On the development of a new approach to the design of lanthanide-based materials for solution-processed OLEDs

Article abstract of DOI:10.1039/c9dt03823j

The targeted design of lanthanide-based emitters for solution-processed organic light-emitting diodes (OLEDs) resulted in obtaining an NIR OLED with one of the highest efficiencies among ytterbium-based solution-processed OLEDs (30 μW W^-1). The design was aimed at the combination of high luminescence efficiency with solubility and charge carrier mobility. The latter was achieved thanks to the introduction of the purposefully selected neutral ligands, which combine electron mobility and the ability to sensitize lanthanide luminescence. Besides, the HOMO and LUMO energies and charge carrier mobility of solution-processed thin films of coordination compounds were measured experimentally for the first time, and novel highly luminescent europium-based materials with PLQYs of up to 80% and purely NIR luminescent ytterbium complexes were obtained.

Full text of DOI:10.1039/c9dt03823j

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
On the development of a new approach to the
design of lanthanide-based materials for solution-
processed OLEDs†
Cite this: Dalton Trans., 2019, 48,
17298
Makarii I. Kozlov, a Andrey N. Aslandukov, a Andrey A. Vashchenko,b
d
Alexey V. Medvedko,a Alexey E. Aleksandrov, c Raitis Grzibovskis,
c
Alexander S. Goloveshkin,e Leonid S. Lepnev,b Alexey R. Tameev,
Aivars Vembris d and Valentina V. Utochnikova
*
a,f
The targeted design of lanthanide-based emitters for solution-processed organic light-emitting diodes
(OLEDs) resulted in obtaining an NIR OLED with one of the highest efficiencies among ytterbium-based
solution-processed OLEDs (30 μW W−1). The design was aimed at the combination of high luminescence
efficiency with solubility and charge carrier mobility. The latter was achieved thanks to the introduction of
the purposefully selected neutral ligands, which combine electron mobility and the ability to sensitize
lanthanide luminescence. Besides, the HOMO and LUMO energies and charge carrier mobility of solu-
tion-processed thin films of coordination compounds were measured experimentally for the first time,
and novel highly luminescent europium-based materials with PLQYs of up to 80% and purely NIR lumi-
nescent ytterbium complexes were obtained.
Received 26th September 2019,
Accepted 25th October 2019
DOI: 10.1039/c9dt03823j
rsc.li/dalton
excited state since according to the spin statistics 75% of
electro-excitons possess triplet nature.5 Nowadays there are
Introduction
Organic light-emitting diode (OLED) technology attracts several types of effective emitters, which are able to provide
increasing attention due to the possibility of obtaining high- triplet harvesting, among which are the widely used iridium
surface area light sources and displays.1 However, to make complexes6–8 and rather promising TADF materials.9–11
them commercially attractive, solution-processed OLEDs, Despite the high efficiencies of OLEDs based on them, these
which are cost-effective and allow high resolution, are impor- types of emitters suffer from two drawbacks: high cost and
tant. Although examples of efficient solution-processed OLEDs high luminescence bandwidth (over 100 nm), which limit the
exist,2,3 allowing us to expect high prospects for this techno- contrast characteristics of the diodes. Another class of emitters
logy, the efficiency of these solution-processed OLEDs is that involve triplet levels in the luminescence process are
usually lower than that of vacuum-deposited OLEDs.2,4 Thus, lanthanide coordination compounds, which are well known
the development of solution-processed OLEDs is an important for their narrow emission bands.12 The position of lanthanide
task nowadays.
ionic luminescence bands depends only on the lanthanide
The design of efficient OLED materials should be aimed at ion, and by changing the lanthanide ion it is possible to
the materials able to provide radiative relaxation of the triplet obtain different colors of luminescence in the visible (for
Sm3+, Eu3+, Tb3+, and Dy3+ ions) and in the NIR ranges (for
Pr3+, Nd3+, Er3+, Tm3+, and Yb3+ ions). Among all the lantha-
nide complexes aromatic carboxylates are the most prospective
aM.V. Lomonosov Moscow State University, 1/3 Leninskye Gory, Moscow, 119991,
Russia. E-mail: valentina.utochnikova@gmail.com
materials for OLED applications due to their high chemical,
bP.N. Lebedev Physical Institute, Leninsky prosp. 53, Moscow, 119992, Russia
thermal, optical, and electrical stability.13 However, until
cA.N. Frumkin Institute of Physical Chemistry and Electrochemistry, Leninsky prosp.
recently, their use in OLEDs has been hampered by their
insufficient solubility and charge transport properties.14,15
Another important challenge is obtaining OLEDs operating in
the near-infrared (NIR) range, which have drawn particular inter-
est in novel and cost-effective applications for emerging photonic
and optoelectronic technologies, such as lab-on-chip platforms
for medical diagnostics, flexible self-medicated pads for photo-
31, bld. 4, Moscow, 119071, Russia
dInstitute of Solid State Physics, University of Latvia, 8 Kengaraga iela, Riga,
LV-1063, Latvia
eA.N. Nesmeyanov Institute of Organoelement Compounds, Vavilova St. 28, Moscow,
119334, Russia
fEVOLED Ltd, 1A-24 Puškina iela, Riga LV-1050, Latvia
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
C9DT03823J
17298 | Dalton Trans., 2019, 48, 17298–17309
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
The individuality and composition of the obtained com-
pounds were determined based on PXRD data and TGA data
along with mass detection and 1H NMR spectroscopy.
Regardless of the initial reagent ratio n, the powder patterns
of the obtained europium complexes were identical (Fig. 2):
the positions of all the reflexes coincided, and a slight differ-
ence in the reflex intensity ratio seemed to be associated with
the different degrees of texturing of the samples. The indivi-
duality of both europium and ytterbium complexes was proved
by the indexing of the PXRD data (see ESI†). The absence of
the reflexes corresponding to either Eu(pfb)3(H2O) and Yb
(pfb)3 or BPhen, PyPhen, DPPZ, and BDPZ indicates the
absence of impurities in these reagents (Fig. 2). Thus, accord-
ing to the PXRD data, individual complexes were formed, and
their composition was independent of the initial ratio of the
reagents.
Fig. 1 The neutral ligands used in the present study.
dynamic therapy, night vision and plastic-based telecommuni-
cations, optical sensors and information processing.16–20
Ytterbium compounds are the most promising NIR emitters for
OLEDs since they combine rather high quantum yields and
narrow emission bands. Despite their obvious importance, the
efficiency of NIR OLEDs is still very low: the highest efficiency of
a Yb-based solution-processed OLED reaches 50 μW W−1 21
,
while
the typical value lies below 10 μW W−1 22–24
.
The TGA data with evolved gas mass-detection and 1H NMR
spectroscopy were used to determine the number of neutral
ligands and solvent molecules in the composition of the com-
plexes. The TGA data of Eu(pfb)3(BPhen)·Solv (Fig. 3a) revealed
that its decomposition occurs in three stages: first, in the
range of 150–200 °C solvent molecules are eliminated, which
results in the formation of Eu(pfb)3(BPhen). Then at 200 °C
neutral ligand elimination begins, and after that thermal
decomposition of the complex occurs with the formation of
the oxide Eu2O3. Mass detection demonstrated the presence of
ethanol molecules in the composition of the complex Eu
(pfb)3(BPhen)·Solv in contrast to other complexes. By analyzing
the first stage of weight loss, we determined that the complex
contains water and ethanol solvent molecules, which corres-
pond to the composition Eu(pfb)3(BPhen)·Solv, Solv = H2O +
EtOH.
