Atmospheric Chemistry of E- and Z-CF3CH CHF ... - ACS Publications

Oct 10, 2017 - knowledge, no photolysis study of E- and Z-CF3CH CHF has been reported. In this work, we present the UV absorption cross sections of ...
1 downloads 0 Views 2MB Size
Subscriber access provided by Service des bibliothèques | Université de Sherbrooke

Article 3

Atmospheric Chemistry of E- and Z-CFCH=CHF (HFO-1234ze): OH Reaction Kinetics as a Function of Temperature and UV and IR Absorption Cross Sections Maria Antiñolo, Ivan Bravo, Elena Jimenez, Bernabe Ballesteros, and Jose Albaladejo J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.7b06174 • Publication Date (Web): 10 Oct 2017 Downloaded from http://pubs.acs.org on October 10, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry A is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Atmospheric Chemistry of E- and Z-CF3CH=CHF (HFO1234ze): OH Reaction Kinetics as a Function of Temperature and UV and IR Absorption Cross Sections

María Antiñoloa,b, Iván Bravoc, Elena Jiméneza,b, Bernabé Ballesterosa,b, José Albaladejoa,b,*

a

Departamento de Química Física. Facultad de Ciencias y Tecnologías Químicas. Universidad de Castilla-La Mancha. Avda. Camilo José Cela, 1B. 13071 Ciudad Real, Spain.

b

Instituto de Investigación en Combustión y Contaminación Atmosférica. Universidad de Castilla-La Mancha. Camino de Moledores, s/n. 13071 Ciudad Real, Spain.

c

Departamento de Química Física. Facultad de Farmacia. Universidad de Castilla-La Mancha. Campus Universitario de Albacete. 02071 Albacete, Spain.

Corresponding author: J. Albaladejo. Phone: +34 926 29 53 00, FAX: +34 926 29 53 18. e-mail: [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 43

Abstract We report here the rate coefficients for the OH-reactions (kOH) with E-CF3CH=CHF and Z-CF3CH=CHF, potential substitutes of HFC-134a, as a function of temperature (263-358 K) and pressure (45-300 Torr) by pulsed laser photolysis coupled to laser induced fluorescence techniques. For the E-isomer, the existing discrepancy among previous results on the T-dependence of kOH needs to be elucidated. For the Z-isomer, this work constitutes the first absolute determination of kOH. No pressure dependence of kOH

was

observed,

while

kOH (E)=(7.6±0.2)×10-13 298 T

kOH (Z)=(1.4±0.1)×10-13 298 T

2.44

1.91

exhibits

kOH exp 

exp 

666±10 T

640±13 T

a

non-Arrhenius



behavior: and

 cm3 molecule-1 s-1, where uncertainties

are 2σ. UV absorption cross sections, σλ, are reported for the first time. From σλ and considering a photolysis quantum yield of 1, an upper limit for the photolysis rate coefficients and lifetimes due to this process in the troposphere are estimated: 3 × 10-8 s-1 and >1 year for E-isomer and 2 × 10-7 s-1 and >2 months for Z-CF3CH=CHF, respectively. Under these conditions, the overall estimated tropospheric lifetimes are 15 days (for E-isomer) and 8 days (for Z-isomer), the major degradation pathway being the OH reaction, with a contribution of the photolytic pathway of less than 3% (for E) and 13% (for Z). IR absorption cross sections were determined both experimentally (5004000 cm-1) and theoretically (0-2000 cm-1). From the theoretical IR measurements, it is concluded that the contribution of 0-500 cm-1 region to the total integrated cross sections is appreciable for E-isomer (9%), but almost negligible for Z-isomer (0.5%). Nevertheless, the impact on their radiative efficiency and global warming potential is negligible.

