Ethylene-promoted versus ethylene-free enyne metathesis

Article abstract of DOI:10.1021/ja207388v

The role of ethylene in promoting metathesis of acetylenic enynes is probed within the context of ring-closing enyne metathesis, using first- and second-generation Grubbs catalysts. Under inert atmosphere, rapid catalyst deactivation is observed by calibr

Full text of DOI:10.1021/ja207388v

COMMUNICATION
pubs.acs.org/JACS
Ethylene-Promoted versus Ethylene-Free Enyne Metathesis
Anne G. D. Grotevendt, Justin A. M. Lummiss, Melanie L. Mastronardi, and Deryn E. Fogg*
Department of Chemistry; Centre for Catalysis Research & Innovation, University of Ottawa, Ottawa, ON, Canada K1N 6N5
S Supporting Information
b
Scheme 1. RCEYM Reaction Explored by MALDI-TOF MS
ABSTRACT: The role of ethylene in promoting metathesis
of acetylenic enynes is probed within the context of ring-
closing enyne metathesis, using first- and second-generation
Grubbs catalysts. Under inert atmosphere, rapid catalyst
deactivation is observed by calibrated GCꢀFID analysis for
substrates with minimal propargylic bulk. MALDI-TOF
mass spectra reveal a Ru(enyne)2 derivative that exhibits
very low reactivity toward both enyne and ethylene. Under
ethylene, formation of this species is suppressed. Enynes
with bulky propargylic groups are not susceptible to this
catalyst deactivation pathway, even under N2 atmosphere.
nyne metathesis offers unparalleled efficiency and atom
E
economy in the assembly of synthetically versatile 1,3-dienes.1
Recent examples of biologically relevant compounds synthesized
by sequences utilizing ring-closing enyne metathesis (RCEYM) as
a key step include the antitumor agents (ꢀ)-Acylfulvene and
(ꢀ)-Irofulven,2 as well as nucleoside analogues to the antiviral
agent Stavudine.3 A breakthrough in this area was reported by Mori
more than 10 years ago, with the discovery that use of ethylene
atmospheres dramatically improves rates and yields in ruthenium-
catalyzed RCEYM.4 A clever “atmosphere-switching” study subse-
quently confirmed that reaction was consistently faster under
ethylene.5 The origin of these effects6 has been much discussed.
Puzzling aspects include the fact that ethylene is not invariably
required for satisfactory outcomes1g (and indeed can sometimes be
detrimental),6b,7 and a striking disparity with olefin metathesis, in
which retention of cogenerated ethylene undermines catalyst
performance.8 Here, we investigate the origin of these inconsisten-
cies, with specific attention to the effect of ethylene on the nature of
the organic and organometallic products, and to the influence of
propargylic substitution.
Figure 1. (a) Rates of RCEYM for the reaction of Scheme 1; (b) product
distribution under C2H4, vs N2 atmosphere; in situ yields determined by
calibrated GCꢀFID analysis.
