Formation of vinyl halides via a ruthenium-catalyzed three-component coupling

Article abstract of DOI:10.1021/ja011426w

The ruthenium-catalized three-component coupling of an alkyne, an enone, and halide ion to form E- or Z-vinyl halides has been investigated. Through systematic optimization experiments, the conditions effecting the olefin selectivity were examined. In general, more polar solvents such as DMF favored the formation of the E-isomer, and less polar solvents such as acetone favored formation of the Z-isomer. The optimized conditions for the formation of E-vinyl chlorides were found to be the use of cyclopentadienyl ruthenium (II) cyclooctadiene chloride, stannic chloride pentahydrate as a cocatalyst, and for a chloride source, either ammonium chloride in DMF/water mixtures or tetramethylammonium chloride in DMF. A range of several other ruthenium (II) catalysts was also shown to be effective. A wide variety of vinyl chlorides could be formed under these conditions. Substrates with tethered alcohols or ketones either five or six carbons from the alkyne portion gave instead diketone or cyclohexenone products. For formation of vinyl bromides, a catalyst system involving the use of cyclopentadienylruthenium (II) tris(acetonitrile) hexafluorophosphate with stannic bromide as a cocatalyst was found to be most effective. The use of ammonium bromide in DMF/acetone mixtures was optimal for the synthesis of E-vinyl bromides, and the use of lithium bromide in acetone was optimal for formation of the corresponding Z-isomer. Under either set of conditions, a wide range of vinyl bromides could be formed. When alkynes with propargylic substituents are used, enhanced selectivity for formation of the Z-isomer is observed. When aryl acetylenes are used as the coupling partners, complete selectivity for the Z-isomer is obtained. A mechanism involving a cis or trans halometalation is invoked to explain formation of the observed products. The vinyl halides have been shown to be precursors to α-hydroxy ketones and cyclopentenones, and as coupling partners in Suzuki-type reactions.

Full text of DOI:10.1021/ja011426w

Published on Web 05/30/2002
Formation of Vinyl Halides via a Ruthenium-Catalyzed
Three-Component Coupling
Barry M. Trost* and Anthony B. Pinkerton
Contribution from the Department of Chemistry, Stanford UniVersity, Stanford, California 94305
Received June 11, 2001
Abstract: The ruthenium-catalyzed three-component coupling of an alkyne, an enone, and halide ion to
form E- or Z-vinyl halides has been investigated. Through systematic optimization experiments, the conditions
effecting the olefin selectivity were examined. In general, more polar solvents such as DMF favored the
formation of the E-isomer, and less polar solvents such as acetone favored formation of the Z-isomer. The
optimized conditions for the formation of E-vinyl chlorides were found to be the use of cyclopentadienyl
ruthenium (II) cyclooctadiene chloride, stannic chloride pentahydrate as a cocatalyst, and for a chloride
source, either ammonium chloride in DMF/water mixtures or tetramethylammonium chloride in DMF. A
range of several other ruthenium (II) catalysts was also shown to be effective. A wide variety of vinyl chlorides
could be formed under these conditions. Substrates with tethered alcohols or ketones either five or six
carbons from the alkyne portion gave instead diketone or cyclohexenone products. For formation of vinyl
bromides, a catalyst system involving the use of cyclopentadienylruthenium (II) tris(acetonitrile) hexa-
fluorophosphate with stannic bromide as a cocatalyst was found to be most effective. The use of ammonium
bromide in DMF/acetone mixtures was optimal for the synthesis of E-vinyl bromides, and the use of lithium
bromide in acetone was optimal for formation of the corresponding Z-isomer. Under either set of conditions,
a wide range of vinyl bromides could be formed. When alkynes with propargylic substituents are used,
enhanced selectivity for formation of the Z-isomer is observed. When aryl acetylenes are used as the
coupling partners, complete selectivity for the Z-isomer is obtained. A mechanism involving a cis or trans
halometalation is invoked to explain formation of the observed products. The vinyl halides have been shown
to be precursors to R-hydroxy ketones and cyclopentenones, and as coupling partners in Suzuki-type
reactions.
Introduction
Background
The efficient formation of functionalized alkenes, especially
vinyl halides, in a stereoselective fashion is an important goal
in organic synthesis. Olefination reactions represent the main
method for formation of such compounds,1 although stoichio-
metric halogenation of carbon-metal bonds is also frequently
utilized.2 The use of transition-metal catalysts wherein vinyl
halides are formed by additions to alkynes is an area of growing
interest.3 Addition reactions to alkynes catalyzed by transition
metals generally lead to either a cis or trans addition. A catalytic
system wherein either isomer can be formed selectively enhances
the utility of that method. In this paper, we describe the
development of a ruthenium-catalyzed three-component coupling
to form either Z- or E-vinyl halides. The development of this
reaction was part of our program to further explore atom-
economical reactions catalyzed by ruthenium complexes.4
During the course of our development of new ruthenium-
catalyzed reactions, it was discovered that terminal alkynes and
enones could be reacted under ruthenium catalysis in DMF/
water mixtures to form 1,5-diketones, as depicted in eq 1.5 The
mechanism that was originally postulated for this reaction is
shown in Scheme 1. When CpRu(COD)Cl (1) was used as a
precatalyst, the initial catalyst generation involves reaction of
the COD ligand in a [2 + 2 + 2] cycloaddition6 with the alkyne
and dissociation of chloride to provide the catalytically active
species, a cationic cyclopentadienyl ruthenium fragment. This
species catalyzes addition of water to the alkyne to form a
ruthenium enolate (3), that can also exist as its O-bound form
* To whom correspondence should be addressed. E-mail: bmtrost@
stanford.edu.
(1) Pine, S. H. Org. React. 1994, 43, 1. Ager, D. J. Org. React. 1990, 38, 1.
Maryanoff, B. E. Chem. ReV. 1989, 89, 863.
(2) Matteson, D. S. Tetrahedron 1989, 45, 1859. Takada, E.; Hara, S.; Suzuki,
A. Tetrahedron Lett. 1993, 34, 7067.
(3) See, for example: Li, J.; Jiang, H.; Feng, A.; Jia, L. J. Org. Chem. 1999,
64, 5984. Hua, R.; Shimada, S.; Tanaka, M. J. Am. Chem. Soc. 1998, 120,
12365.
(4) Trost, B. M. Science 1991, 254, 1471. Trost, B. M. Angew. Chem., Int.
Ed. Engl. 1995, 34, 259.
(5) Trost, B. M.; Portnoy, M.; Kurihara, H. J. Am. Chem. Soc. 1997, 119,
836.
(6) Trost, B. M.; Imi, K.; Indolese, A. F. J. Am. Chem. Soc. 1993, 115, 8831.
9
7376
J. AM. CHEM. SOC. 2002, 124, 7376-7389
10.1021/ja011426w CCC: $22.00 © 2002 American Chemical Society

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
Scheme 1. Proposed Ruthenium Enolate Mechanism for the
1,5-Diketone Synthesis
could occur from 10. Both lead to ruthenium enolate 12, which
is then protonated to release the product and regenerate the
active catalyst.
Differentiation between the ruthenacycle mechanism and the
ruthenium enolate mechanism might be achieved if we could
define the stereochemistry of the addition of water across the
alkyne. A potential approach was revealed in the presence of a
byproduct of the reaction- the formation of a vinyl chloride,
for example, 13 or 14. Scheme 1 requires a trans addition which
would lead to 13, whereas Scheme 2 suggests the cis addition
to be more likely and would lead to 14. We therefore initially
turned to the examination of a three-component coupling
involving a chloride source in lieu of water and its stereochem-
istry.9
Optimization of the Reaction: Vinyl Chloride Formation.
We initially examined the use of various metal halide cocata-
lysts. The reaction chosen for optimization is depicted in eq 2,
Scheme 2. Proposed Ruthenacycle Mechanism for 1,5-Diketone
Formation
with 1-octyne and methyl vinyl ketone (MVK) as the coupling
partner. Two products were formed in this reaction, E-vinyl
chloride 15 and diketone 16. A wide range of initial optimization
experiments (full details of which can be found in the Supporting
Information) was performed. A range of Lewis acid cocatalysts
(InCl3, AlCl3‚6H2O, CeCl3‚7H2O, PbCl2, NiCl2.6H2O, ZnCl2
and SnCl4‚5H2O), chloride sources (LiCl, NH4Cl, N(CH3)4Cl),
additives (NH4PF6, PPh3), and solvents (DMF, DMF/water
mixtures, MeOH, acetone) were examined. Screening deter-
mined that 15 mol % hydrated stannic chloride was the best
cocatalyst, with 3.3 equiv of ammonium chloride in DMF/water,
20/1, at 100 °C. In all cases, except when acetone was used as
a solvent, approximately a 6:1 E/Z mixture of vinyl chlorides
was formed.
Effects of concentration and temperature were then examined
(Table 1). Entry 1 shows the previously optimized conditions
(vide supra). Increasing the concentration of the alkyne to 0.5
M (entry 2) gave an increased yield. A further increase in
concentration to 1 M (entry 4) was detrimental, however.
Unfortunately, even at a higher concentration, the catalyst could
not be lowered to 5 from 10% without a much lower yield (entry
3). Although it was originally believed that high temperatures
were necessary for generation of the active catalyst, we were
delighted to discover that lower temperatures were in fact
beneficial for the reaction, giving now a good yield of the vinyl
4. The enone is then coordinated to the ruthenium, giving 5.
Insertion into the enone leads to another ruthenium enolate 6,
which is then protonated to release product.
