DEVELOPMENT OF REGRESSION MODELS FOR STRENGTH ASSESSMENT OF CONCRETE CONTAINING BINARY, TERNARY AND QUATERNARY CEMENT BLENDS

Abstract

Pozzolans have long been established as viable materials for the partial replacement of cement in concrete. However, the extent to which they can be used is still under investigation. Pozzolans are material which although in themselves do not have cementitious value, in finely divided form and in the presence of moisture react chemically with calcium hydroxide to produce compounds with cementitious properties. When used as binder in concrete, they help reduce water evaporation, heat of hydration, workability and sometimes, increase its compressive strength. Some of the most commonly used pozzolans include bagasse ash, ceramic waste powder, metakaolin, rice husk ash, palm oil fuel ash, glass waste powder, fly ash, etc. This research focuses on three commonly used pozzolans - Rice Husk Ash (RHA), Ceramic Waste Powder (CWP), and Glass Waste Powder (GWP) and their influence on the compressive strength of concrete. It assesses the suitability of using OPC, RHA, CWP and GWP cement blends in binary, ternary and quaternary combination for concrete production. Several tests were carried out, according to BS and ASTM requirements to ascertain the physical and chemical properties of these pozzolans and conventional concrete constituents. Some of these tests include bulk density, setting time, aggregate crushing value (ACV), aggregate impact value (AIV), slump test, X-Ray Fluorescence (XRF) and compressive strength tests. XRF, in particular, is used to determine the oxide compositions of the binders and compressive strength tests are carried out to determine the compressive strengths of the resulting concrete. The results of the compressive strength tests for concrete containing binary, ternary and quaternary combinations of Ordinary Portland Cement (OPC), RHA, CWP and GWP cement blends were analyzed using the Statistical Package for the Social Sciences (SPSS). The results of the study show that RHA, CWP and GWP have higher silica content compared to OPC.These pozzolans, according to ASTM C618-17a, can be classified as Class-N pozzolans since the total percentage composition of silicon, aluminum and iron oxides is greater than 70%. OPC, on the other hand, has a comparatively lower silica content. It is instead richer in calcium oxide compared to these binders. The study also reveals that the binary cement blends of OPC-CWP and OPC-GWP can produce concrete with compressive strengths comparable with that of the control concrete. In fact, OPC-GWP and OPC-CWP concrete mixes at their respective optimal replacement levels (10% and 15%) had higher compressive strengths at 21.90MPa and 22.60MPa respectively compared to the average compressive strength of the control (20.15MPa) at curing age 90. The compressive strength of OPC-RHA concrete (17.93MPa) was only comparable with that of the control up to a replacement level of 10%. In ternary combination, concrete containing OPC-CWP-GWP cement blend performed best. However, the results are only comparable with those of the control after 90 days of curing. The quaternary combination of the cement blends had the least performance, with its highest recorded compressive strength as much as 48.18% lower than that of the control after 90 days of curing (the terminal curing age adopted for this study). This compressive strength was achieved by concrete containing 15% RHA, 5% CWP and 10% GWP. Finally, regression models were developed to predict the compressive strengths of concrete in binary, ternary and quaternary combinations. The model shows that the influence of RHA is most significant in quaternary combination. The models also had an average precision accuracy of 78%. The developed models can be used by concrete industry practitioners to predict the strength of pozzolanic concrete using binary cement combinations of OPC-RHA, OPC-CWP or OPC-GWP. The replacement of OPC with these pozzolans are in part, to conserve cost.