Chemosphere, Vol. 16, Nos. 10-12, pp 2451-2465, 1987

The production of brominated aromatics from combustion was shown to be influenced by the operating conditions. Brominated aromatics also showed high yields compared to their chlorinated analogues.

Chlorinated aromatics are well-known micro-pollutants from combustion processes involving chlorine. Recently brominated aromatics have also attracted attention. Reports have appeared about emissions of brominated and mixed halogenated aromatics in an accidental fire, and from the combustion of leaded gasoline, spill oil and hazardous waste (1 - 3). We have ourselves detected brominated aromatics in flue gases from hazardous waste, municipal waste and peat combustion (4 - 11).

Laboratory experiments have demonstrated the ability to form brominated aromatics from the combustion of propane and HCl contaminated with HBr (12). Polybrominated dioxins and dibenzofurans are formed during the pyrolysis of brominated flame retardants (13 - 17). These laboratory experiments and measurements of automotive emissions show surprisingly high yields of bromine containing aromatics (15, 18, 19).

Here we report results from the test operation in a Swedish hazardous waste incinerator (SAKAB). During these tests the input of bromine with the waste was changed drastically. We also summarize measurement data from previous combustion tests in different plants burning municipal solid waste (MSW) and peat.

Operation

The central industrial waste treatment and disposal facility in Sweden is located at Norrtorp and owned by the Swedish Waste Conversion Company (SAKAB). The incinerator at Norrtorp, designed by von Roll, has a capacity of 20 MW waste heat input. The rotary kiln is equipped with an afterburner chamber, and heat recovery in a steam boiler. The flue gas cleaning system consists of a Niro spray dryer absorber and a three-field electrostatic precipitator for particle removal.

A test program was preformed, within the scope of our continuous surveillance of combustion operation and emissions, to investigate production and emission of brominated and mixed halogenated aromatics.

In three tests the hazardous waste incinerator was fed with a normal input of chlorinated waste (approximately 80 kg of chlorine per hour), mainly solvents. Waste declared as brominated (56 % tetrabutylammonium bromide, 42 % water and 2 % tributylamine) was fed at two input levels, 16 and 35 kg of bromine per hour. Operation conditions were kept stable, with a high combustion efficiency during the testing periods. CO, CO₂, combustion efficiency, afterburner temperature, steam production and flue gas flow were measured continuously and collected with a data logger as 5-minutes arithmetical averages.

The measurement data from municipal waste incinerators reported here derive from both normal operation and starting procedures. The fuel is either municipal solid waste (MSW) or a mixture of MSW with wood chips.

Sampling and analysis

Chlorine and bromine input to the combustion process were measured indirectly by monitoring the HCl and HBr concentrations in the raw gas up-stream the flue gas cleaning system.

Sampling of HCl and HBr was carried out by absorption in aqueous impingers. The flue gases were sampled isokinetically by a sampling probe and passed through a heated filter followed by two impinger bottles containing 0.1 N sodium hydroxide solutions. During the analysis step the pH was adjusted to neutral and chlorides and bromides were co-precipitated with iodide and silver ions. The quantification of the filtrate was carried out with X-ray fluorescence spectrometry. The method of standard additions to the samples was used in order to overcome the matrix and improve the readability.

All-glass sampling equipment was used for collecting samples of organic micro-pollutants in the flue gases (20). The interchangeable probe extended into an oven module which was kept at 160 °C during sampling. The oven contained a high-purity glass fiber thimble filter. After leaving the filter, the flue gas was drawn through a high-efficiency cooler, a condensate collector, and a XAD-2 sorbent trap. At the conclusion of sampling the probe and the cooler were washed with acetone. The acetone wash residue was then treated as part of the sample. The flue gas volumes were between 5 and 10 m³ in each sample.

The sampling equipment fractionated the material into three parts: particles, condensate and matter adsorbed on the XAD-2 resin. All three parts were extracted separately and then all extracts were combined. Acidic components, e g phenols, were extracted with water at pH 12 and recovered by re-extraction at pH 1. The remaining neutral/ basic fraction was analyzed for halogenated aromatic hydrocarbons.