Another potential mechanism for this reaction involves
ruthenacycle formation and hydrolysis. Some evidence for this
ruthenacycle mechanism is found in two related examples, one
involving the use of alkynes with propargylic alcohols7 and the
other an intramolecular version of the reaction depicted in eq
1, wherein either 1,5-diketones or pyrans could be formed.8 As
shown in Scheme 2, and in analogy to the ruthenium-catalyzed
Alder-ene reaction, the mechanism involves initial coordination
to the coordinatively unsaturated ruthenium to form 8. Metal-
lacycle formation then gives 9. At this point, there are several
possibilities for addition of water. Two will be outlined here.
First, coordination of water to the ruthenium can lead to 11.
Alternatively, addition of water across the double bond could
lead to 10. Reductive elimination of the hydroxy group can occur
from 11. A â-hydrogen elimination (from the hydroxyl proton)
(9) For preliminary reports, see: Trost, B. M.; Pinkerton, A. B. J. Am. Chem.
Soc. 1999, 121, 1988; Trost, B. M.; Pinkerton, A. B. Angew. Chem., Int.
Ed. 2000, 39, 360; Trost, B. M.; Pinkerton, A. B., Tetrahedron Lett. 2000,
41, 9627.
(7) Trost, B. M.; Krause, L.; Portnoy, M. J. Am. Chem. Soc. 1997, 119, 11319.
(8) Trost, B. M.; Brown, R. E.; Toste, F. D. J. Am. Chem. Soc. 2000, 122,
5877.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7377

A R T I C L E S
Trost and Pinkerton
Table 1. Concentration and Temperature Effectsa
entry
alkyne concentration (M)
temperature (°C)
16 (%)
15 (%)
1
0.25
0.5
0.5
1.0
0.5
100
100
100
100
60
2
4
6
17
5
48
56
20
45
72
2
3b
4
active compared to the COD catalyst 1. With this catalyst,
reaction of 1-octyne with MVK provides the vinyl chloride 15
in 68% with an E/Z ratio of >15/1. This compares to 72% and
>15/1 for the COD catalyst 1. Therefore, it appears that the
same active species is generated with either 1 or 18 and similar
reactivity profiles are seen.
5
a In all cases, approximately a 6/1 E/Z mixture was obtained for the vinyl
chloride. b Run with 5% 1.
Table 2. Use of Tetramethylammonium Chloride in DMF at 60 °C
entry
chloride source (equiv)
solvent (%)
16 (%)
15 (%)
Substrate Range: Vinyl Chlorides. With optimized condi-
tions in hand, a range of substrates was examined according to
eq 3, with the results summarized in Table 3. Two general
1
2
3
NH4Cl (3)
DMF
DMF
DMF
DMF
DMF
0
0
0
0
0
63a
57b
61b
72b
68b
N(CH3)4Cl (1)
N(CH3)4Cl (2)
N(CH3)4Cl (3)
N(CH3)4Cl (3)
4
5c
a E/Z ratio of 6.2/1. b E/Z ratio of >15/1. c Complex 18 employed as
catalyst.
chloride (entry 5). Indeed, decomposition of the vinyl chloride
product was observed at 100 °C under the reaction conditions.12
Control experiments showed that the product was stable at the
lower temperature.
methods were used. The first, method A, consisted of the use
of ammonium chloride in DMF/water, 20/1, and method B
consisted of the use of tetramethylammonium chloride in DMF.
In general, the use of method B was preferred due to the
generally higher yields and selectivities that were obtained.
As we see from Table 3, a broad range of functionality is
tolerated. Excellent chemoselectvity is observed with nitrile
(entries 4-7), ester (entries 8 and 9), phthalimido (entry 16),
and keto (entry 17) groups. Significantly, isolated, nonconju-
gated olefins (entries 18 and 19) are tolerated, with only bond
formation to the enone observed (although in this case some
unidentified byproducts were isolated). Also, a range of primary
(entries 10, 13, and 20), secondary (entries 11, 12, 14, and 15)
and tertiary alcohols (entries 18 and 19) are compatible with
the reaction conditions. Interestingly, even propargylic alcohols
are tolerated (entry 14). Finally, a range of enones can serve as
partners, with cyclohexylvinyl ketone reacting equally well
(entries 3, 6, and 7). However, acyl oxazolidinone 32 fails to
react under the standard conditions (entry 21). Other partners
such as ethyl acrylate and acrylonitrile also give no products in
this reaction.
Finally, with the milder conditions in hand, we reexamined
the chloride source as well as the solvent (Table 2). It was
initially presumed that some water was necessary for a polar
medium for the active catalyst. Also, we assumed that some
water was necessary for solubility of the ammonium chloride
and stannic chloride. Indeed, the use of straight DMF instead
of DMF/water, 20/1, gave a decreased yield (63% vs 72%),
but no diketone was formed (entry 1). Switching to a more
soluble chloride source, tetramethylammonium chloride, gave
some very nice results (entries 2-4). As shown, 3 equiv of
tetramethylammonium chloride gave a comparable yield of the
vinyl chloride with no diketone (entry 4). Furthermore, the
product was formed as exclusively the E-isomer as determined
by proton NMR spectroscopy (see Supporting Information). The
conditions in entry 4 were therefore used as one of the general
methods for the formation of vinyl chlorides.
Due to the interesting results obtained with the use of acetone
as a solvent in terms of olefin selectivity (vide supra), we further
examined this effect. For example, the use of acetone in place
of DMF as the solvent using our optimized conditions (see eq
2) gave the Z-isomer 17 (Z/E > 15:1, 24% yield). Unfortunately,
diketone formation dominated in this case.13 Other ruthenium
catalysts were also successful in giving the desired vinyl
chloride. However, both a methyallyl ruthenium14,15 catalyst
(CpRu(C4H7)PPh3) and a bis-phosphine ruthenium16 (CpRu-
(PPh3)2Cl) gave inferior results (see Supporting Information).
Last, the tris(acetonitrile) catalyst 1817 was found to be equally
Interestingly, when the standard conditions that were used
to form E-vinyl chlorides were tested with phenylacetylene (eq
4, path a) and 1-ethynylcyclohexanol (eq 4, path b), only the Z
(10) See, for example: Trost, B. M.; Indolese, A. F.; Mu¨ller, T. J. J.; Treptow,
B. J. Am. Chem. Soc. 1995, 117, 615. Trost, B. M.; Mu¨ller, T. J. J.;
Martinez, J. J. Am. Chem. Soc. 1995, 117, 1888.
products 33 and 34, respectively, were obtained.18 Thus, it
appears that such large steric factors override the intrinsic
geometrical preference. It should be noted that in both of these
cases, lower yields were obtained, perhaps reflective of the
competing selectivities.
(11) Either MeOH or DMF/water mixtures appear to be the best solvent for
most applications of this catalyst. See the ref 10.
(12) Approximately 20% of the material was lost when resubmitted to the
reaction conditions.
(13) Thoroughly dried acetone was not examined in this reaction, but later results
showed the necessity of some water for catalytic turnover.
(14) Trost, B. M.; Pinkerton, A. B.; Toste, F. D.; Sperrle, M. J. Am. Chem.
Soc. 2001, 123, 12504.
(15) Albers, M. O.; Robinson, D. J.; Shaver, A.; Singleton, E. Organometallics
1986, 5, 2199.
(16) Joslin, F. L.; Mague, J. T.; Roundhill, D. M. Organometallics 1991, 10,
521.
(17) Gill, T. B.; Mann, K. R. Organometallics 1982, 1, 485.
(18) The olefin geometry was determined by nOe difference experiments. See
Supporting Information.
9
7378 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
Table 3. Examples of Ruthenium-Catalyzed E-Vinyl Chloride
Formationa
with MVK to form vinyl chloride 35 in reasonable yield.
Extension to nonsymmetrically disubstituted alkynes raises the
question of regioselectivity- a question that has not been
addressed. This method offers a nice access to stereodefined
tetrasubstituted olefins.
Several substrates did not form vinyl chlorides under the
standard conditions (see Table 4). Using method A, primary
Table 4. Formation of 1,5-Diketones or Cyclohexenones
alcohol substrates bearing a hydroxy function four or five
carbons removed from the alkyne gave cyclohexenones 36 and
37 as the sole products (entries 1 and 2). These products were
characterized by the proton NMR spectra, which showed singlets
at δ 1.98, indicating an allylic methyl group, and matched the
previously obtained spectra for these compounds.5 They were
also clearly differentiated from the vinyl chlorides by the lack
of vinylic triplets in the proton NMR spectra. It should be noted
that this cyclization reaction is specific to these alcohol
substrates. Clearly, longer or shorter chain alcohols are well
tolerated in the reaction, as evidenced by the results in Table 3.
Ketone substrates with the keto functionality 6 carbons from
the alkyne (entry 4) gave 1,5-diketone such as 39 products, as
did secondary alcohols (entry 3). These products have been
previously described, and the proton and carbon NMR data
correlated. The mechanistic implications of these interesting and
useful reactions will be discussed subsequently (vide infra).
Optimization of the Reaction: Vinyl Bromides. With the
successful formation of vinyl chlorides, we obviously wished
to examine the corresponding formation of vinyl bromides, due
to their greater synthetic utility as partners in cross-coupling
reactions. Although the catalyst 1 employed for the vinyl
chloride reaction possesses a chloride, it was anticipated that
excess bromide would swamp out its competition. Interestingly,
that proved not to be the case; mixtures of vinyl chlorides and
vinyl bromides formed, even at high bromide concentration.
Presumably the chloride came from the catalyst and was very
a All reactions run according to eq 3 with a 1/2 ratio of alkyne to enone,
at 0.5 M in alkyne. b Method A: 20/1 DMF/water and NH4Cl; Method B:
DMF and NMe4Cl.
