Upload
sanjithrnair
View
27
Download
1
Embed Size (px)
DESCRIPTION
geopolymer concrete element journal
Citation preview
7/17/2019 1-s2.0-S0950061815002834-main
1/13
Review
Geopolymer concrete: A review of some recent developments
B. Singh , Ishwarya G., M. Gupta, S.K. Bhattacharyya
CSIR-Central Building Research Institute, Roorkee 247667, India
h i g h l i g h t s
An overview of geopolymer ispresented alongwith its processing
parameters.The hardened properties and
durability of geopolymer concrete are
discussed.
The design guidelines for OPCconcrete are applicable to
geopolymer concrete also.
Geopolymeric building productsdeveloped at CSIR-CBRI are
highlighted.
Ambient cured single componentgeopolymer may enhance its wider
use in the field.
g r a p h i c a l a b s t r a c t
Conversion of fly ash into geopolymers/concrete.
a r t i c l e i n f o
Article history:
Received 26 November 2013
Received in revised form 16 February 2015
Accepted 4 March 2015
Available online 31 March 2015
Keywords:
Geopolymer concrete
Activator
Bond strength
Compressive strength
Durability
a b s t r a c t
An overview of advances in geopolymers formed by the alkaline activation of aluminosilicates is pre-
sented alongwith opportunities for their use in building construction. The properties of mortars/concrete
made from geopolymeric binders are discussed with respect to fresh and hardened states, interfacial
transition zone between aggregate and geopolymer, bond with steel reinforcing bars and resistance to
elevated temperature. The durability of geopolymer pastes and concrete is highlighted in terms of their
deterioration in various aggressive environments. R&D works carried out on heat and ambient cured
geopolymers at CSIR-CBRI are briefly outlined alongwith the product developments. Research findings
revealed that geopolymer concrete exhibited comparative properties to that of OPC concrete which
has potential to be used in civil engineering applications.
2015 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2. An overview of geopolymers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.1. Constituents effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.2. C-S-H phase effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.3. Effect of admixtures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.4. Curing conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3. Geopolymer mortars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4. Geopolymer concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
http://dx.doi.org/10.1016/j.conbuildmat.2015.03.036
0950-0618/ 2015 Elsevier Ltd. All rights reserved.
Corresponding author.
E-mail address:[email protected](B. Singh).
Construction and Building Materials 85 (2015) 7890
Contents lists available at ScienceDirect
Construction and Building Materials
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o n b u i l d m a t
http://dx.doi.org/10.1016/j.conbuildmat.2015.03.036mailto:[email protected]://dx.doi.org/10.1016/j.conbuildmat.2015.03.036http://www.sciencedirect.com/science/journal/09500618http://www.elsevier.com/locate/conbuildmathttp://www.elsevier.com/locate/conbuildmathttp://www.sciencedirect.com/science/journal/09500618http://dx.doi.org/10.1016/j.conbuildmat.2015.03.036mailto:[email protected]://dx.doi.org/10.1016/j.conbuildmat.2015.03.036http://crossmark.crossref.org/dialog/?doi=10.1016/j.conbuildmat.2015.03.036&domain=pdf7/17/2019 1-s2.0-S0950061815002834-main
2/13
4.1. Fresh and hardened properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.2. Interfacial transition zone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.3. Bond between reinforcing bars and geopolymer concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.4. Fire behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5. Durability studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.1. Alkali-silica reaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.2. Effect of acid attack. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.3. Effect of sulphate attack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.4. Carbonation and permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 865.5. Corrosion of steel reinforcement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
6. Research and development at CSIR-CBRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
7. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
1. Introduction
The concrete industry faces challenges to meet the growing
demand of Portland cement due to limited reserves of limestone,
slow manufacturing growth and increasing carbon taxes. It is
reported that the requirement of cement in India is likely to touch550 million tonnes by 2020 with a shortfall of230 million ton-nes (58%) and the demand for cement has been constantlyincreasing due to increased infra-structural activities of the coun-
try[1]. One effort to combat shortfall is the development of alter-
nate binders to Portland cement aiming at to reduce the
environmental impact of construction, use of greater proportion
of waste pozzolan, and also to improve concrete performance.
Search for several alternatives such as alkali-activated cement, cal-
cium sulphoaluminate cement, magnesium oxy carbonate cement
(carbon negative cement), supersulphated cement etc. are being
made with the advantages of Portland cement [2]. As the family
of the alkali-activated cement is growing, the alkaline cement is
classified based on a phase composition of the hydration products:
R-A-S-H (R = Na+ or K+) in the aluminosilicate based systems and R-
C-A-S-H in the alkali-activated slag or alkaline Portland cements
[3]. In recent years, geopolymer has attracted considerable atten-
tion among these binders because of its early compressive
strength, low permeability, good chemical resistance and excellent
fire resistance behaviour [49]. Because of these advantageous
properties, the geopolymer is a promising candidate as an alterna-
tive to ordinary Portland cement for developing various sustain-
able products in making building materials, concrete, fire
resistant coatings, fibre reinforced composites and waste
immobilization solutions for the chemical and nuclear industries.
2. An overview of geopolymers
Geopolymer is considered as the third generation cement afterlime and ordinary Portland cement. The term geopolymer is
generically used to describe a amorphous alkali aluminosilicate
which is also commonly used for to as inorganic polymers,
alkali-activated cements, geocements, alkali-bonded cera-
mics, hydroceramics etc. Despite this variety of nomenclature,
these terms all describe materials synthesized utilising the same
chemistry[4]. It essentially consists of a repeating unit of sialate
monomer (SiOAlO). A variety of aluminosilicate materials
such as kaolinite, feldspar and industrial solid residues such as
fly ash, metallurgical slag, mining wastes etc. have been used as
solid raw materials in the geopolymerization technology. The
reactivity of these aluminosilicate sources depends on their chemi-
cal make-up, mineralogical composition, morphology, fineness and
glassy phase content. The main criteria for developing stablegeopolymer are that the source materials should be highly
amorphous and possess sufficient reactive glassy content, low
water demand and be able to release aluminium easily. The alka-
line activators such as sodium hydroxide (NaOH), potassium
hydroxide (KOH), sodium silicate (Na2SiO3) and potassium silicate
(K2SiO3) are used to activate aluminosilicate materials. Compared
to NaOH, KOH showed a greater level of alkalinity. But in reality,it has been found that NaOH possesses greater capacity to liberate
silicate and aluminate monomers [4]. The properties of geopoly-
mers can be optimised by proper selection of raw materials, correct
mix and processing design to suit a particular application [4].
Viewing the importance of the subject, a collaborative project
sponsored by the European Commission BRITE was undertaken
jointly by France, Spain and Italy on development of Cost-effective
geopolymeric cement for innocuous stabilization of toxic elements
(GEOCISTEM). The project was aimed at manufacturing geopoly-
meric cement by replacing potassium silicate with cheaper alkaline
volcanic tuffs[9].
Geopolymers are synthesized by the reaction of a solid
aluminosilicate powder with alkali hydroxide/alkali silicate [8]. A
schematic representation on formation of fly ash-based geopoly-
mers/concrete is shown inFig. 1. Under highly alkaline conditions,
polymerisation takes place when reactive aluminosilicates are
rapidly dissolved and free [SiO4] and [AlO4]
tetrahedral unitsare released in solution. The tetrahedral units are alternatively
linked to polymeric precursor by sharing oxygen atom, thus form-
ing polymeric SiOAlO bonds. The following reactions occur dur-
ing geopolymerisation[7].
Si2O5Al2O2n H2O OH ! SiOH
4 AlOH
4 1
2
This process releases water that is normally consumed during
dissolution. The water, expelled from geopolymer during the reac-tion provides workability to the mixture during handling. This is in
contrast to the chemical reaction of water in Portland cement mix-
ture during the hydration process. It is reported that the hydration
products of metakaolin/fly ash activation are zeolite type: sodium
aluminosilicate hydrate gels with different Si/Al ratio whereas the
major phase produced in slag activation is calcium silicate hydrate
with a low Ca/Si ratio. Though many physical properties of
geopolymers prepared from various aluminosilicate sources may
appear to be similar, their microstructures and chemical properties
vary to a large extent. The metakaolin-based geopolymer has an
advantage that it can be manufactured consistently, with pre-
dictable properties both during the preparation and development.
However, its plate-shaped particles lead to rheological problems,
increasing the complexity of processing as well as the waterdemand of the system [6]. Contrary to this, the fly ash-based
B. Singh et al. / Construction and Building Materials 85 (2015) 7890 79
http://-/?-http://-/?-7/17/2019 1-s2.0-S0950061815002834-main
3/13
geopolymer is generally more durable and stronger than that of
metakaolin-based geopolymer [4]. The slag-based geopolymer is
considered to have high early strength and greater acid resistance
than those of metakaolin and fly ash-based systems.
2.1. Constituents effect
The most important factors affecting the properties of geopoly-
mer pastes are: SiO2/Al2O3 ratio, R2O/Al2O3 ratio, SiO2/R2O ratio
(R = Na+ or K+) and liquidsolid ratio. The majority of research con-
cluded that an amorphous structure of geopolymers is preferable
in order to achieve desired mechanical strength [1015]. Therelationship between the compressive strength and SiO2/R2O ratio
showed that an increase in alkali content or decrease in silicate
content increases the compressive strength of geopolymers attri-
butable to the formation of aluminosilicate network structures
[10,11]. Geopolymer activated with NaOH alone with Si/Na of 4/4
or less formed the crystalline zeolite (Na96Al96Sr96O384216H2O)
but at a ratio >4/4, nanosized crystals of another zeolite
(Na6[AlSiO4]64H2O) were formed[12]. The addition of even smallamount of sodium silicate to the NaOH significantly reduces crys-
tallite formation due to templating function of silicate units. At low
activator dosage (18%), the pores developed in the fly ash-based
paste were larger and exhibited wider distributions (19.8
2342 A) whereas at higher activator dosage (30%), the pores were
smaller and showed a narrow distribution (19.81155 A) mainlydue to the pore refinement as a result of more dissolution of
particles and formation of reaction products (Fig. 2). The reduced
porosity enhanced the strength of geopolymer pastes [13].
