Fluorene-based phosphine oxide host materials for blue electrophosphorescence: An effective strategy for a high triplet energy level

Article abstract of DOI:10.1002/chem.201003434

A bolt from the blue! Two novel host materials for blue electrophosphorescence were designed and synthesized by modification of fluorene at the 9-position with diphenylphosphine oxide to give high triplet energy levels of 2.99 eV (see scheme), achieved through an indirect linkage of the chromophore and the phosphine oxide moieties. These host materials were also found to exhibit high thermal and morphological stabilities and efficient energy transfer.

Full text of DOI:10.1002/chem.201003434

DOI: 10.1002/chem.201003434
Fluorene-Based Phosphine Oxide Host Materials for Blue
Electrophosphorescence: An Effective Strategy for a High Triplet Energy
Level
Donghui Yu,[a] Yongbiao Zhao,[b] Hui Xu,*[a] Chunmiao Han,[a] Dongge Ma,*[b]
Zhaopeng Deng,[a] Shan Gao,[a] and Pengfei Yan[a]
Electrophosphorescence has attracted much interest, with
the potential of 100% internal quantum efficiency, generat-
ed from both singlet and triplet excitons. Phosphorescent or-
ganic light-emitting diodes (PHOLEDs) based on electro-
phosphorescent materials can be applied to highly energy-
efficient, flat-panel displays and are also promising candi-
dates for the next generation of solid-state lighting.[1] How-
ever, the longer lifetime of triplet excitons increases the pos-
sibility of triplet–triplet annihilation and concentration
quenching. To improve the device performance, an effective
approach is doping the phosphors in host materials.[2] Never-
theless, the creation of stable and efficient blue-emitting
PHOLEDs remains a significant challenge.[3] For blue-emit-
ting electrophosphorescent doping systems, the efficient,
positive-energy transfer to the guest (such as bis(4,6-difluor-
ophenylpyridinato-N,C2)picolinatoiridium (FIrpic)) requires
a very high first triplet energy level (T1) of the host (T1
ꢀ3.0 eV).[4] Besides the high T1, a low operating voltage is
another significant factor that requires an excellent carrier
injection/transporting ability of the host.[3a–e] Usually, a high
T1 requires a small conjugated area, which is detrimental to
the carrier injection and transporting ability. Therefore, the
key issue for high-performance blue-emitting PHOLEDs is
how to develop efficient host materials with a high T1 and
excellent carrier injection/transporting ability.
tetraaryl silane derivatives.[6] However, the poor electron-in-
jection ability of carbazole derivatives, or the electrical iner-
tia of silicon,[7a] induces an unbalanced carrier injection/
transporting ability, which increases the operating voltage.
Recently, a number of aryl phosphine oxides (APO) deriva-
tives have shown excellent host characteristics for blue-emit-
ting PHOLEDs and attracted intense interest.[7] Results in-
dicate that the T1 of the APO hosts is determined by the
chromophores in the molecules. Furthermore, in contrast to
other insulating systems, the P=O moieties can efficiently
polarize the molecules to enable contributions to the lowest
unoccupied molecular orbital (LUMO).[7b] Therefore, APOs
can support both an efficient carrier injection/transporting
ability and a high T1. Nevertheless, for nearly all of the
APO hosts reported so far, the P=O moieties are directly
bonded to the chromophores along the long axis of the mol-
ecules; for example, the 2,7-substitution of fluorene,[7b,f–h] the
3,6-substitution of carbazole,[7a] and the 2,8-substitution of
dibenzofuran.[7c] Such structures are ineffective in maintain-
ing a high T1 because the P=O bond can still slightly reduce
the energy gap and the excited levels.[7a,d,e] Therefore, one of
the key problems for high-performance APO hosts is the de-
velopment of a suitable linkage mode of PO moieties and
chromophores to preserve a high T1 and facilitate further
multifunctionalization of the hosts. Recently, we reported a
novel APO host with an ortho-linked phosphine oxide
moiety.[8] This proved that the unsymmetrical structure is su-
perior in maintaining a high T1 and polarizing the chromo-
phore. To further improve the T1 of the APO hosts, we be-
lieved that an indirect linkage of the chromophore and
phosphine oxide moieties may be another effective strategy.
