Axial tri-tert-butylphosphane coordination to Rh2(OAc)4: Synthesis, structure, and catalytic studies

Article abstract of DOI:10.1021/acs.organomet.6b00477

The introduction of strong σ-donor axial ligands to the Rh-Rh metal bond has been utilized as an effective way to provide new chemical reactivities to bimetallic dirhodium(II) complexes. In this report, Rh₂(OAc)₄ complexes with axial

Full text of DOI:10.1021/acs.organomet.6b00477

Article
pubs.acs.org/Organometallics
Axial Tri-tert-butylphosphane Coordination to Rh2(OAc)4: Synthesis,
Structure, and Catalytic Studies
Jiantao Tan,†,‡ Yi Kuang,†,‡ Yi Wang,§ Qingfei Huang,*,† Jin Zhu,† and Yuanhua Wang*,§
†Chengdu Institute of Organic Chemistry, Chinese Academy of Sciences, Chengdu 610046, People’s Republic of China
‡University of Chinese Academy of Sciences, Beijing 100049, People’s Republic of China
§College of Chemistry, Sichuan University, Chengdu 610046, People’s Republic of China
S
* Supporting Information
ABSTRACT: The introduction of strong σ-donor axial ligands to the
Rh−Rh metal bond has been utilized as an effective way to provide
new chemical reactivities to bimetallic dirhodium(II) complexes. In
this report, Rh2(OAc)4 complexes with axial bulky alkylphosphane
ligands (PR3), in particular P(t-Bu)3, were prepared and characterized.
The net σ-donation from the PR3 to the Rh−Rh bond is the result of
the competition between the electron-donating ability of the R group
and the steric profile of the PR3 at the Rh2 core. Analysis of the crystal structure data showed that the strong σ-donor P(t-Bu)3
coordinates to the rhodium with an amount of σ-donation to the rhodium similar to that of the aryl phosphane ligand PPh3, but
has an unusually long Rh−P bond distance (2.663 Å). During catalytic trials to synthesize 3-aryl-3-hydroxy-2-oxindole by the
addition of arylboronic acids to isatin derivatives, this longer Rh−P bond distance in Rh2(OAc)4(P(t-Bu)3)2 (Cat-1) facilitates
substitution of one of the axial phosphane ligands by the arylboronic acid. This σ-donating effect greatly accelerated the arylation
reaction in comparison to alternative catalysts. Additionally, Rh2(OAc)4 was easily recovered after completion of the reaction.
provide Rh2 complexes with new chemical reactivities.6 This
simple and efficient strategy was developed on the basis of the
observation that during a reaction only one of the two rhodium
INTRODUCTION
Dirhodium(II) complexes have a well-defined bimetallic dimer
■
structure and are well-known for their effectiveness as catalysts
atoms acts as a catalytic site at a time.1,7 There is a high degree
of potential cooperativity and electronic communication
between the two rhodium metal sites, leading to the hypothesis
that a strongly σ-donating ligand axially coordinated to one of
the Rh atoms would influence the electrophilicity of the other
Rh atom across the Rh−Rh bond.6 Though the effect is rather
subtle, axial ligands such as N-heterocyclic carbene (NHC)
ligands,8a,b phosphanes,8c phosphites,8d pyridines,8e−h and so
forth have previously been known to play an important role in
improving the selectivity for classic metal carbene reactions
catalyzed by Rh2 complexes. Due to the unique structure of Rh2
complexes the axial sites are extraordinarily unrestricted,
making it possible for the various σ-donors to approach each
other.1,2 The use of strong σ-donating axial ligands has led to
many new and useful catalytic reactions. One example is the use
of Rh2/NHC complexes to catalyze arylation reactions between
aromatic aldehydes and arylboronic acids in order to synthesize
secondary alcohols, as reported by Gois and co-workers (Figure
2).9 In this case the NHC ligands were strongly σ-donating
toward the Rh−Rh bond and led to a significant effect on the
Rh2 complex reactivity. It was also determined that NHC
ligands with higher stereodemand, such as N,N-(2,6-
diisopropylphenyl)imidazolidene (IPr), are required in order
to support the stability of the Rh2/NHC complexes as well as
for carbene and nitrene species generation in organic synthesis.1
These complexes are air stable and typically have a dimeric
“paddle wheel” structure with D4h symmetry surrounding a
Rh−Rh bond with four bridging ligands and two axial ligands
(Figure 1).1,2 In order to efficiently tune the electronic
Figure 1. Structure of dirhodium(II) carboxylate catalysts.
properties of the Rh2 complexes as well as their catalytic
selectivity, bridging ligands have been modified to generate a
variety of nitrene- and carbenoid-based dirhodium complexes.3
The axial sites of the complexes are electrophilic and are usually
occupied by solvent molecules which are labile and are easily
displaced by alternative reactants.1−4 Historically it was
understood that the axial ligands played a less important role
in catalysis in comparison to the bridging ligands.3,5 However,
recently the introduction of strong σ-donor axial ligands to the
Rh−Rh metal bond has been used as an effective strategy to
Received: June 13, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Arylation of aldehydes catalyzed by Rh2/axial ligand complexes.
prevent the decoordination of the axial NHC ligands. The
various coordination geometries of the selected σ-donors to the
Rh2 core can result in a change in stereoelectronic effects and
lead to different reactivities for the Rh2 complexes. This was the
case in our research, when an unexpected reactivity was
observed for dirhodium(II) acetate with axial n-butylphosphane
ligands used to catalyze the reaction between aromatic
aldehydes and arylboronic acids. Instead of the production of
secondary alcohols as predicted, the final products were ketones
in neat water which were formed via secondary alcohol
intermediates and then slow oxidation to ketone (Figure 2).10
This difference may be explained by the geometry of the σ-
donors: phosphane ligands have a conical shape, as opposed to
NHC ligands, which have a planar shape. This result also
suggests that the alkylphosphane ligands donate higher electron
density in comparison to arylphosphane ligands and therefore
should be selected as the preferred axial ligands to provide
strong σ-donation to the Rh−Rh bond.
