Relay ring-closing metathesis (RRCM): A strategy for directing metal movement throughout olefin metathesis sequences

Article abstract of DOI:10.1021/ja046385t

The title concept involves the use of structurally modified RCM substrates that contain extender arms, terminating in a remote reactive alkene. Initiation of an RCM sequence at that reactive alkene is followed by rapid intramolecular relay of the metal center to an initially less reactive alkene in the parent substrate. This permits one to control the relative timing (or direction) of a metathesis sequence. For example, one can reverse the inherent tendency of an unsymmetrical α,ω-diene substrate to close, say, left-to-right, to that of right-to-left. Four distinct types of application of the RRCM concept are demonstrated. Among other things, they show the preparation of tetrasubstituted electron-deficient alkenes using G1 [(Cy₃P)₂(Cl₂)Ru=CHPh], complementary control of directionality (endedness), auxiliary benefits (enzyme specificity) from the incorporation of additional steric bulk, the activation of otherwise ineffective substrates for RCM closure, the use of unorthodox alkenes as initiation sites for ring closure, and control of product olefin geometry. Copyright

Full text of DOI:10.1021/ja046385t

Published on Web 07/29/2004
Relay Ring-Closing Metathesis (RRCM): A Strategy for Directing Metal
Movement Throughout Olefin Metathesis Sequences
Thomas R. Hoye,* Christopher S. Jeffrey, Manomi A. Tennakoon, Jizhou Wang, and Hongyu Zhao
Department of Chemistry, UniVersity of Minnesota, Minneapolis, Minnesota 55455
Received June 18, 2004; E-mail: hoye@chem.umn.edu
In the past decade, ring-closing metathesis (RCM) has been
established as a powerful and generally applicable method for
construction of carbocycles. However, even with the remarkable
successes emanating from literally thousands of laboratories, the
applicability of RCM is not universal. Limitations are most often
encountered as users attempt to apply the technology to the
construction of increasingly complex molecular targets. More
specifically, RCM can fail when substrate alkenes are either
sterically hindered or electronically deactivated.
To circumvent these types of problems, thereby increasing the
scope of the RCM reaction, we have developed the concept of relay
ring-closing metathesis (RRCM).1 This involves the design of
substrates that, in effect, permit one to dictate the sequence of
metathesis events by choreographing the metal atom (Ru in the
case of G12a and G22b) through the individual steps of the RCM
cascade.3 As we show here, relay RCM (RRCM) permits the
cyclization of many types of otherwise recalcitrant alkene substrates.
The results described here demonstrate how to rationally design
modifications of imperfect RCM substrates so that subsequent
RRCM solves a reactivity or selectivity problem. This constitutes
a new type of substrate control, one in which use of a newly
designed substrate steers the reaction pathway in a preferred
direction.
terminal alkenes.6 In contrast, exposure of the modified relay
substrate 4 to G1 resulted in smooth cyclization to the cyclopentene
derivative 5.7a This and 7 are the only examples we know of
tetrasubstituted alkenes formed by G1-mediated RCM of a simple
R,ω-diene. Thus, introduction of the remote terminal alkene in 4
(readily derived from citronellene) opened a pathway to access 2
by way of 3. A second example involved closure of either of the
isomeric substrates 6 or 8 to the electron-deficient, tetrasubstituted
alkene. Thus, one can access either of two independent pathways
(no identical intermediates)7b from either end of a molecule. These
results also demonstrate that cyclization both onto (during 6 to 7)
and into (during 8 to 7) an electron-deficient alkene is viable.
Although these first successes clearly demonstrated the RRCM
concept, the advent of the second-generation Grubbs initiator (G2)
significantly increased the scope of the traditional RCM reaction
(e.g., 1 can be closed to 5 with G28). Nonetheless, that scope is
finite. The results presented in Scheme 2 constitute definitive
examples in which RRCM provides a level of control that is not
available with traditional RCM. In addition, they demonstrate that
catalyst-to-substrate matching can provide synergistic benefits.
Scheme 2
Our first example of RRCM1a involved the cyclization shown
in Scheme 1.4 Diene 1, bearing two 1,1-disubstituted ethylene
moieties, is known to be unreactive toward the first-generation
Grubbs initiator G15, the ruthenium complex in hand at that time.
