Silver(i), gold(i) and palladium(II) complexes of a NHC-pincer ligand with an aminotriazine core: A comparison with pyridyl analogues

Article abstract of DOI:10.1039/c5dt04213e

Dinuclear silver, di- and tetra-nuclear gold, and mononuclear palladium complexes with chelating C,N,C diethylaminotriazinyl-bridged bis(NHC) pincer ligands were prepared and characterised. The silver and gold complexes exist in a twisted, helical conformation in both the solution- and the solid state. In contrast, an analogous dinuclear gold complex with pyridyl-bridged NHCs exists in a linear conformation. Computational studies have been performed to rationalise the formation of twisted/helical vs. linear forms.

Full text of DOI:10.1039/c5dt04213e

Dalton
Transactions
View Article Online
View Journal
PAPER
Silver(I), gold(I) and palladium(II) complexes of a
NHC-pincer ligand with an aminotriazine core: a
comparison with pyridyl analogues†
Cite this: DOI: 10.1039/c5dt04213e
Jamila Vaughan,a Damien J. Carter,*b Andrew L. Rohl,c Mark I. Ogden,a
Brian W. Skelton,d Peter V. Simpson*a and David H. Brown*a
Dinuclear silver, di- and tetra-nuclear gold, and mononuclear palladium complexes with chelating C,N,C
diethylaminotriazinyl-bridged bis(NHC) pincer ligands were prepared and characterised. The silver and
gold complexes exist in a twisted, helical conformation in both the solution- and the solid state. In
contrast, an analogous dinuclear gold complex with pyridyl-bridged NHCs exists in a linear conformation.
Computational studies have been performed to rationalise the formation of twisted/helical vs.
linear forms.
Received 27th October 2015,
Accepted 20th November 2015
DOI: 10.1039/c5dt04213e
www.rsc.org/dalton
NHCs and it was not until more recently that directly
bonded benzimidazolinylidene-pyridyl pincers (e.g. 2) were
Introduction
N-Heterocyclic carbene (NHC) metal complexes are prevalent reported.23–27
in modern organic and organometallic chemistry.1,2 With
examples that span the majority of the d-block elements,
NHC–metal complexes have been investigated for a range of
applications and uses, including catalysis (particularly Pd and
Ru),3–8 as precursors to other metal complexes (Ag)9,10 and
more recently biomedical applications (mainly Au and Ag) as
antibacterial, anticancer, and antiparasitical agents.11–15
The synthesis of NHCs bearing pyridyl donor groups has
been of considerable interest, and they have been explored in-
depth – either bound directly to the NHC ring (as in 1), or with
a spacer (often methylene).16 In particular, the imidazole-
based NHC-pyridine ‘pincer’ 1 has been coordinated to a
diverse range of metal centres,17–22 and in many cases the
resulting complexes have been used for catalysis applications.
These examples have focused on the use of imidazole derived
aDepartment of Chemistry, Curtin University, GPO Box U1987, Perth WA 6845,
Australia. E-mail: peter.simpson@curtin.edu.au, d.h.brown@curtin.edu.au,
d.carter@curtin.edu.au
bScience & Maths Education Centre, Nanochemistry Research Institute & Department
of Chemistry, Curtin University, GPO Box U1987, Perth WA 6845, Australia
cCurtin Institute for Computation, Nanochemistry Research Institute & Department
of Chemistry, Curtin University, GPO Box U1987, Perth WA 6845, Australia
dCentre for Microscopy, Characterisation and Analysis, The University of Western
Recently, the group of Strassner reported dinuclear silver and
gold complexes containing bis(NHC) ligands incorporating
anionic triazinone bridging units, where the negative charge is
delocalised over the triazinone ring (e.g. 3 and 4a,b).28,29 Short
M–N contacts exist between the metal and the closest triazi-
none nitrogen atom, suggesting a significant interaction that
may contribute to the formation of the twisted “double
helical” conformation observed in the solid-state. This double
Australia, 35 Stirling Highway, Crawley, WA 6009, Australia
†Electronic supplementary information (ESI) available: Full cif files. Crystal and
structure refinement data of all structures and a projection of 5·2PF6. CCDC
1430963 (5·2PF6), 760649 (6·2PF6), 1430964 (7·2PF6), 1430965 (8·2PF6), 1430966
(9·PF6) and 1430967 (11·2BPh4). For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c5dt04213e
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
helical, or twisted, conformation is also commonly observed in water. In the crystal structure of 5·2PF6 (Fig. S1†) the two
for pyridyl-bridged bis(NHC) dinuclear complexes,22,25,30,31 imidazolium units are hydrogen bonded through each H2
and is thought to form via the reaction of the mono-carbene atom to a fluorine atom of the hexafluorophosphate anion,
intermediate with another equivalent of the intermediate. which could give an indication that coordination to a metal at
Here we complement Strassner’s initial study by reporting the this position is favoured upon deprotonation.32 The reaction
preparation of several silver, gold, and palladium complexes of 5·2PF6 with silver oxide in acetonitrile, in the presence of
bearing bis(NHC) diethylaminotriazinyl-bridged pincer 3 Å molecular sieves, afforded the dinuclear silver(I) complex
ligands, as well as a gold complex bearing a bis(NHC) pyridyl- salt 6·2PF6 in 64% yield (Scheme 1). The salt was isolated as a
bridged pincer ligand. We also describe a detailed compu- white powder and could be recrystallised to afford colourless
tational, and solution- and solid-state, investigation of the pro- crystals. The salt readily dissolves in polar solvents such as
pensity of these types of complexes to form “double helical” acetone and acetonitrile.
twisted, or non-twisted linear conformations.
Reaction of the silver(I) complex salt 6·2PF6 with bromido
(tetrahydrothiophene)gold(I) in acetonitrile afforded the gold(I)
complex salt 7·2PF6 as yellow powder (Scheme 2). Recrystallisa-
tion of the yellow powder afforded a mixture of two different
types of crystals. The bulk of the crystals were colourless and
confirmed by NMR and single-crystal X-ray studies (see below)
to be the 7·2PF6. A very minor component of the crystals were
Results and discussion
Synthesis
The reaction of 1-butylbenzimidazole with 1,3-dichloro-5-di-
ethylaminotriazine in acetonitrile afforded the benzimidazo- bright yellow, which were confirmed by single X-ray studies to
be a tetranuclear gold complex 8·2PF6 (Scheme 3, Solid-state
studies section). The crystal structure of 8·2PF6 was modelled
lium salt 5·2Cl in high yield (92%), which was converted to
5·2PF6 by salt metathesis with potassium hexafluorophosphate
Scheme 1 Synthesis of dinuclear silver complex 6·2PF6.
Scheme 2 Synthesis of silver and palladium complexes 7·2PF6 and 9·PF6.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Scheme 3 Tetranuclear gold complex 8 (X = Br/Cl).
as a 50 : 50 mixed Br/Cl, with the chloride impurity possibly
arising from incomplete salt metathesis of 5·2Cl or during the
preparation of bromido(tetrahydrothiophene)gold(I). Numer-
ous attempts to prepare 8 by variations of procedures and
reagent ratios lead only to the formation of 7·2PF6 with only
trace quantities of 8·2PF6. The tetranuclear complex salt 8·2PF6
could only be isolated by manual separation of the yellow crys-
tals that formed as a minor component during the purification
of the bulk material 7·2PF6. The silver complex 6·2PF6 could
also undergo transmetallation with palladium(II) chloride in
acetonitrile to afford the mono-nuclear palladium complex
9·PF6 (Scheme 2).
