Ruthenium-catalyzed transformation of aryl and alkenyl triflates to halides

Article abstract of DOI:10.1021/ja307771d

Aryl triflates were transformed to aryl bromides/iodides simply by treating them with LiBr/NaI and [Cp*Ru(MeCN)₃]OTf. The ruthenium complex also catalyzed the transformation of alkenyl sulfonates and phosphates to alkenyl halides under mild conditions. Aryl and alkenyl triflates undergo oxidative addition to a ruthenium(II) complex to form η'¹- arylruthenium and 1-ruthenacyclopropene intermediates, respectively, which are transformed to the corresponding halides.

Full text of DOI:10.1021/ja307771d

Communication
pubs.acs.org/JACS
Ruthenium-Catalyzed Transformation of Aryl and Alkenyl Triflates to
Halides
Yusuke Imazaki, Eiji Shirakawa,* Ryota Ueno, and Tamio Hayashi*
Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo, Kyoto 606-8502, Japan
S
* Supporting Information
Table 1. Ruthenium-Catalyzed Transformation of 4-
a
ABSTRACT: Aryl triflates were transformed to aryl
Acetylphenyl Triflate to 4′-Bromoacetophenone
bromides/iodides simply by treating them with LiBr/NaI
and [Cp*Ru(MeCN)3]OTf. The ruthenium complex also
catalyzed the transformation of alkenyl sulfonates and
phosphates to alkenyl halides under mild conditions. Aryl
time
b
and alkenyl triflates undergo oxidative addition to a
ruthenium(II) complex to form η1-arylruthenium and 1-
ruthenacyclopropene intermediates, respectively, which are
transformed to the corresponding halides.
entry
[Ru]
solvent
mBr
LiBr
(h) yield (%)
1
2
3
4
5
[Cp*Ru(MeCN)3]OTf DMI
[Cp*Ru(MeCN)3]OTf NMP
[Cp*Ru(MeCN)3]OTf DMF
[Cp*Ru(MeCN)3]OTf DMSO
[Cp*Ru(MeCN)3]OTf 1,4-
12
12
12
12
12
98
97
89
0
LiBr
LiBr
LiBr
LiBr
4
dioxane
ryl halides are versatile substrates under transition-metal
1
6
7
8
9
[Cp*Ru(MeCN)3]OTf DMI
[Cp*Ru(MeCN)3]OTf DMI
[Cp*Ru(MeCN)3]OTf DMI
[(C5Me4CF3)
Ru(MeCN)3]PF6
NaBr
Bu4NBr
KBr
12
12
12
12
98
98
56
36
Acatalysis and convenient precursors for arylmagnesium/
lithium reagents2 and aryl radicals.3 Simple aryl halides are
conventionally prepared by electrophilic aromatic substitution
with halogens4 or the Sandmeyer reaction5 under relatively harsh
conditions. For the synthesis of relatively complicated ones, a
three-step scheme starting from phenols and consisting of
trifluoromethanesulfonylation,6 palladium-catalyzed stannyla-
tion/borylation, and halogenolysis has been widely used as a
reliable method.7−10 As a straightforward method, Buchwald and
co-workers have recently developed a palladium-catalyzed direct
transformation of aryl and alkenyl triflates to the corresponding
bromides and chlorides.11 However, it is not applicable to the
synthesis of aryl iodides. On the other hand, we had already
reported that low-valent ruthenium complexes generated, for
example, from Ru(acac)3 and EtMgBr catalyze the trans-
formation of alkenyl triflates to halides, though aryl triflates are
unreactive under the ruthenium catalysis.12 Here we report that
DMI
LiBr
c
10
11
12
13
14
[Cp*RuCl]4
[Cp*RuCl2]2
[Cp*Ru(MeCN)3]OTf DMI
[Cp*RuCl]4
[Cp*RuCl2]2
DMI
DMI
LiBr
LiBr
LiBr
LiBr
LiBr
12
12
2
96 (2)
c
95 (1)
74
c
DMI
DMI
2
48 (<1)
c
2
5 (<1)
a
The reaction was carried out in a solvent (1.0 mL) under a nitrogen
atmosphere using 1a (0.25 mmol) and mBr (0.38 mmol) in the
b
presence of a ruthenium complex (12.5 μmol of Ru). Determined by
c
1H NMR analysis. The yield of 4′-chloroacetophenone is shown in
parentheses.
electron-donating character of the Cp* ligand is crucial.
