Mechanical properties and microstructure analysis of fly ash geopolymeric recycled concrete

Journal of Hazardous Materials 237–238 (2012) 20–29

Contents lists available at SciVerse ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Mechanical properties and microstructure analysis of fly ash geopolymeric recycled concrete X.S. Shi a , F.G. Collins b , X.L. Zhao b , Q.Y. Wang a,∗ a b

College of Architecture and Environment, Sichuan University, Chengdu, PR China Department of Civil Engineering, Monash University, Clayton, Australia

h i g h l i g h t s    

Sodium silicate solution and sodium hydroxide solution were used to activate fly ash, which substitute cement totally in the concrete. Utilizing two kinds of waste materials (fly ash and recycled aggregates) at the same time. The mechanical properties and microstructures were studied and compared with different recycled aggregates replacement ratios. Such concrete has greater compressive strength and better microstructure than ordinary concrete and also geopolymer concrete.

a r t i c l e

i n f o

Article history: Received 9 March 2012 Received in revised form 6 July 2012 Accepted 8 July 2012 Available online 25 August 2012 Keywords: Recycled aggregate Fly ash Alkali-activated Compressive strength SEM

a b s t r a c t Six mixtures with different recycled aggregate (RA) replacement ratios of 0%, 50% and 100% were designed to manufacture recycled aggregate concrete (RAC) and alkali-activated fly ash geopolymeric recycled concrete (GRC). The physical and mechanical properties were investigated indicating different performances from each other. Optical microscopy under transmitted light and scanning electron microscopy (SEM) coupled with energy dispersive X-ray spectroscopy (EDX) were carried out in this study in order to identify the mechanism underlying the effects of the geopolymer and RA on concrete properties. The features of aggregates, paste and interfacial transition zone (ITZ) were compared and discussed. Experimental results indicate that using alkali-activated fly ash geopolymer as replacement of ordinary Portland cement (OPC) effectively improved the compressive strength. With increasing of RA contents in both RAC and GRC, the compressive strength decreased gradually. The microstructure analysis shows that, on one hand, the presence of RA weakens the strength of the aggregates and the structure of ITZs; on the other hand, due to the alkali-activated fly ash in geopolymer concrete, the contents of Portlandite (Ca(OH)2 ) and voids were reduced, as well as improved the matrix homogeneity. The microstructure of GRC was changed by different reaction products, such as aluminosilicate gel. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Currently, environmental sustainable development has become a worldwide goal for human activities. As the demand for infrastructure development increases, the consumption of concrete as main construction material would keep growing. However, the emission of carbon oxide (CO2 ) from the production process of Portland cement (OPC) is estimated about 1.35 billion tons annually or about 7% of the total greenhouse gas emissions to the earth’s atmosphere [1]. Recently, lots of researches focus on searching for an alternative binder utilized in the concrete to reduce CO2 footprint. The term “Geopolymer” was firstly coined by Davidovits in 1970s [2]. Throughout last a few years’ efforts, research work on

∗ Corresponding author. Tel.: +86 13980955902; fax: +86 28 85405389. E-mail address: [email protected] (Q.Y. Wang). 0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2012.07.070

developing alkali-activated binders which can be made predominantly from industrial waste materials, such as fly ash show enormous potential to become an alternative to OPC, as well as exhibit superior chemical and mechanical properties to OPC [3–8]. On the other hand, the construction and demolition waste disposal problem has become impressing due to the conventional landfill method would take up lots of soil and cause many secondary pollutions. Recycled aggregate concrete (RAC) has been proposed to be an effective way to solve such a problem [9]. The inferior properties of RAC by utilizing recycled aggregate (RA) as the coarse aggregate in the new concrete are attributed to the adhered mortar on the RA which results in lower densities, higher water absorption and weaker strength [10–12]. While, the use of fly ash as a cement addition is found to mitigate its detrimental effect with appropriate mix design [13,14]. As well known, the mechanical behaviors of the materials mostly depend on its intrinsic microstructure. Concrete can be

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29 Table 1 Chemical composition of NA. Chemical composition Weight percentage (wt.%)

O 55.07

Si 25.19

Al 15.70

Ca 7.57

Na 3.95

Fe 0.71

recycled concrete (GRC) can combine the advantages of RAC and geopolymer concrete, as well as compensate their individual disadvantages. Moreover, reusing the waste concrete and fly ash in the new concrete would effectively reduce the environmental pollutions. Taken this into account, the mechanical properties of alkali-activated fly ash recycled concrete with RA replacement ratio of 0%, 50% and 100% were tested. Since, the microstructure of such GRCs is unknown so far, optical microscopy under transmitted light and scanning electron microscopy (SEM) coupled with energy dispersive X-ray spectroscopy (EDX) were conducted to better understand how the differences of the microstructure influence the mechanical properties in such new concretes.

