CHAPTER 2

LITERATURE REVIEW
2.1 Concrete

Concrete is a blend of cement, water, and aggregates, with or without admixtures. The
cement and water will structure a paste that hardens as a effect of a chemical reaction
among the cement and water. The paste acts as glue, binding the aggregates (sand and
gravel or crushed stone) into a solid rock-like mass. The value of the paste and the
aggregates state the engineering properties of this building material. Throughout
hydration and hardening, concrete will develop certain physical and chemical
properties, among others, mechanical strength, low permeability and chemical and
volume stability. Concrete has relatively high compressive strength, but significantly
lower tensile strength, about 10% of the compressive strength.

Table 2.1 Typical Properties of Normal-Strength Portland cement Concrete.

Compressive strength 20-40 MPa

Flexural strength 3-5 MPa

Tensile strength 2-5 MPa

Modulus of elasticity 14000-41000 MPa

Permeability 1x10-10 cm/sec

Coefficient of thermal expansion 10-5/℃

Drying shrinkage 4x10-4 - 8x10-4

Drying shrinkage of Reinforced

2x10-4-3x10-4
Concrete

14
2.2 Cement
Cement may be defined as an adhesive substance capable of uniting fragments or
masses of solid mater to a compact whole. Portland cement was invented in 1824 by
an English mason, Joseph Aspin, who named his product Portland cement because it
produced a concrete that was of the same colour as natural stone on the Isle of
Portland in the English Channel.

Raw materials for manufacturing cement consist of basically calcareous and siliceous
(generally argillaceous) material. The mixture is heated to a high temperature within a
rotating kiln to produce a complex group of chemicals, collectively called cement
clinker (Neville, 1987). Cement is distinct from the ancient cement. It is termed
hydraulic cement for its ability to set and harden under water. Briefly, the chemicals
present in clinker are nominally the four major potential compounds and several minor
compounds. The four major potential compounds are normally termed as Tricalcium

silicate (3CaO.SiO2), Dicalcium silicate (2CaO.SiO2), Tricalcium aluminate

(3CaO.Al2O3) and Tetracalcium aluminoferrite (4CaO. Al2O3.Fe2O3). The American

Society for Testing and Materials (ASTM) Standard C 150, Specification for Portland
cement, provides for the following types of Portland cement:

(a) Type I General Portland cement

(b) Type II Moderate-sulphate-resistant cement
(c) Type III High-early-strength cement
(d) Type IV Low–heat-of-hydration cement
(e) Type V High-sulphate-resistant cement

2.3 Cement Replacement Material

2.3.1 Introduction

With the extensively use of cement in concrete, there has been some environmental

concerns in terms of damage caused by the extraction of raw material and CO 2

emission during cement manufacture. This has brought pressures to reduce the cement
consumption in the industry. At the same time, there are getting more requirements for
enhancement in concrete durability to sustain the changing environment which is
apparently different from the old days.

15
With the development in concrete technology, cement replacement materials (CRM)
have been introduced as substitutes for cement in concrete. Several types of materials
are in common use, some of which are by-products from other industrial processes,
and hence their use may have economic advantages. However, the main reason for
their use is that they can give a variety of useful enhancements or modifications to the
concrete properties. All the materials have two common features:

(i) Their particle size range is similar to or smaller than that of Portland cement.

(ii) They are pozzolan material.

(iii) Pozzolanic Behaviour

A common feature of nearly all CRM is that they exhibit pozzolanic behaviour.

Pozzolanic material is the material which contains active silica (SiO2) and is not
cementitious in itself but will, in a finely divided form and in the presence of moisture,
chemically reacts with calcium hydroxide at ordinary temperatures to form
cementitious compounds (Malhotra et at., 1983).

2.3.2 Types of Cement Replacement Materials

The cement replacement materials that are used in this study are:

2.3.2.1 Silica Fume

2.3.2.1.1 Effect of Silica fume on fresh properties of concrete

Fresh concrete containing silica fume is more cohesive and less prone to segregation
than concrete without silica fume. Concrete containing silica fume shows substantial
reduced bleeding. Additionally silica fume reduces bleeding by physically blocking
the pores in the fresh concrete. Use of silica fume does not significantly change the
unit weight of concrete.

(a) Consistency
Rao (2003) determined the influence of silica fume on the consistency of cement
pastes and mortars. Specific gravity and specific surface of the silica fume were 2.05
and 16,000 m²/kg, respectively. Silica fume was varied from 0 to 30% at a constant
increment of 2.5/5% by weight of cement. Since the SF is finer than the cement, the
specific surface increased with increase in SF content. The standard consistency of

16
pure cement paste was found out to be 31.50%; while at 30% SF, it was 44.25%. It
was observed that the consistency of cement increased with the increase in SF content.
As much as 40% of additional water requirement was observed for cement pastes
containing 20–30% SF.

