Design and synthesis of tris[bis(benzylammonium)aryl]phosphines with bulky meta-substituents

Article abstract of DOI:10.1016/S0040-4020(03)00467-8

A novel 3,5,3′,5′,3″,5″-hexakis(dimethylamino)methyl substituted triphenylphosphine analogue has been prepared. The six amine functionalities of the corresponding phosphine oxide and sulfide were alkylated quantitatively with methyl iodide and benzyl bromide, as well as with G1 and G2 Fréchet dendrons to afford the respective hexa-ammonium triarylphosphine oxides and sulfides. The phosphine sulfide derivatives were deprotected with P(n-Bu)₃ to afford a series of hexa-ammonium triarylphosphines that range from small molecules to first and second generation dendrimers (MW up to 5451.44). Upon formation of the hexa-ammonium salt a clear shift in ³¹P NMR is observed, indicative for opening of the C-P-C angle of the triaryl phosphine. Calculations on the hexacationic phosphines show an increased barrier of rotation around the P-C(aryl) bond with increasing size of the N-substituents. The calculated structure of the G2-dendron alkylated hexa-ammonium triarylphosphine shows that this dendritic phosphine has a disc-like rather than a cone-like structure, with an estimated cone angle of approximately 180°.

Full text of DOI:10.1016/S0040-4020(03)00467-8

Tetrahedron 59 (2003) 3989–3997
Design and synthesis of
tris[bis(benzylammonium)aryl]phosphines with bulky
meta-substituents
*
Robert Kreiter, Robertus J. M. Klein Gebbink and Gerard van Koten
Department of Metal-Mediated Synthesis, Debye Institute, Utrecht University, Padualaan 8, NL-3784 CH Utrecht, The Netherlands
Received 18 November 2002; revised 29 January 2003; accepted 4 February 2003
Abstract—A novel 3,5,30,50,300,500-hexakis(dimethylamino)methyl substituted triphenylphosphine analogue has been prepared. The six
amine functionalities of the corresponding phosphine oxide and sulfide were alkylated quantitatively with methyl iodide and benzyl bromide,
´
as well as with G1 and G2 Frechet dendrons to afford the respective hexa-ammonium triarylphosphine oxides and sulfides. The phosphine
sulfide derivatives were deprotected with P(n-Bu)3 to afford a series of hexa-ammonium triarylphosphines that range from small molecules to
first and second generation dendrimers (MW up to 5451.44). Upon formation of the hexa-ammonium salt a clear shift in 31P NMR is
observed, indicative for opening of the C–P–C angle of the triaryl phosphine. Calculations on the hexacationic phosphines show an
increased barrier of rotation around the P–C(aryl) bond with increasing size of the N-substituents. The calculated structure of the G2-dendron
alkylated hexa-ammonium triarylphosphine shows that this dendritic phosphine has a disc-like rather than a cone-like structure, with an
estimated cone angle of approximately 1808 q 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
type complexes, even in the presence of excess ligand.
Attachment of carbosilane dendrons at the para-position of
all three aryl rings of triphenylphosphine was reported by
the group of Van Leeuwen.10
Dendrimers are widely applied in coordination chemistry
and catalysis.1,2 In most of these cases, ligands are attached
to an (inert) dendrimer backbone, at the periphery, at the
core, or at the branches of the dendrimer. Homogeneous
catalysts based on dendrimers functionalized with phos-
phines at the periphery were reported by many authors.3
Illustrative examples were developed in the groups of Van
Leeuwen,4 Reetz,5 Togni,6 and Majoral,7 as well as in our
group.8 Phosphine ligands of which the phosphorus center is
part of the dendrimer backbone, were reported by the groups
of Dubois and Kakkar, respectively.9 Furthermore, dendron-
enlarged diphosphines derived from DPPF,10 DPPE,11 and
BINAP12 were studied. Surprisingly, dendron attachment to
monodentate phosphines has attracted much less attention.
Catalano reported on (mono-dendron)diphenylphosphines
Recently, we reported on multicationic dendrimers based on
tetrahedral tetrakis(mono- and bis[(dimethylamino)-
methyl]phenyl)silane core molecules. To these core
molecules, dendrons were attached by a quaternization
´
reaction with benzylic bromides derived from Frechet-type
dendritic wedges.15 This yielded a new class of tetra- and
PPh2(G1), and PPh2(G2), where G1 and G2 are first and
13
second generation Frechet-type dendrons, respectively.
´
More recently, Tsuji et al. described the synthesis of
functionalized triphenylphosphines having a dendron on
two of the three phenyl rings, as well as DPPE function-
alized with a dendron on each of the four aryl rings.14
Platinum(0) complexes of the mono-dentate ligands were
studied and revealed a preference for the formation of ML3
Keywords: dendrimers; ammonium salt; phosphine ligand; Dendriphos;
cone angle.
*
Corresponding author. Tel.: þ31-30-2531813; fax: þ31-30-2523615;
Figure 1. Octacationic dendrimer based on a tetrakis(3,5-bis[(dimethyl-
amino)methyl]phenyl)silane core molecule.
e-mail: g.vankoten@chem.uu.nl
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00467-8

3990
R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
octa-cationic dendrimers (for an example see Figure 1),
which were successfully applied as phase-transfer catalysts
in SN2 type reactions. Furthermore, it was shown that these
oligo-cationic dendrimers can serve as host molecules for
the non-covalent attachment of a predefined number of
anions, such as methyl orange, pyrene-1-sulfate, and PF26 .
Moreover, these dendrimers were applied as a soluble
supporting material for anionic organometallic com-
plexes.16 For example, an assembly of eight anionic
organometallic complexes and a single octa-cationic
dendritic support could be constructed via non-covalent
interactions, purified by dialysis and subsequently used as a
Lewis acid catalyst in an aldol condensation.
phosphorus atom and to enable selective N-alkylation
chemistry, crude 2 was first either oxidized with H2O2 in
THF, to afford phosphine oxide 3 (OvP(NCN)3), or reacted
with sulfur in CS2 to afford phosphine sulfide 4 (SvP(NCN)3)
(Scheme 1). These protected phosphines were recrystallized
fromhexane(for3), orwashed withhexane(in thecaseof4) to
afford off-white solids in reasonable overall yields.
OvP(NCN)3 (3) and SvP(NCN)3 (4) were alkylated with
MeI in CH2Cl2 at room temperature (Scheme 2). The
respective products precipitated from the reaction mixtures
as light yellow powders, which could be filtered off.
