Construction and Building Materials 77 (2015) 370–395Contents lists available at ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat Review An overview on the influence of various factors on the properties of geopolymer concrete derived from industrial by-products Part Wei Ken ⇑, Mahyuddin Ramli, Cheah Chee Ban School of Housing, Building and Planning, Universiti Sains Malaysia, 11800 Penang, Malaysia h i g h l i g h t s A review on the latest trends of geopolymer concrete derived from industrial wastes. A critical review on the various influences on the properties of geopolymer concrete. Potential solution for environmental and waste disposal issues in various industries. a r t i c l e i n f o a b s t r a c t Article history: The enormous amount of industrial waste ash generated by power generation industry, timber manufac- Received 9 May 2014 turing industry, iron and steel making industry, rice milling industry, mining industry etc have posed the Received in revised form 21 October 2014 aforementioned industry players a great challenge when it comes to the disposal of these ash materials Accepted 27 December 2014 due to the environmental, health, scarcity of lands and other issues. The best approach in overcoming the Available online 12 January 2015 aforementioned waste management problems is to promote large volume recycling/reuse of these waste materials. In recent years, the rapid growth in research and development related to geopolymer binders Keywords: has indeed indicated that the use of geopolymer offers the greatest potential in solving not only the waste Waste management Geopolymer concrete management problems related to the aluminosilicate solid waste materials generated from various Sustainability industries, but also the environmental degradation related to the use of OPC as primary binder material Recycling in the construction industry. Results of recent studies are indicative that geopolymer concrete fabricated using various industrial by-products exhibited similar or better mechanical, physical and durability prop- erties as compared to OPC concrete. This paper presents a concise review of the current studies on the utilization of industrial by-products as the primary binder materials in the fabrication of geopolymer con- crete. The effects of a number of major factors such as the use of chemical activator, post fabrication cur- ing regime, particle size distribution of source materials, and aggressive environment exposure on the mechanical strength, physical properties, microstructures and durability properties of the geopolymer concrete are exhaustively deliberated. Besides, the current material design, fabrication procedures and post fabrication treatment procedures were rigorously reviewed to identify the limitations of the current geopolymer technology which impede its wide implementation in the construction industry. It has been identified that the high alkaline content in the material design and requirement for elevated temperature treatment of the contemporary geopolymeric binder are among the major technical challenges which resulted in the limited use of the material in the construction industry. Based upon that, numerous strat- egies were proposed to overcome the current limitations of the geopolymer technology towards promot- ing a large scale implementation of the technology in the production of construction materials. Ó 2015 Elsevier Ltd. All rights reserved. Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 2. Effect of chemical activators and curing regime on the mechanical, durability, shrinkage, microstructure and physical properties of geopolymer 372 2.1. Mechanical properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372 ⇑ Corresponding author. Tel.: +60 0164871298. E-mail address:
[email protected] (W.K. Part). http://dx.doi.org/10.1016/j.conbuildmat.2014.12.065 0950-0618/Ó 2015 Elsevier Ltd. All rights reserved. W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 371 2.2. Dimensional stability and durability properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 2.3. Microstructure of geopolymer matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376 2.4. Rheological and physical properties of geopolymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378 3. Effect of particle size distribution of binder phase and additives on the properties of geopolymer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 3.1. Mechanical properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 3.2. Rheological and physical properties of geopolymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382 3.3. Microstructure of geopolymer matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382 3.4. Fourier transform infrared spectroscopy (FTIR) analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 4. The effect of aggressive environmental exposure on properties of geopolymers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 4.1. Mechanical properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 4.2. Microstructure analysis of geopolymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 4.3. Fourier transform infrared spectroscopy (FTIR) analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 4.4. Thermogravimetry (TGA) analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 4.5. Physical properties of geopolymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 5. The effect of water content and forming pressure on the properties of geopolymers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388 5.1. Mechanical properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 5.2. Water absorption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 6. Blended geopolymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390 6.1. Mechanical properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390 6.2. Microstructure of geopolymer matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 6.3. Dimensional stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 7. Summary of the current body of knowledge and identification of challenges faced in future development of geopolymer technology for industrial applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 8. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394 1. Introduction followed by coagulation/gelation; (3) polycondensation to form 3D network of silico-aluminates structures [15]. Based on the types Ordinary Portland cement (OPC) has long been the traditional of resultant chemical bonding, three types of structures can be and widely used binder material in the manufacture of concrete. derived from the 3D aluminosilicate network: poly(sialate) However, the use of OPC as primary construction material has been (–Si–O–Al–O–), poly(sialate-siloxo) (Si–O–Al–O–Si–O) and questioned extensively over the last decades due to the environ- poly(sialate-disiloxo) (Si–O–Al–O–Si–O–Si–O–) [16]. mental impact of clinker production [1,2]. In fact, the production The potential of geopolymer binders to replace the traditional of Portland cement clinker from the cement production plants OPC binders was supported by the fact that there is abundant of worldwide emit up to 1.5 billion tons of CO2 annually, which industrial by-products generated in various industries that was accounts for around 5% of the total man-made CO2 emission and found to be suitable to use as geopolymer source materials, all of if the undesirable trend continues, the figure will rise to 6% by year which are causing problems in term of finding an ideal solution 2015 [3–5]. Apart from OPC, sand and aggregate are also the main for disposal purposes. For instance, pulverized fuel ash (PFA) or constituent source materials in the production of concrete, which more commonly known as fly ash (FA), an industrial by-products originated from the quarrying operations which are both energy of coal burning power plant industry, makes up of 75–80% of global intensive and produces high level of waste materials. Shortage of annual ash production [17], yielded geopolymer concrete with natural resources for construction materials in many developing superior mechanical and durability properties as compared to countries has also led to long distance haulage and thus signifi- OPC concrete [18–20]. GGBFS, by-products of iron pig manufacture cantly increased the production cost of construction materials. from iron ore, has also found significant use in the production of All of the issues mentioned above are against the context of sus- high strength geopolymer concrete [21,22]. The use of palm oil fuel tainable development in construction industry and immediate ash (POFA), waste materials derived from the burning of empty remedy actions must be taken to ensure sustainability in the con- fruit brunches, oil palm shells and oil palm clinker from the oil struction industry [1]. palm industry to generate electricity as geopolymer binder, has The aforementioned issues prompted various researches in an gathered pace in recent years. POFA is widely used as geopolymer attempt to reduce the global carbon footprint ranging from utilizing binder especially in oil palm-rich country such as Malaysia and supplementary cementitious materials (SCMs) as partial cement Thailand due to its increasing amount which rendered the disposal replacement materials [6–10] to developing a whole new cement- method in the mean of landfilling not feasible [4,11,23]. Other less binder, namely geopolymers [11–14]. Geopolymers are alter- industrial by products, for examples rice husk ash (RHA) from native cementitious materials synthesized by combining source the rice milling industry, red mud (RM) from the alumina refining materials which are rich in silica and alumina such as fly ash (FA), industry, copper and hematite mine tailings from the mining ground granulated blast furnace slags (GGBFS) with strong alkali industry etc [14,20,24,25] has also find considerable interest in solutions such as potassium hydroxide (KOH), sodium hydroxide the fabrication of geopolymer concrete. (NaOH) and soluble silicates (in most cases) such as sodium silicate The ever present problem in reducing the use of OPC in con- where the dissolved Al2O3 and SiO2 species undergoes geopolymer- struction industry, coupled with the problems in disposing indus- ization to form a three-dimensional amorphous aluminosilicate trial by-product in various industries, geopolymer binder certainly network with strength similar or higher than that of OPC concrete. has all the potential to replaces OPC as the binder in construction Generally, the mechanism of geopolymerization can be divided into industry. Thus, it is the aim of this paper to review the current three main stages: (1) Dissolution of oxide minerals from the source trends in geopolymer concrete, focusing solely on geopolymeric materials (usually silica and alumina) under highly alkaline condi- binders based on industrial waste materials, along with the various tion; (2) transportation/orientation of dissolved oxide minerals, effects such as chemical activators, curing regime, additives, 372 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 fineness of the binder phases, aggressive environment exposure etc husk ash (RHA)-based geopolymer concluded that higher concen- towards the mechanical, physical, durability and microstructural tration of NaOH had resulted in a decrease in compressive strength properties of the geopolymer concrete as the leveraging of indus- of geopolymer. The possible reasons for the contrasting trends trial waste in developing a whole new binder to replace OPC suited could be attributed to, (i) the high viscosity of NaOH solution due perfectly well in the context of sustainability in cement and con- to the higher concentration distrupts the leaching of Si and Al ions, crete industry. (ii) excessive OH- concentration results in premature precipitation of geopolymeric gels, (iii) the partially reacted/unreacted RHA par- 2. Effect of chemical activators and curing regime on the ticles due to the incomplete dissolution of Si and Al species caused mechanical, durability, shrinkage, microstructure and physical the deterioration in mechanical properties of geopolymer properties of geopolymer produced. Eco-friendly geopolymer bricks were fabricated by Ahmari and The presence of chemical activators such as sodium/potassium Zhang [24] using solely copper mine tailings (MT) and NaOH as hydroxide and soluble silicates in the mix design of geopolymer chemical activator. NaOH concentrations were varied from 10 to has a significant effect on the properties of geopolymers. The 15 M to study the effect of NaOH concentration on the unconfined effects of the addition of chemical activators on the mechanical, compressive strength (UCS) of cured geopolymer bricks. The durability, shrinkage, microstructure and physical properties of authors reported that the UCS of 15 M NaOH specimens is higher geopolymers are deliberated in details as follows. than the 10 M NaOH counterparts for all the mixtures due to the higher NaOH/MT ratio which consequently resulted in a higher 2.1. Mechanical properties Na/Al and Na/Si ratios. This in turn produced a much thicker geo- polymer binder gels which bind the unreacted particles and Chemical activator or alkali activator solution plays a vital role directly contributed to the UCS of geopolymer bricks. in the initiation of the geopolymerization process. Generally, a According to Komljenovic et al. [31], the nature and concentra- strong alkaline medium is necessary to increase the surface hydro- tion of alkali activator is the most dominant parameter in alkali lysis of the aluminosilicate particles present in the raw material activation process. The authors utilized five different type of alkali while the concentration of the chemical activator has a pro- activators i.e. Ca(OH)2, NaOH, NaOH + Na2CO3, KOH and Na2SiO3 nounced effect on the mechanical properties of the geopolymers with various concentrations to fabricate FA-based geopolymer [26,27]. On the other hand, the dissolution of Si and Al species dur- mortars. The curing condition was made constant in order to effec- ing the synthesis of geopolymer is very much dependent on the tively study the influence of types of alkali activators to the concentration of NaOH, where the amount of Si and Al leaching mechanical properties of the geopolymer mortars. Based on the is mostly governed by the NaOH concentration and also the leach- compressive strength results, the alkali activator which possesses ing time [28]. Gorhan and Kurklu [18] investigated the influence of highest activation potential i.e. highest compressive strength was the NaOH solution on the 7 days compressive strength of ASTM Na2SiO3, followed by Ca(OH)2, NaOH, NaOH + Na2CO3 and KOH. Class F FA geopolymer mortars subjected to different NaOH con- The lower activation potential of KOH compared to NaOH was centrations. Three different concentrations of NaOH (3, 6 and due to the difference in ionic diameter difference between sodium 9 M) were used throughout the laboratory work while other and potassium. Regardless of the types of alkali activators used, the parameters such as sand/ash ratio and sodium silicate/NaOH compressive strength generally increased with the increase in acti- (SS/SH) ratio was maintained constant. Based on the compressive vator’s concentration. The authors also concluded the optimum strength results acquired, the optimum NaOH concentration that value of Na2SiO3 modulus was 1.5. Anything higher than the pre- produced highest 7 days compressive strength of 22.0 MPa is scribed modulus will causes deleterious effect to the compressive 6 M. In the aforementioned concentration, an ideal alkaline envi- strength of geopolymer mortars. ronment was provided for proper dissolution of FA particles and The dissolution, hydrolysis and condensation reaction of geo- at the same time polycondensation process was not hindered. polymers are greatly affected by the effective Si/Al ratios. In low When the NaOH concentration is too low at 3 M, it is not sufficient Si/Al ratio geopolymer system, the condensation reaction tends to stimulate a strong chemical reaction while excessively high con- to occurs between aluminate and silicate species, thus resulting centration of NaOH (9 M) resulted in premature coagulation of sil- in mainly poly(sialate) geopolymeric structures. On the other hand, ica which in both cases culminated in lower strength mortars. condensation reaction in high Si/Al system would results in pre- In the other study, Somna et al. [29] studied the compressive dominantly between the silicate species itself, forming oligomeric strength of Ground FA (GFA) cured at ambient temperature by vary- silcates which in turn condense with Al(OH4)4 and forms geopoly- ing the NaOH concentration from 4.5 to 16.5 M. Results showed meric structures of poly(sialate-siloxo) and poly(sialate-disiloxo) that by increasing NaOH concentrations from 4.5 to 9.5 M, signifi- [15,32]. cant increase in the compressive strength of paste samples can be Sukmak et al. [33] studied the effect of sodium silicate/sodium observed. While the variation of NaOH concentrations from 9.5 to hydroxide (Na2SiO3/NaOH) and liquid alkaline activator/FA (L/FA) 14 M also increase the compressive strength of paste samples, but ratios on the compressive strength development of clay-FA geo- in a much lesser extent. The increase in compressive strength with polymer bricks under prolonged curing ages. The Na2SiO3/NaOH the increasing NaOH concentrations is mainly due to the higher ratios used were 0.4, 0.7, 1.0, 1.5 and 2.3 while the L/FA ratios used degree of silica and alumina leaching. The compressive strength were 0.4, 0.5, 0.6 and 0.7 by dry clay mass. The clay-FA geopolymer of GFA hardened pastes start to decline at the NaOH concentrations bricks were compressed using hand-operated hydraulic jack at the of 16.5 M. This decrease in compressive strength is mainly attrib- optimum water content (OWC) to attain maximum dry unit uted to the excess hydroxide ions which caused the precipitation weight. The bricks were left to set at room temperature for 24 h of aluminosilicate gel at very early ages, thus resulting in the forma- before being subjected to oven curing at 75 °C for 48 h. The com- tion of lower strength geopolymers. pressive strength tests were performed on the 7, 14, 28, 60 and While many research papers reported enhancement in compres- 90 days of curing ages. Results showed that L/FA ratios of less than sive strength with the increase in the concentration of chemical 0.3 and greater than 0.8 are not suitable for fabrication of clay-FA activators particularly NaOH [18,24,29,30], some research shows geopolymer bricks as the strength is null at the aforementioned a total contrast in compressive strength development. For example, L/FA ratios. The optimum values of Na2SiO3/NaOH and L/FA ratios a study done by He et al. [14] which focused on red mud (RM)/rice are 0.7 and 0.6, respectively. The optimum Na2SiO3/NaOH ratio of W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 373 0.7 is less than that of FA-based geopolymers as the clay possesses Besides the SS/SH ratio, it is also very important to study the high cation absorption capability and may have absorbed some of effect of silica modulus (Ms) of the activators and its relationship the NaOH added. The significant decrease in strength for clay-FA with SS/SH ratio in order to maximize the strength and also the geopolymer bricks with excessive alkali activator (L/FA > 0.6) was economy aspect in the synthesis of alkali-activated binders claimed to be due to the precipitation of dissolved Si and Al species [37,38]. Ms determined the amount of soluble silicates present at the early stages before the initiation of polycondensation pro- and is crucial in controlling the dissolution rate and also the gela- cess, resulting in formation of cracks on the FA particles. The max- tion process during geopolymerization, which in turns have a sig- imum compressive strength attained at the aforementioned nificant effect towards the strength development of the hardened optimized specimens was approximately 15 MPa at the age of geopolymer mixes. However, the appropriate Ms varied for differ- 90 days. On the other hand, Ridtirud et al. [34] reported the opti- ent geopolymer system comprising of different source materials, mum SS/SH ratio for FA based geopolymer mortar to be 1.5. FA geo- i.e. different chemical composition, indicating the need to under- polymer mortars with SS/SH ratios of 0.33, 0.67, 1.0, 1.5 and 3.0 stand the suitable Ms for each category of geopolymer. For achieved compressive strength of 25.0, 28.0, 42.0, 45.0 and instance, Guo et al. [39] attempted to evaluates the influence of 23.0 MPa respectively. The increasing trend of compressive the Ms and the content of alkali activator on the compressive strength is mainly attributed to the increasing Na content in the strength of CFA-based geopolymer. Mixtures of sodium silicate mixture where Na+ ion plays a critical role in the formation of geo- and sodium hydroxide were used as the alkali activator for the polymer by acting as a charge balancing ions. However, excessive CFA geopolymer. The silica-alkali modulus of the alkali activator silicate in the geopolymer system reduces its compressive strength was varied from 1.0 up to 2.0 while the content of alkali activator as excess sodium silicate hampered the evaporation of water and were based on the mass proportion of Na2O to CFA and