Dimetal complexes of a bibrachial 2+2 thiolate-based macrocycle

Article abstract of DOI:10.1080/10610278.2012.701302

Protocols for accessing the [2+3] and [2+2] tren + thiophenolate-based cryptands and macrocycles, respectively, have been devised; however, a propensity towards incomplete crypt formation is clear: Cd(II)2, Mn(II)2, Ni(II)2 and Pd(II)2 complexes of the [2+2] bibrachial systems in which one arm of each of the tren head groups is pendant were isolated in attempts to make the crypts. The Mn(II)2 complex can be used in transmetallation reactions and these probably occur in a stepwise fashion for the heavier, softer metal ions since the plausible intermediates Mn(II)M(II), M = Cd, Pd, were identified. There is evidence for the formation of [2+3] cryptand systems using templating Ag(I) and Cd(II); however, reaction conditions need to be strictly controlled to ensure isolation of either the [2+3] cryptand or the [2+2] macrocyclic complex since ensuing reactions leading to intractable oligomers are competing. The template and transmetallation experiments described here emphasise the relatively greater hydrolytic sensitivity of the thiophenol-derived imino macrocycles and crypts than their better-known phenol-based counterparts.

Full text of DOI:10.1080/10610278.2012.701302

This article was downloaded by: [University of Waterloo]
On: 12 October 2014, At: 12:29
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Supramolecular Chemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gsch20
Dimetal complexes of a bibrachial 2+2 thiolate-based
macrocycle
Anders Lennartson a , Vickie McKee b , Jane Nelson a b , Michael Arthurs c & Christine J.
McKenzie a
a Department of Physics , Chemistry and Pharmacy, University of Southern Denmark ,
Campusvej 55, 5230, Odense M , Denmark
b Department of Chemistry , Loughborough University , Leicestershire , LE11 3TU , UK
c Department of Chemistry , Queens University, Belfast , BT7 5AG , UK
Published online: 01 Aug 2012.
To cite this article: Anders Lennartson , Vickie McKee , Jane Nelson , Michael Arthurs & Christine J. McKenzie (2012)
Dimetal complexes of a bibrachial 2+2 thiolate-based macrocycle, Supramolecular Chemistry, 24:8, 604-617, DOI:
10.1080/10610278.2012.701302
To link to this article: http://dx.doi.org/10.1080/10610278.2012.701302
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Supramolecular Chemistry
Vol. 24, No. 8, August 2012, 604–617
Dimetal complexes of a bibrachial 2 1 2 thiolate-based macrocycle
Anders Lennartsona, Vickie McKeeb, Jane Nelsona,b, Michael Arthursc and Christine J. McKenziea*
aDepartment of Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark;
c
bDepartment of Chemistry, Loughborough University, Leicestershire LE11 3TU, UK; Department of Chemistry, Queens University,
Belfast BT7 5AG, UK
(Received 4 April 2012; final version received 4 June 2012)
Protocols for accessing the [2 þ 3] and [2 þ 2] tren þ thiophenolate-based cryptands and macrocycles, respectively, have
been devised; however, a propensity towards incomplete crypt formation is clear: Cd(II)2, Mn(II)2, Ni(II)2 and Pd(II)2
complexes of the [2 þ 2] bibrachial systems in which one arm of each of the tren head groups is pendant were isolated in
attempts to make the crypts. The Mn(II)2 complex can be used in transmetallation reactions and these probably occur in a
stepwise fashion for the heavier, softer metal ions since the plausible intermediates Mn(II)M(II), M ¼ Cd, Pd, were
identified. There is evidence for the formation of [2 þ 3] cryptand systems using templating Ag(I) and Cd(II); however,
reaction conditions need to be strictly controlled to ensure isolation of either the [2 þ 3] cryptand or the [2 þ 2] macrocyclic
complex since ensuing reactions leading to intractable oligomers are competing. The template and transmetallation
experiments described here emphasise the relatively greater hydrolytic sensitivity of the thiophenol-derived imino
macrocycles and crypts than their better-known phenol-based counterparts.
Keywords: macrocycle; cryptand; thiophenol; dimetallic; tren
Introduction
thiophenolate S atoms use a pyramidal geometry (2–15).
Thus, the [2 þ 2] systems of the former are planar whereas
they are Vor boat shaped for the latter. The [4 þ 4] system,
known only for X ¼ S, forms a torus structure when four
zinc ions are bound (3). The presence of sulphur will
obviously increase the accessibility to complexes of softer
metal ions.
Macrocycles (Scheme 1, X ¼ O) derived from the dimetal
templated 2 þ 2 condensations of diamines and para-
substituted phenol-1,6-dialdehydes were described by
Robson in 1970 (1). This contribution spurred an on-
going interest in these readily accessible dinucleating
macrocycles which have shown excellent ability to support
homodinuclear, mixed-metal complexes and mixed-
valence complexes. The two phenolate O atoms doubly
bridge the metal ions and, due to their trigonal geometry, the
phenol rings are coplanar with each other and with the four
imine bonds. In principle, macrocyles derived from 3 þ 3,
4 þ 4, etc., Scheme 1(a) n ¼ 3, 4, etc., are possible but
these have not yet been observed by condensation of simple
aliphatic diamines (ethylenediamine and propylenedia-
mine) with the dialdehyde phenol linkers. Notably this
situation is different if thiophenolate replaces phenolate
group in these macrocycles, Scheme 1 X ¼ S. Both [2 þ 2]
(2–6) and [4 þ 4] (2, 3) dimetal and tetrametal templated
macrocycles then become accessible. Replacement of the
bridging phenolate O atom with an S atom in the
construction of macrocycles and crypts has significant
consequences in terms of topology and chemistry. The
effect on structural topology is obvious through the
comparison of the structures of analogous O- and S-hinged
systems, Scheme 1(b). In contrast to the trigonal planar
geometry of hinging phenolato O atoms, hinging
Although thiolate donors in general are of considerable
bioinorganic interest (16), synthetic problems, including the
tendency of these electron-rich donors to aggregation, and
their facile redox activity potentially generating disulphide
links, have delayed progress in this area of chemistry. The
provision of a macrocyclic or better, macrobicyclic frame-
work offers some control over the degree of aggregation and
appears a possibleway to mimic the active site of proteins. In
the synthesis of macrocyclic thiolates, it is necessary to
protect the thiolate function against undesired side reactions.
In earlier studies of thiophenolate ligands, we realised that
although masking of the thio function is necessary during
synthesis of thiolato ligands, in favourable cases unmasking
may subsequently take place spontaneously upon coordi-
nation of the appropriate cation. Thus, where a coordination
set of appropriate geometry requires a nearby thiolate donor
for completion, no deprotection strategy may be necessary.
With such an outcome in mind, and aiming to extend or
define the limitations of the Schiff-base condensation
strategy (17–21) for generation of the macrobicyclic
*Corresponding author. Email: mckenzie@sdu.dk
ISSN 1061-0278 print/ISSN 1029-0478 online
q 2012 Taylor & Francis
http://dx.doi.org/10.1080/10610278.2012.701302
http://www.tandfonline.com

