Highly luminescent 2-phenylpyridine-free diiridium complexes with bulky 1,2-diarylimidazole cyclometalating ligands

Article abstract of DOI:10.1039/C8DT04043E

While a number of highly emissive dinuclear Ir(iii) complexes have been reported, they have generally been restricted to structures based on 2-phenylpyridine (Hppy) cyclometalates. We now present a series of new hydrazide-bridged diiridium complexes (5-8) which incorporate bulky 1,2-diarylimidazole cyclometalating ligands in the place of Hppy. Complexes 6-8 are strongly emissive when doped into poly(methyl methacrylate) (PMMA), displaying the highest PLQYs yet reported for ppy-free diiridium emitters (Φ_PL = 47-55 ± 10%). Notably, complex 8 has an emission peak at 452 nm and CIE_xy colour coordinates in the sky-blue region (0.18, 0.27), which is competitive with state-of-the-art monoiridium analogues. X-ray crystallography and solution-state ¹⁹F NMR spectra reveal the presence of rigidifying intramolecular π-π interactions for complexes 6-8, which explains their improved photophysical performance compared to 5 which does not have these interactions. Structure-property relationships are further rationalised through density functional theory (DFT) and cyclic voltammetry (CV) data. All the complexes studied in this work display aggregation induced phosphorescent emission (AIPE). This series of compounds increases the structural diversity of highly luminescent dinuclear Ir(iii) complexes to include luminophoric ligands that are not restricted to Hppy-type fragments. The colour range accessible to AIPE-active diiridum complexes has also been substantially broadened.

Full text of DOI:10.1039/C8DT04043E

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Highly luminescent 2-phenylpyridine-free
diiridium complexes with bulky 1,2-diarylimidazole
cyclometalating ligands†
Cite this: Dalton Trans., 2018, 47,
16524
Daniel G. Congrave, * Andrei S. Batsanov
and Martin R. Bryce
*
While a number of highly emissive dinuclear Ir(III) complexes have been reported, they have generally
been restricted to structures based on 2-phenylpyridine (Hppy) cyclometalates. We now present a series
of new hydrazide-bridged diiridium complexes (5–8) which incorporate bulky 1,2-diarylimidazole cyclo-
metalating ligands in the place of Hppy. Complexes 6–8 are strongly emissive when doped into poly
(methyl methacrylate) (PMMA), displaying the highest PLQYs yet reported for ppy-free diiridium emitters
(ΦPL = 47–55 10%). Notably, complex 8 has an emission peak at 452 nm and CIExy colour coordinates
in the sky-blue region (0.18, 0.27), which is competitive with state-of-the-art monoiridium analogues.
X-ray crystallography and solution-state 19F NMR spectra reveal the presence of rigidifying intramolecular
π–π interactions for complexes 6–8, which explains their improved photophysical performance compared
to 5 which does not have these interactions. Structure–property relationships are further rationalised
through density functional theory (DFT) and cyclic voltammetry (CV) data. All the complexes studied in
this work display aggregation induced phosphorescent emission (AIPE). This series of compounds
increases the structural diversity of highly luminescent dinuclear Ir(III) complexes to include luminophoric
ligands that are not restricted to Hppy-type fragments. The colour range accessible to AIPE-active
diiridum complexes has also been substantially broadened.
Received 8th October 2018,
Accepted 2nd November 2018
DOI: 10.1039/c8dt04043e
rsc.li/dalton
emitting complexes,14 which have been employed in sensing
applications.23,24 Complementarily, bridging ligands can lead
Introduction
Iridium(III) complexes have been extensively studied over the last to rigid complexes, either covalently through the incorporation
30 years.1 The metal–ligand based photochemistry has enabled of rigid polyaromatic structures (complexes 1, 2 and 3,
their emission colour to be tuned across the entire visible spec- Fig. 1),15,17,18,22,25 or non-covalently through promoting intra-
trum through modular synthesis.2 They typically feature high molecular perfluoroaryl–aryl π–π interactions with peripheral
luminescence quantum efficiencies (Φ), microsecond-scale ligands (complex 4, Fig. 1).26 This has allowed the develop-
phosphorescence lifetimes (τ) and good electrochemical stabi- ment of diiridium complexes that exhibit high solution photo-
lities. These properties are advantageous for applications3 such luminescence quantum yields (PLQYs) (>ca. 50%) from the red
as photocatalysis,4 biological labelling,5 sensing,6 and phos- to the sky-blue regions of the visible spectrum (e.g. complexes
phorescent organic light-emitting devices (PhOLEDs)7,8
1–4). Diiridium complexes have also demonstrated advantages
Interest in phosphorescent dinuclear lr(III) complexes has over their mononuclear counterparts. For example, sub-micro-
recently increased. Unlike their mononuclear analogues, they second phosphorescence lifetimes are obtained alongside
feature ligands which bridge the two metal centres. A wide high PLQYs (complexes 1 and 2) due to high radiative rate con-
variety of bridging ligands has been explored, and they can stants (kr), which is likely due to an improved spin–orbit coup-
heavily influence the photophysical properties of the ling (SOC) effect exerted by two proximal heavy metal
complexes.9–22 For example, flexible bridges impart aggrega- atoms.15,17,25 Chang and co-workers have also recently
tion-induced phosphorescent emission (AIPE) to orange/red- reported sky-blue diiridium phosphors with unity PLQY that
are sublimable.22
2-Phenylpyridine (Hppy)-based cyclometalating ligands
(Fig. 1) are popular for Ir(III) phosphors due to their synthetic
versatility, well-understood structure–property relationships,
Department of Chemistry, Durham University, South Road, Durham DH1 3LE, UK.
E-mail: d.g.congrave@outlook.com, m.r.bryce@durham.ac.uk
†Electronic supplementary information (ESI) available: Synthetic details, NMR
and because they reliably afford complexes with high
spectra, X-ray data. CCDC 1871136 for 7 and 1871137 for 8. For ESI and crystallo-
PLQYs.27–29 Consequently, Hppy-type fragments that form
graphic data in CIF or other electronic format see DOI: 10.1039/c8dt04043e
16524 | Dalton Trans., 2018, 47, 16524–16533
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Fig. 1 Representative highly emissive diiridium complexes containing 2-phenylpyridine (Hppy) fragments with selected solution photo-
luminescence parameters (all obtained in degassed DCM). 2-Phenylpyridine fragments are highlighted in green.
