1. Introduction

Homogeneous catalysts of inorganic salts and organometallic complexes have been used for many bulk and fine chemicals syntheses until now. They are usually dissolved in reaction solutions making all catalytic sites accessible to substrates and show high catalytic activity and selectivity. Despite these advantages, homogeneous catalysts have a share of only ca. 20-30% in industrial processes because catalyst/product(s) separation (problem of product contamination) and reuse of (expensive) catalysts are very difficult. Therefore, the development of easily recoverable and recyclable heterogeneous catalysts has received a particular research interest and can solve the problems of the homogeneous systems. Furthermore, the design of truly efficient heterogeneous catalysts with the activities and selectivities comparable to or higher than those of the corresponding homogeneous analogues is one of the most important challenges in modern organic synthesis, especially for fine chemicals.

The strategy to design the efficient heterogeneous catalysts for various functional group transformations is a creation of the monomerically (at least mainly if not wholly) dispersed ruthenium hydroxide species on the appropriate supports (Figure 1). Of all elements of the Periodic Table, ruthenium has widest scope of oxidation states ranging from -2 to +8 and various coordination geometries in each electron configuration. These properties would lead to the exploitation of efficient and/or novel catalytic transformations. In addition, the monomerically dispersed ruthenium hydroxide species would possess both Lewis acid and Bronsted base sites on the same metal and the various functional group transformations likely be promoted by the "concerted activation" by the Lewis acid and Bronsted base sites.

In this thesis, the highly dispersed ruthenium hydroxide catalysts (Ru(OH)x/support) have been prepared with various supports such as Al2O3 and TiO2. They showed high catalytic performance for liquid-phase hydrogen-transfer reactions such as racemization of chiral secondary alcohols, reduction of carbonyl compounds and allylic alcohols, and aerobic alcohol oxidation. Moreover, they could act as efficient heterogeneous catalysts for the N-imine formation and reductive N-alkylation of primary alcohols with amines, as well as the oxygenation of primary amines to amides. All these catalyses were truly heterogeneous and the catalysts recovered after the reactions could be reused with keeping their high catalytic performance.

2. Preparation and Characterization of Supported Ruthenium Hydroxide Catalysts

The supported ruthenium hydroxide catalysts were prepared with four different supports, anatase TiO2 (denoted as TiO2(A), Ishihara Sangyo Kaisya Ltd., ST-01, 316 m2g-1), anatase TiO2 (TiO2(B), the Catalysis Society of Japan, JRC-TIO-1, 73 m2g-1), rutile TiO2 (TiO2(C), Showa Denko K. K., SUPER-TITANIA G-2, 3.2 m2g-1), and Al2O3 (Sumitomo Chemical Co., Ltd., KHS-24, 160 m2g-1). The catalysts were prepared by the reaction of TiO2 (or Al2O3) with RuCl3 in aqueous medium followed by the treatment with NaOH. A typical procedure is as follows: The powder TiO2 (or Al2O3) (2.0 g) calcined at 550 oC for 3 h was vigorously stirred with 60 mL of an aqueous solution of RuCl3 (8.3 mM, pH ca. 2.0) at room temperature (ca. 22 oC ). After 15 min, the pH of the solution was adjusted to 13.2 by addition of an aqueous solution of NaOH (1 M) and the resulting slurry was stirred for 24 h. The solid was then filtered off, washed with a large amount of water, and dried in vacuo to afford supported ruthenium hydroxide catalysts. The ruthenium contents were 2.1-2.2 wt%.

Figure 2 shows the radial distribution functions (RDFs) from the Fourier transform of the k3-weighed EXAFS for Ru(OH)x/support, Ru(OH)x, and anhydrous RuO2. The first Ru-O shell parameters for Ru(OH)x and Ru(OH)x/support were similar to those for anhydrous RuO2. On the other hand, the RDFs exhibited significant differences for the longer range Ru-Ru environments. The intensities of the Ru-Ru shell signal at r = 2.8 A for Ru(OH)x and Ru(OH)x/support were considerably damped in comparison with that for anhydrous RuO2 and the signals at r = 3.3 A due to the Ru-O-Ru shells were hardly observed for Ru(OH)x and Ru(OH)x/support. The CNs (coordination numbers) of nearest-neighbor Ru atoms suggest the presence of different sizes in the supported hydroxide catalysts. These supported catalysts likely consist of the mixture of the monomeric ruthenium hydroxide species and polylmeric one. Judging from the CNs of the Ru-Ru shell signals, the sizes of the Ru(OH)x decreased in the order of Ru(OH)x > Ru(OH)x/TiO2(C) > Ru(OH)x/Al2O3 > Ru(OH)x/TiO2(B) > Ru(OH)x/TiO2(A). In the ESR measurements, the signal intensity due to Ru3+ species with low spin d5 electron configuration also increased with the decrease in the size of the supported ruthenium hydroxide species.

