Diagnostic organometallic and metallodendritic materials for SO2 gas detection: Reversible binding of sulfur dioxide to arylplatinum(II) complexes

Article abstract of DOI:10.1002/(SICI)1521-3765(20000417)6:8<1431::AID-CHEM1431>3.0.CO;2-D

A series of square-planar platinum(II) complexes of the N,C,N'- terdentate-coordinating monoanionic 'pincer' ligand, [PtX(4-E-2,6- {CH₂NRR'}₂-C₆H₂)] (X = Cl, Br, I, tolyl; R, R' = Et, Me; E = H, OH, OSiMe₂tBu) has been prepared. In the presence of sulfur dioxide, these complexes spontaneously adsorb this gas to form pentacoordinated adducts. Solid-state crystalstructure analyses of the SO₂ adducts 8c (X = I, R = R' = Me, E = OSiMe₂tBu) and 11 (X = Cl, R = R' = Me, E = OH) show a square- pyramidal geometry around the metal center with SO₂ in the apical position. Most interestingly, the adduct 11 forms similar Pt-Cl··· H-O hydrogen- bonded α-type networks as the corresponding SO₂-free complex 5. The conservation of the supramolecular information (hydrogen-bonded self- assembly) throughout a reaction (SO₂ adsorption) is unprecedented in crystal engineering. Adduct formation in the solid state or in solution is fast and reversible and is indicated by a characteristic color change of the material from colorless to bright orange. Since facile methods have been developed to remove SO₂ from the adducts and to regenerate the square-planar starting complexes, these complexes fulfill several essential prerequisites of sensor materials for repeated diagnostic SO₂ detection. The platinum sensors have been found to be highly selective for sulfur dioxide and particularly sensitive for submilimolar to molar gas quantities. Their response capacity is tuneable by electronic and steric modifications of the ligand array by introduction of, for example, different substituents on the nitrogen donors. The periphery of dendrimers is shown to be an appropriate macromolecular support for anchoring the detection-active sites, thus allowing full recovery of the sensor materials for repeated use. By using this concept, metallo- dendrimers 3 and 15 have been prepared. Owing to the dendritic connectivity, these sensors are suitable for repetitive qualitative and quantitative detection of small amounts of SO₂.

Full text of DOI:10.1002/(SICI)1521-3765(20000417)6:8<1431::AID-CHEM1431>3.0.CO;2-D

FULL PAPER
Diagnostic Organometallic and Metallodendritic Materials for SO2
Gas Detection: Reversible Binding of Sulfur Dioxide to
Arylplatinum(ii) Complexes
Martin Albrecht,[a] Robert A. Gossage,[a] Martin Lutz,[b]
Anthony L. Spek,[b] and Gerard van Koten*[a]
Abstract: A series of square-planar
platinum(ii) complexes of the N,C,N'-
terdentate-coordinating monoanionic
ular information (hydrogen-bonded self-
assembly) throughout a reaction (SO2
adsorption) is unprecedented in crystal
engineering. Adduct formation in the
solid state or in solution is fast and
reversible and is indicated by a charac-
teristic color change of the material
from colorless to bright orange. Since
facile methods have been developed to
remove SO2 from the adducts and to
regenerate the square-planar starting
complexes, these complexes fulfill sev-
eral essential prerequisites of sensor
materials for repeated diagnostic SO2
detection. The platinum sensors have
been found to be highly selective for
sulfur dioxide and particularly sensitive
for submilimolar to molar gas quantities.
Their response capacity is tuneable by
electronic and steric modifications of the
ligand array by introduction of, for
example, different substituents on the
nitrogen donors. The periphery of den-
drimers is shown to be an appropriate
macromolecular support for anchoring
the detection-active sites, thus allowing
full recovery of the sensor materials for
repeated use. By using this concept,
metallo-dendrimers 3 and 15 have been
prepared. Owing to the dendritic con-
nectivity, these sensors are suitable for
repetitive qualitative and quantitative
detection of small amounts of SO2.
ªpincerº
ligand,
[PtX(4-E-2,6-
{CH2NRR'}2-C6H2)] (X ˆ Cl, Br, I, tolyl;
R, R' ˆ Et, Me; E ˆ H, OH, OSiMe2-
tBu) has been prepared. In the presence
of sulfur dioxide, these complexes spon-
taneously adsorb this gas to form penta-
coordinated adducts. Solid-state crystal-
structure analyses of the SO2 adducts 8c
(X ˆ I, R ˆ R' ˆ Me, E ˆ OSiMe2tBu)
and 11 (X ˆ Cl, R ˆ R' ˆ Me, E ˆ OH)
show
a square-pyramidal geometry
around the metal center with SO2 in
the apical position. Most interestingly,
the adduct 11 forms similar Pt Cl ´´´
H O hydrogen-bonded a-type networks
as the corresponding SO2-free complex
5. The conservation of the supramolec-
Keywords: ligand design ´ metallo-
dendrimers ´ platinum ´ sensors ´
sulfur dioxide
fuels, the extraction of sulfide ores, or from natural sources
(e.g., volcanoes or oceanic phytoplankton).[1] In particular,
sulfur dioxide (SO2) is thought to contribute significantly to
the production of smog and/or acid rain.[2] In the presence of
hydrogen peroxide (H2O2) or (sun-) light and ozone, SO2
rapidly oxidizes to SO3, which spontaneously reacts with
water to form sulfuric acid. Therefore, the detection (with a
suitable sensor[3]), concentration and subsequent removal of
even minute levels of SO2 is evidently desirable.[4]
Introduction
A major environmental concern is the emission of sulfur-con-
taining compounds during, for example, the combustion of fossil
[a] Prof. G. van Koten, M. Albrecht, Dr. R. A. Gossage
Debye Institute, Department of Metal-Mediated Synthesis
Utrecht University, Padualaan 8
3584 CH Utrecht (The Netherlands)
Fax : (‡ 31)30-2523615
E-mail: g.vankoten@chem.uu.nl
Since the synthesis of the first transition metal complexes of
sulfur dioxide,[5] SO2 has been shown to be a versatile ligand in
inorganic and organometallic chemistry, as it can act as both a
Lewis base or acid.[6] Various binding modes of sulfur dioxide
to a metal center have been realized, making this ligand a very
useful probe of the electronic character of a metal center. The
bonding principles of sulfur dioxide to transition metals has
been subject of various investigations.[7]
[b] Dr. M. Lutz, Dr. A. L. Spek[‡]
Bijvoet Center of Biomolecular Research
Department of Crystal and Structural Chemistry
Utrecht University, Padualaan 8
3584 CH Utrecht (The Netherlands)
Fax : (‡ 31)30-2533940
E-mail: a.l.spek@chem.uu.nl
‡
[ ] Address correspondence pertaining to crystallographic studies to this
author.
With transition metals of a formal d6 or d8 electronic
configuration, the bonding of sulfur dioxide mainly originates
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1431 $ 17.50+.50/0
1431

G. van Koten et al.
FULL PAPER
from an interaction of the dz2 orbital of the metal with the non-
bonding 4a1 orbital of SO2. This overlap is constructive in the
case of d6 metal centers and, consequently, transition-metal-
SO2 complexes have been found in which the metal ± sulfur
vector is coplanar with the sulfur dioxide molecule (planar
binding mode).[8] With d8 metal centers, however, this primary
interaction is unfavorable, since both the metal dz2 and the
ligand 4a1 orbitals are filled. The resulting repulsion becomes
less important upon bending of the SO2 molecule, and a new
interaction involving both the HOMO and LUMO (2b1) of
sulfur dioxide with either the filled metal dxz or dyz orbital is
possible. This theory is supported by the isolation of SO2
complexes with a number of d8 transition metals (NiII, RhI, IrI)
that contain sulfur dioxide in a bent binding mode (i.e., the
metal ± sulfur vector crosses the plane formed by the sulfur
dioxide molecule by an angle between 1808 and 90o).[9]
(CH2NMe2)2-C6H3] (i.e., NCN) ªpincerº ligand reversibly
bind SO2 gas.[10] Reversibility is generally not observed in
transition-metal-SO2 adducts,[7e, 11] but is an inherent prereq-
uisite for the design of carrier and sensor materials. Further-
more, a characteristic color change of the platinum complexes
has been observed upon coordination to SO2. These facts,
namely reversibility and macroscopic change of the materials
properties, suggest that these compounds may be useful as
diagnostic materials for SO2 detection.[12, 13] Modification of
the coordination sphere of the metal center by tuning the
ligand properties can be applied as a methodology to optimize
the (optical) response of these materials to SO2. Fine-tuning
of the system predominantly occurs through change of the
electrophilicity of the metal center either by the introduction
of electron-withdrawing or -releasing para-substituents on the
aryl ring or by changing the basicity of the nitrogen donor
atoms. In addition, this latter method enables the adjustment
of the steric bulk in close proximity to the metal center, thus
strongly influencing the accessibility of this atom to coordi-
nation.[14] On the other hand, fragments that have only a
marginal effect on the binding sensitivity of the metal center
to sulfur dioxide offer potential anchoring points for the
attachment of the sensor to (in)organic supports. Platinum
complexes containing a modified pincer ligand have been
fixed onto dendritic molecules and these organometallic
materials represent the first ªdendrimer sensorsº for toxic
SO2 gas.[12, 15]
It has been previously shown that platinum complexes
containing the terdentate-coordinating monoanionic [2,6-
Abstract in Dutch: De synthese wordt beschreven van een
reeks vlak-vierkante platina(ii)-complexen, die alle een mo-
noanionisch drietandig-gecoördineerd tangligand bevatten,
[PtX(4-E-2,6-{CH2NRR'}2-C6H2)] (X ˆ Cl, Br, I, tolyl; R,
R' ˆ Et, Me; E ˆ H, OH, OSiMe2tBu). Deze complexen
reageren spontaan met zwaveldioxide onder vorming van
adducten met een vijf-gecoördineerd platina-centrum. Uit
kristalstructuuranalyse van twee van deze adducten, 8c (X ˆ
I, R ˆ R' ˆ Me, E ˆ OSiMe2tBu) en 11 (X ˆ Cl, R ˆ R' ˆ Me,
E ˆ OH), blijkt dat SO2 de top-positie van een vierkante
pyramide bezet. Adduct 11 vormt waterstofbruggen door
middel van Pt Cl ´´´ H O interacties; dit zelfde bindings-
patroon wordt ook in het SO2-vrije complex 5 gevonden. Dit
behoud van supramoleculaire informatie (self-assemblage
door waterstofbruggen) voor en na een reactie is uniek en
geeft interessante mogelijkheden voor het bewerken van
kristallijn materiaal (crystal engineering). De selectieve binding
van SO2 aan de platina-complexen vindt zowel in oplossing als
in de vaste toestand plaats en is snel en reversibel. Bij de
adsorptie van SO2 treedt een karakteristieke kleurverandering
op van kleurloos naar oranje. Zelfs zeer lage concentraties SO2
(tot millimolaire hoeveelheden) kunnen door detectie van deze
kleurverandering specifiek worden aangetoond. Milde metho-
den voor de verwijdering van SO2 en het regenereren van de
platina-complexen zijn ontwikkeld. Hiermee wordt aan de
belangrijkste eisen voor het gebruik van deze platina-com-
Herein, we report on the fundamental factors that influence
the formation of sulfur dioxide adducts with platinum
complexes containing the NCN ªpincerº skeleton (Fig-
ure 1).[16] Particularly intriguing is the response of the sensor
capacity of these platinum complexes to various ligand
Figure 1. Potential sites for tuning the steric and electronic impact of the
N,C,N terdentate-coordinating ªpincerº ligand; this offers
a suitable
strategy to modify the physico-chemical properties of the metal center.
modifications (i.e., E, R, R' and X, see Figure 1). As a result of
these studies, we present arylplatinum complexes with
optimized affinity of the metal center towards sulfur dioxide.
Appropriate immobilization of these diagnostic sites allows
the development of the first metallo-dendrimers[17] for
application as fully reversible sensor materials for the
selective and quantitative detection of low levels of SO2 gas.