Herein, two fluorene-based APO hosts, 9-(4’-butylphen-
yl)-9-(diphenylphosphorylphenyl)fluorene (FSPO) and 9,9-
bis(diphenylphosphorylphenyl)fluorene (FDPO) were de-
signed and synthesized (Scheme1). In both FSPO and
FDPO, the diphenylphosphine oxide (DPPO) moieties were
bonded to fluorene through phenyl at the 9-position. This
structure has three advantages: 1) the twisted linkage be-
tween fluorene and APO moieties at the 9-position is effec-
tive in maintaining the excited level; 2) DPPO moieties
bond with fluorene on the short axis of the molecules, which
is more effective for polarization; 3) their unsymmetric
structures and the strong steric effect of DPPOs on both
To expand the conjugated area, most of the hosts are de-
signed to incorporate meso, twisted, or insulating linkages,
such as N,N-dicarbazoyl-3,5-benzene (mCP[4]), 9,9’-(2,2’-di-
methylbiphenyl-4,4’-diyl)bis(9H-carbazole) (CDBP[5]), and
[a] D. Yu, Dr. H. Xu, C. Han, Z. Deng, Prof. S. Gao, Prof. P. Yan
Key Laboratory of Functional Inorganic Material Chemistry
Ministry of Education, Heilongjiang University
74 Xuefu Road, Harbin 150080 (P.R. China)
Fax : (+86)451-86608042
E-mail: hxu@hlju.edu.cn
[b] Y. Zhao, Prof. D. Ma
State Key Laboratory of Polymer Physics and Chemistry
Changchun Institute of Applied Chemistry
Chinese Academy of Sciences
Changchun 130022 (P.R. China)
E-mail: mdg1014@ciac.jl.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003434.
2592
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2592 – 2596

COMMUNICATION
Scheme 1. Procedure used for the synthesis of FSPO and FDPO.
sides of fluorene can strongly reduce the p–p interactions
between two different fluorene rings. Compared to values
reported in the literature,[7c] the T1 of these two fluorene-
based APO hosts are improved by more than 0.2 eV. The
stable and efficient blue electroluminescence (EL) of the
corresponding PHOLEDs was then demonstrated. FSPO
and FDPO were conveniently prepared from fluorenone by
using a three-step procedure, which included a Friedel–
Crafts reaction, a Sandmeyer reaction, and a phosphoryliza-
tion reaction, to give total yields of 38 and 65%, respective-
ly (Scheme 1). The structure characterization was estab-
lished by using mass spectrometry, NMR spectroscopy, and
elemental analysis.
The molecular structures were further confirmed by X-ray
crystallography (Figure 1). The X-ray data reveal that the
unit cell of FSPO contains one molecule each of FSPO, eth-
anol, and n-hexane. It is interesting to note that no hydro-
gen-bonding aromatic and packing interactions exist be-
tween adjacent FSPO molecules (Figure SI1 in the Support-
ing Information). The formation of the crystals can be at-
tributed to a van der Waals force. The unit cell of FDPO
consists of one FDPO molecule and one water molecule.
Two FDPO molecules are interlinked by two water mole-
cules to form a dimer motif (Figure SI2 in the Supporting
Information). Furthermore, adjacent dimers are intercon-
ꢁ
nected by P=O···H C hydrogen bonds between the phenyl
Figure 1. ORTEP diagrams of FSPO (top) and FDPO (bottom) with
thermal ellipsoids at 30% possibility.
ring of fluorene and the P=O fragment, to give a two-dimen-
sional layer structure as shown in Figure SI3 in the Support-
ing Information. The three-dimensional supramolecular net-
ꢁ
works are formed by intermolecular C H···OACTHUNRTGNEUNG(water) hydro-
gen-bonding interactions (Figure SI4 in the Supporting In-
formation). For the 2,7-substituted fluorene derivatives (2,7-
bis(diphenylphosphine)-9,9-dimethylfluorene, PO6),[7b] both
tion is advantageous in forming the stable amorphous state
during film formation and in facilitating the uniform disper-
sion of the guest molecule in the host matrix.
ꢁ
the hydrogen-bonding interaction and the edge-to-face C
The thermal properties of FSPO and FDPO were investi-
gated by thermogravimetric analysis (TGA) and differential
scanning calorimetry (DSC) (Figure SI5 in the Supporting
Information). It is known that the alkyl modification may
improve the compatibility of host with the guest; however,
at the same time the flexible alkyl groups may remarkably
decrease the thermal performance. Even though there is a
butyl group in FSPO, the melting point (Tm) and the temper-
ature of glass transition (Tg) of FSPO are still as high as
H···p contacts provide strong intermolecular interactions
that are detrimental to the morphology stability of its thin
film and the stability of its doping systems. Evidently, the
stronger steric effect of DPPO at the 9-position of FSPO
and FDPO effectively eliminates the intermolecular p con-
tact. In addition, the long n-butyl of FSPO further reduces
the intermolecular interaction because only van der Waals
forces exist. It is known that a weak intermolecular interac-
Chem. Eur. J. 2011, 17, 2592 – 2596
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2593

