Relative Rate Study of the Kinetics, Mechanism, and Thermodynamics

‡System Analytics and Environmental Science Department, Research and Innovation Center, Ford Motor Company, Mail Drop RIC-2122, Dearborn, Michigan 4...
0 downloads 10 Views 948KB Size
Article pubs.acs.org/JPCA

Relative Rate Study of the Kinetics, Mechanism, and Thermodynamics of the Reaction of Chlorine Atoms with CF3CFCH2 (HFO-1234yf) in 650−950 Torr of N2 or N2/O2 Diluent at 296−462 K E. W. Kaiser*,† and T. J. Wallington‡ †

Department of Natural Sciences, University of Michigan-Dearborn, 4901 Evergreen Road, Dearborn, Michigan 48128, United States System Analytics and Environmental Science Department, Research and Innovation Center, Ford Motor Company, Mail Drop RIC-2122, Dearborn, Michigan 48121-2053, United States



ABSTRACT: The rate constant of the reaction Cl + CF3CFCH2 (k1) has been measured relative to several reference species using the relative rate technique with either gas chromatographic analysis with flame-ionization detection (GC/FID) or Fourier transform infrared (FTIR) analysis. Cl atoms were generated by UV irradiation of Cl2/CF3CFCH2/reference/N2/O2 mixtures. At 300−400 K in the presence of >20 Torr O2, k1 = 1.2 × 10−11 e(+1100/RT) cm3 molecule−1 s−1. In N2 diluent, k1 has a sharp negative temperature coefficient resulting from the relatively small exothermicity of the following reactions: (1a) Cl + CF3CFCH2 ↔ CF3CFClCH2(•); (1b) Cl + CF3CF CH2 ↔ CF3CF(•)CH2Cl (reaction 1), which were determined in these experiments to be ∼16.5 (±2.0) kcal mol−1. This low exothermicity causes reaction 1 to become significantly reversible even at ambient temperature. The rate constant ratio for the reaction of the chloroalkyl radicals formed in reaction 1 with Cl2 (k2) or O2 (k3) was measured to be k2/k3 = 0.4 e−(3000/RT) for 300−400 K. At 300 K, k2/k3 = 0.0026. The reversibility of reaction 1 combined with the small value of k2/k3 leads to a sensitive dependence of k1 on the O2 concentration. Products measured by GC/FID as a function of temperature are CF3CFClCH2Cl, CF3COF, and CH2Cl2. The mechanism leading to these products is discussed. The rate constant for the reaction Cl + CF3CFClCH2Cl (k11) was measured as a function of temperature (300−462 K) at 760 Torr to be k11 = 8.2 × 10−12 e−(4065/RT) cm3 molecule−1 s−1. Rate constants relative to CH4 for the reactions of Cl with the reference compounds CH3Cl, CH2Cl2, and CHCl3 were measured at 470 K to resolve a literature discrepancy. (R = 1.986 cal K−1 mol−1).

1. INTRODUCTION The rate constant and mechanism for the addition of Cl atoms to the double bond in 2,3,3,3-tetrafluoropropene (CF3CFCH2, HFO-1234yf) have been examined previously at ambient temperature1,2 and over the temperature range 220−380 K.3 CF3CF CH2 has a global warming potential of 41 and is a possible substitute for HFC-134a as a refrigerant fluid.4 In the present work, we use the relative rate method to examine the rate constant of the potentially reversible addition reactions of Cl to CF3CFCH2 to form alkyl radicals both in the presence and absence of O2 over a temperature range of 296−462 K at 650−950 Torr:

Cl + CF3CF = CH2 ↔ CF3CFClCH2(•)

(1a)

Cl + CF3CF = CH2 ↔ CF3CF(•)CH2Cl

(1b)

If O2 is present in the reactant mixture, the chlorinated radicals formed in reaction 1 can also react with O2 by the potentially reversible reactions

(2a)

CF3CF(•)CH2Cl + Cl2 → CF3CFClCH2Cl + Cl

(2b)

© 2012 American Chemical Society

(3a)

CF3CF(•)CH2Cl + O2 ↔ CF3CF(O2 )CH2Cl

(3b)

Our results for reaction 1 will be compared to the rate constants derived in the references cited above and, more importantly, will also be used to examine the reaction mechanism and the thermodynamics of reaction 1 and the rate constant ratio k2/k3. The relative rate constants measured herein are derived using two analytical techniques. Gas chromatographic analysis with flame-ionization detection (GC/FID) is used with two different reactors to measure rate constants from ambient temperature to 462 K. Measurements have also been performed at ambient

The alkyl radicals formed in reaction 1 subsequently react with Cl2 present in the initial mixture to form a dichloride (CF3CFClCH2Cl) in the absence of O2 via the irreversible reactions CF3CFClCH2(•) + Cl2 → CF3CFClCH2Cl + Cl

CF3CFClCH2(•) + O2 ↔ CF3CFClCH2O2

Special Issue: A. R. Ravishankara Festschrift Received: November 7, 2011 Revised: January 7, 2012 Published: January 9, 2012 5958

dx.doi.org/10.1021/jp210692v | J. Phys. Chem. A 2012, 116, 5958−5971

The Journal of Physical Chemistry A

Article

Mixtures were also placed in the high-temperature reactor without irradiation to determine the degree of thermal reaction that may occur as the temperature is increased. No significant thermal reaction was observed at temperatures up to 462 K. Experiments were performed at ambient temperature in the high temperature reactor, in the 500 cm3 spherical Pyrex flask, and in an FTIR smog chamber as discussed later. The product yields and rate constants were indistinguishable to within experimental error in these three reactors both in the presence and absence of O2, verifying that surface reactions have little effect in these measurements. Identification and calibration of the products were carried out by injecting a known concentration of a pure product species into the GC in the presence of an internal calibration species. This provides a determination of the FID sensitivity and retention time. Pure samples of CF3CFClCH2Cl, CH2Cl2, and CF3COF are available for identification and calibration. CF3CFClCH2Cl is the sole product expected to be formed in the absence of O2. CH2Cl2 and CF3COF are two major products expected to form in the presence of low O2 concentrations as indicated by the mechanisms proposed in refs 2 and 3 as modified in Section 3.1.3. 2.1.2. Sample Stability in the Reactors and in the Sampling Syringe. To explore the effect of the gastight syringe on the measurements, test irradiations were run in the 500 cm3 reactor at ambient temperature in the presence of Cl2 and small amounts of O2. The conditions were chosen such that the three major products (CF3COF, CH2Cl2, and CF3CFClCH2Cl), the reactant (CF3CFCH2), and the calibration species (CF2ClH) were present in this reacted mixture at sufficient concentration to allow reliable measurements. The initial analysis was performed with the typical 20−30 s residence time in the syringe prior to the injection into the GC. A reanalysis was then performed with the residence time in the syringe increased by a factor of 3−25 (up to 12 min.). There was no change in the measured species concentrations to within the typical ±5% experimental error. Based on these results, the syringe sampling method does not introduce significant error. Stability in the reactors was tested in two ways. In the first method, a reaction was performed in the 500 cm3 reactor to generate the products described above. Reanalysis of the gas composition in the 500 cm3 reactor after times up to 2 hours showed no significant change for typical Cl2 mole fractions of 99%), and a kinetic reference compound in N2 (99.999% min) or N2/O2 (99.998%) diluent. Either C2H6 (99.999%), C2H5Cl (99.5%), CH3Cl (99.5%), or CF2ClH (>99%) was added as a reference compound in the relative rate experiments to measure the dependence of the overall rate coefficient for reaction 1 on temperature and the O2 mole fraction. The rate constants of these reference compounds and the rate constant for CH2Cl2, which is needed to correct for secondary consumption, are discussed in Appendix A. The newly measured rate constant for the reaction of Cl with CF3CFClCH2Cl, which is also required for correction of secondary consumption, is discussed in Appendix B. Freeze/ thaw degassing cycles were performed on all condensable reactants. In addition, CH4, CF2ClH, or CCl4 were included for internal calibration of the GC samples. The choice of internal calibrant was such that the consumption of this species was 99%. In smog chamber experiments, it is important to consider potential loss of reactants and products via heterogeneous reactions. As a check for such losses, reaction mixtures were left to stand in the dark in the chamber for 30 min; there was no discernible loss (650 K in these two papers do not agree very well, particularly at elevated temperature. For this reason, a brief series of relative rate experiments has been performed at 470 K to test the internal consistency of these two sets of rate constants. Table 2 presents a comparison Table 2. Rate Constant Ratios Determined at 470 K for CH3Cl, CH2Cl2, and CHCl3 Relative to CH4 species

k/kCH4 (this work)

k/kCH4 (ref 17)

k/kCH4 (ref 19)

CH3Cl CH2Cl2 CHCl3

3.12 1.94 0.58

3.05 1.65 0.55

4.22 3.43 1.57

of our results to relative rate constant ratios calculated at 470 K using the rate constant expressions in refs 17 and 19 for the three chloromethanes relative to CH4. The ratios determined in our experiments agree well with those determined by Bryukov et al.17 The ratios determined from the data of Clyne and Walker19 show relatively poor agreement with our measurements deviating by up to a factor of 3. This indicates that the data of Clyne and Walker are not internally consistent with our data. For this reason, we have chosen to use the temperature-dependent rate constants of Bryukov et al.17 for rate constant determinations and for secondary consumption corrections of the chloromethanes.



APPENDIX B: RATE CONSTANT FOR REACTION 11 (Cl + CF3ClCH2Cl) FROM 297 TO 464 K The rate constant for the hydrogen abstraction reaction Cl + CF3CFClCH2Cl was measured relative to the rate constant of the reaction CF2ClH in N2 to provide the temperaturedependent rate constant needed to correct for secondary consumption of this major product species at elevated temperature. The experimental conditions for these experiments were Cl2 = 700−900 ppm, CF3CFClCH2Cl = 50−60 ppm, CF2ClH = 105−115 ppm, and pressure = 600−760 Torr of N2. CCl4 was used as the internal calibrant in most but not all of these experiments, spanning the full temperature range. The other experiments had no internal calibrant. Experiments at 5970

dx.doi.org/10.1021/jp210692v | J. Phys. Chem. A 2012, 116, 5958−5971

The Journal of Physical Chemistry A

■ ■

ambient temperature were performed in the 500 cm3 reactor. Figure 7 presents the results of these experiments. On the basis of the relative rate ratios and the rate constant for Cl + CF2ClH in Appendix A, the measured rate constant is k11 = 8.2 × 10−12 e(−4065/RT). This rate constant was used to correct the experimental data for secondary consumption of CF3CFClCH2Cl where necessary using the Acuchem chemical kinetics computer program.14

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected].

ACKNOWLEDGMENTS E.W.K. thanks Prof. Craig Donahue, Prof. Ali Bazzi, and the University of Michigan-Dearborn for invaluable assistance during these experiments. We thank Rajiv Singh (Honeywell) for providing the sample of CF3CFClCH2Cl used in these experiments.



APPENDIX C: DISCUSSION OF THE ASSUMPTION THAT k2a/k3a = k2b/k3b The data are not sufficient to prove that k2a/k3a = k2b/k3b as assumed in section 3.1.4, but a strong case can be made that both ratios are ≪1 at ambient temperature. Because of the reversibility of reactions 1a and 1b and their different calculated heats of reaction, in the total absence of O2, the effective rate of reaction 1a will be slowed dramatically relative to 1b (see last paragraph of section 3.2.1). This is because the CF3CFCH2Cl radical formed from 1b is calculated to be more stable than the CF3CFClCH2 formed from 1a.3 This would cause the steadystate concentrations of the two radicals to be different even though the rate constants of the forward reactions 1a and 1b are nearly equal as deduced from measurements in air2,3 (see discussion in the last paragraph of section 3.1.3). However, all measurements of k2/k3 are performed in the presence of O2, and, at ambient temperature, the data in Figure 2 show that for O2 > 200 ppm the effective rate constant k1(eff) remains essentially constant as the O2 mole fraction increases. This indicates that little reversibility is present in either reaction 1a or 1b for O2 = 200 ppm or greater since O2 reacts with both radicals formed in these reactions fast enough to prevent appreciable reverse reaction. Most measurements of k2/k3 at ambient temperature presented in Figure 6 were carried out with O2 > 200 ppm. Therefore, the steady-state concentrations of the CF3CFCH2Cl and CF3CFClCH2 radicals will be similar during the reaction process provided k3a ∼ k3b. As discussed near the end of section 3.1.4, the rate constants for the reaction of O2 with CF3 and with CF3CHF radicals are similar [(3 ± 1) × 10−12 cm3 molecule−1 s−1], and it is likely that this will also be true for the CF3CFCH2Cl and CF3CFClCH2 radicals. The yield of the dichloride, CF3CFClCH2Cl, will then depend primarily on the rate constant ratios k2a/k3a and k2b/k3b. Using experiment 7 in Table 1 as an example, the value obtained assuming k2a/k3a = k2b/k3b is k2/k3 = 0.00274. Assuming that all of the dichloride comes from only one of the radicals (e.g., assume k2a/k3a = 0, implying no reaction occurs between this radical and Cl2), the dichloride yield from channel 1b will actually be twice the yield presented in Table 1 or 3.3%. Applying eq A from section 3.1.4 (k2/k3 = [O2]/[Cl2] {[Y(N2)/Y(O2)] − 1}−1) to this yield results in k2b/k3b = 0.00557. Making the drastic assumption that one of the ratios is zero results in a ratio for the other radical, formed in ∼50% yield during reaction 1, only twice that obtained assuming that k2a/k3a = k2b/k3b. We believe it is more likely that the two ratios will be similar. If they do differ significantly, the ratio k2/k3 for the radical that is more reactive toward Cl2 at ambient temperature will be less than twice that calculated under the k2a/k3a = k2b/k3b assumption. Thus, assuming that k2a/k3a = k2b/k3b seems very reasonable, and under no circumstance will either of the ratios be greater than twice that calculated under this assumption at ambient temperature.



REFERENCES

(1) Nielsen, O. J.; Javadi, M. S.; Sulbaek Andersen, M. P.; Hurley, M. D.; Wallington, T. J.; Singh, R. Chem. Phys. Lett. 2007, 439, 18−22. (2) Hurley, M. D.; Wallington, T. J.; Javadi, M. S.; Nielsen, O. J. Chem. Phys. Lett. 2008, 450, 263−267. (3) Papadimitriou, V. C.; Lazarou, Y. G.; Talukdar, R. K.; Burkholder, J. B. J. Phys. Chem. A 2011, 115, 167−181. (4) Brown, J. S. ASHRAE J. 2009, 51, 22−29. (5) Kaiser, E. W.; Wallington, T. J.; Hurley, M. D. J. Phys. Chem. A 2009, 113, 2424−2437. (6) Wallington, T. J.; Gierczak, C. A.; Ball, J. C.; Japar, S. M. Int. J. Chem. Kinet. 1989, 21, 1077. (7) Kaiser, E. W.; Pala, I. R.; Wallington, T. J. J. Phys. Chem. A 2010, 114, 6850−6860. (8) Kaiser, E. W.; Wallington, T. J. Chem. Phys. Lett. 2011, 501, 187−192. (9) Kaiser, E. W. Int. J. Chem. Kinet. 1993, 25, 667−680. (10) Bilde, M.; Sehested, J.; Nielsen, O. J.; Wallington, T. J.; Meagher, R. J.; McIntosh, M. E.; Piety, C. A.; Nicovich, J. M.; Wine, P. H. J. Phys. Chem. A 1997, 101, 8035−8041. (11) Maricq, M. M.; Szente, J. J.; Kaiser, E. W. Chem. Phys. Lett. 1992, 197, 149−156. (12) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.; Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina, M. J. JPL Publication 97-4; Jet Propulsion Laboratory, California Institute of Technology: Pasadena, CA, 1997. (13) Kaiser, E. W.; Wallington, T. J.; Hurley, M. D. Int. J. Chem. Kinet. 1995, 27, 205−218. (14) Braun, W.; Herron, J. T.; Kahaner, D. K. Int. J. Chemical. Kinet. 1988, 20, 51−62. (15) Bryukov, M. G.; Slagle, I. R.; Knyazev, V. D. J. Phys. Chem. A 2003, 107, 6565−6573. (16) Wine, P. H.; Semmes, D. H. J. Phys. Chem. 1983, 87, 3572− 3578. (17) Bryukov, M. G.; Slagle, I. R.; Knyazev, V. D. J. Phys. Chem. A 2002, 106, 10532−10542. (18) Talhaoui, A.; Louis, F.; Meriaux, B.; Devolder, P.; Sawerysyn, J.-P. J. Phys. Chem. 1996, 100, 2107−2113. (19) Clyne, M. A. A.; Walker, R. F. J. Chem. Soc., Faraday Trans. 1 1973, 69, 1547−1567.

5971

dx.doi.org/10.1021/jp210692v | J. Phys. Chem. A 2012, 116, 5958−5971