ranged also disrupted the formation of three dimensional networks of alu- between 5% and 15%. From the results acquired, both silica-alkali minosilicate geopolymers. modulus and content of alkali activator were proven to be equally Various factors such as liquid alkaline/ash ratio, SS/SH ratio and crucial for the strength development of CFA geopolymers. The opti- NaOH concentration on the compressive strength of Ground Bot- mum modulus and content of alkali activator values were found to tom Ash (GBA) geopolymer mortars were reported by Sathonsaow- be 1.5% and 10%, respectively which yielded 3, 7 and 28 days com- aphak et al. [30]. Besides, water and naphthalene-based pressive strength values of 22.6, 34.5 and 59.3 MPa, respectively superplasticizer (NCP) were also incorporated into the mortar when cured under normal room temperature (23 °C). mixes in an effort to improve the workability while maintaining Law et al. [38] suggested optimum Ms for class F FA based geo- the strength of the mortars. Results shown that the liquid alka- polymer concrete is 1.0, where further increase in Ms does not line/ash ratio, SS/SH ratio and NaOH concentration values of bring about any significant increase in compressive strength. The 0.4209–0.709, 0.67–1.5 and 10 M respectively, produced GBA geo- authors suggested that at Ms > 1.0, either all the FA particles have polymer mortars with high compressive strength and desired been dissolved, or any increase in Ms beyond 1.0 does not results in workability. The authors also stressed that the addition of 10 M further dissolution of the protective crust on the FA particles devel- NaOH solution is essential to the geopolymerization as the Na+ ions oped from the precipitation of the geopolymerization reaction act as charge-balancing ions while the NaOH solution increases the products. Yusuf et al. [37] found that the strength of alkaline acti- dissolution rate of silica and alumina. While for palm oil fuel ash vated ground steel slag/ultrafine palm oil fuel ash (AAGU) with (POFA) geopolymers, the optimum solid to liquid ratios and SS/ various silica modulus (Ms) ranging from 0.915 to 1.635 to be SH ratios to achieve highest compressive strength were reported insignificant, with Ms = 0.915 yielded compressive strength of to be 1.32 and 2.5 respectively [35]. Higher presence of voids in 69.13 MPa while Ms = 1.635 yielded compressive strength of solid to liquid ratio <1.32 will adversely affect the compressive 65 MPa. strength. Also, SS/SH ratios >2.5 will cause excessive sodium sili- Generally, compressive strength can be correlated with the cate which hindered the geopolymerization process. modulus of elasticity where higher degree of geopolymerization A proper adjustment of SiO2/Al2O3 ratio in geopolymers by brings about denser geopolymer matrix, which in turns resulted hybridizing two different source of aluminosilicate source material in higher compressive strength and modulus of elasticity [40]. and adjustment in hybridization ratios can improve the compres- Chemical activator has a pronounced effect on the compressive sive strength of geopolymers [14,20,36]. Nazari et al. [20] proposed strength development of geopolymer concrete. However, other an innovative approach for large volume recycling of rice husk-bark mechanical property such as modulus of elasticity of the resultant ash (RHBA) which is a solid waste generated by biomass power geopolymer concrete does not depend entirely on the chemical plants using rice husk and eucalyptus bark as source of fuel. The activator dosage [41]. It was found that modulus of elasticity was proposed approach requires the blending of RHBA (high silica governed by the amount of aggregate present in the geopolymer source) with FA (FA) in order to modify the chemical composition concrete mixtures. A proper adjustment in the total aggregate con- of the resultant geopolymer. Sodium silicate and varying concen- tent and also the ratio of fine aggregate to total aggregate can tration of NaOH (4, 8 and 12 M) were used as chemical activators result in geopolymer concrete with equal or higher modulus of for solidification and stabilization of the RHBA-FA blend. SS/SH ratio elasticity in compared with OPC concrete [17]. In separate studies, and chemical activator to FA-RHBA mixture were fixed at 2.5 and it was reported that the high silicate content might increase the 0.4 respectively. RHBA was added to the mixture at replacement elasticity of geopolymer concrete subsequently resulted in a lower levels of 20, 30 and 40% by total binder weight. The specimens were modulus of elasticity than OPC concrete [42,43]. Topark-Ngarm tested after being subjected to 80 °C oven curing for 36 h upon pre- et al. [40] reported high calcium FA geopolymer concrete can curing duration of 24 h. The authors concluded that at all FA exhibits similar or higher modulus of elasticity with a lower replacement levels, the compressive strength of FA-RHBA geopoly- SS/SH ratio, corresponding to higher amount of Na2O. mer increased proportionally with the concentration of NaOH. It is well known that conventional geopolymers required heat Moreover, RHBA-FA geopolymer with FA replacement level of 30% treatment in order to attain similar or higher compressive strength using RHBA at any concentration of NaOH exhibited the highest in comparison with OPC concrete [14,22,23,44,45]. Heat treatment compressive strength among the various geopolymer mixes exam- is beneficial towards the dissolution and geopolymerization of alu- ined. The compressive strength of geopolymer with the aforemen- minosilicate gel which results in high early strength gain [46]. It tioned FA replacement level by RHBA and varying NaOH also helps in accelerating the dissolution of silica and alumina spe- concentrations was found to range between 20 and 30 MPa. cies and the subsequent polycondensation process. However, the 374 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 heat curing regime applied must be appropriate in such a way that cured at higher temperature i.e. 90 °C start to decrease after certain it provides an ideal condition for the proper dissolution and precip- period of time. This is because prolonged curing at high tempera- itation of dissolved silica and alumina species. Geopolymerization ture could destroy the granular structure of geopolymer. High tem- might be hindered upon exceeding certain temperature and heat perature curing had also resulted in dehydration of the geopolymer treatment period, depending on the source material, which in turns matrix and subsequently excessive shrinkage due to contraction of adversely affect the mechanical properties of the geopolymers the polymeric gel, (iii) compressive strength as high as 58.9 MPa [14,20,24]. Different type of heat treatment and curing environ- after 28 days of curing was achieved for the geopolymer mixture ment such as steam and autoclave curing [21], saline water curing with FA replacement level of 30% cured at 80 °C for 36 h. [19] and even microwave-assisted curing regime [47] have been Giasuddin et al. [19] discovered the potential of geopolymeric employed in order to maximize the potential and capacity of binder in replacing traditional oil well cements in CO2 geo-seques- geopolymerization. tration in saline aquifer. The slag-added FA based geopolymer were He et al. [14] studied the effect of curing period on the compres- cured in three different curing environments i.e. water curing, sal- sive strength of geopolymer paste derived from two industrial ine water curing (8% and 15% concentration) and sealed curing. API wastes, namely red mud (RM) and rice husk ash (RHA). The geo- recommended class G oil well cement was fabricated for bench- polymer paste was activated using 4 M NaOH with the constant marking purposes. From the 28 days compressive strength results, solution to solid weight ratio of 1.2. The RHA/RM ratio was fixed it can be concluded that, (i) saline water cured geopolymer speci- at 0.4. Upon casting, the geopolymer pastes specimens were left mens exhibited higher compressive strength as compared to water to cured at room temperature and atmospheric pressure for the cured geopolymer specimens while the oil well cement specimens periods of 14, 28, 35, 42 and 49 days before being subjected to showed a contrasting trends as compared to their geopolymer compression test. The following conclusions were derived from counterpart specimens, (ii) for geopolymer specimens, the com- the experimental study, (i) a near constant compressive strength pressive strength of specimens cured in sealed condition is consis- value of 11.7 MPa was obtained at the testing age of 35 days, tently higher as compared to water cured and saline water cured which implies that the geopolymer pastes specimens have only specimens, (iii) the variation in compressive strength with differ- achieved complete geopolymerization at that particular time ent curing environments was attributed to the leaching out of frame, (ii) the slower rate of red mud and RHA geopolymerization alkali activator solution and other useful reactant from the geo- compared with other geopolymer source materials such as FA and polymer specimen to the surrounding curing medium. metakaolin-based geopolymer is due to the dominant crystalline Aydin and Baradan [21] studied the effect of steam and auto- solid phases in RM which acted as unreactive fillers, the larger par- clave curing on the compressive strength of alkali activated slag ticles size of RHA which slows down the rate of dissolution and (AAS) mortars. In this study, ground granulated blast furnace slag also the considerable amount of impurities present in both RM (GGBFS) with low hydration modulus (HM) of 1.33 was used. The and RHA which may cause deleterious effect on the geopolymer- slag was activated using the mixture of NaOH and Na2SiO3 in dif- ization rate. ferent proportions to obtain different Ms values and different An experimental study on the effect of curing temperature on Na2O content. Aggregate to binder (cement or GGBFS) and water the unconfined compressive strength (UCS) of copper mine tailings to binder ratio were fixed at 2.75 and 0.44 respectively for all the (MT)-based geopolymer bricks showed that the optimum curing mixtures. One batch of specimens were kept in humidity cabinet temperature which produced the highest UCS geopolymer bricks for 5 h before being subjected to steam curing at 100 °C for 8 h. is 90 °C. Any temperature higher than that will causes a drastic Another batch of specimens was kept in humidity cabinet for drop in UCS of the geopolymer bricks. The authors concluded that 24 h and subsequently autoclaved at 210 °C and under 2.0 MPa too high a temperature will causes rapid polycondensation process pressure for 8 h. The compressive strength results of AAS mortars and excessive early formation of geopolymeric gels which will hin- were then compared with a standard Portland cement (PC) mortar. der the dissolution of unreacted silica and alumina species. Also, The following conclusions were derived: (i) Compressive strength excessively high curing temperature will cause rapid evaporation values in the range of 15–90 MPa were achieved by steam cured of pore solutions and may result in an incomplete geopolymeriza- AAS mortars, despite using low HM GGBFS, (ii) compressive tion. The optimum curing temperature yielded geopolymer bricks strength values of steam-cured AAS mortars were significantly with UCS of approximately 15 MPa. [24]. On the other hand, Ridti- higher than the PC mortars when the Na2O and Ms values are rud et al. [34] concluded that higher curing temperature i.e. 60 °C higher than 4% and 0.4, respectively, (iii) high performance AAS give rise to a rapid strength development during the early ages mortars with compressive strength value of 70 MPa can be of curing i.e. 7–28 days. Upon 28 days of curing, the strength devel- achieved using a mere 2% Na2O (by weight of slag) under autoclave opment was deemed insignificant for higher curing temperature. curing, (iv) compressive strength of autoclaved PC mortar was The initial curing at elevated temperature accelerates the geopoly- found to be significantly lower than their steam-cured AAS coun- merization reaction and thus the strength of the resultant geopoly- terparts due to the formation of crystalline structure a-calcium sil- mer was improved. icate hydrate (a-C2SH) under high temperature and pressure The effect of curing temperature on the compressive strength of which caused the increase in porosity and reduction in compres- FA-rice husk bark ash (RHBA) geopolymer was studied by Nazari sive strength, (v) autoclave curing was found to be more favorable et al. [20]. FA was replaced by RHBA at three replacement levels for activator solution with low Na2O concentrations and low Ms of 20%, 30% and 40% by total binder weight. The FA-RHBA geopoly- ratios while steam curing is more favorable for activator solution mer mixtures were activated by the sodium silicate and NaOH with with high Ms ratios. sodium silicate/NaOH of 2.5, NaOH concentration of 12 M and Chindaprasirt et al. [47] proposed a method of reducing the heat alkaline activator/ash ratio of 0.4. A precuring period of 24 h was treatment period of high calcium FA geopolymer paste. The results allowed upon casting to increase the homogeneity of geopolymeric showed that by subjecting the paste sample to microwave heating materials before the application of heat curing. After the precuring of 5 min plus conventional oven curing for 6 h at 60 °C, the period, the geopolymer samples were subjected to 50–90 °C oven compressive strength obtained was higher if compared to paste curing for 36 h. From the results of compressive strength acquired, samples cured at 60 °C for 24 h without microwave treatment. (i) the optimum curing temperature for all the mixtures at both 7 Tables 1 and 2 summarize the effect of chemical activator and and 28 days of curing is 80 °C, (ii) compressive strength of samples curing regime on the mechanical properties of geopolymers. W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 375 2.2. Dimensional stability and durability properties However, the variations in Ms was found to have no significant effect on the chloride diffusivity of class F FA based geopolymer The effect of chemical activator (SS/SH ratio, NaOH molarity etc) concrete and are comparable to those of OPC concrete [38]. The and curing regime on the dimensional stability and durability authors predicted that the long term chloride resistance of geo- properties of geopolymers has an indirect relationship with the polymer concrete to be lower if compared with OPC concrete mechanical properties exhibited by the geopolymers. Generally, owing to the lower strength increment over time of the geopoly- geopolymers with excellent mechanical properties will exhibit mer concrete. The attempted investigation of long term chloride superior dimensional stability and durability properties if com- resistance of geopolymer concrete by the authors using rapid chlo- pared with geopolymers with inferior mechanical properties ride permeability test (RCPT) was proven to be futile as the geo- [21,34,38,47]. polymer specimens exhibited rapid rise in temperature during Ridtirud et al. [34] investigated the effect of NaOH concentra- testing which is against the Ohm’s law, implying that RCPT is not tion and sodium silicate to NaOH (SS/SH) ratio on the shrinkage a suitable testing method to evaluate the chloride resistance of of ASTM class C FA geopolymer mortars. In order to determine geopolymer concrete. the influence of NaOH concentration, S/N ratio of 0.67 and NaOH Aydin and Baradan [21] studied the effect of steam curing and concentrations of 7.5, 10.0 and 12.5 M were used. While for the autoclave curing on the drying shrinkage of alkali activated slag influence of S/N ratio, geopolymer containing 10 M of NaOH and (AAS) mortars. A standard OPC mortar was fabricated and used SS/SH ratios of 0.33, 0.67, 1.0, 1.5 and 3.0 were examined. Sand for comparison with all other heat cured AAS mortars. Besides, a to FA ratio of 2.75:1, liquid to ash ratio of 0.6 and curing tempera- group of OPC and AAS mortars were cured in standard water con- ture of 40 °C were used for the aforementioned series of tests. The dition and used as comparison purposes. Upon heat treatment, the authors concluded that the shrinkage of FA geopolymer mortars specimens were left to be cooled to room temperature and the increased with higher NaOH concentration mainly due to the low length changes of the specimens were measured periodically up strength of the corresponding mortar samples. On the other hand, to 6 months. The authors concluded that: (i) drying shrinkages the increase in the SS/SH ratio produces geopolymer with signifi- value of AAS mortars were higher as compared to PC mortars in cantly lower values of shrinkage as the high silica to alumina ratio all curing conditions, (ii) drying shrinkage values of AAS and PC with high S/N ratio gives rise to rapid geopolymerization reaction mortars were reduced upon heat curing, (iii) reduction of drying or condensation of geopolymer. Thus, the increase in shrinkage shrinkage values of AAS mortars upon being subjected to heat cur- was generally associated with the low strength development of ing is more significant as compared to heat-cured PC mortars, (iv) geopolymers. The authors also studied the effect of curing temper- generally, autoclave curing was found to be more effective in ature on the shrinkage properties of ASTM class C FA geopolymer reducing the drying shrinkage of AAS and PC mortars than steam mortars. The results show that FA geopolymer mortars subjected curing. to higher curing temperature i.e. 40 and 60 °C had undergone sig- The sulfate and acid resistance of high calcium FA geopolymer nificantly lower degree of shrinkage as compared geopolymer mor- paste was greatly enhanced by using microwave-assisted heat cur- tars cured at ambient temperature i.e. 23 °C. The lower shrinkage ing method to cure the paste sample. Microwave radiation is at higher curing temperature is associated with the higher strength thought to enhance the dissolution rate of Si and Al species from value as a result of accelerated heat curing for 24 h. The authors the FA particles by the rapid and uniform heating of the aqueous also highlighted the significantly low shrinkage value after 3 weeks alkaline solution by microwave energy, leading to multiple gel for samples cured at all three temperature. This shows that the formation and subsequently a denser, stronger and durable geopolymer mortars have high potential for commercialization matrix was formed if compared to the conventional oven heat cur- especially for the precast structural element industry. ing [47]. Table 1 Effect of chemical activator on the compressive strength of geopolymer concrete. Types of geopolymer Chemical activator Compressive Primary findings strength SS/SH NaOH Ms ratio concentration FA-based Gorhan & Kurklu [18] 0.4–2.3 3–9 M 12–23 MPa 6 M NaOH optimal Sukmak et al. [33] 10 M 4–14 MPa SS/SH ratio of 0.7 optimal Somna et al. [29] 0.33–3.0 4.5–16.5 M 7–25 MPa Strength increased from 4.5–14 M NaOH, but decreased at 16.5 M NaOH Ridtirud et al. [34] 7.5–12.5 M 25–45 MPa SS/SH ratio of 1.5 and 7.5 M NaOH optimal Guo et al. [39] 1.0–2.0 5.0–63.4 MPa Ms of 1.5 optimal Law et al. [38] 10 M 0.75–1.25 39–57.3 MPa Ms of 1.0 optimal RHA/RHBA-based He et al. [14] 2.5 2–6 M 8–15 MPa 2 M NaOH optimal Nazari et al. [20] 4–12 M 20–30 MPa 12 M NaOH optimal Songpiriyakij et al. [48] 0.5–2.5 14 & 18 M 34–56 MPa 18 M NaOH optimal Detphan and Chindaprasirt 1.9–5.5 15–40 MPa SS/SH ratio of 4.0 optimal [49] POFA-based Salih et al. [35] 0.5–3.0 10 M 7–32 MPa SS/SH ratio of 2.5 optimal Yusuf et al. [37] 10 M 0.915– 65–69 MPa Ms of 0.915 optimal but insignificant 1.635 MT-based Ahmari and Zhang [24] 10–15 M 4–34 MPa 15 M NaOH optimal 376 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 Table 2 Effect of curing regime on the compressive strength of geopolymer concrete. Types of Curing regime Compressive Primary findings geoplymer strength FA based Giasuddin 10–12 h room temperature curing upon casting, followed by saline water, 49–91 MPa Sealed condition curing optimal, followed by saline et al. [19] normal water and sealed condition curing water and normal water curing Nazari et al. 24 h precuring period after casting followed by 36 h 50–90 °C oven curing 49–60 MPa 80 °C oven curing optimal [20] Ridtirud et al. 25, 40 and 60 °C curing for 24 h after 1 h of precuring period 22–53 MPa 60 °C oven curing optimal (applicable for 7 and [34] 28 days strength development) Chindaprasirt 24 h 65 °C curing; 5 min microwave curing + 3/6/12 h 65 °C curing; 20–42.5 MPa 5 min microwave curing + 6 h 65 °C curing optimal et al. [47] ambient temperature curing Slag based Aydin and – Steam curing (5 h humidity cabinet curing followed by 100 °C steam 15–90 MPa Steam cured specimens exhibited higher strength Baradan curing for 8 h) than autoclaved specimens [21] – Autoclaved curing (24 h humidity cabinet curing followed by 210 °C, 2.0 MPa autoclaved curing for 8 h) MT based Ahmari and 7 days 60–120 °C oven curing after casting 4–34 MPa 90 °C oven curing optimal Zhang [24] RHBA based He et al. [14] Cured at room temperature after casting for 14, 28, 35, 42 and 49 days 2–12 MPa Optimum strength achieved at 35 days of curing. 