A combined experimental and computational study on the cycloisomerization of 2-ethynylbiaryls catalyzed by dicationic arene ruthenium complexes

Article abstract of DOI:10.1002/chem.201500248

Ruthenium-catalyzed cycloisomerization of 2-ethynylbiaryls was investigated to identify an optimal ruthenium catalyst system. A combination of [η⁶-(p-cymene)RuCl₂(PR₃)] and two equivalents of AgPF₆ effectively converted 2-ethynylbiphenyls into phenanthrenes in chlorobenzene at 120 °C over 20 h. Moreover, 2-ethynylheterobiaryls were found to be favorable substrates for this ruthenium catalysis, thus achieving the cycloisomerization of previously unused heterocyclic substrates. Moreover, several control experiments and DFT calculations of model complexes were performed to propose a plausible reaction mechanism.

Full text of DOI:10.1002/chem.201500248

DOI: 10.1002/chem.201500248
Full Paper
&
Catalysis
A Combined Experimental and Computational Study on the
Cycloisomerization of 2-Ethynylbiaryls Catalyzed by Dicationic
Arene Ruthenium Complexes
Yoshihiko Yamamoto,* Kazuma Matsui, and Masatoshi Shibuya[a]
Abstract: Ruthenium-catalyzed cycloisomerization of 2-ethy-
nylbiaryls was investigated to identify an optimal ruthenium
catalyst system. A combination of [h6-(p-cymene)RuCl2(PR3)]
and two equivalents of AgPF6 effectively converted 2-ethy-
nylbiphenyls into phenanthrenes in chlorobenzene at 1208C
over 20 h. Moreover, 2-ethynylheterobiaryls were found to
be favorable substrates for this ruthenium catalysis, thus
achieving the cycloisomerization of previously unused heter-
ocyclic substrates. Moreover, several control experiments
and DFT calculations of model complexes were performed
to propose a plausible reaction mechanism.
an atom-economical process[2] as one of the ortho CÀH bonds
is directly converted into a CÀC bond with concomitant intra-
molecular H-transfer. However, the thermal cycloisomerization
of 1a to 2a requires a very high reaction temperature (7008C)
that would be incompatible with labile substrates. Moreover,
benzazulene is formed as a side product. Therefore, transition-
metal catalysts have been sought to promote this reaction
under much milder conditions. In their seminal studies, Fꢀrst-
ner and co-workers identified metal salts with carbophilic p-
acid character, such as PtCl2, AuCln (n=1 and 3), GaCl3, and
InCl3, that catalyzed the cycloisomerization of various 2-alkynyl-
Introduction
The cycloisomerization of 2-ethynylbiphenyl (1a) provides
a straightforward access to phenanthrene (2a) because this re-
action proceeds by the intramolecular hydroarylation between
the ethynyl and phenyl moieties connected by a phenylene
tether (Scheme 1a).[1] Thus, this cycloisomerization is essentially
biaryls 1, affording phenanthrenes
2
even at 808C
(Scheme 1b).[3] Notably, internal alkynes including 1-haloal-
kynes (1, terminal group (TG)=Br and I) participated in the cy-
cloisomerization reaction in the presence of an appropriate
catalyst. Moreover, p-acid catalysts also enabled the use of het-
erobiaryl substrates for cycloisomerization as thiophene, pyr-
role, and indole participated in the CÀC bond formation with
pendant alkynes under mild reaction conditions. Since the pio-
neering report from Fꢀrstner’s group, this method has been
extensively applied to the synthesis of polyaromatic molecules
with intriguing structural and optoelectronic properties, such
as (hetero)pyrenes,[4] (hetero)helicenes,[5] and larger polyaro-
matic hydrocarbons.[6] The use of p-acid catalysts with a wide
functional-group compatibility also makes the 2-alkynylbiaryl
cycloisomerization applicable to the syntheses of natural prod-
ucts and bioactive molecules such as aporphine alkaloids,[3b,c]
tylophora alkaloids and analogues,[7] (+)-kibdelone A,[8] and
HIV-1 integrase inhibitors.[9]
Scheme 1. Thermal and transition-metal-catalyzed cycloisomerizations of 2-
alkynylbiphenyls 1.
Despite the significant progress in this field, some limitations
still remain to be solved. The electrophilic activation of alkynes
using p-acid catalysts prefers electron-rich aryl rings (R1 =elec-
tron-donating groups) as the coupling partners; therefore, aryl
groups substituted with electron-withdrawing groups are es-
sentially incompatible because of their insufficient nucleophi-
licity. The cycloisomerization of unsymmetrically substituted
[a] Prof. Dr. Y. Yamamoto, K. Matsui, Dr. M. Shibuya
Department of Basic Medicinal Sciences
Graduate School of Pharmaceutical Sciences, Nagoya University
Chikusa, Nagoya 464-8601 (Japan)
E-mail: yamamoto-yoshi@ps.nagoya-u.ac.jp
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500248.
Chem. Eur. J. 2015, 21, 1 – 12
1
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
substrates suffers from poor regioselectivity. Moreover, the cy-
cloisomerizations involving meta-substituted aryl rings often
lead to a mixture of the expected 6-endo cyclization product 2
and the undesired 5-exo cyclization product 3 (Scheme 1b).[3c]
To solve these problems, new methods have been developed
by investigating under-used metal elements and/or ligands.
For example, the cycloisomerization of electron-deficient biary-
lalkynes to the corresponding phenanthrenes has been ach-
ieved using Fe(OTf)3 as the catalyst, even though the alkyne
moiety is limited to those possessing a phenyl terminal (1,
TG=Ph).[10] In the presence of In(OTf)3 or a cationic gold(I) cat-
alyst with an N-heterocyclic carbene ligand, biarylalkynes pos-
sessing a phenylselenyl terminal (1, TG=SePh) underwent se-
lective 6-endo cycloisomerization irrespective of the substitu-
tion pattern on the aryl rings involved in the CÀC bond forma-
tion, albeit with the concomitant 1,2-migration of the phenyl-
selenyl group via vinylidene intermediates (vide infra) for the
gold-catalyzed reactions.[11] Furthermore, Alcarazo and co-work-
ers reported that platinum(II) and gold(I) complexes selectively
catalyzed the 6-endo cycloisomerization of diverse challenging
ethynylbiaryl substrates under mild reaction conditions. In par-
ticular, the cationic gold catalyst enabled the cycloisomeriza-
tion of 2’,6-disubstituted 2-ethynylbiphenyls, affording highly
strained 4,5-disubstituted phenanthrenes even at room tem-
perature.[12b] In these examples, strong p-acid ligands, 2,3-dia-
lkylaminocyclopropenium-substituted phosphines, play impor-
tant roles in the electrophilic activation of the ethynyl moiety,
thus facilitating the desired 6-endo cycloisomerization. In strik-
ing contrast, Gevorgyan and Chernyak reported that the neu-
tral palladium(II) catalyst with 1,1’-bis(phoshino)ferrocene li-
gands generally catalyzed the 5-exo cycloisomerization of biar-
ylalkynes possessing (hetero)aryl or ethoxycarbonyl terminal
groups on the alkyne moiety (1, TG=aryl or CO2Et), resulting
in the stereoselective formation of alkylidenefluorene deriva-
tives 3.[13] Notably, the palladium-catalyzed 5-exo cycloisomeri-
zation preferred electron-deficient substrates (R1 =electron-
withdrawing substituents) rather than electron-rich substrates,
and an aromatic CÀH activation pathway is proposed based on
kinetic isotope effects observed in deuterium-labeling experi-
ments. Therefore, the reaction efficiency, substrate scope, and
product selectivity should be controlled by the judicious
choice of neutral/cationic metal–ligand combinations.
