CF3 Substitution of [Cu(P^P)(bpy)][PF6] Complexes: Effects on Photophysical Properties and Light-Emitting Electrochemical Cell Performance

Article abstract of DOI:10.1002/cplu.201700501

Herein, [Cu(P^P)(N^N)][PF₆] complexes (P^P=bis[2-(diphenylphosphino)phenyl]ether (POP) or 4,5-bis(diphenylphosphino)-9,9-dimethylxanthene (xantphos); N^N=CF₃-substituted 2,2′-bipyridines (6,6′-(CF₃)₂bpy, 6-CF₃bpy, 5,5′-(CF₃)₂bpy, 4,4′-(CF₃)₂bpy, 6,6′-Me₂-4,4′-(CF₃)₂bpy)) are reported. The effects of CF₃ substitution on their structure as well as their electrochemical and photophysical properties are also presented. The HOMO–LUMO gap was tuned by the N^N ligand; the largest redshift in the metal-to-ligand charge transfer (MLCT) band was for [Cu(P^P){5,5′-(CF₃)₂bpy}][PF₆]. In solution, the compounds are weak yellow to red emitters. The emission properties depend on the substitution pattern, but this cannot be explained by simple electronic arguments. Among powders, [Cu(xantphos){4,4′-(CF₃)₂bpy}][PF₆] has the highest photoluminescence quantum yield (PLQY; 50.3 %) with an emission lifetime of 12 μs. Compared to 298 K solution behavior, excited-state lifetimes became longer in frozen Me-THF (77 K; THF=tetrahydrofuran), thus indicating thermally activated delayed fluorescence (TADF). Time-dependent (TD)-DFT calculations show that the energy gap between the lowest-energy singlet and triplet excited states (0.12–0.20 eV) permits TADF. Light-emitting electrochemical cells (LECs) with [Cu(POP)+(6-CF₃bpy)][PF₆], [Cu(xantphos)(6-CF₃bpy)][PF₆], or [Cu(xantphos){6,6′-Me₂-4,4′-(CF₃)₂bpy}][PF₆] emit yellow electroluminescence. The LEC with [Cu(xantphos){6,6′-Me₂-4,4′-(CF₃)₂bpy}][PF₆] had the fastest turn-on time (8 min), and the LEC with the longest lifetime (t_1/2=31 h) contained [Cu(xantphos)(6-CF₃bpy)][PF₆]; these LECs reached maximum luminances of 131 and 109 cd m^?2, respectively.

Full text of DOI:10.1002/cplu.201700501

Accepted Article
Title: CF3 Substitution of [Cu(P^P)(bpy)][PF6] complexes: Effects on
Photophysical Properties and Light-emitting Electrochemical Cell
Performance
Authors: Catherine E. Housecroft, Sarah Keller, Fabian Brunner, José
M. Junquera, Antonio Pertegás, Maria-Grazia La-Placa,
Alessandro Prescimone, Edwin C. Constable, Henk J. Bolink,
and Enrique Ortí
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPlusChem 10.1002/cplu.201700501
Link to VoR: http://dx.doi.org/10.1002/cplu.201700501
A Journal of
www.chempluschem.org

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
CF3 Substitution of [Cu(P^P)(bpy)][PF6] complexes: Effects on
Photophysical Properties and Light-emitting Electrochemical Cell
Performance
Sarah Keller,[a] Fabian Brunner,[a] José M. Junquera-Hernández,[b] Antonio Pertegás,[b] Maria-Grazia
La-Placa,[b] Alessandro Prescimone,[a] Edwin C. Constable,[a] Henk J. Bolink,[b] Enrique Ortí*[b] and
Catherine E. Housecroft*[a]
displays.1,2,3 Light-emitting electrochemical cells (LECs) are
Abstract: We report [Cu(P^P)(N^N)][PF6] complexes with P^P =
bis(2-(diphenylphosphino)phenyl)ether
(POP)
or
4,5-
less well established, but are emerging devices which, like
OLEDs, are based on the principle of electroluminescence.
Compared to OLEDs, LECs are simpler in setup, more
straight-forward in their processing and therefore also
cheaper in production.4 Whereas polymer-based or purely
organic emitting materials are known for both LECs and
OLEDs, the employment of transition metal complexes has
certain advantages. Depending on the combination of
different ligands and their substitution with functional
groups, the properties of these metal complexes can be
tuned in terms of emission colour, quantum yield and
excited state lifetime.5 , 6 For example, for [Ir(ppy)2(bpy)]+
(Hppy = 2-phenylpyridine, bpy = 2,2'-bipyridine) type
complexes, different emission colours, photoluminescence
quantum yields (PLQY) and device performances have
been obtained upon modification of the cyclometallating
Hppy or the ancillary bpy ligands.5,6 Whereas iridium-based
emitters can be extremely efficient,7,8,9 their replacement by
copper-based compounds has the advantage of higher
abundance and therefore lower marked price of copper
compared to iridium, which also translates to the costs of
the devices.10 Furthermore, many copper complexes are
proven to exhibit thermally activated delayed fluorescence
(TADF), a mechanism which allows the thermal population
of the energetically higher singlet excited state from the
triplet excited state. As a consequence, emission processes
from all excited states are possible and in theory allow for
bis(diphenylphosphino)-9,9-dimethylxanthene (xantphos), N^N
=
CF3-substituted 2,2'-bipyridines (6,6'-(CF3)2bpy, 6-CF3bpy, 5,5'-
(CF3)2bpy, 4,4'-(CF3)2bpy, 6,6'-Me2-4,4'-(CF3)2bpy). We present the
effects of CF3 substitution on structures, and electrochemical and
photophysical properties. The HOMO–LUMO gap is tuned by the
N^N ligand; the largest redshift in the MLCT band is for
[Cu(P^P)(5,5'-(CF3)2bpy)][PF6]. In solution, the compounds are weak
yellow to red emitters. The emission properties depend on the
substitution pattern but this cannot be explained by simple electronic
arguments. For powders, [Cu(xantphos)(4,4'-(CF3)2bpy)][PF6] has
the highest PLQY (50.3%) with an emission lifetime of 12 µs.
Compared to 298 K solution behaviour, excited state lifetimes
lengthen in frozen Me-THF (77 K) indicating thermally activated
delayed fluorescence (TADF). TD-DFT calculations show that the
energy gap between the lowest-energy singlet and triplet excited
states (0.12–0.20 eV) permits TADF. LECs with [Cu(POP)(6-
CF3bpy)][PF6], [Cu(xantphos)(6-CF3bpy)][PF6] or [Cu(xantphos)(6,6'-
Me2-4,4'-(CF3)2bpy)][PF6] emit yellow electroluminescence. A LEC
with [Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6] had the fastest
turn-on time (8 min); the LEC with the longest lifetime (t1/2 = 31 h)
contained [Cu(xantphos)(6-CF3bpy)][PF6]; these LECs reached
maximum luminances of 131 and 109 cd m–2.
Introduction
The development of solid-state lighting based on the widely
distributed organic light-emitting diodes (OLEDs) has had a
massive impact on technology, especially for screens and
quantum yields up to 100% to be reached.11 ,12,13 ,14
We have previously shown for [Cu(P^P)(bpy)][PF6]
complexes
diphenylphosphinophenyl)ether
(where
P^P
(POP)
=
bis(2-
4,5-
or
[a]
S. Keller, F. Brunner, Dr. A. Prescimone, Prof. Dr. E.C. Constable,
Prof. Dr. C.E. Housecroft
Department of Chemistry
bis(diphenylphosphino)-9,9-dimethylxanthene (xantphos)),
that the PLQY of the complex and the efficiency and lifetime
in LEC devices are increased upon addition of methyl or
ethyl groups in one or both 6-positions of the bpy ligand.15,16
Costa et al.17 systematically studied the effect of electron
donating and electron withdrawing substituents at the 4-
positions of bpy in [Cu(P^P)(bpy)][PF6] complexes. The
more negative the σ-Hammett parameter σp, which
describes the σ-donation ability of a given substituent, the
more enhanced is the performance of the LEC employing
University of Basel
BPR 1096, Mattenstrasse 24a, Basel 4058, Switzerland
E-mail: catherine.housecroft@unibas.ch
Dr. J.M. Junquera, Dr. A. Pertegás, M.-G. La-Placa, Dr. H.J. Bolink,
Prof. Dr. E. Ortí
Instituto de Ciencia Molecular,
Universidad de Valencia,
[b]
ES-45980 Paterna, Valencia, Spain
e-mail: enrique.orti@uv.es
the respective compound, within
complexes. We have demonstrated that the incorporation of
a given series of
Supporting information for this article is given via a link at the end of
the document.
1
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
remote fluoro groups in [Cu(POP)(N^N)][PF6] and
with C2v symmetry for the complex cations containing
[Cu(xantphos)(N^N)][PF6], where N^N
=
4,4'-bis(4-
symmetrically substituted bpy ligands.16 The base peaks in the
fluorophenyl)-6,6'-dimethyl-2,2'-bipyridine, leads to an
enhancement of solid state and solution photophysical
properties and is also beneficial to LEC performance.18 This
finding prompted us to investigate the potential positive
effects that the introduction of CF3 substituents may also
have. Substitution with one or more trifluoromethyl groups
is a common motif in coordination chemistry, especially for
N,N'-chelating ligands incorporating pyrrole, pyrazole,
triazole and tetrazole rings, as discussed for pyridyl
azolates by Y. Chi et al.19 Modification with CF3 groups is
often employed in luminescent materials containing Cu(I),
Ir(III) and Pt(II) to Ru(II) and Os(II) complexes. However,
CF3-modified 2,2'-bipyridines are rarely mentioned in the
literature, and copper complexes coordinated by a CF3-
substituted bpy are even more scarce.20 The molar volume
of a CF3 group is significantly larger than for a methyl group
and the steric effect is often comparable to that of an
isopropyl group.21 Furthermore, the electronic properties of
methyl and CF3 differ in that the former acts as a weak σ-
donor whereas the latter has electron-withdrawing
electrospray
mass
spectra
match
the
respective
[Cu(P^P)(N^N)]+ cations, with isotope patterns agreeing with the
calculated ones. Elemental analysis was performed to confirm
the purity of the bulk compounds.
