Enantioselective synthesis of axially chiral biaryl monophosphine oxides via direct asymmetric suzuki coupling and DFT investigations of the enantioselectivity

Article abstract of DOI:10.1021/cs500208n

Direct asymmetric Suzuki coupling between arylboronic acids and 2-diarylphosphinyl-1-naphthyl bromides was successfully developed for the first time with the use of Pd-L1 or Pd-(Cy-MOP) as the catalyst. A variety of axially chiral 2-functionalized-2′-diarylphosphinyl-1,1′-biaryls were afforded in 34-99% yields with up to 94% ee. This methodology provides a highly efficient and practical strategy for the synthesis of novel axially chiral biaryl monophosphine oxides and the corresponding phosphines. The existence of an ortho formyl group in arylboronic acids greatly improves the coupling efficiency and permits further versatile transformations in organic synthesis. Density functional calculations were used to determine the origin of stereoselectivity during the reductive elimination step of the closely related coupling of 2-formylphenylboronic acid with naphthylphosphonate bromide. These studies indicate that both the significant transition metal hydrogen bond between the H atom of the formyl group and palladium(II) and the weak interaction between the Pd center and the phosphoryl oxygen atom in the transition state are crucial for high enantioselectivity of the coupling products.

Full text of DOI:10.1021/cs500208n

Research Article
pubs.acs.org/acscatalysis
Enantioselective Synthesis of Axially Chiral Biaryl Monophosphine
Oxides via Direct Asymmetric Suzuki Coupling and DFT
Investigations of the Enantioselectivity
Yougui Zhou,† Xuepeng Zhang,‡ Huiyi Liang,† Zhenkun Cao,† Xiaoyu Zhao,† Yuwei He,†
Shouliang Wang,† Jiyan Pang,† Zhongyuan Zhou,§ Zhuofeng Ke,*,‡ and Liqin Qiu*,†,∥,⊥
†School of Chemistry and Chemical Engineering, Guangdong Engineering Research Center of Chiral Drugs, Sun Yat-Sen University,
Guangzhou 510275, People’s Republic of China
‡MOE Key Laboratory of Bioinorganic and Synthetic Chemistry/KLGHEI of Environment and Energy Chemistry, Sun Yat-sen
University, Guangzhou 510275, People’s Republic of China
§Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hong Kong, People’s Republic
of China
∥Huizhou Research Institute of Sun Yat-Sen University, Dayawan, Huizhou, People’s Republic of China
⊥Key Laboratory of Synthetic Chemistry of Natural Substances, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, Shanghai People’s Republic of China
S
* Supporting Information
ABSTRACT: Direct asymmetric Suzuki coupling between arylboronic acids and 2-diarylphosphinyl-1-naphthyl bromides was
successfully developed for the first time with the use of Pd−L1 or Pd−(Cy-MOP) as the catalyst. A variety of axially chiral 2-
functionalized-2′-diarylphosphinyl-1,1′-biaryls were afforded in 34−99% yields with up to 94% ee. This methodology provides a
highly efficient and practical strategy for the synthesis of novel axially chiral biaryl monophosphine oxides and the corresponding
phosphines. The existence of an ortho formyl group in arylboronic acids greatly improves the coupling efficiency and permits
further versatile transformations in organic synthesis. Density functional calculations were used to determine the origin of
stereoselectivity during the reductive elimination step of the closely related coupling of 2-formylphenylboronic acid with
naphthylphosphonate bromide. These studies indicate that both the significant transition metal hydrogen bond between the H
atom of the formyl group and palladium(II) and the weak interaction between the Pd center and the phosphoryl oxygen atom in
the transition state are crucial for high enantioselectivity of the coupling products.
KEYWORDS: asymmetric catalysis, biaryls, computational chemistry, phosphine oxides, Suzuki coupling
diarylphosphinic chlorides or diarylphosphine oxides (Scheme
1A);3 however, a multistep synthesis was often required to
obtain these biaryl halide or biaryl triflate substrates, or even
resolution of racemic materials was involved in some
procedures.4 This results in material waste and thus would
INTRODUCTION
■
Axially chiral 2-functionalized-2′-diarylphosphino-1,1′-biaryl
ligands (MOP type ligands) have attracted much attention
because of their high efficiency in transition-metal-catalyzed
asymmetric reactions.1,2 Indeed, the chiral functionalized biaryl
monophosphine oxides are important intermediates for the
preparation of MOP type ligands. The commonly used
synthetic protocol for these compounds relies on a coupling
reaction of axially chiral biaryl halides or biaryl triflates with
Received: February 16, 2014
Revised: March 26, 2014
Published: April 7, 2014
© 2014 American Chemical Society
1390
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
Scheme 1. Synthetic Protocols for Axially Chiral Biaryl
Monophosphine Oxides
Scheme 2. Synthesis of Racemic Biaryl Monophosphine
Oxides
both the coupling racemic products 4aA and 5aA were
obtained smoothly in 90% yield for 72 h at 60 and 90 °C,
respectively.
