SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 1 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
Task 4
PROPERTIES OF HARDENED CONCRETE
Final report
Partner:
Advanced Concrete Masonry Centre
University of Paisley
Scotland, United Kingdom
Authors:
M. SONEBI, P.J.M. BARTOS
W. ZHU, J. GIBBS, A. TAMIMI
May 2000
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 1 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
Contents
1. INTRODUCTION...................................................................................................3
2. SELECTION AND PROPERTIES OF BASIC RAW MATERIALS
2.1 Aggregates.............................................................................................................3
2.2 Cement and fillers .................................................................................................4
2.3 Admixtures............................................................................................................5
2.4 Fibres.....................................................................................................................5
3. MIX PROPORTIONS USED.................................................................................5
4. FRESH CONCRETE..............................................................................................6
4.1 Mixing procedures: ready mix concrete plant.......................................................6
4.2 Test methods for properties of fresh SCC.............................................................6
4.3 Results of tests on fresh concrete: SCC mixes and reference mixes.....................8
4.4 Settlement and bleeding ........................................................................................9
5.
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
PROPERTIES OF HARDENED CONCRETE DETERMINED USING
BS 1881: PART 116 AND OTHER SPECIMENS ..........................................11
Compressive strength..........................................................................................11
Indirect Tensile strength......................................................................................13
Bond strength......................................................................................................14
Modulus of elasticity...........................................................................................17
Drying shrinkage (UoP)......................................................................................18
Shrinkage and creep (LCPC) ..............................................................................18
Water absorption of near surface concrete..........................................................23
Carbonation.........................................................................................................26
Freeze-thaw resistance ........................................................................................26
6.
PROPERTIES OF HARDENED CONCRETE IN FULL-SIZE
STRUCTURAL ELEMENTS ...........................................................................27
6.1 In-situ testing ......................................................................................................27
6.1.1 Test arrangements ............................................................................................28
6.1.2 Results and discussions of in-situ strength ......................................................29
6.1.3 Results and discussions of the other in-situ tests .............................................33
6.1.4 Statistical analysis of the test results................................................................35
6.1.5 Conclusions .....................................................................................................38
6.1.6 Micro-mechanical study of the interfacial zone ...............................................38
6.2 Structural performance of full-scale elements .................................................44
6.2.1 Design...............................................................................................................44
6.2.2 Concrete placing procedures ............................................................................48
6.2.3 Test set-up for columns ....................................................................................50
6.2.4 Test set-up for beams .......................................................................................50
1
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 2 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
6.2.5 Testing columns ...............................................................................................52
6.2.6 Testing beams...................................................................................................53
6.2.7 Cracking patterns and cracking-spacing...........................................................53
6.2.8 Cracking moment .............................................................................................54
6.2.9 Crack-width......................................................................................................55
6.2.10 Load-deflection response ................................................................................56
7. CONCLUSIONS ..................................................................................................56
APPENDICES
A1
A1.1
A1.2
A1.3
A1.4
A1.5
Preliminary investigation...................................................................................60
Materials............................................................................................................60
Observations on effect of materials on concrete properties..............................61
Observations on effect of mixer ........................................................................62
Test methods for fresh properties of SCC.........................................................63
Proportions and fresh concrete properties of preliminary mixes ......................64
A2
Analysis of pull-out test results.........................................................................69
Notation
Reference housing
Self-compacting concrete housing
Reference civil engineering
Self-compacting concrete civil engineering
Fibre self-compacting concrete housing
2
RH
SCCH
RC
SCCC
FSCC
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 3 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
1. INTRODUCTION
The fundamental objective of this task was to provide information on the hardened
properties of self-compacting concrete produced using easily available local raw
materials in Scotland to support the practical work of other partners in assessing the
practicability of actually building with SCC, and to facilitate the introduction of SCC
technology into general construction practice. To this end, the engineering properties
of SCC in sampled specimens and in full size structures were to be compared with
those of normal concrete. Two basic categories of concrete were considered: a
‘housing’ category, and a ‘civil engineering’ category, with characteristic cube
strengths of 35 MPa and 60 MPa respectively. In addition a steel fibre-reinforced
SCC mix was developed.
The concrete was produced in a commercial concrete plant, and the following
mechanical properties of sampled specimens were tested: compressive and flexural
strength, shrinkage, creep, bond strength with reinforcement, and elastic modulus. In
addition, in order to assess durability, longer-term comparative tests were made on
freeze-thaw resistance, water absorption, and carbonation. Testing of fresh concrete
was not part of the task, but necessary for mix design verification and acceptance
purposes. To supplement this, special tests were made on settlement and bleeding.
The full-size elements were tested in two ways. Firstly a number of in-situ tests were
carried out, and secondly, the whole elements themselves were load-tested to failure.
Also, for each type of concrete, a full size element has been left at an outdoor
exposure site for future assessment of durability performance.
Thus the outputs from the task were
• Verification of compliance with specifications for hardened SCC provided by
external supplies.
• Verification of properties of hardened SCC in completed structures using in situ
test methods.
• Comparison of basic mechanical properties of hardened SCC with those of
ordinary concrete.
• (Eventually) Comparison of some aspects of the durability of hardened SCC with
those of ordinary concrete.
2. SELECTION AND PROPERTIES OF BASIC RAW MATERIALS
2.1 Aggregates
A continuously graded crushed microgranite aggregate with a nominal particle size of
20-5mm and a well-graded quartzite sand according to BS 812 grading with a
fineness modulus of 2.74 were used (Table 1). The typical relative densities (SSD) of
the coarse aggregate and sand were 2.65 and 2.56, and typical absorption values 0.8%
and 1%, respectively.
3
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 4 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-5-2
Table 1 - Grading of aggregate
Sieve size (mm)
20 – 5 mm
10 – 5 mm
Sand
28
20
14
10
6.3
5.0
2.36 1.18 0.6
0.3
100
100
100
95.6
100
100
56.6 28.1 7.8 2.2 0.3 ---100 100 43.6 9.6 0.2 ---100 100 100 99.9 85.6 68.1 50.7 21.0
0.15 0.075
--0.7
--0.1
2.2 Cement and fillers
The SCC mixes investigated in this study were prepared with Portland cement, with
either limestone powder (LSP) or ground granulated blast slag (GGBS) added as the
filler. The reference mixes used PC only. The cement and GGBS used conformed to
BS 12:1996 and 6699:1992. The chemical and physical properties of cement, GGBS
and limestone powder are presented in Table 2. All the investigated mixes
incorporated one of the mineral additives (LSP or GGBS) to enhance workability and
segregation resistance. The gradings of the cements and limestone powder are shown
in Fig. 1. The limestone powder was produced from carboniferous limestone of a
very high purity and was finer than cement.
Table 2 - Chemical and physical properties of cement and fillers
Limestone GGBS Cement
powder
SiO 2
-33.5
20.8
Al2 O3
-13.6
5.0
Fe2 O3
-0.62
3.2
CaO
-42.7
63.7
MgO
0.2
6.5
2.6
Na2 O eq.
--0.39
Free CaO
--1.6
LOI
-0.67
0.65
CaCO3
99.0
Relative Density
2.65
2.90
-Bulk Density (loose)
0.90
1.00
-Bulk Density
-1.24
-(compacted)
Specific
Surface
460
385
2
area (m /kg)
Compressive Strength (MPa) (Mortar Prisms)
Age
30% C 100% C
+ 70%
GGBS
7d
-28
41.5
28 d
-51
57.8
4
Cement
Vicat set
times(min)
Initial
Final
119
164
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 5 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-5-2
100
LSP
Passing (%)
80
GGBS
60
PC
40
20
0
0.01
0.1
1
10
100
Sieve size (µ m)
Fig. 1 - Grading of Portland cement, LSP and GGBS
2.3 Admixtures
One of a new generation of copolymer-based superplasticisers, designed for
production of SCC was used (Viscocrete 2). This had a solid content and specific
gravity of 30% and 1.11, respectively. The Viscocrete was used at dosages varying
from 0 to 1%, by mass of fillers.
2.4 Fibres
The steel fibres used in the FSCC mix were Dramix from NV Bekaert, Belgium,
type RC 65/35BN. Fibre length and diameter were 35 mm and 0.55 mm, respectively.
The rate of addition of steel fibres to the mixes was 30 kg/m3 (0.38% by volume).
The suffix letters, BN, mean bright surface and normal tensile strength Dramix fibre
with hooked ends.
3. MIX PROPORTIONS USED
Table 3 summarises the mix proportions of SCCs and reference mixes. The SCC
mixes contain high-volume additions of limestone powder or blast furnace slag to
enhance fluidity and cohesiveness and limit heat generation. Such materials are also
less reactive than cement and can reduce the problems resulting from loss of fluidity
of the rich concrete. The incorporation of one or more powder materials having
different morphology and grain-size distribution can improve particle packing density
and reduce interparticle friction and viscosity, hence improving deformability, selfcompactability, and stability.
5
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 6 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
Table 3 - Mix proportions of concrete, SSD mass, kg/m3
Housing
Civil
FSCC
RH
SCCH
RC
SCCC
Free water
Portland Cement 42.5
Limestone Powder
GGBS
Fibre (RC 65/35BN)
200
295
-----
190
280
245
---
220
515
-----
192
330
--200
160
285
270
-30
Total powder content
295
525
515
530
555
Sand (0-5 mm)
20 mm agg. (*10 mm)
Viscocrete 2, kg
Normal superplasticiser
Water/cement ratio
840
970
---
865
750
4.2
870
750*
5.3
940
715
4.4
0.68
0.68
655
930*
--6.4
0.43
0.58
0.56
Water/powder ratio
0.68
0.36
0.43
0.36
0.29
4. FRESH CONCRETE
4.1
Mixing procedure at the ready mixed concrete plant
The concrete for the full-size pours (2-3m3 ) was mixed in the plant using a 6m3
capacity tilting drum central mixer. The mixing sequence consisted of ribbon feeding
all the materials into the large drum mixer, with the exception of the superplasticiser,
which was added at the end. The concrete was then mixed for three minutes.
4.2
Test methods for assessment of properties of fresh SCC
The test methods used for assessment of fresh properties of SCC were slump flow, Lbox, Orimet and Orimet/ JRing combined. (See also comments in Appendix)
The slump flow test consists of determination of the mean diameter of the concrete
sample spread on a base plate after performing a slump test without any compaction.
This test judges the capability of concrete to deform under its own weight against the
friction on the surface of the base plate with no other external restraint present.
Because of the viscous nature of SCC, the slump flow measurement was made only
when no further discernible movement of the concrete was observed: approximately
60 seconds after the removal of the slump cone. With the same test, the T50 slump
flow time was also measured. The T50 is the time to reach a spread of 500mm.
The dimensions of the L-box used are shown in Fig. 2. It is possible to measure
different properties such as filling and passing abilities and segregation with the Lbox. The vertical part of the box is filled with 12.7 litres of concrete and left to rest
6
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 7 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
for one minute in order to allow any segregation to occur. After that, the gate is
opened and the concrete flows out of the vertical part into the horizontal part through
the reinforcement bars. The gap between the reinforcement bars was 35 and 55 mm
for 10 and 20 mm coarse aggregate, respectively. The times for the leading edge of
the concrete to reach a distance 200 and 400 mm along the horizontal part, and the
heights H1 and H2 of concrete were measured and used to determine the L-box result
h1 /h2 . (Fig.2)
Gaps = 55 mm for SCCH
Gaps = 35 mm for SCCC
h 2 = 150 – H2
h 1 = 600 – H1
Fig. 2 - L-Box test
The Orimet test measures the flow time of concrete through a 120 mm diameter
vertical pipe fitted with an orifice. The orifice of the Orimet was 80 mm. The Orimet
included two horizontal steel bars of 10 mm diameter to obstruct the passage of
concrete through the orifice and thus simulate blocking. Three measurements of the
flow time were made, and the final measurement combined with the JRing. The
slump flow of SCC through the JRing was measured (Fig. 3). The dimensions of the
JRing were 300 mm diameter of its centreline and 100 mm height. The gaps between
the bars were 25 and 55 mm for 10 and 20 mm coarse aggregate, respectively.
Fig 3 - Orimet and JRing
7
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 8 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-5-2
4.3 Results of tests on fresh concrete: SCC mixes and references mixes
The results of tests of the fresh properties of SCCs and reference mixes are given in
Table 4.
Table 4 – Properties of fresh concrete
Housing
Slump Flow at 5 min (mm)
S. Flow at 60 min (mm)
T50 at 5 min (s)
T50 at 60 min (s)
L-box
- T20 and T40 at 10 min (s)
- h2 /h1 at 10 min
Orimet (orifice 80 mm)
- Flow time at 10 min (s)
- Flow time at 70 min (s)
Orimet + JRing spread
- 15 min (mm)
- 75 min (mm)
Air content (%)
Civil
RH
SCCH
RC
SCCC
FSCC
65 slump
50 slump
650
600
1.02
1.66
70 slump
70 slump
690
640
1.95
1.84
665
640
3.0
3.0
1-2
0.81
0.6 – 1.2
0.99
0.90
2.3
3.0
4.0
3.2
3.3
4.1
670
605
1.5
635
635
1.1
650
615
--
1.8
Notes:
SCCC - 10 mm coarse aggregate, clear spacing bars of JRing = 20 - 25 mm
SCCH - 20 mm coarse aggregate, spacing bars of JRing = 50 - 55 mm
Orifice of Orimet used for FSCC = 90 mm
The SCC tests are not appropriate for the reference mixes, which were therefore
assessed by the normal BS1881 slump test. Neither are their fresh properties really
comparable with those of the SCCs, whose fundamentally different mix proportions
make them significantly more flowable and workable. The volume proportion of
coarse aggregate in the reference mixes is significantly higher than that of the SCCs:
the SCCs were designed to have more filler, and the volumes of powder in the SCCH
and SCCC were 46% and 38% greater than those of the reference mixes. The SCC
mixes thus include more paste and less coarse aggregate, and this contributes to the
improved workability. The increase in flowability, however, is essentially due to the
addition of the superplasticiser, Viscocrete 2.
