Solution-processible conjugated electrophosphorescent polymers

Article abstract of DOI:10.1021/ja039445o

We report the synthesis and photophysical study of a series of solution-processible phosphorescent iridium complexes. These comprise bis-cyclometalated iridium units [Ir(ppy)₂(acac)] or [Ir(btp) ₂-(acac)] where ppy is 2-phenylpyridin

Full text of DOI:10.1021/ja039445o

Published on Web 05/12/2004
Solution-Processible Conjugated Electrophosphorescent
Polymers
Albertus J. Sandee,† Charlotte K. Williams,† Nicholas R. Evans,† John E. Davies,†
Clare E. Boothby,‡ Anna Ko¨hler,‡ Richard H. Friend,‡ and Andrew B. Holmes*,†
Contribution from the MelVille Laboratory for Polymer Synthesis, Department of Chemistry,
UniVersity of Cambridge, Lensfield Road, Cambridge CB2 1EW, United Kingdom, and
CaVendish Laboratory, Department of Physics,
UniVersity of Cambridge, Madingley Road, Cambridge CB3 0HE, United Kingdom
Received November 5, 2003; E-mail: abh1@cam.ac.uk
Abstract: We report the synthesis and photophysical study of a series of solution-processible phospho-
rescent iridium complexes. These comprise bis-cyclometalated iridium units [Ir(ppy)2(acac)] or [Ir(btp)2-
(acac)] where ppy is 2-phenylpyridinato, btp is 2-(2′-benzo[b]thienyl)pyridinato, and acac is acetylacetonate.
The iridium units are covalently attached to and in conjugation with oligo(9,9-dioctylfluorenyl-2,7-diyl) [(FO)n]
to form complexes [Ir(ppy-(FO)n)2(acac)] or [Ir(btp-(FO)n)2(acac)], where the number of fluorene units, n, is
1, 2, 3, ∼10, ∼20, ∼30, or ∼40. All the complexes exhibit emission from a mixed triplet state in both
photoluminescence and electroluminescence, with efficient quenching of the fluorene singlet emission.
Short-chain complexes, 11-13, [Ir(ppy-(FO)n-FH)2(acac)] where n ) 0, 1, or 2, show green light emission,
red-shifted through the FO attachment by about 70 meV, but for longer chains there is quenching because
of the lower energy triplet state associated with polyfluorene. In contrast, polymer complexes 18-21 [Ir-
(btp-(FO)n)2(acac)] where n is 5-40 have better triplet energy level matching and can be used to provide
efficient red phosphorescent polymer light-emitting diodes, with a red shift due to the fluorene attachment
of about 50 meV. We contrast this small (50-70 meV) and short-range modification of the triplet energies
through extended conjugation, with the much more substantial evolution of the π-π* singlet transitions,
which saturate at about n ) 10. These covalently bound materials show improvements in efficiency over
simple blends and will form the basis of future investigations into energy-transfer processes occurring in
light-emitting diodes.
Introduction
In light-emitting diodes (LEDs), electrons and holes are
injected from opposite electrodes and combine to form spin-
singlet or spin-triplet excitons. Radiative decay from the singlets
is fast (fluorescence) whereas that from the triplets (phospho-
rescence) is formally forbidden by the requirement of spin
conservation. The inclusion of heavy metal atoms in the
molecular structure can give strong spin-orbit coupling which
renders phosphorescence partially allowed. Models of spin
statistics predict that the electron-hole recombination event
should produce three times as many triplets as singlets,5 and
this has been confirmed experimentally for electroluminescent
devices (OLEDs) fabricated from small molecules.6 For poly-
mers, there is growing evidence that the triplet-singlet ratio
can be as low as 1:17,8 and recent work on polyacetylides
containing heavy metal atoms incorporated into the backbone
has enabled a detailed study of singlet and triplet excited
states.9-12
The discovery of electroluminescence from conjugated poly-
mers has led to the development of and intense interest in the
field of polymer optoelectronics.1-3 The chemical synthesis,
photophysics, and material properties are of interest not only
as an academic curiosity but also as a commercial reality.
Luminescent conjugated polymers show real promise as the
active material in light-emitting display devices for the next
generation of information technology based consumer products.
The principle interest in the use of these polymers lies in the
scope for low-cost, large surface area manufacturing, facilitated
by solution-processing of film-forming materials.4 The wide-
spread commercialization of light-emitting polymers is depend-
ent on improving device efficiencies; this can be achieved both
by improvements in device engineering and by insights into
their fundamental chemistry and physics enabling design of
intrinsically more efficient materials.
† Melville Laboratory for Polymer Synthesis.
(5) Ko¨hler, A.; Wilson, J. S.; Friend, R. H. AdV. Mater. 2002, 14, 701-
707.
‡ Cavendish Laboratory.
(1) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay,
K.; Friend, R. H.; Burn, P. L.; Holmes, A. B. Nature 1990, 347, 539-541.
(2) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem., Int. Ed. 1998,
37, 402-428.
(3) Braun, D.; Heeger, A. J. Appl. Phys. Lett. 1991, 58, 1982-1984.
(4) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.; Marks, R.
N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Bre´das, J. L.;
Lo¨gdlund, M.; Salaneck, W. R. Nature 1999, 397, 121-128.
(6) Baldo, M. A.; O’Brien, D. F.; Thompson, M. E.; Forrest, S. R. Phys. ReV.
B 1999, 60, 14422-14428.
(7) Wilson, J. S.; Dhoot, A. S.; Seeley, A. J. A. B.; Khan, M. S.; Ko¨hler, A.;
Friend, R. H. Nature 2001, 413, 828-831.
(8) Cao, Y.; Parker, I. D.; Yu, G.; Zhang, C.; Heeger, A. J. Nature 1999, 397,
414-417.
(9) Wittmann, H. F.; Friend, R. H.; Khan, M. S.; Lewis, J. J. Chem. Phys.
1994, 101, 2693-2698.
9
10.1021/ja039445o CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 7041-7048
7041

A R T I C L E S
Sandee et al.
