Chemosphere, Vol. 14, No. 2, pp 215-221, 1985

The production and emission of chlorinated aromatics from a hazardous waste incinerator were shown to be influenced by the operating conditions.

Emissions of chlorinated aromatics from different combustion processes have aroused considerable interest in recent years. Several attempts have been made to correlate combustion conditions or fuel parameters to emissions of chlorinated aromatics (1, 2). Despite these efforts it has been difficult to arrive at consistent results in full-scale plants.

Here we report results from a study on a Swedish hazardous waste incinerator.

Several combustion and fuel parameters were varied systematically during the tests. However the variation of these parameters did not take the incinerator out of normal operating conditions (combustion efficiency between 99.89 and 99.98 %).

Flue gas samples were collected up and downstream of the flue gas cleaning system (wet-dry scrubber and electrostatic precipitators). The flue gas samples were collected with an all glass sampling train (3). Chlorinated benzenes, phenols, dibenzodioxins (PCDD) and dibenzofurans (PCDF) were quantified relative to standard mixtures. The chemical analyses of PCDD and PCDF were performed at the university of Umeå, Sweden.

During the entire test programme emissions were monitored closely. In Table 1 the emissions of PCDD and PCDF are presented as weekly samples from this period.

Table 1
Emissions of PCDD and PCDF, 1984-05-08--06-19. All values expressed as µg/Nm³ dry gas at 10% CO2*.

Isomeric group	Mean	n	Standard deviation
Sum PCDD	14	5	16
Sum Tetra-CDD	2	5	1.3
Sum Penta-CDD	8	5	12
Sum PCDF	50	4	37
Sum Tetra-CDF	8	5	8.5
Sum Penta-CDF	18	5	17

These emissions are not anticipated to impose any significant adverse influence on public health (4).

Chlorine in the fuel occurs mainly as solvents (80 - 95 %). The concentrations of PCB and chlorophenols were below 25 and 100 ppm respectively during the test runs.

Several emission parameters were found to correlate closely with the total chlorine input to the kiln. Chlorinated benzenes, PCDD and PCDF show statistically significant correlations to chlorine input, r = 0.88, n = 8, r = 0.62, n = 11 and r = 0.64, n = 11 respectively, p <0.01 and p <0.05. In Figures 1, 2 and 3 mass flows of these substance groups are presented as a function of chlorine flow to the kiln.

Figure 1

 

Figure 2

 

Figure 3

A correlation can also be seen between PCDD and PCDF, r = 0.74, n = 11, p <0.01. A very strong correlation has been found between pentachlorobenzene and pentachlorophenol, r = 0.99, n = 8, p <0.01, see Figure 4.

Figure 4
Pentachlorobenzene as a function of pentachlorophenol, µg/Nm³ dry gas at 10 % CO₂.

Furthermore we can see an influence of the chlorine input on the degree of chlorination of chlorobenzenes, as shown in Figure 5.

Figure 5
* = Hexachlorobenzene, o = Pentachlorobenzene, + = Tetrachlorobenzenes

A high chlorine input favours the formation of more highly chlorinated benzenes.

The results presented here suggest a strong connection between the level of chlorine input and the production of chlorinated aromatics. We can also see connections between the different substance groups. The results presented correspond well with a general formation mechanism of chlorophenols via chlorobenzenes as indicated by others (5):

Organic material + chlorine -> chlorobenzenes -> chlorophenols

Separation of PCDD and PCDF in the flue gas cleaning system was also investigated. Results are reported in Table 2.

Table 2
Separation efficiency (%) of PCDD and PCDF in the flue gas cleaning system.

Isomeric group	Mean	n	Standard deviation
Sum tetra-CDF	65	11	27
Sum-tetra-CDD	48	7	41
Sum penta-CDF	64	11	23
Sum penta-CDF	51	7	29
Sum hexa-CDD	82	11	14
Sum hexa-CDF	73	7	12
Sum hepta-CDD	83	11	15
Sum hepta-CDF	83	11	24
OCDF	85	10	18
OCDD	89	10	22

The separation efficiency in the flue gas cleaning system was shown to correlate to some extent with the load, expressed as steam production from the boiler (t/h). The variations in load influence both the flue gas temperature after the boiler, as well as the flue gas flow rate and retention times in the system. In Figures 6, 7 and 8 the correlations are shown for tetra-CDD, penta-CDD and hepta-CDF.

Figure 6

 

Figure 7

 

Figure 8

The separation efficiency is often dependent on the concentration of the compound to be separated. Better separation efficiencies are achieved at higher concentrations. Such a correlation can also be seen in our data, with correlation coefficients between 0.4 to 0.7 for the different isomers of PCDD and PCDF. These results therefore suggest that an increase in the production of PCDD and PCDF must not necessarily lead to the same increase in emissions.

Our conclusions are that:

• at normal operating conditions with a high combustion efficiency (99.9 %) the production of chloroaromatics in this incinerator can be influenced by varying the total chlorine input.
• high separation efficiencies of PCDD and PCDF can be achieved in this plant at a low load to the kiln.

1. BENFENATI E, GIZZI F, REGINATO R, FANELLI R, LODI M, TAGLIAFERRI R R
   Chemosphere 12, 1151-1157 (1983).
2. GONNARD M F, GUIOCHON G, VIAU A C, KARASEK F W, FINET C
   Presented at the Symposium of the International Solid Wastes and Public Cleansing Association, 21 - 25 May 1984 in Munich.
3. BERGSTRÖM JGT, EKLUND G, TRZCINSKI K
   Presented at the Sixth International Symposium on Polynuclear Aromatic Hydrocarbons, 27 - 29 October 1981 in Columbus, Ohio, USA.
4. ÖBERG T
   MKS Teknisk Rapport MKS-84/38 Environmental Consultants at Studsvik AB September 1984.
5. CHOUDHRY G G, HUTZINGER O
   Toxicol Environ Chem 5, 295-309 (1982).

(Received in The Netherlands 5 January 1985; accepted 10 January 1985)

Reprinted from Chemosphere, Volume 14, Öberg, T., Aittola, J-P., Bergström, J., Chlorinated aromatics from the combustion of hazardous waste, Pages No. 215-221, Copyright (1985), with permission from Elsevier Science. Single copies of the article can be downloaded and printed for the reader's personal research and study.

DOI: 10.1016/0045-6535(85)90100-6