Next Article in Journal
Surface Growth and Diffusion Energetics of Ag Monolayers on Cu (001)
Next Article in Special Issue
Catalytic Reactions over Halide Cluster Complexes of Group 5–7 Metals
Previous Article in Journal
Metal Matrix Composites Reinforced by Nano-Particles—A Review
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Characterization of Catalytically Active Octahedral Metal Halide Cluster Complexes

1
Organometallic Chemistry Laboratory, RIKEN, 2-1 Hirosawa, Wako City, Saitama 351-0198, Japan
2
Advanced Catalysis Research Group, RIKEN Center for Sustainable Resource Science, 2-1 Hirosawa, Wako City, Saitama 351-0198, Japan
3
Graduate School of Science and Engineering, Saitama University, Shimo-Okubo, Sakura-Ku, Saitama City, Saitama 338-8570, Japan
*
Author to whom correspondence should be addressed.
Metals 2014, 4(2), 84-107; https://doi.org/10.3390/met4020084
Submission received: 9 January 2014 / Revised: 4 March 2014 / Accepted: 19 March 2014 / Published: 1 April 2014
(This article belongs to the Special Issue Metal and Molecular Clusters)

Abstract

:
Halide clusters have not been used as catalysts. Hexanuclear molecular halide clusters of niobium, tantalum, molybdenum, and tungsten possessing an octahedral metal framework are chosen as catalyst precursors. The prepared clusters have no metal–metal multiple bonds or coordinatively unsaturated sites and therefore required activation. In a hydrogen or helium stream, the clusters are treated at increasingly higher temperatures. Above 150–250 °C, catalytically active sites develop, and the cluster framework is retained up to 350–450 °C. One of the active sites is a Brønsted acid resulting from a hydroxo ligand that is produced by the elimination of hydrogen halide from the halogen and aqua ligands. The other active site is a coordinatively unsaturated metal, which can be isoelectronic with the platinum group metals by taking two or more electrons from the halogen ligands. In the case of the rhenium chloride cluster Re3Cl9, the cluster framework is stable at least up to 300 °C under inert atmosphere; however, it is reduced to metallic rhenium at 250–300 °C under hydrogen. The activated clusters are characterized by X-ray diffraction analyses, Raman spectrometry, extended X-ray absorption fine structure analysis, thermogravimetry–differential thermal analysis, infrared spectrometry, acid titration with Hammett indicators, and elemental analyses.

Graphical Abstract

1. Introduction

One of the greatest challenges in catalysis is the development of new reactions over new catalysts. Our research on the application of halide cluster complexes to catalysis is reviewed.
Carbonyl clusters have been strongly associated with catalysis, as most of these clusters contain platinum group metals. These metals have versatile catalytic activities in both homogeneous and heterogeneous systems. However, carbonyl clusters are thermally unstable, and almost all decompose above 200 °C due to their inherently weak metal–metal bonds. This dissociation enthalpy for a metal–metal bond in a cluster is two-thirds of that of a metal–CO bond (terminal) [1]. Exceptionally, at >260 °C, [PBu4]2[Os20(CO)40], which has a tetrahedral cubic close-packed (ccp) metal core of twenty Os atoms, was synthesized by the vacuum pyrolysis of [Os3(CO)10(NCMe)2] [2]. Hence, carbonyl cluster catalysis with retention of the cluster framework is difficult, and reactions catalyzed by coordinatively unsaturated carbonyl clusters with intact metal frameworks that have been isolated are rare [3]. Carbonyl clusters have been widely used as catalyst precursors for supported heterogeneous catalysts [4,5,6] or as models of metal catalysts [7,8]. In the latter cases, the carbonyl clusters afford uncoordinated metal species during the course of the reaction with or without decomposition of the metal frameworks, and the isolation and characterization of the adducts of carbonyl clusters with reactants are fairly easy.
Another type of cluster complex is known. The first report of the synthesis of a halide cluster, MoCl2, was published in 1859 [9] (Figure 1). Early on, it was considered to be a trimer; however, the definitive hexameric structure was established by X-ray analysis in 1967 as Mo6Cl12 or, more exactly, [Mo6Cli8]Cla2Cla-a4/2 [10], where “i” and “a” refer to the inner and outer (auβer) coordination spheres, respectively, and “a-a” refers to the corresponding linkage between two clusters. Extensive efforts have been directed at the synthesis of new halide clusters or the combination of known halide clusters with new networks. Presently, more than a hundred types of halide clusters have been synthesized by combining nineteen kinds of Group 3–7 metal atoms and four kinds of π-donor halogen ligands [11,12,13]. Halide clusters are generally synthesized by the comproportionation of a metal with a corresponding mononuclear metal halide at elevated temperatures (typically 600–1300 °C) for several days to weeks in a sealed tube as solid-state clusters. Then, the molecular clusters are synthesized by scission heat-treating with ligands or π-donor solvents (Figure 1). As the metal–metal bonds of the Groups 5 and 6 (4d and 5d) metals are stronger than their platinum group counterparts, an uncoordinated metal center can be revealed if some of the ligands are removed, leaving the thermally stable metal framework intact. Most halide clusters have octahedral metal array structures, consisting of the pure metal crystals with terminally coordinating, bridging, or triply bridging halogen π-donor ligands (Figure 2). These halide clusters have several characteristic features. The clusters are characterized by the middle oxidation states (1+ to 3+) of the metal atoms, which are intermediate between those of the bulk metal and the metal oxide. The catalytic activity, as well as the reactivity, of the metal atoms in the middle oxidation states, have not been reported. Moreover, the concerted effects of the metal–metal bonds of the cluster are also unknown. In multi-center and multi-electron systems, polynuclear coordination could induce reactivity in a ligand that differs significantly from that produced by mononuclear coordination [14]. Thermal stability is the largest advantage of the halide clusters over carbonyl clusters, and halide clusters have low vapor pressures and high melting points as halide complexes. Halide clusters have solid (network) and molecular states, which can be used in heterogeneous and homogeneous systems. A diversity of metal atoms and halogen ligands is also an attribute of halide clusters.
Figure 1. Structure of solid-state molybdenum chloride cluster Mo6Cl12 ([Mo6Cli8]Cla2Claa4/2) and molecular clusters (H3O)2[(Mo6Cl8)Cl6]·6H2O, [(Mo6Cl8)Cl4(H2O)2]·6H2O, and [(Mo6Cl8)Cl4(H2O)2].
Figure 1. Structure of solid-state molybdenum chloride cluster Mo6Cl12 ([Mo6Cli8]Cla2Claa4/2) and molecular clusters (H3O)2[(Mo6Cl8)Cl6]·6H2O, [(Mo6Cl8)Cl4(H2O)2]·6H2O, and [(Mo6Cl8)Cl4(H2O)2].
Metals 04 00084 g001
Figure 2. Two basic structures of halide clusters. Halogen π-donor ligands coordinate in two modes as inner ligands: 12 halogen ligands in a μ2-fashion in [M6X12]2+, and eight halogen ligands in a μ3-fashion in [M6X8]4+.
Figure 2. Two basic structures of halide clusters. Halogen π-donor ligands coordinate in two modes as inner ligands: 12 halogen ligands in a μ2-fashion in [M6X12]2+, and eight halogen ligands in a μ3-fashion in [M6X8]4+.
Metals 04 00084 g002
Studies on the reactivity of halide clusters are quite limited, and have generally been confined to observations of ligand exchange reactions with organic or inorganic reagents, an exception being the report of a simple reaction with hydrogen [15,16]. The applications of these clusters have been explored based on properties such as high temperature superconductivity, metallic conductivity at elevated temperatures, low-dimensional electronic or magnetic properties, and good thermal stability. Halide clusters are promising candidates for catalysts [17]; however, only MoCl2 has been used as a catalyst precursor, although the integrity of the cluster was destroyed during use [18].
The synthesized clusters have no metal–metal multiple bonds or coordinatively unsaturated sites, and accordingly, they are not active catalyst as they stand. Pretreatment for activation is necessary. In the case of a liquid batch-reaction system, the elimination of a halogen ligand by basic reagents or silver(I) ions results in the development of catalytic activity. However, this method is not applicable in a gas–solid reaction system, in which there are no restrictions on the reaction temperature. Therefore, activation in the gas phase has been attempted. The ligands of the coordination complexes are in equilibrium with free molecules or ions, although the dissociation constant is quite low at room temperature. In the case of the molecular molybdenum chloride cluster [(Mo6Cl8)Cl4(H2O)2], the ligands can be removed as neutral molecules at elevated temperatures in the gas phase. Plausible reactions for the activation are given by equations in Figure 3. As shown in Figure 3, a halide cluster was placed in a gas stream to avoid the recoordination of a liberated ligand, and heated gradually to dissociate or change the ligands until catalytic activity developed. Retention of the cluster framework was expected.
Figure 3. Activation and plausible reactions of a halide cluster in a gas stream with increasing temperature.
Figure 3. Activation and plausible reactions of a halide cluster in a gas stream with increasing temperature.
Metals 04 00084 g003
To date, nineteen Groups 3–7 transition metal atoms have been reported to afford halide cluster complexes. The electronegativities of the Groups 3 and 4 transition metals are low and it is difficult to remove the coordinated oxygen atoms of ligands from the cluster complexes. This suggests that oxygen-containing reactants cannot be applied for catalysis. In the case of Groups 6 and 7 transition metals, the coordinated oxygen atoms of ligands can be removed by heating above 200 °C. The preparation of Groups 3 and 4 metal clusters are not easy, since the syntheses are performed in a niobium or tantalum sealed tube. In contrast, Groups 5–7 metal clusters can be synthesized easily, and the clusters are stable in air. Oxygen- or nitrogen-containing reactants can be used over Groups 5–7 metals because of their moderate affinities. Thus, Groups 5–7 transition metal cluster complexes with chlorine ligands were chosen for study as potential catalysts.
Halide clusters, the majority of which have a highly symmetrical and aesthetically pleasing [M6X8] or [M6X12] unit [19], are attractive catalyst candidates. Since the synthesis of halide clusters was first reported over 150 years ago [9], they have been dormant to catalysis, like a “Sleeping Beauty”. In recent years, we have reported a series of reactions catalyzed by these clusters. This review is concerned with the activation and characterization of these cluster catalysts.

