Investigation on the microstructure and mechanism of geopolymer with different proportion of quartz and K-feldspar

Construction and Building Materials 147 (2017) 543–549

Contents lists available at ScienceDirect

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

Investigation on the microstructure and mechanism of geopolymer with different proportion of quartz and K-feldspar Linan Tian, Wuwei Feng ⇑, Hongwen Ma ⇑, Shaogang Zhang, Hao Shi Beijing Key Laboratory of Materials Utilization of Nonmetallic Minerals and Solid Wastes, National Laboratory of Mineral Materials, China University of Geosciences, Beijing 100083, China School of Materials Science and Technology, China University of Geosciences, Beijing 100083, China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Effect of the aggregate contents of

quartz and K-feldspar on the compressive and bending strength is studied.
Determine the optimum modulus of alkaline activator at different proportions of quartz and K-feldspar.
Microstructure and microanalysis of the region between the aggregate and matrix are studied.

a r t i c l e

i n f o

Article history: Received 22 October 2016 Received in revised form 30 March 2017 Accepted 12 April 2017

Keywords: Geopolymeric brick Alkaline activator Aggregate Strength Microstructure

a b s t r a c t Geopolymer appears to be an alternative to ordinary Portland cement as a new sustainable material. The effect of the aggregate contents of quartz and K-feldspar on the microstructure and mechanical properties of Geopolymer samples was investigated in the present study. It was found that the aggregate contents influence the compressive strength and bending strength of geopolymeric bricks. Also, The K-feldspar/quartz mass ratio (F/Q) has a linear relation with the modulus of alkaline activator at the peak value of bending strength. This is due to the fact that the quartz reacts with the sodium hydroxide and forms [H2SiO4]2 which is the main component of sodium silicate, and that the K-feldspar has the ability to provide additional alkaline. The results of electron-microprobe analysis indicated that the matrix phase is composed of more amount of [ASiAOAAlAOASiA]n polymerization monomer with increasing alkaline activator modulus ratio when K+ is not involved in the reaction. However, the polymerization monomer of [ASiAOAAlA]n was only favored despite of variation of alkaline activator modulus when the K+ exists. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction

⇑ Corresponding authors at: Beijing Key Laboratory of Materials Utilization of Nonmetallic Minerals and Solid Wastes, National Laboratory of Mineral Materials, China University of Geosciences, Beijing 100083, China. E-mail addresses: [email protected] (W. Feng), [email protected] (H. Ma). http://dx.doi.org/10.1016/j.conbuildmat.2017.04.102 0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.

Ordinary Portland cement (OPC) is widely used in construction materials in the world, which, however, has high emissions of carbon dioxide (CO2) ranging from 0.66 to 0.82 kg per kilogram of OPC manufactured [1–4]. The main causes of CO2 emissions can be attributed to: (i) the calcination of the key ingredients, limestone;

544

L. Tian et al. / Construction and Building Materials 147 (2017) 543–549

and (ii) the high energy consumption by heating raw materials in a rotating kiln at above 1400 °C [2]. Besides, significant expansion and cracking in concrete structures caused by alkali aggregate reaction (AAR) is one of the main durability problems when OPC is used in concrete [5]. Disposal of piles of tailings has been a worldwide problem. In recent years, with the further continuing exploitation of mineral resources, especially for the mining of low-grade mineral resources, the amount of tailings rose sharply, which became a big threat on the ecological environment. Using feldspar and quartz rich tailings as aggregate to produce building materials can reduce the production cost of building materials and meanwhile also consume piles of tailings, which realize the comprehensive utilization of resources. The contents of feldspar and quartz in tailings vary in a large range, and the effect of different contents of aggregate of feldspar and quartz on the properties of building material is still unknown. As a new sustainable material, geopolymer can provide an alternative to OPC, which is formed by alkali-activated aluminosilicate materials like metakaolin, coal fly ash, and metallurgical slag after curing at ambient temperature or temperature between 60 and 120 °C [6]. Due to the avoidance of calcining limestone and burning fossil fuels, using geopolymer instead of cements reduce the CO2 emissions greatly [1,2]. Although the effects of several parameters, such as aluminosilicate raw material [8–12], curing condition [13– 16], ratio of reactant[17–19], reaction kinetics and mechanisms [20–24], and the composition of polymer-geopolymer [25–28], on the properties of geopolymer have been studied over the last several years, there’s almost no researches reported about the role of aggregate species and contents in geopolymer. It’s still not clear how aggregate behave in the concentrate alkaline paste and influence the macroscopical properties of the geopolymer. In this study the effect of aggregate content of quartz and K-feldspar on the mechanical and microstructural properties of the alkali-activated paste is investigated. In addition, this study is of great significance for the utilization of feldspar quartz rich tailings as the aggregate of geopolymer.

