Tuning the electronic coupling in Mo2-Mo2 systems by variation of the coordinating atoms of the bridging ligands

Article abstract of DOI:10.1039/c4dt00786g

Three novel [Mo₂]-bridge-[Mo₂] complexes were synthesized by a convergent assembling reaction of the dimetal precursor Mo₂(DAniF)₃(O₂CCH₃) (DAniF = N,N′-di(p-anisyl)formamidinate) with the bridging ligands terephthalamidine, terephthalamide and dithioterephthalamide. The structures of these compounds, [Mo₂(DAniF)₃]₂[μ-1,4-{C(E)NH}₂-C₆H₄] (E = NH (1), O (2) or S (3)), were determined, either by X-ray crystallography or ¹H NMR spectroscopy, to be the analogues of the terephthalate bridged dimolybdenum dimer. These compounds are structurally and electronically closely related by having the same structural skeleton and similar bonding parameters, which allowed us to analyze the differences between N, O and S atoms on the bridging ligand in promoting electronic interaction between the two [Mo₂] units. In the electronic spectra, the metal to ligand charge transfer absorption bands, attributed to the HOMO (dδ) → LUMO (pπ) transition, was red shifted as the variable atoms change from N to O to S. The mixed-valence species 1⁺, 2⁺ and 3⁺, generated by one-electron oxidation of the neutral precursors and measured in situ, exhibited characteristic intervalence absorption bands, for which the energy and half-height bandwidth decreased from 1⁺ to 3⁺. Therefore, in comparison to O atoms, S atoms are capable of enhancing the electronic coupling between the two [Mo₂] units, and the incorporation of N atoms to the bridging ligands slightly diminished the metal-metal interaction. The molecular structures and spectroscopic properties of these compounds were simulated by theoretical calculations at DFT level on the simplified models, which gave results consistent with the experimental observations. This journal is

Full text of DOI:10.1039/c4dt00786g

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Tuning the electronic coupling in Mo2–Mo2
systems by variation of the coordinating atoms of
the bridging ligands†
Cite this: Dalton Trans., 2014, 43,
14756
Yao Shu,a Hao Lei,a Ying Ning Tan,a Miao Meng,a Xiao Chun Zhangb and
Chun Y. Liu*a,b
Three novel [Mo2]–bridge–[Mo2] complexes were synthesized by a convergent assembling reaction of the
dimetal precursor Mo2(DAniF)3(O2CCH3) (DAniF = N,N’-di(p-anisyl)formamidinate) with the bridging
ligands terephthalamidine, terephthalamide and dithioterephthalamide. The structures of these com-
pounds, [Mo2(DAniF)3]2[μ-1,4-{C(E)NH}2-C6H4] (E = NH (1), O (2) or S (3)), were determined, either by
X-ray crystallography or 1H NMR spectroscopy, to be the analogues of the terephthalate bridged dimolyb-
denum dimer. These compounds are structurally and electronically closely related by having the same
structural skeleton and similar bonding parameters, which allowed us to analyze the differences between
N, O and S atoms on the bridging ligand in promoting electronic interaction between the two [Mo2] units.
In the electronic spectra, the metal to ligand charge transfer absorption bands, attributed to the HOMO
(dδ) → LUMO (pπ*) transition, was red shifted as the variable atoms change from N to O to S. The mixed-
valence species 1+, 2+ and 3+, generated by one-electron oxidation of the neutral precursors and
measured in situ, exhibited characteristic intervalence absorption bands, for which the energy and half-
height bandwidth decreased from 1+ to 3+. Therefore, in comparison to O atoms, S atoms are capable of
enhancing the electronic coupling between the two [Mo2] units, and the incorporation of N atoms to the
bridging ligands slightly diminished the metal–metal interaction. The molecular structures and spectro-
scopic properties of these compounds were simulated by theoretical calculations at DFT level on the
simplified models, which gave results consistent with the experimental observations.
Received 16th March 2014,
Accepted 5th August 2014
DOI: 10.1039/c4dt00786g
www.rsc.org/dalton
model compounds with varying structural and electronic com-
positions of the D (or A) and B units have been synthesized
Introduction
To a large extent, our understanding of electronic coupling and investigated. To address their electronic configuration,
(EC) and intramolecular electron transfer (ET) are obtained redox and optically active molecular entities are favorable can-
from the studies on simple models that are generally con- didates as electron donor and acceptor for the D–B–A mole-
structed with three units, electron donor (D), bridge (B) and cules. With this prerequisite, majority of the studied
electron acceptor (A), namely, D–B–A assembly. In the theore- compounds are those with dinuclear d6/d5 metal centers, e.g.,
tical aspect, the classical two-state model based on a D–B–A Ru–Ru, Os–Os, Ru–Os and Fc–Fc systems.9–12 Thereafter,
molecular entity has achieved a great success in interpreting bridged triruthenium clusters, Ru3–Ru3, have been studied
and predicting the optical behaviors,1,2 mixed-valence pro- in terms of electronic coupling and electron transfer.13–16
perties,3,4 electron transfer kinetics and mechanism of the Because of their redox and charge transfer properties, specially
experimental systems.5,6 Starting with the Creutz–Taube designed organic radicals are also made use of in this research
complex, [(NH3)5Ru(pyrazine)Ru(NH3)5]5+,7,8 thousands of practice since the 1990s.17–19 Slightly after the aforementioned
work, covalently bonded dimetal units were employed as elec-
tron donor and acceptor for the study in this field mainly by
Cotton20–22 and Chisholm.23,24 The important and unique
feature for the dimetal based D–B–A models is that the elec-
tronic configuration of the donor and acceptor are well
defined. For example, for quadruply bonded [M2]–bridge–[M2]
complexes, only the δ electrons are involved in the charge or
electron transfer process, which greatly facilitates the optical
aDepartment of Chemistry, Jinan University, 601 Huang-Pu Avenue West, Guangzhou
510632, China. E-mail: tcyliu@jnu.edu.cn
bDepartment of Chemistry, Tongji University, 1239 Siping Road, Shanghai 200092,
China
†Electronic supplementary information (ESI) available. CCDC 991923 and
991924. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c4dt00786g
14756 | Dalton Trans., 2014, 43, 14756–14765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 1 Three new Mo2–Mo2 complexes (1–3) with N-containing bridging ligands and related terephthalate derivatives (I–III) ([Mo2] =
[Mo2(DAniF)3]+).
