Heterometallic gold(i)-thallium(i) compounds with crown thioethers

Article abstract of DOI:10.1039/c3dt51334c

The polymeric Au/Tl compounds [{Au(C₆X₅) ₂}Tl]_n (X = Cl, F) react with the crown thioethers 1,4,7-trithiacyclononane ([9]aneS₃), 1,5,8,11- tetrathiacyclotetradecane ([14]aneS₄), and 1,4,7,1

Full text of DOI:10.1039/c3dt51334c

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Heterometallic gold(I)–thallium(I) compounds with
crown thioethers†‡
Cite this: Dalton Trans., 2013, 42, 11559
Alexander J. Blake,a Rocío Donamaría,b Eduardo J. Fernández,b Tania Lasanta,b
Vito Lippolis,*c José M. López-de-Luzuriaga,*b Elena Manso,b Miguel Mongeb and
M. Elena Olmosb
The polymeric Au/Tl compounds [{Au(C6X5)2}Tl]n (X = Cl, F) react with the crown thioethers 1,4,7-trithia-
cyclononane ([9]aneS3), 1,5,8,11-tetrathiacyclotetradecane ([14]aneS4), and 1,4,7,10,13,16,19,22-octathia-
cyclotetracosane ([24]aneS8) in an appropriate molar ratio to afford [{Au(C6X5)2}Tl(L)]2 [L = [9]aneS3,
X = Cl (1), F (4); L = [14]aneS4, X = Cl (2), F (5)], [{Au(C6Cl5)2}2Tl2([24]aneS8)]n (3) or [{Au(C6F5)2}2Tl2-
([24]aneS8)] (6). X-ray diffraction studies of 3, 4 and 6 reveal polymeric (3) or tetranuclear (4, 6) structures
formed via Tl–S bonds and Au⋯Tl or Au⋯Tl and Au⋯Au contacts. All the complexes are luminescent in
the solid state, but not in solution, where the metal–metal interactions, which are responsible for the
luminescence, are no longer present. DFT calculations on representative model systems of complexes 3, 4
and 6 have also been carried out in order to determine the origin of the electronic transitions responsible
for their optical properties.
Received 22nd May 2013,
Accepted 15th June 2013
DOI: 10.1039/c3dt51334c
www.rsc.org/dalton
optical properties.6 Besides, this luminescence strongly
depends on the structural arrangement of the metals, on their
Introduction
The chemistry of heteropolynuclear extended supramolecular environment and on the nature of the ligands present in the
systems built by secondary interactions has attracted physi- complex.7
cists’ and chemists’ interest in recent years. This is mainly due
One of the synthetic methods usually employed in our
to the particular characteristics of the chemical bonding in laboratory for the preparation of heteropolynuclear extended
these systems and to the physical and chemical properties systems is the acid–base strategy, consisting of the treatment
associated with them.1 As is well known, aurophilic inter- of electron-rich neutral molecules or anions with Lewis
actions, which are comparable in strength to hydrogen bonds,2 acids. These reactions lead to an enormous variety of struc-
are the most widely studied nonbonding contacts between tural arrangements, which range from discrete molecules to
closed-shell metal ions. Nevertheless, over the past few years, extended structures of varying dimensionality.8 In particular,
complexes with metallophilic interactions between gold(I) we have extensively employed bis(perhalophenylaryl)aurate(I)
centres and other closed-shell metal atoms (Au(I)⋯M) have salts as basic precursors and closed-shell metal ions such
attracted renewed attention due to their theoretical aspects,3 as Ag(I)9 or Tl(I)6,7 as the acidic ones. For example, treatment of
photophysical properties4 and potential applications.5 In par- NBu4[Au(C6X5)2] (X = Cl, F) with TlPF6 affords the compound
ticular, the luminescence observed in these compounds seems [{Au(C6X5)2}Tl]n, featuring one-dimensional polymeric chains
to be closely related to the presence of metal⋯metal inter- with alternating gold and thallium atoms5a,7c,10 which further
actions, which have in many cases been implicated in these react with different electron donor molecules that may interact
with the thallium centres. In previous work, we have reported
that the changing of the halogens in the bis(perhalophenyl-
aryl)aurate(I) precursor modifies the donor properties of the
aSchool of Chemistry, The University of Nottingham, University Park, Nottingham
NG7-2RD, UK
aurate(I) fragment and, consequently, different structural and
bDepartamento de Química, Universidad de La Rioja, Centro de Investigación en
photophysical properties can be obtained.7a,10 Similarly, the
Síntesis Química (CISQ), Complejo Científico-Tecnológico, 26004-Logroño, Spain.
nature and characteristics of the donor species that bind the
thallium atoms are also of crucial importance in the structural
disposition of the resulting materials. In fact, a “butterfly” dis-
position of metals has only been observed when ketones are
added as ligands.11 Many examples of Au/Tl compounds with
N, O or Se-donor ligands7a,11,12 or even metallic fragments10a
E-mail: josemaria.lopez@unirioja.es
cDipartimento di Scienze Chimiche e Geologiche, Università degli Studi di Cagliari,
S.S. 554 Bivio per Sestu, 09042 Monserrato (CA), Italy. E-mail: lippolis@unica.it
†Dedicated to Professor A. Laguna on the occasion on his 65th birthday.
‡Electronic supplementary information (ESI) available. CCDC 935229–935231.
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c3dt51334c
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11559–11570 | 11559

View Article Online
Paper
Dalton Transactions
bonded to thallium have been reported by our group over the
last few years. However, only a couple of Au/Tl derivatives with
Tl–S bonds are known (from the 1980s), both of them contain-
ing the asymmetric C,S-ligand [CH2P(S)Ph2]−.13
On the other hand, crown thioethers with different
numbers of S-donor atoms have been employed for the syn-
thesis of a large number of coordination compounds with a
great variety of metal centres.14 These macrocycles can form
stable, inert complexes with a wide range of transition metal
ions, sometimes forcing the metal centre to adopt unusual
coordination geometries and/or oxidation states.15 The
majority of these studies are focused on the chemistry of 1,4,7-
trithiacyclononane ([9]aneS3), 1,4,7,10-tetrathiacyclododecane
([12]aneS4), 1,5,9,13-tetrathiacyclohexadecane ([16]aneS4), and
1,4,7,10,13,16-hexathiacyclooctadecane ([18]aneS6), while the
coordination of larger rings, such as 1,4,7,10,13,16,19,22-
octathiacyclotetracosane ([24]aneS8) or 1,4,8,11,15,18,22,25-
octathiacyclooctacosane ([28]aneS8), is less studied. Further-
more, a systematic examination of the coordination chemistry
of p-block metal ions16 and d10 transition metal ions17 with
thioether crowns still remains an open area of investigation.
Specifically, there are only three examples of structurally
characterized thallium(I) complexes with crown thioethers,
namely, [Tl([9]aneS3)]PF6,16e [Tl([18]aneS6)]PF6,18 and [Tl([24]-
aneS8)]PF6.16d In [Tl([9]aneS3)]PF6 the trithia macrocycle is
bound facially to the metal ion and a thioether donor from an
adjacent complex cation makes a longer contact, thus linking
[Tl([9]aneS3)]+ cations into infinite helices. A similar situation
is observed in [Tl([18]aneS6)]PF6. In [Tl([24]aneS8)]PF6, each Tl+
ion is sandwiched between two symmetry-related halves of
two [24]aneS8 molecules within an infinite one-dimensional
sinusoidal polymer.
Taking the above into account, we decided to study the reac-
tivity of the crown thioethers [9]aneS3, [14]aneS4 and [28]aneS8
with the hetero-dimetallic compounds [{Au(C6X5)2}Tl]n
(X = Cl, F) and thereby determine whether the number of
sulphur atoms present in the S-donor ligands modifies the
structural disposition of the two metal centres in the starting
polymeric material, since a behaviour as a terminal or a brid-
ging ligand of the macrocyclic systems is expected to depend
on the number of S-donor centres and the size of the ring
cavity. Additionally, the modification of the halogen present in
the aryl groups bonded to gold may also influence the acidic
properties of the thallium centres. Both factors may therefore
result in different coordination environments at Tl+ and give
rise to different arrangements of metals, which may modify
the number and strength for the unsupported Au⋯Tl inter-
actions and, consequently, their optical properties.
Scheme 1
amounts of the corresponding thioether crown ligand in tetra-
hydrofuran. As expected, the higher number of sulphur
atoms in [24]aneS8 leads to a compound with a different
ratio of metals and ligands, [{Au(C6Cl5)2}2Tl2([24]aneS8)]n (3),
which was obtained by reaction of the same polymetallic chain
[{Au(C6Cl5)2}Tl]n with [24]aneS8 in a 2 : 1 molar ratio using
tetrahydrofuran as a solvent (Scheme 1). Although the syn-
thesis of complexes 1–3 can also be carried out in toluene,
the use of THF leads to higher yields. The substitution of the
chlorine atoms in the aryl groups bonded to gold by fluorine
does not seem to affect the stoichiometry of the resulting com-
pounds, and, thus, similarly, the pentafluorophenyl derivatives
[{Au(C6F5)2}Tl(L)]2 [L
= [9]aneS3 (4), [14]aneS4 (5)] and
[{Au(C6F5)2}2Tl2([24]aneS8)] (6) were obtained by following
the same procedure as for the synthesis of 1–3 using the
same reaction molar ratios of the starting products, but using
[{Au(C6F5)2}Tl]n as a starting material and toluene as a solvent
(see Scheme 1). In the case of complexes 4–6, their synthesis is
also possible in tetrahydrofuran, but unlike the results in the
preparation of 1–3, its employment as a solvent gave rise to the
complexes in lower yield.