Disubstituted alkynes also act as coupling partners in this
reaction, as shown in eq 5. In this example, 5-decyne reacts
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7379

A R T I C L E S
Trost and Pinkerton
Table 5. Initial Screening Experiments for Vinyl Bromide
Formation
Concerned that solubility of the ammonium salts was an issue,
we investigated the use of other bromides. Indeed, lithium
bromide is fully soluble in acetone and gave an excellent yield
of the vinyl bromide with good Z selectivity (entry 17).
Curiously, increasing the amount of lithium bromide to 3 equiv
leads to a decrease in Z selectivity (entry 20). Although neither
catalyst (entry 18) nor cocatalyst loading (entry 19) could be
lowered, we nevertheless adopted these conditions as our optimal
conditions for Z-vinyl bromide formation. Entry 22 shows the
vital nature of the cocatalyst once again, with only low yields
of vinyl bromide obtained in its absence. Finally, a control
experiment with no catalyst gives no product (entry 23).
The results of Table 5 indicate that the E/Z selectivities can
range from 5:1 to >1:15. Ideally, obtaining either geometric
isomer at will is desirable. Therefore, we attempted optimization
of formation of E-vinyl bromides as well. Performing the
reaction of eq 6 using 15 mol % stannic bromide and 3 equiv
of ammonium bromide in 1/1 acetone/DMF at 0.5 M alkyne
with 1.5 equiv of MVK gave a 71% yield of a 2.1:1 E/Z ratio.
Increasing the MVK to 2 equiv increased the E/Z ratio to 3.4:
1. Decreasing the amount of MVK or concentration of alkyne
had no effect. The results suggest that the solvent might be
serving as ligand in competition with the acetonitrile ligands
originally present. Thus, the effects of nitrile concentration and
type of nitrile was examined. Adding 30 mol % excess
acetonitrile had no effect on the reaction in DMF nor in 1/1
acetone/DMF. Use of acetonitrile as solvent shuts down the
reaction. Employing 15 mol % of adiponitrile or glutaronitrile
in 1/1 acetone/DMF increases the E/Z ratio to 3.4 ( 0.2, a
modest increase over the control. At this stage, selectivity for
forming the E-isomer in good yields stands at 2-4:1.
entry cocatalyst (%) bromide source (equiv)
solvent
40 (%)
E/Z
1
SnBr2 (15) NH4Br (3.0)
SnBr2 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
DMF
DMF
DMF
13 3.3/1
14 3.2/1
54 2.1/1
61 1.2/1
11 5.1/1
2b
3
4
5
6
7
8
9
10
11
12
13
14
SnBr4 (15) N(CH3)4Br (5.0) DMF
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
SnBr4 (15) NH4Br (3.0)
DMF/CH3CN
DMSO
acetone
acetone/DMF, 1/1 48 5.1/1
acetone/DMF, 4/1 56 1/1.4
acetone/DMF, 2/1 62 2.2/1
acetone/DMF, 1/2 70 2.4/1
acetone/DMF, 1/4 48 2.6/1
-
-
40 1/10
MeOH
40 1/3.0
35 >1/15
33 1/8.0
SnBr4 (15) N(CH3)4Br (3.0) acetone
15a SnBr4 (15) N(CH3)4Br (1.5) acetone
16
SnBr4 (15) N(CH3)4Br (3.0) acetone/DMF, 1/1 54 1/4.0
17a SnBr4 (15) LiBr (1.5)
18a,c SnBr4 (15) LiBr (1.5)
acetone
acetone
acetone
acetone
88 1/6.6
56 1.3/1
54 1/6.7
87 1/4.0
19a SnBr4 (5)
LiBr (1.5)
20a SnBr4 (15) LiBr (3.0)
21 SnBr4 (15) NaBr (1.5)
22a None
LiBr (1.5)
23d SnBr4 (15) NH4Br (3.0)
acetone/DMF, 1/1 59 1/6.9
acetone
22 1/4.2
acetone/DMF, 1/1
0
-
a Run with 1.5 equiv of MVK. b Run at room temperature for 4 h. c Run
with 5% catalyst. d Run with no catalyst.
effective in competing against the bromide. We therefore turned
to alternative catalysts lacking chloride. Indeed, complex 18,
which is postulated to give the same active intermediate, proved
effective and avoided this problem.
The reaction of eq 6 was examined for optimization and the
Table 5 indicates that the use of tetramethylammonium
bromide significantly increased the Z selectivity but only in
moderate yield. Therefore, an attempt was made to increase the
yield while maintaining the geometrical selectivity using tetra-
alkylammonium salts. Unfortunately, variation of the amounts
of tetramethylammonium bromide, solvent, and the nature of
the alkyl group of ammonium salt did not improve the overall
results. While higher yields were obtained with tetraethyl-
ammonium bromide in DMF (63%), the E/Z selectivity dropped
to 1:2. Curiously, the best yields were obtained with a spiro-
tetraalkylammonium salt (41). Although the E/Z ratio is 1/4.l
to 1/3.l, the yields in both acetone (73%) and DMF (73%) are
good. Nevertheless, these best results using ammonium salts
are inferior to those obtained with using lithium bromide as
the bromide source (Table 5, entry 17).
initial results shown in Table 5. The success of tin salts as
cocatalysts focused our efforts on tin bromide. Initial efforts
used stannous bromide (entries 1 and 2), but only low yields
were obtained. We then turned to stannic bromide as the
cocatalyst for our subsequent experiments. Using ammonium
bromide in DMF, a moderate yield of vinyl bromide was
obtained, also predominately as the E-isomer (entry 3). However,
the E/Z ratio was much lower than for the corresponding vinyl
chloride formation. Use of tetramethylammonium bromide in
DMF (entry 4) did not dramatically improve the yield and
actually lowered the selectivity. Other solvents, such as DMF/
acetonitrile (entry 5) and DMSO (entry 6) gave poor results.
Interestingly, the use of acetone as a solvent (entry 7) with
ammonium bromide gave a moderate yield, but with good Z
selectivity, also an effect that was observed in the vinyl chloride
formation.
Substrate Range: Vinyl Bromides. With several different
sets of conditions in hand, the effect of substrate variation was
examined (see eq 7 and Table 6). The initial set of experiments
We next examined mixtures of DMF and acetone as solvent
in the attempt to have high selectivity as well as an increased
yield. As shown in entries 8-12, when more DMF is present
relative to acetone, more of the E-isomer is formed. Conversely,
when more acetone is present, more of the Z-isomer is formed.
However, in general only moderate yields are obtained in these
cases. The use of methanol as a solvent (entry 13) was only
moderately effective. Interestingly, with tetramethylammonium
bromide in acetone (entries 14 and 15) or acetone/DMF (entry
16), good Z selectivity could be obtained, but only in moderate
yields.
focused on the use of ammonium bromide as the bromide
source, and the formation of either Z- or E-vinyl bromides
depending on the solvent used. In general, a 1/1 mixture of
acetone/DMF (solvent A) was used for E-vinyl bromide
formation and solely acetone (solvent B) or 2-butanone (solvent
9
7380 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
Table 6. Examples of Ruthenium-Catalyzed E- or Z-Vinyl Bromide
(entries 8-10, 18-19) and secondary (entries 6-7, 11-12)
alcohols, esters (entries 13-14, 21-22) and amides (entries 15-
16) are all compatible. Other enones such as cyclohexylvinyl
ketone participate (entries 19-22). Interestingly, a substrate with
a terminal olefin (51) gave no vinyl bromide product. In this
case, it is possible that other Alder-ene type reactions are
occurring as well, and only decomposition products are ob-
served. However, overall the reactions were quite clean, with
the remainder of the alkyne material perhaps oligomerizing.
Although the results in Table 6 are an extremely nice example
of how the same catalyst system, including use of the same
source of bromide, can give different geometrical selectivities
in different solvents, the yields for selective formation of Z-vinyl
bromides were unacceptably low. We therefore ran a second
set of experiments using lithium bromide in acetone as shown
in eq 7. The results are summarized in Table 7. In general, this
was the best method for forming Z-vinyl bromides in good yields
and selectivities.
Formation Using Ammonium Bromidea
The same wide range of substrates was tolerated. In general,
excellent yields were obtained, and for the most part selectivity
for the Z-isomer was moderate to excellent. For example, using
a propargylic alcohol substrate (entry 5), an 1/11.0 E/Z mixture
of vinyl bromide 45 was obtained. The examples also illustrate
the excellent chemoselectivity with hydroxyl, cyano, keto,
carboxy, and phthalimido all compatible. The example of entry
12 is particularly noteworthy since it shows a cis-dialkyl
substituted alkene is unaffected. Other enones such as cyclo-
hexylvinyl ketone (entries 9 and 10) and phenylvinyl ketone
(entries 13 and 14) were compatible.
Disubstituted alkynes also act as coupling partners in vinyl
bromide formation, as shown in eq 8. In this example, 5-decyne
reacts with MVK to form vinyl bromide 56 in moderate yield,
with the expected geometry of the double bond. The broader
general applicability of this reaction will depend on the
regioselectivity with nonsymmetrical disubstituted alkynessan
aspect for future study.
