Trinuclear complexes and coordination polymers of redox-active guanidino-functionalized aromatic (GFA) compounds with a triphenylene core

Article abstract of DOI:10.1021/ic501482u

Herein, we report on the synthesis, redox activity, and coo r d i n a t i o n c h e m i s t r y of 2, 3 , 6 , 7 , 1 0 , 1 1 - h e x a k i s - (tetramethylguanidino)triphenylene. CV measurements indicated that the new compound could be oxidized in three se

Full text of DOI:10.1021/ic501482u

Article
pubs.acs.org/IC
Trinuclear Complexes and Coordination Polymers of Redox-Active
Guanidino-Functionalized Aromatic (GFA) Compounds with a
Triphenylene Core
Anna Lebkucher, Christoph Wagner, Olaf Hubner, Elisabeth Kaifer, and Hans-Jorg Himmel*
̈
̈
̈
Anorganisch-Chemisches Institut, Ruprecht-Karls Universitat Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg, Germany
̈
S
* Supporting Information
ABSTRACT: Herein, we report on the synthesis, redox activity, and
coordination chemistry of 2, 3, 6, 7, 10, 11-hexakis-
(tetramethylguanidino)triphenylene. CV measurements indicated
that the new compound could be oxidized in three separate
reversible two-electron oxidation events. The HOMO and LUMO
energies were estimated from the oxidation wave and the onset of
absorption in the UV/vis spectrum. Our discussion also includes the
related new compound 2,3,6,7,10,11-hexakis(N,N′-
dimethylethyleneguanidino)triphenylene. Then trinuclear CuI and
CuII complexes of the new triphenylene ligands were characterized, and their electronic properties are discussed. In contrast to
previously studied redox-active GFA ligands, oxidation of trinuclear copper(I) iodide complexes with I2 leads to copper instead of
ligand oxidation. In the tetra-coordinated CuII complexes, the coordination mode is intermediate between tetrahedral and square
planar. The optical properties of the complexes were studied, and low-energy electronic transitions were assigned to ligand-to-
metal charge-transfer (LMCT) excitations. We then extended our analysis to trinuclear NiII and CoII complexes. The magnetic
coupling mediated through the triphenylene ligand in the trinuclear CuII and CoII complexes was studied by SQUID
magnetometry, revealing ferromagnetic coupling of the spin centers and different degrees of spin delocalization into the
guanidino groups. Finally, we show that the GFA ligands could be linked to one- or two-dimensional coordination polymers and
porous materials with a layer structure by reaction with silver halides.
Scheme 1. Lewis Structure of 2,3,6,7,10,11-
Hexakis(arylthio)triphenylenes Used for the Construction of
Coordination Polymers
INTRODUCTION
■
Triphenylenes are used for a variety of purposes. They are
applied as components in organic light-emitting diode (OLED)
devices1,2 and transistors,3 and substitution allows control of
their electronic properties.2 A columnar phase of 2,3,6,7,10,11-
hexakis(hexylthio)triphenylene was shown to be suitable for the
fast transport of photogenerated charge carriers.4 Moreover,
ligands with triphenylene backbones have been studied. Very
recently, a tris-carbene ligand with a triphenylene backbone was
synthesized, and its trinuclear metal complexes were applied as
catalysts in coupling reactions.5 2,3,6,7,10,11-Hexakis(arylthio)-
triphenylenes (see Scheme 1) and 2,3,6,7,10,11-hexakis-
(phenylseleno)triphenylene were connected by silver(I) to
three-dimensional coordination networks.6,7 Because of their
rigidity, triphenylenes have also been employed for the
synthesis of porous materials with layer structures.8
Guanidines are versatile ligands and substituents.9 Bicyclic
guanidinates have been used extensively for the stabilization of
dinuclear transition-metal complexes with multiple bonds
between two highly oxidized metal atoms10 and for catalytic
applications.9g,11 Acyclic guanidinates with sterically demanding
organic groups have been used as substituents in dimeric MgI
compounds.12 Neutral hybrid guanidine ligands have been
applied in catalytic reactions such as lactide polymerization.13
Moreover, the first structurally characterized end-on-bonded
copper superoxo complex featured a tripodal guanidine
ligand.14 Guanidino-functionalized aromatic (GFA-n) com-
pounds (where n ≥ 4 denotes the number of guanidino
groups) were established by us as a new class of redox-active
Received: June 20, 2014
Published: September 4, 2014
© 2014 American Chemical Society
9876
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
ligands,15 with 1,2,4,5-tetrakis(tetramethylguanidino)benzene
oxidation of 1(CuI)2 with I2 leads to the chain polymer
{[1(CuI)2](I3)2}n.19 In paramagnetic dinuclear CuII, NiII, or
CoII complexes, weak antiferromagnetic coupling mediated by
the GFA ligand leads to a singlet ground state if the ligand is
neutral or dicationic. On the other hand, a strong ferromagnetic
coupling (between metal and ligand) and a high-spin ground
state occurs if the ligand is oxidized to the radical
monocation.18 GFA compounds are not only interesting for
applications in the area of coordination chemistry, but could
also serve in CH deprotonation21 or photoinduced reductive
coupling reactions.22 Wrapped in a weakly bound dicationic
complex, they can be used for the stabilization of extended
polyanionic networks.23,24 Several methods for varying the
electron-donor capacity have been reported. In particular,
aromatic substitution has a great influence on the energies and
shapes of the frontier orbitals.25 Modifications of the size of the
aromatic system and the guanidino-substiution pattern are
further possibilities for controlling the redox activity. For
example, 1,4,5,8-tetrakis(tetramehylguanidino)naphthalene, 2
(see Scheme 2), has been shown to be a weaker electron
donor than 1.26
(1, see Scheme 2) being the first example.16 In the case of
Scheme 2. Known GFA-4 Compounds 1 and 2 and New
GFA-6 Compounds 3 and 4
Herein, we report on the synthesis, redox behavior and
coordination chemistry of 2,3,6,7,10,11-hexakis-
(tetramethylguanidino)triphenylene, 3. We also synthesized
t he related compound 2, 3, 6, 7, 10, 11-hexaki s-
(dimethylethyleneguanidino)triphenylene, 4 (see Scheme 2).
The guanidino groups perform four roles that are relevant for
potential applications: (i) They shift the highest occupied
molecular orbital (HOMO) of the triphenylene to higher
energies, increasing the electron-donor capacity of the aromatic
compound. (ii) They increase the solubility. (iii) They provide
sites for metal coordination, leading to either trinuclear
complexes or coordination polymers. (iv) They prevent
aggregation through π-stacking. In previous works,
1,4,5,8,9,12-hexaazatriphenylene (HAT) was used as redox-
active ligand and also an n-conducting material.27−31 This
compound is an electron acceptor [the N atoms lower the
dinuclear copper complexes, an almost-complete electron-
transfer series was synthesized. In dinuclear CuI and CuII
complexes, the redox-active ligand 1 could be either neutral
or oxidized to a radical monocation or dication.17,18 In some
cases, oxidation of the ligand is accompanied by polymerization
to give semiconducting coordination polymers.19,20 Hence,
Scheme 3. Synthesis of the New Redox-Active Triphenylene Derivative 3
9877
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
Figure 1. (a) Fraction of the structure of [3H6]Cl6. Vibrational ellipsoids drawn at the 50% probability level. Hydrogen atoms omitted for the sake of
clarity. Selected bond distances (in pm) and angles (in deg): N1C1 1.425(3), N1C19 1.358(3), N2C19 1.334(3), N3C19 1.333(3), N4
C24 1.353(3), N5C24 1.339(3), N6C24 1.331(3), N7C7 1.415(3), N7C29 1.355(3), N8C29 1.334(3), N9C29 1.334(3), N10C8
1.420(3), N10C34 1.348(3), N11C34 1.330(3), N12C34 1.342(3), N13C13 1.422(2), N13C39 1.350(3), N14C39 1.337(3),
N15C39 1.331(3), N16C14 1.423(3), N16C44 1.359(3), N17C44 1.327(3), N18C44 1.339(3), C1C2 1.411(3), C7C8 1.409(3),
C13C14 1.403(3), C1N1C19 126.78(17), C2N4C24 126.12(17), C7N7C29 126.25(17), C8N10C34 121.74(17), C13
N13C39 126.85(17), C14N16C44 124.51(17). (b) Fraction of the structure of [4H3]Cl3. Vibrational ellipsoids drawn at the 50% probability
level. Hydrogen atoms omitted for the sake of clarity. Selected bond distances (in pm) and angles (in deg): N1C1 1.389(2), N1C19 1.311(2),
N2C19 1.370(2), N3C19 1.354(2), N4C2 1.434(2), N4C24 1.343(2), N5C24 1.331(2), N6C24 1.343(2), N7C7 1.393(2), N7
C29 1.292(2), N8C29 1.393(2), N9C29 1.365(2), N10C8 1.429(2), N10C34 1.329(2), N11C34 1.339(2), N12C34 1.340(2),
N13C13 1.441(2), N13C39 1.337(2), N14C39 1.338(2), N15C39 1.341(2), N16C14 1.394(2), N16C44 1.300(2), N17C44
1.367(2), N18C44 1.383(2), C1C2 1.416(2), C7C8 1.420(2), C13C14 1.407(2), C1N1C19 124.31(15), C2N4C24
119.59(15), C7N7C29 124.29(15), C8N10C34 127.32(15), C13N13C39 117.30(14), C14N16C44 120.96(14).
energies of the lowest unoccupied molecular orbital (LUMO)
and HOMO with respect to those of triphenylene]. By contrast,
we will show herein that 3 and 4 are quite strong electron
donors.
in Figure 1. Protonation occurs exclusively at the imino N
atoms and leads to an increase of the imino CN bond
distance (see the structural differences between protonated and
unprotonated guanidino groups in 4·3HCl). The triphenylene
core remains planar. Below, it is shown that coordination can
cause significant distortions of the triphenylene core.
RESULTS AND DISCUSSION
■
The cyclic voltammetry (CV) curve of 3 (see Figure 2a)
shows three reversible redox processes, which we assign to two-
electron redox events (on the basis of the intensities relative to
that of ferrocene as well as comparison with other GFAs). The
curve obtained for 1 is also included for comparison. The
presence of a one-electron wave at higher potential (E1/2 = 0.68
V) in the curve of 1 is in line with the conversion of two
electrons in the wave at lower potential (E1/2 = −0.76 V). A CV
curve of a 1:2 mixture of compound 1 and ferrocene is shown
in Figure S2a of the Supporting Information. The second
reduction wave in the CV curve of 3 has a doublet structure
indicating that reduction from 34+ to 32+ occurs in two
unresolved one-electron steps. This behavior proved to be
independent of the scan rate (see the curves for different scan
rates in Figure S2b of the Supporting Information). The CV
curve of 4 looks similar, but the waves show larger shoulders
(see Figure S3 in the Supporting Information). Table 1
summarizes the potentials measured for 3. The E1/2 value of the
first two-electron redox event is −0.39 V vs Fc+/0. This value
compares with −0.76 V (also vs Fc+/0) determined for 1. The
E1/2 value for the redox couple 4+2/0, as measured by CV in
Synthesis and Characterization of 3 and 4. Starting
with 2,3,6,7,10,11-hexabromotriphenylene, which was prepared
according to the literature method,32 the hexamine derivative
was synthesized in a two-step reaction.33 The hexamine was
then converted to the hexaguanidines 3 and 4 by reaction with
the “activated ureas” or “Vilsmeier” salts 2-chloro-1,1,3,3-
tetramethylformamidinium chloride and 2-chloro-1,3-dimethy-
limidazolinium chloride, respectively (see Scheme 3 for the
synthesis of 3). The new compounds are pale-yellow-colored
solids that must be stored under inert-gas atmosphere to avoid
protonation or slow oxidation by O2. The crystals of the
compounds were unfortunately of poor quality, but never-
theless, the structure of 4 was measured (see Figure S1 in the
Supporting Information) and confirmed the absence of π-
stacking between the triphenylene cores in the solid state as a
result of the steric demand of the guanidino groups. In addition,
it was possible to obtain good-quality single crystals suitable for
X-ray diffraction for the hexa-protonated species 3·6HCl and
for the tris-protonated species 4·3HCl. Both compounds were
prepared by reaction of the bases 3 and 4 with 6 equiv of HCl.
The observation that 4 binds only three protons whereas 3
binds six can be rationalized by the reduced basicity of the
N,N′-dimethylethyleneguanidino group compared with the
tetramethylguanidino group.34 The structures are illustrated
CH2Cl2 solution, is slightly more negative (−0.40 V vs Fc+/0
;
see the CV curve in the Supporting Information). The same
trend was observed upon changing the guanidino groups in
9878
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
0.38 V vs Fc+/0, respectively, for 3 and 0.03 and 0.43 V vs Fc+/0
respectively, for 4 were measured.
,
The CV measurements highlight the difference between the
redox-active guanidino-functionalized triphenylenes 3 and 4
and the previously reported, also redox-active, HAT
(1,4,5,8,9,12-hexaazatriphenylene) ligand.27−31 Whereas 3 and
4 are quite strong organic electron donors, HAT is an electron
acceptor. Figure S4 in the Supporting Information illustrates
the changes in the HOMO energy and the HOMO−LUMO
gap upon substitution of triphenylene with four guanidino
groups calculated at the B3LYP/6-311G** level. Using a simple
relationship,35 the HOMO energy of 3 can be estimated from
the oxidation potential to be ca. −4.8 eV. Hence, as a direct
consequence of the electron-donating guanidino groups, the
HOMO energy is high in comparison with those of typical
triphenylene derivatives used in OLED devices.2 On the other
hand, it is lower than that of 1 (ca. −4.4 eV). The UV/vis
spectrum (see Figure 2b) provides evidence for a very strong
electronic transition centered at 322 nm and weaker electronic
transitions at 356 nm (shoulder) and ca. 215 nm. From the
onset in the absorption spectrum at ca. 420 nm (3 eV), the
LUMO energy can be estimated to be around −1.8 eV.
Compound 3 can be evaporated without decomposition at a
temperature of ca. 450 °C in a vacuum (3 × 10−2 mbar).
Various oxidation agents, such as iodine, bromine,
tetracyanoquinodimethane (TCNQ), and AgPF6, were tested
to chemically oxidize compounds 3 and 4, and distinct color
changes from pale yellow to deep blue, purple, and green,
respectively, indeed signaled reactions leading to the breakup of
the aromatic system. However, product mixtures were formed,
and the oxidized species turned out to be highly unstable, so
that all attempts thus far to isolate a pure species failed.
Therefore, the structures and energies of 3, 32+, 34+, and 36+
were calculated using density functional theory (DFT) methods
(B3LYP functional in combination with the 6-311G** basis
set). The calculated minimum structures are sketched in Figure
2c, and Table 2 provides selected bond distances (as defined in
Scheme 4). It can be seen that the triphenylene core loses its
planarity upon oxidation and becomes increasingly folded with
increasing electron release. The most oxidized species, 36+, is a
radialene derivative. Radialene itself is extremely unstable, and
only a few derivatives have been structurally characterized.36
With increasing charge, the bond distances indeed increasingly
adopt the values expected for the radialene structure. Hence,
the lengths of bonds a, b, d, and g increase, whereas the length
of bond c decreases. Bond e, which is a CN single bond in
neutral 3, decreases in length to a value typical for a CN
double bond. In contrast, bond f (the imino NC double
bond in neutral 3) increases in length to a value typical for a
single bond.
Figure 2. (a) CV curve of 3 (potential vs Fc+/0, CH2Cl2 solution, with
[n-Bu4N][PF6] as the supporting electrolyte, scan speed of 100 mV
s−1). The curve recorded for 1 is included for comparison. (b) UV/vis
spectrum of 3 in CH3CN. (c) Comparison between the structures
calculated for 3, 32+, 34+, and 36+ (B3LYP/6-311G** calculations).
Table 1. Results of the CV Measurements for 3 and 4 in
a
CH2Cl2 Solution
Eox (V)
Ered (V)
E1/2 (V)
Compound 3
−0.25
0.08
first two-e− wave
second two-e− wave
third two-e− wave
−0.52
−0.12
0.32
−0.39
0.02
0.45
0.38
Compound 4
The Gibbs free energy for the gas-phase electron-transfer
reaction between 1 and 32+ (see Scheme 5) was calculated to be
only −1 kJ mol−1, a result that at first glance argues for a similar
electron-donor capacity. On the other hand, the CV data clearly
show 3 to be a weaker electron donor than 1 in CH2Cl2
solution. In 32+, the positive charge is delocalized over a large
number of atoms. The seeming discrepancy can therefore be
explained by the poorer solvent stabilization of the dication 32+
in comparison to 12+. The large solvent influence on redox
reactions with 1 was already discussed by us previously.16
According to our calculations, a hypothetical electron-transfer
reaction in the gas phase between 1, 2-bis-
(tetramethylguanidino)benzene and 32+ (see the second
first two-e− wave
second two-e− wave
third two-e− wave
−0.32
0.02
−0.49
−0.09
0.36
−0.40
0.03
0.51
0.43
a
All potentials given relative to the redox couple Fc+/0 (Ag/AgCl
electrode, [nBu4N][PF6] as the conducting salt).
compound 1.34 The difference between Eox and Ered for the first
two-electron redox event of each 3 and 4 is relatively large,
which might argue for significant structural differences between
the neutral and dicationic species. For the second and third
two-electron redox events of 3 and 4, E1/2 values of 0.02 and
9879
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
a
Table 2. Selected Bond Distances (in Å) as Obtained from DFT Calculations (B3LYP/6-311G**) for Neutral and Oxidized 3
bond
3
32+
34+
36+
a
b
c
d
e
f
1.421−1.428
1.387−1.391
1.409−1.411
1.418
1.429−1.485
1.378−1.429
1.375−1.426
1.411−1.471
1.307−1.381
1.297−1.343
1.415−1.468
1.500−1.502
1.427−1.430
1.381−1.383
1.462−1.468
1.297−1.304
1.352−1.356
1.455−1.457
1.525−1.527
1.463−1.465
1.362−1.364
1.484−1.493
1.275−1.277
1.385−1.388
1.482−1.488
1.394−1.403
1.285−1.286
1.463
g
a
See Scheme 4 for bond notation.
