Characterization of metakaolin and study on early age mechanical strength of hybrid cementitious composites

Construction and Building Materials 121 (2016) 599–611

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Characterization of metakaolin and study on early age mechanical strength of hybrid cementitious composites Mahyuddin B. Ramli, Alonge O. Richard ⇑ School of Housing, Building and Planning, Universiti Sains Malaysia, Penang, Malaysia

h i g h l i g h t s
MK can be produced in the laboratory at 800 °C for 1 h.
MK,CNS and epoxy resin at 10%, 1% and 1% respectively reduces slump value and increase the density as the ages increase.
The flexural strength increases at all ages with incorporation of natural and synthetic fibres.
The relationship between the compressive strength and flexural strength is significant.

a r t i c l e

i n f o

Article history: Received 10 December 2015 Received in revised form 23 May 2016 Accepted 14 June 2016 Available online 18 June 2016 Keywords: Metakaolin ECC Nanosilica Natural fibre Epoxy resin Mechanical strength

a b s t r a c t In this study, the potency of using laboratory produced metakaolin as partial replacement of cement in the evolution of a hybrid cementitious composites (HCC) was investigated. Some chemical composition and phases of metakaolin were appraised with the aid of X-ray fluorescence (XRF) and X-ray diffraction. Particle size analyzer was used to study the physical properties of metakaolin produced in the laboratory from purified Kaolin. Early age mechanical strength and shrinkage properties of structural grade HCC developed by replacing fly ash in the original ECC M45 design with metakaolin (MK), colloidal nanosilica (CNS) and epoxy were examined. The designed mixtures consist of five mixtures with the control and base mix exposed to water and seawater. The control mix consists of only the fine aggregate and cement while the base mix consists of cement, fine aggregate, 10% MK and CNS with epoxy resin at 1% each by weight of the binder. Synthetic barchip fibres and some natural fibres namely, oil palm fruit bunch and coconut fibres were incorporated. The water cement ratio was designed as 0.30 for all mix ratios. The result reveals that the inclusion of MK, CNS and epoxy have a significant effect on the fresh properties of the HCC and at early age of 7, 28 and 90 days. The mechanical strengths were better improved than the standardized M45 ECC at these early ages. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction The impressive headway in the development of highly improved and performance cementitious composite materials cannot be an oversight in these late years. This includes numerous of high strength concrete usually produced with low water binder ratios such as pultruded continuous fibre reinforced concrete which was pioneered by Mobasher et al. [1]. High performance fibre reinforced cementitious composite which has a unique improved strength as easily as high ductility due to a strain hardening response [2], Eco friendly concrete that contain an increase by-products and mineral admixtures which make it to be more ⇑ Corresponding author. E-mail addresses: [email protected] (M.B. Ramli), [email protected] (O.R. Alonge). http://dx.doi.org/10.1016/j.conbuildmat.2016.06.039 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.

environmentally friendlier than the conventional concrete. Green high performance cementitious (GHPC) material is developed in a decade ago as established by Zhongwei [3]. The most important features of GHPC is the use of several admixtures to replace cement content partially [4]. Similarly, this category of concrete material consists of a matrix that does not contain coarse aggregates hence regarded as concrete materials. The most acceptable and widely use mineral admixtures are silica fume (SF), fly ash (FA) and ground granulated blast furnace slag (SL) which are byproducts of industrial process according to Zhu [5]. Nonetheless, other cementitious materials like metakaolin is a natural pozzolanic material derived from calcinated clay would have bigger potential for enhancing engineering properties in concrete. Engineered Cementitious Composites, ECCs as usually called, is a class of high performance fibre reinforced composites and also a green high performance cementitious material due to the nature of

600

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

cement replacement. It is typified by its unique excellent strain capacity under uniaxial tension, with a controlled fibre volume fraction usually 2% or less [6]. ECC is similar to a ductile metal, hence has the capacity to strain harden after first crack with a demonstrated strain capacity of between 300 and 500 times greater than conventional concrete. In contrast to the conventional ordinary concrete, ECC display self-controlled crack width under increasing load. Usually, after initial loading, there is a formation of small numbers of cracks within the material which then spread. The widening progress until an average of about 60 lm which is a product of ECC micro-mechanical tailoring [7]. Better strain capacity and multiple cracking are achieved in ECC through the use of fine sand only [8,9]. The elimination of coarse aggregates in ECC mixtures resulting in a higher content of cement in respective of conventional concrete structures. For instance, a typical ECC can contain cement that is more than 1000 kg/m3. It is established through studies, that each ton of cement produced give rise to an equal quantity of carbon dioxide, which is believed to be the origin of global greenhouse gas emission created by mankind, at least up to 5% [10]. Based on this fact, it is crucial to modify ECC mixture and develop a more environmentally friendly composite material by the incorporation of a natural mineral admixture to partially replace the total cement contents. Likewise, the current trends of ECC make use of Fly ash. It constitutes the necessary ingredients for the formation of ECC to improve the mechanical properties and as well reduce the shrinkage [11,12]. Nonetheless, delay of setting time and depression of early age mechanical strength as well as incompatibility with some other type of cement which may result in early stiffened of concrete materials limit the effectiveness of ECC in some special applications structurally. Hence, this research study takes in an effort to apply a natural fine aggregate with varied size ranges and a natural admixture, MK, produced in the laboratory from purified Kaolin as a partial replacement of cement. The developed HCC is better in early ages mechanical properties. Natural pozzolans generally are included in concrete mixtures to aid in the conversion of calcium hydroxide (CH) which is separated as a less desirable hydration product into the more desirable calcium silicate hydrate (C-S-H). ASTM C 618 spells out class N porcelain as raw or calcined natural pozzolans and the frequently used natural pozzolans in the present dispensation are processed materials, it is treated with heat in a high temperature kiln and then grounded into a fine powdery form. They are metakaolin, calcined clay and calcined shale. Calcined clay is used in general construction like other pozzolans while calcined shale also has some cementing or hydraulic properties. Meanwhile, metakaolin is a product of calcination of high purity kaolin clay at a lowtemperature and then grounded to an average particle size of about 1–2 lm. This is considered to be about ten times finer than cement, but yet 10 times coarser in sizes compares to silica fume [13]. The incorporation of various supplementary cementing materials (SCM’s) in concrete can result to significant effect on early age properties (both fresh and hardened). MK, for instance, was found to improve compressive and indirects tensile strengths, according to past studies, MK mixtures incorporation with between 5% and 10% MK by mass of cement [14]. In another related study, Li and Ding [15] investigated 10% cement weight replacement with MK, combine it with only OPC or with OPC and ultra-fine slag. The results revealed that the compressive strength of the mortar mixture with the incorporation of MK was found to be higher in value than the control mixtures and it was recorded to be approximately 8Mpa higher as at 28 days. Although, it was also recorded that the MK-slag mixtures manifested the highest 28-day strength. These strength contribution were to be assessed as a result of three basic factors which are; the effect of the filler, the OPC hydration

