A highly active catalyst system for intermolecular hydroacylation

Full text of DOI:10.1002/1521-3773(20000901)39:17<3070::AID-ANIE3070>3.0.CO;2-G

COMMUNICATIONS
À81.035 kcalmolÀ1 and the average energy necessary to break a C H
bond is 80.841 kcalmolÀ1. The extent of covalent versus electrostatic
contributions to hydrogen bonding have been the subject of numerous
discussions in the chemical literature (see, for example, J. J. Dannen-
berg, L. Haskamp, A. Masunov, J. Phys. Chem. A. 1999, 103, 7083, and
references therein) and depend on the particular system studied.
Calculated Mulliken partial charges are ‡0.077 for Ha' and À0.13 for
Se indicating an attractive interaction. The difference between an
À
[3]
À
C H bond activation [Eq. (1)]. Herein we report an
efficient intermolecular hydroacylation for which a catalyst
system was designed to ensure facile formation of the
intermediate aldimine.
À
average bond order of C H bonds (0.904; this value is less than the
À
expected due to the destabilization of C H bond by the oxazoline ring
À
system) and of the C Ha' bond (0.888) directly gives the destabiliza-
À
tion by 0.016 bond orders and therefore the C Ha' bond is destabilized
by 1.293 kcalmolÀ1 (C H ˆ 80.841 kcalmolÀ1).[6] The Se ´´´ Ha' inter-
À
action is equal to the total stabilization of the system
(À0.194 kcalmolÀ1) minus the destabilization of the C ´´´ Ha' bond
(1.293 kcalmolÀ1), that is, 1.487kcalmol À1
.
[8] A. Ghosh, M. Bansal, J. Mol. Biol. 1999, 294, 1149; R. A. Musah,
G. M. Jensen, R. J. Rosenfeld, D. E. McRee, D. B. Goodin, S. W.
Bunte, J. Am. Chem. Soc. 1997, 119, 9083, and references therein.
[9] H. Senn and co-workers have observed evidence indicating this type
of interaction (using 2D 1H-77Se HSQC experiment optimized for
relatively small coupling constants) in a protein that contains a Se-Cys
residue. H. Senn, personal communication.
Recently, we found that the reactivity in the hydroacylation
depicted in Equation (1) improved when benzaldehyde (1a)
contaminated by benzoic acid (7)[4] was used as the sub-
strate.[5] Benzoic acid was assumed to catalyze the condensa-
tion of aldehydes 1 with 4 to generate 5; this may be the rate-
determining step of the reaction (see below). This observation
prompted us to search for a way to facilitate the formation of
5, and we found a remarkable enhancement of the reactivity
occurred when aniline (8), as well as 3, 4, and 7, was used as an
additive.
In our experiment 1a was treated with 1-hexene (2a) at
1308C for 1 h in the presence of 2 mol% of [Rh(PPh3)3Cl](3),
20 mol% of 4, 6 mol% of 7, and 60 mol% of 8 as cocatalysts
to give heptanophenone (6a) in 98% yield after chromato-
graphic separation [Eq. (2)]. A significant decrease in reac-
tivity was observed when the reaction was performed under
[10] L. A. Silks, J. Peng, J. D. Odom, R. B. Dunlap, J. Chem. Soc. Perkins
Trans. 1 1991, 2495.
[11] Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication nos.
CCDC-140580 ± 140582. Copies of the data can be obtained free of
charge on application to CCDC, 12 Union Road, Cambridge
CB21EZ, UK (fax: (‡44)1223-336-033; e-mail: deposit@ccdc.cam.
ac.uk).
[12] A. L. Davis, J. Keeler, E. D. Laue, D. Moskau, J. Magn. Reson. 1992,
98, 207.
A Highly Active Catalyst System for
Intermolecular Hydroacylation**
Chul-Ho Jun,* Dae-Yon Lee, Hyuk Lee, and
Jun-Bae Hong
Hydroacylation[1±3] is a useful synthetic method for obtain-
À
ing ketones from aldehydes and olefins by using C H bond
activation by transition metal complexes. Although the intra-
molecular hydroacylation of 4-pentenals has been extensively
studied,[1] only a few successful intermolecular reactions have
been reported.[2a±f] To effect intermolecular hydroacylation,
ethylene,[2a,b] carbon monoxide,[2c] or vinylsilanes with a CoI
catalyst[2d,e] have been used to suppress the decarbonylation
that results in catalytically inactive metal carbonyl species. We
have developed a general intermolecular hydroacylation of
1-alkenes by using an RhI complex and 2-amino-3-picoline as
cocatalyst, whereby aldimine 5 is assumed to be a key
intermediate that suppresses decarbonylation and permits
the same conditions but without 7 and 8 (Figure 1). For
example, while the reaction was complete (100% GC yield)
after 1 h when both 7 and 8 were added, only a 9% yield of 6a
was obtained when the reaction was performed without
additives. The yield increased to 28% with the addition of 7.
