Functionalized tri- and tetraphosphine ligands as a general approach for controlled implantation of phosphorus donors with a high local density in immobilized molecular catalysts

Article abstract of DOI:10.1002/cplu.201402195

Supported phosphine ligands are auxiliaries of topical academic and industrial interest in catalysis promoted by transition metals. However, both controlled implantation and controlled conformation of ligands should be achieved to produce immobilized catalysts that are able to structurally "copy" efficient homogeneous systems. We provide herein a general synthetic strategy for assembling a new class of branched tetraand triphosphine ligands with a unique controlled rigid conformation, and thus providing a high local density of phosphorus atoms for extended coordination to the metal center. We prepared new functionalized cyclopentadienyl (Cp) salts to design polyphosphines that were "ready for immobilization". These protected Cp fragments might also be of synthetic utility for generating other supported metallocenes. Tetra- and triphosphines were then formed and diversely functionalized for immobilization on virtually any kind of support. As evidenced by ³¹P NMR spectroscopy and the strong "through-space" spin- spin ^TSJ(P,P) coupling observed between P atoms, these modified polyphosphines induce, when immobilized on a support, a high local concentration of phosphorus donors that are spatially very close to each other (3-5 A?). These functionalized ligands have been incorporated into polystyrene to give either soluble or insoluble polymers and on inorganic supports such as silica. We report the catalytic performance at 0.05-0.5 mol% low palladium loading of the resulting immobilized polyphosphine ligands. This study provides proof-of-concept of supported catalytic materials built from modular polyphosphines with a novel approach to structurally controlled polydentate heterogeneous catalysts.

Full text of DOI:10.1002/cplu.201402195

DOI: 10.1002/cplu.201402195
Full Papers
Functionalized Tri- and Tetraphosphine Ligands as
a General Approach for Controlled Implantation of
Phosphorus Donors with a High Local Density in
Immobilized Molecular Catalysts
Matthieu Beaupꢀrin,[a] Radomyr Smaliy,[a] Hꢀlꢁne Cattey,[a] Philippe Meunier,[a] Jun Ou,[b]
Patrick H. Toy,[b] and Jean-Cyrille Hierso*[a, c]
Supported phosphine ligands are auxiliaries of topical aca-
demic and industrial interest in catalysis promoted by transi-
tion metals. However, both controlled implantation and con-
trolled conformation of ligands should be achieved to produce
immobilized catalysts that are able to structurally “copy” effi-
cient homogeneous systems. We provide herein a general syn-
thetic strategy for assembling a new class of branched tetra-
and triphosphine ligands with a unique controlled rigid confor-
mation, and thus providing a high local density of phosphorus
atoms for extended coordination to the metal center. We pre-
pared new functionalized cyclopentadienyl (Cp) salts to design
polyphosphines that were “ready for immobilization”. These
protected Cp fragments might also be of synthetic utility for
generating other supported metallocenes. Tetra- and triphos-
phines were then formed and diversely functionalized for im-
mobilization on virtually any kind of support. As evidenced by
31P NMR spectroscopy and the strong “through-space” spin–
spin TSJ(P,P) coupling observed between P atoms, these modi-
fied polyphosphines induce, when immobilized on a support,
a high local concentration of phosphorus donors that are spa-
tially very close to each other (3–5 ꢂ). These functionalized li-
gands have been incorporated into polystyrene to give either
soluble or insoluble polymers and on inorganic supports such
as silica. We report the catalytic performance at 0.05–0.5 mol%
low palladium loading of the resulting immobilized polyphos-
phine ligands. This study provides proof-of-concept of sup-
ported catalytic materials built from modular polyphosphines
with a novel approach to structurally controlled polydentate
heterogeneous catalysts.
Introduction
The separation and recycling of molecular catalysts from reac-
tions is a scientific challenge with environmental and eco-
nomic implications.[1,2] The “holy grail” in molecular catalysis
would arguably be to combine within the same system the vir-
tues of high activity and selectivity attributed to homogeneous
catalysis, and the technical and commercial advantages of cat-
alyst separation and recycling from heterogeneous systems. In
this respect, many techniques have been developed that have
aimed at the immobilization of catalytic systems with the use
of different solid supports,[3,4] and the development of multi-
phasic reaction conditions.[5] These remarkable advances have
shown that heterogenized species can be in many instances
promising challengers to their homogeneous molecular cata-
lyst counterparts. However, a single universal solution to the
problems of separation, catalyst instability and activity, metal
leaching, and control of single-site molecular catalysts has yet
to be reported.
[a] Dr. M. Beaupꢀrin, Dr. R. Smaliy, Dr. H. Cattey, Prof. Dr. P. Meunier,
Prof. Dr. J.-C. Hierso
Universitꢀ de Bourgogne
Institut de Chimie Molꢀculaire (ICMUB)
UMR-CNRS 6302, 9 avenue Alain Savary
21078 Dijon (France)
Fax: (+33)(0)3 80393682
E-mail: hiersojc@u-bourgogne.fr
Among the auxiliaries of easy access and industrial-scale in-
terest, supported phosphine ligands are currently used in tran-
sition-metal catalysis,[6] and some simple examples are com-
mercially available, such as triphenylphosphine polystyrene
[b] Dr. J. Ou, Prof. Dr. P. H. Toy
The University of Hong Kong
Department of Chemistry
(PSÀPPh3)
and
diphenylphosphinomethyl
polystyrene
(PSÀCH2PPh2). In systems that use such auxiliaries, in addition
to mass-transfer problems, the structural similarity between
the homogeneous molecular catalysts and their corresponding
immobilized analogues is often not exact. As a result, this can
negatively affect reactivity and/or selectivity when compared
with the parent homogeneous catalysts.
Pokfulam Road, Hong Kong (P. R. China)
[c] Prof. Dr. J.-C. Hierso
Institut Universitaire de France (IUF)-France
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201402195. It contains the preparation
details for all CpLi salts, functionalized polyphosphines, and immobilized
ligands; spectroscopic data for new compounds; 31P NMR spectroscopic
simulation for second-order spectra and conformational views of
molecular structures.
Efficient palladium-catalyzed cross-coupling in solution is
usually achieved in the presence of a two- to fourfold excess
ChemPlusChem 2015, 80, 119 – 129
119
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
amount of phosphine ligands relative to the metal center.
Phosphines allow for the stabilization of Pd0/PdII as 14 to 18eÀ
species (i.e., “Pd(PR3)n” with n=2–4, R=aryl or alkyl).[7] Con-
versely, whatever the nature of the support for immobilized
monophosphine ligands, the unavoidable high dispersion of
the coordinating sites on the support renders it improbable for
multiple remote donors to coordinate the same metal center.
In more rigid and/or insoluble supports this issue for sup-
ported molecular catalysts is even more pronounced. Thus, to
provide immobilized single-molecular catalysts able to accom-
modate three or four coordinating ligands to a given metal
center, a controlled implantation of donors with a high local
density must be achieved. Finding a general solution to this
challenge is a driving force of the present study.