8
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 9 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
Slump flow: the values for SCCH, SCCC and FSCC at 5 min were 650 to 690 mm.
After 60 min, the slump flow of the SCC mixes was still greater than 600 mm, and the
slump flow loss was thus 7 to 8%: the loss in fluidity was minimal. The SCC mixes
had T50 values at 5 min between 1 to 3 seconds. After 60 min, the T50 values were still
less than 3 seconds, again indicating a satisfactory retention of workability.
The Orimet time at 10 min of the SCC mixes ranged from 2.3 to 4 sec, indicating
excellent deformability without blockage within five seconds.
The retention of workability with the Orimet and the combined Orimet/JRing tests
was variable. Test results are affected by the different aggregate sizes and gap
spacings in the JRing, as well as the different powder types.
The L-box results of the SCCs indicated a good deformability and flowability, and the
blocking ratio (h1 /h2 ) values were greater than 0.80, often considered a critical lower
limit. The blocking ratio was 0.81 and 0.99 for SCCH and SCCC, respectively. The
high value for SCCC indicates excellent deformability, without blockage, through
closely spaced obstacles, and the excellent capability of this highly flowable concrete
to self-compact.
The figures for air content indicate normal values for the SCC mixes.
4.4 Settlement and bleeding
A test used to evaluate the plastic fresh settlement of concrete and its ability to ensure
proper suspension of aggregate and fines is shown in Fig. 4. The stability of fresh
concrete was evaluated by casting concrete in a PVC column measuring 200 mm in
external diameter and 700 mm in height. The settlement was monitored using a linear
dial gauge or LVDT, with a precision of 0.01 mm, fixed on the top of a thin acrylic
sheet that was positioned on the top surface of the concrete. The plate was anchored
by three 50-mm long screws, cast into the fresh concrete. The bleeding and surface
settlement were monitored at set intervals until a steady-state condition was reached.
In the case of settlement, this corresponded to the beginning of set.
In this test, neither the SCC mixes nor the reference mixes exhibited any bleed water
on the surface.
The evolution of settlement of SCCs and the reference mixes are shown in Fig. 5. In
the case of the housing mixes, the maximum settlement of the RH mix was 0.24% and
0.21% of SCCH - 1.2 and 1 mm, respectively. However, in the case of the civil
engineering mixes, the maximum settlement of both mixes also similar. The highest
settlement of all mixes was observed with FSCC mix (0.26%).
9
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 10 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-5-2
Gauge
160 mm
Base-plate
diameter = 150 mm
Width = 6 mm
700 mm
4 screws
h = 40 mm
490 mm
Concrete
PVC tube
200 mm
Fig. 4 - The settlement test layout, using a 700-mm high column
0.3
Settlement (%)
0.25
0.2
0.15
RH
SCCH
0.1
SCCC
0.05
RC
FSCC
0
0
5
10
15
20
Time (h)
Fig. 5 - Plastic settlement of all mixes (Height of column = 50 cm)
10
25
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 11 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
5. PROPERTIES OF HARDENED CONCRETE DETERMINED USING
BS 1881: PART 116 AND OTHER SAMPLED SPECIMENS
5.1 Compressive strength
Standard cubes measuring 150 mm were demoulded one day after casting and
covered with wet hessian and plastic sheeting. Specimens were then cured either in
water or in air at approximately 20 °C, using a curing membrane, until testing was
carried out at 1 d, 7 d, 28 d, 90 d and 180 d.
The results of standard compressive strength at 28 days are presented in Table 4. The
specified characteristic cube strength was 35 MPa for housing mixes and 60 MPa for
civil engineering mixes.
The results in Table 5 indicate that the actual strengths of the SCC mixes were at the
upper end of the normal range for the designed strengths, while the reference mixes
were at the lower end. Such differences in actual strength of the SCC mix and the
reference mixes make direct comparisons difficult.
Table 5 - 28-day compressive strength results of standard specimens
Average of compressive Strength at 28
days, MPa
Concrete Mixes
SCCH
47.0
RH
37.0
SCCC
79.5
RC
61.5
FSCC
63.0
Housing Mixes
Civil Engineering Mixes
Fibre SCC Mix
The strength development up to 6 months is shown in Fig. 6. Rates of strength
development relative to 28-day strengths are shown in Fig. 7. As expected, the
compressive strength was strongly affected by w/c ratio and filler types. Results in
Figures 6 and 7, related to the mix design data in Table 3, indicate that at similar w/c
ratio, strengths of the SCC mixes, using the limestone powder as filler, were
significantly higher than the corresponding reference mixes. It should be emphasised,
however, that both results were within the normal compliance range.
The relatively faster strength development for SCCH and FSCC mixes, particularly at
early ages, is believed to be mainly due to the inclusion of fine limestone powder,
which may have an accelerating effect on C3 S hydration and early strengths [1]. The
SCCC mix, which contained GGBS instead of limestone powder, had a lower strength
at 1 and 7 days than the corresponding reference mix, RC, but developed significantly
higher strength at 28 days and beyond. This is due to the slower, but prolonged
11
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 12 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
reaction (hydraulic and pozzolanic) between cement hydration products and GGBS,
which contributes significantly to strength [2].
Results in Figure 7 also indicate that there was no significant difference in the pattern
of strength development for all the mixes studied.
100
f'cu (MPa)
80
60
40
20
RH
SCCH
RC
SCCC
FSCC
0
0
50
100
150
200
Age (d)
Fig. 6 – Compressive strength development
Strength relative to 28-day
results, %
120
100
80
60
40
SCCH
RH
20
SCCC
RC
FSCC
0
0
50
100
Age (d)
150
200
Fig 7 - Compressive strength at different ages relative to 28-day strength, %
12
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 13 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-5-2
Air cured strength as % of
water cured strength
The effect of curing conditions on strength development of the SCC and reference
mixes were examined, and the air cured strengths relative to water cured strengths are
presented in Fig. 8.
100
90
80
70
60
SCCH
RC
50
RH
FSCC
SCCC
40
0
20
40
60
80
100
Age at testing, days
Fig. 8 - Effect of curing conditions on compressive strength
As expected, the compressive strength of air-cured specimens is lower than that of the
corresponding water-cured specimens. However, the extent of strength reduction due
to the insufficient curing (i.e. in air) up to an age of 90 days depends on the strength
grade and the type of the filler type in the mixes. It appears that the SCC mixes, with
limestone filler (i.e. SCCH and FSCC) are less affected by air curing, and that aircured strengths are reduced less than those of reference concretes. For example, as
shown in Fig. 8, at 28 and 90 d, the relative strength ratios for SCCH were higher than
those for the corresponding reference mix RH (85 vs. 71% at 28 d and 79 vs. 65% at
90 d). This difference could be attributed to the accelerating effect of the limestone
powder and also possibly the enhanced water retentivity of the SCC mixes.
For the SCCC mix which incorporates GGBS, the strengths up to the age of 90 days
are more affected by the air curing, and the strength reduction due to the air curing is
greater than in the corresponding RC reference mixes. Such a difference in
sensitivity to curing conditions is normal for mixes containing GGBS, as continued
presence of water is required for the cement hydrates-GGBS reaction to continue.
5.2 Indirect tensile strength
The splitting tensile strength (f ’t) was determined at 28 days and 6 months on
cylinders measuring 150 x 300 mm and cured in water until the date of test according
the BS 1881: Part 117.
13
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 14 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
Indirect tensile (i.e. splitting) strengths at 28 days and 6 months, tested using φ 150 x
300 mm cylinder specimens, are given in Table 6. For easy comparison, the
tensile/compressive strength ratios are also included in the table. It is worth noting
that each result in Table 9 is the average of only two specimens, and in some cases the
variation of results was as high as 30%. Nevertheless, results in Table 6 indicate that
the relationships of the tensile strength to the compressive strength are of similar
orders for all the mixes studied.
Table 6 - Indirect tensile splitting strength (MPa)
Results
SCCH
RH
SCCC
RC
FSCC
Age of 28-d
3.4
2.4
4.7
4.1
4.0
Age of 180-d
3.3
3.1
5.7
3.9
4.5
7.1
6.5
5.9
6.6
6.4
6.0
7.7
6.2
5.5
6.4
28-d Tensile / Compressive
Strength Ratio, %
180-d Tensile / Compressive
Strength Ratio, %
5.3 Bond strength
The bond strength between reinforcing steel and the concrete was determined by pullout tests carried out at different ages.
Deformed reinforced steel bars with 12 and 20-mm effective diameters were used to
evaluate the bond, in accordance with the recommendation of Rilem set-up TC51ALC, 78-MCA [3]. The test specimen is a prism with a cross-section of 100 x 100
mm and a length of 150 mm. Three specimens were cast per mix. Each specimen
had horizontally bonded reinforcing bars of 12 or 20 mm in diameter and 1 m in
length. A rigid plastic sheathing was tightly attached to the loaded end of each bar to
limit the bond between the bar and concrete to the remaining portion of the bar. The
anchorage length was 120 mm for all bars. The bonded length of each bar was
properly cleaned to ensure an adequate bond with concrete. Average bond stresses
were evaluated by pull-out testing using a 2000-kN hydraulic machine and 200-kN
load cell. The pull-out load is applied progressively up to bond failure and the
deformation of the bar was measured using two LVDT connected to the unloaded end
of the bar (Fig. 9). A data acquisition system was used.
The test was terminated when pull-out failure occurred, the reinforced steel began to
yield, or the surrounding concrete cover failed in split. The average bond strength
was calculated as follows:
τ=
P
π d bls
where P, db , and ls correspond to the applied load, bar diameter and bonded length,
respectively.
14
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 15 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-5-2
The net slip was calculated as the total measured deformation minus the elastic
deformation in the steel and concrete.
150 x 100 x 100 mm
ls
Rigid plastic sheathing
db = 12 and 20 mm
P
Fig. 9 – Pull-out test arrangement used to determine bond strength
It is important to note that with reinforced concrete members, both the concrete and
the steel bars are simultaneously placed in tension in positive moment regions.
However, in the test arrangement adopted in this study, the pulled-out steel bar was
subjected to tension, while the surrounding concrete was in compression. The
confining compressive stresses around the steel bar were therefore reduced by
positioning the bonded region of the bar away from the loaded end of the specimen.
A good reproducibility of bond stress development vs. net slip was obtained for all
mixes and bar sizes. Typical variations in bond stress (τmax) vs. net slip for three
reinforcing bars of 20-mm diameter of SCCH are shown in Fig. 10. The coefficient
of variation (V) of the average bond strength values of the SCC mixes was between 2
to 14%. For all specimens of SCC and the reference mixes with 10 d b and 6 db
development lengths, failure resulted from the splitting of the cover. Given the
absence of stirrups that provide some confinement and delay crack propagation, the
splitting plane propagated through the cover. Therefore in the absence of ultimate
average bond strength values corresponding to a bond failure mode, average bond
stresses at small net slips were considered in the analysis of all mixes. Fig. 11 shows
the variations of the maximum bond strength of all mixes.
15
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 16 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
0.6
SCCH
d = 12 mm
Net slip (mm)
0.80
0.4
0.3
0.2
V = 1.5%
0.60
0.40
0.20
0.1
V = 6.7%
0
0.00
0.00
2.00
4.00
6.00
8.00
10.00
0
5
10
Average bond stress (MPa) at 75 d
Average bond stress (MPa) at 30 d
Fig. 10 – Typical changes of bond stress with net slip of RH and SCCH mixes
As expected, a reduction in bond strength was observed when the diameter of the bar
increased (Fig. 11). The τmax between embedded reinforcement bar and concrete
depends on the diameter of the bar, as well as the mechanical property (e.g. strength)
of the concrete. At 32 ± 4 d, the τmax of SCCC was about 32-45% higher than that of
the reference mix (RC).
40
Z max (MPa)
Net slip (mm)
0.5
1.00
RH mix
d = 20 mm
30
diam. = 12 mm
diam. = 20 mm
36 d
60 d
32 d
20
75 d
30 d
10
0
RH
SCCH
RC
SCCC
FSCC
Fig. 11 - Variations of max bond strength of all mixes
Bond strength is often expressed in terms of tensile strength of the concrete or the
square root of f’c. The τmax values are normalised by dividing them by the square root
of the mean f’c measured on the cubes. The normalised ratios of τmax values of the
SCC housing and civil engineering, and fibre SCC mixes are plotted in Fig. 12. Such
ratios for the reference mixes of housing and civil engineering are included for
comparison.
16
15
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 17 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
For housing mixes, the normalised ratios (τmax /
f'
cu
) for both diameters (12 and
20 mm) of SCCH are higher than those of reference. The SCCH normalised ratio was
about 10% higher than that of reference mix.
