Document Qk8mEGdeg7L9e7oqQaGmzEyD6

ELSEVIER Atmospheric Environment 96 (2014) 145-153 Contents lists available at ScienceDirect Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv ATMOSPHERIC ENVIRONMENT Atmospheric chemistry of HFE-7300 and HFE-7500: Temperature dependent kinetics, atmospheric lifetimes, infrared spectra and global warming potentials Ana Rodriguez Alberto Notario Diana Rodriguez , Araceli Moraleda , Ivan Bravo , Elena Moreno Faculty of Environmental Sciences and Biochemistry, University of Castilla La Mancha, Avenida Carlos III, s/n, 45071 Toledo, Spain b Faculty of Pharmacy, Physical Chemistry Department, University of Castilla La Mancha, Campus de Albacete, Edificio Polivalente, s/n, 02071 Albacete, Spain Institute of Structure of Matter, Molecular Physics Department, CSIC, Serrano 121, 28006 Madrid, Spain d Faculty of Chemical Sciences, University of Castilla La Mancha, Avenida Camilo Jose Cela 10, 13071 Ciudad Real, Spain HIGHLIGHTS . Atmospheric lifetimes and reaction mechanisms of HFE-7300 and HFE7500 were elucidated. Infrared spectra have been measured and radiative forcing efficiencies and global warming potentials were determined. These results enable the discussion of the environmental compatibility of HFE-7300 and HFE-7500. GRAPHICAL ABSTRACT CFACF,),CFCRC IIFE-7300 o:o ?CH2CF13. CF.(CF2)2CFCF(CF,), IIFE7500 +O FESs HFE 7300 0 48 5 24 440 HFE-7500 0.27 CrossMark ARTICLE INFO Article history: Received 28 March 2014 Received in revised form 7 July 2014 Accepted 14 July 2014 Available online 15 July 2014 Keywords: HFE-7300 HFE-7500 Atmospheric lifetimes Reaction mechanism Infrared spectra Radiative efficiencies ABSTRACT The atmospheric degradation of two hydrofluoroethers, HFE-7300 In-C2F5CF(OCH3)CF(CF3)2] and HFE7500 In-C3F7CF(OC2H5)CF(CF3)2] used in industrial applications has been studied. The kinetics and reaction products were determined at atmospheric pressure as a function of temperature in a reaction chamber using GC/FID and GC/MS techniques for the analysis. The following Arrhenius expressions were obtained (in units of cm3 molecule-1 5-1): kHFE-7300 + OH = (5.6 2.0) x 10-13 exp(--(1186 111)/T); kHFE7300 + cl = (3.8 1.3) x 10 12 exp(--(968 101)/T): and kHFE-7500 + OH = (7.6 6.0) x 10-12 exp(--(1163 385)/T) (temperature range 271-333 K). The atmospheric lifetimes calculated from kinetic data for HFE-7300 and HFE-7500 were 5.24 and 0.30 years, respectively. In the oxidation of HFE-7300 with OH and Cl radicals, the only detected product was CF3CF2CF(OCHO)CF(CF3)2, whereas in the oxidation of HFE-7500 by OH radicals the detected products were: C3F7CF(OC(O)CH3)CF(CF3)2 and C3F7CF(OC(O)H)CF(CF3)2. Infrared spectra of the studied HFEs have also been measured and radiative forcing efficiencies were determined. Combining these results with the kinetic data, we estimated 100year time horizon global warming potentials of 440 and 12 for HFE-7300 and HFE-7500, respectively. 2014 Elsevier Ltd. All rights reserved. * Corresponding author. E-mail address: IM@uclm.es (A. Rodriguez). http://dx.doi.org/10.1016/j.atmosenv.2014.07.033 1352-2310/ 2014 Elsevier Ltd. All rights reserved. 146 A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 1. Introduction The adverse environmental impact of chlorinated hydrocarbons on the Earth' ozone layer (Molina and Rowland, 1974; Farman et al., 1985) has focused attention on the effort to replace these compounds with environmentally acceptable substitutes. Hydrofluoroethers (HFEs) have been developed as new alternatives to replace chlorofluorocarbons (CFCs), hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs) in applications such as refrigerants, blowing and cleaning agents, and they are also used as fuels additives, solvents, and medical products (Tsai, 2005). HFE7300 [n-C2F5CF(OCH3)CF(CF3)2] and HFE-7500 [n-C3F7CF(OC2H5) CF(CF3)2] are members of this class of compounds. Thus, HFE-7500 shares many of the inertness and dielectric properties of perfluorinated fluids, and is expected to offer superior environmental behavior. HFE-7300 is effective at mitigating the aggressiveness of solvents and the flammability of blends (3MTM NovecTM 7500 Engineered Fluid). HFE-7300 and 7500 are volatile liquids with a vapor pressure of 5.9 106 and 2.1 103 Pa at 25 C (3MTM NovecTM 7500 Engineered Fluid), respectively, which will be released into the atmosphere during their use. In this sense, it is necessary to perform tropospheric kinetic and product studies of these HFEs, in order to assess the impact of their degradation processes on the air quality and global warming. Generally, HFEs show low surface sticking coefficients and low water solubility. Thus, primary removal of HFEs in the troposphere will mainly be initiated by reaction with OH radicals. Although global atmospheric abundance of OH radical is around 3 orders of magnitude greater than that of chlorine atoms, Cl reactions are generally faster than OH reactions, so their contribution to the degradation of organic compounds may be not negligible compared to the role of OH (Finlayson-Pitts and Pitts, 2000). The contribution of Cl atoms to the oxidation of HFEs could be significant in areas where the concentration of chlorine precursor species has been reported to be high. Thus, Cl atom initiated oxidation may have a significant local impact for very short-lived substances (VSLS) e lifetimes less than a few months e in those areas (WMO, 2011). In this sense, recent studies suggest that in the early morning, the production rate of the Cl atoms exceeds the production of OH radicals for 2e3 h after sunrise, when high concentrations of ClNO2, precursor of the atoms of chlorine through its photolysis, have been measured in different parts of the world (Osthoff et al., 2008; Thornton et al., 2010; Mielke et al., 2011; Phillips et al., 2012). The tropospheric lifetimes for most gases are generally controlled by reactions with OH radicals (Kurylo and Orkin, 2003). However, reactions with other oxidants as Cl radical can also play limited