Chemical Activation Study of the Unimolecular Reactions of

Oct 3, 2016 - Chemically activated C2D5CHCl2 molecules were generated with 88 kcal mol–1 of vibrational energy by the recombination of C2D5 and CHCl...
1 downloads 10 Views 2MB Size
Article pubs.acs.org/JPCA

Chemical Activation Study of the Unimolecular Reactions of CD3CD2CHCl2 and CHCl2CHCl2 with Analysis of the 1,1-HCl Elimination Pathway Allie C. Larkin,‡ Matthew J. Nestler,‡ 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, North Carolina 28804-8511, United States † Department of Chemistry, Kansas State University, Manhattan, Kansas 66506, United States S Supporting Information *

ABSTRACT: Chemically activated C2D5CHCl2 molecules were generated with 88 kcal mol−1 of vibrational energy by the recombination of C2D5 and CHCl2 radicals in a room temperature bath gas. The competing 2,1-DCl and 1,1-HCl unimolecular reactions were identified by the observation of the CD3CDCHCl and CD3CDCDCl products. The initial CD3CD2CCl carbene product from 1,1-HCl elimination rearranges to CD3CDCDCl under the conditions of the experiments. The experimental rate constants were 2.7 × 107 and 0.47 × 107 s−1 for 2,1-DCl and 1,1-HCl elimination reactions, respectively, which corresponds to branching fractions of 0.84 and 0.16. The experimental rate constants were compared to calculated statistical rate constants to assign threshold energies of 54 and ≈66 kcal mol−1 for the 1,2-DCl and 1,1-HCl reactions, respectively. The statistical rate constants were obtained from models developed from electronic-structure calculations for the molecule and its transition states. The rate constant (5.3 × 107 s−1) for the unimolecular decomposition of CHCl2CHCl2 molecules formed with 82 kcal mol−1 of vibrational energy by the recombination of CHCl2 radicals also is reported. On the basis of the magnitude of the calculated rate constant, 1,1-HCl elimination must contribute less than 15% to the reaction; 1,2-HCl elimination is the major reaction and the threshold energy is 59 kcal mol−1. Calculations also were done to analyze previously published rate constants for chemically activated CD2ClCHCl2 molecules with 86 kcal mol−1 of energy to obtain a better overall description of the nature of the 1,1-HCl pathway for 1,1-dichloroalkanes. The interplay of the threshold energies for the 2,1-HCl and 1,1-HCl reactions and the available energy determines the product branching fractions for individual molecules. The unusual nature of the transition state for 1,1HCl elimination is discussed. 1,1-chlorofluoroethane13 using the chemical activation experimental technique with deuterium labeling to identify the pathways. The experimental results were coupled with electronic-structure calculations to characterize models for the transition states and assigned the threshold energies, E0. The goal of the present study was to develop a similar understanding for 1,1-HCl elimination reactions from 1,1-dichloroalkanes. The 1,1-HCl pathway has been identified13 from chemically activated CD3CHFCl with an overall branching fraction of 0.28 and a branching ratio, relative to 2,1-DCl elimination, of 0.5. This favorable outcome for the 1,1-HCl pathway was a consequence of the elevation of E0(2,1-DCl) by the F atom, as well as the stabilization of the CD3C−F carbene. In the present study we have selected CD3CD2CHCl2* for

I. INTRODUCTION The characteristic unimolecular reaction pathway for haloalkanes is 1,2-HX (X = F, Cl, Br) elimination via a fourmembered transition state.1−3 However, the 1,1-HF or HCl elimination reaction pathway from 1,1-dihaloalkanes can compete with 2,1-HF or HCl elimination at high levels of vibrational energy or high temperature.4−8 The 1,1-HX reaction requires a second halogen atom on the terminal carbon atom to stabilize the halocarbene product, which lowers the threshold energy, E0, for the 1,1-elimination reaction. Certain substituents can raise the threshold energy for 2,1-HX elimination, which has the effect of enhancing the 1,1-HX elimination component. However, the possibility of competing halogen-atom interchange reactions, if different halogen atoms are located on adjacent carbon atoms, should be remembered when considering the unimolecular mechanisms for more complex haloalkanes.9−12 We have recently characterized 1,1-HF elimination from 1,1-difluoroethane, 1,1-difluoropropane,8 and © XXXX American Chemical Society