Supramolecular Chemistry
605
Scheme 1. (a) General formula for macrocycles based on propylenediamine and phenols (X ¼ O) or thiophenols (X ¼ S), (b) known
[2 þ 2] macrocycle X ¼ O, S and (c) known [4 þ 4] X ¼ S.
skeleton, we wished to attempt template synthesis of
thiophenolate-incorporatingcryptands, Scheme 2(a), X ¼ S.
The aim was to complement, with a suitable host for soft
cations, the phenolate series studied earlier which have
proved so valuable, even in their hexaimino form, Scheme
2(a), X ¼ O, as hosts for relatively hard cations from the
lanthanide (22, 23) and transition cation (24, 25) series.
The reagents chosen for condensation were tris(2-
aminoethyl)amine (tren), and the 2,6-dialdehyde of p-tBu,
p-Me or p-OMe thiophenol, protected as the formamido-
dimethyl adduct, and using a range of template cations. In
choosing the cations, a number of apparently mutually
contradictory properties were sought: (i) ability to
coordinate thiols, (ii) capacity to unmask the thiol-
protecting group, (iii) relatively mild Lewis acidity to
prevent destruction of the newly formed imino function via
hydrolytic attack (25), (iv) absence of the facile redox
transformation which would permit oxidation of the
deprotected thiolate functions to disulphide links and (v)
the stereochemical plasticity needed to permit adoption of a
range of non-planar coordination geometries. It is clear that
no single cation can meet all these criteria and compromises
will be required. Initially, Ag(I) was selected from the
relatively soft second transition series, because of its well-
established capacity both for binding sulphur and for
coordinating sp2 N in a range of coordination geometries,
although it may perhaps be insufficiently active in criterion
(ii). For better performance on criterion (ii), albeit with
some concern about criterion (iii), another second transition
series ion, Cd(II), was chosen. Among cations from the first
transition series, which are all (in at least one respect) non-
ideal, reasonable performance across the listed criteria is
expected of Ni(II).
Cryptands formed by replacement of the diamine with a
triamine such as tren in Schiff-base condensations with
diformylphenols [Scheme 2(a) X ¼ O (21, 22)] show a
R
R+
N
N
X–
N
N
N
N
X–
X–
H2N
N
N
N
N
N
N
NH2
X–
X–
R
N
N
R
R
Scheme 2. (a) A macrobicycle or [2 þ 3] cryptand and (b) a [2 þ 2] bibrachial macrocycle based on tren and phenols (X ¼ O) or
thiophenols (X ¼ S).

606
A. Lennartson et al.
propensity to form mononuclear complexes, due to the
competition introduced by the potential set of three charge-
assisted H-bonds; under acidic or neutral conditions,
protonation of the three imine protons in the uncoordinated
side of the host cavity occurs. This normally blocks the
second metal coordination site (21, 22). Where basic
conditions are used to generate binuclear complexes
(23, 24), the metal-binding sites typically comprise three
bridging phenolates and three imine donors, while the
tertiary amine remains uncoordinated. Rarely, and from
ostensibly identical reactions, bibrachial complexes have
been isolated (26–29), Scheme 2(b) X ¼ O, although it is
difficult to decidewhether these are the initial product or the
consequence of imine hydrolysis during workup. Although
no analogue cryptand containing three bridging thiophe-
nolates is known, a related acyclic system in which three
thiophenolato groups bridge an iron and a nickel ion shows
that in principle such structures should be accessible (30).
Results and discussion
Scheme 3 illustrates the range of compounds we have
characterised in our attempts to access dinuclear complexes
of the as-yet unknown [2 þ 3] cryptand in Scheme 2(a)
(X ¼ S). We have evidence for the desired 2 þ 3 reactions
and we can report the formation of the [2 þ 2] bibrachial
macrocycle based on tren and thiophenol, Scheme 2(b)
(X ¼ S). Attempts were made to react the dependant
R
R
2+
N
N
SR1
SR1
NH2
NH2
Ag
N
S
O
O
N
N
N
N
+
O
N
H2N
Ag
N
R
SR1
1a, R = Me
1b, R = tBu
N
1c, R = OMe
O
R1 =
N(CH3)2
R
2a R = Me,
2b R = tBu
2c R = OMe
R
R
N
2+
2+
N
S
S
S
N
N
N
N
H2N
S
M
N
N
M
M
N
N
M
NH2
N
N
N
H2
N
H2
R
R
3a M = Cd, R = Me (MACa)
3b M = Cd, R = tBu (MACb)
3c M = Cd, R = OMe (MACc)
4b M = Mn, R = tBu (MACb)
5b M = Ni, R = tBu (MACb)
6b M = Pd, R = tBu (MACb)
Scheme 3. Template reactions of 1a, 1b and 1c and tren in the presence of metal ions to give a [2 þ 3] cryptand without S-deprotection
in the presence of Agþ and [2 þ 2] bibrachial macrocycles with S-deprotection in the presence of Mn2þ and Cd2þ and the
transmetallation reactions of 4b.

Supramolecular Chemistry
607
complexes, 3a–c and 4b, with mono- and dialdehydes such
as benzaldehyde, diformylfuran, isopthaldehyde and
glyoxal as well as with another equivalent of the starting
S-protected dialdehyde, to generate unsymmetric bibra-
chial macrocycles and cryptands, but without success.
assigned to [(2aH-2CON(CH3)2)· ClO4]þ with associated
fragmentation features (Supplementary Information,
1
available online). H NMR spectra showing a clean pair
of 1:1 singlets at 7.52 ppm assigned to the aromatic meta
protons and 8.76 ppm assigned to the imino CH protons.
NOE experiments demonstrated that irradiation into the
imino resonance at 8.76 ppm enhanced the signal from the
meta aromatic protons and a methylene doublet resonance
centred around 3.7 ppm as well as another from the
heavily overlapped aliphatic CH2 cap resonances around
2.4 ppm. This correlates well with observed behaviour of
the imino resonance in the isophthalaldehyde-derived
iminocryptand (33).
Template reactions
Protected [2 þ 3] cryptand
The widespread success of Ag(I) as template in the
assembly of iminocryptands (17–19) via [2 þ 3] Schiff-
base condensation of dialdehydes and triamines made
Ag(I) a natural choice for template cation in attempts to
prepare the simple tren-based cryptand Scheme 2(a)
X ¼ S. Indeed even if the protecting group should be
retained because of the weak capacity of Ag(I) for
promoting its hydrolysis and CZS cleavage, verification
of the macrobicyclic skeleton assembly would represent a
worthwhile target. As usual with silver/base reactions, the
Schiff-base condensation had to be carried out under mild
temperature conditions, with protection from light.
Microcrystalline bronze-coloured products of the conden-
sation reactions between tren and 1b and 1c were best
isolated as triflate salts, with reasonable MeCN solubility
although perchlorate salts were also used. Infrared spectra
showed no NH2 absorptions, and the presence of a broad
high frequency 1660–1690 cm21 tail to the imino CvN
absorption at <1640 cm21 suggests that the CvONMe2
protecting group was still attached to the thiolate function.
The infrared spectrum of the counterion was also present,
with intensity comparable to the ligand absorptions.
The FAB mass spectrometry results (Table S1 of the
[2 þ 2] Bibrachial macrocycle
In the presence of cadmium(II) and manganese(II) using
stoichiometries and reaction conditions appropropriate for
the formation of [2 þ 3] crypts, we isolated dinuclear
complexes of the new [2 þ 2] bibrachial macrocycle
depicted in Scheme 2(b) X ¼ S. Single-crystal X-ray
crystallography, vide infra, shows that the Cd2þ and Mn2þ
complexes, 3c and 4b, have related chemical structures as
drawn in Scheme 3; however, their topologies show
significant differences. The choice of manganese might
seem unusual given the criteria, especially criterion (iv),
listed in the Introduction, and indeed this was not our first
choice when it came to the first row of the transition
metals. We started with Ni(II) for the reasons discussed in
the Introduction. The unexpected success with Mn2þ
apparently stems from the coordinative flexibility of this
ion which by comparison to the success with Cd2þ seems
to be essential. We explored also nickel(II), zinc(II) and
cobalt(II)-templated reactions. Ions consistent with the
respective dimetal complexes of the bibrachial macro-
cycles derived from [2 þ 2] condensations were observed
in otherwise relatively messy ESI-MS of the reaction
mixtures which did not yield tractable solid products.
Subsequently, we were able to access dinickel(II)
complexes via the transmetallation of 4b, vide infra.
In the infrared spectrum, the bright daffodil-coloured
p-OMe dialdehyde (1c)-derived Cd complex, 3c, clearly
shows the presence of nNH2 infrared absorptions at 3275 and
3338 cm21, together with reductionin intensityof the imino
CvN absorption relative to the disilver cryptates. Mass
spectrometric studies explain the origin of these changes. In
the MALDI mass spectrum the base peak, a cluster at m/z
935 in the m/z region 400–1500 corresponds to a
dicadmium complex 3c. We suggest that this is formed
via hydrolysis of one strand of the target cryptand and loss
of a diformylthiophenol fragment [Scheme 3 step (iii)],
rather than directly. The same reaction seems to occur more
slowly with p-Me and p-tBu-dialdehyde thiophenol (1a)
and (1b). Occasionally an amorphous intermediate product
t
Supplementary Information, available online), for the Bu
product, [2b](CF3SO3)2, showed a cluster around m/z
1278.9 as base peak in 400–1800 m/z range corresponding
to the disilver complex of the triply protected macrobi-
cycle, indicating that the silver template strategy was
successful and that the macrobicyclic skeleton had indeed
been assembled as hoped, even though deprotection had
not taken place. Satisfactory microanalytical data were
obtained for [2b](CF3SO3)2 and the p-methoxy analogue
[2c](CF3SO3)2, but we were unable to obtain suitable
crystals for X-ray diffraction studies. It appears that the S
atom is not a donor to the Ag(1) cation. The proposed
structure drawn in Scheme 3 includes a two-coordinate
silver geometry since this arrangement occurs in the
similarly bronze-coloured disilver(I) complex of a
dinucleating imine-based macrocycle which offers very
similar binding sites and topology as those anticipated for
complexes 2b and 2c (31). Of course other options are
possible. The p-Me analogue of these disilver systems, 2a,
the first to be studied, demonstrated cryptand formation
1
via initial H NMR and ESI-MS studies (32). ESI mass
spectra showed a base peak centred at m/z 1109 and

