Ligand-Tuneable, Red-Emitting Iridium(III) Complexes for Efficient Triplet–Triplet Annihilation Upconversion Performance

A Journal of

Accepted Article

Title: Ligand tuneable, red-emitting iridium(III) complexes for efficient

triplet-triplet annihilation upconversion performance

Authors: Simon J. Pope, Thomas Stonelake, Kaitlin Phillips, Simon

Coles, Peter Horton, Shannon Keane, Emily Stokes, Ian

Fallis, Joseph Beames, Jianzhang Zhao, Kepeng Chen, Yuqi

Hou, Andrew Hallett, and Sean O'Kell

This manuscript has been accepted after peer review and appears as an

Accepted Article online prior to editing, proofing, and formal publication

of the final Version of Record (VoR). This work is currently citable by

using the Digital Object Identifier (DOI) given below. The VoR will be

published online in Early View as soon as possible and may be different

to this Accepted Article as a result of editing. Readers should obtain

the VoR from the journal website shown below when it is published

to ensure accuracy of information. The authors are responsible for the

content of this Accepted Article.

To be cited as: Chem. Eur. J. 10.1002/chem.201801007

Link to VoR: http://dx.doi.org/10.1002/chem.201801007

Supported by

Chemistry - A European Journal

10.1002/chem.201801007

Ligand tuneable, red-emitting iridium(III) complexes for efficient triplet-triplet

annihilation upconversion performance

Kaitlin A. Phillips,[a] Thomas M. Stonelake, [a] Kepeng Chen, [b] Yuqi Hou, [b]

Jianzhang Zhao, [b] Simon J. Coles, [c] Peter N. Horton, [c] Shannon J. Keane, [a]

Emily C. Stokes, [a] Ian A. Fallis, [a] Andrew J. Hallett,[d] Sean P. O’Kell,[d] Joseph

M. Beames,* [a] and Simon J.A. Pope*[a]

[

a]

b]

K.A. Phillips, T.M. Stonelake, S. J. Keane, Dr E.C. Stokes, Dr I.A. Fallis, Dr J.M. Beames, Prof. S.J.A. Pope

School of Chemistry, Main Building

Cardiff University

Cardiff, CF10 3AT (UK)

E-mail: beamesj@cardiff.ac.uk; popesj@cardiff.ac.uk

K. Chen, Y. Hou, Prof. J. Zhao

State Key Laboratory of Fine Chemicals

Dalian University of Technology

Dalian 116024 (P.R. China)

[

c]

d]

Prof. S.J. Coles, Dr P.N. Horton

UK National Crystallographic Service, Chemistry,

University of Southampton, Highfield, Southampton, SO17 1BJ (UK)

Dr A.J. Hallett, Dr S. P. O’Kell

STG Aerospace, Brecon House,

Cwmbran, NP44 3AB (UK)

Abstract: A series of substituted 2-phenylquinoxaline ligands have been explored to

finely tune the visible emission properties of a corresponding set of cationic,

cyclometalated iridium(III) complexes. The electronic and redox properties of the

complexes were investigated using experimental (including time-resolved

luminescence and transient absorption spectroscopy) and theoretical methods. The

complexes display absorption and phosphorescent emission in the visible region

attributed to MLCT transitions. The different substitution patterns of the ligands

induce variations in these parameters. TD-DFT studies support these assignments

and show that there is likely to be a strong spin-forbidden contribution to the visible

absorption bands at 500-600 nm. Calculation also reliably predict the magnitude and

trends in triplet emitting wavelengths for the series of complexes. The complexes

were assessed as potential sensitizers in triplet-triplet annihilation upconversion

experiments using 9,10-diphenylanthracene as the acceptor, with the methylated

variants performing especially well with impressive upconversion quantum yields up

to 39.3 %.

ꢀ

1

Chemistry - A European Journal

10.1002/chem.201801007

Introduction

Photoactive metal coordination compounds continue to find broadening application in

several disciplines such as electroluminescence, photovoltaics, photocatalysis, and

bioimaging. In particular, organometallic Ir(III) complexes are extremely attractive as

a wide range of ligand architectures exist and can be developed to allow tuning of the

1

excited state properties of such complexes. A more recent development in their

2

potential application is in the field of triplet-triplet annihilation (TTA) upconversion

3

where such species, through a limited number of studies, appear to be well suited to

use as sensitizers which stimulate fluorescence from an appropriate annihilator

acceptor molecule. TTA upconversion is of great interest due to the benefits that such

processes can provide to the disciplines outlined earlier. A small number of reports

4

have described the use of Ir(III) complexes for TTA upconversion with the best

2 6

performing to date being pyrene-conjugated species of the type [Ir(ppy) (L)]PF which

demonstrated highly efficient upconversion quantum yields of up to 31.6 % (the

5

highest value yet reported). Typically, the sensitizer should possess good molar

6

absorption at the wavelength of excitation and a long triplet lifetime.

Our own work into luminescent Ir(III) complexes has included the development

of low energy emitting species that luminesce in the red part of the visible spectrum.

The requisite cyclometalating ligands are based upon core ligand structures of 2-

7

8

phenylquinoline or 2-phenylquinoxaline and these can provide interesting species

with related pyrene derivatives also showing capability as potent photooxidation

9

sensitizers. Other studies have reported extending the emission wavelengths of Ir(III)

10

complexes into the NIR region.

In the current work, we have focused upon the development of poly-substituted

quinoxaline ligands to tune the emission of a new series of cationic cyclometalated

Ir(III) complexes. The ability to finely tune the excited state character and energy of

the complex is essential when considering applications such as TTA upconversion.

The structural and spectroscopic characterisation of these complexes together with

detailed theoretical analysis has provided further insight into the species and their

application to TTA upconversion. We now report a world-leading TTA upconversion

efficiency of 39.3% from one of the iridium complexes within our series of newly

synthesized phosphors.

ꢀ

2

Chemistry - A European Journal

10.1002/chem.201801007

Results and Discussion

All ligands (Scheme 1) were synthesised by heating the required phenyldiamine with

benzil or 1-phenyl-1,2-propanedione in ethanol in the presence of acetic acid. The

ligands were reacted with iridium trichloride to generate the chloro-bridged dimetallic

1

[

Ir(L)

ethoxyethanol at 120° to yield the corresponding monometallic species

Ir(L) (bpy)]PF following precipitation with NH PF (sat. aq.). These reaction

conditions are known to favour the cis-C,C and trans-N,N coordination mode for the

2 2

-μ-Cl] species, which were subsequently reacted with 2,2’-bipyridine in 2-

[

2

6

4

6

12

cyclometalating ligand at Ir(III) and have been supported by structural data. Other

studies have shown that the mutual cis-C,C and cis-N,N arrangement of certain

13

cyclometalating ligands can be achieved using different reaction conditions. If

required, further purification was achieved using column chromatography (silica) by

eluting a major red band with DCM/MeOH (95/5). All complexes (Scheme 1) were

isolated as reddish brown air-stable solids. L1 has been previously reported, but all

characterisation data are included in the Experimental section for convenience and

14

comparison.

The complexes were characterized using multinuclear NMR, IR, UV-vis.,

transient absorption and luminescence spectroscopies, as well as HRMS, cyclic

1

voltammetry and X-ray diffraction. H NMR spectra provided clear evidence for the

formation of the complexes with characteristic shifts in the various resonances

associated with the quinoxaline ligands. In particular, the proton adjacent to the

cyclometalated carbon was shifted upfield in all cases indicative of the shielding effect

of coordination to iridium(III). For [Ir(L1)

Ir(L3) (bpy)]PF the methyl resonance(s) were shifted away from the corresponding

free ligand values. F{ H} NMR spectroscopy was used to analyse [Ir(L4)

and [Ir(L7) (bpy)]PF and revealed (Table 1) observable differences in the

2

(bpy)]PF

6

, [Ir(L2)

2

(bpy)]PF

6

and

[

2

6

1

9

1

2

(bpy)]PF

6

1

9

2

6

F

resonances in the spectra. Firstly, for the ligand-based fluorines, typical shifts were

observed around -131 ppm; these were subtly shifted upon coordination when

compared to the free ligands. A loss of chemical equivalence in the L7 fluorines was

anticipated upon cyclometalation via the expected coordination mode, and this

3

15

manifested in a JFF coupling of around 22 Hz. Both complexes showed an additional

chemical shift ca. -73 ppm which is assigned to the non-coordinating

hexafluorophosphate. Examples of obtained NMR spectra are shown in the ESI

ꢀ

3

Chemistry - A European Journal

10.1002/chem.201801007

Figures S1-S5. The observations from the NMR data are consistent with the presence

of a single isomer of complex in each case.