Scheme 4. Definitions of Some Bond Distances Used in Table 2 and Lewis Structure of the Species with the Highest Possible
Charge, 36+
consists of both σ- and π-contributions.37 Consequently, the
Scheme 5. Calculated Gas-Phase Redox Reactions to
imino bond length significantly increases upon coordination. In
3(CuCN)3, 3(CuCl)3, and 3(CuBr)3, average NC bond
distances of 1.326, 1.323, and 1.330 Å, respectively, were found.
The electronic absorption spectra were found to be similar
for all four CuI compounds. Figure 4c shows as example the
spectrum of 3(CuI)3 [that of 3(CuCN)3 is included in Figure
S5 of the Supporting Information]. Four strong absorptions
were detected above 300 nm, centered at 313, 342, 380, and
399 nm. For comparison, in the case of 3, only two absorptions
above 300 nm, with maxima at 322 and 356 nm, were found.
The experimentally obtained spectrum was compared with a
simulation based on time-dependent density functional theory
(TD-DFT) calculations (see Figure 4c). We used the B3LYP
method, which was previously shown to provide reasonable
results for guanidine−copper complexes (see the details of the
quantum chemical calculations at the end of this work for more
information).38 The bands labeled 1, 2, and 3 (see Figure 4c)
all belong to transitions with a dominant degree of charge
transfer from the CuI units to the guanidine ligand 3, and in all
cases, the LUMO and LUMO + 1 (both located on the
guanidine ligand) are involved (see the orbital pictures in
Figure 4d and Figures S6 and S7 and Table S1 in the
Supporting Information for more details). The shoulder at 399
nm in the measured spectrum was assigned to two transitions
calculated to appear at 432 nm (labeled 1), in which the
excitations from the HOMO − 2 (located on the Cu and I
orbitals) into the LUMO and LUMO + 1 predominantly
participate.
Compare 3 in Its “Intrinsic” Electron-Donor Capacity with
Other Guanidino-Functionalized Aromatic Compounds
reaction in Scheme 5) is highly endergonic [ΔG = +278 kJ
mol−1 at standard conditions (1 bar, 298 K)]. Hence,
compound 3 cannot be described simply as three (o-
bisguanidino)benzene units fused together by CC single
bonds.
Coordination Chemistry. Reaction of 3 with CuX (X =
CN, Cl, Br, or I) yielded the trinuclear complexes 3(CuX)3.
The highest yield was obtained for X = CN (almost 90%), and
the lowest for X = I (ca. 60%). The complexes 3(CuCN)3,
3(CuCl)3, and 3(CuBr)3 were structurally characterized by
single-crystal X-ray diffraction. Their structures are illustrated in
Figures 3 and 4. In all cases, coordination was accompanied by
a quite large distortion of the triphenylene core, leading to the
loss of planarity of the central C6 ring (see Figure 3b). All Cu
N bonds are very similar in length and adopt average values of
2.021 Å in 3(CuCN)3, 2.033 Å in 3(CuCl)3, and 2.024 Å in
3(CuBr)3. In previous work, the metal−guanidine bonding in
GFA complexes was analyzed, and it was found that the bond
The CuI complexes appear to be generally quite unstable in
solutions of organohalides. When 3(CuI)3 was dissolved in a
CHCl3/Et2O mixture, green-colored crystals of 3(CuCl2)3
formed after some time. However, this reaction is slow and
therefore (at least under the chosen conditions) not suitable for
the synthesis of 3(CuCl2)3 in larger quantities.
9880
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
CuII complexes of the guanidine dication. Only in exceptional
cases was the isolation of dinuclear CuII complexes of neutral 1
possible, for example, the complex 1{(Cu(NO3)2}2.18 The
second, more interesting approach, is oxidation of a trinuclear
CuI complex of 3 (see Scheme 6). Hence, oxidation of 3(CuI)3
with I2 resulted in copper oxidation to give 3(CuI2)3. This
reaction is remarkable in several ways. First, CuII is unstable in
the presence of iodide in aqueous solution.39 On the other
hand, in nonaqueous solutions, some CuII complexes with
iodide ligands have been reported, such as with pyridine
ligands, an example being [Cu(py)4I]I·2py,40 or with β-
diketiminate ligands.41 However, the complex 3(CuI2)3 is
unique in that each copper is coordinated to two iodides with
similar CuI distances (see discussion of the structure below).
The reactivity of 3(CuI)3 toward I2 is in sharp contrast to that
observed for previously studied CuI complexes of redox-active
GFA ligands. Hence, reaction of 1(CuI)2 with I2 leads
exclusively to ligand instead of copper oxidation, yielding the
semiconducting coordination polymer {[1(CuI)2](I3)2}n.19
According to quantum chemical calculations [B3LYP/def2-
SV(P)], the gas-phase ΔG value (at standard conditions) for
oxidation of 3(CuI)3 with I2 to give 3(CuI2)3 (second equation
in Scheme 6) amounts to only −20 kJ mol−1. In solution,
3(CuI2)3 shows no sign of I2 elimination. The complex
decomposes slowly in CHCl3 at 65 °C to unknown products.
We noticed that, in CH2Cl2 solution, the iodide is slowly
exchanged by chloride. The thermogravimetry (TG) curve (see
Figure S8 in the Supporting Information) showed weight loss
(presumably with elimination of I2) in several steps at
temperatures higher than 150 °C.
We achieved the structural characterization of 3(CuCl2)3 and
3(CuI2)3 by single-crystal X-ray diffraction. The structure of
3(CuCl2)3 is displayed in Figure 5. The CuN distances
(1.972 Å on average) are slightly shorter than those in
3(CuCl)3 (2.033 Å on average). A stronger ligand−metal
interaction also manifests itself in the elongated CN bond
(1.344 Å on average). Whereas the triphenylene core is almost
planar in the case of 3(CuCl2)3, it is highly curled for 3(CuCl)3.
A structural detail of importance for the understanding of the
electronic absorption spectra and also the magnetic super-
exchange (see discussion below) is the dihedral angle between
the ClCuCl and NCuN planes, which measures 44°
(Cu1) and 50° (Cu2) and is therefore halfway between the
value of 0° for square-planar coordination and 90° for
tetrahedral coordination. According to theoretical analysis,
CuII prefers a distorted trigonal coordination in the presence of
π-donor ligands,42 in agreement with recent experimental
results.43 The structure of 3(CuI2)3 is shown in Figure S8 in the
Supporting Information. It is similar to that of 3(CuCl2)3. The
average dihedral angle between the NCuN and ICuI
planes measures 55°, thus being even closer to tetrahedral than
to square-planar coordination. In line with the relatively low
coordination number, the CuI distances are relatively short,
measuring 2.559 Å on average. For comparison, for the
complex [Cu(py)4I]I·2py, a CuI distance of 2.951(1) Å was
reported.40 On the other hand, in the β-diketiminate complex
(Me2NN)CuI [Me2NN = 2,4-bis(2,6-dimethylphenylimido)-
pentyl], featuring three-coordinated CuII centers, a shorter
CuI distance of 2.4295(5) Å was measured.41 In contrast to
the situation in the CuI complexes, the triphenylene core is
planar in the CuII complex 3(CuI2)3 [as in 3(CuCl2)3].
Figure 3. (a) Molecular structure of the trinuclear complex
3(CuCN)3. Vibrational ellipsoids drawn at the 50% probability level.
Hydrogen atoms omitted for the sake of clarity. Selected bond
distances (in Å) and angles (in deg): Cu1N1 1.998(3), Cu1N4
2.034(3), Cu1C49 1.862(4), Cu2N7 2.005(3), Cu2N10
2.038(3), Cu2C50 1.860(3), Cu3N13 2.046(3), Cu3N16
2.007(3), Cu3C51 1.867(3), N1C1 1.405(4), N1C19
1.329(4), N2C19 1.366(4), N3C19 1.353(4), N4C2
1.405(4), N4C24 1.327(4), N5C24 1.358(4), N6C24
1.351(4), N7C29 1.320(4), N8C29 1.377(4), N9C29
1.362(4), N10C34 1.330(4), N11C34 1.361(4), N12C34
1.354(4), N13C13 1.409(4), N13C39 1.327(4), N14C39
1.362(4), N15C39 1.363(4), N16C14 1.406(4), N16C44
1.324(4), N17C44 1.357(4), N18C44 1.372(4), C1C2
1.427(4), C7C8 1.422(4), C13C14 1.415(4), N1Cu1N4
82.18(11), N1Cu1C49 150.69(14), N4Cu1C49
126.34(14), Cu1C49N19 173.7(4), N7Cu2N10
82.74(10), N7Cu2C50 145.78(12), N10Cu2C50
131.43(13), Cu2C50N20 176.8(3), N13Cu3N16
81.94(11), N13Cu3C51 135.91(14), N16Cu3C51
142.15(14), Cu3C51N21 177.1(4). (b) View from a different
perspective showing the curled structure of the triphenyl system.
Trinuclear CuII complexes of ligand 3 could be synthesized in
two different ways. The first is direct reaction between a
copper(II) halide and 3. Hence, ligand 3 was found to react
with 3 equiv of CuCl2 or CuBr2 to give the complexes
3(CuCl2)3 or 3(CuBr2)3, respectively. Such a reaction is
generally not possible in the case of compound 1, as it leads to
oxidation of the guanidine ligand and formation of dinuclear
The three CuII complexes 3(CuX2)3 exhibit different optical
properties, as evidenced by their colors in CH2Cl2 solution (see
9881
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
Figure 4. (a) Molecular structure of the trinuclear complex 3(CuCl)3. Vibrational ellipsoids drawn at the 50% probability level. Hydrogen atoms
omitted for the sake of clarity. Selected bond distances (in Å) and angles (in deg): Cu1Cl1 2.1480(11), Cu1N1 2.038(3), Cu1N4 2.013(3),
Cu2Cl2 2.1682(10), Cu2N7 2.003(3), Cu2N10 2.074(3), Cu3Cl3 2.1478(10), Cu3N13 2.047(3), Cu3N16 2.022(3), N1C1
1.401(4), N1C19 1.324(4), N2C19 1.363(4), N3C19 1.359(4), N4C24 1.325(4), N5C24 1.359(4), N6C24 1.364(5), N7C7
1.406(4), N7C29 1.334(4), N8C29 1.355(4), N9C29 1.356(4), N10C8 1.407(4), N10C34 1.317(5), N11C34 1.380(6), N12C34
1.354(6), N13C13 1.401(4), N13C39 1.325(4), N14C39 1.356(4), N15C39 1.364(4), N16C44 1.314(4), N17C44 1.369(4),
N18C44 1.378(5), C1C2 1.431(4), C7C8 1.419(4), C13C14 1.418(4), N1Cu1N4 82.74(11), N7Cu2N10 82.18(11), N13
Cu3N16 82.84(11). (b) Molecular structure of the trinuclear complex 3(CuBr)3. Hydrogen atoms omitted for the sake of clarity. Selected bond
distances (in Å) and angles (in deg): Cu1Br1 2.2701(10), Cu1N1 2.020(5), Cu1N4 2.005(4), Cu2Br2 2.2945(12), Cu2N7 2.009(4),
Cu2N10 2.041(4), Cu3Br3 2.2728(16), Cu3N13 2.069(4), Cu3N16 1.998(4), N1C1 1.402(6), N1C19 1.336(6), N2C19
1.366(7), N3C19 1.354(7), N4C2 1.409(6), N4C24 1.318(7), N5C24 1.369(7), N6C24 1.354(7), N7C7 1.405(6), N7C29
1.335(7), N8C29 1.354(7), N9C29 1.363(6), N10C8 1.402(6), N10C34 1.331(7), N11C34 1.351(6), N12C34 1.370(7), C1C2
1.427(7), C7C8 1.434(7), C13C14 1.420(7), N1Cu1N4 82.98(17), N1Cu1Br1 124.00(12), N4Cu1Br1 152.57(13), N7
Cu2N10 82.62(16), N7Cu2Br2 140.63(12), N10Cu2Br2 136.69(12), N13Cu3N16 83.74(17), N13Cu3Br3 125.62(12),
N16Cu3Br3 150.08(12). (c) Comparison between the experimental UV/vis spectrum of 3(CuI)3 and a simulation on the basis of time-
dependent density functional theory (TD-DFT) calculations. (d) Visualization of the isodensity surfaces for some orbitals involved in the electronic
transitions.
Figure 6a and Figure S9 in the Supporting Information). In
Figure 6b, the UV/vis spectrum of 3(CuI2)3 is compared to
those of 3 and 3(CuI)3. A broad transition near 497 nm with an
extinction coefficient of 8106 M−1 cm−1 and an even broader
absorption at 911 nm with an extinction coefficient of 4281
M−1 cm−1 are present in the spectrum of 3(CuI2)3, but not in
the spectra of 3 and 3(CuI)3. The extinction coefficient at 911
nm was measured for several concentrations of 3(CuI2)3 in
9882
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
Scheme 6. Synthesis of CuII Complexes of 3 by Reaction
a
with CuCl2 or Oxidation of 3(CuI)3
a
The analogous reaction for 1(CuI)2 led to ligand instead of copper
oxidation and the formation of a chain polymer.
Figure 6. (a) Photographs showing the different colors of the
trinuclear CuII complexes 3(CuCl2)3, 3(CuBr2)3, and 3(CuI2)3. (b)
Comparison of the UV/vis spectra of 3, 3(CuI)3, and 3(CuI2)3 in
CH2Cl2 solution.
3(CuBr2)3 (754 nm, ε = 2132 M−1 cm−1) and again to
3(CuI2)3 (911 nm, ε = 4281 M−1 cm−1). At the same time, the
extinction coefficient of the band increases. TD-DFT
calculations were performed for 3(CuI2)3 to allow a better
understanding of these lower-energy electronic transitions. In
Figure 7a, the absorption spectrum in CH2Cl2 is compared with
the calculated transitions (see Figures S11 and S12 and Tables
S2 and S3 in the Supporting Information for more details). All
calculated transitions above 420 nm involve predominantely the
three unoccupied spin orbitals of lowest energy (365β, 366β,
and 366α) as acceptor orbitals (see Figure 7b). These three
orbitals are located on the CuI units. The excitation from spin
orbital 365α into spin orbital 368α, both of which are located at
ligand 3, participates most in the transition at 391 nm, and
therefore, this transition can be described as π−π* excitation.
The transitions within the range 420−1500 nm exhibit
dominant ligand-to-metal charge-transfer (LMCT) character.
Strong transitions were predicted in the range 500−700 nm,
and these were assigned to the broad band near 500 nm in the
experimental spectrum. This band is absent in the spectrum of
3(CuI)3. Two transitions with relatively high oscillator strength
were calculated to appear at 1384 nm. We tentatively assign
these two LMCT transitions to the broad band centered at 911
nm in the experimental UV/vis spectrum, although the level of
agreement is quite poor. In principal, the band could also arise
from a d−d transition. However, according to the TD-DFT
calculations, the d−d transitions of 3(CuI2)3 are much weaker
and exhibit much lower energies (1975−1980 nm).
Figure 5. (a) Molecular structure of the trinuclear complex 3(CuCl2)3.
Vibrational ellipsoids drawn at the 50% probability level. Hydrogen
atoms omitted for the sake of clarity. Selected bond distances (in Å)
and angles (in deg): Cu1Cl1 2.2465(11), Cu1Cl2 2.2372(11),
Cu1N1 1.983(3), Cu1N4 1.958(3), Cu2Cl3 2.2401(12),
Cu2N7 1.974(3), N1C1 1.406(4), N1C10 1.340(4), N2
C10 1.322(5), N3C10 1.345(5), N4C2 1.399(4), N4C15
1.357(4), N5C15 1.342(4), N6C15 1.332(5), N7C9 1.414(4),
N7C20 1.336(5), N8C20 1.340(6), N9C20 1.350(6), C1
C2 1.420(4), C9C9′ 1.404(7), Cl1Cu1Cl2 98.51(4), N1
Cu1N4 82.72(11), Cl3Cu2Cl3′ 100.09(8), N7Cu2N7′
83.55(17). (b) View from a different perspective showing the
coordination mode of the copper ions.
CH2Cl2, and a plot of the extinction coefficient as a function of
concentration confirms a linear relationship in accordance with
the Lambert−Beer law (see Figure S10 in the Supporting
Information). Interestingly, this band shifts considerably toward
lower energy from 3(CuCl2)3 (688 nm, ε = 1486 M−1 cm−1) to
We also studied the redox properties of the complexes by
cyclic voltammetry (CV). Although the CV curves (see Figure
S13 in the Supporting Information) show several features that
are difficult to explain, the major waves can be assigned to
9883
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
measured for the ligand-centered redox processes for
3(CuCl2)3 and 3(CuI2)3 are similar (see Figure S13b in the
Supporting Information), but the wave assigned to metal
reduction occurs at −0.55 V vs Fc+/0 for 3(CuI2)3 and is thus
shifted to higher potentials compared with that of 3(CuCl2)3.