roles. The calculation of atmospheric lifetimes is complicated due several sources as the temporal and spatial variability of the reacting species concentrations what lead to the difficulty to be confident of their absolute concentrations. Besides, accurate determinations of the temperature-dependent rate coefficients with tropospheric oxidants, as OH and Cl, are essentials due the negative vertical temperature profile of the troposphere. In this sense, chemical species of low reactivity will possess lifetimes larger that the characteristic mixing time in the troposphere and will present a relatively homogeneous vertical distribution therein (well-mixed compounds). On the other hand, there are no unique lifetimes for VSLS since they are not "well-mixed" and their lifetimes depends, on the altitude and latitude of emissions along with the chemical and physics conditions of the atmosphere. Therefore, temperaturedependence rate coefficient studies of the reactions of emitted species with atmospheric oxidant are crucial to obtain accurate values for atmospheric lifetimes (WMO, 2011). In this work we report the first study of the temperature dependence of the kinetics of the reactions of OH radicals with HFE-7300 and HFE-7500, and Cl atoms with HFE-7300. To our knowledge, there is only a single previous kinetic study on the OH HFE-7500 reaction at room temperature and low pressure (Goto et al., 2002). It is also here presented the first, as far as we know, experimental study on the degradation pathways and yields of the aforementioned reactions. This work is also the first to report infrared absorption cross section spectra for HFE-7300 and HFE7500 (for HFE-7500 only absorption spectrum was available (Goto et al., 2002)). Using the Pinnock et al. (1995) method together with the spectral data we have estimated the radiative forcing efficiencies. Finally, combining these parameters with kinetic data we have obtained the corresponding global warming potential values; the results enable the discussion of the environmental compatibility of HFE-7300 and HFE-7500. 2. Experimental 2.1. Kinetic study The experimental system used in this study has been previously described in detail (Rodrguez et al., 2010), therefore, only a brief description is given here. Relative-rate experiments were carried out in a 400 L Teflon bag housed in an isothermal cabinet with 6 fluorescent lamps mounted on the walls. The experiments were performed within the temperature range of 271e333 K at 760 torr of total pressure of N2 or synthetic air. Cl and OH radicals were generated by photolysis of Cl2 and H2O2, at lmax 360 nm and lmax 254 nm, respectively. In the smog chamber experiments, the unwanted loss of reactants and products via photolysis, dark chemistry and heterogeneous reactions have to be considered. In this sense, the loss rate was negligible compared to the rate of reaction in all cases. Kinetic data were derived by monitoring the loss of the HFE relative to one or more reference compounds: Cl=OH HFE/Products kHFE Cl=OH Reference/Products kR The decays of the reactant and reference were then plotted using the expression: Ln HHFFEE00 kkHRFE Ln RReeffeerreennccee0t (I) where [HFE]t0, [HFE]t, [Reference]t0, and [Reference]t are the concentrations at times 0 and t, and kHFE and kR are the rate constants for the reactions of the hydrofluoroethers and the reference compound with Cl or OH radicals, respectively. Plots of Ln([HFE]0/ [HFE]t) vs. Ln([Reference]0/[Reference]t) should be linear, pass through the origin and have a slope equal to kHFE/kR. The decay of HFE and reference compound was monitored by gas chromatography, with flame ionization detection, GCeFID (Shimadzu 2010), using a capillary column (size: 30 m 0.32 mm 1 mm. Meta.X5 Teknokroma) maintained isothermal at 40 C. The concentration ranges of the reactants were as follow: (in molecule cm3): HFE (4.2e9.6) 1014, reference compound (1.5e6.4) 1015, H2O2 (4.7e18.4) 1015 and Cl2 (4.2e5.6) 1015 (see Table 1S, in Supporting information). For each mixture of organic compounds, a number of injections of the unreacted mixture, usually 10 or more, were carried out in order to obtain an estimate of the precision associated with the measurements, and to be used in the error analysis. The standard deviations (2s) of these replicate injections were typically 1e2% for both HFEs and reference compounds. A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 147 2.2. Product study 3. Results and discussion The products formed in the reactions of Cl or OH radicals with the studied HFEs were measured by GCemass spectrometry (Shimadzu QP2010) for their identification, and by GCeFID for their quantification. The temperature program used in both chromatographs on a 30 m Meta.X5 column was the following: 40 C (8 min), 10 C/min to 100 C and hold 3 min. All experiments were carried out in the absence and presence of NOx, with initial reactant concentrations as follow: (in molecule cm3): HFE (3.4e7.8) 1014, H2O2 (1.5e4.1) 1015, Cl2 (2.8e4.2) 1015 and NO (1.2e2.4) 1015. The identification of the products was made by analysis of the mass spectrum. In order to determine the concentrations of the HFE and the detected products, the GCeFID response factors of each compound were determined by introducing different known amounts of the commercial sample into the Teflon chamber and conducting several replicate analyses. Product yields were obtained from plots of the amount of the detected product against the amount of lost reactant. For compounds that are not commercially available, the yields were estimated using the effective carbon number method, ECN (Scanlon and Willis, 1985). 