Manganese ate complexes as new reducing agents: Perfectly regiocontrolled generation and reactions of the manganese enolates with electrophiles [9]

Full text of DOI:10.1021/ja964054d

J. Am. Chem. Soc. 1997, 119, 5459-5460
5459
Scheme 1
Manganese Ate Complexes as New Reducing
Agents: Perfectly Regiocontrolled Generation and
Reactions of the Manganese Enolates with
Electrophiles
Makoto Hojo, Hajime Harada, Hajime Ito, and
Akira Hosomi*
corresponding aldol adducts 2 were obtained.6b,d,7 This reaction
is reminiscent of the similar generation of enolates from
R-halogeno ketones with organocuprates shown by Posner8 and
may be possibly related to the partial reductive dimerization of
cyclohexenone at â-position in the 1,4-addition to the enone
using Bu3MnLi reported by Cahiez.9 Selected results are
summarized in Table 1.10
Department of Chemistry, UniVersity of Tsukuba
Tsukuba, Ibaraki 305, Japan
ReceiVed NoVember 25, 1996
Among transition metal ate complexes, the organocuprates
have been predominantly investigated and utilized in organic
synthesis, and reactions using these species are mainly based
on their alkylating abilities.1 The reduction rarely takes place
as a side reaction in these alkylations. This reductive nature of
cuprates has been utilized for the direct generation of organo-
copper reagents in the cases where only a reductive process
occurs.2,3 The organomanganese reagents were also reported
to be alkylating agents.4 We report the generation of the
enolates from the carbonyl compounds bearing a leaving group
such as an acetoxy, siloxy, or halogen group at the R position
by using the organomanganese(II) ate complexes and their
characteristic reactions with electrophiles. For the generation
of the enolates, a new reactivity of the organomanganese ate
complexes as a reductant can thus be exploited (Scheme 1).
We found that “Bu3MnLi” cleanly reduced R-acetoxypro-
piophenone (1b) to generate an enolate of propiophenone, and
it reacted with benzaldehyde to afford the cross aldol adduct
2d in 87% yield (syn/anti ) 87:13).5 Other aromatic ketones
(1) bearing an acetoxy group or even a siloxy group were also
efficiently reduced to the enolates, irrespective of their structures.
After the aldehydes were added to the reaction mixtures, the
The oxidative addition and reductive elimination were thought
to be the key processes in the present reduction by the following
ligand analysis experiments. Protonolysis of lithium tridecyl-
manganate at -20 °C, prior to the addition of a substrate,
resulted in recovery of decane along with a trace amount of
decene and eicosane. This suggests there is no generation of a
low-valent manganese as a reductant for the substrates 1.11 On
the other hand, the decylmanganate complex reduced 1b at the
same temperature (-20 °C), and the resulting enolate reacted
with benzaldehyde to give 2d in 81% yield (syn/anti ) 76:24).
Interestingly, in the reaction mixture, 46% of decene and 31%
of eicosane (based on manganese) were found by NMR and
GC analyses together with decane, while decyl acetate was not
detected. These observations deserve some comments. Lithium
tridecylmanganate, a new manganese ate complex bearing a
ligand other than the butyl group, also acts as a reducing agent
to 1b. The fact that decene and eicosane can be detected only
during the reduction reaction of 1 suggests a kind of oxidative
addition of the substrate to the ate complex in these reductive
generation of the enolates, and the subsequent reductive
elimination of the ligands on manganese are possibly involved.
Since there is no appearance of decyl acetate, a process
equivalent to a “metal-halogen exchange” reaction does not
seem to be significant.
(1) For the ate complexes in organic synthesis, see: (Cu) (a) Lipshutz,
B. H.; Sengupta, S. In Organic Reactions; John Wiley & Sons: New York,
1992; Vol. 41, Chapter 2, pp 135-631. (b) Posner, G. H. In An Introduction
to Synthesis Using Organocopper Reagents; Wiley: New York, 1980.