2. Activation and Characterization of Cluster Complexes Containing Group 6 Metals

2.1. Retention of the Cluster Framework of the (H3O)2[(Mo6Cl8)Cl6]·6H2O Molybdenum Cluster [20]

The undiluted crystalline powder of the molybdenum halide cluster (H3O)2[(Mo6Cl8)Cl6]·6H2O in a glass reaction tube was heated in a helium stream for 1 h at elevated temperature. The X-ray diffraction (XRD) patterns of samples treated at various temperatures are illustrated in Figure 4, in addition to that of independently synthesized [Mo6Cli8]Cla2Cla−a4/2. The cluster chemistry of MoCl2 has been extensively investigated. The XRD pattern of (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at ambient temperature for 1 h exactly matches that of [(Mo6Cl8)Cl4(H2O)2]·6H2O. In this reaction, two Cl ligands in the starting complex are removed as HCl, yielding the aqua complex [(Mo6Cl8)Cl4(H2O)2]·6H2O (Equation (1)). This complex loses its water of crystallization by treatment at 100–150 °C affording (H3O)2[(Mo6Cl8)Cl6] (Equation (2)). The evolution of Cl2 was not detected at this stage. The XRD patterns of samples treated above 250 °C reveal conversion into an extended Mo–Cl–Mo bonded solid-state cluster [Mo6Cl8]Cl2Cl4/2 (Equation (3)). The half-band widths of the peaks at 12.6° and 15.6° grow narrower when the sample is heated from 250 to 400 °C, indicating an increasing crystallinity with increasing treatment temperature.
(H3O)2[(Mo6Cl8)Cl6]·6H2O → [(Mo6Cl8)Cl4(H2O)2]·6H2O + 2HCl (r.t.)
[(Mo6Cl8)Cl4(H2O)2]·6H2O → [(Mo6Cl8)Cl4(H2O)2] + 6H2O (100–150 °C)
[(Mo6Cl8)Cl4(H2O)2] → [Mo6Cl8]Cl2Cl4/2 + 2H2O (>250 °C)
The Raman spectra of the heat-treated samples are shown in Figure 5, in which some of the assigned peaks for the molecular cluster (H3O)2[(Mo6Cl8)Cl6]·6H2O are indicated. The Raman spectra of halide clusters are generally insensitive to ligand exchange. For example, the Raman spectrum of (C6H5CH2NMe3)2[(Mo6Cl8)Cl6] is quite similar to that of [(Mo6Cl8)Cl4(THF)2] [21]. This observation demonstrates that the change in the outer ligands from a Cl to an oxygen donor does not grossly affect the symmetry of the molecule. The spectra of (H3O)2[(Mo6Cl8)Cl6]·6H2O and the samples treated up to 150 °C are very similar; and four assigned peaks are retained. Above 200 °C, the Raman peak originally at 321 cm−1 ascribed to the triple-bridged Mo–Cli breathing vibration was replaced by two peaks, suggesting a change in the coordination mode, and a new peak appeared at 278 cm−1. Catalytic activity for various reactions emerged at 200 °C. The spectra of the samples treated at 200–600 °C are quite similar to that of authentic [Mo6Cli8]Cla2Cla-a4/2 prepared by the reported method [22]. Thus, the results from Raman spectrometry accord with the discussion of the XRD analyses after treatment from room temperature to 600 °C.
Figure 4. XRD patterns of (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at various temperatures in a helium stream for 1 h. The XRD pattern of authentic Mo6Cl12 is also shown. All the identified compounds have two strong diffraction peaks at 12.4°–12.6° and 15.3°–15.6°. Unique diffraction peaks for each compound are indicated: (a) for (H3O)2[(Mo6Cl8)Cl6]·6H2O; (b) for [(Mo6Cl8)Cl4(H2O)2]·6H2O; (c) for [(Mo6Cl8)Cl4(H2O)2]; and (d) for Mo6Cl12.
Figure 4. XRD patterns of (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at various temperatures in a helium stream for 1 h. The XRD pattern of authentic Mo6Cl12 is also shown. All the identified compounds have two strong diffraction peaks at 12.4°–12.6° and 15.3°–15.6°. Unique diffraction peaks for each compound are indicated: (a) for (H3O)2[(Mo6Cl8)Cl6]·6H2O; (b) for [(Mo6Cl8)Cl4(H2O)2]·6H2O; (c) for [(Mo6Cl8)Cl4(H2O)2]; and (d) for Mo6Cl12.
Metals 04 00084 g004
Figure 5. Raman spectra of (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at various temperatures in a helium stream for 1 h. The Raman spectrum of authentic Mo6Cl12 is also shown.
Figure 5. Raman spectra of (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at various temperatures in a helium stream for 1 h. The Raman spectrum of authentic Mo6Cl12 is also shown.
Metals 04 00084 g005
The extended X-ray absorption fine structure (EXAFS) spectra of the near Mo K-edge absorption region for the treated samples are shown in Figure 6a, and the values of the coordination number (C.N.) for Mo–Mo and the bond length determined by curve fitting are plotted in Figure 6b. Only small changes were observed in the EXAFS spectra for treatment up to 400 °C. The value of C.N. for Mo–Mo, which corresponds to the Mo6 octahedron, was unchanged up to 400 °C. The Mo–Mo distance of the cluster framework was also retained up to 450 °C. These findings accord with the change in (H3O)2[(Mo6Cl8)Cl6]·6H2O via [(Mo6Cl8)Cl4(H2O)2]·6H2O and [(Mo6Cli8)Cla4(H2O)2] to [Mo6Cli8]Cla2Cla-a4/2, all of which have the same C.N. and Mo–Mo distance, and, hence, prove that the 400 °C-treated sample retained the Mo6 cluster framework. Treatment above 450 °C caused a significant decrease in both the C.N. and Mo–Mo distance. The decreasing intensity of the XRD peaks above 500 °C, as shown in Figure 4, is attributable to the partial decay of [Mo6Cli8]Cla2Cla-a4/2. No appreciable changes in the EXAFS spectra were observed after subsequent 3 h reaction with diethylamine at 400 °C; the bond distance and the C.N. for Mo–Mo were 0.261 ± 0.002 nm and 4.2 ± 0.4, respectively, after the reaction. This C.N. of the Mo–Mo bond, corresponding to the octahedral metal framework, indicates the retention of the cluster framework throughout the reaction. Thus, these results clearly show the retention of the Mo cluster framework and face-capping Cl ligands during the treatment and the catalytic reaction at 400 °C or above under helium. Under hydrogen, retention of the cluster framework of Mo with face-capping Cl ligands in (H3O)2[(Mo6Cl8)Cl6]·6H2O has similarly been confirmed up to 400 °C [23].
Figure 6. (a) The k3-weighted EXAFS oscillations at the Mo K-edge for (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at various temperatures in a helium stream for 1 h, and (b) values of coordination number (C.N.,○) and bond length (R, ●) determined by curve fitting.
Figure 6. (a) The k3-weighted EXAFS oscillations at the Mo K-edge for (H3O)2[(Mo6Cl8)Cl6]·6H2O treated at various temperatures in a helium stream for 1 h, and (b) values of coordination number (C.N.,○) and bond length (R, ●) determined by curve fitting.
Metals 04 00084 g006

