Chemosphere, Vol. 19, Nos. 1-6, pp 711-716, 1989
The emission and production of chlorinated aromatics from metallurgical industries, e.g. scrap-metal re-melting, show substantial variations due to a number of process factors. The data evaluated indicates clearly that the production of chlorinated aromatics is combustion controlled.
Halogenated aromatics; Organic micro-pollutants; Metallurgical processes; Electric are furnaces.
Halogenated aromatics emitted from different thermal processes have aroused concern during the last decade. Until a few years ago the interest was focused solely on waste combustion. However, numerous investigations have now shown that many other industrial processes are also sources for chlorinated organic micro-pollutants, e.g. scrap-metal re-melting and pulp bleaching.
Here we report data from measurements in the Swedish metallurgical industry and process factors influencing the production of organic micro-pollutants.
The sampling and analytical methods have been described elsewhere (1 - 2). Process gas samples were collected up- and downstream of the flue gas cleaning system.
Measurements of organic micro-pollutants have been performed in nine Swedish steel plants re-melting scrap-metal in electric arc furnaces. In tables 1 and 2 we summarize the variation and annual emissions from these plants (3).
Table 1. Emissions to the air of organic micro-pollutants from electric are furnaces in nine different steel plants. Arithmetic mean, number of samples within brackets.
| Plant | PAH mg/ton* |
Chlorobenzenes mg/ton* |
Chlorophenols mg/ton* |
TCDD-eq (Edaon) µg/ton* |
| 1 | 23 (1) | 15 (1) | 11 (1) | 0.68 (1) |
| 2 | - | 35 (1) | 18 (1) | 7.1 (1) |
| 3 | 180 (3) | 33 (2) | 7 (2) | 4.5 (3) |
| 4 | 120 (2) | 25 (3) | 1.1 (3) | <3 (2) |
| 5 | - | 37 (1) | 16 (1) | - |
| 6 | 84 (2) | 16 (2) | 3.7 (2) | 9.0 (2) |
| 7 | 920 (1) | 3.2 (1) | 2.3 (1) | <0.23 (1) |
| 8 | 240 (2) | 27 (5) | 9.2 (1) | 8.6 (6) |
| 9 | 8.0 (2) | 9.8 (4) | 0.73 (4) | 1.2 (2) |
* Product
Table 2. Calculated annual emissions to the air.
| Plant | PAH kg/year |
Chlorobenzenes kg/year |
Chlorophenols kg/year |
TCDD-eq (Edaon) g/year |
| 1 | 4 | 3 | 2 | 0.1 |
| 2 | - | 7 | 3 | 1 |
| 3 | 70 | 10 | 3 | 2 |
| 4 | 10 | 3 | 0.2 | <0.3 |
| 6 | 30 | 6 | 1 | 3 |
| 7 | 100 | 0.4 | 0.3 | <0.03 |
| 8 | 100 | 10 | 4 | 3 |
| 9 | 0.2 | 0.2 | 0.02 | 0.03 |
The considerable variation in emissions between these steel plants is due to both process variation and different separation efficiencies in the flue gas cleaning systems. However, the separation in the fabric filters is in general high.
Here we will concentrate on examining process parameters studied in two metallurgical industries, A and B. A is a steel plant and B is a non- ferrous metallurgical industry. Both these plants re-melt metals contaminated with organics and chlorine, e.g. paint and cutting fluids.
The four tests carried out in plant A show that optimized oxidation/combustion and low emissions of organic micro-pollutants can be achieved through slow and continuous charging of the scrap-metal (oily cuttings), figure 1.
Figure 1. Sum chlorobenzenes, mg, as a function of charge rate, kg/h.
Each test covered a complete melting cycle. The deviation from a straight line-relation for one of the samples can be explained by a less efficient separation in the flue 5 as cleaning system (wet scrubber) and a resulting higher particle emission, 350 mg/m compared to 140 - 170 in the other tests.
In plant B eight tests were carried out with a factorial design. Samples from a melting furnace were collected from the raw gas before flue gas cleaning. The measurement results show a remarkable resemblance with those obtained from waste combustion (4-5). The two major sources of variation were identified, with principal component analysis, as the load of contaminated material (organics and chlorine) and the oxidation level (carbon monoxide). Multiple regression analysis (MLR) of the measurement data confirmed this observation. In figure 2 we show the calculated MLR-model as a response surface, r = 0.95 F(2.5) = 47.7 p = 0.0006.
Figure 2. Sum chlorobenzenes and chlorophenols, lg mg, as a function of CO, ppm dg, and in-put of contaminated scrap-metal, kg.
The chlorine load also has a marked influence on the isomer pattern of chlorinated aromatics (6 - 7). The low chlorine in-put in metallurgical processes favours the formation of low chlorinated aromatics as exemplified in table 3.
Table 3. Chlorination pattern of polychlorinated dibenzo-p-dioxins from different sources.
| Relative proportion % | Steel plant | Municipal waste combustion | High-chlorinated hazardous waste |
| OCDD | 20 | 36 | 90 |
| HpCDD | 27 | 41 | 9.2 |
| HxCDD | 25 | 14 | 0.3 |
| PeCDD | 13 | 7.6 | <0.1 |
| TCDD | 16 | 1.0 | <0.1 |
Organic chlorine-containing contaminants are not a prerequisite for the formation of chlorinated aromatics in metallurgical processes. Measurements in another plant, C, using mineral and fossil raw materials, will be used as an illustration. In this plant chlorine is supplied to the process as inorganic or bound in fossil raw materials. The emission of HCl and chlorinated aromatics is summarized in table 4.
Table 4. Emission of HCl and chlorinated aromatics, plant C.
| Components | Test 1 | Test 2 | |
| HCl | g/h | 220 | 230 |
| Chlorobenzenes | mg/h | 250 | 150 |
| TCDD-equivalents (Eadon) | µg/h | 13 | 14 |
The results presented indicate that the emissions from metallurgical industries vary considerably. The variation is due to process factors and flue gas cleaning. Important process factors are chlorine load and the oxidation level. Metallurgical processes free from organic contaminants also emit chlorinated aromatics. Efficient final oxidation of the process gas is likely to reduce emissions to a greater extent than melting less contaminated scrap-metals. A high oxidation level can be achieved with an effective afterburner function for the flue gas from the furnace.
A reduction of emissions to the air can also be achieved through optimization of fabric filter operation. We have indications that both increased removal of fine particles as well as feeding efficient sorbents can drastically improve the removal efficiency of organic micro-pollutants. More research is obviously necessary to fully utilize this potential.
The work presented has been funded by the Swedish Steel Producers Association (Jernkontoret) and various Swedish metallurgical industries.
Reprinted from Chemosphere, Volume 19, Öberg, T., Allhammar, G., Chlorinated aromatics from metallurgical industries - Process factors influencing production and emissions, Pages No. 711-716, 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.
|
English homepage