Electrochemical behavior and dioxygen reactivity of tripodal dinuclear copper complexes linked by unsaturated rigid spacers

Article abstract of DOI:10.1039/c2dt31456h

New dinucleating ligands based on two tripodal tris(2-pyridylmethyl)amine (TMPA) units linked by a series of delocalized π-electrons spacers have been synthesized. Their di-Cu^II complexes have been prepared and structurally characterized. As compared to the corresponding monotopic complexes, these dinuclear Cu^II complexes reveal spectroscopic and voltammetric features ascribable to weakly perturbed electronic interactions. In the case of the anthracenyl spacer, observation both in the solid and in solution suggests that the existence of intramolecular π-π stacking interactions influences the geometry of the complex and hence its electronic properties. The bis-Cu^I complexes were prepared electrochemically. In the specific case of the complex bearing a mono-alkyne spacer, addition of dioxygen in acetonitrile leads to the slow formation of a trans-μ-1,2 peroxo Cu₂ complex which shows good stability at 268 K (t_1/2 = 240 s). Analysis of the kinetics of the peroxo formation by UV-vis spectroscopy suggests that the increased activation barrier for intramolecular binding of dioxygen is due to the rigidity of the spacer. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c2dt31456h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Electrochemical behavior and dioxygen reactivity of
tripodal dinuclear copper complexes linked by
unsaturated rigid spacers†
Cite this: Dalton Trans., 2013, 42, 2238
Antoine Gomila, Nicolas Le Poul, Jean-Michel Kerbaol, Nathalie Cosquer, Smaïl Triki,
Bénédicte Douziech, Françoise Conan* and Yves Le Mest*
New dinucleating ligands based on two tripodal tris(2-pyridylmethyl)amine (TMPA) units linked by a
series of delocalized π-electrons spacers have been synthesized. Their di-CuII complexes have been pre-
pared and structurally characterized. As compared to the corresponding monotopic complexes, these
dinuclear CuII complexes reveal spectroscopic and voltammetric features ascribable to weakly perturbed
electronic interactions. In the case of the anthracenyl spacer, observation both in the solid and in solution
suggests that the existence of intramolecular π–π stacking interactions influences the geometry of the
complex and hence its electronic properties. The bis-CuI complexes were prepared electrochemically. In
the specific case of the complex bearing a mono-alkyne spacer, addition of dioxygen in acetonitrile leads
Received 3rd July 2012,
Accepted 2nd October 2012
to the slow formation of a trans-μ-1,2 peroxo Cu2 complex which shows good stability at 268 K (t1/2
=
DOI: 10.1039/c2dt31456h
240 s). Analysis of the kinetics of the peroxo formation by UV-vis spectroscopy suggests that the
increased activation barrier for intramolecular binding of dioxygen is due to the rigidity of the spacer.
www.rsc.org/dalton
ligands have been developed allowing spectroscopic character-
ization of several transient CunO2 reactive species by variation
Introduction
Copper is a central metal in fundamental biological processes of the ligand topology (denticity, electronic and steric
such as electron transfer, O2 transport, substrate mono- or di- environment).4–6
oxygenation or hydroxylation, O2 dismutation and reduction to
The first step of the reaction between O2 and a CuI complex
H2O2 and H2O.1–3 Hence the reproduction of the structural is the formation of a superoxo CuO2 species in an end-on (η1)
and functional properties of Cu enzymes, with the aim of or side-on (η2) coordination of dioxygen (Scheme 1). Then the
synthesizing efficient biomimetic green catalysts and sensors, second step involves a second Cu center with formation of a
has become quite a challenge.4,5 On the basis of the XRD peroxo Cu2O2 species, under control of the ligand topology.
structures of Cu enzymes’ active site, the general strategy has This species may adopt different coordination modes (trans-
been to mimic the biological ligation of the metal by using tri- μ-1,2 peroxo, μ–η2:η2 peroxo or bis(μ-oxo)) (Scheme 1). A large
podal ligands.4,5 Particularly, mononuclear Cu complexes number of studies have shown that tetradentate tripodal
bearing tetradentate pyridyl ligands such as TMPA (TMPA = ligands such as TMPA lead preferentially to CuO2 species with
tris(2-pyridylmethyl)amine) and its derivatives have been end-on coordination of dioxygen, whereas bi- and tridentate
deeply investigated (Chart 1).6 Initially, these complexes were ligands favor side-on coordination. Such a difference is of par-
synthesized to mimic “blue-copper” proteins by constraining ticular importance as it determines the reactivity towards the
the metal in an atypical trigonal geometry and enhancing elec- O2 moiety: the trans-peroxo species exhibits nucleophilic and
tron transfer kinetics.7 Since then, many pyridyl-based tripodal basic behavior whereas μ–η2:η2 peroxo and bis(μ-oxo) show
electrophilic character.5,8 Most of the Cu complexes developed
so far are based on monotopic ligands. They lead in the pres-
ence of O2 to the intermolecular formation of dinuclear
Laboratoire de Chimie, Electrochimie Moléculaires et Chimie Analytique, UMR CNRS
species with dioxygen as a bridging ligand. The successive
evolution of superoxo to peroxo species is usually fast, followed
by rapid degradation of Cu2O2 peroxo species at room temp-
erature. To favor the formation of the dinuclear Cu2O2
adducts, an alternative approach was carried out through the
use of ditopic ligands in which two Cu sites are encapsulated
6521 – Université de Brest – UEB, CS 93837, 6 av. Le Gorgeu, 29238 Brest cedex 3,
France. E-mail: yves.lemest@univ-brest.fr, francoise.conan@univ-brest.fr
†Electronic supplementary information (ESI) available: Details on the EPR and
CV simulations and supplementary experimental data (EPR, solid-state UV-Vis,
peroxo decomposition). CCDC 858000 and 858001 for compounds 2 and 3. For
ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c2dt31456h
2238 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Chart 1
Scheme 1 Formation of superoxo, peroxo and bis-(μ-oxo) Cu complexes.
in a chelating shell bridged by carbonated chains (Chart 1).9 exerted through the linker at the Cu2O2 peroxo level were
The strategy was to enhance thermodynamics of the peroxo invoked to explain such behavior,10 as confirmed by molecular
formation from the superoxo by decreasing the reaction entro- mechanics studies.11 Interestingly, this ligand design enabled
pic cost through pre-organization of the binding site. Such an a dramatic increase of the kinetics of superoxo to peroxo evol-
approach had in fact little effect on thermodynamics because ution. A further stage was completed by better encapsulating
entropy gain was inhibited by enthalpic loss.9 Strain effects the Cu2O2 core to inhibit degradation, while still keeping the
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2239

View Article Online
Paper
Dalton Transactions
dinucleating design for lowering the reaction entropy. This dopamine β-monooxygenase (DβM) and peptidylglycine
was achieved by the synthesis of a specific macrocyclic tetra- α-hydroxylating monooxygenase (PHM), two similar enzymes
pyrazolyl-based ligand by Reedijk et al. which yields room- in which an efficient electron transfer from one site (CuA) to
temperature stabilization of the trans-μ-1,2 peroxo species in the other (CuB) is crucial for the activation of O2.17,18 Cur-
protic media, but with no reversibility for the O2 binding iously, no previous work has been concerned with the associ-
process.12 Alternatively, Comba et al. developed another ation of two non-coupled copper centers within a dinucleating
approach based on the stabilization of the μ-peroxo species ligand to mimic the effect of a proximal Cu site, able to trans-
through the combination of geometric and electronic effects fer an electron, such as observed in DβM and PHM. In this
by using two rigid bispidine backbones.13,14 Indeed this report, we present a new series of dinucleating ligands, based
specific ligand enforces a square-based pyramidal geometry on the TMPA moiety, which differs from analogous bis-TMPA
around Cu. Consequently, the peroxide lies in the equatorial ligands previously developed (Chart 1) by the nature of the
position (instead of the usual axial), and is much less labile. linker fragment, more favorable to electronic delocalization, by
The concept of stabilization by incorporation of a bridge their rigidity as well as by the position of the linkage. The
between two coordinating units was also extended to tridentate 6-pyridyl position for the linker has been preferred to the
tripodal ligands.5,15 For instance, Kodera et al. developed out- 5-position for better accessibility of the synthetic path and
standing models of hemocyanin, which reversibly trap molecu- stability. The ligands DPh,19 DAnt, DYne and DDyne (Chart 2)
lar dioxygen at room temperature.16
have been synthesized; their CuII complexes, [Cu2IIL-
In another connection, the design of the dinucleating (CH3CN)2]4+ isolated as their (CF3SO3)− salts (1–4) in the solid
ligand is a matter of interest conjointly for the preparation of state, have been prepared and fully characterized by X-ray diffr-
dinuclear Cu2 complexes in which the linking arm could action, UV-Vis-NIR and EPR spectroscopies as well as electro-
promote an electron transfer between the two copper ions. chemistry. The nature of the linker was investigated in terms
These systems could act as models for example of the of electron transfer properties. The reactivity towards O2 of the
I
electrogenerated CuI complexes, [Cu2 L(CH3CN)2]2+, has been
scrutinized. In the specific case of [CuI2DYne(CH3CN)2]2+, a
trans-μ-1,2 peroxo Cu2 complex relatively stable at room temp-
erature has been obtained.
Results and discussion
Syntheses
All ligands were prepared from the brominated precursor
6-BrTMPA through cross-coupling reactions catalyzed by palla-
dium complexes. The syntheses of DPh19 and DAnt were per-
formed through a Suzuki–Miyaura reaction between 2 equiv. of
6-BrTMPA and 1 equiv. of the corresponding diboronic acid
(1,4-benzene and 9,10-anthracene diboronic acids, 85% and
50% yield respectively) (Scheme 2). The ligand DYne was
Chart 2
obtained from a Sonogashira reaction (74% yield) between
Scheme 2 Synthetic pathway for the ligands DPh and DAnt.
