Broadly applicable Z- and diastereoselective ring-opening/cross-metathesis catalyzed by a dithiolate Ru complex

Article abstract of DOI:10.1002/anie.201309430

A broadly applicable Ru-catalyzed protocol for Z-selective ring-opening/cross-metathesis (ROCM) is disclosed. In addition to reactions relating to terminal alkenes of different sizes, the first examples of Z-selective ROCM processes involving heteroaryl olefins, 1,3-dienes, and O- and S-substituted alkenes as well as allylic and homoallylic alcohols are reported. Z-Selective transformations with an α-substituted allylic alcohol are shown to afford congested Z alkenes with high diastereoselectivity. Transformations are performed in the presence of 2.0-5.0 mol % of a recently disclosed Ru-based dithiolate complex that can be easily prepared in a single step from commercially available starting materials. Typically, transformations proceed at ambient temperature and are complete within eight hours; products are obtained in up to 97 % yield, >98:2 Z/E, and >98:2 diastereomeric ratio. The present investigations reveal a mechanistically significant attribute of the Ru-based dithiolates that arises from electrostatic interactions with anionic S-based ligands. Finally, the full range: For the first time, Ru-catalyzed olefin metathesis transformations with heteroaryl-substituted alkenes, 1,3-dienes, enol ethers, and allylic and homoallylic alcohols can be performed efficiently and often with >98 % Z selectivity. Hydrogen bonding with the dithiolate ligand likely facilitates ring-opening/cross-metathesis reactions of allylic alcohols. Copyright

Full text of DOI:10.1002/anie.201309430

.
Angewandte
Communications
DOI: 10.1002/anie.201309430
Olefin Metathesis
Broadly Applicable Z- and Diastereoselective Ring-Opening/Cross-
Metathesis Catalyzed by a Dithiolate Ru Complex**
Ming Joo Koh, R. Kashif M. Khan, Sebastian Torker, and Amir H. Hoveyda*
Abstract: A broadly applicable Ru-catalyzed protocol for Z-
selective ring-opening/cross-metathesis (ROCM) is disclosed.
In addition to reactions relating to terminal alkenes of different
sizes, the first examples of Z-selective ROCM processes
involving heteroaryl olefins, 1,3-dienes, and O- and S-substi-
tuted alkenes as well as allylic and homoallylic alcohols are
reported. Z-Selective transformations with an a-substituted
allylic alcohol are shown to afford congested Z alkenes with
high diastereoselectivity. Transformations are performed in the
presence of 2.0–5.0 mol% of a recently disclosed Ru-based
dithiolate complex that can be easily prepared in a single step
from commercially available starting materials. Typically,
transformations proceed at ambient temperature and are
complete within eight hours; products are obtained in up to
97% yield, > 98:2 Z/E, and > 98:2 diastereomeric ratio. The
present investigations reveal a mechanistically significant
attribute of the Ru-based dithiolates that arises from electro-
static interactions with anionic S-based ligands.
trans products with other substrate types.[9] Z-Selective olefin
metathesis with 1,3-dienes is scarce; the existing cases involve
Mo-catalyzed homocoupling[10] and two cross-metathesis
reactions.[4b] Instances of Z- and diastereoselective ROCM
are unknown.
Herein, we report Ru-catalyzed Z-selective ROCM
reactions of aliphatic as well as heteroaryl olefins, 1,3-
dienes, O- and S-substituted alkenes, and allylic and homo-
allylic alcohols. Transformations are performed in the pres-
ence of 2.0–5.0 mol% of a readily accessible Ru dithiolate
complex, typically proceed at ambient temperature, and are
complete within eight hours; desired products are generated
in up to 97% yield, > 98:2 Z/E, and > 98:2 diastereomeric
ratio (d.r.). Our findings suggest that the anionic disulfide
ligand likely facilitates ROCM by forming a proton bridge
with an allylic alcohol derived carbene.
We have demonstrated that Ru carbene 1, accessed in
a single step from a commercially available Ru dichlorocar-
bene and the disodium salt of dithiocatechol, can be used to
promote efficient and exceptionally Z-selective ring-opening
metathesis polymerization and ROCM;[11] reaction rates
appear to be similar to those of processes catalyzed by
commonly used (non-stereoselective) Ru-based dichlorides.
Dithiolate complex 1 was conceived based on the principle
that ruthenacyclobutanes would be formed exclusively syn to
the N-heterocyclic carbene (NHC) (I vs. II or III, Scheme 1),
and that steric repulsion between the metallacycle substitu-
ents and the sizeable NHC leads to a preference for cis
products. We have demonstrated that, in contrast to the other
Z-selective Ru complexes,[6] 1 can be utilized with sterically
hindered alkenes (e.g., styrenes).