acceleration and the pozzolanic reaction of MK with the CH [16]. MK, also caused an increase in the autogenous shrinkage when measured from the age of 24 h according to past studies [17]. 2. Objectives This study aimed to investigate the characteristics of laboratory produced metakaolin (MK) and the early age mechanical strength properties of the hybrid cementitious composites produced by the replacement of binder weight by 10% MK, 1% colloidal nanosilica (CNS) and 1% epoxy resin without hardener. The main focus will be; first, to characterise the MK relatively to particle size distribution, specific surface area, mineralogical phases and chemical composition of the laboratory produced MK, and second, to investigate the early age mechanical strength of hybrid cementitious composites (HCC) with the incorporation of MK, CNS and epoxy. Lastly is to study the drying shrinkage properties of the HCC. 3. Material and methods 3.1. Materials 3.1.1. Metakaolin The MK used in this study was produced in the laboratory using the purified Kaolin supplied by a local supplier called Scancem materials Sdn. Bhd. The calcinations of the Kaolin to metakaolin was done using the ELE International laboratory muffle furnace at 800 °C for 1 h. After the calcinations, the MK was sieved using 150 lm through the EFL 2000 Endecotts siever machine. The particle has 99% particles and its with a mean particle size diameter (d50) of 2.80 lm with a specific gravity of 2.6. The specific surface area is 0.015 m2/kg. The chemical composition of the MK produced in the laboratory is shown in Table 1. 3.1.2. Cement The Ordinary Portland cement that complies with ASTM Type I Portland cement with a median particle size value of 6.14 lm, the specific surface area of 1123.3 m2/kg and the specific gravity of 3.02 were used in this experimental study. The initial setting time and final setting time in minutes after the test are 115 and 310 respectively. The physical and chemical properties of the cement used is in accordance with the specifications in ASTM Standard C150 (ASTM, 1997b). The chemical composition of the OPC is shown in Table 2. 3.1.3. Nanosilica Colloidal nanosilica (CNS) was used in this research study as a cementitious material. The nanosilica used in this study was a product of Sigma–Aldrich product supplied locally. Specifically, it’s Ludox AS-40 colloidal silica 40 wt.% suspension in water (H2O). The total weight is 60.08 g/mol. The physical and chemical properties are as stated in Table 3. 3.1.4. Epoxy resin This research study made use of epoxy resin (CP 370A) bisphenol A. The epoxy resin was used as cementitious material and as a bonding agent in the mixture. The physical composition of the epoxy is highlighted in Table 4.

Table 1 The physical properties of metakaolin produced in the laboratory in % weight. SiO2 53.03

TiO2 0.93

Al2O3 35.63

Fe2O3 1.81

MnO 0.02

MgO 0.57

CaO 0.04

Na2O 0.04

K2O 1.88

P2O5 0.06

The LOI (Loss on ignition) = 1.99% obtained in compliance with BS EN 15935 (2009).

601

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611 Table 2 Chemical composition of ordinary portland cement in mass (%). Constituents

Ordinary Portland Cement

Lime (CaO) Silica (SiO2) Alumina (Al2O3) Iron Oxide (Fe2O3) Magnesia (MgO) Sulphur Trioxide (SO3) N2O Loss of Ignition Lime saturation factor C3S CS C3A C4AF

64.64 21.28 5.60 3.36 2.06 2.14 0.05 0.64 0.92 52.82 21.45 9.16 10.2

Weight Retain (G)

Percentage Retain (%) by mass

Cumulative passing

5.0 2.0 1.18 0.60 0.30 0.15 0.063

– – – 160.8 88.2 44.4 5.3

– – – 54 29 15 1

– – – 54 83 98 99

Table 6 The physical properties of the fine aggregate.

Slightly hazy to hazy Colourless Liquid 9.0–9.5 1.286–1.300 620 cps 39.5–41.0% 0.129–0.155 m2/kg 60.02% 620% 60.10% 20.0–24.0 nm. 60.01%

Table 4 Physical properties and chemical composition of epoxy as supplied by the manufacturer. Properties Colour Mixing ratio (weight) Curing Schedule Gel time (min) Tensile strength (psi) Elongation at break (%) Shore D hardness Tensile strength Volume resistivity Heat deflection temperature (HDT)

Sieve size

The fineness modulus = 99/100 = 0.99.

Table 3 Physical properties and chemical compositions of nanosilica as supplied by the producer. Appearance (clarity) Appearance (colour) Appearance (Form) pH at 25 °C Specific gravity at 60 oF Viscosity at 25 °C Silica Surface area Na2SO4 Assay (%) T Sodium (Na) Particle Size Chloride content

Table 5 Sieve analysis of the fine aggregates.

Epoxy Clear 100 4 h at room temp. 135 9200 4 83 P80 N/mm2 1015 Ohm-cm P70 °C

3.1.5. Aggregates (Fine natural sand) The fine aggregates used in the study were natural sand supplied by local contractor. The sieve analysis of fine aggregate was carried out in accordance with BS EN 12620: 2002 to obtain it grading and specific gravity. The fineness modulus was estimated. The fineness modulus is the total sum of the accumulated percentage retained on the specific sieves divide by 100 as it is specified in ASTM C 33. The fineness modulus of the sand used in this research study is 0.99. The largest particle size was retained in 0.60 mm sieve aperture and a maximum size of 600 lm. The sieve analysis is shown in Table 4 while the properties are shown in Table 5. 3.1.6. Fibres 3.1.6.1. Coconut fibre and oil palm fibre. The coconut fibre and oil palm fruit bunch fibre used in this experimental programme was supplied locally by Fibre X (M) Sdn. Bhd, Malaysia. The two natural fibres were supplied in the form of meat and then chopped manually to predetermine length. The properties are shown in Tables 6 and 7. Their chemical compositions are highlighted in Table 8.

Physical properties

Results

Specific gravity Water absorption

2.62 2.1%

Table 7 The specification of coconut fibre as provided by Fibre – X (M) Sdn. Bhd. Item

Unit

Test Value

Diameter Length Tensile Strength Elongation Toughness (N/mm2) Elastic Modulus Specific Gravity Water absorption

mm mm N/mm2 % N/mm2 GPa – % (24–48 h)

0.10–0.41 8–12 mm 120–200 25.0 3000–3200 19–26 1.13 7.2

Table 8 Specification of oil palm empty fruit bunch. Item

Unit

Test Value

Diameter Length Tensile Strength Elongation Toughness (N/mm2) Elastic Modulus Specific Gravity Water absorption

mm mm N/mm2 % N/mm2 GPa – % (24–48 h)

0.21–0.27 8–12 mm 240–260 14.0 2000–2200 2–6 1.24 0.6

3.1.6.2. Synthetic fibre (Barchip). The synthetic fibre incorporated in the mixtures of this study is Barchip 54 fibres. It is a kind of modified Olefin based synthetic fibre produced and supplied by Elasto Plastic Concrete Inc. This synthetic fibre is inert to most of the aggressive agent attacks. It posses ability not to rust when exposed to air, water or chloride ions, it never oxidizes. Aside from this feature, barchip fibre is devised to draw together with embossed treatment to have surfaced with frictions, like contour so as to maximise the effects of bonding with the concrete matrix. The specification of the barchip fibre is shown in Table 9.

Table 9 The chemical composition of coconut and oil palm empty fruit bunch fibre as provided by the producer. Fibre

Lignin (%)

Cellulose (%)

Hemicellulose (%)

Ash Content (%)