2240 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Scheme 3 Synthetic pathway for ligand DYne.
Scheme 4 Synthetic pathway for ligand DDyne.
6-BrTMPA and the terminal alkyne 6-eTMPA18 (Scheme 3). The
compound DDyne was synthesized by the homo-coupling reac-
tion from 6-ethynylTMPA (62% yield) (Scheme 4). The four
dinuclear CuII complexes 1–4 were prepared by mixing 2 equiv.
of CuII(CF3SO3)2 respectively with the corresponding bisTMPA
ligand L in CH3CN.
Table 1 Pertinent crystal data and structure refinement details for complexes 2
and 3
2
3
Empirical formulaa
C66H64Cu2F12N10O16S4 C54H60Cu2F12N10O14S4
M (g mol−1
T (K)
)
1736.59
170(2)
P1
1556.44
170(2)
P1
ˉ
ˉ
Space group
a (Å)
b (Å)
c (Å)
α (°)
X-ray structures
11.3960(5)
12.4261(5)
14.3449(6)
111.206(4)
100.117(4)
99.357(4)
1806.88(13)
1
11.1778(4)
12.3060(5)
13.5309(5)
82.557(3)
69.394(3)
77.618(3)
1698.65(10)
1
Compounds [CuII2DAnt(CH3CN)2](CF3SO3)4 2 and [CuII2DYne-
(CH3CN)2(CF3SO3)2](CF3SO3)2·Et2O 3 were obtained as deep
blue single crystals suitable for X-ray analysis. Both com-
β (°)
γ (°)
V (Å3)
Z
ˉ
pounds crystallize in the triclinic system according to the P1
space group (Table 1). For each derivative, the asymmetric unit
involves one CuII ion, half of an organic ligand, one CH3CN
molecule, two trifluoromethanesulfonate anions and one mol-
ecule of solvent (EtOAc in 2, disordered Et2O in 3). The linker
being centered on an inversion center, the whole molecule is
then generated by symmetry (Fig. 1 and 2). Selected bond
lengths and angles are gathered in Table 2.
Dcalcd (g cm−3
)
1.596
1.522
μ (cm−1
F(000)
)
8.09
888
8.49
796
Reflections collected 13 683
11 827
5975/0.0226
Independent
reflection/Rint
Reflections with
I > 2σ(I)
7090/0.0332
4905
4671
Within the centrosymmetric [CuII2DAnt(CH3CN)2]4+ cation
of 2, the copper ions are surrounded by the four nitrogen
atoms belonging to the TMPA units and the nitrogen atom
from the coordinated CH3CN solvent molecule. A careful exam-
ination of the bond lengths and angles (Table 2) strongly evi-
dences that each CuII ion displays a distorted trigonal-based
Nv
491
416
c
R1 b/wR2
0.0589, 0.1666
1.031
+0.840/−0.761
0.0555, 0.1798
1.065
1.252/−0.530
GooFd
Δρmax, min (e Å−3
)
a The asymmetric unit contains 0.5 of the chemical formula. b R1 = ∑|
Fo − Fc|/Fo. c wR2 = {Σ[w(Fo − Fc2)2]/Σ[w(Fo2)2]}1/2
.
d GooF = {Σ[w(Fo
−
2
2
Fc2)2]/(Nobs − Nvar)}1/2
.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2241

View Article Online
Paper
Dalton Transactions
Table 2 Selected distances (Å) and angles (°) for a copper environment in com-
plexes 2 and 3
2
3
Cu1–N1
Cu1–N2
Cu1–N3
Cu1–N4
2.035(3)
2.004(3)
2.020(3)
2.140(3)
1.953(3)
2.006(3)
2.022(3)
1.989(3)
2.436(3)
1.993(3)
2.506(3)
83.70(13)
165.17(13)
80.28(11)
96.89(14)
88.19(11)
81.76(12)
80.81(11)
173.09(13)
86.11(11)
100.18(11)
97.20(13)
87.94(12)
106.08(12)
163.39(12)
87.03(12)
Cu1–N5
Cu1–O6ii
N1–Cu1–N2
N1–Cu1–N3
N1–Cu1–N4
N1–Cu1–N5
N1–Cu1–O6ii
N2–Cu1–N3
N2–Cu1–N4
N2–Cu1–N5
N2–Cu1–O6ii
N3–Cu1–N4
N3–Cu1–N5
N3–Cu1–O6ii
N4–Cu1–N5
N4–Cu1–O6ii
N5–Cu1–O6ii
80.80(12)
127.50(13)
117.44(12)
94.78(13)
Fig. 1 ORTEP view of the cationic complex of 2. Code of equivalent positions:
(i) 1 − x, −y, 1 − z.
83.42(13)
80.07(12)
175.23(14)
108.42(12)
97.96(14)
103.71(13)
Code of equivalent positions: (ii) −1 + x, y, z (symmetry label used in
Fig. 2).
Table 3 Metal environment and intramolecular metal–metal distances (Å) in
complexes 1–3
Complex
1a
2
3
Geometry
Distorted
octahedral
(4 + 1 + 1)
9.295(3)
Distorted TBP
(τ = 0.80)
Distorted
octahedral
(4 + 1 + 1)
8.209(1)
Fig. 2 ORTEP view of the cationic complex of 3. Only the coordinated oxygen
atom (O6) of the triflate anion is shown. Code of equivalent positions: (i) 1 − x,
−y, 1 − z; (ii) −1 + x, y, z; (iii) −2 + x, y, z.
Cu⋯Cu distance (Å)
a See ref. 19 and ESI.†
8.470(1)
bipyramid (TBP) geometry (τ = 0.80)20 in which the three
pyridyl nitrogen atoms of the TMPA unit (N1, N3 and N4) are
located in the basal plane. The axial coordination sites are
occupied by the nitrogen atom from the tertiary aliphatic
amine (N2) and the nitrogen atom from a CH3CN group (N5).
This copper geometry is nearly similar to that observed for the
parent mononuclear complex [CuIITMPA(CH3CN)]2+.21 The two
copper sites are located in the trans-position with respect to
the bridging linker. This conformation is likely due to steric
hindrance and electronic repulsion between the copper
centers.4
occupied by the nitrogen atom of the third pyridyl group N4,
with quite a long Cu–N4 bond length (2.436(3) Å), and by one
of the oxygen atoms from an adjacent trifluoromethanesulfo-
nate anion (distance Cu–O6ii = 2.506(3) Å). Whereas compound
3 displays a structure with a distorted CuN5O octahedral
environment, for compound 2 the retention of the TBP geome-
try, observed for the mononuclear TMPA copper complex, is
best explained by intramolecular π-stacking interactions
between a non-substituted pyridyl cycle and the peripheral aro-
matic ring of the anthracenyl linker. These two planes are
almost parallel and form a dihedral angle of 6.3°. The exist-
ence of intramolecular π-stacking interactions in complex 2
induces shorter Cu–N4 distances and consequently a shorter
Cu–Cu distance (Table 3) as compared to the previously
reported compound [CuII2DPh(CH3CN)2]2+, as its perchlorate
salt,19 the bridging length for both phenyl and anthracenyl
linkers being close. It has been verified that the triflate salt
presently isolated as [CuII2DPh(CH3CN)2(CF3SO3)2](CF3SO3)2 1
displays the same spectroscopic and structural features
(see ESI†).
The copper geometry of compound [CuII2DYne-
(CH3CN)2(CF3SO3)2](CF3SO3)2·Et2O 3 can be described as a dis-
torted CuN5O pseudo-octahedral as expected from the Jahn–
Teller effect considering the bond lengths and bond angle
values (Table 2). The plane of the pyramid is well defined by
the nitrogen atom from the amine function N2, two nitrogen
atoms belonging to the non-substituted pyridyl groups N1 and
N3, and the nitrogen atom N5 of the exogenous CH3CN ligand.
Within this plane, the Cu–N bond lengths vary from 1.989(3)
to 2.022(3) Å which is usual for such geometry. Owing to weak
interactions, the two axial coordination sites are respectively
2242 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
EPR and Vis-NIR spectroscopies of the CuII complexes
indicates a less tetragonal geometry (rhombic signal, gx ≠ gy)
in good correlation with the XRD structure (τ = 0.80). Between
X-band EPR spectra of dinuclear complexes [CuII2L(CH3CN)2]4+ 500 and 900 nm, the spectra of the CuII complexes exhibit
were recorded in CH3CN at T = 150 K (Fig. 3A). EPR parameters broad and weak absorption bands ascribable to forbidden d–d
(Table 4) were determined by numeric simulation of the experi- transitions (Fig. 3B). The Vis-NIR spectra of CH3CN solutions
mental spectra (Fig. S2†) on the basis of no exchange or of [CuII2DPh(CH3CN)2]4+, [CuII2DYne(CH3CN)2]4+ and [CuII2D-
dipolar interactions between the two metal centers. Good fits Dyne(CH3CN)2]4+show two distinct d–d bands of nearly equal
could be obtained for all complexes, attesting to the lack of intensities: one around 640 nm (170 ≤ ε ≤ 260 M−1 cm−1) and
interaction. This can be explained by the large intermetallic the other above 800 nm (275 ≤ ε ≤ 345 M−1 cm−1) (Table 4).
distance observed by X-ray diffraction analysis (>8 Å, see These results suggest that these CuII complexes possess dis-
Table 3) and the absence of any through-bridge electronic torted structures in solution, intermediate between the SBP
coupling. As shown in Table 4, compounds [CuII2DPh- and TBP geometry for pentacoordinated cupric complexes. In
(CH3CN)2]4+,
[CuII2DYne(CH3CN)2]4+
and
[CuII2DDyne- the case of the dinuclear compound [CuII2DAnt(CH3CN)2]4+, a
(CH3CN)2]4+ exhibit close parameters. Indeed, in frozen solu- major and large d–d band is observed at 817 nm with a high-
tions, the spectra of these compounds show a typical signal energy shoulder indicating a distorted TBP geometry. These
of CuII species with a hyperfine structure characteristic of a EPR and Vis-NIR spectroscopic observations are in accordance
dx2−y2 ground state corresponding to a tetragonal square-based with those obtained in the solid state by X-ray diffraction
pyramidal (SBP) geometry (gz > gx ≈ gy). Such a result is con- analysis and UV-Vis (Fig. S3†) considering the unbinding of
sistent with the pseudo-octahedral conformation found by the triflate anion to CuII in solution for compounds 1 and 3.
