CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

Cement is the product of the silicate industry and is artificial on the major material in
concrete. It is used extensively in building residential, bridges and other structures. The
raw materials used in making cements are naturally occurring materials which includes
gypsum, CaSO4 + 2H2O, anhydride CaSO4 and limestone rock (Ajiwe et al., 2000).

With high cost of production and energy consumption to produce and also the major
source of greenhouse gas emission, it is time to look into the use of local inexpensive raw
materials or industrial waste materials in replacing cement to build the structures
especially housing for the needy. The utilization of the industrial waste by­ products such
as silica fume, slag and fly ash as well as the agricultural residue such as rice husk ash, as
cement replacement is a new trend in concrete technology in this 21st Century. Having
pozzolanic properties, not only that it gives technical advantage to the resulting concrete,
but it also reduces cement consumption. Thus, the resulting benefits in terms of energy
saving, economy, environmental protection and conservation of resources will be
substantial. The history of the utilization of rice husk ash in concrete goes back to 1946 in
the United States, by work carried out by Mc Daniel (1946). It described the manufacture
and behavior of blocks made from the mixtures of Portland cement, ash and rice husks. In
India, the utilization of RHA has been adapted by the Central Building Research Institute
in 1976 by burning the rice husk balls or cakes bonded in clay producing almost carbon-

9
free ash and then ground with hydrated lime to produce a cementitious material. (Mc
Daniel, 1946).

Cook (1986) quoted that the utilization of the ash in concrete as up to the year 1972 is to
act as a lightweight material and as insulating filler and not as a pozzolan. In 1972, Mehta
applied for a Belgian patent and was issued later in 1973 on the rice husk utilization and
several papers were also published (Mehta, 1983). Most of Mehta's work was on the
studies of the preprocessing parameters and their influence on the ash reactivity. At the
Asian Institute of Technology in Bangkok, studies on the laboratory-burnt RHA has been
investigated as a pozzolana in the form of an additive in cement mortar and concrete
(Columa, 1974). Study by Columa has made on RHA as a good replacement material for
cement in normal concrete. Meanwhile Mehta has carried out other cement-based material
since 1973 at the University of California at Berkeley, USA.

In developing countries especially in rice growing countries such as India, Pakistan and
Malaysia, the concept of cement based on rice husk ash provides impetus for renewed
research. This is revealed by the first workshop outlining the state-of-the-art of RHA
cements conducted in Peshawar, Pakistan in 1979, and a meeting in Malaysia in 1979
(UNIDOIESCAPIRCTI, 1979). Numerous researches were carried out then and
documented and published by the United Nations Industrial Development Organization
(UNIDO) in 1984 and the problems associated with it were also identified. It indicated that
there is a potential use of RHA cement (UNIDOIESCAPIRCTI, 1979). With the
utilization of pozzolanic materials such as silica fume, slag, fly ash, metakaolin and rice
husk ash as partial cement replacement, the demand on Portland cement would be reduced,
and thus the cost of cement can also be reduced.

Concrete structures are generally designed for a service life of 59 years, but experience
shows that in urban and coastal environments many structures begin to deteriorate in 20 to
30 years or even in less time. The lack of durable materials has serious environmental

consequences. Increasing the service life of products is long term and easy solution for
preserving the earth's natural resources (Sampaio, 2000)

10
The world's yearly cement production of 1.6 billion tonnes accounts for about 7% of the
global loading of carbon dioxide into the atmosphere. Portland cement, the principal
hydraulic cement in used today, is not only one of the most energy-intensive materials of
construction but also is responsible for a large amount of greenhouse gases. Producing a
tonne of Portland cement requires about 4GJ energy, and Portland cement clinker
manufacture releases approximately 1 tonne of carbon dioxide into the atmosphere.
Ordinary concrete typically contains about 12% cement and 80% aggregate by mass
(Sampaio, 2000).

During the 21st Century, the increasing demand for cement and concrete must be met by
the use of mineral cement replacement materials. Substantial energy and cost savings can
result when industrial by-products are used as a partial replacement for the
energy/intensive Portland cement. According to Mehta, PK (1994), the cement production
rate of the world is expected to grow exponentially to about 3.5 billion tonnes/year by
2015. According to his projection, most of the increase in cement demand will be met by
the use of mineral cement replacement material. He also suggested that this approach is
necessary to prevent the possible ecological disaster from global warming.

The presence of mineral cement replacement material in concrete is known to impart

significant improvements in workability and durability. A high performance concrete with
good workability and high durability can be made by a cautious choice of minerals cement
replacement material and concrete mix proportions. Some of these materials can be
obtained from by-products. The use of by products is an environmental-friendly method of
disposal of large quantities of materials that would otherwise pollute land, water and air
(Mehta, P.K, 1994). The use of the artificial pozzolans can achieve not only economical
and ecological benefits but technical benefits as well. It is generally agreed that, with
proper selection of admixture, mixture proportioning and curing technique, minerals
additives can greatly improve the durability of concrete.

In this chapter, the definition of High Performance Concrete (HPC) and its importance to
the construction industry is discussed. The principal parameters of HPC in its production,
use and the problems of using HPC is also discussed and reviewed.

11
This chapter also briefly gives an introduction to the pozzolanic materials use and its
performance in the production of HPC. As literature discussion, existing knowledge on the
effect of pozzolanic materials used as replacement material to produce binary blended
cement system in HPC, which are related to this study, are reviewed. Properties and
performance of Ordinary Portland Cement and multiple blended binders cement
containing pozzolans used to produce HPC was discussed. Finally, this chapter also
discusses the role, potential and importance of using multiple blended binders as an
alternative binders to OPC in producing HPC.

The concept of a PFA and cement system is not new and many patents have been
published by many researchers. Chindaprasirt et al, (2007) found that the strength
development of concrete containing fly ash as a cement replacement of 15%, 25%, and
35% was faster than that of the 50% cement replacement, while 25% cement replacement
gave the highest compressive strength at all ages. At 15–35% content replacements, the
compressive strengths were higher than that of control concrete at all ages up to 180 days.
At 28 days, the compressive strength tended to increase with the curing age for all
mixtures and varied from 77.3 MPa in sample of 50% fly ash to 82.5 MPa in sample of
25% fly ash. This is due to the extreme fineness of fly ash that exhibits pozzolanic
properties and packing effect. These characteristics tend to improve concrete strength as
well as its density (Angsuwattana et al, 1998).

In 2004, (Qingge Feng et al, 2004) found that with hydrochloric acid pretreatment of rice
husks, the pozzolanic activity of rice husk ash is not only stabilized, but also the enhanced
the sensitivity of the pozzolanic activity of the rice husk ash to burning conditions is
reduced. The pozzolanic activity of ADR (acid-treated rice husk) is slightly affected by the
change of

maintaining time, but the maintaining time has a great affect on the pozzolanic activity
of RHA (no pretreatment). However, from the age of 7 days, the Ca(OH)2 content in the
cement pastes with RHA and ADR is lower than that of the control paste, though the
cement content in the cement mortar with RHA and ADR at 10% replacement level is
lower than that in the control. This is because of the pozzolanic reaction in the cement
mortar with RHA and ADR (Qingge Feng et al, 2004)

12
According to Nuruddin et al, (2008), compressive strength development of MIRHA
concretes were significantly higher compared to control concrete. The adequate amount of
water and high pozzolanic reactivity were believed to be the main cause of this faster
acceleration of MIRHA concrete. During this stage, MIRHA performed its function both
as a pozzolanic material and filter (Nuruddin et al, 2008).

2.2 Cement

Cement in its general term, can be described as a material with adhesive and cohesive
properties which make it capable of binding the material fragments into the compact
whole. Basically, the raw materials used in the manufacture of Portland cement consist
mainly of lime, silica, alumina and iron oxide. These compounds interact with one another
in the kiln to form a series of more complex products and apart from a small residue of
uncombined lime which has not had sufficient time to react, a state of chemical
equilibrium is reached.

Neville (1997) described cement as a material with adhesive and cohesive properties that
make it capable of bonding mineral fragments (stones, sand, bricks, building block) into a
compact whole. The cement referred here is the hydraulic cement having the property of
setting and hardening under water by virtue of a chemical reaction with it. The principal
constituents of this cement are mainly of silicates and aluminates of lime. This hydraulic
cement which is commonly known as 'Portland' cement is due to its resemblance of the
colour and quality of the hardened cement to Portland stone, limestone quarried in Dorset,

United Kingdom. The various types and classifications of Portland cement and its
properties are stipulated in accordance with ASTM C 150-2005 and BS EN 197-1:2000.

Many types of cements have been developed to ensure good durability of concrete under a
variety of conditions. Table-2.1 shows a list of different types of Portland cement in the
British classification together with the American classification (Neville, 1997)

13
Table-2.1. : Some Main Types of Portland Cement
British Classification American Classification
Description BS Description ASTM
1.) Ordinary Portland 12:1978 Type I C 150-84
2.) Rapid-hardening Portland 12:1978 Type III C 150-84
3.) Low-heat Portland 1370:1979 Type IV C 150-84
4.) Sulphate-Resisting Portland 4027:1980 Type V C 150-84
5.) White Portland 12:1978 Type I P C 150-84
4627:1970

2.2.1. Ordinary Portland Cement

In the new era of concrete industry, Ordinary Portland Cement (OPC) was still remains as
a major binder in hydration process to produce High Performance Concrete. It has been
used as total binder in concrete mixes or as binder proportions in blended cements. The
main compounds of OPC named as tricalcium silicate (C3S), dicalcium silicate (C2S),
tricalcium aluminate (C3A) and tetracalcium aluminoferrite (C4AF) (Handoo, Mahajan
Kaila 2003).

C3S comprises angular crystal content about 52% of OPC volume. It is responsible on
initial setting and rapid strength gain especially to give an early strength (for example, 7
days). C2S

is more rounded crystal content about 19% of OPC volume and it is responsible for long
term strength. It will harden slowly, but contributes notably to strengthen at ages over a
month (Ajiwe et al, 2000). While C3A may be in rectangular or amorphous crystal forms,
it is responsible for rapid setting and C4AF, a non-crystalline composition, is responsible
for grey colour with little contribution to setting or strength as placed surrounding the
cement matrix content about 10% and 8% of OPC volume respectively (Taylor,G.D.

14
2002). The reaction of C3A with water is very violent and leads to immediate stiffening of
the paste, known as flash set. To prevent the flash setting phenomenon, gypsum will
normally be added to cement clinker. The presence of C3A with its rapid setting, high heat
emission and sulphate susceptibility, is undesirable in concrete.

The actual proportions of the various compounds vary considerably from cement to
cement, and indeed different types of cement are obtained by suitable proportioning of the
raw materials. The major composition of OPC is lime, silica, alumina and iron oxide. With
the presence of water these compounds interact with one another to produce hydrated
product which is Calcium Silicate Hydrate (C-S-H) and Calcium Hydroxide (Ca[OH2]).

The C-S-H takes the form of extremely small interlocking crystals which grow out slowly
from cement grains to occupy previously water-filled spaces. The microcrystalline
material is responsible for strength in the harden concrete (Regourd,M.M 1992).

Ca(OH)2 forms in a much larger crystal that acts as fillers in the hardened concrete but do
not interlock to form strength. In the presence of moisture in the concrete matrix, Ca(OH)2
will partly dissolve to form an alkaline solution that is useful to protect reinforcement in
the reinforced concrete structure (Regourd,M.M 1992). The ratio of C-S-H to Ca(OH)2 is
approximately 7:2 by mass of concrete (Neville, 1997).

At any stage of hydration, the hardened paste consists of very poorly crystallized hydrates
of the various compounds. It is referred to, collectively, as gel, crystals of Ca(OH)2, with
some

minor components, unhydrated cement and the residue of water-filled spaces in the fresh
paste (Neville, 1997). As Ca(OH)2 is crystallised in massive superimposed hexagonal
plates, it has created a capillary pore in the cement paste matrix. The capillary pore has
been generated either by Ca(OH)2, air bubbles or micro crack that has become a factor,
attributed to low engineering properties and performance of concrete (Regourd,M.M
1992).

15
The capillary pores represent a part of the gross volume which has not been filled by the
products of hydration. Commonly, the hydration product of OPC occupy twice the volume
of the original solid phase, therefore, the volume of capillary system is reduced with the
progress of hydration (Neville, 1997). The hydration progress depends on water/cement
ratio and on the degree of hydration. Water/cement ratios lower than 0.23 would have self-
desiccation problems and a water/cement ratio higher than 0.36 used the capillary pores
will occur since the volume of the gel is not sufficient to fill all the space available
(Taylor,G.D. 2002).

An improvement has been obtained by several processes which reduce the porosity and the
water/cement ratio. One of the process as introduced is blending OPC with pozzolanics
materials named as pozzolanic cements or pozzolanic blended cement. The small particles
of pozzolans will generate a large surface area for the precipitation of the hydration
product, and make the cement paste become more homogeneous and dense as for the
distribution of the finer pores. This is due to the pozzolanic reactions between the
amorphous silica of the mineral addition and Ca(OH)2 produced by cement hydration
reactions.

Ordinary Portland Cement (Type I) is admirably suitable for use in general concrete
construction when there is no exposure to sulphates in the soil or in ground water. The
specification for this cement is given in BS 12:1978 (British Classification). In addition to
the main compounds mentioned above, there exist minor compounds like manganese
oxide, magnesium oxide, sodium oxide and potassium oxide. They usually amount to less
than a few percent of the weight of cement. It should be pointed out that the terms 'minor

compound' refers primarily to their quantity and not necessarily to their importance.

2.2.2 Chemical Composition of Ordinary Portland Cement

The effect of Portland Cement on concrete strength depends on the chemical composition
and fineness of the cement. There are four main types of compounds which are considered

16
as the major constituents of Portland Cement, and they are :

 Tricalcium silicate, 3CaOSi0 2

 Dicalcium silicate, 2CaOSi02
 Tricalcium aluminate, 3CaO.Al203
 Tetracalcium alumineferrite, 4CaO.Al203.Fe203

2.2.3 Hydration of Cement

This is a reaction by virtue of which Portland cement becomes a binding agent and takes
place in the cement paste. In other words, the presence of water will cause the silicates and
aluminates to form products of hydration which in turn produces a firm and hard mass as
the hardened cement paste.

The hydration reaction for various silicates and aluminates are shown below:
(not to their exact stoichiometric equation)

(i) For tricalcium silicates

2(3CaO.Si02) + 6 H20 = 3Ca0.2Si02. 3H20 + 3 Ca(OH) 2

(ii) For dicalcium silicates

2(2CaO.SiO2)+ 4 H2O = 2Ca0.2Si02.2H2O + 3 Ca(OH)2

(iii) For tricalcium aluminate

3CaO.Al203 + 6 H2O = 3 CaO.Al203.6H20

The hydration of tricalcium silicate and dicalcium silicate is primarily responsible for the
strength of cement paste. Tricalcium silicate can be assumed to contribute most to the
strength development during the first four weeks and dicalcium silicate influences the gain
in strength after the next four weeks. Tricalcium aluminate contributes to the strength of
cement paste at one to three days and possibly longer.

