Mechanistic investigations of the palladium-catalyzed aerobic oxidative kinetic resolution of secondary alcohols using (-)-Sparteine

Article abstract of DOI:10.1021/ja034262n

The mechanistic details of the Pd(II)/(-)-sparteine-catalyzed aerobic oxidative kinetic resolution of secondary alcohols were elucidated, and the origin of asymmetric induction was determined. Saturation kinetics were observed for rate dependence on [(-)-sparteine]. First-order rate dependencies were observed for both the Pd((-)-sparteine)Cl₂ concentration and the alcohol concentration at high and low [(-)-sparteine]. The oxidation rate was inhibited by addition of (-)-sparteine HCl. At low [(-)-sparteine], Pd-alkoxide formation is proposed to be rate limiting, while at high [(-)-sparteine], β-hydride elimination is proposed to be rate determining. These conclusions are consistent with the measured kinetic isotope effect of k_H/k_D = 1.31 ± 0.04 and a Hammett ρ value of -1.41 ± 0.15 at high [(-)-sparteine]. Calculated activation parameters agree with the change in the rate-limiting step by increasing [(-)-sparteine] with ΔH? = 11.55 ± 0.65 kcal/ mol, ΔS? = -24.5 ± 2.0 eu at low [(-)-sparteine], and ΔH? = 20.25 ± 0.89 kcal/mol, ΔS? = -5.4 ± 2.7 eu at high [(-)-sparteine]. At high [(-)-sparteine], the selectivity is influenced by both a thermodynamic difference in the stability of the diastereomeric Pd-alkoxides formed and a kinetic β-hydride elimination to maximize asymmetric induction. At low [(-)-sparteine], the selectivity is influenced by kinetic deprotonation, resulting in lower k_rel values. A key, nonintuitive discovery is that (-)-sparteine plays a dual role in this oxidative kinetic resolution of secondary alcohols as a chiral ligand on palladium and as an exogenous chiral base.

Full text of DOI:10.1021/ja034262n

Published on Web 05/17/2003
Mechanistic Investigations of the Palladium-Catalyzed Aerobic
Oxidative Kinetic Resolution of Secondary Alcohols Using
(-)-Sparteine
Jaime A. Mueller and Matthew S. Sigman*
Contribution from the Department of Chemistry, UniVersity of Utah, 315 South 1400 East,
Salt Lake City, Utah 84112
Received January 21, 2003; E-mail: sigman@chem.utah.edu
Abstract: The mechanistic details of the Pd(II)/(-)-sparteine-catalyzed aerobic oxidative kinetic resolution
of secondary alcohols were elucidated, and the origin of asymmetric induction was determined. Saturation
kinetics were observed for rate dependence on [(-)-sparteine]. First-order rate dependencies were observed
for both the Pd((-)-sparteine)Cl2 concentration and the alcohol concentration at high and low [(-)-sparteine].
The oxidation rate was inhibited by addition of (-)-sparteine HCl. At low [(-)-sparteine], Pd-alkoxide
formation is proposed to be rate limiting, while at high [(-)-sparteine], â-hydride elimination is proposed to
be rate determining. These conclusions are consistent with the measured kinetic isotope effect of kH/kD
)
1.31 ( 0.04 and a Hammett F value of -1.41 ( 0.15 at high [(-)-sparteine]. Calculated activation parameters
agree with the change in the rate-limiting step by increasing [(-)-sparteine] with ∆Hq ) 11.55 ( 0.65 kcal/
mol, ∆Sq ) -24.5 ( 2.0 eu at low [(-)-sparteine], and ∆Hq ) 20.25 ( 0.89 kcal/mol, ∆Sq ) -5.4 ( 2.7
eu at high [(-)-sparteine]. At high [(-)-sparteine], the selectivity is influenced by both a thermodynamic
difference in the stability of the diastereomeric Pd-alkoxides formed and a kinetic â-hydride elimination to
maximize asymmetric induction. At low [(-)-sparteine], the selectivity is influenced by kinetic deprotonation,
resulting in lower krel values. A key, nonintuitive discovery is that (-)-sparteine plays a dual role in this
oxidative kinetic resolution of secondary alcohols as a chiral ligand on palladium and as an exogenous
chiral base.
Introduction
substrate scope, efficiency, and utility of Pd(II)-catalyzed aerobic
oxidation of alcohols.6
Pd(II)-catalyzed oxidations have broad utility in organic
synthesis. An excellent example is the Wacker oxidation1
employed in industry to generate millions of tons of carbonyl
compounds from olefins per year. One of the unique features
of the Wacker oxidation is the use of molecular oxygen as the
terminal oxidant. Molecular oxygen can be considered an ideal
terminal oxidant due to availability and environmentally benign
byproducts produced in the oxidation manifold. In the develop-
ment of simple, cost efficient oxidations, Pd(II)-catalyzed
aerobic oxidations have been extended to more diverse reaction
types including amination of olefins,2 acetoxylation,3 and
transformations of tert-cyclobutanols.4 Of the Pd(II)-catalyzed
aerobic oxidations, the simple conversion of alcohols to ketones
and aldehydes has garnered the most attention. Aerobic oxidation
of alcohols using Pd(II) catalysts was first reported by Schwartz
and Blackburn in 1977.5 Subsequent efforts have extended the
Recently, our research group7 and Stoltz et al.8 simultaneously
reported adaptations of Pd(II)-catalyzed aerobic oxidation of
alcohols6e,g to kinetic resolution as a simple method to access
enantiomerically enriched secondary alcohols (eq 1). In our
reported kinetic resolution, the optimized system employs 5 mol
% Pd(II) catalyst and 20 mol % (-)-sparteine, a commercially
(6) For examples of Pd(II)-catalyzed aerobic oxidations, see: (a) Schultz, M.
J.; Park, C. C.; Sigman, M. S. Chem. Commun. 2002, 3034. (b) Kakiuchi,
N.; Maeda, Y.; Nishimura, T.; Uemura, S. J. Org. Chem. 2001, 66, 6620.
(c) Hallman, K.; Moberg, C. AdV. Synth. Catal. 2001, 343, 260. (d) ten
Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Science 2000, 287, 1636.
(e) Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. J. Org. Chem. 1999,
64, 6750. (f) Ebitani, K.; Fujie, Y.; Kaneda, K. Langmuir 1999, 15, 3557.
(g) Nishimura, T.; Onoue, T.; Ohe, K.; Uemura, S. Tetrahedron Lett. 1998,
39, 6011. (h) A¨ıt-Mohand, S.; Muzart, J. J. Mol. Catal. A 1998, 129, 135.
(i) Peterson, K. P.; Larock, R. C. J. Org. Chem. 1998, 63, 3185. (j) Kaneda,
K.; Fujie, Y.; Ebitani, K. Tetrahedron Lett. 1997, 38, 9023. (k) Kaneda,
K.; Fujii, M.; Morioka, K. J. Org. Chem. 1996, 61, 4502. (l) Aiet-Mohand,
S.; Henin, F.; Muzart, J. Tetrahedron Lett. 1995, 36, 2473. (m) Go´mez-
Bengoa, E.; Noheda, P.; Echavarren, A. M. Tetrahedron Lett. 1994, 35,
7097. (n) Blackburn, T. F.; Schwartz, J. J. Chem. Soc., Chem. Commun.
1977, 157.
(1) For reviews, see: (a) Tsuji, J. Palladium Reagents and Catalysts; InnoVation
in Organic Synthesis; John Wiley & Sons: New York, 1995; pp 19-124.
(b) Trost, B. M.; Verhoeven, T. R. In ComprehensiVe Organometallic
Chemistry; Wilkenson, G., Ed.; Oxford: New York, 1982; Vol. 8, pp 854-
983.