With this approach in hand, we began to seek efficient
asymmetric catalytic systems used in reactions I and II for
affording the axially chiral biaryl monophosphine oxides 4aA
and 5aA. A variety of chiral ligands (Figure 1, L1−L15)9 were
lead to low efficiency of the whole synthetic process. A
convenient preparation for novel axially chiral functionalized
phosphine oxides remains of significance and a challenge.
Enantioselective metal-catalyzed Suzuki coupling has become
one of the most used protocols in the synthesis of axially chiral
biaryls.5 In 2000, Buchwald first reported a Pd-catalyzed
asymmetric Suzuki coupling of dialkoxyphosphinyl (P(O)-
(OR)2)-substituted naphthyl bromides with ortho-alkyl-sub-
stituted arylboronic acids. A number of enantioenriched biaryl
phosphonates were obtained with up to 92% ee.6 The
phosphonate moiety in the coupling product was successfully
converted to the corresponding phosphine oxide, too.
However, the reaction conditions, such as solvent, reaction
temperature and reaction time, were critical for the conversion
to avoid erosion of the enantiomeric purity (Scheme 1B).
We recently established a Pd-(chiral bridged atropisomeric
monophosphine) catalyst system7 and a Pd−(Cy-MOP)
catalyst system8 for asymmetric Suzuki coupling of P(O)-
(OR)2-substituted naphthyl bromides with ortho-substituted
arylboronic acids. These catalyst systems afforded the coupling
products with excellent ee’s (both up to 97% ee). Herein, we
report a subsequent finding that direct asymmetric Suzuki
coupling generates axially chiral biaryl monophosphine oxides
containing various functional groups using these two catalyst
systems (Scheme 1C). To our knowledge, this represents the
first example of enantioselective synthesis of 2-functionalized-
2′-diarylphosphinyl-1,1′-biaryls via this catalytic route.
Figure 1. Monophosphine ligands L1−L15.
examined in toluene with K3PO4 as the base and Pd2(dba)3 as
the palladium source (Table 1). L1 and L15 (Cy-MOP) were
found to be the best choice for reaction I and reaction II,
respectively.
For reaction I, ligands L1−L8 were found to give better ee
values than those of BINOL-derived MOP’s L10−L15 (Table
1, entries 1−11 vs entries 13−18). This demonstrates that the
chiral bridged biphenyl skeleton in L1−L8 is beneficial for
acquiring high enantioselectivity in this reaction. L1 provided
the best result at 30 °C within 48 h (Table 1, entry 1). When
the reaction temperature was decreased to 10 °C, the reaction
became sluggish, and there was no beneficial effect toward the
enantioselectivities (Table 1, entry 2). When the temperature
was increased to 60 °C, 1a was consumed completely in 24 h;
however, the ee value decreased slightly (Table 1, entry 3, 64%
ee). When the reaction was conducted at 30 °C for 24 h, only
40% yield was obtained (Table 1, entry 4). When the 3,5-
(tBu)2-4-MeO-C6H2 group attached to the phosphorus atom in
L1 was changed to a 3,5-(tBu)2-C6H3 group (L2), the same ee
value was given, but the yield dropped (Table 1, entry 4 vs
entry 5). It indicates that a high electron density of the attached
aryl group enhances the reaction rate.10 L3, L4, and L5
possessing less bulky aryl groups at this place provided lower
enantioselectivities (Table 1, entries 6−8). This shows that a
bigger aryl group is somewhat helpful to better enantiose-
lection. The influence of an alkoxy or aryloxy group linked at