XRD analysis for 1 and 3. The complex [CuII2DAnt(CH3CN)2]4+ This suggests that the ligand DAnt favors the TBP-like confor-
exhibits a different EPR signal. The simulation of the spectrum mation around the cupric ion for complex 2, hence that the
Fig. 3 (A) Experimental EPR (T = 150 K) and (B) Vis-NIR spectra of complexes [CuII2L(CH3CN)2]4+ in CH3CN (L = DPh, DAnt, DYne, DDyne).
Table 4 Vis-NIR and EPR (T = 150 K) data for complexes [CuII2L(CH3CN)2]4+ in CH3CN
g (A/G)a
Compound
λmax/nm (εmax/M−1 cm−1
)
gx
gy
gz
[CuII2DPh(CH3CN)2]4+
[CuII2DAnt(CH3CN)2]4+
[CuII2DYne(CH3CN)2]4+
[CuII2DDYne(CH3CN)2]4+
[CuIITMPA(CH3CN)]2+
641 (170)
638 (sh)
639 (260)
641 (170)
630 (sh)
818 (275)
817 (463)
800 (345)
807 (275)
889 (238)
2.060
2.210 (135)
2.060
2.060
2.19 (113)
2.060
2.110 (50)
2.060
2.060
2.19 (113)
2.240 (177)
2.010 (50)
2.235 (177)
2.240 (177)
2.01 (65)
a Estimated by simulation.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2243

View Article Online
Paper
Dalton Transactions
intramolecular π–π interactions must be at work also in towards positive values vs. [Cu(TMPA)]2+ is explained by the
solution.
steric effects induced by a substituent in the 6th position of
one pyridyl ring on the geometry (SBP vs. TBP as compared to
the unsubstituted complexes) and hence on its electronic pro-
perties.22 The more negative potential in the series was
obtained for complex [CuII2DAnt(CH3CN)2]4+ (E1/2 = −0.34 V).
Its value is close to that found for TMPA (−0.40 V) under the
same conditions, emphasizing the geometrical/electronic
dependence of the potential in agreement with the spectro-
scopic results. Exhaustive electrolyses-coulometries of the solu-
tion of the di-CuII complexes under N2 established the
quantitative reduction as a two-electron process. The blue-
green color of the solution disappeared after electro-reduction
yielding a yellow solution. The absence of any EPR signal and
any d–d absorption band in UV-Vis spectroscopy for the
reduced solution confirmed the effective transformation of the
di-CuII complex into di-CuI. Also, the reversibility of the redox
response at any scan rates and before/after electrolysis (Fig. 4
and 5) suggests that the coordination sphere of Cu remains
similar at both redox states (with CH3CN coordinated). This
was confirmed by rotating-disk electrode voltammetry (RDEV)
experiments (Fig. 4B). Such results are in agreement with pre-
vious NMR studies on [CuITMPA(CH3CN)]+ in CH3CN which
Electrochemical behavior
The behavior by cyclic voltammetry (CV) of the bimetallic Cu
complexes [CuII2DPh(CH3CN)2]4+, [CuII2DAnt(CH3CN)2]4+,
[CuII2DYne(CH3CN)2]4+ and [CuII2DDyne(CH3CN)2]4+ in
CH3CN–NBu4PF6 emphasized for all complexes a single quasi-
reversible system (0.01 < v < 5 V s−1), with a half-wave potential
E1/2 value ranging between −0.34 and −0.28 V vs. Fc+/Fc
(Table 5). As has been previously discussed, the shift of E1/2
Table 5 Electrochemical data for compounds [CuII2L(CH3CN)2]4+ and [CuIITMPA-
(CH3CN)]2+ in CH3CNa
Compound
E1/2 (V)
ΔEp (mV)
ipa/ipc
[CuII2DPh(CH3CN)2]4+
[CuII2DAnt(CH3CN)2]4+
[CuII2DYne(CH3CN)2]4+
[CuII2DDYne(CH3CN)2]4+
[CuIITMPA(CH3CN)]2+
−0.32
−0.34
−0.29
−0.28
−0.40
90
0.86
0.89
0.94
0.98
1.03
100
110
90
65
a E1/2 = (Epa + Epc)/2; ΔEp = Epa − Epc; E vs. Fc+/Fc; in CH3CN–NBu4PF6
(0.1 M) at v = 0.1 V s−1 at a Pt electrode.
Fig. 4 (A) Cyclic voltammetry (v = 0.1 V s−1) and (B) rotating disk electrode voltammetry (ω = 1500 t min−1) at a Pt electrode of compound [CuII2DYne(CH3CN)2]4+
(10−3 M) in CH3CN–NBu4PF6 (0.1 M): (a) before and (b) after electrolysis at −0.50 V.
Fig. 5 (A) Cyclic voltammetry at a Pt electrode of compound [CuII2DYne(CH3CN)2]4+ (10−3 M) in CH3CN–NBu4PF6 (0.1 M) at v = 0.1 V s−1 and (B) 0.01 V s−1 < v <
0.2 V s−1 and (C) 0.5 V s−1 < v < 10 V s−1
.
2244 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
demonstrated an axial coordination of a CH3CN molecule, in
dynamic exchange with one pyridyl arm of the TMPA ligand.9
Assuming that the electron transfer kinetics is fast on the
Table 6 Transition MLCT of selected peroxo trans-μ-1,2 Cu complexes
Ligandsa
DYne
λmax/nm ε/(mM−1 cm−1
)
Solvent
Ref.
basis of the [CuII/I(TMPA)(CH3CN)]2+/+ characteristics (ΔEp
=
513 (10.5), 650 (sh)
522 (10.5), 605 (sh)
525 (11.5), 586 (7)
540 (11.1), 600 (9)
535
504 (8), 620 (5)
488 (3), 649 (2)
499 (6.8), 650 (sh)
503 (6.0), 650 (3)
525 (4), 625 (2)
CH3CN
This work
CH3CH2CN
CH3CH2CN
CH3CH2CN
CH3CH2CN 10
CH3CH2CN 14
CH3CH2CN 14
CH3CH2CN 14
CH3CH2CN 14
65 mV), almost the same as ferrocene under similar conditions
(ΔEp = 60 mV), the larger peak separation for the four com-
plexes (ΔEp = 100 10 mV, at v = 0.1 V s−1) suggests a 2 × 1
electron sequential mechanism in which the two separate
redox processes occur at different standard potentials E0, the
reduction of the second center being slightly more difficult
than the first. This assumption is emphasized from the wave
shape of the voltammogram of complex [CuII2DYne-
(CH3CN)2]4+, at 0.1 V s−1, which shows the two close cathodic
and two anodic peaks (Fig. 5A, red arrows). This was fully cor-
roborated by voltammetric simulation of the CV (Fig. S4†),
thus allowing the determination of the standard potential
difference, ΔE0 = 70 mV at 0.1 V s−1, between the microscopic
mono-electronic transfers.23 Half-electrolyses were performed
in order to evaluate the possible formation of a stabilized
mixed-valence (CuIICuI) species. The absence of intervalence
bands in NIR-Vis spectroscopy of the half-electrolyzed solution
clearly indicates that the two Cu centers do not interact elec-
tronically through the π orbitals of the linker. Accordingly, the
EPR spectrum only reflects the presence of the di-CuII form at
half concentration. This indicates that there is no strong stabil-
ization of the mixed-valence through electronic interaction.24
Hence, the low value for the microscopic standard potential
difference ΔE0 (70 mV) and the resulting macroscopic peak
separation ΔEp likely result from slight coulombic repulsions
between Cu centers through space. This is corroborated by the
inverse variation of ΔEp (Table 5) with the Cu–Cu distance
obtained from XRD analysis (Table 3): 1 < 2 < 3.
TMPA
D1
DO
9
9
Bisp
Bisp2Et
Bisp2Pr
Bisp2Xyl
(MePy22Pz)2
(Me-Uns-penp)2Pr 550
BPMAN 505 (10.5), 620 (5.4)
CH3CN
Acetone
CH2Cl2
12
28
29
a Refer to Charts 1 and 2 for representation of the different ligands.
Fig. 6 UV-Vis spectra before (dash black line) and after (red line) addition of
O2 on the electrogenerated reduced form of complex [CuII2DYne(CH3CN)2]4+ in
CH3CH2CN–NBu4PF6 at T = 193 K. Insert: variation of absorbance value at
522 nm with time (experimental points and simulated curve following two
irreversible consecutive first-order reactions, see kinetic part for details).
Interaction of O2 with CuI complexes
most of the Cu2–O2 adducts have been characterized (Table 6).
The reaction of the di-CuI complexes, [CuI2L(CH3CN)2]2+ (L = The addition of dioxygen led to the formation of the purple
DPh, DAnt, DDyne), with O2 was first investigated in CH3CN at species which is relatively stable at 193 K (t1/2 = 54 min for
298 K. The CuI2 complexes were electrogenerated from di-CuII 10 mol. equiv., as shown in Fig. 6, insert). This species is EPR
species in a glovebox under a dry and inert atmosphere as silent (Fig. S5†). In UV-Vis spectroscopy, it displays
a
described above.
maximum peak at λmax = 522 nm together with a shoulder at
Addition of dioxygen to the CuI complexes induced ca. 605 nm. The presence of this shoulder is a discriminating
immediately a change in the color of the solution, from yellow signature between a trans-μ-1,2 peroxo and a μ–η2:η2 peroxo
to green, similarly to what is observed with [CuITMPA]+. This complex. Hence, the O2 adduct of [CuII2DYne(CH3CN)2]4+ can
indicated a fast reaction between the (CuI)2 complex and dioxy- be assigned to the trans-μ-1,2-peroxo complex in good concor-
gen with no detectable intermediate species. Remarkably, the dance for TMPA-related mono and dinuclear complexes in the
room temperature behavior of complex [CuI2DYne(CH3CN)2]2+ same medium (Table 6). The absorption bands at 522 nm and
was different since a purple/red solution could be observed for 605 nm can thus be ascribed to Ligand-to-Metal-Charge-Trans-
several seconds after the addition of a low amount of dioxygen fer (LMCT) πσ* → d and πv* → d transitions. The studies at
(ca. 1 equiv.), suggesting the stabilization of a CuO2 transient room temperature have shown that the di-alkyne complex
species. The CVs recorded after interaction with O2 of the final [CuI2DDYne(CH3CN)2]2+, as well as the other complexes
green solution revealed the presence of a mixture of several [CuI2DPh(CH3CN)2]2+ and [CuI2DAnt(CH3CN)2]2+, display a
reducible species, also detectable by EPR and UV-Vis as much faster kinetics of decomposition after reaction with O2,
different CuII species, which can be ascribed to (bis)hydroxo as is observed with the intermolecular [CuII2(TMPA)2(O2)]
and/or (bis)fluoro CuII complexes (Fig. S6–S8†) as previously adduct complex, contrarily to the [CuI2DYne(CH3CN)2]2+.
observed.25–27 In order to characterize this purple/red inter-
This indicates that a structural specificity of the latter
mediate, UV-Vis experiments were first performed in propio- favors the transient stabilization of the trans-μ-1,2 peroxo
nitrile (CH3CH2CN) at 193 K (−80 °C) under which conditions adduct which suggests that the binding of dioxygen is
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2245

View Article Online
Paper
Dalton Transactions
Fig. 7 UV-Vis spectra of the electrogenerated reduced form of complex [CuII2DYne(CH3CN)2]4+ in CH3CN–NBu4PF6 at T = 268 K after addition of 1 mol. equiv. of
O2: (A) formation (between each curve: Δt = 2 min) and (B) disappearance (between each curve: Δt = 20 min) of the Cu2-peroxo species.
intramolecular for [CuI2DYne(CH3CN)2]2+, and intermolecular geometry around the CuII ions (more tetragonal) and conse-
for the three other complexes. Such an assumption is coherent quently a dramatic decrease of the peroxo maximum wave-
with the existence of a short linker for the DYne ligand. length (λmax = 504 nm and 488 nm for Bisp and Bisp2Et
Indeed, a simple rotation of the XRD structure around the respectively, as shown in Table 6).14 The shoulder at λ =
ethynyl bridge yields a face-to-face configuration in which a 420 nm is typically in the range of that obtained for end-on
Cu–Cu distance (d ≈ 4–4.5 Å) is close to that observed for superoxo Cu : O2 (1 : 1) species.4 On the basis of the sequential
[TMPACuII(O2)CuIITMPA]2+ (d = 4.36 Å).30
pathway for the formation of the Cu2 : O2 (2 : 1) peroxo
The formation and evolution of the O2 adduct of the complex which goes through the superoxo step (Scheme 1),
cuprous complex [CuI2DYne(CH3CN)2]2+ was then monitored such a transition is usually less easy to detect for dinuclear
in CH3CN at 268 K for 1 mol. equiv. of dioxygen (see ESI† for than mononuclear species. Indeed when the complex is well
the experimental procedure). Fig. 7 shows the evolution of the preorganized thanks to dinuclearity, the rate limiting step is
UV-vis spectrum. At this temperature, a slow increase (Fig. 7A), the initial dioxygen binding to CuI (step 1 in Scheme 1)
followed by a slow decrease (Fig. 7B), of the main peak at λmax meaning that this superoxo intermediate does not accumulate
= 513 nm was observed. Concomitantly, two shoulders of low as quickly converted into the intramolecular peroxo form (step
absorbance (vs. the peak at 513 nm) were detected at 420 nm 2). For the complex [CuI2DYne(CH3CN)2]2+, the detection of a
and 650 nm. The peak at λmax = 513 nm is close to that found band at 420 nm concomitantly to the transition at 513 nm
in CH3CH2CN (λmax = 522 nm). As well, the value of the molar indicates that the formation of the trans-μ-1,2 peroxo complex
extinction coefficient at 513 nm (ε = 104 M−1 cm−1) remains from the superoxo is slightly slower than the superoxo for-
specific for a trans-μ-1,2 peroxo Cu2 : O2 adduct.31 However, the mation, as previously observed for the Bisp2Xyl tetradentate
absorbance for the πv* → d transition around 600–650 nm is ligand.14 Indeed the transient formation of a bis-superoxo
significantly lower than that observed in CH3CH2CN. Such an adduct releasing afterwards one of the O2 moieties cannot be
effect has been already reported. For instance, Karlin et al. strictly excluded, but seems difficult to discriminate. By com-
showed that the substitution of one pyridyl arm by a thioether parison to its ethyl (D1) and ether (DO) analogues,9,10 the rigid
for CuTMPA leads to a dramatic decrease of the intensity of monoyne linker seems thus to have a strong impact on the kin-
the πv* → d transition for the peroxo complex, due to distor- etics of the peroxo formation. Likely, the entropic factors are
tion of the geometry.32 It has also been reported that the inten- less determining on the kinetics due to a lower preorganiza-
sity and position of the latter band present large variations tion, refrained by rigidity.
depending both on the solvent and nature of the ligands:
these factors must influence the geometric constraints on the
Cu2–O2 peroxo species, by controlling the orbital overlap and
Kinetic monitoring of the formation/evolution of the peroxo
complex by time-resolved UV-Vis spectroscopy
The kinetics of the reaction of [CuI2DYne(CH3CN)2]2+ with O2
has been scrutinized in order to analyze the effect of the rigid
alkyne bridge around which a rotation is a prerequisite for
intramolecular O2-bridged binding. In complement to the
monitoring of the reaction in a stoichiometric ratio Cu2 : O2
(2 : 1) (Fig. 7), the reaction has also been analyzed in the pres-
ence of an excess of O2, 10 and 50 equiv. (Fig. 8). As indicated
above, the absorption corresponds to the purple trans-μ-1,2
peroxo species. Fig. 9 shows the variation of the absorbance vs.
in consequence the energy and intensity of the πv* → d tran-
sition band.4,14 For the Cu2DYne-based O2 adduct, such an
effect could result from a slight variation of the trigonal geo-
metry adopted by the Cu centers towards a more tetragonal
conformation. This is substantiated by the slight shift of the
LMCT πσ* → d transition from 522 nm in CH3CH2CN to
513 nm in CH3CN, as a result of a less trigonal coordination
mode.33 Indeed, Comba et al. showed that linking two bispi-
dine ligands by an ethyl spacer induced a variation of the
2246 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Fig. 8 Disapperance of the Cu2O2–peroxo species at T = 268 K after addition of (A) 10 mol. equiv. (between each curve, Δt = 75 s), and (B) 50 mol. equiv. (between
each curve, Δt = 20 s) of O2 on the reduced form [CuI2DYne(CH3CN)2]2+ in CH3CN–NBu4PF6.
Fig. 9 Experimental (black dots) and simulated (red line) points for formation/disappearance of the Cu2–peroxo species at T = 268 K after addition of (A) 1 mol.
equiv., (B) 10 mol. equiv., and (C) 50 mol. equiv. of O2 on the reduced form [CuI2DYne(CH3CN)2]2+ in CH3CN–NBu4PF6.
Scheme 5 Formation and disappearance of the intramolecular trans-μ-1,2 peroxo complex B from the reaction of the electrogenerated reduced form [CuI2DYne-
(CH3CN)2]2+ A with O2.
time for the formation/disappearance of the O2 adduct of the the same kinetic law can be used for the three situations. The
reduced form [CuI2DYne(CH3CN)2]2+ in CH3CN for three shape is characteristic of two successive irreversible reactions
different O2/complex concentration ratios. The decrease of the of transformation of a species A into B (reaction 1) then B into
absorption corresponds to the evolution towards oxidation pro- C (reaction 2), as depicted in Scheme 5. This classical formal-
ducts. A major observation is that the shape of the variation ism and associated kinetic equations34 were used to fit the
curves in Fig. 9 remains similar in all cases. This indicates that experimental curves and evaluate the first-order kinetics
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2247

View Article Online
Paper
Dalton Transactions
Table 7 Rates of formation (k1, k1’) and decomposition (k2, k2’, t1/2) for the
peroxo complex derived from [CuI2DYne(CH3CN)2]2+, obtained from the fitting
of experimental data for different O2 vs. Cu ratiosa
Table 8 Second-order rate constant (k1’) for the formation of selected trans-
μ-1,2 peroxo Cu complexes from the CuI reaction with O2
Cu2–O2 complex
k1′/M−1 s−1 T/K Solvent
Ref.
[O2] in mol.
[CuII2DYne(O2)]2+
1
2
268 CH3CN
193 CH3CH2CN This work
183 CH3CH2CN
183 CH3CH2CN
203 Acetone
This work
equiv. vs. Cu k1/s−1
k2/s−1
k1′/M−1 s−1 k2′/M−1 s−1 t1/2/s
[Cu2(TMPA)2(O2)]2+
[Cu2D1(O2)]2+
6.2 × 104
7.1 × 104
9
9
36
9
1
10
50
1.4 × 10−4 9 × 10−5 0.23
2.3 × 10−3 9 × 10−4 1.15
0.15
0.45
0.06
21 600
1900
240
[Cu2(6-MeTMPA)2(O2)]2+ >106
2 × 10−2
2 × 10−3 0.67
[Cu2(TMPAE)2(O2)]2+
4.3 × 103
183 CH3CH2CN
a At 268 K in CH3CN.
obtained with dinucleating ligands (D1, DO), slow formation of
the trans-μ-1,2 peroxo species for the Cu2–DYne complex was
unexpected, since dinuclearity decreases the activation entropy
through enhanced preorganization.9,37 As discussed above
concerning the transient absorption at 420 nm, a best expla-
nation would be the rate-limiting step for the formation of the
peroxo due to a high activation energy required for the rotation
of the initial superoxo around the alkyne bridge to yield the
face-to-face intramolecular peroxo. Another hypothesis, pre-
viously mentioned to rationalize the inertia of the
[CuI2bTMPA]2+ complex vs. O2,38 was the high redox potential
due to a highly stabilizing (poor donating) electronic environ-
ment towards CuI. Such an assumption has to be excluded
here since the complex [Cu2DYne(CH3CN)]4+ displays a redox
potential value that is close to those found for analogous com-
plexes (except bTMPA, Table 9). At last, concerning the stability
of the peroxo species over decomposition, the analysis of the
time-dependent variation of the λmax value gives access to the
decomposition half-time (t1/2) for [Cu2DYne(O2)]2+ in both
nitrile solvents with O2-saturated conditions: t1/2 = 240 s at T =
268 K in CH3CN and t1/2 = 3240 s at T = 193 K in CH3CH2CN.
The value found in CH3CN is in the same range than those
found by Karlin et al. for D1- and DO-based Cu complexes but
under different experimental conditions (t1/2 = 4 s and 20 s,
respectively, at 298 K in acetone).10 On this basis only, the
influence of the linker on the stability of the peroxo is difficult
to rationalize. Indeed, many parameters are involved: the
dinuclear design vs. mononuclear of the N4 ligand is supposed
to enforce stabilization depending on the strain relief for the
peroxo species (distance between Cu),11 together with geo-
metric stabilization by inducing a less labile position for the
peroxide (tetragonal conformation), steric protection against
protons or peroxo species, and lower donation from N-substi-
tuents (strengthening of the Cu–O and O–O bonds).10
Undoubtedly, a short distance between Cu centers by half-
rotation of the complex around the linker is a prerequisite for
the formation of an intramolecular peroxo species in the series
of the unsaturated linkers. We thus suggest that the better
stability of the Cu2DYne peroxo complex compared to the
intermolecular analogous peroxo complexes (DPh, DAnt and
DDYne) is likely due to the shielding effect of the bis-TMPA
core which acts as a semi-cryptand in closed conformation.
Additionally, the rigid mono-linker induces a favored tetra-
gonal geometry, as shown by the low λmax = 513 nm value,
constants k1 and k2, assuming the species A to be the complex
[CuI2DYne(CH3CN)2]2+ in its cuprous reduced form, B the
peroxo intermediate and C the final decomposition species.35
For the three different initial concentration ratios between
species A and O2, reasonable fits could be obtained (Fig. 9).
Particularly, the best match was obtained for 10 mol. equiv. of
dioxygen (Fig. 9B). The kinetics simulations lead approxi-
mately to the same molar extinction coefficient value at
513 nm (10 500 1000 cm−1 M−1) for the three situations. The
simple model used becomes less appropriate in the case of a
2 : 1 Cu2–O2 stoichiometry in which case the peroxo disappear-
ance is slower. The first-order constants k1 and k2 obtained
from these simulations are gathered in Table 7. Since the reac-
tion A → B is bimolecular (it involves A and O2), k1 is a
pseudo-first-order rate constant as expressed by eqn (1):
k1 ¼ k1′½O2ꢀ
ð1Þ
As a consequence, a second-order rate constant k1′ was
determined from k1 and [O2], allowing a comparison between
the three experiments. A relative good agreement between the
three values of k1′ (≈1 M−1 −1) was thus obtained. Interest-
s
ingly, an unexpected acceleration of the kinetics for the second
reaction (k2) is observed when the O2 concentration is
increased. This is also illustrated by the variation of the
decomposition half-time t1/2 with dioxygen content.
k2 ¼ k2′½O2ꢀ
ð2Þ
The relative constancy of the second-order kinetic value k2′
from k2 and [O2], referring to eqn (2), seems to confirm the
implication of dioxygen in this second reaction (Table 7).
Decomposition through oligomeric intermolecular Cun–O2
(n = 1 or 2) complexes, involving a Cu/O2 stoichiometry higher
than 1/2, may occur as already observed for Cu–TMPA and
Cu2–D1.9 As displayed in Table 8, the kinetics of formation of
the trans-μ-1,2 peroxo intermediate for the Cu2–DYne com-
pound is markedly slower than that reported for Cu–TMPA
and its analogous mononuclear and dinuclear complexes in
similar experimental conditions (nitrile solvents, temperature).
For instance, in CH3CH2CN at comparative temperatures
(183–193 K), the second-order rate constant k1′, which includes
the formation of superoxo and trans-peroxo species, is at least
3 orders of magnitude lower for the mono-yne bisTMPA ligand
DYne than for D1 or TMPA. On the basis of the results
2248 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 9 Electrochemical data for [Cu2DYne(CH3CN)2]4+, [CuTMPA(CH3CN)]2+ and other related TMPA-based Cu complexesa
E1/2/V (ΔEp/mV)
Complex
DMF
Ref.
CH3CN
Ref.
[Cu2DYne(CH3CN)]4+
[CuTMPA(CH3CN)]2+
[CuTMPA(Cl)]+
−0.54 (160)
−0.62 (65)
−0.81 (80)
−0.36c (433)
−0.64e (88)
−0.79 (90)
This work
−0.29 (110)
This work
This work
39
40
38
9
40
40
−0.40 (65)
b
b
b
b
b
[Cu2bTMPA(CH3CN)2]4+
[Cu2D1(CH3CN)2]4+
[Cu2DO(Cl2)]2+
[Cu2isoDO(Cl2)]2+
−0.66 (70)
[Cu(6-MeTMPA)(H2O)]2+
[CuTMPAE(CH3CN)]2+
−0.29d (90)
41
b
−0.57e (88)
9
b
a E1/2 = (Epa + Epc)/2; ΔEp = Epa − Epc; in DMF and CH3CN; E/V vs. Fc+/Fc. b Not determined in this solvent. c Potential recalibrated vs. Fc assuming
E0(Fc+/Fc) = +0.54 V vs. Ag/AgCl in DMF (Table 4 in ref. 38). d Potential recalibrated vs. Fc assuming E0(Fc+/Fc) = +0.38 V vs. Ag/AgCl in CH3CN
(ref. 42, p. 163). e Potential recalibrated vs. Fc assuming E0(Fc+/Fc) = +0.02 V vs. Ag/AgNO3 (ref. 9).
which consequently stabilizes the peroxo species as already appears to be of interest for the future design of bio-inspired
observed in the case of the bispidine ligand through possible dinuclear Cu complexes. More generally, an evolution of this
coordination of the peroxide in the equatorial position (less family of dinucleating ligand bearing asymmetric site and
labile).14 Such tetragonal conformation is corroborated by the diverse linking bridges is still under consideration.
solid-state (XRD) and solution (spectroscopies and voltamme-
try) data in comparison to the trigonal analogous TMPA-based
complexes.9 Interestingly, no supplementary signal was
Experimental
Materials and methods
detected for [Cu2DYne(O2)]2+ at a wavelength value close to
that found for TMPA (λmax = 525 nm).9 This is in strong con-
trast with the D1-based bis-Cu complex which forms an inter-
All solvents were thoroughly degassed before use. 6-Bromo-2-
molecular peroxo species. For the [Cu2DYne(O2)]2+ complex,
pyridinecarboxaldehyde (Maybridge) and 2-aminomethylpyri-
dine (Aldrich) were used as received without purification. The
the thermodynamic species seems thus to be the intramolecu-
starting materials bis(2-pyridylmethyl)(6-bromo-2-pyridyl-
lar complex owing to the stronger rigidity of the alkyne bridge
as compared to the alkane linker of D1. Further work is necess-
methyl)amine (6-BrTMPA), bis(2-pyridylmethyl)(6-ethynyl-2-
pyridylmethyl)amine (6-eTMPA) were prepared according to
ary to fully rationalize this behavior.
the literature or adapted methods.43,44 Syntheses of copper
complexes were carried out using standard Schlenk techniques
under purified dinitrogen. Elemental analyses were performed
Conclusions
by the “Service Central d’Analyses du CNRS”, Gif-sur-Yvette,
In summary, we have described the synthesis and full charac- France. Mass analyses were carried out by the “Service Central
terization of a series of TMPA-based dinuclear Cu complexes d’Analyses du CNRS”, Vernaison, France, on an ESI-TOF spec-
bearing rigid π-delocalized linkers. The spectroscopic and elec- trometer. IR spectra (KBr pellets) were recorded on a Nicolet
trochemical investigations have shown that these ligands are Nexus FT-IR instrument. NMR spectra were recorded on a
not suitable for the development of electronically interacting Bruker DRX 500 or Bruker Avance 300 and 500 MHz, J values
Cu ions through the linker. However in one case, an interest- are given in Hz. Electron Paramagnetic Resonance (EPR)
ing control of the Cu coordination by distant π–π stacking spectra were run on a Bruker Elexsys 500 spectrometer
interactions has been observed. The reactivity of the electro- (X-band). EPR simulation was carried out on the XSophe-
generated CuI complexes towards O2 has revealed very specific Sophe-Xeprview software.45 UV-Vis-NIR spectroscopy was per-
behavior for the complex bearing the mono-alkyne linker formed with a JASCO V-670 spectrophotometer with a specific
between the two TMPA units. An intramolecular binding of O2 probe connected through an optic fiber for low-temperature
is proposed in this case, as the distance between both the Cu measurements in a Schlenk tube. The electrochemical studies
centers, after rotation around the alkyne axis, is of the same in organic solvents were performed in a glovebox (Jacomex) (O2
order of magnitude (4.5 Å) than that reported for classical < 1 ppm, H2O < 1 ppm) with a home-designed 3-electrodes cell
trans-μ-1,2-peroxo adducts. The sluggishness of the peroxo for- (WE: Pt; pseudo-RE: Ag wire; CE: Pt). Acetonitrile (CH3CN)
mation is ascribed to the rigidity of the linker, which induces (99.9% BDH, VWR) and propionitrile (CH3CH2CN) (Sigma-
a rotation of half the molecule for O2 binding, hence involving Aldrich, 99%) were used as received and kept under N2 in the
a kinetic barrier. The peroxo complex is relatively stable at glovebox. NBu4PF6 was synthesized from NBu4OH (Fluka) and
268 K (t1/2 = 240 s). It is likely due to the steric hindrance of HPF6 (Aldrich). It was then purified, dried under vacuum for
the bis-tmpa core and/or a stabilizing tetragonal-like confor- 48 h at 100 °C, then kept under N2 in the glovebox. The poten-
mation. The relative stability of the complex in solution tial of the cell was controlled by an AUTOLAB PGSTAT 302
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2249

View Article Online
Paper
Dalton Transactions
(Ecochemie) potentiostat monitored by a computer. Ferrocene (41), 605.27 (8), 606.25 (1), 607.29 (<1); m/z 625.33 ((M + Na)+,
(Fc) was added at the end of each experiment to determine 100%), 626.31 (42), 627.29 (9), 628.28 (2), 629.27 (<1).
redox potential values. Numerical simulations of the voltam-
6,6′-Diethynyl-bis(tris(2-pyridylmethyl)amine): DDyne. This
mograms were performed with the BAS DigiSim© simulator compound was prepared using a procedure similar to that
3.03 (Bioanalytical Systems), using the default numerical used for ligand DYne by self-condensation of 6-ethynylTMPA
options with the assumption of planar diffusion and a Butler– (100 mg, 0.32 mmol) in a mixture of THF–diisopropylamine
Volmer law for the electron transfer.
(15 : 15 mL) with the palladium(II) complex [PdCl2(PPh3)2]
(13 mg, 0.002 mmol) and CuI (6 mg, 0.032 mmol). The
mixture was stirred at room temperature for 40 h, and then fil-
Syntheses of the ligands
6,6′-Phenyl-bis(tris(2-pyridylmethyl)amine): DPh. The syn- tered off. Solvents were removed under low pressure and the
thesis was previously published by Canary et al.19
solid was solubilized with methanol (15 mL). This solution
6,6′-Anthracenyl-bis(tris(2-pyridylmethyl)amine): DAnt. This was then treated with aqueous HCl (5%) and stirred for
compound was prepared according to the procedure described 15 min. After removal of methanol under reduced pressure,
by Canary et al. by reaction of the 9,10-anthracenyl-diboronic impurities were removed from the aqueous solution by extrac-
acid (100 mg, 0.41 mmol) and the 6-BrTMPA derivative tion with CH2Cl2. An aqueous solution of NaOH (10%) was
(303 mg, 0.82 mmol). Yield: (143 mg, 50%). Anal. Calcd for added until a white solid precipitated (pH > 10). This solid
C50H42N8·0.5 CH2Cl2: C, 76.1; H, 5.4; N, 14.1. Found: C, 76.3; was extracted with CH2Cl2 (3 × 20 mL), the solvent volume was
H, 5.5; N, 13.5. IR (cm−1): 3052w, 3010w, 2980w, 2925w, 2819, then reduced under low pressure before purification by chrom-
1587s, 1568s, 1469s, 1435s, 1383w, 1366w, 1310w, 1251w, 1159 atography on neutral alumina (eluent: CH2Cl2–CH3OH 97 : 3).
w, 1119, 1095w, 994, 977w, 889w, 852w, 801, 764s, 751s, 723w, Yield: (60 mg, 62%). Due to a slight decomposition of the
694w, 634w, 607w, 541w, 518w, 404w. 1H NMR (300 MHz, ligand in the solid, good elemental analysis could not be pro-
CDCl3, δH ppm): 3.99 (8 H, s, N(CH2)2), 4.04 (4 H, s, NCH2), vided. IR (cm−1): 3014w, 2926w, 2821, 2200w (CuC, alkyne),
7.15–7.93 (26 H, m, PyH and PhH), 8.55 (4 H, br s, PyH) ppm. 1714w, 1636w, 1590s, 1569s, 1475m, 1448s, 1438s, 1410w,
13C NMR (75.5 MHz, CDCl3, δC ppm): 60.26 (2 N(CH2)2), 60.46 1366, 1312w, 1246w, 1207w, 1152s, 1091s, 986, 879w, 809,
1
(2 NCH2), 121.56 (2 CH, Py), 122.00 (4 CH, Py), 123.06 (4 CH, 773s, 632, 612, 404. H NMR (500 MHz, CDCl3, δH ppm): 3.87
Py), 125.10 (2 CH, Py), 125.36 (4 CH, Ph), 126.28 (4 CH, Ph), (12 H, s, N(CH2)3), 7.13 (4 H, dd, 3JH,H 7.0, 3JH,H 5.0, PyH), 7.37
3
3
129.83 (2 Cipso), 136.45 (4 CH, Py), 136.74 (2 CH, Py), 149.14 (4 (2 H, dd, JH,H 6.5, 4JH,H 2.0, PyH), 7.53 (4 H, d, JH,H 7.5, PyH),
3
4
CH, Py), 157.65 (2 Cipso), 159.43 (4 Cipso–CH2, Py), 159.90 (2 7.60–7.66 (8 H, m, PyH), 8.52 (4 H, dd, JH,H 4.8, JH,H 0.5,
Cipso). ESI-MS (CH2Cl2–MeOH (1 : 9) + TFA): m/z 755.36 ((M + PyH). 13C NMR (125 MHz, CDCl3, δC ppm): 59.94 (2 NCH2),
H)+, 100%), 756.33 (62), 757.33 (19), 758.36 (3), 759.33 (1).
60.24 (2 N(CH2)2), 73.16 (Py–CuC), 81.09 (Py–Cu), 122.06 (4
6,6′-Ethynyl-bis(tris(2-pyridylmethyl)amine): DYne. To a solu- CH, Py), 123.08 (4 CH, Py), 123.20 (2 CH, Py), 126.76 (2 CH,
tion of 6-BrTMPA (388 mg, 1.050 mmol) and 6-ethynylTMPA Py), 136.44 (4 CH, Py), 136.59 (2 CH, Py), 140.99 (2 Cipso–Cu,
(300 mg, 0.905 mmol) in a degassed mixture of 10 mL of THF Py), 149.16 (4 CH, Py), 159.13 (4 Cipso–CH2, Py), 160.66 (2 Cipso
–
and 5 mL of diisopropylamine were added [PdCl2(PPh3)2] CH2, Py). ESI-TOF-MS (MeOH): Calcd, Found: m/z 649.2804,
(40 mg, 0.057 mmol) and CuI (18 mg, 0.095 mmol). The 649.2833 [(M + Na)]+.
mixture was stirred under an atmosphere of dry nitrogen for
Syntheses of the copper complexes
two days then filtered and solvents were removed under
reduced pressure. The brown residue was extracted with Complex [Cu2DPh(CH3CN)2(CF3SO3)2](CF3SO3)2
1 was syn-
CH2Cl2. The crude product obtained after removal of the thesized following the procedure described by Canary et al. for
solvent under vacuum was purified by chromatography on the corresponding perchlorate salt.19 The structural analysis
neutral alumina (eluent: CH2Cl2–CH3OH 97 : 3). Yield: established that in both cases, the copper complexes display
(406 mg, 74%). Anal. Calcd for C38H34N8·CH2Cl2: C, 74.0; H, corresponding features (see ESI†). Complexes [Cu2DAnt-
5.6; N, 18.1. Found: C, 74.2; H, 5.7; N, 17.8. IR (cm−1): 3053w, (CH3CN)2](CF3SO3)4 2, [Cu2DYne(CH3CN)2(CF3SO3)2](CF3SO3)2
3012w, 2921w, 2879w, 2818, 1653w, 1588s, 1568s, 1473w, 3 and [Cu2DDyne(CH3CN)2](CF3SO3)4 4 were prepared under
1460s, 1434, 1365, 1310w, 1259w, 1155w, 1121, 1084w, 1046w, Schlenk conditions according to the following procedure. One
996w, 979w, 881w, 814, 765s, 737w, 632w, 611w, 582w, 565w, equivalent of the bis-TMPA ligand was added to a solution of
402. 1H NMR (500 MHz, CDCl3, δH ppm): 3.87 (8 H, s, N- two molar equivalents of the CuII salt Cu(CF3SO3)2 in CH3CN
(CH2)2), 3.89 (4 H, s, NCH2), 7.11 (4 H, ddd, 3JH,H 7.5, 3JH,H 5.0, (20 mL). The solution was stirred at room temperature over-
3
4
4JH,H 1.0, PyH), 7.46 (2 H, dd, JH,H 7.0, JH,H 1.5, PyH), 7.53 (4 night, and then filtered. After removal of the solvent under low
3
H, d, JH,H 8.0, PyH), 7.59–7.66 (8 H, m, PyH), 8.51 (4 H, dd, pressure, the crude product was dried under vacuum. All
4
3JH,H 4.8, JH,H 0.5, PyH). 13C NMR (125 MHz, CDCl3, δC ppm): copper salts were then recrystallized by layering either diethyl
60.09 (2 NCH2), 60.18 (2 N(CH2)2), 87.86 (–CuC–), 122.03 (4 ether or ethyl acetate onto a CH3CN solution of the products.
CH, Py), 122.57 (2 CHe Py), 123.00 (4 CH, Py), 126.20 (2 CH, For all the derivatives either microcrystalline or crystalline
Py), 136.44 (4 CH, Py), 136.59 (2 CH, Py), 141.81 (2 Cipso–Cu, samples were obtained in quite good yields.
Py), 149.08 (4 CH, Py), 159.16 (4 Cipso–CH2, Py), 160.29 (2 Cipso
–
For compounds 1 and 2, blue crystals were obtained after
CH2, Py). ESI-MS (MeOH): m/z 603.31 ((M + H)+, 100%), 604.28 crystallization in a CH3CN–EtOAc mixture with 60 and 64%
2250 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
yields respectively. The crystallization in
a
CH3CN–Et2O for the hydrogen atoms attached to the ethoxyethane which
mixture, for compound 3, affords light blue crystals (thin were not calculated, the positions of all the other hydrogen
needles) and a dark blue microcrystalline sample. Since this atoms were calculated for the two compounds and included as
crystallization led to a mixture, it was not possible to provide isotropic fixed contributors to Fc. Pertinent crystal data and
accurate elemental analyses for this sample. With the aim to refinement parameters are given in Table 1.
analyze this derivative, a similar reaction was carried out
according to the general procedure but the crystallization was
performed in a mixture of CH3CN–EtOAc. Compound 3 was
Acknowledgements
obtained in quite a good yield (60%) as a microcrystalline
powder which was dried under vacuum before elemental
analysis. Compound 4 was obtained as a blue polycrystalline
powder (55%) after crystallization in a CH3CN–Et2O mixture.
Compound 2. Anal. Calcd for 2·(CH3CO2CH2CH3)2:
C66H64Cu2F12N10O16S4: C, 45.7; H, 3.7; N, 8.1. Found: C, 45.1;
H, 3.6; N, 7.9. IR (cm−1): 3507 (OH, H2O), 3063w, 3028w,
2939w, 2295w, 1734, 1611, 1572, 1477, 1460, 1447, 1438,
1400w, 1364w, 1279s, 1257s, 1162s, 1104w, 1085w, 1030s,
979w, 945w, 864w, 843w, 820w, 792w, 773, 762, 732w, 637s,
573, 517, 492w, 426w.
This research was supported by CNRS, Ministère de la
Recherche et de l’Enseignement Supérieur (MESR grant for
A. G.) and Agence National pour la Recherche (ANR-05-
BLAN-0003). Dr F. Michaud is thanked for XRD analyses. The
authors thank Pr. O. Reinaud, Pr. I. Jabin and Dr B. Colasson
for helpful discussions. C. Calvarin is thanked for the specific
set-up of the low-temperature UV-Vis cell.
Compound 3. Anal. Calcd for 3·(CH3CO2CH2CH3)0.5·(H2O)2:
C48H48Cu2F12N10O15S4: C, 38.7; H, 3.2; N, 9.4. Found: C, 38.2;
H, 2.9; N, 9.0. IR (cm−1): 3552 (OH, H2O), 3073w, 2927w,
2200w, 1615, 1468, 1442, 1288s, 1258s, 1171s, 1105 w, 1031s,
978w, 888w, 812w, 770, 643s, 576w, 519, 491w, 424w.
Notes and references
1 (a) K. D. Karlin and S. Itoh, Copper-Oxygen Chemistry, J. Wiley
& Sons, Inc., Hoboken, New Jersey, 2011; (b) W. Kaim and
J. Rall, Angew. Chem., Int. Ed. Engl., 1996, 35, 43–60.
Compound 4. As observed for the ligand DDyne, compound
4 gives rise to a slight decomposition in the solid state, which
did not allow providing reliable elemental analyses. IR (cm−1):
3555 (OH, H2O), 3075w, 2927w, 2202w (CuC), 1614, 1464,
1448, 1287s, 1262s, 1168s, 1103w, 1032s, 888w, 810w, 770w,
643s, 575w, 519, 491w, 423w.
2 (a) E. I. Solomon, R. K. Szilagyi, S. DeBeer George and
L. Basimallick, Chem. Rev., 2004, 104, 419–458;
(b) C. Dennison, Coord. Chem. Rev., 2005, 249, 3025–3054;
(c) Y. Lu, Cupredoxins, in Comprehensive Coordination
Chemistry II: From Biology to Nanotechnology, ed. J. McClev-
erty and T. J. Meyer, Elsevier, Oxford, 2003, vol. 8 (Bio-
coordination Chemistry, L. Que, Jr. and W. B. Tolman,
Eds.), pp. 91–122; (d) N. M. Marshall, D. K. Garner,
T. D. Wilson, Y.-G. Gao, H. Robinson, M. J. Nilges and
Y. Lu, Nature, 2009, 462, 113–116; (e) K. M. Lancaster,
S. Sproules, J. H. Palmer, J. H. Richards and H. B. Gray,
J. Am. Chem. Soc., 2010, 132, 14590–14595; (f) S. M. Berry,
M. H. Baker and N. J. Reardon, J. Inorg. Biochem., 2010,
104, 1071–1078; (g) J. P. Klinman, J. Biol. Chem., 2006, 281,
3013–3016; (h) S. T. Prigge, R. E. Mains, B. A. Eipper and
L. M. Amzel, Cell. Mol. Life Sci., 2000, 57, 1236–1259.
3 (a) E. I. Solomon, P. Chen, M. Metz, S.-K. Lee and
A. E. Palmer, Angew. Chem., Int. Ed., 2001, 40, 4570–4590;
(b) I. A. Koval, P. Gamez, C. Belle, K. Selmeczi and
J. Reedijk, Chem. Soc. Rev., 2006, 35, 814–840; (c) S. Itoh
and S. Fukuzumi, Acc. Chem. Res., 2007, 40, 592–600; (d) C.
X. Zhang, H.-C. Liang, K. J. Humphreys and K. D. Karlin, in
Advances in Catalytic Activation of Dioxygen by Metal Com-
plexes, ed. L. I. Simándi, Kluwer Academic Publishers,
Dordrecht–Boston–London, 2003, pp. 79–121.
X-ray crystallographic studies
Data for compounds 2–3 were collected on an Xcalibur 2 CCD
Diffractometer (Oxford Diffraction) using graphite monochro-
mated Mo-Kα radiation (λ = 0.71073 Å). The unit cell determi-
nations and data reductions were performed using the
CrysAlis program suite on the full set of data.46 For the two
compounds, the crystal structures were solved by direct
methods and successive Fourier difference syntheses with the
Sir97 program47 and refined on F2 by weighted anisotropic
full-matrix least-squares methods using the SHELXL97
program.48 Both software were used within the WINGX
package.49 After locating all the non-hydrogen atoms in each
structure, the models were refined, first using isotropic and
finally anisotropic thermal displacement parameters, except
for one of the trifluoromethanesulfonate anions (CF3SO3−) of
compound 2 and the ethoxyethane (CH3–CH2–O–CH2–CH3)
solvent molecule of compound 3 which present orientation
disorders: for compound 2, the two triflate orientations appear
in the asymmetric unit as two nearly perpendicular positions
in a 0.69/0.31 occupation ratio; while in compound 3, the two
disordered positions of the ethoxyethane solvent molecule
appear confused but moved by a distance equivalent to a CH2–
CH3 distance. This model led to the occupation ratio of
4 L. M. Mirica, X. Ottenwaelder and T. D. P. Stack, Chem.
Rev., 2004, 104, 1013–1045.
5 E. A. Lewis and W. B. Tolman, Chem. Rev., 2004, 104, 1047–
1076.
6 A. G. Blackman, Eur. J. Inorg. Chem., 2008, 2633–2647.
7 (a) K. D. Karlin, J. C. Hayes, S. Juen, J. P. Hutchinson
and J. Zubieta, Inorg. Chem., 1982, 21, 4106–4108;
−
0.73 : 0.27. In both cases, all the atoms (CF3SO3 in 2 and
CH3–CH2–O–CH2–CH3 in 3) were refined isotropically. Except
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2251

View Article Online
Paper
Dalton Transactions
(b) P. L. Dedert, J. S. Thompson, J. A. Ibers and 26 For comparison, the hydroxo-form of [CuII2DYne-
T. J. Marks, Inorg. Chem., 1982, 21, 969–977.
8 R. A. Himes and K. D. Karlin, Curr. Opin. Chem. Biol., 2009,
13, 119–131.
