Phosphine-catalyzed hydration and hydroalkoxylation of activated olefins: Use of a strong nucleophile to generate a strong base

Article abstract of DOI:10.1021/ja035232n

The direct addition of water and a variety of alcohols to activated olefins was observed in the presence of nucleophilic phosphine catalysts. Unlike existing methods, the reactions proceed at room temperature and in the absence of transition metals, or strong acids or bases. The use of simple commercially available catalysts makes this an attractive method for the preparation of β-hydroxy and β-alkoxy substrates, which are prevalent targets and intermediates in organic synthesis. The scope and mechanism of this reaction has been explored, and the compound that acts as the resting state of the catalyst was synthesized independently. Our mechanism also suggests the possibility of extending the scope of this reactivity to other classes of nucleophiles. Copyright

Full text of DOI:10.1021/ja035232n

Published on Web 06/18/2003
Phosphine-Catalyzed Hydration and Hydroalkoxylation of Activated Olefins:
Use of a Strong Nucleophile to Generate a Strong Base
Ian C. Stewart, Robert G. Bergman,* and F. Dean Toste*
Center for New Directions in Organic Synthesis, Department of Chemistry, UniVersity of California,
Berkeley, California 94720
Received March 19, 2003; E-mail: bergman@cchem.berkeley.edu; fdtoste@uclink.berkeley.edu
Table 1. Catalyst Surveya
â-Hydroxy ketones are ubiquitous intermediates and targets in
organic synthesis. They are typically prepared either via aldol
chemistry or sequential epoxidation and reduction of enones.1,2
Direct preparation by addition of water to an enone is an attractive
alternative, but no method exists for this transformation. Similarly,
preparation of â-alkoxy ketones also represents a challenge in
synthetic organic chemistry. Although methods exist for their
synthesis from enones,3-5 direct preparation via alcohol addition
would be ideal, but has not yet been generally realized.6 Trost7
and Inanaga,8 however, have reported catalytic methods for the
addition of alcohols to alkynoates. Herein we report a general
method for the hydration and hydroalkoxylation of enones and other
R,â-unsaturated substrates by use of a trialkylphosphine catalyst.
During investigations into a transition metal-catalyzed hydration
reaction, we discovered that the phosphine ligand alone was actually
a more active catalyst. In the presence of catalytic amounts of
trimethylphosphine, we observed the direct addition of water and
alcohols to R,â-unsaturated compounds in quantitative conversion.
A number of potential Lewis basic catalysts were tested using the
hydromethoxylation of 1 as a benchmark (Table 1). Trialkylphos-
a Reaction conditions: 1 mmol 4-hexen-3-one, 0.05 mmol catalyst in
0.3 mL CD3OD. b Determined by 1H NMR spectroscopy. c Me-BPE )
1,2-bis(2,5-dimethylphospholano)ethane. d BINAP ) 2,2′-Bis(diphenylphos-
phino)-1,1′-binaphthyl. e Et-DuPhos ) 1,2-Bis(2,5-diethylphospholano)ben-
zene. f 10 mol % catalyst. g DABCO ) bicyclo[2.2.2]-1,4-diazaoctane.
phines (Table 1, entries 1-3) were more reactive than aryl
phosphines (Table 1, entries 4-6) and nitrogen-based nucleophiles
(Table 1, entries 7-8).
A variety of R,â-unsaturated substrates were subjected to
hydromethoxylation conditions.9 Enones, R,â-unsaturated esters, and
R,â-unsaturated nitriles all provided the corresponding hydromethox-
ylated products in good isolated yields. Only the highly conjugated
4-phenyl-3-buten-2-one was unreactive, even at increased temper-
atures and reaction times (Table 2, entry 7).10 Cyclic enones were
found to undergo competitive dimerization reactions in both the
hydration and hydromethoxylation conditions.
Using 1 and methyl vinyl ketone as our standard enones, the
scope of reactive hydroxyl compounds was shown to include water
and primary, secondary, and aryl alcohols. tert-Butyl alcohol did
not undergo addition to either 1 or methyl vinyl ketone; instead,
enone dimerization was observed.11 Surprisingly, no retro-aldol or
ketalization products were ever observed in these reactions. In
contrast, the rates of retro-aldol reactions are competitive with
hydration in analogous base-catalyzed systems.12
Table 2. Phosphine-Catalyzed Hydration and Hydroalkoxylationa
1
entry
EWG
R
ROH
H2O
time (h)
% yieldb
1
COEt
Me
20
24
16
77
85
63
2
MeOH
3c
4
COMe
H
MeOH
Me2CHOH
PhOH
1
1
16
72
56
83
59
0
5c
6c
7
Ph
Me
H
MeOH
8
9
CO2Me
CN
MeOH
MeOH
36
4
71
79
On the basis of experiments described below, we propose the
catalytic cycle shown in Scheme 1.13,14 Initial conjugate addition
by the phosphine to the unsaturated carbonyl compound results in
formation of phosphonium enolate 3, which deprotonates an
equivalent of alcohol. The alkoxide of ion pair 4 then undergoes a
subsequent conjugate addition, generating enolate ion pair 5.15
Protonation of the enolate partner in 5 results in formation of the
â-alkoxy carbonyl and regeneration of 4 to complete the cycle.
An alternative mechanism in which trimethylphosphine (pKa )
9)16 functions simply as a base to deprotonate the alcohol directly
is also possible. However, when catalysis was attempted with
a Reaction conditions: see ref 9. b Isolated yield after purification. c Run
in CH3CN.
triethylamine (pKa ) 11)17 (Table 1, entry 9), no product was
1
observed by H NMR spectroscopy. Additionally, the ease with
which trialkylphosphines are oxidized suggests that phosphine oxide
might be the active catalyst; this was also shown not to be the case
(Table 1, entry 10).
When deuterated protic solvents were used in these reactions,
H/D exchange was observed at both of the R-positions of the R,â-
unsaturated compound. This suggested that a strong base is
9
8696
J. AM. CHEM. SOC. 2003, 125, 8696-8697
10.1021/ja035232n CCC: $25.00 © 2003 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Proposed Catalytic Cycle
Hoff plot confirmed that the reaction was exothermic and entropi-
cally disfavored (∆H° ) -5.0 kcal/mol and ∆S° ) -18 eu).
In conclusion, we have shown that nucleophilic phosphines in
the presence of R,â-unsaturated systems can be used to catalyze
the hydration and hydroalkoxylation of activated olefins in the
absence of added transition metals, or strong acids or bases. This
transformation contrasts the recently reported catalysis of the
Baylis-Hillman reaction of bis(enones) by trialkylphosphines.21,22
Our proposed catalytic cycle suggests that this very practical system
could be extended to other classes of nucleophiles with pKa’s
matched to that of the enolate intermediates. Additionally, the
possibility of using other nucleophilic catalysts would further extend
the applicability and practicality of this genre of reactivity.
generated during the reaction. Trimethylphosphine alone did not
facilitate H/D exchange of 3-pentanone, indicating that it alone is
not the active catalyst. However, addition of a catalytic amount of
an enone and trimethylphosphine caused rapid H/D exchange at
the R-positions (eq 1), clearly supporting the base-generation
hypothesis.
Acknowledgment. We gratefully acknowledge the University
of California, Berkeley for financial support. F.D.T. thanks Merck
Research Laboratories, Eli Lilly & Company, Amgen Inc., and
R.G.B. thanks the National Science Foundation (Grant CHE0094349)
for support of our programs.
Note Added after ASAP: There was an error in Scheme 1 in
the version published on the Web 6/18/2003. The version published
6/20/2003 and the print version are correct.
Supporting Information Available: Experimental procedures and
compound characterization data (PDF). This material is available free
of charge on the Internet at http://pubs.acs.org.
The resting state of the catalyst was investigated by 31P NMR
spectroscopy. When 1 was subjected to a catalytic amount of PMe3
in hydroxylic solvents, a single 31P NMR resonance at 33 ppm was
observed. This suggested the presence of a â-phosphonium ketone,18
which is consistent with 4 being the resting state of the catalyst in
our proposed mechanism.19 This â-phosphonium ketone was
independently synthesized by addition of a full equivalent of PMe3
to 1 in the absence of hydroxylic solvents, followed by addition of
HCl (eq 2). The 31P NMR resonance of this phosphonium salt
matched that observed in the catalytic reactions.
References
(1) Ryberg, P.; Matsson, O. J Am. Chem. Soc. 2001, 123, 2712-2718.
(2) Corey, E. J.; Zhang, F.-Y. Org. Lett. 1999, 1, 1287-1290.
(3) Pd-catalyzed hydration: (a) Ganguly, S.; Roundhill, D. M. Organome-
tallics 1993, 12, 4825-4832. Pd-catalyzed hydroalkoxylation: (b) Miller,
K. J.; Kitagawa, T. T.; Abu-Omar, M. M. Organometallics 2001, 20,
4403-4412.
(4) Use of high pressure for olefin hydration: Jenner, G. Tetrahedron Lett.
2001, 42, 4807-4810. Jenner, G. Tetrahedron 2002, 58, 4311-4317.
(5) Kabashima, H.; Katou, T.; Hattori, H. Appl. Catal., A 2001, 214, 121-
124.
(6) For progress towards a general method, see: Kisanga, P. B.; Ilankumaran,
P.; Fetterly, B. M.; Verkade, J. G. J. Org. Chem. 2002, 67, 3555-3560.
(7) Trost, B. M.; Li, C.-J. J. Am. Chem. Soc. 1994, 116, 10819-10820.
(8) Inanaga, J.; Baba, Y.; Hanamoto, T. Chem. Lett. 1993, 2, 241-4.
(9) Typical experimental procedure: 10 mmol of substrate in 10 mL of alcohol
was subjected to three freeze-pump-thaw cycles, following which PMe3
was added via vacuum-transfer. The flask was sealed and left for the time
indicated. The reaction mixture was then filtered through a pad of silica
and concentrated under reduced pressure. In the case of methanol addition
the desired product was obtained in high purity without chromatography.
All other reactions required flash column chromatography using mixtures
of hexanes and ethyl acetate as the eluent to afford pure product. .
(10) DFT calculations suggest that hydration of 4-phenyl-3-buten-2-one is less
enthalpically favored.
Exchange experiments and variable temperature NMR spectros-
copy were employed to investigate the reversibility of this reaction.
â-Methoxy ketone 6 readily undergoes exchange with perdeuteri-
omethanol in the presence of 5% PMe3, suggesting that the reaction
is reversible (eq 3).20 Variable-temperature NMR spectroscopy was
employed to determine the thermodynamic parameters associated
with this equilibrium. p-Methoxyphenol was chosen for the study
because it gave rise to an observable ratio of unreacted enone to
product at room temperature (eq 4). As expected on the basis of
(11) For another example of this, see: Allen, A. A.; Duffner, R.; Kurzer, F.
Tetrahedron 1978, 34, 1247-1250.
(12) For an example of retro-aldol reactions in related systems, see: Jensen,
J. L.; Hashtroudi, H. J. Org. Chem. 1976, 41, 3299-3302
(13) For a related mechanism, see: White, D. A.; Baizer, M. M. Tetrahedron
Lett. 1973, 14, 3597-3600.
(14) An alternative mechanism could involve direct displacement of PR3 from
the â-phosphonium ketone. Displacements of this type, however, are
extremely rare.
(15) Addition of 5 mol % of NaOH or NaOMe to an aqueous or alcohol solution
of enone also produced the corresponding hydroxylated or hydroalkoxy-
lated product. See ref 11 for more on base-catalyzed hydrations and
competing retro-aldol reactions.
(16) Henderson, W. A.; Streuli, C. A. J. Am. Chem. Soc. 1960, 82, 5791-
5794.
(17) Bruice, P. Y. J. Am. Chem. Soc. 1984, 106, 5959-5964.
(18) Alt, H. G.; Hayen, H. I. J. Organomet. Chem. 1986, 316, 301-313.
(19) This observation is also consistent with 5 being the resting state; however,
the 5 f 4 proton transfer should be exothermic and rapid.
(20) A catalytic amount of enone is likely formed from elimination of methanol
from 2-methoxy-4-hexanone. The mechanism of the elimination is not
necessarily the microscopic reverse of the mechanism for the addition
reactions described in Scheme 1.
(21) Wang, L.-C.; Luis, A. L.; Agapiou, K.; Jang, H.-Y.; Krische, M. J. J.
Am. Chem. Soc. 2002, 124, 2402-2403.
entropy considerations, the ratio of starting materials to products
was increased by raising the temperature from 25 to 75 °C. Upon
cooling, the system returned to its original distribution. A van’t
(22) Frank, S. A.; Mergott, D. J.; Roush, W. R. J. Am. Chem. Soc. 2002, 124,
2404-2405.
JA035232N
9
J. AM. CHEM. SOC. VOL. 125, NO. 29, 2003 8697