In the present paper we propose an approach to the design
of emitter materials for solution-processed lanthanide-based
OLED devices, in particular ytterbium-based. This approach is
based on the purposeful selection of ligands for the lantha-
nide compounds, which are able to ensure (1) efficient lantha-
nide luminescence sensitization, (2) high charge carrier mobi-
lity, and (3) high solubility.
To ensure solubility the pentafluorobenzoate anion was
selected as an anionic ligand, due to the high solubility of
lanthanide pentafluorobenzoates in different organic solvents
and even in water.25 The derivatives of 1,10-phenanthroline
(Fig. 1) were selected as neutral ligands, which combine elec-
tron transport properties and the sensitizing ability.
The chosen ligands BPhen, PyPhen, DPPZ and BDPZ are
expected to possess high electron mobility due to their accep-
tor properties, heteroaromatic nature and high conjugation;
moreover, BPhen is a well-known electron transport material.26
The conjugation degree increases in the PyPhen–BPhen–
DPPZ–BDPZ row, allowing us to expect a decrease in the
triplet energy, and as a result DPPZ and BDPZ sensitize
ytterbium luminescence in the NIR range, while BPhen
and PyPhen sensitize europium luminescence in the visible
region. Thus Yb(pfb)3(DPPZ) and Yb(pfb)3(BDPZ), as well as
Eu(pfb)3(BPhen)·Solv and Eu(pfb)3(PyPhen), were investigated
as materials for solution-processed OLEDs.
The TGA data of Eu(pfb)3(PyPhen), Yb(pfb)3(DPPZ), and Yb
(pfb)3(BDPZ) complexes revealed the absence of coordinated
solvent molecules as confirmed by the absence of weight loss
and the corresponding ionic currents at temperatures below
ca. 200 °C. The decomposition of all three complexes occurs in
two stages: the first stage at ca. 150–250 °C corresponds to the
elimination of the neutral ligands PyPhen, DPPZ and BDPZ,
while at the second stage thermal decomposition of the
complex occurs, and the weight loss ratio of the two stages
corresponds to the ligand ratio [Q] : [pfb−] = 1 : 3 (Fig. 3b–d),
thus showing that only one neutral ligand is coordinated by
the lanthanide ion.
The ratio of [Q] : [pfb−] = 1 : 3 (Q = BPhen, PyPhen, DPPZ,
Results and discussion
Synthesis and characterization
1
BDPZ) in the complex composition was also confirmed by H
NMR spectroscopy. The 1H NMR spectra of Eu
(pfb)3(BPhen)·Solv, Eu(pfb)3(PyPhen) and Yb(pfb)3(Q) (Q =
DPPZ, BDPZ) corresponded to the 1H NMR spectra of Q =
BPhen, PyPhen, DPPZ, and BDPZ, respectively, and were
slightly broadened due to the effect of the Eu3+ and Yb3+ para-
magnetic ions. The absence of protons in the anionic pfb−
ligand does not allow the determination of the [pfb−] : [Q]
ratio directly, and therefore we used 1H NMR spectroscopy
with an external standard.27 Trioctylphosphine oxide (TOPO)
was selected as an external standard, because it contains only
Mixed-ligand complexes Eu(pfb)3(BPhen)·Solv and Eu(pfb)3
(PyPhen) were obtained by the addition reaction by varying the
reagent molar ratio n = 1 or 3:
EuðpfbÞ3ðH2OÞ þ nQ ! EuðpfbÞ3ðQÞ ꢀ Solv #
Mixed-ligand complexes Yb(pfb)3(Q) (Q = DPPZ, BDPZ) were
obtained by the addition reaction with the reagent molar ratio
[Yb(pfb)3] : [Q] = 1 : 1:
YbðpfbÞ3 þ Q ! YbðpfbÞ3ðQÞ #
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17298–17309 | 17299

View Article Online
Paper
Dalton Transactions
Fig. 3 TGA curves (red) of (a) Eu(pfb)3(BPhen)·Solv, (b)
Eu(pfb)3(PyPhen), (c) Yb(pfb)3(DPPZ), and (d) Yb(pfb)3(BDPZ). Normalized
ionic currents are shown in blue (m/z = 18), yellow (m/z = 44), and green
(m/z = 46).
and BDPZ and the signals of TOPO in 1H NMR spectra con-
firmed that all the complexes contained one neutral ligand per
lanthanide ion (Fig. 4 and Table 1).
Fig. 2 Normalized PXRD patterns of (a) Eu(pfb)3(BPhen)·Solv,
(b) Eu(pfb)3(PyPhen), (c) Yb(pfb)3(DPPZ), and (d) Yb(pfb)3(BDPZ).
Luminescence properties
The selection of “neutral ligand–lanthanide ion” pairs is
alkyl protons with the shifts in the range of 0–2 ppm which aimed at the possibility of the neutral ligand to sensitize the
shall not overlap with the signals of the aromatic protons of corresponding lanthanide luminescence, which depends pri-
the neutral ligands. The analysis of the integral intensity ratio marily on the triplet state energy of the neutral ligand.28 The
of the signals of the neutral ligands Q = BPhen, PyPhen, DPPZ, values of triplet state energies (Table 2) were determined from
17300 | Dalton Trans., 2019, 48, 17298–17309
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Fig. 4 1H NMR spectra of (a) Eu(pfb)3(BPhen)·Solv, (b) Eu(pfb)3(PyPhen), (c) Yb(pfb)3(DPPZ), and (d) Yb(pfb)3(BDPZ) with an external standard
(TOPO) in DMSO-d6 (a, b and d) and in CDCl3 (c).
Table 1 Integral intensity ratios of proton signals (calculated and
found)
ligand (Fig. 5a), the latter indicating a slightly different coordi-
nation environment of the europium ions in Eu(pfb)3(Q)·Solv
and the ytterbium ions in Yb(pfb)3(Q).
The Q and TOPO proton signal ratios
The excitation spectra of Eu(pfb)3(BPhen)·Solv and
Eu(pfb)3(PyPhen) indicate the crucial role of the neutral ligand
in luminescence sensitization. Indeed, the spectra are domi-
nated by the through-BPhen (band maximum at 360 nm) and
through-PyPhen (band maximum at 340 nm) bands, while the
through-pfb− excitation band (band maximum at 280 nm) is
much less intense. Narrow excitation bands correspond to the
direct through-europium excitation (Fig. 5b). The difference in
the ratio between the through-europium and through-neutral
ligand bands is probably connected with the different absorp-
Calculated for
Ln(pfb)3(Q)
Calculated for
Ln(pfb)3(Q)2
Found
Q = BPhen
Q = PyPhen
Q = DPPZ
Q = BDPZ
16 : 108
8 : 152
10 : 144
12 : 183
32 : 140
16 : 187
20 : 178
24 : 236
16 : 103
8 : 152
10 : 154
12 : 190
the phosphorescence spectra of gadolinium complexes (see
ESI†) and coincided well with the literature data.29 Due to the tion coefficients of BPhen and PyPhen.
increase of the conjugation length in the row PyPhen–BPhen–
DPPZ–BDPZ, the energy of the triplet state expectedly
The quantum yield calculation demonstrated that the intro-
duction of a neutral ligand results in a 3- to 4-fold increase of
decreases, and therefore both BPhen and PyPhen are able to the luminescence efficiency compared to Eu(pfb)3(H2O)25
sensitize the Eu3+ luminescence (E(5D0) = 17 200 cm−1), while
reaching 55% (Table 3). The main reason for this is the high
DPPZ and BDPZ triplet state energies are too low. However,
sensitization efficiency of BPhen and PyPhen, which reaches
they might be able to sensitize the luminescence of Yb3+ 100% for both ligands (Table 3). The complexes of BPhen,30,31
(E(2F5/2) = 10 000 cm−1).
including fluorobenzoates,32,33 are known for their high
quantum yields due to efficient through-BPhen sensitization,
while PyPhen is less known to be an efficient sensitizer of
europium luminescence.
