Enantioselective synthesis of acyclic allylic esters catalyzed by a palladium/BINAP(S) system

Article abstract of DOI:10.1016/j.tetlet.2004.08.105

The synthesis of chiral nonracemic acyclic allylic pivalates via the Pd-catalyzed allylic substitution of racemic allylic carbonates is presented. Good to excellent enantioselectivities (up to 90%) were observed in several cases. An extraordinarily high preference for the production of the branched regioisomeric product is seen when starting from 3-buten-2-yl and crotyl substrates. A significant kinetic resolution (k_rel = 38) of the 1,3-dimethylallyl substrate was also observed, leading to the production of esters of both enantiomers of an allylic alcohol with a single enantiomer of catalyst.

Full text of DOI:10.1016/j.tetlet.2004.08.105

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 7613–7616
Enantioselective synthesis of acyclic allylic esters
catalyzed by a palladium/BINAP(S) system
J. W. Faller* and Jeremy C. Wilt
Department of Chemistry, Yale University, New Haven, CT 06520-8107, USA
Received 7 July 2004; revised 17 August 2004; accepted 18 August 2004
Abstract—The synthesis of chiral nonracemic acyclic allylic pivalates via the Pd-catalyzed allylic substitution of racemic allylic
carbonates is presented. Good to excellent enantioselectivities (up to 90%) were observed in several cases. An extraordinarily high
preference for the production of the branched regioisomeric product is seen when starting from 3-buten-2-yl and crotyl substrates. A
significant kinetic resolution (krel = 38) of the 1,3-dimethylallyl substrate was also observed, leading to the production of esters of
both enantiomers of an allylic alcohol with a single enantiomer of catalyst.
Ó 2004 Elsevier Ltd. All rights reserved.
The palladium-catalyzed asymmetric allylic substitution
reaction is a powerful synthetic method for the forma-
tion of C–C, C–N, C–S, and C–O bonds.1,2 These reac-
tions constitute a means of transforming an achiral or
racemic allylic ester into a different, enantiomerically
enriched allylic functionality determined by the nucleo-
phile chosen. However, if the desired product is an
enantiomerically enriched allylic alcohol, a different
approach is often taken. Asymmetric additions to ally-
lic aldehydes3–6 and kinetic resolutions of the allylic
esters7–13 or alcohols14–17 are both viable options. How-
ever, the former method is often less selective when an
aliphatic allylic alcohol is desired, and the latter method
has an intrinsic 50% maximum yield.
can be at least partly overcome by the use of allylic
carbonates as substrates, taking advantage of their
greater propensity to oxidatively add to Pd0 than the
produced allylic ester (Scheme 1).
This methodology was initially reported by Trost18 in
the synthesis of cyclic allylic esters from their allylic carbo-
nate counterparts with excellent enantioselectivities. We
wish to report the successful extension of this methodo-
logy to acyclic allylic substrates with our Pd/BINAP(S)
system.
Recently, we reported the palladium-catalyzed kinetic
resolution of several acyclic allylic acetates and benzo-
ates in the presence of the axially chiral (S)-BINAP(S)
ligand 1 (Scheme 2).19
Through the use of a carboxylate as nucleophile in the
palladium-catalyzed asymmetric allylic substitution
reaction, chiral nonracemic allylic esters can be pro-
duced from racemic allylic esters. This method, though
attractive, suffers from an inherent chemoselectivity
problem. Since the product of this reaction can also
serve as the substrate, the rate of reaction of the starting
allylic ester must be significantly greater than that of the
produced allylic ester. If the rates are similar, not only
may the reaction not proceed in the forward direction,
but the produced allylic ester will racemize over the
course of the reaction. This intrinsic selectivity issue
O
OCO2Et
O
t-Bu
R
R
R
R
O
In the presence of
O
t-Bu
L2Pd0, NaO2C(t-Bu)
Keywords: Palladium; Asymmetric allylation; Catalysis.
*
R
R
Corresponding author. Tel.: +1 2034323954; fax: +1 2034326144;
e-mail: jack.faller@yale.edu
Scheme 1.
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.08.105

