Effects of Solvents on the Hydrogenation of Mono-Aromatic

Department of Materials Physics and Chemistry, Graduate School of ... Al2O3 and Pt/Al2O3 catalysts at 120 °C under a hydrogen pressure of 6 MPa...
2 downloads 4 Views 134KB Size
Energy & Fuels 1999, 13, 1191-1196

1191

Effects of Solvents on the Hydrogenation of Mono-Aromatic Compounds Using Noble-Metal Catalysts Hideyuki Takagi, Takaaki Isoda, Katsuki Kusakabe, and Shigeharu Morooka* Department of Materials Physics and Chemistry, Graduate School of Engineering, Kyushu University, Fukuoka 812-8581, Japan Received April 7, 1999

Toluene, phenol, benzyl alcohol, and benzoic acid in alcohols were hydrogenated using Ru/ Al2O3 and Pt/Al2O3 catalysts at 120 °C under a hydrogen pressure of 6 MPa. The Ru/Al2O3 catalyst exhibited a high activity for hydrogenation of mono-aromatics dissolved in alcohols. The hydrogenation activity of the Pt/Al2O3 catalyst was strongly inhibited by the presence of alcohols. To evaluate the role of solvents, nonpolar and polar solvents, some of which contained oxygen, sulfur or nitrogen, were used for the hydrogenation of benzyl alcohol over the Ru/Al2O3 catalyst. A relationship was found between the hydrogenation reactivity of benzyl alcohol in polar solvents and the δ value of the solvent, as defined by the difference between donor number and acceptor number. Nonpolar solvents had no effect on the hydrogenation reaction of benzyl alcohol. Polar solvents with negative δ values, such as methanol, ethanol, and acetic acid, also had no effect. However, polar solvents with positive δ values, such as acetone, tetrahydrofuran, 1,4-dioxane, and diethyl ether, inhibited the hydrogenation of benzyl alcohol. Since benzene was hydrogenated in preference to benzyl alcohol, the hydrogenation of benzyl alcohol was suppressed when benzene was used as the solvent. No hydrogenation reactivity was found when dimethyl sulfoxide, N,Ndimethyl formamide, and N-methyl-2-pyrrolidone were used as the solvent, since the Ru catalysts were inactivated by the solvents containing sulfur or nitrogen. To enhance the hydrogenation reactivity of aromatic compounds, carboxylic acids were added to the reaction system. The hydrogenation reactivity of benzyl alcohol in the polar solvents was greatly increased by the addition of acetic, butyric, or lauric acid. However, the hydrogenation reactivity was completely retarded by the addition of formic acid with a high relative permittivity. This suggests that the effect of carboxylic acids on improving hydrogenation reactivity can be attributed to the decrease in the relative permittivity of the solvent. On the basis of these results, the roles of solvents and additives in the hydrogenation of mono-aromatic ring compounds can be explained by (1) the δ value of the solvent, (2) competitive hydrogenation between substrate and solvent, (3) the relative permittivity of solvent, and (4) deactivation of the catalyst by solvents that contain sulfur or nitrogen.

Hydrogenation of aromatic compounds is a key reaction in processes such as the hydrorefining of heavy oil and the production of petrochemicals. Transition metals, such as nickel and cobalt, as well as platinum-group metals, including ruthenium, rhodium, palladium, and platinum, can be used as the catalyst for this type of hydrogenation. Platinum-group metals have the advantage of high activity for the hydrogenation of aromatic compounds under mild conditions. It has been reported that alkylbenzenes are hydrogenated at 150 °C under a hydrogen pressure of 1 MPa.1 The catalyst activity for the hydrogenation of benzene is reported to be in the order of Rh > Ru . Pt . Pd . Ni > Co.2 The hydrogenation activity of the above catalysts is, however, sometimes decreased by the presence of sol-

vents. Orito et al.3 reported that the hydrogenation reactivity of benzene using a Raney nickel catalyst was not affected by hexane and heptane, but that the reaction was inhibited in the presence of methanol, ethanol, and dioxane. Thus solvent effects can be significant, especially for the case of noble-metal catalysts, which are used for the hydrogenation of asphaltene, coal extracts, and polymers under mild conditions. In a previous study,4 we reported that oxidized Yallourn coal, which had been treated with aqueous H2O2 in the presence of 1-propanol at 70 °C, was solubilized in ethanol at a yield of 80 wt %, based on the dry raw coal mass. The ethanol-solubilized coal was further hydrogenated over a Ru catalyst at 120 °C under a hydrogen pressure of 10 MPa for 72 h.5 As a result of this catalytic hydrogenation, the aromaticity of the coal structure was altered and the pyrolysis reactivity was increased.