This result meets the expectation that copper reduction in the
more electron-rich 3(CuI2)3 complex is easier than that in the
less electron-rich 3(CuCl2)3 complex and lends support to the
assignment of metal- and ligand-centered redox events.
Using the Evans method, the molar susceptibility of
3(CuCl2)3 in CD2Cl2 solution was estimated from the nuclear
magnetic resonance (NMR) chemical shifts to be 5.15 × 10−3
cm3 mol−1. The effective magnetic moment, μeff, was estimated
to be 3.53 μB (see Table 3). Electron paramagnetic resonance
(EPR) spectra of the complexes 3(CuX2)3 (X = Cl, Br, or I),
measured at room temperature in CH2Cl2 solution (ca. 1 mg/
mL), are compared in Figure 8a. At room temperature, the
copper fine structure is not visible. It can be seen that the signal
broadens in the order X = Cl, Br, and I, and the g value slightly
decreases. This trend might point to an increasing degree of
spin delocalization. The calculated spin density in the complex
3(CuI2)3 is illustrated in Figure 8b, and it can be seen that
considerable spin density resides on the guanidino groups.
According to a Mulliken population analysis for 3(CuCl2)3 and
3(CuI2)3, as much as 15.5% and 11.5%, respectively, of the
difference in occupation numbers between α and β electrons is
located on ligand 3. Only 56.5% for 3(CuCl2)3 and 42.5% for
3(CuI2)3 of the surplus of α electrons is located at the copper
atoms, following the trend in g values. This result fits with the
coordination geometries at the copper atoms, which are
between square-planar and tetrahedral, and also with the larger
dihedral angles in 3(CuI2)3 compared with 3(CuCl2)3 (see the
discussion of the crystal structures above). Consequently, the
Cl and I atoms in 3(CuCl2)3 and 3(CuI2)3 carry 28% and 46%,
respectively, of the surplus of α electrons. Hence, the iodine
atoms clearly exhibit more radical character than the chlorine
atoms, in line with the general instability of CuI2 complexes
with respect to CuI and I2 (see discussion above). The
Mulliken charges follow the same trend [0.38 and −0.40 e for
the Cu and Cl atoms, respectively, in 3(CuCl2)3 and 0.27 and
−0.38 e for the Cu and I atoms, respectively, in 3(CuI2)3].
Superconducting quantum interference device (SQUID)
magnetometric measurements were carried out to study the
temperature-dependent magnetism. Magnetization curves were
measured at applied fields of 0.5, 1, 2, 3, 4, and 5 T and
corrected for the underlying diamagnetism (see Figure 8c). The
usual χT versus T plots, which were used to obtain the Curie−
Weiss temperature (ΘCW) and the temperature-independent
paramagnetism (TIP), are provided in Figure S14 in the
Supporting Information. The magnetization curves for all fields
Figure 7. (a) Comparison between the absorption spectrum of
3(CuI2)3 in CH2Cl2 and the transitions calculated by TD-DFT
[B3LYP/def2-SV(P)]. (b) Isodensity surfaces of unoccupied (top)
and occupied (bottom) spin orbitals involved in the electronic
transitions.
metal- or ligand-centered redox processes. In the case of
3(CuCl2)3, on sweeping to lower potentials, reduction of the
CuII ions to CuI gives rise to a broad wave around −0.97 V vs
Fc+/0, a position that is within the broad range typical for the
CuII/I redox couple in copper complexes.44 The corresponding
anodic wave is extremely broad and shifted to −0.14 V. The
three redox couples of the ligand occur at E1/2 values of 0.10,
0.26, and 0.41 V. Similarly to the situation in the free ligand,
these redox processes seem to be reversible. However,
coordination leads to a large shift toward higher potentials,
signaling a strong metal−ligand interaction. The potentials
a
Table 3. Summary of the Results of Magnetic Measurements
[3(CuCl2)3]
[3(CuBr2)3]
[3(CuI2)3]
[3(CoCl2)3]
b
b
b
g1=g2=g3
2.093 (2.086 )
2.086 (2.077 )
2.032 (2.072 )
2.550
0.03
J1 = J2 = J3 (cm−1
)
0.24
0.35
1.46
D1 = D2 = D3 (cm−1
)
−
−
−
−
−1.41
3.9
−
−
−3.80
1.8
46.96
−0.28
−0.61
12.3
E/D1 = E/D2 = E/D3
ΘCW (K)
−0.53
6.5
TIP (10−6 cm3 mol−1
)
c
c
μeff (1000 Oe)
3.64 (3.53 )
3.60
4.17
7.36 (7.65 )
a
b
c
For more details, see the Supporting Information. From electron paramagnetic resonance (EPR) spectroscopy. μeff from Evans NMR analysis.
9884
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
3
Figure 8. (a) EPR spectra of the three complexes 3(CuX2)3 (X = Cl, Br, and I). (b) Calculated spin density for the high-spin state (S = /2) of
3(CuI2)3 (isovalue of 0.00004). (c) Magnetization curves of the three complexes as measured by SQUID together with a curve fit (see text for
details).
are almost identical for 3(CuCl2)3, but they differ significantly
for 3(CuI2)3. This trend indicates an increasing degree of
magnetic superexchange of the three spin centers. The curves
were fitted with the help of the julX program (see Experimental
Details). The spin Hamiltonian is generally expressed as the
sum of the Heisenberg−Dirac−Van Vleck Hamiltonian Hex, the
contribution from zero-field splitting HZFS, and the Zeeman-
interaction term HZee
H = Hex + HZFS + HZee
with
9885
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
ns
⎯⇀ ⎯⇀
Hex = −2
J S ·S
∑
i
j
ij
i<j
ns
⎡
⎤
⎦
E
D
1
3
HZFS
and
HZee
=
D⎢S
−
Si(Si + 1) + i (Sx,i − Sy,i2)⎥
2
2
∑
i
z,i
⎣
i
i=1
ns
=
g[β⎯S⇀·⎯B⎯⇀]
∑
i
i=1
where Jij is the magnetic coupling constant of local spin centers
i and j; ns is the number of spins on each local spin center; and
Di and Ei/Di are the local axial and rhombic zero-field splitting
parameters, respectively.
For the copper complexes, the term HZFE is zero, as each
local spin center is an S = 1/2 system. In the fit, we assumed that
́
all three metals have equal isotropic Lande factors g. In
addition, the three isotropic coupling parameters J in the
trinuclear complex were assumed to be similar. These
assumptions are motivated by the structural data, which show
that all of the metals are coordinated in similar ways. For all
complexes, the coupling constant J turned out to be positive
and to follow the order 3(CuI2)3 (J = 1.46 cm−1) > 3(CuBr2)3
(J = 0.35 cm−1) > 3(CuCl2)3 (J = 0.24 cm−1). The g value
follows the opposite trend, namely, 3(CuI2)3 (g = 2.032) <
3(CuBr2)3 (g = 2.086) < 3(CuCl2)3 (g = 2.093 cm−1), in line
with the results of the EPR measurements. In the past, we
extensively studied the spin delocalization into the guanidino
groups of paramagnetic bisguanidine and GFA complexes and
also analyzed the effect of spin polarization on the aromatic
backbone of bisguanidine ligands, leading to alternating low-
and high-field shifts of the NMR signals.45,46 Direct spin
delocalization and spin polarization should both be of
importance in the magnetic coupling of 3(CuX2)3 complexes.47
The spin delocalization from the copper atoms into the
guanidino groups depends on the CuN bond properties
(covalent character and bond strength) and the angle between
the NC π-orbitals and the Cu magnetic orbital.47,48 The
CuN bond distances in 3(CuCl2)3 and 3(CuI2)3 are similar
[on average, 1.972 Å in 3(CuCl2)3 and 1.974 Å in 3(CuI2)3],
but of course the covalency differs. Presumably, the dihedral
angle between the XCuX and NCuN planes is of
higher importance for the magnetic coupling, because large
dihedral angles [on average, 47° in 3(CuCl2)3 and 55° in
3(CuI2)3] must favor spin delocalization (see also the
discussion in ref 47). For comparison, our B3LYP/def2-
SV(P) calculations predict these angles to be 55° in 3(CuCl2)3
and 62° in 3(CuI2)3. The quantum chemical calculations argue
for significant spin delocalization into the guanidino groups
(see the plotted spin density in Figure 8b). From there, the
magnetic coupling could be mediated by either direct spin
delocalization or spin polarization. Spin polarization leads to an
alternating sign of the spin on the ring carbon atoms. Indeed,
some alternation of the spin on the central C6 ring of the
triphenylene unit could be observed (see Figure 8b).
Figure 9. (a) Molecular structure of the trinuclear complex 3(NiCl2)3.
Vibrational ellipsoids drawn at the 50% probability level. Hydrogen
atoms omitted for the sake of clarity. Selected bond distances (in Å)
and angles (in deg): Ni1Cl1 2.2483(15), Ni1Cl2 2.2401(17),
Ni2Cl3 2.2556(16), Ni2Cl4 2.2380(15), Ni3Cl5 2.2551(17),
Ni3Cl6 2.2341(16), Ni1N1 2.004(4), Ni1N4 1.986(4), Ni2
N7 1.993(4), Ni2N10 2.023(4), Ni3N13 1.996(4), Ni3N16
2.005(4), N1C1 1.410(6), N1C19 1.339(6), N2C19 1.343(7),
N3C19 1.346(6), N4C2 1.405(6), N4C24 1.353(6), N5
C24 1.341(6), N6C24 1.345(7), N7C7 1.404(6), N7C29
1.325(6), N8C29 1.365(7), N9C29 1.340(6), N10C8
1.401(5), N10C34 1.330(6), N11C34 1.369(6), N12C34
1.344(6), N13C13 1.419(6), N13C39 1.343(6), N14C39
1.349(7), N15C39 1.345(6), N16C14 1.413(6), N16C44
1.333(6), N17C44 1.329(7), N18C44 1.359(7), Cl1Ni1Cl2
106.54(6), Cl3Ni2Cl4 113.59(6), Cl5Ni3Cl6 113.56(7),
N1Ni1N4 83.12(15), N7Ni2N10 82.63(15), N13Ni3
N16 83.15(16). (b) View from a different perspective showing the
coordination mode. (c) UV/vis spectrum of 3(NiCl2)3 in CH3CN
solution.
We extended the analysis of trinuclear paramagnetic
compounds of 3 by synthesizing and characterizing the
complexes 3(NiCl2)3, 3(NiBr2)3, and 3(CoCl2)3, which exhibit
six or nine unpaired electrons. Figure 9a,b displays the
molecular structure of 3(NiCl2)3. All Ni atoms are symmetri-
cally coordinated by two imino N atoms, and the average Ni
N distance measures 2.001 Å. Coordination leads to elongation
of the NC bond distances within the guanidino groups to an
average value of 1.337 Å. The tetrahedral angles between the
9886
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
ClNiCl and NNiN planes through the two Ni atoms
are different. The tetrahedral angle at Ni1 measures 65.0°,
whereas the tetrahedral angles at Ni2 and Ni3 are close to 90°
(89.7° and 89.4°, respectively). The structure of 3(CoCl2)3 was
also measured (see Figure S15 in the Supporting Information).
Again, two dihedral planes (ClCoCl, NCoN) are
close to the expected value of 90° (88.9° and 89.5° at Co1 and
Co3, respectively), whereas the third is significantly smaller
(67.3°). The UV/vis spectrum provided evidence for d−d
transitions at 684, 580, and 512 nm for 3(NiCl2)3 (see Figure
9c). As expected, the bands shift to slightly larger values of 708,
ca. 607, and ca. 539 nm for 3(NiBr2)3 (see Experimental
Details). For 3(CoCl2)3, d−d transitions were detected at 664,
622, 568, and 550 nm (see Figure S16 in the Supporting
Information). A first analysis by NMR spectrocsopy using the
Evans method gave molar susceptibilities of 7.7958 × 10−3 and
2.4163 × 10−2 cm3 mol−1 for 3(NiCl2)3 and 3(CoCl2)3,
respectively. The effective magnetic moments, μeff, were
estimated to be 4.35 and 7.65 μB, respectively, for the two
complexes. Using SQUID, a μeff value of 7.36 μB was obtained
for 3(CoCl2)3, in pleasing agreement with the value estimated
in solution by the Evans method (see Table 3). Magnetization
curves were recorded for 3(CoCl2)3 and fitted using the
following parameters: g = 2.55, J = 0.029 cm−1, D = 47.0 cm−1
(all axial zero-field splitting parameters D were assumed to be
equal), E/D = −0.28, together with a temperature-independent
paramagnetism (TIP) of 12.3 × 10−6 cm3 mol−1 (see Figure
S16b in the Supporting Information). Hence, the magnetic
coupling between the three spin centers is very weak, whereas
the zero-field splitting parameter D is relatively large (see a
discussion for other mono- and dinuclear guanidine complexes
in ref 49).
Coordination as a Means to Form Arrays of 3. The new
organic electron donors could be linked by coordinative bonds
to highly ordered arrays. We started our studies in this area with
the reaction beween 3 and AgCN and obtained a molecular
coordination compound, namely, the trinuclear complex
3(AgCN)3. Its crystal structure is shown in Figure 10. All
three silver atoms are 3-fold-coordinated. By contrast, reaction
with silver halides AgX (X = Cl, Br, or I) led to coordination
polymers [3(AgX)3]n. All three compounds crystallized, and the
structures of [3(AgCl)3]n and [3(AgI)3]n are illustrated in
Scheme 7 and Figures 11 and 12 {see Figure S17 In the
Supporting Information for the structure of [3(AgBr)3]n}.
Some degree of disorder of the silver and halogen atoms occurs,
especially in the case of [3(AgI)3]n (see Figure 12). An
interesting detail of all structures is the massive displacement of
the tetrahedrally coordinated Ag atoms from the aromatic plane
of the ligand. In [3(AgCl)3]n, for example, the two tetrahedrally
coordinated Ag atoms are displaced by as much as 1.489 Å
(Ag1) and 1.375 Å (Ag2) from the best plane of the
triphenylene core (see Figure 11). By contrast, the trigonally
coordinated third Ag atom is displaced by not more than 0.009
Å from this plane. As in 3(AgCN)3, the triphenylene core
exhibits a curled structure. For X = Cl or Br, only two of the
three Ag atoms of each complex unit establishes connections to
the next complex units through halogen bridges. Hence, the
structure features both tetrahedrally and trigonally coordinated
Ag atoms. In the case of [3(AgI)3]n, all three Ag atoms are 4-
fold-coordinated and connected through halogen bridges to the
next complex units (see Figure 12a). Such a connectivity
pattern leads to layers with a honeycomb structure, consisting
of hexagons built of six complex units. The arrangement of the
Figure 10. (a) Molecular structure of the trinuclear complex
3(AgCN)3. Vibrational ellipsoids drawn at the 50% probability level.
Hydrogen atoms omitted for the sake of clarity. Selected bond
distances (in Å) and angles (in deg): Ag1N1 2.353(5), Ag1N4
2.202(5), Ag1C49 2.062(8), Ag2N7 2.184(5), Ag2N10
2.382(5), Ag2C50 2.037(7), Ag3N13 2.311(5), Ag3N16
2.261(5), Ag3C51 2.063(7), N1C1 1.400(8), N1C19
1.317(9), N2C19 1.359(9), N3C19 1.374(8), N4C2
1.418(7), N4C24 1.339(8), N5C24 1.373(8), N6C24
1.351(9), N7C7 1.411(8), N7C29 1.328(8), N8C29
1.367(8), N9C29 1.366(8), N10C8 1.398(8), N10C34
1.306(8), N11C34 1.350(9), N12C34 1.367(9), N13C13
1.419(8), N13C39 1.314(8), N14C39 1.367(8), N15C39
1.356(8), N16C14 1.419(8), N16C44 1.311(8), N17C44
1.374(9), N18C44 1.358(9), N19C49 1.132(10), N20C50
1.137(9), N21C51 1.121(9), C1C2 1.414(8), C7C8 1.414(8),
C13C14 1.419(9), N1Ag1N4 74.87(19), N1Ag1C49
121.4(3), N4Ag1C49 163.0(3), Ag1C49N19 170.8(8),
N7Ag2N10 73.94(18), N7Ag2C50 163.5(2), N10
Ag2C50 120.9(2), Ag2C50N20 171.5(7), N13Ag3N16
73.86(18), N13Ag3C51 138.8(2), N16Ag3C51 147.1(2),
Ag3C51N21 172.0(8). (b) View from a different perspective
showing the curled structure of the triphenyl system.
layers is shown in Figure 12b and results in a porous structure
that has some similarities to (but also differences from) the
experimentally verified or predicted structures obtained with
other triphenylene derivatives.8 Using the program Crystal-
Explorer, the pore volume per unit cell was estimated to be
1644.5 Å3, and the pore volume per gram was estimated to be
0.15365 cm3 g−1 (see Figure S18 in the Supporting Information
for the visualization of the voids).50 This is a quite high value
9887
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
a
Scheme 7. Lewis Structures of the Complex Units and Their Connectivity in the Chain Polymer [3(AgCl)3]n and in One of the
Sheets of the Porous Polymer [3(AgI)3]n
a
Similar structure for [3(AgBr)3]n.