Fig. 2. X-ray diffraction pattern of the metakaolin.

quartz and K-feldspar sands were 15.066 m2/g and 19.585 m2/g, respectively. The aggregate used in this study are comprised of quartz and K-feldspar sands, with the mass fraction content of quartz ranging from 20% to 100%. Metakaolin, an aluminosilicate raw material, was obtained by the calcination of kaolinite at 750 °C for 4 hours in a muffle. There is no distinct diffraction peak in the X-ray diffraction pattern of metakaolin except the hump between 20 and 37 degree (Fig. 2), which indicates an amorphous state of metakaolin. In order to demonstrate an apparent effect of aggregate on mechanical properties, the solid mixture was obtained by the mixing of aggregates and metakaolin with a mass ratio of 3 according to our previous experience. The chemical compositions of metakaolin, quartz and K-feldspar determined by X-ray Fluorescence (XRF) are listed in Table 1. Alkaline activators used in geopolymers were consisted of sodium hydroxide (95% of purity, Beijing Chemical Reagent Ltd. China) and water glass (Modulus = 2.46, w(Na2O) = 15.17 and w(SiO2) = 36.22 by mass, Baume degree 50°, Beijing Chemical Reagent Ltd. China). Distilled water was used to dissolve the sodium hydroxide solid to avoid the unknown contaminants in the mixing water [29]. The mixed sodium hydroxide solution was prepared 24 h before addition to solid aluminosilicates. Alkaline activators were added to solid mixture with a L/S (liquid to solid mass ratio) of 0.36. Previous study showed that the modulus ratio of alkaline activators solution between 1.2 and 1.4 lead to high bending and compressive strength of geopolymer [7]. Therefore, the modulus of alkaline activator solution in this study was set to 1.13–1.42.

2. Experimental 2.2. Preparation 2.1. Materials The aggregate used in geopolymer was usually natural rocks like granite and quartzite which mainly consist of silicate and/or aluminosilicate minerals with stable chemical properties, corrosion resistance and high hardness [29]. In order to make certain the aggregate has enough large superficial area to react with the matrix completely, quartz sand and K-feldspar sand were selected and ground finely to a particle size less than 100 lm (Fig. 1). The surface area of the fines of

Metakaolin and aggregate were mixed in a cement mixer for 3 min, followed by the addition of mixed alkaline activators. The ratio of metakaolin (g)/solid powder (g) and alkaline activator (g)/solid powder (g) were 0.25 and 0.36, respectively. The resulting paste was transferred to a stainless steel mould, and then was pressed under an isostatic pressure of 10 MPa for 5 min. After being removed from the mould, the sample was dried at 80 °C in an oven for setting for 24 h, and then maintained in atmosphere for 27 days. 2.3. Methods The compression test was performed on the cylinder samples with diameter of 40 mm and height of 100 mm, and the bending strength was determined on cuboid samples with size of 40 mm  40 mm  160 mm. The failure load of bending test was determined by the load-distortion curve obtained from three-point bending test using an electronic universal testing machine. The standars for measuring the compressive strength and bending strength is based on the Chinese standar GB28635-2012/ Concrete paving bricks. The microstructure of sample was studied by the scanning electron microscope (SEM, Hitachi, S2300), and polarizing microscope analysis (POM, caikon, XP400), respectively. Chemical composition was measured by electron probe microanalysis (EPMA, JXR-8230) with beam spot size of 5 lm at 50 nA and 20 kV.

