Three different window frames having the same layout (Figure 1), design, and thickness but different framing materials were designed. The window system comprised of a two-glass glazing system, an air gap, and a filler material (Figure 1). A frame must provide high thermal insulation and maintain the structural strength and rigidity necessary to support an evacuated glazing over its serviceable life. The material used for window frame and casement (shown in dark blue in Figure 1) were wood, aluminum, and unplasticized polyvinyl chloride (uPVC), respectively for three different window types. Subsequently, another combination of materials was modeled, where the window frame is made out of wood, while the casement is constructed out of aluminum, uPVC, and wood. The material used and their respective conductivities are listed in Table 1. The glazing system used was a double layer glass system with low emissivity layer on inside of the exterior glass. Thickness of the glasses was 4.7 mm each and the air cavity had a thickness of 16.5 mm.

After designing the window frame and the glazing systems, a performance evaluation of the window frames and window system was conducted. Calculation of thermal properties and energy performance was used to evaluate and compare each window type. First, the stand alone U values of the window frames were calculated. The U value of all frames was calculated as per EN ISO 10077-2 [12] with the simulation program Therm [13], which uses heat transfer coefficients prescribed by ISO 15099 [14] to solve for conductive heat transfer. The geometry of the profile (Figure 1) was drawn using computer aided design software and used as an underlay in Therm. Subsequently, the U value of the whole window system for the three different window types was calculated. Therm is based on the equations provided in the standard to calculate total U value for the window.

Following the common LCA methodology [15,16] the scope and goal of the study was to compare the environmental impact of three most common windows frames. Environmental impact of window frames was analyzed with the “Cradle-to-gate” variant, an assessment of a partial product life cycle from manufacture (“cradle”) to the factory gate (i.e., before it is transported to the consumer). The use phase and disposal phase of the product were omitted. The environmental burdens associated with each window frame were considered from raw materials acquisition, through the manufacture/processing stages, accounting for the production and use of fuels, electricity, and heat, as well as taking into account transportation/distribution impacts at all points along the product supply chain. Functional unit for the calculation was determined to be 1 m² of a window frame with the U value of 1.6 W/m²K. Based on the determined goal and scope of the study the life cycle inventory of input/output data for the LCA calculations was compiled. Data of energy inputs, raw materials, products, co-products, waste, and releases to air, water and soil and the upstream life cycle impacts of input materials were not analyzed specifically for this project. Instead, sound secondary life cycle data were sourced from Ecoinvent database 2.0. [17]. The data collected were modeled in Simapro [18].

The dataset included highly automated technology processes in window frame manufacturing, in Switzerland and Germany. For the wood window frame all the processes and material inputs needed to produce a wood window weighing 80.2 kg were included. Processes that were included were timber sawing, varnishing (primer, solvents, paint), selection bar rolling for steel fittings, joining, fitting, all the road transport at different production phases and the disposal of the paint remains. The PVC window frames weighed 94.5 kg, included the following processes: injection molding and extrusion of PVC, section bar rolling for steel fittings, section bar extrusion for aluminum parts, all the road transport at different production phases and the process heat waste. For the aluminum window frames the weight was 50.7 kg, included the following processes: section bar rolling for steel parts and fittings, section bar extrusion for aluminum parts, extrusion of HDPE plastic, surface treatment (powder coating), all the road transport at different production phases, the heat waste and the disposal of the plastic cuttings.

The data collected were modeled in Simapro [18]. Emissions and consumptions were translated into environmental effects, which were grouped and weighed. Carbon footprint was calculated with methodology IPCC 2001 GWP 100a V1.02 [19]. IPCC 2001 [19] contains the climate change factors of IPCC with a timeframe of 100 years. Emissions, which take place in the future, 100 years after the start of the process, were taken into account. IPCC characterization factors for the direct (except CH₄) global warming potential of air emissions were used. They do not include indirect formation of dinitrogen monoxide from nitrogen emissions, do not account for radiative forcing due to emissions of NOx, water, sulphate, etc. In the lower stratosphere as well as upper troposphere, do not consider the range of indirect effects given by IPCC [19], and do not include indirect effects of CO emissions.

The U values of the frames and the corresponding U values of the window systems are presented in Table 2. Moreover, the U values for just the frame alone (U_f) is also presented in Table 2. Considering the U values of the frames alone, the best insulated frame was the wooden frame followed by uPVC, and then aluminum. A high U value of aluminum frame was expected because it is a metal and has high conductivity. However, if only the casement is made out of aluminum with a wooden window frame, then the corresponding U value of the frame reduces significantly, from 11.86 to 3.51 W/m²-K. The impact on the U value of the window system, however, is minimal. Comparing these values to literature Appelfield et al. [8], it is found that window frame design for wood in this study has comparable thermal performance. Appelfield et al. [8] modeled a window with wooden frame and aluminum casement that performed almost identically to the one in this study. The window U values as expected are lower than the frame U values, which is in accordance to the literature. As the glazing systems were similar in all three types of windows, the window with wooden frame performed the best, followed by the uPVC frame.

Carbon footprint calculates the amount of greenhouse gas (GHG) emissions caused by a particular activity or entity, commonly also referred to as global warming potential (GWP). It is measured in tonnes (or kilograms) of carbon dioxide equivalent (CO₂eq.). In this study carbon footprint of different window frames were calculated (Table 3, Table 4 and Table 5). The carbon footprint of the analyzed aluminum window frame was 486.0 kg CO₂e, while the carbon footprint of plastic window frame and wood window frames were 258 and 130 kg CO₂e, respectively.

Carbon footprint calculates the amount of greenhouse gas (GHG) emissions caused by a particular activity or entity, commonly also referred to as global warming potential (GWP). It is measured in tonnes (or kilograms) of carbon dioxide equivalent (CO₂eq.). In this study carbon footprint of different window frames were calculated (Table 3, Table 4 and Table 5). The carbon footprint of the analyzed aluminum window frame was 486.0 kg CO₂e, while the carbon footprint of plastic window frame and wood window frames were 258 and 130 kg CO₂e, respectively.

In the aluminum window frame the biggest emission source was aluminum (69.8%), which was followed by glass fiber reinforced plastic (9.5%), section bar extrusion (8.0%), and powder coating, aluminum sheet (7.6%). All other emission sources together contributed 5.1% to carbon footprint. In the plastic window frame the biggest emission source was polyvinylchloride (45.4%), which was followed by steel (25.2%), extrusion of plastic pipes (8.0%), aluminum (3.6%), and zinc coating (3.6%). Among other emission source of plastic window frames contribution to the total carbon footprint above 2.0% was that of electricity (2.8%) and transport (2.3%). In wood window frame the highest contribution to carbon footprint had electricity (23.4%), which was followed by aluminum (20.1%), sawn timber (16.9%), alkyd paint (11.5%), steel (6.8%), and transport (3.8%).

The results were summarized in Table 6, which provides a side by side comparison of energy performance and carbon footprint of a window design alternative. Results showed that from environmental aspect wood window frames are the best choice among the analyzed window frames (Table 6). Furthermore, the embodied emissions analyzed do not include any offset for the carbon stored in the wood frame. Approximately 50% of dry timber is elemental carbon; thus, 1 kg of wood contains approximately 0.5 kg of carbon, which equates to 1.83 kg of CO₂e [20]. When calculating a carbon footprint, whether to include this stored carbon in wood (and, to a far lesser extent, small amounts of stored carbon in other materials) is a much debated issue [1]. If the carbon storage (sequestration) in the window frames would be included and so called NET carbon footprint calculated, the wood window frames would have even lower CO₂ emissions.

Among the frames studied for a similar thermal performance, aluminum has the highest environmental impacts. This is due to the energy extensive production processes and the pollutants resulting of the processes. Wood window frames performed the best environmentally as expected because processing and production of wooden frames is not as energy intensive as other materials. The results also concur with Asif et al. [9], who concluded that aluminum had the highest environmental impacts, while wooden windows have the lowest impacts. Citherlet [21] also confirmed these trends and showed similar results to the current study. The global warming potential, a sustainability metric, of wood: PVC: aluminum window frames is 1:11:26, respectively [11]. The corresponding carbon footprint ratio for wood: uPVC: aluminum frames as calculated is 1:2:3.74, respectively. Higher environmental performance of wooden frames coupled with the fact that the thermal performance of wooden window frame exceeds the performance of aluminum and uPVC, makes wooden frame a sustainable choice, holistically. This study performed a cradle to gate analysis. A recommendation for future studies will be to analyze the use phase as well as the end of life in order to provide longer-term implication of the choice of window frames.

Considering the carbon footprint and thermal performance of windows together and not in isolation will help make better design choices. Whole building energy simulations reveal that a drop of 31% in U value for window results in an energy saving of 6.5 kWh m⁻² for a building per year [8]. This will not only lead to energy costs reduction, but also reduce the energy production footprint. Incorporating LCA in the design process also leads to more environmentally conscious decisions. For example, a building component with low embodied environmental effects, such as wooden frames, can be replaced many times before totaling the high embodied effects of a material such as aluminum. If the aluminum ends up in landfill after 40 years of use, it was a poor choice from an environmental standpoint. The operational benefits coupled with the fact that carbon footprint of the material was considered in design process will have a significant impact in the sustainability performance of the building.