2.3. Microstructure of geopolymer matrix caused the aluminosilicate gel to precipitate at very early stages, hindering the polycondensation process. Hanjitsuwan et al. [12] reported similar mineralogical phases in Komljenovic et al. [31] analyzed the X-ray diffractograms of dif- high calcium lignite FA with the amorphous phase was indicated ferent sources of FA geopolymer pastes activated using sodium sil- by the broad hump between 20° and 38° while the crystalline icate with different Ms and concentrations. All the geopolymer phases were indicated by sharp peaks mainly consisted of quartz pastes were cured for 1 day at 20 °C followed by 6 days at 55 °C. (SiO2), hematite (Fe2O3), anhydrite (CaSO4), magnesioferrite The Ms used were 0.5 and 1.5 while the concentration of sodium (MgFe2O4) and calcium oxide (CaO). Upon alkali activation by silicate used was 8% and 10%. The Ms of sodium silicate was NaOH (8, 10, 12, 15 and 18 M), the broad hump was shifted to adjusted accordingly by adding NaOH. Zeolite morphological around 25–38° which is indicative of the existence of alkaline alu- phases of faujasite was found in the geopolymer specimens with minosilicate gel and C-S-H gels due to the high calcium content of low Ms of 0.5 but disappeared in the 1.5 Ms specimens, regardless the FA. The new phases occurred in the alkali-activated FA include of sodium silicate concentrations as can be seen in Fig. 1. With the portlandite (Ca(OH)2), sodium sulfate (Na2SO4) and hydrosodalite increase in the concentration of sodium silicate, the crystallization (Na4-Al3Si3O12(OH)). The peak intensity of C-S-H and hydrosodalite of zeolites were slowed down as the alkali activation process was increased as the NaOH concentration increases indicating that both accelerated due to the higher amount of dissolved silicon in the the aforementioned phases are the governing factor in the strength solution, resulting in the formation of amorphous phases as the development of high calcium lignite FA geopolymers. Following only reaction products. The authors also concluded that the reduc- the optimization of the modulus (SiO2/Na2O) and content (Na2O/ tion or absence of crystalline product had contributed to the higher CFA) of alkali activator which were determined to be 1.5 and 10% compressive strength of the geopolymers. respectively, the mineralogical phases of the optimized CFA geo- Salih et al. [35] observed shifting of the diffused halo peak of polymer (CFAG) specimen were evaluated using XRD and com- alkali activated POFA paste sample towards larger angle which cor- pared against the pure CFA sample. The major components of responds to the characteristic of a geopolymer gel. Also, the inten- pure CFA consisted of mullite, quartz, anhydrite and f-CaO. Upon sity of main crystalline peaks originated from the grounded POFA geopolymerization, a broad and amorphous hump between 20° was found to be reduced upon alkali activation process. and 40° (2h) appeared in the XRD pattern of CFAG which include Somna et al. [29] performed FTIR analysis on Ground FA (GFA) both geopolymeric gels and calcium silicate hydrate (C-S-H). This with various NaOH concentrations (4.5–16.5 M) cured at ambient finding suggested that the geopolymeric and hydrate reaction temperature. The IR spectrum of GFA consists of intense band at occurred concurrently in a single CFAG system [39]. Similar find- 450 cm 1 (Si–O–Si bending vibration) and 1180 cm 1 (Si–O–Si ings where the co-existence of geopolymer gel and C-S-H gel were and Si–O–Al asymmetric stretching vibration). A downward shift also observed in high calcium FA based geopolymer system of intensity band at 1180 cm 1 to around 972–990 cm 1 was [12,13,48,49]. observed for all the samples upon the addition of NaOH. This Mineral phases of Ground FA (GFA) using different NaOH con- downward shifting pattern was due to the rise in the tetrahe- centrations (4.5–16.5 M) as chemical activator were investigated drally-positioned Al atom present in the geopolymer system and by Somna et al. [29]. The authors observed that at low NaOH con- is also indicative that geopolymerization of NaOH-activated GFA centrations i.e. 4.5 and 7.0 M, the XRD pattern is somewhat similar has taken place. There is also the existence of newly formed broad to GFA paste without any addition of NaOH, suggesting that the band around 1650 cm 1 and 3480 cm 1 in the NaOH-activated GFA degree of geopolymerization is low. The presence of crystalline sil- samples which is associated to the stretching vibration of –OH and icate and aluminosilicate compounds were detected at higher bending vibration of O–H–O due to the geopolymerization of NaOH concentrations (9.5–14.0 M). At NaOH concentrations of NaOH-activated GFA into geopolymer pastes. Similar adsorption 16.0 M, crystalline products of aluminosilicate compound at bands was reported by other authors using class C FA (CFA) [39]. around 34° and 38° (2 theta degrees) disappeared. The observation The effect of curing temperature and the initial heat curing is probably due to the excess hydroxide ions concentration which durations on the morphology of clay-FA (FA) geopolymer bricks W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 377 especially for Ms values of 0 and 0.4. No significant micro-cracks were observed at the aggregate matrix interface for all the auto- clave-cured specimens. The main reaction products for both steam and autoclave cured AAS mortars are Na-substituted C-S-H, though a lower Ca/Si ratio for the C-S-H was observed for the autoclave- cured AAS mortars. Guo et al. [39] analyzed the micrographs (Fig. 4) of pure class C FA (CFA) and also the optimized CFA geopolymer (CFAG). The opti- mized CFAG was based on the Ms and content (Na2O/CFA) of alkali activator of 1.5% and 10%, respectively. The morphology of CFA (Fig. 4a) showed a series of spherical vitreous particles with vary- ing sizes, similar to that of class F FA. Upon geopolymerization, the partial dissolution of CFA particles can be clearly seen in Fig. 4b and the cavities of the broken CFA particles seem to be filled with large amount of micro particles of the reaction products (Fig. 4c). Also, the energy dispersive X-ray analysis (EDXA) confirmed the main geopolymeric gel existed in CFAG is (Na)-poly(sialate-disil- oxo-), i.e. Nan-(Si–O–Al–O–Si–O–Si–O–)n-. The geopolymeric gels were also found to be co-existed with C-S-H gels and some unre- acted CFA spheres (Fig. 4d). Komljenovic et al. [31] studied the effect of different Ms on the properties of sodium silicate activated FA geopolymers. When the modulus of sodium silicate increased, the Si/Al atomic ratio of the reaction products also increased, while the Na/Si and Na/Al decreased as can be seen in Table 3. Based on the compressive strength results obtained in the study, the authors concluded that Fig. 1. X-ray diffractogram of various FA activated by sodium silicate with various higher compressive strength is directly related to the higher modu- modulus (n) and concentrations [31]. lus value, which in turns is also directly related to the higher Si/Al atomic ratio of the reaction products. Yusuf et al. [37] reported that were evaluated by Sukmak et al. [33]. The formation of micro- Ms has a significant effect on the bond characteristic, structural cracks was clearly seen in specimens cured at high temperature units of reaction products, amorphousity and also the morphology i.e. 85 °C even at relatively short curing duration of 24 h (Fig. 2). of the products. The authors concluded that, (i) high silica modulus Similar micro-cracks development were observed for the speci- (HSM) system contained more polymerized aluminosilicate struc- mens cured at lower temperature (75 °C) but much longer curing ture due to attachment of more Al than the low silica modulus durations (72 h) and was illustrated in Fig. 3. Both the micro-cracks (LSM) system, thus the HSM system has the tendency to form prod- formation was very much related to the substantial loss of mois- ucts resemblance of calcium-(alumino) silicate hydrate (C-(A)-S-H) ture and pore fluid from the geopolymer matrix which induced while LSM system tends to form calcium silicate hydrate (C-S-H), excessive shrinkage during drying and subsequent loss of struc- (ii) HSM system yielded products with higher amorphousity than tural integrity of the clay-FA geopolymer matrix [50,51]. the LSM products, and (iii) HSM paste sample exhibited morphol- SEM analysis was performed by Aydin and Baradan [21] to ogy with denser, more compact and segmental than LSM paste sam- study the microstructure characteristic of steam and autoclave ple due to the lower volume of residual water present in the latter cured alkali activated slag (AAS) mortars. NaOH and Na2SiO3 were system. used as alkali activation agents. Na2O content was fixed at 6% while The effect of curing temperature and duration on the apparent silicate modulus (Ms) values varied between 0 and 1.2. The micro- porosity of ASTM class F FA geopolymer mortars activated by structure of steam-cured AAS mortars was transformed from a por- waterglass and NaOH were evaluated by Gorhan and Kurklu [18]. ous structure into well-packed and homogeneous structure with Geopolymer mortars were subjected to 65 and 85 °C oven curing the increasing Ms value. However, the crack intensity of the matrix for 2, 5 and 24 h after being molded. Results show that both the phase is higher for higher Ms value due to the tension generated curing temperature and time have significant effect on the appar- during shrinkage [52]. As opposed to steam curing, autoclave- ent porosity of geopolymer mortars. Similar findings where water cured specimens exhibited a dense and well-packed structure absorption has a direct influence on the apparent porosity of geo- polymer have been reported [53,54]. From the results acquired, the following conclusions were made: (i) the increase in curing time consistently reduces the apparent porosity of geopolymer mortars cured at constant temperature of 85 °C, (ii) the samples cured at 65 °C have a more porous structure as compared to their counter- parts which were cured at 85 °C, with the former exhibited appar- ent porosity range of 26.1–29.2% while the latter with the apparent porosity range of 25.3–29.8%, (iii) Sufficient while not excessive curing temperature and time play an important role in the pore structure development in geopolymer mortars. The MIP analysis carried out to study the pore size distribution of steam and autoclave cured alkali activated slag (AAS) mortars showed that steam curing caused a coarser pore size distribution (>25 nm). Meanwhile, autoclave cured specimens exhibited a finer Fig. 2. SEM micrograph of 28 days clay-FA geopolymer brick cured at 85 °C for 24 h pore size distribution. This is due to the formation of C-S-H with [33]. lower Ca/Si ratio which has denser matrix structure. Moreover, 378 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 Fig. 3. SEM micrographs of 28 days clay-FA geopolymer bricks cured at 75 °C for (a) 24 h, (b) 48 h, and (c) 72 h [33]. Fig. 4. Scanning electron microscopy (SEM) images of (a) pure CFA powder, (b) the reactive CFA sphere, (c) the reactive area ‘‘A’’ of the CFA sphere in SEM image (b), and (d) CFAG cured at 75 °C for 8 h followed by curing at 23 °C for 28 days [39]. the higher degree of slag grains hydration and the stronger aggre- portionally with the increasing NaOH concentration up to 18 M gate-matrix interface of autoclave cured specimens had also con- of NaOH concentration. The leaching of Ca2+ to the pore solution tributed to the finer pore size distribution [21]. at low NaOH concentration was not disrupted significantly which enable sufficient dissolved Ca2+ in the system for the formation 2.4. Rheological and physical properties of geopolymer of calcium silicate hydrate (C-S-H) and calcium aluminate hydrate (C-A-H) gels. While at higher NaOH concentration, the hardening Rheological properties (initial and final setting time and flow and setting of the paste is governed by the geopolymerization pro- test) and physical properties of geopolymers were found to be cess, which usually occurred in a slower rate as compared with influenced by the nature and complex of geopolymerization pro- CSH and CAH dependent cementitious system, thus resulting in a cess which in turns was governed mostly by the chemical activator higher setting time [12]. and also the curing regime employed[12,55–58]. The setting time Sathonsaowaphak et al. [30] performed flow tests on fresh of high calcium FA geopolymer pasteswas found to increase pro- Ground Bottom Ash (GBA) geopolymer mortars with various liquid W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 379 Table 3 etc [11,13,61,62]. Some researchers utilized ‘‘waste to waste’’ sta- Average content of different elements (at.%) and their ratios in the reaction products bilization technique by incorporating waste materials such as flue of FA + Na2O.nSiO2 (10% Na2O) [31]. gas desulfurization gypsum (FGDG) and Al-rich waste sludge to Elements and ratios SiO2/Na2O achieve a more efficient geopolymerization [49,63]. The addition 0.5 1.5 of different types of fibers, polymer resin, superplasticizers and Si 10.25 17.40 nano-materials was also found to greatly enhance the properties Al 3.95 5.91 of geopolymers, particularly the mechanical properties such as Na 11.66 4.41 flexural strength, splitting tensile strength and also the modulus Ca 2.32 2.62 of elasticity [13,64–67]. Fe 0.87 0.98 The following sections discussed in details on the influence of Si/Al 2.64 3.16 Na/Si 1.98 0.26 particle size distribution of binder phase and additives on the Na/Al 4.47 0.86 mechanical, durability, physical and microstructure properties of geopolymers derived from industrial by-products. alkaline/ash ratio, sodium silicate/NaOH ratio and NaOH concen- 3.1. Mechanical properties tration. The workable range of liquid alkaline/ash ratios was found to be in between 0.429 and 0.709. Higher liquid alkaline/ash ratios In an experimental study involving rice husk ash (RHA), the give rise to a more workable geopolymer mixes due to lower par- compressive strength of blended geopolymer pastes fabricated ticle interference and also larger inter-particle distance. On the from red mud (RM) and RHA was tested at the age of 60 days. other hand, the workable range of sodium silicate/NaOH ratios Two different gradations of RHA samples were used, one with and NaOH concentration lies between 0.67–1.5 and 7.5–12.5 M ‘‘as-received’’ condition and another grounded RHA with 100% par- respectively. The increase in sodium silicate/NaOH ratio and NaOH ticles passing through a #100-mesh (150 lm opening) sieve. A concentration resulted in less workable mortar mixes owing to the constant RHA/RM ratio of 0.4 was used throughout the study. higher viscosity of sodium silicate and NaOH. It is also recom- The geopolymer specimens containing ground RHA exhibited com- mended that the amount of sodium silicate in the mortar mixes pressive strength value of 16.08 MPa, an increase of 37.43% if com- should be kept as low as possible for economic reasons, while at pared with an equivalent unground RHA geopolymer. The the same time does not compromise the workability and also the enhancement in compressive strength was attributed to the higher strength of the geopolymer mortars. degree of geopolymerization achieved by the fine particle size and A higher amount of efflorescence was observed in palm oil fuel the high specific surface area of ground RHA, resulting in a stronger ash (POFA) geopolymer paste samples which contained lower solid geopolymer [14]. Similar findings using grinded RHA in RHA/FA to liquid ratios and lower sodium silicate to sodium hydroxide geopolymers were reported [60]. ratios [35]. Higher Na ions in lower sodium silicate to sodium In the other separate study on rice husk bark ash (RHBA), Nazari hydroxide ratios and lower solid to liquid ratios samples have et al. [20] investigated the effect of particle size distribution on the higher tendency of alkaline leaching out phenomena on the speci- compressive strength of FA (FA)- rice husk bark ash (RHBA) based men’s surface due to the weakly bound Na ions in the nanostruc- geopolymers. The FA and RHBA were sieved and grinded into two ture of geopolymer gel. different particle sizes. The average particle sizes obtained for FA were 75 lm and 3 lm while for RHBA, average particle sizes of 90 lm and 7 lm were obtained. The finer FA and RHBA were 3. Effect of particle size distribution of binder phase and denoted fF and fR while the coarser FA and RHBA were denoted additives on the properties of geopolymer cF and cR. A total of four series of geopolymer samples based on the different particle sizes of FA and RHBA as illustrated in Table 4 Besides the adjustment in the chemical activator and curing were fabricated and subjected to compressive strength tests at the regime, the nature and also the fineness of the geopolymer source curing ages of 7 and 28 days. Generally, specimens made with fine materials play a crucial in the strength development, durability FA and fine RHBA particles (fF-fR series) exhibited highest com- properties and microstructure of the resultant geopolymer matri- pressive strength regardless of curing ages. This is due to the fact ces. Several authors have come to an agreement that the variation that finer particles are more capable of filling pores, hence, resulted of particle size distribution of the binder phase poses a significant in a denser and more compact geopolymer paste structure which effect on the compressive strength, physical properties and micro- can sustain higher applied load prior to ultimate failure. This find- structure of the resultant geopolymer paste [14,59]. Generally, bin- ing was further strengthened when the fF-cR series yielded higher der phase with finer particle size distribution will have a higher compressive strength as compared with the corresponding cF-fR reactivity and subsequently produces geopolymer paste which series. Although fine RHBA was used in cF-fR series, the total per- has denser microstructure, higher compressive strength and centage of finer particles in fF-cR series was much higher due to refined physical properties [20,60]. Chindaprasirt et al. [48] the higher percentage of FA in the FA-RHBA geopolymer mixtures reported an improvement in drying shrinkage for geopolymer mor- and this has proven to be the key factor in determining the tars fabricated using fine high calcium FA as geopolymer source strength of FA-RHBA geopolymers. material. The authors also suggested that high calcium geopolymer Besides FA, the production of coal ash comprised approximately mortars exhibited 1000% improvement in term of drying shrinkage 20% of bottom ash (BA) and as of now most of the BA is disposed of in comparison with OPC mortars, proving the excellent dimen- as landfill due to limited industrial application value if compared sional stability of high calcium FA geopolymers. to FA. However, due to its similarity in silica and alumina content In order to achieve an efficient geopolymer synthesis, one must with FA, with the exception of excessively high carbon content due find or achieve an ideal balancing between the essential elements to incomplete burning and large particle size, several researchers during the geopolymerization i.e. SiO2, Al2O3, Na2O and most have started to incorporate BA in either geopolymer production recently CaO. One of the ways of finding the ideal balancing of or as cement/aggregate replacement materials [30,59,68,69]. Sata the aforementioned elements is by the addition of commercially et al. [59] evaluated the effect of different particle size of bottom available additives such as calcium hydroxide (Ca(OH)2), alumi- ash (BA) on the compressive strength of the BA geopolymer num hydroxide (Al(OH)3), silica fume (SF), nano-SiO2, nano-Al2O3 mortars. Three different fineness of BA was incorporated into the 380 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 mortar specimens: fine BA (15.7 lm), medium BA (24.5 lm) and et al. [13] incorporated 1–3% of nano-SiO2 and nano-Al2O3 by bin- coarse BA (32.2 lm). The compressive strength of the geopolymer der weight to fabricate FA geopolymer pastes. The 7, 28 and mortars were tested at the ages of 7, 28, 90, 180 and 360 days. 90 days compressive strength of the geopolymer pastes were com- Results are indicative that the finer particle sizes of BA give rise pared in reference with an OPC paste. The compressive strength to a higher compressive strength of the hardened geopolymer mor- results showed that the addition of nano-SiO2 and nano-Al2O3 tars at all the curing ages. The geopolymerization rate was regardless of the additive’s dosage give rise to superior compres- increased with the higher fineness and higher specific surface areas sive strength as compared to the reference OPC paste specimens. of fine BA. Besides, as the ‘‘as-received’’ BA contains large portions The dual functionality of the nano-sized additives in FA based geo- of mesopores on the ash surfaces, the grinding process indirectly polymer which provides additional SiO2 and Al2O3 to the geopoly- helps to reduce the porosity of the BA particles and thus reducing mer system and at the same time act as micro filler yielded the water demand of the fresh mortar and contributed to the additional calcium silicate hydrate (CSH) or calcium aluminosili- higher compressive strength of the hardened geopolymer mortar. cate hydrate (CASH) and sodium aluminosilicate hydrate (NASH) The highest compressive strength of fine BA geopolymer mortar gels in the geopolymer matrix and a dense geopolymer structure. is 61.5 MPa, attained at the curing age of 180 days. Similar to OPC concrete, high range water reducing admixtures Mijarsh et al. [11] examined the compressive strength of treated or superplasticizers (SP) can be incorporated into geopolymer sys- palm oil fuel ash (TPOFA) based geopolymer mortars containing tem in order to reduce its water content while maintaining the various proportion of calcium hydroxide (Ca(OH)2), aluminum desired workability thus resulting in a higher strength geopoly- hydroxide (Al(OH)3) and silica fume (SF). Six design factors namely mers [71–73]. Nematollahi and Sanjayan [65] investigated the Ca(OH)2 wt.