Theoretical understanding of the stability of single-atom catalysts

As a new frontier, the rapid development of single-atom catalysts (SACs) in heterogeneous catalysis has attracted extensive attention since the concept of single-atom catalysis was first coined in 2011 [1,2]. Supported metal single atoms usually possess unique chemical and physical properties and have a special local chemical environment that is distinctly different from conventional supported nanoparticles and metal catalysts. Studies in the past several years have shown four main advantages of SACs: high selectivity, possibly high stability, high atomic efficiency and tunable high activity, as shown in Fig. 1. For efficiency, SAC maximizes the utilization of expensive metals by exposing each single metal atom to reactants. The local configurations of the active centers of an ideal SAC can be highly identical, leading to excellent selectivity compared to supported nanoparticles and metal surfaces that often have multiple yet rather diverse active sites. Unique coordination of single metal atoms with neighboring atoms of support may lead to high activity for specific reactions. All the above merits are on the basis of the stability of catalysts that people are most concerned about. To avoid aggregation under catalytic reaction conditions, single metal atoms can be stabilized by anchoring at specific sites on the support, including embedding and surface-adsorbing. Considering the complex realistic conditions, the stability of SACs may also depend on the surface condition, support type, reactant species, and finite temperature and pressure, which are difficult to ascertain and vary from one system to another. It remains a grand challenge to guide the prediction and design of highly stable and reactive SACs today. Examples illustrating the advantages of single-atom catalysts (SACs) include activity, stability, selectivity and atomic efficiency. (a) High loadings of palladium atoms on TiO2 exhibit high catalytic activity in hydrogenation of C = C bonds [3]. (b) Pt1/CeO2 and Au1/FeOx are stable at 800°C and 400°C in oxidation conditions, respectively [4,5]. (c) Iron single atom embedded on a SiO2 surface exhibits high selectivity for conversion of methane to ethylene, aromatics [6]. (d) SACs are less expensive than supported nanoparticles and metal catalysts. Four types of single atoms on an oxide surface are shown in Fig. 2b. The single atom can be adsorbed on perfect or defective surfaces and can also be embedded (doped) into cation or oxygen vacancy. Most fabricated SACs from experiments are the embedded type, with strong covalent metal–support interaction (CMSI), while the sintering of single atoms and dispersion of nanoparticles should involve the diffusion of supported zerovalent single atoms. Therefore, both the supported and embedded single-atom active centers are possible under realistic conditions and worth thorough investigation. The intrinsic stability of SAC arises from the support-assisted lower chemical potential when compared to nanoparticles. When the free-energy change from nanoparticles to single atoms is negative, nanoparticles can be dispersed to single atoms spontaneously, which leads to thermodynamic stability of SACs (Fig. 2a, black curve). However, even if the free-energy change is positive, single atoms can also be stable when the aggregation barrier is high enough to prevent sintering, which is the kinetic stability of SACs (Fig. 2a, red curve). Following the principles of the atomistic theory of Ostwald ripening developed by W-X Li and coworkers, we have achieved a quantitative description and comparison of the two models by taking the reaction environment, particle size and morphology, support type and defects, and metal–reactant interaction into account [7,8]. We have proposed a complete theoretical model to predict the chemical potential of supported single metal atoms and supported nanoparticles, and taken the CO oxidation reaction as an example to guide the design of stable SACs. As a measure of the degree of change of free energy of a system by adding 1 mole of metal atoms to the system at a given temperature and pressure, the chemical potentials for SACs and metal particles were discussed in our recent work [8]. Both thermodynamic and kinetic criteria are considered to determine the stability of SACs. The thermodynamic part includes (i) the energetics of supported metal particles, which is based on the Gibbs−Thomson (G–T) relation with considering the adsorbed reactants and (ii) the chemical potential of monomers (both the metal adatoms and metal–reactant complexes) on supports. And the kinetic part includes (i) the diffusion barrier of monomers on perfect surfaces and defects and (ii) the barrier of moving one metal atom from a supported metal nanoparticle to a substrate surface with corresponding sintering rate equations. Here, we discuss the support effect on the stability of supported metal atoms first using selected examples [9]. We have found that the quantum primogenic effect plays a vital role in determining the valence states and charge distribution of single-atom gold and the adsorption mode of CO on various supports, as shown in Fig. 2d, which is consistent with results from others [10]. Remarkably, Au1 atoms are positively charged on reductive supports (e.g. CeO2, TiO2) by charge transfer from metal adatoms to support cations, which leads to strong interaction between CO's 5σ orbital and 6 s