Document 70RaD7bqkDdOmDQn9Z9OLx72o
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 116, D24307, doi:10.1029/2011JD016518, 2011
OH reaction rate constant, IR absorption spectrum,
ozone depletion potentials and global warming
potentials of 2-bromo-3,3,3-trifluoropropene
Kenneth O. Patten,1 Victor G. Khamaganov,2 Vladimir L. Orkin,2 Steven L. Baughcum,3 and Donald J. Wuebbles1
Received 5 July 2011; revised 1 November 2011; accepted 4 November 2011; published 30 December 2011.
[1] The rate constant for the gas phase reaction of OH radicals with BTP (2-bromo-3,3,3-trifluoropropene, CH2 = CBrCF3) was measured using a flash photolysis resonance-fluorescence technique over the temperature range 220 K to 370 K. The Arrhenius plot was found to exhibit noticeable curvature. The temperature dependence of the rate constant can be represented as kBTP(220 370 K) = 4.85 1013 (T/298)0.92 exp{+613/T} cm3 molecule1 s1. For atmospheric modeling purposes, kBTP(T) can be equally well represented by the standard Arrhenius expression kBTP(220 298 K) = 1.05 1012 exp{+381/T} cm3 molecule1 s1. The IR absorption cross-sections of BTP were also measured between 450 cm1 and 1900 cm1. BTP atmospheric lifetime, Ozone Depletion Potential (ODP), and Global Warming Potential (GWP) were evaluated in the Whole Atmosphere Community Climate Model for land emissions from 30 to 60N and from 60S to 60N. The global, annual average atmospheric lifetime of BTP in the former scenario was 7.0 days, its ODP was 0.0028, and its GWP (100-yr time horizon) was 0.0050; in the latter scenario, the global, annual average BTP lifetime was 4.3 days, ODP was 0.0052, and 100-yr GWP was 0.0028. The short lifetime, low ODP, and low GWP indicate that BTP should have minimal effects on ozone and climate. Little BTP reaches the stratosphere in either emission scenario, but 27% of the ozone loss in the 30 to 60N scenario and 46% of the ozone loss in the 60S to 60N scenario occurs above the tropopause due to Bry from BTP.
Citation: Patten, K. O., V. G. Khamaganov, V. L. Orkin, S. L. Baughcum, and D. J. Wuebbles (2011), OH reaction rate constant, IR absorption spectrum, ozone depletion potentials and global warming potentials of 2-bromo-3,3,3-trifluoropropene, J. Geophys. Res., 116, D24307, doi:10.1029/2011JD016518.
1. Introduction
[2] Since the recognition of adverse environmental effects from the long-lived, fully halogenated gases such as the chlorofluorocarbons (CFCs) and Halons, industry has sought replacement compounds for those gases that perform their functions, but lack adverse ozone or climate effects. The primary approach to this problem has been to test candidate replacement compounds that have short atmospheric lifetimes and lack chlorine or bromine, the halogen atoms that catalyze ozone destruction.
[3] One recently proposed candidate as a possible substitute for Halons in handheld fire extinguishers used in the presence of people in confined spaces such as aircraft cabins is 2-bromo-3,3,3-trifluoropropene (BTP, also named 2-bromo-
1Department of Atmospheric Sciences, University of Illinois at UrbanaChampaign, Urbana, Illinois, USA.
2National Institute of Standards and Technology, Gaithersburg, Maryland, USA.
3Boeing Company, Seattle, Washington, USA.
Copyright 2011 by the American Geophysical Union. 0148-0227/11/2011JD016518
3,3,3-trifluoropropylene, chemical formula CH2 = CBrCF3). The efficiency of bromine at catalyzing ozone destruction, because of the ease in which bromine atom and bromine oxide are regenerated from bromine reservoir species, results in Ozone Depletion Potential (ODP) values for many brominated compounds that are unacceptably high. The presence of a carbon-carbon double bond is expected to render BTP highly reactive toward the hydroxyl radical (OH) in the troposphere, so that relatively little BTP should reach the stratosphere to impact stratospheric ozone.
[4] The reaction rate constant with OH, kBTP(T), for temperatures between 250 and 370 K and the UV absorption cross sections for BTP between 164 and 253 nm have been measured by Orkin et al. [2002]. The BTP atmospheric lifetime was estimated from the OH rate constant expression by the method of Spivakovsky et al. [2000] as 2.8 days. The Orkin et al. [2002] rate constant kBTP(T) was later used by Brioude et al. [2010] in a calculation of BTP atmospheric lifetime and ODP using the FLEXPART Lagrangian model. Brioude et al. obtained BTP lifetimes ranging from 3.1 days for emissions originating in the Indian subcontinent to 7.1 days for emissions originating in Europe, and reported
D24307
1 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Figure 1. OH decay rates due to reaction with CH2 = CBrCF3 versus the reactant concentrations at 220 K, 272 K, 298 K, and 370 K.
annually averaged ODPs ranging from 0.0013 for European emissions to 0.044 for emissions from the Indian subcontinent in Brioude et al.'s [2010] auxiliary material.
[5] The fast reaction with OH radicals in the troposphere is likely to be the major removal process for BTP in the atmosphere which dictates its residence time and, therefore, potential environmental impacts. Therefore, we re-visited this reaction using our recently improved apparatus to provide more accurate kinetic data over the atmospheric temperatures to be used in the following detailed modeling. We also measured IR absorption cross sections of BTP to provide a basis for GWP estimations. Then, we obtain the atmospheric lifetime and ODP using the Whole Atmosphere Community Climate Model (WACCM) for two scenarios of BTP emissions from land surfaces. The first, representing the most likely region of initial use, is from all land surfaces at latitudes 30N to 60N. The second scenario is from all land surfaces at latitudes 60S to 60N, similar to the "global" emissions used for many past ODP studies on short-lived source gases such as npropyl bromide [Wuebbles et al., 2001; Ko et al., 2003; Wuebbles et al., 2011]. The WACCM results are also used with our radiative transfer model to derive estimates of Global Warming Potentials (GWPs) for BTP in those scenarios.
2. Experimental Section
2.1. OH Reaction Rate Constant Measurements
[6] General descriptions of the apparatus and the experimental method used to measure the OH reaction rate constants are given in previous papers [Kurylo et al., 1982; Orkin et al., 1996]. In the present work we used the same principal configuration of the apparatus with a number of modifications made to improve the accuracy of the obtained kinetic data by decreasing and quantifying the instrumental uncertainties. In particular, the gas handling system was completely replaced and a new reaction cell and photomultiplier
were used. Modifications to the apparatus and the measurement procedure were recently described in detail [Orkin et al., 2010, 2011]. The reaction was studied in a doublewalled Pyrex reactor equipped with quartz windows by using a flash photolysis-resonance fluorescence technique. The reactor temperature is controlled with methanol or water circulated between the outer walls. The reaction was studied in argon carrier gas at total pressure of 4.00 kPa (30.0 Torr). Flows of dry argon, argon bubbled through water temperature-controlled at 276 K, and a mixture of BTP diluted with Ar flowed through the reactor at a total flow rate of 1.3 cm3 s1, STP. The BTP/Ar mixtures were prepared in 5 L glass bulbs equipped with Teflon/glass valves. The concentrations of the gases in the reactor were determined by measuring the gas flow rates and the total pressure with MKS Baratron manometers. Flow rates of argon, H2O/Ar mixture, and the reactant/argon mixture were measured and maintained using MKS mass flow controllers directly calibrated for every gas mixture. Hydroxyl radicals were produced by the pulsed flash photolysis of H2O and monitored by their resonance fluorescence near 308 nm as described in detail in earlier publications. The procedure for deriving the reaction rate constant from such data has been described by Orkin et al. [1996, 2010]. Figure 1 gives examples of the derived OH decay rates due to reaction with CH2 = CBrCF3 plotted versus the reactant concentrations at various temperatures. At each temperature the rate constant was determined from a fit to all of the decay rates obtained at that temperature (i.e., the slope of a plot similar to those shown in Figure 1). The temperatures for the measurements were chosen to be approximately equally distant along the Arrhenius 1/T scale so that they would be equally weighted in the fitting procedure. The accuracy of kinetic data was analyzed in detail by Orkin et al. [2010, 2011].
