Molecular nature of support effects in single-site heterogeneous catalysts: silicavs.alumina

Abstract

Well-defined silica- and alumina-supported catalysts can display very different reactivity, which is often attributed to support effects. Advanced spectroscopic studies in combination with computational chemistry show that the origin of these differences is mainly molecular in nature. Silica partially dehydroxylated at 700 °C features mainly isolated silanols (≡SiOH) as surface functional groups, which can be exploited to generate well-defined and site isolated siloxy metal species. In contrast, alumina, a more complex support, exhibits both hydroxyl groups and highly reactive Al,O Lewis acid and base surface sites. Therefore, it is better understood as a multifunctional and adaptable ligand, which can lead to a variety of surface species, often more reactive (cationic) and/or stable than their known silica counterparts.

---

Fernando Rascón, Christophe Copéret and Raphael Wischert

Dr. Fernando Rascón (left), Prof. Christophe Copéret (center) and Dr. Raphael Wischert (right) recently moved to the Department of Chemistry (Institute of Inorganic chemistry) of ETH Zürich (Switzerland) from C2P2/CPE Lyon (France). The group’s research interests lie in the molecular understanding and rational development (structure-property relationship) of heterogeneous catalysts and functional materials. Thus, besides evaluation of properties (e.g. catalytic performances), the research group is involved in material chemistry, controlled surface chemistry, and characterization of surface species and reaction intermediates using the combination of spectroscopy and computational chemistry. For more information, see: http://www.coperetgroup.ethz.ch/.

Introduction

The support plays a protagonist role in heterogeneous catalysis, including the class of so-called single-site catalysts.¹ For instance, clear differences in catalytic performance in terms of activity, selectivity and stability have been reported for supported silicavs.alumina single-site catalysts for various reactions, such as alkene polymerisation,²hydrogenation,³ and metathesis.⁴ Our aim is to discuss the support effects at a molecular level, focusing on selected examples of these systems. This implies not only describing surface functionalities as ligands of the grafted organometallic entities—as in molecular organometallic chemistry—but also including the effect of the presence or absence of (i) interactions between the surface and the grafted molecular entities or the substrate, (ii) Lewis acidic sites, or “defects”. All of these factors can play a major role on the structure of surface species, whether being pre-catalyst (surface complexes) or reaction intermediates. Because of their effect on the catalytic performance, these factors should be taken into account when designing heterogeneous catalysts.

Silica-supported systems

Generalities on silica

Silica is probably one of the most studied oxide supports; it is an amorphous solid composed of a network of tetrahedral [SiO₄]⁴⁻ units, and it can develop large specific surface areas (100–1000 m² g⁻¹).⁵ Crystalline silicas also exist but they are not used as catalyst supports because of their small surface area (few m² g⁻¹). The surface of amorphous silica surfaces is composed of siloxane bridges (≡Si–O–Si≡) and silanol groups (≡SiOH), whose concentration and types (isolated, geminal or vicinal, Fig. 1a) depend on the temperature of treatment under vacuum or gas flow.^5b,5d Upon thermal treatment, the overall density of silanol groups decreases nearly exponentially with temperature by self-condensation (Fig. 1b).^5e This dehydroxylation process yields water and siloxane bridges of decreasing annularity, and it is accompanied by an increased ratio of isolated silanols.

Fig. 1 a) Types of silanols. b) Effect of temperature on surface-OH coverage (θ_OH) and infrared spectra of silica partially dehydroxylated at 200 (SiO_2-(200)) and 700 °C (SiO_2-(700)). c) Types of siloxane bridges.

Partial dehydroxylation takes place without significant loss of specific surface area for temperatures up to around 700 °C for Aerosil® silica. At this temperature, this solid—referred to as SiO_2-(700)—presents a statistical population of isolated silanols with a coverage of ca. 0.8 OH m⁻². Above this temperature, the surface area starts to drop significantly, and reactive, strained siloxane bridges are formed (Fig. 1c). Thermal treatment on mesoporous silica such as MCM-41 can lead to structural collapse, resulting in a sharp loss of specific surface area, if it is performed too fast or at temperatures above 500–600 °C.^5a Thereafter, we will mainly discuss the surface chemistry of Aerosil® silica (200 m² g⁻¹), a support composed of pure silica nanoparticles having a diameter of ca.12 nm. Nonetheless, such discussion can be easily transposed to mesoporous silica, minor differences being the OH density (within 20%), the stability of the material and the roughness of the surface.⁶

Effect on the structure of surface complexes

With a low density of isolated silanols, SiO_2-(700) has been the support of choice for generating well-defined complexes attached by a single covalent bond to the surface (vide infra).^1a,7 In most cases, grafting perhydrocarbyl transition-metal complexes⁸ on SiO_2-(700) leads to the formation of an alkane and a surface species in which one of the M–C bonds has been replaced by a M–O bond without otherwise significantly modifying the structure of the metal fragment (Fig. 2a,b).⁹ Lower dehydroxylation temperatures provide supports with a higher density of vicinal silanols, leading to a higher amount of bis-grafted species; the highest being obtained for silica treated at temperatures of ca. 200 °C.^2a,10 Grafting is not limited to perhydrocarbyl complexes,¹¹ since alkoxides¹² and amides¹³ can also be used in the same way to access well-defined surface species (Fig. 2c, X = OR or NR₂);^10a,14 here an alcohol and an amine are formed and released, but can be easily removed from the rather hydrophobic surface of dehydroxylated silica by washing.

Fig. 2 a) Grafting of an organometallic compound onto a silica surface. b) Perhydrocarbyl silica-supported surface species. c) Examples of amido- and alkoxy-based surface species.

