Grantee Research Project Results
Final Report: Mechanistic Studies of the Transformation of Polychlorinated Dibenzo-p-Dioxins via Hydroxyl Radical Attack
EPA Grant Number: R828189Title: Mechanistic Studies of the Transformation of Polychlorinated Dibenzo-p-Dioxins via Hydroxyl Radical Attack
Investigators: Taylor, Philip H.
Institution: University of Dayton
EPA Project Officer: Hahn, Intaek
Project Period: October 1, 2000 through September 30, 2003
Project Amount: $320,000
RFA: Combustion Emissions (1999) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Air
Objective:
Polychlorinated dibenzo-p-dioxins (PCDD) are considered among the most toxic organic chemicals associated with our industrial society[1-3]. The gas-phase transformation of these chemicals under atmospheric conditions and high-temperature incineration (destruction) conditions is not well understood. Experimental and modeling studies repeatedly have shown that OH radical reactions are among the most important elementary steps under these reaction conditions[4,5]. A review of the literature demonstrates that knowledge of the rate of reaction of OH with dibenzo-p-dioxin (DD) and PCDD is limited to three low temperature experimental studies, or inferred by estimates of room temperature reactivity[6-10]. The mechanism of reaction is largely uncharacterized.
The overall goal of this research project was to determine the rates and mechanisms of hydroxyl radical (OH) reactions with DD and selected PCDD compounds over an extended temperature range. Mechanistic information was gleaned from the temperature dependence of the rate measurements and limited computational modeling. The specific objectives of this research project were to: (1) obtain first absolute rate measurements of the reaction of OH radicals with DD; 2-chloro dibenzo-p-dioxin (CDD); 2,3-dichlorinated dibenzo-p-dioxins (DCDD); 2,7-DCDD; 2,8-DCDD; 1,2,3,4-tetrachlorinated dibenzo-p-dioxin (TCDD); and octachloro dibenzo-p-dioxin (OCDD) over an extended temperature range with determination of accurate Arrhenius or modified Arrhenius parameters; (2) verify previous low temperature measurements for DD[6,9], 2,7-DCDD[8] and 1,2,3,4-TCDD[8]; and (3) investigate mechanisms of the reaction by examining the relative importance of the following pathways through limited theoretical analyses: OH addition, OH addition followed by Cl elimination, and H-atom abstraction.
Summary/Accomplishments (Outputs/Outcomes):
Temperature-dependent experimental measurements for the following reactions were completed:
OH + DD –> products | (k1) |
OH + 2-CDD –> products | (k2) |
OH + 2,3-DCDD –> products | (k3) |
OH + 2,7-DCDD –> products | (k4) |
OH + 2,8-DCDD –> products | (k5) |
OH + 1,2,3,4-TCDD –> products | (k6) |
OH + OCDD –> products | (k7) |
Our experimental efforts represent the first attempt to perform absolute rate measurements for this class of compounds that span the full range in chlorine substitution. Previous studies have utilized the relative rate technique[6-9] and have examined the less chlorinated congeners at temperatures representative of atmospheric conditions.
Experimental Approach and Data Reduction
The experimental procedures developed for pulsed laser photolysis-pulsed laser-induced fluorescence (PLP-PLIF) studies of the reaction of OH radicals with chlorinated dioxins are based on recent studies of chlorinated olefinic compounds[11,12]. The following paragraphs briefly summarize these procedures and discuss modifications to the experimental design that were necessary to perform measurements with halogenated species of much lower vapor pressure.
Three precursors—nitrogen dioxide/water (N2O/H2O) mixtures, hydrogen peroxide (H2O2), and nitrous acid (HNO2), were employed to generate OH radicals. Measurements at lower temperatures were conducted with H2O2 (248 nm) and HNO2 (for DD) (351 nm) to investigate the effect of photolysis wavelength on the rate measurements. The N2O/H2O precursor system was used predominantly at elevated temperatures. To minimize potential photolysis of the substrate, low excimer laser photolysis energies (0.15-0.3 mJ/cm2 at 193 nm and 2.5-5.0 mJ/cm2 at 248 nm and 351 nm) were used in all experiments.
Detection of OH radicals was achieved by PLIF, exciting the (1,0) band of the OH (A2-X2 ) system at 282.2 nm using a pulsed (10 Hz) Nd:YAG pumped-dye laser (Quanta Ray DCR-1/PDL-2). The laser fluence at the entrance of the reactor was about 200µJ/pulse. Broadband fluorescence at 306 nm was collected using a PMT/band-pass filter combination[10-12].