3. Results 3.1. Compressive strength

Fig. 1. Particle size distribution of the K-feldspar and quartz sands.

Compressive strength of the as-prepared geopolymeric bricks is shown in Fig. 3. Note that samples using alkaline activators solution with a modulus ratio of 1.18 and aggregate with K-feldspar/

545

L. Tian et al. / Construction and Building Materials 147 (2017) 543–549 Table 1 The chemical compositions of metakaolin, quartz and K-feldspar (wB%). Sample

SiO2

TiO2

Al2O3

Fe2O3

MnO

MgO

CaO

Na2O

K2O

P2O5

LOI

Total

MKL-2 QC-1 KF-12

49.75 96.33 66.52

1.58 0.03 –

43.23 0.40 18.20

1.35 1.13 0.38

0.02 0.01 0.01

0.03 0.21 –

0.24 0.12 0.48

0.10 – 2.48

0.05 0.08 10.52

0.05 0.05 0.03

2.99 0.05 0.62

99.39 99.66 99.29

Fig. 3. Compressive strength vs alkaline activator modulus ratio curves under different K-feldspar/quartz ratio for geopolymeric bricks.

quartz ratio (F/Q) of 0.2 were named GM 1.18-F/Q 0.2. Similarly, samples with F/Q of 0.4 and activator modulus ratio of 1.18 as well as F/Q of 0.2 and activator modulus ratio of 1.37 were named GM 1.18-F/Q 0.4 and GM 1.37-F/Q 0.2, respectively. The compressive strength of the geopolymeric bricks with different F/Q was very similar when activator modulus varies from 1.13 to 1.42 (Fig. 3), which increased first and then decreased with an increase of activator modulus ratio, and reached the maximum value at the activator modulus of 1.33. Previous studies about the geopolymeric paste showed that SiO2/Na2O molar ratio is a significant parameter which affects the compressive strength [30]. The increase of compressive strength is mainly due to the appropriate SiO2/Na2O molar ratio and an increase of polymerization degree. The increase of NaOH concentration provides better dissolution of silicate and aluminate species and lead to the increase of inter-molecular bond strength. The reduction in strength is probably due to the excess of unreacted NaOH. The shapes of the compressive strength vs alkaline activator modulus curve with different F/Q were almost the same when the SiO2/Na2O molar ratio ranged from 1.24 to 1.42, which did not hold for the activator modulus ratio below 1.24 most probably because the excess sodium hydroxide absorbed carbon dioxide (CO2) from the air and transformed into sodium carbonate. This part of free carbonate appeared at the surface of geopolymeric bricks by the capillarity of transpiration, which is known as ‘‘efflorescence”. The compressive strength decreased with the increase of F/Q when the SiO2/Na2O molar ratio was 1.13 and 1.18. The strength behavior observed here can be explained by the role of the aggregate in geopolymer, which was usually ignored in OPC [31]. In geopolymer the dissolution of aggregate in the presence of alkalis hydroxide will enhance bond intensity at the interface of paste and aggregate. 3.2. Bending strength Bending strength of geopolymeric bricks is shown in Fig. 4. It can be seen that geopolymeric bricks possess very high bending

Fig. 4. Bending strength vs alkaline activator modulus ratio curves under different K-feldspar/quartz ratio in geopolymeric bricks.