The electron-donating ability of alkyl groups increases
according to the following trend: methyl < ethyl < isopropyl
< cyclohexyl (Cy) < tert-butyl (t-Bu).11 Therefore, the strong σ-
donating ability of the phosphane ligands is due to the large
alkyl substituents, which may result in steric repulsion of the
metal. However, the unrestricted nature of the axial sites on the
Rh2 complex would accommodate axial phosphane ligands with
cone angles as great as 180° without steric difficulty. This
allows for preparation of Rh2 complexes with large axial
alkylphosphane ligands, including tri-tert-butylphosphane (P(t-
Bu)3) and tricyclohexylphosphane (PCy3). The axial coordina-
tion of PCy3 to Rh2(OAc)4 has been previously described by Yu
and co-workers and was seen to facilitate ligand exchange of the
bridging ligands as well as affect the oxidation potential of the
complex during oxidative alkenylation of arenes.12 Chang and
co-workers further proposed that PCy3 and the NHC ligands
would axially coordinate to prepared dirhodium(II) complexes
simultaneously as a precatalyst during C−H activation
reactions.13 It was determined that PCy3 acted to stabilize the
active monocoordinated catalytic species Rh2(OAc)4(NHC).
Although these previous studies indicated that axial alkylphos-
phane ligands have strong effects on Rh2(OAc)4, detailed
structures of the adduct complexes have yet to be reported.
In this work, we continue to expand the understanding of the
fine-tuning effects of axial phosphane ligands on dirhodium(II)
complex chemistry. Here we report the efficient preparation of
Rh2(OAc)4 complexes with large, axially ligated alkylphosphane
ligands, specifically the complexes Rh2(OAc)4(P(t-Bu)3)2 (Cat-
1) and Rh2(OAc)4(PCy3)2 (Cat-2). Detailed crystal structures
are provided, and the catalytic performance of these complexes
toward arylation reactions using arylboronic acids and isatin
derivatives are investigated.
from their acid forms in situ by deprotonating the species using
a base.14 To prepare the Rh2(OAc)4(PR3)2 complexes, the air-
stable ligands tri-tert-butylphosphane tetrafluoroborate ([(t-
Bu)3PH]BF4) and tricyclohexylphosphane tetrafluoroborate
([PCy3H]BF4) were reacted with Rh2(OAc)4 in DME
(dimethoxyethane)/H2O under a nitrogen atmosphere and
using K2CO3 as a base. Happily, the red Cat-1 and brown Cat-2
complexes were seen to precipitate directly from the reaction
vessel and were able to be recovered using simple filtration (eq
1). These obtained complexes were tolerant to degradation by
air and could be stored on the bench.
Rh2(OAc)4 + [(t‐Bu)3PH]BF
4
K2CO3
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ Rh2(OAc)4(P(t‐Bu)3)2
DME/H2O, room temp, N2
Cat‐1
96%
Rh2(OAc)4 + [PCy H]BF
4
3
K2CO3
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ Rh2(OAc)4(PCy )2
3
DME/H2O, room temp, N2
Cat‐2
(1)
95%
Crystals suitable for X-ray diffraction were obtained
according to the following procedure: Rh2(OAc)4 (5.8 mg, 1
equiv), [(t-Bu)3PH]BF4 (44 mg, 10 equiv), and K2CO3 (69 mg,
40 equiv) were placed in a Schlenk tube under a nitrogen
atmosphere followed by the addition of Et2O (1 mL) and H2O
(1 mL). The mixture was observed to immediately turn red and
was stirred at room temperature for 30 min. The mixture was
then allowed to stand and separate, and the upper red ether
solution was transferred to a nitrogen-filled Schlenk tube. Cat-1
single crystals were grown by slow evaporation of the ether
(Figure 3).
Figure 3. ORTEP crystal structure of complex Cat-1. The molecular
structure is depicted in an ellipsoid style at the 50% probability level.
Cat-2 single crystals were prepared following a similar
procedure (Figure 4).