608
A. Lennartson et al.
of the cadmium template reaction of tren with these
dialdehydes precipitates but can be easily redissolved on
adding MeCN. When this intermediate is isolated, notably
it displays neither the strong nNH2 peaks in the IR spectrum
nor the bright daffodil colour associated with 3c. In ESI-MS
spectra, peaks commonly observed at masses correspond-
ing to cryptate species encourage us to believe that this
kinetic product is the desired cryptand complex containing
one or both Cd(II) ions as depicted in Scheme 4.1
Microanalytical and other data of 3c after ‘recrystallisation’
accompanied by a colour change to bright yellow and a
lowering in ethanol solubility of this product demonstrate
the formation of a diaminoethyl-dipendant macrocyclic
complex [3b](ClO4)2, analogous to 3c (Scheme 3).
Reaction of tren, 1b and manganese(II) perchlorate in
1:1:1 or 2:3:2 proportions in refluxing methanol under
nitrogen atmosphere produces red solutions. Evaporation
or layering with diethyl ether gave only red oils. In the ESI
mass spectra of these oils, however, a peak at m/z 386 was
seen which could be assigned to dimanganese(II) complex,
4b. Subsequently the red [4b](ClO4)2(H2O)1.5 crystallised
in a few per cent yield from a very concentrated methanol
solution. After isolation, [4b](ClO4)2(H2O)1.5 was only
sparingly soluble in methanol. The reaction of 4b with one
equivalent of 1b was tried in an attempt at realising an
alternative stepwise route to a dimanganese complex of the
[2 þ 3] crypt. Dissolution of [4b](ClO4)2(H2O)1.5 in
refluxing methanol in the presence of 1b indicated reaction;
however, no tractable product could be isolated from the red
solutions. Thus, the low solubility of [4b](ClO4)2(H2O)1.5
facilitates its isolation, as long as further reaction with 1b
was prevented. In order to achieve this, a methanol solution
of 1b was added very slowly to an excess of manganese
perchlorate and tren; the overall proportions being
tren:1b:Mn at 3:1:3. Under these conditions, the inter-
mediate formation of Mn(II) complex of tren was obvious:
depending on how quickly the 1b was added, occasionally a
white, air sensitive, solid containing tren and perchlorate
precipitated. This rapidly redissolves as 1b is added in the
aforementioned amount. During the course of reaction,
solid red [4b](ClO4)2(H2O)1.5 starts to precipitate and is, in
the solid state, fairly well protected from further reaction
with 1b. Using this strategy, [4b](ClO4)2(H2O)1.5 could be
isolated in 85% yield based on 1b. The selective synthesis
of solid [4b](ClO4)2(H2O)1.5 is thus driven by its
favourable crystallisation. This procedure may explain
why we were unable to use nickel(II) to template the
[2 þ 2] bibrachial macrocyle to give 5b. We subsequently
accessed 5b through the transmetallation of 4b, vide infra
and realised that the perchlorate salt of 4b is far more
soluble. Thus, continuation of reaction of 3b and 4b with
1b, probably to form intractable oligomers, is a likely
explanation for why we do not isolate the [2 þ 3] crypt.
These experimental observations lead us to envisage
alternative pathways for formation of the dimanganese(II)
and dicadmium(II) complexes of the [2 þ 2] bibrachial
macrocyles. The former is based on a 2:2 reaction of a
Mn(II)-tren precursor complex with 1, and the latter is
based on intermediate formation of the [2 þ 3] crypt and
subsequent hydrolysis and cleavage of one thiophenol unit,
Scheme 4. Interestingly, the contrasting single-crystal
X-ray structures of 3a and 4b may even reflect their
formation pathway, vide infra.
Solid samples of [4b](ClO4)2(H2O)1.5 are stable in air
for months. However, solutions in methanol or acetonitrile
turn brown overnight when exposed to air, suggesting
oxidation of 4b or of decomplexed Mn(II). A cyclic
voltammogram of 4b run under an inert atmosphere in
acetonitrile did not reveal any redox processes. This
suggests that oxidation is via a decomplexation route.
Certainly a metal-based oxidation of 4b would require
significant reorientation of the coordination geometry
towards a more ideal octahedral. The difficulty of the
requisite ligand reorganisation may prevent this.
R
3+
N
N
N
S
N
2 H2O
M
3 OH–
N
M
3 1 + 2 tren + 2 Cd2+
,
3 M = Cd
S
N
N
R
3 CO(OH)NMe2, 12H+
S
N
R
O
O
SH
R
R = Me (CRYPTa)
R = tBu (CRYPTb)
R = OMe (CRYPTc)
Scheme 4. Proposed templated formation [2 þ 3] cryptand with S-deprotection followed by its hydrolysis to form [2 þ 2] bibrachial
macrocycle when M ¼ Cd2þ
.