For the civil engineering mixes, the normalised ratios of SCCC were also higher than
those of RC mix by 18-38%.. The FSCC exhibited the highest values of normalised
ratios (3.63 for 12 mm and 2.65 for 20 mm).
Thus the SCCH and SCCC mixes exhibited high bond strength, and the normalised
ratio values of both were higher than those of the references mixes (RH and RC) for
both diameters considered in this project (12 and 20 mm).
4
diam. = 12 mm
Zmax / V f' c
3
diam. = 20 mm
2
1
0
RH
SCCH
RC
SCCC
Fig. 12 – Variation of ômax /
f'
cu
FSCC
ratio of mixes tested
5.4 Modulus of elasticity
The modulus of elasticity was determined according to British Standard 1881,Testing
concrete: Part 121: Method for determination of static modulus of elasticity in
compression. End capped φ150 x 300 cylinder specimens were cured in water and
tested at ages of 4 to 13 months for different mixes. Average results obtained from
two individual specimens for each concrete mix are given in Table 7.
For an easier comparison, the ratios of E modulus to square root of cylinder
compressive strength for all the mixes are also included, as a relationship in the form
of E / (fc)0.5 has been widely reported [4].
Table 7 - Static modulus of elasticity
Test results
Age at testing, months
Modulus E, GPa
E/(fc)0.5 ratio
Concrete mix
RH
SCCH
RC
SCCC
FSCC
13
4
34.1
8
34.4
11
41.9
7
37.7
4.92
4.98
4.43
5.43
17
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 18 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
The results in Table 7 indicate that the SCC mixes had the same relationship between
modulus of elasticity and compressive strength as the reference mixes. The E/(fc)0.5
ratio was also close to the value of 4.73 recommended by ACI 318-89 (revised 1992)
for structural calculations, applicable to normal weight concrete.
5.5 Drying shrinkage (U of P)
The dimensions of prisms used to measure the drying shrinkage were 100 x 100 x
400 mm. Drying shrinkage specimens were demoulded after one day, then cured in
water for a further six days. After curing, the specimens were stored in laboratory
environment with temperature of 20 ± 5 ºC. A Demec gauge was used to measure the
surface drying shrinkage on two parallel sides.
The experimental results of drying shrinkage are shown in Fig. 13. The drying
shrinkage of SCCs at 7 d was slightly higher than that of the reference mix but at 28
days and later, the reference mixes exhibited greater shrinkage than SCC mixes. The
drying shrinkages of the reference mixes were 30% (RH) and 35% (RC) higher than
those of the SCCs at 150 days. This is considered to be due to the effect of the volume
of paste and W/P. Similarly, the drying shrinkage of FSCC at 120 d was about 47%
lower that that of RC (490 µm/m vs. 720 µm/m).
1200
Strain (x10-6)
1000
800
600
SCCH
RH
400
SCCC
200
RC
FSCC
0
0
50
100
150
Age (d)
200
250
300
Fig. 13 - Drying shrinkage of mixes tested
5.6 Shrinkage and creep (Results from LCPC)
Experiments carried out at LCPC aimed at producing two different self-compacting
concretes with mean compressive cylinder strengths of 40 MPa and 70 MPa. Two
mix designs derived from the proportions proposed in Task 1 were selected and
shown in Table 8.
18
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 19 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Table 8 - Composition of the concretes tested at LCPC
SCC40
163
319
912
760
192
4.1
169
1.2
0.5287
0.51
Filler (kg/m³)
Cement (kg/m³)
Val de Reuil 0/4 (kg/m³)
Grande Paroisse 4/10 (kg/m³)
Water added (kg/m³)
Sikament 10 (kg/m³)
Free water (kg/m³)
Air (%)
Free water/cement ratio
Filler/cement
SCC70
52
444
874
851
176
9.3
156
0.5
0.3512
0.116
Description of the tests
The mixing process was the following:
− Mixing of dry materials for 1 min.
− Introduction of the water and 1/3 of the superplasticizer during 30 sec.
− Mixing for 2 min.
− Addition of the remaining 2/3 of superplasticizer and mixing for 1 min 30 sec.
Two batches of 140 litres were made in a pan mixer. Slump flow was measured and
specimens for determination of compressive strength at 28 days were made. The
results are summarized in Table 9.
Table 9 -Properties of fresh concretes LCPC test
SCC40
76 x 76
SCC70
80 x 76
T500 (s)
2.8
4.3
Strength at 28 days (MPa)
46.5
62
Slump flow (cm)
Then for SCC40 and SCC70, four Ø 160 x 1000 mm cylinders were cast for shrinkage
and creep measurements:
• one for autogenous shrinkage (AS, without desiccation)
• one for the total shrinkage (TS, with desiccation)
• one for basic creep (BC, without desiccation)
• one for total creep (TC, with desiccation).
In the report the samples are identified by the name of the mix and the type of test (eg:
SCC40-AS for autogeneous shrinkage on mix SCC40). The complete methodology
19
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 20 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
used at LCPC for shrinkage and creep tests is described in [5]. Figures 14 and 15
show the creep and the shrinkage frames. The base of the strain measurements is
50 cm long between the two sets of the 3 invar rods.
1
2
3
4
5
b
6
Legend
b
measurement base =
50 cm
1
LVDT
2
measurement triangle
3
creep frame
4
invar rod
5
concrete specimen
6
nitrogen accumulator
7
steel flat jack
8
3 way valve
9
load cell
10
rod support
7
8
Fig. 14 - Creep frame
Fig. 15 – Shrinkage frame [5]
20
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 21 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
For the tests without desiccation, the columns were wrapped with a double layer of
self-adhesive aluminium sheet just after demoulding.
Table 10 describes procedure used for each concrete cylinder:
Table 10- Planning for the preparation of the cylinders
Mix
Type of
test
Demolding
Aluminum
wrapping
SCC40
AS
at 1 day
at 1 day
SCC40
TS
at 1 day
SCC40
BC
at 1 day
SCC40
TC
at 1 day
SCC 70
AS
at 1 day
SCC 70
TS
at 1 day
SCC 70
BC
at 1 day
SCC 70
TC
at 1 day
Start of
desiccation
Level of
stress (MPa)
Age of
loading
11.51
at 28 days
11.21
at 28 days
19.65
at 28 days
19.69
at 28 days
at 1 day
at 1 day
at 1 day
at 1 day
at 1 day
at 1 day
at 1 day
The shrinkage test specimens (cylinder) were fitted with a device to measure the
internal temperature of the concrete in order to ensure that the strain measurement s
were not disturbed by thermal strains. It can be seen on Figure 16 that after 24 hours,
the temperature was stabilised in all the cylinders.
31
29
SCC40-AS
SCC40-TS
27
Te
mp
era 25
tur
e 23
(°
C) 21
SCC70-AS
SCC70-TS
19
17
0
1
2
3
4
5
6
Time (in days)
Fig. 16 - Temperature of the concrete at early age in the Ø 16 x 100 mm cylinders
21
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 22 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
AS = autogeneous shrinkage (wrapped + sealed)
TS = total shrinkage (in air)
The following Figures 17 and 18 show the results obtained after one year of test.
-700
-600
Strain (µm/m)
-500
SCC40-AS
-400
SCC40-TS
-300
-200
-100
0
0
100
200
300
400
Time (days)
-700
-600
Strain (µm/m)
-500
SCC70-AS
-400
SCC70-TS
-300
-200
-100
0
0
100
200
300
400
Time (days)
Fig. 17 - Strain due to shrinkage without and with desiccation
22
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 23 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
-1200
Strain (µm/m)
-1000
-800
SCC40-BC
SCC40-TC
-600
-400
-200
0
0
100
200
300
400
Time (days)
-1200
Strain (µm/m)
-1000
-800
SCC70-BC
SCC70-TC
-600
-400
-200
0
0
100
200
300
400
Time (days)
Fig. 18 - Strain measured in creep tests with and without desiccation
Note: strain due to shrinkage is deducted, but elastic strain due to loading is included
in the results presented.
5.7 Water absorption of near-surface concrete
It is now recognised that the transport properties of near-surface concrete play a major
role in the durability of reinforced concrete. This is because most deterioration
processes affecting concrete structures involve the transport of aggressive agents (eg.
water, pure or carrying aggressive ions, carbon dioxide and oxygen) into the concrete
and it is usually the case that the more resistant the concrete is to ingress, the more
durable it will be [6]. The transport characteristic of the near-surface concrete is
controlled by three mechanisms/processes, namely, capillary absorption, permeability
and diffusion. As a starting point the ability of near-surface concrete to absorb and
transmit water by capillary action is determined and serves as an indirect measure for
concrete durability performance in this study.
A recent RILEM recommendation, RILEM TC116-PCD: Permeability of Concrete as
a Criterion of its Durability - C: Determination of the capillary absorption of water of
hardened concrete, 1999, was adopted and modified for this study [7]. The
experimental set-up for the capillary absorption test is shown in Figure 19. The
moulded side faces of 150 mm cube specimens were tested, instead of the moulded
bottom face used in the RILEM recommendation. The specimens were cured in water
23
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 24 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
for six months and then put into an oven at 50-55o C for four weeks as
preconditioning. After cooling, the specimens were prepared and tested according to
the test procedures described in the RILEM recommendation. The uptake of water by
capillary absorption is measured through the weight of the specimens at time intervals
of 10 minutes, 0.5 h, 1 h, 2 h, 4 h, and 24 hours of contact with water.
Fig. 19 - Experimental set-up for the capillary absorption test (RILEM TC116-PCD)
The test results for the SCC and references mixes, calculated as water absorption per
unit area of the test surface i (g/m²), for the respective suction periods, t (minute or
hour), are presented in Figure 20 and Table 11. The straight lines in Figure 20 indicate
that there exists a relation of the form: i = St0.5. For the absorption of water, 1 g in
mass is equivalent to 1000 mm³ in volume, thus S (mm/min.0.5 or mm/hour0.5), usually
called sorptivity, can be obtained from the slope of the i vs. t0.5 curve, as described
elsewhere [8]. The results for sorptivity, which is an indicator for the rate of capillary
water absorption of the test concrete surface, are also given in Table 11.
24
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 25 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
Capillary absorption, g/m²
7000
6000
5000
RC
SCCC
SCCH
FSCC
RH
4000
3000
2000
1000
0
0.00
10.00
20.00
30.00
40.00
Time, minutes^1/2
Fig. 20 - Capillary water absorption of concrete surface
Table 11 - Results of capillary absorption of water on concrete surface
Concrete
mixes
RC
SCCC
RH
SCCH
FSCC
Water absorption per unit area of the test surface, g/m²
10 min
30 min
1 hour
4 hours
24 hours
576.5
412.7
752.8
391.1
217.7
867.1
571.4
1192.6
613.3
349.2
1194.5
734.7
1614.4
795.6
471.6
2292.2
1229.0
2956.6
1413.3
1002.2
4584.3
2789.0
6616.2
3173.3
2602.9
Sorptivity,
mm/h0.5
0.8895
0.5282
1.298
0.6148
0.5361
The results in Figure 20 and Table 11 clearly indicate that the capillary absorption
was considerably higher for the reference concrete mixes than for the SCC mixes of
similar compressive strength or w/c ratio. Comparing all the concrete mixes, the rate
of water absorption, indicated by the magnitude of sorptivity was in the order of RH >
RC > SCCH > FSCC > SCCC. Such results suggest that the near-surface concrete
was denser and more resistant to fluid ingress in the SCC mixes than in the
corresponding reference mixes. This is likely to be mainly due to the relatively lower
water-powder ratio of all the SCC mixes, as well as filling effects of the fine
limestone powder in the SCCH and FSCC mixes and the increase of cement hydration
products owing to the GGBS in the SCCC mix. Other factors, such as the better
dispersion of cement and powder particles and better water retention in the fresh SCC
mixes may also contribute to their lower near-surface absorption.
25
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 26 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
5.8 Carbonation
As stated earlier, the durability of the hardened concrete was to be assessed through
external exposure of one column and one beam for each concrete mix. Although this
will provide reliable results, the durability performance (or any deterioration) has to
be monitored over a long period. Since the length of the exposure at this time is
evaluated from 8 to 15 months only it is still too early to obtain a meaningful
evaluation of durability. Assessment of extent of the carbonation carried out at this
stage is only intended to serve as a starting point for further monitoring.
With respect to durability, the importance of carbonation lies in the fact that it reduces
the pH value of the pore solution in hardened Portland cement paste, from above 13 to
a value of about 9. The reduction of pH value to a critical value can result in depassivation of the protective oxide film on the steel and thus the initiation of
corrosion. In this study, the measurement of carbonation was determined using the
common phenolphthalein test. The test procedure is described in RILEM
recommendations CPC-18, Measurement of hardened concrete carbonation depth,
1988 [9].
A limited number of readings on the carbonation depth at different exposure periods
was obtained and is presented in Table 12. It is impossible to draw a definite
conclusion from the very limited results at this early stage. At one year, the C35
grade SCCH and RH mixes both showed a 2 mm carbonation depth, while the C60
grade SCCC and RC showed no sign of carbonation. For the FSCC mix, a
carbonation depth of around 0.5 mm was observed at 8 months.
