Syntheses of a series of S6 thioether cages and their coordination chemistry with Ag+

Article abstract of DOI:10.1039/b606510b

The syntheses of four S₆-cages with different cavity sizes are described. In regard of potential applications for radioimmunotherapy with ¹¹¹Ag, their coordination behaviour towards Ag(I) was evaluated by studies of their reactivity

Full text of DOI:10.1039/b606510b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Syntheses of a series of S6 thioether cages and their coordination
chemistry with Ag+w
Roger Alberto,a Daniela Angst,bc Kirstin Ortner,a Ulrich Abram,d P. August Schubigerc
and Thomas A. Kaden*b
Received (in Montpellier, France) 8th May 2006, Accepted 12th January 2007
First published as an Advance Article on the web 2nd February 2007
DOI: 10.1039/b606510b
The syntheses of four S6-cages with different cavity sizes are described. In regard of potential
applications for radioimmunotherapy with 111Ag, their coordination behaviour towards Ag(I) was
evaluated by studies of their reactivity in solution and of their crystal structures. All cages reacted
differently with Ag(I). Whereas S6-cages based on 3,8,12,17,20,25-hexathiabicyclo[8.8.8]-
hexacosane (14) and on 5,9,17,21,28,31-hexathiabicyclo[11.11.11]pentatriacontane (18) showed
only external Ag(I) coordination, successive adaptation of the cage size, yielded ligand 24,
based on 4,7,13,21,24,25-hexathiabicyclo[8.8.8]hexacosane, which gave a complex with
internal-peripheral Ag(I) binding.
Whereas most Ag+ complexes exhibit a tetrahedral coordi-
Introduction
nation sphere, hexadentate chelators seem to be an alterna-
tive.10 This was shown by Schroder et al.11 who were able to
Targeted radiotherapy is a very important strategy for the
treatment of severe diseases such as various types of cancer.1
Among the different radionuclides of potential use in radio-
therapy, the radiation properties (half life time, type and
energy of decay) of the radioisotope 111Ag make it especially
interesting for applications in radiotherapy.2,3 Large amounts
of high specific activity 111Ag can be prepared by the neutron
irradiation of Pd0 and subsequent separation of 111Ag by
liquid–liquid extraction. If isotope enriched palladium is
irradiated, essentially no carrier added 111Ag can be received,
in the case of natural palladium, 111Ag contains 109Ag as
carrier.4 The applicability of 111Ag in nuclear medicine, how-
ever, relies to a large extent on the possibility to immobilize the
labile Ag+ ion by binding it to a potent ligand, so that
transmetallation in the organism cannot take place. Several
¨
encapsulate the Ag+ with a S6-thioethercrown ligand. Our
efforts to use S6-macrocycles, however, gave mainly tetra-
dentate complexes of Ag+.2 To improve this we present now
a series of S6-cages with varying cage sizes together with the
crystal structures of their Ag+ complexes. We designed a
series of macrobicyclic ligands, abbreviated in the following
as R,R0-[nmn-S6] (Scheme 1) which combine the advantages of
thioether donors with the stabilising chelate effect of cage
compounds. Related cages and their Co(II) and Co(III) coordi-
nation compounds have been presented elsewhere.12 The
synthetic methodology follows an adaptation of the cyclisa-
tion protocol, using Cs2CO3 as a base for high dilution
reactions in DMF.13
approaches have been described in the literature.5–11 So Macke
¨
et al.8 have presented a macrocycle derived from tetraaza-
tetrathiacyclen, which up to now has the highest known
stability constant for Ag+, but apparently still not high
enough to prevent transmetallation under physiological con-
ditions. We also have developed open chain and cage ligands
based on a NS3 donor set to bind Ag+, but were not able to
encapsulate the metal ion into the cages.9 The reason for it is
probably due to the fact that the NS3 cages are too small to
allow incorporation of the Ag+.
Results and discussion
Syntheses and structures
The preparation of CH3,CH3-[131-S6] (9), based on the
3,7,11,15,18,22-hexathiabicyclo[7.7.7]tricosane
has been described by Osvath and Sargeson.12 Thereby
framework,
1,1,1-tris-(para-benzylsulfonylethyl)ethane was reacted with
a Institute of Inorganic Chemistry, University of Zurich,
¨
Winterthurerstrasse 190, CH-8057 Zurich, Switzerland. E-mail:
¨
ariel@aci.unizh.ch; Fax: +41 1 635 6803
b Department of Chemistry, University of Basel, Spitalstrasse 51,
CH-4056 Basel, Switzerland. E-mail: th.kaden@unibas.ch;
Fax: +41 61 267 10 20
c Center for Radiopharmaceutical Science, Paul Scherrer Institute,
CH-5232 Villigen PSI, Switzerland
d Institute of Chemistry, Freie Universitat Berlin, Fabeckstrasse 34-36,
D-14195 Berlin, Germany
¨
w Electronic supplementary information (ESI) available: bond lengths
and angles of 9, 11, [Ag(14)(tosylate)], [Ag(24)(triflate)], and [Ag(18)
(triflate)]. See DOI: 10.1039/b606510b
Scheme 1 Basic structure of the S6-cages, abbreviated as R,R0-[nmn-S6]
(left) and as an example CH3, CH3-[131-S6] (right).
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 409–417 | 409

View Article Online
Fig. 1 Structure of 9 with the exo orientation of the lone pairs of the
thioether sulfur atoms. Details about bond lengths and angles are
contained in Table S1 of the ESIw.
then reacted (iv) with the caps 4–7 under high dilution condi-
tions and in the presence of Cs2CO3 in DMF to give the
bicyclic ligands 9–12 with yields between 10 and 22%. Com-
pound 12 was also converted into 11 by deprotection with
tetrabutylammonium fluoride. From a comparison of the
yields of the cyclisation reaction with 6 (22%) and 7 (15%)
it is obvious that protection of the alcoholic function with a
silyl group results in a lower yield as compared to the reaction
without the protecting group. At a first glance this is un-
expected since an unprotected OH group can potentially
crosslink, however, the sterical bulk of the protecting group
probably hinders the approach of the S6 building blocks and
increases the rate of crosslink, thus, the formation of side
products.
Scheme 2 Synthesis of the cages 9–12 and 14: (i) TsCl in pyridine; (ii)
tert-butyldimethyl-silyl chloride, imidazole in DMF; (iii)
HS–(CH2)3–SH, EtONa in EtOH; (iv) 4–7, Cs2CO3 in DMF; (v)
HS–(CH2)4–SH, EtONa in EtOH; (vi) 6, Cs2CO3 in DMF.
3-mercaptopropionic acid in EtOH in the presence of sodium
to give the triester, which was reduced with LiAlH4 to the triol.
The triol was then converted to the trichloride with SOCl2, which
was doubly cyclised with an excess of 1,1,1-tris(mercapto-
methyl)ethane in the presence of Cs2CO3 in DMF. The purifica-
tion of the cage was achieved through chromatography of the
corresponding Co(III) complex, whereas the free ligand could not
be isolated.