2.3. Infrared spectra Infrared spectra of HFE-7300 and HFE-7500 were measured in our laboratory. Measurements were obtained at 298 K using a 50 L quartz-glass reactor equipped with a White-type multiple-reflection mirror system with a base length of 1.37 m and a total optical path length of 197 m. Details of the experimental set-up can be found elsewhere (Bravo et al., 2013) (see Supporting information as well). The IR spectra were obtained by co-adding 64 scans with a resolution of 1 cm1 in the spectral range 2000e600 cm1, using a Bruker, VERTEX 80V FTIR spectrophotometer, equipped with a MCT (mercury cadmium telluride) detector. The compounds were injected into the reactor at concentrations within (0.4e5) 1013 molecule cm3, and mixed there with synthetic air to a total pressure within (25e150) 2 torr. To ensure that saturation was not a problem in the measurements, for each compound, the peak absorbance was plotted as function of pressure. Points at higher pressures showing non-linear behavior were ignored. Plots of absorbance vs. compound concentration showed good linearity, and zero intercepts (see Fig. S1, Supporting information). The spectra were then normalized to this value for the cross section and the integrated absorption cross section calculated from the average of the spectra. No dependence on total pressure was observed for both studied compounds. The cross section spectra presented here were calculated from the average of the cross section spectra measured within 25e70 torr (we discarded measurements within 70e150 torr because of the noise). Absorption and integrated absorption cross section of HFE-7300 and HFE-7500 at 298 K were then obtained using the corresponding relationships described in literature (Bravo et al., 2013). 2.4. Reagents The chemicals used and their stated purities were as follows: HFE-7300 (>99.5%, 3 M), HFE-7500 (>99%, 3 M), Cl2 (>99,8%, Praxair), hydrogen peroxide (>60%, Fisher Chemical), methane (99%, Air Liquide), chloromethane (99.5%, Aldrich), dichloromethane (>99%, Fluka), ethane (>99%, Aldrich), chloroethane (99.7%, Aldrich) and methyl perfluorooctanoate (98%, Aldrich). Synthetic air (99.999%, Air Liquide) and N2 (99.999%, Air Liquide) were employed as bath gases for the experiments, and He (99.998%, Air Liquide) was used as GC carrier gas. 3.1. Kinetic study The kinetic studies of degradation of HFE-7300 with OH and Cl radicals and HFE-7500 with OH radicals were carried out at atmospheric pressure within the temperature range 271e333 K. Fig. 1 shows the loss of the title HFEs vs. the reference compounds following exposure to OH or Cl radicals at 298 1 K and ~760 torr of air or N2. Consistently with expectations (Eq. (I)), the plots were linear with zero intercept. The reference compounds used in this study were methane, chloromethane, dichloromethane, ethane and chloroethane whose rate constants with OH radicals are as follow: (3.6 0.4) 1014 cm3 molecule1 s1 and (1.0 0.1) 1013 cm3 molecule1 s1 for chloromethane and dichloromethane respectively (Atkinson et al., 2008); (2.4 0.2) 1013 cm3 mol ecule1 s1 for ethane (Atkinson et al., 2006); and (4.2 0.4) 1013 cm3 molecule1 s1 for chloroethane (Cohen and Westberg, 1991). The rate constants of the reactions of Cl atoms with the compounds used as reference were taken from the values reported by Atkinson et al.: (1.0 0.1) 1013 cm3 molecule1 s1 for methane (Atkinson et al., 2006) and (4.8 0.5) 1013 cm3 molecule1 s1 and (3.4 0.3) 1013 cm3 molecule1 s1 for chloromethane and dichloromethane (Atkinson et al., 2008), respectively. A linear least-square analysis of the relative rate plots gives the following rate constants (in units of cm3 molecule1 s1 and errors 2s): kHFE-7300 OH (1.10 0.35) 1014, kHFE-7300 Cl (1.50 0.30) 1013 and kHFE-7500 OH (1.37 0.29) 1013, which are the average of the individual determinations with various reference compounds and air or N2 as bath gases (see more detailed information in Table S1, Supporting information). The obtained results were independent of the bath gas used, so the absence of changes due to the presence of O2 suggests that there is no interference from secondary reactions (Kaiser and Wallington, 1996). For all the studied reactions, the rate constants were found to increase with increasing temperature, as shown in Fig. 2 in the form of Arrhenius plots. The rate constants were measured over the range 271e333 K at atmospheric pressure (see Table 2S, Supporting information), using chloromethane as reference compound in the reactions with OH radicals kCH3ClOH 2:1 1012 exp1210 200=Tcm3 molecule1 s1 (Atkinson et al., 2008), and dichloromethane in the reaction with Cl atoms kCH2Cl2Cl 5:9 1012 exp850200=Tcm3 molecule1 s1 (Atkinson et al., 2008). The linear weighted least-squares analysis of the data yields the activation energies and the pre-exponential factors, and affords the calculation of the kinetic rate constants through the following equations (errors are 2s): kHFE7300OH 5:62:0 1013 exp 1186111=T cm3 molecule1 s1 kHFE7300Cl 3:81:3 1012 exp 968101=T cm3 molecule1 s1 kHFE7500OH 7:66:0 1012 exp 1163385=T cm3 molecule1 s1 This work reports the first temperature dependence study for the degradation of HFE-7300 with OH and Cl radicals and HFE-7500 148 A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 Fig. 1. Relative rate plot at 298 1 K and ~760 torr of the reactions (a) HFE-7300 OH: Left y-axis: Chloromethane (-); Right y-axis: Dichloromethane (B) as reference compounds; (b) HFE-7300 Cl: Methane (:), Chloromethane (-) and Dichloromethane (B) as reference compounds; (c) HFE-7500 OH; Left y-axis: Chloromethane (C); Right yaxis: Ethane () and Chloroethane (:) as reference compounds. with OH radicals. There is a single previous study of the kinetic of the reaction of HFE-7500 with OH radicals at room temperature carried out by Goto et al. (2002). These authors studied the HFE7500 OH reaction at 295 K and 200 torr of total pressure in a relative technique (FTIRephotolysis system), and reported a value of (2.6 0.6) 1014 cm3 molecule1 s1. This value is around five times lower than that reported herein (2.6 vs. 13.7 1014 cm3 molecule1 s1). This fact may suggest a pressure dependence of the rate constant, however, the H-abstraction reactions of other reported HFEs are usually independent of the total Fig. 2. Arrhenius plots at atmospheric pressure of the reactions: (:) HFE-7300 