[Cp*RuCl]4 and [Cp*RuCl2]2 also catalyzed the bromination,
though contamination with 4′-chloroacetophenone was ob-
served (entries 10 and 11). The catalytic activities of the Cp*Ru
complexes were compared using the yields of 2a in a short
reaction period (2 h) (entries 12−14). Moderate conversion was
observed with [Cp*Ru(MeCN)3]OTf or [Cp*RuCl]4, whereas
only a 5% yield of 2a was produced with [Cp*RuCl2]2. It is likely
that the catalytically active species is a Ru(II) complex and that
reluctant reduction of Ru(III) to Ru(II) caused the induction
period with [Cp*RuCl2]2.
A wide variety of aryl triflates were converted into the
corresponding bromides under the conditions for entry 1 of
Table 1. Phenyl triflates having an electron-withdrawing group
such as acetyl, ethoxycarbonyl, nitro, or cyano at the para
position reacted in high yields (Table 2, entries 1−4). Electron-
rich aryl triflates were less reactive. The use of twice the amounts
−
Cp*Ru (Cp* = C5Me5 ) complexes efficiently catalyze the
transformation of aryl triflates to aryl bromides and iodides. The
Cp*Ru complexes were found to be much more catalytically
active than the low-valent ruthenium complexes in the
transformation of alkenyl triflates to halides.
Treatment of 4-acetylphenyl triflate (1a) with [Cp*Ru-
(MeCN)3]OTf (5 mol %) and LiBr (1.5 equiv) in 1,3-
dimethyl-2-imidazolidinone (DMI) at 100 °C for 12 h gave a
98% yield of 4′-bromoacetophenone (2a) (Table 1, entry 1).13
The reaction also proceeded in high yield in N-methyl-2-
pyrrolidone (NMP) and N,N-dimethylformamide (DMF)
(entries 2 and 3). In contrast, almost no reaction took place in
1,4-dioxane or dimethyl sulfoxide (DMSO) (entries 4 and 5).
NaBr and Bu4NBr also worked as bromide sources, whereas KBr
was less effective (entries 6−8). The use of [(C5Me4CF3)Ru-
(MeCN)3]OTf as the catalyst, in which C5Me4CF3 has steric and
electronic effects similar to those of Cp* and Cp, respectively,14
resulted in slow conversion (entry 9), indicating that strong
Received: August 6, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja307771d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
a
Table 2. Transformation of Aryl Triflates to Halides
Table 3. Transformation of Alkenyl Esters to Halides
a
The reaction was carried out in DMI (1.0 mL) at 25 °C under a
nitrogen atmosphere using 4−6 (0.25 mmol) and mX (0.38 mmol) in
b
the presence of [Cp*Ru(MeCN)3]OTf (12.5 μmol). Isolated yields.
c
The reaction was carried out in THF (1.0 mL) at 60 °C using
Ru(acac)3 (12.5 μmol), EtMgBr (50 μmol), and 1,10-phenanthroline
(12.5 μmol). LiBr (1.5 mmol) was used.
d
were converted to −Br (entries 5 and 6). These leaving groups
have economical advantages over triflate. Cyclic alkenyl
bromides having an acetal or steroidal moiety also were obtained
(entries 7 and 8).
