Environ. Sci. Technol. 2003, 37, 1962-1970
Key Parameters for de novo Formation of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans EVALENA WIKSTRO ¨ M , ‡,# S H A W N R Y A N , ‡ ABDERRAHMANE TOUATI,§ AND B R I A N K . G U L L E T T * ,‡ U.S. Environmental Protection Agency, National Risk Management Research Laboratory, Air Pollution Prevention and Control Division, Air Pollution Technology Branch, Research Triangle Park, North Carolina 27711, and ARCADIS Geraghty & Miller, Inc., P.O. Box 13109, Research Triangle Park, North Carolina 27709
De novo formation of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDDs and PCDFs) was investigated in an Entrained Flow Reactor (EFR) to simulate combustion conditions. The parameters investigated were carbon content and nature in fly ash; type of gas-phase environment (oxidative versus reducing conditions) influence of combustion gases such as water, carbon monoxide, and carbon dioxide; amount of gas-phase chlorine; reaction temperature (250-600 °C); and reaction time (minutes vs hours). The comprehensive data set was further evaluated with principal component analysis (PCA) to statistically determine the role and importance of each parameter for de novo formation of PCDDs and PCDFs. Results revealed that an initial fast de novo formation occurs within the first minutes with a formation rate in the orders of hundreds of pmol per minutes; however, the reactivity of the ash was found to decline with time. An average formation rate as low as 3 pmol/min was measured after 6 h. The slower de novo formation of PCDDs and PCDFs was found to be through different reaction mechanisms and, thus, controlled by different parameters. The amount of Cl2 in the gas phase was observed to be an important parameter for PCDFs formation; meanwhile the levels of O2 were not found to be a PCDF rate controlling parameter. The formation rate of PCDDs was significantly lower than the PCDFs, and two mechanisms appear to be controlling the formation, one depending on the amount of O2 and one on the amount of Cl2 present in the gas phase. Overall the most significant parameter for the rate of formation for both PCDDs and PCDFs was revealed to be the reaction temperature. A maximum rate of formation was observed between 300-400 °C for the PCDDs and 400-500 °C for the PCDFs.
Introduction In 1987, Stieglitz and Vogg (1) proposed that a de novo mechanism could be a crucial formation pathway for * Corresponding author phone: (919)541-1534; fax: (919)541-0554; e-mail:
[email protected]. ‡ U.S. Environmental Protection Agency. § ARCADIS Geraghty & Miller, Inc. # Joint program with Oak Ridge Institute for Science and Education postdoctoral program, Oak Ridge, TN. 1962
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003
polychlorinated dibenzo-p-dioxin (PCDDs) and polychlorinated dibenzofurans (PCDFs) in the postcombustion zone (250-450 °C) in waste combustion processes. De novo formation of PCDDs/Fs is defined by Stieglitz (2) as a twostep process involving (i) formation of new carbon-chlorine bonds in a fly ash and (ii) a subsequent release of the newly formed chlorinated carbon structure from the matrix by further carbon degradation. A fly ash carbon matrix can be described as an ensemble of aromatic ring structures orientated in layers, i.e., similar in graphite, which, unlike fly ash, has been shown to be inactive as a carbon source for PCDDs/Fs (2, 3). However, the higher degree of hydrogen, oxygen, chlorine, and functional groups attached to the fly ash carbon ring structures compared to graphite increases the number of active sites (4), which allows the fly ash carbon to be more reactive for oxidation reactions and thus more inclined to participate in de novo formation of PCDDs/Fs. Carbon degradation/oxidation can be described as the result of two successive reactions, i.e., chemisorption of oxygen followed by catalytic or noncatalytic gasification (5). Noncatalyzed gasification has been found to require higher reaction temperatures, around 500 °C, compared to the catalyzed reactions, which are already active around 250300 °C (2). De novo experiments (3, 6, 7) with isotopically labeled 13C have shown that only a few percent of the PCDFs formed incorporated mixed 12C6 and 13C6 aromatic rings; meanwhile 50% of the PCDDs identified had one aromatic ring with either 12C6 or 13C6 (3, 6, 7). This implies that carbon structures similar to the dibenzofuran (DF) structure exist in the ash matrix and thereby act as a carbon source for de novo formation of PCDFs. Moreover the PCDF formation was revealed to be more favorable than PCDDs when a higher reaction temperature than 400 °C was employed (6, 7). Numerous studies have shown that polychlorinated phenols (PCPhs) can form PCDDs via condensation reactions either as gas-phase reactions or fly ash catalyzed reactions; the latter have shown a higher rate of formation than the gasphase formation mechanism at moderate reaction temperature (8-12). The formation via gas-phase reactions have shown to be dependent on O2 in the gas-phase, most likely due to the role of phenoxy radical for the PCDD formation to occur; meanwhile the fly ash catalyzed formation appears to be independent of gas-phase O2 (8, 11). Experiments conducted with isotopically labeled gasphase oxygen (18O) revealed a difference among and within the homologues regarding the incorporation of gas-phase oxygen implying differences in formation pathways for certain PCDD/F homologues (13, 14). The higher chlorinated homologues (hepta and octa) appear mainly to be formed by heterogeneous oxidation reactions with gas-phase oxygen incorporated in the final molecule structure. Meanwhile a direct release from the ash structure without incorporation of gas-phase oxygen was found to be an important formation pathway for the tetra- to hexa-CDDs/Fs homologues. The aim of this study is to investigate a large number of experimental parameters that may affect the formation of mono- to octa-CDDs and CDFs through de novo reactions and then to understand and elucidate the rate-controlling parameter. Parameters that were investigated are the amount and form of carbon present in the fly ash; the gas-phase composition (such as the presence of oxygen and chlorine); the interaction of combustion gases such as water (H2O), carbon dioxide (CO2), and carbon monoxide (CO); reaction temperature; and reaction time. Principal component analysis (PCA) was used to statistically calculate the role and 10.1021/es026240r CCC: $25.00
2003 American Chemical Society Published on Web 04/02/2003
FIGURE 1. Schematic drawing of the entrained flow reactor (EFR) used, not to scale. importance of each parameter for de novo formation of PCDDs and PCDFs.
Experimental Section 1. Experimental Set Up. Experiments were conducted in an Entrained Flow Reactor (EFR) which consists of two concentric quartz reactors, one horizontal (HR) and one vertical (VR) connected in series (Figure 1). The outer and inner concentric quartz tubes in the HR, allowing for gas preheating, are 46 mm and 22 mm inner diameter, respectively, with a length of 1.44 m. Inlets for oxygen/nitrogen (O2/N2) and chlorine (Cl2) gases are located at the end of the outer annulus of the HR. The HR can be equipped with a burner to study the role of combustion gases as well as the fly ash for the formation of PCDDs/Fs. However in this study no burner is employed, and the purpose of the HR is solely to mix and preheat the gases before contact with the fly ash. The VR consists of a similar concentric quartz tube configuration as the HR except with a 14 mm i.d. for the inner tube and an overall length of 3.09 m. To attain well-controlled temperature conditions in the EFR, four electric furnaces, one for the HR, and three for the VR, are employed. An additional gas inlet is located at the bottom of the outer annulus of the VR, which allows the gas to be preheated before mixing with the other constituents. The cap is wrapped in heating tape and connected to a fluidized bed, solids feeder for fly ash addition. Fly ash particles (
