Self-Supported Ligands as a Platform for Catalysis: Use of a Polymeric Oxime in a Recyclable Palladacycle Precatalyst for Suzuki-Miyaura Reactions

Article abstract of DOI:10.1055/s-0033-1341054

A self-supported oxime palladacycle precatalyst for Suzuki Miyaura reactions was synthesized based on the polyether ether ketone architecture. This precatalyst was found to be highly efficient in Suzuki-Miyaura reactions when aryl bromides were used as substrates, but was less efficient in cross-coupling reactions when aryl chlorides were used. The polymeric palladacycle could be recovered and reused up to four times in such reactions, affording excellent yield of the desired product. The approach represents a novel strategy for generating such self-supported complexes for catalysis.Georg Thieme Verlag Stuttgart. New York.

Full text of DOI:10.1055/s-0033-1341054

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
LETTER
▌1319
lSeteterlf-Supported Ligands as a Platform for Catalysis: Use of a Polymeric Oxime
in a Recyclable Palladacycle Precatalyst for Suzuki–Miyaura Reactions
Use of a Polymeric Oxime in Suzuki–Miyaura Reactions
Yun-Chin Yang, Patrick H. Toy*
Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong, P. R. of China
Fax +852 28571586; E-mail: phtoy@hku.hk
Received: 29.01.2014; Accepted after revision: 02.03.2014
This concept of polymeric functional groups has also been
Abstract: A self-supported oxime palladacycle precatalyst for
applied to self-supported organocatalysts in which the re-
Suzuki–Miyaura reactions was synthesized based on the polyether
ether ketone architecture. This precatalyst was found to be highly
efficient in Suzuki–Miyaura reactions when aryl bromides were
used as substrates, but was less efficient in cross-coupling reactions
when aryl chlorides were used. The polymeric palladacycle could
be recovered and reused up to four times in such reactions, affording
excellent yield of the desired product. The approach represents a
novel strategy for generating such self-supported complexes for ca-
talysis.
(a)
+
M
L
L
M
M
L
L
L
L
L
L
L
(b)
Key words: self-supported catalysts, Suzuki–Miyaura reaction, ox-
imes, palladacycles, polyether ether ketones
M
L
M
L
L
L
L
L
Along with the rapid development of new catalysts and re-
agents for organic synthesis in the past few decades, the
recycling of these occasionally precious and/or toxic ma-
terials for economic and environmental reasons has also
attracted attention. A general method with which to
achieve this goal is to immobilize the catalysts and re-
agents on polymer supports.1 Catalysts and reagents are
often anchored either as pendant groups on the support,
e.g., on cross-linked Merrifield-type polystyrene, or at the
end of a linear polymer, for example, on poly(ethylene
glycol) or polyisobutylene oligomers. In the past decade,
the concept of incorporating organometallic catalysts in
the main chain of a linear polymer or a three-dimensional
polymeric network as ‘self-supported’ catalysts has also
received attention.2 As shown in Figure 1, such self-sup-
ported organometallic catalysts can be divided into two
general categories. The first type of polymeric catalyst is
formed by complexation of a metal and a dimeric ligand
(Figure 1, a). For example, Chen and co-workers reported
polymeric thiourea–Pd complex 1 and its use as a recycla-
ble catalyst for Suzuki–Miyaura reactions (Figure 2).3
More recently, Karimi and Akhavan reported recyclable
self-supported N-heterocyclic carbene (NHC)–Pd com-
plex 2, also for Suzuki–Miyaura reactions.4
= metal
= ligand
M
L
Figure 1 General self-supported organometallic catalyst architec-
tures
N
N
N
N
S
S
Cl
Cl
0 or 1
Pd Cl
Cl Pd
n
1
Bn
Bn
Bn
Bn
Br
Pd
Br
N
N
N
N
Br
Br
Br
Br
Pd
Pd
N
N
N
N
Bn
Bn
Bn
Bn
n
2
Ir
Cl
N
N
The second, and much less common, type of self-support-
ed organometallic catalysts is based on the use of a poly-
meric ligand as the support for the metal (Figure 1, b).5,6
For example, polymeric NHC–Ir complex 3 has recently
been reported by Bielawski and co-workers.7 However, its
catalytic activity was not discussed.
C6H13
C6H13
SYNLETT 2014, 25, 1319–1324
Advanced online publication: 07.04.2014
n
3
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
DOI: 10.1055/s-0033-1341054; Art ID: ST-2014-W0084-L
© Georg Thieme Verlag Stuttgart · New York
Figure 2 Examples of self-supported organometallic complexes

