61st Annual Report on Research 2016

The goal of the project is to further characterize the intermediates in Friedel-Crafts alkylation reactions, and the associated ionization processes. The proposed intermediates in these reactions are alkyl halide – Lewis acid Complexes (RX’–MX₃) and in the initial phases of this work, we explored the structural and energetic properties of several such complexes (M = B, Al, Ga; X/X’ = F, Cl; R = CH₃) using density functional theory. At this point we have not only explored structures and binding energies, but also M-X’ potential curves in the gas-phase and bulk dielectric media. This survey-phase of the project is nearly complete, though we are currently exploring systems with larger alkyl groups.

Over the past year, our computational efforts have been focused on exploring the relative energies of two distinct ionization pathways that lead to the key electrophiles in these processes. The first involves a single alkyl halide, e.g.,

RX + MX₃ —> R⁺ + MX₄^– (1)

and could presumably proceed through the 1:1 complex intermediates noted above. The other pathway involves a second alkyl halide intermediate, and leads to the formation of a dialkyl halonium ion, e.g.,

2 RX + MX₃ —> R₂X⁺ + MX₄^– (2)

We have found that (2) is a much lower energy process, both in the gas-phase and in bulk dielectric media, for every permutation of M, X and X’ we have studied.

For example, for the most common reactions involving an alkyl chloride and AlCl₃, e.g.

CH₃Cl + AlCl₃ —> CH₃⁺ + AlCl₄^– ΔE₁ (3)

2 CH₃Cl + AlCl₃ —> (CH₃)Cl⁺ + AlCl₄^– ΔE₂ (4)

we find that ΔE₁ is +154.7 kcal/mol while ΔE₂ is +86.1 kcal/mol, roughly 70 kcal/mol lower in energy. (These particular results are from M06/aug-cc-pVTZ calculations.) In a bulk dielectric medium with ε=10 (PCM/M06/aug-cc-pVTZ), the reaction energies are significantly lower, and ΔE₂ becomes slightly exothermic; ΔE₁ is +51.2 kcal/mol while ΔE₂ is -1.0 kcal/mol.

For the analogous CH₃F/BF₃ processes, e.g.,

CH₃F + BF₃ —> CH₃⁺ + BF₄^– ΔE₁ (5)

2 CH₃F + BF₃ —> (CH₃)F⁺ + BF₄^– ΔE₂ (6)

which are more closely related to our previous work in these systems, we find that ΔE₁ is +134.5 kcal/mol while ΔE₂ is +86.1 kcal/mol, roughly 50 kcal/mol less (via M06/aug-cc-pVTZ). Again, in a bulk dielectric medium with ε=10 (PCM/M06/aug-cc-pVTZ), the reaction energies are significantly lower; ΔE₁ is +67.3 kcal/mol while ΔE₂ is 32.4 kcal/mol. We are currently in the process of validating these results with high-level post Hartree-Fock methods, as well as exploring the free energy changes associated with these reactions, which will provide direct insight into their relative thermodynamic feasibility.

Meanwhile, our experimental efforts have been focused on obtaining low-temperature IR spectra of RX/MX₃ mixtures, in hopes of spectroscopically identifying and characterizing the key intermediates in situ, but in the absence of the aromatic substrate that would be present in an actual reaction mixture. The intent is to design and construct a short-path (0.01 – to 0.1 mm), liquid IR cell with will be housed inside the vacuum chamber of our optical cryostat apparatus, such that it would be filled by condensing RX’ and/or MX₃ vapor from ambient temperature bulbs located outside the main vacuum chamber. Once this capability has been developed, we will collect temperature dependent IR spectra of RX’ (l), MX₃ (l), and RX/MX₃ mixtures, in search of key signatures that are only present in the mixture. These will be identified via comparisons to calculated spectra. At this point, we have developed a prototype that has enabled us to collect the spectrum of (CH₃)­₂CHF at a range of temperatures between 140 and 180 K. Construction of a more robust, permanent cell is currently in progress.