2.2. IR Absorption Cross-Sections Measurements
[7] The IR absorption spectrum was measured by using a Nicolet 6700 FTIR spectrophotometer with spectral resolutions of 0.125 cm1 and 0.25 cm1 (recorded with a step of 0.06 cm1 and 0.12 cm1, respectively). A (10.2 0.05) cm glass absorption cell fitted with KBr windows was fixed in the spectrophotometer to minimize a baseline shift due to changes in the absorption cell position. The temperature of the cell was measured to be (295 1) K. Absorption spectra of the evacuated cell and of the cell filled with a gas sample were alternately recorded several times, and the absorption cross sections at the wave number n (cm1) were calculated. Details and potential complications of these measurements are discussed by Orkin et al. [2010]. The overall instrumental uncertainty associated with the optical path length, pressure measurements, and temperature stability was estimated to be less than 1%.
2.3. Materials
[8] The sample of 2-bromo-3,3,3-trifluoropropene was obtained from SynQuest Laboratories, Inc., and it was degassed through multiple freeze-pump-thaw-boil cycles. The bulk liquid sample had a stated purity of 99% with 0.5% of 1,2-dibromo-3,3,3-trifluoropropane (CH2BrCHBrCF3) and 0.2% of 1-bromo-3,3,3-trifluoroacetone (CH2BrC(O)CF3) being the main impurities. The concentration of these lower volatility compounds in the vapor phase was substantially
2 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Table 1. Rate Constants Measured in the Present Work for the Reaction of OH With CH2 = CBrCF3a
kBTP(T) 1012
[CH2 = CBrCF3]
T, K (cm3 molecule1 s1) (1013 molecule/cm3)
Test Experiment Conditions
220 5.96 0.04 (30) 230 5.48 0.06 (21) 240 5.13 0.06 (9) 250 4.80 0.04 (17) 260 4.56 0.03 (16) 272 4.27 0.05 (17)
4.26 0.06 (15) 285 3.98 0.02 (9) 298 3.79 0.02 (35)
3.76 0.04 (8) 3.75 0.03 (8) 3.79 0.06 (8) 3.86 0.03 (8) 313 3.61 0.02 (17) 330 3.40 0.04 (18) 350 3.24 0.04 (9) 370 3.11 0.02 (28)
0.47-3.63 0.56-4.49 0.62-4.49 0.56-4.49 0.62-5.22 0.78-6.01 0.84-5.34
0.78-6.00 0.41-5.95 0.77-5.93 0.77-5.93 1.10-6.37 0.58-5.34
0.77-5.96 0.77-5.97 0.77-5.94 0.77-5.96
Pressure: 0.93 kPa (7 Torr)
[OH]0: 50% Flash Energy: 200% [OH]0: 200% [OH]0: 400%,
Flash Energy: 200%
aThe uncertainties are two Standard Errors from the least squares fit of a straight line to the measured OH decay rates versus the reactant concentrations. When estimates for instrumental uncertainties are added, the overall uncertainty of the rate constant (95% confidence limit) is estimated to be 2% increasing to 2.5% at lowest temperatures. The boldfaced data are results of the fit to all measurements performed at the particular temperature and 4.00 kPa (30.0 Torr) total pressure using a 0.25% mixture. These data are shown in Figure 1 and were used to derive the temperature dependences. The italicized data are results of some test experiments. The number of experiments for each temperature is included in parentheses.
unsaturated C atom. Therefore, in spite of the multiatomic nature of the BTP molecule there is a possibility that the
reaction is pressure dependent. The test experiments revealed no indication of any variation in the measured rate constants when pressure was changed.
[11] The Arrhenius plot in Figure 2 exhibits a curvature that is clearly resolved in our experiments. Our threeparameter modified Arrhenius expression fit to the data set from Table 1, as shown in Figure 2, is:
kBTP220
370 K
4:85 10 13 T=2980:92 expf613=Tg cm3 molecule1 s1: 1
A goodness of the fit cannot be seen in the Arrhenius plot. Therefore, Figure 2 (bottom) shows the experimental data reported in Table 1 normalized by this expression (1) with error bars representing the relative errors of measurements from Table 1 (filled circles). For atmospheric modeling purposes, kBTP(T) below room temperatures can be represented by the standard Arrhenius expression obtained from the fit to the data between 220 K and 298 K
kBTP220
298 K 1:05 0:04 10 12 expf381 10=Tg cm3 molecule1 s1:
2
smaller (0.02% to 0.05%) resulting in 99.8% purity of the vapor phase, which was used to prepare mixtures for OH reaction rate constant measurements. A few smaller lowvolatile impurities were detected in the liquid phase by a GC-FID analysis. The concentration of these impurities in the vapor phase was also substantially smaller than in the liquid phase. The vapor phase was enriched with the more volatile olefinic impurity, 3,3,3-trifluoropropene (up to 0.06% compared to less than 0.006% in the liquid phase) and a small amount of 1,1,1-trifluoroacetone (0.002%). We used 99.9999% purity argon (Spectra Gases Inc.) as a carrier gas.
3. Experimental Results
3.1. OH Reaction Rate Constant
[9] The rate constants determined for this reaction at various temperatures are presented in Table 1. The bold-highlighted data result from fits to all measurements performed at the indicated temperature under our "standard" conditions: 4.00 kPa (30.0 Torr) total pressure, 6 1014 molecule/cm3 of H2O in the reactor. These data are shown in Figure 2 with filled circles.
[10] To check for any complications, test experiments were performed with variations of H2O concentration, flash energy, and the total pressure in the reactor. The experiments were done to check for a possible influence of the secondary radical or photolysis product reactions on the measured rate constant. Some results of test experiments (italicized) are shown in Table 1. No statistically significant changes of the measured reaction rate constant were observed in these experiments. Note that the title reaction occurs via addition of OH to an
Figure 2. (top) The results of rate constant measurements for the reaction between OH and CH2 = CBrCF3, kBTP(T): squares, Orkin et al. [2002]; circles, this study. The solid line shows a three-parameter fit to the data from this study given by expression (1). (bottom) The ratio of rate constants measured in this study to those calculated from the three-parameter temperature dependence (1) (solid circles) and from the twoparameter Arrhenius dependence (2) (open circles).
3 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Figure 3. IR absorption spectrum of BTP obtained with a spectral resolution of 0.25 cm1.
[12] The experimental data normalized by expression (2) are shown in Figure 2 (bottom) with open circles. Their relative errors are the same as shown for the corresponding filled circles; they are omitted to not overload Figure 2. One can see that the rate constants measured at below room temperatures deviate from the recommended expression (2) by less than 0.5%. However, expression (2) underestimates the rate constant above room temperature.