Supramolecular Chemistry
609
s
s
a
a
b
g
m
u

610
A. Lennartson et al.
C8
(b)
(a)
C7A
O1A
C10
C9
C7
C4
C6A
C8A
C9A
C5A
C4A
C3
C5
C11A
C3A
N3A
C1C
C10
N4A
C6
C11
N1
C12A
C2C
C1
C2A
C2
C13A
C12
C17
N3C
C1A
S1A
N2
N4C
N2
Mn1
N1
Cd2
Cd1
C16
C18
C13C
C13B
C1B
S1
C12C
N4B
S1B
C13
N3
C10B
C6B
C12B
N3B
C3B
C4B
C2B
C11B
C9B
C8B
C7B
C14
N4
C5B
O1B
C15
Figure 1. ORTEP plot of the complex cation in (a) 3c and (b) 4b with the crystallographic numbering schemes. Displacement ellipsoids
drawn at the 50% probability level. All H-atoms have been omitted for clarity.
The crystal structures of 3c and 4b are shown in
Figure 1 and the selected bond distance and angles can be
found in Tables 2 and 3. The manganese cation of 4b lies
on a twofold axis passing through the centre of the Mn2S2
ring; there is no crystallographically imposed symmetry in
the cadmium analogue. The topologies of the complexes
are markedly different, although both complexes have the
same N4S2 donor set, each metal ion is coordinated to two
imine donors, a bridgehead tertiary amine, one pendant
primary amine and two deprotected thiolate donors. The
pendant amine nitrogen donors are separated by
sulphur atoms are almost coplanar with the ring in the Mn
˚
complex (displacement 0.03 A), in the Cd analogue the
two independant sulphur atoms are significantly out of the
˚
plane (0.268 and 0.16 A). The imino and thiolate functions
lie on opposite sides of the aromatic ring as often observed
before (2, 3, 35). As shown in Figure 2, the two structures
are related by ‘inversion’ of the pyramidal geometry at one
of the sulphur donors, which also requires rotation of the
thiophenol group about the vector linking the two imino
donors. The observation that the Cd complex retains what
appears to be a more strained geometry might suggest that
forming this compound is more difficult than the
corresponding Mn system.
˚
approximately 3.4 A in each complex, this is much shorter
than the corresponding imine – imine distances, which are
˚
all over 5.8 A. The geometry at the metal ions is probably
A cryptate precursor to the Cd complex can be
envisaged in which a third diiminothiolate strand links the
pendent arms. The cation would then have approximate
threefold symmetry and the sulphur donors would form a
trigonal planar S3 array. In the case of the Mn complex,
addition of a third strand and formation of a trigonal planar
S3 array would require one of the existing diiminothiolate
rings to rotate as described above (or a considerable
lengthening of the MnZimine bonds). The structure of the
Cd complex is consistent with initial formation of a
cryptate, which then loses one link by imine hydrolysis.
Following the development of their elegant synthetic
work on amine-thiophenolate-derived ligands (36–38)
Kersting et al. (39) have prepared dithiophenolate-based
macrocycles in which the remaining six N donors are
derived from two N,N0,N00-trimethyldiethylenetriamine
units. These contain an auxillary carboxylate bridging
group which is contained in the cleft formed by the
best described as irregular six-coordinate with S values
(21, 34) of 1998 and 2078 for the cadmium ions and 1988
for the manganese (accounting for the stabilisation of the
þ II oxidation state). The metal ions are bridged by the
thiolate donors, generating M2S2 rings; for the manganese
complex this ring is almost planar and the angle between
the two MnS2 planes is 16.25(1)8, but in the cadmium
complex these planes are tilted at 51.56(4)8. In
˚
consequence, the CdZCd distance (3.6401(5)A) is shorter
˚
than the MnZMn distance (3.7887(6) A) and the sulphur
donors make angles about the centroid of the cation of
162.48 and 125.98 for Mn and Cd, respectively (Figure 2).
The geometries of the thiophenolato-diimine groups
are also notably different; in the cadmium complex the
imine nitrogen donors are further displaced from the mean
˚
plane of the aromatic ring (by 0.278 and 0.21 A for Mn and
˚
0.368, 0.398, 0.648 and 0.31 A for Cd). Although the

Supramolecular Chemistry
Table 2. Selected bond lengths (A) and angles (8) for 3c. Table 3. Selected bond lengths (A) and angles (8) for 4b.
611
˚
˚
Cd(1)ZN(3C)
Cd(1)ZN(3A)
Cd(1)ZN(3B)
Cd(1)ZN(1)
Cd(1)ZS(1A)
Cd(1)ZS(1B)
Cd(1)ZCd(2)
Cd(2)ZN(4C)
Cd(2)ZN(4B)
Cd(2)ZN(4A)
Cd(2)ZN(2)
Cd(2)ZS(1A)
Cd(2)ZS(1B)
N(3C)ZCd(1)ZN(3A)
N(3C)ZCd(1)ZN(3B)
N(3A)ZCd(1)ZN(3B)
N(3C)ZCd(1)ZN(1)
N(3A)ZCd(1)ZN(1)
N(3B)ZCd(1)ZN(1)
N(3C)ZCd(1)ZS(1A)
N(3A)ZCd(1)ZS(1A)
N(3B)ZCd(1)ZS(1A)
N(1)ZCd(1)ZS(1A)
N(3C)ZCd(1)ZS(1B)
N(3A)ZCd(1)ZS(1B)
N(3B)ZCd(1)ZS(1B)
N(1)ZCd(1)ZS(1B)
S(1A)ZCd(1)ZS(1B)
N(4C)ZCd(2)ZN(4B)
N(4C)ZCd(2)ZN(4A)
N(4B)ZCd(2)ZN(4A)
N(4C)ZCd(2)ZN(2)
N(4B)ZCd(2)ZN(2)
N(4A)ZCd(2)ZN(2)
N(4C)ZCd(2)ZS(1A)
N(4B)ZCd(2)ZS(1A)
N(4A)ZCd(2)ZS(1A)
N(2)ZCd(2)ZS(1A)
N(4C)ZCd(2)ZS(1B)
N(4B)ZCd(2)ZS(1B)
N(4A)ZCd(2)ZS(1B)
N(2)ZCd(2)ZS(1B)
S(1A)ZCd(2)ZS(1B)
Cd(2)ZS(1A)ZCd(1)
N(4B)ZC(11B)ZC(9B)
2.314(3)
N(1)ZMn(1)
N(2)ZMn(1)i
N(3)ZMn(1)
2.2120(15)
2.2277(15)
2.4041(15)
2.2330(17)
2.5827(5)
2.6025(5)
2.2278(15)
2.6026(5)
94.48(6)
124.48(6)
120.77(7)
75.18(5)
73.05(5)
76.33(6)
80.36(4)
147.76(4)
86.96(5)
134.20(4)
133.87(4)
75.91(4)
97.80(5)
138.96(4)
84.865(16)
2.347(3)
2.365(3)
2.529(3)
2.6350(10)
2.6648(10)
3.6401(5)
2.326(3)
2.340(3)
2.363(4)
2.521(3)
2.5890(10)
2.7326(10)
100.21(12)
134.37(12)
97.15(12)
73.86(11)
71.54(11)
72.30(11)
105.13(9)
77.87(8)
119.70(8)
148.53(8)
101.52(9)
152.94(8)
78.81(8)
N(4)ZMn(1)
S(1)ZMn(1)
S(1)ZMn(1)i
Mn(1)ZN(2)i
Mn(1)ZS(1)i
N(1)ZMn(1)ZN(2)i
N(1)ZMn(1)ZN(4)
N(2)iZMn(1)ZN(4)
N(1)ZMn(1)ZN(3)
N(2)iZMn(1)ZN(3)
N(4)ZMn(1)ZN(3)
N(1)ZMn(1)ZS(1)
N(2)iZMn(1)ZS(1)
N(4)ZMn(1)ZS(1)
N(3)ZMn(1)ZS(1)
N(1)ZMn(1)ZS(1)i
N(2)iZMn(1)ZS(1)i
N(4)ZMn(1)ZS(1)i
N(3)ZMn(1)ZS(1)i
S(1)ZMn(1)ZS(1)i
Note: Symmetry code: (i) 2x þ 2, y, 2z þ 1/2
130.37(8)
81.04(3)
recognised to be more stable, against Lewis-acid-assisted
hydrolytic cleavage of the imine bond, than other
iminocryptands (25), some of which are of little value as
ligands for cations of charge higher than þ1. In contrast,
even tripositive cations are quite stable within phenolate
iminocryptands (21, 22). This stabilisation presumably
derives from diminution of cation Lewis acid activity
consequent on coordination by the high-charge density
RO2 donor, which cannot be matched by the larger low-
charge density RS2 group. Thus, a near-normal hydrolytic
sensitivity of iminocryptands can be expected to apply, and
aminoethyl dipendants may result from such hydrolysis.
120.81(11)
113.14(13)
101.58(12)
73.07(11)
73.37(11)
73.10(11)
125.37(8)
106.47(8)
79.54(8)
151.85(8)
83.37(9)
78.54(8)
159.28(9)
125.63(8)
80.59(3)
Transmetallations
88.34(3)
129.3(4)
Despite the considerable success using Ni(II) as the
templating ion in the construction of the Schiff-base
thiophenolate macrocycles using diamines, Scheme 1(b)
X ¼ S (2–6), its failure to act as template for the tren-based
crypts and bibrachial macrocycles seems now to be
explicable: the bibrachial system can react further. Whether
or not this occurs depends on which metal ion is bound, and
in which geometry. The Schiff-base condensation of the
bibrachial with one more equivalent of 1 will lead either to
ring closure and the crypt or (intractable) oligomers.
Whether or not the crypt is stable will in turn also depend on
the choice of metal ion. However, dinickel(II) and
dipalladium(II) [2 þ 2] bibrachial macrocyclic complexes
were accessible via transmetallation of 4b with Ni(ClO4)2 or
PdCl2(MeCN)2, Scheme 3. The complexes were crystallised
only after manganese perchlorate and excess nickel or
palladium salts had been extracted from the product using
V-shaped Mn2S2 ring units. The corresponding cleft is
much narrower in the case of 4b because the pendant
primary amine donors from the unreacted tren arms are
located on the outside faces of the complex, thus squeezing
the thiophenolate units closer together, not far from
coplanarity, leaving no room for an auxillary bridging
ligand.
As mentioned in the Introduction, the formation of
bibrachial derivatives of Robson-type macrocycles derived
from tren, Scheme 2(b) X ¼ O, is very rare. It seems that
the tren-based cryptands, Scheme 2(a) X ¼ O, are favoured
for the phenol linker. This study indicates the opposite
trend for the thiophenolate-based systems which have not
been investigated previously. Phenolate cryptands are