È
plexen als efficiente sensormaterialen voor de detectie van SO2
voldaan. De detectiegrens kan worden beïnvloed door het
veranderen van de electronische en/of sterische eigenschappen
van het tangligand (ligand tuning). Het gebruik van dendri-
meren als dragermateriaal van deze SO2 detectoren wordt
beschreven. Aangezien metallodendritische macromoleculen
gemakkelijk af te scheiden en terug te winnen zijn is herhaald
gebruik van deze sensormaterialen voor zowel de kwalitatieve
als kwantitatieve detectie van SO2 mogelijk. Het principe van
Results and Discussion
È
het gebruik van dendrimeren als efficiente gas-sensoren wordt
Synthesis: The new arylplatinum(ii) complexes 2b and 2c
were prepared following the synthetic protocol previously
established for the corresponding analogues 1 (Scheme 1).[18]
geïllustreerd met de synthese van de metallo-dendrimeren 3 en
15 en hun SO2 adducten.
1432
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1432 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
Me2
N
ylic acid chloride affords the
hexaplatinated dendrimer 15
(first-generation dendrimer).
This method is similar to the
one applied for the preparation
of the trimetallic complex 3
Cl
Pt
NRR'
Me2N
vi, vii
2a
E
O
O
NMe2
NRR'
O
O
viii
Cl
O
O
Pt
4-E-RR'-H-NCN
(zero-generation
dendri-
NMe2
mer).[22] Note that all difficul-
ties associated with periphery-
functionalization and end-
group determination of the
dendritic materials have ele-
gantly been encompassed by
using a convergent approach
starting from organometallic
platinum(ii) building blocks.
The identity of both the orga-
noplatinum dendron 14 and the
dendrimer 15 was unambigu-
ously confirmed by elemental
analysis (14), mass spectrome-
try (FAB-MS for 14, MALDI-
TOF for 15), and 1H and
13C{1H} NMR spectroscopy.
i, ii
NMe2
HO
Pt Cl
NMe2
3
SO2
NRR'
9
Me2N
Cl
Pt
E
Pt Cl
NRR'
5
N
SO2
Me2
11
iii, iv
or v
complex
E
R, R'
SO2 adduct
X
1a
1b
1c
H
H
H
Me, Me
Me, Me
Me, Me
Cl
Br
I
Cl
Br
I
7a
NRR'
Pt
7b
7c
E
X
2a OSiMe2tBu Me, Me
2b OSiMe2tBu Me, Me
2c OSiMe2tBu Me, Me
8a
8b
8c
NRR'
4a
4b
6
H
H
H
Et, Et
Me, Et
I
Cl
10a
10b
[PtX(4-E-RR'-NCN)]
Me, Me C6H4Me-4 12
Scheme 1. The synthesis of the arylplatinum(ii) complexes 1 ± 6. i) n-BuLi, pentane; ii) [PtCl2(SEt2)2], Et2O;
iii) [AgSO3CF3], acetone; iv) NaX (X ˆ Cl, a; Br, b; I, c), H2O; v) Li-C6H4-Me-4, Et2O; vi) Bu4NF, CH2Cl2; vii)
1,3,5-(COCl)3-C6H3, CH2Cl2; viii) Bu4NF, THF/H2O.
Adduct formation with sulfur
dioxide: The exposure of the
square-planar PtII complexes
1 ± 6 to an atmosphere of SO2 leads to the formation of the
corresponding five-coordinate SO2 adducts 7 ± 12. Platinum(ii)
centers in square-planar complexes that contain the mono-
anionic pincer ligand demonstrate an enhanced nucleophilic-
ity when compared with other d8 metal centers.[23] Conse-
quently, they react with Lewis acids (e.g., SO2) to form
adducts. These reactions occur if the NCN-Pt complexes are in
the solid state or dissolved in an appropriate solvent. Product
formation is instantaneous on the laboratory timescale and is
accompanied by an immediate color change from colorless to
bright orange. When stored in an atmosphere devoid of excess
sulfur dioxide, all these adducts slowly loose SO2 with
regeneration of the corresponding starting materials over a
period of several weeks. Neither oxidation of the SO2 ligand
to sulfate nor insertion products have been found, except for
complex 6 (vide infra). Changing back to an SO2-rich
environment leads again to quantitative adduct formation.
These facts unambiguously demonstrate that the adsorption
reaction is an equilibrium process between free and bound
SO2 complexes [Eq. (1)].
Halide abstraction in [PtCl(4-OSiMe2tBu-2,6-{CH2NMe2}2-
C6H2)] (2a, [PtCl(4-OSiMe2tBu-Me,Me-NCN)]) with silver
trifluoromethane sulfonate (AgSO3CF3) in wet acetone yields
the cationic aqua-complex [Pt(4-OSiMe2tBu-Me,Me-NCN)-
(OH2)][SO3CF3]. Nucleophilic substitution of the aqua ligand
by a halide anion affords the corresponding neutral complexes
[PtX(4-OSiMe2tBu-Me,Me-NCN)] 2b and 2c (X ˆ Br or I,
respectively) in high yields. An asymmetric PtII complex 4b,
[PtCl(4-H-Et,Me-NCN)] was also prepared. The nitrogen
centers in 4b are stereogenic provided that Pt N coordination
occurs. Lithiation of the diaminoaryl ligand precursor 1,3-
(CH2NEtMe)2-C6H4 afforded the 2-lithio derivative,[16b] which
was subsequently transmetallated with [PtCl2(SEt2)2] to give
4b as a pale-yellow air-stable solid. This compound exists as a
mixture of rac and meso forms (ratio 2:3). Attempts to
separate the diastereomers by column chromatography or
repeated recrystallization were not successful. Prolonged
heating of a toluene solution of 4b (1008C) does not change
the meso:rac isomer distribution. This suggests either that
racemization is prevented by a large kinetic barrier of the
process involving reversible Pt N bond dissociation or that
the initially observed diastereomeric ratio represents the
thermodynamic distribution.
[PtX(NCN)] ‡ n(SO2) > [PtX(NCN)SO2)n]
(1)
This equilibrium is shifted towards the four-coordinate SO2-
free complexes, if the sulfur dioxide atmosphere is removed.
Desorption of SO2 from the adducts 7 ± 10 is accelerated by
exposure to air, to solvents free of SO2, or by reducing the
atmospheric pressure, which again quantitatively regenerates
complexes 1 ± 4, respectively. Heating can also be used to
remove SO2 from these platinum complexes. An increase in
temperature (to 408C) leads to the formation of the SO2-free
starting materials within a few minutes. For this latter method
The hexametallic arylester dendrimer 15 was prepared by
using a convergent synthetic approach.[19] Coupling of two
equivalents of 5 (as the peripheral end group) with the oxo-
protected bis(acyl chloride)phenol 13 (as AB2 branching
unit)[21] yieldsÐby double esterificationÐthe dimetallated Pt2
building block 14 (Scheme 2). Deprotection of the phenolic
hydroxy functionality of the platinum dimer 14 and in situ
esterification with the trifurcate core 1,3,5-benzenetricarbox-
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1433 $ 17.50+.50/0
1433

G. van Koten et al.
FULL PAPER
Finally, adsorption of SO2 greatly enhances the solubility
characteristics of all the platinum complexes studied here.
This fact is most prominently demonstrated by the solubility
behavior of 5, which contains a polar organometallic site and a
phenolic hydroxy group. As a four-coordinate complex, it is
virtually insoluble in most common organic and chlorinated
solvents. For example, in benzene, the solubility is lower than
5
O
O
Cl
Cl
NEt3 / CH2Cl2
OSiMe2tBu
13
Me2
Me2
N
N
1
0.2 gL . In SO2-saturated solution, however, dissolved con-
Cl
Pt
Cl
centrations of adduct 11 higher than 20 gL 1 may be obtained.
This increased solubility becomes particularly important for
macromolecular diagnostic materials that contain several gas-
adsorption-active centers. Whereas most of the monomeric
complexes are soluble in THF, the tri- and hexa-organo-
platinum complexes 3 and 15 fail to dissolve in this solvent
and only solubilize as the corresponding sulfur dioxide
adducts 9 and 16, respectively. This is probably due to the
change of the coordination geometry of the platinum center
from square planar to square pyramidal; this changes the
lattice energy of the solid material and reduces the impact of
the phenyl factor.[24]
Pt
O
O
O
Me2N
NMe2
O
OSiMe2tBu 14
O
Cl
NEt3 / CH2Cl2
Cl
O
O
Cl
O
O
Me2
N
Me2
N
Cl
Pt
Cl
Pt
NMe2
O
O
O
O
The reactivity of related PtII complexes such as [PtCl(2-
{CHN(Me)CH2CH2NMe2}-C6H4],[25] (nitrogen donors in cis
positions) and the ionic complexes [Pt(4-E-Me,Me-
NCN)(OH2)][SO3CF3] (with E ˆ H, OH, OSiMe2tBu) with
SO2 were also briefly examined. However, neither NMR nor
UV-visible spectroscopy indicate any change comparable with
that found for the Pt SO2 adsorption products described
above. This suggests that no adduct formation takes place.
The reason for this lack of reactivity may be the result of a
reduced nucleophilicity of these platinum centers relative to
that of the neutral sensor sites in complexes 1 ± 6 and 15. It is
another indication of the special binding properties of the
Me2N
O
Me2
N
Me2
N
O
Cl
Cl
Pt
Pt
O
Me2N
NMe2
O
O
O
O
O
O
O
O
O
15
Me2N Pt NMe2
Cl
Me2N Pt NMe2
Cl
‡
[M X] entity held in the h3-mer NCN ligand manifold (cf.
Scheme 2. Synthetic procedure for the convergent preparation of metallo-
the role of [Ni X] unit in the [NiX(NCN)]-catalyzed atom-
transfer reaction of CCl4 with methyl methacrylate).[26]
dendrimer 15.
to be effective and quantitative, the Pt complexes must be in
solution. In the solid state, thermogravimetric analyses have
shown that they do not loose sulfur dioxide at these temper-
atures. With solid complex 7c, for example, slow desorption
starts at 758C, which becomes faster with increasing temper-
ature. At 1508C, no additional weight is lost; this indicates
complete removal of SO2. This is confirmed by a reduction of
the initial mass to 88%, which clearly corresponds to the
reversion of 7c to 1c (theoretical weight loss: 89%).
Various chemical methods can be used to regenerate the
PtII complexes 1 ± 4 from the SO2 adducts (7 ± 10). These
include the addition of a tertiary amine (e.g., NEt3), a basic
aqueous solution (e.g., 1n NaOH), or a halide source, such as
Bu4NI, to a solution containing the sulfur dioxide adducts.
The SO2 adduct of the 4-hydroxy platinum complex 5 shows
deviating properties with respect to these regeneration
methods. Several of the procedures for removing sulfur
dioxide developed for the other adducts have been found to
be much slower (evaporation in vacuo requires hours rather
than minutes to be complete) or ineffective (refluxing a
solution of [PtCl(4-OH-Me,Me-NCN)(SO2)] 11 in THF for
several hours did not lead to decolorization and reformation
of [PtCl(4-OH-Me,Me-NCN)] 5).
Spectroscopic properties: The characteristic orange color
found for all the SO2 adducts studied here originates from two
Pt !S metal-to-ligand charge-transfer bands that are located
in the UV-visible absorbance region. The observed maxima
for the adducts 7 ± 12 all lie at approximately 350 nm with a
broad shoulder (except complexes 10) at longer wavelength
that is primarily dependent on the metal-bound halide
(Table 1). The series of SO2 complexes that contain the
unsubstituted NCN ligand (complexes 7a ± c) strictly obeys
the spectroscopic series upon change of the Pt-bound halide,
that is, the absorption maximum shifts towards higher wave-
length with increasing electron release of the halide (I > Br>
Cl). Surprisingly, this tendency is not seen in the series of
complexes that contain the siloxy group (i.e., OSiMe2tBu) as
the substituent in the position para to the metal-bound carbon
atom (8a ± c). The bromide complex 8b has a remarkably high
energy gap for the corresponding electronic transition with a
maximum at 333 nm, whereas the maxima of 8a (X ˆ Cl) and
8c (X ˆ I) were found at 350 nm and 368 nm, respectively. A
similar halide dependence is found in the series of complexes
free of SO2 (2a ± c); in contrast to the chloride and iodide
species, the absorption spectrum of the bromide complex 2b
differs considerably from its unsubstituted analogue 1b.
1434
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1434 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
Table 1. UV-visible[a] and relevant IR[b] spectroscopic data for complexes 1 ± 12.
