Assessment on the application of volcanic materials as precursor in alkali-activated binders

Abstract

In 2016 the Paris agreement was adopted by 196 parties. The main goal of this agreement is to obtain a climate-neutral world in 2050. The CO2 emissions should decrease as soon as possible to achieve this goal. Concrete is the world's most used building material, and with good reason, it provides cheap, solid and reliable products used for over 200 years. However, concrete is responsible for 5-10 % of the yearly CO2 emissions worldwide. These emissions are mainly because of cement production, the binding agent of concrete. Therefore, to lower the CO2 emissions from concrete, it is most effective to opt for an alternative binder. This research focuses on alkali-activated materials (AAMs), where an alkali-activated binder has replaced cement. This binder consists of a precursor and an alkaline activator. The materials most commonly used as precursors are blast furnace slag and fly ash, waste products. The production of these materials is not nearly enough to replace cement. Therefore, there is a need for alternative precursors. Volcanic materials can be an alternative precursor, and some have been successfully applied in AAMs before. However, since volcanic materials come in different forms, validation of a wide range of volcanic materials.

This research has investigated four European aluminosilicate volcanic materials: trass, phonolite, perlite and pumice. Each of these materials has been assessed on their particle size, chemical composition and microstructure to determine whether they exhibit (pozzolanic or latent hydraulic) reactivity. All four materials classify as inert. For trass and phonolite, the lack of reactivity is most likely because of their lack of amorphous phases (around 30% and 5%, respectively). For perlite and pumice, it is because of the too coarse particles.
The reactivity of trass and phonolite is enhanced by calcination, and the reactivity of perlite and pumice is enhanced by grinding. Determination of the reactivity shows that calcination had no or a negative effect on the reactivity of trass and phonolite. Most likely because of the lack of calcium and magnesium in these materials. Ground pumice does show an increase in reactivity. However, not sufficient to be classified as reactive. Ground perlite does show reactive behaviour and is classified as pozzolanic, less reactive.

Ground perlite and BFS are used as precursors in mortar to assess the contribution of perlite to compressive strength development. It shows that perlite contributes to the strength of mortar when combined with a sodium silicate and sodium hydroxide activator. Likely, the sodium silicate is necessary to delay the alkalinity of the mixture. This alkaline delay makes that perlite does not form a layer of hydration products over the particles, making it impossible for perlite to react further and contribute to the compressive strength. TGA measurements found that because of the silicates that perlite contributes to the mixture, the hydration products become denser, which seem to cause higher strength. However, it is also found that silicates from sodium silicate solution are more effective in providing silicates than perlite is.

Of all four materials, only perlite seems to be reactive enough to contribute to the strength of alkali-activated binders as precursors. Volcanic materials cannot be considered to be one material, and each one has to be assessed individually.