Stereoselective Synthesis of Exocyclic Tetrasubstituted Vinyl Halides via Ru-Catalyzed Halotropic Cycloisomerization of 1,6-Haloenynes

Article abstract of DOI:10.1021/acs.orglett.7b00879

Herein, a ruthenium-catalyzed cycloisomerization that transforms 1,6-haloenynes into 5-membered carbo- and heterocycles that bear exocyclic, stereodefined, tetrasubstituted vinyl halides is reported. The reaction is insensitive to air and water, tolerates a variety of functional groups, and proceeds with good to excellent stereoselectivity and yield.

Full text of DOI:10.1021/acs.orglett.7b00879

Letter
pubs.acs.org/OrgLett
Stereoselective Synthesis of Exocyclic Tetrasubstituted Vinyl Halides
via Ru-Catalyzed Halotropic Cycloisomerization of 1,6-Haloenynes
Barry M. Trost* and Christopher A. Kalnmals
Department of Chemistry, Stanford University, Stanford, California 94305, United States
S
* Supporting Information
ABSTRACT: Herein, a ruthenium-catalyzed cycloisomerization that
transforms 1,6-haloenynes into 5-membered carbo- and heterocycles that
bear exocyclic, stereodefined, tetrasubstituted vinyl halides is reported.
The reaction is insensitive to air and water, tolerates a variety of
functional groups, and proceeds with good to excellent stereoselectivity
and yield.
ketones. When model substrate 3a was subjected to reaction
ycloisomerizations are attractive reactions due to their
Cability to rapidly build molecular complexity from simpler conditions similar to those employed for the interrupted
metallo-ene reaction, none of the expected ketone product 6
formed (Scheme 1, bottom). Instead, partial conversion to vinyl
chloride 4a was observed, along with a small amount of
dienamide 5. In our prior studies with chlorodienes,
adventitious water was a potent trapping agent, so it was
surprising that no hydration was observed with enynes under
similar conditions.
starting materials while maintaining perfect atom economy.
Transition-metal-catalyzed cycloisomerizations are particular
useful, since judicious choice of the metal can impart differential
reactivity and selectivity. Several reviews on transition-metal-
catalyzed cycloisomerizations,1 their application to total
syntheses,2 and asymmetric variants3 have been published.
After reporting the first transition-metal-catalyzed enyne
cycloisomerization,4 our group has maintained an interest in
this field, particularly in cyclizations catalyzed by Pd5 and Ru.6
Recently, our group reported an asymmetric interrupted
metallo-ene reaction wherein a novel ruthenium-phosphor-
amidite catalyst system transformed 1,6- and 1,7-chlorodienes 1
into five- and six-membered rings 2 bearing exocyclic alcohols
with excellent enantio- and diastereoselectivity (Scheme 1,
top).7 In light of these results, we sought to examine this
methodology further in the context of 1,6-chloroenynes,
envisioning that the additional unsaturation in the enyne
would lead to the formation of the corresponding exocyclic
This intriguing and unexpected result prompted us to
examine this reaction further, particularly since 4a was formed
with good E/Z selectivity. Additionally, a survey of the
literature revealed that while similar cycloisomerization
reactions are precedented for alkynes conjugated with a
carbonyl group, few examples have been reported for
electron-neutral 1,6-chloroenynes. Rh(I)/Ag(I)8 and Pd(II)9
catalyze such cycloisomerizations, but both processes are
limited to allylic chlorides and simple, unfunctionalized alkynes.
Given our previous success with phosphoramidite L1, we
initiated our studies with this ligand. In addition to L1, we
explored several other chiral ligands, but to no avail; vinyl
chloride 4a always formed with high E/Z selectivity, but the
major E isomer was always racemic, leading us to believe that
an external ligand was probably not involved in the
enantiodetermining step. Further exploration of the isomer-
ization with 3b (Table 1) supported this hypothesis, as the
reaction proceeded equally well with or without PPh3. Both
acetone and THF were suitable solvents, and a control
experiment run in the absence of CpRu(MeCN)3PF6 resulted
in no product formation. This reaction was quite robust, and all
reactions were performed in ambient air using bulk solvents.