(2) Fix, S. R.; Brice, J. L.; Stahl, S. S. Angew. Chem., Int. Ed. 2002, 41, 164.
(3) Ebitani, K.; Choi, K. M.; Mizugaki, T.; Kaneda, K. Langmuir 2002, 18,
1849.
(4) Nishimura, T.; Ohe, K.; Uemura, S. J. Org. Chem. 2001, 66, 1455.
(5) Blackburn, T. F.; Schwartz, J. J. Chem. Soc., Chem. Commun. 1977, 157.
(7) Jensen, D. R.; Pugsley, J. S.; Sigman, M. S. J. Am. Chem. Soc. 2001, 123,
7475.
(8) Ferreira, E. M.; Stoltz, B. M. J. Am. Chem. Soc. 2001, 123, 7725.
9
10.1021/ja034262n CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 7005-7013
7005

A R T I C L E S
Mueller and Sigman
available diamine natural product, to oxidatively resolve various
9
racemic benzylic alcohols giving krel values ranging from 9.8
to 23.6 and enantiomeric excess values up to 99%.10,11 The
transformation is performed at mild temperatures, 60-65 °C,
in 1,2-dichloroethane solvent, with a balloon pressure of
molecular oxygen.12
Figure 1. Structures of Pd((-)-sparteine)Cl2 and Pd((-)-sparteine)(OAc)2.
While the Pd(II)/(-)-sparteine-catalyzed aerobic oxidative
kinetic resolution provides a useful synthetic method to access
enantiomerically enriched secondary alcohols, there are limita-
tions. The major drawback of this protocol is the use of (-)-
sparteine as the chiral agent. Since it is widely available only
as a single enantiomer, just one enantiomer of enriched alcohol
is directly accessible.13,14 Additionally, structural changes to (-)-
sparteine for enhancement of selectivity and reactivity of the
catalyst are difficult to envision.15 Therefore, an improved
catalyst system requires replacement of (-)-sparteine with a
chiral agent available in both antipodes that can also be
manipulated to allow for systematic variations of structure.
Successfully replacing (-)-sparteine as the chiral agent requires
a clear understanding of how it affects the reactivity and the
selectivity of the reaction. Accordingly, we sought to elucidate
the precise role(s) of (-)-sparteine and the other reagents in
Pd(II)/(-)-sparteine oxidative kinetic resolution to provide a
platform for the development of an improved catalytic system.
Table 1. Base Screen for Pd-Catalyzed Oxidative Kinetic
Resolution of Alcohols
% conv. (% ee)a
k
rel
b
entry
base
no base
(-)-sparteine
Hunig’s base
Cs2CO3
a
b
c
d
e
0.0 (N.A.)
51.5 (82.4)
28.0 (23.0)
52.9 (53.8)
14.0 (11.7)
N.A.
20.1
5.1
4.7
6.9
KOt-Bu
a Conversion determined using internal standard. b Conversion <10%
gives inaccurate krel values.
alkoxide species, and catalyst regeneration.18 Herein, we disclose
the mechanistic details of the reaction prior to catalyst turnover,
including the origin of asymmetric induction, the clarification
of the active catalyst, and the specific role of each additive.19,20
Various mechanisms have been proposed for Pd(II)-catalyzed
oxidations of alcohols, but have had little evidence to support
them, until recently.16,17 Based on the Wacker oxidation, as well
as other Pd(II)-catalyzed oxidation processes, a plausible
mechanism can be considered. The basic steps include binding
of the alcohol to the catalyst, â-hydride elimination from a Pd-
Results and Discussion
In our reported Pd(II)/(-)sparteine-catalyzed oxidative kinetic
resolution, the catalyst was formed in situ from the appropriate
Pd(II) source and (-)-sparteine. To investigate the nature of
the active catalyst structure, complexes were prepared using Pd-
(OAc)2 and a PdCl2 source. Single crystals were obtained for
each Pd-complex. X-ray analysis revealed (-)-sparteine binds
bidentate to one side of a slightly distorted square planar Pd in
both cases (Figure 1).7,21 To determine if 1 is active in the
catalysis, it was submitted to the standard oxidation reaction
conditions without exogenous (-)-sparteine.22 Surprisingly, no
alcohol oxidation or decomposition was observed (Table 1, entry
a). However, upon addition of 10 mol % exogenous (-)-
sparteine, the catalytic activity was reestablished, and krel values
were measured similar to those in the in situ method (Table 1,
entry b). This observation suggests that exogenous (-)-sparteine
is necessary for oxidation.
A possible role of exogenous (-)-sparteine is as a Brønsted
base to form Pd-alkoxide B from bound alcohol A (Scheme 1).
A simple test of this hypothesis is to evaluate the effectiveness
of other exogenous bases in the oxidative kinetic resolution.
Thus, several bases were tested under the standard conditions
(Table 1). All bases examined yielded a catalytic oxidation. With
the addition of cesium carbonate, a slightly enhanced conversion
as compared to (-)-sparteine was observed; however, (-)-
(9) The krel was calculated using krel ) ln[(1 - C)(1 - ee)]/ln[(1 - C)(1 +
ee)] where C is the conversion and ee is the enantiomeric excess. For an
excellent discussion of kinetic resolutions, see: Kagan, H. B.; Fiaud, J. C.
Kinetic Resolution. Top. Stereochem. 1988, 18, 249.
(10) Alcohols can be resolved by other methods. For catalytic acylation, see:
(a) Lin, M.-H.; RajanBabu, T. V. Org. Lett. 2002, 4, 1607. (b) Vedejs, E.;
MacKay, J. A. Org. Lett. 2001, 3, 535. (c) Copeland, G. T.; Miller, S. J.
J. Am. Chem. Soc. 2001, 123, 6496. (d) Bellemine-Laponnaz, S.; Tweddell,
J.; Ruble, J. C.; Breitling, F. M.; Fu, G. C. Chem. Commun. 2000, 1009.
(e) Jarvo, E. R.; Copeland, G. T.; Papaioannou, N.; Bonitatebus, P. J., Jr.;
Miller, S. J. J. Am. Chem. Soc. 1999, 121, 11638. (f) Vedejs, E.; Daugulis,
O. J. Am. Chem. Soc. 1999, 121, 5813. (g) Sano, T.; Imai, K.; Ohashi, K.;
Oriyama, T. Chem. Lett. 1999, 265. (h) Miller, S. J.; Copeland, G. T.;
Papaioannou, N.; Horstmann, T. E.; Ruel, E. M. J. Am. Chem. Soc. 1998,
120, 1629. (i) Ruble, J. C.; Tweddell, J.; Fu, G. C. J. Org. Chem. 1998,
63, 2794. (j) Ruble, J. C.; Latham, H. A.; Fu, G. C. J. Am. Chem. Soc.
1997, 119, 1492. (k) Kawabata, T.; Nagato, M.; Takasu, K.; Fuji, K. J.
Am. Chem. Soc. 1997, 119, 3169. For epoxidation approach: (l) Martin,
V. S.; Woodward, S. S.; Katsuki, T.; Yamada, Y.; Ikeda, M.; Sharpless,
K. B. J. Am. Chem. Soc. 1981, 103, 6237. Using benzoylation: (m) Iwata,
T.; Miyake, Y.; Nishibayashi, Y.; Uemura, S. J. Chem. Soc., Perkin Trans.
1 2002, 13, 1548.
(11) For recent oxidative approaches, see: (a) Masutani, K.; Uchida, T.; Irie,
R.; Katsuki, T. Tetrahedron Lett. 2000, 41, 5119. (b) Nishibayashi, Y.;
Takei, I.; Uemura, S.; Hidai, M. Organometallics 1999, 18, 2291. (c) Gross,
Z.; Ini, S. Org Lett. 1999, 1, 2077. (d) Hashiguchi, S.; Fujii, A.; Haack,
K.-J.; Matsumura, K.; Ikariya, T.; Noyori, R. Angew. Chem., Int. Ed. Engl.