Organozincates are also known to be alkylating reagents as well as
metalating reagents by metal-halogen exchange reaction. For recently
reported reactions using organozincates, see: (Zn) (c) Harada, T.; Katsuhira,
T.; Osada, A.; Iwazaki, K.; Maejima, K; Oku, A. J. Am. Chem. Soc. 1996,
118, 11377-11390. (d) McWilliams, J. C.; Armstrong, J. D., III; Zheng,
N.; Bhupathy, M.; Volante, R. P.; Reider, P. J. J. Am. Chem. Soc. 1996,
118, 11970-11971. (e) Uchiyama, M.; Koike, M.; Kameda, M.; Kondo,
Y.; Sakamoto, T. J. Am. Chem. Soc. 1996, 118, 8733-8734. For 1,4-
additions of zincates to R,â-enones, see: (f) Kjonass, R. A.; Hoffer, R. K.
J. Org. Chem. 1988, 53, 4133-4135 and references cited therein. For ate
complexes derived from main group elements used in organic synthesis as
alkylation agents, see: (B) (g) Vaultier, M.; Carboni, B. In ComprehensiVe
Organometallic Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G.,
Eds.; McKillop, A., Vol. Ed.; Pergamon Press: Oxford, 1995; Vol. 11, pp
191-276. (Al) (h) Hauske, J. R. In ComprehensiVe Organic Synthesis; Trost,
B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 1, pp 77-
106. (Si) (i) Chuit, C.; Corriu, R. J. P.; Reye, C.; Young, J. C. Chem. ReV.
1993, 93, 1371-1448. (j) Hosomi, A. In ReViews on Heteroatom Chemistry;
Oae, S., Ed.; MYU: Tokyo, 1992; Vol. 7, pp 214-228.
(2) (a) Corey, E. J.; Kuwajima, I. J. Am. Chem. Soc. 1970, 92, 395-
396. (b) Ibuka, T.; Aoyagi, T.; Kitada, K.; Yoneda, F.; Yamamoto, Y. J.
Organomet. Chem. 1985, 287, C18-C22. (c) Krause, N.; Handke, G.
Tetrahedron Lett. 1991, 32, 7229-7232. (d) Fujii, N.; Habasita, H.;
Shigemori, N.; Otaka, A.; Ibuka, T.; Tanaka, M.; Yamamoto, Y. Tetrahedron
Lett. 1991, 32, 4969-4972. See also ref 8.
(3) (a) Hojo, M.; Harada, H.; Hosomi, A. Chem. Lett. 1994, 437-440.
(b) Hojo, M.; Harada, H.; Watanabe, C.; Hosomi, A. Bull. Chem. Soc. Jpn.
1994, 67, 1495-1498. (c) Hojo, M.; Harada, H.; Murakami, C.; Hosomi,
A. J. Chem. Soc., Chem. Commun. 1994, 2687-2688.
(5) The precise structures of the species expressed here as a formula
“R3MnLi” and terms “lithium trialkylmanganate” are not clear at present,
and these are tentatively used. As a control experiment, a reaction of 1b
with 3 equiv of halogen-free BuLi in THF at -20 °C afforded a complex
mixture.
(6) To date, reported methods for the generation of manganese enolates
are as follows. Deprotonation using manganese amides: (a) Cahiez, G.;
Figadere, B.; Tozzolino, P.; Cle´ry, P. Fr. Pat. Appl. 88/15,806, 1988; Eur.
Pat. Appl. EP 373,993, 1990; Chem. Abstr. 1991, 114, 61550y. (b) Cahiez,
G.; Cle´ry, P.; Laffitte, J. A. Fr. Pat. Appl. 90/16,413, 1990; Chem. Abstr.
1993, 118, 169340b. (c) Cahiez, G.; Figade`re, B.; Cle´ry, P. Tetrahedron
Lett. 1994, 35, 3065-3068. Transmetalation using lithium enolates: (d)
Cahiez, G.; Cle´ry, P.; Laffitte, J. A. Fr. Pat. Appl. 91/11,814, 1991; PCT
Int. Appl. WO 93/06,071, 1993; Chem. Abstr. 1993, 119, 116519f. (e)
Cahiez, G.; Chau, K.; Cle´ry, P. Tetrahedron Lett. 1994, 35, 3069-3072.
1,4-Addition to R,â-unsaturated carbonyl compounds: (f) Cahiez, G; Alami,
M. Tetrahedron Lett. 1989, 30, 3541-3544. (g) Takai, K.; Ueda, T.;
Kaihara, H.; Sunami, Y.; Moriwake, T. J. Org. Chem. 1996, 61, 8728-
8729.
(7) The Barbier-type aldol reaction mediated by manganese was re-
ported: Cahiez, G.; Chavant, P.-Y. Tetrahedron Lett. 1989, 30, 7373-
7376.
(8) (a) Posner G. H.; Sterling, J. J. J. Am. Chem. Soc. 1973, 95, 3076-
3077. (b) Wakselman, C.; Mondon, M. Tetrahedron Lett. 1973, 4285-
4288. (c) Posner, G. H.; Sterling, J. J.; Whitten, C. E.; Lentz, C. M.;
Brunelle, D. J. J. Am. Chem. Soc. 1975, 97, 107-118. (d) Lion, C.; Dubois,
J.-E. Tetrahedron 1975, 31, 1223-1226. (e) Dubois, J.-E.; Lion, C.
Tetrahedron 1975, 31, 1227-1231.
(9) Cahiez, G.; Alami, M. Tetrahedron Lett. 1986, 27, 569-572. This
reductive dimerization may be a direct precedent for the present reductive
generation of enolates.
(10) For a general procedure, see: Supporting Information.
(11) As for the stability of Dec3MnLi (Dec ) decyl) in THF, 86% of
decene (based on manganese) was formed after a solution of the ate complex
was stirred at room temperature (rt) for 10 h and 154% at reflux for 4 h
together with <5% of eicosane. When Dec3MnLi was once refluxed for 4
h to yield decene, the reaction of 1b with benzaldehyde under the same
conditions as those in Table 1 resulted in the formation of R-hydroxypro-
piophenone (58%) and the recovery of 1b (32%).
(4) The following are examples for the alkylation with representative
electrophiles using monoalkyl- and dialkylmanganese reagents and trialky-
lmanganate reagents prepared from manganese(II) salts and organolithium
or magnesium reagents: (a) Corey, E. J.; Posner, G. H. Tetrahedron Lett.
1970, 315-318. (b) Cahiez, G.; Masuda, A.; Bernard, D.; Normant, J. F.
Tetrahedron Lett. 1976, 3155-3156. (c) Cahiez, G.; Normant, J. F.
Tetrahedron Lett. 1977, 3383-3384. (d) Cahiez, G.; Alami, M. Tetrahedron
1989, 45, 4163-4176. (e) Dialkylation of gem-dibromocyclopropanes with
trialkylmanganates involving a metalation step was recently reported: Inoue,
R.; Shinokubo, H.; Oshima, K. Tetrahedron Lett. 1996, 37, 5377-5380.
S0002-7863(96)04054-1 CCC: $14.00 © 1997 American Chemical Society

5460 J. Am. Chem. Soc., Vol. 119, No. 23, 1997
Communications to the Editor
ester and a halo ketone to afford the â-keto ester 2n6c and the
â,γ-epoxy ketone 2o, respectively. These characteristic reac-
tivities strongly suggest that the metal counterion of the enolate
in these reactions may be manganese and not lithium.
Table 1. Aldol Reaction of the Ketone Enolates Generated from
the R-Oxy Ketones 1 Using Manganese(II) Ate Complexa
The absence of any polyalkylation products may stem from
the fact that the protonation-deprotonation equilibrium between
the alkylated ketone and the unreacted enolate is negligible under
the alkylation conditions.6 Indeed, the perfectly regiospecific
generation and alkylations of the unsymmetrical and non-
branched aliphatic ketone enolates were achieved (eqs 5-7).
Especially, a regioisomeric pair of the iodo ketones 1g and 1h16
regiospecifically gave the isomeric alkylated products 2p and
2q, respectively.17 In these reactions, the kinetic generation and
the subsequent alkylation of each enolate took place without
accompanying isomerization. Such transformations are difficult
to attain by the deprotonation-alkylation methods under both
kinetic and thermodynamic conditions. This protocol can be
applicable to the generation of enolates from the R-bromo ester
1j and amide 1k (eq 8).
a For the generation of an enolate, see ref 10. To the enolate, an
aldehyde (1 mmol, 4 equiv to 1) was added, and the mixture was stirred
at -78 °C for 30 min. b Isolated yield by column chromatography. The
isomeric ratio is shown in parentheses. c Si ) t-BuMe2Si. d Bu3MnLi
(0.75 mmol, 1.5 equiv to 1) was used. e Bu3MnLi (0.65 mmol, 1.3 equiv
to 1) was used.
Scheme 2
One of plausible mechanisms for this novel reduction of
R-acetoxypropiophenone (1b) with the tridecylmanganese ate
complex is shown in Scheme 2, which is consistent with our
observations, although further investigations are needed for the
elucidation of the precise mechanism.12
The reductant Bu3MnLi was also applicable to the generation
of the aliphatic ketone enolates, although in these cases the oxy
leaving groups were not appropriate because of their low
reactivities, and the halo ketones were used13,14 (eq 1).
As shown above, the manganese(II) ate complexes cleanly
react with carbonyl compounds bearing a leaving group at the
R-position to generate the enolates of the original ketones, esters,
and amides, irrespective of their structures. In these reactions,
the organomanganates serve as a reductant, not an alkylation
agent. This direct generation of the enolates can be efficiently
applied to the regiospecific generation and reactions of a variety
of the enolates of the unsymmetrical and nonbranched ketones.
Further advantages of this new species in organic synthesis
are exhibited by the following reactions. Alkylations using alkyl
halides were attained.6a In these reactions, no polyalkylated
products were found6,15 (eq 2). As shown in eqs 3 and 4, the
enolate derived from 1c reacted chemoselectively with a halo
Acknowledgment. The present work was partly supported by
Grants-in-Aid for Scientific Research on Priority Areas from the
Ministry of Education, Japan and Pfizer Pharmaceuticals Inc. H.H.
acknowledges support from the Research Fellowships of the Japan
Society for the Promotion of Science for Young Scientist.
(12) The order of the reductive elimination of the ligands and the
â-elimination of an acetoxy group is not clear. A mechanism in which a
leaving group is directly substituted by a manganese reagent to generate
the formally similar intermediate to those in substitution reactions of organic
halides with cuprates may be also possible. As pointed out by a referee, an
electron transfer process cannot be ruled out without further study.
(13) From 4-(tert-butyldimethylsiloxy)-5-nonanone (reduced at rt for 2
h) and 4-chloro-5-nonanone (reduced at -20 °C for 1 h), the yields for the
aldol adducts with benzaldehyde (at -78 °C for 1 h) were 38% and 65%,
respectively.
Supporting Information Available: Experimental procedures and
spectral data for the products (15 pages). See any current masthead
page for ordering and Internet access instructions.
JA964054D
(14) When Dec3MnLi was used in the reaction giving 2i, 82% of 2i (syn/
anti ) 70:30) was obtained. In this crude mixture, 24% of 1-decene, 29%
of eicosane, and 27% of 1-bromodecane were found, suggesting the
competition of a metal-halogen exchange (ca. 27%) with the processes
shown in Scheme 2.
(15) Reetz, M. T.; Haning, H. Tetrahedron Lett. 1993, 34, 7395-7398.
(16) For the preparation of 1g-i, see: Supporting Information.
(17) Aldol reactions for the enolates of the unsymmetrical and non-
branched aliphatic ketones generated in a regiospecific manner were also
successful.