RESULTS AND DISCUSSION
■
Establishment of Catalyst System. We have embarked
on a Pd-catalyzed Suzuki coupling of 2-diphenylphosphinyl-1-
naphthyl bromide 1a with 2-methoxyl phenylboronic acid 2A
(Scheme 2, reaction I) and 2-formyl phenylboronic acid 3A
(Scheme 2, reaction II) for the synthesis of racemic
monophosphine oxides 4aA and 5aA. It was observed that,
with the combination of toluene, K3PO4, Pd2(dba)3, and Sphos,
1391
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
Table 1. Ligand Screening
a
a
reaction I
reaction II
b
c
b
c
entry
L
t, °C
time, h
yield, %
ee, %
entry
L
t, °C
time, h
yield %
ee %
d
1
2
L1
30
10
60
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
48
120
24
24
24
48
48
48
48
48
48
48
48
48
48
48
48
48
95
44
95
40
30
95
95
95
95
90
90
66
44
89
79
94
94
80
68 (−)
68 (−)
64 (−)
68 (−)
68 (−)
64 (−)
61 (−)
62 (−)
62 (−)
63 (−)
65 (−)
36 (+)
44 (−)
45 (−)
41 (−)
8 (−)
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
L1
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
70
50
30
72
72
72
72
72
72
72
72
72
72
72
72
72
72
72
72
72
48
69
53
55
45
50
50
20
20
40
45
39
39
80
75
95
95
93
0
80 (84 , R)
L1
L2
80 (R)
84 (R)
86 (R)
85 (R)
83 (R)
77 (R)
79 (R)
67 (R)
67 (R)
65 (R)
55 (R)
77 (R)
79 (R)
84 (R))
89 (R)
3
L1
L3
4
L1
L4
5
L2
L5
6
L3
L6
7
L4
L7
8
L5
L8
9
L6
L9
10
11
12
13
14
15
16
17
18
L7
L10
L11
L12
L13
L14
L15
L15
L15
L15
L8
L9
L10
L11
L12
L13
L14
L15
e
39 (−)
36 (−)
93 (99 , R)
a
b
Reaction conditions: 1a (0.5 mmol), 2A or 3A (1.0 mmol); K3PO4 (1.5 mmol); ligand (4.8 mol %); Pd (4 mol %); toluene (4 mL). Yields were
c
d
e
determined by HPLC and NMR analysis of the crude reaction mixture. Determined by HPLC. The reaction was conducted at 50 °C. 99% ee and
62% overall yield after recrystallization once from methylene chloride and hexanes. The absolute configuration of 5aA was determined by X-ray
crystallography.11
a
Table 2. Optimization Experiments
b
c
entry
1
palladium
Pd2(dba)3
base
solvent
toluene
reaction
yield %
ee %
K3PO4
I
95
93
60
65
77
74
traces
traces
87
80
89
80
0
68 (−)
93 (R)
62 (−)
92 (R)
62 (−)
92 (R)
nd
II
I
2
3
Pd2(dba)3
Pd2(dba)3
PdCl2
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K2CO3
KF
THF
II
I
dioxane
toluene
toluene
toluene
toluene
toluene
toluene
toluene
II
I
4
II
I
nd
5
Pd(OAc)2
Pd(CF3COO)2
68 (−)
93 (R)
68 (−)
92 (R)
nd
II
I
6
II
I
7
II
I
0
nd
8
Pd2(dba)3
Pd2(dba)3
Pd2(dba)3
54
90
43
81
39
85
67 (−)
92 (R)
67 (−)
93 (R)
67 (−)
91 (R)
II
I
9
II
I
10
CsF
II
a
Reaction conditions: 1a (0.5 mmol), 2A or 3A (1.0 mmol), base (1.5 mmol), ligand (4.8 mol %), Pd (4 mol %), solvent (4 mL); reaction I (used
b
L1 as ligand) was run at 30 °C for 48 h, reaction II (used L15 as ligand) was run at 50 °C for 72 h. Yields were determined by HPLC and NMR
analysis of the crude reaction mixture. Determined by HPLC.
c
the 2′-site of the ligands was also tested (L5−L8). When a
MeO or EtO group was introduced (L5 and L6), 4aA resulted,
in 95% yield. When more sterically congested counterparts
were instead employed (L7 and L8), dehalogenation of the
starting material 1a became significant. This resulted in a lower
yield of the coupling product, despite giving higher ee values
(Table 1, entries 8, 9 vs 10, 11). In place of a 3,5-(tBu)2-4-
MeOC6H2 group in L1 with an electron-rich cyclohexyl group,
disappointingly, poor yield and enantioselectivity were
observed (Table 1, entry 12).