analysis. It is believed that the great variety of redox centers O (2) or S (3)) through direct assembly of Mo2(DAniF)3-
and spacer moieties that may be combined in D–B–A systems (O2CCH3) with the corresponding tetradentate ligands. The
1
will help the ongoing refinement of the ET theories.
structures of all the three compounds were determined by H
During the course of investigation on the Mo2–Mo2 NMR spectra and two of them (2 and 3) were characterized by
systems, it is discovered that replacing the O atoms in the crystallographic methods also. In order to evaluate the influ-
parent terephthalate ligand framework ([p-O2C-C6H4-CO2]2−
)
ence of various donor atoms with respect to electronic inter-
with S atoms largely improves the electronic communication actions, their electrochemical and spectroscopic behaviors
between the two Mo2 units, possibly because of the enhanced were compared with those analogues, which are structurally
d(δ)–p(π) orbital interaction.25–28 Obviously, it would be of and electronically closely related (I, II and III in
great interest to further evaluate the electronic effect of intro- Scheme 1).26–28 On this basis, the correlations of the energy of
ducing other heteroatoms into the carboxylate groups of the the metal to ligand charge transfer band (λmax) in the spec-
bridging ligands. For example, full or partial substitution of trum and the potential separation (ΔE1/2) in the cyclic voltam-
the O atoms of the carboxylate ligand with N atoms would mogram (CV) with the extent of metal–metal interaction are
afford amidinate or amidate ligands, respectively. Although discussed. DFT calculations were performed on the simplified
some diamidate bridged dimolybdenum dimers have been model compounds, which further elucidated the electronic
studied, the difference between O and N atoms in affecting effects of coordinating atoms on the EC interactions. From the
electronic coupling is not clarified because of their very close magnetic and optical behaviors of the singly oxidized species
redox and optical behaviors. On the other hand, amidinate 1+, 2+ and 3+, the mixed-valence properties of the system were
and amidate groups are much stronger Lewis bases than the preliminarily evaluated.
carboxylate; thus, it is anticipated that the resulting Mo2–Mo2
complexes with these N-containing bridges would be thermo-
dynamically more stable. Amidinate ligands (especially the
sterically bulky ones) have been widely used in the synthesis
and isolation of various uncommon inorganic/organometallic
Results and discussion
Molecular design and synthesis
structural motifs, including the first and to date only example
of Mo–Mo quintuple bonds.29,30 However, to date, dimers of
dimers synthesized with an amidinate bridging ligand are
scarce because of the synthetic obstacles for ligands and com-
plexes of this type,31 although the most commonly used ancil-
lary ligands in such M2–M2 systems are indeed formamidinate
ligands. In contrast, N-aryl amidate ligands have been
employed to link [Mo2] units to form molecular triangles and
squares,32 and related dimers of dimers have been constructed
using either N-aryl33 or N-alkyl34 amidates. Tetrarhenate
complex [NBu4]2[{Re2Cl6·C3H7NO}2{1,4-(C(O)NH)2C6H4}] is the
only example of M2–M2 compounds having a terephthalami-
date linker,35 which was prepared by hydrolysis of 1,4-dicyano-
benzene in the presence of a dirhenium complex.
As mentioned earlier, our current investigation focused mainly
on the electronic effect of the substitution of N donor atoms in
the bridging ligands of Mo2–Mo2 complexes. Thus, we chose
three new dimers of dimers as our synthetic targets (Scheme 1),
which contain amidinate, amidate and thioamidate bridging
ligands. It is noteworthy that the three new compounds have
similar molecular scaffolds, and the only variation is the type
of the donor atoms of the bridging ligand. It was expected that
such subtle compositional and structural differences would
facilitate the direct comparison of the electron coupling inter-
action within the complex series, including previously reported
terephthalate (I) and thioterephthalate (II and III)
derivatives.26–28 Bridging ligands terephthalamide36 and
dithioterephthalamide37 were prepared in good yields by fol-
lowing the literature procedures. For the preparation of
terephthalamidine dihydrochloride modification was made to
the published methods.38 These new Mo2–Mo2 complexes were
Herein, we report the synthesis and isolation of three di-
molybdenum dimers of dimers [Mo2(DAniF)3]2[μ-1,4-{C(E)NH}2-
C6H4] (DAniF = N,N′-di(p-anisyl)-formamidinate; E = NH (1),
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14756–14765 | 14757

View Article Online
Paper
Dalton Transactions
synthesized by mixing two equiv. of dimetal precursor Mo2-
(DAniF)3(O2CCH3) with one equiv. of bridging ligand in the
presence of NaOC2H5. By following this known synthetic
protocol,39,40 crystalline products of [Mo2(DAniF)3]2[μ-1,4-{C(E)-
NH}2-C6H4] (E = NH (1), O (2) or S (3)) were obtained in high
yields after standard workup and recrystallization procedures.
Compound 1 represents the first example of dimolybdenum
dimers bridged by an amidinate ligand, thus affording a com-
pletely N-coordinated analogue.
Table 1 Crystallographic data and collection parameters for 2·6C4H8O
and 3·5.6C4H8O·2C2H5OH
2·6C4H8O
122H144Mo4N14O20
2510.33
P21/c
19.8102(3)
18.2488(2)
17.1298(3)
90
107.273(2)
90
3·5.6C4H8O·2C2H5OH
Formula
fw
Space group
a (Å)
b (Å)
c (Å)
C
C124.4H152.8Mo4N14O19.6S2
2605.76
ˉ
P1
11.5410(4)
11.7590(4)
24.0969(9)
76.162(3)
86.324(3)
72.438(3)
3027.18(19)
1
α (°)
β (°)
Molecular structures
γ (°)
V (Å3)
Z
5913.35(15)
2
1
The molecular structures of 1–3 were determined by H NMR
spectroscopy. In each of the 1H NMR spectra for 1–3, only a
single resonance at 7.74, 8.04 and 7.88 ppm, respectively, was
observed for the four protons on the bridging aromatic ring.