1
Complex
UV-visible
IR nÄ [cm
]
Ref.
lmax [nm] (e)
nÄs(SO2)
nÄas(SO2)
[PtX(4-E-R,R'-NCN)]
1a
1b
1c
2a
2b
2c
3
4a
4b
5
R ˆ R' ˆ Me, E ˆ H, X ˆ Cl
279 (9.7)
281 (9.8)
285 (12.8)
278 (14.2)
259 (16.5)
283 (16.5)
276 (26.5)
277 (12.2)
277 (7.2)
277 (10.9)
338 (1.0)
[18b]
[18b]
[18b]
[22]
R ˆ R' ˆ Me, E ˆ H, X ˆ Br
R ˆ R' ˆ Me, E ˆ H, X ˆ I
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Br
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ I
[C6H3-{C(O)-O-NCN-PtCl}3]
R ˆ R' ˆ Et, E ˆ H, X ˆ I
[22]
R ˆ Me, R' ˆ Et, E ˆ H, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OH, X ˆ Cl[c]
R ˆ R' ˆ Me, E ˆ H, X ˆ 4-tolyl
6
[PtX(4-E-R,R'-NCN)(SO2)]
7a
7b
7c
8a
8b
8c
R ˆ R' ˆ Me, E ˆ H, X ˆ Cl
352 (6.5), 408 (1.7)
356 (6.9), 410 (1.9)
364 (6.8), 410 (1.4)
350 (7.8), 421 (2.3)
333 (9.3), 395 (2.4)
368 (9.4), 422 (2.2)
351 (14.8), 402 (4.3)
368 (sh; 1.5)
1076
1076
1076
1074
1074
1074
1075 (sh)
1067 (sh)
1078
1238
1230
1226
1226
1215
1215
1226
1219
1234
1236
1215
R ˆ R' ˆ Me, E ˆ H, X ˆ Br
R ˆ R' ˆ Me, E ˆ H, X ˆ I
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Br
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ I
[C6H3-{C(O)-O-NCN-PtCl}3(SO2)3]
R ˆ R' ˆ Et, E ˆ H, X ˆ I
9
10a
10b
11
R ˆ Me, R' ˆ Et, E ˆ H, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OH, X ˆ Cl[d]
R ˆ R' ˆ Me, E ˆ H, X ˆ 4-tolyl
353 (6.3), 401 (1.5)
350 (8.7), 419 (2.0)
357 (4.1), 410 (1.3)
1072
1063
12
1
[a] From CH2Cl2 solution, e [103 m 1cm 1]. [b] In CHCl3. [c] n(O H) 3304 cm 1; [d] n(O H) 3306 cm
.
Time-resolved UV-visible spectroscopic measurements
(stopped flow; benzene solution) indicate that the reaction
of arylplatinum complexes with SO2 is complete within 50 ms.
Decreasing the mixing temperature from ambient temper-
ature to 78C did not significantly slow down the process and,
hence, no definite mechanistic conclusions on the adsorption
reaction (pre-equilibria, site of initial attack) can be drawn
from these measurements.
Infrared spectroscopic investigations have concentrated on
the assignment of the symmetric and asymmetric stretching
vibrations of the bound SO2 molecule (Table 1). In all the
adducts the signals assigned to these vibrations were found
between 1238 and 1214 cm 1 (ns) and 1080 and 1064 cm 1 (nas)
(measured in CHCl3 saturated with SO2). These regions are
characteristic for sulfur dioxide bound to d8 metal centers.[7b]
However, complexes 10a and 10b, which contain ethyl
substituents on the metal-bound amines, do not clearly exhibit
an absorbance at lower wavenumber, probably because of the
overlap with the very strong nas band from free sulfur
dioxide.[7c] Furthermore, the values obtained for complexes
7a ± c in this study corroborate well with the frequencies
found earlier from Nujol measurements of the solid materi-
als.[10]
The NMR characteristics of the square-planar complexes
1 ± 6 have been discussed earlier.[10, 14] For the SO2 adducts 7 ±
12, comparison of the NMR resonance data with those of the
corresponding starting materials 1 ± 6 is illustrative. The
1H NMR spectra of the SO2 adducts 7 ± 9, 11, and 12, reveal
a characteristic down-field shift of the resonance signals of the
NCH3 and CH2N protons (Dd ˆ 0.14 and 0.23 ppm, respec-
tively, in SO2 saturated solutions of CDCl3, with Dd ˆ
d(adduct) d(free complex), Table 2) when compared with the
corresponding SO2 free complexes.[27] The same chemical
shifts of the two benzylic CH2N protons and of both the
nitrogen-bound methyl groups of the (NCN) ligand moiety in
7 have been demonstrated.[10] In an environment which is
subsaturated with sulfur dioxide, one single set of signals is
observed which is positioned between that found for the SO2-
free complex and the adduct.[28] Similar conclusions can be
drawn from the analyses of the 13C{1H} NMR spectroscopic
data (Table 3); adduct formation is pronounced by a notable
shift of the nitrogen-bound methyl carbons towards higher
field, which indicates an enhanced shielding of these nuclei.
The position of the relevant signals in the 1H or 13C{1H} NMR
spectrum is directly dependent on the total concentration of
sulfur dioxide and can therefore be used to indicate and
quantify the presence of dissolved SO2 (Figure 2). Remark-
ably, the trifurcate organoplatinum complex 9 shows one set
of signals only, regardless of the concentration of SO2. This
unambiguously indicates that all three PtII centers are equally
coordinating to SO2.
A particular NMR behavior has been found for the
platinum compounds 4a and 4b with prochiral CH2 groupings
at the nitrogen-bound carbons because of the presence of
ethyl- instead of methyl-amino substituents.[29] In the case of
the diethylamino compound 4a, the multiplets that result
from the NCH2Me protons at d ˆ 3.69 and 2.92 (ABX3
pattern) undergo a shift of 0.06 ppm (high field) and
0.01 ppm (down field), respectively, upon addition of SO2.
Also the benzylic protons do not show a characteristic change
as observed for the adducts 7 ± 9 and, hence, the resonance
signals of 10a appear at virtually the same position as found
for SO2-free 4a. This indicates that coordination of SO2 to 4a
is, if it occurrs at all, very weak. A different situation accounts
for 10b and 4b; from the benzylic protons, which are split into
two sets of AB doublets (assigned to the rac and meso isomers,
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1435 $ 17.50+.50/0
1435

G. van Koten et al.
FULL PAPER
Table 2. 1H NMR data for complexes 1 ± 12.[a]
Complex
Caryl-H
ArCH2N (3J(H,Pt))
NCH3 (3J(H,Pt))
Other H
Ref.
[22]
[PtX(4-E-R,R'-NCN)]
1a
1b
1c
2a
2b
2c
3
R ˆ R' ˆ Me, E ˆ H, X ˆ Cl
R ˆ R' ˆ Me, E ˆ H, X ˆ Br
R ˆ R' ˆ Me, E ˆ H, X ˆ I
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Br
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ I
[C6H3-{C(O)-O-NCN-PtCl}3]
R ˆ R' ˆ Et, E ˆ H, X ˆ I
6.99 (t), 6.80 (d)
7.03 (t), 6.80 (d)
7.03 (t), 6.83 (d)
6.35
6.33
6.73
4.02 (46)
4.02 (46)
4.03 (46)
3.96 (46)
3.96 (45)
3.96 (45)
4.05 (41)
4.06 (37)
3.07 (38)
3.11 (38)
3.18 (39)
3.07 (38)
3.11 (38)
3.16 (39)
3.11 (36)
3.69,[b,c] 2.92 (22)[c]
0.96, 0.16
0.96, 0.16
0.96, 0.16
9.15
6.73
[22]
[29a]
4a
4b
6.97 (t), 6.70 (d)
1.52
R ˆ Me, R' ˆ Et, E ˆ H, X ˆ Cl
major isomer
minor isomer
6.96 (t), 6.78 (d)
6.96 (t), 6.78 (d)
6.39
6.9
4.24 (29), 3.81 (56)
4.21 (38), 3.83 (53)
3.98 (45)
3.13 (39), 3.33 ± 2.80 (m)[b,c]
3.11 (39), 3.33 2.80 (m)[b,c]
3.07 (30)
1.47
1.50
5
6
R ˆ R' ˆ Me, E ˆ OH, X ˆ Cl
R ˆ R' ˆ Me, E ˆ H, X ˆ 4-tolyl
4.16 (43)
2.95 (43)
7.64, 7.02, 2.33
[18b]
[PtX(4-E-R,R'-NCN)(SO2)]
7a
7b
7c
8a
8b
8c
9
10a
10b
R ˆ R' ˆ Me, E ˆ H, X ˆ Cl
7.08 (t), 6.88 (d)
7.08 (t), 6.88 (d)
7.11 (t), 6.87 (d)
6.43
6.42
6.41
4.25 (41)
4.25 (41)
4.25 (42)
4.19 (41)
4.18 (42)
4.20 (41)
4.27 (41)
4.06 (37)
3.22 (32)
3.26 (33)
3.33 (36)
3.22 (32)
3.25 (33)
3.32 (35)
3.24 (31)
3.63,[b,c] 2.93 (21)[c]
R ˆ R' ˆ Me, E ˆ H, X ˆ Br
R ˆ R' ˆ Me, E ˆ H, X ˆ I
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Br
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ I
C6H3-(C(O)-O-NCN-PtCl)3(SO2)3
R ˆ R' ˆ Et, E ˆ H, X ˆ I
0.98, 0.18
0.97, 0.18
0.97, 0.18
9.13
6.84
6.96 (t), 6.69 (d)
1.50
R ˆ Me, R' ˆ Et, E ˆ H, X ˆ Cl
major isomer
6.99 (t), 6.79 (d)
6.98 (t), 6.79 (d)
6.44
4.24 (17), 4.01 (25)
4.17,[b] 4.06 (18)
4.21 (41)
3.14 (36), 3.48,[b,c] 3.15[b,c]
3.13 (36), 3.34,[b,c] 3.08[b,c]
3.22 (31)
1.33
1.38
minor isomer
11
12
R ˆ R' ˆ Me, E ˆ OH, X ˆ Cl
R ˆ R' ˆ Me, E ˆ H, X ˆ 4-tolyl
7.05 (t), 6.89 (d)
4.41 (37)
3.34 (36)
7.70, 7.12, 2.38
[a] Singlet signals from CDCl3 solutions unless stated otherwise; doublet (d), triplett (t), multiplett (m); d (J [Hz]). [b] 3J(Pt,H) not resolved/observed.
3
[c] NCH2Me; ABX3 pattern of NCHAHBMe; 2J(H,H) ˆ 12.3 Hz, J(H,H) ˆ 7.2 Hz.
Table 3. 13C NMR data for complexes 1 ± 12.[a]
Complex
Cipso
Cortho
Cmeta
Cpara
ArCH2N
R, R'
Other C
Ref.