Going forward, we evaluated the scope of this reaction in THF
(Scheme 2).
Scheme 1. Summary of Prior Work and Initial Results
The cycloisomerization does not depend on the starting
olefin geometry; a 4:1 E/Z mixture of 3b afforded identical
results as the pure E isomer. Silyl ether 3c was also a viable
Received: March 24, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00879
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
to reoptimize the reaction conditions (Table 2). Changing from
THF to acetone did not help, and adding PPh3 slowed the
Table 1. Optimization for 1,6-Chloroenynes
a
Table 2. Optimization for 1,6-Bromoenynes
entry
ligand
solvent
result
1
2
3
4
PPh3
PPh3
−
−
−
acetone
THF
84%, >25:1 E/Z
90%, >25:1 E/Z
97%, >25:1 E/Z
Quant., >25:1 E/Z
NR
entry
1
solvent
t (°C)
50
result
acetone
THF
THF
50% conversion
33% conversion
50% conversion
23% conversion
89%, 17:1 Z/E
b
b
2
THF
50
5
THF
a
3
4
5
acetone
DMF
MEK
50
All reactions on 0.05 mmol scale. Yields are isolated yields. Geometry
c
confirmed by NOE and E/Z selectivity determined by 1H NMR.
50 → 70
b
70
Control experiment without CpRu(MeCN)3PF6.
a
All reactions on 0.05 mmol scale. Conversion determined by 1H
a
b
c
Scheme 2. Substrate Scope for 1,6-Chloroenynes
NMR. 6 mol % PPh3. Full conversion after additional 10 h at 70 °C.
reaction further. Although the reaction was slow in DMF at 50
°C, full conversion was achieved when the temperature was
increased to 70 °C, leading us to believe that a higher
temperature was necessary. Indeed, when the reaction was
performed at 70 °C in 2-butanone (MEK), full conversion was
reliably obtained. This observation is consistent with work by
Itoh, who showed that a π-allylruthenium(IV) bromide
underwent reductive C−X bond formation slower than the
corresponding chloride.14
Alkyl-substituted alkynes 7a and 7b were excellent substrates
for this transformation, as were silyl ethers 7c and 7f (Scheme
3). Primary alcohols and alkyl bromides (7g and 7h,
respectively) were also well tolerated. Remarkably, 8d was
obtained cleanly, without any redox isomerization byprod-
ucts.15 While a remote phenyl group (7i) had little impact on
the E/Z selectivity, a conjugated phenyl group (7e) gave an E/
Z ratio of 6:1.
a
All reactions on 0.10 mmol scale. Yields are isolated yields. Geometry
confirmed by NOE, and E/Z selectivity determined by 1H NMR after
b
c
purification. 4b was a 4:1 E/Z mixture. TBSCl/imidazole added.
d
e
Performed at 60 °C. Major isomer assigned by analogy.
Ethereal bromoenynes were also viable substrates, and while
E/Z ratios were slightly lower than those observed for
sulfonamides, it was possible to perform these reactions at
decreased catalyst loading (2.5 mol %). Ethers bearing alkyl
(7j), phenylsulfonyl (7k), silyl ether (7m−n), and aryl (7o)
functionalities were all successfully employed. Unlike 8m, 8n
did not require resilylation, demonstrating the robustness of the
TBDPS group in this reaction. Nitrile 7l, which could have
acted as a Lewis basic catalyst poison, was also amenable to
cycloisomerization when 5 mol % catalyst was used. A variety of
carbon-linked bromoenynes were successfully employed in this
reaction as well; malonates (7p−t) were excellent cyclo-
isomerization substrates, as was bis(sulfone) 7u.
Given the reduced E/Z selectivity observed with phenyl-
acetylene-derived substrates, we were curious whether the
electronic nature of the aryl substituent would affect the
geometrical selectivity of the cycloisomerization (Table 3).