strength compared with OPC. The bending strength also increased first and then decreased with the increase of activator modulus ratio, but with maximum corresponding to different alkaline activator modulus. The bending strength was roughly the same when the activator modulus is 1.37 and 1.42. The geopolymeric bricks with aggregate being composed only of quartz (F/Q = 0) possess highest bending strength when alkaline activator modulus ratio was below 1.37. It was also found that the bending strength decreases with the increase of F/Q. With the increase of F/Q, the alkaline activator modulus corresponding to the maximum of bending strength shifted from 1.24 to 1.37. 3.3. SEM analysis Fig. 5 show SEM micrographs of the sample GM 1.33 - F/Q 0.2. As it can be seen in the figure, the aggregate are of different size, which are wrapped in alumino-silicate gels, sometimes with the unreacted metakaolin. Some extent of corrosion can often be observed on the surface of aggregate, as shown in Fig. 5 (a) and (b). Also, the porous cracks left by the evaporation of water in the matrix can also be seen. The fracture surfaces of geopolymeric brick are shown in Fig. 5(c) and (d). It can be seen that the aggregate are tightly bound to the matrix, and the morphology of the region between the aggregate and matrix is obviously different from that of matrix. The fresh fracture surface indicates that the chemical bonds generated by geopolymerisation between aggregate and matrix are strong and increase the mechanical properties of the sample, which is different from OPC. 3.4. POM and EPMA analysis The polarizing microscope (POM) images of geopolymeric bricks GM 1.33-F/Q 0.2 and GM 1.24-F/Q 0.2 were shown in Fig. 6, in which different mineral phases could be identified by comprehensive analysis. It can be seen that the boundary between quartz and matrix was not clearly, and some quartz particle has

546

L. Tian et al. / Construction and Building Materials 147 (2017) 543–549

Fig. 5. SEM micrograph of geopolymeric brick GM 1.33 - F/Q 0.2. (Note: a and b showed the corrosion surface of aggregate; c and d showed the fresh fracture surface of aggregate.)

(a) GM 1.33-F/Q 0.2

(b) GM 1.24- F/Q 0.2

Fig. 6. Photographs of GM 1.33-F/Q 0.2 and GM 1.24-F/Q 0.2 bricks observed in perpendicular polarized microscope. (Note: Qtz represents quartz; Kfs represents K-feldspar; the elemental composition of the sites marked by A, B, C, D, E and F were analyzed by electron probe microanalysis.)

547

L. Tian et al. / Construction and Building Materials 147 (2017) 543–549 Table 2 Composition of aggregate and matrix (wB%) determined by electron probe microanalysis. Site

SiO2

TiO2

Al2O3

FeO

MnO

MgO

CaO

Na2O

K2O

Total

A B C D E F

46.80 86.51 41.19 39.11 90.85 40.89

0.10 – 0.14 0.55 – 0.21

20.38 4.48 26.14 21.29 0.85 26.69

0.35 – 0.48 0.34 – 0.13

– – 0.10 – – –

0.05 – 0.66 0.12 – 0.31

0.51 – 0.43 0.72 – 0.22

18.69 3.67 11.38 24.47 0.80 11.98

– – 7.11 – – 6.87

86.88 96.44 87.62 86.59 92.51 87.29

Fig. 7. Relations between bending strength and compressive strength according to modulus of alkaline activator in geopolymeric bricks made with different aggregate content.

already corroded into small pieces. The extent of edge corrosion of quartz was greater than that of K-feldspar, which suggests that the quartz formed more chemical bonds with the matrix than Kfeldspar in alkaline solution. This may be the explanation why the geopolymeric bricks with lower F/Q achieve better mechanics performance. In order to analyze the surface reaction between matrix and aggregate, the sites marked by A, B, C, D, E and F, as shown in Fig. 6, were measured by electron probe microanalysis (EPMA) and the results are listed in Table 2. Site A and D represent the matrix which is far from aggregate, site B and E represent the area which is between the matrix and quartz, and site C and F represent the area which is between the matrix and K-feldspar. It can be seen from Table 2 that site A and D are composed of sodium aluminosilicate; site B and E consist mainly of quartz pieces, together with few sodium aluminosilicate; and site C and F consist of sodiumpotassium aluminosilicate. These observations are strong evidences which confirms that the potassium in K-feldspar can be dissolved in alkaline solution and enter into the aluminosilicate matrix.

