Synlett 2014; 25(17): 2390-2414
DOI: 10.1055/s-0034-1379210
account
© Georg Thieme Verlag Stuttgart · New York

Chelation-Assisted Regioselective Catalytic Functionalization of C–H, C–O, C–N and C–F Bonds

Fumitoshi Kakiuchi*
a   Department of Chemistry, Faculty of Science and Technology, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama, Kanagawa 223-8522, Japan   Fax: +81(45)5661591   Email: kakiuchi@chem.keio.ac.jp
b   JST, ACT-C, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
,
Takuya Kochi
a   Department of Chemistry, Faculty of Science and Technology, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama, Kanagawa 223-8522, Japan   Fax: +81(45)5661591   Email: kakiuchi@chem.keio.ac.jp
,
Shinji Murai
c   Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara 630-0192, Japan. Professor Emeritus, Osaka University   Email: murai@ad.naist.jp
› Author Affiliations
Further Information

Publication History

Received: 21 July 2014

Accepted after revision: 29 August 2014

Publication Date:
30 September 2014 (online)

 


Abstract

Transition-metal-catalyzed functionalization of unreactive carbon bonds such as C–H, C–O, C–N, and C–F bonds has been extensively studied over the last two decades. In this account we describe our studies on chelation-assisted catalytic transformations of unreactive carbon bonds in aromatic and olefinic compounds. Various C–C bond forming reactions are achieved by catalytic addition of C–H bonds to alkenes and alkynes, and via coupling reactions such as those of C–H, C–O, C–N, and C–F bonds with boronic esters. Methods for the conversion of C–H bonds into C–Si and C–X (X = halogen) bonds are also developed. The C–C bond forming reactions are applied to short syntheses of polycyclic aromatic hydrocarbons.

1 Introduction

2 Conversion of C–H Bonds into C–C Bonds

2.1 C–H Alkylation

2.1.1 Scope and Limitations

2.1.2 Mechanistic Studies

2.2 C–H Arylation

2.3 C–H Alkenylation

2.4 Introduction of Carbonyl Functionality via C–H Bond Cleavage

3 Conversion of Aromatic C–O, C–N and C–F Bonds into C–C Bonds by Coupling with Organoboronates

3.1 Coupling with Organoboronates via Aromatic C–O Bond Cleavage

3.2 Coupling with Organoboronates via Aromatic C–N Bond Cleavage

3.3 Coupling with Organoboronates via Aromatic C–F Bond Cleavage

4 Introduction of Heteroatoms at C–H Bonds

4.1 Silylation of C–H Bonds

4.1.1 Silylation of Aromatic C–H Bonds with Hydrosilanes

4.1.2 Silylation of Benzylic C(sp3)–H Bonds with Hydrosilanes

4.1.3 C–H Silylation with Vinylsilanes

4.2 Aromatic C–H Halogenation by Palladium-Catalyzed C–H Bond Cleavage and Electrochemical Oxidation

4.2.1 C–H Chlorination and Bromination

4.2.2 C–H Iodination and One-Pot C–H Iodination/Suzuki–Miyaura Coupling by ON/OFF Switching of Electric Current

5 Applications of C–H Functionalization in Short Syntheses of Polycyclic Aromatic Hydrocarbons (PAHs)

5.1 Convenient Synthesis of Multiarylanthracenes

5.2 Short Synthesis of Alkyl-Substituted Anthracenes and Pentacenes

5.3 Short Synthesis of Dibenzo[a,h]anthracenes and Picenes

6 Conclusions


#

Biographical Sketches

Zoom Image

Fumitoshi Kakiuchi was born in Hyogo, Japan, in 1965 and received his B.Sc. in 1988 and Ph.D. in 1993 from Osaka University under the guidance of Prof. Shinji Murai. He was appointed as an Assistant Professor at Osaka University in 1993. He carried out postdoctoral work with Prof. E. N. Jacobsen at Harvard University between 1996–1997. In 2000, he was promoted to an Associate Professor at Osaka University. In 2005, he moved to Keio University as a Professor. His research interests include the development of new transition-metal-catalyzed reactions.

Zoom Image

Takuya Kochi was born in Tokyo, Japan in 1975. He received his undergraduate and master’s degrees from the University of Tokyo, working with Professor ­Masanobu Hidai and Professor Youichi Ishii, and his Ph.D. in chemistry from the University of California at Berkeley, working with Professor Jonathan A. ­Ellman. After carrying out postdoctoral research with Professor Kyoko Nozaki at the University of Tokyo, he joined the group of Professor Fumitoshi Kakiuchi at Keio University as an Assistant Professor in 2007. In 2010, he was promoted to an Assistant Professor/Lecturer at Keio University. His research interests include the development of new reactions and their applications in the synthesis of a wide range of organic molecules.

Zoom Image

Shinji Murai is Emeritus Professor of Chemistry at Osaka University. He was born in Osaka in 1938 and received his education at Osaka University (B.Sc., 1961 and Ph.D., 1966, with Professor S. Tsutsumi). He was appointed research associate in 1966 and then full professor in 1987 until his retirement in 2002. During this period he served as the Dean of the Faculty of Engineering for two years. His research group has actively developed many new synthetic reactions utilizing the characteristic properties of both main group and transition elements. Their breakthrough discovery of a powerful ruthenium-catalyzed process for the functionalization of unactivated C–H bonds was particularly exciting. His present responsibility at JST (Japan Science and Technology Agency, a Government subsidiary for funding) is to identify subjects in the fields of chemistry and nanoscience to be funded. He also works for the Nara Institute of Science and Technology and for Iwatani Corporation. Dr. Murai has served as a President of the Chemical Society of Japan.

1

Introduction

Transition-metal-catalyzed coupling reactions to form carbon–carbon and carbon–heteroatom bonds have been studied extensively, and numerous practical methods have been developed for these transformations.[1] Many of the coupling reactions are initiated by the generation of σ-bonded organotransition metal species via oxidative addition of C–X bonds to transition metals, where X is generally a good leaving group such as iodide, bromide, chloride or triflate. However, the reactive X group has to be pre-installed and typically are not incorporated in the final products.

As another transition-metal-catalyzed coupling approach, catalytic functionalization of carbon–hydrogen bonds has emerged as a powerful synthetic tool which can be employed without sacrificing any functional group.[2] Initial reports on C–H bond cleavage using stoichiometric amounts of transition metal complexes dates back to the 1960s.[3] [4] Kleiman and Dubeck described the first chelation-assisted, regioselective C–H bond cleavage using azobenzene and dicyclopentadienylnickel (Cp2Ni).[3] Chatt and Davidson reported the first direct observation of the oxidative addition of a C–H bond to a low-valent transition metal complex.[4] In the 1980s, significant progress was made on understanding the mechanism of C–H bond cleavage processes such as oxidative addition of C–H bonds, especially sp3 C–H bonds in aliphatic hydrocarbons,[5] whilst organic and organometallic chemists were also motivated to develop transition-metal-catalyzed C–H functionalizations.[2] These transformations are often referred to as C–H bond activation.

It is important to note that several reports on catalytic C–C bond formation, which is considered to proceed via cleavage of C–H bonds by a transition metals complex, had been published by 1980. In 1955, Murahashi reported that the dicobalt octacarbonyl [Co2(CO)8] catalyzed reaction of azobenzenes with carbon monoxide gave benzolactams (Scheme [1]).[6] The mechanism was not described but the reaction appeared to proceed via ortho-selective C–H bond cleavage by the cobalt catalyst. Yamazaki and co-workers have developed several catalytic functionalizations of aromatic C–H bonds, which are considered to proceed via oxidative addition of a C–H bond to a rhodium(0) species (Scheme [2]).[7]

Zoom Image
Scheme 1
Zoom Image
Scheme 2

An early example of the catalytic C–H carbonylation of alkanes under photoirradiation conditions was reported using a rhodium catalyst by Tanaka and co-workers (Scheme [3]).[8a] [b] Their reaction system provided crucial advances in C–H functionalization, though improvements in the regioselectivity and the catalytic efficiency were still desired. Later, Moore described an efficient Ru3(CO)12-catalyzed acylation of C–H bonds in pyridines using carbon monoxide and alkenes.[8c]

Zoom Image
Scheme 3

Nowadays, many catalytic C–H functionalizations are reported each year. These reactions may be categorized into four major classes based on how the C–H bonds are cleaved by transition metal catalysts (Scheme [4]). The first class involves concerted oxidative addition of C–H bonds to low-valent transition metals (path a).[5] [9] In the second, the C–H bond cleavage is initiated by nucleophilic attack of the metal at an electron-deficient carbon, followed by 1,2-hydrogen migration from the carbon atom to the metal center to complete a stepwise oxidative addition (path b).[10,11] In the third class of reaction, the C–H bond cleavage occurs via electrophilic substitution at electron-rich C–H bonds with high-valent oxidation state transition metal complexes, and this process has been well documented as electrophilic metalation (path c).[12] The C–H bond cleavage process in the fourth class of reaction is the concerted metalation–deprotonation (CMD) pathway (alternatively called the base-assisted proton-abstraction pathway), where relatively electron-deficient arenes and heteroarenes react with high-valent oxidation state transition metal complexes and bases (path d).[13]

In paths a and b, the hydrogen atom of the C–H bond remains on the metal as a hydride after the C–H bond cleavage, and can be directly used for the subsequent bond formation via migratory insertion followed by reductive elimination. On the other hand, paths c and d lead to situations where only the carbon atom binds to the metal and the hydrogen is removed as a proton.