Electron-neutral 7w and electron-deficient 7x afforded nearly
identical results. With electron-rich 7y, slightly lower E/Z
selectivity was observed along with a small amount of ketone
9y, which was isolated as a single diastereomer. Compound 7z
cyclized with essentially the same E/Z ratio as 7y, and the yield
of 9 was also identical. Interestingly, when NaHCO3 was added
to the cycloisomerization of 7z, none of the ketone product
(9z) formed (entry 4).
substrate, and although the TBS group was partially hydrolyzed
during the reaction, it was easily reinstalled in the same pot by
adding TBSCl and imidazole once the cycloisomerization was
complete. Alkynes bearing conjugated π-unsaturation such as
enyne 3d and phenylacetylene 3e gave reduced E/Z selectivity,
and in the case of the latter, a higher temperature was required
to achieve full conversion. The reaction was not limited to
sulfonamide-linked enynes, as malonate 3f also cyclized with
excellent yield and Z/E selectivity.10
At this point, we were curious to see whether this
methodology would also apply to 1,6-bromoenynes. Vinyl
bromides are synthetically versatile intermediates, and
furthermore, literature reports of a comparable cycloisomeriza-
tion leading to vinyl bromides are scarce. Jang and co-workers
reported a Sn(II)/Pt(II)-catalyzed cycloisomerization of 1,6-
bromoenynes that was limited to aryl alkynes and gave either
complete complementary geometrical selectivity to that which
we observe or no selectivity at all, depending on the substrate.11
Metallic indium12 and In(III) salts13 also promote the
cyclization of 1,6-bromoenynes. An atom transfer pathway is
invoked, and the identity of the vinyl halide is largely
determined by the choice of halogenated solvent. As with the
Pt(II)-catalyzed cycloisomerization, these indium-based meth-
ods provide a net trans addition of the allylic halide across the
alkyne, whereas our method is selective for the net cis addition.
Since most bromoenynes underwent only partial conversion
under our conditions for chloroenynes, we used bis(sulfone) 7u
We propose the catalytic cycle shown in Scheme 4. The
substrate (A) coordinates to ruthenium via both the alkene and
the alkyne, giving rise to B. While B is likely ligated in its
B
DOI: 10.1021/acs.orglett.7b00879
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Scheme 3. Substrate Scope for 1,6-Bromoenynes
Scheme 4. Proposed Catalytic Cycle
7e, and 7w−z), may increase the lifetime of D and allow for
olefin isomerization prior to reductive elimination, which could
account for the lower selectivities observed with these
substrates. While vinyl iodides have been prepared via metal-
catalyzed radical atom transfer processes, these reactions are
typically unselective with regards to olefin geometry16 or place
the halogen on the less hindered face of the alkene.17 Since our
reaction installs the halogen on the more hindered alkene face,
we do not believe a radical process is operative. Furthermore,
the addition of 5 mol % of either TEMPO or BHT has no effect
on the cycloisomerization of 7b to 8b, providing additional
evidence against a radical mechanism.
While we were unable to render our cycloisomerization
asymmetric, we were curious to see if it might be
diastereoselective, particularly since enantiomerically enriched
propargyl stereocenters are readily accessible. To this end,
chiral propargyl alcohol 10 was prepared via Zn-ProPhenol-
catalyzed alkynylation18 and alkylated under phase-transfer
conditions to afford 7aa, which was transformed into α-
aryltetrahydrofuran 8aa with 5:1 dr (Scheme 5).
a
All reactions on 0.10 mmol scale. Yields are isolated yields. Geometry
1
confirmed by NOE and E/Z selectivity determined by H NMR after
purification. 7j−k and m−o 2.5% Ru at 2.0 M; all others 5% Ru at 1.0
b
c
M. TBSCl/imid. added. 44% yield of 8f + 38% yield of desilylated 8f.
a
d
e
f
Scheme 5. Diastereoselective Cycloisomerization
62% yield, 5:1 E/Z on 1.16 mmol. E/Z by crude NMR. 12% yield of
g
h
pure Z isomer. Major isomer assigned by analogy. 78% yield, >25:1
i
j
Z/E on 0.69 mmol. 12% yield of desilylated 8s. 0.05 mmol scale.
k
25% yield of desilylated 8v.