H. Xu, D. Ma et al.
1658C and 978C, respectively. The Tg is much higher than
that of mCP (608C). Furthermore, the decomposition tem-
perature (Td) of FSPO is 3158C, which makes the device
fabrication feasible through vacuum evaporation. These
data are comparable those found for PO6,[7b] which has a Tg
and Tm of 105 and 1938C. FDPO exhibits much improved
thermal properties, with a Tm of 2738C and a Tg of up to
1428C. In addition, the Td of FDPO is near to 4008C.
FDPO has the best thermal properties, particularly film
morphology stability, compared to other small molecular
host materials.
measured by using 9,10-diphenylanthracene as the standard.
The PH spectrum of fluorene, FSPO, and FDPO showed
that the 0–0 transition band of fluorene is at 412 nm, corre-
sponding to a first triplet energy level (T1) of 3.01 eV. Signif-
icantly, this band was maintained in the PH spectra of FSPO
and FDPO, although it became much weaker with a smaller
redshift of 3 nm. It is evident that the substitution of DPPO
moieties at the 9-position of fluorene can preserve the trip-
let excited state level of the chromophore at 2.99 eV, which
is 0.27 eV higher than that of the 2,7-DPPO-substituted flu-
orine,[7b] therefore proving our molecular design strategy to
be effective.
The UV/Vis absorption spectra and photoluminescent
(PL) spectra of FSPO and FDPO in CH2Cl2 (10ꢁ6 molLꢁ1)
are shown in Figure 2. The absorption and fluorescence
The influence of DPPO substitution on the frontier mo-
lecular orbitals of FSPO and FDPO was investigated by
DFT calculations (Figure 3). Compared with 9,9-diphenyl-
Figure 2. Absorption and emission spectra of FSPO, FDPO, and fluorene
in CH2Cl2 (1ꢁ10ꢁ6 m). Absorption spectra: c&c: FSPO, d&d:
&
*
^
FDPO; g&g: fluorene. FL spectra: FSPO; FDPO; fluorene.
~
!
PH spectra: FSPO; FDPO; 3 fluorene.
(FL) spectra were measured at room temperature, and the
phosphorescent (PH) spectra were measured in CH2Cl2
glass at 77 K. For comparison, the absorption and PL spec-
tra of fluorene was also measured under the same condi-
tions. The absorption spectra of the phosphine oxide substi-
tuted fluorene and unsubstituted fluorene (from 250 to
350 nm) were nearly the same (Figure 2). The redshifts of
all the bands from fluorene in FSPO and FDPO were very
small (750–1100 cmꢁ1) and approximately half of that re-
ported for the 2,7-DPPO-modified fluorene derivative
(1500–2000 cmꢁ1). The main absorption peaks of FSPO and
FDPO are at 233 nm, originating from the DPPO moieties
(Figure 2). The FL spectra of the phosphine oxides consist
of two peaks at 312 and 324 nm and the shape of the FL
curves of FSPO and FDPO are very similar to that of fluo-
rine (Figure 2). Furthermore, the redshift between them is
only 730 cmꢁ1. Notably, the absorption coefficient increased
remarkably from fluorene FSPO to FDPO, which means
that with more phenyl groups, the DPPO moieties at the 9-
position on fluorene can efficiently sensitize the molecules,
but only slightly affect the singlet excited state properties of
the chromophore core fluorene. The relative quantum effi-
ciencies of FSPO and FDPO are 0.63 and 0.56, which were
Figure 3. DFT calculation results of FSPO and FDPO.
fluorene, the lowest unoccupied molecular orbitals (LUMO)
of FSPO and FDPO remarkably decrease to 0.082 and
0.218 eV, respectively. Simultaneously, their highest occu-
pied molecular orbitals (HOMO) also decrease to 0.082 and
0.191 eV, respectively. Furthermore, the LUMOꢁ1 of FSPO
and FDPO are 0.33 and 0.47 eV lower than that of 9,9-di-
phenylfluorene (Figure 3). This proves that the substitution
of DPPO moieties at the 9-position of fluorene can effec-
tively improve the electron-injection ability of the chromo-
phore. The electron-cloud densities of the LUMO and
HOMO of FSPO and FDPO are almost located on the fluo-
rene cores, as is similar with that of 9,9-diphenylfluorene.
