Interaction of C₃–C₅Alkenes with Zeolitic Brønsted Sites: π-Complexes, Alkoxides, and Carbenium Ions in H-FER

Abstract

We use a chemically accurate (4 kJ/mol) hybrid MP2:(PBE+D2) + ΔCCSD(T) method to determine relative stabilities of all possible π-complexes, alkoxides, and carbenium ions formed from propene, butene, and pentene with the Al(2)O(7) Brønsted acid site in H-FER. The energetic order is carbenium ions > tert-alkoxides > π-complexes as well as primary and secondary alkoxide species. Primary carbenium ions are not stationary points on the potential energy surface. The energetically most stable C3, C4, and C5 surface species are 2propoxide, 2-butoxide, and the 2-methyl-2-butene π-complex with energies of −78, −81, and −85 kJ/mol, respectively, for formation from the corresponding alkenes. Compared to the present results, the widely applied PBE+D2 approach overbinds all species, and the energy differences are 18−24, 25−45, and 48−71 kJ/mol for π-complexes, alkoxides, and carbenium ions. Enthalpies and Gibbs free energies are calculated for 323 and 623 K within the harmonic approximation. The calculated adsorption enthalpy of trans-2-pentene, −93 kJ/mol, is in agreement with the experimental value, −92 kJ/mol [Schallmoser et al. J. Am. Chem. Soc. 2017, 139, 8646]. Entropy favors the more mobile species (carbenium ions, π-complexes), and the Gibbs free energy order becomes carbenium ions and tert-alkoxides > primary and secondary alkoxides > π-complexes.

Dimerization of Linear Butenes and Pentenes in an Acidic Zeolite (H‐MFI)

“…For the first step of the surface dimerization (Figure 1), adsorption of two trans ‐2‐pentene molecules on BASs of H‐FER [π‐complex formation, Eq. (1a)], MP2+ΔCC results (−166 kJ mol −1 ) [12] and experiment (−165 kJ mol −1 ) [3] are in very close agreement, thus confirming the accuracy of the hybrid QM:QM method.…”

“…With an estimate of Δ H d =−88 kJ mol −1 for the heat of dimerization in the gas phase yielding a branched alkene the authors arrived at Δ H s =−285 + 88=−197 kJ mol −1 and assigned this enthalpy to the adsorption and formation of a C 10 alkoxide at the BAS, see Figure 1. This assignment was based on hybrid QM:MM predictions that alkoxides are 55–57 kJ mol −1 more stable than the corresponding alkene π‐complexes, [11] whereas later accurate QM:QM calculations showed for butenes and pentenes in H‐FER that alkoxides and π‐complexes are about equally stable [12] …”

“…For smaller cluster models of affordable size CCSD(T) calculations [14] are performed, see Supporting Information for details. Energies obtained this way are termed “MP2+ΔCC”, [12] their estimated uncertainty is ±4 kJ mol −1 [20] . The distribution of BAS over the different crystallographic positions of the MFI framework is not known and varies with the synthesis conditions [21] .…”

Experimental and Theoretical Evidence for the Promotional Effect of Acid Sites on the Diffusion of Alkenes through Small‐Pore Zeolites

“…Our results clearly identify the formation of stable alkene p-complex intermediates which corroborates the findings of previous IR spectroscopy [51][52][53] studies as well as first-principle calculations. [47,[54][55][56] on different zeolite topologies. Thea lkene hopping paths between neighboring cages are illustrated in Figure 5.…”

Dynamic Features of Transition States for β‐Scission Reactions of Alkenes over Acid Zeolites Revealed by AIMD Simulations

“…For cracking reactions, the implication of secondary carbenium ions as virtual reaction products (type B 2 ) needs to convert into a more stable species (here a propene π‐complex), inducing an extra‐barrier with respect to the pure β‐scission cost. Going beyond DFT to reach chemical accuracy may be needed, as GGA functionals overbind carbenium ions and alkoxides with respect to π‐complexes . However, combining this level of calculation with the MD approach is currently prohibitive.…”