Coconut fibre OPEFB fibre

40–45

32–43

0.15-0.25

–

19

65

–

2

602

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

3.1.7. Mixing water and superplasticizer A sulfonated melamine formaldehyde condensates type F superplasticizer was utilized in this study. The superplasticizer was integrated into the HCC mixture at the specific dosages to attain the desire workability. The portable water was locally supplied by the local authority through the main source of supply. 3.2. Methods 3.2.1. The characterization of MK The characterization of MK was determined by investigation the chemical composition of the laboratory produced MK using X-ray spectrometer Rigaku model RIX3000. The mineral phases of all oxide present were detected using the X-ray fluorescence analysis and X-ray diffraction using Bruker X-ray diffractometer. Fingerprint method was engaged in the analysis of the X-ray diffraction patterns [18]. The MK’s Loss on Ignition (LOI) was assessed in conformity with the prescribed ASTM Standard C311 [19]. The median particle size diameter (d50) cum the specific surface area of MK were assessed using Malvern model laser diffraction analyser. The MK’s specific gravity value were assessed with the aid of Le Chatelier Flask and the procedure highlighted in ASTM Standard C188 [20]. 3.2.2. Mixture proportioning and mixing procedure This study’s proportion of mixture are binder, sand and water ration is 1:0.8:0.30 respectively for all the HCC cast. The binder which is cement was replaced with laboratory made MK at replacement level of 10%, CNS at 1% and epoxy at 1% by total weight. The superplasticizer used was at a dose that will enable the mix to attain the desired slump of 150 ± 5 mm. The mix design followed the standard M45 ECC with some modification in terms of material replenishment. The HCC mix components are as detailed in Table 10. 3.2.3. HCC mixing and curing regime A laboratory mechanical mixer was used for each batch of HCC in accordance with ASTM 192. The procedure follows that the fine aggregates was placed first in the mixer, thereafter, the cement was fed into the mixer, then half of the water with other additives and the superplasticizer. The mixing continues for some minutes before various fibres were introduced gradually before the last batch of sand and water are added. This process was followed to achieve the homogenous mix and evenly distribution of the fibres. It complies with the specifications in ASTM Standard C305 [21]. 3.2.4. Flow test The flow test was conducted on the fresh HCC mixes using the flow table in compliance with the specifications highlighted in BS 1881:Part 102 [22]. The fresh density of the HCC was measured in accordance to ASTM C1611.

Table 10 Specification of barchip fibre. Characteristic Base resin Length Tensile Strength Surface texture No. of fibres per kg Specific gravity Young’s modulus Melting Point Ignition Point

Unit

Property

mm N/mm2 – Nos – GPa °C °C

Polyolefin 8–12 mm 640 Continuously embossed P50,000 0.95 8.2 150–165 >450

3.2.5. Bulk density and mechanical strength tests The bulk densities of the hardened HCC mixes were examined in respect of methods as highlighted by BS 1881: Part 114 [23]. HCC mortar samples with 50 mm cube sizes were moulded, cured and tested in compliance with the subroutines defined in ASTM Standard C109 [24] for compressive strength determination. The flexural strength of the HCC mixes was assessed using prism specimen with dimensions of 40  40  160 mm, it was moulded, cured and tested in compliance with ASTM Standard 348 [25]. All the samples are cured in both ordinary water and sea water till the date of test which are 7, 28, and 90 days. The results of the compressive and flexural strengths at any given ages are the mean of three numbers of the samples of HCC tested. At the same time, the ultrasonic pulse velocity pulse transmitted through the hardened HCC mixes was examined with the engagement of an electrical pulse generator and the method of testing highlighted in BS 12504: Part 4 [26]. A big prism samples of size 100  100  500 mm were used for the UPV test. Direct transmission method was used in the processing of assessing the transmission of an ultrasonic pulse through a constant path length of 100 mm between the two transducers. 3.2.6. Shrinkage test In this study, the drying shrinkage test was conducted in compliance with ASTM C 157:2004 [27] and ASTM C596: 2009 [28]. The specimens used for this examination consists of four units sample size of 280 mm  75 mm  75 mm HCC bars. It was moulded using a standard stainless steel mould specially made for shrinkage test. It complies with the description of ASTM C490: 2011 [29] in term of dimension. The HCC bars were formed with a cast-in steel gauge stud through which the measurement of the length change was taken. The HCC samples were de-moulded after 24 h from the time it was cast. The fabricated HCC sample was subjected to drying atmosphere with the temperature of 23 ± 2 °C and a relative humidity of 50 ± 4% in compliance with the conditions highlighted in ASTM C 157 (ASTM, 2004a). The initial measurement of both the reference bar and the HCC specimens were performed after 24 h of casting and subsequent measurement were conducted every day till the 7th day and then 14, 28,56, 90 days of exposure to drying environment. 4. Results and discussion 4.1. Characterisation of MK 4.1.1. The morphology of MK The chemical compositions of the produced MK and the raw clay is shown in Table 11 and it was observed that the chemical compositions include SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO, CaO, Na2O, K2O, P2O5. Out of all the compounds present, it was observed that SiO2 and Al2O3 are the major oxide compounds that are present in both MK and the raw kaolin. The percentage contents are 53.03% and 35.63% respectively for MK, while it was 44.45% and 30.50% respectively for kaolin. This result supported the assertions made by Ambroise et al. [30] as cited by Rashad [31] that MK typically contains between 50% to 55% SiO2 and between 40% to 45% of Al2O3. Other oxide compounds present in the compositions are TiO2, Fe2O3, MnO, MgO, CaO, Na2O, K2O and P2O5. Although, they are presented in small percentages, the difference in the chemical composition, contents shows the effect of the heating on the raw kaolin. The increases in the percentage content of SiO2 and Al2O3 in MK over kaolin are 16% and 15% respectively. This result passes the ASTM C618 because the minimum total expected of SiO2 and Al2O3 compounds was put at 85% and the total SiO2 and Al2O3

603

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611 Table 11 The mix proportion for all mixes. Specimen

OPC (kg/ m3)

MK (kg/ m3)

Nano-SiO2 (kg/m3)

Epoxy (kg/ m3)

Fine aggregate (kg/m3)

Water (kg/ m3)

Superplasticizer (kg/m3)

OPFBF (%)

C BM OPFBF CF BF

82.98 74.67 74.67 74.67 74.67

– 8.29 8.29 8.29 8.29

– 0.01 0.01 0.01 0.01

– 0.01 0.01 0.01 0.01

66.38 66.38 66.38 66.38 66.38

24.90 24.90 24.90 24.90 24.90

0.82 0.82 0.82 0.82 0.82

2%

compounds from the calcinations of purified kaolin in this study is 88.66%. The results also revealed that the purified clay, Kaolinite actually re-crystallizes whereby it was rendered to mullite and spine. The kaolin is a phyllosilicate which consists of alternate layers of ions. It breaks down the kaolin structures so much that the alumina and silica layers become crumple silica and alumina in tetrahedral and octahedral coordination form respectively. The heating process at the 800 °C for 1 h, causes the kaolin to lose 14% of its mass inbound hydroxyl and let loose of their long-range order, hence recrystallize the layers of electrically neutral crystalline structures of the raw kaolin. The kaolinite flakes became more deformed and locally condensed into the boundless and then yield metakaolinite, a material that possesses some degree of order highly reactive transition phase. In the metakaolinite, the Si-O network became more static, hence remains largely intact, whereby the Al-O structure network became well reorganized compared to pre-heating period. This is justified by past studies [31–35]. 4.1.2. X-ray diffraction analysis The Figs. 1 and 2 highlights the X-ray Diffraction patterns of both the purified raw kaolin and calcinated kaolin respectively. According to the relative patterns as revealed, the samples mainly contain Quartz, kaolinite and Muscovite. It demonstrates that the ternary compounds are the major chemical phases of the MK.

CF (%)

BF (%)

BF+OPFBF (%)

BF+CF (%)

2% 2%

The sustainable alteration mineral assemblage generally consists of Muscovite [KAl2(Si3Al)O10(OH)2], the quartz and kaolinite [Al2(Si2O5)(OH)4]. In the result, it is observed that the chemical phases concur with the results obtain in the X-ray Fluorescence analysis highlighted in Table 7. The XRD analyses established the identification of kaolinite, class (O-H), i.e., triclinic 1A, muscovite, class (O-H) monoclinic 2M1 and quartz which is hexagonal and more dominant. The kaolinite is the mineralogical reference term for hydrated aluminium disilicate, that is Al2Si2O5(OH)4. Which is the primary constituent of kaolin, it is between 40% and 70% content according to Moulin etc. [36]. The determination of Muscovite is based on the K2O content of the samples since it is certain that there is no other detectable mineral which contain K. After the extraction of Al2O3 in muscovite, it’s used for kaolinite determination since there is no other Al containing mineral detected in the samples. The quartz contains SiO2. This mineralogy phase, when used as the cementitious supplementary binder with cement enhances the principal reaction between AS2 and CH which is derived from hydration of cement in the presence of water. The reaction yielded an additional cementitious C-S-H gel, together with crystalline products which includes calcium aluminate hydrates as well as aluminosilicate hydrates (C2ASH8, C4AH13 and C3AH6) [37]. It can be observed that the peaks of Quartz are significantly larger than what is obtained in kaolin and this corroborate with

Fig. 1. The XRD of the kaolin.

