Mechanism and enantioselectivity in palladium-catalyzed conjugate addition of arylboronic acids to β-substituted cyclic enones: Insights from computation and experiment

Article abstract of DOI:10.1021/ja401713g

Enantioselective conjugate additions of arylboronic acids to β-substituted cyclic enones have been previously reported from our laboratories. Air- and moisture-tolerant conditions were achieved with a catalyst derived in situ from palladium(II) trifluoroacetate and the chiral ligand (S)-t-BuPyOx. We now report a combined experimental and computational investigation on the mechanism, the nature of the active catalyst, the origins of the enantioselectivity, and the stereoelectronic effects of the ligand and the substrates of this transformation. Enantioselectivity is controlled primarily by steric repulsions between the t-Bu group of the chiral ligand and the α-methylene hydrogens of the enone substrate in the enantiodetermining carbopalladation step. Computations indicate that the reaction occurs via formation of a cationic arylpalladium(II) species, and subsequent carbopalladation of the enone olefin forms the key carbon-carbon bond. Studies of nonlinear effects and stoichiometric and catalytic reactions of isolated (PyOx)Pd(Ph)I complexes show that a monomeric arylpalladium-ligand complex is the active species in the selectivity-determining step. The addition of water and ammonium hexafluorophosphate synergistically increases the rate of the reaction, corroborating the hypothesis that a cationic palladium species is involved in the reaction pathway. These additives also allow the reaction to be performed at 40 C and facilitate an expanded substrate scope.

Full text of DOI:10.1021/ja401713g

Subscriber access provided by University of Sydney Library
Article
Mechanism and Enantioselectivity in Palladium-Catalyzed
Conjugate Addition of Arylboronic Acids to #-Substituted
Cyclic Enones: Insights from Computation and Experiment
Jeffrey C. Holder, Lufeng Zou, Alexander N Marziale, Peng Liu, Yu
Lan, Michele Gatti, Kotaro Kikushima, K. N. Houk, and Brian M. Stoltz
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja401713g • Publication Date (Web): 12 Sep 2013
Downloaded from http://pubs.acs.org on September 16, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
Mechanism and Enantioselectivity in Palladium-Catalyzed Conju-
gate Addition of Arylboronic Acids to β-Substituted Cyclic Enones:
Insights from Computation and Experiment
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Jeffrey C. Holder,† Lufeng Zou,‡ Alexander N. Marziale,† Peng Liu,‡ Yu Lan,‡ Michele Gatti,† Kotaro
Kikushima,† K. N. Houk,*‡ and Brian M. Stoltz*†
†The Warren and Katharine Schlinger Laboratory for Chemistry and Chemical Engineering, Division of Chemistry and
Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
‡Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095, United States
.
ABSTRACT: Enantioselective conjugate additions of arylboronic acids to β-substituted cyclic enones have been reported previ-
ously from our laboratories. Air and moisture tolerant conditions were achieved with a catalyst derived in situ from palladium(II)
trifluoroacetate and the chiral ligand (S)-t-BuPyOx. We now report a combined experimental and computational investigation on
the mechanism, the nature of the active catalyst, the origins of the enantioselectivity, and the stereoelectronic effects of the ligand
and the substrates of this transformation. Enantioselectivity is controlled primarily by steric repulsions between the t-Bu group of
the chiral ligand and the α-methylene hydrogens of the enone substrate in the enantiodetermining carbopalladation step. Computa-
tions indicate that the reaction occurs via formation of a cationic arylpalladium(II) species, and subsequent carbopalladation of the
enone olefin forms the key carbon-carbon bond. Studies of non-linear effects and stoichiometric and catalytic reactions of isolated
(PyOx)Pd(Ph)I complexes show that a monomeric arylpalladium-ligand complex is the active species in the selectivity-determining
step. The addition of water and ammonium hexafluorophosphate synergistically increases the rate of the reaction, corroborating the
hypothesis that a cationic palladium species is involved in the reaction pathway. These additives also allow the reaction to be per-
formed at 40 °C and facilitate an expanded substrate scope.
mation of tertiary stereocenters,11 no conditions were amena-
ble to the synthesis of even racemic quaternary centers until
Introduction
Asymmetric conjugate addition has become a familiar re-
action manifold in the synthetic chemists’ repertoire.1 Though
seminal reports involved highly reactive organometallic nu-
cleophiles,2 systems were rapidly developed that involved
functional-group-tolerant organoboron nucleophiles. Namely,
Hayashi pioneered the use of rhodium/BINAP catalysts for the
asymmetric conjugate addition of a number of boron-derived
nucleophiles.3 As an economical alternative to the rhodium
systems, Miyaura pioneered the use of chiral palladium-
phosphine catalysts to address similar transformations4 and
Minnaard reported a palladium-catalyzed asymmetric conju-
gate addition using a catalyst formed in situ from palladium
trifluoroacetate and commercially available (S,S)-MeDuPhos.5
Lu and coworkers disclosed a dicationic, dimeric palla-
dium/bipyridine catalyst in 2010.12 However, it was not until
our recent report that a palladium-derived catalyst was em-
ployed to generate an asymmetric quaternary stereocenter via
conjugate addition chemistry.13
Scheme 1. Asymmetric conjugate addition with (S)-t-
BuPyOx ligand.
O
N
O
O
Pd(OCOCF3)2 (5 mol %)
Ligand 1 (6 mol %)
ArB(OH)2 (2 equiv.)
N
t-Bu
(S)–t-BuPyOx
Ar
ClCH2CH2Cl, 60 °C, 12 h
R
R
1
Up to 99% yield
Up to 96% ee
More recently, asymmetric conjugate addition has become
a useful strategy for the challenge of constructing asymmetric
quaternary stereocenters.6 Again, many earlier developed
methods involved highly reactive diorganozinc,7 trior-
ganoaluminum,8 and organomagnesium9 nucleophiles, how-
ever, more recently chiral rhodium/diene systems have been
shown to construct asymmetric quaternary stereocenters with
functional group-tolerant organoboron nucleophiles.10 While
rhodium systems are highly developed and exhibit a wide sub-
strate scope, the high cost of the catalyst precursors and oxy-
gen sensitivity of the reactions are undesirable. Despite pro-
gress in palladium-catalyzed conjugate additions for the for-
We employed a catalyst derived in situ from Pd(OCOCF3)2
and a chiral pyridinooxazoline (PyOx) ligand,14 (S)-t-BuPyOx
(ligand 1, Scheme 1). This catalyst facilitates the asymmetric
conjugate addition of arylboronic acids to β-substituted enones
in high yield and good enantioselectivity. Importantly, this
reaction is highly tolerant of air and moisture, and the chiral
ligand, while not yet commercially available, is easily pre-
pared.15 Initial results with the Pd/PyOx system were reported
rapidly due to concerns over competition in the field. Indeed,
recent publications prove palladium-catalyzed conjugate addi-
tion to be a burgeoning field of research.16 After the initial
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 13
disclosure, we observed that in addition to catalyzing conju- oxide for water (Figure 1b) resulted in identical yield, albeit
1
2
3
4
5
6
7
8
gate additions to 5-, 6-, and 7-membered enones, the Pd/PyOx
catalyst successfully reacted with chromones and 4-
quinolones.17 Intrigued by the broad substrate scope and op-
erational simplicity of this highly asymmetric process, we
conducted a thorough study to optimize the reaction condi-
tions, including measures to reduce the catalyst loading, lower
the reaction temperature, and further generalize the substrate
scope. We also performed mechanistic and computational
investigations toward elucidating the catalytic cycle, active
catalyst species, and the stereoelectronic effects on enantiose-
lectivity of this reaction.
with slightly depressed ee observed in ketone 3. Analysis of
ketone 3 by H NMR (Figure 1c) showed significant deute-
1
rium incorporation at the α-position of the carbonyl, even in
the presence of phenylboronic acid.18 As expected, a higher
degree of deuterium incorporation was observed in the reac-
tion where phenyl boroxine was substituted for the boronic
acid, however, the similar level of incorporation in both ex-
periments suggested that the deuterium oxide was the agent
assisting reaction turnover regardless of the use of protic or
aprotic boron reagent.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
(S)-t-BuPyOx (6 mol %)
Pd(OCOCF3)2 (5 mol %)
NH4PF6 (30 mol %)
(a)
O
D
B(OH)2
Results:
ClCH2CH2Cl, 60 °C, 12 h
D2O
(96% yield)
Ph
2
3
1. Effects of Water on Catalyst Turnover
In our initial report,13 we were able to demonstrate that the
addition of up to 10 equivalents of water had no deleterious
effect. Despite this, water was not considered as an important
additive in the initially reported conditions because the
stoichiometric arylboronic acid was believed to be a sufficient
proton source to turn over the catalyst. In considering the
overall reaction scheme, a more precise analysis of the mass
balance of the reaction led us to reconsider the importance of
water as an participant in the overall transformation (Scheme
2a).
O
(S)-t-BuPyOx (6 mol %)
Pd(OCOCF3)2 (5 mol %)
NH4PF6 (30 mol %)
O
(b)
(c)
Ph
O
Ph
D
B
O
B
O
B
ClCH2CH2Cl, 60 °C, 12 h
D2O
(99% yield)
Ph
Ph
2
3
O
H3
integral relative to H5
H1
entry
H2 H3 + H4 H5
H1
H5
boronic acid 0.67 0.52
0.88 1.00
0.66 1.00
1.67 1.00
H4
Ph
boroxine
control
0.52 0.41
0.94 0.90
H2
Figure 1. (a) deuterium incorporation using PhB(OH)2.
(b) deuterium incorporation using (PhBO)3. (c) H NMR data
measuring deuterium incorporation by integral comparison of
α-protons relative to H5, control: treatment of ketone 3 to
deuterium incorporation conditions.