A more detailed understanding of silica supported systems can be illustrated with [(≡SiO)Re(≡CtBu)(=CHtBu)(CH₂tBu)] (Fig. 3a), a well-defined surface complex, fully characterized by IR, solid-state NMR and EXAFS, and selectively obtained from the grafting of [Re(≡CtBu)(=CHtBu)(CH₂tBu)₂] on SiO_2-(700).^9a,15 Formed selectively as the syn-isomer, the resulting surface complex displays a C–H agostic interaction for its alkylidene ligand, as evidenced by the low J_C–H coupling constant. Further thermal or photochemical treatment of the complex leads to a mixture of syn and anti isomers,^15,16 the latter presenting no agostic interaction. Grafting of isoelectronic and isostructural Mo- and W-imido complexes also selectively generates the corresponding syn-isomer (Fig. 3b, X = CH₂tBu and E = NAr).¹⁷

Fig. 3 a) Thermal interconversion of [(≡SiO)Re(≡CtBu)(=CHtBu)(CH₂tBu)] into its syn/anti isomers. b) Representation of d⁰ ML₄ alkylidyne/alkylidene supported olefin metathesis catalysts (M = Mo, W, Re). c) Disposition of the ligands for d⁰ ML₄ complexes.

Computational studies using various models of siloxy ligands [H₃SiO–, POSS– (POSS– = (c-C₅H₉)₇Si₇O₁₂SiO–), and ≡SiO– of cristobalite models] show that the syn isomer is thermodynamically more stable and presents an agostic interaction resulting from the interaction of the σ(C–H) orbital of the alkylidene ligand with the antibonding σ*(M≡E) orbital of the alkylidyne ligand.¹⁸ It is noteworthy that the simplest possible model of a silica surface, based on a H₃SiO– fragment, reproduces quite well the essential properties of the actual system (structure, syn/anti-ratio and spectroscopic data), which shows that the anchoring point on a silica surface can be viewed just as a molecular silanol ligand.^18b

Finally, for d⁰ alkylidene complexes (Fig. 3b), the substitution of one of the neopentyl ligands by a siloxy provokes a change in their structure from an almost perfect tetrahedron into a distorted one, with an opening of the face trans to the remaining neopentyl ligand X ([OEC face], Fig. 3c). This is in line again with the weaker σ-donor character of the siloxy ligand compared to alkyl or other ligands with more electronegative atoms, e.g. N from amido. In fact, this is observed for all d⁰ ML₄ alkylidene complexes having dissymmetric ligand sets (X = CH₂tBu, NR₂).¹⁹ The nature of the siloxy ligand (a weak σ-donor) can be clearly illustrated by the change of geometry and structure upon grafting of Os(=CHtBu)₂(CH₂tBu)₂—a d²ML₄ tetrahedral complex—on SiO_2-(700), yielding (≡SiO)Os(≡CtBu)(CH₂tBu)₂.²⁰ This surface complex adopts a see-saw structure and bears one alkylidyne and two alkyl ligands (Fig. 4).

Fig. 4 a) Grafting of Os(=CHtBu)₂(CH₂tBu)₂ onto SiO_2-(700). b) Optimised geometry of (≡SiO)Os(≡CtBu)(CH₂tBu)₂ on a periodic surface of silica.

Effects on the reactivity

In numerous Lewis acid-catalyzed reactions such as alkene oxidation²¹ as well as alkene/alkyne metathesis,^4,22silica-grafted species based on lanthanides, group 3 or d⁰ transition-metal complexes often display higher catalytic performance (activity and stability) than their molecular counterparts.^{9d, 14b,23} A simple explanation involves the formation of mononuclear and more electrophilic (more Lewis acidic) centres upon including a siloxy rather than an alkoxide ligand. In the particular case of silica-supported alkylidene complexes, theoretical studies have allowed us to delineate the origin of the reactivity in alkene metathesis.^19,24 Here, the siloxy ligand (a weak σ-donor ligand) sensibly affects the structure and the stability of reaction intermediates and transition states (Fig. 5a). The key step in metathesis is the coordination of the incoming alkene, which requires the distortion of the almost tetrahedral species in order to form the π-alkene complex having a TBP geometry. The approach can only take place from one of the two faces presenting the alkylidene ligand in the proper orientation to form the metallacycle intermediate (attack on [OEC] and [XEC] faces, see Fig. 3c). Overall, the coordination of the alkene through the [OEC] face is associated with the lowest energy trigonal bipyramidal (TBP) transition state (TS), because the strong σ-donor ligand (X = alkyl or amide) occupies an apical position trans to the incoming alkene, still far from the metal center. The two other strong σ-donor ligands (alkylidene and E = alkylidyne or imido) share the basal plane of the TBP with the weak σ-donor ligand (siloxy). The other approach (on the [XEC] face) is disfavored because it leads to a TS with three strong σ-donor ligands in the basal plane of the TBP. Finally, note that the preferred face of attack is also the most open one in the alkylidene precursor, thus showing that a disymmetric system prepares the catalyst, which contributes to its higher reactivity. In turn, the analysis of these ligand effects has led to a deeper understanding of the performance of alkene metathesis catalysts.^19,24,25 Based on this structure–activity relationship, better performing silica-supported catalysts (Fig. 5b)^4,14d,14e,26 and even homogeneous systems (Fig. 5c) have been developed.²⁷

Fig. 5 a) Mechanism for alkene metathesis. b) Specific role of each ligand for Mo-catalysts from structure–activity relationship. c) Example of a highly efficient dissymmetric homogeneous alkene metathesis catalyst containing a stereogenic Mo centre.^27b

Effects on the stability

Grafting on silica materials translates into site isolation, which brings higher stability by impeding bimolecular decomposition pathways, including self-aggregation. This has been shown in alkene metathesis by the higher stability of silica supported systems compared to their molecular equivalents (vide supra).^17a Such higher stability also enables supported alkyl-alkylidene metal complexes to be used as pre-catalysts in alkane metathesis, which is performed at 150 °C.^28,29 Further treatment of silica-supported perhydrocarbyl Zr and Ta-complexes with H₂ yields the corresponding monomeric hydrides,³⁰ whose molecular equivalents are not known in the absence of bulky ligands (Fig. 6). This is due to the strong M–O bond early transition metals form with the silica support. The metal centres experience additional stabilisation by the formation of bond(s) with the silica surface upon the opening of adjacent siloxane bridges. These supported hydrides readily activate the C–H bonds of unactivated alkanes at low temperature and display a reactivity unprecedented in molecular chemistry: low temperature hydrogenolysis of alkanes³¹ and polymers,³²alkane metathesis,³³ and homologation of methane.³⁴ Supported tantalum hydrides can even cleave the N₂ molecule.³⁵

Fig. 6 Zirconium and tantalum surface metal hydrides obtained by H₂ treatment.