604

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

Fig. 2. The XRD for the calcinated kaolin, MK produced in the laboratory.

the findings of Seleem et al. [38]. The high SiO2 content in the MK is indicated by the abundance of quartz. 4.1.3. DTA & TGA test of MK The powder samples of MK were also analysed by thermogravimetric analysis (TGA) and differential thermal analysis (DTA) in order to examine the thermal reactions during the heating process from the ambient temperature up to the 1000 °C at the rate of 5 °C per minutes under the atmospheric pressure. It is also used for the purity determination. This thermal behaviour of the raw kaolin is presented in Fig. 3. The major changes detected by TG and DTA are as stated. Between the temperatures of 0 °C to 100 °C, the material was absorbing heat and release of water absorbed into the pores and on the surface. But between 100 °C and 450 °C, the kaolin begins with loss mass and this is attributed to the predehydration process leading to the reorganization in the octahedral layer of the raw kaolin. At temperature between 450 °C and 600 °C, dehydroxylation of kaolinite take place and at the temperature above this, 600 °C and 800 °C, there is formation of metakaolinite. The temperature above this value and upwards to 1000 as resulted in the formation of mullite as denoted by the exothermic peak. On the other hand, the maximum endothermic peak observed was at the 600 °C and this was attributed to dehydroxylation process. The results as noted through the DTA/TGA concur with the result and report of some past studies [39–41].

between 783.1 nm and 2.78 lm. It indicates the layer by layer of MK substructure. The particle shapes, as observed is more like plates with a size of less than 3 lm. Aside from this, some small particles which are more sphere-like morphology is observed. The size of the spherical particles was observed to be less than 800 nm which could be as a result of the calcinations process. These findings of different size particles of the produced MK corroborate the result of the particle size distribution obtained by the laser beam particle size analyser as shown in Table 12. Meanwhile, the plate-like, layer by layer structures are depicted in Figs. 6 and 7. The specific gravity of the produced MK was determined using the Le Chatelier’s flask and the procedure complied with ASTM C188, 1995 [42] (ASTM, 1995). The result was estimated to be 2.6. This conforms with the result obtained by [13,37,41] The loss on ignition of the produced MK in powder form was examined in compliance with the procedures highlighted in ASTM Standard C311 [42]. The result of the findings after calculation was 1.99%. This means that the total CO2 that was loss on ignition in the MK specimen is 1.99%. This is within the range of the findings of other past studies [37,43,44]. When this result was compared with that of the raw purified Kaolin which was estimated to be 10.13%, it was observed that the total CO2 content was very high compared to that of the produced MK. It was about 20% higher. 4.2. Early age mechanical strength of HCC

4.1.4. Particle size, crystal morphology and composition of MK The particle size, the crystal morphology and composition of the MK were also determined using a laser beam particle size analyser (Malvern Mastersizer), scanning electron microscope (SEM) and energy dispersive X-ray (EDX). The SEM images are presented in Figs. 4 and 5. After the heating process, the structure of the MK was seen clearly, using SEM and energy dispersive X-ray, the scale particle of the MK ranges

4.2.1. Flow value of HCC specimen The Flow values of the HCC mixes produced with the dosage of superplasticizer used to reach a targeted value of 150 mm ± 5 mm are detailed in Table 13. The outcome indicates that the base mix (BM) Flow value was a little lower than the control mix with a decrease value of 5 mm which represents close to 3.25%. This may be as a consequence of the incorporation of MK, which is

605

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

Fig. 3. The DTA/TGA curve of the raw Kaolin.

Table 12 The chemical compounds in raw kaolin and calcinated kaolin (MK). Compounds

Kaolin

Metakaolin

SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 Total

44.45 0.85 30.50 1.68 0.02 0.51 0.03 0.04 1.74 0.06 79.88

53.03 0.93 35.63 1.81 0.02 0.57 0.04 0.04 1.88 0.06 94.01

Fig. 4. SEM Image of particle morphology of MK at magnification of 1200.

Fig. 6. SEM image of plate-like structure of MK at magnificent 5000.

Fig. 5. SEM image of particle morphology of MK at magnification of 3000.

known for workability reduction. It can as well be as a result of the incorporation of CNS and epoxy. The CNS was observed to cause reduction in the flow spread of concrete mix [45]. All the remaining

mixes have a consistent Flow value ranges between 140 mm and 146 mm which are far lower than the control and the base mixes. These values stand for between 5% and 9.09% decrease. This may be ascribed to the frictions between the mixtures because of the presence of fibres. The fibres caused an addition in the coherency of the mixtures hence lack of workability. Also, the specific surface of the fibres controls the friction effects yet another factor is the

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

Fig. 7. SEM image of denture-like structures of MK at magnificent 5000.

Table 13 The particle size distribution of MK. Sample

dv10 (lm)

dv50 (lm)

dv90 (lm)

MK

0.87

2.80

9.91

fibre absorption properties. The absorption properties differ from one fibre to the other and this dictates the differences in the Flow values. The absorption property of coconut oil is the highest out of the three fibres, and then the Oil palm waste fruit bunch fibre and lastly the barchip fibre. In all the fibre mixes, then mixes with coconut fibres have the lowest Flow values of 143 mm which is 7.14% lower than the Flow value of the control mix and 4.03% lower than the Flow value of the base mix, BM mix. This is as a consequence of the fact that coconut fibre is highly absorbent. In the track of mixing, the coconut fibre absorbed the free water content of the mix hence increased the water demand of the premix. This in turn reduced the Flow of the mix considerably. The workability value of OPFBF was observed to be better than that of coconut mixes. The Flow value is 145 mm which represents 5.84% lower than the control mix and with 2.68% lower than the workability value of base mix BM. This may be as a result of the fact that OPFBF has lower water absorbing properties than CF. Among the three fibres, Barchip fibres have the least effect on the workability of the HCC mixture. The effect was reflected in the Flow results of 148 mm which represents 3.89% and 0.67%, respectively lower than control mix and base mix (Table 14). This phenomenon can be explained by the fact that when particle size gets finer it necessitate an additional water to facilitate the particle movement on each other, this resulted in the increase of viscosity. Meanwhile, a portion of the water is entrapped in the pores of the agglomerates hence contribute nothing to the flowability. These results compromise the results of past studies on fibres in both concrete and mortar mixes [46–48]. It was noted that the major parameters for the Flow values of all mixtures depend on the w/b, s/b and SP/b. But out of all, the SP/b was considered to be more significant as it contributes to the achievable Flow values. This confirms the past studies on the Flow of HCC mixtures [49].