* Author to whom correspondence should be addressed. Telephone: +81-92-642-3551. Fax: +81-92-651-5606. E-mail: smorotcf@ mbox.nc.kyushu-u.ac.jp. (1) Muroi, T. Catalyst (in Japanese) 1994, 36, 531. (2) Greenfield, H. Ann. N.Y. Acad. Sci. 1973, 214, 233.

(3) Orito, Y.; Imai. S. Catalyst (in Japanese) 1962, 4, 5. (4) Isoda, T.; Tomita, H.; Kusakabe, K.; Morooka, S.; Hayashi, J.-i. Proc. of Int. Conf. Coal Sci. 1997, 2, 581. (5) Isoda, T.; Takagi, H.; Kusakabe, K.; Morooka, S. Energy Fuels 1998, 12, 503.

Introduction

10.1021/ef990061m CCC: $18.00 © 1999 American Chemical Society Published on Web 10/08/1999

1192

Energy & Fuels, Vol. 13, No. 6, 1999

Takagi et al.

Table 1. Properties of Compounds Used compound methanola ethanola 1-propanola 2-propanola hexanea heptanea benzenea acetonea tetrahydrofurana 1,4-dioxanea diethyl ethera dimethyl sulfoxidea N,N-dimethyl formamidea N-methyl-2-pyrrolidonea formic acidb acetic acidb butyric acidb lauric acidb a

abbreviation MeOH EtOH 1-PrOH 2-PrOH

THF dioxane DEE DMSO DMF NMP HCOOH CH3COOH C3H7COOH C11H23COOH

donor number

acceptor number

δ

19.0 20.5

41.3 37.1

-22.3 -16.6

20.0 0.0

33.5 0.0

-13.5 0.0

0.1 17.0 20.0 14.8 19.2 29.8 26.6 27.3

8.2 12.5 8.0 10.8 3.9 19.3 16.0 13.3

-8.1 4.5 12.0 4.0 15.3 10.5 10.6 14.0

52.9

Solvent. b Additive.

However, the hydrogenation reactivity of aromatic compounds in solvents other than ethanol has not yet been investigated. In the present study, therefore, toluene, phenol, benzyl alcohol, and benzoic acid were subjected to hydrogenation using Ru and Pt catalysts at 120 °C under a hydrogenation pressure of 6 MPa. Lower alcohols, nonpolar solvents, and solvents containing oxygen, sulfur, and nitrogen were used in these reactions, and their effect on the efficiency of the reaction was examined. Carboxylic acids were also added to the reaction system, and the effect of these compounds on the hydrogenation reactivity of aromatic compounds was also determined. Experimental Section Chemicals and Catalysts. Toluene, phenol, benzyl alcohol, and benzoic acid were used as substrates. Table 1 shows the abbreviations and properties of the compounds used. Methanol, ethanol, 1-propanol, and 2-propanol were used for alcoholic solvents; hexane, heptane, and benzene were used for nonpolar solvents; acetone, 1,4-dioxane, tetrahydrofuran, and diethyl ether were used for polar solvents containing oxygen; and dimethyl sulfoxide, N,N-dimethyl formamide, and N-methyl-2pyrrolidone were used for polar solvents containing sulfur or nitrogen. Formic acid (HCOOH), acetic acid (CH3COOH), butyric acid (C3H7COOH), and lauric acid (C11H23COOH) were selected for use as added carboxylic acids in the hydrogenation reaction of benzyl alcohol. All chemicals were purchased from Wako Chemical Co. and were used without further purification. An aluminasupported ruthenium catalyst (Ru/Al2O3, Wako Chemical), and an alumina-supported platinum catalyst (Pt/ Al2O3, Wako Chemical) were used as the hydrogenation catalysts. The BET surface areas of these catalysts were calculated from N2 sorption isotherms (ASAP 2000, Shimadzu Micromeritics),6 and these data are shown in Table 2, along with other relevant properties. Reactions. The hydrogenation reactions were performed in a 50-mL batch autoclave equipped with a magnetic stirrer, rotating at 1000 rpm. Three grams of (6) Barret, E. P.; Joyner, L. G.; Halenda, P. P. J. Am. Chem. Soc. 1951, 73, 373.