[PtX(4-E-R,R'-NCN)]
1a
1b
1c
2a
2b
2c
3
R ˆ R' ˆ Me, E ˆ H, X ˆ Cl
144.5
145.7
149.6
152.6
152.6
152.6
147.3
146.2
142.7
142.7
143.4
143.8
143.7
144.0
144.1
145.4
122.6
122.8
123.4
136.2
137.5
141.1
143.1
122.8
118.6
118.7
119.2
111.4
111.3
111.3
112.8
117.4
77.1
76.8
76.9
77.7
77.4
76.9
77.6
69.0
53.7
54.5
56.3
54.5
55.0
56.4
54.5
61.7[c]
[18b]
[18b]
R ˆ R' ˆ Me, E ˆ H, X ˆ Br
R ˆ R' ˆ Me, E ˆ H, X ˆ I
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Br
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ I
C6H3-(C(O)-O-NCN-PtCl)3
R ˆ R' ˆ Et, E ˆ H, X ˆ I
R ˆ Me, R' ˆ Et, E ˆ H, X ˆ Cl
major isomer
25.7, 18.1, 4.4
25.6, 18.0, 4.5
25.5, 18.0, 4.5
163.7, 135.8, 131.3
13.5
[22]
[22]
[29a]
4a
4b
144.1
144.2
154.3
172.0
143.9
143.8
143.4
145.1
122.8
122.8
133.1
122.4
118.6
118.7
106.7
118.1
73.3
73.8
77.3
81.1
59.4,[c] 53.7
59.7,[c] 53.5
54.0
12.9
13.0
minor isomer
[b]
5
6
R ˆ R' ˆ Me, E ˆ OH, X ˆ Cl
[20]
[18b]
R ˆ R' ˆ Me, E ˆ H, X ˆ 4-tolyl
54.9
179.2, 138.9, 129.6,
127.4, 21.0
[PtX(4-E-R,R'-NCN)(SO2)]
7a
7b
7c
8a
8b
8c
9
10a
10b
R ˆ R' ˆ Me, E ˆ H, X ˆ Cl
147.4
150.1
151.9
154.3
154.3
154.3
148.7
146.3
142.4
142.5
142.7
142.9
142.9
143.4
144.7
145.0
125.4
125.6
125.5
138.7
140.3
141.4
143.4
123.0
120.5
120.6
120.4
112.6
112.5
112.4
114.0
117.6
75.9
75.6
75.1
75.7
75.5
75.2
75.9
68.0
53.8
54.4
55.4
53.5
54.3
55.5
54.0
61.7[c]
R ˆ R' ˆ Me, E ˆ H, X ˆ Br
R ˆ R' ˆ Me, E ˆ H, X ˆ I
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Cl
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ Br
R ˆ R' ˆ Me, E ˆ OSiMe2tBu, X ˆ I
[C6H3-{C(O)-O-NCN-PtCl}3(SO2)3]
R ˆ R' ˆ Et, E ˆ H, X ˆ I
25.5, 18.0, 4.5
25.5, 18.0, 4.5
25.6, 18.1, 4.5
163.4, 135.9, 131.2
13.4
R ˆ Me, R' ˆ Et, E ˆ H, X ˆ Cl
major isomer
145.4
145.5
143.5
143.4
143.0
142.8
124.2
124.1
138.0
124.9
119.3
119.2
108.3
120.5
72.4
72.8
75.7
75.4
59.5,[c] 51.7
59.9,[c] 51.8
53.9
12.0
12.5
minor isomer
11
12
R ˆ R' ˆ Me, E ˆ OH, X ˆ Cl
154.7
[d]
[d]
R ˆ R' ˆ Me, E ˆ H, X ˆ 4-tolyl
55.0
138.3, 129.0
126.2, 22.0
[a] CDCl3 solutions unless stated otherwise. [b] In CDCl3/[D6]DMSO. [c] NCH2Me. [d] Cipso not observed.
1436
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1436 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
position with respect to the aryl ring. Hence, the puckering
diminishes and strong coordination of SO2 is hampered. This
conclusion is corroborated by the observation that crystal-
lization from SO2-saturated solutions afforded 4a instead of
10a (vide infra). In 4b, however, axial orientation of the
methyl groups is possible in both the rac and the meso isomers.
In the latter case, two diastereotopic adducts may form in
princple, namely those with either axial or equatorial methyl
groups. However, only one adduct, meso-10b, is observed
spectroscopically (along with rac-10b), which contains the
methyl groupings in pseudo-axial and the ethyl substituents in
equatorial orientation (see 13C NMR data, Table 3). Such a
site-selective bonding of SO2 is expected to occur, when steric
repulsion of large substituents in co-axial positions is domi-
nant.
Figure 2. Diagnostic characteristics of the ligand resonance signals in
1H NMR spectroscopy. The chemical-shift dependence on the concen-
tration of SO2 is shown for the benzylic (top, circles) and methyl (bottom,
triangles) protons of a mixture of 2c and 8c.
When sulfur dioxide is added to the tolyl-complex [Pt(4-
C6H4Me)(NCN-Me,Me)] 6, a set of signals appears, which
points to the formation of a SO2 adduct 12. Besides 12,
another compound is formed gradually that is characterized
by two new AB doublets in the aromatic region (d ˆ 7.93 and
respectively), the resonances at lower frequency are shifted
characteristically upon addition of SO2, whereas the signals
around d ˆ 4.2 are unaffected. A similar behavior is found for
the NCH2Me protons. Additional evidence for sulfur dioxide
bonding and a concomitant change of the geometry around
the metal center in 10b is gained from consideration of the
coupling constants, which decrease substantially when SO2 is
present in solution. These results indicate that in contrast to
4a (which contains two NEt2 groupings), coordination of SO2
to 4b (which bears two NEtMe groupings) occurs for both
isomers. Apparently, one methyl group on the nitrogen donor
already gives space enough for coordination, while two larger
substituents (e.g., NEt2) prevent bonding to SO2, unless their
steric impact is reduced (as, for example, in N(CH2)4).[9e] This
effect is best understood by considering the puckering of the
two five-membered metallacycles Pt-N-C-C-Cipso with the aryl
plane of the NCN ligand in the square-planar PtX(NCN)
complexes (CS symmetry).[14, 29a] Owing to this puckering, the
two alkyl substituents on each nitrogen are positioned
pseudo-axially and pseudo-equatorially, and the two axial
substituents are anti-planar to each other (see Newman
projection in Scheme 3). Coordination of SO2 forces one
metallacycle to invert this relative axial-equatorial orientation
of the substituents (Cv symmetry). Such a structure is
minimized when only little repulsion of the two syn-planar,
co-axially positioned substituents occurs, that is, when at least
one of these substituents is a CH3 group.[14] In the case of 4a,
no such configuration is possible and, as a consequence of the
steric repulsion of the mutually co-axial ethyl groups, the
nitrogen atom is continuously forced to adopt a coplanar
1
7.25) of the H NMR spectrum, together with shifted signals
assigned to the protons of the NCN ligand. In contrast to the
AB patterns for 6 and 12, these new signals, which are due to
the para-tolyl ligand, do not contain characteristic 195Pt ± 1H
couplings. This suggests the formation of a product that results
from the irreversible insertion of SO2 into the metal-carbon
bond.[30] No attempts have been made to isolate and further
identify this product.
Based on these NMR investigations, it becomes evident
that for all [PtX(NCN)] complexes with X ˆ Cl, Br, and I,
adduct formation with SO2 occurs, provided that the ligating
nitrogen atoms contain at least one methyl substituent. Ethyl
substituents are already bulky enough to prevent significant
coordination of SO2 to the metal center. Moreover, complexes
with X ˆ p-tolyl form adducts that bind SO2 as strong as the
corresponding halide complexes, since the chemical shift
differences of the diagnostic protons are in a similar range.
This implies that steric repulsion of the tolyl ligand with
respect to the SO2 molecule is negligible and the oxygen
atoms of SO2 are probably directed towards the NCN-aryl p-
system.
Solid-state studies: In order to unambiguously establish the
nature of 11 and the SO2 complex of 4a, crystal structure
determinations were carried out for complex 11 and attempt-
ed for 10a. Generally, crystals were grown by slow diffusion of
pentane into a SO2-saturated CH2Cl2 solution of the arylpla-
tinum species; this afforded orange needles of 11 and colorless
blocks of 4a suitable for X-ray analysis. For comparison, the
structures of 8c[12] and 7b[10] will also be discussed.
The molecular structure of 8c is shown in Figure 3a (for
selected bond lengths and angles, see Table 4). The sulfur
dioxide molecule is coordinated through S in a h1-bonding
mode to the platinum center. Platinum is in a pseudo-square-
pyramidal environment with SO2 in the apical position. The
angles to the four basal atoms are between 91.4(2)8 and
98.1(1)8. No disorder in the position of SO2 has been detected
that could rise from partial coordination to iodide. This
selective bonding to Pt is remarkable with respect to the
SO2
N
+ SO2
Pt
Pt
C6
N
C6
N
C2
C2
N
- SO2
steric interference
Scheme 3. Newman projection along the Pt Cipso axis, illustrated at the
equilibrium of meso-4b and meso-10b; because of the repulsion of the axial
substituents on nitrogen in the adduct (syn-orientation), only the product is
formed that contains the bulky substituents in equatorial positions.
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1437 $ 17.50+.50/0
1437

G. van Koten et al.
FULL PAPER
Table 4. Selected bond lengths [Š] and angles [8] for 4a and 5,[a] and the
SO2 complexes 8c[b] and 11.
4a (X ˆ I)
8c
5
11
X ˆ Cl
molecule 1 molecule 2 X ˆ I
X ˆ Cl
bond lengths
Pt C1
Pt X
Pt N1
Pt N2
Pt S
1.929(7)
2.7328(6)
2.089(5)
2.089(5)
±
±
±
1.928(6)
2.7300(6)
2.095(5)
2.103(4)
±
±
±
1.946(6)
2.718(1)
2.112(4)
2.113(5)
2.479(2)
1.443(4)
1.433(6)
1.915(9)
2.434(2)
2.094(8)
2.082(8)
±
±
±
1.923(10)
2.423(3)
2.106(8)
2.096(8)
2.531(3)
1.444(9)
1.430(9)
S O1
S O2
bond angles
N1-Pt-N2 165.37(17) 164.7(2)
156.7(2)
163.9(3)
160.5(3)
C1-Pt-X
C1-Pt-N1
C1-Pt-N2
X-Pt-N1
X-Pt-N2
X-Pt-S
C1-Pt-S
N1-Pt-S
N2-Pt-S
O1-S-Pt
O2-S-Pt
O1-S-O2
C4-O-E
180.00(8)
82.68(12)
82.68(12)
97.32(12)
178.15(13) 174.4(2)
177.4(4)
81.6(4)
82.4(4)
173.1(3)
82.3(4)
81.5(4)
97.2(2)
97.7(2)
96.90(11)
90.0(3)
93.2(3)
97.6(2)
105.6(4)
103.6(4)
113.4(6)
109.5
82.1(2)
82.6(2)
81.2(2)
81.5(2)
98.54(16)
96.70(12)
98.15(13) 97.9(2)
97.58(15) 98.2(2)
97.32(12)
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
94.14(4)
91.44(18)
98.1(1)
±
±
±
±
±
±
±
97.8(2)
103.3(2)
105.9(2)
113.6(3)
132.5(4)
(E ˆ Si)
113(9)
(E ˆ H)
(E ˆ H)
[a] From ref. [20]. [b] From ref. [12].
complex 7b the orientation of the sulfur dioxide ligand is less
predictable (the vector of one S O bond being nearly
collinear with that of the Pt Br bond). This may rise from a
more prominent p-donor± acceptor relationship in complex
8c, induced by the halide and the oxo substituent on the aryl
ligand. Probably this is also the reason for the shorter metal-
to-sulfur bond length in 8c (2.479(2) Š) relative to this bond
in 7b (2.613(7) Š). An energy profile of the rotation of the
SO2 ligand around the Pt S bond axis (analysis of the steric
factors by CAChe techniques[33]) does not exceed a maximal
1
difference of 6.0 kJmol . This emphasizes the importance of
Figure 3. Perspective diagrams (50% probability) and numbering schemes
of the structures of the adducts a) 8c [PtI(4-OSiMe2tBu-Me,Me-
NCN)(SO2)] and b) 11 [PtCl(4-OH-Me,Me-NCN)(SO2)] and of the SO2-
free complexes c) 5 [PtCl(4-OH-Me,Me-NCN)] and d) 4a [PtI(Et,Et-
NCN)]. Hydrogen atoms and the second independent molecule in the unit
cell of 4a are omitted for clarity.
electronic arguments and crystal-packing effects in the orien-
tation of the sulfur dioxide molecule in the apical position, as
demonstrated by the crystal structures of 8c and 7b.
The molecular structure of the adduct [PtCl(4-OH-Me,Me-
NCN)(SO2)] 11 displays similar features as described for the
4-OSiMe2tBu complex 8c (Figure 3b; selected bond lengths
and angles in Table 4). Compared with its SO2-free 4-OH
analogue 5 (Figure 3c),[20] only slight differences in bond
lengths and angles are found. Remarkably, even the hydro-
gen-bond motif,[34] which interconnects the single molecules to
linear chains through strong Pt Cl ´´´ H O bonds, is con-
served in the sulfur dioxide adduct 11 (see Table 5). An
expected competitive behavior of iodide in nucleophilicity
and size.[31] Related structures of organometallic PtII-iodide
complexes are known that contain the SO2 molecule coordi-
nated to iodide such as in [Pt(Me)(I-SO2)L2] (L ˆ PPh3).[32]
Nevertheless, the predictions of the geometry for complexes
of the type [ML{SO2}][8] derived from extended Hückel
calculations are clearly demonstrated.[7a] These include
i) selective h1 coordination of sulfur dioxide (note that both
S O bonds are statistically equal in length within experimen-
tal error), ii) a pyramidal geometry of the SO2 plane with
respect to the Pt S axis (q of the plane defined by SO2 with
the Pt S axis is approximately 1168), and iii) bending of the
oxygen atoms away from iodide (p donor) and towards the
aryl unit of the NCN ligand (p acceptor). This last point is
evident in the structure of 8c, whereas in the bromide
Table 5. Hydrogen-bond characteristics for compound 5 and the corre-
sponding SO2 complex 11.
