Embodied Carbon - what is it and how to compare materials? Sustainable building materials, Part 4
Originally posted 16th October 2020
EDIT 24/02/2022: the charts below still a broad overview of the situation, but much newer and more accurate data is available for a number of the materials shown. I’m leaving the charts here for now, but please don’t rely on them for comparison. I’ll try and write a new blog with up to dat info soon.
The charts below were originally compiled for work with the School of Natural Building on the UpStraw project. I shared early versions of the first two on twitter and I am very grateful for the instant peer-review that provided, which informed the versions posted below.
5 minute read
Embodied Carbon
Sustainable building must mean producing as little CO₂ as possible (see more in the 2nd blog of this series https://www.sustainablebuildconsultancy.com/blog/what-does-sustainable-mean). That means low or zero carbon emissions from both construction and use of a building. Choices of material can have a huge impact on this.
The carbon emissions generated through production, transport, use, and disposal of a material are known as Embodied carbon (EC). The amount of energy used in those processes (the embodied energy) - and the fuel used to provide it - determines the embodied carbon, calculated from the carbon intensity of each energy source.
Generally the higher the embodied energy - the higher the embodied carbon, but as more processes are electrified and the carbon intensity of power generation decreases, the two can become disconnected. It is important to use materials that have low embodied energy as well as low embodied carbon, in order to ensure continued energy security (i.e. that there is enough electricity available for all needs at any time). For this blog though, I’m focussing on embodied and operational carbon.
How to compare materials?
As with many aspects of sustainability, meaningfully comparing embodied carbon figures can be tricky. The first chart below shows the embodied carbon of a selection of building materials, given as kgCO₂e/kg (the number of kilos of CO₂ emitted per kilo of material; the ‘e’ stands for ‘equivalent’ - other gases produced are converted to an amount of CO₂ with an equivalent global warming potential). This provides a basic overview of relative emissions but ignores the fact that very different quantities of each material might be used, or that they may have very different material densities. Comparison of kgCO₂/kg of material may make a useful starting point, but should be viewed with caution.
The data in this table is ‘cradle to gate’ CO₂e - the emissions caused by extraction (the cradle), transport and processing of raw materials, and manufacture of the product, up to the point it leaves the factory (the gate). All emissions after that point are ignored, which can give a skewed image (more on this in the next blog).
The data in this and the next chart is taken mainly from the Inventory of Carbon and Energy (ICE) V3.0 unless stated (Available here: http://www.circularecology.com/embodied-energy-and-carbon-footprint-database.html), with further data from Environmental Product Declarations (EPDs). I’ve put the full list of references at the end of the blog.
EDIT (19.10.20): Some of the figures represented will vary a lot depending on exact specification or source used, in particular concrete, where the exact mix and type of concrete can increase the CO₂e significantly. Ideally, the figures for the exact materials used in a build should be checked.
EDIT (16.11.20) kgCO2e/kg chart updated with new clay plaster data from Clayworks who have just released a full EPD for their products. I hope more ‘sustainable/natural materials’ producers will follow suit.
Plant-derived materials store CO₂ that the plants absorbed during their growth. This is known as sequestration of carbon - the capturing and storage of atmospheric carbon. Whether this carbon remains stored depends on what happens to those materials at the end of the life of a building. If the materials are incinerated or allowed to rot then the carbon is released. The chart above shows the embodied carbon for such materials with and without sequestration (the dark grey representing the figures without). Because of the uncertainty about length of sequestration, it is important to use materials that have low embodied carbon regardless of carbon storage, so that the emissions remain lower even at end of life.
For insulation products, a more helpful comparison can be to compare the CO₂e per square metre of wall area, with quantities of each material adjusted to provide the same U value (insulation value, explained more here: https://www.sustainablebuildconsultancy.com/blog/thermal-bridging-in-regs). The operational energy use and associated carbon of a building constructed with that U value would then be the same regardless of chosen insulation material.
The chart below gives kg CO₂e per square metre of wall area, to give a U-value of 0.12 W/m²K. As above, amounts are shown with and without sequestration.
This is more helpful than simple kg CO₂e but is still not perfect. For example, foamed glass block comes out pretty badly from this comparison, but in reality it would never be used as the main insulation material (it is used only in small quantities when something insulating, non-capillary and load-bearing is needed). Some of these materials can function as both insulation and structure simultaneously (reducing need for providing structure separately with potentially higher-carbon materials), while others can only provide insulation.
When I compiled the chart, I was surprised by the figures for extruded polystyrene (XPS) and foamed glass aggregate (FGA). FGA is made from the proportion of recycled glass that can’t be reused as glass, whereas XPS is an oil-derived product. My instinct said FGA: good, XPS: bad. The data made me think again, as clearly XPS comes in as slightly lower embodied carbon than FGA. Both are used to provide load bearing insulation beneath a floor slab. (More on this in the next blog).
Of course, carbon is not the full story. Resource depletion and ecological harm are the other two key indicators of sustainability (https://www.sustainablebuildconsultancy.com/blog/what-does-sustainable-mean). It’s important to consider these aspects too, though it’s beyond the scope of this blog post. Environmental Product Declarations provide information on this, and are a very useful reference where available. More products do have EPDs now, and hopefully the number will keep growing. They provide a useful means to cut through the green-wash and see the data for yourself.
So, how to account for all these nuances and different stages in a way that provides comparable results encompassing them all? The answer is life cycle assessment - the total carbon emissions of a building, including materials, construction, operation, demolition and removal. I discuss this in the next blog, here:
https://www.sustainablebuildconsultancy.com/blog/lca1
Data sources and assumptions:
Embodied carbon equivalent data from Inventory of Carbon and Energy (ICE) V3.0 (Hammond and Jones, 2019) unless stated. Available here: http://www.circularecology.com/embodied-energy-and-carbon-footprint-database.html
Where a range of figures is given in ICE, reported figures are mostly 'General' figures. Specific types may have lower EC, especially if recycled. All info is cradle to gate.
Density data from CIBSE Guide A (2015) or manufacturers data where not listed in CIBSE.
Ecococon data: http://naturalbuilding.fi/wordpress2018/wp-content/uploads/2018/09/EPD-EcoCocon-Straw-Panel_final.pdf
Hempcrete data: https://limecrete.co.uk/hempcrete-factsheet/#more-279
Lime plaster calculated from ICE data for sand and lime, based on plaster/render mixes of 3 parts sand to 1 part lime by volume.
Clay plaster data is average from two plaster types on Clayworks EPD: https://clay-works.com/wp-content/uploads/2020/11/Clayworks_Environmental_Product_Declaration.pdf
Cement render calculated from ICE data for sand, ordinary portland cement, and lime, based on plaster/render mixes of 4 parts sand to 1 part cement and 1 part lime by volume.
Cellulose data: https://www.igbc.ie/wp-content/uploads/2017/10/Warmcel-European-Performance-Declaration.pdf
XPS data: https://www.jackon-insulation.fr/telechargements-et-services/download-settings/Download/download/epd-jackodur-plus/
Woodfibre data: https://www.steico.com/fileadmin/steico/content/pdf/Certificates_-_Documents/English__multiple_markets_/STEICO_EPD-STE-20150327-IBD1-EN.pdfFoamed glass aggregate data: https://belglas.files.wordpress.com/2017/12/glapor-cellular-glass.pdf
Foamed glass block data: https://www.foamglas.com/-/media/project/foamglas/public/shared/files/certifications/b---common-certificates/epd-environmental-product-decl/epd-foamglas-t4plus-2015-en.pdf?la=en-gb
