
10. Jacobs, F., F. Hunkeler, Design of Self-Compacting Concrete For Durable
Concrete Structures, Proceedings of The 1st RILEM Symposium on SelfCompacting Concrete, 1999, pp. 397 407.
11. Xie, Y., B. Liu, J. Yin, S. Zhou, Optimum Mix Parameters of High-Strength
Self-Compacting Concrete with Ultrapulverized Fly Ash, Cement And Concrete
Research, Vol. 32, No. 3, 2002, pp. 477-480.
12. Bouzoubaa, N., M. Lachemi, Self-Compacting Concrete Incorporating High
Volumes of Class F Fly Ash: Preliminary Results, Cement and Concrete
Research, Vol. 31, No. 3, 2001, pp. 413-420.
13. Ho, D.W.S., A.M.M. Sheinn, C.C. Ng, C.T. Tam, The use of Quarry Dust for
SCC Applications, Cement and Concrete Research, Vol. 32, No. 4, 2003, pp. 505
511.
14. Suksawang, N., and Nassif, H.H., Effect of Modulus of Elasticity on Creep
Prediction of High Strength Concrete Containing Pozzolans, Shrinkage and Creep
of Concrete, ACI, SP-227, Farmington Hills, M.I., 2005, pp. 261-284.
15. Nassif, H., Aktas, K., Suksawang, N., Najm, H., (2008) "Self-Compacting
Concrete in New Jersey (Phase I & II)," FHWA-NJ-2007-10, Draft Final Report
Submitted to NJDOT, 51 pp.

- 287 -

Table 1- Mix proportions and test results

Raw Material
(kg/m3)
OPC
SF
F
SL
CA
FA
Water
w/b
SP (ml)
AEA (ml)
VMA (ml)
Slump (mm)*
Air (%)
Time
(Days)
1
3
7
14
28
56
90
Time
(Days)
1
3
7
14
28
56
90
Time
(Days)
1
3
7
14
28
56
90

Mixtures Identification
NC
SCC1 SCC2
477
515
412
-----103
---1022
817
817
701
885
885
186
201
201
0.39
0.39
0.39
2339
5647
3850
94
51
77
---127
699
629
6.00
6.50
5.75

SCC3
386
26
103
-817
885
201
0.39
4620
51
-635
5.00

SCC4
360
--154
817
885
201
0.39
2631
51
-648
6.00

SCC5
335
--85
860
801
148
0.35
3094
445
774
635
6.00

HPC1
382
-95
-1022
701
186
0.39
1476
498
-83
5.00

HPC2
358
24
95
-1022
701
186
0.39
2214
498
-79
5.00

Compressive Strength (MPa)

31.4
28.5
29.5
20.9
38.4
36.0
40.1
29.5
40.8
40.6
47.8
34.6
44.5
45.6
47.9
37.5
50.1
49.4
51.8
40.5
56.2
53.8
57.9
48.3
55.8
56.3
61.8
49.2

19.8
28.9
36.0
40.8
42.4
45.2
43.7

32.6
43.5
48.2
60.6
60.0
62.1
61.8

19.4
29.3
36.0
39.0
44.2
51.0
56.1

22.3
29.3
32.9
41.1
46.0
49.6
53.6

Modulus of Elasticity (MPa)

30980 32723 26916 20565
33528 32634 29871 27193
34218 30192 33299 26124
34343 31547 32802 30910
35299 33824 33647 32210
35954 34716 36077 32649
36100 35540 40998 30856

22848
25407
27303
30395
33336
33636
34888

18332
22572
20893
26146
31271
31018
29489

25209
26078
29556
29035
31637
34656
38322

23722
24013
25177
26716
27719
28871
30270

3.7
3.9
4.2
4.4
5.3
5.4
5.5

N/A
N/A
N/A
N/A
N/A
N/A
N/A

N/A
N/A
N/A
N/A
N/A
N/A
N/A

Tensile Splitting Strength (MPa)

3.7
3.7
3.6
2.6
2.9
4.1
3.9
4.2
3.0
3.2
4.3
4.4
4.7
3.5
3.6
4.3
4.6
4.8
3.9
3.8
4.3
4.7
5.0
4.0
4.0
4.6
5.1
5.9
4.5
4.2
4.8
5.5
5.5
4.6
4.5

- 288 -

Table 2- Sensor locations

Sensor
ID
Sensor 1
Sensor 2
Sensor 3
Sensor 4
Sensor 5

Distance from Top

(ft)
0.25
1
16.75
16.75
32.5

Table 3- Fresh concrete tests performed on concrete for drilled shaft #3

Truck
No.

Slump
Flow
(in)

J-Ring
Flow
(in x in)

1
2
3
4

28.5
24.5
18.5
19.5

27
18.25
-

Difference
from
Slump
Flow (in)
1.5
0.25
-

L Box
h2 (in)

h1 (in)

h2/h1

3.25

4.75

0.684

Table 4- Fresh concrete tests performed on concrete for drilled shaft #1

Truck

Slump
Flow
(in)

J-Ring
Flow
(in x in)

1
2

19
21.5

21.5

Difference
from
Slump
Flow (in)
0

- 289 -

L Box
h2 (in)

h1 (in)

h2/h1

4.75

0.632

Drying Shrinkage (10 mm/mm)

200
0
-200
-400
NC: w/b=0.39, 100%OPC
SCC1: w/b=0.39, 100%OPC
SCC2: w/b=0.39, 20%F
SCC3: w/b=0.39, 5%SF 20%F
SCC4: w/b=0.39, 30%SL
SCC5: w/b=0.35, 20%SL
HPC1: w/b=0.39, 20%F
HPC2: w/b=0.39, 5%SF 20%F

-600
-800
-1000
1

10
Time (Days)

Figure 1- Results from the drying shrinkage tests

- 290 -

100

- 291 -

6000

4000

HPC2: w/b=0.39
5%SF 20%F

7000

HPC1: w/b=0.39
20%F

SCC5: w/b=0.35
20%SL

SCC4: w/b=0.39
30%SL

SCC3: w/b=0.39
5%SF 20%F

SCC2: w/b=0.39
20%F

SCC1: w/b=0.39
100%OPC

NC: w/b=0.39
100%OPC

Charge Passed (Coulombs)

8000

28 Days
56 Days
90 Days

5000

NJDOT RCPT
Specified Passing
Rate

3000

2000

1000

Figure 2- Results from rapid chloride permeability test on various mixes

Sensor 1
Sensor 2

Sensor 4

Sensor 3

Sensor 5

Figure 3- Sensor locations (not to scale)

Figure 4- Model 4200 concrete embedment gage

- 292 -

(a) Spread test

(b) J-Ring Test

(c) L-Box Test

Figure 5- Fresh concrete tests

- 293 -

69

8000

55

6000

41

4000

28

2000
0

Demo Shaft
Shaft #3
Shaft #1
0

10

20
30
40
Time (Hours)

50

14

Compressive Strength (MPa)

Compressive Strength (psi)

10000

0
60

800

600

400

200

1
Shaft #3
Shaft #1

10

20
30
40
Time (Hours)

50

Splitting Tensile Strength (MPa)

Splitting Tensile Strength (psi)

Figure 6- Comparison of compressive strength of field mixes

0
60

Figure 7- Comparison of splitting tensile strength of field mixes

- 294 -

41

5000

34

4000

28

3000

21
Shaft #3
Shaft #1

2000

10

20
30
40
Time (Days)

Modulus of Elasticity (GPa)

Modulus of Elasticity (ksi)

6000

14
60

50

Figure 8- Comparison of modulus elasticity of field mixes

100
Sensor 1
Sensor 2
Sensor 3

Strain (

Sensor 4
Sensor 5

-100
January 8, 2006
-200

-300
December 11, 2006
-400

10
15
20
Time (Days)

25

Figure 9- Strain profile of the demonstration shaft

- 295 -

30

140
Sensor 1
Sensor 2
Sensor 3
Sensor 4
Sensor 5

Temperature (F )

120
100
80
60
40
December 11, 2006
20

January 8, 2006

10
15
20
25
30
Time (Days)
Figure 10- Temperature profile of the demonstration shaft

- 296 -

IN-SITU PROPERTIES OF SELF-CONSOLIDATING

CONCRETE IN A THICK CONCRETE RAFT IN DUBAI.
James Aldred

Synopsis: The use of self-consolidating concrete (SCC) is relatively new in the

Middle East where the initial applications began in the early 2000s, primarily in
piles and raft slabs. There is limited published information on the in-situ
properties of SCC used in the region. This paper presents information on the fresh
and hardened properties of SCC incorporating 40% fly ash replacement and 7%
silica fume used for a thick raft slab as well as mechanical and penetrability data
from cores taken after construction. The results confirm that SCC can provide
excellent in-situ properties which would be expected to achieve demanding
structural and durability requirements.

Keywords: Self-consolidating concrete, in-situ, core, durabilty

- 297 -

James Aldred, Ph.D, CPEng, LEED AP, FACI has over 25 years in the concrete
industry. Dr Aldred is Principal Engineer with GHD Pty Ltd and Manager of the
Independent Verification and Testing Authority for the Burj Dubai. James is a
member of ACI 130, 201, 212, 233, 234, 305 and 365 and authored over 40
technical papers on various aspects of concrete technology.

INTRODUCTION
According to the Council on Tall Buildings and the Urban Habitat by October
2007, there were over 55 super-tall buildings of 300 metres or greater under
construction, over half of which are in the Middle East and being constructed
primarily on reinforced concrete piles and raft foundations (1). However, the
Middle East is not a benign environment for concrete and great care and attention
is necessary to achieve durable concrete in this region. The flowing properties of
self-consolidating concrete (SCC) have been used to minimize the potential for
honeycombing in these elements, particularly in situations where there is a highly
congested reinforcing steel.
One concern regarding the use of SCC has been the requirement for a relatively
high binder and admixture content to achieve the required combination of
flowability and segregation resistance. In the case of foundation elements in the
Middle East, the durability requirements of the specification would normally
require a low water/cementitious ratio and a high binder content, often with
significant replacement with supplementary cementitious materials. Therefore
modification of aggregate grading and the use of suitable viscosity modifying
agent may be the only changes to achieve self-consolidating characteristics at
minimal additional cost. The fact that SCC would greatly reduce the size of the
concrete placing team means that SCC could provide significant cost savings in
these applications.
While there have been numerous investigations on laboratory and mock-up scale
specimens, there is little published information about the in-situ properties of SCC
from actual construction, particularly with respect to penetrability and durability
parameters. This paper presents the data from cores taken from a large raft after
construction.
SPECIFICATION REQUIREMENTS
The required compressive strength for the raft concrete was 50 N/mm2. The
penetrability requirements were 2.0% water absorption in accordance with BS
1881: Pt 122, 1000 coulombs in accordance with ASTM C1202 and 15 mm
penetration in accordance with BS EN 12504-1. The temperature limit was 710C
and the specification called for the use of SCC in the raft with a minimum
slumpflow of 600 mm.

