Leer2 PDF

Volume 53 Number 5 16 January 2017 Pages 793–986
ChemComm
Chemical Communications
rsc.li/chemcomm
ISSN 1359-7345
FEATURE ARTICLE
Adam F. Henwood and Eli Zysman-Colman
Lessons learned in tuning the optoelectronic properties of
phosphorescent iridium(III) complexes
ChemComm
View Article Online
FEATURE ARTICLE View Journal | View Issue
Lessons learned in tuning the optoelectronic

Published on 17 November 2016. Downloaded by Freie Universitaet Berlin on 12/04/2018 09:45:32.
properties of phosphorescent iridium(III)

Cite this: Chem. Commun., 2017,
53, 807 complexes
Adam F. Henwood and Eli Zysman-Colman*
This perspective illustrates our approach in the design of heteroleptic cationic iridium(III) complexes for
optoelectronic applications, especially as emitters in electroluminescent devices. We discuss changes in the
photophysical properties of the complexes as a consequence of modification of the electronics of either the
cyclometalating (C^N) or the ancillary (N^N) ligands. We then broach the impact on these properties as a
Received 16th August 2016, function of modification of the structure of both types of ligands. We explain trends in the optoelectronic
Accepted 17th November 2016 behaviour of the complexes using a combination of rationally designed structure–property relationship
DOI: 10.1039/c6cc06729h studies and theoretical modelling that serves to inform subsequent ligand design. However, we have found
cases where the design paradigms do not always hold true. Nevertheless, all these studies contribute to the
www.rsc.org/chemcomm lessons we have learned in the design of heteroleptic cationic phosphorescent iridium(III) complexes.
1 Introduction
The efficient phosphorescence exhibited by cyclometalated
iridium(III) complexes has made them among the most widely
explored classes of transition metal complexes for photonic and
Organic Semiconductor Centre, EaStCHEM School of Chemistry,
University of St Andrews, St Andrews, Fife, KY16 9ST, UK.
optoelectronic applications1 (behind polypyridyl ruthenium(II)
E-mail: eli.zysman-colman@st-andrews.ac.uk; Web: http://www.zysman-colman.com; complexes), with applications in everything from sensing2 and
Fax: +44 (0)1334 463808; Tel: +44 (0)1334 463826 bioimaging3 to photoredox catalysts for organic transformations4
Adam Francis Henwood graduated Eli Zysman-Colman obtained his

with an integrated Masters degree Ph.D. from McGill University in
from the University of Durham in 2003 under the supervision of
the UK in 2013. He moved north David N. Harpp where he
to St. Andrews to undertake a conducting research in physical
PhD with Eli Zysman-Colman. organic sulfur chemistry. He
His research interests are centred then completed two postdoctoral
on the development of new fellowships, one in supra-
phosphorescent transition metal molecular chemistry with Jay
complexes, such as iridium(III) Siegel at the Organic Chemistry
and platinum(II), for application Institute, University of Zurich and
as phosphors in solid-state the other in inorganic materials
Adam F. Henwood lighting. Eli Zysman-Colman chemistry with Stefan Bernhard
at Princeton University. After six
years in Canada as an assistant professor, in 2013 he moved to the
University of St. Andrews where he is presently Reader in
Optoelectronic Materials. His research program focuses primarily
on the rational design of luminophores for solid-state lighting
based on organic light emitting diode (OLED) and light-emitting
electrochemical cell (LEEC) electroluminescent devices.
This journal is © The Royal Society of Chemistry 2017 Chem. Commun., 2017, 53, 807--826 | 807
View Article Online
Feature Article ChemComm
and as solar fuels.5 In our lab, we have a particular interest in We have often found density functional theory (DFT) calcula-
utilising complexes of the form [Ir(C^N)2(N^N)]PF6, where C^N is tions coupled with a strong background in physical organic
a cyclometalating ligand and N^N is an ancillary diimine ligand, chemistry to be useful tools in aiding us with our interpretations
as materials for solid-state lighting (SSL) technologies, where and have used the insights gained to identify new emitters.
it has been shown that these can act as effective phosphors for However, even with the help of these tools, we have also on
generating light in electroluminescent devices such as organic occasion found that the resultant photophysical properties we
light emitting devices (OLEDs)6 and light-emitting electro- observe are in fact not what we might expect, which keeps this
chemical cells (LEECs).7 field of research both interesting and pertinent.
An important feature of these complexes, particularly for
LEECs, is that their emission can be readily tuned across the

visible spectrum. In many cases, the emissive excited states of 2 Archetypal iridium complexes
these complexes are a mixture of metal-to-ligand charge transfer
(3MLCT) between the metal and N^N ligands and ligand-to- In order to assess the effect of substitution/modification of the
ligand charge transfer (3LLCT) between the phenyl rings of the ligand scaffold about the iridium centre on the optoelectronic
C^N ligands and the N^N ligands. Thus, to a first approximation, properties, it is necessary to define reference complexes
the energies of the HOMOs and LUMOs of these complexes from which to draw comparisons. The cationic iridium(III)
can be independently modulated as a function of appropriate complexes that best fit this purpose are the widely studied
substituent modification of the C^N and N^N ligands. Such complexes 1 and 2, whose photophysical properties were
facile colour tuning is in stark contrast to the narrow orange-red originally reported by Güdel et al.12 We have collated all of
emission range of ruthenium(II) polypyridyl complexes. For the relevant photophysical properties of these two complexes
reference, we include several reviews from other groups exploring in a recent review,7b and thus mention the salient features
colour tuning of iridium complexes.5b,8 only briefly here.
For LEECs, the major challenges that require addressing are Both complexes emit from a mixed 3MLCT/3LLCT state,
the design of red and, especially, blue emitters. The emission luminescing yellow/orange light in MeCN (lPL = 602 nm for
colour perceived by the viewer is normally defined using the 1 and 588 nm for 2; FPL = 6% for 1 and 23% for 2).12e,13 The
system of the Commission Internationale de l’Éclairage (CIE), modest blue-shift in the emission of 2 compared to 1 is due
which plots the colour of the emission as a function of x and y to the inductively electron-donating character of the tert-
coordinates. The CIE coordinates corresponding to deep blue are butyl groups, which destabilise the LUMO located on the
defined as (0.15, 0.06) while the saturated red coordinates are ancillary N^N ligand as well as the first excited triplet state
(0.60, 0.30). To date, no iridium-based emitter in a LEEC has come (T1) of 2 compared to 1. Structural deviation from these
close to the required value for blue – the bluest reported CIE values complexes is normally the result of one or more of three
are (0.20, 0.28)9 – while there is only a single report of a deep blue main substitutions: (1) modification of the aryl rings of
emitting organic fluorophore used for LEECs (CIE: 0.15, 0.09).10 the C^N ligands, (2) modification of the pyridyl or other
For the red emitters, only four devices to date have been reported heterocycle of the C^N ligands and (3) modification of one
to display near the ideal red coordinate, but the lifetimes of these or both of the heterocycles of the N^N ligands (Fig. 1). We will
devices are very short (t1/2 = 2.0–6.3 h, where t1/2 is defined as the study each of these three cases individually in the sections
time taken for the luminance of the device to decay to half to the that follow, before investigating how combining substitution
maximum).11 Thus we have interests in tuning the emission of our at multiple positions about the ligand scaffold can work in
complexes either to the red or to the blue. concert.
Our main philosophy when undertaking any of our studies Fig. 2 illustrates the most common strategy for tuning the
has been to start with understanding the basic properties of emission of heteroleptic cationic iridium complexes. The para-
reference iridium complexes and then in a rational fashion digm for emission tuning described in Fig. 2 works for most,
design a suitable structure–property relationship study where but not all complexes, of the form [Ir(C^N)2(N^N)]+. Density
we can establish trends in behaviour and so better understand Functional Theory (DFT) calculations have shown that the
these new systems. This is the manner in which we will HOMO is comprised of a mixture of the metal and the phenyl
approach this review, by exploring first the photophysics of rings of the C^N ligands.12d,14 The LUMO is localised essentially
archetypal iridium complexes 1 and 2 and then understanding exclusively on the bipyridine in both 1 and 2. Thus, by adding
how these properties change as the deviations in ligand structure the appropriate substituents to the ligands, the emission can
from those present in these traditional complexes become conceivably be tuned from deep red to the blue. However, it is
increasingly elaborate. The strategies for tuning the photo- important to note that as the emission energy is tuned towards
physical properties of this class of iridium complexes will be the extremities of the visible spectrum, the photoluminescence
illustrated largely using example studies coming out of our quantum yield, FPL, of these complexes can often drop off
own group, as this will control for differences in measurement precipitously. For example, the energy gap law states that the
technique or instrumentation from one group to another, rate of non-radiative decay increases exponentially with decreasing
thereby producing a reliable comparison between different emission energy as vibrational modes of the ground state more
complexes. readily couple to those of the excited state,15 which is the primary
808 | Chem. Commun., 2017, 53, 807--826 This journal is © The Royal Society of Chemistry 2017
View Article Online
ChemComm Feature Article

Fig. 1 (a) Structures of [Ir(ppy)2(bpy)](PF6), 1, and its tert-butyl analogue, [Ir(ppy)2(dtbubpy)](PF6), 2, which are two of the most widely studied cationic
iridium complexes. (b) Distinct areas of 1 that are normally individually derivatized, where red is the phenyl ring of the C^N ligands, blue is the pyridyl rings
of the C^N ligands and green is one or both of the pyridines on the N^N ligand. Substitution at each of these regions will be discussed individually,
followed by combinations of substitutions.
Fig. 2 General scheme depicting strategy for colour tuning the emission of [Ir(ppy)2(bpy)]+, 1. DFT calculated Kohn–Sham MOs indicate an electron
density distribution on the HOMO that is largely localised on the metal centre and the phenyl rings of the cyclometalating ligands, and on the LUMO that
is largely localised on the bpy while the calculated spin density of the T1 state is distributed across the entire molecule. This allows for tuning
of the emission by appropriate functionalisation of the ligands. Dashed arrows indicate deactivation from the excited state via non-radiative decay.
DFT [(B3LYP/SBKJC-VDZ for Ir(III)) and (6-31G* for C,H,N)] with CPCM (MeCN).
deactivation mode when tuning the emission towards the red. For emission as a function of the fluorine atoms present on the
iridium complexes that are being tuned towards producing blue C^N ligands, and thus are useful benchmarks for comparing
emission, the energies of the anti-bonding metal-centred (3MC) and new blue emitters. When designing complexes for blue emission,
3
CT (or ligand centred, 3LC) are all very strongly destabilised. the most common strategy is to append electron-withdrawing
However, the relative destabilisation of the 3MC states is less substituents to the phenyl ring of the C^N ligands as this acts
pronounced than the emissive 3CT or 3LC states, resulting in to lower the energy of the HOMO, thereby increasing the
thermal population of these 3MC states becoming increasingly HOMO–LUMO gap and frequently leading to a blue-shift in
likely as a function of an increasing HOMO–LUMO energy gap. the emission. Complex 3 employs the commonly used ligand
Population of these 3MC states leads to efficient and deleterious 2-(2,4-difluorophenyl)pyridine (dFppy) wherein the presence of
quenching of the emission. the two fluorine atoms act to strongly blue-shift the emission
Aside from 1 and 2, we will also consider two additional from the yellow-orange found in 1 and 2 to green (lPL = 515 nm
reference complexes in 3 and 4 that have significantly blue-shifted for 3; FPL = 72% in MeCN).16 It is worth noting that in the LEEC
View Article Online
3 C^N ligand – effects of substitution

