Fs in the Raw and Clean Flue Gas during

ACS2GO © 2018. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to homescreen...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/est

PBDDs/Fs and PCDDs/Fs in the Raw and Clean Flue Gas during Steady State and Transient Operation of a Municipal Waste Combustor Barbara Wyrzykowska-Ceradini,†,§ Brian K. Gullett,*,† Dennis Tabor,† and Abderrahmane Touati‡ †

Office of Research and Development, National Risk Management Research Laboratory, U.S. Environmental Protection Agency, E343-04, Research Triangle Park, North Carolina 27711, United States ‡ ARCADIS U.S. Inc., 4915 Prospectus Drive, Durham, North Carolina 27711, United States § Oak Ridge Institute for Science and Education Research Postdoctoral Program, P.O. Box 117, Oak Ridge, Tennessee 37831, United States

bS Supporting Information ABSTRACT: Concentrations of polybrominated dibenzo-p-dioxins, and -dibenzofurans (PBDDs/Fs) and polychlorinated dibenzo-p-dioxins, and -dibenzofurans (PCDDs/Fs), were determined in the pre- and post-air pollution control system (APCS) flue gas of a municipal waste combustor (MWC). Operational transients of the combustor were found to considerably increase levels of PBDDs/Fs and PCDDs/Fs compared to steady state operation, both for the raw and clean flue gas; ΣPBDDs/Fs increased from 72.7 to 700 pg dscm1 in the raw, pre-APCS gas and from 1.45 to 9.53 pg dscm1 in the post-APCS flue gas; ΣPCDDs/Fs increased from 240 to 960 ng dscm1 in the pre-APCS flue gas, and from 1.52 to 16.0 ng dscm1 in the post-APCS flue gas. The homologue profile of PBDDs/Fs and PCDDs/Fs in the raw flue gas (steady state and transients) was dominated by hexa- and octa-isomers, while the clean flue gas homologue profile was enriched with tetra- and penta-isomers. The efficiency of the APCS for PBDD/F and PCDD/F removal was estimated as 98.5% and 98.7%, respectively. The cumulative TEQPCDD/FþPBDD/F from the stack was dominated by PCDD/F: the TEQ of PBDD/F contributed less than 0.1% to total cumulative toxic equivalency of MWC stack emissions.

’ INTRODUCTION As a result of the regulatory activities and improved air pollution controls atmospheric emissions of PCDDs/Fs from municipal waste combustion in the U.S. have been reduced by approximately 99% between 1987 and 2000 (from 8905 g TEQ/ year to 83.8 g TEQ/year, respectively).1 Even though the PCDDs/Fs emissions from municipal waste combustors (MWCs) are regulated in the U.S., their brominated counterparts  polybrominated dibenzo-p-dioxins and -dibenzofurans (PBDDs/Fs), are rarely reported. This despite several recent reports that PBDDs/Fs may exhibit toxicity similar to PCDDs/Fs.2,3 The 2,3,7,8-substituted isomers of PBDD and PBDF have been already found in several environmental matrices, including air samples.4,5 They have been also found in human milk and adipose tissue at levels that can contribute significantly to the total cumulative toxic equivalency (TEQ).68 Data on PBDD/F emissions from concentrated industrial sources are scarce. The limited data on parallel measurements of chlorinated- and brominated dioxins and furans from full-scale municipal waste combustion showed that levels of PBDDs/Fs in effluent gases from MWCs appear to be very low, sometimes less than 0.1% of the level reported for PCDDs/Fs.911 The most extensive study on PBDD/F emissions from waste incinerators r 2011 American Chemical Society

