Expanding the scope of Lewis acid catalysis in water: Remarkable ligand acceleration of aqueous Ytterbium triflate catalyzed Michael addition reactions

Article abstract of DOI:10.1021/jo051540n

Significant rate acceleration of metal-catalyzed Michael addition reactions in water was observed upon addition of small, dibasic ligands. Ytterbium triflate and TMEDA was the most effective combination leading to a nearly 20-fold faster reaction than in the absence of ligand.

Full text of DOI:10.1021/jo051540n

full scope and potential of Lewis acid catalysis in water is still
unclear. A significant step forward would be the realization of
efficient ligand accelerated catalysis in water.4 When a basic
ligand is added to a catalytic reaction and bound to the metal
the rate of the reaction can be unaffected, accelerated, or
decelerated. For the majority of catalytic processes, no rate
enhancement is observed. However, when a positive ligand
effect on reaction rate can be found, it offers a powerful
approach to increasing reaction efficiency.5
Expanding the Scope of Lewis Acid Catalysis in
Water: Remarkable Ligand Acceleration of
Aqueous Ytterbium Triflate Catalyzed Michael
Addition Reactions
Rui Ding, Kambiz Katebzadeh, Lisa Roman,
Karl-Erik Bergquist, and Ulf M. Lindstro¨m*
Department of Organic Chemistry, Lund UniVersity,
With the goal of developing generally useful and more
efficient Lewis acid catalysts for synthesis in water, we set out
to study the influence of various ligands on the rate of aqueous
Lewis acid catalyzed Michael additions, one of the most
important carbon-carbon bond-forming reactions in organic
synthesis. However, progress toward efficient Lewis acid
catalyzed Michael additions in water has been slow, and only
a few examples exist where carbon nucleophiles are involved.
Keller and Feringa have used Yb(OTf)3 (0.1 equiv) to catalyze
the conjugate addition of â-keto- and R-nitroesters to enones.6
The reactions required 3 equiv of enone and took 3-5 days to
reach completion. Kobayashi et al. used a Lewis acid-surfactant
combined catalyst (LASC) to afford Michael adducts in good
yields.7 The reactions were slow (12-35 h), however, and with
less than 3 equiv of acceptor the reactions became sluggish.
Later, the same research group reported the use of the chiral
catalyst (R)-Tol-BINAP/AgOTf affording the Michael adducts
in good yields and ee’s.8 Nevertheless, the reactions were still
run for 18-36 h using the acceptor in large excess. We now
wish to report remarkable ligand acceleration effects that lead
to much more efficient Lewis acid catalyzed Michael additions
in water, a discovery that is of importance to the overall
development of organic synthesis in aqueous media.
P.O. Box 124, SE-221 00 Lund, Sweden
ulf.lindstrom@organic.lu.se
ReceiVed July 24, 2005
Significant rate acceleration of metal-catalyzed Michael
addition reactions in water was observed upon addition of
small, dibasic ligands. Ytterbium triflate and TMEDA was
the most effective combination leading to a nearly 20-fold
faster reaction than in the absence of ligand.
Reducing the use of hazardous solvents is one of the most
important challenges in the effort to minimize pollution and
risks associated with the production of chemicals. Accordingly,
the development of water as a harmless alternative to organic
solvents as reaction medium for organic synthesis has become