5-membered cyclometalates have been incorporated into the
bridging or peripheral ligands of almost all the significantly
emissive diiridium complexes (PLQY > ca. 30%) reported in
Results and discussion
Design, synthesis and characterisation
the literature (e.g. complexes 1–4).8,9,14,16,18,20,23,25 In contrast, The structures of the complexes 5–8 are shown in Fig. 2. The
non-Hppy components have been seldom explored in diiri- complexes are of the formula [(Ir(C^N)2)2(O^N^N^O)] with
dium complexes. We know of only a single structural type that C^N = 9–11 and O^N^N^O = 12 and 13. The conjugate acids of
does not contain a Hppy fragment and is significantly emissive the ligands in the complexes (H9–H11 and 2H12 and 2H13)
at room temperature: namely, cationic AIPE complexes featur- are also shown in Fig. 2. The mesityl-functionalised 1,2-diaryl-
ing 2-(phenyl)pyrazole (Hppz) cyclometalating ligands which imidazole cyclometalating ligand H9 was selected due to the
exhibit orange/red emission with PLQYs ≤ 31%.23,31 This lack favourable photophysical properties of its homoleptic
of structural diversity is restrictive, especially considering that complex,30,34 and because the steric bulk of the mesityl group
Hppy-based cyclometalating ligands have some drawbacks. For should impart solubility and rigidity without inhibiting the
example, the synthesis of Hppy-based ligands often requires formation of the complexes 5–8. The bridging ligands 2H12
expensive transition metal-catalysed routes.32 Also Hppy- and 2H13 were employed in complexes 5 and 6 to target sky-
derived sky-blue/blue emissive Ir(III) complexes suffer from blue emission and study the effect of intramolecular π–π stack-
poor excited state stability, particularly under PhOLED ing between the perfluoroaryl groups of 12 and the peripheral
operation.27
cyclometalating ligands.26 Due to the enhanced photophysical
Bulky 1,2-diarylimidazole ligands which form 5-membered performance of 6 compared to 5 (discussed below) the per-
cyclometalates are very topical as they have afforded highly fluoroaryl bridge 2H12 was incorporated into complexes 7 and
emissive sky-blue mononuclear Ir(III) complexes that are 8, where incremental fluorination of their cyclometalating
notably more stable under PhOLED operating conditions than ligands (H10 and H11) blue shifts their emission through
Hppy-functionalised analogues.27,30,33–38 They have also been HOMO stabilisation.
incorporated into heteroleptic mononuclear Ir(III) complexes
The bridging ligands 2H12 and 2H13 were synthesised as
that show promising preferential dipole alignment in solution- reported.26 The cyclometalating ligands H9–H11 were accessible
processed films.30 Beneficially, 1,2-diarylimidazole ligands can on a multi-gram scale following Strassner’s one-pot transition
also be synthesised from readily available starting materials metal-free procedure.39 The diiridium complexes 5–8 were then
through condensation chemistry, avoiding transition metal synthesised by cleaving the corresponding μ-Cl dimers with the
catalysis.39 Despite this promise, while 1,2-diarylimidazoles bridging ligands 2H12 or 2H13 under basic conditions.12,13,19,26
have been studied as cyclometalating ligands in homo- and Complexes 6–8 were obtained in ≥ca. 50% yields as diastereo-
heteroleptic monoiridium complexes,38,40–42 to the best of our meric mixtures which were not separated, apart from complex
knowledge they have not previously been applied to phosphor- 5, which was isolated as a single diastereomer. This follows
escent diiridium systems.
The aim of the present work is to diversify phosphorescent methyl bridge 2H13 affords diastereoselectivity.26
dinuclear Ir(III) complexes to include structures that are not Complexes 5–8 show good thermal stability with decompo-
literature precedent, where incorporation of the bis-trifluoro-
restricted to Hppy-based ligands. We present a series of new sition temperatures (Td corresponding to 5% weight loss) of
diiridium complexes (5–8) (Fig. 2) which feature bulky 1,2-di- >400 °C by thermal gravimetric analysis (TGA) (Fig. S6–S9†).
arylimidazole cyclometalating ligands alongside hydrazide
Expansions of the C6F5 regions of the 19F {1H} NMR spectra
bridging ligands. Density functional theory (DFT) calculations, of complexes 6–8 are shown in Fig. 3. The 19F NMR spectra for
X-ray crystallography and electrochemical data provide further the diastereomeric mixtures of 6–8 each display 10 environ-
insight into their structural and optoelectronic properties.
ments in the C6F5 region (5 per diastereoisomer). This is
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16524–16533 | 16525

View Article Online
Paper
Dalton Transactions
Fig. 2 Structures of the complexes and ligands studied in this work.
Fig. 3 Expansions of the C6F5 regions of the 376 MHz 19F {1H} spectra of 6–8 recorded in CDCl3 at 298 K. Diastereomeric ratios from integration: 6
ca. 1 : 1, 7 ca. 1 : 0.6, 8 ca. 1 : 0.9.
greater than the 6 environments that would be expected (3 per X-ray crystal structures
diastereoisomer) for mono-substituted perfluorophenyl
groups, and is due to a lowering of molecular symmetry. This The single-crystal X-ray crystal structures of the meso diastereo-
is ascribed to restriction of rotation of the bridge C6F5 groups mers of 7 and 8 are displayed in Fig. 4. Relevant parameters
due to intramolecular π–π interactions with peripheral cyclo- are listed in Table 1. Presumably, the meso diastereomers pre-
metalating ligands (seen in the solid state for 7 and 8 in the ferentially crystallised from solutions of diastereomeric mix-
X-ray diffraction data below).26,43
tures due to their inversion centre symmetry. Meso 7 crystal-
16526 | Dalton Trans., 2018, 47, 16524–16533
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
lised as a DCM hexasolvate whereas meso 8 crystallised as a
MeOH disolvate.
