A ru-catalyzed four-component coupling [6]

Full text of DOI:10.1021/ja001656v

J. Am. Chem. Soc. 2000, 122, 8081-8082
8081
Scheme 1. Mechanistic Proposal for Four-Component
Coupling
A Ru-Catalyzed Four-Component Coupling
Barry M. Trost* and Anthony B. Pinkerton
Department of Chemistry, Stanford UniVersity
Stanford, California 94305-5080
ReceiVed May 15, 2000
The ability to create molecular complexity rapidly provides
for more efficient syntheses of complex molecules. The more
bonds that can be formed in a single step, the fewer the number
of steps that will be required in a synthetic scheme. Reactions
involving the additions of more than two molecules in a single
step are uncommon; those that involve four components are rare.
The most well-known is the Ugi reaction1 which has found
particular utility in combinatorial strategies.2 As part of a program
to develop atom economical reactions,3 we have developed a four-
component coupling according to eq 1.
Scheme 1 outlines the mechanistic proposal. In our studies of
the addition of HX, alkynes, and vinyl ketones catalyzed by a
ruthenium complex, we proposed that the initial adduct 1
undergoes coordination of a vinyl ketone and migratory insertion
to form a ruthenium enolate 2, which upon protonation forms
the adduct 3 and regenerate the initial ruthenium complex to
initiate another cycle (cycle A).4 Could the initial ruthenium
enolate5 3 undergo capture by an electrophile other than a proton?
An aldehyde seemed to be a reasonable alternative since,
increasingly, the ability of organometallic intermediates to undergo
carbonyl additions in the presence of protonic media is being
developed.6
The initial experiment (eq 2) examined the reaction depicted
in eq 1 utilizing the optimized conditions for E-selective
chloroalkylation4a (3 equiv of (CH3)4NCl, 15 mol % of SnCl4‚
5H2O, 10 mol % of 4) in the presence of 3 equiv of p-
methoxybenzaldehyde (3a) Gratifyingly, the four-component
coupling product 5a7 was obtained in 54% yield as an 8.6:1 E:Z
mixture of alkene isomers but each as only a single diastereomer
as determined by NMR spectroscopy (>10:1dr). In addition, a
19% yield of the three-component coupling product 6 was also
obtained. Switching to tetraethylammonium chloride led to a
poorer ratio of 5:6 (2:1). Using anhydrous stannic chloride
inverted the alkene geometry (1:3.6 E:Z) of 5 and reduced the
yield to 39%; significantly, the product of simple protonation 6
still formed in 15% yield. Use of molecular sieves severely
depressed the yield of 5a (to 15%) but still generated 10% of the
product obtained by protonation of 6. Alternative anhydrous
cocatalysts did not improve the reaction. Increasing the amount
of aldehyde 3 to 6 equiv improved the yield of 5a to 62%;
whereas, the yield of 6 dropped to 13%.
Using the original conditions, the aldehyde 3 was varied. An
aliphatic aldehyde, cyclohexanecarboxaldehyde 3b, gave the four-
component product 5b in 51% (E:Z 8:1) also as a single
diastereomer each in addition to 6 (23%). Cinnamaldehyde also
gave only the “expected” product 5c (48% yield, E:Z 6.8:1, dr >
10:1) wherein only MVK served as the Michael acceptor and the
unsaturated aldehyde as the carbonyl partner. The chemoselec-
tivity of this example stems from the steric sensitivity of this
catalyst whereby monosubstituted double bond substrates react
much faster than those bearing disubstitution.8
(1) For a lead reference on the Ugi reaction, see: Ugi, I. J. Prakt. Chem.
1997, 339, 499. Also: Ugi, I.; Lohberger, S.; Karl, R. In ComprehensiVe
Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1999;
Vol. 2, Chapter 4.6.
(2) For some examples of the Ugi reaction and other multicomponent
reactions in combinatorial chemistry, see: Domling, A. Combinatorial Chem.,
High Throughput Screening 1998, 1, 1. Kobayashi, S. Chem. Soc. ReV. 1999,
28, 1.
(3) Trost, B. M. Science 1991, 254, 1471.