The complex Yb(pfb)3(BDPZ) could only be dissolved upon
heating and at very low concentrations in strong donor and
non-volatile solvents such as DMF and DMSO, which hinders
Indeed, both Eu(pfb)3(BPhen)·Solv and Eu(pfb)3(PyPhen)
possess typical europium ionic luminescence with no
additional bands of ligand luminescence, and Yb(pfb)3(Q) (Q
= DPPZ and BDPZ) possess typical ytterbium ionic lumine-
scence. It should be noted that the shape of the luminescence
spectra depended both on the emitting ion and on the neutral
Table 2 Triplet state energies of the ligands and excited state energies of Eu3+, Yb3+
Eu3+
Yb3+
BPhen
PyPhen
DPPZ
BDPZ
17 200 cm−1
10 000 cm−1
T1 = 19 900 cm−1
T1 = 21 900 cm−1
T1 = 16 400 cm−1
T1 = 14 000 cm−1
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17298–17309 | 17301

View Article Online
Paper
Dalton Transactions
Fig. 5 (a) Normalized luminescence spectra of Eu(pfb)3(BPhen)·Solv, Eu(pfb)3(PyPhen), Yb(pfb)3(DPPZ), and Yb(pfb)3(BDPZ). (b) Normalized exci-
tation spectra of Eu(pfb)3(BPhen)·Solv, Eu(pfb)3(PyPhen), and Eu(pfb)3(H2O).
Table 3 Luminescence properties of europium complexes
either quantum chemical calculations or cyclic voltammetry
(CV).34 However, both methods provide a rather rough esti-
a
b
c
Ln
Ln
%
PLQY,
%
τobs
ms
,
τrad
ms
,
Q
,
ηsens ,
%
mation: theoretical calculations are valid for single molecules
under vacuum, while CV is carried out in solution, which also
provides high error. The most practically significant values of
frontier orbital energies for OLED applications are those
experimentally determined for thin films, since the emitter is
used in OLEDs in thin film form.
Complex
Eu(pfb)3(H2O)25
Eu(pfb)3(BPhen)·Solv
Eu(pfb)3(PyPhen)
15
40
55
0.65
1.0
1.5
1.0
2.2
2.8
65
45
54
23
89
98
a
b
Itot
0ꢁ1
tobs
trad
ꢁ1
Ln
Ln
trad ¼ 1465 sꢁ1 ꢂ n3 ꢂ I
.
Q
¼
ꢂ 100% . c PLQY = ηsens
×
Ln
Ln
Q
.
Therefore, in the present study the HOMO energies of Eu
(pfb)3(BPhen), Eu(pfb)3(PyPhen), and Yb(pfb)3(DPPZ) were
determined as the ionization energies of the films using
the photoemission yield spectroscopy (PYS) method.35 For
Yb(pfb)3(BDPZ) this measurement was impossible due to its
insufficient film thickness and quality resulting from its low
solubility.
the use of solution methods of research. So, Yb(pfb)3(BDPZ)
was excluded from further discussion.
Mobility and boundary orbital energy determination
Materials for OLED emitters must possess not only high
Ionization energies calculated from the photoemission
quantum yield, but also high charge carrier mobility, as well yield spectra Y2/5(hν) of Eu(pfb)3(BPhen), Eu(pfb)3(PyPhen)
as appropriate energies of the HOMO and LUMO. The most (see ESI†), and Yb(pfb)3(DPPZ) (Fig. 6a) are presented in
common methods to determine frontier orbital energies are Table 4. The obtained values are ca. 1.3 eV lower than those
Fig. 6 (a) The photoemission yield of Yb(pfb)3(DPPZ) depending on photon energy. (b) The spectral dependence of the quantum efficiency of
photoconductivity β(hν) of ITO/Yb(pfb)3(DPPZ)/Al.
17302 | Dalton Trans., 2019, 48, 17298–17309
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Table 4 HOMO, LUMO, and Eg energy values
which is in principle suitable for thin films with low charge
mobility, has never been successfully used for the solution pro-
cessed films of the lanthanide complexes.
HOMO, eV
LUMO, eV
Energy gap, eV
1.7a
1.8a
2.7c
—
n.a./3.5b
n.a./3.2b
2.6d/3b
—
In the present work we succeeded in determining both the
electron and hole mobilities of Eu(pfb)3(BPhen), which is the
first example of such an experiment, while larger conjugation
lengths of other neutral ligands allow us to expect that their
mobilities exceed those of Eu(pfb)3(BPhen). The electron
mobility of Eu(pfb)3(BPhen) μe = 3.33 × 10−5 cm2 V−1 s−1 is an
order of magnitude larger than its hole mobility μh = 1.87 ×
10−6 cm2 V−1 s−1. This was expected since the mobility of the
complex is due to the coordinated BPhen molecule, which is
Eu(pfb)3(BPhen)
Eu(pfb)3(PyPhen)
Yb(pfb)3(DPPZ)
Yb(pfb)3(BDPZ)
5.2
5.0
5.3
—
a Calculated as LUMO = HOMO − Eegxp
tion spectra. c Measured from photoconductivity measurements.
d Calculated as Eg = HOMO − LUMOexp
.
b Eegxp estimated from absorp-
.
obtained for free BPhen, PyPhen, and DPPZ molecules,36,37 known to be an electron transport material (μe(BPhen) = 3.22 ×
due to complex formation. Expectedly, the HOMO energies of 10−4 cm2 V−1 s−1).44 It is however important to note that
Eu(pfb)3(BPhen), Eu(pfb)3(PyPhen), and Yb(pfb)3(DPPZ) are BPhen’s electron mobility is larger than that of Eu
mainly localized on the anionic ligand, as observed in ref. 29 (pfb)3(BPhen), which makes the direct measurements of the
for Eu(TTA)3(Q) (Q = PyPhen, DPPZ, BDPZ), where the HOMO mobility of the complex important instead of its estimation as
energies of the complexes were also lower than the HOMO the neutral ligand mobility value. At the same time the
energies of PyPhen, DPPZ, and BDPZ. This indicates that the obtained μe value is high enough and comparable to the μe of
boundary orbital energy values of the neutral ligands cannot electron transport materials such as TPBi.44 Thus, the task of
be used for those of the complexes containing these ligands.