was conducted by the Japanese Ministry of Welfare in 2000.11 Results from 75 incineration plants tested showed that in addition to PCDDs/Fs (which were found in all samples) bromo- and mixed bromochloro-dioxins and furans were also found in effluent gases, bottom and fly ashes, and liquid effluent from MWCs. PBDDs/Fs were found in about half of the analyzed samples (PBDFs in higher concentrations than PBDDs) at about 0.5% of the level reported for chlorinated dioxins (e.g., in the flue gas PBDDs/Fs ranged n.d.-270 ng Nm3, average level of 4 ng Nm3; and PCDDs/Fs were at 0.04815,000 ng Nm3, average of 770 ng Nm3).11 Wang et al.9 reported PBDDs/Fs in the stack gases of large-scale MSWIs in Taiwan at average concentrations of 2.28 pg Nm3 (0.455.46 pg Nm3), accompanied with PCDDs/Fs at average concentration of 3.75 ng Nm3 (0.34 20.4 ng Nm3).9 The PBDD/F and PCDD/F mean stack emission factors reported for two Taiwanese MSWIs in the subsequent study of Wang et al.10 were 14.4 ( 5.67 μg per tonwaste for PCDD/Fs and 0.690 ( 0.787 μg per ton-waste for Received: January 31, 2011 Accepted: May 26, 2011 Revised: May 24, 2011 Published: June 09, 2011 5853

dx.doi.org/10.1021/es200364u | Environ. Sci. Technol. 2011, 45, 5853–5860

Environmental Science & Technology PBDD/Fs.10 The lower levels of PBDDs/Fs, when compared to PCDDs/Fs are usually associated with different formation yields due to the content and origin of halogenated precursors, especially chloro- and bromophenols in the fuel, while de novo formation pathways are suggested to be similar.12 Numerous studies suggest that the PBDD/F formation rate and congener homologue depend on combustion parameters and on ratio and halogens content (Br:Cl) in the fuel.1318 Experiments of S€oderstr€om and Marklund (2002) on pilot-scale fluidized bed incineration of fuels that contain both chlorine and bromine showed the formation of a complex mixture of polybrominated-, polybromochlorinated-, and polychlorinated dibenzo-p-dioxins and dibenzofurans, with the mixed bromo/chlorodioxins and dibenzofurans being the most abundant congeners, and a minor decrease in the levels of PCDD and PCDFs observed when bromine was added to the fuel.14 A similar trend was previously described by Funcke et al. (1997) in the experiments at the pilot MWC 'TAMARA’.16 This study reported that in the reference samples (chlorine-containing input materials) mostly PCDDs/Fs were detected, while in the samples from the cocombustion trials (with chlorine and bromine-containing plastic input materials) up to 45% of the total PXDFs/Ds consisted of PBCDFs/Ds. The percentage of bromo/chloro-congeners correlated well with bromine content of the total halogen input.16 Multiple brominated flame retardants  polybrominated diphenylether (PBDE), tetrabromobisphenol A (TBBP-A), hexabromocyclododecane (HBCD)  have been suggested as precursors of PBDDs/Fs.1218 PBDEs especially are suggested to be potent PBDDs/Fs precursors.12 It is due to the fact that the structure of PBDF is actually contained within PBDE molecule and the reaction to form PBDF is only a ring closure under cleavage of Br2 or HBr.12 In addition, the quantity of PBDDs/Fs formed from combustion of BFRs might be regulated by the presence of water and metals (e.g., antimony(III) oxide Sb2O3, which is often added to BFRs due to its synergistic effect for the fire inhibition).12 In pilot-scale incineration of flame-retarded waste concentrations of PBDDs/Fs ranged from nondetect to 950 ng Nm3 in the flue gas collected at the exit of the secondary combustion chamber and from 47 to 990 ng Nm3 at the exit of the cooling duct.13 In the same study concentrations of PCDDs/ Fs were as follows: 1.17100 ng Nm3 (exit of the secondary combustion chamber) and 2.4230 ng Nm3 (exit of the cooling duct).13 On the other hand, experiments of Vehlow et al. (2000) suggested that co-combustion in MWC might be a beneficial disposal route for limited amounts of electrical and electronic waste plastics.17 This study reported that addition of e-waste increased the burnout of the bottom ashes due to improved heating value of the fuel and did not deteriorate the PCDD/F emission quality.17 The high bromine (Br) loads (up to 12%) caused only limited increase of bromine containing dioxins and furans in the raw gas, with formation of bromine containing furans exceeding that of dioxins.17 The prevailing homologues were those containing one Br atom with virtually no purely brominated homologues detected.17 PBDFs were also found to be impurities of technical formulations of PBDEs, with the concentrations of total tetra- to octaBDFs reported from 260 ng to 50 μg per gram of commercial product.19 This is important information to consider while interpreting results of the present study, as it suggests that fuel of MWCs might contain PBDFs impurities from the brominated flame retardants abundantly used in everyday products. Hence, thermal desorption from waste material might be a potential source of PBDFs in the combustion flue gases.