We have since focused on additional key questions. One is
whether reactions with alkenes bearing relatively small
groups would deliver high Z/E ratios, in spite of the reduced
steric repulsion with mesityl moieties of the NHC ligand (cf. II
and III, Scheme 1). Efficient transformations with hindered
alkenes are readily catalyzed; however, there are complica-
tions associated with reactions of smaller substrates. Unhin-
dered olefins undergo more facile homocoupling to generate
ethylene and the somewhat unstable methylidene complex.
Additionally, processes of the more diminutive substrates
afford relatively exposed disubstituted alkenes that are more
susceptible to post-metathesis isomerization.[4a,c] Another
issue is whether reactions with the sparsely examined enol
ethers, dienes, and heteroaryl alkenes can proceed efficiently
and stereoselectively. Finally, we sought to explore the
possibility of allylic or homoallylic alcohols serving as
effective substrates in ROCM with the new Ru complex.
We first probed the influence of the size of the terminal
alkene cross-metathesis partners on the efficiency and
A
transformative development in olefin metathesis is the
recent emergence of catalysts for efficient synthesis of Z
alkenes.[1] The first advance was reported in 2009 in con-
nection with a monopyrrolide-aryloxide Mo complex pro-
moting ring-opening/cross-metathesis (ROCM) processes
with a strong preference for cis olefin products.[2] Mo- and
W-catalyzed Z-selective homocoupling,[3] cross-metathesis,[4]
and macrocyclic ring-closing metathesis[5] were subsequently
introduced. Since 2011, several Ru carbenes have also been
shown to catalyze different Z-selective olefin metathesis
reactions.[6,7] The complementary profiles of high-oxidation-
state alkylidenes and later-transition-metal carbenes impart
significant scope to this important set of catalytic trans-
formations.[8] The latest progress notwithstanding, major
shortcomings remain unaddressed. One limitation is the
persisting lack of Z-selective reactions with allylic or homo-
allylic alcohols, processes that are in the exclusive purview of
Ru-based catalysis. There are no instances of transformations
that afford cis heteroaryl-substituted alkenes, and the lone
examples of Ru-catalyzed Z-selective ROCM with O- and S-
substituted olefins are promoted by a complex that furnishes
[*] M. J. Koh, R. K. M. Khan, Dr. S. Torker, Prof. A. H. Hoveyda
Department of Chemistry, Merkert Chemistry Center, Boston
College
Chestnut Hill, MA 02467 (USA)
E-mail: amir.hoveyda@bc.edu
[**] Financial support was provided by the NSF (CHE-1111074).
R.K.M.K. is grateful to AstraZeneca for a graduate fellowship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201309430.
1968
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1968 –1972

Angewandte
Chemie
affording Z-3 in 89% yield as the sole isomer (< 2% E); when
the b-substituent is the smaller p-methoxyphenyl group,
ROCM remains efficient and exceptionally Z selective (cf.
4, Scheme 2). g,d-Unsaturated amide 5, in which an additional
=
methylene unit separates the C C bond and the carbonyl-
containing moiety, is formed with complete Z selectivity and
in 65% yield. ROCM is likely less efficient owing to more
competitive homocoupling of the terminal olefins and gen-
eration of the less robust methylidene. In further support of
this hypothesis, under identical conditions, ROCM of 2 with
a homoallylic silyl ether proceeds to 87% conversion,
affording 6 in 68% yield and > 98:2 Z/E. When the least
hindered 1-decene is used, there is 91% conversion (disap-
pearance of 2) and diene 7 is isolated in 58% yield, yet the
corresponding E alkene remains undetected. Transformation
of cyclobutene 8 to diene 9 (58% yield, > 98:2 Z/E) provides
an additional example with a different cyclic olefin. The
above findings indicate that exceptional Z selectivity persists
despite the diminishing size of the substituent of a cross-
metathesis partner. The steric repulsion between a mesityl
moiety of the NHC and the substituents of a ruthenacyclobu-
tane appears to be sufficient for the intermediacy of II or III
to remain non-competitive with I even when R is relatively
small.