Precipitation from MeOH with Et2O afforded pure hexa-
ammonium triaryl phosphine oxide 5a ([OvP(NCN)3·
Me6]I6) and sulfide 6a ([SvP(NCN)3·Me6]I6). These MeI
salts are highly soluble in H2O and moderately soluble in
MeOH, but insoluble in less polar organic solvents, such as
CH2Cl2 and hexane. Likewise, 3 and 4 were converted with
benzyl bromide in CH2Cl2 at room temperature into the
corresponding ammonium salts 5b ([OvP(NCN)3·Bz6]Br6)
and 6b ([SvP(NCN)3·Bz6]Br6) (Scheme 2). Both salts have
solubilities similar to 5a and 6a, but are less soluble in
H2O. These alkylation reactions occur readily at room
temperature and are quantitative. In spite of this, the
products are obtained in lower although still reasonable
yields, which is mainly caused by the required purification
procedure.
In the present study our attention is directed towards the
influence of the stereochemistry of the core molecule on the
binding ability of this kind of oligo-cationic, dendritic
supports. To this end we selected the phosphorus atom of
triphenylphosphine which, like the Si-center in the
dendrimers described above, has a stable tetrahedral
configuration, but has a lone pair instead of a fourth
substituent. Here we report the synthesis of a novel tris(3,5-
bis[(dimethylamino)methyl]aryl)phosphine, providing an
interesting combination of one soft and six hard Lewis
bases. Furthermore, we present the N-alkylation of this
triarylphosphine core molecule to afford several hexa-
ammonium triarylphosphine derivatives. These hexa-
cationic phosphines are potentially useful as supporting
material for a variety of anions, whilst retaining their
phosphine donor ability for coordination to metal sites or for
reaction with electrophiles.
To obtain the hexacationic, dendritic phosphine oxides 5c
([OvP(NCN)3·(G1)6]Br6) and 5d ([OvP(NCN)3·(G2)6]-
Br6) and sulfides 6c ([SvP(NCN)3·(G1)6]Br6) and 6d
([SvP(NCN)3·(G2)6]Br6), 3 and 4 were reacted with the
corresponding benzylic bromides of the first (G1-Br) and
18
second (G2-Br) generation Frechet-type dendritic wedges
´
2. Results and discussion
2.1. Synthesis
in CH2Cl2 at room temperature (Scheme 2). The products
were purified by repeated precipitation with Et2O from
CH2Cl2 and isolated as white solids. Like the reactions with
the smaller alkyl halides, these reactions were quantitative,
and occurred readily at room temperature. The isolated
yields were lowered by the precipitation steps applied in the
purification of the products. These dendritic salts are highly
soluble in chlorinated organic solvents, but are insoluble in
Et2O and alkanes. Furthermore, the G1 derivatives (5c, 6c)
are slightly soluble in MeOH, whereas the G2 derivatives
(5d, 6d) are essentially insoluble in this solvent. These
reactions demonstrate that starting from a single amine
substituted core molecule and a series of alkyl and benzyl
halides, hexacationic compounds can be prepared, varying
from small molecules (MW¼1456.48) to dendrimers of
different generations (MW up to 5451.44). Depending on
the N-substituent these compounds were found to be soluble
in a range of solvents of different polarity. To investigate
whether the dendritic salts are suitable hosts for larger
anions, methyl orange (MO) was exchanged for the bromide
counterions of [OvP(NCN)3·(G1)6]Br6 in a preliminary
experiment. This experiment showed quantitative incorpor-
ation of MO, and confirmed the ability of these dendrimers
to serve as host molecules.
A common route in the preparation of triarylphosphines
involves the coupling of an aryl-Grignard or -lithium
reagent to a phosphorus trihalide. We chose to react
1-bromo-3,5-bis[(dimethylamino)methyl]benzene (Br-NCN,
1)17 with 2 equiv. of t-BuLi to give the 5-lithio-derivative.
Subsequent addition of 1/3 equiv. of phosphorus tribromide
resulted in the formation of tris(3,5-bis[(dimethylamino)-
methyl]phenyl)phosphine (P(NCN)3, 2) (Scheme 1). After
extraction with CH2Cl2 and evaporation of all volatiles, crude
2 was obtained as a yellow oil. In order to protect the
Deprotection of the phosphorus atom in 5a–d and 6a–d
was attempted via several different approaches. Standard
methods in phosphine oxide reduction, such as refluxing
[OvP(NCN)3·(G1)6]Br6 (5c) in a mixture of HSiCl3/NEt3/
Scheme 1. Synthesis of tris(3,5-bis[(dimethylamino)methyl]phenyl)-
phosphine oxide (3) and sulfide (4).

R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
3991
Scheme 2. Alkylation of 3 and 4 to afford a range of hexa-ammonium salts.
C6H6 did not result in reduction. Other methods such as
treatment of one of the oxides 5a–d with bis(catecholato-
O,O0)diboron and Wilkinson’s catalyst in CH2Cl2, DIBAL-
H in CH2Cl2, or Raney Ni in MeOH also did not yield the
free phosphines. Each time, the hexacationic phosphine
oxides were recovered from the reaction mixture, which
points to their high stability under severe conditions. For the
phosphine sulfides 6a–d, methods like reaction with sodium
naphthalenide, treatment with Br2 in Et2O, or heating in
pure P(n-Bu)3 did not result in reduction to the free
phosphine. Obviously, in the latter case this was due to
insolubility of the starting material in P(n-Bu)3. The
hexacationic, dendritic phosphine sulfide [SvP(NCN)3·
(G1)6]Br6 (6c) could, however, be reduced to the free
phosphine following a procedure reported by Gilbertson
et al., which involves sulfur alkylation with MeOTf and
subsequent deprotection with HMPT.19 Whereas this
procedure resulted in reduction of the phosphine sulfide to
the free phosphine, exchange of the bromide counterions for
triflate ions also occurred, affording [P(NCN)3·(G1)6](OTf)6
(7). Because of the toxicity of both reagents (MeOTf and
HMPT) we pursued, in parallel, another method involving
less harmful reagents, and returned to reduction of the
phosphine sulfides with P(n-Bu)3. Finally, it appeared that
reduction of phosphine sulfides 6a–d proceeded smoothly
in a refluxing, homogeneous 2:1 mixture of MeOH and P(n-
Bu)3 (Scheme 3). After removal of the volatiles the obtained
crude phosphines 8a–d were purified by precipitation with
Et2O from MeOH, affording light yellow (8a) to white
solids (8b–d). We call this novel class of dendritic
phosphines Dendriphos by analogy with other common
names for phosphines.
peak is usually the [M23X2]3þ or [M24X2]4þ species (for
a representative spectrum of 6d see Figure 2). For the
dendritic salts (5c,d, 6c,d) additional ion peaks are observed
at lower m/z values than the molecular ion peaks. This could
point to (unexpected) fragmentation or degradation of the
´
Frechet wedges. To investigate whether these additional
ions result from a fragmentation reaction, the cone voltage
was lowered to very low values (10 V). As no change in the
mass spectrum was observed, it was concluded that the
degradation does not occur due to the cone voltage, but
results from the applied capillary voltage in the nano-
electrospray mass spectrometer.
Mass analysis (nano-ESI) of the free phosphines 8a–d
afforded comparable results. The mass spectrum of 8c was
obtained as the phosphine oxide, probably because of
oxidation during sample preparation. For 8a,b, the ions
[M22X2]2þ, [M23X2]3þ, and [M24X2]4þ (X¼I (8a), Br
(8b)) were observed. For 8c,d, the [M23Br2]3þ and
[M24Br2]4þ species were identified.
Furthermore, these triaryl phosphine derivatives were
1
characterized by H, 13C, and 31P NMR spectrometry. To
compare the 31P shifts of the series of phosphine sulfides (4
and hexacationic 6a–c) and of the series of free phosphines
(2, and hexacationic 8a–c) the NMR spectra were all
measured in CD3OD (Table 1). Regrettably, the dendritic
salts 6d and 8d are too insoluble to be measured in this
solvent. Based on these shifts, a clear effect of charge
introduction is observed. Upon quaternization of the amines
in 4 an upfield shift of 22.7 ppm is observed for the
phosphine sulfides (6a–c). The fact that a similar upfield
shift is found for the HCl salt of 4 (d¼40.82 ppm), and a
downfield shift of þ3.4 ppm is found for the free phosphines
(8a–c), indicates that this could be caused by Coulombic
repulsion between the ammonium groups. Most probably,
such repulsion would lead to opening of the C–P–C angles.
Opening of C–P–C angles in free triarylphosphines
resulting from steric effects is known to lead to a downfield
shift in 31P NMR, due to a rehybridization at phosphorus.
This results in an overall increase of electron density at the
phosphorus atom.20 Because of the polarized P–S bond in
phosphine sulfides a similar rehybridization would result in
a decrease in P–S bond order, resulting in an overall
decrease of electron density at phosphorus. Our current
observations appear to be in agreement with this explanation.
2.2. Characterization
Mass spectrometry (nano-ESI) was performed to confirm
the molecular mass of the novel hexa-ammonium phosphine
oxides and sulfides. For all the hexacationic species,
ionization via loss of counterions was observed. In the
mass spectrum of 5a peaks for [M2I2]þ, [M22I2]2þ
[M23I2]3þ, [M24I2]4þ, and in the spectrum of 6a peaks
,
for [M2I2]þ, [M22I2]2þ, [M23I2]3þ, [M24I2]4þ
,
[M25I]5þ, and [M26I2]6þ were observed. For 5b, 6b,
5c, and 6c [M22Br2]2þ, [M23Br2]3þ, [M24Br2]4þ were
observed, while for 5d and 6d [M23Br2]3þ and
[M24Br2]4þ could be identified. In these spectra the base

3992
R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
Scheme 3. Reduction of the phosphine sulfides 6a–d to afford the hexacationic triphenylphosphine analogues 8a–d, and structure of Dendriphos 8d.
Figure 2. Representative mass spectrum of 6d showing characteristic peaks at m/z¼1747.7 [M23Br2]3þ, and 1290.0 [M24Br2]4þ
.

R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
3993
Table 1. 31P NMR data for the series of phosphine sulfides 4 and 6a–d, and free phosphines 2 and 8a–d in CD3OD
[SvP(NCN)3·R6]X6, RX
31P NMR d (ppm)
[P(NCN)3·R6]X6, RX
31P NMR d (ppm)
– (4)
43.40
40.70
40.68
40.58
Insol
– (2)
25.30
21.90
21.82
21.68
Insol.
MeI (6a)
BzBr (6b)
G1Br (6c)
G2Br (6d)
MeI (8a)
BzBr (8b)
G1Br (8c)
G2Br (8d)
Enlargement of the N-substituents in the two series of
ammonium salts seems to enlarge these effects, but only to a
minor extent.
the core aryl-protons occurs, and two separate signals can be
observed. A detailed study on the NMR behavior of these
phosphines at variable temperatures is currently being
carried out and will reported elsewhere. In parallel,
calculations on the rotation of aryl fragments around the
phosphine–aryl bond were performed for the series of
phosphines 2 and 8a–d. These calculations show that the
rotation barriers increase with increasing size of the
N-substituent (Fig. 3). Whereas the parent triarylphosphine,
2 has a rotation barrier of 16 kJ/mol (compare this to the
calculated barrier for triphenylphosphine: 1.5 kJ/mol),
Dendriphos 8c,d have rotation barriers increasing to 82
(G1) and 220 kJ/mol (G2). These gas-phase MM2 calcu-
lations also showed that the rotations of the aryl fragments
are correlated. Moreover, it was observed that the counter-
ions stay in close proximity to the cations, and rotate with
the aryl fragments. From these calculated rotational barriers
can be concluded that for the Dendriphos compounds,
rotation around the phosphorus–aryl bond is virtually
impossible under ambient conditions.
To obtain further information about the electronic effects on
phosphorus, IR spectra of the phosphine oxides and sulfides
were recorded. Unfortunately, the PvO vibrations could
not be distinguished from the aromatic and benzyl–phenyl
ether vibrations. The PvS vibrations are expected around
802–660 and 730–550 cm21 21
In the spectrum of
.
SvP(NCN)3 4, sharp bands are observed that can be
assigned to these vibrations (708 and 558 cm21). Unfortu-
nately, in the spectra of 6a–d these bands are obscured by
the presence of many other vibrations and cannot be
observed.
To gain insight in the size, cone angle, and three-
dimensional structure of these novel hexa-ammonium
salts, we attempted to use X-ray diffraction. Although
recrystallization of OvP(NCN)3 (3) from hexane yielded
analytically pure crystals, diffraction patterns could not be
observed as these crystals underwent a solid–solid phase
change upon cooling. Furthermore, when measured at
2608C, just above the phase change, included hexane
slowly started to evaporate from the crystals leading to
disruption of the 3-D structure. For this reason we turned to
molecular modeling to obtain information on the structural
features of the Dendriphos compounds in particular.
Because of the presence of meta-substituents in these
triarylphosphines, hindered rotation of the aryl rings around
the P–C(aryl) bond would be expected. This is reflected in
their 1H, 13C, and 31P NMR spectra, which become
increasingly broad going from 4 to 6d and from 2 to 8d.
To investigate this, NMR spectra of 8d were recorded at
elevated temperatures. Upon heating, the signals of the
NMe2 and (NMe2)CH2 protons narrow, while the signals
corresponding to the dendron-aryl protons remain broad.
