Controlling Sustainability through Structural Design – Concrete E-Learning Platform

A Performance-Based Analysis of Material Emissions and Data Dispersion

Link zur deutschen Version: Steuerung der Nachhaltigkeit durch den Tragwerksentwurf

The construction industry faces considerable challenges, as construction activity will increase significantly in the coming years to maintain ageing infrastructure and meet the growing global demand for residential and commercial space. At the same time, this opens the door to further developing construction processes sustainably and significantly reducing the industry’s ecological footprint.

The ecological sustainability of structures can be significantly influenced during the design phase. As part of a research project, Professor Walter Kaufmann’s research group at the Institute of Structural Engineering at ETH Zurich is examining sustainability in structural design independently of the materials used. The project was launched in collaboration with industry partners and aims to identify the most important levers available to structural engineers for the design of structures, in a practice-oriented manner, and to quantify their effects using scientific methods. In future, this knowledge will be passed on to both prospective and practising engineers in block and continuing education courses.

To make an accurate assessment of the ecological sustainability of a supporting structure, it is necessary to consider not only the emissions from building materials but also the direct and indirect influences on a building’s environmental impact. These are shown conceptually in Figure 1 using the example of floor slabs [1].

Figure 1: Conceptual equation for assessing emissions using the example of floor slabs [1].

The simplest and probably most commonly used lever in construction today for minimising CO₂ equivalent emissions– also known as global warming potential (GWP) – is the choice of low-emission materials, which, in the conceptual equation according to Figure 1, is particularly reflected in factor (i) emissions per mass. However, factor (ii) mass per floor area is just as important for a representative comparison of different building material and structural variants. It reflects how much mass of a building material, e.g. per floor area, must be used to meet all requirements for structural safety and serviceability. In addition to material density, the volume of the material used also affects this factor. The higher the static performance of a material (in terms of strength and stiffness), and the more efficient the cross-sectional shape or static system, the lower the required volume. Added to this is factor (iii), the indirect influences of a particular load-bearing system [1]. For example, lightweight ceiling systems can significantly reduce a building’s dead weight, resulting in substantial savings in materials and foundation emissions. The remaining factors (iv) floor space per use and (v) service life are also relevant for emissions, but depend primarily on usage requirements and only to a limited extent on the structural design. A regular supporting structure can promote subsequent conversion and thus also a longer service life.

In the following descriptions, CO₂-equivalent (GWP) values are used to quantify environmental impact.

About the sustainability indicators of building materials

The ecological footprint of building materials depends heavily on the production processes, the raw materials used in manufacturing, and the energy requirements and sources. Accordingly, the ecological footprint varies depending on the place of manufacture, manufacturer and product. The characteristic values used in Switzerland today from the list of life-cycle assessment data for the construction sector by KBOB and ecobau [2] represent average environmental impacts, not their dispersion. However, it can be assumed that knowledge of the dispersion and its magnitude would lead to a significantly enhanced understanding of the relationship between materials and their dispersion ranges, and consequently to a more targeted selection of materials. To visualise variation among the building materials currently available on the market, this project compiled sustainability parameters for building products relevant to structures and widely used in Switzerland from environmental product declarations (EPDs) issued by individual manufacturers and associations¹ ².

The dispersion is shown in the following figures for the data basis described for each building material (see Table 1). However, this must be interpreted with caution, as it is – as mentioned – a mixture of EPDs for individual products and average values. Therefore, no qualitative statement can be made about the range of dispersion ³.

Group of construction products	Description and strength classes	Number of EPDs	Origin
Concrete	Structural concrete for buildings NPK A – C	3	CH
Reinforcement steel	B500B	12	E
Timber and wood-based materials	KVH, BSH, BSP GL24h-GL32c und C24	22	EU, CH
Structural steel	S235, S355, S460	12	EU, CH, Turkey
Prestressing and posttensioning steel	Y1770-2160	5	EU

Table 1: Data basis per building material

Figure 2 shows the histograms of the data described above, separated by material. The points represent the values considered for the individual and average EPDs. The grey shaded area contains 80% of the data points (between the 10% and 90% quantiles). For direct comparison, the KBOB values and the average values from the EPDs are shown as vertical lines. The graph shows mean values, since KBOB also operates on a mean-value basis. In contrast to the weighted KBOB mean values, the mean values from the collected EPDs have not been weighted. To give less weight to outliers, medians are used instead of means from Figure 3 onwards.

Figure 2: Dispersion of the emission factor GWP [kg-CO₂-eq/t] of selected building materials.

A comparison of the values shows that the KBOB values for reinforcing steel deviate significantly from the EPD data found. It should be noted that the concrete values are already averaged EPDs. For timber and wood-based materials, the KBOB values fall within the 80% range of the EPDs considered. For structural steel, the KBOB value is very close to the average EPD value used.

It should be noted that the EPDs and the life cycle assessment data published by the KBOB are based on different methodological principles and are therefore not directly comparable. While EPDs are prepared in accordance with the EN ISO 14025 standard and follow a standardised procedure across Europe, the KBOB applies its own methodology, which differs from this standard, particularly with regard to system boundaries and the definition of energy flows. As the KBOB does not present its methodology and the source data used in a complete and independently verifiable manner, it was not possible to examine the deviations in more detail.

In addition to a comprehensible data basis, the project team considers it essential that the existing variation (due to actual differences in production processes) and additional uncertainties (due to incomplete or uncertain databases) are transparently disclosed for each sustainability indicator per building material, in order to enable reliable sustainability comparisons based on this data. This is particularly important given that many stakeholders and decision-makers in the construction industry, such as architects, civil engineers and building contractors, often do not have in-depth training in sustainability assessment, but have to make decisions based on this data – typically in the early planning stages, when no specific products or even manufacturers can be determined, but only the basic choice of materials. Since the variation ranges of individual materials are considerable, life cycle assessments of structures that only account for the average GWP of the respective material do not adequately reflect reality and thus obstruct a well-informed decision.

Comparison of the sustainability and performance of building materials

Figure 3 shows the sustainability of various building materials (as GWP) in relation to their mechanical performance (stiffness and tensile and compressive strength). The GWP was calculated based on the collected EPD data. The variation at material level is illustrated by showing the range between the 25% and 75% quantiles.

For the illustration in Figure 3(a), the GWP is normalised by the stiffness (modulus of elasticity) of the respective material. It can be seen that the concrete type NPK C C30/37 has the best ratio of stiffness (modulus of elasticity) to GWP, followed by structural timber C24. The median for glued laminated timber GL24h is 87% higher than that for NPK C concrete and 25% higher than that for structural timber C24. Reinforcing steel B500B and structural steel S355 have the highest emissions per unit of stiffness, followed by prestressing steel Y1860.

At first glance, stiffer materials (prestressed steel, structural steel, reinforcing steel) appear to have significantly higher GWPs. However, an examination of timber, with less stiffness than concrete, shows that there is no direct correlation with the stiffness of the materials. Figure 4(d) shows a correlation between material density and GWP. The differences in GWP between the various steel processing methods can be explained by the use of primary steel for prestressing steel and secondary steel for reinforcing and structural steel. This clearly shows the potential offered by the use of recycled steel.

Figure 3: Comparison of the GWPs of building materials in relation to (a) stiffness, (b) compressive strength, (c) tensile strength and (d) volume of the respective building material, shown as a box plot with median, 25% and 75% quantiles, upper and lower neighbourhood values and statistical outliers (as circles).

In Figure 3(b) and (c), the GWP is normalised by the design values of the compressive and tensile strengths for the materials under consideration, with only the compressive strength for concrete and only the tensile strength for prestressing steel. It can be seen that structural timber C24 and glued laminated timber GL24h perform best in terms of compressive strength. In terms of tensile strength, prestressed steel has the lowest GWP-to-strength ratio.

The comparison of the volume-based GWP, as shown in Figure 3 (d), does not allow for any performance-based conclusions to be drawn. Only in combination with the requirements for the supporting structure can conclusions be drawn from this information regarding the associated greenhouse gas emissions.

The underlying data vary considerably, leading to substantial overlap among the individual materials. It is also apparent that materials perform differently under different load types and are therefore favourable for the GWP.

As described above, due to the different data bases for each building material with individual product and already averaged EPDs, an interpretation of the actual dispersion of the materials is only possible to a limited extent. The specified dispersion reveals the potential of the different materials, but it is not possible to make a qualitative statement about the range of dispersion.

Comparison of performance in components subjected to normal forces

In the performance-based comparison of different building materials presented below, it is assumed that a CO₂ budget of 100 kg CO₂-eq is available for each material to produce a compression or tension member of 1 m in length. The available volume of the building material determines the respective cross-sectional area of the component (see Figure 4). The resulting load-deformation behaviour can be used to assess the parameters relevant to a supporting structure – strength, stiffness and ductility.

Figure 4 shows the result of such an analysis for (a) tensile and (b) compressive load for a component made of reinforced concrete (NPK C, reinforcement content ρ_s = 8% in tension and ρ_s = 0.6% in compression), prestressed concrete (NPK C, prestressing steel Y1860 prestressed to s_c = 10 MPa with a reinforcement content of ρ_p = 0.9%), structural timber C24, glued laminated timber GL24h and structural steel S355. The median (solid lines) and the 10% and 90% quantiles (dashed lines) of the available EPD values for the building materials were used for the calculations (see Figure 1). In the case of reinforced concrete tension members, the tensile stiffening effect of the concrete (bond between reinforcing steel and concrete) was taken into account by applying the tension chord model. The material law of concrete in the compression zone was modelled in accordance with the fib Model Code 2010 [5]. Any (premature) failure due to stability issues was not accounted for in the compression load.

It is clear that, as a result of the variation in CO₂-eq values, the load-deformation behaviour ranges overlap (with the exception of concrete, for which no variation data is available). Glued laminated timber and structural steel have the highest tensile strength. Timber and prestressed concrete exhibit the greatest stiffness under tensile load in service, while timber and reinforced concrete exhibit the greatest stiffness and a high load-bearing capacity under compressive load. Steel, prestressed concrete and structural steel behave as expected in terms of ductility, while wood-based materials are characterised by brittle failure (brittle fracture at maximum load).

Figure 4 c shows the cross-sections resulting from the available 100 kg CO₂-eq and their dead weight. This shows that significantly different volumes per material are possible with the same CO₂ budget (the maximum ratio of the cross-sectional area of structural timber to prestressed steel is 328:1). The dead weight of the girders also varies greatly. Prestressed concrete is an order of magnitude lighter than the other cross-sections, followed by steel. The timber cross-sections and the reinforced concrete cross-section are comparable, with the reinforcement content or proportion of concrete having a major influence. The deeply reinforced, concrete-intensive cross-section is more than twice as heavy. Such a performance-based comparison enables differentiated evaluation of building materials and their use according to their respective advantages.

Figure 4: Load-deformation diagrams of (a) tension and (b) compression members made of structural steel, reinforced concrete (slack reinforced and prestressed), glued laminated (glulam) timber and structural timber with a consumption of 100 kg CO₂ eq. per running metre; (c) corresponding cross-sections and dead loads.

The distributions shown indicate that manufacturer-specific production parameters can significantly affect the sustainability of the building materials used. Depending on the manufacturer from whom the building material is sourced, it may be at the upper or lower end of the distribution range. In CO₂-conscious planning, there is great potential here to replace a product at the upper limit of the spread with an equivalent product from another manufacturer with a lower GWP by specifically controlling the required maximum CO₂ emissions. This could result in significant savings without changing the materialisation itself. In addition, building owners’ specifications for maximum CO₂ budgets within a material group would create targeted incentives for manufacturers to further reduce the emissions of their products.

Conclusion

Both public and private decision-makers rely on sound expertise and advice from planners to make the right decisions about low-emission and resource-efficient construction in the early stages of a project. A performance-based comparison of building materials enables a differentiated assessment, allowing materials to be used in a targeted manner according to their respective advantages.

It should be noted that at the time of designing a supporting structure, the emissions of the selected building material can only be roughly estimated. The variations within the respective material groups identified in this project show that manufacturer-specific parameters can significantly affect the sustainability of the products used. Failure to account for this variation risks misinterpreting the life-cycle assessments of structures. It should also be borne in mind that considering sustainability at the material level alone is not effective, as building materials have different mechanical properties, which greatly influence material consumption and thus the GWP of the structure.

Knowledge of dispersion within a single material group could also be used to select a product from a manufacturer with a comparatively low GWP. Currently, however, the actual CO₂ equivalence values can only be determined once the products actually used have been specified. And even then, only if separate EPDs are available for these products. Nevertheless, procurement should enable specifying maximum permissible environmental impacts for a particular material to actively control the CO₂ emissions that actually occur.

Outlook

This article deliberately does not consider CO₂-optimised concretes, as these are not currently available in large quantities on the market. It is expected that, with the entry into force of Annex ND of SN EN 206:2013, concretes with significantly reduced emissions will be available in the near future, with an impact on the performance-related sustainability of reinforced and prestressed concrete. These potential developments will be included in future considerations as soon as data becomes available.

In addition to the parameters already discussed, the choice of the structural system has a significant influence on the efficiency and, consequently, on the emissions of the supporting structure. In the case of floor slabs – which are of central importance for sustainability due to their large mass share in most buildings – particular attention must be paid to (impact) sound insulation, fire protection, thermally activatable mass and susceptibility to vibration, see e.g. [4]. If the relevant requirements are not met with a statically optimised design, the component thickness of the supporting structure must be increased or additional non-load-bearing layers must be added to the floor structure, which greatly impairs sustainability.

The next step in the research project is therefore to compare typical ceiling systems made of different materials and examine their advantages and disadvantages across different spans and structural systems. At the same time, a life-cycle assessment of typical structures will be conducted (taking into account previous findings on the dispersion of material characteristics). The aim is to analyse, on the basis of completed structures, the typical emission values of completed buildings and to identify the potential savings that could be achieved through a optimised choice of cross-section, system or material. Indirect influences such as the floor structure or storey height, the presence of a transfer slab, the need for load-bearing partition walls, etc. are also the focus of this part of the project.

The final phase of the project will focus on developing a block and continuing education course for students and planners, as well as creating a design guide to share the knowledge gathered in the previous phases.

Funding

The project is largely financed by project partners. The financing status is currently around 80%, and we are looking for additional partners from the construction industry to cover the remaining 20%. We would be particularly pleased to welcome a partner from the timber construction industry to the project.

A big thank you goes to our existing partners, whose contributions have made this project possible:

Anliker AG
Debrunner Acifer AG
Holcim (Switzerland) AG
Implenia AG
Marti Bauunternehmung AG
TFB AG
Walo Bertschinger AG
armasuisse Real Estate

Research team: Sustainability in Structural Design

References

W. Kaufmann, Ökologisch nachhaltige Geschossdecken, White Paper, 2024, based on L. Gebhard, Reinforcement strategies for digital fabrication with concrete, Diss. Nr. 29201, ETH Zürich, 2023.
Ökobilanzdaten im Baubereich, Version 7, Bundesamt für Umwelt (BAFU), KBOB, Amt für Hochbauten Stadt Zürich und Verein ecobau.
Umwelt-Produktdeklaration nach ISO 14025 und EN 15804, SÜGB, Schwanengasse 12, CH-3011 Bern, 22.05.2023.
R. Ammann, Correlations of requirements and performance metrics for concrete floor slabs, Proceedings of the 15th fib International PhD Symposium in Civil Engineering, 2024.
fib-fédération internationale du béton. fib model code for concrete structures 2010. John Wiley & Sons, 2013.