an important research area.1 Other than the environmental and
economical benefits of using water, it has also been recognized
that water may sometimes even be the preferred solvent in terms
of synthetic efficiency. For example, using water as reaction
solvent can lead to simplified work-up procedures, and the use
of protecting groups to replace acidic hydrogens may be reduced
or avoided completely.2 Metal salts with Lewis acid activity in
water have found extensive use in aqueous catalysis, in
particular, the rare earth metals including the lanthanides (Sc(III),
Y(III), Ln(III)), but also some main group elements (In(III),
Pb(II)) and transition metals (Fe(II), Cu(II), Ag(I), Zn(II),
Cd(II)). Their catalytic capacity in water, as demonstrated in
numerous types of reactions, is ascribed to these metals having
a hydrolysis constant within an optimal range and a fast
exchange rate of coordinated water ligands.3 Nevertheless, the
As a starting point for our investigation we screened various
metal triflates for their ability to catalyze the Michael addition
reaction between ethyl acetoacetate, 1, and methyl vinyl ketone
(MVK) to give the adduct 2 (Table 1).
The reactions were run with 10% of a metal triflate for 16 h
at room temperature, after which time the ratio of adduct 2 to
starting â-ketoester 1 was measured by integrating relevant
peaks in the NMR spectra of the crude products. Rate constants
could then be derived from the second-order rate equation (see
the Supporting Information). In the absence of metal, no trace
of addition product 2 was observed after 16 h (entry 1). Among
the metal triflates tested, all were able to catalyze the reaction
to moderate extent with 10-26% yield (based on NMR) of 1,
which corresponds to second-order rate constants between 2.2
× 10-6 and 6.2 × 10-6 M-1 s-1 (entries 2-5), except In(OTf)3,
the presence of which did not yield any detectable amount of 2
(entry 6). The most effective metal salt was La(OTf)3 (entry
5), which proceeded with approximately three times higher rate
* To whom correspondence should be addressed. Fax: (+46) 46 2228209.
(1) (a) Lindstro¨m, U. M. Chem. ReV. 2002, 102, 2751. (b) Organic
Synthesis in Water; Grieco, P., Ed.; Blackie Academic & Professional:
London, 1998. (c) Li, C. J.; Chan, T. H. Organic Reactions in Aqueous
Media; Wiley: New York, 1997. (d) See also thematic issue: “Organic
Reactions in Water”: AdV. Synth. Catal. 2002, 344 (3-4).
(2) For a recent example, see: Lindstro¨m, U. M.; Ding, R.; Hidestål, O.
Chem. Commun. 2005, 1773.
(4) For examples of modest ligand effects in aqueous media, see: (a)
Kobayashi, S.; Aoyama, N.; Manabe, K. Chirality 2003, 15, 124. (b)
Kobayashi, S.; Aoyama, N.; Manabe, K. Synlett 2002, 483.
(5) Berrisford, D. J.; Bolm, C.; Sharpless, K. B. Angew. Chem., Int. Ed.
1995, 34, 1059.
(6) (a) Keller, E.; Feringa, B. L. Synlett 1997, 842. (b) Keller, E.; Feringa,
B. L. Tetrahedron Lett. 1996, 37, 1879.
(7) Mori, Y.; Kakumoto, K.; Manabe, K.; Kobayashi, S. Tetrahedron
Lett. 2000, 41, 3107.
(8) Kobayashi, S.; Kakumoto, K.; Mori, Y.; Manabe, K. Isr. J. Chem.
2001, 41, 247.
(3) (a) Kobayashi, S.; Manabe, K. Acc. Chem. Res. 2002, 35, 209. (b)
Kobayashi, S.; Nagayama, S.; Busujima, T. J. Am. Chem. Soc. 1998, 120,
8287.
10.1021/jo051540n CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/06/2005
352
J. Org. Chem. 2006, 71, 352-355