612
A. Lennartson et al.
Figure 2. Views of 3c and 4b parallel to the phenyl ring planes.
water. Similarly, transmetallations using more than 3
equivalents of cadmium(II) or cobalt(II) perchlorate also
gave solid products for which ESI mass spectra showed
peaks at m/z 444 and 389, respectively, which can be
assigned to [Cd2(MACb)]2þ and [Co2(MACb)]2þ. If
4b(ClO4)2 and Cd(ClO4)2 or PdCl2(MeCN)2 were mixed
in acetonitrile in a 1:1 ratio at ambient temperature, ESI
mass spectra recorded after approximately 1 h show 4b, the
mixed Mn(II)–M(II) complex [MnM(MACb)]2þ M ¼ Cd,
Pd and the M(II) complex in close to 25%:50%:25%
proportions (Figure 3 and Supplementary Information,
available online). In the case of transmetallation with Ni(II),
approximately equimolar proportions of the homodinuclear
complexes appear only to be present under the same
conditions.
Single crystals of the nickel complex were obtained
from a mixture of N,N-dimethylformamide (DMF)
and methanol, giving crystals of [Ni2(MACb)](ClO4)2
(H2O)(DMF), 5b, and the palladium complex could be
recrystallised from a mixture of acetonitrile and acetone to
give crystals of [Pd2(MACb)](ClO4)2(H2O)5, 6b. When the
nickel complex was recrystallised from aqueous aceto-
nitrile, crystals of poor quality isomorphous with 6b are
obtained. The complex cations in compounds 5b and 6b are
very similar, Figure 4. Selected bond distance and angles
can be found in Tables 4 and 5. Both complexes show two
Figure 3. ESI-MS (CH3CN) of the 1:1 mixture of 4b and Cd(ClO4)2 recorded 1 h. [Mn2(MACb)]2þ, [MnCd(MACb)]2þ and
[Cd2(MACb)]2þ are present in close to 25%:50%:25% proportions.

Supramolecular Chemistry
613
C10
(a)
(b)
C8
C9
C27
C26
C10
C28
C7
C4
C28
C8
C25
C23
C22
C36
C7
C26
C5
C23
C17
C3
C9
C25
C24
C5
C18
C24
C1
C1
C4
N4
C19
C6
N1
C2
C18
C27
C36
C13
N2
C17
C12
C17
C6
C19
C35
N4
C22
C21
N8
C20
N8
C21
C14
C20
C2
C16
C3
C35
C34
N5
Ni2
C1
Ni1
Pd2
C29
S1
N2
C30
S2
C31
S2
C13
C34
C15
C29
N7
N3
N6
C11
Pd1
N1
C12
N5
S1
C14
C32
N3
C30
C31
N7
C32
C15
C33
C33
Figure 4. ORTEP plot of the complex cation in (a) 5b and (b) 6b showing the crystallographic numbering scheme, with displacement
ellipsoids drawn at the 50% probability level. All H-atoms have been omitted for clarity.
crystallographically independant metal atoms having
slightly distorted square planar coordination geometries.
The coordination spheres around the two metal atoms in 5b
and 6b are identical, with each metal atom coordinating one
sulphur atom, one imine nitrogen atom, one tertiary amine
nitrogen atom and one primary amine nitrogen atom. This
leaves two of the imine nitrogen atoms non-coordinating.
The macrocyclic ligand in 5b and 6b is bent to make the
two aromatic rings nearly parallel, but slightly rotated
against each other to avoid eclipse of the bulky tert-butyl
groups. The distance between the centroids of the two
of face-to-face p–p interaction between the two groups.
The crystal structures of 5b and 6b form extensive
hydrogen-bonded networks. Interestingly enough, the unit
cell dimensions are very similar and so are the coordinates
for the complex cations and perchlorate anions. The water
molecule corresponding to O10 in 5b has an analogue O9 in
the crystal structure of 6b. The two water molecules are
oriented in the same fashion in the two unit cells. However,
the DMF molecule in the structure of 5b is replaced by a
cluster of four water molecules in 6b. Thus, the cluster of
water molecules can mimic the shape and hydrogen-
bonding scheme of a DMF molecule and preserve the
overall structure, Figure 5.
˚
aromatic rings in 5b is 3.731(2) A, indicating some degree
˚
Table 4. Selected bond lengths (A) and angles (8) for 5b.
˚
Table 5. Selected bond lengths (A) and angles (8) for 6b.
N(2)ZNi(2)
N(3)ZNi(2)
N(4)ZNi(2)
N(6)ZNi(1)
N(7)ZNi(1)
N(8)ZNi(1)
S(1)ZNi(1)
S(2)ZNi(2)
N(8)ZNi(1)ZN(7)
N(8)ZNi(1)ZN(6)
N(7)ZNi(1)ZN(6)
N(8)ZNi(1)ZS(1)
N(7)ZNi(1)ZS(1)
N(6)ZNi(1)ZS(1)
N(4)ZNi(2)ZN(3)
N(4)ZNi(2)ZN(2)
N(3)ZNi(2)ZN(2)
N(4)ZNi(2)ZS(2)
N(3)ZNi(2)ZS(2)
N(2)ZNi(2)ZS(2)
1.979(2)
1.944(3)
1.862(2)
1.974(2)
1.946(2)
1.862(2)
2.1392(8)
2.1435(8)
171.22(11)
87.72(10)
86.34(10)
97.14(7)
88.82(8)
175.14(7)
170.51(11)
87.85(10)
86.07(10)
97.87(7)
88.45(8)
174.06(7)
N(2)ZPd(2)
N(3)ZPd(2)
N(4)ZPd(2)
N(6)ZPd(1)
N(7)ZPd(1)
N(8)ZPd(1)
S(1)ZPd(1)
S(2)ZPd(2)
N(8)ZPd(1)ZN(7)
N(8)ZPd(1)ZN(6)
N(7)ZPd(1)ZN(6)
N(8)ZPd(1)ZS(1)
N(7)ZPd(1)ZS(1)
N(6)ZPd(1)ZS(1)
N(4)ZPd(2)ZN(3)
N(4)ZPd(2)ZN(2)
N(3)ZPd(2)ZN(2)
N(4)ZPd(2)ZS(2)
N(3)ZPd(2)ZS(2)
N(2)ZPd(2)ZS(2)
2.092(3)
2.073(3)
1.976(3)
2.094(3)
2.064(3)
1.973(3)
2.2484(9)
2.2507(9)
168.40(13)
85.92(12)
84.49(12)
96.64(9)
93.07(10)
177.30(9)
168.35(14)
85.71(12)
84.42(13)
95.89(9)
93.96(10)
178.38(8)