<

Scheme 1>

HRMS data were obtained for each of the complexes confirming the proposed

formulations and showing the expected isotopic distributions in each case; the dichloro

species, in particular, produced distinct spectra. Supporting IR data principally

indicated the presence of the coordinated ligands in each case and the

-

1

hexafluorophosphate counter anion at ca. 840 cm .

1

9

1

Tableꢀ1.ꢀ F{ H}ꢀNMRꢀchemicalꢀshiftꢀandꢀcouplingꢀconstantꢀvaluesꢀforꢀtheꢀfluorinatedꢀligandsꢀ

andꢀcomplexes.ꢀ

ꢀ

[

a]

3

-

1

Compound

Ligand d ( JFF) / ppm

6

PF d ( JFP) / ppm

L4

-130.4 (d, 21 Hz),

-

131.2 (d, 21 Hz)

-131.7 (d, 22 Hz),

133.7(d, 22 Hz)

[

Ir(L4)

2

(bpy)]PF

6

-72.6 (d, 712 Hz)

-72.6 (d, 700 Hz)

-

L7

Ir(L7)

-129.9 (s)

[

-130.3 (d, 22 Hz),

-

132.7 (d, 22 Hz)

ꢀ

X-ray Crystallography

X-ray quality single crystals were obtained for two of the methylated species

Ir(L2) (bpy)]PF6, [Ir(L5) (bpy)]PF and the difluoro variant [Ir(L7) (bpy)]PF via a

[

2

6

2

6

vapour diffusion methodology (diethyl ether into acetonitrile solutions). Data collection

parameters are reported in ESI, Table S1, whilst key bond length/angle data are

shown in ESI, Tables S2 and S3. Each complex adopts a distorted octahedral

geometry at Ir(III) with substituted quinoxaline ligands chelating in a N^C fashion

(Figure 1). The obtained structures confirm the expected cis-C,C and trans-N,N

arrangements of the cyclometalating ligands that are retained irrespective of the

quinoxaline ligand substitution. The observation support previous assertions about the

16

stronger trans influence of the phenyl group over nitrogen heterocycle donors. Bond

ꢀ

4

Chemistry - A European Journal

10.1002/chem.201801007

lengths to Ir(III) lie in the typical range expected for such species. Additional packing

diagrams are shown in the ESI showing that intermolecular pi-pi contacts are present

between phenyl rings on the quinoxaline ligands of neighbouring complexes.

Figure 2 shows a comparison of the experimental crystal structure of

[Ir(L5)

2

(bpy)]PF

6

with the computationally optimized structure. The figure

demonstrates that the computational method chosen serves to adequately reproduce

the crystal structure (RMSD < 0.5 Å) with the majority of the discrepancy being

introduced by the position of the methyl substituent groups, which likely derives from

a combination of errors in crystallography and DFT. This is to be expected given the

low frequency vibrational motions associated with both the flexing and torsional

motions of these methyl groups. Some discrepancy is observed with the bipyridine

2

position, such that the computed geometry is closer to C symmetry.

<

Figure 1>

Figure 2>

Electronic properties of the complexes

The UV-vis. absorption properties of the complexes were determined using chloroform

solutions. Between 260-400 nm the complexes show a composite of high probability

transitions that overlap with one another. The bands are generally ascribed to the

different ligand-centred transitions relating to the coordinated quinoxaline and

bipyridine ligands. They are likely to comprise of various p®p* transitions with the

possibility of weaker n®p* arising from the heterocyclic quinoxaline core. The spectra

show a strong feature at 400-450 nm, and a weaker, broad, lower energy band

3

peaking at 474-498 nm (the associated molar absorption coefficients are ca. 5 ´ 10

-

1

-1

M cm ); transitions which are likely to possess some MLCT character. The

positioning of these bands are clearly dependent upon the substitution of the

cyclometalated quinoxaline ligand: the halogenated complexes show the longest

wavelength values whilst the methylated analogues are, in contrast, hypsochomically

shifted (Figure 3). All visible region features showed a weaker lower energy shoulder

that extended to ca. 600 nm in these complexes. For related iridium(III) complexes

this observation has been previously attributed to the transition to the spin forbidden

ꢀ

5

Chemistry - A European Journal

10.1002/chem.201801007

3

MLCT state. Further discussion of the nature of these electronic transitions is

provided below.

The electrochemical characteristics of the complexes were studied in de-

oxygenated dichloromethane. The cyclic voltammograms were measured using a

-

1

-3

platinum disc electrode (scan rate υ = 200 mV s , 1 × 10 M solutions, 0.1 M

[

NBu ][PF ] as a supporting electrolyte). Each complex showed an irreversible

4

6

3

+ 4+

oxidation process between +1.4 and +1.6 V, which is assigned to a Ir / process.

The methyl-substituted complexes of L1, L2 and L5 possessed the lowest Eox values

consistent with the anticipated electron donating ability of the quinoxaline ligands. An

irreversible signal around -0.9 V was also observed and is assigned to a ligand-centred

reduction.

<

Figure 3>

Density functional theory

Figure S8 (ESI) illustrates the Kohn-Sham frontier orbitals for complex

[

Ir(L1)

geometry on the S

MO) have strong Ir(5d) character, whereas the unoccupied orbitals are more ligand

centred. This is confirmed through MO decomposition analysis (see Table S4, ESI,

2

(bpy)]PF

6

in chloroform. The orbitals are shown at the minimum energy

0

surface. The orbitals illustrate that the occupied molecular orbitals

(

17

GaussSum Package ): the highest occupied molecular orbital (HOMO) has equal MO

contributions from the metal (33%) and the two identical quinoxaline ligands (33% and

3

3%), with a negligible contribution from the bipyridine (1%). In contrast, the lowest

unoccupied molecular orbital (LUMO) is predominantly centred on the quinoxalines

48% and 47%). All complexes exhibit similar orbital contributions, however [Ir(L5-

L7) (bpy)]PF have slightly reduced metal contributions to the HOMO, between 28%

(

2

6

and 30%. The contributions from the quinoxaline ligands are necessarily near

degenerate given the symmetric nature of the system. These orbital descriptions

afford further analysis of the excited states of this system. TD-DFT calculations

suggest the character of all the singlet excited states (that lie in a region of interest,

250 nm < l < 750 nm) are reasonably mixed, however most of the occupied MOs have

sizeable Ir(5d) contributions, the HOMO-1 being the notable exception, and all the

unoccupied

orbitals

ligand

centred,

with

ꢀ

6

Chemistry - A European Journal

10.1002/chem.201801007

<

5% metal contributions, therefore each state will possess a significant degree of

1

MLCT character. The longest wavelength singlet excitation is predicted to be at l =

05 nm, and does not account for the broad structureless feature observed in the

absorption spectrum with lmax ≈ 475 nm, typically attributed to the formally spin

forbidden T ¬ S , which may become weakly allowed due to spin-orbit effects. This

is in reasonable agreement with the comparison between the energy of the T state at

ground state minimum, which corresponds to a predicted

4

1

0

1

the geometry of the S

0

3

MLCT band at l = 514 nm. These transitions all compare qualitatively favourably

(bpy)]PF , as shown in Figure 4.

, affords an examination of the spin-

with the observed UV-vis. spectrum of [Ir(L1)

Optimization of triplet [Ir(L1) (bpy)]PF

2

6

2

6

forbidden emission bands of the complex. This procedure leads to a significant

underestimation of the energy of the spin forbidden band: the computed vertical

transition occurs at 719 nm, with the experimental observation of this band (lmax ≈ 630

nm). This observation could be attributed to the evolution of the system from an initially

prepared singlet photoexcited state into a higher lying triplet state from which emission

occurs (TD-DFT calculations suggest, for example, that there are three additional

1

triplet states within 0.3 eV of T within the Franck-Condon region). However, Kasha’s

3

rule would suggest that MLCT emission occurs after photoexcitation at the S

0

minimum energy geometry; whereupon it is assumed the ISC, IC and IVR processes

occur rapidly in the excited states, such that phosphorescence occurs from the relaxed

1 1

T geometry. The T geometry relaxation primarily involves additional buckling of a

single quinoxaline ligand, and a more subtle change in the bipyridine.

Despite the relatively subtle changes in the triplet state geometry, these complexes

exhibit a range of low frequency vibrational modes, leading to a very broad Franck-

Condon envelope for these electronic transitions. A cursory investigation of the

2 6

emission band profile of [Ir(L1) (bpy)]PF at 298 K, utilizing the Franck-Condon

18

procedures in the Gaussian 09 suite, exclusive of Herzberg-Teller interactions,

shows that indeed the emission profile will be broad, and spans the wavelength region

over which phosphorescence is observed, leading to the conclusion that emission is

1

indeed observed from T . Such a methodology is quantitatively problematic for system

of this size and with such a number of low frequency vibrational modes, but provides

a qualitative depiction of the complex emission. As shown in Figure S9 (ESI), this

simulation provides good agreement with the experimentally recorded emission band,

ꢀ

7

Chemistry - A European Journal

10.1002/chem.201801007

reasonably well reproducing both the lmax value, and the band contour. From the

simulation it can be extracted that the band profile is predominantly formed by a

progression in mixed C-C stretch/C-C-C bend modes localized on the quinoxaline

-

1

ligands (~1400 cm ). This is in good agreement with the electron density change

3

induced by the MLCT emission. These vibronic features are broadened by multiple

combinations of low frequency modes. In addition, the calculations illustrate the

adiabatic transition values are a more appropriate model of emission band positions

(

Table 2). These data illustrate that the dichloro –complexes exhibit substantial

bathochromic shifts, in good agreement with experiment.