In order to better understand the structural differences
induced by the addition of axial bulky alkylphosphane ligands,
the obtained crystal structures were compared with structures
of previously reported axially ligated dirhodium(II) com-
plexes.9,15 Of particular interest to our investigation was a
comparison of the Rh−Rh bond and Rh−P bond lengths. It
had been reported that the strongly σ-donating axial ligands
coordinated to the rhodium led to a shorter Rh−ligand bond,
while also acting to weaken and lengthen the Rh−Rh bond by
RESULTS AND DISCUSSION
■
A challenge of the proposed synthesis is that P(t-Bu)3 and PCy3
are sensitive to oxidant and are difficult to handle. On the basis
of previous reports, it is possible to release P(t-Bu)3 and PCy3
B
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
of PPh3, the longer Rh−P bond distance acts to weaken the σ-
donating abilities of the axial phosphane ligands toward the
Rh−Rh bond and therefore results in the net σ-donating ability
of P(t-Bu)3 or PCy3 being similar to that of PPh3. Furthermore,
in addition to the electron-donating ability, the steric profile of
the σ-donors also has a great influence on the σ-donation
abilities of the ligands to the metal. The Rh2(OAc)4(PR3)2
complexes may have longer Rh−P bond distances due to the
bulkiness of the phosphane ligands. Steric crowding of bulky
phosphane ligands such as P(t-Bu)3 and PCy3 cause either a
distortion of the coordination geometry at the rhodium or an
adjustment of the internal geometry of the phosphane ligands
themselves.17 Because the axial site of the dirhodium core has
square-planar geometry, the bulky P(t-Bu)3 is not hindered
from approaching the rhodium, resulting in a minimal steric
interaction. In order to determine the amount of distortion to
the coordination geometry at the rhodium, Rh−Rh−P angles
were measured. In the case of Cat-1, the Rh−Rh−P linkage
deviated only slightly (3.5°) from linearity (Table 1, entry 5).
This result indicated that internal geometric distortion in the
phosphane ligand was responsible for the relatively longer Rh−
P bond observed for Rh2(OAc)4(PR3)2 complexes. Angular
symmetric deformation coordinates (S4′ values) were also used
to evaluate the geometric distortion at the phosphorus.18 These
values are directly related to the flattening or pyramidality of
the phosphane ligands and correlate to hybridization of the
lone pair at the phosphorus. On the basis of the data collected
from the crystal structures, the calculated S4′ values clearly
showed that a degree of distortion had occurred at the
phosphorus. The least distorted complex, Rh2(OAc)4(P-
(OPh)3)2, had a large S4′ value (56.3°) (Table 1, entry 3),
while Cat-1 had a much smaller S4′ value (13.9°) (Table 1,
entry 5). This small S4′ value is due to the large t-Bu group in
the P(t-Bu)3 ligand forcing the geometry of P(t-Bu)3 to become
more planar. This considerably flattened phosphane ligand
resulted in poor σ-orbital overlap between the phosphorus and
the rhodium and therefore decreased the σ-donation of the P(t-
Bu)3 toward the Rh−Rh core and elongated the Rh−P bond.
This extraordinarily long Rh−P bond distance also suggests
that Rh−P π back-bonding is also very weak and the P(t-Bu)3 is
purely σ-donating.
Figure 4. ORTEP crystal structure of complex Cat-2. The molecular
structure is depicted in an ellipsoid style at the 50% probability level.
adding electrons into the σ*-orbital of the Rh−Rh MO.16
These effects were measured by examining changes to the Rh−
Rh bond length. For instance, in comparison to the
dirhodium(II) parent structure, the strong σ-donor IPr NHC
9
ligated dirhodium(II) complex Rh2(OAc)4(IPr)2 had a longer
Rh−Rh bond length (2.473 Å) as well as a shorter Rh−ligand
bond distance (Rh−L, 2.228 Å).
As shown in Table 1, among the Rh2(OAc)4(PR3)2
complexes examined, the length of the Rh−Rh bond measured
Table 1. Selected Bond Distances, Bond Angles, and Torsion
Angles from Crystal Structure Data of Selected Axially
Ligated Dirhodium(II) Complexes
axial ligand
(L)
Rh−Rh Rh−O Rh−L Rh−Rh−L
entry
(Å)
(Å)
(Å)
(deg)
S4′ (deg)
1
2
3
4
5
H2O15a
2.385
2.443
2.451
2.457
2.454
2.039
2.039
2.045
2.055
2.046
2.310
2.412
2.477
2.509
2.663
176.47
179.90
175.52
176.39
176.52
15b
P(OPh)3
56.3
35.7
27.3
13.9
15b
PPh3
PCy3
P(t-Bu)3
With the structural information in hand, we then set out to
examine the potential of Rh2(OAc)4(PR3)2 complexes as
catalysts for organic reactions. The arylation reaction of N-Bn
isatin (1a) with phenylboronic acid (2a) was used to investigate
how various axially ligated phosphane ligands affected catalytic
performance.19 First, 1 mol % of Rh2(OAc)4 was combined
with 2.5 mol % of PPh3 in DME/water at room temperature in
order to easily generate the Rh2(OAc)4(PPh3)2 complexes as
red precipitates in situ. This step was followed by the addition
of 1 equiv of N-Bn isatin (1a), 1.1 equiv of phenylboronic acid
(2a), and 5 mol % of K2CO3. The reaction was kept at 90 °C
and allowed to proceed for 4 h, after which the desired product
3aa was obtained in 80% isolated yield (Table 2, entry 1).