7614
J. W. Faller, J. C. Wilt / Tetrahedron Letters 45 (2004) 7613–7616
O
O
With our conditions in hand, we then examined the
[(η3-C3H5)Pd(S)-BINAP(S)]SbF6
scope of the reaction with several other allylic carbon-
ates (Table 2). Interestingly, a kinetic resolution was
observed with symmetrical substrate 3a. The selectivity
of a kinetic resolution is measured as the ratio of the rate
constants of the two enantiomers of substrate in ques-
tion, and is known as the selectivity factor, or krel. A krel
of 38 was observed for 3a, which is comparable to the
best reported values8 for derivatives of this allylic alco-
hol. Additionally, the signs of the optical rotations for
product and reactant are opposite, which suggests that
this reaction constitutes a separation of the two enantio-
mers of allylic alcohol with high enantioselectivity. It is
not obvious why this particular substrate should give
rise to a kinetic resolution as opposed to the other sub-
strates tested, although we can say with certainty that
the Pd catalyst must oxidatively add (À)-3a at a greater
rate than it adds (+)-3a or (À)-3b.
O
R3
O
R3
NaCH(CO2Me)2
R1
R2
R1
R1
R2
CH(CO2Me)2
R2
S
PPh2
PPh2
1
Scheme 2.
We were interested in the possibility of combining the
high selectivity that this system had shown for addition
of one enantiomer of starting allylic ester with the pos-
sible enantioselective synthesis of allylic esters from ally-
lic carbonates. Initially, the desired outcome was to
selectively consume a single enantiomer of allylic car-
bonate and produce a single enantiomer of allylic ester;
in essence, a method of preparing both enantiomers of a
given allylic ester with a single enantiomer of catalyst.
Isomeric unsymmetrically substituted substrates 4a and
5a can give rise to both linear and branched allylation
products 4b/4c and 5b/5c. In most Pd-catalyzed reac-
tions, these substrates proceed to give the achiral, linear
substitution product rather than the chiral, branched
regioisomer.21–24 Previously, we have reported that the
Pd/(S)-BINAP(S) catalyzed alkylation of these sub-
strates with sodium dimethylmalonate gave rise to the
branched regioisomeric product in up to a 79:21 ratio.
We were pleased to find that reaction of substrate 4a
under our current conditions gave 4b/4c in a 92:8 regio-
isomeric ratio, with high ee in the branched isomer. We
attribute this preference for formation of the branched
product to both the electronic and steric effects of the
BINAP(S) ligand. Steric effects would tend to place
the more substituted allylic terminus cis to sulfur and
trans to phosphorus. The electronic preference for nucleo-
philic attack trans to phosphorus then gives rise to for-
mation of the branched product. Substrate 5a also
gave 5b/5c with virtually identical regioisomeric distri-
bution (entry 3); however, the reaction only proceeded
to 27% conversion under our standard conditions, even
after prolonged reaction times. Doubling the reactant
To establish general reaction conditions,20 we began our
investigation with the ethyl carbonate of 1,3-diphenyl-
allyl alcohol 2a (Table 1). Unfortunately, the sodium
pivalate nucleophile proved exceedingly insoluble in
thf, and no reaction was observed. Addition of the
cation-solubilizing 18-crown-6 ether to the reaction
mixture did result in conversion to the pivalate product
with good enantioselectivity; however, no kinetic resolu-
tion of the starting allylic carbonate was observed
during the course of the reaction. As the concentration
of 18-crown-6 was increased, a corresponding increase
in reaction rate was observed, with no loss of enantio-
selectivity. This increase in rate can be attributed to an
higher concentration of nucleophile being in solution.
Optimal results were found with 1equiv of 18-crown-6
at 0°C, which gave (À)-2b in 90% ee. It should be noted
that these reactions must be carefully monitored with
respect to percent conversion, and were generally
quenched prior to full conversion to avoid the possibility
of racemization.
Table 1. Optimization of reaction conditions
O
OCO2Et
O
t-Bu
Ph
[(η3-allyl)Pd(S)-BINAP(S)]SbF6
3 NaO2C-t-Bu / 18-crown-6
Ph
Ph
Ph
2a
2b
Entry
Catalyst loadinga
18-c-6a
T, °C
Time, h
% Cb
% eec
Optical rotationd,28 (2b)
1
2
3
4
5
6
0
100%
0
5%
rt
rt
rt
0
24
24
23.5
47
7
0
0
—
—
77
87
89
90
—
—
5%
1%
5%
5%
5%
76
88
90
85
(—)
(—)
(—)
(—)
5%
25%
100%
0
0
2.5
a Mol% with respect to substrate.
b Percent conversion as determined by 1H NMR.
c Determined by chiral shift experiments with (+)-Eu(hfc)3.
d Sign of rotation measured at 20°C and 589nm.

J. W. Faller, J. C. Wilt / Tetrahedron Letters 45 (2004) 7613–7616
7615
Table 2. Reaction scope
O
O
5% [(η3-allyl)Pd(S)-BINAP(S)]SbF6
OCO2Et
O
t-Bu
t-Bu
O
3 NaO2C-t-Bu / 18-crown-6
R1
R2
R1
R2 R1
R2
thf, 0 °C
a
c
b
3: R1=R2=Me
4: R1=Me, R2=H
5: R1=H, R2=Me
6: R1=H, R2=Ph
7: R1=Me, R2=Ph
Entry Substrate 18-c-6a Time, h % Cb Ratio b:cb % ee productc (optical rotation)d % ee starting materialc (optical rotation)d,28
1
2
3
4
5
6
7
3a
4a
5a
5ae
5ae
6a
7a
100%
100%
100%
100%
25%
12
0.5
22
0.5
45
97
27
91
97
91
51
—
87 (À)
85 (+)
0
73 (+) krel = 38
—
92:8f
9:91f
13:87f
9:91f
—
—
0
1
71 (+)
—
—
—
100%
100%
1
0.33
100:0
100:0
43 (À)
24 (+) krel = 2
a Mol% with respect to substrate.
b Percent conversion as determined by 1H NMR.
c Determined by chiral shift experiments with (+)-Eu(hfc)3.
d Sign measured at 20°C and 589nm.
e Reactant concentrations were doubled.20
f Note that 4b = 5c, which are the branched products.
concentration (entry 4) increased the conversion, but led
to a racemic product. By doubling the concentration of
the reactants and reducing the percentage of 18-crown-6
(entry 5), we were able to achieve full conversion to the
pivalate, but the ee was lower than with 4a. This is an
unanticipated result because isomeric 4a and 5a are
often presumed to yield identical p-allyl intermediates;
therefore, they would be expected to give rise to identical
regio- and stereoselectivities in the product. Partial
isomerization or racemization might account for this.
In order to verify that the product ester 5c had not par-
tially racemized over the course of the reaction, the reac-
tion was quenched at various percent conversions under
the conditions for entry 5. Consistent enantioselectivities
were observed at all stages of the reaction under these
conditions.
In summary, we have shown that the Pd/(S)-BINAP(S)
system is effective for the catalytic synthesis of allylic
pivalates from the corresponding racemic carbonates
with high regio- and stereoselectivity. Of particular
interest is the formation of predominantly branched
products from 4a/5a in good to high ee, and the highly
selective kinetic resolution of 3a. Future work will focus
on the many interesting mechanistic facets of this
transformation.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2004.08.105.
Thus reaction conditions for appropriate relative rates
of oxidative addition of linear 5a and subsequent racemi-
zation of 5c must be carefully controlled in order to
achieve high eeÕs in the product.
References and notes
1. Trost, B. M.; Crawley, M. L. Chem. Rev. 2003, 103, 2921–
2943.
Cinnamyl-based substrates 6a and 7a were substituted
exclusively in the less sterically hindered position, which
is contrary to our expectations based on the above inter-
pretation of ligand effects. Regiochemical analysis of the
product dictates that the less substituted allylic position
(H or Me) must be cis to sulfur and trans to phosphorus.
This effectively removes steric arguments from consider-
ation, as the two larger substituents must be cis to one
another in the square-planar metal geometry. A relative
change in the charge distribution of the allylic termini
may be responsible for the observed regioselectivity.
Several reports have documented the preference for
Pd-catalyzed allylic substitution at the more remote ter-
minus from the most electron-withdrawing group.25–27
2. Trost, B. M.; Van Vranken, D. L. Chem. Rev. 1996, 96,
395–422.
3. Pu, L. Tetrahedron 2003, 59, 9873–9886.
4. Wipf, P.; Ribe, S. J. Org. Chem. 1998, 63, 6454–
6455.
5. Miller, K. M.; Huang, W. S.; Jamison, T. F. J. Am. Chem.
Soc. 2003, 125, 3442–3443.
6. Oppolzer, W.; Radinov, R. N. Helvetica Chim. Acta 1992,
75, 170–173.
7. Selvakumar, K.; Valentini, M.; Pregosin, P. S.; Albinati,
A. Organometallics 1999, 18, 4591–4597.
8. Gais, H. J.; Spalthoff, N.; Jagusch, T.; Frank, M.; Raabe,
G. Tetrahedron Lett. 2000, 41, 3809–3812.
9. Hayashi, T.; Yamamoto, A.; Ito, Y. J. Chem. Soc., Chem.
Commun. 1986, 1090–1092.

7616
J. W. Faller, J. C. Wilt / Tetrahedron Letters 45 (2004) 7613–7616
10. Longmire, J. M.; Wang, B.; Zhang, X. M. Tetrahedron
Lett. 2000, 41, 5435–5439.
11. Reetz, M. T.; Sostmann, S. J. Organomet. Chem. 2000,
603, 105–109.
12. Ramdeehul, S.; Dierkes, P.; Aguado, R.; Kamer, P. C. J.;
van Leeuwen, P.; Osborn, J. A. Angew. Chem., Int. Ed.
1998, 37, 3118–3121.
13. Gais, H. J.; Jagusch, T.; Spalthoff, N.; Gerhards, F.;
Frank, M.; Raabe, G. Chem. Eur. J. 2003, 9, 4202–4221.
14. Lussem, B. J.; Gais, H. J. J. Am. Chem. Soc. 2003, 125,
6066–6067.
temperature was then added, and the reaction mixture
stirred for the given amount of time. The reaction was
quenched by adding excess pentane. The mixture was
filtered through a pad of Celite and concentrated under
reduced pressure. Column chromatography over silica gel
(25:1 pentane:Et2O) provided first the pure allylic pivalate,
followed by the allylic carbonate. Column pretreatment
with 3% v/v Et3N/solvent mixture was necessary to
prevent decomposition with 2a. All products were satis-
factorily characterized by comparison of their 1H NMR
spectrum with literature values.
15. Faller, J. W.; Tokunaga, M. Tetrahedron Lett. 1993, 34,
7359–7362.
21. Sato, Y.; Yoshino, T.; Mori, M. Org. Lett. 2003, 5, 31–33.
22. Sjogren, M. P. T.; Hansson, S.; Akermark, B.; Vitagliano,
A. Organometallics 1994, 13, 1963–1971.
23. You, S. L.; Zhu, X. Z.; Luo, Y. M.; Hou, X. L.; Dai, L. X.
J. Am. Chem. Soc. 2001, 123, 7471–7472.
16. Kitamura, M.; Kasahara, I.; Manabe, K.; Noyori, R.;
Takaya, H. J. Org. Chem. 1988, 53, 708–710.
17. Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.;
Ikeda, M.; Sharpless, K. B. J. Am. Chem. Soc. 1981, 103,
6237–6240.
24. Acemoglu, L.; Williams, J. M. J. Adv. Synth. Catal. 2001,
343, 75–77.
18. Trost, B. M.; Organ, M. G. J. Am. Chem. Soc. 1994, 116,
10320–10321.
25. Delbecq, F.; Lapouge, C. Organometallics 2000, 19, 2716–
2723.
19. Faller, J. W.; Wilt, J. C.; Parr, J. Org. Lett. 2004, 6, 1301–
1304.
26. Branchadell, V.; Moreno-Manas, M.; Pajuelo, F.; Pleixats,
R. Organometallics 1999, 18, 4934–4941.
20. General experimental procedure: sodium pivalate (36mg,
290lmol), 18-crown-6, and [(g3-allyl)Pd(S)-BINAP-
(S)]SbF6 (5mg, 4.81lmol) were placed under inert atmos-
phere and 2mL dry thf was added. The dark solution was
stirred at the desired temperature for 10min. A solution of
allylic carbonate (96.3lmol) in 1mL dry thf at the desired
27. Akermark, B.; Hansson, S.; Krakenberger, B.; Vitagliano,
A.; Zetterberg, K. Organometallics 1984, 3, 679–
682.
28. See Supplementary data for tentative assignments of
absolute configurations by comparison of rotations with
allylic methyl carbonates and allylic acetates.