Note on Best Practice: ‘Best practice’ for waste incinerator plants typically refers to the European ‘Best Available Techniques’ (BAT) Reference Document for Waste Incineration.
Note on Units: We have used the units ‘ng/m3’ interchangeably with ‘ng I-TEQ/Nm3’. They are technically slightly different. I-TEQ weights each dioxin and furan variant based on its toxicity relative to the most toxic one (TCDD). It can be thought of like a toxicity multiplier so you get a risk-adjusted total. The Nm3 (as opposed to m3) adjusts the gas volume to standard conditions (0°C and 101.325 kPa) to make measurements comparable across time and location. ng I-TEQ/Nm3 is the more commonly accepted, ‘correct’ unit to use to measure dioxin and furan concentrations, and is ultimately what will probably be used by Energy-to-Waste plants in NSW. However, NSW EPA Policy uses ng/m3.
Emissions Monitoring
Long-term sampling vs short-term sampling
EPA Policy requires Energy from Waste plants in NSW to measure dioxin and furan (PCDD/F) emissions four times in the first year of operation, and twice per year after that. The sampling period is 6-8 hours. This is known as ‘short-term sampling’ or ‘periodic sampling’
The NSW EPA Policy falls below the BAT standard (BAT 4) which calls for monthly sampling with a sampling period of 2-4 weeks (known as ‘long-term sampling’). The BAT then recommends moving to short-term sampling only if emissions are proven to be sufficiently stable.
The NSW EPA decision is based on an independent expert review by authors from the University of Sydney, which claimed that the results from short-term and long-term sampling ‘do not differ significantly’. This is based on a flawed interpretation of a European study. The European study compared emission levels measured with short-term and long-term sampling, with data from 142 reference lines in Belgium and France. The study found that there is little difference between the two sampling methods when measuring low levels of PCDD/F emissions, but high PCDD/F emission levels were more likely to be measured by long-term sampling. The table below summarises the findings.
The NSW EPA Policy falls below the BAT standard (BAT 4) which calls for monthly sampling with a sampling period of 2-4 weeks (known as ‘long-term sampling’). The BAT then recommends moving to short-term sampling only if emissions are proven to be sufficiently stable.
The NSW EPA decision is based on an independent expert review by authors from the University of Sydney, which claimed that the results from short-term and long-term sampling ‘do not differ significantly’. This is based on a flawed interpretation of a European study. The European study compared emission levels measured with short-term and long-term sampling, with data from 142 reference lines in Belgium and France. The study found that there is little difference between the two sampling methods when measuring low levels of PCDD/F emissions, but high PCDD/F emission levels were more likely to be measured by long-term sampling. The table below summarises the findings.
.png)
Long-term sampling is able to capture emissions data during different operating conditions. PCDD/F are more likely to be emitted at high levels during non-steady state operation, especially during cold start-up when combustion is incomplete and the temperature conditions are ideal for PCDD/F formation.
Periodic, short-term sampling does not capture this, because it is undertaken during steady-state operation in normal operating conditions, when PCDD/F emissions are lower. A study on an Energy from Waste Plant in the Netherlands found that significantly higher PCDD/F emissions were recorded using continuous, long-term sampling. Short-term sampling carried out over the same time period did not identify these.
The French data from the above-mentioned European study reported a correlation between elevated values obtained by long-term sampling measurement and the number of start-ups/shut-downs occurring in the time period during which the sampling was performed.
The independent expert reviewer justified their recommendation for twice per year, short-term testing in NSW by saying that it ‘is appropriate for emerging EfW industry in Australia, as it ensures strict guidelines to be complied with while allowing industry to install a simpler testing method’. This reasoning, and the erroneous interpretation of the available data, suggests that the independent expert review was written for the benefit of industry, not as an impartial analysis.
Periodic, short-term sampling does not capture this, because it is undertaken during steady-state operation in normal operating conditions, when PCDD/F emissions are lower. A study on an Energy from Waste Plant in the Netherlands found that significantly higher PCDD/F emissions were recorded using continuous, long-term sampling. Short-term sampling carried out over the same time period did not identify these.
The French data from the above-mentioned European study reported a correlation between elevated values obtained by long-term sampling measurement and the number of start-ups/shut-downs occurring in the time period during which the sampling was performed.
The independent expert reviewer justified their recommendation for twice per year, short-term testing in NSW by saying that it ‘is appropriate for emerging EfW industry in Australia, as it ensures strict guidelines to be complied with while allowing industry to install a simpler testing method’. This reasoning, and the erroneous interpretation of the available data, suggests that the independent expert review was written for the benefit of industry, not as an impartial analysis.
Emissions Standards
The best practice emissions limits for dioxins and furans are found in BAT 30, which gives different emissions limits for:
- Existing plants and new plants
- Short-term sampling and long-term sampling
- PCDD/F and PCDD/F + dioxin-like PCBs
The emission limit that is most relevant for comparison with the NSW EPA Policy limit the the limit for new plants, using short-term sampling, and only measuring PCDD/F. The upper end of this limit is 0.04 ng I-TEQ/Nm3. The NSW EPA Policy limit is 0.1 ng/m3.
The authors of the independent expert review erroneously listed the BAT emission limit as ‘0.01-0.1 ng I-TEQ/Nm3’. As shown in Table 5.7 of the BAT document, this is the emission limit for existing plants, using long-term sampling, measuring PCDD/F + dioxin-like PCBs. The review then concludes that the NSW EPA limit of 0.1 ng/m3 is equivalent to the most stringent, ‘best practice’ limit.
The authors of the independent expert review erroneously listed the BAT emission limit as ‘0.01-0.1 ng I-TEQ/Nm3’. As shown in Table 5.7 of the BAT document, this is the emission limit for existing plants, using long-term sampling, measuring PCDD/F + dioxin-like PCBs. The review then concludes that the NSW EPA limit of 0.1 ng/m3 is equivalent to the most stringent, ‘best practice’ limit.
Summary
(1) PCDD/F: Polychlorinated dibenzo-p-dioxins and -furans(2)After emissions have been proven to be sufficiently stable short-term monitoring still applies (once every six months, 6-8 hour sampling period)
References
Arkenbout et. al, 2018 Emission regimes of POPs of a Dutch incinerator: regulated, measured and hidden issues, Organohalogen Compounds, Vol. 80, 413-416
NSW Chief Scientist & Engineer, May 2020, Energy from Waste Report (with additional advice as at November 2020)
NSW EPA, 2021, NSW Energy from Waste Policy Statement
Neuwahl et al., 2019, Best Available Techniques (BAT) Reference Document for Waste Incineration, Industrial Emissions Directive 2010/75/EU (Integrated Pollution Prevention and Control), JRC Science for Policy Report
Sanjaya E & Abbas A, 2020, Energy-from-Waste: Independent review and expert advice, Waste Transformation Research Hub, School of Chemical and Bimolecular Engineering, University of Sydney