Conventional LCA models include additives such as flame-retardants in consumer products only as a cost, i.e. the environmental benefit of the additive is not taken into account. Recently a Fire-LCA model was developed that also includes fires and their impact on the environment. This model has previously been applied to TV-sets1 and cables2 and has now been applied to furniture.
LCA methodology
LCA methodology is in itself a complex model for evaluating different parts of a given product's whole life-cycle - from the cradle to the grave. Much of the LCA methodology has been defined internationally and it is timely to add new finesses, such as the inclusion of accident analysis. The increased potential fire risk in the absence of flame retardants can be calculated using statistics or creating model systems. This, combined with measurements from full size fire tests, allows the assessment of the environmental effect of flame retardants from a broader perspective than previously possible.
In this project a sofa was chosen as the functional unit. Fire statistics from the UK were used to assess the changes in the number and extent of fires resulting from a move to fire resistant furniture. The UK introduced in 1988 Furniture Fire Safety Regulations requiring domestic upholstered furniture to resist ignition by a small flame, whereas previously in the UK, and still today in the rest of Europe, furniture was only resistant to ignition by a smouldering cigarette and could readily be ignited using a small lighter. Analysis of UK fire statistics trends prior to these Regulations, and from 1988 to 2000, has enabled a robust statistical model to be developed, with results being little affected by the scenarios used.
Data were gathered on furniture production for the sofa model studied, and for end-of-life disposal (with and without flame retardants) to provide LCA input data for all phases of furniture life from the cradle to the grave. The pollutant emissions from accidental fires, of different gravity (extent of fire), involving flame retarded or non-flame retarded furniture were obtained using full scale fire tests on both commercial sofas and on fully furnished test rooms at the SP fire test facilities, Borås, Sweden. A non-flame retarded sofa was used to represent the mainland Europe market and two different (commonly used) flame retardant procedures were used to represent the UK market.
The furniture manufacture phase was the principal source of energy consumption and emissions of CO2, NOx, carbon monoxide, were similar for the flame retarded (FR) and non-FR furniture. Accidental fires, however, were responsible for a significant portion of the emissions of hydrogen cyanide, polycyclic aromatic hydrocarbons and TCDD/TBDD-equivalent (i.e., chlorinated and brominated dioxin/furan equivalents).
Calculated total sofa life cycle TCDD/TBDD-equivalent emissions were higher for the flame retarded sofas, both due to emissions calculated for end-of-life incineration (hypothesis: 30% of sofas go to domestic waste incinerators) and emissions from the few fires containing these sofas. The estimate of emissions from incineration assumed that they are proportional to input Cl/Br levels.
In contrast, the total polycyclic aromatic hydrocarbon (PAH) emissions were higher for the non-FR sofas, because of the higher frequency and gravity of accidental fires. A comparison between the relative importance of PAH and dioxin and furan emissions using a modified cancer risk model clearly shows that the level of PAH emissions is of far greater significance than that of dioxins and furans. Thus, based on PAH emissions the environmental risk associated with flame retarding a sofa is outweighed by the impact of fire emissions associated with non-flame retarded sofas.
Further, the total sofa TCDD emissions calculated for the worst-case over the whole life cycle (manufacture, accidental fires, disposal) correspond to approx. 0.003% of total annual TCDD emissions from all sources. Similar data is not available for TBDD-equivalent emission or PAH emissions.
Results presented in SP report 2003:22
Full details of the project results are available in SP report 2003:22. The report can be ordered from Erika Hjelm at a cost of SEK 500.
Photos from appendix E of SP report 2003:22 can be downloaded here.
A summary report has also been prepared which can be downloaded here.
For further information, please contact Petra Andersson.
Simonson, M., Blomqvist P., Boldizar A., Möller K., Rosell L., Tullin C., Stripple H. and Sundqvist J O, "Fire LCA Model: TV Case Study" SP Report 2000:13
Simonson, M. Andersson P., Rosell L., Emanuelsson, V. and Stripple H., "Fire-LCA Model: Cables Case Study" SP Report 2001:22