Seismic performance of high-strength self-compacting concrete in reinforced concrete structures.
Thesis DisciplineCivil Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Self-compacting concrete (SCC) was first developed in Japan about two decades ago. Since then, it has been offered as a solution to various challenges inherently associated with traditional concrete construction; i.e. quality and speed of construction, impact of unskilled labour force and noise pollution etc. SCC flows into a uniform level under its own weight and fills in all recesses and corners of the formwork even in highly congested reinforcement areas. In recent years the interest in using SCC in structural members has increased manifold; therefore many researchers have started investigating its characteristics. Nevertheless, before this special concrete is widely accepted and globally used in structures, its structural performance under different conditions should be investigated.
This research focuses on investigating the behaviour of high strength self-compacting concrete (HSSCC) in reinforced concrete (RC) structures through a systematic approach in order to bridge part of an existing gap in the available literature. The dissertation is comprised of four main stages; namely, mix design development and mechanical properties of HSSCC, bond performance of deformed bars in HSSCC, experimental investigation on interior RC beam-column joints (BCJs) cast with HSSCC under reversed cyclic excitations, and finally finite element (FE) modelling and analysis of interior BCJs.
First, a HSSCC mix proportion yielding compressive strength greater than 100 MPa was developed in the laboratory using locally available materials in New Zealand. Two benchmark concrete mixes of conventionally-vibrated high-strength concrete (CVHSC) and normal-strength conventionally vibrated concrete (CVC) were also designed for comparison purposes. Material characteristics (such as compressive, splitting tensile and flexural strengths as well as modulus of elasticity, shrinkage and microstructural properties) of all mixes were evaluated. It was found that, once the lower quality of material in normal strength concrete is offset by achieving a denser mix in high-strength concrete, mechanical properties of HSSCC are equivalent to or higher than those in CVHSC.
Given that the performance of RC structures (and in specific BCJs) is highly dependent on bond between reinforcement and concrete, understanding the bond behaviour in HSSCC was an imperative link between the first and third phases of this research. Therefore, the second phase focused on scrutinizing bond properties of deformed bars in HSSCC using monotonic pull-out and innovative cyclic beam tests. Processing of the pull-out results revealed that a shorter development length may be utilized in HSSCC. In addition, the grade (or ductility) of reinforcing steel was found to substantially influence the post-yield bond performance. Important modifications to the bond model used in the CEB-FIP model code and Maekawa’s bond-slip-strain relationship were suggested from the results of this phase. An innovative cyclic beam specimen and test setup were also designed such that a more realistic bond performance could be observed in the laboratory tests compared to that in real RC structures. Deleterious impact of cyclic loading and buckling of reinforcement on bond performance were investigated using this testing protocol.
The third phase of this research focused on the design, fabrication and testing of seven full-size BCJs. BCJs are one of the most critical parts in RC frame structures and their response substantially affects the overall behaviour of the structure. In seismically active regions like New Zealand, the criticality of BCJs is exacerbated with the complexities involved in seismic resistance. The already congested intersection of RC beam and column looks more like a solid steel connection after consideration of earthquake requirements, and placement of concrete becomes problematic in such areas. At the same time, in many of the high-rise structures, normal strength concrete does not meet the capacity requirements; this requires the usage of high-strength concrete. Therefore, once the seismic performance of HSSCC is guaranteed, it can possibly be a solution to both the capacity and compaction problems. Variables such as axial load, concrete type, steel grade, casting direction, and joint shear reinforcement were considered variable in the experimental investigations. It was found that HSSCC has similar seismic performance to that of CVHSC and it can also be incorporated in the joint area of CVC for an enhanced performance.
Finally, DIANA (a nonlinear FE program) was used to simulate the experimental results obtained in the third phase of this research. All BCJs were successfully modelled using their relevant attributes (such as the mechanical properties of HSSCC, steel stress-strain response, test setup and loading protocol) and nonlinear FE analyses (FEA) were performed on each model. FE results were compared to those obtained in the laboratory which showed a reasonable agreement between the two. The capabilities of the FEA were scrutinized with respect to the hysteresis loops, energy dissipation, joint shear deformations, stress development in the concrete and steel, and drift components.
Integrating the results of all stages of this research provided better understanding of the performance of HSSCC both at the material and structural levels. Not only were none of the seismically important features compromised by using HSSCC in BCJs, but also many other associated benefits were added to their performance. Therefore, HSSCC can be confidently implemented in design of RC structures even in seismically active regions of the world.