Understanding the fundamentals of a catalytic process remains an intellectual challenge. Now, a method has been developed that can discriminate mass transport phenomena from reaction kinetics at the single-molecule and single-particle levels.
Heterogeneous catalysis plays a key role in the transition towards a more sustainable society1. To make this move a reality, we should be able to synthesize solid catalysts by rational design. However, this scientific endeavour remains a pipe dream because our fundamental understanding of how real-life solid catalysts work is still relatively limited2. The reasons for this are that solid catalysts are multi-component, hierarchically structured and dynamic materials. They are also non-uniform in space and time, leading to the occurrence of inter- and intraparticle heterogeneities2,3,4,5. Only when we are able to fully capture and appreciate these catalyst heterogeneities under operando conditions can we start to tailor a practical solid catalyst for a particular chemical process6.
Reporting in Nature Catalysis, Wenyu Huang, Ning Fang and colleagues have now made an important step towards this ambition by designing a core–shell nanocatalyst model system and subsequently investigating in great depth variations in molecular diffusion and reaction dynamics7. By using single-molecule fluorescence microscopy, a powerful method that is finding its way in the field of heterogeneous catalysis2,3,5,8, the researchers were able to disentangle reaction rates and diffusion constants in space and time at the single-particle level. Interestingly, an increase in catalytic activity was observed, which was explained in terms of nanoconfinement effects within the core–shell catalyst. Figure 1a outlines the strategy employed and shows the structure of the catalyst. The catalytic material consists of ~5 nm Pt nanoparticles assembled on an optically transparent solid SiO2 core, which served as a template on which a mesoporous SiO2 shell was constructed. Interestingly, Huang, Fang and colleagues were able to align the ~2 nm nanopores within the mesoporous SiO2 material, providing molecular ‘highways’ for the reactants towards the supported Pt nanoparticles.
The crux of the study is the combination of a selective probe reaction, catalysed by the supported Pt nanoparticles, and a very sensitive spectroscopic method, which has high precision in locally determining single molecules. This was demonstrated by taking advantage of total internal reflection fluorescence microscopy and the oxidation of non-fluorescent amplex red into fluorescent resorufin in the presence of hydrogen peroxide (Fig. 1b). As a result, it was possible to locally detect the formation of a fluorescent dye molecule at the Pt nanoparticle, and to follow in space and time how this dye molecule diffuses away from the Pt nanoparticle within the 2-nm-sized nanopores of the core–shell nanocatalyst towards the bulk solution surrounding the nanocatalyst particle (Fig. 1c). The thickness of the mesoporous SiO2 material (expressed as Lp) was varied between 0 and 200 nm, thereby systematically changing the distance between the Pt nanoparticle and the outer surface of the catalyst particle.
This unique set of catalyst materials enabled the researchers to discriminate between mass transport phenomena and reaction kinetics. First, the researchers found that the trajectories of the fluorescent resorufin molecules diffusing through the 2-nm-sized nanopores was diverse, and that the molecules travelled at a much slower speed compared with those in the bulk solution. Furthermore, the molecular transport within the catalyst particle could not be captured by a single diffusion coefficient D, and the observed differences in D were explained by a combination of diffusion and adsorption behaviours associated with variable local environments within the solid catalyst. For example, it is argued that resorufin is adsorbed on the hydrophilic surface of the nanopores. This observation of local diffusion coefficients is in line with a recent single-molecule fluorescence study based on a fluid catalytic cracking (FCC) catalyst particle9. Second, local single-molecule turnovers could be determined at spots near the supported Pt nanoparticles. By using modelling, assuming among other aspects Langmuir–Hinshelwood kinetics, Huang, Fang and colleagues could determine the chemical conversion rate constant (keff) and the adsorption–desorption equilibrium constant (KAR). Their results, as summarized in Fig. 1d, reveal that keff increases with increasing thickness of the mesoporous SiO2 overlayer and reaches an enhancement factor of ~7 at a thickness of 120 nm. At the same time, KAR becomes ~2 times smaller for supported Pt nanoparticles confined in nanopores relative to the catalytic solid with no mesoporous overlayer and thus no nanoconfinement effects. The researchers’ findings were explained by a combination of an increased effective concentration of amplex red in the nanopores near the Pt nanoparticles and the change in the molecular adsorption process of amplex red on the Pt surface. It is proposed that the observed nanoconfinement effects can alter the adsorption strength of molecules, although future theoretical and experimental studies are required to provide further evidence for this hypothesis.
A wide variety of materials must be researched if catalytic performance and mass transfer properties are to be correlated with structure and composition of pore-confined solid catalysts. The size and shape of metal nanoparticles must be changed to alter the catalyst performance, while the composition and porosity of the oxide overlayer must be altered to change the molecular diffusion properties. By doing so, one can investigate (1) how the diffusion coefficient D varies with the nanopore diameter and the exact position of the fluorescent molecule within the mesoporous oxide overlayer and (2) what the specific contribution of the oxide overlayer is to the adsorption chemistry of the fluorescent molecule travelling through the nanopores. However, the ultimate experiment is one in which both the organic and inorganic part of catalytic events can be fully captured down to the single-molecule and single-particle levels. On the one hand, scientists must track both single reactant and product molecules in space and time, while at the same time monitoring locally their conversion near the metal nanoparticle with, for example, fluorescence microscopy. On the other hand, the dynamics of the inorganic part of the catalytic solid should be simultaneously captured with, for example, X-ray and/or electron microscopy. A first study in which single-molecule fluorescence microscopy is integrated within a transmission electron microscope has been recently published10 and illustrates the direction in which this field is heading. It should be stressed that such developments not only require new analytical instrumentation, but also powerful data-mining methods to handle the large amount of data acquired. These recent developments, together with the study of Huang, Fang and colleagues7, show that scientists working in the field of catalysis are on their way to developing relevant structure−mass-transfer−reactivity relationships under operando conditions, a prerequisite for the rational design of practical solid catalysts.