We have previously reported the silver(I) complex
10·2BPh4.25 To enable comparison of the triazinyl-pincer com-
plexes with their pyridyl analogues we also prepared the di-
nuclear gold(I) complex 11·2BPh4 by reaction of 10·2BPh4 with
bromido(tetrahydrothiophene)gold(I) in acetonitrile, which
was isolated as colourless powder/crystals (Scheme 4). The
complex adopt a linear type structure (see Solid state studies
section), similar to other dinuclear gold(I) complexes of bis
(NHC) ligands bridged by para-phenyl, stilbenyl, and anthrace-
nyl groups.33–35
Scheme 4 Synthesis of 11·2BPh4.
the pyridine analogue 10.25 The 13C NMR spectrum of a solu-
tion of 6·2PF6 in CD3CN (or d6-DMSO) displayed two sharp
doublets, centred at δ 194, for the carbene carbons (J = 187
and 215 Hz), consistent with coupling to 107Ag and 109Ag.10 In
addition to the splitting of the carbene carbon signal, for solu-
tions in d6-DMSO, the signals for the bridge-head carbons of
the benzimidazole ring (C8 and C9) appear as doublets (split-
ting 4 and 6 Hz), consistent with three-bond coupling to the
silver. The sharp carbene carbon signals and the long range
coupling suggest that the Ag–carbene bonds are relatively
strong, or that the silver-centres are held in the complex in a
static environment. Previous studies have suggested that the
lack of C–Ag coupling may be due to labile Ag–C bonds.36,37 It
may be possible that weak N⋯Ag interactions involving the tri-
azine in 6 (see below) help to stabilise the complex. Two
similar ‘twisted’ dinuclear structures, 1025 and [12Ag2]2+,31,38
do not display the long-range coupling, and in both cases the
contacts between the pyridyl nitrogens and the silver centres
are longer than those in 6.
NMR studies
1
The H NMR spectrum of a solution of 6·2PF6 in CD3CN dis-
plays signals consistent with the structure, however, of particu-
lar interest is the signal for the methyl groups of the butyl
chain, which has an up-field chemical shift of δ 0.34 (δ 0.26 in
d6-DMSO). This up-field shift, when compared to the ana-
logous signal for 5·2PF6 (δ 0.98 in d6-DMSO), is attributed to
magnetic shielding of the butyl chain by the aromatic groups
on the opposing ligand in the dinuclear structure. The proxi-
mity of the butyl chain and the aromatic groups is observed in
the solid-state structure. These observations indicate that the
‘twisted’ dinuclear structure of 6 is prevalent in solution, and
not just the solid-state. Similar observations were made with
1
The H NMR spectrum of a solution of the gold–triazinyl–
NHC pincer complex 7·2PF6 in CD3CN displays signals similar
to that of the silver analogue 6·2PF6. The signals of the methyl
groups are up-field at δ 0.34, suggesting that the gold complex
also exists in the twisted conformation in solution, as well as
in the solid-state (see Solid-state studies section). As would be
expected for gold(I) complexes, the signal for the carbene
carbons is a singlet at δ 192. Interestingly the 1H NMR spec-
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
trum of a solution of the gold–pyridyl–NHC pincer complex
11·2BPh4 in d6-DMSO displays a signal attributed to the
methyl groups of the butyl chain at δ 0.82, downfield in com-
parison to the analogous signal in the triazinyl complex
7·2PF6. The more conventional chemical shift for an alkyl
methyl group in 11 is consistent with the absence of magnetic
shielding and suggests that in solution the dinuclear gold–
NHC–pyridyl pincer complex 11 adopts a non-twisted confor-
mation, presumably similar to that identified in the solid-state
for 11 (see Solid-state studies section). As expected the 1H
NMR spectrum of a solution of the mononuclear palladium
complex 9·PF6 in d6-DMSO displays a signal attributed to the
methyl groups of the butyl chain at δ 0.87. In this case no mag-
netic shielding of the methyl group is possible.
Luminescence studies
The luminescence of the gold(I) complexes was studied, and is
shown in Fig. 1. It proved extremely difficult to completely
separate the two triazinyl–NHC pincer gold(I) complexes 7·2PF6
and 8·2PF6, with manual separation of a crystalline mixture
proving the only adequate method. Still, quantitative photo-
physical measurements could not be made due to small
amounts of contamination present in the crystals. Solid-state
luminescence studies indicated that the tetra-nuclear complex
8·2PF6 was highly emissive (λem = 528 nm) whereas the di-
nuclear complexes 7·2PF6 and 11·2BPh4 did not exhibit any
significant luminescence. The shorter Au⋯Au contact in 8 is
presumably responsible for the luminescence (see Solid-state
studies).
Fig. 2 Projection of the cation 6. Ellipsoids displayed at 50% probability;
hydrogen atoms have been omitted for clarity. Selected bond dist-
ances (Å) and angles (°): Ag(1)–C(112) 2.098(3), Ag(1)–C(232) 2.100(3),
Ag(1)⋯N(121) 2.786(3), Ag(1)⋯N(221) 2.899(2), Ag(1)⋯Ag(2) 3.0899(3),
Ag(2)–C(132) 2.085(2), Ag(2)–C(212) 2.097(3), Ag(2)⋯N(121) 2.880(2),
Ag(2)⋯N(221) 2.764(2), C(112)–Ag(1)–C(232) 165.51(11), C(132)–Ag(2)–
C(212) 164.74(13).
Solid-state studies
2.764(2) Å; cf. 10:25 Ag⋯N 2.984(1), 3.012(1)]. As a result of
these interactions, the C–Ag–C angles now deviate significantly
from linearity, the C(112)–Ag(1)–C(232) and C(132)–Ag(2)–
C(212) angles being 165.51(11) and 164.74(13)° respectively.
The Ag(1)⋯Ag(2) distance has also been substantially reduced
to 3.0899(3) Å cf. 3.7848(2) Å for 10. The orientations of the
pendant rings in 6 are twisted relative to the central plane, but
less than those of 10. The angles between the planes of the
pendant groups and the central triazine rings are 14.19(9) and
31.32(9)° (ligand 1) and 18.64(8) and 26.29(8)° (ligand 2)
compared with 42.57(4) and 39.91(4)° in 10.
The cation 6 is shown in Fig. 2 (with selected geometries) and
consists of a [Ag2L2]2+ dimer with the benzimidazolin-2-
ylidene groups of each ligand coordinated to different Ag
atoms. There is also one molecule of acetonitrile per dimer in
the lattice. The structure is similar to that of the pyridine ana-
logue 10, except that the N(n21) atoms of the triazine groups
are now weakly coordinated to the Ag atoms [Ag(1)–N(121),
N(221) 2.786(3), 2.899(2) Å and Ag(2)–N(121), N(221) 2.880(2),
The bond lengths of the diethylaminotriazine moiety (see
Fig. 3), and the co-planar arrangement of the N(CH2)2 units
with respect to the triazine rings, suggest the influence of a
pyridone–iminium resonance structure.39 A contribution of
this structure would result in the enhancement of the nucleo-
philicity of the ring-nitrogen para to the diethylamino groups.
Such enhanced nucleophilicity may explain the shorter N⋯Ag
contacts observed in 6, compared to that of 10.
The structure of silver complex 6 begs comparison with
those of triazinone- and pyridyl-bridged bis(NHC)–Ag com-
plexes 4a28 and 10,25 respectively (Fig. 4). Although the NHC
ligands in 4a contain different N-substituents and are derived
from imidazole rather than benzimidazole for 6 and 10,
the electronic effects caused by the three types of bridging
Fig. 1 Emission of 8·2PF6 (red trace) and 7·2PF6 upon excitation at
365 nm.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
relatively weak nucleophilicity of the pyridyl nitrogen in 10 cf.