17
2.3 High Performance Concrete (HPC)

HPC has been widely used in construction industry nowadays. This is due to the
increasing demand for more durable concrete to extend the service life and at the same
time reduces maintenance cost of the concrete structure.

There have been many definitions of HPC. However, the definition has not been
standardized. Nevertheless, there are several definitions of HPC proposed. ACI (2002),
defines HPC as, concrete meeting special combinations of performance and uniformity
requirements that cannot always be achieved routinely using manpowerl constituents and
normal mixing, placing and curing practice. The ACI has also specified the compressive
strength for design of 41 MPa or greater.

HPC has also been defined as concrete with water cementitious materials with ratio less
than or equal to 0.35. The compressive strength characteristic is greater than 35 MPa or
equal after 24 hours and after 28 days the compressive strength should be greater or equal
to 70 MPa. HPC has also been defined based on special requirements and properties such
as, ease to placement and compaction without segregation, enhanced long-term
mechanical properties, high early-age strength, high toughness, volume stability and long
life in severe environments (Hashem et al, 2002).

To achieve an ideal HPC, EG Nawy (2000) has stated five principal parameters namely i)
high performance, ii) economy, iii) resistance to wear and deterioration, iv) resistance to
weathering and chemicals and v) appropriate cement type. Among all parameters stated
above, appropriate cement type has become the main parameter that can influence the
performance of other parameters. Unappropriate type of cement used in producing a HPC
might cause disintegration of the concrete in the structure.

Mailer.Y (1992) introduced two approaches in order to obtain HPC which reduces the
flocculation of cement grains and widens the range of grain size.

18
The advantages of using cement additives in concrete are, mainly, they improve concrete
properties in fresh and hardened states, and economically and ecologically beneficial. The
achievement of these advantages becomes more important for HPC proportioning.
However, the selection of additives needs more attention due to their different properties.

2.3.1 Cement for HPC

In achieving HPC, choosing an appropriate type of cement is important. The type of

cement used depends on the type of structures, the weather and other conditions under
which the structure is built and that it will exist during its life span (EG Nawy (2000).

Concrete which is exposed to the seawater sprays required a sulphate resisting cement type
(Maher, A.Bader 2003). For construction that needs concrete to be quickly hardened,
rapid hardening Portland cement will be essential (Taylor,G.D. 2002). EG Nawy (2000)
also stated that concrete structure which is bulkier and heavier in cross-section needs less
heat of hydration cement to prevent shrinkage problems and surface crack. Blended
pozzolanic cement was found to be necessary in use especially to reduce temperature that
rises during hydration process. At the same time, it improves the durability and
engineering properties of concrete (Sabir et al, 2001).

The chemical compositions of cement have an influence to its properties and

performances, but it also depends on its type and usage. The percentage variation in the
chemical composition of each type gives concise reasons for the difference in reaction
when in contact with water.

The size of the cement particles was also found to give an effect on the rate of reaction of
cement with water (Aitcin, 2003). Larrnard (1992) said that ultra fine particles of cements
will react in two levels which are physical level and chemical level. In physical level the
cement will react as filler between a void in hydrated cement matrix at early age. In the
chemical level the cement will accelerate the hydration process. For blended cement

19
containing pozzolanic materials, the finer particle produces better pozzolanic reaction, but
it is still depending on the quality of pozzolanic materials used. Mazloom et a1, (2004)
also stated that in obtaining HPC, the water/binder ratio must be reduced and binder
content must be increased.

2.4 Cement Replacement

The materials for making concrete consist of cementitious binder, aggregates, water and in
most cases with ready mixed concrete, one or more types in chemical admixtures. Today,
the cementitious binder such as Ordinary Portland Cement or Sulphate Resisting Portland
Cement is often blended with ground granulated blastfurnace slag. In general, pulverized
fuel up to 30% and ground granulated blastfurnace slag up to 70% with the balance that
made up of are used depending on the intended applications.

Within the past decade, silica fume has been introduced, in addition, to the other mineral
admixtures to improve the performance of concrete. These mineral admixtures are
chemicals that provide cementitious gel similar to those produced by the hydration of
Ordinary Portland Cement. Hence they are often called supplementary cementing
materials (Neville, 1997).

In recent years, the use of pozzolanic materials and slag as replacement for cement in
concrete has become more and more widespread throughout the world. Particularly, in
countries where such materials are produced as by-products of the industry, such usage
has the added value of providing a partial solution to the problem of disposal of such
materials. In the following sections, discussion on these supplementary cementing
materials will be confined to, silica fume and ground granulated blastfurnace slag. They
are used together with ordinary Portland cement which provides the calcium hydroxide
needed for the pozzolanic reaction (Neville, 1997) .

20
2.5 Pozzolanic Materials

Pozzolan is a material which, when combined with calcium hidroxide, exhibits

cementitious properties. Pozzolans are commonly used as an addition or as mineral
replacement to Portland cement concrete mixtures to increase the long-term strength and
other material properties of Portland cement concrete. Pozzolans are primarily vitreous
siliceous materials which react with calcium hydroxide to form calcium silicates; other
cementitious materials may also be formed depending on the constituents of the pozzolan
(Agarwal, 2006).

The specific definition of pozzolan was made by ASTM C618-98 and accepted by all
scientists and researchers is a "siliceous or siliceous and aluminous materials, which in
itself possesses little or no cementitious property, but which is finely divided form and in
the presence of moisture, chemically react with calcium hydroxide at ordinary temperature
to form compounds possessing cementing properties". Pozzolan has also been defined as
any materials, regardless of its geologic origin, that possesses hydraulic properties (Ali
Akhbar, 1987). Neville (1997) described pozzolans as a natural or artificial material
containing amorphous silica in a reactive form. The silica can combine with calcium
hydroxide of OPC in the presence of water to form stable calcium silicates which have
cementitious properties.

The first known pozzolan was pozzolana, a volcanic ash, for which the category of
materials was named. The most commonly-used pozzolan today is fly ash (FA), though
silica fume (SF), high reactivity metakalolin (MK), ground granulated blast furnace slag
(GGBS), rice husk ash (RHA) and other materials which are also used as pozzolans. In
this study, it will only cover pozzolans named as silica fume (SF), fly ash (FA),
metakaolin (MK) and rice husk ash(RHA). All of the pozzolans discussed was a pozzolan
which is available in Malaysia. SF was already commercialised and widely used in cement
and concrete industries. Other pozzolans are abundant in the country and their usages are
still very few in the industries.

21
2.5.1 Classification of Pozzolan

ASTM C618-98 classified pozzolans in to Class N, Class F and Class C. Class N is

referred to as natural pozzolan such as diatomaceous earth, opaline cherts, shales, turffs,
volcanic ashes and clays. The pozzolan can be calcined or not depending on its properties
and compositions. Class F pozzolan is referred to as FA produced from burning
bituminous coal and Class C pozzolan is an FA that is produced from subbituminous coal.
Class C pozzolans contain pozzolanic properties and have some cementitious properties.
In Canada, the FA is classified due to the lime content composition. FA with lime content
8%-20% will be classified as CI and classified as C when the lime content is higher than
20%.

European standard ENV 197-1:1992 recognizes two subclasses of pozzolanic cements

named as Portland fly ash cement Class II/A-V and Class II/B-V. Class II/A-V contains 6
to 20% of fly ash and Class II/B- V contains 21 to 35% of fly ash.

Pozzolanic materials are also classified as natural pozzolans and artificial pozzolans.
Natural pozzolans are divided into pyroclastis rocks such as zeolited materials and clastic
rocks such as calcine clay. FA, SF, burnt agro residue and calcined shales are grouped
under artificial pozzolans (Cook D.J, 1986).

Ali Akhbar (1987) stated in his literature on Massazza (1974) paper that the pozzolans are
divided into three groups. The first group includes pyroclastic rock which is material of
volcanic origin. The second group comprises altered materials with high silica content and
the third group includes materials of clastic origin including clays and diatomaceous
earths. Artificial pozzolans such as FA, SF and MK are included in group three pozzolans
in Massazza's classifications and RHA are included in group two due to the alteration
process of rice husk into ash.

Malhotra et al, (2004) classified pozzolanic materials under mineral admixtures that refer
to mineral materials other than aggregates and cement that added immediately before or
during concrete mixing or during manufactured of cement The mineral must have a

22
siliceous or siliceous and aluminous materials that may possess little or no cementitious
property.

2.5.2 Standard Specification of Pozzolan

There are many standard specifications of pozzolanic materials and pozzolanic cement that
have been prepared all around the world. However, for this study, only British Standard
[BS6588:1991 and BS6610:1991] and American Standard [ASTM C618-98]
specifications will be focused

2.5.2.1 British Standard

BS6610:1991 is a British standard specification for Pozzolanic cement. The BS has

defined the pozzolanic cement as blended hydraulic binder comprising homogeneous
mixture of ordinary Portland cement and pozzolana in specific proportions. The
pozzolanic materials are referred as pulverized FA where the proportions of FA are not
more than 50% or less than 3 5% by mass of the total quantity.

To indicate that the pozzolanic cement complies with the standard specifications
requirement, 5 tests on samples of the cement named as proportion of FA by mass to the
nearest 5%, the fineness, the compressive strength at 7 days and 28 days, the initial and
final setting times and the soundness test are to be carried out. The requirements of
standard specifications on all 5 tests are stated in Table 2.2. X is referred to as sample of
pozzolanic cement test.

BS 6588:1991 is a specification standard for Portland pulverized-fuel ash cement. Portland

cement and the pulverized-fuel ash have been thoroughly and intimately mixed together in
a dry state to form a uniform mixture before or after grinding. The specification
requirements of the cement are stated in table 2.3.

23
Table 2.2: BS6610: 1991 Standard Specifications Requirement of Pozzolanic Cement
Type of testing BS6610 requirements
Proportion of FA 35% < x < 50%
Fineness x > 225 m2/kg
Compressive Strength At 7 days: x > 8 N/mm2
At 28 days: x > 16 N/mm2
Setting Time Initial: x > 45 minute
Final: x < 10 hour
Soundness x < l0mm

Table 2.3: BS6588: 1991 Standard Specifications Requirement of Portland Pulverized-fuel

Ash Cement.
Type of testing BS6610 requirements
Proportion of FA 15% < x < 35%
Fineness x > 225 m2/kg
Compressive Strength At 3 days: x > 8 N/mm2
At 28 days: x > 22 N/mm2
Setting Time Initial: x > 45 minute
Final: x < 10 hour
Soundness x < l0mm

2.5.2.2 American Standard

American Society for Testing Materials (ASTM) C6l5-98 is a standard specification for
coal FA and raw or calcined natural Pozzolan for used as a mineral admixture in concrete
where cementitious or pozzolanic action is desired. The standard requirements have been
divided into two groups of requirements that are chemical requirements and physical
requirements.

24
The standard has divided the pozzolans into 3 classes, namely Class N, Class F and Class
C. The requirements of the standard for all pozzolans classes are shown in Table 2.4 and
Table 2.5. All samples and test of the mineral admixture are as in accordance with the
requirements of test methods ASTM C31l. Although ASTM C618-98 has stated that the
loss of ignition (LOI) is less than 6%, a footnote actually allows up to 12%.

ASTM C1240-04 is a standard specification for silica fume used in concrete and other
systems containing hydraulic cement. Requirements stated for SF in this standard is also
based on chemical requirements and physical requirements. Chemical requirements for SF
are based on minimum content of Si02 which is 85%. The moisture content and LOI of SF
has been fixed to be not more than 3% and 6% respectively.

The physical requirements of SF are based on three types of physical requirements namely
maximum size of SF allowed, pozzolanic strength activity index and minimum specific
area of SF used. The maximum size of SF is identified through sieve analysis by
determining the percent of SF retained on 45μm sieve. It was fixed to be not more than
10%.

Pozzolanic strength activity index is determined by comparing the compressive strength of

concrete containing SF with OPC concrete at 7 days. The compressive strength of SF
concrete must be higher than OPC concrete at minimum 5%. Minimum specific surface
area of SF is fixed to be 15m2/g using BET, nitrogen absorption method.

Table 2.4: Chemical Requirements of Pozzolanic Materials (modified from AS'I'M C6l8)

Class of Pozzolanic Materials N F C

SiO2 + Al2O3 + Fe2O3 , min,% 70 70 50
SO3, max,% 4 5 5
Moisture Content, max,% 3 3 30
Loss on ignition, max, % 10 6 6
Na2O, max, % 1.5 1.5 1.5

25
Table 2.5: Physical Requirement of Pozzolanic Materials (modified from ASTM C618)

Class of Pozzolanic Materials N F C

Fineness, amount retained on 45 μm wet sieve (%) 34 34 34
Minimum strength activity index at 7 days 75 75 75
compared with OPC, percent of control

Minimum strength activity index at 28 days 75 75 75

compared with OPC, percent of control

Water requirement, max, percentage of control 115 115 105

Soundness, Autoclave expansion, max, % 0.8 0.8 0.8
Density, max variation from average,% 5 5 5
Percentage retained on 45μm, max, % 5 5 5

2.6 Pozzolanic Cement

When the minerals blended with lime or with Portland cement, they are called pozzolanic
cement or blended cement. The pozzolanic cement comprises a mixture of OPC with any
pozzolanic materials. These materials are to be thoroughly mixed together in dry or wet
state to form a uniform mixture in any proportions. BS6610: 1985 has defined pozzolanic
cement as a blended hydraulic binder comprising a homogenous mixture of Ordinary
Portland Cement and pozzolan in specified proportions. The composition of pozzolan in
the pozzolanic cement has been proposed to be the maxima of 50%. Pozzolanic cement
strength has been reported to be as low 7 and 28 days strength compared to OPC. The
minimum strength on 7days is stated as not less than 8 N/rnm2 and at 28 days the strength
is fixed to be not less than 16N/mm2 as a specification requirement of pozzolanic cement
stated in BS 6610: 1985. ASTM C618-98 has allowed the strength of pozzolanic cement to
be not less than 25% compared to OPC strength at 7 and 28 days as the standard
specifications of pozzolanic cement to be used in the industry.

26
The effect of pozzolanic materials in concrete is generally divided into three effects. They
are the effects of providing a bigger surface area for hydration process, the filler effect and
the pozzolanic reaction effect. These three effects have been performed either during
concrete at fresh stage or at the hardened stage. The effects are also reported to give
effects on the engineering properties and durability properties of concrete.

During fresh stages, the use of pozzolanic materials is generally reported to reduce the
workability and the bleeding of concrete due to the increments of cohesiveness value of
concrete as a result of bigger surface area of the pozzolanic materials compared to OPC.
The fineness of particle of the pozzolanic materials has also led to the filler effect between
the transition zones in the concrete matrix. It has also provided a bigger surface area for
hydration process and reduced the porosity volume that has been responsible for the early
age strength of the concrete.

The pozzolanic reaction may be slower than the rest of the reactions which occur during
cement hydration, and thus the short-term strength of concrete made with pozzolans may
not be as high as concrete made with purely cementitious materials. However, it depends
on the compositions and properties of the pozzolans. On the other hand, highly reactive
pozzolans, such as silica fume and high reactivity metakaolin can produce "high early
strength" concrete that increase the rate at which concrete gains strength (Sabir et al,
2001).