To obtain more detailed information about the conforma-
tion of the S6-cages we grew crystals and solved the structures
of 9 and 11. In the free ligand 9 all non-bonding electron pairs
of the thioether sulfurs are in exo orientation (Fig. 1). Six of
the C–C–S–C torsion angles are in the range of 691 to 951. The
conformation of the other bonds is nearly antiperiplanar.
In 11, the non-bonding electron pairs are also in the exo
conformation and five of the torsion angles are in the range of
731 to 951 (Fig. 2).
Since we are particularly interested in the free ligands, we
decided to use a different approach as shown in Scheme 2. The
synthetic strategy as outlined for cage 9 is the same for all the
cages described below. The caps 4–6 were prepared by tosyla-
tion of 1,1,1-tris(hydroxymethyl)ethane (1), tris(hydroxy-
methyl)nitromethane (2) and pentaerythrite (3) with tosyl
chloride in pyridine. It is essential to have additional function-
alities ‘‘R’’ attached to the carbon-bridge-head in the caps,
since such bifunctional ligands should be attached in a later
step to targeting molecules such as peptides or antibodies. The
tritosylates 4 and 5 could be purified by crystallization.
Compound 6 was purified by column chromatography to
separate the desired product from the di- and tetratosylated
derivatives. The tritosylate 6 was also reacted with tert-butyl-
dimethylsilyl chloride and imidazole in DMF14 to give the
hydroxy-protected product 7.
A close comparison of the two structures shows that they
are nearly superimposable. Therefore we expect that the
coordination reactions of the two ligands with Ag+ will be
The next step (iii) consisted in the alkylation of cap 4 with
1,3-propanedithiol, used in a 15-fold excess to prevent intra-
molecular dialkylation, in EtOH and in the presence of sodium
ethylate as base to give the building block 8. Compound 8 was
Fig. 2 Structure of 11 (one of two independent molecules is shown).
Details about bond lengths and angles are contained in Table S2 of the
ESIw.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
410 | New J. Chem., 2007, 31, 409–417

View Article Online
similar and that the hydroxo group will not influence the
binding of the metal ion. The structure of the free ligand
implies that metal coordination into the cavity of the S6-cages
requires substantial rearrangement of the cage conformation.
From the exo conformation the sulfur donors must change to
an endo conformation, which will require a high activation
energy. The reaction of Ag+ salts with the cage ligands 9–11
resulted in complexes from which no crystals suitable for an
X-ray diffraction analysis could be obtained. Consequently,
other physical chemical methods were used for their structural
characterization. In the ESI-MS spectrum the signal for M+
of [Ag(11)]+ can be detected implying the formation of a 1 : 1
complex in solution. The very low solubility of the Ag+
complexes with all cages 9–11 points towards the formation
of polymers in the solid state. The 1H-NMR spectrum of
[Ag(11)][BF4] in d6-DMSO is very similar to that of the free
ligand with low field shifts o0.1 ppm. This indicates exocyclic
Ag+ coordination and rapid exchange between the six sulfur
atoms taking place. Taking these analytical data all together
and considering the especially unfavourable structure of the
free ligands we conclude that the cages are either to small or
the activation energy for the conformation change is too high
to encapsulate the metal ion.
Fig. 3 Structure of the complex cation of [Ag(14)]tosylate. Selected
bond lengths and angles: Ag–S12 2.563(4), Ag–O40 2.438(8),
Ag–S3202.511(4), Ag–S27002.552(4) A; S12–Ag–O40 106.3(2), S12–
Ag–S320 122.64(13), S12–Ag–S2700 104.07(13)1 (Symmetry operations:
(0) 1ꢁx, ꢁy, ꢁz; (00) ꢁx, ꢁy, ꢁz).
More flexible cages with butylene [141-S6] instead of
propylene [131-S6] bridges should decrease the energy barrier
required for turning the exo sulfur lone pairs into endo
conformation and increase at the same time the cavity size.
Correspondingly, the synthesis of R,R0-[141-S6] 14, based on
the 3,8,12,17,20,25-hexathiabicyclo[8.8.8]hexacosane frame-
work, follows exactly the same strategy as that for 11
(Scheme 2). With 1,4-butanedithiol instead of 1,3-propene-
dithiol 13 was obtained. After cyclisation with 6, the new cage
CH3,CH2OH-[141-S6] 14 was obtained in 29% yield. The
reaction of 14 with Ag[BF4], Ag[F3CSO3] or Ag[O3SC7H7]
gave the corresponding complex [Ag(14)]+ as confirmed by
ES-MS measurements. Again the 1H-NMR spectrum of the
complex is highly symmetrical and indicates either a rapid
exchange of the metal ion outside the ligand or inclusion in the
cage. Crystals suitable for X-ray structure analysis could be
grown and the structure was elucidated. The structure of
[Ag(14)][O3SC7H7] clearly shows that Ag+ is not encapsulated
in the cavity of cage 14 as implied by the ESI-MS measure-
ments but is coordinated ‘‘exo’’ by three thioether sulfur
donors from three different cages. The fourth position of the
distorted tetrahedron is occupied by an oxygen of the tosylate
counter-ion (Fig. 3).
conformationally more flexible. Accordingly, the cage
NO2,NO2-[333-S6] 18, based on the 5,9,17,21,28,32-hexathia-
bicyclo-[11.11.11]pentatriacontane framework, was prepared
by reacting tris-(3-hydroxypropyl)nitromethane (15) with
SOCl2 to afford building block 16, which was then converted
to the corresponding trithiol 17 with an excess of 1,3-propa-
nedithiol. The cage 18 was obtained by cyclisation of 16 with
17 under high dilution condition in the presence of Cs2CO3 in
42% yield (Scheme 3).
The Ag+ complex [Ag(18)]+ was formed by reacting the
ligand with different Ag+ salts in acetonitrile, THF or metha-
nol. ESI-MS gave the mass of a 1 : 1 complex as has been
shown before with [Ag(14)]+. Since ESI-MS does only give the
molecular composition of these complexes and only X-ray
structure data will show whether Ag+ is really encapsulated in
the cage. Single crystals of [Ag(18)][O3SCF3] were grown from
acetonitrile–diethyl ether. The structure of the complex is
shown in Fig. 4.
Each ligand molecule 18 is now connected to two Ag+ ions,
inducing the formation of strings. The coordination geometry
around the Ag+ cation is distorted tetrahedral. Two coordi-
nation sites are occupied by two thioether groups from two
different chains between the bridgehead carbons of the same
ligand, the other two by one thioether group stemming from a
second ligand molecule and by the oxygen of the triflate ion.
Thus, all thioether sulfurs of one ligand are engaged in
coordination to two silver cations. Ag–S bond lengths are in
the normal range 2.48 to 2.62 A, the longest being the one to
the second ligand molecule. The angles around Ag+ are
between 95.381 and 129.81. The structure of the crystals can
be described by polymeric chains (see general discussion).