Typically, the optimum geopolymer strength was reported with
SiO2/Al2O3 ratio in the range of 3.03.8 and Na2O/Al2O3 ratio of1 [14,15]. Changes in SiO2/Al2O3 ratio beyond this range havebeen found to result in low strength. The setting time of geopoly-
mer pastes increased with increasing SiO2/Al2O3ratio of the initial
mixture.
2.2. C-S-H phase effect
The effect of C-S-H phase on the geopolymerization of
aluminosilicates has been studied with a view to know its role in
early age strength [1622]. In metakaoin/slag blends, both C-S-H
phase and aluminosilicate gel (N-A-S-H) co-exist in the paste
[16] as similar to NaOH activated high calcium fly ash-based
geopolymer[17] which are responsible for the strength increase.
The little dissolution of calcium occurs in the case of adding naturalcalcium silicate minerals at lower alkalinity, resulting in less C-S-H
gel formation and subsequent strength reduction of geopolymer
pastes [18]. In the case of fly ash/slag blends, the reaction at
27 C is dominated by the slag activation, whereas the reaction at
60 C is due to combined activation of fly ash and slag. The
improvement in compressive strength of pastes with slag addition
is attributed to its compactness of the microstructure [19]. The
initiation of hardening in fly ash/slag geopolymer made with
potassium silicate and potassium hydroxide was due to C-S-H/C-
A-S-H formation and the hardening continues due to a rapid for-
mation of a C-A-S-H, K-A-S-H and (Ca, K)-A-S-H depending on
the availability of calcium ions and pH of the system. A slower dis-
solution rate of calcium ions effectively increased the compressive
strength as rapid geopolymerization continues for a longer dura-tion[20]. The low pH and limited calcium ion environment facili-
tate the polymerisation reaction between silicate and aluminate
species in high calcium fly ash-based geopolymers producing N-
A-S-H gel [21]. Guo et al. [22] reported 63.4 MPa compressive
strength of class C fly ash-based geopolymer paste showing the
role of calcium participation in the strength development.
2.3. Effect of admixtures
Kusbiantora et al.[23]reported from their studies that admix-
tures such as sucrose and citric acid which act as retarder in OPC
have different mechanism in fly ash-based geopolymers. Sucrose
acted as a retarder since it is absorbed by Ca, Al and Fe ions to form
insoluble metal complexes. On the other hand, citric acid acted asan accelerator reducing the setting time by 9 and 16 min
Fig. 1. Conversion of fly ash into geopolymers/concrete.
Fig. 2. Pore size distribution of fly ash-based geopolymer pastes at different
activator dosages[13].
80 B. Singh et al. / Construction and Building Materials 85 (2015) 7890
7/17/2019 1-s2.0-S0950061815002834-main
4/13
respectively. Amongst the commercial superplasticizers, the naph-
thalene based superplasticizer was effective when single activator
was used rendering 136% increase in relative slump without any
decrease in compressive strength. Modified polycarboxylate based
superplasticizer was efficient one when multi-compound activator
was used with a decrease in compressive strength of 29% [24].
However, retarding effect of polycarboxylate based super-
plasticizer was also reported in fly ash/slag blended system thoughthe improvement in workability was significant compared to naph-
thalene based superplasticizer[25].
2.4. Curing conditions
Several attempts[2631]have been made to study the effect of
different curing conditions on the properties of geopolymer pastes.
The curing temperatures were reported in the range between 40 C
and 85 C for complete geoplymerisation reactions. Palomo et al.
[26] studied curing of alkali activated fly ash (0.25 and 0.30 liq-
uid/solid ratio) at 65 C and 85C. They indicated that the com-
pressive strength of geopolymers (812 M) cured at 85C for
24 h was much higher than those cured at 65C. The rise of
strength was much smaller when curing time was extended after24 h. Perera et al. [27] studied the curing of metakaolin-based
geopolymers under ambient (2123 C) and heat conditions (40
60 C) with a controlled relative humidity (RH) for 24 h and found
that curing at 30% RH was preferable to that at 70% RH. Heah et al.
[28] concluded that the curing of metakaolin-based geopolymers
at ambient temperature was not feasible while increase in tem-
perature (40 C, 60 C, 80 C, 100C) favored the strength gain after
13 days. However, curing at higher temperature for a longer per-
iod of time caused failure of samples at a later age due to the ther-
molysis of SiOAlO bond. Rovnanik[29]reported that curing
of metakaolin based geopolymer at elevated temperature (40
80 C) accelerated the strength development but in 28 days, the
mechanical properties deteriorated in comparison with results
obtained for an ambient or slightly decreased temperature.Ebrahim and Ali [30] prepared three mixes with different for-
mulations and cured hydrothermally at different temperatures
(45, 65, 85C) and time (520 h) after 1 and 7 days of procuring.
Longer procuring at room temperature, before the application of
heat is beneficial for higher strength development. In general, ade-
quate curing of geopolymeric materials is required to achieve opti-
mal mechanical and durability performance to maintain their
structural integrity[31].
3. Geopolymer mortars
Various studies[3240]were conducted on flow and mechani-
cal properties of geopolymer mortars because of their more rele-
vant applications in building construction. The properties ofmortars were optimised with respect to initial flow, aggregate-bin-
der ratio, activator-binder ratio and activator molarity.
Chindaprasirt et al. [32] reported that the compressive strength
of class C fly ash-based geopolymer mortar was 52 MPa when
cured at 70 C for 3 days using sand-fly ash ratio of 2.75 at work-
able flow of 135 5%. Prolonged curing at high temperatures led
to the reduction in the compressive strength because of weakening
of microstructure and increased porosity due to the loss of mois-
ture. In another attempt [33], they produced geopolymer mortar
with a compressive strength of 86 MPa at 28 days with the help
of air classified class C fly ash (4500 cm 2/g fineness) activated with
sodium silicate and NaOH (10 M) at 1:1 mass ratio. The dimen-
sional change in terms of drying shrinkage (161 106 mm/
mm) was insignificant when compared with the Portland cementmortar (700850 106 mm/mm). The geopolymer mortars
(14 M activator solution) with 1030 wt% aggregate exhibited an
acceptable flowability, while the mortars containing 40 & 50 wt%
aggregate were stiff and difficult to pack in the mould. Increasing
aggregate content in the mortar mixes leads to insufficient activa-
tor for complete geopolymerization of fly ash/slag. The activator
may also be utilised for wetting of aggregate leaving less availabil-
ity for dissolution of these fly ash or slag particles. The compressive
strength of geopolymer mortars with high level of aggregate can beachieved by optimising the amount of activator dosage [34].
Khandelwal et al.[35]summarised that the compressive strength,
modulus of elasticity and Poissons ratio of fly ash-based geopoly-
mer mortars increased logarithmically with the increase of strain
rate. These engineering properties of geopolymer mortars com-
pared favourably with those predicted by Standards/Codes for con-
crete mixtures. When bottom ash was used, the geopolymer
mortars exhibited a low compressive strength (20 MPa). With10% replacement of sand by bottom ash, the mix exhibited a com-
parable compressive strength to those made with sand only. The
increase in strength (50100%) of bottom ash mortar was also
reported when the specimens were exposed at 800C probably
due to activation of bottom ash [36]. When lignite bottom ash
was ground to a mean particle size of 15.7 lm (3% retained on
sieve No. 325), the compressive strength of mortars activated with
sodium hydroxide/sodium silicate was 2458 MPa[37].
Brough and Atkinson[38]prepared geopolymer mortars using
slag, sand and activator in a ratio of 1:2.33:0.5. At water-to-total
solid ratio of 0.42, the mortar gained strength of40 MPa. Thesodium silicate activated mortars exhibited higher compressive
strength with low levels of porosity at the interface while KOH
activated mortars were highly porous in the interfacial zone giving
low compressive strength values. Yang et al. [39,40]found that the
flow of alkali-activated mortars increased with the increase of
water-binder ratio and decrease of aggregate-binder ratio. When
the aggregate-binder ratio was larger than 2.5, the flow of mortars
decreased sharply. They also found that slag-based geopolymer
mortars exhibited much higher compressive strength but exhibited
slightly less flow than the fly ash-based geopolymer mortars forthe same mixing condition. The poor compressive strength of fly
ash-based mortars cured at low temperatures is attributed to the
presence of unreacted fly ash particles and large number of voids.
As the aggregate-binder ratio increased, the compressive strength
increased up to a ratio of 2.5 which indicated that the threshold
of aggregates in geopolymer mortars were slightly lower than
OPC mortars. The shrinkage strain of alkali-activated mortars was
also found to be lower than the OPC mortars.
4. Geopolymer concrete
4.1. Fresh and hardened properties
Various mix proportioning of geopolymer concrete (GPC) were
reported with target strength up to 80 MPa. The typical properties
of geopolymer concrete mixes used by the various authors (41, 45,
48, 49, 53) are summarised in Table 1. The properties of mixes
were studied with respect to water-geopolymer solid ratio, activa-
tor strength, water/Na2O ratio, curing time, curing temperature,
and age hardening. The slump of mixes varied depending on the
molarity of activator, workability aids and extra water added to
the mix[41]. The rheological parameters such as yield stress and
plastic viscosity were attempted over slump test of concrete to
assess its workability loss and flow behaviour. Yield stress gives
initial resistance to flow arose from the friction among the solid
particles while plastic viscosity governs the flow after it is initiated
resulting from viscous dissipation due to the movement of water inthe sheared material. Laskar & Bhattacharjee [42] studied the
B. Singh et al. / Construction and Building Materials 85 (2015) 7890 81
http://-/?-http://-/?-7/17/2019 1-s2.0-S0950061815002834-main
5/13
rheology of fly ash-based geopolymer concrete with slump varying
from 25 mm to flowing concrete with 120 M activator strength.
