Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
C. Kouklovsky and G. Vincent
performed in toluene (Table 1, entries 4 and 9), at 808C
(Table 1, entries 4, 7, and 8), and at a concentration of 0.1m
(Table 1, entries 2, 4, and 5). Finally, the addition of G-II
(10 mol%) in two portions proved to be beneficial to the
yield of 9b (Table 1, entry 12).
On the other hand, reaction of the active Ru–carbene
complex with the exocyclic double bond (path B, Scheme 5)
would generate reactive intermediate IV, which would un-
dergo ROM/RCM and finally CM. No ROM/CM products
are expected to arise from this pathway.
In this RRM process different reaction pathways may be
operative, dependent on various factors (Scheme 5).[4e]
Initial cyclobutanametallation of the endocyclic double
bond (path A, Scheme 5, intermediates I–III) would lead to
two regioisomeric ring-opened intermediates, V (path A1,
Scheme 5) and VI or VII (paths A2 and A3, Scheme 5). In-
We propose to examine the variations in the preferred
pathway with the level of ring strain in the bicyclic NDA cy-
cloadduct, the length of the alkene side chain, and the
nature of the CM partner.
With optimized conditions in hand, we turned our atten-
tion to the reactivity of bicycloACTHNUTRGNEUGN[2.2.1]-3,6-dihydro-1,2-oxa-
zine (6) towards various CM
partners (Table 2). In the ab-
sence of an external alkene, the
bicycloACHTNUTRGNEUNG[4.3.0] compound 9a was
obtained in 17% yield (Table 2,
entry 1). Polymeric substances
were the major side products,
which most probably arose
from a ring-opening metathesis
polymerization (ROMP) pro-
cess. Under an ethylene atmos-
phere the polymerization was
suppressed and 9a was ob-
tained with an improved yield
of 54% (Table 2, entry 2). Re-
action with gaseous but-2-ene
at 808C led to the formation of
the ROM/RCM/CM product 9c
(68%, Table 2, entry 3).[17] Sur-
prisingly, a minor amount of
the epimerized compound 24c
was also isolated. The mecha-
nism of formation of this com-
pound is unclear and will be
discussed later (see Scheme 6
below). When the same reac-
tion performed at a lower tem-
perature
entry 4) the formation of 24c
was suppressed and good
yield of 9c was obtained
(608C,
Table 2,
Scheme 5. Potential pathways for the ring-rearrangement metathesis of NDA cycloadducts.
a
termediate V should undergo CM to give 20–23, whereas in-
termediates VI and VII could afford the RCM (9–12) or
CM products (20–23). It is postulated that coordination of
the Ru carbene by the exocyclic carbonyl (VI) would pre-
vent RCM, thus, CM would be favored. Eventually, the
ROM/CM products 20–23 could react further to yield the
RCM compounds 9–12.
Considering the nature of the active catalyst ([Ru]=CHA;
A=H or R), the isolation of significant amounts of ROM/
RCM products 9a and 10a (see Table 3 below) alongside
the ROM/RCM/CM compounds led us to believe that the
ruthenium methylidene ([Ru]=CH2) is a propagating species.
If the ruthenium alkylidene intermediate ([Ru]=CHR) was
the only active catalyst, only ROM/CM/RCM or ROM/CM
products would have been isolated.
(75%). Reaction with allyl acetate (Table 2, entry 5) led to a
1:1 mixture of 9d (ROM/RCM/CM) and 9a (ROM/RCM)
in 52% yield. It is known that CM processes might be im-
proved by using a dimer of one of the partners.[18] Accord-
ingly, we ran the reaction with but-2-enyl diacetate (Table 2,
entry 6). The ratio of 9d/9a was improved to 4:1, but was
accompanied by a decreased yield of 29%. In this case, a
large amount of ROM/CM product 20d was obtained. Allyl-
trimethylsilane (Table 2, entry 7) and oct-1-ene (Table 2,
entry 8) led to good yields of the desired rearranged bicycles
9e and 9 f (ROM/RCM/CM), respectively, as mixtures with
9a (ROM/RCM). When methyl acrylate (Table 2, entry 9)
was tested no ROM/RCM/CM product 9g was recovered.
Only 25% of the ROM/RCM product 9a was obtained,
along with polymer formation, indicative that methyl acry-
2974
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2972 – 2980