Table 3. Examination of Phenylacetylene-Derived
a
Substrates
entry
Ar
result
a
Cycloisomerization performed on 0.05 mmol scale. Stereochemistry
1
1
2
3
Ph (7w)
72% of 8w, 5:1 E/Z
confirmed by NOE; dr determined by crude H NMR.
4-AcC6H4 (7x)
78% of 8x, 5:1 E/Z
4-OMeC6H4 (7y)
50% of 8y, 3:1 E/Z + 11% of 9y
82% of 8z, 3:1 E/Z
b
The products of this cycloisomerization reaction contain a
unique juxtaposition of functionality, the versatility and utility
of which is highlighted by some of the derivitizations we
performed (Scheme 6). Suzuki coupling between 8k and 2-
vinylphenylboronic acid followed by Grubbs ring-closing
metathesis afforded 6−7−5 tricycle 11, which resembles the
skeleton of rubriflordilactone A. A similar compound with a
pyrrolidine in place of the tetrahydrofuran has been reported as
an epithelial sodium channel blocker,19 and 6−7−5 carbocycles
have been studied as steroid mimics.20 By changing the linker in
7, our method could potentially be used to access all three sets
of compounds. Buchwald’s copper-free conditions21 were
uniquely effective for the Sonogashira coupling of 8r, and the
resulting dienyne underwent a Pauson−Khand reaction to
4
3,4,5-(OMe)3C6H2 (7z)
3,4,5-(OMe)3C6H2 (7z)
5
68% of 8z, 3.4:1 E/Z + 10% of 9z
a
All reactions on 0.10 mmol scale. Yields are isolated yields. Geometry
confirmed by NOE, and E/Z selectivity determined by crude 1H
b
NMR. 1.0 equiv of NaHCO3 was added.
resting state, ligand dissociation to a 16-electron complex
probably occurs prior to oxidative addition to form C, which is
consistent with our observation that external ligands do not
affect the outcome of the reaction. Migratory insertion to form
vinylruthenium(IV) halide D followed by syn-reductive
elimination affords the product (E). An R group capable of
stabilizing positive charge, such as an alkene (3d) or arene (3e,
C
DOI: 10.1021/acs.orglett.7b00879
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 6. Derivitizations of Vinyl Bromides
REFERENCES
■
(1) (a) Ojima, I.; Tzamarioudaki, M.; Li, Z.; Donovan, R. J. Chem.
Rev. 1996, 96, 635. (b) Trost, B. M.; Krische, M. J. Synlett 1998, 1998,
1. (c) Lloyd-Jones, G. C. Org. Biomol. Chem. 2003, 1, 215.
(2) Stathakis, C. I.; Gkizis, P. L.; Zografos, A. L. Nat. Prod. Rep. 2016,
33, 1093.
(3) (a) Marinetti, A.; Jullien, H.; Voituriez, A. Chem. Soc. Rev. 2012,
41, 4884. (b) Watson, I. D. G.; Toste, F. D. Chem. Sci. 2012, 3, 2899.
(4) Trost, B. M.; Lautens, M. J. Am. Chem. Soc. 1985, 107, 1781.
(5) Trost, B. M. Acc. Chem. Res. 1990, 23, 34.
(6) For reviews, see: (a) Trost, B. M.; Toste, F. D.; Pinkerton, A. B.
Chem. Rev. 2001, 101, 2067. (b) Trost, B. M.; Frederiksen, M. U.;
Rudd, M. T. Angew. Chem., Int. Ed. 2005, 44, 6630.
(7) Trost, B. M.; Ryan, M. C. J. Am. Chem. Soc. 2016, 138, 2981.
(8) (a) Tong, X.; Li, D.; Zhang, Z.; Zhang, X. J. Am. Chem. Soc. 2004,
126, 7601. (b) Wang, J.; Xie, X.; Ma, F.; Peng, Z.; Zhang, L.; Zhang, Z.