OH; () HFE-7300 Cl; and (C) HFE-7500 OH. pressure (Bravo et al., 2010a; Daz-de-Mera et al., 2009), and therefore the source of this discrepancy is not clear. On the other hand, our results are consistent with kinetic rate constant for fluorinated ethers with similar reactive sites in the molecule, as can be shown in Table 1. For instance, HFE-7300 OH rate constant presented here is in agreement with those found in literature for HFEs with a methyl group in their structure, as C3F7OCH3, (CF3)2CFOCH3, C4F9OCH3 or CF3OCH3. A similar comparison can be made for HFE-7500 OH rate constant presented here with HFEs with a methylene group in their structure, as CF3OC2H5 and C4F9OC2H5. Besides, if it is considered the HFE-7300 Cl reaction, the agreement is also good when we compare to C5F11OCH3, C4F9OCH3, C3F7OCH3 and CF3OCH3, all of them containing a methyl group in their structures. Therefore, a comparable reactivity is obtained with fluorinated ethers of similar structure, showing that the rate constant is not dependent on the number of eCF2e in the perfluorinated chain, even though when this chain is branched (Bravo et al., 2010a; Tokuhashi et al., 1999). On the contrary, the reactivity decrease of the fluorinated ethers compared to non fluorinated ethers presumable reflects the increase in CeH bond strength associated with the presence of multiple fluorinated substituents (Goto et al., 2002; Stein et al., 1991). For instance, in Table 1 we can see that OH rate constant for n-C4H9OC2H5 is more than 200 times larger than that found for the corresponding fluorinated ether C4F9OC2H5. Finally, the replacement of a methyl group in HFE-7300 by an ethyl group in HFE-7500 leads to a substantial reactivity increase towards OH and Cl radicals with kHFE-7500 more than 10 times larger than kHFE-7300 for both radicals: 1.1 vs 13.7 1014 cm3 molecule1 s1 for OH, and 1.5 vs 22.0 1013 cm3 molecule1 s1 for Cl (see Table 1). As is A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 149 Table 1 Rate constants of the reactions of non-fluorinated ethers and fluorinated ethers with OH and Cl radicals. Non fluorinated ether Ether OH CH3OCH3 C2H5(CH3)2OCH3 n-C5H11OCH3 C2H5OC2H5 tert-C4H9OC2H5 n-C4H9OC2H5 Cl CH3OCH3 C4H9OCH3 C2H5OC2H5 C4H9OC2H5 a Bonard et al. (2002). b Teton et al. (1995). c Teton et al. (1996). d Mellouki et al. (1995). e Notario et al. (2000). f This work. g Bravo et al. (2010a). h Stein et al. (1991). i Chen et al. (2001). j Oyaro and Nielsen (2003). k Oyaro et al. (2005). l Diaz-de-Mera et al. (2008). m Aranda et al. (2006). n Nohara et al. (2001). o Christensen et al. (1999). p Daz-de-Mera et al. (2009). q Christense et al. (1998). k (cm3 molecule1 s1) 2.77 1012a 6.08 1012b 6.19 1012c 1.36 1011d 8.80 1012c 2.08 1011c 1.30 1010e 1.40 1010e 2.50 1010e 1.50 1010e Fluorinated ether HFE n-C2F5CF(OCH3)CF(CF3)2 HFE-7300 C3F7OCH3 (HFE-7000) (CF3)2CFOCH3 C4F9OCH3 (HFE-7100) CF3OCH3 (HFE-143a) n-C3F7CF(OC2H5)CF(CF3)2 (HFE-7500) C4F9OC2H5 (HFE-7200) CF3OC2H5 (HFE-263) n-C2F5CF(OCH3)CF(CF3)2 HFE-7300 C3F7OCH3 (HFE-7000) C4F9OCH3 (HFE-7100) C5F11OCH3 CF3OCH3 (HFE-143a) n-C3F7CF(OC2H5)CF(CF3)2 (HFE-7500) C4F9OC2H5 (HFE-7200) k (cm3 molecule1 s1) 1.11 1014f 1.54 1014g 1.52 1014h 1.49 1014g 1.19 1014i 1.37 1013f 1.00 1013j 1.55 1013k 1.50 1013f 1.24 1013l 1.43 1013m 1.03 1013n 1.40 1013o 2.20 1012p 2.70 1012q summarized in Table 1, this behavior was also observed for other HFEs with similar structure; as e.g. HFE-7200 and HFE-7100, where the rate constant with OH and Cl radicals is larger for HFE-7200 than for HFE-7100 (Bravo et al., 2010a; Aranda et al., 2006). All these observations have interesting implications for the design of chemical structures with particular industrial applications. The length (and branching) of the fluorinated part of the ether may be substantially modified to provide the desired physico-chemical behavior while the alkyl part may be modified to increase the gas-phase reactivity and hence ensure efficient elimination from the troposphere. 3.2. Mechanism of the atmospheric oxidation of HFE-7300 Fig. 3a shows the reaction mechanism proposed for the reaction of HFE-7300 with OH and Cl radicals. The atmospheric degradation of HFE-7300 proceeds through the abstraction of an H atom and the alkyl radical formed in this reaction adds O2 to form the corresponding alkyl peroxy radical. Depending on the experimental conditions, absence or presence of NOx, the peroxy radicals may react with NO, HO2, other peroxy radicals or with themselves, giving rise to the corresponding alkoxy radical. This radical can decompose by CeC bond scission or react with O2, leading to the different reaction products. Based on the fragmentation pattern from the electron impact mass spectrum (see Fig. S2, Supporting information), the only product detected by the GCeMS instrument in the oxidation of HFE-7300 by OH and Cl radicals was CF3CF2CF(OCHO) CF(CF3)2 (both in presence and absence of NOx). This compound arises from the reaction of the alkoxy radical with O2. Goto et al. (2002) compared the relative importance of the decomposition channel of the alkoxy radical vs. its reaction with O2 for fluorinated ethers and non fluorinated ethers. This study suggests that the fluorination of alkoxy radical leads to an increase in the reaction with O2. Unfortunately, in this work the yield of the fluorinated formate could not be calculated since this product is not commercial and we have not found a compound of similar structure suitable for our experiments. 