For reaction II, coupling product 5aA was provided with
enantioselectivity in the range of 67−86% ee when the reaction
was conducted at 90 °C for 72 h with chiral bridged L1−L9 as
the ligands. However, the reaction yields of 5aA were poor
because of serious dehalogenation (Table 1, entries 19−27).
Similar phenomena were also observed using BINOL derived
L10−L12 as the ligands (Table1, entries 28−30). Ligands with
1392
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
a
Scheme 3. Synthesis of Axially Chiral Biaryl Monophosphine Oxides via the Suzuki Reaction
a
Reaction conditions: 1 (1.0 mmol), 2 or 3 (2.0 mmol), K3PO4 (3.0 mmol), ligand (4.8 mol %), Pd (4 mol %), toluene (8 mL). Yields are combined
isolated values; ee values were determined by HPLC.
the electron-rich dicyclohexylphosphino group (L13−L15)
gave better yields of 5aA, with 77−84% ee (Table 1, entries
31−33). L15 (Cy-MOP) exhibited the best efficiency among
them (Table 1, entry 33). Using L15 (Cy-MOP) as the ligand,
the reaction proceeded smoothly with excellent enantioselec-
tivity (89% ee) at 70 °C (Table 1, entry 34). Higher ee value
(93% ee) was afforded at a lower temperature 50 °C (Table 1,
entry 35). The reaction did not proceed when the reaction
1393
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
temperature was further decreased to 30 °C, (Table 1, entry
36).
Scheme 4. Synthesis of Axially Chiral Biaryl Monophosphine
Oxides
a
In addition to the ligand screening, the effects of solvents, Pd
sources, and bases were investigated. The results are illustrated
in Table 2. Toluene was found superior to THF and dioxane
solvents. Although the product enantioselectivities were not
affected, dehalogenation of 1a became serious when THF or
dioxane was used (Table 2, entries 1−3). Further survey of
palladium sources indicated that Pd2(dba)3 exhibited better
catalytic performance than Pd(OAc)2, PdCl2, and Pd-
(CF3COO)2 (Table 2, entries 4−6). A control experiment
showed that the reaction did not proceed in the absence of
palladium (Table 2, entry 7). Commonly available inorganic
bases were also screened. K3PO4 was found to be better than
K2CO3, KF, and CsF (Table 2, entries 8−10).
a
Substrate Scope. Having established the optimal con-
ditions, the substrate scope of the reactions was explored
(Scheme 3). Using L1 as the ligand, coupling products 4aB,
4aC, 4aD, and 4aE were afforded in 90−99% yields.
Interestingly, the ee values of 4aB, 4aC, and 4aD were higher
than those of 4aA and 4aE. These results suggest that larger 2-
alkoxy group of the arylboronic acids has a positive influence on
the enantioselectivity; however, for the coupling of 1b (Ar2 = p-
tolyl), 1c (Ar2 = 3,5-xylyl) with 2A, the ee values of the
corresponding products 4bA and 4cA dropped to 64% and
45%, respectively. This implies that the more sterically
congested Ar2 group at the aryl bromide has a negative effect
on the enantioselectivity.
Reaction conditions: 1a (1.0 mmol), boronic acid (2.0 mmol), K3PO4
(3.0 mmol), ligand (4.8 mol %), Pd (4 mol %), toluene (8 mL). Yields
are combined isolated values; ee values were determined by HPLC.
With the aid of this newly developed method, a new chiral
phosphine−oxazoline ligand L16 was prepared from 5aA
(Scheme 5). Oxidation of 5aA by sodium chlorite afforded
Scheme 5. Preparation of Chiral Phosphine−Oxazoline
Ligand L16
When substituted 2-formyl phenylboronic acids with
electron-donating or electron-withdrawing groups (Scheme 3,
3A−3J) were employed as the coupling partners, interestingly,
the coupling products 5aA−5aJ (Scheme 3) were obtained in
moderate-to-good yield with excellent ee’s (90−94% ee’s) with
the use of ligand L15 (Cy-MOP). It suggests that the formyl
group at the 2-position of arylboronic acids shows an
astonishing positive effect on high level of enantioselectivity.
Use of chloro-substituted 2-formyl phenylboronic acids resulted
in relatively lower yields of the coupling products (5aF and
5aI) due to self-coupling reaction of the boronic acids. It is
worth noting that reactions of 1b (Ar2 = p-tolyl) and 1c (Ar2 =
3,5-xylyl) with 3A could also provide the corresponding
products 5bA and 5cA in 89% yields with slightly lower ee’s
(92% and 90% ee’s, respectively) than that of 5aA (93% ee).
This demonstrates that the steric effect of the Ar2 group on
substrate 1 is not significant for the catalytic performance.
Moreover, biphenyl-2-ylboronic acid, 1-naphthenyl boronic
acid, and 4-methylnaphthalen-1-yl boronic acid were also
evaluated (Scheme 4). Ligand L1 showed higher efficacy than
L15 (Cy-MOP) for the coupling reactions of these arylboronic
acids with 1a. The ee values of the corresponding coupling
products 6, 7, and 8 were also lower than those of 5aA−5aJ,
5bA, and 5cA. It also reveals the importance of the formyl
group of the arylboronic acids in acquiring high enantiose-
lectivity in this reaction.
compound 9, which was converted into amide 10 in 82% yield.
The amide was subjected to oxazoline ring formation by
treatment with triethylamine and methanesulfonyl chloride in
dichloromethane to give oxazoline 11. Following reduction of
11 with HSiCl3, a new ligand L16 was obtained in 71% yield.12
COMPUTATIONAL STUDIES ON THE
ENANTIOSELECTIVITIES IN SUZUKI REACTIONS
■
Buchwald’s publication5e,6 suggested that in the asymmetric
Suzuki reaction of 2-alkyl or 2-alkoxy substituted arylboronic
acid with aryl halide, a P(O)(R)2 (R = alky or alkxoy), or
C(O)NRR′ group at the 2-position of the aryl halide is
significant for reaching high levels of enantioselectivity because
of a Pd···O interaction between the palladium(II) atom and the
oxygen atom belonging to the P(O)(R)2 or C(O)NRR′.
However, in our catalyst system, high enantioselectivities were
given in the coupling reactions of 2-diarylphosphinyl-1-
naphthyl bromide (1) with 2-formylphenylboronic acids (3).
In contrast, using 2-alkoxy phenylboronic acids as substrates
instead, only moderate ee values of the coupling products were
observed. It is tempting to speculate that a formyl group at the
2-position of the boronic acids greatly improves the levels of
enantioselectivity.
Application Exploration. To test the practicality of the
methodology, a scaled-up reaction was performed. Coupling of
1a (5 mmol) with 3A (10 mmol) gave product 5aA without
erosion in either the yield (90%) or enantioselectivity (93% ee).
By means of single crystallization in layering dichloromethane/
hexane mixtures, enantiopure crystals of 5aA were obtained
(99% ee). The absolute configuration was confirmed in the R
form by single-crystal X-ray diffraction.11
1394
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
To provide insight into the nature of the 2-position formyl
group and the selectivity in the reaction, computational studies
were performed with the DFT B3LYP methods.13,14 The
effective core potential (ECP) basis sets lanl2dz with double-ζ
basis sets were employed for the Pd and P atoms and the
splitting polarization basis sets 6-31G(d) were adopted for C,
H, and O atoms.
As a starting point of our computational studies, X-ray
crystallographic data were employed in the modeling of the
transition states. We recently reported a closely related reaction
of dimethyl-(1-bromo-2-naphthyl)phosphonate with 2-formyl-
phenylboronic acid catalyzed by the Pd−(Cy-MOP) complex,
which could also afford the coupling product 12 in the R
configuration (Figure 2).8,15 Thus, for the sake of calculation
expense, we chose this closely related reaction to carry out
detailed computational investigations.
The calculated value is 88% ee in favor of R stereochemistry at
25 °C, which is in good agreement with the selectivity found by
experiment of 95% ee.15
The Stability of the Transition State TS2. The
calculation results predict the experimental enantioselectivity
of R well. The ligand in the Pd−(Cy-MOP) complex is more
inclined to coordinate with the palladium(II) ion with both the
P atom and the O atom, which further supports our
experimental finding that one molecule of palladium(II) is
limited to coordinate with one molecule of the ligand for the
sake of steric hindrance and entropy effect. Our calculations
suggest that a transition state with a monodentate ligand is
usually higher in energy than its corresponding bidentate one.5e
Interestingly, TS2 possesses a distorted octahedron config-
urationa in which a bidentate ligand and two carbon atoms bind
to the palladium center on the equatorial plane, and the C−H
motif of the formyl group and the phosphoryl PO of the
phosphonate group are located on two elongated axial position,
respectively. Similar distorted octahedral configurations were
observed from TS6, TS10, and TS14, in which they also have
relatively lower transition state energies. It should be noted that
the fulfilled 4dz2 orbital of the Pd atom is inclined to capture
the electron-deficient H atom in the formyl group, and a typical
three-center, four-electron transition metal hydrogen bond is
generated. As shown in Figure 3, the distance between the H
atom and the Pd atom is 2.45 Å, and that of Pd−C is 2.98 Å in
TS2, which indicates that this significant transition metal
hydrogen bond plays an important role in the stabilization of
TS2. Moreover, there also exists a weak interaction between Pd
atom and the phosphoryl PO group, with a Pd···O distance
of 2.92 Å. NBO population analysis indicates that there is a
remarkable second-order perturbation energy between the
phosphoryl and Pd center to stabilize the transition state
structure. Finally, the origin of the enantioselectivity can be
interestingly figured out by a comparison between TS2 and
TS7 (the former is the most stable transition state for the
product with R stereochemistry, and the latter is the most stable
transition state for the product with S stereochemistry).