Received: July 22, 2016 Revised: October 2, 2016

A

DOI: 10.1021/acs.jpca.6b07368 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

Figure 1. Energy profile for the 2,1-DCl and 1,1-HCl elimination reactions of CD3CD2CHCl2. The listed threshold energies are the experimental assignments. Note that the assigned E0(1,1-HCl) is slightly smaller than the minimum value defined by the thermochemistry for formation of CD3CD2CCl + HCl. The energy profiles for the CD2ClCHCl2 and CHCl2CHCl2 systems are similar except that the ⟨E⟩ from radical association are lower, 86 and 82 kcal mol−1, respectively. The energies for CD2ClCCl (+HCl) and CHCl2CCl (+HCl) are 70 and 72 kcal mol−1, respectively (obtained from DFT calculations of the isomerization energy of the carbenes with an uncertainty of ±2 kcal mol−1). The threshold energies assigned from rate constant experiments for 2,1-DCl (HCl) elimination for CD2ClCHCl2 and CHCl2CHCl2 are 60 and 59 kcal mol−1, respectively; the E0 for 1,1-HCl elimination are 69 and ≈66 kcal mol−1, respectively.

study because of the more favorable pressure range, relative to CD3CHCl2*, for the chemical activation technique. In contrast to a F atom, the second Cl atom of 1,1-dichloroalkanes will lower E0(2,1-DCl),3 and the branching ratio for 1,1-HCl elimination is expected to be lower than that from 1,1chlorofluoroalkanes. Nevertheless, the 2,1-DCl and 1,1-HCl elimination rate constants could be measured from the CD3CDCHCl (+DCl) and CD3CDCDCl (+HCl) products formed from the C2D5CHCl2* molecules that were generated by the recombination of C2D5 and CHCl2 radicals in a room temperature bath gas. The CD3CD2CCl carbene isomerizes to CD3CDCDCl under the conditions of the experiments.14−16 Under thermal equilibrium conditions14 the Z(cis)-isomer of CH3CHCHCl is preferred by a factor of 3 relative to the E(trans)-isomer by the H atom rearrangement of C2H5CCl. Based on the thermochemistry of the recombination reaction, the average vibrational energy of the C2D5CHCl2* molecules is 88 ± 2 kcal mol−1. These results from the chemical activation experiments can be compared to the rate constant from thermal activation studies of C2H5CHCl2.17 Because CHCl2CHCl2* also is present in the experiments, the unimolecular reactions of this molecule formed with 82 kcal mol−1 are analyzed, even though the 1,1-HCl and 1,2-HCl paths cannot be experimentally resolved. The data from the present study of C2D5CHCl2* will be combined with previously published18 results from chemically activated 1,2,2-trichloroethane-d0, -d1, -d2 formed with 86 kcal mol−1 of energy to further define the model for 1,1-HCl elimination. The chloromethyl−chlorocarbene isomerizes to 1,2-dichloroethene, and with deuterium labeling all three reaction channels can be identified. The original assignment18

of the threshold energy for 1,1-HCl elimination from CD2ClCHCl2 needs to be revised upward. The experimental rate constants, kexp, are compared to the calculated statistical rate constants at the average energy, ⟨E⟩, to assign the threshold energies for the 2,1-DCl and 1,1-HCl elimination reactions. Electronic structure calculations using the Gaussian code19 were used to obtain the vibrational frequencies and moments of inertia of the molecules and transition states, which are needed to calculate the rate constants. The MP2 method was selected because density functional theory methods, DFT, did not readily identify the transition states for 1,1-HCl elimination.13 These transition states also serve for the reverse reaction, the addition of the carbene to HCl. CASSCF(6,6) calculations were done to complement the results from MP2 calculations. The vibrationally excited C2D5CHCl2* and CHCl2CHCl2* molecules were generated by the recombination of C2D5 and CHCl2 radicals, which were prepared by mercury photosensitization3 of mixtures of C 2 D5 I and CHCl 3 . The recombination and disproportionation reactions of the C2D5 and CHCl2 radicals are listed in (1)−(3), but recombination is always the dominant process.20a,b The concentration of C2D5 is higher than the concentration of CHCl2 and reactions 1a and 3a are the most important. The disproportionation reactions of CHCl2 have not been studied, but (1c) is much more important than (1b) or (2b), and CCl2 is not an important factor in the experiments. An asterisk denotes vibrational excitation from formation of the carbon−carbon bond. B