Alumina-supported systems

Generalities on alumina

Alumina (Al₂O₃) is a more complex support, in spite of its higher crystallinity. In fact, depending on its preparation, which typically consists in treating aluminium hydroxide precursors at high temperature, numerous polymorphs can be formed.³⁶α-Al₂O₃ (corundum), as the thermodynamically most stable phase is formed above 1000 °C and has a very low surface area (<20 m² g⁻¹). It is used in catalytic processes as a chemically inert support material for metals when high mechanical and thermal resistance is required (for example in the Haber–Bosch process for ammonia synthesis). Metastable aluminas formed by dehydration at 500–800 °C are known as transition or activated aluminas. They are high-surface area (80–250 m² g⁻¹) porous materials which find wide application as catalysts or as catalyst supports, γ-Al₂O₃ being the most widely used.^36,37 Despite considerable efforts devoted to its study, the exact bulk structure and surface sites of γ-Al₂O₃ are still not fully understood and are a matter of debate.³⁸ Extensive investigations by periodic DFT calculations in combination with experimental data have provided a more detailed structural description of the bulk of γ-Al₂O₃³⁹ and its surface sites.^38,40 The bulk contains aluminium ions in tetrahedral (25%) and octahedral (75%) coordination. The unit cell of the most abundant (110) termination (ca. 80%)^40b,41 is composed of one tricoordinated Al_III and two types of tetracoordinated Al_IV sites, namely Al_IVa and Al_IVb (Fig. 7a), resulting from tetrahedral and octahedral bulk Al atoms, while the less abundant (100) termination (ca. 10%) exposes only Al_V sites. The Lewis acidity of these low-coordinated Al atoms follows the order Al_III≫Al_IVb>Al_V>Al_IVa. The surface also exposes oxygen atoms with Lewis basicity O_2a/b > O_3a/b.^40b However, these sites are partially occupied by OH groups or protons (generated by dissociation of H₂O on Al,O sites), even after high temperature pretreatment. The coordination of these hydroxyls (μ₁, μ₂ and μ₃), depends on the type and number of Al sites to which they are bound. In addition, non-dissociated H₂O molecules coordinated to Al sites can also be stabilized on the surface, increasing the number of hydroxyl types (Fig. 7b). In comparison to silica, the corresponding IR spectrum of the OH-region (see for example that for γ-Al₂O_3-(500) in Fig. 8a) is therefore rather complex, but it was nevertheless possible to fully assign the bands with the aid of DFT calculations.⁴⁰

Fig. 7 a–d) Structure of the (110) termination of γ-Al₂O₃: a) fully dehydroxylated surface (s0); b) covered by 3 H₂O per unit cell, ca. 9 OH nm⁻² (s3); c) with free Al_III site, covered by 1 H₂O per unit cell, ca. 3 OH nm⁻² (s1); d) with free Al_III site, covered by 2 H₂O per unit cell, ca. 6 OH nm⁻² (s2). Note that for the most stable surfaces, covered by 1 or 2 H₂O, Al_III, the most Lewis acidic site is occupied by an OH group. Only the top two layers of the periodical slab are represented. A dashed line indicates the unit cell. Al: yellow, O originating from the γ-Al₂O₃ bulk: red, O originating from H₂O dissociation: purple, H: white balls, Al_IVb after surface reconstruction: green circle. All distances are given in Å.

Fig. 8 a) Infrared spectrum of γ-Al₂O₃ dehydroxylated at 500 °C in vacuum (1.77 × 10⁻³ Pa). Assignment of the OH-bands according to Digne et al.^40b b–c) Influence of thermal treatment on the surface of γ-alumina: b) Surface OH coverage (θ_OH, dotted line); Surface area (S_BET, solid line). c) Abundance of Al–CH₃ (C–H activation) upon contact with CH₄. Adapted from ref. 43.

Here, the stabilisation of the surface upon hydration is worth mentioning. In contrast to the less abundant (100), the (110) termination is predicted to be thermodynamically unstable in its fully dehydrated form (s0).⁴⁰

Upon thermal treatment the OH density (θ_OH) decreases exponentially with temperature,^42,43 reaching 2 and 0.6 OH nm⁻² at 500 and 700 °C, respectively (Fig. 8b). Lewis acid sites are formed by removal of chemisorbed H₂O (dissociated and non-dissociated) by thermal treatments above 400 °C. The most reactive, tricoordinated Al_III sites (“defects” or α-sites⁴⁴) are noteworthy in terms of their reactivity: i) the Al_III,O pairs stoichiometrically activate the C–H bond of methane to generate Al–CH₃ and O–H surface species,^33c,43,45 and ii) are therefore most probably involved in the low-temperature H/D-exchange reaction of CH₄ and D₂ into CH₃D and HD mixtures (via C–H activation).^42,46 The density of these sites increases up to about 700 °C, reaching a maximum of 0.03 nm⁻², and then vanishes as the bulk gradually transforms into θ- and α-alumina, the thermodynamically more stable phases. This process is accompanied by a sharp loss of surface area (Fig. 8b).⁴³ Therefore γ-Al₂O₃ is most reactive towards CH₄ dissociation at an OH-coverage of ca. 0.6 OH nm⁻², which strongly suggests the presence of the highly reactive Al_III. This is somewhat counterintuitive as these sites should be in principle occupied by OH groups as discussed above.⁴⁷

Recent studies⁴³ reveal that Al_III sites can indeed exist in the presence of surface hydroxylation up to relatively high OH-coverages of 6 OH nm⁻² (corresponding to a pretreatment temperature of ca. 400 °C, s2, see Fig. 7d). The unexpected occurrence of a stable s2 surface with free Al_III sites is largely due to a reorganization of the surface, with Al sites originally in truncated octahedral geometry ending up upon hydration in a more stable tetrahedral coordination. However, because of the high rigidity of the s2 surface (reconstruction, coordination of Al_III by a 2nd-layer O atom), this Al_III site is unreactive towards CH₄ activation, in line with the experimental findings.^43,45

At lower water coverage (s1, θ_OH = 3 OH nm⁻², Fig. 7c) the probability of finding Al_III sites is lower, however they are highly reactive towards CH₄, as observed experimentally. This is due to the formation of highly reactive frustrated Lewis acid-base pairs:⁴⁸ the Lewis acidity of Al_III is high on s1 and the basicity of the oxygen partner (O_3a, Fig. 7d) is strongly increased by the hydration of the adjacent Al_IVa sites. In summary, the hydroxylation of the surface has a dual role: firstly, as stated, it stabilizes the otherwise unstable (110) termination and thus enables Al_III sites. Second, it maximises the reactivity of these sites at an optimal water coverage, modeled by s1.