The energy of the non emissive triplet state in organic
compounds can be harvested by energy transfer to a phospho-
rescent dopant such as a lanthanide or heavy metal organome-
tallic complex. As early as 1990, Kido and colleagues described
an OLED employing [Tb(acac)3] as the phosphor and giving
green light emission,13 and Wittmann, et al.9 proposed the use
of platinum-containing poly-ynes both as semiconductor and
as triplet emitter. In 1998, Thompson and Forrest and their co-
workers published a seminal paper describing the use of
platinum(II) porphyrins (PtOEP) as the phosphor in an OLED
and were able to obtain higher efficiency red-light emission than
had proved possible with lanthanide dopants.14,15 Energy-transfer
processes and the sites of electron-hole recombination in blends
of PtOEP in conjugated polymer hosts have been studied in
some detail by the Cambridge16-18 and Sheffield19,20 groups.
More recently, Thompson and Forrest and colleagues demon-
strated spectacular enhancements in OLED efficiencies, with
external quantum efficiencies as high as 19%, using Ir(III)
cyclometalated complexes as blends in host luminescent
materials.21-29 However, the processing of such materials can
be complex and costly, often requiring deposition under high
vacuum and controlled temperature, and the use of multiple
layers.
phors into conjugated polymer hosts and the fabrication of LEDs
using these composite materials; however, the improvements
in device efficiency have been modest.16-20,33-40 It has been
proposed that energy is lost by transfer to low-lying triplet states
in the polymer host41 or by triplet-triplet annihilation; under
such conditions, the efficiency is limited by phase separation
and aggregation of dopants even at low-blending concentrations.
It is recognized that this could be suppressed when the
phosphorescent dopant and polymer host have similar surface
functional groups42 or when they are implemented in one
composite material.34,43,44 In a recent paper, statistical copoly-
mers based on dioctylfluorene, with hole-transporting moieties
and phosphorescent Ir(III) complexes attached as pendant groups
to the main backbone, were synthesized and showed some
improvements in device efficiency.45
We aim to attach phosphors covalently to a conjugated
polymer backbone so as to allow efficient energy transfer
between polymer and phosphor and further to minimize ag-
gregation and quenching of phosphorescence. Herein, we
describe the controlled synthesis of such oligomers and polymers
based on 9,9-dialkylfluorene repeat units in conjugation with
bis-cyclometalated iridium(III) acac complexes. The photo-
physics of optical and electrical excitation are presented and
this has enabled insights to be drawn into design criteria for
phosphorescent polymer light-emitting devices.
Spin-coatable solutions of well-defined amorphous iridium
complexes blended in polycarbazole and poly(phenylenevi-
nylene) hosts were described by Bazan and Heeger,30 with the
ultimate production of efficient, but multilayer devices.31,32
Considerable attention has been focused on blending of phos-
Results and Discussion
Synthesis and Characterization of Ir(III) Complexes. The
syntheses of either iridium complexes with 2-phenylpyridinato
(ppy) coupled to well-defined oligofluorenes of chain length
1-3 units or iridium complexes with 2-(2′-benzo[b]thienyl)-
pyridinato (btp) coupled to polyfluorenes of longer chain length
(5-40 units) are described. The syntheses presented are
controlled, high yielding, and general involve either a stepwise
building up of oligofluorenes or the application of the Suzuki
polycondensation reaction to yield well-defined polyfluorene
complexes.46,47
(10) Younus, M.; Ko¨hler, A.; Cron, S.; Chawdhury, N.; Al-Mandhary, M. R.
A.; Khan, M. S.; Lewis, J.; Long, N. J.; Friend, R. H.; Raithby, P. R. Angew.
Chem., Int. Ed. 1998, 37, 3036-3039.
(11) Wilson, J. S.; Kohler, A.; Friend, R. H.; Al-Suti, M. K.; Al-Mandhary, M.
R. A.; Khan, M. S.; Raithby, P. R. J. Chem. Phys. 2000, 113, 7627-7634.
(12) Ko¨hler, A.; Wilson, J. S.; Friend, R. H.; Al-Suti, M. K.; Khan, M. S.;
Gerhard, A.; Ba¨ssler, H. J. Chem. Phys. 2002, 116, 9457-9463.
(13) Kido, J.; Nagai, K.; Ohashi, Y. Chem. Lett. 1990, 657-660.
(14) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.;
Thompson, M. E.; Forrest, S. R. Nature 1998, 395, 151-154.
(15) Kwong, R. C.; Sibley, S.; Dubovoy, T.; Baldo, M.; Forrest, S. R.;
Thompson, M. E. Chem. Mater. 1999, 11, 3709-3713.
(16) Cleave, V.; Yahioglu, G.; Le Barny, P.; Friend, R. H.; Tessler, N. AdV.
Mater. 1999, 11, 285-288.
(17) Cleave, V.; Yahioglu, G.; Le Barny, P.; Hwang, D.-H.; Holmes, A. B.;
Friend, R. H.; Tessler, N. AdV. Mater. 2001, 13, 44-47.
(18) Morgado, J.; Cacialli, F.; Iqbal, R.; Moratti, S. C.; Holmes, A. B.; Yahioglu,
G.; Milgrom, L. R.; Friend, R. H. J. Mater. Chem. 2001, 11, 278-283.
(19) Lane, P. A.; Palilis, L. C.; O’Brien, D. F.; Giebeler, C.; Cadby, A. J.; Lidzey,
D. G.; Campbell, A. J.; Blau, W.; Bradley, D. D. C. Phys. ReV. B 2001,
63, 235206.
(33) McGehee, M. D.; Bergstedt, T.; Zhang, C.; Saab, A. P.; O’Regan, M. B.;
Bazan, G. C.; Srdanov, V. I.; Heeger, A. J. AdV. Mater. 1999, 11, 1349-
1354.
(34) Wong, C. T.; Chan, W. K. AdV. Mater. 1999, 11, 455-459.
(35) Guo, T.-F.; Chang, S.-C.; Yang, Y.; Kwong, R. C.; Thompson, M. E. Org.
Electron. 2000, 1, 15-20.
(36) Virgili, T.; Lidzey, D. G.; Bradley, D. D. C. AdV. Mater. 2000, 12, 58-
62.