2 ACS Paragon Plus Environment

Page 3 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION The implications on the stratospheric ozone (O3) layer of chlorofluorocarbons (CFCs) are widely known; Cl atoms produced by solar ultraviolet (UV) photolysis of these species catalytically destroy the ozone layer that protects Earth surface from harmful UV radiation. Since their recognition as ozone depleting substances, many CFC have been proposed to be replaced by hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs). Due to their higher reactivity in the troposphere, HCFCs and HFCs are less hazardous to ozone depletion, so at the beginning they were thought to be good replacements. However, they are strong greenhouse gases due to their absorption in the infrared (IR) atmospheric window and their still long lifetime. Therefore, species like hydrofluoroolefins (HFOs) have been proposed more recently as replacements of HFCs in applications such as refrigerants, propellants, and solvents. Even though HFOs also present high IR absorption in the bands due to C-F bonds, they are short-lived species, what reduces their contribution to global warming. In particular, the trans and cis isomers of 1,3,3,3-trifluoropropene, E- and Z-CF3CH=CHF, have been proposed as alternatives to HFC-134a (1,1,1,2-tetrafluoroethane) as refrigerants.1-5 Therefore, the evaluation of their impact on the atmosphere is of great importance in order to consider their usage in the future. This assessment consists in carrying out kinetic investigations of their reaction with tropospheric oxidants such as OH radicals, Cl atoms, nitrate (NO3) radicals and O3, and the estimation of atmospheric lifetimes and climate change metrics, such as the global warming potential (GWP). UV photolysis in the solar actinic region (λ>290 nm) can be an additional removal process for these HFOs to consider in lifetime calculations. To our knowledge, no photolysis study of E- and Z-CF3CH=CHF has been reported. In this work, we present the UV absorption

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 43

cross sections of E-CF3CH=CHF and Z-CF3CH=CHF for the first time and the evaluation of the contribution of this process to the atmospheric removal of these HFOs. The gas-phase kinetics of E-CF3CH=CHF and Z-CF3CH=CHF with some of the atmospheric oxidants has been investigated both experimentally and theoretically.1-4, 6, 7

However, there are some discrepancies in the OH-kinetics that need to be elucidated. E-CF3 CH=CHF + OH → Products

(R1)

Z-CF3 CH=CHF+ OH → Products

(R2)

For reaction (R1), Søndergaard et al.1 determined the relative rate coefficients (kOH) in 700 Torr of air at 296 K in smog chamber using FTIR as a detection technique. Using the absolute kinetic technique of flash photolysis-resonance fluorescence, Orkin et al.

2

measured kOH for E-CF3CH=CHF as a function of temperature (220-370 K),

observing a very weak temperature dependence of the rate coefficient, with a noticeable curvature in the Arrhenius plot (lnk vs. 1/T) and a significant discrepancy (30%) with the result at room temperature given by Søndergaard et al.1 More recently, Zhang et al.3 determined the temperature dependence of kOH for E-CF3CH=CHF using a relative technique: smog chamber-gas chromatograph (GC) with flame ionization detection (FID). These authors showed a typical Arrhenius behavior in the 253-328 K range, i.e. no curvature in the Arrhenius plot was observed. Moreover, from the theoretical point of view, Du et al.6 calculated the rate coefficients for reaction (R1) in the 296-500 K range using conventional transition state theory (TST) with Wigner tunneling correction. They reported an Arrhenius behavior of the rate coefficient with a positive activation energy, while Balaganesh and Rajakumar

7

computed kOH for reaction (R1)

between 200 and 400 K by TST and canonical variational TST (CVTST) calculations reporting a curvature in the Arrhenius plots.

4 ACS Paragon Plus Environment

Page 5 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

For reaction (R2), Nilsson et al.4 determined kOH at 296 K in air using a relative kinetic method using a smog chamber coupled to FTIR spectroscopy. Their results and those from Zhang et al.3 obtained at 298 K by GC-FID present a 12% difference. Similarly to the OH-reaction with E-CF3CH=CHF, Zhang et al.3 showed no curvature in the Arrhenius plot for Z-CF3CH=CHF in the 253-328 K range. For reaction (R2), Balaganesh and Rajakumar7 also calculated kOH between 200 and 400 K by CVTST, suggesting a curvature in the Arrhenius plots. We report in this work the absolute rate coefficients of the OH-reactions with E-CF3CH=CHF (R1) and Z-CF3CH=CHF (R2) as a function of temperature (263-358 K) and pressure (45-300 Torr). The pulsed laser photolysis coupled to laser induced fluorescence (PLP-LIF) technique has been used to determine kOH for both reactions. The tropospheric lifetime (τ) for E-CF3CH=CHF and Z-CF3CH=CHF is the estimated considering all possible removal processes, including UV photolysis as stated above. An accurate measurement of the IR absorption cross sections of E-CF3CH=CHF and Z-CF3CH=CHF is crucial in order to calculate their radiative efficiency (RE), which is extremely sensitive to the exact position and the absolute intensity of the band. RE and τ are needed to estimate GWP. In general, the impact of limiting the range of integration to calculate RE to wavenumbers greater than 500 cm-1 is generally fairly small, except for some compounds, such as hydrofluoroethers and hydrofluoropolyethers, where differences of up to about 7-10% have been found (Bravo et al.8). The IR absorption spectra of E-CF3CH=CHF and Z-CF3CH=CHF have been measured between 400 and 2000 cm-1 at different spectral resolutions by Søndergaard et al.1, Orkin et al.2, Zhang et al.3, and Nilsson et al.4. In this work, IR absorption cross sections were also determined (500-4000 cm-1) and computed (0-2000 cm-1) in order to