(see the discussion in ref 50b). Only a small amount of
crystalline material was obtained (according to powder XRD
measurements, the pure material obtained directly upon
removal of the solvent was amorphous), so that we have thus
far not been able to estimate the pore volume from gas
absorption experiments. In ongoing experiments, we are
collecting more crystalline material to study the possibility of
intercalation into the layer structure of [3(AgI)3]n. Because of
the redox activity of the organic linkers in this coordination
polymer, oxidizing guest molecules such as I2 are at the center
of interest in this effort.
some characteristic bands in the electronic absorption spectra
were assigned to metal-to-ligand charge-transfer (MLCT)
transitions (for the CuI complexes) or LMCT transitions (for
the CuII complexes). The magnetic coupling in the CuII
complexes was studied by SQUID magnetometric measure-
ments, and the data indicated weak ferromagnetic coupling
between the three spin centers that increases within the series
3(CuCl2)3 < 3(CuBr2)3 < 3(CuI2)3. The unusual coordination
of the copper atoms [the dihedral angles between the XCu
X (X = halogen) and NCuN planes measure, on average,
47° for 3(CuCl2)2 and 55° for 3(CuI2)3] enables considerable
spin delocalization into the guanidino groups. We extended our
analysis to a trinuclear Co complex of 3 that exhibits a high D
value (zero-field splitting) and a very weak (ferromagnetic)
coupling between the spin centers, as well as to Ni complexes.
Finally, we showed that redox-active 3 can be integrated into
coordination polymers by reaction with silver halides. With
AgCl or AgBr, chain polymers [3(AgCl)3]n or [3(AgBr)3]n,
respectively, are formed, in which two of the three silver atoms
of each complex unit are linked by halogen bridges to the next
complex unit, whereas the third silver atom is trigonal-planar-
coordinated without establishing a link to the next complex
unit. In the case of AgI, the porous material [3(AgI)3]n results
(with an estimated pore volume per unit cell and per gram of
1644.5 Å3 and 0.15365 cm3 g−1, respectively), which consists of
sheets in which all three silver atoms of each complex unit are
linked by halogen bridges to the adjacent complex units. In this
way, a honeycomb structure results, composed of hexagons
formed from six complex units. In all of these structures, the
silver atoms are heavily displaced from the “best plane” of the
triphenylene cores. In ongoing work, we are testing the
possibility of intercalating redox-active guest molecules into the
layer structure of [3(AgI)3]n.
CONCLUSIONS
■
The new redox-active ligands 2,3,6,7,10,11-hexakis-
(tetramethylguanidino)triphenylene (3) and 2,3,6,7,10,11-
hexakis(dimethylethyleneguanidino)triphenylene (4) were pre-
pared in four steps starting from triphenylene. Their redox
activity was analyzed and compared to that of 1,2,4,5-
tetrakis(tetramethylguanidinyl)benzene (1). CV measurements
showed that they could be oxidized reversibly in three two-
electron steps, and the possible structures of the oxidized
species were discussed on the basis of quantum chemical
calculations. With E1/2 = −0.39 V vs Fc+/0 for the first two-
electron redox event in CH2Cl2 solution, its electron-donor
capacity is lower than that of 1 (−0.76 V vs Fc+/0). On the
other hand, quantum chemical calculations predict the gas-
phase reduction of 12+ with 3 to give 1 and 32+ to be associated
with a Gibbs free energy change of almost zero, which implies
that the intrinsic or gas-phase electron-donor capacities are
similar for 3 and 1. This result highlights the massive difference
between intrinsic and experienced electron-donor capacities,
which mainly arises from the solvent stabilization of the
charged species. The new redox-active triphenylene derivative 3
can be evaporated without decomposition.
EXPERIMENTAL DETAILS
■
This work concentrated on the coordination chemistry of the
new guanidino-functionalized aromatic (GFA) compound 3.
Trinuclear CuI and CuII complexes were prepared, and their
structures and electronic properties were compared. The special
complex 3(CuI2)3 can be obtained by oxidation of 3(CuI)3 with
I2. Whereas the triphenylene core in the CuI complexes is
curled, it is planar in the CuII complexes. In the CuI complexes,
the metals are trigonal-planar-coordinated, and in the CuII
complexes, the coordination is intermediate between square-
planar and tetrahedral. With the help of TD-DFT calculations,
General Procedures. All reactions were carried out under inert
gas atmosphere using standard Schlenk techniques. NMR spectra were
measured on a DRX-200 or a Bruker Avance III 600 spectrometer.
UV/vis spectra were recorded using a Cary 5000 spectrophotometer.
For IR spectroscopy, CsI disks of the compounds were measured with
an FTIR Biorad Merlin Excalibur FT 3000 spectrometer. Elemental
analyses were carried out at the Microanalytical Laboratory of the
University of Heidelberg. An EG&G Princeton 273 apparatus was used
for the CV measurements. The curves were recorded at different scan
rates in the range of 50−200 mV s−1 with an Ag/AgCl electrode as the
9888
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
Figure 11. Structure of the coordination polymer [3(AgCl)3]n.
Vibrational ellipsoids drawn at the 50% probability level. Hydrogen
atoms omitted for the sake of clarity. (a) One of the triphenylene units
and the silver chloride coordination. Selected bond distances (in Å)
and angles (in deg): Ag1Cl1 2.4549(15), Ag1Cl1′ 2.7913(18),
Ag1N1 2.332(5), Ag1N4 2.380(5), Ag2Cl2 2.448(2), Ag2
Cl2′ 2.832(2), Ag2N7 2.327(5), Ag2N10 2.372(7), Ag3Cl3
2.336(2), Ag3N13 2.256(5), Ag3N16 2.315(5), N1C1
1.398(6), N1C19 1.316(7), N2C19 1.373(7), N3C19
1.356(7), N4C2 1.391(7), N4C24 1.334(8), N5C24
1.357(8), N6C24 1.362(8), N7C7 1.413(8), N7C29
1.329(9), N8C29 1.362(9), N9C29 1.354(9), N10C8
1.410(7), N10C34 1.321(10), N11C34 1.390(8), N12C34
1.366(10), Cl1Ag1Cl1′ 96.15(5), Cl2AgCl2′ 93.07(7),
N1Ag1N4 71.69(16), N7Ag2N10 71.7(2), N13Ag3
N16 74.22(18). (b) Visualization of one of the chains. Two of the Ag
atoms attached to each triphenylene are tetrahedrally coordinated, and
the third one is trigonally coordinated. (c) View parallel to the
triphenylene plane highlighting the displacement of the silver atoms
from the aromatic plane.
Figure 12. (a) Structure of the coordination polymer [3(AgI)3]n. The
iodine atoms are heavily disordered. Vibrational ellipsoids drawn at the
50% probability level. Hydrogen atoms omitted for the sake of clarity.
Selected bond distances (in Å) and angles (in deg): Ag1I1
2.7306(17), Ag1I1′ 2.9239(17), Ag2I2 2.8179(19), Ag2I4
2.745(2), Ag3I3′ 2.756(3), Ag3I3 2.866(5), Ag1N1 2.333(8),
Ag1N4 2.359(5), Ag2N7 2.383(5), Ag3N1 2.474(8), Ag3
N4 2.307(6), N1C1 1.395(9), N1C10 1.263(10), N2C10
1.327(10), N3C10 1.414(13), N4C2 1.425(9), N4C15
1.316(9), N5C15 1.366(11), N6C15 1.383(12), N7C9
1.407(7), N7C20 1.310(8), N8C20 1.385(8), N9C20
1.340(9), I1Ag1I1′ 99.46(5), I3Ag3I3′ 105.04(12), I4
Ag2I4′ 106.07(5), N1Ag1N4 73.0(2), N1Ag3N4
71.3(2), N7Ag2N7′ 73.3(2). (b) Illustration of the layer structure
of the material. Two layers are shown (top layer in color and bottom
layer in gray).
reference electrode and Bu4NPF6 (Fluka, electrochemical grade) as the
electrolyte. For the SQUID direct-current (dc) measurements, a
Quantum Design MPMS-XL 5 instrument was used. The SQUID
magnetometric data were analyzed with the help of the julX
program.51 High-resolution electrospray ionization (HR-ESI) spectra
were recorded on a Finnigan LCQ quadrupole ion trap. High-
resolution fast atom bombardment (HR-FAB) and liquid-injection
field desorption/ionization (LIFDI) spectra were recorded on a JEOL
JMS-700 magnetic sector.
Materials. All solvents were dried with an MBraun Solvent
Purification System prior to their use. 2,3,6,7,10,11-Hexaaminotriphe-
nylene was synthesized according to the literature method.33
9889
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
Triphenylene was purchased from Aldrich (98%) and used as
delivered. The CuI salts CuCl and CuBr; the AgI salts AgCl, AgBr,
and AgCN; and the salts Ni(dme)Cl2 (dme = dimethoxyethane),
Ni(dme)Br2, and CoCl2 were purchased from ABCR. AgI, CuCN, and
CuI were purchased from Strem and Aldrich, respectively. The CuII
159.67 (NC), 131.93 (Carom), 128.49 (Carom), 119.56 (CHarom),
41.79 (CH3) ppm. IR (CsI): ν = 3454 (m), 3209 (w), 3019 (w), 2963
(m), 2814 (w), 1637 (s), 1551 (s), 1470 (m), 1423 (s), 1405 (s), 1307
(s), 1261 (w), 1231 (m), 1171 (s), 1150 (m), 1106 (w), 1069 (m),
1047 (m), 1040 (w), 1012 (m), 923 (w), 900 (m), 796 (m) cm−1.
UV/vis (CH3CN, c = 2.0229 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1
cm−1) = ca. 332 (49599, shoulder), 303 (125182), 243 (49044) nm.
Crystal data for [3H6]Cl6·5H2O, C48H94Cl6N18O5: Mr = 1216.11, 0.40
× 0.20 × 0.20 mm3, monoclinic, space group P21/n, a = 16.938(3) Å, b
= 12.940(3) Å, c = 30.764(6) Å, β = 90.87(3)°, V = 6742(2) Å3, Z = 4,
dcalc = 1.198 Mg·m−3, Mo Kα radiation (graphite-monochromated, λ =
0.71073 Å), T = 100 K, θrange 1.98−27.60°. Reflns measd 117363,
indep 15489, Rint = 0.0671. Final R indices [I > 2σ(I)]: R1 = 0.0504,
wR2 = 0.1234.
salts CuCl and CuBr were obtained from Riedel-de Haen.
̈
2
Synthe2sis of 2,3,6,7,10,11-Hexaaminotriphenylene·6HCl.
The synthesis started with the preparation of 2,3,6,7,10,11-
hexabromotriphenylene according to a literature protocol,32 which
was further converted to the hydrochloride of 2,3,6,7,10,11-
hexaaminotriphenylene.33
Synthesis of 2-Chloro-1,1′,3,3′-tetramethylformamidinium
Chloride (Activated Urea). N,N,N′,N′-Tetramethylurea (0.670 mL,
0.648 g, 5.581 mmol) was dissolved in 20 mL of CHCl3. Subsequently,
oxalyl chloride (2.436 mL, 3.545 g, 27.926 mmol, 30 equiv) was added
slowly dropwise, and the reaction mixture was stirred under reflux for a
period of 16 h. The solvent was removed under a vacuum, and the
remaining solid was washed three times with 6 mL portions of Et2O.
Finally, the solid was dried and dissolved in 15 mL of CH3CN. This
solution was used without further analysis for the preparation of ligand
3.
S y n t h e s i s o f 2 , 3 , 6 , 7 , 1 0 , 1 1 - H e x a k i s -
(dimethylethylenguanidino)triphenylene, 4. 1,3-Dimethylimida-
zolidine-2-on (0.4218 mL, 0.4463 g, 3.9096 mmol) was dissolved in 20
mL of CHCl3. Then, oxalyl chloride (1.7053 mL, 2.4812 g, 19.5480
mmol) was added dropwise. Subsequently, the mixture was heated to
reflux for a period of 16 h. The solvent was removed under a vacuum,
and the remaining solid was washed three times with 10 mL portions
of Et2O and then dissolved in 12 mL of CH3CN. This solution was
added to a suspension of 2,3,6,7,10,11-hexaaminotriphenylene
hexahydrochloride (0.300 g, 0.5585 mmol) in 12 mL of CH3CN,
and NEt3 (1.3350 mL, 1.0173 g, 10.0532 mmol) was slowly added at a
temperature of 0 °C. The suspension was stirred for a period of 4.5 h,
during which time the reaction mixture was allowed to warm from 0
°C to room temperature. The solution turned beige, and a bright solid
formed. Subsequently, the solvent was removed under a vacuum to
obtain a pale-brown-colored solid, which was dissolved in 18 mL of an
aquous HCl solution (10%). After addition of 18 mL of NaOH (25%),
the solution was extracted with CH3CN until the organic phase
remained colorless. The solvent was then removed in vacuo. Upon
recrystallization of the raw product from CH3CN, compound 4 was
obtained as a beige-brown-colored solid. Yield: 0.282 g (0.3151 mmol,
56%). C 48H66N18 (895.16): calcd C 64.40, H 7.43, N 28.16; found C
63.18, H 7.27, N 27.27. HR-ESI+ (MeOH): m/z = 895.57936 [4 +
H]+, 12%, calcd 895.57961; 448.29314 [4 + 2H]2+, 100.0%, calcd
Synthesis of 2,3,6,7,10,11-Hexakis(tetramethylguanidino)-
triphenylene, 3. The solution of 2-chloro-1,1′,3,3′-tetramethylfor-
mamidinium chloride was slowly added to a suspension of
2,3,6,7,10,11-hexaaminotriphenylene·6HCl (0.500 g, 0.931 mmol) in
15 mL of CH3CN. Triethylamine (2.225 mL, 1.696 g, 16.756 mmol,
18 equiv) was slowly added to the reaction mixture at a temperature of
0 °C. Then, the suspension was stirred for a period of 4 h at 0 °C,
during which time it turned red-brown. The red-brown solid obtained
after removal of the solvent under a vacuum was redissolved in a 10%
aqueous HCl solution (15 mL). Subsequently, a 25% aqueous NaOH
solution (15 mL) was added. The solution was extracted with toluene
until the organic phase remained colorless. The combined organic
phases were dried over K2CO3, and the solvent was removed under a
vacuum. Recrystallization of the crude product from CH3CN yielded
0.784 g (0.864 mmol, 93%) of 3 in the form of a beige-colored solid.
The product was further purified by sublimation and was then isolated
as a pale-yellow-colored solid. Elemental analysis for C48H78N18
(907.35): calcd C 63.54, H 8.67, N 27.79; found C 62.95, H 8.60,
N 27.45. HR-FAB+: m/z = 907.6757 [3 + H]+, 100%, calcd 907.6753;
906.6691 [3]+, 44.4%, calcd 906.6657; 862.6138 [3 − NMe2]+, 26.8%,
1
448.29372; 299.19790 [4 + 3H]3+, 9.2%, calcd 299.19842. H NMR
(600.13 MHz, CD2Cl2): δ = 7.66 (s, 6H, Harom), 3.25 (s, 24H, CH2),
2.67 (s, 36H, CH3) ppm. 13C{1H} NMR (150.90 MHz, CD2Cl2),
HSQC, HMBC: δ = 154.59 (NC), 142.52 (Carom), 124.22 (Carom),
115.16 (CHarom), 49.13 (CH2), 35.23 (CH3) ppm. IR (CsI): ν = 3048
(w), 2933 (w), 2844 (w), 2794 (w), 1650 (m), 1588 (m), 1482 (m),
1438 (m), 1415 (m), 1391 (s), 1281 (s), 1243 (s), 1233(s), 1198 (w),
1163 (m), 1137 (m), 1092 (w), 1075 (m), 1039 (s), 999 (m), 990
(w), 969 (s), 960 (s), 880 (s), 853 (w), 845 (w), 823 (w), 769 (m),
725 (m), 713 m), 680 (w), 669 (w), 646 (w), 610 (m), 595 (w), 586
(w) cm−1. UV/vis (CH 2Cl2, c = 2.6364 × 10−5 mol/L, d = 0.5 cm): λ
(ε, M−1 cm−1) = 317 (96675), ca. 354 (43032, shoulder) nm.