4. Discussion Relations between bending strength and compressive strength of geopolymeric bricks are shown in Fig. 7. The slope of two linear fitting curves is approximately equal to the ratio of bendingcompressive strength. According to literatures, the ratio of bending-compressive strength of Geopolymeric bricks are mainly between these two linear curves, namely, between 0.122 and 0.169. Besides, some of the points lying outside these two linear curves are mainly from samples with the F/Q ratio of 0, 0.2 or 0.4. It is clear that samples with low F/Q ratio mainly lie outside

Fig. 8. Relation between F/Q ratio and modulus of alkaline activator at the peak value of bending strength.

these two linear curves. Therefore, the samples with higher F/Q ratio demonstrate more stable ratio of bending-compressive strength. Besides, the F/Q ratio has a linear relationship with the modulus of alkaline activator which corresponds to the maximum of bending strength, as shown in Fig. 8. Take the aggregate of quartz for example, quartz can be dissolved in the alkaline solution to form [SiO2(OH)2]2, the chemical reaction is given as

SiO2 þ 2OH ! ½SiO2 ðOHÞ2 

2

The modulus of sodium silicate is described as

M ¼ nðSiO2 Þ=nðNa2 OÞ So the modulus of alkaline activator can be described as

M¼

nwg ðSiO2 Þ nwg ðNa2 OÞ þ 2nðNaOHÞ

Considering the [SiO2(OH)2]2 formed by quartz, so

M0 ¼

nwg ðSiO2 Þ þ nð½SiO2 ðOHÞ2  nwg ðNa2 OÞ þ 2nðNaOHÞ

2

When the alkaline solution reacted completely with metakaolin, the geopolymeric bricks achieve the best bending strength, so modulus of alkaline solution (M) is a fixed value, and then

M0 ¼ M þ

2

nð½ðSiO2 ÞððOHÞ2  nwg ðNa2 OÞ þ 2nðNaOHÞ

As is known, the formed [SiO2(OH)2]2 has a quantitative relation (A) with the quartz aggregate, 2

nð½SiO2 ðOHÞ2  Þ ¼ AnðQuartzÞ, and nwg(Na2O) + 2n(NaOH) is a A , so fixed value. Setting the value B ¼ nwg ðNa2 OÞþ2nðNaOHÞ

M0 ¼ M þ BnðQuartzÞ

548

L. Tian et al. / Construction and Building Materials 147 (2017) 543–549

Table 3 Component molar ratio of matrix. Sites

n(Si)/n(Al)

n(K + Na)/n(Al)