Using these same reaction conditions, Rh2(OAc)4 ligated with
the poor σ-donor P(OPh)3 was not able to catalyze this
reaction (Table 2, entry 2). Subsequent control investigations
confirmed that Rh2(OAc)4 and ligands are both necessary for
the reaction to occur (Table 2, entries 3 and 4). Results from
these tests indicate that axial phosphane ligands do act to
increase the reactivity of Rh2(OAc)4. Notably, high yields of
3aa were obtained using Cat-1 and Cat-2 formed in situ (Table
2, entries 5 and 6). Surprisingly, the P(t-Bu)3 ligand accelerated
for the selected axially ligated complexes did indeed increase in
comparison to that of the dirhodium(II) parent structure and
varied overall from 2.443 to 2.457 Å. These data support the
notion that axial ligands with strong σ-donation do indeed
increase the Rh−Rh bond length. However, the data for Rh−P
bond distances measured were not consistent with the previous
report.16 Due to the poor σ-donating ability of P(OPh)3 (Table
15b
1, entry 2), the Rh2(OAc)4(P(OPh)3)2
complex had the
shortest Rh−Rh bond distance measured (2.443 Å) and a
longer Rh−P bond distance (2.412 Å). However, the other
Rh2(OAc)4(PR3)2 complexes with strong σ-donors all had
longer Rh−P bonds (2.477−2.663 Å) (Table 1, entries 3−5) in
comparison to Rh2(OAc)4(P(OPh)3)2. The Rh−P bond
distance of Cat-2 (2.509 Å) (Table 1, entry 4) was slightly
15
longer than that measured for Rh2(OAc)4(PPh3)2 (2.477 Å)
(Table 1, entry 2). It was also noted that the Rh−P bond in
Cat-1 had an unusually long length (2.663 Å) (Table 1, entry
5). As can be seen in Table 1, the Rh−Rh bond distances for
Cat-1 and Cat-2 were 2.454 and 2.457 Å, respectively, and
showed little difference in comparison to the value (2.451 Å)
measured for the analogous PPh3 complex. Although the σ-
donating abilities of P(t-Bu)3 and PCy3 are stronger than that
C
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Table 2. Optimization of Reaction Conditions
b
entry
[Rh2]
ligand
base
time
4 h
yield (%)
1
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4
PPh3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
80
2
P(OPh)3
8 h
trace
n.r.
n.r.
92
3
4 h
4
PPh3
4 h
5
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4
Cat-1
(Cy3PH)BF4
[(t-Bu)3PH]BF4
P(t-Bu)3
4 h
6
20 min
20 min
20 min
4 h
92
7
93
8
96
9
Rh2(OAc)4
Cat-1
[(t-Bu)3PH]BF4
n.r.
n.r.
65
10
11
12
13
14
15
4 h
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4(IPr)2
[(t-Bu)3PH]BF4
[(t-Bu)3PH]BF4
[(t-Bu)3PH]BF4
[(t-Bu)3PH]BF4
KOH
20 min
20 min
20 min
20 min
8 h
NaOH
Na2CO3
NaHCO3
K2CO3
80
90
91
trace
a
Unless otherwise noted, all reactions were carried out with 1a (71.1 mg, 0.30 mmol), 2a (40.0 mg, 0.33 mmol), Rh2(OAc)4 (1.0 mol %), ligand (2.5
b
mol %), and K2CO3 (5.0 mol %) in 1 mL of 1/1 (v/v) DME/H2O at 90 °C under N2. Isolated yield. n.r. = no reaction.
a
Scheme 1. Arylation of Derivatives of Isatins with 2a
a
Unless otherwise noted, all reactions were carried out with 1 (0.30 mmol), 2a (40.0 mg, 0.33 mmol), Rh2(OAc)4 (1.0 mol %), [(t-Bu)3PH]BF4 (2.5
mol %), and K2CO3 (5.0 mol %) in 1 mL of 1/1 (v/v) DME/H2O at 90 °C under N2. Yields of isolated products are reported.
D
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
this arylation reaction and resulted in a reaction time of only 20
min (Table 2, entry 6).20 The end point of the reaction was
determined according to a color change of the reaction mixture
(see the Supporting Information for details). The direct
application of P(t-Bu)3 in the reaction also gave good results
(Table 2, entry 7). The preprepared Cat-1 worked very well as
a catalyst, with an excellent 96% yield of product 3aa obtained
after only 20 min (Table 2, entry 8), demonstrating the
efficiency of this bimetallic Rh(II) catalytic system. In addition,
it was determined that the presence of base was required for the
reaction to progress (Table 2, entries 9 and 10). As a next step,
we briefly screened the base used in the reaction (Table 2,
entries 11−14). The addition of Na2CO3 or NaHCO3 gave
results nearly identical with those seen for K2CO3 (Table 2,
entries 13 and 14), whereas the use of strong bases (KOH and
NaOH) gave lower yields of product (Table 2, entries 11 and
12). Use of Rh2(OAc)4(IPr)2 as a catalyst was also tested and
was shown to be less efficient, a result confirming the
hypothesis that axial phosphane ligands are significantly
different from NHC ligands (Table 2, entry 15).
Scheme 2. Arylation of 1a with Substituted Arylboronic
Acids
a
Once the optimized reaction conditions had been identified,
the next step was to investigate the reaction of isatin substrates
bearing substituents on the nitrogen atom or the aromatic rings
(Scheme 1). The arylation of N-substituted isatins (1b−f)
proceeded efficiently with 2a in DME/H2O and gave the
products in good yields (83−98%) in less than 1 h. For the non
N-substituted isatin 1g, no product 3ga was detected in the
DME/H2O mixture. This is mostly likely due to the inhibiting
effects of the nitrogen atom, which can block the active site of
the Rh2 complex. Next, N-Bn isatin derivatives 1h−n with
various substituents on the aromatic rings were tested under
similar reaction conditions. Derivatives 1h−l reacted smoothly
with 2a to afford the corresponding products 3ha−3la with
yields ranging from 75 to 98%. These results showed that steric
hindrance had obvious effects on the reaction. In the case of N-
Bn-4-bromoisatin (1n) the reaction was unsuccessful, partic-
ularly in comparison to N-Bn-7-fluoro-isatin (1l), which had a
product (3la) yield of 85% in DME/H2O. Notably, the product
of 3ka with the bromo group untouched showed high
selectivity and is therefore attractive for further synthetic
elaboration. It was also determined that the N-Bn isatin
derivatives containing a nitro group (1m) were not tolerated in
this procedure, though with less steric hindrance.