[13] The presence of reactive impurities is a potential source for overestimating the measured reaction rate constant. However, in our case the detected impurities are less reactive toward OH than the target compound, CH2 = CBrCF3. The rate constant of the reaction between OH and CH2 = CHCF3 is smaller than kBTP determined in this study by more than a factor of 2 [Orkin et al., 1997] and, therefore, the presence of this impurity in the amount of less than 0.1% cannot contribute to the measured rate constant. There are no available data on the reactivity of either CH2BrCHBrCF3 or CH2BrC(O)CF3 toward OH. However, the OH reaction rate constants of similar molecules, CH3CH2CF3 and CH3C(O)CF3, are much smaller than kBTP measured in this study. The difference is as large as more than a factor of 60 for CH3CH2CF3 [Rajakumar et al., 2006] and more than a factor of 500 for CH3C(O)CF3 [Carr et al., 2003], respectively at room temperature and increasing at lower temperatures. Also, the test experiments were done to compare the results obtained with mixtures prepared from the vapor phase and liquid phase of the BTP sample, which have substantially different amounts of impurities. The OH reaction rate constants measured with these two
mixtures were statistically indistinguishable with the difference of less than 1%. Thus we are confident that the results of our measurements are not affected by the presence of impurities in the sample.
[14] Results of the only available set of measurements of kBTP(T) are shown in Figure 2 with open squares. These data exceed the prediction of expression (1) by less than 4% and therefore coincide with results of this work within the combined uncertainty. The reported total uncertainty of these older data was estimated to be ca. 6%. The detailed analysis of uncertainties presented by Orkin et al. [2010] allows us to estimate the total instrumental uncertainty of the reaction rate constant measured in the present study to be as small as 2% (room temperature and above) to 2.5% (at lowest measurement temperatures). Based on the available kinetic data we can conclude that impurities found in the BTP sample are less reactive toward OH than the main compound. Therefore, an additional uncertainty of the measured rate constant due to the presence of impurities should not exceed 1%. Although both available data sets for the title reaction were obtained in the same laboratory, the apparatus was completely rebuilt between these studies. Also, different samples of BTP were used in these two studies with the particular attention to their purity and the nature of the impurities. Thus we can conclude that the rate constant of the reaction between OH and BTP is well established.
3.2. IR Absorption Spectra
[15] The IR absorption spectrum of BTP between 450 cm1 and 1900 cm1 is shown in Figure 3. The spectrum was combined from the results of measurements at various pressures of BTP between 0.13 kPa to 27 kPa (1 Torr to 200 Torr). Figure 3 (top) illustrates the main absorption bands while Figure 3 (bottom) shows the same spectrum in log scale to illustrate smaller absorption features. IR absorption cross sections of BTP are available in the auxiliary material.1
[16] Further experiments were performed with spectral resolution of 0.125 cm1 and with CH2 = CBrCF3 pressurized with nitrogen bath gas to test the adequacy of the IR spectrum presented in Figure 3 for atmospheric applications. The presence of either 44.4 kPa (333 Torr) or 88.4 kPa (666 Torr) of nitrogen bath gas changes the integrated intensities of main absorption bands located near 1280 cm1, 1185 cm1, and 1090 cm1 by less than 1%. The spectral resolution of 0.125 cm1 does not affect the integrated band intensities either. Only small narrow absorption peaks (1176.5 cm1, 1088.0 cm1, and 930.9 cm1) at the top of the wide absorption bands have been slightly affected (3% to 12%) by the spectral resolution and the presence of bath gas. However, these absorption features are very small (0.2% to 0.5%) compared to the integrated absorption by BTP.
4. Atmospheric Effects of BTP
4.1. Atmospheric Chemistry-Climate Model
[17] WACCM version 3.5.48, as used in this study and by investigators at the National Center for Atmospheric Research
1Auxiliary material data sets are available at ftp://ftp.agu.org/apend/jd/ 2011jd016518. Other auxiliary material files are in the HTML. doi:10.1029/ 2011JD016518.
4 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
for the SPARC (Stratospheric Processes and their Role in Climate) 2010 Report on the Evaluation of ChemistryClimate Models [Eyring et al., 2010], represents the Earth's atmosphere from the surface to the lower thermosphere (4.5 106 mbar, or roughly 145 km) at a resolution of 2.5 in longitude by 1.9 in latitude. This three-dimensional chemistry-climate model is run in offline mode for this study, in which temperature, winds, specific humidity, and other meteorological fields are supplied by a previous run of WACCM 3.5 representing the year 2000 (R. Garcia, private communication, 2007). WACCM 3.5.48 includes 125 species, of which 15 are relevant mostly to thermospheric processes, and uses the JPL06 chemistry recommendation [Sander et al., 2006] extended for non-methane hydrocarbon chemistry to best represent tropospheric oxidative capacity in a similar way to the tropospheric chemistry in the Model for Ozone and Related Tracers (MOZART) version 3 [Kinnison et al., 2007] and version 4 [Emmons et al., 2010]. The WACCM lightning NOx yield has been scaled globally to produce an annual total NOx of 5.0 TgN, consistent with the recommendation of Schumann and Huntrieser [2007]. Wet deposition is also based on the MOZART 3 approach [Kinnison et al., 2007; Horowitz et al., 2003], in which removal of highly soluble species including HBr, HOBr, and BrONO2 takes place with first-order rates based on the rainfall rate diagnosed within the model convection operator. The model did well in the evaluations for the stratosphere and the upper troposphere and lower stratosphere (UTLS) in CCMVal [Eyring et al., 2010; Gettelman et al., 2010]. The tropospheric chemistry in this model updates the previous version of MOZART2.4 as evaluated and compared with tropospheric observations by Horowitz et al. [2003, 2007], but, while the tropospheric values appear to be similar to the earlier evaluations, this model has not undergone the same level of comparisons for the troposphere as was done in the earlier studies.
[18] BTP is expected to react primarily with OH in the atmosphere. This reaction, with the rate constant given in equation (2) above, is the only BTP loss process represented in WACCM for this study. WACCM upper tropospheric temperatures can range below the 220 K minimum temperature of our measurements, but BTP concentrations are minimal in those regions of the atmosphere as will be shown below. Thus, we expect little effect on our model results from temperature extrapolation below the range of equation (2). Other tropospheric reactions besides that with OH are possible for this brominated alkene, including reactions with such oxidants as O3, NO3, and Cl, but their rate parameters are currently unknown. Photolysis is also a possible loss process, but the absorption cross sections decrease rapidly for wavelengths longer than 220 nm [Orkin et al., 2002] while solar flux at wavelengths shorter than 300 nm at altitudes below stratospheric O3 is negligible, suggesting that photolysis of BTP is unimportant below the stratospheric O3 layer. Any BTP reactions in the troposphere in addition to that with OH will reduce its atmospheric lifetime, decreasing its climate and ozone effects compared to our results below.