2.2. Active Site of Molybdenum Cluster

Figure 7 shows the thermogravimetry–differential thermal analysis (TG–DTA) profiles of (H3O)2[(Mo6Cl8)Cl6]·6H2O under hydrogen [24]. The DTA data showed an endothermic peak from ambient temperature to 60 °C. The total weight loss of 5% corresponds to the loss of hydrogen chloride from the anionic cluster (H3O)2[(Mo6Cl8)Cl6]·6H2O to yield the neutral cluster [(Mo6Cl8)Cl4(H2O)2]·6H2O (Equation (1)). The subsequent endothermic weight loss of 8% from 60 to 117 °C is attributable to the loss of the water of crystallization from [(Mo6Cl8)Cl4(H2O)2]·6H2O (Equation (2)). The next endothermic weight loss above 117 °C can be attributed to the loss of aqua ligands to yield the solid state molybdenum cluster [Mo6Cl8]Cl2Cl4/2 (Equation (3)). However, the evolution of hydrogen chloride was detected above 117 °C and a gradual weight loss was observed from ~190 °C to approximately 450 °C. Hence, by rapid heating, (H3O)2[(Mo6Cl8)Cl6]·6H2O could be decomposed by other routes, as exemplified by Equations (4) and (5).
[(Mo6Cl8)Cl4(H2O)2] → [(Mo6Cl8)Cl3(OH)(H2O)] + HCl (>190 °C)
[(Mo6Cl8)Cl3(OH)(H2O)] → [(Mo6Cl8)Cl2(OH)2] + HCl (>190 °C)
Figure 7. TG and DTA profiles of (H3O)2[(Mo6Cl8)Cl6]·6H2O.
Figure 7. TG and DTA profiles of (H3O)2[(Mo6Cl8)Cl6]·6H2O.
Metals 04 00084 g007
Closer examination of the XRD patterns illustrated in Figure 4 shows that the crystallinity of the heat-treated (H3O)2[(Mo6Cl8)Cl6]·6H2O was poor. This indicated that the coordinated water had not been completely removed by the pretreatment, and that the oxygen of the water molecules had been captured. Elemental analysis of the sample treated at 400 °C in a helium stream revealed that it consisted of Mo (57.6 wt%) and Cl (40.5 wt%), with a residue (1.9 wt%) originating from the coordinated water. These findings can be interpreted in terms of the formation of an imperfect crystal including the defective {[Mo6Cl8]Cla2Cla-a4/2} moiety. That is to say, the catalytically active cluster moiety that has a hydroxo ligand such as {[Mo6Cli8]Cla2Cla−a3/2(OH)} or {[Mo6Cli8]ClaCla-a4/2(OH)} was retained intact in the imperfect crystal lattice of the solid-state cluster, as illustrated in Figure 8. The hydroxo ligands would dissociate H+ as expressed by Equation (6).
{[Mo6Cli8]Cla2Cla-a3/2(OH)} → H+ + {[Mo6Cli8]Cla2Cla-a3/2(O)}
Figure 8. Active sites formed in the imperfect crystal lattice of the solid state cluster Mo6Cl12.
Figure 8. Active sites formed in the imperfect crystal lattice of the solid state cluster Mo6Cl12.
Metals 04 00084 g008
A hydroxo cluster of molybdenum, [(Mo6Cl8)Cl2(OH)2(H2O)2], has been reported to be a weak acid, affording protons in aqueous solution [25]. Consequently, the development of Brønsted acid sites in the poorly crystallized solid-state cluster retaining hydroxo ligands was expected.
A complete description of surface acidity requires the determination of acid strength and the number of acid centers. The titration method cannot be applied to the crushed crystals of (H3O)2[(Mo6Cl8)Cl6]·6H2O, as it changes to a deeply colored crystal upon heat treatment [24]. Therefore, the acidity of (H3O)2[(Mo6Cl8)Cl6]·6H2O supported on silica gel (380 m2/g) in 5.0 wt%, (H3O)2[(Mo6Cl8)Cl6]·6H2O/SiO2, was measured using the n-butylamine titration method with various Hammett indicators [26]. Figure 9 plots n-butylamine titers as a function of the Hammett acidity function (H0). The titer for each plot corresponds to the number of acid sites with acid strengths equal to or greater than the H0 value. No acidity was observed for (H3O)2[(Mo6Cl8)Cl6]·6H2O/SiO2 treated at 150 °C under hydrogen. Treatment at 300 °C resulted in acidity development. In contrast to the crystalline cluster, almost all of the supported cluster molecules on SiO2 were exposed and could not be converted to the solid-state cluster [Mo6Cli8]Cla2Cla-a4/2. The supported clusters activated at 300 °C exhibited an equimolar number of acid sites at H0 ≈ 1.3 with an additional equimolar number of acid sites at H0 ≈ 1.5–2.0. These two acid sites can be attributed to the first and second dissociations of protons from the hydroxo ligands of [(Mo6Cl8)Cl2(OH)2], as exemplified by Equation (6). These sharp distributions of the acid strength and the equivalence of the acid sites on the cluster molecules are characteristic of a molecular catalyst, as solid acids such as silica-alumina and zeolites have a wide distribution of acid strengths because of their inhomogeneous structures. The H0 values of H3BO3 and H3PO4 supported on silica gel at 0.1 mmol g1 are reported to be +1.3 to +1.5 and +1.5 to −3.0, respectively [27].
Figure 9. n-Butylamine titers versus acid strength for (H3O)2[(Mo6Cl8)Cl6]·6H2O/SiO2 treated at 150 °C (∆) and 300 °C (○) in a hydrogen stream. The titer for the same amount of the SiO2 support treated at 300 °C (■) is also shown.
Figure 9. n-Butylamine titers versus acid strength for (H3O)2[(Mo6Cl8)Cl6]·6H2O/SiO2 treated at 150 °C (∆) and 300 °C (○) in a hydrogen stream. The titer for the same amount of the SiO2 support treated at 300 °C (■) is also shown.
Metals 04 00084 g009
The solid-phase synthesis of the well-formed crystalline solid state cluster [Mo6Cl8]Cla2Cla-a4/2 is achieved by slowly raising the temperature (over 2 h) of the crystalline molecular cluster (H3O)2[(Mo6Cl8)Cl6]·6H2O to 200 °C, followed by heating (for 5+ h) at 200 °C under vacuum to constant weight, as illustrated in Figure 1 [22]. In this process, the coordinated water of the intermediate [(Mo6Cl8)Cl4(H2O)2] is removed and replaced with a bridging Cla-a ligand. Rapid heating to higher temperatures, as employed for activation, leads to the formation of a poorly crystallized solid-state cluster containing defective [Mo6Cl8]Cla2Cla-a4/2 moieties, in which molybdenum atoms without bridging Cla-a ligands can be formed. In other words, a coordinatively unsaturated molybdenum atom would be formed by such activation, and would function as the active site of the cluster catalyst. Titration with CO revealed that 0.167 eq. of CO was adsorbed on (H3O)2[(Mo6Cl8)Cl6]·6H2O/SiO2, when the sample had been treated at 300 °C.
The incorporation of dihydrogen into cluster complexes has been reported in the literature. Halide clusters can take up dihydrogen to yield ThI2H0.7, Nb6I11H1.3, or Mo6Cl12H0.7 at temperatures between 150 and 450 °C [15]. Dihydrogen directly converts double-metal-layered ZrX (X = Cl and Br) to the ZrXH phase [16].
Figure 10. Isoelectronic relationships between catalyst metals. In the area of solid metal catalysis, researchers inspect the periodic table longitudinally to find other catalytically active metals for the relevant reaction, as the number of d-electrons plays an important role in catalysis. Iron, ruthenium, and osmium are catalytically active metals for the synthesis of ammonia. In the areas of molecular and complex catalyses, researchers also examine the periodic table laterally, as ligands change the valence electron distribution of the metal atom and sometimes change its electron distribution, such that the metal becomes isoelectronic with that of the rightward metal atoms in the periodic table.
Figure 10. Isoelectronic relationships between catalyst metals. In the area of solid metal catalysis, researchers inspect the periodic table longitudinally to find other catalytically active metals for the relevant reaction, as the number of d-electrons plays an important role in catalysis. Iron, ruthenium, and osmium are catalytically active metals for the synthesis of ammonia. In the areas of molecular and complex catalyses, researchers also examine the periodic table laterally, as ligands change the valence electron distribution of the metal atom and sometimes change its electron distribution, such that the metal becomes isoelectronic with that of the rightward metal atoms in the periodic table.
Metals 04 00084 g010
Therefore, we deduced that a coordinatively unsaturated or, under hydrogen, a hydrido-coordinated Mo atom would be formed as a component of the Mo6 cluster framework, and would be the active site of the cluster catalyst. Isoelectronic metal atoms occasionally exhibit the same catalysis as is exemplified by the nitration of alkylbenzenes catalyzed by Hg(II), Tl(III), or Pb(IV) (Figure 10) [28]. A series of Y(η5-C5H5)2(alkyl), [Zr(η5-C5H5)2(alkyl)]+, and [Nb(η5-C5H5)(η4-C4H6)]+ moieties show similar polymerization activity [29]. It is well known that Ru metals in [RuCl2(PPh3)3] and [RuHCl(PPh3)3] are isoelectronic with Rh metal in [RhCl(PPh3)3], and that these complexes catalyze the hydrogenation of olefinic bonds [30]. The carbides and nitrides of niobium, molybdenum, and tungsten exhibit similar catalysis to iridium metal for hydrazine decomposition in the thruster engines of satellites [31]. This has been explained by the addition of the ligand atomʼs valence electrons to the metal atoms, which affords an electron distribution similar to that of the iridium metal. Consequently, the coordinatively unsaturated or hydrido-coordinated Mo metal atom in the activated molybdenum cluster could be isoelectronic with the platinum group metals by taking two or more electrons from the halogen ligands, and might catalyze hydrogenation, dehydrogenation, and hydrodehydration.