The synthesis of NO2,NO2-[323-S6] (20), based on the
5,8,16,19,26,29-hexathiabicyclo-[10.10.10]dotriacontane fra-
mework, was accomplished in analogy to 18 but using
The Ag–S bond lengths are in the expected range between
2.511 and 2.563 A and the Ag–O bond is 2.44 A. The angles
around the Ag+ ion are in the range 941 to 1271. Since the
Ag+ ion is connected by three cages and these cages again
connect to further Ag(I) cations, a polymeric structure must
result. This feature will be shown schematically and compared
to the other complexes later in the general discussion.
Since increasing the number of carbons and, thus, the bridge
lengths between two vicinal sulfur atoms, did not allow
incorporating the Ag+, the remaining structural elements
which can be varied are the methylene groups connected to
the bridgehead carbon. Introduction of ethylene or propylene
chains will again increase the cavity size and render the cage
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 409–417 | 411

View Article Online
Scheme 4 Synthesis of the cage 24: (i) PBr3–HBr; (ii) HS(CH2)2SH,
EtONa in EtOH; (iii) 22, Cs2CO3 in DMF.
species did not allow the isolation of a pure compound with
chromatographic methods but implied the presence of
molecular species. The dynamic equilibrium points towards a
low stability probably due to a too large cage. This observa-
tion implied the possibility of encapsulating Ag+ with a
smaller ligand than 20 but a larger or more flexible one
than 9. A synthetically achievable combination is an ethylene
spacers for the bridgehead carbons and between the thioether
donors. The synthesis of the underivatised cage [222-S6]
(24) 4,7,13,16,21,24-hexathiabicyclo[8.8.8]hexacosane was per-
formed according to Scheme 4 with the building block 23 and
applying the same methodological strategy as for the other
ligands.
Scheme
3 Synthesis of the cages 18 and 20: (i) SO2Cl; (ii)
HS–(CH2)3–SH, EtONa in EtOH; (iii) 16, Cs2CO3 in DMF; (iv)
HS(CH2)2–SH, EtONa in EtOH; (v) 16, Cs2CO3 in DMF.
1,2-ethanedithiol as bridging component (Scheme 3). The Ag+
complexes were prepared in acetonitrile or EtOH–THF. The
previously described complexes [Ag(14)]+ and [Ag(18)]+ did
not move on TL chromatography, obviously because of their
polymeric structure. In contrast, the complex [Ag(20)]+ gave
chromatograms with several distinct spots which changed in
intensity with time. The dynamic equilibrium between several
Building blocks for a combination of spacer lengths as
required for 24 are not commercially available. Bromination
of 21 with HBr–PBr3 gave the corresponding tribromide 22
and reaction in the usual way with an excess of 1,2-ethane-
dithiol gave the trithiol 23. Compound 23 was then cyclised
with 22 under high dilution condition in the presence of
Cs2CO3 in DMF to give the final product 24 in 20% yield.
The reaction of cage 24 with different Ag(I) salts in acet-
onitrile or EtOH–THF and gave a 1 : 1 complex after 6 h at
50 1C as indicated by ESI-MS. The complex [Ag(24)][O3SCF3]
could be crystallized from acetonitrile and single crystals were
grown by slow diffusion of diethyl ether through the vapour
phase. The overall structure of this complex gives again a
different picture. The Ag+ ion is now coordinated by four
sulfur atoms from one single ligand and by one sulfur from a
second ligand molecule (Fig. 5). This time, the counter-ion
does not participate in coordination to Ag+.
Three out of the five Ag–S bonds are relatively short
(2.51–2.68 A) and two are longer (2.80–2.96 A). The structure
consists of polymeric chains. Still, Ag+ is not in the cavity of
the cage but bound by four sulfurs from the same ligand on
one face of the cage. In the 1H-NMR of the Ag+ complex the
signals are shifted up to 0.23 ppm to lower field as compared
to the signals of the free ligand. The complexation of 24 with
111Ag(I) was studied by TL chromatography. After 30 min in
acetonitrile at 50 1C the TL chromatogram showed one single
peak with Rf = 0.45. No free 111Ag(I) could be observed after
Fig. 4 Structure of the complex cation of [Ag(18)]triflate. Sulfur atom
S14 is disordered over two sites. S14A represents the major site with
occupancy of 0.915 and S14B with 0.085 occupancy. The minor site is
not shown for clarity. Selected bond lengths: Ag–O31 2.526(4), Ag–S8
2.5164(15), Ag–S14A 2.5090(16), Ag–S240 2.6231(15). (Symmetry
operation: ꢁx + 1/2, y ꢁ 1/2, ꢁz + 1/2).
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
412 | New J. Chem., 2007, 31, 409–417

View Article Online
Consequently, we looked at cages with larger caps having
two ([222-S6] as in 24, or [232-S6] as in 20)) or three ([232-S6] as
in 18) carbon atoms. In the free ligand [232-S6] all bond lengths
and angles are close to their optimal values. The electron pairs
of the sulfurs are exo or half exo positioned. The stick model
clearly shows a higher flexibility for [232-S6] than for [131-S6]
and no steric hindrance in the rotations. The calculated
structure of [333-S6] is very similar to that of [232-S6] with
only one close contact between the sulfur atoms (sum of the
van der Waals radii 3.482 A, versus contact 3.438 A).
Because of the structural complexity of cages with six
thioether sulfurs, a prediction for the ‘‘best’’ ligand seems very
difficult. On one side thermodynamic aspects such as preorga-
nization of the ligand and size of the cavity to bind Ag+ are
relevant. On the other side, kinetic aspects also seem very
important, since Ag+ first binds to an exo electron pair of a
sulfur atom then has to move inside the cavity by rotation.
This is only possible if the energy barrier for the rotation is not
too large, which can be achieved by a higher flexibility of the
ligand, which, however, implies that the chelate ring size is also
modified and thus the thermodynamic stability is changed.
Molecular modelling is helpful, but the dynamic required for
inclusion is difficult to predict and calculate. Thus, we pre-
pared a series of S6-cages to test in an experimental way the
consequences of structural changes. The systematic structural
modifications of the S6-cages have given a series of Ag+
complexes which are all polymeric in the solid state with the
metal ion outside of the cage. However, on going from the
[141-S6] cage (14) to [222-S6] cage (24) we observe a systematic
change in the coordination chemistry. With 14, a cage with a
small cap, the Ag+ is coordinated in an monodentate way to
three different macrocycles and the counter ion resulting in a
band structure (Fig. 6). Going to cages with larger caps we
Fig. 5 Structure of the complex cation of [Ag(24)]triflate. Selected
bond lengths and angles: Ag–S4 2.603(3), Ag–S7 2.800(3), Ag–S14
2.967(3), Ag–S17 2.516(3), Ag–S270 2.687(3) A; S4–Ag–S7 81.21(10),
S4–Ag–S14 91.61(10), S4–Ag–S17 163.37(10), S4–Ag–S270 88.36(9),
S7–Ag–S14 150.18(9), S7–Ag–S17 98.39(10), S7–Ag–S270 88.43(8),
S14–Ag–S17 80.48(11), S14–Ag–S270 120.45(9), S17–Ag–S270
108.26(9) (Symmetry operation: ꢁx + 3/2, ꢁy, z + 1/2).
this reaction time. The complex with 111Ag(I) formed in this
way is stable in PBS (phosphate buffered saline, pH 7.6) and
only little decomposition was observed. According to TLC,
the half-life time of this complex is about 20 days.