orbital of Au+, whereas Au1 keeps a zero oxidation state on the irreducible metal dioxides (e.g. ZrO2, HfO2 and ThO2) and has a weak interaction with CO via a bent Au–CO coordination with partial electron donation to the CO π* anti-bonding orbital. For the same substrate, exposing different surfaces also affects the relative stability of single atoms. For example, the stability sequence of doped Pt single atoms on different surfaces of CeO2 follows (110) > (100) > (111) [11]. The high stability of Pt single atoms on (110) surface benefits from the spontaneous formation of an O22− species from two surface oxygen atoms that reduces PtIV to PtII. A different anchoring site for a single atom in a given surface is another significant factor of stability. It has been found that the presence of special defects such as vacancies or steps can improve the stability of an Au single atom on a CeO2(111) surface and the stability sequence is calculated as cation vacancy > steps > oxygen vacancy > perfect (defect-free) surface [8]. Besides the support effect, reactant species also change the stability of SACs. It is now well recognized that sintering is often accelerated by the presence of reducing gases that convert high- and medium-valent metal ions into zerovalent atoms, whereas nanoparticles can also re-disperse to single atoms in some cases. Jones et al. reported that Pt nanoparticles can disperse onto ceria surfaces under an oxygen atmosphere and 800°C [4]. Li and colleagues recently observed noble metal nanoparticles transforming to thermally stable single atoms [12]. Parkinson et al. observed CO-induced coalescence of Pd adatoms supported on the Fe3O4(001) surface at room temperature [13]. Those phenomena can also be attributed to the chemical potential of single atoms affected by reactants. As shown in Fig. 2c, reducing gases such as CO and H2 can react with the lattice oxygen after adsorption onto single metal atoms, which breaks the covalent bond between metal and oxygen, especially for doped atoms at cation vacancy. When all the relevant metal–support covalent bonds are broken by CO, the charge state of Pt changes from positive to neutral. And, simultaneously, the chemical potential of single Pt atoms increases to be higher than Pt nanoparticles. Such Pt0 species become highly mobile with CO adsorption and thus aggregated to particles rapidly by overcoming the diffusion barrier. On the contrary, when Pt nanoparticles are exposed to oxygen atmospheres at high temperature, the chemical potential of separated PtO2 species can become lower than that on nanoparticles. Should there be enough suitable surface sites (such as cation vacancies or steps on ceria) to trap PtO2, Pt nanoparticles will transfer to single atoms on support. (a) Schematic illustration of free-energy diagram of sintering and dispersion processes between Au NPs and SAs. (b) Four types of single atom on oxide support. (c) Aggregation and re-dispersion processes of Pt1/FeOx under CO and O2 atmosphere. (d) Schematic representation of the relationship between Hartree potential and charge transfer on reducible and irreducible substrates, and the molecular orbital of Au+CO species on CeO2 support [9]. (e) Dynamic formation of single-atom catalytic active sites on ceria-supported gold nanoparticles (adapted from [15]). It is worth noting that single metal atoms may also be dynamically generated during the catalytic processes. Based on large-scale ab initio molecular dynamic (AIMD) simulations, we have found that, on reducible oxide-supported gold nanocatalysts [14,15], the gold cation can migrate from the gold nanoparticle to support to catalyse CO oxidation and reintegrate back to the nanoparticle after completing the reaction. This dynamic phenomenon of SACs is named as dynamic single-atom catalysis [15]. Especially, by a combination of ab initio electronic structure and molecular dynamics simulations, as well as a microkinetic simulation, we have shown that, on TiO2-supported Au nanocatalysts, formation of dynamic SACs is the dominant reaction pathway for CO oxidation under oxidizing conditions and T < 400 K [14]. The dynamic formation of single gold atoms under realistic conditions is ultimately attributed to the reducibility of the oxide support that strongly couples with the charge state of gold and makes the single gold atom active during the catalytic processes, which is attributed to dynamic stability. In recent work, we found that dynamic single atoms under reaction conditions account for the notorious size effect in gold nanocatalysts [16]. In summary, intrinsic thermodynamic stability, kinetic stability and dynamic stability are the key factors in determining the reactivity of SACs. Although experimental results present a complicated picture from system to system, it all complies with the rules of chemical potential, which can be quantitatively evaluated from thermodynamic and dynamic aspects. Single-atom catalysis pushes the traditional catalytic theory from a band-structure-based interpretation to a more local chemical-bonding interpretation, where the local chemical coordination and local molecular orbitals dominate the physical and chemical properties, especially the stability and activity. Exploring these intriguing factors on stability of SACs based on theoretical models [7,8] may lead to better and deeper understanding of single-atom catalysis and provide guidance for the rational design of exactly controllable catalytic reactions.