2.3. Tungsten Cluster (H3O)2[(W6Cl8)Cl6]·6H2O

The activation behavior of the tungsten halide cluster (H3O)2[(W6Cl8)Cl6]·6H2O, also based on a Group 6 transition metal, is quite similar to that of the isomorphous molybdenum cluster (H3O)2[(Mo6Cl8)Cl6]·6H2O under both hydrogen [32] and helium [33] streams. One of the differences is the order of the loss of ligands and the transformation temperatures: water of crystallization is lost by treatment at 50 °C to yield (H3O)2[(W6Cl8)Cl6], and then two apical chloro ligands are lost by treatment above 100 °C to afford [(W6Cl8)Cl4(H2O)2]. When the cluster is heated progressively, it changes to an extended W–Cl–W-bonded solid-state compound [W6Cli8]Cla2Cla-a4/2 at 250–300 °C. The other difference is that the activated defective solid-state cluster [W6Cli8]Cla2Cla-a4/2 is stable to temperatures as high as 400 °C under hydrogen [32] and 450 °C under helium [33].
The development of a Brønsted acid site on (H3O)2[(W6Cl8)Cl6]·6H2O/SiO2 was ascertained by measuring the IR spectra of adsorbed pyridine [34]. The IR spectrum of (H3O)2[(W6Cl8)Cl6]·6H2O/SiO2 treated at 450 °C in a hydrogen stream followed by the adsorption of pyridine is illustrated in Figure 11. A band at 1445 cm−1 attributable to pyridine hydrogen-bonded to surface silanol groups is common in each spectrum. The two bands at 1540 and 1489 cm−1, assignable to pyridinium ion in the spectrum of (H3O)2[(W6Cl8)Cl6]·6H2O/SiO2 both before and after the reaction, show the presence of a Brønsted acid site on (H3O)2[(W6Cl8)Cl6]·6H2O. However, there is no band that is characteristic of coordinatively bonded pyridine in the 1449–1460 cm−1 region, and hence, there is no Lewis acid site. Hydroxo or oxo species developed in imperfect crystals of [W6Cl8]Cl2Cl4/2 by treatment at 450 °C exhibited Brønsted acidity, as expressed in Equation (7).
[(W6Cl8)Cl2(OH)2] → H+ + [(W6Cl8)Cl2(OH)(O)]
In summary, when the molybdenum and tungsten clusters were treated with progressively increasing temperatures, they lost two apical chlorido ligands and water of crystallization to yield [(Mo6Cl8)Cl4(H2O)2] and [(W6Cl8)Cl4(H2O)2] up to 200 °C. Then they were transformed to poorly crystallized solid-state clusters containing active sites which comprised Brønsted acid sites and coordinatively unsaturated sites that are stable to 400–450 °C with retention of the cluster framework.
Figure 11. IR spectra of pyridine adsorbed on (H3O)2[(W6Cl8)Cl6]·6H2O/SiO2. Sample powder (15 mg) was pressed into a disk (10 mm diameter) and treated in a hydrogen stream at 450 °C for 1 h (A), followed by introduction of pyridine (B). Catalyst sample after 4 h of reaction of toluene with methanol (C) and unsupported silica gel (D) were treated in the same way.
Figure 11. IR spectra of pyridine adsorbed on (H3O)2[(W6Cl8)Cl6]·6H2O/SiO2. Sample powder (15 mg) was pressed into a disk (10 mm diameter) and treated in a hydrogen stream at 450 °C for 1 h (A), followed by introduction of pyridine (B). Catalyst sample after 4 h of reaction of toluene with methanol (C) and unsupported silica gel (D) were treated in the same way.
Metals 04 00084 g011