(20) Campbell, A. J.; Bradley, D. D. C.; Virgili, T.; Lidzey, D. G.; Antoniadis,
H. Appl. Phys. Lett. 2001, 79, 3872-3874.
(21) Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Nature 2000, 403, 750-
753.
(37) Slooff, L. H.; Polman, A.; Cacialli, F.; Friend, R. H.; Hebbink, G. A.; Van
Weggel, F. C. J. M.; Reinhoudt, D. N. Appl. Phys. Lett. 2001, 78, 2122-
2124.
(22) Adachi, C.; Baldo, M. A.; Forrest, S. R.; Thompson, M. E. Appl. Phys.
Lett. 2000, 77, 904-906.
(23) Adachi, C.; Kwong, R. C.; Djurovich, P.; Adamovich, V.; Baldo, M. A.;
Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 2001, 79, 2082-2084.
(24) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl. Phys.
2001, 90, 5048-5051.
(38) Chen, F.-C.; Yang, Y.; Thompson, M. E.; Kido, J. Appl. Phys. Lett. 2002,
80, 2308-2310.
(39) Higgins, R. W. T.; Monkman, A. P.; Nothofer, H.-G.; Scherf, U. J. Appl.
Phys. 2002, 91, 99-105.
(25) D’Andrade, B. W.; Baldo, M. A.; Adachi, C.; Brooks, J.; Thompson, M.
E.; Forrest, S. R. Appl. Phys. Lett. 2001, 79, 1045-1047.
(26) Holmes, R. J.; Forrest, S. R.; Tung, Y. J.; Kwong, R. C.; Brown, J. J.;
Garon, S.; Thompson, M. E. Appl. Phys. Lett. 2003, 82, 2422-2424.
(27) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Lee, H.-E.;
Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Am. Chem.
Soc. 2001, 123, 4304-4312.
(40) Zhu, W. G.; Mo, Y. Q.; Yuan, M.; Yang, W.; Cao, Y. Appl. Phys. Lett.
2002, 80, 2045-2047.
(41) Sudhakar, M.; Djurovich, P. I.; Hogen-Esch, T. E.; Thompson, M. E. J.
Am. Chem. Soc. 2003, 125, 7796-7797.
(42) Lupton, J. M.; Samuel, I. D. W.; Frampton, M. J.; Beavington, R.; Burn,
P. L. AdV. Funct. Mater. 2001, 11, 287-294.
(43) Iqbal, R.; Yahioglu, G.; Milgrom, L.; Moratti, S. C.; Holmes, A. B.; Cacialli,
F.; Morgado, J.; Friend, R. H. Synth. Met. 1999, 102, 1024-1025.
(44) Morgado, J.; Cacialli, F.; Friend, R. H.; Iqbal, R.; Yahioglu, G.; Milgrom,
L. R.; Moratti, S. C.; Holmes, A. B. Chem. Phys. Lett. 2000, 325, 552-
558.
(28) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Kwong, R.;
Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E. Inorg. Chem.
2001, 40, 1704-1711.
(29) Lamansky, S.; Djurovich, P. I.; Abdel-Razzaq, F.; Garon, S.; Murphy, D.
L.; Thompson, M. E. J. Appl. Phys. 2002, 92, 1570-1575.
(30) Ostrowski, J. C.; Robinson, M. R.; Heeger, A. J.; Bazan, G. C. Chem.
Commun. 2002, 784-785.
(45) Chen, X.; Liao, J.-L.; Liang, Y.; Ahmed, M. O.; Tseng, H.-E.; Chen, S.-A.
J. Am. Chem. Soc. 2003, 125, 636-637.
(46) Rees, I. D.; Robinson, K. L.; Holmes, A. B.; Towns, C. R.; O’Dell, R.
MRS Bull. 2002, 27, 451-455.
(31) Gong, X.; Ostrowski, J. C.; Bazan, G. C.; Moses, D.; Heeger, A. J.; Liu,
M. S.; Jen, A. K.-Y. AdV. Mater. 2003, 15, 45-49.
(47) Towns, C. R.; O’Dell, R. Manufacture of conjugated polymers using boron
derivatives. Cambridge Display Technology PCT Int. Appl. 2000
WO200053656-A1 20000914; Chem. Abstr. 2000, 133, 238529.
(32) Gong, X.; Ostrowski, J. C.; Moses, D.; Bazan, G. C.; Heeger, A. J. AdV.
Funct. Mater. 2003, 13, 439-444.
9
7042 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

Conjugated Electrophosphorescent Polymers
A R T I C L E S
Scheme 1. Synthetic Route to Well-Defined Fluorene Substituted ppyH Ligands 4-9
Reagents and conditions: (a) 2, Pd(PPh3)4, Et4NOH, toluene, 90 °C, 47%; (b) 3, Pd(PPh3)4, Et4NOH, toluene, 90 °C, 88%; (c) ICl, CH2Cl2, 25 °C, 85%;
(d) 2, Pd(PPh3)4, Et4NOH, toluene, 110 °C, 86%; (e) 3, Pd(PPh3)4, Et4NOH, toluene, 90 °C, 88%; ICl, CH2Cl2, 25 °C, 85%; (f) 2, Pd(PPh3)4, Et4NOH,
toluene, 90 °C, 52%.
Oligomers. The synthetic strategy for the oligomers, an
adaptation of a method used to synthesize oligophenylenes,48,49
is shown in Scheme 1. 2-(4′-Bromophenyl)pyridine 1 was used
in sequential Suzuki coupling reactions with orthogonally
protected fluorene monomers, followed by deprotection and
further coupling reactions to give ppyH coupled with one 4
(ppyH-FH), two 7 (ppyH-FO-FH), and three 9 (ppyH-FO-
FO-FH) fluorenyl substituents, respectively (where FH is 9,9-
dihexylfluorene and FO is 9,9-dioctylfluorene).
