Strikingly simple direct α-allylation of aldehydes with allyl alcohols: Remarkable advance in the Tsuji-Trost reaction [8]

Full text of DOI:10.1021/ja011656a

J. Am. Chem. Soc. 2001, 123, 10401-10402
10401
Scheme 1. Pd-Catalyzed Amphiphilic Allylation of a
Secondary Aldehyde with an Allyl Alcohol: Effects of
Additives (1a ) 1.1 mmol, 2b ) 1.0 mmol)
Strikingly Simple Direct r-Allylation of Aldehydes
with Allyl Alcohols: Remarkable Advance in the
Tsuji-Trost Reaction
Masanari Kimura, Yoshikazu Horino, Ryutaro Mukai,
Shuji Tanaka, and Yoshinao Tamaru*
Department of Applied Chemistry
Faculty of Engineering, Nagasaki UniVersity
Bunkyo-machi, Nagasaki 852-8521, Japan
ReceiVed July 9, 2001
Palladium-catalyzed allylic alkylation of active methylene
compounds such as â-ketoesters and malonates (the Tsuji-Trost
reaction) is a reliable and widely used method for C-C bond
formation.1 However, the R-allylic alkylation of nonstabilized
ketones and aldehydes remains to be developed. The alkylation
reported so far requires both reaction partners to be preactivated:
allyl alcohols as their esters and halides and ketones and aldehydes
as their metal enolates,2 enol silyl ethers,3 or enamines.4 In this
context, for the alkylation of ketones, a traditional stoichiometric
alkylation via their metal enolates might still be regarded as an
alternative to be compared favorably.5,6 On the other hand, the
stoichiometric alkylation of aldehydes via their metal enolates
has numerous drawbacks, because of their tendency to undergo
aldol condensation and the Cannizaro and Tishchenko reactions.7
Here we disclose that the R-alkylation of aldehydes can be
readily achieved by direct use of aldehydes and allyl alcohols8
under catalytic conditions with respect to palladium. In most cases,
the reaction is complete within 1 day at room temperature to ∼50
°C and provides the R-allyl aldehydes derived from a variety of
alkyl aldehydes and allyl alcohols in good to excellent yields.
As is shown in run 1 in Scheme 1, under the reaction conditions
developed recently for the R-allylation of active methylene
compounds9 and o-hydroxyacetophenones,10 1a and 2b reacted
to provide a mixture of the expected product 3b along with a
nucleophilic allylation product 4b and a biallyl 5 in comparable
amounts. To improve the yield in favor of 3b, the bases (runs 2
and 3) and chloride ion sources (run 4)11 were examined as
additives, the former to increase the enol content of aldehyde and
the latter to suppress nucleophilic allylation.12b As was expected,
the bases turned out to be very effective at increasing the yield
of 3b and completely suppressed the formation of 5, but 4b was
formed still in considerable amounts (runs 2 and 3). Surprisingly,
however, LiCl11,13 completely inhibited the reaction, and no 3b,
4b, and 5 was produced at all (run 4). In sharp contrast to these
results, to our pleasant surprise, a combination of Et3N and LiCl
turned out to be most satisfactory and furnished 3b selectively in
excellent yield (run 5).
Under the conditions thus established, the R-allylation of a
variety of combinations of secondary aldehydes and allyl alcohols
was examined (runs 1-8, Table 1). As is evident from these
results, the reaction shows quite high generality and provides a
wide structural variety of 3. Only in a limited number of cases
(run 5, Scheme 1 and runs 1 and 2, Table 1) was the reaction
accompanied with nucleophilic allylation, providing 4a-c as
minor products.12 Usually, 10 mol % of Pd(OAc)2 and 20 mol %
of PPh3 were employed (conditions A); however, the results
shown in runs 5-7 indicate that 5 mol % of Pd(OAc)2-10 mol
% of PPh3 might be sufficient to complete the reaction within a
reasonable reaction time at 50 °C (conditions B).14
The reaction feature of primary aldehydes turned out to be quite
different from that of secondary aldehydes and selectively
provided R-allylation products 7 of the self-aldols (runs 9 and
10, Table 1).15 The quantitative conversion of 1-cyclohexenecar-
boxaldehyde (1e) into its R-alkylated products, 7c and 7d, further
supports the above-mentioned sequence (runs 11 and 12).