[19] For this combined ozone and climate effects study, we should consider the possible production of compounds with longer atmospheric lifetimes than BTP, whether or not they contain bromine. While yields of BTP oxidation products under atmospheric conditions have not been measured, we
expect OH to add to either alkene carbon. The most likely initial products of OH addition at the 1-carbon are Br + CF3C (O)CH2OH, and those for OH addition at the 2-carbon are CH2O + CF3CBrO. The rate constant for OH reaction with CF3C(O)CH2OH is currently not available, and the following estimation of its tropospheric chemistry is speculative. The OH reaction rate constant with CH3C(O)CH2OH [Atkinson et al., 2006] (citation as updated on the IUPAC Subcommittee on Gas Kinetic Data Evaluation Web site http://www.iupackinetic.ch.cam.ac.uk/, Data Sheet HOx_VOC32, 18 December 2007) suggests that CF3C(O)CH2OH could react with OH nearly as quickly as does BTP, yielding CH2O plus the CF3C (O)O2 radical. However, OH reaction of CF3C(O)CH3 is at least 10 times slower than that of OH with CH3C(O)CH3 (J. A. Manion et al., NIST Chemical Kinetics Database: NIST Standard Reference Database 17, Version 7.0 (Web), Release 1.4.3, Data version 2008.12, Natl. Institute of Standards and Technology, http://kinetics.nist.gov). If fluorination slows the OH rate constant of CH3C(O)CH2OH by as large a factor, CF3C(O)CH2OH could have an atmospheric chemical lifetime of several months. Even so, CF3C(O)CH2OH is likely to be soluble on cloud water, so that wet deposition should remove it rapidly from the atmosphere. The product of OH addition at the 2-carbon of BTP, CF3CBrO, is a possible bromine carrier for which we have found no photochemical reaction data. The acetyl halides CH3CXO (X = Cl, Br, or I) are known to react readily with liquid water [e.g., Devore and O'Neal, 1969]; by analogy, we expect CF3CBrO should be water-soluble and will react quickly by hydrolysis on clouds or aerosols to yield HBr + CF3C(O)OH. CF3CBrO could also be expected to photolyze to release Br atom and produce CF3C(O)O2, the same result as from oxidation of the other initial BTP + OH product CF3C(O)CH2OH, but the photolysis rate for this additional loss process is unknown. The CF3C(O)O2 radical is expected to mostly produce CF3OH in the atmosphere, but one channel of its reaction with HO2 can produce CF3C(O)OH. In summary, we believe the most important eventual oxidation products to be bromine atom Br plus CH2O (formaldehyde), CF3C(O)CH2OH, CF3OH, (trifluoromethanol) and CF3COOH (trifluoroacetic acid, also called TFA), in proportions that depend on the speed of CF3C (O)CH2OH wet deposition versus OH reaction. CF3C(O)OH is subject to wet deposition, and can react with OH to produce CF3OH. CF3OH is relatively unreactive in the troposphere, and thus should primarily be removed by rainout. We include a diagram of our estimated BTP + OH oxidation scheme in the auxiliary material for this article. Our WACCM runs for this study assume immediate availability of the atomic Br reaction product from BTP upon reaction with OH and neglect other possible products, so that this study considers only direct climate effects of BTP.
[20] The ODP of short-lived gases depends on the latitude at which the gases are emitted. For BTP, we choose two emission scenarios similar to those used in the recent study of the candidate replacement compound n-propyl bromide [Wuebbles et al., 2011]. Although one anticipated use of BTP is in airline cabins, handheld fire extinguishers are only used rarely in flight, so that all BTP emissions are assumed to occur at ground level in this study. BTP is emitted at a total rate of 5.0 Tg yr1 in both scenarios, an emission rate which is chosen to give a numerically significant O3 impact in the atmospheric model for the ODP derivation. Neither
5 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Figure 4. Zonal mean mixing ratios in ppt of 2-bromo-3,3,3-trifluoropropene (BTP) calculated in the Whole Atmosphere Community Climate Model for 5.0 Tg yr1 total flux from land surfaces. For these and Figures 5-7, white dotted lines identify the minimum and maximum pressure of the tropopause.
(a) Annual average, emission from 30N to 60N. (b) Annual average, emission from 60S to 60N. (c) Emission from 30N to 60N, January. (d) Emission from 30N to 60N, July. (e) Emission from 60S to 60N, January. (f) Emission from 60S to 60N, July.
emissions scenario is intended to represent a projection of anticipated use. In each scenario, the BTP emissions are from land surfaces and are constant with time of year. The first scenario emits 1.15 109 molecules BTP cm2 s1 uniformly at all landmasses from 30N to 60N latitude, a latitude range which includes most of the existing
industrialized nations. The second scenario emits 4.60 108 molecules BTP cm2 s1 uniformly at all land surfaces from 60S to 60N latitude, which represents an extension of BTP use worldwide in a way consistent with past "global" n-propyl bromide emissions scenarios [Wuebbles et al.,
6 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Table 2. Atmospheric Burden in Gigagrams of 2-Bromo-3,3,3Trifluoropropene (BTP) Due to Emissions of 5.0 Tg yr1 BTP Into the Whole Atmosphere Community Climate Model (WACCM)a
Troposphere
Stratosphere and Above
Emissions Scenario
Burden (Gg)
Percent of Total
Burden (Gg)
Percent of Total
30N to 60N
95.95
99.71
0.28
0.29
60S to 60N
58.81
99.67
0.19
0.33
aThe tropopause is defined according to the WMO thermal tropopause definition, the altitude at which the decrease of temperature falls below 2 K km1, constrained to a range of 6 to 18 km.
2001]. Each scenario required eight model years to reach near-steady state for ODP calculations.
[21] A separate WACCM scenario used a surface CFC-11 mixing ratio increase of 80 ppt to obtain a percent change in global O3 burden, 0.372%, and a 0.0304 Tg CFC-11 yr1 increase in CFC-11 chemical removal rate (equal to surface flux) for the denominator of the ODP calculation. This and the unperturbed WACCM run required nine years of model time to achieve a year-to-year near-steady state adequate for the ODP calculation.
4.2. BTP in the Atmosphere
[22] Because BTP reacts quickly in the troposphere, its mixing ratios are expected to decrease rapidly away from the region of emissions along both latitude and altitude. Figure 4 shows the BTP zonal average mixing ratios in ppt for the two emission scenarios. In these and following figures of latitude as horizontal axis and approximate pressure (based on the hybrid sigma-pressure coordinates used in the model) as vertical, the two dotted white lines indicate thermal tropopause minimum and maximum pressures over all longitudes and months incorporated in each figure. The annually averaged zonal mean BTP mixing ratio for 30N to 60N land emissions is shown in Figure 4a, and that for 60S to 60N is shown in Figure 4b. The maximum annual and zonal average BTP mixing ratio, 78.4 ppt for 30N to 60N and 31.3 ppt for 60S to 60N, is at the boundary between 50N and 60N in each of the figures. BTP is indeed mostly constrained to the Northern troposphere in the 30N to 60N emissions case, such that the highest BTP mixing ratio at the tropopause maximum pressure in Figure 4a is 2.5 ppt, located at 70N to 80N. For the 60S to 60N emissions case in Figure 4b, the highest BTP mixing ratio at the maximum tropopause pressure is 1.0 ppt located between 55N and 80N.
[23] Because BTP has an atmospheric lifetime much shorter than one year toward reaction with OH, and OH concentrations increase with increased solar flux, the mixing ratio of BTP can be expected to exhibit seasonal dependence opposite in phase to that of the local OH concentration. Figures 4c and 4d show the January and July zonal average BTP mixing ratios for the 30N to 60N, 5.0 Tg yr1 scenario. Throughout the atmosphere, the July mixing ratios for BTP (Figure 4d) are smaller than those in January (Figure 4c). Similarly, the 60S to 60N, 5.0 Tg yr1 BTP emissions result in overall larger mixing ratios in January (Figure 4e) than in July (Figure 4f), which is because most land in the range of emissions latitudes is in the Northern Hemisphere.