<

Figure 4>

A more complete investigation of the vibronic band shapes and Franck-Condon effects

was performed on the S /T ¬ S computed spectral features for [Ir(L1) (bpy)]PF

Photoabsorption cross sections were computed by sampling over the ground state

) geometries accessible at room temperature using a nuclear ensemble method,

1

0

2

6

.

(S

0

1

9,20

implemented in the Newton-X computational suite.

) were sampled from an uncoupled harmonic oscillator Wigner distribution such

that they describe a ground state (S ) quantum distribution, with excitation energies

0n) and oscillator strengths (f0n) for the first ten singlet states (N ) and first 5 triplet

states, computed at each geometry (R ), and then summed (with a d = 0.2 eV

In this case, 2000 geometries

(N

s

0

(

E

n

k

Lorentzian convolution [red]) to construct the photoabsorption cross section (s(E)).

This methodology can be used to provide band shapes and relative cross

sections for electronic transitions, but unlike the Franck-Condon method described

above, does not reproduce vibronic progressions. It is worth noting the refractive index

for all calculations is assumed to be 1. This assessment of band shapes is extremely

2 6

computationally costly and has only been applied to [Ir(L1) (bpy)]PF , in order to

demonstrate the accuracy of the band assignments.

ꢒ

ꢐ_ꢓ

ꢐ

ꢑ

ꢂꢃ^ꢄꢅ

1

ꢌ

ꢀ

ꢁ =

ꢍ ꢎ ꢕ ꢁ − ꢁ ꢎ , ꢖ

ꢉꢌ ꢏ ꢉꢋ ꢌ

2

ꢆꢇꢈ_ꢉ_ꢌ_ꢔ_ꢉ

ꢊ

ꢋ

ꢏ

ꢀ

8

Chemistry - A European Journal

10.1002/chem.201801007

Tableꢀ2.ꢀComputedꢀspectralꢀpropertiesꢀforꢀallꢀcomplexesꢀalongsideꢀexperimentallyꢀobservedꢀ

bandꢀpositionsꢀinꢀparentheses.ꢀTheꢀexperimentalꢀspinꢀallowedꢀabsorptionꢀbandꢀpositionsꢀareꢀ

takenꢀfromꢀtheꢀbandꢀonsets,ꢀwhereasꢀtheꢀspinꢀforbiddenꢀparametersꢀareꢀlmaxꢀvaluesꢀforꢀtheꢀ

respectiveꢀbands.ꢀShadedꢀrowsꢀhighlightꢀL5-L7ꢀphenylꢀsubstitutedꢀanaloguesꢀofꢀL2-L4.ꢀ

Bothꢀverticalꢀandꢀadiabaticꢀemissionꢀenergiesꢀareꢀreported,ꢀtheꢀlatterꢀinꢀitalics.ꢀ

ꢀ

[

a]

[a]

[b]

Compound

S

1

←S

0

(nm)

T

1

←S

0

(nm)

1 0

T → S (nm)

[

Ir(L1)

Ir(L2)

Ir(L3)

Ir(L4)

Ir(L5)

Ir(L6)

Ir(L7)

2

(bpy)]PF

6

405 (372)

402 (402)

419 (406)

409 (394)

406 (417)

423 (428)

414 (418)

541 (477)

533 (474)

565 (500)

552 (480)

540 (479)

572 (501)

555 (483)

691, 607 (627)

698, 605 (617)

708, 630 (634)

692, 612 (624)

723, 618 (624)

730, 642 (645)

716, 626 (632)

[

a] Recorded in chloroform. [b] In chloroform, lex = 355 nm

Given the remarkable agreement between the absorption spectrum of

Ir(L1) (bpy)]PF and the computational spectral simulation, the relative band

positions of all the complexes have been computed, and are displayed in Table 2.

[

2

6

<

Figure 5>

Tableꢀ3.ꢀEmissionꢀdataꢀofꢀtheꢀcomplexesꢀrecordedꢀinꢀchloroform.ꢀShadedꢀrowsꢀhighlightꢀL5-L7ꢀ

phenylꢀsubstitutedꢀanaloguesꢀofꢀL2-L4.ꢀ

ꢀ

-

1

-1

Compound

l

nm

em

/

τ /

µs

a

k

´10

r

/ s

knr / s₆

a,b

a,c

Φ

em

5

´10

[

Ir(L1)

Ir(L2)

Ir(L3)

Ir(L4)

Ir(L5)

Ir(L6)

Ir(L7)

2

(bpy)]PF

6

627

0.46

0.45

0.34

0.40

0.44

0.31

0.54

5.7

5.1

1.8

6.0

6.6

6.4

7.8

1.2

2.1

617

634

624

645

632

1.1

2.1

2.9

2.4

2.1

3.0

1.1

0.32

1.5

2.1

(bpy)]PF

1.4

a

b

c

d

In chloroform; In chloroform, lex = 480 nm; Luminescence lifetime in chloroform, lex = 459 nm; Quantum yield with [Ru(bpy)

3

][PF

6

]

2

as standard

21

(

Δ

Φ = 0.018 in acetonitrile).

The complexes showed emission properties in the visible region (Figure 5). Steady

state measurements in both chloroform (Table 3) and toluene (Table 4) confirmed that

the complexes emit with a broad feature at 550-750 nm. The coordinated quinoxaline

ligands modulate the emission energy of the complexes with the trimethylated species

ꢀ

9

Chemistry - A European Journal

10.1002/chem.201801007

2 6

[Ir(L2) (bpy)]PF revealing the highest energy emission, and the dichloro- analogues

displaying the most bathochromic shift within the series. The shift to longer wavelength

upon chlorination is a result of changing electronic character of the system,

demonstrated clearly by the reproducibility of this trend in the T

1

←S

0

TD-DFT

calculations (Table 2), and equally borne out by the shift in S ←S

1

0

band onsets.

However, the cause of the reduced emission lifetimes for the chlorinated species

cannot be deconvoluted from a mixture of electronic and relativistic effects, and indeed

an increase in ISC rates (supported by increased knr contributions) may be anticipated

for the dichloro- species.

In order to validate the hypothesis that chlorine substitution of the quinoxaline

ligands leads to an increase S

been computed from LR-TDDFT calculations at the T

for [Ir(L1) (bpy)]PF and [Ir(L6) (bpy)]PF6. Values are generated utilising a Breit-

Pauli SO operator based approach, recently implemented in the PySOC

computational suite. These calculations suggest that the SOC elements, < S

>, evaluated as the root of the squared sum of the m sublevels, is larger (5%, 149

than for [Ir(L1) (bpy)]PF6. This

lifetime for the chloro-substituted systems,

0

/T

1

spin orbit couplings (SOC), SOC elements have

1

minimum energy geometries

2

6

2

0

| HSO |

T

1

s

-

1

-1

cm vs 142 cm ) for the chlorinated [Ir(L6)

2

(bpy)]PF

6

2

change is characteristic of a reduced T

1

and along with the electronic shift in the potential energy surfaces, serves to explain

the photophysics of the molecule.

Experimental time-resolved luminescence lifetime measurements showed

monoexponential decay character in each case, consistent with a single emitting state.

These observed lifetimes confirmed the phosphorescent nature of the emission for all

complexes, and corresponding degassed measurements indicated sensitivity to

3

quenching via O

2

, with lifetimes typically extending into the microsecond domain.

Thus, all the photophysical data are consistent with an emitting state that possesses

3

significant MLCT character.

Transient absorption spectroscopy

Transient absorption (TA) spectroscopy was carried out on each of the complexes

using chloroform solutions. All the spectra are similar in appearance, and once again

2 6

complex [Ir(L1) (bpy)]PF is chosen as an example for further discussion. The

spectrum shown in Figure 6 is illustrative of TA spectra for all of the complexes

described herein. From short to long wavelength, the spectrum shows a ground state

ꢀ

10ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

bleach (negative DOD) at l = 355 nm, two features corresponding to putative triplet-

triplet absorptions in the visible (lmax ≈ 430 nm and 560 nm, respectively) and finally

a long wavelength emission attributed to the spin-forbidden T

transition. The two triplet absorption bands are consistent with TD-DFT calculations

which suggests that there are a set of strong T ¬T absorption bands at l < 600 nm.