These considerations proved to be essential during the
scale up of the reaction. Attempts to use the original condi-
tions (with no water added) failed to convert enone 2 effi-
ciently, generating the desired ketone (3) only in moderate
yield (Scheme 2b). We reasoned that when the reaction is per-
formed on a small scale under ambient atmosphere the mois-
ture present in the air and on the glassware could be sufficient
to drive the reaction to completion. On a larger scale, how-
ever, where a more significant quantity of water was neces-
sary, this was no longer true. Gratifyingly, upon the addition
of as little as 1.5 equivalents of water to the reaction mixture,
both reactivity and the enantioselectivity were restored
(Scheme 2c), affording ketone 3 in high yield and ee.
1
2. Effects of Salt Additives on Reaction Rate
Satisfied with our ability to perform the reaction on scale,
we turned our attention toward improving the catalyst activity.
We observed that nearly all previous literature reports regard-
ing palladium-catalyzed conjugate addition utilized cationic
–
precatalysts featuring weakly-coordinative anions (PF6 ,
–
–
SbF6 , BF4 etc.). We reasoned that the substitution of the
trifluoroacetate counterion with a less coordinative species
could lead to an increase in reaction rate. With this goal in
mind, we examined a series of salt additives containing
weakly coordinative counterions. We viewed the strategy for
the in situ generation of the catalyst as the more practical and
operationally simple alternative to the design, synthesis, and
isolation of a new dicationic palladium precatalyst.
Scheme 2. (a) Examination of reaction mass balance. (b)
Absence of water prohibits scale-up. (c) Addition of water
facilitates larger scale reactions.
(a)
O
O
H
PhB(OH)2
H2O
HO B(OH)2
Ph
2
3
We investigated a number of salt additives to test this
mechanistic hypothesis (Table 1). Coordinating counterions
like chloride (entry 1) shut down reactivity. Pleasingly, as per
our hypothesis, weakly coordinating counterions with sodium
cations (entries 2–4) facilitated swift reaction, albeit with de-
pressed ee. Tetrabutylammonium salts (entries 5–6) encoun-
tered slow reaction times, but good enantioselectivity. So-
dium tetraphenylborate (entry 7), however, failed to deliver
appreciable quantities of the quaternary ketone 3, as rapid
formation of biphenyl was observed. Ammonium salts (en-
tires 8–9) provided the desired blend of reaction rate and
enentioselectivity. We concluded that the hexafluorophosphate
anion (entry 9) gave the optimal combination of short reaction
time with minimized loss of enantioselectivity.
(S)-t-BuPyOx (6 mol %)
Pd(OCOCF3)2 (5 mol %)
PhB(OH)2 (2 equiv)
O
O
(b)
Incomplete Conversion
on 1 g scale (9 mmol)
ClCH2CH2Cl, 60 °C, 12 h
No Added H2O
Ph
2
3
(S)-t-BuPyOx (6 mol %)
Pd(OCOCF3)2 (5 mol %)
PhB(OH)2 (2 equiv)
O
O
(c)
22 mmol scale
97% yield, 91% ee
ClCH2CH2Cl, 60 °C, 12 h
5 equiv H2O
Ph
2
3
We next sought to measure deuterium incorporation at the
carbonyl α-position as a method to determine the source of the
proton utilized in reaction turnover. Reactions were per-
formed substituting deuterium oxide for water and observed
by 1H and 2H NMR analysis (Figure 1). Using phenylboronic
acid, the reaction afforded ketone 3 in similar yield and enan-
tioselectivity (Figure 1a). Likewise, substitution of phenyl-
boroxine ((PhBO)3) for phenylboronic acid and deuterium
2
ACS Paragon Plus Environment

Page 3 of 13
Journal of the American Chemical Society
Table 1. Effect of salt additives on reaction rate.a
Table 2. Effect of water and NH4PF6 on reaction rate.
1
2
O
O
(S)-t-BuPyOX, PhB(OH)2
(S)-t-BuPyOx, PhB(OH)2
O
O
Pd(OCOCF3)2
Pd(OCOCF3)2
3
4
5
6
7
8
9
H2O, NH4PF6
ClCH2CH2Cl
additive
ClCH2CH2Cl, 40 °C
Ph
Ph
2
3
2
3
H2O
(equiv.)
NH4PF6
(mol %)
ligand
(mol %)
temp
(°C)
Pd
(mol %)
entrya
1
yield (%)b
ee (%)c
entrya
time (h)
yield (%)b
ee (%)c
additive
time (h)
trace
81d
97
---
1
2
3
24
8
NaCl
10
---
60
60
25
40
40
12
1.5
36
12
12
99
93
98
95
96
91
88
91
91
89
5
5
6
6
6
3
3
NaBF4
88
2
3
4
5
6
7
8
9
5
5
30
30
30
30
NaPF6
5
6
5
87
81
90
4
5
2.5
2.5
NaSbF6
n-(Bu)4PF6
n-(Bu)4BF4
NaBPh4
99
98
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
5
6
24
20
5
30
5
2.5
2.5
2.5
2.5
3
3
3
3
40
40
40
40
12
24
24
12
95
93
93
97
90
90
92
88
95
88
---
89
91
24
24
15
7
8
9
trace
93
5
10
100
NH4BF4
5
NH4PF6
96
12
aConditions: phenylboronic acid (1.0 mmol), 3-
methylcyclohexen-2-one (0.5 mmol), water, NH4PF6,
aConditions: phenylboronic acid (0.5 mmol), 3-
methylcyclohexen-2-one (0.25 mmol), water (5 equiv), additive
(30 mol %), Pd(OCOCF3)2 (5 mol %) and (S)-t-BuPyOX (6 mol
%) in ClCH2CH2Cl (1 mL) at 40 °C . bGC yield utilizng tridecane
standard. cee was determined by chiral HPLC. dReaction checked
at 83% conversion as determined by GC analysis.
b
Pd(OCOCF3)2 and (S)-t-BuPyOx in ClCH2CH2Cl (2 mL). Iso-
lated yield. cee was determined by chiral HPLC.
Though increased rates were observed at 60 °C, the newly-
found ability to perform reactions at 40 °C promoted superior
reactivity of many substrates (Table 3). In fact, many sub-
strates that exhibited high enantioselectivities under the origi-
nal 60 °C reaction conditions suffered from poor yields. Re-
acting these substrates at 40 °C with the addition of ammo-
nium hexafluorophosphate and water promoted significantly
higher isolated yields. Arylboronic acids containing halides,
such as m-chloro- (4a) and m-bromophenylboronic acid (4b)
reacted with good enantioselectivity, but each substrate was
originally marred by low yield using our original conditions.
However, when reacted under the newly optimized reaction
conditions, the isolated yield for the addition of chlorophenyl-
boronic acid increased from 55% to 96% and for bromophen-
ylboronic acid from 44% to 86%. Even m-nitroboronic acid
(4c) reacted with higher isolated yield. Notably, some ortho-
substituted boronic acids, such as o-fluorophenylboronic acid
(4d), reacted more successfully under the milder reaction con-
ditions, leading to increased isolated yield of 70%.
Based on our previous observations regarding the benefi-
cial nature of water as an additive, we next explored the com-
bined effect of water and hexafluorophosphate counterions.
We found addition of both water and ammonium hexafluoro-
phosphate were the most successful for increasing reactivity
(Table 2). Water alone is insufficient to alter reactivity (entry
1), though the use of water with 30 mol % ammonium hex-
afluorophosphate greatly reduced the reaction time (entry 2) to
only 1.5 hours with minimal effect on yield or ee. Further-
more, this combination of additives allowed the reaction to
proceed at temperatures as low as 25 °C with 5 mol % palla-
dium and 6 mol % ligand, and lowering of catalyst loadings to
only 2.5 mol % of palladium and 3 mol % ligand at 40 °C
(entry 3). We determined that optimal conditions for the reac-
tion with lower catalyst loading to be 5 equivalents of water,
30 mol % ammonium hexafluorophosphate at 40 °C (entry 4),
conditions that reproduce the original result at milder tem-
perature and lower catalyst loadings. The reaction was ex-
traordinarily tolerant of the amount of water, with both 10
(entry 5) and 20 (entry 6) equivalents of water having minimal
effect on the yield or ee. Loadings of ammonium hexafluro-
phosphate can be as low as 5 mol % (entry 7) or 10 mol %
(entry 8) with reactions completed in 24 hours. Stoichiometric
additive (entry 9) gave no additional benefit (entry 4). Thus,
we optimized the additive amounts to be 30 mol % ammonium
hexafluorophosphate and 5 equivalents of water.