A Suzuki coupling of 2-(4′-bromophenyl)pyridine 150 with
2-(4′,4′,5′,5′-tetramethyl-1′,3′,2′-dioxaboran-2′-yl)-9,9-dihexyl-
fluorene 251 gave 2-(4′-(9′′,9′′-dihexylfluoren-2′′-yl)phenyl)-
pyridine 4, resulting in the coupling of the ppyH ligand to one
fluorene unit. The key monomeric building block of the stepwise
synthesis, 2-(4′,4′,5′,5′-tetramethyl-1′,3′,2′-dioxaborolan-2′-yl)-
7-trimethylsilyl-9,9-dioctylfluorene 3, was made in two steps
from 2,7-dibromo-9,9-dioctylfluorene via a controlled mono-
lithiation using 1 equivalent of n-BuLi and subsequent reaction
of the lithio-derivative with trimethylsilyl chloride (TMSCl),
followed by bromine-lithium exchange using t-BuLi and
boronation with 2-isopropoxy-4,4,5,5 tetramethyl-1,3,2-dioxa-
borolane. A Suzuki coupling reaction between 2-(4′-bromophen-
yl)pyridine 1 and the trimethylsilyl protected fluorenyl borolane
3 gave the product 5 where ppyH has been coupled with a
trimethylsilyl substituted fluorenyl unit; this was iodo-desilylated
with iodine monochloride (ICl) to give the iodo-substituted
derivative 6. A second Suzuki couping reaction of the iodo-
substituted derivative 6 with the fluorenyl borolane 2 gave 2-(4′-
(7′′-(9′′′,9′′′-dihexylfluoren-2′′′-yl)-9′′,9′′-dioctylfluoren-2′′-yl)-
phenyl)pyridine 7, the second ligand target in which a ppyH
unit is coupled to two fluorene moieties. The final target, 9,
was synthesized by a Suzuki coupling of the iodofluorenyl
substituted ligand 6 and the trimethylsilyl protected fluorenyl
borolane 3, followed by iodo-desilylation to afford the iodo-
derivative 8. A Suzuki coupling of the iodo-derivative 8 with
the fluorenyl borolane 2 gave 2-(4′-(7′′-(7′′′-(9′′′′,9′′′′-dihexyl-
fluoren-2′′′′-yl)-9′′′,9′′′-dioctylfluoren-2′′′-yl)-9′′,9′′ -dioctylfluo-
ren-2′′-yl)phenyl)pyridine 9. The new ligands were characterized
by 1H NMR spectroscopy; the signal due to the pyridyl proton
at δ 8.74 could be integrated and compared with those in the
fluorenyl units at δ 7.85-7.70 thus allowing the number of
fluorene units attached to the ligand to be quantified.
The dibrominated bis-cyclometalated iridium complex 10 was
synthesized by a modification of the route that was recently
reported by Lamansky, et al.27 The chloride bridged dimeric
complexes, from the reaction of IrCl3‚xH2O with excess ligand,
were cleaved by reaction with acetyl acetone to yield monomeric
acetylacetonate (acac) complexes 10-13 as shown in Scheme
1
2. The H NMR spectra showed that the peaks due to the ppy
ligand are more widely spaced compared with the free ligand
and in general these shifted to lower field on increasing the
number of fluorene units. On the basis of the similarity of their
1H NMR spectra to crystallographically characterized com-
pounds,27,28 the complexes are assigned an octahedral coordina-
tion environment with the two nitrogen atoms of the cyclo-
metalating ligands being trans to one another and the two carbon
atoms being cis.27
(48) Henze, O.; Lehmann, U.; Schlu¨ter, A. D. Synthesis 1999, 4, 683-687.
(49) Hensel, V.; Schlu¨ter, A. D. Chem. Eur. J. 1999, 5, 421-429.
(50) Abramovitch, R. A.; Saha, J. G. J. Chem. Soc. 1964, 2175-2187.
(51) Ranger, M.; Leclerc, M. Can. J. Chem. 1998, 76, 1571-1577.
Polymers. Ligands with polyfluorene chains attached to the
ppyH group were synthesized using the Suzuki polycondensation
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 7043

A R T I C L E S
Sandee et al.
Scheme 2. Synthesis of 10-13
Reagents and conditions: (a) IrCl3‚xH2O, 2-ethoxyethanol (aq), reflux, 60-66%; (b) acetyl acetone, Na2CO3, 2-ethoxyethanol, reflux, 50-60%.
Scheme 3. Synthetic Route to Polyfluorene Triplet Emitters 14-21
Reagents and conditions: (a) (Pd(OAc)2, 2 PCy3, Et4NOH, toluene, reflux, 24 h, 50-60%.
protocol developed in Cambridge for poly(9,9-dioctylfluorene)
(PFO)46,47 and terminating the reaction with 2-(4′-bromophenyl)-
pyridine 1. However, the iridium coordination chemistry of these
species was frustrated by gelation of the reaction mixture or
formation of highly insoluble products, leading to incomplete
metal chelation and therefore to lower phosphor loadings than
targeted. A simpler route, shown in Scheme 3, was developed
using Suzuki homo-polymerization of 2-(4′,4′,5′,5′-tetramethyl-
1′,3′,2′-dioxaborolan-2′-yl)-7-bromo-9,9-dioctylfluorene 16, an
AB monomer, and chain extension with the bromo-substituted
cyclometalated iridium complex 10 to give, after reaction with
excess acetyl acetone, polymer complexes composed of 10
fluorene units, (14) [Ir(ppy-(FO)10)2(acac)], and 30 fluorene
units, (15) [Ir(ppy-(FO)30)2(acac)]. The relative chain lengths
signal assigned to the acac methyl groups at δ 1.90, with the
first CH2 groups of the alkyl chains on the 9-position in the
fluorene at δ 2.30-2.00.
The final series of compounds synthesized were Ir(III)-
chelating polyfluorenes for red-light emission. The same
synthetic strategy was applied as shown in Scheme 3; using a
bromo-substituted 2-(2′-benzo[b]thienyl)pyridine (btpH) cyclo-
metalating reagent, Ir(III) complexes of closely analogous
ligands have previously shown red phosphorescence.27 The
modified btpH ligand was prepared via a Suzuki coupling of
2,5-dibromopyridine with 2-(4′,4′,5′,5′-tetramethyl-1′,3′,2′-di-
oxaborolan-2′-yl)benzo[b]thiophene. The ligand was complexed
with IrCl3‚xH2O yielding first a chloride-bridged dimer, which
was then cleaved by reaction with acetyl acetone to give iridium-
(III)bis(2-(2′-benzo[b]thienyl)-5-bromopyridinato-N, C3′)(acetyl-
1
were calculated by comparing the integration of the H NMR
9
7044 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

Conjugated Electrophosphorescent Polymers
A R T I C L E S
acetonate) 17 [Ir(btp-Br)2(acac)]. Crystals suitable for an X-ray
structure determination were grown and a representation of the
structure is shown in the Supporting Information (Figure S1).