Table 14 The flow of the HCC mixes with superplastizer dosage. HCC mix design

Superplasticizer Dosage(kg/m3)

Slump flow (mm)

C BM CF OPFBF BF

0.82 0.82 0.82 0.82 0.82

154 149 143 145 148

4.2.2. Density The results of the bulk density of all specimens, including the control specimens are presented in Fig. 8. The results disclosed that the density of all specimens increases with age. The cause for this might be as a consequence of continuous hydration process of cement in the presence of constant water supply due to the curing regime used in this research, that is, water and sea water curing regime. The bulk density at all ages were in the ranges of 2204 kg/m3–2249 kg/m3 for both water and sea water cured at 7 days apart from the control specimens Then, 2212 kg/m3–2253 kg/m3 for the two curing regime at 28 days apart from the control specimens, 2218 kg/m3–2255 kg/m3 for the same curing regimes at 90 days age. Meanwhile, at all ages, the control specimens have the highest densities. It is within the ranges of 2260 kg/m3–2294 kg/m3 This coincides with the result of past studies [49]. This may be ascribable to the incorporation of MK, CNS and epoxy with short discrete fibres. In respect of MK, the decreases in the densities might be as a consequence of the lower specific gravity because of the finer particles. The same can be predicted for the effect of the CNS on the density. The incorporation of fibres into the HCC has a tendency to trap more air in the course of mixing and vibration. The air voids created, will constitute the matrix to become porous and reduced the HCC density similar to the accounts of past studies [48]. All HCC that contain natural fibre has some significant reduction of density due to the high absorptive properties while the HCC that contains barchip fibre has higher density than those HCC with natural fibres. The barchip fibres have lower absorption properties and also lower aspect ratio; hence there is low tendency to impact any forceful cohesiveness of the mixtures. This makes it easy for the air voids to be eliminated easily in the course of mixing and vibration. The incorporation of both coconut fibres and Oil palm waste fruit bunch fibres resulted in the reduction of the density. This may be due to the differences in the specific gravity of the fibres as well as the differences in the absorption properties. This result confirms the findings of other past experimental studies [50]. When the result of the bulk density is compared based on the environmental vulnerability of the specimens, it will be noted that in most ages, the bulk density value of specimens exposed to sea water is more eminent than that of the specimens exposed to average water. This might be as a consequence of weight gain over ages due to the formation of crystal salts within the microstructures of the HCC specimens. The formation of gypsum and ettringite by the sulphate ions in seawater causes an expansion within the microstructure. This happening is more pronounced in the specimens that are taking in high absorption properties like specimens with coconut fibre and then accompanied by the specimens with Oil palm waste fruit bunch fibres. But amazingly, in this experiment, the specimens with barchip

2280 2260 Bulk density kn/m3

606

C 2240

BM

2220

CF OPFBF

2200

BF

2180

CF+BF OPFBF+BF

2160 7D-W

7-SW

28D-W 28D-SW 90D-W 90D-SW Period of Exposure

Fig. 8. The bulk density of all HCC samples.

607

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

4.2.3. Compressive strength The results of the compressive strengths of HCC mixtures with the incorporation of MK, CNS and epoxy as cement substitutes are presented in Fig. 9. It was observed that all the HCC mixtures have compressive strength that is higher than the control mixture. All the HCC mixtures showed compressive strengths higher than 65 N/mm2 of ECC M45 and 38 N/mm2 of normal concrete at 28 days. This revealed that the HCC showed an increase in compressive strength. Likewise, the compressive strength of all mixtures improved by the age of concrete irrespective of the curing regime, however the rate of the compressive strength development was considered more significant in the first 28 days compared to the strength development after the 28 days. It was observed to be in the ranges of 1.97%–33.41% for 7 days and 28 days respectively above the control specimens for each age, while it was observed to be between the ranges of 18.29%–26.17% for 90 days, 180 days and 365 days. This could be due to the pozzolanic reaction of MK which is higher at the early age compared to later ages. This findings is in agreement with the findings of past studies [51]. The higher compressive strength of the HCC with the incorporation of MK can be attributed to MK particles pores filling effects, the acceleration of cement hydration at the early ages and the pozzolanic reaction of MK with the calcium hydroxide. This assertion was corroborated in literatures [52,53]. In addition to the effect of MK, the CNS also contributed to the strength development in all ages, it react with calcium hydroxide, which is a product of cement hydration to form more C-S-H chain dimension and stiffness at final stages resulted in the significant improvements in the compressive strength at the early ages. This formation of calcium hydroxide through the hydration reaction of cement and the pozzolanic reaction was much higher and faster in CNS added cement than that of plain cement paste and pozzolanic materials like MK. The basic mechanism of this working principle is associated to the nanosilica (CNS) high surface area because it was observed to work as nucleation sites for C-S-H gel precipitation [54]. A significant increase was recorded on the compressive strength of all the hybrid HCC mixtures at 28 days. All the specimen strength is higher than control specimen and it showed that there is a significant impact of MK, CNS and epoxy. The effect of the epoxy resin is to improve the surface bond of all the particles in the matrix with the fibre surfaces hence create a dense matrix. At the age of 28 days as reflected in Fig. 9, the BF specimen has the highest compressive strength with 33.41% above the control specimen and this indicates about 29% above the strength of the same BF at 7 days. CF+BF specimen has a compressive strength of

Compressive strength (N/mm2)

90 80 70

C

60

BM

50

CF

40

OPFBF

30

BF

20

CF+BF

10

33.14% increase over the control specimen and about 28% above the strength gain at 7 days. The BM, CF, OPFBF and OPFBF+BF has 12.21%, 29.93%, 26.26% and 25.89%, respectively, as the increase in the compressive strength of 7 days ages. It was also observed that most of the specimens of HCC with a higher compressive strength were sea water cured. For instance, at the age of 28 days, the highest compressive strength recorded was the specimen BF cured in the sea water, likewise the CF+BF, OPFBF and OPFBF+BF, while BM and CF has the higher compressive strength recorded in samples cured in water. The higher compressive strength might be as a result of the salts crystallisation forming a thin layers over the surface of the samples and lack of degradation of fibres at this age. The compressive strength at 90 days age revealed that BF specimen maintains the highest compressive strength with 80.21 N/ mm2 while CF was having 79.73 N/mm2, which was cured in sea water. The CF+BF has 79.14 N/mm2, OPFBF has 78.95 N/mm2, OPFBF+BF has 78.86 N/mm2 and BM has 77.34 N/mm2. When these values are compared to the 28 days compressive strength percentage increase, it was observed that there is a decrease in the percentage. For instance, the percentage increase in the highest compressive strength which is by BF was 28.19% above the control which signify a percentage decrease. This can be said of all the specimens. In the findings, the percentage increase in strength of all samples decreases as the age grows. This may be due to the early pozzolanic actions of both the MK and the CNS incorporation in the mixtures. The HCC with the incorporation of BF maintained the highest compressive strength in all the remaining ages and higher than those of CF and OPFBF. The BF is a synthetic fibre while CF and OPFBF are natural fibres. This may be due to the established fact that the natural fibre cementitious composites are associated with its deterioration as a result of combined attack of alkali and fibre embrittlement. This arose from hydration product migration. In other word, the mechanism is well correlated to the delignification of the fibre and the wall impregnation by minerals such as magnesium, calcium, silica and aluminium [55]. Likewise, the performance of a cement based fibre reinforced composites of polyolefin based fibres performs very differently, in respect of their mechanical and elastic properties, the surface structure of the fibres is very effective on their performance. Polymer fibres are observed to be very versatile and their performance are quite different from each other [56]. 4.2.4. Flexural strength The Flexural strength results of the Hybrid HCC are highlighted in Fig. 10. The results showed that all specimens’ increases over the ages. The low fibre content reveals long term effectiveness in both curing regimes. Although the aging effects of fibres, especially, the natural fibre, OPFBF in particular, were more obvious in the sea

Flexural Strength (N/mm2)

fibres gain some weight at later ages; it may be ascribable to the continual process of hydration.

12 10 8

7D-W

6

7-SW

4

28D-W

2

28D-SW 90D-W

0

90D-SW

OPFBF+BF

0 7D-W

7D-SW 28D-W 28D-SW 90D-W 90D-SW Period of Exposure

Fig. 9. The compressive strength of HCC samples at all ages.