They found that the yield stress and plastic viscosity were affected
by the molar strength of the sodium hydroxide solution and the
ratio of silicate to hydroxide solution. The setting time of geopoly-
mer concrete was reported up to 120 min. Like Portland cement
pastes and mortars, geopolymers also behave like Bingham fluid
and have a history dependent rheological profile, i.e., geopolymers
may be kept in a fluid form, if subjected to constant shearing for a
certain period of time before initial setting starts [43]. The settingcould be enhanced up to 180 min with the use of naphthalene
based admixture and extended mixing time especially in the case
of slag-based geopolymer which has the potential for a wide range
of technological applications[44].
Hardjito et al. [41]produced fly ash-based GPC with the com-
pressive strength ranging between 30 and 80 MPa with the slump
varied from 100 to 250 mm (activator strength: 814 M). The opti-
mum strength was obtained at 0.18 water-geopolymer solid ratio
cured at 90 C. As the water-geopolymer solids ratio increased,
the compressive strength of GPC decreased analogous to the well
known relationship between compressive strength and water-ce-
ment ratio for OPC concrete. The compressive strength of GPC
remained unchanged with the age when tested after 24 h curing
at elevated temperature. Fernandez-Jimenez et al. [45] made fly
ash-based geopolymer concrete with a compressive strength of
45 MPa at 0.55 liquid/solid ratio cured at 85 C for 20 h. The devel-opment of high early strength in GPC was explained by its compact
microstructure, formation of adequate reaction products, smaller
mean size of the pores and good aggregate-paste bond. They
observed that GPC has a much lower modulus of elasticity
(18.4 GPa) than the OPC concrete (30.3 GPa). Olivia and Nikraz
[46] proportioned fly ash-based geopolymer concrete mix with a
compressive strength of 55 MPa at 28 days and cured at different
temperatures in the range of 6075C. The hardened mix had
higher tensile and flexural strengths, produced less expansion
and showed modulus of elasticity that were 1529% lower than
that of OPC concrete mix. The drying shrinkage (0.025%) of GPC
was less than the OPC concrete (0.09%) after 12 weeks. The mini-
mal shrinkage of GPC may also be due to the significant resistance
offered by its zeolitic microstructure towards drying loss of thewater incorporated during casting[45].
Several attempts [4753] have also been made to establish
correlations within the mechanical properties of geopolymer con-
crete. It was reported that the experimental splitting tensile
strength of fly ash-based GPC was higher than the OPC concrete
(Fig. 3). The increased strength is accounted for a denser interfacial
zone established between the aggregate and geopolymer paste.
The modulus of elasticity increased as the compressive strength
of GPC increased. The modulus of elasticity of GPC was found to
be lower than the values predicted by ACI guidelines for OPC con-
crete. Sofi et al.[48]studied the engineering properties of fly ash/
slag-based GPC. The splitting tensile strength and flexural strength
of GPC were comparable to those models presented by the
Australian Standard (AS 3600) for OPC concrete. Although, the dif-ference between splitting tensile and flexural strength of GPC
mixes has been found to be approximately 2 MPa, similarities
between the strength gain was apparent. Diaz-Loya et al.[49]pro-
posed the equation fr= 0.69p
fc0MPa for correlation between the
flexural strength (fr) and compressive strength and the equation
Ec= 580p
fcMPa for correlation between elastic modulus (Ec)
and compressive strength of GPC (fc= compressive strength).
When compared with the typical Poissons ratio value of OPC con-
crete (0.150.22), the values of GPC appeared to reside toward the
low end of range (0.080.22). Ryu et al. [50]suggested a model for
relationship between compressive strength and splitting tensile
strength (fsp= 0.17 (f0c)
3/4) for fly ash-based GPC. Bondar et al.
[51] reported a relationship between ultrasonic pulse velocity
and compressive strength of GPC. They found that GPC showed a
lower ultrasonic pulse velocity than the OPC concrete even those
Table 1
Typical properties of geopolymer concrete mixes.
Density
(kg/m3)
Molarity
(M)
Slump
(mm)
CS
(MPa)
STS
(MPa)
FS
(MPa)
MOE
(GPa)
Poissons
ratio
Activator/
binder
ratio
Curing
temperature
and time
Hardjito et al. [41] 23302430 1016 60215 3080 3.746 512 2331 0.120.16 0.350.4 6080 C for 24 h
Jimenez et al.[45] NR 8 & 12.5 NR 2943.5 NR 6.86 10.718.4 NR 0.4 & 0.55 85C for 20 h
Sofi et al.[48] 21472408 NR NR 4756.5 2.84.1 4.96.2 2339 0.230.26 0.450.59 23C till testing
Diaz-loya et al.[49] 18902371 14 100150 1080 NR 2.246.41 1.942 0.080.22 0.40.94 60 C for 72 hPan et al.[53] 18762555 8 NR 65.177.9 2.85.1 NR 11.241.2 0.150.19 0.40.65 60 C for 24 h
CS: compressive strength; STS: splitting tensile strength; FS: flexural strength; MOE: modulus of elasticity; NR: not reported.
Fig. 3. Correlations within the mechanical properties of fly ash-based geopolymer
concrete. (a) Splitting tensile strength vs compressive strength. (b) Modulus ofelasticity vs compressive strength[47].
82 B. Singh et al. / Construction and Building Materials 85 (2015) 7890
7/17/2019 1-s2.0-S0950061815002834-main
6/13
with the same or higher compressive strength. It was also reported
[52,53] that GPC was brittle as compared to its OPC counterpart
due to the highly cross-linked framework. The fracture energy of
GPC was also low because of its higher bond with aggregates as
compared to OPC concrete[54].
4.2. Interfacial transition zone
It is well known that the interfacial zone (ITZ) between aggre-
gate and matrix is the weakest link in OPC concrete at which
micro-cracks usually first develop under loads [55]. Investigation
of this zone is very crucial since it is known to have different
microstructure from the bulk of the hardened paste. The high
porosity of ITZ allows the easier penetration of external agents
such as chlorides, oxygen, sulphates, etc. into concrete structure.
Contrary to this, ITZ of GPC has been identified as being dense
and much less microstructurally distinct from the bulk of binder
region[56,57]. The stronger ITZ contributes to higher splitting ten-
sile strength, bond strength and durability of the GPC.
Lee and Deventer[56,57]discussed interface between the natu-
ral siliceous aggregates and paste in GPC using kaolin and albite as
precursors. The increase in concentration of the activating solutionincreased the binding capacity of the gel with natural aggregates.
The presence of chloride salts decreased the interfacial bonding
strength between the paste and aggregate probably by causing
gel crystallisation near the aggregate surfaces which resulted in
debonding. In another attempt, they found that the addition of
0.5 M soluble silicate into an activating solution (10 M NaOH and
2.5 M sodium silicate) facilitates the formation of an aluminium-
enriched aluminosilicate surface onto the aggregates through
accelerated Si-preferential dissolution of kaolin and albite. The sur-
face formed during albite leaching was found to possess a similar
Si/Al ratio to the real interface between a silicious aggregate and
fly ash/metakaolin geopolymer paste activated with 10 M NaOH
solution. Without soluble silicates, no deposited aluminosilicate
interface was observed. This suggested that both high concentra-tion of alkali and soluble silicate are essential for the formation
of a strong interface between silicious aggregates and geopolymer
pastes. Zhang et al.[58]reported that at the beginning, there were
many large voids in the fresh ITZ in potassium poly(sialate)
geopolymer concrete. As hydration proceeded, these voids were
completely filled with the hydration products. At this stage, the
difference in the microstructure between the ITZ and matrix was
hardly distinguishable. The contents of K/Al and Si/Al in the ITZ
were higher than those in the matrix. Demei et al. [59]presented
FESEM analysis of ITZ in the fly ash-based self compacting geopoly-
mer concrete with varying superplasticizer dosages. They reported
that relatively a loose and porous ITZ was found at low super-
plasticizer dosages (3%) whereas a dense ITZ was found between
the aggregate and geopolymer paste at higher dosage (7%). They
also found that the compressive strength increased with decrease
in the thickness of ITZ and this relationship depends on the super-
plasticizer dosage.
4.3. Bond between reinforcing bars and geopolymer concrete
The transfer of forces across the interface between concrete and
reinforcing steel bar is of fundamental importance in the structural
design[60]. Bond stresses in the reinforced concrete arise from two
distinct situations. The first is anchorage or development where
bars are terminated. The second is flexural bond or the change of
force along a bar due to a change in bending moment along the
member. The bond strength of reinforcing bars with concrete is
governed by several factors such as the strength of the concrete,
the thickness of the concrete surrounding the reinforcing bar, the
confinement of the concrete due to transverse reinforcement and
the bar geometry. Generally, the bond strength between the
reinforcing bar and matrix increases with increasing steel bar
diameter and compressive strength of GPC (Fig. 4). There is a
greater amount of slip for larger size rebars in GPC.
Sarker [61]found that the bond strength of fly ash-based GPC
increased with the increase of concrete cover-bar diameter ratio
(1.713.62) and the concrete compressive strength (2529 MPa).
He also observed that GPC has higher bond strength than the
OPC concrete because of higher splitting tensile strength and dense
interfacial transition zone between the aggregate and geopolymer
paste. Bond-slip behaviour[45]of GPC showed that the embedded
steel bar of 8 mm dia broke before slipping and concrete cracking
whereas the bar embedded in OPC concrete slipped. For
16 mm bar, GPC failed by matrix cracking while the bars in OPC
concrete were again observed to slip. Sofi et al. [62]reported that
the values of bond strength of steel bars in fly ash-based GPC werecomparable in both beam-end as well as direct pullout specimen
tests. The normalised bond strength increased with a reduction
in rebar size. The bond strength tested according to AS 3600, ACI
318-02 and EC2 recommendations showed that these Codes are
applicable and also safe to predict the developmental length for
GPC.