1 0

→ S radiative

n

1

The longest wavelength, negative-going peak is assigned based on the similarity

between the TA feature (black) and the emission profile (red).

<

Figure 6>

Each feature, including the ground state bleach and recovery, and the

phosphorescence, exhibit similar TA lifetimes (right of figure 6), suggesting that each

peak relates to the same photoexcitation, intersystem crossing (ISC) process and

deactivation. This is attributed to the formation of the lowest triplet state, via prompt

S

1

/T

emission measurements. Figure 7 displays a comparison of transient absorption

6

spectra for the complexes [Ir(L1) (bpy)]PF , [Ir(L5) (bpy)]PF and [Ir(L6) (bpy)]PF .

1

ISC and compare relatively well to the observed lifetime from time-resolved

22

2

6

2

6

2

Each spectrum shows qualitatively similar features, with the exception of the ground

state bleach <450 nm, which appears to be a clear doublet in the case of

[Ir(L5)

2

(bpy)]PF

6

and [Ir(L6)

2 6

(bpy)]PF , but does not show the same spectral shape

for [Ir(L1)

2

(bpy)]PF

6

. For each of the complexes the temporal evolution (Fig. 8) of the

four transient absorption features described in Figure 7 have been analysed in an

analogous fashion and are shown in Table S5 (ESI), alongside the spin-forbidden

emission lifetimes, duplicated from Table 2.

<

Figure 7>

Figure 8>

TTA upconversion measurements

TTA upconversion luminescence experiments were conducted in degassed toluene

using the complexes as the donor component and 9,10-diphenylanthracene (DPA) as

the acceptor. The spectra in Figure 9 show the recorded emission spectra in degassed

ꢀ

11ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

toluene for these upconversion experiments. In each case the graphs contain

superimposed emission profiles for the native complex and the complex/DPA mixture

following excitation at 510 nm. The DPA triplet excited state is at 700 nm (1.77 eV)

and therefore lies below the triplet emitting level of all of the complexes in the series.

Direct irradiation of DPA using 510 nm does not produce any emission. In contrast,

for the majority of the complex/DPA mixtures, fluorescence from DPA was observed

at 400-500 nm which is therefore indicative of an upconversion process. This was

most pronounced for [Ir(L1)

display

2

(bpy)]PF

6

, [Ir(L2)

2

6 2 6

(bpy)]PF and [Ir(L5) (bpy)]PF which

3

a

concomitant quenching of the

MLCT emission band at

6

00-700 nm.

<

Figure 9>

These observations were quantitatively supported by the measured quantum yields

for upconversion (Table 4) which showed significant variation across the series of

complexes. Interestingly, the methylated variants performed best, amongst which

[

Ir(L1)

9%, respectively. The latter is the highest recorded value for TTA upconversion with

a triplet sensitiser.

2 6 2 6

(bpy)]PF and [Ir(L2) (bpy)]PF displayed remarkable efficiencies of 26% and

3

<

Figure 10>

In comparison, the upconversion performance was contrastingly low (£ 1%) for both

dichloro-species [Ir(L3) (bpy)]PF and [Ir(L6) (bpy)]PF . Previous reports in the

literature have highlighted high molar absorption coefficients in the visible region and

2

6

2

6

23

long triplet excited state lifetimes as advantageous attributes for triplet sensitizers.

It is notable that these complexes possess modest triplet state lifetimes and molar

-

1

-1

absorption values around 5000 M cm in the visible region at ca. 500 nm. The

luminescence data in toluene revealed a larger variation in emission properties with

the methylated variants displaying the highest emission energies and the longest

triplet state lifetimes. The poorly performing dichloro- derivatives show the longest

wavelength emission maxima and shortest triplet lifetimes. The difluoro-substituted

ꢀ

12ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

complexes perform slightly better than the dichloro analogues, probably reflecting the

longer triplet lifetime values.

Table 4. Emission and upconversion data of the complexes recorded in toluene.

Shaded rows highlight L5-L7 phenyl substituted analogues of L2-L4.

b

c

Compound

l

nm

em

/

t

obs / µs

t

obs / µs

UC

Φ /

d

%

a,b

[

Ir(L1)

Ir(L2)

Ir(L3)

Ir(L4)

Ir(L5)

Ir(L6)

Ir(L7)

2

(bpy)]PF

6

618

0.41

0.38

0.39

0.37

0.35

0.38

2.2

2.3

0.8

1.8

2.0

1.8

1.3

25.9

39.3

0.1

609

656

638

617

660

646

2.0

9.6

1.0

4.0

[

a]

−5

[b]

[c]

[d]

In toluene (1.0 × 10 M); Luminescence lifetime in air, lex = 510 nm; Luminescence lifetime in deaerated toluene, lex = 510 nm; TTA

F

upconversion quantum yield (ΦUC) with diiodo-Bodipy as standard (Φ = 0.027 in acetonitrile).

The visual representation of the upconversion was photographically recorded for the

best performing methylated complexes and is shown in Figure 10. The variation in

visual appearance can be plotted using CIE coordinates (Fig. 11) and conveniently

demonstrates the tuneability of the system. The observed red emission of the parent

triplet sensitizers is dramatically shifted upon addition of DPA to give new CIE

coordinates.

<

Figure 11>

Conclusions

The use of substituted quinoxaline ligands as cyclometalating units for iridium(III) has

proven to be a highly efficient route towards the development of high performance

sensitizers for triplet-triplet annihilation upconversion. Methylation of the coordinated

quinoxaline heterocycle increases the energy and the lifetime of the triplet state

emission and thus enhances performance. In contrast, whilst chloro-substitution

bathochromically shifts the absorption and emission profiles, the dramatic reduction in

triplet state lifetime for the complex of L3 proves unfavourable with respect to TTA

upconversion efficiency. We attribute this, with supporting calculations, to the

ꢀ

13ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

enhancement in ISC assisted by the chlorine substituents that leads to more rapid

3

non-radiative deactivation of the MLCT state.

Experimental Section

X-ray crystallography

2 2 6 2 6

Suitable crystals for [Ir(L2) (bpy)]PF6, [Ir(L5) (bpy)]PF and [Ir(L7) (bpy)]PF were

obtained and mounted on a MITIGEN holder in perfluoroether oil on a Rigaku FRE+

equipped with either VHF Varimax confocal mirrors and an AFC12 goniometer and

HyPix 6000 detector diffractometer ([Ir(L5)

2 6

(bpy)]PF ) or HF Varimax confocal mirrors

and an AFC12 goniometer and HG Saturn 724+ detector diffractometer

24

(

[Ir(L2)

2

(bpy)]PF

6

and [Ir(L7)

2

(bpy)]PF

6

). The crystal was kept at T = 100(2) K

25

26

during data collection. Using Olex2 the structure was solved with the ShelXT

structure solution program, using the Intrinsic Phasing solution method. The model

27

was refined with version 2014/7 of ShelXL using Least Squares minimisation. All

non-H atoms were refined anisotropically and difference Fourier syntheses were

employed in positioning idealized hydrogen atoms and were allowed to ride on their

parent C-atoms. For sample [Ir(L5)

assumed to be either MeCN, ether or a mixture) that could not be suitably modelled.

As such solvent masking within Olex2 was applied. Due to the quality of the crystals

2 6

(bpy)]PF there was highly disordered solvent

(

2 6

for sample [Ir(L7) (bpy)]PF a significant number of restraints (RIGU, SADI, BUMP)

were required.

Computational methods

Electronic structure calculations were all performed using density fitted-density

vi

functional theory within the Gaussian 09 computational chemistry suite. All

calculations were performed using the Stuttgart-Dresden (SDD) effective core

potential and basis set in the treatment of the iridium, in combination with a 6-31G*

basis set for all other light atoms. Full geometry optimizations were performed for the

cationic complexes utilizing the self-consistent reaction field model (SCRF) which

treats the solvent implicitly as a dielectric continuum. In all cases the solvent chosen

was chloroform, consistent with that utilized in the both final synthesis and in the

ꢀ

14ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

majority of the spectroscopic measurements. Chloroform is characterized by an

electrical permittivity of ε = 4.7113 within the calculations. This computational method

models the solvent as surrounding a cavity in which the solute resides, and this cavity

is characterized using an integral equation formalism for the polarizable continuum

model (IEFPCM). This model represents the system in equilibrium during, for example,

an optimization routine: in all excited state calculations a non-equilibrium solvent

model is used.

All geometry optimizations were performed using an ultrafine grid and very tight

convergence criteria, and the minima were confirmed as stationary points through the

computation of harmonic vibrational frequencies, each of which showed no imaginary

components. These stationary points were used in single point TD-DFT calculations

to compute vertical excitation energies. All TD-DFT calculations were undertaken

using a linear response approach. All TD-DFT calculations were also performed with

a long range corrected hybrid functional (CAM-B3LYP).