Table 12 - Depth of carbonation of concretes exposed to outdoor environment in UK
Carbonation depth, mm
Exposure
period
6-9 months
10-12 months
13-15 months
RH
0.5
N/A
2.0
Concrete mixes used
SCCH
RC
SCCC
0.5
0
0
N/A
0
0
2.0
---
FSCC
0.5
---
5.9 Freeze-thaw resistance
Two prisms measuring 75 x 75 x 300 mm were cast for each concrete for evaluating
freeze-thaw durability. The specimens were demo ulded after 24 h and cured in limesaturated water at 20 ± 2 ºC for 28 days, then subjected to freeze-thaw cycles. The
prisms were monitored for mass loss and change in dynamic modulus of elasticity.
The daily freeze-thaw cycle consisted of storing the specimens in a freezer at -30 ºC
for 18 hr, then at room temperature for 6 hr to 7 hr.
The results of the freeze-thaw durability tests are shown in Fig. 28. It should be
emphasised that the concrete was not air entrained. There has been no loss of mass to
date in any of the mixes subjected to the freeze-thaw cycles. The civil engineering
mixes, both SCCC and the RC show no drop in ultra-sonic pulse velocity (UPV %),
26
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 27 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
and exhibit better resistance than the two housing concrete mixes. After 125 cycles,
the UPV of SCCC was reduced by only 5%.
Considering the housing mixes, it was observed that the SCC mix exhibited lower
resistance to freeze-thaw than the reference mix. The UPV values of the SCC and
reference mixes after 125 cycles of freeze-thaw were reduced by 71% and 48%
respectively.
After 95 cycles of freeze-thaw, the UPV of FSCC mix was reduced by 32% and
exhibited lower resistance than the civil mixes.
Average UPV (m/s)
6000
SCCH
5000
RH
SCCC
4000
FSCC
RC
3000
2000
1000
0
0
20
40
60
80
100
120
140
160
Number of freeze-thaw cycles
Fig. 21 – Reduction in ultra-sonic pulse velocity (UPV) of all mixes
6. PROPERTIES
OF
HARDENED
CONCRETE
IN
FULL-SIZE
STRUCTURAL ELEMENTS
To permit the range of comparisons required, three full-size beams and/or columns
were cast for the mixes, as follows: SCCH – columns only, RH – beams and columns,
SCCC beams and columns, RC – beams and columns. FSCC – beams only. One of
the three was loaded to failure post-28 days, one was used for in-situ testing, and one
has been left at an outdoor exposure site for future assessment of durability.
6.1 In-situ testing
The quality of the in-situ concrete in the 3-metre high columns and 4-metre long
beams was assessed in three ways:
• Core test - assesses in-situ compressive strength
• Rebound (Schmidt) hammer test - assesses near-surface quality and uniformity
• Pull-out (Lok) test - assesses near-surface quality and uniformity
The full-scale SCC columns were cast by pouring the concrete directly from the top of
the formwork with a free fall of up to 3 metres through the reinforcement (see
Fig. 36). The casting of the SCC beams the concrete was poured from one end and
allowed to flow to the other over the distance of 3.8 metres. To examine if SCC
mixes achieve adequate compaction, and similar uniformity in properties within the
27
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 28 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
element to those obtained with properly compacted reference concrete mixes, tests
were carried out at the top, bottom and middle section for the columns, and both ends
and the middle section for the beams.
6.1.1 Test arrangements
Core tests Tests were carried out according to procedures described in BS1881: Part
120 - Method for determination of the compressive strength of concrete cores. Three
100-mm diameter horizontal cores were taken at each test location within selected
columns and beams. The cores were then cut to the required length and their ends
capped. The prepared concrete cores were tested at the age of roughly 3 months, and
the results were calculated and expressed as estimated in-situ cube strength.
Rebound (Schmidt) hammer test
Tests were carried out according to the procedures described in Draft European
standard prEN12398: Testing concrete – Non-destructive testing - Determination of
rebound number. A digitised apparatus, namely “Digi-Schmidt hammer”, which
automatically records the rebound number for each testing was used in this study.
Testing was carried out at the ages of 7 days and 28 days, and a minimum of 40
readings was taken at each location along the height of columns and along the length
of beams.
Pull out (LOK) test
The Pull out (LOK) test has attracted special attention in recent years because of the
excellent correlation obtained between pullout force and standard compressive strength.
The system is flexible and produces only a minimal destruction to the structure. The
LOK test was selected in this study to provide direct information on the quality of cover
concrete and its uniformity in the structural element, i.e. along the height of columns and
the length of beams.
The test method has been described and standardised in ASTM (ASTM, C-900),
British (BS 1881, part 207, 1992), International (ISO/DIS, 8046), Swedish (SS,
137238), Danish (DS, 423.31) and draft European standards (prEN 12399:1996). In
this study, 4 cast-in inserts were used at each test location (i.e. top, middle, bottom
section of a column, and end A, middle, end B section of a beam). Two of the four
inserts at each location were pulled out at 7 days and the other at 28 days, and the
pullout forces were recorded (Fig. 22).
28
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 29 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Fig. 22 - Lok test
6.1.2 Results and discussions of in-situ compressive strength
The core test results were expressed as estimated in-situ strengths in accordance with
British Standard practice to allow for difference in core dimensions and the presence
of steel bars. The results of the core test at the age of 3 months for the selected
columns and the selected beams are plotted in Figures 23 and 24 respectively. To
permit direct comparisons of the strength variation within columns and beams
between the different grades of concrete, the results have also been expressed as
percentage s of the top section result for columns, and end A result for beams. End A
is the casting end of the beams. These variations are shown in Figures 23 and 24
respectively.
For the in-situ strength variation in columns , Figures 23 and 25 demonstrate higher
strengths at the bottom than the top, as is usual with traditional compacted concrete.
The general profile of the results is similar in all the cases, with the exception of the
RC mix which shows a lower result in the middle section. The results also show that
the strength variation within the column is less significant for C60 strength grade
mixes (both SCCC and RC) than for C35 mixes ( both SCCH and RH). Comparisons
of the relative results in Figure 25 suggest that the variation of strength in both
strength grades is smaller in the SCC mixes than in the corresponding reference
mixes.
29
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 30 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
d from bottom
of column (mm)
3000
Top
RH
2500
RC
SCCH
SCCC
2000
1500
Middle
1000
500
Bottom
0
20
30
40
50
60
70
80
Estimated in-situ cube strength (MPa)
In-situ compressive strength, MPa
Fig. 23 - Variation of in-situ compressive strength along height of columns
80.0
End A
70.0
Middle
End B
SCCC
RC
FSCC
60.0
50.0
40.0
30.0
RH
20.0
0
1000
2000
3000
4000
D from end A of a beam, mm
(End A being casting point for SCC mixes)
Fig. 24 - Variation of in-situ compressive strength along length of beams
30
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 31 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
d from bottom (mm)
3000
Top
2500
2000
Middle
1500
RC
1000
SCCC
RH
500
SCCH
Bottom
0
90
95
100
105
110
115
120
In-situ strength as % of top section of column
Fig. 25 - In-situ compressive strength as % of strength at top section of column
In-situ strength as % of end A
120
RH
RC
115
SCCC
FSCC
110
105
100
95
End A
Middle
End B
90
0
1000
2000
3000
D from end A of a beam, mm
4000
Fig. 26 - In-situ compressive strength as % of end A of beam
Compared to the strength variations along height of columns, results in Figures 24 and
26 demonstrate that the strength variation along the length of beams was much
smaller. For example, a difference of strength up to 15% was observed in the RH
column, while the maximum strength difference in beams was 7%. The general
profile and relative variation of the results for beams are also similar for all the mixes
studied.
31
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 32 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
In-situ strength as % of 28
day cube strength
Comparisons of the in-situ strength re sults with the corresponding 28-day
standard cube strengths can also be made by plotting the relative results, as
presented in Figures 27 and 28 for columns and beams respectively. Results in Figure
26 indicate that the in-situ strength achieved in the columns (all mixes) varies
between 80-100% of the standard cube strength, which is above the average value of
65% reported for such elements [10]. Two other points may be noted: firstly, that the
in-situ strength of the SCCH mix is closer to the 28-day cube strength than that of the
corresponding reference mix RH in the C35 grade; and secondly, in the C60 grade,
that there is no significant difference in the in-situ strength of the two mixes.
105
SCCH
SCCC
RH
RC
95
85
75
top
middle
bottom
Location of cores in column
In-situ strength as % of 28-day
standard cube strength
Fig. 27 - In-situ compressive strength in columns relative to standard 28-day strength
110
RH
RC
SCCC
FSCC
100
90
80
70
End A
Middle
End B
Location of cores in beam
Fig. 28 - In-situ compressive strength in beams relative to standard 28-day strength
32
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 33 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Results in Figure 28, however, show a different picture from that in Figure 27. For
RH and SCCC mixes, the in-situ strength in beams relative to the 28-day cube
strength was significantly lower than in columns cast at the same time. The curing
conditions for individual elements explain the difference in performance. The
columns were cast inside the laboratory building with a relatively constant
temperature of around 20o C all year round, while the beams were cast outdoors with
a range of temperatures depending on the month of the year. The reference beams
were compacted by poker vibrators and cast outdoors to simulate in-situ concrete
construction process (the noise levels were measured in Task 8.5). The casting of RH
mix and SCCC mix was carried out in January and March respectively with an
average temperature of 5 - 10o C, while the RC and FSCC mixes were cast in June and
July with temperature sometimes over 20 degrees in the daytime. Taking the different
casting temperatures into account, the results for the beams appear to be reasonable
and generally in agreement with the results for the columns as the average relative
strength value reported for such elements was 75% [10].
Core testing of larger pours was also carried out by one of the project partners in
Sweden. Although the core results were not directly traceable to the standard cube
results from the pours, they complied with that element in the Swedish standard which
requires a minimum relationship between cores and cubes.
6.1.3 Results and discussions of the other in-situ tests
Results of the Lok test pullout load (Table 25 in appendix) and the Schmidt hammer
rebound number were analysed in the same manner as for the in-situ strengths, and
are presented in Figures 29 and 30 for the test beams and columns respectively.
Results in Figures 29 and 30 demonstrate that the results of the Schmidt hammer and
Lok test tests generally confirm the pattern of results observed for the cores. The
near-surface concrete quality increased towards the bottom of columns, but the
difference between top and bottom of the columns was more pronounced in the
housing mixes (i.e. C35) than in the civil mixes (i.e. C60). The results also indicate
that the concrete properties were less variable along the length of beams than along
the height of columns, particularly for the housing mixes. Furthermore, any
difference in uniformity of the properties of the SCC and the corresponding reference
mixes was not statistically significant, with the uniformity for the SCC mixes being
marginally better.
33
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 34 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
Relative results as %
of end A of beam
Relative results as %
of end A of beam
2000-04-26
core strength
rebound number
120
115
core strength
rebound number
pull-out load
120
115
RH
110
105
110
105
100
100
95
95
90
85
120
115
110
105
100
95
90
85
pull-out load
RC
90
85
120
115
110
105
100
95
90
85
FSCC
End A
Middle
SCCC
End B
End A
Middle
End B
Location of test points in beam
Location of test points in beam
Fig. 29 - Variation of in-situ properties of SCC and reference mixes in beams
Relative results as % of Relative results as % of
top section of column
top section of column
core strength
rebound number
core strength
pull-out load
120
115
120
115
SCCH
110
110
105
105
100
100
95
95
90
90
120
120
115
115
110
pull-out load
rebound number
RH
110
105
105
100
100
95
95
90
SCCC
RC
90
Top
Middle
Bottom
Location of test points in column
Top
Middle
Bottom
Location of test points in column
Fig. 30 - Variation of in-situ properties of SCC and reference mixes in columns
34
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 35 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
6.1.4 Statistical analysis of the test results
In sections 6.1.2 and 6.1.3, the test results were analysed and presented in terms of
variations along the height of columns and along the length of beams. It is essential
to assess the reliability of the results and to determine the statistical significance of
any differences of results within the columns or beams, particularly when the
variations of results at different sections are high. A significantly larger number of
test results on in-situ compressive strength and rebound number was acquired, and has
enabled limited statistical analysis of reliability and variation of the strength and
rebound number results to be made .
The average results of in-situ cube strength obtained from nine cores for each column
or beam, and their variations in the columns and beams for all the SCC and reference
concrete used are calculated and presented in Table 13. The statistical analysis tool ANOVA (analysis of variance) - was also used to determine the statistical
significance of differences of results along the height of column and along the length
of beam.
It was found that at the 5% significance level (related to 95% confidence), α=5%, the
differences of in-situ compressive strength at different sections of the column and
beam were not statistically significant. At the 10% significance level,α=10%,
however, the difference of in-situ strength along the height of column (ie. top, middle
and bottom sections) was statistically significant for the RH, SCCH and RC mixes,
while the differences for the SCCC column and all the beams remained statistically
insignificant.
Table 13 - Statistical analysis of in-situ compressive strength for columns and beams
In-situ
equivalent
cube strength,
MPa
RH
Mean
STD *
V, % **
Mean
STD
V, %
31.35
3.02
9.6
29.28
1.94
6.6
SCCH
44.60
2.96
6.6
N/A
N/A
N/A
RC
57.65
3.25
5.6
60.00
3.40
5.7
SCCC
71.82
4.50
6.3
66.04
5.83
8.8
FSCC
N/A
N/A
N/A
56.13
3.24
5.8
Results for columns
Results for beams
∑ [x − x]
2
*
STD - sample standard deviation (estimate s. d; s =
**
V - coefficient of variation
i
n −1
Similar analysis was applied to the results of rebound hammer testing. Forty readings
were taken at each location along the height of columns or along the length of beams.