It may be pertinent to look at the present unique direct R-allylic
alkylation of aliphatic aldehydes with allyl alcohols, especially
focusing on the dramatic effects of the chloride ion and Et3N
additives (Schemes 1 and 2). Mechanistic studies have clarified
that Pd(OAc)2 is reduced to Pd(0) by trialkyl- and triarylphos-
phines, where 1 equiv of phosphine is oxidized to the corre-
(1) (a) Tsuji, J. Transition Metal Reagents and Catalysts; Wiley: Chichester,
2000. (b) Trost, B. M.; Lee, C. In Catalytic Asymmetric Synthesis; Ojima, I.,
Ed.; Wiley-VCH: New York, 2000; Chapter 8E.
(2) (a) You, S.-L.; Hou, X.-L.; Dai, L.-X.; Zhu, Z.-Z. Org. Lett. 2001, 3,
149-151. (b) Braun, M.; Laicher, F.; Meier, T. Angew. Chem., Int. Ed. 2000,
39, 3494-3497. (c) Trost, B. M.; Schroeder, G. M. J. Am. Chem. Soc. 1999,
121, 6759-6760. (d) Luo, F.-T.; Negishi, E. Tetrahedron Lett. 1985, 26,
2177-2180 and references therein.
(3) (a) Tsuji, J.; Minami, I.; Shimizu, I. Chem. Lett. 1983, 1325-1326.
(b) Trost, B. M.; Keinan, E. Tetrahedron Lett. 1980, 21, 2591-2594.
(4) (a) Hiroi, K.; Abe, J.; Suya, K.; Sato, S.; Koyama, T. J. Org. Chem.
1994, 59, 203-213 and references therein. (b) Huang, Y.; Lu, X. Tetrahedron
Lett. 1988, 29, 5663-5664. (c) Murahashi, S.; Makabe, Y.; Kurita, K. J. Org.
Chem. 1988, 53, 4489-4495.
(5) Ruthenium-catalyzed direct R-propargylation of ketones with propargyl
alcohols: Nishibayashi, Y.; Wakiji, I.; Ishii, Y.; Uemura, S.; Hidai, M. J.
Am. Chem. Soc. 2001, 123, 3393-3394.
(6) Palladium-catalyzed R-arylation (not allylation) of ketones and alde-
hydes has been developed rapidly in recent years: (a) Fox, J. M.; Huang, X.;
Chieffi, A.; Buchwald, S. L. J. Am. Chem. Soc. 2000, 122, 1360-1370 and
references therein. (b) Satoh, T.; Kametani, Y.; Terao, Y.; Miura, M.; Nomura,
M. Tetrahedron Lett. 1999, 40, 5345-5348. (c) Kawatsura, M.; Hartwig, J.
F. J. Am. Chem. Soc. 1999, 121, 1473-1478. (d) Terao, Y.; Satoh, T.; Miura,
M.; Nomura, M. Tetrahedron Lett. 1998, 39, 6203-6206. (e) Muratake, H.;
Hayakawa, A.; Natsume, M. Tetrahedron Lett. 1997, 38, 7577-7580; 7581-
7582.
(7) (a) Smith, M. B.; March, J. AdVanced Organic Chemistry; Wiley: New
York, 2001; Chapter 10-105. (b) Cane, D. In ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 3,
Chapter 1.1.
(11) Tetrabutylammonium chloride (1.0 mmol, completely soluble in the
reaction mixture) showed almost the same reaction behavior as LiCl.
(12) (a) Kimura, M.; Tomizawa, T.; Horino, Y.; Tanaka, S.; Tamaru, Y.