Phosphorescence and spin-forbidden absorption bands were investigated using

unrestricted density functional theory to compute parameters associated with the first

1

triplet state (T ), using an identical methodology as for the singlet states.

Decomposition of the molecular orbital character was performed using the GaussSum

software package. Crystal structure overlays with optimised computational structures

has been performed using the Chimera software package, which has also been used

to calculate root mean squared deviation (RMSD) values for these comparative

28

structures.

Transient absorption measurements

Transient absorption measurements were carried out using an Edinburgh Instruments

LP920 spectrometer. All spectra were collected using a pump wavelength of 355 nm

(

third harmonic of a Continuum Surelite II Nd:YAG laser system). The probe light for

these measurements was a Xenon lamp, affording spectral generation between 300

l < 800 nm. Wavelength dependent spectra were recorded with a 2.05 nm spectral

<

resolution, collected using an Andor ICCD camera, and integrated over the first 500

ns after the pump laser pulse. The spectra are presented as DODXe lamp, which is simply

referred to as DOD. Lifetime data was generated using a photomultiplier to collect time

ꢀ

15ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

resolved signals, with the bandwidth of these data being identical to the camera

resolution (2.05 nm). The lifetime data is fit using the Origin 2017 software package,

and each data set is fit using a monoexponential function, with no evidence of

multiexponential components. Uncertainties in lifetimes are taken from the Least-

Squares fitting algorithm, and are not indicative of the uncertainties in multiple fits or

data sets.

Cyclic voltammetry

Electrochemical studies were carried out using a Parstat 2273 potentiostat in

conjunction with a three-electrode cell. The auxiliary electrode was a platinum wire

and the working electrode a platinum (1.0 mm diameter) disc. The reference was a

silver wire separated from the test solution by a fine porosity frit and an agar bridge

-

3

-3

saturated with KCl. Solutions (10 ml CH

2

Cl

] as the supporting electrolyte. Under these

], added to the test

Unless specified, all electrochemical

2

) were 1.0 × 10 mol dm in the test

-

3

n

4 6

compound and 0.1 mol dm in [NBu ][PF

0

conditions, E ′ for the one-electron oxidation of [Fe(η-C

5

H

5

)

2

[

36]

solutions as an internal calibrant, is +0.46 V.

-

1

values are at υ = 200 mV s .

Triplet-Triplet Annihilation Upconversion

Associated luminescence spectra were recorded on Shimadzu RF-5301PC

spectrofluorometer. The fluorescence and phosphorescence lifetimes were measured

on an OB920 fluorescence/phosphorescence lifetime instrument (Edinburgh, U.K.)

with an EPL picosecond pulsed diode laser (510 nm ± 10 nm, pulse width: 119.9 ps,

maximum average power: 5 mW; All compounds in flash photolysis experiments were

2

deaerated with N for ca. 10 min and the gas was maintained during the measurement.

Continuous laser (510 nm) was used for upconversion and the power of the laser

beam was 5.2 mW. The diameter of the spot of the 510 nm laser was ca. 3 mm. The

mixed solution (with different triplet sensitizers and acceptor) was deaerated for 10

min before experiment, and the gas flow was kept during the measurement. The

upconverted fluorescence was recorded with a RF 5301PC spectrofluorometer. In

order to repress the laser scattering, a small black box was put behind the fluorescent

cuvette as beam dump to trap the laser.

ꢀ

16ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

The upconversion quantum yields (FUC) of all the complexes in toluene were using the

fluorescence quantum yield of diiodoBodipy (F = 2.7% in acetonitrile) as the standard

to be determined. The upconversion quantum yield was using the following equation

Eq. 2) to calculate, where F, A, I and h represent the quantum yield, absorbance,

F

(

integrated photoluminescence intensity and the refractive index, respectively.

Symbols with ‘std’ and 'sam' are the corresponding parameters for the standard used

in the measurement of quantum yield and samples to be measured.

2

-

Astd

æ 1-10

öæ I öæh

ö

÷

ø

sam

^ΦUC = 2Φstd

ç

-A_s_a_m÷ç

÷ç

1

-10

^Istd

^hstd

è

øè

(

Eq. 2)

Synthesis of 2-methyl-3-phenylquinoxaline, L1

-phenyl-1,2-propanedione (2.0 mL, 15 mmol) and 1,2-diaminobenzene (1.60 g, 15

1

mmol) were dissolved in ethanol (30 mL) with acetic acid (1 mL). The reaction mixture

was heated at reflux for 24 hours under a nitrogen atmosphere. The mixture was

cooled to room temperature and the solvent removed in vacuo. The crude product was

taken up in dichloromethane (20 mL) and washed with hydrochloric acid (0.1 M, 2 ´

2

0 mL). The organic phase was dried over MgSO

4

and dried in vacuo to yield a low

): δ 8.10 (1H, d,

1

melting-point, yellow solid (3.26 g, 81 %). H NMR (400 MHz; CDCl

3

H

J

HH = 8.37 Hz), 8.05 (1H, d, JHH = 8.37 Hz), 7.67-7.75 (3H, m), 7.65 (2H, d, JHH = 7.15

1

3

1

Hz), 7.44-7.54 (3H, m), 2.77 (3H, s, Me), 2.51 (3H, s, Me) ppm; C{ H} NMR (75 MHz,

CDCl ): d 155.7, 153.3, 142.0, 141.7, 139.8, 130.5, 130.0, 129.8, 129.7, 129.3, 129.1,

5.2 ppm. LR MS found m/z 220.1072, calcd m/z 220.1073 for C15 . UV-vis.

) lmax (ε / dm mol cm ): 325 (9400) nm. IR (solid) u / cm : 3061, 3032, 2961,

952, 1813, 1686, 1611, 1578, 1557, 1508, 1495, 1483, 1443, 1431, 1395, 1375,

3

C

2

12 2

H N

3

-1

(CHCl

3

1

8

341, 1248, 1217, 1188, 1132, 1117, 1074, 1030, 1005, 993, 974, 950, 921, 897, 868,

18, 797, 708, 679, 619, 608, 575, 559, 496, 467, 436, 409, 401.

Synthesis of L2

As L1 but with 1-phenyl-1,2-propanedione (246 mg, 1.7 mmol) and 1,2-diamino-4,5-

dimethylbenzene (250 mg, 1.8 mmol). Product collected as a low melting-point, brown

1

solid. (361 mg, 86 %). H NMR (300 MHz, CDCl

3

): d

H

7.85 (s, 1H), 7.80 (s, 1H), 7.64

–

7.62 (m, 2H), 7.53 – 7.44 (m, 3H), 2.74 (s, 3H, Me), 2.50 (s, 3H, Me), 2.48 (s, 3H)

ꢀ

17ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

1

3

1

ppm; C{ H} NMR (101 MHz, CDCl

3 C

): d 153.92, 151.29, 140.27, 139.97, 139.62,

1

39.32, 128.97, 128.76, 128.49, 128.28, 127.33, 24.20, 20.44, 20.34 ppm. HRMS

found m/z 249.1385, calcd m/z 249.1386 for C17

H

16

N

2

. UV-vis. (CHCl

3

) lmax (ε /

3

-1

dm mol cm ): 339 (11200), 269 (11400), 262 (21800) nm. IR (solid) u / cm : 3060,

3

1

5

030, 2961, 1654, 1483, 1445, 1398, 1373, 1337, 1252, 1217, 1204, 1157, 1123,

076, 1024, 1003, 988, 920, 876, 870, 858, 785, 768, 739, 706, 696, 644, 629, 610,

59, 532, 494, 478, 440, 420, 403.

Synthesis of L3

As L1 but with 1-phenyl-1,2-propanedione (190 mg, 1.3 mmol) and 1,2-diamino-4,5-

dichlorobenzene (250 mg, 1.4 mmol). Upon cooling to room temperature, a white

precipitate formed and was collected by filtration. The precipitate was washed with

1

methanol to yield the product as a white solid. (283 mg, 73 %). H NMR (300 MHz,

CDCl

3

): d

H

8.22 (s, 1H), 8.17 (s, 1H), 7.66 – 7.62 (m, 2H), 7.58 – 7.52 (m, 3H), 2.77

1

3

1

(

s, 3H) ppm; C{ H} NMR (75 MHz, CDCl

3

): δ

C

155.92, 154.01, 139.97, 139.77,

138.27, 133.69, 129.47, 129.12, 128.91, 128.69, 109.99, 24.54 ppm. HRMS found m/z

3

-1

-

291.0268, calcd m/z 291.0264 for C15

H

10Cl

2

N

2

3

. UV-vis. (CHCl ) lmax (ε / dm mol cm

1

-1

)

: 342 (12800), 268 (30700) nm. IR (solid) u / cm : 3088, 1753, 1697, 1587, 1543,

491, 1441, 1412, 1389, 1371, 1325, 1246, 1209, 1180, 1169, 1107, 1078, 1022,

005, 993, 976, 955, 930, 897, 878, 845, 795, 768, 748, 706, 658, 635, 629, 613, 594,

50, 509, 490, 461, 428, 417.

