Steric vs electronic effects and solvent coordination in the electrochemistry of phenanthroline-based copper complexes

Article abstract of DOI:10.1016/j.electacta.2014.07.086

The present exhaustive electrochemical study proposes a rationalization of the redox properties of 1,10-phenanthroline-based copper complexes as a function of i) ligand molecular structure, evidencing the competition between electronic and steric effects of alkyl/aryl substituents, and ii) nature of working medium in terms of both solvent and supporting electrolyte anion. Occupancy of the 2 and 9 positions of the phenanthroline is a powerful tool to modulate the oxidation potentials of this family of complexes in a wide potential range. Solvent molecules play a key role in the metal-centred oxidative electron transfer process (unlike the optical electron transition), acting as ancillary ligands that allow the transition between tetrahedral four-coordinated Cu(I) state to tetragonal five-coordinated Cu(II). Actually clear evidences of the entry of one solvent molecule in the inner coordination sphere of the complexes are proved by the Kolthoff and Lingane method. Proof of ionic couple formation is also found.

Full text of DOI:10.1016/j.electacta.2014.07.086

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Electrochimica Acta 141 (2014) 324–330
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Steric vs electronic effects and solvent coordination in the
electrochemistry of phenanthroline-based copper complexes
Mirko Magnia,∗,1, Alessia Colomboa,∗, Claudia Dragonettia,b, Patrizia Mussinia,1
a Dipartimento di Chimica dell’Università degli Studi di Milano, UdR-INSTM di Milano, Via Golgi 19, I-20133 Milano, Italy
b Istituto di Scienze e Tecnologie Molecolari del CNR, ISTM-CNR, Via Golgi 19, I-20133 Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 June 2014
Received in revised form 14 July 2014
Accepted 15 July 2014
Available online 1 August 2014
The present exhaustive electrochemical study proposes a rationalization of the redox properties of 1,10-
phenanthroline-based copper complexes as a function of i) ligand molecular structure, evidencing the
competition between electronic and steric effects of alkyl/aryl substituents, and ii) nature of working
medium in terms of both solvent and supporting electrolyte anion. Occupancy of the 2 and 9 positions of
the phenanthroline is a powerful tool to modulate the oxidation potentials of this family of complexes
in a wide potential range. Solvent molecules play a key role in the metal-centred oxidative electron
transfer process (unlike the optical electron transition), acting as ancillary ligands that allow the transition
between tetrahedral four-coordinated Cu(I) state to tetragonal five-coordinated Cu(II). Actually clear
evidences of the entry of one solvent molecule in the inner coordination sphere of the complexes are
proved by the Kolthoff and Lingane method. Proof of ionic couple formation is also found.
© 2014 Elsevier Ltd. All rights reserved.
Keywords:
phenanthroline copper complexes
steric effects
electronic effects
solvent coordination
ion pairing
1. Introduction
class of phenanthroline chelate complexes as a function of their lig-
and structures was still overdue. The present study tries to fill the
Copper complexes have been object of a great variety of studies
in many different fields of chemistry [1,2]; one of the very recent
applications of this class of metal complexes regards the sustainable
energetics concerning the optimization of the third generation of
photovoltaic devices, commonly named dye-sensitized solar cells
(DSSCs) [3], in which copper complexes have been employed both
as sensitizers and as redox mediators acting as “sunlight harvester”
and “electron shuttle” between the two cell electrodes respectively
[4,5].
Despite, at now, the best promising metal-based candidates to
overcome the disadvantage of the most used I3-/I− electron media-
tor couple have been cobalt complexes [6,7], copper-based ones
have started showing their potentiality especially with phenan-
throline ligands as demonstrated by Bai [8] and by our recent paper
[9]; the mediator performances appear to be significantly modu-
lated by the phenanthroline structure [5].
gap, investigating electronic and steric effects in a systematic fam-
ily of alkyl- and aryl-substituted phenanthroline complexes, also
considering solvent and supporting electrolyte effects.
2. Experimental
2.1. Synthesis of copper complexes
The ligands L1-L6 and the related hexafluorophosphate copper
(I) complexes 1-6 object of this work are summarized in the Fig. 1.
The symmetric phenanthrolines L1, L2, L5 and L6 were
purchased from Sigma Aldrich. The asymmetric ligands were
synthesized as follow (Scheme 1): 2-mesityl-4,7-dimethyl-1,10-
phenanthroline, L3, [9] was prepared through
a two-step
procedure: i) nucleophilic aromatic substitution of 4,7-dimethyl-
1,10-phenanthroline (Aldrich) with mesityllithium, obtained react-
ing bromomesitylene (Aldrich) with tert-butyllithium (Aldrich)
and ii) rearomatization of the intermediate species; 2-methyl-
1,10-phenanthroline, L4, was synthesized through a condensation
reaction, starting from commercially available 8-amino-quinoline
(Aldrich) according to the Bernhard and coworkers’ procedure
[11]. Copper complexes 1-6 were synthesized starting from
tetrakis(acetonitrile) copper(I) hexafluorophosphate according to
the literature [9,12]. (Scheme 1)
Notwithstanding the rich literature concerning electron trans-
fer at copper centers [10], a comprehensive rationalization from
the electrochemical point of view of the redox properties of this
∗
Corresponding authors. Tel.: +39 02 50314400.
E-mail addresses: mirko.magni@unimi.it (M. Magni), alessia.colombo@unimi.it
(A. Colombo).
1
ISE member.
http://dx.doi.org/10.1016/j.electacta.2014.07.086
0013-4686/© 2014 Elsevier Ltd. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M. Magni et al. / Electrochimica Acta 141 (2014) 324–330
325
a three-electrode cell with a glassy carbon (Metrohm, surface area
0.071 cm2) as working electrode, a saturated calomel electrode as
operative reference one and a platinum foil as counter-electrode
in a solution well-deaerated by nitrogen bubbling. To avoid con-
tamination of the working solution (i.e. water and chloride ions),
the reference electrode was inserted in a glass jacket (ending in a
glass frit, allowing the electrical continuity) filled with the same
solution of the working medium. An instrumental compensation
(i.e. positive feedback) of the ohmic drop between the working
and reference electrode was carefully performed. The recorded
potentials were all referred to the intersolvental reference redox
couple ferricenium/ferrocene, Fc+/Fc [14]; in our conditions the
half-wave potential of the Fc+/Fc couple, E1/2 Fc+/Fc, was ca. 0.48 V,
0.39 V, 0.44 V vs SCE in N,N’-dimethylformamide, acetonitrile and
dichloromethane respectively.
All solvents (N,N’-dimethylformamide, DMF; acetonitrile, ACN;
dichloromethane, DCM) and tetrabutylammonium-based support-
ing electrolytes (TBABPh4; TBAPF6; TBAClO4) were purchased from
Sigma Aldrich as electrochemical grade chemicals.
UV-visible absorption spectra were recorded using a Jasco V-530
spectrophotometer.
2.2.1. Kolthoff and Lingane method
To evaluate the possible solvent involvement as ancillary lig-
and in copper complexes, an adapted version was applied of
the classical Kolthoff and Lingane approach [15]. Oxidative cyclic
voltammetry patterns were recorded for the complexes in the low
polar solvent DCM as such and with increasing additions of small
volumes of polar DMF (in the 1:1 to 110:1 molar ratio range,
with respect to the copper complex). The negative shifts of the
half-wave potential for the Cu+/Cu2+ process, normalized by k/n
(k = Nernst constant, n = exchanged electrons per mole of complex),
were plotted against logcDMF yielding straight lines with slopes and
intercepts accounting for complexation numbers of DMF and for-
mation constants of DMF: complex adducts, respectively. In order
to rule out possible potential drift effects of the SCE reference elec-
trode during the progressive DMF additions to the working medium
(DCM + TBAPF6 0.1 M), we validated the results adding, during a
second repeated measurement series, an equimolar amount of
Fig. 1. Molecular structures of the hexafluorophosphate Cu(I) complexes 1-6 and
the related ligands L1-L6.
2.2. Electrochemical and spectrophotometric measurements
The experimental apparatus and protocol for electrochemical
study were the same reported in a previous work [13], employing
Scheme 1. Top: reaction scheme for the synthesis of L3 ligand [9]: i) mesityllithium, toluene, 0 ◦C, r.t., 20 h; ii) MnO2 activated, toluene/DCM 5:3, r.t., 20 h. Middle: reaction
pathway for the preparation of the L4 phenanthroline [11]. Bottom: general procedure for the synthesis of copper complexes 1-6 [9,12].