3.3. Mechanism of the atmospheric oxidation of HFE-7500 Mass spectra of the two products detected in the oxidation of HFE-7500 by OH radicals are the same in the absence and presence of NOx (see Fig. S2, Supporting information). One of the detected products was identified as C3F7CF(OC(O)CH3)CF(CF3)2, observing an ion signal at m/z 43 typical of the loss eC(O)CH3 group. The mass spectrum of the other detected product shows ion signals at m/ z 28 and 29, typical of the loss eC(O)$ and eC(O)H (Fig. S2). Taking into account the proposed mechanism, Fig. 3b, two possible compounds can be assigned to this spectrum: C3F7CF(OC(O)H) CF(CF3)2 or C3F7CF(OCH2C(O)H)CF(CF3)2. Although methylene and the terminal methyl groups are both susceptible of H-abstraction, generally, OH and Cl radicals react faster through the abstraction of the most weakly bound hydrogen atoms in the molecule. In this sense, a previous theoretical work involving hydrofluoroethers (Papadimitrou et al., 2004) has shown that HeC bonds are weaker in eCH2 groups than those of eCH3 groups. Furthermore, experimental studies of Cl and OH reactions with C4F9OC2H5 (HFE-7200) (Aranda et al., 2006; Christense et al., 1998) showed that the preferred site of reaction was the eCH2 position. Thus, the detected product arises probably of reaction on the eCH2 group and corresponds to the fluorinated formate: C3F7CF(OC(O)H)CF(CF3)2. Similar results were found in the study of the reaction of HFE-7500 with Cl atoms where the fluorinated acetate was found as the major 150 A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 Fig. 3. Proposed mechanisms for the reactions of (a) HFE-7300 with OH or Cl radicals (X) and (b) HFE-7500 OH; where the detected products are shown in dashed box. product and the fluorinated formate was reported as a minor product (Goto et al., 2002). Even though the fluorinated acetate, C3F7CF(OC(O)CH3)CF(CF3)2, is not commercially available, its yield was estimated from the calibration of a compound with the same molecular formula and commercially available, methyl perfluorooctanoate, using the ECN method (Scanlon and Willis, 1985) (see Fig. S3, Supporting information). The molar yields in the absence and presence of NOx were (59 8%) and (49 9%), respectively. The total error is a combination of systematic errors and the uncertainty of ECN method. These yields were calculated from conversion data below 20% to avoid secondary chemistry contributions. As shown, the yield of this acetate decreased in presence of NOx. This difference can be explained by chemical activation effects in the atmospheric chemistry of different alkoxy radicals, that is to say the differences in the exothermicities of the reactions producing the alkoxy radicals. Peroxy radical self-reactions are generally thermoneutral, while the reactions of peroxy radicals with NO are very exothermic. Thus, the alkoxy radicals (RO) formed by the reaction between RO2 and NO will have higher internal excitation energy and rapid decomposition of these activated alkoxy radicals through CeC bond rupture can compete effectively with collisional thermalization of the radicals. On contrary, the alkoxy radicals formed in the selfreaction of peroxy radicals (RO2) will have little or no excitation energy, and reactions of the radical with O2 become effective, increasing the acetate yield (Hurley et al., 2009; Blanco et al., 2012). Unfortunately, the yield of the other detected product, tentatively considered as the fluorinated formate obtained from decomposition channel, could not be calculated since this compound is not commercially available and we have not found any available chemical of similar structure suitable to apply the ECN method. On the other hand, the yield of C3F7CF(OC(O)CH3)CF(CF3)2 from the reaction of HFE-7500 with Cl has been calculated by Goto et al. (2002). These authors estimate that 68% the alkoxy radicals react with O2 to give the fluorinated acetate. This result is consistent with the yield determined in this work for the fluorinated acetate obtained by OH radical-initiated oxidation (59 8%). In addition, we have to consider that HFE-7500 oxidation may occur at high altitudes and low temperatures in the atmosphere; and, as the decomposition reaction of the alkoxy radical have probably a substantial activation barrier (Goto et al., 2002), a temperature decrease may cause an increase of the yield of the fluorinated acetate in detriment of the fluorinated formate. 3.4. Atmospheric lifetimes For well-mixed compounds e lifetimes greater than a few months e OH-lifetimes are estimated relative to the corresponding lifetime of methyl chloroform (MCF) at 272 K (see WMO, 2011; Spivakovsky et al., 2000 and Hodnebrog et al., 2013 for further detailed discussion): tOH 272 K kCOHH3CCl3 272 KtCH3CCl3 (II) kOH272 K OH where tOH and tCOHH3CCl3 (6.1 year6) are the lifetimes of a given compound and CH3CCl3, respectively, due to the reactions with OH radicals in the troposphere; kOH272 K and kCOHH3CCl3 272 K (6.1 A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 151 1015 cm3 molecule1 s1 (Sander et al., 2011)) are the rate constants for the reactions of these compounds with OH at 272 K. From the data reported in this work for the rate constants and the Arrhenius expressions, the atmospheric lifetimes can be obtained for HFE-7300 and HFE-7500 at 272 K with OH radicals, from equation II. We found lifetimes of 5.24 and 0.37 years for HFE-7300 and HFE-7500, respectively. Since the characteristic mixing time in the troposphere is around 0.5 years, HFE-7300 may be a well-mixed compound, and the lifetime presented here can be considered to be a global lifetime. On the other hand, HFE-7500 is far from being homogeneously distributed and may be considered a VSLS. As was briefly mentioned before, the lifetime for HFE-7500 may depends on when and where the compound is emitted along with the atmosphere conditions (see WMO, 2011). To be consisted with the methodology described by WMO, 2011 for VSLS, we used t 1/k (275 K)[OH] and a global average concentrations [OH] 1 106 molecules cm3 (Prinn et al., 2001) at 275 K for determine an approximate local lifetime of 0.30 years for HFE-7500. Our recommended atmospheric lifetimes for HFE-7300 and HFE-7500 are therefore, 5.24 and 0.30 years, respectively (see Table 2). In order to compare the relative contributions of OH and Cl radicals assuming a global basic reactive removal process, we also determined the lifetimes of the studied compounds at 298 K. We used t 1/k298[oxidant] and a global average concentrations