1

5

Synthesis of L4

As for L1 but with 1-phenyl-1,2-propanedione (230 mg, 1.6 mmol) and 1,2-diamino-

4

,5-difluorobenzene (250 mg, 1.7 mmol). Upon cooling to room temperature, a white

precipitate formed and was collected by filtration and washed with methanol. Product

1

collected as a white solid. (225 mg, 55 %). H NMR (300 MHz, CDCl

3

H

): d 7.87 – 7.77

1

3

1

(

m, 2H), 7.65 – 7.62 (m, 2H), 7.56 – 7.49 (m, 3H), 2.76 (s, 3H) ppm; C{ H} NMR (101

1

9

1

3 C

MHz, CDCl ): d 138.44, 129.29, 128.88, 128.66, 114.78, 114.03, 24.29 ppm; F { H}

3

NMR (376 MHz, CDCl

3 F

): δ -130.38 (d, JFF = 21.2 Hz), -131.17 (d, JFF = 21.2 Hz)

ppm. HRMS found m/z 257.0888, calcd m/z 257.0885. UV-vis. (CHCl

3

) lmax (ε /

3

-1

dm mol cm ): 326 (13000) nm. IR (solid) u / cm : 3030, 1630, 1572, 1553, 1497,

ꢀ

18ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

1

8

450, 1373, 1356, 1339, 1256, 1227, 1200, 1142, 1078, 1015, 1005, 988, 928, 897,

74, 866, 791, 772, 752, 712, 706, 667, 619, 611, 584, 544, 484, 447, 419, 405.

Synthesis of L5

Benzil (357 mg, 1.7 mmol) and 1,2-diamino-4,5-dimethylbenzene (250 mg, 1.8 mmol)

were dissolved in ethanol (15 mL) and acetic acid (1 mL). The reaction mixture was

heated at reflux under a nitrogen atmosphere for 24 hours. The mixture was then

cooled to room temperature and a white precipitate was collected by filtration and

1

washed with methanol. (413 mg, 78 %). H NMR (300 MHz, CDCl

3

): d

H

7.92 (s, 2H),

1

3

1

7

.51 – 7.48 (m, 4H), 7.35 – 7.31 (m, 6H), 2.49 (s, 6H, Me) ppm; C{ H} NMR (75 MHz,

CDCl ): d 152.50, 140.55, 140.21, 139.37, 129.84, 128.53, 128.21, 109.99, 20.50

ppm. HRMS found m/z 311.1542, calcd m/z 311.1543 for C22 . UV-vis. (CHCl

max (ε / dm mol cm ): 356 (13900), 281 (24600), 269 (31800), 254 (44900) nm. IR

3

C

H

18

N

2

3

)

3

-1

l

-

1

(

solid) u / cm : 2974, 1749, 1549, 1531, 1493, 1474, 1460, 1445, 1416, 1400, 1375,

1

8

346, 1335, 1263, 1211, 1179, 1153, 1074, 1059, 1022, 1003, 966, 932, 870, 849,

14, 783, 773, 762, 741, 725, 691, 633, 608, 598, 556, 530, 519, 492, 476, 436, 413.

Synthesis of L6

As L5 but with benzil (273 mg, 1.3 mmol) and 1,2-diamino-4,5-dichlorobenzene (250

1

mg, 1.4 mmol). Product collected as a white solid. (367 mg, 80 %). H NMR (300 MHz,

1

3

1

CDCl

101 MHz, CDCl

HRMS found m/z 351.0450, calcd m/z 351.0450 for C20

3

): d

H

8.29 (s, 2H), 7.52 – 7.49 (m, 4H), 7.37 – 7.35 (m, 6H) ppm; C{ H} NMR

(

3

): d 154.49, 139.95, 138.39, 134.43, 129.80, 129.29, 128.37 ppm.

C

H

12Cl

2

N

2

3

. UV-vis. (CHCl ) lmax

3

-1

(

ε / dm mol cm ): 362 (19400), 254 (61300) nm. IR (solid) u / cm : 3067, 3024, 2980,

1

9

4

589, 1535, 1491, 1452, 1439, 1393, 1337, 1254, 1190, 1109, 1074, 1061, 1020, 999,

64, 920, 883, 874, 831, 814, 766, 733, 719, 692, 640, 621, 606, 598, 546, 511, 488,

80, 444, 426, 419, 409.

Synthesis of L7

As L5, but with benzil (336 mg, 1.6 mmol) and 1,2-diamino-4,5-difluorobenzene (250

1

mg, 1.7 mmol). Product collected as an orange solid. (296 mg, 58 %). H NMR (300

MHz, CDCl

3

): d

H

7.91 (app. td, J = 1.37, 9.35 Hz, 2H), 7.51 – 7.48 (m, 4H), 7.41 – 7.31

ꢀ

19ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

1

3

1

(

m, 6H) ppm; C{ H} NMR (75 MHz, CDCl

50.66, 138.51, 138.48, 129.77, 129.13, 128.37, 114.72 ppm; F{ H} NMR (376 MHz,

CDCl ): δ -129.86 ppm. HRMS found m/z 319.1044, calcd m/z 319.1041 for

3 C

): d 154.31, 154.07, 153.69, 150.89,

1

9

1

3

F

3

-1

C

20

H

12

F

2

N

2

. UV-vis. (CHCl

3

) lmax (ε / dm mol cm ): 343 (14900), 261 (16800) nm. IR

-

1

(

solid) u / cm : 3051, 1597, 1568, 1541, 1456, 1435, 1352, 1342, 1246, 1217, 1194,

1

7

175, 1152, 1142, 1082, 1072, 1055, 1022, 1001, 972, 939, 918, 872, 818, 785, 772,

60, 752, 719, 700, 677, 623, 610, 573, 542, 521, 498, 438, 424, 419.

Complex Synthesis

IrCl .xH O (1 eq.) and ligand, L (2 eq.) were dissolved in 2-ethoxyethanol (10 mL) and

3

2

the reaction mixture heated at reflux for 48 hours. The reaction was then cooled to

room temperature and water (30 mL) was added to form a dark brown precipitate. The

solid was collected by filtration and assumed to yield [Ir(L)

further purification.

2 2

μ-Cl] and used without

2 6

Synthesis of [Ir(L1) (bpy)]PF

[

2 2

Ir(L1) μ-Cl] (100 mg, 0.075 mmol) and 2,2¢-bipyridyl (0.025 g, 0.16 mmol) were

dissolved in 2-ethoxyethanol (10 mL) and heated at reflux for 24 hours under a

nitrogen atmosphere. The reaction mixture was then cooled to room temperature and

a saturated solution of aqueous ammonium hexafluorophosphate was added. Upon

formation of a red precipitate, the mixture was filtered and the precipitate washed with

1

water and diethyl ether to yield the product as a red solid. (0.08g, 68 %). H NMR

(

300MHz, CDCl

2H, d, JHH = 5.32 Hz), 8.01 (2H, app.t, JHH = 7.86 Hz), 7.90 (2H, d, JHH = 8.31 Hz),

.45-7.57 (4H, m), 7.16-7.24 (2H, m), 7.00 (2H, app.t, JHH = 7.69 Hz), 6.86 (2H, app.

3 H

): δ 8.39 (2H, d, JHH = 8.31 Hz), 8.24 (2H, d, JHH = 8.31 Hz), 8.17

(

7

1

3

1

t, JHH = 7.69 Hz), 6.61 (2H, d, JHH = 7.60 Hz), 3.36 (6H, s, Me) ppm; C{ H} NMR

400MHz, CDCl ): δ 163.7, 155.1, 152.6, 152.0, 146.6, 144.0, 140.4, 140.0, 139.7,

35.1, 130.9, 130.5, 130.4, 130.1, 129.2, 127.6, 124.8, 123.6, 123.2, 27.5 ppm. HRMS

(

3

C

1

found m/z 787.2148, calcd m/z 787.2158 for C40

H

30IrN

6

. UV-vis. (CHCl

3

): λmax/nm (ε /

3

-1

dm mol cm ): 477 (2500), 372 (13200), 253 (27900) nm. IR (solid) u / cm : 1605,

1

8

578, 1530, 1449, 1427, 1387, 1348, 1261, 1215, 1196, 1165, 1130, 1016, 1001, 897,

37, 770, 750, 731, 704, 660, 627, 592, 557, 459, 420, 415, 405.