TABLE 1. Observed Yields and Rates for Lewis Acid Catalyzed
Michael Additions in Water in the Absence (k-L) and Presence (k+L
of TMEDAa
)
no TMEDA
TMEDA
k+L
(M-1 s-1
yieldb
k-L
yieldb
(%)
c
c
d
entry
catalyst
(%)
(M-1 s-1
)
)
krel
1
2
3
4
5
6
7
<1
12
10
12
26
<1
<1
<2 × 10-7
2.2 × 10-6
1.9 × 10-6
2.2 × 10-6
6.2 × 10-6
<2 × 10-7
<2 × 10-7
<1
71
26
23
38
<1
<2 × 10-7
Yb(OTf)3
Cu(OTf)2
Sc(OTf)3
La(OTf)3
In(OTf)3
TfOHe
4.3 × 10-5 19.5
6.3 × 10-6
5.2 × 10-6
1.0 × 10-5
<2 × 10-7
3.3
2.4
1.6
a All reactions were performed at least three times, and listed values are
averages. b Yields are based on NMR; products not isolated. NMR detection
limit was estimated to 1% conversion, which corresponds to a rate constant
FIGURE 1. Effect of TMEDA on the reaction rate of Yb(OTf)3-
catalyzed conjugate addition of ethyl acetoacetate to MVK.
of 2 × 10-7
.
c For derivation of rate constants, see the Supporting
Information. d Ligand effect, krel ) k+L/k-L e No metal; only 5% of
.
trifluoromethanesulfonic acid was used.
TfOH (5%). Under these conditions no addition product was
observed after 16 h (entry 7). Finally, we also performed an in
situ NMR analysis that showed the presence of a complex of
ytterbium and TMEDA in water and allowed the estimation of
an equilibrium constant (see the Supporting Information).
Having established a plausible mechanistic origin of the
observed rate accelerations, we went on to further investigate
these ligand effects by screening a number of 1,2-dibasic
compounds for their ability to accelerate the Yb(OTf)3-catalyzed
Michael addition reaction between 1 and MVK. The results are
listed in Table 2.
As can be seen, all diamines and amino alcohols provided
significant rate acceleration (entries 2-12), although we were
not able to find a better ligand than TMEDA among these. The
second best ligand was tris(hydroxyethyl)aminomethane (entry
12), with a relative rate constant of 7.3, which shows that the
observed effect is not limited to 1,2-dibasic compounds and
suggests the possibility that multidentate ligands may be used
with advantage. 1,2-Diols seem to be poor ligands in this
reaction. In fact, catechol had a clearly decelerating effect on
the reaction (entry 13), while 1,2-ethanediol had almost no effect
at all (entry 14).
than either Yb(OTf)3, Cu(OTf)2, or Sc(OTf)3 (entries 2-4).
Next, to monitor the effect that a 1,2-dibasic ligand might have
on the reaction rate, we performed the Lewis acid catalyzed
Michael additions in the presence of 12% of N,N,N′,N′-
tetramethylethylenediamine, TMEDA (Table 1). Gratifyingly,
we found that the catalytic activities of several of the metal
triflates were significantly enhanced upon addition of TMEDA.
Adding equimolar amounts of 1 and MVK to a mixture of Yb-
(OTf)3 (10%) and TMEDA (12%) gave a significantly higher
second-order rate constant of 4.3 × 10-5 M-1 s-1 (entry 2).9
Thus, the ligand acceleration effect obtained from the ratio of
the rate constants is 19.5! This remarkable rate difference offers
great synthetic advantages as illustrated in Figure 1 (see also
Figure S1 in the Supporting Information). In the presence of
TMEDA, the Michael adduct 2 is obtained in 71% yield (based
on NMR) after 16 h as compared to only 12% for the ligand-
free reaction.
To the best of our knowledge, this is the most significant
ligand acceleration effect observed in water to date. The catalytic
activities of Cu(OTf)2, Sc(OTf)3, and La(OTf)3 were also
enhanced by TMEDA, but to a significantly lower extent giving
relative rate constants of 3.3, 2.4, and 1.6, respectively (entries
3-5). Again, no product was detected in the reaction with In-
(OTf)3 (entry 6). In view of the unexpectedly large rate
accelerations, and because positive ligand effects in water are
relatively unexplored, we needed to convince ourselves that what
we observe is true ligand accelerated metal catalysis, i.e., the
active catalyst is the metal-ligand complex and not some other
species formed in the reaction mixture. First, a control experi-
ment confirmed that TMEDA alone does not catalyze the
reaction (entry 1). Furthermore, to ensure that the rate accelera-
tion was not due to trace amounts of trifluoromethanesulfonic
acid (TfOH) being formed upon complexation of the metal
triflate we also made a control experiment in the presence of
Finally, we were able to demonstrate that other 1,3-dicarbonyl
compounds also act as efficient donors in the presence of Yb-
(OTf)3/TMEDA (Table 3).
Ethyl cyclohexanone-2-carboxylate reacted efficiently with
only 1 equiv of MVK to give a quantitative yield of the expected
adduct after 16 h (entry 1). We also performed this reaction
using 3 equiv of MVK in order to make a direct comparison
with previously described methods. Using Yb(OTf)3 as catalyst
and an excess of MVK, this reaction was reported to require 5
days to go to completion.6 In the presence of TMEDA, the
reaction is complete in only 6 h (entry 2)! Ethyl cyclopentanone-
2-carboxylate worked nearly as efficiently providing the ex-
pected adduct in 95% yield (entry 3). Ethyl acetoacetate is
slightly less reactive toward MVK but provides the addition
product in 88% yield (entry 4) after 24 h. The diketone 2,4-
pentanedione added to MVK in 92% yield (entry 5). 3-Methyl-
2,4-pentanedione appeared to proceed equally well, although
(9) For another example of ligand accelerated Yb(OTf)3 catalysis, see:
Aggarwal, V. K.; Mereu, A.; Tarver, G. J.; McCague, R. J. Org. Chem.
1998, 63, 7183.
J. Org. Chem, Vol. 71, No. 1, 2006 353