The high Z selectivity obtained when propargylic substituents
were present (as in entry 12, Table 6; entry 5, Table 7) led us
to postulate that increasing the steric bulk at the propargylic
position should give enhanced selectivity for the Z-isomer. We
therefore examined a range of substrates with substitution at
the propargylic position (see eq 2 and Table 8). When a
substituent even as small as methyl is present at the propargylic
position (entry 1), a very good Z/E ratio of 7.9/1 is observed
for the vinyl bromide. Increasing the substitution to a quaternary
center (entry 2) then produces only the Z-isomer. This example
is particularly noteworthy due to the high propensity of such
compounds to form allenylidene species with ruthenium com-
plexes of the type used here.19 An all-carbon alkyl group (entry
a All reactions run according to eq 7 with a 1/1.5 ratio of alkyne to enone,
at 0.5 M in alkyne with 10% catalyst and 15% stannic bromide in acetone
at 60oC for 2 h unless indicated otherwise. b Solvents: A ) DMF/Acetone
1/1, B ) Acetone, C ) 2-Butanone
C) for Z-vinyl bromide formation. The yields in the acetone/
DMF mixture were higher, indicative of the increased solubility
of the ammonium bromide in the more polar solvent. In general,
this was the most appropriate method for forming E-vinyl
bromides selectively and in good yields.
(19) Trost, B. M.; Flygare, J. A. J. Am. Chem. Soc. 1992, 114, 5476; Selegue,
J. P.; Young, B. A.; Logan, S. L. Organometallics 1991, 10, 1972; Cadierno,
V.; Gamasa, M. P.; Gimeno, J.; Gonzales-Cueva, M.; Lastra, E.; Borge,
J.; Garcia-Granda, S.; Pe´rez-Carren˜o Organometallics 1996, 15, 2137. For
a recent leading reference, see: Nisbibayashi, Y.; Wakiji, I.; Hidai, M. J.
Am. Chem. Soc. 2000, 122, 11019.
A wide range of functionality is tolerated; the reaction is
highly chemoselective. Nitriles (entries 3-5, 19-20), primary
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7381

A R T I C L E S
Trost and Pinkerton
Table 7. Examples of Ruthenium-Catalyzed Z-Vinyl Bromide
Table 8. Enhanced Selectivity in the Ruthenium-Catalyzed Z-Vinyl
Formation Using Lithium Bromidea
Bromide Formation Using Lithium Bromidea
a All reactions run according to eq 7 with a 1/1.5 ratio of alkyne to enone,
at 0.5 M in alkyne with 10% catalyst and 15% stannic bromide in acetone
at 60 °C for 2 h.
with aryl acetylenes compared to the result of entry 1 may seem
counterintuitive at first, as a phenyl group is generally a sterically
less demanding group than an isopropyl group, the effective
steric bulk of the aromatic ring may be larger.20 This effect will
be discussed subsequently. Somewhat surprisingly, neither
alcohol 68 nor ester 69 gave any vinyl bromide product under
the reaction conditions. MVK was replaced with phenylvinyl
a All reactions run according to eq 7 with a 1/1.5 ratio of alkyne to enone,
at 0.5 M in alkyne with 10% catalyst and 15% stannic bromide in acetone
at 60 °C for 2 h.
3) and even a trimethylsilyl group (entry 4) also gives only one
geometric isomer. Alternatively, an aryl substituent shows the
same behavior (entries 5-9). Although the complete selectivity
(20) Conjugation of the aromatic system with the double bond leads to a more
planar geometry and thus increased steric effects.
9
7382 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
Table 9. Conversion of Vinyl Chlorides to Form R-Hydroxyketones
ketone (entries 6, 10, and 11) to produce the corresponding
three-component coupling products with equivalent results.
Extension of the reaction to alkenes other than vinyl ketones
have, sofar, proved elusive. Low yields of the adduct could be
detected in the case of N-acryloyloxazolidin-2-one (e.g., see
entry 15 of Table 7). On the other hand, three-component
coupling products could not be observed with ethyl acrylate,
acrylonitrile, N,N-dimethylacrylamide, ethyl 2-pentenoate, cro-
tonaldehyde, acrolein, nor styrene.
Some Applications
Formation of r-Hydroxyketones Wia Epoxidation or Di-
hydroxylation. The direct availability of vinyl chlorides by this
three-component coupling make them readily available building
blocks. To illustrate, we examined the oxidation of the vinyl
chlorides to take advantage of the different oxidation levels of
the two carbons of the olefin. Initially we examined the
epoxidation of the vinyl halides in an attempt to form halo-
epoxides.21 It was found, however, that such compounds reacted
further to form the R-hydroxyketone either during the reaction
or during workup. The products were characterized by the
presence of a hydroxy stretching frequency of approximately
3500 cm-1 in the infrared spectrum, as well as the presence of
two peaks corresponding to carbonyl groups (at approximately
210 ppm) in the carbon NMR spectra. The reactions were
performed using an excess of m-chloroperbenzoic acid (mCPBA)
buffered with sodium or potassium bicarbonate (eq 9). In
Table 10. Asymmetric Dihydroxylation of Vinyl Halides to Form
R-Hydroxyketones
general, good to moderate yields were obtained (Table 9). An
excellent yield was obtained with vinyl chloride 15 (entry 1),
most likely due to the lower polarity of this compound. The
more functionalized compounds, especially the alcohol 23 (entry
2), gave very polar products. This led to losses of material during
purification (removal of the excess mCPBA as well as m-
chlorobenzoic acid required water washes as well as column
chromatography).
a Ee’s determined by chiral HPLC analysis.
The asymmetric version23 led to its application with the vinyl
chlorides using commercially available AD-mix-â with meth-
anesulfonamide at 0 °C (see eq 10 and Table 10). The ee’s were
An alternative reaction that would lead to the product is a
direct dihydroxylation which has the capacity of being per-
formed asymmetrically rather easily. We therefore turned to
methods for the dihydroxylation of the vinyl chlorides.22 This
proved to be very successful, and in general the reactions were
very clean using quite standard catalytic osmylation conditions
with 5% osmium tetraoxide and 4 equivalents of NMO (eq 9).
As shown in Table 9 good yields were obtained in all cases,
superior to the epoxidation conditions due to ease of workup.
determined by chiral HPLC analysis. These reactions were
somewhat more sluggish than the typical achiral dihydroxyla-
tions using osmium; however, the reactions usually went to
completion although longer reaction times were required. For
(21) March, J. AdVanced Organic Chemistry; Wiley: New York, 1992; pp
1087-1088.
(22) Sharpless, K. B.; Teranishi, A. Y.; Ba¨ckvall, J. E. J. Am. Chem. Soc. 1977,
99, 3120.
(23) Jacobsen, E. N.; Marko´, I.; Mungall, W. S.; Schro¨der, G.; Sharpless, K. B.
J. Am. Chem. Soc. 1988, 110, 1968.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7383

A R T I C L E S
Trost and Pinkerton
easier HPLC analysis, the R-hydroxyketones were converted
to their benzoates, as shown in eq 10. As we can see from Table
10, moderate to good ee’s may be obtained. It should be noted
that no optimization was done for this reaction (i.e., examining
different chiral ligands), and therefore it is possible that much
better ee’s could be obtained.
bromide, and an arylboronic acid in sequential ruthenium- and
palladium-catalyzed reactions. Since either haloolefin geometric
Attempts to increase the ee by lowering the reaction tem-
perature were unsuccessful, as the reaction became too slow.
At -20 °C, the reaction of entry 1 of Table 10 occurred only
to the extent of 10% after 24 h. To ascertain the role, if any,
played by employing E-Z-alkene mixtures, we examined the
reaction of vinyl chlorides 15 and 22, with E/Z ratios greater
than 15/1. As shown in Table 10 (entries 2 and 7), the results
were somewhat surprising; no increase in ee was observed in
either case. Thus, the E-vinyl chlorides behave like Z-1,2-
disubstituted alkenes wherein ee’s are more modest.24 Interest-
ingly, the ee’s are somewhat higher than would be expected
for the Z-1,2-disubstituted alkene, indicating that Cl is playing
some role. The fact that the ee is independent of alkene geometry
precludes the need to use geometrically pure E-alkenes for
optimum ee.
Application of the dihydroxylation reaction to the vinyl
bromides should generate the same products as from the vinyl
chlorides. To examine the effect, if any, of the halide on the
degree of asymmetric induction, we performed the asymmetric
dihydroxylation of 40 as shown in Table 10, entry 3. The initial
R-hydroxyketone was directy benzoylated and analyzed as 74
to be 86% ee. This result is a small but significant improvement
over the corresponding E-vinyl chloride 15 (Table 10, entries
1 and 2). Importantly, the same enantiomer was produced in
both cases even though opposite geometric isomers of alkenes
were employedsthat is, the major peak in the chiral HPLC was
the same in both cases. A similar observation has been noted
in the AD of the geometric isomers of enol silyl ethers.25 The
enhancement in ee might result from the larger size of Br versus
Cl.
isomer is now available, either trisubstituted alkene is available.
For the preparation of the E-alkenes, the E-vinyl bromides may
be used; however, the E-vinyl chlorides are the more practical
precursors.
Cyclopentenone Formation. Access to Z-vinyl bromides
opens the opportunity to use the juxtaposition of the C-Br bond
to the carbonyl group for cyclization, that is, an intramolecular
Barbier reaction. While it is known that metal halogen exchange
with organolithiums may be faster than carbonyl addition,27
attempts to effect cyclization as shown in eq 14 simply by
treating at low temperature with n-butyllithium was messy.
Nozaki and Kishi described the Barbier reaction of vinylbro-
mides with aldehydes using a chromium-nickel system.28 While
ketones normally did not function well in this reaction, the fact
that the reaction of eq 14 was intramolecular encouraged us to
examine it.29 Indeed it works quite well as illustrated in eq 15.