Scheme 1. Ru dithiolate 1, the model for the origin of Z selectivity and
potential intermediates I–III generated in reactions with alkenes.
stereoselectivity of ROCM reactions with diol 2. With
allyltrimethylsilane, where the alkene and the sizeable
moiety are linked by a methylene unit (vs. styrenes or
vinylcyclohexane), reaction proceeds to > 98% conversion,
Ru-catalyzed ROCM reactions can be used to produce
heterocycle-substituted Z alkenes. As represented by syn-
thesis of 10 and 11 (Scheme 3), products are obtained in 93–
97% yield and 93:7 to > 98:2 Z/E in no more than two hours
at ambient temperature. We then examined the correspond-
ing transformations with 1,3-dienes. In the presence of 2.0–
Scheme 2. Ru-catalyzed ROCM of 2 with different cross-metathesis
partners indicates that exceptionally high Z selectivity can be isolated
with diminishing substituent size. TBS=tert-butyl(dimethyl)silyl.
Scheme 3. Z-selective ROCM with heterocyclic alkenes and linear 1,3-
dienes; same conditions used as shown in Scheme 2.
Angew. Chem. Int. Ed. 2014, 53, 1968 –1972
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 1969

.
Angewandte
Communications
5.0 mol% 1, use of (E)-1-methoxy-1,3-butadiene generates
12a in 84% yield and 91:9 Z/E, and 12b is isolated in 80%
yield and > 98% Z selectivity when (E)-deca-1,3-diene is
employed; the latter processes proceed to completion in two
hours at 228C. Two other examples involving cyclobutene 8,
delivering 14a and 14b (88% and 60% yield, respectively)
without a trace of the E,E-diene isomer (< 2%), are shown in
Scheme 3.[12] Thus, reactions with 1,3-dienes, such as 13a,b,
proceed with high Z selectivity despite the involvement of
acyclic substrates that are smaller than the aryl-substituted
alkenes (cf. 10,11).
There is only one report of Z-selective Ru-catalyzed
ROCM reactions of enol ethers or vinyl sulfides.[11] In the said
study, a stereogenic-at-Ru carbene bearing a bidentate N-
heterocyclic ligand was used, and it was the higher energy
carbene diastereomer[13] that promoted the desired trans-
formation. However, the ability of the iodoaryloxide Ru
complex to catalyze Z-selective ROCM processes does not
extend to other classes of alkenes (e.g., aryl or aliphatic
olefins), where usually E isomers are obtained exclusively. In
contrast, dithiolate complex 1 can be employed to generate Z
enol ethers (Scheme 4). Transformations are efficient: with
2.0–5.0 mol% 1, the desired O- or S-substituted carbo- or
heterocyclic (15a,b and 19, Scheme 4) as well as acyclic
products (16 and 17) are obtained in 79–95% yield and 88:12
to > 98:2 Z/E.
A recent investigation regarding Ru-catalyzed Z-selective
ROCM entailed transformations of norbornene-derived bis-
benzyl ether or bis-acetate with allyl acetate;[14] reactions with
either or both of the parent alcohols were not mentioned. We
therefore turned to the possibility of using allylic or homo-
allylic alcohols as cross-metathesis partners. ROCM of
norbornene and allyl alcohol in the presence of 5.0 mol%
1 is complete in two hours (228C), affording 20 in 68% yield
Scheme 5. Z-selective ROCM with allyl alcohol and homoallyl alcohol;
reaction with an allyl ether is inefficient but (pinacolato)allylboron and
homoallyl ether are suitable substrates. B(pin)=(pinacolato)boron.
and 88:12 Z/E (Scheme 5). Likewise, homoallylic alcohol 21 is
formed in 84% yield and 87% Z selectivity. Z-Allylboron
compounds, precursors to useful organic molecules (e.g.,
allylic amines[15]) can be readily accessed: ROCM with
(pinacolato)allylboron generates 22 in 64% yield and 90:10
Z/E.
Contrary to the parent alcohol, ROCM of norbornene
with allyl n-butyl ether gives rise to only 20% conversion to
23 even after 24 h (vs. > 98% conv. in 2.0 h with allyl alcohol);
such a difference in reactivity does not apply to
homoallyl alcohol and its alkyl ether (cf. 21 and 24). In
view of the efficiency of reactions with linear alkenes
(cf. Scheme 2), it appears that insertion of an ether
oxygen at the allylic position (cf. 23) induces an
unfavorable effect, one that is no longer present when
allyl alcohol is used (cf. 20). To gain further insight, we
examined the reaction of an enantiomerically
enriched secondary allylic alcohol, since the ease
with which the sterically congested Z-alkene is
generated could further substantiate the positive
impact of the hydroxyl group. Moreover, whether
the latter ROCM proceeds diastereoselectively would
shed additional light on the nature of the positive
influence of the alcohol unit.