Accordingly, these functionalized ligands have been supported
on polystyrene, thereby providing easy access to either soluble
or insoluble materials. These modular functionalized polyphos-
phines were also used for the controlled implantation of
donors onto commercial silica. We report the catalytic per-
formance at 0.05–0.5 mol% palladium loading of the sup-
ported polyphosphine ligands and their recyclability. CarbonÀ
carbon bond formation by Suzuki reaction and an unprece-
dented heterogeneous direct CÀH functionalization of hetero-
aromatics with demanding chloroarene substrates was also
successfully achieved.
Results and Discussion
Polydentate ligands, which are known to form thermally and
kinetically more stable complexes with metal catalyst than
monodentate ones, have been proposed for attachment to
solid supports by Bianchini et al.,[8] Gade et al.[9] and others,[10]
and triphosphine-supported polymers have been developed.[11]
These pioneering investigations have been limited to
functionalization of the tripodal phosphine ligand 1,1,1-tris[(di-
phenylphosphino)methyl]ethane (tdpme; Scheme 1, com-
The lack of supported polyphosphines reported in the litera-
ture might be attributed to the challenging requirements for
the selective and efficient functionalization of the backbone of
such sophisticated ligands.[13] Indeed, the introduction of phos-
phino groups at an early stage of the design sequence renders
subsequent modifications much more difficult. Protection and
deprotection procedures for phosphorus donors might be nec-
essary, and the loss of phosphine during the overall process
could be considerable. We became interested in the challenge
of providing a general access to supported tri- and tetraphos-
phine ligands of controlled conformation based on the ferro-
cene backbone (Scheme 1C and 1D). To the best of our knowl-
edge, no tetradentate phosphine has been implanted on a sup-
port to date, and we have shown that the ferrocene platform
polyfunctionalized with three or four phosphorus atoms gives
access to molecular palladium catalysts with excellent per-
formance in topical homogeneous CÀC,[14] CÀN,[15] and CÀO
(and CÀS)[16] cross-coupling reactions, as well as direct CÀH
functionalization.[17,18]
Functionalized ferrocenyl tetraphosphine ligands ready for
immobilization
To achieve its anchorage on a solid support, the tetraphos-
phine ligand L1 (Scheme 1, and analogous L2 triphosphines)
needs to be modified by the introduction of a reactive handle.
This point of attachment should preferably be positioned far
enough away from the four phosphino groups to keep the
active catalytic site uncluttered. The steric hindrance induced
by the two bulky tert-butyl groups in L1 is essential to be re-
produced since this feature gives a cisoid conformation to such
type of ligands. This specific conformation is at the origin of
the close proximity of donor atoms in solution and in the solid
state. Accordingly, unique spectroscopic[19] and catalytic prop-
erties[14–18] are associated with the presence of bulky tert-butyl
groups on the ferrocenylphosphines. The preservation and
control of the close proximity of the phosphorus atoms in the
supported ligands is thus critical. Considering these prerequi-
sites, we envisioned that the tetraphosphine derivative
1 would be a valuable target (Scheme 2). Compound 1 possess-
es a carbonyl group that could be readily utilized and modified
to achieve attachment of the tetraphosphine motif to a large
variety of supports.
Scheme 1. Polyphosphines functionalized for immobilization on supports:
(i) State-of-the-art tripodal species, and (ii) rigid and modular branched tri-
and tetraphosphines, a new class of ligands.
pounds A and B) despite the wide variety of linear, tripodal,
and other branched polydentate phosphines available.[12]
We provide herein a general synthetic pathway for assem-
bling a new class of branched tetra- and triphosphine ligands
with a controlled rigid conformation. These species are diverse-
ly functionalized for immobilization on virtually any kind of
support. As demonstrated by 31P NMR spectroscopy and the
strong through-space spin–spin TSJ(P,P) coupling observed,
these modified polyphosphines induce, when immobilized on
a support, a high local concentration of phosphorus donors
spatially very close to each other (Scheme 1, compounds C
and D). We demonstrate that the specific conformation of
polyphosphines in which the phosphorus atoms are in close
spatial vicinity is conserved in the modified polyphosphines.
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
120
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 2. Retrosynthetic analysis for the synthesis of functionalized ferro-
cenyl polyphosphine 1 by assembly of cyclopentadienyl blocks.
We reasoned that the two gem-methyl groups in the linker
attached to the cyclopentadienyl rings (Cp) might serve as tert-
butyl mimetic groups for the control of the ferrocene platform
conformation. An acetal group was chosen to protect the car-
bonyl group since it should be stable towards the strongly
basic conditions usually employed in the assembly of ferrocene
derivatives from cyclopentadienyl compounds. Thus, function-
alized polyphosphine 1 was obtained by treating two equiva-
lents of the diphosphino lithium salt derived from the Cp salt
2 with one equivalent of FeCl2 after cleavage of the acetal pro-
tecting groups (Scheme 2). Lithium or sodium Cp salt 2 was
prepared from the commercially available 2-(2-bromoethyl)-
1,3-dioxane (3), which includes both a masked carbonyl group
and a versatile bromide group. The direct synthesis of com-
pound 2 by addition on 6,6-dimethylfulvene of the Grignard
derivative of 3 (or alternatively the nucleophilic lithium deriva-
tive) did not proceed at all.[20] However, bromide 3 could be
converted into the corresponding Grignard reagent, which in
turn could be directly added to acetyl chloride to afford
ketone 4 in 75% yields after purification (Scheme 3).
Scheme 4. Synthesis of tetraphosphines 1 and 7.
drolyzed using acidic conditions and microwave irradiation to
give the desired tetraphosphine dialdehyde 1 in 90% yield
(Scheme 4). Surprisingly, other methods tested for this depro-
tection were found to be unsuccessful.[21]
Gratifyingly, solution 31P NMR spectroscopy of the functional-
ized polyphosphines 7 and 1 confirmed the success of our
strategy in achieving the desired cisoid conformation of ferro-
cenylphosphines by control with gem-dimethyl groups. Typical
AA’BB’ spin system signatures, which highlight the mutual
proximity of the phosphorus atoms in L1,[19] were obtained for
both functionalized polyphosphines 7 and 1. Only minor differ-
ences in chemical shifts and coupling constants were found in
comparison to L1 (Table 1). The existence of a strong “through-
Table 1. 31P NMR spectroscopic data for L1, 7, and 1.[a]
dAA’ [ppm] dBB’ [ppm] J(A,A’) [Hz] J(A,B)=J(A’,B’) [Hz] J(B,B’) [Hz]
L1
7
1
À28.6
À30.4
À31.0
À32.2
À34.4
À35.0
59.8
64.3
61.0
74.5
74.7
73.4
0
0
0
[a] Spectrum at 297 K in CDCl3; d and J values from simulated spectra at
Æ0.1 ppm and Æ0.1 Hz. Lower-field dAA’ is attributed to internal phospho-
rus atoms and dBB’ to peripheral phosphorus atoms (see 1 in Scheme 4).
All other J coupling constants involving the AA’BB’ 31P spin system have
zero values.
Scheme 3. Synthesis of CpLi salt 2 including an acetal group.