A modified fused silica optical reactor was designed and constructed for this study. Two significant changes to our previous reactor designs[11,12] were made to successfully transport parts per million concentrations of chlorinated dioxins in the gas phase from the point of sample introduction to the reaction volume (intersection of pump and probe laser beams). The unheated Brewster windows attached to the arms of the previous reactor were replaced with fused silica windows installed about 2.2 cm from the reaction volume. The windows were heated to the same temperature as the cylindrical reactor to minimize condensation of the substrate on the optical windows. A fused silica sample inlet and inlet probe were constructed and located immediately beneath the cylindrical reactor to provide a means for rapid and efficient introduction of the substrate vapors to the reaction volume while minimizing condensation of the sample during transport.
Solid substrates were introduced with a fused silica sample inlet probe, 16.4 mm long x 4 mm in diameter. During a typical experiment, 4-5 mg of sample was loaded in the probe, fixed in place with quartz wool. A sample gas flow (argon or helium), controllable from 1.7 to 10.6 mL/minute, provided a means for varying the substrate concentration in the reactor. Substrate concentrations either were determined directly (for DD, 2-CDD, and 1,2,3,4-TCDD) with integrated sampling at the exit of the reactor using Tenax® adsorbent and subsequent gas chromatograph/mass spectrometry analyses, or were calculated (for 2,3-DCDD, 2,7-DCDD, 2,8-DCDD, and OCDD) based on recently published vapor pressure measurements[13-16], the measured substrate temperature, the measured carrier gas flow rate through the sample inlet probe, and the measured carrier gas flow rate through the reactor.
The buildup of reaction products was minimized by conducting experiments under slow flow conditions. Both helium and argon were used as the carrier gas for the kinetic experiments. Total gas flows ranged from 220 to 775 mL/minute; linear gas velocities based on the plug flow assumption ranged from 12.9 to 37.6 cm/second. All experiments were performed at a total pressure of 740±10 torr. The dioxin and chlorinated dioxin substrates were purchased from Ultra Scientific, 98+ percent purity, and used as received.
All experiments were performed under pseudo-first order conditions. Pseudo-first order exponential OH decays were observed, confirming that the substrate concentration was in large excess of OH (at least a factor of 100 larger). OH decays were observed over two to three decay lifetimes over a time interval of 0.5-30.0 milliseconds. The individual bimolecular rate constants were determined from the relation k' = kbi[substrate] + kd, where the bimolecular rate constant, kbi, is the slope of the least-squares fit of k' versus substrate concentration.
Results and Discussion
Prior to conducting the measurements, calibration tests were performed to verify the accuracy of the experimental method. The very low vapor pressures of the substrates did not permit room temperature calibration measurements using this technique. As a result, verification of this sample introduction procedure was obtained by conducting room temperature rate measurements for a structurally similar, solid-phase sample with a higher vapor pressure, phenol. A rate of 2.1±0.14 x 10-11 cm3/molecule-second was measured at 296 K using H2O2 as the OH precursor, in excellent agreement with the previous measurements by Semadeni, et al.[17], and Rinke and Zetzsch[18] (2.32±0.20 x 10-11 and 2.81±0.58 x 10-11 cm3/molecule-second, respectively). The excellent agreement with previous measurements using different experimental methods validated our experimental approach.
Absolute rate measurements for k1-k7 were obtained over a substantial temperature range. Ninety five percent confidence intervals (±2) representing random statistical errors in the data ranged from 5 to 25 percent, with most rate coefficients exhibiting ±2 values between 10 and 15 percent. Changes in carrier gas (helium or argon) had no impact on the rate measurements within statistical uncertainties. Variation in photolysis wavelength (193 and 248 nm, and 351 nm for DD and 2-CDD), photolysis energy (factor of 3), total flow rate (factor of 3), substrate concentration (factor of 2), and initial OH concentration (factor of 2) also had no impact on the rate measurements within statistical uncertainties. The overall uncertainty in the measurements, taking into account systematic errors dominated by uncertainty in the substrate reactor concentration, range from ± a factor of 2 for DD, 2-CDD, 2,3-DCDD, 2,7-DCDD, and 2,8-DCDD to ± a factor of 4 for 1,2,3,4-TCDD and OCDD.
An Arrhenius fit of our extended temperature data for k1-k7 yielded the following expressions (in units of cm3/molecule-second, error bars are 1s):
k1 (326-907 K) = (1.70±0.22) x 10-12 exp(979±55)/T,
k2 (346-905 K) = (2.79±0.27) x 10-12 exp(784±54)/T,
k3 (400-927 K) = (1.83±0.19) x 10-12 exp(742±67)/T,
k4 (390-769 K) = (1.10±0.10) x 10-12 exp(569±53)/T,
k5 (379-931 K) = (1.02±0.10) x 10-12 exp(580±68)/T,
k6 (409-936 K) = (1.66±0.38) x 10-12 exp(713±114)/T,
k7 (514-928 K) = (3.18±0.54) x 10-11 exp(-667±115)/T.
