Novel Geopolymers Incorporating Silicate Waste

Novel geopolymers incorporating silicate waste

Neuartige Geopolymere aus silikatischen Industrieabfällen

Der Technischen Fakultät

der Friedrich-Alexander-Universität Erlangen-Nürnberg

zur Erlangung des Grades

DOKTOR-INGENIEUR

vorgelegt von

Nicoletta Toniolo

aus Velo d´Astico, Italy

Als Dissertation genehmigt von der Technischen Fakultät der Friedrich-Alexander-Universität Erlangen-Nürnberg

Tag der mündlichen Prüfung: 27.07.2018

Vorsitzende des Promotionsorgans: Prof. Dr. –Ing. Reinhard Lerch

Gutachter: Prof. Dr.-Ing. habil. Aldo R. Boccaccini

Prof. Enrico Bernardo

Prof. Cristina Leonelli

“Al mio angelo custode”

Acknowledgements

Firstly, I would like to express my sincere gratitude to Prof. Aldo Boccaccini, supervisor and head of the Institute of Biomaterials, who gave me the opportunity to do my PhD research at his Institute. I want to thank him for all the support during the past 3 years, all the suggestions he gave me and for the opportunity to take part to numerous international conferences, which were important for my knowledge and my personal experience. This research project was carried out within the framework of the poject “CoACH”, funded under the Marie Sklodowska-Curie action of the EU funding program for research and innovation “Horizon 2020”. I am grateful for the financial support it gave me. Thanks to the coordinator of the project Monica Ferraris, Milena Salvo and Cristiana Contardi, thanks for being always available, thanks for organizing meetings, training, etc. I really enjoyed everything. A super thanks to my “CoACH” fellows, I really think that it would have been impossible for me to have better project mates: thanks to Fra, for all the time together, for organizing all the meetings together, for supporting me and listening to my presentation 100 times. Thanks to Cava per le mille chiacchierate! Thanks to Acacio, you have been more than a support for me, thanks for all the scientific support, thanks for the thousand laughs in Sasil, thanks for introducing me to Lindy hop and of course for all the beers together. Thanks to Ale e Rocio, thanks for all the gossip and the evening chats, thanks to Giamma per avermi sopportato un mese a Brno, Min, who taught me that I should be proud of my curly hair , Pablo, Cristian, Hassan, Bhuva, Silviu, Francesco, Katarzyna, Thanks a lot Coachitos, I will really miss our time together! I would like to thank all the PhD students of UNIPD that helped me a lot during my secondment in Padova, especially to Elena, Giamma, Alberto. A special thanks to Prof. Enrico Bernardo for encouraging me to do this PhD and for the support throughout the 3 years, grazie veramente. Thanks also to the members of the IPM group, where I spent one intensive month of mechanical characterizations, thanks to Ivo Dlouhy to supervise my experiments there.

III

I am thankful to Piero Ercole, who took care of my research since the first moment, thanks to Ludovico Ramon, Alessandra La Barbera and all the laboratory staff in Sasil which helped me during the two intensive months I spent in the company. Thanks to my students Philipp Müller and Sebastian Schulte, for helping me with the experiments and for teaching me to be patient. Thanks a lot to the people that helped me with the characterizations: Dr. Judith A. Roether, thanks for your availability and kindness, Harald Rost, Dr. Liliana Liverani, Dr. Yamini Avadhut, Dr. Giulia Zanmarchi and Prof. Jan Dusza. Thanks to all the members of the Biomaterials Institute. Thanks to Julia Will for all the discussions, the important advices and die Deutschunterricht.. Vielen Dank! Thanks to Heinz Maler, always available to solve my problems and to Gerhard Frank, who I disturbed every SEM section. To Frau Bärbel Wust, she was always available to help me with expeditions and travels, moreover thanks for trying to understand my German, Danke. Thanks to Jasmin and Alina for helping me with the lab organization and orders. Thanks to my office colleagues: Valentina, Marfa, Eyerusalem, Ranjot, Diana, thanks for the discussions about geopolymers, thanks for the time together in the lab and of course for the coffee, the cupcakes and all the breaks together, we will start the diet tomorrow ! Thanks to all my colleagues, Barbarita for pushing me all the time towards the goal, Agata, Samira, Lukas, Laura, Marcela, Vera, Florian, Lena, Atiq, Kai, Svenja, Agata, Katharina, Irem, thanks a lot for these three years together, I will never forget you! Thanks to my special friends in Erlangen: Crystel, thanks for trying to improve my English, Andy, Antonina, Sofia, Giacomo, Natascia, thanks for all the adventures, the game evenings, the beers and remember that we are the best to crash party. To all my friends in Italy, for the infinite vocal messages, and in particular for being always ready to party everytime I went back to Italy: Marta, Erica, Roby, Obe, Marta, Grazie ragazze!!! Thanks to my family, for supporting always my choices also when they were not easy to support, a special thanks to my nephews whose video fills my heart. Thanks to Alex, grazie per tutto il supporto, per la pazienza che hai avuto e che avrai, grazie perche´al tuo fianco mi sento una persona migliore!

Abstract

Increasing global warming has raised concerns on the extensive use of Portland cement due to the high amount of carbon dioxide gas associated with its production. For this reason the construction industry is increasingly turning to the use of environmentally friendly materials in order to meet the sustainability targets required for modern infrastructures. Geopolymers are a new class of construction materials developed primarily as an ecofriendly and sustainable alternative to conventional cement-based construction materials. Although research into geopolymers is significantly increasing, most studies have used raw virgin materials or chemical reagents such as metakaolin or sodium silicate, which then raises issues of sustainability and environmental responsibility. To compete with the inexpensive nature of cement and to ameliorate the buildup of waste materials, thus decreasing the strain on landfill space, this study has investigated the use of industrial by-products such as fly ash, waste glass and red mud, in place of virgin raw materials for production of a new family of geopolymers. Initially, fly ash-based geopolymers were developed and the incorporation of red mud or waste glass in the matrix was investigated. The results show that the mechanical properties decrease as the amount of waste glass in the geopolymer increases; on the other hand, the addition of red mud seems to improve the mechanical behavior. Leaching tests were carried out to confirm the capacity of the geopolymer materials to incorporate and stabilize pollutants inside the network. In order to improve the economic and environmental value of geopolymers, the possibility of partially replacing traditional alkaline activators such as sodium silicate, also called waterglass, with urban waste glass was extensively evaluated. Fly ash-based geopolymers and red-mud based geopolymers were developed using waste glass as the silica supplier and sodium hydroxide as the alkaline solution, avoiding altogether the use of waterglass. The incorporation of soda lime glass waste instead of waterglass represents an innovation in geopolymer research. The influence of the addition of waste glass as well as the molarity of the sodium hydroxide solution was investigated for both the microstructure and mechanical strength of the geopolymers. The results suggested that it is possible to incorporate

up to a 60 wt.% of waste glass while still achieving comparable strength to Portland cement. This strength is also attained within a relatively short setting time of circa 7 days. The formation of a geopolymer gel was confirmed by XRD and MAS-NMR analyses. In addition, the XRD technique confirmed the formation of crystalline zeolite phases. Red mud-based geopolymers achieved relatively high compressive strength, 45 MPa, which is extremely interesting given the significant amount of red mud incorporated (60wt.%). Moreover, these geopolymers were developed through an economic process that does not require high temperatures or foaming agents. The light-weight geopolymers, synthesized with the same formulation of the dense materials, have high porosity and acceptable mechanical properties (1 MPa), though, the chemical stability should be improved through future research. In order to expand the range of possible applications, wear resistance and a thermal shock test were carried out obtaining satisfactory results for applications where material degradation in wear and under a sudden temperature change is required. 3D printing and mechanical machining of the developed samples were additionally carried out. The geopolymers were successfully extruded with a 3D printing machine and drilled and worked on the lathe achieving complex shapes and components with accurate surface finish, interesting for expanding the applications of the geopolymers in other industrial sectors beyond constructions. Overall, the outcomes of this study highlight geopolymers as a promising candidate not only for building and construction applications but also for other engineering applications. In particular in areas where specific and accurate shapes are required. The present research has contributed new insights and generated knowledge in the geopolymers field. Geopolymer materials represent an innovative solution for the reuse of waste materials while achieving sound mechanical properties and high chemical stability and due to their processing opportunities geopolymers also promise a wide range of applications.

Zusammenfassung

In Anbetracht der globalen Klimaerwärmung gibt es bei der Verwendung von Portlandzementen zunehmend Bedenken, da bei deren Herstellung hohe Mengen an Kohlendioxid freigesetzt werden. Daher setzt die Bauindustrie verstärkt auf umweltverträgliche, nachhaltige Werkstoffe. Geopolymere sind eine neue Baustoffklasse, die als umweltfreundliche und nachhaltige Alternative zu konventionellen zementartigen Baustoffen entwickelt wurden. Obwohl verstärkt an Geopolymeren geforscht wird, werden in den meisten Studien Metakaolin oder Natriumsilikat als Rohstoffe bzw. Ausgangschemikalien verwendet, was dann wiederum Fragen der Nachhaltigkeit und der Umweltverantwortung aufwirft. In dieser Arbeit werden als Ausgangsstoffe zur Herstllung einer neuen Familie von Geopolymeren industrielle Nebenprodukte wie Flugasche, Altglas und Rotschlamm verwendet und dargestellt, wie diese Materialien in die Zementmatrix eingebaut werden. Damit wird gleichzeitig ein kostengünstiger Herstellungsweg für Zemente und eine Verwendungsmöglichkeit für industrielle Abfallstoffe aufgezeigt. Im ersten Schritte wurden in dieser Arbeit Geopolymere auf der Basis von Flugasche und Geopolymere unter Verwendung von Altglas oder Rotschlamm hergestellt. Es stellte sich heraus, dass die mechanischen Kennwerte der Geopolymere mit zunehmender Menge an Glasabfall im Geopolymer abnehmen, jedoch scheint der Zusatz von Rotschlamm das mechanische Verhalten andererseits zu verbessern. Schadstoffe können in das Netzwerk eingebaut und stabilisiert werden, was durch einen Auslaugungstest bestätigt wurde. Die Möglichkeit wurde umfassend untersucht, traditionelle alkalische Aktivatoren wie Natriumsilikat (Wasserglas) mit Kalk-Natron-Altglas zu ersetzen. Dies stellt eine Innovation in der Geopolymerforschung dar, was den wirtschaftlichen und ökologischen Wert von Geopolymeren enorm verbessert. Auf Flugasche und auf Rotschlamm basierende Geopolymere wurden unter Verwendung von Glasabfall als Kieselsäurelieferant und Natriumhydroxid als alkalische Lösung entwickelt, wobei die Verwendung von Wasserglas insgesamt vermieden wurde. Der Einfluss der Zugabe von Altglas sowie die Molarität der Natronlauge auf die Mikrostruktur als auch auf die Festigkeit der Geopolymere wurde untersucht. Dabei stellte sich heraus, dass es möglich ist, nach

VII

einer relativ kurzen Abbindezeit von ca. 7 Tagen bis zu 60 Gew .-% Altglas einzuarbeiten und eine mit Portlandzement vergleichbaren Festigkeit zu erreichen. Die Bildung eines Geopolymergels wurde durch XRD- und MAS-NMR-Analyse bestätigt. Die Bildung von kristallinen Zeolithphasen wurde mit XRD nachgewiesen. Rotschlamm-basierte Geopolymere erreichten eine relativ hohe Druckfestigkeit von 45 MPa, welche angesichts der signifikanten Menge an eingebautem Rotschlamm (60 Gew .-%) äußerst interessant ist. Darüber hinaus wurden diese Geopolymere durch einen wirtschaftlichen Prozess entwickelt, der keine hohen Temperaturen oder Schaummittel erfordert. Leichte Geopolymere, die vergleichbar mit dichten Materialien synthetisiert wurden, haben eine hohe Porosität und gute mechanische Eigenschaften, obwohl die chemische Stabilität noch durch zukünftige Forschungen verbessert werden sollte. Um den Anwendungsbereich zu erweitern wurden Verschleißfestigkeit- und Thermoschocktests durchgeführt, wodurch zufriedenstellende Ergebnisse für Anwendungen, bei denen ein Materialabbau durch Abrieb und plötzliche Temperaturänderungen erforderlich ist, erreicht werden konnten. Die hergestellten Geopolymere wurden außerdem erfolgreich mit einer 3D- Druckmaschine extrudiert und auf einer Drehmaschine bearbeitet, wodurch komplexe Formen und Komponenten mit einer präzisen Oberflächengüte erreicht wurden, welche für die Erweiterung der Anwendungen der Geopolymeren in anderen industriellen Bereichen außerhalb von Konstruktionen von Interesse sein können Insgesamt weisen die Ergebnisse dieser Studie darauf hin, dass Geopolymere vielversprechende Kandidaten nicht nur für Anwendungen im Bauwesen sind, sondern auch für andere technische Bereichen, insbesondere in Bereichen, in denen spezifische und präzise Formen erforderlich sind. Die vorliegende Forschung konnte daher neue Einsichten und Wissen auf dem Gebiet der Geopolymere generieren. Geopolymere stellen eine innovative Lösung für die Wiederverwendung von Abfallstoffen dar, weisen gute mechanische Eigenschaften und eine sichere chemische Stabilität auf und bieten aufgrund ihrer vielfältigen Verarbeitungsmöglichkeiten eine breite Palette von möglichen Anwendungen.

VIII

Contents

Acknowledgements ...... III Abstract …………………………………………………………………………V Zusammenfassung ...... VII Chapter 1 Introduction ...... 1 1.1 Motivation: Environmental problems related to cement production ...... 1 1.2 Aims and Objectives ...... 2 Chapter 2 State of the art ...... 5 2.1 Introduction ...... 5 2.2 Raw materials used in geopolymerization ...... 6 2.2.1 Fly ash ...... 7 2.2.2 Red mud ...... 9 2.2.3 Waste glass ...... 11 2.2.4 Alkaline activators ...... 14 2.3 Geopolymerization: Reaction mechanism ...... 18 2.4 Geopolymer foams ...... 23 2.5 Properties of geopolymeric materials ...... 25 Chapter 3 Synthesis, characterization and mechanical properties of fly ash- based geopolymers incorporating red mud or waste glass: a comparative study ………………………………………………………………………...31 3.1 Introduction ...... 31 3.2 Materials and Methods ...... 33 3.2.1 Characterization of the raw materials ...... 33 3.2.2 Geopolymers production ...... 37 3.2.3 Test and analysis methods ...... 39 3.3 Results and discussion...... 43 3.3.1 Micromorphology of raw materials ...... 43 3.3.2 Compressive strength ...... 43 3.3.3 Flexural strength ...... 46 3.3.4 Fracture toughness ...... 47 3.3.5 Micro hardness ...... 50 3.3.6 Confocal microscopy ...... 51 3.3.7 Scanning electron microscopy (SEM) ...... 53

3.3.8 IR Spectroscopy ...... 55 3.3.9 Leaching test ...... 56 3.4 Conclusions ...... 57 Chapter 4 Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass ...... 59 4.1 Introduction ...... 59 4.2 Materials and methods ...... 60 4.2.1 Geopolymers preparation ...... 61 4.2.2 Characterization technique ...... 64 4.3 Results discussion ...... 67 4.3.1 Fourier Transform Infrared Spectroscopy ...... 67 4.3.2 X-ray diffraction analysis ...... 69 4.3.3 MAS-NMR technique ...... 71 4.3.4 Density and Porosity ...... 74 4.3.5 Compressive strength ...... 76 4.3.6 Microstructure ...... 80 4.3.7 Leaching test ...... 86 4.4 Conclusions ...... 89 Chapter 5 Red mud-based geopolymers synthesized with waste glass instead of water glass ...... 91 5.1 Introduction ...... 91 5.2 Materials and Methods ...... 92 5.2.1 Geopolymer preparation ...... 92 5.2.2 Test and analysis ...... 93 5.3 Results and discussion ...... 94 5.3.1 Compressive strength test ...... 94 5.3.2 Density and Porosity ...... 95 5.3.3 Fourier Transform Infrared Spectroscopy ...... 97 5.3.4 Microstructure ...... 98 5.3.5 MAS-NMR technique ...... 100 5.3.6 Leaching test ...... 101 5.4 Conclusions ...... 102 Chapter 6 Geopolymers foams from inorganic gel casting process ...... 105 6.1 Introduction ...... 105 6.1.1 Experimental procedure ...... 106 6.1.2 Characterization technique ...... 108

6.2 Results and discussions ...... 109 6.2.1 Fly ash-based geopolymer foams ...... 109 6.3 Red mud-based geopolymer foams ...... 114 6.4 Chemical stability ...... 117 6.5 Conclusions ...... 117 Chapter 7 Thermal shock and wear resistance ...... 119 7.1 Introduction ...... 119 7.2 Thermal shock test ...... 120 7.2.1 Analysis method ...... 120 7.2.2 Results and discussion ...... 121 7.3 Wear test ...... 126 7.3.1 Analysis method ...... 126 7.3.2 Results and discussion ...... 127 7.4 Conclusions ...... 129 Chapter 8 Advanced processing technologies for geopolymers ...... 131 8.1 Introduction ...... 131 8.2 Machinability test ...... 132 8.2.1 Materials and analysis method ...... 132 8.2.2 Results and discussion ...... 132 8.3 Direct ink printing ...... 134 8.3.1 Analysis method ...... 134 8.3.2 Results and discussion ...... 135 8.4 Interlocking pavers ...... 137 8.5 Conclusions ...... 138 Chapter 9 Overall conclusions and future work ...... 141 9.1 Conclusions ...... 141 9.2 Future directions ...... 145 Bibliography ...... 147 List of Abbreviations……...……..………………………….……………………163 Index of Figures………………………………..………………………...………..165 Index of Tables……………………………………………………………………169 List of Publications…………………………………………………………...…..171

Chapter 1 Introduction

1.1 Motivation: Environmental problems related to cement production

Portland cement is one of the construction materials most used in the word. This ubiquitous material is manufactured by heating raw materials up to temperatures as high as 1300-1500°C1. This process requires vast amounts of fuel such as coal and natural gas to be burnt in order to reach the temperatures required, with additional 2 chemical CO2 release as part of the calcination process . The result is that the production of 1 ton of Portland cement produces approximately 1 ton of CO2, 3 without considering the additional CO2 associated with transportation . The 5-7% of 4,5,6 the global CO2 emissions are due to the cement sector , which indicates its negative environmental impact. Geopolymer materials were developed as an eco-friendly alternative to Portland cement thanks to the associated low CO2 emissions. Geopolymers are aluminosilicate inorganic materials, consisting of SiO4 and AlO4 tetrahedral, linked together by sharing oxygen atoms7. Geopolymers, with comparable or better performances than traditional construction materials, allow for reduction of greenhouse emissions thanks to their capability to harden at room temperature, starting from precursors that, in turn, derive from low temperature treatments. In addition, geopolymers may be even more advantageous in terms of low CO2 emissions if produced by replacing virgin materials with inorganic waste (industrial residues) and sodium silicate with silicate waste. The sodium silicate, also known as “water glass” is one of the reagents mostly used during geopolymer synthesis, providing the system with mechanical durability. However, sodium silicate is obtained through an expensive and polluting process because this requires temperatures above 1300°C, which causes the release of CO2. The preparation of geopolymers through the use of waste materials that can not be recycled in conventional processes represents a reduction of emissions coupled with the preservation of valuable resources, moving forward sustainable development. Therefore, the development of geopolymers based on waste materials represents a great benefit from both an economic and an environmental point of view, and such

Chapter 1

geopolymer materials, if exhibiting suitable mechanical properties and durability, can be exploited as an alternative to the Portland cement currently in use.

1.2 Aims and Objectives

The main objectives of this research project are to develop a technology to produce geopolymer materials which should contribute to decrease the CO2 emissions caused by cement production as well as by sodium silicate synthesis and to preserve valuable resource such as metakaolin avoiding landfilling of waste materials. These goals were achieved through the production, at low temperatures, of geopolymers incorporating waste materials not used in other industrial sectors. In particular, the specific goal was to produce geopolymers using soda lime waste glass, fly ash and red mud, avoiding the use of sodium silicate normally employed during geopolymers synthesis. The production of geopolymers with competitive mechanical strength and chemical stability was considered fundamental in this research. Moreover, to draw attention on alternative applications, light-weight geopolymers were developed and to characterize technical properties, specific tests usually carried out in ceramic industry, were performed in order to assess possible application options for the new materials beyond the construction sector. Eventually, the ability to machine and print 3D geopolymers was considered to demonstrate for the first time potential further use of the materials in components of complex shape and surface finish. The thesis is organized in the following manner. Chapter 2 contains the fundamentals introducing the topics of relevance for the research project, including environmental problems correlated with the waste materials used in this project and basic knowledge of geopolymer materials. Chapter 3 presents the fabrication and characterization of geopolymers produced using fly ash, red mud and waste glass as aluminosilicate raw materials and sodium silicate and hydroxide as alkaline solution. The use of waste glass as silica supplier avoiding the consumption of water glass was investigated in chapters 4 and 5 on fly ash-based and red-mud based geopolymers, respectively. The effect of the addition of waste glass and the molarity of the sodium hydroxide solution on the properties of the new materials was investigated in these

Introduction

chapters. Chapter 6 presents the development and characterization of porous geopolymers realized through an inorganic gel casting process. In Chapter 7 the behavior of fly ash-based geopolymers when subjected to thermal shock and wear tests was investigated, while, Chapter 8 studied, for the first time, the machinability of the developed geopolymers. In detail, the capacity to be worked on the lathe and drilled with metal tools was assessed. The first attempt using waste containing geopolymers in 3D printing technology was also introduced in this chapter. Finally, Chapter 9 includes general conclusions, summary of the main achievements and suggestions for further investigations. Figure 1.1 gives a schematic overview of the research tasks carried out in this project in the form of a graphical abstract.

Characterization Foams production Vigorous Mechanical stirring Microscostructure Wear test Thermal shock test Geopolymer development Chemical stability

Fly ash

NaOH Na2SiO3

+ + Coal power plant Red mud 20 µm

50 nm

Waste glass

Alumina plant Glass Lathe container

Drilling

Mechanical machining Direct ink Printing Figure 1.1: Schematic overview of the research project which focused on the development of geopolymers incorporating waste materials. Moreover, foam production, direct ink printing and mechanical machining of geopolymers were parts of this research project.

Chapter 2 State of the art

2.1 Introduction

The term “geopolymers” was introduced by Joseph Davidovits in 1978 to identify a new class of inorganic materials that resulted from mixing naturally aluminosilicate materials at near ambient temperatures in alkaline aqueous solutions7. The word “geopolymers” denotes the mineral nature (`geo`), close for chemical composition and structure to zeolite material, and the polymerization reaction similar to organic polymers (`polymers`). In broad terms geopolymers are a class of synthetic aluminosilicate materials and in the literature the terms “inorganic polymer” or “alkali activated binder” are often used as synonymous for geopolymers. Geopolymers have emerged as novel construction materials substitutes for Portland 8 cement, due to the fact that each ton of cement releases approximately 1 ton of CO2 . Geopolymer technology not only allows to drastically reduce the emissions, but also to use abundant waste materials that are not used in other industrial sectors and are urgent to dispose of, thus geopolymers are an eco-friendly innovative alternative to cement industry9,10. Furthermore, geopolymers can exhibit a wide variety of properties and characteristics depending on the raw materials and the manufacturing conditions11. The high compressive strength and chemical inertness, the excellent fire resistance and the ability to tailor them using different reaction conditions make them potentially useful in many applications beside concrete, such as fire and acid resistance coatings, thermal shock refractories, high-tech resin systems, toxic waste immobilization and storage or replacement for ceramic components12,13,14,15. For these reasons geopolymers have been receiving increasing attention in recent years and as demonstrated in Figure 2.1 the number of publications on geopolymers has increased significantly in the last 10 years.

Chapter 2

250

200

150

100 Publications

0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Years

Figure 2.1: Scientific papers published per year with the keyword “geopolymers” (according to Web of science16)

The further development and understanding of geopolymer technology is of significant commercial interest because these materials can exhibit superior chemical and mechanical performances at a competitive cost.

2.2 Raw materials used in geopolymerization

The raw materials typically used in geopolymer manufacturing are precursors rich in silicon and aluminium in an amorphous form, that are dissolved in a concentrated aqueous solution usually composed of alkali hydroxide and silicate17,18. A key attribute of geopolymer technology is the versatility of the manufacturing process: geopolymers could be synthesized from a wide range of aluminosilicate virgin or waste materials17. Through geopolymerization it is possible therefore to use a large amount of by-products and waste materials and transform them into useful products thanks to the geopolymer capability to immobilize and stabilize waste inside their network 11,19. Several materials have been investigated in the last few years for geopolymer production, such as basalt fibers20, wood21, plasma vitrified air pollution control

State of the art

residues22, blast furnace slag23,24, rice-husk ash25, spent coffee grounds26 and red mud27,28,29,30. Although the most common material among them is metakaolin, the utilization of by-products, such as slag and fly ash, has been considered highly promising due to their abundance and availability worldwide26. For the use of a wider range of aluminosilicate sources, further investigation is needed on the effect of different contaminants on binder formation to obtain specific properties for a given application at a competitive cost. The following section provides a brief overview of the raw materials used in this investigation including the environmental problems related to them and the current knowledge about their use in geopolymer synthesis.

2.2.1 Fly ash

Fly ash is a by-product derived from the combustion of finely ground coal used as fuel in the generation of electric power. At 1500°C the coal burns instantaneously and the remaining matter solidifies by rapid cooling into fine spherical particles. Fly ash is transported by flue gases from the combustion zone to the particle removal system, where it is separated from the combustion gases by mechanical and electrostatic separators31.

Flue gas Coal bunker Electrostatic filter Flue gas

Coal mill

Fire room Precipitator

burner Fly ash silo

Water bath

Figure 2.2: Process of coal fired power plant with generation of fly ash.

Chapter 2

The types and relative amount of incombustible coal and the manufacturing process determine the chemical composition of the fly ash32. In 2011 the global production of fly ash amounted to 620 Mt and according to Heidrich et al.37 just the 53% of it is reused while the rest has to be disposed of with the associated problems of space, risk of air pollution and leaching33. Geopoloymers have been proven to be a valuable path for fly ash reuse. Moreover, combustion residues do not require thermal pre-treatment, adding another energetically attractive advantage for the use of fly ash in this new technology. Fly ash particles are typically spherical, ranging in diameter from 1 to 150 µm34 and the chemical composition is mainly composed of the oxides of silicon (SiO2), aluminium (Al2O3), iron (Fe2O3) and calcium (CaO). Magnesium, sodium, potassium, titanium are present in lower amount, while heavy metals such as Cd, Pb, Cu and Zn are in trace quantity35. Due to the rapid cooling of burned coal in the power plant fly ash consists mostly of amorphous particles, while small amount of crystalline material are usually identified as mullite, haematite, magnetite, quartz and unburned carbon residue7,32. According to ASTM C61836 fly ashes can be divided into two distinct categories: class F and class C, as described in Table 2.1.

Table 2.1: Main categories of fly ashes7. Class F CaO content less than 10% Usually produced from anthraciteand bituminous coals Class C CaO content higher than 10% Usually produced from sub-bitominous and lignite coals

The amount of coal combustion by-products produced worldwide is increasing each year as the global need for energy continues to grow37. In fact the reuse of fly ash in sustainable applications is becoming urgent.

Fly ash in geopolymer synthesis

Fly ash is considered the most suitable waste-material for geopolymers thanks to its great reactivity and availability together with its favorable size and shape 10. Fly ash

State of the art

38 consists of mainly SiO2 and Al2O3 in reactive glassy form that rapidly dissolves under alkaline conditions39. For these reasons the use of fly ash in geopolymer synthesis has been studied recently by many authors, as Palomo et al.33,40 and Duxson et al.41. Several factors, correlated with the nature of the fly ash, are reported to influence the geopolymer reaction. Among them, the most important are summarized as follows:

 SiO2/Al2O3 weight ratio of fly ash should be preferably in the range between 2.0 and 3.5;  Percentage of unburned material lower than 5% because it increases the liquid to solid ratio;

 Fe2O3 content not higher than 10%;  Lower content of CaO, therefore preferably class F;  Content of reactive silica, such as glass, higher than 40%;  Particle size under 100 µm;  High content of reactive vitreous phase. The higher the amount of glass phase in the fly ash, the faster is the dissolution process and higher the degree of reaction. The results achieved so far in literature for fly ash-based geopolymers confirm the good mechanical performance, chemical stability and high temperature resistance, making them suitable for different applications42.

2.2.2 Red mud

Red mud also called bauxite residue is a by-product generated during the Bayer´s process for the production of alumina. Bauxite is processed in a hot NaOH solution that converts alumina to dissolved aluminum hydroxide. The solid impurities that are filtered from the solution are called red mud43 and it is estimated that for each ton of alumina, 1.5-1.6 tons of this by-product are produced44,45,46.

Chapter 2

Bauxite ore

Rod Mill Grinding

Lime, NaOH Digestion Stream

Washing

Filtration Red mud to waste

Alumina to Precipitation calcing

Figure 2.3: Flowsheet depicting the Bayer Process for producing alumina from bauxite47.

According to Nan Ye et al.48 approximately 120 million tons of red mud are generated annually worldwide, so that new applications of this industrial residue are urgently needed. This huge quantity of bauxite residue is classified as toxic waste due to its high basicity and leaching potential that make it difficult to dispose of30. Red mud is usually disposed of in on-site waste lakes for further dewatering, due to the high water content and alkalinity with an average pH value of 11.3, consolidation and storage49. This storage has high maintenance costs, beside serious environmental risks such as caustic exposure for living organisms, leakage of alkaline compounds into the ground water and overflow of materials during storm events43. The mineralogical composition of bauxite residue depends on bauxite source and process methods, but usually comprises mainly oxides and hydroxides of Fe, Al, and 50 Si with minor amount of CaO and TiO2 . Considerable effort has been made to find an option for the reuse of red mud residue, particularly as a construction material, but its reuse is so far rather limited.

State of the art

Red mud in geopolymer synthesis

Through geopolymerization, red mud can be recycled and transformed into useful products, reducing therefore pollution risks of land disposal. Geopolymers have in fact the capability to immobilize and stabilize inside the matrix different pollutants and heavy metals when wastes are used as aluminosilicate source11. Red mud could be considered a potential raw material for geopolymers thanks to its highly alkaline nature and the high amount of Al2O3 and SiO2. Indeed, thanks to the high alkalinity, the incorporation of red mud requires less alkali activator in the solution, thus reducing the use of the most expensive reagent in the geopolymerization process.

Nonetheless, the red mud SiO2/Al2O3 molar ratio is too low to ensure the formation of a geopolymer. For this reason red mud is usually used as aluminosilicate source in combination with metakaolin51 or fly ash52,28. According to the literature53,54,27 , when red mud is added to the mixture as substitute for fly ash or metakaolin, an increment of the mechanical properties is produced, reaching a final value of around 10-20 MPa. This is true up to a maximum of 15-20% of red mud. Exceeding this amount the mechanical properties usually decrease due to the reduction of the amorphous content in the geopolymer53. The geopolymerization degree decreases with the addition of red mud and the activation of fly ash is inhibited implying a low dissolution ratio54. For this reason, further studies on the optimization of the parameters have to be carried out to explore the potential of this raw source and especially to allow the safe use of larger amounts of red mud in geopolymerization processes.

2.2.3 Waste glass

In the glass recycling process, the research and development of sorting strategies for the separation of contaminants from the recyclable glass fraction is of extreme importance55. Since the early 1970s the technology used in glass recycling processes has been evolving, from magnetic/metal detectors in the 1980s until glass cullet color sorting in the 1990s56. Glass fragments resulting from urban waste collection contain several contaminants of ferrous, metallic, wood, paper, plastic and ceramic glass or

Chapter 2

glass-like nature. Such impurities can negatively affect the glass production process and the final product. The presence of non-glass materials in the furnace-ready cullet creates heterogeneities that can escape regular checks and cause fracture of the bottles on the filling lines. At the same time organic compounds react with the sulphates generating SO2 during glass melting, which creates heavy persistent foam on the melt surface57. For these reasons the reduction of the contaminants in the glass recycling process is essential. Automatic sorting of pollutants is usually performed: ferrous and metallic contaminants are detected and discarded by means of magnetic sorters; paper, plastic and other light materials are sorted by different densities, while infrared sensors are usually used to detect opaque cullet such as ceramic and ceramic/glass. Thanks to these developments the recovery of glass in different urban waste separate collection, in order to manufacture new glass containers (“closed loop recycling”), has been implemented with success in the last years. In 2015, for instance, 73% of the overall amount of glass packaging in the European Union was recycled58. The approach is favorable to saving both energy and raw materials but it cannot be extended further due to the decline of urban glass quality caused by the color mixed59 and the excessive quantity of non-glass and organic substances that make the sorting step difficult and expensive60. For these reasons, an accurate recycling process is of high importance and there is always a glass fraction in which impurities are concentrated that remains practically useless and mostly sent to landfill as a consequence55. In Italy in particular, the 70% of the glass collected by urban collectors is recycled in glass production61,62, while for the other 30% new solutions are needed. Sasil s.p.a is an Italian company situated in Brusnengo (Biella) that, thanks to an innovative technology and machines with a most sensitive level, is able to recycle the rejected quantity where impurities are concentrated, to produce the so-called “glassy sand”, which is reused in the glass containers industry63. During this process 85% of the waste glass is converted to glassy sand, characterized by a granulometry between 0.08 – 0.8 mm, while 15% of the glass obtained from this process has a granulometry below 100 µm.