General structural considerations
To better understand the complexation properties of the
S6-cages prepared herein and to better predict a cage structure
able to encapsulate Ag+, simple molecular modelling was used
in addition to the X-ray diffraction studies. Since the length of
the bridges connecting the sulfur atoms and the size of the caps
can be varied to optimize the conformation and flexibility of
the free ligand as well as the size of its cavity, we have studied
these two factors in more detail.
The structures of two derivatives of the basic cage frame-
work [131-S6] (9 and 11) clearly show that the free ligands have
a very unfavourable conformation with the free electron pairs
of all sulfurs being present in the exo position. We therefore
moved to caps with longer bridges such as [141-S6] (14).
Molecular modelling with Hyperchem15 indicates that the
more flexible chains with four carbon atoms now allow a
conformation with three of the sulfur atoms having their free
electron pairs endo, and three exo, which in a way is a better
preorganisation than in [131-S6]. In the calculated structure
there are only two close contacts, which are smaller than the
sum of van der Waals radii by only 0.074 A (S–C-contact)
and 0.087 A (S–S contact). Dreiding models and molecular
modelling showed that in this ligand the conformation
changes necessary to move all electron pairs of six sulfur
atoms in the endo position implied rotations, which are
hindered by interactions between the hydrogen atoms of
the caps.
Fig. 6 Schematic view of the coordination modes of several S6-cages.
Black dots represent thioether sulfur atoms and X the counter ions.
Top: [Ag(24)]triflate; Middle: [Ag(18)]triflate; Bottom: [Ag(14)]
tosylate.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 409–417 | 413

View Article Online
observe a higher tendency to bind Ag(I) by more sulfur atoms
from the same ligand. Hence, the Ag+ complex of NO2,NO2-
[333-S6] cage (18) produces a polymer in which each metal ion
is bound by two sulfur atoms of one ligand, a sulfur of a
second cage and the counter ion. Finally, in [222-S6] cage (24)
Ag+ forms a complex in which four sulfurs of the same ligand
and one of a neighbouring cage are bound. In this last case,
Ag+ is sitting on one of the faces of the cage, but still exposed
to solvent or other ligands. The coordination of Ag+ can be
considered as an intermediate before full inclusion, it was
however not possible to push the cation into the cage, even
under strong thermal conditions or with the aid of microwave
heating.
d = 0.89 (s, 3 H), 2.47 (s, 9 H), 3.77 (s, 6 H), 7.36 (d, J = 8.3
Hz, 6 H), 7.71 (d, J = 8.3 Hz, 6 H). 13C-NMR (75 MHz,
CDCl3): d = 16.0, 21.5, 39.4, 69.8, 128.0, 130.2, 132.1, 145.5.
FAB-MS: m/z 583.0 (100%).
Tris(para-tolylsulfonylmethyl)nitromethane (5). The com-
pound was prepared in analogy to 4 starting from 2 and
crystallized from methanol–dichloromethane with a yield of
57%. Mp. 119–121 1C. (Found: C, 49.01; H, 4.40; N, 2.35; O,
28.43, S, 15.85%, C25H27NO11S3 (M 613.67) requires: C,
48.93; H, 4.43; N, 2.28; O, 28.68; S, 15.68%; IR (KBr,
1
cmꢁ1): 2928 (w), 1598 (s), 1566 (s), 1494 (m). H-NMR (300
MHz, CDCl3): d = 2.48 (s, 9 H), 4.29 (s, 6 H), 7.38 (d, J = 8.4
Hz, 6 H), 7.72 (d, J = 8.4 Hz, 6 H). 13C-NMR (75 MHz,
CDCl3): d = 21.7, 64.3, 87.2, 128.15, 130.3, 131.0, 146.1.
FAB-MS: 614.1 (64.75%).
Conclusions
Although simple molecular modelling studies confirm from a
thermodynamic point of view that the size of the cages would
allow accommodation of Ag+ in their cavities with a reason-
able complex geometry, the kinetics of the formation seems to
be the main obstacle. Once the Ag+ is bound by one of the
sulfur donors a high-energy barrier for the rotation necessary
to bring the metal ion from the exo to the endo position
prevents the encapsulation. Probably a subtle balance between
thermodynamic and kinetic properties of the ligands is needed
to finally bring the Ag+ in the cage.
Pentaerythrite-tris(para-tolylsulfonate) (6). The compound
was prepared in analogy to 4 starting from 3. Chromatogra-
phy on silica with ethyl acetate–hexane (1 : 1) yielded a white
solid. (43%). Mp. 103–105 1C. (Found C, 51.99; H, 5.00; O,
26.58; S, 16.26%, C26H30O10S3 (M 598.7) requires C, 52.16; H,
5.05; O, 26.72; S, 16.07%). IR (KBr, cmꢁ1): 3542 (m), 2980
(w), 1598 (m), 1358 (s), 1192 (s), 1174 (s), 1096 (m), 1064 (m),
1020 (m), 990 (s), 968 (s), 864 (s), 848 (s), 834 (s), 810 (s), 670
(s), 628 (m), 556 (s). 1H-NMR (300 MHz, CDCl3): d = 2.02 (t,
J = 6.2 Hz, 1 H), 2.47 (s, 9 H), 3.52 (d, J = 6.2 Hz, 2 H), 3.92
(s, 6 H), 7.37 (d, J = 8.6 Hz, 6 H), 7.72 (d, J = 8.6 Hz, 6 H).
13C-NMR (75 MHz, CDCl3): d = 21.6, 44.7, 59.3, 66.6, 128.0,
130.2, 131.9, 145.6. FAB-MS: m/z 599.2 (66.78%).
Experimental
General remarks
Pentaerythrite-(tert-butyl-dimethylsilylether)-tris(para-toylsul-
fonate) (7). A solution of 6 (3.80 g, 6.3 mmol) imidazole (0.48 g,
7.0 mmol) and tert-butyl-dimethylchlorosilane (1.05 g, 7.0
mmol) in DMF (5 ml) was stirred under N2 for 24 h at rt.
The solvent was evaporated and the residue taken up in CH2Cl2
and washed with phosphate buffer (1M, pH = 7.4). After
drying the organic phase was evaporated and the residue
chromatographed on silica (ethylacetate : iso-hexane = 1 : 2)
to give a white product, which was dried at 90 1C in high
vacuum. Yield: 81% (3.64 g). Mp. 76–78 1C. (Found C, 54.05;
H, 6.15; S, 13.62%, C32H44O10SiS3 (M 712.96) requires: C,
53.91; H, 6.22; S, 13.49%). IR (KBr, cmꢁ1): 2950 (m), 1598 (m),
All starting materials were purchased either from Fluka or
Aldrich and used without further purification. All reactions
were carried out under an inert atmosphere (N2). NMR
spectra were recorded on a Varian Gemini 2000, 300 MHz
at room temperature and referred to the residual solvent
signal. IR spectra were recorded on a Perkin-Elmer 16 PC
FT-IR spectrometer. MS spectra and elemental analysis were
performed by the analytical service of the ETH Zurich. Melt-
¨
ing points were measured with a Buchi 510 and are un-
¨
corrected. Compounds 157 and 2110 have been prepared as
described elsewhere.
1
1472 (m), 1358 (s). H-NMR (300 MHz, CDCl3): d = ꢁ0.07
Syntheses
(s, 6) 0.73 (s, 9 H), 2.47 (s, 9 H), 3.41 (s, 2 H), 3.85 (s, 6 H), 7.36
(d, J = 8.5 Hz, 6 H), 7.72 (d, J = 8.5 Hz, 6 H). 13C-NMR
(75 MHz, CDCl3): d = ꢁ6.7, 17.8, 21.5, 25.5, 44.4, 59.4, 66.5,
128.0, 130.1, 132.0, 145.4. FAB-MS: m/z 713.2 (23.43%).
1,1,1-Tris(para-tolylsulfonylmethyl)ethane (4). At 0 1C a
solution of para-toluenesulfochloride (47.6 g, 250 mmol) in
absolute pyridine (90 ml) was added to a suspension of 1,1,1-
tris(hydroxymethyl)ethane 1 (10 g, 83 mmol) in absolute
pyridine (70 ml) over 2 h. The reaction mixture was stirred
at rt overnight and hydrolysed by adding ice. The product was
extracted with CH2Cl2 and the organic phase successively
washed with 1M HCl, brine and water. After drying over
Na2SO4 the solvent was evaporated. Recrystallisation from
methanol yielded a white solid. Yield 33.8 g (70%). Mp
107–109 1C. (Found: C, 53.34; H, 5.20, O, 24.51, S, 16.70%
C26H30O9S3 (M 582.72) requires: C, 53.59; H, 5.19; O, 24.71;
S, 16.51%; IR (KBr, cmꢁ1): 1600 (m), 1358 (m), 1178 (s), 1096
(w), 1006 (m), 986 (w), 962 (m), 868 (w), 838 (m), 814 (m), 668
1,1,1-Tris(5-mercapto-2-thiapentanyl)ethane (8). To a solu-
tion of sodium (2.6 g, 113 mmol) in abs. ethanol (200 ml),
1,3-propanedithiol (55.7 g, 0.515 mol) was added at 60 1C. To
this, 4 (20 g, 34 mmol) was added in small portions over 7 h.
The reaction mixture was heated overnight and after cooling,
acidified with 1M HCl. Solvents were evaporated and the
residue dissolved in CH2Cl2. The organic phase was washed
with brine and dried over Na2SO4. The mixture was chromato-
graphed on silica (CH2Cl2 : hexane = 2 : 1) and the compound
was isolated as a colourless oil. Yield 8.3 g (62%). (Found C,
42.98; H, 7.72; S, 49.32%, C14H30S6 (M 390.75) requires:
(m), 600 (w), 558 (m) cmꢁ1 1H-NMR (300 MHz, CDCl3):
.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
414 | New J. Chem., 2007, 31, 409–417

View Article Online
C, 43.03; H, 7.74; S, 49.23%). IR (KBr, cmꢁ1): 2956 (s), 2908
(s), 2548 (m), 1424 (s), 1370 (m), 1344 (m), 1296 (s), 1252 (s),
1204 (m), 866 (m), 754 (m). 1H-NMR (300 MHz, CDCl3): d =
1.05 (s, 3 H), 1.36 (t, J = 8 Hz, 3 H), 1.85 (tt, J = 7 Hz,
6 H), 2.57–2.65 (m, 18 H). 13C-NMR (75 MHz, CDCl3):
d = 23.2, 23.8, 32.1, 33.4, 40.3, 41.5. EI-MS: m/z 390.0
(11.3%).
as eluent. Yield 15%. Mp 78–81 1C. (Found C, 51.05; H, 8.33;
S, 32.68%, C25H50OSiS6 (M 587.12) requires C, 51.14; H, 8.58;
S, 32.77%). IR (KBr, cmꢁ1): 2952 (s), 2924 (s), 2854 (m), 1464
1
(m), 1414 (m), 1250 (m), 1092 (s) 844 (w), 742 (w). H-NMR
(300 MHz, CDCl3): d = 0.05 (s, 6H), 0.89 (s, 9 H), 1.09 (s, 3
H), 1.99 (t, J = 6.6 Hz, 6 H), 2.73 (t, J = 6.6 Hz, 6 H), 2.73 (t,
J = 6.6 Hz, 6 H), 2.87 (s, 6 H), 2.92 (s, 6 H), 3.57 (s, 2 H). 13C-
NMR (75 MHz, CDCl3): d = ꢁ5.71, 18.11,, 24.1, 25.8, 28.4, ,
31.6, 31.8, 38.0, 40.7, 42.1, 45.9, 66.8. FAB-MS: m/z 587.3
(100%).
1,9-Dimethyl-3,7,11,15,18,22-hexathiabicyclo[7.7.7]tricosane
(9). Caesium carbonate (2.84 g, 8.7 mmol) was suspended in
dry DMF (270 ml) under nitrogen. 4 (2.98 g, 5.12 mmol) and 8
(2 g, 5.12 mmol) each dissolved in dry DMF (320 ml) were
added at the same time dropwise under stirring over 72 hours
at 60 1C in a nitrogen atmosphere. After complete addition the
reaction mixture was stirred for an additional day. The solvent
was removed in vacuum and the remaining solid was extracted
several times with dichloromethane. Evaporation of the
solvent gave a brownish semisolid. Column chromatography
over silica with dichloromethane as eluent yielded the
product as a white, crystalline solid. Yield 0.24 g (10%).
Mp. 104–105 1C. (Found C, 49.96; H, 7.76; S, 42.22%,
C19H36S6 (M 456.85) requires: C, 49.95; H, 7.94; S, 42.11%).
IR (KBr, cmꢁ1): 2912 (s), 1430 (m), 1254 (m), 844 (m).
1H-NMR (300 MHz, CDCl3): d = 1.09 (s, 6 H), 1.99 (q, J
= 6.6 Hz, 6 H), 2.75 (t, J = 6.6 Hz, 12 H), 2.87 (s, 12 H).
13C-NMR (75 MHz, CDCl3): d = 24.1, 28.4, 21.6, 40.8, 42.2.
EI-MS: m/z 456.1 (23.34%).
1,1,1-Tris(6-mercapto-2-thiahexanyl)-ethane (13). The com-
pound was prepared from 1,4-butanedithiole and 4 in analogy
to 8. The residue after evaporation of the solvent was sub-
jected to column chromatography on silica (CH2Cl2 : hexane
= 2 : 1). Yield 60%. Found C, 47.20; H, 8.42%, C17H36S6
(M 432.83) requires: C, 47.17; H, 8.38%). IR (KBr, cmꢁ1):
2928 (s), 2848 (m), 2548 (w), 1450 (m), 1370 (m), 1280 (m),
1244 (m), 1202 (w), 860 (w), 738 (w). 1H-NMR (300 MHz,
CDCl3): d = 1.06 (s, 3 H), 1.35 (t, J = 7.9 Hz, 3 H), 1.67–1.71
(m, 12 H), 2.49–2.56 (m, 12 H), 2.63 (s, 6 H). 13C-NMR
(75 MHz, CDCl3): d = 23.7, 24.1, 28.3, 33.3, 32.8, 40.3, 41.6.
