Reinvestigation of the Unimolecular Reactions of ... - ACS Publications

Oct 31, 2016 - Department of Chemistry, University of North Carolina-Asheville, One University Heights, Asheville, North Carolina 28804-8511,. United ...
0 downloads 0 Views 462KB Size
Subscriber access provided by McMaster University Library

Article 2

2

Reinvestigation of the Unimolecular Reactions of CHFCHF: Identification of the 1,1-HF Elimination Component from Addition of CHFCF to trans-2-Butene 2

Caleb A Smith, George L. Heard, Donald W. Setser, and Bert E. Holmes J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.6b07744 • Publication Date (Web): 31 Oct 2016 Downloaded from http://pubs.acs.org on November 17, 2016

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 19

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

Reinvestigation of the Unimolecular Reactions of CHF2CHF2: Identification of the 1,1-HF Elimination Component from Addition of :CFCHF2 to trans-2-Butene. Caleb A. Smith, George L. Heard, D. W. Setser# and Bert E. Holmes* Department of Chemistry, University of North Carolina-Asheville, One University Heights, Asheville, NC 28804-8511, United States and # Kansas State University, Manhattan, Kansas, 66506, United States. ABSTRACT The recombination of •CHF2 radicals in a room temperature bath gas was used to generate CHF2CHF2* molecules with 96 kcal mol-1 of vibrational energy. The CHF2CHF2* molecules decompose by four-centered 1,2-HF elimination and by three-centered 1,1-HF elimination reactions to give HF and either CHF=CF2 or :CFCHF2, respectively. The 1,1-HF component was identified by trapping the :CFCHF2 carbene with trans-2-butene that forms 1-fluoro-1difluoromethyl-2,3-dimethylcyclopropane. The total rate constant for the decomposition of CHF2CHF2* was 6.0 x 105 s-1 and the rate constant for the 1,1-HF pathway forming the carbene, as measured by the 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane yield, was 1.4 x 105 s-1. Based upon matching the experimental rate constants to calculated statistical rate constants, the threshold energies for the four-centered and three-centered reactions are 78 and ≤ 85 kcal mol-1, respectively. 1. INTRODUCTION Numerous thermal1 and chemical activation2-5 experiments have demonstrated that the threshold energy for 1,2-HF elimination from fluorinated alkanes increases with the number of fluorine atoms attached to the two carbon atoms involved in the four-membered ring of the transition state. If two F-atoms are located on the terminal carbon atom of the fluoroalkane, 1,1-HF elimination via a three-membered ring can become competitive with 1,2-HF elimination for high temperature or high vibrational energies. This competition exists because the low vibrational frequencies of the 1,1-HF transition state compensate for the higher threshold energy for the 1,1-HF process. This competition has been experimentally documented from chemical activation studies of CD3CHF25, C2D5CHF25, CD3CHFCl6, and CD2FCHF23. For all of these cases, the 1,1-HF channel was identified from the fluoropropene or fluoroethene corresponding to the D-atom migration in the carbene following 1,1-HF elimination. The 1,1-HF component from CHF2CHF2 and CF3CHF2