The addition was conducted slowly to minimize the compet-
ing formation of 4-bromobutyl acetate from tetrahydrofuran
(THF) ring opening by the organomagnesium intermediate, fol-
lowed by reaction on acetyl chloride. This side reaction was
found to be difficult to avoid since the Grignard reagent could
only be generated in THF. The addition of pyrrolidine on
ketone 4 and subsequent reaction with CpLi gave the 6,6-di-
space” coupling TSJ(A,A’) between internal heteroannular phos-
phorus atoms arises from the overlap of the phosphorus lone
3
3
pairs.[22] In addition, strong J(A,B) and J(A’,B’) values between
homoannular phosphorus are detected, which results mostly
from the overlap of P lone pairs.[23] These coupling constants
dramatically depend on the spatial proximity of the concerned
atoms, and thus attest to the ligand conformation in solution.
Consequently, 31P NMR spectroscopy of samples in solution
supports a similar cisoid conformation for L1, 7, and 1. This
was confirmed in the solid state by an X-ray diffraction struc-
ture of the acetal-functionalized polyphosphine 7 (Figure 1). As
expected, this tetraphosphine bears two acetal groups posi-
tioned on the Cp rings that do not encumber the coordination
sites of the polyphosphine. The ferrocene backbone adopts
a cisoid conformation analogous to the one observed in the
substituted fulvene
5
in 75% yield after purification
(Scheme 3). This diene was treated with MeLi to afford Cp salt
2 in quantitative yield. Two sequential one-pot phosphination
reactions using Cp lithium salt 2 afforded 1,2-bis(diphenylphos-
phino)-4-[4-(1,3-dioxan-2-yl)-2-methylbutyl]
cyclopentadienyl
lithium 6 in 80% overall yield (Scheme 4). Reaction of 6 with
FeCl2 afforded the protected ferrocenyl tetraphosphine 7 in
50% yield after purification. The acetal groups of 7 were hy-
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
121
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
moiety through an ether linkage, but p-vinylbenzylchloride or
iodide did not react appropriately with the corresponding alk-
oxide and instead produced the undesired quaternization of
phosphines (as evidenced by 31P NMR spectroscopy with
a shift from d=À30 to 20 ppm). To avoid tedious sequences
of phosphorus protection/deprotection, we generated the styr-
enyl derivative 9 by means of a Wittig reaction using 1. We al-
ternatively obtained it from the phosphonium salt of p-vinyl-
benzyl iodide, and a Julia–Kocienski olefination with a sulfonyl-
benzothiazole. Polyphosphine 9 was produced in 35–50%
yield using these methods. The preparation of polyphosphine
10 was achieved by reductive amination of 1 with p-vinylben-
zylamine in dichloroethane (DCE; 55% yield). The multinuclear
NMR spectroscopic characterization of the functionalized poly-
phosphines 8–11 confirmed the conservation of the desired
polydentate ligand cisoid conformation for all these com-
pounds.
Although we introduced various functional groups (FGs) effi-
ciently on the metallocene backbone of a tetraphosphine
(Scheme 1), we were also interested in tuning the length of
the linker (Scheme 1) and accessing other useful functionalities.
Therefore, by following a similar strategy, we synthesized vinyl-
functionalized ferrocenyl tetraphosphine 12 (Scheme 6). Such
vinyl groups could potentially allow for further relevant modifi-
cations by means of olefin copolymerization, olefin metathesis,
hydrosilylation, hydroboration, Heck coupling etcetera. In the
tetraphosphine 12 the FG is in the g position, closer to the Cp
ring.
Figure 1. Molecular structure of the acetal-functionalized tetraphosphine 7
(hydrogen atoms are omitted for clarity) showing its cisoid conformation.
Selected geometrical parameters: P1-Ct1-Ct2-P3, À9.90(5)8; P1-Ct1-Ct2-P4,
58.86(5)8; P2-Ct1-Ct2-P3, 57.84(5)8; P1···P3, 3.6737(8) ꢂ.
molecular structure for L1, with small dihedral angles
P1-Ct1-Ct2-P3 of 9.90(5)8 (2.728 for L1; see also the molecular
structure projection in the Supporting Information).
The carbonyl groups of tetraphosphine 1 have to be modi-
fied according to the immobilization mode used and depend-
ing on the solid support. This will guarantee efficient covalent
bonding of the ligand to the
The reaction of 6,6-dimethylfulvene with allylmagnesium
chloride led to the 1-functionalized cyclopentadiene 2a’
(Scheme 6).[24] Our attempts to position the vinyl group in the
b position of the Cp ring by using vinyl-magnesium halides
support. The carbonyl derivative
can be grafted directly on pre-
formed Merrifield-type resins. Al-
ternatively, a styrenyl moiety has
to be introduced for processing
tetraphosphine immobilization
within a polymeric resin, which
would be formed by co-polymer-
ization with styrene. An alkyltrie-
thoxysilane chain has to be
added for anchoring of the
ligand on inorganic supports
such as functionalized silica, or
to alternatively perform the syn-
thesis of a functionalized materi-
al by means of a sol–gel process.
Therefore, tetraphosphine 1 was
modified to give the new func-
tionalized polyphosphines 8–11
(Scheme 5).
Diol 8 was quantitatively ob-
tained from reduction of 1 with
LiAlH4 in THF. From 8, we aimed
to introduce a terminal styrenyl
Scheme 5. Synthesis of tetraphosphines 8–11. Conditions: (1) LiAlH4/THF, 08C, 1 h; (2) triphenyl(4-vinylbenzyl)
phosphonium iodide/NaH, THF, RT (208C), 0.5 h; (3) triacetoxyborohydride, p-vinylbenzylamine, DCE, RT, 15 h; and
(4) from 8: (3-isocyanatopropyl)triethoxysilane/NEt3, DCE, RT, 20 h.
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
122
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 7. Synthesis of the monofunctionalized-tetraphosphines 13–15.
Scheme 6. Synthesis of the vinyl ferrocenyltetraphosphine 12. Conditions:
(1) nBuLi/n-hexane, À808C; (2) ClPPh2/toluene.
symmetric compounds L1 and 7. These phosphines were easily
separated by column chromatography. The 31P spin systems
found for 13–15 support a cisoid conformation for the ferro-
cene backbone. However, they do not resemble the AA’BB’
spin systems of tetraphosphines L1, 7, 1, and 10 (see Table 1).
Indeed, owing to the dissymmetric nature of 13–15, the spin
systems obtained show an additional complexity. This is illus-
trated by very rarely reported intricate ABCD patterns
(Figure 3, top).
failed. Lithiation of 2a’ with nBuLi quantitatively converted it
into 2b. Two successive phosphination/lithiation reactions
using chlorodiphenylphosphine and nBuLi selectively gave 2d
in 70% overall yield. Ferrocenyl polyphosphine 12 was ob-
1
tained by treating FeCl2 with 2d. H and 31P NMR spectroscopy
in solution and single-crystal X-ray diffraction in the solid state
(Figure 2) for the vinylferrocenyl phosphine 12 confirmed the
presence of the vinyl moiety and perfect conservation of
a cisoid conformation for the polyphosphine with a dihedral
angle P1-Ct1-Ct2-P3 of 7.38(5)8 (see also the molecular struc-
ture projection in the Supporting Information).