FAB-MS: m/z 432.0 (100%).
1-Hydroxymethyl-10-methyl-3,8,12,17,20,25-hexathiabicyclo
[8.8.8]hexacosane (14). The compound was prepared starting
from 13 and 6 in analogy to 9. Column chromatography over
silica with dichloromethane : acetonitrile = 9 : 1 as eluent gave
the product as a white, crystalline solid. Yield 29%. M.p.:
80–82 1C. (Found C, 51.47; H, 8.26; O, 3.29; S, 37.11%,
C22H42OS6 (M 514.93) requires C, 51.31; H, 8.22; O, 3.11; S,
37.36%). IR (KBr, cmꢁ1): 3438 (m, br), 2914 (s), 2848 (m),
1420 (m), 1372 (w), 1280 (m), 1240 (m), 1052 (m), 858 (w), 750
(w). 1H-NMR (300 MHz, CDCl3): d = 1.09 (s, 3 H), 1.78–1.83
(m, 12 H), 2.60–2.65 (m, 12 H), 2.77 (s, 6 H), 2.87 (s, 6 H), 3.67
(s, 2 H). 13C-NMR (75 MHz, CDCl3): d = 23.7, 27.0, 27.2,
32.9, 32.9, 37.4, 40.0, 41.4, 44.2, 68.0. FAB-MS: m/z (%) =
514.2 (69.46%).
1-Nitro-9-methyl-3,7,11,15,18,22-hexathiabicyclo[7.7.7]tricosane
(10). The compound was prepared from 5 and 8 in analogy to
the cage 9. The purification was achieved by column chroma-
tography on silica with dichloromethane : iso-hexane = 8 : 1.
Yield 12%. Mp. 117–119 1C. (Found C, 44.40; H, 6.95; N,
2.90; O, 6.45; S, 39.31%, C18H33NO2S6 (M 487.82) requires C,
44.32; H, 6.82; N, 2.87; O, 6.56; S, 39.44%). IR (KBr, cmꢁ1):
2908 (m), 1538 (s), 1406 (m), 1374 (w), 1344 (m). 1H-NMR
(300 MHz, CDCl3): d = 1.10 (s, 3 H), 2.02 (tt, JAB = JBC 6.7
Hz, 6 H), 2.74 (t, J = 6.7 Hz, 6 H), 2.83 (s, 6 H), 3.43 (s, 6 H).
13C-NMR (75 MHz, CDCl3): d = 23.9, 28.4, 31.7, 32.3, 38.3,
40.7, 42.1, 95.6. EI-MS: m/z 487.0 (33.43%), 441.1 (33.29%,
M-NO2).
Tris(3-chloropropanyl)nitromethane (16).
A mixture of
tris(3-hydroxypropanyl)nitromethane 15 (10 g, 43 mmol) and
thionyl chloride (16 ml, 216 mmol) was stirred at rt for one
hour. Excess thionyl chloride was evaporated and the mixture
chromatographed on silica with CH2Cl2 as eluent. The pro-
duct was isolated as a yellow oil. Yield 10.7 g (86%). (Found
C, 41.59; H, 6.26; N, 4.64; O, 11.20%, C10H18Cl3NO2 (M
290.62) requires: C, 41.33; H, 6.24; N, 4.82; O, 11.01%). IR
(KBr, cmꢁ1): 2964 (s), 2872 (m), 1537 (s), 1448 (m), 1354 (m),
1316 (m), 1116 (w), 842 (w), 782 (w), 732 (w), 652 (m).
1H-NMR (300 MHz, CDCl3): d = 1.66–1.75 (m, 6 H),
2.06–2.12 (m, 6 H), 3.56 (t, J = 6.1 Hz, 6 H). 13C-NMR (75
MHz, CDCl3): d = 26.5, 32.8, 44.2, 93.0. EI-MS: m/z 243.0
(12.35%).
1-Hydroxymethyl-9-methyl-3,7,11,15,18,22-hexathiabicyclo
[7.7.7]tricosane (11). The compound was prepared from 8 and
6 in analogy to 9. Purification was achieved by column
chromatography over silica with dichloromethane : aceto-
nitrile = 9 : 1 as eluent. Yield 22%. Mp. 164 1C. (Found C,
48.11; H, 7.61; O, 3.67; S, 40.76%, C19H36OS6 (M 472.85)
requires: C, 48.26; H, 7.67; O, 3.38; S, 40.69%). IR (KBr,
cmꢁ1): 3448 (w), 2950 (m), 2912 (s), 1448 (m), 1430 (m), 1400
(m), 1290 (m), 1254 (m), 1188 (w), 844 (w), 742 (w). 1H-NMR
(300 MHz, CDCl3): d = 1.09 (s, 3 H), 2.00 (tt, J = 6.6 Hz, 6
H), 2.76 (t, J = 6.6 Hz, 12 H), 2.88 (s, 6 H), 2.98 (s, 6 H), 3.65
(s, 2 H). 13C-NMR (75 MHz, CDCl3): d = 24.1, 28.2, 31.5,
31.7, 38.3, 40.9, 42.4, 45.6, 68.0. EI-MS: m/z 472.2 (100%).
Tris(7-mercapto-4-thiaheptanyl)nitromethane (17). The com-
pound was prepared starting from 16 and 1,3-propanedithiole
in analogy to 8. The residue was subjected to column chro-
matography on silica (CH2Cl2 : hexane = 3 : 1). The com-
pound was isolated as a colourless oil. Yield 53%. (Found
C, 45.07; H, 7.74; N, 2.67; O, 6.17; S 38.25%, C19H39NO2S6
1-(tert-Butyl-dimethylsilyloxymethyl)-9-methyl-3,7,11,15,18,
22-hexathia[7.7.7]tricosane (12). The compound was prepared
in analogy to 9 starting from 7 and 8. Purification was
achieved by column chromatography on silica with CH2Cl2
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 409–417 | 415

View Article Online
(M 505.88) requires: C, 45.11; H, 7.77; N, 2.77; O, 6.32;
S, 38.03%). IR (KBr, cmꢁ1): 2928 (s), 2548 (m), 1531 (s),
1452 (m), 1352 (m), 1298 (m), 1260 (m), 842 (m). 1H-NMR
(300 MHz, CDCl3): d = 1.37 (t, J = 8 Hz, 3 H), 1.44–1.54
(m, 6 H), 1.81–1.90 (tt, J = 7 Hz, 6 H), 1.98–2.04 (m, 6 H),
2.51 (t, J = 7 Hz, 6 H), 2.58–2.66 (m, 12 H). 13C- NMR
(75 MHz, CDCl3): d = 23.2, 23.5, 30.3, 31.7, 33.1, 34.4, 93.8.
EI-MS: m/z 505.10 (0.8%).
(75 MHz, CDCl3): d = 30.3, 34.7, 36.1. EI-MS: m/z 256.9
(58% M+ ꢁ Br).
6-(5-Mercapto-3-thiapentanyl)-1,11-dimercapto-3,9-dithiaun-
decane (23). The compound was prepared from 1,2-ethane-
dithiol and 22 in analogy to 8. The raw product was subjected
to gradient column chromatography on silica (CH2Cl2
:
hexane from 1 : 1 to 3 : 1). The compound was isolated as
colourless oil. Yield 27%. IR (KBr, cmꢁ1): 3446 (w, br), 2916
(s), 2850 (m), 2538 (w), 1684 (w), 1654 (w), 1560 (w), 1424 (m),
1268 (m), 1208 (m), 1138 (m), 964 (w), 692 (m). 1H-NMR (300
MHz, CDCl3): d = 1.54–1.61 (m, 6 H), 1.68–1.78 (m, 4 H),
2.55 (t, J = 7.7 Hz, 6 H), 2.71–2.77 (m, 12 H). 13C-NMR
(75 MHz, CDCl3): d = 24.7, 29.4, 33.1, 35.7, 36.3.
1,13-Dinitro-5,9,17,21,28,32-hexathiabicyclo[11.11.11]penta-
triacontane (18). The compound was prepared starting from 17
and 16 in analogy to 9. Column chromatography over silica
with CH2Cl2 as eluent yielded the product as a white, crystal-
line solid. Yield 42%. Mp 75–78 1C. (Found C, 50.66; H, 7.76;
N 4.13; O, 9.26; S, 28.21%, C29H54N2O4S6 (M 687.12)
requires: C, 50.69; H, 7.92; N, 4.08; O, 9.31; S, 28.00%). IR
(KBr, cmꢁ1): 3448 (w, br), 2914 (m), 1534 (s), 1450 (m), 1350
(w), 1280 (w), 834 (w), 792 (w). 1H-NMR (300 MHz, CDCl3):
d = 1.57–1.64 (m, 12 H), 1.82–1.91 (m, 6 H), 2.05–2.16 (m, 12
H), 2.50–2.71 (m, 24 H). 13C-NMR (75 MHz, CDCl3): d =
23.5, 23.7, 24.3, 29.0, 29.1, 30.2, 30.8, 31.0, 31.2, 32.0, 32.3,
34.5, 34.6, 34.9, 93.7. EI-MS: m/z 686.2 (17.37%).
4,7,13,16,21,24-Hexathiabicyclo[8.8.8]hexacosane (24). The
compound was prepared starting from 22 and 23 in analogy to
9. Column chromatography over silica with CH2Cl2 as eluent
yielded the product as a white, crystalline solid. Yield 20%. IR
(KBr, cmꢁ1): 3446 (m, br), 2924 (s), 1684 (m), 1654 (m), 1636
(m), 1560 (m), 1436 (m), 1194 (m). 1H-NMR (300 MHz,
CDCl3): d = 1.55–1.78 (m, 14 H), 2.57–2.72 (m, 12 H), 2.79
(s, 6 H), 2.83 (s, 6 H). 13C-NMR (75 MHz, CDCl3): d = 27.5,
29.0, 30.1, 30.5, 30.7, 32.0, 32.1, 33.6, 33.8, 34.3, 35.0,
36.0, 37.6.
Tris(6-mercapto-4-thiahexanyl)nitromethane (19). The com-
pound was prepared from 1,2-ethanedithiole and 16 in ana-
logy to 8. The residue was subjected to gradient column
chromatography on silica (CH2Cl2 : hexane from 1 : 1 to
1 : 0). The compound was isolated as a colourless oil. Yield
63%. IR (KBr, cmꢁ1): 2922 (s), 2542 (m), 1532 (s), 1452 (s),
1426 (s), 1352 (s), 1268 (s), 1268 (s), 1208 (s), 1140 (m), 964
(m), 842 (m), 696 (m). 1H-NMR (300 MHz, CDCl3): d =
1.54–1.46 (m, 6 H), 1.79–1.75 (m, 3 H), 2.03–1.97 (m, 6 H),
2.56–2.51 (m, 6 H), 2.78–2.68 (m, 12 H). 13C-NMR (75 MHz,
CDCl3): d = 23.6, 24.5, 31.6, 34.3, 36.1, 93.6. FAB-MS: m/z
463.3 (9%).
Table 1 Crystal data and structural refinement for the cages 9 and 11
9
11
Formula
MW
C
19H36S6
C19H36OS6
472.84
293(2)
1.54184
Monoclinic
P2(1)/c
18.186(4)
26.587(4))
10.255(2)
90
456.84
208(2)
1.54184
Monoclinic
Cc
9.432(1)
27.584(2)
9.901(1)
90
Temperature/K
Wavelength/A
Crystal
Space group
A/A
b/A
c/A
a (1)
1,12-Dinitro-5,8,16,19,26,29-hexathiabicyclo[10.10.10]dotria-
contane (20). The compound was prepared starting from 16
and 19 in analogy to 9. Column chromatography over silica
with CH2Cl2 as eluent yielded the product as a colourless oil
that solidified upon standing. Yield 17%. IR (KBr, cmꢁ1):
1
2926 (m), 1534 (s), 1452 (m), 1350 (m) cmꢁ1. H-NMR (300
b (1)
114.26(1)
90
2348.5(4)
4
101.16(1)
90
4864(1)
8
1.291
MHz, CDCl3): d = 1.56–1.65 (m, 12 H), 2.08–2.14 (m, 12 H),
2.52–2.66 (m, 12 H), 2.72–2.82 (m, 12 H). 13C-NMR (75 MHz,
CDCl3): d = 23.4, 24.2, 24.5, 31.8, 32.0, 32.4, 32.6, 32.7, 34.5,
35.0, 93.6. FAB-MS: m/z = 645.0 (8%).
g (1)
Volume/A3
Z
Density calc./mg mꢁ3 1.292
Abs. coefficient/mmꢁ1 5.374
5.239
2032
F(000)
984
0.15 ꢂ 0.15 ꢂ 0.15
Crystal size/mm
Theta range/1
Index ranges
0.30 ꢂ 0.20 ꢂ 0.15
1,5-Dibromo-3-(2-bromoethyl)pentane (22). To ice cooled
3-(2-hydroxyethyl)pentane-1,5-diol (21) (4.7 g, 31.7 mmol)
PBr3 (106.6 g, 37 ml) was slowly added. Then HBr (47%,
8 ml) was added to the mixture, which was stirred for 1 h at
0 1C, then 2 h at rt and finally 12 h at 100 1C. The mixture was
diluted with water (300 ml) and neutralized with NaHCO3.