The Ir centres in both structures display distorted octa-
hedral coordination. The N atoms of the two C^N cyclometalat-
ing ligands occupy axial positions with respect to the bridge
plane, and are trans to one another.8,12 The central hydrazide
moieties of meso 7 and meso 8 are planar, and the aryl substi-
tuents (A) on the bridging ligands are oriented approximately
perpendicular to the hydrazide planes and are stacked face-to-
face (π–π) with the phenyl ring (B) of a cyclometalating ligand
(Fig. 4). The stacking is closer and more parallel in meso 8
compared to meso 7 (Θ = 3.0 vs. 3.5°, D = 3.26 vs. 3.30 Å).
However, the slightly larger centroid–centroid distance for
meso 8 (3.41 Å vs. 3.35 Å for meso 7) indicates greater slippage
of the stacked rings.
Electrochemical study
The oxidation and reduction processes for 5–8 were studied by
cyclic voltammetry (CV). The data are listed in Table 2. The oxi-
dative waves are presented in Fig. 5 and the reductive pro-
cesses are included in Fig. S1.† All complexes display two oxi-
dation waves.
They are assigned to sequential Ir3+/Ir4+ redox couples and
are indicative of electronic communication between the two
centres. Both oxidations are electrochemically reversible for
5–8 based on the equal magnitudes of the coupled oxidation
and reduction peaks. All reduction processes are electro-
chemical irreversible.
Fig. 4 X-ray molecular structures of meso 7 and meso 8. R = mesityl.
Thermal ellipsoids are drawn at the 50% probability level,
H
atoms, solvent of crystallisation and some mesityl groups
are omitted for clarity. Primed atoms are generated by a crystallographic
inversion centre. Vector D identifies intramolecular π–π interactions
(see Table 1).
The first oxidation potential of 5 is more positive than for 6
(by 0.04 V). This suggests that the complexed bis(trifluoro-
methyl) bridge 2H13 is more electron withdrawing than the
bis(pentafluorophenyl) bridge 2H12.44 Sequential fluorination
of the cyclometalating ligands in 6–8 leads to successive
increases in the first oxidation potentials, as expected. The
Table 1 Selected geometrical parameters of 7 and 8 (bond distances in
Å)
peak splitting between the first and second oxidations (ΔE1/2
)
for the series 6–8 also increases incrementally, suggesting that
the bridge HOMO contribution increases along the series.
Complex 5 displays the lowest ΔE1/2. This is because the com-
plexed bridge 2H13 has a shorter conjugation length than
2H12 and therefore is expected to feature a larger energy gap,
decreasing its contribution to the HOMO of 5. These con-
clusions are corroborated by the DFT data below.
The reduction onsets for 5–8 are comparable to the values
for heteroleptic mononuclear complexes functionalised with
2-arylimidazole ligands.41 While the irreversible nature of the
reductions hinders their accurate determination, there are two
clear qualitative trends in the data. (i) The reduction potential
for 5 is less negative than for 6, i.e. 5 is easier to reduce, in-line
with the more electron-withdrawing complexed bridge 2H13
and the higher first oxidation potential of 5. (ii) Sequential
fluorination in the series 6–8 leads to consecutively less nega-
tive reduction potentials as the complexes become more elec-
tron poor. However, the LUMO energies do not drop as signifi-
cantly as the HOMO energies upon fluorination, leading
to sequentially larger electrochemical bandgaps in the order
6 < 7 < 8.
meso 7 ·6CH2Cl2
meso 8 ·2MeOH
Space group
Mol. symmetry
Ir centres
C2/c
Ci
ΔΛ
5.022
2.003(4)
2.012(4)
2.020(3)
2.019(3)
Pbca
Ci
ΔΛ
5.065
2.003(3)
2.016(3)
2.018(2)
2.040(2)
Ir⋯Ir, Å
Ir–C (trans-O)
Ir–C (trans-N)
Ir–N, stacked
Ir–N, non-stacked
Bridge geometry
OCNNCO folding, °
Ir displacement, Å
Ir–O
Planar
0.086
2.119(3)
2.144(3)
1.434(4)
1.306(6)
1.279(4)
Planar
0.208
2.130(2)
2.159(2)
1.437(3)
1.305(4)
1.287(3)
Ir–N
N–N
N–C
C–O
Intramolecular stacking (π–π)
Θ a, °
3.5
3.30
3.35
3.0
3.26
3.41
D b, Å
Centroid–centroid, Å
a Interplanar angle between ring A of the bridging ligand and ring B of
the cyclometalating ligand (see Fig. 4). b Distance between the plane of
ring B and the centroid of ring A.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16524–16533 | 16527

View Article Online
Paper
Dalton Transactions
Table 2 Electrochemical data for complexes 5–8 referenced to E1/2 FcH/FcH+ = 0.00 V
ox(1)/V Epa/Epc [E1/2
]
E
ox(2)/V Epa/Epc [E1/2
]
ΔE1/2 a/V
E
b/V
HOMOc/eV
LUMOd/eV
red
Complex
E
onset
5
6
7
8
0.38/0.29 [0.34]
0.33/0.24 [0.30]
0.55/0.43 [0.49]
0.72/0.63 [0.68]
0.65/0.55 [0.60]
0.73/0.62 [0.67]
0.95/0.84 [0.89]
1.18/1.05 [1.12]
0.26
0.37
0.40
0.44
−2.82
−2.95
−2.89
−2.76
−5.14
−5.10
−5.29
−5.48
−1.98
−1.85
−1.91
−2.04
a Peak splitting between Eox(1) and Eox(2)
.
b All reductions are electrochemically irreversible. c HOMO levels calculated from CV potentials by
HOMO = −4.8 + (−E1/2ox(1)), using ferrocene as the standard. d LUMO levels calculated from CV potentials by LUMO = −4.8 + (−E
), using ferro-
red
onset
cene as the standard.