- 298 -

SCC MIXTURE PROPORTIONS

Cement (MSRPC)
Fly Ash
Silica fume
Aggregate (20 mm)
Aggregate (10 mm)
Washed crushed sand (0-5 mm)
Dune Sand (0 0.6 mm)
Fine Aggregate Component
Polycarboxylate SP
VMA
w/cm

252 kg/m3
168 kg/m3
30 kg/m3
368 kg/m3
368 kg/m3
552 kg/m3
552 kg/m3
60%
4 - 6 L/m3
1 - 1.5 L/m3
0.32

PROPERTIES OF THE TRIAL MIX

The initial slumpflow of the trial mix was 750 mm. The concrete achieved a 28
day strength of 71 N/mm2 and the strength development is shown in Figure 1. The
static modulus was 43 GPa at 28 days as seen in Figure 2. The penetrability
parameters were 1.0% water absorption, 7 mm water penetration and 700
coulombs; all well within the requirements.
The slumpflow at delivery for the largest pour of approximately 5000 cubic
metres is shown in Figure 3. A total of 20 trucks were rejected for non-compliant
slumpflow. The average slumpflow was 700 mm with a standard deviation of 26
mm.
The average compressive strength for the compliance cubes was 57.9 N/mm2 at
28 days and 66.5 N/mm2 at 56 days for this pour. The range is shown in Figure 4.
The overall average was 71 N/mm2 at 56 days.
IN-SITU RESULTS
Temperature Rise:
The maximum temperature within the 4 metre thick raft was 66.90C which was
reached approximately 60 hours after placement as seen in Figure 5. The
temperature at the centre of the raft remained at or above 650C for more than 7
days and had not cooled to ambient after more than 80 days demonstrating the
considerable thermal inertia of large concrete rafts.
Mechanical Properties:
The average compressive strength of the cores based on BS EN 12504-1 was 72.6
N/mm2 slightly higher than the 56 day average for the compliance cubes. Figure 6
shows that there was a slight trend of reducing compressive strength towards the
top of the pour but this was less than would have been expected for normal mix

- 299 -

considering usual bleed effects. The compressive strength of the cores tested to
ASTM C39 were only slightly lower and also showed limited strength reduction
towards the top surface (Figure 7).
The indirect tensile or tensile splitting test results showed greater variability as
would be expected but there was also a greater tendency towards reduced tensile
strength for cores taken from higher levels within the raft (Figure 8). Static
modulus values appeared relatively consistent at around 40 GPa until the top 400
mm where the value dropped by approximately 30 percent (Figure 9). This
suggests that there had been a reduction in aggregate content in the uppermost
samples which was apparent from visual examination.
Penetrability Properties:
Figure 10 shows a considerable variation in the measured coulomb values from
the limited number of tests conducted and a trend towards higher values at the top
surface. The negligible values measured below the top surface were profoundly
less than those for the trial mix and , even the highest value obtained from the
core at the surface was still significantly lower than the specification requirement
and the value for the trial mix.
The water absorption showed a similar trend of lower values below the upper
surface. However the surface value was well within the specification requirement.
Cores were also tested for chloride diffusion using a modified NT Build 443
procedure. The concrete in the cores was tested approximately 3 months after
casting and therefore the chloride diffusion coefficient would have reduced by
approximately 30% to 50% compared to the value measured after 28 days. The
original modeling of service life by GHD had been based on a 28 day chloride
diffusion coefficient (Dce) of 3 x 10-12 m2/s and a mature Dce of 0.5 x 10-12 m2/s.
The average Dce for the five core samples tested was 1.54 x 10-12 m2/s which is
approximately the value anticipated by GHD. The variation in Dce with depth was
only 0.92 1.50 x 10-12 m2/s which is within the range expected.
CONCLUSIONS
A wide range of mechanical and penetrability properties were measured on cores
taken from a thick raft cast with SCC in Dubai. The data presented in this paper
demonstrate that the properties of the hardened concrete exceeded the
specification requirements. There was a trend towards reduced performance in the
surface layer but this was still well within the requirements. The reduction in
many of the parameters was probably less than would be expected due to bleed
effects in normally placed concrete.
REFERENCES
1. Oldfield, P. (2007) The Tallest 20 in 2020 Council on Tall Buildings and
Urban Habitat, November, 2007.

- 300 -

80
70
60

Strength MPa

50
40
30
20
10
0
0

10

15

20

25

30

Age - days
Figure 1: Compressive strength development for C50-20 - SCC (Trial mix)

50
45
40

Modulus GPa

35
30
25
20
15
10
5
0
0

10

15

20

25

Age days
Figure 2: Static modulus for C50-20 - SCC (Trial mix)

- 301 -

30

800

Slumpflow (mm)

775
750
725
700
675
650
625
600
0

10

15

20

25

30

35

Hours

Figure 3: Slumpflow during production

Strength MPa

Central area - 28 & 56 day cube results

90.0
80.0
70.0
60.0
50.0
40.0
30.0
28 day Average 57.9

20.0
10.0
0.0

56 day Average 66.5

No of results

Figure 4: Compliance compressive strength data for C50-20 - SCC

- 302 -

80
A
m
b
i
e
n

70
60

40
30
20
10
0

10

20

30

40

Days

50

60

70

80

Date : Time

Figure 5: Temperature profile through the SCC raft during hydration

BS Core tests
0-400 (44)
Depth into raft (no of results)

Temperature (Degree C)

50

4-800 (5)
8-1200 (9)
12-1600 (2)
16-2000 (2)
20-2400 (0)
24-2800 (4)
28-3000 (3)
0

10

20

30

40

50

60

70

80

90

Sterength MPa

Figure 6: Compressive strength of cores (BS EN 12504-1 - l/d = 1.0)

- 303 -

ASTM Core tests

0-400 (6)
Depth into raft (No of results)

4-800 (3)
8-1200 (1)
12-1600 (4)
16-2000 (0)
20-2400 (1)
24-2800 (2)
28-3000 (2)
0

10

20

30

40
50
Strength MPa

60

70

80

Figure 7: Compressive strength of cores (ASTM C39 l/d = 2.0)

Depth into raft (no of rsults)

Shear
0-400 (4)
4-800 (1)
8-1200 (1)
12-1600 (1)
16-2000 (0)
20-2400 (3)
24-2800 (0)
28-3000 (1)
0

1000

2000

3000

4000

5000

Shear kPa

Figure 8: Tensile splitting strength (ASTM C496)

- 304 -

6000

Modulus

Depth into raft (no of results)

0-400 (2)
4-800 (3)
8-1200 (5)
12-1600 (0)
16-2000 (5)
20-2400 (0)
24-2800 (1)
28-3000 (1)
0

5000

10000

15000

20000

25000

30000

35000

40000

45000

Modulus N/mm2

Figure 9: Modulus of elasticity (ASTM C469)

Depth into raft (no of results)

Rapid Chloride Permeability Test

0-400 (3)
4-800 (1)
8-1200 (1)
12-1600 (1)
16-2000 (0)
20-2400 (0)
24-2800 (0)
28-3000 (1)
0

100

200

300

400

500

Permeability - coulomb

Figure 10: Coulomb values (ASTM C 1202)

- 305 -

600

Depth into raft (no of results)

Wate r Absorption
0-400 (4)
4-800 (1)
8-1200 (1)
12-1600 (1)
16-2000 (0)
20-2400 (0)
24-2800 (0)
28-3000 (0)
0

0.2

0.4

0.6

0.8

1.2

Water absorption - %

Figure 11: 30 minute water absorption (BS 1881:Pt 122)

- 306 -

1.4

ENVIRONMENTALLY SUSTAINABLE SELF COMPACTING

CONCRETE A CASE STUDY

Abu Saleh, Mamo Kebede and Nagham Yacoub

Synopsis: Use of Self Compacting Concrete (SCC) instead of conventional

concrete itself is environmentally sustainable due to its reduced energy
requirement to produce and place. This paper explores whether the inherent
carbon footprint could be further reduced by going beyond normal practices
without compromising basic requirements of concrete structures.
Higher fine requirement of SCC has been substituted by using unwashed Crushed
Rock Fine (CRF) Aggregate of limestone aggregate with 10% dust passing
75micron sieve, saving energy and reducing waste, water consumption and
associated works to dispose the waste.
Maximum replacement of portland cement by Ground Granulated Blast Furnace
Slag (GGBS) and Microsilica has been used ranging from 70% to 100% of total
OPC content. Replacement of up to 80% have been used for Fly Ash mixes.
Total elimination of the use of clinker is a possibility.
A proprietary Polycarboxylate base admixture has been used to facilitate the
hydration of cementitious materials.
Plastic and hardened as well as durability properties of SCC have been measured
and verified against a control mix.

Keywords: Fly ash, Ground Granulated Blast Furnace Slag (GGBS), Self
Compacting Concrete, Sustainability

- 307 -

Abu Saleh Mohammod a civil engineer by training is the Technical Manager of

Xtramix Concrete Solutions, UAE. He is having more than 18 years of
professional experience in Asia, Middle East and Europe. He has an MSc degree
in Concrete & Environment from the University of Dundee and currently
pursuing PhD programme with the University of Bath in Concrete and
Environmental sustainability. He is a Corporate Member of the Institute of
Concrete Technology, UK.
Mamo Kebede is the QC Manager of Xtramix Concrete Solution. He is having
more than 11 years experience and is a graduate of BS Soil and Water
Conservation, Construction Material from Mekelle University, Ethiopia and
completed Advanced Diploma in Advanced Concrete Technology (UK). He is a
corporate member of the Institute of Concrete Technology, UK and a member of
Ethiopian Civil Engineers & Architects, American Concrete Institute, and
International Rock Mechanics Society.
Nagham Yacoub is the Plant Manager of two major batching plants of Xtramix
Concrete Solutions. She has graduated from the Jordan University of Science and
Technology in Civil Engineering in 2008 with good grades. She is an important
member of Xtramix Production team.
INTRODUCTION
One of the key environmental benefits Self Compacting Concrete (SCC) offer is
the energy savings. Mixing, pumping and placing of SCC need much lesser
energy than a conventional concrete. Reduced noise pollution, higher durability
and reduced impact on occupational health hazard by eliminating white finger
syndrome are other significant attributes SCC brings into the Health, Safety and
Environmental aspect of the construction industry. Higher durability as well as
higher strength, a default output of SCC, prolongs service life of structures,
providing inherent benefit of ultimate resource optimization. These environmental
friendly attributes makes SCC as one of the unique building materials for
environmentally sustainable construction process.
BACKGROUND
The requirement of higher paste volume to satisfy the rheology of self compacting
concrete is often met by increasing the mass of total cementitious materials or
adding additional inert fillers. Both approaches are adding additional carbon
footprint to the mix denting the environmental credential of SCC.