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 43

check if the 0-500 cm-1 region is affecting the radiative efficiency of these species. Finally, the atmospheric implications of these species are discussed.

1. METHODS Three experimental set-ups are described in this section: a PLP-LIF system, used in the determination of kOH, a UV absorption set-up, and an IR absorption system. In the last subsection, the method used to theoretically obtain the IR absorption cross sections is also described.

2.1.

Pulsed Laser Photolysis – Laser Induced Fluorescence (PLP-LIF)

Technique The PLP-LIF set-up has been previously described, so only a brief description will be given here.9-13 A jacketed Pyrex cell was used as a reactor, in which pressure was kept between 45 and 300 Torr. Temperature inside the cell was controlled in the 263-358 K range by a thermostatic bath of water (T ≥ 298 K) or ethanol (T < 298 K). Helium, used as the buffer gas, was flowed together with HFO mixtures (containing 1.4%-7.9% of E-CF3CH=CHF or 1.1%-1.5% of Z-CF3CH=CHF diluted in He) and He bubbled through an aqueous solution of H2O2 or HNO3. Total flow rate ranged between 200 and 480 sccm (standard cubic centimeter per minute). Measuring the different mass flow rates of He and reactant mixtures by calibrated mass flow meters and the total pressure with a capacitance manometer, it was possible to determine the concentration of the species inside the reaction cell. OH radicals were produced by pulsed laser photolysis of H2O2 or HNO3 at 10 Hz using a KrF excimer laser (OPTEX, Lambda Physik):

6 ACS Paragon Plus Environment

Page 7 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

H2 O2 +hν (λ = 248 nm) → 2 OH

(R3)

HNO3 +hν (λ = 248 nm) → OH + NO2

(R4)

Upper limits of the initial OH concentration were estimated to be lower than 5 × 1011 molecule cm-3, taking into account the estimated upper limit of precursor concentration, photolysis laser fluence ((4.2-7.9) mJ pulse-1 cm-2), the absorption cross section of H2O2 or HNO3 and their quantum yield for OH production at 248 nm.14, 15 Detection of OH radicals was carried by laser induced fluorescence (LIF). A pulse from a Nd:YAG (Continuum) pumped frequency doubled dye laser excited the (1,0) band of the A2Σ+-X2Π transition of the OH radical at ∼ 282 nm. The probe laser was triggered at a variable time after the photolysis pulse so the LIF from excited OH radicals could be used to monitor the OH decay during the course of the reaction. The signal was averaged by a boxcar unit and recorded onto a computer. Measurements were made under pseudo-first order conditions, in which concentration of reactant exceeds OH concentration ([HFO]0 > 100 [OH]0). Under these conditions, OH LIF decays were fitted to an exponential expression: It = I0 e- k' t

(1)

where It and I0 are the LIF signal at a time t or 0, respectively, and kʹ is the pseudo-first order rate coefficient. Different kʹ were determined by varying the HFO concentration to obtain the second-order rate coefficient, kOH(T), by means of equation (2) or its equivalent (3): k' = kOH (T)[HFO] + k0

(2)

k' − k0 = kOH (T)[HFO]

(3)

k0 is the pseudo-first order rate coefficient determined in the absence of HFO, and it is due to the reaction of OH with its precursor and the diffusion loss of OH. Its value ranged between 70 and 480 s-1, whereas kʹ ranges were 230 – 10800 s-1 for

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 43

E-CF3CH=CHF and 210 – 4200 s-1 for Z-CF3CH=CHF. kOH(T) at a given temperature and pressure were determined from the slope of k′–k0 versus [HFO] plots.

