Download PDF
Synlett 2016; 27(12): 1765-1774
DOI: 10.1055/s-0035-1561617
account
© Georg Thieme Verlag Stuttgart · New York

Aromatic Metamorphosis of Dibenzothiophenes

Hideki Yorimitsu*
a   Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
b   Institute for Molecular Science, Okazaki, Aichi 444-8787, Japan
,
Dhananjayan Vasu
a   Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
,
M. Bhanuchandra
a   Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
,
Kei Murakami
a   Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
c   The Hakubi Center for Advanced Research, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan   Email: yori@kuchem.kyoto-u.ac.jp
,
Atsuhiro Osuka
a   Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
› Author Affiliations
Further Information

Publication History

Received: 12 February 2016

Accepted: 11 March 2016

Publication Date:
18 May 2016 (online)

 
 PDF Download Permissions and Reprints All articles of this category


Abstract

In general, aromatic cores are stable owing to their resonance energies. Different from facile peripheral modifications of aromatic cores, transforming an aromatic core into a different skeleton is ambitious and has attracted only little attention as a general synthetic method. This personal account shows our journey to inventing transformations of dibenzothiophenes into triphenylenes, carbazoles, and spirocyclic diarylfluorenes and to establishing ‘aromatic metamorphosis’ as a useful and game-changing strategy in organic synthesis.

1 Introduction

2 Aromatic Metamorphosis

3 From Dibenzothiophenes to Triphenylenes

4 From Dibenzothiophenes to Carbazoles

5 From Dibenzothiophenes to Spirocyclic Tetraarylmethanes

6 Conclusion


#

Key words

dibenzothiophenes - palladium - arylations - nucleophilic ­aromatic substitution - umpolung

Biographical Sketches

Zoom Image

Hideki Yorimitsu was born in Kochi, Japan, in 1975. He obtained his Ph.D. in 2002 from Kyoto University under the supervision of Professor Koichiro Oshima. He then served as a JSPS (Japan Society for the Promotion of Science) postdoctoral fellow with Professor Eiichi Nakamura at the University of Tokyo. He became an assistant professor in 2003 and an associate professor in 2008 at the Graduate School of Engineering, Kyoto University. He moved to the Graduate School of Science at the same university in 2009 and became a professor in 2015. He is also a visiting professor at the Institute for Molecular Science as well as a project leader of ACT-C, Japan Science and Technology Agency. His research program focuses on the development of new organic transformations for creating new molecules, phenomena, and concepts. He received the Chemical Society of Japan Award for Young Chemists in 2009 and Young Scientists’ Prize from MEXT (Ministry of Education, Culture, Sports, Science, and Technology) in 2011, and will receive the Mukaiyama Award in autumn this year.

Zoom Image

Dhananjayan Vasu was born in Tamil Nadu, India, in 1983. He obtained his B.S. and M.S. from the University of Madras, India, in 2003 and 2005, respectively. He received his Ph.D. in 2011 from National ­Tsing Hua University, Taiwan, under the supervision of Professor Rai-Shung Liu. He worked as a postdoctoral fellow in the same laboratory until 2013 in the field of gold- and platinum-catalyzed new organic transformations. In the same year, he moved to Kyoto University, Japan, as a JSPS postdoctoral fellow under the guidance of Professor Hideki Yorimitsu, where he spent more than two years researching organosulfur-based organic synthesis. After completion of his JSPS fellowship, he moved to the Korea Advanced Institute of Science and Technology (KAIST), Institute of Basic Science (IBS), South Korea, working with Professor Sungwoo Hong. Presently, he is working on transition-metal-catalyzed C–H functionalizations.

Zoom Image

M. Bhanuchandra was born in Andhra Pradesh, India, in 1984. He obtained his Ph.D. in 2014 from the University of Hyderabad under the guidance of Professor Akhila Kumar Sahoo. He was a postdoctoral fellow at Kyoto University from 2014, and he has been an assistant professor at the Central University of Rajasthan since April 2016. He is now working on several projects, such as the development of C–N, C–C, and C–B bond formation reactions using organosulfur compounds.

Zoom Image

Kei Murakami was born in Osaka, Japan, in 1985. He received his Ph.D. degree from Kyoto University in 2012. After his JSPS postdoctoral fellowship at Nagoya University, he started his academic career at the Hakubi Center for Advanced Research/Department of Chemistry, Graduate School of Science, Kyoto University. He moved to the Graduate School of Science, Nagoya University, as an assistant professor. His research focuses on the development of new reactions for making novel functional/bioactive molecules.

Zoom Image

Atsuhiro Osuka was born in Gamagori, Aichi, Japan, in 1954. He received his Ph.D. degree from Kyoto University in 1982. In 1979, he started his academic career in the Department of Chemistry, Ehime University, as an assistant professor. In 1984, he moved to the Department of Chemistry, Kyoto University, where he became a professor in 1996. He was awarded the Chemical Society of Japan Award for Young Chemists in 1988, the Japanese Photochemistry Association Award in 1999, the Nozoe Memorial Lectureship Award in 2009, and the Chemical Society of Japan Award in 2010. He was selected as a project leader of Core Research for Evolutional Science and Technology (CREST) of JST (Japan Science and Technology Agency) during 2001–2006. His research interests cover many aspects of synthetic approaches toward artificial photosynthesis and the development of porphyrin-related compounds with novel structures, electronic systems, and functions.

1

Introduction

One of the authors, H.Y., was interested in the development of novel organic reactions when he was a Ph.D. student and an assistant professor at the Graduate School of Engineering, Kyoto University. Koichiro Oshima was his supervisor and colleague in those days, and taught him how to enjoy research to pursue novel reactions. Novel reactions have always been stimulating to whomever discovers them. However, it is generally no easy task to discover such stimulating reactions. This is also the perspective for H.Y., and he himself has been undoubtedly laboring as well as enjoying such discoveries.

About six years ago, H.Y. moved from the Graduate School of Engineering to the Graduate School of Science at the same university. He belonged to the group of A.O., another author. The group has been focusing on the synthesis of novel porphyrinoids with intriguing properties. Although a number of various porphyrinoids have been synthesized there, the variety of the reactions used is not wide: mostly acid-catalyzed condensation of pyrroles with aromatic aldehydes, cross-coupling, Ir-catalyzed C–H borylation, and electrophilic aromatic substitution reactions. Their chemistry heavily relies on simple, robust, and classical yet reliable reactions. Applying or inventing new reactions is usually unnecessary or, at least, not the highest priority, as is more or less similar to most laboratories aiming at syntheses of bioactive complex molecules. They seldom used bulky phosphines or N-heterocyclic carbene ligands for transition-metal-catalyzed reactions six years ago. This was a big cultural difference for H.Y., who had always primarily pursued the novelty of reactions.

One of H.Y.’s research projects has been the utilization of organosulfur compounds in organic synthesis taking advantage of the intriguing nature of sulfur.[4] His group discovered several Pummerer-type reactions using ketene dithioacetal monoxides as key starting materials.[5] During the course of this project, he was often requested to develop powerful cross-coupling reactions to convert C–S bonds efficiently in a catalytic manner. He indeed found such reactions: the cross-coupling of aryl sulfides with several types of nucleophiles, such as arylzinc reagents,[5f] [6] aryl- and alkylamines,[7] and ketimines.[8] Such cross-coupling reactions are useful especially when combined with sulfur-specific reactions, such as extended Pummerer reactions and nucleophilic aromatic substitution (SNAr) reactions. His subgroup was delighted to find these powerful combinations and has been trying to further pursue different chemistry where organosulfur compounds play key roles that organic halides or pseudohalides cannot.

Soon after H.Y. joined the porphyrin group, he saw an interesting scheme in a handout for a research progress report (Scheme [1]). Because acid-mediated synthesis of subporphyrins[9] did not proceed with aliphatic aldehydes, thiophene-2-carbaldehyde was used as a pentanal equivalent through Ni-mediated desulfurization.[10] [11] In general, breaking a thiophene ring into a saturated alkane can be useless in most cases from the viewpoint of organic synthesis, while desulfurization is an important industrial process in oil refinery.[12] The chemical equation in the handout inspired H.Y. to develop transformations of aromatic thiophenes into compounds bearing no original thiophene skeleton. The chemistry described in this personal account originated from this unexpected encounter with the reaction shown in Scheme [1].

Zoom Image
Scheme 1 Nickel-mediated desulfurization of a meso-tri-2-thienylsubporphyrin

# 2

Aromatic Metamorphosis

Owing to its high resonance energy, an aromatic ring is inherently difficult to cleave. Converting an aromatic ring into another one is even more challenging. However, there are many examples where such transformations have been achieved. As a well-known example, the Diels–Alder reactions of five-membered heteroarenes and six-membered azines with alkynes afford the corresponding bicyclic products that then undergo rearomatization with extrusion of their heteroatomic units.[13] [14] Recently, triazoles were found to undergo very interesting catalytic denitrogenative transannulation with alkynes or nitriles to yield different aromatic skeletons.[15]

In the group of A.O., many porphyrinoids that can change their aromaticity have represented a central subject of research. Among them, expanded porphyrins, which are conveniently defined as porphyrinoids bearing five or more pyrrole subunits, are attractive because their electronically as well as conformationally flexible natures allow the alteration of their aromaticity upon oxidation/reduction, protonation/deprotonative metalation, changing solvents, and so on.[16] As a notable example, an octaphyrin experienced skeletal rearrangement upon treatment with Ni(acac)2 and sodium acetate to afford a porphyrin dimer (Scheme [2]).[17] Although the reaction mechanism is not clear, this unusual rearrangement of the aromatic octaphyrin apparently includes the drastic change of the initial aromatic skeleton and is thus very stimulating.

Zoom Image
Scheme 2 Metamorphosis of an octaphyrin into a porphyrin dimer

As another famous example, Murata and Komatsu developed the ‘molecular surgery’ of fullerene to synthesize H2@C60 and H2O@C60, and so on.[18] Although the molecular surgery required many steps to implant H2 and H2O and C60 is a special aromatic molecule, this was a landmark that showcased the difficulty as well as the importance of converting one aromatic ring into another. Hence, there still remains ample room to invent conversions of ubiquitous aromatic rings into different ones.

We all know a caterpillar becomes a butterfly through a pupa. Metamorphosis of insects or animals is definitely a miracle in nature. We have felt the similarity to this in transforming a stable aromatic skeleton into a different one and we have named this ‘aromatic metamorphosis’.

Motivated by the encounter with Ni-mediated desulfurization of thiophenes in Scheme [1] as well as by our research interest in organosulfur compounds, we envisioned manipulating thiophenes and their π-extended analogues with the aid of transition-metal catalysis for aromatic metamorphosis. To begin with, we focused on dibenzothiophene as a starting point. We initially tried to convert dibenzothiophene into a different aromatic compound in one step. Because it was difficult, we then envisioned a stepwise transformation.


# 3

From Dibenzothiophenes to Triphenylenes

Triphenylene was set as a target skeleton because it has been finding various applications in organic electronics and its simple, symmetrical, planar structure makes it difficult to synthesize multisubstituted triphenylenes in a tailor-made fashion.[19]

Our synthetic plan for the aromatic metamorphosis of dibenzothiophenes to triphenylenes is shown in Scheme [3].[20] Dibenzothiophenes were expected to resist catalytic transformations owing to their aromaticity and serious catalyst poisoning, especially through oxidative addition to a transition metal. We thus envisioned the alkylation of dibenzothiophenes 1 to give their sulfonium salts 2 as the first step, diminishing their aromaticity as well as the Lewis basicity of the sulfur atom.

According to the famous cross-coupling of arene sulfonium salts by Liebeskind,[21] compounds 2 were expected to participate in smooth cross-coupling with an arylmetal reagent. The arylative ring-opening products 3 would be converted into their cyclic sulfonium salts 4. Because salts 4 should show high reactivity and low poisonous character toward transition-metal catalysts, intramolecular C–H aryl­ation of 4 should proceed smoothly to finalize the aromatic metamorphosis, although there had been no reports on catalytic C–H arylation with an aromatic sulfur compound as an electrophilic partner.[6b] [22]

Zoom Image
Scheme 3 Design of the aromatic metamorphosis of dibenzothiophenes to triphenylenes

The first alkylation step was facile, and AgBF4-mediated SN2 reactions of dibenzothiophenes 1 with a 1-chloro-4-halobutane afforded the corresponding salts 2. We then investigated the catalytic ring opening of 2 with arylmetal reagents. However, we soon encountered difficulty. Simply applying the procedure of Liebeskind[21] to the arylative ring opening of 2 resulted in failure. The reaction suffered from the very rapid dealkylative decomposition of 2 to recover 1 of extremely high leaving group ability.

As exemplified in Scheme [4], the model reaction of S-methyldibenzothiophene sulfonium salt with phenylboronic acid afforded 3a-Me in only 44% yield under our best optimized Suzuki–Miyaura conditions. Base-free Migita–Kosugi–Stille conditions were totally ineffective, giving rise to direct dealkylation of 2a-Me with tributylphenylstannane.

After further screening of arylmetal reagents for a half year, we were delighted to find that sodium tetraphenylborate[23] was not so directly reactive with 2, but displayed exquisite reactivity transferring its phenyl group to palladium without any additional activators. Notably, the reaction was highly efficient and went to completion in 30 minutes at 35 °C.

Zoom Image
Scheme 4 Optimization of the arylating agent for ring opening

Varieties of tetraarylborates and dibenzothiophene-related sulfonium salts 2 underwent arylative ring opening to give the corresponding products 3 (Table [1] and Figure [1]), allowing the synthesis of even highly crowded 3l and 3n (Figure [1]). The C–S bond cleavage took place at less-hindered positions to form 3o exclusively and 3p–r with modest regio­selectivities.

Table 1 Arylative Ring Opening of Dibenzothiophene Sulfonium Salt 2a with Various Sodium Tetraarylborates

R

3

Yield (%)

R

3

Yield (%)

H

3a

81

4-CHOa

3f

88

4-Me

3b

80

3-Me

3g

79

4-OMe

3c

83

3-OMe

3h

89

4-F

3d

88

3,4-OCH2O

3i

84

4-Cl

3e

77

(2-naphthyl)b

3j

73

a With NaB[C6H4-4-CH(OMe)2]4; 3f was formed after hydrolysis.

b C6H4R in 3 = 2-naphthyl; with NaB(2-C10H7)4.

Zoom Image
Figure 1 Products of the arylative ring opening [R = (CH2)4Cl]

Treatment of a series of arylated compounds 3 with AgSbF6 led to SN2 cyclization to give the corresponding teraryl-containing sulfonium salts 4. We were hence ready for the last intramolecular C–H arylation, which would be invented especially to this end.

To our delight, a Pd/SPhos [2-(dicyclohexylphosphino)-2′,6′-dimethoxybiphenyl] catalyst was found to be uniquely effective at generating triphenylenes 5 (Table [2] and Figure [2]; numbering of substrates 4 relates to the corresponding structures 3 shown in Table [1] and Figure [1]). Other bulky ligands, including RuPhos and XPhos, as well as simple ligands such as triphenylphosphine were inactive for this reaction.

Several comments are worth noting: (1) The final cyclization reactions were generally efficient, and the four-step overall yields were satisfactory. (2) The cyclization reactions of 3-substituted substrates 4g–j resulted in the preferable formation of the more congested triphenylenes, although the selectivity varied from poor to exclusive. (3) Steric repulsions around the sulfur atoms in 4l, 4n, and 4o had virtually no influence on the efficient formation of 5l, 5n, and 5o, respectively. In contrast, the formation of 5m was hampered probably by steric repulsion in the binaphthyl unit. (4) The reactions of isomeric mixtures of 4p/4p′ and 4q/4q′ led to convergent formations of 5j and 5j′ as the sole products, respectively. In contrast, the reaction of 4j yielded a mixture of 5j/5j′. Thus, our aromatic metamorphosis should be well designed to provide useful synthetic routes to multisubstituted triphenylenes. (5) As we initially designed it, the reaction mechanism of the ring closure starts with oxidative addition of the C–S+(CH2)4 bond to form the corresponding cationic arylpalladium intermediate, which then engages in electrophilic aromatic substitution for C–H palladation.

Table 2 Palladium-Catalyzed Intramolecular Cyclization of Sulfonium Salts 4

R

5

Yield (%) from 4

Overall yield (%) from 1

H

5a

92

70

Me

5b

89

66

OMea

5c

98

73

F

5d

98

70

Cl

5e

84

54

CHO

5f

86

52

a 6 mol% Pd2dba3, 12 mol% SPhos.

Zoom Image
Figure 2 Triphenylenes synthesized

In this aromatic metamorphosis, dibenzothiophene sulfonium salts such as 2 are the key intermediates. Besides starting from dibenzothiophenes, we conceived utilizing the known electrophilic cyclization of biaryl-2-yl sulfoxides 6 by means of triflic acid to synthesize sulfonium salts.[24] This alternative route should be superior in terms of the diversity of accessible sulfonium salts. However, attempted cyclizations of precursors 6, prepared by cross-coupling biaryl synthesis, with triflic acid resulted in no conversions.

We thus decided to use a stronger activator, triflic anhydride (Tf2O). It was indeed successful, and 6 underwent cyclization to provide multisubstituted dibenzothiophene sulfonium triflates 2j–l (Scheme [5]). Sulfonium triflates 2b–d were converted into multisubstituted triphenylenes 5s–u in satisfactory overall yields. Although this route does not represent aromatic metamorphosis, it opens up a new way to assemble three benzene rings into a triphenylene.

Zoom Image
Scheme 5 Triflic anhydride mediated cyclization of sulfoxides followed by transformation into multisubstituted triphenylenes

# 4

From Dibenzothiophenes to Carbazoles

Motivated by the success of the aromatic metamorphosis to triphenylenes, we contemplated possibilities of similar transformations to give different aromatic molecules. Carbazoles are an important class of heteroarenes in medicinal chemistry as well as materials science. Replacing the sulfur atom of dibenzothiophene with nitrogen is very simple with a molecular model, but can be a challenge in flasks through aromatic metamorphosis.

Because primary amines are naturally basic, we could not use sulfonium salts as synthetic intermediates. Instead, we conceived the corresponding sulfones 7 as appropriate starting materials. As shown in Scheme [6], two SNAr reactions of dibenzothiophene dioxides 7 with primary amines were designed to synthesize carbazoles.[25] The initial oxidation would be facile. The subsequent intermolecular SNAr reaction was likely to occur, although similar SNAr reactions of aryl sulfones with amines were reported several decades ago to proceed only under harsh conditions.[26] The most difficult step would be the final intramolecular SNAr reaction since the reaction should formally expel elusive dianionic SO2 2– of high energy.[27]

Zoom Image
Scheme 6 Design of the aromatic metamorphosis of dibenzothiophenes to carbazoles

Despite the predicted difficulty, we were optimistic about tackling this chemistry according to Scheme [6]. Sulfones 7 were synthesized first through oxidation by H2O2. We were lucky to find at an early stage that the expected reactions of 7 with anilines proceeded at 80 °C in the presence of potassium hexamethyldisilazide (KHMDS) as a uniquely effective base (Scheme [7]).

Zoom Image
Scheme 7 Synthesis of carbazoles via aromatic metamorphosis [a1:3 mixture of 9g and the desilylated analogue of 9g; b6 equiv KHMDS and 4 equiv dibenzothiophene dioxide (7a) were added to 1,4-diaminobenzene, 100 °C, 68 h]
Zoom Image
Scheme 8 Regiocontrol through tuning the oxidation state of sulfur

The reactions proceeded efficiently with electron-rich or -neutral anilines to yield carbazoles 9a–h in good to high yields. The vinyl moiety in 9f was compatible, while the triisopropylsilyl protection of 9g partly survived. Bulky mesitylamine participated to yield 9h. A strongly electron-withdrawing cyano group decreased the nucleophilicity of the amino group dramatically to lead to no formation of 9i. In addition, halo groups did not survive under the strongly basic conditions. 1,2-Diaminobenzene underwent twofold carbazole formation to give 9j. π-Extended carbazoles 9k–m were obtained efficiently.

Through oxidation of dibenzothiophenes to give the corresponding sulfones, the electronic nature of the fused dibenzo units dramatically varies, representing an umpolung of an aromatic ring. While electron-rich dibenzothiophene (1a) undergoes electrophilic bromination at the 2,8-positions, dibenzothiophene dioxide (7a) is electron deficient and delivers the 3,7-dibromo product (Scheme [8]). We thus achieved regioselective syntheses of both 2,8- and 3,7-diphenylcarbazoles 9n and 9o, respectively.

Oxidation of benzothiophene into its dioxide does not only result in loss of aromaticity in the thiophene unit, but also modifies the C-2–C-3 bond to be an electron-deficient double bond. Taking this umpolung of the C-2–C-3 bond into consideration, we paid attention to Mohanakrishnan’s Diels–Alder reaction of benzothiophene dioxides with isobenzofurans, e.g. reaction of 10a and 11a to give 12a (Scheme [9]).[28] This reaction interestingly resulted in annulation of a naphthalene unit after acidic workup to synthesize π-extended benzonaphthothiophenes from readily available smaller π-systems. The aromatic metamorphosis of 12a resulted in the formation of benzocarbazole 13a easily. This three-step two-pot sequence was applicable to the synthesis of the more π-extended dibenzoindolo[3,2-b]carbazole 13b starting from 10b in very short steps.

Zoom Image
Scheme 9 Synthesis of π-extended carbazoles via sequential Diels–Alder cycloaddition/aromatization/aromatic metamorphosis

To elucidate the reaction mechanism shown in Scheme [6], we tried to isolate intermediate 8 or its equivalent. Indeed, the reaction of 7a with p-toluidine proceeded even at room temperature to afford zwitterion 14 upon aqueous workup (Scheme [10]). The subsequent intramolecular reaction of 14 with KHMDS required a higher temperature. Although the exact species that departs from 14 is unclear at this moment, the necessity of the high temperature indicates the reluctance of the leaving species.

Zoom Image
Scheme 10 Isolation of zwitterion 14
Zoom Image
Scheme 11 Synthesis of spirocyclic 9,9-diarylfluorenes [a3.5 equiv KHMDS; b2 equiv KHMDS, 15b was formed as a byproduct in 27% yield; cAt 120 °C; dUnder microwave conditions at 150 °C for 12 h; e4 equiv 7a, 6 equiv KHMDS at 100 °C for 72 h; f2.0 equiv KN(SiMe3)2 for 12 h]

# 5

From Dibenzothiophenes to Spirocyclic Tetraarylmethanes

Because the carbazole synthesis was successful, we sought another nucleophile. A promising nucleophile should have two reasonably acidic protons at the nucleo­philic center and yet should be reactive enough to undergo inter- and intramolecular SNAr reactions. After screening many nucleophiles, including a malonate ester, we found cyclic diarylmethanes such as xanthene to be suitably reactive under similar conditions to those shown in Scheme [7] (Scheme [11]).[29] The products were spirocyclic 9,9-diarylfluorenes, which have been attracting increasing attention in organic electronics.[30] In contrast to the conventional Friedel–Crafts-based synthetic strategies,[31] our protocol employs basic conditions.

As listed in Scheme [11, a] variety of spirocyclic 9,9-diarylfluorenes were readily available although base-sensitive functional groups were incompatible. The chloro groups in 15g and 15n partly survived and resisted dechlorination. Unprotected 2-hydroxythioxanthene was converted smoothly with the aid of an additional equivalent of KHMDS to yield 15f. The reactions of acyclic diphenylmethane and fluorene afforded 15i and 15j, respectively, in only moderate yields even at higher temperatures. The reaction of 9,10-dihydroanthracene with an excess amount of 7a led to a quadruple C–C bond-forming event to furnish 15k in 94% yield. Notably, the formations of 15o and 15q were achieved naturally in a regioselective fashion, while the conventional Friedel–Crafts-based syntheses of these molecules should result in regioisomeric mixtures as exemplified in Scheme [12]. The synthesis of 15s (Scheme [11]) was achieved through a similar strategy to that shown in Scheme [9].

Zoom Image
Scheme 12 Friedel–Crafts reaction that results in the formation of ­regioisomers

To feature the advantage of our protocol further, two sequences of transformations were developed.

Benzothieno-fused xanthene 16 reacted with 7a to furnish spirocyclic compound 15t (Scheme [13]). Subsequent oxidation of 15t to give sulfone 17 followed by another SNAr reaction with xanthene gave deviously bispirocyclic compound 15u.

Zoom Image
Scheme 13 Stepwise synthesis of deviously bispirocyclic compound 15u

Moreover, linearly bispirocyclic compounds 15v and 15w were obtained as outlined in Scheme [14]. Oxidation of 15b (see Scheme [11]) afforded sulfone 18. Although six-membered cyclic sulfone 18 was less reactive than five-membered sulfone 7, the desired bispirocyclic compounds 15v and 15w were formed in moderate yields.

Our protocol provided unique pathways to such unusual multispirocyclic compounds, which are otherwise difficult to synthesize.

Zoom Image
Scheme 14 Stepwise synthesis of linearly bispirocyclic compounds 15v and 15w

# 6

Conclusion

Three transformations of dibenzothiophenes into non-thiophene skeletons have been devised. The first aromatic metamorphosis into triphenylenes was accomplished by discovering new palladium-catalyzed arylative ring opening and intramolecular C–H arylation. A variety of substituted triphenylenes have become readily accessible in a tailor-made fashion in satisfactory overall yields. On a negative note, the protocol requires multiple steps and is thus more or less tedious. However, we like this tedious metamorphosis because we could find several new pieces of chemistry through this challenging work. Necessity is indeed the mother of invention. Naturally, we are exploring a more concise route.

The second synthesis to give carbazoles and the final synthesis to form tetraarylmethanes feature transition-metal-free protocols, operational simplicity, and high overall efficiency. One can reach targeted π-rich molecules of synthetic difficulty, especially when these metamorphoses are combined with thiaarene dioxide specific reactions. This strategy consists of oxidation and SNAr reactions and is thus very simple. We believe that it will be a new tool to cut the Gordian knot of synthesizing unprecedented aromatic skeletons.

Considering the central position that aromatic cores occupy in science and technology, we will continue to develop aromatic metamorphosis to be a revolutionary strategy in organic synthesis.


#
#

Acknowledgment

This work was supported by ACT-C, JST, and by Grants-in-Aid for Scientific Research from MEXT (No. 25107002 ‘Science of Atomic Layers’) and from JSPS {Nos. 24685007 [Young Scientists (A)] and 26620081 (Exploratory Research)}. H.Y. acknowledges the Japan Association for Chemical Innovation, the Kansai Research Foundation, the Asahi Glass Foundation, the Naito Foundation, and the Tokuyama Science Foundation for financial support. K.M. and D.V. acknowledge the JSPS Postdoctoral Fellowship program.



   Permissions and Reprints All articles of this category
Zoom Image
Zoom Image
Zoom Image
Zoom Image
Zoom Image
Zoom Image
Scheme 1 Nickel-mediated desulfurization of a meso-tri-2-thienylsubporphyrin
Zoom Image
Scheme 2 Metamorphosis of an octaphyrin into a porphyrin dimer
Zoom Image
Scheme 3 Design of the aromatic metamorphosis of dibenzothiophenes to triphenylenes
Zoom Image
Scheme 4 Optimization of the arylating agent for ring opening
Zoom Image
Figure 1 Products of the arylative ring opening [R = (CH2)4Cl]
Zoom Image
Figure 2 Triphenylenes synthesized
Zoom Image
Scheme 5 Triflic anhydride mediated cyclization of sulfoxides followed by transformation into multisubstituted triphenylenes
Zoom Image
Scheme 6 Design of the aromatic metamorphosis of dibenzothiophenes to carbazoles
Zoom Image
Scheme 7 Synthesis of carbazoles via aromatic metamorphosis [a1:3 mixture of 9g and the desilylated analogue of 9g; b6 equiv KHMDS and 4 equiv dibenzothiophene dioxide (7a) were added to 1,4-diaminobenzene, 100 °C, 68 h]
Zoom Image
Scheme 8 Regiocontrol through tuning the oxidation state of sulfur
Zoom Image
Scheme 9 Synthesis of π-extended carbazoles via sequential Diels–Alder cycloaddition/aromatization/aromatic metamorphosis
Zoom Image
Scheme 10 Isolation of zwitterion 14
Zoom Image
Scheme 11 Synthesis of spirocyclic 9,9-diarylfluorenes [a3.5 equiv KHMDS; b2 equiv KHMDS, 15b was formed as a byproduct in 27% yield; cAt 120 °C; dUnder microwave conditions at 150 °C for 12 h; e4 equiv 7a, 6 equiv KHMDS at 100 °C for 72 h; f2.0 equiv KN(SiMe3)2 for 12 h]
Zoom Image
Scheme 12 Friedel–Crafts reaction that results in the formation of ­regioisomers
Zoom Image
Scheme 13 Stepwise synthesis of deviously bispirocyclic compound 15u
Zoom Image
Scheme 14 Stepwise synthesis of linearly bispirocyclic compounds 15v and 15w