[24] BTP burdens in Gg and the fraction of BTP are given for the model troposphere and for the stratosphere to top of atmosphere in Table 2 for each of the scenarios. For the 30N to 60N scenario, 0.29% of the BTP burden is above the tropopause. For 5.0 Tg yr1 of BTP emitted from land between 60S and 60N, 0.33% of the BTP burden is above the tropopause. The minimal percentage of BTP reaching the stratosphere indicates that most of the BTP is oxidized in the troposphere.
[25] The BTP burden also is the numerator for the atmospheric lifetime calculation; the denominator is the loss rate of BTP due to chemical reaction. Our WACCM runs are set up to include the BTP chemical loss rate in the output, and we have used that information saved from the 5.0 Tg yr1 scenarios to calculate BTP lifetimes shown in Table 3. The 7.0 day lifetime obtained for the 30N to 60N land emissions scenario and the 4.3 day lifetime obtained for the 60S to 60N land emissions scenario are among the shortest we have ever found in an ODP or GWP evaluation for a candidate replacement compound. We also separately tested BTP lifetimes in 2.0 Tg yr1 BTP emission scenarios (not shown in Table 3), and those were higher by less than one percent than the lifetimes obtained from the 5.0 Tg yr1 emissions scenarios, suggesting that the model response to BTP should scale linearly with flux.
4.3. Atmospheric Bromine From BTP
[26] BTP mixing ratios in the model stratosphere (altitudes above the minimum thermal tropopause pressure) do not exceed 1 ppt in either 5.0 Tg yr1 emissions scenario, which suggests that no more than this mixing ratio of inorganic bromine (Bry) can result from direct BTP oxidation in the stratosphere. Figure 5 shows zonal mean Bry perturbations resulting from the 5.0 Tg yr1 BTP emissions scenarios. The annual average, zonal mean Bry perturbation from the 30N to 60N land emissions scenario in Figure 5a shows a nonsymmetric profile, with Bry increases in the Southern and tropical troposphere limited to less than 0.1 ppt. The maximum Bry increase appears as a region starting from the emissions latitudes and reaching to the Northern upper troposphere and lower stratosphere (UTLS). Most of the stratosphere for this scenario has Bry mixing ratio increases from BTP no greater than 1.0 ppt. The 60S to 60N emissions scenario annual and zonal mean in Figure 5b also shows little sign of stratospheric Bry increases from BTP larger than 1 ppt. Within the troposphere, however, the figure is more symmetric with a second region of Bry perturbation in the Southern Hemisphere.
[27] Figure 5 suggests that there will be only a tiny fraction of Bry reaching the stratosphere either directly or indirectly from BTP oxidation. Table 4 lists the Bry burden changes in Gg Br arising from BTP in the model troposphere
Table 3. Atmospheric Lifetimes and Ozone Depletion Potentials (ODPs) of 2-Bromo-3,3,3-Trifluoropropene for 5.0 Tg yr1 Emissions From Land Surfaces in WACCM
Emissions Scenario
Atmospheric Lifetime (days)
ODP
30N to 60N 60S to 60N
7.0
0.0028
4.3
0.0052
7 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Figure 5. Zonal mean changes in mixing ratio (ppt) of inorganic bromine species (Bry) resulting from 5.0 Tg yr1 flux of BTP from land surfaces. (a) Annual average, emission from 30N to 60N. (b) Annual average, emission from 60S to 60N. (c) Emission from 30N to 60N, January. (d) Emission from 30N to 60N, July. (e) Emission from 60S to 60N, January. (f) Emission from 60S to 60N, July.
and above the thermal tropopause for the two emissions
scenarios. For the 30N to 60N, 5.0 ppt scenario, only 3.1% of the Bry perturbation is above the tropopause; for the 60S to 60N, 5.0 ppt scenario, 7.8% of the Bry perturbation is above the tropopause, largely because of the strong transport in the Hadley circulation in the tropics.
[28] The pattern in the annual average, zonal mean Bry perturbations of Figures 5a and 5b could be interpreted to
Table 4. Derived Burden Perturbation of Inorganic Bromine Species (Bry) in Gigagrams Br Due to 5.0 Tg yr1 Surface Emissions of 2-Bromo-3,3,3-Trifluoropropene in WACCM
Emissions Scenario
Troposphere
Stratosphere and Above
Stratosphere Percent of Total
30N-60N
17.93
0.58
3.1
60S-60N
18.68
1.58
7.8
8 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Figure 6. Zonal, monthly mean 2-bromo-3,3,3-trifluoropropene (BTP) chemical loss rates (ppt day1) for 5.0 Tg yr1 flux of BTP from land surfaces at latitudes 30N to 60N. (a) January. (b) July.
indicate that BTP lifted to the UTLS is oxidized there. The January Bry perturbation for the 30N to 60N, 5.0 Tg yr1 land emissions scenario is shown in Figure 5c, and that for July in Figure 5d. July Bry perturbations in the Northern UTLS peak above 5 ppt, markedly larger than the 1 ppt maximum for that region in January, though no more than 1 ppt BTP reaches the UTLS in July (Figure 4d). The January (Figure 5e) and July (Figure 5f) Bry perturbation plots for the 60S to 60N scenario show that the BTP reaches the UTLS in whichever hemisphere is in summer. While these figures do not definitely distinguish transport as BTP followed by oxidation to Bry from BTP oxidation to Bry near the boundary layer followed by transport upward, they indicate that WACCM convection is involved in sending Bry from BTP to the stratosphere. The zonal-mean chemical loss rate of BTP in ppt day1 for the 30N to 60N, 5.0 Tg yr1 land emissions scenario is shown in Figure 6a for January and in Figure 6b for July. In both months, BTP removal rates of tens of ppt day1 occur near the emissions region, and the removal rate decreases with altitude. However, the July BTP removal rate (Figure 6b) decreases less strongly with increased altitude than does that in January (Figure 6a), suggesting contributions from convection of BTP and of Bry produced from BTP near the surface to the Bry perturbation reaching the UTLS.
4.4. Ozone Effects of BTP
[29] Because of the limited fraction of Bry reaching the stratosphere, much of the BTP effect on O3 is derived to occur in the troposphere as shown below. The global and annual mean percentage change of O3 in the model due to 5.0 Tg yr1 BTP is shown in Figure 7. In both the 30N to 60N (Figure 7a) and 60S to 60N (Figure 7b) scenarios, the largest percentage decreases are in the troposphere. The 30N to 60N land emissions scenario in Figure 7a shows small O3 loss percentages in the stratosphere, with the peak values of about 0.5% near the tropopause outside the tropics. The greater fraction of stratospheric Bry (Table 4) for the 60S to 60N scenario leads to O3 losses in Figure 7b of up to 1.0% in the Northern lower stratosphere and nearly 2% in the Southern lower stratosphere. Seasonal dependence of O3 percentage loss due to BTP (Figures 7c and 7d for January and July, 30N to 60N; Figures 7e and 7f for
January and July, 60S to 60N) is limited in the stratosphere, maximized in the summer troposphere for either hemisphere, and minimal in the Southern troposphere for the 30N to 60N scenario.
[30] Though O3 percent losses are largest in the troposphere, those are relative to the small O3 mixing ratios there compared to the stratosphere. The integrated total column of O3 from top of atmosphere determines the extent to which harmful ultraviolet is prevented from reaching the surface, and is therefore the criterion for determining ODP. Changes in Tg of O3 burden for the WACCM troposphere and for the stratosphere and above resulting from 5.0 Tg yr1 of BTP land emissions are listed in Table 5. Despite the low fractions of Bry perturbation reaching the stratosphere given in Table 4, the O3 burden loss above the tropopause is 27% for the 30N to 60N scenario and 46%, or nearly half, for the 60S to 60N scenario.