Testing was carried out on two columns and two beams for each concrete mix, and all
the data points, including outliers were used in the analysis. The mean values and the
35
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 36 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
variations of rebound number along the height of the columns and along the length of
beams for all the SCC and reference concretes used are calculated and presented in
Table 14. A typical set of actual rebound number readings for the SCCH and the RH
concrete column is presented in Figures 31 and 32 respectively.
S-H rebound number
Coef. of variation, %: Top=7.3, Middle=8.1, Bottom=10.8
60
SCCH mix
50
40
30
Top
Middle
Bottom
20
0
10
20
30
40
Test points
Fig. 31 - Typical local variation in rebound number readings in SCCH columns
The ANOVA was also applied to all the test data and it was found that the differences
of rebound number along the height of column were statistically very significant (at
α=1% in most cases) for all the SCC and reference mixes.
For the variation along the length of beams, however, the differences were found to be
not significant even at the 10% significance level, α=10%.
S-H rebound number
Coef. of variation, %: Top=8.3, Middle=12.9, Bottom=15.7
60
RH mix
50
40
30
Top
Middle
Bottom
20
Test points
30
20
0
10
40
Fig. 32 - Typical local variation in rebound number readings in RH columns
36
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 37 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Table 14 - Statistical variation of rebound number for columns and beams at 28 days
Results for column
In-situ rebound
number
RH
SCCH
RC
SCCC
Mean
V, %
Nb of test
points
Top, A*
34.8
7.8
160
Middle
37.3
11.0
160
33.7
8.5
80
Bottom, B**
40.8
15.0
160
33.6
8.4
80
Top
35.7
7.1
160
Middle
35.7
7.7
160
N/A
N/A
N/A
Bottom
40.8
11.3
160
Top, A
44.1
6.9
160
41.4
6.0
80
Middle
46.5
6.9
160
42.3
4.4
80
Bottom, B
46.6
5.5
160
42.4
6.7
80
Top, A
45.0
5.2
120
39.8
7.6
80
Middle
46.0
4.7
120
40.5
8.6
80
Bottom, B
46.3
4.6
120
40.2
5.6
80
42.9
7.5
80
43.3
7.5
80
42.3
5.7
80
End A
FSCC
Middle
N/A
N/A
End B
*
**
Results for beam
Nb of
Mean V, %
test
points
33.3
9.7
80
N/A
Top, A - indicating location of test points were eithe r near the Top of the
columns, or near the end A (ie. the casting end for SCC mixes) of the beams
Bottom, B - indicating location of test points were either near the Bottom of
the columns or near the end B of the beams
In general, the statistical analysis in this section confirmed the validity/reliability of
the results discussed in Sections 6.1.2 and 6.1.3.
It is clear from the outcome of the ANOVA that the differences of in-situ strength and
surface “hardness” in different sections of beams were not statistically significant for
either the SCC or the reference mixes. The differences along the height of columns ,
however, were statistically significant for all the mixes; this is a normal characteristic
of concrete.
The results also suggest that variation of results was generally greater in the C35
mixes than in C60 mixes for both SCC and references mixes.
37
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 38 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
In terms of the differences between the SCC and reference elements, the actual level
of variation of results within columns and beams was similar, the only observation
being that the variations of results within the reference columns was slightly higher
than in the SCC columns.
6.1.5 Conclusions
The quality of the in-situ concrete properties in the columns and beams was assessed
by testing cores for in-situ strength, and pullout of pre-embedded inserts and rebound
hammer number, for uniformity in near-surface properties. From the overall test
results and analysis, the following general conclusions could be drawn.
1) In-situ strength achieved in the columns for all the SCC and reference mixes was
in the range of 80-100% of the 28-day strength for standard cube specimens, the
percentage being generally higher for the C60 grade mixes (SCCC and RC) than
for the C35 grade mixes (SCCH and RH). This is well above the average value of
65% reported for such elements.
In-situ strengths achieved in the beams for the SCC and reference mixes were
affected by the different ambient conditions at casting, but all were above the
reported average of 75% for such elements
2) The SCCH mix containing limestone powder achieved a significantly higher insitu strength than the corresponding reference RH mix, probably due to the
accelerating, and densifying effect of the filler.
3) The in-situ strength and near surface quality of the concrete increased along the
height of the columns towards the bottom, the difference being greater in the C35
housing mixes than the C60 civil engineering mixes. These differences were
statistically significant. The variation of concrete properties was statistically
insignificant along the length of 3.8-metre beams.
4) There were no significant differences in uniformity of in-situ properties between
the SCC mixes and the corresponding reference mixes; the properties of the SCC
mixes were marginally more uniform.
6.1.6 Micro-mechanical study of the interfacial zone
The interfacial transition zone (ITZ) between cement paste and aggregate (or fibre)
and around steel reinforcing bars has been an area of particular interest associated
with engineering and durability properties of cementitious composites and structural
reinforced concrete. As a result of its more porous and heterogeneous/anisotropic
nature, the ITZ was recognised to be the 'weak link' in cement and concrete
composites and to have a considerable influence on the engineering and durability
properties. However, it has been very difficult to assess the micro-mechanical
properties of the ITZ in real concrete specimens, as it requires specialised techniques
and apparatus.
At Paisley, a unique depth-sensing microindentation method was previously used for
micro-strength testing and fibre push-in testing to assess interfacial and bond
properties in glass fibre reinforced cementitious composites. Significant advances
have been made recently in improving the micro/nano-indentation test method and in
theoretical analysis of the test data, and this has allowed assessment of micromechanical properties, such as elastic modulus and micro-strength within the ITZ to
38
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 39 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
be made. The depth-sensing microindentation method now offers great advantages
over the conventional Vickers microhardness testing.
In this preliminary study, the depth-sensing microindentation method was applied to
investigate the ITZ in structural reinforced concrete, particularly the mechanical
properties of the ITZ above and below a horizontal steel reinforcing bar. The
specimens extracted from the SCCC and RC beams were selected for the
investigation, thus enabling possible a comparison of the ITZ properties of the SCC
and the corresponding reference mixes.
Specimens for the microindentation testing were prepared by first drilling φ 100 x
200 mm cores horizontally from the middle section of the beam, the cores intersecting
the three φ 16 reinforcing bars of the upper layer in the tensile zone. Small specimens
(roughly φ 35 x 20 mm) with only one steel bar at the centre were then extracted from
the cores, using a diamond saw. This was then followed by resin embedding,
precision sectioning, grinding, polishing and ultrasonic cleaning to obtain the final
specimens (φ 40 x 15 mm) for the microindentation test.
6.1.6.1 Measurement of elastic modulus and micro-strength
All experiments were performed using a depth-sensing microindentation apparatus,
MicroTest 200, available at the University of Paisley. The load and depth (or
displacement of the indenter) were continuously monitored during a programmed
microindentation load cycling. The operating principle and special features of the
apparatus were described in detail elsewhere [11]. A relatively sharp 90o diamond
indenter/probe in the shape of a corner of a cube was selected for this study so that
smaller but deeper indentations could be made. This would also make the test results
less sensitive to the surface effects of imperfect specimen preparation (e.g. roughness
and hardening/densification).
A typical outcome of the microindentation testing is an indentation load-depth
hysteresis curve as shown in Figure 33. As a load is applied to an indenter in contact
with a specimen surface, an indent/impression is produced which consists of
permanent/plastic deformation and temporary/elastic deformation. Recovery of the
elastic deformation occurs when unloading is started. Determination of the elastic
recovery by analysing the unloading data according to a model for the elastic contact
problem leads to a solution for calculation of elastic modulus E and also microstrength/hardness H of the test area. Details of the theoretical background and
methodology for the elastic modulus determination have been reviewed and presented
elsewhere [12,13].
39
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 40 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
LOAD, P
Pmax
loading
unloading
hp
S
hmax
DISPLACEMENT (DEPTH), h
Fig. 33 - A schematic diagram of an indentation load vs displacement curve
Briefly, the specimen elastic modulus is determined using equations (1) and (2):
S=
dP 2
=
Er A
dh
π
(
) (
1 − v i2
1
1− v 2
=
+
Er
E
Ei
(1)
)
(2)
Where, S = dP/dh is the experimentally measured stiffness of the upper portion of the
unloading data; Er is a reduced elastic modulus defined in equation (2); A is the
projected area of the elastic contact; E and v are Young's modulus and Poisson's ratio
for the specimen; and Ei and vi are the same parameters for the indenter. For the
diamond indenter used in this study, Ei = 1141 GPa and v = 0.07. The projected area
A can be derived from the plastic depth hp obtained using the unloading data and the
indenter shape function, which is dependent on its geometry. For an ideally perfect
900 (corner of a cube) indenter, A = 2.6hp 2 . In the case of imperfect indenter tip
geometry, the shape function can be determined by using electron microscopy
techniques or through calibration of a hardness (or elastic modulus) - plastic depth
curve using a homogeneous material of constant hardness (or elastic modulus). A
special defined parameter - microstrength, H = Pmax/A - can also be calculated,
where Pmax is the maximum load applied. When the Vickers indenter is used the
microstrength is equivalent to the Vicker microhardness.
40
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 41 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
6.1.6.2 Results and Discussions
In this study all testing was programmed in such a way that the loading started when
the indenter came into contact with the test surface and the load increased at a
constant rate of 2.3 mN/s until a depth of penetration of indenter into the specimen
reached a specified value of 10 µm. Then the load was held at its maximum for 10
seconds before unloading at the same constant rate. At completion of unloading, the
specimen was retracted and moved to another test point. With the maximum depth
setting of 10 µm the width of the indent/impression after the load removal was usually
about 20 µm or less. The distance between two adjacent test points was thus set to be
60 µm or greater, to avoid possible overlapping of the areas affected.
Photographs of a test specimen and a magnified tested interfacial area are shown in
Figure 34. To study the interfacial zone around the horizontal reinforcing bar in both
SCC and conventional concrete mixes, a large number of indentation tests were
carried out in the steel-concrete interfacial areas at the top and the bottom sides (ie.
above and below) of the steel bar. Since sand particles also existed in the tested area,
the test points which lie closer to the aggregate particles than to the steel bar were
considered invalid and discarded. A typical set of results is presented in Figure 35 as
plots of modulus of elasticity and microstrength versus distance from the interface.
MICRO-INDENTATION TESTING
MICROINDENTATION
Test point 1
Top
CONCRETE
STEEL BAR
Steel bar
Test point 2
Distance between test points = 80 µ m
Bottom
Test point 3
(a)
(b)
Fig. 34 - Photographs of the tested specimen:
(a) specimen of the indentation test, (b) magnified view of interfacial area after testing
41
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 42 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
1000
under steel bar
above steel bar
40.0
(a)
Microstrength, MPa
E-modulus, GPa
50.0
30.0
20.0
10.0
under steel bar
above steel bar
800
(b)
600
400
200
0
0.0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
Distance from
(b) the interface, um
Distance from(a)
the interface, um
Fig. 35 - A typical set of results versus distance from the interface for:
(a) Elastic modulus, (b) Microstrength
For all the specimens tested in this study, it was found that the micro-mechanical
property (i.e. modulus and strength) profiles showed a trough within the ITZ, with the
minimum values occurring at 10 - 30 µm from the actual steel interface. The values
for their properties then increased when test points moved towards the bulk concrete
matrix and became roughly constant at distances greater than 50 - 70 µm. It was also
noted that the results showed moderate variations even at the same distance from the
interface, and that interference of neighbouring aggregate/sand particles became more
frequent at distances greater than 60 µm from the steel interface. The interference of
steel bar was also present for test points adjacent to or less than 10 µm from the
interface.
For comparison of the micro-mechanical properties of different test areas of the ITZ
and evaluation of their variations, the test points were classified into two groups
according to their distance d, to the actual interface, namely 10 µm < d <30 µm, and
35 µm < d < 60 µm. The results for the 10 - 30 µm distance can be considered to
represent the minimum value within the ITZ, whereas the results for the 35 - 60 µm
distance may provide an indication of properties of the bulk matrix. Average results
and the standard deviations of modulus of elasticity and micro-strength for both the
SCC and reference mixes are presented in Table 15. The relative results of interfacial
properties below and above the steel bar are compared in Figure 36.
42
70
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 43 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
Table 15 - Modulus of elasticity and micro-strength in the ITZ around steel bar
Properties
Modulus of elasticity, GPa
Reference mix (RC)
Test details
SCC mix (SCCC)
Below steel bar
Above steel bar
Below steel bar
Above steel bar
10<d<30 µ m
9.8
17.6
12.9
15.0
St. deviation
4.0
4.4
4.5
5.5
35<d<60 µ m
17.6
21.5
25.2
24.4
St. deviation
5.1
5.9
4.6
3.8
Micro-strength, MPa
10<d<30 µ m
229
329
197
285
St. deviation
77
84
46
74
35<d<60 µ m
372
462
411
418
St. deviation
82
135
125
61
Interfacial properties below
steel bar as % of above the bar
E-modulus
Micro-strength
120
10 < d < 30 um
35 < d < 60 um
100
80
60
40
20
RC
SCCC
RC
Concrete mixes used
SCCC
Fig. 36 - Results of ITZ properties below steel bar relative to those above steel bar, %
43
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 44 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
The results in Table 15 and Figure 36 clearly indicate that the elastic modulus and
micro-strength of the ITZ were significantly lower on the bottom side of the steel bar
than on the top side. The difference was particularly pronounced in the 10 - 30 µm
distance from the actual interface, and for the reference mix. For instance, at the
distance of 10-30 µm, the interfacial E modulus and micro-strength of the reference
mix below the steel bar was only 56% and 70% respectively of those above the bar.