This is consistent with the highly symmetric structures of ami-
dinate bridged complex 1 in solution, and also suggests that
the C–C bond connecting the amidate (2) or thioamidate (3)
group and the phenylene ring can rotate freely at the measure-
ment conditions. In addition, singlet resonances for the
amidate NH protons in compounds 2 and 3 were detected in
T (K)
149(2)
1.401
3.989
0.0484
0.1363
173(2)
1.416
4.228
0.0505
dcalcd (g cm−3
)
μ (mm−1
)
a
R1
b
wR2
0.1271
a R1 = ∑||Fo| − |Fc||/∑|Fo|. b wR2 = [∑[w(Fo2 − Fc2)2]/∑[w(Fo2)2]]1/2
.
Table 2 Selected bond distance (Å) for 2 and 3, in comparison with
the downfield region (9.01 ppm for 2 and 10.29 ppm for 3) and those of I, II and IIIa
integrated into two protons. In contrast, no such resonance
was observed for the previously reported amidate bridged Re2–
2
3
I
II
III
Re2 compound and its absence was explained by the existence
of hydrogen bonding.35 Such comparison indicates that com-
plexes 2 and 3 show discrete structures in the solution and
there is no hydrogen bonding interactions, possibly due to the
steric bulk of the ancillary DAniF ligands on each Mo2 units.
As expected, the 1H NMR spectrum of 1 displayed a singlet
resonance for the four amidinate protons at 8.23 ppm. This
signal is shifted toward upfield relative to that of the ligand
precursor terephthalamidine dihydrochloride (9.00 ppm).
Many efforts have been made to grow single crystals of 1 for
X-ray characterization. Diffusion of ethanol into tetrahydro-
furan (THF) solution of 1 afforded orange yellow coloured
plates. Unfortunately, crystallographic data collection failed
because the crystal was very fragile and extremely air sensitive.
For 2 and 3, the solid-state structures were further confirmed
Mo(1)–Mo(2) 2.0892(5) 2.0943(5)
2.090(1) 2.1051(6) 2.103(1)
Mo(1)–N(1)
Mo(1)–N(3)
Mo(1)–N(5)
Mo(2)–N(2)
Mo(2)–N(4)
Mo(2)–N(6)
Mo(1)–O(7)
Mo(2)–O(8)
Mo(1)–S(1)
Mo(2)–S(2)
Mo(1)–N(7)
Mo(2)–N(8)
Mo2⋯Mo2
2.172(4)
2.152(3)
2.164(3)
2.138(4)
2.139(3)
2.141(3)
2.155(4)
2.150(4)
2.160(4)
2.166(4)
2.188(4)
2.150(4)
2.155(5) 2.133(4)
2.123(5) 2.143(4)
2.152(5) 2.137(4)
2.137(5) 2.161(4)
2.119(5) 2.174(4)
2.137(5) 2.142(4)
2.144(4) 2.112(4)
2.122(4)
2.149(5)
2.153(5)
2.146(5)
2.141(5)
2.164(5)
2.148(5)
2.119(3)
2.4692(13)
2.442(2)
2.445(2)
2.469(1)
2.148(3)
11.36
2.144(4)
11.76
11.24
11.67
12.24
a Data cited from ref. 27.
disorder were detected, even in solution. Moreover, each com-
pound showed typical Mo–Mo quadruple bond lengths,
by X-ray crystallography. Compound 2 crystallizes in the mono-
ˉ
clinic space group P21/c and 3 in the triclinic group P1. The 2.0892(5) Å for 2 and 2.0943(5) Å for 3. The Mo(1)–N(7)
(2.148(3) Å) and Mo(2)–O(8) (2.119(3) Å) distances in 2 are com-
crystallographic data and collection parameters are presented
in Table 1 and selected bond distances and angles are listed in
parable to the corresponding values in the analogues contain-
Table 2. Generally, both complexes show the same structural ing N,N′-diethylterephthalamidate (2.177(5) and 2.112(4) Å)34
or N,N′-diphenylterephthalamidate (2.170(4) and 2.106(3) Å)33
skeleton (Fig. 1) and have similar bond lengths and angles in
bridging ligand. In the solid-state structure of 2, the central
comparison to the terephthalamide and dithioterephthala-
mide analogues.27,28 For instance, the Mo2⋯Mo2 separations p-phenylene group is almost coplanar with the Mo–Mo vectors.
However, for the diamidate analogues with N-substituents,
there is a considerable deviation between the phenylene
in the structures are determined to be 11.36 (2) and 11.76 Å
(3), respectively. These values fall in the range of 11.24–12.24 Å
for the carboxylate and thiocarboxylate derivatives.27 The bridge and the two Mo2 units. For instance, torsion angles of
34.45° and 36.21° were found for the N,N′-diphenyl and N,N′-
similar metal-to-metal distances exclude the differences in the
di(m-trifluoromethyl)phenylterephthalamidate
derivatives,
electrostatic interaction between them and thus would allow
the evaluation of the electronic coupling effect arising from respectively.33 Even more extreme distortion was reported for
the N,N′-diethylterephthalamidate analogue (50.14°).34 Appar-
alternation of the donor atoms. As shown in Fig. 1, for both 2
ently, the steric hindrance of the N-substituents is responsible
and 3, the same donor atoms on the bridging ligand are
located in the trans position and no cis-isomers and structure for the non co-planarity, which weakens the electronic coupling
14758 | Dalton Trans., 2014, 43, 14756–14765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 2 Differential pulse voltammograms (DPVs) (top) and cyclic vol-
tammograms (CVs) (bottom) for complexes 1–3.
Table 3 Electrochemical measurements and the parameters for the
comproportionation equilibrium
E1/2(1)
(mV)
E1/2(2)
(mV)
ΔE1/2
(mV)
ΔGc
Compound
Kc
(cm−1
)
1
2
3
178
327
398
225
468
500
272
411
508
325
584
700
80
96
115
100
116
200
23
42
88
49
91
−645
−774
−928
−807
−935
−1613
Fig. 1 X-ray crystal structures for 2 (top) and 3 (bottom). In the com-
pound formula, [Mo2] represents the dimetal building block,
[Mo2(DAniF)3]+. The hydrogen atoms have been omitted, except for
those on the bridging ligands.