[31] The ODP is obtained through the equation
ODPX DO3DbOur3dbeunrdCeFnCX 11 CFCXflu1x1 flux 3
The change in total O3 burden due to 80 ppt surface mixing ratio increase of CFC-11 in WACCM, relative to the reference atmosphere burden of 3139.0 Tg, was given above as 0.372%, and the corresponding CFC-11 flux increase is 0.0304 Tg yr1. 5.0 Tg yr1 of BTP is applied directly in equation (3), and the percent change in total O3 burden due to BTP is 0.170% for 30N to 60N, 5.0 Tg yr1 and 0.315% for 60S to 60N, 5.0 Tg yr1. These values provide the ODPs in Table 3. The 30N to 60N BTP ODP of 0.0028 is the smallest we have obtained for a brominated species thus far. Even the 60S to 60N ODP of 0.0052 is a factor of nearly 20 times less than the 0.1 limit called for in the U.S. Clean Air Act. A full error analysis of WACCM is not practical at this time. The potential bromine-containing BTP oxidation product CF3CBrO noted in Section 4.1 is likely a major part of the uncertainty in the ODP. However, most BTP oxidation to products including CF3CBrO occurs in the lower to middle troposphere, where clouds are often present and highly soluble species will be removed rapidly in rainout. CF3CBrO is likely to dissolve readily and rapidly
9 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Figure 7. Zonal mean percent changes in mixing ratio of ozone (O3) resulting from 5.0 Tg yr1 flux of BTP from land surfaces. (a) Annual average, emission from 30N to 60N. (b) Annual average, emission from 60S to 60N. (c) Emission from 30N to 60N, January. (d) Emission from 30N to 60N, July. (e) Emission from 60S to 60N, January. (f) Emission from 60S to 60N, July.
hydrolyze on cloud water, which could lower the actual BTP ODP somewhat compared to this study. However, even if CF3CBrO were as long-lived as BTP in the troposphere, it would increase BTP O3 loss and ODP by up to a factor of two compared to our model results. Such an increase in ODP compared to the current study would still indicate minimal O3 loss from BTP.
Table 5. Derived Changes in Ozone (O3) Burdens in Teragrams Due to 5.0 Tg yr1 Surface Emissions of 2-Bromo-3,3,3-Trifluoropropene in WACCM
Emissions Scenario
Troposphere
Stratosphere and Above
Stratosphere Percent of Total
30N-60N
3.92
1.43
27
60S-60N
5.39
4.51
46
10 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
[32] Brioude et al. [2010] recently studied the variation of ODP with emission region and season for several short-lived gases, including BTP, in a Lagrangian model driven by mid2000s meteorological data from the European Centre for Medium-Range Weather Forecasts. The general conclusions of Brioude et al. [2010], including its auxiliary material, were that ODP increases with decreased emissions latitude and with summer compared to winter. Our higher ODPs for land emissions from 60S to 60N than from 30N to 60N are consistent with the Brioude et al. latitude dependence, and our higher values of UTLS BTP, Bry perturbation and % change of O3 in the 30N to 60N scenario for July (Figures 4d, 5d, and 7d, compared to Figures 4c, 5c, and 7c for January) are consistent with the summer ODP increase reported by Brioude et al. While the emission regions used by Brioude et al. [2010] are not identical to those in this study, their annual average ODP reported for midlatitude North America, 0.0035, is slightly higher than our 30N to 60N value of 0.0028. The annually averaged, global BTP lifetime of 4.4 days reported for the midlatitude North America emission region in Brioude et al. is shorter than our 7.0 day lifetime for the 30N to 60N land emissions scenario and more consistent with 4.3 days which we found for the 60S to 60N scenario (Table 3), though the Brioude et al. emission region seems to range from about 23N to 55N. Differences between Brioude et al. and our study may arise from differences in BTP emissions distributions or from dynamic parameterizations in our model versus those from Brioude et al., but may also result from the details of BTP chemical loss processes in the two models: our model evaluates OH loss of BTP within its chemical solver with an external time step of 30 min, while the Brioude et al. [2010] study calculates BTP loss from daily averaged OH concentrations derived from MOZART 4.
4.5. Climate Effects of BTP
[33] We apply the University of Illinois at UrbanaChampaign radiative transfer model (UIUC RTM) to obtain GWPs from the WACCM outputs. The UIUC RTM calculates fluxes of solar and of terrestrial radiation across the tropopause for input atmospheres supplied by models. Earlier versions of the UIUC RTM have been applied to several studies of climate effects and GWPs [Jain et al., 2000; Naik et al., 2000; Youn et al., 2009]. The UIUC RTM was originally based on the radiative transfer models of the Community Climate Model (CCM) [Briegleb, 1992a, 1992b]. The solar model includes 18 spectral bins from 0.2 to 5.0 microns and incorporates absorption by H2O, O3, O2, CO2, clouds, and the surface along with scattering processes by clouds, gas-phase molecules, and the surface. Terrestrial radiation is calculated with a narrow band model of absorptivity and emissivity over the wave number range 0 to 3000 cm1 at resolutions of 5 cm1 for CO2, O3, CH4, and N2O and of 10 cm1 for H2O, CFC-11, and CFC-12. The infrared absorption parameters for these gases are based on the HITRAN 2004 database [Rothman et al., 2005]. The BTP infrared cross sections measured in this study are regridded to 10 cm1 resolution for the UIUC RTM. Observation-based surface albedo and surface emissivity are used with clouds based on the International Satellite Cloud Climatology Project.
[34] To obtain GWP for BTP, we must obtain the change in net downward radiative flux across the tropopause, or tropopause radiative forcing (RF), due to the BTP distribution we have calculated in our model. We apply the UIUC RTM with the BTP infrared cross sections measured above as shown in Figure 3 to our WACCM outputs. The RTM obtains tropopause RF for the reference run (without BTP) and for the two BTP emissions scenarios, and the difference in tropopause RF is used in the further calculations. The tropopause RF change due to BTP in the 30N to 60N land emissions scenario is 0.000112 W m2, and that for the 60S to 60N land emissions scenario is 0.000102 W m2. It must be noted that these are instantaneous RF perturbations; the magnitudes of these perturbations, when compared to net terrestrial radiation fluxes across the tropopause that average some 220 W m2, are small enough that stratospheric temperature adjustment is unlikely to be significant for BTP.
[35] The GWP is calculated as the ratio of tropopause RF integrated over time to a time horizon tH (also known as absolute GWP or AGWP) resulting from a pulse release compound to that from a pulse release of the same mass of CO2. We assume that BTP would be removed with a singleexponential decay at its atmospheric lifetime t if its emissions stopped, so that the AGWP for BTP is given by
AGWP F0th1 exp ttH i 4
where the radiative forcing per kg of BTP F0 is obtained by dividing the tropopause RF perturbation by the annual flux of 5.0 Tg yr1, or 5.0 109 kg yr1. tH frequently is specified as 20, 100, or 500 years for long-lived gases. The lifetime of BTP in the 30N to 60N emissions scenario is 7.0 days and that in the 60S to 60N emissions scenario is 4.3 days (Table 3), so that the exponential term in equation (4) is effectively zero for even tH = 20 years. Thus, the AGWPs for BTP at 20, 100, or 500 years are equal to radiative forcing per kg BTP times lifetime:
30N 60N AGWP 4:3 1016 W m2 yr kg BTP1
60S 60N AGWP 2:4 1016 W m2 yr kg BTP1:
[36] The IPCC 2006 report [Forster et al., 2007] Section 2.10.2 supplies AGWPs for CO2 at tH = 20, 100, and 500 years:
20 year AGWP 2:47 1014 W m2 yr kg CO21
100 year AGWP 8:69 1014 W m2 yr kg CO21
500 year AGWP 28:6 1014 W m2 yr kg CO21
Dividing these into the BTP AGWP provides the GWP values shown on the upper lines in Table 6. These low GWPs suggest that the climate effect of BTP will be markedly smaller than that of an equivalent mass of CO2 from, for example, fossil fuel combustion.