As can be seen from Figure 4, TS2 possesses a distorted
octahedron configuration, whereas TS7 adopts a tetragonal
pyramid configuration without the weak interaction between
the Pd atom and the phosphoryl group (Pd···OP distance,
3.55 Å). TS2 and TS2-O are two types of distorted octahedral
transition states for the product with R stereochemistry;
however, the Pd atom is more inclined to form a transition
metal hydrogen bond (Figure 4, TS2), rather than interact with
the formyl oxygen atom (Figure 4, TS2-O). The lone pair
electrons of the formyl oxygen atom would be repelled by the
electron-occupied 4dz2 orbital of the Pd atom. Furthermore,
NBO population analysis shows that the palladium(II) center
has been partial reduced (the NPA charge on Pd is 0.13), which
decreases the electrostatic interaction with the electronegative
oxygen atom in formyl group.
Figure 2. The Pd−(Cy-MOP)-catalyzed reaction of dimethyl-(1-
bromo-2-naphthyl)phosphonate with 2-formylphenylboronic acid.
DFT calculations were carried out to explore the transition
state structures and their relative free energies. Previous
research indicated that the enantioselectivities of the products
in Suzuki−Miyaura reactions were determined in the step of
the reductive elimination.5e,j,16 In the prototypical transition
state for the reductive elimination, both the formylphenyl and
the naphthylphosphonate addends are coordinated with the
palladium(II) ion. In addition, the formyl and phosphonate
substituents on the formylphenyl and naphthylphosphonate
addends, respectively, can be either above or below the plane
defined by the palladium(II) ion and two bounded carbon
atoms. Meanwhile, the formyl group can be inclined away from
or toward the palladium(II) ion. Finally, the ligand in the Pd−
(Cy-MOP) complex can be monodentate or bidentate.
Consequently, a total of 64 transition structures can be
generated from these orientations discussed above (see the
Supporting Information). However, a general principle can be
conducted on the analysis of the relative free energies of these
transition structures. The computational result indicates that
the Pd−(Cy-MOP) complex prefers a bidentate ligand and the
formyl group is more inclined to interact with the palladium(II)
ion with the H atom (See the Supporting Information).
Therefore, 16 selected transition structures of different
orientations were analyzed (Figure 3).
As shown in Figure 3, the TS2 of enantiomer of R is the most
stable transition state, which is consistent with the config-
uration of the major product in the X-ray crystallography
experiment. In addition, distribution analysis (eq 1) based on
these results was conducted to predict the ee value.
To conclude, the distorted octahedron configuration of the
transition state, that is, the bidentate ligand, the significant
transition metal hydrogen bond between H atom of formyl
group and palladium(II), as well as the weak interaction
between Pd center and the phosphoryl oxygen atom, together
make the transition state TS2 the most stable, giving rise to
product with R stereochemistry.This theoretical study well
explains our experimental finding that the formyl group is
crucial for high enantioselectivity.
CR − CS
ee =
CR + CS
∑
exp(−ΔGR≠/RT) − ∑ exp(−ΔGS≠/RT)
ΔGR≠
ΔGR≠
=
∑
ΔGR≠
exp(−ΔGR≠/RT) + ∑ exp(−ΔGS≠/RT)
ΔGR≠
(1)
1395
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
Figure 3. Transition states and corresponding data for reductive elimination of biaryls from Pd−(Cy-MOP) complexes formed after transmetalation.
The stereochemistry of the product followed by the reductive elimination is denoted in brackets. The relative free energies are shown just following
the brackets (kcal/mol). The distances are in Å.
methodology provides a highly efficient and practical strategy
for the synthesis of novel axially chiral biaryl monophosphine
oxides and the corresponding phosphines. Computational
studies show that the origin of high enantioselectivity come
from the feature of the distorted octahedron configuration of
the reductive elimination transition state. In this transition
state, the significant transition metal hydrogen bond between H
CONCLUSION
■
In summary, direct asymmetric Suzuki coupling reaction of
arylboronic acids with 2-diarylphosphinyl-1-naphthyl bromide
was successfully developed for the first time, affording a variety
of axially chiral 2-functionalized-2′-diarylphosphinyl-1,1′-biaryls
with up to 94% ee using Pd-chiral bridged biphenyl
monophosphine L1 or Pd−(Cy-MOP) as the catalyst. This
1396
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397

ACS Catalysis
Research Article
Zhou, Y.; He, Y.; Cao, R.; Qiu, L.; Zhou, Z. Appl. Catal., A 2013, 458,
201−206.
(3) Uozumi, Y.; Suzuki, N.; Ogiwara, A.; Hayashi, T. Tetrahedron
1994, 50, 4293−4302.
(4) (a) Ding, K.; Wang, Y.; Yun, H.; Liu, J.; Wu, Y.; Terada, M.;
Okubo, Y.; Mikami, K. Chem.Eur. J. 1999, 5, 1734−1737. (b) Wang,
Y.; Guo, H.; Ding, K. Tetrahedron: Asymmetry 2000, 11, 4153−4162.
(c) Luo, Y.; Wang, F.; Zhu, G.; Zhang, Z. Tetrahedron: Asymmetry
2004, 15, 17−19. (d) Hoshi, T.; Hayakawa, T.; Suzuki, T.; Hagiwara,
H. J. Org. Chem. 2005, 70, 9085−9087. (e) Hoshi, T.; Sasaki, K.; Sato,
S.; Ishii, Y.; Suzuki, T.; Hagiwara, H. Org. Lett. 2011, 13, 932−935.
(5) (a) Cammidge, A. N.; Crep
1723−1724. (b) Genov, M.; Almorín, A.; Espinet, P. Chem.Eur. J.
2006, 12, 9346−9352. (c) Bermejo, A.; Ros, A.; Fernandez, R.;
́
y, K. V. L. Chem. Commun. 2000, 18,
́
Lassaletta, J. M. J. Am. Chem. Soc. 2008, 130, 15798−15799.
(d) Uozumi, Y.; Matsuura, Y.; Arakawa, T.; Yamada, Y. M. A.
Angew. Chem., Int. Ed. 2009, 48, 2078−2710. (e) Shen, X.; Jones, G.
O.; Watson, D. A.; Bhayana, B.; Buchwald, S. L. J. Am. Chem. Soc.
2010, 132, 11278−11287. (f) Yamamoto, T.; Akai, Y.; Nagata, Y.;
Suginome, M. Angew. Chem., Int. Ed. 2011, 50, 8844−8847. (g) Wu,
W.; Wang, S.; Zhou, Y.; He, Y.; Zhuang, Y.; Li, L.; Wan, P.; Wang, L.;
Zhou, Z.; Qiu, L. Adv. Synth. Catal. 2012, 354, 2395−2402. (h) Lundin,
P. M.; Fu, G. C. J. Am. Chem. Soc. 2010, 132, 11027−11029.
(i) Wilsily, A.; Tramutola, F.; Owston, N. A.; Fu, G. C. J. Am. Chem.
Soc. 2012, 134, 5794−5797. (j) Tang, W.; Patel, N. D.; Xu, G.; Xu, X.;
Savoie, J.; Ma, S.; Hao, M.-H.; Keshipeddy, S.; Capacci, A. G.; Wei, X.;
Zhang, Y.; Gao, J. J.; Li, W.; Rodriguez, S.; Lu, B. Z.; Yee, N. K.;
Senanayake, C. H. Org. Lett. 2012, 14, 2258−2261. (k) Zhang, S.-S.;
Wang, Z.-Q.; Xu, M.-H.; Lin, G.-Q. Org. Lett. 2010, 12, 5546−5549.
Figure 4. Comparison between TS2 and TS7 transition structures.