State of the art

Primary 25% 79% Glass powder treatment <100 m

Glass container 75% 21%

Glass factory to Glassy sand produce new glass containers

Figure 2.4: Schematic representation of glass recycling process in Sasil S.p.a.

This latter fraction of glass, due to its high specific surface, creates foam problems during glass processing in industrial furnaces. For this reason Sasil s.p.a is forced to reject the production of glass with a granulometry size below 100 µm, which amounts to 60-70000 tons/year. The necessity to find a solution for this part of soda lime waste glass together with the high surface energy and, consequently its high reactivity, are the motivation for the use of this waste glass as raw material in this research project.

Waste glass in geopolymer synthesis

Since 1960s many studies have been conducted to assess the possibility of different waste glass applications in the construction industry, as lightweight engineering material, cementitious material, pavement materials and aggregate in concrete59. Soda lime glass waste has been the most studied, while not many studies have investigated the use of other types of waste glass as aggregate or substitute for cement in concrete. Thanks to the high proportion of alkali and silicate, not only could recycled glass be a valid candidate for the production of geopolymers, but moreover the amount of water glass and sodium hydroxide normally used during the synthesis could be reduced or completely eliminated64. With the increase of interest in this new technology, different types of glasses have been exploited in geopolymer

Chapter 2

formulation: waste glass fibers65, borosilicate glass from pharmaceutical package66, fluorescent lamps67, solar panel waste glasses, glass produced by DC plasma treatment of waste68, thin-film transistor liquid display panels and glasses derived from the vitrification of municipal residues69. To compensate the low amount of

Al2O3 in glass composition, glass is usually mixed with fly ash or metakaolin to obtain the silica alumina network characteristic of the geopolymer material. Different studies demonstrated that the introduction of waste glass decreases the degree of polymerization, thus normally no more than 30% of glass is used during geopolymer synthesis64,70. In the last few years Tchakoute´ et al.25 and mainly Carrasco et al.71,72,73 have demonstrated the possibility of producing alkaline solution from the waste glass as alternative to the water glass. Water glass, whose synthesis entails high energy and environmental costs, is used in geopolymer synthesis because it facilitates the formation of stronger and more durable geopolymers. In all these studies the waste glass is produced by mixing the glass with NaOH or Na2CO3 solution and subsequently filtering, in order to obtain an alkaline solution ready to be mixed with raw materials. In this way the water glass obtained from the waste cullet allows to save an expensive chemical reagent but, after filtering, the glass is so far disposed of, meaning that the approach does not lead to complete elimination of the glass residue.

2.2.4 Alkaline activators

In addition to the aluminosilicate raw materials described in the sub-chapter above, a chemical activator is required to initiate the geopolymerization reaction. The alkaline solution is a key factor in the geopolymer synthesis and for this reason it needs to be set up to allow a rapid dissolution of the raw materials that leads to a rapid increase in solution of aluminate and silicate units. The chemical activator is usually an alkaline liquid mainly composed of hydroxide and silicate solutions74.

Hydroxide solution

Any ion of the I group can theoretically be used as the alkali element (M) in the geopolymer activating solution, and the choice depends on the type and composition

State of the art

of the source material as well as the application of the final geopolymer. The presence of cations in the original material as impurities or added as metal hydroxides is also considered to be important due to their potential catalytic role75. Optimum geopolymer properties are obtained when the M+ concentration is sufficient to provide a charge balancing mechanism for the substitution of tetrahedral Si with Al. However, an excess of alkali concentration in the geopolymer framework produces unwanted by-products such as sodium carbonate by atmospheric 17,76 carbonation or efflorescence . For these reasons the M2O/Al2O3 molar ratio should be optimally close to 1. The metal cations mainly used in hydroxide solution are sodium or potassium. Both ions have different effect on the geopolymerization process and on the final structural properties. Smaller sodium cations increase the rate of dissolution and better stabilize the silicate monomers and dimers in the solution. Whereas the larger size of potassium cations increases the condensation rate and accelerate the reaction kinetics. Moreover, potassium cations facilitate the formation of larger silicate oligomers and more precursors providing better setting and stronger compressive strength13. Duxson et al.77 observed that potassium is incorporated in the geopolymeric network in preference over sodium and the structure containing potassium exhibits a lower degree of crystallinity. NaOH and KOH are the most commonly used hydroxide solutions. The use of sodium solution is more widespread in the literature due to its low cost and wide availability, whereas the potassium based solution displays more favorable phase behavior and rheology78.

Silicate solution

Sodium silicate is a generic name for a series of compounds with the Na2O·nSiO2 formula, available in aqueous solution and in solid form79. The sodium silicates with different n values can have different properties with varied industrial applications. As mentioned before, together with hydroxide solutions, silicate solutions are the most used in the alkaline dissolution of aluminosilicate raw materials, because they facilitate the formation of stronger and more durable geopolymers. The properties of

Chapter 2

soluble silica are of extreme importance since they affect the workability, composition and microstructure of the geopolymer material. Sodium silicate was first discovered by Van Helmont when he combined together silica with an excess of alkali. Afterwards J. N. von Fuchs discovered that by dissolving silica in caustic potash, a product with some properties very similar to those of glass in a solution, could be obtained. For this reason sodium silicate is also called water glass or water liquid80. Sodium silicates are produced with the combination, in different proportions, of high purity silica sand (SiO2) and sodium carbonate (Na2CO3). The fusion of these materials at temperatures higher than 1000°C produces amorphous solid glass cullet. The cullet is fed into a reactor and mixed with water and steam to create a high pressure environment in which it dissolves to obtain the liquid silicate.

Commercial grades of liquid sodium silicate can be produced with mass ratio of SiO2 to Na2O from 1.60 to 3.80. The structure and properties of vitreous and liquid silicates also vary with their composition. Water glass is normally used in liquid form with modulus ranging between 1.6 and 3.3. The dissolution of solid sodium silicate is an endothermic reaction and the dissolution rate and the solubility of vitreous silicates decreases as the modulus increases. The process of synthesis of water glass is costly and has negative consequences on the environment because it is necessary to reach high temperatures in the 71 decarbonation of Na2CO3 . Indeed, it is estimated that emissions amount approximately to 1.514 kg of CO2 per kilogram of sodium silicate produced without considering the energy expended during the raw materials extraction81. For these reasons in the last years some studies have investigated the opportunity to obtain geopolymers with high mechanical performance without using water glass or producing water glass from recycled materials.

Other activators

Many researchers have studied the use of alkali carbonate solutions for activating slags, but their use in the activation of fly ash is much less widespread78. The use of mixed hydroxide carbonates activating solutions gives a poorly reacted product, due to the lower reactivity of fly ash compared to slags82. For this reason,

State of the art

when fly ash is used as aluminosilicate source, a higher level of available alkalinity is required, so a stronger base than M2CO3 (M: Na or K) is required. Phair et al.83 produced geopolymers by activation of fly ash with aqueous sodium aluminate, while Brew et al.84 used aqueous sodium aluminate in combination with silica fume to produce a high purity geopolymer. Some works are aimed at the use of solid sodium aluminate in “just add water” type one-part mix geopolymers, using solid geothermal silica as a silicate source85. This research is at the early stage of development but further investigations into these methods would be justified, considering the difficulties associated with the treatment of concentrated caustic activating solutions and the environment issue.

Water

Water is of absolute importance during geopolymer synthesis and its quantity depends from different factors such as the source materials, the particle size, shape, distribution and specific surface area17. The total mass of water is the sum of the water contained in the hydroxide and silicate solutions and the mass of extra water, when this is added to the mixture. The presence of water in geopolymerization is important to control the speed of the reaction that has to be slow enough to allow the complete dissolution of the raw materials, but fast enough to allow the formation of a dense network with a reasonable compressive strength. Water takes part in the dissolution, hydrolysis and polycondensation reactions during the geopolymer synthesis providing the medium for the dissolution of aluminosilicate and the transfer of various ions and hydrolysis of Al3+ and Si4+ 86. As a result, water has a strong effect on the geopolymer formation, structure of the geopolymer gel and final properties. Due to the fact that water is indispensable during geopolymer synthesis, it is important to preserve the water evaporation during the curing process. As a general trend confirmed from different authors87,88,89, less water leads to a higher compressive strength, while increasing the H2O/Na2O ratio the dissolution and reaction rates decrease. Moreover, an increase in water content results in higher porosity 90.

Chapter 2

As a conclusion, water and environmental conditions especially humidity and temperatures should be taken into consideration because they can drastically affect the reaction parameters91.

2.3 Geopolymerization: Reaction mechanism

Geopolymers are classified as inorganic polymers consisting of repeating molecular units called monomers. The process by which the monomers react with each other to form a 3D network is called “geopolymerization”. Geopolymer synthesis, as described in the previous paragraph, involves an aluminosilicate source and an alkaline solution usually composed of hydroxide and silicate solutions92.

Geopolymers consist of SiO4 and AlO4 tetrahedra, linked together by sharing oxygen atoms, analogous to those of natural zeolites. The substitution of Si4+ ion with the Al3+ ion creates a negative charge that is balanced by positive ions such as Na+, K+. These oligomer structures are defined by Davidovits7,93 as “polysialate”, in which sialate is an abbreviation for aluminosilicate oxide.

O O O O O K+ + (-) Na Si Al Ca+ Li+ + O O { H3O

Figure 2.5: Tetrahedral configuration of sialate Si-O-Al-O.

Polysialates have the following empirical formula:

Mn(-(SiO2)z-AlO2)n,wH2O Where “z” could be 1, 2, 3 or higher; M is a monovalent cation such as potassium or sodium and “n” is the degree of polycondensation. The polysialate molecular structures include at least four elementary units classified according to the Si:Al atomic ratio: Si:Al = 1, poly(sialate); Si:Al = 2, poly(sialate-siloxo);

State of the art

Si:Al = 3, poly(sialate-disiloxo); Si:Al > 3, poly(sialate-multisiloxo);

tri-sialate di-sialate sialate

Si : Al = 1

sialate-siloxo di(sialate-siloxo)

di(sialate-siloxo) Si : Al = 2

sialate-disiloxo sialate-disiloxo sialate-disiloxo Si : Al = 3

sialate-link

Si : Al > 3

Figure 2.6: Terminology of poly(sialate) geopolymers. Reproduced from ref7 with permission from Institut Geopolymere. Reproduced from ref.7 with permission from J. Davidovits.

The geopolymerization mechanism is a complex multiphase process, involving a quite fast chemical reaction. The exact process is currently not fully understood, mainly due to the extremely rapid dissolution and polymerization in the initial stages, which impede quantitative analysis94. Additionally, many variables such as precursors and activation solution greatly affect the geopolymerization mechanisms95.

Chapter 2

A schematic conceptual model for the geopolymerization reaction, Figure 2.7, was presented by Provis et al.96,97. This model was originally proposed for the metakaolin geopolymerization but it is also applicable to other raw materials sources.

Aluminosilicate source

Dissolution

Silicate monomer Aluminate monomer

Oligomerisation

Aluminosilicate oligomers

Polymerization Nucleation Aluminosilicate Aluminosilicate polymer `nuclei`

Gelation Crystallisation Transformation Aluminosilicate Geopolymers gel

Figure 2.7: Schematic reaction sequences proposed by Provis et al. for the geopolymeric process, (Figure reproduced with permission from Elsevier96).

The model presents different stages that mainly occur simultaneously during the reaction but can hardly be separated98. Although the exact process has not been sufficiently understood yet, Davidovits7 divided it into four different phases, namely: 1. Dissolution: dissolution of Si and Al from the aluminosilicate source in a strongly alkaline aqueous solution by alkaline hydrolysis (consuming water), producing aluminate and silicate species; 2. Nucleation: formation of oligomers in the aqueous phase; 3. Gelation: condensation of oligomers and creation of a three-dimensional aluminosilicate structure; during this process the water consumed during the dissolution stage is released41.

State of the art

4. Polymerization: after gelation the system continues to rearrange, as the connectivity of the gel network increases, resulting in a three-dimensional aluminosilicate network normally attributed to geopolymers. Nucleation and dissolution steps are highly dependent on thermodynamic and kinetic parameters. Chemically the geopolymer reaction can be described by the following chemical reactions, reproduced with permission from7: 1. Dissolution: a. Creation of the tetravalent Al in the next sialate group:

(1a)

b. Alkaline dissolution by linking OH- anion to the silicon atom, which changes its valence to pentavalent state:

(1b)

2. Nucleation: Bond cleavage of Si-O-Si group to form an intermediate silanol Si-OH on the one hand and a basic siloxo Si-O- on the other hand:

(2)

Chapter 2

3. Gelation: a. Further formation of silanol Si-OH groups and isolation of the ortho-sialate molecule as the elementary unit in geopolymerization:

(3a)

Reaction of the basic siloxo with the cation Na+ and formation of Si-ONa terminal bond:

(3b)

4. Polymerization: a. Condensation of ortho-sialate units via Si-ONa and OH-Al reactive groups, to create the cyclo-tria-sialane structure. During this step NaOH is liberated and reacts again supporting polycondensation.

(4a)

Na-poly(sialate) Nepheline framework

State of the art

b. Condensation of ortho-sialato-disiloxo cyclic structure in the presence of water glass, with release of NaOH:

(4b)

c. Na-poly (sialate-disiloxo) condensation into feldspar crankshaft chain structure:

(4c)

Feldspar crankshaft

chain

2.4 Geopolymer foams

Geopolymeric cements or resins can expand or be foamed to produce lightweight materials with thermal insulating properties, representing an alternative to the traditional foam materials7. Geopolymer foams are very promising materials since

Chapter 2

they are synthesized at temperature below 100 °C and they possess properties similar to foamed glass or foamed ceramics, both of which are produced at higher temperatures, above 900 °C99. Thanks to the inorganic structure, geopolymer foams resist high temperatures, up to 800-900 °C, acting as a thermal barrier. Thanks to this property geopolymers are more efficient as fireproof materials than organic foams representing a unique material in terms of mechanical properties and fire safety100. Therefore they may be applied in various industry branches, especially in construction applications involving thermal gradients. As there are no standard procedures and parameters in the process of production of geopolymer foams, such parameters should be selected individually considering the available raw materials and final application. The successful production of foamed geopolymers requires a delicate optimization of two parameters7: - Kinetics of peroxides decomposition with production of oxygen; - Increase in viscosity of the geopolymer. One of the greatest advantages of this technology is the simplicity of the process that allows controlling the pore size and distribution during its synthesis. Beside the intrinsic porosity characteristic of the geopolymeric material, with a different methodology it is possible to include also a variable. In this way geopolymeric foams can be achieved with a total porosity close to 70% or more, thanks to the introduction of macro and micro pores. The introduction of a foam agent in the geopolymeric mixture creates a material with a cellular structure and low density; the same result can be achieved with chemical agents able to release gas molecules in the geopolymeric matrix. Generally, by increasing the foaming agent, both the total porosity and the dimension of the pores increase. Pore dimensions and distribution depend on the foaming agent and on the process used.

Foaming techniques

Geopolymer foams can be produced either by the pre-foaming method or by mixed- foaming process101. In the pre-foaming method, a suitable foaming agent is mixed with water and after that the foam is combined with the geopolymer mixture.

State of the art

Meanwhile, in the mixed-foaming method, the foaming agent is added into the geopolymeric slurry. Different chemical reagents could be used in the production of geopolymer foams: - Fillers containing impurities capable of generating porosity, such as "silica fume" or silicon carbides. - Metal powders such as metallic aluminum or zinc powder that, when reacting in alkaline solution, release hydrogen following the equation102: - - 8Al + 2OH + 2H2O → 4Al2O + 3H2 (1) - Hydrogen peroxide and organic peroxides, that through an exothermic 103 reaction decompose in water and oxygen . Bubbles of O2 are trapped within the paste, expanding and increasing the volume. Hydrogen peroxide is thermodynamically unstable and can be easily decomposed into water and oxygen: - - H2O2 + OH → HO2 + H2O (2) - - HO2 + H2O2 → H2O + O2 + OH (3) The synthesis using hydrogen peroxide is determined by the kinetic of peroxide decomposition which influences the increase in viscosity, pore size and pore distribution. Geopolymers foams can also be produced by introducing a large volume fraction of air bubbles into the mixture, mainly by using organic foaming agents such as soap, detergents and hydrolyzed proteins99. It can be concluded that the manufacturing of stable foams depends on many factors, such as the selection of the foaming agent, the methodology used to prepare the foam and the selection of raw materials.

2.5 Properties of geopolymeric materials

Geopolymeric materials, with a smaller CO2 footprint than traditional Portland cement, usually display very good strength and chemical resistance properties as well as a variety of other potentially valuable characteristics described in the next sub- paragraphs. These properties vary according to different factors including source materials, alkali activator type and content, curing parameters, water amount and mixing parameters.

Chapter 2

Compressive and flexural strength

The improvement of mechanical properties in geopolymeric materials is of extreme importance because their main application is in the construction industry. The compressive strength depends from the alkali solutions, Si/Al molar ratio, calcium content and curing parameters104. Davidovits7 reported that geopolymers possess high strength up to 20 MPa after only 4 hours at 20 °C. The compressive strength continues to increase with time, but it is reported that in the first 3 days all geopolymer mixes achieve more than 75% of their total strength105. Moreover, increasing the temperature accelerates the geopolymerization reaction thanks to the intensification of the reactivity of the aluminosilicate precursors. Improvements in the degree of the compressive strength were also observed when increasing the alkali content of the reactive solution106,107. In addition, water glass is usually used in combination with NaOH to increase the compressive strength, thanks to the high viscosity that helps the formation of a compact geopolymeric gel. Fly ash, activated with solution of sodium hydroxide and silicate, after 28 days of storage can reach compressive strength values close to 60-80 MPa108. Raw materials and alkali solutions determined the Si/Al ratios. Higher Si/Al ratios increase the amount of –Si-O-Si bonds to obtain a higher compressive strength, since the –Si-O-Si bonds are stronger than –Si-O-Al- and –Al-O-Al bonds. Geopolymers suffer from brittle failure with low fracture toughness. For this reason fibers can be incorporated in the geopolymeric matrix to increase the fracture toughness and to reinforce the flexural strength. Geopolymers could be reinforced with cotton fibers, carbon fibers, wood, wool and cellulose109,110,111. By incorporating, for example, 8% of cotton fibers, Alomayri at al.110 demonstrated that flexural strength increases from 8 to 32 MPa, moreover the presence of the fibers significantly reduces crack propagation showing a less brittle behavior.

State of the art

Resistance to chemical attack

Different authors tested the behavior of geopolymers in low pH acids that are responsible for the highest chemical attack. The aluminosilicate chemical network is resistant to any chemicals that are usually harmful for organic polymers7. Geopolymers have been proven to perform satisfactorily when exposed to hydrochloric and sulfuric acids112,113, sea water114, nitric acid115, sulfate and acetic acid116. For all the acids geopolymers demonstrated far better resistance than Portland cement. Geopolymeric materials immersed in the acids reported mass losses of only 6-7%. Portland cement contains CH, which is the most soluble among the hydration products, and when exposed to water it dissolves increasing the porosity and vulnerability of the paste. Thanks to the lower Ca/Si molar ratio geopolymers are more stable in acid medium and in particular geopolymers manufactured with sodium hydroxide and cured at elevated temperature demonstrate the highest chemical stability114.

Resistance to high temperature and to fire

Traditional Portland cement shows weak performance when subjected to thermal treatment: when the temperature rises to 300°C, it starts to disintegrate with rapid deterioration of its compressive strength. Geopolymeric materials on the contrary show high stability when subjected to high temperatures, even around 1000°C117. Rovnanik et al.118 reported that geopolymers, under extreme heat load exhibit a morphological ability to form a new microstructure based on akermanite, which possesses superior mechanical and thermal resistance and for this reason can be considered a prime candidate for fire protection materials. Geopolymers under heat load do not release toxic fumes and after exposure at 600°C-800°C they show low weight loss, between 5-12%. Fernandez J. A.108 et al. demonstrated that resistance to bending remains constant after thermal treatment up to 400 °C while at higher temperatures the resistance decreases to a third of the initial one. However, this value is still much higher than the one obtained for a traditional Portland cement. In addition, geopolymers at 600°C – 800°C preserve a high compressive strength value (40-50 MPa)119 because of relatively stable contraction of geopolymer paste at

Chapter 2

those temperature ranges. They also exhibit extremely low shrinkage in comparison to Portland cement. It can be concluded that geopolymers are suitable materials in applications with high fire risk.

Freeze-thaw resistance

Mass loss due to freezing and thawing is a typical problem for Portland cement concrete, while geopolymers have been proven to be resistant to the repetitive freezing and thawing cycles. Degirmenci et al.120 observed that after 25 freeze-thaw cycles no body disintegration or deformation is visible. Geopolymers could support up to 300 freeze-thaw cycles with the loss of 5% of the original compressive strength121, which means an irrelevant weight loss. Gifford et al.122 found out that the durability depends mainly on the air content and air bubbles distribution, so the poor performance of materials is attributed to this phenomenon.

Abrasion resistance

The resistance to abrasion is an important property due to the fact that the abrasion of the surface can expose the inner part of the structure to the environment. This of course can influence the properties of the materials. Ramujee et al.123 proved that the wear resistance of geopolymer material is better than Portland cement´s and in particular cement´s depth of wear resistance is 61% higher compared to geopolymer for a 12-hours duration and 64% higher for a 24-hours duration. Moreover, better results could be achieved when geopolymers are used as matrix in composite materials124.

Heavy metals incorporation

Heavy metals are components of many industrial residues used as raw materials in geopolymers production as mine tailings, electric furnace slag, electroplating sludge and municipal solid waste incineration fly ash125. This type of wastes are rich in heavy metals and should not be let to infiltrate into surface or ground water, which

State of the art

can lead to environmental problems. Fly ash-based geopolymers provide a satisfactory method for heavy metals immobilization with a low permeability and long durability126. Metals such as Co, Cu, Pb, Cd, Ni, Zn, Pd, As, Ra and U can be incorporated in the 3 dimensional geopolymer network reducing the mobility of the heavy ions through metal hydroxide precipitation, ion substitution or physical encapsulation127,128,129. The amount of ions that a matrix can incorporate without losing structural integrity is limited and determined by chemical and physical properties of each element130. In most cases the metals are not incorporated in the crystalline matrix but in the amorphous one131. Metal ions influence the geopolymer structure differently, mainly because of differences in their atomic ratio, electronegativity and reactivity. The pH of the alkali solution, used during geopolymer synthesis, strongly affects the heavy metals incorporation. An increase in alkali solution pH improves ions dissolution and consequently their immobilization inside the matrix132. Research results indicate that NaOH compared to Na-silicate has more effective activation properties for ion immobilization. This correlates with the higher alkalinity of NaOH which favors minerals dissolution and consequently the creation of new phases involving ions inside the structure133. Also the Si/Al molar ratio is important for the ion immobilization. For example, it was shown that leaching of Cr and Cu decreased as the Si/Al ratio increased, while Zn reached its lowest point at an intermediate Si/Al ratio129,134. In addition, a Si/Al ratio near 2 was found to be more suitable for binding Pb in the geopolymer gel135. The immobilization efficiency is strongly related to the binder microstructure as well as to pore size distribution, pore shape and total porosity136. From the results collected so far in the literature it is apparent that heavy metals can be effectively immobilized in geopolymer matrices due to the participation of heavy metal cations in the balance of negative charge of Al in the frameworks, which indicates an important application of geopolymers in the management of hazardous heavy metal containing wastes126.

Chapter 3 Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

3.1 Introduction

In recent years, many countries have developed strict regulations for the disposal of industrial by-products, in order to minimize the impact of exploitation of virgin materials on the environment and to improve the management of wastes. Red mud is the major industrial waste generated during the Bayer process for the extraction of alumina from bauxite. Depending on the quantity and purity of the bauxite, the amount of red mud generated varies between 55% and 65% of the processed bauxite137, therefore worldwide a huge amount of red mud is produced worldwide. Due to the presence of an excessive amount of dissolved sodium hydroxide, used to extract silicates and alumina, red mud is characterized by a strong alkalinity that makes its disposal difficult from both a safe and an economic point of view. Although in the past extensive research has focused on the reuse of red mud for construction applications or ceramic components, a widely accepted technology is not available at the present moment138. In the last decades the amount of urban waste glass has increased worldwide and it is estimated that a city of 1 million habitants generates approximately 400 000 tons of waste glass per year81. For this reason recycling waste glass from urban collection continues to be of extreme importance in the world. Waste glass from urban collection consists mainly of containers, window panels, flat glass and glass contaminated with ceramic and metal impurities. The different glass compositions and colors hinder its reuse in conventional technologies, for this reason between 6 and 30% of the collected glass cannot be reused in glass production139. Geopolymers were primarily developed for the construction industry as substitutes for Portland cement but thanks to their attractive properties, they are being taken into consideration for many other applications such as refractory adhesive for glass, ceramics, composites, fire-resistant materials and high-tech weight structures13, 140.

Chapter 3

The growing interest in the geopolymer field is not only attributed to the eco-friendly and low cost technology resulting from the low temperature processing but it is also due to the possibility of incorporating in the geopolymer network waste materials which would be otherwise disposed of 10. Fly ash, a by-product of coal combustion in thermal power plants, is the most suitable material for geopolymerization because of its suitable alumino-silicate composition, availability worldwide and its pozzolanic property. This chapter explores the possibility to reuse also red mud and waste glass in geopolymer production considering fly ash as one of the components. The advantage of red mud lies in its high alkalinity and high content of alumina, while cullet glass takes advantage of its high amount of silica and high surface reactivity. The incorporation of different amounts of red mud and waste glass has already been evaluated by Bobirica et al.141, Novais et al.67, Kuman et al.44 and Mucsi et al.53, who reported a decrease in the geopolymer performance after the incorporation of such type of waste materials. In this investigation the fly ash used in the production of geopolymers was substituted with 10, 20, and 30 wt.% waste glass or red mud to study the possibility of creating a valuable product starting from a combination of problematic wastes. The different performances obtained with the introduction of red mud or waste glass were investigated, as well as the effect of the incorporation of different amounts of waste material. An in-depth comparison between the mechanical properties of the achieved geopolymers and the construction materials currently present in the market was carried out. At the end through the leaching results, the capability of geopolymers to immobilize heavy metal ions inside the matrix was also assessed. This work was made in collaboration with the Institute of Physics of Materials (IPM) in Brno. The results presented in this chapter are part of a previous publication led by the author 142.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

3.2 Materials and Methods

3.2.1 Characterization of the raw materials

Low calcium fly ash (FA) (Class F according to ASTM C618), red mud (RM) and soda lime waste glass (SLG) were used as raw materials in this investigation. Fly ash with a mean particle size d(0.5) of 21µm was provided by Steag power mineral GmbH, Dinslaken, Germany. Waste glass is a soda lime glass with mean particle size d(0.5) of 24µm obtained from Sasil S.p.A Biella, Italy, after being processed as described in chapter 2.2.4. Red mud, collected by Alteo, Gardanne, France, was dried at 100 °C for 3 hours to take out the humidity and then sieved to a particle size < 75 µm. Laser particle size analyzer Hydro 2000 MU was used to determine the particle size distribution of the fly ash and waste glass (Figure 3.1).

Figure 3.1: The volume-mean particle sizes for fly ash (a) and waste glass (b).

Chapter 3

The chemical composition of the raw materials was determined using X-ray fluorescence analysis (XRF) at a Spectro Xepos He energy dispersive X-ray fluorescence spectrometer (Spectro Analytical Instruments GmbH)a. For the analyses, fusion tablets were fabricated by adding 4.830 g of Li2B4O7 and 230 mg

I2O5 as flux to 1 g of sample powder. An Oxiflux fusion system (CRB Analyse Service GmbH) was used for fusion tablet fabrication. The results are given in Table 3.1.

Table 3.1: Chemical composition of raw materials (wt. %) determined by XRF. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol. Constituent Fly ash Red Mud Glass

SiO2 54.36 5.21 70.50

TiO2 1.07 8.05 0.07

Al2O3 24.84 15.21 3.20

Fe2O3 8.28 52.94 0.42

MnO 0.009 0.05 - MgO 2.06 0.38 2.3 CaO 2.56 2.95 10.00

Na2O 0.83 2.40 12.00

K2O 3.03 0.63 1.00

P2O5 0.38 0.54 -

Loss of ignition 2.10 10.77 0.40

The raw materials’ quantitative phase composition was determined by powder X-ray diffraction (XRD) combined with Rietveld refinement and the G-factor method143. XRD measurements were performed using a D8 Advance with DaVinci design diffractometer (Bruker AXS). The following measurement parameters were applied: range 7°-70° 2θ; step size 0.01125° 2θ, integration time 0.4 s; radiation: copper Kα; generator settings: 40 mA, 40 kV; divergence slit: 0.3°. Three independent samples were measured for fly ash and waste glass. Red mud was analysed five times because the quantification of some phases was less reproducible in this sample. Rietveld refinement was performed with the software TOPAS 4.2 (Bruker AXS). a The XRF analysis has been done by Dr. Katrin Hurle at the Institute of Mineralogy, University of Erlangen-Nuremberg.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

The G-factor method144, an external standard method, was applied to obtain the absolute contents of crystalline phases. The method is described in detail by Zhang et al.143. The amorphous content of the samples could be determined as the difference of the sum of all crystalline phases quantities to 100 wt. %. The G-factor, which is a device-specific calibration constant, was obtained by measuring a quartzite slice (fine grained rock of almost pure quartz) as an external standard under identical measurement conditions as the samples. The quartzite was previously calibrated with pure crystalline silicon powder, NIST Si Standard 640d. The mass attenuation coefficients (MACs) of the powder samples were determined from the chemical composition obtained through XRF data and the MACs of the elements presented in the International Tables for Crystallography145. The quantitative phase compositions of fly ash, red mud and glass, determined by G-factor method, are shown in Table 3.2, 3.3 and 3.4 respectively.

Table 3.2: Quantitative phase composition of fly ash, determined by G-factor method. The structures given in the references for each phase were used for Rietveld refinement. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol. Mineral Formula Fly Ash [wt.%] 146 Mullite 3Al2O32SiO2 18.5 ± 0.4 147 Quartz SiO2 17.2 ± 0.3 148 Hematite Fe2O3 2.08 ± 0.07 149 Magnetite Fe3O4 1.64 ± 0.06 150 Anhydrite CaSO4 0.88 ± 0.08 Periclase151 MgO 0.44 ± 0.02 Lime152 CaO 0.24 ± 0.02 Amorphous fraction - 59.0 ± 0.9

Chapter 3

Table 3.3: Quantitative phase composition of red mud, determined by G-factor method. The structures given in the references for each phase were used for Rietveld refinement. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

Mineral Formula Red Mud [wt.%] Goethite153 FeO(OH) 30.07 ± 0.80 148 Hematite Fe2O3 24.5 ± 0.4 154 Cancrinite Na8(AlSiO4)6(CO3)(H2O)2 10.2 ± 0.6 155 Gibbsite Al(OH)3 6.6 ± 0.9 156 Boehmite AlO(OH) 6.5 ± 0.2 157 Rutile TiO2 3.9 ± 0.3 158 Larnite Ca2SiO4 3.4 ± 0.2 159 Carnegieite NaAlSiO4 2.8 ± 0.5 160 Muscovite K[Al2(OH)2/AlSi3O10] 1.1 ± 0.6 Amorphous fraction - 7.2 ± 5.4

Table 3.4: Quantitative phase composition of waste glass, determined by G-factor method. The structures given in the references for each phase were used for Rietveld refinement. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

Mineral Formula Glass [wt.%] 147 Quartz SiO2 1.3 ± 0.2 161 Albite NaAlSi3O8 0.9 ± 0.2 162 Microcline K(AlSi3O8) 1.1 ± 0.2 163 Calcite CaCO3 0.3 ± 0.02 Amorphous fraction - 96.1 ± 0.6

The reproducibility of the three independent fly ash measurements was satisfactory. Fly ash is mainly composed of an amorphous fraction (59.0 ± 0.9 wt.%) and two main crystalline phases, namely mullite and quartz (Table 3.2). The mineralogical composition of red mud is rather complex, with goethite and hematite as main mineral phases (Table 3.3). The quantification of some minor phases was not sufficiently reproducible, probably due to the inhomogeneous distribution of these

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study minerals in the sample. The amorphous fraction of red mud is rather low while, as expected, glass is characterized by a highly predominant amorphous phase (96.1 ± 0.6 wt.%) (Table 3.4). In geopolymerization, the amorphous content plays a predominant role, since it is usually the most reactive part and the fastest to dissolve and takes part in the polymerization process. The high percentage of amorphous phase in fly ash is one of the reasons why this raw material is highly suitable for the production of geopolymers164.

