Chemo- and Regioselective Rhodium(I)-Catalyzed [2+2+2] Cycloaddition of Allenynes with Alkynes

Article abstract of DOI:10.1002/chem.201602239

A highly chemo- and regioselective partially intramolecular rhodium(I)-catalyzed [2+2+2] cycloaddition of allenynes with alkynes is described. A range of diverse polysubstituted benzene derivatives could be synthesized in good to excellent yields, in which the allenynes served as synthetic equivalent to the diynes. A high regioselectivity could be observed when allenynes were treated with unsymmetrical alkynes.

Full text of DOI:10.1002/chem.201602239

DOI: 10.1002/chem.201602239
Full Paper
&
Regioselectivity
Chemo- and Regioselective Rhodium(I)-Catalyzed [2+2+2]
Cycloaddition of Allenynes with Alkynes
Shigeo Yasuda, Yasuaki Kawaguchi, Yuta Okamoto, and Chisato Mukai*[a]
Abstract: A highly chemo- and regioselective partially intra-
molecular rhodium(I)-catalyzed [2+2+2] cycloaddition of al-
lenynes with alkynes is described. A range of diverse poly-
substituted benzene derivatives could be synthesized in
good to excellent yields, in which the allenynes served as
synthetic equivalent to the diynes. A high regioselectivity
could be observed when allenynes were treated with un-
symmetrical alkynes.
Introduction
us to investigate the cycloaddition reaction between allenynes
(allene–alkyne derivatives)[8] and alkynes. We now describe
some promising results of the rhodium(I)-catalyzed partially in-
tramolecular [2+2+2] cycloaddition of allenynes with al-
kynes[9–15] in which a variety of polysubstituted benzene deriva-
tives could be synthesized in satisfactory yields and with high
regioselectivity (Scheme 2).
The transition-metal-catalyzed [2+2+2] cycloaddition of al-
kynes[1] provides the most straightforward as well as a powerful
methodology for the synthesis of polysubstituted benzene de-
rivatives. It is well known that the cycloaddition of two or
three different alkynes[2] generally suffers from a poor chemo-
and/or regioselectivity. As one of the promising methods to
solve such problems, intramolecular approaches employing
tethered alkynes, diynes, or triynes, have been developed.[1] Al-
though the partially intramolecular cycloaddition of diynes
with the appropriate monoalkynes often led to a high chemo-
selectivity, the control of the regioselectivity still remains an
unsolved issue upon utilizing unsymmetrical alkynes as an ex-
ternal p component.[1] Indeed, diynes A possessing a fairly
bulky R such as trimethylsilyl group at an alkyne terminus un-
derwent the [2+2+2] cycloaddition with unsymmetrical mono-
alkynes B to produce the cycloadducts C with high regioselec-
tivity in many cases. The preferential formation of C over C’
would be attributed to avoidance of the steric hindrance be-
tween a substituent of diyne A and that of B (Scheme 1).[3] This
was, however, not the case in the reactions of A having a less
bulky R with monoalkynes B resulting in nonregioselective pro-
duction of C and C’.[4]
On the other hand, Cheng et al.[5a] reported a new and relia-
ble procedure using the allene functionality instead of an
alkyne as the external p component, in which the nickel cataly-
sis worked well for the [2+2+2] cycloaddition between the
diyne and allene groups, resulting in the formation of polysub-
stituted benzene derivatives in a highly chemo- and regioselec-
tive manner (Scheme 2). Our continuous interests in the field
of rhodium(I)-catalyzed cycloadditions and cycloisomerizations
of allene species with additional p components[6,7] prompted
Results and Discussion
Our initial attempt used the phenylsulfonylallene-alkyne 1a for
screening the reaction conditions. Treatment of 1a and the
symmetrical diester 2a with 10 mol% of the [Rh(cod)2]BF4/
BINAP complex,[16] in 1,2-dichloroethane at 608C for 1 h pro-
duced the tetrahydronaphthalene derivative 3aa in 58% yield
(Table 1, entry 1). Compound 3aa must be formed through 4,
followed by aromatization. In the absence of BINAP, the reac-
tion was less efficient and the yield decreased (Table 1,
entry 2). RhCl(PPh3)3 with or without a silver salt afforded simi-
lar results (Table 1, entries 3 and 4). A neutral RhI complex, ad-
justed from [RhCl(cod)]2 and BINAP, was found to be effective
for our purpose providing 3aa in a quantitative yield (Table 1,
entry 5). The reaction without BINAP was very sluggish and the
yield was rather low (Table 1, entry 6). Other biaryl biphosphine
ligands, such as tol-BINAP, xyl-BINAP, Segphos, DTBM-Segphos,
and BIPHEP, were screened and confirmed to be effective
except for BIPHEP (Table 1, entries 7–11).
The best conditions (5 mol% [RhCl(cod)]2 and 10 mol%
BINAP in 1,2-dichloroethane at 608C)[17] were then used for the
reaction of several allenynes and symmetrical alkynes (Table 2).
The simple carbon tether analogue 1b without the gem-disub-
stituent effect[18] was treated with 2a to afford 3ba in 84%
yield (Table 2, entry 1). The nitrogen congener 1c produced
the aza-compound 3ca in 98% yield (Table 2, entry 2).
[a] Dr. S. Yasuda, Y. Kawaguchi, Y. Okamoto, Prof. Dr. C. Mukai
Division of Pharmaceutical Sciences, Graduate School of Medical Sciences
Kanazawa University, Kakuma-machi, Kanazawa 920-1192 (Japan)
E-mail: mukai@p.kanazawa-u.ac.jp
A phenylsulfonyl substituent on the allenyl moiety was
found not to be essential for this transformation. Indeed, both
allenynes 1d and 1e produced 3da (64%) and 3ea (60%), re-
spectively (Table 2, entries 3 and 4). These results indicate that
the electron-withdrawing groups such as phenylsulfonyl, ethyl
Supporting information for this article can be found under
http://dx.doi.org/10.1002/chem.201602239.
Chem. Eur. J. 2016, 22, 1 – 9
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Scheme 1. Regioselectivity in transition-metal-catalyzed partially intramolecular [2+2+2] cycloaddition between unsymmetrical diyne and alkyne.
Scheme 2. Regioselective partially intramolecular [2+2+2] cycloaddition producing polysubstituted benzene derivative.
Table 1. Optimization of reaction conditions for rhodium(I)-catalyzed
[2+2+2] cycloaddition of allenyne 1a with alkyne 2a.
Table 2. Rhodium(I)-catalyzed [2+2+2] cycloaddition of allenyne 1 with
alkyne 2.
RhI complex
Ligand
t
[h]
Yield
[%][a]
Entry
1
R1
X
Y
2
t
[h]
Yield
[%][a]
Entry
1
2
3
4
1b
1c
1d
1e
1 f
1g
1h Me
1i
1a
H
H
H
H
H
H
CH2
NTs
SO2Ph
SO2Ph
2a
2a
2a
1.5 3ba: 84
0.5 3ca: 98
1.5 3da: 64
1
2
3
[Rh(cod)2]BF4
[Rh(cod)2]BF4
RhCl(PPh3)3
RhCl(PPh3)3
[RhCl(cod)]2
[RhCl(cod)]2
[RhCl(cod)]2
[RhCl(cod)]2
[RhCl(cod)]2
[RhCl(cod)]2
[RhCl(cod)]2
BINAP
–
–
–
BINAP
–
tol-BINAP
xyl-BINAP
Segphos
DTBM-Segphos
BIPHEP
1
1
5
5
0.5
6
1
1
1
58
44[b]
51
62
99
45
98
97
98
83
45
C(CO2Me)2 CO2Me
C(CO2Me)2 P(O)(OEt)2 2a
4[c]
5
2
0.2
7
3ea: 60
[b]
5
C(CO2Me)2 Me
C(CO2Me)2
C(CO2Me)2 SO2Ph
CH2OBn C(CO2Me)2 SO2Ph
C(CO2Me)2 SO2Ph
2a
2a
2a
2a
2b
–
6[c]
7[c]
8[c]
9
H
–
[b]
6[d]
7
2.5 3ha: 72
5.5 3ia: 64
8
9
10
11
H
7
3ab: 54
1.5
24
[a] Yield of the isolated product. [b] An intractable mixture was obtained.
[c] Heated at reflux in 1,2-dichloroethane (DCE).
[a] Yield of the isolated product. [b] Yield was determined by 1H NMR
analysis with CHBr3 as the internal standard. [c] Reaction was carried out
with 20 mol% AgBF4 in benzene. [d] Heated at reflux in DCE. DCE=1,2-
dichloroethane; cod=1,5-cyclooctadiene.
moiety. Thus, the reaction of the methyl derivative 1 f and 2a
was carried out under the standard conditions to furnish an in-
tractable mixture (Table 2, entry 5). The unsubstituted substrate
1g (Y=H) also gave the similar result upon treating with 2a
(Table 2, entry 6). As a result, it might be concluded that the
electron-withdrawing group is essential for this [2+2+2] cyclo-
addition. Although higher reaction temperatures were needed,
the allene-internal alkynes 1h and 1i afforded the correspond-
ing ring-closed products 3ha (72%) and 3ia (64%; Table 2, en-
tries 7 and 8). Bis(benzyloxymethyl)acetylene (2b) was treated
with 1a to afford 3ab in 54% yield (Table 2, entry 9). In sharp
contrast to the result of the 1,1-disubstituted allenes 1a–1i,
phosphonate, and methoxycarbonyl groups on the allenyl
moiety served as a suitable substituent. We further examined
the cycloaddition of two additional substrates 1 f (Y=Me) and
1g (Y=H) to see the effect of the substituent on the allenyl
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the allenyne 5 having a methyl group at the allenic terminus
(1,1,3-trisubstituted allene) did not yield the desired product at
all, but the benzene derivative 7 in 78% yield, which should
have arisen from the cycloaddition between the alkyne moiety
of 5 and two molecules of the external alkyne 2a [Eq. (1)].
The rhodium(I)-catalyzed [2+2+2] cycloaddition of allenynes
with unsymmetrical alkynes was the next subject (Table 3).
Treatment of 1a with ethyl phenylpropiolate (8a) under the
standard conditions provided a mixture of two tetrahydro-
naphthalene derivatives 9aa/10aa in 53% yield in the ratio of
62 to 38 (Table 3, entry 1).[19] The alkyl analogue 8b showed
a better reactivity and regioselectivity furnishing 9ab/10ab
(82:18 mixture) in 85% yield (Table 3, entry 2). The terminal al-
kynes 8c and 8d having ester functionality and simple alkyl
group both reacted with 1a to afford the cycloadducts 9ac/
10ac (84%, 87:13 mixture) and 9ad/10ad (72%, 89:11 mix-
ture), respectively (Table 3, entries 3 and 4). On the other hand,
the reaction of the allene-internal alkyne substrate 1h with 8b
produced a mixture of 9hb/10hb in 88% NMR yield in a ratio
of 88 to 12 (Table 3, entry 5 vs. entry 2),[20] a slightly better re-
gioselectivity compared to that of 9ab/10ab (82:18; Table 3,
entry 2). Furthermore, the reaction of 1h with 8c smoothly
proceeded to exclusively furnish 9hc in 90% yield (Table 3,
entry 6). Similarly, a high regioselectivity was observed in the
reactions of 1h (93:7), 1j (91:9) with 8d (Table 3, entries 7 and
8). Thus, the preferential production of 9 over 10 was consis-
tently recorded. In particular, the results of entries 3, 4 and 6–8
(Table 3) obviously indicated that higher regioselectivity was
attained with terminal alkynes.
It should be emphasized that the reverse regioselectivity
was observed when the propargyl alcohol derivative 11 a was
treated with 1a to provide a mixture of 12aa/13aa in 85%
yield in the ratio of 91:9 (Scheme 3). The major product 12aa
has a benzyloxymethyl group at the p-position to the phenyl-
sulfonylmethine residue. This is obviously different from the re-
sults in Table 3, in which cycloadducts having the substituent
at the m-position to the phenylsulfonylmethine residue were
consistently obtained as the major product. Furthermore, prop-
argyl alcohol (11 b) itself exhibited a similar behavior, exclusive-
ly forming 12ab in 94% yield.
Table 3. Rhodium(I)-catalyzed [2+2+2] cycloaddition of allenyne 1 with unsymmetrical alkyne 8.
Entry
1
R1
8
R2
R3
t
Yield [%] (9+10)[a]
Rs (9/10)[b]
[h]
1[c]
2[c]
3
1a
1a
1a
1a
1h
1h
1h
1j
H
H
H
H
8a
8b
8c
8d
8b
8c
8d
8d
CO2Et
CO2Et
CO2Me
nBu
CO2Et
CO2Me
nBu
Ph
Me
H
H
Me
H
11
4
1
1.5
6
0.5
11
7
53 (9aa+10aa)
85 (9ab+10ab)
84 (9ac+10ac)
72 (9ad+10ad)
88 (9hb+10hb)[d]
90 (9hc)
62:38
82:18
87:13
89:11
88:12
9hc only
93:7
4
5[c]
6
Me
Me
Me
CH2OH
7
8[c]
H
H
72 (9hd+10hd)
70 (9jd+10jd)
nBu
91:9
1
[a] Yield of the mixture of 9+10. [b] Regioselectivity (Rs) determined by H NMR analysis of the mixture of 9+10. [c] Heated at reflux in DCE. [d] Yield was
1
determined by H NMR analysis with CHBr3 as the internal standard.
Scheme 3. Rhodium(I)-catalyzed [2+2+2] cycloaddition of allenyne 1a with propargyl alcohol (11 b) and its derivative 11 a.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Several propargyl alcohol analogues were examined to de-
termine the scope and limitations (Table 4). Treatment of 1a
with homopropargyl alcohol (11 c) produced 12ac in a highly
regioselective manner (12ac/13ac=94:6) and 90% yield
(Table 4, entry 1). The two-carbon homologated substrate 11 d
again regioselectively furnished 12ad (12ad/13ad=93:7;
Table 4, entry 2). The secondary and tertiary propargyl alcohols
11 e and 11 f did not disturb the reaction and selectively pro-
duced the cyclized products 12ae/13ae (96:4)[21] and 12af/
13af (83:17; Table 4, entries 3 and 4). Similarly, the exclusive
production of the nitrogen analogue 12ag was achieved in
a quantitative yield (Table 4, entry 5). The reaction of 2-butyn-
1-ol (11 h) proceeded in line with the preferential formation of
12ah over 13ah (75:25), but a prolonged reaction time (4 h)
was required and the selectivity decreased (Table 4, entry 6).
into the Csp3ÀRh bond resulting in the formation of the two
other intermediates III and III’. These four intermediates could
subsequently collapse to the final tetrahydronaphthalenes
through reductive elimination, followed by aromatization. If
the reaction proceeded through the insertion of alkyne into
the Csp3ÀRh bond, the observed complementary regioselectivi-
ty can be reasonably explained as follows. Alkyne derivatives 8
would approach to the Csp3ÀRh bond of the intermediate I in
which sterically less hindered substituent (RS) on the triple
bond should be taken a place close to Rh atom (coordinating
with biphosphine ligand) as described in II to avoid a nonbond-
ing interaction resulting in the formation of the seven-mem-
bered rhodacycle III. Consecutive reductive elimination of III
and aromatization would then lead to the formation of 9. The
opposite approach of 8 to the Csp3ÀRh bond of I (i.e., II’), how-
ever, would suffer from unfavorable steric repulsion between
RL and biphosphine ligand on Rh. As a result, the reaction
pathway through the intermediates II’ and III’ must be the un-
favorable and minor one, thereby the alternative reaction
route through II and III would become more favored route.
The situation must be drastically changed when alkyne deriva-
tives 11 possessing hydroxyl functionality were used. Prior to
direct insertion of the triple bond of 11 into the Csp3ÀRh bond
of I, it is reasonable to consider coordination of Rh atom of I
with hydroxyl functionality of 11 to form VI. Thus, the insertion
of the triple bond of 11 would subsequently occur from the in-
termediate VI to afford the seven-membered rhodacycle VII. Fi-
nally, reductive elimination of VII was followed by aromatiza-
tion to afford 12 in a highly regioselective manner. Although
the possibility of the route that involves the intermediacy of V
and V’ through IV and IV’, respectively, cannot be completely
ruled out, we now do not have good and clear explanation of
the observed complementary regioselectivity. In the reaction
of the trisubstituted allene species 5, the formation of the de-
sired tetrahydronaphthalene derivative 6 could not be ob-
served. The benzene derivative 7 was obtained instead (see
above). This result might be rationalized by considering that
the key rhodabicyclic intermediate such as I could not be so
easily formed because of the sterically congested trisubstituted
allene functionality and the conventional [2+2+2] cycloaddi-
tion of three alkyne components preferentially occurred. The
similar interpretation for the by-production of the benzene de-
rivative 17’ can be made (see below). The formation of seven-
membered ring is known to be more difficult than the six-
membered one. Thus, the formation of seven-membered rho-
dabicyclic intermediate (one-carbon homologated I) was re-
tarded in part and the desired allene-two alkynes [2+2+2] cy-
cloaddition reaction became the minor pathway.
Table 4. Rhodium(I)-catalyzed [2+2+2] cycloaddition of allenyne 1a with
alkyne 11.
Entry 11
R1
R2
t
Yield [%]
Rs
(9/10)[b]
[h] (12+13)[a]
1
2
11 c
(CH2)2OH
H
H
1
1
90 (12ac+13ac)
88 (12ad+13ad)
94:6
93:7
11 d (CH2)3OH
3
4
11 e
11 f
H
H
1
95 (12ae+13ae)[c] 96:4
1.5 86 (12af+13af)
83:17
5
6
11 g CH2NHTs
11 h CH2OH
H
Me
1
4
quant. (12ag)
71 (12ah+13ah)
12ag only
75:25
[a] Yield of the mixture of 12+13. [b] Determined by 1H NMR analysis of
the mixture of 12+13. [c] Compounds 12ae and 13ae were obtained as
a 1:1 mixture of diastereomers.
The regioselectivity in the [2+2+2] cycloaddition of alle-
nynes with unsymmetrical alkynes could be highly controlled
by proper choice of alkynes (with or without a hydroxyl
group). Treatment of 1 with unsymmetrical alkynes 8 without
a hydroxyl functionality predominantly produced 9 (Table 3),
whereas the reaction with unsymmetrical alkynes 11 having
a hydroxyl group (e.g., propargyl alcohol) furnished 12 in
a highly reverse regioselective manner (Scheme 3 and Table 4).
We tentatively interpreted these experimental results based on
the following plausible mechanism (Scheme 4). The reaction
would initiate the formation of the rhodabicyclic intermediate
I,[7e,f,j] which was postulated as a key intermediate in the most
cases of our previous works.[7e,f,j] There might be two possible
pathways for the following insertion process of the external
alkyne 8. One is the insertion into the Csp2ÀRh bond leading to
the two intermediates V and V’ and the other is the insertion
Some additional experiments were carried out to reveal the
scope of this method (Scheme 5). Exposure of the one-carbon
shortened substrate 14 and 2a to the standard conditions pro-
vided the indane derivative 15 in 78% yield. Construction of
the larger-sized 17 (67%) could also be attained along with
the formation of the benzene derivative 17’ (15%), the latter
of which should have been derived from the cycloaddition be-
tween 16 and two molecules of 2a, similar to the formation of
7 from 5.
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 4. Explanation for the observed regioselectivity on the basis of the plausible mechanism.
A phenylsulfonyl group can be easily removed by the con-
ventional procedure[22] to afford 18 in 72% yield when com-
pound 3ba was exposed to Raney nickel in ethanol[22c] heated
at reflux [Eq. (2)]. Exposure of 3ba to Friedel–Crafts alkylation
conditions in benzene effected substitution of a phenyl group
for a phenylsulfonyl moiety to afford 19 in 80% yield
[Eq. (3)].[23]
The allenynes could be regarded as the synthetic equivalent
of the corresponding diynes (proton tautomer of propargyl
moiety). As indicated in Table 4, the ring-closing reactions of al-
lenynes with propargyl alcohol derivatives proceeded in
a highly regioselective manner to produce the tetrahydronaph-
thalene derivatives 12. It would be of particular interest from
a synthetic point of view to see how the reaction proceeds
Scheme 5. Synthesis of indane derivative 15 and benzocycloheptane derivative 17.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
ing to the efficient formation of polysubstituted benzene-fused
bicyclic derivatives. The combination of allenynes with alkynes,
especially with propargylic alcohols, was found to produce the
desired products with an excellent regioselectivity as well as
high chemical yields. This is not the case of the corresponding
diyne, a synthetic equivalent of the allenyne, affording the tet-
rahydronaphthalenes in a high yield but in a nonregioselective
way. Thus, we could demonstrate the superiority of the alle-
nyne in place of the diyne as the p component in the [2+2+2]
cycloaddition. We are currently investigating the scope and
limitations of this method as well as the development of other
[2+2+2] cycloadditions utilizing allene species.
with diynes instead of allenynes. Thus, the reaction of the alle-
nyne 20 with propargyl alcohol (11 b) under the standard con-
ditions furnished 21 in a highly regioselective manner (21/22=
93:7) and 92% yield, which was then transformed into the de-
sulfonylated one 25 in 69% yield (59% overall yield from 20;
Scheme 6). Similar treatment of the corresponding diyne 27
with 11 b in the presence of H8-BINAP[16] nonregioselectively
produced a mixture of 25 and its regioisomer 28 (25/28=
50:50) in 93% yield. Another control experiment using ester-
substituted alkyne 8c instead of 11 b was conducted. Treat-
ment of 20 with 8c resulted in the preferential formation of
24 over 23 (87%, 23/24=24:76), which was subsequently de-
sulfonylated to afford 26 in 72% yield (48% overall yield from
20). On the other hand, treatment of 27 with 8c provided
a 1:1 mixture of 29 and 26 in 66% yield. These experiments
would strongly indicate the superiority of the allenyne instead
of the diyne in terms of regioselectivity.
Experimental Section
General methods
Melting points were measured with YANAGIMOTO micro melting
point apparatus, and are uncorrected. Infrared spectra were mea-
sured with a SHIMADZU FTIR-8700 spectrometer for samples in
1
CHCl3. H NMR spectra were measured with JNM-ECS400 or JNM-
ECA600 spectrometers for samples in [D]-chloroform (CDCl3), using
either tetramethylsilane (for compound with a phenyl group, d=
0.00 ppm), CHCl3 (d=7.26 ppm) as an internal reference. 13C NMR
spectra were measured with JNM-ECS400 or JNM-ECA600 spec-
trometers for samples in CDCl3 (d=77.0 ppm) as an internal refer-
ence. High-resolution mass spectra were measured with JMS-
T100TD (DART) mass spectrometers, and mass spectra were mea-
sured with JMS-T100TD (DART) mass spectrometers. Commercially
available [RhCl(cod)]2 (Kanto Chemical) and [Rh(cod)2]BF4 (Wako
Pure Chemical Industry) were employed for reactions. Silica gel
(Silica gel 60 N, 40–50 mm, Kanto Chemical) was used for chroma-
tography. All reactions were carried out under N2 atmosphere. Or-
ganic extracts were dried over Na2SO4. Synthetic procedure and
characterization data of allenynes (1, 5, 14, 16, 20) and diyne 27
are summarized in Supporting Information. Characterization data
Conclusions
We have developed a highly chemo- and regioselective new
type of [2+2+2] cycloaddition of allenynes with alkynes lead-
Scheme 6. Control experiments using allenyne 20 and the corresponding diyne 27.
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
of the products of the [2+2+2] cycloaddition other than 3aa, 9ab,
9ac, 9 fc, and 12ab are summarized in the Supporting Informa-
tion.
6-(Hydroxymethyl)-3,3-bis(methoxycarbonyl)-1-(phenylsulfonyl)-
1,2,3,4-tetrahydronaphthalene (12ab): Colorless powder. M.p.
147–1498C (hexane/AcOEt); 1H NMR (600 MHz, CDCl3): d=7.63–
7.60 (m, 1H), 7.56–7.55 (m, 2H), 7.46–7.40 (m, 3H), 7.18–7.17 (m,
1H), 7.05 (s, 1H), 4.77 (dd, 1H, J=9.6, 7.2 Hz), 4.67 (s, 2H), 3.77 (s,
3H), 3.45 (s, 3H), 3.12 (dd, 1H, J=15.3, 2.9 Hz), 3.00 (ddd, 1H, J=
14.8, 9.6, 2.9 Hz), 2.68 (d, 1H, J=15.3 Hz), 2.41 (dd, 1H, J=14.8,
7.2 Hz), 1.75 ppm (brs, 1H); 13C NMR (151 MHz, CDCl3): d=170.9,
170.1, 141.7, 136.1, 136.0, 133.9, 131.3, 129.5, 128.9, 127.2, 125.5,
Representative procedure for rhodium(I)-catalyzed [2+2+2]
cycloaddition of allenynes with alkynes
[RhCl(cod)]2 (1.2 mg, 2.5 mmol) and rac-BINAP (3.1 mg, 5.0 mmol)
were dissolved in DCE (0.20 mL), and the mixture was heated at
608C for 0.5 h under an atmosphere of Ar. A solution of the alle-
nyne 1a (18 mg, 0.05 mmol) and the alkyne 2a (36 mg, 0.25 mmol)
in DCE (0.80 mL) was added to the reaction mixture, which was
stirred at 608C for 0.5 h. The residue was chromatographed with
hexane/AcOEt to afford 3aa (25 mg, 99%).
~
125.3, 64.6, 63.5, 54.5, 53.2, 52.8, 35.2, 29.7 ppm; IR: n=3603, 1734,
1308, 1146 cmÀ1; DART MS: m/z 419 (M+ +1, 14.5); DART HRMS:
m/z calcd for C21H23O7S: 419.1165; found: 419.1161.
Representative procedure for desulfonylation of 1,2,3,4-tet-
rahydronaphthalenes
3,3,6,7-Tetrakis(methoxycarbonyl)-1-(phenylsulfonyl)-1,2,3,4-tet-
rahydronaphthalene (3aa): Colorless plate. M.p. 178–1818C
(hexane/AcOEt); 1H NMR (600 MHz, CDCl3): d=7.67–7.64 (m, 2H),
7.59–7.57 (m, 2H), 7.50–7.48 (m, 2H), 7.40 (s, 1H), 4.77 (dd, 1H, J=
9.6, 7.0 Hz), 3.90 (s, 3H), 3.87 (s, 3H), 3.79 (s, 3H), 3.47 (s, 3H), 3.21
(dd, 1H, J=15.6, 2.9 Hz), 3.04 (ddd, 1H, J=14.8, 9.6, 2.9 Hz), 2.81
(d, 1H, J=15.6 Hz), 2.49 ppm (dd, 1H, J=14.8, 7.0 Hz); 13C NMR
(151 MHz, CDCl3): d=170.4, 169.7, 167.5, 166.9, 139.6, 135.6, 134.2,
132.3, 131.6, 130.1, 129.9, 129.5, 129.2, 63.3, 54.2, 53.4, 53.1, 52.8,
An excess amount of Raney Ni was added to a solution of 3ba
(19 mg, 0.05 mmol) in EtOH (1.0 mL). The reaction was conducted
at reflux condition with vigorous stirring for 36 h. Then, the reac-
tion mixture was cooled to room temperature and filtered through
a pad of Celite. The filtrate was quenched with brine, and then ex-
tracted with CH2Cl2. The extract was washed with water and brine,
dried and concentrated to dryness. The residue was chromato-
graphed on silica gel with hexane/AcOEt to afford 18 (8.9 mg,
72%).
52.7, 35.0, 29.2 ppm; IR: n=1736, 1310, 1148 cmÀ1; DART MS m/z
~
505 (M+ +1, 100); DART HRMS: m/z calcd for C24H25O10S: 505.1168;
6,7-Bis(methyloxycarbonyl)-1,2,3,4-tetrahydronaphthalene (18):
Colorless oil. 1H NMR (600 MHz, CDCl3): d=7.42 (s, 2H), 3.88 (s,
6H), 2.81–2.79 (m, 4H), 1.82–1.80 ppm (m, 4H); 13C NMR (151 MHz,
found: 505.1158.
7-(Ethoxycarbonyl)-3,3-bis(methoxycarbonyl)-6-methyl-1-(Phe-
nylsulfonyl)-1,2,3,4-tetrahydronaphthalene (9ab): Colorless amor-
phous solid. 1H NMR (600 MHz, CDCl3): d=7.70 (s, 1H), 7.64–7.61
(m, 1H), 7.59–7.57 (m, 2H), 7.48–7.45 (m, 2H), 6.94 (s, 1H), 4.73
(dd, 1H, J=10.0, 6.9 Hz), 4.33–4.24 (m, 2H), 3.78 (s, 3H), 3.46 (s,
3H), 3.14 (dd, 1H, J=15.5, 3.1 Hz), 3.04 (ddd, 1H, J=15.1, 10.0,
3.1 Hz), 2.81 (d, 1H, J=15.5 Hz), 2.55 (s, 3H), 2.52 (dd, 1H, J=15.1,
6.9 Hz), 1.35 ppm (t, 3H, J=7.2 Hz); 13C NMR (151 MHz, CDCl3): d=
170.7, 170.0, 166.7, 140.9, 139.9, 135.9, 134.0, 133.2, 132.3, 129.6,
128.9, 128.5, 124.0, 63.1, 60.7, 54.4, 53.2, 52.9, 35.0, 29.1, 21.4,
~
~
CDCl3): d=168.3, 140.8, 129.7, 128.9, 52.5, 29.2, 22.6 ppm; IR: n=
1720 cmÀ1; DART MS m/z 249 (M+ +1, 100); DART HRMS: m/z
calcd for C14H17O4: 249.1127; found: 249.1116.
Transformation of 3ba by Friedel–Crafts alkylation [Eq. (3)]
7,8-Bis(methoxycarbonyl)-1-phenyl-1,2,3,4-tetrahydronaph-
thalene (19)
AlCl3 (20 mg, 0.15 mmol) was added to a solution of 3ba (19 mg,
0.050 mmol) in benzene (1.0 mL) at room temperature. After being
stirred for 6 h at the same temperature, the reaction was quenched
by addition of 1m aqueous NaOH and the mixture was extracted
with AcOEt. The extract was washed with water and brine, dried
and concentrated to dryness. The residue was chromatographed
with hexane/AcOEt (7:1) to afford 19 (13 mg, 80%) as a colorless
14.3 ppm; IR: n=1735, 1306, 1143 cmÀ1; DART MS m/z 475 (M+
+
1, 81.9); DART HRMS: m/z calcd for C24H27O8S: 475.1427; found:
475.1414.
3,3,7-Tris(methoxycarbonyl)-1-(phenylsulfonyl)-1,2,3,4-tetrahy-
dronaphthalene (9ac): Colorless needle. M.p. 171–1738C (hexane/
1
AcOEt); H NMR (600 MHz, CDCl3): d=7.91 (dd, 1H, J=7.9, 1.7 Hz),
7.88 (s, 1H), 7.65–7.62 (m, 1H), 7.56–7.54 (m, 2H), 7.47–7.44 (m,
2H), 7.14 (d, 1H, J=7.9 Hz), 4.76 (dd, 1H, J=10.0, 6.5 Hz), 3.88 (s,
3H), 3.78 (s, 3H), 3.43 (s, 3H), 3.20 (dd, 1H, J=15.5, 3.1 Hz), 3.05
(ddd, 1H, J=14.8, 10.0, 3.1 Hz), 2.80 (d, 1H, J=15.5 Hz), 2.53 ppm
(dd, 1H, J=14.8, 6.5 Hz); 13C NMR (151 MHz, CDCl3): d=170.6,
169.9, 166.3, 141.3, 135.8, 134.1, 132.3, 129.9, 129.5, 129.3, 129.0,
1
amorphous solid: H NMR (600 MHz, CDCl3): d=7.50 (s, 1H), 7.30–
7.27 (m, 2H), 7.25 (s, 1H), 7.23–7.20 (m, 1H), 7.04–7.03 (m, 2H),
4.14 (t, 1H, J=6.2 Hz), 3.89 (s, 3H), 3.79 (s, 3H), 2.95 (dt, 1H, J=
17.5, 6.5 Hz), 2.87 (dt, 1H, J=17.5, 5.8 Hz), 2.18–2.14 (m, 1H), 1.92–
1.86 (m, 2H), 1.79–1.72 ppm (m, 1H); 13C NMR: d=168.4, 168.0,
146.0, 142.8, 141.4, 130.9, 129.6, 129.5, 129.0, 128.7, 128.5, 126.3,
~
127.0, 63.6, 54.5, 53.3, 52.8, 52.1, 35.4, 29.3 ppm; IR: n=1721, 1294,
52.5, 52.4, 45.3, 32.7, 29.6, 20.1 ppm; IR: n=1721 cmÀ1; DART MS
~
1143 cmÀ1; DART MS m/z 447 (M+ +1, 100); DART HRMS: m/z calcd
m/z 325 (M+ +1, 100); DART HRMS m/z calcd for C20H21O4:
for C22H23O8S: 447.1114; found: 447.1115.
325.1440; found: 325.1433.
3,3,7-Tris(methoxycarbonyl)-5-methyl-1-(phenylsulfonyl)-1,2,3,4-
tetrahydronaphthalene (9 fc): Colorless powder. M.p. 1618C
(hexane/AcOEt); 1H NMR (600 MHz, CDCl3): d=7.79 (s, 1H), 7.65–
7.62 (m, 2H), 7.54–7.53 (m, 2H), 7.47–7.44 (m, 2H), 4.74 (dd, 1H,
J=10.0, 6.2 Hz), 3.86 (s, 3H), 3.79 (s, 3H), 3.42 (s, 3H), 3.37 (dd, 1H,
J=15.8, 2.7 Hz), 3.05 (ddd, 1H, J=14.8, 10.0, 2.7 Hz), 2.58 (dd, 1H,
J=14.8, 6.2 Hz), 2.55 (d, 1H, J=15.8 Hz), 2.28 ppm (s, 3H); 13C NMR
(151 MHz, CDCl3): d=170.8, 170.0, 166.5, 140.3, 136.7, 135.8, 134.0,
131.1, 130.2, 129.5, 128.9, 128.0, 126.8, 64.0, 54.5, 53.3, 52.8, 52.1,
Acknowledgements
This work was financially supported by JSPS KAKENHI Grant
Numbers 15H02490 and 15K18826, for which we are thankful.
Keywords: alkynes · allenes · cycloaddition · regioselectivity ·
rhodium
31.4, 28.9, 19.5 ppm; IR: n=1734, 1308, 1148 cmÀ1; DART MS: m/z
~
461 (M+ +1, 4.15); DART HRMS: m/z calcd for C23H25O8S: 461.1270;
found: 461.1287.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[1] For selected recent reviews, see: a) Transition-Metal-Mediated Aromatic
Ring Construction (Ed.: K. Tanaka), Wiley, Hoboken, 2013, chap. 1–11;
b) D. L. J. Broere, E. Ruijter, Synthesis 2012, 2639–2672; c) Y. Shibata, K.
Tanaka, Synthesis 2012, 323–350; d) N. Weding, M. Hapke, Chem. Soc.
Rev. 2011, 40, 4525–4538; e) K. Tanaka, Chem. Asian J. 2009, 4, 508–
518; f) T. Shibata, K. Tsuchikama, Org. Biomol. Chem. 2008, 6, 1317–
1323.
[10] For nickel-catalyzed partially intramolecular [2+2+2] cycloaddition of
allene–alkene with alkene, see: N. N. Noucti, E. J. Alexanian, Angew.
Chem. Int. Ed. 2013, 52, 8424–8427; Angew. Chem. 2013, 125, 8582–
8585.
[11] Rhodium(I)-catalyzed partially intramolecular [2+2+2] cycloaddition of
allene–allene with allene was also reported, see: P. Lu, S. Ma, Org. Lett.
2007, 9, 5319–5321.
[2] a) N. Mori, S. Ikeda, K. Odashima, Chem. Commun. 2001, 181–182; b) Y.
Ura, Y. Sato, H. Tsujita, T. Kondo, M. Imachi, T. Mitsudo, J. Mol. Catal. A
2005, 239, 166–171; For a stoichiometric reaction, see: c) T. Takahashi,
Z. Xi, A. Yamazaki, Y. Liu, K. Nakajima, M. Kotora, J. Am. Chem. Soc. 1998,
120, 1672–1680.
[3] For selected examples, see: a) B. Witulski, T. Stengel, Angew. Chem. Int.
Ed. 1999, 38, 2426–2430; Angew. Chem. 1999, 111, 2521–2524; b) B. Wi-
tulski, C. Alayrac, Angew. Chem. Int. Ed. 2002, 41, 3281–3284; Angew.
Chem. 2002, 114, 3415–3418; c) T. Konno, K. Moriyasu, R. Kinugawa, T.
Ishihara, Org. Biomol. Chem. 2010, 8, 1718–1724; d) M. A. Zotova, D. V.
Vorobyeva, P. H. Dixneuf, C. Bruneau, S. N. Osipov, Synlett 2013, 1517–
1522; e) R. W. Foster, C. J. Tame, H. C. Hailes, T. D. Sheppard, Adv. Synth.
Catal. 2013, 355, 2353–2360.
[12] For intramolecular [2+2+2] cycloaddition of allene–alkyne–alkyne with
a stoichiometric amount of Co complex, see: a) D. Llerena, O. Buisine, C.
Aubert, M. Malacria, Tetrahedron 1998, 54, 9373–9392; b) M. Petit, C.
Aubert, M. Malacria, Tetrahedron 2006, 62, 10582–10593; For rhodiu-
m(I)-catalyzed reaction of alkyne–alkyne–allene, see: c) B. Bennacer, M.
Fujiwara, S.-Y. Lee, I. Ojima, J. Am. Chem. Soc. 2005, 127, 17756–17767.
[13] For other examples of transition-metal-catalyzed intramolecular
[2+2+2] cycloaddition of allene compounds, see: a) N. Saito, T. Ichi-
maru, Y. Sato, Chem. Asian J. 2012, 7, 1521–1523; b) Y. Oonishi, Y.
Kitano, Y. Sato, Tetrahedron 2013, 69, 7713–7718; c) E. Haraburda, ꢃ.
Torres, T. Parella, M. Solꢄ, A. Pla-Quintana, Chem. Eur. J. 2014, 20, 5034–
5045; d) E. Haraburda, A. Lledꢂ, A. Roglans, A. Pla-Quintana, Org. Lett.
2015, 17, 2882–2885.
[4] For regioselective partially intramolecular [2+2+2] cycloaddition be-
tween unsymmetrical diynes without bulky substituent at their alkyne
terminus and monoalkynes, see: a) B. Witulski, T. Stengel, J. M. Fernꢁn-
dez-Hernꢁndez, Chem. Commun. 2000, 1965–1966; b) Y. Yamamoto, T.
Arakawa, R. Ogawa, K. Itoh, J. Am. Chem. Soc. 2003, 125, 12143–12160.
[5] a) M. Shanmugasundaram, M.-S. Wu, M. Jeganmohan, C.-W. Huang, C.-
H. Cheng, J. Org. Chem. 2002, 67, 7724–7729; For related studies, see:
b) M. Shanmugasundaram, M.-S. Wu, C.-H. Cheng, Org. Lett. 2001, 3,
4233–4236; c) J.-C. Hsieh, D. K. Rayabarapu, C.-H. Cheng, Chem.
Commun. 2004, 532–533.
[6] For recent reviews on transition-metal-catalyzed cycloaddition or cycloi-
somerization of allene compounds, see: a) G. Chen, X. Jiang, C. Fu, S.
Ma, Chem. Lett. 2010, 39, 78–81; b) F. Lꢂpez, J. L. MascareÇas, Chem.
Eur. J. 2011, 17, 418–428; c) C. Aubert, L. Fensterbank, P. Garcia, M. Mal-
acria, A. Simonneau, Chem. Rev. 2011, 111, 1954–1993; d) F. Inagaki, S.
Kitagaki, C. Mukai, Synlett 2011, 594–614.
[7] a) C. Mukai, I. Nomura, K. Yamanishi, M. Hanaoka, Org. Lett. 2002, 4,
1755–1758; b) F. Inagaki, C. Mukai, Org. Lett. 2006, 8, 1217–1220; c) F.
Inagaki, S. Narita, T. Hasegawa, S. Kitagaki, C. Mukai, Angew. Chem. Int.
Ed. 2009, 48, 2007–2011; Angew. Chem. 2009, 121, 2041–2045; d) F. In-
agaki, K. Sugikubo, Y. Miyashita, C. Mukai, Angew. Chem. Int. Ed. 2010,
49, 2206–2210; Angew. Chem. 2010, 122, 2252–2256; e) F. Inagaki, K.
Sugikubo, Y. Oura, C. Mukai, Chem. Eur. J. 2011, 17, 9062–9065; f) C.
Mukai, Y. Ohta, Y. Oura, Y. Kawaguchi, F. Inagaki, J. Am. Chem. Soc. 2012,
134, 19580–19583; g) T. Iwata, F. Inagaki, C. Mukai, Angew. Chem. Int.
Ed. 2013, 52, 11138–11142; Angew. Chem. 2013, 125, 11344–11348;
h) K. Sugikubo, F. Omachi, Y. Miyanaga, F. Inagaki, C. Matsumoto, C.
Mukai, Angew. Chem. Int. Ed. 2013, 52, 11369–11372; Angew. Chem.
2013, 125, 11579–11582; i) Y. Kawaguchi, S. Yasuda, A. Kaneko, Y. Oura,
C. Mukai, Angew. Chem. Int. Ed. 2014, 53, 7608–7612; Angew. Chem.
2014, 126, 7738–7742; j) Y. Ohta, S. Yasuda, Y. Yokogawa, K. Kurokawa,
C. Mukai, Angew. Chem. Int. Ed. 2015, 54, 1240–1244; Angew. Chem.
2015, 127, 1256–1260.
[14] For nickel-catalyzed intermolecular [2+2+2] cycloaddition of allene
with isocyanate, see: T. Miura, M. Morimoto, M. Murakami, J. Am. Chem.
Soc. 2010, 132, 15836–15838.
[15] Only one example of the partially intramolecular [2+2+2] cycloaddition
of an allenyne with an alkyne was reported by Ma and co-workers, see
reference [8c].
[16] [Rh(cod)2]BF4/BINAP and [Rh(cod)2]BF4/H8-BINAP are known as effective
catalysts for the various [2+2+2] cycloadditions of diynes with monoal-
kynes, see reference [1c,d].
[17] We chose BINAP as the best phosphine ligand for the following experi-
ments because of its highest yield as well as easy availability.
[18] M. E. Jung, G. Piizzi, Chem. Rev. 2005, 105, 1735–1766.
[19] The regiochemistry of 9aa, 9ab, 9ac, 9hb, 9hd, and 12ab were deter-
mined based on the NOE experiments (see the Supporting Information
for details). The regiochemistry of the other cyclized products obtained
by the reactions with unsymmetrical alkynes were determined by com-
parison of 1H NMR spectra of them with those of 9ac, 9hd, and 12ab.
[20] Yield of 9hb/10hb was determined by 1H NMR analysis because 9hb/
10hb could not be purified effectively. Compound 9hb/10hb was
therefore obtained as a mixture with a slight amount of some impuri-
ties.
[21] Ratio of 12ae to 13ae was determined by 1H NMR analysis of the mix-
ture of their oxidized derivatives 30 and 31, which were obtained by
a standard oxidation condition using 2-iodoxybenzoic acid (IBX) as an
oxidant (see the Supporting Information for details).
[8] For selected examples of rhodium(I)-catalyzed cycloisomerization or cy-
cloaddition of allenynes, see: a) K. M. Brummond, H. Chen, P. Sill, L. You,
J. Am. Chem. Soc. 2002, 124, 15186–15187; b) T. Shibata, Y. Takesue, S.
Kadowaki, K. Takagi, Synlett 2003, 268–270; c) X. Jiang, S. Ma, J. Am.
Chem. Soc. 2007, 129, 11600–11607; d) Y. Oonishi, A. Hosotani, Y. Sato,
J. Am. Chem. Soc. 2011, 133, 10386–10389; e) Y. Oonishi, Y. kitano, Y.
Sato, Angew. Chem. Int. Ed. 2012, 51, 7305–7308; Angew. Chem. 2012,
124, 7417–7420; f) Y. Oonishi, T. Yokoe, A. Hosotani, Y. Sato, Angew.
Chem. Int. Ed. 2014, 53, 1135–1139; Angew. Chem. 2014, 126, 1153–
1157.
[22] For reviews on the transformation of sulfones, see: a) C. Nꢁjera, M. Yus,
Tetrahedron 1999, 55, 10547–10658; b) N. S. Simpkins, Sulfones in Or-
ganic Synthesis, Pergamon, Oxford, 1993; For desulfonylation of the
compounds, which are structurally similar to 3ba, using Raney nickel,
see: c) T. Zhang, X. Huang, J. Xue, S. Sun, Tetrahedron Lett. 2009, 50,
1290–1294.
[23] For similar Friedel–Crafts alkylation with sulfone, see: T. Niwa, H. Yori-
mitsu, K. Oshima, Tetrahedron 2009, 65, 1971–1976.
[9] Rhodium(I)-catalyzed partially intramolecular [2+2+2] cycloaddition of
allene–alkene with allene was reported, see: a) A. T. Brusoe, E. J. Alexani-
an, Angew. Chem. Int. Ed. 2011, 50, 6596–6600; Angew. Chem. 2011,
123, 6726–6730; For asymmetric version, see: b) A. T. Brusoe, R. V. Ed-
wankar, E. J. Alexanian, Org. Lett. 2012, 14, 6096–6099.
Received: May 11, 2016
Published online on && &&, 2016
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Regioselectivity
S. Yasuda, Y. Kawaguchi, Y. Okamoto,
C. Mukai*
&& – &&
Chemo- and Regioselective
Rhodium(I)-Catalyzed [2+2+2]
Cycloaddition of Allenynes with
Alkynes
Under control: A highly chemo- and re-
gioselective partially intramolecular rho-
dium(I)-catalyzed [2+2+2] cycloaddition
of allenynes with alkynes is described
(see scheme). A range of diverse poly-
substituted benzene derivatives could
be synthesized in good to excellent
yields, in which the allenynes served as
synthetic equivalent to the diynes. A
high regioselectivity could be observed
when allenynes were treated with un-
symmetrical alkynes.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