2.6.1 Engineering Performance

The effect of pozzolans on engineering properties of HPC containing pozzolans is referred

to effects of pozzolans studied on properties of fresh concrete and hardened concrete. The
properties of fresh concrete were focused on the effect on water demand, workability,
bleeding and setting time. Hardened concrete properties are referred to as compressive
strength, Yong's modulus of elasticity, porosity and expansion or shrinkages.

27
2.7 Silica Fume (SF)

SF is a by-product of the manufacturing process of silicon or various silicon alloys and

silicon metal. ASTM C1240-04 has defined SF as a very fine pozzolanic material,
composed mostly of amorphous silica produced by electrical arc furnace as a byproduct of
the production of elemental silicon or ferro-silicon alloys. It is also known as condense
silica fume and microsilica.

The chemical composition and properties of SF depend on the composition of the principal
product being made by the furnace and furnace design Malhotra et al, (2004). Usually, SF
contains more than 80% to 85% of silica in non-crystalline form and has a spherical shape
with average particles size of 0.1- 0.5 µm and nitrogen BET surface of 20,000 m2/kg
(Yajun et al, 2003).

The use of SF as minerals admixture or as replacement materials in concrete industry has

been increasingly used in many mega projects all around the world. Many researchers
reported the SF is highly reactive pozzolan that has improve the concrete properties. The
effectiveness of using SF in increasing the engineering properties and concrete durability
makes the SF selective materials especially to be used in obtaining high performance
concrete.

SF is dry, densified and used as mineral admixture formulated to produce concrete with
special performance qualities. It has been found to improve the properties of hardened
concrete. These improvements are in two ways. Firstly it acts as filler between cement
particles and secondly, it acts as pozzolanic materials which react chemically within the
concrete to increase the amount of calcium silicate hydrate gel formed, thus improving the
strength and reducing the permeability of the concrete.

Although SF could impart significant contributions to the strength and chemical resistance
of concrete, it is also reported leading to increases in water demand, placing difficulties
and plastic shrinkage problems in concrete (Thomas et al., 1999). Due to the high demand

28
for this material in the concrete industry, SF has become significantly expensive. The
increase of construction costs due to the increasing price of imported silica fume compared
to other admixtures that has led researchers to turn their interest toward other
supplementary cementitious materials and technique to produce similar strengthening
effects as silica fume.

2.7.1 Chemical Composition of SF

SF is a byproduct of the manufacturing process of silicon or various silicon alloys, having

a high content of Si02 in amorphous form. Chemical composition of SF depends on the
composition of the principal product being made by the furnace and furnace design
(Mehta, 1986). A furnace which is equipped with a heat recovery system produces SF with
lower value of ignition loss (LOI) or carbon content. The LOI of SF is on the range of 2.41
% to 2.75%.

Even though Table 2.5 stated that the SiO2 content of SF is 85.49%, it may contain more
than 80% of Si02. The amount of SiO2 in SF was reported to vary depending on silicon
content in the alloy production. The higher the silicon content in the alloy used, the higher
the silica content in the resulting silica fume (Neville, 1997). The other compositions
contained in SF such as Al2O3, Fe2O3, CaO and alkali contents are low. Even though there
is MgO content in a compound of SF, it was reported not to be deleterious to concrete.

The chemical composition of SF was not affected by time. The constancy in chemical
composition is due to the relatively pure raw materials used in a production of silicon
metal or ferrosilicon alloys (Mehta, 1989).

2.7.2 Physical Characteristic of SF

The particle size of SF was reported to be from 0.1 to 0.5 µm and the nitrogen BET
surface is 20,000 m2/kg (Yajun et al., 2003).

29
SF has a low bulk density around 2.0 to 3.0 kg/m3. The low bulk density of SF may cause
difficulty in transporting and handling. In order to improve the handling and transport
properties, SF was processed to increase its bulk density by densified, sluried or alletized
process.

Although SF could impart significant contributions to strength and chemical resistance of

concrete, it could also lead to increase in water demand, placing difficulties and plastic
shrinkage problems in concrete (Thomas et al., 1999). The agglomeration particle of SF in
cement paste or mortar was also reported to decrease the chemical reaction of SF during
hydration process. This agglomeration of SF can reduced its effectiveness through having
a larger particle diameter, a smaller surface area (SSA) and a lower pozzolanic reactivity
than the unitary grains (Boddy et al, 2000).

2.7.3 Effects of SF on Fresh Concrete

The utilization of SF reported to give a significant effect to the properties of fresh

concrete. The high specific surface area of SF that is bigger than OPC particles becomes
the main factor of the effect. The large specific area of SF increased their capacity to
absorb water that result more water being used in the mixture to maintain the workability
of concrete. Malhotra et al, (2004) reported that the utilization of SF in concrete mixes
would give a net effect of increased water requirement compared to OPC concrete with a
same level of consistency.

Mazloom et al (2004) in their study also indicate that as the proportion of silica fume
increased, the workability of concrete decreased (Megat Johari et al, 2002). They report
that the effect of replacing part of cement with 5% to 10% SF is to improve the
workability but as a replacement level of SF increase the workability is reduced.

The water absorption effects due to large specific surface area of SF will also control the
mixes with SF to have bleeding problems. This is due to small particles of SF that will fill
in the voids between the large particles of cement and absorb the entrapped water during

30
hardening of concrete. In the concrete mix with high cementations content and low
water/binder ratio, the cohesiveness of the paste increased due to the reduction of internal
bleeding in the mixture (Edward, 2001). The high cohesiveness of concrete containing SF
may result the concrete to having very little bleeding or even none.

The effect of using SF to the setting time of cement paste or concrete depended on level of
replacement and water/binder ratio Ali et al (2007) in his investigation reported that the
addition of 5 to 10% of SF had no significant effected on setting time but in a concrete
with water/cement ratio of 04 and 15% replacement level of SF the setting time was
delayed. The retardation effect (initial and final setting time) will increase with higher
replacement of SF to OPC and the retardation effect of SF to cement paste or concrete is
due to the decreased in cement content which is responsible for early stiffening of the
paste (Alshamsi et al, 1997).

The investigation of De Almeida and Goncalves (1990), as reported by Megat Azmi

(2000), showed that there is no significant difference in setting times of concrete when SF
introduced to non­superplasticised concrete, but for a concrete with low water/binder ratio
containing superplasticize, the effect of SF was to reduce the setting time of concrete
compare to control concrete with superplastized. The reduction of setting time in SF
concrete is due to an interaction between SF and the superplasticized used.

2.7.4 Effects of SF on Hardened Concrete

The effect of SF on the properties of harden concrete was established by many researcher.
SF was recognized as materials that can contributed significantly to the compressive
strength development of concrete. The effective pozzolanic reactions and the filler effect
between the transition zones were known as a reason to a better compressive strength
performance. The main contribution of SF to compressive strength development at normal
temperatures takes places between ages of about 3 and 28 days. The overall strength
development of SF concrete are varying according to concrete proportions, composition
and curing conditions Malhotra et al, (2004).

31
The compressive strength development of SF concrete was reported to be high during 3 to
28 days of curing period. However from 28 to 90 days the relative strength of the concrete
is relatively low (Mehta, 1986). Mazloom et al., (2004) also indicated that the compressive
strength of concrete mixtures containing silica fume did not increase after the age of 90
days.

The effect of SF as cement replacement materials to compressive strength of concrete

reported to improve the short-term mechanical properties such as 28-day compressive
strength. This effect is achieved through the combination effect of chemical and physical
effects provided by SF. The chemical effect of SF that is referred to the pozzolanic
reaction during cement hydration process will change the calcium hydroxide to calcium
silicate hydrate faster than OPC because of high surface area and high content of
amorphous silica (Peter, 1998). The fine particles of SF played a role to fill in voids
contents in the concrete matrix has also been reported as a factor of a better compressive
strength of the SF concrete (Roszilah et al., 2002).

The addition of silica fume to HPC seems to reduce the rate of increase of the modulus of
elasticity with age. The reason for this is due to the high rate of hydration of concrete
containing silica fume. At an early age, silica fume concrete has higher strength gained but
gradually decreases to ordinary concrete. Hence, the elastic modulus at an early age is
higher, with a gradual decrease over time. Hani et aI., (2005) have their investigation
results show that adding silica fume resulted in an increase in strength and modulus at
early age However, there has been no change in the modulus at 28 and 56 days.

The reduced of bleeding in SF concrete can lead to plastic shrinkage cracking under
drying condition. Results indicated that the plastic and drying shrinkage of concrete with
silica fume cement concrete specimens were more than those in the plain cement concrete
specimens (Neville, 1997). Therefore, proper curing technique was proposed to prevent
the problem from occuring when SF is used. The shrinkage strains in SF cement concrete
specimens cured by continuous water-pending were reported to be less than that in similar
concrete specimens cured by covering them with wet burlap (Al­ Amoudi et al., 2007).

32
The percentages of SF replacement did not give a significant influence on total shrinkage;
however, the autogenous shrinkage of concrete increased as the amount of SF increased.
The results have also indicated that the increase of the proportion of SF lowers the amount
of expansion. (Mazloom et al., 2004)

2.8 Fly Ash (FA)

FA is a finely divided residue that is a by-product of the combustion of ground or

powdered coal exhaust fumes of coal-fired power stations. In certain places, FA is also
known as

pulverized fuel ash (PFA) and it was found to possess pozzolanic properties due to
contents of SiO2 and Al2O3 (Xinghua et al., 2002). This material represents a substantial
reserve of pozzolanic materials if it can be fully recovered. It is generally finer than
cement and consists mainly of glassy-spherical particles.

2.8.1 Chemical Composition of FA

Combustion of ground or powdered coal will produce a residue named fly ash (FA). The
chemical composition of FA may vary from one batch to another. This, however depends
on the minerals associated with the coal and the burning condition during the combustion
of FA. In general, FA contains SiO2, Al2O3 and Fe2O3 and the amount of these three
compositions has been stated as the main requirement of ASTM C 618-94 in determining
the classification of FA. Ravindra, (1986) stated that the classification of FA through the
amount of SiO2, Al2O3 and Fe2O3 is confusing, since it was found that many class C FA
meet class F requirements. It is also found that FA relatively contains more alumina and
less silica as compared to other pozzolans (McCarthy et al., 2005).

33
CaO is also contained in FA but its composition depends on the type of coal used.
Thecomposition of CaO was found to be low, as less than 10% in bituminous coal and
more than 10% in sub bituminous coal (Neville, 1997). The CaO composition will
establish either the FA having a cementitious property or not. FA with high level of CaO
will have cementitious properties as an additional to pozzolanic properties while FA with
low level of CaO only has a pozzolanic properties (Ravindra, 1986). The amount
composition of CaO is also used to identify the classifications of FA. FA with low CaO
has been classified as class F and high CaO content classified as Class C.

Other than the four composition discussed before others, composition such as MgO, Fe2O3
alkalise and carbon have been also determined. The carbon content is assumed to be equal
to the LOI. Even the amount of MgO of FA is the highest amount the pozzolans but the
MgO is not harmful because it exists in a non-reactive form (Neville, 1997).

2.8.3 Physical Properties of FA

FA consists of glassy spherical particles with some crystalline matter and carbon in the
form of un-burnt coal which varies from plant to plant (Ali Akhbar, 1987). The particle
diameter of FA is between less than l μm and 1000 μm with an average size of 20μm, and
the specific surface measure of FA is usually between 250 and 600 m2/kg and the overall
value of specific gravity is 2.35 (Neville, 1997).

Mineralogy analysis of FA typically contains about 50%-90% of glass. The reactivity of

glass in FA depends on the chemical composition especially the CaO content. The typical
crystalline minerals of low calcium FA are quarzt, mullite, sillimanite, hematite and
magnetite. These minerals do not posses any pozzolanic activity. High calcium FA
contains minerals that may react with water which is tricalcium aluminate, calcium
aluminosulfate, anhydrite, free CaO, and alkali sulfates. High calcium FA also contains
quarzt and periclase but these two minerals do not give any effects to the reactions
(Malhotra et al, 2004).

34
2.9 Rice Husk Ash as Pozzolanic Material

Rice husk ash is a general term used to describe the type of ash produced from burning
rice husk. Rice husks are the shells produced during the de-husking operation of paddy
rice, which is the by-product of rice paddy milling industries. Studies conducted by Ankra
(1976) at the University of California, Berkeley as reported by Ismail and Waliuddin
(1996), have indicated that the silica of soil migrates in the plant in the shape of
monosilicic acid by evaporation, and under electron microscope studies, it showed the
dispersion of silica throughout the cellular structure of the husk. Study by Hwang and
Chandra (1997) shows that the un-burnt rice husks contain about 50% cellulose (C5H10O5),
25-30% lignin (C7H10O3) and 15-20% of silica (SiO2). Burning Of rice husk will remove
the cellulose and lignin, thus leaving behind silica ash.

Pozzolanic materials primarily consist of SiO2, Al2O3 and Fe2O3, the total of these
specified to be a minimum of 70% for mineral admixture Class F or N and 50% for
mineral admixture Class C for use in concrete (ASTM 618, 2003). When these are mixed
with Portland cement and water, the oxides which are in an amorphous structure will react
chemically at ordinary temperature with the calcium hydroxide (produced by the hydration
of calcium compounds in Portland cement) to form calcium silicate hydrate (C-S-H)
compound, thus possessing cementing properties. The reaction is secondary to the
hydration of the clinker compounds and occurs at a slower rate than the main chemical
reaction. Therefore, it is expected that as the cement replacement increases, the initial
strength of the cement - pozzolana material decreases, but at later ages it will increase
more rapidly than Portland cement as cementitious material has replaced the calcium
hydroxide (Cook et al., 1976).

ACI Committee 232 (2002) defines pozzolan as "a siliceous or siliceous and aluminous
material, which itself possesses little or no cementitious value but will, in finely divided
form and in the presence of moisture, chemically react with calcium hydroxide at ordinary
temperature to form compounds possessing cementitious properties."

35
With respect to the main C-S-H forming reaction, a comparison between Portland cement
and Portland pozzolan cement blended can be made and given as follows as described by
Mehta and Monteiro (1993) :-

Portland cement :- C3S + H fast C-S-H + CH ...................... Eq.2.1

Portland Pozzolan cement :- Pozzolan + CH + H slow C-S-H….…….. Eq. 2.2

Where, the symbols used are. C = CaO H = H2O S=SiO2

James and Subba (1986) quoted that the above reactions sequence will explain the setting
process that leads to either development of strength or its enhancement. They also stated
that the long terms decrease in strength of RHA-cement might be due to changes in
morphology or crystalline of C-S-H and unreacted silica.

RHA is a general term describing all types of ash produced from burning rice husk which
is a waste product of rice industry. It has been reported to have very high silica content as
high as that of silica fume after burning process at certain temperature (Kartini et al.,
2005). The silica ash produced varies from gray to black depending on the inorganic
impurities and unburned carbon amounts (Della et al., 2002). The silica in the ash
undergoes structural transformations depending on the temperature regime it undergoes
during combustion. At 550°C - 800°C, amorphous silica is formed and at greater
temperatures, crystalline silica is formed, the type of RHA suitable for pozzolanic activity
is amorphous rather than crystalline (Metha, 1994). The specific surface area of RHA can
be as high as 50 000m2/kg, even though the particle size is larger which is 10 to 75 µm.
This may be due to the particle complex shapes of RHA as reflecting their original plant
origins (Neville, 1997).

