Material characterisation of existing masonry: A strategy to determine strength, stiffness and toughness properties for structural analysis

S. Jafari

Research output: ThesisDissertation (TU Delft)

303 Downloads (Pure)

Abstract

Understanding the behaviour of unreinforced masonry (URM) structures requires in-depth insight into the mechanical properties of its constituents and their interaction under compression, tension, and shear loading. As a result, a complete picture of masonry characteristics, accounting for its full nonlinear response and its statistical distribution, has long been of scientific research interest worldwide. This has become a necessity for the Netherlands in recent years because of the induced seismicity affecting the vulnerable masonry building stock in the province of Groningen. The more rigorously the mechanical properties are determined, the more they engender confidence in the reliability of structural analyses. This calls for an interdisciplinary approach, whereby experimental research focuses on meeting the demands of structural analysts and numerical modellers. As a result, laboratory and in-situ testing campaigns can provide input parameters and a basis for the validation and calibration of the various models to be used in deterministic or probabilistic settings. Nevertheless, a thorough characterisation of the material properties of existing unreinforced masonry structures is widely acknowledged as a challenging task, given the quasi-brittle nature of masonry, a very diverse range of masonry types, a lack of reliable in-situ semi-invasive testing methods, and a large number of tests required to be carried out. To this end, both invasive and semi-invasive testing methods are adopted in this thesis. The former refers to tests on medium-sized samples, which follow the guidelines of the European standards for the selection of each sample’s size. The latter points to a novel testing method, whereby small-diameter cores are extracted perpendicular to the wall surface. In this thesis, the studied masonry types are either laboratory-made, replicating the five most typical Dutch masonry types, or field-extracted, from nineteen different URM dwellings and schools in the Groningen region, built between 1910 and 2010. The typical masonry types include clay brick masonry, calcium-silicate brick masonry and calcium-silicate element masonry. By performing 218 invasive tests on laboratory-made replicated masonry specimens, this thesis provides comprehensive insight into the behaviour of five masonry types and its constituents. It covers several aspects that have been addressed only partially in the literature: 1) investigating the response of masonry under different stress states, i.e. compression, bending, and shear; 2) analysing the complete nonlinear behaviour not only in the pre-peak but also in the post-peak regime; and 3) exploring the orthotropic response under compression and bending loads. Moreover, this lab-testing campaign has studied the influence of brick type, number of wythes, and joint thickness on the response of masonry. By performing 478 invasive tests on field-extracted masonry specimens, this thesis offers a regional dataset of material properties. This dataset, along with previously available data from the literature, has been transferred into material properties table that have been incorporated into the Dutch standard for the assessment of existing URM buildings subjected to induced earthquakes. Moreover, this research provides new insights into the inter-building variability of material properties and their statistical distribution. This thesis also investigates the suitability of a novel semi-invasive testing method, whereby 167 small-diameter cores were subjected to compression as well as shear-sliding load. To this end, a comparative study was conducted, in which the material properties obtained from tests on cores, namely strength, stiffness, and toughness, are correlated with those found from tests on companion specimens. Promising conclusions are drawn, indicating that this testing method can be regarded as a reliable and practical alternative to conventional flat-jack based in-situ testing methods. Considering the comprehensive dataset established and the study of different testing methods, this thesis ultimately formulates a strategy to characterise the material properties for assessment of existing URM structures. To this end, this research investigated the presence of relationships between different material properties, thus offering recommendations to indirectly derive elastic and toughness properties as a function of easy-to-obtain properties, i.e. strength found from testing cores and bond wrench tests. Moreover, this study introduced improved constitutive functions for compression, bending, and shear loading. Therefore, following this strategy, an acceptable level of knowledge on material properties can be gained, while intrusiveness and damage due to sampling remain limited.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Delft University of Technology
Supervisors/Advisors
  • Rots, J.G., Supervisor
  • Esposito, R., Advisor
Thesis sponsors
Award date24 Sept 2021
Print ISBNs978-94-6421-457-4
DOIs
Publication statusPublished - 2021

Keywords

  • Dutch Unreinforced Masonry
  • Material Characterisation
  • Experiments
  • Compression
  • Bending
  • Shear
  • Clay Brick Masonry
  • Calcium Silicate Element Toughness Masonry
  • Orthotropic Behaviour
  • Core Testing Methods
  • Statistical Analysis
  • Correlation Study

Fingerprint

Dive into the research topics of 'Material characterisation of existing masonry: A strategy to determine strength, stiffness and toughness properties for structural analysis'. Together they form a unique fingerprint.

Cite this