most of that ultimately formed (92% in situ yield; Figure 1b). In
striking contrast, the reaction under N2 shows only 23% conver-
sion at 1 h, and the yield of 1a0 is just 14%, the mass balance being
due to involatile species arising from intermolecular reaction
(vide infra). The rate of ensuing reaction declines sharply; over
24 h, conversions increase by only ca. 10%, with a final diene yield
of 18%. Independent experiments at enyne concentrations of
0.3 M exhibit still lower RCEYM selectivity.10
MALDI-TOF mass spectra for these experiments are shown in
Figure 2. For reactions under ethylene, the spectrum at 1 h is
dominated by the radical cation for methylidene Ru-2, the
expected resting state of the catalyst in the presence of C2H4
(m/z 746.3, Figure 2a). A minor signal at higher mass (Ru-3, m/z
878.3) grows in over 24 h. Importantly, this signal dominates the
spectrum for the reaction under N2, even at 1 h, when enyne
consumption is just beginning to plateau (compare Figures 2b,
1a). Complex Ru-3 thus corresponds to the major ruthenium
species present at the onset of catalyst deactivation. Its mass and
isotope pattern indicate the presence of two enyne-derived
repeat units, as discussed below. Despite the low reactivity
implied by its formation under C2H4, and indicated by the rate
curve of Figure 1a, Ru-3 undergoes conversion into higher
oligomers via slow, sustained consumption of 1a. After 24 h
We began by examining the effect of the headspace atmo-
sphere in reactions of enyne 1a with Ru-1 (Scheme 1), a
substrate-catalyst combination known to afford higher RCEYM
yields under ethylene.4 Experiments were carried out in a glove-
box in sealed Schlenk tubes under C2H4 or N2, in solutions
presaturated with the headspace gas. Samples were removed at
intervals for parallel assessment of the organic and the ruthenium
constituents by, respectively, GCꢀFID and anaerobic MALDI-
TOF mass spectrometry. We have described elsewhere the
power of charge-transfer (CT) MALDI MS methods for analysis
of neutral transition-metal complexes, including the Grubbs
catalysts.9
For reactions carried out under C2H4, conversions of 1a
reached 37% after 1 h, and 98% after 24 h (Figure 1a). The
expected 1,3-diene 1a0 is the sole initial product, and accounts for
Received: August 5, 2011
Published: September 15, 2011
r
2011 American Chemical Society
15918
dx.doi.org/10.1021/ja207388v J. Am. Chem. Soc. 2011, 133, 15918–15921
|

Journal of the American Chemical Society
COMMUNICATION
1a occurs in the initial cross-metathesis step that generates
intermediate Ru-A. A key branch point is then possible. Thus,
reaction of Ru-A with the olefinic end of a new enyne substrate
would liberate the product 1a0, while competing reaction with the
alkynyl end would generate a ruthenium species bearing two
enyne-derived repeat units (e.g., Ru-B). CꢀH activation and
loss of HCl from Ru-B would afford Ru-3, a possible structure
for which is depicted. The stability of Ru-3 could be consistent
with delocalization in the π-system of the ruthenacycle, or,
alternatively, with a coordinatively saturated π-allyl complex.11,12
Irrespective of the specific structure of Ru-3, its molecular mass is
strong evidence against the “yne-first” pathway (Scheme 2b), the
complete atom-economy of which would result in a higher-mass
Ru intermediate.
Additional isomers of Ru-3 may arise from head-to-head
versus head-to-tail cycloaddition, or intermolecular reaction
prior to the cyclization step indicated. In situ NMR experiments
1
13
(1H; Hꢀ C HMBC) reveal multiple RudCRR0 species.
Detailed structural characterization of Ru-3 is complicated by
this apparent abundance of isomers, by their further evolution
over a time scale of hours, and by difficulties in isolating the Ru
products by silica gel chromatography. The foregoing is never-
theless important in revealing a previously unsuspected bias of
these Ru catalysts toward sequential, intermolecular enyne metathesis
under inert atmosphere. We conclude that a key benefit of ethylene
lies in its capacity to suppress (or reverse) uptake of a second enyne
unit, and ensuing catalyst deactivation. Of note, addition of ethylene
after 24 h effects a minor increase in RCEYM yields, but does not
restore activity to initial levels, nor generate Ru-2. Likewise (in
contrast with ring-closing olefin metathesis),13,14 oligomers15,16
cannot be recycled into the desired cyclic products by diluting the
reaction to trigger a concentration-dependent cyclodepolymeriza-
tion. The exothermic nature of the enyne metathesis reaction
precludes many of the equilibria characteristic of olefin metathesis.17
Competing alkyne polymerization is widely recognized as a
challenge in enyne metathesis promoted by group 6 complexes.16
With a few notable exceptions,18 it has been little considered for
the Ru systems, despite Mori’s early suggestion that it might be
operative.19 A probable contributor to the belief that alkyne
polymerization was not an issue for the Ru catalysts was
pioneering work by Blechert and co-workers on intermolecular
enyne metathesis, showing that polymerization by Ru-1 is slow
relative to productive metathesis. In this case, it should be noted,
however, that a 2- to 3-fold excess of alkene was used.20
Figure 2. CT-MALDI mass spectra (pyrene matrix) showing Ru
species formed in the reactions of Scheme 1; (a) at 1 h under C2H4;
(b) at 1 h under N2; (c) at 24 h under N2. Insets: isotope patterns (top,
simulated; bottom, observed). [*] = ion formed by gas-phase loss of
chloride from Ru-3. In (c), E denotes the molecular formula for enyne
1a or its product 1a0: C13H18O4.