Scheme 2. Catalytic cycloisomerizations using [(p-cymene)RuCl2(PR3)] as pre-
catalysts.
ruthenium catalysts for the cycloisomerization of 2-ethynyl-
3’,5’-dimethoxybiphenyl (1i) with a 5 mol% catalyst loading in
toluene at 808C (Scheme 2c).[3c] A complete consumption of 1i
was observed by using a dicationic catalyst, which was derived
in situ from [(p-cymene)RuCl2(PCy3)] (4a·PCy3) and two equiva-
lents of AgBF4, even though the 6-endo/5-exo selectivity was as
low as 3:7 and the isolated yield of the major 5-exo product 3i
was only 17%. Independently, Liu and co-workers reported
that
a
different
cationic
ruthenium
catalyst,
[TpRu(PPh3)(CH3CN)2PF6] (Tp=tris(pyrazolyl)borate), efficiently
catalyzed the 6-endo cycloisomerization of 2-ethynylbiphenyl
(1a), affording phenanthrene (2a) in a high yield.[6b] This indi-
cates that an appropriate combination of ancillary ligands, pos-
itive ionic charges, and counterions possibly improves the effi-
ciency and substrate scope of the cycloisomerization of 2-alky-
nylbiaryls.
Therefore, we decided to reinvestigate the ruthenium cata-
lyst system based on [(h6-arene)RuCl2(PR3)] (4·PR3) because
they are readily prepared,[15] and the catalytic efficiency can be
readily modulated by altering the h6-arene and phosphine li-
gands as well as silver additives. In this paper, we report the
development of ruthenium catalysts for the catalytic cycloiso-
merization of diverse 2-ethynylbiaryls.
With this background, we focused on the ruthenium-cata-
lyzed cycloisomerizations of biarylalkynes. In a seminal study
by Merlic and Pauly, a monocationic catalyst with an arene
ligand, which was derived in situ from [(p-cymene)RuCl2(PPh3)]
(4a·PPh3) and NH4PF6, was reported to be an efficient catalyst Results and Discussion
for
the
cycloisomerization
of
heteroarylenynes
5
Screening of ruthenium catalysts
(Scheme 2a).[14] However, the ruthenium-catalyzed cycloisome-
rizations of 2-ethynylbiaryls have not been investigated. Subse-
quently, Donovan and Scott reported the cycloisomerization of
enediyne 6 to pentacycle 7 in an excellent yield using 4a with-
out any additive, even though a similar reaction of an oxa ana-
logue of 6 afforded the corresponding product in <10% yield
(Scheme 2b).[6a] This substrate-dependent variation of the cata-
lyst efficiency is a severe drawback for the ruthenium catalyst
system. Moreover, Fꢀrstner and co-workers screened several
At the outset of this study, a ligand set and a silver additive
were optimized for the ruthenium-catalyzed cycloisomerization
of ethynylbiphenyl 1b as a representative substrate (Table 1).
In the absence of a silver salt, a solution of 1b and 5 mol%
4a·PPh3 in chlorobenzene was heated at 1208C (Table 1,
entry 1). However, no reaction occurred within 20 h. Then, the
reaction was repeated by adding 11 mol% AgBF4, resulting in
the complete consumption of 1b (entry 2). Purification using
&
&
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
2
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Substrate scope and limitations
Table 1. Optimization of ruthenium catalyst 4·PR3/nAgX.
The general applicability of the ruthenium-catalyzed cycloiso-
merization was investigated using 5 mol% 4a·PPh3 and
11 mol% AgPF6 in PhCl at 1208C for 20 h as the standard reac-
tion conditions (Table 2). 2-Ethynylbiphenyl 1a and 4’-chloro
analogue 1c underwent cycloisomerization to afford phenan-
Entry
Ru complex[a]
AgX
2b Isolated yield [%]
1
4a·PPh3
–
no reaction
2
3
4
5
6
7
8
9
10
11
12
13
14
15
4a·PPh3
4a·PPh3
4a·PPh3
4a·PPh3
AgBF4
AgPF6
AgPF6
74
77
44
59
Table 2. Cycloisomerizations of 2-ethynylbiphenyls using 4a·PAr3/AgPF6.[a]
[b]
Entry
Substrate
Product, yield [%][b]
AgOTf
AgNTf2
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
AgPF6
4a·PPh3
complex mixture
4a·P(p-MeOC6H4)3
4a·P(p-FC6H4)3
4a·P(p-F3CC6H4)3
4a·PCy3
4a·P(OMe)3
4a
4b·PPh3
4c·PPh3
[RuCl2(PPh3)3]
74
79
73
53
72
21
68
73
30
1
2
3
4
5
[a] Arene ligands are p-cymene, benzene, and hexamethylbenzene for 4a,
4b, and 4c, respectively. [b] 7.5 mol%.
silica gel chromatography afforded phenanthrene 2b in 74%
yield. The use of AgPF6 instead of AgBF4 resulted in a slightly
better yield (entry 3), whereas the loading of AgPF6 was de-
creased to half, leading to an incomplete reaction (entry 4).
These results clearly indicate that the dicationic ruthenium spe-
cies of the type [(h6-arene)Ru(PR3)]2+ are competent catalysts.
The use of AgOTf or AgNTf2 as the additives caused inferior re-
sults, which indicated that less coordinating counterions are
essential (entries 5 and 6). Next, the effect of phosphine li-
gands was investigated. When P(p-MeOC6H4)3 and P(p-FC6H4)3,
with similar cone angles but electron-donating and -withdraw-
ing, respectively, compared to PPh3, were used, comparable
yields were obtained (entries 7 and 8). Because the latter gave
6
7
8
9
a
slightly higher yield, more electron-withdrawing P(p-
CF3C6H4)3 was investigated to improve the yield (entry 9). How-
ever, the yield was slightly lowered. Moreover, the yield was
considerably decreased, when bulky and strongly electron-do-
nating PCy3 was the phosphine ligand, which was previously
used by Fꢀrstner and co-workers (entry 10).[3c] Small and elec-
tron-withdrawing P(OMe)3 was a better ligand than PCy3
(entry 11). The importance of phosphine ligands was unambig-
uously shown by the reaction conducted using 4a without
PPh3, affording 2b in only 21% yield (entry 12). In addition to
p-cymene complex 4a·PPh3, similar complexes 4b·PPh3 and
4c·PPh3, with more electron-donating hexamethylbenzene and
less electron-donating benzene as the h6-arene ligands, respec-
tively, were used as the precatalysts, resulting in slightly lower
yields of 2b (entries 13 and 14). Thus, p-cymene was the opti-
mal h6-arene ligand. The necessity of the h6-arene ligand was
also confirmed by the reaction using [RuCl2(PPh3)3] as the pre-
catalyst (entry 15). In this case, 2b was obtained in only 30%
yield. Finally, AgPF6 exhibited no catalytic activity.
10
11
12
2b, 50
2g, 18
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
3
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
tions, affording phenanthrene 2b in 50% yield (Table 2,
entry 11). Moreover, the reaction of 5-methoxy analogue 1m
afforded a complex mixture of products, and 2g was isolated
in a low yield (entry 12). These results are in striking contrast
to the fact that the same products 2b and 2g were obtained
in much better yields from substrates 1b and 1g, respectively.
Thus, the electron-donating substituents on the ethynyl-substi-
tuted phenyl ring negatively affected the cyclization efficiency.
On the other hand, the reaction of 5-fluoro derivative 1n af-
forded the corresponding phenanthrene 2l in a higher yield of
57% (entry 13). However, substrate 1o possessing substituents
on both phenyl rings resulted in a complex product mixture
(entry 14).
Table 2. (Continued)
Entry
Substrate
Product, yield [%][b]
13
14
complex mixture
[a] Standard conditions: 5 mol% 4a·PPh3, 11 mol% AgPF6, PhCl, 1208C,
20 h. [b] Isolated yields. Yields obtained using P(p-MeOC6H4)3 as the
ligand are shown in parentheses. [c] An inseparable mixture of 2g and its
1-methoxy isomer 2g’ with 13:1 ratio. [d] 1h was recovered in 13% yield.
[e] An inseparable mixture with the 2i/3i ratio of 4:1.
Subsequently, the cycloisomerizations by intramolecular het-
eroarene CÀH alkenylation were investigated as summarized in
Table 3. Under the standard reaction conditions, benzofuran
derivative 8a efficiently underwent cycloisomerization to
threnes 2a and 2c in 72 and 74% yields, respectively (Table 2,
entries 1 and 2). On the other hand, the reaction of more elec-
tron-deficient 4’-fluoro substrate 1d afforded 2d, albeit in
a moderate yield of 61% (entry 3). These results indicate that
the substituents on the phenyl ring that undergo intramolecu-
lar CÀH alkenylation have a significant electronic impact on
the reaction efficiency, and an electron-withdrawing group
lowers the product yield. This electronic effect was more pro-
nounced in the cycloisomerizations of substrates possessing
methoxycarbonyl and acetyl substituents at the 4’ positions, af-
fording 2e and 2 f in 56 and 47% yields, respectively (entries 4
and 5).
Table 3. Cycloisomerizations of 2-ethynylheterobiaryls using 4a·PPh3/
AgPF6.[a]
Entry
1
Substrate
Product, yield [%][b]
2
3
4
The effect of the substitution pattern was then investigated
by using 2-ethynylbiphenyls 1g and 1h possessing a methoxy
group at the 3’ or 2’-positions, respectively, as the substrates
(Table 2, entries 6 and 7). The reaction of 1g afforded an insep-
arable mixture of 3-methoxyphenanthrene (2g) and 1-methox-
yphenanthrene in 74% combined yield with a 3-MeO/1-MeO
ratio of 13:1, whereas the reaction of 1h afforded 4-methoxy-
phenanthrene 2h in a moderate yield (65%). The lower yield
of 2h was caused by the unfavorable steric repulsion between
the methoxy group and hydrogen atom at the 4- and 5-posi-
tions, respectively.[12b] The reactivity of problematic 3’,5’-dime-
thoxy-substituted substrate 1i was reinvestigated (entry 8), be-
cause Fꢀrstner and co-workers previously used 4a·PCy3/AgBF4
as the catalyst in toluene at 808C to obtain 5-exo cycloisomeri-
zation product 3i in 17% yield as the major isomer (Sche-
me 2c).[3c] The new catalyst system in this study produced 2i
and 3i as an inseparable mixture in 41% combined yield with
a 2i/3i ratio of 4:1. Thus, the different phosphine ligands,
counterions, and reaction conditions not only improved the
catalytic efficiency, but also inversed the 6-endo/5-exo selectivi-
ty. In contrast to 1i, 3’,4’-dimethoxy analogue 1j and ben-
zo[1,3]dioxole derivative 1k underwent selective 6-endo cycloi-
somerization to afford the corresponding phenanthrenes 2j
and 2k in 62 and 67% yields, respectively (entries 9 and 10).
Furthermore, the cycloisomerization of 2-ethynyl-4-methoxy-
biphenyl 1l was carried out under the standard reaction condi-
5
6
7
8
9
&
&
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
4
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
parent thiophene analogue 8i and 2-chloro and 2-acetyl ana-
logues 8j and 8k also afforded the desired products in good
yields (72–77%, entries 9–11). Moreover, 1H-benzo[g]indole 9l
was successfully obtained in 74% yield by the one-pot cyclo-
isomerization/deprotection of N-Boc pyrrole derivative 8l
(entry 12). In contrast to the above examples, the cycloisomeri-
zations of 2-ethynyl-3-(4-aryl)thiophenes 8m and 8n were
problematic. The reactions of these substrates afforded 7-me-
thoxynaphtho[2,1-b]thiophene (9m) and benzo[1,2-b:4,3-b’]di-
thiophene (9n), albeit in 49 and 44% yields, respectively (en-
tries 13 and 14). As a whole, the ruthenium-catalyzed cycloiso-
merization was highly efficient for 2-ethynylheterobiaryl sub-
strates except for the 2-ethynylthiophene derivatives.
Table 3. (Continued)
Entry
Substrate
Product, yield [%][b]
10
11
12
Mechanistic considerations
13
14
As shown in Scheme 1b, several reaction pathways were pro-
posed for the transition-metal-catalyzed cycloisomerizations of
2-alkynylbiphenyls 1, affording phenanthrenes 2 and alkylide-
nefluorenes 3. Based on the recent theoretical studies, the
electrophilic activation of the alkyne moiety by the formation
of p-alkyne complexes IM1 has been postulated as the initial
step of cycloisomerizations using p-acid platinum, gold, and
indium catalysts.[12,16] A subsequent intramolecular Friedel–
Crafts-type cyclization followed by sequential hydrogen shifts
afford phenanthrenes 2 or alkylidenefluorenes 3. Alternatively,
the involvement of the intramolecular cyclopropanation of IM1
was also proposed for the platinum-catalyzed cycloisomeriza-
tion, affording phenanthrenes 2.[12,16] When the cycloisomeriza-
tions of 2-(2-haloalkynyl)biphenyls 1 (TG=I or Br) were per-
formed using AuCl as the catalyst, the corresponding phenan-
threnes 2’ were obtained rather than isomeric 2.[3] Although
a pathway involving the formation of vinylidene complexes
IM2 by 1,2-halide migration was initially proposed to explain
this phenomenon, an alternative mechanism comprising the
Friedel–Crafts-type cyclization of p-alkyne complexes IM1 and
a subsequent 1,2-halide migration was later proposed based
on theoretical studies.[16] Therefore, the p-alkyne pathway has
been accepted as the general mechanism for the cycloisomeri-
zations using p-acid catalysts.
[a] Standard conditions: 5 mol% 4a·PPh3, 11 mol% AgPF6, PhCl, 1208C,
20 h. [b] Isolated yields. [c] Compound 10 was also formed in 7% yield.
[d] Two-step yields after removal of the Boc group.
afford the desired benzo[d]naphtho[1,2-b]furan (9a) in 85%
yield (entry 1). The cycloisomerizations of similar substrates 8b
and 8c possessing electron-donating methoxy and electron-
withdrawing fluoro substituents, respectively, on the ethynyl-
substituted phenyl ring were conducted without difficulty, af-
fording the corresponding products 9b and 9c in high yields
of over 80% (entries 2 and 3). These results are in striking con-
trast to the fact that the reactions of biaryls 1l–n resulted in
lower yields compared to other biaryl substrates (Table 2, en-
tries 11–13). Although the reaction of benzothiophene deriva-
tive 8d also afforded benzo[d]naphtho[1,2-b]thiophene (9d) in
90% yield (Table 3, entry 4), 11H-benzo[a]carbazole (9e) was
obtained in 67% yield along with small amounts of intramolec-
ular hydroamination byproduct 10 from indole-derived sub-
strate 8e (entry 5). This result is consistent with that of the pre-
vious report: the platinum(II)-catalyzed cycloisomerization of
8e also afforded 9e and 10.[3c] Thus, a similar substrate 8 f
with N-Boc protection was subjected to the present cycloiso-
merization conditions to avoid the undesired side reaction
(entry 6). Because a partial deprotection occurred, the crude
products were directly treated with trifluoroacetic acid (TFA) at
608C for 3 h. This one-pot procedure successfully afforded the
desired benzocarbazole derivative 9 f in 82% yield. Similarly, 2-
methylfuran derivative 8g and 2-methylthiophene derivative
8h were efficiently converted into 2-methylnaphtho[1,2-
b]furan (9g) and 2-methylnaphtho[1,2-b]thiophene (9h) in 92
and 82% yields, respectively (entries 7 and 8). The reactions of
To elucidate the mechanism of the cycloisomerization cata-
lyzed by the dicationic ruthenium complex, several control ex-
periments were performed (Scheme 3). First, the substrates
possessing an internal alkyne moiety were subjected to the
standard cycloisomerization conditions. In the presence of
5 mol% 4a·PPh3 and 11 mol% AgPF6, 11 a and 11 b with
methyl and p-tolyl terminal groups, respectively, were sepa-
rately heated in chlorobenzene at 1208C for 20 h, resulting in
the 91 and 93% recovery of intact 11 a and 11 b, respectively
(Scheme 3a). These results indicate that this cycloisomerization
of 2-ethynylbiaryls proceeds by the electrocyclization of vinyl-
idene intermediates such as IM2. Moreover, the vinylidene
mechanism was also proposed for the previous ruthenium-cat-
alyzed cycloisomerizations.[6a,b,14,17]
However, evidence that contradicts the vinylidene mecha-
nism was also obtained. This ruthenium(II)-catalyzed cycloiso-
merization of 2-ethynylbiaryls should be performed under rig-
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
5
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
computational efficiency because the electronic effects of the
h6-arene and phosphine ligands are less important for the cyc-
lization efficiency (for details of computation, see the Support-
ing Information). Three starting complexes relevant to the
present reaction were located (Figures 1–3). Among these in-
termediates, p-alkyne complex A1 and h1-allenylruthenium
complex B1 have similar energy, and vinylidene complex C1
was located approximately 3 kcalmolÀ1 higher than A1 and B1.
Subsequent calculations indicate that the 6-endo cyclization to
afford a phenanthrene complex occurs from complexes A1
and C1 (Figures 1 and 2), while complex B1 ultimately evolves
to
a
methylenefluorene complex by 5-exo cyclization
(Figure 3).