Furthermore, decoalescence of the signals corresponding to
Molecular modeling was further applied to study the three-
dimensional structure of the hexacationic phosphines. These
studies showed that the phosphorus atom becomes increas-
ingly buried in the organic bulk with increasing size of the
N-substituents. For the Dendriphos compounds 8c,d this
results in disc-like rather than cone-like structures. In the
case of [P(NCN)3·(G2)6]Br6 (8d), a nanosize disc is
obtained with a diameter of 3.5 nm and a height of
approximately 2.0 nm† (Fig. 4). These discs consist of a
rigid triarylphosphine core, surrounded by a ring of ionic
groups. The ammonium groups are shielded by a soft shell
of dendrons with increased peripheral conformational
freedom. Based on our calculations, a cone angle20 of
1808 could be tentatively assigned to this macromolecular
phosphine.
3. Conclusions
In summary, a new bifunctional hexa-amine triaryl-
phosphine core molecule was prepared and converted to
the corresponding phosphine oxide and phosphine sulfide.
The resulting protected phosphines were used as core
molecules for the synthesis of hexa-ammonium salts via
alkylation of the amine functionalities. The hexacationic
phosphine sulfides could be reduced to the free phosphines
via a general method. In this way, a series of hexacationic
phosphines has become available, starting from relatively
straightforward building blocks, and following an easy
protection, alkylation, deprotection reaction sequence. The
†
Figure 3. Energy plot of the rotation barrier of the aryl rings around the
P–C(aryl) bond in triarylphosphines against the type of meta-substituent.
Obtained by taking the mean value of several distance measurements in
the structure.

3994
R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
Figure 4. Top and side view of [P(NCN)3(G2)6]Br6 (8d), showing the disc-like character and dimensions of this Dendriphos. The P-atom is shown in orange,
the Br2 counterions are shown in green. Three of the Br2 ions are found on top of and three below the disc.
solubility of these hexacations dramatically changes by
varying the N-substituent, which hints at the application of
these compounds in both aqueous and non-aqueous media.22
From 31P NMR data it can be concluded that the C–P–C
angles of these meta-substituted Dendriphos ligands
increase with increasing size of the (meta-)N-substituent,
opening the way for tuning the properties of metal
complexes derived from these ligands by means of variation
of this substituent. The position and size of the counterions
in such complexes could be of additional importance to the
final C–P–C angle and to the effective cone angle of these
hexacationic ligands. Both the physical properties as well as
the coordination behaviour of this series of phosphines are
the subject of current investigations.
4.1.1. Synthesis of OvP(NCN)3 (3). To a solution of
1-bromo-3,5-bis[(dimethylamino)methyl]benzene (Br-NCN,
1) (4.84 g, 17.9 mmol) in Et2O, 2 equiv. of t-BuLi (1.5 M in
pentane, 23.8 mL, 35.7 mmol) were added dropwise under N2
at2788C. The resultinglight brown suspension was stirred for
1 h, after which PBr3 (0.56 mL, 5.96 mmol) was added
dropwise. The mixture was allowed to warm up to room
temperature, andwasstirredfor16 h.DegassedH2O(100 mL)
was added and the organic layer was separated. The aqueous
layer was extracted twice with CH2Cl2 (2£75 mL), the
combined organic fractions were dried over MgSO4 and
evaporated in vacuo, affording crude 2 as a yellow oil. To this
oil, THF (100 mL) was added and the solution was cooled to
08C.H2O2 (35%inH2O, 5 mL)wasaddedandthemixturewas
stirredinairfor3 h.AqueousNaOH(1 M,100 mL)wasadded
and the mixture was concentrated. The product was extracted
with CH2Cl2 (2£100 mL), the combined organic layers were
dried over MgSO4, and evaporated to dryness. The crude
product was further purified by crystallization from hexane
4. Experimental
4.1. General
1
and obtained as colorless crystals (1.21 g, 33%). H NMR
(200 MHz, CDCl3): d¼7.49, 7.43 (overlapping s and d, 9H,
All lithiation reactions and manipulations of the phosphines
were performed using standard Schlenk techniques. Et2O
was dried over Na/benzophenone and distilled prior to use.
1-Bromo-3,5-bis[(dimethylamino)methyl]benzene, 117 and
Ar-H), 3.37 (s, 12H, Ar–CH2), 2.15 (s, 36H, N(CH3)2); 13
C
{1H} NMR (50 MHz, CDCl3): d¼139.61 (d, Ar–C,
2JCP¼12.0 Hz), 133.62 (s, Ar–C), 133.49 (s, Ar–C), 131.67
3
(d, Ar–C, JCP¼10.1 Hz), 64.00 (s, ArCH2), 45.48 (s,
the benzylic bromides of the Frechet dendrons (G1 and
N(CH3)2); 31P {1H} NMR (81 MHz, CDCl3): d¼30.13 (s);
MALDI-TOF MS: m/z 620.39 Mþ (100%). Anal. calcd for
C36H57N6OP: C 69.64, H 9.25, N 13.54, P 4.99. Found: C
69.64, H 9.21, N 13.40, P 4.91.
´
G2)18 were prepared according to literature procedures.
PBr3 was distilled prior to use. All other starting materials
were obtained from Acros Chimica or Sigma-Aldrich and
1
were used without further purification. H, 13C {1H}, and
31P {1H} NMR measurements were carried out on a Varian
Inova 200 MHz spectrometer at 258C and all chemical shifts
(d) are given in ppm. The 1H NMR spectra were referenced
against the solvent residual peak and the 31P NMR spectra
against 85% H3PO4 as external standard. MALDI-TOF
mass spectra were recorded on a Voyager-DE BioSpec-
trometry Workstation mass spectrometer. Nano-ESI mass
spectra were recorded on a Micromass LC-TOF mass
spectrometer. IR spectra were recorded on a PE-system
2000 FT-IR spectrometer. Elemental analyses were per-
formed by Kolbe, Mu¨lheim a/d Ruhr, Germany.
4.1.2. Synthesis of SvP(NCN)3 (4). SvP(NCN)3 was
prepared similarly to the synthesis described above for
OvP(NCN)3, starting from Br-NCN (1) (4.54 g,
16.75 mmol) and PBr3 (0.44 mL, 4.68 mmol). To the
resulting crude P(NCN)3 (2), sulfur (0.54 g, 16.84 mmol)
dissolved in CS2 (50 mL) was added in one portion. The
mixture was stirred for 16 h and evaporated to dryness,
yielding a yellow sticky solid. The crude mixture was
purified by addition of HCl (1 M, 100 mL) followed by
washing with CH2Cl2, and addition of NaOH (4 M, 150 mL)
followed by extraction of the product with CH2Cl2

R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
3995
1
(2£100 mL). The product containing organic fractions were
combined, dried over MgSO4, and evaporated to dryness.
The obtained sticky material was washed twice with cold
Ar–C, JCP¼105.30 Hz), 133.48 (s, Ar–C), 130.78 (s,
Ar–C), 130.39 (d, Ar–C, 2JCP¼13.3 Hz) 129.11 (s, Ar–C),
127.62 (s, Ar–C), 67.17 (s, ArCH2), 48.72 (s, N(CH3)2); 31
P
1
pentane to afford a light yellow solid (2.98 g, 84%). H
{1H} NMR (81 MHz, CD3OD): d¼32.93 (s); nano-ESI MS:
m/z 743.25 [M22Br]2þ (18%), 469.18 [M23Br2]3þ (92%),
331.72 [M24Br2]4þ (100%). Anal. calcd for C78H99Br6-
N6OP: C 56.88, H 6.06, N 5.10, P, 1.88. Found: C 57.02, H
6.12, N 5.13, P 1.98.
NMR (200 MHz, CDCl3): d¼7.51, 7.44 (overlapping s and
d, 9H, Ar-H), 3.36 (s, 12H, Ar–CH2), 2.16 (s, 36H,
N(CH3)2); 13C {1H} NMR (50 MHz, CDCl3): d¼139.55 (d,
Ar–C, 2JCP¼12.4 Hz), 133.17 (s, Ar–C), 131.68 (d, Ar–C,
3
1JC–P¼84.3 Hz), 131.81 (d, Ar–C, JCP¼11.1 Hz), 64.00
(s, ArCH2), 45.49 (s, N(CH3)2); 31P {1H} NMR (81 MHz,
CDCl3): d¼43.40 (s); MALDI-TOF MS: m/z 637.09 Mþ
(100%). Anal. calcd for C36H57N6PS: C 67.89, H 9.02, N
13.19, P 4.86. Found: C 67.86, H 8.95, N 13.08, P 4.99.
4.1.6. Synthesis of [SvP(NCN)3·Bz6]Br6 (6b). Similar to
the procedure described for 5b, starting from SvP(NCN)3
(4, 0.25 g, 0.39 mmol) and benzyl bromide (0.30 mL,
2.51 mmol), 6b was obtained as a light yellow solid
1
(0.61 g, 94%). H NMR (200 MHz, CD3OD): d¼8.79 (d,
4.1.3. Synthesis of [OvP(NCN)3·Me6]I6 (5a). To a
solution of OvP(NCN)3 (3, 0.33 g, 0.53 mmol) in CH2Cl2
(50 mL) was added MeI (0.40 mL, 6.42 mmol) and the
mixture was stirred for 4 h, during which the product
precipitated. The solid material was collected and washed
with CH2Cl2 (2 x 30 mL), and further purified by
precipitation from MeOH (10 mL) with Et2O (75 mL),
affording a light yellow solid (0.54 g, 68%). 1H NMR
6H, Ar-H, 3JHP¼13.6 Hz), 8.14 (s, 3H, Ar-H), 7.8–7.4 (m,
30H, CH2Ar–H), 4.86 (s, 12H, ArCH2), 4.79 (s, 12H,
ArCH2), 3.241 (s, 36H, N(CH3)2); 13C {1H} NMR (50 MHz,
CD3OD): d¼141.42 (s, Ar–C), 139.308 (d, Ar–C,
3JCP¼11.9 Hz), 136.32 (overlapping d, Ar–C), 133.47 (s,
Ar–C), 130.78 (s, Ar–C), 130.10 (d, Ar–C, 2JCP¼13.3 Hz),
129.11 (s, Ar–C), 127.62 (s, Ar–C), 67.16 (s, ArCH2),
48.70 (s, N(CH3)2); 31P {1H} NMR (81 MHz, CD3OD):
d¼40.68 (s); nano-ESI MS: m/z 751.24 [M22Br]2þ (28%),
447.50 [M23Br2]3þ (100%), 335.70 [M24Br2]4þ (71%).
Anal. calcd for C78H99Br6N6PS: C 34.26, H 5.13, N 5.71, P,
2.10. Found: C 34.21, H 5.20, N 5.64, P1.97.
2
(200 MHz, D2O): d¼8.69 (d, 6H, Ar-H, JHP¼12.6 Hz),
8.13 (s, 3H, Ar-H), 4.74 (s, 12H, Ar–CH2), 3.24 (s, 48H,
N(CH3)3; 13C {1H} NMR (50 MHz, D2O): d¼142.71 (s,
Ar–C), 138.68 (d, Ar–C, 3JCP¼10.6 Hz), 132.17 (d, Ar–C,
2
1JC–P¼107.3 Hz), 130.55 (d, Ar–C, JCP¼12.9 Hz), 67.71
(s, ArCH2), 53.30 (s, N(CH3)3); 31P {1H} NMR (81 MHz,
D2O): d¼32.93 (s); nano-ESI MS: m/z 1345.58 [M2I2]þ
(1%), 609.11 [M22I2]2þ (28%), 363.79 [M23I2]3þ(82%),
241.24 [M24I2]4þ (100%), 167.76 [M25I2]5þ (46%),
118.81 [M26I2]6þ (7%). Anal. calcd for C42H75I6N6OP: C
34.26, H 5.13, N 5.71, P, 2.10. Found: C 34.21, H 5.20, N
5.64, P 1.97.
4.1.7. Synthesis of [OvP(NCN)3·(G1)6]Br6 (5c). To a
solution of OvP(NCN)3 (3, 0.55 g, 0.89 mmol) in CH2Cl2
(100 mL) was added G1Br (2.04 g, 5.32 mmol) in one
portion after which the mixture was stirred at room
temperature for 16 h. The resulting solution was concen-
trated and the product was precipitated with Et2O (75 mL).
The crude product was further purified by precipitating three
times from CH2Cl2 (20 mL) with Et2O (75 mL) affording 5c
as a white solid (1.95 g, 75%). 1H NMR (200 MHz, CDCl3):
d¼8.81 (broad, 9H, overlapping ArH), 7.8–7.0 (m, 60H,
ArH), 6.94 (broad s, 16H, ArH), 6.60 (broad s, 8H, ArH),
5.4–4.5 (broad, 48H, overlapping ArCH2), 3.05 (broad s,
36H, N(CH3)2); 13C {1H} NMR (50 MHz, CDCl3):
d¼160.13, 142.88, 139.46, 136.50, 133.66 (Ar–C) 130.20
4.1.4. Synthesis of [SvP(NCN)3·Me6]I6 (6a). Similar to
the procedure described for 5a, starting from SvP(NCN)3
(4, 0.20 g, 0.31 mmol) and MeI (0.20 mL, 3.21 mmol), 6a
was obtained as a light yellow solid (0.36 g, 76%). 1H NMR
2
(200 MHz, D2O): d¼8.51 (d, 6H, Ar-H, JHP¼13.6 Hz),
8.08 (s, 3H, Ar-H), 4.75 (s, 12H, ArCH2), 3.25 (s, 48H,
N(CH3)3); 13C {1H} NMR (50 MHz, D2O): d¼142.77 (s,
Ar–C), 138.70 (d, Ar–C, 3JCP¼11.0 Hz), 132.17 (d, Ar–C,
2
(d, Ar–C, JCP¼13.3 Hz), 130.05–127.97 (m, overlapping
Ar–C), 112.69, 104.71 (Ar–C) 70.48 (OCH2), 66.98
(broad, CH2NCH2) 49.36 (N(CH3)2); 31P {1H} NMR
(81 MHz, CDCl3): d¼26.01 (broad s); nano-ESI MS: m/z
1380.63 [M22Br]2þ (3%), 893.46 [M23Br2]3þ (40%),
650.36 [M24Br2]4þ (100%). Anal. calcd for C162H171Br6-
N6O13P: C 66.62, H 5.90, N 2.88, P 1.06. Found: C 66.48,
H6.03, N 2.96, P 1.13.
2
1JC–P¼107.3 Hz), 130.55 (d, Ar–C, JCP¼13.4 Hz), 67.69
(s, ArCH2), 53.30 (s, N(CH3)3); 31P {1H} NMR (81 MHz,
D2O): d¼44.29 (s); nano-ESI MS: m/z 1361.64 [M2I2]þ
(1%), 617.17 [M22I2]2þ (21%), 368.63 [M23I2]3þ
(100%), 244.22 [M24I2]4þ (32%). Anal. calcd for
C42H75I6N6PS: C 33.89, H 5.08, N 5.65, P 2.08. Found: C
34.08, H 5.20, N 5.57, P 2.07.
4.1.8. Synthesis of [SvP(NCN)3·(G1)6]Br6 (6c). Similar to
the procedure described for 5c, starting from SvP(NCN)3
(4, 0.23 g, 0.36 mmol) and G1Br (0.83 g, 2.17 mmol), 6c
was obtained as a white solid (0.98 g, 92%). 1H NMR
(200 MHz, CDCl3): d¼8.79 (broad, 9H, overlapping ArH),
7.6–7.0 (m, 60H, ArH), 6.94 (broad s, 16H, ArH), 6.63
(broad s, 8H, ArH), 5.4–4.5 (broad, 48H, overlapping
ArCH2), 3.05 (broad s, 36H, N(CH3)2); 13C {1H} NMR
(50 MHz, CDCl3): d¼160.12, 142.87, 139.46, 136.42,
4.1.5. Synthesis of [OvP(NCN)3·Bz6]Br6 (5b). To a
solution of OvP(NCN)3 (3, 1.11 g, 1.79 mmol) in CH2Cl2
was added benzyl bromide (1.4 mL, 11.71 mmol) and the
mixture was stirred for 16 h. The yellow suspension was
evaporated to dryness and the resulting solid was pre-
cipitated twice from MeOH (10 mL) using Et2O (75 mL),
affording a light yellow solid (2.13 g, 72%). 1H NMR
(200 MHz, CD3OD): d¼8.69 (d, 6H, Ar-H, 3JHP¼12.6 Hz),
8.21 (s, 3H, Ar-H), 7.8–7.4 (m, 30H, CH2Ar–H), 4.85 (s,
12H, ArCH2), 4.74 (s, 12H, ArCH2), 3.10 (s, 36H,
N(CH3)2); 13C {1H} NMR (50 MHz, CD3OD): d¼142.38
2
133.65 (Ar–C) 130.14 (d, Ar–C, JCP¼13.3 Hz), 130.08–
127.96 (m, Ar–C), 112.65, 104.79 (Ar–C) 70.61 (OCH2),
67.02 (broad, CH2NCH2) 49.43 (N(CH3)2); 31P {1H} NMR
(81 MHz, CDCl3): d¼40.11 (broad s); nano-ESI MS: m/z
3
(s, Ar–C), 139.24 (d, Ar–C, JCP¼11.0 Hz), 133.75 (d,

3996
R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
1388.68 [M22Br]2þ (7%), 898.82 [M23Br2]3þ (56%),
654.37 [M24Br2]4þ (57%). Anal. calcd for C162H171Br6-
N6O12PS: C 66.26, H 5.87, N 2.86, P 1.05. Found: C 66.18,
H 5.98, N 2.91, P 1.06.
and washing with Et2O (2£10 mL), HMPT (0.18 mL,
1.0 mmol) was added and the mixture was stirred for 1 h.
After removal of all volatiles, the product was precipitated
twice from CH2Cl2 with Et2O, affording a white solid
(0.095 g, 72%). 1H NMR (200 MHz, CDCl3): d¼8.18
(broad, 6H, ArH), 8.00 (broad s, 3H, ArH), 7.5–7.0 (m,
60H, ArH), 6.65 (broad, 24H, overlapping ArH), 4.97
(broad s, 24H, ArCH2), 4.54, 4.36 (broad, 24H, overlapping
ArCH2), 2.78 (broad s, 36H, N(CH3)2); 31P {1H} NMR
(81 MHz, CDCl3): d¼0.10 (broad s).
4.1.9. Synthesis of [OvP(NCN)3·(G2)6]Br6 (5d). Similar
to the procedure described for 5c, starting from
OvP(NCN)3 (3, 0.18 g, 0.29 mmol) and G2Br (1.42 g,
1.75 mmol), 5d was obtained as a white solid (0.89 g,
55%).1H NMR (200 MHz, CDCl3): d¼9.0–8.6 (broad, 9H,
overlapping ArH), 7.8–6.8 (m, 120H, ArH), 6.64, 6.50, 6.47
(broad, 54H, overlapping ArH), 5.3–4.5 (broad, 84H,
overlapping ArCH2), 3.01 (broad s, 36H, N(CH3)2); 13C
{1H} NMR (50 MHz, CDCl3): d¼160.23, 160.05, 139.06,
136.98, 136.53 (Ar–C), 128.79–127.87 (m, Ar–C), 112.55,
106.85, 104.82, 101.79 (Ar–C), 70.25 (OCH2), 67.50
(broad, CH2NCH2), 49.38 (N(CH3)2; 31P {1H} NMR
(81 MHz, CDCl3): d¼25.24 (broad s); nano-ESI MS: m/z
1742.71 [M23Br]3þ (18%), 1286.57 [M24Br2]4þ (100%).
Anal. calcd for C330H315Br6N6O37P: C 72.49, H 5.81, N
1.54, P 0.57. Found: C 72.34, H 5.93, N 1.58, P 0.63.
4.4. General procedure for the reduction of the
hexacationic phosphine sulfides
The phosphine sulfide (6a–d) was refluxed in a 2:1 mixture
of MeOH and P(n-Bu)3 for 16 h. The mixture was
concentrated by distilling off all volatiles, after which
Et2O was added to afford the crude product as a light yellow
or white precipitate. The SvP(n-Bu)3 side product was
removed by repeated precipitation of the product from
MeOH with Et2O.
4.1.10. Synthesis of [SvP(NCN)3·(G2)6]Br6 (6d). Similar
to the procedure described for 5c, starting from
SvP(NCN)3 (4, 0.23 g, 0.36 mmol) and G2Br (0.83 g,
2.17 mmol), 6d was obtained as a white solid (0.98 g, 92%).
1H NMR (200 MHz, CDCl3): d¼9.2–8.6 (broad, 9H,
overlapping ArH), 7.8–7.0 (m, 120H, ArH), 6.92, 6.62,
6.42 (broad, 54H, overlapping ArH), 5.3–4.4 (broad, 84H,
overlapping ArCH2), 2.92 (broad s, 36H, N(CH3)2); 13C
{1H} NMR (50 MHz, CDCl3): d¼160.23, 160.05, 139.06,
136.97 (Ar–C), 128.80–127.86 (m, Ar–C), 112.52, 106.75,
104.91, 101.87 (Ar–C), 70.23 (OCH2), 67.51 (broad,
CH2NCH2), 49.46 (N(CH3)2; 31P {1H} NMR (81 MHz,
CDCl3): d¼40.42 (broad s); nano-ESI MS: m/z 1747.79
[M23Br]3þ (100%), 1290.69 [M24Br2]4þ (45%). Anal.
calcd for C330H315Br6N6O36PS: C 72.28, H 5.79, N 1.53, P
0.56. Found: C 72.20, H 5.71, N 1.60, P 0.61.
4.4.1. Analysis of [P(NCN)3·Me6]I6 (8a). Starting from 6a
1
(0.13 g, 0.087 mmol), to yield 0.11 g 8a (86%); H NMR
(200 MHz, D2O): d¼8.02 (d, 6H, Ar-H, 2JHP¼7.4 Hz), 7.73
(s, 3H, Ar-H), 4.65 (s, 12H, ArCH2), 3.24 (s, 48H,
N(CH3)3); 13C {1H} NMR (50 MHz, D2O): d¼140.44 (d,
1
2
Ar–C, JCP¼19.9 Hz), 139.08 (d, Ar-H, JCP¼13.1 Hz),
138.72 (s, Ar–C), 129.68 (d, Ar–C, 3JCP¼7.3 Hz), 68.24 (s,
ArCH2), 53.10 (s, N(CH3)3); 31P {1H} NMR (81 MHz,
D2O): d¼23.92 (s); nano-ESI MS: m/z 601.23 [M22I2]2þ
(23%), 358.86 [M23I2]3þ (100%), 237.61 [M24I2]4þ
(75%).
4.4.2. Analysis of [P(NCN)3·Bz6]Br6 (8b). Starting from 6b
1
(60 mg, 0.036 mmol), to yield 51 mg 8b (88%); H NMR
2
(200 MHz, CD3OD): d¼8.41 (d, 6H, Ar-H, JHP¼4.8 Hz),
7.88 (s, 3H, ArH), 7.8–7.4 (m, 30H, ArH), 4.78 (s, 12H,
C
ArCH2), 4.72 (s, 12H, ArCH2), 3.06 (s, 36H, N(CH3)2); 13
4.2. Incorporation of MO in [OvP(NCN)3·(G1)6]Br6
(5c)
{1H} NMR (50 MHz, CDCl3): d¼140.08 (d, Ar–C,
2JCP¼20.9 Hz), 138.83 (s, Ar–C,), 133.43, 130.74 (s,
3
Ar–C), 129.79 (d, Ar–C, JCP¼13.3 Hz), 129.07 (s,
A previously reported procedure was applied,15 starting
from a solution of 6c (0.36 g, 0.12 mmol) in CH2Cl2
(50 mL) and a solution of MO (0.24 g, 0.74 mmol) in H2O
(50 mL). The two solutions were mixed and stirred
vigorously for 5 min, after which the layers were allowed
to separate. An orange CH2Cl2 layer and a slightly yellow
H2O layer were obtained. The organic layer was separated,
washed with H2O, dried over Mg2SO4, filtered, and
evaporated to dryness, affording a bright orange solid
Ar–C), 127.69 (s, Ar–C), 67.64, 67.17 (s, ArCH2), 48.75
(s, N(CH3)2); 31P {1H} NMR: d¼21.82 (s); nano-ESI MS:
m/z 735.17 [M22Br2]2þ (12%), 464.16 [M23Br2]3þ
(68%), 328.10 [M24Br2]4þ (100%).
4.4.3. Analysis of [P(NCN)3·(G1)6]Br6 (8c). Starting from
6c (0.10 g, 0.034 mmol), to yield 61 mg 8c (62%); 1H NMR
(200 MHz, CDCl3): d¼8.39 (broad, 9H, overlapping ArH),
7.7–7.1 (m, 60H, ArH), 7.00 (broad s, 16H, ArH), 6.62
(broad s, 8H, ArH), 5.3–4.5 (broad, 48H, overlapping
ArCH2), 3.05 (broad s, 36H, N(CH3)2); 13C {1H} NMR
(50 MHz, CDCl3): d¼160.08, 136.45 (Ar–C) 129.28–
127.96 (m, Ar–C), 112.69, 104.56 (Ar–C), 70.49 (OCH2),
67.55 (broad, CH2NCH2) 49.52 (N(CH3)2); 31P {1H} NMR
(81 MHz, CDCl3): d¼23.88 (broad s); nano-ESI MS,
measured as the phosphine oxide: m/z 893.39
[Mox23Br2]3þ (53%), 650.39 [Mox24Br2]4þ (41%).
1
(0.51 g, 96%). H NMR (200 MHz, CDCl3): d¼9.0–8.6
(broad, 9H, overlapping ArH), 8.0–7.6 (broad, 36H,
overlapping ArH(MO)) 7.5–7.0 (m, 60H, ArH), 6.68
(overlapping s, 28H, ArH(MO), ArH), 6.54 (broad s, 8H,
ArH), 5.2–4.4 (broad, 48H, overlapping ArCH2), 3.03 (s,
36H, NMe2(MO)), 3.05 (broad s, 36H, N(CH3)2).
4.3. Reduction of 6c to afford [P(NCN)3·G1)6]OTf6 (7)
To a solution of 6c (0.115 g, 0.039 mmol) in CH2Cl2
(10 mL) was added MeOTf (0.15 mL, 1.3 mmol) and the
mixture was stirred for 30 min. After removal of all volatiles
4.4.4. Analysis of [P(NCN)3·(G2)6]Br6 (8d). Starting from
6d (0.25 g, 0.046 mmol), to yield 0.21 g 8d (84%); 1H NMR
(200 MHz, CDCl3): d¼9.0–8.2 (broad, 9H, overlapping

R. Kreiter et al. / Tetrahedron 59 (2003) 3989–3997
3997
Kamer, P. J. C.; van Leeuwen, P. W. N. M.; Reek, J. N. H.
Chem. Rev. 2002, 102, 3717–3756.
ArH), 7.6–7.0 (m, 120H, ArH), 6.89, 6.60, 6.41 (broad,
54H, overlapping ArH), 5.3–4.4 (broad, 84H, overlapping
ArCH2), 2.94 (broad s, 36H, N(CH3)2); 13C {1H} NMR
(50 MHz, CDCl3): d¼160.15, 159.97, 138.97, 136.94,
136.46 (Ar–C) 129.26–127.82 (m, Ar–C), 112.53,
106.80, 104.77, 101.83 (Ar–C), 70.25 (OCH2), 67.42
(broad, CH2NCH2), 49.47 (N(CH3)2; 31P {1H} NMR
(81 MHz, CDCl3): d¼23.83 (broad s); nano-ESI MS: m/z
1735.33 [M23Br2]3þ (79%), 1281.94 [M24Br2]4þ (67%).
3. For a review on phosphorus containing dendrimers: Majoral,
J.-P.; Caminade, A.-M. Top. Curr. Chem. 1998, 197, 79–124.
4. De Groot, D.; Reek, J. N. H.; Kamer, P. C. J.; Van Leeuwen,
P. W. N. M. Eur. J. Org. Chem. 2002, 6, 1085–1095.
5. Reetz, M. T.; Giebel, D. Angew. Chem Int. Ed. 2000, 39,
2498–2501.
¨
6. Kollner, C.; Pugin, B.; Togni, A. J. Am. Chem. Soc. 1998, 120,
10274–10275.
4.5. Computational details
7. Maraval, V.; Caminade, A-M.; Majoral, J.-P. Organometallics
2000, 19, 4025–4029.
Geometry optimizations were performed on a molecular
mechanics level using the SPARTAN 5.1.1 (UNIX)
package,23 with a MMFF94 force field. The input data
was derived from the builder routine in the package.
Calculations on the rotation around the P–C bonds were
carried out using the coordinate driving routine in the
package on a molecular mechanics level. The energy
profiles of a rotation around 7208 were determined and for
the barrier of such a rotation, the difference between
minimum and maximum energy was taken. In these
calculations, the individual aryl rings of the triarylphosphine
core were forced in a planar geometry using the constrain
dihedral option, to prevent deformations.
8. (a) Hovestad, N. J.; Eggeling, E. B.; Heidbu¨chel, H. J.;
Jastrzebski, J. T. B. H.; Kragl, U.; Keim, W.; Vogt, D.; van
Koten, G. Angew. Chem. Int. Ed. 1999, 38, 1655–1658.
(b) Wijkens, P.; Jastrzebski, J. T. B. H.; van der Schaaf, P.;
Kolly, R.; Hafner, A.; van Koten, G. Org. Lett. 2000, 2,
1621–1624.
9. (a) Miedaner, A.; Curtis, C. J.; Barkley, R. M.; Dubois, D. L.
Inorg. Chem. 1994, 33, 5482–5490. (b) Petrucci-Samija, M.;
Guillemette, V.; Dasgupta, M.; Kakkar, A. K. J. Am. Chem.
Soc. 1999, 121, 1968–1969.
10. Oosterom, G. E.; Steffens, S.; Reek, J. N. H.; Kamer, P. C. J.;
Van Leeuwen, P. W. N. M. Top. Catal. 2002, 19, 61–73.
11. Brunner, H. J. Organomet. Chem. 1995, 500, 39–46.
12. Fan, Q-H.; Chen, Y-M.; Chen, X-M.; Jiang, D-Z.; Xi, F.;
Chan, A. S. C. Chem. Commun. 2000, 789–790.
13. Catalano, V. J.; Parodi, N. Inorg. Chem. 1997, 36, 537–541.
14. Balaji, B. S.; Obora, Y.; Ohara, D.; Koide, S.; Tsuji, Y.
Organometallics 2001, 20, 5342–5350.
Acknowledgements
The authors like to acknowledge C. Versluis (Utrecht
University, Department of Biomolecular Mass Spec-
trometry) for conducting the mass spectrometry experi-
ments and Dr T. Visser (Utrecht University, Department of
Inorganic Chemistry and Catalysis) for help with the IR
measurements.
15. Kleij, A. W.; van de Coevering, R.; Klein Gebbink, R. J. M.;
Noordman, A. M.; Spek, A. L.; van Koten, G. Chem. Eur. J.
2001, 7, 181–192.
16. Van de Coevering, R.; Kuil, M.; Klein Gebbink, R. J. M.; van
Koten, G. Chem. Commun. 2002, 1636–1637.
17. Steenwinkel, P.; James, S. L.; Grove, D. M.; Veldman, N.;
Spek, A. L.; van Koten, G. Chem. Eur. J. 1996, 2, 1440–1445.
´
18. Hawker, C. J.; Frechet, J. M. J. J. Am. Chem. Soc. 1990, 112,
7638–7647.
References
1. For general reviews on dendrimers see (a): Newkome, G. R.;
19. Gilbertson, S. R.; Collibee, S. E.; Agarkov, A. J. Am. Chem.
Soc. 2000, 122, 6522–6523.
¨
Moorefield, C. N.; Vogtle, F. Dendrimers and Dendrons:
Concepts, Syntheses and Applications; VCH: Weinheim,
20. Tolman, C. A. Chem. Rev. 1977, 77, 313–348.
21. Bellamy, L. J. The Infra-red Spectra of Complex Molecules;
Wiley: New York, 1975; pp 348–349, 360–361.
22. Aqueous-Phase Organometallic Chemistry; Concepts and
Applications; Cornils, B., Herrman, W. A., Eds.; Wiley-
VCH: Weinheim, 1998.
¨
2001. (b) Fischer, M.; Vogtle, F. Angew. Chem. Int. Ed.
1999, 38, 884–905. For a review on dendrimer applications
see: (c) Bosman, A. W.; Janssen, H. M.; Meijer, E. W. Chem.
Rev. 1999, 99, 1665–1688.
2. Recent reviews on dendrimers in catalysis: (a) Kreiter, R.;
Kleij, A. W.; Klein Gebbink, R. J. M.; van Koten, G. Top.
Curr. Chem. 2001, 217, 163–199. (b) Astruc, D.; Chardac, F.
Chem. Rev. 2001, 101, 2991–3024. (c) van Heerbeek, R.;
23. SPARTAN SGI version 5.1.1; Wavefunction Inc.: 18401 Von
Karman Ave., Ste. 370 Irvine, CA 92612 USA.