- 308 -

In UAE the bulk source of both coarse and fine aggregates are from crushed
limestone mainly from Ras Al Khaimah and Fujairah limestone deposits. The 05mm crushed rock fine (CRF) contains a very good source of filler passing below
75micron sieve. Traditionally, large amount of this filler would be removed from
the CRF by washing to satisfy ASTM C33 which allows maximum of 5 to 7
percent passing 75 micron sieve. Most of the local specification would also put a
limit of 5% maximum allowable fine content for CRF. Generally unwashed
CRF would contain 10 % of fine limestone dust which is below 75micron, a
good source of filler material contributing to the aggregate packing improving
strength, durability and the flowability of concrete. All these attributes are
equally important for SCC. An attempt has been made to compare the
performance of all these three criteria of SCC made of washed and unwashed
CRF. Industry would save a significant amount of resources should the
acceptable minimum dust content in CRF be raised to a much higher level such as
10% without compromising basic requirement of concrete characteristics. This
would also add value to its environmental credential as a significant amount of
waste would be reduced and expensive energy would be saved by not requiring of
washing CRF.
Finding novel materials alternative to portland cement to reduce CO2 emission has
become an essential field of research interest to the scientific community. While
the doors of possibilities are wide open, actual available alternatives are limited.
Ground Granulated Blast Furnace Slag (GGBS), Fly ash and Microsilica have
been well proven to be partial alternative to portland cement. In the Middle East
they have been widely used. It is not necessarily due to Environmental point of
view, rather to meet durability requirements of the concrete. Traditionally for
GGBS, portland cement are being replaced from 55 to 70% by mass depending on
the durability requirements. 30% replacement for Fly ash is widely accepted.
Microsilica depending on both durability and strength could range from 4 to 12%
replacement of portland cement. As none of these materials are locally produced
an inherent carbon footprint due to shipment is unavoidable. Reprocessing works
to make them useable materials also add additional carbon footprint.
Nevertheless, considering from current scenario GGBS, Fly ash and Microsilica
are the only viable alternative to portland cement in the Middle East.
Strength development of alternative cementitious materials
One of the inherent problems with the alternative cementitious materials is loss of
strength when using beyond their usual prescribed limits. The chemical reaction
resulting subsequent development of strength enhancing hydrate phases depends
on the amount of portland cement present in the system. Secondly, presence of
Ca(OH)2 is a requirement for any pozzolanic reaction to occur, which is being

- 309 -

reduced with the reduction of the mass of portland cement from the cement paste,
hence loss of strength.
An alkaline activator such as NaOH or waterglass (SiO2.nNa2O.mH2O + NaOH)
with pH higher than 13 can be used to activate slag or Fly ash. The high alkaline
environment of the activator will break the bonds of Ca-O, Si-O and Al-O from
surface of slag creating reactive silicate and aluminate species. When the
concentration of these species reaches the product of solubility of the different
solid compounds, they precipitate creating CSH and CAH phases.[1]
The exact mechanism, however, of the setting of alkali activating slag/Fly ash is
not fully understood. Most proposed mechanism consists of a dissolution,
transportation or poly-condensation. Temperature is also considered as one of the
factors that affects to any polymerization. [2] One of the major issues of alkali
activation is the requirement of higher temperature for the strength development,
which often is not practical for actual application. In this study a special
proprietary admixture from Chryso has been used to boost the hydration process
and enhance the strength gain.
OBJECTIVES
The objective of this case study to explore the strength development and
subsequent durability performance of concrete containing very high replacement
of alternative cementitious materials such as GGBS and Fly ash while
maintaining their self compacting capabilities. The impact of excess filler present
in the unwashed CRF in the physical properties of the SCC is also of interest.
This study will demonstrate the environmental credential of SCC using alternative
cementitious materials and unwashed sand to save energy and reduce waste.
TRIALS
GGBS and Fly ash have been considered to test their maximum allowable limit to
replace portland cement. Microsilica has been added in to both GGBS and Fly ash
mixes to observe the improvement in performance, if any.
Each GGBS and Fly ash series examines the effect of filler content in the CRF
sand by using unwashed and washed 0-5mm sand made of limestone fines.
GGBS Mix:
A mix containing 70% GGBS + 30% portland cement has been chosen to be the
control mix as this proportion is being widely used in the Middle East. The
proportion of GGBS content has been increased by 10% for next 3 mixes

- 310 -

containing 80%, 90% and 100% GGBS, reducing the portland cement content at
the same rate.
A 5% of Microsilica has been added for additional 3 mixes replacing same
amount of GGBS while keeping the portland cement content at the same rate.
This is to observe whether Microsilica could help to boost the strength as
increasing GGBS would naturally reduce the compressive strength.
Both unwashed and washed sands have been used for above two sets of concrete
mix designs creating total 4 sets of concrete to compare their performance with
each other.
Total cementitious content of each mix is 400 kg with 152kg of water providing
w/b of 0.38. This is same for both GGBS and Fly ash series.
The concrete has been designed to be self compacting with having initial flow of
700mm. The expected flow at 90 minutes to be more than 600mm.
Total 14 mixes have been prepared for GGBS series.
Table 1 shows the GGBS containing concrete mix designs.
Fly Ash Mix
A 30% replacement by Fly ash with 70% portland cement is the control mix for
the Fly ash series. Similar to the GGBS mixes, a 10% increment of Fly ash has
been chosen for up to maximum 80% replacement.
Microsilica has been added at a 5% rate replacing same amount of Fly ash to
observe whether there is any significant boost of strength mainly due to particle
packing phenomenon.
Similar to the GGBS series both unwashed and washed CRF have been used,
creating 4 sets of concrete. All concrete have been designed to be self
compacting with similar flow requirement of GGBS mixes. Total 22 numbers of
concrete mixes using Fly Ash have been produced.
Table 2 shows the Fly Ash series concrete mix designs.
Cement and cementitious materials
Among all cementitious materials only portland cements are locally produced.
GGBS are locally ground while classified Fly ash and densified Microsilica are
imported.
The chemical compositions of all cementitious materials are illustrated in Table 3.

- 311 -

Aggregate and Aggregate Grading

A combination of 10mm and 20mm crushed Limestone aggregates have been
used as coarse aggregate together with the combination of limestone CRF and fine
dune sand. CRF is washed and unwashed. Washed CRF contains fines finer than
75micron less than 5% whereas unwashed contains around 10%.
The combined grading of the aggregates containing unwashed CRF and washed
CRF are shown in Figs. 1 and 2 respectively.
Tables 4 10 displays the individual and combined grading of aggregates
including 10mm, 20mm, 5mm (washed and unwashed) and dune sands.
Chemical Admixture
Chryso Fluid Optima 245 EMx a new generation admixture from Chryso for high
volume fly ash and slag concrete has been used in this study. According to Chryso
this new admixture combines their latest proprietary polymers along with most
advanced developments in terms of hydration rate enhancement.
It seems to be a polycarboxylate admixture with pH range around 6.5 with special
capabilities to boost the hydration process for high volume fly ash and slag
concrete. Lower pH suggests that it is not a typical alkali activator.
TESTS
Slump flow and L-box tests have been carried out to check the flowability and
passing-ability of the self compacting concrete.
Flow retention for maximum 90 minutes has been measured taking readings for
every 30 minutes.
Cube compressive strengths for 3, 7 and 28 days have been tested.
Rapid Chloride permeability test (ASTM 1202), water permeability test as per
DIN 1048 and water absorption test as per BS1881: Part 122 have been adopted to
measure the durability of the concrete. These tests have been chosen because they
are widely accepted in the local industry to measure the durability of concrete
subject to harsh environmental exposure.
RESULTS & DISCUSSIONS
Slump flow
GGBS mix: The initial slump flow for all the mixes have been found to be above
700mm. The average admixture dosage range between 1.0 to 1.25% by weight of

- 312 -

cementitious materials. The retention of the flow at 90 minutes at room

temperature is also satisfactory with all having more than 600mm of flow.
Figure 3 & 4 illustrate the slump flow behaviour of GGBS mix for unwashed and
washed CRF respectively. Visually the unwashed mixes demonstrate more robust
mix with little or no bleeding compared with washed CRF mixes. This is obvious
due to higher presence of finer material. Presence of Microsilica has also
positively affected the rheology of the mix as expected. The concrete with GGBS
has been appeared to be stickier than required.
Fly Ash mix: As expected Fly ash mixes demonstrate much better flow properties
in terms of initial and final flow as well as concrete appearance and ease of
handling. The concrete was not sticky like the GGBS series. Admixture dosage
is also lesser which is between 0.625 to 1.0% by weight of total cementitious
content.
Figure 5 & 6 show the slump flow retention of Fly ash mixes.
L-Box test
The passing ability index for Fly ash mixes are far better than GGBS mixes with
no specific influence being observed either by the presence of excess filler of
unwashed sand or Microsilica. In average it is 0.87 and 0.96 for GGBS and Fly
ash respectively.
Table 11 and 12 presents the slump flow retention and L box test passing ability
values for GGBS and Fly ash respectively.
Compressive Strength test
GGBS Mixes: As expected mixes containing higher portland cement gain higher
compressive strength compared to mixes with higher GGBS content. The control
mix with 30% Portland cement + 70% GGBS gains the most while the mix
containing zero portland cement achieved the least.
However the strength gain for mixes without any portland cement is very
remarkable. Mix no. 4 with 100% GGBS achieved 16, 22.5 and 30 MPa at 3, 7
and 28 days respectively, which is quite significant. The drop of strength is
approximately 50% compared to the control mix with 70% GGBS mix.
Mixes containing Microsilica showed similar strength gain behaviour of mixes
without Microsilica, showing no advantages. One explanation could be due to
lower mixing capability of laboratory mixer to evenly break-off the densified
particles of Microsilica to its actual particle sizes. Undensified Microsilica may
provide effective particle packing benefit for lab trials.