%, Al(OH)3 wt.%, SF wt.%, NaOH concentration (molar- effect of different type of commercial SP to the compressive ity), Na2SiO3/NaOH (weight ratio) and alkali activator-solid materi- strength of class F FA geopolymer paste. The SP used in the exper- als (weight ratio) were examined at five levels using the Taguchi imental study consisted of naphthalene (N), melamine (M) and experimental design method to obtain the optimum mix propor- modified polycarboxylate (PC) based SPs, each of them was added tion. A total of 25 trial mixes were fabricated and tested in accor- to the fresh geopolymer mixture at a dosage of 1% by mass of FA. dance to the L25 array proposed by Taguchi method. The TPOFA Although the compressive strength of all the SP-added geopolymer was obtained by first separate the incompletely combusted fibers paste decrease as compared to the control mix (without the addi- and kernel shells from the raw POFA by using a 300 lm sieve. Then tion of SP), the authors concluded that PC based SP is the most suit- the POFA was heated at 500 °C for 1 h to remove the unburned car- able type of commercial SP to be incorporated into class F bon before being subjected to secondary grinding to obtain the geopolymer paste activated using multi compound activator (Na2- TPOFA. The sand to binder material mass ratio for all the mixes SiO3/NaOH = 2.5), showing the least reduction in compressive was fixed at 1.5. Immediately after molding, the specimens were strength (16–29%). However, Puertas et al. [74] reported the addi- wrapped using a cling film and left to cured for 1 h before being tion of vinyl copolymer and polyacrylate copolymer based SPs into subjected to oven curing at 75 °C for 48 h. The compressive FA based geopolymer paste and mortar does not bring about any strength of all the specimens was then tested at 1, 3 and 7 days significant changes on the compressive strength nor the workabil- of curing ages. Compressive strength result of the 25 trial mixes ity of the resultant geopolymers. ranged from 15.67 to 44.74 MPa at 1, 3 and 7 days of curing. From Nath and Kumar [75] utilized two types of iron making slags the trial mix results, the optimum level of substitutions or ratios of namely granulated blast furnace slag (GBFS) and granulated corex various factors examined i.e. additive materials (20 wt.% Ca(OH)2, slag (GCS) in FA (FA) based geopolymer system in the range of 0– 5 wt.% SF and 10 wt.% Al(OH)3) and alkaline activators (10 M NaOH, 50% by weight of binder. GCS is produced during Corex process in Na2SiO3/NaOH = 2.5 and alkaline activator/solid = 0.47) was iron making, having similar chemistry and phase composition with obtained and fabricated to test the resultant compressive strength. GBFS. Prior to be used as mix constituents, both GBFS and GCS The optimum TPOFA geopolymer mortar exhibited compressive were milled for 2 h in a ball mill to obtain the desired fineness of strength of 47.27 ± 5.0 MPa at 7 days of curing which is higher than d50 = 18.49 lm and 18.53 lm, respectively. The compressive all the 25 trial mixes for the same curing duration. strength of hardened geopolymers was tested on the 7 and 28 days The addition of small amount of nano-sized additives such as of curing. The following conclusion can be made based on the com- nano-SiO2 and nano-Al2O3 was known to effectively enhanced pressive strength results obtained, (i) both geopolymers, GBFS-FA the compressive and tensile strength of concrete by the mean of and GCS-FA exhibited an increase in compressive strength with additional pozzolanic and filler effects [70]. Phoo-ngernkham the increasing slag content, (ii) the increase in compressive Table 4 Mix design of FA-RHBA based geopolymer specimens [20]. Sample Weight percent of fine FA (fF Weight percent of coarse FA Weight percent of fine RHBA Weight percent of coarse RHBA SiO2/AlO3 designation wt.%) (cF wt.%) (fR wt.%) (cR wt.%) ratio fF-fR-1 60 0 40 0 3.81 fF-fR-2 70 0 30 0 2.99 fF-fR-3 80 0 20 0 2.38 fF-cR-1 60 0 0 40 3.81 fF-cR-2 70 0 0 30 2.99 fF-cR-3 80 0 0 20 2.38 cF-fR-1 0 60 40 0 3.81 cF-fR-2 0 70 30 0 2.99 cF-fR-3 0 80 20 0 2.38 cF-cR-1 0 60 0 40 3.81 cF-cR-2 0 70 0 30 2.99 cF-cR-3 0 80 0 20 2.38 Alkali activator (waterglass + sodium silicate) to FA-RHBA mixture ratio is 0.4. W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 381 strength rate is more pronounced at slag addition content above splitting tensile strength of the geopolymer paste sample by a mas- 20% for both geopolymers, (iii) GCS addition generally has resulted sive 36% as compared to unreinforced paste sample. The authors in higher 7 and 28 days compressive strength in comparison with also observed a change of failure mode from brittle to ductile fail- GBFS addition, (iv) the strength increment is mainly due to the for- ure with the incorporation of sweet sorghum fibers and the effect mation of C-S-H cementitious gel which occupied the pore space are similar with those geopolymers incorporated by synthetic and subsequently improved the density of the resultant geopoly- fibers. In a separate study, the splitting tensile strength of FA based mer binder matrix, (v) compressive strength as high as 93.4 and self-compacting geopolymer concrete increased by 12.8% by the 91.2 MPa was obtained for 50% addition of GCS and GBFS respec- addition of 10 wt.% silica fume, in comparison with the control tively after 28 days of curing. mix in which no silica fume is added into the FA based geopolymer Boonserm et al. [63] attempted to improve the geopolymeriza- concrete mixtures [79]. tion of bottom ash (BA) by incorporating ASTM class C FA (FA) and The addition of 2% nano-SiO2 and nano-Al2O3 by weight was flue gas desulfurization gypsum (FGDG). BA:FA as the blended found to significantly enhance the flexural strength in high calcium source materials with ratios of 100:0, 75:25, 50:50, 25:75 and FA based geopolymer paste system [13]. The increase in reaction 0:100 were used. The source materials were then replaced by products such as CSH, CASH and NASH due to the addition of nano FGDG in 0%, 5%, 10% and 15%. Sodium silicate/NaOH ratio, liquid/ particles led to remarkable enhancement in the interfacial transi- ash ratio, sand/ash ratio were fixed at 1.0, 0.6 and 2.75 respectively tion zone within the geopolymer matrix. The authors found that for all the geopolymer mixes. The fresh geopolymer mixes were the flexural strength of the resultant geopolymer paste samples casted in 5 5 5 cm cubic molds and were subjected to 40 °C increased linearly with the square root of ultimate compressive electric oven curing for 48 h. All the geopolymer mortars were sub- strength and the flexural strength values obtained generally higher jected to compressive strength test at the age of 7 days. From the than that of OPC concrete as given by ACI 318. Fig. 1 showed the experimental results, it can be concluded that, (i) degree of geopo- relationship between the compressive strength and flexural lymerization of FA is higher than that of BA geopolymer owing to strength of the resultant geopolymer pastes in comparison with the high glassy mineral phase content of FA and also the additional the values given by ACI 318. Similar finding was also reported calcium silicate hydrate (CSH) gel formed as a result of the reaction where the addition of 10 wt.% silica fume in FA based self-com- between Ca2+ and silicate from the FA, thus the strength of blended pacting FA based geopolymer concrete (SCGC) was found to geopolymer mortars increased with the increase in FA content, (ii) increases the flexural strength by 11.09% if compared to the non- the incorporation of 5–10% of FGDG shown significant effect on the silica fume added FA based SCGC [79]. blended geopolymer mortars with low FA replacement level i.e. 0%, Flexural strength of geopolymers can be greatly enhanced by 25% and 50%. The aforementioned phenomenon was due to the incorporating different types of short synthetic fibers such as additional CSH gel formed as a result of increase in the concentra- PVA fibers, polypropylene (PP) and carbon fibers through bridging tion of Ca2+ ions and also significantly higher dissolution rate of effect during the micro and macro-cracking of the geopolymer Al3+ in BA due to the presence of SO2- 4 ions in the system, (iii) the matrix under flexure stresses [67]. However, the use of natural addition of 15% of FGDG caused adverse effects to all the geopoly- fibers to reinforce geopolymers are gaining wide interest due to mer mortars. High FGDG content obstruct the geopolymerization its environmental friendly and cost efficient characteristics [80]. process especially in geopolymer mortars with high FA content In one of those studies, cotton fabric was used to reinforce ASTM and as a result a thenardite phase which existed as impurity pres- class F FA based geopolymer composites [64,81,82]. Results ence in the geopolymer system and caused all the geopolymer showed that the addition of cotton fabric greatly enhanced the mortars to exhibit very low compressive strength ranging from flexural strength of the geopolymer composites. The flexural 0.3 to 1.0 MPa. strength was enhanced by almost 3-fold with the optimum cotton Reuse and recycling of industrial wastes are the ideal solution in fabric addition which is 8.3 wt.%, in comparison with the unrein- the current waste management problems and Chindaprasirt et al. forced geopolymer composites. On the other hand, the addition [49] found a way to utilize Al-rich waste originated from a waste- of 2 wt.% of sweet sorghum fiber enhanced the flexural strength water treatment unit of a polymer processing plant as additive in of ASTM class F FA based geopolymer paste samples by almost the fabrication of high calcium FA based geopolymer mortar. The 40%. Higher fiber content will resulted in a significant decrease in raw Al-rich wastes were dried, grounded and calcined at tempera- flexural due to entrapment of air bubbles in the geopolymer com- ture ranging from 400 to1000 °C before it can be used as additive. posite resulted from the poor workability and fiber agglomeration Results showed that active h-Al2O3 can be obtained at high calcina- [66]. tion temperature, i.e. 1000 °C. 7 days compressive strength of The addition of water-soluble organic polymers was found to 34.2 MPa could be obtained by adding 2.5 wt.% of Al-rich waste improve the mechanical properties of geopolymers through the calcined at 1000 °C. Further increase in the additive’s dosage modification of microstructure and pore size distribution of the resulted in reduction in compressive strength due to the excess resultant geopolymer matrix. For instance, Zhang et al. [67] Al species which act as non-functional filler in the geopolymer improved the mechanical performance i.e. flexural strength of system. metakaolin (MK)/ground blast furnace slag(GBFS) based geopoly- The brittle nature of geopolymers can be inhibited or overcome mer composites by the incorporation of 1–15 wt.% of polymer by incorporating different types of fibers such as polyvinyl alcohol resin. The authors concluded that the addition of a mere 1% resin (PVA) fibers [76], carbon fibers [77] and plain woven stainless steel by weight percentage improved greatly i.e. 41% the flexural mesh [78] to improves the ductility of the geopolymer composite. strength of the geopolymer composites. Higher dosage of polymer The aforementioned fibers are produced by energy intensive pro- resin incorporation resulted in decrease in flexural strength due to cess and are therefore against the geopolymer context of sustain- the coating effect of resin has on MK and GBFS fine particles thus ability and environmental friendly product. In light of this, Chen resulted in significant reduction in the binder’s reactivity. The et al. [66] investigated the effect of sweet sorghum fiber which is authors have also evaluated the effect of elevated temperature a natural fibers on the splitting tensile strength of FA based geo- towards the resin-reinforced MK/GBFS based geopolymer compos- polymer paste. Prior to be incorporated into the geopolymer mix- ites in a separate study [83]. The geopolymer composites exhibited ture, the fibers were first treated with alkaline solution in order an increase in flexural strength when exposed to temperature 150– to improve the adhesion between the fiber and the matrix. Results 300 °C due to the enhancement in polycondensation reaction. showed that the inclusion of 2% sweet sorghum fiber increase the However, when exposed to elevated temperature ranging from 382 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 450 to 850 °C, the flexural strength significantly reduced due to the needed additional aluminum compound in order to achieve higher dehydration of the geopolymer matrix, thermal decomposition of geopolymerization efficiency [84]. Rattanasak et al. [62] studied the added resin and some phase transformation. the effect of Al(OH3) substitution and curing temperature on the The modulus of elasticity of high calcium FA geopolymer paste disintegration of rice husk ash (RHA) based geopolymer paste in sample cured at ambient temperature increased by 30% by the boiling water. RHA was replaced at 2.5, 5, 10, 20 and 30 wt.% by addition of 2% nano-SiO2 and nano-Al2O3 by weight of binder. E- Al(OH3). 10 M NaOH, Na2SiO3/NaOH = 1.5, and solid/total mix- value as high as 17.65 GPa is obtained after 90 days of curing, com- ture = 0.6 was used throughout the experimental study. 1 wt.% of parable with that of OPC concrete. The observed enhancement in E- boric acid (H3BO3) powder was added to each mixture to solve values of the geopolymer paste samples was due to the denser and the problem of disintegration in water. Upon completion of com- stronger matrix formed with the addition of the aforementioned paction into steel molds, the geopolymer paste were wrapped with nano-particles. The E-values obtained in the study though are cling film and cured at 70, 85, 100 and 115 °C for 48 h before being slightly lower if compared with OPC paste and blended cement subjected to boiling test. Al(OH)3 is needed in RHA based geopoly- paste samples and it was concluded that the absence of heat treat- mer because RHA contains mostly silica (SiO2). The added Al(OH)3 ment in the current study is the main factor contributing to the served as aluminum source for RHA geopolymer. Results showed observed lower E-value [13]. that all the pastes cured at 70, 85 and 100 °C disintegrated upon being subjected to boiling test while specimens with 2.5–30 wt.% 3.2. Rheological and physical properties of geopolymer of Al(OH)3 cured at 115 °C maintained their structural integrity after immersion in boiling water. At curing temperature of ASTM class F FA has been widely used as geopolymer source 115 °C, which is above the dehydration temperature of H3BO3 materials [18,33,65]. However, the aforementioned ash material (110 °C), H3BO3 starts to react with SiO2 and other constituents requires heat treatment to achieve high early strength and faster in a high silica aluminosilicate material and formed a more stable setting time of the resultant geopolymer pastes which is a draw- geopolymer matrix. The H3BO3 is relatively inert at temperature back as heat treatment induced additional energy requirement below 110 °C in RHA geopolymer. Although the RHA geopolymer for the fabrication of geopolymer with competitive strength if with 20 and 30 wt.% of Al(OH)3 does not disintegrate in boiling compared to OPC concrete. Several researchers incorporated addi- water, swelling of specimens was observed due to the excessive tional additives such as Ca(OH)2 [11], nano-SiO2 and nano-Al2O3 Al(OH)3 content. Shorter setting time with the addition of alumi- [13] or by using or hybridize high calcium source materials such num compound was also observed by De Silva et al. [15]. as GGBFS and ASTM class C FA in order to achieve fast setting time Generally finer geopolymer source materials bring about higher and strength development while not compromising the strength degree of reactivity and vigorous geopolymerization due to higher potential of the geopolymer system [21,34,49,75]. For instance, in specific surface area attained in finer source materials which in high calcium FA based geopolymer paste, the addition of nano- turn will have a shorter setting time and higher early strength SiO2 up to 3% by binder weight has resulted in decrease of initial development. However, significant enhancement in mechanical, and final setting time as compared to OPC paste while the addition durability and microstructure properties can only be achieved if of nano-Al2O3 has resulted in a slight reduction in initial and final the balancing between the corresponding higher water require- setting time in reference with the OPC paste. The reduced setting ment and source materials fineness is achieved. In one particular time by the addition of nano-SiO2 was due to the much faster rate study, original FA (CFA), medium fineness FA (MFA) and fine FA of activation with the readily available free calcium ions in the high (FFA) with corresponding Blaine fineness of 2700, 3900 and calcium FA and formed additional CSH gels [13]. Similar finding 4500 cm2/g were utilized in fabricating high strength geopolymer was also reported by Chindaprasirt et al. [58]. using high calcium FA as the source material [48]. FFA yielded Nematollahi and Sanjayan [65] studied the effect of different shorter setting time compared to CFA and MFA due to the higher commercial superplasticizers (SPs) addition to the relative slump specific surface area and the presence of larger content of amor- of class F FA geopolymer pastes. Naphthalene (N), melamine (M) phous phase which increase the reactivity of FFA. The relatively and modified polycarboxylate (PC) based SPs were used at 1% by long setting time behavior of the three fineness of high calcium mass of FA throughout the experimental study. Two types of chem- FA suggested that the aforementioned geopolymer products can ical activators were used in the study, i.e. 8 M concentration of be suitably use for industrial application for the ease of handling, NaOH and multi compound activator consisted of Na2SiO3 and transporting, placing and compaction prior to curing. The authors 8 M NaOH with Na2SiO3/NaOH = 2.5. The relative slump of the also concluded that the particle shape of FA regardless of its fine- fresh geopolymer pastes were measured using the mini slump test ness plays an important role in improving the workability of the method in accordance to ASTM C1437. For geopolymer paste acti- fresh geopolymer paste samples. It has been agreed upon that fine vated using only 8 M NaOH, N-based SP was found to be the most FA with a spherical shape and a smooth surface would yield the effective high range water reducing admixture as the addition of best ball bearing effect and thus increase the flow of the subse- 1% of N-based SP increased the relative slump of fresh geopolymer quent fresh mixes without the need of additional water or water- pastes by 136% in reference with the control mix (without SP). For reducing admixtures. multi compound activators-activated geopolymer pastes, PC-based Table 5–7 summarized the effect of particle size distribution of SP is the most effective high range water reducing admixture with the binder phase and additives towards the mechanical, rheologi- 46% increase of the relative slump, followed by N-based SP with cal and physical properties of geopolymers derived from various increment of 8% while M-based SP exhibited a decrease of relative industrial by-products. slump by 3%. Though all the commercial SPs used were found to be chemically unstable in high basic media (NaOH + Na2SiO3) which 3.3. Microstructure of geopolymer matrix reduce their plasticizing effect, PC-based SP showed a more prom- inent plasticizing effect as compared to N and M-based SP owing to Another little known industrial by product namely flue gas the existence of lateral ether chains in its structure which resulted desulfurization gypsum (FGDG) has found its way into the fabrica- in steric repulsion in addition to electrostatic repulsion effect pos- tion process of geoplolymers. This waste gypsum which originated sessed by N, M and PC based SP. from the coal burning industry has the potential of enhancing the Being a silica rich source material with over 90% of SiO2 pre- geoplymerization process as in the case of enhancing hydration of sents in the chemical composition, the alkali activation of RHA OPC concrete by pure gypsum [85,86]. However, the aforemen- W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 383 tioned waste materials have so far found limited research in the Rice husk ash (RHA) is another industrial by-product which emerging geopolymer field [63]. In one of the few research that uti- possessed great potential to be utilized as construction and build- lized FGDG in the fabrication of geopolymers, Boonserm et al. [63] ing materials. Its high silica content (>80 wt.