Qing et al (2007) examined the influence of nano-Si (NS) addition on consistency of

cement paste incorporating NS or silica fume. The influence of NS or silica fume
addition on consistency and setting time of fresh pastes is given in Table 2.2. It was
found that with a increase in the NS content, fresh pastes for sample A-series grew
thicker gradually and their penetration depths (consistency value) decreased gently as
compared with that of control sample (CO). While with increasing the silica fume
content, the pastes for sample B-series grew thinner and their depths increased. They
concluded that silica fume makes cement paste thinner as compared with NS.

Table 2.2 Mix proportions, consistency of pastes made of cement and NS and
silica fume (Quing et al, 2007)

Mix proportion in mass Consistency

Sample

Cement Nano Silica Water SM (mm)

silica fume

CO 100 0 0 22 2.5 34

A1 99 1 0 22 2.5 34

A2 98 2 0 22 2.5 33

A3 97 3 0 22 2.5 33

A5 95 5 0 22 2.5 32

B2 98 0 2 22 2.5 35

B3 97 0 3 22 2.5 35

B5 95 0 5 22 2.5 36

17
(b) Setting Times
Alshamsi et al (1993) reported that addition of micro-silica lengthened the setting
time of pastes. This was expected since micro-silica replaces part of the OPC, reducing
the early stiffening potential. While the addition of micro-silica (10%) had little effect
on setting times, higher percentages produced significant influences. There was 6–
20% increase in setting times when OPC was replaced with 20% micro-silica.

Lohtia and Joshi (1996) concluded that the addition of silica fume to concrete in the
absence of water-reducer or super plasticizer causes delay in setting time, compared to
non-silica fume concrete of equal strength, especially when the silica fume content
was high. The additions of 5 - 10% silica fume to either super plasticized or non-super
plasticized concrete with W/C (SF) ratio of 0.40 did not exhibit any significant
increase in setting time. However, when 15% silica fume was added with super
plasticizer, both the initial and final setting times were delayed by approximately 1 and
2 h, respectively. The observed delay was attributed to the relatiely high dose of super
plasticizers needed for the high amount of silica fume added to concrete.

(c) Workability
Sellevold et al. (1983) reported that there is net decrease in water requirements in
concretes containing high concentration of silica fume and water-reducer or super
plasticizers. The addition of water-reducer or a super plasticizer causes the dispersion
of cement and silica fume particles and reduces the concentration of contact points
between the different grains; resulting in less water requirement to achieve a given
consistency.

Khayat and Aitcin (1993) reported that addition of 10% silica fume in a lean concrete
(100 kg/• ) of cement reduced the water demand. However, it exhibited poor durability
against freeze–thaw attack. In normal structure concrete, even with 5% silica fume
addition, the water demand is increased to maintain constant slump. For producing
very high strength and durable concrete, silica fume up to 10% is added as an
admixture and use of super-plasticizer to maintain specified slump is found necessary.
When no plasticizers are used an additional 1 l/• of water should be used for every 1
kg/• of silica fume addition to maintain constant level of fluidity.

Wong and Razak (2005) studied the cementing efficiency factor (k) of silica fume.
Specific gravity of silica fume was 2.22. Three water-to-cementitious material ratios

18
(w/cm) of 0.27, 0.30 and 0.33 were used in concrete mixtures. At each w/c ratio,
cement was replaced with 0, 5, 10, and 15% silica fume. Slump and Vebe time results
are shown in Table 2.3. It could be seen from this table that mixtures achieved slump
values ranging from 30 to 260 mm, while Vebe time was in the range of 1–15 s. The
large variation of workability across mixtures was due to the constant super plasticizer
dosage used for mixtures with the same w/c ratio.

Table 2.3 Workability characteristics (Wong et al., 2005)

Slump Vebe Slump Vebe Slump Vebe

Mixtu w/c w/c w/c
(mm) (s) (mm) (s) (mm) (s)
re

C 0.27 165 8 0.30 225 3 0.33 240 1

SF 5 100 8 215 3 180 3

SF10 50 12 117 5 100 6

SF15 35 15 30 16 35 16

2.3.2.1.2 Effect of Silica Fume on the Hardened Properties of Concrete

(a) Compressive strength

Bentur et al. (1989) reported that the strength of silica fume concrete is greater than
that of silica fume paste which they attributed to the change in the role of the
aggregate in concrete. In cement concrete, the aggregate functions as inert filler but
due to the presence of weak interfacial zone, composite concrete is weaker then

19
cement paste. But, in silica fume concrete, the presence of silica fume eliminates this
weak link by strengthening the cement paste aggregate bond and forming a less porous
and more homogenous microstructure in the interfacial region. Thus, silica fume
concrete is stronger than silica fume cement paste, taking into account that the strength
of aggregate exceeds the strength of cement paste.