2.2.

UV Absorption Spectroscopy The UV absorption spectra of E- and Z-CF3CH=CHF were measured in the

200 – 350 nm range using a single beam apparatus described in previous works.9, 10, 12, 13 Gas-phase species were introduced in a Pyrex absorption cell with a 107 cm path length, ℓ, in which total pressure was measured by a capacitance pressure transducer. An Oriel deuterium lamp was employed as radiation source and the detector consisted in a coupled-charge device cooled by a Peltier system. A 0.5-m spectrograph equipped with a 300 grooves/mm grating with an instrumental resolution of 0.18 nm was placed between the absorption cell and the detector. Wavelengths were calibrated daily using a pen-ray Hg lamp within an uncertainty of less than 0.06%. Spectra of the evacuated cell and of the cell filled with the gas species were alternately recorded several times by changing the pressure of pure HFO inside the cell (P = 0.93 – 80.6 Torr). The absorption cross sections at the wavelength λ (in base e), σλ, were obtained from the slope of the Beer-Lambert plots as defined by equation (4): Aλ = σλℓ[HFO]

(4)

Absorbance, Aλ, was determined from the transmitted light in absence and presence of HFO. Beer-Lambert plots were linear in the pressure range studied for both isomers, as shown in Figure S1 for E-CF3CH=CHF and Z-CF3CH=CHF.

2.3.

IR Absorption Cross Sections Determination IR absorption spectra were experimentally recorded using a set-up presented in

previous works.10, 12, 13 It consists on a Fourier Transform Infrared (FTIR) spectrometer

8 ACS Paragon Plus Environment

Page 9 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(TENSOR 27, Bruker) coupled to a White-type gas cell with a fixed optical path length, ℓ, of 800 cm. Spectra were measured in the wavenumber range of 500-4000 cm-1 at a spectral resolution of 1 cm-1 by the averaging of 16 interferograms. IR spectra were acquired under static conditions from diluted mixtures of E- and Z-CF3CH=CHF in helium, the dilution factors being between 0.07% and 0.46%. Inside the FTIR cell, total pressure ranged from 1.3 to 50 Torr, so concentrations of HFO were in the range (1.0-12) × 1014 molecule cm-3. The IR absorption cross sections at every wavenumber (in base e), σ(ν ), were determined by employing Beer-Lambert Law, as described for σλ: A(ν ) = σ(ν )ℓ[HFO]

(5)

A(ν ) were plotted versus the HFO concentration (see Figure S2 for Beer-Lambert plots at selected wavenumbers), showing an excellent linearity in the concentration range. Similarly, integrated cross sections, Sint , were determined for different wavenumber ranges using the Beer-Lambert law in terms of the integrated absorbance Aint : Aint = Sint ℓ[HFO]

(6)

The Gaussian09 software package16 was used to perform computational calculations to compute σ(ν ) in the 0-2000 cm-1 range. Molecular structures were first optimized at the B3LYP/6-31G** level of theory, followed by calculation of vibrational frequencies. No symmetry constraints were imposed, and for all molecules, the absence of negative frequencies confirmed that we had obtained a minimum on the potential energy surface. When more than one minimum was found, the structure with the lowest Gibbs free energy was used for the calculation of σ(ν ). The methods used for the calculation of infrared spectra have been described in Bravo et al. 8, 17 Accurate σ(ν ) can only be obtained for fluorinated compounds if the position of the main C-F stretching

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 43

vibrational mode is determined accurately. According to Bravo et al.8, 17 the expression (7) can be used to obtain scaled wavenumbers that are used for the σ(ν ) calculations, where ν calc is the calculated vibrational mode wavenumber and ν scal is the empiricallycorrected value. ν scal (cm-1 )= 0.977 ν calc + 11.664

(7)

A detailed description of the scaling procedure of the calculated wavenumbers is given by Bravo et al.

8, 17

The theoretical calculations provide wavenumber positions and

absorption cross sections, σ(ν ). The overall integrated cross section is the sum of the Sint over the appropriate wavenumber range (500-2000 cm-1 and 0-2000 cm-1, in this work).