RHA has many applications due to the variation in its properties. It is an active pozzolan
and has several applications in cement and concrete industry. It is also highly absorbent,

36
and it is used to absorb oil on hard surfaces and potentially to filter arsernic from water
(Metha., 1994). It has been reported to be produced in Malaysia as almost 526,000 tonnes
per year of rice husk from 2.813 million tonnees of Malaysian paddy industry, and this
may contribute to a total of 102,000 tonnes of RHA per year (Kartini et al., 2005).

The utilization of RHA as cement replacement materials has been accepted as an

alternative to transform rice industrial residue into useful raw materials and at the same
time avoiding damage to the environment.

2.9.1 Production of Rice Husk Ash

2.9.1.1 Burning Temperature

The objective of burning rice husk is to remove by controlled oxidation the cellulose and
lignin present, while maintaining and preserving the original cellular structure of the rice
husk (Mehta, 1992). It has been established that the silica in the ash undergoes structural
transformation under varied temperature conditions, thus influencing both the pozzolanic
activity of the ash and its grinding ability.

When rice husks were first heated, weight loss due to evaporation of absorbed water
occurs at temperature up to 100°C and at 350°C. Further weight loss has occurred and the
husks than commence to burn. From 400°C to 500°C, the residual carbon oxidizes and the
silica in the ash is still in an amorphous form. Majority of weight loss occurred at this
period. As the temperature increased, the conversion to other crystalline forms of silica
progresses and further prolonged in incinerator at temperatures beyond 800°C produces
crystalline silica that is normally whitish in colour. It was suggested that the best burning
temperature for keeping the silica in amorphous state which is grey in colour and highly
cellular form was at 500°C to 600°C (Cook, 1986).

Mazlum and Uyan (1992) reported that burning the husk at 400°C and 500°C for one and
half (1½) hour produced amorphous silica. Cook et al. (1976) reported that the combustion

37
of rice husks at 450°C for four hours produces a material that conforms to the ASTM
C618-2003 Class N pozzolan.

Mehta (1992) and Hwang and Wu (1989) stated that in order to produce an ash of high
pozzolanic activity, the burning temperature must be such that the silica should be in a
non-crystalline state and in a highly micro-porous structure. They also concluded that the
best contribution of RHA on strength of concrete was obtained when the RHA was in
amorphous

form obtained by low burning of rice husks and with the holding time less than one (<1)
minute. Feasibility study conducted by Rego et al. (2004) on composing Portland cement
with the incorporation of amorphous and crystalline RHA revealed that amorphous RHA
yielded significant increased in compressive strength as compared to crystalline RHA.

Patel (1988) reported that the temperature of carbonization is preferably below 700°C to
avoid any transformation of amorphous to crystalline form. Cook et al. (1976, 1986),
Chopra et al. (1981) and Ibrahim and Helmy (1981) stated that when there has been over
burning of RHA, the transformation of crystallization scheme of the amorphous silica
(Si0 2) will change into cristobalite, quartz and tridymite as shown below:-

Quartz Cristobalite Tridymite

Amorphous (low temp. (temp. greater
crystalline-about (temp. beyond
than 800°C) 1000°C)
600°C)

However, when it is burnt at low temperatures much un-burnt carbon will remain and
normally the ash is blackish in colour. The influence of carbon content as investigated by
Cook and Sumanvitaya (1981) up to 20% by weight of the ash did not significantly
influence the strength development and beyond 30% the strength decreased.

38
Boateng and Skeete (1990) reported that the incineration of rice husks obtained from
Guyana at temperature ranges of 550°C to 700°C generally produced amorphous silica,
while a temperature in excess of 900°C produced unwanted crystalline form and for a
temperature of about 800°C maintained for 12 hours gave a small proportion of crystalline
silica.

Above all, the quality of RHA depends on the method of ash incineration, the degree of
grinding and on the preservation of cellular structure and the extent of amorphous
materials within the structure. Figure 2.1 shows the diffusion process for obtaining a
reactive cellular rice husk based on the data obtained from Ankra (1976), Mehta and Pitt
(1976) and Hwang and Chandra (1997).

From the figure, it shows that the optimum incineration condition is important to obtain
reactive RHA with microporous and cellular structure. In order to produce an ash with
high pozzolanic activity, the silica should be held in a non-crystalline state and in a highly
microporous structure, as only the active amorphous silica contributes to the development
of cementitious properties. Therefore, many researchers (Hwang and Chandra, 1997;
Mazlum and Uyan, 1992; Mehta, 1992; Sugita et al., 1992; Hwang and Wu, 1989; Patel,
1988; Chopra et al., 1981) have put their effort to produce amorphous silica by controlling
the burning temperature.

Figure 2.1: The optimum incineration condition curve for obtaining reactive cellular RHA
Adapted from Ankra (1976), Mehta and Pitt (1976) and Hwang and Chandra (1997)

39
2.9.2 Physical Properties of Rice Husk Ash

The physical properties of ash need to be determined, as it is an early indication of the

quality of ash to be used in the concrete. The physical properties commonly determined
for the ash are fineness, setting time and soundness. The physical requirements for
pozzolans can be obtained from ASTM C618-2003 and BS 3892: Part 1: 1997.

The physical properties of RHA are influenced by the condition of pyroprocessing. Cook
(1986) quoted that the compacted unit mass of the RHA ranges from 200 to 400 kg/m3 and
the bulk density ranges from 2000 to 2300 kg/m3. It is also known that the specific gravity
of RHA is much less than that of OPC; therefore weight-to-weight replacement would
therefore cause an increase in the volume of cementitious material. This would require
additional water for lubrication that would tend to lower the strength.

The physical characteristic of RHA can be altered depending on the production of RHA
technique. RHA has complex particle shapes in amorphous stage with a porous structure
that reflects its plants origins (Neville, 1997). In order to obtain a high specific surface
area the milling process was proposed (Della et al., 2002).

Generally, the particle mean size of RHA is around 33μm, but after the milling process,
the particle mean size of RHA may decrease to 0.681m-1m. The high surface area of RHA
prepares a bigger surface area for chemical reaction, during the hydration process of
cement. The bigger surface area may also give effects on increasing the water requirement
and water absorption of the cement paste (Singh et al., 2002).

2.9.2.1 Fineness

RHA of essentially pure silica in non-crystalline form are highly cellular (with fineness
about 600000 cm2/g using nitrogen absorption method) and are very fine particle with
higher specific surface as compared to the OPC (about 3020-3350 cm2/g).

40
The influence of fineness as reported by Shimizu and Jorillo (1990) and Chopra et al.
(1981) on the compressive strength showed that the increase in fineness resulted in the
increase of compressive strength. The relationship between specific surface, w/c ratio and
compressive strength based on Chopra et al. report is shown in Table 2.6. The findings
agreed with Lao et al. (1984) that stated that the strength of RHA concrete is higher than
the control mix with the same w/c ratio if the fineness of RHA is higher than 85% passing
No. 325 sieve. They further concluded that because of higher water requirements in
making RHA concrete, therefore, its strength is lower than the normal mix with similar
workability.

However, Mehta (1992) stated that the RHA particles should not be very fine in order to
develop pozzolanic activities in the presence of Portland cement. The reason was the
source of high surface area in RHA is in the microporous structure of individual particles
and with RHA particles in the range 10-75 μm, it exhibits satisfactory pozzolanic reaction.

Cook (1986) reported that-die reactivity of the ash is related to its surface area and the
amount of amorphous silica. Ash reactivity has to be balanced against the water demands,
as the high specific surface area of RHA will significantly increase the amount of water
required to produce a workable concrete.

Al-Khalaf and Yousif (1984) described the effect of burning and grinding parameters on
the properties of RHA concrete, in which rice husks when burnt at temperature ranged
between 450 0C to 8500C for periods ranging from 0.5 to 5 hours will produce Blaine
surface area of 5000 - 21000 cm2/g.

Table 2.6: Relationship between Specific Surface, Water/Binder Ratio and Compressive
Strength of OPC - RHA Blended Cement

Mixture Specific w Compressive Strength (N/mm2 )

designation surface c + RHA 3 days 7 days 28 days
(cm2/g)
Control 3780 0.42 24.7 30.7 43.9
LA - 1 5500 0.43 17.8 26.7 42.6
LA - 2 7000 0.45 25.5 37.9 51.7
LA - 3 8500 0.47 27.5 48.1 59.6
Adapted from Chopra et al. (1981)

41
2.9.2.2 Setting Time

Neville (1997) defined the term setting as ‘the stiffening of the cement paste from a fluid to
a rigid state and the terms initial set and final set are used to describe the arbitrarily
chosen stages of setting’. It is a transformation in the cement paste, in the mortar or
concrete, from fluid material to one that is solid and rigid. The setting process is
accompanied by the temperature changes in the cement paste. The initial setting is the start
of solidification which indicates the point at which the paste is no longer workable, while
the time required to fully solidify the mortar indicates the moment when setting ends.

The study on the rate of hydration of paste with RHA is important in predicting its
strength development. Study conducted by Hwang and Wu (1989), revealed that the heat
evolution curve of cement paste with RHA is similar in shape to that without RHA. They
also revealed that addition of RHA in cement performs as a kind of accelerator that
shortens the setting time through the absorption of the surrounding water, that is,
increasing the amount of RHA and w/b ratio in the mix resulted in shortening of setting
time. When comparing with the OPC control, the setting time for OPC-RHA paste is
shorter and as the percentage of replacement increases from 5% to 20%, the setting time
reduces. The reason is due to the water adsorption ability of the cellular form of RHA,
hence reducing the surrounding w/c ratio.

Neville (1997) quoted that the reaction between cement and water was exothermic and
because of its higher cement content, the plain cement paste would evolve greater amount
of heat.

The result of Ikpong (1993) agreed with the findings of Stroeven et al. (1999) and Cook et
al. (1976) who reported increased setting times of OPC-RHA pastes over those of plain
cement paste, well above the level indicated in ASTM C595-2003. The studies also
indicated that up to 20% cement replacement could be achieved without any significant
adverse effect on concrete properties.

42
Cook and Suwanvitaya (1981) also reported that the carbon content in the ash influences
the setting time as tabulated in Table 2.7. From the table, it can be seen that an increase in
the carbon content decreases both the initial and final setting time. They also reported that
the presence of up to 20% carbon (by weight) in the ash does not significantly influence
compressive strength. The strength reduction only becomes noticeable at carbon contents
of 30% and higher; this is mainly due to a corresponding reduction in silica content.

Table 2.7: Influence of Setting Time with Percentage of RHA in OPC Paste
Binder Mix (%) Ash carbon Setting times
References content w Initial Final
Portland RHA (%) c + RHA (min) (min)

Cook and 100 0 - - 40 80

Suwanvitaya 80 20 42 - 50 170
(1981 ) 80 20 14 - 120 190
60 40 42 - 35 50
60 40 14 - 70 110

100 0 - 240 600

95 5 - 180 540
90 10 - 0.35 150 480
85 15 - 120 390
Hwang 80 20 - 90 330
andWu
(1989) 100 0 - 180 480
95 5 - 180 450
90 10 - 0.47 150 480
85 15 - 120 390
80 20 - 30 330

95 5 - 0.33 214 259

90 10 - 0.35 222 282
Mahyuddin 85 15 - 0.45 273 314
(1993) 80 20 - 0.47 280 325
70 30 - 0.54 295 365
60 40 - 0.61 310 420

100 0 - - 125 243

Ikpong 70 30 - - 208 272
(1993) 60 40 - - 245 339
50 50 - - 281 392

Singh et al. 90 10 - 0.42 225 320

(2002)

43
Study by Mahyuddin (1993) as indicated in Table 2.9 revealed that the reactivity of RHA
cement depends upon the specific surface area, hence, RHA cement with finer particles
exhibit superior setting behaviour and compressive strength. He suggested that the
increase in percentages of ash in OPC would result in an increase of w/c ratio, thereby
increases the initial and final setting time of cement paste.

His finding is agreeable with Ikpong (1993) who suggested that higher water to cement
ratio tends to increase the setting time because of less contact between the open matrix and
the silica cellular structure, thus causes a reduction in early strength development. Study
conducted by Singh et al. (2002) on effect of replacing 5, 10, 15, 20, 25 and 30% by
weight of OPC with RHA on setting times has shown that at 10% replacement with RHA
and with w/b ratio of 0.42, the initial and final setting times are at maximum (that is 225
min and 320 min), and it is considered to be an optimum limit of RHA concentration.

2.9.2.3 Soundness

Soundness is the ability of a cement paste, mortar or concrete to withstand internal stresses
generated during cement hydration, without cracking (Neville, 1997). This expansion may
take place due to delay or slow hydration of some compounds present in the hardened
cement, namely free lime, magnesia, and calcium sulfate. Soundness of cement can be
determined from the test devised by Le Chatelier as prescribed by BS EN 196-3: 1995.
The values of expansion for RHA by various researchers are given in Table 2.1. The BS
EN 197-1:2000 states that the expansion of pozzolanic cement should not exceed 10 mm.

Literature on soundness with regards to RHA blended mortar can be hardly found. No test
was carried on the use of RHA blended cement. However, fly ash blended with OPC has
found to be advantageous in reducing expansion phenomena probably to the fact that the
concrete contains a lower amount of Portland cement than concrete without fly ash.

44
2.9.3 Chemical Composition of Rice Husk Ash

Chemical composition of RHA depends on the type of rice husk used and the combustion
process of the rice husk in producing RHA. Generally, RHA consists 80% to 90% of
silica. The ash produced by controlled burning of the rice husk between 550°C and 700°C
incinerating temperature for 1 hour transforms the silica content of the ash into amorphous
phase (Boateng, and Skeete, 1990). The rice husk ash produced by burning temperature
lower than 500°C was reported to contain unburnt carbon around 30% that adversely
affects the pozzolanic activity of RHA (Cook., 1986). The burning of RHA in air was
reported to lead the formation of silica ash but it depends on the inorganic impurities and
unburnt carbon amounts (Della et al., 2002). Della, 2002 in his work also reported that the
RHA with 95% of silica powder could be produced after the RHA calcined at 700°C for 6
hours.

The loss on ignition value for RHA may vary and it depends on the combustion process.
The resulting RHA is whitish gray in colour. Colour changes are associated with the
completeness of the combustion process as well as structural transformation of the silica in
the ash. Thus, whitish gray colour of the resulting ash is an indication of complete
oxidation of the carbon in the ash.

RHA has been reported to contain alkalise of high pH that may give an effect on the
expansion of cement (Neville, 1997). Singh et al., (2002) stated that the presence of high
pH alkalise in RHA may help breaking the glass structure to dissolve faster and enhances
its reactivity with calcium hydroxide. It also reacts with calcium hydroxide resulting into
complete elimination of Ca(OH).