ꢀ

20ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

2 6

Synthesis of [Ir(L2) (bpy)]PF

1

Product collected as a red solid (133 mg, 94 %). H NMR: (400 MHz, Acetone-d

6

): δ

.57 – 8.54 (2 H, m), 8.49 (2 H, dd, JHH = 8.3, 1.2 Hz), 8.41 (2 H, app. dt, JHH = 8.2,

.0Hz), 8.16 (2 H, ddd, JHH = 8.3, 7.6, 1.6 Hz), 7.86 – 7.81 (2 H, m), 7.66 (2 H, s), 7.24

H

8

1

(2 H, ddd, JHH = 8.3, 7.1, 1.3 Hz), 7.17 (2 H, s), 6.86 – 6.79 (2 H, m), 6.70 (2 H, dd,

1

3

1

J

HH = 7.7, 1.3 Hz), 3.34 (6 H, s, Me), 2.30 (6 H, s, Me), 1.81 (6 H, s, Me) ppm; C { H}

(101 MHz, Acetone-d

6

): δ 164.56, 156.84, 154.00, 153.38, 149.62, 146.39, 142.69,

C

1

8

42.14, 141.53, 140.70, 140.05, 136.48, 132.09, 131.63, 130.00, 129.66, 125.61,

25.24, 124.33, 28.18, 20.36, 20.11 ppm. HRMS found m/z 843.2783 calcd m/z

3

-1

6 3

43.2784 for C44H38IrN . UV-vis (CHCl ) lmax (ε / dm mol cm ): 474 (4800), 391

-

1

(

22100), 376 (23900), 309 (19300), 390 (32300), 268 (47200) nm. IR (solid) u / cm :

1

601, 1582, 1560, 1526, 1483, 1445, 1396, 1375, 1342, 1323, 1267, 1219, 1171,

134, 1063, 993, 835, 795, 768, 737, 702, 660, 627, 556, 474, 434, 420, 403.

2 6

Synthesis of [Ir(L3) (bpy)]PF

As [Ir(L1)

2

(bpy)]PF

6

but with [Ir(L3)

2

Cl]

2

(100 mg, 0.06 mmol) and 2,2¢-bipyridine (20

1

mg, 0.1 mmol). Product collected as a red solid (61 mg, 46 %). H NMR (400 MHz,

Acetone-d

.26 – 8.18 (2 H, m), 8.13 (2 H, d, JHH = 1.2), 7.92 – 7.83 (2 H, m), 7.59 (2 H, s), 7.31

2 H, dd, JHH = 8.3, 6.7), 6.97 – 6.89 (2 H, m), 6.86 (1 H, s), 6.84 (1 H, s), 3.41 (6 H, s)

6 H

): δ 8.60 (2 H, d, JHH = 8.4), 8.51 (3 H, dd, JHH = 5.6, 3.6), 8.48 (1 H, s),

8

(

1

3

1

ppm. C { H} NMR (126 MHz, DMSO): δ

C

165.56, 155.14, 154.80, 153.75, 147.29,

43.91, 140.79, 138.80, 138.42, 135.28, 133.16, 132.60, 132.14, 131.53, 129.82,

29.31, 125.05, 124.96, 123.41, 27.34 ppm. HRMS found m/z 925.0548, calculated

1

3

-1

4 6 3

m/z 925.0558 for C40H26Cl IrN . UV-vis. (CHCl ) lmax (ε / dm mol cm ): 500 (4500),

-

1

3

1

8

83 (23700), 298 (28900), 266 (48200) nm. IR (solid) u / cm : 1603, 1576, 1528,

464, 1447, 1381, 1315, 1265, 1188, 1165, 1132, 1113, 1061, 1026, 1009, 962, 895,

70, 843, 824, 772, 739, 729, 673, 664, 646, 637, 608, 556, 467, 428, 419, 403.

2 6

Synthesis of [Ir(L4) (bpy)]PF

1

Product collected as a red solid (133 mg, 97 %). H NMR (400 MHz, Acetone-d

6

): δ

H

8

.58 (2 H, dd, JHH = 8.3, 1.2), 8.54 (2 H, ddd, JHH = 5.5, 1.6, 0.8), 8.47 (2 H, dt, JHH

=

.2, 1.0), 8.25 – 8.18 (2 H, m), 7.92 – 7.81 (4 H, m), 7.34 – 7.20 (4 H, m), 6.95 – 6.88

ꢀ

21ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

1

3

1

(

2 H, m), 6.84 – 6.79 (2 H, m) ppm. C { H} NMR (101 MHz, Acetone-d

6 C

): δ 155.55,

152.77, 147.77, 144.41, 140.52, 135.29, 131.51, 131.13, 128.91, 124.79, 123.34 ppm.

1

9

6 F

F NMR (376 MHz, Acetone-d ): δ -72.63 (d, J = 711.7 Hz), -131.73 (d, J = 21.9 Hz),

-

133.10 (d, J = 21.9 Hz) ppm. HRMS found m/z 859.1780, calculated m/z 859.1781

3

-1

26 4 6 3

for C40H F IrN . UV-vis. (CHCl ) lmax (ε / dm mol cm ): 480 (2400), 375 (13200),

-

1

3

1

7

11(11800), 288 (15400), 262 (25800) nm. IR (solid) u / cm : 1065, 1578, 1533, 1501,

447, 1341, 1331, 1252, 1233, 1196, 1128, 1063, 1036, 997, 878, 841, 795, 772, 741,

31, 689, 660, 638, 586, 557, 476, 451, 428, 422, 407.

2 6

Synthesis of [Ir(L5) (bpy)]PF

As [Ir(L1)

2

(bpy)]PF

6

but with [Ir(L5)

2

Cl]

2

(150 mg, 0.09 mmol) and 2,2¢-bipyridine (29

1

mg, 0.2 mmol). Product collected as a red solid (124 mg, 63%). H NMR (400 MHz,

6 H

Acetone-d ): δ 9.10 (2 H, dt, JHH = 5.1, 2.2), 8.61 – 8.52 (2 H, m), 8.34 – 8.25 (2 H,

m), 8.24 – 8.15 (2 H, m), 7.96 (4 H, dt, JHH = 6.7, 2.7), 7.78 (2 H, s), 7.74 – 7.65 (6 H,

m), 7.34 (2 H, s), 7.18 (2 H, ddd, JHH = 8.2, 2.9, 1.5), 6.78 – 6.72 (2 H, m), 6.69 – 6.63

1

3

1

(

2 H,m), 6.61 – 6.54 (2 H, m), 2.35 (6 H, s), 1.94 (6 H, s) ppm. C { H} NMR (101

MHz, Acetone-d ): δ 162.58, 156.09, 153.60, 152.81, 149.26, 144.93, 142.80,

41.53, 140.81, 140.42, 139.91, 139.43, 135.23, 131.40, 130.46, 130.06, 129.58,

29.44, 129.21, 124.84, 123.98, 122.37, 19.43, 18.99 ppm. HRMS found m/z

6

C

1

9

3

-1

67.3086, calcd m/z 967.3099 for C54

H

42IrN

6

. UV-vis. (CHCl

3

) lmax (ε / dm mol cm ):

-

479 (6500), 400 (29300), 362 (24800), 297 (49000), 269 (71300) nm. IR (solid) u / cm

1

:

1603, 1580, 1479, 1447, 1348, 1321, 1234, 1207, 1159, 1134, 1074, 1024, 1001,

9

74, 833, 810, 775, 748, 737, 729, 700, 658, 640, 608, 577, 557, 542, 446, 432, 415.

2 6

Synthesis of [Ir(L6) (bpy)]PF

As [Ir(L1)

2

(bpy)]PF

6

, but with [Ir(L6)

2

Cl]

2

(100 mg, 0.05 mmol) and 2,2¢-bipyridine (19

1

mg, 0.1 mmol). Product collected as a red solid (48 mg, 37 %). H NMR (300 MHz,

Acetone-d

6

H

): δ 8.98 (2 H, ddd, JHH = 5.5, 1.7, 0.7), 8.60 (2 H, dt, JHH = 8.2, 1.0), 8.30

(2 H, td, JHH = 7.9, 1.6), 8.23 – 8.15 (4 H, m), 7.99 – 7.90 (4 H, m), 7.69 (8 H, q, JHH

=

1

3

1

2

.2, 1.8), 7.25 – 7.15 (2 H, m), 6.81 – 6.62 (6 H, m) ppm. C { H} NMR (101 MHz,

): δ 166.36, 157.25, 156.93, 155.10, 149.90, 145.27, 142.19, 141.00,

40.74, 140.57, 136.52, 136.16, 133.44, 132.75, 131.91, 131.67, 130.77, 130.55,

Acetone-d

6

C

1

26.65, 126.43, 123.84 ppm. HRMS found m/z 1049.0839, calculated m/z 1049.0872

ꢀ

22ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

for C50

H30Cl

4

6 3

IrN . UV-vis. lmax / nm (CHCl ) 501(7300), 404 (35300), 299 (54600), 268

-

1

(

87500) nm. IR (solid) u / cm : 1603, 1576, 1524, 1493, 1445, 1433, 1406, 1383,

1

8

342, 1317, 1258, 1186, 1165, 1132, 1115, 1072, 1045, 1026, 1001, 961, 880, 839,

26, 766, 734, 698, 673, 648, 635, 606, 577, 557, 532, 517, 486, 474, 451, 434, 419.