The iridium atom has octahedral geometry with the pyridyl
nitrogen atoms occupying a trans disposition and the benzo[b]-
thienyl carbon atoms being cis to one another. There are some
distortions from ideal octahedral geometry exemplified by the
largest bond angle being between the two carbon atoms [92.1-
(8)°] and the smallest being between carbon and nitrogen atoms
on the same ligand [78.9(8)°]. The bond lengths and angles are
within the range found for other such iridium complexes.27,52-57
The polymers, 18-21, were synthesized by a Suzuki poly-
merization reaction of 2-(4′,4′,5′,5′-tetramethyl-1′,3′,2′-dioxa-
borolan-2′-yl)-7-bromo-9,9-dioctylfluorene 16 as the monomer
and in the presence of bromo-substituted bis-cyclometalated
iridium complex 17. Chain lengths from 5 to 40 units were
obtained by varying the reaction stoichiometry and were
calculated by integration of signals due to the acac methyl
groups at δ 1.92, and the first CH2 groups of the alkyl chains
1
on the nine-position in the fluorene at δ 2.30-2.00, in the H
NMR spectra. Gel permeation chromatography was run in
chloroform against polystyrene standards and confirmed the high
molar masses expected. Molecular weights higher than predicted
were found, probably because of the rigid rod nature of the
polymers in solution, and showed polydispersities between 2
and 3, as predicted for condensation polymerizations.
Figure 1. (a) The thin-film room temperature absorption spectra of
compounds 10-15. 10 [Ir(ppy-Br)2(acac)] dashed line with no markers,
11 [Ir(ppy-FH)2(acac)] has filled squares, 12 [Ir(ppy-FO-FH)2(acac)] has
open circles, 13 [Ir(ppy-(FO)2-FH)2(acac)] has filled triangles, 14 [Ir(ppy-
(FO)10)2(acac)] (same spectrum as 15 [Ir(ppy-(FO)30)2(acac)]) has crosses,
and PFO has a solid line with no markers. (b) The thin-film room
temperature absorption spectra of compounds 17-21, 17 [Ir(btp-Br)2(acac)]
dashed line with no markers, 18 [Ir(btp-(FO)5)2(acac)] has filled squares,
19 [Ir(btp-(FO)10)2(acac)] has open circles, 20 [Ir(btp-(FO)20)2(acac)] has
filled triangles, 21 [Ir(btp-(FO)40)2(acac)] has crosses, and PFO has a solid
line with no markers.
Optical Spectroscopy
Absorption. The thin-film absorption spectra of the series
of ppy substituted iridium complexes, 10-15, and btp substituted
iridium complexes, 17-21, are shown in Figure 1. Complex
10 [Ir(ppy-Br)2(acac)] shows several low-intensity metal to
ligand charge transfer (MLCT) transitions from 2.50 to 3.60
eV; the lowest energy transitions being the MLCT d-π*(ppy)
3
1
excitations, with MLCT triplet states at 2.50 eV and MLCT
singlet states at 2.65 eV. The high-intensity peak centered at
4.70 eV is assigned to the πfπ* transition of the ppy
ligands.27,28,58 Increasing the conjugation of the cyclometalating
ligand passing from 10 [Ir(ppy-Br)2(acac)] to 17 [Ir(btp-Br)2-
(acac)] causes a red shift in the πfπ* peak intensity of 0.60
eV and a diminution in peak intensity. The lower energy of the
btp transition may lead to stronger mixing with the iridium
d-orbitals, and this in turn may account for the increased
intensity of the MLCT transitions from 2.40 to 3.60 eV in 17
compared to 10. In contrast to the πfπ* transition, the low-
energy MLCT transitions shift only by 0.2 eV between the two
compounds and this reflects their more localized character.
When complex 10 is coupled to a single fluorene unit to give
11 [Ir(ppy-(FH))2(acac)], the πfπ* peak at 4.70 eV shifts to
3.65 eV, consistent with the extended conjugation length of the
ligand. As the number of fluorene units is increased from 1 in
complex 11 to 30 in complex 15, the main πfπ* peak shifts
further to the red and increases in oscillator strength, tending
toward the absorption spectrum of PFO. In a similar way, the
πfπ* peak at 4.10 eV in 17 [Ir(btp-Br)2(acac)] shifts to 3.20
eV in 18 [Ir(btp-(FO)5)2(acac)]. No further red shift occurs on
increasing the number of fluorene units from five in complex
18 to forty in complex 21, indicating that five fluorene units
are already sufficient to shift the πfπ* peak into that of PFO.
For both series, the intensity of the MLCT transitions decreases
in the larger oligomers and polymers because of the decreased
loading of the complex in the materials. The absorption spectra
of ppy substituted iridium complexes 11-15 and btp substituted
iridium complexes 18-21 do not correspond to the sum of the
individual absorption spectra of 10 or 17 and PFO, but instead
show a shift of the iridium complex πfπ* transition upon
coupling with fluorene units. This suggests that the energy levels
of the iridium complex and the short fluorene chains cannot be
separated but that the whole oligomer forms a single conjugated
entity with a common π-π* singlet state.
(52) Dedeian, K.; Djurovich, P. I.; Garces, F. O.; Carlson, G.; Watts, R. J. Inorg.
Chem. 1991, 30, 1685-1687.
Photoluminescence. The photoluminescence (PL) spectra of
the series taken from thin films at 5 K are shown in Figure 2a
and b, respectively, along with the PL spectra of blends of [Ir-
(ppy)2(acac)] or [Ir(btp)2(acac)] and PFO.
The PL spectrum for complex 10 [Ir(ppy-Br)2(acac)] is
characterized by a 0-0 peak at 2.28 eV and broad vibronic
(53) Grushin, V. V.; Herron, N.; Le Cloux, D. D.; Marshall, W. J.; Petrov, V.