The simulation of these second-order spectra (Figure 3,
middle) allowed us to determine the 31P NMR spectroscopic
chemical shifts and coupling constants, as exemplified for
tetraphosphine 13 (Figure 3). Thus, four different but proxi-
mate chemical shifts are obtained for the anisochronous phos-
The monofunctionalized analogues 13–15 of the difunction-
alized tetraphosphines 1, 7, and 10 were also synthesized
(Scheme 7). Compound 13 was formed from the appropriate
Cp lithium salt mixture, with a statistical coproduction of the
3
3
phino groups (Figure 3, bottom). Different J(A,B), J(C,D), and
TSJ(B,C) values were obtained, which are consistent with the
values obtained for TSJ(A,A) and 3J(A,B)=3J(A’,B’) in the sym-
metrical ferrocenylphosphine analogues; this ascertained the
controlled conformation of the monofunctionalized tetraphos-
phines 13 and 15 (second-order 31P NMR spectroscopic simula-
tions in the Supporting Information).
Immobilization of functionalized tetraphosphines onto
organic and inorganic supports
To test the reactivity of the modified polyphosphines, we first
achieved the direct anchoring of dialdehyde 1 on a polymeric
support. Compound 1 was treated with a commercially avail-
able mesoporous aminomethylpolystyrene resin (PL-AMS) in
the presence of sodium triacetoxyborohydride in DCE. Since
1 has a deep red color, the originally white polymer beads dis-
played a pink coloration at the end of the reaction that was
conserved after repeated washing of the resin. The grafting
ratio was determined by phosphorus elemental analysis and
a loading of 0.99% P was achieved (0.32 mmolgÀ1). Solid-state
31P NMR (cross-polarization magic angle spinning, CP-MAS) was
performed. This established that no oxidation of phosphine oc-
curred during the immobilization. The corresponding spectrum
showed a broad singlet centered at d=À27.0 ppm (see the
Supporting Information), which roughly corresponds to the
average chemical shift of phosphorus atoms in the original
ligand. From this insoluble resin the presence of only one
Figure 2. Molecular structure of the vinyl-functionalized tetraphosphine 12
(disorder of 1,1-dimethylbut-3-enyl groups and hydrogen atoms are omitted
for clarity). Selected geometrical parameters: P1-Ct1-Ct2-P3, 7.38(5)8; P1-Ct1-
Ct2-P4, À60.11(5)8; P2-Ct1-Ct2-P3, À60.66(5)8; P1···P3, 3.5987(8) ꢂ.
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
123
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 8. Synthesis of polystyrene-supported tetraphosphine 16.
Conversely, styrenyl tetraphosphine 10 was found to be suita-
ble to form with styrene either soluble or insoluble supported
polyphosphine, depending on the experimental conditions.
The radical polymerization of 10 with styrene and azobisisobu-
tyronitrile (AIBN) initiator in toluene at 858C for 48 hours af-
forded, after removal of the solvent, compound 16 (Scheme 8)
as the first non-cross-linked polystyrene-supported tetraphos-
phine. This orange polymer is soluble in most common organic
solvents except for methanol and n-hexane.
The elemental analysis of 16 indicated a phosphorus loading
of 1.05% (0.34 mmolgÀ1), and 31P NMR spectroscopy in CD2Cl2
showed two broad signals centered at d=À30.0 and
À34.0 ppm (Figure 4). These chemical shifts are fully consistent
with those of the non-supported parent ligand, thereby attest-
ing to the non-oxidized state of the phosphorus atoms. The
presence of two signals evidenced in the supported ligand
a differentiation between internal and peripheral phosphorus
atoms. Gratifyingly, this confirmed the expected conservation
of the specific cisoid conformation of the tetraphosphine in
the polymer.
Figure 3. 31P NMR spectra of the ABCD spin system for the tetraphosphine
13 (top), and experimental and simulated enhancement spectra (middle).
Chemical shift (d) for 13 (Æ0.1 ppm) and J constants (Æ0.6 Hz): dA =À33.9,
dB =À29.9, dC =À30.3, dD =À34.0 ppm, J(A,B)=73.3, J(B,C)=61.6,
J(C,D)=74.2 Hz. All other J coupling constants involving the ABCD 31P spin
system have zero values.
strong phosphorus signal is informative of the success of im-
mobilization; however, no fine structure could be obtained to
definitively confirm the preservation of the ligand conforma-
tion.
With a view toward having better control over the immobili-
zation process on a polymeric support, and to tune the solu-
bility properties of the resulting resins, we performed radical
copolymerizations with styrene of the styrenyl-functionalized
tetraphosphines 9 and 10. Copolymerization of styrene with
vinyl-functionalized tetraphosphine 12 was unsatisfactory,
likely owing to reaction kinetics being too different for the
structurally dissimilar monomers, and thus no detectable incor-
poration of phosphine ligands was achieved. Because of the
cross-linking internal double bond in the d position to Cp, mo-
nomer 9 was difficult to incorporate properly into a polystyrene
matrix, and only traces of phosphorus were detected by ele-
mental analysis of the resulting resins formed (<30 mmolgÀ1).
Figure 4. 31P NMR spectra of a solution of polystyrene-supported tetraphos-
phine 16 in CD2Cl2. Distinct signals for peripheral and internal phosphorus
are found at d=À30.0 and À34.0 ppm, respectively. Less than 10% traces of
phosphine oxide are detectable at d=25.0 ppm.
This conformation control for supported phosphines pro-
vides a localized high density of four proximate phosphorus
atoms in the designed polymeric material. Additionally, the for-
mation of an insoluble polystyrene-supported resin was ach-
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
124
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
ieved by incorporating a supplementary cross-linker such as di-
vinylbenzene or the Jandajel cross-linker into the reacting mix-
ture of 15 and styrene.[25] Accordingly, the morphology of solu-
ble and insoluble polymer-supported tetraphosphines (named
15-PS) is illustrated in Figure 5.
To further explore the versatility of our modified polyphos-
phines, we also achieved their immobilization on an inorganic
support. The concept of dialkoxysilane 11 was inspired by
silica-based condensation technology.[26] Phosphine 11 was im-
mobilized on silica gel (0.035–0.070 mm, 60 ꢂ) to give brown-
red beaded powder (11-SiO2), which were found insoluble in
all the solvents tested (details in the Supporting Information).
The corresponding solid-state 31P NMR CP-MAS spectrum indi-
cated a broad singlet centered at À30 ppm.
Stimulated by this preliminary proof-of-concept, we used
our functionalized Cp salts to design dissymmetric modified tri-
phosphines that were “ready for immobilization”, in which the
nature of the substituents on phosphino groups might differ
(Scheme 1D). Such a powerful approach of modular formation
of triphosphine towards immobilized ligands is described
below.
Functionalized ferrocenyltriphosphines as modular ligands
ready for immobilization
Based on the assembly of two different substituted Cp alkali-
metal salts with iron dichloride, the triphosphines 17a,b–
19a,b (Scheme 9) were prepared. Great modularity is achieva-
Scheme 9. Synthesis of modular functionalized triphosphines 17a,b–19a,b
with various PR2 groups (R=aryl, alkyl).
ble by this method of combining the protected salt 6 with an-
other substituted cyclopentadienyl fragment. The characteriza-
tion of functionalized ligands demonstrated the success of
these syntheses with the typical chemical shift for the phos-
phorus atoms, as well as the signature of the FGs (see the Sup-
porting Information). Functionalized triphosphine 19a was
then immobilized by radical copolymerization with styrene
(60 equiv) and divinyl benzene (7 equiv) to give an insoluble
resin with a 1.05% phosphorus content. The strategy we devel-
oped is thus applicable to the entire class of triphosphine li-
gands we have developed in recent years, and which contain
the
1,2-bis(diphenylphosphino)-4-tert-butylcyclopentadienyl
fragment.[27] Clearly, these cyclopentadienyl fragments and
their derivatives might also be of synthetic utility to form other
supported metallocenes such as titanocenes,[28] zircono-
cenes,[29] lanthanocenes[30] etcetera.