Sobolev (2004) studied the compressive strength of high performance concretes. It

was observed that:

(i) Increase in super plasticizer dosage from 8 to 18% led to a reduction of w/c from
0.31 to 0.26 and improved the concrete compressive strength from 86 to 97 MPa.

(ii) Maximum compressive strength of 91 MPa was obtained at 15% silica fume.

(iii) Lower strength value of 90 MPa occurred at 10 and 20% silica fume.

(iii) Reduction of w/c 0.32–0.19 increased the compressive strength of cement

concrete and resulted in super high strength concrete having strength up to 135 MPa.

Table 2.4 Details of HPC mixtures (Sobolev, 2004)

Proportions
SF (5%) SF (10%) SF (15%) SF (20%)
(kg/m ) 3

Cement 426 449 449 478

Silica fume 22 50 83 120

Age Compressive strength (MPa)

1 day 16.8 24.1 34.4 45.1

3 days 28.6 42.2 63.0 84.9

7 days 50.1 67.2 84.8 102.5

28 days 60.0 80.0 100.0 120.0

20
Koksal et al. (2008) studied the compressive strength of steel fibre reinforced
concrete with silica fume. Cold drawn steel fibres with hooked ends were used. Aspect
ratios (l/d) of fibres were 65 and 80 and volume fractions (Vf) of steel fibres were 0.5
and 1%. Silica fume content was 0, 5, 10, and 15% weight of cement. They observed
that (i) a considerable increase in the compressive strength of the concretes without
steel fibres by increasing the silica fume content. The increases were 12, 73.4 and
85.5% for 5, 10 and 15% silica fume, respectively. These result were clearly
dependant on increasing bond strength of cement paste–aggregate interface by means
of filling effect of silica fume; and (ii) compressive strengths of concretes produced by
additions of both steel fibre and silica fume had higher than the ones containing silica
fume only.

Salim et al (2016) studied the hardened properties of concrete using Silica fume as a
replacement material at 7, 14 and 28 days. They replaced 5, 10, 15, 20, 30, 40 and
50% of cement by Silica fume and they found that at 10% replacement level, the
compressive strength was 37.74 MP a after 28 days.

(b) Tensile Strength

Hooton (1993) reported the splitting tensile strength of silica fume concretes up to the
age of 182 days (Table 2.5). It can be seen that except at 28 days, the splitting tensile
strength was not improved for silica fume concrete mixes. Also it was observed that
with increasing replacement of silica fume split tensile strength decreased.

Table 2.5 Splitting tensile strength of concrete, MPa (Hooton, 1993)

Test age
Concrete mixes
(days)

Control 10% SF 15% SF 20% SF

28 5.2 6.3 6.2 4.6

91 6.8 6.7 6.2 5.6

182 7.1 6.2 6.5 5.6

21
Tanyildizi and Coskun (2008) studied the effect of silica fume on tensile strength of
lightweight concrete exposed to high temperature. In mixtures containing silica fume,
0, 10, 20 and 30% of Portland cement by weight was replaced with silica fume. They
reported that the tensile strength started to drop with temperature starting from 2008C.
The reduction in splitting tensile strength of lightweight concrete containing 10%
silica fume was 3.11, 11.46 and 80.15% at the 200, 400 and 800 respectively. The
reduction in splitting tensile strength of lightweight concrete containing 20% silica
fume was 4.69, 12.91 and 78.87% at the 200, 400 and 800 respectively. And the
reduction in splitting tensile strength of lightweight concrete containing 30% silica
fume was 5.8, 40.62 and 75.08% at the 200, 400 and 800 , respectively. They
concluded that addition of admixture silica fume prevented the decrease in the tensile
strength of concrete.

Gonzalez and Martinez (2008) studied the properties of concrete using recycled
aggregates from Spanish demolition debris (RC mixes) and the impact of the addition
of silica fume on the properties of recycled concrete (RCS mixes). A comparison was
made between both these materials and standard conventional concrete (CC mixes),
which was also modified by adding silica fume (CCS mixes). For the test, they made
10 CC, 7CCS, 10RC and 6RCS mixes. The results of splitting tensile strength are
shown in Table 2.6. They stated that neither the addition of recycled aggregates nor
the addition of silica fume had any particular impact on the tensile strength of the
concrete types. The reason behind this was that silica fume mainly affected mechanical
properties like compressive strength of high strength concretes.