Measurements at lower temperatures were not possible because of the inability to establish pseudo-first order conditions in the reactor because of the low substrate concentrations. Measurements at elevated temperatures were limited by the onset of thermal decomposition of N2O, the photochemical precursor to OH radical generation, and the onset of thermal decomposition of the substrate[19]. For our reaction conditions, this limiting temperature is approximately 1,000 K.
Comparison of our absolute rate measurements for DD, 2,7-DCDD, and 1,2,3,4-TCDD with previous relative rate measurements generally were within combined experimental uncertainties of the respective measurements. As for other aromatic compounds at low temperatures, the gas-phase reactions of OH with DD, 2-CDD, 2,3-DCDD, 2,7-DCDD, 2,8-DCDD, and 1,2,3,4-TCDD likely proceed by initial addition of the OH radical to the aromatic rings to form a hydroxycyclohexadienyl-type radical that subsequently is stabilized by collisions with the bath gas[6,7,10,20]. The magnitude and negative temperature dependence of our measurements are consistent with this mechanism.
At elevated temperatures (> 500 K) relevant to combustion and post-combustion conditions, our data does not exhibit any evidence for a change in reaction mechanism from the formation of a stabilized OH addition adduct to H abstraction (large decrease in observed rate with a reversal in temperature dependence) as has been observed in our laboratories for chlorinated substrates containing a double bond[11,12]. Our theoretical estimates of the rate of H abstraction suggest that this reaction does not contribute to the observed rate at temperatures below 1,000 K. Further details regarding these calculations are provided in the final report. Our experimental measurements indicate that H abstraction is not a significant pathway below 1,000 K for all chlorinated congeners investigated. It is possible that at higher temperatures, H abstraction may be a significant, if not dominant, mechanism for all dioxin congeners, excluding OCDD.
Our results generally are consistent with previous results at low temperatures indicating a decrease in reactivity with increasing Cl substitution[6,9]. We observed a similar rate of reaction for DD and 2-CDD at room temperature. Multiple Cl substitution on a single aromatic ring resulted in a measurable (factor of 2) decrease in OH reactivity. Cl substitution on both aromatic rings resulted in a much larger (factor 5 to 13) decrease in room temperature reactivity. These results generally are consistent with the OH addition mechanism and indicate a significant steric effect when Cl atoms are present on both aromatic rings.
The reaction rate and temperature dependence of the OCDD + OH reaction was different from all other chlorinated dioxin congeners that were examined in this study. OCDD exhibited a much slower reaction rate and a positive temperature dependence throughout the temperature range investigated. The measured activation energy (1.3±0.2 kcal/mol) is similar to that measured for OH + C2Cl4 (1.5±0.2 kcal/mol)[21]. Quantum RRK modeling of the OH + C2Cl4 reaction indicated that OH addition followed by Cl elimination was the dominant reaction pathway at all temperatures[21]. A similar mechanism is likely for the reaction of OH with OCDD. This is consistent with the observed temperature-dependent behavior for OCDD + OH. OH addition followed by Cl elimination is not significant for the other chlorinated dioxin congeners because of the much faster OH addition to the nonchlorinated carbon sites.
References:
1. Baker JI, Hites RA. Is combustion the major source of polychlorinated dibenzo-p-dioxins and dibenzofurans to the environment? A mass balance investigation. Environmental Science and Technology 2000;34:2879-2886.
2. National Center for Environmental Assessment, Office of Research and Development, U.S. Environmental Protection Agency. The inventory of sources of dioxin in the United States. Washington, DC, April 1998, External Review Draft, EPA/600/P-98/002Aa.
3. Kociba RJ, Cabey O. Comparative toxicity and biologic activity of chlorinated dibenzo-p-dioxins and furans relative to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Chemosphere 1985;14:649-660.
4. Senkan SM. In: Gardiner Jr. WC, ed. Gas-phase Combustion Chemistry. New York, NY: Springer-Verlag, 2000, Chapter 4.
5. Finlayson-Pitts BJ, Pitts Jr. JN. Chemistry of the Upper and Lower Atmosphere. San Diego, CA: Academic Press, 1999.
6. Kwok ESC, Arey J, Atkinson R. Gas-phase atmospheric chemistry of dibenzo-p-dioxin and dibenzofuran. Environmental Science and Technology 1994;28:528-533.