Halogenated benzenes and phenols were analyzed by high-resolution gas chromatography and electron capture detector (HRGC-ECD) in our investigations prior to 1987. As from 1987 these compounds are determined by high-resolution gas chromatography and Selected Ion Monitoring Mass Spectrometry (HRGC-MS/SIM).

Deuterium labeled 1,4-dichlorobenzene and ¹³C-labelled 1,2,4,5-tetrachlorobenzene, 2,4,5-trichlorophenol and pentachlorophenol were added to the samples before the extraction. The results were then compensated for losses in the extraction and clean-up procedure.

Halogenated dibenzo-p-dioxins and dibenzofurans were determined in the neutral/ basic fraction after HPLC clean-up on a Spherisorb A54, 25 cm, 4.6 mm column (purchased from Phase Sep). Fractions containing halogenated dioxins and dibensofurans were analyzed by HRGC-MS/SIM. At least two ions were monitored for each isomer using the electron impact mode.

A 60 m DB-5, 0.25 Wm film thickness, 0.25 mm ID (J&W) was utilized as a general purpose column. For the isomer specific separation of the tetrachlorinated dibenzo-p-dioxins and dibenzofurans a 60 m SP-2331, 0.25 µm film, 0.25 mm ID column was used.

Quantification was done by the external standard method. The recoveries were calculated for ¹³C- and ³⁷Cl-labelled PCDD/PCDF surrogates added to the samples prior to sampling (filter and XAD-2) and prior to extraction and clean-up. Amount of PCDD/PCDF were compensated for recoveries of the filter sampling spikes (¹³C-2,3,7,8-TCDD, ¹³C-1,2,3,7,8-PeCDD and ¹³C-OCDD) with the same number of chlorine atoms per molecule (hexa and hepta isomers were compensated for recoveries of PeCDD and OCDD). Reference compounds of halogenated dibenzo-p-dioxins and dibenzofurans were from Cambridge Isotope Laboratories (Woburn, USA) and Wellington Laboratories (Guelph, Canada).

Operating conditions - hazardous waste incinerator

The production of organic micro-pollutants from a combustion process is determined by the operating conditions (21 - 24). In table 1 we summarize results for the main operating parameters.

Table 1
Operating conditions

Test		1	2	3
Steam production	t/h	19.1	19.5	20.7
Flue gas flow	m3/h*	37900	40200	37900
CO2	vol-% dry gas	6.1	6.1	6.0
Combustion efficiency CE=(1-CO/CO2)*100	%	99.99	99.97	99.96
Afterburner temperature	°C	1030	1020	1000
Chlorides as HCl
Raw gas	mg/m3*	2250	1800	2300
Clean gas	mg/m3*	45	32	67
Separation efficiency	%	98	98	97
Bromides as HBr
Raw gas	mg/m3*	32	1100	530
Clean gas	mg/m3*	3	23	11
Separation efficiency	%	91	98	98
Sampling time	h	5	5	4

* standard dry gas at 10 % CO₂.

Halogenated aromatics from the incineration of hazardous waste

Due to the lack of published quantitative data we present a complete table for our measurements of halogenated aromatics in the raw gas (up-stream from the electrostatic precipitator), table 2.

Table 2
Flue gas concentrations (raw gas) of halogenated aromatics in ng/m³ standard dry gas at 10 % CO₂. Analyzed by HRGC-MS/SIM.