1997, 36, 288. (e) Rychnovsky, S. D.; McLernon, T. L.; Rajapakse, H. J.
Org Chem. 1996, 61, 1194.
(12) When this transformation is run in open air, significant catalyst decomposi-
tion is observed.
(13) A total synthesis of (+)-sparteine was reported recently, but it is not yet
practical, see: Smith, B. T.; Wendt, J. A.; Aube, J. Org. Lett. 2002, 4,
2577-2579.
(18) Mueller, J. A.; Jensen, D. R.; Sigman, M. S. J. Am. Chem. Soc. 2002, 124,
8202.
(19) For a study of Pd(II)-catalyzed oxidation of alcohols in DMSO with rate-
limiting catalyst turnover, see: Steinhoff, B. A.; Fix, S. R.; Stahl, S. S. J.
Am. Chem. Soc. 2002, 124, 766.
(14) This Pd(II)/(-)-sparteine-catalyzed OKR has been reported in the synthesis
of pharmaceuticals where they used a Mitsunobu inversion to access the
other needed enantiomer of alcohol, see: Ali, I. S.; Sudalai, A. Tetrahedron
Lett. 2002, 43, 5435.
(20) For mechanistic insight into the oxygenation of Pd(0)/bathocuproine, see:
Stahl, S. S.; Thorman, J. L.; Nelson, R. C.; Kozee, M. A. J. Am. Chem.
Soc. 2001, 123, 7188.
(15) Sparteine analogues have been synthesized and proven to access the other
enantiomer of product with lower krel values, see: Dearden, M. J.; Firkin,
C. R.; Hermet, J.-P. R.; O’Brien, P. J. Am. Chem. Soc. 2002, 124, 11870.
(16) Steinhoff, B. A.; Fix, S. R.; Stahl, S. S. Org. Lett. 2002, 4, 4179.
(17) ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. AdV. Synth. Catal.
2002, 344, 355.
(21) See Supporting Information for Pd((-)-sparteine)(OAc)2 X-ray crystal
analysis.
(22) Standard conditions for this experiment include 0.25 M sec-phenethyl
alcohol, 5 mol % 1, and 10 mol % base in 1 mL of 1,2-dichloroethane
(DCE) under balloon pressure of O2, at 65 °C.
9
7006 J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003

Investigations of Pd-Catalyzed Oxidations
A R T I C L E S
Scheme 1. Pd(II)-Catalyzed Oxidation of Alcohols
sparteine gives considerably higher krel values than the other
bases tested. These results reinforce the proposal that a Brønsted
base is needed for catalysis. Furthermore, the selectivity is
dependent on the nature of the exogenous base, indicating (-)-
sparteine could have an additional role as a chiral Brønsted base
in an asymmetric deprotonation.
A revised mechanism is proposed that includes roles for
(-)-sparteine as both a chiral ligand on palladium and a
chiral Brønsted base (Scheme 1). After alcohol binding to the
Pd((-)-sparteine)Cl2 catalyst (1), exogenous (-)-sparteine
deprotonates the bound alcohol (A) to form the Pd-alkoxide
species (B). The Pd-alkoxide undergoes â-hydride elimination
to release the ketone product and generate the Pd-hydride
catalytic species (C) that is recycled by oxygen.
Mechanism Elucidation. Each step of the proposed mech-
anism involves a single Pd-species. If this proposed mechanism
is plausible, a first-order dependence in Pd[(-)-sparteine]Cl2
concentration should be observed. Thus, the dependence of
oxidation rate on Pd[(-)-sparteine]Cl2 concentration was evalu-
ated from 0.5 to 5 mM (0.5-5 mol %), and a first-order
dependence was observed.23 Unless specified, for all kinetic
studies, initial rates were measured, and single enantiomers of
sec-phenethyl alcohol were utilized to avoid complications
associated with competitive binding.24
Figure 2. Rate dependence of sec-phenethyl alcohol oxidation on [(-)-
sparteine].
Table 2. Influence on Rate-Limiting Step
entry dominant term
rate expression
k1[A][catalyst]
k3[catalyst]
rate-limiting step
1
2
3
4
k3
k1[A]
alcohol binding
â-hydride elimination
k-2[BHCl] k1k3[A][catalyst]/k-2[BHCl] â-hydride elimination
none
k1k3[A][catalyst]/
â-hydride elimination
k1[A] + k3 + k-2[BHCl]
alcohol concentration. The observed saturation kinetics implicate
another step, either alcohol binding, â-hydride elimination, or
Pd(II) regeneration with O2, becomes rate limiting at high [(-)-
sparteine]. Probing the dependence on alcohol concentration
should distinguish these possibilities. A rate dependence on
alcohol concentration should be observed for either rate-limiting
alcohol binding or â-hydride elimination but not for Pd(II)
regeneration with O2. At both low and saturating (-)-sparteine
conditions,26 a first-order [alcohol] dependence was observed
within the concentration range from 0.02 to 0.2 M. These results
confirm rate-limiting alkoxide formation at low [(-)-sparteine]
and eliminate Pd(II) regeneration with O2 as rate determining
at high [(-)-sparteine].
A rate law was derived to describe the overall catalytic
process and to provide assistance in the differentiation of the
two possible rate-limiting scenarios at high [(-)-sparteine]:
alcohol binding and â-hydride elimination (eq 3, where [A] is
[alcohol] and [BHCl] is [(-)-sparteine HCl]).27 At saturation,
the denominator of the derived rate law can be simplified to
include only terms associated with k2[(-)-sparteine] (k3, k1[A],
and k-2[BHCl]). Depending on which of these terms dominate,
different mechanistic possibilities arise (Table 2). If the rate
constant associated with â-hydride elimination, k3, is large,
alcohol binding will be rate limiting at high [(-)-sparteine]
(entry 1). In contrast, â-hydride elimination becomes rate
determining at high [(-)-sparteine] for the remaining possibili-
ties (entries 2-4). If the term associated with alcohol binding,
k1[A], is dominant, the rate expression simplifies to k3[catalyst],
implying a simultaneous alcohol saturation event should be
To probe the role(s) of exogenous (-)-sparteine, the rate
dependence on the concentration of (-)-sparteine (0.002-0.069
M, 2-69 mol %) was assessed (eq 2). For oxidation of each
enantiomer of alcohol, a nonlinear relationship of rate to [(-)-
sparteine] was observed (Figure 2). The data were fit using
nonlinear least squares to an equation describing saturation with
excellent agreement.25
The first-order (-)-sparteine dependence at low [(-)-
sparteine] suggests deprotonation of the Pd-bound alcohol (A)
is rate limiting. If this is the case, according to the proposed
mechanism, a first-order dependence should be observed in
(23) See Supporting Information for the plots and data analysis.
(24) Four reactions were analyzed simultaneously under the same reaction
conditions (oxygen atmosphere and temperature fluctuations). See Sup-
porting Information for apparatus. Conditions were slowed from the
optimized synthetic conditions to enable the hand sampling of all four
reactions.
(25) Fit to: kobs ) k′[(-)-sparteine]/(k′′ + [(-)-sparteine]). For the Michaelis-
Menten equation, see: Walsh, C. Enzymatic Reaction Mechanisms; W. H.
Freeman and Company: New York, 1979.
(26) In these studies, 4 mol % (-)-sparteine is used as low base concentration
conditions, and 50 mol % is used as saturating conditions.
(27) For derivation of the rate law, see Supporting Information.
9
J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003 7007

A R T I C L E S
Mueller and Sigman
Figure 4. Benzyl alcohol oxidation Hammett plot at high [(-)-sparteine].