Attempts were also made to study behaviour of reinforced fly
ash-based GPC beams and columns with respect to longitudinal
tensile reinforcement ratio and concrete compressive strength as
test variables [6366]. Sumajouw et al. [63] reported that the
flexural capacity of beams increased with the increase in tensile
reinforcement (0.642.69%) but the effect of concrete compressive
strength was marginal. The ductility index increased significantly
for beams having longitudinal reinforcement ratio less than 2%.They also studied the strength of reinforced GPC slender columns
with respect to the compressive strength of concrete, longitudinal
reinforcement ratio and load eccentricity. The design provisions
mentioned in the Standards for OPC concrete can be used for
designing geopolymer concrete columns also. Dattatreya et al.
[64]found that the load carrying capacity of reinforced slag-based
GPC beams was 17.7% more than the Portland pozzolana cement
concrete beams at 2.68% tension reinforcement. Yost et al. [65]
indicated that loaddeflection behaviour of GPC beam was identi-
cal to OPC beam. The maximum strain obtained for under-rein-
forced beam was less than 3000 microstrains which is generally
assumed for design work. The predicted neutral axis depth was
15% less than the experimentally achieved value for GPC. The
Whitneys stress block for strength calculation was found applic-able for GPC also. Ng et al. [66]investigated potential use of steel
Fig. 4. Bond strength of fly ash-based geopolymer concrete as a function of steel bardiameter[47].
B. Singh et al. / Construction and Building Materials 85 (2015) 7890 83
http://-/?-http://-/?-7/17/2019 1-s2.0-S0950061815002834-main
7/13
fibres (up to 1.5 wt%) to replace conventional shear reinforcement
in GPC beams of 2250 mm span length. They found that the
increase in fibre volume led to an increase in the cracking load
and the ultimate shear strength. A good correlation of test data
was observed with the predictive fib Model Code 2010.
4.4. Fire behaviour
In general, concrete has good property with respect to fire resis-
tance. However, it is known that the residual strength of OPC con-
crete after firing between 800 C and 1000 C does not exceed 20
30% normally because of dehydration and destruction of C-S-H &
other crystalline hydrates, aggregate types, permeability etc. Fire
introduces high temperature gradient and as a result, the hot layer
tends to separate and spall from the cooler interior layer of the
body [67]. Contrary to this, geopolymers possess good fire resis-
tance at elevated temperature because of the existence of highly
distributed nano-pores in the ceramic like microstructure that
allows physically and chemically bonded water to migrate and
evaporate without damaging the aluminosilicate network [4].
During fire, several events such as evaporation of water adsorbed
by N-A-S-H gel, formation of anhydrous products, crystallizationof stable anhydrous phases and melting (sintering) leading to
destruction generally occurred. The phase transformation of
geopolymers during fire is depicted below.
Kong et al. [68] found that the residual strength of fly ash-based
geopolymer pastes increased by 6% after exposure to 800C,
whereas the strength of metakaolin-based geopolymer pastes was
reduced by 34%. During heating, the high permeability of fly ash-
based geopolymer provides the escape route for moisture in the
matrix, thereby decreasing the damage. The strength increase is
also partly attributed to the sintering reaction of unreacted fly ash
particles. Geopolymer pastes made with metakaolin and potassium
based activator showed an enhanced post-elevated temperature
performance compared to sodium based activator system. The
strength deterioration reduced with increasing Si/Al ratio (>1.5)
[69]. Aggregate size larger than 10 mm resulted in good strength
performance in both ambient and elevated temperature (800 C).
The strength loss in fly ash-based geopolymer concrete at elevated
temperatures is attributed to thermal mismatch between the
geopolymer paste and aggregate [70]. No spalling was reported in
the samples by Zhao and Sanjayan [71] when fly ash-based GPC
with compressive strength ranging from 40 to 100 MPa was
exposed to 850 C. They also found that at the same strength level,
GPC possessed higher spalling resistance under fire than the OPC
concrete due to its increased porosity.
5. Durability studies
One of the major problems associated with OPC concrete is its
long term durability which had always been an issue against
aggressive environments. The deterioration of concrete is usuallyassessed forsulphate attack,chloride induced corrosion, atmospheric
carbonation, alkali-silica reaction and freezethaw attack.In view of
this, several studies are being carried out to understand the beha-
viour of geopolymers exposed to these conditions.
5.1. Alkali-silica reaction
Alkali-silica reaction (ASR) causes gradual but severe deteri-
oration of hardened Portland cement concrete in terms of itsstrength loss, cracking, volume expansion etc. It involves the reac-
tion between the hydroxyl ion in the pore solution within the con-
crete matrix and reactive silica of the aggregate. In general terms,
the reactions will proceed in stages, with the first stage being the
hydrolysis of reactive silica by hydroxyl ions to form alkali-silica
gel and a later secondary overlapping stage being the absorption
of water by the gel, which will result in increase of volume[72].
(i) Acid-based reaction
H0:38SiO2:19 0:38NaOH!Na0:38SiO2:19 0:38H2O 3
(ii) Attack of the siloxane bridges and disintegration of the silica
Na0:38SiO2:19 1:62NaOH!2Na2 H2SiO
24
4
In geopolymer concrete, the un-utilised alkali after geopolymer-
ization of aluminosilicates is expected to react with the silica of the
aggregates causing disruption of their siloxane bridges. It is
reported that geopolymer mortars using aggregates of different
reactivities expanded less than the corresponding Portland cement
mortars[73]. The geopolymer mortars appeared to be sound with-
out any surface cracking. The cause of expansion in slag-based
geopolymer mortars is the formation of sodium calcium silicate
hydrate reaction product with rosette-type morphology [74].
Contrary to this, there was no significant expansion in fly ash-
based geopolymer mortars. The formation of crystalline zeolites
was very slow and since these minerals are usually found in thegaps of the matrix, the existence of stress that might generate
cracking is unlikely [75]. Geopolymer mortar bars made with fly
ash/slag blends expanded less than 0.1% limit prescribed in ASTM
C1260-07 after 16 days (Fig. 5). At 90 days exposure, these mortars
failed to meet the specified criteria. Increasing slag content in fly
ash/slag mix increased the expansion of resulting systems [76].
ASR has also been claimed to be helpful in providing a strong bond
at the paste-aggregate interface, thus enhancing the tensile
strength of GPC[8]. Patil et al.[73]indicated that sandstone, quartz
and limestone aggregates in geopolymer concrete were not prone
Fig. 5. Alkali-silica reaction in various geopolymer and OPC mortars under anaccelerated condition (1 M NaOH) at 80 C[76].
84 B. Singh et al. / Construction and Building Materials 85 (2015) 7890
http://-/?-http://-/?-7/17/2019 1-s2.0-S0950061815002834-main
8/13
to ASR. During accelerated mortar bar test, a slight expansion was
noticed because of re-initiation of the geopolymerization process
of unreacted fly ash particles leading to lower porosity and higher
strength. The lower sensitivity of reactive aggregates in GPC pro-
vides economic advantages in areas where high quality deposits
of aggregates have been depleted.
5.2. Effect of acid attack
The acid resistance of geopolymer pastes/concrete was studied
by several authors[7784]. The extent of degradation depends on
the concentration of acid solution and period of exposure.
Davidovits et al. [8] indicated that metakaolin-based geopolymer
pastes showed only 7% mass loss when sample was immersed in
5% H2SO4 for 30 days. It was also reported that fly ash-based
geopolymer pastes retained a dense microstructure after 3 months
exposure in HNO3. Temuujin et al. [77] concluded that acid and
alkaline resistance of fly ash-based geopolymer strongly depend
on its mineralogical composition. High solubility of Al, Si and Fe
ions was obtained in both strong alkali and acid solutions. The per-
formance of fly ash-based geopolymer pastes when exposed to 5%acetic acid and 5% H2SO4 solutions was superior to ordinary
Portland cement pastes. The deterioration in pastes was connected
to depolymerisation of the aluminosilicate network and formation
of zeolites[78].
5
Wallah and Rangan[41]found that the reduction in compres-
sive strength of fly ash-based GPC in 0.5% H2SO4 solution was
20% after 12 months exposure. This value was52% and65%respectively when samples exposed to 1% and 2% H2SO4 solution.
Pitting and erosion on the surface of the concrete were also
observed. The loss in strength of concrete is mainly due to the
degradation in the geopolymer matrix rather than the aggregate.
They concluded that the acid resistance of GPC was superior to
OPC concrete. Ariffin et al. [79] exposed GPC made with a blend
of pulverized fuel ash and palm oil fuel ash in 2% solution of sul-
phuric acid for 18 months. The weight loss in GPC was 8% whileOPC concrete exhibited 20% weight loss. The strength reduction
in GPC was 35% in 18 months as against 68% strength loss in OPC
concrete after 30 days and was severely deteriorated after
18 months. The C-S-H could have severe deleterious effect on
OPC concrete while N-A-S-H gel appeared to have little effect on
the structure of GPC. Sathia et al. [80]reported the weight loss in
concrete samples was less than 5% after 3 months exposure in 3%
H2SO4 solution. Bakharev et al. [81] found that slag-based GPC
(40 MPa) exhibited33% reduction in strength compared to 47%in OPC concrete when exposed in acetic acid solution (pH 4) for
12 months. The slag particles and low calcium C-S-H with average
Ca/Si ratio of 1 were more stable in the acid solution than the con-
stituents of the OPC pastes. During immersion in 2% H2SO4 solu-
tion, the strength loss was11% compared to 36.2% for OPCconcrete.
5.3. Effect of sulphate attack
Fly ash-based geopolymer pastes did not deteriorate signifi-
cantly, under the influence of water, sodium sulphate (4.4%)
and ASTM sea water [82]. Only some fluctuations in flexural
strength were observed between 7 days and 3 months exposures.
The least strength change was observed in the pastes exposed in
the 5% Na2
SO4
and 5% MgSO4
solutions while most significant
deterioration was observed in the 5% mixed sulphate solution
Fig. 6. Atomic force microscope images of fly ash-based geopolymer exposed under
sulphate after 4 months[86].