Ia
IIb
IIIc
2405
a Data cited from ref. 21. b Data cited from ref. 27. c Data cited from ref. 26.
between the two [Mo2] units. For the thioamidate compound
(3), the steric repulsion between the sulfur atoms and the
neighboring phenyl protons also causes a N(8)–C(1)–C(2)–C(3)
torsion angle of 25.62°.
ca. 80 mV, indicating that compared to O and S atoms, nitro-
gen donor atoms on the bridging ligand has an effect of
diminishing the EC interaction. The thiolated analogue (3), on
the other hand, has the largest potential separation, i.e., ΔE1/2
= 115 mV, which is consistent with the previous studies.27,28
Compound 2, which has O/N mixed donors, gave a ΔE1/2 value
(96 mV) in the middle. A possible explanation on the coordi-
nating atom effect is that the softness of the atoms E increases
the covalent character of the M–E bonds, consequently, enhan-
cing the coupling interaction between the [Mo2] units in the
order of N < O < S.
The difference between O and N atoms in mediating elec-
tronic interaction is further manifested by comparing the ΔE1/2
values for the three closely related compounds I, 2 and 1,
which can be viewed to be the derivatives of stepwise replace-
ment of the O donor atoms of the terephthalate analogue by
N atoms. As indicated in Table 3, as a result of increasing the
N donors, the potential separation ΔE1/2 is lowered in order,
i.e., from 100 mV (for I) to 96 mV (for 2) to 80 mV (for 1).
These results demonstrate that among the three types of donor
Electrochemical studies
For complexes 1–3, the electrochemical process in CH2Cl2 were
recorded by cyclic voltammograms (CVs) and differential pulse
voltammograms (DPVs), as shown in Fig. 2, which involves two
redox events corresponding to the stepwise removal of one δ
electron from each of the two bridged [Mo2]4+ units. The
potential separation (ΔE1/2) between the two redox couples was
employed to probe the relative strength of the electronic com-
munication between the two redox sites. However, for these
compounds, the two redox cycles were not resolved in the CVs
because of the weak EC interaction. The small ΔE1/2 values
for 1–3 were determined from their DPVs by Richardson
and Taube’s methods,41 and the electrochemical data are sum-
marized in Table 3. For easier comparison, ΔE1/2 values for the
terephthalate (I),21 dithioterephthalate (II)27 and tetrathioter-
ephthalate (III)26 analogues from literatures are also included.
As it can be seen, compound 1 has the smallest ΔE1/2 value,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14756–14765 | 14759

View Article Online
Paper
Dalton Transactions
atoms we have investigated to date, the electronic effects for HOMO−1 are metal based orbitals. More specifically, the
enhancing the electronic communication in the Mo2–Mo2 HOMO mainly consists of the out-of-phase combination of Mo
systems are in the sequence of S > O > N. It should be noted δ orbitals (δ − δ), and the HOMO−1 mainly consists the corres-
that the ΔE1/2 value (96 mV) for 2 is compatible to those ponding in-phase combination (δ + δ). Our previous study
(100 mV) for the N,N′-diarylterephthalamidate complexes,33 demonstrated that in Mo2–Mo2 systems, HOMO–LUMO energy
but larger than that (85 mV) for the N,N′-diethylterephthalami- gap determines the electron transfer rate between the two
date analogue.34 For the N/O complexes, the N-substituted di- dimetal sites.27,28 A system with smaller HOMO–LUMO energy
amidate analogues are expected to have a larger ΔE1/2 because gap would show a higher electron transfer rate between the
of the electron donating property of the substituent groups. donor and acceptor sites. Hence, the calculated HOMO–LUMO
This explains that the N,N′-diarylterephthalamidate complexes energy gap for complexes 1′–3′ (Table 4) suggests that the
(100 mV) has the ΔE1/2 value slightly larger than that for 2. In enhancement of the electronic interaction by the coordinating
comparison, the diethyl substituted analogue has an excep- atoms follows the sequence of S > O > N, supporting our elec-
tionally small potential separation. This is likely due to the trochemical results.
large deviation (50.14°) between the two planes defined by the
The three compounds show distinct colors, i.e. orange
bridging phenyl ring and the associated dimetal chelating yellow (1), red (2) and purple (3), despite the subtle difference
ring, which diminishes the π conjugation through the bridge. on the bridging ligands. Consistent with the different colors
Therefore, for these structurally and electronically similar com- for these compounds, in the spectra, each of them exhibits a
plexes, comparison of the ΔE1/2 values indeed shows the rela- characteristic absorption band with similar intensity at 465
tive strength of the metal–metal electronic interaction, even (1), 490 (2) and 560 nm (3) (Fig. 4). This absorbance should be
though the difference caused by O/N atomic alternation is assigned to the metal to ligand charge transfer (MLCT). Simu-
small.
lations from the DFT calculations generated spectra having
For this dimetal system, the equilibrium constant (Kc) for similar variation tendency for the band energies (Fig. 4), con-
the comproportionation of the neutral [Mo2–Mo2] and the firming that the absorption corresponds to the HOMO →
doubly oxidized [Mo2–Mo2]2+ to the mixed-valent species LUMO transition. The transition energy for 2 is comparable to
[Mo2–Mo2]+ can be derived from the redox potential separation that found for the related N,N′-diarylterephthalamidate com-
ΔE1/2 by the expression Kc = exp(ΔE1/2/25.69).41 Accordingly, plexes (aryl = Ph (462 nm) and m-CF3C6H4 (473 nm)).33
the free energy change (ΔGc) for the comproportionation reac- However, the N,N′-diethylterephthalamidate analogue showed
tion can be determined electrochemically (Table 3). For the a considerably less intense absorption at shorter wavelength
same reason mentioned above, in this system, ΔGc can be (442 nm; ε = 3200 M−1 cm−1). This result is consistent with the
used to evaluate the extent of electronic delocalization. The electrochemical measurement, which shows small potential
similar Kc and ΔGc values for 2 and I are indicative of compar- separation (ΔE1/2 = 84 mV).34 The absorption bands of 1–3, as
able metal–metal interactions in these N- or O-donor well as I, II and III, are summarized in Table 5. In consistence
compounds.
with our previous finding,27,28 stepwise introduction of
S atoms to the bridging amidinate ligand shifts the MLCT
band to 560 nm for 3 from 465 nm for 1, and then further to
715 nm for III. In contrast, it appears that substitution of
O atoms with N atoms only slightly shifts the band in the
opposite direction, from 492 nm (I) to 490 nm (2) and 465 nm
(1). Therefore, it is evident that introducing N donor to the
bridging ligand increases the MLCT energy, while S donor
atoms on the bridging ligand lower the MLCT energy. These
observed trends are well stated by theoretical prediction.