ARTICLE

’ METHODS Sampling. Field sampling was performed in December 2006 at a U.S. municipal waste combustor (MWC) firing refuse derived fuel (RDF) in a grate furnace technology. The MWC system schematics and operational characteristics as well as the RDF production overview are given in the Appendix of the Supporting Information. Sampling was performed during various stages of normal (“steady state”) and transient (“shutdown” and “startup”) boiler operation. The shutdowns and startups of the boiler were scheduled at the request of the sampling team. Samples of raw flue gases were taken isokinetically according to a modified version of U.S. EPA Method 23 (M-23) prior to the APCS using “short” (5 min) and “long” (>1 h) sampling durations. Samples of the stack flue gases were collected in time regimens corresponding to the total “long” sampling length of the raw flue gases (stack emissions during monitored shutdowns and following startups were sampled jointly and are together referred as post-APCS transient samples). The concentrations in the flue gases are expressed in picograms (pg) or nanograms (ng) per normalized dry standard cubic meter (dscm)  normalized metered volume of sample run at ambient temperature of 20 °C (293.15 K) and under atmospheric pressure of 101.325 kPa (1 atm); the corresponding emissions are calculated in micrograms per minute (μg min1). The M-23 sampling protocol,20 routinely used for determination of PCDD/F from municipal waste combustors, was modified for simultaneous sampling of bromoorganics by addition of 13C12-labeled BDE and BDF presampling spikes and by measures to avoid photodegradation and debromination due to ultraviolet light exposure.21 Analysis and Quality Control/Quality Assurance. Extraction of train samples was performed by means of sequential Soxhlet extraction using methylene chloride (3.5 h) with restricted exposure to light, followed by 16 h extraction with toluene to ensure removal of other target PCDDs/Fs from the carbonaceous matrix. All raw extracts were concentrated using three-ball Snyder columns, filtered, and concentrated further with nitrogen to 0.5 mL using an automated evaporator (Zymark Turbovap). The samples were cleaned and fractionated using an automated liquid chromatography multicolumn (multilayer silica, basic alumina) Power Prep Dioxin System (FMS Fluid Management Systems, Inc., USA). Details on the concentrations and volumes of eluting solvents are given in Figure S-1 of Supporting Information. Concentrations of PBDDs/Fs and PCDDs/Fs were determined by the high resolution gas chromatography/high resolution mass spectrometry (HRGC/HRMS) analysis using Hewlett-Packard gas chromatograph 6890 Series equipped with a CTC Analytics Combi PAL autosampler (CTC Analytics, Switzerland) and coupled to a Micromass Premiere (Waters Inc., UK) double-focusing high resolution mass spectrometer operated in electron impact (35 eV and 650 μA current) selective ion recording (SIR) mode at resolution R > 10000 (5% valley). Details on the HRGC/HRMS operating conditions are given elsewhere.21 The recovery rates were 79% ((17.1% SD) for tetra- through octa-13C12-labeled PCDDs/Fs and 85.8% ((14% SD) for tetrathrough octa-13C12-labeled PBDDs/Fs, respectively (the mean recovery of tri-13C12-labeled BDF internal standard was below 60%). Reported values are recovery compensated. Routine procedural blanks were performed and analyzed. PBDDs/Fs were not detected (