All the complexes are stable to air and moisture for long
periods. They are insoluble in dichloromethane, acetonitrile
and diethyl ether, but soluble in O-donor solvents such as
tetrahydrofuran or acetone.
Their elemental analyses and spectroscopic data are in
accordance with the proposed stoichiometries (see the Experi-
mental section). Their IR spectra in nujol display, among
19
20
others, absorptions arising from the C6F5
and C6Cl5
groups bonded to gold(I) at approximately 1500, 950 and
780 cm−1 or about 834 and 614 cm−1, respectively. The pres-
ence of the [Au(C6F5)2]− fragment in 4–6 is evident in their
19F NMR spectra, which resemble that of the precursor
complex NBu4[Au(C6F5)2], and seem to indicate that a dissocia-
tive process giving rise to aurate(I) anions and thallium(I)
1
cations takes place in solution. Regarding the H NMR spectra
of complexes 1–6 in [D8]tetrahydrofuran, they all show the
resonances corresponding to the crown thioethers at similar
chemical shifts to those found for the free ligands. Therefore,
the coordination of the macrocyclic ligands to thallium does
not significantly affect the position of the resonances observed
in their 1H NMR spectra, nor does the dissociative process
Results and discussion
Synthesis and characterization
The heterometallic complexes [{Au(C6Cl5)2}Tl(L)]2 [L = [9]aneS3
(1), [14]aneS4 (2)], were obtained by reaction of the polymetal-
lic chain compound [{Au(C6Cl5)2}Tl]n with equimolecular
11560 | Dalton Trans., 2013, 42, 11559–11570
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
affect the S-donor molecules. Thus, the 1H NMR spectra of
2 and 5 display three signals at 2.01 (q, 4H, S–CH2–CH2–CH2–S),
2.73 (t, 8H, S–CH2–CH2–CH2–S) and 2.85 ppm (s, 8H, S–CH2–
CH2–S), while in the 1H NMR spectra of the rest of the pro-
ducts a unique singlet at 3.20 (1 and 4) or at 2.92 ppm (3 and
6) is observed. The molar conductivity measurements of solu-
tions of the complexes in acetone are of special relevance,
since they are consistent with an ionic formulation as 1 : 1 elec-
trolytes for complexes 1, 2, 4 and 5, and as 2 : 1 electrolytes for
complexes 3 and 6. This suggests that the units [Au(C6X5)2]−
and [Tl(L)]+ (1, 2, 4 and 5) or [Tl2(L)]2+ (3 and 6) are the species
present in solution. Finally, in their mass spectra (MALDI−) a
peak due to the unit [{Au(C6X5)2}2Tl]− at m/z = 1595 (1–3) or
1266 (4–6), as well as a signal corresponding to [Au(C6X5)2]− at
m/z = 695 (1–3) or 531 (4–6) are observed, the latter appearing
as a parent peak. In the MALDI+ mass spectra peaks due to
the fragments [Tl(L)]+ at m/z = 385 (1, 4), 473 (2, 5) or 685 (3, 6)
are observed with the experimental isotopic distributions in
agreement with the theoretical ones.
Fig. 2 Partial view of polymeric heterobimetallic chains in compound 3. Hydro-
gen atoms are omitted for clarity. Colour code: gold: yellow; thallium: blue;
sulphur: orange; carbon: grey; chlorine: green.
X-ray structural determination of derivatives 3, 4 and 6
Single crystals suitable for X-ray diffraction studies of com-
plexes 3, 4 and 6 were obtained by slow diffusion of n-hexane
into saturated solutions of the complexes in tetrahydrofuran.
Although they all contain similar structural motifs:
[Au(C6X5)2]− fragments linked to Tl(I) centres through Au⋯Tl
contacts, and each thallium connected to the macrocyclic
ligand through the sulphur atoms, the three crystal structures
are different. This is due to the behaviour of the crown
thioether as a terminal or as a bridging ligand as well as to the
presence or absence of aurophilic interactions. Therefore, in
the cases of complexes 3 and 6, where the ligand [24]aneS8
coordinates two Tl(I) centres, an Au–Tl–L–Tl–Au disposition
is observed (Fig. 1). However, while the crystals of 6 contain
discrete tetranuclear [{Au(C6F5)2}2Tl2([24]aneS8)] molecules,
the solid state structure of 3 displays polymeric chains as a
result of unsupported aurophilic contacts between the Au(I)
centres (see Fig. 2). In contrast, in 4 the crown thioether
[9]aneS3 acts as a terminal ligand and Au⋯Au interactions are
also observed, which results in tetranuclear discrete molecules
Fig. 3 Molecular structure of compound 4 with the labelling scheme for the
atom positions. Hydrogen atoms are omitted for clarity and ellipsoids are drawn
at the 30% level. #1 −x +1, −y + 1, −z.
with a L–Tl–Au–Au–Tl–L disposition, i.e., the crystal structure
and L/Tl ratio differ from those in 3 and 6 (Fig. 3).
In view of these crystal structures, we can confirm our
hypothesis about the influence of the nature of the ligands at
thallium on the metal ion arrangements in the final products,
since the vast majority of Au/Tl complexes with Au⋯Tl con-
tacts display polymeric networks with alternating gold and
thallium atoms,5b,c,6a,7,10,13,21 even with bulky N- or O-donor
ligands at thallium.6a A polymeric disposition similar to that
found in 3 has only been previously observed in the complex
[Tl(bipy)]2[Au(C6F5)2]2 which contains bridging 4,4′-bipyridine
ligands and Au⋯Au contacts, while the pentachlorophenyl
derivative displays the usual Au–Tl–Au–Tl disposition of
metals and the bipyridine ligands bridging adjacent poly-
metallic chains.7a The electronic effects associated with the
modification of the halogens in the aryl groups could be
responsible for the formation of a tetranuclear molecule in 6
instead of a polymeric chain through Au⋯Au contacts, as in
complex 3, since the smaller fluorine atoms in 6 should favour
the presence of aurophilic interactions in the latter if steric
effects were the dominant ones. Thus, the structure of 6
represents a very unusual arrangement only found in the
[{Au(C6Cl5)2Tl (toluene)}2(dioxane)] complex, where the dioxane
molecule bridges both thallium centres and no aurophilic con-
tacts are observed (the Au⋯Au separation is 3.876 Å).22
Fig. 1 Part of the crystal structure of compound 3 (left) and molecular struc-
ture of compound 6 (right) with the labelling scheme for the atom positions.
Hydrogen atoms are omitted for clarity and ellipsoids are drawn at the 30%
level. #1 −x + 2, −y + 2, −z + 1 #2 −x, −y + 1, −z.
On the other hand, complex 4 represents the first example
of a Au/Tl discrete molecule with a L–Tl–Au–Au–Tl–L sequence
of ligands and metals.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11559–11570 | 11561

View Article Online
Paper
Dalton Transactions
In all the structures the gold centres are linearly co- structure of 6 a higher number of interactions is observed:
ordinated to the ipso carbon atoms of the perhalophenyl thus, there are four weak Tl–S bonds [3.201(2)–3.418(3) Å]
ligands, with normal Au–C bond distances between 2.046(7) and two even weaker interactions [Tl–S(3)#2 = 3.500(3) and
and 2.056(7) Å, and with a maximum deviation from linearity Tl–S(4) = 3.744(2) Å]. In addition, the Tl⋯Au contact of
of 6.3° seen in 6. Additionally, the crystal structures of both 3 3.1483(4) Å and two Tl⋯F interactions of 3.110(4) and 2.968(6) Å
and 4, which contain different aryl groups and macrocycles, give rise to a more complex environment for thallium.
display aurophilic interactions of 3.3497(11) and 3.3294(6) Å,
respectively (Tables 2 and 3). These Au–Au distances are
Photophysical properties
shorter than those in [Tl(bipy)]2[Au(C6F5)2]2 (3.4092(3) Å),7a
or in other related polymeric Au/Tl complexes,5b,c,6a,7,10,13,21 all
The absorption spectra of complexes 1–3 show similar features
displaying two bands at 235 and 287 nm. The first one, which
is also present in the precursor complex NBu4[Au(C6Cl5)2], is
assigned to a π→π* transition in the pentachlorophenyl rings.
A similar assignment was previously made for related com-
plexes.5c The possibility of an n→σ* transition in the thioether
ligands cannot be ruled out since the energy of such tran-
sitions has been reported at similar energies24 and the ligands
show one absorption at 234 nm, although with less intensity.