Fig. 5 Cyclic voltammograms in 0.1 M n-Bu4NPF6/DCM showing the
oxidation processes for complexes 5–8. The current range for each vol-
tammogram is 10 to −10 μA.
Photophysical properties
The absorption spectra of 5–8 are shown in Fig. S2† and the data
are listed in Table S1.† The spectra display features typical of
cyclometalated iridium complexes: there are intense bands below
ca. 300 nm which correspond to population of LC states, while
the weaker bands extending to ca. 450 nm are assigned to 1MLCT
and 3MLCT transitions.45,46 The extinction coefficients are higher
than for similar mononuclear 2-phenylimidazole complexes,41
ascribed to a larger number of cyclometalating ligands and Ir
atoms per complex. An increase in the intensity of the MLCT
bands compared to mononuclear analogues may be due to
improved spin–orbit coupling in dinuclear complexes.15,17
The emission spectra for 5–8 doped into poly(methyl meth-
Fig. 6 Emission spectra for complexes 5–8. (Top) Spectra of complexes
doped into PMMA at 1 wt% at room temperature. (Bottom) Spectra of
complexes in 2-MeTHF glasses at 77 K (λexc 355 nm).
3
acrylate) (PMMA) at room temperature, and in 2-MeTHF at complex 8 has an emission peak (452 nm) and CIExy coordi-
77 K are shown in Fig. 6. Photoluminescence data are tabu- nates (0.18, 0.27) that are competitive with the bluest mono-
lated in Table 3. Complexes 5–8 are non-emissive in DCM solu- iridium analogues that contain arylimidazole ligands.35,38 The
tion at room temperature with PLQYs ≤ 0.05%. In contrast, comparatively broad emission of 5–8 is reminiscent of mono-
they are emissive (PLQY = 11–55
10%) in the blue-green/ nuclear analogues.27,30,33–38 While we cannot completely
sky-blue regions at room temperature in dilute PMMA films disregard any effects of diastereomeric mixtures on the opto-
(1 wt% complex). To the best of our knowledge, complexes 6–8 electronic properties of 6–8, there are literature precedents
exhibit the highest PLQYs yet reported for ppy-free diiridium that diiridium diastereomers display very similar photo-
complexes (PLQY = 47–55 10%).23,31 Sequential fluorination physical properties.15,25
of the cyclometalating ligands in the series 6–8 leads to incre-
The matrix-dependent emission properties of 5–8 are
mental blue shifts in their emission, as expected. We note that ascribed to non-radiative emission quenching in solution. As
16528 | Dalton Trans., 2018, 47, 16524–16533
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Table 3 Summary of the key photoluminescence data for complexes 5–8
Doped into PMMA 1 wt%a
2-MeTHF glassb
Complex
λmax em/nm [CIExy]
PLQY/% ( 10%)
τ/μs
kr/×105 s−1
knr/×105 s−1
λmax em/nm (λ10% em/nm)c [ET /eV]d
τ/μs
5
6
7
8
469sh, 500 [0.20, 0.40]
470sh, 501 [0.20, 0.39]
456sh, 486 [0.18, 0.31]
452sh, 480 [0.18, 0.27]
11
55
47
52
1.82
2.80
4.15
4.55
0.60
1.96
1.13
1.14
4.89
1.61
1.28
1.05
466 (458) [2.71]
469 (459) [2.70]
456 (442) [2.81]
449sh, 480 (440) [2.82]
3.88
4.02
5.35
5.21
sh = Shoulder. a Measured in an integrating sphere under air using an excitation wavelength of 355 nm. b Measured at 77 K using an excitation
wavelength of 355 nm. c Wavelength at 10% intensity on the blue edge of the spectrum obtained at 77 K. d Estimated using ET = hc/λ10% em.τ =
1/knr + kr.
this is suppressed in a PMMA matrix, it is evident that the 5–8) have SM values ≥0.7, and those that are weakly emissive
dominant pathway for non-radiative decay in solution likely (PLQY = ca. 1–5%) have intermediary SM values of 0.5–0.7.
involves significant molecular motion, rather than any other
The PLQY of 5 in PMMA is notably lower than for 6–8 (11
process, for example C–C bond stretching.47 There are 10% vs. ca. 50 10%). This is ascribed to an absence of rigidi-
examples of homo- and heteroleptic monoiridium complexes fying intramolecular π–π interactions in 5 as its knr value is
functionalised with bulky 1,2-diarylimidazole ligands that are notably large (4.89 vs. 1.61 × 105 s−1 for 6). The phosphor-
highly emissive in solution,34,37,41 which indicates that the escence lifetimes of 6–8 are long for blue-green iridium
bridging ligands of 5–8 are likely to be the structural feature phosphors.53–57 For example, while 6 has a very similar ET to
responsible for their non-emissive behaviour in solution. the Hppy-derived complex 4 (2.70 vs. 2.72 eV) and a similar
Therefore, we conclude that non-radiative decay through PLQY in PMMA (55 vs. 65 10%), its τ is over twice as long
motion of the bridging ligand is responsible for the quenching (2.80 μs for 6 vs. 1.19 μs for 4). This is related to the substan-
of solution phosphorescence, which is reinforced by literature tially lower kr of 6 (1.96 × 105 s−1 vs. 5.46 × 105 s−1 for 4) which
precedents.13,14 There is precedent from work on other di- is likely a consequence of a lower MLCT contribution to the
iridium complexes that a rigid polymer matrix such as the excited state of 6. Therefore, the long phosphorescence life-
cycloolefin polymer zeonex could also lead to a similar times and low radiative rates of 5–8 are likely to stem from
emission enhancement.17
This property could be anticipated for complex 5 which from the well-resolved vibronic features in their emission
does not feature rigidifying intramolecular interactions to spectra recorded in PMMA at room temperature: the v0,0, v0,1
high LC contributions to their excited states. This is evident
,
restrict bridge motion.13,26,48,49 However, it is more surprising v0,2 and v0,3 bands are all reasonably well-resolved. This con-
for 6–8, for which intramolecular π–π interactions are observed clusion is also supported by minimal rigidochromic shifts in
in their solution 19F NMR spectra (Fig. 3) (similar to complex the emission spectra of 5–8 upon cooling (mentioned above).