3.2.2 Geopolymers production

Sodium hydroxide solution 8M was prepared by dissolution of NaOH pellets (Merk 99.5%) in distilled water, the water glass solution (PQ Corporation) has modulus 3 (SiO2/Na2O ratio) = 2.0, density at 20°C = 1.4 g/cm and viscosity at 20°C=200mPa.

The alkaline solution was prepared by mixing Na2SiO3 and NaOH solutions in a ratio of 2.5. This solution was prepared one day before the use to allow the exothermically heated solution to cool down to room temperature. Geopolymer samples were prepared by mixing fly ash, red mud or waste glass and alkaline solution with a head stirrer at 85 rpm until a homogeneous slurry was obtained. After mechanical mixing, a vibrating table was used to remove entrained air before being casted in polyethylene molds and sealing from the atmosphere. The cylindrical mould had a diameter of 14 mm and height of 31.5 mm. Samples were cured in a laboratory oven at 60°C for 24 hours. After curing, the samples were maintained at room temperature until they were mechanically tested.

Chapter 3

Aluminosilicate raw

Alkaline solution C] °

material [ NaOH sol. Fly ash 60°C- 24h + + + Casting

Na2SiO3 sol. Red mud or glass Temperature Time [h] Mechanical mixing at 85 rpm for 10 min

Figure 3.2: Schematic representation of the synthesis route for geopolymers.

One mixture, as reference, was prepared using only fly ash as aluminosilicate raw material, while three batches were prepared by replacing 10wt.%, 20wt.% and 30wt.% of fly ash with red mud and other three batches replacing 10wt.%, 20wt.% and 30wt.% of fly ash with waste glass (Figure 3.3).

10 wt.% 20 wt.% 30 wt.%

Red Mud Red

Waste Waste Glass 12 mm

Figure 3.3: Geopolymer samples prepared with 10-20-30 wt.% of red mud or waste glass.

For all the mixtures the ratio between the liquid alkaline solution and the raw materials was kept constant at 0.4. The samples designation and proportion are given in Table 3.5. Table 3.5: Summary of mixture proportions used in the experimental trial. Designation Fly ash Glass Red mud Alkaline sol./ Na2SiO3sol./

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

(wt.%) (wt.%) (wt.%) raw mat. NaOH sol. FA 100 - - 0.4 2.5 10SLG 90 10 - 0.4 2.5 20SLG 80 20 - 0.4 2.5 30SLG 70 30 - 0.4 2.5 10RM 90 - 10 0.4 2.5 20RM 80 - 20 0.4 2.5 30RM 70 - 30 0.45 2.5

3.2.3 Test and analysis methods

Compression test Compression test was performed on cylindrical samples with height twice the diameter. The tests were performed using a universal testing machine (Zwick Roell, Ulm, Germany, Series Z050) at a cross-head speed of 0.5 mm/min, with load cell of 30 kN. The compressive strength was defined as the maximum stress of the linear elastic part of the stress-strain curve. For each batch a minimum of 10 samples was tested to evaluate the 28-day strength of the specimens.

Flexural strength and fracture toughness Three point bending strength tests and chevron notch for fracture toughness determination tests were carried out using a Zwick/Roell Z50 machine. Ten specimens, with the geometry 4 x 3 x 20 mm3 and a minimum span length of 16 mm, were tested for each formulation type. The samples were cut and ground under drying conditions. For flexural strength determination, the specimens were loaded in three-point bending at room temperature at a constant cross-head speed of 10 µm/min. The flexural strength (σ) was calculated using the following equation165:

(3.1)

Chapter 3

where F is the fracture load, L is the span length, B the width and W the height of the specimen166. To determine the fracture toughness the chevron notch technique was used according to the methodology described in the literature 167,168. A chevron notch with angles of 90°C as shown in Figure 3.4 was cut into the samples using a diamond wheel.

Chevron notched geometry

Figure 3.4: Schematic representation of the chevron-notch (left) and the specimen in three point bending according to Dlouhy 167. Reproduced from ref.167 with permission from Elsevier.

Each sample with notch was placed in the three point bending machine and loaded up to fracture initiation. The tests were carried out at room temperature in normal atmosphere. Figure 3.5 shows the samples after fracture toughness measurements. In the middle the chevron notch and the propagation of the crack is visible.

1 cm

Figure 3.5: Samples used to perform fracture toughness measurements.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

169 The fracture toughness (KIC) was calculated with the following equation :

(3.2)

* where Y min is a geometrical compliance function, FM is the maximum load, which was obtained from load deflection traces as the maximum flexural load. The tests were carried out at room temperature; the cross-head speed applied was 0.5 mm/min.

Micro hardness Vickers hardness was measured by using a screw driven testing machine Zwick Z2.5 equipped with micro hardness head ZHU0.2. The hardness of the material is defined as the ratio between the applied load P and a representative contact area between the indenter and the material. The representative contact area depends on the shape of the indenter. For Vickers hardness test, the indenter has the shape of a pyramid with a square base with an angle of 136° between the opposite faces at the vertex. The corresponding hardness number HV in N/mm2 is obtained via equation:170

HV = 1.8544. (3.3)

where P is the indenting force in N and d is the average diagonal distance of the indentation measured, expressed in mm. The diagonal distance is calculated by means of a microscope. The standard notation for Vickers measurement reporting is HV followed by a number representing the indentation load in kg. In this study HV 0.2 was used for the tests, so a force of 1.961 N was applied.

Confocal microscope The measurement of surface patterns and roughness was carried out with a LEXT OLS3100 confocal laser scanning microscope, which allows to obtain live images and to measure the surface area texture. Thanks to the surface profile measurements it is possible to calculate the roughness profile of the geopolymer sample. Samples after three point bending were examined with the confocal microscope, in particular the chevron notch area where the fracture propagation proceeded.

Chapter 3

Scanning Electron Microscopy (SEM) The qualitative microstructural evaluation of the raw materials and final geopolymer samples was performed via scanning electron microscopy (SEM), (LEO 435 VP, LEO Electron Microscopy Ltd., Cambridge, UK and Ultra Plus, Zeiss, Jena, Germany) in order to evaluate the homogeneity of the waste glass and red mud incorporated in the geopolymer matrix. Pieces of samples after the compression strength test were used to analyze the microstructure.

Fourier Transform Infrared Spectroscopy (FT-IR) Fourier transform infrared spectra were recorded using a Nicolet 6700 device in the range between 4000 cm-1 and 400 cm-1. The specimens for FTIR were prepared using the KBr pellet technique: 2mg of sample were mixed with 200mg of potassium bromide before being compressed into pellets with an electro-hydraulic press applying a force of 50 kN.

Leaching test The release of heavy metals was evaluated according to the European standard for waste toxicity evaluation (EN 12457-2). Pieces under 4 mm were placed in distilled water, with a liquid/solid ratio of 10, and softly stirred at 25°C for 24 h. The resulting solutions were filtered through a 0.6 m filter and analysed using inductively coupled plasma (ICP; SPECTRO Analytical instruments GmbH, Kleve, Germany).

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

3.3 Results and discussion

3.3.1 Micromorphology of raw materials

a b c

10 µm 100 µm 10 µm

Figure 3.6: SEM images of the raw materials involved in the geopolymer synthesis: (a) fly ash, (b) red mud and (c) waste glass. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

Fly ash, Figure 3.6 a, is mainly composed of hollow spherical particles of different dimensions bonded together. The irregularly-shaped particles observed by the morphology analysis in SEM are identified as unburned particles or agglomerated minerals171. The morphology of red mud, Figure 3.6 b, is mainly composed of flake-shaped particles with visible agglomerates that stick together. Soda lime waste glass powder consists of a heterogeneous distribution of irregularly-shaped, smooth fragments of variable dimensions (Fig. 1c).

3.3.2 Compressive strength

In order to evaluate the compressive strength, 10 samples for each mixture were tested after 28-day ageing. Figure 3.7 shows the geopolymer sample after fracture. After failure, the sample fracture presents a conical shape that from ASTM C 39 Normative172 is stated as the fracture typically observed in concrete specimens, confirming the fragile nature of geopolymer samples.

Chapter 3

a b

12 mm

Figure 3.7: Image of compression test of 10SLG geopolymer (a) and geopolymer conical fracture after compression test (b).

The average results of compressive strength values are presented in Figure 3.8.

100 Fly ash 90 Glass 80 Red mud

Compressive(MPa) strength 20

0 0 5 10 15 20 25 30 Concentration of waste material (wt.%)

Figure 3.8: Compressive strength average of fly ash-based geopolymers and geopolymers were 10- 20-30 wt.% of fly ash is substituted with red mud or waste glass. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

The geopolymers prepared using only fly ash as raw material exhibited the highest compressive strength, with value of 75 ± 14 MPa. The incorporation of red mud or waste glass shown in Figure 3.8 indicates the same trend in strength development.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

The addition of both raw materials caused the compressive strength to decrease, while the amount of recycled glass and red mud added to the mixture increased. In particular the mixture containing 20 wt.% of glass cullet displayed the lowest mechanical strength, with a reduction of 40% of the initial compressive strength. The substitution of 10 wt.% of fly ash with waste glass affected considerably the compressive strength compared with red mud, while for the other percentages of substitution of red mud and glass the compressive strength values were in the same range. Many authors agree that higher red mud content is lead to overall lower compressive strength28. For example Kumar44, Mucsi53, He28 et al. demonstrated that with a ratio of 80:20 fly ash to red mud, a maximum value of ca. 25 MPa can be achieved. Moreover, increasing the amount of red mud over 20% was seen to cause a drastic decrease in the compressive strength of the material. Figure 3.8 shows, on the contrary, that there is not a significant difference with the addition of 20 or 30% wt. red mud to the original mixture. Besides, in both cases the mechanical strength maintains a satisfactory value of 60 MPa. Due to the high amount of silica in glass and of alumina in red mud, it is expected that the initial SiO2/Al2O3 molar ratio increases as the amount of glass rises and decreases when red mud is added to the geopolymer. Si-O-Si bonds are stronger than Si-O-Al and Al-O-Al bonds, for this reason a geopolymer richer in silica than in aluminum could exhibit a better mechanical performance173. Different authors have already tested the possibility of incorporating different types of waste glass in an alkaline activated matrix. The introduction of waste glass into fly ash-based geopolymers usually causes a decrease in mechanical properties, as confirmed by the literature. Bobirica173 and Novais67 and al. determined a decrease of up to 55% of the initial mechanical strength after introducing 10 and 20% wt. of fluorescent lamp waste glass, with a maximum value of 19 MPa. In this study, with the introduction of soda lime glass, a decrease in compressive strength close to 45% is observed, however a considerably higher value of 50 MPa was obtained in comparison with literature results. Therefore, it can be concluded that the compressive strength of the fly ash-based geopolymers with red mud and waste glass (45-60 MPa) was lower than that of pure fly ash geopolymer (75 MPa), but is still satisfactory for sustainable applications.

Chapter 3

3.3.3 Flexural strength

The flexural test is one of the most common tests conducted on hardened concrete, to determine the load at which the structural element cracks. The flexural properties were measured by 3 point bending test on beams having cross section of 3x4 mm2.

18 Fly ash 16 Red mud Glass 14

Load (N) Load 8

0 0,000 0,005 0,010 0,015 0,020 0,025 0,030 0,035 0,040 Deflection (mm)

Figure 3.9: Load – deflection curves for the different geopolymers investigated. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

In Figure 3.9, typical load-deflection curves for the samples with fly ash, fly ash plus waste glass (10 wt.%) and fly ash plus red mud (10 wt.%) are plotted in the same graph. The curves show a linear behavior until brittle fracture occurs followed by a sudden drop of load when fracture strength is reached. Different slopes of the linear part, even though affected by the real cross-section area of samples, reflect differences in the Young’s modulus of the materials.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

25 Fly ash Glass 20 Red mud

10 Flexural strength (MPa)Flexural strength 5

0 0 5 10 15 20 25 30 35 Concentration of waste material (wt.%)

Figure 3.10: Flexural strength average for all the samples investigatedb. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

Figure 3.10 shows the values of flexural strength for geopolymers made only with fly ash and geopolymers containing red mud or waste glass. The effect of the glass or red mud addition on the flexural strength is quite different. The incorporation of waste glass has a negative effect on the flexural strength which decreases as the amount of waste glass increases. On the contrary, the addition of red mud provides an improvement in flexural strength from 11 ± 2 MPa to roughly 15 ± 2 MPa related to fly ash-only geopolymers. Moreover, a high value of flexural strength is maintained nearly constant for high content of red mud up to (30 wt.%).

3.3.4 Fracture toughness

The chevron notch test by using the critical stress intensity factor (KIC) allows to determine the fracture toughness that indicates the level of stress concentration at the crack tip needed for crack initiation from the notch174.

KIC was calculated for 10 samples of each batch using equation (3.2). The notch depth, necessary for the calculation, was measured using an image analysis software

b The tests were carried out at the Institute of Physics of Materials in Brno (Head: Prof. I. Dlouhy).

Chapter 3

from optical microscope pictures. Examples of macrographs of fracture surfaces are shown in Figure 3.11.

0.82 mm 0.88 mm

Figure 3.11: Notch depth determined in samples containing 20 wt.% of waste glass (left) and 10% of red mud (right). Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

The average fracture toughness values for each mixture are summarized in Figure

3.12. For each sample the KIC result is the average between the two values calculated from the two parts of the samples obtained after rupture. For most of the samples the

KIC of the right part equaled the one of the left part of the specimen. Only for few samples there was a small variation in the value, probably due to inhomogeneous voids or inaccuracy during measurements. The statistical test made using ten samples for each composition demonstrated a good consistency in KIC values with low variability. This result indicates a high reproducibility of the measurements. The low scattering of KIC values might have been caused by the fact that samples exhibit the some porosity and microstructural inhomogeneity.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

0,4 Fly ash Glass

Red mud

) 1/2

0,3

(Pa*m

IC K

0,2 0 5 10 15 20 25 30 Concentration of waste material (wt. %)

Figure 3.12: Mean fracture toughness values of the different geopolymers investigatedc. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

The fracture toughness does not change significantly by substituting a part of fly ash for soda lime glass or red mud. From the figure, it is evident that all values are lying in the range between 0.25 and 0.35 MPa*m1/2. However, increasing soda lime glass content seems to reduce to a higher extent the rate of deformation energy dissipation, while the introduction of red mud does not affect the fracture toughness. This result can be explained considering that an increase in the waste glass content could result in a sample affected by more defects and cracks. The distribution of glass particles in the matrix could induce a crack deflection mechanism but this effect was probably insufficient to balance the negative effect caused by porosity and cracks already present in the matrix. Moreover, dissolution and polycondensation, the main phases in geopolymerization, are heterogeneous processes, for this reason it is difficult to control the final uniformity of the matrix175. Geopolymers incorporating red mud are thus expected to exhibit a more homogeneous final structure that can explain the improvement in fracture toughness.

c The tests were carried out at the Institute of Physics of Materials in Brno (Head: Prof. I. Dlouhy).

Chapter 3

3.3.5 Micro hardness Hardness is one of the most frequently measured properties of ceramic materials. The square-based pyramidal indenter, in Vickers indentation, creates a small deep impression that allows to determine the hardness.

Figure 3.13: Vickers indentation in the geopolymer sample 10SLG (left) and load-indentation curves (right)d.

In Figure 3.13 right, typical load-indentation depth curves on a geopolymer samples 10SLG are presented, which in agreement with the report of Fischer-Cripps et al.176, are usually observed in brittle materials. The hardness of ceramic materials or more generally speaking of brittle materials is difficult to test and in some cases, the indentation realized in the material can not be measured, unless very low indentation loads are used, especially in the case of materials containing porosity177. This was probably the problem encountered in the measurement of geopolymers hardness. The indentation shape, on the present sample, as shown in Figure 3.13 left, was difficult to measure with the microscope, and consequently the hardness of the materials could not be determined accurately. From the picture, it can be seen that the sample presents microstructural defects, in particular voids that affect the indentation shape especially when this is made above the defect.

d The tests were carried out at the Institute of Physics of Materials in Brno (Head: Prof. I. Dlouhy).

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

For this reason measuring the size of the square diagonal of the indentation was difficult and so it was to obtain accurate values of the hardness. The measurements for all the batches show important data scattering, with values in the range between 30 and 50 HV0.2. A positive factor is that after indentation the geopolymer material does not show crack or damage, as often occurs when ceramic materials are tested. It can be concluded that further tests are necessary in order to increase the reliability of the measurement. Increasing the load applied could make the indentation more visible and consequently the measurement of the diagonal would be more precise, but this could create damage in the structure.

3.3.6 Confocal microscopy

20 µm

Figure 3.14: Confocal microscope images for samples 30SLG (a and b) and 30RM (c and d).

Images of geopolymer sample surfaces after rupture (30SLG and 30RM) obtained with the confocal microscope are reported in Figure 3.14. In the images a) and c), of both samples, big bubbles are visible. These could be air entrapped during the preparation of the slurry and not released during the curing process. Since cracks propagate preferentially through inhomogeneities and defects of the microstructure

Chapter 3

the presence of such large pores is undesired as they weaken the fracture resistance of the material. In images b) and d) the surface of fracture propagation, where the chevron notch was carved. The surface of the sample incorporating waste glass (sample 30SLG, Figure 3.14 b) appears rather homogeneous whereas the surface of the RM sample is more irregular with different surface levels, indicating possible changes in crack path and velocity. If the fracture surface is irregular and rough, it indicates that the crack deflection toughening mechanism probably took place during the fracture of the material 178.

The roughness of the fractured area (Ra arithmetic average) was determined with the confocal microscope and is represented in Figure 3.15 for all the samples.

5,0 Glass Red mud 4,5

4,0

3,5

3,0

Roughness Number Roughness 2,5

2,0

1,5 10 20 30 40 Concentration of waste material (wt.%)

Figure 3.15: Average roughness number for all specimense.

It can be seen that samples incorporating red mud have an higher roughness in comparison with the samples incorporating waste glass. By comparing these results with the results obtained from the fracture toughness test (Figure 3.12) it can be seen that as the fracture surface roughness increases, KIC increases. Since fracture involves the propagation of the crack through a stressed material, the magnitude of fracture surface area and the degree of mechanical interlock are important factors that affect the energy required for fracture179. For this reason, when the fracture

e The tests were carried out at the Institute of Physics of Materials in Brno (Head: Prof. I. Dlouhy).

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study surface of the geopolymer material is more tortuous and rugged, it is reasonable to expect KIC to be higher. Summarizing the results obtained from the samples incorporating red mud, it can be concluded that the latter presents a higher level of roughness that could be correlated with a difficult path for the crack to propagate, resulting in samples with higher resistance to fracture propagation in comparison with samples incorporating waste glass.

3.3.7 Scanning electron microscopy (SEM)

The microstructure of the geopolymers was evaluated by scanning electron microscopy (SEM). Figure 3.16 shows the SEM images, at the same magnification, for specimens containing fly ash plus 10, 20 or 30 wt.% of red mud or soda lime glass.

Waste Glass Red Mud

10wt.%

20μm 20μm

20wt.%

20μm 20μm

30wt.%

20μm 20μm

Figure 3.16: SEM images of geopolymer fracture surfaces after compression strength tests with different replacement level of recycled glass or red mud. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol.

Chapter 3

In all the microstructures spherical particles of variable dimensions are visible, these are unreacted fly ash particles that did not take part in the geopolymerization. On the other hand when these spherical particles dissolve, they create porosity in the matrix, as shown in Figure 3.16. The spaces left after fly ash dissolution and water evaporation during the curing process are responsible for the porosity of the geopolymer material. The micrographs of the geopolymers containing waste glass present unreacted glass fragments and fly ash in the likely amorphous geopolymer matrix. In particular, by increasing the amount of glass an increase of relatively large cracks and pores is produced, larger than the one produced after red mud addition; particularly at 30% loading. The addition of glass requires more water from the system compared to red mud, as a consequence geopolymers incorporating waste glass present a higher shrinkage with subsequent reduced interface adhesion between the initial raw phases. The amount of dissolved silica and alumina, as mentioned before, correlates with the amorphous fraction in the raw materials. It is indeed not expected that all glass particles used as raw material dissolve when taking part in the geopolymerization process. This is confirmed by the glass fragments visible in the microstructure (Figure 3.16). Moreover, by increasing the content of glass the amount of unreacted silica should increase, in particular if the reaction product covers the glass particles, making it more difficult for them to dissolve inside the alkali solution. Particle size is a key factor in glass solubility, as has been extensively documented in literature. Carrasco71 and Cry64 et al. tested the influence of glass granulometry in geopolymerization. The results show that the lower the particle size range of the glass, the higher the reactivity and consequently the compressive strength. In particular, reducing the fineness from 75 µm to 15 µm for the glass average grain size leads to an increase in the mechanical strength up to five times the original value. The solubility could also be improved by decreasing the particle size of the glass but this implies adding a pretreatment before using the raw materials. For economic and environmental reasons soda lime waste glass and red mud are used directly as received by waste processing facilities.

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

The SEM images reveal a homogeneous, dense structure for geopolymer specimens incorporating red mud. This is consistent with the optimal measured compressive strength of the samples (Figure 3.8). Red mud consists of particles that have a good capacity to assemble together filling the spaces left by fly ash particles dissolution to obtain a compact final microstructure. Micrographs in Figure 3.16 indicate that increasing the amount of red mud does not affect the mesostructural homogeneity of the final geopolymer.

3.3.8 IR Spectroscopy

SLG10

SLG20

SLG30

RM10

RM20

RM30

3460

1430

1645

555

783

453 1018

4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1)

Figure 3.17: FTIR spectra of fly ash-based geopolymers with 10, 20 and 30% addition of waste glass or red mud. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol. (The relevant peaks are discussed in the manuscript).

The FTIR spectra in Figure 3.17 show the same trend for mixtures containing red mud and recycled glass. Bands at 453 cm-1 and 455 cm-1 belong to Si-O and Si-O-Si

Chapter 3

bending vibration respectively7. The small bands between 790 and 800 cm-1 are attributed to the bands present in the fly ash source as quartz or mullite173. The main absorption band is located between 900 and 1100 cm-1, which is attributed to the Si- O-Al and Si-O-Si asymmetric stretching vibrations. The band at 555 cm-1 is assigned to Si-O-Al bending vibration; this band is more intense for geopolymers containing red mud, thanks to the high amount of alumina that facilitates its formation. The band identified at 1430 cm-1 can be related to O-C-O stretching vibration of a carbonate phase, which might be formed from remained unreacted activator and 69 -1 CO2 . Bands at 3390-3370 cm are related to –OH, H-O-H bonds stretching vibration, while the band between 1600 cm-1 and 1650 cm-1 corresponds to H-O-H bending vibration. Since these bands are generated by water molecules, they are indicators of the hydration of the geopolymer53. Figure 3.17 shows how the incorporation of waste glass and red mud did not inhibit the formation of an aluminosilicate gel, but only changed the intensity of the bands related to the Si and Al vibration.

3.3.9 Leaching test A leaching test following EN 12457-2 Normative was carried out on all samples to ascertain their durability.

Table 3.6: Leaching test results from raw materials and geopolymer samples. Thresholds for inert and non-hazardous material from reference180 Element (ppm) As Cd Cr Cu Mo Pb Se Zn

10RM 0,76 <0.002 0,028 0,0072 0,666 0,005 0,0017 <0.203

20RM 0,9488 <0.002 0,065 0,0032 0,667 0,004 0,2016 <0.203

30RM 0,919 <0.002 0,0424 0,0094 0,4903 0,0048 0,1664 <0.203

10SLG 1,25 <0.002 0,011 0,002 0,958 0,006 0,276 <0.203

20SLG 1,021 <0.002 0,0064 0,0017 0,657 0,009 0,226 <0.203

30SLG 0,774 <0.002 0,0043 0,001 0,383 0,008 0,145 <0.203

FA <0.049 <0.002 0,467 0,028 0,898 <0.047 0,221 <0.203

RM 0,015 0,008 0,599 0,027 1,021 0,13 0,049 <0.203

SLG <0.049 0,001 0,043 0,036 0,007 0,018 0,018 <0.203

Inert material 0,05 0,04 0,5 2 0,5 0,5 0,1 4 Non- hazardous 2 1 10 50 10 10 0,5 50 material

Synthesis, characterization and mechanical properties of fly ash-based geopolymers incorporating red mud or waste glass: a comparative study

From Table 3.6 it can be observed that all the samples exceed the thresholds for the relevant elements for inert materials but are below the limitations for non-hazardous materials. In particular As, Mo and Se are the most difficult heavy metals to stabilize into the geopolymeric matrix. Se is abundant in fly ash and Mo coming from red mud, they were partially stabilized after geopolymerization. In particular, the introduction of red mud and soda lime waste glass in the geopolymer formulation generally reduces the leaching of such heavy metals. Samples incorporating low amounts of red mud or waste glass are more prone to ion release in comparison with samples incorporating more by-products. The values below the threshold for non- hazardous materials confirm the good capacity of geopolymers to stabilize heavy ions in the matrix also when incorporating different types of waste materials.

3.4 Conclusions

In this chapter the production and characterization of geopolymers based on the combination of different wastes such as fly ash, red mud and waste glass, are presented. The study investigated the different properties obtained by incorporating red mud or waste glass in a fly ash- based geopolymer, with a particular focus on the variations in the mechanical behavior. The introduction of recycled glass and red mud causes a decrease in the mechanical properties compared with the geopolymer synthesized only with fly ash. Despite this, the performances obtained with the incorporation of waste glass or red mud are satisfactory for many fields of application. Summing up the results from mechanical tests and microstructural analysis, it can be concluded that the loss of compressive strength verified in geopolymers created with waste glass is related with the presence of remaining unreacted glass particles and micro-crack and porosity development. The replacement of fly ash with red mud implied a negligible decrease in compressive strength while results of bending test showed no substantial differences originated by the substitution of fly ash with the other by-products. Furthermore, red mud addition seems to have better influence on fracture toughness and flexural strength compared to waste glass incorporation.

Chapter 3

Comparing mechanical testing results with microstructure images leads to conclude that the loss of compressive strength after inclusion of glass correlates with the development of microcracks. From this study it can be concluded that fly ash in geopolymer synthesis can be used in combination with waste glass or red mud without drastically modifying the final properties. Moreover the properties achieved are within the ranges required in construction applications, thus the present approach represents a new end-use for these waste products.

Chapter 4 Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

4.1 Introduction

Geopolymers were primarily developed for the construction industry thanks to their good mechanical and chemical properties and their low carbon dioxide emissions compared to Portland cement. Geopolymers are mainly produced using metakaolin as aluminosilicate source dissolved into an alkaline solution usually composed of solutions of sodium hydroxide and sodium silicate. As introduced in chapter 2, geopolymers have gained interest in different fields of application, such as ceramic components, high-tech materials, fire-resistant materials, etc181. Moreover, due to the enhancement of European regulations on waste discharge and CO2 emissions, geopolymers are becoming promising materials thanks to the possibility of being fabricated with different types of industrial wastes. In fact geopolymers can be produced using different sources of alumina and silica such as fly ash, red mud, waste glass, bottom ash, etc. Thus geopolymer technology enables waste materials currently sent to landfill to be converted in valuable products. The present study is essentially aimed at evaluating the feasibility of using waste glass instead of sodium silicate, also called water glass, normally used as alkaline activator. More in detail, soda-lime glass residue coming from urban waste collection in Italy was used, in particular the glass fraction in which plastic and ceramic impurities were concentrated. This glass cullet with a size under 100 µm could generate foaming problems in an industrial furnace, for this reason it is currently landfilled. Replacing sodium silicate with waste glass is an innovative approach that could represent an innovative eco-friendly solution because it allows not only to avoid the use of an expensive chemical reagent but also to incorporate in a valuable product a waste material with disposal difficulties.

Chapter 4

Tchakoute et al.25 and Torres-Carrasco et al.72 have already published studies dealing with the replacement of sodium silicate with waste glass. In these cases a “waste glass sodium silicate solution” was obtained by mixing sodium hydroxide solution with waste glass and subsequently filtering the solution. The aim of this work is to recycle waste glass not only by producing a silicate solution but also by incorporating the glass in the geopolymer network, which would represent an innovative way to synthesize eco-friendly geopolymers. The influence that the addition of waste glass has on the strength and microstructure of fly ash based-geopolymers was investigated. Leaching analyses were also carried out to examine the capacity of stabilizing heavy metal ions in the geopolymer matrix. Moreover, XRD analyses were carried out to detect the crystalline zeolite phases formed after geopolymerization, while infrared spectroscopy was used to identify the characteristic bands of geopolymers. In addition to X-ray diffraction and FTIR spectroscopy, magic-angle spinning MAS-NMR spectroscopy provided useful structural data to prove the formation of silico-aluminate geopolymers. This work was carried out in collaboration with the University of Padova (Italy) in the framework of CoACH EU project182. The results presented in this chapter are part of previous publication.

4.2 Materials and methods

Low calcium fly ash (FA) class F (ASTM C 618), provided by Steag Power Minerals (Gladbeck, Germany), with a mean particle size of 20 µm, and soda-lime waste glass (SLG), coming from municipal waste collection and provided by the company SASIL S. p. a. (Brusnengo, Biella, Italy) in the form of fine powders with particle size < 30 µm, were used as raw materials.

Table 4.1: Chemical composition of raw materials determined by XRF. wt% SiO2 Al2O3 Na2O K2O CaO MgO Fe2O3 TiO2 FA 54.36 24.84 0.83 3.03 2.56 2.06 8.28 1.07 SLG 70.5 3.2 12 1 10 2.3 0.42 0.07

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Table 4.1 summarises the chemical composition of fly ash and waste glass determined by means of X-ray fluorescence (XRF, Spectro Analytical Instruments GmbH with a Spectro Xepos He energy-dispersive X-ray fluorescence spectrometer). The alkaline activator was a sodium hydroxide solution, prepared at 3, 5, 8, 10 and 12 M, using sodium hydroxide flakes (Merck 99.5%) dissolved in distilled water. The alkaline solutions were prepared one day before the use to allow the exothermically heated solution to cool down to room temperature.

4.2.1 Geopolymers preparation

In order to assess the possibility of using recycled waste glass instead of water glass in geopolymers preparation, a systematic study was carried out by modifying the molarity of NaOH solution and the theoretical molar ratio between SiO2 and Al2O3.

The initial mixtures were produced by fixing the SiO2 /Al2O3 theoretical molar ratio of the final geopolymers at 5, 6 and 7, which represent mixtures of FA/SLG in wt.% of 76/24, 64/36, 54/46 respectively. NaOH aqueous solutions with molarity 3, 5, 8, 10 and 12 were used as alkali activators in all the mixtures, while the liquid solid ratio was fixed at 0.45 to obtain a good workability. The samples label for further characterization has been established to be xSyM where “x” is associated with SiO2

/Al2O3 molar ratio, while “y” refers to the molarity of the activating solution. Table 4.2 reports the composition of the geopolymers synthesised.

61 Chapter 4

Table 4.2: Summary of mixture proportion used in the experimental trial. Alkaline sol./ Designation FA (wt.%) SLG (wt.%) NaOH (M) raw materials 5S3M 76 24 3 0.45 6S3M 64 36 3 0.45 7S3M 54 46 3 0.45 5S5M 76 24 5 0.45 6S5M 64 36 5 0.45 7S5M 54 46 5 0.45 5S8M 76 24 8 0.45 6S8M 64 36 8 0.45 7S8M 54 46 8 0.45 5S10M 76 24 10 0.45 6S10M 64 36 10 0.45 7S10M 54 46 10 0.45 5S12M 76 24 12 0.45 6S12M 64 36 12 0.45 7S12M 54 46 12 0.45

To synthesize the geopolymer, the raw materials were kept under mechanical stirring for 4 hours at 2000 rpm in order to obtain a homogeneous slurry of partially dissolved glass powders and activated fly ash. The mixture was cast into polyethylene-sealed moulds with diameter of 12 mm and height of 20 mm. The geopolymers were than vibrated for 5 minutes to remove trapped air before curing in an oven at 60 °C for 48 h to complete the polycondensation reaction. The schematic geopolymer preparation and final samples are shown in Figure 4.1 and Figure 4.2, respectively.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Aluminosilicate source Alkaline solution

Fly ash Waste glass NaOH Water

Raw materials

Mixing for 4h at 2000rpm

Casting in polyethylene moulds

Vibrating with a vortex for 5 min.