O H
H ´´´ Cl
O ´´´ Cl
O H ´´´ Cl
5[a]
11[b]
0.84(13)
0.84
2.32(13)
2.31
3.127(8)
3.126(8)
161(15)
163.9
[a] The hydrogen atom was located on the difference Fourier map and
further refined (from ref. [20]). [b] Located on the difference Fourier map
and refined with a rotating model.
1438
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1438 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
intriguing feature of the crystal structure of 11 is the
considerably short distance between the sulfur nucleus and
Thermodynamics: To get further information concerning the
nature and the stability of the Pt SO2 binding in the
complexes 7 ± 11 in solution, UV-visible spectrophotometric
measurements at varying concentrations of SO2 were carried
out (Figure 5). The overlapping spectra do not show an
isosbestic point, probably because of the coinciding strong
absorbance of free SO2 (relevant below 320 nm), which
entirely covers the absorption bands of the SO2-free com-
plexes 1 ± 5. Increasing the concentration of sulfur dioxide
leads to a higher absorbance at lmax (Figure 5).
the metal-bound chloride of
a neighboring molecule
(3.253(4) Š).[35] This gives rise to a b-type network with
Pt ´´´ S ´´´ Cl interactions that are perpendicular to the (for-
mer) a-type network induced by Pt Cl ´´´ H O hydrogen
bonding. The result is a two-dimensional supramolecular
assembly as depicted in Figure 4, due to molecular recognition
Figure 4. The crystal structure of 11 [PtCl(4-OH-Me,Me-NCN)(SO2)].
The view along the y axis shows the supramolecular self-assembly of 11 in
the solid state by directional Pt Cl ´´´ H O hydrogen bonding and super-
imposed Pt ´´´ S ´´´ Cl interactions, which results in the formation of a b-type
network.
Figure 5. The response of the complex 2a on changes in the concentration
of SO2, monitored by UV-visible spectroscopy. The spectra correspond to
2a (dashed line) and to mixtures of 2a and 8a at 40, 80, 160, 300, 540,
1000mm, and saturated solutions of SO2. Inset: Variation of the extinction
at 350 nm.
(hydrogen-bonded network) and substrate-mediated (SO2)
interlocking of these chains. Materials such as 11 are
particularly attractive for crystal engineering, since they
combine the directional properties of the a-type interaction
and the self-organization through reversibly formed b-type
interactions.
Equilibrium constants Kf of the equilibrium between the
SO2 adducts 7 ± 11 and their four-coordinate starting materials
were determined by titration of solutions of complexes 1 ± 5,
respectively, with SO2 (dissolved in benzene). The constants
Kf were evaluated by recording Ac (absorbance lmax of the
adduct at a SO2 concentration c, relative to A1, the
absorbance in SO2-saturated benzene solution) versus the
concentration of sulfur dioxide. These plots are shown
in Figure 6 for the reactions of sulfur dioxide with
[PtI(4-OSiMe2tBu-Me,Me-NCN)] 2c (as representative for
Attempts to crystallize 10b failed. In all cases, crystals of 4a
were obtained exclusively. However, the structural details of
the molecular geometry of 4a in the solid state gives valuable
information concerning the weak tendency of 4a to form SO2
complexes. Two inequivalent molecules of 4a were found in
the crystal lattice; one is C2v symmetric (4a-C2) and the
second is asymmetric (4a-C1). The molecular structure (Fig-
ure 3d) is characterized by a platinum(ii) metal center in a
slightly distorted square-planar geometry, bound to the h3-
mer-coordinating NCN ligand and an iodide ion. Bond lengths
and angles (Table 4) are closely related to the values found for
similar complexes.[9e, 18b, 36] The puckering of the five-mem-
bered metallacycles with the plane of the aryl ring, charac-
terized by the torsion angle of N1-Pt-C1-C2, does not exceed
4.3(3)8. This is significantly smaller than the values found
typically in the corresponding NMe2 complexes (around 108).
This increased coplanarity of the aryl plane and the square
plane around the platinum center probably originates from
steric repulsion of the four ethyl substituents and results in a
flattening of the five-membered chelate rings, that is, in a less
pronounced axial/equatorial orientation of these ethyl groups.
Consequently, replacment of both nitrogen-bound methyl by
ethyl groups gives rise to an increased hindrance of the
potential coordination sites at the platinum center.[9e] Note
that the NMR studies already indicated that 4a did not show
any notable interaction with SO2 in solution (vide supra).
Figure 6. Titration curves for various mono- (2c, circles; 4b, triangles) and
multimetallic sensor materials (3, bars) against SO2.
the monometallic NMe2 complexes 1, 2, and 5), the dendritic
Pt3 species 3, and [PtI(4-H-Et,Me-NCN)] 4b. The molarity in
SO2 was adjusted to be always at least ten times higher than
the actual concentration of the platinum complexes.[37] This
allows us to neglect the change in concentration of SO2 due to
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1439 $ 17.50+.50/0
1439

G. van Koten et al.
FULL PAPER
adduct formation. Hence, from Equation (2), Equation (3)
can be deduced and was applied to fit the obtained data
points.
steric rather than the electronic arguments in these platinu-
m(ii) complexes has arisen from closely related work that
includes I2 as Lewis acid.[39] Data from the X-ray structure
determination of platinum complexes with pincer derivatives
similar to 4b, but with NtBuMe instead of NEtMe donors,
confirmed that the metal center is still accessible by I2.
Adducts that display a square-pyramidal geometry were
isolated only when the methyl groups were in pseudo anti-
conformation with respect to the iodine ligand (pseudo-axial
position). A different isomer containing both the tBu groups
in the pseudo anti-position (methyl groups equatorial) is not
stable, probably owing to steric repulsion of the large amine
substituents. Hence, the meso-conformer of 4b is supposed to
coordinate SO2 selectively on one site, thus forming a product
with axially positioned methyl groups, whereas in rac-4b, no
such site-discrimination is expected. This model is further
confirmed by comparison of the spectroscopic data of 10a and
10b (vide supra).
‰PtXꢀNCN†ꢀSO2† Š
n
Kf ˆ
(2)
n
‰PtXꢀNCN†Š‰SO2Š
n
‰PtXꢀNCN†ꢀSO2† Š
Kf ‰SO2Š
n
c
ˆ
(3)
n
‰PtXꢀNCN†SO2† Š
Kf ‰SO2Š ‡ 1
n
1
The values for the formation constants Kf and the
corresponding Gibbs free energies DG8 are summarized in
Table 6. For the reactions starting with monoplatinum species
[PtX(4-H-Me,Me-NCN)] 1 or the corresponding 4-substitut-
Table 6. Equilibrium constants Kf and DG8.[a]
1
1
Kf [m
]
DG8 [kJmol ]
1a
1b
1c
2a
2b
2c
3
10.42 (0.42)
9.97 (0.93)
9.80 (1.36)
14.81 (0.35)
8.48 (0.34)
9.51 (1.53)
9.90 (1.26)
2.16 (0.09)
9.70 (0.63)
5.71 (0.10)
5.60 (0.22)
5.56 (0.32)
6.57 (0.06)
5.21 (0.10)
5.49 (0.36)
5.59 (0.29)
1.88 (0.10)
5.54 (0.15)
Sensor capacity of the square-planar PtII complexes: The
platinum compounds 1, 2, 4b, and 5 and the dendritic Pt3
complex 3 are all very sensitive towards the presence of low
concentrations of sulfur dioxide. Sulfur dioxide coordination
is fast (less than 50 ms at room temperature) and can be
monitored by various analytical techniques like UV-visible or
NMR spectroscopy. No pretreatment of the sensor materials
is necessary to activate the platinum complexes. Their
selectivity for sulfur dioxide is illustrated by the fact that no
evidence for the formation of either adducts with oxygen or
with other gases present in the atmosphere (e.g., N2, CO,
CO2) has been found. Consequently, these platinum-based
sensor materials do not need particular protection (e.g., from
humidity or air) or pre-activation. None of the frequently
observed reactions occur that follow sulfur dioxide coordina-
tion on organometallic complexes [e.g., insertion of SO2 into a
metal-carbon (NCN-aryl) bond,[30] oxidation of SO2 to sulfate
in the presence of air or oxygen[40]]. Evident exceptions are
complexes that contain unstabilized metal ± carbon bonds,
like the Pt Ctolyl bond in intermediate complex 12, in which
slow insertion of sulfur dioxide and formation of Pt-S(O)2-
Ctolyl bonds most probably takes place. A similar insertion into
the metal ± carbon bond of the Pt(NCN) unit is inhibited,
owing to the h3-mer-N,C,N chelation of the NCN ligand that
stabilizes the platinum ± carbon bond to a large extent.[16]
A variety of facile and non-destructive methods have been
elaborated to recycle the sensor materials, including chemical
(e.g., competitive sulfur dioxide transfer to halide ions[41] or
amines[42]) and physical (e.g., heating, pressure reduction)
removal of sulfur dioxide. This broad choice of procedures
underlines the suitability of these compounds as sensor
devices for repeated and efficient use in various environ-
mental situations.
4b
5
[a] Esds in parantheses; for platinum complex 4a: see text.
ed complexes 2 or 5, these data are entirely consistent with the
formation of a 1:1 adduct with SO2 in solution and, hence, are
in agreement with the Dd data obtained by 1H NMR
spectroscopy. Despite an earlier hypothesis,[10] no indication
for the formation of an octahedral product containing two
coordinated molecules of SO2 has been found. Neither the
substituent H or OR on the aryl ligand nor the metal-bound
halide seem to affect the equilibrium constant substantially,
and, thus, the values found for Kf are all in the same range.
The equilibrium between the trimetallic species 3 and 9 is of
a different nature and is best described by a 1:3 complex
between 3 and SO2. The similarity of the titration curves of
the monometallic and trimetallic sensors suggests that in 3,
the diagnostic platinum centers are virtually independent
from each other and behave similar to monomers.[38] This
emphasizes the applicability of dendritic materials as useful
sensor materials.
Compared with the complexes that contain the pincer
ligand with two methyl substituents on the amine donor, for
example, complexes 1 or 2, a significantly lower equilibrium
constant has been found for complex 4b, which contains a
bulkier substituent on the amine ligand. The titration
measurements still point to a 1:1 adduct with SO2, whereas
in the corresponding NEt2 complex 4a a linear relationship
between the spectroscopic absorbance and the concentration
of SO2 has been found; this points to a very small equilibrium
constant. These observations are remarkable in terms of
electronic factors, since the basicity of the nitrogen donor is
increased upon changing the substituents in the order
Me < Et(<iPr). As a consequence, the nucleophilicity on
the metal center and, therefore, its affinity to Lewis acids
(e.g., SO2) is increased. Evidence for the predomination of the
Even traces of sulfur dioxide gas are sufficient to induce the
characteristic color change. By using a solution of 2a in
CH2Cl2 (40mm), it is possible to detect SO2 concentrations as
low as 200 mm by means of UV-visible spectroscopy; this
results in a Pt/SO2 ratio of 1:5. In contrast to these
observations, the threshold value for detecting SO2 has been
found to be higher when the sensors are used in the solid state.
1440
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1440 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
A cellulose surface coated with dendritic macromolecule 3 in
tions.[44] Most importantly, the presence of well-defined
connectivity patterns and exact number of functional (sensor)
sites in dendrimers allows qualitative and quantitative sub-
strate detection without preceding calibration. This is partic-
ularly desirable for solution measurements and contrasts with
heterogenous and polymeric sensors, for which the loading of
the sensor units must be determined for each batch.
By using this principle,[15] the nanosize dendrimer 15, which
contains six peripheral sensor units, has been tested in the
reaction with sulfur dioxide (Scheme 4).[45] When this metallo-
dendrimer 15 is exposed to an atmosphere of SO2, an
2
a density of 20 nmolmm is enough for the detection of an
1 [12]
SO2 concentration of 8.5(Æ0.5) mgL .
Interestingly, the
threshold values reported here have been determined by
routine measurements with common laboratory equipment.
This implies that the sensitivity of these sensors must be
considerably higher and that threshold limits can be lowered
when signal detection is performed on dedicated instruments.
Moreover, improvement of the sensor capacity may be
achieved by means of signal (light) amplification techni-
ques.[43]
The full reversibility of the SO2 adsorption is demonstrated
by time-resolved photospectroscopic monitoring of a solution
of 2c in dependence of the gaseous environment. Saturation
of a solution of 2c in CH2Cl2 (0.1mm) with SO2 results in full
conversion of the complex into its SO2 adduct 8c, which is
visualized by maximal absorbance at 368 nm. Continuous
addition of air to this solution (slowly) regenerates the four-
coordinate starting complex 2c, which does not absorb at this
wavelength. Repetition of this adsorption-desorption cycle by
saturating the solution with sulfur dioxide and air, respec-
tively, does not cause a decrease of the observed extinction
coefficients, even after several cycles. Figure 7 shows these
spectral changes of the detector sites at a fixed wavelength
(lmax of the corresponding adduct 8c, E ˆ OSiMe2tBu, X ˆ I)
as a diagnostic response on variation of the atmospheric
constitution. These results further provide evidence that there
is no loss of SO2 binding activity, a result which demonstrates
the resistance of the Pt-NCN unit towards the applied
sensoring conditions. Moreover, the fast response of the
detector device to the presence of SO2 is emphasized by an
instantaneous signal transmission.
SO2
Cl
NMe2
Pt
SO2
Pt
Me2N
Cl
Me2N
NMe2
O
O
Me2
N
O
O
SO2
O
O
O
O
Pt Cl
NMe2
+ SO2
- SO2
O
O
15
O
O
O
O
Me2
N
O
O
Me2N
O2S
SO2
Cl
Pt
N
Pt
Cl
Me2
O
O
NMe2
Me2N
O2S
Pt
16
N
Cl
Me2
Scheme 4. The activity of multimetallic sensors towards sulfur dioxide
adsorption.
Another major characteristic of the sensor materials
presented here is their activity towards sulfur dioxide both
as solid materials or in liquid (in solution or as suspension).
instantaneous color change is likewise observed and spectro-
scopic analyses are consistent with the formation of the hexa-
adduct 16. The shift difference of the characteristic protons in
the 1H NMR spectrum (Dd ˆ 0.12 and 4.20 ppm for the NCH3
and the benzylic CH2N protons, respectively) and the
extinction coefficient of 16 reveal that all platinum centers
in the macromolecule are active on SO2 binding. Reformation
of the SO2-free material is easily achieved by any of the
desorption methods described above. The shape-persistence
of the core[46] of these nanosize metallo-dendrimers makes
them good candidates for suc-
Dendrimer-supported sensor devices: Complete recovery of
the diagnostic materials requires a suitable support for the
active sites of the sensor. Dendritic supports have an out-
standing potential for such purposes, since they combine all
advantages of the monomeric complexes with those typically
found for heterogeneous or polymeric materials (e.g., easy
recovery). Nanofiltration techniques have been successfully
used to separate dendritic materials from (reaction) solu-
cessful recovery from solutions
by nanofiltration.
Conclusion
Multimetallic metallo-dendri-
mers and monometallic organo-
platinum complexes that con-
tain bis-chelating diaminoaryl
pincer ligands are efficient sen-
sor materials for the detection
of minute levels of sulfur diox-
ide. Any presence of this gas is
Figure 7. Change of the absorbance of a solution of 2c at a fixed wavelength (368 nm, lmax of 8c) as a direct
response on the change of the atmosphere. Repetitive cycles do not decrease the activity of the sensor materials.
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1441 $ 17.50+.50/0
1441

G. van Koten et al.
FULL PAPER
(3 Â 50 mL) and dried in vacuo for 4 h, to afford 17.6 g of a yellow solid
(99%). Spectroscopic analyses (IR, 13C{1H} NMR) indicated a mixture of
trans- and cis-[PtCl2(SEt2)2] and were consistent with literature values.[48]
indicated by various analytical techniques, most prominently
by a characteristic color change of the initially colorless sensor
devices to orange. Adsorption of SO2 is fully reversible and is
fast on the NMR and UV-visible timescale, as the equilibrium
of the adsorption reaction is reached within less than 50 ms at
temperatures above 58C. Further characteristics of these
sensor devices comprise facile quantification of the amount of
SO2 present and excellent chemical resistance properties (e.g.,
towards acids).The diagnostic platinum sites demonstrate
outstanding selectivity for SO2-gas binding. Sensoring of these
organometallic bases (the [PtX(NCN)] systems) occurs by a
highly specific molecular recognition process of a single acid
(SO2), even in the presence of other (stronger) acids such as
HCl, CO2, or H2O.
Sensor-active metallo-dendrimers, detector materials which
can be recycled easily and quantitatively, are accessible when
detection-insensitive sites of the sensor device are coupled to
a dendritic framework. Suitable ligand functionalization has
been used as the methodology to anchor organoplatinum sites
to dendritic cores. This afforded multimetallic sensor materi-
als that are highly efficient and recoverable as exemplified by
the metallo-dendrimers 3 and 15. The well-defined structures
of these macromolecules makes them useful for repetitive
qualitative and quantitative diagnostic gas sensor applications
Moreover, a new synthon for organometallic crystal
engineering has been developed. In the solid state, the SO2
molecule in complex 11, which contains a metal-bound
chloride, binds to the metal center in a classical bonding
mode (i.e., through h1-Pt S bond formation) and, in addition,
is stabilized by bridging coordination of sulfur to a metal-
bound chloride nucleus of a different molecule, hence
inducing b-type network formation of 11. This novel supra-
molecular synthon has potential application for directing the
self-assembly of organometallic building blocks in the solid
state.
[PtBr(4-OSiMe2tBu-2,6-{CH2NMe2}2-C6H2] (2b): A modified literature
procedure was used.[18b] AgCF3SO3 (76 mg, 0.30 mmol) was added to a
solution of 2a (165 mg, 0.30 mmol) in acetone (5 mL) and water (0.2 mL).
After stirring for 0.5 h under strict exclusion of light, the suspension was
filtered through a short pad of Celite. All volatiles were removed in vacuo,
and the residue was taken up in water (8 mL). Solid NaBr (87 mg,
0.85 mmol) was added to this suspension and stirring maintained for 0.5 h.
The precipitate was collected by filtration, washed with Et2O (4 mL) and
extracted with small portions of CH2Cl2 (12 mL total). The combined
organic layers were then dried in vacuo, leaving a white solid. Yield: 156 mg
(87%); elemental analysis calcd (%) for C18H33BrN2OPtSi (596.6): C 36.24,
H 5.58, N 4.70; found C 36.39, H 5.51, N 4.78.
[PtI(4-OSiMe2tBu-2,6-{CH2NMe2}2-C6H2] (2c): This compound was pre-
pared in a similar way as described for 2b by using NaI (141 mg,
0.94 mmol). Yield: 176 mg (91%); elemental analysis calcd (%) for
C18H33IN2OPtSi (643.6): C 33.59, H 5.17, N 4.35, I 19.72; found C 33.65,
H 5.08, N 4.26, I 19.59.
1,3-(CH2NEtMe)2-C6H4: HNEtMe (5.00 g, 83 mmol) was added to a,a'-
dibromo-m-xylene (5.28 g, 20 mmol) dissolved in Et2O (30 mL) at 08C. The
suspension was stirred for 3 h at 08C and 12 h at ambient temperature, then
poured into NaOH (1m, 100 mL). Extraction with pentane (3 Â 100 mL)
and subsequent drying of the combined organic fractions over K2CO3
yielded, after solvent evaporation in vacuo, a colorless oil, which was
purified by bulb-to-bulb distillation. Yield: 3.74 g (84%); 1H NMR: d ˆ
7.23 (t, 3J(H,H) ˆ 7 Hz, 1H; ArH), 7.22 (s, 1H; ArH), 7.19 (d, 3J(H,H) ˆ
7 Hz, 2H; ArH), 3.47 (s, 4H; ArCH2N), 2.44 (q, 3J(H,H) ˆ 7.2 Hz, 4H;
NCH2Me), 2.18 (s, 6H; NCH3), 1.09 (t, 3J(H,H) ˆ 7.2 Hz, 6H; NCH2CH3);
13C{1H} NMR (75 MHz): d ˆ 139.0, 129.8, 128.0, 127,7 (all Ar), 61.9
(ArCH2N), 51.2 (NCH2Me), 41.7 (NCH3), 12.4 (NCH2CH3); MS (70 eV,
‡
‡
EI): m/z (%): 220 (9) [M] , 163 (100) [M NEtMe] ; elemental analysis
calcd (%) for C14H24N2 (220.4): C 76.31, H 10.98, N 12.71; found C 76.17, H
11.09, N 12.63.
[PtCl(2,6-{CH2NEtMe}2-C6H3)] (4b): A solution of n-BuLi (1.6 mL, 1.66m
in hexane, 2.7 mmol) was added dropwise to
a solution of 1,3-
(CH2NEtMe)2-C6H4 (0.59 g, 2.7 mmol) in pentane (8 mL) at 788C. The
reaction mixture was allowed to warm to room temperature overnight, and
all volatiles were removed in vacuo. The residue was redissolved in Et2O
(10 mL), added to a suspension of [PtCl2(SEt2)2] (1.20 g, 2.7 mmol) in Et2O
(30 mL) and stirred for 16 h. All volatiles were then removed in vacuo, and
the residue was washed with pentane (3 Â 30 mL) and extracted with
CH2Cl2 (3 Â 30 mL). Concentration of the combined chlorinated fractions
to 5 mL followed by the addition of pentane (70 mL) afforded a colorless
precipitate, which was isolated by centrifugation and dried in vacuo. Yield:
0.79 g (66%); elemental analysis calcd (%) for C14H23N2ClPt (449.9): C
37.38, H 5.15, N 6.23; found C 37.20, H 5.09, N 6.12.
Experimental Section
General comments: All reactions were performed by using standard
Schlenk techniques unless stated otherwise. Sulfur dioxide, PtCl2, SEt2,
Bu4NF (1 M in THF), HNEtMe, and a,a'-dibromo-m-xylene were obtained
commercially and used without further purification; 1,3,5-benzenetricar-
boxylic acid chloride was recrystallized from hot hexane prior to use.
Benzene, pentane, and Et2O were distilled from Na/benzophenone, and
CH2Cl2 from CaH2. Elemental analyses were performed by Kolbe,
Mikroanalytisches Laboratorium (Mülheim, Germany). 1H and 13C{1H}
NMR spectra were recorded on a Bruker AC300 or a Varian Inova 300
spectrometer. All spectra were recorded in CDCl3 solution at 258C, unless
otherwise stated, and referenced to external TMS (d ˆ 0.00 ppm, J in Hz;
see Tables 2 and 3). Spectroscopic studies (UV-visible) were carried out on
a Varian Cary 1 spectrophotometer with degassed benzene or CH2Cl2 as
solvent and standard quartz cuvettes (Table 1). Infrared measurements
were performed on a Perkin ± Elmer FT-IR spectrometer with CH2Cl2 as
solvent. Thermogravimetric analyses were carried out in platinum contain-
ers (under a continuous flow of nitrogen) on a Perkin ± Elmer TGS-2
apparatus using a heating rate of 58Cmin 1. The PtII complexes 1,[18b] 2a,[22]
3,[22] 4a,[29a] 5,[20] and 6[18b] and the AB2 branching unit 13[21] were prepared as
described.
Synthesis of 14 (see Scheme 2): NEt3 (5 mL, excess) was added dropwise to
a suspension of 5 (1.32 g, 3.0 mmol) and 13 (0.50 g, 1.5 mmol) in CH2Cl2
(30 mL) over a period of 30 minutes. The reaction mixture slowly became
clear and was stirred for additional 24 h before adding H2O (20 mL). The
aqueous phase was separated and extracted with CH2Cl2 (2 Â 30 mL). The
combined organic layers were dried (Na2SO4) and evaporated to dryness.