4, Fig. 1). The non-emissive nature of complexes 6–8 in solu- Blue shifting the emission in the series 6–8 through fluorina-
tion could be related to their high triplet energies (ET) (≥2.70 tion of the cyclometalating ligands of 7 and 8 also leads to
eV), as we have previously noted that intramolecular π–π inter- incremental increases in τ (i.e. for 8 τ = 4.55 μs and kr = 1.96 ×
actions become less effective at suppressing the non-radiative 105 s−1). This fits a typical trend in Ir(III) phosphors, where the
decay of hydrazide-bridged diiridium complexes in solution as LC character of the excited state increases upon blue shifting
their emission energies increase.26
the emission.29,58,59
The emission spectra of 5–8 at 77 K in 2-MeTHF are rela-
Complexes 5–8 are emissive under UV irradiation (365 nm)
tively broad and show distinct vibronic features. Minimal rigi- in the solid state as powders. Typical aggregation-induced
dochromic shifts are observed on cooling (≤3 nm) compared phosphorescent emission (AIPE) behaviour47,60 is observed by
to the room temperature emission spectra recorded in PMMA. titrating water into THF solutions of the complexes to induce
This implies a strong LC contribution to the excited states of precipitation/aggregation, which promotes emission. The
5–8.50 The Huang–Rhys factors (SM) (estimated from the ratio emission intensity increases as the THF fraction decreases.
of the v0,0 and v0,1 band intensities51,52) are also large: 5 = 0.8, Spectra for complexes 6 and 8 are shown in Fig. 7. Spectra for
6 = 0.8, 7 = 1.0 and 8 > 1.0 (1 s.f.). These values imply that the 5 and 7 are included in Fig. S3.† The mechanism which results
excited states of 5–8 are highly distorted compared to their in solid state emission from 5–8 is, in principle no different
ground states,45 and are related to their non-emissive pro- from that which promotes emission in dilute PMMA films, as
perties in solution. Comparing these data with those obtained evident from near-identical spectral profiles. In the solid state
for previous complexes,13,26 there is a rational inverse relation- intramolecular motion is restricted due to interactions
ship between the Huang–Rhys factor and the solution PLQY between neighbouring complexes, rather than between the
for hydrazide-bridged diiridium complexes. Complexes with complexes and a PMMA host. This property is enabled by the
SM values of <0.5 (e.g. complex 4) tend to be highly emissive bulky (and ancillary – see DFT below) mesityl groups. In the
(PLQY = ≥50%), those that are non-emissive in solution (e.g. solid state they increase the distances between the emissive
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16524–16533 | 16529

View Article Online
Paper
Dalton Transactions
The predicted frontier molecular orbital (FMO) contri-
butions are listed in Tables S2 and S3.† Generally, there is a
good agreement between diastereomers, and so FMO plots for
meso 5–8 are presented in Fig. 8 (the FMO plots for the rac dia-
stereomers are included in Fig. S4 and S5†). For the diastereo-
mers of 5, the HOMOs are primarily localised on the Ir centres
and the cyclometalating ligands, with some contribution from
the central hydrazide fragments of the bridging ligands, while
the LUMOs are cyclometalating ligand-based. The spatial sep-
aration of the FMOs on the cyclometalating ligands is less
defined than for typical Hppy-based complexes,45,52,63 i.e. the
LUMO contribution is split nearly equally between the phenyl
and imidazole moieties. This is in good agreement with
studies on mononuclear complexes with similar 1,2-diarylimid-
azole cyclometalating ligands.40–42
For 6–8 the HOMOs are mainly localised on the Ir centres
and the central hydrazide fragments of the bridging ligands.
Interestingly, rather than being based on the cyclometalating
ligands, the LUMOs are primarily localised on the bridge
pentafluorophenyl groups for 6–8. This contrasts with the data
reported for Hppy-based analogues such as 4, for which the
pentafluorophenyl groups are ancillary.26 Presumably, this is
due to the more electron rich nature of the imidazole hetero-
cycles compared to pyridine, which forces the LUMO onto the
strongly electron accepting pentafluorophenyl groups. As a
result, the cyclometalating ligands of 6–8 are not major FMO
contributors (their HOMO and LUMO contributions are
Fig. 7 Emission spectra for THF solutions of complexes 6 and 8 upon
incremental titration of water to induce precipitation (λexc 355 nm). THF
fraction is percentage volume.
‘cores’ of the complexes, suppressing triplet–triplet annihil-
ation as
a quenching pathway (which dominates for
unshielded diiridium complexes13,61,62). Complexes 5–8 greatly
extend the spectral range of diiridium AIPE emitters – the
most prominent literature examples are orange/red14,23 while 8
is sky-blue (CIExy = 0.18, 0.27 in PMMA).
Computational study
Electronic structure calculations were carried out on 5–8 at the
B3LYP/LANL2DZ:3–21G* level12,13,19,26 to gain insight into the
photophysical properties of the complexes. In the optimised
structures of 5–8 the central hydrazide fragments are predicted
to be close to planar for both the meso and rac diastereomers.
This contrasts with previously reported analogues,12,13,26 for
which the rac forms tend to be folded, and is assigned to the
highly congested nature of the structures. The optimised struc-
tures of meso 7 and meso 8 are in good agreement with the
X-ray data. However, the optimised geometries of the rac forms
of 5–8 cannot be compared with X-ray data as no rac structures
have been solved. Such similar optimised geometries for the
rac and meso diastereomers of 6–8 may explain why they could
not be separated.
Fig. 8 Molecular orbital compositions for meso 5–8. The orbital contri-
butions are percentages and the HOMO and LUMO energies were cal-
culated at B3LYP/LANL2DZ:3–21G*. Bridge = central bridge OCNNCO
fragment; F5 = fluorinated bridge aryl rings; Ph = cyclometalating ligand
phenyl groups; Im = cyclometalating ligand imidazolyl groups.