material design, combining high quantum yield and mobility,
The LUMO energy is calculated as LUMO = HOMO − Eg, was fulfilled.
where Eg is the HOMO–LUMO energy gap. The simplest
method to determine Eg is from the beginning of the first
Host selection
absorption band,38,39 but usually this value is overestimated if
According to the mobility data complexes possess high elec-
the direct HOMO–LUMO transition is forbidden. A more
correct Eg value may be obtained from photoconductivity (PC)
measurements,40 however, this method requires thick
(>500 nm) high quality smooth films, which cannot be
obtained for nonvolatile compounds. In the present work we
succeeded in obtaining films of enough quality for the photo-
conductivity measurements, but only for Yb(pfb)3(DPPZ)
(Fig. 6b), while the Eg values of Eu(pfb)3(BPhen) and
Eu(pfb)3(PyPhen) were obtained from absorption spectra (see
ESI†). To verify the correctness of this experiment for this class
of compounds, i.e. phenanthroline derivative containing
ternary lanthanide pentafluorobenzoates, the Eg value of
Yb(pfb)3(DPPZ) was calculated from both PC and absorption
experiments. As expected, the real Eg value of Yb(pfb)3(DPPZ),
obtained from PC data, is 0.4 eV lower than the value obtained
from the absorption spectrum. This value can be considered
as the experimental error of LUMO energy determination for
the europium complexes.
Determination of charge carrier mobility for nonvolatile
lanthanide complexes is usually impossible. Indeed, the “time-
of-flight” method requires high thickness of the film
(>1 µm),41 which is impossible to obtain with high quality
from solutions of small molecules. The mobility measure-
ments of organic field-effect transistors (FETs) require the
absence of charge carrier traps;42 moreover, the FET mobility
reflects the transport of charge carriers along the substrate,
while in OLEDs charge carriers move across the film in the
perpendicular direction to the substrate. Due to the mor-
phology of the film (molecular orientation, packing, etc.), the
mobility of charge carriers can be different in mutually per-
pendicular directions.43 Even the photo-CELIV method,41
tron and low hole mobilities. To increase the latter and thus to
shift the recombination zone in the emitter film from the
emitter/hole transport layer (HTL) border host : emitter films
containing hole transporting hosts have to be used. In
addition, to achieve good transport properties, these hosts
have to ensure high efficiency energy transfer to the emitter
material, which was investigated in the present section for the
example of the Eu(pfb)3(BPhen) film.
As hole-transporting hosts commonly used CBP and TCTA
were selected, as well as composite hosts CBP : OXD-7
(67 : 33 wt%) and CBP : TCTA (67 : 33 wt%), developed for solu-
tion-processed emitters.45 Although OXD-7 is electron trans-
porting and cannot be used in OLEDs alone, it was also tested
to determine its sensitizing properties. Thus composite
films host : Eu(pfb)3(BPhen) (10 wt%), as well as the pure
Eu(pfb)3(BPhen) film, were deposited from chloroform.
The quantum yield of the pure Eu(pfb)3(BPhen) film equals
80%, which is higher than that of Eu(pfb)(BPhen)·Solv
powder, obviously due to the absence of the coordinated
quenching water molecules. The composite films demon-
strated much lower quantum yields, with the highest yield
(53%) obtained for OXD-7 and the lowest yield (∼0%) obtained
for TCTA, while for CBP, CBP : OXD-7, and CBP : TCTA the
quantum yield was equal to 7% (Table 5). Therefore, CBP and
Table 5 Quantum yields of host : emitter films, 5%
Host
Pure
80
CBP
7
TCTA
0
OXD-7
53
CBP : OXD-7
7
CBP : TCTA
7
QY, %
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17298–17309 | 17303

View Article Online
Paper
Dalton Transactions
CBP : OXD-7 were selected as host materials for OLED fabrica-
tion (Fig. 7).
OLED fabrication
Among the four obtained ternary complexes, one europium
complex Eu(pfb)3(BPhen)·Solv and one ytterbium complex
Yb(pfb)3(DPPZ) were tested in OLEDs, because they combined
all the required properties: high luminescence intensity, high elec-
tron mobility, and high solubility in a number of low boiling
temperature solvents, i.e. chloroform, acetone, THF, etc.
Initially red-emitting Eu(pfb)3(BPhen)·Solv was tested in
OLEDs; HTL and ETL layers were selected based on the values
of HOMO and LUMO energies (Table 6). OLED A, based on a
host-free Eu(pfb)3(BPhen) emissive layer, demonstrated typical
ionic europium luminescence and rather high current density
(Fig. 9) thanks to the purposeful combination of high
quantum yield and electron mobility. However, its luminance
was low due to both solution processing and low hole mobility.
Indeed, the use of the composite CBP : Eu(pfb)3(BPhen)
(50 wt%) film (OLED B) demonstrated higher luminance
despite a much lower quantum yield, showing the crucial
importance of the transport properties of the materials for
OLED applications. However, an additional wide luminescence
band in the EL spectrum of OLED B, corresponding to the
superposition of the luminescence of TPBi and poly-TPD
layers, and the presence of a hump in the I–V curve of OLED B
indicate imbalanced hole and electron currents, showing that
Fig. 7 Normalized (a) excitation and (b) luminescence spectra of com-
posite films.
Table 6 Europium-based OLED heterostructures, as well as switch-on voltage and maximum luminance values
Device Heterostructure
Uon, V Maximum luminance, cd m−2
A
B
C
D
ITO/PEDOT : PSS/poly-TPD/Eu(pfb)3(BPhen)/TPBi/LiF/Al
ITO/PEDOT : PSS/poly-TPD/CBP : Eu(pfb)3(BPhen) (50 wt%)/TPBi/LiF/Al
ITO/PEDOT : PSS/poly-TPD/CBP : Eu(pfb)3(BPhen) (50 wt%)/OXD-7/LiF/Al
ITO/PEDOT : PSS/poly-TPD/CBP (60 wt%):OXD-7 (30 wt%):Eu(pfb)3(BPhen) (10 wt%)/OXD-7/LiF/Al
7
4
5
5
n/a
2 (10–12 V)
2 (10–12 V)
9 (17 V)
Fig. 8 The energy diagram of europium OLEDs, Red = Eu(pfb)3(BPhen).
17304 | Dalton Trans., 2019, 48, 17298–17309
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
the OLED heterostructure was not optimal. The replacement of
Similarly, OLED F, based on the pure Yb(pfb)3(DPPZ) film,
the ETL material TPBi by OXD-7, which possesses suitable demonstrated typical ytterbium luminescence in the NIR
HOMO and LUMO energies (Fig. 8), allowed us to obtain range and high current density, while the introduction of the
OLED C, which demonstrated pure ionic europium lumine- hole-transporting host material CBP led to an increase in its
scence and balanced currents. Further optimization, i.e. host efficiency of up to 30 μW W−1 (Fig. 10). This is one of the
and host : emitter ratio variation, allowed us to achieve a lumi- highest values for the solution-processed Yb-based OLEDs,
nance of 9 cd m−2 for almost pure europium emission of which was exceeded only once in ref. 21, where an efficiency of
OLED D (Table 6).
50 μW W−1 was reached (Table 7).