Table 3. Increased reaction yields with different arylboronic
acid substrates under new reaction conditions.a
O
(S)-t-BuPyOx (6 mol %)
Pd(OCOCF3)3 (5 mol %)
NH4PF6, H2O
O
R
R
ClCH2CH2Cl
40 °C, 12 h
B(OH)2
2
4
5
(HO)2B
Cl
(HO)2B
Br
(HO)2B
NO2
(HO)2B
F
4a
55 % Yield 97 % ee
4b
44 % Yield 86 % ee
4c
40 % Yield 92 % ee
4d
32 % Yield 77 % ee
96 % Yield 96 % ee
84 % Yield 84 % ee
81 % Yield 91 % ee
70 % Yield 77 % ee
aBlue font: reported yield and ee of 5 in the absence of NH4PF6
and water with reactions performed at 60 °C; red font: yield and
ee of 5 with additives. Conditions: boronic acid (1.0 mmol), 3-
methylcyclohexen-2-one (0.5 mmol), NH4PF6 (30 mol %), water
(5 equiv.), Pd(OCOCF3)2 (5 mol %) and (S)-t-BuPyOx (6 mol %)
in ClCH2CH2Cl (2 mL) at 40 °C . Isolated yield reported, ee was
determined by chiral HPLC.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 13
plexes are the active species in the catalytic cycle. The car-
3. Non-Linear Effect Correlation of Catalyst and Prod-
uct Enantioenrichment
1
2
3
4
5
6
7
8
bopalladation is calculated to be the rate- and stereoselectiv-
ity-determining step (Scheme 3). Now, we have performed
DFT investigations on the catalytic cycle of reactions with the
Pd/PyOx manifold and the effects of substituents and ligand
on reactivity and enantioselectivity. The calculations were
performed at the theoretical level found satisfactory in our
previous study of the Pd/bipyridine system. Geometries were
optimized with BP8628 and a standard 6-31G(d) basis set
(SDD basis set for palladium). Solvent effects were calculated
with single point calculations on the gas phase geometries
with the CPCM solvation model in dichloroethane. All calcu-
lations were performed with Gaussian 03.29
Despite optimization of catalytic conditions for this highly
enantioselective process, we were unsure of the nature of the
active catalyst. For example, some rhodium conjugate addi-
tion systems have been shown to involve trimeric ligand/metal
complexes.19 Furthermore, Lu and coworkers reported the use
of the palladium dimer [(bpy)Pd(OH)]2•2BF4 as a precatalyst
for conjugate addition.12 We aimed to rule out the in situ for-
mation and kinetic relevance of such dimers in our system. In
seeking to support our hypothesized monomeric ligand-metal
complex, we performed a non-linear effect study to determine
the relationship between the ee of the ligand and the ee of the
generated product.20 The endeavor was to exclude dimeric
(ML)2 complexes from kinetic relevance, clarifying the
monomeric nature of the active catalyst.21 Five reactions were
performed using a catalyst with different level of enantiopurity
(racemic, 20%, 40%, 60% and 80% ee), and the obtained
enantioselectivities were plotted against ee of the catalyst
mixture (Figure 2). The obtained data clearly demonstrates
that a non-linear effect is not present, and this observation
strongly supports the action of a single, monomeric (ML)-type
Pd/PyOx catalyst as the kinetically relevant species.21 While
the precise nature of the active catalyst species is unknown,
isolated (PyOx)Pd(OCOCF3)2 serves as an identically useful
precatylst, delivering ketone 3 in 99% yield and 92% ee.22
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Enantioselectivity-determining step in asymmetric
conjugate addition of arylboronic acids to cyclic enones.
O
N
N
t-Bu
O
O
1
Pd(OCOCF3)2
B(OH)2
ClCH2CH2Cl, 60 °C
2
3
O
O
H
N
N
Pd
Enantioselectivity-Determining Step
The computed potential energy surface for the catalytic cy-
cle is shown in Figure 3. To simplify the computations of the
mechanisms, a model ligand, in which the t-Bu group on the t-
BuPyOx ligand was replaced by H, was used in the calcula-
tions of the mechanisms and the full ligand was used in the
calculations of enantioselectivities which will be discussed
below. Calculations on the reaction mechanism with the full
ligand scaffold, however, generated a similar reaction dia-
gram, and the rate- and stereo-determining steps were un-
changed (see Figure S4 in the Supporting Information). The
first step involves transmetallation of cationic Pd(II)-
phenylborate complex 6 to generate a phenyl palladium com-
plex. Transmetallation requires a relatively low free energy
barrier of 15.6 kcal/mol (7-ts) with respect to complex 6 and
leads to a phenyl palladium complex (8). Complex 8 under-
goes ligand exchange to form a more stable phenyl palladium-
enone complex 12, in which the palladium binds to the enone
oxygen atom. Complex 12 isomerizes to a less stable π com-
plex 13 and then undergoes carbopalladation of the enone (14-
ts) to form the new carbon–carbon bond. The carbopalladation
step requires an activation free energy of 21.3 kcal/mol (12 →
14-ts), and is the stereoselectivity-determining step. The re-
gioisomeric carbopalladation transition state 16-ts requires 5.6
kcal/mol higher activation free energy than 14-ts, indicating
the formation of the α-addition compound 17 is unlikely to
occur. Coordination of one water molecule to 15 leads to a
water-palladium enolate complex 18, and finally facile hy-
Figure 2. Determination of linearity between catalyst ee and
product ee.
4. Computational Investigations of the Reaction
Mechanism
Despite the results of the non-linear effect study agreeing
with the proposed monomeric Pd/PyOx catalyst, no formal
exploration of the mechanism of this transformation has been
reported. Our initial hypothesis concerning the mechanism of
the Pd/PyOx-catalyzed asymmetric conjugate addition were
well informed by the seminal work of Miyaura,23 however, the
heterogeneous nature of the reaction medium, undefined na-
ture of the precise catalyst,24 and complicating equilibrium of
organoboron species make kinetic analysis and thorough
mechanistic study extremely challenging.25
Previously, we performed density functional theory (DFT)
calculations to investigate the mechanism of palladium-
catalyzed conjugate addition of arylboronic acids to enones,
explicitly studying a catalytic palladium(II)/bipyridine system
,
in MeOH solvent similar to that developed by Lu.12
Cal-
,26 27
culations indicated that the mechanism involves three steps:
transmetallation, carbopalladation (i.e. alkene insertion), and
protonation with MeOH. Monomeric cationic palladium com-
4
ACS Paragon Plus Environment

Page 5 of 13
Journal of the American Chemical Society
drolysis of 18 via 19-ts affords product complex 20. Libera- We also considered an alternative pathway involving direct
1
2
3
4
5
6
7
8
tion of the product 3 from 20 and coordination with another
molecule of phenyl boronic acid regenerates complex 6 to
complete the catalytic cycle. The computed catalytic cycle
demonstrates some similarities with the Pd/bipyridine system
in our previous computational investigation, which also in-
volves monomeric cationic palladium as the active species and
a catalytic cycle of transmetallation, carbopalladation, and
protonation (with MeOH instead of H2O).
nucleophilic attack of the phenyl boronic acid at the enone
while the Pd catalyst is acting as a Lewis acid to activate the
enone and directs the attack of the nucleophile (9-ts, Figure 3).
This alternative pathway requires an activation free energy of
58.9 kcal/mol, 43.3 kcal/mol higher than the transmetallation
transition state 7-ts. Thus, this alternative pathway was ex-
cluded by calculations.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
N
N
Pd
HO
O
ΔGsol
(ΔHsol
HO
OH
)
B
kcal/mol
58.9
(42.6)
9-ts
9-ts
9-ts
O
N
N
19-ts
16-ts
14-ts
Pd
B
HO
OH
OH
O
N
O
N
O
N
O
N
15.6
(14.4)
N
O
N
N
N
Pd
Pd
Pd
7-ts
7-ts
O
Pd
H
O
1.3
(-0.4)
0.8
(-12.9)
0.0
H
(0.0)
7-ts
O
16-ts
-4.3
(-6.8)
O
O
O
O
Ph
19-ts
10
6
14-ts
-11.0
(-11.3)
N
N
17
-9.0
(5.1)
16-ts
-10.6
Pd
-12.4
(-24.0)
B(OH)3
-13.9
(-15.3)
O
N
HO
(-23.7)
O
14-ts
Ph
11
H2O
HO
B
-17.4
(-15.3)
19-ts
13
N
2
18
3
17
-19.0
(-21.2)
HO
O
-22.3
(-35.0)
Pd
B
O
N
-22.7
O
N
8
-25.6
HO
HO
OH
N
N
(-33.2)
15
(-24.2)
O
N
6
N
Pd
N
10
20
O
12
Pd
B(OH)2
Pd
O
N
O
H
N
O
O
O
Pd
6
H
N
N
N
O
11
N
N
O
Pd
Pd
Ph
13
HO
Pd
OH
Ph
18
O
B
OH
15
HO
8
Ph
20
12
Figure 3. Computed potential energy surface of the catalytic cycle (shown in black), the alternative direct nucleophilic addition
pathway (via 9-ts, shown in blue), and the isomeric carbopalladation pathway (via 16-ts, shown in green).
Bu on the ligand and the phenyl group, as indicated by the C-
H and C-C distances labeled in Figure 4. 1-TS-B and 1-TS-D
lead to the minor (S)-product, which are ~ 3 kcal/mol less
stable than 1-TS-A as a result of the repulsions between the t-
Bu on the ligand and the phenyl group. In 1-TS-B, the cyclo-
hexenone ring is syn to the t-Bu group on the ligand. The
shortest H–H distance between the ligand and the enone is
2.30 Å, suggesting some steric repulsions. In contrast, no
ligand-substrate steric repulsions are observed in 1-TS-A, in
which the cyclohexenone is anti to the t-Bu. 1-TS-A is also
stabilized by a weak hydrogen bond between the carbonyl
oxygen and the hydrogen geminal to the t-Bu group on the
oxazoline. The O–H distance is 2.16Å. Therefore, the enthalpy
of 1-TS-B is 2.3 kcal/mol higher than that of 1-TS-A. This
corresponds to an ee of 94%, which is very similar to the ex-
perimental observation (93%). Enantioselectivities were com-
puted from relative enthalpies of the transition states. The
selectivities computed from Gibbs free energies are very
similar and are given in the SI.
5. Experimental and Computational Investigations of
the Enantioselectivities
With the aforementioned optimized reaction conditions and
computational elucidation of the mechanism and stereoselec-
tivity-determining transition states, we explored the effects of
ligand and substrate on enantioselectivities by both experiment
and computations. The enantioselectivity-determining alkene
insertion step involves a four-membered cyclic transition state,
which adopts a square-planar geometry. When a chiral biden-
tate ligand, such as (S)-t-BuPyOx, is employed, there are four
possible isomeric alkene insertion transition states. The 3D
structures of the alkene insertion transition states in the reac-
tion of 3-methyl-2-cyclohexenone with (S)–t-BuPyOx ligand
are shown in Figure 4. In 1-TS-A and 1-TS-B, the phenyl
group is trans to the chiral oxazoline on the ligand, and in 1-
TS-C and 1-TS-D, the phenyl group is cis to the oxazoline. 1-
TS-A, which leads to the predorminant (R)-product, is the
most stable as the t-Bu group is pointing away from other
bulky groups. 1-TS-C leads to the same enantiomer, but with
an activation enthalpy 2.6 kcal/mol higher than 1-TS-A. The
difference is likely to result from steric effects between the t-
We then investigated the effects of substituents on the
ligand, in particular, at the 4 position of the oxazoline. The
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 13
activation enthalpies of four alkene insertion pathways and the Similarly, when the (S)-i-BuPyOx or (S)-PhPyOx ligands
1
2
3
4
5
6
7
8
computed and experimental ee for the reaction of 3-methyl-2-
cyclohexenone and phenyl boronic acid are summarized in
Table 4. The t-Bu substituted PyOx ligand is found to be the
optimum ligand experimentally (Table 4, entry 1). Replacing
t-Bu with smaller groups, such as i-Pr, i-Bu, or Ph, dramati-
cally reduces the ee.
are used, the enantioselectivity is further decreased to 0.8
kcal/mol (52% ee) for (S)-i-BuPyOx and 1.0 kcal/mol (65%
ee) for (S)-PhPyOx. (Table 4, entries 3 and 4). These results
agree well with the experimental trend.
Table 4. Activation energies and enantioselectivities of
alkene insertion with (S)-t-BuPyOX, (S)-i-PrPyOX, (S)-i-
BuPyOX, and (S)-PhPyOx ligands.
The bulky t-Bu substituent on the ligand is essential not
only to discriminate the diastereomeric transition states 1-TS-
A and 1-TS-B, but also fix the orientation of the ligand to
point the chiral center cis to the cyclohexenone. The energy
difference between 1-TS-C and 1-TS-D, in which the chiral
center on the ligand is trans to the cyclohexenone, is dimin-
ished.
9
O
O
O
O
O
O
H
H
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H
N
N
N
N
N
N
N
N
Pd
Pd
Pd
R1
O
R1
R1
Pd
O
R1
n-TS-A
n-TS-B
n-TS-C
n-TS-D
ΔH‡ a
ee b
TS
R1
TS-A
19.2
TS-B
21.5
TS-C
TS-D
94%
[93%]
67%
[40%]
52%
[24%]
65%
[52%]
t-Bu
i-Pr
i-Bu
Ph
1
2
3
4
21.8
22.2
18.6
18.5
18.0
19.7
19.3
19.0
20.1
20.4
19.6
20.0
20.4
20.2
1-TS-A
ΔH‡ = 19.2 kcal/mol
1-TS-B
ΔH‡ = 21.5 kcal/mol
aThe values are activation enthalpies in kcal/mol calculated at the
BP86/6-31G(d)–SDD level and the CPCM solvation model in
b
dichloroethane. Experimental ee were obtained under standard
conditions and are given in square brackets.
Electronically differentiated PyOx ligands were also stud-
ied, and the results are summarized in Table 5. Electron-
withdrawing or donating groups at the 4-position of the PyOx
ligand showed minimal effects on the activation barriers, and
were calculated to have minimal effect on product ee. With
the electron-withdrawing CF3 and the electron-donating OCH3
on the 4-position of the ligand, the activation enthalpies of
alkene insertion increase by only 0.3 kcal/mol and 0.1
kcal/mol, respectively. The calculated ee are essentially iden-
tical among these three ligands. Experimentally, depressed ee
was observed with both the 4-CF3 and the 4-OCH3 substituted
ligands (entries 5 and 6). This confirms that the enantioselec-
tivity is mainly attributed to the ligand/substrate steric repul-
sions.
1-TS-C
ΔH‡ = 21.8 kcal/mol
1-TS-D
ΔH‡ = 22.2 kcal/mol
Figure 4. The optimized geometries of transition states in the
enantioselectivity-determining alkene insertion step of the
reaction of 3-methyl-2-cyclohexenone and phenyl boronic
acid with (S)–t-BuPyOx ligand. Selected H–H, C–H, and O–H
distances are labeled in Å.
When the (S)-i-PrPyOx ligand is used, the alkene insertion
transition states with phenyl trans to the oxazoline (2-TS-A
and 2-TS-B) are also preferred. Thus, the enantioselectivity is
determined by the energy difference between 2-TS-A and 2-
TS-B. The (R)-product (via 2-TS-A) is favored with a com-
puted ee of 67%, slightly higher than the experimental ee
(40%). The optimized geometries of 2-TS-A and 2-TS-B are
shown in Figure 5 and the activation energies of all four tran-
sition states are shown in Table 4, entry 2. The i-Pr substituted
ligand manifests via similar steric effects to (S)-t-BuPyOx,
with, as expected, slightly weaker steric control. The lower
enantioselectivity is attributed to the weaker steric repulsions
between the i-Pr and the cyclohexenone in 2-TS-B than those
with the t-Bu in 1-TS-B. The shortest distance between the
hydrogen atoms on the ligand and the cyclohexenone is 2.35 Å
in 2-TS-B, slightly longer than the H–H distance in 1-TS-B
(2.30 Å). Less steric repulsions with the (S)-i-PrPyOx lead to
2.8 kcal/mol lower activation barriers for 2-TS-B compared to
1-TS-B. The ligand steric effects on the activation energies of
the major pathway TS-A are smaller; the i-Pr substituted 2-
TS-A is only 0.6 kcal/mol more stable than the t-Bu substi-
tuted 1-TS-A.
Table 5. Remote ligand substituent effects on activation en-
ergies and enantioselectivities of alkene insertion.
R2
R2
H
R2
R2
O
O
O
O
O
O
H
H
N
N
N
N
N
N
H
N
N
Pd
Pd
Pd
Pd
O
O
n-TS-A
n-TS-B
n-TS-C
n-TS-D
ΔH‡ a
ee b
TS
R2
H
TS-A
TS-B
21.5
TS-C
TS-D
94%
[93%]
93%
[81%]
93%
[78%]
1
5
6
19.2
19.5
19.3
21.8
21.8
21.5
22.2
CF3
OCH3
21.7
21.5
22.3
22.2
aThe values are activation enthalpies in kcal/mol calculated at the
BP86/6-31G(d)-SDD level and the CPCM solvation model in
dichloroethane. bExperimental ee are given in square brackets.
6
ACS Paragon Plus Environment

Page 7 of 13
Journal of the American Chemical Society
Table 6. Activation energies and enantioselectivities of
alkene insertion with substrates varying at the α'-position.
(a)
1
2
3
4
5
6
7
8
9
O
O
O
O
X
O
O
X
H
H
H
N
N
N
N
N
N
H
N
N
Pd
Pd
Pd
Pd
O
O
X
X
n-TS-A
n-TS-B
n-TS-C
n-TS-D
2-TS-A
ΔH‡ = 18.6 kcal/mol
2-TS-B
ΔH‡ = 19.7 kcal/mol
ΔH‡ a
ee b
TS
X
TS-A
19.2
TS-B
21.5
TS-C
TS-D
(b)
10
94%
[93%]
88%
[57%]
99%
CH2
O
1
7
8
21.8
22.2
11
12
13
14
15
16
17
18
19
16.7
17.9
18.6
19.4
20.0
19.6
C(CH3)2
22.0
20.8
[90%]
aThe values are activation enthalpies in kcal/mol calculated at the
BP86/6-31G(d)-SDD level and the CPCM solvation model in
dichloroethane. bExperimental ee are given in square brackets.
Experimental isolated yields: TS-1: 99%, TS-7: 49%, TS-8: 9%.
7-TS-A
ΔH‡ = 16.7 kcal/mol
7-TS-B
ΔH‡ = 18.6 kcal/mol
Figure 5. The optimized geometries of transition states in the
enantioselectivity-determining alkene insertion step of the reac-
tion of (a) 3-methyl-2-cyclohexenone and phenyl boronic acid
with (S)-i-PrPyOx ligand; (b) 3-methyl-δ-2-pentenolide and
phenyl boronic acid with (S)-t-BuPyOx ligand. Selected H–H,
C–H, and O–H distances are labeled in Å.
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We also considered the electronic effects of arylboronic
acids on enantioselectivity (Table 7). Computations predicted
that para-electron-withdrawing substituents lead to increases
in the activation barrier in alkene insertion, probably due to
the electrophilicity of the β-carbon of the enone, and thus are
predicted to afford slightly decreased enantioselectivities.
Both para-acetylphenylboronic acid (9-TS-A) and para-
trifluoromethylphenylboronic acid (10-TS-A) are predicted to
react with 92% ee. However, both excellent enantioselectiv-
ities (96% ee) and excellent yields (99% isolated yield) are
observed experimentally. Thus, the electronic effects of
phenyl substituents on enantioselectivities are minimal, though
slightly increased enantioselectivities are observed experi-
mentally with the use of electron-withdrawing substituents.
The transition state structures shown in Figure 5 indicate
that the steric control mainly arises from the repulsion of the
Cα hydrogens on the cyclohexenone with the ligand. We then
'
investigated the effects of substitution at the α' position and
replacement of the CH2 group with O. The reactivity and
enantioselectivity of the reactions of lactone (Table 6, entry 7)
and α',α'-dimethylcyclohexenone (Table 6, entry 8) with the
(S)-t-BuPyOx ligand were computed. The enantioselectivity of
lactone is predicted to be lower than cyclohexenone. The en-
thalpy of 7-TS-B is 1.8 kcal/mol higher than that of 7-TS-A,
corresponding to an ee of 88%. Experimentally, the ee of the
lactone product is 59%, also significantly lower than that with
cyclohexenone. The optimized geometries of 7-TS-A and 7-
TS-B are shown in Figure 5b. Replacing the CH2 group with
O decreases the ligand–substrate steric repulsion in 7-TS-B is
smaller than that in 1-TS-B. This results in decreased enanti-
oselectivity.
