Effects of recycled glass on properties of architectural mortar before and after exposure to elevated temperatures

• After exposure to 800 °C, increasing RG content improved the compressive strength.
• Increasing RG induced a more obvious improvement on the elastic modulus at 800 °C.
• The melting and pore-filling property of RG was responsible for the healing effect.
• SCAM with 100% RG retained a pleasant appearance even after exposure to 800 °C.

When put into practical uses, self-compacting architectural mortars (SCAM) containing recycled glass (RG) may face the risk of exposure to fire. It is of great interest to investigate the effect and the roles of RG acting in the SCAM after exposing to high temperatures. In this study, RG was used to replace river sand at ratios of 0%, 25%, 50%, 75% and 100% by weight. The results showed that at room temperature, the increase in RG content slightly reduced the thermal conductivity, compressive strength and elastic modulus of the SCAM samples, but had a negligible influence on the chloride-ion penetration rate. Although increasing the exposure temperatures resulted in reductions in compressive strength and elastic modulus of the SCAM samples, when the temperature was increased to 800 °C, the incurred melting and re-solidification process of the RG enabled a pore-filling effect to fill up micro-cracks and pores in the mortar matrix. Such beneficial effects were more obvious for elastic modulus than for compressive strength. It was also found that the 100% RG incorporated SCAM was able to maintain a pleasant appearance (negligible changes were induced in its appearance) after undergoing 800 °C exposure. This study demonstrated that the introduction of 100% RG in the SCAM allowed an attractive combination of high strength (45.7 MPa) and aesthetically pleasing appearance even after exposure to 800 °C. This paved way for further and broader applications of RG based SCAM products in the market. However, the melting properties of RG may incur adverse effects on the mechanical properties of the SCAM sample under loading conditions, which necessitate further studies to clarify such a potential limit.

Figure optionsDownload as PowerPoint slide