on the phenyl ring
In this section we will explore the effect on the photophysics of
the complex when the C^N ligand is substituted with electron-
withdrawing or electron-donating functionalities. Of particular
importance here is the regiochemistry of the substituent with
respect to the Ir–C bond. Hammett values (sm/p), which para-
meterise the electron-withdrawing or donating character of a
substituent based on whether it is situated in a meta or para

position to a particular functional group of interest, are useful
tools in understanding these electronic effects, although there
are instances where these relationships break down.19 We note
that Baranoff and co-workers have established a quantifiable
relationship between Hammett values and several optoelectronic
Fig. 3 Structure of [Ir(dFppy)2(dtbubpy)](PF6), 3, which is a widely studied
properties of neutral iridium(III) complexes.20
green-emitting iridium complex, and its mesityl-substituted analogue 4,
[Ir(dFMesppy)2(dtbubpy)](PF6), which shows the same electronic properties
Many groups have explored incorporating electron-withdrawing
as 3 but with enhanced FPL. substituents onto the C^N ligands in place of or in addition to
fluorine atoms with the goal of blue-shifting the emission.21
Several examples of these groups include: trifluoromethyl,22
device, this complex is among the most efficient emitters sulfonyl,23 phosphonium,24 diarylphosphine oxide23b and nitrile.21,25
reported to date, with an external quantum efficiency (EQE) In addition, there have also been efforts to change the aryl ring of the
of 14.9% (Fig. 3).17 C^N ligands altogether from phenyl to a heterocycle such as
Complex 4 is the final reference complex, and it is structu- pyridine,26 pyridinium27 or even pyrimidine.28 The nitrogen
rally identical to 3 except for the addition of the mesityl units at atoms in these heterocycles stabilise the HOMO by inductively
the 4-position of the pyridine rings on the C^N ligands. This withdrawing electron density away from the metal centre (Fig. 5).
addition has no impact on the optoelectronic properties of Although many blue-emitting iridium complexes that are
the complex, as the mutually orthogonal conformation of the reported contain fluorine atoms appended to the C^N ligands,
mesityl and pyridine rings disrupts any formal conjugation there has been a concerted effort to move away from fluorinated
between the two. Furthermore, the steric bulk of the mesityl C^N ligands. The F–Caryl bond has been identified as a point of
group inhibits intermolecular quenching of the emission in electrochemical instability in these complexes, especially when
solution, resulting in an enhanced FPL observed for 4 compared placed on the C^N ligands, as this can activate the ring to
with 3, while maintaining the same emission colour (lPL = 515 nm, degradation by such processes as nucleophilic aromatic substi-
FPL = 80% for 4).16 This mesityl substitution strategy was first tution in the device.29 Exploring new and more chemically
reported by Bryce and co-workers18 and we have since found it to be robust functional groups or heterocyclic rings is thus an impor-
applicable to a wide variety of systems, many of which will be tant avenue of research. Complexes 5, 6, 7 and 11 are some
discussed here (Fig. 4). examples of blue to sky-blue-emitting complexes reported that
do not contain F–Caryl bonds while complexes 8–10 combine
both fluorine and other electron-withdrawing groups on the C^N
ligands.
We have also targeted strongly electron-withdrawing sub-
stituents with large Hammett values as a strategy to blue-shift
emission. Complexes 12–15 are four examples that contain
trifluoromethyl-type substituents at the 4-position of the cyclo-
metalating phenyl ring. Our recent study explored whether
spacing the trifluoromethyl unit with oxygen (13), sulfur (14)
or sulfonyl (15) would lead to an increased blue-shift in the
emission compared to that observed for the reference complex
12. The emission maxima of the complexes correlated only
somewhat to their corresponding Hammett values of the four
functional groups, despite their HOMO energies being strongly
correlated. We observed that for complexes 12–14, the emission
maxima (lPL = 484, 516 nm for 12, 527 nm for 13 and 491,
Fig. 4 Emission spectrum of 3 and 4 in MeCN solution. Inset: Photograph
525 nm for 14, in MeCN) followed the trend of the Hammett
of MeCN solutions of 3 and 4 showing their green colour under UV values (sm = 0.43, 0.38 and 0.40 for 12, 13 and 14, respectively),
photoexcitation. with 12 the bluest of these complexes. However, to our surprise,
View Article Online

Fig. 5 Literature examples of different blue-to-sky-blue-emitting iridium complexes for use in LEECs (5–7) or OLEDs (8–11). Their relevant
photophysical properties in DCM (5, 8, 11) MeCN (6, 7), chloroform (9) or 2-MeTHF (10) solution are given for reference.
complex 15, which has the most electron-withdrawing substituent generally accepted substitution paradigms employed for blue-
(sm = 0.83 for the –SO2CF3 group), is in fact the most red-shifted of shifting the emission of these materials. Firstly, substituting
the four complexes (lPL = 515, 545 nm), illustrating the limits of pyridines for more electron-rich heterocycles, such as pyrazoles,
this approach (Fig. 6). on the C^N ligands generally blue-shifts emission due to the
Another strongly electron-withdrawing substituent that we increased p-accepting character of this ring, which leads to a
have studied is the pentafluorosulfanyl group (–SF5), which stabilisation of the HOMO.11d Indeed, we observed by cyclic
possesses a large associated Hammett value (sm = 0.61, voltammetry that this predicted electronic tuning was opera-
sp = 0.68). These substituents are underexplored as motifs in tional, with the pyridyl complexes showing lower positive oxida-
coordination compounds despite their extensive use in medicinal tion potentials (Eox = 1.57 V for 16 and 1.58 V for 17, vs. SCE) than
and materials chemistry applications.30 Complexes 16–19 represent their pyrazole analogues (Eox = 1.64 V for 18 and 1.67 V for 19, vs.
the first examples of iridium complexes bearing –SF5 substituents SCE). These observations point to the difficulty of correlating
(Fig. 7).31 electrochemistry – a ground state experiment – with the excited
All four complexes emit blue-green light that is modestly state properties characterised by emission.32 Secondly, we also
blue-shifted compared to 3 in MeCN solution (lPL = 485–503 nm). unexpectedly observed that the meta regiochemistry of the –SF5
We were surprised to find that of these four complexes, 16 shows unit to the Ir–C bond in 16 actually produced a greater blue-shift
the bluest emission profile. This observation goes against two than the para relationship in 17 when frequently the opposite
View Article Online

Fig. 8 Emission in solution of complexes 16–19 (in order from left to

right) in MeCN.
Fig. 6 Iridium complexes bearing a variety of trifluoromethyl-type

substituents on the C^N ligands.
requires electrochemical reversibility to be effective. By contrast,
these compounds exhibited green electroluminescence in solution-
processed OLEDs (EQE = 0.2–1.7%), since the role of charge
transport is instead handled primarily by the molecules of the host
matrix within the emissive layer.
Similarly to complexes 10,26d 1128a and others,26a–c,28b we
have also explored replacing the phenyl ring for a cyclometalating
heterocycle (complexes 20–23). The presence of the nitrogen
atom para to the Ir–C bond leads to a strong stabilisation of the
HOMO energy and is primarily responsible for the blue-shifted
emission in solution (lPL = 510–517 nm in MeCN for 20–23).33
Although not observed in solution for this family of complexes,
others have shown that alkoxy groups are useful substituents in
blue-shifting the emission.20,26c,28b While their para Hammett
values are negative (sp = 0.27), in a meta position these values
are positive (sm = 0.61), allowing for stabilisation of the HOMO
through inductive electron-withdrawing effects (Fig. 9).
In the neat film the emission of 20 is strongly red-shifted
(lPL = 547 nm) compared to 21–23 (lPL = 514–525 nm) and is
Fig. 7 Iridium complexes bearing pentafluorosulfanyl substituents on the also significantly less emissive (FPL = 2%) compared with the
C^N ligands. other three complexes (FPL = 19–32%) suggesting that in
addition to maintaining bluer emission, the added steric bulk
of the methoxy groups helps to reduce concentration quenching
trend is observed (cf. Hammett values for a measure of the in the neat film. It is worth noting that there is still a lack
relative electron-withdrawing power of the group) (Fig. 8).23a,27a of understanding regarding the changes in the photophysical
Although these complexes showed reasonably reversible properties when transitioning from solution to solid state, and
oxidation waves (attributable to the IrIII/IrIV redox couple), the this is a topic that requires addressing in future molecular design
reductions of all four complexes, in the region of 1.30 to of emitters for LEEC devices, where packing in the film becomes
1.40 V vs. SCE, are multi-electron and highly irreversible in important. Ultimately, this leads to poor LEEC device perfor-
nature, which was attributed to direct electrochemical decom- mance for 20, but better LEEC efficiencies for complexes 21–23
position of the –SF5 unit. This is compared with that of 3 (EQE = 2.0–2.8%), as well as emission very close to the ideal green
(Ered = 1.36 V vs. SCE), which shows good reversibility charac- CIE coordinate (CIE: 0.30, 0.60) for complex 21 (CIE: 0.31, 0.57)
teristic of reduction of the bpy ligand to its radical anion.16 (Fig. 10).
The absence of electroluminescence in the LEEC devices was Despite the many examples reporting the use of alkoxy
attributed to the electrochemical instability of these complexes, groups for blue-shifting emission, the ‘intuitive’ strategy of
since the emitters in LEECs are required to perform the dual using these groups to red-shift emission has in fact not been
role of light emission and charge transport, the latter of which widely explored. An early example of the use of methoxy groups
View Article Online