Phenyl vinyl ketone (7) reacts equally well (eq 3). With the
cyanoalkyne 8a and p-anisaldehyde 3a, a 58% yield of the four-
component coupling product 9a (E:Z 8:1) as a single diastereomer
in each case in addition to the product of simple protonation 10
(21%) was obtained. Use of the aliphatic aldehydes 3b or 3d with
cyanoalkyne 8a and methoxycarbonylalkyne 8b respectively gave
the desired adducts 9b (E:Z 7.1:1) and 9c (E:Z 12.5:1) as single
diastereomers in 44% and 42% yields respectively with 10a and
10b being isolated in 23% and 28% yields, respectively.
The complementary cis-bromoalkylation4b was also examined
in the four-component coupling as shown in eq 4 and Table 1.
(4) (a) Trost, B. M.; Pinkerton, A. B. J. Am. Chem. Soc. 1999, 121, 1988.
(b) Trost, B. M.; Pinkerton, A. B. Angew. Chem., Int. Ed. 2000, 39, 360.
(5) (a) For some examples of Ru enolates, see: Hartwig, J. F.; Bergman,
R. G.; Anderson, R. A. Organometallics 1991, 10, 3326. Rasley, B. T.; Rapta,
M.; Kulawiec, R. J. Organometallics 1996, 15, 2852. (b) For some examples
of metal-catalyzed addition to Michael acceptors followed by aldol reactions,
see: Taylor, S. J.; Morken, J. P. J. Am. Chem. Soc. 1999, 121, 12202. Kiyooka,
S.; Shimizu, A.; Torri, S. Tetrahedron Lett. 1998, 39, 5237. Matsuda, I.;
Takahashi, K.; Sato, S. Tetrahedron Lett. 1990, 31, 5331. Revis, A.; Hilty, T.
Tetrahedron Lett. 1987, 28, 4809. (c) For an example of a Ru-catalyzed
addition to a Michael acceptor, see: Yi, C. S.; Liu, N. J. Organomet. Chem.
1998, 553, 157.
(6) For a recent example, see: Loh, T. P.; Zhou, J. R. Tetrahedron Lett.
1999, 40, 9115.
(7) All new compounds have been characterized spectroscopically, and
elemental composition has been established by combustion analysis or high-
resolution mass spectroscopy.
(8) Trost, B. M.; Indolese, A.; Mu¨ller, T. J. J.; Treptow, B. J. Am. Chem.
Soc. 1995, 117, 615.
10.1021/ja001656v CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/06/2000

8082 J. Am. Chem. Soc., Vol. 122, No. 33, 2000
Communications to the Editor
Table 1. Four-Component Coupling via cis-Bromoalkylationa
tetraalkylammonium salt 16 in acetone also gave only a single
aldol diastereomer for the initial reaction but in a somewhat lower
E:Z ratio (entry 9).
isolated yields
11
12
E:Z
14
entry
1c 3a n-C6H13
2
3
4
5
6
7
3
(R1)