ACS Paragon Plus Environment

1

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 19

has not been identified, because the deuterium labeling technique cannot be utilized. However, :CFCHF2 and :CFCF3 have slow rates of isomerization, and chemical trapping techniques can be used to identify the 1,1-HF component in carefully designed experiments.7-10 In the present study we have used trans-2-butene to trap the :CFCHF2 carbene from chemically activated CHF2CHF2* formed by the recombination of •CHF2 radicals in a room temperature bath gas. This study is a continuation of the work described in the report of the reactions of C2D5CHF25, that report contains the analysis of the thermochemistry and the computational background needed to describe the reactions of CHF2CHF2. In the chemical activation study of the C2D5CHF2* system5, •CHF2 radicals were needed to generate C2D5CHF2*, and preliminary efforts were made to observe the unimolecular reactions of the CHF2CHF2*molecules, which contain 96 kcal mol-1 of vibrational energy. Experiments with added cis-2-butene suggested that some of the :CFCHF2 carbene from 1,1-HF elimination could be trapped as a cyclopropane adduct, which was identified from a mass spectrum obtained from gas chromatographic analysis with a mass spectrometer as a detector(GC-MS) . However, at that time we did not have a method to calibrate the analysis for the ratio of the cyclopropane adduct to the collisionally stabilized CHF2CHF2 or the CHF=CF2 products. In the present experiments we have used trans-2-butene as the trapping agent in order to avoid the cis- and trans-isomer complication of the cyclopropane adduct, and we have devised a calibration method for the GCMS analysis. We also acquired a new decomposition to stabilization plot of CHF=CF2/CHF2CHF2 to obtain the experimental rate constant for CHF=CF2 formation. The new plot served to identify an error in the data published in reference 5. A mistake had been made in obtaining the GC-MS calibration factor for the CF2=CHF/CHF2CHF2 ratio, which gave a rate constant for CF2=CHF formation that was too small by a factor of 2.8. In the present study, we have selected CHF2I as the photolytic system to generate •CHF2 radicals in the presence of trans-2-butene. The principal reactions are shown below; M denotes a bath gas molecule and an asterisk denotes vibrational excitation. The disproportionation to combination ratio11 of •CHF2 radicals is 0.15. The :CF2 will mainly recombine to C2F4 or combine with •CHF2 to give •CF2CHF2, which will abstract an Iatom from CHF2I or recombine with another •CHF2 radical. The free-radical chemistry in this system is much simpler than in the scheme5 used for the C2D5CHF2* study. CHF2I + hν

→ •CHF2 + •I

ACS Paragon Plus Environment

1.

2

Page 3 of 19

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

2 •CHF2

CHF2CHF2*

+ M :CFCHF2 + trans-2-C4H8 :CFCHF2

→ CHF2CHF2*

2a.

→ :CF2 + CH2F2

2b.

→ HF + CF2=CHF

3a.

→ HF + :CFCHF2

3b.

→ CHF2CHF2 + M

3c.

→ cyclo-CHF2- CF-C4H8

4.

→ CF2=CHF

5.

Trans-2-butene was selected as the trapping agent based on the work of Haszeldine and coworkers7-10, who studied the gas phase reactions of :CFCHF2 with numerous substrate molecules at 1500 C. They always found a competition between unimolecular reaction by H-atom rearrangement and bimolecular reaction with the substrate. The calculated (B3PW91/6311+G(2d,p))5 potential energy barrier for rearrangement of :CFCHF2, (eq 5) is 10 kcal mol-1. The excess energy released to HF + :CFCHF2 is approximately 18 kcal mol-1, and this energy will be distributed to the two products. The energy disposal will be analyzed more thoroughly in the Discussion Section. However, the possibility of isomerization of :CFCHF2, reaction 5, should be remembered, and the cyclopropane adduct from reaction 4 will be a lower limit measurement to the actual 1,1-HF process. The experiments consist of photolysis of CHF2I with and without trans-2-butene for a range of pressures followed by GC-MS or GC-FID analysis for CHF2CHF2, CF2=CHF and 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane. The experimental rate constants are compared to the calculated statistical rate constants to assign threshold energies. The molecular models used for the calculations of the rate constants are the same as employed in reference 5. Those transition states were based on electronic structure calculations using Density Functional methods with the Gaussian 09 code. In particular, B3PW91/cc-pVDZ was found to be satisfactory for describing the 1,2- and 1,1-HF elimination transition states. A detailed description of the models was provided in ref. 5. Those calculated rate constants were confirmed to be correct in the present study. 2. EXPRIMENTAL METHODS AND RESULTS.