Fig. 9 Iridium complexes based on cyclometalated 2,3 0 -bipyridines.
to mesomerically red-shift emission was reported by Watts et al.,34

who showed that the emission of the tris-cyclometalated complex
fac-[Ir(ppy)3] (24) could be red-shifted if –OMe groups were
appended para to the Ir–C bond (25) or blue-shifted if they were
in a meta relationship (26) to the iridium centre (lPL = 494 nm
for 24, 539 nm for 25 and 489 nm for 26 MeOH/EtOH glass at
77 K). More recently, Davies and co-workers35 reported a family
of iridium complexes bearing 1-phenyl-1H-pyrazole (ppz) and
various ppz derivatives as C^N ligands. They observed that Fig. 10 Neutral (24) and charged (27) reference complexes to MeO-
compared to the reference complex 27 (27, lPL = 557 nm in substituted neutral (25 and 26) and charged (28) iridium complexes.
DCM) the complex bearing –OMe groups, 28, displayed a strongly
red-shifted emission (28, lPL = 615 nm in DCM).
Our study explored decoration of the C^N ligands with an analogous fashion to the photophysical properties of the
increasing number of –OMe groups in a systematic fashion to emitters 20–23 (Fig. 11).
red-shift the emission and explore the competing s-withdrawing The electrochemistry follows the trend found for the emission
effects with the p-donating effects of these substituents (29a–32b).36 properties. For all of these complexes, the reduction is largely
The emission profiles of 29a and 29b are the most blue-shifted unchanged compared to 1 and 2, pointing towards emission
in the study (lPL = 618 and 595 nm for 29a and 29b in MeCN, tuning being a HOMO-dominated effect. Accordingly, there
respectively) but nevertheless are red-shifted compared with 1 is stark contrast in the oxidation potentials, which vary signifi-
and 2 (lPL = 602 and 585 nm for 1 and 2 in MeCN, respectively). cantly across the series. The most stabilised values were
For complexes 30a and 30b, the importance of the substitution observed for 29a and 29b (Eox = 1.15 and 1.13 V vs. SCE, for
regiochemistry becomes clear, with the emission of these com- 29a and 29b, respectively), whereas the most red-shifted
pounds greatly red-shifted (lPL = 710 and 680 nm for 30a and 30b emitters, 31a and 31b also showed the lowest oxidation
in MeCN, respectively) compared to 29a and 29b. For these two potentials (Eox = 0.84 and 0.78 V vs. SCE, for 31a and 31b,
complexes, the mesomeric effects of the –OMe groups govern the respectively). Apart from 29a and 29b (FPL = 6 and 15% for 29a
emission energy. The situation with complexes 31a–32b is some- and 29b in MeCN, respectively), all of these complexes were
what less clear. Complexes 31a and 31b are the most red-shifted poorly emissive (FPL o 1%) presumably as a result of the
among all eight of this family (lPL = 730 nm for 31a and 700 nm energy gap law. Thus, in the LEECs their performances were
for 31b), but this red-shift is only relatively small compared to limited, with 31b showing the best device performance (EQE =
the large red-shift in emission achieved by the substitution at the 0.05%, t1/2 = 2 h, CIE: 0.61, 0.38). These numbers are low
5-position of the phenyl rings in 30a and 30b. Furthermore, compared to champion orange-red LEECs in terms of efficiencies
adding a third –OMe substituent induces a small blue-shift (EQE = 3.3–9.5%)11b,d,37 and device lifetimes (t1/2 = 10–37 h)37,38
(lPL = 720 nm for 32a and 685 nm for 32b), in a somewhat reported for this colour.
View Article Online
regiochemistry of a wide variety of different C^N ligands

(Fig. 12).44
We have shown theoretically that phenyl-1,2,3-triazole (phtl)
ligands, such as those used in complexes 35–37, are important
nitrogen-rich C^N ligands that can blue-shift emission.45
We studied the complexes in Fig. 13, using 1-benzyl-4-phenyl-
1H-1,2,3-triazole (phtl, complexes 40a and 40b) and 1-benzyl-4-
(2,4-difluorophenyl)-1H-1,2,3-triazole (dFphtl, complexes 41a
and 41b).
In contrast to ppz, which stabilises the HOMO, we observed

that these phtl ligands intrinsically tune the LUMO, despite the
LUMO showing electron density largely localised on the ancillary
N^N ligand. Consider the electrochemical data shown in Table 1.
Comparing the HOMO energies of 2 with 40b (EHOMO = 5.71 eV
for 2 and 5.66 eV for 40b) or 3 with 41b (EHOMO = 6.02 eV for 3
and 6.00 eV for 41b) shows that the HOMO is rather insensitive
to modification of the coordinating heterocycle of the C^N
ligands. By contrast, the LUMO changes significantly, with the
LUMOs of 40b (ELUMO = 2.88 eV) and 41b (ELUMO = 2.95 eV)
strongly destabilised compared with their pyridyl analogues
(ELUMO = 3.04 and 3.06 eV for 2 and 3, respectively). Although
Fig. 11 Cationic iridium complexes bearing an increasing number of
the effect on the HOMO energies is non-zero, demonstrating that
methoxy substituents on the C^N ligands.
with all these substitutions one is never exclusively modulating
the energy of only one set of orbitals without affecting the other
To conclude this section, the features to note concern the set, this design strategy clearly produces a LUMO dominant
crucial role that regiochemistry of the substituents on the effect. DFT computations reveal a non-zero orbital contribution
C^N ligands plays in influencing the photophysics of these from the metal to the LUMO, which accounts for the strong
complexes, and how this can (in a large number of cases) be sensitivity of the LUMO to the nature of the heterocycle on the
correlated with the corresponding Hammett values of the C^N ligand.
substituent. Furthermore, when there are multiple substituents The photophysical data also demonstrate the blue-shifting
on the C^N ligands the emission can be tuned further, but the nature of the triazoles. The emission maxima of 40a and 40b
effects on the photophysics are not necessarily linearly additive (lPL = 580 and 575 nm, respectively in MeCN) are blue-shifted
with each substituent added, making their influence sometimes compared to 1 and 2 (lPL = 602 and 591 nm, respectively
difficult to discern. in MeCN). The addition of fluorine atoms to the C^N ligands
blue-shifts the emission further, with 41b showing sky-blue
emission, rather than the green emission of 3 (see Fig. 14;
4 C^N ligand – effects of substitution/ lPL = 515 nm for 3 and 498 nm for 41b).
modification of the pyridyl ring An important consideration then is how employing a different
heterocycle to pyridine in the C^N ligands tunes the emission.
Rather than decorating the pyridyl ring of the C^N ligands with We have striven to show here that this is not immediately obvious,
substituents, modification of this ring is typically achieved by with pyrazoles exerting HOMO-tuning effects and triazoles modu-
utilising an entirely different heterocycle. To the best of our lating the LUMO. These are important considerations that should
knowledge, the most common ligand framework used to this inform future molecular design.
end is 1-phenyl-1H-pyrazole (ppz). By electrochemistry, it has
been demonstrated that the second nitrogen atom in the five-
membered heterocycle blue-shifts emission through stabilisa- 5 N^N ligand – effect of substitution/
tion of the HOMO of these complexes.11d,23a,c,39 Other C^N modification of the bpy
ligands coordinated to iridium(III) containing nitrogen-rich
heterocycles such as imidazoles;40 1,2,3-triazoles;41 1,2,4- In this section, we explore different ways of modifying the N^N
triazoles;42 and 1,2,3,4-tetrazoles43 have also been reported, ligand to shift the emission either to the blue or to the red.
and an important feature to note from these studies is that These strategies are based on the fact that the LUMO is typically
they do not always modulate the HOMO energy. In particular localised on this ligand. For shifting the emission towards the
we will show that 1,2,3-triazoles strongly influence the LUMO blue, there are two main strategies designed to raise the energy
energy instead. For reference, we point the reader to Ortı́ and of the LUMO. The first strategy, which will be discussed in
co-workers’ recent extensive theoretical study into the effect on Section 6, is to decorate bipyridines with strongly electron-
the photophysical properties of varying the nitrogen count and donating groups such as amines. The second, and demonstrably
View Article Online

Fig. 12 Iridium complexes bearing a variety of different azoles within the C^N ligand framework. Their relevant photophysical properties in DCM
(33–35) MeCN (36, 37, 39) and toluene (38) solution are given for reference.
more fruitful strategy, involves replacing the coordinating pyridine tuning effect is not specific to 1,2,3-triazoles and has also been
rings with more strongly s-electron donating heterocycles, as has shown for other heterocycles as well. Indeed, comparing, for
been reported with the complexes 42–46. This second strategy will example, otherwise analogous complexes wherein pyrazole
be the main focus of this section. Red emission, by contrast, can be rings have been included into the ligand framework either of
accomplished in one of two main ways: (1) appending electron- the N^N ligand (42) or the C^N ligands (33), the complex
withdrawing groups onto the N^N ligand to stabilise the LUMO; or bearing the modified N^N ligand generates an invariably bluer
(2) increasing the conjugation of the N^N ligands to narrow the emission (lPL = 452, 480 nm for 42 and 492 nm for 33).7b,9,39c
HOMO–LUMO gap, examples of both strategies will be highlighted Thus we designed complex 48, which employs a bis(triazole),
here (Fig. 15). btl, ancillary ligand. As with 43, inclusion of triazole motifs on
Given the efficacy of triazoles in blue-shifting emission for the N^N ligand, rather than the C^N ligands produced an even
complexes 40a–41b, we wanted to explore their effects when more pronounced destabilization of the LUMO.49 Complex 48
incorporated within the ancillary ligand. De Cola and co-workers48 shows structured LC-type emission (lPL = 481, 511 nm) while
showed that using these heterocycles as part of the ancillary ligand the emission spectra of 40a and 40b are broad, unstructured
framework is even more effective in blue-shifting the emission than and substantially red-shifted in colour (lPL = 580 and 575 nm,
when they are incorporated within the C^N ligands. This enhanced respectively) and predominantly CT in nature. Achieving such a
View Article Online

Fig. 13 Iridium complexes bearing 1,2,3-triazoles within the cyclometalating

C^N ligand framework.
Table 1 Electrochemical data comparing complexes bearing pyridyl

(2 and 3) and 1,2,3-triazoles (40b and 41b) within the C^N ligand
frameworka,b
E1/2 (V vs. SCE) Eorbital (eV)

Complex C^N ligand Eox Ered HOMO LUMO Ref.
2 ppy 1.29 1.38 5.71 3.04 13a
3 dFppy 1.60 1.36 6.02 3.06 16
40b phtl 1.24 1.54 5.66 2.88 45
41b dFphtl 1.58 1.47 6.00 2.95 45
a
Measurements were carried out in de-aerated MeCN and are refer-
enced vs. SCE (Fc/Fc+ = 0.38 in MeCN).46 b EHOMO/LUMO = [Eox/red vs.
Fc/Fc+ + 4.8] eV.47
Fig. 15 Blue-emitting iridium complexes with modified ancillary ligands.