Interestingly, both (E)- and (Z)-1-octen-1-yl triflate (4d) were
transformed to (E)-1-bromo-1-octene (7d) upon reaction at
−20 °C for 1 h (Table 4, entries 1 and 2). At 25 °C, E/Z
Table 4. Stereoselective Transformation of Acyclic Alkenyl
a
Triflates to Halides
a
The reaction was carried out in DMI (1.0 mL) at 100 °C under a
nitrogen atmosphere using 1 (0.25 mmol) and mX (0.38 mmol) in the
b
c
presence of [Cp*Ru(MeCN)3]OTf (12.5 μmol). Isolated yields. At
d
e
120 °C. [Cp*Ru(MeCN)3]OTf (25 μmol) was used. mX (0.75
mmol) was used. [Cp*Ru(MeCN)3]OTf (37.5 μmol) was used. mX
f
g
(1.5 mmol) was used.
of the ruthenium catalyst and LiBr was required to convert tert-
butyl- and methoxy-substituted triflates to bromides within
acceptable reaction periods (entries 5 and 6). o-Cyanophenyl
triflate (1g) underwent the bromination (entry 7). Naphthyl
triflates were readily transformed to naphthyl bromides (entries 8
and 9). Chloroarene, ketone, and acetal moieties remained intact
(entries 9−11). Heteroaryl triflates were converted into
bromides (entries 12−14). In contrast to the observation that
the palladium catalysis does not afford aryl iodides,11a,b the
ruthenium catalysis gave aryl iodides when NaI was used instead
of LiBr (entries 15−18). Electron-rich aryl triflates also showed
low reactivities for the iodination (entry 18).
Alkenyl triflates were transformed to halides much more
smoothly than in our previous report.12a In the presence of
[Cp*Ru(MeCN)3]OTf (5 mol %), 4-tert-butylcyclohexen-1-yl
triflate (4a) was transformed to bromide 7a in 10 min at 25 °C
(Table 3, entry 1). In contrast, 12 h was required when a low-
valent ruthenium complex was used, even at 60 °C (entry 2).12a
Transformations of 4a to the iodide and chloride also took place
in high yields (entries 3 and 4). In addition, −OTs and
−OPO(OPh)2, which are poorer leaving groups than −OTf,
a
The reaction was carried out in THF (1.0 mL) at −20 °C under a
nitrogen atmosphere using 4 (0.25 mmol) and LiBr (0.38 mmol) in
b
the presence of [Cp*Ru(MeCN)3]OTf (12.5 μmol). Isolated yields.
c
d
At 25 °C. LiBr (0.75 mmol) and [Cp*Ru(MeCN)3]OTf (25 μmol)
e
f
were used. (E,E)/(E,Z)/(Z,Z). LiCl (0.38 mmol) was used instead of
LiBr, and the alkenyl chloride (9j) was obtained.
B
dx.doi.org/10.1021/ja307771d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
isomerization of 7d occurred, and the E/Z ratio became constant
at 36/64 (entries 3−5). These results indicate that (E)-7d is the
kinetically favored product and that it is generated from the same
intermediate in the reactions of both (E)-4d and (Z)-4d. The
isomerization is ascribed to reentry of the product into the
ruthenium catalysis. 1-Alken-1-yl triflates are much more easily
prepared as mixtures of the stereoisomers than in an E-pure
form.15 Thus, this transformation is especially useful for the
preparation of (E)-haloalkenes. Stereoisomeric mixtures of
alkenyl triflates having a cycloalkane, benzene, alkene, or
chloroalkane moiety were transformed to (E)-alkenyl bromides
in the reaction at −20 °C (entries 6−9). A ditriflate underwent
dibromination (entry 10). An α,β-disubstituted (E)-vinyl
chloride also was obtained (entry 11).
Scheme 3. Plausible Mechanism for the Transformation of
Alkenyl Triflates to Halides
D′ rather than D is considered to be the intermediate,17 taking
into account high ring strain probably induced in a form of D.
