Controlled insertion of tin atoms into zeolite framework vacancies and consequences for glucose isomerization catalysis

- Dichloromethane-assisted reflux enables grafting tin within all vacancy defects in Beta zeolites.
- Open Sn sites are incorporated preferentially at low Sn densities via post-synthetic grafting.
- Glucose isomerization rates (per open Sn) are invariant with Si/Sn among post-synthetic Sn-Beta.
- Methanol micropore filling can probe silanol defect density within micropores.
- Glucose isomerization rates are 15-50× higher (373 K) on hydrothermally-synthesized than on post-synthetic Sn-Beta.

The incorporation of tin heteroatoms within framework vacancy defects of dealuminated Beta zeolites via stannic chloride reflux in dichloromethane solvent (333 K) enables preparing Sn-Beta zeolites comprising a wide range of Sn content (Si/Sn = 30-144; 1.4-6.1 wt% Sn) and >4× higher framework Sn densities than reported (Si/Sn > 120) for reflux in isopropanol solvent (383 K). Silanol defects form hydrogen bonds with isopropanol but not with dichloromethane under reflux conditions, evident in the evolution of adsorbed isopropanol (∼5× per vacancy) but not dichloromethane during temperature programmed desorption of dealuminated zeolites saturated with either solvent at ambient, and in IR features for perturbed hydroxyl stretches (∼3350 cm−1) and CH stretches (∼2700-3000 cm−1) that were retained after dealuminated zeolites were saturated with isopropanol and evacuated at 383 K. Dichloromethane-assisted tin grafting provides a route to precisely control the density of framework Sn sites and residual vacancy defects in Sn-Beta, up to the point of grafting virtually every framework vacancy with Sn. Open Sn sites ((OSi)3Sn(OH), ν(CD3CN): 2316 cm−1), quantified from IR spectra collected after CD3CN titration (303 K), are preferentially incorporated over closed Sn sites (Sn(OSi)4, ν(CD3CN): 2308 cm−1) at low Sn densities via grafting in dichloromethane solvent, suggesting that this preparation method can tune the open-to-closed framework Sn site ratio more systematically than hydrothermal synthesis methods. Open Sn sites are dominant active sites for aqueous-phase glucose-fructose isomerization; consequently, isomerization turnover rates (per total Sn) decrease systematically with increasing Sn content. Initial isomerization rates (per open Sn) are invariant (within ∼2×, 373 K) among twelve Sn-Beta samples of varying Sn content (Si/Sn = 30-144), the behavior expected of a single-site catalyst in which open framework Sn sites are the loci of catalytic reactivity. First-order isomerization rate constants are 15-50× lower (373 K) when open Sn sites are confined within high-defect than within low-defect Beta micropores, consistent with previous reports that aqueous-phase sugar isomerization cycles turn over at faster rates within hydrophobic, low-defect micropores. These findings clarify the consequences of liquid-phase reflux procedures on the coordination of tin heteroatoms incorporated within zeolite framework vacancies, and underscore the requirement to quantify putative active site structures in order to rigorously normalize measured rate data prior to kinetic or mechanistic interpretation.

78