TABLE 2. Effects of Various Ligands on the Yb(OTf)3-Catalyzed
Michael Addition between Ethyl Acetoacetate and MVK in Watera
TABLE 3. Michael Additions to MVK in Water Catalyzed by
Yb(OTf)3/TMEDAa
a For reaction conditions, see Table 1. b Isolated yields. c Three equiva-
lents of MVK. d Performed at 45 °C. e Some impurities (<10%) remained
even after repeated attempts to purify the product.
ated and the basic catalytic processes for several of the metals
studied provide excellent opportunity for chiral catalysis. It is
reasonable to assume that our findings on ligand accelerated
catalysis reported herein can be used to improve also other Lewis
acid catalyzed reactions in water and become a valuable tool
in the development of asymmetric catalysis in water.
Experimental Section
Determination of Rate Constants. All reactions were performed
at 25 ( 0.5 °C. Ligand (0.06 mmol, 0.12 equiv) and ytterbium
triflate (0.05 mmol, 0.10 equiv) were mixed by stirring for 15 min.
Water (0.5 mL) was added and the mixture stirred for another 15
min. Ethyl acetoacetate (0.5 mmol, 1 equiv) and MVK (0.5 mmol,
1 equiv) were then added. After being stirred vigorously for 16 h,
the reactions were stopped by dilution with water and extraction
with EtOAc (5 mL). Concentration afforded crude mixtures that
were analyzed by 1H NMR (acetone-d6) to give the ratios of ethyl
acetoacetate 1 to adduct 2. All experiments were performed at least
three times, and listed values are averages. The reaction rates were
calculated using the formula k′ ) [2]16h/([1]16ht), which is derived
from the second-order rate equation (see the Supporting Informa-
tion). When plotting 1/[1] against time, a straight line typical for a
second-order reaction is obtained (see the Supporting Information).
The data for the ligand-free reactions were obtained in an identical
a For reaction conditions, see Table 1. All reactions were performed at
least three times. Listed rate constants are averages. b For derivation of rate
constants, see the Supporting Information. c Relative to the ligand-free
reaction (entry 1), krel ) k+L/k-L
.
problems with purification resulted in a slightly lower yield
(84%) of a product containing 5-10% of unknown impurity
(entry 6). To clearly demonstrate the potential of this method
for preparative purposes, we performed the reaction between
ethyl acetoacetate and MVK on a gram scale. With gentle
heating (45 °C), the adduct was obtained in 85% isolated yield
after only 2.5 h reaction time. Unfortunately, some other
potential Michael acceptors such as acrylonitrile or ethyl acrylate
were unreactive to the conditions described.
way. Relative rates were then obtained using the formula krel
k′+L/k′-L
)
.
Typical Procedure for Yb(OTf)3/TMEDA-Catalyzed Michael
Addition. TMEDA (0.014 g, 0.12 mmol) and ytterbium triflate
(0.062 g, 0.10 mmol) were stirred together for 15 min at ambient
temperature, after which water (1 mL) was added and stirring
continued for another 15 min. Ethyl cyclohexanone-2-carboxylate
(0.170 g, 1.0 mmol) and 3-buten-2-one (0.070 g, 1.0 mmol) were
then added. After being stirred for 16 h, the reaction was diluted
with water and extracted with ethyl acetate (10 mL). The organic
phase was dried over MgSO4, filtered, and concentrated. The crude
product was purified by flash chromatography on a silica gel column
pretreated with triethylamine, eluting with heptane/ethyl acetate 1:1,
In summary, we have developed an efficient, catalytic method
of achieving carbon-carbon bond-forming Michael addition
reactions in water. This was made possible by the pivotal find
that small, dibasic ligands have remarkable, positive effects on
the catalytic activity of several metal triflates in water, ytterbium
triflate, in particular, which in combination with TMEDA gave
an almost 20 times faster reaction compared to the ligand-free
reaction. Other than providing a method that is relatively fast
and high-yielding, even with equimolar amounts of donor and
acceptor, the large rate differences between the ligand acceler-
354 J. Org. Chem., Vol. 71, No. 1, 2006

which afforded the desired Michael adduct (0.239 g, 100%). The
analytical data of the product were identical to those reported in
the literature.10
Supporting Information Available: Experimental procedure
optimized for preparative purposes. Sample of NMR used for
determination of rate constants. Derivation of second-order rate
equation. Graph showing fit of experimental data to the rate
equation. Details of in situ NMR analysis of complexation. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Acknowledgment. This work was supported by the Swedish
Research Council and the Crafoord Foundation. R.D. gratefully
acknowledges a post-doctoral stipend from the Carl Trygger
Foundation.
(10) Christoffers, J. J. Chem. Soc., Perkin Trans. 1 1997, 3141.
JO051540N
J. Org. Chem, Vol. 71, No. 1, 2006 355