The chemical composition of the rice husk varies depending on the type of paddy species,
and upon burning most of the evaporable components that is. cellulose and lignin will be
removed, leaving behind the silica ash (Mehta, 1992; Hwang and Wu, 1989). Therefore,
the characteristics of the ash obtained depend on the components of the rice husk,
temperature and time of burning. Table 2.10 shows the chemical compositions of RHA

45
produced due to different burning temperature as reported by Hwang and Wu (1989).
From the table, it shows that the higher the burning temperature, the greater the silica (Si)
content in the ash and the K, S, Ca, Mg as well as several other compounds are revealed to
be volatile, thus, affecting the quality of the ash produced.

Table 2.8: Chemical Composition of RHA (Taiwan) under Different Burning

Temperatures

Component Temperature (0C)

<300 400 600 700 1000

Si 81.90 80.43 81.25 86.71 92.7

K 9.58 11.86 11.80 7.56 2.57
Ca 4.08 3.19 2.75 2.62 1.97
Element Na 0.96 0.92 1.33 1.21 0.91
(%) Mg 1.25 1.20 0.88 0.57 0.66
S 1.81 1.32 1.30 1.34 0.16
Ti 0.00 0.00 0.00 0.00 0.45
Fe 0.43 1.81 0.68 0.00 0.68

SiO2 88.01 88.05 88.67 92.15 95.4

Oxide MgO 1.17 1.13 0.84 0.51 0.59
(%) SO3 1.12 0.83 0.81 0.79 0.09
CaO 2.56 2.02 1.73 1.60 1.16
K2O 5.26 6.48 6.41 3.94 1.28
Na2O 0.79 0.76 1.09 0.99 0.73
Fe2O3 0.29 0.74 0.46 0.00 0.43

Adapted from Hwang and Wu (1989)

Chemically, RHA is a heterogeneous mixture of crystalline and amorphous compounds of

SiO2, Al2O3, etc. with varying proportion of CaO and other minor components as shown in
Table 2.8. From the table, it is clearly shown that the SiO2 in the RHA presents as primary
constituent. The range of SiO2 content in RHA obtained from various researches is
between 80% and 97%. The wide difference in the percentage of SiO2, is due to the type
of paddy species, the temperature and time of burning as suggested by Hwang and Wu
(1989).

46
The reaction of the Ca(OH)2 with silica rich RHA in blended concrete will form extra C­ -
H and less portlandite which contributes to improvement in the strength of the concrete
and its resistance to acid attack, carbonation and penetration. A study by Qiyun et al.
(1999) shows that the presence of RHA as pozzolan accelerates the early hydration of C3S
and C3A due to depression of Ca2+ ionic concentration in the liquid phase. RHA will
stimulate the dissolution of C3S by the absorption of Ca2+ ion on the surface of RHA
particles and provide increasing site for the precipitation of C-S-H.

A study conducted by Yu et al. (1999) on the reaction between RHA and Ca(OH)2
solution at 40°C temperature found that after adding RHA to the Ca(OH)2 solution, the
concentration of Ca2+ and OH- were reduced with time due to the interaction between
Ca(OH)2 and RHA. This was due to the pH value that was directly related to the amount
of free OH- anions in the solution. It is also shown that the electrical conductivity was
mainly dependent on the effective concentration of Ca2+ and OH- ions in the solution.

Investigations had been conducted in many countries incorporating RHA in concrete as

cement replacement, and it is found that it could improve numerous properties in concrete.
This is due to the presence of SiO2 in RHA and Ca(OH)2 in cement paste, which react to
produce C-S-H that will contribute to the improvement in the strength and durability of
concrete.

2.10 Characteristics of Fresh Concrete with Rice Husk Ash

2.10.1 Effects on Fresh Concrete

The addition of RHA as partially cement replacement materials to OPC has been reported
to increase the water requirement of cement paste to become consistent. Since RHA take
some time in participating in the hydration process, they simply adsorb water because of
higher

47
surface area and hence, the water consistency of the composite becomes higher as
compared to that of OPC alone. The water required for standard consistency linearly
increases with an increase in RHA content. As the specific surface area of RHA is much
higher than cement, it needs more water (Singh et al, 2002). Cook (1986) also reported the
result on water demand increment of RHA concrete. The increment of water demand of
RHA may due to the carbon content that may also influence its setting times.

One of the effects of RHA on the properties of fresh concrete is reported to reduce the
workability of concrete. The used of RHA also required more superplasticizer and more
air-entraining admixture compared with the control OPC concrete in order to obtain the
same slump and air content (Malhotra et al., 2004). The air entraining agent for RHA
concrete mixtures was also found to be higher than that of SF concrete mixtures and this
can likely be due to its higher carbon content compared to SF (Nehdi et al.,2003)

The addition of RHA as cement replacement materials in cement paste was found to
increase the initial and final setting times. During this period RHA do not contribute in the
setting process and simply act as fillers and causes dilution effect, because of that , setting
times of the mixes are enhanced. Ganesan et al ; (2007) in his study found that up to 15%,
increasing the RHA level increases the initial setting time. At 20%, 25%, 30% and 35%,
there is a decrease in initial setting time

The initial setting time measured for RHA blended cements up to 35% is higher than that
of control OPC. On the other hand, the final setting time continues to decrease with the
increase in RHA up to 35%. The similar conclusion also made by Cook,(1986) that
noticed the carbon content in RHA below 20% (by weight) does not significantly
influenced the setting time but the increase of setting time was noticeable when the carbon
content in RHA is 30% and higher,

48
2.10.2 Effect on the Workability

Workability as defined in ASTM C 125 (2003) is "property determining the effort required
to manipulate a freshly mixed quantity of concrete with minimum loss of homogeneity".
Whereas, the ACI definition of workability, given in ACI 116R-90 (ACI, 1994) is "that
property of freshly mixed concrete or mortar which determines the ease and homogeneity
with which it can be mixed, placed, consolidated, and finished",

Concrete behaviour in terms of its consistency is normally dominated by the relative

proportions of aggregate, cement and water. In order to measure this behaviour, slump test
as in accordance to BS EN 12350-2: 2000 need to be carried out.

The characteristics of fresh concrete with RHA at early age depend on the w/b ratio, the
percentage proportion of RHA, the amount and fineness of cement, aggregate type and
grading admixtures and mixture proportion of concrete. As quoted by Mehta (1983), at a
given w/c ratio, small addition of RHA which is less than 2% to 3% by weight of cement
will improve the stability and workability of concrete by reducing the tendency towards
bleeding and segregation. The reason for this is due to the large surface area of RHA that
is in the range of 500000 cm2/g to 600000 cm2/g measured by nitrogen absorption (Mehta,
1992).

However, due to the adsorptive character of cellular RHA particles, concrete containing
RHA requires more water for a given consistency. Thus, large additions would produce
dry or unworkable mixtures (Hwang and Wu, 1989; Shafiq et al., 1988). Study conducted
by Shafiq et al. (1988) showed a decrease of 60% in the flow value when the percentage
content of RHA was 40% as compared to the control mortar. Beside the cellular structure,
the reason for this is that the RHA is much finer than that of ordinary Portland cement.
Mehta (1978) reported a reduction in slump from 95 to 12.5mrn when 30% of Portland
cement in a concrete mix was replaced with RHA.

49
A study conducted by Shimizu and Jorillo (1990) showed that increase in the water
requirement for a constant consistency of RHA concrete mix resulted in decrease in
strength. They also revealed that as the percentage of replacement of RHA increased, the
mixture possessed harsh characteristics. It was due to the overall decrease in the amount of
the binding medium, change in overall fineness of cement and change in the water
requirement. Sumadi and Abu (1989) showed that increase in w/b ratio resulted in better
compaction as a result of improved workability, thus this resulted in increase in strength of
the mortar. Table 2.9 showed some of the results obtained from their studies.

Figure 2.2: Workability of cement paste and mortar with RHA

Adapted from Hwang and Wu (1989)

50
Table 2.9: Water/binder ratio and compressive strength of RHA cement mortar cube
Compressive Strength at
Mixture w 28 days
Designation c + RHA (N/mm2)

OPC 10.89
RHA20 0.40 4.35
RHA35 1.18

OPC 14.44
RHA20 0.50 16.32
RHA35 15.03
OPC 10.17
RHA20 0.60 26.63
RHA35 29.24

OPC 13.05
RHA20 0.65 25.00
RHA35 26.40

OPC 13.50
RHA20 0.70 20.50
RHA35 23.00
Adapted from Sumadi and Abu (1989)

2.10.3 Effect of Superplasticizer on the Workability

Superplasticizing (Sp) admixture is a functional extension of normal plasticizing

admixture in that it disperses cement agglomerates efficiently. Sp is anionic surfactant
that is negatively charged. It will chemically adsorb onto the cement agglomerates, and the
resulting repulsion will cause dispersion of the cement flocks. This dispersion allows high
workability concrete to be obtained without increasing water content, thus giving flowing
concrete.

Neville (1997) quoted that Sp does not alter fundamentally the structure of the hydrated
cement paste, and the main effect being a better distribution of cement particles and
consequently better hydration. There are two main types of Sp, namely, the sulphonated
melamine-formaldehyde condense (SMF) and the sulphonated naphthalene-formaldehyde
condense (SNF).

51
Singh et at. (2002) stated that the effect of adding 1% of Sp on the consistency of 10%
OPC-RHA blended resulted in reduction in w/b ratio. Wada et at. (2000) also reported that
due to large specific surface area of the RHA, the Sp content of the mortar and concrete
mixtures increased with increasing RHA content to achieve a given consistency.

Shafiq et al. (1988) in their investigation revealed that by adding 2% of Sp in the 40%
RHA mortar mixes the flow increased by 40%, whereas with 30% RHA the flow increased
by about 50%. However, for the control mortar with the addition of Sp, the mortar became
fluid in which its flow value could not be determined. Brookes et al. (1979) revealed that
inclusion of Sp did not influence shrinkage, creep and modulus of elasticity of concrete.

2.11 Characteristics of Hardened Concrete and Cement Paste with Rice Husk Ash

The characteristics of hardened concrete with RHA emphasize more on the strength of the
concrete itself. The effect of RHA on the strength properties in terms of its compressive
strength, modulus of rupture, tensile splitting strength, static and dynamic modulus of
elasticity, Poisson's ratio, ultrasonic pulse velocity value, rebound number, drying
shrinkage, expansion and durability criteria will be discussed in various sub­ headings.

The effect of curing on the characteristic strength of hardened concrete will also be
discussed. The term curing is defined by Garnbhir (2006) is "a process of creation of an
environment (that is. the desirable condition such as suitable temperature, provision of
ample moisture or prevention of loss of moisture, and the avoidance of premature stressing
or disturbance) during a relatively short period immediately after placing and compaction
of the concrete, favourable to the setting and hardening of concrete". Curing is essential in
obtaining satisfactory long-term durability performance, particularly in hot and dry
climates (McCarter and Ben-Saleh, 2001).

A study by Patel et al. (1988) showed that hydration is greatly reduced when the relative
humidity within the capillary pores drops below 80% and suggested that hydration at a
maximum rate can proceed only under condition of saturation. The prevention of loss of

52
water from the concrete is of importance not only because the loss adversely affects the
development of strength, but also because it leads to plastic shrinkage, increased
permeability and reduced resistance to abrasion.

2.11.1 Compressive Strength

A study conducted by Cook et al. (1976) in terms of the compressive strength of cement
paste with RHA revealed that increase in the replacement level of more than 20% resulted
in decreasing compressive strength at 28 days as compared to 100% of OPC cement. In
their study, it was also highlighted that the increase in strength was greater for all the
replacement level tested (10%, 20%, 30% and 40%) from 28 to 90 days compared to the
100% OPC concrete mix. This is expected because the pozzolanic activity which is related
to the ability of silica to combine with Ca(OH)2 to produce cementitious C-S-H occurred
at a slower rate. As the cement replacement increased, the initial strength of the cement
replacement material decreased.

Table 2.10: Compressive Strength of Cement-RHA Blended Concrete

Compressive
Mix Description w Strength
Cement + RHA : Sand: Gravel c + RHA (N/mm2)
(1 : 1.68 : 2.53) 28 days 90 days

C100:0-1 0.45 35.85 38.72

C100:0-2 0.55 35.16 37.80
C100:0-3 0.65 26.89 29.17

C90:10-1 0.49 40.67 45.55

C90:10-2 0.55 33.09 36.73
C90:10-3 0.65 27.58 31.03

C80:20-1 0.53 33.78 37.00

C80:20-2 0.62 33.44 36.28
C80:20-3 0.69 26.20 29.89

C70:30-1 0.75 18.61 21.03

C70:30-2 0.93 14.48 16.22

Adapted from Cook et al. (1976)

53
Table 2.12 shows the result of their study on the compressive strength of blended cement
RHA concrete. In 1981, Cook and Suwanvitaya reported that for OPC replacement of
more than 40% with RHA, the compressive strength is significantly reduced at 365 days.
They also concluded that as w/(c + RHA) ratio increased, the strength decreased, and this
finding is in accordance with the conventional w/c ratio concept of strength.

Sugita et al. (1997) also reported that the addition of RHA resulted in increased in
compressive strength and at the RHA blending ratio of 20%, the concrete attained the
maximum increment in strength. They also suggested that the 28 days compressive
strength of concrete with different w/c ratios increased with the addition of RHA. Their
results are tabulated in Table 2.11.

Their reasons for these were included that; the ground RHA is a porous material and with
large number of mesopores, resulting in RHA having high water absorption characteristic.
Therefore , the practical w/c ratio in RHA blended mortar is less than the used because a

portion of free water has been absorbed in the mesopores in RHA. Besides, the RHA used
in their study have high content of Si02 ,that is about 91.9% , and is very active. Thus, it
can quickly react with the Ca2+ and OR ions in the liquid phase of hydrating cement.
Therefore, in comparison with the normal concrete, there will be more C-S-H and less
portlandite in the RHA blended, thus these contributed to an increase in the strength of
concrete.

Giaccio et al. (2007) also reported that the incorporation of 10% RHA at lower w/b ratio
(0.4 - 0.5) resulted in higher compressive strength compared to control OPC concrete, with
relative compressive strength of RHA concrete increase to near 9% and 7% at 28 days and
90 days respectively. Study by Mehta and Pritz (1978), showed that in a high strength
concrete mixture containing 30% of RHA , by weight of total cementing material, gives
about 8% higher strength at the age of 28 days compared with those of control Portland
cement concrete. Moreover, the adiabatic temperature rise is lower for the RHA concrete.