%) renders it as poten- studied the microstructure characteristic of bottom ash (BA): FA tial source material for geopolymer fabrication. Due to its low (FA) geopolymer paste with different BA:FA ratio and also different aluminum content, an external aluminum source such as Al(OH)3 replacement level by FGDG. Sodium silicate/NaOH ratio and liquid/ is usually added to RHA based geopolymer to enhances the geopo- ash ratio were fixed at 1.0, and 0.6 respectively for all the geopoly- lymerization process. Rattanak et al. [62] performed FTIR analysis mer paste. Freshly mixed pastes were subjected to 40 °C electric to determines the effect of Al(OH)3 substitution to the RHA-based oven curing for 48 h and were analyzed using scanning electron geopolymer. RHA was replaced at 2.5, 5 and 10 wt.% by Al(OH)3. microscope (SEM) on the age of 28 days. SEM micrograph as per For each mix proportion, 1 wt.% of boric acid (H3BO3) was added Fig. 5 shows that the geopolymerization products of BA:FA blended to the mixture to overcomes the problem of disintegration in water mortars consisted of well-connected structures, mostly glassy of the hardened geopolymer pastes. A mixture of NaOH and Na2- phase structure with no definite grain boundary. Without FGDG, SiO3 was used as the chemical activator solution for the geopoly- very dense geopolymer matrices were obtained with mixes con- mer paste. 10 M of NaOH and Na2SiO3/NaOH of 1.5 was fixed for taining high FA content i.e. 100% and 70% of FA as in Fig. 5a4 and all the geopolymer mixes. The freshly casted geopolymer pastes a5. The matrices were less dense and less homogeneous with low were subjected to oven curing at 115 °C for 48 h before being FA content as shown in Fig 5a1–a3. On the other hand, the addition tested for FTIR analysis. Si–O bending and stretching peaks were of 5% and 10% of FGDG adversely affected the microstructure of observed at 470 cm 1 and 1100 cm 1 for the raw RHA sample. geopolymer containing high FA content (50%, 75% and 100% FA; Upon incorporation of Al(OH)3, reduction of Si–O bending peak at Fig 5c4 and c5) while improved microstructure were observed with 470 cm 1 and occurrence of new peak at 780 cm 1 which is mixes containing low FA content (0% and 25% FA; Fig 5a1–c1 and assigned to Si–Al–O symmetric stretching were observed. The a2–c2). The enhancement of microstructure observed in low FA reduction in Si–O bending peak is more pronounced with the content is attributed to the ability of sulfate ions in FGDG to dis- increase in Al(OH)3 content, suggesting that more silica has reacted solve Al3+ ions in BA while the weakened matrices observed in high with aluminum and formed geopolymer gels. The main spectra FA mixes with the addition of FGDG is due to the presence of the- band which was observed at 1100 cm 1 of the raw RHA sample, nardite phase which existed as impurity in the geopolymer system was shifted to a lower frequency of 1040 cm 1 (Si–Al bonding) which obstructed the geopolymerization reaction. for all the composites. The peak intensity at 1040 cm 1 increased Addition of 2.5 wt.% of Al-rich waste sludge calcined at 1000 °C as the Al(OH)3 content increased and this is associated with the was found to produce a homogeneous matrix if compared to non- increase in mean chain length of the aluminosilicate composite calcined Al-rich waste with the same dosage in the fabrication of (ASC). high calcium FA based geopolymers as can be seen in Fig. 6. The high calcination temperature transformed the inactive boehmite 4. The effect of aggressive environmental exposure on to active alumina (h-Al2O3), which in turn can be used to adjust properties of geopolymers the SiO2/Al2O3 ratio in the high calcium FA based geopolymer and enhanced the geopolymeric gel formation in the resultant Like OPC concrete, geopolymers concrete will be exposed to matrix [49]. severe environments such as marine environment where sulfate and acid attacks are pre-dominant if geopolymer binders are suc- 3.4. Fourier transform infrared spectroscopy (FTIR) analysis cessfully implemented to replace OPC. Therefore, thorough under- standing of the effect of geopolymer binders exposed to aggressive FTIR provides useful information about the vibrational transi- environment is imminent. Thus far, due to the total contrast of tions and rigidity of chemical bonds present in organic and inor- hydration reaction and the reaction products where OPC yields ganic materials [87,88]. In the emerging field of geopolymer, the C-S-H gels while geopolymer binders yield N-A-S-H, C-A-S-H or changes in chemical bonds and spectrum upon the alkali activation C-S-H gels, geopolymer binders were reported to have superior process can be identified and comparative study can be done effi- resistance towards sulfate and acid attacks by various researchers ciently using FTIR [49,63]. [89–92]. The current literatures reported the sulfate, acid and chlo- Boonserm et al. [63] analyzed the effect of flue gas desulfuriza- ride resistance of geopolymeric binders by using various methods tion gypsum (FGDG) on the IR spectra of bottom ash (BA) and FA and analytical techniques, such as the direct immersion in pre- (FA) geopolymer pastes. The FGDG was added in 0%, 5%, 10% and determined sulfate and acid solution followed by subsequent mea- 15% to the BA and FA geopolymer pastes, respectively. Both BA surement of strength and mass loss [90], measurement of corroded and FA geopolymer pastes were activated using sodium silicate depth [93] and accelerated laboratory electrochemical method and NaOH, and were cured at 40 °C in an electric oven for 48 h. [94]. The 28 days samples were then grinded to particle size less than In the following sections, the effect of aggressive environmental 75 lm and used for FTIR analysis at the range of 4000–400 cm 1. exposure on properties of geopolymers will be deliberated in terms FTIR results showed considerably broad bands at 3700– of the influence of environmental exposure condition on the 2200 cm 1 and 1700–1600 cm 1 for all geopolymer pastes which mechanical properties, physical properties and microstructure are assigned to O–H stretching and H–O–H bending were due to changes. the weakly bound water molecules which were adsorbed on the surface or trapped in large pores between the rings of geopolymer products. SO24 bonding was detected at wave numbers of 1200 4.1. Mechanical properties and 636 cm 1 and this suggested that there is a reaction between SO24 ions and the alkaline solution which form SO4 compound. Apart from its environmental friendliness, the distinct advan- With the exception of BA geopolymer paste with 5% FGDG, all tage of geopolymer binders is its excellent acid resistance prop- other BA and FA geopolymer pastes exhibited distinctive peaks of erties if compared to ordinary Portland cement (OPC) binders SO4 especially for the 10% and 15% FGDG added FA geopolymer [3]. Ariffin et al. [90] exposed geopolymer concrete based on pastes. The large quantity of SO4 compound detected was in line blended pulverized fuel ash (PFA) and palm oil fuel ash (POFA) with the low strength exhibited by FA geopolymer mortars. to 2% solution of sulfuric acid up to 18 months. The ratio of 384 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 Table 5 Effect of particle size distribution of binder phase on the compressive strength of geopolymer concrete. Types of geopolymer Fineness Compressive Primary findings strength FA based Nazari et al. [20] FA: 75 lm and 3 lm RHBA: 90 lm and 7 lm 15–45 MPa Combination of fine FA and RHBA give rise to highest strength Chindaprasirt et al. [48] FA with Blaine fineness of 2700, 3900 and 39–75 MPa FA with highest Blaine fineness give rise to optimum 4500 cm2/g strength RHA based He et al. [14] RHA: 100% passes 150 lm sieve 16 MPa 37.43% strength increment as compared to unground samples Detphan and Chindaprasirt RHA: 5, 3 and 1% retained on No. 325 sieve 34.5–43.0 MPa Finer RHA gives rise to highest compressive strength [60] BA based Sata et al. [59] BA: 15.7 lm, 24.5 lm and 32.2 lm 35–61.5 MPa Finer BA give rise to higher strength Table 6 Effect of additives on the compressive strength and physical properties of geopolymer concrete. Types of Types of additives Compressive Rheological & physical Primary findings geopolymer strength properties BA based Boonserm et al. ASTM Class C FA (0, 25, 50, 75, 100%), FGDG (0, 5, 5–55 MPa 50% FA and 5% FGDG replacement level optimal [63] 10, 15%) FA based Nematollahi and 1% addition by mass of binder of N, M and PC 47–81.3 MPa Relative slump: PC based SP showed highest plasticizing effect Sanjayan [65] based SP PC > M > N and least strength reduction Puertas et al. [74] 0.5–1.5% addition of vinyl copolymer and 30–35 MPa Insignificant changes in strength and polyacrylate copolymer based SP workability with addition of both SPs Phoo-ngernkham 1–3% of nano-SiO2 and nano-Al2O3 by binder 20.2– Decrease of initial and 2% nano-SiO2 and 1% nano-Al2O3 optimal et al. [13] weight 56.4 MPa final setting time Nath and Kumar 0–50% GBFS and GCS replacement by weight of 8.5–93.4 MPa – 50% GBFS and GCS replacement optimal [75] binder – FA-GCS yielded higher strength than FA- GBFS Chindaprasirt et al. Addition of 2.5–5 wt.% of Al-rich waste calcined 27.4– Al-rich waste with 2.5 wt.% and 1000 °C [49] at 400, 600, 800 and 1000 °C 34.2 MPa calcined temperature optimal POFA based Mijarsh et al. [11] 15–25 wt.% Ca(OH)2, 5–10 wt.% Al(OH)3, 2.5– 15.67– 20 wt.% Ca(OH)2, 5 wt.% SF and 10 wt.% Al(OH)3 7.5 wt.% SF 44.74 MPa optimal Table 7 Effect of additives on the flexural strength and splitting tensile strength of geopolymer concrete. Types of Types of additives Flexural Splitting tensile Primary findings geopolymer strength strength FA based Chen et al. [66] 1, 2 and 3 wt.% addition of sweet sorghum fibers 3.2–5.6 MPa 2.2–3.4 MPa Addition of 2 wt.% of sweet sorghum fibers optimal Alomayri et al. 0–8.3 wt.% addition of horizontally and vertically 8–32 MPa 8.3 wt.% and horizontally oriented cotton fabric [81] oriented cotton fabric optimal Memon et al. [79] 0–15 wt.% addition of silica fume 4.09– 4.14–4.67 MPa 10 wt.% of silica fume optimal 4.56 MPa Chindaprasirt 0–3% addition of nano-SiO2 and nano-Al2O3 3.66– 1 wt.% SiO2 and 2 wt.% Al2O3 optimal et al. [13] 5.12 MPa Slag based Zhang et al. [67] 1–15 wt.% addition of polymer resin 4.8–8.6 MPa 41% enhancement in flexural strength by 1 wt.% resin addition PFA to POFA used was 70:30. The blended ash geopolymer (BAG) exposure. Based on the compressive strength results, BAG con- concrete was activated using commercial grade sodium hydrox- crete exhibited superior resistance to sulfuric acid as compared ide (NaOH) and sodium silicate (Na2SiO3) with alkaline solution to OPC concrete, with an average of 7.3% strength loss after a to blended ash ratio of 0.4 and NaOH: Na2SiO3 of 2.5 by mass. month, 1.6% strength loss for the subsequent months and a total An OPC concrete with water to cement ratio of 0.59 was fabri- of 35% strength loss after 18 months of exposure. In contrast, the cated and used as the control specimen. Upon casting, both compressive strength of OPC concrete was severely affected with the BAG and OPC concrete were subjected to room temperature a total of 68% strength loss after 18 months of sulfuric acid expo- (28 °C) curing for 28 days before immersion in sulfuric acid. sure. The findings were in agreement with other similar Compressive strength of both BAG and OPC concrete was exam- researches where geopolymer concrete exhibited minimal ined before and after 1, 3, 6, 12 and 18 months of sulfuric acid strength loss upon prolonged acidic environment exposure W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 385 Fig. 5. SEM micrograph of fractured BA and FA geopolymer pastes (BA:FA:FGDG) [63]. Fig. 6. SEM images of Al-rich waste added high calcium FA geopolymer: (a) 2.5 wt.% non-calcined Al-rich waste sample; (b) 2.5 wt.% 1000 °C calcined Al-rich waste sample [49]. 386 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 which justified its superior acid resistance in comparison with strength of the more porous samples instead show a steady OPC concrete [59,95]. increase in strength along the exposure period up to 365 days, Rattanasak et al. [62] studied the effect of sulfuric acid (H2SO4) although the strength first drop during the 28 days exposure per- and magnesium sulfate (MgSO4) immersion on the compressive iod. The authors concluded that the unusual observation was most strength of rice husk ash (RHA) based geopolymer mortars. RHA probably attributed to the continuing alkali activation in the sul- was replaced by Al(OH)3 at 2.5, 5 and 10 wt.%. 10 M NaOH and Na2- fate solution by the more porous geopolymer mortars. The afore- SiO3 were used as chemical activator and the Na2SiO3/NaOH mass mentioned findings were in total contrast to the current body of ratio was fixed at 1.5. Sand to powder mass ratio of 2:1 was used knowledge where a denser matrix or reduced porosity either geo- for all the specimens. Upon casted into 50 mm cubic molds, the polymers or OPC concrete resulted in enhancement in the durabil- fresh geopolymer mortars were wrapped with cling film to prevent ity performance [62,89,98]. moisture loss and were subjected to oven curing at 115 °C for 48 h. After being cooled to room temperature, all the specimens were 4.2. Microstructure analysis of geopolymer then immersed in 5 wt.% MgSO4 and 3 vol.% H2SO4 for 90 days, after which the compressive strength loss for each specimens were Bhutta et al. [91] investigated the mineralogical phase changes determined. For the unexposed specimens, the compressive of PFA-POFA based blended fuel ash geopolymer concrete (BFAGC) strength was observed to increase as the Al(OH)3 content increases before and after immersion in 5% sodium sulfate (Na2SO4) solution with the highest compressive strength was recorded at 20.0 MPa. for a period of 18 months. As can be seen in Fig. 7, after 18 months This was due to the lower Si/Al ratio as a result of Al(OH)3 addition of 5% Na2SO4 exposure, the semi-crystalline aluminosilicates gel which in turns promotes the formation of more cross-linked alumi- (N-A-S-H) which is presence before the immersion remains intact nosilicate structure and thus leading to higher strength. After the and showed little changes. On the other hand, the OPC specimen immersion periods, specimens that was subjected to 5 wt.% MgSO4 which was used as the control specimen showed significant immersion showed a higher rate of strength losses as compared to changes in the diffraction pattern after 18 months of 5% Na2SO4 specimens that were immersed in 3 vol.% of H2SO4. This phenom- exposure due to the formation of gypsum and ettringite which sub- enon was due to the formation of magnesium hydroxide sequently lead to expansion and spalling of surface layers. (Mg(OH)2) resulted from the reaction between hydroxyl ions The effect of 2% sulfuric acid immersion up to 18 months on the (OH ) of the aluminosilicate composite (ASC) and the magnesium XRD pattern of blended ash geopolymer concrete based on PFA and ions from MgSO4 and caused the migration of hydroxide ions POFA activated by mixture of NaOH and Na2SiO3 solution was towards the surface of the specimen which produced the insoluble studied by Ariffin et al. [90]. OPC concrete was used as the control brucite. Furthermore, in a high silica system, the Mg(OH)2 reacted specimen. Before the acid exposure, the primary mineral phase with the silica gels and formed the hydrated magnesium silicate detected in BAG concrete was a crystalline N-A-S-H phase while which possesses no binding capability and resulted in sharp drop a different crystalline phase of C-S-H was detected in the OPC con- in compressive strength. The results justified that there is some crete. After 18 months of sulfuric acid exposure, the main phases distinct correlation between the mechanical, microstructure and detected in BAG concrete e.g. sodalite, gmelinite, natrolite and N- durability properties of RHA based geopolymers [14,60,96]. A-S-H were still intact and some traceable amount of gypsum Ahmari and Zhang [92] measured the unconfined compressive was also detected as a result of reaction with atmospheric CO2. strength (UCS) of copper mine tailings (MT)-based geopolymer As expected, no ettringite was detected in BAG concrete as the Al bricks after immersion in in pH 4(nitric acid) and pH 7 solutions ions participated in the formation of N-A-S-H gels, thus making up to 4 months. NaOH concentration and curing temperature of the available Al ions insufficient in the formation of ettringite as 15 M and 90 °C respectively were fixed throughout the experimen- in Portland cement binder system. In the other study, Bascarevic tal study. Based on the previous study [24], MT-based geopolymer et al. [97] investigated the effect of sodium sulfate solution (50 g/ bricks with initial water content/forming pressure of 12%/25 MPa l) exposure on the mineralogical phase changes of FA based geo- and 16%/0.5 MPa were selected to study the effect after immersion polymers up to 365 days. XRD analysis showed that no new phases in pH 4 and pH 7 solutions. Based on the compressive strength were formed in the geopolymer sample even after 365 days of results, the authors carved out the following conclusions, (i) MT- exposure period, implying the superior sulfate resistance charac- based geopolymer bricks with water content/forming pressure of teristic by the FA based geopolymers. 12%/25 MPa exhibited UCS loss of 59.3% at pH 4 and 53.3% at pH XRD patterns of copper mine tailings (MT)-based geopolymer 7. Meanwhile 16%/0.5 MPa MT-based geopolymer bricks exhibited bricks subjected to immersion in pH 4 (nitric acid) and pH 7 solu- UCS loss of 78.4% for pH = 4 and 75.2% for pH = 7, respectively after tions for 4 months showed increase in the crystalline peaks inten- immersion periods of 4 months, (ii) the substantial strength loss by sity after immersion. However, the crystalline phases before and both 12%/25 MPa and 16%/0.5 MPa MT-based geopolymer bricks is after immersion showed very little differences. This is due to the caused by incomplete geopolymerization of MT, modification of effect of immersion in pH 4 and pH 7 solutions for a prolonged per- chemical compositions of the geopolymer gels formed, high Si/Al iod of time which caused the dissolution of geopolymer and subse- ratio and also high degree of unreacted alkali in the specimens, quently leads to the exposure of unreacted crystalline MT grains (iii) 12/25 specimens exhibited lower strength loss compared with [92]. 