DOI: 10.1021/acs.jpca.6b07368 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A C2D5 + CHCl 2 → CD3CD2 CHCl 2* → CD3CD2 H + CCl 2

(1b)

→ CD2 CD2 + CDHCl 2

(1c)

2CHCl 2 → CHCl 2CHCl 2* → CH 2Cl 2 + CCl 2 2C2D5 → C4 D10 * → CD2 CD2 + CD3CD3

preparation was 13.022 cm3 and the quartz vessels ranged in size from 6.5095 to 1056.6 cm3. Photolysis times were between 10 s and 7 min depending on the volume of the reaction vessel and pressure of reactants. After photolysis, the contents of the vessel were transferred to the vacuum line of a Shimadzu QP5000 GC-MS, which had a 105m RTX-1701 column, and analyzed with peak areas integrated using the software supplied by Shimadzu. Calibration mixtures were prepared to replicate the composition of the photolyzed samples to obtain response factors to convert the apparent ratios of CD3CDCHCl/ CD3CD2CHCl2 and CHClCCl2/CHCl2CHCl2 to relative concentrations. The GCMS signals were recorded as integrated areas of mass peaks. The 1-chloropropene and 1,1-dichloropropane were monitored by masses corresponding to the parent ion with 35Cl and parent ion minus one Cl atom, but with the 37Cl isotopic mass, respectively. The calibration factor, based on five trials from four separate mixtures of the available nondeuterated compounds, equaled 0.547 ± 0.068. The Z- and E-CH3CHCHCl geometric isomers, which were separated by the GC, were calibrated individually. The isotopic isomers, CD3 CDCDCl and CD3 CDCHCl are distinguished directly by the mass spectrometer and their ratios require no calibration. Yields of CHClCCl2 and CHCl2CHCl2 were measured by the mass spectrometer using m/e = 134 and 168, respectively. The calibration factor for CHClCCl 2 / CHCl2CHCl2 was 0.389 ± 0.042 on the basis of four mixtures prepared to mimic actual product yields. Preparation of calibration mixtures for the CHCl2CHCl2 system was difficult because of the low vapor pressures, and the results for CHCl2CHCl2 are less reliable than for C2D5CHCl2. No attempt was made to identify the products3 arising from the reactions of Cl atoms with CF3CHCH2. The experimental rate constants, kexp, for the excited molecules are obtained from experiments in which the ratio of the concentration of the decomposition products (D) to the collisionally stabilized product (S) is measured as a function of pressure. The ratio of D/S is equal to the ratio of rate constants, kexp/kM[M]; kM is the collision rate constant and [M] is the concentration of bath gas, which is represented as pressure for convenience of discussion. This formulation for efficient collisional deactivation should be an acceptable approximation for collisions of C2D5CHCl2* or CHCl2CHCl2* with C2D5I, CHCl3, and CF3CHCH2 at room temperature. The average energy lost from C2D5CHCl2* per collision will be 6−8 kcal mol−1 by analogy to similar systems.21a,b The rate constants decline rapidly with reduced energy and the probability for additional reaction after one collision is not significant. The collision rate constant was calculated from the standard equation, kM = πdM2(8kT/πμM)1/2Ω2,2(T); the properties for the collision partners are given in the footnotes of Table 1.