Reagents. Liquified gas E-CF3CH=CHF (99%) and Z-CF3CH=CHF (99%) were purchased to Apollo Scientific Inc. HNO3 (65%, Panreac) was used as supplied, and H2O2 (50%, Scharlau) was pre-concentrated by flowing He through it for several days. He (99.999%, Praxair) was used as provided.

3. RESULTS AND DISCUSSION 3.1. OH Reaction Rate Coefficients Examples of temporal profiles of the LIF signal are depicted in Figure S3 in logarithmic form for OH loss in absence and presence of reactant (E-CF3CH=CHF and Z-CF3CH=CHF). kʹ was determined at different HFO concentrations, as described in section 2.1. Figure 1 shows the plots of kʹ-k0 versus [HFO] at 298 K and different pressures between 44 and 302 Torr. No pressure dependence was observed in this range. Therefore, an average of all kOH(T) obtained at different total pressures for (R1) and (R2) are reported in Tables 1 and 2, respectively. For example, the results at 298 K were:

10 ACS Paragon Plus Environment

Page 11 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

kOH(E-CF3CH=CHF) = (7.1 ± 0.3) × 10-13 cm3 molecule-1 s-1 kOH(Z-CF3CH=CHF) = (1.2 ± 0.1) × 10-12 cm3 molecule-1 s-1

Figure 1. Pseudo-first order plots of E-CF3CH=CHF and Z-CF3CH=CHF at 298 K and different total pressures.

In Figures 2 and 3, the averaged kOH(T) for E-CF3CH=CHF and Z-CF3CH=CHF in the temperature range investigated are depicted. As can be seen in the figures, a noticeable, but slight, curvature in the Arrhenius plot is observed. The experimental results are well n

described by a three-parameter expression: kOH (T) =A 298 exp - RT. The n value T

Ea

obtained in this analysis is fixed to reduce the statistical uncertainty in the pre-

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 43

exponential factor, A, and the activation energy, Ea. The best fit in the temperature range 263-358 K provides the following expressions (errors are ±2σ): kOH (E- CF3CH=CHF)=(7.6±0.2)×10-13  kOH (Z- CF3CH=CHF)=(1.4±0.1)×10-13 



T

2.44

298 T



298

1.91

exp  exp 

666±10 T

 cm3 molecule-1 s-1 (8)

640±13 T

 cm3 molecule-1 s-1 (9)

Table 1. Experimental conditions and results for the reaction between OH and E-CF3CH=CHF. Uncertainties are ± 2σ [HFO] / Average 1014 kOH(T) / 10-13 cm3 T/ kOH(T)/ 10-13 cm3 P /Torr molecule molecule-1 s-1 molecule-1 s-1 K cm-3 263 60.1 3.1 – 29 7.18 ± 0.23 7.07 ± 0.27 99.3 3.3 – 32 6.93 ± 0.28 182 2.0 - 19 7.03 ± 0.31 270 49.5 1.8 – 18 6.95 ± 0.31 7.04 ± 0.21 102 5.2 – 49 7.14 ± 0.38 182 2.0 - 19 7.09 ± 0.38 55.4 5.3 – 34 7.02 ± 0.42 278 7.04 ± 0.04 107 1.4 – 13 7.06 ± 0.31 199 5.0 - 49 7.03 ± 0.28 287 55.3 3.5 – 33 6.99 ± 0.17 6.97 ± 0.05 101 1.3 – 12 7.00 ± 0.22 194 7.4 - 70 6.95 ± 0.14 298 43.7 0.78 – 7.3 7.22 ± 0.32 7.06 ± 0.27 70.8 1.1 – 11 7.08 ± 0.29 105 3.2 – 13 7.28 ± 0.18 120 3.5 – 34 7.22 ± 0.17 160 9.3 – 96 6.99 ± 0.09 193 7.1 – 67 6.94 ± 0.19 295 2.8 - 27 6.88 ± 0.32 313 59.4 2.3 – 9.0 7.13 ± 0.20 65.6 3.8 – 36 7.24 ± 0.12 101 1.3 – 12 7.19 ± 0.18 7.22 ± 0.17 136 6.6 – 68 7.05 ± 0.41 203 1.9 - 18 7.30 ± 0.19 328 56.4 0.91 – 8.6 7.53 ± 0.13 7.30 ± 0.37 102 1.1 - 11 7.29 ± 0.22 122 5.7 - 58 7.17 ± 0.10 175 3.9 - 15 7.31 ± 0.46 343 61.6 0.86 – 8.1 7.59 ± 0.29 7.45 ± 0.27 97.6 1.1 - 11 7.53 ± 0.19 140 6.2 – 64 7.32 ± 0.17 211 4.5 - 18 7.61 ± 0.54 358 53.1 0.71 – 6.7 7.76 ± 0.12 7.64 ± 0.23 65.7 3.3 – 32 7.68 ± 0.20 12 ACS Paragon Plus Environment