Table 2. Properties of Catalysts Used catalyst

support

metal content [wt %]

surface area [m2/g]

Ru/Al2O3 Pt/Al2O3

γ-Al2O3 γ-Al2O3

5.0 5.0

110 97

substrate and 0.5 g of catalyst were used. The amount of carboxylic acid was 3 g unless otherwise specified. The amount of solvent was varied over the range of 0.75-6 g. The temperature was increased from room temperature to the final temperature, 120 °C, at a rate of 15 °C/min. The reaction was assumed to start when the temperature reached 115 °C, after which the reaction was continued for 0.5-2 h. The time required to reach 115 °C from room temperature was 6 min, and the conversion of a substrate in this period was less than 10 mol %. The initial H2 pressure was 6 MPa. The pressure increased from the initial value to 6.1-6.6 MPa during the heating step and approached a final pressure, 3-6 MPa, which was dependent on hydrogenation reactivity. After the hydrogenation, the catalyst was separated by centrifugation. Products were qualitatively and quantitatively analyzed using a GC-FID (Shimadzu, GC-14A,) and a GC-MS (Shimadzu, QP-5000), equipped with a capillary column. The hydrogenation conversion of substrates was calculated using the following equation:

conversion [mol %] ) (1 - N/100) × 100

(1)

where N [mol %] is the yield of unreacted substrate. The δ value of a solvent is defined by

δ ) DN - AN

(2)

where DN is the donor number, and AN is the acceptor number of the solvent.7-10 The relative permittivity for a mixture of solvents i and j, Pmix, was calculated from the following equation.

Pmix ) (PiXi + PjXj)/(Xi + Xj)

(3)

where Pi and Pj are the relative permittivities of solvent (7) Gutmann, V. Angew. Chem., Int. Ed. Engl. 1970, 9, 843. (8) Marzec, A.; Juzwa, M.; Betlej, K.; Sobkowiak, M. Fuel Proc. Technol. 1979, 2, 35. (9) Szeliga, J.; Marzec, A. Fuel 1983, 62, 1229. (10) Van Krevelen, D. W. COAL; Elsevier: Amsterdam, 1993; p 569.

Solvent effects on Hydrogenation of Mono-Aromatics

Energy & Fuels, Vol. 13, No. 6, 1999 1193

Table 3. Hydrogenation Conversions of Model Compounds Dissolved in Alcohols over Ru/Al2O3 and Pt/Al2O3 Catalysts conversion of substrate [mol %] solvent substrate

catalyst

MeOH

EtOH

1-PrOH

toluene phenol benzyl alcohol benzoic acid toluene phenol benzyl alcohol benzoic acid

Ru/Al2O3a Ru/Al2O3a Ru/Al2O3a Ru/Al2O3a Pt/Al2O3c Pt/Al2O3c Pt/Al2O3 Pt/Al2O3b

95 100 87 100 0 5 0c 0

100 100 100 100 0 74 21a 0

100 100 100 100 0 86 26a 10

c

a substrate:solvent ) 3 g:6 g. b substrate:solvent ) 3 g:1.5 g. substrate:solvent ) 3 g:0.75 g.