5
5
4
3
4
3
C
C
O
O
PPh2
PPh2
PPh2
PPh2
xantphos
Ph = ring D
POP
4
4
4
3
3
3
5
5
5
5
B
A
A
N
N
N
N
6
6
4
CF3
bpy
6-CF3bpy
4
3
3
properties and therefore
parameter σp than alkyl groups.
Our aim in the present investigation was to prepare a
series of copper(I) complexes of the type
a more positive σ-Hammett
B
F3C
CF3
B
N
N
N
N
6
F3C
CF3
6,6'-(CF3)2bpy
5,5'-(CF3)2bpy
F3C CF3
[Cu(POP)(bpy)][PF6] and [Cu(xantphos)(bpy)][PF6] with bpy
ligands that are substituted with CF3 groups in the 4-, 5- or
6-positions. The chemical structure of the POP, xantphos
and bpy ligands used is given in Scheme 1. The
electrochemical and photophysical properties of the new
complexes are compared with those employing
unsubstituted bpy as model compounds and with the Cu(I)
complexes with alkyl-substituted bpy ligands,16 and are
interpreted with the help of density functional calculations.
Those complexes with more promising photophysical
F3C
CF3
3
3
5
5
B
N
B
N
N
N
6
6,6'-Me2-
4,4'-(CF3)2bpy
4,4'-(CF3)2bpy
Scheme 1. Structures of ligands with ring and atom labels for NMR
spectroscopic assignments.
properties are tested in LEC devices
.
Because of the constraints of the xanthene unit, the
xantphos ligand is less sterically demanding than POP. As
a consequence, the addition of the POP to [Cu(MeCN)4]+
leads to [Cu(POP)(MeCN)]+ or [Cu(POP-P,P')(POP-κ1P)]+.
On the other hand, xantphos reacts with [Cu(MeCN)4]+ to
give [Cu(xantphos-κ2P)2]+.23 This difference in behaviour
leads to varying approaches to the preparation of
Results and Discussion
Synthesis, stability and characterization of
[Cu(P^P)(N^N][PF6] complexes
The [Cu(P^P)(N^N)][PF6] complexes with P^P = POP and
xantphos and N^N = bpy, 6-CF3bpy, 5,5'-(CF3)2bpy and 4,4'-
(CF3)2bpy, as well as the [Cu(xantphos)(6-Mebpy)][PF6],
[Cu(xantphos)(6,6'-Me2bpy)][PF6] and [Cu(xantphos)(6,6'-Me2-
4,4'-(CF3)2bpy)][PF6] complexes (see Scheme 1 for ligands),
were synthesized following the standard procedures15,16,22 and
were isolated as bright yellow to orange solids with yields of 52
to 96%. The formation of heteroleptic [Cu(P^P)(bpy)][PF6]
complexes was confirmed by one- and two-dimensional NMR
spectroscopic techniques (1H, 31P, 19F, 13C, COSY, NOESY,
HMQC, HMBC), which allowed for the unambiguous assignment
of all signals. On the NMR spectroscopic timescale, the spectra
of the compounds dissolved in acetone-d6 are in accordance
[Cu(POP)(N^N)][PF6]
and
[Cu(xantphos)(N^N][PF6].
Whereas for complexes with POP, the bpy ligand was
added after an initial reaction of POP and [Cu(MeCN)4][PF6]
in CH2Cl2 (sequential addition),15,16 for complexes with
xantphos, a mixture of both the bisphosphane and the N^N
ligand was added to the solution of [Cu(MeCN)4][PF6] to
avoid formation of [Cu(xantphos)2]+ and facilitate the
formation of heteroleptic [Cu(xantphos)(bpy)]+.16
Substitution at the 6,6'-positions of the bpy ligand has a
significant effect on both the photophysics and the stability
of the [Cu(P^P)(bpy)]+ complexes. Substituents in 6,6'-
positions shield the copper(I) centre from solvent attack and
2
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
therefore reduce quenching by avoiding the so-called
“solvent-related excited-state relaxations”.12,26 In general,
large sterically demanding ligands in copper(I) complexes
prevent a geometrical rearrangement of the tetrahedral
cation towards more flattened structures upon excitation
and thus help elongating excited state lifetimes. This
structural effect was studied in detail for [Cu(phen)2]+
complexes.24 ,25 ,26 However, if the substituents in the 6,6'-
positions are too large or electronically repulsive, the
exclusive formation of heteroleptic [Cu(P^P)(N^N)]+ is not
achievable and, instead, mixtures of homoleptic [Cu(N^N)2]+
and [Cu(xantphos)2]+ are obtained. This phenomenon has
also been observed for complexes with phenanthrolines of
different steric demand in the 2,9-positions together with a
series of P^P chelating ligands.27 Interestingly, substituents
in the 4,4'-positions of the bpy ligand also appear to have
an influence on the ligand redistribution (Scheme 2). For
6,6'-Me2bpy, with 1.0 equivalents of xantphos and 1.2
equivalents of POP, respectively, an exclusive formation of
the heteroleptic [Cu(P^P)(bpy)]+ complexes was achieved
(removal of excess P^P chelating ligand by subsequent
layer crystallization (CH2Cl2/Et2O).16 In the case of 6,6'-Me2-
4,4'-(CF3)2bpy pure heteroleptic [Cu(xantphos)(6,6'-Me2-
4,4'-(CF3)2bpy)][PF6] was obtained with 1.2 equivalents of
xantphos (and following recrystallization to remove excess
xantphos). However, the analogous reaction with POP
leads to a mixture of free 6,6'-Me2-4,4'-(CF3)2bpy and
[Cu(POP)2][PF6] with heteroleptic [Cu(POP)(6,6'-Me2-4,4'-
(CF3)2bpy)][PF6], which we were not able to isolate without
the side products and was therefore not characterized
further (see Figure S1–S3 for NMR spectra).
spectroscopic data for the crude product showed a mixture
of [Cu(xantphos)(MeCN)2]+ and [Cu(6,6'-(CF3)2bpy)2][PF6].28
In the electrospray mass spectrum, peak envelopes at m/z
641.3 and 1219.7 were assigned to [Cu(xantphos)]+ and
[Cu(xantphos)2]+,
but
no
ion
attributed
to
[Cu(xantphos)(6,6'-(CF3)2bpy)]+ was observed. Layer
recrystallization (CH2Cl2/Et2O) of the crude material led to a
mixture of colourless (dominant) and orange crystals, which
could be manually separated. While the orange crystals
were identified as homoleptic [Cu(6,6'-(CF3)2bpy)2][PF6],28
analysis of the colourless crystals showed the formation of
a one-dimensional coordination polymer [{Cu(xantphos)(µ-
PO2F2)}n], with the copper centres linked by µ-PO2F2 units
which stem from partial hydrolysis of the [PF6]− anion.29
Closer inspection of the spatial properties of the CF3
group helps us to understand why the formation of the
heteroleptic [Cu(P^P)(6,6'-(CF3)2bpy)][PF6] complexes was
not successful. Whereas the calculated van der Waals
radius for a methyl group lies between 1.715 and 2.230 Å,
the reported values for a CF3 group are between 2.107 and
2.743 Å. 30 Although the van der Waals radius is a
reasonable parameter to compare, it is only an intrinsic
property. In order to determine the steric effect, which is an
extrinsic phenomenon, coulombic interactions between all
the atoms or groups involved in the interaction have to be
taken into consideration. A number of different approaches
to analyse the steric effect of typical substituents show that
a CF3 group is comparable with an isopropyl group.21,31
Considering that two ethyl groups in the 6,6'-positions at the
bpy are already too sterically demanding to allow the
exclusive formation of heteroleptic [Cu(P^P)(6,6'-Et2bpy)]+
cations, the steric requirements of two CF3 groups are way
beyond what these type of heteroleptic [Cu(P^P)(bpy)][PF6]
complexes can offer.
P
P
P
P
P
P
N
N
N
N
N
N
Cu
Cu
Cu
+
In
the
case
was
of
unsubstituted
bipyridine,
and
contrast,
[Cu(POP)(bpy)][PF6]
characterized.Error!
previously synthesized
Bookmark
not
defined.
In
Scheme 2. Ligand redistribution results in an equilibrium between heteroleptic
and homoleptic cations.
[Cu(xantphos)(bpy)]+ has only been reported as the [BF4]− salt.22
We
decided
that
[Cu(POP)(bpy)][PF6]
and
[Cu(xantphos)(bpy)][PF6] would serve well as reference
complexes to compare the effects of attaching CF3 and methyl
The sensitivity of the ligand redistribution equilibrium
(Scheme 2) towards substituents in the 6,6'-positions of the
bpy ligand was again corroborated by the unsuccessful
groups
in
different
positions
in
the
bpy
ligand.
[Cu(xantphos)(6,6'-Me2bpy)][PF6]
and
[Cu(xantphos)(6-
Mebpy)][PF6] were investigated in an earlier study16 and are
included here for comparative purposes.
attempts
to
isolate
[Cu(P^P)(6,6'-(CF3)2bpy)][PF6]
complexes. NMR spectra of the crude product of the
reaction of [Cu(MeCN)4][PF6] with POP (1.0 as well as 1.1
equivalents) in CH2Cl2 and subsequent addition of 6,6'-
(CF3)2bpy identify the material as
a
mixture of
[Cu(POP)2][PF6], [Cu(POP)(MeCN)2][PF6] and [Cu(6,6'-
(CF3)2bpy)2][PF6].28 In the mass spectrum, the base peak at
m/z 601.2 was assigned to [Cu(POP)]+, and no peak
envelope arising from [Cu(POP)(6,6'-(CF3)2bpy)]+ was
detected.