Table 2.6 Average tensile splitting tensile strength (MPa) for the various test ages
(Gonzalez et al., 2008)

Average splitting tensile strength (MPa)

Cube
7 days 28 days 115 days

CC 3.12 3.15 3.32

CCS 3.04 3.15 3.31

RC 3.17 3.00 3.37

RCS 3.24 3.36 3.35

22
Salim et al (2016) studied the hardened properties of concrete using Silica fume as a
replacement material at 7, 14 and 28 days. They replaced 5, 10, 15, 20, 30, 40 and
50% of cement by Silica fume and they found that, at 10% replacement level, the
tensile strength had a maximum value of 9.12 MPa after 28 days which was very high
as compared to that of control mix (6.89 MPa).

c) Sulphate Resistance

According to ACI Committee 234, the effect of silica fume on sulphate resistance is
due more to the reduction in permeability than to dilution of the C3A content because
of the relatively low doses of silica fume used in practice.

Sellevold and Nilsen (1987) reported field studies of concretes with and with out 15%
silica fume. After 20 years’ exposure to ground water containing 4 g/L sulphate and
2.5–7.0 pH, the performance of the silica fume concrete was found equal to that of the
concretes made with sulphate-resisting Portland cement, even though the
water/cementitious materials ratio was higher for silica fume concrete (0.62) than for
control (0.50).

Cohen and Bentur (1988) studied the effect of 15% silica fume replacement of Types
I and V Portland cement on the resistance to sulphate attack in magnesium and sodium
sulphate solutions. The water–cementitious materials ratio was 0.3. In the sodium
sulphate solutions, the silica fume concrete specimens were resistant to sulphate
attack. In the magnesium sulphate solutions, all the specimens expanded, with the
Type I cement specimens (with or without silica fume) expanding more than Type V
cement specimens (with or without silica fume). Since specimens were thin (6 mm),
the authors attributed the effect of silica fume on sulphate resistance more to chemical
effects than to reduced permeability.

Irassar et al. (1996) investigated the sulphate resistance of concrete made with silica
fume. Concrete specimens were half-buried in sulphate soil for 5 years. Mineral
admixtures were used as a partial replacement for ordinary Portland cement (C3A =
8.5%), and the progress of sulphate attack was evaluated by several methods (visual
rating, loss in mass, dynamic modulus, strength, X-ray analysis). Results showed that
silica fume improved the sulphate resistance when the concrete was buried in the soil.

23
However, concretes with high content of silica fume exhibited a greater surface scaling
over soil level due to the sulphate salt crystallization.

Shannag et al. (2003) prepared high-performance concrete mixes containing various

proportions of natural pozzolan and silica fume (up to 15% by weight of cement).
They were stored in sodium and magnesium sulphate solutions, in Dead Sea and Red
Sea waters. After 1 year immersion in sulphate solution and sea water, the concrete
mix containing a combination of 15% silica fume, and 15% natural pozzolan (by
weight of cement) showed a maximum protection against sulphate attack compared to
those investigated in the study. This mix retained more than 65% of its strength after 1
year of storage in sulphates solutions and sea waters. The superior resistance of that
mix against sulphate attack was attributed to the pore refinement process and further
densification of the transition zone occurring due to the conversion of lime forming
from the hydration of cement into additional binding material through lime-pozzolan
reaction. The results also showed that magnesium sulphates had a more damaging
effect than sodium sulphates; which was consistent with the data available in the
literature.

Lee et al. (2005) studied the effectiveness of silica fume in controlling the damage
arising from sulphate attack. The water/cementitious materials ratios (w/cm) of the
mortar mixtures were 0.35, 0.45 and 0.55. Under this sulphate environment, the
incorporation of 10% silica fume in OPC matrix showed no evidence of spalling and
cracking up to about 1 year of exposure, and strength loss increased as the w/cm ratio
increased; and the total strength loss as well as that between different w/cm ratio levels
was greater in mortar specimens without silica fume compared to those with silica
fume.

2.3.2.2 Metakaolin

2.3.2.2.1 Effect of Metakaolin on Fresh Properties of Concrete

Wild et al. (1996) studied the workability of metakaolin concretes. Replacement

levels of ordinary Portland cement (OPC) by metakaolin (MK) were 0, 5, 10, 15, 20,
25, and 30% and water-binder ratio (w/b) was 0.45. OPC concrete mixture proportion
was 1:2.3:3.4.Workability results (slump, compacting factor and vebe time) are given
in Table 2.7.