(1a)

(2a) (2b) (3a) (3b)

Reaction 1a is the process of primary interest, but, because CHCl2CHCl2* is present in the system, it was included in the study. Although (1a) and (2a) are reversible in principle, the elimination reactions shown in (4) and (5) have lower threshold energies, and dissociation of C2D5CHCl2* and C2H2Cl4* is not competitive. The CD3CD2CHCl2* molecules will decompose by either 2,1-DCl or 1,1-HCl elimination unless they are deactivated by collisions with molecules of the bath gas. Reaction 4d is the isomerization of CD3CD2C−Cl, which has a threshold energy15,16 of 6 kcal mol−1. CD3CD2 CHCl 2* → DCl + Z ‐ and E‐CD3CDCHCl (4a) → HCl + CD3CD2 CCl + M → CD3CD2 CHCl 2 + M CD3CD2 CCl → Z ‐ and E‐CD3CDCDCl

(4b) (4c) (4d)

Similar reactions describe the unimolecular decomposition of CHCl2CHCl2*, but the 1,1-HCl and 1,2-HCl channels cannot be distinguished experimentally. The rearrangement of CHCl2CCl via H atom migration has a potential barrier15,16 of 4−5 kcal mol−1. The possibility of carbon−chlorine rupture as a third decomposition path for CHCl2CHCl2* will be considered in the Discussion. CHCl 2CHCl 2* → CHClCCl 2 + HCl

(5a)

→ CHCl 2CCl + HCl

(5b)

+ M → CHCl 2CHCl 2 + M CHCl 2CCl → CHClCCl 2

(5c) (5d)

The thermochemistry of the C 2 D 5CHCl 2 system is summarized in Figure 1, and the relevant information for CHCl2CHCl2 and CD2ClCHCl2 is given in the caption of Figure 1. The minimum threshold energy for the 1,1-HCl elimination is provided by the enthalpy of formation of carbene + HCl, which can be obtained from the sum of enthalpy of formation of the product olefin + HCl and the enthalpy for rearrangement of the carbene.

3. EXPERIMENTAL RESULTS The decomposition to stabilization ratios and the branching fractions for the CD3CD2CHCl2 system are displayed in Figures 2 and 3. The numerical results, including starting gas composition and vessel size, as well as the decomposition and stabilization ratios, are given in the Supporting Information. The sum of the 1,1-HCl and 2,1-DCl decomposition products, Dtotal, is divided by the stabilized product, S, and plotted vs 1/ pressure to obtain the total average rate constant. The slope of this plot is 2.15 ± 0.11 Torr, which corresponds to a rate constant of 3.1 × 107 s−1 when the collision parameters in

2. EXPERIMENTAL METHODS The experiments consist of loading quartz vessels with CD3CD2I, CHCl3, and CF3CHCH2 in a ratio of 5:2:4 and a droplet of Hg, followed by photolysis at room temperature with the 253.7 nm output from a 15 W germicidal lamp. CF3CHCH2 was added to protect the 1-chloropropene decomposition product from reactions with Cl atoms that are produced from the interaction of Hg(3P1) atoms with CHCl3. The use of CF3CHCH2 as a scavenger for Cl atoms is described in ref 3. The calibrated volume for sample C

DOI: 10.1021/acs.jpca.6b07368 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

Table 1. Rate Constants and Threshold Energies for C2D5CHCl2 (⟨E⟩ = 88 kcal mol−1), C2H2Cl4 (⟨E⟩ = 82 kcal mol−1), and CD2ClCHCl2 (⟨E⟩ = 86 kcal mol−1) experimental resultsa,b

calculated results

slope, Torr

kExp, s−1

k⟨E⟩, s−1

E0, kcal/mol

C2D5CHCl2* → CD3CDCHCl (DCl)

1.83 ± 0.09

(2.7 ± 0.3) × 107

CD3CDCDCl (HCl)

0.32 ± 0.02

(0.47 ± 0.05) × 107

3.8 ± 0.4

(5.3 ± 0.6) × 107

3.3 × 107 2.2 × 107 0.19 × 107 0.33 × 107 3.6 × 107 0.6 × 107

54 55 67 66 59c ≤69c

4.0 ± 0.8 2.0 ± 0.4