Sealing

Curing (60°C 48h)

Geopolymers

Figure 4.1: Scheme for the preparation of geopolymers samples of composition shown in Table 4.2.

63 Chapter 4

SiO2/Al2O3 (molar ratio) 7 6 5 10

[mol/L] 8

5 Molarity

3 NaOH 12 mm

Figure 4.2: Geopolymers prepared with different SiO2/Al2O3 molar ratio and NaOH molarity.

4.2.2 Characterization technique

Fourier transform infrared spectroscopy Fourier transform infrared spectroscopy (FTIR) studies were performed on powder by using a Shimadzu model IRAffinity-1S. The spectra were collected in transmittance mode with a resolution index of 4 and 24 numbers of scan.

X-Ray diffraction (XRD) The mineralogical analysis was conducted by X-Ray diffraction (XRD) on powdered samples (Bruker D8 Advance, Karlsruhe, Germany – CuKα radiation, 0.15418 mm, 40 kV-40mA, 2Ɵ =10-70°, step size 0.05°, 2s counting time). The phase Identification was performed by means of the Match!® program package (Crystal Impact GbR, Bonn, Germany), supported by data from PDF-2 database (ICDD- International Centre for Diffraction Data, Newtown Square, PA).

Solid-State Magic Angle Spinning NMR

The solid-state Magic Angle Spinning (MAS) NMR spectra were recorded on an Agilent DD2 500WB spectrometer equipped with a commercial 3.2 mm triple

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass resonance MAS probe at 27Al and 29Si resonance frequencies of 130.24, and 99.30 MHz, respectively. The one-dimensional (1D) 27Al MAS spectra were obtained at a sample spinning frequency of 15 kHz with a typical 90° pulse lengths for the 27Al central transition 1.625 us and recycle delays of 1.0 s. The chemical shifts of 27Al is referenced to a 1.1 mol/kg solution of Al(NO)3 in D2O on a deshielding scale. Saturation combs were applied prior to all repetition delays. Total 2000 numbers of scans were acquired for 27Al.

Density and porosity calculation

The geometric or bulk density (ρb) was evaluated by considering the mass to volume ratio of samples. The apparent (ρa) and the true density (ρt) were measured by using a helium pycnometer (Micromeritics AccuPyc 1330, Norcross, GA), operating on bulk or on finely crushed samples, respectively. The three density values were used to compute the amounts of open (OP), closed (CP) and total porosity (TP) with the following formulas:

Compressive strength The compressive strength of cylindrical samples was measured by using a universal testing machine (Zwick Roell, Ulm, Germany Series Z050) with a load speed of 0.5 mm/min. A minimum of 10 samples for each batch was tested to evaluate the 28 days strength of the specimens.

Scanning Electron Microscopy (SEM)- Energy dispersive X-ray spectroscopy(EDS) A qualitative morphological evaluation of the raw materials and final geopolymers was performed via scanning electron microscopy (LEO 435, LEO Electron Microscopy Ltd., Cambridge, UK and Ultra Plus, Zeiss, Jena, Germany) in order to assess the homogeneity of the geopolymers. Pieces of samples after the compression test (fracture surfaces) were used to analyse the microstructure. The EDS analyses were done at a fix working distance of 6 mm and a voltage up to 20kV.

65 Chapter 4

Inductively coupled plasma (ICP) The release of heavy metals was evaluated according to European Standard for waste toxicity evaluation (EN 12457-2). Fragments under 4 mm were placed in an extraction solution consisting of distilled water, with a pH value of ~7, for a liquid/solid ratio of 10, and softly stirred at 25 °C for 24h. The resulting solutions were filtered through a 0.6 µm filter and analysed using inductively coupled plasma (ICP; SPECTRO Analytical instruments GmbH, Kleve, Germany). Samples presenting the best overall mechanical properties were subjected to the leaching test.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

4.3 Results discussion

4.3.1 Fourier Transform Infrared Spectroscopy

Glass

Fly ash

5S5M 5S8M

5S10M

6S5M

6S8M 6S10M

Trasmittance(%) 7S5M

7S8M

7S10M

O-H C-O Al-O-Si Si-O-Si Si-O Al-O

4000 3500 3000 2500 2000 1500 1000 500 -1 Wavenumber (cm ) Figure 4.3: FTIR spectra of raw materials and geopolymers (compositions shown in Table 4.2). The relevant peaks are discussed in the manuscript.

FTIR spectra of raw materials and geopolymers with different SiO2/Al2O3 molar ratio and NaOH molarity are represented in Figure 4.3. Bands between 3000 and 3700 cm-1 are associated to stretching vibration of O-H groups, while those between 1600 cm-1 and 1650 cm-1 are associated with O-H bending183. As visible in Figure 4.3 these bands increase intensity with higher morality, as a further dissolution of the initial materials leads to the formation of more hydrated compounds. The presence of carbonate traces at 1515 cm-1 is identified by the C-O bond stretching184. Bands at 420 cm-1 and 690 cm-1 are connected to Si-O-Si bending vibrations and symmetric stretching185. The bands

67 Chapter 4

detected between 1200 cm-1 and 900 cm-1 are associated with the Si–O-Si asymmetric vibration and Al-O-Si symmetric bending vibration53,186. The geopolymerization reaction could be followed by the shift and the increase of intensity of the Si-O stretching vibration.

SLG

6S5M 6S8M 6S10M

1019 cm-1

1034 cm-1 Trasmittance (%) Trasmittance

963 cm-1 961 cm-1 960 cm-1

1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 Wavenumber (cm-1)

Figure 4.4:FTIR spectrum with focus placed on wavenumbers between 400 and 1500 cm-1. (The peaks are discussed in the text).

Figure 4.4 focuses on this band for the initial raw materials and selected samples with SiO2/Al2O3 molar ratio of 6 at different molarities. The waste soda lime glass (SLG) broad signal becomes narrower in the geopolymeric sample due to the more ordered structure, while the fly ash (FA) Si-O stretching vibration initially situated at 1025 cm-1 shifts to lower frequencies187. This shift indicates the activation of the fly ash amorphous phase with the formation of the aluminosilicate gel. Torres Carrasco et al.72 proposed that the band shifts to lower frequency due to the rise in tetrahedral aluminum content. For all the geopolymer samples this band is detected at around 960 cm-1. The intensity of the band is attributed to the amount of Si-O or Al-O in tetrahedral configuration53, wider bands represent a more disordered structure. As the molarity increases this band becomes narrower as a result of a short-range ordering that occurs due to geopolymerization.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

4.3.2 X-ray diffraction analysis

Quartz Mullite Cancrinite Sodalite

7S12M

7S10M

7S8M Intensity (a.u.) Intensity

7S5M

7S3M

Fly ash

Glass

10 15 20 25 30 35 40 45 50 55 60

2 (deg.)

Figure 4.5: XRD patterns of raw materials and geopolymers with SiO2/Al2O3 molar ratio of 7 at different molarity.

X-ray powder diffraction patterns for raw materials and geopolymeric samples, with different molarity, but the same SiO2/Al2O3 molar ratio, are shown in Figure 4.5. A diffuse halo between 2Ɵ 25° and 45° appears in all XRD diffractograms; the shift of the halo to higher angles 2Ɵ in the geopolymer samples is indicative of the geopolymeric reaction between fly ash and recycled glass67.

69 Chapter 4

The crystalline phases, quartz (SiO2, PDF#086-1560) and mullite (Al4SiO8, PDF#073-1389), detected in the original fly ash, remained apparently unaltered with activation. This confirms that, during the geopolymerization synthesis the amorphous part is the most reactive phase involved in the process, whereas the crystalline phases remain basically unchanged. The presence of zeolitic phases, such as sodalite and cancrite, identified in the geopolymer samples, demonstrates the formation of an alumino-silicate 3D structure. The intensity of the peaks for these crystalline phases was seen to increase with the molarity of the alkaline solution. The samples 7S3M and 7S5M present very low intensity for cancrite and sodalite, while increasing the molarity from 8 to 12 more intense and defined peaks are visible. This is more evident in particular for the peaks situated at 14 and 24 deg. A higher alkalinity of the solution promotes more silica dissolution from the waste glass during the geopolymer synthesis, inducing enhanced zeolite formation. The same trend was detected also for the mixtures at SiO2/Al2O3 molar ratio of 5 and 6, hereby not represented. Moreover, increasing the SiO2/Al2O3 molar ratio at constant molarity also caused a partial increase of zeolite crystalline phases thanks to the higher amount of dissolved silica, as visible in Figure 4.6.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Quartz Mullite Cancrinite Sodalite

7S8M

6S8M Intensity (a.u.) Intensity

5S8M

Fly ash Glass

10 15 20 25 30 35 40 45 50 55 60 2 (deg.)

Figure 4.6: XRD patterns of raw materials and geopolymers with 8 molarity and different SiO2/Al2O3 molar ratio.

4.3.3 MAS-NMR technique

Solid state magic angle spinning nuclear magnetic resonance (MAS-NMR) was performed on the same geopolymer samples to understand the structure, composition and connectivity of the aluminum and silicon species in the geopolymer. Davidovits7 has underlined the importance of this technique to prove the formation of a geopolymeric structure. He observed that the Al(V) and Al(VI) converted mostly into Al(IV) after geopolymerization. 27 Figure 4.7 represents the Al spectra of the samples with the same SiO2/Al2O3 molar ratio and different NaOH molarity, while Figure 4.8 shows 27Al spectra with different molarity but same SiO2/Al2O3 molar ratio.

71 Chapter 4

59 7S10M

58 7S8M

56 7S5M

54 Glass

FA 150 100 50 0 -50 Al chemical shift (ppm)

27 f Figure 4.7: Al MAS-NMR spectra of raw materials and geopolymers with 7 SiO2/Al2O3 molar ratio .

f The analyses have been done at the Erlangen Catalysis Resource Center in collaboration with Dr. Y.S. Avadhut.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

60 7S8M

59 6S8M

56 5S8M

54 Glass

FA 150 100 50 0 -50 Al chemical shift (ppm)

Figure 4.8: 27Al MAS-NMR spectra of raw materials and geopolymers synthesized with 8 molarity NaOH solutionf.

All the geopolymer samples exhibit a dominant 27Al peak situated between 58 and 60 ppm confirming the tetrahedral structure of alumina in the geopolymer material. 27Al NMR peak at about 2 ppm, corresponding to Al(VI), more pronounced in fly ash, is weaker in geopolymer samples indicating that mostly all Al is (IV) coordinated. Moreover, peaks for fly ash and glass are very broad due to the disorder of the structure. After geopolymerization peaks become noticeably narrower and sharper, indicating a higher degree of polymerization and structural order, similar to peaks that are associated with tetrahedral aluminium zeolite188.

73 Chapter 4

4.3.4 Density and Porosity

The density of geopolymers is an important factor in the determination of porosity, assessment of durability and strength. The density of the geopolymer samples and the related porosity values are reported in Table 4.3 and Table 4.4, respectivelyg.

Table 4.3: Density of geopolymer samples after 28-day ageing. Density (g/cm3)

Bulk [ρb] Apparent [ρa] True [ρt] 5S3M 1.35± 0.01 2.23± 0.01 2.41 ± 0.01

5S5M 1.52± 0.01 2.27 ± 0.01 2.42±0.01

5S8M 1.63 ± 0.01 2.19 ± 0.01 2.33 ± 0.01

5S10M 1.55 ± 0.01 2.20 ± 0.06 2.37 ± 0.01

6S3M 1.21 ± 0.01 2.31 ± 0.01 2.44 ± 0.05

6S5M 1.51 ± 0.01 2.25 ± 0.01 2.39 ± 0.01

6S8M 1.93 ± 0.01 2.06 ± 0.01 2.33 ± 0.04

6S10M 1.18 ± 0.04 2.25 ± 0.06 2.32 ± 0.01

7S3M 1.36 ± 0.01 2.26 ± 0.01 2.47 ± 0.01

7S5M 1.55 ± 0.02 2.30 ± 0.09 2.38 ± 0.01

7S8M 1.80 ± 0.01 2.28 ± 0.01 2.35 ± 0.01

7S10M 1.78 ± 0.01 2.33 ± 0.03 2.36 ± 0.01

g The density and porosity analysis has been determined by Acacio Rincon at the University of Padova.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Table 4.4: Porosity of the geopolymer samples after 28-day ageing. Porosity (%)

Total porosity [TP] Open porosity [OP] Closed porosity [CP]

5S3M 44 ± 8 39 ±1 4.5 ± 0.6

5S5M 37 ± 3 33 ± 2 4.1 ± 0.4

5S8M 30 ± 1 26 ± 1 4.2 ± 0.7

5S10M 21 ± 2 16 ± 3 6 ± 0.3

6S3M 50 ± 2 48 ± 4 2.7 ± 0.2

6S5M 36 ± 1 32 ± 2 4.2 ± 0.3

6S8M 17 ± 1 6.3 ± 0.2 11.1 ± 0.3

6S10M 22 ± 2 19 ± 5 2.7 ± 3

7S3M 45 ± 1 40 ±1 5.1 ± 0.1

7S5M 34 ± 1 31 ± 6 3 ± 5

7S8M 23 ± 6 21 ± 6 2.4 ± 0.01

7S10M 24 ± 5 23 ± 9 1 ± 1 The inherent porosity arises from the mesoporous nature of the geopolymeric gel, the unreacted fly ash particles or entrapped air189. All synthesized samples show high density and low porosity, roughly between 20 and 35%, in agreement with values found in literature190. In comparison with density values of Portland cement, usually between 2200-2550 kg/m3 191, geopolymers present a reduced density, indicating a good geopolymerization of the final material. We may observe that samples activated with low molarity present the highest total porosity values as a result of poor reaction of the initial materials and lack of cohesion between the unreacted particles, as the geopolymeric gel is not produced in a sufficient extension. This trend is more visible in Figure 4.9.

75 Chart Title Chapter 4

5 SiO2/Al2O3 0,60 6 SiO2/Al2O3 7 SiO2/Al2O3 0,50

0,40

0,30

0,20 Porosity values Porosity

0,10

0,00 φtotal φopen φclose φtotal φopen φclose φtotal φopen φclose φtotal φopen φclose 3M 5M 8M 10M

Figure 4.9: Histogram representation of geopolymers porosity for different SiO2/Al2O3 ratios and different molarities.

The increase in the SiO2/Al2O3 molar ratio causes the total porosity to decrease; this could be seen as an indicator of enhanced geopolymerization. These results are in agreement with the slight improvement of the mechanical properties following the increase of molarity and amount of glass in the mixture. The optimum molarity was found to be equal to 8 with a reduction of porosity of 17% and the best mechanical performance due to the formation of an almost flawless geopolymer structure.

4.3.5 Compressive strength

Figure 4.10 shows the compressive strength values of geopolymer samples after 7 days obtained by changing NaOH solution molarity and silica to alumina molar ratio. The dashed line shows the evolution trends depending on the molarity of the sodium hydroxide solution.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

120 5S 6S 100 7S

20 Compression strength [MPa] strength Compression

2 3 4 5 6 7 8 9 10 11 12 13 Molarity

Figure 4.10: Compressive strength of geopolymer samples of different molarity and SiO2/Al2O3 molar ratio after 7 days.

Both NaOH molarity and SiO2/Al2O3 molar ratio influence the mechanical properties of the final geopolymers. The compressive strength tends to rise as the molarity of the solution as well as the SiO2/Al2O3 molar ratio increase.

As the SiO2/Al2O3 molar ratio of the mixtures increases, a higher replacement level of waste glass is introduced in the samples. When glass starts to dissolve, under alkaline conditions, a silica-rich gel is deposited on the surface of the unreacted particles, hindering the complete dissolution of the initial waste glass. This phenomenon can be evidently compensated by increasing the amount of glass. From the results represented in Figure 4.10, it is obvious that a 3 molarity NaOH solution is not “aggressive” enough to dissolve the raw materials and to provide samples with sufficient mechanical resistance. NaOH solution at 5M involves also in this case a low dissolution ratio of the raw materials, so that as a consequence a large amount of unreacted particles likely remained in the final material. On the other hand, solutions at 8M and 10M determined an increased dissolution of components leading to more geopolymeric precursors in the initial slurry and less unreacted

77 Chapter 4

particles. We can deduce that stronger mechanical values are achieved with higher molarity, but this is not demonstrated in the case of geopolymers activated with 12M. The use of a more alkaline solution does not increase the dissolution ratio and consequently the mechanical strength. In Figure 4.11 the results of the compressive strength test after 28-day storage are shown.

120 5S 6S 100 7S

20 Compressive strength (MPa) strength Compressive 0

2 3 4 5 6 7 8 9 10 11 12 13 Molarity

Figure 4.11: Compressive strength of geopolymer of different molarity and SiO2/Al2O3 after 28 days.

There are no significant variations between the results obtained after 7 or 28 days of aging. This result confirms the short setting time characteristic of the geopolymer material, different from cement, which needs 28 days to complete the hardening192. Even if there are no important time-related differences in mechanical performance, sample 7S10M involves a strength loss. The value of 100 MPa achieved after 7 days decreases drastically to 36 MPa after a storage time of 28 days. The same behavior was detected by Novais et al.67, who reported that a high alkalinity of the solution could cause the alumino-silicate gel to disintegrate after a long period. This result indicates that the geopolymer network present after 7 days may decompose as an effect of alkali excess. From this result, it can be concluded that a 10 M alkali

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass activation, not necessary to activate the raw materials, is not convenient from both an economic and an environmental point of view. For this reason the samples activated with a molarity of 12 were not analyzed in all the tests. 6S8M and 7S8M samples present outstanding mechanical properties reaching values of 45 MPa, even after a setting time of 7 days. The values remain constant with time. These results suggest that a molarity of 8 is adequate to dissolve almost completely the initial raw materials, leading to a correct ratio between silica and aluminum precursors in the geopolymeric gel. The samples developed in this research possess a compressive strength comparable to that of traditional Portland cement, normally used in the construction industry78. Previous studies have already demonstrated the possibility of incorporating up to 30 wt.% waste glass in the geopolymer network, but with final compressive strength values not exceeding 20 MPa70,193,. Conversely, the results achieved in this research show the possibility of substituting water glass with waste soda lime glass obtaining satisfactory values of compressive strength.

79 Chapter 4

4.3.6 Microstructure

3M 5M

20μm 20μm

8M 10M

20μm 20μm

12M

20μm

Figure 4.12: Microstructure of fractured surfaces of geopolymers with SiO2/Al2O3 molar ratio of 5 and different NaOH molar ratio. Spherical particles are undissolved fly ash particles and particles with angles are unreacted glass particles.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

3M 5M

20μm 20μm

8M 10M

20μm 20μm

12M

20μm

Figure 4.13: Microstructure of fractured surfaces of geopolymers with SiO2/Al2O3 molar ratio of 6 and different NaOH molar ratio. Spherical particles are undissolved fly ash particles and particles with angles are unreacted glass particles.

81 Chapter 4

3M 5M

2200μμm 20μm

8M 10M

20μm 20μm

12M

20μm

Figure 4.14: Microstructure of fractured surfaces of geopolymers with SiO2/Al2O3 molar ratio of 7 and different NaOH molar ratio. Spherical particles are undissolved fly ash particles and particles with angles are unreacted glass particles.

The mechanical properties can be correlated with the microstructure of the geopolymer samples, as shown in Figure 4.12, Figure 4.13 and Figure 4.14. There are significant changes in microstructure with variation in molarity and

SiO2/Al2O3 molar ratio. In all samples we can distinguish spherical particles as well as particles with sharp angles that could be identified respectively as undissolved fly ash and unreacted glass194. These particles appear well distributed in the amorphous geopolymeric matrix. After dissolution in 3M and 5M NaOH solutions, a large amount of unreacted particles remained poorly encapsulated in the geopolymeric gel.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Working with low alkalinity solutions implies a moderate reaction in the mixture so that there was poor adhesion between unreacted or partially reacted particles, and consequently low mechanical strength. In sample 5S8M (Figure 4.12), although activated with a high molarity solution, limited silica precursors were likely involved in the geopolymerisation due to the lower amount of glass in the mixture, resulting in poor homogeneity. On the other hand, samples activated with 8 and 10M NaOH solutions with theoretical SiO2/Al2O3 molar ratios of 6 and 7 (Figure 4.13, Figure 4.14), caused a more efficient dissolution, characterized by largely homogeneous gel containing few unreacted particles and small micro-cracks in the matrix. The formation of a more compact matrix corresponds to a higher reaction degree that consequently results also in improved mechanical properties. Geopolymers synthesised with 10 and 12 molarity solutions, although exhibiting good homogeneity and reduced amount of unreacted particles, exhibited large cracks and pores distributed throughout the surface. This aspect could definitely correlate with the reduction in the mechanical performances verified by increasing the molarity. The decrease in mechanical properties with time (Figure 4.11), experienced by sample 7S10M sample, could be explained by the morphological changes occurring with ageing. Figure 4.15 compares the microstructure evolution from day 7 (Fig. 4.15a) to day 28 (Fig.4.15b).

Figure 4.15: SEM images of 7S10M after 7 (a) and 28 (b) days, showing the sponge-like structure on the left and a more compact structure on the right.

83 Chapter 4

The structure after 7 days, due to the excessive alkalinity, presents a sponge-like gel structure, already described in the literature195,196. The uniformity of the gel could explain the good mechanical properties measured after short time. After 28 days of storage the microstructure is more compact, however it also includes long visible cracks and pieces of unreacted glass. These defects inside the microstructure reasonably caused the strength decrease. These aspects underline again the drawbacks in the use of an excess of alkali in the solution. Figure 4.16 shows the EDX analysis of sample 7S10M. It is clear from the spectra a and b that the sponge-like structure presents an excess of sodium, whose peaks are abundantly higher than the Al ones. Considering that sodium ions should balance the alumina in the structure, a molar ratio between Al and Na of 1 is considered ideal. Dissolving the raw materials in an alkaline solution of 10 or 12 M, as shown by these results, leads to an excessive increase of alkali in the structure that with time could influence negatively the properties.

O a O b Si

a.u. a.u. Na Na

Al Mg Al Ca K Ca Ca Mg K Ca

0 1 2 3 4 5 0 1 2 3 4 5 keV keV

Figure 4.16: EDX spectra of 7S10M, after 7 days, in points a) and b) on the SEM micrograph.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Figure 4.17 presents the EDX analysis of sample 6S8M (featuring the best overall properties), intended to clarify the nature of the developed geopolymeric gel. Spectrum (a), collected from the spherical particle, confirms the attribution to fly ash, according to the Si/Al ratio (in agreement with the initial fly ash composition, shown in Table 4.1). The irregular particle with sharp angles (identified by ‘b’), corresponds to an unreacted glass particle, since the main signals in the EDX spectrum match with the elements in a glass with soda-lime composition. The matrix (area ‘c’) corresponds to amorphous geopolymeric gel: the relative height of Al- and Si-related peaks is consistent with a Si/Al molar ratio in the range between 1 and 3, as expected from the formation of a geopolymer material197. The Na-related peak, nearly as high as the Al-related one, is consistent with a Al/Na molar ratio close to 1, in turn corresponding to the formation of AlO4 tetrahedra, mixed with SiO4 tetrahedra, with Na in a charge balancing role198. Furthermore, the amount of Ca in the gel is limited, so that a limited amount of calcium C-S-H (calcium silicate hydrate) gel could form.

Al Si O O a b c Si

a. u. a. a.u. O a.u.

Na Al

Na Ca K Mg Ca Ca Ca Na Ca Fe K 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 keV KeV keV

Figure 4.17: EDX spectra of 6S8M geopolymer obtained on the points a), b), c) of the SEM image.

85 Chapter 4

4.3.7 Leaching test

As described in the “State of the art” section, one of the most important characteristics of geopolymer samples is the capacity to stabilize heavy metal ions within the polymeric network, although the immobilization mechanism is still under discussion7. The release of heavy metal ions upon leaching tests under legislation limits represents a fundamental requirement for waste-derived marketable products199. Images of the solutions after being stirred for 24 hours are reported in Figure 4.18, Figure 4.19 and Figure 4.20.

Figure 4.18: Water and fragments of geopolymers synthesized with 5M solution, after 24h of stirring.

Figure 4.19: Water and fragments of geopolymers synthesized with 6M solution, after 24h of stirring.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass

Figure 4.20: Water and fragments of geopolymers synthesized with 7M solution, after 24h of stirring.

After a first analysis of the images, it can be observed that the samples with a

SiO2/Al2O3 molar ratio of 5 contain turbid water, whereas the samples with a higher molar ratio contain purer water. At the same time, it is clear from the images that water containing fragments of geopolymers synthesized at 5 M is the darkest and most contaminated water. 6S8M is the sample presenting the best aspect after stirring for 24 hours, in fact the water appears clear and transparent. These deductions, made after observing the images, are confirmed by the heavy metal release results reported in Table 4.5.

87 Chapter 4

Table 4.5: Leaching test results from raw materials and geopolymer samplesh. Element (ppm) As Cd Cr Cu Mo Pb Se Zn

5S5M 0.63 0.005 0.035 0.098 0.86 0.152 0.066 <0.203

5S8M 1.177 0.003 0.0101 0.111 1.071 0.07 0.039 <0.203

5S10M 1.291 0.005 0.0101 0.154 1.187 0.056 0.041 <0.203

6S5M 0.174 0.007 0.0251 0.11 0.363 0.1 0.038 <0.203

6S8M 0.134 0.004 0.0101 0.115 0 0.078 0.048 <0.203

6S10M 1.587 0.08 0.0117 0.295 1.02 0.096 0.059 <0.203

7S5M 1.07 0.007 0.0059 0.271 0.792 0.079 0.044 <0.203

7S8M 0.147 0.008 0.0052 0.197 0.996 0.117 0.048 <0.203

7S10M 1.145 0.019 0.0073 0.188 0.924 0.14 0.044 1.1

FA <0.049 <0.002 0.467 0.028 0.898 <0.047 0.022 <0.2

SLG <0.049 0.0011 0.0434 0.0365 0.0073 0.018 0.018 0.0882

Inert material 0.5 0.04 0.5 2 0.5 0.5 0.1 4 Non-hazardous 2 1 10 50 10 10 0.5 50 material

The content of heavy metals in the initial soda-lime glass does not exceed the thresholds according to the normative EN 12457, while fly ash passes the limits set for a material to be classified as ‘inert’ concerning Mo. In general, samples activated at low molarity are more prone to ion release since a limited amount of geopolymeric gel has formed. The reduced leaching rate found in samples activated with 8 molarity is consistent with the effective formation of a geopolymer matrix that entraps heavy metals in three-dimensional zeolitic-like alumino-silicate ‘cages’. Cr, released in high amount in fly ash, was successfully immobilized in the geopolymer network. Samples activated with 10 M show the highest leaching values in agreement with the previous observations on the depolymerisation caused by alkali excess. It may be concluded that the concentration of alkali activator plays a fundamental role in the efficiency of the immobilization process. The release of toxic elements still exceeds the limits for inert materials in all samples except 6S5M and 6S8M. However, all the values are below the limits for non–hazardous materials. Geopolymers formed in these conditions present thus

h ICP test was carried out at the University of Padova by Dr. Giulia Zanmarchi.

Development and characterization of innovative fly ash-based geopolymers using waste glass instead of chemical water glass excellent chemical stability, confirming the capacity of geopolymer technology to produce a valuable product from waste materials.

4.4 Conclusions

In this study geopolymers incorporating fly ash and soda lime waste glass were developed and characterized. The innovation of this work consists in the capacity of producing waste-based geopolymer materials without using sodium silicate normally used in geopolymer production. Sodium hydroxide was used as the only chemical reagent while fly ash and waste glass were brought to the system silica and alumina. This new approach is an interesting way of recycling a fraction of glass currently landfilled and represents a significant economic advantage as well as an environmentally friendly solution to landfill problems. According to the results obtained, it is possible to realize geopolymers incorporating glass cullet instead of commercial sodium silicate. The mechanical properties increased with the increasing molarity of the activating solution. The same trend was observed with the increase of the SiO2/Al2O3 theoretical molar ratio. The highest compressive strength was performed by the sample incorporating circa 50 wt.% of glass, indicating the relatively high amount of waste glass incorporated in the geopolymers. An excessive molarity of NaOH solution was found to cause strength loss with time and worsening of the geopolymer performance. For this reason, for both an economic and an environmental point of view, it is not convenient to use a solution with a molarity higher than 8. NMR, XRD and EDX analyses indicated the formation of a geopolymeric gel with zeolite crystalline phases. The leaching test, moreover, confirms the capacity of the geopolymers to encapsulate the heavy ions in the matrix. Geopolymers meet the current regulatory requirements for non-hazardous materials, despite the innovative formulation. Summing up, the results confirm the possibility of creating geopolymer materials using industrial waste or aluminosilicate sources and avoiding the use of expensive commercial sodium silicate. Last but not least, the developed geopolymers possess an overall performance comparable to that of traditional Portland cement

Chapter 5 Red mud-based geopolymers synthesized with waste glass instead of water glass

5.1 Introduction

Red mud is the highly alkaline, toxic residue of the aluminium ore bauxite generated during extraction of aluminium through the Bayer process47. According to the website of the International Aluminium Institute, in 2017 Europe produced 16 thousand metric tonnes of alumina200. Considering that the production of 1 tonne of aluminium requires 4.6 tonnes of bauxite198, it is easy to calculate the huge amount of red mud that is produced worldwide. In addition, the aluminium industry is expected to continue to increase due to the large number of industrial applications of Al in transport, construction, packaging, electrical and other sectors. Due to its high basicity and leaching potential, the storage of red mud entails a substantial environmental problem. Construction materials such as cements, lightweight aggregates and geopolymers are considered interesting solutions for red mud safe disposal201. The possibility of producing geopolymer materials using red mud as raw source has been investigated from different authors. Red mud has been combined with electric slag202, fly ash52, metakaolin203,204, rice husk ash30 and ground blast furnace slag19. The compressive strength of the resulting materials is extremely variable, but almost alway in the range of 5-20 MPa. In this study, geopolymers with superior mechanical properties were developed using red mud and waste soda lime glass, in particular the glass fraction where plastic and ceramic impurities are concentrated and therefore is otherwise mostly landfilled. Different formulations using red mud as source of alumina and waste glass as silica supplier were developed, using sodium hydroxide as the only non-waste material. The formation of a homogeneous geopolymeric material is confirmed by solid-state NMR and EDX analysis. Moreover, the stabilization of possible pollutants is proved by leaching test. The results presented in this chapter have been partially publish205.

Chapter 5

5.2 Materials and Methods

Red mud (RM) with a particle size under 75 µm (Alteo, Gardanne, France) and soda lime glass (SLG) with particle size < 30 µm (Sasil s.p.a., Italy) were used as raw materials. The chemical composition of such residues is shown in Table 5.1.

Table 5.1: Chemical composition of the raw materials (wt.%) determined by XRF. wt% SiO2 Al2O3 Na2O K2O CaO MgO Fe2O3 TiO2 RM 5.21 15.21 2.40 0.63 2.95 0.38 52.94 8.05 SLG 70.5 3.2 12 1 10 2.3 0.42 0.07

Sodium hydroxide solutions prepared at 4 and 6 molarity by using sodium hydroxide flakes (Merck 99.5%) dissolved in distilled water were used as alkaline activators. The alkaline solutions were prepared one day before the use to allow the exothermically heated solution to cool down to room temperature.

5.2.1 Geopolymer preparation

The SiO2/Al2O3 theoretical molar ratio of the final geopolymers was fixed at 5, 6 and 7, which represents mixtures of SLG/RM (in wt%) of 40/60, 45/55, 50/50, respectively. The samples designation was established as in chapter 4, by means of the code: xSyM, where “x” is associated with the SiO2/Al2O3 molar ratio, while “y” indicates the molarity of the NaOH solution, as shown in Table 5.2.