B. Singh et al. / Construction and Building Materials 85 (2015) 7890 85
http://-/?-http://-/?-7/17/2019 1-s2.0-S0950061815002834-main
9/13
(Na2SO4+ MgSO4) after 5 months exposure [83]. In fly ash/slag
system, the extensive physical deterioration of pastes was
observed during immersion in MgSO4 solution after 3 months
exposure but not in Na2SO4 solution. The calcium sulphate dihy-
drate formed in paste was identified as being particularly dama-
ging to the materials in MgSO4 [84]. Atomic force microscopic
images of fly ash-based geopolymer pastes exposed to sulphate
environment are shown inFig. 6. In the case of Na2SO4 solution,only exposition of grains was clearly visible while in MgSO4 solu-
tion, both exposition of grains and dissolved aluminosilicate
matrix were observed showing severity of MgSO4 attack [85].
The deterioration is considered mainly due to the destruction of
aluminosilicate skeleton, liberation of silicic acid, leaching of
sodium ion etc. [86]. These reactions seem to have significant
effect on the mechanical strength. The geopolymer prepared with
NaOH activator had the best performance over those made with a
synergistically used sodium silicate and NaOH/KOH activators,
which is attributed to its stable cross-linked aluminosilicate poly-
mer structure.
Several attempts [41,87] have been made to study sulphate
resistance of GPC. The deterioration in concrete was evaluated in
terms of its visual appearance, weight loss and change in compres-
sive strength. Hardjito et al. [41]observed that there was no sig-
nificant effect of 5% Na2SO4 solution in the compressive strength,
the weight loss and the dimension of fly ash-based GPC after
3 months exposure. Rajamane et al. [87] reported sulphate resis-
tance of fly ash-based GPC for 3 months in 5% Na2SO4 and 5%
MgSO4 solutions. The weight loss in samples was 2.4% only.
There was 229% loss of compressive strength as compared to 9
38% in the OPC concrete. The deterioration of OPC concrete can
be attributed to the formation of expansive gypsum and ettringite
which can cause expansion, cracking and spalling in the concrete.
Contrary to this, GPC in general do not contain Ca(OH) 2and mono-
sulphoaluminate in the matrix to cause expansion.
5.4. Carbonation and permeability
Bernal et al.[88]studied slag/metakaolin-based GPC (w/b ratio
0.47) under an accelerated carbonation test using CO2 concentra-
tion of 3.0 0.2% at 20 C for 28 days. They found that the com-
pressive strength decreased monotonically as the carbonation
proceeds. The relationship between the pore volume and extent
of carbonation was much more similar with samples with differ-
ent percentages of metakaolin contrary to the slag-based samples.
This suggested that porosity is not the only parameter controlling
the strength loss of the carbonated binder. There must be a
convoluting effect due to the binder gel chemistry, which deter-
mines the residual level of strength after an accelerated carbona-
tion. Olivia and Nikraz [46] reported lower water permeability
(2.464.67 1011
m/s) of GPC (activator-fly ash ratio, 0.300.40cured at 60 C for 24 h) than the OPC concrete due to its denser
paste and smaller pore inter-connectivity. They also reported that
the water-geopolymer solids ratio was the most influential
parameter that affects the properties of GPC. Bondar et al. [51]
studied the oxygen and chloride permeability of alkali-activated
concrete made with the Iranian natural pozzolan (Taftan andesite
and Shahindej dacite). They concluded that alkali-activated natu-
ral pozzalona concrete has 1035% lower oxygen permeability at
normal curing conditions for 90 days compared with the OPC con-
crete. The rapid chloride permeability test gave high values for
the alkali-activated concrete. This is probably due to the very
high alkali ion concentration in the pore solution promoting
higher electrical conductivity in the GPC. This effect seems to
reduce with age due to a change in the porosity of the GPCmicrostructure.
5.5. Corrosion of steel reinforcement
Corrosion potential is a technique used to detect the state of
reinforcement without disturbing the structures. This is important
because the intensity of corrosion of steel in concrete is generally
known only after the concrete has cracked or disrupted. Various
studies[46,80,89]were reported to estimate the corrosion poten-
tial of steel within the GPC as per ASTM C876. Olivia and Nikraz[46] reported that the half cell potential of GPC was lower than
the specified value of404 mV mentioned in the Standard for sev-ere corrosion after 91 days. Sathia et al. [80] also reported corro-
sion potential up to 300 mV which showed a probablecorrosion indication due to the lower pH of concrete during the
half-cell potential measurement. Accelerated corrosion results
showed that GPC mixes exhibited low level corrosion activity
and time to failure that were 3.865.70 times longer than those
of the OPC concrete. Under impressed voltage, a crack appeared
suddenly in the concrete when time to failure was reached and this
was followed immediately by high current reading. The large
amounts of fly ash and alkaline activators in the GPC mix increased
the availability of ions that can produce high electrical resistance at
high impressed voltage. This enhanced the cathodic reaction and
reduces the rate of corrosion, which in turn, reduces the tensile
stress of the specimens, thus decreasing the risk of cracking and
clearly extending the time to failure [46]. Reddy et al. [89]com-
pared the durability of GPC with that of OPC concrete exposed to
marine environment for a period of 21 days. The initial corrosion
current measured for GPC (7191 mA) was much lower than that
of OPC concrete (772 mA). The OPC specimens initially recorded
decrease in the current but later started increasing while the GPC
current never showed significant increase.
6. Research and development at CSIR-CBRI
A systematic R&D work is initiated at CSIR-Central Building
Research Institute, Roorkee on the development of heat and ambi-ent cured geopolymers using fly ash, slag and other aluminosili-
cates as precursors. In view of variability in the constituents of
fly ash, the property optimisation of geopolymeric pastes was car-
ried out as a function of activator concentration and its dosage,
water-geopolymer solid ratio, curing time and curing temperature
[13]. Geopolymerisation reaction, thermal stability, identification
of bond linkages and microstructural features were analysed by
various techniques such as quasi isothermal DSC, TGA, FTIR and
FESEM. The durability of geopolymer pastes/mortars was also
studied in terms of alkali-silica reaction and also in acidic and sul-
phate environments for 4 months [76,86]. The suitability of these
geopolymer pastes was assessed in making various geopolymeric
products such as mortars & concrete, bricks, solid & hollow blocks,
insulation concrete, foam, sandwich composites and temperatureresistant coatings (Fig. 7(ac)). Attempt was also made to utilise
lime sludge, a waste from paper industry with the geopolymeric
binders for making paving blocks.
Fly ash-based GPC mixes were made with the compressive
strength of 2555 MPa using absolute volume method adopted
for OPC concrete mixes. The strength of GPC increased with
decreasing water-geopolymer solid ratio as it is said analogous to
the water-cement ratio of the OPC concrete. The compressive
strength increased with increasing molarity of the activator (10
16 M) probably due to the formation of stable aluminosilicate net-
works following the dissolution of silica and alumina in the solu-
tion from the fly ash. It was found that the splitting tensile
strength of GPC was more than those of predicted values as per
ACI 318 guideline and other existing empirical equations. A trendline curve between the compressive strength and modulus of
86 B. Singh et al. / Construction and Building Materials 85 (2015) 7890
7/17/2019 1-s2.0-S0950061815002834-main
10/13
elasticity showed that the elastic modulus was lower (17%) than
the one predicted by Ivan Diaz-Loya et al. for GPC and also the val-
ues obtained with ACI guidelines. As expected, the bond strength of
steel bar embedded in GPC increased with increasing steel bar
diameter and compressive strength of concrete. It was noted that
the bond strength between geopolymer paste and reinforcing bars
was found to be higher than the OPC concrete[47].
Light weight geopolymer concrete was proportioned with the
help of fly ash, activators, expanded polystyrene beads (EPS up
to 3 wt% or 91 vol%), admixtures and fine & coarse aggregates
(Fig. 7(a)). It was noted that a decrease in the strength was more
when larger size of EPS beads (
7/17/2019 1-s2.0-S0950061815002834-main
11/13
(15 MPa) as specified (13.1 MPa) in ASTM C 90. Regarding fireperformance, the samples were non-ignitable and exhibited Class
I-very low spread of flame as per BS EN-476 part 7 (Fig. 8). Itwas noted that the fire propagation index of the samples was
7/17/2019 1-s2.0-S0950061815002834-main
12/13
[4] Duxon P, Fernandez-Jiminez A, Provis JL, Luckey GC, Palomo A, Van DeventureJSJ. Geopolymer technology: the current state of the art. J Mater Sci2007;42:291733.
[5] Provis JL, VanDeventer JSJ, editors. Geopolymers, structure, processing,properties and application. UK: Woodhead Publishing Limited; 2009.
[6] Li C, Sun H, Li L. A review: the comparison between alkali-activated slag (Si+Ca)and metakaolin (Si+Al) cements. Cem Concr Res 2010;40:13419.
[7] Komnitsas KA. Potential of geopolymer technology towards green buildingsand sustainable cities. In: International conference on green buildings andsustainable cities, procedia engineering, vol. 21; 2011: p.102332.
[8] Davidovits J. Geopolymers: inorganic polymeric new materials. J Therm Anal1991;37:163356.
[9] Davidovits J, Buzzi L, Rocher P, Marini DGC, Tocco S. Geopolymeric cementbased on low cost geologic materials geocistem. In: second internationalconference geopolymer 99, France; 1999.
[10] Van Jaarsveld J, Van Deventure J. Effect of alkali metal activators on theproperties of fly ash-based geopolymer. Ind Eng Chem Res 1999;38:3932 .
[11] Xu H, Van Deventure JSV. Geopolymerisation of multiple minerals. Min. Eng.2002;15:11319.
[12] Zhang B, MacKenzie KJD, Brown IWM. Crystalline phase formation inmetakaolinite geopolymers activated with NaOH and sodium silicate. JMater Sci 2009;44:466876.
[13] Ishwarya G. Development of geopolymer concrete cured at ambienttemperature [M. Tech thesis]. Roorkee, India: CSIR-Central Building ResearchInstitute (AcSIR); 2013.