6 and 4a leads to longer N–Ag contacts and a greater Ag–Ag
distance. For completeness, it is worth noting that although
Ag–N interactions are present in these pincer compounds, in
other cases involving close d10 metal–arene contacts, proximity
alone does not always indicate the presence of agnostic
interactions.40
The cation of 7 is shown in Fig. 5 and consists of a
[Au2L2]2+ dimer with the benzimidazolin-2-ylidene groups of
each ligand coordinated to different Au atoms. There is also
one molecule of acetonitrile per dimer in the lattice. The struc-
ture is isomorphous with that of the silver analogue 6. The tri-
azine groups are weakly coordinated to the Au atoms, Au(1)–
N(121), N(221) 2.902(4), 2.945(3) Å and Au(2)–N(121), N(221)
2.937(3), 2.890(3) Å, while the Au(1)⋯Au(2) distance is 3.3955(4)
Å. In the same way that the aminotriazinyl- and triazinone-
bridged silver complexes displayed similar core geometries,
the analogous distances in 7 are similar to those for gold tri-
azinone complex 4b, also reported by Strassner and co-
workers.29 As a result of these interactions, the C–Au–C angles
now deviate significantly from linearity, the C(112)–Au(1)–
C(232) and C(132)–Au(2)–C(212) angles being 170.50(15) and
170.58(17)°, respectively, but with less deviation than the
Fig. 3 Bond lengths for the aminotriazine moieties of 6, and a possible
resonance contribution to the structure.
heterocycles contribute significantly to the overall geometry corresponding values in the Ag analogue 6. The orientations of
around the Ag–Ag core. The enhanced nucleophilicity of the the pendant rings are twisted relative to the central plane. The
triazinyl nitrogen in 6, due to the pyridine–iminium resonance angles between the planes of the pendant groups and the
form, leads to short N–Ag contacts, and also contribute to a central triazine rings are 16.0(1) and 32.9(1)° (ligand 1) and
shortening of the Ag–Ag bond. Similarly, in 4a, the formal 20.3(1) and 27.5(1)° (ligand 2) and similar to those of 6.
negative charge on the triazinyl nitrogen presumably leads to
The cation 11, where pyridyl groups now bridge the NHC
short N–Ag contacts and a short Ag–Ag bond. Therefore, the units, adopts a very different conformation to that of 7, and is
core geometries around 6 and 4a are very similar, with only a seen in Fig. 6. The cation consists of a [Au2L2]2+ dimer situated
slight shortening of the Ag–Ag bond in 4a relative to 6. The on a crystallographic inversion centre with the benzimidazo-
Fig. 4 Projections and selected bond distances (Å) of (a) aminotriazinyl-bridged cation 6; (b) triazinone-bridge complex 4a from Strassner and co-
workers;28 and (c) pyridyl-bridged cation 10 from Brown and co-workers.25 Ellipsoids displayed at the 50% probability level; hydrogen atoms and the
(benz)imidazolyl N-substituents have been omitted for clarity.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 6 Projections of the cation 11 showing (a) the top view of the
molecule and (b) the side view, highlighting the ‘V’ shaped nature of the
benzimidazolyl units. Ellipsoids are displayed at the 50% probability
level; hydrogen atoms and the butyl chains in (b) have been omitted for
clarity. Selected bond distances (Å) and angles (°): Au(1)–C(12) 2.006(2),
Au(1)–C(32’) 2.012(3), C(12)–Au(1)–C(32’) 175.43(9).
Fig. 5 Projection of the cation 7. Ellipsoids are displayed at the 50%
probability level; hydrogen atoms have been omitted for clarity. Selected
bond distances (Å) and angles (°): Au(1)–C(112) 2.023(4), Au(1)–C(232)
2.025(4), Au(1)⋯N(121) 2.902(4), Au(1)⋯N(221) 2.945(3), Au(1)⋯Au(2)
3.3955(5), Au(2)–C(132) 2.009(3), Au(2)–C(212) 2.027(3), Au(2)⋯N(121)
2.937(3), Au(2)⋯N(221) 2.890(4), C(112)–Au(1)–C(232) 170.50(15), C(132)–
Au(2)–C(212) 170.58(17).
In the structure of tetranuclear complex 8, shown in
Fig. 7(a), each ligand bridges between two gold atoms with the
lin-2-ylidene groups of each ligand coordinated to different Au terminal gold atoms capped by halide ions, thus forming an
atoms. Thus, the coordination about the Au atom is essentially extended helical structure. Each of the gold atoms is two co-
linear, with the C12–Au1–C32′ angle being 175.43(9)°, the ordinate with Au–C distances of 1.965(10) and 1.976(9) Å for the
prime referring to the centrosymmetrically related atom at 1 − terminal gold atoms with those to the central gold atoms lying
x, 1 − y, 1 − z. Relevant geometries are given in the caption of in the range 2.017(10)–2.033(9) Å. There are, however, weak
Fig. 6. Since the molecule is situated on an inversion centre, bridging interactions between the nitrogen atom of each of the
the four coordinated carbon atoms are coplanar, the Au(1) triazine groups and adjacent gold atoms; the Au–N distances
deviation from this plane being 0.080(2) Å. The ligand is not lie in the range 2.846(7)–2.976(7) Å. The coordination about
planar, but forms a ‘V’ shape (see Fig. 6(a and b)), with the the gold atoms deviates significantly from linearity with the
pyridyl ring at the base. The angles between the coordination C–Au–C angles being 175.8(4) and 176.1(4)° and the C–Au–X
Au2C4 plane and the planes of the two pendant rings are 27.93 angles 170.9(3) and 175.5(3)°. The four gold atoms are not line-
(5)° and 28.53(5)°. The Au(1)–N(22) distances are greater than arly arranged with the Au⋯Au⋯Au angles being 143.70(2) and
4 Å, indicative of no interaction, while the Au1⋯Au1′ distance 147.07(2)°. The Au⋯Au distances are Au(1)⋯Au(2) 3.2101(10),
of 6.6193(5) Å is too great to lead to any aurophilic inter- Au(2)⋯Au(3) 3.2709(7) and Au(3)⋯Au(4) 3.1395(8) Å. A more in
actions. A similar linear conformation of gold atoms albeit depth discussion of the structure of 8 can be found in the ESI.†
with
a
different pyridyl arrangement has recently been
The coordination around the palladium centre in complex
reported for an analogous gold complex with hexyl groups 9, seen in Fig. 8(a), is distorted square planar. The three rings
instead of butyl groups on the benzimidazolyl units,41 and so are essentially coplanar with the atoms with the maximum
the helical/twisted nature of observed in 7 may be attributable deviations being C(36) (−0.083(1) Å) and C(37) (−0.071(1) Å)
to the aminotriazinyl bridge. Furthermore, it is interesting to with the Pd deviation being 0.017(1) Å. The dihedral angles
note that similar dinuclear gold complexes where the NHCs between the coordination plane and the three rings 1–3 are
are derived from imidazole instead of benzimidazole display a 1.55(2), 2.26(3) and 3.05(3)°. There is a weak interaction
twisted conformation similar to that of 6 and 7.22,41
between the PF6− anion and the Pd atom, the distance Pd(1)–F(1)
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 7 Projections of (a) cation 8 and (b) crystal packing showing the cations lying in sheets. Ellipsoids are displayed at the 30% probability level;
hydrogen atoms and the butyl chains have been omitted for clarity. Selected bond distances (Å) and angles (°): Au(1)–C(112) 1.965(10), Au(1)–Br, Cl(1)
2.3686(14), Au(1)⋯Au(2) 3.2101(10), Au(2)–C(132) 2.033(9), Au(2)–C(212) 2.033(8), Au(2)⋯Au(3) 3.2709(7), Au(3)–C(232) 2.017(10), Au(3)–C(312)
2.022(8), Au(3)⋯Au(4) 3.1395(8), Au(4)–C(332) 1.976(9), Au(4)–Br, Cl(2) 2.3838(13), C(112)–Au(1)–Br, Cl(1) 170.9(3), C(132)–Au(2)–C(212) 175.8(4),
Au(1)⋯Au(2)⋯Au(3) 143.70(2), C(232)–Au(3)–C(312) 176.1(4), Au(4)⋯Au(3)⋯Au(2) 147.07(2), C(332)–Au(4)–Br, Cl(2) 175.5(3).