16530 | Dalton Trans., 2018, 47, 16524–16533
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Table 4 Summary of the TD-DFT data for complexes 6 and 8
6
8
meso
rac
meso
rac
Main orbital
Transition contribution
λ/
Main orbital
λ/
Main orbital
λ/
Main orbital
λ/
nm
nm contribution
nm contribution
nm contribution
S0 → T1
S0 → T2
S0 → T3
S0 → T4
S0 → T5
HOMO → LUMO
429 HOMO−1 → LUMO+3,
HOMO → LUMO+1
426 HOMO → LUMO
420 HOMO−1 → LUMO+3
409
409
403
403
402
HOMO−2 → LUMO+3,
HOMO → LUMO+1
HOMO → LUMO+3,
HOMO−2 → LUMO+2
HOMO−1 → LUMO+2
425 HOMO−1 → LUMO+1,
425 HOMO → LUMO+1
410 HOMO−1 → LUMO+2,
HOMO → LUMO+3
HOMO → LUMO+3
423 HOMO−1 → LUMO+4,
HOMO−2 → LUMO+1
416 HOMO−2 → LUMO+4
417 HOMO → LUMO+3,
HOMO−2 → LUMO+1
408 HOMO−1 → LUMO+4,
HOMO−2 → LUMO+5
417 HOMO−1 → LUMO+4,
HOMO−2 → LUMO+2
402 HOMO−2 → LUMO+4
HOMO−1 → LUMO+4
415 HOMO → LUMO+17
413 HOMO−1 → LUMO+2,
HOMO−2 → LUMO+4
401 HOMO → LUMO+8
≤20%). Nevertheless, complexes 6–8 are emissive despite their FMO analysis.35,40 This analysis is in agreement with the
unusual FMO distributions, highlighting the versatility of photophysical data above. For example, the observation that
hydrazide bridges as a platform for obtaining emissive diiri- the profiles of the PL spectra for 6–8 are very similar to those
dium complexes.
of homoleptic 2-phenylimidazole complexes30,33–38 (which do
To determine the significance of the unusual FMO distri- not feature bridging ligands) signifies that their excited states
butions of 6–8, a time-dependent density functional theory should possess a high degree of LC character on the cyclome-
(TD-DFT) study was carried out to gain insight into the nature talating ligands.
of their lowest energy excited states. This is because a simple
consideration of the FMOs is not necessarily sufficient to
predict the localisation of the lowest energy triplet states of
Conclusions
Ir(III) complexes.64–66 The data for both diastereomers of the
least (6) and most (8) fluorinated derivatives are presented in A new family of hydrazide-bridged diiridium complexes 5–8
Table 4 (the data for 7 show the same trends and are included based on bulky 1,2-diarylimidazole cyclometalating ligands
in Table S4†). The two largest contributing transitions to each has been studied in detail.
state (≥ca. 20%) are included. The TD-DFT data for 5 are
included in Table S4.†
The complexes 6–8 are strongly emissive when doped into
PMMA. Significantly emissive ppy-free diiridium complexes
Both diastereomers of 6–8 feature 5 triplet states that are are rare, and of these 6–8 exhibit the highest PLQYs yet
relatively close in energy (≤20 nm). A number of these states reported (ΦPL = 47–55 10%).23,31 They display emission peaks
may, therefore, be relevant when considering their emissive as blue shifted as 452 nm and complex 8 has CIExy colour
states.64 Such a large number of near-degenerate states, many coordinates in the sky-blue region (0.18, 0.27). Prior to this
of which have significant contributions from multiple tran- work there have been only two reports of sky-blue diiridium
sitions, complicates detailed analysis of the data. However, it complexes,22,26 both containing Hppy ligands. We have shown,
can be noted that as well as a HOMO → LUMO transition, therefore, that Hppy is not essential for obtaining highly emis-
many of the relevant transitions in Table 4 involve contri- sive diiridium complexes, and also Hppy is not required for
butions from higher energy unoccupied (LUMO+1–LUMO+3), the challenging task of shifting their emission into the sky-
and lower energy occupied (HOMO−1 and HOMO−2) orbitals. blue region. Moreover, complex 8 is as blue as the bluest
Contributions for the HOMO−5–LUMO+5 orbitals are tabu- mono-Ir complexes yet reported based on arylimidazole
lated in Tables S2 and S3.† Particularly, while the LUMO+2 ligands.35,38
and LUMO+3 orbitals of 6–8 generally include a degree of pen-
X-ray crystallography and solution-state 19F NMR spectra
tafluorophenyl character, their cyclometalating ligand charac- reveal the presence of rigidifying intramolecular π–π inter-
ter is much higher than for the LUMOs (≥50%, as high as actions for complexes 6–8, which explains their improved
100%). Also, the HOMO−1 and HOMO−2 orbitals for 6–8 are PLQYs compared to 5. The rather long phosphorescence life-
3
almost exclusively distributed between the Ir atoms and cyclo- times of 6–8 have been attributed to the high LC character of
metalating ligands (ca. 50 : 50 in all cases), in contrast to the their excited states, which is corroborated by TD-DFT.
HOMO orbitals that are mainly Ir and hydrazide based.
The complexes also display AIPE behaviour. This substan-
Therefore, while it is likely that the pentafluorophenyl substi- tially broadens the colour range that can now be accessed by
tuents are somewhat involved in the excited states of 6–8, AIPE diiridium emitters towards the sky-blue and should
TD-DFT predicts that the cyclometalating ligands are more provide added versatility in applications such as anti-counter-
involved in their emitting states than is implied by a simple feiting.67 This study considerably increases the scope of dinuc-
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16524–16533 | 16531

View Article Online
Paper
Dalton Transactions
lear Ir(III) complexes to include luminophoric ligands that are 17 R. E. Daniels, S. Culham, M. Hunter, M. C. Durrant,
not restricted to Hppy-type cyclometalates, and provides a
foundation for further diversification of emissive diiridium
complexes away from conventional Hppy architectures.