of 1 106 molecules cm3 (Prinn et al., 2001) and 1 103 molecules cm3 (Rudolph et al., 1996) for OH and Cl, respectively (See Table 2). It is clear from the estimated lifetime at 298 K that reactions with OH will play a major role in the atmospheric destruction of the studied HFEs during the day, although the reaction with Cl atoms could also be locally competitive, particularly for HFE-7500. Lifetime obtained here for HFE-7300 and HFE-7500, evidence that these materials are relatively short-lived and their contribution to the warming may be low even though they are highly fluorinated. In this sense, atmospheric lifetime plays an important role in the determination of the radiative efficiency and global warming potentials. As they depend on the atmospheric compound location, a unique radiative efficiency cannot be defined for very short-lived compounds without a detailed knowledge of the spatial and temporal emission pattern (see section 3.6). 3.5. Infrared absorption cross-section measurements Infrared spectra for HFE-7300 and HFE-7500 are illustrated in Fig. 4 (cross-section data are summarized in the Supporting information). As expected, all spectra show strong bands at the CeF stretching region within the 1200e1300 cme1 range. Integrating the spectra between 600 and 2000 cm1 gives the integrated absorption cross sections listed in Table 2. To our knowledge, here we present the first cross section spectrum recorded for HFE-7300. For HFE-7500, Goto et al. (2002) showed the absorption spectra for this molecule but it was not converted into cross section. Therefore, here we present the first data of integrated cross section for the related compound. As was expected, HFE-7500 shows a stronger absorption than HFE-7300 due the presence of more number of CeF bond in the molecule e with maximum absorption cross section in the region 7e8 1018 cm2 molecule1 e what lead to a larger integrated cross section (46.1 vs. 50.5 1017 cm2 molecule1 cm1). 3.6. Radiative forcing efficiencies and global warming potentials To assess the impact of the release to the atmosphere of HFE7300 and HFE-7500 has on the radiation budge, and hence in the context to evaluate the climate impact of emission of these gases, the radiative forcing efficiencies (REs) have been calculated using the methodology described by Pinnock et al. (1995) along with the updated 1 cm1 resolution radiative forcing function recently presented by Hodnebrog et al. (2013). In Table 2 we report the REs results assuming perfect mixing. However, the HFEs studied here present relatively short lifetimes, as was commented in section 3.4, and are unlikely to be well-mixed (mainly HFE-7500). Therefore, their mixing ratio will probably fall off rapidly in the stratosphere, and the use of lifetime-correction methodology in the calculation of REs will be necessary. Here we use the recently described S-shaped lifetime-correction proposed by Hodnebrog et al. (2013) for lifetimes within 104e104 years. This was based on an S-shaped fit to chemical-transport model calculations to crudely account for the fact that short-lived gases are not well-mixed in the atmosphere. The "well-mixed" RE value is multiplied by a factor 2:962t0:9312=1 2:994t0:9302., where t is the atmospheric lifetime in years. "Well-mixed" and lifetime-corrected RE values are summarized in Table 2. Up to our knowledge, we present the first RE literature data for HFE-7300: 0.52 and 0.48 (in W m2 ppbv1) for "wellmixed" and lifetime-corrected REs, respectively. For HFE-7500, only the experimental data reported by Goto et al. (2002) was found in literature (0.37 W m2 ppbv1). This value was calculated using the Pinnock et al. (1995) method without lifetime-correction within the 900e1900 cme1 range. Here we obtained values of 0.56 and 0.27 (in W m2 ppbv1) for wellmixed and lifetime-corrected REs, respectively (see Table 2). Using the outdated radiative forcing function by Pinnock to compare with Goto et al. data, we obtained a well-mixed RE value for HFE- Table 2 Atmospheric lifetimes, integrated Absorption Cross Sections (S), Radiative Forcing Efficiencies (RE) d for a constant mixing ratio profile and lifetime-corrected d and GWPs relative to CO2, for HFE-7300 and HFE-7500. As is recommended by Hodnebrog et al. (2013), we considered an increase of 10% due to stratospheric temperature adjustment for the RE calculations. For the GWP calculations we used the lifetime corrected REs along with the updated values of AGWP for CO2 presented by Hodnebrog et al. (2013) 2.495 1014, 9.171 1014 and 32.17 1014 W m2 yr (kgCO2)1 for time horizons of 20, 100 and 500 years, respectively. t (298 K)a year Recommended Sc,e lifetime (t)b/year REd Lifetime corrected REd GWPs 20 yr 100 yr 500 yr HFE-7300 OH Cl HFE-7500 OH Cl 2.96 211.40 0.23 14.41 5.24 0.30 46.1 50.5 0.52 0.56 0.48 0.27 1620 440 126 45 3 12 a t (298 K) 1/kx[X] where X OH or Cl and kx were obtained from this work and Daz-de-Mera et al. (2009). b Lifetime calculation is detailed in the text. c Units are 1017 cm2 molecule1 cm1. d Units are W m2 ppbv1. e 600e2000 cm1 and 298 K. 152 A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 Fig. 4. Infrared spectra of: (a) HFE-7300 and (b) HFE-7500. Spectra have been smoothed to ca. 2 cm1 resolution using a sliding average method. 