Photophysics reported in MeCN (42, 45–47) or DCM (43, 44).
cationic iridium complexes. The potency of both the parent

1H,1H 0 -2,2 0 -biimidazole (H2biim)50 and the dimethylated ana-
log 1,1 0 -dimethyl-2,2 0 -biimidazole (Me2biim)11a had previously
been demonstrated with complexes 44 and 45 emitting in the
Fig. 14 Relevant photophysical data for 41b in MeCN. Inset: Emission in blue-green to blue regime in MeCN solution, respectively.
solution of 3 (left) and 41b (right) to demonstrate blue-shift achieved by However, both 44 (FPL = 10% in DCM) and 45 (FPL = 5% in
substituting the pyridyl ring on the C^N ligand for a 1,2,3-triazole ring. MeCN) were poorly emissive. In addition, the high orbital
energies associated with these biimidazole-type N^N ligands
renders them essentially ‘non-chromophoric’ in nature, resulting
large blue-shift in the emission without concomitant modifica- in both complexes displaying predominantly LC-type emission
tion of the C^N ligands is normally difficult, and important localised on the C^N ligands rather than mixed MLCT/LLCT
when pursuing fluorine-free blue emitters. Attempts to further emission that typically characterises complexes such as found
blue-shift the emission through judicious decoration of the for 3. This interplay between MLCT/LLCT and LC excited states is
C^N ligands however produced photochemically unstable com- a common feature of blue-emitting complexes with modified
plexes (see Section 7). ancillary ligands.
In a similar fashion to the btl ligand, biimidazole (biim) We hypothesized that the reason for the low FPL was due to
ligands are also effective scaffolds for blue-shifting emission of the presence of an undesired twisting of the biimidazole in the
View Article Online

Fig. 16 Iridium complex bearing a bis-1,2,3-triazole ancillary ligand.
excited triplet state, which would subsequently twist back into

planarity upon relaxation to the ground state via a non-radiative
pathway.51 DFT modelling supported this proposal whereby in
particular the steric bulk of the methyl groups in 50 enforced a
twisting out of planarity of the ancillary ligand (Fig. 16). Indeed,
both 49 and 50 show blue, ligand-centred emission (lPL = 464,
490 nm for 49 and 457, 486 nm for 50, in MeOH) but with
expectedly low photoluminescence quantum yields (FPL = 20%
for 49 and 2% for 50) and short emission lifetimes (te = 1.6 ms
for 49 and just 91 ns for 50), particularly considering the LC
nature of the emission. In order to mitigate this non-radiative
pathway, we proposed that a strategy to tether the distal
nitrogen atoms of the ancillary ligands could rigidify the ligand
framework and thus restrict the distortions observed for 50 in
Fig. 18 (a) Calculated geometries of 50 and 51 in the T1 state, with
the excited state. Our lead candidate was 51, which, aside from
zoomed images showing the distortion of the biimidazole invoked by
the rigidity imparted by the o-xylylene linker, was also predicted to the steric bulk of the clashing methyl groups (50) and the lack of such
maintain the same optoelectronic properties, since the methylene distortion when tethered (51). DFT [(B3LYP/SBKJC-VDZ for Ir(III)) and
groups impede further conjugation between the tethered phenyl (6-31G* for C,H,N)] with CPCM (MeCN). (b) Emission spectra of 49–51 in
group and the remainder of the ligand framework (Fig. 17). deaerated MeOH, with the emission colour in solution inset.
The emission of 51 in MeOH showed essentially the same

profile and energy as that of 50. More importantly, and gratify-
ingly, the FPL of 51 was greatly enhanced (FPL = 68%) compared Although this strategy proved effective, the poor solubility of
to both 49 and 50, demonstrating the viability of this strategy 51 in most organic solvents precluded its viability as a solution-
and corroborating our hypothesis. The emission lifetime of processable emitter for LEECs. We thus targeted a mesitylated
51 (te = 3.8 ms) was also significantly longer than that observed analogue based on a design similar to 4 (Fig. 19).16 Complex 52
for either 49 or 50. By comparison, the te at 77 K for all three is both more soluble in polar organic solvents, and also more
complexes are all similar (te = 3.7–4.0 ms for 49–51) suggesting emissive in solution (FPL = 82% in MeOH and 90% in MeCN)
that the rigidified local environment at low temperature than 51. This FPL value for a cationic blue emitter (lPL = 459 nm)
impedes any distortions of the biimidazole from occurring in solution is among the highest ever reported.9,11a,52 The archi-
and thus further substantiating our assertions that quenching tecture of 52 demonstrates the importance of rational ligand
of 49 and 50 is attributable to vibrational processes (Fig. 18). design that combines elements to effectively tune the electronics
Fig. 17 Iridium complexes bearing various biimidazole-type ligands.
View Article Online
thus reduces the conjugation length of the ligand.54 In particular,

complexes 53 and 54 were targeted to produce blue emission. Two
important optoelectronic effects result from the incorporation of
gpy onto the complex. Firstly, due to the saturation present in the
guanidyl ring, the LUMO becomes localised largely on the C^N
ligands (as indicated by DFT and CV), rather than on the N^N
ligand as is observed for complexes such as 2 and 3. Secondly, the
strongly electron-releasing nature of the coordinating nitrogen
atom of guanidyl ring acts to destabilise both the HOMO and
the LUMO, but the greater destabilisation of the LUMO and also
the triplet state results in a net blue-shifted emission. A comparison
of the emission maxima in MeCN solution of 53 (lPL = 503 nm) with
2 (lPL = 585 nm) and 54 (lPL = 470, 498 nm) with 3 (lPL = 515 nm)
brings the electronic effect of the gpy ligand into stark focus. By
contrast to 53 and 54, 55 displays strongly red-shifted and broad,
mixed CT-type emission. DFT calculations point to a LUMO that is
largely localized on the electron-poor pyrazine ring, resulting in both
Fig. 19 Solubilised iridium complex bearing a substituted biimidazole ligand. a much lower emission energy (lPL = 640 nm) and a very low FPL of
0.2% (Fig. 20).
Replacement of the coordinating nitrogen atoms for those
of the complex with control over intramolecular quenching that are at once more s-donating and less p-accepting accom-
processes (the tethered biimidazole) and intermolecular quench- plishes the same effect. In particular, complexes bearing P^P
ing processes (the bulky mesityl substituent). This complex was ligands, such as 47, have been explored in this vein.55 Complex
studied in solution-processed OLEDs, giving sky blue emission 47 is poorly emissive in MeCN solution (FPL = 0.4%), but
(CIE: 0.21, 0.37) and reasonable efficiency (EQE = 3.42%) for a nevertheless is a deep blue emitter (lPL = 459 nm). We observed
cationic complex employed in an OLED.16 similar emission profiles when we explored a large family of
A complementary strategy for ancillary ligand design in the complexes of the form [Ir(ppy)2(P^P)]PF6 where the bite angle of
context of blue emission is to reduce the p-accepting strength of the P^P ligand was varied systematically (Fig. 21). In all cases,
the ligand while maintaining the strong s-donating character deep blue emission (lPL = 444–485 nm for the E0–0 band
of the coordinating nitrogen atoms. One method to implement in MeCN solution) was observed for the six ppy-containing
this strategy is to partially saturate the ligand scaffold while complexes studied, regardless of the nature of the P^P ligand.
nevertheless maintaining sp2-nitrogen coordinating atoms. However, each of these complexes also showed low FPL values
One way in which this was done previously was by utilising (FPL = 0.3–4.2%).
bis-N-heterocyclic carbene (NHC) ligands, which are bridged By contrast, the nature of the C^N ligands impacted more
by methylene groups such as complex 46.52a These ligands greatly both the emission energy and, importantly, the FPL of
combine the exceptional s-donating capabilities of NHCs with the complexes. When dFppy was employed as the C^N ligand
a lack of conjugation across the ligand scaffold, allowing these (56), the expected blue-shift in the emission (lPL = 445 nm in
complexes to achieve very deep blue emission (lPL = 440 nm in MeCN for the E0–0 band for 56 compared to 457–485 nm for the
MeCN for 46).53 As with the 49–52, the ancillary bis-NHC ligand ppy complexes) was observed, and this coincided with a signi-
of 46 is non-chromophoric, with this complex showing highly ficant enhancement of the photoluminescence quantum yield
structured LC emission. (FPL = 18%). Although this result is in line with the energy gap
For our strategy we used guanidylpyridine, gpy, (53 and 54) law, it is still nevertheless contradictory to what we would have
or guanidylpyrazine, gpz, (55) ancillary ligands, wherein one of expected, given that MC states should be more thermally
the coordinating heterocyclic rings is partially saturated and accessible for 56 than for the other complexes (Fig. 22).
Fig. 20 Iridium complexes bearing guanidylpyridine (53 and 54) and guanidylpyrazine ancillary ligands (55).
View Article Online

Fig. 21 General structures of iridium complexes explored with various P^P ancillary ligands.
reported previously.55b We also targeted 57 for solution-processed

OLEDs and showed that cationic iridium complexes could produce
blue emission with CIE coordinates of (0.20, 0.29). However, despite
the use of a mixed host-dopant configuration (host: mCP and
OXD7) in the emissive layer to improve charge balance, the
efficiency was still low (EQE = 0.21%). Nevertheless, this example
demonstrates the importance of optimising not only each ligand
separately but also the combination of the ligands in order to
achieve the desired properties of the complex. This point will be
discussed further in Section 6.
Complex 55 illustrated the efficacy of incorporating within
the ancillary ligand electron-withdrawing groups to red-shift
the emission. Our study of complexes 58–61 combined this
feature with increased conjugation across the ancillary ligand
scaffold. The complexes possess an additional pyridine ring at
the 5 0 -position of the N^N ligand. Two families were studied,
Fig. 22 Iridium complexes bearing P^P type ancillary ligands.
with complexes 58 and 59 containing a 2,2 0 :5’,200 -terpyridine
(2,5-tpy) ancillary ligand and 60 and 61 containing a 2,5-
di(pyridin-2-yl)pyrazine (2,5-dpp) ancillary ligand. Comparison
Accounting for the deep blue emission of all of these P^P of the photophysics of 58 and 59 demonstrates the expected
complexes but their poor photoluminescence quantum yields blue-shift of the emission as a function of the presence of the
led us to perform a substitution study to explore if we could fluorine atoms on the C^N ligands, with the emission of 59
maintain the deep blue colour but enhance the FPL to appreci- (lPL = 544 nm in MeCN) being blue-shifted compared to that of
able levels. When studying the effects of the addition of a 58 (lPL = 605 nm in MeCN). Complex 59 is a very efficient
mesityl ring at the 4-position of the pyridine of the C^N ligands emitter in solution (FPL = 93%), particularly compared with
(2-phenyl-4-(2,4,6-trimethylphenyl)pyridine, Mesppy) we observed 58 (FPL = 6%). A comparison of the emission maxima of 58 and
only an incremental enhancement of the FPL. Similarly, when 1 (lPL = 602 nm in MeCN) suggests that the additional pyridine
dFMesppy was used as the C^N ligand with wide bite-angle P^P ring has a negligible effect on the electronics. However, the
ligands poorly emissive complexes were still observed (FPL o 1.0%). analogous comparison between 59 and 3 (lPL = 515 nm for 3 in
However, combining dFMesppy with the smallest bite angle MeCN) reveals a red-shift of the emission in the former
dppe ligand in 57 gratifyingly afforded both a bright and deep compared to the latter.11c
blue emission (lPL = 444 nm, FPL = 52% in MeCN). When used By contrast, complexes 60 and 61, containing the pyrazine
in LEECs in a host–guest configuration, we found that 57 gave ring, are significantly red-shifted in their emission compared to
blue electroluminescence (lEL = 479 nm) but with low 58 and 59. Despite containing dFMeppy C^N ligands, 61 shows
luminance (Lmax = ca. 7.5 cd m2) and poor stability (t1/2 = 0.5 h). essentially the same emission profile as 58 (lPL = 604 nm in
Such instability of bis-phosphine complexes in LEECs has been MeCN for 61), demonstrating that the nitrogen in the pyrazine
View Article Online