Test	1	2	3
Chlorobenzenes
1,3-	1700	120	700
1,4-	91	36	140
1,2-	770	120	480
1,3,5-	22	11	240
1,2,3-	1100	630	3100
1,2,4-	170	190	1800
1,2,3,5-	150	180	1200
1,2,4,5-	81	80	550
1,2,3,4-	240	420	1800
Penta-	790	1300	4800
Hexa-	850	1400	4600
Sum	5900	4400	19000
Monobromochlorobenzenes
1-Br-3-	38	50	94
1-Br-4-	22	44	50
1-Br-2	28	40	160
1-Br-3,5-	1.2	17	24
1-Br-2,5- / 1-Br-3,4-	9.3	65	220
1-Br-2,6-	16	110	340
Bromobenzenes
1,3-	1.7	84	31
1,4-	46	460	120
1,2-	32	250	89
1,3,5-	<0.2	6.3	1.4
1,2,4-	0.6	44	31
Sum Br3-			48
1,2,4,5-	0.2	19	7.9
Sum Br4-		97	48
Monobromotoluenes	1700	11000	130
Chlorophenols
2,4-	190	8.4	50
2,6-	8.7	1.7	7.2
2,4,6-	240	92	650
2,4,5-	6.9	5.5	41
2,3,6-	4.0	3.2	21
2,3,5,6-	5.9	3.8	<2
2,3,4,5-	22	6.5	50
2,3,4,6-	540	610	5300
Penta-	198	290	2400
Sum	1200	1000	8400
Monobromochlorophenols
2-Br-4-	67	21	29
4-Br-3-	26	6.7	1.1
4-Br-2,6-	20	42	110
Sum BrCl2-	180	170	620
Bromophenols
2-	36	110	16
3-/4-	24	230	31
2,4-	18	210	36
2,6-	<4	29	17
2,4,6-	<14	380	260
Chlorinated dibenzo-p-dioxins
2,3,7,8-	0.50	0.34
Sum Cl4-	11	4.6
1,2,3,7,8-	0.42	0.13
Sum Cl5-	4.0	3.6
1,2,3,4,7,8-	0.26	0.36
1,2,3,6,7,8-	0.55	0.44
1,2,3,7,8,9-	0.87	0.48
Sum Cl6-	6.1	3.8
Sum Cl7-	71	9.4
Cl8-	100	17
Brominated and mixed halogenated dibenzo-p-dioxins*
8-Br-2,3,7-Cl3-	<0.08	<0.08	<0.1
Sum BrCl3-	<	1.3	<
2,3-Br2-7,8-Cl2-	<0.4	<0.4	<0.5
Sum Br2Cl2-	<	<	<
2,3,7,8-Br4-	<0.2	0.2	<0.2
Sum Br4-	<	<	<
1,2,3,7,8-Br5-	<0.3	<0.3	<0.4
Sum Br5-	<	<	<
Chlorinated dibenzofurans
2,3,7,8-	4.0	2.1
Sum Cl4-	59	21
Sum Cl4-	7.9	4.2
1,2,3,7,8-	9.5	7.1
Sum Cl5-	100	61
1,2,3,4,7,8-	20	9.9
1,2,3,6,7,8-	7.5	8.0
1,2,3,7,8,9-	42	38
2,3,4,6,7,8-	50	40
Sum Cl6-	260	190
Sum Cl7-	460	73
Cl8-	100	29
Monobromotrichlorodibenzofurans*
8-Br-2,3,4-Cl3-	<0.05	0.086	<0.06
Sum BrCl3-	<	1.3	<

* Results are not compensated for recoveries from extraction and clean-up.

Monobromotrichlorodibenzo-p-dioxins and dibenzofurans could only be detected in the raw gas sample from test 2. However samples taken down-stream from the flue gas cleaning system show detectable quantities of monobromotrichlorodibenzofurans in all samples, table 3. The increase in flue gas concentrations of monobromotrichlorodibenzofurans cannot be interpreted further at this stage. The separation efficiency for tetrachlorinated dioxins and dibenzofurans varied between 50 - 90 %.

Table 3 Monobromotrichlorodibenzo-p-dioxins and dibenzofurans in clean gas, ng/m³ standard dry gas at 10 % CO₂. Analyzed by HRGC-MS/SIM.

Test	1	2	3
Sum BrCl3DD	nd	0.068	0.15
8-Br-2,3,4-CDF	<0.06	0.26	0.39
Sum BrCl3DF	0.20	2.4	4.5

A close correlation has previously been demonstrated between chlorinated aromatics and the in-put of chlorine to the combustion process (13, 15). This is also the case for brominated aromatics, figure 1 - 3.

Figure 1
Brominated phenols as a function of HBr, in raw gas. o = 2,4,6-tribromophenol, + = 2,4-dibromopheriol and * = 2-bromophenol.