Mg 0.18

Table 2 Chemical composition of fly ash. Constituent

Al2 O3

Property (%) 30.50

SiO2

CaO

Fe2 O3

48.30 2.80 12.10

K2 O

MgO Na2 O SO3

0.40 1.20

0.20

21

Loss on ignition 0.30 1.70

regarded as a three paste system composed with coarse aggregate, mortar matrix with fine aggregate, and interfacial transition zones (ITZs). The studies on microstructure of RAC have found that, the ITZ takes crucial role in the concrete performances [15–17]. Considering the geopolymer based on fly ash is a very heterogeneous material, the microstructure of geopolymer concrete has been intensively studied recently [6,18–22]. In most cases, most of the authors agree with the main hydration products of binders based on alkali-activated slag and fly ash are calcium-silicate-hydrates (C-S-H) and complex amorphous gel, such as aluminosilicate gel. However, the influence of recycled aggregates in geopolymer concrete on microstructure has not been reported yet. From the above viewpoints, to utilize RA as the coarse aggregate in fly ash based geopolymer concrete to form geopolymeric

2. Experimental program 2.1. Materials 2.1.1. Aggregates All the recycled aggregates were supplied by Alex Fraser P/L and the fine particles smaller than 4.75 mm were removed from the RA by sieving in the laboratory. The natural coarse aggregates (NA) were crushed old basalt aggregates from a local quarry in Pakenham, Victoria and fine aggregates consisted of river sand from Lyndhurst, Victoria, Australia. Recycled aggregate is made by washing and crushing from old concrete. According to EDX

Table 3 Summary of concrete mixture proportions (kg/m3 ). Mixture

RA ratio

RA/NA

Cement

Sand

RAC0 RAC50 RAC100 GRC0 GRC50 GRC100

0% 50% 100% 0% 50% 100%

0/1294 647/647 1294/0 0/1294 647/647 1294/0

364 364 364 – – –

554 554 554 554 554 554

Fly ash

NaOH solution

Na2 SiO3 solution

Added water

W/G ratio

368 368 368

– – – 53 53 53

– – – 131 131 131

182 182 182 0 0 0

0.50 0.50 0.50 0.50 0.50 0.50

Table 4 Compressive strength of different concretes with ages. Concrete

RAC0 RAC50 RAC100 GRC0 GRC50 GRC100

Compressive strength (MPa) 3-Day

SD

7-Day

SD

28-Day

SD

60-Day

SD

90-Day

SD

22.51 22.41 22.04 74.37 61.11 47.77

0.13 0.19 0.38 2.31 1.81 1.27

36.94 36.08 36.37 80.58 67.72 50.83

0.94 0.20 0.35 1.87 0.29 0.37

49.98 44.9 44.58 85.66 71.59 54.66

0.41 0.76 0.18 1.58 1.66 1.73

51.72 46.51 45.6 86.15 71.72 54.95

0.67 0.64 0.70 0.90 1.57 0.97

62.42 53.63 50.79 88.22 71.97 55.09

0.44 0.62 0.70 1.07 1.91 1.27

Fig. 1. Cracks inside of recycled concrete: (a) aggregates in RAC0; (b) aggregates in RAC100.

22

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

Fig. 2. Paste inside of RAC and GRC: (a) RAC100 under crossed polarized light; (b) RAC100 under plane polarized light; (c) GRC100 under crossed polarized light; (d) GRC100 under plane polarized light.

result of nature aggregate in Table 1, the main chemical composition is Si and Al, only a few quantities of Ca and Na. The tests to investigate the properties of aggregates on grading, densities and water absorption were in accordance with Australian Standard AS1141. The nominal sizes of RA and NA were 20 mm and 14 mm, respectively. The densities and water absorption of RA and NA were 2433 t/m3 , 2850 t/m3 and 5.60%, 1.08%, respectively. All the aggregates conformed to the requirements of AS2758.1 (1998) for use in production in concrete. The concrete mixture proportions were based on aggregates in saturated surface dry (SSD) conditions.