Therefore, it can be masked by the more intense π→π* tran-
sition of the C6Cl5 ligands. In the case of the pentafluoro-
phenyl derivatives 4–6, the spectra are featureless, showing a
broad band between 225 and 275 nm that is likely to contain
the two peaks at 241 and 263 nm of the gold precursor
NBu4[Au(C6F5)2].
As we expected, all the complexes display rich optical be-
haviour in the solid state. Thus, they display emissions
between 445 nm and 526 nm at room temperature, and
between 441 and 599 nm when the measurements are carried
out at liquid nitrogen temperature (77 K) (see Table 5). In most
cases the spectra show a single emission, but in the case of
complexes 1 and 4 a low-intensity shoulder is evident at low
temperature (see Fig. 4). By contrast, all the complexes display
a single emission in butyronitrile glass at 77 K and none of
them show luminescence in solution at room temperature. All
these data seem to be related to the different structures of the
complexes in the solid state, which depend on the sulphur
donor ligand (see above), or on the different disposition that
they adopt in the different media. For example, the fact that
none of the complexes is luminescent in solution is likely to
be related to the rupture of the metal–metal interactions
of them showing Au–Au separations longer than 3.5 Å (3.825 Å
in 6), thereby excluding the possibility of significant inter-
action between gold atoms.
Although they are inequivalent, the Au–Tl distances are con-
sistent with the presence of metallophilic interactions in the
three structures. Therefore, the pentachlorophenyl complex 3,
which shows a weaker aurophilic contact than 4, displays
the strongest Au–Tl interaction, with a Au–Tl distance of
2.8881(8) Å. The strongest metallophilic contacts are found in
complex 4 [d(Au–Tl) = 3.0360(4), d(Au–Au) = 3.3294(6) Å], while
compound 6 displays the longest Au–Tl distance of the three
complexes [3.1483(4) Å]. In spite of these differences, all Au–Tl
distances lie within the range of Au–Tl distances found in rele-
vant structures: the distances vary from 2.804 Å in the metallo-
cryptand [AuPdTl(P2phen)3][BF4]2·2.5CH2Cl223 to 3.4899(6) Å in
the polymeric [Tl(1,10-phen)][Au(C6F5)2]7b (average Au–Tl dis-
tance 3.070 Å).
Again, the influence of the ligands bonded to thallium is
revealed by the Tl–S distances found in the three complexes,
since the structure of compound 4, which contains the
thioether [9]aneS3, displays shorter Tl–S bond distances than
3 or 6, in which the bulkier macrocycle [24]aneS8 is present.
Thus, the Tl–S bond lengths in 4 range from 3.0246(17) to
3.1154(19) Å, the longer one being of the same order as those
16e
described for [Tl([9]aneS3)]PF6
(from 3.092(3) to 3.114(3) Å
within each molecule), which contains the same ligand. In
contrast, complexes 3 and 6 show weaker interactions between
the thallium centres and the sulphur atoms of the thioether
[24]aneS8, displaying Tl–S distances between 3.256(7) and
3.587(7) Å in 3 and between 3.201(2) and 3.418(3) Å in 6,
values that compare well with those found in the related
complex [Tl([24]aneS8)]PF616d [3.2413(11)–3.4734(14) Å].
The stronger Tl–S interactions in 4 correspond well with the
lower coordination number that would be expected for Tl(I)
when [9]aneS3 is employed as a ligand and, accordingly, only
three Tl–S bonds are present in 4, but three additional Tl⋯F
contacts between 3.257(5) and 3.336(6) Å, as well as the metal-
lophilic Au⋯Tl interaction of 3.0360(4) Å increase the coordi-
nation number to 7. The environment of thallium could be
described as distorted pentagonal bipyramidal with Au and
S(1) occupying the apical positions. In the case of complex 3,
where the octathia macrocycle [24]aneS8 is employed, five Tl–S
interactions of different strengths appear (see Table 2), and no
significant Tl⋯F contact is observed: this results in a pyrami-
dal pentagonal environment for thallium with the gold centre
Fig. 4 Excitation and emission spectra for complex 4 in the solid state at RT
occupying the vertex of the pyramid. In contrast, in the crystal and 77 K.
11562 | Dalton Trans., 2013, 42, 11559–11570
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
promoted by the solvent, which is compatible with the conduc-
tivity values found in all cases (see the Experimental section).
Therefore, the presence of intermetallic interactions in the
solid state and their strength seem to be the key to explain the
luminescence of these complexes. In the case of the emissions
in butyronitrile glass at 77 K the values of energy are almost
identical to those of the [{Au(C6Cl5)2}Tl]n chains in the same
solvent. Therefore, the butyronitrile molecules probably dis-
place the sulphur donor ligands from the coordination sphere
of thallium giving rise to the same species.
Another factor that should be considered is the disposition
of the metal centres in the structures. In this sense, what is
known from the X-ray diffraction studies is that two different
structural dispositions are observed: that present in complex 4,
which shows a Tl–Au–Au–Tl unit, and those of complexes 3
and 6, in which two Au–Tl moieties are connected by a
[24]aneS8 thioether ligand (see X-ray structures discussion).
In the case of complex 4, its emission spectrum shows a
broad band at 501 nm at room temperature and two bands at
77 K, one of high intensity placed at 441 nm and another one
slightly shifted to the red (at 505 nm) of lower intensity (see
Fig. 4). These bands are likely to arise from two connected
excited states since both are obtained with the same excitation
energy (363 nm). In this sense, the band placed at higher
energy is assigned to arise from a metal–metal (gold) to ligand
(π* orbital from C6F5 rings) charge transfer (MMLCT) in the
interacting [Au(C6F5)2]− units, since the gold precursor
NBu4[Au(C6F5)2] displays an emission at 433 nm at low temp-
erature and it is not luminescent at room temperature.
The slight shift of the emission in complex 4 can be attributed
to the interaction between two of these units, which
destabilize the HOMO and stabilize the LUMO reducing the
HOMO–LUMO band gap.25 On the other hand, the band that
appears at lower energy is likely to arise from a ligand–metal
([Au(C6F5)2]−2) to metal–metal (gold–thallium) charge transfer
transition (LMMCT). In fact, a tetranuclear gold/thallium
complex which displays a similar distribution of metals
([Tl(bipy)]2[Au(C6F5)2]2) shows an emission at similar energy,
which was attributed to this transition based on previous
TD-DFT (Time Dependent-Density Functional Theory)
calculations.7a
In the case of the pentachlorophenyl complex 1 that con-
tains the same sulphur donor ligand, the emission spectra
also show a similar appearance, i.e., one emission band at
room temperature (525 nm) and an intense band at 471 nm
and a shoulder at 512 nm at 77 K. In this case, the pentachloro-
phenylgold(I) precursor displays an emission at 514 nm at
77 K; therefore we assign this emission to a transition that
arises from a metal–metal (gold) to ligand (π* orbital from the
C6Cl5 rings) charge transfer (MMLCT). By contrast, we assign
the emission that appears at higher energy as arising from a
ligand–metal ([Au(C6F5)2]−2) to metal (thallium or gold–thal-
lium) charge transfer transition (LMMCT). In this we can
anticipate that the structure of this complex will show a longer
Au–Tl distance, since the shorter the Au⋯Tl interaction is, the
smaller the HOMO–LUMO gap is.9b,26 Therefore, taking into
Fig. 5 Excitation and emission spectra for complex 3 in the solid state at RT
and 77 K.
account the previous comments we can propose for complex 1
a structure similar to that found for complex 4, but with a
longer gold–thallium interaction.
In the case of complexes 3 and 6, where both structures are
known, a direct comparison can be made. Thus, both com-
pounds are almost isostructural showing two [{Au(C6Cl5)2}Tl]
moieties with each thallium interacting with five sulphur
atoms of a common [24]aneS8 ligand. Both of them show a
single emission in their spectra at room temperature and at
low temperature, showing a shift of the emission to the red at
77 K (see Table 5 and Fig. 5). In this case we can propose that
the emission arises from a LMMCT (ligand (C6X5)–metal (gold)
to metal (thallium) charge transfer) in the same way as those
described above for complexes 1 and 4. In this case the impor-
tance of the strength of the Au⋯Tl interaction is highlighted;
since a shorter Au–Tl distance (2.89 Å) gives rise to a less
energetic emission (526 nm) in 3, while the more energetic
emission in complex 6 (471 nm) corresponds to a longer Au–Tl
distance (3.15 Å). Similarly, the shift to higher energy of the
emissions at room temperature when compared to the corres-
ponding low temperature measurements in both spectra is
a consequence of the thermal expansion that larger gold–
thallium distances and larger HOMO–LUMO gaps provoke.
For complexes 2 and 5 bearing the macrocycle [14]aneS4,
both spectra present a single asymmetric emission, whose
shape could mask a second emission, as observed in the
spectra of 1 and 4. Nevertheless, in the case of complex 2, this
asymmetry disappears at low temperature, shifting the emis-
sion to the red (from 507 to 599 nm), while for complex 5 the
asymmetry remains at low temperature and, as well as the
main peak, is shifted to the red (see ESI‡). Both observations
clearly differ from what is seen for complexes 1 and 4, so their
solid state structures should be somehow different. Accord-
ingly, we tentatively propose a similar Tl–Au–Au–Tl disposition
of metals, but with a different environment for the thallium
centres, since the ligands in 2 and 5 contain a fourth sulphur
atom that can also bind this metal centres. In addition, the
molar conductivity measurements for complexes 2 and 5 show
values typical of 1 : 1 electrolytes (see the Experimental
section); this is in accordance with the proposed structure.
