Experimental and analytical investigation of high performance concrete columns at elevated temperatures

In recent years, local construction industry has shown great interest in applications of high performance concrete (HPC) due to its superior performance such as higher strength and longer durability compared with other building materials. In the design of reinforced concrete (RC) structures, code pr...

Full description

Saved in:
Bibliographic Details
Main Author: Du, Panwei
Other Authors: Tan Kang Hai
Format: Thesis-Doctor of Philosophy
Language:English
Published: Nanyang Technological University 2021
Subjects:
Online Access:https://hdl.handle.net/10356/152911
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:In recent years, local construction industry has shown great interest in applications of high performance concrete (HPC) due to its superior performance such as higher strength and longer durability compared with other building materials. In the design of reinforced concrete (RC) structures, code provisions require all structural members to have adequate fire resistance under fire conditions. After a comprehensive review of previous research works, the author identified that there were limited studies on effects of load eccentricity (both uniaxial and biaxial), column dimensions and axial restraint for HPC columns. Moreover, existing analytical models and codes of practice mainly address the fire design of normal strength concrete (NSC) columns. To address these research gaps, a comprehensive study was undertaken to investigate structural behaviour of HPC columns at elevated temperatures through fire tests and analytical modelling. To strengthen the understanding of fire performance of HPC columns, two series of tests consisting of ten full-scale HPC columns were conducted under the standard ISO 834 fire curve. The first series including six columns was aimed to investigate the effects of uniaxial load eccentricity and column dimensions, while the second series of another four columns focused on the effects of biaxial bending and axial restraint. Arising from the test programme, there were new experimental data on the structural behaviour of HPC columns including axial deformation, mid-height column deflection, thermal-induced restraint force, failure modes and fire endurance. It should be highlighted that no explosive spalling was observed in all the HPC columns, even under complex loading conditions such as eccentric loading and axial restraint due to the addition of hybrid polypropylene and steel fibres. A fire design model incorporating heat transfer analysis, fibre reinforced concrete design and second-order effect of slender columns was proposed. A comparison with test data showed that the fire design model with 400 °C isotherm depth could provide safe and accurate predictions on failure time and failure load. After a comprehensive assessment of structural behaviour of HPC columns at elevated temperatures subjected to different loading conditions, a simple yet universally applicable model was proposed to analyse the behaviour of RC columns under ambient and fire conditions. The novelty of the proposed model is that it transformed the cross-sectional capacity into actual column capacity by introducing a stability term. The proposed model in the reciprocal form provided a theoretical background to the classical Rankine formula. It was validated against a large set of data including 47 specimens tested at ambient temperature and 72 specimens under fire conditions. The comparison with test results showed that the proposed model could capture well the mid-height deflections of RC columns at both ambient and elevated temperatures. Moreover, accurate and conservative predictions were achieved on peak loads at ambient temperature with a mean value of 0.96 and a COV of 0.13, and fire resistance at elevated temperatures with a mean value of 0.97 and a COV of 0.19. The model had a wide range of applicability including NSC and HPC columns with compressive strength varying from 24.1 to 97.2 MPa, pure axial loading and eccentric loading with load eccentricity (e/B) from 0.1 to 0.5, and short and slender columns with slenderness ratio (L/r) ranging from 20.8 to 98.9. A series of parametric studies were conducted to investigate the effects of transient strain (ɛtr,c) and creep strain (ɛcr,c) on mid-height deflection. From the analyses, it was found that effects of ɛtr,c and ɛcr,c are more significant in more slender columns, or columns under a lower load level, or with a smaller width (higher slenderness ratio), or under a longer heating duration (creep strain). The proposed model was further extended to circular confined RC columns which greatly simplified the calculation process by omitting finite element modelling and regression analysis. The simplified method was validated against three series of test data including a total number of 40 specimens collected from the literature. In general, the predictions using the proposed method were accurate and conservative. The predicted mid-height deflection yielded an overall mean of 0.942 and a COV of 0.194, while the predictions of load-carrying capacity had a mean of 0.977 and a COV of 0.076. This approach was suitable for both NSC and HPC and axially- and eccentrically-loaded columns with a broad range of slenderness ratios (L/r = 12 to 40).