Scheme 2. (a) Ene-First versus (b) Yne-First RCEYM,
Showing Productive Cyclization, Uptake of a Second Enyne,
and (for the Ene-First Pathway), Formation of Deactivated
Ru-3a
Given the evidence above that intermolecular reaction of
Ru-A with “yne” leads to catalyst deactivation, and that this
can be blocked by competing reaction with ethylene, we specu-
lated that enynes for which ethylene is not required might be ones
for which approach of the alkyne to Ru-A is sterically inhibited.
This could account for the fact that “Mori conditions” are not
required for internal alkynes. We suspected that propargylic
functionalization might account for many of the remaining
examples21,22 (that is, the same steric factors that accelerate retro-
addition1e could also inhibit approach of the alkyne to Ru-A).
To test this hypothesis, we undertook a systematic examina-
tion of the effect of enyne substitution on RCEYM yields in the
absence of ethylene (Table 1). We also sought to clarify whether
the preponderance of Ru-NHC catalysts in ethylene-free
RCEYM21 implies that such catalysts are immune to the pro-
blems that the Mori conditions were designed to address. We
therefore included the important second-generation Grubbs
catalyst RuCl2(H2IMes)(PCy3)(dCHPh) Ru-4 in this study.
a [Ru] = RuCl2(PCy3); E = C(CO2Et)2. For a complete depiction of the
atom-efficient initiation in (b), see Supporting Information.
under N2, the dominant Ru species contains six enyne-derived
repeat units (Figure 2c). Not seen, even at 1 h, is Ru-2 (the
anticipated resting state in the “yne-first” mechanism for enyne
metathesis). Nor is Ru-A (the first cycloaddition product in the
“ene-first” mechanism: see Scheme 2a; m/z 676.2) or its PCy3-
bound resting state (m/z 956.4).
The excellent match between calculated and observed isotope
patterns supports identification of Ru-3 as a “RuCl(PCy3)-
(E)2ꢀCH3” species. We envisage its formation via “ene-first”
reaction of Ru-1 with 1a (Scheme 2a). Formal loss of CH2 from
15919
dx.doi.org/10.1021/ja207388v |J. Am. Chem. Soc. 2011, 133, 15918–15921

Journal of the American Chemical Society
COMMUNICATION
Table 1. Ethylene-Free RCEYMa
In summary, we find that acetylenic enynes with minimal
propargylic bulk perform poorly in ethylene-free RCEYM using
first- or second-generation Grubbs catalysts, owing to rapid
catalyst deactivation. MALDI-MS experiments with Ru-1 reveal
that the onset of deactivation correlates with uptake of a second
equivalent of enyne by the key vinylalkylidene intermediate Ru-
A. The two enyne-derived repeat units in the deactivation
product Ru-3 may form a delocalized metallacyclohexene ring,
or a coordinatively saturated π-allyl structure, either of which
could account for low reactivity toward both enyne and ethylene.
Use of ethylene atmosphere suppresses this pathway and hence
catalyst deactivation: we suggest that this is a major contributor
to the beneficial effect of the “Mori conditions” in Ru-catalyzed
enyne metathesis. Propargylic bulk also inhibits formation of
Ru-3 and thus enables ethylene-free enyne metathesis.