 

Figure 2
Brominated benzenes as a function of HBr, raw gas. * = 1,2-dibromobenzene, + = 1,2,4-tribromobenzene and o = 1,2,4,5-tetrabromobenzene.

 

Figure 3
The ratio between monobromotrichloro- and tetrachlordibenzofuran, clean gas, as a function of HBr (molar percent of total hydrogen halides), raw gas.

HBr has a low thermal stability compared to HCl. This stability difference will result in comparatively high concentrations of both bromine molecules and radicals, and possibly brominated aromatics. The measurement results for halogenated aromatics reported here (table 2), show a significantly higher yield of brominated compounds comparing the input of bromine with chlorine. This increased reactivity of bromine is especially pronounced for halogenated phenols, table 4 and figure 4.

Table 4
Distribution of dihalogenated phenols and hydrogenhalide in the raw gas, molar percent

Test	1	2	3
HCl	99	82	92
HBr	0.63	18	8.4
Sum dichlorophenols	70	5	47
Sum monobromomonochlorophenols	26	12	28
Sum dibromophenols	4	83	95

Figure 4
Fraction of dihalogenated phenols (molar percent) as a function of HBr (molar percent of total hydrogen halides). * = dichlorophenols and + = dibromophenols.

Bromine and brominated aromatics from the combustion of municipal solid waste (MSW) and peat

We have reported elsewhere the detection of brominated aromatics from MSW incineration and peat combustion (6 - 11). Here we wish to bring some of these data together and also add information concerning bromine in the fuel.

In table 5 we present measurement data for di- and trihalogenated phenols from three different MSW-incinerators during normal operation and start-up procedures.

Table 5
Di- and trihalogenated phenols from MSW incineration, ng/m³ standard dry gas at 10 % CO₂. Analyzed by HRGC-ECD.

Plant	A	B	C:1	C:2
Operation	Start	Normal	Normal / disturbed	Normal
Fuel	MSW	MSW	MSW	MSW / Wood waste
2,4-Dichlorophenol	15000	140	1300	440
2,4-Dibromophenol	2600	58	12	16
2,4,6-Trichlorophenol	55000	6600	3300	1800
2,4,6-Tribromophenol	340	na	4	5

These brominated phenols are found at lower levels than those substituted with chlorine in the same positions. However, the relative concentrations of brominated compared with chlorinated phenols indicate either unexpectedly high levels of bromine in MSW, or a higher yield from the same amount of halogen (assuming binomial distribution between possible isomers).

Literature data concerning the amount of bromine present in MSW seems to be scarce. We have therefore recently measured both HCl and HBr in the flue gas from a MSW incinerator, with the purpose of establishing a basis for comparison, table 6.

Table 6
HCl and HBr in the flue gas from a MSW incinerator, mg/m³ standard dry gas at 10 % CO₂.

Test	1	2	3	4
Fuel	MSW	MSW	MSW	MSW / Wood chips
HCl	720	730	660	450
HBr	25	25	22	17

The average concentration of bromine in these measurements was 3.5 % (w/w) relative to chlorine, or twice the relative concentration in the "normal" mixture of hazardous waste (see test in table 1).

In table 7 we present measurement data for di- and trihalogenated phenols from peat combustion.

Table 7
Di- and trihalogenated phenols from peat combustion, ng/m³ standard dry gas. Analyzed by HRGC-ECD.

Plant	D	E	F
2,4-Dichlorophenol	<5	3	<1
2,4-Dibromophenol	290	60	25
2,4,6-Trichlorophenol	1100	40	<1
2,4,6-Tribromophenol	<5	60	14

The average concentration of bromine in peat was 26 ppm (wet weight) in eight samples from Sweden and Finland, or 9.6 % (w/w) of the chlorine concentration (25).

Assuming binomial distribution between possible isomers both MSW-incineration and peat combustion result in higher yield, in relation to halogen-input, of these brominated phenols compared with chlorinated phenols.