7. Kwok ESC, Atkinson R, Arey J. Rate constants for the gas-phase reactions of the OH radical with dichlorobiphenyls, 1-chlorodibenzo-p-dioxin, 1,2-dimethoxybenzene, and diphenyl ether: estimation of OH radical reaction rate constants for PCBs, PCDDs, and PCDFs. Environmental Science and Technology 1995;29:1591-1598.
8. Brubaker Jr. WW, Hites RA. Polychlorinated dibenzo-p-dioxins and dibenzofurans: gas-phase hydroxyl radical reactions and related atmospheric removal. Environmental Science and Technology 1997;31:1805-1810.
9. Brubaker Jr. WW, Hites RA. OH reaction kinetics of polycyclic aromatic hydrocarbons and polychlorinated dibenzo-p-dioxins and dibenzofurans. Journal of Physical Chemistry A 1998;102:915-921.
10. Atkinson R. In: Hester RE, Harrison RM, eds. Issues in environmental science and technology. Cambridge, United Kingdom: The Royal Society of Chemistry, 1996, Vol. 6, p. 53.
11. Yamada T, El-Sinawi A, Siraj M, Taylor PH, Peng J, Hu X, Marshall P. Rate coefficients and mechanistic analysis for the reaction of hydroxyl radicals with 1,1-dichloroethylene and trans-1,2-dichloroethylene over an extended temperature range. Journal of Physical Chemistry A 2001;105;7588-7597.
12. Yamada T, Siraj M, Taylor PH, Peng J, Hu X, Marshall P. Rate coefficients and mechanistic analysis for reaction of OH with vinyl chloride between 293 and 730 k. Journal of Physical Chemistry A 2001;105:9436-9444.
13. Rordorf BF. Containment of dioxin emissions from refuse fired thermal processing units - prospects and technical issues. Thermochimica Acta 1985;85:435-438.
14. Rordorf BF. Thermodynamic and thermal properties of polychlorinated compounds: The vapor pressures and flow tube kinetics of ten dibenzo-para-dioxines. Chemosphere 1985;14:885-892.
15. Rordorf BF, Sarna LP, Webster GRB. Vapor pressure determination for several polychlorodioxins by two gas saturation methods. Chemosphere 1986;15:2073-2076.
16. Mader BT, Pankow JF. Vapor pressures of the polychlorinated dibenzodioxins (PCDDs) and the polychlorinated dibenzofurans (PCDFs). Atmospheric Environment 2003;37:3103.
17. Semadeni M, Stocker DW, Kerr JA. International Journal of Chemical Kinetics 1995;27:287.
18. Rinke M, Zetzsch C. Berichte der Bunsen-Gesellschaft-Physical Chemistry 1984;99:55.
19. Rordorf BF, Marti E. Thermodynamic properties of polychlorinated compounds: the vapor pressures and enthalpies of sublimation of ten dibenzo-para-dioxines. Thermochimica Acta 1985;85:439-442.
20. Atkinson R. Kinetics and mechanisms of the gas-phase reactions of the hydroxyl radical with organic compounds. Journal of Physical and Chemical Reference Data 1989, Monograph No.1, 246 pp., ISBN 0-88318-720-5.
21. Tichenor LB, Graham JL, Yamada T, Taylor PH, Peng J, Hu X, Marshall P. Kinetic and modeling studies of the reaction of hydroxyl radicals with tetrachloroethylene. Journal of Physical Chemistry A 2000;104:1700-1707.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
Other project views: | All 9 publications | 1 publications in selected types | All 1 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Taylor PH, Yamada T, Neuforth A. Kinetics of OH radical reactions with dibenzo-p-dioxin and selected chlorinated dibenzo-p-dioxins. Chemosphere 2005;58(3):243-252. |
R828189 (2002) R828189 (Final) |
Exit |
Supplemental Keywords:
dioxin, combustion, oxidation, modeling, exposure, air pollution, hazardous air pollutants, hydroxyl radical, OH, OH reactions, dibenzo-p-dioxin, DD, polychlorinated DD, PCDD, 1,2,3,4-TCDD., RFA, Scientific Discipline, Toxics, Waste, Chemical Engineering, Environmental Chemistry, pesticides, Chemistry, Incineration/Combustion, Environmental Engineering, dioxin, gas-phase transformation, industrial waste, chemical contaminants, analytical chemistry, hydrocarbons, toxic organic chemicals, mechanistic study, fused silica test cell, incineration, combustion contaminants, laser photolysisProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.