Table 5.2: Designation of the geopolymer samples produced with different amount of RM, SLG and different NaOH solution molarity. Alkaline sol./ Designation RM (wt.%) SLG (wt.%) NaOH (M) raw materials 5S4M 60 40 4 0.55/0.50 6S4M 55 45 4 0.55/0.50 7S4M 50 50 4 0.55/0.50 5S6M 60 40 6 0.55/0.50 6S6M 55 45 6 0.55/0.50 7S6M 50 40 6 0.55/0.50

Red mud-based geopolymers synthesized with waste glass instead of water glass

The geopolymers were produced with a ratio of 0.55 and 0.50 between the alkaline solution and the raw materials, in order to determine the best mixture proportion. Geopolymer samples were prepared by mechanically mixing red mud and waste glass in the sodium hydroxide solution at 2000 rpm for 4 hours in order to obtain an homogeneous slurry. The mixture was cast into polyethylene cylindrical moulds and was vibrated for 5 minutes to remove trapped air before curing at 75 °C for 10 days. Typical geopolymer samples are shown in Figure 5.1.

NaOH molarity 4 6

5 ratio

molar 6

/Al 2

SiO 7 1 cm

Figure 5.1: Geopolymers developed with red mud and waste glass with alkali sol. to raw materials ratio of 0.50.

5.2.2 Test and analysis

Compressive strength The compressive strength was measured by using an Instron 1121 UTS (Danvers, MA) testing machine, operating at a cross-head speed of 0.5 mm/min. 10 cylindrical samples, with diameter of 14 mm and height of 20 mm, where tested for each batch.

Density and Porosity The geometric density was evaluated by considering the mass to volume ratio of samples. The apparent and true densities were measured by using a helium pycnometer (Micromeritics AccuPyc 1330, Norcross, GA), operating on bulk or on

Chapter 5

finely crushed samples, respectively. The three density values were used to compute the amounts of open and closed porosity.

Fourier transform infrared spectroscopy (FTIR) FTIR studies were performed on powder by using a Shimadzu model IRAffinity-IS. The spectra were collected in transmittance mode with a resolution index of 4 and 24 numbers of scan.

Scanning electron microscopy The morphological evaluation of the final geopolymers was performed via scanning electron microscopy (LEO 435, LEO Electron Microscopy Ltd., Cambridge, UK and Ultra Plus, Zeiss, Jena, Germany), equipped with EDX. Fracture surfaces of samples after compression test were used to analyse the microstructure.

Solid-state NMR spectroscopy Solid-state Magic Angle Spinning (MAS) NMR spectra were recordered on an Agilent DD2 500WB spectrometer equipped with a commercial 3.2 mm triple resonance MAS probe at 27Al resonance frequency of 130.24 MHz.

Inductively coupled plasma (ICP) The heavy metal release was evaluated according to the European Standard for waste toxicity evaluation (EN 12457-2) and analysed using Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES) (SPECTRO Analytical Instrument GmbH, Kleve, Germany).

5.3 Results and discussion

5.3.1 Compressive strength test

Figure 5.2 shows the compressive strength values of the samples synthesized at 4 and 6 NaOH solution molarity and changing the amount of waste glass in the mixture with consequent change of the SiO2/Al2O3 molar ratio. Geopolymers were realized using a liquid to solid weight ratio of 0.50 and 0.55 to individuate the best mixture proportions. The dashed line in Figure 5.2 shows the evolution trends depending on the molarity of the sodium hydroxide solution.

Red mud-based geopolymers synthesized with waste glass instead of water glass

60 60 5S 0.50 5S 0.55 6S 6S 50 7S 50 7S

40 40

30 30

20 20

10 10

Compression strength [MPa] strength Compression Compression strength [MPa] strength Compression

0 0 3 4 5 6 7 3 4 5 6 7 Molarity Molarity

Figure 5.2: Compressive strength average results as function of NaOH solution molarity and SiO2/Al2O3. On the far left geopolymers synthesized with li./solid ratio of 0.50, while on the far right the liq./solid ratio used was 0.55. Reproduced from ref.205 with permission from Elsevier.

It was found that geopolymers developed with liquid to solid ratio of 0.55 are not affected either when the molarity of the sodium hydroxide solution changes or when the amount of waste glass and red mud varies in the investigated range. All compressive strength values are in the range between 10 and 20 MPa. These results accord with the values found in literature for geopolymers developed using red mud and fly ash as raw materials and NaOH and water glass as alkaline solution27,28. From Figure 5.2 it can be observed that reducing the alkaline solution and consequently the liquid to solid ratio to 0.50 allows to enhance the mechanical properties. In particular, this is more evident for the geopolymers synthesized with 6M NaOH solution. Samples 5S6M and 6S6M present high compressive strength close to 45 MPa. These are extremely interesting values considering the significant amount of red mud incorporated in the mixture. Moreover, a high SiO2/Al2O3 ratio corresponded to an enhanced content of waste glass in the mixture and the evolution of the compressive strength in Figure 5.2 suggested an upper limit to glass cullet incorporation.

5.3.2 Density and Porosity

Table 5.3 and Table 5.4 reported the values of density and porosity after 28 days of ageing.

Chapter 5

Table 5.3: Density and porosity of red mud-based geopolymers with liq./solid ratio of 0.50. Total porosity Open porosity Closed porosity Liq./sol. 0.50 Bulk [ρ ] Apparent [ρ ] True [ρ ] b a t [TP] [OP] [CP] 5S4M 1.65 ± 0.01 2.58 ± 0.01 2.61 ± 0.01 37 ± 1 36 ± 1 1.7 ± 0.3 6S4M 1.68 ± 0.01 2.72 ± 0.01 2.73 ± 0.01 38 ± 1 38 ± 1 1.0 ± 0.2 7S4M 1.65 ± 0.02 2.49 ± 0.01 2.52 ± 0.01 35 ± 2 34 ± 2 1.0 ± 0.3 5S6M 1.69 ± 0.01 2.75 ± 0.02 2.78 ± 0.01 39 ± 2 39 ± 1 1.3 ± 0.3 6S6M 1.71 ± 0.01 2.72 ± 0.03 2.75 ± 0.02 38 ± 1 37 ± 1 1.3 ± 0.5 7S6M 1.78 ± 0.01 2.64 ± 0.02 2.68 ± 0.01 34 ± 2 33 ± 2 1.2 ± 0.4

Table 5.4: Density and porosity of red mud-based geopolymers with liq./solid ratio of 0.55. Total porosity Open porosity Closed porosity Liq./sol. 0.55 Bulk [ρ ] Apparent [ρ ] True [ρ ] b a t [TP] [OP] [CP] 5S4M 1.46 ± 0.03 2.62 ± 0.05 2.65 ± 0.01 45 ± 2 44 ± 1 2.0 ± 0.2 6S4M 1.58 ± 0.01 2.72 ± 0.02 2.75 ± 0.04 43 ± 2 42 ± 1 1.0 ± 0.2 7S4M 1.48 ± 0.02 2.39 ± 0.01 2.42 ± 0.01 39 ± 1 38 ± 1 1.0 ± 0.5 5S6M 1.58 ± 0.01 2.73 ± 0.02 2.76 ± 0.01 43 ± 2 42 ± 1 1.3 ± 0.3 6S6M 1.50 ± 0.01 2.47 ± 0.03 2.51 ± 0.02 40 ± 1 39 ± 1 1.8 ± 0.5 7S6M 1.56 ± 0.02 2.44 ± 0.02 2.47 ± 0.01 37 ± 1 36 ± 2 1.2 ± 0.2

The porosity of the geopolymers is due to the mesoporous nature of the geopolymeric gel, the unreacted particles inside the matrix or the entrapped air after polycondensation. The bulk density of the geopolymers, in the range between 1.45- 1.70 g/cm-1, is in agreement with the literature for red mud-based geopolymers29. The porosity in the geopolymer samples is mainly open porosity, while the close porosity is low in all samples. No significant difference is observed either when the amount of waste glass or the molarity of the NaOH solution change. What is important to observe is that increasing the liquid to solid molar ratio from 0.50 to 0.55 leads to an increase in the total porosity of the geopolymer material. This is in agreement with the values of mechanical strength discussed in the paragraph 5.3.1. As already studied by Liew et al.206, Hamzah et al.207 and other authors, increasing the amount of solid in the geopolymer mixture usually causes the mechanical properties to grow thanks to the formation of a denser structure. The limit of the ratio between solid and liquid is usually determined by the viscosity and workability, since increasing too much the solid amount implies the impossibility of mixing and subsequently casting the material.

Red mud-based geopolymers synthesized with waste glass instead of water glass

The mechanical and porosity results confirm that working with more amount of solid allows to improve the performance of the geopolymers. For this reason further characterization will involve investigating geopolymers developed with a solid to liquid ratio of 0.50.

5.3.3 Fourier Transform Infrared Spectroscopy

Figure 5.3 shows the FTIR spectra of geopolymers with different SiO2/Al2O3 molar ratio and NaOH solution molarity.

7S6M 6S6M

5S6M

7S4M

6S4M

5S4M Trasmittance (%) Trasmittance RM

SLG

Si-O Si-Al 965 Si-O-Si 425

2000 1900 1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 Wavenumber (cm-1)

Figure 5.3: FTIR spectra of geopolymer samples and raw materials. The peaks of relevance are discussed in the text.

It is known from literature that the main characteristic bands for geopolymer materials are present in the region between 900 and 1200 cm-1 54. In fact, in this region the bends associated with the Si-O-Si asymmetric vibration and Al-O-Si symmetric bending vibration are situated 53,186. The main red mud band in the FTIR spectra is positioned at 980 cm-1, representing Si-O vibration208. As detected in Figure 5.3, the FTIR spectra regarding geopolymers are not a linear combination of the spectra associated with the raw materials, indicating that the raw materials

Chapter 5

dissolved and polymerized to create the geopolymers. Moreover, the red mud band, detected at 980 cm-1 shifts to lower frequencies after geopolymerization. In agreement with the values in the literature72, the Si-O and Al-O bands in geopolymer materials are individuated around 965 cm-1 and the intensity of these bands is attributed to the amount of Si-O or Al-O in tetrahedral configuration. No significant variation in the FTIR spectra is observed in the red mud-based geopolymers when changing the SiO2/Al2O3 ratio and the amount of waste glass, all samples present the typical bands after geopolymerization.

5.3.4 Microstructure

4M 6M

20μm 20μm

Figure 5.4: SEM images of samples realized with NaOH molarity at 4 and 6 and SiO2/Al2O3 molar ratio of 5,6 and 7 showing unreacted particles surrounded by amorphous geopolymeric gel.

Red mud-based geopolymers synthesized with waste glass instead of water glass

Figure 5.4 presents the microstructure of sample pieces (fracture surfaces) after compression test. Sample 5S4M shows on the surface visible flake-shaped particles that could be identified as unreacted red mud.

Increasing the SiO2/Al2O3 molar ratio corresponds to a decrease in red mud amount and to fewer unreacted red mud particles that remain in the geopolymer microstructure. Moreover, in all samples irregularly-shaped fragments of unreacted waste glass are present in the microstructure. However, the presence of unreacted material is reasonable considering that no water glass is used and that NaOH solution molarity is low compared to standard geopolymer formulations. The increase of waste glass amount leads to a slight formation of cracks but a relatively scarce quantity of cracks and voids is visible in Figure 5.4. Sample 5S6M presents the most homogeneous microstructure in accord with the results from the mechanical characterization. No inhomogeneities, like unreacted particles, cracks or voids are visible, as a consequence of a good geopolymerization reaction. Sample 5S7M, due to the high amount of waste glass, shows unreacted fragments that remained poorly encapsulated in the geopolymeric gel, this is visible from the large voids left on the surface after the compression test. However, we can conclude that a higher alkalinity allows decreasing the amount of unreacted materials and facilitates the formation of the geopolymeric gel that consequently results in improved mechanical properties. The formation of a geopolymeric gel with glass particles embedded in the structure is confirmed on the basis of high magnification details of the microstructure in Figure 5.4 and EDX analysis on selected points. Spectrum “a” collected in the matrix corresponds to a truly geopolymeric gel, considering the Si/Al and Al/Na molar ratio close to 2 and 1, respectively, as expected from the formation of a geopolymer material7. The spectrum “b” corresponds to an unreacted glass fragment, considering that the main signals in EDX spectrum match with the characteristic elements in the soda-lime glass composition.

Chapter 5

O a

a.u. a.u. Si

Na Al

Fe Fe Ca Ca

0 1 2 3 4 5 6 7 a KeV

b Si b

5 µm a.u.

Ca O

Na Al K

0 1 2 3 4 5 6 7 KeV Figure 5.5: Higher magnification SEM image of the sample 5S6M and EDX analysis. Showing the geopolymeric gel (a) and a fragment of unreacted glass (b). Reproduced from ref.205 with permission from Elsevier.

5.3.5 MAS-NMR technique

6S6M

9 5S6M 61

Red Mud

100 90 80 70 60 50 40 30 20 10 0 -10 -20 Al chemical shift (ppm)

Figure 5.6: 27Al MAS NMR spectra of raw materials and geopolymers activated with 6M NaOH solutioni, showing the tetrahedral coordination of alumina in the geopolymer samples. Reproduced from ref.205 with permission from Elsevier.

iThe analyses have been done at the Erlangen Catalysis Resource Center in collaboration with Dr. Y.S. Avadhut.

100

Red mud-based geopolymers synthesized with waste glass instead of water glass

The formation of an aluminosilicate geopolymeric gel was confirmed also by 27Al MAS NMR analysis represented in Figure 5.6. Red mud shows two small signals at 61 and 9 ppm, attributed to tetrahedral and octahedral coordinated Al, respectively209. After geopolymerization Al(VI) was mostly converted into Al(IV): the tetrahedral coordination is known to be essential for developing a ´zeolite-like´ geopolymeric structure7. Moreover, after raw materials have undergone geopolymerization, the peaks become noticeably narrower and sharper indicating a higher degree of polymerization and structural order, similar to the peaks that are associated with tetrahedral aluminium zeolites. From Figure 5.6 it is also noticeable that when the SiO2/Al2O3 molar ratio increases, the Al tetrahedral peak becomes narrower and sharper. The small amount of Al(VI) visible in the geopolymer spectra is a result of unreacted particles that remain inside the material. 29Si NMR was also carried out but was greatly affected by the presence of iron, resulting in very noisy spectra and in some cases no signals were obtained.

5.3.6 Leaching test

Table 5.5: Leaching test results from raw materials and geopolymer samples. Threshold values for inert and non-hazardous materials from ref. 180,j. Reproduced from ref.205 with permission from Elsevier. Element (ppm) As Ba Cd Cr Cu Mo Pb Zn

564M 1.550 <0.020 0.013 0.080 0.229 0.299 0.328 <0.202

5S6M 1.549 <0.020 0.012 0.102 0.246 0.151 0.304 <0.203

6S4M 1.545 <0.020 0.01 0.027 0.201 0.188 0.093 <0.203

6S6M 1.549 <0.020 0.016 0.062 0.251 0.565 0.162 <0.203

7S4M 1.620 <0.020 0.017 0.059 0.205 0.117 0.178 <0.203

7S6M 1.550 <0.020 0.016 0.037 0.298 0.113 0.187 <0.203

Red Mud 0.015 <0.020 0.008 0.599 0.027 1.021 0.130 <0.203

Waste Glass <0.049 <0.020 0.001 0.043 0.036 0.007 0.018 0.088

Inert material 0.5 20 0.04 0.5 2 0.5 0.5 4 Non-hazardous 2 100 1 10 50 10 10 50 material

j ICP test was carried out at the University of Padova by Dr. Giulia Zanmarchi.

101

Chapter 5

The potential applications of waste-derived geopolymeric materials require the complete stabilization of the pollutants within the geopolymeric network. Table 5.5 shows that red mud after stabilization inside the geopolymer matrix reduced its leaching of some heavy metals, such as Cr and Mo. The leaching from the developed geopolymers actually exceeded the thresholds for inert materials, according to European Norm EN 12457 in particular for As, however, the leaching was below the limit for non-hazardous materials. The obtained values are analogous to those found in literature even considering the high amount of red mud incorporated29. Lower red mud incorporation will probably lead to better leaching results. Moreover a possible alternative could be to use a red mud with lower amount of heavy metals, derived from a preliminary recovery process. In any case the proposed approach has high potential for stabilization of heavy metals onto the geopolymer network.

5.4 Conclusions

This chapter presents an experimental study that investigated the potential reuse of red mud, an industrial waste from alumina refining, via geopolymerization reactions with another waste, soda lime glass. Sodium hydroxide was the only non-waste material used as alkaline solution to dissolve the raw materials. Different ratios between alkaline solutions and solid raw materials were evaluated pointing out that the best mechanical performance is obtained by keeping this ratio at the lowest level as long as the workability allows to mix the blend. This new study is interesting not only because it demonstrates that the use of chemical water glass can be avoided thanks to the use of waste glass, but also because red mud a by-product with well known disposal problems, is used in high amount. By incorporating 60% wt. of red mud and 40% wt. of waste glass a compressive strength value of 45 MPa is obtained. This is a satisfactory value comparable to that of Portland cement and it acquires added value considering that it has been obtained starting from industrial waste material. The formation of an aluminosilicate geopolymer was confirmed by MAS-NMR analysis and by Si/Al and Na/Al ratios, estimated by the EDX measurements.

102

Red mud-based geopolymers synthesized with waste glass instead of water glass

SEM micrographs also confirm the composite nature of the final geopolymer product which incorporates some unreacted particles. A leaching test was performed to assess the material safety and leaching potential. The results were in accord with values found in literature, despite the relatively high amount of red mud in these new formulations. The results achieved entail significant environmental and economic impact if the materials are developed further not only in the construction industry but also in the recycling management.

103

Chapter 6 Geopolymers foams from inorganic gel casting process

6.1 Introduction

The use of thermal insulation materials constitutes the most effective way of reducing heat losses in buildings and therefore reducing heat energy needs. With the exception of expanded cork, which is based on recyclable material, all the other insulation materials are potentially toxic210. Polystyrene, for example, contains anti- oxidant additives and ignition retardant, moreover, its production involves the generation of benzene and chlorofluorocarbons. Polyurethane obtained from isocyanates could be as well really dangerous and toxic, because isocyanate remains one of the most commonly reported causes of asthma worldwide211. For these reasons, the development of alternative insulating materials is becoming important in the construction sector. The interest in geopolymer technology, in the recent years, has fuelled the development of different types of geopolymer products, including highly porous geopolymer-based foams171,212. Foamed geopolymers are unique in terms of mechanical properties, high temperature resistance and fire safety100. Therefore, they may be applied in various branches of industry as lightweight pre-fabricated components, acoustic panels, membrane supports, catalytic supports, heavy metal adsorbents, filters, etc 213,99. Another advantage of geopolymer foams is that they are produced at temperatures below 100 °C, representing a cost-effective energy production technology in comparison with foamed glass or ceramic, both of which are produced at high temperatures, above 900 °C 214,215. Geopolymer foams are typically created by adding foaming agents, which decompose and release gases into the geopolymer mixture: finely divided metallic aluminum, hydrogen peroxide, sodium peroxide, sodium perborate and metal silicon, have been widely used in literature as foaming agents216. In particular, aluminum, 103,217,218 silicon and H2O2 are the most used in the fabrication of geopolymer foams , howevers the pores that they generate are mostly closed, then leading to closed cell

105

Chapter 6

foams thereby limiting the range of applications. For this reason, different authors are investigating the production of geopolymer foams in alternative ways. Rincon and al.58 developed a novel “inorganic gel casting” process that consists in vigorously stirring the geopolymer gel and the surfactant added. Thanks to the stirring more bubbles are incorporated in the mixture and then setting determines the freezing of the cellular structure. Surfactants are used as surface-active agents to stabilize the wet foams, e.g. to stabilize the liquid-gas interface, decreasing in this way the surface tension of the system: the long-chain molecules adsorb at the gas bubble surface with their hydrophilic tail in contact with the aqueous phase. The type of surfactant could influence the pores size, size distribution and degree of the interconnection among adjacent pores. This novel foaming method was used in this study to produce fly ash and red mud- based geopolymer foams. The formulations developed in chapter 4 for fly ash-based geopolymers and in chapter 5 for red mud-based geopolymers, after the addition of surfactants, were foamed to produce porous geopolymers. The geopolymer mixtures were intensively mechanical stirred at room temperature to create the cellular structure, without the need of stabilization additives. A non-ionic surfactant such as triton and a more economic anionic surfactant as sodium lauryl sulfate were used. Compressive stregth test, density and porosity calculation and microstructure evaluation were carried out on the foamed geopolymers. The boiling test was used to verify the chemical stabilization of the foams. This work was carried out in collaboration with the University of Padova (Italy) in the framework of CoACH EU project 182.

6.1.1 Experimental procedure The geopolymer mixtures were synthesized as described in chapters 4 and 5, using fly ash, soda lime glass waste and red mud. Fine powders made of fly ash and waste glass and red mud and waste glass, respectively, were mixed in NaOH solutions for 4 hours in order to obtain an homogeneous mixture. The obtained blends were cast in polystyrene cylindrical moulds (60 mm diameter). Non-ionic surfactant Triton X-100 (polyoxyethylene octyl phenyl ether –C14H22O(C2H4O)n, n=9-10, Sigma-Aldrich, Gillingham, UK) or anionic sodium lauryl sulfate (SLS) (CH3(CH2)11OSO3Na Carlo Erba) in an aqueous

106

Geopolymer foams from gel casting inorganic process

solution 1/10 in weight for an amount of 4 wt% were added to the geopolymeric mixture and vigorously mixed (2000 rpm) for 5 min with a head stirrer in the first step of geopolymerization. The so-prepared wet foams were kept in an oven at 60°C for 48 h to complete the geopolymer reaction and successively demolded. A schematic representation of the processing procedure is depicted in Figure 6.1.

NaOH pellets Waste glass + + Distilled water Fly ash or red mud

4 hours stirring

Geopolymer slurry

Geopolymer slurry + 5 min stirring Surfactant

60°C oven (sealed mold 1d) + (open mold 1d)

Demolding

Figure 6.1: Schematic diagram of the production of geopolymer foams using inorganic gel casting process.

The sample labels used for the foams follow the same format as the one used for the dense geopolymers: “xSyM” where “x” is associated with SiO2/Al2O3 molar ratio and “y” refers to the molarity of the NaOH solution. All the fly ash-based formulations developed in chapter 4 were foamed with SLS, while only the formulation activated with NaOH 5 molarity was foamed with Triton. The red mud-based geopolymers, which were activated with 6M NaOH solution and presented the best mechanical performance, were foamed with SLS.

107

Chapter 6

6.1.2 Characterization technique

Density evaluation

The bulk density (b) was evaluated by considering the mass to volume ratio. The apparent (a) and true density (t) were measured by using a helium pycnometer (Micromeritics AccuPyc 1330, Norcross, GA), operating on bulk or on finely crushed samples, respectively. The three density values were used to compute the amounts of open and closed porosity.

Compressive strength The foams were cut with a metal saw and polished to obtain samples of about 10mm x 10mm x 10mm as shown in Figure 6.2.

Figure 6.2: Geopolymer foams were cut to obtain cubes for further characterizations.

The obtained cube-shaped foams were subjected to compressive strength tests by using an Instron 1121 UTS (Instron, Danvers, MA) machine, with a crosshead speed of 0.5 mm/min. For each formulation 5 samples were tested.

Morphological and microstructural characterization The morphological and microstructural characterization was performed by optical stereomicroscope (M50 Mikrosysteme Vertrieb GmbH) and scanning electron microscopy (LEO 435, LEO Electron Microscopy Ltd., Cambridge, UK and Ultra Plus, Zeiss, Jena, Germany). Pieces of foams after compressive strength were analyzed.

108

Geopolymer foams from gel casting inorganic process

Chemical stability evaluation A becher containing distilled water was heated with hot plate in order to bring the water to boiling temperature. When the water was boiling the geopolymer foams were immersed for 20 minutes to verify their chemical stability.

6.2 Results and discussions

6.2.1 Fly ash-based geopolymer foams

The foams were created starting from the geopolymer binders activated with NaOH solution with molarity 3, 5, 8 and 10. From the binders with higher alkalinity foams could not be produced. The excessive viscosity, in samples with 8 and 10 molarity, hindered the encapsulation of air in the slurry, consequently preventing the formation of a foamed material. Figure 6.3 shows the foams created with 8 and 5 molarity NaOH solutions, on the top and on the bottom respectively. The one on the top is not foamed and dense, as visible from the microscope picture, while the foam created with NaOH 5 molarity, on the bottom, resulted in a foamed material with pores well distributed in the matrix.

a b

1 cm 2 mm

c d

1 cm

Figure 6.3: Geopolymer foams created with SiO2/Al2O3 molar ratio of 7 and NaOH molarity of 8 a) and b) and 5 c) and d).

109

Chapter 6

For these reasons the fly ash-based foams activated with NaOH molarity over 5 were not characterized. However, the use of weaker alkali concentration allows to obtain less expensive and more environmentally friendly geopolymer foams. Table 6.1 shows the values of density and porosity after 28-day aging for mixtures incorporating SLS, while in Table 6.2 the values for the foams with triton are reported.

Table 6.1: Density and porosity of fly ash-based geopolymers foams with SLS. Density (g/ ) SLS Porosity (%) SLS Total porosity Open porosity Closed Bulk [ρ ] Apparent [ρ ] True [ρ ] b a t [TP] [OP] porosity [CP] 5S3M 0.40± 0.03 2.59± 0.04 2.63 ± 0.01 85 ± 8 85 ± 8 0.2 ± 0.2

6S3M 0.39 ± 0.04 2.44 ± 0.01 2.46 ± 0.01 84.1 ± 1 84 ± 1 0.1 ± 0.1

7S3M 0.41 ± 0.01 2.44 ± 0.04 2.48 ± 0.04 83 ± 3 83 ± 4 0.24 ± 0.02

5S5M 1.38 ± 0.05 2.32 ± 0.03 2.38 ± 0.03 42 ± 1 40 ± 1 1.50 ± 0.30

6S5M 0.56 ± 0.01 2.33 ± 0.01 2.50 ± 0.04 77.3 ± 3 76 ± 4 1.63 ± 0.04

7S5M 0.83 ± 0.01 2.36 ± 0.02 2.50 ± 0.01 67 ± 2 65 ± 3 1,90 ± 0.10

Table 6.2: Density and porosity of developed fly ash-based geopolymers foams with triton. Density (g/ ) Triton Porosity (%) Triton Total porosity Open porosity Closed porosity Bulk [ρ ] Apparent [ρ ] True [ρ ] b a t [TP] [OP] [CP] 5S5M 0.93 ± 0.01 2.18 ± 0.06 2.27 ± 0.01 59 ± 1 57 ± 1 1.7 ± 0.3

6S5M 0.88 ± 0.01 2.22 ± 0.03 2.50 ± 0.04 65 ± 1 60 ± 1 4.3 ± 1.0

7S5M 0.89 ± 0.02 2.24 ± 0.02 2.50 ± 0.01 64 ± 2 60 ± 2 4.2 ± 1.0

All samples with SLS, except 5S5M, present values of bulk density between 0.40 and 0.80 g/cm3. These values are comparable to the values of fly ash-based geopolymers foams created using Al powder or H2O2, which are determined by Ducman et al.99 to be in the range between 0.60 and 1 g/cm3. Samples incorporating triton, due to their higher viscosity as no water is incorporated with the surfactant, show higher density than the ones with SLS, but also in this case below 1 g/cm3. From the values in Table 6.1 it can be observed that the 80% of total porosity in the foams is mainly composed of open porosity while the closed porosity is less than 2%.

110

Geopolymer foams from gel casting inorganic process

As expected, by increasing the molarity of the NaOH solution the porosity of the geopolymer foams decreases, due to the previous increase in viscosity. When comparing the porosity values using anionic surfactant and non-ionic surfactant it can be noticed that triton, due to the difference in chemical structure, enhances the viscosity of the geopolymer mixture, inhibiting the formation of pores. For this reason, the results indicate that SLS was a more effective surfactant because it allowed to stabilize a larger amount of bubble surfaces. In Figure 6.4 the microstructure for the geopolymer foam 6S3M can be seen. The same morphology was observed for the other samples activated with NaOH solution of 3 molarity.

1 mm

100 µm

40 µm

Figure 6.4: Microstructure of the samples 6S3M with SLS surfactant. At higher magnification particles of fly ash and unreacted glass fragments are visible.

111

Chapter 6

The microstructure presents big pores distributed uniformly over the material. In the images at higher magnification it can be noticed that the level of geopolymerization is low. Spherical particles, identified as fly ash, and sharply particles associated with waste glass are visible in the walls of the pores. As already demonstrated in the previous chapters a low alkalinity is not enough to obtain a high level of geopolymerization. Figure 6.5 shows the microstructures of the geopolymer foams with SLS on the left and triton on the right synthesized with NaOH 5 molarity. Also in these microstructures spherical unreacted fly ash particles are visible, but in comparison with Figure 6.4 the amount of unreacted material is lower.

Figure 6.5: Microstructures of foamed geopolymers created with 5M NaOH solution and incorporating SLS (left) and triton (right).

112

Geopolymer foams from gel casting inorganic process

Except for the foam 5S5M, the foam microstructures revealed that the samples with different SiO2/Al2O3 molar ratio and different surfactant have similar pore morphology. Samples with higher SiO2/Al2O3 molar ratio (e.g. higher glass content) seem to have more pores but with a smaller size in comparison with samples with less amount of glass. Figure 6.6 exhibits the compressive strength of samples with different surfactants: on the left (a) the foams with SLS while on the right (b) foams with triton.

11 3 5S 5S 10 6S 6S 9 7S 7S

8 2 7

4 1 3

Compressive strength [MPa] strength Compressive Compressive strength [MPa] strength Compressive 1

0 0 2 3 4 5 6 4 5 6 Molarity Molarity

Figure 6.6: Compressive strength of geopolymer foams with SLS (a) and triton (b).

An increase of NaOH molarity is associated with an increase in the compressive strength as expected after analysis of the microstructure. Except for the sample 5S5M, which presents high strength value due to the low porosity, we can affirm that increasing the amount of glass causes the compressive strength to grow. This result correlates with the decrease of density obtained by increasing the amount of glass in the structure and consequently the viscosity of the binder. Foams with triton are not affected by the change in SiO2/Al2O3 molar ratio, but in comparison with the foams realized with SLS the values are lower. Moreover the values of compressive strength reached with this eco-friendly formulations are comparable to those of fly ash

218, 100, 219 geopolymers foamed with H2O2 or Al powder reported in literature .

113

Chapter 6

6.3 Red mud-based geopolymer foams

Table 6.3 reports the values of density and porosity for the red mud-based geopolymer foams with SLS surfactant.

Table 6.3: Density and porosity of red mud-based geopolymers. Density (g/ ) SLS Porosity (%) SLS Red Mud Total porosity Open porosity Closed porosity Bulk [ρ ] Apparent [ρ ] True [ρ ] b a t [TP] [OP] [CP] 5S6M 0.92 ± 0.02 2.89 ± 0.02 2.91 ± 0.01 68 ± 3 68 ± 3 1.0 ± 0.3

6S6M 0.93 ± 0.01 2.67 ± 0.05 2.88 ± 0.01 68 ± 1 65 ± 3 2.5 ± 1.2

7S6M 0.97 ± 0.02 2.65 ± 0.05 2.76 ± 0.01 65 ± 2 63 ± 3 1.44 ± 0.8

From the results it is evident that there are not significant variations of density when changing the SiO2/Al2O3 molar ratio in the geopolymer binder. The geometric density for the red mud foams is lower than 1 g/cm3. The porous geopolymers developed with red mud have lower porosity compared with the fly ash-based foams. For these geopolymers, as for the fly ash-based geopolymers, the porosity is nearly totally composed of open porosity, while the closed porosity is between 1 and 2%. The results achieved are in accord with the values reported in literature for metakaolin and fly ash-based geopolymers213, 220, while no red mud foams are still being explored in the literature.

114

Geopolymer foams from gel casting inorganic process

Figure 6.7: Microstructure of red mud-based geopolymer foams. Increasing the amount of glass more unreacted particles are visible. All the formulations present a homogeneous distribution of the pores.

Figure 6.7 shows the microstructure of the foamed geopolymers. Increasing the amount of glass seems to enhance the amount of unreacted materials, in fact 7S6M foam shows a less homogeneous microstructure, while 5S6M presents more definite pores with less unreacted particles on the walls of the pores. All the foams show a fairly uniform porosity with pores smaller than 50 µm. In addition, no formation of cracks is observed in the material after the geopolymerization of the slurry. As confirmed by the values of porosity the pores are mainly interconnected inside the structure. In Figure 6.8 the cube-shaped foams used for the compressive strength test are presented, while Figure 6.9 shows the compressive strength values obtained.