Suzuki-Type Cross-Coupling Reactions. The vinyl chlorides
were also examined in Suzuki-type cross-coupling reactions.
Using the recently developed coordinatively unsaturated pal-
ladium catalyst,26 Suzuki coupling of chloride 15 with an
arylboronic acid proceeds with complete integrity of alkene
geometry to give trisubstituted alkene 78 (eq 11), an observation
Thus, treating a 5.6/1 Z/E ratio of vinyl bromide 40 under
standard Nozaki-Kishi conditions provided the cyclopentenol
81 in 70% yield. Presumably, the minor E-vinyl bromide cannot
cyclize, but instead leads to reductive dehalogenation. Thus,
the yield should be adjusted to reflect the Z/E ratio of the starting
material to 83%. Such tertiary allylic cyclopentenols can be
envisioned to participate in numerous reactions, for example,
Claisen and related rearrangements, allyl coupling, and so forth.
A simple application is their oxidation concomitant with allylic
rearrangement to cyclopentenone 82.30
Discussion
that validates this approach as an excellent strategy to trisub-
stituted alkenes of defined geometry. Vinyl bromides also react
readily as cross-coupling partners. For example, the Z-vinyl
bromide 40 underwent Suzuki coupling with both an electron-
rich and electron-poor aryl boronic acid (eq 12) to give Z-alkenes
79a and 79b. The excellent chemoselectivity is highlighted by
the example of eq 13 wherein the trisubstituted Z-alkene 80
derives in two steps from 5-cyano-1-pentyne, MVK, lithium
The formation of either vinyl halide isomer conflicts with a
metallacycle mechanism of the type proposed in Scheme 2. The
(27) See: Smith, M. B. Organic Synthesis; McGraw-Hill: New York, 1994;
pp 719-727.
(28) (a) For a review, see: Cintas, P. Synthesis 1992, 248. (b) For a recent
intramolecular example and representative procedure, see: Chen, X.-T.;
Bhattacharya, S. K.; Zhou, B.; Gutteridge, C. E.; Pettus, T. R. R.;
Danishefsky, S. J. J. Am. Chem. Soc. 1999, 121, 6563.
(29) Using a recently disclosed bipyridyl ligand system, ketones have been shown
to be reactive. See: Chen, C. Synlett 1998, 1311. See also: Chen, C.;
Tagami, K.; Kishi, Y. J. Org. Chem. 1995, 60, 5386.
(30) Trost, B. M.; Pinkerton, A. B. Org. Lett. 2000, 2, 1601. Dauben, W. G.;
Michno, D. M. J. Org. Chem. 1977, 42, 682. Majetich, G.; Song, J.-S.;
Leigh, A. J.; Condon, S. M. J. Org. Chem. 1993, 58, 1030. Majetich, G.;
Condon, S.; Hull, K.; Ahmad, S. Tetrahedron Lett. 1989, 30, 1033.
(24) For a review, see: Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. ReV. 1994, 94, 2483.
(25) Morikawa, K.; Park, J.; Andersson, P. G.; Hashiyama, T.; Sharpless, K. B.
J. Am. Chem. Soc. 1993, 115, 8463.
(26) Littke, A. F.; Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2000, 122, 4020.
9
7384 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
Scheme 3. Mechanistic Rational for Vinyl Halide Formation
ability to use the ruthenium complexes (CpRu(C4H7)PPh3) and
(CpRu(PPh3)2Cl) suggests that perhaps only one open coordina-
tion site may be necessarysagain inconsistent with the metal-
lacycle mechanism. The results clearly indicate that there is a
delicate balance between a trans- and a cis-haloruthenation to
initiate this three-component coupling. While two totally dif-
ferent mechanistic pathways are possible, the most attractive
proposal is outlined in Scheme 3. The catalytically active
species, a cationic, coordinatively unsaturated ruthenium (which
is derived from either 1 or 18), coordinates with the alkyne and
a halide to form two possible species, 83 and 86. Species 83, a
more ionic species, leads to external halide attack and a trans
halometalation to give 84. Upon trapping with the enone, 84
gives E-vinyl halide 85. Conversely, the more covalent species
86 leads to an internal, cis halometalation,31 giving the vinyl
ruthenium 87. This then leads to Z-vinyl halide 88 upon trapping
with an enone. This mechanism supports the observation that
reaction with bromide leads to more of the cis halometalation,
which is in line with the weaker nucleophilicity of bromide. In
contrast, chloride, which is more nucleophilic,32 gives more of
the external attack (i.e., to form 84) for a net trans halometa-
lation.33,34 Also, the overall effect of solvent strongly supports
this mechanism. In both the chloride and bromide cases, the
use of a less polar solvent (such as acetone) favors the more
covalent species 86, and thus more Z-vinyl halide is produced.
Conversely, a more polar solvent (i.e., DMF) supports the more
ionic species 83, and thus more E-vinyl halide is formed.
However, one interesting aspect is that the Ru-Cl bond is
stronger than the Ru-Br bond,35 and thus one might expect
that 86 should be more favored for vinyl chloride formation
(which in general it is not). Therefore, it appears that other
factors such as the difference in nucleophilicity outweigh this
consideration.
Several other observations support this mechanism. First,
during the optimization of the vinyl bromide reaction, it was
noted that when the amount of lithium bromide was lowered to
1.5 equiv from 3, the Z/E ratio increased from approximately
4/1 to almost 7/1. This would be explained by the fact that
having a higher concentration of bromide should lead to more
external attack (i.e., from 83) and thus more trans-halometalation
and E-vinyl bromide. Generally speaking, the use of more
soluble halide sources, such as tetramethylammonium chloride
(vs ammonium chloride) not only increases the yield but also
increases the selectivity for the E-isomer. This is also consistent
as above with having a higher concentration of halide ion in
solution and thus favoring trans attack to form 84.
Another effect seen that supports this mechanism is the use
of alkynes with bulky propargylic substituents. Steric interac-
tions between the R group and ruthenium in 84 disfavor this
pathway. Thus, the reaction via 87 becomes more favorable.
This is clearly the case when bulky alkynes are used, as
exclusively the Z-isomer resulting from reaction via 87 is
observed. The inability of other coupling partners to be effective
in the ruthenium-catalyzed three-component coupling may also
disfavor a ruthenacycle mechanism. If a ruthenacycle mechanism
were operative, there is no a priori reason that other olefin
partners should be unreactive.36 Conversely, if the mechanism
is that outlined in Scheme 3, then it is more understandable
that insertion of a vinylmetal species should occur much more
readily onto an enone instead of, say, styrene. The failure of
acrylates implies more reactive Michael acceptors are required.
Another problem with acrylates is their stability under the
reaction conditions. The failure of more substituted enones to
be reactive can be explained in both mechanisms by steric
effects.
The formation of the products of Table 4 provide additional
support for the mechanism of Scheme 3 as delineated in Scheme
4. The juxtaposition of a free OH three or four carbons removed
from the double bond sets the stage for an intramolecular
nucleophilic addition of the OH37 to compete with the inter-
molecular addition of chloride, and the former predominates.
Complexation with MVK and migratory insertion creates the
enol ether 89. Under these acidic conditions in the presence of
(31) For several examples of cis halometalations see: Dietl, H.; Reinheimer,
H.; Moffatt, J.; Maitlis, P. M. J. Am. Chem. Soc. 1970, 92, 2276; Kaneda,
K.; Uchiyama, T.; Fujiwara, Y.; Imanaka, T.; Teranishi, S. J. Org. Chem.
1979, 44, 55; Hua, R.; Shimada, S.; Tanaka, M. J. Am. Chem. Soc. 1998,
120, 12365. See also: Hara, S.; Dojo, H.; Takinami, S.; Suzuki, A.
Tetrahedron Lett. 1983, 24, 731. Hara, S.; Satoh, Y.; Ishiguro, H.; Suzuki,
A. Tetrahedron Lett. 1983, 24, 735.
(32) Carey, F. A.; Sundberg, R. J. AdVanced Organic Chemistry: Part A, 3rd
ed.; Plenum: New York, 1990; p 289.
(33) For a trans chloroalkylation see: Wang, Z.; Lu, X. Chem. Commun. 1996,
535.
(36) Terminal olefins are quite reactive in the Ru-catalyzed Alder-ene reaction.
See: Trost, B. M.; Indolese, A. F.; Mu¨ller, T. J. J.; Treptow, B. J. Am.
Chem. Soc. 1995, 117, 615. Trost, B. M.; Mu¨ller, T. J. J.; Martinez, J. J.
Am. Chem. Soc. 1995, 117, 1888.
(37) For some examples of -OH addition to alkynes catalyzed by Ru, see: Trost,
B. M.; Rhee, Y. H. J. Am. Chem. Soc. 1999, 121, 11680. By Pd: Nan, Y.;
Miao, H.; Yang, Z. Org. Lett. 2000, 2, 297. See also: Alper, H.;
Despeyroux, B.; Woell, J. B.Tetrahedron Lett. 1983, 24, 5691. Luo, F.-T.;
Schreuder, I.; Wang, R.-T. J. Org. Chem. 1992, 57, 2213. Marshall, J. A.;
Yanik, M. M. Tetrahedron Lett. 2000, 41, 4717.
(34) For trans addition of carboxylate nucleophiles, see Rothman, E. S.; Hecht,
S. S.; Pfeffer, P. E.; Silbert, L. S. J. Org. Chem. 1972, 37, 3551. Roten,
N.; Shvo, Y. Organometallics 1983, 2, 1689. Mitsudo, T.; Hori, Y.;
Yamakawa, Y.; Watanabe, Y. J. Org. Chem. 1987, 52, 2230. Bruneau, C.;
Neveux, M.; Kabouche, Z.; Ruppin, C.; Dixneuf, P. H. Synlett 1991, 755.