3. Size-dependent Catalytic Activity of Supported Ruthenium Hydroxides for Hydrogen Transfer Reactions and Aerobic Alcohol Oxidation

The catalytic activities of a series of the supported ruthenium hydroxide catalysts for the racemization of (R)-1-phenylethanol (>99% ee) were examined. No reaction proceeded in the absence of the catalyst, or in the presence of Al2O3, TiO2, or Ru(OH)x. Among the supported ruthenium hydroxide catalysts tested, the reaction rates decreased in the order of Ru(OH)x/TiO2(A) (TOF = 380 h-1) > Ru(OH)x/TiO2(B) (120 h-1) > Ru(OH)x/TiO2(C) (3 h-1), while the selectivities to the racemization were 88-95% and did not much change. The TOFsobs (observed turnover frequencies) monotonically increased with the decrease in CNs (sizes of ruthenium hydroxide species) as shown in Figure 3(a), suggesting that the present hydrogen transfer reactions depend on the sizes of supported Ru(OH)x species. In the presence of the most active Ru(OH)x/TiO2(A) catalyst, the scope of the present system towards various kinds of structurally diverse chiral secondary alcohols was examined (Table 1). Ru(OH)x/TiO2(A) showed high catalytic activities for the racemization of activated chiral secondary alcohols as well as non-activated ones. A major advantage of using Ru(OH)x/TiO2(A) is the heterogeneous nature, which was confirmed by the fact that the reaction did not proceed after removal of the catalyst at the reaction temperature and no leaching of the ruthenium species was observed. It was confirmed by XANES and EXAFS spectra that the local structure of the ruthenium species in the used Ru(OH)x/TiO2(A) catalyst was the same as that in the fresh catalyst. In addition, the Ru(OH)x/TiO2(A) catalyst could be reused for the racemization of (R)-1-phenylethanol at least three times with retention of its high catalytic performance (<1% ee, 86% select. for the 3rd reuse experiment). The Ru(OH)x/TiO2(A) catalyst could also act as an effective heterogeneous catalyst for the reduction of carbonyl compounds to alcohols [Eq. (1)] and allylic alcohols to saturated alcohols [Eq. (2)] in the presence of 2-propanol as a hydrogen source.

Next, the catalytic activities of a series of the supported ruthenium hydroxide catalysts for the aerobic oxidation of 1-phenylethnaol were examined. The TOFsobs were also dependent on the sizes of the ruthenium hydroxide species and increased with the decrease in CNs. However, they reached maximum with Ru(OH)x/TiO2(B), and then decreased (Figure 3(b)). A reaction mechanism involving the Ru alcoholate formation/β-elimination has been proposed for the present catalytic oxidations. Both of them are reversible reactions. The kinetic isotope effects (kH/kD = 5.3) indicate that the C-H bond breaking is included in the rate-determining step. Intrinsically, the catalytic activities of Ru(OH)x for both the hydride abstraction and hydride re-addition increased with the decrease in the CNs. The size effect on the hydride re-addition was lager than that on the hydride abstraction in the CNs from 0.76 to 0.37. As a result, the Ru(OH)x/TiO2(B) catalyst with the suitable size showed the highest catalytic activity for the aerobic alcohol oxidation.

Table 2 shows the results of the aerobic oxidation of various alcohols with the most active Ru(OH)x/TiO2(B) catalyst. All primary and secondary benzylic alcohols were converted into the corresponding carbonyl compounds, respectively, in almost quantitative yields. The a,B-unsaturated alcohol afforded the corresponding enal without the intramolecular hydrogen transfer. The present system could also oxidize secondary aliphatic and heteroatom-containing alcohols in high yields. The oxidation of radical clock substrate of cyclopropylphenyl carbinol exclusively produced cyclopropylphenyl ketone without the ring-opened products, suggesting that free-radical intermediates are not involved in the present alcohol oxidation. The present alcohol oxidation was also heterogeneously catalyzed and the Ru(OH)x/TiO2(B) catalyst could be reused for the oxidation of benzyl alcohol with retention of its high catalytic performance (>99% yield for the 3rd reuse experiment).