Compared with 9,9-diphenylfluorene, with the exception of
9,9-diphenyl, DPPO moieties give remarkable contributions
to the electron-cloud density of the LUMOꢁ1 of FSPO and
FDPO. Thus, the pendent DPPO moieties may play an im-
portant role in electron transport. It is also noticeable that
2594
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2592 – 2596

Fluorene-Based Phosphine Oxide Host Materials
COMMUNICATION
the energy gap (Eg) of FSPO is nearly the same as that of
9,9-diphenylfluorene; however, the Eg of FDPO is slightly
smaller than those of FSPO and 9,9-diphenylfluorene
(Figure 3). It is known that a smaller host Eg is beneficial to
the balanced carrier injection/transport in the emitting
layers. Therefore, although the effect of DPPO moieties on
Eg is not very strong, a more balanced carrier injection/
transporting of FSPO and FDPO can be expected.
rier injection and transport in the emitting layer, as evi-
denced by the lower operating voltages and higher current
densities of B. Furthermore, the brightness of B was much
higher than that of A at the same voltages. This implies that
the charge-trapping in FIrpic was the main mechanism for
the emission process of the devices based on FSPO, and had
an important effect on improving the EL performance.
However, for FDPO, increasing the doping concentration
did not remarkably affect the EL performance, therefore
the energy transfer is the main mechanism for C and D,
which have the same turn-on voltages. The higher doping
concentration of D facilitated the formation of an exciton,
which induces its slightly higher current density and bright-
ness compared to C. The maximum power and current effi-
ciency of B were 12.5 lmWꢁ1 and 14.5 cdAꢁ1, respectively
(Figure 4 top right). The highest maximum efficiencies were
achieved from D, which were 13.8 lmWꢁ1 and 14.6 cdAꢁ1.
When increasing the doping concentrations, it was found
that the efficiencies of the devices did not decrease; in fact,
the efficiencies of B were nearly twice of those of A at same
current densities. This demonstrates that FSPO and FDPO
can effectively disperse the dyes uniformly so as to reduce
the concentration quenching and triplet–triplet anihilation.
The efficient energy transfer from FSPO and FDPO to
FIrpic was proved by the EL spectra of the devices
(Figure 4 bottom), in which only the emission from FIrpic
was observed. The spectra were very stable during the in-
crease in voltage.
To investigate the performance of FSPO and FDPO as
hosts, the devices with the configuration of ITOjMoO3
(10 nm)jNPB (80 nm)jTCTA (5 nm)jHOST:x% FIrpic
(20 nm)jTPBi (45 nm)jLiF (1 nm)jAl were fabricated, in
which MoO3 and LiF served as hole- and electron-injecting
layers, respectively, NPB served as hole-transporting layer
(HTL), TCTA served as exciton-blocking layer, and TPBi
was used as electron-transporting and hole-blocking materi-
al. Devices A and B were based on FSPO doping with 5 and
7% FIrpic, respectively, whereas devices C and D were
based on FDPO doping with 6 and 8% FIrpic, respectively.
The turn-on voltage of A was 3.5 V, whereas those of B–D
were about 3.3 V (Figure 4 top left). The maximum bright-
ness of A–D were 3132 (11.1 V, 250.6 mAcmꢁ2), 6977
(11.3 V, 304.7 mAcmꢁ2), 6962 (11.7 V, 386.8 mAcmꢁ2), and
7682 cdmꢁ2 (11.7 V, 435.1 mAcmꢁ2), respectively. It should
be noted that, in addition to the increasing of the dye con-
centration, the current-density–voltage (J–V) curves of the
devices based on FSPO shifted to low voltages and the in-
creasing of the concentration of FIrpic can improve the car-
Figure 4. Top left: Voltage/luminance (filled symbols)/current-density (open symbols) characteristics of the devices based on FSPO and FDPO with dif-
ferent doping concentrations. Top right: Current- and power-efficiency curves of the devices. Bottom left: EL spectra of the devices based on FSPO
(7%) at 4, 6, 8 and 10 V. Bottom right: EL spectra of the devices based on FDPO (8%) at 4, 6, 8 and 10 V.
Chem. Eur. J. 2011, 17, 2592 – 2596
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2595

H. Xu, D. Ma et al.
In conclusion, two novel APO hosts for blue electrophos-
phorescence were designed and synthesized according to an
effective strategy of the indirect linkage of the chromophore
and the phosphine oxide moieties. With the 9-substituted
structure, the T1 triplet state of the chromoophore fluorene
is almost preserved. The energies of the T1 states of FSPO
and FDPO are 2.99 eV, which are 0.27 eV higher than PO6
and ensure the positive energy transfer from the hosts to
FIrpic. The high thermal and morphology stability and the
effect of APO moieties on the carrier injection/transporting
ability of the hosts were proved by thermal analysis and
Gaussian simulation. The efficient energy transfer from
FSPO and FDPO to FIrpic was further demonstrated
through the doping of OLEDs. These results suggest that
the indirect linkage of the chromophore and P=O moieties
is advantageous in preserving T1 of the chromophore and
superior in constructing high-performance hosts for blue
electrophosphorescence.