On the other hand, the transformation of aryl triflates is
unlikely to include D or D′ because the loss of aromaticity upon
its formation would be unsuitable.18 Alternatively, activation of
aryl triflates through a different mode of oxidative addition to
Ru(II) could possibly be operative. Thus, a catalytic cycle
including η1-arylruthenium(IV) triflate F (Scheme 4) is a
A similar characteristic stereochemical outcome has been
reported in osmium-mediated substitution reactions of alkenyl
ethers.16 Thus, both (E)- and (Z)-1-ethoxypropenes coordinated
to osmium undergo acid-promoted substitution of EtO with
PPh3 to give (E)-propenylphosphonium salts (Scheme 1). The
Scheme 1. Osmium-Mediated Substitution Reaction of
Alkenyl Ethers
Scheme 4. Plausible Mechanism for the Transformation of
Aryl Triflates to Halides
Scheme 2. Competition Reaction of Alkenyl Phosphates
possibility.19 Oxidative addition of ArOTf 1 to Cp*RuX B
generates F.20,21 After coordination of X− to F, reductive
elimination of aryl halide 2/3 from η1-arylruthenium(IV)
complex G regenerates B. The higher reactivities of electron-
deficient aryl triflates compared with electron-rich ones (Table
2) are consistent with this type of oxidative addition. The result
that aryl iodide 3o was more reactive than triflate 1p in the
competition reaction (Scheme 5) is in good agreement with the
exclusive formation of the E isomer is ascribed to the
intermediacy of 1-osmacyclopropene A, which accepts attack
of PPh3 at cationic C2 selectively from the opposite site of Me on
C3. The result that 2-methoxypropene was much more reactive
than methoxyethene in substitution in the coordination sphere of
osmium can also be rationally understood by the involvement of
a 1-osmacyclopropene, which would be stabilized by an α-alkyl
substituent. A similar substituent effect was observed in the
ruthenium-catalyzed transformation of alkenyl triflates to halides
(Scheme 2). α-Alkyl vinyl phosphate 6k was preferentially
consumed over non-α-substituted vinyl phosphate 6d in a
competition reaction.
Scheme 5. Effect of Leaving Groups
The observed similarities prompt us to propose that the
present reaction also proceeds through a 1-metallacyclopropene
intermediate, as shown in Scheme 3. Coordination of alkenyl
triflate 4 to ruthenium to give C followed by oxidative addition
upon elimination of −OTf gives 1-ruthenacyclopropene D.
Attack on D by an outer sphere X− or a 1,2-shift of X− from Ru to
C within D gives ruthenium−haloalkene complex E, which
undergoes elimination to give alkenyl halide 7−9, regenerating
B. For cyclic alkenyl triflates, η2-alkenylruthenium(IV) complex
general reactivity order observed in oxidative addition of aryl
electrophiles to transition-metal complexes such as palladium(0)
complexes.22 The opposite reactivity order was observed with
alkenyl electrophiles, where alkenyl iodide 8l was less reactive
than triflate 4d.23 These results support the conclusion that the
reaction mechanism with aryl triflates is different from that with
alkenyl triflates.
C
dx.doi.org/10.1021/ja307771d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Zhang, Y.; García-Fortanet, J.; Kinzel, T.; Buchwald, S. L. Science 2009,
325, 1661.
In conclusion, we have developed a ruthenium-catalyzed
transformation of aryl and alkenyl triflates to the corresponding
bromides and iodides. The strong electron-donating character of
the Cp* ligand likely contributes to the high catalytic activity by
facilitating oxidative addition of aryl and alkenyl triflates to
ruthenium(II) complexes to give arylruthenium(IV) and 1-
ruthenacyclopropene complexes, respectively.
(12) (a) Shirakawa, E.; Imazaki, Y.; Hayashi, T. Chem. Commun. 2009,
5088. The low-valent ruthenium catalysis is also effective for the
reaction of alkenyl triflates with zinc thiolates to give alkenyl sulfides.
See: (b) Imazaki, Y.; Shirakawa, E.; Hayashi, T. Tetrahedron 2011, 67,
10212.
(13) The most efficient conditions found in the previous study12a
[Ru(acac)3 (3 mol %), EtMgBr (12 mol %), and 3,4,7,8-tetramethyl-
1,10-phenanthroline (3 mol %) in 1,4-dioxane at 120 °C] were totally
ineffective for the transformation of 1a to 2a. Also, bromide 2a was not
produced at all in the reaction in DMI.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and characterization data for all the
products. This material is available free of charge via the Internet
at http://pubs.acs.org.
(14) Gassman, P. G.; Mickelson, J. W.; Sowa, J. R., Jr. J. Am. Chem. Soc.
1992, 114, 6942.
(15) Mixtures of stereoisomers can readily be obtained from aldehydes
simply by treatment with Tf2O and a base such as 2,6-di-tert-butyl-4-
methylpyridine. See ref 6 and: (a) Stang, P. J.; Treptow, W. Synthesis
1980, 283. (E)-1-Alken-1-yl triflates can be synthesized from the
corresponding silyl enolates in a stereoretentive manner. However, the
stereoselective synthesis of (E)-silyl enolates is generally not facile. See:
(b) Matsuzawa, S.; Horiguchi, Y.; Nakamura, E.; Kuwajima, I.
Tetrahedron 1989, 45, 349. (c) Ohmura, T.; Yamamoto, Y.; Miyaura,
N. Organometallics 1999, 18, 413.
AUTHOR INFORMATION
Corresponding Author
■
shirakawa@kuchem.kyoto-u.ac.jp; thayashi@kuchem.kyoto-u.
ac.jp
Notes
The authors declare no competing financial interest.
(16) Chen, H.; Harman, W. D. J. Am. Chem. Soc. 1996, 118, 5672.
(17) The existence of an apparent border between 1-metal-
lacyclopropenes and η2-alkenylmetals has been debated. See:
(a) Casey, C. P.; Brady, J. T.; Boller, T. M.; Weinhold, F.; Hayashi, R.
K. J. Am. Chem. Soc. 1998, 120, 12500. (b) Frohnapfel, D. S.; Templeton,
J. L. Coord. Chem. Rev. 2000, 206−207, 199. To the best of our
knowledge, no reports of the observation of 1-ruthenacyclopropenes or
η2-alkenylrutheniums are available. However, both of these have been
proposed as intermediates in antihydrosilylation of alkynes catalyzed by
[Cp*Ru(MeCN)3]+ on the basis of DFT calculations. See: (c) Chung,
L. W.; Wu, Y.-D.; Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2003, 125,
11578.
(18) Naphthyl triflates are more reactive than phenyl triflates, probably
because a 10-π-electron (10π) aromatic system is much less susceptible
to the loss of π-bond character than a 6π system. Naphthyl triflates are
likely to be transformed via D′ in a similar manner as alkenyl triflates.
(19) The result that [Cp*Ru(η6-1e)]OTf showed no catalytic activity
under the conditions of Table 2, entry 5 excludes SNAr pathways
accelerated by π complexation with the metal. Ruthenium complexes are
known to catalyze or mediate SNAr reactions of aryl halides through
ruthenium−η6-haloarene complexes. See: (a) Otsuka, M.; Endo, K.;
Shibata, T. Chem. Commun. 2010, 46, 336. (b) Otsuka, M.; Yokoyama,
H.; Endo, K.; Shibata, T. Synthesis 2010, 2601. (c) Dembek, A. A.; Fagan,
P. J. Organometallics 1996, 15, 1319. (d) West, C. W.; Rich, D. H. Org.
Lett. 1999, 1, 1819.
ACKNOWLEDGMENTS
■
This work was financially supported in part by a Grant-in-Aid for
the Global COE Program “Integrated Materials Science” from
the Ministry of Education, Culture, Sports, Science and
Technology of Japan. Y.I. thanks the JSPS for a Research
Fellowship for Young Scientists.
REFERENCES
■
(1) Hartwig, J. F. Organotransition Metal Chemistry: From Bonding to
Catalysis; University Science Books: Sausalito, CA, 2010; pp 877−965.
(2) (a) Rieke, R. D. Science 1989, 246, 1260. (b) Knochel, P.; Dohle,
W.; Gommermann, N.; Kneisel, F. F.; Kopp, F.; Korn, T.; Sapountzis, I.;
Vu, V. A. Angew. Chem., Int. Ed. 2003, 42, 4302.
(3) (a) Jasperse, C. P. Chem. Rev. 1991, 91, 1237. (b) Rossi, R. A.;
Pierini, A. B.; Penenory, A. B. Chem. Rev. 2003, 103, 71.
́
̃
̃
(4) Larock, R. C. Comprehensive Organic Transformations, 2nd ed.;
Wiley-VCH: New York, 1999; pp 619−626.
(5) Hodgson, H. H. Chem. Rev. 1947, 40, 251.
(6) Ritter, K. Synthesis 1993, 735.
(7) Via alkenylstannanes: (a) Wulff, W. D.; Peterson, G. A.; Bauta, W.
E.; Chan, K.-S.; Faron, K. L.; Gilbertson, S. R.; Kaesler, R. W.; Yang, D.
C.; Murray, C. K. J. Org. Chem. 1986, 51, 277. (b) Grunewald, G. L.;
Seim, M. R.; Regier, R. C.; Criscione, K. R. Bioorg. Med. Chem. 2007, 15,
1298. (c) Rawat, M.; Prutyanov, V.; Wulff, W. D. J. Am. Chem. Soc. 2006,
128, 11044. Via alkenylboronates: (d) Thompson, A. L. S.; Kabalka, G.
W.; Akula, M. R.; Huffman, J. W. Synthesis 2005, 547.
(8) Aryl triflates are efficient alternatives to aryl halides in transition-
metal-catalyzed reactions. However, they cannot be used as precursors
for arylmagnesium/lithium reagents and aryl radicals.
(9) A few examples of the direct synthesis of aryl halides from phenols
are available, but they require forcing conditions. See: (a) Wiley, G. A.;
Hershkowitz, R. L.; Rein, B. M.; Chung, B. C. J. Am. Chem. Soc. 1964, 86,
964. (b) Bay, E.; Bak, D. A.; Timony, P. E.; Leone-Bay, A. J. Org. Chem.
1990, 55, 3415.
(10) Strongly electron-deficient aryl triflates are transformed to the
halides by nucleophilic aromatic substitution (SNAr). For recent
examples, see: (a) Kundu, S. K.; Tan, W. S.; Yan, J.-L.; Yang, J.-S. J.
Org. Chem. 2010, 75, 4640. (b) Trost, B. M.; O’Boyle, B. M. Org. Lett.
2008, 10, 1369. (c) Wang, Z.; Shangguan, N.; Cusick, J. R.; Williams, L.
J. Synlett 2008, 213.
(20) An aryl ruthenium(II) complex is known to react with
bromobenzene to give the corresponding phenylarene. The reaction is
considered to proceed through oxidative addition and reductive
elimination. See: (a) Oi, S.; Funayama, R.; Hattori, T.; Inoue, Y.
Tetrahedron 2008, 64, 6051. For a review of the related catalytic
reactions, see: (b) Ackermann, L.; Vicente, R.; Kapdi, A. R. Angew.
Chem., Int. Ed. 2009, 48, 9792.
(21) Reversible oxidative addition and reductive elimination between
allyl halides and Cp*Ru(II) complexes are known. See: (a) Nagashima,
H.; Mukai, K.; Itoh, K. Organometallics 1984, 3, 1314. (b) Nagashima,
H.; Mukai, K.; Shiota, Y.; Yamaguchi, K.; Ara, K.; Fukahori, T.; Suzuki,
H.; Akita, M.; Moro-oka, Y.; Itoh, K. Organometallics 1990, 9, 799.
(22) Jutand, A.; Mosleh, A. Organometallics 1995, 14, 1810.
(23) In oxidative additions of alkenyl electrophiles (C to D in Scheme
3), the reactivity is likely to be governed simply by the stability of the
−
leaving anion, where OTf is more stable than I−. For relative leaving
group abilities in SN1 reactions, see: Noyce, D. S.; Virgilio, J. A. J. Org.
Chem. 1972, 37, 2643.
(11) (a) Shen, X.; Hyde, A. M.; Buchwald, S. L. J. Am. Chem. Soc. 2010,
132, 14076. (b) Pan, J.; Wang, X.; Zhang, Y.; Buchwald, S. L. Org. Lett.
2011, 13, 4974. The transformation of aryl triflates to aryl fluorides has
also been reported. See: (c) Watson, D. A.; Su, M.; Teverovskiy, G.;
D
dx.doi.org/10.1021/ja307771d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX