Amino-acid containing metallomonomers copolymerized into porous organic polymers: Applicability to allylic alkylation catalysis

Article abstract of DOI:10.1016/j.ica.2004.04.017

This paper reports the synthesis of a polymerizable dppe-derivative with eight pendent styrenyl units linked to the dppe core through four tyrosine residues. This diphosphine ligand was utilized in the synthesis of a series of P₂PdX₂ complexes (X₂=(R)-BINOL, (S)-BINOL, Cl₂ and π-1,3-Ph₂-allyl⁺). The compounds were used as comonomers for the synthesis of porous organic polymers (poly EDMA). Molecular imprinting effects on X₂-ligand removal were investigated. Overall, the presence of a cavity of significant size was found to be beneficial to the rate of the allylic alkylation reaction, however, chiral BINOL shaped cavities did not influence the enantio-selectivity of the reaction.

Full text of DOI:10.1016/j.ica.2004.04.017

Inorganica Chimica Acta 357 (2004) 3854–3864
www.elsevier.com/locate/ica
Amino-acid containing metallomonomers copolymerized into
porous organic polymers: applicability to allylic alkylation catalysis
*
€
Hassan Aıt-Haddou, Shannon M. Leeder, Michel R. Gagne
ꢀ
Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-3290, USA
Received 15 March 2004; accepted 12 April 2004
Available online 26 June 2004
Dedicated to Prof. Tobin J. Marks on his 60th birthday
Abstract
This paper reports the synthesis of a polymerizable dppe-derivative with eight pendent styrenyl units linked to the dppe core
through four tyrosine residues. This diphosphine ligand was utilized in the synthesis of a series of P2PdX2 complexes (X2 ¼ (R)-
BINOL, (S)-BINOL, Cl2 and p-1,3-Ph2-allylþ). The compounds were used as comonomers for the synthesis of porous organic
polymers (poly EDMA). Molecular imprinting effects on X2-ligand removal were investigated. Overall, the presence of a cavity of
significant size was found to be beneficial to the rate of the allylic alkylation reaction, however, chiral BINOL shaped cavities did not
influence the enantio-selectivity of the reaction.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Catalysis; Molecularly imprinted polymers; Allylation
1. Introduction
speculated that organized functionality might provide a
means to improve the molecular recognition properties of
the artificial active sites [3,4]; a hypothesis grounded in the
fact that metalloenzymes utilize functional groups to
construct and outfit their active sites [5–7]. To this end we
have synthesized a series of diphosphine complexes that
contain polymerizable amino acid units [8,9] for copoly-
merizing into the matrix of the host polymer, hoping that
the basic carbonyl or H-bonding amide might provide the
desired functionality to the active site [10]. We report
herein the synthesis and characterization of several
P2PdX2 [11] complexes, their polymerization into the
matrix of highly crosslinked methacrylate polymers and
their subsequent activation towards catalysis of the allylic
alkylation reaction [12–16].
The synthesis of artificial active sites [1] for hosting
transition metal catalysts with unique reactivity profiles is
an important problem, but one also fraught with technical
difficulties. Our first forays into this area focused on the
molecular imprinting of transition metal complexes into
styrene/divinylbenzene or ethylene dimethacrylate
(EDMA)-based networks for creating artificial ‘‘active
sites’’, wherein the metal complex had an associated chiral
cavity whose topology could be influenced by the shape of
a large, removable imprinting ligand on the metallomo-
nomer (Scheme 1). When (R)-BINOL shaped cavities
were created, the metal sites were capable of differentiat-
ing between (R)- and (S)-BINOL in rebinding experiments
(85:15 R:S). In fact some of the sites were capable of
providing up to 97:3 selectivities, though others were only
2:1 selective; the metal fragment is of course achiral [2].
Since these cavities were generally constructed of non-
functionalized monomers (styrene, EDMA, etc.), we
2. Results and discussion
2.1. Synthesis of a tyrosine-containing, polymerizable
diphosphine ligand [17]
* Corresponding author. Tel.: +9199626341; fax: +9199626342.
Synthesis of the desired compounds began with a
suitably protected reactive diphosphine reagent for
ꢀ
E-mail address: mgagne@unc.edu (M.R. Gagne).
0020-1693/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.04.017

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3855
Bu
Bu
3% P2Pt(Bu-BINOL)
97% EDMA
equal voume PhCl
AIBN
P
P
P
P
O
O
O
O
R
Pt
R
Pt
Bu
Bu
- R-Bu2BINOL
R-BINOL
Shaped
Cavity
P
P
L
L
Pt
Scheme 1.
coupling to an amino acid ester. The diphosphine 2 was
synthesized in high yield by the addition of the Grignard
generated from the 4-iodobenzoate ester 1 [18] to
bis(dichlorophosphino)ethane and borane protection
(Scheme 2). The tetra ester was transformed to its cor-
responding acid 3 by the action of TBAF and acidifi-
Scheme 2.

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3856
cation. The amino-acid bearing polymerizable arm 5
was prepared in two steps from N-Boc- -tyrosine, first
2.2. Synthesis of palladium metallomonomers
L
by double alkylation with 4-vinylbenzyl chloride then
Boc removal. Coupling of this amino ester with the di-
phosphine tetra acid chloride generated from oxalyl
chloride/DMF provided the borane protected product,
which was most conveniently deprotected with ethyl-
amine to yield 7. This sequence provided multigram
quantities of diphosphine 7. Double benzylation of the
amino acid was found to be critical to providing di-
phosphines that were soluble in organic solvents; other
amino acids and alkylation schemes were abandoned
due to insufficient solubility.
The reaction of 7 with (CH3CN)2PdCl2 gave the
metallomonomer 8 (Scheme 3), which provided the ðRÞ-
or ðSÞ-BINOL-based complexes 9-ðRÞ and 9-ðSÞ, re-
spectively, on reacting with enantiomerically pure
Na2BINOL. These P2Pd(BINOL) complexes were stable
in THF, acetone and toluene but were very sensitive to
chlorinated solvents. For example, 9 decomposed com-
pletely after 1 h in chlorobenzene, accompanied by
formation of palladium black, diphosphine oxide and
BINOL. Although formally diastereomeric, no spec-
troscopic differences were noted for the ðRÞ- and ðSÞ-
OCH Sty
OCH Sty
2
2
O
O
OCH Sty
2
OCH Sty
2
OCH Sty
OCH Sty
2
2
HN
HN
O
O
O
O
StyH CO
2
StyH CO
2
N
H
N
H
O
2
O
P
P
P
P
Cl
O
O
Pd
Pd
(CH CN) PdCl
3
2
Cl
Na (R)-BINOL
2
7
O
O
NH OCH Sty
NH OCH Sty
2
2
NH
NH
O
O
O
O
StyH CO
StyH CO
2
2
O
O
StyH CO
StyH CO
2
2
StyH CO
StyH CO
2
2
8
9
P2PdCl2
P2Pd(R-BINOL)
Scheme 3.
OCH Sty
2
-
O
BF
4
OCH Sty
2
OCH Sty
2
HN
O
O
StyH CO
2
N
H
Ph
Ph
Ph
Ph
O
P
P
Cl
1) AgBF
+
Pd
4
Pd
Pd
Cl
2) 7
Ph
Ph
O
NH OCH Sty
2
NH
(S)* Ar
(S)* Ar
Ar *(S)
O
P
P
O
Ar *(S)
Ph
(R)
StyH CO
2
10
O
Pd
Ph
StyH CO
P2Pd(π-allyl)+
2
(S)
StyH CO
2
Scheme 4.

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3857
BINOL containing diastereomers 9-ðRÞ and 9-(S).
Apparently the chirality of the amino acid arms does
not efficiently communicate with the BINOL stereo-
chemistry [19].
an allylic alkylation reaction (Scheme 6) [22]. The fourth
polymer was considered a control polymer and used the
P2PdCl2 metallomonomer 8, formally differing from the
others in that an imprinting ligand was not present
during polymer formation (Scheme 7).
A metallomonomer that would be an intermediate in
the catalytic cycle of the allylic alkylation reaction was
ꢀ
also desired. Synthesis of P2Pd(p-allyl)þ BF4 complex
[20] 10 was easily achieved by reacting the free diphos-
phine 7 with (1,3-diphenyl-p-allyl)palladium chloride
dimer [21], after activation with AgBF4 (Scheme 4). The
31P NMR (Fig. 1) of this compound was unusual in that
it displayed two P-resonances with a solvent-dependent
chemical shift. For example in acetone, the resonances
were exactly coincident, whereas in DMSO, THF and
chloroform, the chemical shifts were increasingly dif-
ferent and AB quartets were observed. This P-inequiv-
alence reflects diastereotopic Ps since the p-allyl
fragment has mirror symmetry, while the phosphine has
C2 symmetry (Scheme 4, inset).
2.3. Synthesis of molecularly imprinted polymers
To generate a variety of active site structures, the
available metallomonomers were used to synthesize four
different polymers. The first pair used the (R)- and (S)-
BINOL containing metallomonomers 9-ðRÞ and 9-ðSÞ,
BINOL cleavage of which (HCl) [2,10] provided
P2PdCl2 catalyst precursors that should contain (R)- and
(S)-BINOL shaped cavities in the outer sphere [2,15a],
respectively (Scheme 5). The third polymer used the
[P2Pd(p-allyl)]BF4 metallomonomer 10, and was de-
signed to provide a direct entry into the catalytic cycle of
Fig. 1. 31P NMR of 10 in: (a) acetone-d6, (b) DMSO-d6, (c) THF-d8
and (d) CDCl3. The chemical shifts have been artificially shifted for
visualization purposes. The scale corresponds to spectrum A. The
chemical shifts for B–D are ꢁ1.0 ppm of that indicated.
Scheme 5.

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3858
Scheme 6.
OCH Sty
2
O
OCH Sty
2
OCH Sty
2
HN
O
O
StyH CO
2
N
H
O
P
P
Cl
Cl
EDMA (98.5%)
Pd
P
P
Cl
Cl
No
Cavity
Pd
AIBN
60 oC, THF
O
NH OCH Sty
2
NH
O
O
P3
StyH CO
2
O
StyH CO
2
8
StyH CO
2
Scheme 7.
As described elsewhere [23], the highly crosslinked
polymers used in molecular imprinting are porous and
have high specific surface areas (approach 500–600 m2/
g). The key to obtaining these desirable properties (from
an accessibility to the interior perspective) is to carry out
the free radical polymerization in the presence of a
ꢂ50% by volume diluent or porogenic solvent [24]. The
role of this solvent is to create the pore structure during
the gelling and subsequent phase separation of the ma-
terial. The procedure generates monolithic, infinitely
cross-linked networks that are insoluble in all solvents,
but whose interiors can be penetrated by diffusion
through a combination of macro-, meso- and microp-
ores [25]. The molecularly imprinted P2PdX2 active sites
were obtained by heating (N2, 24 h, 60 °C) a mixture of
the metallomonomer (1.5 mol%), ethylene dimethacry-
late (EDMA) and AIBN (1 wt%) with the porogenic
1
solvent THF (1:1 by weight to all monomers). This
leads to brittle polymers that are crushed into manage-
able pieces, soxhlet washed with THF and dried under
vacuum. In the case of the BINOL containing metal-
lomonomers 9-ðRÞ and 9-(S), the light brown polymers
were first crushed and the BINOL then removed by
1
Typically chlorobenzene is used as a porogen, however, it was
observed that P2Pd-BINOL complexes are unstable in chlorinated
solvents. This was confirmed upon soxhlet washing with CH2Cl2 (prior
to HCl cleavage), which led to the observation of free BINOL,
quantified to be 10% and 15% of the total amount of polymers ðRÞ-P1
and ðSÞ-P1, respectively.

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3859
Table 1
The allylic alkylation of rac-1,3-diphenyl-2-propenyl acetate 11a
Entry
[Pd] catalyst
Catalyst (mol%)
Time (h)
Conversion (%)b;c
1
2
8
10
1
1
1
2
1
1
1
2
14
4.5
72
32
72
10
72
32
72
75
95
97
75
98
78
54d
25
95
85
20
3
P1-(R)
P1-(R)
P1-(S)
P2
P3
4
5
6
7
8
P1-(R) (first run)
P1-(R) (second run)
P1-(R) (third run)
9
10
a Conditions: rac-1,3-diphenyl-2-propenyl acetate (0.1 mmol); sodium dimethyl malonate (0.2 mmol); THF (2 mL); 60 °C.
b The conversion was determined by proton NMR.
c In all cases, racemic product was obtained.
d After 30 h, there was no additional conversion.
treating with 4 N HCl in dioxane, followed by soxhlet
washing with THF and drying under vacuum, to yield
the beige-yellow monolithic polymers, P1-ðRÞ and P1-
viding the expected products. The slower activity of 8
presumably reflected the fact that in situ reduction was
necessary for entry into a productive cycle, while 10
should have already been in the cycle. Complicating the
desire to call these ‘‘homogeneous’’ control experiments
was the observation that 8 and 10 appeared to poly-
merize as the reaction proceeded (increasing solution
viscosities). Given the low concentration of catalyst in
these experiments, these results were a bit surprising,
never the less, the effect was reproducible. Not surpris-
ingly, the products of both catalysts were racemic.
As each catalyst was reactive towards alkylation, the
next step was to examine the molecularly imprinted
catalyst systems (Table 1, entries 3–7) [3]. Compared to
results obtained with soluble catalysts, the reactivity of
the polymeric catalysts was diminished. The BINOL-
cavity containing catalysts P1-ðRÞ and P1-ðSÞ had nearly
identical activities (75% and 78% conversion, respec-
tively, entries 3 and 5), reflecting the fact that little
communication was observed between the chirality on
the ligand and the BINOLs. Like the solution analogs,
the products were racemic, and increasing the catalyst
loading to 2 mol% (entry 4) increased the rate.
Most surprising was the lack of reactivity of P2 (entry
6). Like the solution catalyst 10, it was expected that this
system would be more reactive since the catalyst was
already on the catalytic cycle. However, this catalyst
inexplicably stops at only 54% conversion to product. In
the case of P3, which does not have a defined space
created by the imprinting process, we expected to see a
diminished activity and this was indeed the case (entry 7,
25% conversion, 72 h). This effect has been previously
ascribed to crowding of the active site by the polymer
[26]. When large ligands are used in the imprinting
process (and subsequently removed), equally large
spaces are vacated, which ensure that the reactants can
be accommodated in the site. A series of recycle exper-
iments were also carried out to determine if catalyst
reuse was feasible (entries 8–10). Unfortunately the re-
2
ðSÞ (Scheme 5). The [P2Pd(p-allyl)]BF4 metallomono-
mer 10 yielded, after soxhlet washing with THF and
drying, a bright yellow polymer, P2 (Scheme 6) that was
ready for direct introduction into catalytic reaction
conditions. The non-imprinted reference polymer, P3,
was synthesized under identical conditions from 8 and
provided an immobilized catalyst that was composi-
tionally identical to P1 but presumably lacking a defined
cavity (Scheme 7).
2.4. Catalysis of the allylic alkylation reaction [12b]
The effects of the cavity size/shape were measured in
the allylic alkylation reaction. The alkylation of both
1,3-diphenyl propenyl acetate 11 (Eq. (1)) and 3-phenyl-
propenyl-acetate 12 (Eq. (2)) with sodium dimethyl
malonate were examined with the described Pd catalysts
under homogeneous and heterogeneous conditions.
MeO C
CO Me
2
2
OAc
O
O
NaH
THF, 60 oC
Pd - catalyst
(1 mol %)
ð1Þ
MeO
OMe
11
MeO C
CO Me
2
2
CO Me
2
O
O
NaH
THF
Pd - catalyst
(1 mol %)
OAc
CO Me
2
MeO
OMe
12
Linear
Branched
β
l
ð2Þ
To measure the inherent reactivity of the new amino-
acid containing diphosphine, compounds 8 and 10 were
tested under homogeneous conditions (1 mol% Pd,
THF, 60 °C) for the process in Eq. (1). As shown in
Table 1, 8 and 10 were both competent catalysts, pro-
2
Generally, HPLC analysis shows that 75–80% of the theoretical
amount of BINOL is removed under these conditions.

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3860
Table 2
The allylic alkylation of 3-phenyl propenyl acetate 12a
Entry
[Pd] catalyst
Catalyst (mol%)
Time (h)
Conversion (%)c
l : bd
1
2
3
4
5
6
7
8
8
10
1
1
1
1
1
1
1
1
14
10 min
>98
>99
98
98:02
80:20
71:29
70:30
75:25
73:27
75:25
74:26
P1-(R)
P1-(S)
P1-(R)b
P2
6
6
95
90
14
50
72
24
74
87
56e
P3
P(iPrdppePdCl2)
a Conditions: 3-phenylpropenyl acetate (0.1 mmol); sodium dimethyl malonate (0.2 mmol); THF (2 mL); 60 °C.
b 3-Phenylpropenyl acetate (0.2 mmol); sodium dimethyl malonate (0.1 mmol); THF (2 mL); 60 °C.
c The conversion was determined by GC.
d The linear:branched (l : b) ratio was determined by GC.
e Conversion measured again at 48 h showed no additional activity.
activity of P1-ðRÞ diminished by 2/3 after the first run
and was nearly unreactive by the third run.
were used as comonomers in the synthesis of perma-
nently porous EDMA-based polymers; the polymers in
turn served as catalysts for the alkylation of allylic
acetates.
Next, the activity of the soluble catalysts was exam-
ined for the alkylation of 3-phenyl-propenyl acetate 12
(Eq. (2), Table 2). In general, the catalysts were more
active toward 12 than 11, with p-allyl 10 being especially
so. Both 8 and 10 displayed viscosity spikes, however,
the reaction with 10 was completed (<10 min) before
These studies represent one of the few approaches to
the functionalization of transition metal active sites,
though it appears that allylic alkylation reactions don’t
respond to this feature in an obviously interpretable
manner. What is apparent from these experiments,
however, is the benefit of the sacrificial ligand approach
to catalyst synthesis. Removal of large ligands from the
metallomonomer post-polymerization ensures that the
catalyst has sufficient room in the active site to accom-
modate the reactants necessary to accomplish catalysis
[15a,26]. Contrasting this general observation is the be-
havior of the p-allyl metallomonomer 10, which we ex-
pected to provide active sites perfectly able to
accommodate the structure of an intermediate p-allyl
during catalysis. Solving this apparent dichotomy is the
subject of continuing efforts.
3
catalyst polymerization apparently begins. The lin-
ear:branched (l : b) ratio for 8 was also higher than that
observed with any of the other catalysts. The poly-
4
meric catalysts, P1-ðRÞ and P1-ðSÞ again displayed
similar reactivities (entries 3 and 4), P2 again being
slower than P1.
To gain insight into the possible role of the amino
acid unit on catalytic reactivity, a second P2PdCl2 met-
allomonomer was examined. The readily synthesized
[2,16f] (iPrdppe)PdCl2 contains polymerizable groups
directly bonded to the dppe and like P3 is meant to
represent an active site with no imprinted cavities, but
one also lacking the polar substructure of the amino
acid modified metallomonomer. In the event
P(iPrdppe)PdCl2 had an activity very similar to P3 (entry
8) except that it ceased functioning at 56% conversion
while P3 continued on to completion. We do not have a
satisfactory explanation for this observation.
4. Experimental
4.1. Compound 1
Oxalyl chloride (21.0 mL, 243 mmol) was added to a
solution of 4-iodobenzoic acid (20.0 g, 80.9 mmol) in
dichloromethane (150 mL) at room temperature, fol-
lowed by the addition of few drops of dry DMF. The
reaction mixture was then stirred overnight and di-
chloromethane and excess of oxalyl chloride were re-
moved under reduced pressure. The white solid was
dried en vacuo for 24 h. This residue was dissolved in
dichloromethane (100 mL). 2-Trimethylsilyl ethanol
(12.2 g, 101 mmol) was added and the mixture was
cooled to 0 °C. Pyridine (10.1 mL, 121 mmol) was added
slowly over 30 min and the reaction mixture was allowed
to warm to room temperature and was stirred for an
additional 4 h. Dichloromethane was evaporated and
3. Summary
This paper reports the synthesis of a polymerizable
amino acid containing diphosphine ligand along with
its coordination chemistry to the PdCl2, Pd(BINOL)
and Pd(p-allyl)þ fragments. These metallomonomers
3
Use of [dppePd(hexenyl-p-allyl)]þ generated in situ from hexenyl
acetate and Pd(DBA)2 gave 96:4 l : b selectivity and 98.5% conversion
after 5 h, see [20a].
4
As a comparison, the observed regioselectivity with a dppePd(p-
allyl) catalyst generated in situ was 89/11, see [22b].

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3861
the residue was re-dissolved in diethyl ether (150 mL).
4.4. Compound 4
The pyridinium salt was then filtered off and the ethereal
solution was washed with 2 N HCl (2 ꢃ 50 mL), water
(2 ꢃ 50 mL), saturated NaCl (50 mL), dried over mag-
nesium sulfate, filtered and concentrated. The residue
was purified by flash chromatography on silica (EtOAc/
Hexanes; 1:3) to give 25.35 g of the desired product.
A solution of N-Boc-
L
-tyrosine (5.00 g, 18.0 mmol) in
dry DMF (80 mL) at room temperature was treated with
potassium carbonate (6.23 g, 44.5 mmol; 2.50 eq), 4-vi-
nyl benzyl chloride (9.0 mL, 53 mmol; 3.0 eq) and
tetrabutyl ammonium iodide (1.20 g, 3.00 mmol). The
resulting reaction mixture was stirred at room temper-
ature overnight. Water (150 mL) was added and the
solution mixture was extracted with diethyl ether
(5 ꢃ 100 mL). The combined ether layers were washed
with 1 N HCl (100 mL), water (100 mL), brine (50 mL),
dried over magnesium sulfate, filtered and concentrated.
The product was precipitated by addition of a mixture
of ethanol and hexanes (1:1). The white solid was wa-
shed with hexanes and dried to give 8.70 g of the desired
product. Yield: 95%; 1H NMR (400 MHz, CDCl3) d
7.46 (m, 6H), 7.30 (dd, J ¼ 8:0 and 6.10 Hz, 2H), 7.02
(d, J ¼ 8:0 Hz, 2H), 6.90 (d, J ¼ 8:0 Hz, 2H), 6.82–6.74
(m, 2H), 5.82 (d, J ¼ 17:0 Hz, 2H), 5.33 (dd, J ¼ 17:0
and 6.1 Hz, 2H), 5.18 (m, 3H, 5.05 (s, 2H), 4.66 (m, 1H),
3.09 (m, 2H), 1.9 (s, 9H); 13C NMR (100 MHz, CDCl3)
d 137.8, 137.3, 136.6, 136.4, 136.3, 134.8, 130.4, 128.8,
128.2, 127.6, 126.4, 126.3, 114.9, 114.4, 114.0, 79.8, 69.7,
54.7, 37.4, 28.3.
1
Yield: 90%; H NMR (400 MHz, CDCl3) d 7.82 (dd,
J ¼ 1:95 and 6.60 Hz, 2H), 7.77 (dd, J ¼ 1:95 and 6.60
Hz, 2H), 4.44 (t, J ¼ 8:50 Hz, 2H), 1.15 (t, J ¼ 8:50 Hz,
2H), 0.63 (s, 9H); 13C NMR (100 MHz, CDCl3) d 165.7,
133.5, 133.2, 78.2, 63.8, 17.5, )1.5.
4.2. Compound 2
To a solution of 1 (5.0 g, 14 mmol) in dry THF (25
mL) isopropyl magnesium bromide 1 M in THF (15.8
mL, 1.10 eq) was added over 5 min at )25 °C. The re-
action mixture was stirred for an additional hour at this
temperature. A solution of bis(dichlorophosphino)eth-
ane (1.08 g, 3.60 mmol) in dry and degassed THF (5 mL)
was added at )25 °C and the reaction mixture was
stirred for 3 h at the same temperature, and then
quenched with degassed methanol (0.5 mL). BHÅ3THF (8
mL, 8 mmol) was added at )25 °C and the reaction
mixture was allowed to warm to room temperature and
stirred overnight. THF was removed under reduced
pressure and the product was precipitated by the addi-
tion of ethanol. The white precipitate was filtered, wa-
shed with ethanol and dried on the vacuum line for 24 h
4.5. Compound 5
Compound 4 (20.0 g, 38.9 mmol) was added in portions
over 30 min at room temperature to 4 N HClðaqÞ in 1,4-
dioxane (concentrated HCl/dioxane; 1:2; 444 mL). The
reaction mixture was then stirred for 4 h. The precipitated
white solid was filtered, washed with water (2 ꢃ 50 mL),
diethyl ether (2 ꢃ 50 mL) and dried on the vacuum line for
24 h to give 15.5 g of the desired product. Yield: 88%; 1H
NMR (400 MHz, DMSO-d6) d 8.55 (s, 3H, NH3), 7.4 (m,
6H), 7.26 (d, J ¼ 8:0 Hz, 2H), 7.1 (d, J ¼ 8:5 Hz, 2d), 6.92
(d, J ¼ 8:5 Hz, 2H), 6.75 (d, J ¼ 17:6 Hz, 2H), 5.86 (d,
J ¼ 17:6 Hz, 2H), 5.29 (m, 2H), 5.15 (s, 2H), 5.07 (s, 2H),
4.3 (s, 1H), 3.24 (m, 2H). 13C NMR (100 MHz, DMSO-
d6) d 169.2, 158.0, 137.6, 137.2, 137.1, 136.7, 136.6, 134.8,
130.9, 129.1, 128.2, 127.1, 126.6, 126.5, 115.3, 115.2,
114.8, 69.4, 67.1, 53.8, 35.5.
1
to give 3.30 g of the desired product. Yield: 92%; H
NMR (400 MHz, CDCl3) d 8.12 (d, J ¼ 8:13 Hz, 2H),
7.73 (m, 8H), 4.46 (t, J ¼ 8:34 Hz, 8H), 2.43 (d, J ¼ 2:40
Hz, 4H), 1.15 (t, J ¼ 8:34 Hz, 8H), 0.11 (s, 36H); 13C
NMR (100 MHz, CDCl3) d 165.5, 133.9, 132.6, 132.1 (t,
J ¼ 4:9 Hz), 129.9 (t, J ¼ 4:9 Hz), 77.1, 19.2 (t, J ¼ 18:4
Hz), 17.4, )1.3, )1.5, )1.8; 31P NMR (161.86 MHz,
CDCl3) d 21.0 (s, 2P).
4.3. Compound 3
Tetrabutyl ammonium fluoride (1 M in THF) (16.0
mL, 16.0 mmol) was added at 0 °C to a solution of 2
(2.10 g, 2.09 mmol) in THF (20 mL). The cooling bath
was then removed and the reaction mixture was stirred
overnight at room temperature. About 75% of the THF
was removed under reduced pressure and a solution of 1
N HCl (50 mL) was added at 0 °C. The white precipitate
was collected by filtration, washed with water followed
by diethyl ether and dried overnight on the vacuum line
to give 1.20 g of the desired product. This material was
used in the next step without any further purification.
4.6. Compound 6
Oxalyl chloride (1.62 mL, 18.6 mmol; 8 eq) was added
slowly under argon to a suspension of 3 (1.40 g, 2.32
mmol) in dichloromethane (35 mL) at 0 °C. The reaction
mixture was stirred for 30 min and then a few drops of
dry DMF were added, and stirring was pursued for an
additional 2 h at 0 °C. The evolution of the reaction was
monitored by 31P NMR (d(31P) of the tetraacid chloride
product is 22.5 ppm). The reaction mixture was con-
centrated under reduced pressure and the residue was
1
Yield: 95%; H NMR (400 MHz, CD3OD) d 8.10 (d,
J ¼ 7:8 Hz, 8H), 7.76 (m, 8H), 2.48 (s, 4H); 31P NMR
(161.86 MHz, CD3OD) d 17.6 (s, 2P).

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3862
dried on the vacuum line for 24 h. This material was
added in dichloromethane (60 mL) to a solution of 5
(5.22 g, 11.6 mmol) and triethylamine (3.6 mL, 23 mmol)
in dichloromethane (60 mL) at 0 °C and under argon.
The reaction mixture was stirred at the same temperature
for 5 h. About 3/4 of CH2Cl2 was removed and methanol
(100 mL) was added. The beige solid precipitate was
filtered, washed with water (50 mL), washed with
methanol (2 ꢃ 50 mL) and dried on the vacuum line
overnight to give 3.75 g of the desired product. Yield:
temperature and under argon a solution of 8 (0.39 g,
0.16 mmol) in THF (5 mL). The reaction mixture was
then stirred overnight. THF was removed under reduced
pressure and the residue was washed with ether (25 mL),
methanol (50 mL), and dried on the vacuum line for 24
h to give 0.42 g of the desired product as a brown solid.
Yield: 99%; H NMR (400 MHz, THF-d8) d 7.95–6.25
(m, 94H), 5.75 (m, 8H), 5.38–4.75 (m, 28H), 3.2 (m, 6H);
31P NMR (161.86 MHz, THF-d8) d ¼ 52:9 (s, 2P).
1
1
75%; H NMR (400 MHz, CDCl3) d 7.76 (m, 8H), 7.64
(m, 8H), 7.36 (m, 32H), 6.94 (m, 8H), 6.76 (m, 20H), 5.78
4.10. Compound 10
(m, 2H), 5.33–4.98 (m, 28), 3.2 (m, 8H), 2.45 (m, 4H); 31
P
To a slurry of 7 (0.91g, 0.42 mmol) in dried, distilled
acetone (20 mL) was added a slurry of AgBF4 (84 mg,
0.43 mmol) in acetone (20 mL) at room temperature
under N2. The reaction was stirred for 1 h. The phos-
phine slurry went from yellow to brownish-green upon
addition of AgBF4. The solution was filtered over Celite,
washed with EtOAc (50 mL) and purified by column
chromatography (9:1 EtOAc:hexanes) to yield a yellow-
green solid. Yield: 80%; 1H NMR (400 MHz, THF-d8) d
8.15–6.75 (m, 76H), 6.75–6.60 (m, 8H), 5.80–5.65 (m,
8H), 5.25–4.85 (m, 24 H), 4.10–3.95 (m, 1H), 3.52–3.50
(d, J ¼ 8 Hz, 1H), 3.35–3.15 (m, 8H), 2.75–2.40 (m, 8H),
1.80 (d, J ¼ 8 Hz, 1H) 31P NMR (161.86 MHz, THF-d8)
d 46.5 (dd, J ¼ 63:2 Hz, 2P); 31P NMR (161.86 MHz,
acetone-d6) d 49.2 (s); 31P NMR (161.86 MHz, DMSO-
d6) d 49.1 (d, J ¼ 5:3 Hz); 31P NMR (161.86 MHz,
CDCl3) d 49.2 (dd, J ¼ 97:9 Hz).
NMR (161.86 MHz, CDCl3) d 20.88 (s, 2P).
4.7. Compound 7
The solution of 6 (2.20 g, 1.01 mmol) in degassed THF
(10 mL) and degassed ethylamine (15 mL) was heated at
55 °C under argon for 3 h. Methanol (10 mL) was added
to complete the precipitation of the formed product. The
beige solid was filtered, washed with methanol (25 mL)
and dried on the vacuum line for 16 h to give 2.05 g of
pure free phosphine-based monomer. Yield 94%; 1H
NMR (400 MHz, CDCl3) d 7.65 (m, 8H), 7.42 (m, 42H),
6.98 (m, 8H), 6.74 (m, 18H), 5.78 (m, 8H), 5.31–5.04 (m,
18H), 4.96 (s, 8H), 3.18 (m, 8H), 2.11 (m, 4H), 1.61 (m,
2H); 13C NMR (100 MHz, CDCl3) d 171.4, 166.5, 166.3,
157.9, 137.9, 137.3, 136.4, 136.2, 134.5, 134.3, 132.9,
132.6, 130.3, 128.9, 128.8, 128.0, 127.9, 127.5, 127.0,
126.3, 114.9, 114.5, 114.0, 77.1, 69.7, 67.0, 53.8, 36.9; 31
P
4.11. Compund P1
NMR (161.86 MHz, CDCl3) d )11.7 (s, 2P).
Metallomonomer 9 (231 mg, 0.087 mmol, 1.5 mol%),
AIBN (10.7 mg, 1 wt%), EDMA (1.18 g, 5.72 mmol,
98.5 mol%) and dried THF (1400 mg) were combined in
a 20-mL scintillation vial under N2 and sealed with a
Teflon-lined cap. The vial was heated to 60 °C for 24 h,
yielding a hard, transparent, light-brown polymer. After
Soxhlet washing with THF for 8 h and drying on the
vacuum line at 50 °C overnight, 1375 mg of polymer was
obtained (63.3 lmol of Pd/g of polymer). A suspension
of this polymer (687.5 mg, 31.6 lmol of Pd) in THF (25
mL) was treated under argon with 4 N HCl in dioxane
(2 mL). The mixture was stirred at room temperature for
8 h. The reaction was then filtered and the polymer was
Soxhlet washed with THF for 16 h. Quantitative anal-
ysis of the extract showed that 83% of the BINOL was
recovered from the polymer (generating 83% of PdCl2
sites from the total Pd sites). The polymer was dried on
the vacuum line at 60 °C for 24 h, 675 mg (26.2 lmol of
Pd/g polymer) of beige-yellow P1 was obtained.