is 3.2928(7) Å. In the unit cell, there are close approaches of have been used to study a range of solid state systems such as
atoms of pairs of centrosymmetrically related cations resem- monosaccharides,42 amino acids,43 epoxydihydroarsanthrene
bling π-stacking, with the shortest distances being Pd(1)⋯N analogues,44 polycyclic aromatic hydrocarbons,45 and many
(21) 3.4456(9), N(13)⋯C(32) 3.383(1), N(31)⋯C(13A) 3.391(1) Å others.46–53
(Fig. 8(b)).
There have been a number of gas–phase studies of large
supramolecular systems54–59 using vdW-corrected DFT, with a
number of these studies using the S12L set of large dispersion-
bound host–guest systems.58 There are a limited number of
studies of large crystalline supramolecular systems, such as
calculations of p-tert-butylcalix[4]arene inclusion com-
pounds,60,61 metal organic framework materials62,63 and rotax-
ane structures.64
In Table 1 we report the lattice parameters of DFT opti-
mised crystal structures of 10·2BPh4 and 11·2BPh4 containing
either Ag or Au metal atoms. The calculated structures
compare closely to the corresponding observed experimental
crystal structures, with a variation of 0.8–1.5% in the lattice
Computational studies
In an effort to find the origin of the intriguing difference in
conformation of the silver- and gold–pyridyl–NHC pincer com-
plexes (10 and 11 respectively), computational studies were
carried out. The use of traditional density functional theory
(DFT) methods to study soft matter or molecular crystals has
in the past been hampered by the lack of potentially important
van der Waals (vdW) or dispersion forces. In recent years, the
development and improvement of methods to include dis-
persion forces in DFT now makes this routinely possible. van
der Waals corrected DFT calculations of molecular crystals
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 1 DFT optimised lattice parameters of 10·2BPh4 and 11·2BPh4
with both Ag and Au as the coordinated atom, compared to the respect-
ive experimental crystal structures
Metal
atom
a (Å)
b (Å)
c (Å)
β (°)
10·2BPh4 structure
Experiment
DFT
Ag
Ag
24.003
24.323
(+1.33%)
24.357
17.332
17.562
(+1.33%)
17.540
20.468
20.630
(+0.79%)
20.624
104.16
104.57
(+0.39%)
104.31
DFT
Au
11·2BPh4 structure
Experiment
DFT
Au
Au
13.415
13.610
(+1.45%)
13.690
17.869
18.106
(+1.32%)
18.012
17.508
17.731
(+1.27%)
17.700
100.72
100.46
(−0.26%)
100.06
DFT
Ag
Table 2 Relative interaction energies of Ag and Au in isolated com-
plexes and the crystal structures of 10·2BPh4 and 11·2BPh4
Relative interaction
energy (kJ mol−1
per metal atom
)
Metal atom
Ag
Ligand structure
Isolated
Crystal
10·2BPh4 (twisted)
11·2BPh4 (non-twisted)
0.00
+20.27
0.00
+1.00
Au
10·2BPh4 (twisted)
11·2BPh4 (non-twisted)
0.00
+13.31
+12.21
0.00
atoms by 20.27 and 13.31 kJ mol−1 (per metal atom), respect-
ively. In the crystal structure, the relative interaction energies
show that Ag is more stable in the twisted 10·2BPh4 structure
by 1.00 kJ mol−1 (per metal atom), and Au is more stable in
the non-twisted 11·2BPh4 structure by 12.21 kJ mol−1 (per
metal atom). This trend in relative interaction energies
matches exactly with the experimental observations of the
stable crystal structures, and solution phase studies, observed
for Ag and Au atoms with the two different ligand environ-
ments. The 1 kJ mol−1 preference of Ag metal for the twisted
10·2BPh4 structure is significant, as although we found an
average lattice energy error of 3.85 kJ mol−1 for the C21 refer-
ence set of molecule crystals,42 for systems containing the
same species in the same amounts, the relative error will be
much smaller due to error cancellation.
Fig. 8 Projection of (a) the cation 9 and (b) close contacts between
centrosymmetrically related molecules of 9. Ellipsoids are displayed at
the 50% probability level; hydrogen atoms have been omitted for clarity.
Selected bond distances (Å) and angles (°): Pd(1)–N(21) 1.9497(8), Pd(1)–
C(12) 2.0317(10), Pd(1)–C(32) 2.0384(10), Pd(1)–Cl(1) 2.2935(3), N(21)–
Pd(1)–C(12) 78.60(3), N(21)–Pd(1)–C(32) 78.01(3), C(12)–Pd(1)–C(32)
156.59(4), N(21)–Pd(1)–Cl(1) 178.61(3), C(12)–Pd(1)–Cl(1) 100.67(3), C(32)–
Pd(1)–Cl(1) 102.74(3).
Examining in more detail the isolated ligand structures, it
parameters. We have previously reported the excellent perform- is perhaps not unsurprising to see that the twisted 10 structure
ance of the vdW-DF2(rPW86) functional for the C21 reference is preferred for both Ag and Au. In the isolated case, the
set of molecular crystals and monosaccharide structures,42 ligand backbone is free to contort to accommodate the co-
with mean deviations of the lattice parameters of ∼1.2%.
ordinated metal atoms, without any interaction with the
In Table 2 we report the relative interaction energies, cor- crystal environment or solvation considerations. The metal
rected for basis set superposition error, for Ag and Au in the atom–metal atom distances in the isolated complexes are
isolated 10·2BPh4 and 11·2BPh4 ligand structures and the noticeably shorter than those found in the corresponding cal-
crystal structures. For the isolated complexes, calculations culated crystal structure, with the Ag–Ag separation distance
predict the twisted 10 structure is preferred for both Ag and Au shortening from 4.14 to 3.16 Å and the Au–Au separation
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
distance shortening from 4.23 to 3.54 Å. The shorter metal (tetrahydrothiophene)gold(I)69 and the silver complex
25
atom–metal atom distances indicate a stronger “metallophil- 10·2BPh4 were prepared by literature procedures.
lic” interaction between the metal atoms in the isolated struc-
Synthesis of 1,1′-[2,6-(4-diethylaminotriazinyl)]di(3-n-butyl-
tures. These metallophillic interactions between two d10 metal benzimidazolium) dichloride 5·2Cl. A solution of 1-n-butyl-
centres have been recently reviewed.65,66 In the less-favoured, benzimidazole (1.25 g, 7.2 mmol) in acetonitrile (5 mL) was
isolated, non-twisted 11 structure, the metal atom–metal atom added to a solution of 1,3-dichloro-5-diethylaminotriazine
distances increased slightly compared to those in the corres- (0.58 g, 2.6 mmol) in acetonitrile (5 mL). The resulting solu-
ponding calculated crystal structure, with the Ag–Ag separation tion was heated at reflux overnight. The solution was then
distance increasing from 6.78 to 7.16 Å and the Au–Au separ- cooled and diluted with diethyl ether (20 mL). The resulting
ation distance increasing from 6.83 to 7.15 Å. The metal atom–N precipitate was collected, washed with diethyl ether
atom distances and metal atom–C atom distances in both the (3 × 15 mL) and dried under vacuum to afford a white powder
isolated complexes and crystal structures are largely unaffected (1.45 g, 92%). δH (200 MHz; d6-DMSO) 0.96 (6 H, t, J 8,
by the choice of Ag or Au coordinating atoms, with only very 2 × CH2CH2CH3), 1.40 (6 H, t, J 6, 2 × NCH2CH3), 1.44 (4 H, m,
minor variations in separation distances. Based on these calcu- 2 × CH2CH2CH3), 2.05 (4 H, m, 2 × NCH2CH2), 3.96 (4 H, q, J 6,
lations, the crystal environment (or solvation in solution) plays 2 × NCH2CH3), 4.81 (4 H, t, J 8, 2 × NCH2CH2), 7.80–7.96
a significant but subtle role in determining the conformation (4 H, m, 4 × Ar CH), 8.31 (2 H, d, J 8, 2 × Ar CH), 8.72 (2 H, d,
of these metal complexes.
J 8, 2 × Ar CH) and 12.32 (2 H, s, NCHN); δC (50 MHz;
d6-DMSO) 12.3 (CH3), 13.3 (CH3), 19.0 (CH2), 30.5 (CH2),
43.4 (CH2), 47.5 (CH2), 114.6 (CH), 117.0 (CH), 127.5 (CH),
128.7 (CH), 128.8 (C), 132.2 (C), 144.7 (CH), 160.7 (C) and
164.0 (C); (Found: C, 57.41; H, 7.00; N, 18.58. C29H38N8Cl2·2H2O
Conclusion
We have reported the synthesis of several silver, gold, and pal- requires C, 57.52; H, 6.99; N, 18.50%).
ladium bis(NHC) complexes containing a pyridyl- or diethyl-
Synthesis of 1,1′-[2,6-(4-diethylaminotriazinyl)]di(3-n-butyl-
aminotriazinyl-bridge. The introduction of the triazine moiety benzimidazolium) bis(hexafluoro-phosphate) 5·2PF6. A solu-
has resulted in significant structural changes compared to a tion of 5·2Cl (0.23 g, 0.40 mmol) in water (10 mL) was added
pyridine analogue. Solution- and solid-state studies have to a solution of KPF6 (0.45 g, 2.4 mmol) in water (10 mL).
revealed that the triazinyl containing gold complexes adopt a The resulting precipitate was collected, washed with water
twisted, “helical” conformation, while simple pyridyl-bridged (2 × 10 mL) and diethyl ether (3 × 10 mL), and dried to afford a
systems are linear. Computational studies suggest that the white powder (0.26 g, 82%). δH (200 MHz; d6-DMSO) 0.98 (6 H,
twisted conformation is generally more stable in vacuo, and t, J 7, 2 × CH2CH2CH3), 1.38 (6 H, t, J 7, 2 × NCH2CH3), 1.49
the crystal environment stabilises the linear conformation in (4 H, m, 2 × CH2CH2CH3), 2.02 (4 H, m, 2 × NCH2CH2), 3.97
the gold pyridyl-bridged systems; an exception being in gold (4 H, q, J 7, 2 × NCH2CH3), 4.73 (4 H, t, J 8, 2 × NCH2CH2),
complexes containing imidazole-based carbene ligands, which 7.82–7.97 (4 H, m, 4 × Ar CH), 8.33 (2 H, d, J 7, 2 × Ar CH), 8.76
also crystallise in the twisted form. In contrast, the analogous (2 H, d, J 7, 2 × Ar CH) and 10.77 (2 H, s, NCHN); (Found:
silver complex is more stable in the helical conformation in C, 44.32; H, 4.90; N, 14.23. C29H38N8P2F12 requires C, 44.17;
isolation, and in the crystal environment. The remarkable H, 4.86; N, 14.21%).
tetranuclear gold complex 8 was also synthesised in trace
Synthesis of bis{2,6-di(3-n-butylbenzimidazolin-2-ylidene)-
amounts as a by-product in the preparation of 7, and was 4-diethylaminotriazine}disilver(I) bis(hexafluorophosphate)
characterised by X-ray studies.
[(CtzNC)2Ag2](PF6)2 6·2PF6. A mixture of 5·2PF6 (0.12 g,
0.15 mmol), Ag2O (0.048 g, 21 mmol) and 3 Å molecular sieves
in acetonitrile (15 mL) was stirred under nitrogen, in darkness,
for 3 days. The mixture was then filtered through a plug of
Celite and the filtrate was evaporated to dryness. The residue
was triturated in ethyl acetate (15 mL) and the solid was col-
Experimental
General methods
Nuclear magnetic resonance spectra were recorded using lected and dried to afford a white powder (0.072 g, 64%). δH
1
Varian Gemini 200 (200 MHz for H, 50 MHz for 13C), Bruker (400.1 MHz; CD3CN) 0.34 (12 H, t, J 8 Hz, 4 × CH2CH2CH3),
AVN 400 (400.1 MHz for 1H, 100.6 MHz for 13C), Bruker Avance 0.89 (8 H, m,
4 × CH2CH2CH3), 1.12 (4 H, m, 4 ×
500 (500.1 MHz for H, 125.8 MHz for 13C) or Bruker Avance NCH2CHHCH2), 1.36 (4 H, m, 4 × NCH2CHHCH2), 1.46 (12 H,
600 (600.1 MHz for 1H, 150.9 MHz for 13C) spectrometers at t, J 8 Hz, 4 × NCH2CH3), 3.98 (8 H, m, 4 × NCH2CH3), 4.20
ambient temperature. 1H and 13C chemical shifts were refer- (8 H, m, 4 × NCH2CH2CH2), 7.54–7.65 (12 H, m, 12 × ArH),
enced to residual solvent resonances. Microanalyses were per- 8.66 (4 H, d, J 8 Hz, 4 × ArH); δC (100 MHz, CD3CN) 13.2
formed by the Central Science Laboratory at the University of (CH2CH2CH3), 14.5 (NCH2CH3), 20.1 (CH2CH2CH3), 32.6
Tasmania or by Mr Robert Herman at the Department of (NCH2CH2CH2), 44.4 (NCH2CH3), 51.1 (NCH2CH2CH2), 113.5
Chemistry, Curtin University. UV-vis data was collected on a (benzimidazolyl CH), 117.7 (benzimidazolyl CH), 126.7 (benz-
PerkinElmer LAMBDA 35 UV/VIS spectrometer. 1-n-Butylbenz- imidazolyl CH), 127.3 (benzimidazolyl CH), 132.9 (benz-
imidazole,67 1,3-dichloro-5-diethylaminotriazine,68 bromido- imidazolyl C), 135.4 (benzimidazolyl C), 164.4 (triazinyl C)
1
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
165.5 (triazinyl C) and 193.6 (two doublets, 1J