The corresponding results for the SCC mix were 86% and 69%, which suggests that
the ITZ properties were more uniform in the SCC mix than in the reference mix. This
may be due to the lower water content and higher powder volume in the fresh SCC
mix. The average results for the distance of 35-60 µm from the interface showed that
both properties on the bottom side achieved 80% of those on the top side for the
reference mix, whereas for the SCC mix almost identical values were obtained on
both sides. This was mainly due to the fact that for the reference mix the weak
interfacial zone was found to be wider on the bottom side than on the top side of the
bar. Such a phenomenon was not evident in the SCC mix.
6.1.6.3 Conclusions
1) The results obtained for both the SCC and the reference mixes revealed a
distribution of micro-mechanical properties within ITZ with a common profile,
namely with a trough, or a minimum, occurring at 10-30 µm from the actual steel
interface. The width of the ITZ was found to be between 50 and 70 µm, beyond
which the micro-mechanical properties became less variable.
2) The results indicated that the elastic modulus and the micro-strength of the ITZ
were significantly lower on the bottom side of a horizontal steel reinforcement bar
than on the top side. This supported the common perception that the ITZ is
weaker underneath a large aggregate particle or steel bar than above it, owing to
internal bleeding and settlement of particles during concrete placing. This effect
was particularly pronounced at the 10-30 µm distance from the actual interface.
3) The difference of ITZ properties between top and bottom side of the horizontal
steel bar seemed to be generally less pronounced for the SCC mix than for the
reference mix.
6.2 Structural performance of full-scale elements
6.2.1 Design
Columns
The full-size columns are shown in Fig. 37 and measured 300 x 300 x 3000 mm. Two
arrangements of links, Type A and B were adopted for housing and civil engineering,
respectively (Fig. 37).
The reinforcement details are shown in Table 16. All columns for housing and civil
engineering were made with the same percentage of longitudinal reinforcement
(ñg = 1.4 and 1.9% for Type A and Type B) and the same amount of lateral
reinforcement, ñh = 0.66% for link configuration Type A and ñh = 1.53% for
configuration Type B. Link spacings of 100 mm for housing and 75 mm for civil
engineering were used.
44
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 45 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Deformed steel bars were used for longitudinal reinforcement and plain undeformed
bars for the lateral reinforcement. Conventional strength steel was used for the
longitudinal and confining link steel reinforcement, with nominal yield values of 460
and 250 MPa, respectively.
Table 16 – Reinforcement details of columns
Column
Longitudinal reinforcement
ñg
(%)
fyk
(MPa)
Lateral reinforcement
Dh
(mm)
S
(mm)
Housing
(Type A),
Reference
and
SCCH
4 No16 + 4 No 12
1.4
460
10
100
Civil
(Type B),
Reference
and
4 No16 + 8 No 12
1.9
460
10
75
SCCC
fyk : Yield stress of steel reinforcement in compression (MPa)
ñh : Percentage of longitudinal reinforcement (%)
ñh : Percentage of lateral reinforcement (%)
Dh : Nominal diameter (mm)
S : Stirrup spacing (mm)
ñh
(%)
fyk
(MPa)
0.66
250
1.53
250
Beams
Two configurations of reinforcement bars were used for housing and civil
engineering. Beams for housing are reinforced with 2 bars of 16 mm diameter for
tension, 2 of 12 mm for compressive and links spaced at 250 mm. Beams for civil
engineering are reinforced with 6 bars of 16 mm diameter for tension and links spaced
at 180 mm. Details of reinforcement are shown in Fig. 38. The same configuration of
reinforcement bars used for housing was used for the steel fibre mix, with links
spaced at 300 mm.
45
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 46 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Structural elements : COLUMNS
450 mm
250 mm
Section of housing column – Type A
300 mm
40 mm
Link Φ = 8 mm,
spacing = 100 mm
300 mm
40 mm
4 Φ = 12 mm
4 Φ = 16 mm
3m
300 mm
2500 mm
Section of civil engineering column – Type B
300 mm
40 mm
Link Φ = 8 mm,
spacing = 75 mm
300 mm
40 mm
8 Φ = 12 mm
4 Φ = 16 mm
250 mm
300 mm
Fig.3
Fig. 37 - Details of reinforcement and dimensions of the columns
46
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 47 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
BEAM
3800 mm
200 mm
Section of civil engineering
Steel stirrup Φ = 8 mm,
Spacing 180 mm
b = 200 mm
Steel compression
reinforcement, Φ = 16 mm
Steel tensile
reinforcement,
3 + 3 Φ 16 mm
h = 300 mm
50 mm
25 mm
25 mm
Section of housing
b = 200 mm
Steel compression
reinforcement, Φ = 12 mm
Steel tensile
reinforcement, Φ = 16 mm
h = 300 mm
40 mm
Steel stirrup Φ = 8 mm,
Spacing = 250 mm RH and SCCH
Spacing = 350 mm for Fibre mix
40 mm
Fig. 38 - Details of reinforcement and dimensions of beams
47
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 48 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
6.2.2 Concrete placing procedures
Columns . The set up for the placing is shown in figures 39-41.
The reference columns were cast in three lifts and compacted using an external
vibrator. In the case of the SCC columns, the concrete was cast through a hopper at
the top of the formwork without internal or external compaction (Fig. 40). An acrylic
plate was provided on one side of the formwork of one of the three columns to permit
observation of concrete flow during the placement (Fig. 41).
Fig. 39 – Columns with reinforcement detail prior to pour.
Note: the gaps in the reinforcement of the left-hand column are to permit subsequent
access for coring.
Fig. 40 – Casting SCC with skip from the top of formwork
48
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 49 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
Fig 41 - Acrylic plate used for central column to provide observation of flow of
concrete during the placement
Beams . The reference civil beam was cast and consolidated using a poker vibrator. In
the case of the SCC beams, the concrete was poured directly into one end of the
formwork from the discharge chute of the truck mixer, and allowed to flow along the
length of the beam (Fig. 42).
Fig. 42 – Casting SCC beams
49
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 50 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
6.2.3 Test set-up for columns
Thin layers of cement were used as capping over the top and bottom ends of each
column to ensure parallelism of specimen end surfaces and uniform distribution of the
load during testing. To ensure that the failure would occur in the central instrumented
region on the column, the tapered ends were further confined with bolted boxes made
from thick steel plates. Two hydraulic jacks, each with maximum compressive load
capacity of 4000 kN were used. The axial displacement of specimens was recorded
using three LVDTs located at three corners of the column and attached to upper and
lower clamped steel collars with vertical bars fixed to the gauge length of 100 mm.
The columns were subjected to monotonic concentric axial loading until failure, to
determine the overall load carrying capacity, and hence the overall in-situ
compressive strength (Fig. 43).
Fig. 43 – Load testing of column
6.2.4 Test set-up for beam
The beams were instrumented with a LVDT at mid-span to monitor deflection. The
beam, spanning 3800 mm, was then subjected to four-point flexural testing as
illustrated in Fig. 44. An automatic data acquisition system was used to monitor
loading and mid-span deflection. The load was applied step-by-step to the beam at a
rate of 5 KN per step by means of two 220 kN-hydraulic jacks and measured with a
load cell (Fig. 45). At the end of each step, cracks were sketched and crack-width
measured using a microscope (Fig. 46).
50
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 51 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
1.1 m
P
P
3.3 m
3.8 m
Fig. 44 - Four point-flexural test on beams.
Fig. 45 – Set-up for full-scale test of beam
Fig 46 – Measurement of crack-width
51
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 52 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
6.2.5 Testing of columns
The experimental results obtained from the testing of four reinforced columns are
shown in Table 17, and Figure 47 shows the axial load versus strain curves of the four
confined columns.
Table 17 – Compressive test results of columns
Columns
Axial load
AC
Housing
Reference
88743
SCCH
88743
Civil Eng.
Reference
88291
SCCC
88291
Po = 0.67 f’cu Ac + fy Asc
where:
Age
Pmax (kN)
Pmax /Po
90 d
90 d
3420
3020
2.07
1.18
6 months
5203
5414
1.40
1.35
6 months
Ac = b2 - Asc
Po = calculated ultimate capacity of the column
f’cu = in-situ compressive strength of the concrete (note: air cured strength
used for columns)
fy = yield stress of steel reinforcement in compression
Ac = area of concrete (net)
Asc = area of embedded steel
6000
Housing section
300 mm
40 mm
5000
Column axial load, P (KN)
D = 8 mm
(S=100 mm)
SCCC
RC
4 D = 16
mm
300 mm
4000
40 mm
4 D = 12 mm
RH
3000
SCCH
Civil section
300 mm
2000
40 mm
D = 8 mm,
(S=75 mm)
300 mm
1000
40 mm 8 D = 12 mm
4 D = 16 mm
0
0
0.0005
0.001
0.0015
0.002
0.0025
Concrete axial strain
Fig. 47 – Comparison of concrete axial load vs. axial strain for columns
52
0.003
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 53 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
In this figure, Pmax is the maximum load sustained by the reinforced columns; Po is the
calculated ultimate capacity of the column (Po = 0.67 f’cu Ac + f y Asc, the compressive
axial strength of the column calculated in accordance the BS 8110 standard).
A number of observations may be made:
•
•
•
•
The maximum load, Pmax, varied from 3020 to 5415 kN. In all cases this exceeded
the design load.
The ratio Pmax/Po was lower in the SCC mixes, though only slightly so in the case
of the civil engineering category.
Examining the behaviour of the C60 (civil engineering) columns, the high
observed toughness of both SCC and reference is due to the close spacing of the
lateral confining reinforcement, which delays the buckling of the longitudinal bars
and permits the full development of the steel strength of the links. The maximum
load of the SCCC column was slightly higher than the RC column (reference)
(5415 kN vs. 5203 kN), a consequence of the higher compressive strength of
concrete at the time of testing (66 MPa vs. 61 MPa). The SCCC column did not
behave in the same manner as the RC column in terms of initial stiffness and axial
strain at the maximum load. The RC exhibited considerably greater toughness
than the SCCC column.
Examining, now, the behaviour of the C35 (housing) mixes, the maximum load of
the SCCH was slightly lower than the RH column (3020 kN vs. 3420 kN) but both
columns exhibited similar behaviour in terms of initial stiffness, axial strain at the
maximum load, and toughness.
6.2.6 Testing beams
The flexural behaviour of reinforced SCC and conventionally-compacted beams was
investigated in terms of cracking behaviour, load carrying capacities and mode of
failure, and load deflection response.
6.2.7 Cracking patterns and cracking-spacing
Typical cracking patterns of beams at ultimate load are shown in Figure 48. At the
service load level, cracking in the flexural span is composed predominantly of vertical
cracks perpendicular to the direction of maximum principal stress induced by pure
moment. Cracking outside the pure bending zone starts similarly to flexural cracking
but, as the load is increased, other cracks are formed and as shear stresses become
important, more inclined cracks appear. Table 18 presents the average crack-spacing
at 90% ultimate load. The average crack-spacing of SCC civil and RC is around 80
and 160 mm. The crack-spacing of FSCC was higher because the link spacing of this
beam was 350 mm compared to 250 mm which was used for other beams.
53
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 54 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
Table 18 – Average crack-spacing, mm
Beam
Average crack spacing at
90% Multimate
RH
RC
SCCC
FSCC
138
160
82
270
Fig. 48 – Typical failure mode of SCC beam
6.2.8 Cracking moment
Experimental and theoretical results for moments at the first cracking are given in
Table 19. The theoretical cracking-moment, Mcr, is computed in two different ways,
(a) and (b) below:
[a]
Mcr = (fr I) / yt
Where:fr is the modulus of rupture of concrete (see [b])
I is the moment of inertia of beam section
yt is the distance from the centroid to extreme tension layer of the section.
Values in Table 19 entitled Mcr-th1 are calculated by taking into account the
moment of inertia of all reinforcement (tension and compressive reinforcement) and
using the equation [2] according to principles in EC2 as well as CEB-FIP model Code
1990:
54
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 55 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
fr = 0.40 (f’cu)2/3
[b]
Where f’cu is the compressive strength of concrete on cubes). Those in Table 19
entitled Mcr-th2 represent the cracking-moment calculated as recommended by BS
code; that is neglecting the moments of inertia of all reinforcement and assuming the
section of the beam made only of concrete. Comparison between Mcr-th1 and Mcr-th2
shows that the effect of the reinforcement is not neglected and the computation of the
moment of inertia is very simplified when the reinforcement is neglected. The
average value of Mcr-th2 is seen to be close to Mcr-exp. Note that the experimental
cracking moment includes the self-weight of the beam.
Table 19 – Experimental and theoretical moments at first cracking (kN.m)
Beam
RH
RC
SCCC
FSCC
M cr-exp
6.6
23.1
22.0
15.4
M cr-th1
10.6
26.2
31.4
22.1
M cr-th2
9.6
20.3
24.4
19.9
6.2.9 Crack-width
Figure 49 shows the maximum observed crack-widths plotted against the applied
load. Comparing the beams in the civil engineering category, it can be seen that the
cracks in the reference beam are wider than those in the SCCC beam.
Crack-width, ( µm)
1400
FSCC
1200
1000
RC
800
600
400
200
SCCC
0
0
50
100
150
200
250
Maximum applied moment, (KN.m)
Fig. 49 – Variation of crack-width vs. applied load
55
300
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 56 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-04-26
Table 20 presents the average-crack width at different load levels. At the same
moment (< 110 kN.m), the cracks in the fibre FSCC beam are less wide than those in
SCCC and RC beams. This difference in crack-widths can be attributed to the use of
the fibre in concrete, which affects the mechanism of development of cracks.
Table 20 - Average crack-width, ìm at different load levels
Beam
RC
SCCC
FSCC
Moment
88 kN.m
110 kN.m
100
160
100
120
28
40
66 kN.m
100
55
20
132 kN.m
180
120
1200
6.2.10 Load-deflection response
Figure 50 traces typical experimental load-deflection of RC, SCCC, FSCC and RH.
Initially, the beams are uncracked and stiff. With further loading, cracking occurs at
mid-span when the applied moment exceeds the cracking moment Mcr, causing a
reduction in stiffness.
350
SCCC (6 months)
RC (4.5 months)
300
Applied moment (KN.m)
250
200
FSCC (3 months)
150
RH (6 months)
100
50
0
0
20
40
60
80
100
120
Midspan-deflection (mm)
Fig. 50 – Load-deflection of RH, FSCC, RC and SCCC beams
7. CONCLUSIONS
The workability requirements for successful placement of SCC necessitate that the
concrete exhibits excellent deformability and proper stability to flow under its own
weight through closely spaced reinforcement without segregation and blockage.
Insuring high stability is important to limit bleeding, segregation, and surface
settlement of concrete after placement and secure uniform properties of the hardened
56
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 57 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
concrete, including bond to embedded reinforcement. In general, the SCC exhibits
low yield value and adequate cohesiveness (moderate viscosity). In addition to slump
flow test used to evaluate deformability, the L-box and the Orimet with JRing test
should be used to evaluate the ability to achieve smooth flow through restricted
spacing without blockage.
The ultimate load of the SCC civil column was slightly higher than the reference civil
column (RC) given the higher compressive strength at the time of testing (66 MPa vs.
61 MPa). The SCC civil column did not seem to behave in the same manner as the
RC column in terms of initial stiffness and axial strain at the maximum load. The
difference could be due to the difference of elastic modulus and the presence of
greater density of microcracking in concrete that affects mostly tensile properties and
stiffness. The SCC housing column seemed to have the same behaviour in terms of
initial stiffness and axial strain at the maximum load.
Typical cracking patterns of SCC and reference beams at ultimate load are very
similar. With four point-flexural test of beam, at the service load level, cracking in
the flexural span is composed predominantly of vertical cracks perpendicular to the
direction of maximum principal stress induced by pure moment. Cracking outside the
pure bending zone starts similarly to flexural cracking but, as the load is increased,
other cracks are formed and as shear stresses become important, more inclined cracks
appear.
With the comparison of SCC and reference civil beams, it can be seen that the number
of cracks in the reference beam were higher than those in the SCC civil beam. The
cracks in fibre FSCC beam were smaller than those in SCC and reference civil beams.
This difference in crack-widths can be attributed to the use of the fibre in concrete
which affect the mechanism of development of cracks. The beam with SCC civil
developed similar load deflection response to that of the reference beam mix.
(Further conclusions will be added)
57
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 58 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
REFERENCES
[1]
Pera, J., Husson, S., and Guilhot, B. ‘Influence of finely ground limestone on
cement hydration’, Cement & Concrete Composites, 1999 (21), 99-105.
[2]
Hogan, F.J. and Meusel, J.W., 'Evaluation for durability and strength
development of a ground granulated blast furnace slag’, Cement, Concrete and
Aggregate, 3, No.1, (1981), pp.40-52.
[3]
Rilem recommendation,
reinforcement’, 1992.
[4]
Neville, A. M., ‘Properties of concrete’, Edition Longman, 1995.
[5]
R. Le ROY, J-M CUSSAC, and O. MARTIN: ‘Structures sensitive to creep:
from laboratory experimentation to structural design. The case of the Avignon
high-speed rail viaduct’, Special Issue of Revue Francaise du Génie Civil:
Creep and Shrinkage of Concrete, vol 3 n°3-4, ed. by Ulm, Prat, Calgaro and
Carol, pp 133-157, 1999.
[6]
Buenfeld, N.R., ‘Measuring and modelling transport phenomena in concrete
for life prediction of structures’, Chapter 5 of Prediction of Concrete
Durability, Edited by J. Glanville and A.M. Neville, E& FN Spon, London,
1997.
[7]
RILEM TC116-PCD: 'Permeability of concrete as a criterion of its durability,
C: determination of the capillary absorption of water of hardened concrete',
Materials and Structures. 32, April 1999, pp178-179.
[8]
McCarter, W.J. and Ezirim, H., 'Influence of constituents on the surface
absorption characteristics of concrete', Proceedings of 9th International
Congress on the Chemistry of Cement, pp39-45.
[9]
RILEM recommendations CPC-18, ‘Measurement of hardened concrete
carbonation depth’, Materials and Structures, 1988.
[10]
Bungey, J.H. and Millard, S.G., ‘Testing of concrete in structures’, 3rd Edition,
Blackie Academic & Professional, London, 1996.
[11]
Zhu, W and Bartos, PJM, 'Assessment of interfacial microstructure and bond
properties in aged GRC using a novel microindentation method', Cement and
Concrete Research, 27 (1997), pp.1701-1711.
TC51-ALC,
58
78-MCA,
‘Pull-out
test
for
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 59 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-04-26
[12]
Oliver, WC and Pharr, GM, 'An improved technique for determining hardness
and elastic modulus using load and displacement sensing indentation
experiments', Journal of Materials Research, 7 (1992), pp.1564-1579.
[13]
Doerner, MF and Nix, WD, 'A method for interpreting the data from depth
sensing indentation instruments', Journal of Materials Research, 1 (1986),
pp.601-609.
59
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 60 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-03-3
APPENDIX
A1 Preliminary investigation
A preliminary investigation was necessary in order to determine suitable materials
and mix designs for the full-scale work. These had to comply with the specified
requirements of the overall program.
A1.1 Materials
The characteristics of local aggregates are given in Table 21. Figures 51 and 52
present the grading of cement and additive powders and aggregates used in
preliminary mixes. Mix proportions are given in Table 22 and test method details of
fresh concrete in paragraph 5. Three types of filler, namely PFA (Scottish Power,
Longannet ), GGBS (Castle GX5), and LSP (limestone powder, Longcliffe), were
used. The PFA was classified as BS3892: Part 1, GGBS and LSP were finely ground.
The coarse aggregate (both sizes) was microgranite (Cloburn quarry), sand 1 a
quartzite building sand, and sand 2 a coarser quartzite concrete sand (both
Douglasmuir quarry). Two superplasticisers, Sikament 10 and Viscocrete 2 were
used.
Table 21 – Properties of coarse aggregate and sand
Properties
Relative density (Oven Dry), kg/m³
Relative density (SSD), kg/m³
Concrete shrinkage
Water absorption, %
Water absorption (from room dry), %
B.D. Uncompacted, kg/m³
B.D. Compacted, kg/m³
Loose Packing Density (SSD)
(without tamping)
Coarse
aggregate
2.60
2.62
0.029
0.80
0.20
1446
N/A
0.552
Sand 1
Sand 2
2.47
2.53
N/A
2.40
1.70
1550
N/A
0.614
2.51
2.56
N/A
1.00
0.10
1610
1730
0.629
Note: The loose packing density is the volume occupied/unit volume
60
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 61 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
100
Passing, %
80
60
PC
40
LSP
GGBS
20
PFA
0
1
10
100
Particle Size, µ m
Fig. 51 – Particle size distribution of cement, GGBS, PFA and LSP
Grading curve for aggregates
Grading curve for sands
100
80
10 mm
Passing, %
Passing, %
100
60
20 mm
40
20
0
Sand 1
80
60
Sand 2
40
20
0
1
10
100
10
100
1000
10000
Size, µ m
Size, mm
Fig. 52 – Grading of coarse aggregates and sands
A1.2
Effect of local materials and superplasticisers on concrete properties
A considerable amount of preliminary work was necessary to assess the materials’
performance in and suitability for fresh SCC. The main conclusions were:
61
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 62 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-03-3
•
Generally, at equal water content and slump flow value, PFA mixes were more
cohesive than GGBS mixes. In other words, GGBS mixes have lower viscosity
than corresponding PFA mixes; thus the GGBS mixes flow faster, but tend to
segregate at a lower slump flow value. Compared to GGBS and PFA mixes, LSP
mixes require lower water content and SP dosage for achieving the same slump
flow.
•
Content of fines in sand had a very significant effect on SCC mix proportions.
Fine sand requires more water and SP, but less filler than coarse sand. The SP
dosage, water content and cement/filler content could be adjusted by treating the
fines (<150 µm) in sand as part of the filler.
•
The mixes using Sikament 10 lost flowability faster than mixes using Viscocrete
2. Compared to Sikament 10, the Viscocrete 2 produced more cohesive mixes,
but at dosages higher than 1.2% (by weight of powder) large quantities of air
bubbles and low early strength were observed (see Table 18). This was
particularly true with the combination of PFA and Viscocrete 2.
•
Segregation tendency is significantly higher for mixes with large size (20mm)
aggregate than small aggregate, and the self-compacting properties of such mixes
tend to be more sensitive to small changes in mix proportions. It seems that
greater slump flow and cohesiveness are required for mixes using 20mm
aggregate.
•
Satisfactory compressive strength results could be obtained for both SCC housing
and civil engineering mixes, by adjusting the W/C ratio and the dosage of
superplasticizer. 28-day cube strengths were ~55 MPa and ~70 MPa respectively
for housing and civil engineering SCC mixes. At this preliminary stage, it was
found difficult to reduce the 28-day strength of the housing SCC mixes to the
target, c45Mpa. This is mainly due to the requirements of large amount of fillers
(450-550 kg/m³) and water content (<195 kg/m³) for achieving satisfactory
performance in fresh SCC mixes.
A.13
Effect of mixer
All mixes in the laboratory were prepared in 25 litre batches and mixed in an open pan
mixer. The mixing sequence consisted of mixing the sand and the coarse aggregate
with ¼ of water content, waiting 5 minutes for absorption, addition of cementitious
materials, addition of the rest of the water with most of the superplasticizer. The
concrete was mixed for five minutes then after measuring slump flow the dosage of
superplasticizer was adjusted to achieve slump flow greater that 600 mm. In the case
of fibre SCC, the fibre was added after all other materials
Three mixer types were used: a small free-fall drum mixer and two different sizes of
pan mixer.
The type of mixer had a strong effect on the water and SP requirement in SCC mixes.
The pan mixer, due to its strong shear action, appeared to produce SCC mixes of
62
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 63 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-03-3
equal slump flow value at water content 10 kg/m³ lower, or SP dosage a third less,
than the small lab drum mixer. The Viscocrete 2 seemed to be more sensitive to the
mixer types than Sikament 10. There were even noticeable, though minor differences
between the two pan mixers. The stronger action pan mixer seemed to promote
aggregate segregation and faster loss of workability with time.
A1.4
Test methods for fresh SCC
Test methods used on the fresh concrete were: slump flow, L-box, Orimet, Japanese
ring (JRing), and sieving test for assessing coarse aggregate segregation.
Modifications were made on L-box, Orimet, and JRing tests.
a) Slump flow – easy to carry out, can provide an indication of flowability;
requirement on flow and T50 values should be different for different maximum
sizes/shapes of aggregates, and admixtures; difficult to assess the
segregation/settlement tendency.
b) L-box – not easy to carry out; simulates better the real flow/blocking condition; if
bar spacing is selected appropriately it can provide information on self-levelling
and on likelihood of blocking – but the blocking ratio depends on slump flow and
surface condition/wall effect, as well as on the geometry and bar spacing of the Lbox; difficult to identify segregation/settlement.
c) Orimet – easy and quick to carry out, can provide an indication of flowability and
cohesiveness/viscosity; possible to assess segregation and blocking tendency by
appropriately selecting/adjusting the size of orifice and by adding steel bars to
limit the passage of materials.
d) JRing – easy to carry out, represents an improvement on slump flow, simulates
better the real flow condition; possible to assess blocking if the clear gap is
selected/adjusted properly.
e) Visual assessment of segregation – difficult to assess, often time consuming;
large dependence on sampling and efficiency of the mixer used.
f) Combination of Orimet and JRing – seems to be able to provide all the
information about flowability, cohesiveness/viscosity, likelihood of blocking; with
further modification on Orimet, may be able to assess segregation/settlement.