These findings illustrate the richness and complexity of the alumina surface. Like silica, alumina features OH groups but these exhibit different coordinations, in contrast to silica (unless the latter is treated at low temperature). The main difference, however, consists in the presence of Lewis acidic and basic Al and O sites. Unlike silica, where all the silicon atoms are fixed in a rather regular, tetrahedral environment and are therefore chemically inert, alumina exhibits highly reactive, Lewis acidic Al atoms thanks to its configurational flexibility. At the same time, the crucial role of oxygen basicity should be reemphasized for reactions where Al,O pairs are involved, such as the activation of C–H bonds.

Overall, γ-alumina exhibits a complex and flexible surface with a rich palette of surface sites and functionalities; it is therefore not surprising that this support exerts a profound effect on the structure, reactivity and stability of surface species involved in catalytic events, contributing to the rationale of the so-called support effect.

Effects on the structure of surface complexes

The surface hydroxyls are involved in the grafting of organometallic complexes in several cases.^{2b, 49} On alumina (either γ or δ) partially dehydroxylated at 500 °C, where θ_OH is ca. 2 OH nm⁻², the grafting of Zr(CH₂tBu)₄ leads to the full consumption of surface OH groups and to the formation of 2 equiv of tBuCH₃, yielding the bis-grafted species (Al_SO)₂Zr(CH₂tBu)₂.⁵⁰ A detailed study combining IR, EXAFS, solid-state NMR spectroscopy and computational modelling shows that several surface species are present, but in particular, that cationic Zr complexes are formed, as evidenced by the formation of neopentyl aluminate (Fig. 9a).

Fig. 9 a) Grafting of M(CH₂tBu)₄ onto Al₂O_3-(500) leading to cationic species. b) Interactions of (Al_SO)W(≡CtBu)(CH₂tBu)₂ with neighboring surface hydroxyls on alumina. c) Grafting of CH₃ReO₃ onto Al₂O_3-(500).

This behaviour evidences the multifunctional nature of alumina surfaces. In fact, analogous cationic surface species have also been observed with other group 4 transition metal complexes, such as Cp*ZrMe₃⁵¹ or Hf(CH₂tBu)₄,⁵² as well as lanthanide and actinide complexes.^{3a, 53} In contrast, grafting of W(≡CtBu)(CH₂tBu)₃—containing a less electrophilic metal centre—leads to partial consumption of surface OH groups and to the selective grafting on Al_IVOH, generating the monografted surface species (Al_SO)W(≡CtBu)(CH₂tBu)₂, where the hydrocarbyl moieties interact with the remaining surface hydroxyls (Fig. 9b). Here, the formation of cationic species is not observed.⁵⁰ Grafting of CH₃ReO₃ is also noteworthy: it does not undergo chemical reaction with OH-groups but attaches to the surface via multiple Lewis acid–base interactions involving the coordination of its oxo ligands to Al sites on the one hand and that of the Re center to surface O atoms on the other hand (Fig. 9c). Additionally, Al_SCH₂ReO₃ species are formed via C–H activation of the methyl ligand of CH₃ReO₃ on Al,O sites.⁵⁴

Effects on the reactivity

As stated in the introduction, major differences in reactivity have been found for silicavs.alumina supported single-site catalysts. In the field of polymerisation, in particular for catalysts based on d⁰ early transition-metals, the major differences arise from the fact that supported systems on silica generate neutral species, while cationic species are formed on alumina (vide supra). The same is probably true for the hydrogenation catalysts, based on actinide, lanthanide, group 3 or early transition-metals.^{3a, 51a, 53}Similar strong support effects are known for Re-based catalysts in the field of alkene metathesis. While Re₂O₇/SiO₂ is unreactive, one of the best performing systems is Re₂O₇/γ-Al₂O₃,⁵⁵ which even becomes tolerant to functionalized alkenes such as methyl oleate when Me₄Sn is used as an additive.⁵⁶ Despite years of research,^55a poor molecular insight has emerged. In fact, the key alkylidene species which would explain the olefin metathesis activity has never been observed, and its formation from the Re=O species remains obscure. However, accumulated evidence suggests that it could be generated through a pseudo-Wittig reaction.⁵⁷ In particular, the origin of the reactivity of Re₂O₇/γ-Al₂O₃ has been linked to the Lewis acidic properties of γ-alumina,⁵⁵ but so far it has been difficult to obtain a more detailed understanding of the actual structure of the active sites or of the ultimate role of the alkyl tin additive. One of the standing difficulties studying this catalytic system is the low content of active sites, accounting for only 2% of the surface Re centres.^{57d, 58} This corresponds to a density of ca. 0.04 active sites per nm², a value coinciding with the number of “defect” sites present on γ-alumina partially dehydroxylated at 500 °C (0.03 nm⁻², vide supra). While possibly fortuitous, such correspondence suggests the implication of “defects” for the generation of the active sites. In this respect, studies on a related model system, CH₃ReO₃ supported on γ-Al₂O₃ dehydroxylated at 500 °C (CH₃ReO₃/γ-Al₂O₃),^54,59 are insightful, as both systems possess the same selectivity in 2-butene metathesis at low contact times (Z/E = 0.4) and thus possibly similar active sites.^59c For CH₃ReO₃/γ-Al₂O₃ it was shown that the active sites—or more precisely—the precursor and resting state of the active site is (Al_s)CH₂ReO₃, formed by the C–H activation of CH₃ReO₃ on reactive Al,O pairs upon grafting (Fig. 9c).^54,59c It is noteworthy that CH₃ReO₃/γ-Al₂O₃, where the key alkylidene ligand is already in place in a masked form as (Al_s)CH₂ReO₃, does not require activating agents to be tolerant to functionalized alkenes.^59a This points to the role of Me₄Sn in the metathesis of functionalized alkenes with Re₂O₇/γ-Al₂O₃, viz generating the alkylidene ligand. In the absence of Me₄Sn, Re₂O₇/γ-Al₂O₃ is inactive with such polar substrates because of the poisoning of the necessary Lewis acid sites by the ester group—a Lewis base. These findings suggest that the beneficial effect of additives like Me₄Sn on Re₂O₇/γ-Al₂O₃ arises from the in situ formation and subsequent grafting of CH₃ReO₃ or species alike. Indeed, CH₃ReO₃ can be efficiently prepared by the reaction of Me₄Sn with Re₂O₇.⁶⁰ The systems CH₃ReO₃/γ-Al₂O₃ and Re₂O₇/γ-Al₂O₃ stand as examples of how catalytic activity originates from the “defect” sites of the support.