3. Activation and Characterization of Cluster Complexes Containing Group 5 Metals

3.1. Retention of the Cluster Framework of the Niobium Cluster [(Nb6Cl12)Cl2(H2O)4]·4H2O [20]

The undiluted crystal powder of the niobium halide cluster [(Nb6Cl12)Cl2(H2O)4]·4H2O in a glass reaction tube was treated in a helium stream for 1 h at an elevated temperature. The XRD patterns of the treated samples at various temperatures are depicted in Figure 12. The changes in the pattern reveal that treatment above 100 °C transformed the cluster into an amorphous compound due to the loss of water of crystallization (Equation (8)).
[(Nb6Cl12)Cl2(H2O)4]·4H2O → [(Nb6Cl12)Cl2(H2O)4] + 4H2O (>100 °C)
The evolution of HCl gas was detected when the cluster was treated above 150 °C, and a violent evolution began when treated above 200 °C. However, Cl2 gas was not detected up to 500 °C. The amorphous compound was not the extended solid-state cluster Nb6Cl14 ([Nb6Cli10Cli-a2/2] Cla-i2/2Cla-a4/2) that was prepared by the direct reaction of Nb metal with NbCl5[35].
The Raman spectra of the heat-treated samples under helium are shown in Figure 13. The spectra markedly changed at 250 °C with increasing temperature. The band at 239 cm−1 is assigned to the breathing motion of the Nb6 octahedron (A1g), and those at 161 and 152 cm−1 are attributed to edge bridging Nb–Cli breathing vibrations (Eg and T2g, respectively) [36]. Crystal structural data revealed that the differences in M–M distances parallel the observed variations in the Raman shift [37]. The variations in the observed shift of the M6 A1g band are almost entirely accounted for by differences in the M–M force constant that accompany changes in the M–M bond distances [37]. The band of the Nb6 A1g vibration mode begins to shift to higher energy upon treatment and has a maximum shift at 250 °C. The loss of some ligands can increase the Nb–Nb bond order of the Nb6 octahedron, which may be ascribed to the higher energy shift [37]. Nevertheless, the Nb6 metal framework remains intact. Above 250 °C the two Raman bands originally at 161 and 152 cm−1 due to edge-bridging Nb–Cli breathing vibrations are replaced with a new band, which suggests that the coordination mode of the ligands changes above this temperature. Loss of the Cl atom as HCl was detected and the origin of the H atom in HCl can be an aqua ligand. No appreciable changes in the Raman spectra were observed after 3 h catalytic reaction with phenyl acetate at 150 °C. The Raman spectra of the samples heat-treated under hydrogen were the same as those treated under helium [32].
Figure 12. XRD patterns of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a helium stream for 1 h.
Figure 12. XRD patterns of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a helium stream for 1 h.
Metals 04 00084 g012
Figure 13. Raman spectra of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a helium stream for 1 h.
Figure 13. Raman spectra of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a helium stream for 1 h.
Metals 04 00084 g013
The EXAFS spectra of the near Nb K-edge absorption region for the helium-treated samples, and the Nb−Nb C.N. values and bond lengths determined by curve fitting are plotted in Figure 14. The Nb−Nb bond length of 0.291 nm, which corresponds to the Nb6 octahedron, was maintained up to 400 °C and the C.N. values for Nb−Nb were retained up to 250 °C with increasing treatment temperature. Accordingly, the octahedral cluster framework was completely retained to 250 °C, then began to disintegrate with increasing temperature, and disintegrated above 400 °C.
Figure 14. The k3-weighted EXAFS oscillations at the Nb K-edge for [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a helium stream for 1 h, and (b) values of coordination number (C.N., ○) and bond length (R, ●) determined by curve fitting.
Figure 14. The k3-weighted EXAFS oscillations at the Nb K-edge for [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a helium stream for 1 h, and (b) values of coordination number (C.N., ○) and bond length (R, ●) determined by curve fitting.
Metals 04 00084 g014
The elemental analysis data for the Nb and Cl atoms in the treated samples revealed that the Cl content of [(Nb6Cl12)Cl2(H2O)4]·4H2O, and consequently the relative amount of Cl to Nb, definitely decreased above 200 °C with increasing treatment temperature. The origin of the hydrogen in the liberated hydrogen chloride was the aqua ligand. Assuming that the remainder was composed of hydroxyl groups, the Nb6Cl11.2(OH)4.8 species, on average, was formed by treatment at 250 °C. Since the outer halogen ligands are generally labile, and, hence, are replaced prior to the inner ligands, the loss of the inner Cl ligands commenced at this stage, which is consistent with the results obtained by Raman and IR (Section 3.1) analyses. Above this temperature, the substantial loss of the Cl ligand began. Thus, when [(Nb6Cl12)Cl2(H2O)4]·4H2O was treated with progressively increasing temperature, its crystal structure decayed above 100 °C by the loss of water of crystallization (Equation (8)). Loss of the Cl ligand began above 150 °C with retention of the Nb6 cluster framework. Above 400 °C, the Nb6 octahedron practically decayed.