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
326
M. Magni et al. / Electrochimica Acta 141 (2014) 324–330
Table 1
Electrochemical data for free-phenanthrolines L1-L6 and related complexes 1-6 in DMF.a
species
Ep,I Rd/V
|∂Ep,I Rd/∂log␯|/V
ꢀEp,I
L-C/V b
Ep,II Rd/V
|∂Ep,II Rd/∂log␯|/V
Ep,Ox/V
|∂Ep,Ox/∂log␯|/V
Rd
L1
L2
L3
L4
L5
L6
1
2
3
4
5
−2.54
−2.59
−2.57
−2.59
−2.62
−2.43
−2.06
−2.19
−2.26
−2.09
−2.13
−2.01
0.032
0.025
0.025
0.026
0.042
0.038
0.015
0.023
0.005
0.006
0.005
0.003
−2.72
−2.80
−2.76
−2.82
−2.80
−2.56
−2.23
−2.34
−2.48
−2.29
−2.35
−2.20
0.028
0.002
0.004
0.015
0.031
0.014
0.011
0.005
0.001
0.002
0.003
0.006
−0.48
−0.40
−0.31
−0.49
−0.49
−0.42
−0.31
−0.36
−0.13
−0.04
0.21
0.011
0.020
0.021
0.048
0.016
0.025
6
0.22
a
Well deaerated solutions of 1·10−3 M sample concentration with TBAPF6 0.1 M as supporting electrolyte; potential scan rate 0.02 ÷ 2 V s−1 on GC electrode. Ep referred to
Fc+/Fc redox couple.
b
ꢀEp,I
stands for Ep,I Rd Ligand–Ep,I
.
Rd Complex
L-C
Rd
ferrocene to the copper complex solution; the half-wave potential
of the Fc+/Fc couple remained invariant after all the DMF additions.
reduced charge density on the ligands upon copper chelation. Such
Ep,I Rd Ligand–Ep,I difference can be regarded as an index
Rd Complex
of the complexation capability of each ligand and suggests that the
ligand L3 is characterized by the weakest complexation strength
(Table 1), tentatively assigned to the high steric hindrance offered
by the mesityl ring.
3. Results and discussion
3.1. Ligands
The anodic window is characterized by a single redox pro-
cess that can be attributed to Cu2+/Cu+ metal-centred chemically
reversible and electrochemically quasi-reversible electron trans-
fer (Figs. 2 and 3 and Table 1). Differently from the ligand-based
reactions, this process is mainly influenced by steric effects of the
chelating ligands, overshadowing electronic ones. In fact, only in
the case of 2, with methyl groups in external 4 and 7 positions, the
expected electron donating effect is observed, smoothed respect to
the case of the (closer) ligand-based reduction.
Instead substituents placed in the internal phenanthroline posi-
tions (i.e. adjacent to its nitrogen atoms) chiefly influence the
behaviour of the Cu2+/Cu+ redox couple through their steric hin-
drance, affecting the geometry around the metal centre. Respect
to previous works [19,20], thanks to the synthesis of asymmetric
phenanthrolines (i.e. L3 and L4), we are now able to demonstrate
that the energetics of the electron transfer occurring at the metal
core is strictly proportional to the number of occupied internal posi-
tions, making possible a fine tuning of the half-wave potential of
the Cu2+/Cu+ couple. In fact considering the oxidation half-wave
potentials the complexes appear clearly partitioned in three dis-
tinct groups (Figs. 2 and 3): 1 and 2 (no internal substituent), 3 and
4 (one internal substituent), 5 and 6 (two internal substituents).
Comparing 1, 4 and 5 complexes, the subsequent addition of methyl
groups in the 2 and 9 positions regularly shifts the anodic peak
at more positive potentials by ca. 0.25 V per each substituent, in a
direction opposite to that virtually induced by their electron releas-
ing nature.
All the six phenanthrolines are stable upon oxidative cycling but
reveal two reduction peaks (Table 1 and Fig. 2), the first of which,
monoelectronic, electrochemically quasi-reversible and chemically
irreversible, can be tentatively attributed to the formation of a rad-
ical anion species as suggested in some literature works [16].
Referring to the unsubstituted phenanthroline, L1, addition of
a methyl group in position 2, L4, and a second one in position 9,
L5, occurs in a progressive negative shift of the first peak potential
(0.03-0.05 V per each methyl) according to the electron releasing
nature of the alkyl group. The same effect, albeit smoothed (ca.
