Methodology for on-site assessment of the reuse potential of post-consumer tempered glass

Abstract

Flat glass is a widely used material in the construction industry. However, its production has a relatively high carbon footprint due to the high melting temperature required, and most of it is still sent to landfill or down-cycled at end-of-life. The reuse of glass therefore has a great potential, given its lower carbon footprint and the number of existing buildings with glass panes that would be suitable for reuse.
This research project aims to propose a novel methodology for estimating the potential of tempered glass for adaptive reuse. It is illustrated with a case study of a historic building in Paris, built over 50 years ago with about 2500 fully toughened glass panes for potential reuse.
Firstly, a brief environmental impact assessment quantifies the impact of reuse compared to recycling and down-cycling to provide a more critical understanding of the benefits of reuse.
Secondly, a probabilistic equivalence of the Heat-Soak Test (HST) is justified for thermally toughened glass. Based on statistical analysis of spontaneous breakage due to nickel sulphide (NiS) inclusions over time, a safety level over time is estimated. This level of safety can already be considered quite high from 20 years after installation and is equivalent to an HST after 50 years.
Finally, the residual strength of aged tempered glass is estimated. This strength is based on the combination of the pre-stress level and Linear Elastic Fracture Mechanics (LEFM). The pre-stress level and the largest flaw depth are measured on-site using portable optical devices on a representative sample. A visual inspection procedure is then defined to speed up the evaluation of the flaw depth. Once trained, the observer should be able to qualify the glass panes relatively quickly following different 'Quality Levels' with associated characteristic strengths (95% confidence level).
This methodology was applied to the case study. Pre-stress and flaw depth measurements were carried out on 70 glass panes. Visual inspection was tested and the margins calibrated to reach a 95% confidence level. The glass panes have an average surface pre-stress level of 81.41 MPa with a standard deviation of 4.17 MPa, which is consistent with other values reported in the literature for fully toughened glass. The largest flaw depths ranged from 4 µm to 147 µm, also in agreement with other publications. All of the glass panes analysed are suitable for reuse as heat-strengthened glass, and potentially better for some of them.
The proposed methodology has been summarised to be generalised to other reuse projects. Typical fully toughened, heat-strengthened and annealed glass (average pre-stressing level and surface quality based on previous measurements) have a great potential for reuse at the same characteristic strength if the surface quality estimated by visual inspection has flaw depths smaller than 20-40 µm. Qualification for reuse of these three types of glass with a reduced strength is also proposed. This could allow reusing glass panes with flaw depths of up to 500 µm for several design scenarios.