A B C D E F

1.94 19.98 1.33 1.56 95.76 1.29

1.50 2.01 1.01 1.89 1.74 1.02

Therefore, the F/Q ratio has a linear relationship with modulus of alkaline activator in theory. And the reaction of K-feldspar with alkaline solution is similar to quartz. The structure of K-feldspar can be destroyed by the concentrated alkaline solution and the K+ spread into the aluminosilicate gels to balance the charge of [Al(OH)4]. Therefore, the K-feldspar have the ability to provide additional alkaline, and thus can decrease the consumption of sodium hydroxide in geopolymerization. The three dimensional aluminosilicate framework could be classified into four monomers types according to the different n (Si)/n(Al) ratios in matrix [6]: (1) poly (sialate) (n(Si)/n(Al) = 1, PS); (2) poly (sialate-siloxo) (n(Si)/n(Al) = 2, PSS); (3) poly (sialate-disiloxo) (n(Si)/n(Al) = 3, PSDS); and (4) poly (sialatemultisiloxo) (n(Si)/n(Al) > 3, PSMS). The component molar ratio of matrix was shown in Table 3. The n(Si)/n(Al) ratio for site A with activator modulus of 1.33 is approximately 2, which means the polymerization monomer was mainly PSS type ([ASiAOAAlAOASiA]n). The site D with activator modulus of 1.24 has a n(Si)/n(Al) of 1.56, which suggest that it consists of both [ASiAOAAlAOASiA]n and [ASiAOAAlA]n polymerization monomers. Therefore, the [ASiAOAAlAOASiA]n polymerization monomer was easier formed under higher alkaline activator modulus ratio. The n(Si)/n(Al) value for sites B and E are very larger probably due to probing mainly on quartz. The n(Si)/n(Al) for sites C and F have the similar values, 1.33 and 1.29, which is greatly smaller than the n(Si)/n(Al) of 3 for K-feldspar. It means sites C and F belong to K-containing matrix phase, rather than locating on the surface of K-feldspar. In addition, the similar n(Si)/n(Al) value which approximates to 1 for sites C and F indicated that both sites are mainly composed of PS type ([ASiAOAAlA]n) polymerization monomer. Therefore, different from observation in the area of matrix, variation of activator modulus does not exert substantial effect on the polymerization monomer type for the K-containing area between matrix and K-feldspar. In geopolymer, alkali metal ions are mainly distributed in interstitial sites of the three-dimensional framework of aluminum silicate to balance the negative charge of alumina tetrahedron. Due to the incompleteness of polymerization reaction, part of alkali metal ions usually still exists in the skeleton of aluminum silicate. Therefore, n(K + Na)/n(Al) is usually between 1 and 1.6 in a normal geopolymer materials [7]. Sites A and D represent matrix phase which does not contain K ion. The A site with activator modulus of 1.33 has the n(Na)/n(Al) of 1.50. As for site A, the alkali metal ions are not excess, which exists in forms of interstitial ion and SiAOANa, AlAOANa bonds. Therefore, GM 1.33-F/Q 0.2 sample demonstrated good mechanical property. With decreasing activator modulus to 1.24 (Site D), the n(Na)/n(Al) increases to 1.94, indicating existence of excess alkali metal ions which will precipitate from the sample surface in a free form. This explained the lower mechanical property of GM 1.24-F/Q 0.2 sample. Sites B and E does not contain K ion, in which the existence of excess Na+ is due to the formation of poly-sodium silicate, the product of reaction between quartz and sodium hydroxide solution. Sites C and F are near Kfeldspar and therefore contain K+. The n(K + Na)/n(Al) is 1.01 and 1.02 for both sites, respectively, which is lower than the counter-

part of matrix. This is due to the fact that K+ is more capable of entering into the matrix to balance the negative charge of polysialate than Na+, and thus increases the rate of ionization and dissolution. The matrix of geopolymer is mainly composed of amorphous aluminosilicate gel, which has a short-range order threedimensional network, such as nano-zeolite [32]. The matrix will partially evolves into a small mounts of zeolites when curing time was prolonged [33]. These zeolites may be crystal-structurally similar with analcite and hydroxysodalite according to the result of EPMA. Base on our experimental results and thermodynamic calculation, the main chemical reactions that may occurr in this research are given as below.