Synthesis of 4·3HCl. 4 (296.1 mg, 0.3308 mmol) was dissolved in
8 mL of CH3CN, and 0.99 mL of HCl·Et2O (2 N solution, 1.9847
mmol, 6 equiv) was added. Initially, the formation of a bright solid was
observed, but it later disappeared. After the mixture had been stirred
for a period of 1 h at room temperature, the solvent was removed, and
the product was crystallized by slow diffusion of Et2O into a CH3CN
solution of 4·3HCl. A total of 170.3 mg (0.1609 mmol, 49%) of clean
product (4·3HCl·3H2O) was obtained. 4·3HCl·3H2O,
C48H75N18Cl3O3 (1058.59): calcd C 54.46, H 7.14, N 23.82; found
C 53.43, H 7.12, N 23.99. HR-ESI+ (CH3CN): m/z = 895.58043 ([4 +
H]+, 100.0%, calcd 895.57906), 448.29340 ([4 + 2H]2+, 51.1%, calcd
448.29317), 299.19808 ([4 + 3H]3+, 11.5%, calcd 299.19787). 1H
NMR (600.13 MHz, CD3CN): δ = 10.92 (s, 3H, NH), 8.16 (s, 6H,
Harom), 3.41 (s, 24H, CH2), 2.71 (s, 36H, CH3) ppm. 13C{1H} NMR
(150.90 MHz, CD3CN), HSQC, HMBC: δ = 157.35 (NC), 118.43
(CHarom), 49.30 (CH2), 35.04 (CH3) ppm. IR (CsI): ν = 3453 (m),
3236 (w), 3135 (w), 2937 (m), 2876 (m), 1632 (s), 1585 (s), 1491
(s), 1448 (s), 1419 (s), 1398 (s), 1376 (m), 1301 (s), 1289 (s), 1239
(m), 1142 (w), 1075 (w), 1040 (s), 1009 (m), 967 (s), 886 (w), 844
(w), 795 (w), 772 (w), 748 (w), 698 (m) cm−1. UV/vis (CH3CN, c =
1
calcd 862.6157. H NMR (600.13 MHz, CD3CN): δ = 7.31 (s, 6H,
Harom), 2.69 (s, 72H, CH3) ppm. 13C{1H} NMR (150.90 MHz,
CD3CN), HSQC, HMBC: δ = 159.19 (NC), 144.79 (Carom), 124.48
(Carom), 114.20 (CHarom), 39.95 (CH3) ppm. 1H NMR (600.13 MHz,
C6D6): δ = 8.08 (s, 6H, Harom), 2.57 (s, 72H, CH3) ppm. 13C{1H}
NMR (150.90 MHz, C6D6), HSQC, HMBC: δ = 158.08 (NC),
144.30 (Carom), 125.44 (Carom), 114.77 (CHarom), 39.75 (CH3) ppm.
1H NMR (600.13 MHz, CD2Cl2): δ = 7.37 (s, 6H, Harom), 2.71 (s,
72H, CH3) ppm. 13C{1H} NMR (150.90 MHz, CD2Cl2), HSQC,
HMBC: δ = 158.17 (NC), 143.41 (Carom), 123.47 (Carom), 113.34
(CHarom), 39.15 (CH3) ppm. IR (CsI): ν = 2997 (w), 2926 (m), 2886
(m), 2804 (w), 1594 (s), 1483 (s), 1457 (m), 1424 (m), 1408 (m),
1373 (s), 1294 (w), 1232 (m), 1220(m), 1137 (s), 1062 (w), 1027 (s),
986 (m), 922 (w), 874 (m), 833 (w), 788 (w), 763 (w), 747 (w), 736
(w), 701 (w), 669 (w), 630 (w), 599 (w) cm−1. UV/vis (CH3CN, c =
2.5793 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) = ca. 215 (66076),
322 (94581), ca. 356 (45808) nm.
Synthesis of 3·6HCl. 3 (77.3 mg, 0.0852 mmol) was dissolved in 7
mL of Et2O. Then, 0.51 mL of HCl·Et2O (1 N solution, 0.5112 mmol,
6 equiv) was added. A colorless solid precipitated. After 1 h, the
solvent was removed by filtration, and the solid washed three times
with 3 mL portions of Et2O. Finally, the product was dried to obtain
82.7 mg (0.0734 mmol, 86%) of 3·6HCl, which was recrystallized from
a CH3CN solution. 3·6HCl·3H2O, C48H90N18Cl6O3 (1180.07): calcd
C 48.85, H 7.69, N 21.36; found C 49.20, H 7.89, N 21.49. HR-ESI+:
m/z = 907.67279 ([3 + H]+, 78.6%, calcd 907.67296), 454.34018 ([3
+ 2H]2+, 100.0%, calcd 454.34067). 1H NMR (600.13 MHz, CD3CN):
δ = 11.31 (s, 6H, NH), 8.38 (s, 6H, Harom), 3.10 (s, 72H, CH3)
ppm. 13C{1H} NMR (150.90 MHz, CD3CN), HSQC, HMBC: δ =
9890
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
1.2469 × 10−5 mol/L, d = 1.0 cm): λ (ε, M−1 cm−1) = ca. 352 (42425,
shoulder), 315 (79669) nm. Crystal data for [4H3]Cl3·H2O·CH3CN,
C50H74Cl3N19O: Mr = 1063.63, 0.40 × 0.35 × 0.20 mm3, monoclinic,
space group P21/c, a = 8.599(2) Å, b = 26.487(5) Å, c = 24.255(5) Å,
β = 96.62(3)°, V = 5487.5(19) Å3, Z = 4, dcalc = 1.287 Mg·m−3, Mo Kα
radiation (graphite-monochromated, λ = 0.71073 Å), T = 100 K, θrange
2.29−27.46°. Reflns measd 24582, indep 12516, Rint = 0.0276. Final R
indices [I > 2σ(I)]: R1 = 0.0465, wR2 = 0.1235.
2797 (w), 1629 (m, shoulder), 1572 (s), 1522 (s), 1490 (s), 1466 (s),
1419 (s), 1402 (s), 1331 (m), 1315 (m), 1275 (m), 1232 (m), 1188
(m), 1155 (s), 1106 (w), 1064 (m), 1035 (s), 1002 (m), 926 (w), 915
(w), 866 (s), 845 (m), 826 (m), 805 (m), 755 (w), 737 (m), 726 (m),
702 (m), 662 (w), 630 (m), 602 (w) cm−1. UV/vis (CH2Cl2, c =
2.4202 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) = 397 (47237),
377 (56400), 339 (86659), 311 (61938), 237 (57365) nm. Crystal
data for 3(CuCl)3·1.7CH2Cl2, C49.70H81.40Cl6.40Cu3N18: Mr = 1348.63,
0.40 × 0.30 × 0.22 mm3, monoclinic, space group C2/c, a = 30.408(6)
Å, b = 17.433(4) Å, c = 29.641(6) Å, β = 120.29(3)°, V = 13568(5)
Å3, Z = 8, dcalc = 1.320 Mg·m−3, Mo Kα radiation (graphite-
monochromated, λ = 0.71073 Å), T = 100 K, θrange 2.35−27.53°.
Reflns measd 15539, indep 15539, Rint = 0.0259. Final R indices [I >
2σ(I)]: R1 = 0.0570, wR2 = 0.1746.
Synthesis of [3(CuCN)3]. CuCN (29.9 mg, 0.3340 mmol, 3 equiv)
was added to a solution of 3 (101.0 mg, 0.1113 mmol) in 7 mL of
toluene. The suspension was stirred for a period of 2.5 h at a
temperature of 110 °C. Then, the formed pale-yellow precipitate was
separated by filtration, washed three times with 5 mL portions of Et2O,
and dried. The raw product (115.8 mg, 89%, 0.0985 mmol) was
recrystallized from a CH2Cl2/Et2O mixture. 3(CuCN)3·2CH2Cl2,
C53H82N21Cu3Cl4 (1345.81): calcd C 47.30, H 6.14, N 21.86; C 47.40,
H 6.28, N 22.05. HR-ESI+ (MeOH): m/z = 1176.47175 [3(CuCN)3 +
H]+, 3.0%, calcd 1176.47065; 1149.46224 [3(CuCN)2 + Cu]+, 4.3%,
calcd 1149.45972; 1085.54057 [3(CuCN)2 + H]+, 48.4%, calcd
1085.53886; 1060.53016 [3(CuCN) + Cu]+, 20.0%, calcd 1060.52737;
996.60697 [3(CuCN) + H]+, 79.8%, calcd 996.60619; 530.76771
[3(CuCN) + Cu + H]2+, 28.9%, calcd 530.76760; 498.80724
[3(CuCN) + 2H]2+, 100.0%, calcd 498.80701. 1H NMR (600.13
MHz, CD2Cl2): δ = 7.23 (s, 6H, Harom), 2.93 (s, 72H, CH3) ppm.
13C{1H} NMR (150.90 MHz, CD2Cl2), HSQC, HMBC: δ = 163.96
(NC), 141.95 (Carom), 124.45 (Carom), 110.94 (CHarom), 40.69
(CH3) ppm. IR (CsI): ν = 3065 (w), 3002 (w), 2935 (m), 2874 (m),
2796 (w), 2124 (w, shoulder), 2104 (m), 1516 (s), 1489 (s), 1467 (s),
1420 (s), 1392 (s), 1333 (m), 1313 (m), 1274 (m), 1231 (m), 1188
(m), 1155 (s), 1107 (w), 1063 (m), 1034 (s), 1000 (s), 917 (m), 869
(s), 849 (m), 809 (s), 756 (m), 728 (m), 703 (m), 664 (w), 630 (m),
601 (w) cm−1. IR (CHCl3 solution): ν = 3000 (m), 2945 (w), 2891
(w), 2870 (w), 2798 (w), 2135 (w), 2100 (m), 1531 (s), 1489 (s),
1468 (m), 1420 (s), 1407 (w), 1393 (s), 1335 (w), 1311 (w), 1273
(w), 1244 (w), 1154 (m), 1141 (m), 1106 (w), 1063 (w), 1034 (m),
1000 (m) cm−1. UV/vis (CH2Cl2, c = 2.8913 × 10−5 mol/L, d = 0.5
cm): λ (ε, M−1 cm−1) = 397 (38643), 379 (45428), 340 (71506), 311
(55128), ca. 228 (67249) nm. Crystal data for 3(CuCN)3·4.6CH2Cl2,
C53.30H82.60Cl4.60Cu3N21: Mr = 1377.69, 0.35 × 0.20 × 0.15 mm3,
Synthesis of 3(CuBr)3. A solution of 3 (47.9 mg, 0.0527 mmol) in
6 mL of toluene was added to a Schlenk flask containing CuI (22.7 mg,
0.1582 mmol, 3 equiv). The suspension was stirred in boiling toluene
for a period of 2.5 h. The pale-brown-colored precipitate was separated
by filtration, washed two times with 3 mL portions of toluene and then
three times with 3 mL portions of Et2O, and subsequently dried. The
product was isolated as a yellow solid in a yield of 66% (46.3 mg,
0.0346 mmol). C48H78N18Cu3Br3 (1337.60): calcd C 43.10, H 5.88, N
18.85; found C 42.45, H 5.67, N 18.81. HR-ESI+ (CH2Cl ): m/z =
2
1336.21038 [3(CuBr)3]+, 3.7%, calcd 1336.20625; 1257.29367
[3(CuBr)2 + Cu]+, 40.0%, calcd 1257.28819; 1195.36980 [3(CuBr)2
+ H]+, 100.0%, calcd 1195.36656; 1051.52683 [3(CuBr) + H]+, 3.5%,
calcd 1051.52028; 557.22727 [3(CuBr) + Cu + H]2+, 12.2%, calcd
557.22487; 526.26615 [3(CuBr)2 + 2H]2+, 8.8%, calcd 526.26405. 1H
NMR (600.13 MHz, CD2Cl2): δ = 7.27 (s, 6H, Harom), 2.93 (s, 72H,
CH3) ppm. 13C{1H} NMR (150.90 MHz, CD2Cl2), HSQC, HMBC: δ
= 163.31 (CN), 141.60 (Carom), 124.50 (Carom), 111.27 (CHarom),
40.45 (CH3) ppm. IR (CsI): ν = 3066 (w), 3001 (w), 2930 (m), 2872
(m), 2793 (w), 1567 (s, shoulder), 1533 (s), 1520 (s), 1491 (s), 1466
(s), 1417 (s), 1395 (s), 1388 (s), 1332 (s), 1315 (m), 1275 (m), 1227
(m), 1191 (w), 1152 (s), 1105 (w), 1064 (w), 1033 (s), 1001 (s), 970
(w), 918 (w), 861 (m), 848 (m), 815 (m), 805 (m), 755 (w), 737 (w),
725 (m), 699 (m), 630 (w), 601 (w) cm−1. UV/vis (CH Cl2, c =
2
1.8690 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) = 397 (50691),
377 (60644), 341 (92132), 310 (60925), 231 (66648) nm. Crystal
data for 3(CuBr)3·2.8CH2Cl2, C50.80H83.60Br3 Cl 5.60Cu3N18, Mr =
triclinic, space group P1, a = 15.275(3) Å, b = 15.419(3) Å, c =
̅
17.478(4) Å, α = 104.69(3)°, β = 108.99(3)°, γ = 109.05(3)°, V =
3368.1(12) Å3, Z = 2, dcalc = 1.358 Mg·m−3, Mo Kα radiation
(graphite-monochromated, λ = 0.71073 Å), T = 100 K, θrange 2.17−
29.10°. Reflns measd 33156, indep 17928, Rint = 0.0380. Final R
indices [I > 2σ(I)]: R1 = 0.0548, wR2 = 0.1463.
1575.43, 0.40 × 0.30 × 0.25 mm3, triclinic, space group P1, a =
̅
15.234(3) Å, b = 15.525(3) Å, c = 17.541(4) Å, α = 109.20(3), β =
3
104.86(3)°, γ = 109.32(3), V = 3372.3(12) Å , Z = 2, dcalc = 1.552
Mg·m−3, Mo Kα radiation (graphite-monochromated, λ = 0.71073 Å),
T = 100 K, θ range 2.48−27.62°. Reflns measd 62325, indep 15523, Rint
= 0.1014. Final R indices [I > 2σ(I)]: R1 = 0.0718, wR2 = 0.1888.
Synthesis of 3(CuI)3. A solution of compound 3 (50.0 mg, 0.0551
mmol) in 4 mL of toluene was added to a suspension of CuI (31.5 mg,
0.1653 mmol, 3 equiv) in 2 mL of toluene. The reaction mixture was
stirred for a period of 3 h at a temperature of 110 °C. The pale-yellow
precipitate was separated from the mixture by filtration, washed first
two times with 2 mL portions of toluene and then three times with 2
mL portions of Et2O, and finally dried. A yield of 52.0 mg (64%,
0.0352 mmol) was obtained in the form of a yellow solid.
C48H78N18Cu3I3 (1478.60): calcd C 38.99, H 5.32, N 17.05; found
C 38.82, H 4.86, N 17.20. HR-ESI+ (CH2Cl2, MeOH): m/z =
1289.34489 [3(CuI)2 + H]+, 6.4%, calcd 1289.34101; 1225.42371
[3(CuI) + I + 2H]+, 9.4%, calcd 1225.41987; 1097.50696 [3(CuI) +
H]+, 8.3%, calcd 1097.50758; 549.25753 [3(CuI) + 2H]2+, 100.0%,
Synthesis of 3(CuCl)3. 3 (35.6 mg, 0.0394 mmol) and CuCl (11.7
mg, 0.1182 mmol, 3 equiv) were suspended in 6 mL of toluene and
stirred for a period of 3 h at 110 °C. A beige precipitate formed that
was separated by filtration, washed first two times with 3 mL portions
of toluene and then three times with 3 mL portions of Et2O, and
finally dried. The raw product, obtained in this way in the form of a
yellow solid in 70% yield (22.4 mg, 0.0277 mmol), was recrystallized
from a CH2Cl2/Et2O mixture. 3(CuCl)3·CH2Cl2, C49H80N18Cu3Cl5
(1289.18): calcd C 45.65, H 6.25, N 19.56; found C 45.66, H 6.31, N
19.84. HR-ESI+ (MeCN): m/z = 1241.33935 [3(CuCl)3 + Cl + 2H]+,
2.1%, calcd 1241.33910; 1203.336608 [3(CuCl)3 + H]+, 1.3%, calcd
1203.36537; 1169.41784 [3(CuCl)2 + Cu + 2H]+, 2.4%, calcd
1169.40628; 1141.44425 [3(CuCl)2 + Cl + 2H]+, 5.8%, calcd
1141.44536; 1105.46783 [3(CuCl)2 + H]+, 8.2%, calcd 1105.46891;
1043.54575 [3(CuCl) + Cl + 2H]+, 12.0%, calcd 1043.54725;
1005.57172 [3(CuCl) + H]+, 8.7%, calcd 1005.57196; 907.67329 [3
+ H]+, 14.4%, calcd 907.67351; 536.26336 [3(CuCl) + Cu + 3H]2+,
6.4%, calcd 536.25802; 522.27653 [3(CuCl) + Cl + 3H]2+, 14.8%,
calcd 522.27754; 503.28948 [3(CuCl) + 2H]2+, 21.3%, calcd
503.28989; 454.34015 [3 + 2H]2+, 100.0%, calcd 454.34067;
1
calcd 549.25770. H NMR (600.13 MHz, CD2Cl2): δ = 7.25 (s, 6H,
Harom), 2.96 (s, 72H, CH3) ppm. 13C{1H} NMR (150.90 MHz,
CD2Cl2), HSQC, HMBC: δ = 163.60 (NC), 141.89 (Carom), 124.31
(Carom), 111.14 (CHarom), 40.63 (CH3) ppm. IR (CsI): ν = 3065 (w),
3004 (w), 2936 (w), 2873 (w), 2793 (w), 1566 (s), 1534 (s), 1522 (s),
1490 (s), 1466 (s), 1418 (s), 1391 (s), 1333 (m), 1314 (m), 1274 (w),
1232 (m), 1190 (w), 1154 (s), 1106 (w), 1063 (w), 1033 (m), 1001
(m), 925 (w), 917 (w), 867 (w), 847 (w), 813 (w), 804 (w), 754 (w),
738 (w), 725 (w), 701 (w), 627 (w), 602 (w), 504 (w) cm−1. UV/vis
(CH2Cl2, c = 1.8125 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) =
1
303.22929 [3 + 3H]3+, 32.9%, calcd 303.22972. H NMR (600.13
MHz, CD2Cl2): δ = 7.26 (s, 6H, Harom), 2.91 (s, 72H, CH3) ppm.
