Refine
Document Type
- Doctoral Thesis (2)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- Röntgendiffraktometrie (2) (remove)
Institute
- Institut für Geographie und Geologie (2) (remove)
Autoclaved aerated concrete (AAC) is a building material that combines heat insulation
properties with sufficient mechanical strength for masonry construction. Compared to
ordinary concrete, the matrix is highly porous (>50%) and hardened by a hydrothermal curing
process at 150°C - 200°C. During this process, quartz sand and portlandite react to form first
calcium silicate hydrates (C-(A)-S-H) with Ca/Si ratios <1.3 and then tobermorite. Especially
tobermorite, which has a much larger crystallite size than C-(A)-S-H, provides improved
mechanical strength. This reaction sequence is influenced by many parameters and
additives of which calcium sulfate is probably the most important. Despite several attempts to
investigate these hydrothermal reactions, the actual reaction mechanism involved when
adding sulfate ions is not fully understood. It has been suggested that the addition of ca.
1.5 wt% significantly improves the mechanical properties due to the enhanced formation and
arrangement of tobermorite in the porous matrix. Since the sulfate content in AAC waste is
exceeding regulatory threshold for low-quality reuse in some countries, the aim of this study
was to investigate in detail the reaction mechanisms involving sulfate addition. Such
knowledge may open up the possibility to improve AAC production and to avoid the need for
sulfate addition. To achieve this goal, this research work focused on investigating the
hydrothermal curing process to determine the sequence of hydrothermal reactions and the
spatial distribution of the phases formed. For this purpose, a new setup for in situ X-ray
diffraction was specifically designed to study hydrothermal reactions and to conduct time
intensive experiments on a normal laboratory diffractometer. In order to quantitatively
evaluate the in situ measurements by Rietveld analysis using TOPAS, it was also necessary
to develop atomistic structure models for C-(A)-S-H phases. This was made possible by
adopting a supercell approach that was previously used to describe turbostratically stacked
clay minerals. The structure models, derived from tobermorite, are placed in an otherwise
empty supercell to simulate the C-(A)-S-H nanostructure. Adopting these methodological
advances, it was possible to obtain absolute phase quantities from in situ data and to track
the reaction kinetics of the hydrothermal curing process. These results were then combined
with ex situ X-ray diffraction and scanning electron microscopy. Confirming previous studies,
the major effect of sulfate ions was the formation and decomposition of hydroxylellestadite. It
was further revealed that C-(A)-S-H formation was delayed during hydroxylellestadite
formation, which is supposed to support the silicate ion diffusion and hence the tobermorite
formation at a stage critical for improved hardening of AAC. This can be linked to the
formation of lower amounts of capillary pores in the range of 1-5 µm, as observed by
scanning electron microscopy, and therefore a lower concentration of inherent defects that
resulted in the improved mechanical properties. This research work highlights how important the spatial distribution of crystallites is for the properties of a building material and how this
distribution can be influenced by small alterations in reaction chemistry.
Geopolymers (GPs) are inorganic binders created by adding alkaline solution (e.g. KOH) to silicates such as furnace slag, fly ash or clay to dissolve Si and Al that polymerises and precipitates to form an inorganic binder material while hardening. GP properties are similar to ordinary Portland cement regarding their high compressive strength or low shrinkage but they are particularly notable for a high resistance to acid and fire. However, the most significant advantage of GP cements is their low CO2 footprint. The most common clay used as GP raw material is kaolin. The aim of this study is to investigate the suitability of illitic clays as a cheaper alternative to kaolin and determine the necessary preparation steps required to produce effective GP binder materials. Three clays dominated by dioctahedral 2:1 layer silicates, in particular interstratifications of mica and smectite were investigated: (1) Illitic clay from Friedland, Northern Germany, containing an irregularly stacked illite-smectite interstratification (R0 I-S), (2) rectorite from Arkansas, USA, as a regular interstratification of mica and smectite, and (3) clay stated as “sárospatakite” from Füzérradvány clay deposit, Northern Hungary, containing a long range ordered I-S (R3). The three types of I-S interstratification-rich clays were extensively characterised and the Friedland clay, as the most probable raw material for GP production, was studied in more detail including several size fraction analyses. These results are used to investigate and determine the parameters necessary to produce suitable precursors for GP binders. Different approaches of clay activation to yield a highly reactive material by milling and heating were examined. Milling was found to be suitable as a preparation step after heating breaking up sintering aggregates to create pathways for the alkaline solution, but not as a substitute for heating. Important parameters for the precursor design such as temperature, time, and heating rate are determined and discussed. Geopolymerisation is considered to be a multi-parameter system and is influenced strongly by the degree of dehydroxylation, Si:Al ratio, or amount of 5-fold coordinated Al. However, in contrast to kaolin-based systems, none of these parameters explain why the illitic Friedland clay heated to 875 °C was found to be most suitable for GP binders. Based on leaching experiments and specific surface area (AS) measurements of the heated Friedland clay, a conceptual model is presented to explain the observed relationship between the heating temperature and the subsequent compressive strength of the GP cement. An optimum between the counteracting reactions of decreasing AS (fewer particles must be covered with GP phase) and decreasing Si+Al dissolved (less GP phase created) is necessary, which exists at 875 °C for the Friedland clay. In this state enough GP phase is created to bind all remaining sintering aggregates to form a cement with high compressive strength. This relationship can be expressed as (Si+Al) / AS (sum of dissolved Si and Al divided by the surface area of grains that must be covered with GP phase), and can be used as a predictive tool for determining the optimal heating temperature. The results presented in this thesis indicate that illitic clays are suitable raw materials as GP binders if the necessary preparation steps of dehydroxylation, sintering and grinding are made. Proxies used to evaluate the optimal conditions for making GP binders are determined including the (Si+Al) / AS ratio as a key relationship that controls the cementation process and determines its ultimate hardness.