The postulated mechanism is depicted in Scheme 1. Cy-
cle A represents the mechanism for the catalyst system
consisting of 3 and 4. The first step is believed to be the
formation of aldimine 9 from 1a and 4. Aldimine 9 reacts with
2a to yield ketimine 10 by hydroiminoacylation.[6] The
resulting ketimine 10 is hydrolyzed by H2O, generated from
[*] Prof. Dr. C.-H. Jun, D.-Y. Lee, H. Lee, J.-B. Hong
Department of Chemistry
Yonsei University
Seoul, 120-749 (Korea)
Fax : (‡82)2-364-7050
E-mail: junch@alchemy.yonsei.ac.kr
[**] This work was supported by the Brain Korea 21 Project.
3070
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3917-3070 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 17

COMMUNICATIONS
Figure 1. Conversion (yield of 6a) versus time plot for the hydroacylation
of 2a with 1a in the of presence 2 mol% of 3, and 20 mol% of 4 in toluene
&
*
at 1308C. Results for the reaction with 7 and 8 ( ), the reaction with 7 ( ),
~
and the reaction without any additives ( ) are shown (y ˆ yield of 6a
Scheme 2. The reactions of 2a with aldimines 11 and 14 and aldehydes 1a
and 1b in the presence of 3, 4, and 7.
determined by GC).
mechanism of cycle A in Scheme 1.[12] These results imply that
the transimination of 11 with 4 is more facile than the direct
condensation of 1a and 4 as a route to 9.
To examine the effect of carboxylic acid on transimination
an alkylation of 11 with 2a was carried out in the presence of a
cocatalyst system comprising 3 and 4 [Eq. (3)]; the results are
shown in Figure 2. With the addition of 7, the reactivity
increased and the reaction was complete in 3 h, while it took
more than 18 h to obtain a 90% yield of 12 in the absence of 7.
This result strongly suggests that transimination is catalyzed
by carboxylic acid.
Scheme 1. The postulated mechanism for the hydroacylation of 2a with
1a.
the reaction of 1a and 4, to produce 6a with the regeneration
of 4. The role of benzoic acid might be to catalyze the
condensation of 4 and 1a to generate intermediate 9, which
resulted in an increased reactivity.[7]
Cycle B depicts the reaction mechanism with a cocatalyst
system consisting of 8, 3, 4, and 7, in which imine 11 may also
be involved.[8] The initial formation of 11 from 1a and 8 is
followed by transimination[9] of 11 with 4 to give 9. A similar
type of semicarbazone formation from carbonyl compounds
catalyzed by aniline in an acidic medium was previously
reported.[10a] This process, known as nucleophilic catalysis,[10]
consists of the initial formation of the imine compound and
consecutive transimination with semicarbazide to form semi-
carbazone. We also demonstrated the conversion of 11 into
ketimine 12 by transition metal catalyzed alkylation through
transimination with 4.[11] To confirm this mechanism, the
following experiments were carried out.
When 2a was treated with aldimine 11 and anisaldehyde 1b
at 1308C for 5 min in the presence of a catalyst system
consisting of 3, 4, and 7, a mixture of 6a and 13 was isolated in
21 and 5% yield, respectively, after hydrolysis (Scheme 2). In
contrast, the reaction of aldimine 14 and benzaldehyde 1a
with 2a produced a mixture of 6a and 13 in 5 and 26% yield,
respectively. Regardless of the substituents, aldimines that
undergo hydroacylation by transimination (cycle B of
Scheme 1) are more reactive than aldehydes that follow the
Figure 2. The effect of 7 on the alkylation of imine 11 with 2a in the
presence of 2 mol% of 3, amd 20 mol% of 4 [Eq. (3)]. Results for the
&
*
reaction with 7 ( ) and the reaction without 7 ( ) are shown (y ˆ yield of 12
determined by GC).
The reactions of various aldehydes and olefins are sum-
marized in Table 1. All terminal olefins were hydroacylated in
fairly good yields within 1 h. However, trimethylvinylsilane
(2e) was so reactive that the reaction went to completion with
only 1.1 equivalents of 2e (entry 5). The reactions of penta-
fluorostyrene (2 f) and allyl phenyl ether (2g) were complete
after 40 min (entries 6 and 7). However, internal olefins such
as cyclohexene and 2-pentene failed to undergo hydroacyla-
tion. The reaction with aldehydes bearing various substituents
gave moderate yields of the corresponding ketones (en-
tries 8 ± 11). Even aliphatic aldehyde underwent hydroacyla-
tion to give ketones in moderate yield (entry 12).