13C{1H} NMR (150.90 MHz, CD2Cl2), HSQC, HMBC: δ = 163.28
(NC), 141.56 (Carom), 124.42 (Carom), 111.31 (CHarom), 40.50
(CH3) ppm. IR (CsI): ν = 3066 (w), 3004 (w), 2936 (m), 2880 (m),
9891
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
399 (40375), 380 (47672), 342 (75425), 313 (50722), 237 (60425)
nm.
CH2Cl2 and recrystallized by slow diffusion of Et2O and CHCl3 into
the solution. C48H78N18Cu3I6 (1859.32): calcd C 31.01, H 4.23, N
13.56; found C 30.45, H 4.35, N 13.22. HR-ESI+ (CH2Cl2 + MeOH):
m/z = 1287.34364 [3(CuI)2 + H]+, 7.6%, calcd 1287.34165;
1197.40610 [3(CuI)(CuCl) + H]+, 2.9%, calcd 1197.40486;
1097.50971 [3(CuI) + H]+, 3.4%, calcd 1097.50758; 643.67160
[3(CuI)2 + H]2+, 43.6%, calcd 643.67082; 598.70275 [3(CuI)(CuCl)
+ H]2+, 20.6%, calcd 598.70243; 580.21949 [3(CuI) + Cu + H]2+,
8.2%, calcd 580.21859; 549.25793 [3(CuI) + 2H]2+, 100.0%, calcd
549.25770. IR (CsI): ν = 3006 (w), 2929 (w), 2889 (w), 2863 (w),
2790 (w), 1617 (w, sholder), 1570 (s), 1513 (s), 1489 (s), 1465 (m),
1457 (m), 1415 (s), 1399 (s), 1327 (m), 1321 (m), 1274 (w), 1231
(m), 1222 (m), 1189 (w), 1161 (s), 1139 (w), 1104 (w), 1062 (w),
1040 (m), 1005 (m), 923 (w), 911 (w), 862 (m), 821 (m), 803 (w),
763 (w), 735 (w), 720 (w), 701 (w), 668 (w), 626 (w), 503 (w) cm−1.
UV/vis (CH2Cl2, c = 1.3123 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1
cm−1) = 911 (4281), 667 (2606), ca. 497 (8106, shoulder), 400
(53747), 380 (63305), 345 (88618), 310 (43963), 234 (58719,
shoulder) nm. Crystal data for 3(CuI2)3·4.45CHCl3·0.8CH2Cl2,
C53.25H84.05Cl14.95Cu3I6N18: Mr = 2458.43, 0.30 × 0.20 × 0.15 mm3,
monoclinic, space group P21/c, a = 28.958(6) Å, b = 17.326(4) Å, c =
20.477(4) Å, β = 107.67(3)°, V = 9789(3) Å3, Z = 4, dcalc = 1.668 Mg·
m−3, Mo Kα radiation (graphite-monochromated, λ = 0.71073 Å), T =
Synthesis of 3(CuCl2)3. A solution of 3 (87.5 mg, 0.0964 mmol)
in 4 mL of CH3CN was added to a solution of CuCl2 (37.6 mg, 0.2797
mmol, 2.9 equiv) in 4 mL of CH3CN. The deep-green-colored
reaction mixture was stirred at room temperature for a period of 1 h.
Then, the solvent was removed in vacuo, and the green solid was
washed first two times with 4 mL portions of toluene and then three
times with 4 mL portions of Et2O. After drying, the product was
obtained as a green solid in a yield of 82% (103.1 mg, 0.0787 mmol).
The product was recrystallized from a CH2Cl2/C6H14 (hexane)
mixture. C48H78N18Cu3Cl6 (1310.61): calcd C 43.99, H 6.00, N 19.24;
found C 43.72, H 5.94, N 18.79. HR-ESI+ (CH3CN): m/z =
1309.26450 [3(CuCl2)3]+, 2.4%, calcd 1309.26348; 1274.29475
[3(CuCl2)2(CuCl)]+, 3.3%, calcd 1274.29482; 1174.39875
[3(CuCl2)2]+, 6.7%, calcd 1174.39840; 1139.43018 [3(CuCl2)-
(CuCl)]+, 14.8%, calcd 1139.42971; 1039.53460 [3(CuCl2)]+, 2.4%,
calcd 1039.53299; 1026.33537 [3(CuCl2)(CuCl) + H − NC-
(NMe2)2]2+, 4.8%, calcd 1026.33429; 620.16706 [3(CuCl2)(CuCl)2
+ H]2+, 18.7%, calcd 620.16701; 587.19930 [3(CuCl2)2]2+, 17.4%,
calcd 620.19920, 552.23051 [3(CuCl)2]2+, 38.9%, calcd 552.23054,
520.76547 [3(CuCl2)]2+, 34.6%, calcd 520.76580, 502.78619
[3(CuCl) + H]2+, 67.9%, calcd 502.78598. IR (CsI): ν = 3011 (w),
2932 (m), 2893 (w), 2870 (w), 2795 (w), 1624 (w, shoulder), 1577
(s), 1515 (s), 1491 (s), 1460 (m), 1418 (s), 1400 (s), 1329 (m), 1319
(m), 1277 (m), 1231 (m), 1189 (w), 1163 (m), 1140 (w), 1105 (w),
1064 (m), 1041 (m), 1006 (m), 924 (w), 912 (w), 886 (w), 865 (m),
824 (m), 806 (w), 766 (w), 737 (w), 721 (w), 701 (w), 668 (w), 665
(w), 632 (w), 628 (w) cm−1. Evans NMR (CD2Cl2): Δν = 28.225 Hz,
χMol = 5.1478 × 10−3 cm3/mol, μeff = 3.53 μB, [μSO(3e−) = 3.87 μB].
UV/vis (CH2Cl2, c = 1.9228 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1
cm−1) = 1132 (832), 688 (1486), ca. 490 (1551, shoulder), 397
(53500), 376 (63760), 341 (97461), 307 (60956), 240 (48683) nm.
100 K, θrange 0.74−27.70°. Reflns measd 155781, indep 22583, Rint
0.0668. Final R indices [I > 2σ(I)]: R1 = 0.0612, wR2 = 0.1769.
=
Synthesis of 3(NiCl2)3. A solution of 3 (32.6 mg, 0.0359 mmol) in
3 mL of CH2Cl2 was cooled to −78 °C and then added to a
suspension of NiCl2·dme (22.8 mg, 0.1041 mmol, 2.9 equiv) in 3 mL
of CH2Cl2 that had also been cooled to −78 °C. Then, the reaction
mixture was slowly allowed to warm to room temperature overnight.
After removal of the solvent under a vacuum, the red-brown residue
was washed three times with 3 mL portions of C6H14 (hexane) and
subsequently dried. The product was obtained in a yield of 38.2 mg
(0.0295 mmol, 82%). It was recrystallized from CH2Cl2/C6H14.
C48H78N18Ni3Cl6 (1296.05): calcd C 44.48, H 6.07, N 19.45; found C
44.00, H 6.29, N 18.01. LIFDI (CH2Cl2): m/z = 1259.3 [3(NiCl2)3 −
Cl]+, 19%; 1129.5 [[3(NiCl2)2 − Cl]+, 30%; 1037.5 [[3(NiCl2) + H]+,
53%; 999.4 [[3(NiCl2) − Cl]+, 43%; 907.6 [3 + H]+, 100%. 1H NMR
(200 MHz, CD2Cl2): δ = 45.85 (s, CH3), 9.17 (s, Harom). IR (CsI): ν =
3009 (w), 2940 (m), 2893 (w), 2800 (w), 1558 (s), 1522 (s), 1491
(s), 1466 (m), 1420 (s), 1410 (s), 1398 (s), 1333 (m), 1314 (m),
1273 (w), 1238 (m), 1188 (w), 1161 (m), 1142 (m), 1109 (w), 1065
(w), 1040 (m), 1005 (w), 986 (w), 914 (w), 864 (m), 822 (m), 762
(w), 737 (w), 723 (w), 708 (w), 669 (w), 635 (w), 602 (w) cm−1.
UV/vis (CH3CN, c = 1.88 × 10−5 mol/L, d = 1 cm): λ (ε, M−1 cm−1)
= 684 (274), 580 (618), 512 (1350), ca. 453 (3691, shoulder), 365
(49395), 331 (83446), 235 (51455) nm. Crystal data for 3(NiCl2)3·
4CH2Cl2, C52H86Cl14N18Ni3, Mr = 1635.82, 0.40 × 0.30 × 0.30 mm3,
Crystal data for 3(CuCl2)3·3CH2Cl2, C51H84Cl12Cu3N18, Mr
=
1565.38, 0.50 × 0.50 × 0.45 mm3, monoclinic, space group C2/c, a
= 28.949(6) Å, b = 17.537(4) Å, c = 15.714(3) Å, β = 96.85(3)°, V =
7921(3) Å3, Z = 4, dcalc = 1.313 Mg·m−3, Mo Kα radiation (graphite-
monochromated, λ = 0.71073 Å), T = 100 K, θrange 2.63−27.49°.
Reflns measd 17404, indep 8979, Rint = 0.0207. Final R indices [I >
2σ(I)]: R1 = 0.0625, wR2 = 0.1835.
Synthesis of 3(CuBr2)3. Compound 3 (50.0 mg, 0.0551 mmol)
was dissolved in 3 mL of CH3CN, and the solution was added to a
solution of CuBr2 (36.9 mg, 0.1653 mmol, 3 equiv) in 3 mL of
CH3CN. The initially shiny deep-green-colored solution turned olive
green-brown, and a dark solid precipitate formed after some time. The
reaction mixture was stirred at room temperature for a period of 1.5 h,
and subsequently, the solvent was removed under a vacuum. The solid
was washed three times with 2 mL portions of Et2O and dried. The
product was isolated as an olive-green solid in a yield of 96% (83.0 mg,
0.0526 mmol). C48H78N18Cu3Br6 (1577.32): calcd C 36.55, H 4.98, N
15.98; obsd C 36.42, H 4.75, N 15.81. FAB+: m/z = 1576.9
[3(CuBr2)3]+, 5.5%; 1498.1 [3(CuBr2)2 + CuBr]+, 10.0%; 1417.2
[3(CuBr2)2 + Cu]+, 12.0%; 1355.2 [3(CuBr2)2 + H]+, 10.5%; 1273.3
[3(CuBr2) + CuBr]+, 22.5%; 1194.4 [3(CuBr2) + Cu]+, 32.0%; 1050.5
[3(CuBr)]+, 24.0%; 906.6 [3]+, 74.0%. HR-FAB+: m/z = 1273.2692
[3(CuBr2) + CuBr]+, 74.0%, calcd 1273.2768; 1194.3564 [3(CuBr2) +
Cu]+, 100.0%, calcd 1194.3588. IR (CsI): ν = 3010 (w), 2930 (m),
2895 (w), 2870 (w), 2796 (w), 1624 (w, shoulder), 1574 (s), 1513
(s), 1490 (s), 1459 (s), 1417 (s), 1399 (s), 1327 (m), 1277 (m), 1232
(m), 1188 (w), 1162 (m), 1138 (m), 1103 (w), 1063 (m), 1040 (m),
1006 (m), 923 (w), 911 (w), 864 (m), 823 (m), 804 (w), 762 (w),
737 (w), 719 (w), 702 (w), 662 (w), 628 (w), 502 (w) cm−1. UV/vis
(CH2Cl2, c = 1.7371 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) =
1152 (1097), 754 (2132), ca. 490 (3519, shoulder), 399 (53447), 377
(64169), 342 (93861), 310 (55362), 240 (46440) nm.
triclinic, space group P1, a = 16.234(3) Å, b = 17.325(4) Å, c =
̅
17.430(4) Å, α = 93.05(3)°, β = 106.58(3)°, γ = 117.31(3)°, V =
4076.1(14) Å3, Z = 2, dcalc = 1.333 Mg·m−3, Mo Kα radiation
(graphite-monochromated, λ = 0.71073 Å), T = 100 K, θrange 2.03−
28.50°. Reflns measd 69667, indep 20626, Rint = 0.0556. Final R
indices [I > 2σ(I)]: R1 = 0.0917, wR2 = 0.2696.
Synthesis of 3(NiBr2)3. A solution of 3 (100.0 mg, 0.1102 mmol)
in 6 mL of CH2Cl2 was cooled to −78 °C and then added to a
suspension of NiBr2·dme (100.9 mg, 0.3269 mmol, 2.97 equiv) in 5
mL of CH2Cl2, also kept at a temperature of −78 °C. Then, the
reaction mixture was slowly allowed to warm to room temperature
overnight. The red-brown suspension was filtered, and the solvent was
removed from the filtrate under a vacuum. The brown solid was
washed three times with 4 mL portions of C6H14 and then dried. The
product was obtained in a yield of 132.0 mg (0.0845 mmol, 77%).
C48H78N18Ni3Br6 (1562.98): calcd C 36.89, H 5.03, N 16.13; found C
1
36.83, H 5.14, N 15.71. H NMR (200 MHz, CD2Cl2): δ = 46.28 (s,
Synthesis of 3(CuI2)3. A solution of I2 (25.7 mg, 0.1014 mmol) in
2.5 mL of CH3CN was added to a suspension of 3(CuI)3 (50.0 mg,
0.0338 mmol) in 7 mL of CH3CN. The dark-brown reaction mixture
was stirred at room temperature for a period of 1.5 h and then filtered.
The black residue (40.0 mg, 0.0215 mmol, 64%) was dissolved in
CH3), 9.94 (s, Harom). IR (CsI): ν = 3007 (w), 2934 (m), 2887 (w),
2799 (w), 1552 (s), 1522 (s), 1491 (s), 1466 (m), 1418 (s), 1410 (s),
1398 (s), 1333 (m), 1317 (m), 1277 (m), 1233 (m), 1188 (w), 1161
(m), 1142 (m), 1107 (w), 1063 (w), 1040 (m), 1007 (m), 916 (w),
864 (m), 824 (m), 802 (w), 760 (w), 737 (w), 723 (w), 704 (w), 669
9892
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
(w), 633 (w), 602 (w) cm−1. UV/vis (CH3CN, c = 1.30 × 10−5 mol/L,
d = 1 cm): λ (ε, M−1 cm−1) = ca. 708 (296), ca. 607 (987), ca. 539
(630), ca. 466 (3317), 364 (51868), 332 (84890) nm.
1337.29419 [3(AgCl)3 + H]+, 0.8%, calcd 1337.28890; 1193.42227
[3(AgCl)2 + H]+, 4.5%, calcd 1193.420811; 1087.52383 [3(AgCl) +
Cl + 2H]+, 10.3%, calcd 1087.52063; 1051.54752 [3(AgCl) + H]+,
5.6%, calcd 1051.547229; 907.67473 [3 + H]+, 8.9%, calcd 907.67351;
543.26629 [3(AgCl) + Cl + 3H]2+, 13.6%, calcd 543.26598; 526.27739
[3(AgCl) + 2H]2+, 100.0%, calcd 526.27753; 454.34068 [3 + 2H]2+,
43.0%, calcd 454.34067; 303.22945 [3 + 3H]3+, 23.6%, calcd
303.22972. 1H NMR (600.13 MHz, CD2Cl2): δ = 7.25 (s, 6H,
Harom), 2.87 (s, 72H, CH3) ppm. 13C{1H} NMR (150.90 MHz,
CD2Cl2), HSQC, HMBC: δ = 163.65 (NC), 140.90 (Carom), 124.60
(Carom), 112.65 (CHarom), 40.58 (CH3) ppm. IR (CsI): ν = 3061 (w),
3011 (w), 2995 (w), 2932 (m), 2869 (m), 2792 (w), 1564 (s), 1539
(s), 1519 (s), 1492 (s), 1468 (s), 1417 (s), 1396 (m), 1388 (s), 1333
(m), 1311 (m), 1275 (m), 1229 (s), 1194 (m), 1150 (s), 1105 (w),
1058 (m), 1034 (s), 1001 (s), 920 (w), 863 (m), 846 (m), 811 (m),
799 (m), 754 (w), 736 (w), 725 (m), 701 (m), 677 (w), 672 (w), 665
(w), 659 (w), 630 (w), 621 (w), 599 (m) cm−1. UV/vis (CH2Cl2, c =
1.5610 × 10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) = ca. 394 (39327,
shoulder), 374 (52493), 336 (91060), 308 (57287), 245 (46739) nm.
