Hilal Ucar links glass performance to embodied carbon, showing that product selection can reduce emissions at the insulating-glass-unit level while still meeting solar and thermal targets. The key is to consider carbon alongside coating, thickness and build-up rather than after specification.
Carbon has become another glass-design parameter, but Hilal Ucar’s argument is more precise than simply choosing a product with a lower number. The effect of the base glass has to be read together with coating, laminate, pane thickness, manufacturing yield and the performance of the complete insulating-glass unit. A useful specification therefore asks not only how the glass performs in the building, but what material route produced that performance.
Lower-carbon glass changes the starting point
Saint-Gobain introduced ORAÉ in 2022 as a lower-carbon float-glass substrate. In Ucar’s figures, the 4 mm product combines 64% recycled content with renewable electricity and carries a reported footprint of 6.64 kg CO2e per square metre, compared with about 11 kg CO2e/m² for the conventional reference shown in the same comparison. That is roughly a 42% reduction before the glass is coated, laminated or assembled into an IGU. The distinction matters because base glass is often a large share of an IGU’s footprint. Ucar’s example assigns about 65% to the base glass, around 2% to coating and about 12% to lamination, with the balance coming from the rest of the build-up and production effects.
Closed-loop cullet is valuable, and scarce
The circularity challenge is not whether glass can be recycled, but whether clean end-of-life façade glass actually returns to flat-glass furnaces. Ucar describes a European market producing about 10 million tonnes of flat glass a year, with roughly 80% going into buildings. Against an estimated 1.5 million tonnes of end-of-life building glass, only a small fraction is recycled and an even smaller fraction returns in a closed loop to new flat glass.
That gap makes demolition sorting and deconstruction part of façade decarbonisation. Clean cullet can replace virgin raw material and reduce furnace energy. Ucar’s rule-of-thumb is that one tonne of cullet can avoid about 1.2 tonnes of raw-material input and around 700 kg of CO2.
The IGU has to be optimised as a system
A lower-carbon substrate does not remove the need for conventional façade engineering. Solar control, thermal transmittance, daylight, safety, acoustics and wind loading still determine the build-up. The carbon benefit can also be diluted if a low-carbon pane is paired with unnecessary thickness, excessive processing or poor cutting yield.
This is where specification becomes design rather than substitution. Ucar describes project examples in which changing the substrate while keeping required optical and thermal performance produced reductions in the order of 35–44% at IGU level; in a more optimised example, the combined effect reached roughly 60%. Those results are configuration-specific, not universal percentages.
Renovation can feed the next production cycle
The most consequential shift is to connect specification with recovery. Renovation produces a stream of old façade glass; processors need clean, traceable material; furnaces need cullet of suitable quality; and designers can create demand for products that use it. Each part of that chain affects whether circular glass remains a theory or becomes a repeatable supply route.
For design teams, the practical lesson is to put carbon data beside the familiar glass schedule. Compare product-specific declarations, avoid unnecessary mass and processing, and ask how demolition glass will be recovered. Lower-carbon glass becomes most useful when it is treated as one variable in a performance-and-material system rather than a late substitution.
Cutting yield belongs in the carbon conversation
Ucar also warns that the declared footprint of a square metre of glass is not the same as the footprint of the glass that finally reaches the façade. Large or irregular pane geometry can create offcuts, and every discarded area has already carried raw-material and furnace energy. A nominally low-carbon product can therefore lose part of its advantage if the cutting pattern is inefficient. That makes optimisation a geometry exercise as well as a procurement decision. Module sizes, pane nesting and the ability to return clean process cullet to the furnace should be considered together. Carbon data is strongest when it follows the real material flow from batch to coated sheet, cut pane, IGU and eventually back to recovery.