(R2) 13
14
15
drb
1:1
CH3 LiBr 60% 20%
1:1
3a n-C6H13
3b n-C6H13
3c n-C6H13
3d Ph
3a NC(CH2)3 Ph 16
3b CH3O2C- Ph 16
(CH2)6
CH3 16
CH3 16
CH3 16
CH3 16
52% 20%
55% 18%
63% 15%
40% 11% <2:>98
46% 15%
59% 11%
1:3.7 >10:1d
1:5.7 >10:1d
1:3.4
7:1
9:1
8:1
1:4.4
1:3.1 >10:1d
In all cases, excellent diastereoselectivity for the aldol step was
observed.9,10 The relative stereochemistry of the aldol product as
syn was established in the case of 17. Diastereoselective reduction
with triacetoxyborohydride11 followed by acetonide formation
gave an acetonide 19 consistent with the structure depicted.12
Thus, the syn-aldol geometry implies good selectivity in the
migratory insertion of the vinyl ketone to form a Z-enolate 2013
which undergoes reaction via a typical Zimmerman-Traxler14
transition state 21 to produce the syn-aldol product 22. While
8
3d NC(CH2)3 Ph 16
70% 15%
51% 19%
1:5.5 >10:1d
1:2.0 >10:1d
9e 3a n-C6H13
CH3 16
a All reactions run at 0.5 M in DMF as outlined in eq 4 unless noted
otherwise. b Diastereomeric ratio of aldol adducts which is independent
of alkene geometry. c Reaction performed in acetone. d Only one
diastereomer seen for each alkene isomer by 1H NMR spectroscopy at
500 MHz. e Reaction performed at 0.5 M in acetone.
2
the total suppression of the three-component coupling product,
which derived from a proton serving as the electrophile, remains
a goal, the four-component coupling product can be obtained in
synthetically useful yields (40-70% yields) especially considering
how many bonds are being formed in a single step. The high syn
selectivity in the ruthenium-catalyzed aldol reaction is also
noteworthy.10 The kinetically formed enolate is captured without
loss of regioselectivity. The utility of ruthenium enolates5,13 in
aldol reactions clearly merits further investigations which are
underway in these laboratories.
Utilizing our conditions for the three-component coupling (10%
4, LiBr, at 0.5 M in acetone) but in the presence of an aldehyde
gave a 60% yield of the desired four-component coupling product
but, disappointingly, with no diastereoselectivity (entry 1). The
striking contrast in diastereoselectivity with the chloroalkylation
sequence, which gave only one diastereomer, could arise from
the effect of the change in halide, in cation, or in solvent since
the chloroalkylation reaction was performed in DMF with
tetramethylammonium chloride. Changing the solvent to DMF
led to a dilemma since the three-component bromoalkylation
sequence gave very poor E:Z selectivity in this solvent. We
therefore reexamined the bromoalkylation reaction in DMF.
Changing the bromide source from lithium bromide to tetraethyl-
ammonium bromide improved the E:Z ratio to 1:2. Curiously,
the use of the spirotetraalkylammonium bromide 16 gave the best
E:Z selectivity (1:3) in DMF. As entry 2 illustrates, running the
same four-component coupling in DMF with 16 as the bromide
source gave a 52% yield of 14 as a 1:3.7 E:Z ratio wherein only
a single diastereomer of each was detected by 500 MHz 1H NMR
spectroscopy. Use of an aliphatic aldehyde gave good results also
(entry 3). Like in the trans-chlororuthenation process, the cis-
bromoruthenation version also showed excellent chemoselectivity
in using an R,â-unsaturated aldehyde as the carbonyl component
(entry 4). With phenylacetylene, only the Z alkene isomer was
isolated as a single diastereomer (entry 5). The reaction also gave
somewhat improved E:Z ratios favoring the Z-isomer with
5-hexynonitrile (entries 6 and 8). Alternatively, using the spiro-
Acknowledgment. We thank the National Institutes of Health, General
Medical Sciences, and the National Science Foundation for their generous
support of our work. A.B.P. has received an ACS Organic Division
fellowship sponsored by Bristol Myers Squibb. Mass spectra were
provided by the Mass Spectrometry Facility, University of San Francisco,
supported by the NIH Division of Research Resources.
Supporting Information Available: An Appendix containing ex-
perimental details and results (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
JA001656V
(9) For an overview of the aldol reaction, see: Heathcock, C. H. In
ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Perga-
mon: Oxford, 1991; Vol. 2, Chapters 1.5 and 1.6.
(10) For generation and use of rhodium enolates in aldol reactions, see:
Slough, G. A.; Bergman, R. G.; Heathcock, C. H. J. Am. Chem. Soc. 1989,
111, 938. Reetz, M. T.; Vougioukas, A. E. Tetrahedron Lett. 1987, 28, 793.
Sato, S.; Matsuda, I.; Izumi, Y. Tetrahedron Lett. 1986, 27, 5517. For Pd and
Pt, see: Fujimura, O. J. Am. Chem. Soc. 1998, 120, 10032. Hagiwara, E.;
Fujii, A.; Sodeoka, M. J. Am. Chem. Soc. 1998, 120, 2747. Sodeoka, M.;
Ohrai, K.; Shibasaki, M. J. Org. Chem. 1995, 60, 2648.
(11) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc.
1988, 110, 3560.
(12) See Supporting Information. A similar procedure giving the same
relative stereochemistry was peformed for the product (14) from entry 3, Table
1.
(13) Transmetalation of the Ru enolate to form a tin enolate prior to the
aldol reaction cannot be ruled out.
(14) Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957, 79, 1920.