7500 of 0.54 (in W m2 ppbv1). This value is 31% larger than that obtained by Goto et al.: 0.54 vs. 0.37 (in W m2 ppbv1). This discrepancy may come from the restricted wavenumber interval used for the RE calculation for Goto et al. compared to that used in the present work since the radiative forcing function is strong within 700e900 cm1 (Pinnock et al., 1995). To clarify, we have calculated the RE within 900e1900 cm1 range and we found that our value was still 16% larger than that reported by Goto et al. (2002) (0.44 vs. 0.37 W m2 ppbv1). Differences within 14e25% of existing experimental values provide a valuable data for the REs (Blowers et al., 2007; Bravo et al., 2010b), however, a discrepancy of 16% is high when the same Pinnock's method was used in both studies. Besides, the RE value reported by Goto et al. (2002) for HFE7500 (n-C3F7CF(OC2H5)CF(CF3)2) is surprisingly low when compared to HFEs of the same series but with a lower number of CeF bonds. For instance, the reported RE value for HFE-7200 (C4F9CO2H5) is 0.42 W m2 ppbv1 (Bravo et al., 2010a). Theoretical methods for the determination of REs may be useful this time to investigate where the discrepancies come from. In this sense, we used the methodology described by Bravo et al. (2010b, 2011a) to determine the RE of HFE-7500. Using this approaching we found REs of 0.55 and 0.43 W m2 ppbv1 for the 0e2500 and 900e1900 cme1 wavenumber intervals, respectively, which are in excellent agreement with the experimental values presented here for well-mixed REs: 0.54 and 0.44 W m2 ppbv1 for the 600e2000 and 900e1900 cme1 intervals respectively. Therefore, it is unclear why the RE obtained by Goto et al. (2002) is lower; this might reflect differences in the spectral variation of the absorption crosssection they used. The global warming potential (GWP) of a greenhouse gas takes into account both the lifetime and the RE of the gas and represents its contribution to the warming compared to carbon dioxide. The 100 year GWP is the common scale generally used and IPCC (see e.g. Forster et al., 2007) regularly report 20, 100 and 500 year GWP values for a large number of gases. Table 2 presents the 20, 100 and 500 year GWPs obtained here for HFE-7300 and HFE-7500. Note that these values were calculated using the OH lifetimes at 272 K obtained in section 3.4 and lifetime corrected RE values given above to get more realistic results. We can see that the 100 year GWP is considerably higher than that of CO2 for both gases, particularly for HFE-7300 because of its longer lifetime. However, these values are much smaller than those of the most common CFCs that HFEs use to replace d for instance, the 100 year GWP for CFC-11 is 4660 (Hodnebrog et al., 2013). Here we report the first GWP values for HFE-7300. For HFE-7500, Goto et al. (2002) reported values of GWP relatives to CFC-11 instead of CO2. When it is converted into GWP relative to CO2 using the RE and lifetime values they reported for HFE-7500 (2.2 years and 0.37 W m2 ppbv1), leads to a 100 year GWP for HFE-7500 of 110 e almost 10 time larger than that reported herein. Therefore, we consider the GWP values obtained in the present work for HFE-7500 are also the first data currently reported in literature. To conclude, we have carefully examined the atmospheric lifetimes and REs along with the GWP values derived, and therefore we recommend the data presented here for HFE-7300 and HFE-7500 are the best currently available. 3.7. Atmospheric implications Proposed H-abstraction mechanisms described in Fig. 3 suggest that oxidation of HFE-7300 and HFE-7500 leads to the corresponding fluorinated ester. According with its short atmospheric lifetime (0.30 year), HFE-7500 rapidly reacts with OH or Cl radicals in the troposphere leading to the corresponding fluorinated formate (C3F7CF(OC(O)H)CF(CF3)2) and acetate (C3F7CF(OC(O)CH3) CF(CF3)2). As shown in section 3.3, the acetate is presumed to be the main oxidation product, while the remainder could decompose to give the fluorinated formate. Because of its short lifetimes, HFE7500 may be involved in local smog processes. On the other hand, HFE-7300 possess a larger atmospheric lifetime compared to HFE-7500, which suggests that it may remain in the troposphere for a longer time while its oxidation with OH and Cl occurs. As commented in section 3.2, the corresponding fluorinated formate (C2F5CF(OCHO)CF(CF3)2) is supposed to be the unique direct oxidation product because of the presence of the eCH3 group in the structure. GWP values presented in section 3.6 are direct-GWP, which take into account only the direct effect of the emitted compound. Nevertheless, fluorinated formates and acetates are potentially greenhouse gases since they absorb infrared radiation strongly within the atmospheric windows, due to the presence of CeF bonds in the molecule. Many fluorinated esters contribute more actively to global warming than the parent HFEs, even when they both have the same number of CeF bonds in the molecular structure (Bravo et al., 2011b). It is worth noting, that the HFEs studied here may contribute more to the warming than expected from their direct GWP values due the cumulative effect of fluorinated esters formed during their atmospheric oxidation. To shed light on the A. Rodrguez et al. / Atmospheric Environment 96 (2014) 145e153 153 contribution to the warming of these oxidation products, we have theoretically estimated their radiative behavior (see Table S3, Supporting information for details). In this sense, we estimated a RE value of 0.58 W m2 ppbv1 for the fluorinated formate coming from the HFE-7300 oxidation, what led to an indirect and net GWP100 of 310 and 750, respectively. For the fluorinated acetate and formate coming from HFE-7500 we predicted REs of 0.68 and 0.61 W m2 ppbv1, respectively; and our estimated indirect and net GWP100 for HFE-7500 are 110 and 125. From this outlook we clearly see as the contribution to the warming of HFE-7300 and HFE-7500 may increase within 70e80% when we take into account this cumulative effect from the oxidation products when compare to the direct GWP estimations. Acknowledgments We thank the Spanish Ministry of Science e Education (CGL2011-24799) and the Castilla-La Mancha Science e Education Council (PEII09-0262-2753) for their financial support. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atmosenv.2014.07.033. References Aranda, A., Diaz-de-Mera, Y., Bravo, I., Rodriguez, D., Rodrguez, A., Martnez, E., 2006. Environ. Sci. Technol. 40, 5971e5976. Atkinson, R., Baulch, D., Cox, R., Crowley, J.N., Hampson, R., Hynes, R., Jenkin, M., Rossi, M., Troe, J., 2006. Atmos. Chem. Phys. 6, 3625e4055. Atkinson, R., Baulch, D., Cox, R., Crowley, J.N., Hampson, R., Hynes, R., Jenkin, M., Rossi, M., Troe, J., Wallington, T.J., 2008. Atmos. Chem. Phys. 8, 4141e4496. Blanco, M.B., Bejan, I., Barnes, I., Wiesen, P., Teruel, A.M., 2012. Environ. Sci. Technol. 46, 8817e8825. Blowers, P., Moline, D.M., Tetrault, K.F., Wheeler, R.R., Tuchawena, S.L., 2007. J. Geophys. Res. 112, D15108. Bonard, A., Daele, V., Delfau, J.L., Vovelle, C., 2002. J. Phys. Chem. A 106, 4384e4389. Bravo, I., Daz-de-Mera, Y., Aranda, A., Smith, K., Shine, K.P., Marston, G., 2010a. Phys. Chem. Chem. Phys. 12, 5115e5125. Bravo, I., Aranda, M.D., Hurley, G., Marston, D.R., Nutt, K.P., Shine, K., Smith, Wallington, T.J., 2010b. J. Geophys. Res. 115, D24317. Bravo, I., Marston, G., Nutt, D.R., Shine, K.P., 2011a. J. Quant. Spectrosc. Radiat. Transf. 112, 1967e1977. Bravo, I., Daz-de-Mera, Y., Aranda, A., Moreno, E., Nutt, D.R., Marston, G., 2011b. Phys. Chem. Chem. Phys. 13, 17185e17193. Bravo, I., Rodrguez, A., Rodrguez, D., Diaz-de-Mera, Y., Notario, A., Aranda, A., 2013. ChemPhysChem 14, 3834e3842. Chen, L., Kutsuna, S., Nohara, K., Takeuchi, K., Ibusuki, T., 2001. J. Phys. Chem. A 105, 10854e10859. Christense, L.K., Sehested, J., Nielsen, O.J., Bilde, M., Wallington, T.J., Guschin, A., Molina, L.T., Molina, M.J., 1998. J. Phys. Chem. A 102, 4839e4845. Christensen, L.K., Wallington, T.J., Guschin, A., Hurley, M.D., 1999. J. Phys. Chem. A. 103, 4202e4208. Cohen, N., Westberg, K.R., 1991. J. Phys. Chem. Ref. Data 20, 1211e1311. Diaz-de-Mera, Y., Aranda, A., Bravo, I., Rodrguez, D., Rodrguez, A., Moreno, E., 2008. Environ. Sci. Pollut. Res. 15, 584e591. Daz-de-Mera, Y., Aranda, A., Bravo, I., Moreno, E., Martnez, E., Rodrguez, A., 2009. Chem. Phys. Lett. 479, 20e24. Farman, J.D., Gardiner, B.G., Shanklin, J.D., 1985. Nature 315, 207e210. Finlayson-Pitts, B.J., Pitts Jr., J.N., 2000. Chemistry of the Upper and Lower Atmo- sphere: Theory, Experiments, and Applications. Academic Press, San Diego, CA. Forster, P.M.D., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D.W., Haywood, J., Lean, J., Lowe, D.C., Myhre, G., Nganga, J., Prinn, R., Raga, G., Schulz, M., Van Dorland, R., 2007. In: Solomon, S. (Ed.), Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge. Goto, M., Inoue, Y., Kawasaki, M., Guschin, A.G., Molina, L.T., Molina, M.J., Wallington, T.J., Hurley, M.D., 2002. Environ. Sci. Technol. 36, 2395e2402. Hodnebrog, ., Etminan, M., Fuglestvedt, J.S., Marston, G., Myhre, G., Nielsen, C.J., Shine, K.P., Wallington, T.J., 2013. Rev. Geophys. 51, 300e378. Hurley, M.D., Wallington, T.J., Laursen, L., Javadi, M.S., Nielsen, O.J., Yamanaka, T., Kawasaki, M., 2009. J. Phys. Chem. A. 113, 7011e7020. Kaiser, E.W., Wallington, T.J., 1996. J. Phys. Chem. A. 100, 4111e4119. Kurylo, M.J., Orkin, V.L., 2003. Chem. Rev. 103, 5049e5076. Mellouki, A., Teton, S., LeBras, G., 1995. Int. J. Chem. Kinet. 27, 791e805. Mielke, L.H., Furgeson, A., Osthoff, H.D., 2011. Environ. Sci. Technol. 45 (20), 8889e8896. Molina, M.J., Rowland, F.S., 1974. Nature 249, 810e812. Nohara, K., Toma, M., Kutsuna, S., Takeuchi, K., Ibusuki, T., 2001. Environ. Sci. Technol. 35, 114e120. Notario, A., Mellouki, A., Le Bras, G., 2000. Int. J. Chem. Kinet. 32, 105e110. Osthoff, H.D., Roberts, J.M., Ravishankara, A.R., Williams, E.J., Lerner, B.M., Sommariva, R., Bates, T.S., Coffman, D., Quinn, P.K., Dibb, J.E., Stark, H., Burkholder, J.B., Talukdar, R.K., Meagher, J., Fehsenfeld, F.C., Brown, S.S., 2008. Nat. Geosci. 1 (5), 324e328. 3MTM NovecTM 7500 Engineered Fluid, Product Information and 3MTM NovecTM 7300 Engineered Fluid, Product Information. Oyaro, N., Nielsen, C.J., 2003. Asian Chem. Lett. 7, 119e122. Oyaro, N., Sellevaag, S.R., Nielsen, C.J., 2005. J. Phys. Chem. A. 109, 337e346. Papadimitrou, V.C., Kambanis, K.G., Lazarou, Y.G., Papagiannakopoulos, P., 2004. J. Phys. Chem. A 108, 2666e2674. Phillips, G.J., Tang, M.J., Thieser, J., Brickwedde, B., Schuster, G., Bohn, B., Lelieveld, J., Crowley, J.N., 2012. Geophys. Res. Lett. 39, L10811. Pinnock, S., Hurley, M.D., Shine, K.P., Wallington, T.J., Smyth, T.J., 1995. J. Geophys. Res. 100, 23227e23238. Prinn, R.G., Huang, J., Weiss, R.F., Cunnold, D.M., Fraser, P.J., Simmonds, P.G., McCulloch, A., Harth, C., Salamch, P., O'Doherty, S., Wang, R.H.J., Porter, L., Miller, B.R., 2001. Science 292, 1882e1888. Rodrguez, A., Rodrguez, D., Garzon, A., Soto, A., Aranda, A., Notario, A., 2010. Phys. Chem. Chem. Phys. 12, 12245e12258. Rudolph, J., Hoppman, R., Plass-Dlmer, C.H., 1996. Atmos. Environ. 30, 1887e1894. Sander, S.J., Abbatt, J.P.D., Barker, J.R., Burkholder, J.B., Friedl, R.R., Golden, D.M., Huie, R.E., Kolb, C.E., Kurylo, M.J., Moortgat, G.K., Orkin, V.L., Wine, P.H., 2011. Chemical Kinetics and Photochemical Data for Use in Atmospheric Studies, Evaluation No. 17. JPL Publication 10-6. Jet Propulsion Laboratory, Pasadena. http://jpldataeval.jpl.nasa.gov. Scanlon, J.T., Willis, D.E., 1985. J. Chromatogr. Sci. 23, 333e340. Spivakovsky, C.M., Logan, J.A., Montzka, S.A., Balkanski, Y.J., Foreman-Fowler, M., Jones, D.B.A., Horowitz, L.W., Fusco, A.C., Brenninkmeijer, C.A.M., Prather, M.J., Wofsy, S.C., McElroy, M.B., 2000. J. Geophys. Res. 105, 8931e8980. Stein, S.E., Rukkers, J.M., Brown, R.L., 1991. NIST Standard Reference Database 25: Structures and Properties. NIST, Gaithersburg, MD. Teton, S., Mellouki, A., LeBras, G., Sidebottom, H., 1995. Int. J. Chem. Kinet. 28, 291e297. Teton, S., Mellouki, A., LeBras, G., Sidebottom, H., 1996. Int. J. Chem. Kinet. 28, 291e297. Thornton, J.A., Kercher, J.P., Riedel, T.P., Wagner, N.L., Cozic, J., Holloway, J.S., Dube, W.P., Wolfe, G.M., Quinn, P.K., Middlebrook, A.M., Alexander, B., Brown, S.S., 2010. Nature 464 (7286), 271e274. Tokuhashi, K., Nagai, H., Takahashi, A., Kaise, M., Kondo, S., Sekiya, A., Takahashi, M., Gotoh, Y., Suga, S., 1999. J. Phys. Chem. A 103, 2664e2672. Tsai, W.T., 2005. J. Hazard. Mater. A 119, 69e78. WMO, 2011. Scientific Assessment of Ozone Depletion: 2010, Global Ozone Research and Monitoring Project e Report No. 50. World Meteorological Organization, Geneva, Switzerland, p. 572 (Rep).