[14] Silva PD, Crenstil KS, Sirivivatnanon V. Kinetics of geopolymerization: role ofAl2O3 and SiO2. Cem Concr Res 2007;37:5128.
[15] de Vargas AS, DalMolin DCC, Vilela ACV, da Silva FJ, Pavo B, Veit H. The effectsof Na
2O/SiO
2 molar ratio, curing temperature and age on compressive
strength, morphology and microstructure of alkali-activated fly ash-basedgeopolymers. Cem Concr Comp 2011;33:65360.
[16] Yip CK, Luckey GC, Provis JL, van Deventure JSV. The coexistence ofgeopolymeric gel and calcium silicate hydrate at the early stage of alkalineactivation. Cem Concr Res 2005;35:168897.
[17] Somna K, Jaturapitakkul C, Kajitvichyanukul P, Chindaprasirt P. NaOH-activated ground fly ash geopolymer cured at ambient temperature. Fuel2011;90:211824.
[18] Yip CK, Luckey GC, Provis JL, van Deventure JSV. Effect of calcium silicatesources on geopolymerisation. Cem Concr Res 2008;38:55464.
[19] Kumar S, Kumar R, Melhotra SP. Influence of granulated blast furnace slag onthe reaction, structure and properties of fly ash-based geopolymer. J Mater Sci2010;45:60715.
[20] Puligilla S, Mondal P. Role of slag in microstructural development andhardening of fly ash-slag geopolymer. Cem Concr Res 2013;43:7080 .
[21] Chindaprasirt P, Silva PD, Crentsil KS, Hanjitsuwan S. Effect of SiO 2 and Al2O3on the setting and hardening of high calcium fly ash-based geopolymersystems. J Mater Sci 2012;47:487683.
[22] Guo X, Shi H, Dick WA. Compressive strength and microstructuralcharacteristics of class C fly geopolymer. Cem Concr Comp 2010;32:1427 .
[23] Kusbiantoro A, Ibrahim MS, Muthusamy K, Alias A. Development of sucroseand citric acid as natural based admixture for fly ash based geopolymer. ProcEnviron Sci 2013;17:596602.
[24] Nemotallahi B, Sanjayan JG. Effect of different superplasticizer and activatorcombinations on workability and strength of fly ash based geopolymer. MaterDes 2014;57. 667-67.
[25] Jang JG, Lee NK, Lee HK. Fresh and hardened properties of alkali-activatedfly ash/slag pastes with superplasticizers. Constr Build Mater2014;50:16976.
[26] Palomo A, Grutzeck MW, Blanco MT. Alkali-activated fly ashes: a cement forthe future. Cem Concr Res 1999;29:13239.
[27] Perera DS, Uchida O, Vance ER, Finnie KS. Influence of curing schedule on theintegrity of geopolymer. J Mater Sci 2007;42:3099106.
[28] Heah CY, Kamarudin H, Al Bakri AMM, Binhussain M, Luqman M, Nizar IK, et al.Effect of curing profile on Kaolin-based geopolymers. Phys Procedia2011;22:30511.
[29] Rovnanik P. Effect of curing temperature on the development of hard structure
of metakaolin-based geopolymer. Constr Build Mater 2010;24:117683.[30] Kani EN, Allahverdi A. Effects of curing time and temperature on strength
development of inorganic polymeric binder based on natural pozzolan. J MaterSci 2009;44:308897.
[31] Van Jaarsveld JGS, Van Deventer JSJ, Lukey GC. The effect of composition andtemperature on the properties of fly ash and kaolinite based geopolymers.Chem Eng J 2002;89:6373.
[32] Chindaprasirt P, Chareerat T, Sirivivatnanon V. Workability and strength ofcoarse high calcium fly ash geopolymer. Cem Concr Comp 2007;29:2249 .
[33] Chindaprasirt P, Chareerat T, Hatanaka S, Cao T. High strength geopolymerusing fine high-calcium fly ash. J Mater Civ Eng 2011;23:26470.
[34] Temuujin J, van Riessen A, Mackenzie. Preparation and characterisation of flyash-based geopolymer mortars. Constr Build Mater 2010;24:190610.
[35] Khandelwal M, Ranjith PG, Pan Z, Sanjayan JG. Effect of strain rate on strengthproperties of low-calcium fly ash-based geopolymer mortar under drycondition. Arabian J Geosci 2013;6:23839.
[36] Hardjito D, Fung SS. Fly ash-based geopolymer mortar incorporating bottomash. Mode Appl Sci 2010;4:4452.
[37] Sathonsaowaphak A, Chindaprasirt P, Pimraksa K. Workability and strength oflignite bottom ash geopolymer mortar. J Hazard Mater 2009;168:4450.
[38] BroughAR, Atkinson A. Sodium silicate-based alkali-activated slag mortars PartI. Strength, hydration and microstructure. Cem Concr Res 2002;32:86579.
[39] Yang KH, Song JK, Lee KS, Ashour AF. Flow and compressive strength of alkali-activated mortars. ACI Mater J 2009;106:508.
[40] Yang KH, Song JK. Workability loss and compressive strength development ofcementless mortars activated by combination of sodium silicate and sodiumhydroxide. J Mater Civ Eng ASCE 2009;21:11927.
[41] Hardjito D, Wallah SE, Sumajouw DMJ, Rangan BV. On the development of flyash-based geopolymer concrete. ACI Mater J 2004;101:46772.
[42] Laskar AI, Bhattacharjee R. Rheology of fly ash-based geopolymer concrete. ACI
Mater J 2011;108:53642.[43] Montes C, Allouche EN. Rheological behaviour of fly ash-based geopolymers.
STP 1566 on Geopolymer Binder Systems 2013, ASTM: p. 7284.[44] Palacios M, Banfill Phillip FG, Puertas F. Rheology and setting of alkali-
activated slag pastes and mortars: effect of organic admixture. ACI Mater J2008;105:1408.
[45] Fernandez-Jiminez AM, Palomo A, Lopez-Hombrados C. Engineering propertiesof alkali-activated fly ash concrete. ACI Mater J 2006;103:10612.
[46] Olivia M, Nikraz H. Properties of fly ash geopolymer concrete designed bytaguchi method. Mater Des 2012;36:1918.
[47] Singh A. Engineering properties of reinforced geopolymer concrete [B. Techproject report]. Roorkee, India. CSIR-Central Building Research Institute; 2012.
[48] Sofi M, Van Deventer JSJ, Mendis PA, Lukey GC. Engineering properties ofinorganic polymer concretes. Cem Concr Res 2007;37:2517.
[49] Diaz-Loya EI, Allouche EN, Vaidya S. Mechanical properties of fly ash-basedgeopolymer concrete. ACI Mater J 2011;108:3006.
[50] Ryu GS, Lee YB, Koh KT, Chung YS. The mechanical properties of fly ash-basedgeopolymer concrete with alkaline activators. Constr Build Mater2013;47:40918.
[51] Bondar D, Lynsdale CY, Milestone NB, Hassani N, Ramezanianpour AA.Engineering properties of alkali-activated natural pozzolan concrete. ACIMater J 2011;108:6472.
[52] Yost JR, Radlinska A, Ernst S, Salera M. Structural behavior of alkali activatedfly ash concrete. Part 1: mixture design, material properties and samplefabrication. Mater Struct 2012;46:43547.
[53] Pan Z, Sanjayan JG, Rangan BV. Fracture properties of geopolymer paste andconcrete. Mag Concr Res 2011;63:76377.
[54] Sarker PK, Haque R, Ramgolam KV. Fracture properties of heat cured fly ash-based geopolymer concrete. Mater Des 2013;44:5806.
[55] Mehta PK, Monteiro PJM. Concrete: microstructure, properties andmaterials. USA: The McGraw Hill companies, Inc.; 2006.
[56] Lee WKW, Van Deventure JSJ. The interface between natural siliceousaggregates and geopolymers. Cem Concr Res 2004;34:195206.
[57] Lee WKW, Van Deventure JSJ. Chemical interactions between siliceousaggregates and low-Ca alkali-activated cements. Cem Concr Res2007;37:84455.
[58] Zhang YS, Sun W, Li JZ. Hydration process of interfacial transition in potassium
polysialate (K-PSDS) geopolymer concrete. Mag Concr Res 2005;57:338.[59] Demie S, Nuruddin MF, Shafiq N. Effects of micro-structure characteristics of
interfacial transition zone on the compressive strength of self-compactinggeopolymer concrete. Constr Build Mater 2013;41:918.
[60] ACI 408R-03. Bond and development of straight reinforcing bars in tension.[61] Sarker PK. Bond strength of reinforcing steel embedded in fly ash-based
geopolymer concrete. Mater Struct 2011;44:102130.[62] Sofi M, Van Deventer JSJ, Mendis PA, Lukey GC. Bond performance of
reinforcing bars in inorganic polymer concrete (IPCs). J Mater Sci2007;42:310716.
[63] Sumajouw DMJ, Hardjito D, Wallah SE, Rangan BV. Fly ash-based geopolymerconcrete: study of slender reinforced columns. J Mater Sci 2007;42:312430 .
[64] Dattatreya JK, Rajamane NP, Ambily PS. Structural behaviour of reinforcedgeopolymer concrete beams and columns. CSIR-SERC Research Report RR-6;2009.
[65] Yost JR, Radlinska A, Ernst S, Salera M. Structural behaviour of alkali activatedfly ash concrete. Part 2: structural testing and experimental findings. MaterStruct 2012;46:44962.
[66] Ng TS, Amin A, Foster SJ. The behaviour of steel-fibre-reinforced geopolymer
concrete beams in shear. Mag Concr Res 2013;65:30818.[67] Neville AM. Properties of concrete. 4th ed. India: Dorling Kindersley
Publishing, Inc.; 1997.[68] Kong DLK, Sanjayan JG, Crentsil KS. Comparative performance of geopolymers
made with metakaolin and fly ash after exposure to elevated temperature.Cem Concr Res 2007;37:15839.