Tetrahedron Lett. 2000, 41, 3627-3629. (b) Kimura, M.; Kiyama, I.;
Tomizawa, T.; Horino, Y.; Tanaka, S.; Tamaru, Y. Tetrahedron Lett. 1999,
40, 6795-6798. For a discussion of umpolung of π-allylpalladium, see: (c)
Tamaru, Y. J. Organomet. Chem. 1999, 576, 215-231.
(13) A large portion of LiCl remains undissolved in the reaction mixture.
(14) Interestingly, under the same conditions as run 1 in Table 1, both
allyl chloride and acetate were unreactive and provided 3a in much lower
isolated yield (ca. 35%) after 80 h at room temperature.
(8) Allyl alcohols as π-allylpalladium precursors: (a) Araki, S.; Kamei,
T.; Hirashita, T.; Yamamura, H.; Kawai, M. Org. Lett. 2000, 2, 847-849.
(b) Yang, S.-H.; Hung, C.-W. J. Org. Chem. 1999, 64, 5000-5001. (c) Xiao,
W.-J.; Alper, H. J. Org. Chem. 1998, 63, 7939-7944. Also see references
cited in refs 9, 10, and 12b.
(15) For the reaction with relatively unreactive allyl alcohols (e.g., R- and
γ-methylallyl alcohols), a mixture of a self-aldol condensation product 6 and
its R-allylation product 7e was obtained in comparable amounts.
(9) Tamaru, Y.; Horino, Y.; Araki, M.; Tanaka, S.; Kimura, M. Tetrahedron
Lett. 2000, 41, 5705-5709.
(10) Horino, Y.; Naito, M.; Kimura, M.; Tanaka, S.; Tamaru, Y. Tetra-
hedron Lett. 2001, 42, 3113-3116.
10.1021/ja011656a CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/27/2001

10402 J. Am. Chem. Soc., Vol. 123, No. 42, 2001
Communications to the Editor
Table 1. Pd-Catalyzed R-Allylation of Aldehydes with Allyl
Alcohols
observed in runs 4 and 5 (Scheme 1) might be rationalized by
taking the contrasting reactivity of Pd(OAc)2 and PdCl2 into
consideration. In the absence of Et3N, no reaction takes place,
because PdCl2 is tough under the conditions and remains intact
(run 4);18 however, in the presence of Et3N, PdCl2 might be
reduced to a Pd(0) species and the reaction recovers (run 5). In
run 5, an allyl alcohol might take the place of water to produce
a Pd(0) species from PdCl2 and Ph3P. However, note that 1 equiv
of water forms in every one catalytic cycle of all the pathways
a-c (Scheme 2).19
Oxidative addition of an allyl alcohol activated by the
coordination with BEt3 to the Pd(0) species leads to an intermedi-
ate I, which would enjoy three different pathways: path a, an
allyl-ethyl exchange to form allyldiethylborane and EtPdOH,12
path b, nucleophilic attack by an aldehyde enol, and path c, a
counterion exchange to form a π-allylpalladium chloride species
II, which further reacts with the aldehyde enol to form the final
product 3.20
In the absence of LiCl (runs 1-3, Scheme 1), either pathway
a predominates over pathway b and provides a mixture of 4 and
5 as the major products (run 1),12b,c or pathway b predominates
over pathway a in the presence of a base, since the base might
increase an enol content (runs 2 and 3). On the other hand, in the
presence of LiCl and Et3N, all the pathways a-c vie with one
another; consequently, the contribution of pathway a might
become smaller and negligible, resulting in the selective formation
of 3 (run 5).
Although in the mechanism outlined in Scheme 2 the roles of
Et3B were explicitly referred to as a Lewis acid to activate both
an alcohol and an aldehyde and also as an ethyl group source to
generate EtPdOH and allylborane (path a), we believe that Et3B
might play some presently unknown, decisively important role
to promote the present catalytic reaction effectively. It should be
noted also that Et3N apparently reduces its ability as a base, not
only because it forms a Lewis acid-base complex with Et3B that
is present in excess in the reaction mixture,21 but also because it
quenches HCl and HOAc generated upon the reduction of PdCl2
and Pd(OAc)2, respectively; nevertheless, Et3N is absolutely
indispensable to promote the selective R-alkylation of aldehydes
(run 5, Scheme 1).