Page 13 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

104 125 185

2.7 – 10.4 5.3 – 55 1.7 - 16

7.82 ± 0.47 7.54 ± 0.12 7.55 ± 0.20

Table 2. Experimental conditions and results for the reaction between OH and Z-CF3CH=CHF. Uncertainties are ± 2σ [HFO] / Average T/ 1014 kOH(T)/ / 10-12 cm3 kOH(T)/ 10-12 cm3 P /Torr K molecule molecule-1 s-1 molecule-1 s-1 -3 cm 263 53.5 0.78 – 7.4 1.29 ± 0.08 1.27 ± 0.04 92.7 1.7 – 16 1.26 ± 0.11 180 2.3 – 22 1.26 ± 0.08 270 49.0 0.71 – 6.8 1.23 ± 0.06 1.25 ± 0.03 97.6 1.7 – 16 1.24 ± 0.04 171 2.1 - 21 1.26 ± 0.02 278 52.1 1.2 – 12 1.25 ± 0.03 1.25 ± 0.03 100 1.5 – 14 1.25 ± 0.05 197 2.2 - 21 1.22 ± 0.06 287 45.3 0.71 – 6.7 1.25 ± 0.04 1.24 ± 0.03 100 1.5 – 13.9 1.24 ± 0.02 199 2.3 - 22 1.22 ± 0.03 298 45.7 3.4 – 33 1.26 ± 0.09 1.21 ± 0.03 58.5 0.94 – 8.9 1.27 ± 0.05 82.8 1.5 – 14 1.21 ± 0.01 98.5 1.1 – 10 1.23 ± 0.13 200 2.1 – 20 1.28 ± 0.06 302 3.4 - 33 1.21 ± 0.04 313 51.4 0.84 – 7.9 1.19 ± 0.03 1.20 ± 0.04 100 1.2 – 11 1.22 ± 0.05 190 1.6 - 15 1.22 ± 0.04 328 54.5 0.73 – 6.9 1.19 ± 0.07 1.19 ± 0.02 101 1.1 – 11 1.18 ± 0.08 200 1.9 - 18 1.20 ± 0.10 343 55.8 0.77 – 7.3 1.20 ± 0.03 1.20 ± 0.02 107.6 1.2 – 12 1.19 ± 0.07 186.2 1.4 - 14 1.17 ± 0.09 358 50.8 0.73 – 6.9 1.21 ± 0.06 1.21 ± 0.03 98.4 0.91 – 8.7 1.23 ± 0.05 206 1.7 - 16 1.19 ± 0.05

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 43

3.1.1. Comparison of the OH-reactivity with propene and other fluoropropenes at room temperature When the reactivity towards OH of E-CF3CH=CHF and Z-CF3CH=CHF are compared, it is observed that Z-CF3CH=CHF reacts faster. This might be due to steric effects, that make more difficult that the OH radical approaches the double bond in E-CF3CH=CHF than in Z-CF3CH=CHF. A comparison of kOH(298 K) for several propenes with different amount of F atoms can be done, observing the following trend: H

H C

H

H

C H

H

C

C

> H

H

2.57×10-11

C H

F

C

>

F

H

F

F

C F

∼1.25×10-12



F

C

>

F

F

F

F

F

C F

1.21×10-12

>

H F

H

F

F

C F

C

F

C F

∼1.1×10-12

F

C

>

F

∼1.4×10-12

F C

H C

>

F

C

2.15×10-12 H

C

H

C

F

H C

F C

∼2.3×10-12 H

C

H

C F

H

F C

F C

1.48×10-11 F

>

F

H

H C

> H

C F

F

C F

7.06×10-13

It is clear that propene is more reactive18 than fluorinated propenes,19-28 indicating that F atoms cause a deactivation in reactivity due to their electron withdrawing inductive effect. Thus, it is expected that the more F atoms, the lower kOH(298 K) would be. However, in this comparison we found out that perfluoropropene is the most reactive among the fluorinated propenes,20,