We next explored the performance of our set of immobilized
polyphosphine ligands as catalytic materials in palladium-cata-
lyzed reactions for which the homogeneous counterparts have
been notably efficient at low catalyst loading.
Figure 5. Polymeric materials: (a) Standard insoluble cross-linked polystyrene
beads that do not incorporate a polyphosphine ligand (homogeneous dis-
persion 400–500 mm beads); (b) insoluble cross-linked polystyrene beads 15-
PS incorporating tetraphosphine ligand 15 (P loading 0.30 mmolgÀ1, homo-
geneous dispersion 250–300 mm beads); (c) compound 16, as soluble poly-
styrene powder incorporating tetraphosphine ligand 10 (P loading
0.34 mmolgÀ1).
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
125
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Palladium-catalyzed CÀC bond formation with immobilized
reaction products in moderate (50%) to fairly good yield (65%)
by using [Pd/16]. Significantly better performance was ach-
ieved by using the silica-supported system [Pd/11-SiO2]
(formed in situ), with yields above 80% for the coupling of 4-
bromoanisole and 4-bromoacetophenone to phenyl boronic
acid. The performance of supported triphosphine 19a-PS was
investigated in the copper-free Pd-catalyzed Sonogashira cou-
pling of bromobenzene with phenylacetylene; it showed excel-
lent performance (95%) and good potential for recycling. In
general, efficient recycling of the catalysts after six runs has
not been established. This is mainly due to accumulative loss
of material during filtration. For the insoluble catalysts used at
polyphosphine catalysts
Palladium-catalyzed carbonÀcarbon bond formation is a power-
ful set of synthetic tools, which includes reactions between or-
ganometallic nucleophiles and organic halides such as the
Suzuki and Sonogashira cross-couplings. Owing to the necessi-
ty of limiting costly and potentially toxic stoichiometric metal-
lic reagents, research focus has recently shifted to the direct
functionalization of substrates by CÀH activation using aro-
matic and heteroaromatic starting materials. Recoverable or
heterogeneous catalytic systems for direct arylation of CÀH
bonds are limited to very few ex-
amples.[31] The performance of
the new supported polyphos-
Table 2. CÀC bond formation from aryl halides and nucleophiles with immobilized polyphosphine catalysts.[a]
phines at low palladium-catalyst
loading was probed in Suzuki
coupling of aryl bromides with
phenyl boronic acid, in copper-
free Sonogashira coupling with
phenylacetylene, and in the
direct CÀH arylation of furane,
thiophene, and pyrrole sub-
strates with chloroarenes. With
satisfactory conditions estab-
lished for the use of the support-
ed tetraphosphines 16 and 11-
SiO2, and triphosphine 19a-PS,
pertinent results are summarized
in Table 2. Noteworthy, the per-
formance of the insoluble resin
15-PS was poorer under similar
conditions of solvent and tem-
perature (conversion <50%).
Further study is needed to assert
the stability and activity conser-
vation of these catalysts.[1b,32]
Nevertheless, these preliminary
results show satisfactory consis-
tency with the parent homoge-
neous polyphosphine catalysts,
which inspired the present
study, as illustrated in Table 2 for
head-to-head comparisons with
homogeneous counterparts and
with PPh3 and bis-(diphenyl-
phosphino)ferrocene (dppf) li-
gands.
Catalyst [%]
Halide
Nucleophile
Product
Yield [%] (run)
TONs
0.5 mol% Pd/PPh3
0.5 mol% Pd/dppf
0.5% mol% Pd/dppe
0.5% mol% L1
PhCl
PhCl
PhCl
PhCl
PhBr
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhÀPh
PhÀPh
PhÀPh
PhÀPh
PhÀPh
5
38
44
95
80 (1)
77 (2)
82 (3)
75 (4)
72 (5)
68 (6)
10
76
88
190
908
0.5 mol% [Pd/16]
50 (1)
49 (2)
37 (3)
0.5 mol% [Pd/16]
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
272
1300
1700
510
0.05 mol% [Pd/16]
0.05 mol% [Pd/11-SiO2]
0.5 mol% [Pd/11-SiO2]
65
85
85 (1)
87 (2)
83 (3)
95 (1)
92 (2)
90 (3)
0.5 mol% [Pd/19a-PS]
PhBr
554
0.5 mol% [Pd/19a-PS]
0.5 mol% [Pd/19a-PS]
0.5 mol% [Pd/19a-PS]
68[b]
49[b]
23[b]
136
98
The cross-coupling of phenyl-
boronic acid with bromoben-
zene was most efficiently
achieved by using [Pd/16]
(formed in situ; see the Experi-
mental Section and the Support-
ing Information) at 0.5 mol%
(80%). Electron-rich and elec-
tron-neutral bromoarenes gave
46
[a] Conditions: [PdCl(h3-C3H5)]2, 1208C, DMF, 20 h. [b] Pd(OAc)2, 1508C for 40 h in N,N-dimethylacetamide,
2 equiv of tetra-n-butylammonium, yields based on GC analysis and NMR spectroscopy are the average value
of two or more runs.
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
126
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
0.5 mol%, inductively coupled plasma atomic emission spec-
troscopy (ICP-AES) analysis of the filtrate indicated no signifi-
cant leaching (<10 ppm, less than 0.1% of the initial Pd load-
ing). Accordingly, the hot filtration test showed the termination
of the cross-coupling reactions by use of the filtrate recharged
with reagents and bases.
nonequivalent nuclei attest to the conformations, which were
confirmed by X-ray structures. This study finally provides
proof-of-concept of the synthetic utility of these new support-
ed catalytic materials with a localized high density of donor
atoms in CÀC bond-formation reactions.[33] Further study in
this area will aim to optimize the supports and supporting
methods.
The resulting cumulative turnover numbers (TONs; 270–
1700) of the catalysts are satisfactory relative to their homoge-
neous counterparts tetraphosphine L1 and triphosphine L2
when used with chlorides (Table 2, first entries). However, L1
and L2 yielded TONs over 10000 with bromides under homo-
geneous conditions not reached by the supported ligands to
date.[14]
Experimental Section
General procedures
All reactions were performed under an argon atmosphere using
Schlenk techniques. Full details of the preparation and characteri-
zation of modular ligands, their support, and catalytic use are pro-
vided as Supporting Information. The homogeneous polyphos-
phine counterparts to the modular functionalized polyphosphines
presented herein are commercially available under the name
HiersoPHOS (Strem Chemicals).