The attempted synthesis of [Cu(xantphos)(6,6'-
(CF3)2bpy)][PF6] from [Cu(MeCN)4][PF6], 6,6'-(CF3)2bpy and
xantphos (1.0 as well as 1.2 equivalents) in CH2Cl2 yielded
a
pale orange solid upon solvent removal. NMR
3
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
where the angle comes closest to 90° are [Cu(POP)(bpy)]+
(88.52°) and [Cu(xantphos)(6-Mebpy)]+ (87.92°). It appears that
the dihedral angle is predominantly influenced by packing effects
since no clear trend within the series of bpy ligands or upon
exchange of POP and xantphos could be identified. The dihedral
N–C–C–N angle defining the interring rotation of the bpy ligands
varies from no torsion at all (0(1)°) for [Cu(POP)(4,4'-
(CF3)2bpy)]+ to
a
significant torsion of 20.5(2)° for
[Cu(xantphos)(bpy)]+. It is worth pointing out that the parameters
for [Cu(xantphos)(6-CF3bpy)]+ are very close to those of
[Cu(xantphos)(6-Mebpy)]+, which is also illustrated in the
structure overlay of the two cations in Figure 1.
Table 1. Comparison of structural parameters of the [Cu(P^P)(N^N)]+ cations
P–Cu–P
chelating
angle / deg
N–Cu–N
chelating
angle /
deg
Angle
N–C–C–
N torsion
angle
Complex cation
Figure 1. Overlay of the molecular structures of [Cu(xantphos)(6-Mebpy)]+
(light blue) and the major conformation of [Cu(xantphos)(6-CF3bpy)]+ (purple)
with ellipsoids plotted at 50% probability level. Only the ipso-C atoms of the
PPh2 phenyl rings are shown and H atoms are omitted, with exception of the
methyl group at the bipyridine to allow for a better comparison with the CF3
group. The Cu atoms, pairs of corresponding N atoms and corresponding P
atoms were overlaid.
between
P–Cu–P
and N–
Cu–N
planes /
deg
/deg
[Cu(POP)(bpy)]+ [a]
119.47(3)
113.816(14)
115.68(3)
113.55(3)
79.66(7)
79.32(5)
80.35(10)
79.93(9)
88.52
79.63
79.03
86.61
–2.8(3)
20.5(2)
–18.3(4)
0.6(4)
Structural characterization
[Cu(xantphos)(bpy)]+
[Cu(POP)(6-CF3bpy)]+
X-ray quality crystals of [Cu(xantphos)(bpy)][PF6], [Cu(POP)(6-
CF3bpy)][PF6]·1.3Et2O·0.35H2O,
CF3bpy)][PF6]·2Et2O·1.5CH2Cl2,
[Cu(xantphos)(6-
[Cu(POP)(4,4'-
[Cu(xantphos)(6-
CF3bpy)]+
(CF3)2bpy)][PF6]·0.5CH2Cl2, [Cu(xantphos)(4,4'-(CF3)2bpy)][PF6]
and [Cu(POP)(5,5'-(CF3)2bpy)][PF6]·0.5Et2O were grown by
layering Et2O over CH2Cl2 solutions of the compounds. ORTEP-
style diagrams of the cations in the complexes are illustrated in
Figure S4-S9 in the Supporting Information. We compare these
[Cu(POP)(5,5'-
(CF3)2bpy)]+
111.87(3)
113.02(5)
122.58(4)
113.38(3)
119.47(3)
79.25(9)
79.7(2)
83.54
85.83
84.63
87.92
85.48
6.2(3)
0(1)
[Cu(POP)(4,4'-
(CF3)2bpy)]+
structures
to
those
published
for
[Cu(xantphos)(4,4'-
(CF3)2bpy)]+
Bookmark
not
defined.
[Cu(POP)(bpy)][PF6]·CHCl3,Error!
79.63(13)
80.97(12)
79.66(9)
–17.2(5)
1.7(5)
6.7(3)
[Cu(xantphos)(6-Mebpy)][PF6]·CH2Cl2·0.4Et2O
and
[Cu(xantphos)(6,6'-Me2bpy)][PF6].16 Most of the complexes
crystallize in the triclinic space group P−1, with the exception of
[Cu(xantphos)(bpy)][PF6] and [Cu(POP)(bpy)][PF6]·CHCl3 (both
monoclinic P21/n) and [Cu(POP)(5,5'-(CF3)2bpy)][PF6]·0.5Et2O
(monoclinic P21/c). The Cu–P and Cu–N bond distances show
little variation and are found to be within 2.2159(11) and
2.2841(10) Å, and between 2.014(2) and 2.1523(19) Å,
respectively. Whereas the N–Cu–N angles in all of the
complexes stay very close to 80° (79.25(9) – 80.97(12)°), the P–
Cu–P chelating angles range from 111.87(3) to 119.47(3)° for
the complexes with POP and from 113.38(3) to 122.58(4)° for
those with xantphos. In the case of the unsymmetrical
[Cu(xantphos)(6-CF3bpy)]+ cation, the 6-CF3bipy ligand is
disordered over two orientations with occupancies of 0.75 (CF3
group facing towards the xanthene “bowl”) and 0.25 (CF3 away
from the xanthene “bowl”), respectively. The major conformation
is identical to the one reported for the [Cu(xantphos)(6-Mebpy)]+
cation.16 In Table 1, the dihedral angle between the planes
through N–Cu–N and P–Cu–P illustrates the distortion from the
orthogonal coordination of the two ligands. Whereas
[Cu(xantphos)(bpy)]+ and [Cu(POP)(6-CF3bpy)]+ show the
largest distortion (79.63 and 79.03°, respectively), the cations
[Cu(xantphos)(6-
Mebpy)]+ [a]
[Cu(xantphos)(6,6'-
Me2bpy)]+ [a]
[a] Published data for [Cu(POP)(bpy)][PF6]·CHCl3,Error!
[Cu(xantphos)(6-Mebpy)][PF6]·CH2Cl2·0.4Et2O
Bookmark not defined.
and
[Cu(xantphos)(6,6'-
Me2bpy)][PF6].16
Electrochemistry
Cyclic voltammetry was used to characterize the redox
properties of the copper(I) cations (Table 2), and
a
representative cyclic voltammogram of [Cu(POP)(4,4'-
(CF3)2bpy)][PF6] is illustrated in Figure S10 in the Supporting
Information. The lowest oxidation potential E1/2 (vs. Fc+/Fc),
ox
which corresponds to a Cu+/Cu2+ process, was observed at
+0.72 and +0.76
V
for the reference complexes
[Cu(POP)(bpy)][PF6] and [Cu(xantphos)(bpy)][PF6], respectively.
The substituted complexes present E1/2 values between +0.85
and +0.96 V, the highest value corresponding to
ox
4
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6]. In comparison to
slightly overestimating (~0.1 Å) the reported X-ray values (Table
1). The N−Cu−N and P−Cu−P chelating angles present values
between 75.38 and 77.26º and between 113.71 and 116.89º,
respectively, and slightly underestimate the X-ray values. The
angle between the planes through N–Cu–N and P–Cu–P
illustrating the distortion from the orthogonal coordination of the
two ligands is in a range between 85 and 90º, which is slightly
narrower than that observed experimentally. It should be
stressed that the theoretical geometries correspond to minimum-
energy structures optimized in solution and do not take into
account the packing forces acting in the solid state. These
forces tend to reduce the coordination distances and to increase
the chelating angles.
the other complexes, which have potential separations (Epc
–
Epa) between 100 and 180 mV, a significantly larger separation
of 280 mV was recorded for [Cu(xantphos)(6,6'-Me2-4,4'-
(CF3)2bpy)][PF6]. All complexes show a second oxidation peak at
around +1.2 V which corresponds to an oxidation of the
phosphane ligand (Figure S10). The second reduction peak at
around +0.1 V is connected to this overoxidation; it is not visible
when the scan is recorded only up to +1.0 V and only the first
oxidation is accessed. The complexes with CF3-modified bpy
ligands show reduction processes in addition to the typical
oxidation process, which is in contrast to similar complexes with
unsubstituted or alkyl-substituted bpy ligands. For the
complexes with two CF3 groups at the bipyridine, this reduction
is located around –1.6 V, whereas for the complexes with mono-
substituted 6-CF3bpy this value is cathodically shifted being
around –1.9 V.
Figure 2 sketches the evolution of the energy calculated
for the highest-occupied (HOMO) and lowest-unoccupied
molecular orbital (LUMO) along the series of complexes studied.
The topology of the molecular orbitals does not vary significantly
along the series, so only the contour plots for the reference
complexes [Cu(xantphos)(bpy)]+ and [Cu(POP)(bpy)]+ are shown
in Figure 2. As previously found for this type of complexes,16,Error!
Bookmark not defined. the HOMO appears mainly centred on the metal
with a small contribution from the phosphorus atoms, whereas
the LUMO spreads over the bpy ligand. The energy of the
HOMO slightly changes along both series being between –6.03
and –6.19 eV. This small change is an expected result, because
the HOMO is centred on a region of the complex that remains
structurally unchanged along the series, and is consistent with
the small variation observed in the oxidation potentials of the
substituted complexes (0.85–0.96 V). The attachment of CF3
groups in 4,4’- and 5,5’-positions of the bpy causes a small
stabilization of the HOMO (~0.1 eV) in good agreement with the
higher oxidation potentials measured experimentally for these
complexes (Table 2). Substitution of POP by xantphos also
leads to a small stabilization of the HOMO (~0.05 eV) in accord
with the slightly more positive oxidation potentials recorded for
the xantphos derivatives.
Table 2. Cyclic voltammetric data for [Cu(N^N)(P^P)][PF6] complexes
referenced to internal Fc/Fc+ = 0.V; CH2Cl2 (freshly distilled) solutions with
[nBu4N][PF6] as supporting electrolyte and scan rate of 0.1 V s–1. Processes
are quasi-reversible.