Figure 3. Rate inhibition by (-)-sparteine HCl.
measured at high [(-)-sparteine]. This value is consistent with
KIE values measured for other processes, implicating â-
hydride elimination as rate determining including a KIE of
1.4 ( 0.1 reported for the decomposition of trans-[Pd(CH2-
CD3)2(PMePh2)2],29,30 a KIE of 1.3 ( 0.1 recently reported by
Stahl et al. in evaluating Uemura’s Pd(II)/pyridine-catalyzed
oxidation of benzyl alcohol (at 0.1 M alcohol),16 and a KIE of
1.4 reported by Sheldon et al. for the Pd(II)/bathophenanthroline
disulfonate-catalyzed oxidation of sec-phenethyl alcohol.17
Hammett Studies. To continue to characterize the unusual
change in the rate-determining step by increasing [(-)-
sparteine], Hammett plots were constructed by measuring the
rate of oxidation of various substituted benzylic alcohols at high
and low [(-)-sparteine] (eq 5).31 These experiments were
observed upon saturation in [(-)-sparteine]. However, saturation
in alcohol was not observed over the range of alcohol
concentration evaluated. On the basis of these observations, two
situations remain in which â-hydride elimination can limit the
rate: (1) when the term associated with reprotonation of the
bound alkoxide, k-2[BHCl], is dominant or (2) when a
combination of terms dominates. In both cases, at high [(-)-
sparteine], the rate of oxidation is inversely proportional to [(-)-
sparteine HCl]. Therefore, evaluation of [(-)-sparteine HCl]
dependence can distinguish between alcohol binding and
â-hydride elimination becoming rate determining at high [(-)-
sparteine]. If alcohol binding becomes rate limiting, the addition
of (-)-sparteine HCl should not influence the reaction rate (entry
1), but if â-hydride elimination becomes rate limiting, adding
(-)-sparteine HCl should inhibit the reaction.
Reaction rates were measured with added (-)-sparteine HCl
from 0.001 to 0.012 M (1-12 mol %). A significant retardation
of rate was seen upon increase of [(-)-sparteine HCl] (Figure
3). Nonlinear least-squares analysis was used to fit the data to
a simplified derived rate law with good agreement.28 The
observed inhibition is consistent with â-hydride elimination
becoming rate limiting at high [(-)-sparteine].
Kinetic Isotope Effect. On account of only small amounts
of (-)-sparteine-HCl presumably being present under normal
reaction conditions, there is a possibility that the addition of
(-)-sparteine-HCl may perturb the reaction mechanism. Thus,
the kinetic isotope effect (KIE) at the â-position of the alcohol
was investigated to unambiguously differentiate between alcohol
binding and â-hydride elimination. If â-hydride elimination does
indeed become the rate-limiting step at high [(-)-sparteine],
substituting the â-hydrogen of sec-phenethyl alcohol (3) for
deuterium (4) and measuring kH/kD should show a primary
isotope effect. At low [(-)-sparteine], when deprotonation is
rate limiting, no kinetic isotope effect should be observed.
accomplished using in situ IR to monitor the rate of appearance
of the carbonyl absorbance of the corresponding aldehyde
products. At high [(-)-sparteine], the initial rates of product
appearance were plotted against their σ and σ+ values (Figures
4 and 5).32 The calculated F and F+ values, -1.41 ( 0.15 and
-1.00 ( 0.13, respectively, indicate that substituents influence
the rate through both induction and resonance. The negative F
value signifies a positive charge buildup in the transition state
consistent with â-hydride elimination being rate limiting at high
[(-)-sparteine]. The negative F+ value specifies that direct
resonance donation stabilizes the positive charge buildup at the
benzylic center during rate-limiting â-hydride elimination at high
[(-)-sparteine]. The same benzylic alcohols were also oxidized
at low [(-)-sparteine].33 When the Hammett plot was con-
structed with the calculated kobs values, no correlation was
(29) Ozawa, F.; Ito, T.; Yamamoto, A. J. Am. Chem. Soc. 1980, 102, 6457.
(30) KIEs on other d8 metals have been reported in which â-hydride elimination
has been implicated as the rate-determining step, see: (a) Evans, J.;
Schwarts, J.; Urquhart, P. W. J. Organomet. Chem. 1974, 81, C37. (b)
Ikariya, T.; Yamamoto, A. Organomet. Chem. 1976, 120, 257.
(31) These experiments were performed using in situ IR to measure the initial
rates of formation of the aldehyde products. The initial absorbencies were
measured and converted to concentration of aldehyde product using the
Beer-Lambert relationship. Low [(-)-sparteine] is 2 mol % for studies
using in situ IR.
(32) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 3rd ed.; Harper and Row: New York, 1987; p 144.
(33) For these studies, 2 mol % (-)-sparteine is used for low base concentration
conditions, and 50 mol % (-)-sparteine is used for saturation conditions.
Upon evaluation, a negligible kinetic isotope effect of 1.04
( 0.06 was measured at low [(-)-sparteine] (eq 4), consistent
with rate-limiting deprotonation. A KIE of 1.31 ( 0.04 was
(28) The data were fit to kobs ) k′/(k′′ + [(-)-sparteine HCl]), where k′ and k′′
are constants. See Supporting Information for the statistical analysis.
9
7008 J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003

Investigations of Pd-Catalyzed Oxidations
A R T I C L E S
Figure 5. Benzyl alcohol oxidation Hammett plot with σ+ values at high
[(-)-sparteine].
Figure 7. Eyring plots of benzyl alcohol oxidation at high and low [(-)-
sparteine].
Table 3. Activation Parameters for Benzyl Alcohol Oxidation
activation parameters
2 mol % (−)-sparteine
15 mol % (−)-sparteine
∆Gq at 60 °C
∆Hq
19.71 ( 0.94 kcal/mol
11.55 ( 0.65 kcal/mol
-24.5 ( 2.0 eu
22.04 ( 1.26 kcal/mol
20.25 ( 0.89 kcal/mol
-5.4 ( 2.7 eu
∆Sq
at low [(-)-sparteine] and rate-limiting â-hydride elimination
at high [(-)-sparteine]. Other mechanistic scenarios can be
considered for this system that may show saturation in (-)-
sparteine. One possibility entails preactivation of the alcohol
through hydrogen bonding to (-)-sparteine, followed by catalyst
coordination and deprotonation.36 For this case, saturation in
alcohol should be seen simultaneous with saturation in (-)-
sparteine. Also, if it is assumed that preactivation is rate limiting
at low [(-)-sparteine], there should be no dependence on
[Pd((-)-sparteine)Cl2]. This possible scenario is effectively ruled
out by the observation of a first-order rate dependence on
[Pd((-)-sparteine)Cl2] as well as no observed saturation in
alcohol over the concentration range evaluated.
Figure 6. Benzyl alcohol oxidation Hammett plot at low [(-)-sparteine].
observed (Figure 6).34 This illustrates at low [(-)-sparteine] that
the transition state does not change consistently with substituent
changes on the phenyl ring.
Eyring Studies. Further characterization of the rate-limiting
step as a function of [(-)-sparteine] was achieved by measuring
the activation parameters at high and low [(-)-sparteine] for
benzyl alcohol oxidation (Figure 7).35 At low concentrations of
(-)-sparteine, the ∆Hq was measured to be 11.55 ( 0.65 kcal/
mol (Table 3). This relatively low activation energy is consistent
with deprotonation being rate limiting. The ∆Sq was measured
to be -24.5 ( 2.0 eu. This large negative value is consistent
with the bimolecular nature of intermolecular deprotonation. At
high [(-)-sparteine], the ∆Hq is considerably higher at 20.25
( 0.89 kcal/mol. This increased value is congruent with rate-
limiting â-hydride elimination due to the higher energy required
to break a C-H bond. The ∆Sq at high [(-)-sparteine] was
measured to be -5.4 ( 2.7 eu, a value significantly larger than
that observed at low [(-)-sparteine]. This is consistent with the
unimolecularity of â-hydride elimination versus the bimolecu-
larity of deprotonation. It is notable that both the entropy and
the enthalpy of activation play a crucial role in [(-)-sparteine]
causing a significant change in rate-limiting events.
Additionally, two possible pathways can be considered
involving the coordination of a second (-)-sparteine onto the
catalyst that may account for saturation in (-)-sparteine. The
first scenario entails coordination of a second (-)-sparteine to
[Pd((-)-sparteine)Cl2], followed by alcohol coordination and
deprotonation to form the Pd-alkoxide. In this case, at low [(-)-
sparteine], there should be no dependence on alcohol concentra-
tion. Thus, it is effectively ruled out due to the observed alcohol
concentration dependence under these conditions.37 The second
possibility is coordination of a second (-)-sparteine after alcohol
coordination followed by intramolecular deprotonation by (-)-
sparteine. This scenario seems unlikely due to the structurally
large nature of (-)-sparteine, and upon alcohol coordination,
the resulting complex would be dicationic. Therefore, the
original mechanistic proposal (Scheme 1) is the most compatible
with kinetic data presented.
Origin of Enantioselectivity. Key considerations for the
development of new catalysts are the influence on enantiose-
lectivity of the change in rate-limiting steps from deprotonation
Other Mechanistic Possibilities. Thus far, all evidence
supports a mechanism that consists of rate-limiting deprotonation
(34) p-Nitrobenzyl alcohol formed the aldehyde initially, but then formed another
side product. Because of the determination of the rate by aldehyde
formation, this substrate was not used for the Hammett plot at low [(-)-
sparteine].
(36) (-)-Sparteine has been used to thermodynamically resolve tertiary acetylenic
alcohols, see: Toda, F.; Tanaka, K.; Ueda, H.; Oshima, T. J. Chem. Soc.,
Chem. Commun. 1983, 743.
(37) For this scenario, a second-order dependence in [(-)-sparteine] should be
observed at low [(-)-sparteine]. Additionally, NMR experiments showed
no evidence of another species other than Pd[(-)-sparteine]Cl2.
(35) For oxidations using benzyl alcohol, 15 mol % (-)-sparteine is suitable
for saturation.
9
J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003 7009

A R T I C L E S
Mueller and Sigman
Table 4. Intrinsic krel and Racemate krel Values at High and Low
[(-)-Sparteine]
(−)-sparteine
rate (R)
rate (S)
intrinsic
racemate
(mol %)
(M/min)
(M/min)
k
rel
k
rel
4
50
1.91 × 10-5
3.11 × 10-6
6.1
11
7.6 ( 2.0
25 ( 4.6
7.51 × 10-5
7.11 × 10-6
to â-hydride elimination and the overall origin of asymmetric
induction. Three steps which could possibly influence the
selectivity of the Pd-catalyzed aerobic oxidative kinetic resolu-
tion are (1) alcohol binding, (2) alkoxide formation, and/or (3)
â-hydride elimination (see Scheme 1). Catalyst turnover by
oxygen does not presumably involve the alcohol; therefore, it
should not influence selectivity. At low [(-)-sparteine], possible
influences on selectivity are thermodynamic binding and kinetic
deprotonation. At high [(-)-sparteine], thermodynamic binding,
thermodynamic deprotonation, and kinetic â-hydride elimination
can affect selectivity.
Figure 8. Reaction coordinate of oxidative kinetic resolution.
Table 5. Activation Parameters for sec-Phenethyl Alcohol
Oxidation
An apparent discrepancy is revealed when the intrinsic krel
values, calculated from the relative oxidation rates for each
enantiomer of sec-phenethyl alcohol, are compared to the krel
values calculated from oxidative kinetic resolution of the racemic
alcohol at both high and low [(-)-sparteine] (Figure 2). The
intrinsic krel values arise from asymmetric discrimination due
to the kinetic steps alone because a single enantiomer is used
in these experiments. At low [(-)-sparteine], the intrinsic krel
is 6.1, solely due to the selectivity of kinetic deprotonation
(Table 4). At high [(-)-sparteine], the intrinsic krel value is 11
and is only attributable to the selectivity of kinetic â-hydride
elimination. Unlike the intrinsic krel values, the krel values
calculated from the Pd-catalyzed oxidative kinetic resolution
of sec-phenethyl alcohol at both high and low [(-)-sparteine]
can include thermodynamic events in addition to the known
kinetic contributions (Table 4). At low [(-)-sparteine], the
racemate krel value is 7.6, within error of the intrinsic krel value
of 6.1. This implies that, at low [(-)-sparteine], deprotonation
is enantiodetermining and alcohol binding does not play a
significant role in asymmetric induction. In contrast, the
racemate krel increases to 25 at high [(-)-sparteine], a value
2.3 times higher than the intrinsic krel value of 11. This increase
in selectivity can be explained by a thermodynamic difference
in stability between the diastereomeric alkoxide species, ∆∆Gf,
formed prior to kinetic â-hydride elimination (Figure 8). A
difference of 0.52 kcal/mol between the diastereomeric alkoxides
with the R-alkoxide being favored corresponds to a selectivity
factor of 2.3. When the ∆∆Gf (0.52 kcal/mol) is added to ∆∆Gq
associated with kinetic â-hydride elimination (1.59 kcal/mol,
krel ) 11), the total difference in free energy is 2.11 kcal/mol
corresponding to the ultimately calculated krel value of 25. In
this system, the krel value is not only defined by ∆∆Gq, the
activation parameters
(R)-sec-phenethyl alcohol
(S)-sec-phenethyl alcohol
∆Gq at 60 °C
∆Hq
22.48 ( 0.64 kcal/mol
16.76 ( 0.45 kcal/mol
-17.1 ( 1.4 eu
23.77 ( 2.55 kcal/mol
14.31 ( 1.79 kcal/mol
-28.4 ( 5.4 eu
∆Sq
Table 6. Oxidative Kinetic Resolution Changing Pd Catalyst
Counterion
Pd(II) catalyst
krel value
Pd((-)-sparteine)Cl2
Pd((-)-sparteine)(OAc)2
17.5
13.0
a selective kinetic component attributable to rate-limiting
â-hydride elimination and a selective thermodynamic component
due to the difference in stability of the diastereomeric alkoxide
complexes formed.
Additionally, the activation parameters for each enantiomer
of sec-phenethyl alcohol were determined at high [(-)-sparteine]
(Table 5). Eyring plots were constructed by measuring the rates
of reaction over a temperature range from 30 to 65 °C. These
data indicate selectivity is entropically derived due to the
significant difference in ∆∆Sq of 11.3 eu. Qualitatively, the free
energy of the transition state of the R-alkoxide is smaller than
that of the S-alkoxide, consistent with the proposed model.
Unfortunately, the margins of error are too great to draw specific
conclusions.
Counterion Effect. In the development of Pd-catalyzed
oxidative kinetic resolution of secondary alcohols, it was
observed that the krel values when using Pd((-)-sparteine)Cl2
were consistently higher than those measured when using Pd-
((-)-sparteine)(OAc)2 (Table 6). The difference between these
precatalysts is the counterion: chloride versus acetate. Studying
this intriguing effect could lead to a better understanding of
the role of the counterion in oxidative kinetic resolution. Thus,
the dependence of krel on [(-)-sparteine] was investigated using
Pd[(-)-sparteine](OAc)2. A pronounced enhancement of krel was
observed as [(-)-sparteine] was increased from 0 to 50 mol %
(Figure 9). Of note is the observation of catalysis (35%
conversion in 22 h) and a measurable krel value of 1.4 in the
q
difference in ∆GRq and ∆GS , but it is also influenced by ∆∆Gf,
thus providing a supplementary source of selectivity from the
difference in the zero point energies of the diastereomeric
alkoxides. Therefore, the krel or selectivity is a combination of
∆∆Gq and ∆∆Gf in accordance with the Curtin-Hammett
principle.38,39 These experiments implicate that the optimal
selectivity is achieved at high [(-)-sparteine] when there is both
(38) Curtin, D. Y. Rec. Chem. Prog. 1954, 15, 111.