2.1.2. Fly ash Fly ash (ASTM Class F) was used as the main aluminum and silicate source for synthesizing of geopolymeric binder. It is mainly glassy with some crystalline inclusions of mullite, hematite and quartz. The chemical composition of the fly ash used in this experimental program is shown in Table 2.

3. Specimen preparation 3.1.1. Mixture design and manufacturing process Six mixtures were designed to compare the influence of different RA replacement ratios on RAC and GRC concrete. The concrete mixture proportions are summarized in Table 3. The OPC used in the experiments is Type GP sourced from Cement Australia, West Footscray, and conforms to the requirements of general purpose cement as defined by the AS3972. W/G is the ratio of total water to geopolymeric binder solids. The mixing of the concrete was conducted in a mechanical mixer according to the procedure in AS1012.2 (1994). After mixing, the mixture was poured into steel cylinder moulds with 100 mm diameter × 200 mm length for compressive strength by applying 20 manual strokes per layer in three equal layers on a vibration table.

3.1.2. Curing condition 2.1.3. Alkali solution Sodium silicate solution (Na2 SiO3 ) with specific gravity of 1.53 and sodium hydroxide (NaOH) flakes of 98% purity were supplied by PQ Australia. The chemical composition of the sodium silicate solution was Na2 O = 14.7%, SiO2 = 29.4% and water = 55.9% by mass. Sodium hydroxide was dissolved using distilled water to provide 8 molarity alkaline solutions. Na2 SiO3 and NaOH solutions were prepared one day prior to usage.

The specimens of RAC were cured under polyethylene sheets for 24 h in the laboratory environment. And then, demould and were transferred to a tank of saturated limewater at 23 ± 2 ◦ C as the moist-curing regime to satisfy AS1012.8.1 (2000). The specimens of GRC were sealed by plastic sheets to prevent excessive evaporation during curing at 80 ◦ C for 24 h in the oven, followed by removal of specimens from the moulds and storage in the ambient condition.

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

23

Fig. 3. Interfacial transition zone in RAC and GRC: (a) RAC0; (b) RAC100; (c) GRC0; (d) GRC100.

3.1.3. Petrographic thin sections The samples were taken from the core of the concrete cylinders. Thin sections of 75 mm × 50 mm in size with thickness of 30 ␮m were prepared in Ballarat University, Australia. The experiments were carried out under transmitted light on Nikon (Opitphot2-POL) in Monash University, Australia. 3.1.4. SEM samples The specimens were selected from the broken section of the 90days compressive experiments and studied with Hitachi S-4800 in Sichuan University, China. In order to investigate the ITZs, the specimens should be chosen with typical interfaces. All the specimens

were soaked in the absolute alcohol for one day, following by drying at 60 ◦ C for 15 min. The specimens were coated with Au prior to the analysis. 4. Results and discussions 4.1. Compressive strength The compressive strength tests were performed on 1800 kN electro-hydraulic mechanical testing machine in accordance with AS1012.9 (1999). Triplicate cylinder samples were sulfur-capped for each kind of concrete. The results of compressive strength of RAC and GRC with different RA replacement ratios at ages of 3, 7, 28, 60 and 90-days are shown in Table 4.

Fig. 4. SEM images of fly ash.

24

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

Fig. 5. The typical morphology of reaction products in the RAC and GRC: (a) RAC0; (b) RAC100; (c) GRC0; (d) GRC100.