Effects on the stability

While silica and alumina-supported tantalum alkane metathesis catalysts do not display substantial differences in terms of support effects, tungsten-based systems are a noteworthy example, where the alumina-supported systems outperform their silica-supported counterparts, the latter being almost inactive.⁶¹ Here, the difference is mainly due to a higher stability of the active sites. This is particularly evident during the synthesis of silica-supported tungsten hydrides, for which only a small fraction of the (≡SiO)W(≡CtBu)(CH₂tBu)₂ species is converted into the corresponding hydrides on silica because of sintering and formation of aggregates, while rather well-defined tungsten hydrides are generated from the corresponding isostructural alumina supported system, (Al_sO)W(≡CtBu)(CH₂tBu)₂. Such difference has been explained by the stabilization of tungsten oxo species by adjacent Lewis acidic sites.^33b,62 This shows that Lewis acidic sites do not only provide a way to activate metal centers but can also participate in stabilizing reactive species.

Conclusions and perspectives

In the selected examples presented, we have shown the molecular nature of the interaction of surface functionalities of oxides, namely silica and alumina, with metal centers. These interactions can lead to highly reactive, yet stable species,⁶³ whose catalytic performance can surpass those found in molecular chemistry. While partially dehydroxylated silica can be viewed in a simplified way as a molecular siloxy ligand (≡SiOH) with the advantage of site isolation, alumina exhibits a more complex surface chemistry because of its multifunctional nature by the presence of Brønsted (OH) and Lewis (Al) acidic as well as Lewis basic (O) sites. These diverse environments lead to a variety of effects ranging from the formation of unusual species (cationic centres in group 4 polymerisation catalysts and masked carbenes “Al_s(CH₂)ReO₃” in CH₃ReO₃/γ-Al₂O₃) to the stabilization of surface complexes (e.g.Group 4–6 metal hydrides) with unprecedented reactivity. We have demonstrated that understanding supported catalysts requires a deep knowledge of the surface at a molecular level. Indeed, many heterogeneous catalysts display support effects, but their origin remains poorly understood. Considering the complexity of heterogeneous catalysts, it is clear that future research efforts will have to involve the development of yet more advanced spectroscopic techniques (in situ/operando) in combination with computational chemistry, along with detailed kinetic investigations and more controlled preparation processes. The better understanding of surfaces and interfaces in catalytic systems is crucial for the rational development of catalytic materials, which are the key for sustainable development.