187 Hz, was added 6·2PF6 (0.23 g, 0.29 mmol) and the resulting
107Ag,C
1J
215 Hz, C–Ag); δH (600.1 MHz; d6-DMSO) 0.26 (12 H, t, mixture was stirred at room temperature for 3 d in darkness,
109Ag,C
J 7 Hz, 4 × CH2CH2CH3), 0.83 (8 H, m, 4 × CH2CH2CH3), 0.99 under nitrogen with 3 Å molecular sieves. The solution was fil-
(4 H, m, 4 × NCH2CHH), 1.27 (4 H, m, 4 × NCH2CHH), 1.48 tered through Celite, which was rinsed with acetonitrile
(12 H, t, J 7 Hz, 4 × NCH2CH3), 3.98 (4 H, m, 4 × NCHHCH3), (15 mL). The yellow filtrate was concentrated in vacuo to afford
4.08 (4 H, m, 4 × NCHHCH3), 4.33 (4 H, m, 4 × NCHHCH2), a yellow powder (0.43 g, 95%). Crystals suitable for X-ray
4.40 (4 H, m, 4 × NCHHCH2), 7.62 (4 H, apparent t, splitting diffraction studies were grown by diffusion of vapours between
8 Hz, 4 × Ar CH), 7.73 (4 H, apparent t, splitting 8 Hz, 4 × neat diethyl ether and an acetonitrile solution of the salt
Ar CH), 7.89 (4 H, d, J 8, 4 × Ar CH) and 8.67 (4 H, d, J 8, 4 × (0.16 g, 36%). δH (400 MHz, d6-DMSO) 0.87 (6 H, t, J 7.4 Hz, 2 ×
Ar CH); δC (150.9 MHz; d6-DMSO) 12.63 (CH2CH2CH3), 12.64 CH2CH3), 1.32–1.45 (10 H, m, 2 × CH2CH2CH3, 2 × NCH2CH3),
(NCH2CH3), 18.8 (CH2CH2CH3), 31.5 (NCH2CH2), 43.1 1.75 (4 H, m, 2 × NCH2CH2CH2), 3.96 (4 H, q, J 7.2 Hz, 2 ×
(NCH2CH3), 49.7 (NCH2CH2), 113.0 (benzimidazolyl CH), 116.2 NCH2CH3), 4.56 (4 H, t, J 7.6 Hz, 2 × NCH2CH2CH2), 7.65 (2 H,
(benzimidazolyl CH), 125.9 (benzimidazolyl CH), 126.2 (benz- apparent t, 7.2 Hz splitting, 2 × ArH), 7.74 (2 H, apparent t,
3
imidazolyl CH), 131.4 (d, JAg,C 4 Hz, benzimidazolyl C), 134.2 7.6 Hz splitting, 2 × ArH), 7.88 (2 H, d, J 8.4 Hz, 2 × ArH) and
3
(d, JAg,C 6, benzimidazolyl C), 163.0 (triazinyl C1 and C3), 8.30 (2 H, d, J 8.0 Hz, 2 × ArH); δC (100 MHz, d6-DMSO) 12.6
163.9 (triazinyl C5) and 192.4 (two doublets, 1J
188, (CH3), 13.6 (CH3), 19.2 (CH2), 31.6 (CH2), 44.5 (CH2), 47.0
107Ag,C
1J
217, C–Ag); Analytically pure colourless crystals were (CH2), 113.7 (benzimidazolyl CH), 113.8 (benzimidazolyl CH),
109Ag,C
grown by the diffusion of vapours between neat ethyl acetate 127.1 (benzimidazolyl CH), 127.7 (benzimidazolyl CH), 128.8
and a solution of the salt in acetonitrile. (Found: C, 46.60; (benzimidazolyl C), 133.4 (benzimidazolyl C), 160.0 (triazinyl
H, 4.68; N, 14.63. C58H72N16Ag2P2F6 requires C, 46.47; H, 4.84; C1 and C3), 162.8 (triazinyl C5) and 174.0 (C–Pd). (Found:
N, 14.97%). Colourless crystals of crystallographic quality were C, 44.39; H, 4.57; N, 14.15. C29H36N8PdClPF6 requires C, 44.46;
grown by the diffusion of vapours between neat diethyl ether H, 4.63; N, 14.30%).
and a solution of the salt in acetonitrile.
Synthesis of bis{2,6-di(3-n-butylbenzimidazolin-2-ylidene)-
Synthesis of bis{2,6-di(3-n-butylbenzimidazolin-2-ylidene)- pyridine}digold(I) bis(tetraphenyl-borate) [(CpyNC)2Au2](BPh4)2
4-diethylaminotriazine}digold(I) bis(hexafluorophosphate) 11·2BPh4. A solution of bromido(tetrahydrothiophene)gold(I)
[(CtzNC)2Au2](PF6)2 7·2PF6. A solution of 6·2PF6 (0.12 g, (0.035 g, 0.1 mmol) in acetonitrile (1 mL) was added to a solu-
0.08 mmol) and bromido(tetrahydrothiophene)gold(I) (0.04 g, tion of 10·2BPh4 (0.09 g, 0.05 mmol) in acetonitrile (20 mL).
0.11 mmol) in acetonitrile (15 mL) was stirred at room temp- The mixture was stirred for 2 h and then filtered through
erature for 3 d in darkness, under nitrogen with 3 Å molecular Celite. The filtrate was concentrated in vacuo. The residue was
sieves. The solution was filtered through Celite, which was dissolved in CHCl3 and then diluted with hexanes to afford a
then rinsed with acetonitrile (15 mL). The yellow filtrate was colourless powder (0.070 g, 65%). δH (500 MHz, d6-DMSO)
concentrated in vacuo affording a yellow powder (0.10 g, 74%). δ 0.82 (12 H, t, J 7.4 Hz, 4 × CH2CH3), 1.30 (8 H, m, 4 ×
δH (400 MHz, CD3CN) δ 0.34 (12 H, t, J 7.2 Hz, 4 × CH2CH3), 1.80 (8 H, m, 4 × NCH2CH2), 4.64 (8 H, m, 4 ×
CH2CH2CH3), 0.89 (8 H, m, 4 × CH2CH2CH3), 1.17 (4 H, m 4 × NCH2CH2), 6.81 (8 H, m, para BPh4), 6.95 (16 H, m, meta
NCH2CHHCH2), 1.38 (8 H, m, 4 × NCH2CHHCH2), 1.46 (12 H, BPh4), 7.21 (16 H, m, ortho BPh4), 7.59 (apparent t, splitting
t, J 7.0 Hz, 4 × NCH2CH3), 3.99 (m, 8H, 4 × NCH2CH3), 4.28 7.5 Hz, 4H, 4 × benzimidazolyl H), 7.65 (apparent t, splitting
(4 H, m, 4 × NCHHCH2CH2), 4.41 (4 H, m, 4 × NCHHCH2CH2), 7.3 Hz, 4H, 4 × benzimidazolyl H), 7.84 (d, J 8.2 Hz, 4H, 4 ×
7.56–7.62 (12 H, m, 12 × ArH), 8.66 (4 H, d, J 8.4 Hz, 4 × ArH); benzimidazolyl H), 8.02 (d, J 8.2 Hz, 4H, 4 × benzimidazolyl
δC (100 MHz, CD3CN) δ 12.7 (CH2CH2CH3), 12.9 (NCH2CH3), H), 8.19 (d, J 7.9 Hz, 4H, 2 pyridyl H3), 8.44 (t, J 7.8 Hz, 2H, 2 ×
19.7 (CH2CH2CH3), 32.0 (NCH2CH2CH2), 44.0 (NCH2CH3), 50.4 pyridyl H4); δC (125 MHz, d6-DMSO) δ 13.4 (CH2CH3), 19.3
(NCH2CH2CH2), 113.1 (benzimidazolyl CH), 117.0 (benzimida- (CH2CH3), 31.7 (NCH2CH2), 48.5 (NCH2CH2), 112.5 (CH, benz-
zolyl CH), 126.9 (benzimidazolyl CH), 127.1 (benzimidazolyl imidazolyl), 112.9 (CH, benzimidazolyl), 121.2 (CH, pyridyl C3),
CH), 132.0 (benzimidazolyl C), 134.5 (benzimidazolyl C), 163.4 121.5 (CHpara, BPh4), 125.3 (q, JC–B 2.7 Hz, CHmeta, BPh4), 125.4
(triazinyl C) 165.2 (triazinyl C) and 191.7 (C–Au); (Found: C, (CH, benzimidazolyl), 125.6 (CH, benzimidazolyl), 131.2
41.93; H, 4.23; N, 13.31. C58H72N16Au2P2F6·0.5CH3CN requires (C, benzimidazolyl), 133.4 (C, benzimidazolyl), 135.5 (CHortho
,
C, 41.74; H, 4.36; N, 13.61%). Colourless crystals of crystallo- BPh4), 143.8 (CH, pyridyl C4), 148.5 (CH, pyridyl C2), 163.3
graphic quality were grown by the diffusion of vapours (q, JC–B 49 Hz, Cipso, BPh4), 188.6 (C–Au); (Found: C, 64.96;
between neat diethyl ether and a solution of the salt in aceto- H, 5.07; N, 7.42. C102H98N10Au2B2 requires C, 65.18; H, 5.26;
nitrile. During the crystallisation process a very small quantity N, 7.45%). Colourless crystals of crystallographic quality
of yellow crystals also deposited from solution. These crystals were grown by the diffusion of vapours between a solution of
were identified as 8·2PF6 by single crystal X-ray studies.