Through our studies on the dimers of dimers with conju-
gated bridging ligands, we found that the MLCT energy is cor-
related with the extent of electron delocalization and for the
mixed-valence species, the HOMO–LUMO gap determines the
rate of electron transfer from one site to the other.28 These
results are reflection of superexchange mechanism, by which
electron hopping from one dimetal center to the other via the
bridge. Note that sulfur atoms as the coordination donors are
capable of enhancing the electronic delocalization and increas-
ing the electron transfer rate. It is confirmed that nitrogen-
chelating groups would weaken the Mo2–Mo2 interaction and
lower the electron transfer rate. As is well known, oxygen (O),42
nitrogen (N)43 and sulfur (S)44,45 are the major donor atoms in
naturally occurring metal enzyme systems, which may function
Electronic structures and spectroscopic properties
To further investigate the electronic structures of compounds
1–3, DFT calculations were performed on the simplified model
compounds [Mo2(NHCHNH)3]2[μ-1,4-{C(E)NH}2-C6H4] (E = NH
(1′), O (2′) or S (3′)), where the p-anisyl groups in the DAniF
ligands of 1–3 were replaced with hydrogen atoms. The calcu-
lated bond distances are in good agreement with those found
by X-ray crystallography (Table 4). The calculated frontier mole-
cular orbitals for model complexes 1′–3′ are presented in
Fig. 3. As can be seen, the LUMO is mainly based on the conju-
gated π orbitals of the bridging ligand, while the HOMO and
Table 4 Calculated bond distances and HOMO–LUMO energy gap for
1’–3’
Bond distances (Å)
HOMO–
LUMO
energy
Model
Mo–Mo
Mo2⋯Mo2
Mo–O
Mo–S
Mo–N
gap (eV)
1′
2′
3′
2.131
2.129
2.133
11.50
11.47
11.90
2.149
2.164
2.150
2.868
2.791
2.682
2.130
2.524
14760 | Dalton Trans., 2014, 43, 14756–14765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 3 Illustrations of the 0.04 contour surface of the TD-DFT calculated frontier molecular orbitals for model compounds 1’–3’ (O (red), S (yellow)
and N (blue)).
as electron transporter or involve electron transfer processes.
Hence, the understanding obtained from this study has the
significance of elucidating the electronic properties of some
biochemical systems.
Magnetic and optical properties of the mixed-valence species
Single-electron oxidation of compounds 1–3 with ferrocenium
hexafluorophosphate (FcPF6) afforded the expected mixed-
valence complexes 1+–3+ as evidenced by the measured EPR
spectra (Fig. 5). The main EPR signal in the spectra unambigu-
ously confirmed the existence of an unpaired electron in each
of the oxidized complexes. The corresponding g values fell in
the range of 1.940–1.948, suggesting that the unpaired electron
resides mainly on a Mo-based orbital. These g values are very
close to those obtained for complexes I+–III+ 27 and the biphe-
nylene bridged analogues (1.942 to 1.947).46 Such a narrow
range of g values in both series of mixed-valence compounds
indicates that all the molecular orbitals containing the
unpaired electron have very similar electronic structures. The
hyperfine structure in the EPR spectra is due to the coupling
of the unpaired electron with 95Mo or 97Mo (I = 5/2) nuclei.
Table 5 Experimentally observed and theoretically predicted absorp-
tion bands for 1–3 in comparison with those for I–IIIa
MLCT (nm)
1
2
3
I
II
III
Exp.
Cal.
465
517
490
520
560
555
492
548
618
652
715
685
Fig. 4 Electronic absorption spectra of 1–3 (top) and the simulated
electronic absorption spectra from the time-dependent DFT (TD-DFT)
calculations on the models 1’–3’ (bottom).
a Data cited from ref. 27.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14756–14765 | 14761

View Article Online
Paper
Dalton Transactions
Fig. 6 Visible and near-infrared spectra for the MV complexes 1+–3+,
along with the spectra of the neutral compounds 1–3. The intervalence
absorption bands are simulated with Gaussian-shaped curves as shown
by the dashed line profiles.
Table 6 Spectroscopic data for the mixed-valence 1+–3+, along with
those for I+–III+ for comparisona
EIV
εIV
Exp. Δν1/2
Cald Δν1/2
Complex
(cm−1
)
(M−1 cm−1
)
(cm−1
)
(cm−1
)
Γ
1+
2+
3+
I+
4980
4651
3182
4240
3440
2640
520
1171
3589
1470
3690
8840
5242
3688
4410
3290
1770
3392
3278
2711
3190
2820
2470
−1.45
−0.60
−0.36
−0.17
−0.14
0.30
II+
III+
Fig. 5 X-band EPR spectra of the radical cations generated by single
oxidation of the neutral compounds. Samples were measured in CH2Cl2
solution at 173 K.
12 660
a Data cited from ref. 27.
In the spectra from visible to near-infrared region, the in
situ generated samples of 1+–3+ display characteristic inter- lower intensities, and the bandwidth becomes narrower as the
valence (IV) transition absorption bands, in addition to the donor atoms vary from N to O to S (Table 6). The measured
MLCT absorptions similar to those observed for the neutral half-height bandwidths (Δν1/2) are much larger than the calcu-
complexes (Fig. 6). For the series, the band energy decreases lated values (2310νmax
)
1/2. According to Γ = 1 − (Δν1/2)/
a Γ value of −1.45, −0.60 and −0.36 cm−1 is
but the intensity increases in order from 1+ to 3+. As shown in (2310νmax
)
,
1/2 47
Table 6, the IV transition for the amidinate bridged complex found for 1+, 2+ and 3+, respectively, indicating that the mixed-
1+ has the highest energy (4980 cm−1). This band is slightly valence series belongs to the weakly coupled Class II in Robin–
red shifted to 4651 cm−1 for the amidate analogue (2+), and Day’s scheme. The variation trends of the IV band energy and
further down to 3182 cm−1 for the thioamidate derivative (3+). intensity as well as width are consistent with our observation
In comparison with I+–III+, which have similar Mo2⋯Mo2 dis- for the related series of I+–III+ and the biphenylene bridged
tance, complexes 1+–3+ present the IV bands with considerably analogues.27,45 These mixed-valence behaviors of 1+–3+
14762 | Dalton Trans., 2014, 43, 14756–14765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
conform well to the electrochemical and spectroscopic pro- ing to published methods. Terephthalamidine dihydrochloride
perties for 1–3 described above. Thus, it is confirmed that vari- was synthesized by modification of a reported procedure.38
ation of the coordinating atoms from N to O and further to S
Physical measurements
would induce the enhancement of the electronic coupling in
the [Mo2]–bridge–[Mo2] systems.