115

Chapter 6

5S6M 6S6M 7S6M

1 cm

Figure 6.8: Geopolymer cube-shaped foams used for the compression test.

In the foams realized with red mud and waste glass, contrariwise to the fly ash-based ones, increasing the amount of glass allows the compressive strength to rise. This phenomenon is not so pronounced, since all the values are in same range. The foam values, in the range between 1 and 2 MPa, are relatively high considering the high porosity of this lightweight geopolymers.

3 5S 6S 7S

1 Compression strength [MPa] strength Compression

0 5 6 7 Molarity

Figure 6.9: Compressive strength for red mud geopolymer foams when changing the SiO2/Al2O3 molar ratio.

116

Geopolymer foams from gel casting inorganic process

6.4 Chemical stability

To test the chemical stability of the realized foams in a qualitative manner the boiling test in water was carried out. According to Davidovits7, the boiling test is one of the most severe tests because samples often disaggregate totally during the test.

Figure 6.10: Samples after boiling test, 6S6M red mud on the far left, 7S3M in the middle and 7S5M on the far right, showing impurities in the water, except for the sample 7S5M where the water maintain its transparency.

After observing Figure 6.10 we can conclude that geopolymer 7S5M shows a good stability in water, while foams 7S3M fly ash and 6S6M red mud present cloudy water and deposition of material in the bottom after the test. To better stabilize the chemical stability of the foams, in particular at low molarity, they were cured in the oven at 800°C. Foams after curing treatment, carried out at the University of Padova, are not reported in this thesis.

6.5 Conclusions

In this study, porous fly ash- and red mud-based geopolymers without sodium silicate, were developed through a novel inorganic gel casting process. The foamed geopolymers can be a promising alternative to other foamed materials thanks to the suitable mechanical properties achieved together with the eco-friendly and economic fabrication process. The developed geopolymer foams posses low density and suitable mechanical properties as reported in summed up in Table 6.4.

117

Chapter 6

Table 6.4: Summary of the density and compressive strength of the developed geopolymer foams. They present high open porosity, low density and reasonable mechanical properties. Samples Bulk Density (g/ ) Compressive stregth (Mpa)

5S3M 0.40± 0.03 0.11 ± 0.02

6S3M 0.39 ± 0.04 0.24 ± 0.01

7S3M 0.41 ± 0.01 0.26 ± 0.02

FlySLS ash 5S5M 1.38 ± 0.05 8.85 ± 1.17

6S5M 0.56 ± 0.01 0.99 ± 0.06

7S5M 0.83 ± 0.01 3.65 ± 0.51

5S6M 0.92 ± 0.02 0.93 ± 0.13

6S6M 0.93 ± 0.01 1.42 ± 0.31

Red mud SLS Red 7S6M 0.97 ± 0.02 1.68 ± 0.23

The results demonstrate the possibility to create porous materials just with the introduction of SLS and a vigorous stirring. Moreover SLS shows better results in comparison with triton surfactant. The geopolymer foams were developed using an extremely economic process that does not require high temperatures or foaming agents or additives. This technology not only solves the problem of reusing waste materials, but it also addresses a new strategy for obtaining insulating materials at low price. The obtained results from this study are a first attempt to create lightweight materials by incorporating extensive amount of waste materials and without introducing chemical reagents. Starting from the results presented in this chapter further characterization on these foams was carried out by the University of Padova, whose investigation revealed other interesting properties not already published.

118

Chapter 7 Thermal shock and wear resistance

7.1 Introduction

Ceramic materials, thanks to their high melting point, are used in many applications where a high-temperature resistance is required, such as high-temperature furnaces, heat exchangers, gas turbine engines, oxide fuel cell, catalyst supports, etc 221. However, due to their low toughness, low thermal conductivity and high Young´s modulus 222, ceramics are particularly vulnerable to thermal shock failure, considered a crucial factor for their durability. During the process of fast heating or cooling, the temperature changes rapidly with time, producing thermal stresses223. In fact, when a ceramic material is exposed to high temperatures, the surface of the material reaches a high temperature relatively fast, whereas the temperature of the internal portion of the sample remains lower. This result in internal stresses related to the thermal expansion of the ceramic surface compared to the relatively intact interior223. In contrast, on cooling, the surface experiences tensile stress while the interior experiences compressive stress. When these thermal stresses exceed the strength of the materials they initiate and propagate surface cracks, often severe enough to induce a catastrophic fracture. For the above-mentioned reasons the susceptibility of ceramic materials to thermal shock is one of the main factors limiting their lifetime performance and applications. As already mentioned in the previous chapters geopolymers were primarily developed as substitute for Portland cement material but thanks to their satisfactory mechanical properties, chemical stability and thermal resistance they are becoming of interest also for other applications including fire protection panels224,225. In order to evaluate the possibility of using geopolymer materials for applications where the temperature changes rapidly the thermal shock test was carried out. Two different compositions developed in chapter 4 were subjected to heating at 200 °C, 500°C and 800°C followed by quick cooling in water (quenching test). Compression test was performed to evaluate the mechanical resistance, XRD and SEM analyses were completed to verify the changes in microstructure and crystalline phases respectively, after thermal shock test.

119

Chapter 7

Furthermore, ceramic materials such as Al2O3, Si3N4, SiC and ZrO2 are known to possess excellent wear resistance and for this reason there ceramic materials are involved in many applications where tribological aspects are important124. At the moment there is a strong interest in applications of ceramic materials as components in machines without lubrication, for this reason research on wear resistance behavior of ceramics is being carried out in many research groups226. Wang et al.124 tested for the fist time the wear resistance of metakaolin-based geopolymer composites. He demonstrated that the introduction of PTFE prevents geopolymers from severe wear obtaining satisfactory results. To the author knowledge, no other information is available in the literature concerning the wear properties of geopolymer materials. In this chapter, geopolymers developed with an unconventional formulation as described in chapter 4, and in particular some of the geopolymers exhibiting the best mechanical properties have been tested with the pin-on-disc test in order to evaluate the wear resistance.

7.2 Thermal shock test

7.2.1 Analysis method

Thermal shock test was performed on fly ash-based geopolymers 6S8M and 5S10M, prepared as described in chapter 4. The characterization was carried out after a 28-day ageing period. The thermal shock resistance behavior of the geopolymers was studied by measuring the retained compressive strength (Zwick Roell, Ulm, Germany Series Z050), the change in porosity (Micromeritics AccuPyc 1330, Norcross, GA) and the microstructure (LEO 435, LEO Electron Microscopy Ltd., Cambridge, UK and Ultra Plus, Zeiss, Jena, Germany). Samples of 12 mm diameter and 17 mm in height (5 for each temperature) were placed in the furnace at 200°C, 500°C and 800°C for a soaking period of 10 minutes and then they were rapidly quenched in cold water for 10 minutes complying with ASTM C1171 procedure.

120

Thermal shock and wear resistance

7.2.2 Results and discussion

Figure 7.1 and Figure 7.2 show the samples 5S10M and 6S8M, respectively, after thermal shock test at 200°C, 500°C, 800°C followed by cooling in cold water.

0°C 200°C 500°C 800°C

1 cm

Figure 7.1: Samples 5S10M after thermal shock at 200°C, 500°C and 800°C.

0°C 200°C 500°C 800°C

1 cm

Figure 7.2: Samples 6S8M after thermal shock at 200°C, 500°C and 800°C.

After the thermal shock, the samples do not show the formation of large cracks or defects, which usually implies a catastrophic failure observed in ceramic materials when subjected to thermal stresses. The samples heated at 800°C show dimensional changes and in detail they transform into geopolymer foam when heated to high temperatures. This transformation could be considered an advantage in construction application. In fact, foamed geopolymers have usually better thermal resistance than unfoamed geopolymers as pores provide space to counteract the internal damage227. Moreover, in case of fire, foamed panels act as flame retardant providing valuable protection from heat or delay of mechanical deterioration. Figure 7.3 shows the porosity development for sample 6S8M. As described above, when the temperature during the test increases the total porosity of the geopolymer

121

Chapter 7

5 samples grows. This is more evident after thermal shock at 800°C, confirming the formation of a foam-like structure212.

0,70

0,60

0,50

0,40

Porosity 0,30

0,20

0,10

5 0,00 φtotal φopen φclose φtotal φopen φclose φtotal φopen φclose φtotal φopen φclose 0°C 200°C 500°C 800°C

Figure 7.3: Porosity development after thermal shock test for geopolymers 6S8M.

A slight decrease in close porosity with the increase of the thermal shock temperature can be observed in Figure 7.3. This suggested that probably there was a sintering effect and partial melting due to the elevated temperature which allowed viscous flow to fill pores or voids present in the structure228.

70 6S8M 5S10M 60

20 Compressive strength (MPa) strength Compressive 10

0 0 200 400 600 800 ΔT

Figure 7.4: Compressive strength after thermal shock test at different temperatures.

Figure 7.4 reports the compressive strength values of the geopolymer samples after thermal shock test at different temperatures. Samples heated at 200 °C and then quenched in water showed mechanical strength similar to the non-quenched samples. After thermal shock test the loss in resistance is merely 6-17% compared to the untested geopolymer. When the testing temperature increased, the loss of mechanical

122

Thermal shock and wear resistance

properties became more pronounced. Geopolymers quenched after 800°C exhibited compressive strength values around 9 MPa. These are still satisfactory values considering the foam nature after exposition at such high temperatures, e.g. the samples did not break catastrophically as a consequence of the thermal shock loading. No substantial differences were observed when changing the geopolymer mixture, confirming the comparable thermal shock resistance of the developed geopolymers.

0°C 200°C

500°C 800°C

Figure 7.5: Microstructure of samples 5S10M before and after thermal shock test at 200°C, 500°C and 800°C.

123

Chapter 7

0°C 200°C

500°C 800°C

Figure 7.6: Microstructure of samples 6S8M before and after thermal shock test at 200°C, 500°C and 800°C.

Figure 7.5 and Figure 7.6 show the microstructure of the geopolymers before and after thermal shock test at different temperatures. The untested fly ash-based geopolymer matrix appeared homogeneous surrounding some unreacted fly ash particles. Few coarse pores can be seen in the geopolymer matrix, while a higher amount of cracks is distributed in the matrix. After thermal shock test there is a slight increase in the porosity distributed throughout the matrix. This is in agreement with the porosity values reported in Figure 7.3. Large pores visible for example in the sample 6S8M after exposition to 200°C followed by water cooling could be the result of water evaporation after temperature exposure227. Observing Figure 7.5 and Figure 7.6 no deterioration of the microstructures is visible after thermal shock test at 200°C and 500°C. Besides, minor cracks are observed, however, after thermal shock test at 800°C the microstructure of a foam is evident, the voids present different dimensions but no crack are visible on the geopolymer foams. Clearly the sample has expanded (foamed) and there is evidence also of a molten or soften phase.

124

Thermal shock and wear resistance

Quartz Nepheline Mullite

800°C

500°C Intensity (a.u.) Intensity

200°C

0°C

15 20 25 30 35 40 45 50 55 60 65 70 2(deg.)

Figure 7.7: XRD patterns of 5S10M geopolymer before and after thermal shock test.

Figure 7.7 shows the XRD patterns of the samples after thermal shock test. In the crystalline phases no changes are observed after thermal shock test at 200°C and 500°C, where the diffractogram shows the presence of quartz and mullite. As reported in literature, raising the temperature over 500°C causes an increase of the propensity towards the formation of stable crystalline phases. Cheng-Yong et al.227 demonstrated the presence of nepheline, anorthite and cristobalite after exposition of geopolymeric material to 800°C. Timakul et al.229 found belite crystalline phase in fly ash-based geopolymer composite after thermal shock test at 800°C. In Figure 7.7 it can be seen that after thermal shock test at 800°C the crystalline phases of the present geopolymer change. In particular nepheline crystalline phase, no detected until 500°C, forms after thermal shock test at 800°C. In general, the crystallization would most probably enhance the mechanical properties of geopolymers since crystalline phases might act as a filler to reinforce the geopolymer matrix 227,

125

Chapter 7

however this possible effect is compromised in the present case by the notable increase of porosity of the sample, leading to a reduction of strength.

7.3 Wear test

Wear tests were carried out at the Slovak Academy of Sciences, Bratislava. (Contact: Prof. Dr. Jan Dusza, Head of the Centre for Advanced Materials and Technologies).

7.3.1 Analysis method The pin-on-disc dry sliding wear tests (tribometer, THT, CSM Instruments, Switzerland) were performed at the load of 5N and linear speed of 0.1 m/s. Alumina ball with a diameter of 6 mm was used as counter body during the test. The wear track is a ring with 2 mm radius as shown in Figure 7.8 and the test lasted until the total sliding distance of 100 m was achieved. The friction coefficients were recorded continuously during the experiment. Wear volume on each specimen was calculated from the surface profile traces recorded across the wear track (Sensofar PLu neox, Spain).

F=5 N

Wear track Alumina ball

v=0,1 m/s

Figure 7.8: Scheme of the pin-on-disc wear tester.

126

Thermal shock and wear resistance

7.3.2 Results and discussion

0,9 6S8M 0,8 5S10M

0,7

0,6

0,5

0,4

0,3 Friction coefficient Friction

0,2

0,1

0,0 Samples

Figure 7.9: Friction coefficient of samples 6S8M and 5S10M.

0,0010 6S8M 5S10M

0,0008 /N.m) 3 0,0006

0,0004 Wear rate (mm rate Wear 0,0002

0,0000 Samples

Figure 7.10: Wear rate of samples 6S8M and 5S10M.

127

Chapter 7

Figure 7.9 and Figure 7.10 report the friction coefficient and wear rate, respectively, for samples 6S8M and 5S10M. The different geopolymer formulations show the same values of friction coefficient, while the samples with a higher amount of glass present a lower wear rate. It is likely that the increase of unreacted glass fragments in the geopolymer structure can have a positive effect on the wear test. The wear resistance of geopolymer materials has been tested only by Wang et al.124 who reported a friction coefficient of 0.7 and a wear rate of 260. 10-5 (mm3/m.N) for metakaolin-based geopolymers. Figure 7.9 and Figure 7.10 show a friction coefficient in the same range of the values reported by Wang et al. but a much lower wear rate showing a better wear behavior for waste-based geopolymers, compared to metakaolin ones. The friction coefficient determined for the developed geopolymers in this study is in the range of values reported in literature for different types of dense glass-ceramic materials, which is around 0.8 230.

1,0 0,9 a b 0,8

0,7

0,6

0,5

0,4

Friction coefficient Friction 0,3

0,2

0,1

0,0 0 10 20 30 40 50 60 70 80 90 100 Distance (m)

Figure 7.11: Friction coefficient vs sliding distance for all the geopolymers tested (a), confocal microscope image for sample 6S8M (b)k.

It can be observed in Figure 7.11 (a) that a steady state friction coefficient is attained after an initial short length of about 0.7 meters. The evolution of µ before and after reaching the steady state is very similar for all geopolymer samples tested indicating high reproducibility. Moreover, no noticeable fluctuation appears throughout the

k The test has been done by Dusza Jan at the Slovak Academy of Sciences, Bratislava.

128

Thermal shock and wear resistance

steady-state region of the geopolymers. In Figure 7.11 (b) the surface of the material after the wear test (confocal microscopy image) appears homogeneous with the area subjected to friction well delineated.

7.4 Conclusions

In this chapter, thermal shock and wear tests were carried out on fly ash-based geopolymers developed using waste glass instead of water glass. While these tests are unconventional for geopolymer materials, and there are no reports of such characterization in the literature, the tests used here are widely used to characterize ceramic components. Thermal shock and wear tests were therefore carried out in order to evaluate the possibility of using geopolymers where ceramic materials are currently employed. The thermal shock results prove the satisfactory ability of the geopolymer materials to counteract thermal stresses created after a sudden change in temperature. Moreover, geopolymers preserve high compressive strength after being exposed to 200°C, 500°C and 800°C heat treatment followed by quenching in water. The decrease in compressive strength after 200°C treatment is between 6 and 17% and the compressive strength after thermal shock at 800°C is close to 10 MPa. With increasing temperature, the porosity increases and at 800°C a foam-like geopolymer is formed, moreover according to XRD analysis, a nepheline crystalline phase is formed at this high temperature, indicating that geopolymer components could be subjected to thermal shock loading with ΔT<800°C. The friction coefficient values around 0.7 and 0.8, detected to the present geopolymers, are in accordance with metakaolin-based geopolymer values reported in the literature. Besides, these friction coefficient values are in the range of some glass ceramic materials. The wear rate shows some changes with the adjustment of the amount of glass in the geopolymer microstructure and in detail, it seems to decrease with an increase in the amount of glass. These first attempts at measuring technical properties on waste-based geopolymers show satisfactory results that require further investigation. Understanding the thermal shock and wear resistance of geopolymers the use of geopolymer materials in a wide

129

Chapter 7

range of applications, beyond construction/building materials, in particular for uses in mechanical engineering as part/components replacing potentially in some cases conventional ceramics.

130

Chapter 8 Advanced processing technologies for geopolymers

8.1 Introduction

As discussed in Chapter 2, geopolymers have arisen increasing interest thanks to their good properties in combination with the ecofriendly fabrication process. Despite that, the applications of geopolymer materials so far are still restricted to the building and construction field. Fabricating components of complex and intricate shape for applications beyond building materials may represent a useful approach to broaden geopolymer applications, as anticipated in Chapter 7. Little attention has been placed on developing components of complex shape through new processing technologies or on investigating machining techniques to achieve parts of superior surface finish and shape accuracy which could open the possibility of applications in other sectors, such as automotive components or mechanical engineering. Franchin et al.231 and Xia et al.232 already proved the possibility of 3D printing geopolymer slurries obtaining satisfactory results in terms of mechanical properties, shape and surface finish. No studies on the machinability of geopolymer materials are reported so far in the literature. The aim of this part of the present research project is to highlight the positive factors that favor geopolymer technology, including not only low cost processing and stabilization of recycled waste materials but also the possibility to process geopolymer components with unconventional or innovative technologies. In the present proof-of-concept experiments we investigated the applicability of two standard machining processes, namely machining on the lathe and drilling, on cylindrical geopolymer samples. The results show good geopolymer machinability without need of adding lubricant, additives or stabilizers. Moreover, the fly ash based-geopolymers, created in the laboratory (see chapter 4), were developed in an industrial plant, to obtain samples in a larger scale, to confirm

131

Chapter 8

that the parameters fixed in laboratory scale could be easily applied to industrial production.

8.2 Machinability test

8.2.1 Materials and analysis method Cylindrical samples of diameter 14mm and height 16mm were prepared by mixing 64 wt.% of fly ash and 36wt.% of waste glass with NaOH at 8 molarity, as described in chapter 4. They were conventionally clamped on a lathe and subsequently machined, at a speed of 800 rpm using a Weiler Matador vs2 machine (Weiler Werkzeugmaschinen GmbH, Emskirchen, Germany). The samples were machined using turning tool made of hard metal. In order to drill geopolymer cylinders, these were machined using a stainless steel tool, 5 mm diameter, at a speed of 800 rpm. No lubricant was used during the process.

8.2.2 Results and discussion Figure 8.1 (a) and (b) show the initial and desired shape respectively, while in the images at the bottom the cylindrical geopolymer samples worked on the lathe (c) and the samples after processing (d) can be seen.

a b

8 mm 8

16 mm 16 8 mm 8

14 mm 14 mm

c d

Figure 8.1: Example of geopolymer sample fabricated on the lathe. (a) Original geopolymer shape, (b) desired final shape, (c) sample on the lathe and (d) final geopolymer component after machining.

132

Advanced processing technologies for geopolymers

It was demonstrated that geopolymer samples could be machined by using a lathe and traditional metal tooling to obtain the desired final shape without failure. The microscope image in Figure 8.1 (d) shows that the final shape of the sample was highly accurate and well defined. Moreover no cracks propagated during the work on the lathe. In the central part of the geopolymer some bubbles are visible: they are probably air bubbles that were entrapped in the samples during the synthesis, but are not related to the machining process. Furthermore, the geopolymer showed uniform dimensional accuracy and smooth surface; i.e. neither roughness nor defects were visible after the process. Cylindrical geopolymer samples were also drilled with a metal tool, as shown in Figure 8.2, to realize a hole of 5 mm in diameter in the middle of the sample.

a b

16 mm 16 16 mm 16

14 mm 14 mm

c d

Figure 8.2: Example of geopolymer sample drilled with a metal tool. (a) represents the original geopolymer shape, (b) the desired final shape, (c) the samples on the lathe and (d) the final geopolymers.

In Figure 8.2 (d) the precision of the shape of the hole can be appreciated: the inner wall has a good surface finish. Also in this case the drilling operation did not break the sample and did not initiate the formation of cracks in the material. The sample

133

Chapter 8

appears without defects and just a small bubble is visible on the surface, probably already present before processing and not related to the drilling operation. In an early study, Boccaccini233 claimed that to achieve a good machinability in ceramics material, low brittleness index (B) is required, which means reaching a compromise between hardness and fracture toughness (B is the ratio of hardness to fracture toughness). Thanks to the excellent machinability, demonstrated in this study through drilling technology and work on the lathe, the relationship hardness vs toughness appears to be suitable in the case of the developed geopolymers. Moreover, these results, even if qualitative, indicate increased versatility and the possibility of innovative industrial applications for geopolymer materials as components of complex shape.

8.3 Direct ink printing

8.3.1 Analysis method 3D geopolymer porous structures were fabricated through additive manufacturing using a bio plotter (type BioScaffolder 2.1; GeSiM, Großerkmannsdorf, Germany) (Figure 8.3).

a b

15mm

Figure 8.3: Bioplotter (a) and the syringe used to print the slurry (b).

This bioplotter is equipped with a printing head movable in three room space axes, a pressurized air system and a static platform. The air system is compatible with the cartridges (used as a reservoir for the plotting material). Micro-nozzles were

134

Advanced processing technologies for geopolymers

purchased from Nordson EFD (Switzerland). This system is conventionally used to fabricate scaffolds for biomedical applications234 but it was conveniently applied here for geopolymer parts. To define the dimension and the shape of the scaffolds, the “Scaffold generator” software was used. With this software the height of the different layers in the z-direction, the edge lengths of the scaffold as well as the plotting speed and the number of plotted struts could be defined. The scaffolds were directly plotted into a plastic substrate (culture plate) (VWR chemicals), which was placed on the static platform. The scaffolds were cured at room temperature. As a proof-of-concept experiment, a grid-like scaffold was plotted using a plotting speed of 20 mm/s, 3 layers and 5 struts per layer. The two chosen needles had inner diameters of 0,84 mm and 1,36 mm. The morphology of the scaffolds was investigated by an optical microscope, Zeiss Stereolupe Stemi 508. The processing parameters were selected for the geopolymer considered in this study by simple trial and errors.

8.3.2 Results and discussion Direct ink printing is a challenging task for geopolymers because the slurry is subjected to ongoing polycondensation reactions which continuously modify the rheological properties over time231. For this reason, different parameters such as alkaline solution viscosity, particle size, printing speed and pressure should be arranged to obtain an extruded sample. At the beginning, to evaluate the possibility to create an ink with 6S8M fly ash-based geopolymers, a syringe was filled with the blend and the material was manually extruded. The extruded ink behaved like a gel that is typical of geopolymers after dissolution of the raw materials and the structure fabricated kept the shape without collapse. Figure 8.4 shows on the left the filament created by manual extrusion with a plastic syringe, while on the right the scaffold-structure after 28 days of aging at room temperature.

135

Chapter 8

a b

1 cm

Figure 8.4: Geopolymer ink extruded with a syringe on the left (a) and the final scaffold-structure after 28-day aging on the right (b).

Thanks to this satisfactory first attempt, further tests were carried out using the bioplotting machine illustrated in Figure 8.3. It is important to point out that the slurry was used after mechanical mixing of the raw materials, without adding rheology modifier.

Needle: 1.36 mm

1 mm 2mm

Needle: 0.84 mm

500 m 2 mm

Figure 8.5: Microscope images of the direct ink printed scaffolds, using needles of different diameters, (a) and (b) printed with the 1.36 mm needle, while (c) and (d) with the 0.84 needle.

The geopolymer slurry, realized as described in chapter 4, was extruded through two needles of different size i.e. 1.36 mm and 0.84 mm. The microscope images are displayed in Figure 8.5. The ink was able to retain the original shape after each filament was deposited over the other, which confirms a quick viscosity recovery.

136

Advanced processing technologies for geopolymers

Despite this, after 28 days, the printed scaffolds, especially those extruded through the 1.36 needle, showed signs of collapse that should be improved. The small degree of collapse could be caused by weakly crosslinked structures comparable to those occurring in a soft gel or non-efficient material dispersion. Besides that, Figure 8.5 shows filaments with limited sagging and deformation. Moreover final scaffolds do not present cracks or surface defects. In Figure 8.5 (b) some pores in the structure are visible, probably caused by water evaporation, entrapped air during the preparation of the slurry or through filling of the cartridge with the slurry for the printing process. Moreover, the side lattice remains open with a square shape, especially using the 0.84 mm needle, which confirmed that the rheological properties of the ink were suitable for printing porous structures.

8.4 Interlocking pavers

Interlocking pavers have been used since the era of the Ancient Romans 2000 years ago to cover the floor of squares and pathways. Nowadays, they are applied to garden borders, pathways and garden steps. With the interlocking mechanism a paver is interlocked with the other. In this way the pavement is more stable. They are easy to install and repair235. Interlocking pavers can have different shape and color and most commonly they are made of concrete and aggregate but sometimes also of natural stone or clay236. The formulation 6S8M, developed extensively in laboratory scale in chapter 4, was synthesized in a larger scale to create an interlocking paver. The blend, made of 64 wt.% of fly ash, 36 wt.% of waste glass and 8M NaOH solution, was created with an industrial mix and subsequently cast in a mould. The sample was cured in an oven at circa 100°C for 2 days.

137

Chapter 8

a b

5 cm

Figure 8.6: Geopolymer interlocking pavers inside the mould (a) and outside the mould (b).

In Figure 8.6 (a), representing the final geopolymers inside the mould, it can be seen that the shrinkage is minimal. The mass loss, calculated by weighing the sample before and after the curing, amounted to just 10%. Figure 8.6 (b) shows the interlocking geopolymer paver: neither cracks nor bubbles are visible on the surface, which looks smooth. This work thus indicates the possibility to create interlocking pavers but also other building materials using waste materials and preserving in this way the use of virgin raw materials.

8.5 Conclusions

This study investigated the possibility of producing geopolymer parts by direct ink printing and machining waste-based geopolymers. The experiments demonstrated that fly ash-based geopolymers incorporating waste glass could be 3D printed to obtain porous grid-like components. This proof-of-concept test confirmed the suitability of the geopolymer ink with limited collapse and structural defects; future work should be done to link the processing parameters with the rheological properties of the ink to achieve specific 3D shapes. Moreover, the geopolymers were successfully worked using a lathe and drilled with standard metal tools to obtain different shapes with a good surface quality and dimensional accuracy. These first experiments showed the good machinability of the geopolymer material. Eventually,

138

Advanced processing technologies for geopolymers

a geopolymer sample was cast in a mould to realize an interlocking paver. The final material shows good surface finishing and complete absence of defects. Thanks to the precise design and the possibility to fabricate complex shapes, the range of applications of geopolymers will expand, especially in the field of hardmaterials where machining is a challenging process. Geopolymers offer the possibility to combine good mechanical properties, eco-friendly production and the possibility to apply alternative shaping and working techniques to produce complex shaped or intricate parts, as demonstrated in this chapter. Applications in mechanical engineering or in other sectors requiring designed components of certain structured integrity can be considered.

139

Chapter 9 Overall conclusions and future work

9.1 Conclusions

The present project was designed to develop different formulations of geopolymers incorporating waste materials to produce, at low temperatures and with an economic process, low environmental impact materials with competitive mechanical properties and satisfactory chemical stability. Aiming for low environmental impact and low raw materials cost the specific goal of this project became thus the fabrication of geopolymers using waste or by-product materials from different industrial sectors, which must otherwise be disposed of. In this way, the use of conventional chemical reagents such as water glass or virgin materials such as metakaolin, that are normally employed in geopolymer preparation, can be avoided. Within this work fly ash-based geopolymers incorporating red mud and soda lime waste glass were developed and characterized, with a particular focus on the correlation of the mechanical behavior with the composition and microstructure of the samples. The introduction of red mud and waste glass leads, in general, to a decrease in mechanical properties of fly ash-based geopolymers, however, despite this the results achieved are satisfactory for different applications. The leaching of heavy metals (from the used wastes) was also evaluated to confirm the stabilization of the heavy ions in the geopolymer matrix. Summing up the results from mechanical tests and microstructural analysis, it could be concluded that the loss of compressive strength, with the addition of waste glass and red mud, was mainly related to the remaining unreacted particles, in particular waste glass fragments. To promote first the dissolution of the waste glass cullet and further the incorporation of the silica in the geopolymer network a new methodology was studied. Sodium silicate, commonly called water glass, is a chemical reagent normally used during geopolymer synthesis. This water glass was completely removed from the process and substituted with waste glass. In this way, only sodium silicate solution is used as the alkaline solution and the silica required for geopolymerization comes from the dissolution of cullet glass. This approach is highly innovative. Previous literature

141

Chapter 9

indicates that either water glass has been used during the geopolymer synthesis or waste glass was dissolved in an alkaline solution and subsequently filtered without being introduced directly in the geopolymer material. In this research project different geopolymer formulations were developed changing the molarity of the sodium hydroxide solution and the molar ratio between silica and alumina. This molar ratio was changed modifying the amount of raw materials used during the synthesis, in particular, increasing the amount of waste glass leads to an increase of the molar ratio. Fly ash-based geopolymers employing fly ash and waste glass and red mud-based geopolymers using red mud and waste glass were successfully developed. This work demonstrated the possibility to create geopolymer materials avoiding the use of sodium silicate and instead using only waste materials as sources of alumina and silica. The formation of a geopolymer gel or rather a three- dimensional network structure, consisting of SiO4 and AlO4 tetrahedra, linked together by sharing oxygen atoms, was confirmed by NMR and EDX analyses. In both fly ash-based geopolymers and red mud-based geopolymers the alumina was mainly in tetrahedral coordination with Si/Al molar ratio in the exact range expected for the formation of a “zeolite-like” geopolymer structure. These results were confirmed also by means of XRD analysis reporting the formation of zeolite crystalline phases. Fly ash-based geopolymers presented outstanding mechanical properties reaching values as high as 45 MPa, even after a setting time of only 7 days confirming the characteristically short setting time of geopolymer materials. The mechanical properties of these geopolymers increased with the enhancement of the molarity of the sodium hydroxide solution until a maximum value of 8 molarity. In comparison to the values of molarity, normally cited in the literature, 8 is considered a rather low value, confirming the reduced environmental impact of the formulations developed. The mechanical values achieved are comparable to those of traditional Portland cement, normally used in construction industry. In addition, the experiments, contrary to what is reported in the literature, confirm the possibility to incorporate an amount of waste glass superior to 30 wt.%. Red mud-based geopolymers also demonstrated competitive compressive mechanical strength achieving 30-45 MPa. This relatively high compressive strength is

142

interesting considering the significant amount of red mud incorporated, approximately 60 wt%. Despite the relatively high amount of waste materials incorporated in the formulation, the leaching results were below the limit for non- hazardous materials, confirming the safe incorporation of heavy metals inside the geopolymeric matrix and the superior durability of the new geopolymer materials. With the formulations developed for the production of dense geopolymer materials, geopolymer foams were also realized in this project using an innovative “inorganic gel casting technique”. Foaming was introduced in the slurry through vigorous mechanical stirring, avoiding the introduction of porous agents. Fly ash-based geopolymer foams with an open porosity of 80% had a compressive strength around 2 MPa, while the foams developed with red mud and waste glass presented a lower open porosity, around 68%, and a compressive strength of 1.5 MPa. Fly ash- and red mud-based porous materials have a bulk density inferior to 1 g/cm3 confirming the light-weight character of the material. Due to the porous nature, the resistance to chemical attack and degradation is weaker in comparison to bulk materials and further tests should be carried out in order to improve the chemical durability of the foams. In addition to the normal characterization usually carried out to test the properties of geopolymer materials, alternative tests such as thermal shock and wear resistance tests were carried out on the fly ash-based geopolymers. The results showed a good resistance to thermal stress caused by quenching samples from elevated temperatures in water. The geopolymers tested did not present cracks after quenching in water in comparison to ceramic materials which normally fail catastrophically in this environment. After exposure to 800°C the geopolymers maintained a compressive resistance close to 10 MPa. Wear and abrasion resistance tests carried out, by pin-on- disc method, confirmed a wear resistance for geopolymer materials in the range of glass-ceramic materials. These results are important when evaluating the range of applications for these new materials that are still in the phase of research and development. To increase the range of applications for geopolymer materials, different processing technologies such as 3D printing and machining were studied. The results showed the possibility to extrude the fly ash-based geopolymer slurry to obtain a 3D structure by 3D printing. This proof-of-concept test confirmed the

143

Chapter 9

suitability of the geopolymer ink by produced 3D structures by additive manufacturing with limited collapse and structural defects. Moreover, for the first time, cylindrical geopolymer specimens were perforated and work on the lathe using metal tools to create components of specific shape with a good surface quality and dimensional accuracy. These innovative processes confirm that the relationship between hardness and toughness in the present geopolymers is suitable to machine the material with conventional tools. The geopolymer samples were first created in laboratory scale, given the success of the initial testing a further sample was produced at a larger scale equivalent to interlocking pavers. This experiment confirmed that the parameters fixed in small scale may also be easily applied to industrial applications. In summary, geopolymers incorporating relatively large amount of waste glass and fly ash or red mud were successfully developed using sodium hydroxide as the only virgin, raw material. The main output from this study is the possibility to create industry-ready geopolymers using fly ash, waste glass and red mud. The results represent an important perspective shift as by-products (in particular red mud) previously seen as landfill problems will instead be regarded as valuable raw materials. Red mud-based geopolymers developed without the use of sodium silicate present superior mechanical properties in comparison with geopolymers synthesized using sodium silicate27,28,30,52. Moreover, in comparison with the results reported in the literature237, geopolymers developed in this research project incorporated a larger amount of red mud (45% wt.) preserving the chemical stability. Likewise, the developed fly ash-based geopolymers, in comparison to geopolymers incorporating waste glass reported in the literature67,70, incorporate a much larger fraction of glass (46 wt.%) preserving suitable mechanical characteristics. Torres-Carrasco81 and Tchakoute25 already developed geopolymers substituting water glass with a solution realized by filtering waste glass after dissolution in distilled water or sodium hydroxide. In comparison to these works, the progress described in the previous chapters demonstrates in the capacity to incorporate the glass inside the structure without the need to filtering and subsequently disposing of this industrial waste glass.