Period of Exposure(days) Fig. 10. The flexural strength of HCC at various ages.

water curing. At the early age of 7 days, all the specimens attained greater flexural strength than the control samples and the BM samples, an indication that the incorporation of fibre in the HCC increases the flexural strength. This can be traced to fibre impact by an advance in the toughness matrix of the HCC and as well as the compactness and homogeneity of fibre distribution in the HCC. This observation supports the statement made by Felekoglu et al. [56]. The BF HCC specimens have the highest flexural strength with 29.58% above that of control specimens and 26.88% above the BM. The specimen with the lowest value of flexural strength was OPFBF cured in water with a flexural value of 6.36 N/mm2, which is 3.92% above the control specimen value and 1.59% above the BM HCC specimen. At this age, it can be observed that the HCC specimen, BM and CF cured in water has values which are above the sea water cured while at all other ages, the HCC specimens cured in the sea water have higher flexural strength values. At age 28 days, the HCC with the incorporation of BF, cured in water causes the highest flexural strength value. It is 52.77% above the control specimen and 27.30% above the BM specimens. When this is compared to the age 7 days value of the same BF HCC specimen, it can be noted that the specimen attained 23.19% and 0.42% increase in flexural strength value achieved in 7 days. There is a significant improvement above the control and BM HCC specimens. At the same 28 days age, the HCC with the incorporation of CF has the second highest value of HCC, it was 9.97 N/mm2 which represents 40.57% above the control specimen and 18.40% above the BM specimen of HCC. Among all the specimens with fibre reinforcement, the HCC with OPFBF incorporation cured in water has the lowest Flexural value of 9.07 N/mm2 which represents 28.65% above the control specimen and 8.36% above the BM specimen of the Hybrid HCC. At age 90 days, the same phenomenon was discovered, only the increment above the 28 days value was not equally high as that of the 28 days ages against 7 days. The HCC specimen with the incorporation of BF still maintained the lead in the flexural strength value and this occurred throughout the ages. This may be ascribable to the high tensile ductility properties of the barchip fibre compared to the natural fibres which have lower tensile ductile properties. More often than not, it was thought that the flexural response of HCC reflects the tensile ductility [57]. Likewise, in addition to the barchip fibre mechanical and elastic properties, the surface structure is more impactive in their performance. Relative to experimental results obtained, it was noted that an empirical correlation exists between the compressive strength and flexural strength of HCC mixes. This suspected relationship between the compressive strength and flexural strength of HCC mixes is basically expressed as detailed in an equation, Eq. (1) as proposed by [58]. a

f cf ¼ kf cu

ð1Þ 2

where fcf is recognised as the flexural strength in N/mm , fcu is referred to as the compressive strength in N/mm2. The k and a are considered as the coefficients. By utilising the water and seawater curing specimens values of compressive strength and flexural strength of age at 28 days, the regression analysis was evaluated for the determination of the values for the coefficients k and a. The result is presented in Figs. 11 and 12. From Figs. 11 and 12, it follows that the relationship between the compressive strength and flexural strength in both saltwater and water cured for up till 28 days can be evinced in the following two equations;

F cf ¼ 0:1626f cu 2:747; R2 ¼ 0:8419 ðwater cured samplesÞ

ð2Þ

Flexural Strength (N/mm2)

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

12

fcf = 0.1626fcu- 2.747 R² = 0.8419

10 8 6 4 2 0 50

60

70

80

Compressive Strength (N/mm2)

Fig. 11. The correlation between compression and flexural strength of samples cured in water at 28 days.

12

Flexural strength (N/mm2)

608

fcf= 0.1385fcu - 0.99 R² = 0.8064

10 8 6 4 2 0 50

60

70

80

90

100

Compressive strength (N/mm2)

Fig. 12. The correlation between compression and flexural strength of samples cured in seawater at 28 days.

F cf ¼ 0:1385f cu 0:99; R2¼ 0:8064 ðseawater cured samplesÞ

ð3Þ

The regression analysis implies that the HCC specimens compressive strength and the flexural strength have a significant relationship. Meanwhile, the regression value in water cured is higher than that of seawater cured indicating that the relationship is higher. The regression, R2 value for water cured system is 0.8419 while that sea water is 0.8064. 4.2.5. Ultra Pulse Velocity (UPV) In this experimental study, UPV was used to monitor the behaviour of the Hybrid HCC specimen subjected to sea water and water curing regime over a period of ages up to 90 days. The ultrasonic pulse velocity result is shown in Fig. 13. The results of the UPV in this experimental study reveal that there was an increment trend in UPV results of all specimens and this is similar to the trend observed in the compressive strength and the dynamic modulus. The observed results show that all the specimens fall within the excellent and good class of the UPV values classification [59]. At the early age of 7 days, all the specimens have UPV value which is within the ranges of 4.01–4.36 km/s of water cured and 4.02–4.38 km/s of sea water cured specimens. This signifies that the incorporation of MK, CNS and Epoxy resin combined, have a significant effect on the matrix and it enhances the microstructure of the mix hence the difference in the control mix and the base mix. The percentage difference between the BM that comprises MK, CNS and Epoxy without any fibres are 5.86% and 5.07% above the water and sea water cured specimens of the control specimen respectively. This also suggested that there was a reduction of porosity in the mix due to the pozzolanic nature of both MK and the CNS coupled with the bonding properties of

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

Ultra Pulse Velocity (Km/s)

5 4.5 4

C

3.5

BM

3

CF

2.5

OPFBF

2 1.5

BF

1

CF+BF

0.5

OPFBF+BF

0 7D-W

7-SW

28D-W 28D-SW 90D-W 90D-SW

Period of Exposure (days) Fig. 13. The ultra velocity pulse of HCC at all ages.

epoxy resin, a polymer structure. The increase observed in the specimen may be as a result of the nature and properties of natural fibres which benefited the microstructure control of the low volume fibre as they were assumed to resist the initiation and propagation of cracks in the matrix [60]. The higher values of UPV displayed by specimen cured in the sea water might be as a result of crystallized salt within the structure to fill the empty spaces like the fibre/matrix interfacial zones. At age of 28 days, the BF HCC specimen cured in sea water has the highest value of UPV, it has 4.78 km/s above all specimens cured in both water and sea water. This represents 8.88% above the control cured in sea water and 7.17% above BM mixes cured in sea water. Meanwhile, the lowest value of UPV of this age was recorded by CF HCC, it has 4.40 km/s, about 8.6% lower than the highest recorded by BF. But at the age of 90 days onward, the hybrid fibres HCC, CF+BF and was having higher UPV above others especially those specimens cured inside sea water. Consistent increase was observed in the UPV of all specimens as the age grows. These results are in agreement with the result of the experiment conducted earlier by other researchers [53]. 4.2.6. Shrinkage test The result, as presented in Fig. 14 displayed clearly the significant effect of MK, CNS and Epoxy apart from the influence of the different fibres. At early age of 1 day to 7 days, it was obvious that the Control mix was having a high percentage increase of shrinkage compared with the BM mix that comprises of MK, CNS and epoxy. This may be due to the reduction in the rate of evaporation of water from the mix as a result of the refined pore structure and lower permeability of the mixes caused by formation of larger amounts of secondary C-S-H mineral resulted from the early pozzolanic reactions of both MK and CNS and the bonding features of Epoxy resins. When the hydration process proceeds, the products’ volume becomes smaller than that of the reactants and 0.14

Drying Shrinkage (%)

0.12 0.1

C BM

0.08

CF 0.06

OPFBF

0.04

BF CF+BF

0.02

OPFBF+BF

0 1d

2d

3d

4d

5d

6d

7d

28d

90d

Period of Exposure (days) Fig. 14. Drying shrinkage of all HCC samples at all ages.