i and solvent j, respectively.11 Xi and Xj are the mass fractions of solvent i and solvent j, respectively, (Xi + Xj ) 1). Results Hydrogenation Reactivity of Mono-Aromatics. Table 3 shows the hydrogenation conversions of toluene, phenol, benzyl alcohol, and benzoic acid in alcohols over the Ru/Al2O3 and Pt/Al2O3 catalysts. The reaction time was 2 h, and the other reaction conditions were as described in the caption. For the case of the Ru/Al2O3 catalyst, complete conversion was observed for all substrates when ethanol and 1-propanol were used as the solvent. The conversion of phenol and benzoic acid was also 100 mol % in methanol, while that of toluene and benzyl alcohol was 95 and 87 mol %, respectively. The hydrogenation of toluene was strongly inhibited in the presence of methanol over the Pt/Al2O3 catalyst. The conversion was negligible when 0.75 g of methanol was added to the reaction system. The hydrogenation conversion of benzyl alcohol and benzoic acid over the Pt/ Al2O3 catalyst was also decreased by the addition of alcohols, resulting in a conversion of 21 mol % for benzyl alcohol with 6 g of ethanol and 0 mol % for benzoic acid with 1.5 g of ethanol. The conversion of toluene (86 mol %) was slightly affected by the presence of 1-propanol. Hydrogenation Reactivity of Benzyl Alcohol over the Ru/Al2O3 Catalyst. Figure 1 shows product distributions for the hydrogenation of benzyl alcohol in the presence of the various solvents over the Ru/Al2O3 catalyst. The reaction time was 0.5 h, and 6 g of each solvent was used. The total conversion of benzyl alcohol was 85 mol % with no solvents and additives being present and decreased to 76 mol % in methanol, 82 mol % in ethanol, and 81 mol % in acetic acid. This indicates that methanol, ethanol, and acetic acid had little or no effect on the hydrogenation activity over the Ru/Al2O3 catalyst. The yield of cyclohexanecarbaldehyde was 50 mol % in the solvent-free case. When methanol and ethanol was used as the solvent, however, the yield of cyclohexanecarbaldehyde was decreased to 33 and 37 mol %, respectively, and the yield of toluene was higher than that observed for the solvent-free case. When 2-propanol was used as the solvent, the total conversion of benzyl alcohol was decreased to 35 mol %, and the (11) Fowler, F. W.; Katritzky, A. R.; Rutherford, R. J. D. J. Chem. Soc. (B) 1971, 460.

Figure 1. Product distributions for the hydrogenation of benzyl alcohol over the Ru/Al2O3 catalyst.

yield of toluene was much smaller than that in the presence of methanol and ethanol. When hexane and heptane, which are nonpolar solvents, were used as the solvent, the hydrogenation conversion of benzyl alcohol was 75 and 61 mol %, respectively. The nonpolar solvents had no effect on the hydrogenation reaction. However, the conversion of benzyl alcohol was decreased to 31 mol % by the addition of benzene. When acetone, THF, dioxane, and diethyl ether were used as the solvent, the conversion of benzyl alcohol was 31, 36, 46, and 28 mol %, respectively. These solvents strongly retarded the hydrogenation activity over the Ru/Al2O3 catalyst. The hydrogenation of benzyl alcohol was completely inhibited when DMSO, DMF, and NMP were used as the solvent. Effect of Additives on the Hydrogenation of Benzyl Alcohol. Figure 2 shows the product distributions for the hydrogenation of benzyl alcohol when dissolved in polar solvents (ethanol, acetone, THF) in the presence of carboxylic acids. In this case, the Ru/ Al2O3 catalyst was used, along with 6 g of each solvent and 3 g of each carboxylic acid being added. The reaction time was fixed at 0.5 h in this case. When ethanol was used as the solvent, the conversion of benzyl alcohol was 82 mol % in the absence of carboxylic acid, and the hydrogenation of benzyl alcohol was completed in the presence of acetic acid, butyric acid, and lauric acid. The yield of methylcyclohexane was 36, 51, and 43 mol % in acetic acid, butyric acid, and lauric acid, respectively, and 11 mol % using ethanol as the solvent with no carboxylic acid present. The added carboxylic acids remained unchanged during the hydrogenation reaction under the above reaction conditions. However, formic acid strongly inhibited the hydrogenation of benzyl alcohol. When THF and acetone were used as the solvent, the addition of acetic acid increased the hydrogenation reactivity, as observed in ethanol. The hydro-

1194

Energy & Fuels, Vol. 13, No. 6, 1999

Figure 2. Product distributions for the hydrogenation of benzyl alcohol in polar solvents in the presence of carboxylic acids

Takagi et al.

Figure 4. Product yields for the hydrogenation of benzyl alcohol as a function of total hydrogenation conversion.

Figure 5. Pathways for the hydrogenation of benzyl alcohol over the Ru/Al2O3 catalyst.