144

Geopolymers developed using “waste” materials perform not only satisfactorily in terms of mechanical properties and chemical stability, but show the possibility to be foamed, extruded or machine worked to obtain different products. It can be concluded that not only an innovative class of geopolymers was developed, but they was linked with specific properties and machinability in order to expand the range of applications. The results contribute thus to developing mechanically sound geopolymer with large economic and environmental advantages for the construction field and in the ceramic industry in general.

9.2 Future directions

The development of fly ash-based and red mud-based geopolymers, which are the main output of the present research project, opens many possibilities for future research in this field. As these geopolymers are still in the initial phase of development many of their characteristics are still unknown. Future research will need to examine tensile strength and fracture toughness. Careful evaluation of the geopolymers’ mechanical properties will determine the opportunities for application in the building and construction industries. In this study, sodium hydroxide solutions were used to dissolve the raw materials due to the favorable price, but future research may also study the properties of the geopolymers when using potassium hydroxide, which is reported in the literature to perform with higher mechanical resistance. Additional studies on the developed geopolymer gel should be carried out through MAS-NMR analysis. In particular, peak deconvolution in Si-NMR could give a better understanding of the structure formed, at the same time Na-NMR could provide important information to analyze the structure. The preliminary thermal shock test carried out showed encouraging results, but more tests, particularly changing the number of thermal shock cycles, should be performed to ensure resistance to repetitive thermal stress, thermal fatigue, which may be relevant for applications involving high temperatures. Impact testing will also give more information about the mechanical resistance of the materials, for particular applications while cytotoxicity tests on waste-based

145

Chapter 9

geopolymers should be carried out to evaluate the environmental safety of these materials in a more conclusive way. The current geopolymers were synthesized mixing the powder raw materials with the alkaline solution, commonly called two-part-mixing geopolymer. It may prove interesting to evaluate the possibility of creating one-part-mixing geopolymers, producing geopolymers from a ready-mix precursor that can directly mix with water like Portland cement does. Porous geopolymers were realized using the same formulation developed for the dense samples. More studies should be carried out varying the content of raw materials or molarity of the activating solution if necessary, to obtain a valuable lightweight product with possible application as an acoustic or insulating panel. The rheology of the material should be accurately studied to control the size and pore distribution in the material. Besides that, and in particular with red mud, more tests should be run to optimize the curing temperature and time for improved chemical stability of the geopolymer foams. In terms of machining and 3D printing, only preliminary tests were carried out in this study. More extensive testing could improve the rheology of the ink to avoid the possible collapse of the 3D printed structure and to further increase the mechanical properties of the final structure. With the production of different shapes and components, new applications for geopolymers should be addressed. A study of possible applications and consequently of properties requested will be another important step in the development of geopolymers with the aim to expand geopolymer technologies for applications in mechanical engineering for example. The substitution of waste glass, for water glass, in combination with other by- products such as fly ash and red mud, is arguably the most innovative aspect of this study. This innovation could be applied to other geopolymers´ formulations incorporating metakaolin, or other types of wastes such as slags and bottom ash, among others.

146

Bibliography

1. Almenares R. S., Vizcaíno L. M., Damas S., Mathieu A., Alujas A. & Martirena F. Industrial calcination of kaolinitic clays to make reactive pozzolans. Case studies in construction materials, 2017, 6, 225-32. 2. Imbabi M. S., Carrigan C. & McKenna S. Trends and developments in green cement and concrete technology. International Journal of Sustainable Built Environment, 2012, 1, 194-216. 3. Okoye F., Durgaprasad J. & Singh N. Effect of silica fume on the mechanical properties of fly ash based-geopolymer concrete. Ceram. Int., 2016, 42, 3000-6. 4. Zhang J., Liu G., Chen B., Song D., Qi J. & Liu X. Analysis of CO2 emission for the cement manufacturing with alternative raw materials: a LCA-based framework. Energy Procedia, 2014, 61, 2541-5. 5. Bosoaga A., Masek O. & Oakey J. E. CO2 capture technologies for cement industry. Energy Procedia, 2009, 1, 133-40. 6. Díaz Y. C., Berriel S. S., Heierli U., et al. Limestone calcined clay cement as a low-carbon solution to meet expanding cement demand in emerging economies. Development Engineering, 2017, 2, 82-91. 7. Davidovits J. Geopolymer chemistry and applications. 3rd ed. Geopolymer Institute. Saint-Quentin France, 2011. 8. Ostwal T., Chitawadagi M. V. Experimental investigations on strength, durability, sustainability & economic characteristics of geo-polymer concrete blocks. International Journal of Research in Engineering and Technology, 2014, 3, 115- 22. 9. Duxson P., Provis J. L., Lukey G. C. & Van Deventer J. S. The role of inorganic polymer technology in the development of ‘green concrete’. Cem. Concr. Res., 2007, 37, 1590-7. 10. Toniolo N., Boccaccini A. R. Fly ash-based geopolymers containing added silicate waste. A review. Ceram. Int., 2017,43, 14545-14551. 11. Komnitsas K., Zaharaki D. Geopolymerisation: A review and prospects for the minerals industry. Minerals Eng, 2007, 20, 1261-77. 12. Van Jaarsveld J., Van Deventer J. & Lorenzen L. The potential use of geopolymeric materials to immobilise toxic metals: Part I. Theory and applications. Minerals Eng, 1997, 10, 659-69. 13. Xu H., Van Deventer J. The geopolymerisation of alumino-silicate minerals. Int. J. Miner. Process., 2000, 59, 247-66. 14. Temuujin J., Rickard W., Lee M. & van Riessen A. Preparation and thermal properties of fire resistant metakaolin-based geopolymer-type coatings. J. Non Cryst. Solids, 2011, 357, 1399-404. 15. Davidovits J. Geopolymers: ceramic-like inorganic polymers. Journal Of Ceramic Science and Technology, 2017, 8, 335-50. 16. www.webofknowledge.com 17. Luukkonen T., Abdollahnejad Z., Yliniemi J., Kinnunen P. & Illikainen M. One- part alkali-activated materials: A review. Cem. Concr. Res., 2017, 103, 21-34.

147

18. Abdul Rahim R., Rahmiati T., Azizli K. A., Man Z., Nuruddin M. F. & Ismail L. Comparison of using NaOH and KOH Activated Fly Ash-based Geopolymer on the Mechanical Properties. In Materials Science Forum, ed. Anonymous, 2014, 803, 179-184. 19. Ye N., Yang J., Ke X., et al. Synthesis and characterization of geopolymer from Bayer red mud with thermal pretreatment. J Am Ceram Soc, 2014, 97, 1652-60. 20. Rill E., Lowry D. & Kriven W. M. Properties of basalt fiber reinforced geopolymer composites. Strategic Materials and Computational Design: Ceramic Engineering and Science Proceedings, Volume 31, 2010, 57-67. 21. Ribeiro R., Sa Ribeiro M. & Kriven W. A Review of Particle-and Fiber- Reinforced Metakaolin-Based Geopolymer Composites. Journal of Ceramic Science and Technology, 2017, 8, 307-22. 22. Kourti, I., Rani, D. A., Deegan, D., Cheeseman, C. R., & Boccaccini, A. R. Development of geopolymers from plasma vitrified air pollution control residues from energy from waste plants. Design, Development, and Applications of Engineering Ceramics and Composites, 2010, 295-304. 23. Kumar S., Kumar R. & Mehrotra S. Influence of granulated blast furnace slag on the reaction, structure and properties of fly ash based geopolymer. J. Mater. Sci., 2010, 45, 607-15. 24. Allahverdi A., Kani E. N. & Yazdanipour M. Effects of blast-furnace slag on natural pozzolan-based geopolymer cement. Ceramics-Silikáty, 2011, 55, 68-78. 25. Tchakouté H. K., Rüscher C. H., Kong S., Kamseu E. & Leonelli C. Geopolymer binders from metakaolin using sodium waterglass from waste glass and rice husk ash as alternative activators: A comparative study. Constr. Build. Mater., 2016, 114, 276-89. 26. Kua T., Arulrajah A., Horpibulsuk S., Du Y. & Shen S. Strength assessment of spent coffee grounds-geopolymer cement utilizing slag and fly ash precursors. Constr. Build. Mater., 2016, 115, 565-75. 27. Zhang G., He J., Gambrell R., Synthesis, Characterization, and Mechanical Properties of Red Mud-Based Geopolymers. Transp. Res. Rec. J. Transp. Res. Board., 2010, 2167, 1-9. 28. He J., Zhang G. Geopolymerization of red mud and fly ash for civil infrastructure applications. In Geo-Frontiers 2011: Advances in Geotechnical Engineering, ed. Anonymous 2011, 1287-1296. 29. Zhang M., Zhao M., Zhang G., Mann D., Lumsden K. & Tao M. Durability of red mud-fly ash based geopolymer and leaching behavior of heavy metals in sulfuric acid solutions and deionized water. Constr. Build. Mater., 2016, 124, 373-82. 30. He J., Jie Y., Zhang J., Yu Y. & Zhang G. Synthesis and characterization of red mud and rice husk ash-based geopolymer composites. Cement and Concrete Composites, 2013, 37, 108-18. 31. Al Bakri A. M., Abdulkareem O. A., Rafiza A., Zarina Y., Norazian M. & Kamarudin H. Review on Processing of low calcium fly ash geopolymer concrete. Australian Journal of Basic and Applied Sciences, 2013, 7, 342-9. 32. Ramezanianpour A. A., Cement Replacement Materials: Properties, Durability, Sustainability. Spring, 2013.

148

33. Fernández-Jiménez A., Palomo A. Characterisation of fly ashes. Potential reactivity as alkaline cements. Fuel, 2003, 82, 2259-65. 34. Fulekar M., Dave J. Disposal of fly ash—an environmental problem. Int. J. Environ. Stud., 1986, 26, 191-215. 35. Ye T. M., Wang W., Gao X. B., Wan X. & Wang F. Characterization and heavy metals leaching toxicity of fly ash from municipal solid waste incinerators in China. Huan Jing Ke Xue., 2007, 28, 2646-50. 36. ASTM C 618-05, Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use in Concrete, 2005. 37. Heidrich C., Feuerborn H. & Weir A. Coal combustion products: a global perspective. In Proceedings of the World of Coal Ash Conference, ed. Anonymous 2013, 22-25. 38. Sear, L. K. Properties and use of coal fly ash: a valuable industrial by-product. Thomas Telford, 2001. 39. Jha B., Singh D. N. Fly ash zeolites: innovations, applications, and directions, Springer, 2016. 40. Fernández-Jiménez A., De La Torre A., Palomo A., López-Olmo G., Alonso M. & Aranda M. Quantitative determination of phases in the alkali activation of fly ash. Part I. Potential ash reactivity. Fuel, 2006, 85, 625-34. 41. Duxson P., Fernández-Jiménez A., Provis J. L., Lukey G. C., Palomo A. & van Deventer J. S. Geopolymer technology: the current state of the art. J. Mater. Sci., 2007, 42, 2917-33. 42. Alehyen S., Achouri M. & Taibi M. Characterization, microstructure and properties of fly ash-based geopolymer. Journal of materials and environmental sciences, 2017, 8, 21-34. 43. Gomes H. I., Mayes W. M., Rogerson M., Stewart D. I. & Burke I. T. Alkaline residues and the environment: a review of impacts, management practices and opportunities. J. Clean. Prod., 2016, 112, 3571-82. 44. Kumar A., Kumar S. Development of paving blocks from synergistic use of red mud and fly ash using geopolymerization. Constr. Build. Mater., 2013, 38, 865- 71. 45. Pan Z., Li D., Yu J. & Yang N. Properties and microstructure of the hardened alkali-activated red mud–slag cementitious material. Cem. Concr. Res., 2003, 33, 1437-41. 46. Wang P., Liu D. Physical and chemical properties of sintering red mud and bayer red mud and the implications for beneficial utilization. Materials, 2012, 5, 1800- 10. 47. Hairi M., Syazana S. N. Synthesis and Applications of Inorganic Materials Derived from Aluminium Smelting Waste. Victoria University of Wellingtonm 2014. 48. Ye N., Chen Y., Yang J., et al. Co-disposal of MSWI fly ash and Bayer red mud using an one-part geopolymeric system. J. Hazard. Mater., 2016, 318, 70-8. 49. Zhang M., El-Korchi T., Zhang G., Liang J. & Tao M. Synthesis factors affecting mechanical properties, microstructure, and chemical composition of red mud–fly ash based geopolymers. Fuel, 2014, 134, 315-25.

149

50. Gonda1–Zsombor A. K., Székely1–Balázs F. T. & Kovács Z. M. Physical and chemical characterization of red mud in terms of its environmental effects. Geosciences and Engineering, 2014, 3, 129-137. 51. Hajjaji W., Andrejkovičová S., Zanelli C., et al. Composition and technological properties of geopolymers based on metakaolin and red mud. Materials & Design (1980-2015), 2013, 52, 648-54. 52. He J., Zhang J., Yu Y. & Zhang G. The strength and microstructure of two geopolymers derived from metakaolin and red mud-fly ash admixture: a comparative study. Constr. Build. Mater., 2012, 30, 80-91. 53. Mucsi G., Szabo R., Racz Á., Molnar Z., Kristály F. & Kumar S. Influence of red mud on the properties of geopolymer derived from mechanically activated lignite fly ash. In Proceedings of the Bauxite Residue Valorisation and Best Practices Conference, 5–7/10/2015, Leuven, ed. Anonymous 2015, 211-217. 54. Kaya K., Soyer-Uzun S. Evolution of structural characteristics and compressive strength in red mud–metakaolin based geopolymer systems. Ceram. Int., 2016, 42, 7406-13. 55. Bonifazi G., Serranti S. Imaging spectroscopy based strategies for ceramic glass contaminants removal in glass recycling. Waste Manage., 2006, 26, 627-39. 56. Dhir R. K., Dyer T. D. Recycling and reuse of glass cullet: proceedings of the international symposium organised by the Concrete Technology Unit and held at the University of Dundee, Scotland, UK on 19-20 March 2001, Thomas Telford, 2001. 57.http://ec.europa.eu/environment/life/project/Projects/index.cfm?fuseaction=home.s howFile&rep=file&fil=MEIGLASS-project.pdf 58. Rincón A., Giacomello G., Pasetto M. & Bernardo E. Novel ‘inorganic gel casting’process for the manufacturing of glass foams. Journal of the European Ceramic Society, 2017, 37, 2227-34. 59. Mohajerani A., Vajna J., Cheung T. H. H., Kurmus H., Arulrajah A. & Horpibulsuk S. Practical recycling applications of crushed waste glass in construction materials: A review. Constr. Build. Mater., 2017, 156, 443-67. 60. Farcomeni A., Serranti S. & Bonifazi G. Non-parametric analysis of infrared spectra for recognition of glass and glass ceramic fragments in recycling plants. Waste Manage., 2008, 28, 557-64. 61.https://www.inabottle.it/en/news/italy-recycles-70-its-glass 62.http://www.lastampa.it/2017/10/10/scienza/ambiente/focus/italia-prima-in-europa- per-riciclo-rifiuti-al-WrOlRC1h3RzXnQChxI2MdM/pagina.html 63.http://ec.europa.eu/environment/life/project/Projects/index.cfm?fuseaction=search. dspPage&n_proj_id=3067&docType=pdf 64. Cyr M., Idir R. & Poinot T. Properties of inorganic polymer (geopolymer) mortars made of glass cullet. J. Mater. Sci., 2012, 47, 2782-97. 65. Rashidian-Dezfouli H., Rangaraju P. R. & Kothala V. S. K. Influence of selected parameters on compressive strength of geopolymer produced from ground glass fiber. Constr. Build. Mater., 2018, 162, 393-405. 66. Taveri G., Tousek J., Bernardo E., Toniolo N., Boccaccini A. & Dlouhy I. Proving the role of boron in the structure of fly-ash/borosilicate glass based geopolymers. Mater Lett, 2017, 200, 105-8.

150

67. Novais R. M., Ascensão G., Seabra M. & Labrincha J. Waste glass from end-of- life fluorescent lamps as raw material in geopolymers. Waste Manage., 2016, 52, 245-55. 68. Kourti I., Devaraj A. R., Bustos A. G., Deegan D., Boccaccini A. R. & Cheeseman C. R. Geopolymers prepared from DC plasma treated air pollution control (APC) residues glass: properties and characterisation of the binder phase. J. Hazard. Mater., 2011, 196, 86-92. 69. Lin K., Shiu H., Shie J., Cheng T. & Hwang C. Effect of composition on characteristics of thin film transistor liquid crystal display (TFT-LCD) waste glass-metakaolin-based geopolymers. Constr. Build. Mater., 2012, 36, 501-7. 70. Bobirică C., Shim J., Pyeon J. & Park J. Influence of waste glass on the microstructure and strength of inorganic polymers. Ceram. Int., 2015, 41, 13638- 49. 71. Torres-Carrasco M., Palomo J. & Puertas F. Sodium silicate solutions from dissolution of glasswastes. Statistical analysis. Materiales de Construcción, 2014, 64, 014. 72. Torres-Carrasco M., Puertas F. Waste glass in the geopolymer preparation. Mechanical and microstructural characterisation. J. Clean. Prod., 2015, 90, 397- 408. 73. Torres-Carrasco M., Rodríguez-Puertas C., del Mar Alonso M. & Puertas F. Alkali activated slag cements using waste glass as alternative activators. Rheological behaviour. Boletín de la Sociedad Española de Cerámica y Vidrio, 2015, 54, 45-57. 74. Panagiotopoulou C., Kontori E., Perraki T. & Kakali G. Dissolution of aluminosilicate minerals and by-products in alkaline media. J. Mater. Sci., 2007, 42, 2967-73. 75. Sperinck S. Metakaolin as a model system for understanding geopolymers, PhD Thesis. Curtin University. 2012, 76. Khale D., Chaudhary R. Mechanism of geopolymerization and factors influencing its development: a review. J. Mater. Sci., 2007, 42, 729-46. 77. Duxson P., Lukey G., Separovic F. & Van Deventer J. Effect of alkali cations on aluminum incorporation in geopolymeric gels. Ind Eng Chem Res, 2005, 44, 832- 9. 78. Provis J. L., Van Deventer, Jan Stephanus Jakob. Geopolymers: structures, processing, properties and industrial applications, Elsevier, 2009, 79. Gilliam, T. M., & Wiles, C. C. (1996). Stabilization and solidification of hazardous, radioactive, and mixed wastes. ASTM. 80. Shi, C., Roy, D., & Krivenko, P. Alkali-activated cements and concretes. CRC press, 2003. 81. Torres Carrasco M. Reutilización de residuos vítreos urbanos industriales en la fabricación de cementos alcalinos: Activación, comportamiento y durabilidad. PhD Thesis, Instituto de Ciencias de la Construcción Eduardo Torroja, 2015. 82. Fernández-Jiménez A., Palomo A. & Criado M. Alkali activated fly ash binders. A comparative study between sodium and potassium activators. Materiales de Construcción, 2006, 56, 51-65..

151

83. Phair J., Van Deventer J. Characterization of fly-ash-based geopolymeric binders activated with sodium aluminate. Ind Eng Chem Res, 2002, 41, 4242-51. 84. Brew D., MacKenzie K. Geopolymer synthesis using silica fume and sodium aluminate. J. Mater. Sci., 2007, 42, 3990-3. 85. Hajimohammadi A., Provis J. L. & Van Deventer J. S. One-part geopolymer mixes from geothermal silica and sodium aluminate. Ind Eng Chem Res, 2008, 47, 9396-405. 86. Zuhua, Z., Xiao, Y., Huajun, Z., & Yue, C. Role of water in the synthesis of calcined kaolin-based geopolymer. Applied Clay Science, 2009, 43(2), 218-223. 87. Vora P. R., Dave U. V. Parametric studies on compressive strength of geopolymer concrete. Procedia Engineering, 2013, 51, 210-9. 88. Kupaei R. H., Alengaram U. J. & Jumaat M. Z. The effect of different parameters on the development of compressive strength of oil palm shell geopolymer concrete. ScientificWorldJournal, 2014, 2014, 898536. 89. Satpute Manesh B., Wakchaure Madhukar R. & Patankar Subhash V. Effect of duration and temperature of curing on compressive strength of geopolymer concrete. International Journal of Engineering and Innovative Technology (IJEIT) Volume, 2012, 1, 90. Ranjbar N., Mehrali M., Alengaram U. J., Metselaar H. S. C. & Jumaat M. Z. Compressive strength and microstructural analysis of fly ash/palm oil fuel ash based geopolymer mortar under elevated temperatures. Constr. Build. Mater., 2014, 65, 114-21. 91. Zuhua Z., Xiao Y., Huajun Z. & Yue C. Role of water in the synthesis of calcined kaolin-based geopolymer. Appl. Clay. Sci., 2009, 43, 218-23. 92. Abdullah, M. M. A., Hussin, K., Bnhussain, M., Ismail, K. N., & Ibrahim, W. M. W.Mechanism and chemical reaction of fly ash geopolymer cement-a review. Int. J. Pure Appl. Sci. Technol. 2011, 6(1), 35-44. 93. Davidovits J. Geopolymers: inorganic polymeric new materials. Journal of Thermal Analysis and calorimetry, 1991, 37, 1633-56. 94. Weng L., Sagoe-Crentsil K. Dissolution processes, hydrolysis and condensation reactions during geopolymer synthesis: Part I—Low Si/Al ratio systems. J. Mater. Sci., 2007, 42, 2997-3006. 95. Kanuchova M., Kozakova L., Drabova M., et al. Monitoring and characterization of creation of geopolymers prepared from fly ash and metakaolin by X-ray photoelectron spectroscopy method. Environmental progress & sustainable energy, 2015, 34, 841-9. 96. Provis J., Duxson P., Van Deventer J. & Lukey G. The role of mathematical modelling and gel chemistry in advancing geopolymer technology. Chem. Eng. Res. Design, 2005, 83, 853-60. 97. Van Deventer J., Provis J., Duxson P. & Lukey G. Reaction mechanisms in the geopolymeric conversion of inorganic waste to useful products. J. Hazard. Mater., 2007, 139, 506-13. 98. Yao X., Zhang Z., Zhu H. & Chen Y. Geopolymerization process of alkali– metakaolinite characterized by isothermal calorimetry. Thermochimica Acta, 2009, 493, 49-54.

152

99. Ducman V., Korat L. Characterization of geopolymer fly-ash based foams obtained with the addition of Al powder or H2O2 as foaming agents. Mater Charact, 2016, 113, 207-13. 100. Łach M., Korniejenko K. & Mikuła J. Thermal insulation and thermally resistant materials made of geopolymer foams. Procedia Engineering, 2016, 151, 410-6. 101. Abdullah, Mohd Mustafa Al Bakri, Hussin K., Bnhussain M., Ismail K. N., Yahya Z. & Abdul Razak R. Fly ash-based geopolymer lightweight concrete using foaming agent. International journal of molecular sciences, 2012, 13, 7186-98. 102. Masi G., Rickard W. D., Vickers L., Bignozzi M. C. & Van Riessen A. A comparison between different foaming methods for the synthesis of light weight geopolymers. Ceram. Int., 2014, 40, 13891-902. 103. Hlaváček P., Šmilauer V., Škvára F., Kopecký L. & Šulc R. Inorganic foams made from alkali-activated fly ash: Mechanical, chemical and physical properties. Journal of the European Ceramic Society, 2015, 35, 703-9. 104. Zhuang X. Y., Chen L., Komarneni S., et al. Fly ash-based geopolymer: clean production, properties and applications. J. Clean. Prod., 2016, 125, 253-67. 105. Malkawi A. B., Nuruddin M. F., Fauzi A., Almattarneh H. & Mohammed B. S. Effects of Alkaline Solution on Properties of the HCFA Geopolymer Mortars. Procedia engineering, 2016, 148, 710-7. 106. Ekaputri J. J., Priyanka N. F. The Effect of Temperature Curing on Geopolymer Concrete. In MATEC Web of Conferences, ed. Anonymous EDP Sciences, 2017, 01005. 107. Wang H., Li H. & Yan F. Synthesis and mechanical properties of metakaolinite- based geopolymer. Colloids Surf. Physicochem. Eng. Aspects, 2005, 268, 1-6. 108. Fernández-Jiménez A., Palomo Á. Propiedades y aplicaciones de los cementos alcalinos. Revista ingeniería de construcción, 2009, 24, 213-32. 109. Zhang H., Kodur V., Cao L. & Qi S. Fiber reinforced geopolymers for fire resistance applications. Procedia Engineering, 2014, 71, 153-8. 110. Alomayri T., Shaikh F. & Low I. M. Synthesis and mechanical properties of cotton fabric reinforced geopolymer composites. Composites Part B: Engineering, 2014, 60, 36-42. 111. Alzeer M., MacKenzie K. J. Synthesis and mechanical properties of new fibre- reinforced composites of inorganic polymers with natural wool fibres. J. Mater. Sci., 2012, 47, 6958-65. 112. Davidovits J., Comrie D. C., Paterson J. H., & Ritcey, D. J. Geopolymeric concretes for environmental protection. Concrete International, 1990, 12(7), 30- 40. 113. Palomo A., Blanco-Varela M. T., Granizo M., Puertas F., Vazquez T. & Grutzeck M. Chemical stability of cementitious materials based on metakaolin. Cem. Concr. Res., 1999, 29, 997-1004. 114. Palomo A., Grutzeck M. & Blanco M. Alkali-activated fly ashes: a cement for the future. Cem. Concr. Res., 1999, 29, 1323-9. 115. Shi D., Shi L., Zhang J. Y. & Cheng J. Preparation and properties of a novel nonflammable thermal insulation material. In Advanced Materials Research, ed. Anonymous Trans Tech Publ, 2012, 1504-1512.

153

116. Bakharev T., Sanjayan J. & Cheng Y. Sulfate attack on alkali-activated slag concrete. Cem. Concr. Res., 2002, 32, 211-6. 117. Pawlasová S., Skavara F. High-temperature properties of geopolymer materials. In Alkali activated materials-research, production and utilization 3rd conference, Prague, Czech Republic, ed. Anonymous 2007, 524. 118. Rovnaník P., Bayer P. & Rovnaníková P. Characterization of alkali activated slag paste after exposure to high temperatures. Constr. Build. Mater., 2013, 47, 1479-87. 119. Shaikh F., Vimonsatit V. Compressive strength of fly ash-based geopolymer concrete at elevated temperatures. Fire Mater., 2015, 39, 174-88. 120. Degirmenci F. N., Freeze-thaw and fire resistance of geopolymer mortar based on natural and waste pozzolans. Ceramics–Silikáty, 2018, 62, 41-9. 121. Sun P., Wu H. Chemical and freeze–thaw resistance of fly ash-based inorganic mortars. Fuel, 2013, 111, 740-5. 122. Gifford P. M., Gillott J. E. Freeze-thaw durability of activated blast furnace slag cement concrete. Materials Journal, 1996, 93, 242-5. 123. Ramujee K., Potharaju M. Abrasion resistance of geopolymer composites. Procedia Materials Science, 2014, 6, 1961-6. 124. Wang H., Li H. & Yan F. Reduction in wear of metakaolinite-based geopolymer composite through filling of PTFE. Wear, 2005, 258, 1562-6. 125. Qian G., Yang X., Dong S., et al. Stabilization of chromium-bearing electroplating sludge with MSWI fly ash-based Friedel matrices. J. Hazard. Mater., 2009, 165, 955-60. 126. El-Eswed B., Yousef R., Alshaaer M., Hamadneh I., Al-Gharabli S. & Khalili F. Stabilization/solidification of heavy metals in kaolin/zeolite based geopolymers. Int. J. Miner. Process., 2015, 137, 34-42. 127. Guo X., Hu W. & Shi H. Microstructure and self-solidification/stabilization (S/S) of heavy metals of nano-modified CFA–MSWIFA composite geopolymers. Constr. Build. Mater., 2014, 56, 81-6. 128. Van Jaarsveld J., Van Deventer J. & Lorenzen L. The potential use of geopolymeric materials to immobilise toxic metals: Part I. Theory and applications. Minerals Eng, 1997, 10, 659-69. 129. Zheng L., Wang W. & Shi Y. The effects of alkaline dosage and Si/Al ratio on the immobilization of heavy metals in municipal solid waste incineration fly ash-based geopolymer. Chemosphere, 2010, 79, 665-71. 130. Van Jaarsveld J., Van Deventer J. The effect of metal contaminants on the formation and properties of waste-based geopolymers. Cem. Concr. Res., 1999, 29, 1189-200. 131. Komnitsas K., Zaharaki D. & Bartzas G. Effect of sulphate and nitrate anions on heavy metal immobilisation in ferronickel slag geopolymers. Appl. Clay. Sci., 2013, 73, 103-9. 132. Nikolici I., Durovic D., Tadic M., Blecic D. & Radmilovic V. Immobilization of zinc from metallurgical waste and water solutions using geopolymerization technology. In E3S Web of Conferences, ed. Anonymous EDP Sciences, 2013,

154

133. Phair J., Van Deventer J. & Smith J. Effect of Al source and alkali activation on Pb and Cu immobilisation in fly-ash based “geopolymers”. Appl. Geochem., 2004, 19, 423-34. 134. Waijarean N., Asavapisit S. & Sombatsompop K. Strength and microstructure of water treatment residue-based geopolymers containing heavy metals. Constr. Build. Mater., 2014, 50, 486-91. 135. Lee S., Van Riessen A., Chon C., Kang N., Jou H. & Kim Y. Impact of activator type on the immobilisation of lead in fly ash-based geopolymer. J. Hazard. Mater., 2016, 305, 59-66. 136. Zhang J., Provis J. L., Feng D. & van Deventer J. S. Geopolymers for immobilization of Cr6 , Cd2 , and Pb2. J. Hazard. Mater., 2008, 157, 587-98. 137. Paramguru, R. K., P. C. Rath, and V. N. Misra. Trends in red mud utilization–a review. Mineral Processing & Extractive Metall., 2004, 26.1,1-29. 138. Liu X., Zhang N. Utilization of red mud in cement production: a review. Waste Manage. Res., 2011, 29, 1053-63. 139. http://www.europarl.europa.eu/EPRS/EPRS-Briefing-564398-Understanding- waste-streams-FINAL.pdf 140. Kupaei R. H., Alengaram U. J. & Jumaat M. Z. The effect of different parameters on the development of compressive strength of oil palm shell geopolymer concrete. ScientificWorldJournal, 2014, 2014, 898536. 141. Bobirică C., Shim J., Pyeon J. & Park J. Influence of waste glass on the microstructure and strength of inorganic polymers. Ceram. Int., 2015, 41, 13638-49. 142. Toniolo N., Taveri G., Hurle K., et al. Fly-Ash-Based Geopolymers: How the Addition of Recycled Glass or Red Mud Waste Influences the Structural and Mechanical Properties. Journal of Ceramic Science and Technology, 2017, 8, 411-20. 143. Zhang M., El-Korchi T., Zhang G., Liang J. & Tao M. Synthesis factors affecting mechanical properties, microstructure, and chemical composition of red mud–fly ash based geopolymers. Fuel, 2014, 134, 315-25. 144 Jansen, D., Goetz-Neunhoeffer, F., Stabler, C., & Neubauer, J. A remastered external standard method applied to the quantification of early OPC hydration. Cement and Concrete Research, 2011, 41(6), 602-608. 145 International Union for Crystallography, Prince E., International Tables for Crystallography, Volume C. Mathematical Physical and Chemical Tables. Third Ed.,Wiley, 2004. 146. Ban T., Okada K. Structure refinement of mullite by the Rietveld method and a new method for estimation of chemical composition. J Am Ceram Soc, 1992, 75, 227-30. 147. Le Page Y., Donnay G. Refinement of the crystal structure of low-quartz. Acta Crystallographica Section B: Structural Crystallography and Crystal Chemistry, 1976, 32, 2456-9. 148. Sawada H. An electron density residual study of α-ferric oxide. Mater. Res. Bull., 1996, 31, 141-6. 149. Bragg W. The structure of magnetite and the spinels. Nature, 1915, 95, 561.

