Repository logo
 

Blast Resilience of Glazed Façades: Towards a New Understanding of the Post-Fracture Behaviour of Laminated Glass


Type

Thesis

Change log

Authors

Angelides, Socrates 

Abstract

Laminated glass panels are increasingly used to improve the blast resilience of glazed facades, as part of efforts to mitigate the threat posed to buildings and their occupants by terrorist attacks. These composite ductile panels offer superior blast resistance and result in reduced glass-related injuries, compared to the inherently brittle, monolithic glass, which has historically been used in building facades. This is due to the interlayer’s ability to both provide residual resistance, following the fracture of the glass layers, and retain glass fragments. Therefore, such panels can act as the first barrier of defence during an external explosion to prevent the blast waves from penetrating the building interior and protect occupants.

The blast response of these ductile panels is still only partially understood, with an evident knowledge gap between fundamental behaviour at the material level and observations from full-scale blast tests. To enhance our understanding, and help bridge this gap, this thesis adopts a ‘first principles’ approach to investigate the effects of high strain-rate and inertia loading associated with blast loading. The former is studied by developing simplified analytical beam models, for all stages of deformation, with the focus on laminated glass with polyvinyl butyral, as this is the most commonly used interlayer in building facades. The models account for the enhanced properties of both the glass and the interlayer at high strain-rates. This enhances the residual post-fracture bending moment capacity, arising from the combined action of the glass fragments in compression and the interlayer in tension, which is considered negligible under low strain-rates. The post-fracture resistance is significantly improved by the introduction of in-plane restraint, due to the membrane action associated with panel stretching under large deflections. This is demonstrated by developing a yield condition that accounts for the relative contributions of bending and membrane action, and applying the upper bound theorem of plasticity, assuming a tearing failure of the interlayer.

To validate the post-fracture capacity predicted by the derived analytical models, three- and four-point bending tests are performed at low temperature on specimens pre-fractured before testing. The pre-fracture ensures controlled and repeatable fracture patterns, and the low temperature simulates the effects of the high strain-rates that result from short-duration blast loads by taking advantage of the time–temperature dependency of the viscoelastic interlayer. A new time–temperature mapping equation is derived from experimental results available in the literature, to relate the temperatures and strain-rates that result in the same interlayer yield stress. The results of the low-temperature tests demonstrate an enhancement of the ultimate load capacity of the fractured glass by two orders of magnitude, compared to that at room temperature. Due to the time–temperature dependency of the interlayer, a similar enhancement is therefore anticipated at the high strain-rates associated with typical blast loading, as predicted by the analytical models. Additionally, comparable moment capacities were observed between the experimental results and the analytical models, and tearing failure consistently occurred in the experiments at the plastic hinge locations predicted by the analytical models. The experiments also demonstrated that the post-fracture moment capacity is unaffected by the number and size of the glass fragments. However, the moment capacity is influenced by the crack alignment between the glass layers, with significantly higher moments recorded for specimens with misaligned cracks compared to specimens with aligned cracks.

To investigate the effects of inertia that are known to be significant under the accelerations experienced by a panel during a typical blast event, laminated glass specimens were impacted with polymer foam projectiles, launched from a gas gun. These tests aimed to simulate the loading from a blast pulse. The dynamic bending response was subsequently recorded with a high-speed camera. Nine different test types were performed by varying the boundary conditions, impacting both intact and pre-fractured specimens and testing glass specimens with different cross-section sizes. It was found that the collapse mechanisms formed in laminated glass specimens under short-duration pulses depend on the intensity of the loading. Under high intensity loading, the panel can resist pressures greater than the static collapse load, due to the effects of inertia. This results in a more localised collapse mechanism, compared to quasi-static loading, and explains the repeated failure pattern observed in blast tests of laminated glass panels. The incorporation of axial restraint results in a significant membrane contribution to the response, and therefore reduced deflections.

Description

Date

2022-01-31

Advisors

Talbot, James

Keywords

Blast, Laminated glass, Post-fracture

Qualification

Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
Sponsorship
EPSRC (1817334)
Engineering and Physical Sciences Research Council (EPSRC) funding through the EPSRC Centre for Doctoral Training in Future Infrastructure and Built Environment (FIBE CDT) at the University of Cambridge (EPSRC Grant Reference No. EP/L016095/1). ICE Research & Development Enabling Fund.