 |
Toolkit for Identification and Quantification of Releases of Dioxins, Furans and Other Unintentional POPs PART III Annexes |
Annex 20 Complementary information to source category 2e Aluminum Production
Overview of recent revisions
New data confirmed the majority of PCDD/PCDF emission factors for this source category, with slightly higher values for residues (class 3). New emission factors are proposed for HCB and PCBs.
Table III.20.1 PCB emission factors for source category 2e Aluminum Production
2e Aluminum Production Emission Factors (µg TEQ/t aluminum) Classification Air Water Land Product Residue 1 Thermal processing of scrap Al, minimal treatment of inputs and simple dust removal 40 2 Thermal Al processing, scrap pre-treatment, well-controlled, fabric filters with lime injection 0.1 20 3 Optimized for PCDD/PCDF control – afterburners, lime injection, fabric filters and active carbon 0.02 4 Shavings/turning drying (simple plants) 5 Thermal de-oiling of turnings, rotary furnaces, afterburners, and fabric filters 6 Primary Al production Table III.20.2 HCB emission factors for source category 2e Aluminum Production
2e Aluminum Production Emission Factors (µg/t aluminum) Classification Air Water Land Product Residue 1 Thermal processing of scrap Al, minimal treatment of inputs and simple dust removal 500 2 Thermal Al processing, scrap pre-treatment, well-controlled, fabric filters with lime injection 500 3 Optimized for PCDD/PCDF control – afterburners, lime injection, fabric filters and active carbon 500 4 Shavings/turning drying (simple plants) 5 Thermal de-oiling of turnings, rotary furnaces, afterburners, and fabric filters 6 Primary Al production Emission factors for PCB/HCB are provided with:
- A high level of confidence for class 2 air emissions (PCB), as emission factors are derived from a broad geographical coverage and are based on a low data range and not on expert judgment;
- A medium level of confidence for class 2 air emissions (HCB), as emission factors are not based on expert judgment but are not derived from a broad geographical coverage;
- A low level of confidence for all other classes (PCB/HCB), as emission factors are based on extrapolations and expert judgment.
Derivation of emission factors
Release to Air
Several steps in the processing of aluminum scrap can lead to the release of PCDD/PCDF to air. Thermal pretreatment of input materials, scrap melting and metal refining using chlorine or hexachloroethane (as a degasifying agent) can all lead to releases of PCDD/PCDF to air.
Emissions to air vary greatly depending on the nature of the scrap, pre-cleaning of the feed and the type of furnace and gas cleaning system applied. Older technology furnaces fitted with fabric filters had emissions of 146 to 233 μg TEQ/t of product. Concentrations and volumes of flue gas vary considerably; concentrations up to 10 ng I-TEQ/m³ were reported (SCEP 1994). Drum furnaces using aluminum turnings seemed to produce high emissions. For systems using contaminated scrap (such as scrap with cutting oils, plastics) with simple controls and gas cleaning consisting of cyclones or basic fabric filters an emission factor of 100 μg TEQ/t of product should be used.
The class 2 emission factor of 3.5 μg TEQ/t is taken from recent measurements at two European plants and are for well-controlled modern plants with scrap treatment, fabric filters, and lime injection. The Italian study gave 5.2 μg TEQ/t of Al. Class 2 emission factor was confirmed by a number of publications released between 2005 and 2010. Emission factors of 0.3 to 8.6 µg TEQ/t were derived in Poland (Grochowalski et al. 2007), 2.9 µg TEQ/t in New Zealand (Graham and Bingham 2010), 7-8 µg TEQ/t in Japan (Iwata et al. 2008), 1.24 µg TEQ/t in South Korea (Yu et al. 2006), 2.65 µg TEQ/t in China (Ba et al. 2009).
Class 3 emission factor should be applied for plants equipped with dioxin reducing technology, especially optimized flue gas cleaning systems. Class 3 emission factors are confirmed by measurements carried out on industrial plants in New Zealand (Merz 2004) and in Italy (Pitea et al. 2008). In New Zealand, emission factors ranging from 0.0027 to 0.5 µg TEQ/t were calculated. In Italy, an emission factor of 0.35 µg TEQ/t was derived from various measurements on a plant optimized for PCDD/PCDF controls. In Taiwan, PCDD/PCDF emissions from secondary smelters fed with aluminum ingots and very clean scrap were in the range of 0.025-0.441 µg TEQ/t, bringing additional confirmation for class 4 emission factor.
Classes 4 and 5 emission factors address plants for de-oiling and drying of Al turnings: class 4 emission factor of 5 μg TEQ/t applies to the drying of Al shavings and turnings in rotary drums or similar equipment and class 5 emission factor applied to thermal de-oiling of turnings in rotary kilns with afterburners and fabric filters (ENEA 2003).
Regarding PCB emissions, from measurements carried out in Poland (Grochowalski et al. 2007), in China (Ba et al. 2009) and in South Korea (Yu et al. 2006) on plants assigned to class 2 emission factors ranging from 0.04 to 0.81 µg TEQ/t were derived. Thus, an emission factor of 0.1 µg TEQ/t is proposed for class 2. Emission factors for class 1 are based on Japanese data (Takeuchi et al. 2009) and Spain (Sanz et al. 2010) where respective concentrations of 7 and 3 ng TEQ/Nm³ are reported, which correspond to an average emission factor of 40 µg TEQ/t. Class 3 PCB emission factor is estimated from class 2 PCB emission factor and class 3 and class 2 PCDD/PCDF emission factors.
Regarding HCB emissions, data reported in the literature and derived from Japanese plants assigned to classes 1, 2 and 3 were very similar. Class 3 emission factors were in the range of 23-7,200 µg/t (Iwata et al. 2008). Class 2 emission factors assessed were in the range of 23-3,600 µg/t (Ota 2005). Class 1 emission factors assessed were in the range 450-1,300 µg/t (Takeuchi et al. 2009). Therefore, the same emission factor is proposed for classes 1 to 4 (500 µg/t).
Release to Water
Releases to water may result where wet scrubbers or other processes have liquid effluents. There is insufficient information to estimate emission factors. Any liquid effluents should be noted and their source recorded.
Release to Land
No release to land is expected.
Release in Products
No releases to with the products are expected.
Release in Residues
Residues from the process are expected to contain PCDD/PCDF. The highest contamination is expected to be associated in dust and sludge from flue gas treatment. The amounts of such dust and sludge should be recorded and any use in other processes may lead to transfer of PCDD/PCDF. Melting in rotary drum furnaces generates 300-500 kg salt slag per ton of Al and 10-35 kg filter dust/t Al. Dross generated at ca. 25 kg/t Al can be reused in rotary drum furnaces (UBAVIE 2000).
Concentrations of PCDD/PCDF in filter dusts have been recorded from 3 to 18,000 ng TEQ/kg (SCEP 1994, Bremmer et al. 1994). Filter dusts are produced at a rate of approximately 8% of the metal production (Dyke et al. 1997). Combined with an average concentration of 5,000 ng TEQ/kg, this gives an emission factor of 400 μg TEQ/t of product for class 2 (fine particulates). Measurement data obtained both in the European Union (BREF 2009) and in China (Ba et al. 2009) from plants assigned to class 3 show concentrations of 4-5 ng TEQ/g of waste. Assuming the abovementioned 8% ratio, such concentrations correspond to an emission factor of 350-400 µg TEQ/t. Therefore, an emission factor of 400 µg TEQ/t is proposed for class 2.
The Italian study gave 183 μg TEQ/t and thus, the emission factor of class 1 has been changed to 200 μg TEQ/t of Al (ENEA 2003). Class 1 emission factor was confirmed. For class 3, optimized PCDD/PCDF control is implemented (including clean scrap), the lower factor of 100 μg TEQ/t should be applied to make initial estimates.
PCB concentrations measured at a Chinese plant related to class 3 were 0.4 ng TEQ/g of dust on average (Ba et al. 2009), which would correspond to an emission factor of 20 µg TEQ/t.