9 D. H. Lee, N. Wei, N. N. Murthy, Z. Tyeklár, K. D. Karlin,
S. Kaderli, B. Jung and A. D. Zuberbühler, J. Am. Chem.
Soc., 1995, 117, 12498–12513.
10 K. D. Karlin, D. H. Lee, S. Kaderli and A. D. Zuberbühler,
Chem. Commun., 1997, 475–476.
11 P. Comba, P. Hilfenhaus and K. D. Karlin, Inorg. Chem.,
1997, 36, 2309–2313.
(CH3CN)2]4+ after reaction with a stoichiometric amount of
NBu4OH in CH3CN displays a green solution with similar
spectroscopic features (EPR signal centered at g = 2.21;
UV-Vis absorbance at λ = 365, 647 nm and 801 nm
(shoulder)) as well as an irreversible reduction peak at Epc
=
−0.67 V. This seems to indicate that the final complex
resulting from the reaction of O2 with the reduced form
[CuI2DYne(CH3CN)2]2+ is a hydroxo CuII species. Interest-
ingly, its UV-Vis response is also close to that found for bis-
μ-hydroxo Cuanalogous complexes with a band at ca λ =
350 nm corresponding to an OH → CuII charge transfer
transition.16a,27 See ESI† for experimental data.
12 J. E. Bol, W. L. Driessen, R. Y. N. Ho, B. Maase, L. Que, Jr.
and J. Reedijk, Angew. Chem., Int. Ed. Engl., 1997, 36, 998–
1000.
27 L. Casella, O. Carugo, M. Gulloti, S. Garofani and
P. Zanello, Inorg. Chem., 1993, 32, 2056–2067.
13 H. Börzel, P. Comba, C. Katsichtis, W. Kiefer, A. Lienke,
V. Nagel and H. Pritzkow, Chem.–Eur. J., 1999, 6, 1716– 28 C. Würtele, F. W. Heinemann and S. Schindler, J. Coord.
1721. Chem., 2010, 63, 2629–2641.
14 H. Börzel, P. Comba, K. S. Hagen, M. Kerscher, 29 C. He, J. L. DuBois, B. Hedman, K. O. Hodgson and
H. Pritzkow, M. Schatz, S. Schindler and O. Walter, Inorg.
Chem., 2002, 41, 5440–5452.
S. J. Lippard, Angew. Chem., Int. Ed., 2001, 40, 1484–
1487.
15 (a) Y. Tachi, Y. Matsukawa, J. Teraoka and S. Itoh, Chem. 30 R. R. Jacobson, Z. Tyeklár, A. Farooq, K. D. Karlin, S. Liu
Lett., 2009, 38, 202–203; (b) L. Tahsini, H. Kotani, and J. Zubieta, J. Am. Chem. Soc., 1988, 110, 3690–3692.
Y.-M. Lee, J. Cho, W. Nam, K. D. Karlin and S. Fukuzumi, 31 Both trans-μ-1,2 peroxo and μ:η2:η2 peroxo species are
Chem.–Eur. J., 2012, 18, 1084–1093.
characterized by a maximum UV-Vis absorption peak
between 480 and 550 nm. They however differ by their
molar extinction coefficient at this wavelength (10 000 M−1
cm−1 and 1000 M−1 cm−1 respectively).4.
16 (a) M. Kodera, K. Katayama, Y. Tachi, K. Kano, S. Hirota,
S. Fujinami and M. Suzuki, J. Am. Chem. Soc., 1999, 121,
11006–11007; (b) M. Kodera, Y. Kajita, Y. Tachi,
K. Katayama, K. Kano, S. Hirota, S. Fujinami and 32 (a) L. Q. Hatcher, D.-H. Lee, M. A. Vance, A. E. Milligan,
M. Suzuki, Angew. Chem., Int. Ed., 2004, 43, 334–337.
17 (a) H. R. Lucas, L. Li, A. A. N. Sarjeant, M. A. Vance,
E. I. Solomon and K. D. Karlin, J. Am. Chem. Soc., 2009,
131, 3230–3245; (b) C. Würtele, O. Sander, V. Lutz, T. Waitz,
F. Tuczek and S. Schindler, J. Am. Chem. Soc., 2009, 131,
7544–7545.
R. Sarangi, K. O. Hodgson, B. Hedman, E. I. Solomon and
K. D. Karlin, Inorg. Chem., 2006, 45, 10055–10057;
(b) Y. Lee, D.-H. Lee, G. Y. Park, H. R. Lucas, A. A. Narducci
Sarjeant, M. T. Kieber-Emmons, M. A. Vance,
A. E. Milligan, E. I. Solomon and K. D. Karlin, Inorg. Chem.,
2010, 49, 8873–8885.
18 A. Kunishita, M. Kubo, H. Sugimoto, T. Ogura, K. Sato, 33 The shift of λmax with the solvent was previously detected
T. Takui and S. Itoh, J. Am. Chem. Soc., 2009, 131, 2788–
2789.
for other Cu tripodal tetradentate peroxo complexes. For
instance, Cu–TMPA exhibits λmax = 525 nm and 522 nm in
CH3CH2CN and THF respectively.37 Also, Comba et al.
observed an increase of the absorbance for [Cu2Bisp2Et-
(O2)]4+ when substituting CH3CH2CN by acetone (482 nm
and 488 nm, respectively).14.
19 This ligand and its CuII complex have already been syn-
thesized, but its redox behavior and reactivity towards O2
were not investigated: L. Zhu, O. dos Santos, C. W. Koo,
M. Rybstein, L. Pape and J. W. Canary, Inorg. Chem., 2003,
42, 7912–7920.
34 R. A. Alberty and R. J. Silbey, Physical Chemistry, J. Wiley &
Sons, New York, 2nd edn, 1996.
20 A. W. Addison, T. N. Rao, J. Reedjik, J. van Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1356. 35 The irreversibility of the dioxygen binding to the
21 B. S. Lim and R. H. Holm, Inorg. Chem., 1998, 37, 4898–
4908.
22 (a) C.-L. Chuang, K. Lim, Q. Chen, J. Zubieta and
[CuIDYne]2+ complex at 268 K in CH3CN was confirmed by
addition of PPh3 or oxygen release (vacuum cycling)
experiments.
J. W. Canary, Inorg. Chem., 1995, 34, 2562–2568; 36 K. Uozomi, Y. Hayashi, M. Suzuki and A. Uehera, Chem.
(b) C.-L. Chuang, K. Lim and J. W. Canary, Supramol.
Chem., 1995, 5, 39–43.
23 A. J. Bard and L. R. Faulkner, Electrochemical Methods,
J. Wiley & Sons, New York, 2001.
Lett., 1993, 963–966.
37 C. X. Zhang, S. Kaderli, M. Costas, E.-I. Kim,
Y.-M. Neuhold, K. D. Karlin and A. D. Zuberbühler, Inorg.
Chem., 2003, 42, 1807–1824.
24 M. B. Robin and P. Day, Adv. Inorg. Chem. Radiochem., 38 D.-H. Lee, N. Murthy and K. D. Karlin, Inorg. Chem., 1997,
1967, 10, 247–403. 36, 5785–5792.
25 R. R. Jacobson, Z. Tyeklár, K. D. Karlin and J. Zubieta, 39 N. Le Poul, B. Douziech, J. Zeitouny, G. Thiabaud, H. Colas,
Inorg. Chem., 1991, 30, 2035–2040.
F. Conan, N. Cosquer, I. Jabin, C. Lagrost, P. Hapiot,
2252 | Dalton Trans., 2013, 42, 2238–2253
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
O. Reinaud and Y. Le Mest, J. Am. Chem. Soc., 2009, 131, 44 A. Gomila, N. Le Poul, N. Cosquer, J.-M. Kerbaol,
17800–17807.
J.-M. Noël, M. T. Reddy, I. Jabin, O. Reinaud, F. Conan and
40 B. Lucchese, K. J. Humphreys, D.-H. Lee, C. D. Incarvito,
Y. Le Mest, Dalton Trans., 2010, 39, 11516–11518.
R. D. Sommer, A. L. Rheingold and K. D. Karlin, Inorg. 45 G. R. Hanson, K. E. Gates, C. J. Noble, M. Griffin,
Chem., 2004, 43, 5987–5998.
41 H. Nagao, N. Komeda, M. Mukaida, M. Suzuki and
K. Tanaka, Inorg. Chem., 1996, 35, 6809–6815.
A. Mitchell and S. Benson, J. Inorg. Biochem., 2004, 98, 903–
916.
46 Oxford Diffraction Ltd, Xcalibur CCD system, CrysAlis Soft-
ware system, 2005, Version 1.171.
42 P. Zanello, Inorganic Electrochemistry, Theory, Practice and
Application, The Royal Society of Chemistry, Cambridge, 47 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
2003.
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori
43 (a) C.-L. Chuang, O. dos Santos, X. Xu and J. W. Canary,
and R. Spagna, J. Appl. Crystallogr., 1999, 32, 115–119.
Inorg. Chem., 1997, 36, 1967–1972; (b) M. Merkel, 48 G. M. Sheldrick, SHELX97. Programs for Crystal Structure
D. Schnieders, S. M. Baldeau and B. Krebs, Eur. J. Inorg.
Chem., 2004, 783–790.
Analysis 1997, University of Göttingen, Göttingen, Germany.
49 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2238–2253 | 2253