Elemental analyses were performed by an Elementar Vario EL
elemental analyzer. UV-vis spectra were obtained on a Shi-
madzu UV-3600 UV-vis-NIR spectrophotometer in CH2Cl2 solu-
tion. 1H NMR spectra were recorded on
a Bruker-400
Conclusions
spectrometer. Cyclic voltammograms (CVs) and differential
pulse voltammograms (DPVs) were obtained using a CH
Instruments model-CHI660D electrochemical analyzer in
0.10 M CH2Cl2 solution of nBu4NPF6, with Pt working and
auxiliary electrodes, an Ag/AgCl reference electrode, and a scan
In summary, the synthesis and isolation of three Mo2–Mo2
complexes with amidinate, amidate and thioamidate bridges
are described. Many reported dimolybdenum dimers are bridged
by dicarboxylate and several by thiodicarboxylate, and still
others have mixed-donor (O, N and S) bridges; however, com-
pound 1 is the first complete nitrogen-coordinated analogue.
Structural characterization in both solid-state and solution
indicates that all three complexes share the same molecular
structural skeleton. Electrochemical and spectroscopic studies
of 1–3 suggest that while S donor atoms induce the largest
enhancement of the electronic communication, N atoms on
the bridge promote, relative to O atoms, slightly weaker metal–
metal interaction. This trend is consistent with the variation of
the coordinating atoms E in softness. More covalent character
for the Mo–S bonds enhances the d(δ)–p(π) conjugation and
promotes stronger metal–metal interaction, while harder
N atoms take effect oppositely. The mixed-valence species 1+–3+,
generated by single oxidations of the corresponding neutral
compounds and measured in situ, displayed an isotropic EPR
signal with a g value of 1.942–1.947 and a broad metal to
metal intervalence transition band in near-IR region. The
energy and half-height width of the IV bands decrease in order
from 1+ to 3+. This demonstrates that simply by varying the
functional groups of the bridging ligand, the electronic coup-
ling between the two Mo2 units could be modulated in a rela-
tively wide range. Subtly tuning the electron coupling in
mixed-valence compounds of this type may be achieved by
alternation of O/N donors, which provides valuable guidance
for the development of experimental models for the refinement
of electron-transfer theories. Furthermore, the molecular struc-
tures of 1–3 and the electronic spectra could be reproduced by
DFT calculation using simplified model compounds. The
trend of MLCT energy variation predicted by TD-DFT calcu-
lation was consistent with the experimental results, on the
basis of which modulation of electron coupling by varying the
coordinating atoms (S > O > N) is confirmed.
rate of 100 mV s−1
.
X-ray structure determinations
Single-crystal data for 2·6C4H8O were collected on an Agilent
Gemini S Ultra diffractometer with Cu-Kα radiation (λ =
1.54178 Å) at 149(2)
K
and single-crystal data for
3·5.6C4H8O·2CH3CH2OH were collected on an Agilent Xcalibur
Nova diffractometer with Cu-Kα radiation (λ = 1.54178 Å) at
173(2) K. For both, the empirical absorption corrections were
applied using spherical harmonics, implemented in the
SCALE3 ABSPACK scaling algorithm.49 All the structures were
solved using direct methods, which yielded the positions of all
non-hydrogen atoms. Hydrogen atoms were placed in calcu-
lated positions in the final structure refinement. Structure
determination and refinement were carried out using
SHELXS-97 and SHELXL-97 programs, respectively.50 For the
two measured crystal structures, the solvent molecules were
disordered in multiple orientations, which were refined isotro-
pically. All non-hydrogen atoms were refined with an isotropic
displacement parameters.
Computational details
All DFT (density functional theory) calculations were per-
formed with the hybrid B3LYP51,52 functional implemented in
the Gaussian 09 package (Revision A0.2).53 The model com-
plexes were fully optimized. The standard 6-31G* basis set
were used for H, C and N atoms, and aug-CC-pvDZ basis set
for S and O atoms of the bridging ligands. The SDD basis set
together with SDD pseudo-potential were used for the heavy
metal Mo atoms. Time-dependent DFT (TD-DFT) calculations
were carried out to obtain 60 excitations for all the model com-
pounds. By replacing the p-anisyl groups on [Mo2(DAniF)3]+
with hydrogen atoms, the employed calculation models have
[Mo2(NHCHNH)3]+ units as the building blocks.
Preparation of terephthalamidine dihydrochloride
Experimental
Materials and methods
To a solution of benzene-1,4-dicarbonitrile (1.24 g, 9.7 mmol)
in 20 mL of THF, 40 mL of 1 M LiN(SiMe3)2 solution was
All manipulations were performed in a nitrogen-filled glove added dropwise in 30 min at 0 °C. The mixture was stirred at
box or by using standard Schlenk-line techniques. All solvents room temperature for 3 h and then cooled to 0 °C in an ice
were freshly distilled and dried over appropriate drying agents bath. The reaction was quenched by careful addition of 6 M
under N2. Terephthalamide,36 dithioterephthalamide,37 HCl–EtOH (35 mL) and the mixture was set aside for several
HDAniF,48 and Mo2(DAniF)3(O2CCH3),39 were prepared accord- hours. The precipitate was then filtered, washed with Et2O,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14756–14765 | 14763

View Article Online
Paper
Dalton Transactions
and then the powder was recrystallized from H2O–EtOH 4H, aromatic H), 3.72 (s, 24H, OCH3), 3.70 (s, 6H, OCH3), 3.63
mixture. Yield: (1.74 g, 96%). 1H NMR δ (ppm in DMSO-d6): (s, 6H, OCH3). UV-vis, λmax nm (ε, M−1 cm−1): 560 (2.6 × 104).