Fig. 24 Di-nuclear iridium complex bridged by 2,5-di(pyridin-2-yl)pyrazine.
Fig. 23 Iridium complexes bearing 2,2 0 :5’,200 -terpyridine (58 and 59) or
2,5-di(pyridin-2-yl)pyrazine (60 and 61) as the ancillary ligand.
Fig. 25 Mono-nuclear iridium complexes bearing alkynyl substituents.
ring is more influential to the photophysics than the extended

conjugation, and that the inclusion of the pyrazine ring towards typical 3LLCT/3MLCT emission. This emission is red-
counterbalances the HOMO-stabilizing effect of the fluorinated shifted compared to 1 and 3, while also showing lower photo-
cyclometalating ligands. The emission in 60 is by far the most luminescence quantum yields (FPL = 9% and 6% for 63 and 64,
red-shifted of this family of complexes (lPL = 666 nm in MeCN), respectively). The low FPL values, particularly for 64, were
although it is also the least emissive (FPL = 2.6%). This deep red attributed to vibrations of the alkynyl group that provided a
emission translates to one of only four LEEC devices with CIE non-radiative decay outlet (Fig. 25).
coordinates (CIE: 0.68, 0.33) that coincide with the ideal red The dinuclear complexes were obtained by homocoupling 63
coordinate. However, the lifetime of this device is short or 64 under Glaser conditions using CuCl. Comparing the
(t1/2 = 6.3 h), and the efficiency is low (EQE = 0.08%),11c pointing corresponding dinuclear complexes 65 and 66, we observed a
towards the need for improved molecular and device design surprising evolution in the photophysical properties. For
(Fig. 23). complex 65, the emission in 2-MeTHF showed the expected
Another important aspect of these ancillary ligands is that red-shift (lPL = 647 nm) in emission compared to the mono-
they can function as bridging ligands. Previous work had nuclear parent with a corresponding decreased FPL = 1%,
shown that using 1,4-di(pyridine-2-yl)benzene as an ancillary pointing towards increased conjugation across the bipyridine
ligand allowed the formation of neutral dinuclear iridium ligand scaffold, and thus electronic communication between
complexes.56 In our efforts, we used the 2,5-dpp ligand as a the two metal centres. However, to our surprise the dFMeppy
bridging motif. It has been shown that adding multiple metals analogues showed a different trend, with 66 somewhat blue-
around a conjugated core can lead to a strong red-shift in the shifted (lPL = 558 nm) compared to its corresponding monomer,
emission.57 In this case, no emission could be detected at room 64. At 77 K, however, the expected pattern was indeed restored,
temperature from the resultant dinuclear complex 62, and thus with the monomeric complex 64 (lPL = 490, 527 nm at 77 K for 64)
it was unclear if the emission had been too strongly red-shifted undergoing a much more pronounced hypsochromic shift than
into the NIR, or if it was totally quenched. However, at 77 K in a its dimeric counterpart, 66 (lPL = 560, 605 nm at 77 K for 66)
2-MeTHF glass emission from 62 (lPL = 715 nm) could be (Fig. 26).
detected, and indeed this was greatly red-shifted compared to As we have seen, the mesityl substitution for complexes such
the emission of 60 under identical conditions (lPL = 589 nm at as 3 and 52 can have a profound impact on the FPL of the
77 K in 2-MeTHF glass) (Fig. 24). complex without impacting the emission energy. We13a inves-
A different strategy that we have employed for constructing tigated the effects on the optoelectronic properties of a series of
bi- or multimetallic systems employs the 5-ethynyl-2,2 0 -bipyridine complexes bearing diaryl-substituted bipyridine ligands (Fig. 27
ligand, as with complexes 63 and 64.58 In 2-MeTHF solution the and 28). Complexes 67–69 show modestly red-shifted emission
emission of both complexes 63 and 64 (lPL = 623 and 561 nm for compared to 1, with 68 (lPL = 613 nm in MeCN) the bluest
63 and 64, respectively) is broad and unstructured, pointing of these followed by 67 (lPL = 623 nm in MeCN) and 69
View Article Online

Fig. 26 Di-nuclear iridium complexes bearing alkynyl bridging units.
Fig. 27 Iridium complexes bearing diaryl substituted bipyridine ancillary ligands.
(lPL = 659 nm in MeCN). While 67 and 68 emit from a mixed

CT state and are reasonably emissive (FPL = 17% for 67 and
FPL = 21% for 68 compared to 6% for 1), the strongly electron-
donating character of the pendant dimethylamine groups in 69
forces a change in the excited state to that of an intraligand
charge transfer (ILCT) state, with a correspondingly low FPL of
0.7%. These results suggest that the strategy of incorporating
steric bulk at the 5 and 5 0 positions of the bipyridine helps to
enhance FPL without modulating significantly the emission
energy, assuming mixed CT emission is maintained, in an
analogous fashion to the mesityl-substituted complexes dis-
cussed above. In addition, 67 is notable for its performance in a
LEEC, which demonstrated greatly enhanced device stability
under constant current driving compared to a control based on
2 (t1/2 = 1.3 h for 2 and 110 h 67).59
We explored the effects of decoupling mesomeric con-
jugation completely through the addition of ortho,ortho-
dimethyl groups on the aryl substituents in 70–73. Such a
substitution resulted in a series of complexes with slightly
blue-shifted emission maxima compared to 67–69 and 1
(lPL = 592–613 nm in MeCN) coupled with an enhancement
in the FPL. For example, complexes 70 (FPL = 22%) and 71
Fig. 28 Iridium complexes bearing bulky diaryl substituted bipyridine (FPL = 21%) are brighter than 67, while 73 (FPL = 2%) is also
ancillary ligands. enhanced compared 69.
View Article Online
Thus, we have seen examples of both deep blue and deep red deep blue, highly emissive cationic iridium complexes functioning
emitters in solution. We have further illustrated examples of with good efficiency and stability in the device.
how careful molecular design can lead to efficient (blue) The first example of our efforts in this endeavour is based on
emitters in solution and demonstrated that incorporation of complex 4, originally reported by Nazeeruddin and co-workers.60
bulky groups inhibit intermolecular quenching processes that This complex incorporates strongly electron-donating dimethyl-
can reduce the FPL. However, achieving a particular colour is amino groups at the 4,4 0 -positions of the bipyridine ligands
now not enough. For blue emission, an important challenge designed to destabilise the LUMO and shift the emission to the
that remains to be addressed is controlling the interplay blue. The blue-shifting capacity of this ancillary ligand is
between LC and CT states governing the emission. For device evidenced by comparing the emission of reference complexes
applications, CT states are typically preferred as the nature of 74 and 75 with that of 3. Complexes 74 and 75 both show
3
these excited states mean that they generally display shorter LC-type emission characterised by high energy shoulders in the
lifetimes (reduced triplet quenching) and narrower emission region of 460 nm as well as principal emission bands in the
profiles (improved colour purity).6c However, as we have seen, 490 nm regime (lPL = 464, 490 nm for 74 and 466, and 494 nm
maintaining a mixed CT triplet state becomes increasingly for 75 in MeCN) compared to the red-shifted 3CT emission of
difficult as N^N ligands with higher orbital energies are 3 (lPL = 515 nm for 3 in MeCN) (Fig. 29).
employed. This behaviour in part accounts for the lack of a In an effort to blue-shift this emission further, we incorpo-
deep blue LEEC device reported to date. For red emission, rated the dFphtl ligand first reported coincidentally by De Cola
although there are examples of ‘true red’ LEEC devices, these and co-workers for 36 and 3741b and by us for 41a and 41b61 in
devices are invariably unstable, and poorly efficient. This is at order to take advantage of its blue-shifting properties relative
least in part a consequence of the energy gap law, which to dFppy. However, to our surprise 76 exhibits sky-blue,
negatively impacts the FPL of the emitter. unstructured emission (lPL = 495 nm in MeCN) that is essen-
tially the same colour as 74 and 75. To rationalize this result,
we carried out DFT calculations to understand the effect of
6 C^N and N^N ligand – combining changing the ancillary ligand when the same dFphtl ligand was
substitution on both ligands used (Fig. 30). Comparing 76 with 41a and 41b, we observed
that the calculated energies of the HOMO orbitals were largely
We have seen that modification of individual components of similar, with that of 76 slightly destabilised in comparison to
either the C^N or N^N ligands can modulate quite effectively the other two, as expected for substitution changes on the
the optoelectronic properties of the emitters. However, in order ancillary ligand. The LUMO energies increase with increased
to optimize these properties it is frequently necessary to install electron density on the N^N ligands, leading to an overall
both bespoke C^N and N^N ligands. Indeed, many of the increase in the band gap (DEHOMO–LUMO: 76 4 41b 4 41a).
examples presented from other groups such as 10, 11 or 46 Thus, on the basis of the band gaps calculated for the three
explore deviations both from the traditional C^N-type ligands complexes we would have expected the emission of 76 to be
(ppy and dFppy) and the conventional N^N ligand (bpy). We much bluer than both 41a and 41b, but this is not observed.
have also explored these strategies, primarily within the context Instead, DFT calculations reveal that the energy of the T1 states
of designing deep blue emitters, and frequently found that for the three complexes remain essentially isoenergetic (modelled
these structure–property relationship studies are not straight- as the HSOMO energy in Fig. 30), indicating that the triplet energy
forward to interpret. As we observed with the red-emitting is largely unaffected by the substitution of the –NMe2 on the N^N
complexes 29a–32b, there is a limit to how far we can tune ligand.
the photophysics (in either the red or blue direction) of these As the strategy of adding the most electron-donating sub-
cationic complexes by simple ligand substitution. This has so stituent did not have the desired blue-shifting effect, we turned
far precluded others and us from realising the goal of developing our attention to modifying the coordinating heterocycles of the
Fig. 29 Iridium complexes bearing 4,4 0 -dimethylamino-2,2 0 -bipyridine as the ancillary ligand.
View Article Online
a photolysed sample showed that the primary degradation

product was [Ir(dFphtl)2(NCMe)2]+. These and 1H NMR analyses
pointed towards a photo-ejection of the btl ligand from the
complex. Elliott and co-workers62 independently corroborated
this decomposition mechanism.
Despite the room temperature photoinstability of 78, we
were able to measure the emission of this complex at 77 K.
Comparing the low temperature emission maxima of 78 with
77 and 48, 78 is by far the bluest under these conditions
(lPL = 470 nm for 78, 407 nm for 77 and 393 nm for 78 in

