A new palladium(II)-catalyzed asymmetric chlorohydrin synthesis

Full text of DOI:10.1021/jo972015u

2790
J . Org. Chem. 1998, 63, 2790-2791
Sch em e 1
A New P a lla d iu m (II)-Ca ta lyzed Asym m etr ic
Ch lor oh yd r in Syn th esis
Arab El-Qisairi, Othman Hamed, and
Patrick M. Henry*
Department of Chemistry, Loyola University of Chicago,
Chicago, Illinois 60626
Sch em e 2
Received November 4, 1997
The Pd(II)-catalyzed oxidation of ethene in aqueous solu-
tion (Wacker reaction) gives exclusively acetaldehyde at low
[Cl-] (1.0 M). At high [Cl-] (>2.5 M) and high [CuCl2] (>3
M), formation of ethylene chlorohydrin occurs to an ap-
preciable extent (Scheme 1).1,2
Previous studies in these laboratories have shown that
substitution of chloride by pyridine in the coordination
2-
sphere of PdCl4 to give PdCl3(pyridine)- resulted in the
formation of chlorohydrin at [Cl-] as low as 0.2 M.3,4 At this
Synthesis of the tetrasulfonated ligand involved treatment
of Tol-BINAP with H2SO4 containing 20% SO3 at room
temperature for 24 h followed by neutralization with NaOH.7
Recrystallization from methanol provided pure samples.
Reaction with K2PdCl4 or PdCl2(PhCN)2 provided the chiral
catalyst. 31P NMR measurements confirmed that all the
tolyl rings were sulfonated. In these catalysts X ) Cl.
Treatment of a solution of [Pd(CH3CN)4]BF4 in CH3CN
with 1,3,5-pentanetriones gave the bimetallic triketone
complexes. Addition of the chiral diphosphine ligand pro-
vided the chiral bimetallic complex. After purification, the
2-
low [Cl-], PdCl4 gives only acetaldehyde at any [CuCl2].
This finding opens up the possibility of an asymmetric
chlorohydrin synthesis with R-olefins. Pd(II) catalysts
containing chiral auxiliaries should produce optically active
products. The logical starting point is the replacement of
pyridine with a chiral amine L*. Scheme 2 outlines the
general reaction scheme with monodentate chiral amines
such as (CH3)2C*NH(CH3)Ph. As shown, the two positional
isomers, 2 and 3, arise from the two possible modes of
hydroxypalladation. The ratio 2/3 of about 4 for propene
and 1-pentene is typical for a number of catalysts.
As expected, the optical purity of 2 was low, with ee’s of
10-15%. Catalysts with chiral chelating diphosphines
should give much higher optical purities. However, the
monometallic Pd(II) complex containing a diphosphine ligand
is a neutral species and thus insoluble in the reaction media.
The solution to this problem involved two different ap-
proaches. One approach used sulfonated chiral ligands
while the other involved bimetallic complexes with a bridg-
ing diphosphine ligand.6 The structures of the two catalytic
systems are shown below.
1
products were characterized by elemental analysis and H-
13C and 31P NMR. As initially prepared, X is the CH3CN
ligand. However, in the actual reaction mixtures, CH3CN
is almost certainly replaced by Cl-.
Gas uptake measurements using gas burets monitored the
progress of the reactions.8,9 Propene oxidation was moni-
tored by propene consumption. With the other nongaseous
olefins the gas was dioxygen. This is possible because the
CuCl formed in the first oxidation step to form chlorohydrin
readily reacts with dioxygen to give CuCl2. Thus, as with
the Wacker reaction, the chlorohydrin synthesis is a net air
oxidation.
(3) Francis, J . W.; Henry, P. M. J . Mol. Catal. A: Chem. 1995, 99, 77.
(4) A recent report of the oxychlorination of allylic amines and sulfides
is another example of the effect of a neutral ligand on the chlorohydrin
formation.5
(5) Lai, J .-Y.; Wang, F.-S.; Guo, G.-Z.; Dai, L.-X. J . Org. Chem. 1993, 58,
6944.
(6) These bimetallic catalysts are homogeneous analogues of the bimetal-
lic heterogeneous catalysts studied previously in these laboratories. See:
(a) Zaw, K.; Henry, P. M. J . Mol. Catal. A: Chem. 1995, 101, 187. (b) Henry,
P. M.; Ma, X.; Noronha, G.; Zaw, K. Inorg. Chim. Acta 1995, ICA240/1-2,
205.
(7) Amrani, Y.; Lecomte, L.; Sinou, D.; Bakos, J .; Toth, I.; Heil, B.
Organometallics 1989, 8, 542.
(8) In a typical experiment, a 250-mL two-necked cone-shaped flask, with
indented sides to increase the efficiency of stirring, was equipped with a
magnetic stirring bar, subseal septum, and vacuum adapter. The flask was
charged with 20 mL of H2O, 5 mL of THF, 10.09 g (75 mmol) of CuCl2,
0.212 g (5 mmol) of LiCl, and 0.1 mmol of catalyst. The flask was then placed
in a constant-temperature bath at 25 °C and connected to the gas uptake
system.9 The system was evacuated for 10 min on the vacuum line with
the stirrer running. The stirring was stopped and the system pressured to
1.0 atm with dioxygen. The olefin was added to the reaction mixture by
syringe. The mercury in the gas buret and the leveling bulb were equalized,
and a reading was taken. The stirrer was turned on to start the reaction.
The pressure was kept constant at 1 atm by continuously leveling the
mercury in the gas buret and bulb. Gas uptake readings were taken at
regular intervals. The reaction was allowed to run until the reaction mixture
was at least 0.25 M in total oxidation product. The oxidation product was
separated from the reaction mixture by continuous extraction with ether
overnight. The ether was dried over anhydrous MgSO4 and removed by
(1) For general discussion and references see: Henry, P. M. Palladium
Catalyzed Oxidation of Hydrocarbons; D. Reidel: Dordrecht, Holland, 1980;
pp 41-84.
(2) (a) Stangl, H.; J ira, R. Tetrahedron Lett. 1970, 3589. (b) Ba¨ckwall, J .
E.; Åkermark, B.; Ljunggren S. O. J . Am. Chem. Soc. 1979, 101, 2411.
distillation. Analysis of the product was carried out by VPC and 1H and13
C
NMR.
(9) For a similar apparatus see: ref 1, p 57.
S0022-3263(97)02015-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/07/1998

Communications
J . Org. Chem., Vol. 63, No. 9, 1998 2791
a
Ta ble 1. Resu lts for th e Oxid a tion of Sever a l Olefin s by Ar -BINAP Ca ta lysts in th e P r esen ce of Cu Cl2
run
Ar
cat. (config)
[LiCl] (M)
substrate
propene
CH2dCHC(O)CH3
CH2dCHCH2OPh
propene
1-pentene
1-hexene
CH2dCHCH2OPh
CH2dCHCH2O(1-Naph)
2/3 ratiob
% ee of 2
turnovers
1
2
3
4
5
6
7
8
4
4
4
5
5
5
5
5
mono(R)
mono(R)
mono(R)
bim(S)e
bim(S)g
bim(S)g
bim(S)g
bim(S)g
0.20
0.20
0.20
0.10
0.22
2.20
0.30
0.20
12
>95
>95
3.5
46
76
68
94f
81
87
93
80
60c
65d
72d
195c
200d
155d
170d
175d
4.0
18.7
>95
>95
a
All runs contain 0.1-0.3 mmol of chiral catalyst in 25-50 mL of solvent and are 3-5 M in CuCl2. Temperature ) 25 °C. The solvent
b
was a H2O/THF mixture containing 30-90% THF by volume. The reaction mixture also contained varying amounts of the Wacker
ketone product (5-30%). c Measured by propene uptake using gas burets. Dioxygen is oxidant; turnovers measured by O2 uptake using
d
gas burets. In calculating turnovers dioxygen is assumed to be a four-electron oxidant. e R1 ) R2 ) CH3. f Absolute configuration was
g
determined to be (S) by converting to the epoxide and comparing with an authentic sample. R1 ) CH3, R2 ) Ph.
Table 1 lists some representative results. As discussed
above, all reactions were catalytic oxidations. The catalytic
turnovers, listed in the last column, do not represent the
maximum number obtainable. In all cases, the reaction rate
had not diminished at the termination point, and many more
turnovers are possible.
Previous studies suggested that the degree of sulfonation
could be an important factor in determining ee values.7,10
In the hydrogenation of imines by rhodium(I) catalysts
containing a sulfonated chiral diphosphine, it was found that
the optical purity depended on the degree of sulfonation.10
When the degree of sulfonation was about 4 the % ee of the
product was only 19%. However, when the degree of
sulfonation was about 1.5 the optical purity rose to 96% ee.
Thus, the use of catalysts with monosufonated BINAP
probably would increase ee values with the monometallic
sulfonated catalysts in the present studies.
The 2/3 ratio for the monometallic catalyst (runs 1-3) was
high, indicating the formation of 3 is not a serious problem.
Only propene gave any detectable amount of 3. Even in this
case it was less than 10% of the total. Steric factors must
account for this trend in 2/3 ratios. For the bimetallic
catalyst, the 2/3 ratio is about 4 for propene and 1-pentene
(runs 4 and 5) but increases to almost 20 for 1-hexene (run
6). The reason for the latter jump is not certain but could
result from the change in [Cl-]. As with the sulfonated
catalyst, the allyl ethers gave no 3 (runs 7 and 8).
The optical purities with the bimetallic catalyst were
higher than with the sulfonated monometallic catalyst,
ranging from modest to good. In all cases, the ee was over
80% and rose to over 90% in two runs (4 and 7). Simple
steric factors cannot explain these trends. Thus, the 94%
ee obtained for propene in run 4 is the highest value
obtained. In run 8, these factors would predict a higher %
ee for the 1-naphthyl ether.
Without the free rotation of the monodentate ligand in 1
the asymmetric induction is increased. The second factor
is the coordination sphere of the Pd(II). With a bidentate
â-diketone and a phosphine ligand, there is only one open
coordination site available to the olefin. The site is adjacent
to the chiral phosphine ligand where the asymmetric induc-
tion will be the highest. Note that in catalyst 1 the site trans
to the chiral ligand would experience little chiral induction.
The discovery of a new asymmetric chlorohydrin synthesis
increases the present stable of transition-metal-catalyzed
chiral syntheses. It differs from previous asymmetric oxida-
tion reactions, such as asymmetric epoxidation of allylic
alcohols,11a epoxidation of unfunctionalized olefin,11b and
asymmetric dihydroxylation11c in that it is a catalytic air
oxidation. The closest comparison to the present reaction
is the Pd(II)-catalyzed synthesis of chiral chlorhydrins using
an olefin containing a chiral allylic amine ligand.5 This very
interesting conversion gives a chiral chlorohydrin containing
the chiral amine in poor to modest optical purities (ee )
1-77%). As the chiral agent is monodentate, the system is
analogous to catalyst 1 in Scheme 2. The fact that the
optical yields were generally higher than those obtained with
1 can be rationalized by the fact the system is rigid in the
same fashion as the bimetallic catalyst in the present
studies. In addition, as opposed to catalyst 1, the olefin is
forced next to the chiral center.
An interesting possibility is the formation of other asym-
metric products by variation of reaction conditions. The
reaction consists of two distinct steps, addition and decom-
position. In chloride-free media, nucleophiles other than
chloride could be involved in the decomposition step, and
better nucleophiles than water could be the attacking species
in the first step. Thus, preparation of chiral bromohydrins
and iodohydrins as well as chiral dibromides is a viable
possibility.
The high asymmetric induction obtained with the bime-
tallic catalysts is probably the most unexpected and inter-
esting result of the present work. Formally, the bimetallic
catalyst, with one phosphine per Pd(II), is analogous to the
monometallic catalyst 1 in Scheme 2. Two factors almost
certainly explain the high enantioselectivity observed. The
first is the rigid structure of the bridging diphosphine ligand.
Probably the main practical application of this new
technology will be the preparation of chlorohydrins from allyl
ethers. Thus, the chlorohydrin from allyl-1-naphthyl ether
is a precursor for (-)-propranolol, the â-adrenergic blocker.12
Future work will concentrate on defining the scope of the
reaction and improving optical yields. The latter effort
involves the testing of other chiral auxiliaries and determi-
nation of the effect of [Cl-] and solvent composition on %
ee’s.
(10) Bakos; J .; Orosz, A.; Heil, B.; Laghmari, M.; Lhoste, P.; Sinou, D. J .
Chem. Soc., Chem. Commun. 1991, 1684.
(11) Ojima, I., Ed.; Catalytic Asymmetric Synthesis; VCH: New York,
1993. (a) J ohnson, R. A.; Sharpless, K. B. Chapter 4.1. (b) J acobsen, E. N.
Chapter 4.2. (c) J ohnson, R. A.; Sharpless, K. B. Chapter 4.4.
(12) Aitken, R. A.; Kile´nyi, S. N., Ed. Asymmetric Synthesis; Chapman
& Hall: New York, 1992; p 220.
Su p p or tin g In for m a tion Ava ila ble: Text describing experi-
mental procedures and characterization data (6 pages).
J O972015U