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
1320
Y.-C. Yang, P. H. Toy
LETTER
peating units are the catalytic groups themselves (Figure ether ketone (PEEK) architecture12 could be used as a cat-
3). For example, polyNHC 4, which was used to prepare alyst platform. The most commonly used PEEK material,
3, was employed directly as a recyclable catalyst for ben- poly(oxy-1,4-phenyleneoxy-1,4-phenylenecarbonyl-1,4-
zoin condensation reactions.8 Similarly, recyclable self- phenylene) (6), is inexpensive and commercially avail-
supported ionic polymers bearing chiral quaternary am- able, and has been used as a thermoplastic in numerous
monium salts 5 were prepared by Itsuno and co-workers, applications, including in fuel cell ion exchange mem-
and used to catalyze asymmetric benzylation of a glycine branes (Scheme 1).13 We envisioned that the benzophe-
tert-butyl ester derivative (Figure 4).9,10
none groups of 6 could be converted into oxime groups in
7, and wondered whether this could be used to prepare ox-
ime palladacycle 8. In this way, self-supported oxime li-
gand 7 could serve as a platform for a precatalyst in Pd-
catalyzed cross-coupling reactions.
C
C
C
C
C
C
C
C
= catalyst
Figure 3 General self-supported organocatalyst architecture
O
O
••
O
n
N
N
6
HO
N
Li2PdCl4
O
C6H13
C6H13
O
n
7
HO
Cl
N
Pd
O
n
4
N
O
n
8
R1O
R2
Scheme 1 Proposed synthesis of 8 based on PEEK 6
O3S
SO3
H
N
R3
H
OR1
N
Among the known Pd catalysts and precatalysts, oxime
palladacycles have been demonstrated to be air- and mois-
ture-stable precatalysts that release active palladium spe-
cies into reaction mixtures.14,15 Given the nature of such
palladacycles as reservoirs of catalytically active Pd(0)
species, attaching them to a support or phase-tag facili-
tates the recovery and reuse of the catalysts. Indeed, pal-
ladacycles have been supported on silica (9),
poly(ethylene glycol) (10), cross-linked polystyrene (11),
and other materials.16 Additionally, an oxime palladacycle
with fluorous tails (12) has been reported,17 as has the star-
shaped, self-supported, and recyclable oxime palladacy-
cle precatalyst 13.18 Selected examples of these palladacy-
cles are shown in Figure 5. These supported oxime
palladacycles were effective and recyclable precatalysts
in Suzuki–Miyaura, Heck, Sonogashira, Stille, and
Kumada reactions. However, due to the mass associated
with the supports, the Pd loading of these palladacycles
was quite low — usually less than 0.5 mmol/g. Due to the
structurally simple and efficient PEEK architecture, our
proposed self-supported oxime palladacycle should have
the advantage of higher Pd loading, which, in turn, means
that less catalyst is required for reactions.
X
Na
N
n
5
R1 = H or allyl
R2
=
or
or
R3
=
or
or
or
or
or
Figure 4 Examples of self-supported organocatalysts
Our group has had a long-term interest in investigating the
design and use of new polymer-supported reagents and
catalysts for synthesis.11 We were inspired by the above
reports of self-supported organocatalysts and were inter-
ested to see if a polymeric ligand based on the polyether
Initially we attempted to convert 6 into 7 as outlined in
Scheme 1. However, due the insolubility of 6 in common
organic solvents, this transformation was unsuccessful.
Synlett 2014, 25, 1319–1324
© Georg Thieme Verlag Stuttgart · New York

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
LETTER
Use of a Polymeric Oxime in Suzuki–Miyaura Reactions
1321
N
Cl
OH
OH
Pd
N
O
O
SiO2
or
MCM-41
Si
S
O
11
OH
O
O
OMe
N
9
OH
N
Cl
HO
N
Cl
OH
Pd
Pd
N
O
O
Pd
Cl
O
O
O
O
O
O
O
O
O
136
O
10
O
Cl
OH
Pd
N
O
Cl
O
O
O
Pd
NO2
O
O
N
PS
OH
11
N
OH
2
OH
13
Cl Pd
N
OH
C8F17
3 O
O
3 C8F17
N
n
12
Figure 5 Examples of supported and phase-tagged oxime-palladacycle precatalysts
Therefore, we envisioned that a flexible alkane linker be- bromides (entries 7 and 8), excellent yields were obtain-
tween the benzophenone moieties may make the resulting able within 12 hours. It should be noted that these yields
polymer soluble, and allow for oxime formation and solu- are similar to those reported using other supported oxime-
tion-state spectroscopic characterization of the resulting palladacycle precatalysts shown in Figure 5. The biphenyl
polymer. The synthesis of our designed self-supported ox- products 21 isolated after aqueous workup were essential-
ime palladacycle therefore started with the preparation of ly pure according to 1H and 13C NMR spectroscopic anal-
14 from 4,4′-dihydroxybenzophenone (15) and 1,12-di-
bromododecane (16; Scheme 2).19 The ketone groups of
O
14 were then transformed into oximes in 17, which was
Br(CH2)12Br (16)
then converted into the desired palladacycle 18. It should
be noted that 18 is only slightly soluble in DMF, THF, di-
Cs2CO3, DMF
80 °C, 16 h
30%
HO
OH
15
chloromethane, and chloroform, and is insoluble in water,
ether, and hexanes. Molecular weight determination of 14
by GPC and MALDI-TOF MS was inconclusive due to its
low solubility and lack of suitable standards and matrix.
Importantly, the Pd loading level of 18 was determined by
ICP-MS to be 0.77 mmol/g.
O
H2NOH
pyridine, EtOH
100 °C, 12 h
95%
O
O(CH2)12
n
14
We then set out to examine the performance of 18 as a
self-supported precatalyst for the Suzuki–Miyaura reac-
tions of aryl bromides 19 and phenylboronic acid (20). For
these reactions, a catalytic quantity of 18 was suspended
in the reaction mixture to allow release of the catalytically
active Pd species (Table 1). When the reaction was com-
plete, the reaction mixture was cooled to room tempera-
ture, and ether and water were added to facilitate
precipitation of 18. In this manner 18 could be easily re-
covered by filtration.20 The reactions proceeded smoothly
with activated electron-poor aryl bromides and bromo-
benzene (entries 1–6), and were complete within two
hours. For reactions using deactivated electron-rich aryl
HO
HO
N
Li2PdCl4
THF, r.t., 12 h
90%
O
O(CH2)12
n
17
Cl
HO
N
N
Pd
O
O(CH2)12
O
O(CH2)12
1
1.9
18
Scheme 2 Synthesis of 18
© Georg Thieme Verlag Stuttgart · New York
Synlett 2014, 25, 1319–1324