Figure 3. Product yields for the hydrogenation of benzyl alcohol in THF over Ru/Al2O3 catalyst as a function of the amount of acetic acid.

genation conversion, which was 36 mol % without acetic acid, was increased to 88 mol % in the presence of acetic acid in THF. Figure 3 shows the product yields for the hydrogenation of benzyl alcohol in THF using the Ru/Al2O3 catalyst. The solvent was 6 g of THF, and the reaction time was 0.5 h. The amount of acetic acid was varied and was expressed in wt % with respect to the initial mass of benzyl alcohol in this case. The conversion of benzyl alcohol was 36 mol % for the reaction without acetic acid and was increased to 84 mol % in the presence of 17 wt % acetic acid. The yields of toluene and cyclohexanecarbaldehyde, which were products of

the hydrogenation of benzyl alcohol, remained unchanged even if the amount of added acetic acid was increased beyond 17 wt %. The yield of methylcyclohexane increased with increasing amounts of acetic acid, reaching 15 mol % when acetic acid was added to a level of 200 wt % with respect to the benzyl alcohol. Figure 4 shows the product yields for the hydrogenation of benzyl alcohol as a function of the total conversion of benzyl alcohol. The yield of cyclohexanecarbaldehyde increased with increasing conversion of benzyl alcohol in the range below 60 mol % and remained unchanged at higher conversions. The yield of toluene reached a maximum value for a benzyl alcohol conversion of 80 mol %. The yield of toluene decreased, and that of methylcyclohexane increased, with increasing conversion of benzyl alcohol in the range above 80 mol %. Discussion Pathway for Hydrogenation of Benzyl Alcohol over Ru Catalyst. Figure 5 shows the pathways for the hydrogenation of benzyl alcohol over the Ru/Al2O3 catalyst at 120 °C at an initial hydrogen pressure of 6 MPa. The hydrogenation of the aromatic ring (route 1)

Solvent effects on Hydrogenation of Mono-Aromatics

Energy & Fuels, Vol. 13, No. 6, 1999 1195

Figure 6. Relationship between hydrogenation conversion of benzyl alcohol and δ values of solvents.

and the hydrogenolysis of hydroxy group (route 2) proceed competitively. In route 1, four hydrogen atoms are introduced into benzyl alcohol, giving rise to 1-cyclohexenylmethanol. The resulting 1-cyclohexenylmethanol is then converted to cyclohexanecarbaldehyde.12 Toluene is produced via route 2, and methylcyclohexane is produced via the hydrogenation of toluene. The yield of toluene, which is formed by the hydrogenolysis of benzyl alcohol, is increased by the addition of carboxylic acids, which appear to promote this hydrogenolysis reaction. However, the data herein are not comprehensive and should not be used as absolute proof of this conclusion. Solvent Effects. Noble-metal catalysts selectively adsorb solvents which contain oxygen, nitrogen, and sulfur. The adsorption of the solvents is caused by the interaction between the metals and the unshared pairs of electrons of the oxygen, nitrogen, and sulfur atoms of the solvent molecules.13,14 Thus, the hydrogenation reactivity of aromatic compounds in those solvents over noble-metal catalysts is decreased by the competitive adsorption of the substrate and the solvents to the catalyst surface. The Pt/Al2O3 catalyst interacts with the solvents more strongly than the Ru/Al2O3 catalyst. As a result, the hydrogenation reactivity of the substrate is significantly inhibited by the presence of such solvents over the Pt/Al2O3 catalyst. Figure 6 shows the relationship between the hydrogenation conversion of benzyl alcohol and the δ values of the solvents. The donor number represents the parameter which is associated with the solvent basicity, while the acceptor number is associated with the solvent acidity.10 Thus, the effect of solvents on the hydrogenation reactivity of benzyl alcohol using the Ru catalyst can be classified into four categories as follows: (1) Solvents with negative δ values are capable of accepting electrons, and methanol, ethanol, acetic acid, and hexane are classified into this group. The AN value of acetic acid is higher than that of the other solvents. However, the conversion with respect to the hydrogena(12) Nishimura, S.; Hama, M. Bull. Chem. Soc. Jpn. 1966, 39, 2467. (13) Mitsui, S.; Senda, Y.; Shimodaira, T.; Ichikawa, H. Bull. Chem. Soc. Jpn. 1965, 38, 1897. (14) Mitsui, S.; Sakai, T.; Saito, H. J. Chem. Soc. Jpn. 1965, 86, 409.