Tetrahedron 2010, 66, 4212.
(9) Zhu, G.; Zhang, Z. J. Org. Chem. 2005, 70, 3339.
(10) The E/Z designation changes based on the linker due to the
Cahn−Ingold−Prelog convention. For heteroatom-linked substrates,
the major olefin isomer is E; for carbon-linked substrates, the major
olefin isomer is Z.
afford 12 as a single diastereomer. This sequence is noteworthy,
as 12 maps nicely onto the carbon skeleton of several natural
products, including hypnophilin and complicatic acid. Hydro-
boration−oxidation of 8i was highly selective for the terminal
alkene and afforded 13 with the vinyl bromide intact and
primed for further functionalization.
In conclusion, we have developed a Ru-catalyzed halotropic
cycloisomerization that transforms 1,6-haloenynes into five-
membered rings that bear exocyclic tetrasubstituted vinyl
chlorides and bromides. The reaction proceeds with good to
excellent stereoselectivity and affords the products cleanly in
high yields. Inert conditions are not required, and the reaction
tolerates a variety of functionality, including protected and
unprotected alcohols, alkyl bromides, nitriles, and aromatic
rings. The vinyl bromides accessible using this cycloisomeriza-
tion are versatile building blocks and can be rapidly
transformed into complex carbon skeletons that may have
utility in the synthesis of natural products and pharmaceuticals.
(11) Jang, M.-S.; Wang, X.; Jang, W.-Y.; Jang, H.-Y. Organometallics
2009, 28, 4841.
(12) Bhatti, N. H.; Salter, M. M. Tetrahedron Lett. 2004, 45, 8379.
(13) Cook, G. R.; Hayashi, R. Org. Lett. 2006, 8, 1045.
(14) Nagashima, H.; Mukai, K.; Shiota, Y.; Yamaguchi, K.; Ara, K.-I.;
Fukahori, T.; Suzuki, H.; Akita, M.; Moro-oka, Y.; Itoh, K.
Organometallics 1990, 9, 799.
(15) Trost, B. M.; Livingston, R. C. J. Am. Chem. Soc. 2008, 130,
11970.
(16) (a) Monks, B. M.; Cook, S. P. Angew. Chem., Int. Ed. 2013, 52,
14214. (b) Shen, Y.; Cornella, J.; Julia-
Catal. 2017, 7, 409.
́ ́
Hernandez, F.; Martin, R. ACS
(17) Curran, D. P.; Chen, M.-S.; Kim, D. J. Am. Chem. Soc. 1986, 108,
2489.
(18) Trost, B. M.; Weiss, A. H.; von Wangelin, A. J. J. Am. Chem. Soc.
2006, 128, 8.
(19) Bhalay, G.; Budd, E.; Bloomfield, C. G.; Collingwood, S. P.;
Dunstan, A.; Edwards, L.; Gedeck, P.; Howsham, C.; Hunt, P.; Hunt,
T. A.; Oakley, P.; Smith, N. Organic Compounds. U.S. Pat. Appl. US
2010/0130506 A1, May 27, 2010.
(20) Dodd, J. H.; Dixon, L. A.; Bullington, J. L.; Schwender, C. F.
Substituted Dibenz (A F)Azulenes and Methods of Preparation. U.S.
Patent 5,834,521, Nov 10, 1998.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b00879.
(21) Gelman, D.; Buchwald, S. L. Angew. Chem., Int. Ed. 2003, 42,
5993.
Experimental procedures and characterization data for all
compounds and 1H/13C NMR spectra for 3, 4, 7, 8, and
11−13 (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: bmtrost@stanford.edu.
ORCID
Barry M. Trost: 0000-0001-7369-9121
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the NSF (CHE-1360634) and the NIH (GM-
033049) for financial support of our programs. C.A.K. thanks
Michael C. Ryan (University of WisconsinMadison) for
helpful discussions.
D
DOI: 10.1021/acs.orglett.7b00879
Org. Lett. XXXX, XXX, XXX−XXX