Neveux, M.; Seiller, B.; Hagedorn, F.; Bruneau, C.; Dixneuf, P. H. J.
Organomet. Chem. 1993, 451, 133.
(35) Luo, L.; Li, C.; Cucullu, M. E.; Nolan, S. P. Organometallics 1995, 14,
1333.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7385

A R T I C L E S
Trost and Pinkerton
Scheme 4. Mechanistic Rationale for Cyclohexenone Formation
water, two paths to product appear reasonable. Conceptually
simplest is hydrolysis to diketone 90 which then undergoes
standard aldol condensation to the observed products. Alterna-
tively, simple acid-catalyzed isomerization of 89 to the ther-
modynamically more stable enol ether 91 then would permit
direct carbonyl addition to the final product.
The reaction of entry 3 of Table 4 in which only diketone 38
is observed is interesting in this regard. The strain of the enol
ether 92 (eq 16) may disfavor its formation and thereby allow
The first experiment involved running the reaction under the
standard lithium bromide/acetone conditions, then quenching
the reaction with deuterated acetic acid. No deuterium was
incorporated (determined by integration of the proton NMR
signals, which were assigned above). This indicates that the
protonation occurs during the reaction itself and that an enolate
is not being stoichiometrically formed. We then investigated
whether the proton came from the acetone solvent, that is, a
metal (ruthenium?) enolate deprotonating the acetone to release
the vinyl halide product. This appears not to be the case either,
as when deuteroacetone was used as solvent, no deuterium
incorporation was observed. Finally, we investigated the addition
of D2O in place of water, because if water in the reaction mixture
was responsible for the protonation of the proposed ruthenium
enolate, addition of D2O should lead to some deuterium
incorporation. Expectedly, some deuterium incorporation (∼30%)
at the methylene position R in the product was observed when
the reaction was run in the presence of 1 equiv of D2O.38 The
modest level of deuterium incorporation derives from the
presence of normal water arising from the water of hydration
of stannic bromide.
the competitive hydrolysis of 92 to diketone 38 to dominate.
While aldol condensation of 38 might be anticipated to be slower
than for diketones 90 (n ) 1 and 2), it is somewhat surprising
that it does not occur at all. Thus, this observation may suggest
that the endocyclic enol ethers 91 are the actual precursors of
the cyclohexenones rather than the diketones.
Judicious placement of a carbonyl group in the side chain
appears to be able to compete effectively with external chloride
for the ruthenium-complexed alkyne (eq 17). Hydration of the
resultant adduct 94 then forms a ruthenium enolate 95 which
then combines with MVK to form triketone 39. This route is
very reminiscent of the three-component coupling of water,
alkynes, and vinyl ketones to form diketones.5 On the other
hand, the intermediate 94 may hydrate to 96 and then add MVK
faster than collapse of this lactol to generate product. The ability
of an oxygen of a carbonyl group to totally divert the course of
reaction indicates that the ability of chloride to function as a
nucleophile is marginal. These neighboring-group effects have
real synthetic value. Thus, a very simple cyclohexenone syn-
thesis emerges as summarized in eq 18.
Conclusions
In conclusion, we have developed a system for the formation
of stereodefined vinyl halides via a three-component coupling
process catalyzed by cationic cyclopentadienyl ruthenium spe-
cies. The versatility of vinyl halides as cross-coupling partners
make this method a useful addition to synthetic methodology.39
Furthermore, such compounds have been shown to be R-hy-
droxyketone equivalents as well as precursors to 3-hydroxycy-
clopentenes, cyclopentenones, and substituted alkenes of defined
geometry.
The catalyst system described herein represents the first
example where either isomer of a vinyl halide can be accessed
in a single catalyst system, depending on the counterion and
solvent. Although there are a number of palladium-40 and
rhodium-catalyzed41 cis additions of halides to alkynes, the only
examples of trans addition involve the addition to alkynes
bearing electron-withdrawing groups.42 The other methods for
the formation of vinyl halides involve use of stoichiometric
Another question that must be answered for both the vinyl
bromide and the vinyl chloride reaction is where the proton
comes from to protonate the proposed ruthenium enolate and
release the product. Adventitious water in the solvent or the
salts used is the likely source for both reactions. It is clearly
the case in the vinyl chloride reaction, where hydrated tin
chloride is used. In the case of the vinyl bromides, a series of
deuteration experiments were done to determine where the
proton came from.
(38) For capture by an aldehyde as an electrophile leading to four-component
coupling, see: Trost, B. M.; Pinkerton, A. B. J. Am. Chem. Soc. 2000,
122, 8081.
(39) For some recent examples, see: Negishi, E.; Alimardanov, A.; Xu, C. Org.
Lett. 2000, 2, 65. Williams, D. R.; Meyers, B. J.; Mi, L. Org. Lett. 2000,
2, 945. Maleczka, R. E., Jr.; Gallagher, W. P.; Terstiege, I. J. Am. Chem.
Soc. 2000, 122, 384.
(40) Kaneda, K.; Uchiyama, T.; Fujiwara, Y.; Imanaka, T.; Teranishi, S. J. Org.
Chem. 1979, 44, 55. Li, J.; Jiang, H.; Feng, A.; Jia, L. J. Org. Chem. 1999,
64, 5984.
(41) Hua, R.; Shimada, S.; Tanaka, M. J. Am. Chem. Soc. 1998, 120, 12365.
(42) Wang, Z.; Lu, X. Chem. Commun. 1996, 535.
9
7386 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
E-6-Chloro-8-hydroxy-oct-5-en-2-one (25): light yellow oil. Rf )
0.30 (1/1 petroleum ether/ethyl acetate). IR (neat): 3415, 2959, 2926,
1713, 1657, 1410, 1368, 1262, 1233, 1166, 1107, 1050 cm-1. 1H NMR
(300 MHz, CDCl3): δ 5.65 (t, J ) 7.7 Hz, 1H), 3.81 (t, J ) 5.9 Hz,
2H), 2.62 (t, J ) 6.0 Hz, 2H), 2.54 (t, J ) 6.9 Hz, 2H), 2.33 (q, J )
7.1 Hz, 2H), 2.13 (s, 3H). 13C NMR (75 MHz, CDCl3): δ 209.7, 133.1,
130.6, 61.1, 44.1, 38.5, 31.5, 24.1. Anal. Calcd for C8H13ClO2: C, 54.40;
H, 7.42. Found: C, 54.33; H, 7.34.
Figure 1. Formation of E- or Z-Vinyl Halides.
A larger scale example is given in the following. 4-Pentyn-2-ol (252
mg, 3.0 mmol) and methylvinyl ketone (423 mg, 0.50 mL, 6.0 mmol)
were added to a solution of CpRu(COD)Cl (90 mg, 0.3 mmol), tin
(IV) chloride pentahydrate (158 mg, 0.45 mmol), and ammonium
chloride (529 mg, 9.9 mmol) in DMF/water 20/1 (6 mL) in a pressure
tube. The tube was capped and then heated to 60 °C for 4 h. It was
then cooled to room temperature and poured into saturated sodium
bicarbonate (100 mL). The aqueous layer was extracted with ether
(2 × 100 mL). The organic layer was dried over magnesium sulfate
and the solvent removed by rotary evaporation. The crude mixture was
analyzed by proton NMR and then subjected to silica gel chromatog-
raphy (1/1 petroleum ether/ethyl acetate) to give 383 mg of vinyl
chloride 24 (66%) as a 6.9/1 E/Z-mixture as determined by integration
of the vinylic triplets.
compounds, for example by olefination reactions or stoichio-
metric halogenation of carbon-metal bonds.43 The selective
formation of Z-vinyl bromides (and chlorides) by bromoboration
is a useful method and somewhat more selective than the
ruthenium-catalyzed reactions.44 However, it involves the use
of multiple equivalents of a very harsh reagent, boron tribromide,
which is clearly not compatible with such groups as free alcohols
and methyl esters. Both of these groups are tolerated under the
much milder ruthenium-catalyzed conditions developed here.
Retrosynthetically, E- or Z-vinyl halides can now be envisioned
to come from alkynes and enones as outlined in Figure 1.
Experimental Section
E-6-Chloro-8-hydroxy-non-5-en-2-one (24): light yellow oil. Rf )
0.29 (1/1 petroleum ether/ethyl acetate). IR (neat): 3425, 2969, 2929,
1714, 1655, 1409, 1370, 1264, 1166, 1111, 1075, 940 cm-1. 1H NMR
(300 MHz, CDCl3): δ 5.65 (t, J ) 7.6 Hz, 1H), 4.13 (m, 1H), 2.65
(dd, J1 ) 14 Hz, J2 ) 8.5 Hz, 1H), 2.54 (t, J ) 6.8 Hz, 2H), 2.45-
2.24 (m, 4H), 2.13 (s, 3H), 1.24 (d, J ) 6.2 Hz, 3H). 13C NMR (75
MHz, CDCl3): δ 209.5, 133.2, 130.6, 66.9, 44.9, 44.0, 31.4, 24.4, 24.2.
Anal. Calcd C9H15ClO2: C, 56.69; H, 7.93. Found: C, 56.49; H, 7.79.
