Chemical Physics Letters 439 (2007) 18–22 www.elsevier.com/locate/cplett Atmospheric chemistry of CF3CF@CH2: Kinetics and mechanisms of gas-phase reactions with Cl atoms, OH radicals, and O3 O.J. Nielsen a,*, M.S. Javadi a, M.P. Sulbaek Andersen a, M.D. Hurley b, T.J. Wallington b,*, R. Singh c a Department of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark b Physical and Environmental Sciences Department, Ford Motor Company, Mail Drop SRL-3083, Dearborn, MI 48121, USA c Honeywell International Inc., 101 Columbia Road, Morristown, NJ 07962, USA Received 31 January 2007; in final form 6 March 2007 Available online 21 March 2007 Abstract À11 Long path length FTIR-smog chamber techniques were used to determine k(Cl + CF3CF@CH2) = (7.03 ± 0.59) · 10 , À12 À21 3 À1 À1 k(OH + CF3CF@CH2) = (1.05 ± 0.17) · 10 , and k(O3 +CF3CF@CH2) = (2.77 ± 0.21) · 10 cm molecule s in 700 Torr of N2,N2/O2, or air diluent at 296 K. CF3CF@CH2 has an atmospheric lifetime of approximately 11 days and a global warming potential (100 yr time horizon) of four. CF3CF@CH2 has a negligible global warming potential and will not make any significant contribution to radiative forcing of climate change. Ó 2007 Elsevier B.V. All rights reserved. 1. Introduction tion with ozone and (iv) atmospheric implications. Results are reported herein. Recognition of the adverse environmental impact of chlorofluorocarbon (CFC) release into the atmosphere 2. Experimental [1,2] has led to an international effort to replace these com- pounds with environmentally acceptable alternatives. Experiments were performed in a 140-liter Pyrex reactor Unsaturated fluorinated hydrocarbons are a class of com- interfaced to a Mattson Sirus 100 FTIR spectrometer [3]. pounds which have been developed to replace CFCs and The reactor was surrounded by 22 fluorescent blacklamps saturated hydrofluorocarbons in air conditioning units. (GE F15T8-BL), which were used to photochemically initi- Prior to their large-scale industrial use an assessment of ate the experiments. Chlorine atoms were produced by the atmospheric chemistry, and hence environmental photolysis of molecular chlorine. impact, of these compounds is needed. To address this need Cl2 þ hv ! Cl þ Cl ð1Þ the atmospheric chemistry of CF3CF@CH2 was investi- gated. Smog chamber/FTIR techniques were used to deter- OH radicals were produced by photolysis of CH3ONO in mine the following properties for this compound: (i) the presence of NO in air. kinetics of its reaction with chlorine atoms, (ii) kinetics of its reaction with hydroxyl radicals, (iii) kinetics of its reac- CH3ONO þ hv ! CH3O þ NO ð2Þ CH3O þ O2 ! HO2 þ HCHO ð3Þ HO2 þ NO ! OH þ NO2 ð4Þ * Corresponding authors. E-mail addresses: [email protected] (O.J. Nielsen), [email protected] In the relative rate experiments the following reactions take (T.J. Wallington). place. 0009-2614/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2007.03.053 O.J. Nielsen et al. / Chemical Physics Letters 439 (2007) 18–22 19 Cl þ Reactant ! products ð5Þ 3.5 Cl þ Reference ! products ð6Þ 3.0 OH þ Reactant ! products ð7Þ ) C H t ] 2 2 OH þ Reference ! products ð8Þ 2 2.5 It can be shown that CF=CH 3 ½Reactant k ½Reference Ln t0 Reactant Ln t0 9 2.0 ¼ ð Þ /[CF ½Reactant kReference ½Reference t0 t t ] 2 where [Reactant]t0, [Reactant]t, [Reference]t0, and [Refer- 1.5 C H ence]t are the concentrations of reactant and reference at 2 4 CF=CH times t0 and t, and kReactant and kReference are the rate 3 1.0 constants for the reactant and the reference. Plots of Ln[Reactant]t0/[Reactant]t) vs. Ln([Reference]t0/[Refer- Ln ([CF 0.5 ence]t) should be linear, pass through the origin, and have a slope of kReactant/kReference. The kinetics of the O3 reac- tion were studied using an absolute rate method in which 0.0 0.00.51.01.52.02.5 the pseudo first-order loss of CF3CF@CH2 was measured Ln ([Reference]t0/[Reference]t) in the presence of excess O3. O was produced from O via silent electrical discharge 3 2 Fig. 1. Decay of CF3CF@CH2 vs. C2H4 and C2H2 in the presence of Cl using a commercial O3 ozonizer. CH3ONO was synthesized atoms in 700 Torr of either air (open symbols) or N2 (filled symbols) at by the drop wise addition of concentrated sulfuric acid to a 296 ± 2 K. saturated solution of NaNO2 in methanol. Other reagents were obtained from commercial sources. Experiments were to the data in Fig. 1 gives k10/k11 = 0.76 ± 0.04 and k10/ conducted in 700 Torr total pressure of N2,orN2/O2 dilu- k12 = 1.38 ± 0.06. À11 ent at 296 ± 1 K. Using k11 = (9.29 ± 0.51) · 10 [4] and k12 = (5.07 ± À11 Concentrations of reactants and products were moni- 0.34) · 10 [4] (700 Torr, 295 K) gives k10 = (7.06 ± tored by FTIR spectroscopy. IR spectra were derived from 0.54) · 10À11 and (7.00 ±0.56) · 10À11 cm3 moleculeÀ1 sÀ1. 