SiO2 þ 2OH ! ½H2 SiO4 

2

ð1Þ 

Al2 Si2 O7 þ 6OH þ 3H2 O ! 2½AlðOHÞ4  þ 2½H2 SiO4  

2

KAlSi3 O8 þ 6OH þ 2H2 O ! Kþ þ ½AlðOHÞ4  þ 3½H2 SiO4  

ð2Þ 2

ð3Þ

Naþ þ ½AlðOHÞ4  þ 2½H2 SiO4 

2

! NaAlSi2 O6  H2 O þ 4OH þ 2H2 O 

ð8  zÞNaþ þ zKþ þ 6½AlðOHÞ4  þ 6½H2 SiO4 

ð4Þ 2

! ðNa8z Kz ÞAl6 Si6 O24 ðOHÞ2  12H2 O þ 10OH

ð5Þ

5. Conclusion This work presents studies on effect of different F/Q ratio and activator modulus ratio on the microstructure and mechanical properties of geopolymeric bricks. Main conclusions are summarized as follows: 1. The aggregate contents of quartz and/or K-feldspar do not influence the compressive strength of geopolymeric bricks notably. the geopolymeric bricks made with lower F/Q ratio achieve better bending strength. It means that the quartz might forms more chemical bonds with the matrix than K-feldspar in alkaline solution. 2. The relation between F/Q ratio and modulus of alkaline activator at the peak value of bending strength is linear. Because the potassium in K-feldspar can be dissolved in alkaline solution and enter into the aluminosilicate matrix, and thus can decrease the consumption of sodium hydroxide needed for geopolymerization. Besides, the quartz reacts with the sodium hydroxide and forms [H2SiO4]2 which is the main component of sodium silicate. 3. The matrix phase formed more [ASiAOAAlAOASiA]n polymerization monomer with higher alkaline activator modulus ratio when K+ is not involved in the reaction. However, only the polymerization monomer of [ASiAOAAlA]n was favored against change of alkaline activator modulus when the K+ exists.

Acknowledgement This work was financially supported by the funds from the China Geological Survey Project (Grant no. 12120113087700). References [1] L.K. Turner, F.G. Collins, Carbon dioxide equivalent (CO2-e) emissions: a comparison between geopolymer and OPC cement concrete, Constr. Build. Mater. 43 (2013) 125–130.

L. Tian et al. / Construction and Building Materials 147 (2017) 543–549 [2] E. Gartner, Industrially interesting approaches to ‘‘low-CO2” cements, Cem. Concr. Res. 34 (2004) 1489–1498. [3] J.X. Peng, L. Huang, Y.B. Zhao, et al., Modeling of carbon dioxide measurement on cement plants, Adv. Mater. Res. Trans. Tech. Publ. 610 (2012) 2120–2128. [4] C. Li, X.Z. Gong, S.P. Cui, et al., CO2 emissions due to cement manufacture, Mater. Sci. Forum 685 (2011) 181–187. [5] C. Tennakoon, A. Shayan, J.G. Sanjayan, Influence of binder on alkali reactivity of aggregates in geopolymer concrete, in: Concrete Institute of Australia Conference, 27th, 2015, Melbourne, Victoria, Australia, 2015. [6] J. Davidovits, 30 years of successes and failures in geopolymer applications. Market trends and potential breakthroughs. in: Keynote Conference on Geopolymer Conference, 2002. [7] J.L. Provis, S.A. Bernal, Geopolymers and related alkali-activated materials, Annu. Rev. Mater. Res. 44 (2014) 299–327. [8] A. Palomo, M.W. Grutzeck, M.T. Blanco, Alkali-activated fly ashes: a cement for the future, Cem. Concr. Res. 29 (1999) 1323–1329. [9] H. Xu, J.S.J. Van Deventer, The geopolymerisation of alumino-silicate minerals, Int. J. Miner. Process. 59 (2000) 247–266. [10] H. Xu, J.S.J. Van Deventer, Geopolymerisation of multiple minerals, Miner. Eng. 15 (2002) 1131–1139. [11] A. Buchwald, M. Hohmann, K. Posern, et al., The suitability of thermally activated illite/smectite clay as raw material for geopolymer binders, Appl. Clay Sci. 46 (2009) 300–304. [12] N. Bouzón, J. Payá, M.V. Borrachero, et al., Refluxed rice husk ash/NaOH suspension for preparing alkali activated binders, Mater. Lett. 115 (2014) 72– 74. [13] T. Bakharev, Geopolymeric materials prepared using Class F fly ash and elevated temperature curing, Cem. Concr. Res. 35 (2005) 1224–1232. [14] G. Kovalchuk, A. Fernández-Jiménez, A. Palomo, Alkali-activated fly ash: effect of thermal curing conditions on mechanical and microstructural development–Part II, Fuel 86 (2007) 315–322. [15] Ö. Arıöz, K. Kilinç, M. Tuncan, et al., Physical, mechanical and micro-structural properties of F type fly-ash based geopolymeric bricks produced by pressure forming process, Adv. Sci. Technol. 69 (2011) 69–74. [16] P. Chindaprasirt, U. Rattanasak, S. Taebuanhuad, Role of microwave radiation in curing the fly ash geopolymer, Adv. Powder Technol. 24 (2013) 703–707. [17] A. Fernández-Jiménez, A. Palomo, Composition and microstructure of alkali activated fly ash binder: effect of the activator, Cem. Concr. Res. 35 (2005) 1984–1992.