Table 7. Activation energies and enantioselectivities of
alkene insertion with various boronic acids.
O
O
O
O
O
O
H
H
H
N
N
N
N
N
N
H
N
N
Pd
Pd
Pd
Pd
O
O
R3
R3
R3
R3
n-TS-A
n-TS-B
n-TS-C
n-TS-D
Methyl substitution at the α' position of cyclohexenone in-
creases the steric repulsion with the t-Bu group on the ligand.
Computations predicted increased enenatioselectivity with
α',α'−dimethylcyclohexenone (99% ee, Table 6, entry 8).30
However, experimentally, the ee is comparable with the reac-
tion of 3-methylcyclohexenone.
ΔH‡ a
ee b
TS
R3
H
TS-A
19.2
TS-B
TS-C
TS-D
94%
[93%]
92%
[96%]
92%
1
9
21.5
21.8
22.2
CH3CO 21.6
CF3 21.3
23.7
23.4
24.5
23.6
24.8
24.2
10
[96%]
aThe values are activation enthalpies in kcal/mol calculated at the
BP86/6-31G(d)-SDD level and the CPCM solvation model in
dichloroethane. Experimental ee are given in square brackets.
Experimental isolated yields: TS-1: 99%, TS-9: 1%, TS-10: 99%,
TS-11: 99%.
b
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 13
Table 8. Enantioselectivity trends in pyridinooxazoline and
6. Experimental Investigation of Arylpalladium(II) In-
termediates and Formation of the Key C–C Bond
related ligand frameworks.a
1
2
3
4
5
6
7
8
Experiments aiming to corroborate the calculated mecha-
nism have been performed. We sought to observe the forma-
tion of the key C–C bond between an arylpalladium(II) species
and the enone substrate in the absence of exogenous phenylbo-
ronic acid. Complexes 29 were synthesized as an intractable
mixture of isomers, and were treated with AgPF6 in situ to
generate the [(PyOx)Pd(Ph)]+ cation. Gratifyingly, complexes
29 serve as a competent precatalyst at 5 mol % loading, and
affords ketone 3 in 96% yield and 90% ee (Table 9, entry 1).
Varying the amount of complex utilized in proportion with
phenylboronic acid, however, leads to significant production
of biphenyl above 5 mol % (Table 9, entries 2–5). Utilizing
even 25 mol % (entry 2) resulted in significant increase in
biphenyl production (16% yield), and reduction in yield of the
desired ketone 3 to 79%. Increasing complex loadings to 45
and 65 mol % (entries 3 and 4) leads to negligible production
of ketone 3, and nearly quantitative formation of biphenyl
relative to catalyst loading. Furthermore, attempts to use the
complex as a stoichiometric reagent in the place of PhB(OH)2
lead to no observed product (entry 5), and exclusive formation
of biphenyl. We hypothesize that quantitative generation of
the reactive arylpalladium cation intermediate in high relative
concentration leads quickly to disproportionation and forma-
tion of biphenyl and palladium(0). Omission of the AgPF6 in
favor of 30 mol % NH4PF6 leads to isolation of only 22%
yield of the conjugate addition product (entry 6). Finally, a
control experiment demonstrates that AgPF6 is incapable of
catalyzing the reaction itself (entry 7).31 This control further
supports the computational results, which indicate a transmet-
allation-based mechanism as opposed to a Lewis acid-
catalyzed pathway.
O
O
ligand (6 mol %)
Pd(OCOCF3)2 (5 mol %)
B(OH)2
+
ClCH2CH2Cl, 60 °C, 12 h
Ph
2
3
O
O
O
O
N
N
N
N
Ph
N
N
N
N
9
i-Pr
H
Ph
i-Bu
23
95% yield
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
99% yield
40% ee
22
99% yield
52% ee
24
99% yield
2% ee
24% ee
CF3
OMe
O
O
O
N
N
N
N
N
N
t-Bu
t-Bu
t-Bu
1
25
26
99% yield
99% yield
96% yield
93% ee
81% ee
78% ee
O
O
N
N
N
N
27
13% yield
28
14% yield
t-Bu
i-Pr
20% ee
0% ee
aConditions: 3-methylcyclohexen-2-one (0.25 mmol), phenylbo-
ronic acid (0.5 mmol), ClCH2CH2Cl (1 mL). Yields are isolated
yields, ee determined by chiral HPLC.
Permutations of the pyridinooxazoline ligand framework
corroborate the calculated data and suggest that a number of
factors affect enantioselectivity. First, the steric demand of
the chiral group on the oxazoline greatly impacts the observed
enantioselectivity in the reaction (Table 8). Only t-BuPyOx
(1) yields synthetically tractable levels of enantioselectivity,
while the less sterically demanding i-PrPyOx (21), PhPyOx
(22), and i-BuPyOx (23) all exhibit greatly diminished selec-
tivity. Oxazoline substitution patterns also affect enantioselec-
tivity. Substitution at the 4-position appears to be required for
high selectivity, as substitution at the 5-position yields practi-
cally no enantioselectivity (ligand 24). Electronic variation in
the PyOx framework was observed to have a large effect on
the rate of the reaction but, disappointingly, led to depressed
stereoselectivity. CF3-t-BuPyOx (25) afforded the conjugate
addition product in 99% yield and 81% ee. Surprisingly,
MeO-t-BuPyOx (26) afforded the product in similar yield and
only 78% ee. Finally, substitution at the 6-position of the
pyridine (ligands 27 and 28) greatly diminished both reactivity
and selectivity, perhaps due to hindered ligand chelation to
palladium.
Table 9. Arylpalladium(II) catalyzed conjugate addition.a
O
O
B(OH)2
Complex 29, AgPF6
H2O (5 equiv.)
ClCH2CH2Cl, 40 °C, 12 h
Ph
2
3
yield (%)
biphenyl
entry
AgPF6 (mol %) PhB(OH)2 (mol %)
complex 29 (mol %)
3
1
2
5
10
50
100
80
96
79
3
25
45
16
3
90
60
11
53
4
5
65
130
210
40
0
5
0
58
87
105
6b
7
5
0
0
200
200
0
0
22
0
20
Thus, we have concluded that enantioselectivity is con-
trolled by the steric repulsion between the substituent on the
chiral pyridinooxazoline ligand and the cyclohexyl ring. The
bulkier t-Bu substituent on the (S)-t-BuPyOx ligand leads to
greater enantioselectivity than the reactions with (S)-i-PrPyOx
or (S)-PhPyOx. Similarly, substrates with less steric demand
adjacent the carbonyl exhibit lower enantioselectivities; for
example the reaction of a lactone substrate (Table 6, TS-7)
yields lower enantioselectivity due to smaller repulsions be-
tween the lactone oxygen and the t-Bu group.
a
Conditions: 3-methylcyclohexen-2-one (0.25 mmol), pheylbo-
ronic acid (equiv as stated), ClCH2CH2Cl (1 mL). Yields are
isolated yields, ee determined by chiral HPLC. bReaction per-
formed with 30 mol % NH4PF6
Concerned by our inability to observe consistent product
formation at 40 °C, we sought alternative experimental verifi-
cation that a putative cationic arylpalladium(II) species is ca-
pable of reacting to form conjugate addition products. Thus,
we performed the stoichiometric reaction of the isomeric
phenyl palladium iodide complexes (29) with AgPF6 and 3-
methylcyclohexenone at cryogenic temperatures, allowing the
mixture to warm slowly to room temperature before quench-
ing with trifluoroethanol.32 We observed both conjugate addi-
8
ACS Paragon Plus Environment

Page 9 of 13
Journal of the American Chemical Society
tion product and biphenyl, with the desired adduct (3) isolated Pd/PyOx system operates under a similar manifold, and dem-
1
2
3
4
5
6
7
8
in 30% yield (Scheme 4a). Curiously, the conjugate addition
adduct was isolated in only 55% ee. We considered that the
isomeric mixture of phenyl palladium iodide isomers formed
configurationally stable cationic species at cryogenic tem-
onstrated a significant energy difference (transmetallation is
favored by over 40 kcal/mol, Figure 4) between the potential
roles of the palladium catalyst, suggesting that the role of the
palladium species is not simply that of a Lewis acid. Further-
more, it is difficult to rationalize the high degree of enanti-
oselectivity imparted by the chiral palladium catalyst if it is
assumed that the metal acts only as a Lewis acid and is not
directly mediating the key C–C bond-forming step.36 Finally,
a number of Lewis- and π−acidic metal salts were substituted
for palladium with no product observed, further indicating that
palladium-catalyzed conjugate addition is likely not a Lewis
acid-catalyzed process.37
,
perature.33 Repeating the experiment, we substituted triphen-
ylphosphine for methylcyclohexenone and observed the reac-
1
tion at low temperature utilizing H and 31P NMR (Scheme
4b). Indeed, two 31P signals corresponding to phosphine-
bound palladium(II) species (30) were observed at 28 and 34
ppm. No isomerization was observed upon warming the iso-
meric mixture to room temperature.34 Indeed, we were able to
isolate and characterize the mixture of arylpalladium-
phosphine cations. With evidence for the configurationally
stable arylpalladium cation, we rationalized the observed 55%
ee for the direct reaction of mixture 29 with methylcyclohex-
enone corresponds directly to the isomeric ratio of the com-
plex: a ratio of 1.3:1 represents a 56:44 ratio of isomers. As-
suming that the major isomer reacts with 92% ee, and the mi-
nor isomer reacts with no stereoselectivity to give racemic
products,35 a net stereoselectivity of 53.7% ee would be pre-
dicted for the product. Presuming configurational stability of
the arylpalladium(II) cation as suggested by the triphenyl-
phosphine trapping experiment, the diminished enantioselec-
tivity observed in this result is unsurprising. Thus, we have
obtained experimental verification of the key C–C bond
forming event of the Pd/PyOx asymmetric conjugate addition
occurring from a quantitative generated arylpalladium(II)
cation in the presence of enone substrate and absence of an
exogenous arylboronic acid.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
While highly Lewis-acidic, the role of cationic palla-
dium(II) is to provide a vacant coordination site for the enone
substrate to approach the catalyst. The presence of a cationic
intermediate is further supported by the observed rate accel-
eration of non-coordinating anionic additives such as PF6– and
BF4– salts (Table 1). Conversely, the addition of coordinating
anions, such as chloride, inhibited the reaction (Table 1, entry
1). This counterion effect was evident even from choice of
palladium(II) precatalysts.13 For instance, Pd(MeCN)2Cl2 was
not a suitable precatalyst, nor were any palladium(II) halides.