2-MeTHF at 77 K), thus demonstrating the merits (from a
colour tuning point of view) of ‘optimising’ the electronics of
these complexes by addressing each individual ring (Fig. 32).
To establish a compromise between the blue-shifting quali-
ties of the btl ligand with the associated photoinstability of its
complexes, we synthesised complexes 79–82, which contain
pyridyltriazole ancillary ligands, similar to complex 43 and its
analogues.48 We explored two different pyridyltriazole ligands:
one where the distal nitrogen was N-phenylated (79 and 81) and
Fig. 30 DFT calculated energies of the HOMO4 to LUMO+4 for com- one where the distal nitrogen was N-benzylated (80 and 82).
plexes 41a, 41b and 76, as well as the highest singly occupied molecular Each of these complexes employ either Fphtl or dFphtl as the
orbital (HSOMO) in the T1 state. DFT [(B3LYP/SBKJC-VDZ for Ir(III)) and C^N ligands, with only the number of fluorine atoms appended
(6-31G* for C,H,N)] with CPCM (MeCN).
to the phenyl ring being varied. As would be expected, the
complexes where the phenyl ring has been substituted with
only one fluorine atom (79 and 80) are not as blue in MeCN
solution (lPL = 487 nm and 485 nm for 79 and 80, respectively)
than when there are two fluorine atoms present as in 81 and 82
(lPL = 461 nm and 452 nm for 81 and 82, respectively).
Fig. 31 Iridium complexes bearing 1,2,3-triazoles within the C^N and

N^N ligand frameworks.
ancillary ligand. Complexes 77 and 78 employ the same bis-

triazole ancillary ligand as in 48 (Fig. 31).49
Comparing 48 and 77, the emission of 77 is moderately
bluer, showing 3CT-type emission in the sky-blue region
(lPL = 495 nm in MeCN), compared to the more structured
3
LC-type emission of 48 (lPL = 511 nm in MeCN with a high
energy shoulder at 481 nm). This change in the nature of the
emission is due to the capacity of the phtl ligand to increase
the energy difference between the N^N-centred LUMO and
C^N-centred LUMO+1 orbitals, thereby contributing to a more
diffuse spin density distribution in the T1 state and rendering
the emission more CT-like. When fluorine atoms are incorpo-
rated on the C^N ligands in 78 we noticed that the complex
becomes unusually photo-unstable, with the presence of two
emission bands at room temperature at 352 nm and 505 nm, Fig. 32 Iridium complexes bearing pyridyltriazole ancillary ligands and
the latter of which diminished over time. LC-HRMS analysis of Fphtl and dFphtl C^N ligands.
View Article Online
In addition, complexes 79 and 81, where the phenyl rings are based on 80 is comparable to some of the best and bluest LEECs
conjugated into the N^N ligand framework, are slightly reported to date. In terms of CIE values, only a small number of
red-shifted in emission compared to their N-benzyl substituted iridium complexes have been reported to date that show bluer CIE
counterparts. Finally, complex 43 is in fact blue-shifted coordinates in a LEEC. Complex 42 has the bluest CIE value
(lPL = 452, 483 nm) compared with complexes 79–82, albeit reported to date (CIE: 0.20, 0.28),9 while strongly blue-shifted
with the caveat that the photophysics of 42 were investigated in emission is also observed from complex 33 (CIE: 0.20, 0.36),39c
DCM rather than MeCN for complexes 79–82. 83 (CIE: 0.21, 0.33)63 and 84 (CIE: 0.20, 0.34).53 Finally,
Although the FPL of all the complexes were low (FPL = 0.03–0.3% although the CIE values for the LEEC with 43 and its analogues
in MeCN), particularly compared to 43 (FPL = 22%), the strongly were not reported, the significantly bluer electroluminescence
blue-shifted emission of 81 and 82 in particular led us to maxima reported for these complexes (lEL = 456–460 nm)48
explore the application of all four of these complexes as compared to 79–82 suggests that the CIE coordinates of these
emitters in LEECs. Studying their properties in the solid state, complexes are likely bluer in the device as well, despite these
we found that the neat film photophysics were similar to those complexes bearing dFppy as the cyclometalating ligand, rather
in the solution state, with low FPL values (FPL = 1–2% for than dFphtl (Fig. 33).
complexes 79–82) but with strongly blue-shifted emission Given the very small change in photoluminescence properties
colour (lPL = 451–473 nm for complexes 79–82); powder FPL in moving from solution to the solid state for all four of these
for 80 and 82 were however enhanced at 12 and 10%, respectively. complexes, the red-shift observed in the electroluminescence of
Within the LEEC, we were surprised that the electroluminescence complexes 79–82 is unlikely to be due to film morphology,
measurements were not consistent with the photoluminescence assuming similar solution-processing protocols. As has been
measurements. All of the complexes displayed red-shifted electro- identified in the literature, there are normally two possible
luminescence profile compared to their photoluminescence in explanations for this kind of phenomenon: (1) the electrolumi-
both solution or the solid state. Furthermore, we observed that the nescence mechanism gives rise to excimers, in which the effect is
trend in emission colour as a function of fluorine content of the stronger for the difluorinated complexes 80 and 81, or (2) in the
C^N ligands had reversed: complexes 79 and 80, which contained device there is a reordering of the relative energies of multiple
mono-fluorinated C^N ligands, in fact produced bluer emitting emissive triplet states, which although close in energy to each
devices [lEL = 508 nm and 487 nm for 79 and 80 respectively; CIE: other, may in fact differ significantly in their respective emission
(0.31, 0.44) for 79 and (0.26, 0.36) for 80] than the difluorinated energies.64 Regarding the first explanation, there are several
analogues [lEL = 569 nm and 508 nm for 81 and 82, respectively; reports of excimer formation in LEECs (such as complexes 85
CIE: (0.37, 0.45) and (0.28, 0.45) for 81 and 82]. Although this red- and 86), but frequently there is little to no characterisation of how
shift was undesired, the emission colour produced from the LEEC such excimers form in the device, particularly given the pseudo-
spherical nature of the Ir iTMC, making it difficult to assess the
validity of this explanation.37,65 The second explanation has been
invoked for 8764 and other complexes66 where one of multiple
different emissive triplet states (which differ significantly in
their respective emission energies) are operative based on their
environment. It is unclear which of these possibilities are
operative in the case of complexes 79–82, but these examples
serve to illustrate the difficulty in translating the performance
of these complexes in the solution state, all the way to the
device (Fig. 34).
In this section we have seen how unusual, unpredictable
Fig. 33 Iridium complexes displaying blue emission in LEECs. and ‘non-linear’ the evolution of the photophysical properties
Fig. 34 Iridium complexes reported to display significant red-shift in their electroluminescence profiles. For complexes 85 and 86, this was attributed to
excimer formation,37 while for 87 this was explained to be a result of reordering of multiple emissive states which are close in relative energy with respect
to each other, but differ greatly in terms of their emission energies.64
View Article Online
of these complexes can be when departing heavily from the 5 (a) Y. J. Yuan, Z. T. Yu, J. G. Cai, C. Zheng, W. Huang and Z. G. Zou,
archetypal structures of complexes 1 and 2. Complex 76 hit a ChemSusChem, 2013, 6, 1357–1365; (b) M. S. Lowry and S. Bernhard,
Chem. – Eur. J., 2006, 12, 7970–7977; (c) J. M. Thomsen, D. L. Huang,
‘photophysical wall’ when blue-shifting emission; 78 was found R. H. Crabtree and G. W. Brudvig, Dalton Trans., 2015, 44,
to be photophysically unstable; complexes 79 and 80 were bluer 12452–12472.
than their analogues 81 and 82 in the solid state. Although 6 (a) For a recent review on iridium-based LEECs see: C. Ulbricht,
B. Beyer, C. Friebe, A. Winter and U. S. Schubert, Adv. Mater., 2009,
we have been able to explain and rationalise the properties 21, 4418–4441; (b) H. Yersin, Highly Efficient OLEDs with Phosphorescent
underpinning some of these complexes, they remain interesting Materials, Wiley-VCH, Weinheim, 2008; (c) H. Yersin, A. F. Rausch,
examples that do not fit the traditional models. R. Czerwieniec, T. Hofbeck and T. Fischer, Coord. Chem. Rev., 2011, 255,
2622–2652.
7 (a) R. D. Costa, E. Ortı́, H. J. Bolink, F. Monti, G. Accorsi and