549

[18] D. Sabitha, J.K. Dattatreya, N. Sakthivel, et al., workability and strength of potassium versus sodium-activated high volume fly ash-based geopolymers, Curr. Sci. (Bangalore) 103 (2012) 1320–1327. [19] P. Duxson, J.L. Provis, G.C. Lukey, et al., Understanding the relationship between geopolymer composition, microstructure and mechanical properties, Colloids Surf. A 269 (2005) 47–58. [20] W.I. Mastura, H. Kamarudin, I.K. Nizar, et al., Mechanical performances of fly ash geopolymer bricks, Adv. Sci. Lett. 19 (2013) 186–189. [21] P. De Silva, K. Sagoe-Crenstil, V. Sirivivatnanon, Kinetics of geopolymerization: role of Al2O3 and SiO2, Cem. Concr. Res. 37 (2007) 512–518. [22] M.R. North, T.W. Swaddle, Kinetics of silicate exchange in alkaline aluminosilicate solutions, Inorg. Chem. 39 (2000) 2661–2665. [23] M. Lizcano, H.S. Kim, S. Basu, et al., Mechanical properties of sodium and potassium activated metakaolin-based geopolymers, J. Mater. Sci. 47 (2012) 2607–2616. [24] H. Xu, J.S.J. van Deventer, The effect of alkali metals on the formation of geopolymeric gels from alkali-feldspars, Colloids Surf. A 216 (2003) 27–44. [25] J.A. Rossignolo, M.V.C. Agnesini, Durability of polymer-modified lightweight aggregate concrete, Cement Concr. Compos. 26 (2004) 375–380. [26] V.F.F. Barbosa, K.J.D. MacKenzie, C. Thaumaturgo, Synthesis and characterisation of materials based on inorganic polymers of alumina and silica: sodium polysialate polymers, Int. J. Inorg. Mater. 2 (2000) 309–317. [27] C. Ferone, G. Roviello, F. Colangelo, et al., Novel hybrid organic-geopolymer materials, Appl. Clay Sci. 73 (2013) 42–50. [28] T.B. Lewis, L.E. Nielsen, Dynamic mechanical properties of particulate-filled composites, J. Appl. Polym. Sci. 14 (1970) 1449–1471. [29] F. Pacheco-Torgal, J. Castro-Gomes, S. Jalali, Investigations about the effect of aggregates on strength and microstructure of geopolymeric mine waste mud binders, Cem. Concr. Res. 37 (2007) 933–941. [30] S. Salvador, O. Pons, A semi-mobile flash dryer/calciner unit to manufacture pozzolana from raw clay soils application to soil stabilisation, Constr. Build. Mater. 14 (2000) 109–117. [31] C. Shi, P. Xie, Interface between cement paste and quartz sand in alkaliactivated slag mortars, Cem. Concr. Res. 28 (1998) 887–896. [32] J.L. Provis, G.C. Lukey, J.S.J. van Deventer, Do geopolymers actually contain nanocrystalline zeolites? A reexamination of existing results, Chem. Mater. 17 (2005) 3075–3085. [33] M. Criado, A. Fernndez-Jimnez, A.G. De la Torre, et al., An XRD study of the effect of the SiO2/Na2O ratio on the alkali activation of y ash, Cem. Concr. Res. 37 (2007) 671–679.