(39) For a manifestation of the Curtin-Hammett principle in which the minor
diastereomeric adduct gives rise to the major enantiomer of product, see:
Halpern, J. Science 1982, 217, 401.
9
7010 J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003

Investigations of Pd-Catalyzed Oxidations
A R T I C L E S
describing saturation with good agreement. This is reasonable
initial evidence that there are no significant changes in mech-
anism when the solvent is changed from 1,2-dichloroethane to
tert-butanol. While a precise explanation for the increase in krel
values by switching to tert-butyl alcohol solvent is unclear, a
simple solvent polarity argument may be made in which tert-
butyl alcohol more readily supports any cationic intermediates.
Conclusions
The mechanistic details for the Pd-catalyzed aerobic oxidative
kinetic resolution of secondary alcohols have been elucidated.
Saturation kinetics were observed for the dependence of
oxidation rate on (-)-sparteine concentration. At low [(-)-
sparteine], when the dependence on [(-)-sparteine] is first order,
deprotonation is proposed to be rate determining due to a first-
order dependence in both alcohol concentration and Pd((-)-
sparteine)Cl2 concentration. At saturation, when the dependence
on [(-)-sparteine] is zero order, KIE experiments, [(-)-sparteine
HCl] inhibition, and Hammett correlations support â-hydride
elimination being the rate-limiting step. These overall observa-
tions are further supported by the measurement of activation
parameters at high and low [(-)-sparteine].
Figure 9. Selectivity dependence on [(-)-sparteine] using Pd((-)-
sparteine)(OAc)2.
The origin of asymmetric induction was also elucidated under
both conditions. At low [(-)-sparteine], it was found that the
intrinsic krel was experimentally equivalent to the calculated
racemate krel, indicating that deprotonation is both enantio- and
rate determining. In contrast, the intrinsic and racemate krel
values were significantly different at high [(-)-sparteine]. This
discrepancy is proposed to arise from both selective kinetic
â-hydride elimination and a thermodynamic difference in the
stability of the diastereomeric alkoxides formed in the reaction.
A key, nonintuitive discovery from this investigation is (-)-
sparteine plays a dual role in the oxidative kinetic resolution of
secondary alcohols as a chiral ligand on palladium and as a
chiral Brønsted base. This insight has led to new catalysts for
oxidative kinetic resolution in which chiral carbene ligands
successfully replace (-)-sparteine.42 Further studies are ongoing
regarding replacement of (-)-sparteine as the chiral base using
information acquired from these investigations.
Figure 10. Rate dependence of sec-phenethyl alcohol oxidation on [(-)-
sparteine] in tert-butyl alcohol.
absence of exogenous (-)-sparteine. Recall in the initial studies
that Pd[(-)-sparteine]Cl2 was found to be incompetent without
additive (-)-sparteine. These results implicate that acetate can
compete with (-)-sparteine as a Brønsted base through a much
less selective pathway and only when significant levels of
exogenous (-)-sparteine exist can an effective kinetic resolution
result. Therefore, the difference in counterions is directly related
to their basicity given that chloride is a poor Brønsted base and
its use limits any nonselective pathways.40
Oxidative Kinetic Resolutions in tert-Butyl Alcohol Sol-
vent. Using 1,2-dichloroethane as the solvent, we found that
the oxidative kinetic resolution of aliphatic substrates, such as
3,3-dimethyl-2-butanol, proceeds with significantly lower krel
values than the oxidative kinetic resolution of benzylic sub-
strates. When the solvent is switched to tert-butyl alcohol, the
Experimental Section
General. All reagents, unless specified, were bought from com-
mercial sources and used without further purification. It was necessary
to purify (R)-(+)-sec-phenethyl alcohol purchased from Alpha Aesar
by flash chromatography (20% ethyl acetate/hexanes) and distillation.
(-)-Sparteine was prepared by dissolving (-)-sparteine hydrogensulfate
pentahydrate salt in water, adding sodium hydroxide pellets until the
pH > 10, and extracting the free-base product with ether (three times).
The extract was dried over sodium sulfate and filtered, and the solvent
was removed in vacuo. 1,2-Dichloroethane (DCE) and trimethylsilyl
chloride (TMSCl) were distilled from calcium hydride. Methanol was
distilled from potassium carbonate. Unless otherwise specified, analysis
of reactions was performed using GC equipped with a chiral column
(autosampling 6890 GC fitted with a Hewlett-Packard HP Chiral 20%
permethylated â-cyclodextrin column (0.25 µm film thickness, 30 m
length, phase ratio 320) using a 2.0 mL/min hydrogen flow rate).
Tetradecane was used as an internal standard. A 120 °C constant
temperature program for 9.5 min was used to separate (R)-(+)- from
(S)-(-)-sec-phenethyl alcohol and the internal standard. In situ IR was
used to analyze reactions for certain experiments (ASI ReactIR 1000).
A three-neck jacketed apparatus equipped with a condenser, an O2-
krel for the oxidative kinetic resolution of 3,3-dimethyl-2-butanol
increases from 7.6 to 17.4.41 Such a dramatic change in
enantioselectivity brings into question whether the solvent
change also causes a change in mechanism. Thus, the depen-
dence of the oxidation rate on [(-)-sparteine] was determined
in tert-butyl alcohol (Figure 10). As observed in 1,2-dichloro-
ethane, a nonlinear relationship of rate to [(-)-sparteine] was
obtained and fit using a nonlinear least squares to an equation
(40) Upon comparing the pKa values, we found that acetate (pKa of 12.3 in
DMSO) can function as a base much more readily than chloride (pKa )
1.8 in DMSO).
(41) Benzylic alcohols also resolve with good krel values using tert-butyl alcohol
as a solvent. Mandal, S. K.; Jensen, D. R.; Pugsley, J. P.; Sigman, M. S.
J. Org. Chem. 2003, ASAP.
(42) Jensen, D. R.; Sigman, M. S. Org. Lett. 2003, 5, 63.
9
J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003 7011

A R T I C L E S
Mueller and Sigman
filled balloon, and a Schlenk sidearm was utilized for these kinetic
experiments.
KIE Studies. Preparation of (S)-r-D-sec-Phenethyl Alcohol. To
a 100 mL round bottom flask was added 12.85 mL of methanol. To
this was added 3 mL of acetophenone (25.7 mmol, 1 equiv) by syringe.
Sodium borodeuteride, 1.19 g (28.3 mmol, 1.1 equiv), was added over
20 min while the mixture was stirred under nitrogen at 0 °C. After 2
h, the reaction was worked up by concentrating in vacuo, partitioning
between water and dichloromethane, and extracting into dichloro-
methane three times. The resulting oil was further purified by bulb-
to-bulb distillation. A 78% yield was obtained. Purity was determined
by GC equipped with a chiral column to be >99% with 93% deuterium
incorporation by GC mass spectroscopy.