0.025 V per methyl group), is observed when the same substituents
functionalize the positions 4 and 7, L2, suggesting a greater dis-
tance of these from the reduction redox site. Upon addition of
two phenyl groups in L6 the extension of the ␲-conjugated sys-
tem, bringing to an energy stabilization of the LUMO energy, neatly
prevails on the above inductive effect, resulting in a ca. 0.2 V posi-
tive shift of the reduction peaks respect to L5 analogue. Instead in
L3 the addition of a mesityl group only results in a 0.02 V positive
shift of the peak potential respect to its L2 analogue; this posi-
tive shift is much lower than that expected for a single aryl group
(ca. 0.1 V, vide infra, considering the nearly unperceivable effects
of mono-methylation on the Hammett constants of the phenyl
substituent [17]), suggesting some torsional angle between the
aromatic substituent and the phenanthroline core that hampers ␲
conjugation.
Interestingly, contrary to expectations, the entity of the stabi-
lization of the cuprous state is mainly due to the number of internal
substituents, while their nature and even their actual steric hin-
drance appear less influential.
3.2. Complexes
All complexes exhibit two reduction peaks approaching elec-
trochemical and chemical reversibility (Fig.
2 and Table 1).
The powerful steric effect of substituents in positions 2 and 9
confirms the metal-centred nature of the redox process occurring in
the anodic window, during which the metal has to switch between
Cu(I) and Cu(II) state characterized by different preferred geome-
tries: tetrahedral, with a 4-coordinate environment for the Cu(I)
state and distorted tetragonal, with a 5- or 6-coordinate geometry
for the Cu(II) one [10,21,22]. However, in our case, the presence of
one or two “kiss lock” substituents in 2,9 positions could increas-
ingly hamper such transition, forcing Cu(II) to maintain a geometry
similar to the preferred Cu(I) one; this would result in destabiliza-
tion of the Cu(II) product, and therefore in a less thermodynamically
Convolution analysis points to each of them involving the exchange
of one electron per complex molecule, by comparison with
the surely monoelectronic oxidation peak corresponding to the
Cu2+/Cu+ couple. Accordingly, the two peaks should correspond to
the formation of two stable radical anions on each phenanthroline
group, behaving as equivalent but reciprocally interacting redox
sites [18].
Metal complexation not only results in radical anion stabiliza-
tion (i.e. improved chemical reversibility), but also in a neat positive
shift of the reduction peak potentials with respect to the cor-
responding free-ligand cases, which is easily attributable to the

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M. Magni et al. / Electrochimica Acta 141 (2014) 324–330
327
1
2
3
4
5
6
1 unit
free-ligands
complexes
-3.3
-2.8
-2.3
-1.8
-1.3
-0.8
-0.3
0.2
0.7
(Fc+/Fc)/V
E vs
Fig. 2. Synopsis of normalized cyclovoltammograms of free-ligands L1-L6 (black thin lines) and the corresponding copper complexes 1-6 (blue thick lines) on GC electrode
in DMF with TBAPF6 0.1 M. Sample concentration 0.001 M; scan rate 0.2 V s−1
.
favoured process, consistently with the remarkable positive shift of
the reversible cyclic voltammetric peak.
up to 440÷480 nm (Table 2), related to a metal-to-ligand charge
transfer transition, MLCT, [23] more intense for the complex 6 due
to the more extended ␲–system, guaranteed by the two phenyl
rings, that also induces a decrease of the HOMO-LUMO energy gap
responsible for its absorption at the highest wavelength. For all
complexes the UV-visible spectra do not show any solvatochromic
effect, passing from DMF, a polar solvent, to DCM, a solvent of low
polarity (Table 2, Fig. 5).
3.2.1. The role of the solvent and of the supporting electrolyte
anion
A comparative spectroscopic and electrochemical study has also
been performed in different media.
From the spectroscopic point of view all the complexes exhibit
only one broad absorption band in the visible region (Fig. 4), ranging
On the contrary, and consistently with former literature [10],
the solvent appears clearly non-innocent in the Cu2+/Cu+ electron
DMF
DCM
1
2
3
4
5
1 unit
6
-1.0
-0.8
-0.6
-0.4
-0.2
E vs
0.0
0.2
0.4
0.6
0.8
1.0
(Fc+/Fc)/V
Fig. 3. Comparison of normalized cyclovoltammograms of 1-6 complexes recorded in two solvents of decreasing polarity: DMF (blue lines) and DCM (green lines). Supporting
electrolyte: TBAPF6 0.1 M; sample concentration 0.001 M; GC electrode; scan rate 0.2 V s−1
.