Reactions between several arylboronic acids 2b−i and 1a
were also investigated under the previously determined optimal
conditions (Scheme 2). In the cases of highly electron
withdrawing 4-chlorophenylboronic acid 2e and 4-trifluorome-
thylphenylboronic acid 2g, products 3ae,ag were obtained in
excellent yields. Interestingly, electron-rich 4-methoxyphenyl-
boronic acid 2d also gave good yields (77%) of the arylation
product 3ad. However, 4-cyanophenylboronic acid 2f was an
exception and resulted in only 40% yield of the addition
product 3af. Incubation of 2-hydroxyphenylboronic acid 2b
with 1a afforded the product 3ab in low yields (<20%),
indicating that this particular reaction was not compatible with
a hydroxyl substituent on the aromatic ring of the phenyl-
boronic acid. Bulkier arylboronic acids such as 2-methoxyphe-
nylboronic acid (2c) and 1-naphthylboronic acid (2h) were
reacted with 1a and afforded the desired products 3ac,ah in
good yields. This result demonstrated that there is little steric
hindrance during this transformation, and thus it is possible to
obtain higher yields. The reaction of trans-2-phenylvinylboronic
a
All reactions were carried out with 1a (71.1 mg, 0.3 mmol), 2 (0.33
mmol), Rh2(OAc)4 (1.0 mol %), [(t-Bu)3PH]BF4 (2.5 mol %), and
K2CO3 (5.0 mol %) in 1 mL of 1/1 (v/v) DME/H2O at 90 °C under
N2. Yields of isolated products are reported.
acid 2i with 1a was also investigated, and the desired product
3ai was obtained in high yield (95%).
The next step was to determine the reaction mechanisms
utilized by the catalyst during these reactions. Investigation of
the reactive intermediate was of greatest importance, due to the
possibility that the presence of the axial ligands may cause the
Rh−Rh bond to undergo oxidative cleavage and generate Rh(I)
or Rh(III) species through disproportionation during the
reaction.3 We therefore attempted to recover the Rh2 complex
for analysis after the reaction was complete. Because P(t-Bu)3 is
easily oxidized to PO(t-Bu)3 and is weakly coordinated to the
dirhodium, we added ethyl acetate to the reaction mixture at
the end of the reaction in order to free the Rh2(OAc)4 from the
attached axial ligands. The mixture was stirred in an air
atmosphere for a couple of hours and then allowed to stand and
separate into two phases. The water phase of the mixture
showed a green color, suggesting that Rh2(OAc)4 may be
present in the water.1 UV/vis spectral analysis of the water
phase showed evidence of the characteristic Rh−Rh bond of
Rh2(OAc)4 with a band at 592 nm due to the Rh2 π* to Rh2 σ*
transition.3,21 This spectral feature confirmed that the Rh−Rh
bond remained intact during the reaction (Figure 5).
The fact that the catalyst is water soluble makes recovery
simple and easy, as described in the following steps. The water
phase was collected and washed several times with ethyl acetate
followed by the addition of an equal volume of DME as well as
additions of [(t-Bu)3PH]BF4 and K2CO3 while under nitrogen.
E
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
On the basis of these observations, it was suggested that the
31P NMR signals in Figure 6b would coincide with eq 4. The
signal at 64.8 ppm in Figure 6b may originate from Cat-1, the
signal at 69.5 ppm corresponds to the P−B complex, and the
signal at 67.7 ppm was assigned to the complex A, which results
from 2a reacting with one of the rhodium atoms through a
substitution reaction.
The following 1H NMR study (see the Supporting
Information for details) confirmed our conclusion (Figure 7).
When Cat-1 reacted with 2 equiv of 4-methoxyphenylboronic
acid 2d, the signal for the protons at the (t-Bu)3P group of Cat-
1 (peak b in Figure 7a) split into two peaks (peaks b and d in
Figure 7b), suggesting that the mono- and bis-(t-Bu)3P groups
of Rh2(OAc)4 adducts were in equilibrium. Meanwhile, the
observed peak e in Figure 7b also suggested that the (t-Bu)3P
ligands dissociated from Cat-1. It is worth noting that no
substitution was observed for the same reaction run with
phenylboronic acid ester (Figure 7c). Conversely, we observed
that the signal for protons at the OMe group of 2d (peak c in
Figure 7b) split into three peaks. Considering the result of 31P
NMR, this indicated that the arylbronic acid 2d exists in three
forms in solution: complex A, the P−B complex, and free 2d
(Figure 7b). These results indicated that the arylboronic acid
ligates to the rhodium through dissociative loss of one of the (t-
Bu)3P ligands.
On the basis of the NMR study and activation mechanism for
boronic acid previously described by Gois and co-workers for a
Rh2(OAc)4(NHC)2 catalyst,6,9 we proposed that with the
addition of the axial (t-Bu)3P ligand the dirhodium catalyst
activates the arylboronic acid in the other axial position across
the Rh−Rh bond. Next the aryl group is transferred directly
from the boron to the isatin derivatives through the transition
state (Scheme 3). This procedure is similar to the Petasis
reaction.24 The large acceleration effect due to the P(t-Bu)3
ligand can be attributed to the fact that the much longer Rh−P
bond distance (2.663 Å) in Cat-1 leads to easy and rapid
dissociation of the phosphane ligand.
Figure 5. UV/vis spectra analysis of the water phase (λmax 592 nm).
To our delight, the Cat-1 complex (as confirmed by NMR)
precipitated as a red solid with a 70% recovery yield. Successful
catalyst recovery indicated that the Rh2(OAc)4 remained stable
during the reaction and the addition of axial ligands acted to
activate the bimetallic compound in order to catalyze the
reaction. Alternatively, we also attempted catalyst recycling
experiments in an operationally simple procedure. The
collected water phase containing Rh2(OAc)4 could be used
directly in the next run. The catalyst was still active after three
reaction cycles (see the Supporting Information for details).
On the basis of this understanding, we continued to evaluate
the possible reaction mechanism utilized during catalysis. In
particular, it was important to understand the behavior of the
catalyst during the reaction. It has been reported that the trans-
influence effect influences Rh2(OAc)4(PR3)2 complexes. This
effect takes place when one of the axial phosphane ligands
causes the other phosphane ligand across the Rh−Rh bond to
become much more labile.22 This causes the substrate to react
with one of the rhodium atoms through a substitution reaction
(eq 2).
CONCLUSION
■
In order to investigate this trans-influence effect, we used a
31P NMR study (see the Supporting Information for details) to
observe phosphorus signals during the possible substitution
reaction between Cat-1 and 2a (Figure 6).
In this work, we have developed a convenient method for the
preparation of bimetallic dirhodium(II) acetate complexes with
axially ligated bulky alkylphosphane ligands. Although P(t-Bu)3
would be expected to be a very strong σ-donor, the net donor
ability of the P(t-Bu)3 is the result of competition between the
electron-donating ability of the t-Bu group and the steric profile
of the P(t-Bu)3 on the Rh2 core. From analysis of the collected
crystal structure data, the P(t-Bu)3 ligand coordinates to the
rhodium with an unusually long Rh−P bond distance but still
has a similar amount of σ-donation to the rhodium in
comparison to PPh3. The Cat-1 complex was successfully
applied as a catalyst for the synthesis of 3-aryl-3-hydroxy-2-
oxindole by the addition of arylboronic acids to isatin
derivatives. The long Rh−P bond distance in the Cat-1
complex facilitates substitution of one of the axial phosphorus
ligands by arylboronic acid. The arylboronic acid in this
position was activated due to fine-tuning effects from the
opposite axial phosphane ligand across the Rh−Rh bond.
As shown in Figure 6a, the 31P NMR signal of Cat-1 in
CDCl3 was observed at 65.7 ppm. The change in the 31P NMR
spectra was monitored when Cat-1 reacted with 2 equiv of 2a
(Figure 6b), and three 31P NMR signals appeared at 64.8, 67.7,
and 69.5 ppm. A further experiment showed that P(t-Bu)3 also
would react with 2a to form the P−B complex (eq 3),23 which
had a 31P NMR signal at 70.2 ppm (Figure 6c) that was close to
the signal at 69.5 ppm in Figure 6b.
F
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 6. 31P NMR study of the substitution reaction between Cat-1 and 2a: (a) 31P NMR of Cat-1; (b) 31P NMR of Cat-1 with 2 equiv of 2a; (c)
31P NMR of P(t-Bu)3 with 2 equiv of 2a.
Figure 7. 1H NMR study of the substitution reaction between Cat-1 and 2d: (a) 1H NMR of Cat-1; (b) 1H NMR of Cat-1 with 2 equiv of 2d; (c)
1H NMR of Cat-1 with 2 equiv of phenylboronic acid neopentylglycol ester.
shifts (δ) are reported in ppm and coupling constants (J) in Hz. The
following abbreviations were used to designate the multiplicities: s =
singlet, d = doublet, dd = doublet of doublets, t = triplet, m =
multiplet. The high-resolution mass spectra (HRMS) were obtained
with a Waters-Q-TOF Premier (ESI) instrument. Elemental analyses
were performed with a Vario MICRO select instrument.
Future investigation of the catalytic potential of these
complexes will be extended to determine the application of
this effective catalyst system to chemoselective reactions.
EXPERIMENTAL SECTION
■
General Considerations. All manipulations were carried out
under a nitrogen atmosphere. Unless otherwise noted, all reagents
were obtained from commercial suppliers and used without further
purification. Rh2(OAc)4, tri-tert-butylphosphane, and [(t-Bu)3PH]BF4
were purchased from Admas-beta. Rh2(OAc)4(IPr)2 was prepared
according to the published methods.9 Column chromatography was
generally performed on silica gel (200−300 mesh), and reactions were
monitored by thin-layer chromatography (TLC) using UV light to
Synthesis of Cat-1. Rh2(OAc)4 (79 mg, 1.0 equiv), [(t-
Bu)3PH]BF4 (129 mg, 2.5 equiv), and K2CO3 (123 mg, 5.0 equiv)
were placed in a Schlenk tube under a nitrogen atmosphere followed
by the addition of DME (1 mL) and H2O (1 mL). The reaction
mixture was stirred at room temperature for 30 min. The red complex
was precipitated and separated by filtration. The solid was washed with
DME/H2O (1/1 v/v) and dried under vacuum to give Cat-1 as a red
solid. Yield: 146 mg (96%). Anal. Calcd for Rh2(OAc)4(P(t-Bu)3)2 (M
1
visualize the course of the reactions. H NMR and 13C NMR spectra
1
= 846.23 g/mol): C, 45.40; H, 7.86. Found: C, 45.38; H, 7.80. H
were measured on a 300 or 400 MHz Bruker spectrometer using
CDCl3 or d6-DMSO as the solvent at room temperature. The chemical
NMR (400 MHz, CDCl3): δ 1.82 (s, 12H), 1.50 (s, 54H).
