New porous materials such as zeolites, metal-organic frameworks and mesostructured oxides are of immense practical utility for gas storage, separations and heterogeneous catalysis. Their extended pore structures enable selective uptake of molecules or can modify the product selectivity (regioselectivity or enantioselectivity) of catalyst sites contained within. However, diffusion within pores can be problematic for biomass and fine chemicals, and not all catalyst classes can be readily synthesized with pores of the correct dimensions. Here, we present a novel approach that adds reactant selectivity to existing, non-porous oxide catalysts by first grafting the catalyst particles with single-molecule sacrificial templates, then partially overcoating the catalyst with a second oxide through atomic layer deposition. This technique is used to create sieving layers of Al(2)O(3) (thickness, 0.4-0.7 nm) with 'nanocavities' (<2 nm in diameter) on a TiO(2) photocatalyst. The additional layers result in selectivity (up to 9:1) towards less hindered reactants in otherwise unselective, competitive photocatalytic oxidations and transfer hydrogenations.

The effect of outer-sphere environment on alkene epoxidation catalysis using an organic hydroperoxide oxidant is demonstrated for calix[4]arene-Ti single-sites grafted on amorphous vs crystalline delaminated zeotype (UCB-4) silicates as supports. A chelating calix[4]arene macrocyclic ligand helps enforce a constant Ti inner-sphere, as characterized by UV-visible and X-ray absorption spectroscopies, thus enabling the rigorous comparison of outer-sphere environments across different siliceous supports. These outer-sphere environments are characterized by solid-state H NMR spectroscopy to comprise proximally organized silanols confined within 12 membered-ring cups in crystalline UCB-4, and are responsible for up to 5-fold enhancements in rates of epoxidation by Ti centers.

The effect of dynamic reorganization and confinement of isolated TiIV catalytic centers supported on silicates is investigated for olefin epoxidation. Active sites consist of grafted single-site calix[4]­arene–TiIV centers or their calcined counterparts. Their location is synthetically controlled to be either unconfined at terminal T-atom positions (denoted as type-(i)) or within confining 12-MR pockets (denoted as type-(ii); diameter ∼7 Å, volume ∼185 Å3) composed of hemispherical cavities on the external surface of zeotypes with *-SVY topology. Electronic structure calculations (density functional theory) indicate that active sites consist of cooperative assemblies of TiIV centers and silanols. When active sites are located at unconfined type-(i) environments, the rate constants for cyclohexene epoxidation (323 K, 0.05 mM TiIV, 160 mM cyclohexene, 24 mM tert-butyl hydroperoxide) are 9 ± 2 M–2 s–1; whereas within confining type-(ii) 12-MR pockets, there is a ∼5-fold enhancement to 48 ± 8 M–2 s–1. When a mixture of both environments is initially present in the catalyst resting state, the rate constants reflect confining environments exclusively (40 ± 11 M–2 s–1), indicating that dynamic reorganization processes lead to the preferential location of active sites within 12-MR pockets. While activation enthalpies are ΔH ‡ app = 43 ± 1 kJ mol–1 irrespective of active site location, confining environments exhibit diminished entropic barriers (ΔS ‡ app = −68 J mol–1 K–1 for unconfined type-(i) vs −56 J mol–1 K–1 for confining type-(ii)), indicating that confinement leads to more facile association of reactants at active sites to form transition state structures (volume ∼ 225 Å3). These results open new opportunities for controlling reactivity on surfaces through partial confinement on shallow external-surface pockets, which are accessible to molecules that are too bulky to benefit from traditional confinement within micropores.

Although essentially molecular noble metal species provide active sites and highly tunable platforms for the design of supported catalysts, the susceptibility of the metals to reduction and aggregation and the consequent loss of catalytic activity and selectivity limit opportunities for their application. Here, we demonstrate a new construct to stabilize supported molecular noble-metal catalysts, taking advantage of sterically bulky ligands on the metal that serve as surrogate supports and isolate the active sites under conditions involving steady-state catalytic turnover in a reducing environment. The result is demonstrated with an iridium pair-site catalyst incorporating P-bridging calix[4]arene ligands dispersed on siliceous supports, chosen as prototypes because they offer weakly interacting surfaces on which metal aggregation is prone to occur. This catalyst was used for the hydrogenation of ethylene in a flow reactor. Atomic-resolution imaging of the Ir centers and spectra of the catalyst before and after use show that the metals resisted aggregation and deactivation, remaining atomically dispersed and accessible for catalysis. This strategy thus allows the stabilization of the catalysts even when they are weakly anchored to supports.

Nests of three silanol groups are located on the internal pore surface of calcined zeolite SSZ‐70. 2D 1H double/triple‐quantum single‐quantum correlation NMR experiments enable a rigorous identification of these silanol triad nests. They reveal a close proximity to the structure directing agent (SDA), that is, N,N′‐diisobutyl imidazolium cations, in the as‐synthesized material, in which the defects are negatively charged (silanol dyad plus one charged SiO− siloxy group) for charge balance. It is inferred that ring strain prevents the condensation of silanol groups upon calcination and removal of the SDA to avoid energetically unfavorable three‐rings. In contrast, tetrad nests, created by boron extraction from B‐SSZ‐70 at various other locations, are not stable and silanol condensation occurs. Infrared spectroscopic investigations of adsorbed pyridine indicate an enhanced acidity of the silanol triads, suggesting important implications in catalysis.