A.; Wang, Y. Chem. Commun. 2001, 1494-1495.
(54) Colombo, M. G.; Gu¨del, H. U. Inorg. Chem. 1993, 32, 3081-3087.
(55) Colombo, M. G.; Gu¨del, H. U. Inorg. Chem. 1993, 32, 3088-3092.
(56) Colombo, M. G.; Brunold, T. C.; Riedener, T.; Gu¨del, H. U.; Fo¨rtsch, M.;
Bu¨rgi, H.-B. Inorg. Chem. 1994, 33, 545-550.
(57) Lo, K. K.-W.; Chung, C.-K.; Zhu, N. Chem Eur. J. 2003, 9, 475-483.
(58) Hay, P. J. J. Phys. Chem. A 2002, 106, 1634.
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 7045

A R T I C L E S
Sandee et al.
Table 1. EL and PL Efficiencies in Thin Films at Room
Temperature for Compounds 10-15, Blend A, Blend B, and PFO
total PLQY (%)
molar %
total
triplet
singlet
1%
triplet
singlet
#
of Ir complex ELQY (%) ELQY(%) ELQY (%)
in PS PLQY (%) PLQY (%)
10
11
12
13
14
15
A
100.0
33.3
20.0
14.3
4.8
1.6
4.8
1.6
0
0.012
0.030
0.110
0.115
0.045
0.070
0.025
0.055
0.250
0.012
0.030
0.110
0.115
0.034
0.048
0.023
0.006
0.2 40.0
0.8
0.8 26.0
0.5
3.0
0.2
0.8
0.8
0.5
2.3
0.2
0.011
0.022
0.002
0.049
0.7
1.8
2.2
7.2
50.0
2.0
2.2
7.2
4.0
B
PFO
0.250 50.0
Table 2. EL and PL Efficiencies in Thin Films at Room
Temperature for Compounds 17-21, Blend C, Blend D, and PFO
total
ELQY
(%)
triplet
ELQY
(%)
singlet
ELQY
(%)
total
PLQY
(%)
triplet
PLQY
(%)
singlet
PLQY
(%)
molar % of
Ir complex
material
17
18
19
20
21
C
100.0
9.1
4.8
2.4
1.2
2.4
1.2
0
0.002
0.150
0.200
0.450
1.500
0.70
0.002
0.150
0.200
0.450
1.500
0.686
0.480
0.3
9.9
0.3
9.9
12.1
21.8
26.6
25.5
23.0
50.0
11.7
20.7
23.1
24.0
20.7
0.4
0.9
3.5
1.5
2.3
0.014
0.020
0.250
D
PFO
0.50
0.250
50.0
Figure 2. (a) The thin-film photoluminescence spectra at 5 K of compounds
10 [Ir(ppy-Br)2(acac)], 11 [Ir(ppy-(FH))2(acac)], 12 [Ir(ppy-FO-FH)2(acac)],
13 [Ir(ppy-(FO)2-FH)2(acac)], 14 [Ir(ppy-(FO)10)2(acac)], 15 [Ir(ppy-
(FO)30)2(acac)], blend A (same molecular percentage of [Ir(ppy)2(acac)]
in PFO as for 14), and blend B (same molecular percentage of [Ir(ppy)2-
(acac)] in PFO as for 15). Spectra are normalized and offset along the
vertical axis for ease of comparison. The dotted line indicates the position
of the triplet state in PFO polymer at 2.1 eV and trimer at 2.3 eV. (b) The
thin-film photoluminescence spectra at 5 K of compounds 17 [Ir(btp-Br)2-
(acac)], 18 [Ir(btp-(FO)5(acac)], 19 [Ir(btp-(FO)10)2(acac)], 20 [Ir(btp-
(FO)20)2(acac)], 21 [Ir(btp-(FO)40)2(acac)], blend C (same molecular
percentage of [Ir(btp)2(acac)] in PFO as for 20), and blend D (same
molecular percentage of [Ir(btp)2(acac)] in PFO as for 21). Spectra are
normalized and offset along the vertical axis for ease of comparison. The
dotted line indicates the position of the triplet state in PFO polymer at 2.1
eV.
and D). Polyfluorene fluorescence is also strongly quenched in
the blends (Tables 1 and 2) as there is good spectral overlap
between the emission of PFO and the MLCT absorptions of
[Ir(ppy)2(acac)] and this allows for efficient Fo¨rster transfer from
PFO to the complexes.
In contrast, in a conjugatively linked system, a mixed triplet
state may be formed between the iridium complex and the
polyfluorene, as both triplet states are energetically similar. We
would therefore expect phosphorescence to occur in both series,
and this is indeed observed. On the coupling of a fluorene unit
to 10 to give complex 11 [Ir(ppy-(FH))2(acac)], the emission
band shifts from 2.28 to 2.21 eV, and shows a well-resolved,
narrower vibronic structure. The increase in structure indicates
that the emission is dominated by a πfπ* transition and that
the fluorene units are indeed electronically coupled to the
complex. This emission displays strong temperature dependence
in thin films, is quenched by the presence of oxygen, and is
energetically placed near the triplet states of both 10 and
polyfluorene; therefore, it is attributable to a triplet excited state.
The energy shifts from 10 to 11 and 12 are to lower energies,
even though the triplet state of polyfluorene trimer is at higher
energy. This confirms the mixed nature of the state and also
suggests the triplet wave function is sensitive to oligomer length
despite being a localized state. No further shift is observed on
increasing the number of fluorene units from 2 in the oligomer
12 to 30 in the polymer 15, even though the triplet level in
polyfluorene is at lower energy. This suggests that phospho-
rescence occurs from the first few fluorene units nearest the
iridium and that more distant units have little effect on the
emission maximum. In a similar way, the triplet emission shifts
by 0.05 eV to lower energies when a fluorene-pentamer is
coupled to 17, even though the triplet state in polyfluorene itself
is at higher energy than in 17.
structure, while in the PL spectrum of 17 [Ir(btp-Br)2(acac)],
the 0-0 peak is at 1.95 eV with a narrower and better resolved
vibronic structure. The small Stokes shift between absorption
and emission maxima for 10 indicates emission from a mixed
MLCT/πfπ* state whose character is predominately that of
the MLCT transition, as observed for the related complex [Ir-
(ppy)2(acac)].28 The PL emission maximum for 17, on the other
hand, has a larger Stokes shift, inconsistent with a localized
charge-transfer transition. This transition therefore has a strong
πfπ* character, as has been observed for the related complex
[Ir(btp)2(acac)] by Lamansky et al.27
The triplet states of a fluorene trimer and polymer are reported
at about 2.30 and 2.10 eV, respectively.59 Therefore, while the
triplet state of 10 is equal to or slightly higher than that of the
polyfluorene oligomers, the triplet state in 17 is at lower energy.