Angew. Chem. Int. Ed. 2000, 39, No. 17
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3917-3071 $ 17.50+.50/0
3071

COMMUNICATIONS
Table 1. The reaction of various aldehydes and olefins.
152; Angew. Chem. Int. Ed. 1998, 37, 145 ± 147; d) C.-H. Jun, H.-S.
Hong, C.-W. Huh, Tetrahedron Lett. 1999, 40, 8897± 8900; e) C.-H.
Jun, J.-B. Hong, D.-Y. Lee, Synlett 1999, 1 ± 12.
[4] The oxidation of bezaldehyde occurs spontaneously on contact with
Â
air: M. Hudlicky, Oxidations in Organic Chemistry, American
Chemical Society, Washington, DC, 1990, p. 174.
[5] To confirm the effect of carboxylic acid, the hydroacylation of
1-hexene with freshly purified benzaldehyde was performed with
benzoic acid under the reaction conditions depicted in Equation (1).
While it took 24 h to obtain a 72% yield without benzoic acid, the
reaction time was shortened to 6 h with benzoic acid, and a 75% yield
of heptanophenone 6a was obtained.
[6] a) J. W. Suggs, J. Am. Chem. Soc. 1979, 101, 489; b) C.-H. Jun, J.-B.
Kang, J.-Y. Kim, Tetrahedron Lett. 1993, 34, 6431 ± 6434; c) C.-H. Jun,
J.-S. Han, J.-B. Kang, S.-I. Kim, J. Organomet. Chem. 1994, 474, 183 ±
189.
Entry
R1 (1)
R2 (2)[a]
Product
Yield [%][b]
1
2
3
4
5
Ph (1a)
Ph (1a)
Ph (1a)
Ph (1a)
Ph (1a)
Ph (1a)
n-C4H9 (2a)
n-C3H7 (2b)
n-C6H13 (2c)
tBu (2d)
Me3Si (2e)
C6F5 (2 f)
PhOCH2 (2g)
n-C4H9 (2a)
n-C4H9 (2a)
n-C4H9 (2a)
n-C4H9 (2a)
n-C4H9 (2a)
6a
6b
6c
6d
6e
6 f
6g
13
6h
6i
98 (100)
83 (86)
99 (100)
84 (87)
95 (100)[c]
98 (100)[d]
95 (100)[d]
79 (80)
71 (86)
60 (64)
95 (98)
71[e]
6
7Ph (
8
9
10
11
12
1a)
pMeOC6H4 (1b)
pCF3C6H4 (1c)
pMe2NC6H4 (1d)
PhC6H4 (1e)
PhCH2CH2 (1 f)
[7] The reactivity of 2a with 9 towards hydroiminoacylation was not
enhanced by the addition of 7. This implies that carboxylic acid does
not affect the hydroiminoacylation step. Therefore, we assume that
the rate-determining step is the formation of 9.
6j
6k
[8] It is also possible that 1a condenses with 4 to form 9. However, the
condensation of 1a with 8 is more facile than with 4. When an
equimolar mixture of 1a, 4, and 8 was heated at 1308C for 10 min 1a
was completely consumed and the ratio of 9:11 was determined as
10:90 by GC.
[a] Five equivalents based on aldehyde were used. [b] Yield of product
after isolation; GC yields are given in parenthesis. [c] 1.1 equivalents of 2e
was used. [d] Reaction time was 40 min. [e] 10% of the aldol condensation
product of 1 f was obtained.
[9] a) P. Zandbergen, A. M. C. H. van den Nieuwendijk, J. Brussee, A.
van den Gen, C. G. Kruse, Tetrahedron 1992, 48, 3977 ± 3982; b) E.
Hulsbos, J. Marcus, J. Brussee, A. van den Gen, Tetrahedron: Asym-
metry 1997, 8, 1061 ± 1067; c) E. F. J. de Vries, P. Steenwinkel, J.
Brussee, C. G. Kruse, A. van den Gen, J. Org. Chem. 1993, 58, 4315 ±
4325.
[10] a) E. H. Cordes, W. P. Jencks, J. Am. Chem. Soc. 1962, 84, 826 ± 831;
b) T. H. Lowry, K. S. Richardson, Mechanism and Theory in Organic
Chemistry, 2nd ed., Harper & Row, New York, 1981, p. 641; c) J. Hine,
R. C. Dempsey, R. A. Evangelista, E. T. Jarvi, J. M. Wilson, J. Org.
Chem. 1977, 42, 1593 ± 1599.