M. R. Probert, W. Clegg, J. A. G. Williams and
V. N. Kozhevnikov, Dalton Trans., 2016, 45, 6949–6962.
18 X. Yang, Z. Feng, J. Zhao, J.-S. Dang, B. Liu, K. Zhang and
G. Zhou, ACS Appl. Mater. Interfaces, 2016, 8, 33874–33887.
19 A. M′hamedi, M. A. Fox, A. S. Batsanov, H. A. Al-Attar,
A. P. Monkman and M. R. Bryce, J. Mater. Chem. C, 2017, 5,
6777–6789.
20 M. Y. Wong, G. Xie, C. Tourbillon, M. Sandroni,
D. B. Cordes, A. M. Z. Slawin, I. D. W. Samuel and
E. Zysman-Colman, Dalton Trans., 2015, 44, 8419–8432.
21 G. Li, D. G. Congrave, D. Zhu, Z. Su and M. R. Bryce,
Polyhedron, 2018, 140, 146–157.
22 J. L. Liao, P. Rajakannu, P. Gnanasekaran, S. R. Tsai,
C. H. Lin, S. H. Liu, C. H. Chang, G. H. Lee, P. T. Chou,
Z. N. Chen and Y. Chi, Adv. Opt. Mater., 2018, 6, 1800083.
23 G. Li, W. Guan, S. Du, D. Zhu, G. Shan, X. Zhu, L. Yan,
Z. Su, M. R. Bryce and A. P. Monkman, Chem. Commun.,
2015, 51, 16924–16927.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We thank EPSRC for funding equipment used in this study
(grant EP/K0394/23/1).
Notes and references
1 Iridium(III) in Optoelectronic and Photonics Applications, ed.
E. Zysman-Colman, John Wiley & Sons, Chichester, 2017.
2 Y. Chi and P.-T. Chou, Chem. Soc. Rev., 2010, 39, 638–655.
24 Y. Jiang, G. Li, D. Zhu, Z. Su and M. R. Bryce, J. Mater.
Chem. C, 2017, 5, 12189–12193.
3 Z. Q. Chen, Z. Q. Bian and C. H. Huang, Adv. Mater., 2010, 25 X. Yang, X. Xu, J. Dang, G. Zhou, C.-L. Ho and W.-Y. Wong,
22, 1534–1539. Inorg. Chem., 2016, 55, 1720–1727.
4 M. S. Lowry and S. Bernhard, Chem. – Eur. J., 2006, 12, 26 D. G. Congrave, Y.-T. Hsu, A. S. Batsanov, A. Beeby and
7970–7977. M. R. Bryce, Dalton Trans., 2018, 47, 2086–2098.
5 K. K.-W. Lo, K. H.-K. Tsang, K.-S. Sze, C.-K. Chung, T. 27 Y. Im, S. Y. Byun, J. H. Kim, D. R. Lee, C. S. Oh, K. S. Yook
K.-M. Lee, K. Y. Zhang, W.-K. Hui, C.-K. Li, J. S.-Y. Lau, D. and J. Y. Lee, Adv. Funct. Mater., 2017, 27, 1603007–1603031.
C.-M. Ng and N. Zhu, Coord. Chem. Rev., 2007, 251, 2292– 28 Y. Chi and P.-T. Chou, Chem. Soc. Rev., 2010, 39, 638–655.
2310.
29 J. Li, P. I. Djurovich, B. D. Alleyne, M. Yousufuddin,
N. N. Ho, J. C. Thomas, J. C. Peters, R. Bau and
M. E. Thompson, Inorg. Chem., 2005, 44, 1713–1727.
30 J. Zhuang, W. Li, W. Wu, M. Song, W. Su, M. Zhou and
Z. Cui, New J. Chem., 2015, 39, 246–253.
31 G. Li, X. Ren, G. Shan, W. Che, D. Zhu, L. Yan, Z. Su and
M. R. Bryce, Chem. Commun., 2015, 51, 13036–13039.
32 R. Davidson, Y.-T. Hsu, T. Batchelor, D. Yufit and A. Beeby,
Dalton Trans., 2016, 45, 11496–11507.
33 J. Zhuang, W. Li, W. Su, Y. Liu, Q. Shen, L. Liao and
M. Zhou, Org. Electron., 2013, 14, 2596–2601.
34 K. Udagawa, H. Sasabe, C. Cai and J. Kido, Adv. Mater.,
2014, 26, 5062–5066.
6 R. Gao, D. G. Ho, B. Hernandez, M. Selke, D. Murphy,
P. I. Djurovich and M. E. Thompson, J. Am. Chem. Soc.,
2002, 124, 14828–14829.
7 X. Yang, G. Zhou and W.-Y. Wong, Chem. Soc. Rev., 2015,
44, 8484–8575.
8 C. Ulbricht, B. Beyer, C. Friebe, A. Winter and
U. S. Schubert, Adv. Mater., 2009, 21, 4418–4441.
9 T. Hajra, A. J. K. Bera and V. Chandrasekhar, Aust. J. Chem.,
2011, 64, 561–566.
10 A. M. Prokhorov, A. Santoro, J. A. G. Williams and
D. W. Bruce, Angew. Chem., Int. Ed., 2012, 51, 95–98.
11 M. Graf, R. Czerwieniec and K. Sünkel, Z. Anorg. Allg.
Chem., 2013, 639, 1090–1094.
35 H. Cho, J. Lee, J. I. Lee, N. S. Cho, J. H. Park, J. Y. Lee and
Y. Kang, Org. Electron., 2016, 34, 91–96.
12 Y. Zheng, A. S. Batsanov, M. A. Fox, H. A. Al-Attar,
K. Abdullah, V. Jankus, M. R. Bryce and A. P. Monkman, 36 Y. Zhang, J. Lee and S. R. Forrest, Nat. Commun., 2014, 5,
Angew. Chem., Int. Ed., 2014, 53, 11616–11619. 1–7.
13 D. G. Congrave, Y. Hsu, A. S. Batsanov, A. Beeby and 37 K. Udagawa, H. Sasabe, F. Igarashi and J. Kido, Adv. Opt.
M. R. Bryce, Organometallics, 2017, 36, 981–993. Mater., 2016, 4, 86–90.
14 G. Li, Y. Wu, G. Shan, W. Che, D. Zhu, B. Song, L. Yan, 38 Y. Kwon, S. H. Han, S. Yu, J. Y. Lee and K. M. Lee, J. Mater.