General Procedure for Ruthenium-Catalyzed E- or Z-Vinyl
Bromide Formation Using Ammonium Bromide (Table 6). The
alkyne (0.25 mmol) and enone (0.5 or 0.375 mmol) were dissolved in
the appropriate solvent (0.5 mL) (see below) and then added to CpRu-
(CH3CN)3PF6 (10.9 mg, 0.025 mmol), stannic bromide (16.4 mg, 0.0375
mmol), and ammonium bromide (73.4 mg, 0.75 mmol) in a pressure
tube. The tube was capped and then heated to 60 °C for 2 h. It was
then cooled to room temperature and applied directly to a silica gel
column. The eluting solvent for each case is the same solvent used for
Rf determination.
A typical example is given in the following. 10-Undecyn-1-ol (42.1
mg, 0.25 mmol) and methylvinyl ketone (26.5 mg, 0.032 mL, 0.375
mmol) were dissolved in the acetone/DMF 1/1 (0.5 mL) and then added
to CpRu(CH3CN)3PF6 (10.9 mg, 0.025 mmol), stannic bromide (16.4
mg, 0.0375 mmol), and ammonium bromide (73.4 mg, 0.75 mmol) in
a pressure tube. The tube was capped and then heated to 60 °C for 2
h. It was then cooled to room temperature and applied directly to a
silica gel column (1/1 petroleum ether/ethyl acetate) to give 60 mg
(75%) of vinyl bromide 48, as a 2.7/1 E/Z-mixture, as determined by
integration of the two vinylic triplets at δ 5.80 and 5.67 (for the E-
and Z-isomers respectively) in the proton NMR spectra.
General Procedure for E-Vinyl Chloride Formation (Table 3).
Method A. The alkyne (0.25 mmol) and enone (0.5 mmol) were added
to a solution of CpRu(COD)Cl (7.7 mg, 0.025 mmol), tin (IV) chloride
pentahydrate (13.2 mg, 0.0375 mmol), and ammonium chloride (44.1
mg, 0.825 mmol) in DMF/water 20/1 (0.5 mL) in a pressure tube. The
tube was capped and then heated to 60 °C for 4 h. It was then cooled
to room temperature and poured into saturated sodium bicarbonate (25
mL). The aqueous layer was extracted with ether (2 × 25 mL). The
organic layer was dried over magnesium sulfate and the solvent removed
by rotary evaporation. The crude mixture was analyzed by proton NMR
and then subjected to silica gel chromatography.
Method B. The alkyne (0.25 mmol) and enone (0.5 mmol) were
added to a solution of CpRu(COD)Cl (7.7 mg, 0.025 mmol), tin (IV)
chloride pentahydrate (13.2 mg, 0.0375 mmol), and tetramethyl-
ammonium chloride (82.2 mg, 0.75 mmol) in DMF (0.5 mL) in a
pressure tube. The tube was capped and then heated to 60 °C for 4 h.
It was then cooled to room temperature and poured into saturated
sodium bicarbonate (25 mL). The aqueous layer was extracted with
ether (2 × 25 mL). The organic layer was dried over magnesium sulfate
and the solvent removed by rotary evaporation. The crude mixture was
analyzed by proton NMR and then subjected to silica gel chromatog-
raphy.
A typical example is given in the following. 3-Butyn-1-ol (17.5 mg,
0.019 mL, 0.25 mmol) and methylvinyl ketone (35.3 mg, 0.042 mL,
0.5 mmol) were added to a solution of CpRu(COD)Cl (7.7 mg, 0.025
mmol), tin (IV) chloride pentahydrate (13.2 mg, 0.0375 mmol), and
ammonium chloride (44.1 mg, 0.825 mmol) in DMF/water 20/1 (0.5
mL) in a pressure tube. The tube was capped and then heated to 60 °C
for 4 h. It was then cooled to room temperature and poured into
saturated sodium bicarbonate (25 mL). The aqueous layer was extracted
with ether (2 × 25 mL). The organic layer was dried over magnesium
sulfate and the solvent removed by rotary evaporation. The crude
mixture was analyzed by proton NMR and then subjected to silica gel
chromatography (1/1 petroleum ether/ethyl acetate) to give 30.1 mg of
vinyl chloride 25 (68%) as a 6.7/1 E/Z-mixture as determined by
integration of the vinylic triplets.
E-6-Bromo-15-hydroxy-pentadec-5-en-2-one (E-48): light yellow
oil. Rf ) 0.38 (1/1 petroleum ether/ethyl acetate). IR (neat): 3412, 2928,
2855, 1717, 1646, 1464, 1427, 1409, 1365, 1198, 1164, 1056 cm-1
.
1H NMR (500 MHz, CDCl3): δ 5.80 (t, J ) 7.7, 1H), 3.65 (t, J ) 6.6,
2H), 2.53 (t, J ) 7.3, 2H), 2.44 (t, J ) 7.2, 2H), 2.29 (q, J ) 7.5, 2H),
2.17 (s, 3H), 1.59-1.51 (m, 5H), 1.36-1.25 (m, 10H). 13C NMR (125
MHz, CDCl3): δ 207.6, 130.2, 127.3, 63.0, 42.6, 35.4, 32.7, 30.0, 29.4,
29.3, 29.2, 28.5, 27.9, 25.7, 23.6. HRMS: Calcd for C15H27BrO2-
H2OBr: 221.1904. Found: 221.1905.
(43) For some examples, see: Jung, M. E.; Light, L. A. Tetrahedron Lett. 1982,
23, 3851. See also: Corey, E. J.; Ulrich, P.; Fitzpatrick, J. M. J. Am. Chem.
Soc. 1976, 98, 222. Ensley, H. E.; Buescher, R. R.; Lee, K. J. Org. Chem.
1982, 47, 404. Takahashi, T.; Xi, C.; Ura, Y.; Nakajima, K. J. Am. Chem.
Soc. 2000, 122, 3228. See also: Knochel, P. In ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991;
Vol. 4, Chapter 4, pp 865-911.
Z-6-Bromo-15-hydroxy-pentadec-5-en-2-one (Z-48): light yellow
oil. Rf ) 0.38 (1/1 petroleum ether/ethyl acetate). IR (neat): 3412, 2928,
2855, 1717, 1646, 1464, 1427, 1409, 1365, 1198, 1164, 1056 cm-1
.
1H NMR (500 MHz, CDCl3): δ 5.67 (t, J ) 6.7, 1H), 3.66 (t, J ) 6.7,
2H), 2.56 (t, J ) 7.2, 2H), 2.44-2.39 (m, 4H), 2.17 (s, 3H), 1.60-
1.52 (m, 5H), 1.40-1.25 (m, 10H). 13C NMR (125 MHz, CDCl3): δ
(44) Satoh, Y.; Serizawa, H.; Hara, S.; Suzuki, A. J. Am. Chem. Soc. 1985,
107, 5225.
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7387

A R T I C L E S
Trost and Pinkerton
208.1, 129.8, 126.4, 63.0, 42.2, 41.4, 32.7, 29.8, 29.4, 29.3, 29.2, 28.3,
28.0, 25.7, 25.6. HRMS: Calcd for C15H27BrO2-H2OBr: 221.1904.
Found: 221.1905.
saturated aqueous sodium chloride (25 mL). The organic layer was
separated, washed two times with saturated aqueous sodium chloride,
dried over magnesium sulfate, and concentrated by rotary evaporation
to yield crude product, which was purified with silica gel chromatog-
raphy (1/2 petroleum ether/ether), giving 25 mg of hydroxyketone 73
(64%). Some of this material (10 mg, 0.04 mmol) was dissolved in
pyridine (1 mL), and benzoyl chloride (0.013 mL, 0.11 mml) was added
at room temperature. The reaction was stirred for 16 h and then stopped
by pouring it into ether (25 mL) and water (25 mL). The organic layer
was separated, washed two times with 1 N aqueous hydrochloric acid,
dried over magnesium sulfate, and concentrated by rotary evaporation
to yield product, which was purified with silica gel chromatography
(1/1 petroleum ether/ether) giving 9.2 mg of 77(82%). The ee was
86.3% determined by chiral HPLC analysis (compared to the racemic
material), separated on a Chiralpak AD column, eluting 90/10 heptane/
2-propanol, with the major isomer eluting in 15.36 min, and the minor
in 20.13 min.
1-Acetoxy-6-hydroxydecan-5,9-dione (73): colorless oil, Rf ) 0.15
(1/2 petroleum ether/ether). IR (neat): 3474, 2923, 1714, 1434, 1366,
1241, 1164, 1105, 1039, 804, 737 cm-1. 1H NMR (300 MHz, CDCl3):
δ 4.19-4.13 (m, 1H), 4.07 (t, J ) 6.0 Hz, 2H), 3.50 (d, J ) 4.9 Hz,
1H), 2.73-2.29 (m, 4H), 2.21-2.19 (m, 1H) 2.17 (s, 3H), 2.05 (s,
3H), 1.70-1.60 (m, 5H).13C NMR (75 MHz, CDCl3): δ 211.6, 208.1,
171.1, 75.3, 63.9, 38.5, 37.3, 30.0, 28.1, 27.4, 20.9, 20.0. Full
characterization was done for the benzoylated compound.
General Procedure for Ruthenium-Catalyzed Z-Vinyl Bromide
Formation Using Lithium Bromide (Table 7). The alkyne (0.25
mmol) and enone (0.375 mmol) were dissolved in the acetone (reagent
grade, not distilled, 0.5 mL) and then added to CpRu(CH3CN)3PF6 (10.9
mg, 0.025 mmol), stannic bromide (16.4 mg, 0.0375 mmol), and lithium
bromide (32.6 mg, 0.375 mmol) in a pressure tube. The tube was capped
and then heated to 60 °C for 2 h. It was then cooled to room temperature
and applied directly to a silica gel column. The eluting solvent for each
case is the same solvent used for Rf determination.