609

excepts there is a water supplied from an outside source, the variance in volume will bring around the formation of empty pores within the microstructure of the cement paste. The empty porosity leads to a decrease in the relative humidity of the paste and a measurable autogenous shrinkage of the material [61]. Likewise, the microstructure of CNS concrete or mortar is more uniform and more compact than normal concrete [62]. The difference in the shrinkage values of all other HCC mixes compared to BM at day 1 to day 7 signify that the incorporation of fibres caused a certain degree of reduction in the shrinkage. From 1st day till 7th day, BM has between 25.81% and 67.72% reduction in shrinkage compared to that of the Control specimen. Meanwhile, at this age 1 day to 7 days, CF mix has the lowest reduction percentage compared to other BM mixes. It reduction percentage is within the ranges of 3.19%–7.83%. This may be due to the high absorptive property of Coconut fibre which does not support shrinkage reduction. BF mix has a better reduction value than OPFBF at the early ages. Although, these effects were partially overcome by the addition of Epoxy resin. Broadly speaking, in all ages up till the 90 days, the effect of fibre was significant, although, a the percentage decrease of shrinkage reduces as the specimen grows in age. The outcome of this experimental test complies with other past studies [63]. 5. Conclusions The following conclusion can be deduced from the laboratory experiment; 1. The preparation of MK in the laboratory using laboratory muffle furnace instead of the kiln, at the 800 °C temperature for 1 h can produce a highly reactivity metakaolin. 2. The HCC with the incorporation of 10% MK, 1% CNS and 1% epoxy resin of the total weight of the binder reduces the slump value of the mix compare to the control by 3.25%. Meanwhile, the incorporation of natural fibres causes further reduction compare to the control specimen with about 7.14%. 3. The density of all HCC samples in this experiment with the incorporation of 10% MK, 1% CNS and 1% epoxy resin of the total weight of the binder increases as the ages increase. 4. The HCC with the incorporation of 10% MK, 1% CNS and 1% epoxy resin of the total weight of the binder showed a high compressive strength at the early age of the mix. It is higher than 65 N/mm2 of M45 ECC and 38 N/mm2 of normal concrete at 28 days. At the same time, the compressive strength of all mixtures improved by the age of concrete irrespective of the curing regime. The strength increase above that of control at 28 days was 33.41%. 5. The incorporation of both natural and synthetic fibres at 2% volume of the weight of the binder in HCC mixes with the incorporation of 10% MK, 1% CNS and 1% epoxy causes the flexural strength to increase at all ages. The barchip reinforced HCC mixes have the highest flexural strength with 29.58% above that of control and 26.88% above the base mix. 6. The relationship between the compressive strength and flexural strength of the HCC with the incorporation of 10% MK, 1% CNS and 1% epoxy showed a high value of R2. An indication that the results of the two tests have a significant value to each other, the R2 values are 0.8419 and 0.8064 for water and sea water cured specimen respectively. 7. At 28 days, the HCC mixes that contain barchip fibre with the incorporation of 10% MK, 1% CNS and 1% epoxy and cured in sea water have the highest value of UPV, it was 8.88% above the control mixes cured in sea water and 7.17% above the base mix cured in sea water.

610

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611

8. The incorporation of 10% MK, 1% CNS and 1% epoxy in HCC shows significant decrease in drying shrinkage. In the first 7 days, the base mix, BM has between 25.81% and 67.72% reduction of drying shrinkage when compared to the control specimen. The HCC with coconut fibre, CF, mix has the lowest reduction percentage compared to base mix, BM mixes. The reduction was in the range of 3.19%–7.83%. 9. The optimum temperature and hour for the calcinations of MK is 800 °C and 1 h.

Acknowledgements The authors acknowledge the Universiti Sains Malaysia research scheme RU-PRGS No-1001/PPBGN/846112 and RUI grant No. 1001/PPBGN/814212. References [1] B. Mobasher, A. Peled, J. Pahilajani, Distributed cracking and stiffness degradation in fabric-cement composites, Mater. Struct. 39 (3) (2006) 317– 331. [2] A. Namman, H. Reinhardt, Setting the stage, toward performance based classification of FRC composites, High Performance Fiber Reinforced Cement Composties (HPFRCC 4), in: Proc. of the 4th Int. RILEM Workshop, 2003. [3] W. Zhongwei, Green high performance concrete-the trend of concrete development, Chinal Concr. Cem. Prod. (1998) 1. [4] B. Chen, J. Liu, Experimental application of mineral admixtures in lightweight concrete with high strength and workability, Constr. Build. Mater. 22 (6) (2008) 1108–1113. [5] Y. Zhu, Y. Yang, Y. Yao, Use of slag to improve mechanical properties of engineered cementitious composites (ECCs) with high volumes of fly ash, Constr. Build. Mater. 36 (2012) 1076–1081. [6] V.C. Li, Engineered Cementitious Composites (ECC) Material, Structural, and Durability Performance, 2008. [7] M. S ß ahmaran, V.C. Li, Durability properties of micro-cracked ECC containing high volumes fly ash, Cem. Concr. Res. 39 (11) (2009) 1033–1043. [8] J. Zhang et al., Engineered cementitious composite with characteristic of low drying shrinkage, Cem. Concr. Res. 39 (4) (2009) 303–312. [9] V.C. Li, D.K. Mishra, H.-C. Wu, Matrix design for pseudo-strain-hardening fibre reinforced cementitious composites, Mater. Struct. 28 (10) (1995) 586–595. [10] H. Van Oss, A.C. Padovani, Cement manufacture and the environment, J. Ind. Ecol. 6 (1) (2002) 89–106. [11] Y. Zhu et al., Mechanical properties of engineered cementitious composites with high volume fly ash, J. Wuhan Univ. Technol. 24 (89) (2009) 166–170. [12] Y. Zhu, Y.Z. Yang, Y. Yao, Effect of High Volumes of Fly Ash on Flowability and Drying Shrinkage of Engineered Cementitious Composites, in: Materials Science Forum, Trans Tech Publ., 2011. [13] S. Mindess, J.F. Young, D. Darwin, Concrete, 2003. [14] P. Taylor, Integrated Materials and Construction Practices for Concrete Pavement, Federal Highway Administration Office of Pavement Technology, 2007. [15] J.T. Ding, Z.J. Li, Effects of metakaolin and silica fume on properties of concrete, ACI Mater. J. 99 (4) (2002) 393–398. [16] S. Wild, J. Khatib, A. Jones, Relative strength, pozzolanic activity and cement hydration in superplasticised metakaolin concrete, Cem. Concr. Res. 26 (10) (1996) 1537–1544. [17] J. Kinuthia et al., Self-compensating autogenous shrinkage in Portland cement—metakaolin—fly ash pastes, Adv. Cem. Res. 12 (1) (2000) 35–43. [18] P. Whitfield, L. Mitchell, Phase Identification and Quantitative Methods, Principles and Applications of Powder Diffraction, 2008, pp. 226–260. [19] ASTM C311, Test Method for Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland Cement Concrete, vol. 04.01, Annual Book of ASTM Standard, 2004. [20] ASTM C188, Test Method for Density of Hydraulic Cement, vol. 04.01, Annual Book of ASTM Standards, 1995. [21] ASTM C305, Practice for Mechanical Mixing of Hydraulic Cement Pastes and Mortars of Plastic Consistency, vol. 04.01, Annual Book of ASTM Standards, 1994. [22] BS 1881: Part 102, Testing Concrete. Method for Determination of Slump, British Standards Institution, London, 1983. [23] BS 1881: Part 122, Testing Concrete. Methods for Determination of Density of Hardened Concrete, British Standards Institution, London, 1983. [24] ASTM C109, Test Method for Compressive Strength of Hydraulic Cement Mortars (Using 2-in. Or [50-mm] Cube Specimens), vol. 04.01, Annual Book of ASTM Standards, 1998. [25] ASTM C348, Test Method for Flexural Strength of Hydraulic Mortar, vol. 04.01, Annual Book of ASTM Standards, 1997.

[26] BS EN 12504. Part 4. Testing Concrete. Determination of Ultrasonic Pulse Velocity, British Standards Institution, London, 1983. [27] American Society of Testing Materials, ASTM C157:2004. Standard Test Method for Length Change of Hardened Hydraulic Cement Mortar and Concrete. Annual Book of ASTM Standards, Vol. 04.02, ASTM, West Conshohocken, PA, USA, 2004. [28] American Society of Testing Materials, ASTM C596:2009. Test Method for Drying Shrinkage of Mortar Containing Portland Cement. Annual Book of ASTM Standards, vol. 04.01, ASTM, West Conshohocken, PA, USA, 2009. [29] American Society of Testing Materials, ASTM C490:2011. Standard Practice for Use of Apparatus for the Determination of Length Change of Hardened Cement Paste, Mortar, and Concrete. Annual Book of ASTM Standards, vol. 04.01, ASTM, West Conshohocken, PA, USA, 2011. [30] J. Ambroise, M. Murat, J. Pera, Investigations on synthetic binders obtained by middle-temperature thermal dissociation of clay minerals, Silic. Indus. 51 (7– 8) (1986) 99–107. [31] A.M. Rashad, Metakaolin as cementitious material: history, scours, production and composition–A comprehensive overview, Constr. Build. Mater. 41 (2013) 303–318. [32] J. Ambroise, S. Maximilien, J. Pera, Properties of metakaolin blended cements, Adv. Cem. Based Mater. 1 (4) (1994) 161–168. [33] W. Sha, G. Pereira, Differential scanning calorimetry study of ordinary Portland cement paste containing metakaolin and theoretical approach of metakaolin activity, Cement Concr. Compos. 23 (6) (2001) 455–461. [34] C. Li, H. Sun, L. Li, A review: the comparison between alkali-activated slag (Si+ Ca) and metakaolin (Si+ Al) cements, Cem. Concr. Res. 40 (9) (2010) 1341– 1349. [35] E.G. Badogiannis et al., Durability of metakaolin self-compacting concrete, Constr. Build. Mater. 82 (2015) 133–141. [36] E. Moulin, P. Blanc, D. Sorrentino, Influence of key cement chemical parameters on the properties of metakaolin blended cements, Cement Concr. Compos. 23 (6) (2001) 463–469. [37] R. Siddique, J. Klaus, Influence of metakaolin on the properties of mortar and concrete: a review, Appl. Clay Sci. 43 (3) (2009) 392–400. [38] H.E.-D.H. Seleem, A.M. Rashad, B.A. El-Sabbagh, Durability and strength evaluation of high-performance concrete in marine structures, Constr. Build. Mater. 24 (6) (2010) 878–884. [39] M. Hajjaji, S. Kacim, M. Boulmane, Mineralogy and firing characteristics of a clay from the valley of Ourika (Morocco), Appl. Clay Sci. 21 (3) (2002) 203– 212. [40] K.A.A. Hadi, The genesis and characteristics of primary kaolinitic clay occurrence at Bukit Lampas, Simpang Pulai, Ipoh, Bull. Geol. Soc. Malaysia 54 (2008) 9–16. [41] B.R. Ilic´, A.A. Mitrovic´, L.R. Milicˇic´, Thermal treatment of kaolin clay to obtain metakaolin, Hemijska Industrija 64 (4) (2010) 351–356. [42] ASTM C311, Test Method for Sampling and Testing Fly Ash or Natural Pozzolans for Use in Portland Cement Concrete, vol. 04.02, Annual Book of ASTM Standards, 2004. [43] A.L. Velosa, F. Rocha, R. Veiga, Influence of chemical and mineralogical composition of metakaolin on mortar characteristics, Acta Geodyn. Geomater. 6 (1) (2009) 153. [44] J. Han, Z. Shui, G. Wang, Research on the reactivity of Metakaolin with different grade, Sustain. Constr. Mater. (2012) 173–180. [45] L. Singh et al., Beneficial role of nanosilica in cement based materials–A review, Constr. Build. Mater. 47 (2013) 1069–1077. [46] M. Ramli, K.W. Hoe, Influences of short discrete fibers in high strength concrete with very coarse sand, Am. J. Appl. Sci. 7 (12) (2010) 1572. [47] F. Kontoleontos et al., Influence of colloidal nanosilica on ultrafine cement hydration: physicochemical and microstructural characterization, Constr. Build. Mater. 35 (2012) 347–360. [48] P. Dinakar, P.K. Sahoo, G. Sriram, Effect of metakaolin content on the properties of high strength concrete, Int. J. Concr. Struct. Mater. 7 (3) (2013) 215–223. [49] M. Sßahmaran et al., Improving the workability and rheological properties of Engineered Cementitious Composites using factorial experimental design, Compos. B Eng. 45 (1) (2013) 356–368. [50] O. Karahan, C.D. Atisß, The durability properties of polypropylene fiber reinforced fly ash concrete, Mater. Des. 32 (2) (2011) 1044–1049. [51] J. Khatib, J. Hibbert, Selected engineering properties of concrete incorporating slag and metakaolin, Constr. Build. Mater. 19 (6) (2005) 460–472. [52] M. Said-Mansour et al., Influence of calcined kaolin on mortar properties, Constr. Build. Mater. 25 (5) (2011) 2275–2282. [53] R. Madandoust, S.Y. Mousavi, Fresh and hardened properties of selfcompacting concrete containing metakaolin, Constr. Build. Mater. 35 (2012) 752–760. [54] L. Zapata et al., Rheological performance and compressive strength of superplasticized cementitious mixtures with micro/nano-SiO2 additions, Constr. Build. Mater. 41 (2013) 708–716. [55] R.D. Toledo Filho et al., Durability of alkali-sensitive sisal and coconut fibres in cement mortar composites, Cement Concr. Compos. 22 (2) (2000) 127–143. [56] B. Felekog˘lu, K. Tosun, B. Baradan, Effects of fibre type and matrix structure on the mechanical performance of self-compacting micro-concrete composites, Cem. Concr. Res. 39 (11) (2009) 1023–1032. [57] M. Kunieda, K. Rokugo, Recent progress on HPFRCC in Japan required performance and applications, J. Adv. Concr. Technol. 4 (1) (2006) 19–33.

M.B. Ramli, O.R. Alonge / Construction and Building Materials 121 (2016) 599–611 [58] M. Ramli, Durability Properties of Polymer Modified Cement System, University of Sheffield, 1997. [59] R. Solis-Carcaño, E.I. Moreno, Evaluation of concrete made with crushed limestone aggregate based on ultrasonic pulse velocity, Constr. Build. Mater. 22 (6) (2008) 1225–1231. [60] F. de Andrade Silva, B. Mobasher, R.D. Toledo Filho, Cracking mechanisms in durable sisal fiber reinforced cement composites, Cement Concr. Compos. 31 (10) (2009) 721–730.

611

[61] D.P. Bentz et al., Effects of cement particle size distribution on performance properties of Portland cement-based materials, Cem. Concr. Res. 29 (10) (1999) 1663–1671. [62] G. Quercia, H. Brouwers, Application of nano-silica (nS) in concrete mixtures, 8th fib PhD symposium in Kgs. Lyngby, Denmark, 2010. [63] E. Güneyisi et al., Strength, permeability and shrinkage cracking of silica fume and metakaolin concretes, Constr. Build. Mater. 34 (2012) 120–130.