References

1. (a) C. Copéret, M. Chabanas, R. Petroff Saint-Arroman and J.-M. Basset, Angew. Chem., Int. Ed., 2003, 42, 156; (b) J. M. Thomas, R. Raja and D. W. Lewis, Angew. Chem., Int. Ed., 2005, 44, 6456; (c) M. Tada and Y. Iwasawa, Coord. Chem. Rev., 2007, 251, 2702.
2. (a) D. G. H. Ballard, Adv. Catal., 1973, 23, 263; (b) D. G. H. Ballard, Coord. Polym., 1975, 223; (c) Y. I. Yermakov, B. N. Kuznetsov and V. A. Zakharov, Stud. Surf. Sci. Catal., 1981, 8, 59.
3. (a) T. J. Marks, Acc. Chem. Res., 1992, 25, 57; (b) C. Copéret, in The Handbook of Homogeneous Hydrogenation, ed. Johannes G. de Vries and C. J. Elsevier, Wiley-VCH, Weinheim, 2008, 111–151.
4. C. Copéret, Dalton Trans., 2007, 5498.
5. (a) A. P. Legrand, The Surface Properties of Silica, Wiley, New York, 1998; (b) B. A. Morrow, in Stud. Surf. Sci. Catal., ed. J. L. G. Fierro, Elsevier, 1990, vol. 57, Part 1, A161–A224; (c) C. Setzer, G. van Essche and N. Pryor, in Handbook of Porous Solids, ed. F. Schüth, K. Sing and J. Weitkamp, Wiley-VCH, Weinheim, 2008, 1543–1591; (d) B. A. Morrow and A. J. Mcfarlan, J. Non-Cryst. Solids, 1990, 120, 61; (e) L. T. Zhuravlev, Colloids Surf., A, 2000, 173, 1.
6. (a) I. G. Shenderovich, G. Buntkowsky, A. Schreiber, E. Gedat, S. Sharif, J. Albrecht, N. S. Golubev, G. H. Findenegg and H. H. Limbach, J. Phys. Chem. B, 2003, 107, 11924; (b) J. Gath, G. L. Hoaston, R. L. Vold, R. Berthoud, C. Copéret, M. Grellier, S. Sabo-Etienne, A. Lesage and L. Emsley, Phys. Chem. Chem. Phys., 2009, 11, 6962.
7. J.-M. Basset, J. P. Candy and C. Copéret, in Comprehensive Organometallic Chemistry III, ed. H. C. Robert and D. M. P. Mingos, Elsevier, Oxford, 2007, 499–553.
8. (a) R. R. Schrock and G. W. Parshall, Chem. Rev., 1976, 76, 243; (b) R. R. Schrock, Acc. Chem. Res., 1979, 12, 98; (c) R. R. Schrock, Acc. Chem. Res., 1986, 19, 342.
9. (a) M. Chabanas, A. Baudouin, C. Copéret and J.-M. Basset, J. Am. Chem. Soc., 2001, 123, 2062; (b) R. Petroff Saint-Arroman, M. Chabanas, A. Baudouin, C. Copéret, J.-M. Basset, A. Lesage and L. Emsley, J. Am. Chem. Soc., 2001, 123, 3820; (c) M. Chabanas, E. A. Quadrelli, B. Fenet, C. Copéret, J. Thivolle-Cazat, J.-M. Basset, A. Lesage and L. Emsley, Angew. Chem., Int. Ed., 2001, 40, 4493; (d) R. Petroff Saint-Arroman, J.-M. Basset, F. Lefebvre and B. Didillon, Appl. Catal., A, 2005, 290, 181; (e) E. Le Roux, M. Taoufik, M. Chabanas, D. Alcor, A. Baudouin, C. Copéret, J. Thivolle-Cazat, J.-M. Basset, A. Lesage, S. Hediger and L. Emsley, Organometallics, 2005, 24, 4274.
10. (a) A. O. Bouh, G. L. Rice and S. L. Scott, J. Am. Chem. Soc., 1999, 121, 7201; (b) L. Lefort, M. Chabanas, O. Maury, D. Meunier, C. Copéret, J. Thivolle-Cazat and J.-M. Basset, J. Organomet. Chem., 2000, 593, 96.
11. (a) R. R. Schrock, Chem. Rev., 2002, 102, 145; (b) R. R. Schrock, Chem. Rev., 2009, 109, 3211.
12. N. Y. Turova, E. P. Turevskaya, V. G. Kessler and M. I. Yanovskaya, The Chemistry of Metal Alkoxides, Springer, US, 2002.
13. M. Lappert, A. Protchenko, P. Power and A. Seeber, Metal Amide Chemistry, John Wiley & Sons, Ltd, Chinchester, UK, 2008.
14. (a) C. Roveda, T. L. Church, H. Alper and S. L. Scott, Chem. Mater., 2000, 12, 857; (b) J. Jarupatrakorn and T. D. Tilley, J. Am. Chem. Soc., 2002, 124, 8380; (c) H. Weissman, K. N. Plunkett and J. S. Moore, Angew. Chem., Int. Ed., 2006, 45, 585; (d) F. Blanc, J. Thivolle-Cazat, J.-M. Basset, C. Copéret, A. S. Hock, Z. J. Tonzetich, A. Sinha and R. R. Schrock, J. Am. Chem. Soc., 2007, 129, 1044; (e) F. Blanc, R. Berthoud, A. Salameh, J.-M. Basset, C. Copéret, R. Singh and R. R. Schrock, J. Am. Chem. Soc., 2007, 129, 8434; (f) F. Blanc, R. Berthoud, C. Copéret, A. Lesage, L. Emsley, R. Singh, T. Kreickmann and R. R. Schrock, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 12123.
15. M. Chabanas, A. Baudouin, C. Copéret, J.-M. Basset, W. Lukens, A. Lesage, S. Hediger and L. Emsley, J. Am. Chem. Soc., 2003, 125, 492.
16. A. Lesage, L. Emsley, M. Chabanas, C. Copéret and J.-M. Basset, Angew. Chem., Int. Ed., 2002, 41, 4535.
17. (a) F. Blanc, C. Copéret, J. Thivolle-Cazat, J.-M. Basset, A. Lesage, L. Emsley, A. Sinha and R. R. Schrock, Angew. Chem., Int. Ed., 2006, 45, 1216; (b) B. Rhers, A. Salameh, A. Baudouin, E. A. Quadrelli, M. Taoufik, C. Coperet, F. Lefebvre, J.-M. Basset, X. Solans-Monfort, O. Eisenstein, W. W. Lukens, L. P. H. Lopez, A. Sinha and R. R. Schrock, Organometallics, 2006, 25, 3554.