G1 is not sufficiently active to engage geminally substituted
Scheme 1
Tandem enyne metathesis9 substrates 9 (Scheme 2)4 can cyclize
with either a left-to-right or right-to-left endedness to give isomeric
dienes 11-ltr (via 10-ltr) or 11-rtl (via 10-rtl). Closure of parent
dienyne 9a with G2 provided the benchmark value of 1.0:2.0 for
the ratio of products 11-ltr to 11-rtl. The related relay substrates
9b vs 9c (now containing an allylic ether relay moiety) provided
improved, but imperfect, selectivity. RRCM closure of each with
G2 gave 1.0:7.0 vs 4.7:1.0 ratios of 11-ltr to 11-rtl, respectively.
Thus, G2 may not be highly discriminating of the two termini in
9b or 9c.7c If true, then the less reactive G1 should be superior.
Consistent with this analysis, relay substrate 9b, when treated with
G1, gave the bicyclic diene 11-ltr very selectively (26:1.0).
9
10210
J. AM. CHEM. SOC. 2004, 126, 10210-10211
10.1021/ja046385t CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3
task for a researcher to modify the appropriate alkene to incorporate
the relay extension even if commitment to the RRCM strategy is
not made from the outset of a synthetic plan.
The specific examples of relay-driven ring closures presented
here demonstrate a number of strategic advantages. Tetrasubstituted,
electron-deficient alkenes were prepared using G1 (Scheme 1), fully
complementary control of directionality (endedness) was achieved
(Scheme 2), nonobvious auxiliary benefits (enzyme specificity) from
the incorporation of additional steric bulk were recognized (Scheme
3), mechanistic insight (not discussed) has emerged, ineffective
substrates for traditional RCM closures were turned on (Schemes
1-4), and unorthodox alkenes could be used as initiation sites for
ring closure (Scheme 4). From these examples, one can see that
relay ring-closing metathesis is complementary to traditional RCM.
It represents enabling technology for molecular construction at the
strategic level. We predict that many applications will emerge as
investigators contemplate their own uses of RRCM.
Moreover, analogous treatment of the isomeric 9c (the “endomer”
of 9b) was highly complementary, giving 11-rtl nearly exclusively
(1.0:45).
RRCM also offers nonobvious advantages. In the course of
preparing a nonracemic sample of 15 (Scheme 3),4 the anticipated
product of a silicon-tethered cross-metathesis of 12 with 13, we
observed that the lipase resolution of carbinol (()-12a was not
serviceable (40% ee). Recognizing the expendable nature of the
remote alkene atom (• in 12) and all that it bears during RRCM,
we capitalized on the greater size difference of the groups flanking
the carbinol center in (()-12b to achieve a much more efficient
lipase differentiation (g90% ee). RRCM of 14 with G2 then led
to 15 in good (58%) isolated yield.1b
The RRCM reactions shown in Scheme 44 are instructive in
different ways. Substrates 16a-c are armed with a relay moiety,
ready to pass the metal into an otherwise less accessible (electroni-
cally and/or sterically deactivated) site (cf. the atypical Ru-
alkylidene intermediates 17a-c). When each of the polyenes 16a-c
Acknowledgment. These studies were supported by grants
awarded by the DHHS (GM-65597 and CA-76497).
Supporting Information Available: Spectral characterization data
for compounds 4, 5, 6-8, 9a-c, 11-ltr, 11-rtl, 12a,b, 14, 15, 16a-c,
and 18a-c; reaction conditions for all RRCM reactions; yield data for
1
all reactions; and a representative sequence of H NMR spectra from
monitoring tandem dienyne metathesis cyclization (9b to 11-ltr + 11-
rtl) (PDF). This material is available free of charge via the Internet at
http://pubs.acs.org.
Scheme 4
References
(1) (a) Zhao, H., Ph.D. Thesis, University of Minnesota, Minneapolis,
Minnesota, 2000. (b) Tennakoon, M. A., Ph.D. Thesis, University of
Minnesota, Minneapolis, Minnesota, 2001. (c) Danielson, M. E., Ph.D.
Thesis, University of Minnesota, Minneapolis, Minnesota, 2003. (d) Hoye,
T. R.; Wang, J. Abstracts of Papers, 226th National Meeting of the
American Chemical Society, Sept 7-11, 2003, New York; American
Chemical Society: Washington, DC, 2003; ORGN-670.