[1] M. A. Baldo, D. F. OꢂBrien, Y. You, A. Shoustikov, S. Sibley, M. E.
Thompson, S. R. Forrest, Nature 1998, 395, 151–154.
[2] S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B.
Lussem, K. Leo, Nature 2009, 459, 234–238.
[3] a) M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson, S. R.
Forrest, Appl. Phys. Lett. 1999, 75, 4–6; b) C. Adachi, R. C. Kwong,
P. Djurovich, V. Adamovich, M. A. Baldo, M. E. Thompson, S. R.
Forrest, Appl. Phys. Lett. 2001, 79, 2082–2084.
[4] R. J. Holmes, S. R. Forrest, Y.-J. Tung, R. C. Kwong, J. J. Brown, S.
Garon, M. E. Thompson, Appl. Phys. Lett. 2003, 82, 2422–2424.
[5] S. Tokito, T. Iijima, Y. Suzuri, H. Kita, T. Tsuzuki, F. Sato, Appl.
Phys. Lett. 2003, 83, 569–571.
[6] a) X. Ren, J. Li, R. J. Holmes, P. I. Djurovich, S. R. Forrest, M. E.
Thompson, Chem. Mater. 2004, 16, 4743–4747; b) M.-H. Tsai, H.-W.
Lin, H.-C. Su, T.-H. Ke, C.-C. Wu, F.-C. Fang, Y.-L. Liao, K.-T. Wong,
C.-I Wu, Adv. Mater. 2006, 18, 1216–1220; c) L. X. Xiao, S.-J. Su, Y.
Agata, H. Lan, J. Kido, Adv. Mater. 2009, 21, 1271–1274.
[7] a) S. O. Jeon, K. S. Yook, C. W. Joo, J. Y. Lee, Adv. Funct. Mater.
2009, 19, 3644–3649; b) A. B. Padmaperuma, L. S. Sapochak, P. E.
Burrows, Chem. Mater. 2006, 18, 2389–2396; c) P. A. Vecchi, A. B.
Padmaperuma, H. Qiao, L. S. Sapochak, P. E. Burrows, Org. Lett.
2006, 8, 4211–4214; d) F.-M. Hsu, C.-H. Chien, P. I. Shih, C. F. Shu,
Chem. Mater. 2009, 21, 1017–1022; e) E. Polikarpov, J. S. Swensen, N.
Chopra, F. So, A. B. Padmaperuma, Appl. Phys. Lett. 2009, 94,
223304; f) S. E. Jang, C. W. Joo, S. O. Jeon, K. S. Yook, J. Y. Lee, Org.
Electron. 2010, 11, 1059–1065; g) J. Lee, J.-I. Lee, J. Y. Lee, H. Y.
Chu, Appl. Phys. Lett. 2009, 95, 253304; h) J. Lee, J. I. Lee, J. Y. Lee,
H. Y. Chu, Org. Electron. 2009, 10, 1529–1533; i) H.-H. Chou, C.-H.
Cheng, Adv. Mater. 2010, 22, 2468–2471; j) A. L. Von Ruden, L. Co-
simbescu, E. Polikarpov, P. K. Koech, J. S. Swensen, L. Wang, J. T.
Darsell, A. B. Padmaperuma, Chem. Mater. 2010, 22, 5678–5686.
[8] C. M. Han, G. H. Xie, H. Xu, Z. S. Zhang, D. H. Yu, Y. Zhao, Z. P.
Deng, Q. Li, P. F. Yan, Chem. Eur. J. 2011, 17, 445–449
Acknowledgements
This project was financially supported by NSFC (50903028 and
20972043), Science and Technology Bureau of Heilongjiang Province
(QC08C10), Education Bureau of Heilongjiang Province (10td03), Inno-
vation Fund of KLFIMC, the Supporting Program of Innovation teams
of HLJU (hdtd2010-08) and Heilongjiang University Funds for Distin-
guished Young Teachers.
Keywords: aryl phosphine oxide
·
energy transfer
·
Received: November 28, 2010
fluorene · phosphorescence · UV/Vis spectroscopy
Published online: January 24, 2011
2596
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2592 – 2596