[37] We are unaware of other reports of GWP for BTP, but the GWPs reported in this study are considerably lower than
11 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Table 6. Global Warming Potentials (GWPs) for 2-Bromo-3,3,3Trifluoropropene (BTP) for Time Horizons tH of 20, 100, and 500 Years Derived From WACCM and the UIUC RTMa
Emissions Scenario
tH = 20 yr
tH = 100 yr
tH = 500 yr
30N to 60N 60S to 60N Well-mixed estimate
0.017 0.0097
2.1
0.0050 0.0028 0.59
0.0015 0.00084
0.18
aBTP GWPs based on a tropospherically well-mixed approximation are also shown for comparison.
GWPs reported for many other short-lived halocarbons. Such GWPs are usually based on a well-mixed approximation in which the halocarbon is assumed to be thoroughly mixed in the troposphere. However, this approximation is not realistic for compounds with atmospheric lifetimes markedly shorter than one year such as BTP. Figure 4 illustrates that BTP mixing ratios derived in WACCM do not satisfy the well-mixed approximation. BTP mixing ratios in the upper troposphere and lower stratosphere where most of its radiative forcing effect is expected are markedly lower than near the surface emissions region. We show well-mixed approximation GWPs for BTP in Table 6 for comparison to our model-derived GWPs. The well-mixed BTP GWPs are based on a radiative efficiency of 0.193 W m2 ppb1 obtained from the IR cross sections in this report by the method of Pinnock et al. [1995] and a well-mixed approximation atmospheric lifetime of 3.0 days estimated by scaling the OH lifetime of methyl chloroform [Prinn et al., 2001; Spivakovsky et al., 2000] by the ratio of OH rate constants at an average tropospheric temperature of 270 K. These wellmixed approximation GWPs are over 100 times larger than our model-derived GWPs. Many GWPs previously reported for short-lived compounds are upper limits based on the well-mixed approximation, and it should be emphasized that model-based profiles for very short-lived compounds could result in markedly smaller GWPs than have been previously reported.
5. Conclusions
[38] The OH reaction rate constant for temperatures from 220 to 370 K and the IR absorption cross sections of 2-bromo-3,3,3-trifluoropropene (BTP) were measured in this study. The measured OH reaction rate constant can be represented as kBTP(T) = 4.85 1013 (T/298)0.92 exp(+613/T) cm3 molec1 s1 over the entire T range of measurements. For atmospheric modeling, we recommend the Arrhenius expression kBTP(T) = 1.05 1012 exp(+381/T) cm3 molec1 s1. We have also evaluated the potential effects of BTP on ozone and climate in the form of ODPs and GWPs for assumed land-based emissions at latitudes 30N to 60N and at latitudes 60S to 60N with a rate of 5.0 Tg yr1 (selected to provide an ozone burden decrease of at least 0.1%, not as a reflection of anticipated BTP use) with the Whole Atmosphere Community Climate Model. BTP is among the shortest-lived of bromocarbons evaluated thus far, with a lifetime of 7.0 days for 30N to 60N emissions and of 4.3 days for 60S to 60N emissions. The Ozone Depletion Potential (ODP) obtained for the 30N to 60N emissions scenario is 0.0028, one of the lowest of any candidate
replacement compound we have studied, and that and the 0.0052 ODP for 60S to 60N emissions indicate that BTP would have an insignificant effect on ozone. The direct Global Warming Potential (GWP) of BTP at a 100-year time horizon is estimated to be 0.0050 for the 30N to 60N emissions scenario and 0.0028 for the 60S to 60N emissions scenario. These small ODP and GWP values suggest that the environmental effects of BTP will be minimal for any reasonable level of emission expected for this candidate fireextinguishing replacement compound.
[39] Acknowledgments. This study was supported by the Boeing
Company and by the Upper Atmosphere Research Program of the National Aeronautics and Space Administration. V.L.O. also acknowledges the support of NATO CLG Program, grant ESP.EAP.CLG.983035. We thank D. Kinnison and F. Vitt of the National Center for Atmospheric Research for WACCM 3.5.48. We also thank Michael Danilin for comments and suggestions on an earlier draft, and we thank three anonymous reviewers for their comments. Certain commercial equipment, instruments, or materials are identified in this article in order to adequately specify the experimental procedure. Such identification does not imply recognition or endorsement by the National Institute of Standards and Technology or by the University of Illinois, nor does it imply that the material or equipment identified are necessarily the best available for the purpose.
References
Atkinson, R., D. L. Baulch, R. A. Cox, J. N. Crowley, R. F. Hampson, R. G. Hynes, M. E. Jenkin, M. J. Rossi, and J. Troe (2006), Evaluated kinetic and photochemical data for atmospheric chemistry: Volume II-Gas phase reactions of organic species, Atmos. Chem. Phys., 6, 3625-4055.
Briegleb, B. P. (1992a), Delta-Eddington approximation for solar radiation in the NCAR Community Climate Model, J. Geophys. Res., 97, 7603-7612.
Briegleb, B. P. (1992b), Longwave band model for thermal radiation in climate studies, J. Geophys. Res., 97, 11,475-11,485.
Brioude, J., R. W. Portmann, J. S. Daniel, O. R. Cooper, G. J. Frost, K. H. Rosenlof, C. Granier, A. R. Ravishankara, S. A. Montzka, and A. Stohl (2010), Variations in ozone depletion potentials of very short-lived substances with season and emission region, Geophys. Res. Lett., 37, L19804, doi:10.1029/2010GL044856.
Carr, S., D. E. Shallcross, C. Canosa-Mas, J. C. Wenger, H. W. Sidebottom, J. J. Treacy, and R. P. Wayne (2003), A kinetic and mechanistic study of the gas-phase reactions of OH radicals and Cl atoms with some halogenated acetones and their atmospheric implications, Phys. Chem. Chem. Phys., 5, 3874-3883, doi:10.1039/b304298g.
Devore, J. A., and H. E. O'Neal (1969), Heats of formation of the acetyl halides and of the acetyl radical, J. Phys. Chem., 73, 2644-2648, doi:10.1021/j100842a031.
Emmons, L. K., et al. (2010), Description and evaluation of the Model for Ozone and Related chemical Tracers, version 4 (MOZART-4), Geosci. Model Dev., 3, 43-67, doi:10.5194/gmd-3-43-2010.
Eyring, V., T. G. Shepherd, and D. W. Waugh (Eds.) (2010), Stratospheric processes and their role in climate report on the evaluation of chemistryclimate models (SPARC CCMVal), WMO/TD 1526, World Meteorol. Org., Geneva, Switzerland. [Available at http://www.atmosp.physics. utoronto.ca/SPARC.]
Forster, P., et al. (2007), Changes in atmospheric constituents and in radiative forcing, in Climate Change 2007: The Physical Science Basis, Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, edited by S. Solomon et al., pp. 129-234, Cambridge Univ. Press, New York.