Consequently, Figure 2 and Tables 1 and 2 show how thin-
film phosphorescence is efficiently quenched for [Ir(ppy)2(acac)]
blended in PFO (Table 1, entries A and B), while this is not
the case for [Ir(btp)2(acac)] blended in PFO (Table 2, entries C
We have been able to follow the evolution of the energy of
the triplet state centered on the Ir complex as a function of the
(59) Hertel, D.; Setayesh, S.; Nothofer, H.-G.; Scherf, U.; Mu¨llen, K.; Ba¨ssler,
H. AdV. Mater. 2001, 13, 65-70.
9
7046 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

Conjugated Electrophosphorescent Polymers
A R T I C L E S
In addition, the PL quantum yield of 15 is quenched by a factor
of 10 upon raising the temperature. For the longer oligomers,
polyfluorene segments far away from the iridium site may have
an electronic structure close to that of polyfluorene polymer
itself, with a non-emissive triplet energy level below that of
the mixed state. It is therefore possible that thermally activated
exciton diffusion may occur to such electronically decoupled
polyfluorene units where phosphorescence is quenched in the
same way as has been observed in the blends of the [Ir(ppy)2-
(acac)] with PFO. A comparison of the PL spectrum of 14 with
the corresponding blend A shows that the blend spectrum is
dominated by fluorescence, while 14 is dominated by phos-
phorescence. When the polyfluorene chain is extended in 15,
both the conjugated compound and the corresponding blend B
are dominated by fluorescence. This indicates that 10 fluorene
units correspond to the extent of conjugation before electronic
decoupling occurs, in agreement with measurements.61
For the series of compounds coupled to complex 17, the triplet
energy level is below that of the polyfluorene oligomers and
so the ratio of triplet to singlet state emission is hardly affected
by temperature. In addition, we observe similar fluorescence
and phosphorescence quantum yields for the blends C and D
as for the corresponding compounds 20 and 21 (in contrast to
the EL quantum yields, as detailed further below).
Electroluminescence. The electroluminescence (EL) spectra
of the compounds (Figure 4) were measured at room temperature
in a device structure of ITO/PEDOT:PSS/compound/Ca/Al as
described in the Supporting Information. For the short-chain
complexes of ppy ligands 10-13, all of the emission occurs
from the triplet state. Longer chain complexes 14 and 15, and
the blend with high iridium complex loading, A, show mostly
triplet emission, but a small peak due to polyfluorene singlet
emission is also observed. The blend with low iridium complex
loading, B, on the other hand gives almost exclusively singlet
emission. This demonstrates the potential for these polymer
complexes to phosphoresce even at iridium loadings which lead
to quenching of such emission in blended materials. The series
of complexes with fluorene units bound to btp ligands, 17-21,
all show exclusive triplet emission in their EL spectra pointing
to charge trapping at iridium being the dominant process under
EL excitation.
Efficiencies. The photoluminescence and electroluminescence
efficiencies (PLQY and ELQY respectively) of the series linked
to 10 are low (Table 1) for several reasons. For the short
oligomers, concentration quenching occurs in the neat films.
The PL efficiencies of 10 and 12, for example, increased from
0.2% to 40% and from 0.8% to 26% when blended in a
polystyrene (PS) matrix at 1% concentration. For longer
oligomers, concentration quenching should be less significant
but triplet efficiencies are now low because there is triplet energy
transfer onto FO segments. This study of the energy transfer in
the solid state shows that the triplet energy reduces with
increasing length of the FO segment, and transfer from the
iridium site to the FO chromophore then becomes an exothermic
and thus more efficient process. This was previously noted by
Thompson and co-workers, when studying the triplet transfer
in solution between iridium-containing phosphors with a range
of triplet energies and a fluorene trimer.41 Thus, for the long
Figure 3. (a) The thin-film photoluminescence spectra at 290 K of
compounds 10 [Ir(ppy-Br)2(acac)], 11 [Ir(ppy-(FH))2(acac)], 12 [Ir(ppy-
FO-FH)2(acac)], 13 [Ir(ppy-(FO)2-FH)2(acac)], 14 [Ir(ppy-(FO)10)2(acac)],
15 [Ir(ppy-(FO)30)2(acac)], blend A, and blend B. Spectra are normalized
and offset along the vertical axis for ease of comparison. (b) The thin-film
photoluminescence spectra at 290 K of compounds 17 [Ir(btp-Br)2(acac)],
18 [Ir(btp-(FO)5)2(acac)], 19 [Ir(btp-(FO)10)2(acac)], 20 [Ir(btp-(FO)20)2-
(acac)], 21 [Ir(btp-(FO)40)2(acac)], blend C, and blend D. Spectra are
normalized and offset along the vertical axis for ease of comparison.
extension of conjugation of the polyfluorene chain and find that
this is sensitive to only the first attached fluorene group. This
indicates that this triplet state is much more strongly localized
than is the case for singlet states on conjugated chains, as can
be seen here by the considerable downward evolution of the
π-π* singlet transition energies with chain length (Figure 1)
saturating at about 3.2 eV for n ) 10.
Singlet emission is also observed at 2.90 eV for the materials
with the longest fluorene chains, with the energy and the very
sharply resolved vibronic structure typical for polyfluorene. The
presence of singlet emission is most likely due to insufficient
singlet energy transfer from polyfluorene segments to the
iridium.