Subjection of commercially available allylic alco-
hol 25 (96:4 enantiomeric ratio) with cyclic alkene 2
(5:1) in the presence of 5.0 mol% 1 affords triol 26 in
67% yield as a single diastereomer (> 98:2 Z/E and
> 98:2 d.r.; Scheme 6a). The identity of the major
isomer was established by the X-ray structure of the
phenylboronate derivative 27 (Scheme 6b). When 25
is treated with cyclopropene 28 (1:2), under otherwise
the same conditions, the congested cis alkene 29 is
obtained in 78% yield, 91% Z selectivity and 93:7 d.r.
Scheme 4. ROCM reactions with O- or S-substituted alkenyl ethers proceed
efficiently and with high Z selectivity.
1970
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1968 –1972

Angewandte
Chemie
containing a proximal hydroxy group, likely because the
sulfide group positioned opposite to the donor NHC ligand
participates in favorable electrostatic interactions.
Development of more efficient catalysts and other types
of stereoselective olefin metathesis reactions are in progress.
Received: October 29, 2013
Keywords: alkenes · olefin metathesis ·
.
ring-opening/cross-metathesis · ruthenium · synthetic methods
[1] A. Fꢀrstner, Science 2013, 341, 1357 – 1364.
[2] a) I. Ibrahem, M. Yu, R. R. Schrock, A. H. Hoveyda, J. Am.
Chem. Soc. 2009, 131, 3844 – 3845; For subsequent studies
involving enol cross-metathesis partners, see: b) M. Yu, I.
Ibrahem, M. Hasegawa, R. R. Schrock, A. H. Hoveyda, J. Am.
Chem. Soc. 2012, 134, 2788 – 2799.
[3] a) A. J. Jiang, Y. Zhao, R. R. Schrock, A. H. Hoveyda, J. Am.
Chem. Soc. 2009, 131, 16630 – 16631; b) E. M. Townsend, R. R.
Schrock, A. H. Hoveyda, J. Am. Chem. Soc. 2012, 134, 11334 –
11337.
[4] a) S. J. Meek, R. V. OꢁBrien, J. Llaveria, R. R. Schrock, A. H.
Hoveyda, Nature 2011, 471, 461 – 466; b) E. T. Kiesewetter, R. V.
OꢁBrien, E. C. Yu, S. J. Meek, R. R. Schrock, A. H. Hoveyda, J.
Am. Chem. Soc. 2013, 135, 6026 – 6029; c) T. J. Mann, A. W. H.
Speed, R. R. Schrock, A. H. Hoveyda, Angew. Chem. 2013, 125,
8553 – 8558; Angew. Chem. Int. Ed. 2013, 52, 8395 – 8400.
[5] a) M. Yu, C. Wang, A. F. Kyle, P. Jakubec, D. J. Dixon, R. R.
Schrock, A. H. Hoveyda, Nature 2011, 479, 88 – 93; b) C. Wang,
M. Yu, A. F. Kyle, P. Jakubec, D. J. Dixon, R. R. Schrock, A. H.
Hoveyda, Chem. Eur. J. 2013, 19, 2726 – 2740; c) C. Wang, F.
Haeffner, R. R. Schrock, A. H. Hoveyda, Angew. Chem. 2013,
125, 1993 – 1997; Angew. Chem. Int. Ed. 2013, 52, 1939 – 1943.
[6] For representative reports regarding homocoupling and cross-
metathesis processes, see: a) K. Endo, R. H. Grubbs, J. Am.
Chem. Soc. 2011, 133, 8525 – 8527; b) B. K. Keitz, K. Endo, P. R.
Patel, M. B. Herbert, R. H. Grubbs, J. Am. Chem. Soc. 2012, 134,
693 – 699; c) M. B. Herbert, V. M. Marx, R. L. Pederson, R. H.
Grubbs, Angew. Chem. 2012, 124, 317 – 319; Angew. Chem. Int.
Ed. 2012, 51, 310 – 314; d) G. Occhipinti, F. R. Hansen, K. W.
Tçrnroos, V. R. Jensen, J. Am. Chem. Soc. 2013, 135, 3331 – 3334.
[7] For Ru-catalyzed Z-selective macrocyclic ring-closing metathe-
sis (RCM), see: L. E. Rosebrugh, M. B. Herbert, V. M. Marx,
B. K. Keitz, R. H. Grubbs, J. Am. Chem. Soc. 2013, 135, 1276 –
1279.
[8] A. H. Hoveyda, A. R. Zhugralin, Nature 2007, 450, 243 – 251.
[9] R. K. M. Khan, R. V. OꢁBrien, S. Torker, B. Li, A. H. Hoveyda, J.
Am. Chem. Soc. 2012, 134, 12774 – 12779.
[10] See Ref. [3b].
[11] R. K. M. Khan, S. Torker, A. H. Hoveyda, J. Am. Chem. Soc.
2013, 135, 10258 – 10261.
[12] The precise origin of the lower stereoselectivity in the formation
of 12a (91:9 Z/E) versus 12b and 14a,b (> 98:2 Z/E) is unclear.
However, examination of selectivity as a function of time
indicates that E-12a is not generated by post-metathesis isomer-
ization.
[13] R. K. M. Khan, A. R. Zhugralin, S. Torker, R. V. OꢁBrien, P. J.
Lombardi, A. H. Hoveyda, J. Am. Chem. Soc. 2012, 134, 12438 –
12441.
[14] J. Hartung, R. H. Grubbs, J. Am. Chem. Soc. 2013, 135, 10183 –
10185.
[15] S. N. Mlynarski, A. K. Karns, J. P. Morken, J. Am. Chem. Soc.
2012, 134, 16449 – 16451, and references therein.
Scheme 6. a) Ru-catalyzed ROCM with enantiomerically enriched allylic
alcohol 25 bears significant mechanistic implications and demon-
strates that congested Z alkenes can be accessed. b) X-ray structure of
the phenylboronate derivative 27. c) Intermedate A. e.r.=enantiomeric
ratio; d.r. =diastereomeric ratio.
There is < 5% conversion after 24 h (228C) when methyl
ether 30 is used with either 2 or 28.
The observations depicted in Scheme 6 underscore a key
mechanistic factor. We have previously shown that allylic
alcohols react significantly faster and with higher diastereo-
selectively than their protected variants in transformations
catalyzed by Ru dichloride complexes.[16] Electrostatic attrac-
tion (“H-bonding”) between the hydroxyl unit and the
anionic halides was put forth as the principal reason for the
rate acceleration and elevated d.r. values. The data in
Scheme 6 indicate that similar principles are likely operative
here (see A, Scheme 6c).[17] The resulting structural organ-
ization leads to high diastereofacial differentiation; addition-
ally, it stabilizes the ruthenacyclobutane and the preceding
transition state by minimizing the trans influence that arises
from the placement of the NHC and sulfide groups. What is
more, the proton bridge dispenses with the electron–electron
repulsion that otherwise exists between the heteroatom-
containing carbene substituent and the nearby sulfide (cf. 23,
Scheme 5).[18] The collective consequence is the distinctive
ability of a Ru dithiolate to catalyze reactions with alkenes
[16] a) A. H. Hoveyda, P. J. Lombardi, R. V. OꢁBrien, A. R. Zhugra-
lin, J. Am. Chem. Soc. 2009, 131, 8378 – 8379; For related studies,
Angew. Chem. Int. Ed. 2014, 53, 1968 –1972
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1971

.
Angewandte
Communications
see: b) T. R. Hoye, H. Zhao, Org. Lett. 1999, 1, 1123 – 1125; c) H.
Fuwa, A. Saito, M. Sasaki, Angew. Chem. 2010, 122, 3105 – 3108;
Angew. Chem. Int. Ed. 2010, 49, 3041 – 3044; d) Y. A. Lin, J. M.
Chalker, B. G. Davis, J. Am. Chem. Soc. 2010, 132, 16805 – 16811;
e) T. J. Donohoe, J. A. Basutto, J. F. Bower, A. Rathi, Org. Lett.
2011, 13, 1036 – 1039.
Org. Chem. 2005, 4995 – 5000; c) E. Schreiner, N. N. Nair, R.
Pollet, V. Staemmler, D. Marx, Proc. Natl. Acad. Sci. USA 2007,
104, 20725 – 20730; d) P. Zhou, F. Tian, F. Lv, Z. Shang, Proteins
Struct. Funct. Bioinf. 2008, 76, 151 – 163.
[18] Similar unfavorable repulsive forces between the oxygen
subsituent and the sulfide ligand are not likely in effect for
vinyl (Scheme 4) and homoallylic ethers (cf. 6 in Scheme 2 and
24 in Scheme 5); it appears that it is the heteroatom within an
allylic ether that engenders repulsive electronic interactions.
[17] For studies on H-bonding interactions with S-containing func-
tional groups, see: a) F. Wennmohs, V. Staemmler, M. Schindler,
J. Chem. Phys. 2003, 119, 3208 – 3218; b) S. B. Tsogoeva, D. A.
Yalalov, M. J. Hateley, C. Weckbecker, K. Hutchmacher, Eur. J.
1972
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1968 –1972