We have previously reviewed[14] our earlier work regarding catalytic additions of aromatic and olefinic C–H bonds to carbon–carbon multiple bonds using ruthenium and rhodium catalysts.[15] [16] In this account, we describe our research on catalytic transformations of unreactive carbon bonds including examples of more recent work such as the conversion of C–H bonds into C–C bonds using organo­boron compounds,[17] alkenyl esters,[18] acid chloride derivatives,[19] into C–Si bonds using hydrosilanes or vinylsilanes,[20] and into C–X (X = Cl, Br, I) bonds by means of a combination of C–H bond cleavage and electrochemical oxidation.[21] We also describe catalytic coupling of aryl ethers,[22] amines,[23] and fluorides[24] with organoboron compounds and the applications of these C–C bond formation reactions in short syntheses of organic electronic materials.[17c] [25]


# 2

Conversion of C–H Bonds into C–C Bonds

2.1

C–H Alkylation

2.1.1

Scope and Limitations

Catalytic alkylation of aromatic C–H bonds using alkenes are valuable reactions for syntheses of alkylarenes without loss of any atoms. When we began our C–H alkylation studies, there were only two reports on the regioselective C–H alkylation of a limited sets of substrates.[26] [27a] The first described the ethylation of phenols with ethene using a ruthenium phosphite complex as the catalyst (Scheme [5]).[26] Exchange of the aryloxy moieties of the phosphite ligands with phenols occurs during the reaction and coordination of the phosphorus atom in the phosphite assists the C–H bond cleavage.

Zoom Image
Scheme 4
Zoom Image
Scheme 5

The second report outlined the zirconium(IV)-catalyzed alkylation of α-C–H bonds in α-picoline with terminal alkenes under a hydrogen atmosphere, which kept the catalyst active, but resulted in consumption of the starting alkene by hydrogenation (Scheme [6]).[27a] This reaction was later extended to an asymmetric version using a chiral zirconocene derivative.[27b]

Zoom Image
Scheme 6

In 1993, we reported the first general regioselective catalytic C–H functionalization method using aromatic ketones and alkenes.[15a] For example, when the reaction of 2′-methylacetophenone (1) with triethoxyvinylsilane (2) was carried out in the presence of RuH2(CO)(PPh3)3 (3) as the catalyst in toluene at reflux temperature, the ortho alkylation product 4 was obtained in 93% yield as the sole product (Scheme [7]). A C–C bond was formed between the ortho carbon of aromatic ketone 1 and the terminal carbon of alkene 2 to give linear alkylation product 4. One of the most important features of this result is that complete regioselectivity was achieved by employing an acyl group as a directing group. That is, coordination of the ketone carbonyl oxygen brings the ruthenium close to the ortho C–H bond, and formation of a five-membered ruthenacycle after the C–H bond cleavage stabilizes the Ar–Ru–H intermediate, which, in general, is thermally less stable without the chelate formation. In this reaction, several alkenes such as vinylsilanes, 3,3-dimethylbut-1-ene, styrenes, hex-1-ene, and norbornene can be used, and other than complex 3, Ru(CO)2(PPh3)3 and RuH2(PPh3)4 also function as catalysts. Coupling of acetophenones possessing two ortho C–H bonds with alkenes, except with styrenes, gave 2′,6′-dialkylation products selectively.[14] The coupling with terminal alkenes having allylic C–H bonds, such as hex-1-ene, resulted in low yields under reflux in toluene, but almost quantitative yields could be achieved when the reaction temperature was increased to 170 °C.[25b] The chelation-assistance strategy was also applied to the reactions of various aromatic and olefinic compounds bearing ketone, ester, aldehyde, nitrile, imine, and hydrazine moieties and nitrogen-containing heterocycles.[15] The chelation-assistance strategy can be employed in all of the C–H bond cleavage pathways shown in Scheme [4], and has been used by many research groups to achieve regioselective C–H functionalization since our initial report was published.[28]

Zoom Image
Scheme 7

# 2.1.2

Mechanistic Studies

The mechanism of the C–H alkylation was investigated by deuterium-labeling experiments and kinetic studies.[10] [15c] [h] The outline of our proposed mechanism is shown in Scheme [8]. The catalytic cycle starts with coordination of the carbonyl oxygen to the ruthenium center. Oxidative addition of the ortho C–H bond to the metal center provides a ruthenacycle intermediate. Next, hydroruthenation of the olefin followed by reductive elimination gives the alkylation product. Theoretical calculations by Koga, ­Morokuma and co-workers supported the catalytic cycle and also indicated that the C–H bond cleavage proceeds through nucleophilic attack of a ruthenium(0) species at the ortho carbon, followed by a 1,2-hydrogen shift onto the ruthenium center.[11] The reductive elimination is also suggested to occur through alkyl migration from the ruthenium center onto the ortho carbon.[11b] [15h]

Zoom Image
Scheme 8

Stoichiometric reactions of ruthenium catalyst 3 with olefins gave new ruthenium species, which were catalytically active toward C–H alkylation. Treatment of catalyst 3 with trimethylvinylsilane (5) at 90 °C for 1.5 hours afforded activated ruthenium catalysts, Ru(o-C6H4PPh2)(H)(CO)(PPh3)2 (6) as a mixture of four geometric isomers 6ad (Scheme [9]). The activated complex 6 showed high catalytic activity for the C–H alkylation with alkenes, and it turned out that the reaction of 1 with 5 proceeded at room temperature to give the corresponding ortho alkylation product 7 in 99% yield after 48 hours (Scheme [10]).[10] Activated complex 6 also showed excellent catalytic activity (turnover number = 994) at a high (120 °C) reaction temperature and low catalyst loading (0.1 mol%).

Zoom Image
Scheme 9
Zoom Image
Scheme 10

The 1H and 31P NMR spectroscopic studies showed that complex 6 reacts with ketone 1 to give ortho-ruthenated acetophenone complex 8, which is supposedly formed via oxidative addition of an ortho C–H bond to the ruthenium center (Figure [1]). Both complexes 6 and 8 were observed in the reaction mixture while the alkylation progresses. When ketone 1 had been completely consumed, complex 8 disappeared and 6 became the main species, but addition of 1 to the mixture led to the regeneration of 8. These observations indicate that complexes 6 and 8 are relevant intermediates in the C–H alkylation.

Zoom Image
Figure 1 Structure of ortho-ruthenated acetophenone complex 8
Zoom Image
Scheme 11

Deuterium-labeling experiments on the C–H alkylation were performed using acetophenone-d 0 (9) and acetophenone-d 5 (9-d 5 ) at room temperature under pseudo-first-­order kinetic conditions (Scheme [11]). The observed kinetic isotope effect (KIE), k obs(9)/k obs(9-d 5 ) was 1.02, and this small KIE value indicates that the C–H bond cleavage step is not turnover-limiting.[10]

Partial incorporation of deuterium atoms in exchange for each of the three vinylic hydrogens was detected by 2H NMR experiments on the C–H alkylation of 9-d 5 with alkene 5 (Scheme [12]). The partial H/D exchange between the ketone and the vinylsilane supports the proposed mechanism where the turnover-limiting step is reductive elimination, and each prior step is fast even at room temperature.

Zoom Image
Scheme 12
Zoom Image
Scheme 13

Deuterium-labeling experiments were also conducted on the alkylation of aromatic esters using methyl benzoate-d 5 (10-d 5 ) (Scheme [13]). In addition, 13C kinetic isotope effects[29] were measured at natural abundance using methyl benzoate (11) as the substrate (Table [1]).[15h] The results of these experiments suggest that the reductive elimination step is turnover-limiting in the C–H alkylation of aromatic esters as well.

