Alkali activated earth construction materials

Abstract

There is a shortage of adequate housing in urban areas of less economically developed countries (LEDCs), which is likely to be exacerbated by high population growth. Conventional walling materials typically used for this application include unstabilised earth, fired brick and concrete block. None of these conventional materials are sufficient to meet the multi-dimensional requirements for sustainable urban growth. There is therefore a demand for new materials that fulfil the criteria of being practical, sustainable and affordable, which is currently unmet. Alkali-activated earth materials are an emerging category of construction materials which could have the potential to fulfil these criteria. In these materials, the clay minerals in soil are transformed into a stabilising phase by the addition of an alkaline activator, in order to give the soil greater strength and durability. These materials have two main potential advantages over conventional walling materials. Firstly, soils are low cost, low environmental impact precursors; secondly, alkali-activated stabilisation has the potential for lower environmental impact than Portland cement stabilisation as it does not require high temperatures or the direct release of CO2 in the life cycle. Despite these potential advantages, there is a significant knowledge gap around which soils are suitable to use in alkali-activated earth materials. The aims of this thesis are firstly, to establish a fundamental understanding of which soil compositions are suitable for alkali activation, and secondly, to assess the overall viability of alkali-activated earth materials as walling materials suitable for mass housing in this application.

An experimental programme was devised to understand the behaviour of the different components of soil in alkali activation. In this programme, the complexity of the precursors was built up progressively, starting from individual clay minerals commonly found in soils (kaolinite, montmorillonite and illite), followed by mixtures of these clay minerals, natural and synthetic soils, and finishing with soils containing an addition of aggregate. In a simple production process, clay or soil precursors were activated using an aqueous solution of NaOH and then cured at a low temperature of 80°C. Phase formation behaviour was investigated using a range of characterisation techniques. Constraints were specified to make the systems relevant to construction in urban areas of LEDCs. Firstly, an innovative consistency constraint was used, to ensure that the mixes would be appropriate for brickmaking processes. Secondly, the clay and soil precursors were used in their uncalcined form to minimise both the environmental impacts and the technological complexity of the process.

The findings from each experimental stage were used to inform the understanding of the next stage in the series. At the start of the series - the individual clay minerals, the product phases formed by alkali activation were hydrosodalite for kaolinite, a N-A-S-H or (N,C)-A-S-H geopolymer for montmorillonite, and illite did not form a product phase but underwent alteration. Under the range of conditions used, the clay minerals were never fully consumed. For the mixtures of clay minerals, phase formation behaviour deviated from an ideal rule of mixtures model, which suggested there was a hierarchy of reactivity and influence between the individual clay minerals. For the natural and synthetic soils, it was shown that the clay mineralogy largely determined phase formation behaviour. In contrast, the non-clay components generally had little or no effect on phase formation behaviour, although they did produce a retarding effect on geopolymer formation in one natural soil. In addition, the plasticity of soil was shown to be an important factor in the practical suitability of soils for alkali activation. For the soil mixed with aggregate, it was shown that neither the production of a larger sample, nor the addition of inert aggregate, made any fundamental differences to the alkali activation process.

From this improved technical understanding, it can be stated that using this production process, kaolinitic soils are suitable for alkali activation, whilst montmorillonitic and illitic soils are unsuitable. However, building on this improved fundamental understanding, there is scope for a wider range of soils to be used by tailoring their composition with reactive additives and admixtures. Future research should develop how to tailor soils in this way, and also lower the environmental and financial cost of NaOH-based activators. This research has made an important contribution to the fundamental understanding of how the different components of soil behave in the alkali activation process. Going forward, alkali-activated earth materials have the potential to be part of the solution in providing practical, sustainable and affordable walling materials for housing.

Date of Award	3 Apr 2019
Original language	English
Awarding Institution	University of Bath
Supervisor	Andrew Heath (Supervisor), Mark Evernden (Supervisor), Pete Walker (Supervisor) & Pascaline Patureau (Supervisor)