- 313 -

Figure 7 & 8 compares the strength gain of GGBS mixes with and without
Microsilica for both unwashed and washed sand respectively.
Figure 9 & 10 compares the effect of sand on the strength gain at 28 days. Mixes
containing unwashed crashed rock fine perform better than mixes containing same
sand but washed to satisfy ASTM C33. The unwashed sand contains
approximately 9-10% fines smaller than 75 micron. Higher compressive strength
at all age levels for both mixes with and without Microsilica have been observed
for mixes containing unwashed sand compared to washed sand mixes.
Table 13 displays the results of compressive strength of GGBS mixes for all ages.
Fly Ash Mixes: Fly ash series produced very similar results like the GGBS mixes.
The control mix with 70% Portland cement + 30% Fly ash is having the best
compressive strength. The strength reduction is almost linear to the rate of
reduction of portland cement from the concrete. However, mix containing 80%
Fly ash with 20% portland cement (unwashed sand) shows a significantly
improved strength of 30 MPa at 28 days. There is no significant impact on
Microsilica additions was found, perhaps due to the same reasons as mentioned
earlier.
Similar to GGBS mix, concrete containing unwashed sand shows better strength
enhancement compare to concrete with washed sand in the Fly ash mixes.
Figs. 11, 12, 13, 14 and Table 14 illustrate and compare the compressive strength
development of Fly ash mixes.
Durability Tests:
Concrete containing Microsilica for both GGBS and Fly Ash mixes were not
tested for durability due to disappointing compressive strength results.
Rapid Chloride Permeability Test: RCPT values as per ASTM 1202 are excellent
for both GGBS and Fly ash mixes especially for GGBS mixes. No significant
differences have been observed among washed and unwashed sand mixes, though
overall unwashed mixes performed better than washed mixes. All values are
within 1000 coulomb indicating excellent chloride resisting capabilities with
concrete containing very high replacement of GGBS or Fly ash. GGBS concrete
containing 100% slag and unwashed sand is having RCPT value of 257 coulombs
which is the lowest. Most of the RCPT values for GGBS mixes are below 500
coulombs. For Fly ash mixes, the RCPT values hover around 600 coulombs.
Water Permeability Test: Water permeability test according to DIN 1048 method
shows no sign of water penetration for Fly Ash mixes. No graph could be plotted
due to zero penetration of water through the specimen. For GGBS mixes except
the two control mixes (30% Portland cement + 70% GGBS, washed and

- 314 -

unwashed), all other results are below 10mm, indicating excellent water-tightness
of the concrete. There is no significance of the influence of unwashed/washed
sand have been observed.
Water Absorption (BS 1881: Part 122): Fly ash mixes are having better water
absorption values mostly less than 1% compared to GGBS mixes which stay
between 1% and 1.2%. Like Permeability test, there is no influence of the
presence of filler provided by unwashed sand.
In general it has been seen that higher replacement of portland cement gives better
durability values with no significance in durability if unwashed sand is used.
Tables 15 & 16 and Figs. 15, 16, 17, 18 & 19 present the durability values of both
GGBS and Fly ash mixes.
ENVIRONMENTAL BENEFITS
Results from the concrete trials using high volume GGBS and Fly ash show that
structural grade concrete where very high strength is not a requirement can be
produced readily with no or very little requirement of ordinary portland cement.
Enhanced durability performance would effectively increase the service life of
structures providing opportunities to reduce the demand of raw materials.
Producing better concrete in terms of strength and durability using unwashed
crushed sand raises questions on the validity of the requirement of ASTM C33 on
the limitation on maximum fines. A huge amount of energy could be saved and
fewer or no waste would be generated as the filler materials would be turned into
useful concrete ingredient.
Use of SCC made of higher alternative cementitious materials and unwashed CRF
as sand would reduce the net carbon footprint of projects.
FUTURE WORKS
The strength gain of concrete without or minimum portland cement can be further
enhanced should the w/c can be reduced further for a given consistency. Concrete
with 0.34 w/c or less need to be investigated to produce concrete with higher
grade.
The scope of physical tests should also be widened. Tensile and flexural strength,
E-modulous and shrinkage of concrete should be investigated.

- 315 -

Actual chloride diffusion tests beside indirect test (RCPT) should be included.
Change in length in sulphate solution and other aggressive elements such as sea
water should also be investigated.
Effect of Microsilica should be re-examined by incorporating undesified
Microsilica for lab mixes. Other pozzolanic materials such as reactive rice husk
ash could be of interest.
Effect of recycled aggregate with high volume alternative slag or other pozzolanic
materials should also be investigated.
Actual carbon footprint needs to be calculated to measure the net gain in terms of
the environmental sustainability.
CONCLUSIONS
Two sets of self compacting concrete trials have been conducted using GGBS and
Fly ash as replacement of portland cement for each. For GGBS mixes the
maximum replacement was 100% and minimum 70% and for Fly ash mixes it
was 80% and 30% respectively. A new proprietary admixture from Chryso has
been used to boost the hydration of the cementitious materials.
Two types of CRF sand have been used for each set. Washed CRF to remove
excessive fine materials to meet ASTM C33 requirement with less than 5% fines
and unwashed limestone CRF with approximately 10% fines smaller than 75
micron.
Microsilica has been added additionally to boost the strength gain, which was not
materialized perhaps due to the usage of densified Microsilica.
Self compacting concrete with an average flow of more than 700mm and L-Box
passing ability index of more than 0.90 have been achieved even without any
presence of portland cement.
Compressive strength of 30 MPa has been achieved for concretes with 100%
GGBS + 0% Portland cement and 80% Fly ash + 20% Portland cement mixes.
Durability test results of RCPT (ASTM 1202), Water Permeability (DIN 1048)
and Water Absorption (BS1881:Part 122) positively indicate that concrete
containing little or no portland cement with higher GGBS or Fly Ash provide
better durability performance.
It is possible to produce structurally sound self compacting concrete without using
portland cement. The compressive strength achieved is 30MPa with much
improved durability performance without removing important additional filler

- 316 -

from the crushed fine aggregates. This enhances the environmental credential of
these concrete tremendously.
REFERENCES
1.

Puertas F, Fernandez-Jimenze A and Blanco-Varela M.T, Pore solution in

alkali-activated slag cement paste. Relation to the composition and structure
of calcium silicate hydrate, Cement and Concrete Research, 34 (2004) pp.
139 148

2.

Songpiriyakij S, Alkali activated Fly ash concrete, King Mongkuts

Institute of Technology, North Bangkok, Thailand

- 317 -

Table 1: Concrete Mix Designs: GGBS series

Mix
No.

Concrete Grade

Cementitious
Materials, kg/m3
OPC GGBS MS

1
2
3
4
5
6
7
8
9
10
11
12
13
14

UW(30%OPC+
70%GGBS)
UW(20%OPC+
80%GGBS)
UW(10%OPC+
90%GGBS)
UW(0%OPC+
100%GGBS)
UW(20%OPC+
75%GGBS+5%MS)
UW(10%OPC+
85%GGBS+5%MS)
UW(95%GGBS+
5%SF)
W(30%OPC+
70%GGBS)
W(20%OPC+
80%GGBS)
W(10%OPC+
90%GGBS)
W(0%OPC+
100%GGBS)
W(20%OPC+
75%GGBS+5%MS)
W(10%OPC+
85%GGBS+5%MS)
W(95%GGBS+5%
MS

Coarse
Fine
Admixture
Aggregate, Aggregate, Water
(Chryso)
kg/m3
kg/m3
20
5 Dune
Dosage,
10 mm
kg/m3
mm
mm Sand
L/m3

120

280

241

565

530

530

152

6.0

80

320

240

563

528

528

152

5.0

40

360

239

562

526

526

152

5.0

400

239

560

525

525

152

4.5

80

300

20

239

561

526

526

152

4.0

40

340

20

238

560

525

525

152

4.0

380

20

238

559

524

524

152

4.8

120

280

241

565

530

530

152

4.2

80

320

240

563

528

528

152

3.9

40

360

239

562

526

526

152

3.0

400

239

560

525

525

152

3.8

80

300

20

239

561

526

526

152

4.0

40

340

20

238

560

525

525

152

380

20

238

559

524

524

152

4.5

Note: UW= Unwashed sand ; W = Washed sand

- 318 -

Table 2: Concrete Mix Designs: Fly Ash Series

Coarse
Fine
Cementitious
Admixture
3 Aggregate, Aggregate, Water
Materials, kg/m
(Chryso )
Mix
kg/m3
kg/m3
Concrete Grade
No.
20
10
5 Dune
Dosage,
OPC FA MS
kg/m3
mm mm mm* Sand
L/m3
1 UW(70%OPC+ 30%FA) 280 120 0 239 563 527 527 152
4.0
2

UW(60%OPC+ 40%FA)

240

160

238

559

524

524

152

4.0

UW(50%OPC+ 50%FA)

200

200

236

555

520

520

152

2.72

UW(40%OPC+ 60%FA)

160

240

235

551

517

517

152

3.2

UW(30%OPC+ 70%FA)

120

280

233

547

513

513

152

2.5

UW(20%OPC+ 80%FA)
UW(20%OPC+
7
75%FA+5%MS)
UW(30%OPC+
8
65%FA+5%MS)
UW(40%OPC+
9
55%FA+5%MS)
UW(50%OPC+
10
45%FA+5%MS)
UW(60%OPC+
11
35%FA+5%MS)
12 W(70%OPC+ 30%FA)

80

320

231

543

509

509

152

2.5

80

300

20

231

543

509

509

152

2.8

120

260

20

233

547

513

513

152

3.1

160

220

20

234

551

516

516

152

3.4

200

180

20

236

555

520

520

152

3.6

240

140

20

238

558

523

523

152

3.8

280

120

239

563

527

527

152

3.5

13 W(60%OPC+ 40%FA)

240

160

238

559

524

524

152

3.3

14 W(50%OPC+ 50%FA)

200

200

236

555

520

520

152

2.8

15 W(40%OPC+ 60%FA)

160

240

235

551

517

517

152

2.1

16 W(30%OPC+ 70%FA)

120

280

233

547

513

513

152

1.8

17 W(20%OPC+ 80%FA)
W(20%OPC+
18
75%FA+5%MS)
W(30%OPC+
19
65%FA+5%MS)
W(40%OPC+
20
55%FA+5%MS)
W(50%OPC+
21
45%FA+5%MS)
W(60%OPC+
22
35%FA+5%MS)

80

320

231

543

509

509

152

1.8

80

300

20

231

543

509

509

152

2.5

120

260

20

233

547

513

513

152

160

220

20

234

551

516

516

152

200

180

20

236

555

520

520

152

240

140

20

238

558

523

523

152

4.6

- 319 -

Table 3: Chemical Composition of cementitious materials used

Items
Silicon Dioxide
Aluminum Trioxide
Ferric Oxide
Calcium Oxide
Magnesium Oxide
Sulphate
Potassium Oxide
Sodium Oxide
Titanium Oxide
Manganese Oxide
Chloride
Insoluble Residue
Loss On Ignition
Tricalcium
Aluminates
Alkalis Equivalent
Specific surface area
Moisture

SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
K2O
Na2O
TiO2
Mn2O3
Cl
IR
LOI
C3 A
AE
m2/kg
H2O