Figure 1 shows the free energy surface for the conversion of
p-alkyne complex A1 to h2-phenanthrene complex A5. In A1,
the ethyne moiety was significantly bent by the coordination
of the ruthenium center as shown by the bond angles of C2-
Ca-Cb=148.3 and Ca-Cb-H=139.38 and the CaÀCb bond
length of 1.317 ꢂ is similar to that of a typical Csp2=Csp2
double bond. The 6-endo cyclization of A1 proceeded by tran-
sition-state TSA1A2 with an activation energy of DG° =
+10.1 kcalmolÀ1. On transferring from A1 to TSA1A2, the CbÀ
C2’ distance decreased from 3.002 to 2.212 ꢂ. The cyclization
product A2, which is a distorted h1-alkenyl complex with
a RuÀC single-bond length of 2.108 ꢂ and bond angles of Ru-
Ca-Cb=99.7 and Ru-Ca-C2=135.18 had a slightly higher
energy than A1 (+0.003 kcalmolÀ1); thus, A1 and A2 are in
equilibrium. A subsequent 1,2-shift of the C2’ proton occurred
by transition-state TSA2A3 with a smaller activation energy of
DG° =+7.9 kcalmolÀ1 than that of TSA1A2. The formation of
carbene complex A3 with a short Ru=Ca bond length of
1.958 ꢂ is highly exergonic (À24.3 kcalmolÀ1). The newly
formed CbÀH2 bond is oriented perpendicular to the Ru=Ca
bond (Ru-Ca-Cb-H2=94.08), while the original CbÀH1 bond
makes a very weak agostic interaction with the ruthenium
center (RuÀH1=2.358 ꢂ). Complex A3 and its conformational
isomer A4 are in equilibrium with a very small activation barri-
er of approximately 2 kcalmolÀ1. In A4, the CbÀH2 bond
makes a very weak agostic interaction with the ruthenium
center (RuÀH2=2.320 ꢂ). Finally, A4 evolved to h2-phenan-
threne complex A5 by the 1,2-shift of the original Ca proton
H1 by transition-state TSA4A5. The activation energy of this step
is DG° = +16.7 kcalmolÀ1, and the formation of A5 is exergon-
ic (À16.2 kcalmolÀ1). Overall, the initial 6-endo cyclization fol-
lowed by the first 1,2-H shift occurs with relatively small activa-
tion barriers. Although the activation barrier of the second 1,2-
H shift is the largest, this is small enough to override under
the experimental conditions. The formation of A3 can be con-
sidered as irreversible because this step is highly exergonic,
and transition state TSA1A2 has the highest energy. Thus, the 6-
endo cyclization step is the product- and rate-determining
step. Because the activation barriers are <20 kcalmolÀ1, the
ruthenium-mediated transformation of 2-ethynylbiphenyl to
phenanthrene should proceed even at room temperature. This
was indeed confirmed by the stoichiometric reaction of
4a·P(OMe)3 with 1b as shown in Scheme 3c.
Scheme 3. Control experiments.
orously anhydrous conditions because of the hydration of the
alkyne moiety when the reaction was carried out in the pres-
ence of adventitious water. In fact, the reaction of representa-
tive substrate 1b was performed under the standard cycloiso-
merization conditions except for adding one equivalent of
H2O, affording the known ketone 12[18] as the major product
(68% NMR) along with phenanthrene 2b (11% NMR) and
intact 1b (20% NMR) as shown in Scheme 3b. The formation
of 12 can be attributed to the Markovnikov hydration via a p-
alkyne intermediate, and the anti-Markovnikov hydration prod-
uct 13 was not detected. These observations are in contrast to
those reported by Wakatsuki and co-workers, who showed
that the ruthenium-catalyzed anti-Markovnikov hydration of
terminal alkynes proceeded via vinylidene intermediates to
afford aldehydes.[19]
Furthermore, a stoichiometric reaction of representative sub-
strate 1b with p-cymene ruthenium complex 4a was carried
out (Scheme 3c). P(OMe)3 was selected as the supporting
1
ligand for a clear H NMR spectroscopic analysis. When a 1:1
mixture of 4a·P(OMe)3 and 1b in CDCl3 was treated with two
equivalents of AgPF6 at room temperature, a spontaneous
change in the color of the solution from red to brown and
1
concomitant precipitation were observed. The H NMR spectro-
scopic analysis of the resulting crude reaction mixture showed
the clean formation of phenanthrene 2b in 85% NMR spectro-
scopic yield without any noticeable side product other than
the decomposed complex. Therefore, an elevated reaction
temperature of 1208C is probably required for the catalytic re-
actions to restore the catalytically active species from the rest-
ing states stabilized by the coordination of the product (i.e.
product inhibition). In contrast, the cycloisomerization of sub-
strate 11 a with an internal alkyne failed under the same stoi-
chiometric reaction conditions.
Then, the DFT calculations of model complexes were per-
formed, using h6-benzene and PH3 as the simplified ligands for
&
&
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
6
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 1. Energy surface for conversion of A1 to A5 with relative Gibbs free energies in chlorobenzene at 298 K.
Next, the energy surface for the conversion of p-alkyne com-
plex A1 to h2-methylenefluorene complex B4 was calculated as
shown in Figure 2. Because the attempt to locate a transition
state for the direct 5-exo cyclization from A1 failed, an alterna-
tive route from h1-allenyl complex B1 was investigated. The ro-
tation around the C1ÀC1’ bond of A1 induced the isomeriza-
tion to B1, which may be a delocalized h1-alkyne complex lo-
cated À0.1 kcalmolÀ1 below A1. Complexes A1 and B1 are in
equilibrium with very small activation barriers of less than
+2 kcalmolÀ1. The 5-exo cyclization of B1 proceeded via transi-
tion-state TSB1B2 with an activation energy of DG° = +14.8 kcal
molÀ1, and the formation of h1-alkenyl complex B2 was 2.7 kcal
molÀ1 endergonic. A subsequent 1,2-H shift from B2 occurred
via transition-state TSB2B3 with an activation energy of DG° =
+14.7 kcalmolÀ1, which is larger than that of the retrocycliza-
tion from B2 to B1 (DG° = +12.1 kcalmolÀ1). The formation of
carbene complex B3 with a Ru=Cb bond length of 1.898 ꢂ is
11.8 kcalmolÀ1 exergonic. The second 1,2-H shift from B3 af-
forded the final h2-methylenefluorene complex B4 with a large
exergonicity of 24.0 kcalmolÀ1. This step proceeded via transi-
tion-state TSB3B4 with an activation energy of +13.7 kcalmolÀ1
,
which is comparable to that of the first 1,2-H shift. According
to these analyses, the 5-exo cyclization route from h1-allenyl
complex B1 to h2-methylenefluorene complex B4 is less effi-
cient than the above 6-endo cyclization from p-alkyne complex
A1, because the activation barrier of the rate-determining 5-
exo cyclization step is much larger than that of the 6-endo cyc-
lization from A1 (DG° = +14.8 vs. +10.1 kcalmolÀ1) and the
subsequent 1,2-H shift is also less facile than that in the 6-endo
route (DG° = +14.7 vs. 7.9 kcalmolÀ1). Therefore, it is conclud-
ed that the 6-endo cyclization mode is more favorable than the
5-exo cyclization mode for the ruthenium-catalyzed cycloiso-
merization of 2-ethynylbiaryls.
The involvement of a vinylidene intermediate was also inves-
tigated (Figure 3). Because a transition state for the direct con-
version of p-alkyne complex A1 to a vinylidene complex could
not be located, the 1,2-H shift from h1-allenyl complex B1 was
investigated, resulting in the identification of transition-state
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
7
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 2. Energy surface for conversion of A1 to B4 with relative Gibbs free energies in chlorobenzene at 298 K.
