Figure 4. (a) Molar formation rate (per total Fe) of CH3OH and CH3OD formed over MIL-100(Fe) when exposed to D2O at 473 K following reaction. (b) Effect of the molar ratio of H2O and D2O fed during the product extraction step on the relative amount of CH3OH and CH3OD formed. (c) Molar formation rate (per total Fe) of CH316OH and CH318OH over MIL-100(Fe) when exposed to H218O at 473 K following reaction. (1.6 kPa N2O, 1.5 kPa CH4, 473 K, 2 h). Figure 4b reproduced from ref. 35 Copyright © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
Reaction with water to form methanol from intermediates that are not desorbed from the surface in the absence of water mirrors several observations reported in prior literature. Ethane oxidation over MOF-74(Mg,Fe) to produce ethanol, acetaldehyde, and diethyl ether required extraction with CD3CN following reaction at 348 K.36 Propane/ethane oxidation over MIL-100(Fe) also required oxygenated products to be extracted with D2O, with only unsaturated ethene/propene products desorbing into the gas phase in the absence of D2O.39 DFT calculations reported by Vitillo et al. evaluating the radical rebound mechanism for methanol formation suggest that the step involving formation of the Fe(IV)=O intermediate over MIL-100(Fe) carries the highest activation barrier (140.5 kJ mol-1).40 The authors suggest that alcohol desorption may not readily occur under reaction conditions due to the high activation barrier for the desorption for methanol (91.5 kJ mol-1) in comparison to the heat of adsorption of N2O (30 kJ mol-1). A competing pathway to radical rebound to form the surface bound methanol product is one in which the radical dissociates from the active center.55,56 The activation energy for radical desorption from Fe nodes in MIL-100(Fe) MOFs was predicted to be only slightly greater (~ 5 kJ mol-1greater) than the barrier for radical rebound.40 We note in this context that experimental evidence for catalyticmethane hydroxylation over MIL-100(Fe) has not yet been reported in the literature. Though methoxy intermediates are identified in our study as the predominant species formed prior to exposure to water vapor, the identity of elementary steps that form them remain unclear, and the possibility of minor quantities of methanol being formed is challenging to disprove given the plausibility of methanol reacting with open-metal iron sites, as demonstrated in the discussion that follows.
To test for the plausibility of methoxy formation mediated by either adsorbed or gas phase methanol, thermally-activated MIL-100(Fe) was first exposed to CH3OH at 373 K, purged for 6 h under inert flow at 473 K to remove excess CH3OH, and then exposed to D2O. 0.23 mol CH3OD (mol Fe)-1 were measured upon introduction of D2O subsequent to exposure to methanol, a value coinciding closely with that formed following reaction with methane and N2O (0.27 mol CH3OD (mol Fe)-1)- Figure 5- evidencing the plausibility of methoxy formation through methanol dissociation over Fe2+ sites. The methanol dissociation observed is analogous to water reacting with open-metal sites to reform hydroxyl anions that have to be eliminated during thermal activations steps in MIL-100(Fe) (Figure S11, SI), as also reported previously over Cr2+ sites in MIL-100(Cr).45 The susceptibility of methanol towards dissociation over Fe2+ sites suggests that the formation of methanol intermediates in our experiments cannot be excluded. Regardless of the identity of steps mediating methoxy formation, its stoichiometric formation exclusively over Fe2+ sites appears to precede methanol formation upon extraction with water vapor. A 1:1 correspondence between methoxy concentrations and Fe2+site densities across a range of thermal activation conditions35 suggests that methoxy formation involves the participation of only one active center, and contrasts with prior reports for iron-exchanged zeolites that propose the involvement of two active ’α-oxygen’ sites per methoxy formed (CH4 + 2(O)α → (OH)α + (OCH3)α).57,58