Synthesis of [Ir(L7)

As [Ir(L1) (bpy)]PF

2

(bpy)]PF

6

2

6

but with [Ir(L7)

2

Cl]

2

(100 mg, 0.06 mmol) and 2,2’-bipyridine (19

1

mg, 0.1 mmol). Product collected as a red solid (108 mg, 83 %). H NMR (400 MHz,

Acetone-d

.4), 8.22 (2 H, m), 8.01 (6 H, tt, JHH = 8.5, 4.4), 7.79 – 7.68 (6 H, m), 7.55 (1 H, dt,

HH = 8.2, 1.5), 7.46 – 7.37 (2 H, m), 7.28 – 7.20 (2 H, m), 6.87 – 6.79 (2 H, m), 6.77

6 H

): δ 9.10 – 9.06 (2 H, m), 8.63 (2 H, d, JHH = 8.2), 8.35 (2 H, tt, JHH = 8.0,

1

J

–

d

1

6

1

3

1

6.72 (2 H, m), 6.69 (2 H, dt, JHH = 7.8, 1.3) ppm. C { H} NMR (101 MHz, Acetone-

): δ 157.50, 150.05, 142.41, 140.79, 136.71, 133.25, 132.55, 131.76, 130.75,

26.56, 124.00, 117.87 ppm. F{ H} NMR (376 MHz, Acetone-d

99.9 Hz), -130.30 (d, J = 21.7 Hz), -132.66 (d, J = 21.7 Hz) ppm. HRMS found m/z

6

C

1

9

1

6

F

): δ -72.64 (d, J =

3

-

9

30 4 6 3

83.2088, calculated m/z 983.2095 for C50H F IrN . UV-vis. (CHCl ) lmax (ε / dm mol

1

-1

cm ): 483 (4500), 396 (21600), 367 (20200), 297 (33100), 265 (48600) nm. IR (solid)

-

1

u / cm : 1603, 1578, 1503, 1447, 1429, 1335, 1275, 1260, 1223, 1204, 1163, 1126,

1

4

072, 1043, 1026, 980, 874, 835, 810, 758, 739, 700, 660, 640, 623, 615, 557, 536,

98.

Acknowledgements

Cardiff University (Knowledge Economy Skills Scholarship to KAP) and STG

Aerospace are thanked for financial support. We thank the staff of the Engineering

and Physical Sciences Research Council (EPSRC) Mass Spectrometry National

Service (Swansea University) for providing MS data and the Engineering and Physical

Sciences Research Council (EPSRC) UK National Crystallographic Service at the

University of Southampton. JMB acknowledges Dr. James A. Platts for helpful

discussions regarding the Gaussian calculations.

Keywords: Iridium complex • Spectroscopy • Upconversion • Phosphors • DFT

ꢀ

23ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

Notesꢀandꢀreferencesꢀ

‡

CCDC1825271, 1825273 and 1825272 contains supplementary X-ray

crystallographic data for [Ir(L2)

2

(bpy)]PF

6

, [Ir(L5)

2

(bpy)]PF

6

and [Ir(L7)

2

(bpy)]PF

6

respectively.

This

data

can be

obtained

free

of charge

via

http://www.ccdc.cam.ac.uk/structures/, or from the Cambridge Crystallographic Data

Centre, Union Road, Cambridge, CB2 1EZ; fax(+44) 1223-336-033 or email:

deposit@ccdc.cam.ac.uk.

CorrespondingꢀAuthors:ꢀ

Prof.ꢀSimonꢀJ.ꢀA.ꢀPope:ꢀ

E-mail:ꢀPopeSJ@cardiff.ac.ukꢀ

ꢀ

DrꢀJosephꢀM.ꢀBeames:ꢀ

E-mail:ꢀBeamesJ@cardiff.ac.ukꢀ

ꢀ

24ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

ꢀ

FiguresꢀandꢀCaptionsꢀ

ꢀ

25ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

2 6 2 6

Figure 1. X-ray structures of the complexes [Ir(L2) (bpy)]PF , [Ir(L5) (bpy)]PF and

2 6

[Ir(L7) (bpy)]PF (top to bottom). Anions and solvents of crystallisation are omitted.

Figure

2.

Overlay

of

the

crystal

structure

(blue)

and

(bpy)]PF .

DF-DFT//B3LYP/6-31G*(SDD) optimised structures (brown) for [Ir(L5)

2

6

The structures exhibit an RMSD of 0.3 Å.

6

[

Ir(L2)2(bpy)]PF6

Ir)L3)2(bpy)]PF6

Ir(L7)2(bpy)]PF6

5

4

3

2

1

0

250

300

350

400

450

500

550

600

wavelengthꢀ/ꢀnm

Figure 3. Absorption spectra of [Ir(L2)

2

(bpy)]PF

6

,

2 6

[Ir(L3) (bpy)]PF ,

2 6

[Ir(L7) (bpy)]PF (in chloroform).

ꢀ

26ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

Figure 4. Comparison of the experimental absorption spectrum of [Ir(L1)

(

2

(bpy)]PF

black) with the TDDFT//CAM-B3LYP/6-31G*(SDD) convoluted absorption spectrum

dashed red), computed the method described in the text. The red line is a summation

6

of spin allowed and spin forbidden transition energies, where all spin forbidden

transitions are assigned an identical oscillator strength, with the total summative

transition strength chosen to best illustrate the spectrum.

Figure 5. Normalized emission spectra of complexes (in chloroform,

l

ex = 355 nm).

ꢀ

27ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

Figure 6. Left: transient absorption spectrum of [Ir(L1)

2

(bpy)]PF

6

shown in black,

overlaid with the emission spectrum of the same complex shown in red. Right:

transient absorption lifetime measurements made at selected wavelengths,

highlighted as grey bars in the right hand figure. The red traces indicate

monoexponential fits to these measurements, with corresponding lifetimes displayed

in each panel. Recorded in chloroform, lex = 355 nm.

Figure 7. Transient absorption spectra of several sample complexes. The spectra

show

qualitatively

similar

features.

Recorded

in

chloroform,

l

ex = 355 nm.

ꢀ

28ꢀ

Chemistry - A European Journal

10.1002/chem.201801007

2 6

Figure 8. Sample time resolved transient absorption data for [Ir(L6) (bpy)]PF prior

to spectral deconvolution. The data is shown from t = 1800 ns for clarity purposes only.

4

3

2

1

90

40

90

40

90

40

90

40

990

890

790

690

590

490

390

290

190

90

C1

C1ꢀ+ꢀDPA

C2

C2ꢀ+ꢀDPA

ꢀ

9

0

4

0

-

10

-10

3

50

450

550

650

750

850

3

50

450

550

650

750

850

wavelengthꢀ/ꢀnm

7

6

5

4

3

2

1

90

2

90

240

1

90

40

C5

C7

C7ꢀ+ꢀDPA

1

C5ꢀ+ꢀDPA

90

40

ꢀ

90

-10

3

50

450

550

650

750

850

350

450

550

650

750

850

wavelengthꢀ/ꢀnm

ꢀ

Figure 9. Clockwise from top left: the upconversion fluorescence spectra of C1

[

Ir(L1)

Ir(L5)

2

(bpy)]PF

6

,

C2 [Ir(L2)

2

(bpy)]PF

6

,

C7 [Ir(L7)

2

(bpy)]PF

6

,

and C5

2

6

as photosensitizer in toluene. DPA (9,10-diphenylanthracene) was

the acceptor. Excitation was done with a continuous laser at 510 nm (noted as the

incident peak on the spectra) and power density of 5.2 mW under deaerated

-

5

-3

atmosphere. c (sensitizer) = 1.0 ´ 10 M, c (DPA) were 1.6 ´ 10 M, 1.6 ´ 10 M,

2

-

4

-4

.6 ´ 10 M, 2.0 ´ 10 M, respectively, 20 °C.

ꢀ

29ꢀ

Article Doi

Ligand-Tuneable, Red-Emitting Iridium(III) Complexes for Efficient Triplet–Triplet Annihilation Upconversion Performance

DOI: 10.1002/chem.201801007

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201801007

Full text of DOI:10.1002/chem.201801007

Products guided by the article

R&D Labs maybe for 10130-23-1

Relevant to this article

Hot Product