15/0.5 specimens due to the more prevalent effect of reduction in mix porosity due to the exerted compression pressure over com- 4.3. Fourier transform infrared spectroscopy (FTIR) analysis pact structure resulted from geopolymer gels towards acid attack resistance. The aforementioned phenomenon is also related to Ariffin et al. [90] performed FTIR analysis on the blended ash the Na/Al ratio of 12/25 specimens which is closer to unity than geopolymer (BAG) concrete based on PFA and POFA before and 16/0.5 specimens and thus less geopolymer gels were dissolved after immersion in sulfuric acid environment for up to 18 months. in the acid solution. Mixture of NaOH and Na2SiO3 solution were used as chemical acti- An interesting phenomena has been observed by Bascarevic vator. OPC concrete was used as the control specimen. Major bands et al. [97] in assessing the sulfate resistance of two type of FA based at approximately 3440, 1645, 1425, 1010 cm 1 and 3465, 1645, geopolymer mortars. While the lower porosity FA based geopoly- 1425, 1040 cm 1 were detected in unexposed OPC and BAG con- mer mortars only exhibited noticeable reduction in compressive crete, respectively. The stretching band of O–H, the bending of strength after 180 days of exposure period, the compressive chemically bonded H–O–H and carbonate in the system were W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 387 located at 3200–3700 cm 1, 1645 cm 1 and 1425 cm 1 respec- tively. The main binder gel band for OPC which is the asymmetric stretching mode of C-S-H structure was detected at 1010 cm 1 while the N-A-S-H gels formed in the geopolymer binder system appeared at 1040 cm 1. Upon exposure to 2% solution of sulfuric acid for 18 months, FTIR spectra of BAG concrete showed little or no difference from their unexposed counterpart. In contrast, marked decomposition of C-S-H gel and O–H phases was detected in the OPC concrete after exposure to acid environment. Water component was shifted from 3435 cm 1 to 3405–3555 cm 1 and the chemically bonded water molecules changed from 1625 cm 1 to 1625–1690 cm 1 due to the presence of gypsum. Meanwhile the presence of calcite shifted the C-S-H gel phase from 1010 cm 1 to 1145 cm 1 which indicates the decomposition of the main binder system in OPC concrete. The authors also con- cluded that the decomposition of the C-S-H gel has provided addi- tional calcium for the formation of gypsum in the exposed OPC concrete. Ahmari and Zhang [92] performed FTIR analysis on copper mine tailings (MT)-based geopolymer bricks after immersion in pH = 4 (nitric acid) and 7 solutions for a period of 4 months. Two batches of MT-based geopolymer bricks were fabricated, one with initial Fig. 7. XRD of BFAGC and OPC concrete before and after immersion in 5% Na2SO4 for 18 months [91]. water content/forming pressure of 12%/25 MPa and another with ratios of 16%/0.5 MPa. The IR spectra of the MT powder, geopoly- mer bricks before immersion and after immersion in different solu- their unexposed counterpart. In contrast, the mass loss of exposed tions are shown in Fig. 8. Significant difference was observed for OPC concrete was much higher at 18–22% in comparison with the geopolymer bricks before and after immersion, though there is unexposed OPC samples. The temperature by which the evapora- not much of difference in the IR spectra between specimens tion of free water occurred was higher for both OPC (140 °C) and immersed in pH 4 (nitric acid) and pH 7 solutions. For MT powder, BA (122 °C). Furthermore, the intensity of mass loss was so much after geopolymerization, the band around 1000 cm 1 which is higher in the exposed OPC samples. In the same study conducted, associated to the stretching vibrations of Si–O–T (T = Al or Si) the authors concluded that gypsum is the dominant product in the bonds shifted to a lower wave numbers while the weak shoulder samples exposed to sulfuric acid and the majority of the mass loss at 1070 cm 1 becomes stronger. Upon immersion in the aforemen- in the aforementioned intense peaks could be attributed to the tioned solutions for 4 months, the amorphous geopolymer gels decomposition of the resultant gypsum as gypsum was known to were partially dissolved and the underlying crystalline phases of decompose from temperature 110–150 °C. The results showed that MT grains were exposed, leading to the Si–O–T bonds becomes the BAG concrete held upper hand in term of sulfuric acid resis- sharper and shifted towards higher wave numbers. Also, the weak tance compared to the OPC concrete. shoulder around 1070 cm 1 also becomes weaker as a conse- quence of geopolymer gels dissolutions. Carbonates compounds formed due to geopolymerization at around 1450 cm 1 was untraceable after immersion in pH = 4 (nitric acid) and 7 solutions 4.5. Physical properties of geopolymer due to carbonates have been dissolved in the solutions. PFA-POFA based blended fuel ash geopolymer concrete (BFAGC) 4.4. Thermogravimetry (TGA) analysis showed superior resistance to prolonged sulfate environment exposure as compared to OPC concrete, as can be seen in Fig. 10. In order to study the effect of sulfuric acid exposure of blended Also, the decrease in mass of OPC concrete after prolonged expo- ash geopolymer (BAG) concrete based on PFA and POFA, Ariffin sure to sulfate environment was as high as 20% as compared to et al. [90] performed TGA on BAG concrete before and after 4% decrease in mass of BFAGC. It was found that the superior dif- 18 months of 2% solution of sulfuric acid exposure. OPC concrete ference in mass change of OPC concrete and BFAGC was due to was used as the control specimen. The grinded BAG and OPC sam- the low calcium content in BFAGC which renders it to be more ples were held under isothermal condition for 60 min at 40 °C and resistance towards sulfate attack [91]. However, the mass change then heated to 900 °C at 10 °C/min in a nitrogen environment. method in examining the sulfate and acid resistance of geopoly- Before exposure to acid environment, the mass loss in the measure meric binders [91,99] has been casted into doubt as there are study period for OPC and BAG concrete was 18% and 10%, respectively. that claimed that the results obtained from the mass change meth- Fig. 9 showed there were 4 distinct peaks observed in the differen- ods are not representative for assessing the sulfate and acid resis- tial thermograms (DTG) at various temperatures for both OPC and tance of geopolymeric binders [93]. Lloyd et al. [93] claimed the BAG concrete. The first two peaks, which occurred below 100 °C, utilization of corroded depth method over the conventional change were attributed to the removal of free evaporable water trapped in mass method as a more effective and representative testing in the pores of binder’s gel i.e. C-S-H or N-A-S-H gel system. The method to determine the acid resistance of inorganic polymer dehydration of calcium-rich silicate gel was detected at peak just binders. This is due to the tendency of formation of apparently below 200 °C while the mass loss at 264 °C was attributed to the intact, but physically weak and porous reaction product on the dehydration of gypsum (OPC) and gmelinite (BAG). Broad mass loss sample surface upon attack by acid substances, a phenomena was detected at approximately 704 °C for OPC concrete and was which differs from the acid attack mechanism for other type of assigned to the decomposition of the carbonate minerals. Upon binders in which the a complete disappearance of the binder phase 18 months of 2% sulfuric acid immersion, the TGA and DTG dia- were observed. The aforementioned statement needs further clar- gram of BAG concrete showed little or no difference compared to ification in order to justify the methods currently being employed 388 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 by various researchers in assessing the acid resistance of geopoly- rebars were fabricated and tested for the corrosion test using the meric binders are valid. aforementioned method after curing for 28 days. Results showed Sata et al. [59] evaluated the effect of 5% sodium sulfate and 3% that geopolymer concrete possessed superior resistance towards sulfuric acid immersion over 360 days on the physical properties of salt attack relatively with OPC concrete. OPC beams starts to crack BA geopolymers mortars. PC mortar with the same binder to sand after 60 h of accelerated corrosion test while geopolymer concrete ratio was used as the control specimen and was subjected to the showed no sign of cracking even after the end of the accelerated same environment as the BA geopolymer mortars specimens. BA corrosion test. The superior durability performance of geopolymer geopolymer mortar exhibited excellent resistance after immersion concrete was further verified where no mass loss were recorded for in 5% sodium sulfate for 360 days where the length changes were the rebars upon the completion of accelerated corrosion test where only 65–121 microstrain. On the other hand, the PC mortars which for OPC concrete, the maximum mass loss of rebars were recorded exhibited expansion of 7600 microstrain for the same duration of as 71.2% after the completion of the test. immersion in 5% sodium sulohate. Deterioration of PC mortar in sulfate environment was due to the formation of gypsum and ettringite from the reaction of calcium hydroxide and calcium 5. The effect of water content and forming pressure on the monosulfoaluminate which leads to the expansion and cracking properties of geopolymers of the mortar’s surface layers. Under 3% sulfuric acid solution, BA geopolymer mortar again exhibited superior resistance if com- Water content and forming pressure has a significant effect on pared to PC mortar, where the weight loss of BA geopolymer mor- the mechanical strength and sorptivity performance of geopoly- tars after 360 days immersion in 3% sulfuric acid were only 1.4– mer, particularly in the fabrication of geopolymer pressed block 3.6%, which is relatively insignificant as compared to PC mortar [24]. This is because both water content and forming pressure have which recorded weight loss of 95.7%. The weight loss in acid envi- a direct influence on the total porosity of the geopolymer matrix. ronment is very much related to the calcium content in the system, Generally, higher water content will result in increased total poros- where higher calcium content leads to higher amount of calcium ity [100]. On the other hand, higher forming pressure will reduce hydroxide, which in turns reacts with the acid solution form salt the total porosity of the geopolymer matrix. Also, the utilization crystal. The salt crystal formed within the paste matrix induces of pressed forming methods in fabricating geopolymer allows sig- internal tension stress which causes the formation of crack and nificant reduction in water requirement in comparison with vibra- scaling within the paste matrix. tory forming methods where order suitable workability of fresh Reddy et al. [94] attempted to simulate the exposure of FA mixtures must be achieved for proper compaction [101]. In this based geopolymer concrete to marine environment by utilizing section, the effect of water content and forming pressure on the an accelerated laboratory electrochemical method for the corro- properties of geopolymers is deliberated in depth. sion test, relative to OPC concrete. 150 150 525 mm geopoly- mer and OPC concrete beams with centrally reinforced 13 mm Fig. 8. IR spectra of MT powder, (a) 12/25 specimens and (b) 16/0.5 specimens Fig. 9. Thermogravimetry (TGA–DTG) curves for OPC and BAG concretes. (a) before and after immersion in pH = 4 (nitric acid) and 7 solutions for 4 months [92]. Themogravimetric data (TGA). (b) Differential thermograms (DTG) [90]. W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 389 5.1. Mechanical properties out of the binder matrix. Thus, bricks with lowest possible level of porosity and higher strength can be fabricated using the pres- In alkali activation process, the geopolymerization reaction pri- sure forming method [100,101]. An experimental study was con- marily involves the chemical reaction between the dissolved spe- ducted by Ahmari and Zhang [24] to study the effect of forming cies of silicates and aluminates in the presence of highly alkaline pressure on the unconfined compressive strength (UCS) of copper environment. The presence of water in the geopolymer system mine tailings (MT)-based geopolymer bricks. The MT-based geo- merely acts as a transport medium between the dissolved silicates polymer bricks were activated with 15 M of NaOH and curing tem- and aluminates ions. Besides, mixing water also provide workabil- perature of 90 °C based on the optimization results from the same ity to the freshly mixed geopolymer mortars, as it does not partic- research work. Various forming pressures were applied and the ipates directly in the geopolymeric reaction [102]. However, geopolymer bricks were tested after 7 days of oven curing. From Komljenovic et al. [31] reported there are some effect, if not signif- the UCS results, it can be concluded that, (i) all the MT-based geo- icant, of the water/FA (FA) ratio to the strength of geopolymer mor- polymer bricks showed an increasing trend of UCS up to a certain tars, depending on the type of activators used. Generally, for NaOH, level of forming pressure, (ii) the decline in UCS for high forming Na2SiO3 and Ca(OH)2 activated FA geopolymer mortars, the com- pressure was due to the loss of NaOH solution which was squeezed pressive strength increases with the decrease in water/FA ratio. out during the forming process at high forming pressure, hence For KOH activated geopolymer mortars, the geopolymer mortars reducing the degree of geopolymerization, (iii) forming pressure exhibited low compressive strength even at low water/FA ratio. is related to the initial water content of geopolymer mix, i.e. higher This is indicative of the low activation potential of KOH for FA acti- forming pressure may result in lower initial water content. vation as compared with other alkali activators. Zhao et al. [100] studied the effect of forming pressure on the However, the importance of water was highlighted in a study compressive strength of autoclaved bricks made from low silicon where the liquid medium is crucial for the diffusion of the dis- tailings. Alkali activated cementitious materials based on slag solved alumina and silica species. Ahmari and Zhang [24] studied and FA were incorporated into the mixture in an attempt to pro- the effect of initial water content on the unconfined compressive duce high strength low silicon autoclaved bricks with load bearing strength (UCS) of copper mine tailings (MT)-based geopolymer capacity. Water content in materials of 7.5%, tailings to cementing bricks. A numbers of geopolymer bricks were fabricated based on material mass ratio of 85:15, autoclaved curing regime of 2-8-2 6 initial water contents i.e. 8%, 10%, 12%, 14%, 16% and 18% and (temperature rising-holding-dropping stages) and autoclaved tested on the 7th day of curing duration. The initial water content steam pressure of 1.0 MPa was fixed throughout the study. Results was referred to the mass ratio between the water in the activating showed that the compressive strength of bricks increased with the solution (NaOH) and the solid content of the mixture. The fresh forming pressure. However, the magnitude of strength increase in geopolymer pastes were put inside a steel mold and compressed forming pressure exceeding 20 MPa was deemed insignificant. to an extent whereby it reached saturation state. The results indi- Thus, the optimum forming pressure should ranges between 18 cated that the UCS of MT-based geopolymer bricks increased with and 20 MPa in consideration of producing bricks with adequate higher initial water content. The aforementioned observation was mechanical strength with minimal energy usage. The compressive attributed to the role of water as a liquid medium during geopoly- strength of autoclaved bricks obtained using the aforementioned merization. Besides, it is also related to the availability of sufficient forming pressure range exceeds the target strength of 15.0 MPa amount of NaOH in liquid phase during geopolymerization. The for load bearing brick. amount of NaOH, in turn, is strongly dependent on the Na/Al and Na/Si ratios as shown by other related researches [15,57,103]. UCS as high as 33.7 MPa was obtained at 18% initial water content 5.2. Water absorption and 0.2 MPa forming pressure. In another experimental work, the compressed autoclaved bricks based on low silicon tailings and Water absorption is a very important parameter for the fabrica- alkali activated cementing materials (slag and FA) attained the tion of geopolymer bricks as it indicates the permeability and the optimum compressive strength of 16.0 MPa at the water content degree of reaction for geopolymer bricks. Generally higher degree range of between 6.5–8.0% and forming pressure of 20 MPa. Any of geopolymerization gives rise to a less porous and permeable values in excess of the aforementioned range of water content geopolymer matrix. Ahmari and Zhang [24] evaluated the effect and forming pressure had resulted in a decrease in compressive of forming pressure on the water absorption of copper mine tail- strength [100]. ings (MT)-based geopolymer bricks. NaOH concentration, initial In the fabrication of compressed building bricks, forming pres- water content, and curing temperature were made constant at sure plays an important role in achieving optimum densification 15 M, 16% and 90 °C respectively. The freshly mixed MT-based geo- of intra- or inter-particles packing by pushing the entrapped air polymer paste was compressed at five different forming pressure namely 0.5, 1.5, 3, 5 and 15 MPa to form the geopolymer bricks. The authors found that the water absorption after 4 days of soaking varies from 2.26% to 4.73% corresponding to forming pressure from 0.5–15 MPa. The increase in water absorption with the increasing forming pressure was attributed to a higher amount of NaOH solu- tion being squeezed out at higher forming pressure. Under such circumstance, the geopolymerization reaction is hindered, hence, less geopolymer gels were formed and subsequently resulting in higher porosity of the geopolymer matrix. On the other hand, Frei- din et al. [104] reported water absorption less than 10% for FA based geopolymer bricks upon the addition of hydrophobic addi- tives. All the water absorption values of the MT-based geopolymer bricks were well below the maximum water absorption value allowable for different kind of bricks as in accordance to various Fig. 10. Visual appearance of OPC concrete and BFAGC after 18 months of 5% ASTM standards namely ASTM C34-03, C62-10, C126-99, C216- Na2SO4 immersion [91]. 07a and C902-07. 390 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 6. Blended geopolymer mortars were investigated. The following conclusions were made as a result of the laboratory investigation, (i) compressive strength Blended geopolymers is a new category of geopolymeric binder range from 13 to 42 MPa can be achieved using sodium silicate to which is produced by the combination of two or more industrial sodium hydroxide ratio from 1.5 to 5.9, with the optimum ratio at waste ashes followed by subsequent stabilization and solidification 4.0 regardless the RHA/FA hybridization ratios, (ii) a reduction in using chemical activators. The dual advantages of waste utilization compressive strength was observed with the increasing SiO2/ and more importantly the alteration in Si/Al and Ca/Si in geopoly- Al2O3 ratio mainly due to the cellular structure of RHA which mer system prompted sudden rush in the amount of researches in resulted in high water uptake of the corresponding mortar mixes, the field of blended geopolymer during recent years (iii) a delay time of 1 h, curing period of 48 h and curing tempera- [20,22,23,101,105,106]. Ever since it was known that C-S-H gel ture of 60 °C was found to be the most ideal curing regime to can be co-exist with geopolymeric gel in a single system and con- achieve high strength, with the effect of curing temperature was tributes to the overall strength gain [39], various researchers has more prevalent to mortar mixes which contain high FA dosage. utilized high calcium waste material such as GGBFS and ASTM Rice husk and bark ash (RHBA) is generated in biomass electric- class C FA to blend with ASTM class F FA in order to achieve a ity power plant where rice husk and bark were burned at temper- higher early strength gain and shorter setting time which is bene- ature around 400 °C, yielding silica-rich ash with similar chemical ficial for the application in pre-cast industry [30,75] Canfield et al. composition with RHA, with the exception of slightly lower silica [107] investigated the role of calcium in FA based geopolymers by and higher calcium compound. Songpiriyakij et al. [108] investi- blending high and low calcium FA. It was found that calcium gated the effect of SiO2/Al2O3 ratios towards the compressive played two major roles during the geopolymerization of the strength development of RHBA/FA blended geopolymers. Due to blended FA geopolymer specimen, (i) calcium was found to aids the silica rich RHBA, a wide range of SiO2/Al2O3 ratios i.e. 4.03– the dissolution of silica and alumina species from the FA particles, 1035 were obtained by hybridizing the two aforementioned base yielding higher tetrahedral silicate and aluminate monomer con- materials. Results shown that the optimum compressive strength centration as shown in FTIR, XRD and TGA/DSC results, (ii) calcium at 28 days of curing obtained is 51.0 MPa for SiO2/Al2O3 ratio of also functions as a counter-balancing cation when incorporated 15.9 due to the formation of stronger Si–O–Si bonds contributed into the geopolymer pore structure. In a separate study, The work- by the addition of silica rich RHBA. However, the authors observed ability of fresh geopolymer concrete consisted of GGBFS and FA expansion and cracking on specimens with SiO2/Al2O3 ratios showed a decreasing trend with the increase in slag content and greater than 15.9 over time. Also, specimen failure mode transition decrease of SS/SH ratio due to the enhancement in reactivity in from crushing to deformation was also observed for SiO2/Al2O3 the presence of GGBFS in FA based geopolymer system [105]. Other greater than 15.9. The authors also concluded that the other than waste materials such as RHA and POFA have also found consider- the reactivity of the source materials, the quality of the matrix ably interest among geopolymer researchers in blended geopoly- phase developed is also an essential factor which contributes mer field [11,37,60,62]. towards the compressive strength development of the RHBA/FA The following section covers the mechanical, durability and blende geopolymer paste. The strength obtained was in agreement microstructure properties of a numbers of emerging blended with another related study [20]. geopolymers. While the addition of wastepaper sludge from the paper recy- cling industries in OPC concrete brings about adverse effect to var- 6.1. Mechanical properties ious properties[109,110], Yan and Sagoe-Crentsil [36] attempted to incorporate dry wastepaper sludge addition to FA-based geopoly- The 60 days compressive strength of geopolymer pastes derived mer mortars on the mechanical properties. The dry wastepaper from rice husk ash (RHA) and red mud (RM) with varying RHA/RM sludge was added to the geopolymer system as a sand replacement ratios (0.3, 0.4, 0.5 and 0.6) was examined by He et al. [14]. All the material in the range of 0–10 wt.