N. Armaroli, Angew. Chem., Int. Ed., 2012, 51, 8178–8211; (b) A. F.
7 Conclusions Henwood and E. Zysman-Colman, Top. Curr. Chem., 2016, 374, 1–41;
(c) J. D. Slinker, J. Rivnay, J. S. Moskowitz, J. B. Parker, S. Bernhard,
Within this perspective article we have explored in a systematic H. D. Abruña and G. G. Malliaras, J. Mater. Chem., 2007, 17,
fashion the evolution in photophysical properties of cationic 2976–2988.
8 (a) For recent reviews see: L. Flamigni, A. Barbieri, C. Sabatini,
iridium complexes as the C^N or N^N ligands are changed B. Ventura and F. Barigelletti, Top. Curr. Chem., 2007, 281, 143–203;
individually. We have also examined cases where modulation of (b) K. P. S. Zanoni, R. L. Coppo, R. C. Amaral and N. Y. Murakami
the structure of both ligands works to tune the optoelectronic Iha, Dalton Trans., 2015, 44, 14559–14573; (c) Y. Chi and P.-T. Chou,
Chem. Soc. Rev., 2010, 39, 638–655.
properties in concert. A combination of theoretical modelling, 9 L. He, L. Duan, J. Qiao, R. Wang, P. Wei, L. Wang and Y. Qiu,
physical organic chemistry and knowledge of the prior art have Adv. Funct. Mater., 2008, 18, 2123–2131.
proved to be useful tools in this endeavour. However, even with 10 K. Shanmugasundaram, M. S. Subeesh, C. D. Sunesh and Y. Choe,
RSC Adv., 2016, 6, 28912–28918.
this confluence of knowledge and experience, we have found on 11 For representative of white-emitting LEECs see: (a) L. He, J. Qiao,
more than one occasion that the properties of the bespoke L. Duan, G. Dong, D. Zhang, L. Wang and Y. Qiu, Adv. Funct. Mater.,
complexes do not always fit the expected trends, illustrating the WILEY-VCH Verlag, 2009, 19, 2950–2960; (b) R. D. Costa, F. J.
Cespedes-Guirao, E. Orti, H. J. Bolink, J. Gierschner, F. Fernandez-
challenges faced when designing new materials. Even in Lazaro and A. Sastre-Santos, Chem. Commun., 2009, 3886–3888;
instances when complexes behave in an expected way in one (c) K. Hasan, L. Donato, Y. Shen, J. D. Slinker and E. Zysman-
medium, such as in solution, their properties may not necessarily Colman, Dalton Trans., 2014, 43, 13672–13682; (d) A. B. Tamayo,
S. Garon, T. Sajoto, P. I. Djurovich, I. M. Tsyba, R. Bau and
translate in a predictable fashion to other media such as thin M. E. Thompson, Inorg. Chem., 2005, 44, 8723–8732.
films or in electroluminescent devices. A greater understanding of 12 (a) M. G. Colombo and H. U. Güdel, Inorg. Chem., 1993, 32,
how the optoelectronic properties of these emitters change upon 3081–3087; (b) M. G. Colombo, A. Hauser and H. U. Güdel, Inorg.
Chem., 1993, 32, 3088–3092; (c) J. D. Slinker, A. A. Gorodetsky,
moving from the solution state to the solid state is certainly M. S. Lowry, J. Wang, S. T. Parker, R. Rohl, S. Bernhard and
required in order to design improved emitters that produce EL G. G. Malliaras, J. Am. Chem. Soc., 2004, 126, 2763; (d) R. D. Costa,
devices with enhanced performance metrics. The breadth of E. Ortı̀, H. J. Bolink, S. Graber, S. Schaffner, M. Neuburger,
C. E. Housecroft and E. C. Constable, Adv. Funct. Mater., 2009, 19,
examples illustrated herein should demonstrate unequivocally
3456–3463; (e) R. D. Costa, E. Ortı́, D. Tordera, A. Pertegás,
the potential of cationic iridium complexes and their use in H. J. Bolink, S. Graber, C. E. Housecroft, L. Sachno, M. Neuburger
optoelectronic applications. and E. C. Constable, Adv. Energy Mater., 2011, 1, 282–290.
13 (a) S. Ladouceur, D. Fortin and E. Zysman-Colman, Inorg. Chem.,
2010, 49, 5625–5641; (b) D. Tordera, M. Delgado, E. Ortı́, H. J.
Bolink, J. Frey, M. K. Nazeeruddin and E. Baranoff, Chem. Mater.,
Acknowledgements 2012, 24, 1896–1903.
14 S. Ladouceur and E. Zysman-Colman, Eur. J. Inorg. Chem., 2013,
EZ-C acknowledges past and present group members for 2985–3007.
their tireless efforts and enthusiasm and the University of 15 (a) M. Bixon, J. Jortner, J. Cortes, H. Heitele and M. E. Michel-Beyerle,
St Andrews for financial support. We thank Johnson Matthey J. Phys. Chem., 1994, 98, 7289–7299; (b) J. V. Caspar and T. J. Meyer,
J. Phys. Chem., 1983, 87, 952–957.
and Umicore AG for the gift of materials. We would like to 16 A. F. Henwood, A. K. Bansal, D. B. Cordes, A. M. Z. Slawin, I. D. W.
thank the Engineering and Physical Sciences Research Council Samuel and E. Zysman-Colman, J. Mater. Chem. C, 2016, 4,
for financial support for Adam Henwood (EPSRC DTG Grants: 3726–3737.
17 H. J. Bolink, E. Coronado, R. n. D. Costa, N. Lardiés and E. Ortı̀,
EP/J500549/1; EP/K503162/1; EP/L505097/1). Inorg. Chem., 2008, 47, 9149–9151.
18 V. N. Kozhevnikov, Y. Zheng, M. Clough, H. A. Al-Attar, G. C.
Griffiths, K. Abdullah, S. Raisys, V. Jankus, M. R. Bryce and
References A. P. Monkman, Chem. Mater., 2013, 25, 2352–2358.
19 A. K. Pal, D. B. Cordes, A. M. Z. Slawin, C. Momblona, E. Ortı,
1 Y. You and W. Nam, Chem. Soc. Rev., 2012, 41, 7061–7084. I. D. W. Samuel, H. J. Bolink and E. Zysman-Colman, Inorg. Chem.,
2 (a) V. Guerchais and J.-L. Fillaut, Coord. Chem. Rev., 2011, 255, 2016, 55, 10361–10376.
2448–2457; (b) X. D. Wang and O. S. Wolfbeis, Chem. Soc. Rev., 2014, 20 J. Frey, B. F. E. Curchod, R. Scopelliti, I. Tavernelli, U. Rothlisberger,
43, 3666–3761. M. K. Nazeeruddin and E. Baranoff, Dalton Trans., 2014, 43,
3 (a) K. Lo, K. Tsang, K. Sze, C. Chung, T. Lee, K. Zhang, W. Hui, C. Li, 5667–5679.
J. Lau and D. Ng, Coord. Chem. Rev., 2007, 251, 2292–2310; (b) K. K.-W. 21 X. Yang, X. Xu and G. Zhou, J. Mater. Chem. C, 2014, 3, 913–944.
Lo, M.-W. Louie and K. Y. Zhang, Coord. Chem. Rev., 2010, 254, 22 N. M. Shavaleev, R. Scopelliti, M. Grätzel, M. K. Nazeeruddin,
2603–2622; (c) K. K. Lo, Acc. Chem. Res., 2015, 48, 2985–2995. A. Pertegás, C. Roldán-Carmona, D. Tordera and H. J. Bolink,
4 (a) K. Teegardin, J. I. Day, J. Chan and J. Weaver, Org. Process Res. J. Mater. Chem. C, 2013, 1, 2241.
Dev., 2016, 20, 1156–1163; (b) C. K. Prier, D. A. Rankic and 23 (a) C. D. Ertl, J. Cerdá, J. M. Junquera-Hernández, A. Pertegás, H. J.
D. W. C. MacMillan, Chem. Rev., 2013, 113, 5322–5363. Bolink, E. C. Constable, M. Neuburger, E. Ortı́ and C. E. Housecroft,
View Article Online
RSC Adv., 2015, 5, 42815–42827; (b) C. Fan, Y. Li, C. Yang, H. Wu, 43 F. Monti, A. Baschieri, I. Gualandi, J. J. Serrano-Perez, J. M. Junquera-
J. Qin and Y. Cao, Chem. Mater., 2012, 24, 4581–4587; (c) D. Tordera, Hernandez, D. Tonelli, A. Mazzanti, S. Muzzioli, S. Stagni, C. Roldan-
A. M. Bünzli, A. Pertegás, J. M. Junquera-Hernández, E. C. Carmona, A. Pertegas, H. J. Bolink, E. Ortı́, L. Sambri and N. Armaroli,
Constable, J. A. Zampese, C. E. Housecroft, E. Ortı́ and H. J. Inorg. Chem., 2014, 53, 7709–7721.
Bolink, Chem. – Eur. J., 2013, 19, 8597–8609. 44 P. Pla, J. M. Junquera-Hernandez, H. J. Bolink and E. Ortı́,
24 H. J. Bolink, L. Cappelli, E. Coronado, A. Parham and P. Stossel, Dalton Trans., 2015, 44, 8497–8505.
Chem. Mater., 2006, 18, 2778–2780. 45 For examples see: S. Ladouceur, D. Fortin and E. Zysman-Colman,
25 J.-H. Lee, G. Sarada, C.-K. Moon, W. Cho, K.-H. Kim, Y. G. Park, Inorg. Chem., 2011, 50, 11514–11526.
J. Y. Lee, S.-H. Jin and J.-J. Kim, Adv. Opt. Mater., 2015, 3, 211–220. 46 V. V. Pavlishchuk and A. W. Addison, Inorg. Chim. Acta, 2000, 298,
26 (a) S. J. Lee, K.-M. Park, K. Yang and Y. Kang, Inorg. Chem., 2008, 48, 97–102.
1030–1037; (b) Y. Kang, Y.-L. Chang, J.-S. Lu, S.-B. Ko, Y. Rao, 47 C. M. Cardona, W. Li, A. E. Kaifer, D. Stockdale and G. C. Bazan,
M. Varlan, Z.-H. Lu and S. Wang, J. Mater. Chem. C, 2013, 1, Adv. Mater., 2011, 23, 2367–2371.
441–450; (c) H. Oh, K.-M. Park, H. Hwang, S. Oh, J. H. Lee, J.-S. 48 M. Mydlak, C. Bizzarri, D. Hartmann, W. Sarfert, G. Schmid and
Lu, S. Wang and Y. Kang, Organometallics, 2013, 32, 6427–6436; L. De Cola, Adv. Funct. Mater., 2010, 20, 1812–1820.
(d) C.-H. Yang, M. Mauro, F. Polo, S. Watanabe, I. Muenster, 49 L. Donato, P. Abel and E. Zysman-Colman, Dalton Trans., 2013, 42,
R. Fröhlich and L. De Cola, Chem. Mater., 2012, 24, 3684–3695. 8402–8412.
27 (a) B. J. Coe, M. Helliwell, S. Sanchez, M. K. Peers and N. S. Scrutton, 50 N.-R. Jung, E.-J. Lee, J.-H. Kim, H.-K. Park, K.-M. Park and Y.-J. Kang,
Dalton Trans., 2015, 44, 15420–15423; (b) B. J. Coe, M. Helliwell, Bull. Korean Chem. Soc., 2012, 33, 183–188.
J. Raftery, S. Sanchez, M. K. Peers and N. S. Scrutton, Dalton Trans., 51 A. F. Henwood, S. Evariste, A. M. Z. Slawin and E. Zysman-Colman,
2015, 44, 20392–20405. Faraday Discuss., 2014, 174, 165–182.
28 (a) C. H. Chang, Z. J. Wu, C. H. Chiu, Y. H. Liang, Y. S. Tsai, 52 (a) S. B. Meier, W. Sarfert, J. M. Junquera-Hernández, M. Delgado,
J. L. Liao, Y. Chi, H. Y. Hsieh, T. Y. Kuo, G. H. Lee, H. A. Pan, D. Tordera, E. Ortı́, H. J. Bolink, F. Kessler, R. Scopelliti, M. Grätzel,
P. T. Chou, J. S. Lin and M. R. Tseng, ACS Appl. Mater. Interfaces, M. K. Nazeeruddin and E. Baranoff, J. Mater. Chem. C, 2013, 1, 58;
2013, 5, 7341–7351; (b) T. Duan, T. K. Chang, Y. Chi, J. Y. Wang, (b) J. M. Fernandez-Hernandez, S. Ladouceur, Y. Shen, A. Iordache,
Z. N. Chen, W. Y. Hung, C. H. Chen and G. H. Lee, Dalton Trans., X. Wang, L. Donato, S. Gallagher-Duval, M. de Anda Villa, J. D.
2015, 44, 14613–14624. Slinker, L. De Cola and E. Zysman-Colman, J. Mater. Chem. C, 2013,
29 (a) V. Sivasubramaniam, F. Brodkorb, S. Hanning, H. P. Loebl, 1, 7440–7452; (c) C. D. Sunesh, K. Shanmugasundaram, M. S.
V. van Elsbergen, H. Boerner, U. Scherf and M. Kreyenschmidt, Subeesh, R. K. Chitumalla, J. Jang and Y. Choe, ACS Appl. Mater.
J. Fluorine Chem., 2009, 130, 640–649; (b) Y. Zheng, A. S. Batsanov, Interfaces, 2015, 7, 7741–7751.
R. M. Edkins, A. Beeby and M. R. Bryce, Inorg. Chem., 2012, 51, 53 C.-H. Yang, J. Beltran, V. Lemaur, J. Cornil, D. Hartmann, W. Sarfert,
290–297; (c) D. Tordera, J. J. Serrano-Pérez, A. Pertegás, E. Ortı́, R. Fröhlich, C. Bizzarri and L. De Cola, Inorg. Chem., 2010, 49,
H. J. Bolink, E. Baranoff, M. K. Nazeeruddin and J. Frey, Chem. 9891–9901.
Mater., 2013, 25, 3391–3397. 54 K. Hasan, A. K. Pal, T. Auvray, E. Zysman-Colman and G. S. Hanan,
30 (a) S. Altomonte and M. Zanda, J. Fluorine Chem., 2012, 143, 57–93;
Chem. Commun., 2015, 51, 14060–14063.
(b) P. R. Savoie and J. T. Welch, Chem. Rev., 2014, 115, 1130–1190.
55 (a) M. S. Lowry, W. R. Hudson, R. A. Pascal Jr. and S. Bernhard,
31 N. M. Shavaleev, G. Xie, S. Varghese, D. B. Cordes, A. M. Z. Slawin,
J. Am. Chem. Soc., 2004, 126, 14129–14135; (b) E. C. Constable,
C. Momblona, E. Ortı́, H. J. Bolink, I. D. W. Samuel and E. Zysman-
C. E. Housecroft, E. Schönhofer, J. Schönle and J. A. Zampese,
Colman, Inorg. Chem., 2015, 54, 5907–5914.
Polyhedron, 2012, 35, 154–160; (c) S.-X. Luo, L. Wei, X.-H. Zhang,
32 J.-L. Bredas, Mater. Horiz., 2014, 1, 17.
M. H. Lim, K. X. V. Lin, M. H. V. Yeo, W.-H. Zhang, Z.-P. Liu,
33 S. Evariste, M. Sandroni, T. W. Rees, C. Roldan-Carmona, L. Gil-Escrig,
H. J. Bolink, E. Baranoff and E. Zysman-Colman, J. Mater. Chem. C, D. J. Young and T. S. A. Hor, Organometallics, 2013, 32, 2908–2917.
2014, 2, 5793–5804. 56 A. Tsuboyama, T. Takiguchi, S. Okada, M. Osawa, M. Hoshino and
34 K. Dedeian, P. I. Djurovich, F. O. Garces, G. Carlson and R. J. Watts, K. Ueno, Dalton Trans., 2004, 1115–1116.
Inorg. Chem., 1991, 30, 1685–1687. 57 (a) S. Culham, P. H. Lanoe, V. L. Whittle, M. C. Durrant, J. A.
35 D. L. Davies, M. P. Lowe, K. S. Ryder, K. Singh and S. Singh, Dalton Williams and V. N. Kozhevnikov, Inorg. Chem., 2013, 52, 10992–11003;
Trans., 2011, 40, 1028–1030. (b) V. N. Kozhevnikov, M. C. Durrant and J. A. G. Williams, Inorg.
36 K. Hasan, A. K. Bansal, I. D. W. Samuel, C. Roldán-Carmona, Chem., 2011, 50, 6304–6313.
H. J. Bolink and E. Zysman-Colman, Sci. Rep., 2015, 5, 12325. 58 A. M. Soliman, D. Fortin, P. D. Harvey and E. Zysman-Colman,
37 J. Zhang, L. Zhou, H. A. Al-Attar, K. Shao, L. Wang, D. Zhu, Z. Su, Dalton Trans., 2012, 41, 9382–9393.
M. R. Bryce and A. P. Monkman, Adv. Funct. Mater., 2013, 23, 59 L. Sun, A. Galan, S. Ladouceur, J. D. Slinker and E. Zysman-Colman,
4667–4677. J. Mater. Chem., 2011, 21, 18083–18088.
38 (a) J. L. Rodriguez-Redondo, R. D. Costa, E. Orti, A. Sastre-Santos, 60 (a) M. K. Nazeeruddin, R. T. Wegh, Z. Zhou, C. Klein, Q. Wang,
H. J. Bolink and F. Fernandez-Lazaro, Dalton Trans., 2009, 9787–9793; F. DeAngelis, S. Fantacci and M. Grätzel, Inorg. Chem., 2006, 45,
(b) S. Graber, K. Doyle, M. Neuburger, C. E. Housecroft, E. C. Constable, 9245–9250; (b) F. De Angelis, S. Fantacci, N. Evans, C. Klein,
R. n. D. Costa, E. Ortiz, D. Repetto and H. J. Bolink, J. Am. Chem. Soc., S. M. Zakeeruddin, J.-E. Moser, K. Kalyanasundaram, H. J. Bolink,
2008, 130, 14944–14945. M. Gratzel and M. K. Nazeeruddin, Inorg. Chem., 2007, 46,
39 (a) C. D. Sunesh, G. Mathai, Y.-R. Cho and Y. Choe, Polyhedron, 5989–6001.
2013, 57, 77–82; (b) C. D. Sunesh, M. Chandran, G. Mathai and 61 (a) K. N. Swanick, S. Ladouceur, E. Zysman-Colman and Z. Ding,
Y. Choe, Opt. Mater., 2013, 35, 407–413; (c) H.-B. Wu, H.-F. Chen, C.-T. Angew. Chem., Int. Ed., 2012, 51, 11079–11082; (b) S. Ladouceur,
Liao, H.-C. Su and K.-T. Wong, Org. Electron., 2012, 13, 483–490. K. N. Swanick, S. Gallagher-Duval, Z. Ding and E. Zysman-Colman,
40 H. J. Bolink, F. De Angelis, E. Baranoff, C. Klein, S. Fantacci, Eur. J. Inorg. Chem., 2013, 5329–5343.
E. Coronado, M. Sessolo, K. Kalyanasundaram, M. Gratzel and 62 C. E. Welby, G. K. Armitage, H. Bartley, A. Sinopoli, B. S. Uppal and
M. K. Nazeeruddin, Chem. Commun., 2009, 4672–4674. P. I. Elliott, Photochem. Photobiol. Sci., 2014, 13, 735–738.
41 (a) B. Beyer, C. Ulbricht, D. Escudero, C. Friebe, A. Winter, L. González 63 B. Chen, Y. Li, W. Yang, W. Luo and H. Wu, Org. Electron., 2011, 12,
and U. S. Schubert, Organometallics, 2009, 28, 5478–5488; (b) J. M. 766–773.
Fernández-Hernández, C.-H. Yang, J. I. Beltrán, V. Lemaur, F. Polo, 64 H. J. Bolink, L. Cappelli, S. Cheylan, E. Coronado, R. D. Costa,
R. Frõhlich, J. Cornil and L. De Cola, J. Am. Chem. Soc., 2011, 133, N. Lardies, M. K. Nazeeruddin and E. Orti, J. Mater. Chem., 2007, 17,
10543–10558. 5032–5041.
42 (a) S.-C. Lo, C. P. Shipley, R. N. Bera, R. E. Harding, A. R. Cowley, 65 E. C. Constable, C. E. Housecroft, G. E. Schneider, J. A. Zampese,
P. L. Burn and I. D. W. Samuel, Chem. Mater., 2006, 18, 5119–5129; H. J. Bolink, A. Pertegas and C. Roldan-Carmona, Dalton Trans.,
(b) C.-H. Yang, S.-W. Li, Y. Chi, Y.-M. Cheng, Y.-S. Yeh, P.-T. Chou, 2014, 43, 4653–4667.
G.-H. Lee, C.-H. Wang and C.-F. Shu, Inorg. Chem., 2005, 44, 66 Z. Wang, L. He, L. Duan, J. Yan, R. Tang, C. Pan and X. Song, Dalton
7770–7780. Trans., 2015, 44, 15914–15923.