Gettelman, A., et al. (2010), Upper troposphere and lower stratosphere, in Stratospheric Processes And Their Role in Climate Report on the Evaluation of Chemistry-Climate Models (SPARC CCMVal), edited by V. Eyring, T. G. Shepherd, and D. W. Waugh, pp. 253-304, WMO/TD 1526, World Meteorol. Org., Geneva, Switzerland. [Available at http://www. atmosp.physics.utoronto.ca/SPARC.]
Horowitz, L. W., et al. (2003), A global simulation of tropospheric ozone and related tracers: Description and evaluation of MOZART, version 2, J. Geophys. Res., 108(D24), 4784, doi:10.1029/2002JD002853.
Horowitz, L. W., A. M. Fiore, G. P. Milly, R. C. Cohen, A. Perring, P. J. Wooldridge, P. G. Hess, L. K. Emmons, and J.-F. Lamarque (2007), Observational constraints on the chemistry of isoprene nitrates over the eastern United States, J. Geophys. Res., 112, D12S08, doi:10.1029/ 2006JD007747.
12 of 13
D24307
PATTEN ET AL.: BTP k(OH), IR ABSORPTION, ODP, AND GWP
D24307
Jain, A. K., B. P. Briegleb, K. Minschwaner, and D. J. Wuebbles (2000), Radiative forcings and global warning potentials of 39 greenhouse gases, J. Geophys. Res., 105, 20,773-20,790, doi:10.1029/2000JD900241.
Kinnison, D. E., et al. (2007), Sensitivity of chemical tracers to meteorological parameters in the MOZART-3 chemical transport model, J. Geophys. Res., 112, D20302, doi:10.1029/2006JD007879.
Ko, M. K. W., et al. (2003), Very short-lived halogen and sulfur substances, in Scientific Assessment of Ozone Depletion: 2002, Global Ozone Res. Monit. Project Rep., 47, pp. 2.1-2.57, World Meteorol. Org., Geneva, Switzerland.
Kurylo, M. J., K. D. Cornett, and J. L. Murphy (1982), The temperature dependence of the rate constant for the reaction of hydroxyl radicals with nitric acid, J. Geophys. Res., 87, 3081-3085, doi:10.1029/JC087iC04p03081.
Naik, V., A. K. Jain, K. O. Patten, and D. J. Wuebbles (2000), Consistent sets of atmospheric lifetimes and radiative forcings on climate for CFC replacements: HCFCs and HFCs, J. Geophys. Res., 105, 6903-6914, doi:10.1029/1999JD901128.
Orkin, V. L., R. E. Huie, and M. J. Kurylo (1996), Atmospheric lifetimes of HFC-143a and HFC-245fa: Flash photolysis resonance fluorescence measurements of the OH reaction rate constants, J. Phys. Chem., 100, 8907-8912, doi:10.1021/jp9601882.
Orkin, V. L., R. E. Huie, and M. J. Kurylo (1997), Rate constants for the reactions of OH with HFC-245cb (CH3CF2CF3) and some fluoroalkenes (CH2CHCF3, CH2CFCF3, CF2CFCF3, and CF2CF2), J. Phys. Chem. A, 101, 9118-9124, doi:10.1021/jp971994r.
Orkin, V. L., F. Louis, R. E. Huie, and M. J. Kurylo (2002), Photochemistry of bromine-containing fluorinated alkenes: Reactivity toward OH and UV spectra, J. Phys. Chem. A, 106, 10,195-10,199, doi:10.1021/jp014436s.
Orkin, V. L., L. E. Martynova, and A. N. Ilichev (2010), High-accuracy measurements of OH reaction rate constants and IR absorption spectra: CH2 = CF-CF3 and trans-CHF = CH-CF3, J. Phys. Chem. A, 114, 5967-5979, doi:10.1021/jp9092817.
Orkin, V. L., V. G. Khamaganov, L. E. Martynova, and M. J. Kurylo (2011), High accuracy measurements of OH reaction rate constants and IR and UV absorption spectra: Ethanol and partially fluorinated ethyl alcohols, J. Phys. Chem. A, 115, 8656-8668, doi:10.1021/jp202099t.
Pinnock, S., M. D. Hurley, K. P. Shine, T. J. Wallington, and T. J. Smyth (1995), Radiative forcing of climate by hydrochlorofluorocarbons and hydrofluorocarbons, J. Geophys. Res., 100, 23,227-23,238, doi:10.1029/ 95JD02323.
Prinn, R. G., et al. (2001), Evidence for substantial variations of atmospheric hydroxy radicals in the past two decades, Science, 292, 1882-1888, doi:10.1126/science.1058673.
Rajakumar, B., R. W. Portmann, J. B. Burkholder, and A. R. Ravishankara (2006), Rate coefficients for the reactions of OH with CF3CH2CH3 (HFC263fb), CF3CHFCH2F (HFC-245eb), and CHF2CHFCHF2 (HFC-245ea) between 238 and 375 K, J Phys. Chem. A, 110, 6724-6731, doi:10.1021/ jp056248y.
Rothman, L. S., et al. (2005), The HITRAN 2004 molecular spectroscopic database, J. Quant. Spectrosc. Radial. Transfer, 96, 139-204, doi:10.1016/ j jqsrt.2004.10.008.
Sander, S. P., et al. (2006), Chemical Kinetics and Photochemical Data for Use in Atmospheric Studies, Evaluation Number 15, JPL Publ. 06-2, Jet Propul. Lab., Calif. Inst. of Technol., Pasadena, Calif. [Available at http://jpldataeval.jpl.nasa.gov.]
Schumann, U., and H. Huntrieser (2007), The global lightning-induced nitrogen oxides source, Atmos. Chem. Phys., 7, 3823-3907, doi:10.5194/ acp-7-3823-2007.
Spivakovsky, C. M., et al. (2000), Three-dimensional climatological distribution of tropospheric OH: Update and evaluation, J. Geophys. Res., 105, 8931-8980, doi:10.1029/1999JD901006.
Wuebbles, D. J., K. O. Patten, M. T. Johnson, and R. Kotamarthi (2001), New methodology for ozone depletion potentials of short-lived compounds: N-Propyl bromide as an example, J. Geophys. Res., 106, 14,551-14,571, doi:10.1029/2001JD900008.
Wuebbles, D. J., K. O. Patten, D. Wang, D. Youn, M. Martinez-Aviles, and J. S. Francisco (2011), Three-dimensional model evaluation of the ozone depletion potentials for n-propyl bromide, trichloroethylene and perchloroethylene,Atmos. Chem. Phys., 11, 2371-2380, doi:10.5194/acp-11-2371-2011.
Youn, D., K. O. Patten, J.-T. Lin, and D. J. Wuebbles (2009), Explicit calculation of indirect global warming potentials for halons using atmospheric models, Atmos. Chem. Phys., 9, 8719-8733, doi:10.5194/acp-9-8719-2009.
S. L. Baughcum, Boeing Company, PO Box 3707, MC OR-MT, Seattle, WA 98124, USA.
V. G. Khamaganov and V. L. Orkin, National Institute of Standards and Technology, 100 Bureau Dr., Stop 6320, Gaithersburg, MD 20899-6320, USA.
K. O. Patten and D. J. Wuebbles, Department of Atmospheric Sciences, University of Illinois at Urbana-Champaign, 105 S. Gregory St., Urbana, IL 61801-3070, USA. (M@atmos.uiuc.edu)
13 of 13