%. Laboratory grade sodium sili- geopolymer pastes were cured under room temperature and atmo- cate solution and sodium hydroxide pellets were used as activating spheric pressure until the specified testing age. NaOH was used as solutions and the SiO2/Na2O molar ratio and H2O/Na2O ratio were the chemical activator and the concentration and liquid to solid fixed at 1.5 and 11, respectively. Also, the other constant parame- weight ratio were fixed at 4 M and 1.2 throughout the experimen- ters in the experimental study are the sand/FA ratio of 3 and the tal study. The aforementioned RHA/RM ratios of 0.3, 0.4, 0.5 and liquid/solids ratio of 0.2. The freshly casted specimens were ini- 0.6 employed in this study were corresponding to Si/Al ratios of tially cured in a steam chamber at 60 °C for 8 h before being 1.68, 2.24, 2.80 and 3.35. Results showed compressive strength demolded, followed moist curing (sealed in plastic bags at room increased up to RHA/RM ratio of 0.5 before decreased for the paste temperature) until the testing ages. The authors concluded that specimens with RHA/RM ratio of 0.6. The compressive strength the decrease in compressive strength of the geopolymer mortars obtained ranged from 3.2 to 20.5 MPa. The enhancement in up to 10 wt.% of dry waste paper sludge was due to the presence strength was due to the increasing amount of reactive silica and of surfactants (dissolved lignin residues) in dry wastepaper sludge higher specific surface area of RHA. Meanwhile the deterioration which act as air-entraining admixtures and altered the porosity in strength observed in specimens with RHA/RM ratio of 0.6 was and pore size distribution of the geopolymer matrix, resulting in due to the high amount of unreacted RHA particles in the mixture. a lower density geopolymer mortars as the percentage of dry Also, the higher concentration of soluble Si ion in that particular wastepaper sludge increases. The average 91 days compressive mixture which hinders the restructuring of Si and Al geopolymer strength of geopolymer mortars containing 2.5 wt.% and 10 wt.% network, subsequently result in the formation of weaker geopoy- of dry wastepaper sludge were 55.7 MPa and 31.2 MPa respec- mer matrix. In another study involved RHA, Detphan & Chindapra- tively, retaining 92% and 52% of the control mortars strength which sirt [60] studied the feasibility of producing geopolymers based on attained 60.6 MPa. the hybridization of rice husk ash (RHA) and FA (FA). The hybrid- Due to the emergence of fluidized bed combustion (FBC) as a ization ratios of RHA/FA used to fabricate the geopolymer mortars promising clean coal technology compared to the traditional pul- were 0/100, 20/80, 40/60 and 60/40. The effect of sodium silicate to verized coal combustion (PCC) method due to the reduction of sodium hydroxide ratio, the SiO2/Al2O3 ratio and curing regime SO2 and NOx gasses emitted in flue gas, Chindaprasirt et al. [111] (delay time, curing temperature and curing period) towards the utilized FA obtained from FBC as a source material for geopolymer compressive strength development of the RHA/FA geopolymer fabrication. Geopolymer mortars consisted of FBC-FA and PCC-FA W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 391 with mass ratios 0:100, 20:80, 40:60, 60:40, 80:20, 100:0 were CASH) with the inclusion of GGBFS in the blended system [113]. activated by 10 M NaOH and sodium silicate with Na2SiO3 to NaOH Yusuf et al. [114] recommended the partial replacement of POFA ratio of 1.5 by weight. Liquid content for each mixes was adjusted by GGBFS to be at 20% in order to maintain excellent mechanical accordingly to achieve workable mix due to the irregular shape of properties profile. POFA when in used as replacement material in FBC-FA. Sand to ash ratio of 2 was maintained for all the mixes. FA based geopolymers reduces the early strength and delayed Freshly mixed geopolymer was poured into 25 mm diameter and the geopolymerization process [23]. However, POFA/FA geopoly- 25 mm height mold, oven-cured at 65 °C for 48 h and cured contin- mers exhibited a gain in strength even upon subjected to elevated uously at controlled room temperature of 25 °C until the testing temperature as high as 500 °C. The authors also reported the addi- ages. The compressive strength results at 7 days specimen age tion of POFA into FA based geopolymer mortars reduced the com- were indicative that the compressive strength of geopolymer mor- pressive strength and density of the mixtures [45]. Excellent tar decreased as the amount of FBC-FA increased. This is due to the compressive and flexural strength was observed in a ternary lower reactivity of FBC-FA as compared to primary PCC-FA. The low blended geopolymer system comprising POFA, FA and RHA [116]. amorphous phase and the porous nature of FBC-FA is the governing reason for the aforementioned observation. PCC-FA to FBC-FA ratio 6.2. Microstructure of geopolymer matrix of 60:40 which exhibited 7 days compressive strength of 30.0 MPa was recommended as the mix proportion for the fabrication of geo- The microstructure of FBC-FA and PCC-FA blended geopolymer polymer due to the relatively high strength and also the significant paste was studied by Chindaprasirt et al. [111]. The geopolymer amount of FBC-FA used. paste were activated using NaOH to Na2SiO3 ratio of 1.5 and were GGBFS is the one of the most popular source materials to be cured in oven at 65 °C for 48 h and followed by subsequent curing blended with other geopolymer source materials as the addition at controlled temperature of 25 °C. The morphology of the blended of GGBFS increase the amount of amorphous silica and alumina geopolymer paste was examined at the age of 7 days. Fig 11 (a) and and also the CaO in the resultant geopolymer system which in turn (b) show the morphology of 60:40 PCC-FA: FBC-FA blend and its will greatly improves the mechanical performance of the blended geopolymer paste counterpart. The spherical shape of PCC-FA geopolymers. For instance, different grades of ground granulated induced the ball bearing effect and improved the workability of blast furnace slags (GGBFS) i.e. 80, 100 and 120 were incorporated the resultant paste as compared with FBC-FA which consisted of into FA based geopolymers in a fixed FA/GGBFS weight ratio of 5/3. irregular and porous particles. The morphology of the 60:40 Results suggested that without the incorporation of external amor- blended geopolymer paste, PCC-FA geopolymer paste (Fig 11c) phous silica source, higher grade GGBFS is only beneficial for the and FBC-geopolymer (Fig 11d) paste show continuous mass of alu- early strength development of FA geopolymer. The reactivity of mino-silicate, indicating a relatively well developed geopolymer the higher grade GGBFS can be exploited by the enhancement in network. However, the unreacted/partially reacted grains of irreg- SiO2/Al2O3 and SiO2/Cao ratios by the incorporation of additional ular FBC-FA is much more porous than PCC-FA which culminated amorphous silica source [106]. GGBFS/FA geopolymer concrete in lower strength of blended geopolymer paste with higher FBC- with 28-days compressive strength as high as 51 MPa could be FA content. achieved with the GGBFS/FA hybridization ratio of 20/80, 40% of XRD analysis of various ratios of PCC-FA/FBC-FA blended geo- activator liquid and SS/SH ratio of 1.5 when cured at ambient tem- polymer pastes was conducted by Chindaprasirt et al. [111]. At perature [105]. the age of 7 days, blended geopolymer pastes with high amount Deb et al. [105] investigated the splitting tensile strength of of PCC-FA (60%, 80% and 100% PCC-FA) showed high amount of GGBFS/FA based geopolymer concrete cured at ambient tempera- amorphous phases and trace amount of crystalline products. ture. The tensile strength increased with increasing slag content Meanwhile geopolymer pastes with high amount of FBC-FA (60%, and decreasing SS/SH ratio, providing a strong correlation with 80% and 100% FBC-FA) exhibited intense peaks of crystalline the corresponding compressive strength development. GGBFS/FA phases with reduced amount of amorphous gel. Calcium silicate geopolymer concrete with 20% of GGBS content and SS/SH ratio similar to the hydration product of Portland cement was detected of 1.5 exhibited 55% higher 28-days tensile strength than geopoly- in all the blended hardened geopolymer pastes. mer concrete mixture with 10% GGBFS and SS/SH ratio of 2.5. The The pore size of POFA/FA geopolymer mortars significantly ranges of tensile strength obtained were also consistently higher increased upon subjected to elevated temperature beyond 800 °C. than the OPC concrete specimens. High POFA content in the blended geopolymer mortar mixes The numbers of researches utilizing palm oil fuel ash (POFA) in deformed at 800 °C while FA based geopolymer mortars maintain the fabrication of geopolymers or blended geopolymers have risen their structural integrity up to temperature of 1000 °C, suggesting in recent years due to the abundance of POFA wastes especially in lower termal stability upon addition of POFA in FA bsed geopoly- South East Asian countries for example Malaysia and Thailand mer mortars [45]. In a separate study, the addition of POFA in FA [112–116]. Approximately 3 million tonnes and 0.1 million tonnes geopolymer system increase the porosity in the resultant blended of POFA were produced annually in Malaysia and Thailand, respec- geopolymer matrix [23]. This is due to the unreacted POFA parti- tively [4,117]. The aluminosilicate waste material was obtained cles having the tendency to trap air because of its inherent crum- from the palm oil industry and usually undergone pre-treatment bled shape. process involving calcination, grinding and sieving before it can be suitably used as geopolymer feedstocks [11]. As of now, 6.3. Dimensional stability researchers are focusing on the optimization of blended geopoly- mers involving GGBFS/POFA [22,114], FA/POFA [23,45,90,91]. A The drying shrinkage behavior of dry wastepaper sludge added- ternary blended geopolymer system of GGBFS/POFA/RHA was also FA geopolymer mortars was monitored up to 91 days by Yan and proposed [116]. Various factors such as H2O/Na2O ratio [113], Ms Sagoe-Crentsil [36]. The dry wastepaper sludge was used in the [37], curing regime and chemical activator dosage [115] were range of 0–10 wt.% as sand replacement material. Results showed investigated on GGBFS/POFA blended geopolymer system and that the addition of dry wastepaper sludge up to 10 wt.% reduces excellent mechanical properties were obtained from the optimum the drying shrinkage of the resultant geopolymer mortars com- mixes from each of the factors studied. The excellent mechanical pared with the reference mortars which contain only FA as the bin- properties were due to the formation of dense geopolymer matrix der phase material. The drying shrinkage of 91 days cured mortars and the co-existence of C-S-H gels and geopolymeric gels (NASH/ specimens of 10 wt.% dry wastepaper sludge is 492 le, which cor- 392 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 respond to a remarkable 64% of reduction if compared with the ref- (9) Generally, N-based superplasticizers is the most effective erence mortars (1346 le). The reduction in drying shrinkage with high range water reducing admixtures for geopolymer system the increasing dry wastepaper sludge was due to the presence of activated by NaOH only. For geopolymer system activated by cellulose fibers, which existed predominantly in dry wastepaper both Na2SiO3 and NaOH, PC-based plasticizers exhibited the sludge. The expansion of cellulose fibers in the presence of mois- best plasticizing effect among other commercial superplasti- ture compensated the actual drying shrinkage of geopolymer mor- cizes e.g. N-based and M-based superplasticizers. tars specimens incorporating dry wastepaper sludge, which is (10) Hybridization of different source of binder materials with expected to exhibits a higher degree of drying shrinkage based the optimized proportion of chemical activators and opti- on the results of moisture loss analysis. mum curing method depending on the suitability of source Yusuf et al. [112] investigated various factor influencing the materials can lead to the production of geopolymer concrete shrinkage behavior of GGBFS/POFA blended geopolymers mortar. with superior mechanical, physical, durability and micro- The factors studied included the effect of GGBFS addition, effect structural properties. of SS/SH ratio and also the effect of pore sizes and volume of base (11) Geopolymer concrete exhibited superior resistance upon materials. The following conclusions can be drawn from the study, exposure to aggressive environment such as sulfuric acid, (i) internal micro-cracks and enhancement in pore filling and pore magnesium sulfate, nitric acid immersion etc with an aver- refinement which resulted in significant reduced shrinkage can be age of 12–40% of compressive strength loss compared to achieved with the addition of GGBFS up to 60%, (ii) SS/SH ratio of OPC concrete which generally exhibited 40–65% of compres- 2.5 was recommended for GGBFS/POFA blended geopolymer sys- sive strength loss in a same exposure period. tem, where enhancement in product glassy phase, reduction in car- (12) The effect of water content in geopolymer concrete towards bonation and tendency of C-A-S-H gel formation resulted in the strength development generally governed by the type of significantly reduced shrinkage. chemical activators used. For example, for FA-based geo- polymer, the strength increased with the decrease water/ 7. Summary of the current body of knowledge and FA ratio for NaOH, Na2SiO3 and Ca(OH)2 activated FA-based identification of challenges faced in future development of geopolymer. On the other hand, the reversed trend was geopolymer technology for industrial applications observed for FA-based geopolymer activated by KOH. (13) In the fabrication of press-formed geopolymer bricks, an Following the reviews performed on the essential factors influ- appropriate forming pressure that minimized the interparti- encing the properties of geopolymers derived from industrial by- cle spacing in the interfacial transition zone (ITZ) and also products, the following summaries can be made: the leaching out of the alkaline pore solution will yield geo- polymer bricks with excellent compressive strength (1) Generally the inclusions of 2–14 M of NaOH as chemical (P30 MPa) and low water absorption (2–5%). activators into the geopolymer matrix increased the com- pressive strength of the hardened geopolymer concrete. OPC concrete has received wide criticisms over the past decades (2) The multi-compound chemical activators consisted of Na2- due to its inherent high embodied energy and carbon footprint SiO3/NaOH was found to be the most effective chemical acti- which prompted the rise of geopolymers technology in recent years. vators to be added into the geopolymer matrix for the Geopolymer was seen as the solution to both construction industry purpose of mechanical strength enhancement. and also the waste management issues suffered by various indus- (3) Coupled with the addition of chemical activators, the appli- tries such as coal burning industry, palm oil, rice milling industry cation of heat curing in terms of duration and temperature is etc. However, despite the intensified researches performed over essential in accelerating the early age strength development the years, geopolymers technology is still no way near in replacing of geopolymer concrete. A maximum duration of 24 h and a OPC in actual industrial practices. Based on the reviews performed, heat curing temperature range of 50–90 °C was found to be the most influencing factors which governed the mechanical, phys- beneficial towards the short and long term strength develop- ical, durability and microstructure performance of geopolymers are ment and stability of geopolymer concrete. the alkaline activator, curing regime and also the source material’s (4) Higher liquid alkaline/ash ratio generally resulted in higher physical properties and chemical composition. workability of the fresh geopolymer mixture. Higher Na2- A recent study which compares the carbon footprint of OPC and SiO3/NaOH ratio generally reduced the workability of the geopolymer binders has casted doubt on the actual carbon foot- fresh mixtures due to the higher viscosity of Na2SiO3. While print of geopolymer binders [118,119]. It was reported the actual higher concentration of NaOH was found to increases the difference in carbon footprint between geopolymer binders and setting time of the resultant geopolymer mixture. OPC binders is just 9%, as opposed to widespread claim that the (5) Generally the shrinkage of geopolymer concrete is governed production of geopolymer binders reduced the carbon footprint by the corresponding strength i.e. increase in shrinkage is of OPC binders by as high as 60% [120]. In order to achieve similar associated with the low strength development of the or better properties in comparison with OPC concrete, most of the geopolymers. geopolymers require either the utilization of high dosage of alka- (6) In calcium-added or high calcium geopolymeric system, C- line activator [13] or heat treatment [1], or in many cases the appli- S-H gels were found to be co-existed with the geopoly- cation of both [4,18,45]. It is believed that the aforementioned meric gels and enhanced the microstructural and mechan- factors proved to be the stumbling blocks in the transition of geo- ical strength development of the resulting geopolymer polymers technology from research basis towards industrial appli- concrete. cation. The energy incurred in producing the alkaline activators (7) High strength (>60 MPa) geopolymer concrete can be fabri- namely sodium silicate and sodium hydroxide is intensive, cated with the higher fineness of the binder phase materials. reported to be 5.37 and 20.5 MJ/kg, respectively, which is higher (8) Small amount (63% by binder weight) nano-sized particle if compared to the embodied energy of OPC. The alkaline activators additions to the geopolymer matrix resulted in significant constituted approximately 60% of the total carbon footprint in the enhancement in mechanical and microstructural properties manufacture of geopolymer concrete [118]. Also, the highly alka- of the geopolymer concrete and caused reduction in both line in nature of the fresh geopolymer mixtures may poses prob- initial and final setting time of the geopolymer paste. lems during handling process. The necessity of heat treatment to W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 393 Fig. 11. Morphology of 7 days specimens: (a) 60/40 blend PCC-FA: FBC-FA; (b) 60/40 PCC-FA: FBC-FA blended geopolymer; (c) PCC-FA geopolymer paste; (d) FBC-FA geopolymer paste [111]. achieve desired strength further adds to the energy requirement of Utilization of advanced analytical methods such as nuclear geopolymer concrete fabrication. magnetic resonance (NMR) technique to elucidate the struc- Another potential challenges in commercializing geopolymers tural unit of the amorphous geopolymer products formed by technology is the inconsistency in properties and performance single or hybridized source materials which can’t be derived shown by various geopolymer source materials. It is well known quantitatively using other analytical methods such as XRD. that the performance of geopolymers is very much governed by Life cycle assessment (LCA) of each geopolymers concrete the properties of the source material itself i.e. chemical composi- developed should be derived in order to truly justify the envi- tion and physical properties. Popular geopolymer source materials ronmental and economical benefits offered by geopolymeric originated from industry wastes such as FA, GGBFS, RHA and POFA binder over OPC concrete. have their own unique chemical composition and physical proper- ties thus require distinctly different alkaline activator dosage and 8. Conclusions processing methods in order to achieve similar performance. Fur- thermore, the properties of the same source materials, but from The present work reviews and summarizes the essential factors different location possessed different characteristics in term of which have considerable effect on the properties of geopolymers chemical composition and physical properties. The aforemen- derived from various industrial by-products. Besides, the numer- tioned variations will definitely pose problems when transferring ous challenges and issues faced by the practitioner of geopolymers geopolymer knowledge to the industrial practitioners. technology and the steps needed to overcome the barriers were In light with the various challenges faced in the field of geopoly- highlighted and discussed as well. mers concrete, the following strategies are proposed to facilitate The present literatures provide a detailed elucidation of various the implementation of geopolymers technology in industrial factors that influence the properties of geopolymers concrete. With applications: reference to the current body of knowledge in geopolymer technol- ogy, rigorous amount of study has been performed to cover the Future researches should focus on fabricating geopolymers con- various aspects of established geopolymers such as FA and GGBFS crete with minimal alkaline activators dosage and high temper- based geopolymers. However, there is still a significant gap of ature treatment in order to produce a sustainable product with knowledge related to the geopolymerization reaction kinetics, low embodied energy, low carbon footprint, low in production material properties and rheological behavior of a number of cost and safe for site handling. emerging geopolymers such blended geopolymers and biomass Development of new activation and curing methods such as uti- ash geopolymers. Hence, detailed studies such as those related to lizing waste materials e.g. sodic waste or other additives which the derivation and modeling of reaction kinetics under various possessed similar properties with the commercial alkaline treatment and fabrication conditions of the emerging class of geo- activators. polymers source materials such as POFA, RHA and blended geo- Establishment of standard specification and testing methods polymers are required. Besides, contradicting findings in terms of designed specifically for geopolymers might be one of the few embodied energy and carbon footprint of geopolymers in compar- steps in convincing the widespread acceptance of geopolymers ison with conventional Portland cement must be addressed. Hence, technology in replacing OPC concrete. the development of new geopolymer material design, fabrication A complete elucidation and modeling of geopolymerization and post fabrication treatment technology which are oriented reaction kinetics and chemistry based on different source mate- towards minimizing the production cost, embodied energy and rials to serves as a general guideline for the geopolymers carbon footprint is essential to ensure the sustainability and future researchers in identifying the crucial parameters and factors implementation of geopolymers technology as a useful construc- to be considered during the design and fabrication stage. tion and manufacturing material. 