32 coadded interferograms with a spectral resolution of We choose to cite a final value which is the average of the À1 0.25 cm and an analytical path length of 27.1 m. To individual determinations together with error limits which check for unwanted loss of reactants and reference com- encompass the extremes of the determinations, hence À11 3 À1 À1 pounds via heterogeneous reactions, reaction mixtures k10 = (7.03 ± 0.59) · 10 cm molecule s . While there were left to stand in the chamber for 60 min. There was have been no previous studies of k10, we can compare our À10 no observable (<2%) loss of any of the reactants or prod- result with k(Cl + CH3CH@CH2) = 2.4 · 10 [5], k(Cl + À11 ucts in the present work. Unless stated otherwise, quoted CF3CH@CH2) = (9.07 ± 1.08) · 10 [6], k(Cl + À11 uncertainties are two standard deviations from least C4F9CH@CH2) = (8.9 ± 1.0) · 10 [7], k(Cl+C6F13CH@ À11 squares regressions. CH2) = (9.1 ± 1.0) · 10 [7],andk(Cl + CF3CF@ À11 3 À1 À1 CF2) = (2.7 ± 0.3) · 10 cm molecule s [8]. The 3. Results and discussion reaction of Cl atoms with propene proceeds via electro- philic addition to the >C@C< double bond. The presence @ 3.1. Kinetics of the Cl + CF3CF CH2 reaction of electron withdrawing fluorine substituents is expected to lead to decreased reactivity with Cl atoms. Consistent The rate of reaction (10) was measured relative to reac- with expectations, the reactivity of CF3CF@CH2 reported tions (11) and (12): here lies between those of CF3CH@CH2 and CF3CF@CF2 reported previously. Cl þ CF3CF@CH2 ! products ð10Þ Cl þ C2H4 ! products ð11Þ 3.2. Kinetics of the OH + CF3CF@CH2 reaction Cl þ C2H2 ! products ð12Þ Reaction mixtures consisted of 19.1–26.2 mTorr of The rate of reaction (13) was measured relative to reac- tions (14) and (15): CF3CF@CH2, 104–133 mTorr Cl2, and either 4.33– 30.2 mTorr C2H4, or 2.35–8.5 mTorr C2H2, in 700 Torr OH þ CF3CF@CH2 ! products ð13Þ of air, or N , diluent. The observed loss of CF CF@CH 2 3 2 OH þ C H ! products ð14Þ vs. those of the reference compounds is plotted in Fig. 1. 2 4 As seen from Fig. 1, there was no discernable difference be- OH þ C2H2 ! products ð15Þ tween the results obtained in N2, or air, diluent. A linear Initial reaction mixtures consisted of 17.6–18.1 mTorr of least squares fit (unweighted, not forced through the origin) CF3CF@CH2, 110–200 mTorr CH3ONO, and 3.38 mTorr 20 O.J. Nielsen et al. / Chemical Physics Letters 439 (2007) 18–22 0.5 reactivity of CF3CF@CF2 appears to be anomalously high. A computational study of the reaction of OH radicals with CF3CF@CF2 would be of interest to shed further light on ) C H t 2 2 ] the mechanism of these reactions. 2 0.4 3.3. Absolute Rate Study of k(O3 +CF3CF@CH2) CF=CH 3 0.3 The kinetics of reaction (16) were studied by observing /[CF @ t0 the decay of CF CF CH when exposed to ozone in the ] 3 2 2 reaction chamber. Reaction mixtures consisted of 14– 0.2 28 mTorr CF3CF@CH2, 30–46 mTorr cyclohexane, and CF=CH 180–1890 mTorr O3 in 700 Torr of air diluent. Cyclohexane 3 was added to avoid potential problems associated with the C2H4 0.1 loss of CF3CF@CH2 via reaction with OH radicals formed Ln ([CF in reaction (16). Variation of the [cyclohexane]/ [CF3CF@CH2] ratio over the range 1–3 had no discernable @ 0.0 effect on the observed decay of CF3CF CH2 suggesting 0.0 0.5 1.0 1.5 2.0 that loss via reaction with OH radicals is not a significant @ Ln ([Reference] /[Reference] ) complication. The loss of CF3CF CH2 followed pseudo t0 t first-order kinetics in all experiments (see insert in Fig. 3). Fig. 2. Decay of CF3CF@CH2 vs. C2H4 and C2H2 in the presence of OH Fig. 3 shows a plot of the pseudo first-order loss of radicals in 700 Torr of air at 296 ± 2 K. CF3CF@CH2 vs. O3 concentration. The line through the À21 3 À1 À1 data gives k16 = (2.77 ± 0.21) · 10 cm molecule s .