4. Catalytic Tandem One-pot Syntheses of Aldimines and Secondary Amines from Primary Alcohols and Amines

The tandem one-pot synthesis of aldimines from primary alcohols and amines could be efficiently promoted by supported ruthenium hydroxide catalysts, in particular Ru(OH)x/TiO2(B). The various combinations of substrates (seven alcohols and six amines) with the most active catalyst, Ru(OH)x/TiO2(B), have been investigated and all reactions efficiently proceeded to afford the corresponding aldimines in high yields. The catalyst (B) could be applied to the large-scale production and the TOF remarkably reached up to 182 h-1, showing the highest value among the heterogeneous systems to this day. Furthermore, the tandem syntheses of benzylidenealonitrile and 2-phenylbenzimidazole could be realized with good yields in the presence of supported ruthenium hydroxide catalysts.

The selective N-alkylation of aromatic and heteroaromatic amines with alcohols was described. A variety of aromatic and heteroaromatic amines in the presence of Ru(OH)x/Al2O3 were selectively converted into the corresponding secondary amines in high to excellent yields without any co-catalysts such as bases and ligands. The present N-alkylation likely proceeds via the following three sequential reactions; the oxidative dehydrogenation of an alcohol to a carbonyl compound occurred to form the corresponding imine followed by the hydrogen transfer reaction of the imine to afford the corresponding secondary amine.

5. Oxygenation of Primary Amines to Amides by Supported Ruthenium Hydroxide Catalysts

Amides are very important class of compounds in chemistry and biology that have been used as intermediates in peptide and protein synthesis, intensifiers of perfume, color pigments for inks, detergents, and lubricants. Traditionally, amides have been synthesized by the reaction of activated carboxylic acid derivatives with amines including ammonia or by the acid-catalyzed rearrangements of ketoximes. However, these traditional methods often produce vast amounts of toxic chemical wastes. Therefore, the development of efficient procedures for the synthesis of amides that avoids wasteful use of stoichiometric reagents and/or acidic and basic media is highly desirable and still a great challenge in modern organic synthesis. In general, it is very difficult to oxygenate the a-methylene groups of amines to the corresponding amides and a stoichiometric oxidant of RuO4 have been used with the protection of amino groups, for example. To date, efficient catalytic oxygenations of primary amines to amides including homogeneous as well as heterogeneous ones have never been reported. In the presence of supported ruthenium hydroxide catalysts, the oxygenation of primary amines to amides with molecular oxygen (air) in water could be realized. This is the first example of catalytic oxygenation of amines to amides with molecular oxygen.

In this transformation, the Ru(OH)x/Al2O3 catalyst showed the highest selectivity to the desired amides because the hydrolytic decomposition of amides (and/or nitrile intermediates) to carboxylic amides proceeded to some extent in the presence of supported ruthenium hydroxide catalysts on TiO2. In the presence of Ru(OH)x/Al2O3, various primary amines including aromatic, aliphatic, and heterocyclic could be converted into the corresponding amides in high yields (Table 3). After the reaction was completed, the Ru(OH)x/Al2O3 catalyst could be easily retrieved from the reaction mixture and reused without significant loss of its catalytic performance.

The Ru(OH)x/Al2O3-catalyzed transformation of benzylamine in 18O-labeled water (18O content:≧98%) led to the selective formation of 18O labeled benzamide in almost quantitative yield, suggesting that the oxygen atom incorporated into benzamide originates from water. The reaction profile for the oxygenation of benzylamine showed that benzonitrile was initially formed, followed by the formation of benzamide, indicating that the Ru(OH)x/Al2O3-catalyzed transformation possibly proceeds by a tandem process of the oxidative dehydrogenation of amines to nitriles followed by hydration to produce the corresponding amides.