Synthesis of chlorido{2,6-di(3-n-butylbenzimidazolin-2-ylidene)-
4-diethylaminotriazine}-palladium(II) hexafluorophosphate
hexanes and a solution of the salt in acetone.
X-ray structure determinations
[(CtzNC)PdCl]PF6 9·PF6. A mixture of palladium(II) chloride Crystallographic data for the structures were collected at 100(2)
(0.14 g, 0.79 mmol) in acetonitrile (10 mL) was heated at reflux K (180(2) K for 7·2PF6) on an Oxford Diffraction Gemini or an
for 2 h. The solution was filtered and cooled. To the filtrate Oxford Diffraction Xcalibur diffractometer fitted with Mo Kα
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
radiation. Following face-indexed absorption corrections
and solution by direct methods, the structures were refined
against F2 with full-matrix least-squares using the program
SHELXL-2014.70 Anisotropic displacement parameters were
employed for the non-hydrogen atoms. All hydrogen atoms
were added at calculated positions and refined by use of a
riding model with isotropic displacement parameters based
on those of the parent atom.
2 P. de Frémont, N. Marion and S. P. Nolan, Coord. Chem.
Rev., 2009, 253, 862–892.
3 E. A. B. Kantchev, C. J. O’Brien and M. G. Organ, Angew.
Chem., Int. Ed., 2007, 46, 2768–2813.
4 S. Díez-González, N. Marion and S. P. Nolan, Chem. Rev.,
2009, 109, 3612–3676.
5 V. Dragutan, I. Dragutan, L. Delaude and A. Demonceau,
Coord. Chem. Rev., 2007, 251, 765–794.
6 E. Colacino, J. Martinez and F. Lamaty, Coord. Chem. Rev.,
2007, 251, 726–764.
7 N. Marion and S. P. Nolan, Chem. Soc. Rev., 2008, 37, 1776–
1782.
8 C. Samajłowicz, M. Bieniek and K. Grela, Chem. Rev., 2009,
109, 3708–3742.
9 I. J. B. Lin and C. S. Vasam, Coord. Chem. Rev., 2007, 251,
642–670.
Density functional theory calculations
Density functional theory (DFT) calculations were performed
using the SIESTA code.71 The electronic wave functions were
expanded in a basis set of numerical atomic orbitals of
double-zeta plus polarization (DZP) quality. The effective
potentials due to the nucleus and core electrons were
described using norm-conserving Troullier and Martins72
pseudopotentials. The electron density was represented using
an auxiliary basis consisting of a real-space mesh with a
kinetic-energy cutoff of 300 Ry. Exchange–correlation
was treated using the non-local van der Waals method
vdW-DF2(rPW86).73 Atomic coordinates and unit cells
(where applicable) were fully optimized in all calculations to
a force tolerance of 0.01 eV Å−1. For the isolated complexes
calculations were performed in a cubic box with sides of
30 Å to avoid interactions of molecules with their periodic
images.
10 J. C. Garrison and W. J. Youngs, Chem. Rev., 2005, 105,
3978–4008.
11 K. M. Hindi, M. J. Panzner, C. A. Tessier, C. L. Cannon and
W. J. Youngs, Chem. Rev., 2009, 109, 3859–3884.
12 M.-L. Teyssot, A.-S. Jarrousse, M. Manin, A. Chevry,
S. Roche, F. Norre, C. Beaudoin, L. Morel, D. Boyer,
R. Mahiou and A. Gautier, Dalton Trans., 2009, 6894–6902.
13 L. Oehninger, R. Rubbiani and I. Ott, Dalton Trans., 2013,
42, 3269–3284.
14 P. V. Simpson, C. Schmidt, I. Ott, H. Bruhn and
U. Schatzschneider, Eur. J. Inorg. Chem., 2013, 2013, 5547–5554.
15 T. Bernardi, S. Badel, P. Mayer, J. Groelly, P. de Frémont,
B. Jacques, P. Braunstein, M.-L. Teyssot, C. Gaulier,
F. Cisnetti, A. Gautier and S. Roland, ChemMedChem, 2014,
9, 1140–1144.
16 F. E. Hahn, M. C. Jahnke, V. Gomez-Benitez, D. Morales-
Morales and T. Pape, Organometallics, 2005, 24, 6458–6463.
17 D. Pugh and A. A. Danopoulos, Coord. Chem. Rev., 2007,
251, 610–641.
18 M. Poyatos, J. A. Mata and E. Peris, Chem. Rev., 2009, 109,
3677–3707.
We have previously shown that this approach can accurately
model geometries and energies of molecular crystals and
molecules.42,60 For molecular crystals, the interaction energy
(Einteraction) was calculated using:
Ecrystalþmetal
Einteraction
¼
ꢀ Emetal ꢀ Emolecule
;
n
where Ecrystal+metal is the energy of the optimised molecular
crystal containing the metal atom, Emetal is the energy of an iso-
lated metal atom, Emolecule is the energy of an isolated molecule
from the molecular crystal, and n is the number of molecules in
the unit cell. As the counter-ions were not required for calcu-
lations of the isolated complexes, the cubic cell was charged
accordingly. Basis set superposition errors (BSSE) were
accounted for using the Counterpoise (CP) correction method.74
19 E. Peris and R. H. Crabtree, Coord. Chem. Rev., 2004, 248,
2239–2246.
20 D. Pugh, A. Boyle and A. A. Danopoulos, Dalton Trans.,
2008, 1087–1094.
21 K. Inamoto, J.-I. Kuroda, E. Kwon, K. Hiroya and T. Doi,
J. Organomet. Chem., 2009, 694, 389–396.
22 F. Jean-Baptiste dit Dominique, H. Gornitzka, A. Sournia-
Saquet and C. Hemmert, Dalton Trans., 2009, 340–352.
23 T. Tu, J. Malineni and K. H. Dötz, Adv. Synth. Catal., 2008,
350, 1791–1795.