3.2. Active Site of Niobium Cluster

The TG curve for [(Nb6Cl12)Cl2(H2O)4]·4H2O illustrated in Figure 15 shows a clear two-step weight loss [33]. The ~7% weight loss and associated endothermic effect between ambient temperature and 150 °C can be ascribed to the loss of the crystallization water (Equation (8)). A large weight loss between 150 and 250 °C was observed in the TG analysis. The DTA curve indicates a large exothermic effect. Evolution of HCl was observed when [(Nb6Cl12)Cl2(H2O)4]·4H2O was treated above 150 °C. The bond energy of O–H (427.6 kJ mol−1) is comparable to that of Cl–H (431.6 kJ mol−1). The bond energy of Nb–O (771.5 kJ mol−1) is fairly large, though no data has been reported for that of Nb–Cl. The electronegativity of niobium is greater than that of chlorine. The exothermic effect is attributable to the oxidation of the niobium atom. Accordingly, the 22.5% weight loss between 150 and 250 °C corresponds to the loss of hydrochloric acid from the aqua and chloro ligands to form the hydroxo clusters (Equations (9) and (10)).
[(Nb6Cl12)Cl2(H2O)4] → [(Nb6Cl12)Cl(OH)(H2O)3] + HCl (150–250 °C)
[(Nb6Cl12)Cl(OH)(H2O)3] → [(Nb6Cl12)(OH)2(H2O)2] + HCl (150–250 °C)
Figure 15. TG and DTA profiles of [(Nb6Cl12)Cl2(H2O)4]·4H2O.
Figure 15. TG and DTA profiles of [(Nb6Cl12)Cl2(H2O)4]·4H2O.
Metals 04 00084 g015
The diffuse reflectance FT-IR spectra of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures are presented in Figure 16 [38]. All the samples exhibited O–H stretching vibrations in the region of 3600 cm−1, suggesting the retention of the hydroxy group even after the reaction of aniline with acetone at 300 °C for 3 h. In the region of 397 cm−1, the Nb6–OH stretching vibration of a hydroxo ligand appears upon treatment, whereas the Nb6–OH2 stretching vibration of an aqua ligand exhibits no peaks. As exemplified by Equations (9) and (10), a hydroxo ligand appears upon treatment above 250 °C and is retained after the reaction of aniline with acetone at 300 °C for 3 h, indicating the conservation of the hydroxo cluster.
Figure 16. IR spectra of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a hydrogen stream for 1 h. IR spectrum of [(Nb6Cl12)Cl2(H2O)4]·4H2O subjected to 3 h reaction of aniline with acetone at 300 °C is also shown.
Figure 16. IR spectra of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at various temperatures in a hydrogen stream for 1 h. IR spectrum of [(Nb6Cl12)Cl2(H2O)4]·4H2O subjected to 3 h reaction of aniline with acetone at 300 °C is also shown.
Metals 04 00084 g016
Then, the appearance of the acid site was detected. The IR spectrum of crystalline [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at 250 °C in a helium stream followed by the adsorption of pyridine is illustrated in Figure 17 [20]. The band at 1441 cm−1 was assigned to a hydrogen bonded pyridine, as untreated [(Nb6Cl12)Cl2(H2O)4]·4H2O has water of crystallization and aqua ligands. Three bands at 1634, 1536, and 1486 cm−1 assignable to pyridinium ion in the spectrum of [(Nb6Cl12)Cl2(H2O)4]·4H2O treated at 250 °C show the presence of a Brønsted acid site. Furthermore, there is no band that is characteristic of coordinatively bonded pyridine in the 1447–1460 cm−1 region, and, hence, there is no Lewis acid site.
Figure 17. IR spectra of pyridine adsorbed on [(Nb6Cl12)Cl2(H2O)4]·4H2O. Crystals of [(Nb6Cl12)Cl2(H2O)4]·4H2O (A) were treated in a helium stream at 250 °C for 1 h (B), followed by introduction of pyridine (C). Pyridine was introduced to the crystals of [(Nb6Cl12)Cl2(H2O)4]·4H2O without thermal activation (D).
Figure 17. IR spectra of pyridine adsorbed on [(Nb6Cl12)Cl2(H2O)4]·4H2O. Crystals of [(Nb6Cl12)Cl2(H2O)4]·4H2O (A) were treated in a helium stream at 250 °C for 1 h (B), followed by introduction of pyridine (C). Pyridine was introduced to the crystals of [(Nb6Cl12)Cl2(H2O)4]·4H2O without thermal activation (D).
Metals 04 00084 g017
The acidity of [(Nb6Cl12)Cl2(H2O)4]·4H2O/SiO2 was measured using the n-butylamine titration method with various Hammett indicators. The cluster activated at 150 °C exhibited an equimolar number of acid sites at 1.1 < H0 ≤ 1.5. The acid sites can be attributed to the dissociation of protons from the hydroxo ligands of [(Nb6Cl12)(OH)2(H2O)2], as exemplified by Equation (11). No acidity in [(Nb6Cl12)Cl2(H2O)4]·4H2O/SiO2 was observed without thermal treatment. No acid sites with an H0 stronger than +4.0 were detected on the silica gel support.
[(Nb6Cl12)(OH)2(H2O)2] → H+ + [(Nb6Cl12)(OH)(O)(H2O)2]
The activation behavior of the niobium cluster under a hydrogen stream is almost the same as that under helium streams [32]. Hydrogen served to physically rather than chemically remove the produced H2O and HCl. The activation behavior of a tantalum halide cluster [(Ta6Cl12)Cl2(H2O)4]·4H2O, also a Group 5 transition metal, is quite similar to that of the isomorphous niobium cluster [(Nb6Cl12)Cl2(H2O)4]·4H2O under both helium and hydrogen streams [39]. Formation of Brønsted acid sites was confirmed, and no basic sites were detected by CO2 adsorption on the activated tantalum cluster. The tantalum cluster was rather thermally stable.
In summary, the active sites of niobium and tantalum clusters are Brønsted acid sites that are stable to 350–400 °C with retention of the cluster framework. The formation of coordinatively unsaturated sites seems more difficult than for molybdenum and tungsten clusters, since removal of the coordinated oxygen atoms of the aqua ligands is difficult even under a hydrogen stream because of the large bond dissociation enthalpies of Nb–O (771.5 kJ mol−1) and Ta–O (799.1 kJ mol−1).

3.3 Rhenium Cluster Re3Cl9 ([Re3Cli3]Cla3Cla-a6/2)

The XRD patterns of the powdered crystals of the rhenium halide cluster Re3Cl9 treated at various temperatures in helium streams for 1 h are shown in Figure 18, and the Raman spectra of these samples are illustrated in Figure 19 [40]. The crystal structure and unit cell dimensions of Re3Cl9 were retained up to 400 °C, while the crystallinity decreased above 350 °C. The Raman spectrum noticeably changed on treatment at 300 °C. At this temperature, the peaks attributable to the Re–Cl and Re–Re vibration modes nearly disappeared. The samples treated at 350 °C and 400 °C were notable because they mostly maintained the original crystal structure but the constituents were no longer the same as those of Re3Cl9. This apparent disagreement can be interpreted by the partial decomposition of the component cluster [Re3Cl9] unit, which remains in the same position of the crystal lattice. The generation of Cl2 gas was not detected throughout the treatment. However, the slight evolution of HCl gas was observed at the vent when Re3Cl9 was treated in a stream of helium at 150 °C, and strong evolution began when treated above 200 °C.
Figure 18. XRD patterns of Re3Cl9 treated at various temperatures under a helium stream for 1 h. Top pattern is that of Re3Cl9 treated in a hydrogen stream for 1 h at 300 °C showing reduction of Re3Cl9 to metallic rhenium.
Figure 18. XRD patterns of Re3Cl9 treated at various temperatures under a helium stream for 1 h. Top pattern is that of Re3Cl9 treated in a hydrogen stream for 1 h at 300 °C showing reduction of Re3Cl9 to metallic rhenium.
Metals 04 00084 g018
Figure 19. Raman spectra of Re3Cl9 treated at various temperatures under helium for 1 h.
Figure 19. Raman spectra of Re3Cl9 treated at various temperatures under helium for 1 h.
Metals 04 00084 g019
It is difficult to dissolve the freshly prepared solid-state cluster Re3Cl9 ([Re3Cli3]Cla3Cla-a6/2), which consists of a [Re3Cli3]Cla3 moiety linked by weak halide bridges Cla-a6/2 (Figure 20). However, after exposure to a solvent followed by removal of the solvent, the materials becomes much more soluble in many solvents. This is due to the initial disruption of the tight Re3Cl9 structure, which remains disrupted after removal of the first solvent [41]. The Re3Cl9 cluster can be thermally stable for a short time at up to 450 °C, at which temperature it can be purified by vacuum sublimation, presumably by breaking the Cla-a6/2 cluster bridges in the solid state. Although Re3Cl9 decomposes in aqueous solution, it reacts with a number of Lewis bases with retention of the triangular metal framework to form complexes of the type L3Re3Cl9, where L = triphenylphosphine, pyridine, etc. [42]. Trace amounts of water contained in the reaction system could be incorporated in this solid cluster as coordinated water. Hence, Re3Cl9 can be activated by treatment under a stream of helium to form Re–OH or Re–O species, as expressed in Equations (12)−(14).
Re3Cl9 + H2O → Re3Cl9(H2O)
Re3Cl9(H2O) → Re3Cl8(OH) + HCl
Re3Cl8(OH) → Re3Cl7O + HCl
Thus, the catalytically active site of the rhenium cluster would be attributed to the acid site of the metal–OH moiety (hydroxo cluster) or metal–O moiety (oxo cluster) with retention of the metal cluster framework, which is stable at least up to 300 °C under inert atmosphere. Additionally, a coordinatively unsaturated site can be developed by removal of the chloro ligands or by breaking the Cla-a6/2 cluster bridges in the solid state. Under hydrogen, in contrast, the cluster is reported to be reduced to metallic rhenium at 250–300 °C, which has been confirmed by XRD analysis (Figure 18).
Figure 20. Structure of Re3Cl9 ([Re3Cli3]Cla3Claa6/2).
Figure 20. Structure of Re3Cl9 ([Re3Cli3]Cla3Claa6/2).
Metals 04 00084 g020