614
A. Lennartson et al.
a
a
0
0
b
b
Figure 5. Packing in 5b and 6b showing that a DMF molecule in the former and four water molecules in the latter occupy the same
position so that packing of dinickel(II) and dipalladium(II) complexes is similar.
Conclusions
diformyl-4-methylphenyl)dimethylcarbamothioate (1a)
and S-(2,6-diformyl-4-tert-butylphenyl)dimethylcarba-
mothioate (1b) were prepared using published procedures
(2). CAUTION! Although we encountered no problems
during our work, perchlorate salts of metal complexes are
potentially explosive and should be handled with caution
in small quantities.
The template and transmetallation experiments described
here emphasise that the thiophenol-derived imino ligand
has relatively greater hydrolytic sensitivity than the
phenolate analogue. Even the relatively weak Lewis acid
Cd2þ carries out this hydrolytic reaction with ease,
although it appears to take place perceptibly more slowly
on the p-tBu and p-Me substituted thiophenolate systems.
The soft Lewis acid Ag(I), however, fails to effect any
hydrolysis of these systems, and although in the absence of
suitable crystals for X-ray diffraction studies it is not
possible to confirm the donor set of the cation, it seems
likely that the still protected thiolate function is not among
these. Comparison of structures 3c, 4b, 5b and 6b
illustrates the possibly desirable flexibility achievable by
dipendant systems. Structure 4b adopts a topology in
which the two bridging thiolate donors lie syn, in contrast
to 3c in which they lie anti to each other. Structures 5b and
6b do not utilise bridging thiolate, instead relying on
terminal thiolate coordination to achieve the low-spin
coordination set preferred by d8 cations in strong field.
This close juxtaposition of potentially catalytically active
d8 sites may prove to be of value in future development of
this chemistry.
2-Hydroxy-5-methoxybenzene-1,3-dicarbaldehyde, 1c
2,6-Diformyl-4-methoxyphenol was converted to 1c by a
modification of the technique of Newman and Karnes (40),
as follows.
To a cooled stirred suspension of 2,6-diformyl-4-
methoxyphenol (20 g, 110 mmol) in dry DMF (dried by
standing over freshly activated sieves under dry N2 for a
few days and storing over newly activated sieves after
distillation) was added 60% sodium hydride in mineral oil
(4.88 g, 122.2 mmol). The ice bath was removed after
evolution of hydrogen ceased, and the orange mixture
was allowed to warm to ambient temperature, when a
solution of N,N-dimethylthiocarbamylchloride (16.14 g,
133.2 mmol) in dry DMF was added and the mixture
stirred for 24 h at 358C. Water 1 dm3 was added, generating
a buff-coloured precipitate, which was filtered under
vacuum and washed with cold water (2 £ 100 cm3). The
precipitate was extracted with CH2Cl2 and dried over
MgSO4 before removing the solvent. Flash chromatog-
raphy was then applied on a silica column, using a
concentrated CH2Cl2 solution, which was eluted with 9:1
CH2Cl2 –hexane to which 1% MeOH was added
(Rf < 0.3). The resulting white crystals were obtained in
68% yield. This product (15 g, 56.1 mmol) was stirred
rapidly under N2 at 1808C for 20 min and allowed to cool
Experimental
All experiments were carried out under nitrogen
atmosphere using Schlenk techniques. Methanol and
acetonitrile were degassed by bubbling with a stream of
dry nitrogen. Tris(2-aminoethyl)amine, manganesese(II)
perchlorate hydrate, nickel(II) perchlorate hexahydrate
silver triflate, cadmium perchlorate and bis(acetonitrile-
palladium(II) chloride were used as received. S-(2,6-

Supramolecular Chemistry
615
to ambient temperature under N2. Flash chromatography
was then applied to the solution in CH2Cl2 on before
(Rf < 0.1). Rotary evaporation of the combined appro-
priate fractions yielded 11.1 g off-white solid in 74% yield,
which could be purified by trituration with cold ether
generating 10.2 g white crystals, which were collected in
vacuum filtration and dried in vacuo.
protected p-methoxy, p-tBu- or p-methyl-thiophenoldial-
dehyde (0.12 mmol) is added dropwise under inert
atmosphere, generating a yellow colour, which intensified
somewhat on refluxing for 1 h. In the case of 3c, the
solution was allowed to stand overnight under N2 and the
bright yellow solid filtered off in 80% yield was
recrystallised from EtOH–MeCN mixture by slow
evaporation of solvent in air to give crystals, suitable for
X-ray diffraction study. For the analogous reactions of tren
with 1b and 1a on Cd(II), recrystallisation of the ill-
defined initial products from MeCN–EtOH gave 3b and
3a. 3c: anal. calc. for Cd2C34H54S2Cl2N8O13: C, 36.30; H,
4.83; N, 9.66. Found: C, 36.39; H, 4.30; N, 9.66. IR (KBr):
3339, 3275 m, 1642 ms, 1597 m, 1310 m, 1090 s, 1034 mw,
625 m. MALDI: 935 (100%, [Cd2(MACc)(ClO4)]þ). 3b:
anal. calc. for Cd2C36H60S2Cl2N8O11: C, 37.93; H, 5.30;
N, 9.83. Found: C, 37.86; H, 5.14; N, 9.76. IR (KBr): 3290,
3289 m, 1636 ms, 1542 m, 1090 s, 625 m. ESI-MS
(CH3CN, m/z) of product formed via the transmetallation
route from 4b: 444.1017 (100%, [Cd2(MACb)]2þ),
987.1431 (30%, [Cd2(MACb)(ClO4)]þ). 3a: anal. calc.
for Cd2C32H48N8S2Cl2O9: C, 36.66; H, 4.61; N, 10.69.
Found: C, 36.21; H, 4.05; N, 10.81. ESI-MS (MeCN m/z)
402, (95%, [Cd2(MACa)]2þ; 437, (100%, [Cd2(MACa)
DMF]2þ; 903, (22%, [Cd2(MACa)ClO4]þ.
[Ag2(CRYPTb 1 3CONMe2)](CF3SO3)2 · (H2O)3 (2b)
and [Ag2(CRYPTc 1 3CONMe2)](CF3SO3)2 · (EtOH)2
(2c)
Ag(CF3SO3) (26 mg, 0.1 mmol) in 5 cm3 MeOH was added
to deoxygenated p-methoxo- (40 mg, 0.15mmol) or p-tBu-
thiophenoldialdehyde (42 mg, 0.15mmol) dimethylcarba-
mothioates followed by dropwise addition of tren (15 mg,
0.1 mmol) in 5 cm3 MeOH with good stirring, under N2
protection at 508C. The addition of the amine was
accompanied by development of a bronze-yellow colour,
and after addition the mixture was lowered to ambient
temperature and solvent reduced under a good flow of N2 in
the dark over several hours. The acetonitrile-soluble dark
bronze solids were obtained which was washed well with
MeOH. 2b(CF3SO3)2 · 3H2O: Yield 70mg, 90%. Anal. calc.
for Ag2C59H87S5N11O12F6: C, 43.42; H, 5.37; N, 9.44.
Found: C, 43.27; H, 5.13; N, 9.81. FAB–MS: m/z 1278.9
(100%, inthemassrange abovem/z 1000[Ag2(CRYPTb þ 3
CONMe2)]þ); IR (KBr): 3401ms, 2956, 2934m, 1665mw,
1631ms, 1540, 1363m, 1252, 1223, 1108, 1028s, 639 mw.
FAB–MS data for the perchlorate salt of2b are also supplied
in supplementary material. 2c(CF3SO3)2 · 2EtOH: 60mg,
90%. Anal. calc. for Ag2C54H75S5N11O11F6: C, 38.8; H, 4.5;
N, 9.9; S, 10.4. Found: C, 39.2; H, 4.4; N, 9.7; S, 10.2.
MALDI-MS: 1426, (100%, in the range above m/z 1000
[Ag2(CRYPTc þ 3H þ 3CONMe2) · DMF· (CF3SO3)]þ);
IR (KBr): 3422br, 1636ms, 1591m, 1542m, 1457m,
1369m, 1258s, 1160ms, 1031s, 639 mw.
In all three cadmium products made by the attempted
template route, mass spectroscopic evidence was
obtained, in the initial products at least, for the presence
of remnant cryptate species. Thus, solvated or aquated
clusters, such as [H(CRYPTb)· H2O]þ, [Cd(CRYPTa)]þ
or [Cd2CRYPTa · H2O]þ indicative of the intact or
monohydrolysed crypt, are seen with significant intensity
(Supplementary Information, available online). These
appear to be transient rather than stable products capable
of isolation.
The reaction carried out using 1a gave an initial sticky
product for which elemental analysis is more consistent
with formation of the dicadmium complex of the cryptand,
[Cd2(CRYPTa)](ClO4)2 than the bibrachial macrocylic
complex 3a. Anal. calc. for [Cd2(CRYPTa)](ClO4)2,
Cd2C39H49N8S3Cl2O10: C, 39.66; H, 4.18; N, 9.49.
Found: C, 39.53; H, 4.44; N, 9.59. Attempted recrystalli-
sation of this product from MeOH resulted in dihydrolysis
and release of one thiophenolate linker yielding 3a.
[Ag2(CRYPTa þ 3CONMe2)](ClO4)2 (2a) was pre-
pared similarly using AgNO3 as the starting source for
Ag(I) and perchlorate for precipitation; however, elemen-
tal analyses and IR spectra indicated the presence of both
perchlorate and nitrate in the isolated solid product. This
was not pursued. Supplementary material provides FAB–
MS data of a sample partly deprotected using ground
NaOH and gave ions appropriate for the formation of a
disilver complex of the mono- or diprotected crypt.
[Mn2(MACb)](ClO4)2(H2O) (4b)
Manganese(II) perchlorate hexahydrate (2.16 g, 6 mmol)
was dissolved in 100 ml degassed methanol under nitrogen
atmosphere. Tris(2-aminoethyl)amine (0.90 ml, 6 mmol)
was added, and the solution was heated to reflux under
stirring. A solution of S-(4-(tert-butyl)-2,6-diformylphe-
nyl) dimethylcarbamothioate (590 mg, 2 mmol) in 100 ml
degassed methanol was added dropwise over 7 h, where-
upon an orange precipitate formed. The mixture was
[Cd2(MACc)](ClO4)2(EtOH)2 (3c)
[Cd2(MACb)](ClO4)2(H2O)3 (3b),
[Cd2(MACa)](ClO4)2(EtOH) (3a)
Cd(ClO4)2·2H2O (0.1 mmol) in 5 cm3 MeCN and tren
(0.1 mmol) were mixed and the mixture deoxygenated
with N2. Then a deoxygenated alcoholic solution of