25-28

except 3-fluoropropene.19

This was explained by Søndergaard et al.1 by a mechanism through which the OH radical forms a hydrogen bond with F (this will be more important if there are more F atoms), making the approach of OH to the double bond easier. 3-fluoropropene reacts faster than perfluoropropene probably because the influence of the electron withdrawing inductive effect of the F atom over the double bond is very weak. This would also explain why 3,3,3-trifluoropropene is more reactive than other fluoropropenes.20, 21 A possible explanation for the higher reactivity of E-CF3CF=CHF compared to its isomer 14 ACS Paragon Plus Environment

Page 15 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Z-CF3CF=CHF might be the steric effects.22,

23

The lower steric effects would make

easier that the OH radical approaches the double bond in E-CF3CF=CHF.

3.1.2. Comparison with previous studies on the OH+ E-CF3CH=CHF reaction. In Table 3 the room temperature rate coefficient, kOH(298 K), and the temperature dependence parameters (A, n and Ea/R) from other experimental studies are compared with our results. kOH(298 K) obtained in this work is in good agreement with that reported by Orkin et al.2 (absolute kinetic technique) and Zhang et al.3 (relative kinetic technique), whereas the first report on relative kOH(298 K) from Søndergaard et al.1 is 1.3 times higher and lies outside uncertainties. Theoretical predictions of kOH(298 K) from Du et al.6 and Balaganesh and Rajakumar7 are also higher with respect to our value (21% and 15%, respectively). Regarding the temperature dependence of kOH(T), as can be seen in Figure 2, the T-trend observed in this work is in agreement with that reported by Orkin et al.2, while the tendency observed by Zhang et al.3 only agrees at T> 298K. However, the difference at lower temperatures with our kOH(T) is less than 5%. Only kOH(T=253 K) reported by Zhang et al.3 is much lower than the obtained by Orkin et al.2 As shown in Table 3, Zhang et al.3 reported an Arrhenius behavior for kOH(T) with a positive activation energy that contrasts with our findings and those from Orkin et al.2 Computed kOH(T) for E-CF3CH=CHF also show a different behavior as a function of temperature compared to experimental works (see Figure 2).6,7 These theoretical calculations are far from the trend observed experimentally in this work and by Orkin et al.2 On one hand, Du et al.6 reported a positive activation energy of 3.3 kJ mol-1 using an Arrhenius expression (i.e, with n=0), while Balaganesh and Rajakumar7 stated that the energy barrier was of 1.1 kcal/mol (i.e, 4.7 kJ mol-1). We have fitted the

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 43

theoretical data from Balaganesh and Rajakumar7 between 200 and 400 K to the threeparameter expression, obtaining (±2σ): kOH (E- CF3CH=CHF)=(1.7±1.2)×10-13 298 T

2.7±0.8

467±206

exp 

T

 cm3 molecule-1 s-1 (10)

Even though the activation energy is –(3.9±1.8) kJ mol-1, in agreement with the experimental results, it must be noted that the difference in n with respect to ours is 11% but the pre-exponential factor A is more than twice ours, yielding to the fast increase in kOH(T) at high temperatures.

Figure 2. Rate coefficients determined in this work for E-CF3CH=CHF as a function of temperature together with results obtained in previous studies.

16 ACS Paragon Plus Environment

Page 17 of 43

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

3.1.3. Comparison with previous studies on the OH+ Z-CF3CH=CHF reaction As shown in Table 3, kOH(298 K) from the present work is in very good agreement with the previous experimental value reported by Nilsson et al.4, while that reported by Zhang et al.3 is in reasonable agreement (14% difference in the mean value with respect to ours). The theoretical calculation of kOH(298 K), 1.44×10-12 cm3 molecule-1 s-1, from Balaganesh and Rajakumar7 is 20% higher than ours (see Figure 3). As it can be seen in Figure 3, the temperature dependence of kOH(T) observed by Zhang et al.3 is fairly consistent with our results at T