Additionally, Pd(OAc)2 only afforded modest yield of conju-
gate adducts, whereas the less coordinating counterion present
in Pd(OCOCF3)2 afforded complete conversion.
Satisfied that the palladium catalyst was not acting as a
Lewis acid, we next sought to demonstrate the viability of the
hypothesized arylpalladium(II) species as a catalyst (Table 9,
entries 1–5). While serving as a suitable precatalyst under
reaction conditions, arylpalladium(II) mixture 29 failed to
facilitate the conjugate addition reaction when used in
stoichiometric quantities in the presence of AgPF6 (entry 5).
We rationalize this outcome to be the result of the highly re-
active nature of the quantitatively generated arylpalladium(II)
cation. Significant biphenyl formation- occuring even under
dilute conditions representative of the catalytic reaction itself-
suggest that alternative reaction pathways, such as dispropor-
tionation, readily out compete the desired insertion reaction.
This reactive nature of the arylpalladium(II) cation led us to
propose performing the stoichiometric reactions at low tem-
peratures (Scheme 4). Successfully demonstrating that the key
C–C bond could be generated, albeit in modest yield, from
(PyOx)Pd(Ph)(I) (29) corroborates both the precise role of the
arylpalladium(II) cation in the calculated mechanism as well
as the transition state put forth in the calculations. However,
the modest yield of this process and requisite cryogenic tem-
peratures prompted, again, consideration of the role of the
arylboronic acid.38 Calculations suggest that the presence of
boronic acids as Lewis basic entities may serve to stabilize
these reactive intermediates under the reaction conditions
(Figure 3, cationic arylpalladium 8).39 This suggestion is con-
sistent with the successful use of arylpalladium(II) mixture 29
as a precatalyst in the presence of arylboronic acids (Table 9,
entry 1).
Scheme 4. (a) Direct formation of C–C bond from arylpalla-
dium(II) cation. (b) Triphenylphosphine trapping experiments
demonstrates configurational stability of arylpalladium cation.
(a)
O
O
O
1. 3-methylcyclohexenone,
AgPF6, CH2Cl2, –78 °C
N
N
I
N
I
N
Pd
Pd
t-Bu
t-Bu
2. CF3CH2OH, 20 °C
Ph
3
30% yield
55% ee
29
mix of isomers
(b)
O
O
–
+
PF6
N
N
I
N
N
Pd
Pd
t-Bu
t-Bu
PPh3
AgPF6, PPh3
29
mix of isomers
30
Two 31P Signals Observed
CD2Cl2, –78 °C
O
O
–
+ PF6
N
I
N
N
N
Pd
Pd
t-Bu
t-Bu
Ph3P
Summary and Discussion
Lastly, water (or another proton source) is required for effi-
cient turnover of the reaction. Considerations of reaction scale
(Scheme 2) and deuterium incorporation experiments (Figure
1) suggest that water is the likely protonation agent, despite
numerous other protic sources in the heterogenous reaction
mixture. Attempts to use other, miscible proton sources
(MeOH, phenol, t-BuOH, etc.) typically resulted in 10–15%
less enantioselectivity observed.40 2,2,2-Trifluoroethanol can
be successfully substituted for water with minimal loss of
enantioselectivity, however it affords no supplementary bene-
Several experimental results have been described that sup-
port the DFT-calculated mechanism for the Pd/PyOx-
catalyzed asymmetric conjugate addition of arylboronic acids
to cyclic enone (Figure 4); specifically, the role of the palla-
dium catalyst has been addressed. Calculations and previous
experimental work by Miyaura on palladium-catalyzed conju-
gate addition suggest that a transmetallation event occurs to
transfer the aryl moiety from the boronic acid species to the
palladium catalyst.4 Our calculations indicated that the
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 13
fit and water is the preferred additive for all reactions.25 Water
rate- and enantioselectivity-determining carbopalladation of
1
2
3
4
5
6
7
8
in combination with NH4PF6 serves to facilitate milder reac-
tion conditions (Table 2), which in turn greatly increases the
synthetic scope with respect to challenging arylboronic acid
nucleophiles (Table 3). Synthetic yields were observed to
double in many cases, greatly increasing the utility of these
transformations. The functional group tolerance of the
Pd/PyOx system is unprecedented for asymmetric conjugate
addition; it encompasses a wide array of halides, carbonyl
functional groups, protected phenols, acetamides, free hy-
droxyl groups, and even nitro groups. Many of these groups
are incompatible with traditional rhodium- and copper-
catalyzed conjugate additions due to the reactivity with the
nucleophiles used or the strong coordination of the functional
group to the metal catalyst.
the enone olefin by a cationic palladium species, and pro-
tonolysis of the resulting palladium-enolate. We have taken
advantage of these mechanistic insights to develop a modified
reaction system whereby the addition of water and ammonium
hexafluorophophate increase reaction rates, and can facilitate
lower catalyst loadings. The modified conditions have opened
the door to new substrate classes that were inaccessible by the
initially published reaction conditions. Furthermore, we have
demonstrated that this operationally simple reaction is tolerant
of ambient atmosphere and capable of producing enantioen-
riched, β-quaternary ketones on multi-gram scale. The steric
and electronic effects of the boronic acid and enone substrates
and the ligand on enantioselectivities were elucidated by a
combined experimental and computational investigation. The
enantioselectivity is mainly controlled by the steric repulsion
of the t-Bu substituent of the oxazoline on the ligand and the
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
+
N
L
B(OH)2
N
X
Pd
Cα position hydrogens of the cyclohexenone substrate in the
'
alkene insertion transition state.
Ph
3
31
protonation
H2O
Further investigations of both the scope of this transforma-
tion and its application toward natural product synthesis are
current underway in our laboratories.
transmetallation
+
N
N
Pd
+
O
O
N
L
N
+
Pd
N
L
32
N
Pd
L
Ph
Ph
ASSOCIATED CONTENT
34
35
Supporting Information. Experimental details, optimized Carte-
sian coordinates and energies, and complete reference of Gaussian
03. This material is available free of charge via the Internet at
http://pubs.acs.org.
insertion
+
N
2
Pd
N
O
coordination
O
33
AUTHOR INFORMATION
Figure 6. Proposed catalytic cycle for Pd/PyOx-catalyzed
conjugate addition of arylboronic acids to cyclic enones.
Corresponding Author
* stoltz@caltech.edu; houk@chem.ucla.edu
The combined results described herein have allowed us to
put forth the following catalytic cycle (Figure 6). The cationic
catalyst, represented as (PyOx)Pd(X)(L) (31), undergoes
transmetallation with an arylboronic acid to yield cationic
(PyOx)Pd(Ar)(L) (32). Substrate coordination forms cationic
arylpalladium(II) 33, which undergoes rate and enantioselec-
tivity-determining insertion of the aryl moiety into the enone
π-system to afford C-bound palladium enolate 34. Tautomeri-
zation to the O-bound palladium enolate (35), or direct pro-
tonolysis of the C-bound enolate (34), liberates the conjugate
addition product (3) and regenerates a cationic palladium(II)
species for reentry into the catalytic cycle.
ACKNOWLEDGMENT
The authors thank NIH-NIGMS (B.M.S., R01 GM080269–01,
K.N.H. GM36700), NSF (K.N.H., CHE-0548209), Caltech, Am-
gen, the American Chemical Society Division of Organic Chem-
istry (predoctoral fellowship J.C.H.), the Swiss National Science
Foundation (postdoctoral fellowship M.G.) and the Japan Society
for the Promotion of Science (postdoctoral fellowship K.K.) for
financial support. A.N.M is grateful for a research fellowship by
the German National Academy of Sciences Leopoldina (LPDS
2011-12). Prof. Theodor Agapie (Caltech) is thanked for helpful
discussions. Dr. David VanderVelde of the Caltech NMR facility
is thanked for invaluable assistance with NMR experiments and
helpful discussions. Lawrence Henling and Dr. Michael K. Ta-
kase (Caltech) are gratefully acknowledged for X-ray crystallo-
graphic structural determination. The Bruker KAPPA APEXII X-
ray diffractometer was purchased via an NSF CRIF:MU award to
the California Institute of Technology, CHE-0639094. The Var-
ian 400 MHz NMR spectrometer at Caltech was purchased via an
NIH grant (RR027690). Calculations were performed on the
Hoffman2 cluster at UCLA and the Extreme Science and Engi-
neering Discovery Environment (XSEDE), which is supported by
the NSF.