Crystal data for [3(AgCl)3]n, C48H78Ag3Cl3N18: Mr = 1337.24, 0.40 ×
Synthesis of 3(CoCl2)3. A solution of ligand 3 (54.9 mg, 0.0606
mmol) in 3.0 mL of CH3CN was added to a solution of CoCl2 (22.8
mg, 0.1756 mmol, 2.9 equiv) in 2.0 mL of CH3CN. The deep-blue-
colored solution was stirred at room temperature for a period of 2 h,
during which time a blue precipitate formed. Then, the solvent was
removed under a vacuum, and the solid was washed first two times
with 3 mL portions of toluene and then two times with 3 mL portions
of Et2O. The product was obtained in a yield 57.2 mg (0.0441 mmol,
73%). It was recrystallized from CH2Cl2/Et2O. C48H78N18Co3Cl6
(1296.77): calcd C 44.46, H 6.06, N 19.44, obsd C 44.20, H 6.14,
N 19.12. HR-ESI+ (CH3CN): m/z = 1260.30662 [3(CoCl2)3 − Cl]+,
21.9%, calcd 1260.30740; 1167.41237 [3(CoCl2)2 + H]+, 23.6%, calcd
1167.41336; 612.66889 [3(CoCl2)3 − 2Cl]2+, 29.1%, calcd 612.66935;
565.22298 [3(CoCl2) + Cl + H]2+, 100.0%, calcd 565.22323. 1H NMR
(200 MHz, CD2Cl2): δ = 31.06 (s, CH3), 22.19 (s, Harom). IR (CsI): ν
= 3010 (w), 2939 (w), 2800 (w), 1558 (s), 1521 (s), 1491 (m), 1495
(m), 1420 (s), 1407 (s), 1399 (s), 1331 (m), 1314 (m), 1270 (m),
1239 (s), 1188 (w), 1160 (m), 1141 (m), 1108 (w), 1064 (w), 1039
(m), 1004 (m), 985 (m), 913 (w), 863 (m), 820 (m), 724 (w), 708
0.30 × 0.30 mm3, triclinic, space group P1, a = 13.099(3) Å, b =
̅
17.406(4) Å, c = 17.403(4) Å, α = 61.82(3)°, β = 71.90(3)°, γ =
74.91(3)°, V = 3293.8(11) Å3, Z = 2, dcalc = 1.348 Mg·m−3, Mo Kα
radiation (graphite-monochromated, λ = 0.71073 Å), T = 100 K, θrange
2.26−27.52°. Reflns measd 27392, indep 14858, Rint = 0.0360. Final R
indices [I > 2σ(I)]: R1 = 0.0855, wR2 = 0.2166.
(w), 665 (w) cm−1. UV/vis (CH CN, c = 1.59 × 10−5 mol/L, d = 1
3
−1
cm): λ (ε, M
cm−1) = 664 (1195), 622 (1066), 568 (824), 550
(786), 373 (64906), 331 (103836) nm. Crystal structure measured
(see Supporting Information).
Synthesis of 3(AgCN)3. 3 (40 mg, 0.0441 mmol) and AgCN (17.7
mg, 0.1322 mmol, 3 equiv) were suspended in 5 mL of toluene. The
reaction mixture was stirred in refluxing toluene for a period of 2.5 h.
Then, the pale-yellow-colored precipitate was separated by filtration,
washed two times with 2 mL portions of toluene and three times with
2 mL portions of Et2O, and subsequently dried. The raw product (46.4
mg, 0.0354 mmol, 80%) was finally recrystallized from a CH2Cl2/Et2O
mixture. 3(AgCN) 3·CH2Cl2, C52H80N21Ag3Cl2 (1393.84): calcd C
44.81, H 5.79, N 21.10; found C 44.71, H 5.79, N 21.45. HR-ESI+
(MeOH): m/z = 1308.40301 [3(AgCN)3 + H]+, 9.8%, calcd
1308.39805; 1281.39241 [3(AgCN)3 − CN]+, 10.6%, calcd
1281.387136; 1175.49601 [3(AgCN)2 + H]+, 27.2%, calcd
1175.49003; 1148.48384 [3(AgCN)2 − CN]+, 39.0%, calcd
1148.479114; 1042.58486 [3(AgCN) + H]+, 100.0%, calcd
1042.582270; 907.67809 [3 + H]+, 9.4%, calcd 907.6735; 520.79592
Synthesis of [3(AgBr)3]n. AgBr (50.9 mg, 0.2711 mmol, 3 equiv)
was added under exclusion of light to a solution of 3 (82.0 mg, 0.0904
mmol, 1 equiv) in 8 mL of toluene. The suspension was stirred
overnight at a temperature of 110 °C. The pale-yellow-colored flaky
precipitate was separated by filtration, washed first three times with 2
mL portions of toluene and then three times with 2 mL portions of
Et2O, and subsequently dried. The product was obtained in a yield of
122.8 mg (0.0835 mmol, 92%). It was recrystallized from CH2Cl2/
Et2O. C48H78N18Ag3Br3 (1470.57): calcd C 39.20, H 5.35, N 17.14;
found C 38.76, H 5.41, N 17.06. HR-ESI+ (CH3CN): m/z =
1445.16614 [3 + (AgBr)2 + 2Br + 3H]+, 2.7%, calcd 1445.16601;
1175.42101 [3(AgBr) + Br + 2H]+, 3.5%, calcd 1175.42200;
1095.49507 [3(AgBr) + H]+, 4.2%, calcd 1095.49628; 907.67261 [3
+ H]+, 19.9%, calcd 907.67351; 588.21390 [3(AgBr) + Br + 3H]2+,
19.5%, calcd 588.21491; 548.25119 3(AgBr) + 2H]2+, 24.1%, calcd
548.25201; 454.33998 [3 + 2H]2+, 100.0%, calcd 454.34067;
303.22913 [3 + 3H]3+, 95.9%, calcd 303.22972. FAB+: m/z =
1390.4 [3(AgBr)2 + Ag]+, 1283.5 [3(AgBr)2 + H]+, 1203.5 [3(AgBr) +
1
[3(AgCN) + 2H]2+, 19.8%, calcd 520.79475. H NMR (600.13 MHz,
CD2Cl2): δ = 7.20 (s, 6H, Harom), 2.86 (s, 72H, CH3) ppm. 13C{1H}
NMR (150.90 MHz, CD2Cl2), HSQC, HMBC: δ = 164.08 (NC),
141.28 (Carom), 124.59 (Carom), 112.34 (CHarom), 40.63 (CH3) ppm.
IR (CsI): ν = 3061 (w), 3002 (w), 2932 (m), 2875 (m), 2795 (w),
2126 (m), 1533 (s), 1489 (s), 1469 (s), 1420 (s), 1390 (s), 1333 (m),
1306 (m), 1275 (m), 1232 (m), 1190 (w), 1152 (s), 1107 (w), 1062
(m), 1035 (s), 1000 (s), 920 (w), 869 (m), 846 (m), 801 (m), 753
(w), 737 (w), 726 (w), 702 (m), 668 (w), 661 (w), 629 (w), 600 (m)
cm−1. UV/vis (CH 2Cl2, c = 2.0475 × 10−5 mol/L, d = 0.5 cm): λ (ε,
M−1 cm−1) = ca. 394 (37702), 374 (49211), 336 (84641), 308
(54546), ca. 231 (60829) nm. Crystal data for [3(AgCN)3]·1.15CH
2Cl2, C52.15H80.30Ag3Cl2.30N21: Mr = 1406.62, 0.22 × 0.20 × 0.20 mm3,
Ag]+, 1094.9 [3(AgBr)] , 1014.3 [3 + Ag + H]+. H NMR (600.13
MHz, CD2Cl2): δ = 7.25 (s, 6H, Harom), 2.87 (s, 72H, CH3) ppm.
13C{1H} NMR (150.90 MHz, CD2Cl2), HSQC, HMBC: δ = 163.75
(NC), 140.99 (Carom), 124.58 (Carom), 112.64 (CHarom), 40.65
(CH3) ppm. IR (CsI): ν = 3060 (w), 3000 (w), 2930 (m), 2880 (m),
2793 (w), 1539 (s), 1490 (s), 1466(s), 1457 (w), 1437 (w), 1420 (s),
1388 (s), 1334 (m), 1309 (m), 1275 (m), 1228 (m), 1192 (w), 1151
(s), 1106 (w), 1064 (m), 1034 (s), 1000 (s), 921 (w), 869 (m), 845
(m), 810 (m), 798 (m), 754 (w), 736 (m), 725 (m), 702 (m), 668
(w), 659 (w), 630 (w), 600 (m) cm−1. UV/vis (CH 2Cl2, c = 1.7952 ×
10−5 mol/L, d = 0.5 cm): λ (ε, M−1 cm−1) = 373 (51976), 335
(91534), 308 (59006), ca. 244 (56924) nm. Crystal data for
[3(AgBr)3]n·1.75CH2Cl2, C49.75H81.50Ag3Br3Cl3.50N18: Mr = 1619.24,
0.30 × 0.28 × 0.25 mm3, monoclinic, space group P21/ c, a =
13.158(3) Å, b = 18.088(4) Å, c = 28.453(6) Å, β = 96.22(3)°, V =
6732(2) Å 3, Z = 4, dcalc = 1.598 Mg·m−3, Mo Kα radiation (graphite-
+
1
triclinic, space group P1, a = 17.508(4) Å, b = 17.583(4) Å, c =
̅
25.244(5) Å, α = 79.13(3)°, β = 73.60(3)°, γ = 60.14(3)°, V =
6455(2) Å 3, Z = 4, dcalc = 1.447 Mg·m−3, Mo Kα radiation (graphite-
monochromated, λ = 0.71073 Å), T = 100 K, θ
Reflns measd 56287, indep 29645, Rint = 0.0524. Final R indices [I >
2σ(I)]: R1 = 0.0683, wR2 = 0.1567.
0.84−27.57°.
range
Synthesis of [3(AgCl)3]n. Compound 3 (98 mg, 0.1080 mmol)
was dissolved in 9 mL of toluene, and then AgCl (46.4 mg, 0.3240
mmol, 3 equiv) was added to this solution. The suspension was stirred
in refluxing toluene for a period of 3 h. The flaky pale-yellow
precipitate was separated by filtration. Then, it was washed five times
with 4 mL portions of toluene and subsequently five times with 4 mL
portions of Et2O, and dried. The raw product (79.1 mg, 0.0592 mmol,
55%) was recrystallized from a CH2Cl2/Et2O mixture. 3(AgCl)3·
CH2Cl2, C49H80N18Ag3Cl5 (1422.15): calcd C 41.38, H 5.67, N 17.73;
found C 41.28, H 5.41, N 17.75. HR-ESI+ (MeCN + MeOH): m/z =
monochromated, λ = 0.71073 Å), T = 100 K, θ
Reflns measd 34711, indep 17371, Rint = 0.0219. Final R indices [I >
2σ(I)]: R1 = 0.0515, wR2 = 0.1412.
2.11−28.69°.
range
Synthesis of [3(AgI)3]n. Compound 3 (58.3 mg, 0.0643 mmol)
was dissolved in 7 mL of toluene, and AgI (45.3 mg, 0.1928 mmol, 3
equiv) was added to this solution. The suspension was stirred in
boiling toluene for a period of 9 h. The pale-yellow precipitate was
separated by filtration, washed first three times with 3 mL portions of
toluene and then three times with 3 mL portions of Et2O, and finally
dried to give 73.5 mg of product (0.0456 mmol, 71%). Small crystals
9893
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
were grown by slow diffusion of Et2O into a CH2Cl2 solution.
C48H78N18Ag3I3 (1611.57): calcd C 35.77, H 4.88, N 15.64; found C
35.63, H 4.76, N 15.89. HR-ESI+: m/z = 1143.48973 [3(AgI) + H]+,
14.8%, calcd 1143.48363; 907.67443 [3 + H]+, 100.0%, calcd
907.67296. FAB+: m/z = 907.7 [3 + H]+, 49.1%, 807.0 [3(AgI)3 +
CH2Cl2; details on the TD-DFT calculations of 3(CuI2)3; CV
curves for 3(CuCl2)3 and 3(CuI2)3; χT−T curves as derived
from SQUID measurements for 3(CuCl2)3, 3(CuBr2)3,
3(CuI2)3, and 3(CoCl2)3; molecular structure, UV/vis
spectrum, and magnetization curves of 3(CoCl2)3; structure
of [3(AgBr)3]n; visualization of the voids in [3(AgI)3]n using
the program CrystalExplorer; details of the quantum chemical
calculations (electronic energies; zero-point energy corrections;
and x, y, and z coordinates for the optimized structures); and
X-ray crystallographic data in cif format. This material is
available free of charge via the Internet at http://pubs.acs.org.
CCDC entries 999187 (3·6HCl), 999188 (4·3HCl), 999189
[3(CuCN)3], 999190 [3(CuCl)3], 999191 [3(CuBr)3], 999192
[3(CuCl2)3], 999193 [3(CuI2)3], 999194 [3(NiCl2)3], 999195
[3(AgCN)3], 999196 [3(AgCl)3], 999197 [3(AgBr)3], and
999198 [3(AgI)3] contain the supplementary crystallographic
data for this work.
1
2H]2+, 32.1%, 750.6 [3(AgI)2 + I]2+, 30.2%. H NMR (200 MHz,
CD2Cl2): δ = 7.27 (s, 6H, Harom), 2.88 (s, 72H, CH3) ppm. 13C{1H}
NMR (150.90 MHz, CD2Cl2), HSQC, HMBC: δ = 163.71 (NC),
112.43 (CHarom), 40.82 (CH3) ppm. IR (CsI): ν = 3059 (w), 2999
(w), 2929 (w), 2883 (m), 2794 (w), 1622 (m, shoulder), 1564 (s,
shoulder), 1534 (s), 1487 (s), 1466(m), 1420 (s), 1388 (s), 1334 (w),
1302 (m), 1273 (w), 1233 (m), 1223 (m), 1189 (w), 1152 (s), 1106
(w), 1063 (m), 1033 (s), 997 (s), 920 (w), 869 (m), 841 (m), 805
(m), 796 (m), 752 (m), 736 (m), 726 (m), 704 (m), 667 (w), 631
(w), 599 (m) cm−1. UV/vis (CH2Cl2, c = 1.7498 × 10−5 mol/L, d =
0.5 cm): λ (ε, M−1 cm−1) = ca. 395 (38536, shoulder), 374 (51655),
336 (87982), 308 (54687), ca. 233 (64693, shoulder) nm. Crystal data
for [3(AgI)3]n, C48H78Ag3I3N18: Mr = 1611.59, 0.25 × 0.20 × 0.20
mm3, monoclinic, space group I2/m, a = 14.89120(12) Å, b =
29.8096(3) Å, c = 18.5920(2) Å, β = 94.5880(9)°, V = 8226.56(15)
Å3, Z = 4, dcalc = 1.301 Mg·m−3, Mo Kα radiation (graphite-
monochromated, λ = 0.71073 Å), T = 100 K, θrange 3.23−25.12°.
Reflns measd 116949, indep 7491, Rint = 0.0390. Final R indices [I >
2σ(I)]: R1 = 0.0765, wR2 = 0.2310.
AUTHOR INFORMATION
Corresponding Author
*Fax: +49-6221-545707. E-mail: hans-jorg.himmel@aci.uni-
■
heidelberg.de.
Crystal Structure Determination. Suitable crystals were
removed directly from the mother liquor, immersed in perfluorinated
polyether oil, and fixed on top of a glass capillary. Measurements were
made with a Nonius-Kappa CCD diffractometer with a low-
temperature unit using graphite-monochromated Mo Kα radiation.
The temperature was set to 100 K. The data collected were processed
using the standard Nonius software.52 All calculations were performed
using the SHELXT-PLUS software package. Structures were solved by
direct methods with the SHELXS-97 program and refined with the
SHELXL-97 program.53,54 In the case of [3(AgI)3]n, attempts to find a
satisfactory model for the solvent present in the structure were
unsuccessful, and a diffuse solvent correction was applied to the
intensity data of this compound using the SQUEEZE program of the
Platon software suite.55 Graphical handing of the structural data during
solution and refinement was performed with XPMA.56 Atomic
coordinates and anisotropic thermal parameters of non-hydrogen
atoms were refined by full-matrix least-squares calculations.
Details of the Quantum Chemical Calculations. DFT and TD-
DFT calculations were carried out with the Gaussian 0957 and
TURBOMOLE58 program packages. For all calculations, the B3LYP
functional59 was used, in combination with the 6-311G** basis set60
for 3, 32+, 34+, and 36+ and with the def2-SV(P) basis set61 for 3(CuI)3
and 3(CuI2)3. Calculations with B3LYP/6-311G** and B3LYP/def2-
SV(P) on the free ligand 3 gave very similar structures. The smaller
def2-SV(P) basis set was chosen for the complexes with relatively large
numbers of atoms leading to large computational times. Where
applicable, the calculations on 3(CuI)3 and 3(CuI2)3 were carried out
with the resolution of the identity approximation for the coulomb
integrals (RI-J)58c using the appropriate auxiliary SV(P) basis sets.62
Additionally, the multipole accelerated resolution of the identity
approximation (MARI-J) was employed.58d For the optimized
structures of 3(CuI)3 and 3(CuI2)3, TD-DFT calculations58e were
performed. The excitation energies for 100 states were determined.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors gratefully acknowledge continuous financial
support by the Deutsche Forschungsgemeinschaft (DFG).
■
REFERENCES
■
(1) Wettach, H.; Jester, S. S.; Colsmann, A.; Lemmer, U.; Rehmann,
N.; Meerholz, K.; Hoger, S. Synth. Met. 2010, 160, 691−700.
(2) Thiessen, A.; Wettach, H.; Meerholz, K.; Neese, F.; Hoger, S.;
̈
̈
Hertel, D. Org. Electron. 2012, 13, 71−83.
(3) Lang, N. D.; Solomon, P. M. ACS Nano 2009, 3, 1437−1440.
(4) Adam, D.; Schuhmacher, P.; Simmerer, J.; Haussling, L.;
̈
Siemensmeyer, K.; Etzbach, K. H.; Ringsdorf, H.; Haarer, D. Nature
1994, 371, 141−143.
(5) Gonell, S.; Poyatos, M.; Peris, E. Angew. Chem., Int. Ed. 2013, 52,
7009−7013.
(6) Li, K.; Xu, Z.; Xu, H.; Carroll, P. J.; Fettinger, J. C. Inorg. Chem.
2006, 45, 1032−1037.
