Glass is produced by melting raw materials (mainly silica, soda ash and limestone), and often cullet (recycled glass), in glass furnaces of different sizes and technologies. Typical furnace designs include: cross-fired or end-fired with regenerative air preheat, recuperative heat recovery and fuel-oxygen firing (EU-BREF Glass, 2001). The industry is capital intensive, furnaces have a lifetime of up to 12 years and there are a limited number of technology providers. Natural gas and fuel oil are the main fuels used by the glass industry. Reliable international statistics on glass production are not available. The global glass industry is dominated by the production of container glass and flat glass. According to industry estimates the global production of container glass was 57 Mt in 2001 (ISO, 2004); production of flat glass was 38 Mt in 2004 (Pilkington, 2005). The production volumes of special glass, domestic glass, mineral wool and glass fibres are each smaller by roughly an order of magnitude.
Beerkens and van Limpt (2001) report the energy intensity of continuous glass furnaces in Europe and the USA as 4 to 10 GJ/t of container glass and 5 to 8.5 GJ/t of flat glass, depending on the size and technology of the furnace and the share of cullet used. The energy consumption for batch production is higher, typically 12.5 to 30 GJ/t of product (Römpp, 1995). Assuming an average energy use of 7 GJ/t of product, half from natural gas and half from fuel oil, yields an emission factor of 450 kg energy related CO2/t of product. Globally, energy used in the production of container and flat glass results in emissions of approximately 40 to 50 MtCO2 (11 to 14 MtC) per year. Emissions from the decarbonisation of soda ash and limestone can contribute up to 200 kg CO2/t (55 kgC/t) of product depending on the composition of the glass and the amount of cullet used (EU-BREF Glass, 2001).
The mid-term emission potential for energy efficiency improvements is less than half of what corresponds to the range of efficiencies reported by Beerkens and van Limpt (2001), which also reflect differences in product quality and furnace age. The global potential for emissions reduction from fuel switching is unknown. The main mitigation options in the industry include: improved process control, increased use (up to 100%) of cullet (Kirk-Othmer, 2005), increased furnace size, use of regenerative heating, oxy-fuel technology, batch and cullet pre-heating, reduction of reject rates (Beerkens and van Limpt, 2001), use of natural gas instead of fuel oil, and CO2 capture for large oxy-fuel furnaces. High caloric value biogas could be used to reduce net CO2 emissions, but potential new break-through technologies are not in sight.