7
Hoveyda-Grubbs Ru complex (3) toward such electron-
deficient substrates is well recognized (cf. Table 1, entries
Table 2. Ru-Catalyzed CM between Allyl Cyanide and Type I
and Type II Olefins Promoted by Ru Complex 2a
8
1
-3). The results illustrated in Table 1 indicate that, in
addition to 6a, unsaturated nitriles 6b and 6c can be effectiVe
cross partners with certain Type I olefin partners proVided
that the reaction concentration is optimized. Contrary to CM
reactions involving acrylonitrile, to the best of our knowl-
edge, neither allyl cyanide (6b) nor homoallyl cyanide (6c)
9
have been reported as effective cross partners thus far.
The critical effect of reaction concentration is illustrated
by the observation that the yield of product 7b with Ru
complex 2 is nearly doubled to ∼80% when a 0.5 M solution
1
0
is used (vs a 0.05 M solution; Table 1, entries 5-6). Such
modifications of the reaction conditions allowed us to prepare
substantial amounts of CM products in a reliable fashion.
Thus, when the catalytic CM shown in entry 6 of Table 1
was carried out in a 5 g scale process, the desired product
7b was isolated in 75% yield (unoptimized) after purification.
It should be noted that when Ru complex 3 was used for the
same reaction (Table 1, entry 6), the trend regarding solution
concentration did not hold; instead, a similar or slightly lower
yield of the desired CM product was obtained upon
concentrating the reaction mixture (Table 1, entries 7 and
8). As illustrated in entry 9 of Table 1, in catalytic CM
involving homoallyl cyanide 6c, a 0.5 M solution used in
reactions involving Ru complex 2 afforded high product
selectivity (Table 1, entry 9).11 Enhancement of the CM
product selectivity through control of a simple parameter
a
Reaction conditions: allyl cyanide (2 equiv), CM partner (1 equiv),
(solution concentration) again proved to be sufficient to allow
b
Ru complex 2 (5 mol %), CH2Cl2, reflux, [reaction] ) 0.5 M, 2 h. Yields
c
us to employ the strategy shown in Scheme 1 by preparing
multigram quantities of the intermediate 7b through Ru-
catalyzed CM. Finally, regarding the data summarized in
Table 1, it should be noted that Ru-catalyzed CM reaction
with allyl cyanide proceeded with greater (and opposite)
stereoselectivity (E/Z g 6:1) than that typically obtained with
acrylonitrile (E/Z e 1:3) (cf. Tables 1 and 2).12
are based on isolated purified products. E:Z ratios were determined by
1
H NMR. d Percent conversion for this entry was determined by H NMR
1
after partial purification.14
tions between allyl cyanide and terminal hydroxy-olefin
substrates, formation of transient five- or six-membered
hydroxy-alkylidene Ru chelates may serve to enhance
product selectivity significantly.
Two further experiments were performed using pent-4-
With the above encouraging results in hand, we carried
out additional experiments to define the scope and limitations
of these findings. Thus, catalytic CM reactions between allyl
cyanide and several other Type I and II olefins were
conducted under the optimized conditions in the presence
17
en-1-ol (5c) as the cross partner with either its O-trityl-
protected form 8 or with styrene (Table 3). In both cases,
18
product yields were below the anticipated 66% statistical
yield for the 1:2 stoichiometry employed. Thus, the potential
Ru-oxygen chelation alone is not sufficient to rationalize the
unexpectedly high CM product selectivity obtained in the
foregoing studies with pent-4-en-1-ol (5c).
A few comments on the workup following the CM reaction
are noteworthy. To minimize secondary metathesis processes
that can adversely affect the yield and purity during the
13
of Ru complex 2 (cf. Table 1, entry 6). Among hydroxy-
lated terminal olefin partners examined, only homoallylic
alcohol 5b underwent selective CM with 6b (77% yield)
(
Table 2, entry 2); reaction of allyl alcohol (5a) and dec-9-
en-1-ol (5d) proved to be significantly less efficient (Table
, entries 1 and 3). Interestingly, it was observed that O-trityl
2
protection of pen-4-en-1-ol adversely affects product selec-
tivity (cf. 23% yield, Table 2, entry 4 vs 81% yield, Table
1, entry 6).
(17) For studies on chelation effects in CM, see: (a) BouzBouz, S.;
On the other hand, matching allyl cyanide against Type
Cossy, J. Org. Lett. 2001, 3, 1451-1454. (b) Engelhardt, F. C.; Schmitt,
M. J.; Taylor, R. E. Org. Lett. 2001, 3, 2209-2212. (c) Smulik, J. A.; Diver,
S. T. Org. Lett. 2000, 2, 2271-2274. (d) For the effect of free allylic
hydroxyl group on the RCM reaction rates, see: Hoye, T. R.; Zhao, H.
Org. Lett. 1999, 1, 1123-1125.
II hydroxy-olefin partners such as a secondary allylic alcohol
10, Table 2, entry 5) or a disubstituted allylic alcohol (12,
entry 6), resulted in substatistical product yields (21%, Table
(
(
18) In addition to the CM product (Table 3, entry 2), a side-product
1
5
16
2
, entry 5 and 46%, Table 2, entry 6 ). Catalytic CM of
that tenaciously coeluted was also detected (∼15%) and was characterized
1
1
vinyl dioxolane 13 (Table 2, entry 7), considered a Type II
partner, proceeded similarly (26% yield). Collectively, the
above observations suggest that in Ru-catalyzed CM reac-
as Ph-CHdCH-(CH2)2-OH ( H- H COSY, GC-MS). This adventitious
product is likely the result of olefin isomerization (in 5c) followed by CM.
For other related examples, see: Alcaide, B.; Almendros, P. Chem. Eur. J.
2003, 9, 1259-1262 and references therein.
Org. Lett., Vol. 7, No. 11, 2005
2115