[69] Kong DLK, Sanjayan JG, Crentsil KS. Factors affecting the performance ofmetakaolin geopolymers exposed to elevated temperatures. J Mater Sci2008;43:82431.
[70] Kong DLY, Sanjayan JG. Effect of elevated temperatures on geopolymer paste,mortar and concrete. Cem Concr Res 2010;40:3349.
[71] Zhao R, Sanjayan JG. Geopolymer and Portland cement concretes in simulatedfire. Mag Concr Res 2011;63:16373.
[72] Swamy RN, editor. The alkali-silica reaction in concrete. UK: Blackie and SonsLimited; 1992.
[73] Patil KK, Allouche EN. Impact of alkali silica reaction on fly ash-basedgeopolymer concrete. J Mater Civ Eng ASCE 2013;25:1319.
[74] Fernandez-Jimenez A, Puertas F. The alkali-silica reaction in alkali-activated
granulated slag mortars with reactive aggregate. Cem Concr Res2002;32:101924.
B. Singh et al. / Construction and Building Materials 85 (2015) 7890 89
http://refhub.elsevier.com/S0950-0618(15)00283-4/h0020http://refhub.elsevier.com/S0950-0618(15)00283-4/h0020http://refhub.elsevier.com/S0950-0618(15)00283-4/h0020http://refhub.elsevier.com/S0950-0618(15)00283-4/h0025http://refhub.elsevier.com/S0950-0618(15)00283-4/h0025http://refhub.elsevier.com/S0950-0618(15)00283-4/h0025http://refhub.elsevier.com/S0950-0618(15)00283-4/h0030http://refhub.elsevier.com/S0950-0618(15)00283-4/h0030http://refhub.elsevier.com/S0950-0618(15)00283-4/h0030http://refhub.elsevier.com/S0950-0618(15)00283-4/h0040http://refhub.elsevier.com/S0950-0618(15)00283-4/h0040http://refhub.elsevier.com/S0950-0618(15)00283-4/h0050http://refhub.elsevier.com/S0950-0618(15)00283-4/h0050http://refhub.elsevier.com/S0950-0618(15)00283-4/h0050http://refhub.elsevier.com/S0950-0618(15)00283-4/h0055http://refhub.elsevier.com/S0950-0618(15)00283-4/h0055http://refhub.elsevier.com/S0950-0618(15)00283-4/h0060http://refhub.elsevier.com/S0950-0618(15)00283-4/h0060http://refhub.elsevier.com/S0950-0618(15)00283-4/h0060http://refhub.elsevier.com/S0950-0618(15)00283-4/h0060http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0070http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0075http://refhub.elsevier.com/S0950-0618(15)00283-4/h0080http://refhub.elsevier.com/S0950-0618(15)00283-4/h0080http://refhub.elsevier.com/S0950-0618(15)00283-4/h0080http://refhub.elsevier.com/S0950-0618(15)00283-4/h0085http://refhub.elsevier.com/S0950-0618(15)00283-4/h0085http://refhub.elsevier.com/S0950-0618(15)00283-4/h0085http://refhub.elsevier.com/S0950-0618(15)00283-4/h0090http://refhub.elsevier.com/S0950-0618(15)00283-4/h0090http://refhub.elsevier.com/S0950-0618(15)00283-4/h0095http://refhub.elsevier.com/S0950-0618(15)00283-4/h0095http://refhub.elsevier.com/S0950-0618(15)00283-4/h0095http://refhub.elsevier.com/S0950-0618(15)00283-4/h0100http://refhub.elsevier.com/S0950-0618(15)00283-4/h0100http://refhub.elsevier.com/S0950-0618(15)00283-4/h0100http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0110http://refhub.elsevier.com/S0950-0618(15)00283-4/h0110http://refhub.elsevier.com/S0950-0618(15)00283-4/h0110http://refhub.elsevier.com/S0950-0618(15)00283-4/h0115http://refhub.elsevier.com/S0950-0618(15)00283-4/h0115http://refhub.elsevier.com/S0950-0618(15)00283-4/h0115http://refhub.elsevier.com/S0950-0618(15)00283-4/h0120http://refhub.elsevier.com/S0950-0618(15)00283-4/h0120http://refhub.elsevier.com/S0950-0618(15)00283-4/h0120http://refhub.elsevier.com/S0950-0618(15)00283-4/h0125http://refhub.elsevier.com/S0950-0618(15)00283-4/h0125http://refhub.elsevier.com/S0950-0618(15)00283-4/h0125http://refhub.elsevier.com/S0950-0618(15)00283-4/h0130http://refhub.elsevier.com/S0950-0618(15)00283-4/h0130http://refhub.elsevier.com/S0950-0618(15)00283-4/h0135http://refhub.elsevier.com/S0950-0618(15)00283-4/h0135http://refhub.elsevier.com/S0950-0618(15)00283-4/h0140http://refhub.elsevier.com/S0950-0618(15)00283-4/h0140http://refhub.elsevier.com/S0950-0618(15)00283-4/h0140http://refhub.elsevier.com/S0950-0618(15)00283-4/h0145http://refhub.elsevier.com/S0950-0618(15)00283-4/h0145http://refhub.elsevier.com/S0950-0618(15)00283-4/h0150http://refhub.elsevier.com/S0950-0618(15)00283-4/h0150http://refhub.elsevier.com/S0950-0618(15)00283-4/h0150http://refhub.elsevier.com/S0950-0618(15)00283-4/h0155http://refhub.elsevier.com/S0950-0618(15)00283-4/h0155http://refhub.elsevier.com/S0950-0618(15)00283-4/h0155http://refhub.elsevier.com/S0950-0618(15)00283-4/h0160http://refhub.elsevier.com/S0950-0618(15)00283-4/h0160http://refhub.elsevier.com/S0950-0618(15)00283-4/h0165http://refhub.elsevier.com/S0950-0618(15)00283-4/h0165http://refhub.elsevier.com/S0950-0618(15)00283-4/h0170http://refhub.elsevier.com/S0950-0618(15)00283-4/h0170http://refhub.elsevier.com/S0950-0618(15)00283-4/h0170http://refhub.elsevier.com/S0950-0618(15)00283-4/h0175http://refhub.elsevier.com/S0950-0618(15)00283-4/h0175http://refhub.elsevier.com/S0950-0618(15)00283-4/h0175http://refhub.elsevier.com/S0950-0618(15)00283-4/h0180http://refhub.elsevier.com/S0950-0618(15)00283-4/h0180http://refhub.elsevier.com/S0950-0618(15)00283-4/h0185http://refhub.elsevier.com/S0950-0618(15)00283-4/h0185http://refhub.elsevier.com/S0950-0618(15)00283-4/h0185http://refhub.elsevier.com/S0950-0618(15)00283-4/h0190http://refhub.elsevier.com/S0950-0618(15)00283-4/h0190http://refhub.elsevier.com/S0950-0618(15)00283-4/h0195http://refhub.elsevier.com/S0950-0618(15)00283-4/h0195http://refhub.elsevier.com/S0950-0618(15)00283-4/h0200http://refhub.elsevier.com/S0950-0618(15)00283-4/h0200http://refhub.elsevier.com/S0950-0618(15)00283-4/h0200http://refhub.elsevier.com/S0950-0618(15)00283-4/h0205http://refhub.elsevier.com/S0950-0618(15)00283-4/h0205http://refhub.elsevier.com/S0950-0618(15)00283-4/h0205http://refhub.elsevier.com/S0950-0618(15)00283-4/h0210http://refhub.elsevier.com/S0950-0618(15)00283-4/h0210http://refhub.elsevier.com/S0950-0618(15)00283-4/h0220http://refhub.elsevier.com/S0950-0618(15)00283-4/h0220http://refhub.elsevier.com/S0950-0618(15)00283-4/h0220http://refhub.elsevier.com/S0950-0618(15)00283-4/h0225http://refhub.elsevier.com/S0950-0618(15)00283-4/h0225http://refhub.elsevier.com/S0950-0618(15)00283-4/h0230http://refhub.elsevier.com/S0950-0618(15)00283-4/h0230http://refhub.elsevier.com/S0950-0618(15)00283-4/h0240http://refhub.elsevier.com/S0950-0618(15)00283-4/h0240http://refhub.elsevier.com/S0950-0618(15)00283-4/h0245http://refhub.elsevier.com/S0950-0618(15)00283-4/h0245http://refhub.elsevier.com/S0950-0618(15)00283-4/h0250http://refhub.elsevier.com/S0950-0618(15)00283-4/h0250http://refhub.elsevier.com/S0950-0618(15)00283-4/h0250http://refhub.elsevier.com/S0950-0618(15)00283-4/h0255http://refhub.elsevier.com/S0950-0618(15)00283-4/h0255http://refhub.elsevier.com/S0950-0618(15)00283-4/h0255http://refhub.elsevier.com/S0950-0618(15)00283-4/h0260http://refhub.elsevier.com/S0950-0618(15)00283-4/h0260http://refhub.elsevier.com/S0950-0618(15)00283-4/h0260http://refhub.elsevier.com/S0950-0618(15)00283-4/h0265http://refhub.elsevier.com/S0950-0618(15)00283-4/h0265http://refhub.elsevier.com/S0950-0618(15)00283-4/h0270http://refhub.elsevier.com/S0950-0618(15)00283-4/h0270http://refhub.elsevier.com/S0950-0618(15)00283-4/h0270http://refhub.elsevier.com/S0950-0618(15)00283-4/h0275http://refhub.elsevier.com/S0950-0618(15)00283-4/h0275http://refhub.elsevier.com/S0950-0618(15)00283-4/h0275http://refhub.elsevier.com/S0950-0618(15)00283-4/h0280http://refhub.elsevier.com/S0950-0618(15)00283-4/h0280http://refhub.elsevier.com/S0950-0618(15)00283-4/h0280http://refhub.elsevier.com/S0950-0618(15)00283-4/h0285http://refhub.elsevier.com/S0950-0618(15)00283-4/h0285http://refhub.elsevier.com/S0950-0618(15)00283-4/h0285http://refhub.elsevier.com/S0950-0618(15)00283-4/h0290http://refhub.elsevier.com/S0950-0618(15)00283-4/h0290http://refhub.elsevier.com/S0950-0618(15)00283-4/h0295http://refhub.elsevier.com/S0950-0618(15)00283-4/h0295http://refhub.elsevier.com/S0950-0618(15)00283-4/h0295http://refhub.elsevier.com/S0950-0618(15)00283-4/h0305http://refhub.elsevier.com/S0950-0618(15)00283