(7) Huang, G.; Xu, H.; Zhou, X.-P.; Xu, Z.; Li, K.; Zeller, M.; Hunter,
A. D. Cryst. Growth Des. 2007, 7, 2542−2547.
(8) (a) Hmadeh, M.; Lu, Z.; Liu, Z.; Gandara, F.; Furukawa, H.; Wan,
́
S.; Augustyn, V.; Chang, R.; Liao, L.; Zhou, F.; Perre, E.; Ozolins, V.;
Suenaga, K.; Duan, X.; Dunn, B.; Yamamto, Y.; Terasaki, O.; Yaghi, O.
M. Chem. Mater. 2012, 24, 3511−3513. (b) Cui, J.; Xu, Z. Chem.
Commun. 2014, 50, 3986−3988.
(9) (a) Edelmann, F. T. Adv. Organomet. Chem. 2008, 57, 183−352.
(b) Edelmann, F. T. Coord. Chem. Rev. 1994, 137, 403−481.
(c) Barker, J.; Kilner, M. Coord. Chem. Rev. 1994, 133, 219−300.
(d) Bailey, P. J.; Price, S. Coord. Chem. Rev. 2001, 214, 91−141.
(e) Herres-Pawlis, S. Nachr. Chem. 2009, 57, 20−23. (f) Coles, M. P.
Dalton Trans. 2006, 985−1001. (g) Coles, M. P. Chem. Commun.
2009, 3659−3676. (h) Himmel, H.-J. In Modeling of Molecular
Properties; Comba, P., Ed.; Wiley-VCH: Weinheim, Germany, 2011;
Chapter 26, pp 425−453. (i) Jones, C. Coord. Chem. Rev. 2010, 254,
1273−1289. (j) Edelmann, F. T. Adv. Organomet. Chem. 2013, 61, 55−
374.
ASSOCIATED CONTENT
■
S
* Supporting Information
Molecular structure of 4; CV curves for 1 and 4 in CH2Cl2
solution; effects of the guanidino groups in 3 and 4 on the
HOMO and LUMO energies of triphenylene; UV/vis spectra
of 3(CuCN)3 and 3(AgCN)3 in CH2Cl2; details on the TD-
DFT calculations of 3(CuI)3; molecular structure of 3(CuI2)3;
comparison of the UV/vis spectra of 3(CuCl2)3, 3(CuBr2)3,
and 3(CuI2)3 (in CH2Cl2 solution); validity check of the
Lambert−Beer law for the complex 3(CuI2)3 dissolved in
(10) (a) Cotton, F. A.; Matonic, J. H.; Murillo, C. A. J. Am. Chem.
Soc. 1997, 119, 7889−7890. (b) Bear, J. L.; Li, Y.; Han, B.; Kadish, K.
́
M. Inorg. Chem. 1996, 35, 1395−1398. (c) Clerac, R.; Cotton, F. A.;
Daniels, L. M.; Donahue, J. P.; Murillo, C. A.; Timmons, D. J. Inorg.
Chem. 2000, 39, 2581−2584. (d) Cotton, F. A.; Murillo, C. A.; Wang,
X.; Wilkinson, C. C. Inorg. Chim. Acta 2003, 351, 191−200.
9894
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
(11) Day, B. M.; Mansfield, N. E.; Coles, M. P.; Hitchcock, P. B.
Chem. Commun. 2011, 47, 4995−4997.
(33) Chen, L.; Kim, J.; Ishizuka, T.; Honsho, Y.; Saeki, A.; Seki, S.;
Ihee, H.; Jiang, D. J. Am. Chem. Soc. 2009, 131, 7287−7292.
(34) Emeljanenko, D.; Peters, A.; Vitske, V.; Kaifer, E.; Himmel, H.-J.
Eur. J. Inorg. Chem. 2010, 4783−4789.
(12) (a) Green, S. P.; Jones, C.; Stasch, A. Science 2007, 318, 1754−
1757. (b) Westerhausen, M. Angew. Chem., Int. Ed. 2008, 47, 2185−
2187.
(35) Zacharias, P.; Gather, M. C.; Rojahn, M.; Nuyken, O.; Meerholz,
K. Angew. Chem., Int. Ed. 2007, 46, 4388−4392.
(13) (a) Borner, J.; Florke, U.; Huber, K.; Doring, A.; Kuckling, D.;
̈
̈
̈
(36) (a) Hopff, H.; Wick, A. K. Helv. Chim. Acta 1961, 44, 19−24.
(b) Hopff, H.; Wick, A. K. Helv. Chim. Acta 1961, 44, 380−386.
Herres-Pawlis, S. Chem.−Eur. J. 2009, 15, 2362−2376. (b) Borner, J.;
̈
dos Santos Vieira, I.; Pawlis, A.; Doering, A.; Kuckling, D.; Herres-
(c) Trætteberg, M.; Bakken, P.; Hopf, H.; Hopfner, T. J. Mol. Struct.
Pawlis, S. Chem.−Eur. J. 2011, 17, 4507−4512. (c) Borner, J.; dos
̈
̈
1998, 445, 99−105. (d) Iyoda, M.; Tanaka, S.; Otani, H.; Nose, M.;
Santos Vieira, I.; Jones, M. D.; Doring, A.; Kuckling, D.; Florke, U.;
̈
̈
Oda, M. J. Am. Chem. Soc. 1988, 110, 8494−8500. (e) Stanger, A.;
Herres-Pawlis, S. Eur. J. Inorg. Chem. 2011, 4441−4456. (d) dos Santos
Vieira, I.; Herres-Pawlis, S. Eur. J. Inorg. Chem. 2012, 765−774. (e) dos
Santos Vieira, I.; Herres-Pawlis, S. Z. Naturforsch. B 2012, 67, 320−
330.
Ashkenazi, N.; Boese, R.; Blaser, D.; Stellberg, P. Chem.−Eur. J. 1997,
̈
3, 208−211. (f) Shinozaki, S.; Hamura, T.; Ibusuki, Y.; Fujii, K.;
Uekusa, H.; Suzuki, K. Angew. Chem., Int. Ed. 2010, 49, 3026−3029.
(37) Roquette, P.; Maronna, A.; Peters, A.; Kaifer, E.; Himmel, H.-J.;
Hauf, Ch.; Herz, V.; Scheidt, E.-W.; Scherer, W. Chem.−Eur. J. 2010,
16, 1336−1350.
(14) (a) Wurtele, C.; Gaoutchenova, E.; Harms, K.; Holthausen, M.
̈
C.; Sundermeyer, J.; Schindler, S. Angew. Chem., Int. Ed. 2006, 45,
3867−3869. (b) Maiti, D.; Lee, D.-H.; Gaoutchenova, K.; Wurtele, C.;
̈
(38) Jesser, A.; Rohrmuller, M.; Schmidt, W. G.; Herres-Pawlis, S. J.
̈
Holthausen, M. C.; Sarjeant, A. A. N.; Sundermeyer, J.; Schindler, S.;
Karlin, K. D. Angew. Chem., Int. Ed. 2007, 47, 82−85. (c) Lanci, M. P.;
Smirnov, V. V.; Cramer, C. J.; Gauchenova, E. V.; Sundermeyer, J.;
Roth, J. P. J. Am. Chem. Soc. 2007, 129, 14697−14709. (d) Maiti, D.;
Comput. Chem. 2014, 35, 1−17.
(39) Burdett, J. K.; Sevov, S. J. Am. Chem. Soc. 1995, 117, 12788−
12792.
(40) Bowmaker, G. A.; Di Nicola, C.; Pettinari, C.; Skelton, B. W.;
Somers, N.; White, A. H. Dalton Trans. 2011, 40, 5102−5115.
(41) Melzer, M. M.; Mossin, S.; Dai, X.; Bartell, A. M.; Kapoor, P.;
Meyer, K.; Warren, T. H. Angew. Chem., Int. Ed. 2010, 49, 904−907.
(42) (a) Ryde, U.; Olsson, M. H. M.; Ross, B. O.; Borin, A. C. Theor.
Chem. Acc. 2001, 105, 452−462. (b) Bryantsev, V. S.; Diallo, M. S.;
Goodard, W. A., III. J. Phys. Chem. A 2009, 113, 9559−9567.
Lee, D.-H.; Gaoutchenova, K.; Wurtele, C.; Holthausen, M. C.;
̈
Sarjeant, A. A. N.; Sundermeyer, J.; Schindler, S.; Karlin, K. D. Angew.
Chem., Int. Ed. 2008, 47, 82−85. (e) Peterson, R. L.; Ginsbach, J. W.;
Cowley, R. E.; Qayyum, M. F.; Himes, R. A.; Siegler, M. A.; Moore, C.
D.; Hedman, B.; Hodgson, K. O.; Fukuzumi, S.; Solomon, E. I.; Karlin,
K. D. J. Am. Chem. Soc. 2013, 135, 16454−16467. (f) Saracini, C.;
Liakos, D. G.; Zapata Rivera, J. E.; Neese, F.; Meyer, G. J.; Karlin, K.
D. J. Am. Chem. Soc. 2014, 136, 1260−1263.
(43) Hoffmann, A.; Binder, S.; Jesser, A.; Haase, R.; Florke, U.;
̈
Gnida, M.; Stagni, M. S.; Meyer-Klaucke, W.; Lebsanft, B.; Grunig, L.
̈
(15) Himmel, H.-J. Z. Anorg. Allg. Chem. 2013, 639, 1940−1952.
(16) Peters, A.; Kaifer, E.; Himmel, H.-J. Eur. J. Org. Chem. 2008,
5907−5914.
E.; Schneider, S.; Hashemi, M.; Goos, A.; Wetzel, A.; Rubhausen, M.;
̈
Herres-Pawlis. Angew. Chem., Int. Ed. 2014, 53, 299−304.
(44) Xiao, Z.; Donnelly, P. S.; Zimmermann, M.; Wedd, A. G. Inorg.
Chem. 2008, 47, 4338−4347.
(17) Peters, A.; Trumm, C.; Reinmuth, M.; Emeljanenko, D.; Kaifer,
E.; Himmel, H.-J. Eur. J. Inorg. Chem. 2009, 3791−3800.
(45) Roquette, P.; Maronna, A.; Reinmuth, M.; Kaifer, E.; Enders,
M.; Himmel, H.-J. Inorg. Chem. 2011, 50, 1942−1955.
(18) Trumm, C.; Hubner, O.; Kaifer, E.; Himmel, H.-J. Eur. J. Inorg.
̈
Chem. 2010, 3102−3108.
(46) Maronna, A.; Hubner, O.; Enders, M.; Kaifer, E.; Himmel, H.-J.
̈
(19) Emeljanenko, D.; Peters, A.; Wagner, N.; Beck, J.; Kaifer, E.;
Himmel, H.-J. Eur. J. Inorg. Chem. 2010, 1839−1846.
(20) Trumm, C.; Stang, S.; Eberle, B.; Kaifer, E.; Wagner, N.; Beck,
Chem.−Eur. J. 2013, 19, 8958−8977.
(47) Glaser, T.; Heidemeier, M.; Strautmann, J. B. H.; Bogge, H.;
̈
Stammler, A.; Krickemeyer, E.; Huenerbein, R.; Grimme, S.; Bothe, E.;
Bill, E. Chem.−Eur. J. 2007, 13, 9191−9206.
J.; Bredow, T.; Meyerbroker, N.; Zharnikov, M.; Hubner, O.; Himmel,
̈
̈
H.-J. Eur. J. Inorg. Chem. 2012, 3156−3167.
(48) Bencini, A.; Gatteschi, D.; Totti, F.; McCleverty, J. A.; Ward, M.
D. J. Phys. Chem. A 1998, 102, 10545−10551.
(21) Emeljanenko, D.; Peters, A.; Vitske, V.; Kaifer, E.; Himmel, H.-J.
Eur. J. Inorg. Chem. 2010, 4783−4789.
(49) (a) Roquette, P.; Konig, C.; Hubner, O.; Wagner, A.; Kaifer, E.;
̈
̈
(22) Stang, S.; Kaifer, E.; Himmel, H.-J. Chem.−Eur. J. 2014, 20,
5288−5297.
Enders, M.; Himmel, H.-J. Eur. J. Inorg. Chem. 2010, 4770−4782.
(b) Vitske, V.; Roquette, P.; Leingang, S.; Adam, C.; Kaifer, E.;
Wadepohl, H.; Himmel, H.-J. Eur. J. Inorg. Chem. 2011, 1593−1604.
(50) (a) Wolff, S. K.; Grimwood, D. J.; McKinnon, J. J.; Turner, M.
J.; Jayatilaka, D.; Spackman, M. A. CrystalExplorer 2.2; University of
Western Australia: Perth, Australia, 2010; available at http://
hirshfeldsurface.net. (b) Turner, M. J.; McKinnon, J. J.; Jayatilaka,
D.; Spackman, M. A. CrystEngComm 2011, 13, 1804−1813.
(23) Vitske, V.; Hermann, H.; Enders, M.; Kaifer, E.; Himmel, H.-J.
Chem.−Eur. J. 2012, 18, 14108−14116.
(24) Herrmann, H.; Ziesak, A.; Wild, U.; Leingang, S.; Schrempp, D.;
Wagner, N.; Beck, J.; Kaifer, E.; Wadepohl, H.; Himmel, H.-J.
ChemPhysChem 2014, 351−365.
(25) Peters, A.; Herrmann, H.; Magg, M.; Kaifer, E.; Himmel, H.-J.
Eur. J. Inorg. Chem. 2012, 1620−1631.
(26) Vitske, V.; Konig, C.; Kaifer, E.; Hubner, O.; Himmel, H.-J. Eur.
(51) Bill, E. julX v.1.4.1Simulation of Molecular Magnetic Data;
̈
̈
Max-Planck Institut for Bioinorganic Chemistry: Mulheim/Ruhr,
̈
J. Inorg. Chem. 2010, 115−126.
Germany, 2008.
(27) Kitagawa, S.; Masaoka, S. Coord. Chem. Rev. 2003, 246, 73−88.
(28) Patra, S.; Sarkar, B.; Ghumaan, S.; Fiedler, J.; Kaim, W.; Lahiri,
G. K. Dalton Trans. 2004, 754−758.
(52) DENZO-SMN, Data Processing Software; Nonius: Delft, The
Netherlands, 1998; available at http://www.nonius.com.
(53) (a) Sheldrick, G. M. SHELXS-97, Program for Crystal Structure
(29) Piglosiewicz, I. M.; Beckhaus, R.; Wittstock, G.; Saak, W.; Haase,
D. Inorg. Chem. 2007, 46, 7610−7620.
Solution; University of Gottingen: Gottingen, Germany, 1997; available
̈
̈
(30) Herrera, J.-M.; Ward, M. D.; Adams, H.; Pope, S. J. A.; Faulkner,
S. Chem. Commun. 2006, 1851−1853.
at http://shelx.uni-ac.gwdg.de/SHELX/index.html. (b) Sheldrick, G.
M. SHELXL-97, Program for Crystal Structure Refinement; University of
(31) Klivansky, L. M.; Hanifi, D.; Koshkakaryan, G.; Holycross, D.
R.; Gorski, E. K.; Wu, Q.; Chai, M.; Liu, Y. Chem. Sci. 2012, 3, 2009−
2014.
Gottingen: Gottingen, Germany, 1997; available at http://shelx.uni-ac.
̈
̈
gwdg.de/SHELX/index.html.
(54) International Tables for X-ray Crystallography; Kynoch Press:
(32) (a) Breslow, R.; Jaun, B.; Kluttz, R. Q.; Xia, C.-Z. Tetrahedron
Birmingham, U.K., 1974; Vol. 4.
1982, 38, 863−867. (b) Yatabe, T.; Harbison, M. A.; Brand, J. D.;
(55) (a) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13. (b) Spek, A.
L. PLATON, A Multipurpose Crystallographic Tool; Utrecht University:
Utrecht, The Netherlands, 2005.
Wagner, M.; Mullen, K.; Samori, P.; Rabe, J. P. J. Mater. Chem. 2000,
̈
10, 1519−1525.
9895
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896

Inorganic Chemistry
Article
(56) Zsolnai, L.; Huttner, G. XPMA; University of Heidelberg:
Heidelberg, Germany, 1994; available at http://www.uni-heidelberg.
de/institute/fak12/AC/huttner/soft-ware/software.html.
(57) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, J. Bloino, A. F.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, E. Brothers, J. J.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian Inc.: Wallingford, CT, 2009.
(58) (a) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Chem.
̈
̈
̈
Phys. Lett. 1989, 162, 165−169. (b) Treutler, O.; Ahlrichs, R. J. Chem.
̈
Phys. 1995, 102, 346−354. (c) Eichkorn, K.; Treutler, O.; Ohm, H.;
Haser, M.; Ahlrichs, R. Chem. Phys. Lett. 1995, 242, 652−660.
̈
(d) Sierka, M.; Hogekamp, A.; Ahlrichs, R. J. Chem. Phys. 2003, 118,
9136−9148. (e) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett.
1996, 256, 454−464.
(59) (a) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J.
J. Chem. Phys. 1994, 98, 11623−11627. (b) Becke, A. D. J. Chem. Phys.
1993, 98, 5648−5652. (c) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B
1988, 37, 785−789.
(60) (a) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650−654. (b) McLean, A. D.; Chandler, G. S. J. Chem.
Phys. 1980, 72, 5639−5648.
(61) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(62) Weigend, F. Phys. Chem. Chem. Phys. 2006, 8, 1057−1065.
9896
dx.doi.org/10.1021/ic501482u | Inorg. Chem. 2014, 53, 9876−9896