Chemosphere, Vol. 19, Nos. 1-6, pp 317-322, 1989
The production of halogenated aromatics has been investigated in the SAKAB hazardous waste incinerator, Norrtorp, Sweden. The measurement results show that there is a substantial net production of chlorinated aromatics in the boiler.
Halogenated aromatics, Production, Hazardous Waste; Boiler.
The production of halogenated aromatics in combustion processes is controlled by the combustion conditions and fuel composition (1-3). This does not necessarily mean that the final formation of halogenated aromatics takes place in the furnace. Studies recently performed both in a municipal waste incinerator and in the laboratory indicate that formation can also take place in the boiler (4-6).
In order to obtain a better understanding of where the formation takes place, the Environmental Consultants together with SAKAB have carried out this study in the SAKAB hazardous waste incinerator, Norrtorp. Sweden.
The SAKAB incinerator plant in Norrtorp is designed to handle about 33,000 tons of waste a year. The incinerator plant consists of an input system, a control unit, a rotary kiln, a post-combustion chamber, an exhaust gas boiler, a flue gas cleaning system and a chimney stack. The pumpable waste is fed in through the front end of the rotary kiln through a burner and "lances", and through input locks for drums and solid waste. The design capacity of the plant is corresponds to a heat input of 20 MW.
Sampling was carried out on three days with ordinary operation of the plant. The incinerator was operated with the same procedural order during the tests. The preset waste feed is presented in table 1.
Table 1. Waste feed to the incinerator.
| Waste type | Feed rate m3/h |
Heat value MJ/kg |
Cl % |
S % |
| Solvents | 0.8 | 24 | 2.9 | 0.5 |
| Sludge | 1.2 | 25 | 1.1 | 0.5 |
| Waste oil | 0.2 | 41 | 0.7 | 0.6 |
| Drums | 5-7 per hour |
The incineration was performed under very good conditions and the emissions were well below the permitted levels.
Some operational parameters during the tests are presented in table 2.
Table 2. Operational parameters.
| Test | 1 | 2 | 3 | |
| Produced steam | t/h | 19.6 | 19.7 | 20.1 |
| Gas temperature: | ||||
| Post-combustion chamber | °C | 958 | 953 | 1011 |
| Before boiler | °C | 764 | 762 | 774 |
| After boiler | °C | 277 | 282 | 281 |
| Gas before cleaning: | ||||
| CO2 | vol-% | 6.8 | 6.9 | 7.1 |
| CO | ppm | <5 | <5 | <5 |
| HCl | mg/nm3 dry gas | 2100 | 2400 | 2700 |
| Flue gas flow in the stack | nm3/h | 64400 | 64300 | 63500 |
Samples were collected simultaneously at four different places in the boiler.
The flow rates through the sample probes were adjusted so that the residence times of the sampled gas were equal in all four probes. The probe inlets were situated 1 m from the boiler wall inside the boiler. The probes were connected directly to large impinger bottles filled with water. The aim of such a sampling concept was to cool down the sampled gas and particles rapidly. The sampling train used for collecting the samples was an all glass system. After the rapid cooling in the impinger the gas was passed through a XAD-2 sorbent trap and a high purity tissue quartz thimble filter. The sampling time was approximately four hours and the gas volume in each sample varied between two and five m3 standard dry. Temperatures in all four sample probes were measured during the sampling. The gas volumes were measured with calibrated gas meters. All glassware used for sampling was carefully rinsed and then heated to overnight 450 °C. The quartz fiber filter was purified by heating to 530 °C. Amberlite XAD-2 was purified by successive washing with water, methanol and acetone. The resin was then packed in the sampling ampoule and soxhlet-extracted with acetone and dichloromethane for 48 h with intermediate changes of the solvent. These blank extracts were preconcentrated and examined using gas chromatography to check for impurities.
Is order to determine the recovery in the sampling and clean-up steps known amounts of the 13C-labeled PCDD/PCDF surrogates were added to the filter and XAD-2 sorbent before the sampling equipment left the laboratory.
The glassware containing the samples was transported to the laboratory and the extraction took place within a few days. The extraction and clean-up procedures have been described elsewhere (2-3).
All compounds were analyzed by gas chromatography and mass spectrometry with single ion monitoring, electron impact mode. At least two ions were monitored for each compound except for PAHs, for which only the molecular ions were examined. A 30 m 0.25 mm id, DB-5 column (J&W Scientific) was used as a general purpose column. Tetra and penta chlorinated PCDD/PCDF were separated on a 60 m, 0.25 id, SP-2331 column (Supelco).
Quantification was carried out by the internal standard method. 13C labeled internal standards of chlorobenzenes, chlorophenols and PCDD/PCDF were added to the samples before the extraction. The recoveries were calculated for the surrogates added to the samples and to compensate the corresponding isomers.
Samples from test two were analyzed in more detail than other samples. The analytical data available are presented below, figures 1 - 5.
The amount of chloride in the impinger bottles was determined for each sample.
During the entire test program the combustion conditions were excellent, with a combustion efficiency of or above 99.9 %. These optimized conditions and the resulting low levels of organic micro-pollutants can be assumed to weaken relations between process parameters and the production of organics due to sampling and analytical error. However this investigation shows the expected relation between chlorine in-put and the production of chlorinated aromatics. Furthermore the measurement results show that there is a substantial net production of chlorinated aromatics in the boiler, figures 1 and 2.
Figure 1. Sum chlorinated benzenes, µg/m3 sdg at 10% CO2, as a function of temperature, degrees centigrade, and HCl, mg/m3 sdg at 10% CO2.
Figure 2. TCDD-equivalents (Eadon), ng/m3 sdg at 10% CO2 , as a function of temperature degrees centigrade.
The production of micro-pollutants in the boiler is even more obvious when examining the brominated aromatics, figure 3. Here brominated toluenes are the dominating components at or above a concentration of 1 mg/m3 dry gas at 10 % CO2 in the flue gas.
Figure 3. Brominated toluenes and phenols, µg/m3 sdg at 10% CO2, as a function of temperature, degrees centigrade.
All groups of micro-pollutants do not show this behavior in the post-combustion zone and boiler, figures 4 and 5.
Figure 4. Sum chlorinated phenols, µg/m3 sdg at 10% CO2, as a function of temperature, degrees centigrade.
Figure 5. Naphtalene and sum PCB (Arochlor 1242), ng/m3 sdg 10% CO2, as a function of temperature, degrees centigrade.
We expected naphtalene to decline or not to be influenced by passage through the boiler. It is more difficult to explain the lack of co-variation in the boiler between different chloroaromatics.
The measurement results show that there is a substantial net production of halogenated aromatics in the boiler. Halogenated aromatics were present already in the boiler inlet. These results agree with earlier observations that a high combustion efficiency limits the production of all organic micro-pollutants. The test results emphasize the need to reduce the in-put of organics to the boiler by optimizing the combustion conditions in the kiln.
Efforts are now made to adapt the boiler design, with the aim of avoiding substantial formation of halogenated aromatics. The critical temperature interval must be more precisely defined and the effect of residence time in the boiler evaluated. This work will be carried out in collaboration with SAKAB.
We thank the staff at SAKAB for skilled technical assistance and enjoyable cooperation.
Reprinted from Chemosphere, Volume 19, Öberg, T., Warman, K., Bergström, J., Production of chlorinated aromatics in the post-combustion zone and boiler, Pages No. 317-322, Copyright (1989), with permission from Elsevier Science. Single copies of the article can be downloaded and printed for the reader's personal research and study.
|
Swedish homepage