Z. Su and M. R. Bryce, Chem. Commun., 2014, 50, 6977–
6980.
15 P.-H. Lanoë, C. M. Tong, R. W. Harrington, M. R. Probert,
Chem. C, 2018, 6, 4565–4572.
39 M. Micksch, M. Tenne and T. Strassner, Eur. J. Org. Chem.,
2013, 6137–6145.
W. Clegg, J. A. G. Williams and V. N. Kozhevnikov, Chem. 40 T. Karatsu, M. Takahashi, S. Yagai and A. Kitamura, Inorg.
Commun., 2014, 50, 6831–6834. Chem., 2013, 52, 12338–12350.
16 V. Chandrasekhar, T. Hajra, J. K. Bera, S. M. W. Rahaman, 41 E. Baranoff, S. Fantacci, F. De Angelis, X. Zhang,
N. Satumtira, O. Elbjeirami and M. A. Omary, Inorg. Chem.,
2012, 51, 1319–1329.
R. Scopelliti, M. Grätzel and M. K. Nazeeruddin, Inorg.
Chem., 2011, 50, 451–462.
16532 | Dalton Trans., 2018, 47, 16524–16533
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
42 S. Takizawa, J. Nishida, T. Tsuzuki, S. Tokito and 55 T. Duan, T.-K. Chang, Y. Chi, J.-Y. Wang, Z.-N. Chen,
Y. Yamashita, Inorg. Chem., 2007, 46, 4308–4319.
43 D. G. Congrave, A. S. Batsanov, M. Du, Y. Liu, D. Zhu and
M. R. Bryce, Inorg. Chem., 2018, 57, 12836–12849.
44 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91,
165–195.
W.-Y. Hung, C.-H. Chen and G.-H. Lee, Dalton Trans., 2015,
44, 14613–14624.
56 H. J. Park, J. N. Kim, H. Yoo, K. Wee, S. O. Kang and
D. W. Cho, J. Org. Chem., 2013, 78, 8054–8064.
57 T. B. Fleetham, L. Huang, K. Klimes, J. Brooks and J. Li,
Chem. Mater., 2016, 28, 3276–3282.
45 A. B. Tamayo, B. D. Alleyne, P. I. Djurovich, S. Lamansky,
I. Tsyba, N. N. Ho, R. Bau and M. E. Thompson, J. Am. 58 M. A. Baldo, S. R. Forrest and M. E. Thompson, in Organic
Chem. Soc., 2003, 125, 7377–7387.
Electroluminescence, ed. Z. H. Kafafi, CRC and SPIE Press,
46 S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq,
2005.
H. E. Lee, C. Adachi, P. E. Burrows, S. R. Forrest and 59 L. Yang, F. Okuda, K. Kobayashi, K. Nozaki, Y. Tanabe,
M. E. Thompson, J. Am. Chem. Soc., 2001, 123, 4304–4312.
47 J. Mei, N. L. C. Leung, R. T. K. Kwok, J. W. Y. Lam and
B. Z. Tang, Chem. Rev., 2015, 115, 11718–11940.
48 C. E. Housecroft and E. C. Constable, Coord. Chem. Rev.,
2017, 350, 155–177.
49 R. D. Costa, E. Ortí, H. J. Bolink, F. Monti, G. Accorsi and
N. Armaroli, Angew. Chem., Int. Ed., 2012, 51, 8178–8211.
50 E. S. Andreiadis, D. Imbert, J. Pécaut, A. Calborean,
Y. Ishii and M.-A. Haga, Inorg. Chem., 2008, 47, 7154–
7165.
60 L. Ravotto and P. Ceroni, Coord. Chem. Rev., 2017, 346, 62–
76.
61 S. Reineke, K. Walzer and K. Leo, Phys. Rev. B: Condens.
Matter Mater. Phys., 2007, 75, 125328.
62 N. C. Giebink and S. R. Forrest, Phys. Rev. B: Condens.
Matter Mater. Phys., 2008, 77, 235215.
I. Ciofini, C. Adamo, R. Demadrille and M. Mazzanti, Inorg. 63 H. Benjamin, Y. Zheng, A. S. Batsanov, M. A. Fox, H. A. Al-
Chem., 2011, 50, 8197–8206.
51 G. Li, T. Fleetham, E. Turner, X. C. Hang and J. Li, Adv.
Opt. Mater., 2015, 3, 390–397.
52 J. Li, P. I. Djurovich, B. D. Alleyne, M. Yousufuddin,
N. N. Ho, J. C. Thomas, J. C. Peters, R. Bau and
M. E. Thompson, Inorg. Chem., 2005, 44, 1713–1727.
Attar, A. P. Monkman and M. R. Bryce, Inorg. Chem., 2016,
55, 8612–8627.
64 H. J. Bolink, F. De Angelis, E. Baranoff, C. Klein,
S. Fantacci, E. Coronado, M. Sessolo, K. Kalyanasundaram,
M. Grätzel and M. K. Nazeeruddin, Chem. Commun., 2009,
4672–4674.
53 A. F. Henwood, A. K. Bansal, D. B. Cordes, A. M. Z. Slawin, 65 E. Baranoff and B. F. E. Curchod, Dalton Trans., 2015, 44,
I. D. W. Samuel and E. Zysman-Colman, J. Mater. Chem. C,
2016, 4, 3726–3737.
54 C.-H. Yang, M. Mauro, F. Polo, S. Watanabe, I. Muenster,
8318–8329.
66 X. Gu, T. Fei, H. Zhang, H. Xu, B. Yang, Y. Ma and X. Liu,
J. Phys. Chem. A, 2008, 112, 8387–8393.
R. Frohlich and L. De Cola, Chem. Mater., 2012, 24, 3684– 67 Y. Jiang, G. Li, W. Che, Y. Liu, B. Xu, G. Shan, D. Zhu, Z. Su
3695.
and M. R. Bryce, Chem. Commun., 2017, 53, 3022–3025.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16524–16533 | 16533