616
A. Lennartson et al.
˚
refluxed for 1 h and allowed to cool to ambient temperature
and the solution was left at ambient temperature overnight.
The volume was reduced to c. 50 ml in vacuo. The mixture
was filtered, the precipitate washed with methanol (3 ml)
and dried by suction. Yield: 0.83 g (85%). IR (KBr): 3429
(br), 2959 (s), 2866 (m), 1631 (s), 1536 (m), 1466 (m),
1394 (m), 1363 (w), 1281 (w), 1229 (m), 1107 (s), 990
(m), 890 (m), 815 (w), 739 (w), 688 (w), 624 (m), 510 (w)
cm21. Anal. calc. for Mn2C36H56S2Cl2N8O9: C, 43.68; H,
5.70; N, 11.32. Found: C, 43.18; H, 5.41; N, 11.08. ESI-
MS (CH3CN): 386.1261(100%, [Mn2(MACb)]2þ),
871.2226 (30% [Mn2(MACb)(ClO4)]þ).
(l ¼ 0.71073 A). Empirical absorption corrections were
applied using the SADABS program. Crystal structures
were solved using the programs SIR-92 (41) or TREF (42)
and refined using full-matrix least squares calculations on
F 2 using the SHELXL-97 (41) program. All non-hydrogen
atoms were refined anisotropically, and C-bound hydrogen
atoms were included in calculated positions and refined
using a riding model. Structures were drawn using
SHELXTL (41) or ORTEP3 (43) for Windows and
PLUTON (44) SIR-92, SHELXL-97 and ORTEP3 were
enclosed in the WinGX-software package (45).
Supplementary Information
[Ni2(MACb)](ClO4)2(H2O)3 (5b)
Crystal structures have been deposited with the CCDC
with entry numbers: 87,4623–87,4626. Supporting
information includes FAB–MS data for 2a and 2b and
the ESI-MS data for transmetallation reactions and for
initial reaction products from the preparations of 3a and
3b.
[Mn2(MACb)](ClO4)2(H2O) (49 mg, 0.05 mmol) and
Ni(ClO4)2 £ 6H2O (73 mg, 0.20mmol) were dissolved in
acetonitrile(6 ml). After 1 h, the solution was evaporatedand
water (2 ml) was added. The mixture was filtered and the
brown precipitate was washed with water (2 £ 2 ml). The
precipitate was dissolved in dimethylformamide (0.5ml)
and methanol(7 ml)was added. Thesolutionwas evaporated
slowly in a vial, whereupon brown crystals formed. The
product was dried in vacuo. Yield: 52mg (98%). IR (KBr):
3435 (br), 2961 (m), 1657 (s), 1630 (s), 1530 (m), 1463 (m),
1437 (m), 1392 (m), 1362 (m), 1277 (w), 1237 (m), 1144 (s),
1091 (s), 921 (w), 884 (w), 752 (w), 728 (w), 661 (w), 627 (s)
cm21. Anal. calc. for Ni2C36H60S2Cl2N9O11: C, 41.84; H,
5.85; N, 10.84. Found: C, 41.39; H, 5.67; N, 11.34. ESI-MS
(CH3CN): 390.1343 (100%, [Ni2(MACb)]2þ).
Acknowledgements
We thank the Danish Council for Independant Research –
Natural Sciences for financial support (Grant No. 272-08-0436).
We acknowledge the early contributions of Joanne Coyle, PhD
thesis, Open University, 1999 and Jefferson Revell, undergradu-
ate project work, University of Coventry, 2000, for spectroscopy
of the silver templated para-methyl thiophenolate cryptand and
synthesis of the para-methoxy thiophenoldialdehyde (3c),
respectively. The assistance of the EPSRC National Mass
Spectrometry service at University of Swansea is acknowledged
in respect of the silver and cadmium mass spectra reported here.
[Pd2(MACb)](ClO4)2(H2O)5 (6b)
[Mn2 (MACb)](ClO4)2(H2O) (49 mg, 0.05 mmol) and
[PdCl2(MeCN)2] (39 mg, 0.15 mmol) were dissolved in
40 ml acetonitrile. After 2 h, the solution was evaporated
and water (2 ml) was added. The brown precipitate was
filtered and washed with water (2 ml). The precipitate was
dissolved in 35 ml acetone and 35 ml acetonitrile, the
solution filtered and evaporated in a beaker. Yield: 51 mg
(88%). IR (KBr): 3434 (br), 2961 (s), 1631 (s), 1526 (w),
1464 (m), 1394 (w), 1363 (w), 1274 (w), 1234 (m), 1121
(s), 954 (w), 751 (w), 726 (w), 625 (s) cm21. Anal. calc.
for Pd3C36H60S2Cl5N8O7 (bulk seems contaminated by
heavy species, thus one counter perchlorate ion was
replaced with [PdCl4]22 and the bulk formulated [Pd2
(MACb)](ClO4)2(H2O)3): C, 34.53; H, 4.35; N, 8.95.
Found: C, 34.28; H, 4.67; N, 8.24. ESI-MS (CH3CN):
438.1069 (100%, [Pd2 (MACb)]2þ).
Note
1. Indeed the main product obtained on reaction of tren with 1a
appears to be a lightly hydrolysed (opened at one imine
function) form of the cryptate. The dependant macrocyclic
complex 3a was isolated on recrystallisation of this product
from MeCN–EtOH.
References
(1) Pilkington, N.H.; Robson, R. Aust. J. Chem. 1970, 23,
2225–2236.
(2) Christensen, A.; Jensen, H.S.; McKee, V.; McKenzie, C.J.;
Munch, M. Inorg. Chem. 1997, 36, 6080–6085.
(3) Christensen, A.; Mayer, C.; Jensen, F.; Bond, A.D.;
McKenzie, C.J. Dalton Trans. 2006, 35, 108–120.
(4) Branscombe, N.D.J.; Blake, A.J.; Martin-Becerra, A.;
Li, W.S.; Parsons, S.; Ruiz-Ramirez, L.; Schroder, M.
Chem. Commun. 1996, 2573–2574.
(5) Brooker, S.; Caygill, G.B.; Croucher, P.D.; Davidson, T.C.;
Clive, D.L.J.; Magnuson, S.R.; Cramer, S.P.; Ralston, C.Y.
Dalton Trans. 2000, 3113–3121.
(6) Brooker, S. Coord. Chem. Rev. 2001, 222, 33–56.
(7) McKenzie, C.J.; Robson, R. Inorg. Chem. 1987, 26,
3615–3621.
X-ray crystallography
The crystals were mounted on glass fibres and data were
collected using a Bruker-Nonius APEX II diffractometers
with graphite monochromated Mo-Ka radiation