24
 Table 2.7 Workability of metakaolin concretes (Wild et al., 1996)

Metakaolin (%) Super plasticizer Slump (mm) Compacting Vebe time

(%) factor (sec)

0 0 5 0.81 26

5 0.6 10 0.84 15

10 1.2 15 0.88 10

15 1.8 25 0.89 9

20 2.4 75 0.89 7

25 3.0 75 0.89 4

30 3.6 90 0.90 5

Asbridge et al. (2001) showed that MK reduced the rate of chloride diffusion through
the hydrated cement matrix and tended to enhance the resistance to chloride transport
of the ITZ material within the mortars.

Brooks and Johari (2001) reported the slump and setting times of concretes
containing 0, 5, 10, and 15% metakaolin (MK). Control concrete mixture proportion
was 1:1.5:2.5 with water-binder ratio of 0.28. Slump and setting times results are
given in Table 2.8. It can be seen that slump decreased and setting times increased
with the increase in MK content.

25
 Table 2.8 Workability, setting times of MK concretes (Brooks and Johari, 2001)

Initial setting time Final setting

Concrete mixes Slump
(hours) time (hours)

OPC 100 5 7.7

MK5 30 6.42 8.82

MK10 20 6.98 9.42

MK15 5 6.45 9.31

Li and Ding (2003) investigated the fluidity of Portland cement containing metakaolin
(MK) and ultra-fine slag (S). Four series of pastes were made. First series contained
100% cement, 0% MK, and 0% slag; second had 90% cement and 10% MK, 0% slag;
third had 70% cement, 10% MK, and 20% slag, and the fourth series had 60% cement,
10% MK, and 30% slag. Description of Physical properties of cement paste is given in
Table 2.9.

(i) After incorporation of MK into PC, the fluidity of MK blended cement became
poorer than that of PC at the same dosage of super plasticizer and the same
water/binder ratio.

(ii) Incorporation of 10% MK and 20% or 30% ultra-fine slag jointly into PC,
improved the fluidity of blended cements, also the 28-day compressive strength.

(iii) Both MK and slag can react with CH released by cement clinker hydration to
produce secondary C-S-H gel inside the cement paste. The secondary formed C-S-H
gel improved the microstructure of cement paste matrix.

26
 Table 2.9 Physical properties of cement paste (Li and Ding, 2003)

Water required
Mixture Setting times (hours:
Composition for standard
type mins)
consistency (%)

Cement MK Slag Initial Final

PO 100 0 0 26.3 1:36 3:27

MI 90 10 0 29.1 1:11 2:12

S@ 70 10 20 28.4 1:55 4:34

S# 60 40 30 28.0 2:19 6:15

Bai et al. (2003) used neural networks to predict workability of concrete incorporating
metakaolin and fly ash. The predictions reflected the effect of graduated variations in
pozzolanic replacement in Portland cement up to 15% MK and 40% FA. The results
showed that the models are reliable and accurate and illustrate how neural networks
27
can be used to beneficially predict the workability parameters of slump, compacting
factor and Vebe time across a wide range of PC-FA-MK compositions. On the basis of
the models developed, effects of MK and FA on workability were analyzed and iso -
slump maps were plotted for w/b 0.25, 0.4 and 0.5. Through this, it is possible for the
designer to produce mixtures with various blend compositions for a given range of
slumps.

Li and Ding (2003) investigated the consistency and setting times of Portland cement
containing metakaolin (MK) and ultra-fine slag (S). Four series of pastes were made.
First series contained 100% cement, 0% MK, ad 0% slag; second had 90% cement and
10% MK, 0% slag; third had 70% cement, 10% MK, and 20% slag, and the fourth
series had 60% cement, 10% MK, and 30% slag. Test results of water consistency and
setting times are presented in Table 2.12. They concluded that:

(a) Standard consistency of cement paste was in the range of 26.3–29.1%. Mixture PO
had the lowest water requirement, but MK blended cement M1 had the highest Water
requirement. After the slag was mixed with MK blended cement, the water
requirement of cement mixture S2 and S3 was reduced, but it was still higher than that
of PC.

(b) Initial and final setting times of the cement incorporating 10% MK were shorter
than those of PC, while the initial and final times of S2 and S3 cements were longer
than those of PC.

Badogiannis et al. (2005) reported the results of water demand and setting times of
cements containing five metakaolins. The metakaolinite contents in metakaolins MK1,
MK2, MK3, and MK4 (derived from poor Greek kaolins) were 36, 37, 71, and 49%,
respectively, where as it was 95% in a commercial metakaolin (MKC) of high purity.
Cement was replaced with 0, 10, and 20% Metakaolin. They concluded that:

(i) Blended cements demanded significantly more water than the relatively pure
cement. With a metakaolin content of 10%, the water demand varied from 29.0% to
32.5%, while the PC had a water demand of 27.5%. With a metakaolin content of
20%, the water demand varied from 31.5% to 41.5%. Metakaolins MK1 and MK2
showed the best behaviour and increase in water demand was less. The increase in

28
water demand could be attributed to the high fineness of metakaolin as well as to their
narrow particle size distribution.
(ii) The initial and final setting time of metakaolin cements was affected by the
metakaolin content. Cements with 10% metakaolin, generally, exhibited similar setting
times to that of PC, while for 20% metakaolin content there was a delay in the setting.
MK4 showed the greatest effect on the setting delay of the cements.
In general, metakaolin offers a set of benefits similar to those imparted by silica fume,
including comparable strengths, permeability, chemical resistance, and drying
shrinkage resistance (Justice et al. 2005).

The experimental results obtained in (Vejmelkova et al. 2010) show that the
replacement of PC by 10% of MK leads in most cases either to improvements or at
least does not significantly impair substantial properties of the analyzed high
performance concrete.

Fadhil et al. (2014) investigated the microstructure of concrete using Metakaolin and
polyvinyl alcohol (PVA) fibres as mineral admixture. They found that the
microstructure of concrete with MK has been refined and packing of material is much
better with MK which is also reported by Poon et al. There is none to negligible voids
in concrete with 10% MK.

2.3.2.2.2 Effect of Metakaolin on the Properties of Hardened Concrete

(a) Compressive Strength

Wild et al. (1996) investigated the effect of varying percentages of metakaolin on the
compressive strength of concrete mixtures. Replacement levels of ordinary Portland
cement (OPC) by metakaolin (MK) were 0, 5, 10, 15, 20, 25, and 30% and water-
binder ratio (w/b) was 0.45. OPC concrete mixture proportion was 1:2.3:3.4.
Compressive strength tests were conducted up to the age of 90 days, and results are
presented in Table 2.10. They concluded that inclusion of MK as partial replacement
of cement enhanced the compressive strength of concrete at all ages, but the optimum
replacement level of OPC by MK to give maximum long term strength enhancement
was about 20%.

29
Table 2.10 Compressive strengths and densities of metakaolin concretes (Wild et
al., 1996)

Compressive strength (MPa)

Density

MK (%)
(kg/m3)
1 day 7 days 14 days 28 days 90 days

0 2490 19.07 50.23 57.10 62.60 72.43

5 2440 21.50 53.80 58.97 63.50 71.63

10 2460 22.43 62.30 69.23 71.00 80.07

15 2470 20.23 64.80 74.67 76.00 83.70

20 2480 19.33 66.47 75.73 82.47 85.13

25 2470 15.73 62.50 69.77 73.93 82.23

30 2480 14.53 60.53 72.33 76.73 81.80

Curcio et al. (1998) examined the role of metakaolin as pozzolanic micro filler for
high-performance mortars. They characterized super-plasticized mortars containing
metakaolin (MK) as 15% replacement of cement and with a water/binder ratio of 0.33.
Four commercially available MK samples and one silica fume sample were studied for
comparison. They concluded that:

30
(a) Specimens containing three of the four metakaolin samples had a higher rate of
compressive strength development as compared to that of the control at ages below 28
days, a consequence of the higher hydration rate. Silica fume and the fourth and
coarser metakaolin had a less pronounced effect.
(b) At 90 and 180 days metakaolin and silica fume specimens gave similar strengths.
The difference in the compressive strength between the specimens with micro fillers
and the control decreased after 28 days, because of a smaller slowdown of the
hydration rate in the control.
DTA calcium hydroxide peak heights on mortars and the pozzolanic activity test on
mixtures of 80% cement and 20% micro filler showed that three of the four metakaolin
had a remarkable pozzolanic activity comparable to that of silica fume, even if slightly
lower at longer ages. The fourth and coarser metakaolin sample was less reactive. The
amount of water that was lost after 24 h at 90◦C was different for all types of
specimens and was dependent on the differences of the hydration products formed and
on the permeability of the materials. It can be related to the fineness of the micro filler
in the specimens with metakaolin.

Poon et al. (2001) showed that the cement pastes containing 5% – 20% MK had
higher compressive strengths than the control at all ages from 3 to 90 days, with the
paste containing 10% MK performing the best. The cement pastes containing SF or
FA had lower compressive strength than the control at early ages. The SF
replacements resulted in higher compressive strengths than the control at 28 and 90
days, while the FA replacement resulted in a higher compressive strength than the
control only at 90 days. The above results indicated that at early ages, the MK
contributed better to the compressive strength development of high-performance
cement pastes than the SF.

Brooks and Johari (2001) determined the compressive strength of concretes

containing 0, 5, 10, and 15% metakaolin (MK). Mixture proportion of OPC concrete
was 1:1.5:2.5 with a water-to-cement ratio of 0.28. 28-day Compressive strength
results of MK concretes are given in Table 2.11. It is evident from the table that
compressive strength increased with the increase in the metakaolin content.

Table 2.11 Compressive strengths of MK concrete mixes (Brooks and Johari,

2001)

31
 Compressive strength
Concrete mixes
(MPa)

OPC 87.0

MK5 91.5

MK10 104.0

MK15 103.5

Bai et al. (2002) reported the influence of the composition of Portland cement–
pulverized fuel ash–metakaolin binders on strength development of PC–PFA–MK
concrete cured both in air and water. Concrete mixtures were made with four different
total cement replacement levels (10, 20, 30 and 40%) for PC–PFA–MK concrete with
various MK/PFA proportions and cured in water and air up to 18 months.

In water-cured concrete made with PC–PFA–MK binder, the MK enhanced early (28
days) strength, and PFA retarded early strength. Air-cured concrete showed a loss in
strength relative to equivalent concrete that was water cured and the strength
difference increased with curing period. The difference was enhanced in concrete
made with PC–PFA binder at high replacement levels, which showed a much reduced
strength gain with time when air cured, whereas for PC–PFA–MK concrete, this
difference was reduced as the MK content increases.

Jin and Li (2003) investigated the effect of mineral admixtures such as silica fume,
slag, fly ash, and metakaolin on the mechanical behaviour of young concrete under
either uniaxial compression or tension. Tests were conducted at ages of 1/2, 1, 2, 3, 7,
and 28 days. They reported that:

32
(i) Stress-strain (deformation) curves for young concretes were different from mature
concretes in their capability in resisting the deformation. More ductile behaviour was
observed for a young concrete. However, the complete stress-deformation curves for
young concrete under uniaxial tension showed less ductility as compared to those
under uniaxial compression.
(ii) Incorporation of metakaolin, silica fume, and slag into concrete can prompt both
the strength and modulus of elasticity development of a young concrete.
(iii) Metakaolin showed the best enhancement on the mechanical properties of young
concrete. The reactivity of slag modified concrete was speeded up after one day.

Cachim et al. (2010) observed that the addition of MK has a beneficial effect on the
strength at 28 days and beyond, for all types of curing conditions. Cement pastes
containing 5% to 20% MK had higher compressive strengths than the control
specimens at all ages from 3 to 90 days, with the paste containing 10% MK
performing the best (Poon et al. 2001).

Fadhil et al (2014) investigated the compressive strength of concrete using

Metakaolin and polyvinyl alcohol (PVA) fibres as mineral admixture. They found that
the compressive strength of concrete with 5% and 10% MK has been increased about
5% and 16.5% at age of 28 days as compared to control mix, respectively.

(b) Tensile strength

Qian and Li (2001) investigated the tensile strength of concrete incorporating 0, 5, 10,
and 15% metakaolin as partial replacement of cement. Metakaolin had specific surface
area 12000 /kg, and its average particle diameter was 2.23 µm. 300×100×20 mm
samples were tested under direct tension. Tests were conducted at the age of 28 days.
Tensile strength test results are presented in Table 2.12.

The results showed that tensile strength of concrete increased systematically with
increasing metakaolin replacement level. The average tensile strength increases were
7% (5% metakaolin), 16% (10% metakaolin), and 28% (15% metakaolin), and the
average ultimate strain increases were 3% (5%metakaolin), 19% (10% metakaolin),
and 27% (15% metakaolin).

33
The descending area of over-peak stress was less steep when metakaolin replacement
was 5% and 10% whereas with 15% metakaolin it was similar to that for concrete
without metakaolin. The tensile elasticity modulus for these specimens is in the range
from 26 to 27 GPa.

Table 2.12 Tensile strength of concrete with different metakaolin replacements

(Qian and Li, 2001)

Tensile strength (MPa)

Age (days)
MK (0%) MK (5%) MK (10%) MK (15%)

28 3.35 3.58 3.88 4.29

The splitting tensile strength results also generally showed increases with use of MK
compared to ordinary concrete controls (Justice et al. 2005).

Brooks and Johari (2001) found that metakaolin reduces the drying shrinkage as well
as creep. This effect can be related to reduction of C-S-H content due to formation of
hydrogarnet phases (Nita et al. 1998).

Fadhil et al. (2014) studied the tensile strength of concrete using Metakaolin and
polyvinyl alcohol (PVA) fibres as mineral admixture. They found that inclusion of 5
and 10% MK caused 16.5% and 24% increase in splitting tensile strength at age of 28
days as compared to control mix.