TSB1C1. The activation energy for
this step was estimated as
DG° = +18.2 kcalmolÀ1, and the
formation of C1 is endergonic
by 3.5 kcalmolÀ1. Therefore, the
formation of vinylidene complex
C1 is less favorable, although
the subsequent cyclization from
C1 proceeded via transition-
state TSC1C2 with an activation
energy (DG° = +11.5 kcalmolÀ1
)
that is comparable to that of the
6-endo cyclization from A1. The
final 1,2-H shift from C2 to h2-
phenanthrene complex A5 via
TSC2A5
was
facile
(DG° =
+8.4 kcalmolÀ1) and highly ex-
ergonic (41.1 kcalmolÀ1). Conse-
quently, the vinylidene route can
be excluded because of the un-
favorable formation of vinyli-
dene intermediate C1.
The key 6-endo cyclization
step was further evaluated by
performing the calculations for
model complex D1 with the
Figure 3. Energy surface for conversion of A1 to A5 via C1 with relative Gibbs free energies in chlorobenzene at
298 K.
&
&
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
8
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
a substituent at the 5-position were more endergonic than
that of D2. In particular, the formation of 5-methoxy analogue
H2 was highly endergonic as the activation energy of the re-
verse process, DG° = +4.4 kcalmolÀ1, was much smaller than
that of the forward process, DG° = +15.4 kcalmolÀ1. The low-
yield formation of 2g from 1m (Table 2, entry 12) is consistent
with this theoretical prediction.
The 6-endo cyclization of furyl analogue J1 proceeded with
an activation energy of DG° = +13.8 kcalmolÀ1, which is com-
parable to that for D1. In contrast to the cyclizations of the
above biphenyl models, the formation of J2 is slightly exergon-
ic. These results indicate that a heterocyclic analogue with
lesser aromaticity is a better substrate for the present cycliza-
tion. This is consistent with the experimental results that heter-
oaryl substrates are superior to biphenyl substrates. In striking
contrast, the activation energy for the 6-endo cyclization of K1
with an internal alkyne (DG° = +24.7 kcalmolÀ1) is considera-
bly larger than those of terminal alkyne complexes. Moreover,
the formation of K2 is significantly endergonic (+23.2 kcal
molÀ1). Thus, the inverse process is more feasible as the corre-
sponding activation barrier is much smaller (DG° = +1.5 kcal
molÀ1). Therefore, it can be concluded that the cyclization of
internal biphenylalkynes is not facile both kinetically and ther-
modynamically. This conclusion is consistent with the experi-
mental results for internal alkyne substrates 11 a and 11 b
(Scheme 3a).
Scheme 4. 6-Endo cyclizations of model complexes E1–K1.
PPh3 ligand (Scheme 4). The cyclization of 2-ethynylbiphenyl
complex D1 proceeded with an activation energy of DG° =
+14.6 kcalmolÀ1. This value is 4.5 kcalmolÀ1 larger than that of
A1 with the PH3 ligand, but is still small enough for the cycliza-
tion to proceed at ambient temperature. Moreover, the forma-
tion of D2 was endergonic (+5.9 kcalmolÀ1). The effect of the
structural variation was further evaluated by computing vari-
ous models. The activation energy for the cyclization of E1
with an electron-donating methoxy substituent on the 4’ posi-
tion was estimated as DG° = +11.8 kcalmolÀ1, which is 2.8 kcal
molÀ1 lower than that of D1. Conversely, a significantly larger
activation energy of DG° = +17.9 kcalmolÀ1 was obtained for
the cyclization of F1 with an electron-withdrawing fluoro sub-
stituent at the 4’-position. More-
Synthetic application: Synthesis of polyheteroaromatic mol-
ecules by tandem cycloisomerizations
Polyaromatic compounds containing dibenzofuran and diben-
zothiophene moieties are attractive synthetic targets because
of their optoelectronic properties.[21,22] Thus, the tandem cyclo-
isomerizations of 2-ethynylheterobiaryls to polyheteroarenes
were investigated to demonstrate the synthetic potential
of the present ruthenium-catalyzed cycloisomerization
(Schemes 5 and 6).
The Sonogashira coupling of 1,4-dibromo-2,5-diiodobenzene
with 2.2 equivalents of o-siloxyphenylacetylene 14, which was
prepared from commercially available 5-tert-butyl-2-iodophe-
over, the latter case was approxi-
mately 4 kcalmolÀ1 more ender-
gonic. These results indicate the
electrophilic character of the
cyclization mediated by the di-
cationic ruthenium complex. On
the other hand, introduction of
substituents on the ethynyl-sub-
stituted phenyl ring has an insig-
nificant effect on the activation
barrier of the cyclizations: The
calculated activation energies for
models G1, H1, and I1 were
~15 kcalmolÀ1, which are com-
parable to that of D1. However,
the formations of H2 and I2 with Scheme 5. Synthesis of anthra[1,2-b:5,6-b’]bisbenzofuran 18 via Ru-catalyzed tandem cycloisomerization.
Chem. Eur. J. 2015, 21, 1 – 12 www.chemeurj.org ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9
&
&
These are not the final page numbers! ÞÞ

Full Paper
Experimental Section
General procedure for ruthenium-catalyzed cycloisomeriza-
tion—Synthesis of 2b
AgPF6 (6.7 mg, 0.027 mmol) was added to a solution of [h6-(p-cym-
ene)RuCl2(PPh3)] (4a·PPh3) (6.8 mg, 0.012 mmol) in dry degassed
chlorobenzene (2 mL) and the mixture was stirred under an argon
atmosphere at room temperature for 5 min. To this catalyst solu-
tion was added a solution of 2-ethynylbiphenyl 1b (50.0 mg,
0.24 mmol) in dry chlorobenzene (1 mL) at room temperature, and
the reaction mixture was degassed three times at À788C. The reac-
tion mixture was then heated at 1208C under argon for 20 h. After
having been cooled to room temperature, the reaction mixture
was concentrated in vacuo, and the crude material was purified
with silica gel column chromatography (hexane/AcOEt=50:1) to
give 2b[23] (38.3 mg, 77% yield) as a colorless solid (m.p. 93.0–
94.08C, lit. m.p. 93.5–94.58C[23]). 1H NMR (400 MHz, CDCl3, 258C):
d=3.97 (s, 3H), 7.26–7.32 (m, 2H), 7.51–7.56 (m, 1H), 7.60–7.66 (m,
1H), 7.67 (d, J=9.0 Hz, 1H), 7.73 (d, J=9.0 Hz, 1H), 7.87 (d, J=
7.6 Hz, 1H), 8.59 ppm (d, J=8.8 Hz, 2H); 13C NMR (100 MHz, CDCl3,
258C): d=55.3, 108.5, 117.0, 122.1, 124.2, 124.6, 125.5, 126.4, 126.6,
127.5, 128.5, 130.4, 131.0, 133.4, 158.2 ppm.
Scheme 6. Synthesis of dinaphtho[1,2-b:2’,1’-d]thiophene (21) via Ru-cata-
lyzed tandem cycloisomerization.
nol in three steps, afforded diyne 15 in 98% yield (Scheme 5).