E1/2ox / V
red
E1/2
/
ΔE
Complex cation
(Epc – Epa
/
V
/V
mV)
[Cu(POP)(bpy)]+
+0.72 (110)
+0.76 (110)
+0.90 (170)
+0.92 (100)
+0.89 (150)
+0.94 (180)
–
–
[Cu(xantphos)(bpy)]+
[Cu(POP)(6-CF3bpy)]+
[Cu(xantphos)(6-CF3bpy)]+
[Cu(POP)(5,5'-(CF3)2bpy)]+
–
–
–1.94
–1.89
–1.59
–1.55
2.84
2.81
2.48
2.49
[Cu(xantphos)(5,5'-
(CF3)2bpy)]+
[Cu(POP)(4,4'-(CF3)2bpy)]+
+0.88 (110)
+0.92 (140)
–1.66
–1.62
2.54
2.54
[Cu(xantphos)(4,4'-
(CF3)2bpy)]+
[Cu(xantphos)(6,6'-Me2-4,4'-
As shown in Figure 2, the energy of the LUMO features
larger changes because the attachment of electron-
withdrawing CF3 groups to the bpy ligand, where the LUMO
is located, provokes the stabilization of this orbital. The
addition of a single CF3 group stabilizes the LUMO by
around 0.2 eV, whereas the introduction of a second group
causes an additional stabilization of 0.3 eV. The effect is
slightly larger (0.07eV) when substitution is made in 5,5'-
positions compared with 4,4'-positions. The introduction of
electron-donor methyl groups in [Cu(xantphos)(N^N)]+ is the
contrary, inducing a small destabilization of the LUMO of
0.04 eV upon introduction of the first Me in passing from
[Cu(xantphos)(bpy)]+ to [Cu(xantphos)(6-Mebpy)]+, and of
0.12 eV after introducing the second group in
[Cu(xantphos)(6,6'-Me2bpy)]+. In the complex cation
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)]+, where both CF3
and Me groups are added, the effects sum up and the
LUMO appears 0.12 eV higher in energy than that of
+0.96 (280)
–1.67
2.63
(CF3)2bpy)]+
[Cu(xantphos)(6-Mebpy)]+
+0.85 (100)
+0.90 (150)
–
–
–
–
[Cu(xantphos)(6,6'Me2bpy)]+
Ground state theoretical calculations
The geometry of all the [Cu(N^N)(P^P)]+ cations in their ground
electronic state S0 was optimized at the DFT B3LYP/(6-
31**G+LANL2DZ) level without imposing any symmetry
restriction. Table S1 in the Supporting Information summarizes
the values calculated for selected structural parameters.
Calculations satisfactorily reproduce the distorted-tetrahedral
configuration defined by the P^P and N^N ligands around the
metal centre. The Cu−P bond lengths range from 2.368 to 2.433
Å, and the Cu−N bond lengths are between 2.152 and 2.276 Å,
5
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
Figure 2. Energy diagram showing the energies calculated for the HOMO and LUMO of [Cu(POP)(N^N)]+ and [Cu(xantphos)(N^N)]+
complexes. The HOMO−LUMO energy gap is also quoted. Isovalue contour plots (±0.03 a.u.) are shown for the HOMO and LUMO of the
reference complexes (N^N = bpy)
[Cu(xantphos)(4,4'-(CF3)2bpy)]+. The trends predicted for
the energy of the LUMO perfectly explain the reduction
potentials discussed above for complexes including CF3
groups (Table 2). In summary, complexes incorporating the
6-CF3bpy ligand present a more negative potential than
complexes bearing two CF3 groups, and, within the latter,
complexes substituted in 4,4'-positions show less negative
potentials (by 0.07 V) than complexes substituted in 5,5'-
positions due to the LUMO stabilization (0.07 eV) in passing
from 4,4'- to 5,5'-substituted derivatives.
type of complexes. On this basis, unsubstituted and Me-
substituted complexes will be the ones featuring S1 and T1
at higher energies and bluer absorption/emission
wavelengths. These wavelengths will shift to the red as CF3
groups are added, the maximum shift being expected for
complexes bearing the 5,5'-(CF3)2bpy ligand which show
the lowest HOMO−LUMO gap. However, it has to be
considered that, although this energy ordering may be
correct at the ground state optimum geometries (Franck-
Condon region), geometry relaxation of the excited state
cannot be ignored when dealing with emission processes
as discussed below.
The smallest HOMO−LUMO energy gap (3.24 eV) is
obtained
for
[Cu(POP)(5,5'-(CF3)2bpy)]+
and
[Cu(xantphos)(5,5'-(CF3)2bpy)]+,
followed
by
those
calculated for [Cu(xantphos)(4,4'-(CF3)2bpy)]+ (3.31 eV),
[Cu(POP)(4,4'-(CF3)2bpy)]+
[Cu(xantphos)(6,6’-Me2-4,4'-(CF3)2bpy)]+
Photophysical properties and excited states
(3.33
eV)
and
The UV-Vis absorption spectra of CH2Cl2 solutions of the
[Cu(P^P)(bpy)][PF6] complexes show, in addition to ligand-
centred bands around 280 nm, very broad bands in the
region between 350 and 480 nm, that are assigned to
MLCT transitions (Figure 3, Table 3 and Figure S11). For a
given bpy, the MLCT absorption bands of the respective
complexes with POP and xantphos are very similar in both
the value of λmax and the shape of the band. This suggests
that the energy difference between the HOMO and the
LUMO is mainly determined by the N,N'-chelating ligand in
accord with the DFT results discussed above. Since the
HOMO is fully located on the {Cu(P^P)} domain and is
similar for all the complexes, we can directly observe the
effect of the substitution pattern in the bpy ligand on the
(3.41
eV).
Complexes [Cu(xantphos)(6-CF3bpy)]+ and [Cu(POP)(6-
CF3bpy)]+ with a single CF3 group feature intermediate gap
values of 3.50 and 3.54 eV, respectively, and the widest
gaps correspond to unsubstituted or Me-substituted
complexes with values in the 3.70–3.78 eV range (Figure
2). These trends correctly reproduce the relative order of
the electrochemical gaps inferred from redox potentials
(Table 1).
The HOMO−LUMO gap can be used, in a first
approach, to predict the relative energy of the lowest-
energy singlet (S1) and triplet (T1) electronic excited states,
usually described by the HOMO→LUMO excitation in this
6
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
HOMO−LUMO gap from the MLCT maxima. Because of the
similarity of the absorption spectra between the respective
pairs of POP and xantphos complexes, we focus only on
the series of xantphos complexes and compare this group
to [Cu(xantphos)(bpy)][PF6] with unsubstituted bpy as the
model compound. All of the complexes with one or more
CF3 groups at the bpy show a redshift with respect to
[Cu(xantphos)(bpy)][PF6], corroborating the energy lowering
of the LUMO and the smaller HOMO–LUMO gap predicted
theoretically. The largest redshift of 400 meV (54 nm) is
observed for [Cu(xantphos)(5,5'-(CF3)2bpy)][PF6] and the
smallest comes to 153 meV (19 nm) for [Cu(xantphos)(6-
CF3bpy)][PF6] (see also Table S2), in perfect agreement
with the HOMO−LUMO gaps calculated for these
complexes. Substitution of the bpy ligand with one or two
methyl groups in the 6-positions results in a blueshift of 17
and 69 meV (2 and 8 nm) for the respective complex, in
accord with the destabilization of the LUMO and the
increase of the HOMO–LUMO distance.
Figure 3. Expansion of the lowest-energy MLCT region of the normalized
solution absorption spectra of the [Cu(P^P)(N^N)][PF6] complexes (CH2Cl2, 2.5
× 10−5 mol dm−3). For the full spectrum see Figure S10, and for a comparison
of the maxima, see Table S2.
To gain a deeper insight into the nature of the electronic
excited states giving rise to the absorption spectra, singlet (Sn)
and triplet (Tn) excited states were calculated for all the
complexes using the time dependent DFT (TD-DFT) approach.
Table 4 summarizes the vertical excitation energies computed
for the lowest-energy singlet (S1) and triplet (T1) states at the
optimized geometry of S0. For all the complexes, both S1 and T1
result from the HOMO→LUMO monoexcitation with
a
contribution always exceeding 90%. This supports the MLCT
Table 3. Absorption and emission maxima, photoluminescence quantum yields (PLQY) and lifetimes (τ1/2) for [Cu(P^P)(N^N)][PF6] complexes.
Complex cation
CH2Cl2 solution[a]
Powder[b]
PLQY /
%
Me-THF at 77 K
max
max
λemmax / nm
τ1/2
(non-
deaerated /
deaerated)
/ ns
λem
/
τ1/2 / µs λem
/
τ1/2 /
µs
UV-Vis
MLCT
λ
PLQY
(non-
λexc
nm
/
nm
nm
max / nm
deaerated /
deaerated) /
%
[Cu(POP)(bpy)]+ [e]
388
383
399
402
441
388
390
380
380
-
618, 649[a]
620, 650[a]
618, 646[a]
622, 647[a]
-
0.4/0.5
0.5/0.5
0.7/0.7
0.6/0.6
-/-
43/46
581[c]
587[c]
575[c]
581[c]
648
3.0[c]
1.7[c]
6.2[c]
11.1[c]
0.5[d]
0.5[d]
0.5[d]
0.9[d]
1.5[c]
1.3[c]
610
613
610
595
656
16
11
45
31
−−
[Cu(xantphos)(bpy)]+
[Cu(POP)(6-CF3bpy)]+
[Cu(xantphos)(6-CF3bpy)]+
[Cu(POP)(5,5'-(CF3)2bpy)]+
[Cu(xantphos)(5,5'-
(CF3)2bpy)]+
75/104
95/119
84/99
2.9[c]
2.9[c]
-/-
0.185[d]
437
436
433
-
-
-/-
-/-
-/-
-/-
-/-
-/-
647
0.251[d]
0.096[d]
0.579[d]
646
650
652
−−
3
667, 697[b]
(very weak)
667, 705[b]
(very weak)
[Cu(POP)(4,4'-(CF3)2bpy)]+
430
430
664[d]
632[d]
[Cu(xantphos)(4,4'-
(CF3)2bpy)]+
5
[Cu(xantphos)(6,6'-Me2-
421
381
375
400
379
379
612, 637[a]
605, 635[a]
606, 635[a]
0.5/0.5
1.0/1.8
1.6/10.0
39/39
27/78
517[c]
547[c]
539[c]
50.3[c]
33.8
37.3[c]
12[c]
9.6[c]
11.4[c]
604
567
551
42
46
88
4,4'-(CF3)2bpy)]+
[Cu(xantphos)(6-Mebpy)]+
[Cu(xantphos)(6,6'-
Me2bpy)]+
451/3406
[a]Solution concentration = 2.5 × 10–5 mol dm–3 [b]Solution concentration = 5.0 × 10–5 mol dm–3 [c]λexc = 365 nm. [d]λxc = 405 nm. Deaeration was by flow of argon.