394 W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 Successful utilization of geopolymers derived from industrial by [24] Ahmari S, Zhang L. Production of eco-friendly bricks from copper mine tailings through geopolymerization. Constr Build Mater 2012;29:323–31. products in actual industrial application will bring about numerous [25] Chen Y, Zhang Y, Chen T, Zhao Y, Bao S. Preparation of eco-friendly benefits towards construction industry and solve various industry construction bricks from hematite tailings. Constr Build Mater waste management issues. It is a promising sign that the 2011;25(4):2107–11. researches in geopolymers especially in the utilization of industrial [26] de Vargas AS, Dal Molin DCC, Vilela ACF, Silva FJD, Pavao B, Veit H. The effects of Na2O/SiO2 molar ratio, curing temperature and age on compressive waste materials have been intensified and it is certainly a step for- strength, morphology and microstructure of alkali-activated fly ash-based ward in full implementation of geopolymers in the construction geopolymers. Cem Concr Compos 2011;33(6):653–60. industry. [27] Hu M, Zhu X, Long F. Alkali-activated fly ash-based geopolymers with zeolite or bentonite as additives. Cem Concr Compos 2009;31(10):762–8. [28] Panias D, Giannopoulou IP, Perraki T. Effect of synthesis parameters on the mechanical properties of fly ash-based geopolymers. Colloids Surf A Acknowledgement 2007;301(1–3):246–54. [29] Somna K, Jaturapitakkul C, Kajitvichyanukul P, Chindaprasirt P. NaOH- activated ground fly ash geopolymer cured at ambient temperature. Fuel The authors would like to acknowledge the fundings of 2011;90(6):2118–24. the study by University Sains Malaysia Research University [30] Sathonsaowaphak A, Chindaprasirt P, Pimraksa K. Workability and strength of (RU-1001/PPBGN/814211) grant and Malaysian Ministry of Educa- lignite bottom ash geopolymer mortar. J Hazard Mater 2009;168(1):44–50. tion under the Fundamental Research Grant Scheme (FRGS-203/ [31] Komljenovic M, Bascarevic Z, Bradic V. Mechanical and microstructural properties of alkali-activated fly ash geopolymers. J Hazard Mater PPBGN/6711347) . 2010;181(1–3):35–42. [32] North MR, Swaddle TW. Kinetics of silciate exchange in alkaline aluminosilicate solutions. Inorg Chem 2000;39(12):2661–5. References [33] Sukmak P, Horpibulsuk S, Shen S-L. Strength development in clay-fly ash geopolymer. Constr Build Mater 2013;40:566–74. [34] Ridtirud C, Chindaprasirt P, Pimraksa K. Factors affecting the shrinkage of fly [1] Ahmari S, Ren X, Toufigh V, Zhang L. Production of geopolymeric binder from ash geopolymers. Int J Miner Metall Mater 2011;18(1):100–4. blended waste concrete powder and fly ash. Constr Build Mater [35] Salih MA, Abang Ali AA, Farzadnia N. Characterization of mechanical and 2012;35:718–29. microstructural properties of palm oil fuel ash geopolymer cement paste. [2] Schneider M, Romer M, Tschudin M, Bolio H. Sustainable cement production – Constr Build Mater 2014;65:592–603. present and future. Cem Concr Res 2011;41(7):642–50. [36] Yan S, Sagoe-Crentsil K. Properties of wastepaper sludge in geopolymer [3] Davidovits J. Global warming impact on the cement and aggregate industries. mortars for masonry applications. J Environ Manage 2012;112:27–32. World Resour Rev 1994;6:263–78. [37] Yusuf M, Megat Johari M, Ahmad Z, Maslehuddin M. Impacts of silica [4] Yusuf MO, Johari MAM, Ahmad ZA, Maslehuddin M. Strength and modulus on the early strength of alkaline activated ground slag/ultrafine microstructure of alkali-activated binary blended binder containing palm palm oil fuel ash based concrete. Mater Struct 2014:1–9. oil fuel ash and ground blast-furnace slag. Constr Build Mater 2014; [38] Law D, Adam A, Molyneaux T, Patnaikuni I, Wardhono A. Long term durability 52:504–10. properties of class F fly ash geopolymer concrete. Mater Struct 2014:1–11. [5] Damtoft JS, Lukasik J, Herfort D, Sorrentino D, Gartner EM. Sustainable [39] Guo X, Shi H, Dick WA. Compressive strength and microstructural development and climate change initiatives. Cem Concr Res 2008; characteristics of class C fly ash geopolymer. Cem Concr Compos 38(2):115–27. 2010;32(2):142–7. [6] Cheah CB, Ramli M. Mechanical strength, durability and drying shrinkage of [40] Topark-Ngarm P, Chindaprasirt P, Sata V. Setting time, strength and bond of structural mortar containing HCWA as partial replacement of cement. Constr high calcium fly ash geopolymer concrete. J Mater Civ Eng 2014. Build Mater 2012;30:320–9. [41] Khandelwal M, Ranjith PG, Pan Z, Sanjayan JG. Effect of strain rate on strength [7] Kroehong W, Sinsiri T, Jaturapitakkul C, Chindaprasirt P. Effect of palm oil fuel properties of low-calcium fly-ash-based geopolymer mortar under dry ash fineness on the microstructure of blended cement paste. Constr Build condition. Arab J Geosci 2013;6(7):2383–9. Mater 2011;25(11):4095–104. [42] Olivia M, Nikraz H. Properties of fly ash geopolymer concrete designed by [8] Nath P, Sarker P. Effect of fly ash on the durability properties of high strength Taguchi method. Mater Des 2012;36:191–8. concrete. Proc Eng 2011;14:1149–56. [43] Sofi M, van Deventer JSJ, Mendis PA, Lukey GC. Engineering properties of [9] Cheah CB, Ramli M. The engineering properties of high performance concrete inorganic polymer concretes (IPCs). Cem Concr Res 2007;37(2):251–7. with HCWA-DSF supplementary binder. Constr Build Mater 2013;40:93–103. [44] Ryu GS, Lee YB, Koh KT, Chung YS. The mechanical properties of fly ash-based [10] Cheah CB, Ramli M. The fluid transport properties of HCWA-DSF hybrid geopolymer concrete with alkaline activators. Constr Build Mater supplementary binder mortar. Composites B Eng 2014;56:681–90. 2013;47:409–18. [11] Mijarsh MJA, Megat Johari MA, Ahmad ZA. Synthesis of geopolymer from [45] Ranjbar N, Mehrali M, Alengaram UJ, Metselaar HSC, Jumaat MZ. Compressive large amounts of treated palm oil fuel ash: application of the Taguchi method strength and microstructural analysis of fly ash/palm oil fuel ash based in investigating the main parameters affecting compressive strength. Constr geopolymer mortar under elevated temperatures. Constr Build Mater Build Mater 2014;52:473–81. 2014;65:114–21. [12] Hanjitsuwan S, Hunpratub S, Thongbai P, Maensiri S, Sata V, Chindaprasirt P. [46] Khater HM. Effect of calcium on geopolymerization of aluminosilicate wastes. Effects of NaOH concentrations on physical and electrical properties of high J Mater Civ Eng 2011;24(1):92–101. calcium fly ash geopolymer paste. Cem Concr Compos 2014;45:9–14. [47] Chindaprasirt P, Rattanasak U, Taebuanhuad S. Resistance to acid and sulfate [13] Phoo-ngernkham T, Chindaprasirt P, Sata V, Hanjitsuwan S, Hatanaka S. The solution of microwave-assisted high calcium fly ash geopolymer. Mater effect of adding nano-SiO2 and nano-Al2O3 on properties of high calcium fly Struct 2013:375–81. ash geopolymer cured at ambient temperature. Mater Des 2014;55:58–65. [48] Chindaprasirt P, Chareerat T, Hatanaka S, Cao T. High strength geopolymer [14] He J, Jie Y, Zhang J, Yu Y, Zhang G. Synthesis and characterization of red mud using fine high calcium fly ash. J Mater Civ Eng 2010;23(3):264–70. and rice husk ash-based geopolymer composites. Cem Concr Compos [49] Chindaprasirt P, Rattanasak U, Vongvoradit P, Jenjirapanya S. Thermal 2013;37:108–18. treatment and utilization of Al-rich waste in high calcium fly ash [15] Silva PD, Sagoe-Crenstil K, Sirivivatnanon V. Kinetics of geopolymerization: geopolymeric materials. Int J Miner Metall Mater 2012;19(9):872–8. role of Al2O3 and SiO2. Cem Concr Res 2007;37(4):512–8. [50] Bakharev T. Geopolymeric materials prepared using Class F fly ash and [16] Davidovits J. Properties of geopolymer cements, alkaline cements and elevated temperature curing. Cem Concr Res 2005;35(6):1224–32. concretes. Kiev, Ukraine; 1994. [51] Perera DS, Uchida O, Vance ER, Finnie KS. Influence of curing schedule on the [17] Joseph B, Mathew G. Influence of aggregate content on the behaviour of fly integrity of geopolymers. J Mater Sci 2007;42(9):3099–106. ash based geopolymer concrete. Sci Iranica 2012;19(5):1188–94. [52] Puertas F, Fernandez-Jimenez A, Blanco-Varela MT. Pore solution in alkali- [18] Gorhan G, Kurklu G. The Influence of the NaOH solution on the properties of activated slag cement pastes. Relation to the composition and structure of the fly ash – based geopolymer mortar cured at different temperatures. calcium silicate hydrate. Cem Concr Res 2004;34(1):139–48. Composites B Eng 2013. [53] Mishra A, Choudhary D, Jain N, Kumar M, Sharda N, Dutt D. Effect of [19] Giasuddin HM, Sanjayan JG, Ranjith PG. Strength of geopolymer cured in concentration of alkaline liquid and curing time on strength and water saline water in ambient conditions. Fuel 2013;107:34–9. absorption of geopolymer concrete. ARPN J Eng Appl Sci 2008; [20] Nazari A, Bagheri A, Riahi S. Properties of geopolymer with seeded fly ash and 3(1):14–8. rice husk bark ash. Mater Sci Eng A 2011;528(24):7395–401. [54] Jeyasehar CA, Saravanan G, Ramakrishnan Kandasamy AK. Strength and [21] Aydin S, Baradan B. Mechanical and microstructural properties of heat cured durability studies on fly ash based geopolymer bricks. Asian J Civ Eng alkali-activated slag mortars. Mater Des 2012;35:374–83. 2013;14(6):797–808. [22] Islam A, Johnson Alengaram U, Jumaat MZ, Bashar II. The development of [55] Rattanasak U, Pankhet K, Chindaprasirt P. Effect of chemical admixtures on compressive strength of ground granulated blast furnace slag-palm oil fuel properties of high-calcium fly ash geopolymer. Int J Miner Metall Mater ash-fly ash based geopolymer mortar. Mater Des 2014;56:833–41. 2011;18(3):364–9. [23] Ranjbar N, Mehrali M, Behnia A, Johnson Alengaram U, Jumaat MZ. [56] Alonso S, Palomo A. Calorimetric study of alkaline activation of calcium Compressive strength and microstructural analysis of fly ash/palm oil fuel hydroxide-metakaolin solid mixtures. Cem Concr Res 2001;31(1):25–30. ash based geopolymer mortar. Mater Des 2014;59:532–9. W.K. Part et al. / Construction and Building Materials 77 (2015) 370–395 395 [57] Rattanasak U, Chindaprasirt P. Influence of NaOH solution on the syntheis of [90] Ariffin MAM, Bhutta MAR, Hussin MW, Mohd Tahir M, Aziah N. Sulfuric acid fly ash geopolymer. Miner Eng 2009;22(12):1073–8. resistance of blended ash geopolymer concrete. Constr Build Mater [58] Chindaprasirt P, De Silva P, Sagoe-Crentsil K, Hanjitsuwan S. Effect of SiO2 2013;43:80–6. and Al2O3 on the setting and hardening of high calcium fly ash-based [91] Bhutta M, Hussin M, Ariffin M, Tahir M. Sulphate resistance of geopolymer geopolymer systems. J Mater Sci 2012;47(12):4876–83. concrete prepared from blended waste fuel ash. J Mater Civ Eng 2014. [59] Sata V, Sathonsaowaphak A, Chindaprasirt P. Resistance of lignite bottom ash [92] Ahmari S, Zhang L. Durability and leaching behavior of mine tailings-based geopolymer mortar to sulfate and sulfuric acid attack. Cem Concr Compos geopolymer bricks. Constr Build Mater 2013;44:743–50. 2012;34(5):700–8. [93] lloyd RR, Provis JL, van Deventer JSJ. Acid resistance of inorganic binders. 1. [60] Detphan S, Chindaprasirt P. Preparation of fly ash and rice husk ash Corrosion rate. Mater Struct 2012;45:1–14. geopolymer. Int J Miner Metall Mater 2009;16(6):720–6. [94] Reddy DV, Edouard JB, Sobhan K. Durability of fly-ash based geopolymer [61] Rashad AM. A comprehensive overview about the influence of different structural concrete in the marine environment. J Mater Civ Eng admixtures and additives on the properties of alkali-activated fly ash. Mater 2012;25(6):781–7. Des 2014;53:1005–25. [95] Bakharev T. Resistance of geopolymer materials to acid attack. Cem Concr Res [62] Rattanasak U, Chindaprasirt P, Suwanvitaya P. Development of high volume 2005;35(4):658–70. rice husk ash alumino silicate composites. Int J Miner Metall Mater [96] Chindaprasirt P, Rukzon S. Strength, porosity and corrosion resistance of 2010;17(5):654–9. ternary blend Portland cement, rice husk ash and fly ash mortar. Constr Build [63] Boonserm K, Sata V, Pimraksa K, Chindaprasirt P. Improved Mater 2008;22(8):1601–6. geopolymerization of bottom ash by incorporating fly ash and using waste [97] Bascarevic Z, Komljenovic M, Miladinovic Z, Nikolic V, Marjanovic N, Petrovic gypsum as additive. Cem Concr Compos 2012;34(7):819–24. R. Impact of sodium sulfate solution on mechanical properties and structure [64] Alomayri T, Shaikh FUA, Low IM. Synthesis and mechanical properties of of fly ash based geopolymers. Mater Struct 2014:1–15. cotton fabric reinforced geopolymer composites. Composites B Eng [98] Maes M, De Belie N. Resistance of concrete and mortar against combined 2014;60:36–42. attack of chloride and sodium sulphate. Cem Concr Compos 2014;53:59–72. [65] Nematollahi B, Sanjayan J. Effect of different superplasticizers and activator [99] Rangan BV, Wallah SE. Low calcium fly ash based geopolymer concrete: long combinations on workability and strength of fly ash based geopolymer. Mater term properties. Research Report GC2, Faculty of Engineering, Curtin Des 2014;57:667–72. Univeristy of Technology, Perth, 2006. [66] Chen R, Ahmari S, Zhang L. Utilization of sweet sorghum fiberto reinforce fly [100] Zhao F-q, Zhao J, Liu H-j. Autoclaved brick from low-silicon tailings. Constr ash-based geopolymer. J Mater Sci 2014;49:2548–58. Build Mater 2009;23(1):538–41. [67] Zhang YJ, Wang YC, Xu DL, Li S. Mechanical performance and hydration [101] Cheah CB, Noor Shazea AN, Part WK, Ramli M, Kwan WH. The high volume mechanism of geopolymer composite reinforced by resin. Mater Sci Eng A reuse of hybrid biomass ash as a primary binder in cementless mortar block. 2010;527(24–25):6574–80. Am J Appl Sci 2014;11(8):1369–78. [68] Chindaprasirt P, Jaturapitakkul C, Chalee W, Rattanasak U. Comparative study [102] Chindaprasirt P, Chareerat T, Sirivivatnanon V. Workability and strength of on the characteristics of fly ash and bottom ash geopolymers. Waste Manage coarse high calcium fly ash geopolymer. Cem Concr Compos 2007;29:224–9. 2009;29(2):539–43. [103] Davidovits J. Mineral polymers and methods of making them. US Patent [69] Jaturapitakkul C, Cheerarot R. Development of bottom ash as pozzolanic 43493861982. material. J Mater Civ Eng 2003;15:48–53. [104] Freidin C. Cementless pressed blocks from waste products of coal-firing [70] Li G. Properties of high-volume fly ash concrete incorporating nano-SiO2. Cem power station. Constr Build Mater 2007;21(1):12–8. Concr Res 2004;34(6):1043–9. [105] Deb PS, Nath P, Sarker PK. The effects of ground granulated blast-furnace slag [71] Kong DLY, Sanjayan JG. Effect of elevated temperatures on geopolymer paste, blending with fly ash and activator content on the workability and strength mortar and concrete. Cem Concr Res 2010;40(2):334–9. properties of geopolymer concrete cured at ambient temperature. Mater Des [72] Palacios M, Puertas F. Effect of superplasticizer and shrinkage-reducing 2014;62:32–9. admixtures on alkali-activated slag pastes and mortars. Cem Concr Res [106] Xu H, Gong W, Syltebo L, Izzo K, Lutze W, Pegg IL. Effect of blast furnace slag 2005;35(7):1358–67. grades on fly ash based geopolymer waste forms. Fuel 2014;133:332–40. [73] Palacios M, Houst YF, Bowen P, Puertas F. Adsorption of superplasticizer [107] Canfield GM, Eichler J, Griffith K, Hearn JD. The role of calcium in blended fly admixtures on alkali-activated slag pastes. Cem Concr Res 2009;39(8):670–7. ash geopolymers. J Mater Sci 2014;49:5922–33. [74] Puertas F, Palomo A, Fernandez-Jimenez A, Izquierdo M, Granizo M. Effect of [108] Songpiriyakij S, Kubprasit T, Jaturapitakkul C, Chindaprasirt P. Compressive superplasticisers on the behaviour and properties of alkaline cements. Adv strength and degree of reaction of biomass- and fly ash-based geopolymer. Cem Res 2003;15(1):23–8. Constr Build Mater 2010;24(3):236–40. [75] Nath SK, Kumar S. Influence of iron making slags on strength and [109] Yan S, Sagoe-Crentsil K, Shapiro G. Reuse of de-inking sludge from microstructure of fly ash geopolymer. Constr Build Mater 2013;38:924–30. wastepaper recycling in cement mortar products. J Environ Manage [76] Sun P, Wu H-C. Transition from brittle to ductile behavior of fly ash using PVA 2011;92(8):2085–90. fibers. Cem Concr Compos 2008;30(1):29–36. [110] Naik T. Greener concrete using recycled materials. Concr Int 2002;7:45–9. [77] He P, Jia D, Lin T, Wang M, Zhou Y. Effects of high-temperature heat treatment [111] Chindaprasirt P, Rattanasak U, Jaturapitakkul C. Utilization of fly ash blends on the mechanical properties of unidirectional carbon fiber reinforced from pulverized coal and fluidized bed combustions in geopolymeric geopolymer composites. Ceram Int 2010;36(4):1447–53. materials. Cem Concr Compos 2011;33:55–60. [78] Zhao Q, Nair B, Rahimian T, Balaguru P. Novel geopolymer based composites [112] Yusuf MO, Megat Johari MA, Ahmad ZA, Maslehuddin M. Shrinkage and with enhanced ductility. J Mater Sci 2007;42(9):3131–7. strength of alkaline activated ground steel slag/ultrafine palm oil fuel ash [79] Memon FA, Nuruddin MF, Shafiq N. Effect of silica fume on the fresh and pastes and mortars. Mater Des 2014;63:710–8. hardened properties of fly ash-based self-compacting geopolymer concrete. [113] Yusuf MO, Megat Johari MA, Ahmad ZA, Maslehuddin M. Effects of H2O/Na2O Int J Miner Metall Mater 2013;20(2) [205-3]. molar ratio on the strength of alkaline activated ground blast furnace slag- [80] Ramakrishna G, Sundararajan T. Impact strength of a few natural fibre ultrafine palm oil fuel ash based concrete. Mater Des 2014;56:158–64. reinforced cement mortar slabs: a comparative study. Cem Concr Compos [114] Yusuf MO, Megat Johari MA, Ahmad ZA, Maslehuddin M. Evolution of alkaline 2005;27(5):547–53. activated ground blast furnace slag-ultrafine palm oil fuel ash based concrete. [81] Alomayri T, Shaikh FUA, Low IM. Effect of fabric orientation on mechanical Mater Des 2014;55:387–93. properties of cotton fabric reinforced geopolymer composites. Mater Des [115] Yusuf MO, Megat Johari MA, Ahmad ZA, Maslehuddin M. Influence of curing 2014;57:360–5. methods and concentration of NaOH on strength of the synthesized alkaline [82] Alomayri T, Low IM. Synthesis and characterization of mechanical properties activated ground slag-ultrafine palm oil fuel ash mortar/concrete. Constr in cotton fiber-reinforced geopolymer composites. J Asian Ceram Soc Build Mater 2014;66:541–8. 2013;1(1):30–4. [116] Karim MR, Zain MFM, Jamil M, Lai FC. Fabrication of a non-cement binder [83] Zhang YJ, Li S, Wang YC, Xu DL. Microstructural and strength evolutions of using slag, palm oil fuel ash and rice husk ash with sodium hydroxide. Constr geopolymer composite reinforced by resin exposed to elevated temperature. J Build Mater 2013;49:894–902. Non-Cryst Solids 2012;358(3):620–4. [117] Chindaprasirt P, Homwuttiwong S, Jaturapitakkul C. Strength and water [84] Fletcher RA, MacKenzie KJD, Nicholson CL, Shimada S. The composition range permeability of concrete containing palm oil fuel ash and rice husk bark ash. of aluminosilicate geopolymers. J Eur Ceram Soc 2005;25:1471. Constr Build Mater 2007;21(7):1492–9. [85] Gutti CS, Roy A, Metcalf JB, Seals RK. The influence of admixtures on the [118] Turner LK, Collins FG. Carbon dioxide equivalent (CO2-e) emissions: a strength and linear expansion of cement-stabilized phosphogypsum. Cem comparison between geopolymer and OPC cement concrete. Constr Build Concr Res 1996;26(7):1083–94. Mater 2013;43:125–30. [86] Talero R. Comparative XRD analysis ettringite originating from pozzolan and [119] Habert G, d’Espinose de Lacaillerie JB, Roussel N. An environmental from portland cement. Cem Concr Res 1996;26(8):1277–83. evaluation of geopolymer based concrete production: reviewing current [87] Socrates G. Infrared and Raman characteristic group frequencies. third research trends. J Cleaner Prod 2011;19(11):1229–38. ed. England: John Wiley & Sons; 2001. [120] Duxson P, Provis JL, Lukey GC, van Deventer JSJ. The role of inorganic polymer [88] Komnitsas K, Zaharaki D. Geopolymerisation: a review and prospects for the technology in the development of ‘green concrete’. Cem Concr Res minerals industry. Miner Eng 2007;20(14):1261–77. 2007;37(12):1590–7. [89] Bakharev T. Durability of geopolymer materials in sodium and magnesium sulfate solutions. Cem Concr Res 2005;35(6):1233–46.