’ ASSOCIATED CONTENT
S
Supporting Information. Mechanistic details related to
b
Scheme 2, experimental procedures, characterization data, spec-
tra and rate curves. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
dfogg@uottawa.ca
’ ACKNOWLEDGMENT
This paper is dedicated to Christian Bruneau on the occasion
of his 60th birthday. We thank NSERC of Canada for financial
support in the form of Discovery and Accelerator grants.
’ REFERENCES
(1) Recent reviews of enyne metathesis: (a) Fischmeister, C.; Bruneau,
C. Beilstein J. Org. Chem. 2011, 7, 156–166. (b) Mori, M. Materials 2010,
3, 2087–2140. (c) Villar, H.; Frings, M.; Bolm, C. Chem. Soc. Rev. 2007,
36, 55–66. (d) Mori, M. Adv. Synth. Catal. 2007, 349, 121–135. (e) Diver,
S. T. Coord. Chem. Rev. 2007, 251, 671–701. (f) Hansen, E. C.; Lee, D. Acc.
Chem. Res. 2006, 39, 509–519. (g) Diver, S. T.; Giessert, A. J. Chem. Rev.
2004, 104, 1317–1382.
(2) Siegel, D. S.; Piizzi, G.; Piersanti, G.; Movassaghi, M. J. Org.
Chem. 2009, 74, 9292–9304.
(3) Vuong, S.; Rodriguez-Fernandez, M. M.; Renoux, B.; Len, C.
Carbohydr. Res. 2010, 345, 324–329.
(4) Mori, M.; Sakakibara, N.; Kinoshita, A. J. Org. Chem. 1998,
63, 6082–6083.
(5) Lloyd-Jones, G. C.; Margue, R. G.; de Vries, J. G. Angew. Chem.,
Int. Ed. 2005, 44, 7442–7447.
(6) General rate improvements associated with use of ethylene in
enyne metathesis have been attributed to competition with PCy3 for
binding to the four-coordinate active species, thus inhibiting formation
of the less reactive RuCl2(PCy3)(H2IMes)(dCH2) resting state (see
(a) Galan, B. R.; Giessert, A. J.; Keister, J. B.; Diver, S. T. J. Am. Chem.
Soc. 2005, 127, 5762ꢀ5763) and to faster turnover by reaction of
ethylene, vs enyne, as discussed below (see (b) Giessert, A. J.; Brazis,
N. J.; Diver, S. T. Org. Lett. 2003, 5, 3819ꢀ3822, and ref 5.)
(7) Mori, M.; Kitamura, T.; Sakakibara, N.; Sato, Y. Org. Lett. 2000,
2, 543–545.
a Conditions: N2 atmosphere, 24 °C, 0.1 M enyne, CH2Cl2; in situ yields
by calibrated GCꢀFID analysis, (2.5%. One mole percent Ru for 1aꢀc,
except 1c with Ru-4: 0.5 mol % for the latter and 2ꢀ8.
For enynes devoid of propargylic substituents, competing oligo-
merization is observed with both first- and second-generation
catalysts, irrespective of the nature of the homoallylic moiety (1aꢀc,
entries 1ꢀ3). Indeed, Ru-4 is even more susceptible to deactivation
by such substrates than Ru-1, affording poorer conversions and very
low RCEYM yields. Within the context of RCM, we have noted a
greater tendency of Ru-NHC complexes, versus Ru-1, toward
oligomerization; a similar bias appears operative in the present case.
With increasing propargylic bulk, a steady improvement in RCEYM
yields is seen (entries 3ꢀ9). Gem-dialkyl functionalization is insuffi-
cient to completely inhibit deactivation, but replacement of even one
alkyl group by phenyl enables complete conversion of enyne 6 to the
1,3-diene (entry 8). Reaction rates also increase. This probably
reflects reduced rates of deactivation, as well as sterically accelerated
retro-addition and inhibited reuptake of PCy3.1e,g
Once uptake of a second enyne is suppressed, the superior
reactivity of Ru-4 is manifested: this catalyst effects complete
RCEYM of diphenyl-substituted 7 within 1 h at RT, versus >18 h
with Ru-1. Finally, allylic bulk (entry 10) confers no beneficial
effect, presumably because it impedes access of “ene” to the
Ru center. The very low conversions of 8 are consistent with
previous findings1g that allylic bulk inhibits reaction.
(8) For the longevity of ruthenium catalysts commonly used for
olefin metathesis, see: (a) Ulman, M.; Grubbs, R. H. J. Org. Chem. 1999,
64, 7202–7207. For selected studies demonstrating the detrimental
effect of ethylene on rates and yields in olefin cross-metathesis and
15920
dx.doi.org/10.1021/ja207388v |J. Am. Chem. Soc. 2011, 133, 15918–15921

Journal of the American Chemical Society
COMMUNICATION
ring-closing metathesis, see: (b) Burdett, K. A.; Harris, L. D.; Margl, P.;
Maughon, B. R.; Mokhtar-Zadeh, T.; Saucier, P. C.; Wasserman, E. P.
Organometallics 2004, 23, 2027–2047. (c) Lysenko, Z.; Maughon, B. R.;
Mokhtar-Zadeh, T.; Tulchinsky, M. L. J. Organomet. Chem. 2006,
691, 5197–5203. (d) Monfette, S.; Eyholzer, M.; Roberge, D. M.; Fogg,
D. E. Chem.—Eur. J. 2010, 11720–11725.
2001, 42, 8231–8233. (b) F€urstner, A.; Ackermann, L.; Gabor, B.;
Goddard, R.; Lehmann, C. W.; Mynott, R.; Stelzer, F.; Thiel, O. R.
Chem.—Eur. J. 2001, 7, 3236–3253. (c) Saito, N.; Sato, Y.; Mori, M. Org.
Lett. 2002, 4, 803–805. (d) Giessert, A. J.; Snyder, L.; Markham, J.;
Diver, S. T. Org. Lett. 2003, 5, 1793–1796. (e) Park, S.; Kim, M.; Lee, D.
J. Am. Chem. Soc. 2005, 127, 9410–9415. (f) Clavier, H.; Nolan, S. P.
Chem.—Eur. J. 2007, 13, 8029–8036. (g) Castarlenas, R.; Eckert, M.;
Dixneuf, P. H. Angew. Chem., Int. Ed 2005, 44, 2576–2579. Most of these
examples involve Ru-NHC catalysts, perhaps contributing to the
perception that such catalysts do not require use of ethylene for
successful RCEYM; the latter involves the in situ Dixneuf catalyst.
(22) RCEYM yields were previously shown to be improved by
increasing propargylic substitution for reactions under ethylene. See:
Kitamura, T.; Sato, Y.; Mori, M. Adv. Synth. Catal. 2002, 344, 678–693.
(9) Eelman, M. D.; Blacquiere, J. M.; Moriarty, M. M.; Fogg, D. E.
Angew. Chem., Int. Ed. 2008, 47, 303–306.
(10) Consistent with this concentration-dependence is behaviour
observed by Hoye and co-workers, who recognized in early, ethylene-
free work that the ene-first mechanism (see later) necessitates reaction
with incoming enyne to liberate the diene product, and therefore
proposed that increased enyne concentrations might be beneficial.
Conversions were indeed dramatically improved by slow addition of
Ru-1 to enyne, but yields suffered. These reaction conditions promote
the formation of oligomers. See: Hoye, T. R.; Donaldson, S. M.; Vos,
T. J. Org. Lett. 1999, 1, 277–279.
(11) The likelihood of forming an allylic dichlororuthenium com-
plex by rearrangement of the vinyl ruthenacyclobutane intermediate
itself has been examined computationally (ref 12); this was found to be a
high-energy process. Here, we envisage formation of an allylic species via
CꢀH activation of a pendant olefin.
(12) Lippstreu, J. J.; Straub, B. F. J. Am. Chem. Soc. 2005, 127, 7444–
7457.
(13) Monfette, S.; Fogg, D. E. Chem. Rev. 2009, 109, 3783–3816.
(14) Conrad, J. C.; Eelman, M. D.; Duarte Silva, J. A.; Monfette, S.;
Parnas, H. H.; Snelgrove, J. L.; Fogg, D. E. J. Am. Chem. Soc. 2007,
129, 1024–1025.
(15) We use the term oligomers for convenience, but note that these
will include any dimers that are insufficiently volatile to emerge under
our GC conditions. Liberation of dimeric products by competing
turnover of Ru-B with enyne is highly probable, and indeed such
products have been isolated for enyne metathesis in the absence of
ethylene (see (a) Diver, S. T.; Kulkarni, A. A.; Clark, D. A.; Peppers, B. P.
J. Am. Chem. Soc. 2007, 129, 5832–5833). Enyne-derived oligomers have
been isolated in Group 6 catalysis; see ref 16. Symmetrical dimers can
also form (if the catalyst is sufficiently reactive) by cross-metathesis of
the conjugated dienes, but this is slow even in refluxing CH2Cl2 (see,
e.g., (b) Poulsen, C. S.; Madsen, R. J. Org. Chem. 2002, 67, 4441–4449).
We observe no such behaviour at room temperature (see, for example,
the rate curves for RCEYM via Ru-4 in the Supporting Information).
(16) For leading references to alkyne polymerization by group 6
catalysts, see: (a) Lee, Y.-J.; Schrock, R. R.; Hoveyda, A. H. J. Am. Chem.
Soc. 2009, 131, 10652–10661. (b) Zhao, Y.; Hoveyda, A. H.; Schrock,
R. R. Org. Lett. 2011, 13, 784–787.
(17) The overall process of enyne metathesis can be reversed under
some conditions, however. See: Lee, H.-Y.; Kim, B. G.; Snapper, M. L.
Org. Lett. 2003, 5, 1855–1858.
(18) In their 2005 computational study, Lippstreu and Straub
explicitly treated alkyne polymerization as a potential side-reaction in
Ru-catalyzed enyne cross-metathesis; ref 12. A side-reaction yielding
alkyne cyclodimers was observed by Diver and co-workers, for which an
initial sequence corresponding to Scheme 2a was proposed; ref 15a. The
susceptibility of the Ru catalysts to alkyne polymerization was accepted
by Hoveyda and Schrock (ref 16), and segregation of the catalyst in a
deactivation process was suggested. Alkyne polymerization should be
distinguished from potentially competing ring-opening metathesis po-
lymerization in intermolecular enyne metathesis of cycloalkenes. See,
e.g., (a) Kulkarni, A. A.; Diver, S. T. Org. Lett. 2003, 5, 3463–3466.
(b) Banti, D.; North, M. Adv. Synth. Catal. 2002, 344, 694–704.
(19) Kinoshita, A.; Sakakibara, N.; Mori, M. J. Am. Chem. Soc. 1997,
119, 12388–12389.
(20) Stragies, R.; Schuster, M.; Blechert, S. Angew. Chem., Int. Ed.
Engl. 1997, 36, 2518–2520.
(21) For selected examples of successful RCEYM in the absence of
ethylene, in which propargylic substitution emerges as a common
structural motif, see ref 6b and: (a) Timmer, M. S. M.; Ovaa, H.;
Filippov, D. V.; van der Marel, G. A.; van Boom, J. H. Tetrahedron Lett.
15921
dx.doi.org/10.1021/ja207388v |J. Am. Chem. Soc. 2011, 133, 15918–15921