. .
[e][Cu(POP)(bpy)][PF6] was prepared according to the literature (ref. 43) but measurements were made for the present work.
7
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
character of the S1 and T1 states since the HOMO→LUMO
excitation implies an electron transfer from the {Cu(P^P)}
environment to the bpy ligand (see Figure 2). The oscillator
strength (f) calculated for the electronic transition to the S1 state
lies between 0.06 and 0.13 (Table 4), the next singlet excited
state with f values higher than 0.01 being around 0.9 eV above
S1. Excited states with high oscillator strengths (~0.40), centred
on the ligands, are found around 285 nm (~4.35 eV) in good
agreement with the intense bands observed in this region in the
absorption spectrum. These results identify the S1 state as
responsible of the low-energy absorption band observed in the
spectra in the 350–500 nm region (Figures 3 and S11). The
vertical excitation energies calculated for S1 (Table 4) are in
good agreement with the absorption maxima correctly
reproducing the experimental trends (Table 3). Complexes with
Me substituents feature S1 energies blue shifted with respect to
the reference complexes, whereas S1 states of complexes with
one or two CF3 groups appears gradually shifted to the red. The
lowest excitation energy is predicted for [Cu(POP)(5,5'-
(CF3)2bpy)]+ (2.65 eV, 468 nm) and [Cu(xantphos)(5,5'-
(CF3)2bpy)]+ (2.66 eV, 467 nm) in very good agreement with
experimental λmax values (441 and 437 nm, respectively). The
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)]+ complex presents an
excitation energy (2.80 eV, 442 nm) lower than the
[Cu(xantphos)(4,4'-(CF3)2bpy)]+ complex (2.71 eV, 457 nm) due
to the presence of methyl groups. The energy ordering of the S1
states also agrees with that expected from the MO analysis and
the electrochemical gaps, and corroborates that light absorption,
which takes place around the ground state optimal geometry,
can be explained based on electronic factors without considering
the flattening effects that the HOMO→LUMO excitation has on
the molecular geometry of the excited states as explained below.
The T1 states are computed 0.16–0.20 eV below S1 (Table 4)
and the vertical excitation energies to T1 follow the same trends
discussed above for S1.
[Cu(xantphos)(6,6'-Me24,4'-
2.802 / 442 (0.11)
2.639
(CF3)2-bpy)]+
[Cu(xantphos)(6-Mebpy)]+
3.088 / 402 (0.09)
3.145 / 394 (0.06)
2.896
2.985
[Cu(xantphos)(6,6'-Me2bpy)]+
Figure
4 illustrates the normalized solution emission
spectra of the [Cu(P^P)(bpy)][PF6] complexes in CH2Cl2 upon
excitation in the region of their respective MLCT band; values of
max
λem
are presented in Table 3. Due to their poor emissive
behaviour, the spectra of the complexes with 4,4'-(CF3)2bpy are
not included in Figure 3. The complexes with 5,5'-(CF3)2bpy
were non-emissive in solution. With emission maxima between
605 and 705 nm, the complexes are yellow to red emitters in
solution. The bands are structured with two maxima, and as in
the absorption spectra, the emission bands are only slightly
affected by a change from POP to xantphos for a given N^N
ligand. In contrast to the absorption spectra where a redshift was
observed, the emission of complexes with 6-CF3bpy remains
almost unchanged with respect to the unsubstituted complexes
and that of complex with 6,6'-Me2-4,4'-(CF3)2bpy is slightly
blueshifted. The largest blueshift is recorded for the complexes
with 6-Mebpy and 6,6'-Me2bpy.
Table 4. Vertical excitation energies (E) calculated at the TD-DFT B3LYP/(6-
31G**+LANL2DZ) level for the lowest singlet (S1) and triplet (T1) excited states
of complexes [Cu(N^N)(P^P)]+ in CH2Cl2 solution. S0→S1 oscillator strengths
(f) are given within parentheses.
Figure 4. Solution emission spectra of [Cu(P^P)(N^N)][PF6] complexes
(CH2Cl2, 2.5 × 10−5 mol dm−3). For λexc see Table 3.
Complex cation
S1
T1
The solid-state (powder) emission spectra for all
complexes are shown in Figure 5. The only complexes where
the emission is redshifted with respect to the spectra of the
complexes with unsubstituted bpy are those with 5,5'-(CF3)2bpy
and 4,4'-(CF3)2bpy. The redshift is more pronounced for the
complexes with POP, being 83 nm for [Cu(POP)(4,4'-
(CF3)2bpy)][PF6] and 67 nm for [Cu(POP)(5,5'-(CF3)2bpy)][PF6]
(Table 3). The emission maxima of [Cu(POP)(6-CF3bpy)][PF6]
and [Cu(xantphos)(6-CF3bpy)][PF6] both are shifted 6 nm to
shorter wavelengths. In contrast to the solution emission spectra,
where complexes with 6-Mebpy and 6,6'-Me2bpy exhibit the
largest shift to shorter wavelengths, the complex with the most
blueshifted (70 nm) solid state emission is [Cu(xantphos)(6,6'-
Me2-4,4'-(CF3)2bpy)][PF6].
E (eV/nm) (f)
E (eV)
2.906
2.893
2.772
2.704
2.484
2.483
2.531
2.512
[Cu(POP)(bpy)]+
3.089 / 401 (0.08)
3.085 / 402 (0.09)
2.930 / 423 (0.06)
2.874 / 431 (0.07)
2.647 / 468 (0.06)
2.655 / 467 (0.07)
2.739 / 453 (0.09)
2.713 / 457 (0.13)
[Cu(xantphos)(bpy)]+
[Cu(POP)(6-CF3bpy)]+
[Cu(xantphos)(6-CF3bpy)]+
[Cu(POP)(5,5'-(CF3)2bpy)]+
[Cu(xantphos)(5,5'-(CF3)2bpy)]+
[Cu(POP)(4,4'-(CF3)2bpy)]+
[Cu(xantphos)(4,4'-(CF3)2bpy)]+
8
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
Figure 6. Powder samples of [Cu(xantphos)(N^N)][PF6] complexes under
normal light (top) and under UV light (λexc = 365 nm, bottom). From left to right:
[Cu(xantphos)(bpy)][PF6], [Cu(xantphos)(6-CF3bpy)][PF6], [Cu(xantphos)(4,4'-
(CF3)2bpy)][PF6], [Cu(xantphos)(5,5'-(CF3)2bpy)][PF6] and [Cu(xantphos)(6,6'-
Me2-4,4'-(CF3)2bpy][PF6]
.
The highest powder PLQY was measured for
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6] (PLQY
=
50.3%). Due to this high quantum yield, the emission in the
photograph in Figure 6 appears to be almost white,
whereas according to the CIE coordinates (0.449, 0.532,
see also Figure S12), the colour of the emitted light is
between the green and the yellow region. For all the
complexes, the emission maxima are blueshifted on going
from solution to solid state, which has been observed in
earlier studies of similar compounds.16 In general, the
emissive properties of the complexes are enhanced in solid
state compared to solution. The only complex with a
noteworthy quantum yield in solution is [Cu(xantphos)(6,6'-
Me2bpy)][PF6] with 1.6 versus 10.0% for non-deaerated and
deaerated solutions. The higher solution quantum yield is
usually attributed to the steric protection of the copper(I)
centre by the methyl groups attached to the bpy, which also
help to avoid tetrahedron flattening. It is therefore surprising
Figure 5. Normalized powder emission spectra of [Cu(P^P)(N^N)][PF6]
complexes. For λexc see Table 3
.
In order to visualize the solid state emission, photographs
of the powder samples of [Cu(xantphos)(N^N)][PF6] under
normal and under UV light (λexc = 365 nm) are shown in Figure 6.
Solid [Cu(xantphos)(5,5'-(CF3)2bpy)][PF6] appears nearly non-
emissive, (consistent with the low solid-state PLQY, Table 3).
Overall, the photophysical properties of both POP and xantphos
complexes with 5,5'-(CF3)2bpy are impaired with respect to their
respective reference complexes with naked bpy, and we have to
conclude that a modification with CF3 groups in this position is
not beneficial for emissive applications. With PLQY = 0.9%, the
weak red emission of solid [Cu(xantphos)(4,4'-(CF3)2bpy)][PF6] is
just visible by eye (Figure 6). However, even for a red emitter
this value is too low to qualify this complex as luminophore, and
as a result the 4,4'-substitution of the bpy ligand with CF3 groups
appears detrimental. In contrast, [Cu(POP)(6-CF3bpy)][PF6] and
[Cu(xantphos)(6-CF3bpy)][PF6] have increased PLQY and
lifetime values, both in solution and in powder, when compared
to their respective reference complexes [Cu(POP)(bpy)][PF6]
and [Cu(xantphos)(bpy)][PF6]. Whereas the powder PLQY is
only doubled for [Cu(POP)(6-CF3bpy)][PF6] (6.2% in comparison
to 3.0% for the model compound), the value is more than six
times higher for [Cu(xantphos)(6-CF3bpy)][PF6] (11.1% versus
1.7%). Although the CF3 group in the 6-position of the bpy
appears to be beneficial for the photophysical properties, it is
less efficient than a methyl group in this position (PLQY 9.5% for
[Cu(POP)(6-Mebpy)][PF6],15 and 33.8% for [Cu(xantphos)(6-
Mebpy)][PF6]).16
that
the
PLQY
of
[Cu(xantphos)(6,6'-Me2-4,4'-
(CF3)2bpy)][PF6] in solution (0.5%, Table 3) is two orders of
magnitude lower than in powder (50.3%) and also
significantly lower than the PLQY of [Cu(xantphos)(6,6'-
Me2bpy)][PF6]. It appears that although the methyl groups
should retain a beneficial effect, this is cancelled out in
solution by the CF3 groups in 4,4'-positions, which seem to
offer additional non-radiative pathways in solution and thus
quench the excited state. The effect on the photophysical
properties of substitution with CF3 groups in the 4,4'-
positions of the bpy is therefore rather ambiguous. Whereas
in solution the CF3 groups lead to a weakening and redshift
of the emission, in solid state the emission is only redshifted
and less intense for [Cu(POP)(4,4'-(CF3)2bpy)][PF6] and
[Cu(xantphos)(4,4'-(CF3)2bpy)][PF6],
but
for
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6] it is blueshifted
and increased even in comparison to [Cu(xantphos)(6,6'-
Me2bpy)][PF6].