ACS Paragon Plus Environment

3

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 19

The experimental measurements consisted of three series of photolytic experiments at room temperature; one with just CHF2I, the second with added trans-2-butene in a 1:2 ratio of trans-2butene to CHF2I, and the third with a 10 fold excess of trans-2-butene relative to CHF2I. For all experiments a small amount of Hg2I2 was added to control the free iodine.12 Reagents were measured on a grease-free vacuum system and placed in Pyrex vessels of known volume to achieve the desired pressures. Irradiation times depended on vessel volume and were between 25 minutes using an Oriel 8510-4 high-pressure mercury lamp, which typically gave less than 10% conversion of the CHF2I to products. After photolysis, the contents of the vessels were frozen using liquid N2 into an injection loop of a 6-port Valco gas value connected to a 0.25 mm diameter by 105 m long RTX-624 column contained in Shimadzu QP-5000 GC-MS. The GC operating conditions were an initial temperature of 30°C for a period of 10 minutes, at which point the temperature increased at a rate of 4°C/min. to a final temperature of 150°C. The GCMS interface temperature was 250°C and UHP helium carrier gas was used with the following parameters: column inlet pressure of 172.4 kPa, a column flow of 1.00 mL/min, linear velocity of 19.2 cm/sec, and the split ratio was 10. The MS used electron impact ionization set at 70eV. Calibrations of the response of the GC-MS were made from pre-prepared mixtures made to replicate the photolytic samples. The results from each series of experiments are described separately. The experiments with CHF2I alone were done to measure the yield of the product stabilized by collision ([CHF2CHF2] = S) and the product from unimolecular decomposition ([CHF=CF2] = D). The D/S ratios were measured for a range of pressures in order to obtain the experimental rate constant for 1,2-HF elimination; kexp/kM[M] =D/S, where kM is the collision rate constant. The data are shown in Figure 1 and Figure 2. The three plots of [CHF=CF2]/ [CHF2CHF2] = D/S versus P-1 are linear and passes through the origin, and the mean slope is 0.038 ± 0.002 Torr. Evaluation of the collision rate constant using the collision diameters and ε/k values given in the caption of Figure 1 gives a rate constant of kexp = 4.6 ± 0.5 x 105 s-1.. The three measurement of the D/S ratio have slopes, see Figures 1 and 2 and the captions, of 0.0360, 0.0407 and 0.0385 and illustrate that the slope is not effected by added trans-2-butene or using either the total ion-count (TIC) or m/z = 83 for the yield of the eluted CHF2CHF2. In all plots m/z = 82 was used for CHF=CF2. All data used to construct Figures 1 and 2 are given in Supporting Information. The

ACS Paragon Plus Environment

4

Page 5 of 19

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

calibration factor for the GC-MS needed to convert the apparent ratio to a true concentration ratio was obtained from a pre-prepared mixture of known composition, and the average value was 1.416 ± 0.12 for comparison of m/z = 82/83 and 11.212 ± 0.097 for 82/TIC. The raw data from the current measurements and those reported in Ref. 5 are in agreement; however, the calibration factor used in Ref 5 was incorrect, and 4.6 x 105 s-1 is the correct rate constant for the formation of CHF=CF2, which is the 1,2-HF rate constant plus any contribution from :CFCHF2 that might have isomerized. A small impurity that was present in the trans-2-butene sample overlapped the eluted CHF2CHF2, and mass 83 was used to monitor CHF2CHF2 in experiments with trans-2-butene, see Supporting Information for GC traces. Therefore, the calibration using mass 83 for CHF2CHF2 was tested using the D/S data in Figure 1, and those points also are shown in the plot. The agreement between the two calibration methods is satisfactory. In the prior study5 with C2D5CHF2, we had identified the cyclopropane adduct from the addition of :CFCHF2 with cis-2-butene from the mass spectrum of a product that matched a report from reference 10. We decided to use trans-2-butene as the trapping agent in the present study, because 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane consists of a single isomer; the mass spectrum of 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane is shown in Table 1. In the future we will frequently refer to the 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane as the cyclopropane adduct. In the first series of experiments a 1:2 ratio of trans-2-butene to CHF2I was used, and the total pressure ranged from 0.02 to 0.15 torr. A 10:1 ratio of trans-2-butene to CHF2I was used for the last set of experiments. Due to the restriction associated with size of the photolysis vessels, the pressure range for the second series of experiments was 0.5 to 0.04 torr. The 10:1 data set is of better quality than the 1:2 series, because we had improved the experimental technique during the course of the study. However, no significant distinction exists between the two data sets, and all the data were combined to assign slopes to the two D/S plots shown in Fig. 2. The rate constant for formation of CHF=CF2 with trans-2-butene as the bath gas was essentially the same as that with CHF2I as the bath gas, which was shown in Figure 1. The slopes from the D/S plots of the cyclopropane adduct were 0.010 ± 0.002 (GC-MS calibration) and 0.012 ± 0.001 (GC-FID calibration) torr. The slopes from Figure 2 give a branching ratio of 0.31 for 1,1-HF elimination vs. 1,2-HF elimination. Combining this branching ratio with the rate constant for CHF=CF2 formation from Figure 1 gives a rate constant of 1.4 x 105 s-1 for 1,1-HF elimination. In principle12, ring-opening of 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane

ACS Paragon Plus Environment

5

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 19

formed by addition of the carbene to trans-2-butene could form hexenes. We found no experimental evidence for hexenes in the GC-MS records. Furthermore, if there were significant isomerization, the product ratio of 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane to CHF=CF2 would not be constant with pressure. Based upon these experimental observations, we conclude that isomerization of the cyclopropane adduct was not important for our experimental conditions. The reliability of the 1,1-HF rate constant depends on the calibrations for the GC-MS response, and those efforts will be described in some detail. A sample of 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane was not available to calibrate the GC-MS, and it was necessary to use proxy molecules for calibration of the [cyclopropane adduct]/[CHF2CHF2] and [cyclopropane adduct]/[CHF=CF2] ratios. We selected 1,1,1trifluorohexane and 1,1,1-trifluoro-3-methylpentane as proxies, because they have the same number of carbon and fluorine atoms as the cyclopropane adduct molecule. The mass spectrum of these two molecules is compared to the 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane in Table 1. The total ion-count (TIC) of the two C6F3H11 molecules differed by only ± 5%. The average of the total ion-count from these molecules was assumed to be equal to that of the cyclopropane adduct for calibration of the cyclopropane adduct/CHF2CHF2 ratio. In order to confirm the results from the GC-MS measurements, experiments were done with analysis by flame ionization (FID) as the detector for the GC. Based upon considerable experience (see references 5 and 12 and references therein), we expected the FID response for CHF2CFCH(CH3)CH(CH3) to be similar to that for CF3(CH2)4CH3 and they differed by less than 4%. Several photolytic experiments were done, see Figure 2, and mixtures of CHF2CHF2 and CF3(CH)4CH3 were prepared to obtain the calibration factor. In addition, the response of the GCFID for 1,2-dimethylcyclopropane was compared to n-pentane to account for the smaller number of H-atoms and the ring strain in the response of a substituted cyclopropane. The linear C-5 compound had a 18.5 ± 1.5% larger FID response than the 1,2-dimethylcyclopropane. The D/S points from the GC-FID analysis shown in Figure 2 are in close agreement with the results based on total ion-count from the GC-MS analysis. The uncertainty in the calibration based on the surrogate molecules contributes to the statistical uncertainty of the rate constant measurements for 1,1-HF elimination, and we estimate an overall uncertainty of ± 20%.

ACS Paragon Plus Environment

6

Page 7 of 19

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

Since formation of :CFCHF2 compromises about 23% of the decomposition of CHF2CHF2*, the chemical reactions of :CFCHF2 in the absence of a trapping agent are of general interest. Since a 1:2 ratio of trans-2-butene to CHF2I was sufficient to trap :CFCHF2, the reaction rate of :CFCHF2 with CHF2I must be relatively slow. One possible reaction for :CFCHF2 is with CHF=CF2, according to the studies of Haszeldine7-10 at higher temperature. Therefore, we searched for the presence of the addition product 1,2,2,3-tetrafluoro-1-difluoromethyl-cyclopropane and the likely decomposition product (:CF2 + CHF= CFCHF2) in experiments with added CHF=CF2. We found no evidence for either of these compounds. For our experimental conditions the carbene seems to recombine with •CHF2 to give •CF(CF2H)2 radicals, which can decompose via loss of an H-atom to give CF2=CFCHF2 or react with another •CHF2 radical to give CF2HCHFCF2H or (CF2H)3CF. The lack of pure compounds for identification of these possible products and the relatively small yields at the required low pressures prevented a more quantitative study. 3. DISCUSSION The trapping experiments with trans-2-butene provide a lower limit of 0.31 to the branching ratio for 1,1-HF to 1,2-HF elimination for CHF2CHF2, assuming that our calibration for 1-fluoro-1difluoromethane-2,3-dimethylcyclopropane is reliable. Based on the experimental rate constants and the calculated statistical rate constants from ref. 5, the threshold energies correspond to 78 and 85 kcal mol-1 for 1,2- and 1,1-HF elimination, respectively. However, the possibility that some of the :CFCHF2 carbene may actually have isomerized to CHF=CF2 needs to be evaluated. The excess energy (96-85 = 11 kcal mol-1) at the transition state will be distributed statistically among the vibrational modes, and the average fraction retained as vibrational energy by :CFCHF2 will be 7-8 kcal mol-1. The potential energy (≈ 3-5 kcal mol-1) released in passing from the transition state to HF + :CFCHF2 will go mainly to relative translational energy.14 According to this qualitative analysis, a small fraction of the energy distribution for :CFCHF2 could extend above the 10 kcal mol-1 barrier or be sufficient for tunneling and; thus, contribute to the yield of CHF=CF2. For this reason we believe that 0.31 is a lower limit to the branching ratio and that 85 kcal mol-1 is an upper limit to E0(1,1-HF). The thermochemical lower bound to the formation17 of HF + :CFCHF2 is 79 kcal mol-1. For a 3 kcal mol-1 threshold energy for addition of :CFCHF2 to HF the lower limit becomes 82 kcal mol-1. The E0 assignments also depend on the . The selfconsistent set of thermochemistry values for C1 and C2 fluorinated species from Haworth et al.17

