UC Berkeley

It is well known that the efficacy of acidic zeolite catalysts for the cracking of hydrocarbons originates from the shape and size of the zeolite pores. However, the mechanisms by which changes in pore structure influence cracking kinetics are not well understood or exploited. The aim of this dissertation is to use experiments and simulations to shed light on the ways by which zeolite structure and acid site location affect the apparent and intrinsic kinetics of n-alkane monomolecular cracking and dehydrogenation. In the rate-determining step of these processes, C-C or C-H bonds are cleaved catalytically by Brønsted protons. Thus, the kinetics of monomolecular activation reactions are useful for characterizing the influence of active site structural environment on catalysis.

In Chapter 2, the effects of active site distribution on n-butane monomolecular activation kinetics are investigated for commercial samples of MFI having a range of the Si/Al ratio. Based on UV-visible spectroscopic analyses of (Co,Na)-MFI, it is inferred that, with increasing Al concentration, the fraction of Co—and, by extension, Brønsted protons in H-MFI—located at channel intersections increases relative to the fraction located at channels. Concurrently, the first-order rate coefficients (kapp) for cracking and dehydrogenation, the selectivity to terminal cracking versus central cracking, and the selectivity to dehydrogenation versus cracking increase. The stronger dependence of the selectivity to dehydrogenation on Al content is attributed to a product-like transition state, the stability of which is more sensitive to confinement than the stabilities of cracking transition states, which occur earlier along the reaction coordinate. For terminal cracking and dehydrogenation, the intrinsic activation entropy (∆Sint) increases with Al content, consistent with the larger dimensions of intersections relative to channels. Surprisingly, the rate of dehydrogenation is inhibited by butene products. Theoretical calculations suggest that this effect originates from the adsorption of isobutene at channel intersections, indicating that dehydrogenation occurs with stronger preference for these locations than does cracking.

In order to analyze the effects of zeolite structure on monomolecular alkane activation reactions, it is necessary to separate the contributions of the adsorption and reaction steps to observed kinetics. A method is developed in Chapter 3 for obtaining the enthalpy and entropy changes for adsorption of n-alkanes from the gas phase onto Brønsted protons (ΔHads‑H+ and ΔSads‑H+) using configurational-bias Monte Carlo (CBMC) simulations. Simulated values of ΔHads‑H+ and ΔSads‑H+ for H-MFI are in good agreement with those determined from experimental measurements at 300-400 K. However, the simulations account correctly for the redistribution of alkanes to protons at less confining parts of the zeolite with increasing temperature. Values of ∆Hint and ∆Sint for the cracking of n-alkanes, determined using previously reported kinetic data and simulated values of ΔHads‑H+ and ΔSads‑H+, both corresponding to 773 K, agree well with values obtained independently from quantum mechanics/molecular mechanics calculations. It is found that the observed increase in kapp with increasing chain length is caused by a decrease in ∆Hint and that ∆Sint is insensitive to chain length. These results contrast those reported by other authors, who used values of ΔHads‑H+ and ΔSads‑H+ measured at 323 K to extract ∆Hint and ∆Sint from the same measured kinetic data and concluded that the increase in kapp with alkane size is caused by an increase in ∆Sint.

In Chapter 4 the effects of zeolite structural confinement on n-butane cracking and dehydrogenation are characterized for zeolites that differ predominately in the size and abundance of cavities. Values of ∆Hads-H+ and ∆Sads-H+ are obtained from CBMC simulations and used to extract intrinsic rates and activation parameters. As ∆Sads-H+ (a proxy for confinement) becomes more negative, ∆Hint and ∆Sint decrease for terminal cracking and dehydrogenation when the channel topology (e.g., straight, sinusoidal) is fixed. This observation, as well as positive values for ∆Sint, indicate that the transition states for these reactions resemble the products. For central cracking (an early transition state), ∆Hint remains similar while ∆Sint increases with confinement because less entropy is lost upon transfer of a proton to an adsorbed n-butane molecule. For zeolites having straight channels, the increase in ∆Sint is large enough to cause kint to also increase. For terminal cracking and dehydrogenation, concurrent decreases in ∆Hint and ∆Sint cause kint to increase less strongly, and selectivities to these reactions decrease with increasing confinement. Depending on channel topology, changes in kapp are driven by changes in kint or by changes in the adsorption equilibrium constant (Kads H+), which is not, in general, dominated by either ∆Hads-H+ or ∆Sads-H+. These findings differ from earlier reports that ∆Hint and ∆Sint are structure-insensitive, and that Kads-H+ is dominated by the value of ∆Hads-H+.

Finally, in Chapter 5 the influence of channel and cage topology on n-alkane adsorption are characterized for zeolites and zeotypes with one-dimensional pore systems. When cages are not present, ∆Hads-H+ and ∆Sads-H+ at fixed pore-limiting diameter (PLD; the diameter of the largest sphere that can traverse the pores) decrease in magnitude as the ratio of the smallest to largest channel diameter decreases and the pore become less circular. The higher entropy of alkanes in non-circular pores is attributed to greater freedom of movement and can cause the free energy to be lower in these environments relative to circular pores. The addition of cages to straight channels at fixed PLD generally decreases confinement and the magnitudes of ∆Hads-H+ and ∆Sads-H+. Replacing straight channels with cages of the same diameter does not change ∆Sads-H+ significantly when the channel PLD exceeds the length of the alkane, but lowers ∆Hads-H+ and the free energy due to the greater surface area and curvature of cages relative to channels. In zeolites that lack cages, the selectivity to adsorption via a central C-C bond vs. a terminal bond exhibits a minimum at PLDs near the length of the alkane. When cages are present, the selectivity to adsorption via a central bond exhibits a minimum with respect to cage size, occurring at a diameter larger than that observed in the absence of cages. This result is attributed to a greater ability of cages to stabilize configurations in which the alkane backbone is oriented perpendicular to the cage wall.