G
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 3. Possible Mechanism
REFERENCES
■
(1) (a) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic
Methods for Organic Synthesis with Diazo Compounds; Wiley: New
York, 1998. (b) Ye, T.; McKervey, M. A. Chem. Rev. 1994, 94, 1091−
1160. (c) Doyle, M. P.; Duffy, R.; Ratnikov, M.; Zhou, L. Chem. Rev.
2010, 110, 704−724. (d) Davies, H. M. L.; Morton, D. Chem. Soc. Rev.
2011, 40, 1857−1869.
(2) Paulissen, R.; Reimlinger, H.; Hayez, E.; Hubert, A. J.; Teyssie, P.
Tetrahedron Lett. 1973, 14, 2233−2236.
(3) (a) Timmons, D. J.; Doyle, M. P.; Cotton, F. A.; Murillo, C. A.;
Walton, R. A. Multiple Bonds between Metal Atoms; Springer Science
and Business Media: New York, 2005. (b) Catalysis by Di- and
Polynuclear Metal Cluster Complexes; Adams, R. D., Cotton, F. A., Eds.;
Wiley-VCH: New York, 1998.
(4) Boyar, E. B.; Robinson, S. D. Coord. Chem. Rev. 1983, 50, 109−
112.
(5) Doyle, M. P.; Ren, T. Prog. Inorg. Chem. 2001, 49, 113−168.
(6) Trindade, A. F.; Coelho, J. A. S.; Afonso, C. A. M.; Veiros, L. F.;
Gois, P. M. P. ACS Catal. 2012, 2, 370−383.
(7) (a) Pirrung, M. C.; Liu, H.; Morehead, A. T. J. Am. Chem. Soc.
2002, 124, 1014−1023. (b) Pirrung, M. C.; Morehead, A. T. J. Am.
Chem. Soc. 1996, 118, 8162−8163.
Synthesis of Cat-2. Rh2(OAc)4 (79 mg, 1.0 equiv), (Cy3PH)BF4
(160 mg, 2.5 equiv), and K2CO3 (123 mg, 5.0 equiv) were placed in a
Schlenk tube under a nitrogen atmosphere followed by the addition of
DME (1 mL) and H2O (1 mL). The reaction mixture was stirred at
room temperature for 30 min. The brown complex was precipitated
and separated by filtration. The solid was washed with DME/H2O (1/
1 v/v) and dried under vacuum to give Cat-2 as a brown solid. Yield:
170 mg (95%). Anal. Calcd for Rh2(OAc)4(PCy3)2 (M = 1002.33 g/
(8) (a) Gomes, L. F. R.; Trindade, A. F.; Candeias, N. R.; Veiros, L.
F.; Gois, P. M. P.; Afonso, C. A. M. Synthesis 2009, 2009, 3519−3526.
(b) Gomes, L. F. R.; Trindade, A. F.; Candeias, N. R.; Gois, P. M. P.;
Afonso, C. A. M. Tetrahedron Lett. 2008, 49, 7372−7375. (c) Wynne,
D. C.; Olmstead, M. M.; Jessop, P. G. J. Am. Chem. Soc. 2000, 122,
7638−7647. (d) Zaykov, A. N.; Ball, Z. T. Tetrahedron 2011, 67,
4397−4401. (e) Saito, H.; Iwai, R.; Uchiyama, T.; Miyake, M.; Miyairi,
S. Chem. Pharm. Bull. 2010, 58, 872−874. (f) Saito, H.; Uchiyama, T.;
Miyake, M.; Anada, M.; Hashimoto, S.; Takabatake, T.; Miyairi, S.
Heterocycles 2010, 81, 1149−1155. (g) Marcoux, D.; Lindsay, V. N. G.;
Charette, A. B. Chem. Commun. 2010, 46, 910−912. (h) Lindsay, V. N.
G.; Nicolas, C.; Charette, A. B. J. Am. Chem. Soc. 2011, 133, 8972−
8981.
1
mol): C, 52.70; H, 7.84. Found: C, 52.67; H, 7.79. H NMR (400
MHz, CDCl3): δ 2.40 (m, 6H), 2.10−2.07 (m, 12H), 1.78−1.56 (m,
48H), 1.36−1.26 (m, 12H).