Note that such a slight difference could be significant enough
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11559–11570 | 11563

View Article Online
Paper
Dalton Transactions
to modify the optical properties of these compounds. In fact,
as mentioned in the introduction, in previous papers we have
described some examples of Au/Tl complexes in which the
luminescent properties are determined by the structural dis-
position and the Tl(I) environment.7
Finally, if we observe the data of all the complexes we can
establish some interesting trends when we compare similar
structures or S-donor ligands: (1) Complexes with C6F5 groups
give shorter lifetime values than C6Cl5 complexes. (2) Com-
plexes with C6F5 groups emit at shorter wavelengths than the
corresponding C6Cl5 ones. (3) The quantum yields in the solid
state are higher in the case of the C6Cl5 complexes. All of these
characteristics can be incorporated into design strategies that
permit the rational synthesis of complexes for practical
applications.
DFT calculations
The rich structure-dependent emissive properties of complexes
1–6 in the solid state prompted us to study from a theoretical
point of view the origin of the electronic transitions respon-
sible for this behaviour. We have therefore carried out DFT
and TD-DFT calculations on representative model systems of
complexes 3, 4 and 6 (models 3a, 4a and 6a, respectively) that
display the two types of structural disposition determined for
these Au(I)–Tl(I) systems with different ring size crown
thioethers. Since theoretical models incorporating all the
metallophilic interactions are too big for optimization runs,
we have built up the model systems directly from the X-ray
diffraction data. Model 4a represents the uncommon Tl–Au–
Au–Tl discrete disposition found in the corresponding struc-
ture of complex 4; meanwhile, models 3a and 6a represent
the metallophilic Au⋯Au and Au⋯Tl interactions in a model
system that leaves the [24]aneS8 thioether ligand only co-
ordinated to one Tl(I) center (see Fig. 6). These tetranuclear
models 3a and 6a consist of the molecular fragments that
contain the repeat unit of the polymeric complex 3 displaying
all the interactions present in the solid state.
Fig. 7 Frontier molecular orbitals HOMO and LUMO for model system 4a.
from the C6F5 ligands is found. On the other hand, the popu-
lation analysis of the lowest empty orbitals shows a main con-
tribution from the Tl(I) centres to the LUMO orbital (43%), but
also from the [Au(C6F5)2]− units (50%) and a main contri-
bution from the Tl(I) centres (>80%) to the LUMO + 1 and
LUMO + 2 orbitals. In view of these results the lowest elec-
tronic transitions should be of the charge transfer type since a
clear difference in the composition of the highest occupied
MOs and the lowest empty ones is observed. The shape of the
orbitals is shown in Fig. 7.
The first ten excitation energies of models 4a have been cal-
culated at the TD-DFT level of theory as described in the com-
putational methods section. We have carried out an analysis of
the energy, strength, and orbitals involved in the first singlet–
singlet excitations, which reproduce the vertical electronic exci-
tations that could be related with the luminescent behaviour
observed for complex 4. The results are depicted in Table 7.
Among the first 10 singlet–singlet excitations the ones giving
the highest oscillator strengths, i.e., excitations 1, 3 and 7, are
studied in detail. Thus, the lowest theoretical singlet–singlet
excitation appears at 323 nm (experimental excitation at
363 nm at 77 K) and displays a large oscillator strength value.
This excitation consists of an electronic transition between
HOMO and LUMO MOs and, as we have commented above, it
can be considered as a charge transfer from the interacting [Au
(C6F5)2]−⋯[Au(C6F5)2]− units that mainly contribute to the
HOMO (79%), to the Tl⋯Au interacting metals that mainly
contribute to the LUMO orbital (71%), with some participation
from the C6F5 ligands (22%). This transition can be assigned
to the one observed experimentally that is responsible for the
low energy emission of complex 4 at 77 K (experimental emis-
sion at 505 nm). If we analyze the 3rd and 7th theoretical
singlet–singlet excitations that appear at 302 and 296 nm, we
observe a slight difference with the first one that is a different
origin of electronic excitation. The starting orbitals in both
A study of the molecular orbitals (MOs) along with a popu-
lation analysis was used to check the contribution of each
atom to each occupied orbital for model 4a (see Table 6). The
population analysis of the highest occupied MOs shows that
while the HOMO orbital is mainly located at the [Au(C6F5)2]−
units (79%) with a small contribution from the Tl(I) centres
(14), from HOMO − 1 to HOMO − 4 a clear main contribution
Fig. 6 Theoretical model systems [Au2(C6F5)4Tl2([9]ane-S3)2] (4a) (left),
[Au2(C6Cl5)4Tl2([24]ane-S8)2] (3a) (center) and [Au2(C6F5)4Tl2([24]ane-S8)2]
(6a) (right).
11564 | Dalton Trans., 2013, 42, 11559–11570
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
cases (HOMO − 2 and HOMO − 4) are mainly located at the
Thus, in the case of model 3a the frontier orbitals HOMO
anionic C6F5 ligands of the interacting [Au(C6F5)2]−⋯ and LUMO are mostly located at the [Au(C6F5)2]− units and the
[Au(C6F5)2]− units, while the arriving orbital is the same as in Tl(I) centres, the next occupied MOs (H − 1, H − 8 to H − 11),
the first singlet–singlet excitation, i.e., the LUMO orbital. In are mostly ligand-based. This trend is in agreement with short
view of these results we can state that the vertical excitations of intermetallic distances that are directly related to a reduction
singlet parentage mainly arrive into the lowest empty orbital of the metal-based HOMO–LUMO gap. In the case of model
LUMO. However, the experimental observation of a high- 6a, the presence of larger intermetallic distances changes the
energy emission at 441 nm assigned to a metal–metal (gold) to HOMO, which is now placed at the thioether ligand, with
ligand (π* orbital from C6F5 rings) charge transfer (MMLCT) is HOMO − 2 and LUMO now associated with the aurate-
not easy to assign from the calculation of the vertical excitation thallium based MOs (see Table 6). Nevertheless, although the
from the ground state (TD-DFT approach). The fact that two population analysis of the frontier orbitals shows some differ-
emissions appear by excitation at the same energy and change ences between these complexes, the lowest singlet–singlet
their relative intensities with temperature seems to indicate theoretical excitations calculated at the TD-DFT level display a
the presence of two interconnected singlet excited states similar behaviour (Table 7). In the case of model 3a the most
(experimental lifetimes for the emissions at 77 K are in the intense theoretical excitation (among the first 10 excitations) is
nanosecond range). It seems likely that upon vertical excitation the first one (theoretical excitation at 387 nm, experimental
the molecular excited state structure could be relaxed towards continuum profile in the 300–450 nm) and consists of a verti-
two different local minima from which the emissions would cal excitation between HOMO and LUMO orbitals, that is, a
occur, depending on the temperature. The different structural charge transfer from the interacting [Au(C6F5)2]−⋯[Au(C6F5)2]−
distortions would be explained by the full optimization of both units that mainly contribute to the HOMO (67%), to the
excited states but, unfortunately, the large size of the theore- Tl⋯Au interacting metals that mainly contribute to the LUMO
tical models precludes these calculations. Therefore, although orbital (54%), with an important participation from the C6Cl5
the assignment of the high energy emission is reasonable ligands (42%) (Table 7). This result is similar to that obtained
from an experimental viewpoint it cannot be evaluated at this for model 4a and explains the origin of the emission proposed
stage from a theoretical one, due to a very large computational experimentally for complex 3. If we analyze the singlet–singlet
−
cost.
electronic excitations calculated for model 6a we observe now
The second part of the theoretical study has been focused that the main excitation is the first one (theoretical excitation
on the analysis of the electronic excitations observed in the at 298 nm, experimental maximum at 358 nm) and it consists
Au–Tl complexes bearing the [24]aneS8 thioether ligand. of a mixed electronic transition mainly involving aurate-
As has been shown in the structural studies section, both thallium-based HOMO − 2 and LUMO MOs with the contri-
complexes [{Au(C6Cl5)2}2Tl2([24]aneS8)]n (3) and [{Au(C6F5)2}2- bution of a HOMO–LUMO transition (Table 7). In view of the
Tl2([24]aneS8)] (6) display analogous structures with the excep- character of the involved orbitals we propose that when the
tion of different perhalophenyl ligands, C6Cl5 (3) and C6F5 (6) intermetallic distances are longer the main electronic singlet–
and the presence of aurophilic interactions in 3 that are singlet excitation is still a charge transfer between the aurate
absent in 6 and result in a polymer chain or a discrete tetra- units that mainly contribute to the HOMO − 2 (61%), and the
nuclear molecule, respectively. The study of the electronic Tl⋯Au interacting metals (60%), although a ligand ([24]aneS8)
structure and population analysis of model systems 3a and 6a to metal–metal (gold–thallium) charge transfer transition con-
shows significant differences that could be attributed to the tribution cannot be excluded. The relationship between the
main difference between both complexes (Table 6 and Fig. 8), larger intermetallic distances and the different contributions
that is, the presence of much shorter intermetallic distances to the origin of the emissive behaviour for models 3a and 6a
in the case of complex 3 than the ones observed in complex 6.
would explain the different emission energies found for com-
plexes 3 and 6.