Table 1 13C KIE Measurements at Natural Abundance

Entry

Conversion

KIEs

C1

C2

C3 + C4

C6

1

64%

0.996(1)

1.001(3)

1.003(1)

1.034(1)

2

69%

0.999(2)

0.996(3)

1.000(2)

1.032(2)

3

79%

0.997(1)

0.996(1)

1.004(1)

1.034(1)


#
# 2.2

C–H Arylation

Catalytic cross-couplings of aryl halides and pseudo-halides with organometallic reagents have revolutionized the way to construct biaryl frameworks. But the necessity of pre-installation of reactive functional groups such as Cl, Br, I, OTf, or N2 +, which are not retained in the final products, has motivated researchers to develop more efficient routes to prepare biaryls. In this context, direct C–H arylations of arenes have emerged as a powerful tool for biaryl synthesis.[30] To date, a large number of catalytic C–H arylation methods have been developed, and this is perhaps the most studied area in catalytic C–H functionalization.

When we initiated our research on the C–H arylation with arylboronates,[17a] there was only one report on catalytic C–H arylation using an aryl–metal reagent. In 1998, Oi and co-workers had published their results on the rhodium-catalyzed ortho-selective C–H arylation of arylpyridines with tetraarylstannanes using 1,1,2,2-tetrachloroethane (CHCl2CHCl2) as the solvent.[31]

Zoom Image
Scheme 14 Proposed catalytic cycle for the C–H arylation of acetophenone with arylboronates

We developed the catalytic C–H arylation of aromatic ketones with arylboronates, in which the C–H bond was supposedly cleaved via oxidative addition by a low-valent ruthenium species. The proposed catalytic cycle for this C–H arylation is shown in Scheme [14].[17b] The reaction starts with coordination of the aromatic ketone to ruthenium and oxidative addition of the ortho C–H bond to the ruthenium(0) species. The Ru–H bond of the resulting ­ortho-ruthenated complex 12 then adds to the carbonyl group of another ketone molecule to form ruthenium alkoxide species 13. Transmetalation between 13 and the arylboronate gives the diaryl ruthenium complex 14, and subsequent reductive elimination yields the coupling product along with regeneration of the ruthenium(0) species.

The reaction of ketone 1 with one equivalent of phenylboronate 15 using 3 as the catalyst, performed in toluene at reflux temperature, gave the ortho phenylation product 16 in 47% yield along with the product of reduction 17 (Table [2], entry 1). This result suggested that ketone 1 functioned as the substrate as well as the hydride acceptor.

Table 2 Reactions of Ketone 1 with Phenylboronate 15 in the Presence of Catalyst 3

Entry

1 (equiv)

15 (equiv)

Solvent

Yield of 16 and 17

1

1

1

toluene

47%a

40%a

2

2

1

toluene

80%b

71%b

3

1

1.1

pinacolone

85%a

trace

a Based on 1.

b Based on 15.

The use of two equivalents of 1 with respect to phenylboronate 15 improved the yield to 80% (Table [2], entry 2). The drawback of this C–H arylation is that a half of the aromatic ketone is lost during the reaction. Screening of various reaction conditions led us to the finding that some aliphatic ketones such as pinacolone and acetone could be used as hydride acceptors.[17b] When the reaction of 1 with 1.1 equivalents of 15 was carried out in the presence of catalyst 3 in pinacolone under reflux for one hour, coupling product 16 was obtained in 85% isolated yield based on 1 (Table [2], entry 3).

The C–H arylation was applicable to a variety of aromatic ketones such as acetophenone derivatives having electron-donating or electron-withdrawing groups, aceto­naphthone, α-tetralone, and 1-benzosuberone (Figure [2]). In all cases, the corresponding ortho-arylated aromatic ketones were obtained in high to excellent yields. The presence of both electron-donating and electron-withdrawing groups, such as Me, OMe, NMe2, and CF3, on the arylboronates was tolerated. Furthermore, heteroarylboronates could also be used in this reaction.

Zoom Image
Figure 2 The generality of the C–H arylation of aromatic ketones with arylboronates

As shown in Figure [2], the reaction of acetophenone (9) gave the diphenylation product in high yield, and in this case, the second arylation was so fast that it was difficult to obtain the corresponding monoarylation product under the reaction conditions, even after shorter reaction times. Selective synthesis of both mono- and diarylation products is one of the challenges in C–H arylation chemistry, especially when the second arylation is not much slower than the first. According to the literature, control of the product selectivity was achieved using different directing groups[32] and leaving groups,[33] but methods to prepare both mono- and diarylation products from the same sets of substrates are desired.

In this context, screening of additives for the C–H arylation of 9 was conducted to prepare the monophenylation product, selectively. As a result, the unique effect of styrene (18) as an additive on controlling the selectivity of the mono- and diarylation products was realized.[17d] When the C–H arylation was performed using a 3:1 ratio of ketone 9 with phenylboronate 15, monophenylation product 19a was formed in 15% yield along with a 64% yield of the diarylation product 20a (Table [3], entry 1). However, simple addition of one equivalent of 18 switched the product selectivity to give 19a and 20a in 56% and 15% yields, respectively (Table [3], entry 2). These results indicate that 18 retards the second C–H arylation step. The couplings with a 4-N,N-dimethylaminophenylboronate provided the corresponding monoarylation products in high yields. In these reactions, C–H alkylation products with styrene were also formed as by-products. Coordination of styrene (18) to the ruthenium center is thought to facilitate reductive elimination and stabilize the electron-rich ruthenium(0) species formed (Scheme [15], step a). In addition, the second C–H bond cleavage is suppressed due to the π-acidic character of 18 (Scheme [15], step b).

Table 3 Selective Monoarylation of Acetophenone Derivatives

Entry

R

Ar

18 (equiv)

Yield of 19 and 20

1

H (9)

Ph

none

15% (19a)

64% (20a)

2

H (9)

Ph

1

56% (19a)

15% (20a)

3

H (9)

4-Me2NC6H4

1

78%

11%

4

OMe

4-Me2NC6H4

1

65%

5

t-Bu

4-Me2NC6H4

1

83%

Zoom Image
Scheme 15 Proposed mechanism for the C–H arylation in the presence of styrene (18)

The good ability of the catalyst system to affect multiple arylation of aromatic ketones led us to examine the C–H arylation of anthraquinone (21). When the reaction of 21 with p-tolylboronate 22a was performed in pinacolone under reflux, the arylation proceeded at all of the four ­ortho C–H bonds to give 1,4,5,8-tetra-p-tolylanthraquinone (23a) in 77% isolated yield (Scheme [16]).[17c] A variety of arylboronates including 3,5-dimethyl- (22b), 4-n-hexyl- (22c), 4-trifluoromethyl- (22d), 4-methoxy- (22e), and 4-N,N-dimethylamino-phenylboronate (22f) could be used in this reaction to give the corresponding tetraaryl products 23 in good to high yields.

Zoom Image
Scheme 16

# 2.3

C–H Alkenylation

Alkenylation of C–H bonds in aromatic and olefinic compounds provides an efficient route to styrene derivatives and conjugated dienes. There are four major strategies used for catalytic C–H alkenylation: (1) addition of C–H bonds to alkynes;[16] [34] [35] (2) oxidative C–H alkenylation with alkenes;[15j–m] , [36,37] (3) coupling with alkenyl halides and pseudohalides;[18,38] and (4) oxidative coupling with alkenyl metal reagents.[17e] [f] In this section, we describe our approaches for the alkenylations of aromatic and olefinic C–H bonds: (1) C–H/alkyne coupling,[16] (2) dehydrogenative coupling with alkenes,[15j] [k] [l] [m] (3) coupling with alkenyl esters,[18] and (4) coupling with alkenylboronates.[17e] [f]

The addition of C–H bonds to alkynes can be considered as the most simple C–H alkenylation method.[16] The reaction of aromatic ketones with internal alkynes using 3 as the catalyst in toluene under refluxing conditions gives ­ortho-alkenylated products. Trimethylsilylacetylenes 25ac react with α-tetralone (24) regioselectively to provide β-silylstyrene derivatives 26ac (Scheme [17]). In the case of 1-trimethylsilylpropyne (25c), the reaction proceeded regio- and stereoselectively to give E-isomer 26c as the sole product.

Zoom Image
Scheme 17
Zoom Image
Scheme 18

Similarly, α,β-unsaturated ketones 27 can be used for the C–H alkenylation with internal alkynes.[16b] The reaction of 1-pivaloyl-1-cyclohexene (27a) with diphenylacetylene (25d) in the presence of catalyst 3 afforded conjugated dienone 28ad in 85% yield as a mixture of E- and Z-isomers (Scheme [18]). Oxygen-containing cyclic enones 27b,c and silylacetylene 27b were also applicable for this alkenylation.