Conclusions
In conclusion, we have reported experimental and compu-
tational results that corroborate a single PyOx ligand/metal
complex as the active catalytic species in the palladium-
catalyzed asymmetric conjugate addition of arylboronic acids
to enones for the construction of quaternary stereocenters. We
have used computational models to rule out a suggested alter-
native mechanism in which the palladium catalyst is acting as
a Lewis acid to activate the enone. The preferred mechanism
involves formal transmetallation from boron to palladium,
1
(a) P. Perlmutter, in Conjugate Addition Reactions in Organic
Synthesis, Tetrahedron Organic Chemistry Series 9; Pergamon,
Oxford, 1992; (b) K. Tomioka, Y. Nagaoka, in Comprehensive
Asymmetric Catalysis, Vol. 3 (Eds: E. N. Jacobsen, A. Pfaltz, H.
References
10
ACS Paragon Plus Environment

Page 11 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Yamamoto, Springer-Verlag, New York, 1999; Chapter 31; (c)
Gini, F.; Hessen, B.; Feringa, B. L.; Minnaard, A. J. Chem. Com-
R.; Takeda, M.; Nishimura, T.; Hayashi, T. Angew. Chem., Int. Ed.
2010, 49, 3969–3971.
11
mun. 2007, 710–712.
For excellent review articles, see: (a) Gutnov, A. Eur. J. Org.
2
For reviews, see: (a) Harutyunyan, S. R.; den Hartog, T.; Geurts,
Chem. 2008, 4547–4554. (b) Christoffers, J.; Koripelly, G.; Rosiak,
A.; Rössle, M. Synthesis 2007, 1279–1300.
K.; Minnaard, A. J.; Feringa, B. L. Chem. Rev. 2008, 108,
2824–2882. (b) Alexakis, A.; Backvall, J. E.; Krause, N.; Pamies,
O.; Dieguez, M. Chem. Rev. 2008, 108, 2796–2893. (c) Lopez, F.;
Minnarard, A. J.; Feringa, B. L. Acc. Chem. Res. 2007, 40,
179–188. (d) Hayashi, T.; Yamasaki, K. Chem. Rev. 2003, 103,
2829–2844. (e) Feringa, B. L. Acc. Chem. Res. 2000, 33,
346–353. (f) Rossiter, B. E.; Swingle, N. M. Chem. Rev. 1992, 92,
12 Lin, S.; Lu, X. Org. Lett. 2010, 12, 2536–2539.
13
Kikushima, K.; Holder, J. C.; Gatti, M.; Stoltz, B. M. J. Am.
Chem. Soc. 2011, 133, 6902–6905.
14
PyOx ligand scaffolds are increasingly common in asymmetric
catalysis, see: (a) Podhajsky, S. M.; Iwai, Y.; Cook-Sneathen, A.;
Sigman, M. S. Tetrahedron 2011, 67, 4435–4441. (b) Aranda, C.;
Cornejo, A.; Fraile, J. M.; García-Verdugo, E.; Gil, M. J.; Luis, S.
V.; Mayoral, J. A.; Martínez-Merino, V.; Ochoa, Z. Green Chem.
2011, 13, 983–990. (d) Pathak, T. P.; Gligorich, K. M.; Welm, B.
E.; Sigman, M. S. J. Am. Chem. Soc. 2010, 132, 7870–7871. (d)
Jiang, F.; Wu, Z.; Zhang, W. Tetrahedron Lett. 2010, 51,
5124–5126. (e) Jensen, K. H.; Pathak, T. P.; Zhang, Y.; Sigman, M.
S. J. Am. Chem. Soc. 2009, 131, 17074–17075. (f) He, W.; Yip, K.-
T.; Zhu, N.-Y.; Yang, D. Org. Lett. 2009, 11, 5626–5628. (g) Dai,
H.; Lu, X. Tetrahedron Lett. 2009, 50, 3478–3481. (h) Linder, D.;
Buron, F.; Constant, S.; Lacour, J. Eur. J. Org. Chem. 2008,
5778–5785. (i) Schiffner, J. A.; Machotta, A. B.; Oestreich, M.
Synlett 2008, 2271–2274. (j) Koskinen, A. M. P.; Oila, M. J.; Tois,
J. E. Lett. Org. Chem. 2008, 5, 11–16. (k) Zhang, Y.; Sigman, M.
S. J. Am. Chem. Soc. 2007, 129, 3076–3077. (l) Yoo, K. S.; Park,
C. P.; Yoon, C. H.; Sakaguchi, S.; O’Neill, J.; Jung, K. W. Org.
Lett. 2007, 9, 3933–3935. (m) Dhawan, R.; Dghaym, R. D.; St.
Cyr, D. J.; Arndtsen, B. A. Org. Lett. 2006, 8, 3927–3930. (n) Xu,
W.; Kong, A.; Lu, X. J. Org. Chem. 2006, 71, 3854–3858. (o)
Malkov, A. V.; Stewart Liddon, A. J. P.; Ramírez-López, P.; Ben-
dová, L.; Haigh, D.; Kocovsky, P. Angew. Chem., Int. Ed. 2006, 45,
1432–1435. (p) Abrunhosa, I.; Delain-Bioton, L.; Gaumont, A.-C.;
Gulea, M.; Masson, S. Tetrahedron 2004, 60, 9263–9272. (q)
Brunner, H.; Kagan, H. B.; Kreutzer, G. Tetrahedron: Asymmetry
2003, 14, 2177–2187. (r) Cornejo, A.; Fraile, J. M.; García, J. I.;
Gil, M. J.; Herrerías, C. I.; Legarreta, G.; Martínez-Merino, V.;
Mayoral, J. A. J. Mol. Catal. A: Chem. 2003, 196, 101–108. (s)
Zhang, Q.; Lu, X.; Han, X. J. Org. Chem. 2001, 66, 7676–7684. (t)
Zhang, Q.; Lu, X. J. Am. Chem. Soc. 2000, 122, 7604–7605. (u)
Perch, N. S.; Pei, T.; Widenhoefer, R. A. J. Org. Chem. 2000, 65,
3836–3845. (v) Bremberg, U.; Rahm, F.; Moberg, C. Tetrahedron:
Asymmetry 1998, 9, 3437–3443. (w) Brunner, H.; Obermann, U.;
Wimmer, P. Organometallics 1989, 8, 821–826.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
771–806.
3
(a) Shintani, R.; Duan, W.-L.; Hayashi, T. J. Am. Chem. Soc.
2006, 128, 5628–5629. (b) Hayashi, T.; Takahashi, M.; Takaya, Y.;
Ogasawara, M.; J. Am. Chem. Soc. 2002, 124, 5052–5058. (c)
Takaya, Y.; Ogasawara, M.; Hayashi, T. J. Am. Chem. Soc. 1998,
120, 5579–5580.
4
(a) Nishikata, T.; Yamamoto, Y.; Miyaura, N. Angew. Chem., Int.
Ed. 2003, 42, 2768–2770. (b) Nishitaka, T.; Yamamoto, Y.; Mi-
yaura, N. Chem. Lett. 2007, 36, 1442–1443. (c) Nishitaka, T.; Ki-
yomura, S.; Yamamoto, Y.; Miyaura, N. Synlett 2008, 2487–2490.
5
Gini, F.; Hessen, B.; Minnaard, A. J. Org. Lett. 2005, 7,
5309–5312.
6
For reviews on the synthesis of quaternary stereocenters, see: (a)
Denissova, I.; Barriault, L. Tetrahedron 2003, 59, 10105–10146.
(b) Douglas, C. J.; Overman, L. E. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 5363–5267. (c) Christoffers, J.; Baro, A. Adv. Synth.
Catal. 2005, 347, 1473–1482. (d) Trost, B. M.; Jiang, C. Synthesis
2006, 369–396. (e) Mohr, J. T.; Stoltz, B. M. Chem. Asian J. 2007,
21, 1476–1491. (f) Cozzi, P. G.; Hilgraf, R.; Zimmermann, N. Eur.
J. Org. Chem. 2007, 5969–5994.
6
For an excellent comprehensive review, see: Hawner, C.; Alex-
akis, A. Chem. Commun. 2010, 46, 7295–7306.
7
(a) Wu, J.; Mampreian D. M.; Hoveyda, A. H. J. Am. Chem. Soc.
2005, 127, 4584–4585. (b) Hird, A. W.; Hoveyda, A. H. J. Am.
Chem. Soc. 2005, 127, 14988–14989. (c) Lee, K.-S.; Brown, M. K.;
Hird, A. W.; Hoveyda, A. H. J. Am. Chem. Soc. 2006, 128,
7182–7184. (d) Brown, M. K.; May, T. L.; Baxter, C. A.; Hoveyda,
A. H. Angew. Chem., Int. Ed. 2007, 46, 1097–1100. (e) Fillion, E.;
Wilsily, A. J. Am. Chem. Soc. 2006, 128, 2774–2775. (f) Wilsily,
A.; Fillion, E. J. Org. Chem. 2009, 74, 8583–8594. (g) Dumas, A.
M.; Fillion, E. Acc. Chem. Res. 2010, 43, 440–454. (h) Wilsily, A.;
Fillion, E. Org. Lett. 2008, 10, 2801–2804.
8 (a) d’Augustin, M.; Palais, L.; Alexakis, A. Angew. Chem., Int. Ed.
2005, 44, 1376–1378. (b) Vuagnoux-d’Augustin, M.; Alexakis, A.
Chem.–Eur. J. 2007, 13, 9647–9662. (c) Palais, L.; Alexakis, A.
Chem.–Eur. J. 2009, 15, 10473–10485. (d) Fuchs, N.; d’Augustin,
M.; Humam, M.; Alexakis, A.; Taras, R.; Gladiali, S. Tetrahedron:
Asymmetry 2005, 16, 3143–3146. (e) Vuagnoux-d’Augustin, M.;
Kehrli, S.; Alexakis, A. Synlett, 2007, 2057–2060. (f) May, T. L.;
Brown, M. K.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2008, 47,
7358–7362. (g) Ladjel, C.; Fuchs, N.; Zhao, J.; Bernardinelli, G.