4. Conclusions

When molecular halide clusters of niobium, tantalum, molybdenum, and tungsten ([(M6X12)X2(H2O)4]·4H2O (M = Nb, Ta; X = Cl, Br), (H3O)2[(M6X8)X6]·6H2O (M = Mo, W; X = Cl, Br)) possessing octahedral metal frameworks are treated in a hydrogen or helium stream, the metal framework is retained up to 350–450 °C. Above 150–250 °C, two kinds of catalytically active sites develop in the clusters. One is a weak Brønsted acid resulting from a hydroxo ligand that is formed by the elimination of hydrogen halide from the halogen and aqua ligands. The other is a coordinatively unsaturated metal atom formed by the removal of an aqua ligand. This coordinatively unsaturated metal atom in the cluster could be isoelectronic with the platinum group metals by taking two or more electrons from the halogen ligands. Thus, activated halide clusters of Groups 5 and 6 metals are a bi-functional catalysts. In the case of the rhenium chloride cluster Re3Cl9, the cluster framework is stable to at least 300 °C under inert atmosphere; however, it is reduced to metallic rhenium at 250–300 °C under hydrogen. Catalytic reactions that occur over halide clusters will be presented in the subsequent paper [43].

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Mingos, D.M.P.; Wales, D.J. Introduction to Cluster Chemistry; Prentice-Hall: Upper Saddle River, NJ, USA, 1990; p. 6. [Google Scholar]
  2. Gade, L.H.; Johnson, B.F.G.; Lewis, J.; McPartlin, M.; Powell, H.R.; Raithby, P.R.; Wong, W.T. Synthesis and structural characterisation of the osmium cluster dianions [Os17(CO)36]2− and [Os20(CO)40]2−. J. Chem. Soc. Dalton Trans. 1994, 1994, 521–532. [Google Scholar]
  3. Chihara, T.; Yamazaki, H. Hexaruthenium carbido carbonyl methyl cluster [PPN][Ru6C(CO)16(CH3)] as catalyst precursor for hydrogenation of olefins. Syntheses and structures of unsaturated and saturated hexaruthenium hydrido clusters [PPN][Ru6C(CO)15H] and [PPN][Ru6C(CO)16H]. J. Organomet. Chem. 1994, 473, 273–284. [Google Scholar] [CrossRef]
  4. Lamb, H.H. Engineering catalyst surfaces with metal carbonyl clusters. Catal. Today 1993, 18, 3–19. [Google Scholar] [CrossRef]
  5. Braunstein, P.; Rosé, J. Heterometallic clusters for heterogeneous catalysis. In Catalysis by Di- and Polynuclear Metal Cluster Complexes; Adams, R.D., Cotton, F.A., Eds.; Wiley-VCH: New York, NY, USA, 1998; Chapter 13; p. 443. [Google Scholar]
  6. Gates, B.C. Metal cluster catalysts dispersed on solid supports. In Catalysis by Di- and Polynuclear Metal Cluster Complexes; Adams, R.D., Cotton, F.A., Eds.; Wiley-VCH: New York, NY, USA, 1998; Chapter 14; p. 509. [Google Scholar]
  7. Zuffa, J.L.; Gladfelter, W.L. On the mechanism of homogeneous catalytic hydrogenation using anion-promoted metal clusters. J. Am. Chem. Soc. 1986, 108, 4669–4671. [Google Scholar] [CrossRef]
  8. Adams, R.D.; Babin, J.E.; Tasi, M.; Wang, J.G. Catalyst design. The activation of a trinuclear metal cluster complex by metal atom substitution. Organometallics 1988, 7, 755–764. [Google Scholar] [CrossRef]
  9. Blomstrand, W. Ueber unorganische Haloidverbindungen, die sich wie Radicale verhalten. J. Prakt. Chem. 1859, 77, 88–119. [Google Scholar] [CrossRef]
  10. Schafer, H.; Schnering, H.G.; Tillack, J.; Kuhnen, F.; Wohrle, H.; Baumann, H. Neue Untersuchungen über die Chloride des Molybdäns. Z. Anorg. Allg. Chem. 1967, 353, 281–310. [Google Scholar] [CrossRef]
  11. Lee, S.C.; Holm, R.H. Nonmolecular metal chalcogenide/halide solids and their molecular cluster analogues. Angew. Chem. Int. Ed. Engl. 1990, 29, 840–856. [Google Scholar] [CrossRef]
  12. Cotton, F.A.; Hughbanks, T.; Runyan, C.E., Jr.; Wojtczak, W.A. Halide-suppored octahedral clusters of zirconium: Structural and bonding questions. In Early Transition Metal Clusters with π-Donor Ligands; Chisholm, M.H., Ed.; VCH Publishers: New York, NY, USA, 1995; Chapter 1; p. 1. [Google Scholar]
  13. Saito, T. Chalcogenide cluster complexes of the early transition metals. In Early Transition Metal Clusters with π-Donor Ligands; Chisholm, M.H., Ed.; VCH Publishers: New York, NY, USA, 1995; Chapter 3; p. 63. [Google Scholar]
  14. Muetterties, E.L. Metal clusters in catalysis VIII. Reduction of triple bonds. Bull. Soc. Chim. Belg. 1976, 85, 451–470. [Google Scholar] [CrossRef]
  15. Struss, A.W.; Corbett, J.D. Reaction of hydrogen with metallic and reduced halides. The requirement of delocalized electrons for reaction. Inorg. Chem. 1978, 17, 965–969. [Google Scholar] [CrossRef]
  16. Imoto, H.; Corbett, J.D.; Cisar, A. Synthesis by hydrogen-driven disproportionation reactions. Synthesis and structure of the hexazirconium dodecahalide clusters Zr6Cl12 and Zr6Br12 and the double salt Zr6Cl12.M2ZrCl6 (M = sodium, potassium, cesium). Inorg. Chem. 1981, 20, 145–151. [Google Scholar] [CrossRef]
  17. Miller, G.J. Chemistry and properties of novel niobium cluster compounds. J. Alloys Compd. 1995, 229, 93–106. [Google Scholar] [CrossRef]
  18. Boorman, P.M.; Chong, K.; Jasim, K.S.; Kydd, R.A.; Lewis, J.M. A study of supported molybdenum cluster complexes as catalyst precursors: Part II. Evaluation of the catalytic activity of clusters supported on γ-alumina and fluorided γ-alumina. J. Mol. Catal. 1989, 53, 371–380. [Google Scholar] [CrossRef]
  19. Prokopuk, N.; Shriver, D.F. The octahedral M6Y8 and M6Y12 clusters of group 4 and 5 transition metals. Adv. Inorg. Chem. 1999, 46, 1–49. [Google Scholar]
  20. Kamiguchi, S.; Mori, T.; Watanabe, M.; Suzuki, A.; Kodomari, M.; Nomura, M.; Iwasawa, Y.; Chihara, T. Retention of the octahedral metal framework of Nb and Mo halide clusters in catalytic decomposition of phenyl acetate to phenol and ketene. J. Mol. Catal. A 2006, 253, 176–186. [Google Scholar] [CrossRef]