The results of compressive strength show that, the overall strengths of GRC are significantly higher than that of RAC, especially for 3-days’ strength of GRC0 which exhibited about 3 times higher than RAC0. As the ages increasing, the strengths of RAC increased faster than GRC did. This is because polymerization reactions are highly temperature dependent. The elevated temperature curing condition accelerated the polymerization between fly ash with alkali solution to form binder in the early ages. After the concrete were removed from the oven to the ambient condition, the polymerization reaction became very slowly indicating the strengths grew inconspicuously. While, the strength improved gradually as the time went by for RACs. As expected, most of the strength development occurred during the first 7 days, followed by lower rate of strength development at later ages. Overall, the influence of RA replacement ratio to RAC’s compressive strength presented not very great. The decrease of compressive strength between RAC100 and RAC0 is about 10% for 28-day, and less than 20% for 90-day. However, the influence of RA replacement ratio to the GRC compressive strength is much greater, with the decline of 16% and 36% compared with ordinary geopolymer concrete for GRC50 and GRC100 respectively. The strengths between RAC50 and RAC100 are very close for all the ages, while, the decrease between GRC 50 and GRC100 always keeps about 23%. It is inferred that the recycled aggregates could not make positive effect in the chemical reaction process to improve the strength of geopolymer concrete, which is different from what they might be in RAC. 4.2. Petrographic analysis With optical microscope, plane polarized light and crossed polarized light were employed. The following were mainly analyzed: aggregates, cement paste, ITZs including interfaces between

new paste with coarse aggregates, new paste with old paste and old paste with coarse aggregate. Crossed polarized light was employed to define the natural aggregates mineralogy. It is inferred the granite aggregates composition was basically the feldspar group and quartz. The recycled aggregates were observed with far more cracks than natural aggregates, due to some damages were produced during the crushing procedure. This is the evident that the recycled aggregate was inferior to conventional aggregates. See Fig. 1. The paste in RAC and GRC are different due to different matrix formation process. In RAC, the paste is mainly the production of cement hydration, and some white shining crystals could be recognized as Portlandite (CH). However, in GRC, because of the chemical reaction between fly ash and alkali solution, the polymerization process involves a substantially fast chemical reaction under alkaline condition on Si Al minerals, that results in a three-dimensional polymeric chain and ring structure consisting of Si O Al O bonds, as follows [18]: Mn [ (SiO2 )z AlO2 ]n·wH2 O

(1)

where, M is the alkaline element; the symbol indicates the presence of a bond, z is 1, 2, or 3, and n is the degree of polymerization. The schematic formation of geopolymer material can be shown as described by the following Eqs. (2) and (3) [23]: n(Si2O5,Al2O2)+2nSiO2+4nH2O+NaOH

K++n(OH)3-Si-O-Al- -O-Si-(OH)3 (OH)2 (Geopolymer precursor)

(2)

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

25

Fig. 6. The SEM images of past in different concretes: (a) RAC0; (b) RAC50; (C) RAC100; (d) GRC0; (e) GRC50; (f) GRC100.

n(OH)3-Si-O-Al- -O-Si-(OH)3 + NaOH (OH)2

(Na+)-(-Si-O- Al- -O-Si-O-)+4nH2O O

O

O

(Geopolymer backbone)

(3) Fig. 2 shows typical microstructure obtained using transmitted light microscopy on thin sections at 100× magnification. The white small shining spots uniformly distributed in the paste or/and around the aggregates are Portlandite, while, the paste in the GRC shows a mostly homogeneous matrix. Compared with images employed by plane polarized light and crossed polarized light, the

holes can be figured out as the black one in crossed polarized picture but white one in plane polarized picture. Overall, the paste in GRC presented much denser with fewer pores. The interfacial transition zone (ITZ) is considered to have a fundamental affect on the strength of the concrete, which is the weakest point in the concrete with RA due to the adhered old mortar. There are two new ITZs in the concrete with recycled aggregates both in RAC and GRC. One is the interface between the adhered old mortar and new cement paste or the binder, the other one is between the adhered old mortar and original aggregate. In Fig. 3 ITZs in RAC and GRC are compared. With respect to recycled aggregates in the concrete, the ITZ presented not as dense as that in conventional concrete. Some cracks and voids could be observed in the ITZ between RA and the paste. During the hydration process, the Portlandite crystallizes in the voids of the cement matrix which is most porous at ITZ. Compared with RAC and GRC, the ITZ

26

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

Fig. 7. SEM images of ITZs in RAC: (a) RAC0-1; (b) RAC0-2; (c) RAC50-1; (d) RAC50-2; (e) RAC100-1; (f) RAC100-2.