In order to probe the emission processes further, low
temperature lifetime and emission spectra of the complexes
were recorded. Solutions of the compounds in Me-THF form a
glass at 77 K and this approximates to the situation in the solid
state. The emission spectra of the complexes with xantphos
(with the exception of [Cu(xantphos)(5,5'-(CF3)2bpy)][PF6], see
above) are illustrated in Figure 7 and the maxima and lifetime
values are summarized in Table 3 and compared to the room
temperature values of the powder (Table S4).
9
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
maximum is basically unchanged, and [Cu(POP)(4,4'-
(CF3)2bpy)][PF6], with a blueshift of 14 nm (40 meV). The
redshift for the other complexes indicates that at room
temperature, the majority of the emission stems from the
singlet state. The largest redshift is observed for
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6] (345 meV, 87
nm).
To understand the emission processes more thoroughly,
the effects of geometry relaxation on the lowest-energy S1
and T1 states were investigated theoretically. The
geometries of the T1 states were first optimized at the spin-
unrestricted UB3LYP level and they feature relevant
differences with respect to those calculated for the ground
state S0. As explained above, T1 originates in the
HOMO→LUMO excitation and implies a charge transfer
from a d orbital of the Cu atom to a molecular orbital
centred on the bpy moiety of the complexes. The metal
centre is hence partially oxidized and tends to adopt a
squared-planar coordination sphere, typical of four-fold
coordinated d9 copper complexes, instead of the tetrahedral
coordination preferred by d10 copper complexes. This
tendency is clearly illustrated by the angle formed by the N–
Cu–N and P–Cu–P planes that changes from values close
to 90º in S0, typical of a tetrahedral coordination in which
both ligands are orthogonal, to values as low as 58º in T1
(Table S1). The presence of substituents in 6,6'-positions
hinders the bending of the bpy moiety and limits the
flattening distortion of the tetrahedron in T1. Thus,
complexes with one of those positions substituted by a Me
or a CF3 present an angle between the planes through N–
Cu–N and P–Cu–P between 70.3 and 71.4º, which is
significantly larger than for complexes with no substituent in
Figure 7. Normalized emission spectra of [Cu(P^P)(N^N)][PF6] complexes in
Me-THF at 77 K, λexc = 410 nm
.
The excited state lifetimes are found to be lengthened
for all the complexes, which indicates that they are, as
confirmed for similar compounds, TADF emitters.11, 32 , 33
TADF describes the emission from the singlet excited state
S1 which has been (re)populated from the long-lived triplet
excited state T1 by making use of the available thermal
energy kBT. This process is favourable for the application of
a molecule as an emitter for two reasons. First, the
repopulation of the singlet excited state allows the
harvesting of, in theory, 100% of all photons, which equals
a PLQY of 100%, respectively all excitons when the
situation in the electroluminescent device is considered.
Second, TADF processes lead to a shortening of the
excited state lifetime. Phosphorescence from T1 can be a
very slow process in comparison to fluorescence from S1
and is therefore not ideal when it comes to the application
of the molecule in light-emitting devices.
the
6,6'-positions
(~50º).
The
complexes
[Cu(xantphos)(6,6'-Me2bpy)]+ and [Cu(xantphos)(6,6'-Me2-
4,4'-(CF3)2bpy]+ with both positions substituted by Me
groups present even less flattened structures with angles of
74.3 and 74.4º, respectively (Table S1).
At 77 K, TADF is reduced or even completely impeded,
depending on the energy gap between T1 and S1. As a
consequence,
the
contribution
of
the
slower
phosphorescence to the emission is increased, which can
be directly evidenced by the significant elogation of the
lifetime. This is up to 88 µs for [Cu(xantphos)(6,6'-
Me2bpy)][PF6], which is an almost eight-fold increase with
respect to powder at room temperature (11.4 µs). For
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6], which has a
powder lifetime (12.0 µs) similar to [Cu(xantphos)(6,6'-
Me2bpy)][PF6], the lifetime at 77 K is only less than four-fold
increased to 42 µs (Table S4). This could be either due to a
shorter phosphorescence lifetime or a smaller energy gap
between the triplet and singlet excited states. This energy
gap can also be inferred by comparing the powder emission
maxima at room temperature and the emission in the frozen
Me-THF glass at 77 K. As with lower temperature the
proportion of triplet emission increases, the emission
maxima should normally be moved to longer wavelengths.
This is the case for all the complexes reported here, except
for [Cu(xantphos)(5,5'-(CF3)2bpy)][PF6], where the emission
The flattening of the complex structure is accompanied
by a large stabilization of the T1 state that, for complexes
with no substituent in 6,6'-positions of the bpy, amounts to
0.8–0.9 eV with respect to the energies at the equilibrium
geometry of S0. This relaxation energy decreases to ~0.6
eV
for
[Cu(xantphos)(6,6'-Me2bpy)]+
due
and
the
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)]+
to
hindering effect of the Me groups that limits the geometrical
relaxation. Therefore, the energy position of the T1 state
relative to S0 not only depends on the electron-donating or
electron-withdrawing character of the substituent groups but
also on the positions on the ligands where they are
introduced due to the purely structural effects they induce.
This justifies the fact that the emission maxima recorded for
the family of complexes studied does not follow the trends
observed for absorption (see above), and also explains that
the TD-DFT energies computed for T1 (Table 4) do not
10
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
reproduce the experimental trends observed in the
emission spectra because they are based only on electronic
considerations (they are calculated at the geometry of S0)
with no geometry relaxation. When the emission energies
from T1 are determined as the vertical energy difference
between the T1 and S0 states at the T1 relaxed geometry,
they fully support the experimental trends observed at 77 K
where TADF is suppressed and emission mainly results
from phosphorescence from T1. [Cu(POP)(bpy)]+,
in the HOMO→LUMO excitation. The values of ΔE(S1 – T1)
were computed for a set of representative complexe cations
([Cu(POP)(bpy)]+, [Cu(xantphos)(bpy)]+, [Cu(POP)(4,4'-
(CF3)2bpy)]+,
[Cu(xantphos)(4,4'-(CF3)2bpy)]+
and
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)]+) and in all cases
are found in the 0.12–0.20 eV range. These values are
significantly lower than the 0.37 eV (3000 cm–1) proposed
by Yersin and coworkers.12,34 to allow the population of S1
from T1 at room temperature and, therefore, contribution
from S1 to the emission by TADF should be expected at
room temperature. The ΔE values calculated for the
[Cu(xantphos)(bpy)]+ and
[Cu(POP)(6-CF3bpy)]+
are
calculated to emit at very similar wavelengths (730, 737 and
735 nm, respectively) whereas the emission of
[Cu(xantphos)(6-CF3bpy)]+ is blueshifted (691 nm) and that
of complexes with 4,4'- and 5,5'-(CF3)2bpy is redshifted
(1.34–1.41 eV, 920–875 nm). Complex [Cu(xantphos)(6,6'-
Me24,4'-(CF3)2bpy)]+, whose absorption wavelength and S1
energy were shifted to the red with respect to
[Cu(xantphos)(bpy)]+, features a T1 energy (1.80 eV, 688
nm) and a emission maximum at 77 K (2.05 eV, 604 nm)
bluer than the reference complex (1.68 and 2.01 eV,
respectively). The theoretical values underestimate the
experimental emission energies (Table 3) because they are
calculated at the fully relaxed geometry of T1 whereas this
relaxation is expected to be severely restricted in the glass
at 77 K. As a conclusion, the emission energies of
[Cu(P^P)(bpy)]+ complexes with substituents at 6,6'-
positions of the bpy do not correspond to those expected
reference
complexes
[Cu(POP)(bpy)]+
and
[Cu(xantphos)(bpy)]+ are 0.177 and 0.185 eV, respectively,
slightly above the ideal value of 0.12 eV for TADF being
exploited in electroluminescent devices, but still in the
range of other Cu complexes for which TADF has been
reported previously. The inclusion of CF3 groups in the 4,4'-
positions has little effect on ΔE that slightly increases for
[Cu(POP)(4,4'-(CF3)2bpy)]+
(0.189
eV)
and
[Cu(xantphos)(4,4'-(CF3)2bpy)]+ (0.197 eV). However, the
addition of Me groups in 6,6'-positions favours a decrease
in ΔE, and the value of 0.110 eV computed for
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)]+ points to this
complex as the one expected to feature TADF at lower
temperatures.
Electroluminescent devices
from
electronic
considerations
(MO
analysis
or
Light-emitting electrochemical cells (LECs) were fabricated with
electrochemical and optical absorption gaps) because 6,6'-
substitution hinders the tetrahedron flattening associated to
T1 relaxation and limits its stabilization. The T1 state
therefore stays at higher energies than in complexes with
no substituent at 6,6'-positions, thus leading to a bluer
emission than expected.
complexes
[Cu(POP)(6-CF3bpy)][PF6],
[Cu(xantphos)(6-
CF3bpy)][PF6] and [Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6]
because only these CF3-substituted complexes show significant
PLQY values in powder (6.2, 11.1 and 50.3%, respectively,
Table 3). The LECs were fabricated in
architecture, by depositing
a
double layer
poly(3,4-
a
Finally, the S1 and T1 states were fully optimized using
the TD-DFT approach to evaluate the adiabatic energy
difference (ΔE(S1 – T1)) between these states at their
respective minimum-energy geometries. It should be
mentioned that the complexes in the S1 state undergo
flattening distortions of the tetrahedral structure similar to
those discussed above for T1 because both states originate
ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS)
layer and the emissive layer sandwiched between indium tin
oxide (ITO) and aluminium electrodes. The active layer
contained the copper(I) complex mixed with the ionic liquid (IL)
1-ethyl-3-methylimidazolium hexafluoridophosphate [Emim][PF6]
at a 4:1 (Cu complex:IL) molar ratio. The IL was added to
shorten the turn-on time of the LEC by increasing the
Table 5. Performance of ITO/PEDOT:PSS/[Cu(P^P)^N^N)][PF6]:[Emim][PF6] 4:1 molar ratio/Al LECs measured using a pulsed current driving (average current
density 100 A m–2, 1 kHz, 50% duty cycle, block wave).
b
c
f
max
Complex
tona / min
Lum0
Lummax
t1/2d / h
EQEmaxe / %
PCEmax
Efficacymax
cd A–1
0.7
/
λEL
/
/cd m–2
/cd m–2
65
/lm W–1
0.2
nm
595
589
593
[Cu(POP)(6-CF3bpy)][PF6]
[Cu(xantphos)(6-CF3bpy)][PF6]
[Cu(xantphos)(6,6'-Me2-4,4'-
(CF3)2bpy)][PF6]
22
137
8
39
5
59
8.5
31.0
2.0
0.4
0.5
0.6
109
131
0.4
0.4
1.1
1.3
a Time to reach the maximum luminance. b Initial luminance. c Maximum luminance reached. d Time to reach one-half of the maximum luminance. eMaximum
external quantum efficiency reached. f Maximum power conversion efficiency reached.
11
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
concentration of ionic species and thereby the ionic mobility in
the light-emitting layer.35,36 To enhance the device response and
lifetime, LECs were operated using a block-wave pulsed current
of 100 A m‒2 (1 kHz and 50% duty). The LEC characteristics are
summarized in Table 5 and the luminance and average voltage
versus time plots are depicted in Figure 8 and S13, respectively.
of the maximum luminance. The LEC containing
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6]
shortest ton (8 min) and t1/2 (2h). When the active material is
[Cu(POP)(6-CF3bpy)][PF6] or [Cu(xantphos)(6-
has
the
CF3bpy)][PF6], the device response is slower with ton values
of 22 and 137 min, respectively. Following this trend, the
respective LEC lifetimes are 8.5 and 31 hours when using
these two complexes (Table 5). These values indicate a
clear link between the ton and t1/2; the faster the ton the
shorter is t1/2
.
On the one hand, the LEC with
[Cu(xantphos)(6-CF3bpy)][PF6] (with shorter ton and t1/2
)
rapidly reaches its minimum voltage (~4.8 V) which then
increases over time (see Figure S13). This voltage profile is
indicative of a fast ionic mobility at the beginning of
operation. However, over time, charge transport is hindered
as evidenced by an increasing resistance during operation.
On the other hand, the LEC with [Cu(xantphos)(6-
CF3bpy)][PF6] shows a voltage profile which decreases over
time during device operation.The luminance as well as the
efficacy of the LECs are in agreement with the trend
obtained for the PLQY in thin film. [Cu(xantphos)(6,6'-Me2-
4,4'-(CF3)2bpy)][PF6] with a PLQY of 16% achieves a
maximum luminance of 131 cd m‒2 and a maximum efficacy
of 1.3 cd A‒1). [Cu(POP)(6-CF3bpy)][PF6] with a PLQY of
4% achieves 65 cd m‒2 and 0.7 cd A‒1 in the LEC.
However, the LEC using [Cu(xantphos)(6-CF3bpy)][PF6],
with a PLQY of 5% in thin film similar to [Cu(POP)(6-
Figure 8. Luminance versus time characteristics for
ITO/PEDOT:PSS/[Cu(P^P)^N^N)][PF6]:[Emim][PF6] 4:1/Al LECs operated at
pulsed current (average current density 100 A m‒2, 1 kHz, 50% duty cycle,
block wave).
All three LECs have orange electroluminescence (Figure
S14) with maximum emission in the 589‒595 nm range and
a shoulder around 650 nm similar to that observed for the
spectra in solution (Figure 4). The emission maxima are
slightly blue-shifted with respect to the photoluminescence
spectra recorded for the active thin film (Cu complex:IL at a
4:1 molar ratio) with maxima in the 596‒606 nm range
(Figure S15). It should be noted that the PLQY significantly
decreases in passing from powder to the active thin film
composition for both [Cu(POP)(6-CF3bpy)][PF6] (PLQY =
4%), [Cu(xantphos)(6-CF3bpy)][PF6] (5%) and especially for
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6] (16%). This
decrease in PLQY has been observed for Cu-based
emitters and is related to the different environment
surrounding the complex.16,18 In thin amorphous films, the
structural rearrangement of the complex upon excitation is
less hindered than in powder which leads to poorer
emissive properties. The presence of the CF3 group results
in a slight red-shift of the electroluminescence when
CF3bpy)][PF6], exhibits
a higher performance with a
luminance (109 cd m‒2) and a efficacy (1.1 cd A‒1) similar to
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6]. Considering
the PLQY values in thin film and a typical outcoupling of
20%, the theoretical maximum external quantum efficiency
(EQEmax
CF3bpy)][PF6]
(CF3)2bpy)][PF6] when all injected electrons and holes
combine is and 3.2%, respectively, whereas the
)
predicted for LECs with [Cu(xantphos)(6-
and [Cu(xantphos)(6,6'-Me2-4,4'-
1
experimental values are 0.5 and 0.6% (Table 5). The
smaller difference found between the theoretical and the
experimental value for the LEC with [Cu(xantphos)(6-
CF3bpy)][PF6] indicates a lower exciton-quenching, which is
probably related with a better balance between electrons
and holes in the device.38 With this in mind, together, the
time-dependence characteristics and the voltage profile
indicate that the characteristics of the LEC with
[Cu(xantphos)(6-CF3bpy)][PF6] are limited by permanent
degradation while with [Cu(xantphos)(6-CF3bpy)][PF6] the
comparing to the analogous complexes substituted with
max
methyl groups [Cu(POP)(6-Mebpy)][PF6] (λEL
nm)15 and [Cu(xantphos)(6-Mebpy)][PF6] (λEL
nm).16
= 574
= 583
max
doping-induced quenching is the main decay factor.
Once biased, the electrical resistance of the device is
reduced due to the presence of ions in agreement with the
operational mechanism established for LECs.37 Hence, the
luminance is initially low and rises up gradually (Figure 8).
The time needed to achieve the maximum luminance (ton) is
an indicator of the device response, and the operational
lifetime (t1/2) is usually defined as the time to reach one-half
Similar complexes with xanthphos, [Cu(xantphos)(6-
Mebpy)][PF6] and [Cu(xantphos)(6,6'-Me2bpy)][PF6], have been
previously characterized with the same composition and
architecture in LECs but operated with a lower pulsed current
(50 A m‒2).16 Compared with [Cu(xantphos)(6-CF3bpy)][PF6],
substitution of the CF3 group by a Me group leads to a LEC
device with a slightly shorter ton (102 min) but also a reduced t1/2
12
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
(15 h). The attachment of the second methyl group (6,6'-
Me2bpy) reduces the response time (10 min) but also the lifetime
t1/2 (0.8 h) of the LEC. Further substitution with CF3 groups in the
4,4'-positions (6,6'-Me2-4,4'-(CF3)2bpy) has benefits on the time-
dependence characteristics of the LEC because both the ton (8
min) and t1/2 (2 h) are improved with respect to
[Cu(xantphos)(6,6'-Me2bpy)][PF6], even if the device is operated
at higher current densities (100 A m‒2 vs. 50 A m‒2). However,
the presence of CF3 groups is detrimental for both the luminance
and the efficiency of the device, which are less for the LEC with
[Cu(xantphos)(6,6'-Me2bpy-4,4'-(CF3)2)][PF6] (131 cd m‒2 and
1.3 cd A‒1, respectively) than with [Cu(xantphos)(6,6'-
Me2bpy)][PF6] (145 cd m‒2 and 3.0 cd A‒1).16 This negative effect
is due to the lower PLQY of the CF3-substituted emitters in the
active films.
turn-on time (8 min), whereas the longest lived LEC (t1/2 = 31 h)
contained [Cu(xantphos)(6-CF3bpy)][PF6]; these LECs reached
maximum luminances of 131 and 109 cd m–2 respectively.
Although the device with [Cu(xantphos)(6-CF3bpy)][PF6] was
operated at higher current density (100 A m‒2 vs. 50 A m‒2), its
lifetime t1/2 is more than twice as long as for the device with the
respective [Cu(xantphos)(6-Mebpy)][PF6] complex (31 vs. 15 h).
However, compared to LECs with CF3-free bpy-based
[Cu(POP)(N^N)]+ or [Cu(xantphos)(N^N]+ complexes, those
incorporating CF3 groups performed less well; CF3 substituents
are detrimental to both the luminance and the efficiency of the
LEC.
Acknowledgements
Financial support from the Swiss National Science Foundation
(Grant number 162631), the University of Basel, the MINECO
of Spain (CTQ2015-71154-P and Unidad de Excelencia María
de Maeztu MDM-2015-0538), the Generalitat Valenciana
(PROMETEO/2016/ 135) and European FEDER funds
(CTQ2015-71154-P) is acknowledged. Prof. Dr. Oliver S.
Wenger and Dr. Christopher Bryan Larsen are thanked for use
of their LP920-KS instrument from Edinburgh Instruments,
including Quantel Brilliant b Nd:YAG laser and iCCD camera
from Andor, for low-temperature emission and lifetime
Conclusions
We
have
investigated
a
series
of
heteroleptic
[Cu(POP)(N^N)][PF6] and [Cu(xantphos)(N^N)][PF6] complexes
in which the N^N ligand is a bpy substituted with CF3 groups in
either the 6-, 5- or 4-positions. The effects of incorporating both
methyl and trifluoromethyl into the bpy domain on the structural,
electrochemical and photophysical properties of the complexes
have been studied. The single crystal structures of
[Cu(xantphos)(bpy)][PF6],
[Cu(xantphos)(4,4'-(CF3)2bpy)][PF6]
measurements
.
[Cu(POP)(6-CF3bpy)][PF6]·1.3Et2O·0.35H2O, [Cu(xantphos)(6-
CF3bpy)][PF6]·2Et2O·1.5CH2Cl2,
(CF3)2bpy)][PF6]·0.5CH2Cl2
[Cu(POP)(4,4'-
[Cu(POP)(5,5'-
and
(CF3)2bpy)][PF6]·0.5Et2O have been determined; each copper(I)
centre is in a distorted tetrahedral environment.