Leer2 PDF

Uploaded by

Copyright:

Available Formats

Leer2 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Leer2 PDF

Uploaded by

Copyright:

Available Formats

Volume 53 Number 5 16 January 2017 Pages 793–986

Lessons learned in tuning the optoelectronic

properties of phosphorescent iridium(III)

Adam Francis Henwood graduated Eli Zysman-Colman obtained his

Feature Article ChemComm

LEECs, is that their emission can be readily tuned across the

ChemComm Feature Article

Feature Article ChemComm

3 C^N ligand – effects of substitution

substituent based on whether it is situated in a meta or para

ChemComm Feature Article

Feature Article ChemComm

Fig. 8 Emission in solution of complexes 16–19 (in order from left to

Fig. 6 Iridium complexes bearing a variety of trifluoromethyl-type

ChemComm Feature Article

Fig. 9 Iridium complexes based on cyclometalated 2,3 0 -bipyridines.

to mesomerically red-shift emission was reported by Watts et al.,34

Feature Article ChemComm

regiochemistry of a wide variety of diﬀerent C^N ligands

In contrast to ppz, which stabilises the HOMO, we observed

ChemComm Feature Article

Feature Article ChemComm

Fig. 13 Iridium complexes bearing 1,2,3-triazoles within the cyclometalating

Table 1 Electrochemical data comparing complexes bearing pyridyl

E1/2 (V vs. SCE) Eorbital (eV)

Fig. 15 Blue-emitting iridium complexes with modified ancillary ligands.

cationic iridium complexes. The potency of both the parent

ChemComm Feature Article

Fig. 16 Iridium complex bearing a bis-1,2,3-triazole ancillary ligand.

excited triplet state, which would subsequently twist back into

The emission of 51 in MeOH showed essentially the same

Fig. 17 Iridium complexes bearing various biimidazole-type ligands.

Feature Article ChemComm

thus reduces the conjugation length of the ligand.54 In particular,

ChemComm Feature Article

reported previously.55b We also targeted 57 for solution-processed

Feature Article ChemComm

Fig. 24 Di-nuclear iridium complex bridged by 2,5-di(pyridin-2-yl)pyrazine.

Fig. 25 Mono-nuclear iridium complexes bearing alkynyl substituents.

ring is more influential to the photophysics than the extended

ChemComm Feature Article

Fig. 26 Di-nuclear iridium complexes bearing alkynyl bridging units.

Fig. 27 Iridium complexes bearing diaryl substituted bipyridine ancillary ligands.

(lPL = 659 nm in MeCN). While 67 and 68 emit from a mixed

Feature Article ChemComm

ChemComm Feature Article

a photolysed sample showed that the primary degradation

(lPL = 470 nm for 78, 407 nm for 77 and 393 nm for 78 in

Fig. 31 Iridium complexes bearing 1,2,3-triazoles within the C^N and

ancillary ligand. Complexes 77 and 78 employ the same bis-

Feature Article ChemComm

ChemComm Feature Article

7 (a) R. D. Costa, E. Ortı́, H. J. Bolink, F. Monti, G. Accorsi and

Feature Article ChemComm

You might also like