Fig. 9 (a) Electroluminescence spectra and (b) I–V curves of OLEDs A, B, C, D, and E.
Fig. 10 (a) Electroluminescence spectra and (b) I–V curves of OLEDs E, F, and G.
Table 7 Ytterbium-based OLED heterostructures, as well as switch-on voltage and maximum efficiencies
Device
Heterostructure
Uon, V
Efficiency, µW W−1
E
F
G
ITO/PEDOT : PSS/poly-TPD/Yb(pfb)3(DPPZ)/TPBi/LiF/Al
ITO/PEDOT : PSS/poly-TPD/CBP : Yb(pfb)3(DPPZ) (50 wt%)/TPBi/LiF/Al
ITO/PEDOT : PSS/CBP : Yb(pfb)3(DPPZ) (50 wt%)/TPBi/LiF/Al
6
4
5
n/a
30
n/a
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17298–17309 | 17305

View Article Online
Paper
Dalton Transactions
2 nm was changed using an MYM-1 diffraction grating mono-
chromator. The light/dark cycles were switched using a
Newport 76993 shutter. The sample was irradiated through
Experimental section
Materials and methods
All solvents and chemicals were purchased from commercial the 2 × 15 mm slit in the copper electrode. A short focal
sources. length cylindrical lens was placed between the monochroma-
1H NMR spectra were recorded at 25 °C using an Agilent tor and the quartz window of the vacuum cryostat providing
400 MR spectrometer with an operating frequency of illumination of the 5 × 15 mm large area of the sample. A
400.130 MHz. Chemical shifts are reported in ppm relative to voltage of 50 V was applied to the electrodes to improve the
Me4Si (1H). Thermal analysis was carried out on a thermoana- precision of the measurement as the signal was amplified by
lyzer STA 409 PC Luxx (NETZSCH, Germany) in the tempera- one order of magnitude compared to the signal without
ture range of 20–1000 °C in air and at a heating rate of 10° applied voltage. The ITO electrode under the organic film pre-
min−1. The evolved gases were simultaneously monitored cluded the charging of the sample which could have
during the TA experiment using a coupled QMS 403C Aeolos decreased the signal quality. A Keithley617 electrometer with
quadrupole mass spectrometer (NETZSCH, Germany). The a built-in voltage source was used to apply the voltage as well
mass spectra were registered for the species with the following as to measure the electrical current. Photoemission yield
m/z values: 18 (corresponding to H2O), 44 (corresponding to Y(hν) was calculated as:
CO2) and 46 (corresponding to C2H5OH).
IðhvÞ
Powder X-ray diffraction (PXRD) was performed by using
YðhvÞ ¼
;
PðhvÞ
Bruker D8 Advance [λ(Cu-Kα) = 1.5418 Å; Ni filter] and Bruker
D8 Advance Vario diffractometers [λ(Cu-Kα1) = 1.54060 Å;
Ge(111)-monochromator] with a step size of 0.020°. The patterns
were indexed by using SVD-Index46 as implemented in the
TOPAS 4.2 software.47 Then, the powder patterns were refined
by using the Pawley method.
where I(hν) is the number of emitted electrons and P(hν) is
the number of incident photons with the energy of hν.
The relationship between photoemission yield and ioniza-
tion energy Eioniz can be expressed as a power law:
n
YðhνÞ ¼ αðhν ꢁ EionizÞ ;
Emission and excitation spectra were recorded using a
FluoroMax Plus spectrophotometer upon excitation with a
xenon lamp. Luminescence lifetime measurements were
recorded and detected on the same system. The lifetimes are
the averages of at least three independent measurements. All
luminescence decays proved to be perfect single-exponential
functions. Photoluminescence quantum yields in the visible
range were determined with the FluoroMax Plus spectrophoto-
meter at room temperature upon excitation into ligand states
according to an absolute method using an integration sphere.
The modified de Mello et al.48 method requires the measure-
ment of (i) La, the integrated intensity of light exiting the
sphere when the empty cuvette is illuminated at the excitation
wavelength (Rayleigh scattering band); (ii) Lc, the same inte-
grated intensity at the excitation wavelength when the sample
is introduced into the sphere; (iii) Ea the integrated intensity of
the entire emission spectrum of the empty cuvette; and (iv) Ec
the integrated intensity of the entire emission spectrum of the
cuvette with the sample. The absolute quantum yield is then
given by:
where α is the constant showing the amplitude of the signal
and n = 1…3 depending on the studied materials.49 The value
of n = 2 is used in the case of metals,50,51 while n = 2.5…3 is
used in the case of semiconductors.52,53 The value n = 2.5 was
used in the present paper. To obtain Eioniz, Y1/n(hν) was calcu-
lated and plotted depending on photon energy. The linear part
of the Y1/n(hν) curve was extrapolated until Y1/n(hν) = 0 which
allowed the calculation of the ionization energy of the studied
compound.
The absorption spectra of the thin films were recorded
using an Ocean Optics HR4000CG-UV-NIR spectrometer in the
measurement range between 200 nm and 1100 nm.
The adiabatic energy gap Eg of the studied compounds
was determined from the photoconductivity measurements
using a similar setup as for the PYS measurements. Instead of
the electron collecting electrode, we used an aluminum electrode
deposited on the film of the investigated complex (ITO/complex/
Al).
The threshold energy of photoconductivity Eth can be esti-
mated from the spectrum of the quantum efficiency of the
photoconductivity β(hν). The value of β(hν) was estimated
according to the equation:54
Ec ꢁ Ea
PLQY ¼
ꢂ 100%
La ꢁ Lc
Photoemission yield spectroscopy (PYS) was carried out
using a custom-made system to determine the ionization
energy of the thin films. The sample and a copper electrode
which collected the emitted electrons were placed in a
jphðhv; UÞ
βðhv; UÞ ¼
;
ð1Þ
kðhvÞIðhvÞgðhvÞ
vacuum chamber. During the measurement a pressure of 1 × where jph is the density of photocurrent at a given photon
10−5 mbar was ensured in the chamber. There was an about energy hν and applied voltage U, I(hν) is the intensity of light
2 cm gap between the copper electrode and the sample. An (phot per cm2 per s), k(hν) is the transmission of the semitran-
ENERGETIQ Laser Driven Light Source (LDLSEQ-99) was used sparent electrode, g(hν) is the coefficient that characterizes the
as the light source. The wavelength with a spectral width of absorbed light in the organic layer.
17306 | Dalton Trans., 2019, 48, 17298–17309
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
It has been shown that in anthracene-type crystals the spec- Finally, emission layers were spin-coated from chloroform (c =
tral dependence of β(hν) in the near threshold region can be 7 g L−1 for europium complexes and 5 g L−1 for ytterbium com-
approximated using:54
plexes) at 1500 rpm for 1 min.
After deposition the substrates were transferred into an
argon glove box where emission layers were dried at 80 °C for
20 minutes. The ∼20 nm-thick electron-transporting/hole-
blocking layers TPBi (Lumtec Corp.) and OXD-7 (Ossila) were
thermally evaporated (Univex-300, LeybordHeraeus) followed
by an ∼1 nm-thick LiF layer and a >100 nm-thick aluminum
layer as the cathode under a pressure of below 10−5 mbar. The
thicknesses of the layers were controlled by a quartz micro-
balance resonator pregraded by atomic force microscopy.