Figure 7. Relationship between hydrogenation conversion of benzyl alcohol and relative permittivity of solutions. Solvents: (a) EtOH, (b) acetone, (c) THF.

tion of benzyl alcohol is approximately 80 mol % for all these solvents. Thus, the negative δ values of the solvents have no effect on the hydrogenation reactivity of benzyl alcohol over the Ru catalyst. (2) Benzene and 2-PrOH possess negative δ values and accept electrons. However, these solvents decrease the hydrogenation reactivity of benzyl alcohol. Benzene is completely converted to cyclohexane under the experimental conditions used in this study. Thus, hydrogenation reactivity is controlled by the competitive reaction between benzyl alcohol and benzene, regardless of the negative δ value of benzene. In the case of 2-PrOH, as shown in Figure 2, the yield of toluene is decreased in the presence of 2-PrOH to a greater extent than by methanol and ethanol. Meanwhile, the yield of cyclohexanecarbaldehyde in 2-PrOH is the same as that in methanol and ethanol. Thus, the hydrogenolysis of benzyl alcohol is suppressed by the addition of 2-PrOH. (3) Solvents with positive δ values are capable of transferring electrons; acetone, THF, dioxane, and di-

1196

Energy & Fuels, Vol. 13, No. 6, 1999

ethyl ether can be classified into this group. In the presence of these solvents, the hydrogenation conversion decreases with increasing δ values of the solvents. Ruthenium adsorbs oxygen-containing solvents13,14 via an interaction between the surface of the ruthenium and the pair of unshared electrons of the oxygen atom of the solvents. The active sites of the catalyst are occupied by electron donor solvents to a greater extent than electron acceptor solvents, and, as a result, the hydrogenation conversion of the substrate is decreased in the presence of the members of this solvent group. (4) The activity of the Ru catalyst is inactivated in the presence of DMSO, DMF, and NMP, which contain either sulfur or nitrogen. Figure 7 shows the relationship between the hydrogenation conversion of benzyl alcohol and the relative permittivity of the solutions. The relative permittivity is related to solvent polarity.11 The relative permittivity is 13.1 for benzyl alcohol, 24.3 for ethanol, 20.7 for acetone, and 7.4 for THF. The relative permittivity of acetic acid is 6.2, which is much smaller than that of formic acid, 58.0. Thus, the polarity of mixed solvents is decreased by the addition of acetic acid, and increased by that of formic acid. By the addition of acetic acid, the hydrogenation conversion of benzyl alcohol is increased from 82 to 100 mol % in ethanol, from 31 to 75 mol % in acetone, and from 36 to 88 mol % in THF. However, the hydrogenation reactivity in ethanol, acetone, and THF is diminished by the addition of formic acid. Since formic acid is a solvent as protogenic as acetic acid, it is an electron acceptor. In addition, the relative permittivity of formic acid is much larger than that of acetic acid. Thus, the hydrogenation reactivity is affected by relative permittivity rather than the δ value for the case of formic acid. Conclusions (1) The Ru/Al2O3 catalyst was effective for the hydrogenation of mono-aromatic rings when lower alcohols

Takagi et al.

were used as solvents, while the Pt/Al2O3 catalyst was strongly inhibited by these alcohols. (2) A relationship between the hydrogenation reactivity of benzyl alcohol in polar solvent and the δ value of the solvent was found. Solvents with negative δ values, i.e., methanol, ethanol, acetic acid, and hexane, did not affect the hydrogenation of benzyl alcohol over the Ru catalyst. However, solvents with positive δ values, i.e., acetone, THF, dioxane, and diethyl ether, suppressed the hydrogenation of benzyl alcohol. (3) Benzene was hydrogenated in preference to benzyl alcohol. As a result, the hydrogenation of benzyl alcohol was decreased by the presence of benzene. (4) DMSO, DMF, and NMP, which contained sulfur or nitrogen, deactivated the Ru catalyst. (5) The hydrogenation reactivity of benzyl alcohol in polar solvents was increased by the addition of acetic acid, butyric acid, and lauric acid. The hydrogenation reactivity was related to the relative permittivity of the solution and was completely lost by the addition of formic acid with a high permittivity. Acknowledgment. This research was supported by Ministry of Education, Science, Sports and Culture, Japan Society for the Promotion of Science (JSPS), Organization of New Energy and Industrial Technology Development Organization (NEDO), and Center for Clean Coal Utilization, Japan (CCUJ). Support by Research for the Future Projects (Coordinator, M. Iino, Tohoku University), as well as International Joint Research Program (Coordinator, M. Nomura, Osaka University), is especially acknowledged. EF990061M