Influence of ASR Degradation on Structural Behavior of Concrete Structures

Master Thesis (2019)
Author(s)

N. Lin (TU Delft - Civil Engineering & Geosciences)

Contributor(s)

C.B.M. Blom – Mentor (TU Delft - Concrete Structures)

M. Lukovic – Mentor (TU Delft - Concrete Structures)

D. A. Hordijk – Graduation committee member (TU Delft - Concrete Structures)

R. Esposito – Graduation committee member (TU Delft - Applied Mechanics)

Faculty
Civil Engineering & Geosciences
Copyright
© 2019 Nan Lin
More Info
expand_more
Publication Year
2019
Language
English
Copyright
© 2019 Nan Lin
Graduation Date
25-01-2019
Awarding Institution
Delft University of Technology
Programme
['Civil Engineering | Structural Engineering | Concrete Structures']
Faculty
Civil Engineering & Geosciences
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

A lot of concrete structuressufferingfrom ASR show a decrease in material properties. However, it is not clear what might be the influence of ASR on structural behavior. Influence of restraint is often ignored in material property measurement, while reinforcement is present as a restraint of ASR expansion in most structures.Thisresults in so called chemical prestresseffect, whichshould not be neglected. In the presented study,the result indicatesthat neglecting the chemical prestresswould miss the failure mechanism transition and lead to a lowerbeam stiffness. Therefore, a newway to model ASR-affected structures is proposed, called the ASR-layered model.The numerical analysis is done with ATENA.Depending on the reinforcement arrangementand amount, with longitudinal reinforcement acting as a restraint, expansion to the greatest extend occurs in the direction of least confinement and the cracks become parallel to reinforcement. Considering the potential to connect with each other and form long horizontal cracks, the ASR crackingin restrained structuresare modelled as horizontal layers with reduced material properties. The propertyreduction is concentrated in ASR layers only, this local reduction is more realistic than the traditional method done by (Ferche, Sheikh, & Vecchio, 2017)in which a global reductionon material propertiesis applied. For chemical prestress, it is difficult to separate the influence of prestressin experiments,so experimentalresult is always affected both by the reduction of properties and prestress and probably some other phenomena that exist. The chemical prestress effect is modelledherein the way of physical prestress, but the expansion of ASR gel in chemical prestress would possibly influence the bond strength between concrete and reinforcement, so bond model is also considered.Some conclusions can be drawn on this study. The material reduction caused by ASR would only result in a decrease of load capacity. While in combination with prestress, the load capacity of ASR affected beam may be increased to be even higher than the unaffected beam. The application of prestress could also lead to the possible change of failure mechanism from shear to bending failure, whichis also observed in experiments. Including the bond-slip effect in prestressed ASR-layered model provides a better fitting with experimental results,but the influence of ASR on bond strength remains unclear. With parameter study, no obvious influence on ASR-affected structures could be linked to a/d ratio and reinforcement ratio. 1 or 2 ASR layers will only influence slightly on the ultimate load. The limitations of this model exist, the expansion caused by ASR is not directly modelled. Information on ASR layer property is quite limited, it would be difficult to measure the cracking part of ASR affected structures alone. In addition, the chemical prestress and possible cracking depend on the amount of ASR expansion as well as the reinforcement arrangement and amount, but they are not considered to correlate with each other in this model.

Files

NanLin_MasterThesis.pdf
(pdf | 8.38 Mb)
License info not available