9.00 (s, 8H, NH), 8.06 (s, 4H, aromatic H).
Anal. Calcd (%) for C98H96Mo4N14O12S2: C, 55.80; H, 4.59; N,
9.29. Found: C, 55.52; H, 4.58; N, 9.41.
[Mo2(DAniF)3]2[μ-1,4-{C(NH)NH}2-C6H4] (1)
To a mixture of Mo2(DAniF)3(O2CCH3) (0.406 g, 0.40 mmol)
and terephthalamidine dihydrochloride (0.0472 g, 0.20 mmol),
30 mL of THF was added. 4.0 mL of 0.1 M NaOC2H5 solution
in ethanol was added slowly with stirring, and the reaction
mixture was stirred at room temperature for 4 h to produce a
red solution. The solvent was removed under reduced pressure,
and the residue was washed with ethanol (3 × 20 mL) and then
dried under vacuum. Diffusion of ethanol into a dichloro-
methane solution of the compound yielded orange yellow
coloured rhombus crystals. Yield: 0.32 g (77%). 1H NMR
δ (ppm in CDCl3): 8.45 (s, 2H, –NCHN–), 8.26 (s, 4H, –NCHN–),
8.23 (s, 4H, C(NH)NH), 7.74 (s, 4H, aromatic H), 6.62 (d, 16H,
aromatic H), 6.47 (d, 8H, aromatic H), 6.38 (d, 16H, aromatic
H), 6.25 (d, 8H, aromatic H), 3.71 (s, 24H, OCH3), 3.68 (s, 12H,
OCH3). UV-vis, λmax nm (ε, M−1 cm−1): 465 (2.35 × 104). Anal.
Calcd (%) for C98H98Mo4N16O12: C, 56.71; H, 4.76; N, 10.79.
Found: C, 56.39; H, 4.66; N, 10.97.
Acknowledgements
We gratefully thank the National Natural Science Foundation
of China (90922010, 21371074 and 21301070) and Jinan Uni-
versity for financial support and the Supercomputer Center at
Jinan University for computational resources.
Notes and references
1 N. S. Hush, Prog. Inorg. Chem., 1967, 8, 391–444.
2 N. S. Hush, Electrochim. Acta, 1968, 13, 1005–1023.
3 M. B. Robin and P. Day, Adv. Inorg. Chem. Radiochem.,
1967, 10, 247–422.
4 C. Creutz, Prog. Inorg. Chem., 1983, 30, 1–73.
5 R. A. Marcus, J. Chem. Phys., 1956, 24, 966–978.
6 C. Creutz, M. D. Newton and N. Sutin, J. Photochem. Photo-
biol., A, 1994, 82, 47–59.
7 C. Creutz and H. Taube, J. Am. Chem. Soc., 1969, 91, 3988–
3989.
8 C. Creutz and H. Taube, J. Am. Chem. Soc., 1973, 95, 1086–
1094.
[Mo2(DAniF)3]2[μ-1,4-{C(O)NH}2-C6H4] (2)
To a mixture of Mo2(DAniF)3(O2CCH3) (0.406 g, 0.40 mmol)
and terephthalamide (0.0272 g, 0.20 mmol), 30 mL of THF was
added. 4.0 mL of 0.1 M NaOC2H5 solution in ethanol was
added slowly with stirring, and the reaction mixture was
stirred at room temperature for 4 h. The solvent was removed
under reduced pressure, and the red residue was washed with
9 C. Creutz, Prog. Inorg. Chem., 1983, 30, 1–73.
ethanol (3 × 20 mL) and then dried under vacuum. The solid 10 R. J. Crutchley, Adv. Inorg. Chem., 1994, 41, 273–325.
was dissolved in 15 mL of tetrahydrofuran, and the solution 11 D. E. Richardson and H. Taube, Coord. Chem. Rev., 1984,
was layered with hexane. Red crystals formed in 5 days. Yield:
0.309 g (75%). 1H NMR δ (ppm in CDCl3): 9.01 (s, 2H, C(O)- 12 W. Kaim and G. K. Lahiri, Angew. Chem., Int. Ed., 2007, 46,
NH), 8.47 (s, 2H, –NCHN–), 8.32 (s, 4H, –NCHN–), 8.04 (s, 4H, 1778–1796.
aromatic H), 6.62 (d, 24H, aromatic H), 6.50 (d, 4H, aromatic 13 T. Ito, T. Hamaguchi, H. Nagino, T. Yamaguchi,
60, 107–129.
H), 6.43 (d, 4H, aromatic H), 6.38 (d, 8H, aromatic H), 6.30 (d,
4H, aromatic H), 6.22 (d, 4H, aromatic H), 3.71 (s, 24H, OCH3),
J. Washington and C. P. Kubiak, Science, 1997, 277, 660–
663.
3.68 (s, 6H, OCH3), 3.66 (s, 6H, OCH3). UV-vis, λmax nm (ε, M−1 14 T. Ito, T. Hamaguchi, H. Nagino, T. Yamaguchi, H. Kido,
cm−1): 490 (2.3 × 104). Anal. Calcd (%) for C98H96Mo4N14O14:
C, 56.66; H, 4.66; N, 9.44. Found: C, 56.39; H, 4.60; N, 9.57.
I. S. Zavarine, T. Richmond, J. Washington and
C. P. Kubiak, J. Am. Chem. Soc., 1999, 121, 4625–4632.
15 T. Ito, N. Imai, T. Yamaguchi, T. Hamaguchi,
C. H. Londergan and C. P. Kubiak, Angew. Chem., Int. Ed.,
2004, 43, 1376–1381.