-4/h0305http://refhub.elsevier.com/S0950-0618(15)00283-4/h0305http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0315http://refhub.elsevier.com/S0950-0618(15)00283-4/h0315http://refhub.elsevier.com/S0950-0618(15)00283-4/h0325http://refhub.elsevier.com/S0950-0618(15)00283-4/h0325http://refhub.elsevier.com/S0950-0618(15)00283-4/h0325http://refhub.elsevier.com/S0950-0618(15)00283-4/h0330http://refhub.elsevier.com/S0950-0618(15)00283-4/h0330http://refhub.elsevier.com/S0950-0618(15)00283-4/h0335http://refhub.elsevier.com/S0950-0618(15)00283-4/h0335http://refhub.elsevier.com/S0950-0618(15)00283-4/h0340http://refhub.elsevier.com/S0950-0618(15)00283-4/h0340http://refhub.elsevier.com/S0950-0618(15)00283-4/h0340http://refhub.elsevier.com/S0950-0618(15)00283-4/h0345http://refhub.elsevier.com/S0950-0618(15)00283-4/h0345http://refhub.elsevier.com/S0950-0618(15)00283-4/h0345http://refhub.elsevier.com/S0950-0618(15)00283-4/h0350http://refhub.elsevier.com/S0950-0618(15)00283-4/h0350http://refhub.elsevier.com/S0950-0618(15)00283-4/h0355http://refhub.elsevier.com/S0950-0618(15)00283-4/h0355http://refhub.elsevier.com/S0950-0618(15)00283-4/h0360http://refhub.elsevier.com/S0950-0618(15)00283-4/h0360http://refhub.elsevier.com/S0950-0618(15)00283-4/h0365http://refhub.elsevier.com/S0950-0618(15)00283-4/h0365http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0370http://refhub.elsevier.com/S0950-0618(15)00283-4/h0365http://refhub.elsevier.com/S0950-0618(15)00283-4/h0365http://refhub.elsevier.com/S0950-0618(15)00283-4/h0360http://refhub.elsevier.com/S0950-0618(15)00283-4/h0360http://refhub.elsevier.com/S0950-0618(15)00283-4/h0355http://refhub.elsevier.com/S0950-0618(15)00283-4/h0355http://refhub.elsevier.com/S0950-0618(15)00283-4/h0350http://refhub.elsevier.com/S0950-0618(15)00283-4/h0350http://refhub.elsevier.com/S0950-0618(15)00283-4/h0345http://refhub.elsevier.com/S0950-0618(15)00283-4/h0345http://refhub.elsevier.com/S0950-0618(15)00283-4/h0345http://refhub.elsevier.com/S0950-0618(15)00283-4/h0340http://refhub.elsevier.com/S0950-0618(15)00283-4/h0340http://refhub.elsevier.com/S0950-0618(15)00283-4/h0340http://refhub.elsevier.com/S0950-0618(15)00283-4/h0335http://refhub.elsevier.com/S0950-0618(15)00283-4/h0335http://refhub.elsevier.com/S0950-0618(15)00283-4/h0330http://refhub.elsevier.com/S0950-0618(15)00283-4/h0330http://refhub.elsevier.com/S0950-0618(15)00283-4/h0325http://refhub.elsevier.com/S0950-0618(15)00283-4/h0325http://refhub.elsevier.com/S0950-0618(15)00283-4/h0325http://refhub.elsevier.com/S0950-0618(15)00283-4/h0315http://refhub.elsevier.com/S0950-0618(15)00283-4/h0315http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0310http://refhub.elsevier.com/S0950-0618(15)00283-4/h0305http://refhub.elsevier.com/S0950-0618(15)00283-4/h0305http://refhub.elsevier.com/S0950-0618(15)00283-4/h0295http://refhub.elsevier.com/S0950-0618(15)00283-4/h0295http://refhub.elsevier.com/S0950-0618(15)00283-4/h0295http://refhub.elsevier.com/S0950-0618(15)00283-4/h0290http://refhub.elsevier.com/S0950-0618(15)00283-4/h0290http://refhub.elsevier.com/S0950-0618(15)00283-4/h0285http://refhub.elsevier.com/S0950-0618(15)00283-4/h0285http://refhub.elsevier.com/S0950-0618(15)00283-4/h0285http://refhub.elsevier.com/S0950-0618(15)00283-4/h0280http://refhub.elsevier.com/S0950-0618(15)00283-4/h0280http://refhub.elsevier.com/S0950-0618(15)00283-4/h0275http://refhub.elsevier.com/S0950-0618(15)00283-4/h0275http://refhub.elsevier.com/S0950-0618(15)00283-4/h0270http://refhub.elsevier.com/S0950-0618(15)00283-4/h0270http://refhub.elsevier.com/S0950-0618(15)00283-4/h0265http://refhub.elsevier.com/S0950-0618(15)00283-4/h0265http://refhub.elsevier.com/S0950-0618(15)00283-4/h0260http://refhub.elsevier.com/S0950-0618(15)00283-4/h0260http://refhub.elsevier.com/S0950-0618(15)00283-4/h0260http://refhub.elsevier.com/S0950-0618(15)00283-4/h0255http://refhub.elsevier.com/S0950-0618(15)00283-4/h0255http://refhub.elsevier.com/S0950-0618(15)00283-4/h0255http://refhub.elsevier.com/S0950-0618(15)00283-4/h0250http://refhub.elsevier.com/S0950-0618(15)00283-4/h0250http://refhub.elsevier.com/S0950-0618(15)00283-4/h0250http://refhub.elsevier.com/S0950-0618(15)00283-4/h0245http://refhub.elsevier.com/S0950-0618(15)00283-4/h0245http://refhub.elsevier.com/S0950-0618(15)00283-4/h0240http://refhub.elsevier.com/S0950-0618(15)00283-4/h0240http://refhub.elsevier.com/S0950-0618(15)00283-4/h0230http://refhub.elsevier.com/S0950-0618(15)00283-4/h0230http://refhub.elsevier.com/S0950-0618(15)00283-4/h0225http://refhub.elsevier.com/S0950-0618(15)00283-4/h0225http://refhub.elsevier.com/S0950-0618(15)00283-4/h0220http://refhub.elsevier.com/S0950-0618(15)00283-4/h0220http://refhub.elsevier.com/S0950-0618(15)00283-4/h0220http://refhub.elsevier.com/S0950-0618(15)00283-4/h0210http://refhub.elsevier.com/S0950-0618(15)00283-4/h0210http://refhub.elsevier.com/S0950-0618(15)00283-4/h0205http://refhub.elsevier.com/S0950-0618(15)00283-4/h0205http://refhub.elsevier.com/S0950-0618(15)00283-4/h0200http://refhub.elsevier.com/S0950-0618(15)00283-4/h0200http://refhub.elsevier.com/S0950-0618(15)00283-4/h0200http://refhub.elsevier.com/S0950-0618(15)00283-4/h0195http://refhub.elsevier.com/S0950-0618(15)00283-4/h0195http://refhub.elsevier.com/S0950-0618(15)00283-4/h0190http://refhub.elsevier.com/S0950-0618(15)00283-4/h0190http://refhub.elsevier.com/S0950-0618(15)00283-4/h0185http://refhub.elsevier.com/S0950-0618(15)00283-4/h0185http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0180http://refhub.elsevier.com/S0950-0618(15)00283-4/h0180http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0175http://refhub.elsevier.com/S0950-0618(15)00283-4/h0175http://refhub.elsevier.com/S0950-0618(15)00283-4/h0175http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0170http://refhub.elsevier.com/S0950-0618(15)00283-4/h0170http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0165http://refhub.elsevier.com/S0950-0618(15)00283-4/h0165http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0160http://refhub.elsevier.com/S0950-0618(15)00283-4/h0160http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0155http://refhub.elsevier.com/S0950-0618(15)00283-4/h0155http://refhub.elsevier.com/S0950-0618(15)00283-4/h0155http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0150http://refhub.elsevier.com/S0950-0618(15)00283-4/h0150http://refhub.elsevier.com/S0950-0618(15)00283-4/h0150http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0145http://refhub.elsevier.com/S0950-0618(15)00283-4/h0145http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0140http://refhub.elsevier.com/S0950-0618(15)00283-4/h0140http://refhub.elsevier.com/S0950-0618(15)00283-4/h0140http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0135http://refhub.elsevier.com/S0950-0618(15)00283-4/h0135http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0130http://refhub.elsevier.com/S0950-0618(15)00283-4/h0130http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0125http://refhub.elsevier.com/S0950-0618(15)00283-4/h0125http://refhub.elsevier.com/S0950-0618(15)00283-4/h0125http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0120http://refhub.elsevier.com/S0950-0618(15)00283-4/h0120http://refhub.elsevier.com/S0950-0618(15)00283-4/h0120http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0115http://refhub.elsevier.com/S0950-0618(15)00283-4/h0115http://refhub.elsevier.com/S0950-0618(15)00283-4/h0115http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0110http://refhub.elsevier.com/S0950-0618(15)00283-4/h0110http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://refhub.elsevier.com/S0950-0618(15)00283-4/h0105http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0100http://refhub.elsevier.com/S0950-0618(15)00283-4/h0100http://-/?-http://refhub.elsevier.com/S0950-0618(15)00283-4/h0095http://refhub.elsevier.com/S0950-0618(15)00283-4/h0095http://refhub.elsevier.com/S0950-0618(15)00283-4/h0095http://-/?-http://refhub