Supramolecular Chemistry
617
(8) Bond, A.M.; Haga, M.A.; Creece, I.S.; Robson, R.;
Wilson, J.C. Inorg. Chem. 1988, 27, 712–721.
(9) Bond, A.M.; Haga, M.A.; Creece, I.S.; Robson, R.;
Wilson, J.C. Inorg. Chem. 1989, 28, 559.
(10) Iliopoulos, P.; Murray, K.S.; Robson, R.; Wilson, J.C.;
Williams, G.A. J. Chem. Soc., Dalton Trans. 1987,
1585–1591.
(11) Brooker, S.; Croucher, P.D.; Davidson, T.C.; Dunbar, G.S.;
Beck, C.U.; Subramanian, S. Eur. J. Inorg. Chem. 2000,
169–179.
(12) Brooker, S.; Croucher, P.D.; Roxburgh, F.M. J. Chem. Soc.,
Dalton Trans. 1996, 3031–3037.
(24) Drew, M.G.B.; Harding, C.J.; McKee, V.; Morgan, G.
J. Chem. Soc., Chem. Commun. 1995, 10, 1035–1038.
(25) Harding, C.J.; Lu, Q.; Malone, J.F.; Mars, D.J.; Martin, N.;
McKee, V.; Nelson, J. J. Chem. Soc., Dalton Trans. 1995,
1739–1747.
(26) Wang, J.; Xu, X.; Chen, J.; Luo, Q.; Shen, M.; Huang, X.;
Wu, Q. Inorg. Chim. Acta 1997, 256, 121–124.
(27) Xu, X.Y.; Luo, Q.H.; Shen, M.C.; Huang, X.Y.; Wu, Q.J.
Polyhedron 1997, 16, 915–919.
(28) Xu, X.Y.; Luo, Q.H.; Shen, M.C.; Huang, X.Y.; Wu, Q.J.
J. Coord. Chem. 1997, 42, 25–32.
(29) Xu, X.Y.; Wang, Z.L.; Luo, Q.H.; Shen, M.C.; Zhou, X.G.;
Zhou, Z.Y. J. Coord. Chem. 1998, 43, 81–88.
(30) Steinfeld, G.; Kersting, B. Chem. Commun. 2000, 205–206.
(31) McKenzie, C.J.; Nielsen, L.P.; Sotofte, I. Acta Chem.
Scand. 1998, 52, 571–577.
(32) Coyle, J.L. Ph.D. thesis, Dicopper cryptands as biosite
mimics, Open University, UK, 1999.
(33) McKee, V.; Robinson, W.T.; McDowell, D.; Nelson, J.
Tetrahedron Lett. 1989, 30, 7453–7456.
(34) McCrea, J.; McKee, V.; Metcalfe, T.; Tandon, S.S.;
Wikaira, J. Inorg. Chim. Acta 2000, 297, 220–230.
(35) Kruger, P.E.; McKee, V. J. Chem. Soc., Chem. Commun.
1997, 1341–1342.
(36) Kersting, B.; Siebert, D.; Volkmer, D.; Kolm, M.J.;
Janiak, C. Inorg. Chem. 1999, 38, 3871–3882.
(37) Kersting, B.; Steinfeld, G.; Fritz, T.; Hausmann, J.
Eur. J. Inorg. Chem. 1999, 2167–2172.
(38) Kersting, B.; Steinfeld, G. Chem. Commun. 2001,
1376–1377.
(39) Journaux, Y.; Glaser, T.; Steinfeld, G.; Lozan, V.;
Kersting, B. Dalton Trans. 2006, 1738–1748.
(40) Newman, M.S.; Karnes, H.A. J. Org. Chem. 1966, 31,
3980–3984.
(13) Atkins, A.J.; Blake, A.J.; Schroder, M. J. Chem. Soc.,
Chem. Commun. 1993, 1662–1665.
(14) Hoskins, B.F.; McKenzie, C.J.; Robson, R.; Zhenrong, L.
J. Chem. Soc., Dalton Trans. 1990, 2637–2641.
(15) Hoskins, B.F.; McKenzie, C.J.; MacDonald, I.A.S.;
Robson, R. J. Chem. Soc., Dalton Trans. 1996, 2227.
(16) (a) See, for example: Iwata, S.; Ostermeier, C.; Ludwig, B.;
Michel, H. Nature 1995, 376, 600–669; (b) Volbeda, A.;
Charon,M.H.;Piras,C.;Hatchikan,E.C.;Frey,M.;Fontecilla-
Camps, J.C. Nature 1995, 373, 580–587; (c) Frey, M. Str.
Bond 1998, 90, 97–126; (d) Cammack, R. Nature 1998, 397,
214–215.
(17) Nelson, J.; McKee, V.; Morgan, G.G. Prog. Inorg. Chem.
1998, 47, 167–316.
(18) McKee, V.; Nelson, J.; Speed, D.; Town, R.M. J. Chem.
Soc., Dalton Trans. 2001, 3641–3646.
(19) Coyle, J.; Nelson, J.; McKee, V. J. Chem. Soc., Chem.
Commun. 1998, 709–710.
(20) Morgan, G.; McKee, V.; Nelson, J. Inorg. Chem. 1994, 33,
4427–4428.
(21) Drew, M.G.B.; Howarth, O.W.; Harding, C.J.; Martin, N.;
Nelson, J.; McKee, V. J. Chem. Soc., Chem. Commun.
1995, 903–906.
(41) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Gualardi, A.
J. Appl. Cryst. 1993, 26, 343–350.
(22) Drew, M.G.B.; Howarth, O.W.; Martin, N.; Nelson, J.
J. Chem. Soc., Dalton Trans. 2000, 1275–1278.
(23) Drew, M.G.B.; McKee, V.; Morgan, G.; Nelson, J. J. Chem.
Soc., Dalton Trans. 1994, 3149–3158.
(42) Sheldrick, G.M. Acta Cryst. 2008, A64, 112–122.
(43) Farrugia, J. J. Appl. Cryst. 1997, 30, 565.
(44) Spek, A.L. Acta Cryst. 1990, A46, 194–201.
(45) Farrugia, J. J. Appl. Cryst. 1999, 32, 837–838.