Brominated and mixed halogenated aromatics are of concern because of the potential health effects. The relatively high yield of brominated aromatics from combustion accentuates this problem. The operating conditions have a marked influence on the production of halogenated aromatics. Emissions of brominated and mixed halogenated aromatics, and limiting factors for the production, should therefore be evaluated carefully.

5 000 halogenated dioxins and dibenzofurans containing chlorine and/or bromine, not to mention other compound groups, pose an enormous analytical problem. We suggest therefore the use of halogenated benzenes and phenols as indicator parameters. This approach has, been applied with promising results for chlorinated dioxins and dibenzofurans (26).

1. Buser, H-R
   Anal Chem 58, 2913-2919 (1986).
2. Oehme M, Benestad Ch
   Presented at ISWA/DAKOFA/WHO Specialized Seminar: Emission of Trace Organics from Municipal Solid Waste Incinerators, Copenhagen 20 - 22 January 1987.
3. Schäfer W, Ballschmiter K
   Chemosphere 15, 755-763 (1986).
4. Andersson C
   Technical report EBM-83/104 Studsvik Energiteknik AB 1983. In Swedish.
5. Björner B
   Technical report MKS-86/75.
6. Andersson C Report STUDSVIK/EK-81/140 Studsvik Energiteknik AB, 1981 In Swedish
7. Larsson G
   Technical report MKS-85/105 Miljökonsulterna i Studsvik AB, 1985. In Swedish.
8. Öhrström T
   Technical report MKS-86/10 Miljökonsulterna i Studsvik AB, 1986. In Swedish.
9. Warman K, Hydén E
   Technical report MKS-85/2 Miljökonsulterna i Studsvik AB, 1985. In Swedish.
10. Warman K, Hydén E
    Technical report MKS-85/3 Miljökonsulterna i Studsvik AB, 1985. In Swedish.
11. Warman K, Hydén E
    Technical report MKS-85/4 Miljökonsulterna i Studsvik AB, 1985. In Swedish.
12. Eklund G, Pedersen J, Strömberg B
    Chemosphere 16, 161-166 (1987).
13. Tiernan T O, Ramalingam B, Garret J H, Mazer T, Wagel D J, Malloy C M, Taylor M
    Presented at 5th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth (FRG), 16 - 19 September 1985.
14. Buser H-R
    Environ Sci Technol 20, 404-408 (1986).
15. Thoma H, Rist S, Hauschulz G, Hutzinger O
    Chemosphere 15, 649-652 (1986).
16. Thoma H, Hauschulz G, Knorr E, Hutzinger O
    Chemosphere 16, 277-285 (1987).
17. Thoma H, Hauschulz G, Hutzinger O
    Chemosphere 16, 297-307 (1987).
18. Eklund G
    Personal communication.
19. Buser H-R
    Chemosphere 16, 713-732 (1987).
20. Bergström J G T, Eklund G, Warman K
    Presented at the Sixth International Symposium on Polynuclear Aromatic Hydrocarbons, Columbus, Ohio (USA), 27 - 29 October 1981.
21. Bergström J G T, Lundqvist J
    Presented at the Symposium of the International Solid Waste and Public Cleansing Association - ISWA, Munich, 21 - 25 May 1984.
22. Öberg T, Aittola J-P, Bergström J G T
    Chemosphere 14, 215-221 (1985).
23. Öberg T, Bergström J G T
    Chemosphere 15, 2041-2044 (1986).
24. Bergström J G T, Warman K
    Presented at ISWA/DAKOFA/WHO Specialized Seminar: Emission of Trace Organics from Municipal Solid Waste Incinerators, Copenhagen, 20 - 22 Januari 1987.
25. Blomqvist G, Fredriksson D, Thun R, Kallio M
    Report STUDSVIK/EP-86/38 Studsvik Energiteknik AB, 1986. In Swedish.
26. Öberg T, Bergström J G T
    Chemosphere, in press.

(Received in Germany 6 June 1987; accepted 19 August 1987).

Reprinted from Chemosphere, Volume 16, Öberg, T., Bergström, J., Brominated aromatics from combustion, Pages No. 2451-2465, Copyright (1987), 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(87)90304-3