Figure 3 shows the photoluminescence spectra of thin films
of the new materials at room temperature. In both series,
inhomogeneous broadening occurs, yet the energy of the 0-0
peak in the triplet emission remains unaltered, which further
underscores the notion that phosphorescence occurs localized
on the conjugated segment involving the iridium site and
immediately adjacent ligands. The singlet emission observed
for the longer chain complexes and blends shifts to slightly
higher energies when the temperature is raised from 5 to 290
K possibly because of thermally activated exciton diffusion
between segments with different singlet energies.60
In the series of compounds where fluorene units are coupled
to complex 10, there is a higher fraction of singlet emission at
290 K than at 5 K in the longer polymer complexes 14 and 15.
(60) Ba¨ssler, H.; Schweitzer, B. Acc. Chem. Res. 1999, 32, 173-182.
(61) Scherf, U.; List, E. J. W. AdV. Mater. 2002, 14, 477-487.
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 7047

A R T I C L E S
Sandee et al.
fluorene chains with carbazole units, and variation of chain
length may enable optimization of these promising initial re-
sults.
Conclusions
Controlled, high-yielding, and general synthetic routes have
been developed to a series of well-defined oligo and polyfluo-
renyl bis-cyclometalated iridium complexes. The addition of
fluorene substituents to ppyH and btpH ligands under optimized
Suzuki coupling conditions and their coordination complexes
with Ir(III) yielded materials that are readily spin-coated from
solution as neat materials. Photophysical studies revealed that
there is mixing of the triplet levels of the fluorene and
cyclometalating ligand to give a hybrid mixed triplet state as
the lowest energy level. Photoluminescence emission occurs
from this mixed triplet state and its wavelength may be tuned
from green to red by judicious choice of cyclometalating ligand
and fluorene chain length. The EL spectra are dominated by
triplet emission, even at low iridium loadings, indicating that
charge trapping at the metal center may be the dominant
mechanism. The efficiencies of the green devices are moderate,
yet the devices still represent improvements over blended
composite materials of organometallic phosphors in polyfluorene
host. The red devices, on the other hand, were designed to have
improved energy matching between iridium and fluorene energy
levels and show significant improvements in device effici-
ency.
Figure 4. (a) Room temperature electroluminescence spectra of 10 [Ir-
(ppy-Br)2(acac)], 11 [Ir(ppy-(FH))2(acac)], 12 [Ir(ppy-FO-FH)2(acac)], 13
[Ir(ppy-(FO)2-FH)2(acac)], 14 [Ir(ppy-(FO)10)2(acac)], 15 [Ir(ppy-(FO)30)2-
(acac)], blend A, and blend B. Spectra are normalized and offset along the
vertical axis for ease of comparison. The LEDs were made in a layer
structure of ITO/PEDOT:PSS/compound/Ca/Al. (b) Room temperature
electroluminescence spectra of 17 [Ir(btp-Br)2(acac)], 18 [Ir(btp-(FO)5)2-
(acac)], 19 [Ir(btp-(FO)10)2(acac)], 20 [Ir(btp-(FO)20)2(acac)], 21 [Ir(btp-
(FO)40)2(acac)], blend C, and blend D. Spectra are normalized and offset
along the vertical axis for ease of comparison. The LEDs were made in a
layer structure of ITO/PEDOT:PSS/compound/Ca/Al.
The new solution-processible triplet emitters have the po-
tential for further optimizations by structural modifications to
the iridium polyfluorene complexes which can be accomplished
by adapting the synthetic strategies described herein. In addition,
the new polymers represent a fascinating class of materials in
their own right for further study of the fundamental photophysics
of light-emitting polymers.
oligomer 15, energy transfer occurs onto electronically de-
coupled polyfluorene segments and singlet emission is now seen
with modest efficiency. Nevertheless, electroluminescence from
the triplet state is more efficient than in the corresponding
blends, where it occurs predominately from the singlet
state.
Acknowledgment. We thank the Engineering and Physical
Sciences Research Council U.K. (EPSRC) for financial support
and provision of the Swansea Mass Spectrometry Service, the
Commission of the EU (STEPLED, IST-2001-37375), the Royal
Society (University Research Fellowship to A.K.), the Com-
monwealth Scholarship Commission, Cambridge Common-
wealth Trust, and the Ramsay Memorial Trust (studentship to
N.R.E.) for generous financial support. We thank Cambridge
Display Technology Limited for the provision of substrates and
some analytical measurements.
The compounds linked to 17 show high PL and EL yields
(Table 2). The PL yields increase along the series, which we
attribute to reduced self-quenching. The same trend is observed
for the EL yields, with values up to 1.5% for 21, which exceeds
the 0.5% reported for nonconjugated analogs.45 The conjugated
polymer complex 20 has a slightly lower electroluminescence
quantum yield (ELQY) than the blended device of equivalent
iridium loading, blend C, although the film quality of the blends
is poor with crystalline phase separated regions visible by eye
and inhomogeneous light emission. On the other hand, the longer
chain complex 21 shows 3 times the efficiency of the corre-
sponding blend, D. These results demonstrate the subtle
influences of phosphor concentration on device efficiency, a
phenomenon which has previously been observed but as yet
remains poorly understood.62 It is clear that further fine-tuning
of the polymer structure, by, for example, substitution of the
Supporting Information Available: General experimental
techniques and optical measurements. Experimental procedure
for the synthesis of 1, 3-21, diiridium(III) di-µ-chlorotetrakis-
(2-(4′-bromophenyl)pyridinato-N,C2′), 2-(4′,4′,5′,5′-tetramethyl-
1′,3′,2′-dioxaborolane)benzo[b]thiophene, 2-(2′-benzo[b]thienyl)-
5-bromopyridine, diiridium(III) di-µ-chlorotetrakis(2-(2′-benzo-
[b]thienyl)-5-bromopyridinato-N,C3′). The X-ray CIF file and
representation of the crystal structure of 17 (the data has been
deposited with the Cambridge Crystallographic Database: CCDC
223879). This material is available free of charge via the Internet
at http://pubs.acs.org.
(62) Baldo, M. A.; Adachi, C.; Forrest, S. R. Phys. ReV. B 2000, 62, 10967-
10977.
JA039445O
9
7048 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004