The stereochemistry of the product followed by the reductive
elimination is denoted in brackets. The relative free energies are
shown just following the brackets (kcal/mol). In TS2-O, the Pd atom
is bound to O atom in the formyl group.
atom of the formyl group and palladium(II) and the weak
interaction between Pd center and the phosphoryl oxygen atom
play important roles in stabilizing the R type transition state
TS2.
ASSOCIATED CONTENT
* Supporting Information
■
́
́
(l) Ros, A.; Estepa, B.; Bermejo, A.; Alvarez, E.; Fernandez, R.;
S
Lassaletta, J. M. J. Org. Chem. 2012, 77, 4740−4750. (m) Xu, G.; Fu,
W.; Liu, G.; Senanayake, C. H.; Tang, W. J. Am. Chem. Soc. 2014, 136,
570−573.
(6) Yin, J.; Buchwald, S. L. J. Am. Chem. Soc. 2000, 122, 12051−
12052.
(7) Wang, S.; Li, J.; Miao, T.; Wu, W.; Li, Q.; Zhuang, Y.; Zhou, Z.;
Qiu, L. Org. Lett. 2012, 14, 1966−1969.
(8) Zhou, Y.; Wang, S.; Wu, W.; Li, Q.; He, Y.; Zhuang, Y.; Li, L.;
Pang, J.; Zhou, Z.; Qiu, L. Org. Lett. 2013, 15, 5508−5511.
(9) L1−L9 were prepared according to the literature procedure.7
L10−L15 were prepared in analogy to the literature method.3
(10) Martin, R.; Buchwald, S. L. Acc. Chem. Res. 2008, 24, 1461−
1473.
Detailed experimental procedures and compound character-
ization data. Energies and Cartesian coordinates of stationary
points from B3LYP calculations. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: qiuliqin@mail.sysu.edu.cn.
*E-mail: kezhf3@mail.sysu.edu.cn.
■
Notes
The authors declare no competing financial interest.
(11) For crystal data of (R)-5aA, see the Supporting Information.
(12) Ogasawara, M.; Yoshida, K.; Kamei, H.; Kato, K.; Uozumi, Y.;
Hayshi, T. Tetrahedron: Asymmetry 1998, 9, 1779−1787.
(13) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(14) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(15) The related reaction was also tested at 25 °C, and the same
enantioselectivity (95% ee) was obtained as at 50 °C. For crystal data
of (R)-12, see the Supporting Information.
ACKNOWLEDGMENTS
■
We are grateful for financial support from the National Natural
Science Foundation of China (NSFC) (20972196, 21272283,
J1103305), Doctoral Fund of Ministry of Education of China
(RFDP) (200805580009), Guangdong Provincial Natural
Science Foundation (S2011010001305), Dayawan District
Science and Technology Program (201202002), and College
Students’ Innovative Experiment Projects of Guangdong
Province.
(16) Joncour, A.; Decor, A.; Liu, J. M.; Dau, M.; Baudoin, O.
Chem.Eur. J. 2007, 13, 5450−5465.
REFERENCES
■
(1) For reviews, see: (a) Hayashi, T. Acc. Chem. Res. 2000, 33, 354−
362. (b) Hayashi, T. Catal. Today 2000, 62, 3−15. (c) Gibson, S. E.;
Rudd, M. Adv. Synth. Catal. 2007, 349, 781−795. (d) Hargaden, G. C.;
Guiry, P. J. Chem. Rev. 2009, 109, 2505−2550.
(2) For selected references, see: (a) Hamada, T.; Chieffi, A.; Åhman,
J.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 1261−1268.
(b) RajanBabu, T. V.; Nomura, N.; Jin, J.; Nandi, M.; Park, H.; Sun, X.
J. Org. Chem. 2003, 68, 8431−8446. (c) Kawashima, Y.; Okano, K.;
Nozaki, K.; Hiyama, T. Bull. Chem. Soc. Jpn. 2004, 77, 347−355.
(d) Xie, X.; Zhang, T. Y.; Zhang, Z. J. Org. Chem. 2006, 71, 6522−
6529. (e) Han, J. W.; Hayashi, T. Tetrahedron: Asymmetry 2010, 21,
2193−2197. (f) Huang, Z.; Liu, Z.; Zhou, J. J. Am. Chem. Soc. 2011,
133, 15882−15885. (g) Li, Q.; Wan, P.; Wang, S.; Zhuang, Y.; Li, L.;
1397
dx.doi.org/10.1021/cs500208n | ACS Catal. 2014, 4, 1390−1397