Measurements of the OLED characteristics were performed
in the argon glove box. Electroluminescence spectra were
obtained using an Ocean Optics Maya 2000 Pro CCD spectro-
meter sensitive within 200–1100 nm. I–V curves were measured
using two DT 838 digital multimeters. The luminance was
n
βðhνÞ ¼ Aðhν ꢁ EthÞ ;
ð2Þ
where n ≈ 5/2, Eth is the threshold of intrinsic conductivity and
A is the coefficient. The value of the photoconductivity threshold
Eth correlates with the value of the adiabatic energy gap.54
The mobility of the thin films was measured using the
MIS-CELIV method, which consists in registering a transient
current signal with a linearly increasing voltage of the electric
field applied to the sample.55,56 Samples for determining
mobility were prepared as follows: a glass substrate with
indium tin oxide (ITO) coated with a dielectric layer of SiO2
(70 nm) was successively applied as a layer of the test com-
pound (d = 100 nm) and an aluminum layer (80 nm) as the
opposite electrode. The SiO2 layer is blocking for both types of
charge carriers, i.e. it prevents carrier injection from ITO.
When measuring the transient current of holes, a linearly
increasing positive potential was applied to ITO; therefore,
extraction of holes took place on the aluminum electrode. The
SiO2 layer prevented the injection of holes from the ITO elec-
trode. In this case, the contribution of the electron current to
the signal can be neglected, since the SiO2 layer also blocked
it. To measure the electron mobility, the polarity of the voltage
applied to the sample was reversed.
measured using
a
TKA-PKM luminance meter (TKA
Instruments). OLED NIR luminance power was determined
using a precalibrated Coherent FieldMaxII Laser Power Meter
with an optical filter removing the visible part of the spectra.
NIR efficiency was calculated as the ratio of measured lumi-
nance power to consumed electrical power.
Synthesis
Synthesis of PyPhen, DPPZ, and BDPZ ligands. The starting
compound 1,10-phenanthroline-5,6-dione was synthesized
according to the literature57 with 64% yield.
The characteristic time tmax corresponding to the maximum
transient current value was determined from the transient
current curve. Typical transient current signals in the layers of
organic semiconductors prepared by casting from a solution
were previously described.56 The mobility of charge carriers (µ)
was calculated according to the formula (3).
PyPhen.58
1,10-Phenanthroline-5,6-dione
(876
mg,
4.17 mmol) was stirred with 1,2-ethylenediamine (0.877 ml,
13.14 mmol) and methanol (44 ml) for 24 h at rt. The red solu-
tion was evaporated to dryness. The residue was recrystallized
from toluene. Yield 741 mg (77%) of a yellow solid.
2 ꢂ d2
A ꢂ ðtmax
3
1H-NMR (CDCl3) 7.80 (dd, J 8.0, 8.2 Hz, 2H), 9.00 (s, 2H),
μ ¼
;
ð3Þ
2
9.30 (dd, 3J 4.4 Hz, 2H), 9.50 (dd, 3J 8.0, 8.2 Hz, 2H).
Þ
DPPZ.59 1,10-Phenanthroline-5,6-dione (1.050 g, 5 mmol) was
mixed with o-phenylenediamine (0.595 g, 5.5 mmol) and
p-TsOH·H2O (9.5 mg, 0.05 mmol) and refluxed in dry ethanol
(15 ml) for 8 h. The reaction mixture was cooled down to rt and
filtered off. The obtained powder was recrystallized from the
where d is the thickness of the compound layer and A is the
applied voltage ramp rate.
OLED manufacture
Prepatterned indium tin oxide coating with 15 Ohm per sq on ethanol–water mixture. Yield 1.222 g (87%) of yellow powder.
the glass substrates (Lumtec Corp.) were used as anodes. were
1H-NMR (DMSO-d6) 7.88–7.91 (m, 2H), 8.01–8.03 (m, 2H),
used as anodes. The substrates were sequentially washed in an 8.30–8.33 (m, 2H), 9.17–9.18 (m, 2H), 9.43–9.45 (m, 2H).
ultrasonic bath with NaOH solution (30 min), distilled water BDPZ. 1,10-Phenanthroline-5,6-dione (0.840 g, 4 mmol) was
(10 min), and 2-propanol (10 min) and then dried under a flow mixed with naphthalene-1,2-diamine (0.695 g, 4.4 mmol) and
of N2. Then a 30 minute UV-treatment was performed in order p-TsOH·H2O (7.6 mg, 0.04 mmol) and refluxed in dry ethanol
to remove residual organic impurities.
(35 ml) for 8 h. The reaction mixture was cooled down to rt, fil-
All the spin-coating deposition processes were performed tered off, and washed with 30 ml of methanol. An orange
under ambient air conditions. A 50 nm-thick PEDOT : PSS powder was recrystallized from DMF, filtered off and washed
(poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate),
Lumtec Corp.) hole-injection layer was deposited by pouring
with diethyl ether. Yield 1.106 g (83%) of orange foam.
1H-NMR (CDCl3) 7.62–7.66 (m, 2H), 7.80 (dd, 3J 8.1 Hz, 2H),
3
300 μl of the solution onto the substrate, followed by rotation 8.20–8.23 (m, 2H), 8.96 (s, 2H), 9.25–9.26 (dd, J 4.4 Hz, 2H),
for 1 min at 2000 rpm. The obtained film was dried at 100 °C 9.65 (dd, 3J 8.1 Hz, 2H).
for 30 min. A 20 nm-thick hole-transporting poly-TPD (Ossila)
Synthesis of Eu(pfb)3(H2O) and Yb(pfb)3 complexes. An
was spin-coated from chlorobenzene (5 g L−1 concentration) at excess of concentrated aqueous solution of NH3 (6.5 ml, 20.0
2000 rpm for 1 min and then dried at 100 °C for 10 min. equiv.) was added to an aqueous solution of LnCl3·6H2O (Ln =
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17298–17309 | 17307

View Article Online
Paper
Dalton Transactions
Eu, Yb) (5 mmol, 1 equiv.) in water (25 ml). The mixture was gated complexes were measured, as well as charge carrier
stirred for 30 min, and the precipitate of Ln(OH)3 (Ln = Eu, mobility, which has never been measured for non-volatile
Yb) was centrifuged and washed with water until the pH of the lanthanide complexes before. Both the proposed design of
washing solution became neutral. A small excess of freshly pre- efficient lanthanide-based emitters and OLED heterostructure
pared Ln(OH)3 was placed in a beaker, and a solution of optimization resulted in obtaining an NIR OLED with the
2,3,4,5,6-pentafluorobenzoic acid (12.5 mmol, 2.5 equiv.) in second highest efficiency among ytterbium-based solution-pro-
ethanol (25 ml) was added. The reaction mixture was stirred cessed OLEDs (30 μW W−1).
for 1 h at 60 °C and the unreacted components were separated
by filtration followed by the evaporation of the clear solution
to dryness. Yield 90% of white powder. – CHN: Eu(pfb)3(H2O), Conflicts of interest
calcd, % C 31.40, H 0.25, found C 31.32, H 0.34, Yb(pfb)3,
There are no conflicts to declare.
calcd, % C 31.28, found C 31.38.