ACS Paragon Plus Environment

7

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 19

would suggest that could be 98 kcal mol-1, which would lower our E0 assignments by 1 kcal mol-1. The third possible reaction of CHF2CHF2 with 96 kcal mol-1 of energy is the rearrangement to CH2F2 + :CF2. There was no experimental support for a pressure dependent yield of CH2CF2 yield. We also examined this possibility by evaluating the threshold energy by computations with the B3PW91/cc-pVDZ method. The calculated threshold energy was more than 10 kcal mol-1 higher than E0(1,1-HF), and this reaction need not be considered for CHF2CHF2*. The computed transition state geometry for this reaction is given in Supporting Information. The calculated E0(1,1-HF) according to the B3PW91/cc-pVDZ method was 77 kcal mol-1. There was little variation across basis sets so the cc-pVDZ basis set was used for further calculations. The MP2 method gave higher E0 (1,1-HF) values ≈ 84 kcal mol-1, but those E0(1,2-HF) results were unreasonably high. The calculated E0(1,1-HF) by M06-2X and CAM-B3LYP methods with the cc-pvDZ basis set were 82.5 and 79 kcal mol-1, respectively. More work is needed to identify the best method and basis sets for calculating simultaneously the threshold energies for 1,1-HF and 1,2-HF elimination reactions, although all methods give very similar structures for the transition states. The structures of the transition states for 1,1-HF and 1,2-HF elimination from CHF2CHF2 were discussed in ref. 5. Based upon comparison to the rate constant (2.8 x 105 s-1)12 for the cyclopropane formed from the reaction of :CH2 with CF3CH2CH2CH=CH2, some isomerization of CHF2-cycloCFCH(CH3)CH(CH3) could have been anticipated. As already noted in the Experimental Section, no evidence for isomerization was observed. The explanation is that the addition of :CFCHF2 to trans-2-butene is less exoergic than the reaction of :CH2 with olefins. DFT calculation for the addition of :CFCHF2 to ethene gave an enthalpy of reaction which is at least 20 kcal mol-1 lower than for the reaction with methylene, verifying that no decomposition of the cyclopropane would occur. Haszeldine and coworkers10 attempted to measure the unimolecular rate of isomerization of :CFCHF2 in competition with bimolecular addition reactions with added substrate molecules over the 440-623 K temperature range. The source of the carbene was the pyrolysis of trifluoro1,1,2,2-tetrafluoroethylsilane. From indirect measurements of the isomerization rate constant vs.