General Procedure for Dirhodium(II)-Catalyzed Arylation of
Isatin Derivatives. A mixture of added isatins (0.30 mmol, 1.0
equiv), arylboronic acid (0.33 mmol, 1.1 equiv), Rh2(OAc)4 (0.003
mmol, 1.0 mol %), [(t-Bu)3PH]BF4 (0.0075 mmol, 2.5 mol %), and
K2CO3 (0.015 mmol, 5.0 mol %) was evacuated and purged three
times with N2. DME/H2O (1 mL, 1/1 v/v) was added, and the
mixture was stirred at 90 °C. After the reaction was complete (as
monitored by TLC), the reaction mixture was extracted with ethyl
acetate (3 × 5 mL) and washed with water (2 × 10 mL). The ethyl
acetate layer was separated and dried over Na2SO4. After evaporation
of the solvent, the residue was purified by flash column
chromatography (ethyl acetate/hexane) to give the desired products.
(9) (a) Trindade, A. F.; Gois, P. M. P.; Veiros, L. F.; Andre, V.;
Duarte, M. T.; Afonso, C. A. M.; Caddick, S.; Cloke, F. G. N. J. Org.
Chem. 2008, 73, 4076−4086. (b) Gois, P. M. P.; Trindade, A. F.;
Veiros, L. F.; Andre, V.; Duarte, M. T.; Afonso, C. A. M.; Caddick, S.;
Cloke, F. G. N. Angew. Chem., Int. Ed. 2007, 46, 5750−5753.
(10) Kuang, Y.; Wang, Y. H. Eur. J. Org. Chem. 2014, 2014, 1163−
1166.
(11) Tolman, C. A. Chem. Rev. 1977, 77, 313−348.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00477.
(12) Vora, H. U.; Silvestri, A. P.; Engelin, C. J.; Yu, J. Q. Angew.
Chem., Int. Ed. 2014, 53, 2683−2686.
■
S
(13) Kwak, J.; Kim, M.; Chang, S. J. Am. Chem. Soc. 2011, 133,
3780−3783.
(14) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3, 4295−4298.
(15) (a) Cotton, F. A.; DeBoer, B. G.; LaPrade, M. D.; Pipal, J. R.;
Ucko, D. A. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem.
1971, 27, 1664−1671. (b) Christoph, G. G.; Halpern, J.; Khare, G. P.;
Koh, Y. B.; Romanowski, C. Inorg. Chem. 1981, 20, 3029−3037.
(16) (a) Berry, J. F. Dalton Trans. 2012, 41, 700−713. (b) Christoph,
G. G.; Koh, Y. B. J. Am. Chem. Soc. 1979, 101, 1422−1434.
(17) Huheey, J. E.; Keiter, E. A.; Keiter, R. L.; Medhi, O. K. Inorganic
Chemistry: Principles of Structure and Reactivity; Pearson Education:
London, 2006.
Experimental procedures, pictures of the reaction mixture
of 1a and 2a, NMR studies, catalyst recycling experi-
ments, experimental characterization data for the
products, and H and 13C spectra for the products
(PDF)
X-ray crystallographic data for Cat-1 and Cat-2 (CIF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail for Q.H.: qfhuang@cioc.ac.cn.
*E-mail for Y.W.: yhwang@scu.edu.cn.
(18) (a) Cooney, K. D.; Cundari, T. R.; Hoffman, N. W.; Pittard, K.
A.; Temple, M. D.; Zhao, Y. J. Am. Chem. Soc. 2003, 125, 4318−4324.
(b) Dunne, B. J.; Morris, R. B.; Orpen, A. G. J. Chem. Soc., Dalton
Trans. 1991, 653−661. (c) Kendall, A. J.; Zakharov, L. N.; Tyler, D. R.
Inorg. Chem. 2016, 55, 3079−3090.
(19) (a) Lai, H. S.; Huang, Z. Y.; Wu, Q.; Qin, Y. J. Org. Chem. 2009,
74, 283−288. (b) Yamamoto, Y.; Yohda, M.; Shirai, T.; Ito, H.;
Miyaura, N. Chem. - Asian J. 2012, 7, 2446−2449. (c) Shintani, R.;
Inoue, M.; Hayashi, T. Angew. Chem., Int. Ed. 2006, 45, 3353−3356.
(d) Zhang, J. L.; Chen, J. X.; Ding, J. C.; Liu, M. C.; Wu, H. Y.
Tetrahedron 2011, 67, 9347−9351. (e) Toullec, P. Y.; Jagt, R. B. C.; de
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful for financial support from National Science
Foundation of China (Grant No. 21272162).
■
H
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Vries, J. G.; Feringa, B. L.; Minnaard, A. J. Org. Lett. 2006, 8, 2715−
2718.
(20) Ueda, M.; Miyaura, N. J. Org. Chem. 2000, 65, 4450−4452.
(21) Kornecki, K. P.; Berry, J. F.; Powers, D. C.; Ritter, T. Prog. Inorg.
Chem. 2014, 58, 225−302.
(22) Aquino, M. A. S.; Macartney, D. H. Inorg. Chem. 1987, 26,
2696−2699.
(23) Holthausen, M. H.; Bayne, J. M.; Mallov, I.; Dobrovetsky, R.;
Stephan, D. W. J. Am. Chem. Soc. 2015, 137, 7298−7301.
(24) (a) Petasis, N. A.; Akritopoulou, I. Tetrahedron Lett. 1993, 34,
583−586. (b) Petasis, N. A.; Zavialov, I. A. J. Am. Chem. Soc. 1997,
119, 445−446. (c) Petasis, N. A.; Zavialov, I. A. J. Am. Chem. Soc.
1998, 120, 11798−11799.
I
DOI: 10.1021/acs.organomet.6b00477
Organometallics XXXX, XXX, XXX−XXX