A typical example is given in the following. 5-Cyanopentyne (23.4
mg, 0.25 mmol) and methylvinyl ketone (26.5 mg, 0.032 mL, 0.375
mmol) were dissolved in the acetone (reagent grade, not distilled, 0.5
mL) and then added to CpRu(CH3CN)3PF6 (10.9 mg, 0.025 mmol),
stannic bromide (16.4 mg, 0.0375 mmol), and lithium bromide (32.6
mg, 0.375 mmol) in a pressure tube. The tube was capped and then
heated to 60 °C for 2 h. It was then cooled to room temperature and
applied directly to a silica gel column (1/1 petroleum ether/ethyl ether)
to give 55 mg of vinyl bromide 42 (90%) as a 3.3/1 Z/E mixture, as
determined by integration of the two vinylic triplets at δ 5.89 and 5.80
(for the E- and Z-isomers respectively) in the proton NMR spectra.
E-5-Bromo-9-oxo-dec-5-enenitrile (E-42): light yellow oil. Rf )
0.15 (2/1 petroleum ether/ether). IR (neat): 3529, 2925, 2361, 2247,
1
1713, 1426, 1363, 1165, 1100, 924, 860, 747 cm-1. H NMR (300
1-Acetoxy-6-benzoyloxydecan-5,9-dione (77): yellow oil, Rf ) 0.18
(1/1 petroleum ether/ether). IR (neat): 3064, 2924, 2853, 1722, 1680,
1602, 1452, 1367, 1316, 1272, 1247, 1112 cm-1. 1H NMR (300 MHz,
CDCl3): δ 8.06 (dd, J1 ) 8.2, J2)1.1, 2H), 7.61 (t, J ) 7.3, 1H), 7.47
(t, J ) 7.8, 2H), 5.24 (dd, J1 ) 8.1 Hz, J2 ) 4.4 Hz, 1H), 4.06 (t, J )
6.0 Hz, 2H), 2.67-2.59 (m, 4H), 2.30-2.22 (m, 1H) 2.17 (s, 3H), 2.15-
2.05 (m, 1H), 2.03 (s, 3H), 1.71-1.61 (m, 4H). 13C NMR (75 MHz,
CDCl3): δ 207.0, 206.5, 171.2, 166.0, 133.6, 129.8, 129.1, 128.6, 77.7,
64.1, 38.6, 38.1, 30.1, 27.9, 24.2, 21.0, 19.6. Anal. Calcd for
C19H26O6: C, 65.50; H, 6.94. Found: C, 65.64; H 7.17.
MHz, CDCl3): δ 5.89 (t, J ) 7.7, 1H), 2.62 (t, J ) 7.0, 2H), 2.54 (t,
J ) 7.1, 2H), 2.39-2.31 (m, 4H), 2.14 (s, 3H), 1.93 (quint, J ) 7.0,
2H) 13C NMR (75 MHz, CDCl3): δ 207.0, 133.0, 123.7, 119.3, 42.2,
33.4, 30.0, 25.5, 23.6, 15.7. Anal. Calcd for C10H14BrNO: C, 49.20;
H, 5.78; N, 5.74. Found: C, 49.15; H, 6.00; N, 5.50.
Z-5-Bromo-9-oxo-dec-5-enenitrile (Z-42): light yellow oil. Rf )
0.15 (2/1 petroleum ether/ether). IR (neat): 3529, 2925, 2361, 2247,
1
1713, 1426, 1363, 1165, 1100, 924, 860, 747 cm-1. H NMR (300
MHz, CDCl3): δ 5.80 (t, J ) 6.8, 1H), 2.63-2.52 (m, 4H), 2.41 (q,
J ) 6.8, 2H), 2.32 (t, J ) 7.0, 2H)), 2.16 (s, 3H), 1.91 (quint., J ) 7.1,
2H) 13C NMR (75 MHz, CDCl3): δ 207.6, 129.5, 126.1, 119.2, 41.8,
39.6, 29.8, 25.5, 23.4, 15.5. Anal. Calcd for C10H14BrNO: C, 49.20;
H, 5.78; N, 5.74. Found: C, 49.15; H, 6.00; N, 5.50.
A larger-scale example is given in the following. Phenylacetylene
(510 mg, 5 mmol) and phenylvinyl ketone (990 mg, 7.5 mmol) were
dissolved in the acetone (reagent grade, not distilled, 10 mL) and then
added to CpRu(CH3CN)3PF6 (218 mg, 0.5 mmol), stannic bromide (320
mg, 0.75 mmol), and lithium bromide (652 mg, 7.5 mmol) in a pressure
tube. The tube was capped and then heated to 60 °C for 2 h. It was
then cooled to room temperature and applied directly to a silica gel
column (1/1 petroleum ether/ethyl ether) to give 1.13 g of vinyl bromide
62 (70%). Only the Z-isomer was observed by the presence of a single
vinylic triplet in the proton NMR spectra.
Non-Racemic: separated on Chiralpak AD column (90/10 heptane/
2-propanol, 1 mL/min, 254 nm detection); first enantiomer: 15.36 min
(major); second enantiomer: 20.13 min.
Experimental Details for Equation 11: Cross-Coupling Reaction
of Vinyl Chloride 15. Following the published procedure,26 vinyl
chloride 15 (22 mg, 0.1 mmol) p-acetylbenzeneboronic acid (33 mg,
0.2 mmol), potassium fluoride (20 mg, 0.33 mmol), and Pd2dba3‚CHCl3
(2.6 mg, 0.0025 mmol) were added to a test tube. The tube was sealed
and placed under argon. THF (0.25 mL) was added to the test tube
purged with argon for 5 min. Then, tri-tert-butylphosphine (2.6 mg,
0.0015 mL, 0.006 mmol) was added, and the reaction stirred at room
temperature for 16 h. The reaction was next poured into ether (25 mL)
and extracted three times with water, and then the organic layer was
dried over magnesium sulfate. The ether was removed by rotary
evaporation to give a crude material that was purified by silica gel
chromatography (10/1 petroleum ether/ethyl acetate) to give 22 mg 78
(73%) as the E-isomer. The other isomer was not isolated.
Z-5-Bromo-1,5-diphenyl-pent-4-en-1-one (62): yellow oil. Rf )
0.35 (12/1 petroleum ether/ethyl acetate). IR (neat): 3059, 1683, 1598,
1
1489, 1445, 1405, 1361, 1234, 1178, 1074, 993, 754 cm-1. H NMR
(500 MHz, CDCl3): δ 8.02 (d, J ) 8.1, 2H), 7.61-7.59 (m, 1H), 7.56-
7.49 (m, 4H), 7.37-7.31 (m, 3H), 6.39 (t, J ) 7.1, 2H), 3.24 (t, J )
7.1, 2H), 2.83 (q, J ) 7.1, 2H). 13C NMR (125 MHz, CDCl3): δ 199.0,
136.7, 133.2, 133.0, 130.0, 128.7, 128.6, 128.4, 128.2, 128.1, 127.5,
37.0, 27.0. Anal. Calcd for C17H15BrO: C, 64.78; H, 4.80. Found: C,
64.80; H, 4.73.
General Procedure for Asymmetric Dihydroxylation of Vinyl
Chlorides To Form r-Hydroxyketones (Table 10). A representative
example is given in the following. AD-mix-â (238 mg) and methane-
sulfonamide (16.2 mg, 0.17 mmol) were mixed in tert-butyl alcohol/
water, 1/1 (1 mL), at room temperature and stirred for 30 min. The
reaction was then cooled to 0 °C, and vinyl chloride 22 (41.9 mg, 0.17
mmol) was added. The mixture was stirred at 0 °C for 24 h. The reaction
was stopped by pouring it into ether or ethyl acetate (25 mL) and
6-(4-Acetyl-phenyl)-dodec-5-en-2-one (78): colorless oil. Rf ) 0.17
(15/1 petroleum ether/ether). IR (neat): 2957, 2929, 2858, 2358, 1719,
1687, 1456, 1361, 1264, 1162, 1123, 1100 cm-1. 1H NMR (500 MHz,
CDCl3): δ 7.99-7.97 (m, 1H), 7.60-7.57 (m, 1H), 7.51-7.47 (m,
2H), 5.55 (t, J ) 7.7, 1H), 2.63 (s, 3H), 2.52 (t, J ) 7.3, 2H), 2.37-
2.29 (m, 4H), 2.17 (s, 3H), 1.58-1.52 (m, 2H), 1.35-1.27 (m, 6H),
0.91 (t, J ) 6.9, 3H). 13C NMR (125 MHz, CDCl3): δ 207.5, 198.2,
137.1, 135.4, 133.1, 128.6, 128.3, 125.9, 42.9, 33.6, 31.6, 30.0, 28.4,
27.2, 26.6, 22.6, 22.5, 14.0. HRMS: Calcd for C20H28O2: 300.2089.
Found 300.2095.
Acknowledgment. We thank the National Science Foundation
and the National Institutes of Health, General Medical Sciences,
for their generous support of our programs. Mass spectra were
9
7388 J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002

Vinyl Halides Formed by Ru-Catalyzed Reaction
A R T I C L E S
provided by the Mass Spectrometry Regional Center of the
University of California-San Francisco, supported by the NIH
Division of Research Resources. A.B.P. thanks Abbott Labo-
ratories, Glaxo-Wellcome, the ACS Division of Organic Chem-
istry (sponsored by Bristol-Myers Squibb), and Eli Lilly for
fellowship support.
Supporting Information Available: Experimental details for
optimization reactions as well as characterization data (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
JA011426W
9
J. AM. CHEM. SOC. VOL. 124, NO. 25, 2002 7389