In conclusion, we have disclosed for the first time the direct
R-allylic alkylation of aldehydes with allyl alcohols, which pro-
ceeds catalytically with respect to palladium (5-10 mol %) in
the presence of stoichiometric amounts of Et3B, Et3N, and LiCl
at room temperature to ∼50 °C. The mildness of the reaction
conditions, wide applicability to a variety of combinations of alde-
hydes and allyl alcohols, the ease with which the reaction can be
performed, and, most importantly, the fact that there is no need
for preactivation of both reaction partners (atom economy)22 all
combine to contribute to greatly advance the Tsuji-Trost reaction.
a Conditions A: 1 (1.1 mmol), 2 (1.0 mmol), Pd(OAc)2 (0.1 mmol),
PPh3 (0.2 mmol), Et3B (2.4 mmol, 1 M hexane), Et3N (1.2 mmol),
LiCl (1.0 mmol) in dry THF (5 mL) under N2. Conditions B: 1 (1.1
mmol), 2 (1.0 mmol), Pd(OAc)2 (0.05 mmol), PPh3 (0.10 mmol), Et3B
(2.4 mmol, 1 M hexane), Et3N (1.2 mmol), LiCl (1.0 mmol) in dry
THF (5 mL) under N2. Conditions C: 1 (2.0 mmol), 2 (1.2 mmol),
Pd(OAc)2 (0.1 mmol), PPh3 (0.2 mmol), Et3B (2.4 mmol, 1 M hexane),
Et3N (1.2 mmol), LiCl (1.0 mmol) in dry THF (5 mL) under N2.
b Diastereomer ratio.
Scheme 2. Plausible Mechanism for Pd-Catalyzed
Amphiphilic Allylation of Aldehyde
Acknowledgment. We thank Mr. Y. Ohhama, NMR Facility, for his
outstanding technical assistance. Financial support by the Ministry of
Education, Culture, Sports, Science, and Technology, Japanese Govern-
ment, is gratefully acknowledged.
Supporting Information Available: Experimental procedures and
characterization data for 3a-i, 4a-c, 6, and 7a-e (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
JA011656A
(18) Amatore, C.; Azzabi, M.; Jutand, A. J. Am. Chem. Soc. 1991, 113,
8375-8384.
sponding phosphine oxide.16 To perform the reduction of PdCl2
to Pd(0), on the other hand, requires water and an amine base in
addition to a phosphorus ligand.17 The contrasting phenomena
(19) See Supporting Information for further discussion.
(20) Another pathway is conceivable that involves Pd-catalyzed Claisen
rearrangement of vinyl allyl ethers that might be formed in situ via Et3B-
catalyzed condensation of allyl alcohols and aldehydes; however, the fact that
cinnamyl alcohol provides selectively R-cinnamyl aldehydes (e.g., 3b,c) clearly
excludes the possibility.
(16) (a) Mandai, T.; Matsumoto, T.; Tsuji, J.; Saito, S. Tetrahedron Lett.
1993, 34, 2513-2516. (b) Amatore, C.; Jutand, A.; M’Barki, M. A.
Organometallics 1992, 11, 3009-3013. (c) Ozawa, F.; Kubo, A.; Hayashi,
T. Chem. Lett. 1992, 2177-2180.
(17) (a) Meier, M.; Basolo, F. Inorg. Synth. 1972, 13, 113-115. (b) Mason,
M. R.; Verkade, J. G. Organometallics 1992, 11, 2212-2220.
(21) Brown, H. C. J. Am. Chem. Soc. 1945, 67, 374-378.
(22) For a general discussion of atom economy in organic synthesis, see:
(a) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259-281. (b) Trost,
B. M. Science 1991, 254, 1471-1477.