Synthesis of Eu(pfb)3(BPhen)·Solv and Eu(pfb)3(PyPhen).
The solution of Eu(pfb)3(H2O) (1 mmol) in 20 ml of ethanol
was added to the solution of one or three equivalents of Q =
Acknowledgements
BPhen and PyPhen (1 mmol or 3 mmol) in 30 ml of ethanol.
The reaction mixture was refluxed for 24 h and then was evap-
This work was supported by Russian Foundation for Basic
Research (18-33-20210). UV thanks President’s grant
orated to half-volume until precipitate formation. The precipi-
(MK-2799.2019.3) for financial support.
tate was filtered off, washed with ethanol, and dried in air.
Yield 70% of white powders Eu(pfb)3(BPhen)·Solv and Eu
(pfb)3(PyPhen). – CHN: Eu(pfb)3(BPhen)·Solv (1 : 1), calcd, % C
48.36, H 1.44, N 2.51, found C 48.48, H 1.60, N 2.43,
Notes and references
Eu(pfb)3(PyPhen) (1 : 1), calcd, % C 41.32, H 0.79, N 5.51,
found C 41.19, H 0.96, N 5.66, Eu(pfb)3(BPhen)·Solv (1 : 3),
calcd, % C 48.36, H 1.44, N 2.51, found C 48.54, H 1.53,
N 2.62, Eu(pfb)3(PyPhen) (1 : 3), calcd, % C 41.32, H 0.79,
N 5.51, found C 41.22, H 0.88, N 5.44.
1 N. Thejo Kalyani and S. J. Dhoble, Renewable Sustainable
Energy Rev., 2015, 44, 319–347.
2 L. Duan, L. Hou, T.-W. Lee, J. Qiao, D. Zhang, G. Dong,
L. Wang and Y. Qiu, J. Mater. Chem., 2010, 20, 6392.
3 L. Deng, W. Li and J. Li, Displays, 2013, 34, 413–417.
4 T. W. Lee, T. Noh, H. W. Shin, O. Kwon, J. J. Park,
B. K. Choi, M. S. Kim, D. W. Shin and Y. R. Kim, Adv. Funct.
Mater., 2009, 19, 1625–1630.
5 L. Wang, Z. Zhao, C. Wei, H. Wei, Z. Liu and Z. Bian, Adv.
Opt. Mater., 2019, 7, 1801256.
6 Y. Chi, T.-K. Chang, P. Ganesan and P. Rajakannu, Coord.
Chem. Rev., 2017, 346, 91–100.
7 M. L. P. Reddy and K. S. Bejoymohandas, J. Photochem.
Photobiol., C, 2016, 29, 29–47.
8 J. Jayabharathi, K. Jayamoorthy and V. Thanikachalam,
J. Organomet. Chem., 2014, 761, 74–83.
9 Y. Li, J. Y. Liu, Y. Di Zhao and Y. C. Cao, Mater. Today, 2017,
20, 258–266.
Synthesis of Yb(pfb)3(Q) (Q = DPPZ, BDPZ). The solution of
Yb(pfb)3 (1 mmol) in 20 ml of ethanol was added to the solu-
tions of Q = DPPZ and BDPZ (1 mmol) in 30 ml of ethanol. The
reaction mixture was refluxed for 24 h and then was evaporated
to half-volume until precipitate formation. The precipitate was
filtered off, washed with ethanol, and dried in air. Yield 60% of
white powder Yb(pfb)3(DPPZ) and 50% of orange powder
Yb(pfb)3(BDPZ). – CHN: Yb(pfb)3(DPPZ), calcd, % C 43.03,
H 0.93, N 5.15, found C 43.18, H 0.85, N 5.28, Yb(pfb)3(BDPZ),
calcd, % C 45.36, H 1.06, N 4.92, found C 45.23, H 1.15, N 4.99.
Synthesis of GdCl3(Q) (Q = BPhen, PyPhen, DPPZ, BDPZ).
GdCl3(Q) (Q = BPhen, PyPhen, DPPZ, BDPZ) were synthesized
by reaction between solutions of GdCl3 and BPhen, PyPhen,
DPPZ, and BDPZ in ethanol. The reaction mixture was refluxed 10 Z. Liu, J. Qiu, F. Wei, J. Wang, X. Liu, M. G. Helander,
for 24 h and then was evaporated to half-volume until precipi-
tate formation. The precipitate was filtered off, washed with
ethanol, and dried in air.
S. Rodney, Z. Wang, Z. Bian, Z. Lu, M. E. Thompson and
C. Huang, Chem. Mater., 2014, 26, 2368–2373.
11 S. Gong, J. Luo, Z. Wang, Y. Li, T. Chen, G. Xie and C. Yang,
Dyes Pigm., 2017, 139, 593–600.
12 V. R. Sastri, J. R. Perumareddi, V. R. Rao, G. V. S. Rayudu
and J. C. Bünzli, Modern aspects of rare earths and their com-
plexes, Elsevier, Amsterdam, 2003.
Conclusions
Thus, we proposed the targeted design of efficient lanthanide- 13 V. Utochnikova, A. Kalyakina and N. Kuzmina, Inorg. Chem.
based emitters for solution-processed OLEDs, which was Commun., 2012, 16, 4–7.
aimed at the combination of high luminescence efficiency 14 V. V. Utochnikova, A. S. Kalyakina, I. S. Bushmarinov,
with solubility and charge carrier mobility. New highly lumi-
nescent europium-based materials with PLQYs of up to 80%
and purely NIR luminescent ytterbium complexes were
A. A. Vashchenko, L. Marciniak, A. M. Kaczmarek, R. Van
Deun, S. Bräse and N. P. Kuzmina, J. Mater. Chem. C, 2016,
4, 9848–9855.
obtained and thoroughly characterized. The HOMO and 15 A. S. Kalyakina, V. V. Utochnikova, E. Y. Sokolova,
LUMO energies of solution-processed thin films of the investi-
A. A. Vashchenko, L. S. Lepnev, R. Van Deun, A. L. Trigub,
17308 | Dalton Trans., 2019, 48, 17298–17309
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Y. V. Zubavichus, M. Hoffmann, S. Mühl and 35 R. Grzibovskis and A. Vembris, Study of the P3HT/PCBM
N. P. Kuzmina, Org. Electron., 2016, 28, 319–329.
16 H. Suzuki, J. Photochem. Photobiol., A, 2004, 166,
155–161.
17 Á. L. Álvarez and C. Coya, Lanthanide-Based Multifunctional
Materials, Elsevier, 2018, pp. 133–170.
Interface Using Photoemission Yield Spectroscopy, 2016, vol.
9895, pp. 1–8.
36 C. Li, M. Schwab, Y. Zhao, L. Chen, I. Bruder, I. Münster,
P. Erk and K. Müllen, Dyes Pigm., 2013, 97, 258–261.
37 L. Zhang, S. Yue, B. Li and D. Fan, Inorg. Chim. Acta, 2012,
384, 225–232.