(c) Sulphate Resistance

Sulphate attack is one of the most aggressive environmental deteriorations that affect
the long-term durability of concrete structures. The sulphate attack on concrete leads
to expansion, cracking, and deterioration of many civil engineering structures exposed
to sulphate environment such as piers, bridges, foundations, concrete pipes, etc. The
sulphate ions in solution, which come from the soil, ground water, and seawater, are
found in combination with other ions such as sodium, potassium, magnesium and

34
calcium ions. The sulphate attack is generally attributed to the reaction of sulphate
ions with calcium hydroxide and calcium aluminate hydrate to form gypsum and
ettringite. The gypsum and ettringite formed as a result of sulphate attack is
significantly more voluminous (1.2–2.2 times) than the initial reactants. The formation
of gypsum and ettringite leads to expansion, cracking, deterioration, and disruption of
concrete structures. In addition to the formation of ettringite and gypsum and its
subsequent expansion, the deterioration due to sulphate attack is partially caused by
the degradation of calcium silicate hydrate (C–S–H) gel through leaching of the
calcium compounds. This process leads to loss of C–S–H gel stiffness and overall
deterioration of the cement paste matrix.

Khatib and Wild (1998) evaluated the effect of metakaolin on the sulphate resistance
of mortar. Two types of cements were used; high C3A and intermediate C3A content.
Cement was replaced with 0, 5, 10, 15, 20, and 25% of metakaolin (MK). Specimens
of size 150 × 150 × 150 mm were moist-cured in air for 14 days, and their length was
measured before immersing in 5% Na2SO4 solution. Test results demonstrated that:

(a) Expansion decreased systematically with the increase in MK content for both
types of cement.

(b) For mortars containing high C3A cement, the control mortar and mortars
containing 5% and 10% MK showed rapid expansion and deterioration between 40
and 70 days exposure to sulphate solution. Mortars containing 15% and 20% MK
showed small but quite sharp expansions also between 40 and 70 days, but
subsequently stabilized. By 520 days, mortar containing 20% and 25% MK showed no
significant changes in length.

(c) With intermediate C3A content in cement, trend was similar to the mortars made
with high C3A content. The control mortar (0% MK) showed significant expansion
followed by 5% MK mortar. Mortar containing 10% MK exhibited little expansion
and other mortar specimens containing 15, 20 and 25% were found to be extremely
stable, showing little overall change and no sign of deterioration after 520 days of
exposure in sodium sulphate solution.

Lee et al. (2005) reported an experimental study on the magnesium sulphate resistance
of mortar and paste specimens incorporating varying percentages of metakaolin (MK).

35
The resistance of mortar specimens was evaluated by expansion measurements. Four
different mortar mixtures were prepared. Cement was partially replaced with 0, 5, 10
and 15% of MK by mass. The water–cementitious material ratio (w/cm) was fixed at
0.45 by mass. The cementitious material/fine aggregate was invariant at 2.0 in all
mortar mixtures. Prisms (25 × 25 × 285 mm) were cast for expansion measurements.
All mortar and paste samples were immersed in magnesium sulphate solution (Mg) for
360 days. Sulphate solution concentrations used were 0.42% (3380 ppm), 1.27%
(10,140 ppm) and 4.24% (33,800 ppm), respectively.

They concluded that:

(i) At low solution concentration (0.42%) of magnesium sulphate (MgSO4), only the
MK5 mortar specimen showed an expansion value of over 0.1%.
(ii) At magnesium sulphate solution concentration of 1.27%, the mortar specimens
with 5% to 15% replacement levels of MK exhibited poor resistance to magnesium
sulphate attack. After 360 days of exposure, expansion values were 0.343, 0.339 and
0.469% for MK5, MK10 and MK15 mortar specimens, respectively, whereas the MK0
mortar specimens exhibited about 0.185% expansion after the same period of
exposure.
(iii) At high solution concentration of magnesium sulphate (4.24% MgSO4), the
expansion values for the MK0 mortar was about 0.244% after 360 days of exposure,
whereas mortar specimens with partial replacement levels of MK showed expansions
of 0.360%, 0.410% and 0.556% for MK5, MK10 and MK15 mortar mixtures,
respectively.
The sulphate resistance of MK concretes increase with the decreasing w/b ratio due to
the fact that the capillary pore volume decreases with decreasing the w/b ratio
(Akhras et al. 2006). The sulphate resistance of both non-air-entrained and air-
entrained concrete was found to increase with the MK replacement level (from 5% to
10% and 15%); based on measured expansion of concrete prisms and compressive
strength reduction of concrete tubes (Akhras 2006).

36