6. Conclusion

The supported ruthenium hydroxide catalysts showed high catalytic activities and selectivities for hydrogen transfer reactions, aerobic alcohol oxidation, N-imine formation and reductive N-alkylation, and oxygenation. The present transformations were intrinsically heterogeneous and the supported ruthenium hydroxide catalysts could be reused with retention of their high catalytic performance.

Figure 1.The strategy to design efficient heterogeneous catalysts.

Figure 2. The RDFs from the Fourier transformation of the k3-weighted EXAFS for (a) Ru(OH)x/TiO2(A), (b) Ru(OH)x/TiO2(B), (c) Ru(OH)x/Al2O3, (d) Ru(OH)x/ TiO2(C), (e) Ru(OH)x, and (f) anhydrous RuO2. The phase shift was not corrected.

Figure 3. Relationship between TOFsobs and CNs of nearest-neighbor Ru atoms for (a) the racemization of (R)-1-phenylethanol and (b) the aerobic oxidation of 1-phenylethanol by various ruthenium hydroxide catalysts. (1): Ru(OH)x/TiO2(A), (2): Ru(OH)x/TiO2(B), (3): Ru(OH)x/ Al2O3, (4): Ru(OH)x/TiO2(C), and (5): Ru(OH)x.

Table 1. Ru(OH)x/TiO2(A)-catalyzed racemization of various chiral secondary alcohols.[a]

Table 2. Ru(OH)x/TiO2(B)-catalyzed aerobic oxidation of various alcohols.[a]

Table3. Ru(OH)x/Al2O3-catalyzed transformation of various primary amines to amides.[a]

本論文は「Study on Highly Functionalized Heterogeneous Catalysts Based on Supported Ruthenium Hydroxides for Green Functional Group Transformations (環境調和型官能基変換反応のための高機能担持水酸化ルテニウム触媒開発に関する研究)」と題し、全8章で構成されている。第1章は序論であり、固体触媒開発の重要性、金属水酸化物の特徴、本研究における触媒設計指針が述べられている。第2章では、種々の酸化物担体を用いて担持ルテニウム水酸化物触媒を調製し、XAFS、ESR、XPSなどの分光学的手法によりその構造を明らかにしている。また、様々なサイズを有する担持ルテニウム水酸化物触媒の作りわけにも成功している。第3章では、担持ルテニウム水酸化物触媒が、Meerwein-Ponndorf-Verley型還元反応、ラセミ化反応などの水素移行型還元反応に対する優れた不均一系触媒となることを見出している。触媒の構造を反応活性の関係を詳細に検討した結果、EXAFS、ESRから推定されるルテニウム水酸化物種のサイズが小さくなるにつれ、これらの水素移行型還元反応に対する活性が高くなることも明らかにしている。第4章では、担持ルテニウム水酸化物触媒が、分子状酸素を酸化剤としたアルコールのカルボニル化合物への選択酸化反応に対する優れた不均一系触媒となることを見出している。本酸化反応は構造敏感反応でありルテニウム水酸化物種の最適なサイズが存在すること、その理由も明らかにしている。第5および6章では、前章で得られた担持ルテニウム水酸化物触媒の特長を生かし、アルコールと1級アミンからの酸化的アルジミン合成およびアルコールと1級アミンからの還元的2級アミン合成などのワンポット反応系の開発に成功している。第7章では、担持ルテニウム水酸化物触媒を用いると、芳香族、脂肪族、ヘテロ原子を有する種々の1級アミンの酸素化反応が空気を酸化剤として水溶媒中で効率よく進行することを明らかにしている。本反応はこれまでに先例のない触媒反応である。さらに、反応速度論、18Oによるラベル化実験などから本反応は、アミンのニトリルへの酸化的脱水素反応、続くニトリルの水和反応により対応するアミドが生成することを明らかにしている。第3~7章で開発に成功した反応はすべて触媒固体表面上で進行する不均一系触媒反応であり、いずれにおいても反応後回収した触媒は、活性・選択性を低下させることなく再使用が可能であった。最後に、第8章として本論文の総括を行ってある。

以上のように、本論文では金属水酸化物の特長を生かした高機能な固体触媒および触媒反応系の開発に成功し、さらに反応機構、活性点構造に対する考察を行っており、固体触媒設計に対して重要な知見を与えるものである。

よって本論文は博士(工学)の学位請求論文として合格と認められる。