The crude product was purified by column chromatography (SiO2; CH2Cl2/
acetone 5:1) to afford 14 as an off-white solid (1.05 g, 62%). 1H NMR: d ˆ
3
8.53 (t, 3J(H,H) ˆ 1.4 Hz, 1H; ArH), 7.83 (d, J(H,H) ˆ 1.4 Hz, 2H; ArH),
3
6.69 (s, 4H; ArH), 4.03 (s, 8H; J(Pt,H) ˆ 38 Hz, ArCH2N), 3.09 (s, 24H;
3J(Pt,H) broad, not resolved, NCH3), 1.01 (s, 9H; SiCCH3), 0.26 (s, 6H;
13
1
ˆ
SiCH3); C{ H} NMR (75 MHz): d ˆ 164.4 (C O), 156.2 (C O), 147.5,
144.0, 142.6 (C Pt), 131.6, 126.2 (C H), 124.4 (C H), 112.9(C H) (all
ArC), 77.4 (CH2N), 54.4 (NCH3), 25.6 (SiCCH3), 18.2 (SiCMe3), 4.4
(SiCH3); FAB-MS: m/z: found 1135.24 (calcd 1135.04); elemental analysis
calcd (%) for C38H53Cl2N4O5Pt2Si (1135.0): C 40.21, H 4.79, N 4.94; found C
40.37, H 4.97, N 4.85.
[Pt(Cl2)(SEt2)2]: According to a modified literature procedure,[47] SEt2
(40 mL, 440 mmol) was poured to a suspension of PtCl2 (10.6 g, 39.9 mmol)
in benzene (100 mL). Within 0.5 h, all solids had dissolved. The solution
was stirred for an additional 0.5 h and subsequently filtered through Celite
and evaporated to dryness. The yellow solid was washed with cold pentane
Synthesis of 15 (see Scheme 2): Bu4NF (0.75 mL, 1m in THF, 0.75 mmol)
was added to a solution of 14 (0.85 g, 0.75 mmol) in CH2Cl2 (10 mL). After
stirring for 30 minutes, 1,3,5-benzenetricarboxylic acid chloride (58 mg,
0.22 mmol) was added and the mixture stirred for 3 days. All volatiles were
then removed in vacuo, and the residue washed with THF (2 Â 20 mL).
1442
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1442 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
Then the residual solid was dissolved in CH2Cl2 (5 mL) and THF was added
(20 mL), which caused precipitation of a fluffy solid after 1 h. The residue
was collected and precipitated once more by the same procedure, to afford
an off-white solid (0.28 g, 39%). 1H NMR: d ˆ 9.25 (s, 3H; ArH), 8.82 (brs,
3H; ArH), 8.26 (brs, 6H; ArH), 6.71 (s, 12H; ArH), 4.04 (s, 24H; 3J(Pt,H)
Table 7. Crystallographic data for complexes 4a and 11.
4a
11
color, shape
empirical formula
formula weight
crystal system
space group
a [Š]
b [Š]
c [Š]
V [Š3]
colorless cube
C16H27IN2Pt
569.38
orthorhombic
Pbcn (No. 60)
28.3122(3)
14.2257(1)
13.7374(1)
5532.89(8)
12
orange needle
C12H19ClN2O3PtS
501.89
orthorhombic
Pna2l (No. 33)
16.837(3)
10.2189(16)
9.0231(10)
1552.5(4)
4
broad, not resolved, ArCH2N), 3.08 (s, 72H; 3J(Pt,H) broad, not resolved,
1
NCH3); 13C{ H} NMR (75 MHz): d ˆ 166.5, 166.4 (both C O), 159.5, 155.1
ˆ
(both C O), 147.5, 143.8, 141.9, 132.3, 113.1, 108.2, 106.5 (all ArC; C Pt not
observed), 77.5 (CH2N), 54.4 (NCH3); Maldi-TOF: m/z: found 3188.1 [M
Cl] ; calcd 3185.9.
‡
Preparation of the SO2 adducts (7 ± 12, 16): Bubbling SO2 (g) over or
through a CH2Cl2 solution of the corresponding platinum(ii) complex 1 ± 6
or 15 afforded an instantaneous color change from colorless to orange
(yellowish, when 4a is used). The adducts precipitate readily upon addition
of pentane or Et2O. Filtration yielded quantitatively the title compounds
7 ± 12, and 16 as orange solids (10a: yellowish). Drying of any of these SO2
adducts in vacuo is not possible, because removal of SO2 immediately takes
place, thus regenerating starting complexes 1 ± 6 and 15, respectively. The
spectroscopic characteristics of the adducts 7 ± 12 can be found in Tables 1,
2 and 3. Analytical data for 16: 1H NMR: d ˆ 9.26 (s, 3H; ArH), 8.81 (brs,
3H; ArH), 8.32 (brs, 6H; ArH), 6.79 (s, 12H; ArH), 4.24 (s, 24H; 3J(Pt,H)
broad, not resolved, ArCH2N), 3.20 (s, 72H; 3J(Pt,H) broad, not resolved,
NCH3); UV/Vis (CH2Cl2): lmax (e) ˆ 354 nm (sh, 42000m 1cm 1), 405 nm
(11000m 1cm 1).
Z
3
1calcd [gcm
m [mm 1] (MoKa
F(000)
]
2.0506(1)
9.28
3216
2.1473(6)
9.35
960
)
crystal size [mm]
0.25 Â 0.38 Â 0.50
0.03 Â 0.08 Â 0.25
1
(sinq/l)max [Š
]
0.65
0.65
reflections collected
unique reflections
Rint
40529
6341
0.086
11626,
2166
0.106
0.736 ± 1.223
182/79
0.0350
0.0558
1.054
s2(Fo2† ‡ (0.0146P)2
transmission range
parameters/restraints
0.050 ± 0.480
310/138
0.0359
[a]
R1
[b]
wR2
0.095
When a SO2 saturated solution of 10a or 11 in CH2Cl2 was exposed to
pentane, crystals were obtained that were suitable for single-crystal X-ray
structure determination. The crystals of 10a, however, were colorless and
not orange, as expected from the other sulfur dioxide adducts (11: orange
needles). Indeed, no SO2 molecules were found in the unit cell, which
contains two independent molecules. The crystals, although grown in the
presence of excess sulfur dioxide, contained exclusively compound 4a.
Goodness of fit
w
1.02
1[c]
s2(Fo2† ‡ (0.0415P)2
‡ 8.5439P
1.14/ 0.98
3
residual electron density [eŠ
]
0.97/ 1.01
[a] R1 ˆSjjFO j jFC jj/SjFO j, for all I > 4s(I). [b] wR2 ˆ[S[w(Fo2 Fc2†2]/
S[w(Fo2†2]]1/2. [c] Pˆ(max(Fo2,0)‡2Fc2†/3.
Recovery of the starting complexes 1 ± 6 or 15 in solution is achieved by
i) continuous bubbling of N2, O2, or air through a solution of 7 ± 12 or 16,
respectively, in CH2Cl2 for 0.5 h (3 h for 11); ii) heating of a solution to
408C for 0.5 h (not suitable for 11); iii) addition of 30 mol equiv NEt3 or
Bu4NX (X ˆ Cl, Br, I) to a solution containing the SO2 adducts; iv)
extraction with aqueous NaOH (1m); v) washing the adducts with hexane
or Et2O; vi) evacuation.
293 K (4a) or 150 K (11), with graphite-monochromated MoKa radiation
(l ˆ 0.71073 Š). Absorption corrections (DIFABS/PLATON[50]
) were
applied for both complexes. The structures were solved by using automated
Patterson methods and subsequent difference Fourier techniques (DIR-
DIF-92[51]). The structures were refined on F 2 using full-matrix least-
squares techniques (SHELXL-96[52]); no observance criterion was applied
during refinement. Hydrogen atoms were included in the refinement on
calculated positions, riding on their carrier atoms, except the oxygen-bound
hydrogen in 11, which was located on the difference Fourier map and
subsequently refined with a rotating model. All non-hydrogen atoms were
refined with anisotropic thermal parameters. Neutral atomic scattering
factors and anomalous dispersion corrections were taken from the Interna-
tional Tables of Crystallography. Geometrical calculations and illustrations
were performed with PLATON.[50] Crystallographic data (excluding
structure factors) for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data Centre as supple-
mentary publication nos. CCDC-133528 (4a) and CCDC-133527 (11).
Copies of the data can be obtained free of charge on application to CCDC,
12 Union Road, Cambridge CB21EZ, UK (fax: (‡44)1223-336-033;
e-mail: deposit@ccdc.cam.ac.uk).
Solution measurements: Measurements for the determination of the
equilibrium constants Kf were performed by two different methods.
Method A: By UV-visible photospectroscopy: A solution of the appro-
priate PtII complex (1 ± 6) in degassed benzene (typically 80 mM) was
titrated with a solution of SO2 of well-defined concentration (benzene).[37]
The absorption spectrum was measured by using standard quartz cuvettes
in the wavelength range of 700 to 320 nm at typically ten different
concentrations of SO2. From the absorbance at lmax the concentration of
the corresponding SO2 adduct was calculated (relative to the total
concentration of platinum complexes) and compared with the actual
concentration of sulfur dioxide. Correlation of the measured curve with an
idealized 1:1 adduct of SO2 and Pt was higher than 0.99 for all recorded
equilibria except for 4a.
Method B: By 1H NMR spectroscopy: Aliquots of gaseous SO2 (portions of
1 mL, 45 mmol, typically) were added to a well-defined amount of the
appropriate PtII complex (<0.1 mmol) dissolved in CDCl3 (0.7 mL). The
shifts of the resonance signals for the CH2N and the NCH3 groups were
recorded in relation to those of the SO2-free complexes and plotted against
the actual concentration of SO2; this resulted in curves that correspond well
with the data obtained by method A.
Acknowledgement
We thank Prof. Dr. R. van Eldik and Dr. H. C. Bajaj (kinetic experiments)
and Prof. Dr. E. J. Baerends (bonding principles) for stimulating discus-
sions. This work was supported in part (M.L. and A.L.S.) by the
Netherlands Foundation for Chemical Sciences (C.W.) with financial aid
from the Netherlands Organization for Scientific Research (N.W.O.).
Kinetic measurements were carried out by using routine stopped-flow
techniques. The very first UV/Vis spectrum, obtained 0.5 ms after mixing a
solution of 1a with a solution of SO2 (both in benzene), was identical with
the final spectrum of 7a. The reverse reaction, that is, mixing of a solution
of 7a with 1a (which has to be ca. 10 times slower, see values of Kf) did not
show any kinetic trace and only the final spectrum has been observed.
Structure determination and refinement of 4a and 11: Crystals suitable for
X-ray structure analysis were glued onto the tip of a glass fiber, and
transferred into the cold nitrogen stream of a Nonius k-CCD (4a) or an
Enraf-Nonius CAD4-T diffractometer (11). Reduced-cell calculations did
not indicate higher lattice symmetry.[49] Crystal data and details on data
collection and refinement are collected in Table 7. Data were collected at
[1] a) J. Lelieveld, J. Heintzenberg, Science 1992, 258, 117; b) R. J.
Charlson, J. E. Lovelock, M. O. Andreae, S. G. Warren, Nature 1987,
326, 655; c) K. Capaldo, J. J. Corbett, P. Kasibhatia, P. Fischbeck, S. N.
Pandis, Nature 1999, 400, 743; d) R. J. Charlson in, The Changing
Atmosphere (Eds.: F. S. Rowland, I. S. A. Isaksen), Wiley, Chichester,
1988, p. 79; e) M. O. Andreae in, The Biogeochemical Cycling of
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1443 $ 17.50+.50/0
1443

G. van Koten et al.
FULL PAPER
Sulfur and Nitrogen in the Remote Atmosphere (Eds.: J. N. Galloway,
R. J. Charlson, M. O. Andreae, H. Rohde), Reidel, Dordrecht, 1985,
p. 5; f) P. Brimblecombe, C. Hammer, H. Rohde, A. Ryaboshapko,
C. F. Boutron in, Evolution of the Global Biogeochemical Sulphur
Cycle (Eds.: P. Brimblecombe, A. Y. Lein), Wiley, NewYork, 1989,
p. 77; g) European Union, Agreement on the Reduction of Emissions
of Fuels, June 1998.
[25] I. C. M. Wehman-Ooyevaar, G. M. Kapteijn, D. M. Grove, W. J. J.
Smeets, A. L. Spek, G. van Koten, J. Chem. Soc. Dalton Trans. 1994,
703.
[26] a) R. A. Gossage, L. A. van de Kuil, G. van Koten, Acc. Chem. Res.
 Ã
Â
1998, 31, 423; b) C. Granel, P. Dubois, R. Jerome, P. Teyssie,
Macromolecules 1996, 29, 8576.