ACS Paragon Plus Environment

8

Page 9 of 19

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

temperature, they assigned 23 kcal mol-1 to the activation energy for the isomerization of :CFCHF2. However, the DFT calculated5 (B3PW91/6-311+G(2d,p)) value is 9.9 kcal mol-1. Calculations for the isomerization energy for various fluoroethylidines5,6,14,15 by different computational methods give consistent results, and the calculated threshold energy of 10 kcal mol-1 for isomerization of :CFCHF2 can be accepted. The small rate constant for 1,2-HF elimination yielding CHF=CF2 from chemically activated CHF2CHF2* has been verified by four independent experiments2,5, including this work. The assigned threshold energy of 78 kcal mol-1 is ≈ 10 kcal higher than for CH3CF3 or CH2FCHF2.2 This large increase in threshold energy for 1,2-HF elimination enables the 1,1-HF elimination to compete with the 1,2-HF process from CHF2CHF2. The systematic increase in threshold energy for 1,2-HF elimination with the number of F-atoms in the molecule (about 5 kcal mol-1 per Fatom) correlates with the decrease in bond dissociation energy of the corresponding product olefin, which, for example, is 190, 140 and 110 kcal mol-1 for CH2= CH2, CH2=CF2, and CHF=CF2, respectively. DFT calculations for 1,2-HF elimination from fluoroethanes generally follow the trend of increasing threshold energies with number of F-atoms. A survey of methods (B3LYP, B3PW-91, M06, CAM-B3LYP) with several basis sets (6-31G(d',p'), 6-311++G(2d,p), cc-pVDZ, cc-pVTZ, 6-311++G(3df,3pd)) gave 1,2-HF threshold energies for CHF2CHF2 ranging from 75 to 79 kcal mol-1. These results are similar to the exhaustive computational survey18 for the reaction of CH3CF3. The small rate constants for 1,1-HF and 1,2-HF elimination from chemically activated CHF2CHF2* molecules are mainly the result of high threshold energies; however, another contributing factor is the enhanced density of states of the molecule when Fatoms are exchanged for H-atoms.12 4. CONCLUSIONS The total unimolecular rate constant of CHF2CHF2* with 96 kcal mol-1 of vibrational energy was measured as 6.0 ± 0.6 x 105 s-1. The branching fractions for 1,1-HF and 1,2-HF elimination were 0.23 and 0.77, respectively. This branching fraction for 1,1-HF elimination may be a lower limit, if some of the :CFCHF2 carbenes isomerized to CHF=CF2. Because nearly one-fourth of CHF2CHF2 molecules decompose by the 1,1-HF elimination pathway scientists developing models and procedures for destruction of HCFCs and HFCs that contain a terminal carbon with the –CHF2 structure must account for the production and subsequent reactions of fluorocarbenes.

ACS Paragon Plus Environment

9

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 19

The 1,1-HF channel was identified by trapping the :CFCHF2 carbene with trans-2-butene, and the rate constant was evaluated from measuring the 1-difluoromethyl-1-fluoro- 2,3dimethylcyclopropane/CHF2CHF2 ratio versus pressure-1. Fitting these rate constants to calculated statistical RRKM rate constants of ref. 5 gave threshold energies of 78 and 85 kcal mol-1 for 1,2-HF and 1,1-HF elimination, respectively. A lower limit of 82 kcal mol-1 can be estimated for 1,1-HF elimination from the enthalpy of formation of HF + :CFCHF2 plus 3 kcal-1 as the threshold energy for addition of HF to the carbene. These threshold energies are considerably higher than those for CH3CF3 or CH2FCHF2. The rate constant reported in reference 5 for the formation of CHF=CF2 from CHF2CHF2* was revised upward because the GC-MS calibration for the CHF=CF2/CHF2CHF2 ratio that was used in that work was incorrect. 5. AUTHOR INFORMATION Corresponding Author. * BEH E-mail: [email protected] Telephone: 828-232-5168 6. ACKNOWLDEGMENTS. Financial support from the National Science Foundation (CHE-1111546 and CHE-1229406) is gratefully acknowledged. 7. SUPPORTING INFORMATION. Supporting Information contains the calculated transition state geometry using M062X/cc-pVDZ for the reaction of CHF2CHF2 forming CF2H2 and :CF2. The experimental data tables used to construct Figures 1 and 2 are given. Sample GC chromatograms for runs with 2:1 ratio of trans-2-butene:CHF2I, 10:1 ratio of trans2-butene:CHF2I and with no trans-2-butene are also shown. This information is available free of charge via the Internet at http://pubs.acs.org.

ACS Paragon Plus Environment

10

Page 11 of 19

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

Table 1. Mass-Spectral fragmentation data at 70eV (mass charge ratio and relative abundance, R.A.). 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropanea m/z

R.A.

Ion

59

100

C3H4F+

27

88

C2H3+

41

54

C3H5+

87

51

C4HF2+

39

47

C3H3+

51

26

CHF2+

26

25

C2H2+

77

19

C3F2H3+

31

16

CF+

a) The two adducts formed by the addition of CHF2CF: to cis-2-butene had virtually identical fragmentation patterns and were observed at 36.6 and 42.1 minutes (see the experimental section) in a 2.28 ± 0.26 ratio, respectively. Three distinctive ion fragments were the parent ion, M+ = 138 with RA = 2, m/z = 123 with RA = 4 from M+ - CH3 and m/z = 103 with RA = 7 from M+ -HF and -CH3. 1,1,1-Trifluorohexane m/z

R.A.