Standard solutions were prepared for use in kinetic experiments. A
standard solution of (-)-sparteine was prepared by adding the desired
amount of (-)-sparteine to a 2 mL volumetric flask and diluting with
DCE to 2 mL. A 0.0123 M standard solution of Pd ((-)-sparteine)Cl2
was prepared in situ by adding 12.8 mg of Pd(MeCN)2Cl2 (1 equiv,
0.0494 mmol) and 300 µL of 0.1822 M (-)-sparteine solution (1.1
equiv, 0.0547 mmol) to a 4 mL volumetric flask. DCE was used to
dilute to 4 mL. The NMR spectrum of the catalyst solution prepared
in the described manner is identical to the NMR spectrum of
recrystallized Pd((-)-sparteine)Cl2 (see Supporting Information). Stan-
dard solutions of the alcohols were prepared by adding sec-phenethyl
alcohol and 0.1 mol equiv of tetradecane (internal standard) to a 2 mL
volumetric flask and diluting to volume with freshly distilled DCE.
The concentrations of the alcohols were determined by UV-vis
spectroscopy (HP 8452A diode array spectrophotometer).
Recrystallization of Pd((-)-sparteine)(OAc)2. In a vial were
dissolved PdOAc2 (1.0 equiv) and (-)-sparteine (1.1 equiv) in 1,2-
dichloroethane. To this was added hexanes. The solvent volume was
reduced, and the vial was placed in a jar containing hexanes. A single
yellow crystal suitable for X-ray diffraction was obtained after 1 week.
See Supporting Information for X-ray analysis.
External Base Screen. In a 5 mL round-bottom flask was dissolved
4.1 mg (0.01 mmol, 0.05 equiv) of Pd((-)-sparteine)Cl2 in 0.5 mL of
DCE. To this, 0.02 mmol, 0.1 equiv, of external base was added. The
reaction flask was attached to a condenser, and the system was
evacuated of air and refilled with oxygen from a balloon. The evacuation
and refill was repeated three times. The flask was placed in a 65 °C
constant temperature oil bath. To this, 24 µL (0.20 mmol, 1.0 equiv)
of the alcohol in 0.5 mL of DCE was added with a small amount of
tetradecane as internal standard, and a small sample was saved as a
time zero reference. An aliquot (∼0.1 mL) was taken from the reaction
via syringe and quenched by diluting in a methanol/dichloromethane
solution. Conversion was measured relative to the tetradecane internal
standard. The krel was then calculated.
Standard Kinetic Protocol. Four kinetic experiments were per-
formed simultaneously. In each 10 mL Schlenk tube, equipped with a
stir bar, the desired amounts of Pd((-)-sparteine)Cl2 (0.001 mmol, 0.01
equiv) solution and excess (-)-sparteine were added and brought to a
volume of 0.5 mL by addition of DCE. Condensers were attached to
the Schlenk tubes to prevent loss of solvent. When the apparatus was
assembled with the four Schlenk tubes attached, it was evacuated of
air by water aspiration and filled with oxygen from a balloon. This
process was repeated three times. After the four reactions were allowed
to stir for 30 min while heating in a 60 ( 0.1 °C oil bath, 0.5 mL of
sec-phenethyl alcohol (0.1 mmol, 1 equiv) solution, containing tet-
radecane as internal standard (0.01 mmol, 0.1 equiv), was syringed
into each reaction vessel. A small sample of alcohol solution was
retained as a time zero reference. Aliquots (25-50 µL) were syringed
out as the reactions progressed and quenched with 1 mL of 2% TFA
in methanol. The aliquots were analyzed by GC, and the reactions were
followed to 5-15% conversion to determine the initial rates. This
protocol was used for [Pd((-)-sparteine)Cl2] dependence, [(-)-
sparteine] dependence, [alcohol] dependence, and [(-)-sparteine HCl]
dependence.
Preparation of (-)-Sparteine HCl. To a 10 mL round-bottom flask
was added 1.5 mL of DCE. To this were added 110 mg of (-)-sparteine
(0.5 mmol, 1 equiv), 21 µL of methanol (0.55 mmol, 1.1 equiv), and
58.5 µL of trimethylsilyl chloride (0.46 mmol, 0.98 equiv). This mixture
was stirred for 5 min, and the solvent was removed in vacuo. A white
precipitate formed upon removal of the solvent (97% yield). 1H NMR
(300 MHz (CD3)2SO): δ 10.99 (s, 1H), 4.00 (dd, J ) 13.0 Hz, 2H),
3.84-2.77 (m, 10H), 2.71-1.15 (m, 14H). 13C NMR (75 MHz, (CD3)2-
SO): δ 17.61, 22.13, 23.01, 23.13, 24.26, 25.87, 28.43, 32.20, 32.26,
46.21, 52.18, 55.21, 60.10, 61.40, 65.02.
The racemic deuterated product was then subjected to an oxidative
kinetic resolution to obtain the (S)-(-)-enantiomer of R-D-sec-phenethyl
alcohol. To a 50 mL Schlenk flask was added 274 mg of Pd(OAc)2
(1.22 mmol, 0.1 equiv). To this were added 840 µL of (-)-sparteine
(3.66 mmol, 0.3 equiv) and 16 mL of DCE. The flask was fitted with
a condenser and a balloon filled with oxygen. The reaction vessel was
evacuated and filled with oxygen from the balloon three times and
allowed to stir at 60 ( 0.1 °C for 30 min. A solution of 1.5 g of R-D-
sec-phenethyl alcohol (12.2 mmol, 1 equiv) and 335 µL of tetradecane
(1.22 mmol, 0.1 equiv) in 6 mL of DCE was then added by syringe.
After 3 days, the reaction was concentrated by water aspiration and
purified by column chromatography (20% ethyl acetate/hexanes). The
deuterated alcohol was further purified by bulb-to-bulb distillation to
yield 235 mg of clear oil (15% yield). The isolated material was
determined to be >99% ee.
Standard solutions of (S)-R-D-sec-phenethyl alcohol and (S)-R-sec-
phenethyl alcohol were made up with 1 equiv alcohol and 0.1 equiv of
tetradecane in DCE. The molarity of these solutions was confirmed by
UV-vis. The absorbance was recorded at 270 nm at various concentra-
tions of sec-phenethyl alcohol to construct a calibration curve.
To each of four 10 mL Schlenk tubes was added 81 µL of 0.0123
M Pd((-)-sparteine)Cl2 solution (0.001 mmol, 0.01 equiv). To two
reaction flasks were added 19 µL of 0.2122 M (-)-sparteine solution
(0.004 mmol, 0.04 equiv) and 400 µL of freshly distilled DCE to bring
the reaction volume to 0.5 mL. To the other two Schlenk tubes were
added 235 µL of 0.2122 M (-)-sparteine solution (0.05 mmol, 0.5
equiv) and 184 µL of freshly distilled DCE to bring the reaction volume
to 0.5 mL. The four reactions were allowed to stir under oxygen at 60
( 0.1 °C for 30 min, and 500 µL of either (S)-sec-phenethyl alcohol
or (S)-R-D-sec-phenethyl alcohol solution (0.1 mmol, 1 equiv) was
added to each reaction so that both substrates were reacted at both (-)-
sparteine concentrations (4 mol %, 50 mol %). Aliquots (25-50 µL)
were syringed out as the reactions progressed, quenched with 1 mL of
2% TFA in methanol, and analyzed by GC. The disappearance of
substrate was plotted against time for each substrate at each concentra-
tion. The rate was calculated for each reaction. kD values were corrected
for incomplete deuterium incorporation43 (93%). The ratios of kH/kD
were calculated for each set of experiments. At 4 mol % (-)-sparteine,
these reactions were run in quadruplicate, while at 50 mol % (-)-
sparteine, they were run in triplicate.
Hammett Studies. To the reaction apparatus attached to the in situ
IR probe was added 133 µL of Pd((-)-sparteine)Cl2 (0.00504 mmol,
0.0168 equiv) standard solution (0.0365 M in DCE). To this were added
34 µL of (-)-sparteine (0.148 mmol, 0.5 equiv) and 533 µL of DCE
for high base conditions. A 50 ( 0.1 °C oil bath was elevated to
immerse the reaction vessel. The apparatus was evacuated of air by
water aspiration and refilled with oxygen from the balloon three times.