  21. Guggenberger, L.J.; Sleight, A.W. Structural and bonding characterizations of molybdenum dibromide, Mo6Br12·2H2O. Inorg. Chem. 1969, 8, 2041–2049. [Google Scholar] [CrossRef]
  22. Nannelli, P.; Block, B.P. Molybdenum(II) halides. Inorg. Synth. 1970, 12, 170. [Google Scholar]
  23. Kamiguchi, S.; Kondo, K.; Kodomari, M.; Chihara, T. Catalytic ring-attachment isomerization and dealkylation of diethylbenzenes over halide clusters of group 5 and group 6 transition metals. J. Catal. 2004, 223, 54–63. [Google Scholar] [CrossRef]
  24. Nagashima, S.; Kamiguchi, S.; Ohguchi, S.; Chihara, T. Gas-phase alkylation of pyridine and phenol with alcohols over halide clusters of group 5–7 transition metals as solid acid catalysts. J. Clust. Sci. 2011, 22, 647–660. [Google Scholar] [CrossRef]
  25. Brosset, C. The structure of complex compounds of bivalent molybdenum. III. X-ray analysis of the structure of the chloro acid in alcohol solution. Arkiv. Kemi. 1949, 1, 353. [Google Scholar]
  26. Johnson, O. Acidity and polymerization activity of solid acid catalysts. J. Phys. Chem. 1955, 59, 827–831. [Google Scholar]
  27. Benesi, H.A. Acidity of catalyst surfaces. I. Acid strength from colors of asdorbed indicatios. J. Am. Chem. Soc. 1956, 78, 5490–5494. [Google Scholar] [CrossRef]
  28. Greenop, M.W.; Thomas, C.B. Nitration of alkylbenzenes catalyzed by mercury(II), thallium(III) and lead(IV). J. Chem. Soc. Perkin Trans. 1995, 2, 1595–1599. [Google Scholar] [CrossRef]
  29. Mashima, K. Catalytic diversity of diene complexes of niobium and tantalum on polymerizations of ethylene, norbornene, and methyl methacrylate. Macromol. Symp. 2000, 159, 69–76. [Google Scholar] [CrossRef]
  30. Bennett, M.A.; Matheson, T.W. Catalysis by ruthenium compounds. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F.G.A., Abel, E.W., Eds.; Plenum Press: Oxford, UK, 1982; Volume 4, p. 931. [Google Scholar]
  31. Chen, X.; Zhang, T.; Zheng, M.; Wu, Z.; Wu, W.; Li, C. The reaction route and active site of catalytic decomposition of hydrazine over molybdenum nitride catalyst. J. Catal. 2004, 224, 473–478. [Google Scholar] [CrossRef]
  32. Kamiguchi, S.; Noda, M.; Miyagishi, Y.; Nishida, S.; Kodomari, M.; Chihara, T. Catalytic isomerization of 1-hexene to 2-hexene by halide clusters of Nb, Mo, Ta and W possessing an octahedral metal core. J. Mol. Catal. A 2003, 195, 159–171. [Google Scholar] [CrossRef]
  33. Kamiguchi, S.; Nagashima, S.; Komori, K.; Kodomari, M.; Chihara, T. Thermal activation of molecular tungsten halide clusters with the retention of an octahedral metal framework and the catalytic dehydration of alcohols to olefins as a solid acid catalyst. J. Clust. Sci. 2007, 18, 414–430. [Google Scholar] [CrossRef]
  34. Kamiguchi, S.; Nishida, S.; Kurokawa, H.; Miura, H.; Chihara, T.T. Formation of Brønsted acid site on halide clusters of group 5 and 6 transition metals. Catalytic methylation and demethylation of methylbenzenes with methanol. J. Mol. Catal. A 2005, 226, 1–9. [Google Scholar]
  35. Simon, A.; Schnering, H.G.; Wöhrle, H.; Schäfer, H. Nb6Cl14–Synthese, Eigenschaften, Struktur. Z. Anorg. Allg. Chem. 1965, 339, 155. [Google Scholar] [CrossRef]
  36. Harder, K.; Preetz, W. Schwingungsspektren der Clusterverbindungen (M6Xi12)Xa2·8H2O, M = Nb, Ta; Xi = Cl, Br; Xa = Cl, Br, I. Z. Anorg. Allg. Chem. 1990, 591, 32–40. [Google Scholar] [CrossRef]
  37. Schoonover, J.R.; Zietlow, T.C.; Clark, D.L.; Heppert, J.A.; Chisholm, M.H.; Gray, H.B.; Sattelberger, A.P.; Woodruff, W.H. Resonance Raman spectra of [M6X8Y6]2− cluster complexes (M = Mo, W; X, Y = Cl, Br, I). Inorg. Chem. 1996, 35, 6606–6613. [Google Scholar] [CrossRef]
  38. Kamiguchi, S.; Takahashi, I.; Kurokawa, H.; Miura, H.; Chihara, T. Vapor-phase synthesis of 1,2-dihydro-2,2,4-trimethylquinolines from anilines and acetone over group 5–7 metal halide clusters as catalysts. Appl. Catal. A 2006, 309, 70–75. [Google Scholar]
  39. Kamiguchi, S.; Takaku, S.; Kodomari, M.; Chihara, T. Variable catalytic behavior of Nb, Mo, Ta, W, and Re halide clusters. Isomerization of alkynes to conjugated dienes under nitrogen and hydrogenation to alkenes under hydrogen. J. Mol. Catal. A 2006, 260, 43–48. [Google Scholar] [CrossRef]
  40. Kamiguchi, S.; Watanabe, M.; Kondo, K.; Kodomari, M.; Chihara, T. Catalytic dehydrohalogenation of alkyl halides by Nb, Mo, Ta, and W halide clusters with an octahedral metal framework and by a Re chloride cluster with a triangular metal framework. J. Mol. Catal. A 2003, 203, 153–163. [Google Scholar] [CrossRef]
  41. Cotton, F.A.; Wilkinson, G.; Murillo, C.A.; Bochemann, M. Advanced Inorganic Chemistry, 6th ed.; Wiley: New York, NY, USA, 1999; p. 981. [Google Scholar]
  42. Cotton, F.A.; Lippard, S.J.; Mague, J.T. Preparation and spectra of some adducts of the nonachlorotrirhenium and nonabromotrirhenium groups; evidence for the Re3Br9 group. Inorg. Chem. 1965, 4, 508–514. [Google Scholar] [CrossRef]
  43. Nagashima, S.; Kamiguchi, S.; Chihara, T. Halide cluster catalysis. Catalytic reactions. Metals submitted for publication. 2014. [Google Scholar]

Share and Cite

MDPI and ACS Style

Kamiguchi, S.; Nagashima, S.; Chihara, T. Characterization of Catalytically Active Octahedral Metal Halide Cluster Complexes. Metals 2014, 4, 84-107. https://doi.org/10.3390/met4020084

AMA Style

Kamiguchi S, Nagashima S, Chihara T. Characterization of Catalytically Active Octahedral Metal Halide Cluster Complexes. Metals. 2014; 4(2):84-107. https://doi.org/10.3390/met4020084

Chicago/Turabian Style

Kamiguchi, Satoshi, Sayoko Nagashima, and Teiji Chihara. 2014. "Characterization of Catalytically Active Octahedral Metal Halide Cluster Complexes" Metals 4, no. 2: 84-107. https://doi.org/10.3390/met4020084

Article Metrics

Back to TopTop