Conclusions
The use of crown thioethers (L) against [{Au(C6X5)2}Tl]n
affords compounds of stoichiometries [{Au(C6X5)2}Tl(L)]2 or
[{Au(C6F5)2}2Tl2(L)] depending on the size of the macrocycle,
which can act as a terminal or as a bridging S-donor ligand.
The formation of Tl–S bonds and Au⋯Tl or Au⋯Tl and Au⋯Au
contacts leads to unusual structures for Au/Tl complexes with
polymeric [–L–Tl–Au–Au–Tl–]n (3) or tetranuclear [L–Tl–Au–Au–
Tl–L] (4) or Au–Tl–L–Tl–Au (6) dispositions, clearly influenced
by the nature of the ligands at the metal centres.
The metal–metal interactions are the origin of the lumine-
scence observed for all the complexes in the solid state, a
Fig. 8 Frontier molecular orbitals and HOMO and LUMO for model systems 3a
and 6a.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11559–11570 | 11565

View Article Online
Paper
Dalton Transactions
property that strongly depends on the halogens present in the solution was concentrated under vacuum. Finally, the addition
aryl groups: fluorinated compounds display shorter lifetimes of n-hexane led to the precipitation of product 2 as a yellow
and emission wavelengths, as well as lower quantum yields solid that was filtered and washed with n-hexane (98.2 mg,
than the chlorinated ones.
DFT calculations on representative model systems of some (1168.53): C 22.61, H 1.73, S 10.98. Found: C 22.34, H 1.73,
of the complexes also agree with the assignation of the origin S 9.70. ΛM: 117 Ω−1 cm2 mol−1 1H NMR (400 MHz, [D8]-
of their optical properties and show a good correlation tetrahydrofuran, ppm): δ 2.01 (q, 4H , S–CH2–CH2–CH2–S,
76% yield). Elemental analysis (%) calcd for C22H20AuCl10S4Tl
.
3
between the intermetallic distances and emission energies.
3J(H–H) = 2 Hz), 2.73 (t, 8H, S–CH2–CH2–CH2–S, J(H–H) = 2 Hz),
2.85 (s, 8H, S–CH2–CH2–S). MALDI-TOF(−) m/z (%): 695
[Au(C6Cl5)2]− (100), 1595 [{Au(C6Cl5)2}Tl]− (16). MALDI-TOF(+)
m/z (%): 473 [Tl([14]aneS4)]+ (100). FTIR (Nujol):
Experimental
General
ν([Au(C6Cl5)2]−) at 837 and 614 cm−1
.
[{Au(C6Cl5)2}2Tl2([24]aneS8)]n (3). To a well stirred solution
of [{Au(C6Cl5)2}Tl]n (85 mg, 0.095 mmol) in tetrahydrofuran,
[24]aneS8 (23 mg, 0.047 mmol) was added. The mixture was
stirred at room temperature for 3 h, and then the solvent was
partially removed under reduced pressure and, finally, the
addition of n-hexane led to the precipitation of product 3 as a
green solid that was filtered and washed with n-hexane
(93.4 mg, 87% yield). Elemental analysis (%) calcd for
C40H32Au2Cl20S8Tl2 (2279.27): C 21.13, H 1.42, S 11.26. Found
Thioether crown ligands,27 [{Au(C6Cl5)2}Tl]n5b and [{Au(C6F5)2}Tl]n
were prepared according to the literature.
10a
Instrumentation
Infrared spectra were recorded in the 4000–200 cm−1 range on
a Nicolet Nexus FT-IR using Nujol mulls between polyethylene
sheets. C,
H and S analyses were carried out with a
Perkin-Elmer 240C microanalyzer. Molar conductivities were
measured in ca. 5 × 10−4 M acetone solutions with a Jenway
4510 conductivity meter. Mass spectra were recorded with
a Bruker Microflex MALDI-TOF using dithranol (DIT) or
11-dicyano-4-tert-butylphenyl-3-methylbutadiene (DCTB) as the
C 21.10, H 1.52, S 11.96. ΛM: 220 Ω−1 cm2 mol−1 1H NMR
.
(400 MHz, [D8]tetrahydrofuran, ppm): δ 2.92 (s, 32H, CH2).
MALDI-TOF(−) m/z (%): 695 [Au(C6Cl5)2]− (100), 1595
[{Au(C6Cl5)2}Tl]− (14). MALDI-TOF(+) m/z (%): 685 [Tl([24]-
aneS8)]+ (66). FTIR (Nujol): ν([Au(C6Cl5)2]−) at 834 and
1
matrix. H and 19F NMR spectra were recorded with a Bruker
614 cm−1
.
Avance 400 in [D8]tetrahydrofurane. Chemical shifts are
quoted relative to SiMe4 (1H, external) and CFCl3 (19F, exter-
nal). Excitation and emission spectra in the solid state were
recorded with a Jobin-Yvon Horiba Fluorolog 3–22 Tau-3
spectrofluorimeter. Lifetime measurements were recorded
with a Datastation HUB-B with a nanoLED controller and
DAS6 software. The nanoLED employed for lifetime measure-
ments was one of 370 nm with pulse lengths of 0.8–1.4 ns. The
lifetime data were fitted with the Jobin-Yvon software package.
Measurements at 77 K were done with an Oxford Cryostat
Optistat DN with an accessory for solid samples.
[{Au(C6F5)2}Tl([9]aneS3)]2 (4). [9]aneS3 (24.5 mg, 0.136 mmol)
was added to a suspension of [{Au(C6F5)2}Tl]n (100 mg,
0.136 mmol) in toluene. After 2 h of stirring, the solution was
concentrated under vacuum. Finally, the addition of n-hexane
led to the precipitation of product 4 as a white solid that was
filtered and washed with n-hexane (74.5 mg, 60% yield).
Elemental analysis (%) calcd for C18H12AuF10S3Tl (915.82):
C 23.61, H 1.32, S 10.50. Found C 24.73, H 1.88, S 10.44. ΛM:
133 Ω−1 cm2 mol−1 1H NMR (400 MHz, [D8]tetrahydrofuran,
.
ppm): δ 3.21 (s, 12 H, CH2). 19F NMR (400 MHz, [D8]tetrahydro-
3
furan, ppm): δ −112.0 (m, 2F, Fo), −159.8 (t, 1F, Fp, J(Fp–Fm)
=
Synthesis
19.3 Hz), −161.3 (m, 2F, Fm). (MALDI-TOF(−) m/z (%): 531
[{Au(C6Cl5)2}Tl([9]aneS3)]2 (1). To a well stirred solution [Au(C6F5)2]− (100), 1267 [{Au(C6F5)2}Tl]− (20). MALDI-TOF(+)
of [{Au(C6Cl5)2}Tl]n (100 mg, 0.111 mmol) in tetrahydrofuran, m/z (%): 385 [Tl([9]aneS3)]+ (15). FTIR (Nujol): ν([Au(C6F5)2]−)
[9]aneS3 (20 mg, 0.111 mmol) was added. The mixture was at 1502, 952 and 785 cm−1
.
stirred at room temperature for 3 h, and then the solvent was
[{Au(C6F5)2}Tl([14]aneS4)]2
(5). [14]aneS4
(36.5
mg,
partially removed under reduced pressure and, finally, the 0.136 mmol) was added to a suspension of [{Au(C6F5)2}Tl]n
addition of n-hexane led to the precipitation of product 1 as a (100 mg, 0.136 mmol) in toluene. After 2 h of stirring, the solu-
yellow solid that was filtered and washed with n-hexane tion was concentrated under vacuum. Finally, the addition of
(86.4 mg, 72% yield). Elemental analysis (%) calcd for n-hexane led to the precipitation of product 5 as a white solid
C18H12AuCl10S3Tl (1080.36): C 22.01, H 1.12, S 8.90. Found: that was filtered and washed with n-hexane (120 mg, 88%
C 22.71, H 1.20, S 8.90. ΛM: 117 Ω−1 cm2 mol−1 1H NMR yield). Elemental analysis (%) calcd for C22H20AuF10S4Tl
.
(400 MHz, [D8]tetrahydrofuran, ppm): δ 3.20 ppm (s, 12 H, (1003.99): C 26.32, H 2.01, S 12.78. Found C 27.62, H 2.70,
1
CH2). MALDI-TOF(−) m/z (%): 695 [Au(C6Cl5)2]− (100), 1595 S 11.52. ΛM: 108 Ω−1 cm2 mol−1. H NMR (400 MHz, [D8]tetra-
3
[{Au(C6Cl5)2}Tl]− (9). MALDI-TOF(+) m/z (%): 385 [Tl([9]aneS3)]+ hydrofuran, ppm): δ 2.01 (q, 4H , S–CH2–CH2–CH2–S, J(H–H)
=
3
(100). FTIR (Nujol): ν([Au(C6Cl5)2]−) at 834 and 617 cm−1
.
2 Hz), 2.73 (t, 8H, S–CH2–CH2–CH2–S, J(H–H) = 2 Hz), 2.85
[{Au(C6Cl5)2}Tl([14]aneS4)]2 (2). [14]aneS4 (30 mg, 0.111 mmol) (s, 8H, S–CH2–CH2–S). 19F NMR (400 MHz, [D8]tetrahydrofuran,
was added to
solution of [{Au(C6Cl5)2}Tl]n (100 mg, ppm): δ –112.0 (m, 2F, Fo), −159.9 (t, 1F, Fp, 3J(Fp–Fm) = 19.8 Hz),
0.111 mmol) in tetrahydrofuran. After 3 h of stirring, the −161.4 (m, 2F, Fm). MALDI-TOF(−) m/z (%): 531 [Au(C6F5)2]−
a
11566 | Dalton Trans., 2013, 42, 11559–11570
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 1 Data collection and structure refinement details for 1–3
Compound
3
4
6
Chemical formula
Crystal habit
Crystal size/mm
Crystal system
C
40H32Au2Cl20S8Tl2
C36H24Au2F20S6Tl2
Colourless block
0.45 × 0.3 × 0.2
Monoclinic
P21/c
C40H32Au2F20S8Tl2
Pale yellow plate
0.3 × 0.2 × 0.1
Triclinic
Green plate
0.14 × 0.09 × 0.03
Triclinic
ˉ
P1
ˉ
Space group
P1
a/Å
b/Å
c/Å
α/°
10.6829(11)
11.9086(10)
14.4564(12)
66.786(8)
78.615(8)
66.326(9)
1546.3(2)
1
10.9933(3)
11.7372(4)
17.4726(6)
90
92.688(2)
90
3058.51(19)
2
2.701
9.0826(5)
12.5273(7)
12.9677(5)
115.204(3)
98.272(3)
103.410(3)
1248.28(11)
1
β/°
γ/°
V/Å3
Z
Dc/g cm−3
2.449
2.596
M
F(000)
2280.80
1060
1831.58
1672
1951.81
900
T/°C
2θmax/°
22
50
11.091
8439
5438
0.031
0.0516
0.136
−100
−80
55
55
μ(Mo-Kα)/mm−1
No. of refl. measured
No. of unique refl.
Rint
14.019
29 515
5174
0.0803
0.0402
0.1090
12.735
19 740
5679
0.0601
0.0389
0.1068
R [F > 2σ(F)]a
Wr [F2, all refl.]b
No. of refl. used
[F > 2σ(F)]
4340
322
309
1.02
1.98
4437
299
97
1.097
1.562
4564
352
125
1.033
1.506
No. of parameters
No. of restraints
Sc
Max. residual electron density/e Å−3
a R(F) = Σ||Fo| − |Fc||/Σ|Fo|. b wR(F2) = [Σ{w(Fo2 − Fc2)2}/Σ{w(Fo2)2}]0.5; w−1 = σ2(Fo2) + (aP)2 + bP, where P = [Fo2 + 2Fc2]/3 and a and b are constants
adjusted by the program. c S = [Σ{w(Fo2 − Fc2)2}/(n − p)]0.5, where n is the number of data and p the number of parameters.
Table 4 Selected bond lengths [Å] and angles [°] for complex 6
Table 2 Selected bond lengths [Å] and angles [°] for complex 3
Tl–S(1)
Tl–S(2)
Tl–S(3)
Tl–S(4)#1
Tl–F(1)
3.201(2)
3.258(2)
3.418(3)
3.277(2)
3.110(4)
Tl–F(10)
Au–Tl
Au–C(1)
Au–C(11)
2.968(6)
3.1483(4)
2.047(7)
2.046(7)
Tl(1)–S(1)
Tl(1)–S(4)
Tl(1)–S(7)
Tl(1)–S(10)
Tl(1)–S(1)#2
3.327(7)
3.319(6)
3.256(7)
3.366(6)
3.587(7)
Au(1)–Tl(1)
Au(1)–Au(1)#1
Au(1)–C(20)
Au(2)–C(30)
2.8881(8)
3.3497(11)
2.056(11)
2.046(12)
C(11)–Au–C(1)
172.7(3)
C(30)–Au(1)–C(20)
175.6(5)
Tl(1)–Au(1)–Au(1)#1
147.89(3)
Symmetry transformations used to generate equivalent atoms:
#1 −x + 2, −y + 2, −z + 1.
Symmetry transformations used to generate equivalent atoms:
#1 −x + 1, −y, −z #2 −x, −y + 1, −z.
Table 3 Selected bond lengths [Å] and angles [°] for complex 4
[{Au(C6F5)2}2Tl2([24]aneS8)]
(6). [24]aneS8
(32.7
mg,
Tl–S(1)
Tl–S(2)
Tl–S(3)
Tl–F(2)#2
Tl–F(8)#3
3.1154(19)
3.0584(20)
3.0246(17)
3.257(5)
Tl–F(9)#3
Au–Tl
Au–Au#1
Au–C(1)
Au–C(11)
3.285(5)
3.0360(4)
3.3294(6)
2.056(7)
2.052(7)
0.068 mmol) was added to a suspension of [{Au(C6F5)2}Tl]
(100 mg, 0.136 mmol) in toluene. After 2 h of stirring, the
solution was concentrated under vacuum. Finally, the addition
of n-hexane led to the precipitation of product 6 as a pale
yellow solid that was filtered and washed with n-hexane
(121 mg, 91% yield). Elemental analysis (%) calcd for
C40H32Au2F20S8Tl2 (1951.87): C 24.60, H 1.65, S 13.14. Found
3.336(6)
C(11)–Au–C(1)
176.9(3)
Tl–Au–Au#1
158.410(14)
Symmetry transformations used to generate equivalent atoms:
#1 −x + 1, −y + 1, −z #2 −x, −y + 1, −z #3 −x + 1, −y, −z.
C 23.52, H 1.64, S 13.83. ΛM: 280 Ω−1 cm2 mol−1 1H NMR
.
(400 MHz, [D8]tetrahydrofuran, ppm): δ 2.92 (s, 32 H, CH2).
(100), 1267 [{Au(C6F5)2}Tl]− (24). MALDI-TOF(+) m/z (%): 473
[Tl([14]aneS4)]+ (30). FTIR (Nujol): ν([Au(C6F5)2]−) at 1508, 955
19F NMR (400 MHz, [D8]tetrahydrofuran, ppm): δ −112.0 (m,
3
2F, Fo), −160.0 (t, 1F, Fp, J(Fp–Fm) = 19.8 Hz), −161.5 (m, 2F,
and 785 cm−1
.
Fm). MALDI-TOF(−) m/z (%): 531 [Au(C6F5)2]− (100), 1267
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11559–11570 | 11567

View Article Online
Paper
Dalton Transactions
Table 5 Photophysical properties of complexes 1–6
Solid (RT) em (exc)
525 (405)
507 (369)
526 (433)
Solid (77 K) em (exc)
Glassa (77 K) em (exc)
τb (ns)
Φb
[{Au(C6Cl5)2}Tl([9]aneS3)]2 (1)
[{Au(C6Cl5)2}Tl([14]aneS4)]2 (2)
[{Au(C6Cl5)2}Tl([24]aneS8)]n (3)
NBu4[Au(C6Cl5)2]
471, 512 (sh) (360)
599 (369)
534 (374)
529 (362)
533 (370)
517 (363)
629.6
139.9
3069.2
26
37
89
514 (369)
[{Au(C6Cl5)2}Tl]n
534 (370)
422 (296)
433 (292)
437 (271)
[{Au(C6F5)2}Tl([9]aneS3)]2 (4)
[{Au(C6F5)2}Tl([14]aneS4)]2 (5)
[{Au(C6F5)2}2Tl2([24]aneS8)] (6)
NBu4[Au(C6F5)2]
501 (371)
445, 507 (sh) (394)
471 (372)
441, 505 (sh) (363)
460, 541 (sh) (379)
484 (358)
339.1
78.2
484.8
22
25
25
433 (285)
[{Au(C6F5)2}Tl]n
431 (274)
a In butyronitrile (2 × 10−4 M). b Solid state at room temperature.
[{Au(C6F5)2}Tl]− (18). MALDI-TOF(+) m/z (%): 685 [Tl([24]- alternative orientations refined to 0.53 and 0.47. In 6, S3, C23
aneS8)]+ (100). FTIR (Nujol): ν([Au(C6F5)2]−) at 1502, 953 and and C24 are also disordered over two different positions
779 cm−1
.
with occupancies refined to 0.8 and 0.2. Further details of the
data collection and refinement are given in Table 1. Selected
bond lengths and angles are presented in Tables 2–4 and
crystal structures of complexes 3, 4 and 6 are shown in
Fig. 1–3.
Crystallography
Crystals were mounted in inert oil on glass fibres and trans-
ferred into the cold gas stream of an Oxford Cryosystems open-
flow cryostat mounted on a Nonius Kappa CCD diffractometer.
Data were collected using graphite-monochromated Mo Kα
radiation (λ = 0.71073 Å). Scan type ω and ϕ. Absorption correc-
tions: semiempirical (based on multiple scans). The structures
were solved by direct methods and refined on F2 using the
program SHELXL97.28 All non-hydrogen atoms were refined
anisotropically with the exception of C8, C9, C11 and C12 in 3,
and all hydrogen atoms were included using a riding model.
Disorder of the C8, C9, C11 and C12 methylene groups of the
macrocycle in 3 was modelled with occupancies of the two
Computational details
All calculations were carried out using the Gaussian09
program package.29 DFT and time-dependent DFT calculations
were carried out using the PBE functional.30 The following
basis set combinations were employed for the metals Au and
Table 7 TD-DFT first singlet excitation calculation for model systems 3a, 4a
and 6a. Only excitations with larger oscillator strengths are included among the
first 10 excitation calculations
Model Exc.
λcalc (nm) f (s)a
Contributionsb
Table 6 Population analysis for model systems 3a, 4a and 6a. Contribution
from each part of the molecule to the occupied orbitals (%)
3a
1
9
386.7
303.7
0.5874
0.0236
HOMO → LUMO (100)
HOMO(−11) → LUMO (44.2)
HOMO(−9) → LUMO (55.8)
HOMO(−11) → LUMO (13.0)
HOMO(−10) → LUMO (61.5)
HOMO(−9) → LUMO (11.8)
HOMO(−8) → LUMO (11.3)
HOMO → LUMO (100)
HOMO(−2) → LUMO (100)
HOMO(−4) → LUMO (93.6)
HOMO → LUMO(+2) (6.4)
HOMO(−1) → LUMO(+1) (81)
HOMO(−4) → LUMO(+2) (12)
HOMO(−2) → LUMO (65.4)
HOMO → LUMO (31.9)
HOMO(−5) → LUMO (27.4)
HOMO(−2) → LUMO(+2) (16.2)
HOMO → LUMO (14.0)
HOMO(−3) → LUMO(+1) (11.9)
HOMO → LUMO(+1) (11.2)
HOMO(−11) → LUMO (13.0)
HOMO(−10) → LUMO (61.5)
HOMO(−9) → LUMO (11.8)
HOMO(−8) → LUMO (11.3)
Model
Orbital
Tl
Au
S-ligand
C6X5
10
303.2
0.1358
3a
LUMO
HOMO
28
12
1
1
1
2
2
83
81
43
14
0
0
0
0
74
70
34
2
7
6
26
54
0
4
6
0
2
0
0
28
54
11
2
2
15
0
2
26
6
36
13
4
4
21
94
35
37
41
64
16
17
7
7
0
1
1
42
13
5
61
55
57
32
1
HOMO(−1)
HOMO(−8)
HOMO(−9)
HOMO(−10)
HOMO(−11)
LUMO(+2)
LUMO(+1)
LUMO
4a
6a
1
3
7
323.5
302.5
295.8
0.8299
0.0414
0.2048
4a
6a
8
1
6
285.5
298.2
278.2
0.0019
0.4651
0.1903
2
22
25
88
98
98
84
9
4
26
4
25
10
20
48
HOMO
HOMO(−1)
HOMO(−2)
HOMO(−3)
HOMO(−4)
LUMO(+2)
LUMO(+1)
LUMO
1
17
24
14
88
32
72
76
28
7
277.5
0.0712
HOMO
HOMO(−2)
HOMO(−3)
HOMO(−4)
HOMO(−5)
0
4
a Oscillator strength (f) shows the mixed representation of both
velocity and length representations. b Value is |coeff|2 × 100.
20
11568 | Dalton Trans., 2013, 42, 11559–11570
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Tl: the 19-VE and 21-VE pseudo-potentials from Stuttgart and
the corresponding basis sets augmented with two f polariz-
ation functions were used,31 respectively. The rest of the atoms
were treated with SVP basis sets.32,33
6 See for example: (a) E. J. Fernández, A. Garau, A. Laguna,
T. Lasanta, V. Lippolis, J. M. López-de-Luzuriaga,
M. Montiel and M. E. Olmos, Organometallics, 2010, 29,
2951; (b) E. J. Fernández, A. Laguna, T. Lasanta,
J. M. López-de-Luzuriaga, M. Montiel and M. E. Olmos,
Inorg. Chim. Acta, 2010, 363, 1965.
7 (a) E. J. Fernández, P. G. Jones, A. Laguna, J. M. López-de-
Luzuriaga, M. Monge, J. Pérez and M. E. Olmos, Inorg.
Chem., 2002, 41, 1056; (b) E. J. Fernández, P. G. Jones,
A. Laguna, J. M. López-de-Luzuriaga, M. Monge,
M. Montiel, M. E. Olmos and J. Pérez, Z. Naturforsch., B:
Chem. Sci., 2004, 59, 1379; (c) E. J. Fernández, A. Laguna,
J. M. López-de-Luzuriaga, M. Montiel, M. E. Olmos and
J. Pérez, Organometallics, 2005, 24, 1631.
Acknowledgements
D.G.I.(MEC)/FEDER (project number CTQ2010-20500-C02-02)
and MIUR (project PRIN 2009–2009Z9ASCA) are acknowledged
for financial support.
8 (a) M. F. Hawthorne and Z. Zheng, Acc. Chem. Res., 1997,
30, 267; (b) M. F. Hawthorne, Pure Appl. Chem., 1994, 66,
245; (c) J. D. Wuest, Acc. Chem. Res., 1999, 32, 81;
(d) J. Vaugeois, M. Simard and J. D. Wuest, Coord. Chem.
Rev., 1995, 95, 55.
9 (a) T. Lasanta, M. E. Olmos, A. Laguna, J. M. López-de-
Luzuriaga and P. Naumov, J. Am. Chem. Soc., 2011, 133,
16358; (b) A. Laguna, T. Lasanta, J. M. López-de-Luzuriaga,
M. Monge, P. Naumov and M. E. Olmos, J. Am. Chem. Soc.,
2010, 132, 456; (c) E. J. Fernández, A. Laguna, J. M. López-
de-Luzuriaga, M. E. Olmos and R. C. Puelles, Z. Natur-
forsch., B: Chem. Sci., 2009, 64, 1500; (d) E. J. Fernández,
J. M. López-de-Luzuriaga, M. Monge, M. E. Olmos,
R. C. Puelles, A. Laguna, A. A. Mohamed and J. P. Fackler
Jr., Inorg. Chem., 2008, 47, 8069.
Notes and references
1 (a) R. Hoffmann, Angew. Chem., Int. Ed. Engl., 1987, 26, 846;
(b) J. M. Williams, A. J. Schultz, A. E. Underhill and
K. Carneiro, in Extended Linear Chain Compounds, Plenum
Press, New York, vol. 1–3, 1982.
2 (a) H. Schmidbaur, in Gold: Progress in Chemistry, Bio-
chemistry and Technology, John Wiley & Sons, New York,
1999; (b) H. Schmidbaur, Nature, 2001, 413, 31;
(c) H. Schmidbaur, Gold Bull., 1990, 23, 11;
(d) H. Schmidbaur, Gold Bull., 2000, 33, 3;
(e) F. Scherbaum, A. Grömann, B. Huber, C. Krüger
and H. Schmidbaur, Angew. Chem., Int. Ed. Engl., 1988,
27, 1544; (f) F. Scherbaum, A. Grömann, G. Müller
and H. Schmidbaur, Angew. Chem., Int. Ed. Engl., 1989,
28, 463; (g) H. Schmidbaur, F. Scherbaum, B. Hubert 10 (a) E. J. Fernández, A. Laguna, J. M. López-de-Luzuriaga,
and G. Müller, Angew. Chem., Int. Ed. Engl., 1988, 27,
419.
3 (a) P. Pyykkö, Chem. Rev., 1997, 97, 597; (b) P. Pyykkö,
Angew. Chem., Int. Ed., 2004, 43, 4412; (c) P. Pyykkö, Inorg.
M. Monge, M. Montiel, M. E. Olmos and J. Pérez, Organo-
metallics, 2004, 23, 774; (b) E. J. Fernández, A. Laguna,
T. Lasanta, J. M. López-de-Luzuriaga, M. Montiel and
M. E. Olmos, Organometallics, 2008, 27, 2971.
Chim. Acta, 2005, 358, 4113; (d) P. Pyykkö, Chem. Soc. Rev., 11 (a) E. J. Fernández, J. M. López-de-Luzuriaga, M. Monge,
2008, 37, 1967.
M. E. Olmos, J. Pérez and A. Laguna, J. Am. Chem. Soc.,
2002, 124, 5942; (b) E. J. Fernández, J. M. López-de-Luzur-
iaga, M. E. Olmos, J. Pérez, A. Laguna and M. C. Lagunas,
Inorg. Chem., 2005, 44, 6012.
4 (a) J. M. Forward, J. P. Fackler Jr. and Z. Assefa, in Opto-
electronic Properties of Inorganic Compounds, ed.
D. M. Roundhill and J. P. Fackler Jr., Plenum, New York,
1999, pp. 195–226; (b) J. M. López-de-Luzuriaga, in Modern 12 M. Arca, T. Aroz, M. C. Gimeno, M. Kulcsar, A. Laguna,
Supramolecular Gold Chemistry, ed. A. Laguna. Wiley-VCH,
Weinheim, 2008, p. 347; (c) E. J. Fernández, A. Laguna and
J. M. López-de-Luzuriaga, Gold Bull., 2001, 34, 14, and
references therein.
5 (a) E. J. Fernández, A. Laguna, J. M. López-de-Luzuriaga
and M. Monge, Spanish Patent, P200001391, 2003;
(b) E. J. Fernández, J. M. López-de-Luzuriaga, M. Monge,
T. Lasanta, V. Lippolis, J. M. López-de-Luzuriaga,
M. Monge and M. E. Olmos, Eur. J. Inorg. Chem., 2011,
2288.
13 (a) S. Wang, J. P. Fackler Jr., C. King and J. C. Wang, J. Am.
Chem. Soc., 1988, 110, 3308; (b) S. Wang, G. Garzón,
C. King, J. C. Wang and J. P. Fackler Jr., Inorg. Chem., 1989,
28, 4623.
M. E. Olmos, J. Pérez, A. Laguna, A. A. Mohammed and 14 See for example: (a) S. R. Cooper and S. C. Rawle, in Bio-
J. P. Fackler Jr., J. Am. Chem. Soc., 2003, 125, 2022;
(c) E. J. Fernández, J. M. López-de-Luzuriaga, M. Monge,
M. Montiel, M. E. Olmos, J. Pérez, A. Laguna,
F. Mendizábal, A. A. Mohamed and J. P. Fackler Jr., Inorg.
Chem., 2004, 43, 3573; (d) E. J. Fernández, J. M. López-de-
Luzuriaga, M. Monge, M. E. Olmos, R. C. Puelles,
A. Laguna, A. A. Mohamed and J. P. Fackler Jr., Inorg.
Chem., 2008, 47, 8069.
inorganic Chemistry, Springer-Verlag, Berlin, Heidelberg,
1990, vol. 72; (b) S. R. Cooper, Acc. Chem. Res., 1988, 21,
141; (c) A. J. Blake and M. Schröder, Adv. Inorg. Chem.,
1990, 35, 1; (d) T. Adachi, M. D. Durrant, D. L. Hughes,
C. J. Pickett, R. L. Richards, J. Talarmin and T. Yoshiba,
J. Chem. Soc., Chem. Commun., 1992, 1464; (e) C. Landgrafe
and W. S. Sheldrick, J. Chem. Soc., Dalton Trans., 1996, 989;
(f) R. D. Adams, S. B. Fallon, J. L. Perrin, J. A. Queisser and
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11559–11570 | 11569

View Article Online
Paper
J. H. Yamamoto, Chem. Ber., 1996, 129, 313; (g) H.-J. Kim,
J.-H. Lee, I.-H. Suh and Y. Do, Inorg. Chem., 1995, 34, 796.
15 J. L. Shaw, J. Wolowska, D. Collison, J. A. K. Howard,
E. J. L. McInnes, J. McMaster, A. J. Blake, C. Wilson and
M. Schröder, J. Am. Chem. Soc., 2006, 128, 13827.
Dalton Transactions
J. Pérez, Inorg. Chim. Acta, 2005, 358, 4293;
(e) E. J. Fernández, A. Laguna, J. M. López-de-Luzuriaga,
M. Montiel, M. E. Olmos and J. Pérez, Organometallics,
2006, 25, 1689.
22 E. J. Fernández, A. Laguna, J. M. López-de-Luzuriaga,
M. E. Olmos and J. Pérez, Chem. Commun., 2003, 1760.
16 (a) G. H. Robinson, H. Zhang and J. L. Atwood, Organo-
metallics, 1987, 6, 887; (b) K. George, M. Jura, W. Levason, 23 V. J. Catalano and M. A. Malwitz, J. Am. Chem. Soc., 2004,
M. E. Light, L. P. Ollivere and G. Reid, Inorg. Chem., 2012, 126, 6560.
51, 2231; (c) C. Gurnani, M. Jura, W. Levason, R. Ratnani, 24 E. Pretsch, P. Bühlmann, C. Affolter, A. Herrera and
G. Reid and M. Webster, Dalton Trans., 2009, 1611;
(d) A. J. Blake, D. Fenske, W.-S. Li, V. Lippolis and
M. Schröder, J. Chem. Soc., Dalton Trans., 1998, 3961;
R. Martínez, in Structure Determination of Organic Com-
pounds, Springer-Verlag, Berlin, Heidelberg, New York,
2000.
(e) A. J. Blake, J. A. Greig and M. Schröder, J. Chem. Soc., 25 E. J. Fernández, M. C. Gimeno, A. Laguna, J. M. López-de-
Dalton Trans., 1991, 529; (f) F. Cheng, A. L. Hector,
W. Levason, G. Reid, M. Webster and W. Zhang, Chem.
Luzuriaga, M. Monge, P. Pyykkö and D. Sundholm, J. Am.
Chem. Soc., 2000, 122, 7287.
Commun., 2008, 5508; (g) G. R. Willey, A. Jarvis, J. Palin and 26 E. J. Fernández, A. Laguna and J. M. López-de-Luzuriaga,
W. Errington, J. Chem. Soc., Dalton Trans., 1994, 255; Dalton Trans., 2007, 1969.
(h) H.-J. Kuppers, K. Wieghardt, B. Nuber and J. Weiss, 27 (a) D. Sellmann and L. Zapf, Angew. Chem., 1984, 96, 799;
Z. Anorg. Allg. Chem., 1989, 577, 155; (i) G. R. Willey,
M. T. Lakin, M. Ravindra and N. W. Alcock, J. Chem. Soc.,
Chem. Commun., 1991, 271.
(b) J. Buter and R. M. Kellog, J. Org. Chem., 1981, 46, 4481;
(c) J. J. H. Edema, J. Buter and R. M. Kellog, Tetrahedron,
1991, 50, 2095.
17 (a) Sanaullah, K. Kano, R. S. Glass and G. S. Wilson, J. Am. 28 G. M. Sheldrick, SHELXL97, Program for Crystal Structure
Chem. Soc., 1993, 115, 592; (b) E. Wyer, G. Giuccardo, Refinement, University of Göttingen, Germany, 1997.
V. Leigh, H. Muller-Bunz and M. Albrecht, J. Organomet. 29 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
Chem., 2011, 696, 2882; (c) A. J. Blake, N. R. Champness,
S. M. Howdle and P. E. Webb, Inorg. Chem., 2000, 39, 1035;
(d) D. Parker, P. S. Roy, G. Ferguson and M. M. Hunt,
Inorg. Chim. Acta, 1989, 155, 227; (e) W. N. Setzer, Q. Guo,
G. J. Grant, J. L. Hubbard, R. S. Glass and
D. G. VanDerveer, Heteroat. Chem., 1990, 1, 317;
(f) M. L. Helm, L. L. Hill, J. P. Lee, D. G. VanDerveer and
G. J. Grant, Dalton Trans., 2006, 3534; (g) M. Wilhelm,
S. Deeken, E. Berssen, W. Saak, A. Lutzen, R. Koch and
H. Strasdeit, Eur. J. Inorg. Chem., 2004, 2301.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
GAUSSIAN 09 (Revision A.1), Gaussian, Inc., Wallingford CT,
2009.
18 A. J. Blake, J. A. Greig and M. Schröder, J. Chem. Soc.,
Dalton Trans., 1992, 2987.
19 R. Usón, A. Laguna, M. Laguna, B. R. Manzano, P. G. Jones
and G. M. Sheldrick, J. Chem. Soc., Dalton Trans., 1984, 285.
20 R. Usón, A. Laguna, M. Laguna, B. R. Manzano and
A. Tapia, Inorg. Chim. Acta, 1985, 101, 151.
21 (a) E. J. Fernández, A. Laguna, J. M. López-de-Luzuriaga,
M. E. Olmos and J. Pérez, Dalton Trans., 2004, 1801;
(b) O. Crespo, E. J. Fernández, P. G. Jones, A. Laguna, 30 C. Adamo and V. Barone, J. Chem. Phys., 1999, 110,
J. M. López-de-Luzuriaga, A. Mendía, M. Monge and 6158.
M. E. Olmos, Chem. Commun., 1998, 2233; 31 D. Andrae, U. Haeussermann, M. Dolg, H. Stoll and
(c) E. J. Fernández, A. Laguna, J. M. López-de-Luzuriaga, H. Preuss, Theor. Chim. Acta, 1990, 77, 123.
F. Mendizabal, M. Monge, M. E. Olmos and J. Pérez, 32 A. Schäfer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992,
Chem.–Eur. J., 2003, 9, 456; (d) E. J. Fernández, A. Laguna, 97, 2571.
J. M. López-de-Luzuriaga, M. Montiel, M. E. Olmos and 33 T. H. Dunning Jr., J. Chem. Phys., 1994, 100, 5829.
11570 | Dalton Trans., 2013, 42, 11559–11570
This journal is © The Royal Society of Chemistry 2013