Keywords: copper • 2,2'-bipyridine • chelating bis(phosphane) •
trifluoromethyl • light-emitting electrochemical cell
ox
Introducing the CF3 groups pushes the value of E1/2 for
the Cu+/Cu2+ process to higher potentials (+0.85 to +0.96 V) with
respect to [Cu(POP)(bpy)]+ and [Cu(xantphos)(bpy)]+, and the
ox
observed trends in E1/2 are consistent with results of DFT
1
N. T. Kalyani, H. Swart, S. J. Dhoble, Principles and Applications
of Organic Light Emitting Diodes (OLEDs), Woodhead Publishing,
Duxford, United Kingdom, 2017.
calculations. The HOMO–LUMO separation is significantly
altered by the nature of the N^N ligand, with the largest redshift
in the MLCT band being for [Cu(P^P)(5,5'-(CF3)2bpy)]+; this
observation is corroborated by DFT calculations. In solution, the
compounds are weak yellow to red emitters, but the emission is
2
3
Yi-Lu Chang, Efficient Organic Light Emitting-Diodes (OLEDs),
CRC Press Taylor&Francis group, Boca Raton, Florida, 2016.
H. Yersin, Highly Efficient OLEDs with Phosphorescent Materials,
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany,
2008.
enhanced on going to the solid state. Powdered
max
[Cu(xantphos)(4,4'-(CF3)2bpy)][PF6] (λem
= 517 nm) exhibits
the highest PLQY (50.3%). The emission properties strongly
depend on the substitution pattern and cannot be explained by
simple electronic considerations due to the flattening of the
tetrahedral structure experienced by the complex upon
excitation. Compared to solution behaviour at 298 K, excited
state lifetimes lengthen for all complexes in frozen Me-THF (77
K), which is indicative of TADF. TD-DFT calculations reveal that
ΔE(S1 – T1) lies in the range 0.12–0.20 eV, which is a small
enough energy to allow TADF.
4
5
6
7
8
H. Rudmann, S. Shimada, M. F. Rubner, J. Am. Chem. Soc. 2002,
124, 4918-4921.
R.D. Costa, E. Ortí, H.J. Bolink, F. Monti, G. Accorsi, N. Armaroli,
Angew. Chem. Int. Ed. 2012, 51, 8178-8211.
C. E. Housecroft, E.C. Constable, Coord. Chem. Rev. 2017, 350,
155-177.
H. J. Bolink, E. Coronado, R. D. Costa, N. Lardiés, E. Ortí, Inorg.
Chem. 2008, 47, 9149-9151.
LECs were fabricated with [Cu(POP)(6-CF3bpy)][PF6],
[Cu(xantphos)(6-CF3bpy)][PF6] or [Cu(xantphos)(6,6'-Me2-4,4'-
R. D. Costa, E. Ortí, H. J. Bolink, S. Graber, C. E. Housecroft, E. C.
Constable, Adv. Funct. Mater. 2010, 20, 1511-1520.
(CF3)2bpy)][PF6] in the emissive layer. All showed yellow
max
electroluminescence (λem
= 589–595 nm). The LEC with
[Cu(xantphos)(6,6'-Me2-4,4'-(CF3)2bpy)][PF6] had the fastest
13
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
9
K. J. Suhr, L. D. Bastatas, Y. Shen, L. A. Mitchell, B. J. Holliday, J.
25 M. Tromp, A. J. Dent, J. Headspith, T. L. Easun, X.-Z. Sun, M. W.
George, O. Mathon, G. Smolentsev, M. L. Hamilton, J. Evans, J.
Phys. Chem. B 2013, 117, 7381-7387.
D. Slinker, ACS Appl. Mater. Interfaces 2016, 8, 8888-8892.
10 See for example: C. E. Housecroft and E. C. Constable, Chem.
Soc. Rev. 2015, 44, 8386-8398.
26 T. J. Penfold, S. Karlsson, G. Capano, F. A. Lima, J. Rittmann, M.
Reinhard, M. H. Rittmann-Frank, O. Braem, E. Baranoff, R.
Abela, I. Tavernelli, U. Rothlisberger, C. J. Milne, M. Chergui, J.
Phys. Chem. A 2013, 117, 4591-4601.
11 R. Czerwieniec, H. Yersin, Inorg. Chem. 2015, 54, 4322-4327.
12 R. Czerwieniec, M. J. Leitl, H. H. H. Homeier, H. Yersin, Coord.
Chem. Rev. 2016, 325, 2-28.
27 A. Kaeser, M. Mohankumar, J. Mohanraj, F. Monti, M.Holler, J.-
J.Cid, O. Moudam, I.Nierengarten, L. Karmazin-Brelot, C.
Duhayon, B. Delavaux-Nicot, N. Armaroli, J.-F. Nierengarten,
Inorg. Chem. 2013, 52, 12140-12151.
13 L. Bergmann, G. J. Hedley, T. Baumann, S. Bräse, I. D.W.
Samuel, Sci. Adv. 2016, 2, e1500889.
14. M. Osawa, M. Hashimoto, I. Kawata, M. Hoshino, Dalton Trans. 2017,
46, 12446-12455.
28 F. Brunner, Y. M. Klein, S. Keller, C. D. Morris, A. Prescimone, E.
C. Constable, C. E. Housecroft, RSC Adv. 2015, 5, 58694-58703.
29 S. Keller, F. Brunner, A. Prescimone, E. C. Constable, C. E.
Housecroft, Inorg. Chem. Comm. 2015, 58, 64-66.
15 S. Keller, E. C. Constable, C. E. Housecroft, M. Neuburger, A.
Prescimone, G. Longo, A. Pertegás, M. Sessolo, H. J. Bolink,
Dalton Trans. 2014, 43, 16593-16596.
30 M. Charton, J. Am. Chem. Soc. 1969, 91, 615-618.
31 K. Uneyama, Organofluorine Chemistry, Blackwell Publishing,
Oxford, UK, 2006.
16 S. Keller, A. Pertegás, G. Longo, L. Martinez, J. Cerdá, J. M.
Junquera-Hernández, A. Prescimone, E. C. Constable, C. E.
Housecroft, E. Ortí, H. J. Bolink, J. Mater. Chem. C. 2016, 4,
3857-3871.
32 M. Y. Wong, E. Zysman-Colman in Light-Emitting Electrochemical
Cells, ed. R. D. Costa, Springer International Publishing AG,
Cham, Switzerland, 2017, pp. 237-266.
17 M. D. Weber, M. Viciano-Chumillas, D. Armentano, J. Cano, R. D.
Costa, Dalton Trans. 2017, 46, 6312-6323.
33 M. Elie, S. Gaillard, J.-L. Renaud, in Light-Emitting
Electrochemical Cells, ed. R. D. Costa, Springer International
Publishing AG, Cham, Switzerland, 2017, pp. 287-327.
34 M. J. Leitl, V. A. Krylova, P. I. Djurovich, M. E. Thompson, H. Yersin,
J. Am. Chem. Soc. 2014, 136, 16032-16038.
18 F. Brunner, L. Martínez-Sarti, S. Keller, A. Pertegás, A.
Prescimone, E. C. Constable, H. J. Bolink, C. E. Housecroft,
Dalton Trans. 2016, 45, 15180-15192.
19 Y. Chi, B. Tong, P.-T. Chou, Coord. Chem. Rev. 2014, 281, 1-25.
20 F. Brunner, Y. M. Klein, S. Keller, C. D. Morris, A. Prescimone, E. C.
Constable, C. E. Housecroft, RSC Adv. 2015, 5, 58694-58703.
35 S. T. Parker, J. D. Slinker, M. S. Lowry, M. P. Cox, S. Bernhard, G.
G. Malliaras, Chem. Mater. 2005, 17, 3187-3190.
21 D. O’Hagana, H. S. Rzepa, Chem. Commun. 1997, 645-652 and
references therein.
36 R. D. Costa, A. Pertegás, E. Ortí, H. J. Bolink, Chem. Mater. 2010, 22,
1288-1290.
22 I. Andrés-Tomé, J. Fyson, F.B. Dias, A.P. Monkman, G.
Iacobellis, P. Coppo, Dalton Trans. 2012, 41, 8669-8674.
23 J. Yuasa, M. Dan, T. Kawai, Dalton Trans. 2013, 42, 16096-
16101.
37 S. van Reenen, P. Matyba, A. Dzwilewski, R. A. J. Janssen, L.
Edman, M. Kemerink, J. Am. Chem. Soc., 2010, 132, 13776-13781.
38 H.-C. Su, J.-H. Hsu, Dalton Trans. 2015, 44, 8330-8345.
24 M. W. Mara, K. A. Fransted, L. X. Chen, Coord. Chem. Rev.
2015, 282-283, 2-18.
14
This article is protected by copyright. All rights reserved.

10.1002/cplu.201700501
ChemPlusChem
ChemPlusChem
FULL PAPER
Sarah Keller, Fabian Brunner, José M.
Junquera-Hernández, Antonio Pertegás,
Maria-Grazia La-Placa, Alessandro
Prescimone, Edwin C. Constable, Henk J.
Bolink, Enrique Ortí* and Catherine E.
Housecroft*
Text for Table of Contents
Page No. – Page No.
CF3 Substitution of [Cu(P^P)(bpy)][PF6]
Complexes: Effects on Photophysical
Properties and Light-emitting
Electrochemical Cell Performance
[a]
[b]
S. Keller, F. Brunner, Dr. A. Prescimone, Pro
Prof. Dr. C.E. Housecroft
Department of Chemistry
University of Basel
BPR 1096, Mattenstrasse 24a, Basel 4058,
E-mail: catherine.housecroft@unibas.ch
Dr. J.M. Junquera-Hernández, Dr. A. Perteg
H.J. Bolink, Prof. Dr. E. Ortí
Instituto de Ciencia Molecular,
Universidad de Valencia,
ES-45980 Paterna, Valencia, Spain
e-mail: enrique.orti@uv.es
Supporting information for this article is give
the document.((Please delete this text if not
15
This article is protected by copyright. All rights reserved.