[Mo2(DAniF)3]2[μ-1,4-{C(S)NH}2-C6H4] (3)
To a mixture of Mo2(DAniF)3(O2CCH3) (0.406 g, 0.40 mmol)
and dithioterephthalamide (0.039 g, 0.20 mmol), 30 mL of 16 J. C. Salsman, C. P. Kubiak and T. Ito, J. Am. Chem. Soc.,
THF was added. 4.0 mL of 0.1 M NaOC2H5 solution in ethanol 2005, 127, 2382–2383.
was added slowly with stirring, and the reaction mixture was 17 C. Lambert and G. Nöll, J. Am. Chem. Soc., 1999, 121, 8434–
stirred at room temperature for 4 h to produce a purple solu- 8442.
tion. The solvent was removed under reduced pressure. The 18 S. F. Nelsen, R. F. Ismagilov and D. A. Trieber, Science,
residue was washed with ethanol (3 × 20 mL) and then dried 1997, 278, 846–849.
under vacuum. Diffusion of ethanol into a tetrahydrofuran 19 A. Heckmann and C. Lambert, Angew. Chem., Int. Ed., 2012,
solution of the compound yielded purple crystals. Yield: 51, 326–392.
0.333 g (80%). H NMR δ (ppm in CDCl3): 10.29 (s, 2H, C(S)- 20 F. Albert Cotton, C. Lin and C. A. Murillo, J. Chem. Soc.,
NH), 8.43 (s, 2H, –NCHN–), 8.31 (s, 4H, –NCHN–), 7.88 (s, 4H, Dalton Trans., 1998, 3151–3154.
aromatic H), 6.63 (d, 16H, aromatic H), 6.54 (d, 16H, aromatic 21 F. A. Cotton, J. P. Donahue and C. A. Murillo, J. Am. Chem.
1
H), 6.37 (d, 8H, aromatic H), 6.29 (d, 4H, aromatic H), 6.09 (d,
Soc., 2003, 125, 5436–5450.
14764 | Dalton Trans., 2014, 43, 14756–14765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
22 F. A. Cotton, J. P. Donahue, C. A. Murillo and L. M. Pérez, 42 E. I. Solomon, T. C. Brunold, M. I. Davis, J. N. Kemsley,
J. Am. Chem. Soc., 2003, 125, 5486–5492.
S.-K. Lee, N. Lehnert, F. Neese, A. J. Skulan, Y.-S. Yang and
J. Zhou, Chem. Rev., 2000, 100, 235–350.
43 E. I. Solomon, P. Chen, M. Metz, S.-K. Lee and
A. E. Palmer, Angew. Chem., Int. Ed., 2001, 40, 4570–4590.
23 R. H. Cayton, M. H. Chisholm, J. C. Huffman and
E. B. Lobkovsky, J. Am. Chem. Soc., 1991, 113, 8709–8724.
24 R. H. Cayton, M. H. Chisholm, J. C. Huffman and
E. B. Lobkovsky, Angew. Chem., Int. Ed. Engl., 1991, 30, 862– 44 J. C. Fontecilla-Camps, A. Volbeda, C. Cavazza and
864. Y. Nicolet, Chem. Rev., 2007, 107, 4273–4303.
25 M. H. Chisholm and N. J. Patmore, Dalton Trans., 2006, 45 C. Tard and C. J. Pickett, Chem. Rev., 2009, 109, 2245–
3164–3169. 2274.
26 M. J. Han, C. Y. Liu and P. F. Tian, Inorg. Chem., 2009, 48, 46 X. Xiao, M. Meng, H. Lei and C. Y. Liu, J. Phys. Chem. C,
6347–6349. 2014, 118, 8308–8315.
27 X. Xiao, C. Y. Liu, Q. He, M. J. Han, M. Meng, H. Lei and 47 B. S. Brunschwig, C. Creutz and N. Sutin, Chem. Soc. Rev.,
X. Lu, Inorg. Chem., 2013, 52, 12624–12633. 2002, 31, 168–184.
28 C. Y. Liu, X. Xiao, M. Meng, Y. Zhang and M. J. Han, 48 C. Lin, J. D. Protasiewicz, E. T. Smith and T. Ren, Inorg.
J. Phys. Chem. C, 2013, 117, 19859–19865. Chem., 1996, 35, 6422–6428.
29 Y.-C. Tsai, H.-Z. Chen, C.-C. Chang, J.-S. K. Yu, G.-H. Lee, 49 CrysAlis RED, Version 1.171.31.7, Oxford Diffraction Ltd,
Y. Wang and T.-S. Kuo, J. Am. Chem. Soc., 2009, 131, 12534–
12535.
30 H.-G. Chen, H.-W. Hsueh, T.-S. Kuo and Y.-C. Tsai, Angew.
Chem., Int. Ed., 2013, 52, 10256–10260.
31 W.-Z. Chen and T. Ren, Inorg. Chem., 2006, 45, 9175–9177.
32 F. A. Cotton, C. Y. Liu, C. A. Murillo and X. Wang, Inorg.
Chem., 2006, 45, 2619–2626.
Abington, U.K., 2006.
50 G. M. Sheldrick, SHELXTL, Version 6.12, Bruker Analytical
X-ray Systems, Inc., Madison, WI, 2000.
51 A. J. Cohen and N. C. Handy, Mol. Phys., 2001, 99, 607–615.
52 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.
Matter, 1988, 37, 785–789.
53 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaus-
sian 09, Revision A.02, Gaussian, Inc., Wallingford, CT,
2009.
33 F. A. Cotton, L. M. Daniels, J. P. Donahue, C. Y. Liu and
C. A. Murillo, Inorg. Chem., 2002, 41, 1354–1356.
34 F. A. Cotton, Z. Li, C. Y. Liu and C. A. Murillo, Inorg. Chem.,
2006, 45, 9765–9770.
35 K. J. Nelson, R. W. McGaff and D. R. Powell, Inorg. Chim.
Acta, 2000, 304, 130–133.
36 M. Ghaemy and F. H. Nasr, J. Appl. Polym. Sci., 2012, 124,
1707–1715.
37 S. Dixon and R. J. Whitby, Tetrahedron Lett., 2006, 47,
8147–8150.
38 G.-L. Song, H.-J. Zhu, L. Chen, S. Liu and Z.-H. Luo, Helv.
Chim. Acta, 2010, 93, 2397–2405.
39 F. A. Cotton, C. Y. Liu, C. A. Murillo, D. Villagrán and
X. Wang, J. Am. Chem. Soc., 2003, 125, 13564–13575.
40 F. A. Cotton, Z. Li, C. Y. Liu and C. A. Murillo, Inorg. Chem.,
2007, 46, 7840–7847.
41 D. E. Richardson and H. Taube, Inorg. Chem., 1981, 20,
1278–1285.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 14756–14765 | 14765