Portland
Cement
20.12
4.84
3.53
63.44
1.36
2.95
0.51
0.20

0.02
0.48
2.84
6.86
0.54
334

Fly
Ash
57.66
30.35
4.92
1.71
0.70
0.16
1.14
0.01
1.64
0.04
0.01
0.63
0.97

GGBS
31.6
16.58
1.28
37.20
10.34
0.24
0.35
0.24
0.48
0.01
0.60
1.0

Microsilica
90.53

2.38

0.76
444
0.3

0.03

0.87

Table 4: 20mm aggregate grading

Sieve Size,
(mm)
Wt. Retained,
(g)
Cum. Wt., (g)
% Retained
% Passing

37.50

20.00

14.00

10.00

5.00

0.075

Pan

419.4

2026.6

1200.5

105.1

6.9

1.5

419.4
11.2
88.8

2446
65.1
34.9

3646.5
97
3

3751.4
99.8
0.2

3758.5

3760

- 320 -

Table 5: 10mm aggregate grading

Sieve Size, (mm)
Wt. Retained, (g)
Cum. Wt., (g)
% Retained
% Passing

20.00

14.00 10.00
385.9
385.9
17.6
82.4

5.00
1665.6
2051.5
93.5
6.5

2.36
121.5
2173
99.0
1.0

0.075
9.1
2182.1
99.4
0.6

Pan
12.9
2195

Table 6: Unwashed CRF grading

Sieve Size, (mm)
Wt. Retained, (g)
Cum. Wt., (g)
% Retained
% Passing

5.00
59.1
59.1
4.4
95.6

2.36 1.18
0.60
0.30 0.150 0.075
490.8 323.4 160.1 87.6
44.4
59.3
549.9 873.3 1033.4 1121 1165.4 1224.7
40.8 64.8
76.7
83.2
86.5
90.9
59.2 35.2
23.3
16.8
13.5
9.1

Pan
122.6
1347.3

Table 7: Washed CRF grading

Sieve Size, (mm)
Wt. Retained, (g)
Cum. Wt., (g)
% Retained
% Passing

5.00
58.5
58.5
5
95

2.36
454
512.5
43.5
56.5

1.18
280.6
793.1
67.3
32.7

0.60
155.9
949
80.6
19.4

0.30
81.3
1030.3
87.4
12.6

0.150
55.1
1085.4
92.2
7.8

0.075
42.3
1127.7
95.7
4.3

Table 8: Dune Sand grading

Sieve Size, (mm)
Wt. Retained, (g)
Cum. Wt., (g)
% Retained
% Passing

0.60
0
0
0
100

0.30
4.8
4.8
0.8
99.2

- 321 -

0.150
296.9
301.7
53.1
46.9

0.075
262
563.7
99.2
0.8

Pan

Pan

Table 9: Combined grading with Unwashed CRF

SIEVE
SIZES
37.5
20
5
0.600
0.150

20 mm
100
88.8
0.2
0
0

10 mm
100
100
6.5
1
0

5 mm
100
100
95
23.3
13.5

D/SAND COMBINED
100
100
100
99
100
57
100
35
46.9
17

Table 10 : Combined grading with washed CRF

SIEVE
SIZES
37.5
20
5
0.600
0.150

20 mm
100
88.8
0.2
0
0

10 mm
100
100
6.5
1
0

5 mm
100
100
95
19.4
7.8

- 322 -

D/SAND COMBINED
100
100
100
99
100
57
100
34
46.9
16

Table 11: Slump flow retention and L-box test results GGBS mixes
Mix
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14

T0
725
710
745
735

T30
715
680
730
725

T60
700
650
710
680

T90
620
610
665
610

LBox
h1/h2
0.82
0.80
0.89
0.84

760

740

700

680

0.9

730

720

700

680

0.88

750
750
710
720
745

730
740
700
700
720

660
660
680
680
680

600
640
650
650
620

0.87
0.86
0.89
0.88
0.85

720

700

660

610

0.9

730

705

675

610

0.88

735

710

665

620

0.9

Slump Flow, mm

Concrete Grade
UW(30%OPC+ 70%GGBS)
UW(20%OPC+ 80%GGBS)
UW(10%OPC+ 90%GGBS)
UW(0%OPC+ 100%GGBS)
UW(20%OPC+
75%GGBS+5%MS)
UW(10%OPC+
85%GGBS+5%MS)
UW(95%GGBS+5%MS)
W(30%OPC+ 70%GGBS)
W(20%OPC+ 80%GGBS)
W(10%OPC+ 90%GGBS)
W(0%OPC+ 100%GGBS)
W(20%OPC+
75%GGBS+5%MS)
W(10%OPC+
85%GGBS+5%MS)
W(95%GGBS+5%MS)

- 323 -

Table 12: Slump flow retention and L-box test results Fly ash mix
Mix
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

Concrete Grade
UW(70%OPC+ 30%FA)
UW(60%OPC+ 40%FA)
UW(50%OPC+ 50%FA)
UW(40%OPC+ 60%FA)
UW(30%OPC+ 70%FA)
UW(20%OPC+ 80%FA)
UW(20%OPC+
75%FA+5%MS)
UW(30%OPC+
65%FA+5%MS)
UW(40%OPC+
55%FA+5%MS)
UW(50%OPC+
45%FA+5%MS)
UW(60%OPC+
35%FA+5%MS)
W(70%OPC+ 30%FA)
W(60%OPC+ 40%FA)
W(50%OPC+ 50%FA)
W(40%OPC+ 60%FA)
W(30%OPC+ 70%FA)
W(20%OPC+ 80%FA)
W(20%OPC+
75%FA+5%MS)
W(30%OPC+
65%FA+5%MS)
W(40%OPC+
55%PFA+5%MS)
W(50%OPC+
45%FA+5%MS)
W(60%OPC+
35%FA+5%MS)

T0
720
740
760
750
750
760

Slump Flow, mm
T30
T60
720
710
740
735
745
730
740
740
745
730
760
760

T90
700
730
710
720
710
745

L-Box
h1/h2
0.9
0.9
0.9
0.9
0.9
0.86

740

730

720

700

0.98

750

730

725

700

0.99

740

715

700

690

0.97

750

720

695

690

0.92

730

700

690

670

0.98

710
770
770
790
770
750

710
760
760
770
770
730

695
750
745
740
750
710

690
730
715
730
730
670

0.99
0.99
0.99
0.98
0.98
0.99

750

750

730

710

0.98

750

740

710

690

0.99

730

705

680

665

0.98

720

715

710

700

0.97

715

710

700

660

0.98

- 324 -

Table 13: Compressive Strength GGBS Mixes

Mix
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14

Concrete Grade
UW(30%OPC+ 70%GGBS)
UW(20%OPC+ 80%GGBS)
UW(10%OPC+ 90%GGBS)
UW(0%OPC+ 100%GGBS)
UW(20%OPC+ 75%GGBS+5%MS)
UW(10%OPC+ 85%GGBS+5%MS)
UW(95%GGBS+5%MS)
W(30%OPC+ 70%GGBS)
W(20%OPC+ 80%GGBS)
W(10%OPC+ 90%GGBS)
W(0%OPC+ 100%GGBS)
W(20%OPC+ 75%GGBS+5%MS)
W(10%OPC+ 85%GGBS+5%MS)
W(95%GGBS+5%MS)

- 325 -

Compressive Strength,
MPa
3d
7d
28d
32
48.0
58.5
27
43.0
56.0
22
39.5
47.0
16
22.5
30.0
27
40.5
54.0
22
36.5
43.0
15
19.5
26.5
28
43.0
56.0
28
45.0
54.5
24
31.0
46.0
16
19.0
25.5
25
34.5
47.0
20
34.0
39.0
16
18.0
23.5

Table 14: Compressive Strength Fly ash mixes

Mix
Concrete Grade
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

UW(70%OPC+ 30%FA)
UW(60%OPC+ 40%FA)
UW(50%OPC+ 50%FA)
UW(40%OPC+ 60%FA)
UW(30%OPC+ 70%FA)
UW(20%OPC+ 80%FA)
UW(20%OPC+ 75%FA+5%MS)
UW(30%OPC+ 65%FA+5%MS)
UW(40%OPC+ 55%FA+5%MS)
UW(50%OPC+ 45%FA+5%MS)
UW(60%OPC+ 35%FA+5%MS)
W(70%OPC+ 30%FA)
W(60%OPC+ 40%FA)
W(50%OPC+ 50%FA)
W(40%OPC+ 60%FA)
W(30%OPC+ 70%FA)
W(20%OPC+ 80%FA)
W(20%OPC+ 75%FA+5%MS)
W(30%OPC+ 65%FA+5%MS)
W(40%OPC+ 55%PFA+5%MS)
W(50%OPC+ 45%FA+5%MS)
W(60%OPC+ 35%FA+5%MS)

- 326 -

Compressive Strength,
MPa
3
7
28
32.5
42.0
58.0
32.5
40.0
53.0
27.0
35.5
54.5
22.0
29.0
49.0
14.0
19.5
38.5
11.0
16.0
30.0
13.5
19.0
28.5
16.5
21.0
38.0
20.5
27.5
42.0
19.0
30.5
45.0
27.0
40.0
51.0
27
37.0
53.0
27.5
35.5
51.0
26
33.0
49.0
21.0
23.0
42.0
15.0
19.2
33.0
7.0
11.0
20.5
9.5
12.0
24.0
17
21.0
37.5
16.5
24.0
40.0
22.5
31.0
44.0
24.0
36.0
48.0

Table 15: Durability Tests GGBS mix

Mix
No.
1
2
3
4
8
9
10
11

Concrete Grade
UW(30%OPC+
70%GGBS)
UW(20%OPC+
80%GGBS)
UW(10%OPC+
90%GGBS)
UW(0%OPC+
100%GGBS)
W(30%OPC+
70%GGBS)
W(20%OPC+
80%GGBS)
W(10%OPC+
90%GGBS)
W(0%OPC+
100%GGBS)

Water
Water
Permeability Absorption
Coulomb
mm
%
RCPT

377

15

1.1%

315

10

1.2%

513

0.9%

257

3.5

1.0%

297

15

1.0%

442

0.7%

560

1.2%

355

1.2%

Table 16: Durability Tests Fly Ash mix

Mix
No.
1
2
3
4
5
6
12
13
14
15
16
17

Concrete Grade
UW(70%OPC+ 30%FA)
UW(60%OPC+ 40%FA)
UW(50%OPC+ 50%FA)
UW(40%OPC+ 60%FA)
UW(30%OPC+ 70%FA)
UW(20%OPC+ 80%FA)
W(70%OPC+ 30%FA)
W(60%OPC+ 40%FA)
W(50%OPC+ 50%FA)
W(40%OPC+ 60%FA)
W(30%OPC+ 70%FA)
W(20%OPC+ 80%FA)

RCPT
Coulomb
806
578
542
579
670
502
970
864
574
528
452
516

- 327 -

Water
Permeability
mm
0
0
0
0
0
0
0
0
0
0
0
0

Water
Absorption
%
1
0.2
1
0.8
0.8
0.9
1.1
0.9
0.3
0.6
0.8
1.4

PASSING %

Combined Aggregate Grading Chart

Unwashed CRF
100
90
80
70
60
50
40
30
20
10
0
0.1

10

100

B.S SIEVE (m m )

Fig.1: Combined aggregate grading unwashed CRF

PASSING %

Combined Aggregate Grading Chart

Washed CRF
100
90
80
70
60
50
40
30
20
10
0
0.1

10
B.S SIEVE (m m )

Fig. 2: Combined aggregate grading washed CRF

- 328 -

100

Flow Retention
GGBS - Unwashed CRF

800

UW(30%OPC+
70%GGBS)
750

UW(20%OPC+
80%GGBS)
UW(10%OPC+
90%GGBS)

mm

700

UW(0%OPC+
100%GGBS)

650

UW(20%OPC+
75%GGBS+5%MS)
600

UW(10%OPC+
85%GGBS+5%MS)

550
T0

T30

T60

T90

UW(95%GGBS+5%
MS)

Time

Fig. 3: Flow retention for GGBS mixes with unwashed CRF

Flow retention
GGBS - Washed CRF
800
W(30%OPC+
70%GGBS)
750

W(20%OPC+
80%GGBS)
W(10%OPC+
90%GGBS)

700

mm

W(0%OPC+
100%GGBS)
W(20%OPC+
75%GGBS+5%MS)

650

W(10%OPC+
85%GGBS+5%MS)
600

W(95%GGBS+5%
MS)

550
T0

T30

T60

T90

Time

Fig. 4: Flow retention for GGBS mixes with washed CRF

- 329 -

Flow Retention
Fly Ash - Unwashed CRF
780
760
740

mm

720
700
680
660
640
620
600
T0

T30

T60

T90

UW(70%OPC+
30%FA)
UW(60%OPC+
40%FA)
UW(50%OPC+
50%FA)
UW(40%OPC+
60%FA)
UW(30%OPC+
70%FA)
UW(20%OPC+
80%FA)
UW(20%OPC+
75%FA+5%MS)
UW(30%OPC+
65%FA+5%MS)
UW(40%OPC+
55%FA+5%MS)
UW(50%OPC+
45%FA+5%MS)
UW(60%OPC+
35%FA+5%MS)

Time

Fig. 5: Flow retention for Fly ash mixes with unwashed CRF

Flow Retention
Fly Ash - Washed CRF
800

W(60%OPC+ 40%FA)

780

W(50%OPC+ 50%FA)

760

W(40%OPC+ 60%FA)

740

W(30%OPC+ 70%FA)

720

mm

W(70%OPC+ 30%FA)

W(20%OPC+ 80%FA)

700

W(20%OPC+ 75%FA+5%MS)

680
660

W(30%OPC+ 65%FA+5%MS)

640

W(40%OPC+ 55%PFA+5%MS)

620

W(50%OPC+ 45%FA+5%MS)

600
T0

T30

T60

T90

W(60%OPC+ 35%FA+5%MS)

Time

Fig. 6: Flow retention for Fly ash mixes with washed CRF

- 330 -

Compressive Strength
GGBS - Unwashed CRF
With & without MS
UW(30%OPC+
70%GGBS)

60
55

UW(20%OPC+
80%GGBS)

50
45

UW(20%OPC+
75%GGBS+5%
MS)
UW(10%OPC+
90%GGBS)

MPa

40
35
30

UW(10%OPC+
85%GGBS+5%
MS)
UW(0%OPC+
100%GGBS)

25
20
15

UW(95%GGBS+
5%MS)

10
0

14

21

28

Days

Fig. 7: Compressive strength development of GGBS mixes with unwashed

sand and with & without Microsilica
Compressive Strength
GGBS - Washed CRF
With & without MS

60

W(30%OPC+
70%GGBS)
W(20%OPC+
80%GGBS)

50

W(20%OPC+
75%GGBS+5%
MS)
W(10%OPC+
90%GGBS)

mm

40
30

W(10%OPC+
85%GGBS+5%
MS)
W(0%OPC+
100%GGBS)

20
10

W(95%GGBS+5
%MS)

14 Days 21

28

35

Fig. 8: Compressive strength development of GGBS mixes with unwashed

sand and with & without Microsilica

- 331 -

28D Compressive Strength

Unwashed vs Washed w/o MS

70.0
60.0

58.5

56.0 56.0 54.5

W(30%OPC+ 70%GGBS)
47.0 46.0

50.0
MPa

UW(30%OPC+
70%GGBS)

UW(20%OPC+
80%GGBS)

40.0

W(20%OPC+ 80%GGBS)
30.0

30.0

25.5

20.0

UW(10%OPC+
90%GGBS)
W(10%OPC+ 90%GGBS)

10.0

UW(0%OPC+
100%GGBS)

0.0

W(0%OPC+ 100%GGBS)

GGBS

Fig. 9: Compressive strength at 28 days: Unwashed vs. Washed

GGBS without Microsilica

28D Compressive Strength

Unwashed vs Washed w/MS
60.0
50.0

MPa

40.0

UW(20%OPC+
75%GGBS+5%MS)

54.0
47.0

W(20%OPC+
75%GGBS+5%MS)

43.0
39.0

UW(10%OPC+
85%GGBS+5%MS)

26.5

30.0

23.5

W(10%OPC+
85%GGBS+5%MS)

20.0

UW(95%GGBS+5%MS)

10.0

W(95%GGBS+5%MS)

0.0
GGBS

Fig. 10: Compressive strength at 28 days: Unwashed vs. Washed GGBS

with Microsilica

- 332 -

Compressive Strength
Fly Ash - Unwashed CRF
With & Without MS
60
UW(70%OPC+ 30%FA)

55

UW(60%OPC+ 40%FA)

50

UW(60%OPC+
35%FA+5%MS)
UW(50%OPC+ 50%FA)

45

MPa

40
UW(50%OPC+
45%FA+5%MS)
UW(40%OPC+ 60%FA)

35
30

UW(40%OPC+
55%FA+5%MS)
UW(30%OPC+ 70%FA)

25

UW(30%OPC+
65%FA+5%MS)
UW(20%OPC+ 80%FA)

20
15

UW(20%OPC+
75%FA+5%MS)

10
0

14

21

28

Days

Fig. 11: Compressive strength development of Fly ash mixes with unwashed
sand and with & without Microsilica
Compressive Strength
Fly Ash - Washed CRF
With & Without MS

60
W(70%OPC+ 30%FA)
W(60%OPC+ 40%FA)

50

W(60%OPC+
35%FA+5%MS)
W(50%OPC+ 50%FA)

40

MPa

W(50%OPC+
45%FA+5%MS)
W(40%OPC+ 60%FA)

30

W(40%OPC+
55%PFA+5%MS)
W(30%OPC+ 70%FA)

20

W(30%OPC+
65%FA+5%MS)
W(20%OPC+ 80%FA)

10

W(20%OPC+
75%FA+5%MS)

0
0

10

15

20

25

30

Days

Fig. 12: Compressive strength development of Fly ash mixes with unwashed
sand and with & without Microsilica

- 333 -

28d Compressive Strength

Unwashed vs Washed w/o MS
70.0

60.0

UW(70%OPC+ 30%FA)
W(70%OPC+ 30%FA)

58.0
53.0

53.0

54.5
51.0

UW(60%OPC+ 40%FA)
49.0

50.0

49.0

W(60%OPC+ 40%FA)
42.0

UW(50%OPC+ 50%FA)
38.5

MPa

40.0

W(50%OPC+ 50%FA)

33.0
30.0

UW(40%OPC+ 60%FA)

30.0
20.5
20.0

W(40%OPC+ 60%FA)
UW(30%OPC+ 70%FA)
W(30%OPC+ 70%FA)

10.0

UW(20%OPC+ 80%FA)
W(20%OPC+ 80%FA)

0.0
Fly Ash

Fig. 13: Compressive strength at 28 days: unwashed vs. washed Fly ash
without Microsilica

28d Compressive Strength

Unwashed vs Washed w/MS
UW(60%OPC+ 35%FA+5%MS)
60.0
W(60%OPC+ 35%FA+5%MS)
51.0
50.0

48.0
45.0

UW(50%OPC+ 45%FA+5%MS)
44.0

42.0

40.0

40.0

38.0

W(50%OPC+ 45%FA+5%MS)
37.5

MPa

UW(40%OPC+ 55%FA+5%MS)
28.5

30.0

W(40%OPC+ 55%PFA+5%MS)
24.0
UW(30%OPC+ 65%FA+5%MS)

20.0
W(30%OPC+ 65%FA+5%MS)
10.0

UW(20%OPC+ 75%FA+5%MS)
W(20%OPC+ 75%FA+5%MS)

0.0
Fly Ash

Fig. 14: Compressive strength at 28 days: unwashed vs. washed Fly ash with
Microsilica

- 334 -

RCPT ASTM 1202

Unwashed vs Washed
560

600

513

Coulomb

500
400

UW(30%OPC+ 70%GGBS)

442
377

355
297

315

UW(20%OPC+ 80%GGBS)

257

300

W(30%OPC+ 70%GGBS)

W(20%OPC+ 80%GGBS)

200

UW(10%OPC+ 90%GGBS)

100

W(10%OPC+ 90%GGBS)

UW(0%OPC+ 100%GGBS)
W(0%OPC+ 100%GGBS)

GGBS

Fig. 15: RCPT comparison of GGBS mixes for unwashed vs. washed sand

Permeability DIN 1048

Unwashed vs Washed
16

15

15

14
12

UW(30%OPC+ 70%GGBS)
10

W(30%OPC+ 70%GGBS)

mm

10
8

UW(20%OPC+ 80%GGBS)

6
3.5

4
2

W(20%OPC+ 80%GGBS)
UW(10%OPC+ 90%GGBS)
W(10%OPC+ 90%GGBS)

UW(0%OPC+ 100%GGBS)

W(0%OPC+ 100%GGBS)
GGBS

Fig. 16: Permeability comparison of GGBS mixes for unwashed vs. washed
sand

- 335 -

Water absorption, BS 1881: Part 122

Unwashed vs Washed
UW(30%OPC+ 70%GGBS)

1.4%

1.2%

1.2%

1.2%

W(30%OPC+ 70%GGBS)

1.1%

1.2%

1.0%

1.0%
0.9%

1.0%

UW(20%OPC+ 80%GGBS)

0.7%

W(20%OPC+ 80%GGBS)

0.8%
0.6%

UW(10%OPC+ 90%GGBS)

0.4%

W(10%OPC+ 90%GGBS)

0.2%

UW(0%OPC+ 100%GGBS)

0.0%

W(0%OPC+ 100%GGBS)

GGBS

Fig. 17: Water absorption comparison of GGBS mixes for unwashed vs.
washed sand
RCPT ASTM 1202
Unwashed vs Washed
1200
UW(70%OPC+ 30%FA)
W(70%OPC+ 30%FA)

970

1000

UW(60%OPC+ 40%FA)

864

806

W(60%OPC+ 40%FA)

Coulomb

800

UW(50%OPC+ 50%FA)

670
600

578

542

574

579

W(50%OPC+ 50%FA)

528
452

502

516

UW(40%OPC+ 60%FA)
W(40%OPC+ 60%FA)

400

UW(30%OPC+ 70%FA)
W(30%OPC+ 70%FA)

200

UW(20%OPC+ 80%FA)
W(20%OPC+ 80%FA)

0
Fly Ash

Fig. 18: RCPT comparison of Fly Ash mixes for unwashed vs. washed sand

- 336 -

Water Absorption BS 1881:Part 122

Unwashed vs Washed
1.6

1.4

1.4
1.2

W(70%OPC+ 30%FA)
UW(60%OPC+ 40%FA)

1.1

1
0.9

0.8

0.8

0.8
0.6

0.6
0.4

UW(70%OPC+ 30%FA)

0.2

0.8

0.9

W(60%OPC+ 40%FA)
UW(50%OPC+ 50%FA)
W(50%OPC+ 50%FA)
UW(40%OPC+ 60%FA)

0.3

W(40%OPC+ 60%FA)

0.2

UW(30%OPC+ 70%FA)

W(30%OPC+ 70%FA)
UW(20%OPC+ 80%FA)
Fly Ash

W(20%OPC+ 80%FA)

Fig. 19: Water absorption comparison of GGBS mixes for unwashed vs.
washed sand

- 337 -

- 338 -

RHEOLOGY OF SELF COMPACTING CONCRETE AND Eco-SCC

Olafur H. Wallevik, Florian V. Mueller, Bjrn Hjartarson
ICI Rheocenter, Reykjavik University, Innovation Center Iceland

Synopsis: Eco-SCC is greener alternative to conventional self compacting

concrete (SCC), where the total powder content (cement, fly ash, silica fume,
limestone filler) is 315 kg/m3 or less. Its rheological properties are quite different
having very low plastic viscosity from those of normal SCC, having high powder
content. It is an economical and environmental advantageous concrete. Its
stability is accomplished by special grading of the aggregates. The grading is
based on so-called particle lattice effect, where theoretically the volume of
aggregate in a given particle size range shall be alike or higher than the content in
the corresponding upper adjacent range. Generally, it is characterized by
relatively continuous grading, which is rich in mid size and contains relatively
low amount of stone near the maximum size. It is also beneficial to use high fines
content, stabilizers (VMA) or silica fume to increase its stability in respect to
segregation and to reduce bleeding.

Keywords: Self compacting concrete, Rheology, Viscosity, Eco-SCC, Mixdesign

- 339 -

INTRODUCTION
General
The main intention by the Eco-SCC project is to develop concrete that contains
less cement (up to one third less) than the concrete traditionally used in the
industry today, without loss in quality. The production of 1 metric ton of cement
releases almost 1 ton of CO2 to the atmosphere. Thus a significant reduction in
release of CO2 to the atmosphere (due to cement production) will be attained.
Additionally, the concrete is to maintain the properties of normal SCC[1-4] and its
superiority to conventional vibrated concrete (CVC). Among other benefits is less
material cost, less shrinkage in concrete and thereby reduced risk for crackformation (due to reduced paste volume in the concrete). The approach used is
new, based on rheology. Other known methods, like replacing cement with fly ash
or blast furnace slag have been used.
In short[14], Eco-SCC is an economical and environmentally advantageous
alternative to conventional vibrated concrete, in which the total powder content
(cement, fly ash, silica fume, lime stone filler) is 315 kg/m3 or below.
Eco-SCC compared to conventional SCC
Eco-SCC has very low content of cementitious material compared to traditional
SCC. For instance it is common to use about 600 kg/m3 of cementitious materials
in SCC in Germany (and most of Asia) whereas it is as low as 360 kg/m 3 in
Denmark. Danish SCC is generally stiff (according to our definition it is a highyield SCC), but on the other hand it has acquired some 30% market share in
Ready-Mix concrete.
As Eco-SCC has so low cement content its rheology is quite different from other
SCCs, in particular as it has very low plastic viscosity. This makes Eco-SCC very
easy to place; on the other hand it is difficult to ensure stability (in respect of
segregation) with such a low viscosity.
In a survey by Domone[3] on sixty-eight case studies of applications of selfcompacting concrete (SCC) the average powder content was about 500 kg/m3
which is well over 50% higher than in conventional vibrated concrete, see figure
1. It is noteworthy that relative to distribution, the mixes containing stabilizers
(VMA) contains only 20 kg/m3 less powder than mixes without stabilizers, see
Figure 1a.

- 340 -

According to figure 1 the lowest content of powder in SCC is about 450 kg/m 3.
The Eco-SCC has much less or only 2/3 of that. The ICI Rheocenter
classification[14] of SCC relative to its powder content is as follow:
Rich SCC

550

kg/m3

or from 550 and upward

Regular powder content SCC: 500 50 kg/m3

or from 450 to 550

415 35 kg/m3

or from 380 to 450

or from 315 to 380

Lean SCC
Green SCC

350 35 kg/m

Eco-SCC

315

kg/m3

up to 315 kg/m3

Actually, the definition should be based on clinker content or CO2 per produced
ton of cement rather than powder content, but it is more complicated (also as the
clinker content in cement is so variable). In general SCC that has less than 380
kg/m3 of powder should be considered green alternative, in particular if or more
of it is not based on clinker and thus the Eco-SCC is very green.
The aggregates is the key
When the cement content in SCC is reduced the concrete is apt to become
unstable. In Eco-SCC the stability is accomplished by employing aggregate
grading which enhances the so-called particle-lattice effect. This grading is
characterized by high fines content, although preferably without colloidal
particles. Simply put, this is continuous grading with relatively high content of
mid size aggregate and relatively low content of stone near the maximal aggregate
size. The maximum aggregate size is generally 16 mm but can be reduced to 12
mm if necessary to obtain stable mix in respect of segregation, alternatively
enlarged to 22 mm in case of compliant aggregates.
In figure 2 are given particle size distributions of typical Eco-SCC in use, X-crete
(which is a SCC with 220 kg/m3 of cement) as well as theoretical distribution for
Eco-SCC with 16 and 22mm maximum aggregate size.
The principle of the particle-lattice concept (see Figure 2) implies that the content
of particles by volume in a given particle size range shall always be alike or
higher than the content in the corresponding upper adjacent range.
The theoretical grading would be a straight line across the diagram in figure 3 at
about 11%. The added powder shall be included (except for pure Portland cement
or the clinker in the cement) in that volume. As mentioned the aggregates shall
preferably not contain a significant amount of colloidal particles. If needed, these
can be removed from the sand/aggregates by simple floating process.

- 341 -

Naturally, it is not only the particle size distribution of the aggregates which is of
importance, but also their shape and surface properties, but they will not be
commented here.
In addition the Eco-SCC may also contain stabilizers (VMA) to increase its
stability and to reduce bleeding. Silica fume as powder added to the mix is
recommended as it adds volume to the binder and is at least as effective as most
chemical stabilizers.
A general comment regarding the aggregates; one can surely use worse grading
or quality but that will necessitate higher cement content. So the choice between
suitable grading or more cement can be a question of economy.
THE RHEOLOGY WAS VITAL PART OF THE DEVELOPEMENT
The flow characteristics
Fresh concrete requires application of force (or shear stress) in order to flow. In
other words, it has a certain resistance to flowing. The force required to initiate
flow is called the yield value (0), see figure 4a. Increasing the rate of flow (rate
of shear) causes the resistance (shear stress) to increase, as shown in figure 4b.
The slope () of the line is an expression of the plastic viscosity of the mass.

0 - The threshold value, 0 is a measure of the force required to initiate

flow.
- The slope, is a measure of the resistance to increase in the rate of
flow.
The yield value and the plastic viscosity of fresh concrete can be measured with
the help of a viscometer (or rheometers like the Tattersall two-point workability
test apparatus or the ConTec Rheometer-4SCC). Such a workability test apparatus
records the shear resistance (stress) of the concrete at two (or more) flow rates.
Figure 4 is based on linear relation between the rate of shear and shear stress
(Bingham behavior) but sometimes the relation can be more complex for example
in case of shear thickening behavior. References 6 and 7 provide a detailed
description.
A portable rheometer, named Rheometer-4SCC[8], has been developed at IBRI
(now ICI Rheocenter) which is intended for SCC and is at least as accurate as the
rheometers mentioned above. It is quite easy to use, requires about 7 liters of
concrete for testing and the time from filling the material container, through
testing and data processing, to emptying the container, is about 1 minutes (even
shorter than needed for the simple slump test).
Rheology applied to SCC
The rheology of self-compacting concrete differs from normal concrete. An SCC
exhibits almost no yield value (or about zero to 60 Pa, compared to normal
concrete where it ranges from a couple of hundreds to several thousands Pa). The

- 342 -

applied viscosity is very variable (especially between countries). It varies from

20 Pas to well over 100 Pas (in extreme cases the viscosity can be more than 200
Pas).
Rheological definition of SCC
The proposed area for SCC in a rheograph is shown in figure 6. If the plastic
viscosity is low or below some 40 Pas, the SCC should have significant yield
value (depending on the viscosity). On the other hand if the SCC is viscous i.e.
showing plastic viscosity over 70 Pas, the yield value has to be about zero. The
inner box (red-coloured) in figure 6 represents the recommended values for pairs
of yield and plastic viscosity values for SCC. The figure shows also the
necessary slump-flow depending on plastic viscosity to achieve self-compacting
concrete.
The applied plastic viscosity can differ considerably among SCCs and the author
has experienced viscosity in the range of 7 to 160 Pas. The pioneers like Japan,
Sweden and the Netherlands generally utilize very high viscosity due to the high
powder content. The yield value in these countries is normally negligible or near
zero. Norway and Switzerland (which often use very good aggregates) usually
apply very low plastic viscosity; on the other hand they normally employ a
significant yield value (typically 20 to 40 Pa). The Eco-SCC is similar in
rheology with typically values of yield value (Pa) and plastic viscosity (Pas) of
30 15.
COMPARISON OF FIVE DIFFERENT MIXES
The mixes
In order to demonstrate the difference in rheology of SCC and Eco-SCC, five
mixes where made. Four of these comprise (according to Wallevik) typical
composition of Danish, Norwegian, Swedish and German SCC mix-design and
the fifth is an Eco-SCC mix design. The composition of the binder/matrix is given
in table 1.
The volume of the binder in the five mixes is very varying. The so-called Danish
(which according to IBRIs (now ICI) definition is a semi-self compacting) mix
and Eco-SCC has the lowest volume of binder, see figure 8.
The rheology of these five mixes
The rheology of these five mixes was measured with two instruments, the
Rheometer-4SCC and the ConTec Viscometer 5[2,10,11]. The viscometer measures
the yield value (Pa) and the plastic viscosity (Pas) whereas the Rheometer-4SCC
measures the G-yield and H-viscosity. The results together with slump-flow
values are shown in figure 9.

- 343 -

The variability in the slump-flow of these five mixes

In order to demonstrate the very low plastic viscosity of the Eco-SCC relative to
for instance the very viscous German SCC, videos were made of the slump-flow
measurements. Snapshots from the videos are shown in figures 10 and 11.
ROBUSTNESS
Several projects have been conducted at ICI Rheocenter where one conducts socalled robustness-test on the Eco-SCC. Robustness-test is evaluation the
sensitivity of the workability/rheology of the concrete mix when water and
dispersion admixture (SP) content is gradually increased and decreased. Here, the
reporting of Mueller et al. will be confer. This paper[15] evaluates the robustness
of fresh concrete in respect to variation of water content and SP content. The main
emphasis is on rheological response, but some hardened concrete properties are
included in the investigation. Based on one reference mix, the water content is
varied by 5 l/m and 10 l/m, whereas the SP content is varied by 10% and
20%. Generally the mixes showed a systematic response to the varied mixdesign parameter. The compressive strength as a fuction of water/cement ratio is
shown in figure 12. Only in one case, when increasing water content by some 10
l/m, the mix revealed a certain risk of segregation but not to a great extent. Only
in two extreme cases, the mixes reveal semi-SCC properties, e.g. some energy
might be required for full consolidation. The variation of the PCE based SP did
not reveal significant effect on hydration process and no significant retardation
was observed, which leads to similar early strength. Shrinkage data of beams
analogous to ASTM C157 (placed in RH 50% at 1 day after de-molding, not
subjected to moist curing prior to beginning of drying) are low relative to data
from conventionally vibrated concrete (CVC) in Iceland, where the shrinkage is
typically over 1 mm/m after 3 months (see figure 12).
A PRACTICAL APPLICATION
Eco-SCC was used as casing around 430 m long vertical pipes (penstock) at
Krahnkar hydro-power plant in Iceland, where several thousand cubic meters of
Eco-SCC were placed. The total binder (80% cement + 20% fly ash) content in
this Eco-SCC was initially 290 kg/m3 and was increased to 320 kg/m3 when the
production was moved to another concrete plant (with no earlier experience with
SCC). Due to narrowness around the pipe, it would have been very difficult to
vibrate the concrete around the pipe and therefore SCC was chosen. The Eco-SCC
was here selected as it is environmental friendly and had low shrinkage (the paste
content is so low).

- 344 -

FINAL REMARKS
The key to make the Eco-SCC is the particle-lattice effect of aggregates. That
effect depends upon grading where the coarser particles are suspended in blend of
progressively smaller particles thus reducing the driving force of segregation.
The grading is characterized by relatively continuous grading, rich in mid size and
contains relatively low amount of stone near the maximum size. It is also
beneficial to use high fines content. Theoretically, the volume of aggregate in
given particle size range shall be alike or higher than the content in the
corresponding upper adjacent range.
As the total powder content is so very low (cement + fly ash + silica fume +
limestone filler is totally 315 kg/m3 or less) and concrete is so low viscous, the
Eco-SCC is environmentally friendly both in respect to greenhouse gasses and
hard labor work.
In addition, the concrete may contain stabilizers (VMA) or silica fume to increase
its stability in respect to segregation and to reduce bleeding.
ACKNOWLEDGEMENT
The authors gratefully acknowledges the support of (in alphabetical order):
CEMEX Research Group AG (Swiss), Icelandic Housing Fund, Landsvirkjun
(Iceland), MEST Steypustin (Iceland), Norstone (Norway) and Icelandic
Research Foundation Rannis for the project Eco-SCC. In addition, the
colleagues at the Institute and in particular S. Oesterheld are acknowledged for
their contributions to the project.

REFERENCES
1. Wallevik. O., Nielsson, I., Self Compacting Concrete A Rheological
Approach, International Workshop on SCC, Kochi, Japan {JSCE Concrete
Engineering Series no. 30}, August, 1998, pp. 136-159.
2. Wallevik, O.H., Rheology A Scientific Approach to Develop SelfCompacting Concrete, paper submitted to 3rd Int. Symp. on SCC, Rilem,
Reykjavik, Iceland, August, 2003, p. 10.
3. Nielsson, I., Wallevik, O.H., Mix Design of HS-SCC and Practical
Application, paper submitted to 3rd Int. Symp. on SCC, Rilem, Reykjavik,
Iceland, August, 2003, p. 8.

- 345 -

4. Wallevik, O.H., Practical description of rheology of SCC, SF Day at the Our

World of Concrete, Singapore, August, 2002, p. 42.
5. Domone, P.L., Self-compacting concrete: An analysis of 11 years of case
studies, Cement & Concrete Composite, 28, 2006, pp. 197-208.
6. Tattersall, G.H., Banfill, P.F.G., The rheology of fresh concrete, Pitman,
London, 1983, p. 356.
7. Barnes, H.A., Hutton, J.F., Walters, K., An introduction to rheology,
Elsevier, Amsterdam, 1989, p. 199.
8. Wallevik, O.H., Hjartarson, B., A novel field instrument to measure
rheological properties of self-compacting concrete, 3rd North America
Conference on Self-Compacting Concrete, USA, October, 2008, p. 6.
9. Wallevik, O.H., Rheology A New Dimension in Concrete Technology,
Bausil, Weimar, September, 2006, pp. 2_1417-2_1430.
10. Wallevik, O.H., Description of fresh concrete properties by use of two-point
workability test instrument, Master thesis, NTH, Trondheim, 1983, p. 136. (in
Norwegian).
11. Wallevik, O. H., Gjrv, O. E., Development of a Coaxial Cylinder
Viscometer for Fresh Concrete, Properties of Fresh Concrete proceedings of the
Rilem Colloquium. Chapman & Hall. Hanover, October, 1990, pp. 213-224.
12. Wallevik O.H., Kubens S., Mller F., Influence of cement-admixture
interaction on the stability of production properties of SCC, Fifth International
RILEM Symposium, Volume 1, Ghent, 2007, pp. 211-216.
13. Wallevik O.H., Mller F., Hjartarson, B., Eco-SCC, an environmental and
economical alternative, ICCX Oceania 2009, Sydney, 2009, pp. 16-20.
14. Wallevik, O.H., F.V. Mueller, and B. Hjartarson, Kubens, S., The green
alternative of self compacting concrete; Eco-SCC, iBausil, Germany, September,
2009, p. 12.
15. Mueller, F.V., Wallevik, O.H., Robustness of very low binder, self
compacting concrete: Eco-SCC, Rilem Conference Rheology of cement
suspension such as fresh concrete, August, Iceland, 2009, p. 121-132.

- 346 -

Table 1- Composition of the binder in five mixes of different origin

Cement

[kg/m3]

Silica fume
Fly ash
Limestone filler
Sum powder
Water
Glenium
51
PC
Gl.
stream
Stabilizer

[kg/m3]
[kg/m3]

GER

SWE

DK

NO

ECO

DK rapid

SWE bygg

DK rapid

NO standard

NO-FA

315

330

315
11
27

330
27

235*

353
150
1,5

357
175
2,6

275

[kg/m3]
[kg/m3]
[kg/m3]
[kg/m3]
[kg/m3]

185
515
180
3,5

590
165
5,3
0,34

50*
285*
185
2,9
3,42

*cement and FA milled together: FA-cement

Figure 1- Distribution of powder content in SCC[5] as a function of a) stabilizers

(VMA) and b) aggregates

Eco-SCC

X-crete

Theory-16

Theory-22

Percentage passing [%]

100

80

60

40

20

0
0,063

0,125

0,250

0,500

1,000

2,000

4,000

8,000 11,200 16,000 22,400 32,000

Sieve size [mm]

Figure 2- Grading curves for Eco-SCC with particle-lattice effect

- 347 -

40

CVC

Content of aggregates [%]

Theoretical Eco-SCC
30

Practical Eco-SCC
20

10

0,0010,250,063 0,0630,125 0,1250,25

0,5
0.063 0.125 0.250 0.50

0,51

12

24

48

816

1632

1.0

16

32

Aggregate fraction [mm]

Figure 3- A suitable particle size distribution of aggregates for Eco-SCC (in

green) compared to grading used in conventional vibrated concrete (CVC) having
particle gap around 6 mm and no filler

Rate of shear (1/s)

Shear stress (Pa)

c)

Shear stress (Pa)

b)

Rate of shear (1/s)

Rate of shear (1/s)

Figure 4- A simple flow-curve of fresh concrete according to Bingham model

- 348 -

Figure 5. The ConTec rheometer-4SCC at building site [9]

Yield value (Pa)

160

120

80

550 mm

Min. Slump-flow to obtain SCC

Eco 600 mm

40

650 mm

700 mm

0
0

30

60

90

120

Plastic viscosity (Pa s)

Figure 6- Proposed area for SCC in a rheograph [4] including a area for Eco-SCC

- 349 -

Figure 7- Eco-SCC is easy to place due to its very low plastic viscosity
(Lj on the book means poem)

40%
Paste volume (%)

Paste volume (%)

40%

35%

30%

30%
20%
10%
0%

25%

NO DK Eco SE

NO DK Eco SE

Figure 8- Volume of binder/matrix (without air) in the five mixes [13]

Start
3 sec. after start
Figure 10- At start of the slump-flow measurements on four of the five mixes
(left) and the slump-flow three seconds later (right). The Eco-SCC (upper left
corners) is fairly advanced in three seconds [13]

- 350 -

8 sec. after start, Eco is finished

15 sec. The DK/NO is finished

28 sec. The SE is finished

70 sec. The D is almost finished
Figure 11- The positions of the same mixes 8, 15, 28 and 70 seconds after start.
The mixes are given a gray-bluish shade when finished. The Eco-SCC (upper left
corners) was finished at 8 seconds after start while the German mix is still
advancing at 70 seconds [13]

- 351 -

2,5

.
2,0

120

G-Yield (A)

Yield value (Pa)

160

80
40

1,5
1,0
0,5
0,0

0
0

40
80
120
160
Plastic viscosity (Pa s)

200

8
12
16
H-Viscosity (A s)

20

24

2,0

120

G-Yield (A)

Yield value (Pa)

2,5
160

1,5

80

1,0

40

0,5

0
400

500

600
700
800
Slum p flow (m m)

0,0
400

900

500

600
700
800
Slum p flow (m m )

900

80

1,2

60

Shrinkage [mm/m]

Compressive strength [MPa]

Figure 9- Results from rheological measurements on five distinctive mixes [13]

Legend: Green Eco-SCC; orange NO; red DK; yellow SE and blue, D

28 d
2d

40
20
0
0,4

0,5
0,6
w/b-ratio [%]

0,7

+ 10 l
+5

0,8

Ref
-5l

- 10 l

0,4

IS-CVC
s_1

0,0
0

56

112

168

Time [d]

Figure 12- Compressive strength and shrinkage data of 5 mixes from the robusttest having water content of 178, 181, 186, 191 and 200 kg/m3 [15]

- 352 -

430m
high

Figure 12- A sketch of the 430m long vertical pipes (penstock) at Krahnkar
hydro-power plant

- 353 -

- 354 -