Next, 300 µL of alcohol solution (0.3 mmol of substrate alcohol) was
added to begin the reaction. The reactions were monitored using in
situ IR to measure the appearance of the carbonyl stretch band for each
aldehyde product. For low base conditions, the same procedure was
followed except 45 µL of 0.133 M (-)-sparteine solution (0.006 mmol,
0.02 equiv) was added with 522 µL of DCE. Rates from 5% to 7.5%
(43) Lewis, E. S.; Funderburk, L. J. Am. Chem. Soc. 1967, 89, 2322.
9
7012 J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003

Investigations of Pd-Catalyzed Oxidations
A R T I C L E S
conversion were measured. The data were plotted log kobs versus σ (or
σ+); see Supporting Information for data and Sigma Plot analysis.
Eyring Studies. Benzyl Alcohol Oxidation. To the reaction
apparatus attached to the in situ IR probe was added 133 µL of Pd-
((-)-sparteine)Cl2 (0.00504 mmol, 0.0168 equiv) standard solution
(0.0365 M in DCE). To this were added 34 µL of (-)-sparteine (0.148
mmol, 0.5 equiv) and 533 µL of DCE for high base conditions. An oil
bath, varied in temperature from 30 to 65 ( 0.1 °C, was elevated to
immerse the reaction vessel. The apparatus was evacuated of air by
water aspiration and refilled with oxygen from the balloon three times.
Next, 300 µL of alcohol solution (0.3 mmol of substrate alcohol in
DCE) was added to begin the reaction. The reactions were monitored
using in situ IR to measure the appearance of the carbonyl stretch band
for benzaldehyde at 1706.1 cm-1. For low base conditions, the same
procedure was followed except 45 µL of 0.133 M (-)-sparteine solution
(0.006 mmol, 0.02 equiv) was added with 522 µL of DCE. Rates from
5 to 7.5% conversion were measured for benzyl alcohol. The data were
plotted ln(kobs/T) versus 1/T. See Supporting Information for data and
Sigma Plot analysis.
sec-Phenethyl Alcohol Oxidation. To the apparatus was added 26
µL of Pd((-)-sparteine)Cl2 (0.001 mmol, 0.010 equiv) standard solution
(0.0365 M in DCE). To this were added 11.5 µL of (-)-sparteine (0.050
mmol, 0.5 equiv) and 662 µL of DCE for high base conditions. An oil
bath, varied in temperature from 30 to 65 ( 0.1 °C, was elevated to
immerse the reaction vessel. The apparatus was evacuated of air by
aspiration and refilled with oxygen from the balloon three times. Next,
300 µL of alcohol solution (0.3 mmol of (R)- or (S)-sec-phenethyl
alcohol in DCE) was added. The reactions were monitored using in
situ IR to measure the appearance of the carbonyl stretch band for
acetophenone at 1686.8 cm-1. Rates from 3% to 8% conversion were
measured for (R)-sec-phenethyl alcohol and (S)-sec-phenethyl alcohol.
The data were plotted ln(kobs/T) versus 1/T. See Supporting Information
for data and Sigma Plot analysis.
Oxidative Kinetic Resolutions with Pd((-)-sparteine)Cl2. To 10
mL Schlenk tubes was added 81 µL of 0.0123 M Pd((-)-sparteine)Cl2
solution (0.001 mmol, 0.01 equiv). To one reaction flask were added
7.5 µL of (-)-sparteine solution (0.004 mmol, 0.04 equiv) and 411 µL
of freshly distilled DCE to bring the reaction volume to 0.5 mL. To
the other Schlenk tube were added 97 µL of (-)-sparteine solution
(0.05 mmol, 0.5 equiv) and 322 µL of freshly distilled DCE to bring
the reaction volume to 0.5 mL. The flasks were evacuated by water
aspiration and filled with oxygen from the balloon three times. The
reactions had been stirred under oxygen at 60 ( 0.1 °C for 30 min,
and 0.5 mL of sec-phenethyl alcohol solution (0.1 mmol, 1 equiv),
containing tetradecane as internal standard, was syringed into each
reaction. Aliquots (25-50 µL) were syringed out as the reactions
progressed, quenched with 1 mL of 2% TFA in methanol, and analyzed
by GC. Samples were taken every 12 h, and relative rates were
calculated as a function of conversion and enantiomeric excess at each
time.
(OAc)2 solution (0.004 mmol, 0.05 equiv). To this were added 25 µL
of 0.326 M (-)-sparteine solution (0.00815 mmol, 0.1 equiv) and 195
µL of freshly distilled DCE to bring the reaction volume to 300 µL.
This was repeated for the three other reactions for each run, keeping
the amount of catalyst constant while varying the amount of (-)-
sparteine from 0.0 to 0.04 mmol (0.0-0.5 equiv) and adding DCE to
a total volume of 300 µL. The apparatus with flasks attached was
evacuated by water aspiration and filled with oxygen from the balloon.
This process was repeated three times. After the reactions had been
stirred under oxygen at 60 ( 0.1 °C for 30 min, 100 µL of sec-phenethyl
alcohol solution (0.08 mmol, 1 equiv with 0.008 mmol, 0.1 equiv of
tetradecane standard) was syringed into each reaction. Aliquots (25-
50 µL) were syringed out as the reactions progressed, quenched with
1 mL of 2% TFA in methanol, and analyzed by GC. Samples were
taken after 22 h, and the relative rates were calculated as a function of
conversion and enantiomeric excess.
[(-)-Sparteine] Dependence in tert-Butyl Alcohol. To the apparatus
was added 5.1 mg of Pd((-)-sparteine)Cl2 (0.0123 mmol, 0.05 equiv).
To this were added 11.5 µL of (-)-sparteine (0.05 mmol, 0.2 equiv),
958 µL of tert-butyl alcohol, and 25 mg of 3 Å activated MS. For
other reactions, [(-)-sparteine] was varied from 2.5 to 50 mol %
(0.005-0.125 mmol, 0.025-0.5 equiv), adding tert-butyl alcohol to
bring the reaction volume to 0.97 mL. A 65 ( 0.1 °C oil bath was
elevated to immerse the reaction vessel. The apparatus was evacuated
of air by water aspiration and refilled with oxygen from a balloon three
times. Next, 30 µL of alcohol (R)-sec-phenethyl alcohol (0.25 mmol)
was added. The reactions were monitored using in situ IR to measure
the appearance of the carbonyl stretch band for acetophenone at 1683.0
cm-1. Rates from 3% to 8% conversion were measured.
Acknowledgment. This work was supported by the National
Institutes of Health (NIGMS #RO1 GM63540) and supported
by a Research Innovation Award sponsored by the Research
Corporation. We would like to thank the University of Utah
Research Foundation, Merck Research, Rohm and Haas Re-
search, and Invenux Inc. for partial support of this research.
We thank the donors of the Petroleum Research Fund, admin-
istered by the ACS, for partial support of this research
(PRF#37536-G1). We thank Johnson Matthey for supplies of
various palladium salts. We also thank David R. Jensen,
Professor Joel Harris, and Professor Gary Keck for insightful
discussions. The X-ray crystal structure analysis was performed
by Atta Arif.
Supporting Information Available: Statistical analyses, rate
tables, rate law derivation, X-ray crystallography analysis, and
NMR spectra (PDF and CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
Oxidative Kinetic Resolutions with Pd((-)-sparteine)(OAc)2. To
a 10 mL Schlenk tube was added 80 µL of 0.05 M Pd((-)-sparteine)-
JA034262N
9
J. AM. CHEM. SOC. VOL. 125, NO. 23, 2003 7013