looks much denser with fewer voids in GRC than that in RAC. With the strength of the concrete decreasing, the size and quantity of the voids increased, which would be more easily to form a weak transition zone. 4.3. SEM examination For SEM analysis, samples were taken from specimens that had been fractured during compression test, mounted in epoxy, polished, and Au coated. Fly ash (FA) is generally observed as a very heterogeneous material, which consists of fine, amorphous and reactive aluminosilicate particles. Class F fly ash, which is high in aluminosilicate and low in calcium, is preferred grade for inorganic polymerization. The chemical components are listed in Table 1, and SEM images of the fly ash precursor are shown in Fig. 4. It is shown that fly ash consisted of spherical solid particles, cenospheres

(hollow particles) and plerospheres with a broad size distribution, the majority of the particles ranging from 1 to 40 ␮m. The cenospheres present black spot on the top of the particles, while plerospheres are hollow particles of large diameter filled with smaller size particles. The color of fly ash is light grey which led to GRC looks darker than ordinary concrete. It is reported that the compressive strength of geopolymers predominantly depended on the content of fly ash fine particles, which smaller than 43 ␮m would be more active in alkali-activation process [22]. This is due to the finer particles have higher amorphous content which can get more chances to react with solution. Accordingly, the Class F fly ash used in this study is quite benefit for geopolymerization. 4.3.1. Matrix of the concrete The matrix formation process in GRC is different from that in OPC which results in different reaction products, and in turns

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

27

Fig. 8. SEM images of ITZs in GRC: (a) GRC0-1; (b) GRC0-2; (c) GRC50-1; (d) GRC50-2; (e) GRC100-1; (f) GRC100-2.

influences the compressive strength on macro scale. The primary phase composition of hydrated concrete includes the amorphous aluminosilicate gel and poorly crystalline phases with needle or column shape. In Fig. 5, the typical reaction products in RAC and GRC show that, the use of alkali-activated fly ash as replacement to cement led to significant changes in physical and chemical characteristics of the concrete. In the case of fly ash, the matrix appears denser in GRC. The fly ash particles are well embedded and connected to the matrix, and some of them are partly covered with needle-like phases. In recycled concrete, the CH crystalline presents as some nubby matrix in a layered structure with higher orientation. The C-S-H gel exhibits various phases, such as fibriform, floc, coralloid or snow flower, which interweave together. The structure of matrix in RAC is much looser with more voids than that in GRC matrix.

The morphology of matrix in RAC and GRC with different RA replacement ratios is shown in Fig. 6. All the tested samples were taken from the compression test with 90 days. Although with different RA contents, the reaction products are basically the same in the same concrete type. With the increasing of RA replacement ratio, the dense of the paste decreased. More hydrates with various appearances and more pores are observed in RAC100 and GRC100. The hydrated cement paste adhered to the surface of RA effects the mechanical and chemical performance of RA in new concrete. The pores increase the water absorption of RA and decrease the compressive strength of the concrete. In GRC concretes, many unreacted fly ash are embedded in the matrix that fills the pore space. On the other hand, due to the reaction of fly ash in alkali-activated solution, amorphous aluminosilicate gels are produced which are more stable hydration products, and also

28

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

Fig. 9. EDX analysis of the matrix in RAC and GRC ITZs: (a) EDX pattern of the calcium hydroxide in RAC0 ITZ; (b) EDX pattern of alkali silicate gel in GRC100 ITZ.

results in more uniform and homogenous matrix. Therefore, forming a striking contrast to RAC, there are less crystalline CH and much more gel-like phases with lower porosity in GRC, which induces the significant higher strength than corresponding RAC concretes. 4.3.2. Interfacial transition zone (ITZ) The ITZ is the weakest area in conventional concretes due to accumulation of water at this point. While, because of the adhered mortar onto the RAs, the ITZs in concrete with RAs would be with more complex microstructures and structural variety. The morphology of ITZs in RAC and GRC are shown in Figs. 7 and 8, respectively. Fig. 7 shows that in the vicinity of ITZ, some calcium silicate hydrates, amounts of crystalline calcium hydroxide and many pores could be observed. Compared with ordinary concrete RAC0, the recycled concrete’s ITZ appears more pores and looser structure, due to the adhesive mortar on the RA. Many CH crystals with high orientation are present in the ITZ. These crystals are easily to be broken because the directional alignment structure would easily to form the track of the crack. The cracks almost happened on the interface between aggregate and the paste, and then extended to the paste. In RAC100, the cracks also developed to the aggregate, due to the weaker strength of RA. Generally, the structure in ITZs in RACs is loose and mostly with a lot of voids. The differences in strength development between various RA content concretes are due to the differences in both the strength of coarse aggregate and microstructure properties of the ITZs. In Fig. 8, it can be found that with different RA contents, the morphology of ITZ in GRCs is similar. There are voids in the binder, some of which are filled with fly ash particles. And also, many

unreacted fly ash particles are visible in the paste. With more RAs in the concrete, the microstructure of ITZ consists of more pores and more reaction gels in the pattern of white granules. This is inferred because some adhered mortar on the RA which could result in more water and alkali solution accumulated around the aggregate. In this case, the reactive silica on the surface of the aggregate could be dissolved to form an alkali silica gel. The binder presents denser in the concrete with less RAs. The crack development in GRCs is similar to that in RACs, which almost happened on the interface and then extended to the paste. The composition of the reaction products in RAC and GRC is different according to the EDX analysis. Fig. 9 shows that, in RAC, Portlandite present much more in the paste which results in the composition of calcium is much higher. However, in GRC, the main reaction product is alkali silica reaction gel in which silica and aluminum take up the most percentage. Whereas, the concentration of silicon is the key factor in determining the strength of alkaliactivated fly ash based concrete which results in high compressive strength. Compared with RACs, the paste of GRC is denser with fewer pores. The particles in the ITZ are smaller and attach more tightly to the aggregate in GRC. Meanwhile, the fly ash particles could take the role of filler in the paste in GRC. Thus, from the microstructure point of view, it can be seen that GRC is stronger than the correspondence RAC. Overall, with respect to the use of alkali-activated fly ash as binder in the recycled concrete, the physical and chemical characteristics of concrete are changed. The physical changes include a significant decrease in the volume of voids, and an overall improvement of the microstructure density and homogeneity. The formation of new hydration products and the absence of calcium hydroxide are the main chemical changes that were observed.

X.S. Shi et al. / Journal of Hazardous Materials 237–238 (2012) 20–29

These changes result in such geopolymeric recycled concrete with higher compressive strength than OPC and RAC. 5. Conclusions The following broad conclusions can be drawn from the above results: 1. The compressive strengths of GRC with different RA contents are significantly higher than that of the corresponding RAC. With more RAs, the compressive strengths of RAC and GRC are decreased. The influence of RA on the strength is greater on GRC than that on RAC in this case. 2. Aggregate components, reaction products such as CH and C-S-H, pores, cracks and ITZs could be identified by optical microscope under transmitted light and SEM. Many CH white spots are in the paste and on the border of the aggregate in RAC. The appearance of RAC is quite different from that of GRC. The RAC paste is mainly composed of the amorphous or poorly crystalline calcium silicate hydrate (C-S-H) and layered calcium hydroxide (CH). In GRC, the reaction products are mainly amorphous aluminosilicate gels. Moreover many fly ash particles are well embedded in paste and fill up the pores. Comparatively, the paste of GRC is denser and homogeneous, which led to higher strength. 3. With more RAs in the concrete, the microstructure is looser with more pores and cracks. The cracks always appear on the interface between the aggregate and the paste, and then develop into the paste. The absence of CH and the effect of fly ash as the filler of the pores improve the mechanical properties of GRC. The compositions of the paste near the ITZ are identified by EDX analysis indicating more silicon and aluminum, but less calcium existing in GRC paste, which proves the fact that CH is absent and aluminosilicate gel is produced. Acknowledgments Authors gratefully acknowledge the support from The National Natural Science Foundation of China (No. 10925211), Chinese Scholarship Council and Monash-Sichuan University Strategic Fund for sponsorship to provide the study resources for this research work. The efforts and assistance with the laboratory work provided by Long Kim Goh, Jeff Doddrell, Kevin Nievaart, Mark Taylor and Lei Zhou are also gratefully acknowledged. References [1] C. Meyer, The greening of the concrete industry, Cem. Concr. Res. 31 (2009) 601–605.

29

[2] J. Davidovits, Geopolymers – inorganic polymeric new materials, J. Therm. Anal. 37 (1991) 1633–1656. [3] J.L. Provis, P. Duxson, J.S.J. Van Deventer, Geopolymer technology and the search for a low-CO2 alternative to concrete, in: 2007 AIChE Annual Meeting, November 4–9, 2007, American Institute of Chemical Engineers, Salt Lake City, UT, United states, 2007. [4] P. Duxson, J.L. Provis, G.C. Lukey, J.S.J. Van Deventer, The role of inorganic polymer technology in the development of ‘green concrete’, Cem. Concr. Res. 37 (2007) 1590–1597. [5] P. Duxson, A. Fernandez-Jimenez, J. Provis, G. Lukey, A. Palomo, J. van Deventer, Geopolymer technology: the current state of the art, J. Mater. Sci. (2007) 2917–2933. [6] T. Bakharev, Geopolymeric materials prepared using Class F fly ash and elevated temperature curing, Cem. Concr. Res. 35 (2005) 1224–1232. [7] B.V. Rangan, D. Hardjito, S.E. Wallah, D.M.J. Sumajouw, Properties and applications of fly ash-based concrete, in: Materials and Testing Conference 2005, October 30–November 2, 2005, Institute of Metals and Materials Australasia, Fremantle, WA, Australia, 2006, pp. 170–175. [8] M. Sofi, J.S.J. van Deventer, P.A. Mendis, G.C. Lukey, Engineering properties of inorganic polymer concretes (IPCs), Cem. Concr. Res. 37 (2007) 251–257. [9] T.C. Hansen, Recycled aggregate concrete second state-of-art report developments 1945–1985, Mater. Struct. 19 (1986) 201–246. [10] S.W. Tabsh, A.S. Abdelfatah, Influence of recycled concrete aggregate on strength properties of concrete, Constr. Build. Mater. 23 (2009) 1163–1167. [11] X.-S. Shi, Q.-Y. Wang, C.-C. Qiu, X.-L. Zhao, Experimental study on the properties of recycled aggregate concrete with different replacement ratios from earthquake-stricken area, Sichuan Daxue Xuebao (Gongcheng Kexue Ban)/J. Sichuan Univ. (Eng. Sci. Ed.) 42 (2010) 170–176. [12] X.-S. Shi, Q.-Y. Wang, Research review and application outlook of RAC from post-earthquake reconstruction, Sichuan Daxue Xuebao (Gongcheng Kexue Ban)/J. Sichuan Univ. (Eng. Sci. Ed.) 41 (2009) 301–310. [13] S.C. Kou, C.S. Poon, D.C. Chen, Influence of fly ash as cement replacement on the properties of recycled aggregate concrete, J. Mater. Civ. Eng. 19 (2007) 709–715. [14] S.C. Kou, C.S. Poon, D.C. Chan, Influence of fly ash as a cement addition on the hardened properties of recycled aggregate concrete, Mater. Struct. 41 (2008) 1191–1201. [15] C.S. Poon, Z.H. Shui, L. Lam, Effect of microstructure of ITZ on compressive strength of concrete prepared with recycled aggregates, Constr. Build. Mater. (2004) 461–468. [16] W.Y.T. Vivian, Carbonation around near aggregate regions of old hardened concrete cement paste, Cem. Concr. Res. 35 (2005) 1180–1186. [17] W.Y.T. Vivian, Micro structural analysis of recycled aggregate concrete produced from two-stage mixing approach, Cem. Concr. Res. 35 (2005) 1195–1203. [18] A. Palomo, M.W. Grutzeck, M.T. Blanco, Alkali-activated fly ashes: a cement for the future, Cem. Concr. Res. 29 (1999) 1323–1329. [19] M. Steveson, K. Sagoe-Crentsil, Relationships between composition, structure and strength of inorganic polymers – part 2 – fly ash-derived inorganic polymers, J. Mater. Sci. 40 (2005) 4247–4259. [20] S. Diamond, N. Thaulow, The patch microstructure in concrete: evidence that it exists and is not a backscatter SEM artifact, Cem. Concr. Comp. 28 (2006) 606–612. [21] E. Hewayde, M. Nehdi, E. Allouche, G. Nakhla, Effect of geopolymer cement on microstructure, compressive strength and sulphuric acid resistance of concrete, Mag. Concr. Res. 58 (2006) 321–331. [22] M. Komljenovic, Z. Bascarevic, V. Bradic, Mechanical and microstructural properties of alkali-activated fly ash geopolymers, J. Hazard. Mater. 181 (2010) 35–42. [23] J. Davidovits, Chemistry of geopolymeric systems terminology, in: Geopolymer’99 International Conference, France, 1999.