[27] Furthermore, in all adduct complexes the satellite signals due to 195Pt-
1H coupling are more clearly resolved upon addition of sulfur dioxide;
see Supporting Information on the WWW under http://www.wiley-
vch.de/home/chemistry/.
[2] a) J. R. Holum, Elements of General and Biological Chemistry, 4th ed.,
Wiley, New York, 1975, p. 188; b) I. S. A. Isaksen in, The Changing
Atmosphere (Eds.: F. S. Rowland, I. S. A.Isaksen), Wiley, Chichester,
1988, p. 141.
[3] a) J. Janata, Principles of Chemical Sensors, Plenum, New York, 1989;
b) R. M. Crooks, A. J. Ricco, Acc. Chem. Res. 1998, 31, 219; c) J.
Mitrovics, H. Ulmer, U. Weimar, W. Göpel, Acc. Chem. Res. 1998, 31,
307.
[28] Despite an earlier statement,[10] it is impossible to see both the SO2
adduct and the free complex simultaneously by two sets of NMR
signals at any concentration of sulfur dioxide.
[29] a) J. A. M. van Beek, G. van Koten, G. P. C. M. Dekker, E. Wissing,
M. C. Zoutberg, C. H. Stam, J. Organomet. Chem. 1990, 394, 659;
b) M. Oki, Pure Appl. Chem. 1989, 61, 699; c) M. Oki, M. Ohira,
Chem. Lett. 1982, 1267.
[30] a) A. Wojcicki, Acc. Chem. Res. 1971, 4, 344; b) F. Faraone, L.
Silvestro, S. Sergi, R. Pietropaolo, J. Organomet. Chem. 1972, 46, 379;
c) A. Wojcicki, Adv. Organomet. Chem. 1974, 12, 31; d) S. L. Miles,
D. L. Miles, R. Bau, T. C. Flood, J. Am. Chem. Soc. 1978, 100, 7278;
e) Y.-R. Hu, A. Wojcicki, M. Calligaris, G. Nardin, Organometallics
1987, 6, 1561.
[4] a) P. Bühlmann, E. Pretsch, E. Bakker, Chem. Rev. 1998, 98, 1593;
b) C. A. Daws, C. L. Exstrom, J. R. Sowa, K. R. Mann, Chem. Mater.
1997, 9, 363.
[5] a) K. Gleu, W. Breuel, K. Rehm, Z. Anorg. Allg. Chem. 1938, 235, 201;
b) K. Gleu, W. Breuel, K. Rehm, Z. Anorg. Allg. Chem. 1938, 235, 211;
c) K. Gleu, K. Rehm, Z. Anorg. Allg. Chem. 1936, 227, 237.
[6] a) G. J. Kubas, Acc. Chem. Res. 1994, 27, 183; b) W. A. Schenk, Angew.
Chem. 1987, 99, 101; Angew. Chem. Int. Ed. Engl. 1987, 26, 98.
[7] a) D. M. P. Mingos, Transition Met. Chem. 1978, 3, 1; b) R. R. Ryan,
G. J. Kubas, D. C. Moody, P. G. Eller, Struct. Bonding 1981, 46, 47;
c) G. J. Kubas, Inorg. Chem. 1979, 18, 182; d) J. Reinhold, M. Schüler,
T. Hoffmann, E. Wenschuh, Inorg. Chem. 1992, 31, 559; e) L. Vaska,
Acc. Chem. Res. 1968, 1, 334.
[8] see for example: a) L. H. Vogt, J. L. Katz, S. E. Wiberley, Inorg. Chem.
1965, 4, 1157; b) R. R. Ryan, G. J. Kubas, Inorg. Chem. 1978, 17,
637; c) G. J. Kubas, G. D. Jarvinen, R. R. Ryan, J. Am. Chem. Soc.
1983, 105, 1883; d) E. H. Braye, W. Hübel, Chem. Ber. 1964, 97,
1871.
[9] see for example: a) S. J. La Placa, J. A. Ibers, Inorg. Chem. 1966, 5,
405; b) K. W. Muir, J. A. Ibers, Inorg. Chem. 1969, 8, 1921; c) T. E.
Nappier, D. W. Meek, R. M. Kirchner, J. A. Ibers, J. Am. Chem. Soc.
1973, 95, 4194; d) P. G. Eller, R. R. Ryan, Inorg. Chem. 1980, 19, 142;
e) U. Schimmelpfennig, R. Zimmering, K. D. Schleinitz, R. Stösser, E.
Wenschuh, U. Baumeister, H. Hartung, Z. Anorg. Allg. Chem. 1993,
619, 1931.
[10] J. Terheijden, P. W. Mul, G. van Koten, F. Muller, H. C. Stam,
Organometallics 1986, 5, 519.
[11] a) L. Vaska, S. S. Bath, J. Am. Chem. Soc. 1966, 88, 1333; b) R. R.
Ryan, G. J. Kubas, Inorg. Chem. 1978, 17, 637.
[12] M. Albrecht, R. A. Gossage, A. L. Spek, G. van Koten, Chem.
Commun. 1998, 1003.
[13] J. H. Holtz, S. A. Asher, Nature 1997, 389, 829.
[14] J. A. M. van Beek, G. van Koten, M. J. Ramp, N. C. Coenjaarts, D. M.
Grove, K. Goubitz, M. C. Zoutberg, C. H. Stam, W. J. J. Smeets, A. L.
Spek, Inorg. Chem. 1991, 30, 3059.
[15] M. Albrecht, G. van Koten, Adv. Mater. 1999, 11, 171.
[16] a) G. van Koten, Pure Appl. Chem. 1989, 61, 1681; b) M. H. P.
Rietveld, D. M. Grove, G. van Koten, New. J. Chem. 1997, 21, 751.
[17] For other applications of metallo-dendrimers, see for example:
a) E. C. Constable, Chem. Commun. 1997, 1073; b) M. A. Hearshaw,
R. J. Moss, Chem. Commun. 1999, 1; c) G. R. Newkome, E. He, C. N.
Moorefield, Chem. Rev. 1999, 99, 1689.
[31] R. D. Shannon, C. T. Prewitt, Acta Crystallogr. Sect. B 1969, 25,
925.
[32] a) M. R. Snow, J. A. Ibers, Inorg. Chem. 1973, 12, 224; b) M. R. Snow,
J. McDonald, F. Basolo, J. A. Ibers, J. Am. Chem. Soc. 1972, 94, 2526.
[33] The SO2 molecule were rotated around the Pt S axis in intervals of 38.
The energies of the conformationally minimalized structures (mm2
level) were calculated by extended Hückel methods with Alvarez
collected parameters.
[34] a) J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995; b) C. B. Aakeroy, K. R. Seddon, Chem. Soc.
Rev. 1993, 397; c) J. C. MacDonald, G. M. Whitesides, Chem. Rev.
1994, 94, 2383; d) A. D. Burrows, C.-W. Chan, M. M. Chowdry, J. E.
McGrady, D. M. P. Mingos, Chem. Soc. Rev. 1995, 329; e) S. C.
Zimmerman, F. Zeng, D. E. C. Reichert, S. V. Kolotuchin, Science
1996, 271, 1095.
[35] A similar network mediated by bridging SO2 has been found in
inorganic coordination compounds, see: M. Y. Darensbourg, T.
Tuntulani, J. H. Reibenspies, Inorg. Chem. 1995, 34, 6287, but not in
related platinum compounds containing NCN-type ligands, see for
example, ref [10].
[36] W. J. J. Smeets, A. L. Spek, A. J. M. Duisenberg, J. A. M. van Beek, G.
van Koten, Acta Crystallogr. Sect. C 1987, 43, 463.
[37] C. L. Young in, Solubility Data Series, Vol. 12 (Ed.: C. L. Young),
Pergamon, Oxford, 1983, p. 145.
[38] In a previous report,[15] qualitatively different but incorrect data were
presented for the equilibrium of dendritic 3 with SO2.
[39] a) J. A. M. van Beek, G. van Koten, W. J. J. Smeets, A. L. Spek, J. Am.
Chem. Soc. 1986, 108, 5010; b) R. A. Gossage, A. D. Ryabov, A. L.
Spek, D. J. Stufkens, J. A. M. van Beek, R. van Eldik, G. van Koten, J.
Am. Chem. Soc. 1999, 121, 2488.
[40] a) E. Lindner, M. Haustein, R. Fawzi, M. Steimann, P. Wegner,
Organometallics 1994, 13, 5021; b) E. Lindner, M. Geprägs, K.
Gierling, R. Fawzi, M. Steimann, Inorg. Chem. 1995, 34, 6106.
[41] a) D. F. Burow, Inorg. Chem. 1972, 11, 573; b) E. J. Woodhouse, T. H.
Norris, Inorg. Chem. 1971, 10, 614; c) J. Jander, G. Tuerk, Angew.
Chem. 1963, 75, 792; Angew. Chem. 1963, 2, 548.
[18] a) D. M. Grove, G. van Koten, J. N. Louwen, J. G. Noltes, A. L. Spek,
H. J. C. Ubbels, J. Am. Chem. Soc. 1982, 104, 6609; b) J. Terheijden, G.
van Koten, F. Muller, D. M. Grove, K. Vrieze, J. Organomet. Chem.
1986, 315, 401.
[42] a) J. D. Childs, D. van der Helm, S. D. Christian, Inorg. Chem. 1975,
14, 1387; b) D. van der Helm, J. D. Childs, S. D. Christian, J. Chem.
Soc. Chem. Commun. 1969, 887.
Â
[19] C. J. Hawker, J. M. J. Frechet, J. Am. Chem. Soc. 1990, 112, 7638.
[20] P. J. Davies, N. Veldman, D. M. Grove, A. L. Spek, B. T. G. Lutz, G.
van Koten, Angew. Chem. 1996, 108, 2078; Angew. Chem. Int. Ed.
Engl. 1996, 35, 1959.
[21] T. M. Miller, E. W. Kwock and T. X. Neenan, Macromolecules 1992,
25, 3143.
[43] M. Berggren, A. Dodabalapur, R. E. Slusher, Z. Bao, Nature 1997,
389, 466.
[44] a) U. Kragl, C. Dreisbach, C. Wandrey, in Applied Homogeneous
Catalysis with Organometallic Compounds, Vol. 2 (Eds.: B. Cornilis,
W. A. Herrmann), VCH, Weinheim, 1996, p. 832.
[22] P. J. Davies, D. M. Grove, G. van Koten, Organometallics 1997, 16, 800.
[23] J. C. Muijsers, J. W. Niemantsverdriet, I. C. M. Wehman-Ooyevaar,
D. M. Grove, G. van Koten, Inorg. Chem. 1992, 31, 2655.
[24] I. Dance in, The Crystal as a Supramolecular Entity (Ed.: G. R.
Desiraju), Wiley, Chichester, 1996, p. 137.
[45] Molecular modeling (mm2 level) demonstrates a disk-like structure
for 15 with a maximal Cl Cl distance of 3.2 nm.
[46] J. S. Moore, Acc. Chem. Res. 1997, 30, 402.
[47] P. K. Byers, A. J. Canty, H. Jin, D. Kruis, B. A. Markies, J. Boersma, G.
van Koten, Inorg. Synth. 1998, 32, 162.
1444
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1444 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Metallodendritic Materials
1431±1445
[48] a) G. B. Kauffman, D. O. Cowan, Inorg. Synth. 1960, 21, 211; b) R.
Roulet, C. Barbey, Helv. Chim. Acta 1973, 56, 2179; c) P. L. Goggin,
R. J. Goodfellow, S. R. Haddock, F. J. S. Reed, J. G. Smith, K. M.
Thomas, J. Chem. Soc. Dalton Trans. 1972, 1904.
Program System. Technical Report of the Crystallography Laboratory,
University of Nijmegen (The Netherlands), 1992.
[52] G. M. Sheldrick, SHELXL-96 Program for Crystal Structure Refine-
ment, Beta test version, University of Göttingen (Germany), 1996.
[49] A. L. Spek, J. Appl. Crystallogr. 1988, 21, 578.
[50] A. L. Spek, Acta Crystallogr. Sect. A 1990, 46, C34.
[51] P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, S. Garcia-
Granda, R. O. Gould, J. M. M. Smits, C. Smykalla, The DIRDIF
Received: August 23, 1999 [F1998]
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1445 $ 17.50+.50/0
1445