(2) (a) G1 [(Cy3P)2(Cl2)RudCHPh]: Schwab, P.; Grubbs, R. H.; Ziller, J.
W. J. Am. Chem. Soc. 1996, 118, 100-110. (b) G2 [(H2IMes)(Cy3P)-
(Cl2)RudCHPh]: Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org
Lett. 1999, 1, 953-956.
(3) For recent reports where this strategy was used to an advantage in
improving the efficiency (directionality) of metathesis cross-coupling and
macrocylization reactions, respectively, see: (a) Hansen, E. C.; Lee, D.
Org. Lett. 2004, 6, 2035-2038. (b) Wang, X.; Bowman, E. J.; Bowman,
B. J.; Porco, J. A., Jr. Angew. Chem., Int. Ed. 2004, 43, 3601-3605. See
also (c) Randl, S.; Gessler, S.; Wakamatsu, H.; Blechert, S. Synlett 2001,
430-432. (d) She, J.; Crimmins, M. T. Abstracts of Papers, 226th National
Meeting of the American Chemical Society, Sept 7-11, 2003, New York;
American Chemical Society: Washington, DC, 2003; ORGN-193. (e)
Robinson, J.; Piscopio, A. D. Abstracts of Papers, 226th National Meeting
of the American Chemical Society, Sept 7-11, 2003, New York;
American Chemical Society: Washington, DC, 2003; ORGN-118.
(4) For reaction conditions see Supporting Information.
(5) Kirkland, T. A.; Grubbs, R. H. J. Org. Chem. 1997, 62, 7310-7318.
(6) Ulman, M.; Grubbs, R. H. Organometallics 1998, 17, 2484-2489.
(7) (a) Trace amount (<2%) of the “truncation” byproduct 1 (E ) CO2Et)
was observed, presumably arising from cross-metathesis of 2 with either
4 or ethylene. A control experiment in which 4 (E ) CO2Et) was doped
with 1 (E ) CO2Me) gave none of 5 (E ) CO2Me), demonstrating that
the truncated byproduct was not part of a productive pathway. (b) A control
experiment showed that methallyl methacrylate was, as expected, unre-
active and, therefore, not a common intermediate. (c) An alternative
explanation is that a portion of 9b or 9c was being truncated to parent
9a, thereby eroding selectivity. In situ NMR monitoring (see Supporting
Information) suggested that this was not the case. (d) Without active
removal of ethylene and propylene (N2 sparging and vigorous reflux),
truncation products such as 19 were frequently formed as terminal events.
(8) Ref 2b: and Ackermann, L.; Fu¨rstner, A.; Weskamp, T.; Kohl, F. J.;
Herrmann, W. A. Tetrahedron Lett. 1999, 40, 4787-4790.
was subjected to G2 in toluene at 110 °C (with continuous N2
sparging), it was consumed within minutes.7d The cyclized product
18, a 14-membered lactone, was formed as the exclusive [18a (Z-
only) and 18b (E-only)] or major product [18c (E-only) + ∼20%
byproducts]. Importantly, a control experiment demonstrated that
the independently prepared truncation product 19, corresponding
to substrate 16a, did not give 18a under the reaction conditions,
ruling out its intermediacy in the conversion of 16a to 18. Ring
closure originating from the opposite end of 16a-c would likely
suffer from low reactivity and/or regioselectivity issues inherent
to the alkenes in the CdC(Me)ABCO2R subunit. Finally, since most
acyclic RCM substrates are synthesized by convergent strategies
in modular fashion (e.g., 16a-c were all easily assembled by
straightforward esterification), it will usually be a relatively simple
(9) (a) Kim, S.-H.; Bowden, N.; Grubbs, R. H. J. Am. Chem. Soc. 1994, 116,
10801-10802. (b) Poulsen, C. S.; Madsen, R. Synthesis 2003, 1-18. (c)
Mori, M. In Handbook of Metathesis; Grubbs, R. H., Ed.; Wiley-VCH:
Weinheim, Germany, 2003; Vol. 2, pp 176-204. (d) Diver, S. T.; Giessert,
A. J. Chem. ReV. 2004, 104, 1317-1382 and references therein.
JA046385T
9
J. AM. CHEM. SOC. VOL. 126, NO. 33, 2004 10211