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
1322
Y.-C. Yang, P. H. Toy
LETTER
Table 2 Suzuki–Miyaura Reactions of 22 with 20a
ysis. These results indicate that 18 is an efficient
precatalyst that can be separated from the desired reaction
product by simple filtration.
18 (1.5 mol%)
Cs2CO3
Cl
B(OH)2
+
R
DMF–H2O, 100 °C
24 h
Table 1 Suzuki–Miyaura Reactions of 19 with 20a
R
22
20
21
18 (1.5 mol%)
Cs2CO3
Br
B(OH)2
Entry
R
Product
Yield (%)b
65
+
R
DMF–H2O, 80 °C
R
O
19
20
21
1
2
3
COMe
NO2
Entry
R
Time (h) Product
Yield (%)b
21a
21b
O
NO2
1
COMe
2
99
40
63
21a
O
NO2
2
3
NO2
2
99
97
COPh
Ph
21b
21c
O
O
H
COPh
2
Ph
4
5
CHO
H
42
0
21c
21d
21f
O
H
4
CHO
2
95
21d
a Reaction conditions: 22 (1.0 mmol), 20 (1.2 mmol), 18 (0.015
mmol), Cs2CO3 (1.5 mmol), DMF (2.5 mL), H2O (2.5 mL), 100 °C,
CN
5
6
7
8
CN
H
2
99
99
95
96
24 h.
b Isolated yield.
21e
2
After examining the reactivity of 18 in a range of Suzuki–
Miyaura reactions, we then examined its recyclability. We
chose 4-bromoacetophenone (19a) as the substrate for
this, using the reaction conditions detailed in Table l. The
self-supported oxime palladacycle 18 collected by filtra-
tion after reaction was directly used for the following run
(Table 3). In this way, 18 was recovered and reused six
times, producing excellent yields of the desired product in
the first five runs. To further investigate the recyclability
of 18, samples of freshly prepared 18 and recovered 18 af-
ter the first run were analyzed by ICP-MS. The results in-
dicated no significant decrease of Pd loading, with used
18 maintaining a loading level of 0.75 mmol/g. The aver-
age Pd content of the products of the first five runs was
0.28 ppm, suggesting that 18 was efficiently removed
from the products during workup. The diminished yields
from run 6 seemed to arise from accumulated physical
loss of 18 during workup of previous runs, for example,
loss of 18 on the filter paper. In addition, gradually formed
inactive palladium black, which was observed during the
course of recycling experiments, may also be partially re-
sponsible for the lower yields after run 5.
21f
Me
OMe
12
21g
OMe
12
21h
a Reaction conditions: 19 (1.0 mmol), 20 (1.2 mmol), 18 (0.015
mmol), Cs2CO3 (1.5 mmol), DMF (2.5 mL), H2O (2.5 mL), 80 °C for
the indicated reaction time.
b Isolated yield.
When aryl chlorides 22 were used as the substrate, the re-
actions proceeded much more sluggishly (Table 2).
Therefore, higher temperatures and longer times were re-
quired to push the reactions towards completion. Howev-
er, even after 24 hours, only activated aryl chlorides
afforded the desired biphenyl products in moderate yields
(entries 1–4). The reaction with chlorobenzene did not
proceed under such conditions (entry 5).
In conclusion, we have prepared a self-supported oxime
palladacycle 18 based on the PEEK architecture as a prec-
atalyst for Suzuki–Miyaura reactions, using readily avail-
Synlett 2014, 25, 1319–1324
© Georg Thieme Verlag Stuttgart · New York