Ion

43

100

C3H7+

27

53

C2H3+

41

52

C3H5+

29

47

C2H5+

39

26

C3H3+

28

21

C2H4+

47

17

C2H4F+

ACS Paragon Plus Environment

11

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 19

56

16

C4H8+

55

13

C4H7+

42

11

C3H6+

69

9

CF3+

1,1,1-Trifluoro-3-methylpentane m/z

R.A.

Ion

57

100

C4H9+

29

73

C2H5+

41

65

C3H5+

27

58

C2H3+

47

39

C2H4F+

39

37

C3H3+

28

27

C2H4+

91

23

C4F2H5+

77

14

C3F2H3+

61

12

C3FH6+

56

12

C4H8+

ACS Paragon Plus Environment

12

Page 13 of 19

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

FIGURE CAPTIONS Figure 1. Plot of CHF=CF2/CHF2CHF2 vs. pressure-1 from photolysis of CHF2I. Two sets of data points are shown corresponding to whether the total ion-count ( and solid line) or mass 83 (☐ and dotted line) from CHF2CHF2 was used to convert the raw product ratios from the GC-MS measurements to actual concentration ratios of the products. As explained in the text and Supporting Information, it was necessary to use mass 83 for the GC-MS experiments with trans-2-butene shown in Figure 2. The slope and intercept values are 0.0360 ± 0.0019 and 0.038 ± 0 0.105 () and 0.0407 ± 0.0015 and -0.055 ± 0.084 (☐). The collision rate constant, kM, was calculated from the following collision diameters and ε/k values: CHF2I (4.6 Å; 298 K), CHF2CHF2 (5.2 Å; 201 K); kM = 3.96 x 10-10 cm3 molecule-1 s-1. Figure 2. Plots of CHF=CF2/CHF2CHF2 and 1-fluoro-1-difluoromethyl-2,3-dimethylcyclopropane /CHF2CHF2 vs pressure-1 from photolysis of CHF2I with added trans-2-butene. The data include experiments with 1:2 and 10:1 mixtures of trans-2-butene to CHF2I using GC-MS analysis for CHF=CF2/CHF2CHF2; (● and solid line) and (cyclopropane adduct)/CHF2CHF2 (

☐ and dotted line). Data from photolysis of 10:1 mixtures using GC-FID analysis are shown for cyclopropane/CHF2CHF2 ( and dashed line). The slopes and intercept values are 0.0101 ± 0.0004 and 0.0186 ± 0 0.0047 ( and dashed line), 0.0121 ± 0.0009 and 0.0037 ± 0.0155 (☐ and dotted line) and 0.0385 ± 0.0007 and -0.009 ± 0.009 (● and solid line).

ACS Paragon Plus Environment

13

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 19

References 1. Robinson, P. J.; Holbrook, K. A. Unimolecular Reactions, 1972, Wiley-Interscience, N Y, see Table 7.20. 2. Holmes, B. E.; Setser, D. W.; Pritchard, G. O. Energy Disposal in the Three-Centered Elimination of DF from 1,1,2-Trifluoroethane-1-D1. Int. J. Chem. Kinet. 1976, 8, 215-234. 3.Duncan, J. R.; Roach, M. S.; Stiles, B. S.; Holmes, B. E. Unimolecular Rate Constant and Threshold Energy for the HF Elimination from Chemically Activated CF3CHFCF3. J. Phys. Chem. A 2010, 114, 6996-702. 4. Duncan, J. R.; Solaka, S. A.; Setser, D. W.; Holmes, B. E. Unimolecular HCl and HF Reactions of 1,2-Dichloroethane, 1,2-Difluoroethane and 1,2-Chlorofluoroethane: Assignment of Threshold Energies. J. Phys. Chem. A 2010, 114, 794-803. 5. Wormack, L. N.; McGreal, M. E.; McClintock, C. E.; Heard, G. L.; Setser, D. W.; Holmes, B. E. Characterization of the 1,1-HF Elimination Reaction from the Competition between the 1,1HF and 1,2-DF Unimolecular Elimination Reactions of C2D5CHF2. J. Phys. Chem. A 2015, 119, 3887-3896. 6. Brown, T. M.; Nestler, M. J.; Rossabi, S. M.; Heard, G. L.; Setser, D. W.; Holmes, B. E. Characterization of the 1,1-HCl Elimination Reaction of Vibrationally Excited CD3CHFCl Molecules and Assignment of Threshold Energies for 1,1-HCl and 1,2-DCl plus 1,1-HF and 1,2-DF Elimination Reactions. J. Phys. Chem. A 2015, 119, 9441-9451. 7. Haszeldine, R. N.; Robinson, P. J.; Williams, W. J. The Kinetics of the Reactions of Silicon Compounds. Part VIII. The Gas-phase Thermal Decomposition of Trifluoro-1,1,2,2Tetrafluoroethylsilane. J. Chem. Soc. Perkins 2 1973, 1013-1019. 8. Haszeldine, R. N.; Rowland, R.; Speight, J. G.; Tipping, A. E. Carbene Chemistry. Part 11. Insertion Reactions of 1,2,2-Trifluoroethylidine into Carbon-Hydrogen Bonds of Alkanes, Cycloalkanes and Diethyl Ether. J. Chem. Soc. Perkins 1 1979, 1943-1947.

