Role of the morphology and the dehydroxylation of metakaolins on geopolymerization V. Medri a, , S. Fabbri a , J. Dedecek b , Z. Sobalik b , Z. Tvaruzkova b , A. Vaccari c a ISTEC-CNR, Via Granarolo 64, 48018 Faenza, Italy b J. Heyrovsky Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, Dolejskova 2155/3, 18 223, Prague 8, Czech Republic c Dipartimento di Chimica Industriale e dei Materiali, Alma Mater Studiorum, Università di Bologna, Viale Risorgimento 4, 40136, Bologna, Italy abstract article info Article history: Received 7 July 2010 Received in revised form 5 October 2010 Accepted 12 October 2010 Available online 20 October 2010 Keywords: Metakaolin Geopolymerization Microstructure NMR spectroscopy FTIR Two commercial metakaolins were tested during partial geopolymerization with potassium silicate in order to emphasize the different surface reactivities. Both manual and short-term mechanical stirring were used for slurry preparation, while radiation, infrared, and microwave heating were used for curing. The metakaolins had similar compositions and specic surface areas, but different morphologies and dehydroxylation degrees due to different calcination kiln technologies. The degree of geopolymerization was checked by SEM and N 2 adsorption (BET), FTIR and 27 Al MAS NMR spectroscopy. While the dehydroxylation degrees were different, the metakaolins had similar reactivity. The metakaolin powder with rounded agglomerates and lower water demand was more sensitive to the various geopolymerization conditions than the ne dispersed lamellar one, thus giving rise to very different micro- and macrostructures of the partially geopolymerized samples. IR heating seemed to increase the geopolymerization degree slightly, while MW heating induced the fast evaporation of the water, forming porous samples. © 2010 Elsevier B.V. All rights reserved. 1. Introduction Differences in the durability of ancient cements and modern concretes have been investigated by Glukhovsky et al. (1980). This work led to the synthesis of various aluminosilicate binders from clays, feldspars, volcanic ashes and slags. Later, Davidovits used the word geopolymersfor a similar class of inorganic polymeric materials (Davidovits, 1991; Davidovits, 2002), as an alternative to organic matrices for composites. Geopolymers can be generally dened as amorphous to semi- crystalline materials with a three-dimensional aluminosilicate net- work formed by condensation polymerization. Polymeric bonds of SiOAlO form under alkaline conditions in the presence of alu- minosilicate oxides. The nal microstructure of a fully reacted geopolymeric resin consists of nanoparticulates ranging from 5 to 15 nm, separated by nanopores of 3 to 10 nm (Kriven et al., 2003). At the atomic scale, the geopolymer amorphous network is formed by SiO 4 and AlO 4 tetrahedra connected by oxygen corners. Recent results indicate that these tetrahedra form rings of various sizes in the network and endow the geopolymer matrix with ion exchange properties (Dedecek et al., 2008). All monovalent cations balancing the negative charge of AlO 4 tetrahedra (Na + or K + ) can be replaced by small monovalent (Li + and NH 4 + ) or divalent (Co 2+ ) ions. On the other hand, large cations as Cs + can replace Na + or K + ions only partially. The ion exchange properties of geopolymers, together with the spectra of dehydrated Co(II) exchanged geopolymers in visible region, resulted in the suggestion that their network is formed pre- dominantly by six-member, eight-member, and larger rings (Dedecek et al., 2008). Therefore, from some standpoints, geopolymers may be regarded as X-ray amorphous analogues and precursors of tetrahedral aluminosilicate framework zeolites. To produce a geopolymer, the aluminosilicate raw materials are mixed with an alkaline aqueous solution, usually a sodium or potassium hydroxide and/or silicate aqueous solution (Kriven et al., 2003; Panagiotopoulou et al., 2007; Nair et al., 2007). The geopoly- merization mechanism is quite complex (Davidovits, 2008), but may be simplied into a few steps (Duxon et al., 2007a): dissolution (consuming water); speciation equilibrium; gelation; reorganization; polymerization, and hardening. The rst step is the dissolution of the solid aluminosilicates by alkaline hydrolysis, producing aluminate and silicate species. Small size cations such as Na + increase the dis- solution rate (Panagiotopoulou et al., 2007; Xu and Van Deventer, 2000), while alkaline metal cations with a larger atomic size (such as K + ) stimulate condensation and promote the geopolymerization to a stage of completion (Comrie and Kriven, 2003). Many AlSi containing source materials may be used to produce geopolymers (Xu and Van Deventer, 2000; Panagiotopoulou et al., 2007; Duxon et al., 2007b; Fletcher et al., 2005; Rahier et al., 1996). When aluminosilicates come into contact with alkaline solutions, Applied Clay Science 50 (2010) 538545 Corresponding author. Tel.: +39 0546699723; fax: +39 054646381. E-mail address: valentina.medri@istec.cnr.it (V. Medri). 0169-1317/$ see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.clay.2010.10.010 Contents lists available at ScienceDirect Applied Clay Science journal homepage: www.elsevier.com/locate/clay