The aqueous phase was extracted with CH2Cl2. The organic
phase was washed with water, dried with Na2SO4 and evapo-
rated to dryness. The yellow oil was dried in high vacuum at 80
1C, whereby the nearly pure product was obtained. Yield 84%
(9.03 g, 26.8 ml). IR (KBr, cmꢁ1): 2966 (s), 2932 (s), 1654 (m),
1560 (m), 1448 (s), 1260 (s), 1228 (s). 1H-NMR (300 MHz,
CDCl3): d = 1.89 (m, 7 H), 3.41 (t, J = 7.0, 6 H). 13C-NMR
5.39–64.00
5.23–65.00
ꢁ10 o = h o = 10, ꢁ21 o = h o = 21,
ꢁ2 o = k o = 32, ꢁ31 o = k o = 0,
ꢁ11 o = l o = 11 ꢁ1 o = l o = 12
Reflections collected
4265
9737
Independent reflections 3831
Absorption correction None
8238
psi scan
Tmax/Tmin
—
0.9000/0.5877
Full matrix
0.997
Refinement method
Goodness on F2
Final R indices
[>I 4 2 sigma(I)]
R indices (all data)
Largest diff. peak
and hole
Full matrix
0.931
0.0754/0.1989
0.0552/0.1378
0.0957/0.2235
0.428/–0.433
0.0899/0.1604
0.754/–0.331
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
416 | New J. Chem., 2007, 31, 409–417

View Article Online
Table 2 Crystal data and structural refinement for [Ag(14)]tosylate, [Ag(24)]triflate and [Ag(18)]triflate ꢃ 2CH3CN
[Ag(14)]tosylate
[Ag(24)]triflate
[Ag(18)]triflate ꢃ 2CH3CN
Formula
MW
Temperature/K
Wavelength/A
Crystal
Space group
C29H49AgO4S7
793.97
208(2)
0.71073
Monoclinic
P21/c
C21H38AgF3O3S7
727.80
208(2)
0.71073
Orthorhombic
P212121
C35H60Ag2F6O10N4S8
1283.09
208(2)
0.71073
Monoclinic
C2/c
A/A
b/A
c/A
a/1
10.707(4)
17.737(4)
19.081(8)
90
10.3960(10)
15.421(2)
18.421(1)
90
21.614(8)
13.491(2)
18.364(8)
90
b/1
g/1
103.61(2)
90
3522(2)
4
1.497
90
90
2953.2(9)
4
1.637
109.70(2)
90
5041(3)
4
1.691
Volume/Aꢁ3
Z
Density calc./mg mꢁ3
Abs. coefficient/mmꢁ1
F(000)
1.020
1656
1.219
1496
1.185
2616
Crystal size/mm3
Theta range/1
Index ranges
0.15 ꢂ 0.15 ꢂ 0.10
0.40 ꢂ 0.30 ꢂ 0.15
3.24–26.49
0 o = h o = 13,
0 o = k o = 19,
ꢁ23 o = l o = 23,
6656
6098
Psi scan
0.9552/0.9050
Full matrix
1.029
0.20 ꢂ 0.15 ꢂ 0.15
3.02–26.96
ꢁ27 o = h o = 16,
ꢁ15 o = k o = 17,
ꢁ22 o = l o = 23
12148
5466
Psi scan
0.87360/0.75096
Full matrix
1.005
3.02–24
ꢁ1 o = h o = 12,
0 o = k o = 20, ꢁ21
o = l o = 21,
6398
5419
Difabs
1.000/0.8721
Full matrix
0.956
Reflections collected
Independent reflections
Absorption correction
Tmax/Tmin
Refinement method
GoF
Final R indices [I 4 2 s(I)]
R indices (all data)
Largest diff. peak and hole
0.0776/0.0755
0.3303/0.1173
0.534/–0.970
0.0675/0.1353
0.1410/0.1668
0.823/–0.759
0.0476/0.1055
0.1007/0.1235
1.488/–0.627
X-ray crystallographic study
References
X-ray crystal structure analyses were performed on
CAD-4 Enraf Nonius diffractometers with Cu Ka (l =
1.54184 A) (compounds 9 and 11) and Mo Ka radiation
(l = 0.71073 A) ([Ag(14)]tosylate, [Ag(24)]triflate and
[Ag(18)]triflate).
1 M. R. Zalutsky, in Handbook of Nuclear Chemistry, ed. F. Rosch,
¨
Kluwer Academic Publishers, 2003, vol. 4, pp. 315–348.
2 P. A. Schubiger, R. Alberto and A. Smith, Bioconjugate Chem.,
1996, 7, 165.
3 W. A. Volkert and T. J. Hoffmann, Chem. Rev., 1999, 99,
2269–2292.
4 R. Alberto, P. Blauenstein, I. Novak-Hofer, A. Smith and P. A.
¨
Schubiger, Appl. Radiat. Isot., 1991, 43, 869.
5 R. Alberto, W. Nef, A. Smith, T. A. Kaden, M. Neuburger, M.
Zehnder, A. Frey, U. Abram and P. A. Schubiger, Inorg. Chem.,
1996, 35, 3420.
6 R. Alberto, D. Angst, U. Abram, K. Ortner, T. A. Kaden and P.
A. Schubiger, Chem. Commun., 1999, 1513.
7 D. Angst, Ph.D. Thesis, University of Basel, 1998.
All structures were solved by direct methods using the
program SHELXS-97 and refined using the program
SHELXL-97.16 The refinement was performed with aniso-
tropic thermal parameters for all non-hydrogen atoms
except of two restrained C atoms in the structure of [Ag(14)]-
tosylate. Hydrogen atom positions were included at idealized
positions and treated by the ‘riding model’ option of
SHELXL-97. Some disorder positions were considered in
the refinement of the structures of compound 11 and
[Ag(18)]triflate ꢃ 2CH3CN. More details about crystal data
and the structure refinements are contained in Table 1 and
Table 2.z
8 T. Gyr, H. R. Macke and H. Henning, Angew. Chem., Int. Ed.,
1998, 36, 2786.
¨
9 J. M. Baumeister, R. Alberto, K. Ortner, B. Spingler, P. A.
Schubiger and Th. A. Kaden, J. Chem. Soc., Dalton Trans.,
2002, 4143.
10 Comprehensive Coordination Chemistry II, ed. D. E. Fenton,
Elsevier, 2004, vol. 6, p. 911.
11 J. Blake, R. O. Gould, A. J. Holder, T. I. Hyde and M. Schroder,
¨
Polyhedron, 1989, 8, 513.
12 (a) P. Osvath and A. M. Sargeson, J. Chem. Soc., Chem. Commun.,
1993, 40; (b) J. Buter and R. M. Kellogg, J. Chem. Soc., Chem.
Commun., 1980, 466.
Acknowledgements
13 W. H. Kruizinga and R. M. Kellogg, J. Am. Chem. Soc., 1981, 103,
5183.
The authors express their thanks to the Swiss National Science
Foundation (project No. 20-58958.99) for supporting this
research.
14 E. J. Corey and A. Venkateswarlu, J. Am. Chem. Soc., 1972, 94,
6190.
15 Hyperchems, Computer program, Hypercube, Inc., Waterloo,
1994.
16 G. M. Sheldrick, SHELXS-97 and SHELXL-97-Programs for
the solution and refinement of crystal structures, University of
z CCDC reference numbers 292370–292372 and 633283 and 633284.
For crystallographic data in CIF or other electronic format see DOI:
10.1039/b606510b
Gottingen, Germany, 1997.
¨
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 409–417 | 417