The supported triphosphine 19a-PS was also investigated in
the coupling of chloroarenes to furane, thiophene, and pyrrole
substrates by direct C2 CÀH functionalization. These reactions
were inspired by the excellent performance of the ferrocenyl-
triphosphines reported in these relevant reactions that elimi-
nate the need for prefunctionalization of substrates for cou-
pling.[17,18] We were glad to observe that these direct arylation
reactions of heteroaromatics, which are typically promoted by
triphosphines under homogeneous conditions,[17] were also
achievable under heterogeneous conditions. The arylation of
2-n-butylfurane with 4-chlorobenzonitrile (68%) and the cou-
pling of 2-n-butylthiophene to 4-chloronitrobenzene (49%)
were achieved more efficiently than the C2 arylation of 1-
methyl-2-formylpyrrole with 4-chlorobenzaldehyde (23%). All
these coupling reactions tolerate a functionalization both on
the chloroarene and heteroaromatic substrates and represent
promising results towards difficult CÀH functionalization with
demanding chloroarenes under “greener” heterogeneous con-
ditions.
Synthesis of 1,1’,2,2’-tetrakis(diphenylphosphino)-4,4’-di[4-
(1,3-dioxan-2-yl)-2-methylbut-2-yl]ferrocene (7)
A solution of 6 (3.1 g, 5.2 mmol) in toluene (10 mL) was added
dropwise at 208C to a suspension of FeCl2 (0.45 g, 3.55 mmol) in
toluene (5 mL). The reaction mixture was then heated to reflux
overnight and removed by filtration. Solvents were removed under
reduced pressure, and the residue was washed with cold ethanol.
When impurities were still present, purification by flash column
chromatography on silica gel (eluent: AcOEt/heptane 1:4) was per-
1
formed to yield 7 (3.2 g, 50%) as an orange-red powder. H NMR
(CDCl3, 600 MHz): d=8.42–6.45 (m, 40H; Ph), 4.47 (m, 2H; OCH1),
4.15–4.06 (m, 8H; 4HCP+ 4OCH6eq,8eq), 3.79 (m, 4H; OCH6ax,8ax), 2.11
(m, 2H; CH7eq), 1.51–1.21 (m, 10H; 2CH7ax +4H2, 4H3), 0.94,
0.10 ppm (s, 6H each, Me); 31P{1H} NMR (CDCl3, 242.9 MHz): d=
À30.4 (AA’ spin system, 2P), À34.4 ppm (BB’ spin system, 2P);
13C NMR (CDCl3, 151 MHz): d=137.5–127.1 (m, 48C), 105.4 (s, 2C),
102.8 (s, 2C), 87.9 (dd, J(C,P)=35.5 Hz, J(C,P)=13 Hz, 2C), 79.4 (m,
2C), 72.7, 72.0 (s, 2C each), 67.0 (s, 4C), 41.7 (s, 2C), 32.8 (s, 2C),
31.0 (s, 2C), 26.9, 28.7 (s, 2C each), 26.0 ppm (s, 2C); MS: m/z calcd
for C76H78FeO4P4 (1235.17): 1257.4092 [M+Na+]; found: 1257.4037.
s=0.030, err=4.42 ppm.
Conclusion
In this study, we have delivered a general strategy for synthe-
sizing unprecedented localized high-density phosphorus donor
atoms in immobilized catalysts. Several reactive functional
groups have been selectively introduced onto cyclopentadien-
yl rings, and these in turn allowed new synthetic pathways to
the first supported rigid branched tetraphosphines, and to sev-
eral modular triphosphines in which different phosphino
groups can be introduced. These cyclopentadienyl fragments
might also be of synthetic utility to form other supported
metallocenes of interest. As a proof-of-concept, the modified li-
gands were supported on organic and inorganic supports ela-
borated by following different methodologies, which include
copolymerization with styrene (with or without a cross-linker),
direct grafting onto preformed polymers, and inorganic sol–gel
condensation. The resulting materials were successfully used
and recycled in low-metal-loading cross-coupling reactions. In
particular, successful results have been obtained in direct CÀH
bond arylation with chloroarenes.
Synthesis of 1,1’,2,2’-tetrakis(diphenylphosphino)-4,4’-di(4-
oxo-2-methylbut-2-yl)ferrocene (1)
A hydrochloric acid solution (2n, 5 mL) was added to a solution of
7 (0.3 g, 0.24 mmol) in THF (15 mL). The reaction mixture was
stirred for 20 min under microwave irradiation (125 W), quenched
by a solution of saturated aqueous sodium hydrogenocarbonate
and extracted with CH2Cl2 (2ꢄ25 mL). The organic layer was dried
over MgSO4, and the solvents were removed under reduced pres-
sure. The crude product was purified by flash column chroma-
tography on silica gel (eluent: AcOEt/heptane 1:4) to yield
1 (0.245 g, 90%) as a red powder. 1H NMR (CDCl3, 300 MHz): d=
9.77 (s, 2H), 8.37–6.48 (m, 40H), 4.18 (s, 2H), 4.04 (s, 2H), 2.20 (m,
4H), 1.54 (m, 4H), 0.97, 0.21 ppm (s, 6H each); 31P{1H} NMR (CDCl3,
121.48 MHz) : d=À31.0 (AA’, 2P), À35.0 ppm (BB’, 2P); 13C NMR
(CDCl3, 75 MHz): d=202.0 (s, 2C), 138.7–127.4 (m, 48C), 105.1 (s,
2C), 88.8 (m, 2C), 80.0 (m, 2C), 72.4, 71.7 (s, 2C each), 40.2 (s, 2C),
37.7 (s, 2C), 32.8 (s, 2C), 28.0, 27.4 ppm (s, 2C each); FTIR: n˜ =
1721.8 cmÀ1 (CO) (full spectrum in the Supporting Information).
Overall, this study provides a new and general approach for
supporting tetraphosphine ligands and modular triphosphines
with highly controlled conformations. The use of 31P NMR spec-
troscopy and “through-space” spin coupling of magnetically
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
127
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Owing to the reactivity of the formyl function, proper mass analy-
ses were not obtained.
Acknowledgements
Support provided by the ANR program for Sustainable Chemistry
Development (ANR-09-CP2D-03 CAMELOT), the Rꢀgion Bourgogne
(3MIM program), and the CNRS is gratefully acknowledged. We
thank Johnson Mattey for a generous gift of palladium. Thanks
also go to B. Hanquet, J. Roger, A. Job, and S. Royer for their
technical assistance.
Synthesis of 1,1’,2,2’-tetrakis(diphenylphosphino)-4,4’-bis{5-
[(3-triethoxysilyl)propylcarbamoyloxy]-2-methyl}pent-2-ylfer-
rocene (11)
Tetraphosphine 1 (50 mg, 44 mmol) was dissolved in CH2Cl2 (1 mL).
The addition of (3-isocyanatopropyl)triethoxysilane (24 mL,
44 mmol) to the reaction mixture was complemented with one
drop of triethylamine. After 20 h at 208C, the solvent of the reac-
tion was evaporated under reduced pressure to yield 11 (70 mg,
Keywords: heterogeneous catalysis
·
immobilization
·
metallocenes · phosphane ligands · through-space interactions
1
99%) as a red powder. H NMR (CDCl3, 500 MHz, 298 K): d=8.42–
3
[1] a) R. Chen, R. P. J. Bronger, P. C. J. Kamer, P. W. N. M. van Leeuwen,
J. N. H. Reek, J. Am. Chem. Soc. 2004, 126, 14557; b) C. W. Jones, Top
Catal. 2010, 53, 942; c) A. E. Collis, I. T. Horvꢅth, Catal. Sci. Technol. 2011,
1, 912; d) Heterogeneized Homogeneous Catalysts for Fine Chemicals Pro-
duction: Materials and Processes (Eds.: P. Barbaro, F. Liguori), Springer,
Dordrecht, 2010; e) Recoverable and Recyclable Catalysts (Ed.: M. Bena-
glia), Wiley, Chichester, 2009.