155

150. Kirfel A., Will G. Charge density in anhydrite, CaSO4, from X-ray and neutron diffraction measurements. Acta Crystallographica Section B: Structural Crystallography and Crystal Chemistry, 1980, 36, 2881-90. 151. Sasaki S, Fujino K, Takeuchi Y. X-ray determination of electron-density distributions in oxides, MgO, MnO, CoO, and NiO, and atomic scattering factors of their constituent atoms. Proc Japan Acad , Ser B 1979;55:43-8. 152. Shen C, Liu R, Lin J, Huang C. Phase stability study of La1. 2Ca1. 8Mn2O7. Mater Res Bull 2001;36:1139-48. 153. Yang H. X., Ren L., Downs R.T., Goethite, alpha-FeO(OH), from single-crystal data, Acta Cryst., 2006, vol. E62, i205-i252. 154. Smolin Y. I., Shepelev Y. F., Butikova I. K., The crystal structure of cancrinite, Kristallografiya, 1981, 26, 63-66. 155. Saalfeld H, Wedde M. Refinement of the crystal structure of gibbsite, A1 (OH) 3. Zeitschrift für Kristallographie-Crystalline Materials 1974;139:129-35. 156. Farkas L., Gado P., Werner P.E., The structure refinement of boehmite (AlO(OH)-gamma) and the study of ist structural variability based on Guinier – Haegg power data, Mater. Res. Bull. 1977, 12, 1213-1219. 157. Abrahams S, Bernstein J. Rutile: normal probability plot analysis and accurate measurement of crystal structure. J Chem Phys 1971;55:3206-11. 158. Jost K, Ziemer B, Seydel R. Redetermination of the structure of β-dicalcium silicate. Acta Crystallographica Section B: Structural Crystallography and Crystal Chemistry 1977;33:1696-700. 159. Withers R, Thompson J. Modulation wave approach to the structural parameterization and Rietveld refinement of low carnegieite. Acta Crystallographica Section B: Structural Science 1993;49:614-26. 160. Gatineau L., Localisation des replacements isomorphiques dans la muscovite, C. R. Hebd. Seances Acad. Sci., 1963, 256, 4648-4649. 161. Phillipsi M, Colville A, Ribbe P. The crystal structures of two oligoclases: A comparison with low and high albite. Zeitschrift für Kristallographie-Crystalline Materials 1971;133:43-65. 162. Blasi A, De Pol Blasi C, Zanazzi PF. A re-examination of the Pellotsalo microcline; mineralogical implications and genetic considerations. The Canadian Mineralogist 1987;25:527-37. 163. Maslen E, Streltsov V, Streltsova N, Ishizawa N. Electron density and optical anisotropy in rhombohedral carbonates. III. Synchrotron X-ray studies of CaCO3, MgCO3 and MnCO3. Acta Crystallographica Section B: Structural Science 1995;51:929-39. 164. Temuujin J, Williams R, Van Riessen A. Effect of mechanical activation of fly ash on the properties of geopolymer cured at ambient temperature. J Mater Process Technol 2009;209:5276-80. 165. Yang H, Jiang L, Zhang Y, Pu Q. Flexural strength of cement paste beam under chemical degradation: Experiments and simplified modeling. J Mater Civ Eng 2012;25:555-62. 166. Pernica D, Reis P, Ferreira J, Louda P. Effect of test conditions on the bending strength of a geopolymer-reinforced composite. J Mater Sci 2010;45:744.

156

167. Dlouhý I, Boccaccini A. Reliability of the chevron notch technique for fracture toughness determination in glass composites reinforced by continuous fibres. Scr Mater 2001;44:531-7. 168. Boccaccini A, Rawlings R, Dlouhý I. Reliability of the chevron-notch technique for fracture toughness determination in glass. Materials Science and Engineering: A 2003;347:102-8. 169. White, K. W., Bradt, R. C., Sakai, M., & Munz, D. (Eds.). (2005). Fracture Mechanics of Ceramics: Active Materials, Nano-scale Materials, Composites, Glass, and Fundamentals. Springer. 170. Ghorbal GB, Tricoteaux A, Thuault A, Louis G, Chicot D. Comparison of conventional Knoop and Vickers hardness of ceramic materials. Journal of the European Ceramic Society 2017;37:2531-5. 171. Ziegler D, Formia A, Tulliani J, Palmero P. Environmentally-friendly dense and porous geopolymers using fly ash and rice husk ash as raw materials. Materials 2016, 9, 466. 172. ASTM (2012) C39: Standard test method for compressive strength of cylindrical concrete specimens. ASTM International, West Conshohoken, PA, USA 173. Bobirică C, Shim J, Pyeon J, Park J. Influence of waste glass on the microstructure and strength of inorganic polymers. Ceram Int 2015;41:13638- 49. 174. Sarker PK, Haque R, Ramgolam KV. Fracture behaviour of heat cured fly ash based geopolymer concrete. Mater Des 2013;44:580-6. 175. Kamseu E, Bignozzi MC, Melo U, Leonelli C, Sglavo VM. Design of inorganic polymer cements: Effects of matrix strengthening on microstructure. Constr Build Mater 2013;38:1135-45. 176. Fisher-Cripps, A., and IBIS Handbook Of Nanoindentation Book. Fisher-Cripps Laboratories Pty. Limited. New South Wales, 2009, 4-15. 177. Quinn GD, Gettings R, Ives LK. A standard reference material for Vickers hardness of ceramics and hardmetals. JIS, 2000,45,2831. 178. Gonzaga C. C., Cesar P. F., Miranda W. G.,Jr & Yoshimura H. N. Slow crack growth and reliability of dental ceramics. Dent. Mater., 2011, 27, 394-406. 179. Lange D. A., Jennings H. M. & Shah S. P. Relationship between fracture surface roughness and fracture behavior of cement paste and mortar. J Am Ceram Soc, 1993, 76, 589-97. 180. Council Decision. 33/EC of 19 December 2002 establishing criteria and procedures for the acceptance of waste at landfills pursuant to Article 16 of and Annex II to Directive 1999/31/EC, Off. J. Europ. Communities, (2003) L11/27- L11749. 181. Davidovits J. Properties of geopolymer cements. In First international conference on alkaline cements and concretes, ed. Anonymous Scientific Research Institute on Binders and Materials Kiev State Technical University, Ukraine, 1994, pp. 131-149. 182. http://www.coach-etn.eu/ 183. Rattanasak U., Chindaprasirt P. Influence of NaOH solution on the synthesis of fly ash geopolymer. Minerals Eng, 2009, 22, 1073-8.

157

184. Barbosa V. F., MacKenzie K. J. Thermal behaviour of inorganic geopolymers and composites derived from sodium polysialate. Mater. Res. Bull., 2003, 38, 319-31. 185. Lee Y., Peng Y. L. & Tomozawa M. IR reflection spectroscopy of a soda-lime glass surface during ion-exchange. J. Non Cryst. Solids, 1997, 222, 125-30. 186. Longhi MA, Rodríguez ED, Bernal SA, Provis JL, Kirchheim AP. Valorisation of a kaolin mining waste for the production of geopolymers. J Clean Prod 2016, 115, 265-72. 187. Kramar S, Ducman V. Mechanical and microstructural characterization of geopolymer synthesized from low calcium fly ash. Chemical Industry and Chemical Engineering Quarterly 2015, 21, 13-22. 188. Kim E., Understanding effects of silicon/aluminum ratio and calcium hydroxide on chemical composition, nanostructure and compressive strength for metakaolin geopolymers, Master Thesis, University of Illinois at Urbana- Champaign, 2012. 189. Sindhunata , Van Deventer J, Lukey G, Xu H. Effect of curing temperature and silicate concentration on fly-ash-based geopolymerization. Ind Eng Chem Res 2006;45:3559-68. 190. Tamburini S, Natali M, Garbin E, Panizza M, Favaro M, Valluzzi MR. Geopolymer matrix for fibre reinforced composites aimed at strengthening masonry structures. Constr Build Mater 2017, 141, 542-52. 191. Fapohunda C, Akinbile B, Shittu A. Structure and properties of mortar and concrete with rice husk ash as partial replacement of ordinary Portland cement– A review. International Journal of Sustainable Built Environment 2017. 192. Živica V, Palou MT, Križma M. Geopolymer Cements and Their Properties: A Review. Building Research Journal 2014, 61,85-100. 193. Arulrajah A, Kua T, Horpibulsuk S, Phetchuay C, Suksiripattanapong C, Du Y. Strength and microstructure evaluation of recycled glass-fly ash geopolymer as low-carbon masonry units. Constr Build Mater 2016, 114,400-6. 194. Kong DL, Sanjayan JG, Sagoe-Crentsil K. Comparative performance of geopolymers made with metakaolin and fly ash after exposure to elevated temperatures. Cem Concr Res 2007, 37, 1583-9. 195. Kamarudin H, Al Bakri AM, Binhussain M, Ruzaidi C, Luqman M, Heah C et al. Preliminary study on effect of NaOH concentration on early age compressive strength of kaolin-based green cement. 2011,10 ,18-24. 196. En X, Zhang L, Ramey D, Waterman B, Ormsby S. Utilization of aluminum sludge (AS) to enhance mine tailings-based geopolymer. J Mater Sci 2015, 50, 1370-81. 197. Rowles MR, O'Connor BH. Chemical and structural microanalysis of aluminosilicate geopolymers synthesized by sodium silicate activation of metakaolinite. J Am Ceram Soc, 2009, 92, 2354-61. 198. Stone WE, El Shafei GM, Sanz J, Selim SA. Association of soluble aluminum ionic species with a silica-gel surface: a solid-state NMR study. J Phys Chem 1993, 97, 10127-32.

158

199. Xu J, Zhou Y, Chang Q, Qu H. Study on the factors of affecting the immobilization of heavy metals in fly ash-based geopolymers. Mater Lett, 2006, 60, 820-2. 200. http://www.world-aluminium.org/statistics/alumina-production/#data 201. Luukkonen, T., Abdollahnejad, Z., Yliniemi, J., Kinnunen, P., & Illikainen, M.. One-part alkali-activated materials: A review. Cement and Concrete Research, 2017,103, 21-34.

202. Komnitsas, K. and D. Zaharaki, Utilisation of low-calcium slags to improve the strength and durability of geopolymers., in Geopolymers: Structures, Processing, Properties and Industrial Applications J.L. Povis and J.S.J. van Deventer, Editors. 2009, 343-375. 203. Dimas D. D., Giannopoulou I. P. & Panias D. Utilization of alumina red mud for synthesis of inorganic polymeric materials. Mineral processing & Extractive metallurgy review, 2009, 30, 211-39. 204. Pridobivanje U., Blata E. Utilization of geopolymerization for obtaining construction materials based on red mud. Materiali in tehnologije, 2013, 47, 99- 104. 205. Toniolo, N., Rincón, A., Avadhut, Y. S., Hartmann, M., Bernardo, E., & Boccaccini, A. R. Novel geopolymers incorporating red mud and waste glass cullet. Materials Letters, 2018, 219, 152-154. 206. Liew Y., Kamarudin H., Al Bakri A. M., et al. Influence of solids-to-liquid and activator ratios on calcined kaolin cement powder. Physics Procedia, 2011, 22, 312-7. 207. Hamzah H. N., Abdullah, Mohd Mustafa Al Bakri, Yong H. C., Zainol, Mohd Remy Rozainy Mohd Arif, Nor A. M. & Wazien A. W. Correlation of the Na2SiO3 to NaOH Ratios and Solid to Liquid Ratios to the Kedah’s Soil Strength. In MATEC Web of Conferences, ed. Anonymous EDP Sciences, 2016, 01071. 208. Pişkin S., Figen A. K. Structural Characterization of Seydişehir Red Mud to Utilization in Roof Tile Manufacturing. IFAC Proceedings Volumes, 2013, 46, 484-7. 209. Singh P. S., Bastow T. & Trigg M. Structural studies of geopolymers by 29Si and 27Al MAS-NMR. J. Mater. Sci., 2005, 40, 3951-61. 210. Rincón A., Desideri D. & Bernardo E. Functional glass-ceramic foams from ‘inorganic gel casting’and sintering of glass/slag mixtures. J. Clean. Prod., 2018, 187, 250-6. 211. Lockey J. E., Redlich C. A., Streicher R., et al. Isocyanates and human health: multistakeholder information needs and research priorities. J. Occup. Environ. Med., 2015, 57, 44-51. 212. Natali Murri A., Medri V., Papa E., Laghi L., Mingazzini C. & Landi E. Porous Geopolymer Insulating Core from a Metakaolin/Biomass Ash Composite. Environments, 2017, 4, 86. 213. Bai C, Franchin G, Elsayed H, Conte A, Colombo P. High strength metakaolin- based geopolymer foams with variable macroporous structure. Journal of the European Ceramic Society 2016, 36, 4243-9.

159

214. Cellular Ceramics: Structure, Manufacturing, Properties and Applications. 215. Tulyaganov D., Fernandes H., Agathopoulos S. & Ferreira J. Preparation and characterization of high compressive strength foams from sheet glass. Journal of Porous Materials, 2006, 13, 133-9. 216. Papa E, Medri V, Kpogbemabou D, Morinière V, Laumonier J, Vaccari A et al. Porosity and insulating properties of silica-fume based foams. Energy Build 2016, 131, 223-32. 217. Prud’homme E, Michaud P, Joussein E, Peyratout C, Smith A, Arrii-Clacens S et al. Silica fume as porogent agent in geo-materials at low temperature. Journal of the European Ceramic Society 2010, 30, 1641-8. 218. Feng J, Zhang R, Gong L, Li Y, Cao W, Cheng X. Development of porous fly ash-based geopolymer with low thermal conductivity. Materials & Design, 2015, 65, 529-33. 219. Zhang Z., Provis J. L., Reid A. & Wang H. Geopolymer foam concrete: An emerging material for sustainable construction. Constr. Build. Mater., 2014, 56, 113-27. 220. Cilla M. S., Colombo P. & Morelli M. R. Geopolymer foams by gelcasting. Ceram. Int., 2014, 40, 5723-30. 221. Li K, Wang D, Chen H, Guo L. Normalized evaluation of thermal shock resistance for ceramic materials. Journal of Advanced Ceramics, 2014, 3, 250-8. 222. Han J, Wang B. Thermal shock resistance of ceramics with temperature- dependent material properties at elevated temperature. Acta Materialia 2011, 59, 1373-82. 223. Panda P., Kannan T., Dubois J., Olagnon C. & Fantozzi G. Thermal shock and thermal fatigue study of ceramic materials on a newly developed ascending thermal shock test equipment. Science and Technology of Advanced Materials, 2002, 3, 327-34. 224. Zhao Y., Jow J., Cai X. & Lai S. Fly ash based geopolymer foam technology for thermal insulation and fire protection applications. Proceedings of the 2015 World of Coal Ash (WOCA), Nashville, TN, USA, 2015, 5-7. 225. Luna-Galiano Y., Cornejo A., Leiva C., Vilches L. & Fernández-Pereira C. Properties of fly ash and metakaolín based geopolymer panels under fire resistance tests. Materiales de Construcción, 2015, 65, 059. 226. Pasaribu H. R. Friction and wear of zirconia and alumina ceramics doped with CuO, PhD Thesis, University of Twente, 2005. 227. Cheng-Yong H., Yun-Ming L., Abdullah, Mohd Mustafa Al Bakri & Hussin K. Thermal resistance variations of fly ash geopolymers: foaming responses. Scientific Reports, 2017, 7, 45355. 228. Vickers L., Pan Z., Tao Z. & Van Riessen A. In situ elevated temperature testing of fly ash based geopolymer composites. Materials, 2016, 9, 445. 229. Timakul P., Rattanaprasit W. & Aungkavattana P. Enhancement of compressive strength and thermal shock resistance of fly ash-based geopolymer composites. Constr. Build. Mater., 2016, 121, 653-8. 230. Sola D., Conde A., García I., Gracia-Escosa E., de Damborenea J. J. & Peña J. I. Microstructural and wear behavior characterization of porous layers produced

160

by pulsed laser irradiation in glass-ceramics substrates. Materials, 2013, 6, 3963-77. 231. Franchin G., Scanferla P., Zeffiro L., et al. Direct ink writing of geopolymeric inks. Journal of the European Ceramic Society, 2017, 37, 2481-9. 232. Xia M., Sanjayan J. Method of formulating geopolymer for 3D printing for construction applications. Mater Des, 2016, 110, 382-90. 233. Boccaccini A. Machinability and brittleness of glass-ceramics. J. Mater. Process. Technol., 1997, 65, 302-4. 234. Zehnder T., Freund T., Demir M., Detsch R. & Boccaccini A. R. Fabrication of Cell-Loaded Two-Phase 3D Constructs for Tissue Engineering. Materials, 2016, 9, 887. 235. http://www.tiletechpavers.com/interlocking-pavers-introduction/ 25th May 2018 235. https://www.thisoldhouse.com/ideas/all-about-concrete-pavers 25th May 2018 235. Hajjaji W., Andrejkovičová S., Zanelli C., et al. Composition and technological properties of geopolymers based on metakaolin and red mud. Materials &Design (180-2015), 2013, 52, 648-54.

161

162

List of abbreviations

3D Three dimensional

Al2O3 Aluminium oxyde FA Fly Ash FTIR Fourier Transform Infrared ICP Inductively Coupled Plasma MAS-NMR Magic Angle Spinning NaOH Sodium hydroxide

Na2SiO3 Sodium silicate or waterglass RM Red mud

SiO2 Silicon dioxide SLG Soda lime glass SLS Sodium Lauryl Sulfate XRD X-Ray Diffractometry

163

164

Index of Figures

Figure 1.1: Schematic overview of the research project which focused on the development of geopolymers incorporating waste materials. Moreover, foam production, direct ink printing and mechanical machining of geopolymers were parts of this research project...... 3 Figure 2.1: Scientific papers published per year with the keyword “geopolymers” (according to Web of science) ...... 6 Figure 2.2: Process of coal fired power plant with generation of fly ash...... 7 Figure 2.3: Flowsheet depicting the Bayer Process for producing alumina from bauxite...... 10 Figure 2.4: Schematic representation of glass recycling process in Sasil S.p.a...... 13 Figure 2.5: Tetrahedral configuration of sialate Si-O-Al-O. (Figure modified from Davidovits7)...... 18 Figure 2.6: Terminology of poly(sialate) geopolymers. Figure modified from Davidovits7...... 19 Figure 2.7: Schematic reaction sequences proposed by Provis et al. for the geopolymeric process (figure modified from Provis et al.96)...... 20 Figure 3.1: The volume-mean particle sizes for fly ash (a) and waste glass (b)...... 33 Figure 3.2: Schematic representation of the synthesis route for geopolymers...... 37 Figure 3.3: Geopolymer samples prepared with 10-20-30 wt.% of red mud or waste glass...... 38 Figure 3.4: Schematic representation of the chevron-notch (left) and the specimen in three point bending according to Dlouhy 167...... 39 Figure 3.5: Samples used to perform fracture toughness measurements...... 40 Figure 3.6: SEM images of the raw materials involved in the geopolymer synthesis: (a) fly ash, (b) red mud and (c) waste glass. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol...... 42 Figure 3.7: Image of compression test of 10SLG geopolymer (a) and geopolymer conical fracture after compression test (b)...... 43 Figure 3.8: Compressive strength average of fly ash-based geopolymers and geopolymers were 10-20-30 wt.% of fly ash is substituted with red mud or waste glass...... 43 Figure 3.9: Load – deflection curves for the different geopolymers investigated. .... 45 Figure 3.10: Flexural strength average for all the samples investigated...... 46 Figure 3.11: Notch depth determined in samples containing 20 wt.% of waste glass (left) and 10% of red mud (right)...... 47 Figure 3.12: Mean fracture toughness values of the different geopolymers investigated...... 48 Figure 3.13: Vickers indentation in the geopolymer sample 10SLG (left) and load- indentation curves (right)...... 49 Figure 3.14: Confocal microscope images for samples 30SLG (a and b) and 30RM (c and d)...... 50 Figure 3.15: Average roughness number for all specimens...... 51

165

Figure 3.16: SEM images of geopolymer fracture surfaces after compression strength tests with different replacement level of recycled glass or red mud...... 52 Figure 3.17: FTIR spectra of fly ash-based geopolymers with 10, 20 and 30% addition of waste glass or red mud. Reproduced from ref.142 with permission from J. of Ceramic. Sci. Technol. (The relevant peaks are discussed in the manuscript)...... 54 Figure 4.1: Scheme for the preparation of geopolymers samples of composition shown in Table 4.2...... 63 Figure 4.2: Geopolymers prepared with different SiO2/Al2O3 molar ratio and NaOH molarity...... 64 Figure 4.3: FTIR spectra of raw materials and geopolymers (compositions shown in Table 4.2). The relevant peaks are discussed in the manuscript...... 67 Figure 4.4:FTIR spectrum with focus placed on wavenumbers between 400 and 1500 cm-1. (The peaks are discussed in the text)...... 68 Figure 4.5: XRD patterns of raw materials and geopolymers with SiO2/Al2O3 molar ratio of 7 at different molarity...... 69 Figure 4.6: XRD patterns of raw materials and geopolymers with 8 molarity and different SiO2/Al2O3 molar ratio...... 71 Figure 4.7:27Al MAS-NMR spectra of raw materials and geopolymers with 7 SiO2/Al2O3 molar ratio...... 72 Figure 4.8: 27Al MAS-NMR spectra of raw materials and geopolymers synthesized with 8 molarity NaOH solutionf...... 73 Figure 4.9: Histogram representation of geopolymers porosity for different SiO2/Al2O3 ratios and different molarities...... 76 Figure 4.10: Compressive strength of geopolymer samples of different molarity and SiO2/Al2O3 molar ratio after 7 days...... 77 Figure 4.11: Compressive strength of geopolymer of different molarity and SiO2/Al2O3 after 28 days...... 78 Figure 4.12: Microstructure of fractured surfaces of geopolymers with SiO2/Al2O3 molar ratio of 5 and different NaOH molar ratio. Spherical particles are undissolved fly ash particles and particles with angles are unreacted glass particles...... 80 Figure 4.13: Microstructure of fractured surfaces of geopolymers with SiO2/Al2O3 molar ratio of 6 and different NaOH molar ratio. Spherical particles are undissolved fly ash particles and particles with angles are unreacted glass particles...... 81 Figure 4.14: Microstructure of fractured surfaces of geopolymers with SiO2/Al2O3 molar ratio of 7 and different NaOH molar ratio. Spherical particles are undissolved fly ash particles and particles with angles are unreacted glass particles...... 82 Figure 4.15: SEM images of 7S10M after 7 (a) and 28 (b) days, showing the sponge- like structure on the left and a more compact structure on the right...... 83 Figure 4.16: EDX spectra of 7S10M, after 7 days, in points a) and b) on the SEM micrograph...... 84 Figure 4.17: EDX spectra of 6S8M geopolymer obtained on the points a), b), c) of the SEM image...... 85 Figure 4.18: Water and fragments of geopolymers synthesized with 5M solution, after 24h of stirring...... 86 Figure 4.19: Water and fragments of geopolymers synthesized with 6M solution, after 24h of stirring...... 86

166

Figure 4.20: Water and fragments of geopolymers synthesized with 7M solution, after 24h of stirring...... 87 Figure 5.1: Geopolymers developed with red mud and waste glass with alkali sol. to raw materials ratio of 0.50...... 93 Figure 5.2: Compressive strength average results as function of NaOH solution molarity and SiO2/Al2O3. On the far left geopolymers synthesized with li./solid ratio of 0.50, while on the far right the liq./solid ratio used was 0.55...... 95 Figure 5.3: FTIR spectra of geopolymer samples and raw materials. The peaks of relevance are discussed in the text...... 97 Figure 5.4: SEM images of samples realized with NaOH molarity at 4 and 6 and SiO2/Al2O3 molar ratio of 5,6 and 7 showing unreacted particles surrounded by amorphous geopolymeric gel...... 98 Figure 5.5: Higher magnification SEM image of the sample 5S6M and EDX analysis. Showing the geopolymeric gel (a) and a fragment of unreacted glass (b).100 Figure 5.6: 27Al MAS NMR spectra of raw materials and geopolymers activated with 6M NaOH solution, showing the tetrahedral coordination of alumina in the geopolymer samples...... 100 Figure 6.1: Schematic diagram of the production of geopolymer foams using inorganic gel casting process...... 107 Figure 6.2: Geopolymer foams were cut to obtain cubes for further characterizations...... 108 Figure 6.3: Geopolymer foams created with SiO2/Al2O3 molar ratio of 7 and NaOH molarity of 8 a) and b) and 5 c) and d)...... 109 Figure 6.4: Microstructure of the samples 6S3M with SLS surfactant. At higher magnification particles of fly ash and unreacted glass fragments are visible...... 111 Figure 6.5: Microstructures of foamed geopolymers created with 5M NaOH solution and incorporating SLS (left) and triton (right)...... 112 Figure 6.6: Compressive strength of geopolymer foams with SLS (a) and triton (b)...... 113 Figure 6.7: Microstructure of red mud-based geopolymer foams. Increasing the amount of glass more unreacted particles are visible. All the formulations present a homogeneous distribution of the pores...... 115 Figure 6.8: Geopolymer cube-shaped foams used for the compression test...... 116 Figure 6.9: Compressive strength for red mud geopolymer foams when changing the SiO2/Al2O3 molar ratio...... 116 Figure 6.10: Samples after boiling test, 6S6M red mud on the far left, 7S3M in the middle and 7S5M on the far right, showing impurities in the water, except for the sample 7S5M where the water maintain its transparency...... 117 Figure 7.1: Samples 5S10M after thermal shock at 200°C, 500°C and 800°C...... 121 Figure 7.2: Samples 6S8M after thermal shock at 200°C, 500°C and 800°C...... 121 Figure 7.3: Porosity development after thermal shock test for geopolymers 6S8M...... 122 Figure 7.4: Compressive strength after thermal shock test at different temperatures...... 122 Figure 7.5: Microstructure of samples 5S10M before and after thermal shock test at 200°C, 500°C and 800°C...... 123

167

Figure 7.6: Microstructure of samples 6S8M before and after thermal shock test at 200°C, 500°C and 800°C...... 124 Figure 7.7: XRD patterns of 5S10M geopolymer before and after thermal shock test...... 125 Figure 7.8: Scheme of the pin-on-disc wear tester...... 126 Figure 7.9: Friction coefficient of samples 6S8M and 5S10M...... 127 Figure 7.10: Wear rate of samples 6S8M and 5S10M...... 127 Figure 7.11: Friction coefficient vs sliding distance for all the geopolymers tested (a), confocal microscope image for sample 6S8M (b)...... 128 Figure 8.1: Example of geopolymer sample fabricated on the lathe. (a) Original geopolymer shape, (b) desired final shape, (c) sample on the lathe and (d) final geopolymer component after machining...... 132 Figure 8.2: Example of geopolymer sample drilled with a metal tool. (a) represents the original geopolymer shape, (b) the desired final shape, (c) the samples on the lathe and (d) the final geopolymers...... 133 Figure 8.3: Bioplotter (a) and the syringe used to print the slurry (b)...... 134 Figure 8.4: Geopolymer ink extruded with a syringe on the left (a) and the final scaffold-structure after 28-day aging on the right (b)...... 136 Figure 8.5: Microscope images of the direct ink printed scaffolds, using needles of different diameters, (a) and (b) printed with the 1.36 mm needle, while (c) and (d) with the 0.84 needle...... 136 Figure 8.6: Geopolymer interlocking pavers inside the mould (a) and outside the mould (b)…………………………………………………………………………...138

168

Index of Tables

Table 2.1: Main categories of fly ashes7...... 8 Table 3.1: Chemical composition of raw materials (wt. %) determined by XRF. .... 34 Table 3.2: Quantitative phase composition of fly ash, determined by G-factor method. The structures given in the references for each phase were used for Rietveld refinement...... 35 Table 3.3: Quantitative phase composition of red mud, determined by G-factor method. The structures given in the references for each phase were used for Rietveld refinement...... 35 Table 3.4: Quantitative phase composition of waste glass, determined by G-factor method. The structures given in the references for each phase were used for Rietveld refinement...... 36 Table 3.5: Summary of mixture proportions used in the experimental trial...... 38 Table 3.6: Leaching test results from raw materials and geopolymer samples. Thresholds for inert and non-hazardous material from reference ...... 55 Table 4.1: Chemical composition of raw materials determined by XRF...... 60 Table 4.2: Summary of mixture proportion used in the experimental trial...... 62 Table 4.3: Density of geopolymer samples after 28-day ageing...... 74 Table 4.4: Porosity of the geopolymer samples after 28-day ageing...... 75 Table 4.5: Leaching test results from raw materials and geopolymer samples...... 88 Table 5.1: Chemical composition of the raw materials (wt.%) determined by XRF. 92 Table 5.2: Designation of the geopolymer samples produced with different amount of RM, SLG and different NaOH solution molarity...... 92 Table 5.3: Density and porosity of red mud-based geopolymers with liq./solid ratio of 0.50...... 96 Table 5.4: Density and porosity of red mud-based geopolymers with liq./solid ratio of 0.55...... 96 Table 5.5: Leaching test results from raw materials and geopolymer samples. Threshold values for inert and non-hazardous materials from ref. 180,...... 101 Table 6.1: Density and porosity of fly ash-based geopolymers foams with SLS. ... 110 Table 6.2: Density and porosity of developed fly ash-based geopolymers foams with triton...... 110 Table 6.3: Density and porosity of red mud-based geopolymers...... 114 Table 6.4: Summary of the density and compressive strength of the developed geopolymer foams. They present high open porosity, low density and reasonable mechanical properties...... 118

169

170

List of Publications

 Toniolo N., Rincon A., Avadhut Y., Hartmann M., Bernardo E., Boccaccini A.R., Novel geopolymers incorporating red mud and waste glass cullet, Materials Letter, 2018, 219, 152-154, DOI:10.1016/j.matlet.2018.02.061.

 Toniolo N., Boccaccini A.R., Fly ash-based geopolymers containing added silicate waste. A review, Ceramic International, 2017,43, 14545-14551, DOI:10.1016/j.ceramint.2017.07.221.

 Toniolo N., Taveri G., Hurle K., Roether J., Ercole P., Dlouhy I., Boccaccini A.R., Fly-ash-based geopolymers: How the addition of recycled glass or red mud waste influences the structural and mechanical properties, Journal of Ceramic Science and Technology, 2017, 8, 411-419. DOI: 10.4416/JCST2017-00053.

 Toniolo N., Rincon Romero A., Marangoni M., Binhussain M., Boccaccini A.R., Bernardo E., Glass-ceramic proppants from sinter-crystallisation of waste-derived glasses, Advances in Applied Ceramics, 2017, 117, 127-132. DOI: 10.1080/17436753.2017.1394019.

 Taveri G., Toušek J., Bernardo E., Toniolo N., Boccaccini A.R., Dlouhy I., Proving the role of boron in the structure of fly-ash/borosilicate glass based geopolymers, Materials Letters, 2017, 200, 105-108. DOI: 10.1016/j.matlet.2017.04.107.

171