18. (a) X. Solans-Monfort, E. Clot, C. Copéret and O. Eisenstein, Organometallics, 2005, 24, 1586; (b) X. Solans-Monfort, J. S. Filhol, C. Copéret and O. Eisenstein, New J. Chem., 2006, 30, 842; (c) A. Poater, X. Solans-Monfort, E. Clot, C. Copéret and O. Eisenstein, Dalton Trans., 2006, 3077.
19. X. Solans-Monfort, C. Copéret and O. Eisenstein, J. Am. Chem. Soc., 2010, 132, 7750.
20. R. Berthoud, N. Rendon, F. Blanc, X. Solans-Monfort, C. Copéret and O. Eisenstein, Dalton Trans., 2009, 5879.
21. S. T. Oyama, in Mechanisms in Homogeneous and Heterogeneous Epoxidation Catalysis, ed. S. T. Oyama, Elsevier, Amsterdam, 2008, 3–99.
22. W. Zhang and J. S. Moore, Adv. Synth. Catal., 2007, 349, 93.
23. (a) G. Gerstberger and R. Anwander, Microporous Mesoporous Mater., 2001, 44, 303; (b) R. Petroff Saint-Arroman, B. Didillon, A. de Mallmann, J.-M. Basset and F. Lefebvre, Appl. Catal., A, 2008, 337, 78.
24. (a) X. Solans-Monfort, E. Clot, C. Copéret and O. Eisenstein, J. Am. Chem. Soc., 2005, 127, 14015; (b) A. Poater, X. Solans-Monfort, E. Clot, C. Copéret and O. Eisenstein, J. Am. Chem. Soc., 2007, 129, 8207.
25. A. M. Leduc, A. Salameh, D. Soulivong, M. Chabanas, J.-M. Basset, C. Copéret, X. Solans-Monfort, E. Clot, O. Eisenstein, V. P. W. Böhm and M. Röper, J. Am. Chem. Soc., 2008, 130, 6288.
26. (a) F. Blanc, N. Rendon, R. Berthoud, J.-M. Basset, C. Copéret, Z. J. Tonzetich and R. R. Schrock, Dalton Trans., 2008, 3156; (b) N. Rendon, R. Berthoud, F. Blanc, D. Gajan, T. Maishal, J.-M. Basset, C. Copéret, A. Lesage, L. Emsley, S. C. Marinescu, R. Singh and R. R. Schrock, Chem.–Eur. J., 2009, 15, 5083.
27. (a) S. J. Malcolmson, S. J. Meek, E. S. Sattely, R. R. Schrock and A. H. Hoveyda, Nature, 2008, 456, 933; (b) M. M. Flook, A. J. Jiang, R. R. Schrock, P. Müller and A. H. Hoveyda, J. Am. Chem. Soc., 2009, 131, 7962.
28. (a) C. Copéret, O. Maury, J. Thivolle-Cazat and J.-M. Basset, Angew. Chem., Int. Ed., 2001, 40, 2331; (b) E. Le Roux, M. Chabanas, A. Baudouin, A. de Mallmann, C. Copéret, E. A. Quadrelli, J. Thivolle-Cazat, J.-M. Basset, W. Lukens, A. Lesage, L. Emsley and G. J. Sunley, J. Am. Chem. Soc., 2004, 126, 13391; (c) F. Blanc, C. Copéret, J. Thivolle-Cazat and J.-M. Basset, Angew. Chem., Int. Ed., 2006, 45, 6201; (d) F. Blanc, J. Thivolle-Cazat, J.-M. Basset and C. Copéret, Chem.–Eur. J., 2008, 14, 9030.
29. This reaction can also be performed with the combination of dehydrogenation and alkene metathesis catalysts, see: (a) R. L. Burnett and T. R. Hughes, J. Catal., 1973, 31, 55; (b) A. S. Goldman, A. H. Roy, Z. Huang, R. Ahuja, W. Schinski and M. Brookhart, Science, 2006, 312, 257.
30. (a) V. A. Zakharov, V. K. Dudchenko, E. Paukstis, L. G. Karakchiev and Y. I. Yermakov, J. Mol. Catal., 1977, 2, 421; (b) J. Schwartz and M. D. Ward, J. Mol. Catal., 1980, 8, 465; (c) Y. I. Yermakov, Y. A. Ryndin, O. S. Alekseev, D. I. Kochubei, V. A. Shmachkov and N. I. Gergert, J. Mol. Catal., 1989, 49, 121; (d) V. Vidal, A. Theolier, J. Thivolle-Cazat, J.-M. Basset and J. Corker, J. Am. Chem. Soc., 1996, 118, 4595; (e) F. Rataboul, A. Baudouin, C. Thieuleux, L. Veyre, C. Copéret, J. Thivolle-Cazat, J.-M. Basset, A. Lesage and L. Emsley, J. Am. Chem. Soc., 2004, 126, 12541.
31. (a) C. Lécuyer, F. Quignard, A. Choplin, D. Olivier and J. M. Basset, Angew. Chem., Int. Ed. Engl., 1991, 30, 1660; (b) M. Chabanas, V. Vidal, C. Copéret, J. Thivolle-Cazat and J.-M. Basset, Angew. Chem., Int. Ed., 2000, 39, 1962.
32. V. R. Dufaud and J.-M. Basset, Angew. Chem., Int. Ed., 1998, 37, 806.
33. (a) V. Vidal, A. Theolier, J. Thivolle-Cazat and J.-M. Basset, Science, 1997, 276, 99; (b) J.-M. Basset, C. Copéret, D. Soulivong, M. Taoufik and J. Thivolle-Cazat, Acc. Chem. Res., 2010, 43, 323; (c) C. Copéret, Chem. Rev., 2010, 110, 656.
34. (a) D. Soulivong, C. Copéret, J. Thivolle-Cazat, J.-M. Basset, B. M. Maunders, R. B. A. Pardy and G. J. Sunley, Angew. Chem., Int. Ed., 2004, 43, 5366; (b) D. Soulivong, S. Norsic, M. Taoufik, C. Copéret, J. Thivolle-Cazat, S. Chakka and J.-M. Basset, J. Am. Chem. Soc., 2008, 130, 5044.
35. P. Avenier, M. Taoufik, A. Lesage, X. Solans-Monfort, A. Baudouin, A. de Mallmann, L. Veyre, J.-M. Basset, O. Eisenstein, L. Emsley and E. A. Quadrelli, Science, 2007, 317, 1056.
36. P. Euzen, P. Raybaud, X. Krokidis, H. Toulhoat, J.-L. Le Loarer, J.-P. Jolivet and C. Froidefond, in Handbook of Porous Solids, ed. F. Schüth, K. Sing and J. Weitkamp, Wiley-VCH, Weinheim, 2008, 1591–1677.
37. L. K. Hudson, C. Misra, A. J. Perrotta, K. Wefers and F. S. Williams, in Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH, Weinheim, 2000, vol. 2, 339–358.
38. M. Digne, P. Raybaud, P. Sautet, B. Rebours and H. Toulhoat, J. Phys. Chem. B, 2006, 110, 20719.
39. X. Krokidis, P. Raybaud, A. E. Gobichon, B. Rebours, P. Euzen and H. Toulhoat, J. Phys. Chem. B, 2001, 105, 5121.
40. (a) M. Digne, P. Sautet, P. Raybaud, P. Euzen and H. Toulhoat, J. Catal., 2002, 211, 1; (b) M. Digne, P. Sautet, P. Raybaud, P. Euzen and H. Toulhoat, J. Catal., 2004, 226, 54.
41. (a) J. P. Beaufils and Y. Barbaux, J. Chim. Phys. Phys.- Chim. Biol., 1981, 78, 347; (b) P. Nortier, P. Fourre, A. B. M. Saad, O. Saur and J. C. Lavalley, Appl. Catal., 1990, 61, 141.
42. H. Knözinger and P. Ratnasamy, Catal. Rev. Sci. Eng., 1978, 17, 31.
43. R. Wischert, C. Copéret, F. Delbecq and P. Sautet, Angew. Chem., Int. Ed., 2011, 50, 3202.
44. J. Peri, J. Phys. Chem., 1966, 70, 3168.
45. J. Joubert, A. Salameh, V. Krakoviack, F. Delbecq, P. Sautet, C. Copéret and J.-M. Basset, J. Phys. Chem. B, 2006, 110, 23944.
46. (a) J. G. Larson and W. K. Hall, J. Phys. Chem., 1965, 69, 3080; (b) P. J. Robertson, M. S. Scurrell and C. Kemball, J. Chem. Soc., Faraday Trans. 1, 1975, 71, 903; (c) L. Quanzhi and Y. Amenomiya, Appl. Catal., 1986, 23, 173; (d) J. S. J. Hargreaves, G. J. Hutchings, R. W. Joyner and S. H. Taylor, Appl. Catal., A, 2002, 227, 191.
47. A recent investigation has highlighted the role of these Lewis acid sites in the non-dissociative adsorption of CO and H₂ on alumina pre-treated at high temperature: E. N. Gribov, O. Zavorotynska, G. Agostini, J. G. Vitillo, G. Ricchiardi, G. Spoto and A. Zecchina, Phys. Chem. Chem. Phys., 2010, 12, 6474.
48. (a) W. Tochtermann, Angew. Chem., Int. Ed. Engl., 1966, 5, 351; (b) D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2010, 49, 46.
49. (a) R. L. Burwell, J. Catal., 1984, 86, 301; (b) D. G. H. Ballard, J. Polym. Sci., Polym. Chem. Ed., 1975, 13, 2191; (c) P. J. Toscano and T. J. Marks, J. Am. Chem. Soc., 1985, 107, 653.
50. J. Joubert, F. Delbecq, P. Sautet, E. Le Roux, M. Taoufik, C. Thieuleux, F. Blanc, C. Copéret, J. Thivolle-Cazat and J.-M. Basset, J. Am. Chem. Soc., 2006, 128, 9157.
51. (a) K. H. Dahmen, D. Hedden, R. L. Burwell and T. J. Marks, Langmuir, 1988, 4, 1212; (b) M. Jezequel, V.r. Dufaud, M. J. Ruiz-Garcia, F. Carrillo-Hermosilla, U. Neugebauer, G. P. Niccolai, F. Lefevre, F. Bayard, J. Corker, S. Fiddy, J. Evans, J.-P. Broyer, J. Malinge and J.-M. Basset, J. Am. Chem. Soc., 2001, 123, 3520.
52. M. Delgado, C. C. Santini, F. Delbecq, R. Wischert, B. Le Guennic, G. Tosin, R. Spitz, J.-M. Basset and P. Sautet, J. Phys. Chem. C, 2010, 114, 18516.
53. M. S. Eisen and T. J. Marks, J. Am. Chem. Soc., 1992, 114, 10358.
54. (a) A. Salameh, J. Joubert, A. Baudouin, W. Lukens, F. Delbecq, P. Sautet, J.-M. Basset and C. Copéret, Angew. Chem., Int. Ed., 2007, 46, 3870; (b) A. Salameh, A. Baudouin, J.-M. Basset and C. Copéret, Angew. Chem., Int. Ed., 2008, 47, 2117; (c) R. Wischert, C. Copéret, F. Delbecq and P. Sautet, ChemCatChem, 2010, 2, 812.
55. (a) J. C. Mol, Catal. Today, 1999, 51, 289; (b) J. C. Mol, Top. Catal., 2004, 27, 97; (c) J. C. Mol, J. Mol. Catal. A: Chem., 2004, 213, 39.
56. E. Verkuijlen, F. Kapteijn, J. C. Mol and C. Boelhouwer, J. Chem. Soc., Chem. Commun., 1977, 198.
57. (a) X. Y. Chen, X. Y. Zhang and P. Chen, Angew. Chem., Int. Ed., 2003, 42, 3798; (b) X. Zhang, X. Chen and P. Chen, Organometallics, 2004, 23, 3437; (c) S. Narancic and P. Chen, Organometallics, 2004, 24, 10; (d) A. Salameh, C. Copéret, J.-M. Basset, V. P. W. Böhm and M. Röper, Adv. Synth. Catal., 2007, 349, 238.
58. (a) F. Kapteijn, H. L. G. Bredt, E. Homburg and J. C. Mol, Ind. Eng. Chem. Prod. Res. Dev., 1981, 20, 457; (b) Y. Chauvin and D. Commereuc, J. Chem. Soc., Chem. Commun., 1992, 462.
59. (a) W. A. Herrmann, W. Wagner, U. N. Flessner, U. Volkhardt and H. Komber, Angew. Chem., Int. Ed. Engl., 1991, 30, 1636; (b) F. E. Kühn, A. Scherbaum and W. A. Herrmann, J. Organomet. Chem., 2004, 689, 4149; (c) A. Salameh, A. Baudouin, D. Soulivong, V. Böhm, M. Röper, J.-M. Basset and C. Copéret, J. Catal., 2008, 253, 180.
60. W. A. Herrmann, J. G. Kuchler, J. K. Felixberger, E. Herdtweck and W. Wagner, Angew. Chem., Int. Ed. Engl., 1988, 27, 394.
61. (a) E. Le Roux, M. Taoufik, C. Copéret, A. de Mallmann, J. Thivolle-Cazat, J.-M. Basset, B. M. Maunders and G. J. Sunley, Angew. Chem., Int. Ed., 2005, 44, 6755; (b) M. Taoufik, E. Le Roux, J. Thivolle-Cazat, C. Copéret, J.-M. Basset, B. Maunders and G. J. Sunley, Top. Catal., 2006, 40, 65.
62. J. Joubert and P. Sautet, unpublished results.
63. C. Copéret, Pure Appl. Chem., 2009, 81, 585.

---