[2] a) N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102, 3217; b) C. A. McNa-
mara, M. J. Dixon, M. Bradley, Chem. Rev. 2002, 102, 3275; c) Q.-H. Fan,
Y.-M. Li, A. S. C. Chan, Chem. Rev. 2002, 102, 3385; d) C. E. Song, S.-G.
Lee, Chem. Rev. 2002, 102, 3495; e) Z. Wang, G. Chen, K. Ding, Chem.
Rev. 2009, 109, 322; f) J. M. Fraile, J. I. Garcꢆa, J. A. Mayoral, Chem. Rev.
2009, 109, 360; g) A. F. Trindade, P. M. Gois, C. A. Afonso, Chem. Rev.
2009, 109, 418; h) P. Barbaro, F. Liguori, Chem. Rev. 2009, 109, 515; i) R.
Akiyama, S. Kobayashi, Chem. Rev. 2009, 109, 594.
6.45 (m, 40H), 4.7 (s, 2H), 4.17, 4.04 (s, 2H each), 4.02 (t, J=6.5 Hz,
3
3
4H), 3.82 (q, J=7 Hz, 12H), 3.20 (q, J=6.5 Hz, 4H), 1.65 (m, 4H),
3
1.30 (m, 4H), 1.24 (t, J=7 Hz, 18H), 0.99, 0.13 (s, 6H each), 0.87
(m, 4H), 0.63 ppm (s, 4H); 31P{1H} NMR (CDCl3, 202.5 MHz, 298 K):
d=À30.7 (AA’, 2P), À34.6 ppm (BB’, 2P); 13C NMR (CDCl3,
125.7 MHz, 298 K): d=157.1 (s, 2C), 139.1–127.5 (m, 48C; Ph),
105.8 (s, 2C), 88.5 (dd, J(C,P)=13.8, 35.2 Hz, 2C), 79.9 (t, J(C,P)=
18.9 Hz, 2C), 72.7, 72.2 (s, 2C each), 65.6 (s, 2C), 58.8 (s, 6C), 43.8
(s, 2C), 43.3 (s, 2C), 35.7 (s, 2C), 28.9, 27.2 (s, 2C each), 25.1 (s, 2C),
23.7 (s, 2C), 18.6 (s, 6C), 8.0 ppm (s, 2C); MS: m/z calcd for
C90H112P4FeO10N2Si2 (1617.77): 1616.61537 [M+]; found: 1616.61284.
s=0.1892, err=1.3 ppm.
[3] For recent examples, see: a) D. T. Genna, A. G. Wong-Foy, A. J. Matzger,
M. S. Sanford, J. Am. Chem. Soc. 2013, 135, 10586; b) Y. M. A. Yamada, Y.
Yuyama, T. Sato, S. Fujikawa, Y. Uozomi, Angew. Chem. 2014, 126, 131;
Angew. Chem. Int. Ed. 2014, 53, 127; c) T. Iwai, T. Harada, K. Hara, M. Sa-
wamura, Angew. Chem. 2013, 125, 12548; Angew. Chem. Int. Ed. 2013,
52, 12322; d) T. Iwai, R. Tanaka, T. Harada, M. Sawamura, Chem. Eur. J.
2014, 20, 1057.
[4] a) M. Waki, Y. Maegawa, K. Hara, Y. Goto, S. Shirai, Y. Yamada, N. Mizoshi-
ta, T. Tani, W.-J. Chun, S. Muratsugu, M. Tada, A. Fukuoka, S. Inagaki, J.
Am. Chem. Soc. 2014, 136, 4003; b) D. Astruc, F. Chardac, Chem. Rev.
2001, 101, 2991.
[5] a) G.-J. ten Brink, I. W. C. E. Arends, R. A. Sheldon, Science 2000, 287,
1636; b) I. T. Horvath, J. Rabai, Science 1994, 266, 72.
[6] M. Guinꢇ, K. K. (Mimi) Hii, Chem. Soc. Rev. 2007, 36, 608.
[7] a) C. Amatore, M. Azzabi, A. Jutand, J. Am. Chem. Soc. 1991, 113, 8375;
b) A. Jutand, Chem. Rev. 2008, 108, 2300.
Grafting of triethoxysilane ligand 11 on silica gel (11-SiO2)
In a Dean–Stark apparatus, silica gel (1.0 g, 16.64 mmol; Acros,
0.035–0.070 mm, 60 ꢂ) was suspended in toluene (50 mL) and
heated to reflux for 20 h. The silane ligand 11 (0.3 g, 185 mmol) dis-
solved in toluene (15 mL) was then added, and the mixture was
heated to reflux for 5 h. After evaporation of toluene, the resulting
mixture was treated with dichloromethane (20 mL), filtered, dried,
and washed several times with dichloromethane, then dried under
vacuum to give a red powder (1.06 g). The grafted silica was in-
soluble in all the organic solvent tested (THF, CHCl3, CH2Cl2, tolu-
ene, Et2O, ethyl acetate, n-hexane, methanol, N,N-dimethylforma-
mide (DMF), dimethyl sulfoxide (DMSO), N-methyl-2-pyrrolidone
(NMP)). The grafting ratio of 11 was determined, and a 0.84%
phosphorus loading was achieved (0.272 mmolgÀ1 of P,
0.068 mmolgÀ1 of ligand). 31P NMR (CP-MAS; 121.48 MHz, 298 K):
d=À30.0 ppm.
[8] a) C. Bianchini, D. G. Burnaby, J. Evans, P. Frediani, A. Meli, W. Oberhaus-
er, R. Psaro, L. Sordelli, F. Vizza, J. Am. Chem. Soc. 1999, 121, 5961; b) P.
Barbaro, C. Bianchini, V. Dal Santo, A. Meli, S. Moneti, R. Psaro, A. Scaffi-
di, L. Sordelli, F. Vizza, J. Am. Chem. Soc. 2006, 128, 7065.
[9] R. A. Findeis, L. H. Gade, Eur. J. Inorg. Chem. 2003, 99.
[10] P. Schober, G. Huttner, L. Zsolnai, A. Jacobi, J. Organomet. Chem. 1998,
571, 279.
[11] C. Bianchini, A. Meli, M. Peruzzini, F. Vizza, F. Zanobini, Coord. Chem. Rev.
1992, 120, 193.
Direct CÀH arylation of heteroaromatics with chloroarenes
using Pd/polystyrene 19a-PS
[12] a) J.-C. Hierso, R. Amardeil, E. Bentabet, R. Broussier, B. Gautheron, P.
Meunier, P. Kalck, Coord. Chem. Rev. 2003, 236, 143; b) H. A. Mayer, W. C.
Kaska, Chem. Rev. 1994, 94, 1239.