54
Table 2.11: Compressive Strength of Cement-RHA Concrete Cured in Water for 28 days

Cement RHA w Compressive Strength, N/mm2

(kg/m3) 3
(kg/m ) (%) c + RHA (interpolate from graph)

283.8 0 0 22.0
255.4 28.4 10 0.65 23.0
198.6 85.2 30 27.0

245.9 0 0 16.0
221.3 24.6 10 22.5
196.7 49.2 20 0.75 24.5
172.2 73.8 30 21.0
123.0 123.0 50 16.5

233.1 0 0 13.0
209.8 23.3 10 0.80 19.5
186.5 46.6 20 22.0
163.2 69.9 30 21.5
116.6 116.6 50 17.0

Adapted from Sugita et al. (1997)

Adding the correct amount of RHA is important to achieve high strength. A study
conducted by Hwang and Wu (1989) showed that large amount of RHA have an adverse
effect and reduce strength. Highly active RHA are effective in amounts as low as 10% and
tend to increase rather than decrease the strength at early ages. In fact, the early strength of
concrete is a function of w/b ratio. The early strength of concrete will be similar as long as
the w/b ratio is kept constant, but the. ultimate strength will be enhanced due to pozzolanic
reaction. However, Hwang and Chandra (1997) showed that at high w/b ratio, the addition
of RHA to cement paste will not only reveal a significant effect on strength at early ages,
but the strength at later ages also tend to be higher than those with lower w/b ratios.
Shimizu and Jorillo (1990) in their study reported that as the percentage of RHA increases,
the amount of water needed to maintain a constant consistency of 100 mm slump increases
and this undoubtedly resulted in decrease in strength. The results of their study are best
translated as in Table 2.12 .

55
Table 2.12: Compressive Strength at 28 Days of Cement-RHA Blended Concrete

Compressive
Mix Code Cement RHA w Strength at 28 days
(kg) (kg) c + RHA
kg/cm2 N/m2
MG - 0 (Control) 313 - 0.58 283 27.8
MG - 80:20 250 39.49 0.68 214 21.0
MG -70:30 219 59.23 0.83 156 15.3
MG -60:40 187 78.97 0.93 116 11.4

HG - 0 (Control) 356 - 0.50 352 34.5

HG - 80:20 285 44.98 0.59 259 25.4
HG -70:30 249 64.45 0.74 169 16.6
HG -60:40 214 89.96 0.81 144 14.1
HG -50:50 178 112.45 0.87 106 10.4

Note: HG represents high grade

MG represents medium grade
Adapted from Shimizu and Jorillo (1990)

A research conducted by Coutinho (2003) on the compressive strength of RHA concrete

on 10%, 15% and 20% replacement of cement with RHA taken at 80 days showed
strengths of 41.5, 41.7 and 43.0 N/mm2 respectively as compared to control mix which
showed a strength of 35.1 N/mm2. All the mixes have a constant percentage of Sp and w/b
ratios of 1.2% and 0.43% respectively. This showed that the replacement of cement with
RHA has an influence on its strength and the higher the replacement; the higher will be the
strength at later ages (that is. based up to 20% replacement). Zhang et al. (1996a) reported
that the higher compressive strength of RHA concrete is due to the reduced porosity,
reduced Ca(OH)2, and reduced width of the interfacial zone between the paste and the
aggregate.

Chindaprasirt et al. (2005) in their report on the compressive strength of concrete with
40% and 60% replacement of RHA from total cementitious content of 333 kg/m2 plus Sp
showed that at higher replacement level, lower compressive strength resulted. Both RHA
concrete exhibited lower strength compared to the OPC concrete. Their reason was owing
to the low amount of Portland cement and hence, low calcium hydroxide such that both

56
the initial and the subsequence reactions were lower. The SiO2 of OPC and RHA used in
their study were 20.90% and 88.33% respectively.

A study conducted by Ikpong (1993), on various grades of concrete from 15 N/mm2 to 40

N/mm2 with varying levels of cement substitution by volume, taken up to age of 90 days,
reveals that each of the compressive strength increased with age but decreased with RHA
content. Proportion by volume was adapted because of the remarkable difference in
specific gravities of the RHA and cement (2.12 for RHA and 3.15 for cement). The SiO2
of RHA used in his study was 82.12% with L.O.I of 8.60. The variation of compressive
strength with age and levels of replacement are presented in Table 2.13.

Table 2.13: Variation with Age of Compressive Strength of RHA Concrete

Design % Cement RHA Water Compressive Strength

Strength RHA Content (kg) (kg)
(N/mm2) content (kg) 7 days 14 days 28 days 90 days

0 13.1 17.8 25.8 28.3

15 30 N/A N/A N/A 12.1 16.0 25.0 27.6
40 8.9 11.8 19.9 24.7
50 5.7 10.3 13.7 17.8

0 270 - 185 17.5 23.9 33.1 37.1

20 30 189 81 197 15.8 18.2 29.1 35.2
40 162 108 204 13.4 15.2 27.2 33.9
50 135 135 211 10.6 12.7 23.0 28.7

0 321 - 192 28.4 37.6 43.5 46.3

30 30 224.7 96.3 196 22.6 32.0 38.3 43.2
40 192.6 128.4 203 16.7 27.5 31.8 37.1
50 160.5 160.5 216 9.1 16.2 19.6 22.1

0 32.4 38.8 44.2 47.6

40 30 N/A N/A N/A 24.0 33.5 39.4 44.9
40 18.5 29.3 34.6 38.9
50 10.0 18.1 22.5 22.9

Adapted from Ikpong (1993)

57
Mazlum and Uyan (1992) in their study showed that the compressive strength of mortar
with cement replacement of 10%, 20% and 30% with RHA was greater than that of the
control mortar until the 56 days and remains constant beyond that time. As shown in,
Table 2.14, the compressive strength increased remarkably in the course of time
proportional to the increase in the amount of RHA, with 20% replacement showed the
most effective RHA percentage replacement.

However, Mahyuddin (1993) in his study reported that the strength of cement mortar
blended with RHA replacement up to 25% gave good compressive strength; but beyond
that level of replacement, the strength of cement mortar blended with RHA is lower
compared to mortar without RHA. He also stated that the effectiveness of the grinding and
the temperature of burning have significant effect on the surface reactivity.
Table 2.14: Compressive Strength of Mortars

Compressive Strength, N/mm2

Cement RHA w Sp
(kg) (kg) c + RHA (%) 28 days 56 days 84 days 112 days

450 0 34.3 35.5 41.5 38.5

405 45 36.7 37.4 42.9 42.0
360 90 0.55 2 36.8 39.7 42.4 42.9
315 135 34.2 38.4 41.3 40.9

Adapted from Mazlum and Uyan (1992)

Wada et al. (2000) in their research on the RHA concrete subjected to different curing
conditions that is moist-cured at 20°C and 85% relative humidity for 1 day and water-
cured at 20°C until the age of 28 days showed that irrespective of the curing condition, the
compressive strength of the RHA concrete is higher than that of the control concrete. The
reason was probably due to the higher pozzolanic reactivity of RHA concrete. Saricimen
et al. (2000) in their study have shown that exposure to water curing of plain and
pozzolanic cement concrete resulted in an increase in the compressive strength.

58
RHA was also known as a pozzolanic material that can improve the engineering properties
of concrete especially at harden stated of concrete. The high silica content at amorphous
stage of the RHA was reported to be the contributory factor of the improvement. The
effectiveness of using RHA as cement replacement materials or as additive in the concrete
mixes depends on factors named as type of rice husk used, incineration degree of RHA,
fineness of RHA and replacement level of RHA to OPC.

The increase in compressive strength of concrete with RHA produced by controlled

incineration is mainly due to the pozzolanic effect. Gemma, (2006), study on the effect of
incineration degree of RHA to the concrete engineering properties concluded that, RHA
that produced by open burning provides a positive effect on the compressive strength and
had a comparable effect as the controlled incineration RHA at early ages, but in the long
term, the behaviours of the concretes with RHA produced by controlled incineration was
more significant.

The rate of hydration in concrete made with part replacement of cement by RHA is slow
as compared to concrete with OPC only. This fact is very dominant during the initial three
days of age of concrete. This rate of slow hydration also affects the 150 day strength of
concrete made by part replacement of cement by RHA in the mix (Muhammad et al,1996).

The RHA concrete was found had higher compressive strength at 91 days in comparison
with that of the concrete without RHA, although at 7 and 28 days a different behaviour
was observed between the control concretes with the RHA concrete. The increase in
compressive strength of concretes with residual RHA is better justified by the filler effect
(physical) at early age than by the pozzolanic effect (chemical/physical) at late age (Bui et
al, 2005).

The results indicate that optimum replacement of cement by RHA will be around 10% to
20% with finely ground RHA. It simply indicates that in the presence of RHA, some
alkalise are liberated in the solution that further breaks the glassy network and enhances its
reactivity with calcium hydroxide. RHA also reacts with calcium hydroxide resulting into
complete elimination of Ca(OH) (Singh et al, 2002).

59
The strength development in cement is mainly due to the formation of calcium silicate
hydrate. The increase in strength in the presence of RHA is due to the formation of
additional amount of calcium silicate hydrate as a result of pozzolanic reaction (Singh et
al, 2002).

The cement paste incorporating RHA did not show any increase in the compressive
strength compared with that of the control portland cement paste. The higher compressive
strength of the RHA concrete compared with that of the control concrete is due probably
to its reduced

porosity, reduced Ca(OH), content, and the reduced width of the interfacial zone between
the paste and the aggregate (Zhang et al, 1996).

Zhang et al (1996) as illustrated in Figure 2.3 also stated that the use of RHA as cement
replacement materials had comparable effects to SF at 10% replacement level on the
compressive strength at 90 and 180 days but lower than SF before 28 days.

The reduction of concrete durability due to the reaction of sulphate is to leach the Ca+ ion
from the concrete matrix and replace with Mg ++ ion, increased the volume of concrete
with sulphate crystal and increased the internal strain of concrete matrix that will generate
more micro crack (Neville, 1997). Magnesium sulphate chemically reacts with calcium
hydroxide in the presence of water to gypsum and magnesium hydroxide. The latter salt is
relatively insoluble and poorly alkaline (pH < 10.4). At this low pH, the stability of
calcium silicate hydrate (C-S-H) is reduced. The decomposition of C-S-H may proceed to
completion and all the calcium content of the binder may be replaced by magnesium,
thereby leading to disintegration of concrete.

2.11.2 Modulus of Rupture

Flexural strength is also expressed in terms of the modulus of rupture (MOR). In 1990,
Shimizu and Jorillo reported that the modulus of rupture for RHA concrete is lower

60
compared to OPC concrete. The result of their study on the modulus of rupture of cement-
RHA blended concrete is presented in Table 2.15.

Table 2.17: Modulus of Rupture of Cement-RHA Blended Concrete

Modulus of
Mix Code Cement RHA w Rupture at 28 days
Cement : RHA (kg) (kg) c + RHA
kg/cm2 N/m2

MG - 0 (Control) 313 - 0.58 28.5 2.79

MG - 80:20 250 39.49 0.68 22.1 2.17
MG -70:30 219 59.23 0.83 23.3 2.29
MG -60:40 187 78.97 0.93 15.5 1.52

HG - 0 (Control) 356 - 0.50 35.1 3.44

HG - 80:20 285 44.98 0.59 27.4 2.69
HG -70:30 249 64.45 0.74 25.9 2.54
HG -60:40 214 89.96 0.81 21.9 2.15
HG -50:50 178 112.45 0.87 19.4 1.90

Note: HG represents high grade MG represents medium grade

Adapted from Shimizu and Jorillo (1990)

The ash was obtained from the open burning in the field of variety of rice husk in the
Philippines. The ash used in Shimizu and Jorillo's study was at its natural fineness with
47.67% and 25.29% passing No. 100 and No. 200 sieve respectively. From the table, it can
be seen that irrespective of the grade, the MOR for RHA concrete is smaller compared to
the control concrete, and it reduces as the amount of replacement with RHA increases.

Not many work were obtained on the MOR of cement-RHA blended concrete. However,
there are few research work carried out on MOR of cement-RHA blended mortar.
Investigation conducted by Mazlum and Uyan (1992) showed that the modulus of rupture
of mortar with RHA has generally increased with the increase in the amount of RHA and
that mortar with 30% RHA has the greatest modulus of rupture. This can be seen in Table
2.18.

61
Table 2.16: Flexural Strength of Cement-RHA Blended Mortar

Flexural Strength, N/mm2

Cement RHA w Sp
(kg) (kg) c + RHA (%) 28 days 56 days 84 days 112 days
450 0 5.6 6.2 6.4 3.6
405 45 6.4 7.4 7.4 7.5
360 90 0.55 2 6.4 7.4 7.9 7.8
315 135 7.0 8.0 8.4 7.9
Note * Relative values calculated in relation to control mortar at 28 days
Adapted from Mazlum and Uyan (1992)

A study conducted by Kilinckale (1997) on mortar containing the 20% RHA showed that
the 28 days and 56 days modulus of rupture were higher than the control mortar. However,
a study conducted by Mahyuddin (1993) on basic mix composition of mortar of 1:2.5 (1
part of cement to 2.5 part of sand by weight) with 0.5 w/c ratio and replacing cement with
RHA by 5, 10, 15, 20, and 30% has indicated that the modulus of rupture of these cement-
RHA blended mortar have lower strength than the control (that is, 5 N/mm2) and 30%
RHA replacement showed the lowest strength achieved.

The values of the flexural strength for these cement-RHA blended mortars were 4.8, 4.0,
3.7, 3.5 and 2.8 N/mm2 respectively. It was reported that the reasons for the discrepancy
were due to the difference in fineness of RHA used and different burning temperature
adapted in producing the ashes.

2.11.3 Tensile Splitting Strength

The common tests used to estimate the tensile strength of concrete is the splitting tension
test. Compared to direct tension, the splitting tensile strength is known to overestimate the
tensile strength of concrete by 10 - 15% as reported by Mehta and Monteiro (1993). Table
2.17 shows the results obtained from the study conducted by Shimizu and Jorillo (1990)
on the

62
RHA concrete subjected to tensile splitting test. From the table, it can be derived that
irrespective of the grade, the tensile splitting strengths of cement-RHA blended were
lower than the control concrete. It also shows that as the amount of replacement with RHA
increases, the tensile splitting strength decreases, and agrees with Cook's finding.

Table 2.17: Tensile Splitting of Cement-RHA Blended Concrete

Modulus of
Mix Code Cement RHA w Rupture at 28 days
Cement : RHA (kg) (kg) c + RHA
kg/cm2 N/m2

MG - 0 (Control) 313 - 0.58 30.1 2.95

MG - 80:20 250 39.49 0.68 23.8 2.33
MG -70:30 219 59.23 0.83 19.2 1.88
MG -60:40 187 78.97 0.93 16.8 1.65

HG - 0 (Control) 356 - 0.50 36.5 3.61

HG - 80:20 285 44.98 0.59 29.7 2.91
HG -70:30 249 64.45 0.74 22.6 2.22
HG -60:40 214 89.96 0.81 20.1 1.97
HG -50:50 178 112.45 0.87 15.5 1.52

Note: HG represents high grade

MG represents medium grade
Adapted from Shimizu and Jorillo (1990)

Cook et al. (1976) in their research on cement replacement with RHA of between 10% and
40% showed that increase in replacement level results in decrease in the splitting tension
when tested at 28 days. For the 30% and 40% replacement level, the 28 days tensile
splitting strength was considerably lower than the plain cement concrete. The result of
their work is presented in Table 2.18. Karasudhi and Nimityongskul (1979) also reveal
that addition of RHA with decreasing amount of cement in the mix decreases the concrete
splitting tensile strength.

63
Table 2.18: Tensile Splitting Strength of Cement-RHA Blended Concrete

Mix Description Splitting Strength

Cement + RHA : Sand: Gravel w (N/mm2 )
(1 : 1.68 : 2.53) c + RHA 28 days

C100:0-1 0.45 2.93

C100:0-2 0.55 2.75
C100:0-3 0.65 2.17

C90:10-1 0.49 3.45

C90:10-2 0.55 3.15
C90:10-3 0.65 2.69

C80:20-1 0.53 2.69

C80:20-2 0.62 2.79
C80:20-3 0.69 2.28

C70:30-1 0.75 1.74

C70:30-2 0.93 1.52

C60:40-1 0.85 1.17

C60:40-2 1.09 0.96

Adapted from Cook et al. (1976)

2.11.4 Non-Destructive Test on Rice Husk Ash Concrete

The term "non-destructive testing" is taken in its broad sense to include methods that cause
localized surface damage, but are non-destructive in relation to the body of the concrete
under examination. Non-destructive testing may be applied to both new and existing
structures. For new structures, the principal applications are for quality control, monitoring
of strength development or resolution of doubts about the quality of materials or
construction. For existing structures, normally it will be related to an assessment of
structural integrity or adequacy, and material deterioration.

64
2.11.4.1 Ultrasonic Pulse Velocity

The pulse velocity method has been used successfully both in laboratory as well as field to
evaluate the quality of concrete. It involves the use of sonic waves to detect internal
structure changes in concrete, to obtain the dynamic modulus of elasticity, Poisson's ratio
and to estimate the strength of concrete test specimens resulting in no damage to the
concrete element being tested. Neville (1997) defined the pulse velocity through concrete
as the outcome of the time taken by the pulse to travel through the hardened cement paste
and through aggregate.

Naik and Malhotra (1991) reported that there are many factors that affect the pulse
velocity; it can either be the factors affecting the concrete properties or factors affecting
the pulse velocity measurement regardless of the properties of concrete. The pulse velocity
for ordinary concrete is typically in the order of 3660 m/sec, and it is not dependent upon
the size and the shape of a specimen.

Sturrup et al. (1984) revealed that there is no unique relation between ultrasonic pulse
velocity and compressive strength exist. However, the rate of hydration is different for
different cements and as the degree of hydration increase the strength will increase and the
pulse velocity will also increase (Facaoaru, 1970).

Neville (1997) concluded that for a given aggregate and a given richness of the mix, the
ultrasonic pulse velocity of the concrete is affected by changes in the hardened cement
paste, such as a change of w/c ratio, that is as the w/c ratio increases, the density,
compressive and the modulus of rupture and the corresponding pulse velocity decreases.
Naik and Malhotra (1991) and Bungey (1989) stated that the effect of curing conditions on
the pulse velocity through saturated concrete was higher (can be up to 5%) than the same
concrete in a dry condition.

A study conducted by Ikpong (1993) revealed that increased in the gel-space ratio of
concrete at advanced age was markedly different from that of the same concrete at an early

65
age. The former consists of a compact network of aggregates strongly held together by
cement gel that accounted for the increased strength of such concrete, whilst the latter
consists of a near discrete pack of aggregates, stiffened cement paste, capillary water and a
relatively small proportion of cement gel. His research on RHA concrete showed that the
initial gel-space ratio is invariably much less than that of a corresponding conventional
concrete.

However, the gel-space ratio at advanced age was very close to that of the early age. The
results of his study on the compressive strength and the pulse velocity are shown in Table
2.19.

Table 2.19: Variation with Age of Compressive Strength and Pulse Velocity of Concrete
Designed Strength of 30 N/mm2
Compressive Strength, fc
% Cement RHA Water (N/mm2)
RHA Content (kg) (kg)
content (kg) 7 days 28 days 90 days

0 321 - 192 28.4 43.5 46.3

30 224.7 96.3 196 22.6 38.3 43.2
40 192.6 128.4 203 16.7 31.8 37.1
50 160.5 160.5 216 9.1 19.6 22.1

Pulse Velocity, V
% Cement RHA Water (km/sec)
RHA Content (kg) (kg)
content (kg) 7 days 28 days 90 days

0 321 - 192 4.38 4.70 4.74

30 224.7 96.3 196 4.23 4.63 4.68
40 192.6 128.4 203 4.00 4.44 4.60
50 160.5 160.5 216 3.15 3.61 3.81

Note:- For the mixes containing RHA, the combination of cement and RHA was proportioned by
volume, because of the remarkable difference in the specific gravity of the RHA and cement, that
is. 2.12 and 3.15 respectively.
Adapted from Ikpong (1993)

66
From the table, it can be seen that the compressive strength and pulse velocity of RHA
concrete decreased with increased in RHA content. These were due to the slower
hydration of cement and the secondary pozzolanic reaction involving the silica and
alumina present in the ash. In addition, increase in RHA required increase in water content
for equal workability (Singh et aI., 2002; Wada et aI., 2000; Shimizu and Jorillo, 1990;
Hwang and Wu, 1989). Thus, this has resulted in more capillaries being left in RHA
concrete as hydration progresses. Therefore, this increases the time of travel of the pulse
and hence, decreases the calculated velocity.

The above statement agreed with Neville (1997) who stated that the speed of pulse
propagation is much less through voids, than through solid matter and in fact, the pulse
cannot travel directly through the voids but is deflected round it so that the pulse can be
transmitted through the solid concrete matter. He also stated that pulse travels faster
through a water-filled void than through an air-filled. Table 2.20 shows the pulse velocity
classification established for concrete as suggested by Neville (1997).

Table 2.20: Classification of the Quality of Concrete on the Basis of Pulse Velocity

Pulse velocity, V (km/sec) Quality of concrete

≥ 4.5 Excellent
3.5 - 4.5 Good
3.0 - 3.5 Doubtful
2.0 -3.0 Poor
< 2.0 Very poor
Adapted from Neville (1997)

2.11.4.2 Rebound Hammer

The rebound hammer test or also known as the Schmidt rebound hammer test is based on
the principle that the rebound of an elastic mass depends on the hardness of the surface
against which the mass impinges. It is useful in the assessment of uniformity of concrete
within a structure, and can also be used to establish whether the rebound number has

67
reached a value known to correspond to the desired strength (Neville, 1997; Malhotra,
1991). The rebound number is an arbitrary measure because it depends on the energy
stored in the given spring and on the size of the mass. The rebound number will be
governed by the energy absorbed by the concrete which is both related to its strength and
stiffness. However, the results are very difficult to predict because it is affected by the
smoothness of the test surface, size, shape and rigidity of the specimens, the age of the
specimen, surface and internal moisture conditions of the concrete, type of coarse
aggregate and cement and type of mould and carbonation of the concrete surface, to name
a few.

There was a wide degree of disagreement on the correlation between compressive strength
of concrete and the hammer rebound number among various researchers. However,
Malhotra (1991) reported that by consensus, the accuracy of the estimation of compressive
strength of test specimens cast, cured and tested under laboratory conditions by a properly
calibrated hammer lies between ± 15% and ± 20% and probable accuracy of estimation of
concrete strength in a structure is ± 25%.

Extensive researches have been conducted on the normal strength concrete incorporating
OPC in terms of its rebound numbers. However, no research has been conducted and
reported for concrete incorporating MIRHA.

2.12 Durability of Concrete with Rice Husk Ash

Durability as defined by ACI Committee 201(1991) is its ability to resist weathering

action, chemical attack, abrasion, or any other process of deterioration; that is, durable
concrete will retain its original form, quality and serviceability when exposed to its
environment.

Water is the primary agent of most durability problems in concrete. In porous solids, water
is known to be the cause of many types of physical and chemical processes of degradation.
The physical-chemical phenomena associated with water movements in porous solid are

68
controlled by the permeability of the solid (Mehta and Monteiro, 1993). The rate of
chemical deterioration of concrete would depend on whether the chemical attack is
confined to the surface of concrete or inside the material itself, or by the type of
concentration of ions in water and by the chemical composition of the solid. Mehta (1977)
in his report stated that OPC contains approximately 60% to 65% of CaO, about 25% is
released upon hydration as free Ca(OH)2 and it is this product which makes Portland
cement prone to deterioration in acidic environment.

The physical effects of the chemical reaction between Portland-RHA cement will produce
pore-size refinement and grain-size refinement (Hwang and Chandra, 1997; Mehta, 1992).
The pore-size refinement is referred to the process of transformation of the secondary
hydration products. These are mainly C-S-H around the pozzolan particles which contain
large capillary voids into a microporous product containing numerous fine pores.
Meanwhile, grain-size refinement is referred to as the process of transformation of a
system containing large grains of a component into a product containing smaller grains.
Both the pore-size refinement and grain-size refinement processes strengthen the transition
zone, thus reducing the microcracking and increasing impermeability of the concrete.

Speare et al. (1999) also studied the durability of mortars containing RHA. They noted an
increase in compressive strength, an improvement in absorption characteristics and a
reduction in oxygen permeability for RHA mortars.

Cook (1986) suggested that the improved overall durability of Portland cement-RHA is
related to the physical structure of the hydrated cements as well as the reduction in the
amount of Ca(OH)2 with regards to structure. A study conducted by Zhang et al. (1996a)
on Ca(OH)2 content as shown in Figure 2.3 shows that RHA paste has a lower Ca(OH)2
than the control cement paste. These lower Ca(OH)2 and lower unhydrated cement content
for RHA clearly indicate the pozzolanic reactivity of the RHA.

69
 x-axis Age y-axis Control y-axis RHA
1 11 9
10 14 8.8
55 13 8
100 15 7
Adapted from Zhang et al. (l996a)

Figure 2.3: Calcium Hydroxide Content in OPC and RHA paste

Zhang et al. (1996a) through their research demonstrated that incorporation of RHA in
cement paste and concrete will reduce its porosity. The Ca(OH)2 amount, as a result of
hydration in the interfacial zone and the width of the interfacial zone between the
aggregate and the paste, also reduced, compared with that of the control Portland cement
composite.

Their study on the porosity of the paste as a function of the distance from the aggregate
surface on test specimens cured for 28 days is shown in Table 2.21. From the table, it
indicates that the incorporation of RHA reduced the porosity in the interfacial zone
compared with that of the control Portland cement composite.

70
Table 2.21: Porosity of OPC and RHA paste as a function of the distance from aggregate
surface on the test specimen

Distance from the aggregate surface (micron)

Paste 20 33 43
Porosity (%)
OPC 25 17 11
RHA 18 9 5
Adapted from Zhang et al (l996a)

Sugita et al. (1997) and Gambhir (2006), in their study, found out that the amount of
Ca(OH)2 in the hardened cement paste and concrete added with RHA, was greatly reduced
compared with the paste and concrete without RHA addition. This is due to the paste or
concrete blended with RHA that reacts quickly with the Ca2+ and OH- ions released from
the hydration of cement and as a result, more C-S-H gel formed. With higher formation of
C­ S-H gel and lower content of portlandite in the RHA blended concrete, resulted in
superior to normal concrete in terms of resistance to acid attack, carbonation and
penetration of chloride ion. However, the above results could be achieved provided proper
control of temperature to produce the amorphous silica in the ash is made.

Many studies have shown that RHA improves the durability of concrete by modifying the
micro and macro structure of cement paste (Speare et al. (1999); Sugita et al. (1997);
Zhang et al. (1996a); Hwang and Chandra (1997); Gambhir (2006); Mehta (1992);
Manmohan and Mehta, (1981)).

Hwang and Chandra (1997) reported that addition of pozzolanic materials could affect
both the strength and permeability by strengthening the aggregate-cement paste interface
and by blocking the large voids in the hydrated cement paste through pozzolanic reaction.
Pozzolanic reaction occupies the empty space in the pore structure, thus making the paste
or concrete dense. They also reported that above 54 kg/m3 RHA addition, there is no
influence on the strength; however, it decreases the permeability of concrete. Mehta
(1992) in his study

71
on cement paste with RHA additions shows that there is a significant reduction in the
porosity of the cement paste and also refinement in the pore structure.

The fine and cellular RHA particles will have direct beneficial effect in reducing bleeding
and segregation characteristic of concrete mixtures due to its large water absorption
ability, high internal surface area as well as microporous and amorphous particles.

Reduction in bleeding on the other hand would result in a stronger transition zone between
solid matters and cement paste and thus, leads to a more impermeable and durable
concrete.

2.12.1 Permeability

Permeability as defined by Mehta and Monteiro (1993) is the property that governs the
rate of flow of a fluid into a porous solid. Permeability that is dictated by the
microstructure of concrete controls the ingress of moisture or gases into the concrete, thus
revents chemical degradation. Chemical degradation as a result of reaction between an
external agent and the ingredients of concrete decreases as the permeability of concrete
reduces. Therefore, it is generally observed that the less permeable the concrete, the
greater will be its resistance to aggressive solutions or pure water.

In concrete technology, permeability normally refers to a pressure induced flow, as in D'

Arcy's permeability. In the determination of concrete durability in aggressive
environments, permeability of concrete is one of the important properties. The study of
permeability of concrete is important because the penetration by materials in solution may
adversely affect the durability of concrete. As the concrete permeability is lowered, the
resistance to the penetration by these aggressive agents increases. Therefore, the more
permeable the concrete, the lower will be its resistance to deterioration.

72
The parameter that has the largest influence on durability is w/c or w/b ratio. Permeability
of concrete does not only depend on porosity, but also on the size, distribution, shape and
continuity of the pores (Mehta and Monteiro, 1993). When the w/c or w/b ratio decreases,
the porosity of the paste decreases and hence, the concrete decreases its permeability. Low
cement content, inadequate curing and poor consolidation also cause permeable concrete.
The permeability of cement paste decreases with age because gel gradually fills the
original water filled space. Permeability also varies with the degree of hydration. For
pastes hydrated to the same degree of hydration, the permeability is lower with lower w/c
ratio or higher cement content. For the same w/c ratio, the permeability of paste with
coarser cement particles is higher than that with the finer cement.

Neville (1997) quoted that the existence of the pores in concrete is of different kinds.
Therefore, it is important to distinguish between porosity and permeability. Porosity as
defined by Neville (1997) is a measure of the proportion of the total volume of concrete
occupied by pores, and is usually expressed in per cent If the porosity is high and the pores
are interconnected, they contribute to the transport of fluids through concrete, therefore, its
permeability is also high. However, if the pores are discontinuous or otherwise ineffective
with respect to the transport, then the permeability of the concrete is low, even if its
porosity is high.

Water permeability property as indicated by saturated water absorption of RHA blended

concrete specimens after 28 and 90 days of curing shows that at 28 days curing, the
percentage of water absorption increases with RHA content up to 35%. This is due to the
fact that RHA is finer than OPC and also it is hygroscopic in nature. When the curing time
was increased to 90 days the percentage of water absorption values decreased considerably
with increase in RHA content up to 25%. Even at 30% RHA, the value was lower
compared to that of control. Obviously, with prolonged curing, addition of RHA leads to a
reduction of permeable voids (Ganesan et al., 2007).

Ganesan et al., (2007) also found that the chloride diffusion coefficients of RHA blended
concrete specimens are continuously decreased with the increase in RHA content up to
25% of RHA. At 30% and 35% RHA, there is an increase in diffusion coefficient and

73
these values are also lower compared to that of control. This observation is true for both
28 and 90 days cured specimens. Thus, chloride diffusion is also considerably reduced by
partial replacement of OPC with RHA. There is a 28% reduction in chloride diffusion
coefficient for 30% RHA blended concrete compared to control concrete. The transport of
chloride ions through RHA blended concretes depends on the pore structure of the
concrete while the electrical conduction depends on both pore structure characteristics and
electrical conductivity of the pore solution.

Cook (1986) has also reported that highly reactive pozzolana, such as RHA is able to
reduce the size of voids in hydrated cement pastes, thus, making them almost impermeable
even at early age (7-28 days). Zhang et al. (1996), who also studied the incorporation of
the RHA in concrete stated that the used of RHA as cement replacement materials
successfully reduced the porosity and the Ca(OH) content. The amount in the interfacial
zone and the width of the interfacial zone between the aggregate and the cement paste
compared with the control portland cement composite was also reduced. However, the
porosity of the rice-husk ash composite in the interfacial zone was higher than that of the
silica fume composite.

2.12.1.1 Gas Permeability

Manmohan and Mehta (1981) in their investigation confirmed that in cement pastes
containing 10% to 30% of RHA, after 28 days of hydration the concrete was almost
impermeable. This is due to pore refinement attributed to RHA fineness or a
transformation of large permeable pores to small impermeable pores. This statement
agreed with the study conducted and reported by Mehta (1992), RHA being a pozzolanic
material is capable of pore refinement especially the paste in the immediate vicinity of the
zone surrounding coarse aggregate particles in concrete, thus, reducing the permeability of
concrete with increasing RHA.

Mehta (1992) also stated that each percent of RHA can improve at least 0.6 times of
permeability at year one and the permeability of the cement paste with RHA is in the

74
range of 1x10 -11 cm/sec. Therefore, there is a potential usefulness of RHA as a cement or
concrete additive for applications, where the corrosion of reinforcing steel is a major
concern, especially in those areas that need water resistance and good durability such as
that in the marine environment.

Another study conducted by Mehta (1989) showed that the addition of pozzolanic
admixtures (finely divided siliceous materials) to a concrete mixture improved the
properties of concrete mainly by modification of the transition zone structure. There is a
reduction in permeability of fresh concrete, causing a corresponding reduction in internal
bleeding which helps to decrease the porosity of the transition zone.

However, a study corroborated by Cook (1984) showed that the total porosity of mixture
containing RHA is greater although the permeability has considerably reduced.

The addition of microporous RHA absorbs large amount of water surrounding the solid
matter, thus , producing a low w/c ratio and refining the pore structure. Thus, the
pozzolanic reaction between cement and RHA modifies the pore structure; that is, it
occupies the empty space in the pore structure that reduces the water permeability of
concrete.

This statement agreed with Sugita et al. (1997) in which they stated that the more the
added RHA, the less the coefficient of air permeability and water permeability. They also
stated that the higher the w/c ratio (65% or 75%), the coefficient of air permeability and
coefficient of water permeability of RHA blended concrete are much smaller than those of
concrete without RHA addition. They quoted three reasons that attributed to such a
condition: the practical w/c ratio of concrete is decreased due to the addition of RHA;
more C-S-H gel and less portlandite are formed in concrete; and the average pore size of
concrete is reduced.

Speare et al. (1999) in their determination in terms of oxygen permeability found that with
higher content of cement (400 kg/m3), the coefficient of permeability is lower and with the

75
presence of RHA, the effect of using high cement content was found to be not significant
as compared to control mix. Table 2.22 shows the average results of coefficient of
permeability for RHA concrete cured in water for 28 days and stored at 20 ± 2°C and 65 ±
5% RHA at the age of 37 days.

Table 2.22: Oxygen Permeability Coefficient of Cement-RHA Blended Concrete

Mixture . w . Coefficient, K
Designation c + RHA (m2 )
OPC 0.32 4.62 x 10 -17
5% RHA 0.34 4.10 x 10 -17
10% RHA 0.37 4.36 x 10 -17
Adapted from Speare et al., (1999)

2.13 Microwave Incinerated Rice Husk Ash (MIRHA) as Pozzolanic Material

This is a study on the usefulness of using an agro-waste, known as microwave incinerated

rice husk ash (MIRHA) as part replacement of cement with locally available ingredients.
The performance will be compared with that specimen using silica fume which is widely
used in producing high strength concrete (HSC) (Salmia Beddu et al., (2009).

In this study, an effort was made to evaluate the usefulness of using MIRHA as part
replacement of cement with locally available ingredients. According to Nuruddin et al
(2008), compressive strength development of MIRHA concretes has been significantly
higher compared to control concrete. The adequate amount of water and high pozzolanic
reactivity were believed to be the main cause of this faster acceleration of MIRHA
concrete. During thus stage, MIRHA performed its function both as a pozzolanic material
and filler (Nuruddin et al (2008).

76
Table 2.23 Percentage Increment in Strength between MIRHA and SF Concrete

% increment in strength
Mix w/b
3 to 7 days 7 to 28 days
MIRHA500a 0.3 14 5
SF500a 0.27 22 26
MIRHA500b 0.3 21 15
SF500b 0.27 39 39
MIRHA550a 0.3 19 16
SF550a 0.27 10 7
MIRHA550b 0.3 62 1
SF550b 0.27 53 19
Adapted from S.Beddu et al. (2009)

A study by (Salmia Beddu et al., (2009) as indicated in Table 2.23, MIRHA gave higher
early age strength compared to silica fume. The highest increment of compressive strength
of MIRHA and silica fume from 3 to 7 days is 62% and 53% respectively. The highest
increment between 7 and 28 days for MIRHA and SF concrete is 16% and 39%
respectively. The finer particles size enables MIRHA to act as filler that seeps into the tiny
spaces between cement particles and as well as spaces between cement particles and
aggregate. A greater surface area providing space for nucleation of C-S-H and calcium
hydroxide Ca(OH)2. This will accelerate the reactions and form smaller calcium hydroxide
crystals.

The amount of alkalis in MIRHA used, referred to K2O is 3.77%. The presence of alkalis
(high pH) will break the glassy structure of Ca(OH) and that become easier and the
dissolution will become fast. Singh et al. (2002) indicates that in the presence of RHA,
some alkalis are liberated in the solution which further breaks the glassy network of
GGBS and enhances its reactivity with calcium hydroxide. RHA also reacts with calcium
hydroxide resulting into complete elimination of Ca(OH). The effect of alkalis has also
been found to increase early strength development and decrease the long term strength
(Neville, 1997).

77
2.13.1 Production of Microwave Incinerated Rice Husk Ash (MIRHA)

By burning method, the ash produced contains about 95% pure silica and if properly
prepared, it is in an active form which behaves very much like cement (United Nations
Industrial Development Organization,1984). Green Concrete as the name suggests is eco
friendly and saves the environment by using waste products generated by industries in
various forms like rice husk ash, micro silica to name a few, to make resource-saving
concrete structures. The use of green concrete helps in saving energy, emissions and waste
water. Green concrete is often also cheap to produce as it uses waste products directly as a
partial substitute for cement, thus saving energy consumption in production of per unit of
cement. Over and above all green concrete have greater strength and durability than the
normal concrete.

Microwaves are part of the electromagnetic spectrum and are located between 300 MHz
and 300 GHz. Microwave heating is defined as the heating of a substance by
electromagnetic energy operating in that frequency range. There is a fundamental
difference in the nature of microwave heating when compared to conventional methods of
heating material. Conventional heating relies on one or more of the beat transfer
mechanisms of convection, conduction, or radiation to transfer thermal energy into the
material. In all three cases, the energy is deposited at the surface of the material and the
resulting temperature gradient established in the material causes the transfer of heat into
the core of the object. Thus, the temperature gradient is always into the material with the
highest temperatures being at the surface. In microwave heating, the microwave energy
not only interacts with the surface material but also penetrates the surface and interacts
with the core of the material as well. Energy is transferred from the electromagnetic field
into thermal energy throughout the entire volume of the material that is penetrated by the
radiation.

Microwave heating does not rely on conduction from the surface to bring heat into the
core region. Since the heating rate is not limited by conduction through the surface layer,
the material can be heated quicker. Another important aspect of microwave heating is that
it results in a temperature gradient in the reverse direction compared to conventional

78
heating. That is to say, the highest temperature occurs at the centre of the object and heat
is conducted to the outer layer of the material..

Advantages of microwave firing can be summarised as: Short cycle and uniform heating
of even the special shaped products can be realised thanks to internal heat generation of
products by microwave and high energy efficiency can be realised as the heating of kiln
body is not required and finally, The environment-friendly heat treatment can be realised
with less noise and exhaust gas.

Due to the husk that is hard to handle, people normally burn it and bury under paddy fields
as organic manure. By controlling the burning temperature, RHA can be a green material
and re-utilized in construction materials. Therefore, burning procedure to obtain RHA that
is highly reactive that needs to be established, so that it can be used as cement
replacement. Hence, the optimum replacement percentage of OPC by RHA that improves
the quality of concrete need to be identified.

Burning the RHA with higher temperature increases the SiO2 content. However, it is not
suggested to burn rice husk above 800°C longer than one hour, because it tends to cause a
sintering effect (coalescing of fine particles) and is indicated by a dramatic reduction in
the specific surface (Hwang and Chandra, 1997). In order to produce MIRHA with high
reactive silica content, controlled combustion of rice husk the equipment used for this
experiment was specially designed and developed at the Universiti Teknologi Petronas.

The UTP Microwave Incinerator (UTPMI) used in the research adapted the Air Cooled
Magnetron system with an overall dimension of 2.3m(H) x 4m(W) x 4m(L) with a
chamber capacity of 1m3. Ceramic filter is used in the emission and ash control system
with PLC mode of operation. Flue Gas Filter equipped with the microwave incinerator
provides significant positive effect to the environment. It distils all the dust and ashes that
are resulted from rice husk incineration, hence the air pollution from burning process can
be reduced (Hwang and Chandra, 1997). The temperature ranges to 1600°C with operating
temperature of 800 oC.

79
The effect of burning temperature and percentage inclusion of MIRHA on normal strength
concrete to obtain quality rice husk ash that can significantly improve the concrete
compressive strength. The utilization of Microwave Incinerated Rice Husk Ash (MIRHA)
into the concrete mix proportions has given various effects to the concrete properties. The
improvement of compressive strength results of MIRHA concrete samples was influenced
by the quality of MIRHA and mix proportion used.

Based on the study done by Kamal et al, (2008), the effect of microwave incinerated rice
husk ash (MIRHA), burnt at different burning temperatures on the compressive strength of
concrete. The performance of concrete mixes was determined by the compressive strength
test at ages 3, 7 and 28 days.

The test results showed that 5% replacement of MIRHA with 800°C burning temperature
provided good acceleration in compressive strength development in concrete. It was also
shown that the inclusion of 10% replacement of MIRHA with 700°C burning temperature
improved the strength of concrete. MIRHA with burning temperature 600°C also revealed
that the optimum replacement of MIRHA was 5%. The results were compared to the
normal concrete without MIRHA.

2.13.2 MIRHA high strength concrete

(Salmia Beddu et al., (2009) in their investigation confirmed that the highest increment of
compressive strength of MIRHA and silica fume from 3 to 7 days is 62% and 53%
respectively. The highest increment between 7 and 28 days for MIRHA and SF concrete is
16% and 26% respectively. The finer particles size enables MIRHA to act as filler that
seeped into the tiny spaces between cement particles and as well as spaces between
cement

particles and aggregate. A greater surface area providing space for nucleation of C-S-H
and calcium hydroxide Ca(OH)2. This will accelerate the reactions and form smaller

80
calcium hydroxide crystals. It is possible to get strength above 70 MPa using MIRHA. The
performance of MIRHA is almost similar to silica fume even though MIRHA need higher
water demands. At early age the performance of MIRHA concrete is better than silica
fume concrete which is the maximum increment of 62% compared to 39% respectively.

2.13.3 MIRHA foamed concrete

Nuruddin and Bayuaji, (2009) in their study found that the excess non­ evaporable water
(NEW) in MIRHA foamed concrete demonstrated that the hydration of MIRHA foamed
concrete, which produced calcium silicate hydrates, was more than the normal plain
foamed concrete. This in turn would enhance the strength and durability of the foamed
concrete. Nevertheless, this was only true for 5% inclusion which was the optimum level.
For 10% and 15% samples, the capability to react with Ca(OH)2 available from the cement
hydration process was quite limited since the amount of Ca(OH)2 present in the paste was
sufficient for 5% inclusion.

It can be said that for foamed concrete modified with 5% level of MIRHA provided some
favorable effects. From the raw data, the 5% MIRHA foamed concrete has on average
across the board, 13% higher compressive strength compared to plain foamed concrete.
Definitely, this is a way forward for MIRHA foamed concrete to be further developed and
to be used as structural members in the concrete construction industry. (Nuruddin and
Bayuaji, 2009).

A study by Nuruddin and Bayuaji, (2009) reported that the investigation carried out is to
better understand the effect of the MIRHA, water binder ratio, and sand cement ratio on
the foamed concrete. The experimental work (using the Taguchi method) is designed to
give the optimum working conditions of the parameter that affects the physical properties
of concrete

mixtures. One of the advantages of Taguchi method over the conventional design, in
addition to keeping the experiment cost at the minimum level, is that it minimizes the
variability around the investigated parameters when bringing the performance value to

81
target value. Its other advantage is that the optimum working conditions determined by the
laboratory work can also be reproduced in the real production environment.

A new method is established using taguchi method for determination of the optimum
composition of material proportion and the effect of MIRHA properties and durability of
foamed concrete. Lightweight Foamed Concrete (LWFC) consists of many components; it
is critical to use a systematic approach to identify optimal mixes and investigate the most
effective factors under a set of constraints (Fadhil and Ridho, 2009).

2.13.4 Effect of Used Engine Oil on MIRHA Concrete

Nuruddin, Shafiq, and Beddu, (2009) reported that the strength increment of MIRHA
concrete is due to the high amorphous silica and fineness of MIRHA used. The effective
burning method of UTP microwave incinerator adapted to produce amorphous silica in the
rice husk ashes is instrumental in creating good quality MIRHA concrete. The addition of
MIRHA causes an increment in the compressive strength. This is due to the capacity of
pozzolan, consuming the calcium hydroxide generated during reactions of hydrated
cement paste. The end product of the pozzolan reacts to the formation of C-S-H gel.

Compressive strength of concrete containing 0.15%-0.5% used engine oil that gives
almost similar result and starts to decrease at 0.8% of usage. It was believed that the
decrease of compressive strength due to the heavy metal content can delay the setting and
lower the strength. Used engine oil can act as chemical plasticizer the same as
superplasticizer to improve workability, air content and strength of MIRHA concrete. It
was found that 0.5% inclusion of used engine oil demonstrated better concrete with
respect to slump value, air content and compressive strength compared to MIRHA control
mix.

82