63
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 64 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-03-3
A1.5 - Table 21 - Mix proportions and fresh concrete properties in preliminary
investigation
Test Series
No.1
% of Filler in Powder
Type of Mixer Used
No.2
No.3
No.3x
No.4
No.5
No.6
30%
30%
30%
30%
30%
25%
25%
GGBS
GGBS
LSP
LSP
LSP
PFA
PFA
blue pan blue pan blue pan blue pan blue pan blue pan blue pan
Mix Proportions, kg/m³
Filler
GGBS
160
144
-
-
-
-
-
PFA
-
-
-
-
-
110
115
LSP
-
-
140
145
135
-
-
Portland Cement, 42.5
368
332
325
338
315
330
340
Coarse Aggregate, 6-20 mm
785
782
790
745
790
760
760
Sand1, 0-5 mm, (6.6%<150 um)
800
876
885
920
910
910
875
Sand2, 0-5 mm, (0.7%<150 um)
-
-
-
-
-
-
-
Effective Water
200
188
182
188
176
185
190
Total Powder
528
476
465
483
450
440
455
Total Powder +fines in sand
581
534
523
544
511
501
513
Sikament 10
8.04
7.14
7.00
7.88
7.20
8.80
7.96
Viscocrete 2
-
-
-
-
-
-
-
Orimet Test (80 mm orifice)
Average flow time,
~ 10 min.
-
-
-
-
-
-
-
seconds
~ 90 min.
-
-
-
-
-
-
-
Max and Min flow ~ 10 min.
time,
seconds
~ 90 min.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
~ 15 min.
595
572
713
722
613
545
688
~ 90 min.
-
-
-
-
-
-
-
~ 15 min.
-
-
-
-
9.2
16.96
7.73
~ 90 min.
-
-
-
-
-
-
-
~ 15 min.
-
47
30
30
45
45
35
~ 90 min.
-
-
-
-
-
-
-
~ 15 min.
-
12
18
12
13
10
15
~ 90 min.
-
-
-
-
-
-
-
JRing Test (~40 mm gaps)
slump flow, mm
T50, seconds
Hin, mm
H50, mm
64
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 65 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-03-3
Table 21 - Mix proportions etc in preliminary investigation (continued)
Test Series
No.7
No.8
No.9
No.10
No.11
No.12
25%
PFA
Yellow
pan
35%
GGBS
Yellow
pan
35%
LSP
yellow
pan
25%
PFA
yellow
pan
30%
PFA
yellow
pan
35%
GGBS
yellow
pan
-
174
-
-
-
185
PFA
115
-
-
115
140
-
LSP
-
-
182
-
-
-
Portland Cement, 42.5
335
330
318
335
320
330
Coarse Aggregate, 6-20 mm
750
750
757
750
750
750
Sand1, 0-5 mm, (6.6%<150 um)
900
895
919
-
-
-
Sand2, 0-5 mm, (0.7%<150 um)
-
-
-
900
900
900
Effective Water
185
185
168
185
180
180
Total Powder
450
504
500
450
460
515
Total Powder +fines in sand
509
563
561
456
466
521
Sikament 10
7.65
8.46
7.56
6.75
-
-
Viscocrete 2
-
-
-
-
5.75
5.15
% of Filler in Powder
Type of Mixer Used
Mix Proportions, kg/m³
Filler
GGBS
Orimet Test (80 mm orifice)
Without internal bars
With two 16 mm internal bars
Average flow time,
~ 10 min.
4.39
3.35
5.08
4.85
6.29
6.34
seconds
~ 90 min.
5.07
3.36
4.02
-
-
-
Max and Min flow ~ 10 min.
time,
seconds
~ 90 min.
4.88,4.0
6
5.16,4.9
8
JRing Test
slump flow, mm
T50, seconds
Hin, mm
H50, mm
3.63,3.21 5.70,4.70 5.74,4.17 6.85,5.82
3.41,3.30 4.03,4.01
~40 mm gaps
-
-
6.67,5.92
-
~50mm gaps, combined with
Orimet
738
673
575
~ 15 min.
660
580
535
~ 90 min.
-
570
635
-
-
-
~ 15 min.
9.74
7.48
8.91
-
-
-
~ 90 min.
-
11.45
10.71
-
-
-
~ 15 min.
40
52
55
30
-
47
~ 90 min.
-
-
45
-
-
-
~ 15 min.
16
11
10
14
-
15
~ 90 min.
-
-
15
-
-
-
65
SCC
Self Compacting Concrete
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 66 (73)
final report t4
Task 4 – Properties of Hardened Concrete
2000-03-3
Table 21 - Mix proportions etc preliminary investigation (continued)
Test Series
No.13
No.13x
No.14
No.14x
No.15
No.15x
% of Filler in Powder
30%
PFA
yellow
pan
30%
PFA
yellow
pan
30%
PFA
yellow
pan
30%
PFA
yellow
pan
40%
GGBS
yellow
pan
40%
GGBS
yellow
pan
-
-
-
-
220
219
PFA
139
139
150
150
-
-
LSP
-
-
-
-
-
-
Portland Cement, 42.5
318
318
350
350
330
328
Coarse Aggregate, 6-20 mm
746
746
750
750
750
746
Type of Mixer Used
Mix Proportions, kg/m³
Filler
GGBS
Sand1, 0-5 mm, (6.6%<150 um)
-
-
-
-
-
-
Sand2, 0-5 mm, (0.7%<150 um)
896
896
865
865
860
856
Effective Water
184
184
190
190
185
189
Total Powder
457
457
500
500
550
547
Total Powder +fines in sand
463
463
506
506
556
553
Sikament 10
-
-
-
-
-
-
Viscocrete 2
3.66
4.57
5.00
6.25
5.5
5.5
4.66
-
4.80
Orimet Test (80 mm orifice)
With two 16 mm internal bars
Average flow time, ~ 10 min.
-
seconds
3.26
-
~ 90 min.
-
-
-
-
-
-
Max and Min flow ~ 10 min.
time,
seconds
~ 90 min.
-
3.41,3.09
-
4.88,4.15
-
5.24,4.41
-
-
-
-
-
-
JRing Test
~50mm gaps, combined with Orimet
slump flow, mm
T50, seconds
Hin, mm
H50, mm
~ 15 min.
-
658
-
658
-
595
~ 90 min.
-
-
-
-
-
-
~ 15 min.
-
-
-
-
-
-
~ 90 min.
-
-
-
-
-
-
~ 15 min.
-
33
-
-
-
-
~ 90 min.
-
-
-
-
-
-
~ 15 min.
-
17
-
-
-
-
~ 90 min.
-
-
-
-
-
-
66
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 67 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
Table 21- Mix proportions etc preliminary investigation (continued)
Test Series
No.16
No.17
No.17x
No.18
No.19
No.20
% of Filler in Powder
25%
PFA
small
drum
30%
PFA
small
drum
30%
PFA
small
drum
33%
PFA
small
drum
38%
GGBS
small
drum
37%
GGBS
small
drum
-
-
-
-
199
190
PFA
114
140
140
159
-
-
LSP
-
-
-
-
-
-
Portland Cement, 42.5
333
320
320
318
318
325
Coarse Aggregate, 6-20 mm
746
750
750
746
746
745
Sand1, 0-5 mm, (6.6%<150 um)
896
900
900
871
886
885
Sand2, 0-5 mm, (0.7%<150 um)
-
-
-
-
-
-
Effective Water
189
180
180
185
185
185 (+5)
Total Powder
447
460
460
477
517
515
Total Powder +fines in sand
506
519
519
534
575
573
Sikament 10
7.65
-
-
-
-
-
Viscocrete 2
-
5.75
6.50
7.50
8.00
8 (+0.75)
Type of Mixer Used
Mix Proportions, kg/m³
Filler
GGBS
Orimet Test (80 mm orifice)
With two 16 mm internal bars
Average flow time,
~ 10 min.
-
-
-
seconds
~ 90 min.
-
-
-
Max and Min flow ~ 10 min.
time,
seconds
~ 90 min.
-
-
-
-
-
-
JRing Test
slump flow, mm
T50, seconds
Hin, mm
H50, mm
With two 16 mm internal bars
13.00
10.97
-
-
-
8.05
14.81,11.2
8
-
13.48,9.38
-
-
9.34,7.24
~50 mm gaps, not combining with Orimet
~50mm gaps with
orimet
685
-
~ 15 min.
670
-
610
525
~ 90 min.
-
-
-
-
-
700
~ 15 min.
8.79
-
8.56
11.71
-
-
~ 90 min.
-
-
-
-
-
-
~ 15 min.
35
-
45
60
40
-
~ 90 min.
-
-
-
-
-
35
~ 15 min.
22
-
12
12
18
-
~ 90 min.
-
-
-
-
-
12
67
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 68 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
Table 21- Mix proportions etc in preliminary investigation (continued)
Test Series
No.21
No.21x
No.22
No.22x
No.23
No.24
% of Filler in Powder
28%
PFA
small
drum
28%
PFA
small
drum
27%
PFA
small
drum
27%
PFA
small
drum
Ref. C30
Ref. C60
small
drum
small
drum
-
-
-
-
-
-
PFA
130
129
120
119
-
-
LSP
-
-
-
-
-
-
Portland Cement, 42.5
330
327
330
328
350
445
Coarse Aggregate, 6-20 mm
750
743
750
746
1225
1195
Sand1, 0-5 mm, (6.6%<150 um)
890
881
900
896
575
510
Sand2, 0-5 mm, (0.7%<150 um)
-
-
-
-
-
-
Effective Water
185
194
185
189
196
196
Total Powder
460
456
450
447
350
445
Total Powder +fines in sand
519
514
509
506
388
479
Sikament 10
-
-
-
-
-
-
Viscocrete 2
6.5
8 (+0.75)
8.16
8.16
-
-
Type of Mixer Used
Mix Proportions, kg/m³
Filler
GGBS
Orimet Test (80 mm orifice)
With two 16 mm internal bars
Average flow time,
~ 10 min.
-
-
-
6.84
seconds
~ 90 min.
-
7.53
-
-
Max and Min flow ~ 10 min.
time,
seconds
~ 90 min.
-
-
-
7.11,6.42
-
8.46,6.92
-
-
JRing Test
slump flow, mm
T50, seconds
Hin, mm
H50, mm
~50 mm gaps, combined with last orimet test
~ 15 min.
-
-
-
~ 90 min.
-
680
-
-
~ 15 min.
-
-
-
9.87
~ 90 min.
-
-
-
-
~ 15 min.
-
-
-
40 (45)
~ 90 min.
-
35
-
-
~ 15 min.
-
-
-
20 (14)
~ 90 min.
-
13
-
-
68
653 (565)
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 69 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
A2 - Table 23 - Statistical analysis of the pull out test results carried out by Lok test
(British standard 1881 part 207, 1992 and ISO/DIS 8046)
1. HOUSING REFERENCE MIX
A. 7 DAYS
TEST No.
COLUMN
(kN)
BEAM
(kN)
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
15
18
18
17
1.732
10.1
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
((kN)
C.O.V
(%)
18
18
23
24
22
20
20.83
2.56
24
18
12.2
16
15
14
17
18
18
16.33
1.63
18
14
9.9
B 28 DAYS
TEST No.
COLUMN
(kN)
BEAM
(kN)
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
22
22
23
22.33
0.577
2.5
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
30
30
29
--
36
28
30.6
3.13
36
28
10.2
24
26
20
24
27
30
25.166
3.37
30
20
13.3
2. HOUSING SCC MIX
A. 7 DAYS
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
20
21
22
21.00
1.00
4.8
69
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 70 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
TEST No.
COLUMN
(kN)
BEAM
(kN)
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
28
30
31
34
35
32
31.66
2.58
35
28
8.2
33
30
30
30
27
30
30.00
1.89
33
27
6.3
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
40
34
37
34
38
32
35.83
2.99
40
32
8.3
35
35
36
32
37
34
34.83
1.72
37
32
4.9
B 28 DAYS
TEST No.
COLUMN
(kN)
BEAM
(kN)
3. CIVIL REFERENCE MIX
A. 7 DAYS
TEST No.
COLUMN
(kN)
BEAM
(kN)
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
23
27
25
25.00
2.00
8
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
40
38
35
42
38
40
38.83
2.40
42
35
6.2
25
26
30
26
28
25
26.26
1.96
30
25
7.5
B 28 DAYS
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
30
28
30
29.33
1.15
3.9
TEST No.
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
COLUMN
(kN)
40
38
35
35
49
44
40.16
5.49
49
35
13.6
70
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 71 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
4. CIVIL SCC MIX
A. 7 DAYS
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
34
34
34
34.00
0.00
0
TEST No.
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
COLUMN
(kN)
36
40
40
48
48
44
42.66
4.84
48
36
11.3
B 28 DAYS
TEST No.
COLUMN
(kN)
BEAM
(kN)
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
44
40
--
--
--
--
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
56
48
58
54
52
48
52.66
4.13
58
48
7.8
38
42
40
42
38
36
39.33
2.42
42
36
6.2
5. SFRSCC MIX
A. 7 DAYS
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
30
32
30
30.66
1.15
3.8
TEST No.
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
BEAM
(kN)
30
38
38
36
33
36
35.16
3.12
38
30
8.8
71
SCC
Brite EuRam Proposal No. BE96-3801
Brite EuRam Contract No. BRPR-CT96-0366
Page 72 (73)
final report t4
Self Compacting Concrete
Task 4 – Properties of Hardened Concrete
2000-03-3
B 28 DAYS
CUBES
1
2
3
MEAN
(kN)
S.D.
(kN)
C.O.V
(%)
(kN)
35
34
32
33.67
1.52
4.5
TEST No.
1
2
3
4
5
6
MEAN
(kN)
S.D.
(kN)
HIGH
(kN)
LOW
(kN)
C.O.V
(%)
BEAM
(kN)
40
38
38
38
40
37
38.50
1.22
40
37
3.2
72