18 M. A. Katkova, A. P. Pushkarev, T. V. Balashova,
A. N. Konev, G. K. Fukin, S. Y. Ketkov and M. N. Bochkarev, 38 S. Janietz, D. D. C. Bradley, M. Grell, C. Giebeler and
J. Mater. Chem., 2011, 21, 16611–16620. M. Inbasekaran, Appl. Phys. Lett., 1998, 2453, 24–27.
19 T. V. Balashova, N. A. Belova, M. E. Burin, D. M. Kuzyaev, 39 E. Girotto, A. Pereira, C. Arantes, M. Cremona,
R. V. Rumyantcev, G. K. Fukin, A. P. Pushkarev,
V. A. Ilichev, A. F. Shestakov, I. D. Grishin and
M. N. Bochkarev, RSC Adv., 2014, 4, 35505–35510.
A. J. Bortoluzzi, C. A. M. Salla, I. H. Bechtold and
H. Gallardo, J. Lumin., 2019, 208, 57–62.
40 R. Grzibovskis, J. Latvels and I. Muzikante, IOP Conf. Ser.:
Mater. Sci. Eng., 2011, 23, 12021.
20 V. A. Ilichev, A. V. Rozhkov, R. V. Rumyantcev, G. K. Fukin,
I. D. Grishin, A. V. Dmitriev, D. A. Lypenko, E. I. Maltsev, 41 A. Kokil, K. Yang and J. Kumar, J. Polym. Sci., Part B: Polym.
A. N. Yablonskiy, B. A. Andreev and M. N. Bochkarev,
Dalton Trans., 2017, 46, 3041–3050.
21 A. Kovalenko, P. O. Rublev, L. O. Tcelykh,
A. S. Goloveshkin, L. S. Lepnev, A. S. Burlov,
Phys., 2012, 50, 1130–1144.
42 N. Karl, K. Kraft, J. Marktanner, M. Münch, F. Schatz,
R. Stehle, H. Uhde, F. Schatz, R. Stehle and H. Uhde, J. Vac.
Sci. Technol., A, 1999, 17, 2318–2328.
A. A. Vashchenko, Ł. Marciniak, A. M. Magerramov, 43 H. Sirringhaus, P. J. Brown, R. H. Friend, M. M. Nielsen,
N. G. Shikhaliyev, S. Z. Vatsadze and V. V. Utochnikova,
Chem. Mater., 2019, 31, 759–773.
22 T.-S. Kang, B. S. Harrison, T. J. Foley, A. S. Knefely,
K. Bechgaard, B. M. W. Langeveld-Voss, A. J. H. Spiering,
R. A. J. Janssen, E. W. Meijer, P. Herwig and D. M. de
Leeuw, Nature, 1999, 401, 685–688.
J. M. Boncella, J. R. Reynolds and K. S. Schanze, Adv. 44 S. H. Rhee, K. b. Nam, C. S. Kim, M. Song, W. Cho, S.-H. Jin
Mater., 2003, 15, 1093–1097. and S. Y. Ryu, ECS Solid State Lett., 2014, 3, R19–R22.
23 T.-S. Kang, B. S. Harrison, M. Bouguettaya, T. J. Foley, 45 X. Zhang, X. Guo, Y. Chen, J. Wang, Z. Lei, W. Lai, Q. Fan
J. M. Boncella, K. S. Schanze and J. R. Reynolds, Adv. Funct.
Mater., 2003, 13, 205–210.
and W. Huang, J. Lumin., 2015, 161, 300–305.
46 A. A. Coelho, J. Appl. Crystallogr., 2003, 36, 86–95.
24 Z. Li, H. Zhang and J. Yu, Thin Solid Films, 2012, 520, 47 Bruker, TOPAS 4.2 User Manual, Bruker AXS GmbH,
3663–3667. Karlsruhe, Germany, 2009.
25 A. S. Kalyakina, V. V. Utochnikova, I. S. Bushmarinov, 48 J. C. De Mello, H. F. Wittmann and R. H. Friend, Adv.
I. V. Ananyev, I. L. Eremenko, D. Volz, F. Rönicke, Mater., 1997, 9, 230–232.
U. Schepers, R. Van Deun, A. L. Trigub, Y. V. Zubavichus, 49 E. O. Kane, Phys. Rev., 1962, 127, 131–141.
N. P. Kuzmina and S. Bräse, Chem. – Eur. J., 2015, 21, 50 C. W. Ow-yang, J. Jia, T. Aytun, M. Zamboni, A. Turak,
17921–17932.
26 I. Wu, P. Wang, W. Tseng, J. Chang and C. Wu, Org. 51 H. Monjushiro, I. Watanabe and Y. Yokoyama, Anal. Sci.,
Electron., 2012, 13, 13–17. 1991, 7, 543–547.
27 S. K. Bharti and R. Roy, TrAC, Trends Anal. Chem., 2012, 35, 52 K. Kanai, M. Honda, H. Ishii, Y. Ouchi and K. Seki, Org.
5–26. Electron., 2012, 13, 309–319.
K. Saritas and Y. Shigesato, Thin Solid Films, 2014, 559, 58–63.
28 M. Latva, H. Takalo, V.-M. Mukkala, C. Matachescu, 53 Y. Gao, Mater. Sci. Eng., R, 2010, 68, 39–87.
J. C. Rodríguez-Ubis and J. Kankare, J. Lumin., 1997, 75, 54 E. A. Silinsh and V. Capek, Organic Molecular Crystals,
149–169.
Interaction, localization and transport phenomena, AIP Press,
New York, 1994.
55 A. Armin, G. Juska, M. Ullah, M. Velusamy, P. L. Burn,
P. Meredith and A. Pivrikas, Adv. Energy Mater., 2014, 4,
1300954.
29 X.-N. Li, Z.-J. Wu, Z.-J. Si, L. Zhou, X.-J. Liu and H.-J. Zhang,
Phys. Chem. Chem. Phys., 2009, 11, 9687–9695.
30 S. Zhang, G. A. Turnbull and I. D. W. Samuel, Org.
Electron., 2012, 13, 3091–3096.
31 T. W. Canzler and J. Kido, Org. Electron., 2006, 7, 56 V. V. Malov, T. Ghosh, V. C. Nair, M. M. Maslov, K. P. Katin,
29–37.
K. N. N. Unni and A. R. Tameev, Mendeleev Commun., 2019,
29, 218–219.
32 V. V. Utochnikova, N. N. Solodukhin, A. N. Aslandukov,
L. Marciniak, I. S. Bushmarinov, A. A. Vashchenko and 57 C. Wang, L. Lystrom, H. Yin, M. Hetu, S. Kilina,
N. P. Kuzmina, Org. Electron., 2017, 44, 85–93.
33 A. S. Kalyakina, V. V. Utochnikova, M. Zimmer, F. Dietrich,
S. A. McFarland and W. Sun, Dalton Trans., 2016, 45,
16366–16378.
A. M. Kaczmarek, R. Van Deun, A. A. Vashchenko, 58 A. Delgadillo, P. Romo, A. M. Leiva and B. Loeb, Helv.
A. S. Goloveshkin, M. Nieger, M. Gerhards, U. Schepers
and S. Bräse, Chem. Commun., 2018, 54, 5221–5224.
34 K. G. von Eschwege, Polyhedron, 2012, 39, 99–105.
Chim. Acta, 2003, 86, 2110–2120.
59 W. Yang, W. Yang, W. Liu and W. Qin, Spectrochim. Acta,
Part A, 2013, 104, 56–63.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17298–17309 | 17309