ACS Paragon Plus Environment

14

Page 15 of 19

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

9. Haszeldine, R. N.; Rowland, R.; Speight, J. G.; Tipping, A. E. Cyclopropane Chemistry. Part 4. The Reactions of 1,2,2-Trifluoroethylidine with Alkenes and Pyrolysis of the Resulting Cyclopropanes. J. Chem. Soc. Perkins 1 1980, 314-324. 10. Haszeldine, R. N.; Parkinson, C.; Robinson, P. J.; Williams, W. J. Carbene Chemistry. Part 12. Kinetics of the Isomerisation, Addition, and Insertion Reactions of 1,2,2Trifluoroethylidene. J. Chem. Soc. Perkins 2 1979, 954-961. 11. Nilsson, W. B.; Pritchard, G. O. Disproportionation Reactions between CF2H and C2H5 Radicals in the Gas Phase. Int. J. Chem. Kinet. 1982, 14, 299-323 12. Holmes, B. E.; Paisley, S. D.; Rakestraw, D. J.; King, E. E. Generation of the Ground Electronic State Haloalkyl Radicals in the Gas Phase. Int. J. Chem. Kinet. 1986, 18, 639- 649. 13. Dorer, F. H.; Rabinovitch, B. S. An Experimental Generalization of Quantum Statistical Weight Effects in Nonequilibrium Unimolecular Reactions. J. Phys. Chem. 1965, 69, 19731980.

14. Arunan, E.; Wategaonkar, S, J.; Setser, D. W. HF/HCl Vibrational and Rotational Distributions from Three- and Four-Centered Unimolecular Elimination Reactions. J. Phys. Chem. 1991, 95, 1539-1547. 15. Hu, C-H. Density Functional Study on the Reactivity of Carbenes Toward 1,2-H Shifts. J. Chinese Chem. Soc. (B), 2001, 48, 5-12. 16. Bacskay, G. B. Quantum Chemical Characterization of the Ground and Lowest Excited Singlet and Triplet States of CH3CF and CF3CF and their Photochemical Isomerization and Dissociation Pathways. Mol. Phys. 2003, 101, 1955-1965. 17. Haworth, N. L.; Smith, M. H.; Bacskay G. B.; Mackie, J. C. Heats of Formation of Hydrofluorocarbons Obtained by Gaussian-3 and Related Quantum Chemical Calculations. J. Phys. Chem. A, 2000, 104, 7600-7611.

ACS Paragon Plus Environment

15

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 19

18. Martell, J. M.; Beaton, B. T.; Holmes, B. E. Comparison between Density Functional Theory and Conventional ab Initio Methods for 1,2-HF Elimination from CH3CF3: Test Case Study for HF Elimination from Fluoroethanes. J. Phys. Chem. A 2002, 106, 8471-8478.

ACS Paragon Plus Environment

16

Page 17 of 19

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

Table of Contents Graphic

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry

Figure  1.    

5

2

4

3

2

2

CHF=CF /CHF CHF

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 18 of 19

2

1

0

0

20

40

60

80

1/P (Torr)

ACS Paragon Plus Environment

100

120  

Page 19 of 19

Figure  2    

1

2

0.8

0.6

i

2

D /CHF CHF

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

0.4

0.2

0

0

5

10

15

20

25

1/P (Torr)

ACS Paragon Plus Environment

30

35

40