Transition-metal-catalyzed dehydrogenative couplings of arenes with alkenes have been known for several decades. Reaction systems using palladium(II) catalysts and oxidants were developed by Moritani, Fujiwara, and co-workers more than 40 years ago,[36] and related catalytic reactions have been studied extensively.[37] In these reactions, stoichiometric amounts of oxidants such as copper(II) salts, silver(I) salts, and benzoquinone are essential in order to perform the reactions catalytically.

We have developed another type of oxidative arene/alkene coupling where the alkene functions as a coupling partner as well as a hydrogen scavenger.[15j] [k] [l] [m] When the reaction of o-tolyloxazoline 29 with vinylsilane 2 was carried out in the presence of Ru3(CO)12 (30) as the catalyst in toluene at reflux temperature, the dehydrogenative coupling product 31a was formed in 87% yield as the major product, along with a 10% yield of the alkylation product 32a (Scheme [19]).[15j] The reaction with 18 gave stilbene derivative 31b in 45% yield along with a nearly equimolar amount of ethylbenzene, which shows that 18 was not just a substrate but also a hydrogen scavenger.

Zoom Image
Scheme 19

Alkenyl acetates can be used as alkenylating agents for the C–H alkenylation of arenes bearing directing groups.[18a] [b] The most effective catalyst for this transformation was found to be Ru(cod)(cot) (33) (cod = 1,5-cyclooctadiene, cot = 1,3,5-cyclooctatriene). The reaction of 3-methyl-2-phenylpyridine (34) with β-styryl acetate (35) (E/Z = 58:42) using catalyst 33 provided alkenylation product 36 in 93% yield (Scheme [20]). A variety of alkenyl acetates and aromatic compounds having nitrogen-containing heterocycles can be used for this reaction. Increasing the steric bulk of the substituent at the β carbon of the alkenyl esters improved the E/Z selectivity. This reaction provides a halogen-, base-, and oxidant-free C–H alkenylation method.

Zoom Image
Scheme 20
Zoom Image
Scheme 21 Proposed mechanism for the C–H alkenylation with alkenyl acetate 35

The proposed reaction mechanism for the C–H alkenylation using an alkenyl acetate is summarized in Scheme [21]. The mechanistic studies of the Ru(cod)(cot)-catalyzed C–H alkenylation with alkenyl acetates involved deuterium-labeling experiments, kinetic studies, and structural determination of intermediates.[18b] The deuterium-labeling experiments suggested that the C–H bond cleavage takes place via oxidative addition and is not turnover-limiting. The kinetic studies showed that the reaction rate seems to be independent of the concentration of the arylpyridine during the reaction and is first order with respect to the catalyst 33 and the alkenyl acetate. The structures of two types of key intermediates were determined by NMR and single crystal X-ray diffraction analyses. Both types of intermediate contain an ortho-ruthenated arylpyridine moiety and an acetate ligand, but one (such as 37) has a COD ligand, while the other (such as 38) possesses the C–H alkenylation product as a bidentate ligand. Complex 37 showed similar catalytic activity to 33. However, lower activity was observed for 38 compared with 33 and 37, but intermediate 38 can be converted into 37 in the presence of COD and acetic acid. On the basis of the experimental results, it was indicated that the turnover-limiting step was exchange of the coupling product with the COD ligand.

Alkenyl carbonates are also effective substrates in Ru(cod)(cot)-catalyzed aromatic C–H alkenylation.[18c] In the case of alkenyl acetate, acetic acid was produced as a by-product, but in this case, the corresponding by-product was an alcohol, which is much less acidic. Therefore, in addition to arylpyridines, the more basic aryloxazoline 29 can be used as a substrate without the need to add an external base, and when coupled with ethyl styryl carbonate (39), was converted into ortho-styrylation product 40 in 77% yield (Scheme [22]).

Zoom Image
Scheme 22

The alkenylation of aromatic ketones was achieved using alkenylboronates.[17e] [f] The reaction of pivalophenone (41) with 2-propenylboronate 42 using catalyst 3 in pinacolone under reflux afforded ortho alkenylation product 43 in 73% yield (Scheme [23]).

Zoom Image
Scheme 23

The C–H alkenylation of 2′-methoxyacetophenone (44) with alkenylboronates proceeds chemoselectively at the ortho C–H bond without sacrificing the ortho methoxy group.[17f] When 44 was reacted with boronate 42 in pinacolone under reflux, the alkenylation took place selectively at the ortho C–H bond leaving the methoxy group intact to give monoalkenylation product 46a in 58% yield (Scheme [24]). In this case, a small amount of an alkenylation product without the methoxy group (47a) was formed. The coupling with β-E-styrylboronate 45 provided the corresponding alkenylation product 46b in 76% yield. Methoxy alkenylation product 47 is thought to be produced by alkenylation via cleavage of the ortho C–O bond, which will be discussed in Section 3.1

Zoom Image
Scheme 24

A stoichiometric reaction of activated complex 6 with o-aryloxyacetophenone 49 at room temperature led to exclusive formation of hydrido complex 50 (Scheme [25]).[17f] Subsequently, an excess amount of alkenylboronate 42 was added and the mixture heated at 80 °C. The 31P NMR spectra of the resulting mixture indicated that most of complex 50 had been converted into ruthenium complex 51, in which the alkenylation product coordinates to the ruthenium through the carbonyl oxygen and the olefin moiety. This observation suggests that coordination of the alkenyl moiety to the ruthenium center prevents the C–O bond cleavage.

Zoom Image
Scheme 25

# 2.4

Introduction of Carbonyl Functionality via C–H Bond Cleavage

Introduction of carbonyl functionalities via aromatic C–H bond cleavage provides an efficient route to prepare versatile synthetic intermediates and can be highly useful in organic synthesis. When we started our project regarding direct, catalytic introduction of carbonyl groups via C–H bond cleavage, various methods had been reported for the corresponding direct introductions of carboxy,[39] amide,[7] [19a] [40] ester,[19a,41] acyl,[8c,19b,42] and formyl[43] groups.

In order to develop methods for the introduction of carbonyl groups via C–H bond cleavage, the use of acid chlorides was examined because a variety of acid chlorides derived from carboxylic, carbamic, and carbonic acids are easily accessible and can be used for the facile incorporation of various types of carbonyl functionalities. When benzo[h]quinoline (52) was reacted with N,N-dimethylcarbamoyl chloride (53a) using RuCl2(PPh3)3 (54) as the catalyst in the presence of potassium carbonate (K2CO3) in toluene at 120 °C, ortho-aminocarbonylation product 55a was obtained in 94% isolated yield (Scheme [26]).[19a] The reactions using N,N-diphenylcarbamoyl chloride (53b) and 4-morpholinecarbonyl chloride (53c) afforded the corresponding amides 55b and 55c in 97% and 90% isolated yields, respectively. The aminocarboxylation was also applicable to other arylpyridines and an arylimidazole.

Zoom Image
Scheme 26

Alkoxycarbonylation of arylpyridines is also possible under similar reaction conditions.[19a] The reaction of 2-phenylpyridine 56 with ethyl and n-butyl chloroformates (57a,b) afforded the corresponding monoesters 58a and 58b in 50% and 56% yields, respectively (Figure [3]). The product selectivity of the alkoxycarbonylation, interestingly, was different from that of the aminocarboxylation. Arylpyridines having two ortho C–H bonds reacted with ethyl chloroformate (57a) to give the monoalkoxycarbonylation products 58 as the sole products, while the aminocarbonylation afforded the corresponding difunctionalized products. It is worth noting that Friedel–Crafts methods are not applicable for the reaction with alkyl chloroformates due to their rapid decarboxylation.[44]

Zoom Image
Figure 3 Products formed by the C–H alkoxycarbonylation of arylpyridines

An important feature of the alkoxycarbonylation is that 2-(3-trifluoromethylphenyl)pyridine (59a) reacted faster than 2-(3-methylphenyl)pyridine (59b), which is more electron-rich, in a competition experiment (Scheme [27]). Therefore, we proposed that the C–H bond cleavage step in this reaction proceeded via a concerted metalation–deprotonation (CMD) pathway, in which the more acidic C–H bonds are cleaved faster (Scheme [4]).[13a] [45] [46] However, alternative mechanisms are still possible.

Zoom Image
Scheme 27

The acylation of arylpyridines was similarly achieved by the ruthenium-catalyzed coupling using acyl chlorides. The reaction of 52 with benzoyl chloride (61a) led to aroylation product 62a in 95% isolated yield (Scheme [28]).[19b] Alkenoyl groups can also be introduced to the benzene rings of the arylpyridines using tigloyl chloride (62b) and 1-cyclohexenecarbonyl chloride (62c).

Zoom Image
Scheme 28

When ortho-ruthenated complex 63, as reported by Sirlin, Pfeffer, and co-workers,[47] was reacted with compound 64 at 120 °C, the corresponding acylation product 65 was formed in 92% yield according to GC (Scheme [29]). The result shows that the acylation product can be formed by the reaction of an acyl chloride with a ruthenacycle(II) species generated from an arylpyridine. Therefore, the catalytic acylation is considered to proceed via ortho-­selective C–H bond cleavage, followed by reaction with the acyl chloride.

Zoom Image
Scheme 29

#
# 3

Conversion of Aromatic C–O, C–N, and C–F Bonds into C–C Bonds by Coupling with Organoboronates

As discussed in the previous sections, our ongoing studies on the catalytic functionalization of C–H bonds showed that complex 3 exhibits high catalytic activity toward various C–H functionalizations, which proceed efficiently and regioselectively owing to chelate formation.[15] [16] [17] The chelation-assistance strategy was also found to be applicable to catalytic functionalization of unreactive carbon bonds other than C–H bonds. The ruthenium-catalyzed aromatic C–H arylation of aromatic ketones and esters with organoboronates was extended to the couplings of aryl ethers,[22] anilines,[23] and aryl fluorides[24] via C–O, C–N, and C–F bond cleavage.

3.1

Coupling with Organoboronates via Aromatic C–O Bond Cleavage

In order to achieve efficient oxidative addition of aromatic C–O bonds to low-valent transition metals, it is generally required that the oxygen atom is attached to an electron-withdrawing group such as trifluoromethylsulfonyl or p-tolylsulfonyl. Although it is challenging to cleave aromatic C–O bonds in aryl ethers in a similar manner, the chelation-assistance strategy has enabled the catalytic conversion of aromatic C–O bonds in aryl ethers into C–C bonds using organoboronates, via oxidative addition of aromatic C–O bonds.[22] Prior to our study of the C–O functionalization, there was only one previous example describing catalytic C–C bond formation via aromatic C–O bond cleavage in ethers, in which the catalytic coupling of methoxyarenes with Grignard reagents was achieved using a nickel catalyst.[48]

Treatment of 2′-methoxy-6′-methylacetophenone (66) with phenylboronate 15 in the presence of catalyst 3 and toluene under refluxing conditions gave the ortho phenylation product 16 in excellent yield (Scheme [30]). Table [4] shows representative results from the coupling of 2′-methoxypivalophenone (67) with organoboronates.

Zoom Image
Scheme 30

Table 4 Ruthenium-Catalyzed Coupling of 67 with Organoboronates

Entry

R

R′

Time

Product

Yield

1
2
3
4
5

NMe2
OMe
F
CF3
CH=CH2

20 h
 3 h
30 h
 3 h
 2 h

68a
68b
68c
68d
68e

89%
76%
82%
86%
79%

6

CH=CMe2

 2 h

68f

81%

7

Me

 1 h

68g

83%

8

Bn

20 h

68h

71%

We have proposed that this coupling reaction proceeded via oxidative addition of the aromatic carbon–oxygen bond to a low-valent ruthenium complex.[22] In order to gain an insight into the reaction mechanism, stoichiometric reactions of ruthenium catalyst 3 with aryl ether substrates were examined. The reaction of a pivalophenone bearing a p-tolyloxy group at the ortho position (69) with complex 3 in toluene at reflux temperature for 20 hours led to the formation of aryloxo ruthenium complex 70, presumably via oxidative addition of the C–O bond (Scheme [31]).[22b] Single crystal X-ray diffraction analysis of complex 70 showed that a five-membered ruthenacycle was formed by conversion of the ortho C–O bond in 69 into a C–Ru bond and coordination of the carbonyl group; the CO and aryloxy ligands are located cis and trans to the ruthenated aryl carbon, respectively. This was the first direct observation of the oxidative addition of the C–O bond to a transition metal complex. In 1997, Milstein and co-workers reported the formation of an arylrhodium complex via C–O bond cleavage in aromatic ethers bearing phosphine moieties at the 2- and 6- positions.[49] The C–O bond is considered to be cleaved by oxidative addition, but the corresponding alkoxo ligand was not directly observed in this case.

Zoom Image
Scheme 31

The relative reactivity of C–H and C–O bonds toward the ruthenium complexes was examined by reacting with an acetophenone bearing a p-tolyloxy group at the ortho position (71) (Scheme [32]).[22b] At room temperature, the 1H and 31P NMR spectra indicated that C–H bond cleavage took place to give the hydride complex 72 as the major product. Heating the reaction mixture at 80 °C for three hours resulted in the complete disappearance of 72 and generation of the C–O bond cleavage product 73. These results suggested that the C–H bond cleavage is faster, but thermodynamically the C–O bond cleavage is more favored.

Zoom Image
Scheme 32

Chemoselective C–C bond formation using vinylsilanes and organoboronates was possible via tandem C–H and C–O bond cleavage. When the reaction of 2′-methoxy­acetophenone (44) with vinylsilane 5 and phenylboronate 15 was carried out in refluxing toluene, the tandem C–H alkylation/C–O arylation product 74 was formed exclusively in 93% yield (Scheme [33]). This result indicates that the C–H/olefin coupling proceeds much faster than the C–H arylation in toluene.

Zoom Image
Scheme 33

# 3.2

Coupling with Organoboronates via Aromatic C–N Bond Cleavage

Conversion of amino groups into better leaving groups is necessary for the cleavage of aromatic carbon–nitrogen bonds. Fujiwara and co-workers reported that reactions of anilines with olefins in the presence of stoichiometric amounts of palladium salts and carboxylic acids, performed under air, resulted in the conversion of the amino groups into alkenyl groups.[50] The first direct observation of the cleavage of aromatic C–N bonds in anilines was achieved by Wolczanski and co-workers using a tantalum(III) complex.[51] However, catalytic functionalization of aromatic C–N bonds in anilines had not been achieved until our group reported the catalytic arylation via cleavage of the aromatic C–N bond in anilines.[23a]

The reaction of 2′-dimethylamino-6′-methylacetophenone (75) with 15 provided ortho phenylation product 16 in 84% yield via aromatic C–N bond cleavage (Scheme [34]). This reaction is considered to proceed via oxidative addition of the C–N bond to a low-valent ruthenium species, followed by transmetalation between an amido ruthenium species with an arylboronate.

Zoom Image
Scheme 34

Various o-acylanilines containing NMe2, N(CH2)4, NMe(allyl), NMeAc, NHMe (76ae), and NH2 (77) groups can be used in the cross-coupling with organoboronates. The phenylation product 78 was obtained in good to high yields from o-acylanilines 76 and 77 (Scheme [35]). The substituent on the nitrogen atom does not largely affect the yields of the coupling products, except for that of the substrate with an NMeAc group, which gave a lower yield (51%). In addition, o-aminobenzoates are effective substrates for this coupling.

Zoom Image
Scheme 35

The wide applicability of organoboronates for the C–N arylation has been demonstrated and selected results are listed in Table [5]. Arylboronates with various electron-­donating or electron-withdrawing groups gave high yields of the corresponding products. Similarly high yields were obtained with sterically-demanding arylboronates such as 2-tolyl- and 1-naphthylboronates. The reaction with 4-vinylphenylboronate gave the corresponding product containing a vinyl group in 76% yield. Heteroarylboronates including 2-furyl- and 2-thienylboronates were also applicable to the C–N arylation. In addition, alkenyl- and al­kylboronates could be employed for the incorporation of propenyl, styryl, benzyl, β-phenethyl, and cyclopropyl groups. The reaction was also performed using a trimethylsilylmethylboronate, which is rarely applicable in ­Suzuki–Miyaura cross-coupling reactions, to give the corresponding alkylation product.

Table 5 Ruthenium-Catalyzed Coupling of 76a with Organoboronates

Entry

R

R′

Product

Yield

1
2
3
4

4-NMe2
4-CF3
4-CH=CH2
2-Me

68a
68d
68e
68i

98%
89%
76%
87%

5

2-furyl

79a

81%

6

2-thienyl

79b

65%

7

CH=CH2

79c

82%

8

CH2SiMe3

79d

59%

9

cyclo-C3H5

79e

82%

Stoichiometric reactions of o-acylanilines with ruthenium complexes have been examined.[23b] The reaction of ruthenium complex 3 with o-acetylaniline (80) in toluene at 120 °C for 20 hours gave amido hydrido complex 81 in 56% isolated yield (Scheme [36]). The molecular structure of 81 was confirmed by X-ray crystallography, which showed that this complex possessed a six-membered chelate formed from 80, and the nitrogen atom of the amide was located cis to the hydride. When o-pivaloylaniline (77) was used for the reaction with 3, the corresponding amido hydrido complex 82 was similarly obtained.

Zoom Image
Scheme 36

When the reaction of complex 3 with o-acylaniline 77 was carried out in the presence of vinylsilane 5 at 120 °C, the generation of aryl amido complex 83HH was observed, and after three days, complex 83HH was the only ruthenium species observed by 31P NMR spectroscopy (96% yield by 1H NMR spectroscopy), and was isolated in 56% yield (Scheme [37]). The molecular structure of 83HH was confirmed by X-ray crystallography. Complex 83HH was considered to be formed by chelation-assisted oxidative addition of the C–N bond of 77, followed by exchange of the NH2 ligand with another equivalent of 77.

Zoom Image
Scheme 37

The amido ligand of 83HH is easily exchanged with o-acylanilines. Therefore, aryl amido complexes 83HR can be generated by the reaction of 83HH with o-acylanilines 84 having Me, MeO, F, or CF3 groups at the para position relative to the NH2 group (Scheme [38]). The reaction of 83HMe with phenylboronate 15 gave products 78 and 85 in 97% and 3% yields, respectively (Scheme [39]). The selective formation of 78 indicated that the ortho-ruthenated aryl ligand, originally formed by C–N bond cleavage, was used for the C–C bond formation.

Zoom Image
Scheme 38
Zoom Image
Scheme 39

The relative reactivity of o-acylanilines toward activated ruthenium complex 6 was examined by competition experiments between 77 and para-substituted acylanilines 84.[23c] The results suggested that the order of the relative reactivity appeared to be 84MeO > 84Me > 84F > 77 ~ 84CF3 and that electron-donating substituents such as OMe and Me groups facilitated the cleavage of the C–N bonds. The substituent effects observed on our C–N bond cleavage were opposite to that of the general tendency of oxidative addition of polar aromatic C–X bonds.[52]

Zoom Image
Scheme 40 Possible explanation of the substituent effect on the C−N bond cleavage

This unusual substituent effect can be explained as follows (Scheme [40]): the N–H bond cleavage to form amido hydrido complex 82 and C–N bond cleavage to give aryl amido complex 87 proceed via the same intermediate, 86. The amide ligand of 82 undergoes rapid exchange with another molecule of 84, and complexes 82r1 , 82r2 , 86r1 , and 86r2 are rapidly interconverted. If R1 is more electron-donating than R2, the concentration of 86r1 is expected to be higher than that of 86r2 because 84r1 can bind more strongly to the ruthenium. The higher concentration of 86r1 leads to the faster rate of C–N bond cleavage.

Reductive deamination of o-acylanilines possessing alkyl group(s) on the amine nitrogen atom was achieved by heating with catalyst 3.[23d] The reaction of N-alkylacetylaniline 76a with catalyst 3 gave the corresponding reductive deamination product 41 in 90% yield, supposedly via oxidative addition of the C–N bond, followed by β-hydride elimination and reductive elimination. In the presence of olefin 5, o-acylaniline 76a was converted into silane 88 via reductive deamination/alkylation (Scheme [41]).

Zoom Image
Scheme 41

Deuterium-labeling experiments using an o-acylaniline containing an N(CD3)2 group (77-d 6 ) suggested that β-hydride elimination from the ruthenium amide intermediate occurred and that the C–H bond cleavage and the hydrometalation of the alkene were reversible (Scheme [42]).

Zoom Image
Scheme 42

# 3.3

Coupling with Organoboronates via Aromatic C–F Bond Cleavage

Cross-couplings using haloarenes play very important roles in modern organic chemistry,[53] but the use of aryl fluorides as substrates is still challenging.[54] We found that our chelation-assistance strategy for the catalytic functionalization of unreactive carbon bonds was also applicable to cross-coupling between aromatic ketones bearing fluoro groups at the ortho position(s) and organoboronates using ruthenium catalyst 3.[24] When the reaction of 2′,6′-difluoroacetophenone (89) with phenylboronate 15 in toluene was carried out using catalyst 3 at 130 °C for 24 hours in the presence of vinylsilane 5 and cesium fluoride (CsF) as additives, compound 20 was formed in 77% yield via carbon–fluorine bond cleavage (Scheme [43]). Coupling of ketone 90, which has a hydrogen and a fluorine at the ortho positions, with phenylboronate 15 and vinylsilane 5 afforded the tandem C–F arylation/C–H alkylation product 74 in 79% yield, along with an 8% yield of the diphenylated product 20 (Scheme [44]).

Zoom Image
Scheme 43
Zoom Image
Scheme 44

#
# 4

Introduction of Heteroatoms at C–H Bonds

The catalytic conversion of C–H bonds into carbon–heteroatom bonds has been widely studied, not only because carbon–heteroatom bonds are present in various molecules of biological and industrial interest, but since many of them give different and improved reactivity that can be used for further transformations. The introduction of heteroatoms via catalytic carbon–hydrogen bond cleavage involves straightforward methods that can, in theory, be achieved for diverse heteroatoms. To date, a variety of heteroatoms, including halogens,[21] [55] oxygen,[56] sulfur,[57] nitrogen,[58] phosphorus,[59] boron,[2k] [60] silicon,[2k] [20] [61] [62] and tin[63] can be introduced selectively to aromatic carbon atoms via C–H bond cleavage with transition metal catalysts. In this section, we describe our results on the transition-metal-catalyzed introduction of heteroatoms to aromatic rings.[20] [21] [22] [23] [24]

4.1

Silylation of C–H Bonds

Organosilicon compounds are not just a valuable class of compounds, as they are also widely applicable synthetic intermediates which can be used for the synthesis of alcohols and olefins, and in cross-coupling reactions. Reactions of carbanions with silyl halides and the hydrosilylations of alkenes and alkynes are often employed for the synthesis of organosilanes, but catalytic C–H silylations have also emerged as useful methods. Several silicon reagents have been used for the C–H silylation such as hydrosilanes,[20b] [c] [d] [e] , [61] vinylsilanes,[20a] and disi­lanes.[62]

4.1.1

Silylation of Arene C–H Bonds with Hydrosilanes

Dehydrogenative silylation of C–H bonds with hydrosilanes is usually an endothermic process, but the reaction may become exothermic when combined with the hydrogenation of alkenes. Reaction of 2-phenyloxazole 91 with triethylsilane (Et3SiH) (92) using Ru3(CO)12 (30) as the catalyst proceeded under thermal conditions to give ortho silylation product 93; notably, in the presence of t-butyl­ethylene (a hydrogen acceptor), the yield was improved to 93% (Scheme [45]).[20b] Various electron-donating and electron-withdrawing groups tolerate the C–H silylation. Even when the substrate had two ortho C–H bonds, only the monosilylation product was obtained. Complex 30 showed higher catalytic activity than 3 for this C–H silylation.

Zoom Image
Scheme 45

Our C–H silylation was applied to other aromatic compounds bearing directing groups containing sp2 nitrogen atoms. The reaction of arylaldimine 94 with Et3SiH (92) using norbornene (95) as a hydrogen acceptor gave the monosilylation product 96 in almost quantitative yield (Scheme [46]).[20c] Various sp2 nitrogen-containing heteroarenes such as pyridine, pyrazole, imidazole, triazole, and tetrazole can function as directing groups for the C–H silylation.[20d] Depending on the size of the directing group, mixtures of mono- and disilylation products were obtained.

Zoom Image
Scheme 46

In our catalytic C–C bond formation via aromatic C–H bond cleavage, it is important for the directing groups to be π-conjugated to the aromatic ring (Figure [4a]). In contrast, this C–H silylation with hydrosilanes can take place without the π-conjugation between the directing group and the aromatic ring (Figure [4b]).

Zoom Image
Figure 4 Types of directing groups used in the chelation-assisted C–H bond cleavage

The reaction of N,N-dimethylbenzylamine (97) with Et3SiH (92) in the presence of norbornene (95) and a catalytic amount of 30 provided ortho silylation product 98 in 58% yield (Scheme [47]). Aryl 2-pyridyl ethers, 2-benzylpyridine, 1-phenylpyrazole, 1-methyl-4-phenylimidazole, and 1-methyl-4-phenyl-1,2,3-triazole also reacted with 92 to give the corresponding ortho-silylated products in good yields.

Zoom Image
Scheme 47

# 4.1.2

Silylation of Benzylic C(sp3)–H Bonds with Hydrosilanes

The Ru3(CO)12-catalyzed C–H silylation can be extended to benzylic sp3 C–H bonds. Silylation of 2-(2,6-dimethylphenyl)pyridine (99) with Et3SiH (92) occurred at the methyl groups to give mono- (100) and disilylation (101) products in 30% and 55% yields, respectively (Scheme [48]).[20e] In the reaction with a 2-phenylpyridine bearing a methyl and an ethyl group at the ortho positions, the silylation took place exclusively at the methyl group to give 102 in 53% yield (Scheme [49]). Several other benzylic methyl groups were also silylated in high yields.

Zoom Image
Scheme 48
Zoom Image
Scheme 49

# 4.1.3

C–H Silylation with Vinylsilanes

While studying the ruthenium-catalyzed alkylation of acylthiophenes with vinylsilanes, we found that triorganovinylsilane functioned as a silylating reagent.[20a] Coupling of 3-acetylthiophene (103) and vinylsilane 5 using 3 as the catalyst yielded predominantly the alkylation product 104 (Scheme [50]). However, the use of ruthenium carbonyl catalyst 30 in place of 3 led to the complete suppression of the C–H alkylation and the exclusive formation of α-silylation product 105, which was formed via β-silyl elimination from a β-silylethylruthenium intermediate.[64] Substrates used for this C–H silylation with vinylsilanes are limited to five-membered heteroaromatic compounds having keto, ester, and amide moieties as directing groups. RuHCl(CO)(PPh3)3 was also found to be an effective catalyst for this process.

Zoom Image
Scheme 50

#
# 4.2

Aromatic C–H Halogenation by Palladium-Catalyzed C–H Bond Cleavage and Electrochemical Oxidation

Aryl halides are versatile synthetic intermediates in organic synthesis. There are various classical methods for the synthesis of aryl halides, but recently, transition-metal-catalyzed C–H halogenation has attracted considerable attention as a selective halogenation method.[55] Sanford and co-workers have developed palladium-catalyzed chlorination, bromination, and iodination using halogenating agents such as N-halosuccinimides (NXS, X = Cl, Br, I).[55a] [b] [c] [d] [e] Yu and co-workers reported copper-catalyzed chlorination with tetrachloroethane and palladium-catalyzed iodination with molecular iodine (I2) and iodobenzene diacetate [(PhI(OAc)2].[55i] [j] [k] There have also been reported examples of other catalytic C–H halogenations of aromatic compounds including fluorination.[55l] [m] [n] These catalytic C–H halogenations can introduce halogen atoms regioselectively, but stoichiometric amounts of halogenating agents with high oxidizing ability are required, and the remaining halogenating reagents and the stoichiometric amounts of by-products generated from the halogenating reagents have to be removed after the reaction.

Electrochemical reactions have been used as a method to avoid the use of the stoichiometric amounts of reactive species by generating them in situ from practical reagents. Halogenation of arenes with X+ species generated by electrochemical oxidation has been reported,[65] however, these reactions proceed via electrophilic aromatic substitution mechanisms and it is hard to control the regioselectivity.

We have developed C–H halogenation methods by combination of palladium-catalyzed C–H bond cleavage and electrochemical oxidation of X or X2 into X+.[21] It is not necessary to use stoichiometric amounts of reactive halogenating reagents because they are generated in situ by electrochemical oxidation.

4.2.1

C–H Chlorination and Bromination

The C–H chlorination was performed in an H-type divided cell with an anion-exchange membrane. A DMF solution of the substrate and a catalytic amount of palladium(II) chloride (PdCl2) was added to the anionic chamber, while the cathodic chamber was charged with an aqueous solution of HCl. When the reaction of benzo[h]quinoline (52) was carried out for five hours at 90 °C under constant current electrolysis conditions at 20 mA, the 10-position of 52 was chlorinated regioselectively to afford product 106 in a quantitative yield (Scheme [51]).[21a] No reaction was observed in the absence of either the catalyst or the electric current. In this reaction, the aqueous HCl solution was used as both an electrolyte and a source of chlorine.

Zoom Image
Scheme 51

The C–H chlorination can be applied to a variety of arylpyridines having electron-donating and electron-withdrawing groups (Scheme [52]). When an aqueous solution of hydrogen bromide (HBr) and a catalytic amount of palladium(II) bromide (PdBr2) were used for the reaction, C–H bromination proceeded regioselectively at the ortho position.

Zoom Image
Scheme 52

One of advantages of electrochemical methods is that the rate of generation of reactive species can be easily controlled by adjusting the electric current to an appropriate level. For example, under the electrochemical oxidation conditions using a 20 mA electric current, 2-arylpyridine 107, which has an electron-donating methoxy group, was converted into the desired product 108, along with a small amount of dichloride 109, formed by further electrophilic chlorination (Scheme [53]). However, simple reduction of the electric current to 10 mA led to the exclusive formation of the desired product 108 in 92% yield.

Zoom Image
Scheme 53

# 4.2.2

C–H Iodination and One-Pot C–H Iodination/Suzuki–Miyaura Coupling by ON/OFF Switching of Electric Current

Modification of the reaction conditions was necessary to achieve C–H iodination, and molecular iodine (I2) was found to be an effective iodine source. When the reaction of 2-(2-methylphenyl)pyridine (110) with I2 was performed using palladium(II) acetate [Pd(OAc)2] as the catalyst in MeCN with charging sulfuric acid in the cathodic chamber, the C–H iodination took place at the ortho position of 110 to give iodide 111 in 88% yield (Scheme [54]).

Zoom Image
Scheme 54

When the electric current was turned on, the palladium-catalyzed electrochemical C–H halogenation proceeds via C–H bond cleavage with the palladium(II) species, and the constant provision of the X+ species does not allow generation of the Pd(0) species under the reaction conditions (Scheme [55]). However, when the electricity is turned off, the formation of X+ stops and the reaction system becomes free of strong oxidants. Therefore, a different catalytic reaction proceeding via a Pd(0)/Pd(II) cycle would be achieved. Based on this hypothesis, we have developed a one-pot C–H iodination/Suzuki–Miyaura coupling by ON/OFF switching of the electric current. The electrochemical C–H iodination of 110 was performed, and after the reaction, the generation of I+ was stopped by turning off the electric current. Phenylboronic acid and K2CO3 were added to the mixture, which was then heated for two hours (without supplying any electric current) to afford phenylation product 112 in 84% yield (Scheme [56]). This result demonstrated that the ON/OFF switching of two different catalytic cycles could be accomplished using the same palladium catalyst in a one-pot fashion.[21b]

Zoom Image
Scheme 55 Proposed reaction pathway for the sequential C–H iodination/Suzuki–Miyaura coupling
Zoom Image
Scheme 56

#
#
# 5

Applications of C–H Functionalization in Short Syntheses of Polycyclic Aromatic ­Hydrocarbons (PAHs)

Polycyclic aromatic hydrocarbons (PAHs) have attracted significant attention due to their intriguing electronic properties and potential utility as organic semiconducting materials.[66] In recent years, considerable efforts have been devoted to the development of methods to prepare various derivatives of the desired PAHs, because the introduction of appropriate substituents might improve the electronic properties and/or the solubility. In this context, our group has been making efforts to develop methods for short syntheses of PAHs using our ruthenium-catalyzed C–H functionalization of aromatic ketones. Our strategy for PAH synthesis has several advantages including: (1) ready accessibility to diverse aromatic ketone substrates, (2) the wide variety and broad scope of the ruthenium-­catalyzed C–H functionalization methods, and (3) the versatility of the carbonyl-directing group for further manipulation. In this section, we describe our recent efforts on the short syntheses of PAHs using our C–H functionalization methods.[17c] [25]

5.1

Convenient Synthesis of Multiarylanthracenes

Anthraquinone and its derivatives are convenient precursors of anthracene derivatives. Simple reduction of anthraquinone frameworks gives access to anthracene structures, while substituted anthracenes are obtained by a nucleophilic addition/reduction sequence. However, these methods only allow for the introduction of substituents at the carbonyl carbons. We envisioned that more highly substituted anthracenes could be prepared by application of our ruthenium-catalyzed C–H functionalization to anthraquinone and its derivatives. Four C–H bonds are present at ortho positions relative to the carbonyls in the anthraquinone structure and these are difficult to functionalize by methods other than catalytic C–H functionalization.

First, we developed a short, straightforward method for the synthesis of tetra- and hexaarylanthracenes using the RuH2(CO)(PPh3)3-catalyzed C–H arylation of anthraquinone (21) with arylboronates 22.[25a] As described in Scheme [14, a] variety of aryl groups could be introduced at the four ortho C–H bonds of 21 to give the corresponding products in good to high yields.[25a] Reduction of the carbonyl groups in tetra(p-tolyl)anthraquinone 23a by a modified version of Pascal’s method[67] using aqueous hydrogen iodide (HI) and acetic acid (AcOH) at 140 °C in a sealed tube afforded 1,4,6,9-tetra(p-tolyl)anthracene (113) in 83% yield (Scheme [57]).[68] This result illustrates that the synthesis of a tetraarylanthracene can be completed in two steps.

Zoom Image
Scheme 57

The conversion of 23 into hexaarylanthracenes 115 was examined next.[17] Tetraarylanthraquinones 23a,g were reacted with an aryllithium reagent in THF to give diols 114, which were then converted into hexaarylanthracenes 115 with sodium iodide (NaI) and sodium hypophosphite (NaH2PO2) in AcOH at reflux temperature (Scheme [58]). Single crystal X-ray diffraction analysis of hexaarylanthracene 115a showed that the anthracene framework was strongly twisted, as was observed for decaphenylanthracene in the report of Pascal and co-workers.[69] The end-to-end twist of 115a was 58°, which was slightly smaller than the 63° reported for Pascal’s decaphenylanthracene.

Zoom Image
Scheme 58

# 5.2

Short Synthesis of Alkyl-Substituted Anthracenes and Pentacenes

A similar strategy can be used for the short synthesis of tetrasubstituted pentacenes. The reaction of pentacenequinone (116) with triethylvinylsilane at 115 °C gave tetraalkylated pentacenequinone 117a in 36% isolated yield (Scheme [59]).[25b] The C–H alkylations of 116 with hex-1-ene and 2-methylstyrene at 180 °C gave the corresponding tetraalkylated pentacenequinones 117b,c in 69% and 74% isolated yields, respectively.

Zoom Image
Scheme 59

Reductive aromatization of 117a can be completed essentially in one step using lithium aluminum hydride (LAH) to afford tetraalkylpentacene 118a in 45% yield as a blue oil (Scheme [60]). Tetra(n-hexyl)pentacene 118b was similarly prepared in 39% yield as a blue solid.

Zoom Image
Scheme 60

# 5.3

Short Synthesis of Dibenzo[a,h]anthracenes and Picenes

Dibenzo[a,h]anthracenes[70] and picenes[71] have recently shown some promise as organic field-effect transistor (OFET) materials. As described in Scheme [61], we have developed an efficient route for their syntheses, consisting of the coupling of acetophenone derivatives with 1,4-arenediboronates (step 1), the transformation of the acetyl groups into ethynyl groups (step 2),[72] and the platinum-catalyzed cycloaromatization to give dibenzo[a,h]anthracenes 119 and picenes 120 (step 3).[73]

Zoom Image
Scheme 61

The synthesis of dibenzo[a,h]anthracene derivatives 119 was explored using 2,5-dimethyl-1,4-benzenediboronate 121a. The C–H arylation of acetophenone derivatives with 121a gave teraryl product 122. Treatment of diacetylterphenyls 122 with lithium diisopropylamide (LDA) and diethyl chlorophosphate [(EtO)2P(O)Cl] provided diethynylterphenyls, and subsequent platinum-catalyzed cycloaromatization afforded the dibenzo[a,h]anthracene derivatives 119 (Scheme [62]).[25a] Measurement of the OFET properties of bis(trimethylsilyl)dibenzo[a,h]anthracene 119d revealed that the OFET fabricated with a bottom-contact configuration showed moderate hole mobility (μ FET = 3.0 × 10–4 cm2 V–1 s–1). This OFET property of 119d is insufficient for it to be used as a device, but the result suggests that C–H arylation is appropriate to be used as a tool for the synthesis of polycyclic aromatic hydrocarbons in a short three-step sequence.

Zoom Image
Scheme 62

The synthesis of picene derivative 120a was possible by changing the arenediboronate from 121a to 2,3-dimethyl-1,4-benzenediboronate 121b. The C–H arylation of acetophenone 1 with 121b afforded teraryl product 123 in 14% yield. Transformation of the acetyl groups to ethynyl groups led to the corresponding diethynylteraryl compound, which was then converted into picene 120 by ­cycloaromatization (Scheme [63]). When 2,5-thiophenediboronate was used for the C–H arylation, sulfur-containing derivatives were obtained (Figure [5]).

Zoom Image
Scheme 63
Zoom Image
Figure 5 Substituted 1,2,7,8-dibenzo-9-thiafluorenes

#
# 6

Conclusions

Over the past 20 years, considerable progress has been made in the field of transition-metal-catalyzed functionalization of unreactive carbon bonds such as C–H, C–O, C–N, and C–F bonds. The C–H bond functionalization is often called C–H bond activation. We have shown that the chelation-assistance strategy is applicable to a variety of reactions which convert carbon–hydrogen and other un­reactive carbon bonds into carbon–carbon and carbon–­heteroatom bonds. We have also demonstrated that the reactions can be used for efficient syntheses of substituted polycyclic aromatic hydrocarbons.

Catalytic transformations of unreactive carbon bonds are now recognized as atom- and/or step-economical methods and are used widely in organic synthesis. The chelation-assistance strategy is recognized as being a highly reliable method for cleaving and making new bonds with high selectivity and efficiency. These reactions have been applied to efficient syntheses of organic electronic materials, natural products, and bioactive compounds. Further developments on the selective catalytic functionalization of unreactive carbon bonds and their creative applications will certainly lead to even more significant advances in synthetic organic chemistry.


#
#

Acknowledgment

We started the research described in this account in 1992 while F.K. was at Osaka University. Since then, many people including co-workers at Keio University and Osaka University have contributed to this work, which would never have existed without their excellent talents, hard work and fruitful discussions. We wish to express our deep appreciation to our co-workers for their contributions. We are also grateful for the financial support, in part, by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan



Zoom Image
Zoom Image
Zoom Image
Zoom Image
Scheme 1
Zoom Image
Scheme 2
Zoom Image
Scheme 3
Zoom Image
Scheme 4
Zoom Image
Scheme 5
Zoom Image
Scheme 6
Zoom Image
Scheme 7
Zoom Image
Scheme 8
Zoom Image
Scheme 9
Zoom Image
Scheme 10
Zoom Image
Figure 1 Structure of ortho-ruthenated acetophenone complex 8
Zoom Image
Scheme 11
Zoom Image
Scheme 12
Zoom Image
Scheme 13
Zoom Image
Scheme 14 Proposed catalytic cycle for the C–H arylation of acetophenone with arylboronates
Zoom Image
Figure 2 The generality of the C–H arylation of aromatic ketones with arylboronates
Zoom Image
Scheme 15 Proposed mechanism for the C–H arylation in the presence of styrene (18)
Zoom Image
Scheme 16
Zoom Image
Scheme 17
Zoom Image
Scheme 18
Zoom Image
Scheme 19
Zoom Image
Scheme 20
Zoom Image
Scheme 21 Proposed mechanism for the C–H alkenylation with alkenyl acetate 35
Zoom Image
Scheme 22
Zoom Image
Scheme 23
Zoom Image
Scheme 24
Zoom Image
Scheme 25
Zoom Image
Scheme 26
Zoom Image
Figure 3 Products formed by the C–H alkoxycarbonylation of arylpyridines
Zoom Image
Scheme 27
Zoom Image
Scheme 28
Zoom Image
Scheme 29
Zoom Image
Scheme 30
Zoom Image
Scheme 31
Zoom Image
Scheme 32
Zoom Image
Scheme 33
Zoom Image
Scheme 34
Zoom Image
Scheme 35
Zoom Image
Scheme 36
Zoom Image
Scheme 37
Zoom Image
Scheme 38
Zoom Image
Scheme 39
Zoom Image
Scheme 40 Possible explanation of the substituent effect on the C−N bond cleavage
Zoom Image
Scheme 41
Zoom Image
Scheme 42
Zoom Image
Scheme 43
Zoom Image
Scheme 44
Zoom Image
Scheme 45
Zoom Image
Scheme 46
Zoom Image
Figure 4 Types of directing groups used in the chelation-assisted C–H bond cleavage
Zoom Image
Scheme 47
Zoom Image
Scheme 48
Zoom Image
Scheme 49
Zoom Image
Scheme 50
Zoom Image
Scheme 51
Zoom Image
Scheme 52
Zoom Image
Scheme 53
Zoom Image
Scheme 54
Zoom Image
Scheme 55 Proposed reaction pathway for the sequential C–H iodination/Suzuki–Miyaura coupling
Zoom Image
Scheme 56
Zoom Image
Scheme 57
Zoom Image
Scheme 58
Zoom Image
Scheme 59
Zoom Image
Scheme 60
Zoom Image
Scheme 61
Zoom Image
Scheme 62
Zoom Image
Scheme 63
Zoom Image
Figure 5 Substituted 1,2,7,8-dibenzo-9-thiafluorenes