4.8. Compound 8
To a suspension of (CH3CN)2PdCl2 (0.12 g, 0.46
mmol) in dry THF (10 mL) at room temperature was
added under argon 7 (1.00 g, 0.462 mmol) in one portion.
The reaction mixture was stirred for 2 h. The solution
was filtered on Celite and concentrated under reduced
pressure to give 1.08 g of the desired product. Yield 98%;
1H NMR (400 MHz, CDCl3) d 7.75 (m, 16H), 7.44 (m,
36H), 7.02 (m, 8H), 6.74 (m, 16H), 5.74 (d, J ¼ 17:5 Hz,
8H), 5.28–4.83 (m, 28H), 3.41 (m, 8H), 2.5 (m, 4H); 13
C
NMR (100 MHz, CDCl3) d 171.9, 171.4, 165.8, 165.6,
157.9, 157.7, 137.8, 137.2, 136.4, 136.2, 136.1, 134.5,
134.0, 133.5, 130.7, 130.2, 128.7, 128.1, 127.9, 127.6,
126.7, 126.4, 126.3, 115.0, 114.8, 114.5, 114.4, 114.0,
113.9, 77.2, 69.6, 67.8, 67.2, 67.0, 54.2, 54.0, 36.9; 31P
NMR (161.86 MHz, CDCl3) d 64.3 (s, 2P).
4.9. Compound 9
4.12. Compound P2
To a solution of sodium binolate, prepared from the
reaction of sodium tert-butoxide (0.35 mmol) and BI-
NOL (0.17 mmol) in THF (10 mL), was added at room
Compound P2 was prepared under the reaction
conditions described for P1: (0.203 mg 10, 0.0798 mmol,

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3863
[10] For our first report on non-functionalized dendritic diphosphine
1.5 mol%), AIBN (0.021 mg, 1 wt%), EDMA (1.05 g,
5.30 mmol) and dried THF (1.06 g) were combined in a
20-mL scintillation vial under N2 and sealed with a
Teflon-lined cap. The vial was heated to 60 °C for 24 h,
yielding a green-yellow polymer. The resulting polymer
was Soxhlet extracted with THF for 8 h and dried on the
vacuum line at 60 °C for 16 h; 1.27 g of P2 (62.6 lmol of
Pd/g of polymer) was obtained.
ꢀ
imprinting, see: J.J. Becker, M.R. Gagne, Organometallics 22
(2003) 4984.
[11] For an example of a dendritic Pt complex, see: B.S. Balaji, Y.
Obora, D. Ohara, S. Koide, Y. Tsuji, Organometallics 20 (2001)
5342.
[12] (a) For other reports on P2PdX2 catalyzed allylic alkylation, see:
T. Hayashi, A. Yamamoto, T. Hagihara, J. Org. Chem. 51 (1986)
723, and references cited within;
(b) B. Trost, C. Lee, Asymmetric allylic alkylation reactions, in: I.
Ojima (Ed.), Catalytic Asymmetric Synthesis, Wiley-VCH, New
York, 2000, pp. 593–649.
4.13. Compound P3
[13] For an example of Pd catalyzed allylic alkylation with dendridic
phosphines: G.E. Oosterom, R.J. van Haaren, J.N.H. Reek,
P.C.J. Kamer, P.W.M. van Leeuwen, Chem. Commun. (1999)
1119.
Compound P3 was prepared under the reaction
conditions described for P1: (98.3 mg 8, 0.0454 mmol),
AIBN (7.2 mg, 1 wt%), EDMA (615 mg, 2.98 mmol)
and dried THF (714 mg) were combined in a 20-mL
scintillation vial under N2 and sealed with a Teflon-lined
cap. The vial was heated to 60 °C for 24 h, yielding a
beige-yellow polymer. The resulting polymer was
Soxhlet extracted with THF for 8 h and dried on the
vacuum line at 60 °C for 16 h; 700 mg of P3 (64.8 lmol
of Pd/g of polymer) was obtained.
[14] (a) Catalysis within imprinted media has recently been reviewed,
see: G. Wulff, Chem. Rev. 102 (2002) 1;
(b) C. Alexander, L. Davidson, W. Hayes, Tetrahedron 59 (2003)
2025;
(c) M. Tada, Y. Iwasawa, J. Mol. Catal. A: Chem. 3953 (2003)
1.
[15] (a)For other examples of transition metal catalysis within
imprinted polymers, see: J.H. Koh, A.O. Larsen, P.S. White,
ꢀ
M.R. Gagne, Organometallics 21 (2002) 7;
(b) K. Polburn, K. Severin, Chem. Commun. (1999) 2481;
(c) K. Polburn, K. Severin, Chem. Eur. J. 6 (2000) 4604;
(d) A.N. Cammidge, N.J. Baines, R.K. Bellingham, Chem.
Commun. (2001) 2588.
Acknowledgements
[16] (a) For examples of metal-mediated catalysis in porous polymers
that do not incorporate imprinting per se, see: Ru catalyzed
oxidation O. Nestler, K. Severin, Org. Lett. 3 (2001) 3907, for
hydroformylation;
We thank the National Science Foundation (CHE-
0315203 and CHE-0075717) for generous support of this
research. M.R.G. is a Camille-Dreyfus Teacher Scholar.
(b) K. Nozaki, Y. Itoi, F. Shibahara, E. Shirakawa, T. Ohta, H.
Takaya, T. Hiyama, J. Am. Chem. Soc. 120 (1998) 4051;
(c) K. Nozaki, S. Fumitoshi, Y. Itoi, E. Shirakawa, T. Ohta, H.
Takaya, T. Hiyama, Bull. Chem. Soc. Jpn. 72 (1999) 1911;
(d)for transfer hydrogenation, see: S. Wendicke, E. Burri, R.
Scopelliti, K. Severin, Organometallics 22 (2003) 1894, for olefin
hydrogenation;
References
[1] (a) For reviews on functionalized dendrimers for use as biological
mimics, see: D.K. Smith, F. Diederich, Chem. Eur. J. 4 (1998)
1353;
ꢀ
(e) S.M. Vinson, M.R. Gagne, Chem. Commun. (2001) 1130;
(f) R.A. Taylor, B.P. Santora, M.R. Gagne, Org. Lett. 2 (2000)
ꢀ
(b) S. Hecht, J.M.J. Frechet, Angew. Chem., Int. Ed. 40 (2001) 74;
(c) F. Zeng, S.C. Zimmerman, Chem. Rev. 97 (1997) 1681.
ꢀ
1781, for Ti catalyzed Diels-Alder reactions;
ꢀ
(g) B.P. Santora, P.S. White, M.R. Gagne, Organometallics 18
ꢀ
[2] N. Brunkan, M.R. Gagne, J. Am. Chem. Soc. 122 (2000) 6217.
[3] For an alternative approach using an organized crown-ether in the
(1999) 2557;
(h) B.P. Santora, A.O. Larsen, M.R. Gagne, Organometallics 17
ꢀ
active site, see: F. Viton, P.S. White, M.R. Gagne, Chem.
Commun. (2003) 3040.
ꢀ
(1998) 3138.
[17] (a) For an overview of dendrimer structure illustrating both
[4] For a recent review on the synthesis of synthetic hosts by
imprinting, see S.C. Zimmerman, N.G. Lemcoff, Chem. Commun.
(2004) 5.
ꢀ
synthesis and application, see: S.M. Grayson, J.M.J. Frechet,
Chem. Rev. 101 (2001) 3819;
(b) A.W. Bosman, H.M. Janssen, E.W. Meijer, Chem. Rev. 99
(1999) 1665.
[5] For an example of functionalized dendridic phosphine ligands in
asymmetric catalysis, see: H. Brunner, J. Organometallic Chem.
500 (1995) 39.
[18] P. Knochel, W. Dohle, N. Gommermann, F.F. Kneisel, F. Kopp,
T. Korn, I. Sapountzis, V.A. Vu, Angew. Chem., Int. Ed. 42
(2003) 4302.
[6] (a) For reviews illustrating the use of dendrimers in catalysis, see:
L.J. Twyman, A.S.H. King, I.K. Martin, Chem. Soc. Rev. 31
(2002) 69;
[19] These observations are generally consistent with the work of
Brunner who showed that ‘expanded’-chirality had little effect on
(b) D. Astruc, F. Chardac, Chem. Rev. 101 (2001) 2991.
review on chiral dendrimers for use in asymmetric
€
reaction stereochemistry. See for example: H. Brunner, J. Furst,
Tetrahedron 50 (1994) 4303.
[7] For
a
catalysis, see: B. Romagnoli, W. Hayes, J. Mater. Chem. 12
(2002) 767.
[20] (a) There are several reports examining alkylation with
similar P2Pd(p-allyl)þ complexes, see: M. Kranenburg, P.C.J.
Kamer, P.W.N.M. van Leeuwen, Eur. J. Inorg. Chem. (1998)
25;
[8] Although the described complexes don’t have any dendritic
character, they are related to a class of amino-acid containing
dendrimers, see: Y. Kim, F. Zeng, S.C. Zimmerman, Chem. Eur.
J. 5 (1999) 2133.
(b) or for a general review, see: J.A. Davies, in: E.W. Abel,
F.G.A. Stone, G. Wilkinson (Eds.), Comprehensive Organo-
metallic Chemistry II, vol. 9, Pergamon, Oxford, 1995
(Chapter 6).
[9] For other examples of amino acid modified phosphine ligands,
see: A. Agarkov, E.W. Uffman, S.R. Gilbertson, Org. Lett. 5
(2003) 2091, and references therein.

€
H. Aıt-Haddou et al. / Inorganica Chimica Acta 357 (2004) 3854–3864
3864
[21] W.T. Dent, R. Long, A.J. Wilkinson, J. Chem. Soc. (1964) 1585.
[22] (a) For references addressing the mechanism of asymmetric
alkylation, see: P.B. Mackenzie, J. Whelan, B. Bosnich, J. Am.
Chem. Soc. 107 (1985) 2046;
(b) B. Sellegren, K.J. Shea, J. Chromatogr. 635 (1993) 31.
[24] D.C. Sherrington, Chem. Commun. (1998) 2275.
[25] We have shown that even fluorinated solvents will penetrate the
network and concentrate analytes/reactants into the matrix, see:
ꢀ
S.M. Leeder, M.R. Gagne, J. Am. Chem. Soc. 125 (2003) 9048,
and Ref. [16e].
(b) T. Hayashi, M. Kawatsura, Y. Uozumi, J. Am. Chem. Soc.
120 (1998) 1681.
ꢀ
[23] (a) B.P. Santora, M.R. Gagne, K.G. Maloy, N.S. Radu, Macro-
molecules 34 (2001) 658;
[26] K. Severin, Curr. Opin. Chem. Biol 4 (2000) 710, For additional
discussions, see [15a].