In summary, we have presented an efficient catalytic system
for intermolecular hydroacylation. Further work is now
directed toward understanding the mechanistic details of this
reaction.
Experimental Section
Typical procedure for preparation of ketone 6a (Table 1, entry 5): A screw-
capped pressure vial (1 mL) was charged with freshly purified benzalde-
hyde (1a, 0.5 mmol), 2-amino-3-picoline (4, 0.1 mmol), benzoic acid (7,
0.03 mmol), aniline (8, 0.3 mmol), 1-hexene (2a, 2.5 mmol), and toluene
(80 mg). After the mixture had been stirred at room temperature for
several minutes, [Rh(PPh3)3Cl] (3, 0.01 mmol) was added, and then it was
stirred at 1308C for 1 h. After cooling the reaction mixture to room
temperature, it was purified by column chromatography (SiO2, n-hexane/
ethyl acetate 4/1) to yield pure 6a (0.49 mmol, 98% yield).
[11] C.-H. Jun, J.-B. Hong, Org. Lett. 1999, 1, 887± 889.
[12] It can be explained by the fact that the protonated aldimine is more
electrophilic than the aldehyde. Since the aldimine is more basic than
the aldehyde, it is the dominant reactant in the presence of acid.
Received: March 7, 2000 [Z14820]
Asymmetric Synthesis of a Chiral Secondary
Grignard Reagent**
[1] a) K. Sakai, J. Ide, O. Oda, N. Nakamura, Tetrahedron Lett. 1972,
1287± 1290; b) R. E. Campbell, C. F. Lochow, K. P. Vora, R. G. Miller,
J. Am. Chem. Soc. 1980, 102, 5824 ± 5830; c) R. C. Larock, K. Oertle,
G. F. Potter, J. Am. Chem. Soc. 1980, 102, 190 ± 197; d) D. Milstein, J.
Chem. Soc. Chem. Commun. 1982, 1357± 1358; e) D. P. Fairlie, B.
Bosnich, Organometallics 1988, 7, 936 ± 945; f) D. P. Fairlie, B. Bosnich,
Organometallics 1988, 7, 946 ± 954; g) R. W. Barnhart, B. Bosnich,
Organometallics 1995, 14, 4343 ± 4348; h) R. W. Barnhart, D. A.
McMorran, B. Bosnich, Chem. Commun. 1997, 589 ± 590; i) B.
Bosnich, Acc. Chem. Res. 1998, 31, 667± 674, and references therein.
[2] a) K. P. Vora, C. F. Lochow, R. G. Miller, J. Organomet. Chem. 1980,
192, 257± 264; b) T. B. Marder, D. C. Roe, D. Milstein, Organo-
metallics 1988, 7, 1451 ± 1453; c) T. Kondo, M. Akazome, Y. Tsuji, Y.
Watanabe, J. Org. Chem. 1990, 55, 1286 ± 1291; d) C. P. Legens, M.
Brookhart, J. Am. Chem. Soc. 1997, 119, 3165 ± 3166; e) C. P. Legens,
P. S. White, M. Brookhart, J. Am. Chem. Soc. 1998, 120, 6965 ± 6979;
f) T. Kondo, N. Hiraishi, Y. Morisaki, K. Wada, Y. Watanabe, T.
Mitsudo, Organometallics 1998, 17, 2131 ± 2134, and references there-
in.
Reinhard W. Hoffmann,* Bettina Hölzer,
Oliver Knopff, and Klaus Harms
Chiral organometallic reagents are of interest in stereo-
selective synthesis. This holds in particular for chiral a-
heterosubstituted organolithium and Grignard reagents.[1]
However, their reactions with electrophiles do not always
take a stereochemically homogenous pathway. It is not clear
[*] Prof. Dr. R. W. Hoffmann, Dipl.-Chem. B. Hölzer,
Dipl.-Chem. O. Knopff, Dr. K. Harms
Fachbereich Chemie
Philipps-Universität Marburg
Hans-Meerwein-Strasse, 35032 Marburg (Germany)
Fax : (‡49)6421-282-8917
E-mail: rwho@chemie.uni-marburg.de
[**] This study was supported by the Deutsche Forschungsgemeinschaft
(SFB 260 and Graduiertenkolleg ªMetallorganische Chemieº) and the
Fonds der Chemischen Industrie.
[3] a) C.-H. Jun, H. Lee, J.-B. Hong, J. Org. Chem. 1997, 62, 1200 ± 1201;
b) C.-H. Jun, D.-Y. Lee, J.-B. Hong, Tetrahedron Lett. 1997, 38, 6673 ±
6676; c) C.-H. Jun, C.-W. Huh, S.-J. Na, Angew. Chem. 1998, 110, 150 ±
3072
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3917-3072 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 17