Subsequent removal of the TBS groups with tetrabutylammo-
nium fluoride (TBAF) in refluxing THF converted 15 to bisben-
zofuran 16 in 84% yield. Finally, the Sonogashira coupling with
trimethylsilylacetylene followed by desilylation afforded the de-
sired substrate 17 in 90% yield. The final tandem cycloisomeri-
zation of 17 was carried out under the standard reaction con-
ditions, affording the expected product 18, with the hitherto
unknown anthra[1,2-b:5,6-b’]bisbenzofuran framework, in 59%
yield.
Acknowledgements
This work was partially supported by the Platform for Drug Dis-
covery, Informatics, and Structural Life Science and Grant-in-
Aid for challenging Exploratory Research from the Ministry of
Education, Culture, Sports, Science and Technology, Japan.
Furthermore, dinaphtho[1,2-b:2’,1’-d]thiophene (21) was ob-
tained in 41% yield via the tandem cycloisomerization of diyne
20, which was straightforwardly prepared by the tandem
Suzuki–Miyaura coupling of commercially available 2,5-dibro-
mothiophene with o-formylphenylboronic acid followed by
Ohira–Bestmann alkynylation of 19 (Scheme 6).
Keywords: ethynylbiaryls · cycloisomerization · heterocycles ·
phenantherenes · ruthenium catalyst
[1] I. D. Mackie, R. P. Johnson, J. Org. Chem. 2009, 74, 499–503, and refer-
ences cited.
Conclusion
[2] B. M. Trost, Science 1991, 254, 1471–1477.
[3] a) A. Fꢀrstner, V. Mamane, J. Org. Chem. 2002, 67, 6264–6267; b) A.
Fꢀrstner, V. Mamane, Chem. Commun. 2003, 2112–2113; c) V. Mamane,
P. Hannen, A. Fꢀrstner, Chem. Eur. J. 2004, 10, 4556–4575; d) A. Fꢀrst-
ner, V. Mamane, G. Seidel, D. Laurich, Org. Synth. 2006, 83, 103–110.
[4] a) M. J. D. Bosdet, W. E. Piers, T. S. Sorensen, M. Parvez, Angew. Chem.
Int. Ed. 2007, 46, 4940–4943; Angew. Chem. 2007, 119, 5028–5031;
b) D. B. Walker, J. Howgego, A. P. Davis, Synthesis 2010, 3686–3692; c) T.
Matsuda, T. Moriya, T. Goya, M. Murakami, Chem. Lett. 2011, 40, 40–41;
d) M. M. Machuy, C. Wꢀrtele, P. R. Schreiner, Synthesis 2012, 44, 1405–
1409; e) D. Lorbach, M. Wagner, M. Baumgarten, K. Mꢀllen, Chem.
Commun. 2013, 49, 10578–10580; f) K. Yamamoto, N. Oyamada, S. Xia,
Y. Kobayashi, M. Yamaguchi, H. Maeda, H. Nishihara, T. Uchimaru, E.
Kwon, J. Am. Chem. Soc. 2013, 135, 16526–16532; g) J. H. Delcamp, A.
Yella, T. W. Holcombe, M. K. Nazeeruddin, M. Grꢃtzel, Angew. Chem. Int.
Ed. 2013, 52, 376–380; Angew. Chem. 2013, 125, 394–398.
We successfully identified the optimal ruthenium catalyst
system for the cycloisomerization of 2-ethynylbiaryls as [h6-(p-
cymene)RuCl2(PR3)] (4a·PR3) with two equivalents of AgPF6. As
the phosphine ligand, PPh3, and its analogues and P(OMe)3
were comparably efficient. Under the optimized reaction con-
ditions (5 mol% catalyst, chlorobenzene, 1208C, 20 h), o-ethy-
nylbiphenyls were converted to the corresponding phenan-
threnes in various yields. Moreover, this ruthenium catalyst was
more efficient for the cycloisomerization of heteroaryl ana-
logues, affording the corresponding polyheterocycles in higher
yields. Based on several control experiments and the DFT cal-
culations of model complexes, the electrophilic 6-endo cycliza-
tion involving a p-alkyne complex and subsequent consecutive
1,2-H shifts are proposed as the most probable mechanism for
this ruthenium-catalyzed cycloisomerization of 2-ethynylbiaryls.
As the demonstration of the synthetic potential of thus devel-
oped ruthenium-catalyzed cycloisomerization, a concise syn-
thesis of the anthra[1,2-b:5,6-b’]bisbenzofuran framework was
accomplished for the first time, and a three-step route to di-
naphtho[1,2-b:2’,1’-d]thiophene from commercially available
starting materials was also established using tandem cycloiso-
merizations.
ˇ
[5] a) J. Storch, J. Cermꢄk, J. Karban, Tetrahedron Lett. 2007, 48, 6814–6816;
ˇ
´
ˇ
˚ˇ ˇ
b) J. Storch, J. Sykora, J. Cermꢄk, J. Karban, I. Cꢅsarovꢄ, A. Ruzicka, J. Org.
ˇ
ˇ
Chem. 2009, 74, 3090–3093; c) J. Storch, J. Cermꢄk, J. Karban, I. Cꢅsar-
´
ovꢄ, J. Sykora, J. Org. Chem. 2010, 75, 3137–3140; d) M. Weimar, R. C.
da Costa, F.-H. Lee, M. J. Fuchter, Org. Lett. 2013, 15, 1706–1709; e) K.
Yamamoto, M. Okazumi, H. Suemune, K. Usui, Org. Lett. 2013, 15, 1806–
1809; f) H. Oyama, K. Nakano, T. Harada, R. Kuroda, M. Naito, K. Nobusa-
wa, K. Nozaki, Org. Lett. 2013, 15, 2104–2107.
[6] a) P. M. Donovan, L. T. Scott, J. Am. Chem. Soc. 2004, 126, 3108–3112;
b) H.-C. Shen, J.-M. Tang, H.-K. Chang, C.-W. Yang, R.-S. Liu, J. Org. Chem.
2005, 70, 10113–10116; c) T.-A. Chen, T.-J. Lee, M.-Y. Lin, S. M. A. Sohel,
E. W.-G. Diau, S.-F. Lush, R.-S. Liu, Chem. Eur. J. 2010, 16, 1826–1833;
d) B. S. Shaibu, S.-H. Lin, C.-Y. Lin, K.-T. Wong, R.-S. Liu, J. Org. Chem.
&
&
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
10
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
2011, 76, 1054–1061; e) T.-A. Chen, R.-S. Liu, Chem. Eur. J. 2011, 17,
8023–8027; f) K. Kitazawa, T. Kochi, M. Nitani, Y. Ie, Y. Aso, F. Kakiuchi,
Chem. Lett. 2011, 40, 300–302; g) C. L. Eversloh, C. Li, K. Mꢀllen, Org.
Lett. 2011, 13, 4148–4150; h) T.-A. Chen, R.-S. Liu, Org. Lett. 2011, 13,
4644–4647; i) T. Nakae, R. Ohnishi, Y. Kitahata, T. Soukawa, H. Sato, S.
Mori, T. Okujima, H. Uno, H. Sakaguchi, Tetrahedron Lett. 2012, 53,
1617–1619; j) A. Wu, D. Xu, D. Lu, T. M. Penning, I. A. Blair, R. G. Harvey,
Tetrahedron 2012, 68, 7217–7233; k) D. Wu, H. Zhang, J. Liang, H. Ge, C.
Chi, J. Wu, S. H. Liu, J. Yin, J. Org. Chem. 2012, 77, 11319–11324; l) T.-L.
Wu, H.-H. Chou, P.-Y. Huang, C.-H. Cheng, R.-S. Liu, J. Org. Chem. 2013,
78, 267–274.
wa, M. Shido, K. Maeyama, H. Kusama, J. Am. Chem. Soc. 2000, 122,
10226–10227; d) K. Ohe, K. Miki, T. Yokoi, F. Nishino, S. Uemura, Organo-
metallics 2000, 19, 5525–5528; e) T. Miura, N. Iwasawa, J. Am. Chem.
Soc. 2002, 124, 518–519; f) K. Sangu, K. Fuchibe, T. Akiyama, Org. Lett.
2004, 6, 353–355.
[18] P. Gandeepan, K. Partharathy, C.-H. Cheng, J. Am. Chem. Soc. 2010, 132,
8569–8571.
[19] a) M. Tokunaga, Y. Wakatsuki, Angew. Chem. Int. Ed. 1998, 37, 2867–
2869; Angew. Chem. 1998, 110, 3024–3027; b) T. Suzuki, M. Tokunaga, Y.
Wakatsuki, Org. Lett. 2001, 3, 735–737; c) T. Suzuki, M. Tokunaga, Y. Wa-
katsuki, Tetrahedron Lett. 2002, 43, 7531–7533.
[7] a) A. Fꢀrstner, J. W. J. Kennedy, Chem. Eur. J. 2006, 12, 7398–7410; b) W.
Gao, A. P.-C. Chen, C.-H. Leung, E. A. Gullen, A. Fꢀrstner, Q. Shi, L. Wei,
K.-H. Lee, Y.-C. Cheng, Bioorg. Med. Chem. Lett. 2008, 18, 704–709.
[8] D. K. Winter, M. A. Endoma-Arias, T. Hudllicky, J. A. Beutler, J. A. Por-
co, Jr., J. Org. Chem. 2013, 78, 7617–7626.
[9] H. Sharma, T. W. Sanchez, N. Neamati, M. Detorio, R. F. Schinazi, X.
Cheng, J. K. Buolamwini, Bioorg. Med. Chem. Lett. 2013, 23, 6146–6151.
[10] K. Komeyama, R. Igawa, K. Takaki, Chem. Commun. 2010, 46, 1748–
1750.
[20] Gaussian 09, Revision C.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, ꢆ. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
[11] W. Lim, Y. H. Rhee, Eur. J. Org. Chem. 2013, 460–464.
[12] a) J. Carreras, M. Patil, W. Thiel, M. Alcarazo, J. Am. Chem. Soc. 2012, 134,
16753–16758; b) J. Carreras, G. Gopakumar, L. Gu, A. Gimeno, P. Linow-
ˇ
ski, J. Petuskova, W. Thiel, M. Alcarazo, J. Am. Chem. Soc. 2013, 135,
18815–18823.
[13] a) N. Chernyak, V. Gevorgyan, J. Am. Chem. Soc. 2008, 130, 5636–5637;
b) N. Chernyak, V. Gevorgyan, Adv. Synth. Catal. 2009, 351, 1101–1114.
[14] C. A. Merlic, M. E. Pauly, J. Am. Chem. Soc. 1996, 118, 11319–11320.
[15] a) M. A. Bennett, A. K. Smith, J. Chem. Soc. Dalton Trans. 1974, 233–241;
b) S. A. Serron, S. P. Nolan, Organometallics 1995, 14, 4611–4616.
[16] a) E. Soriano, J. Marco-Contelles, Organometallics 2006, 25, 4542–4553;
b) G. Huang, B. Cheng, L. Xu, Y. Li, Y. Xia, Chem. Eur. J. 2012, 18, 5401–
5415.
[21] a) W.-Y. Chai, E.-Q. Yang, Y.-L. Zhang, A.-L. Xie, X.-P. Cao, Synthesis 2012,
44, 439–445; b) M. Furusawa, T. Imahori, K. Igawa, K. Tomooka, R. Irie,
´
´
Chem. Lett. 2013, 42, 1134–1136; c) S. Radenkovic, J. Kojic, J. Petronijev-
´
´
ic, M. Antic, J. Phys. Chem. A 2014, 118, 11591–11601.
[22] K. Takimiya, M. Nakano, M. J. Kang, E. Miyazaki, I. Osaka, Eur. J. Org.
Chem. 2013, 217–227.
[23] F. B. Mallory, M. J. Rudolph, S. M. Oh, J. Org. Chem. 1989, 54, 4619–
4626.
[17] For selected examples of electrocyclization of vinylidene intermediates,
see: a) K. Ohe, M. Kojima, K. Yonehara, S. Uemura, Angew. Chem. Int. Ed.
Engl. 1996, 35, 1823–1825; Angew. Chem. 1996, 108, 1959–1962; b) K.
Maeyama, N. Iwasawa, J. Org. Chem. 1999, 64, 1344–1346; c) N. Iwasa-
Received: January 19, 2015
Published online on && &&, 0000
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
11
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Catalysis
Y. Yamamoto,* K. Matsui, M. Shibuya
&& – &&
Polyaromatics: A combination of [h6-(p-
cymene)RuCl2(PPh3)] and two equiva-
lents of AgPF6 effectively converted di-
verse 2-ethynylbiaryls into polyaromat-
ics in chlorobenzene at 1208C for 20 h
(see scheme). Several control experi-
ments and DFT calculations of model
complexes were performed to propose
a plausible reaction mechanism.
A Combined Experimental and
Computational Study on the
Cycloisomerization of 2-Ethynylbiaryls
Catalyzed by Dicationic Arene
Ruthenium Complexes
&
&
Chem. Eur. J. 2015, 21, 1 – 12
www.chemeurj.org
12
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!