Author: Khashayar Jafari

Publisher:

Published: 2021

Total Pages:

ISBN-13:

Supplementary cementitious materials (SCMs) are a key ingredient of today's concrete and can vastly improve the durability and sustainability of concrete mixtures. While the demand for traditional pozzolans (i.e., pulverized coal fly ash and slag) continue to grow, the supply of high-quality and economically available pozzolans has been shrinking. As such, there is a significant need and interest to search for alternative SCM sources due to supply-and-demand concerns in the future. One of the most promising alternative sources are calcined clays since they have not yet reached their full potential as cement replacement and clay is an abundant and widespread material resource. Calcined clays other than metakaolin (high purity calcined kaolinite clay) are not currently used as SCMs in the concrete industry due to the complexity of clay minerals and insufficient knowledge of the underlying reaction mechanisms. Also, their performance as concrete pozzolans are not well understood. To address these knowledge gaps, the purpose of this research is to identify, characterize, improve, and facilitate the use of impure calcined clays (CC) as viable SCMs in Portland cement concrete and as a precursor in geopolymer concrete. To increase the service life and expand the use of high-performance concrete, it is critical to provide a high-quality, durable, and cost-competitive concrete; as such, a stable and abundant supply of SCM is required. Toward this objective, three calcined clays were obtained from eastern, central, and western United States which had different amorphous contents while the balance included minerals such as quartz, feldspar, iron oxides, and other clay minerals. The key starting point of this study consists of a detailed chemical, mineralogical and physical characterization of impure calcined clays (CC) and evaluation of their performance as a pozzolan in Portland cement concrete or as an aluminosilicate source for geopolymer concrete. In addition, to assess the pozzolanic reactivity of CCs, isothermal calorimetry and bound water were used to measure the heat of hydration and chemically bound water, respectively, in the lime-pozzolan paste systems (aka R3 test, ASTM C1897-20). The results were compared with quantifying the unreacted pozzolan and as a function of time (by acid dissolution) and portlandite consumption (by TGA) within the R3 paste. A low-purity clay was obtained from a large commercial aggregate production facility. Several methods included centrifuging to classify based on particle size and density, membrane filtration to separate based on particle size, and dispersant-assisted sedimentation were used for separating the clay and non-clay minerals and enrichment of low-purity kaolinite clays to improve their reactivity and performance in concrete. Both low-purity and purified clay were calcined and tested in mortar and concrete mixtures for their fresh properties (slump, air content, and time of setting), hardened properties (air content, compressive strength) and durability (drying shrinkage, alkali-silica reaction, and rapid chloride permeability). The results were compared to control mixture (100% Portland cement, PC). Further, the low-purity CC was used in combination with slag cement as a partial replacement of PC (~50%) in ternary concrete mixtures. The performance of ternary binder was evaluated in mortar and concrete mixtures and the results were compared with control (100% PC) and two binary (20% CC and 50% slag) mixtures. Also, the possibility to make a coupled substitution of PC with CC and limestone (LS) was tested to allow higher levels of cement replacement up to 55%. For this purpose, a statistical design of experiment (DoE) approach combined with machine learning (i.e., artificial neural network, ANN) was adopted to determine the optimum PC, CC, and LS contents in mortar mixtures, leading to the best strength versus time. The best mixture was selected considering the compressive strength, cost, and CO2 emission of each mixture. Besides, the possibilities for eliminating the use of PC, and thus minimizing the CO2 footprint of concrete, were explored via a one-part liquid (just add water) geopolymer binder. Some routine tests were performed on mortar mixtures such as flowability, flow retention, time of setting, compressive strength, and volume stability. This study can provide sufficient knowledge on reactivity of impure CC sources and facilitates large-scale and high-volume usage of these sources in concretes to ensure their long-term performance and low life-cycle cost, and improved sustainability. The impact of this research is to advance the science, technology, and practice to facilitate doubling the supply of concrete pozzolans in the U.S. within the next decade. The research work in this dissertation will result in six journal publications, and several conference presentations/posters.