The aryl chloride (1 mmol), heteroaromatic derivative (2 mmol),
KOAc (2 mmol, 196 mg), and Bu4NBr (1 mmol, 322 mg) were intro-
duced into a Schlenk tube equipped with a magnetic stirring bar
and were purged several times with argon. In another tube,
Pd(OAc)2 (0.005 mmol, 1.12 mg) and 19a-PS (0.005 mmol, 69.3 mg)
were mixed in degassed dimethylacetamide (DMAc; 3 mL) and
stirred for 30 min. After addition of the reagents to the catalyst,
the Schlenk tube was placed in an oil bath that had been preheat-
ed to 1508C, and the mixture was stirred for 40 h. The cooled mix-
ture was filtered on paper to recover the catalyst. The solvent was
removed by heating of the reaction vessel under vacuum, and the
residue was charged directly onto a silica gel column. The products
were eluted using diethyl ether/pentane.
[13] R. A. Findeis, L. H. Gade, Dalton Trans. 2003, 249.
[14] a) J.-C. Hierso, A. Fihri, R. Amardeil, P. Meunier, H. Doucet, M. Santelli, B.
Donnadieu, Organometallics 2003, 22, 4490; b) J.-C. Hierso, A. Fihri, R.
Amardeil, P. Meunier, H. Doucet, M. Santelli, V. V. Ivanov, Org. Lett. 2004,
6, 3473.
[15] a) M. Sophal, M. Platon, H. Cattey, H. J. Spencer, P. J. Low, J.-C. Hierso,
Catal. Commun. 2014, 51, 10; b) M. Platon, J. Roger, S. Royer, J.-C.
Hierso, Catal. Sci. Technol. 2014, 4, 2072; c) A. Fihri, J.-C. Hierso, A. Vion,
D.-H. Nguyen, M. Urrutigo¨ıty, P. Kalck, R. Amardeil, P. Meunier, Adv.
Synth. Catal. 2005, 347, 1198.
[16] a) M. Platon, L. Cui, S. Mom, P. Richard, M. Saeys, J.-C. Hierso, Adv. Synth.
Catal. 2011, 353, 3403; b) M. Platon, N. Wijaya, V. Rampazzi, L. Cui, Y.
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
128
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Rousselin, M. Saeys, J.-C. Hierso, Chem. Eur. J. DOI: 10.1002/
chem.201403337.
[17] D. Roy, S. Mom, M. Beaupꢀrin, H. Doucet, J.-C. Hierso, Angew. Chem.
2010, 122, 6800; Angew. Chem. Int. Ed. 2010, 49, 6650.
[18] D. Roy, S. Mom, D. Lucas, H. Cattey, J.-C. Hierso, H. Doucet, Chem. Eur. J.
2011, 17, 6453.
[19] a) J.-C. Hierso, A. Fihri, V. V. Ivanov, B. Hanquet, N. Pirio, B. Donnadieu, B.
Rebiꢁre, R. Amardeil, P. Meunier, J. Am. Chem. Soc. 2004, 126, 11077;
b) J.-C. Hierso, D. Evrard, D. Lucas, P. Richard, H. Cattey, B. Hanquet, P.
Meunier, J. Organomet. Chem. 2008, 693, 574.
[20] This is the classical way to produce tert-butyl cyclopentadienyl lithium,
which is employed for the synthesis of L1.
[21] The other conditions we tested were unsuccessful or led to decomposi-
tion of polyphosphines.
[22] a) J.-C. Hierso, Chem. Rev. 2014, 114, 4838; b) J.-C. Hierso in High Resolu-
tion Nuclear Magnetic Resonance Parameters: Understanding Molecules
and their Electronic Structures (Ed: R. H. Contreras), Elsevier, Amsterdam,
2014.
[29] a) S. Knꢉppel, G. Erker, R. Frçhlich, Angew. Chem. 1999, 111, 2048;
Angew. Chem. Int. Ed. 1999, 38, 1923; b) B. Bosch, G. Erker, R. Froehlich,
O. Meyer, Organometallics 1997, 16, 5449; c) G. Erker, G. Kehr, R. Frçh-
lich, J. Organomet. Chem. 2004, 689, 1402; d) J. A. Pool, E. Lobkovsky,
P. J. Chirik, J. Am. Chem. Soc. 2003, 125, 2241.
[30] a) P. B. Hitchcock, M. F. Lappert, S. Tian, Organometallics 2000, 19, 3420;
b) M. R. Douglass, T. J. Marks, J. Am. Chem. Soc. 2000, 122, 1824.
[31] For the first heterogeneous CÀH bond arylation of furans and thio-
phenes with bromoarenes, see the recoverable system PdNPs@polypyr-
role: a) V. A. Zinovyeva, M. A. Vorotyntsev, I. Bezverkhyy, D. Chaumont,
J.-C. Hierso, Adv. Funct. Mater. 2011, 21, 1064; for C3 regioselective
direct CÀH functionalization of benzo[b]thiophenes with chloroarenes,
see: b) D.-T. D. Tang, K. D. Collins, F. Glorius, J. Am. Chem. Soc. 2013, 135,
7450; for CÀH bond functionalization with iodobenzene see:
c) Ref. [3b]; for CÀH phenylation of naphthalene using diphenyliodoni-
um, see: d) T. H. Park, A. J. Hickman, K. Koh, S. Martin, A. G. Wong-Foy,
M. S. Sanford, A. J. Matzger, J. Am. Chem. Soc. 2011, 133, 20138.
[32] a) M. Lamblin, L. Nassar-Hardy, J.-C. Hierso, E. Fouquet, F.-X. Felpin, Adv.
Synth. Catal. 2010, 352, 33; b) N. T. S. Phan, M. Van Der Sluys, C. W.
Jones, Adv. Synth. Catal. 2006, 348, 609.
[23] O. L. Malkina, V. G. Malkin, Angew. Chem. 2003, 115, 4471; Angew. Chem.
Int. Ed. 2003, 42, 4335.
[24] D. Vos, P. Jutzi, Synthesis 2000, 357.
[25] J. Lu, P. H. Toy, Chem. Rev. 2009, 109, 815.
[26] F. Hoffmann, M. Cornelius, J. Morell, M. Frçba, Angew. Chem. 2006, 118,
3290; Angew. Chem. Int. Ed. 2006, 45, 3216.
[27] V. V. Ivanov, J.-C. Hierso, R. Amardeil, P. Meunier, Organometallics 2006,
25, 989.
[33] These functionalized polyphosphines, the methods for implantation of
P donors on supports, and the catalytic applications at low loading are
patented: J.-C. Hierso, M. Beaupꢀrin, P. Meunier, FR Patent 2961813 A1
20111230, 2011; WO 2012001601 A1 20120105, 2012; US 2013158219
(A1), 2013.
[28] a) W. Wang, M. Fujiki, K. Nomura, J. Am. Chem. Soc. 2005, 127, 4582;
b) A. Gansꢈuer, C. Kube, K. Daasbjerg, R. Sure, S. Grimme, G. D. Fianu,
D. V. Sadasivam, R. A. Flowers, J. Am. Chem. Soc. 2014, 136, 1663.
Received: June 28, 2014
Published online on August 26, 2014
ChemPlusChem 2015, 80, 119 – 129
www.chempluschem.org
129
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim