Communication
Table 1. Optimization of the benzo/[7+1] reaction conditions.[a]
Entry
CO [atm]
x [mol%]
Solvent
T [8C]
t [h]
Yield [%][b]
1
2
3
4
5
6
7
8
1
1
1
1
1
1
0.5[c]
0.2[c]
0.5[c]
1
10
10
10
5
2.5
1
5
5
5[d]
10[e]
5
dioxane
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
nBu2O
100
140
130
140
140
140
140
140
140
140
140
110
85
120
6
9
6
4
6
7
8
22
20
10
72
70
N.R.
84
52
79
84
72
83
49
0
9
10
11
12
13
0
Scheme 2. Proposed benzo/[7+1] reaction pathway of CP-BCB with CO
under transition-metal catalysis.
1
1
1
72
57
N.R.
5
5
PhCH3
DCE
sertion then transforms III into intermediate IV or IV’. Finally
reductive elimination gives the benzocyclooctenone (BCO).
This designed benzo/[7+1] reaction involves two CÀC bond
cleavages: one is through the thermal opening of the four-
membered ring in benzocyclobutene, the other through transi-
tion-metal-catalyzed cyclopropane opening.[8–10]
[a] All reactions were carried out on a 0.2 mmol scale in 4 mL solvent.
[b] Yield of isolated product. [c] Here 0.5 or 0.2 atm. CO means that we
used balloon pressured mixed gas of CO/N2 (ratio is 1:1 for 0.5 atm. CO
and 1:4 for 0.2 atm. CO), whereas 1 atm. CO indicates using ballon pres-
sured gas of CO. [d] 10 mol% PPh3 was added. [e] [RhCl(PPh3)3] was used.
DCE=1,2-dichloroethane, N.R.=no reaction.
However, we had several concerns at the outset of this
work. First, the opening of cyclobutene ring in CP-BCBs could
be difficult because usually such processes were carried out
under very harsh conditions, under which the starting materi-
als may decompose. Therefore choosing an appropriate R
group in CP-BCBs to facilitate the ring opening of benzocyclo-
butene would be the primary requirement for the success of
the designed benzo/[7+1] reaction. Suzuki and co-workers
have demonstrated that CP-BCBs can give benzocycloheptenes
(structure is not given here) under thermal reaction conditions
without catalyst, even though they had to use more reactive
substrates in which the four-membered ring had two alkoxyl
substituents.[11] As a result, the second obstacle of the de-
signed benzo/[7+1] reaction is that, once intermediate III is
generated, it could also undergo direct reductive elimination
to give benzocycloheptene as a side product. In addition, b-hy-
drogen elimination could also happen for all possible inter-
mediates and this may mess up the reaction.
viously reported by our group (Table 1).[4b] However, the reac-
tant remained intact on heating for five days (Table 1, entry 1).
Considering the difficulties associated with ring-opening of
BCB, we surmised that a higher reaction temperature was
needed. When the reaction was conducted at 1408C in p-
xylene, to our delight, the desired benzo/[7+1] cycloaddition
product was obtained in a good yield (entry 2). Lowering the
reaction temperature decreased the yield remarkably (entry 3).
Reducing catalyst loading had little effect on the reaction yield
(entries 4–6). It was noticed that the benzo/[7+1] reaction with
2.5 mol% loading of catalyst gave a comparable reaction yield
as with 10 mol% catalyst loading (entry 5 vs. 2), but much
lower loading (1 mol%, entry 6) led to a little lower yield.
1 and 0.5 atm. CO were both good to the reaction, but
0.2 atm. CO reduced the yield significantly (entries 7–8). Extra
phosphine ligand was found to be detrimental to the reaction
(entry 9). Wilkinson’s catalyst cannot promote the reaction at
all (entry 10). We found that the benzo/[7+1] reaction can also
be carried out in a high boiling point solvent, such as n-butyl
ether (nBu2O), with a slightly lower yield (entry 11). In contrast,
using toluene as the solvent gave a much lower yield
(entry 12), and 1,2-dichloroethane (DCE) cannot be used as the
reaction medium (entry 13). Consequently, the optimal reaction
conditions of the benzo/[7+1] cycloaddition were determined:
balloon pressured 1 atm. CO gas, 2.5 mol% [RhCl(CO)2]2 as the
catalyst, p-xylene as the solvent at the temperature of 1408C.
We must point out here that the TBS protecting group is re-
quired for the success of the present reaction. Substrate with
a free alcohol group gave the Murakami-type CÀC cleavage
product 3a under [RhCl(CO)2]2 catalysis [Eq. (1)].[3d,14] In addi-
tion, we did not observe benzocycloheptene in the reaction
system, which suggests reductive elimination from III is slower
It was reported that the electron-donating group could pro-
mote BCB’s ring opening,[1b] so we envisioned the substituent
(R-, Scheme 2) in the BCB ring of the substrate had better be
some electron-donating group (here we chose the trimethyl-
siloxy (TMSO) or tert-butyldimethylsiloxy (TBSO) group). We
chose Rh complexes as the catalysts since Rh-catalyzed cyclo-
propane opening from intermediates similar to II had been re-
ported.[12] In addition, Rh-catalyzed cycloadditions of various
species with CO had been well documented.[13] Previously it
has been shown that [RhCl(CO)2]2 is an excellent catalyst in
several CO carbonylation reactions;[6,13] therefore, we chose it
as the catalyst to test our designed benzo/[7+1] reaction.
Our research was commenced with the model substrate 1a,
which was subjected to the reaction conditions used for the
[7+1] cycloaddition of butandienylcyclopropanes and CO pre-
&
&
Chem. Eur. J. 2015, 21, 1 – 6
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!