24 T. Tu, X. Bao, W. Assenmacher, H. Peterlik, J. Daniels and
K. H. Dötz, Chem. – Eur. J., 2009, 15, 1853–1861.
25 D. H. Brown, G. L. Nealon, P. V. Simpson, B. W. Skelton
and Z. Wang, Organometallics, 2009, 28, 1965–1968.
26 D. H. Brown and B. W. Skelton, Dalton Trans., 2011, 40,
8849–8858.
27 K. D. M. MaGee, G. Travers, B. W. Skelton, M. Massi,
A. D. Payne and D. H. Brown, Aust. J. Chem., 2012, 65,
823–833.
Acknowledgements
We thank Curtin University for a Research and Teaching Fel-
lowship (to D. H. B.). Dr Max Massi is thanked for useful dis-
cussions. This work was supported by computational
resources provided by the Australian Government through the
Pawsey Centre under the National Computational Merit Allo-
cation and the Pawsey Partner schemes.
References
1 F. E. Hahn and M. C. Jahnke, Angew. Chem., Int. Ed., 2008,
47, 3122–3172.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
28 A. Poethig and T. Strassner, Organometallics, 2011, 30, 52 A. M. Reilly and A. Tkatchenko, J. Chem. Phys., 2013, 139,
6674–6684. 024705–024713.
29 A. Poethig and T. Strassner, Organometallics, 2012, 31, 53 J. van de Streek and M. A. Neumann, Acta Crystallogr., Sect.
3431–3434. B: Struct. Sci., 2010, 66, 544–558.
30 J. C. C. Chen and I. J. B. Lin, J. Chem. Soc., Dalton Trans., 54 J. Antony, R. Sure and S. Grimme, Chem. Commun., 2015,
2000, 839–840. 51, 1764–1774.
31 A. Caballero, E. Díez-Barra, F. A. Jalón, S. Merino, 55 T. Bereau and O. A. von Lilienfeld, J. Chem. Phys., 2014,
A. M. Rodríguez and J. Tejeda, J. Organomet. Chem., 2001,
627, 263–264.
32 W. Zuo and P. Braunstein, Organometallics, 2012, 31, 2606–
2615.
33 Y.-F. Han, G.-X. Jin, C. G. Daniliuc and F. E. Hahn, Angew.
Chem., Int. Ed., 2015, 54, 4958–4962.
34 Y.-F. Han, G.-X. Jin and F. E. Hahn, J. Am. Chem. Soc., 2013,
135, 9263–9266.
35 A. Rit, T. Pape and F. E. Hahn, Organometallics, 2011, 30,
6393–6401.
36 D. Tapu, D. A. Dixon and C. Roe, Chem. Rev., 2009, 109,
3385–3407.
37 H. M. J. Wang and I. J. B. Lin, Organometallics, 1998, 17,
972–975.
38 A. Caballero, E. Díez-Barra, F. A. Jalón, S. Merino and
J. Tajeda, J. Organomet. Chem., 2001, 617–618, 395–398.
39 K. A. Wheeler and B. M. Foxman, Mol. Cryst. Liq. Cryst.,
1992, 211, 347–360.
141, 034101–034112.
56 L. Kronik and A. Tkatchenko, Acc. Chem. Res., 2014, 47,
3208–3216.
57 A. Otero-de-la-Roza and E. R. Johnson, J. Chem. Theor.
Comput., 2015, 11, 4033–4040.
58 T. Risthaus and S. Grimme, J. Chem. Theor. Comput., 2013,
9, 1580–1591.
59 M. J. Turner, S. Grabowsky, D. Jayatilaka and
M. A. Spackman, J. Phys. Chem. Lett., 2014, 5, 4249–4255.
60 D. J. Carter and A. L. Rohl, J. Chem. Theor. Comput., 2012,
8, 281–289.
61 M. Ogden, A. Rohl and J. Gale, Chem. Commun., 2001,
1626–1627.
62 L. Kong, G. Román-Pérez, J. Soler and D. Langreth, Phys.
Rev. Lett., 2009, 103, 096103.
63 A. M. Walker, B. Civalleri, B. Slater, C. Mellot-Draznieks,
F. Corà, C. M. Zicovich-Wilson, G. Román-Pérez, J. M. Soler
and J. D. Gale, Angew. Chem., Int. Ed., 2010, 49, 7501–
7503.
40 X. Liu, R. Pattacini, P. Deglmann and P. Braunstein, Orga-
nometallics, 2011, 30, 3302–3310.
41 A. Herbst, C. Bronner, P. Dechambenoit and O. S. Wenger,
Organometallics, 2013, 32, 1807–1814.
64 F. Malberg, J. G. Brandenburg, W. Reckien, O. Hollóczki,
S. Grimme and B. Kirchner, Beilstein J. Org. Chem., 2014,
10, 1299–1307.
42 D. J. Carter and A. L. Rohl, J. Chem. Theor. Comput., 2014, 65 H. Schmidbaur and A. Schier, Angew. Chem., Int. Ed., 2015,
10, 3423–3437. 54, 746–784.
43 R. Sabatini, E. Küçükbenli, B. Kolb, T. Thonhauser and 66 S. Sculfort and P. Braunstein, Chem. Soc. Rev., 2011, 40,
S. de Gironcoli, J. Phys.: Condens. Matter, 2012, 24,
424209.
44 A. Otero-de-la-Roza, V. Luaña, E. R. T. Tiekink and
2741–2760.
67 Q.-X. Liu, L.-N. Yin, X.-M. Wu, J.-C. Feng, J.-H. Guo and
H.-B. Song, Polyhedron, 2008, 27, 87–94.
J. Zukerman-Schpector, J. Chem. Theor. Comput., 2014, 10, 68 J. T. Thurston, J. R. Dudley, D. W. Kaiser,
5010–5019.
I. Hechenbleikner, F. C. Schaefer and D. Holm-Hansen,
J. Am. Chem. Soc., 1951, 73, 2981–2983.
69 R. Uson, A. Laguna and M. Laguna, Inorg. Synth., 1989, 26,
85–91.
45 S. Ehrlich, J. Moellmann and S. Grimme, Acc. Chem. Res.,
2013, 46, 916–926.
46 K. Berland, Ø. Borck and P. Hyldgaard, Comput. Phys.
Commun., 2011, 182, 1800–1804.
70 G. M. Sheldrick, Acta. Crystallogr., Sect. C, 2015, 71, 3–8.
47 B. Civalleri, C. M. Zicovich-Wilson, L. Valenzano and 71 J. Soler, E. Artacho, J. Gale, A. García, J. Junquera,
P. Ugliengo, CrystEngComm, 2008, 10, 405–410.
48 M. Del Ben, J. Hutter and J. VandeVondele, J. Chem. Theor.
Comput., 2012, 8, 4177–4188.
P. Ordejón and D. Sánchez-Portal, J. Phys.: Condens. Matter,
2002, 14, 2745.
72 N. Troullier and J. L. Martins, Phys. Rev. B: Condens. Matter,
1991, 43, 8861–8869.
49 S. Feng and T. Li, J. Chem. Theor. Comput., 2006, 2, 149–156.
50 K. D. Nanda and G. J. O. Beran, J. Chem. Phys., 2012, 137, 73 K. Lee, É. Murray, L. Kong, B. Lundqvist and D. Langreth,
174106–174112. Phys. Rev. B: Condens. Matter, 2010, 82, 081101.
51 M. A. Neumann and M.-A. Perrin, J. Phys. Chem. B, 2005, 74 F. B. van Duijneveldt, J. G. C. M. van Duijneveldt-van de
109, 15531–15541.
Rijdt and J. H. van Lenthe, Chem. Rev., 1994, 94, 1873–1885.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015