Alexakis, A. Eur. J. Org. Chem. 2009, 4949–4955. (h) Palais, L.;
Mikhel, I. S.; Bournaud, C.; Micouin, L.; Falciola, C. A.; Vuag-
noux-d’Augustin, M.; Rosset, S.; Bernardinelli, G.; Alexakis, A.
Angew. Chem., Int. Ed. 2007, 46, 7462–7465. (i) Hawner, C.; Li,
K.; Cirriez, V.; Alexakis, A. Angew. Chem., Int. Ed. 2008, 47,
8211–8214. (j) Müller, D.; Hawner, C.; Tissot, M.; Palais, L.; Alex-
akis, A. Synlett 2010, 1694–1698. (k) Hawner, C.; Müller, D.;
Gremaud, L.; Felouat, A.; Woodward, S.; Alexakis, A. Angew.
15 The ligand was prepared as described in the literature, see: Brun-
ner, H.; Obermann, U. Chem. Ber. 1989, 122, 499–507. See Sup-
porting Information for experimental details.
16
(a) Xu, Q.; Zhang, R.; Zhang, T.; Shi, M. J. Org. Chem. 2010,
75, 3935–3937; (b) Zhang, T.; Shi, M. Chem.–Eur. J. 2008, 14,
3759–3764; (c) Gottummukkala, A. L.; Matcha, K.; Lutz, M.; de
Vries, J. G. ; Minnaard, A. J. Chem.–Eur. J. 2012, 18, 6907–6914.
17
Holder, J. C., Marziale, A. N., Gatti, M.; Mao, B.; Stoltz, B. M.
Chem.–Eur. J. 2013, 19, 74–77.
18
We believe the small amount of deuterium incorporation at the
methylene adjacent the enone occurs via substrate enolization dur-
ing the extended reaction times under the mild reaction conditions.
19
Duan, W.-L.; Iwamura, H.; Shintani, R.; Hayashi, T. J. Am.
Chem. Soc. 2007, 129, 2130–2138.
20 Guillaneux, D.; Zhao, S.-H.; Samuel, O.; Rainford, D.; Kagan, H.
B. J. Am. Chem. Soc. 1994, 116, 9430–9439. Further evidence for
the monomeric nature of the Pd/PyOx catalyst was provided by
comparison of diffusion rates of various PyOx/Pd complexes by
diffusion oriented spectroscopy (DOSY NMR). These studies sug-
gest an upper bound for the molecular weight of the solution species
of the catalyst, demonstrating that dimeric (ML)2 complexes are not
Chem., Int. Ed. 2010, 49, 7769–7772.
9
(a) Martin, D.; Kehrli, S.; d’Augustin, M.; Clavier, H.; Mauduit,
M.; Alexakis, A. J. Am. Chem. Soc. 2006, 128, 8416–8417. (b)
Kehrli, S.; Martin, D.; Rix, D.; Mauduit, M.; Alexakis, A.
Chem.–Eur. J. 2010, 16, 9890–9904. (c) Hénon, H.; Mauduit, M.;
Alexakis, A. Angew. Chem., Int. Ed. 2008, 47, 9122–9124. (d)
Matsumoto, Y.; Yamada, K.-I.; Tomioka, K. J. Org. Chem. 2008,
present. See Supporting Information.
21
(a) Inanaga, J.; Furuno, H.; Hayano, T. Chem. Rev. 2002, 102,
2211–2226. (b) Kagan, H. B. Synlett 2001, 888–899. (c) Girard, C.;
Kagan H. B. Angew. Chem., Int. Ed. 1998, 37, 2922–2959.
73, 4578–4581.
10
22
(a) Shintani, R.; Tsutsumi, Y.; Nagaosa, M.; Nishimura, T.; Ha-
For preparation and use of (PyOx)Pd(OCOCF3)2 see Supporting
yashi, T. J. Am. Chem. Soc. 2009, 131, 13588–13589. (b) Shintani,
Information.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 13
1
2
23 Nishitaka, T.; Yamamoto, Y.; Miyaura, N. Organometallics 2004,
39
The suggestion of boronic acid stabilization of arylpalladium
23, 4317–4324.
3
4
5
6
7
8
9
cationic intermediates is consistent with the observation that boron
species lacking hydroxyl groups serve as poor substrates for the
reaction. For example, greatly diminished yields (and high degrees
of biphenyl formation) are observed with the use of NaBPh4 or
KF3BPh as the phenyl donor species. See Supporting Information.
40 See Supporting Information of ref. 13 for details on the sensitivity
of enantioselectivity of the Pd/PyOx system to polar, coordinating
solvents. The addition of 5 equiv. of alcoholic co-solvent as a pro-
ton source is generally detrimental to the enantioselectivity and,
occasionally, to the yield of the reaction.
24
The catalyst itself is known to be soluble and not zero valent
palladium nanoparticles due to exclusion by a mercury drop test, see
ref 17. For examples of the mercury drop test, see: (a) Ines, B.;
SanMartin, R.; Moure, M. J.; Dominguez, E. Adv. Synth. Catal.
2009, 351, 2124–2132. (b) Ogo, S.; Takebe, Y., Uehara, K., Yama-
zaki, T., Nakai, H., Watanabe, Y., Fukuzumi, S, Organometallics
2006, 25, 331–338.
25
Attempts to study the reaction by NMR initially saw no reaction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
progress due to the inability to stir the immiscible reaction mixture
in an NMR tube. Substitution of 2,2,2-trifluoroethanol for water as
the super stoichiometric proton source facilitated the reaction to
proceed in the absence of stirring with no detriment to the observed
enantioselectivity, however the reaction was necessarily performed
at sufficiently elevated temperature that observation of the initial
rate was not tenable and, thus, kinetic study was abandoned.
26 Lan, Y.; Houk, K. N. J. Org. Chem. 2011, 76, 4905–4909.
27
For related computational studies on palladium-catalyzed conju-
gate additions of arylboronic acids to enones: (a) Nishikata, T.;
Yamamoto, Y.; Gridnev, I. D.; Miyaura, N. Organometallics 2005,
24, 5025–5032. (b) Sieber, J. D.; Liu, S.; Morken, J. P. J. Am.
Chem. Soc. 2007, 129, 2214–2215. (c) Dang, L.; Lin, Z.; Marder, T.
B. Organometallics 2008, 27, 4443–4454.
28 (a) Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377. (b) Becke,
A. D. J. Chem. Phys. 1993, 98, 5648–5652. (c) Perdew, J. P.;
Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.;
Singh, D. J.; Fiolhais, C. Phys. Rev. B 1992, 46, 6671–6687. (d)
Perdew, J. P. Phys. Rev. B 1986, 33, 8822–8824.
29
Gaussian 03, Revision C.02: Frisch, M. J et al. Gaussian, Inc.,
Wallingford CT, 2004.
30
Here, it is assumed the isomeric trans and cis phenylpalladium
complexes cannot undergo rapid isomerization before alkene inser-
tion and thus the enantioselectivity is determined by the energy
difference between 8-TS-A and 8-TS-B, since the transmetallation
leading to the trans isomer is favored (see SI). If trans/cis isomeri-
zation is faster than alkene insertion, the enantioselectivity will be
determined by the energy difference between 8-TS-A and 8-TS-D
(98% ee).
31
Other Lewis acids also proved incapable of catalyzing the reac-
tion, including AlCl3 and Sn(OTf)2.
32 These experiments were followed by NMR (1H, 13C ), however no
discrete intermediates were successfully characterized.
33
Attempts to separate the isomeric mixture of phenylpalladium
iodide complexes failed by both conventional silica gel flash chro-
matography and preparatory HPLC.
34
Optimization of the cis/trans isomerization transition state of the
cationic phenyl palladium(II) complex failed to locate a TS. Scan of
the reaction coordinate indicates the barrier of isomerization is
higher than 10 kcal/mol with respect to the cationic phenyl palla-
dium complex 11. This suggests the cis/trans isomerization via the
dissociative mechanism via isomerization of the tri-coordinated
complex 11 cannot occur. See Supporting Information for details.
35 Computations indicated that alkene insertion to the minor isomer
of phenyl palladium complex, in which the Ph is cis to the oxa-
zoline, yields very low enantioselectivity. The chiral oxazoline is
now trans to the enone, and thus the stereocontrol is diminished (see
1-TS-C and 1-TS-D in Figure 5).
36
A similar argument about imparted enantioselectivity can be
made to rule out the catalytic activity of palladium(0) nanoparticles.
Additionally, a mercury(0) poisoning test has ruled out the activity
of palladium(0) nanoparticles in the Pd/PyOx manifold. See ref. 17.
37
Brønsted acid catalysis was also ruled out, as the substitution of
trifluoroacetic acid for Pd(OCOCF3)2 proved unable to catalyze the
reaction. Protic acids are not tenable catalysts in the absence of
palladium salts.
38 We have computed the effects of coordination with phenylboronic
acid to activate the carbonyl on the enone in the alkene insertion
step. No acceleration on alkene insertion was observed computa-
tionally with either Lewis acid or hydrogen bonding coordination.
See Supporting Information.
12
ACS Paragon Plus Environment

Page 13 of 13
Journal of the American Chemical Society
1
2
3
O
N
O
O
N
R2
t-Bu
B(OH)2
Pd(OCOCF3
)
2
R2
4
ClCH2CH2Cl, 60 °C
R1
R1
5
6
7
8
O
O
H
N
N
Pd
R1
R2
Enantioselectivity-Determining Step
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment