Achieving high-efficiency non-doped blue organic light-emitting diodes:
charge-balance control of bipolar blue fluorescent materials with reduced
hole-mobility†
Chih-Chin Chi,
aChih-Long Chiang,
abShun-Wei Liu,
acHan Yueh,
dChin-Ti Chen*
aband Chao-Tsen Chen*
d Received 11th February 2009, Accepted 11th May 2009First published as an Advance Article on the web 19th June 2009 DOI: 10.1039/b902910a
We found an unusual way in improving electroluminescence efficiency of blue organic light-emitting
diodes (OLEDs). Two electron deficient 4,5-diazafluorene- or di(2,20-pyridyl)-containing blue
fluorophores, PhSPN2DPV (4,5-diaza-20-diphenylamino-70-(2,200-diphenylvinyl)-9,90-spirobifluorene)
and PhFpy2DPV (N-[7-(2,2-diphenylvinyl)-9,90-di(2,200-pyridyl)-2-fluorenyl]-N,N-diphenylamine), were
synthesized and characterized for non-doped blue OLEDs. Whereas PhFpy2DPV OLED performs
ordinarily, PhSPN2DPV OLED outperforms previously known PhSPDPV
(2-diphenylamino-7-diphenylvinyl-9,90-spirobifluorene) OLED significantly: maximum external quantum efficiency of5%
(4.6% at 20 mA cm2) and the peak electroluminance of 60510 cd m2(1810 cd m2at 20 mA cm2)
versus 3.4% (2.9% at 20 mA cm2) and 33020 cd m2(910 cd m2at 20 mA cm2) of PhSPDPV OLED.
We attribute the superior performance of PhSPN2DPV OLED to the good charge balancing, which is in
turn due to the very low hole mobility of PhSPN2DPV. The experimental results reveal that the
electron-deficient moiety, 4,5-diazafluorene or di(2,20-dipyridyl), increases electron affinity but reduces
the hole mobility. Electron mobility, determined by time-of-flight (TOF) method, is 5 105and 5
104cm2V1s1(at an electric field of 4.9 105V cm1) for PhSP
N2DPV and PhFpy2DPV, respectively.
Surprisingly, they are not higher than 8 104cm2V1s1of nonpolar PhSPDPV. On the other hand,
hole mobility is 2 106and 2 104cm2V1s1for PhSP
N2DPV and PhFpy2DPV, respectively, and
they are both significantly lower than 6 103cm2V1s1of PhSPDPV. For PhSP
N2DPV and
PhFpy2DPV bipolar blue fluorophores, we have demonstrated that electron-transporting and
light-emitting functions involve different molecular halves. The design of such molecular halves greatly facilitates the optical and electronic optimizations of fluorophores for high-performance OLEDs.
Introduction
Having a three times higher limit of theoretical electrolumines-cence (EL) efficiency, electrophosphoreselectrolumines-cence-based organic light-emitting diodes (OLEDs) are usually more favorable than
electrofluorescence-based OLEDs.1 However, blue
electro-fluorescence materials have revived and become highly deman-ded by the energy-saving solid-state lighting (SSL) of white OLEDs, which have been prevailed with troublesome
electro-phosphorescence materials recently.2In addition to the common
problem of short operational lifetimes2a,3and the high material
cost, electrophosphorescence-based OLEDs with acceptable blue
color purity (a Commission Internationnale de l’Eclairage,
CIEx,y, coordinate of y <0.25) are relatively rare when compared
with large numbers of green or red
electrophosphorescence-based OLEDs.4Electrophosphorescence-based OLEDs showing
true blue (CIE, coordinate of y <0.20) remain challenging and
they always have not so high external quantum efficiency (hEXT)
of 5.8–8.5%.5More critically, blue electrophosphorescence-based
OLEDs very often exhibit serious efficiency roll-off at elevated
driving current density,5due to the T
1-T1annihilation nature of
electrophosphorescence material. For practical SSL application, white OLEDs require electroluminance (L, or brightness) equal
to or greater than 1000 cd m2, which is usually corresponding to
a driving current density (J) over 20 mA cm2 for
electro-phosphorescence-based OLEDs.5 At such a current density,
hEXTs of blue electrophosphorescence-based OLEDs are usually
low, often less than 5.5% of the conventional upper limit of the
electrofluorescence efficiency of OLEDs.6 Under such
circum-stances, electrophosphorescence-based blue OLEDs have very little advantage over electrofluorescence ones
In recent years, a growing number of literature examples can be found for blue or sky blue electrofluorescence-based OLEDs
having a super high hEXTof 5 8%.7These
electrofluorescence-based OLEDs rival the best of true blue
electrophosphorescence-based OLEDs. Except of bis(4-phenylquinoline)-based
fluorophores,7ethese super efficient fluorophores are either
non-hetroatom-containing polycyclic aromatic hydrocarbons (PAHs) or arylamine-substituted PAHs, although insightful explanations
of super high hEXT’s are seldom reported.7Charge-transporting
aInstitute of Chemistry, Academia Sinica, Taipei, Taiwan, 11529, Republic
of China. E-mail: [email protected]; Fax: +886 2 27831237; Tel: +886 2 27898542
bDepartment of Applied Chemistry, National Chiao Tung University,
Hsin-Chu, Taiwan, 30050, Republic of China
cGraduate Institute of Photonics and Optoelectronics and Department of
Electrical Engineering, Taipei, Taiwan, 106, Republic of China
dDepartment of Chemistry, National Taiwan University, Taipei, Taiwan,
106, Republic of China. E-mail: [email protected]; Fax: +886 2 33664200 † CCDC reference numbers 682902 and 682903. For crystallographic data in CIF or other electronic format see DOI: 10.1039/b902910a
ª The Royal Society of Chemistry 2009
PAPER www.rsc.org/materials | Journal of Materials Chemistry
and charge-balancing of OLEDs are two factors that seem to be crucial in high efficiency blue electrofluorescence-based OLEDs. Previous investigations have reported the injection of electrons or electron-transporting properties found for fluorene-based
blue-emitting polymers and oligomers.8In addition,
arylamine-substituted PAHs, such as 4,40
-bis[N-(1-naphthyl)-N-phenyl-amino]biphenyl (NPB), have been shown for similar hole- and
electron-mobility.9 However, the high energy gaps (>3 eV)
essential for blue emission often result in low electron affinities
(EAs < 2.5 eV) of these blue-light-emitting materials.7e This
hampers charge injection, unbalances the charge carriers (hole and electron), and impairs the efficiency of blue OLEDs. Very often, elaboration of device fabrication (i.e., doping process) and complication of device architecture (i.e., multilayer device) are necessary to improve OLED performance.
In this paper, we report the synthesis, characterization, and EL property of a couple of novel blue fluorescent materials,
4,5-diaza-20-diphenylamino-70-(2,200-diphenylvinyl)-9,90
-spirobi-fluorene (PhSPN2DPV) and N-[7-(2,200-diphenylvinyl)-9,90
-di(2,200-pyridyl)-2-fluorenyl]-N,N-diphenylamine (PhF
py2DPV)
(Scheme 1). Due to the electron-rich diphenylamine substituent and electron-deficient nitrogen-containing heterocyclic moieties,
PhSPN2DPV and PhFpy2DPV are bipolar in nature. They are the
structurally modified version of virtually non-polar
2-dipheny-lamino-7-(2,200-diphenylvinyl)-9,90-spirobifluorene (PhSPDPV),
an arylamine-substituted PAH previously known as the blue
fluorophore for high efficiency non-dopant blue OLEDs.10
Bipolar PhSPN2DPV and PhFpy2DPV are special for two reasons.
First, structural moieties associated with opposite polarity (elec-tron-rich vs. electron-deficient) are separated from each other, i.e., they are not connected to each other by p-conjugation. Second, as we demonstrate in this paper, decoupled molecular functions, light-emitting and electron-transporting, are realized in these bipolar blue emitters. For OLED application, we have
found that PhSPN2DPV is significantly more efficient than
PhSPDPV for non-dopant blue OLEDs: maximum hEXT is
approaching 5.0% at about 100 cd m2(1 mA cm2), 4.6% at
1000 cd m2(19 mA cm2), and 4.2% at 3000 cd m2(50 mA
cm2). Furthermore, efficiency roll-off is much lighter than what
is commonly seen in electrophosphorescence-based blue OLEDs. For fundamental significance, we have demonstrated that the charge balancing control of EL devices, namely the reduction of the hole mobility in matching the electron mobility, is effective in achieving high EL efficiency of the bipolar blue fluorophores.
Results and discussion
Synthesis and characterizationAs shown in Scheme 2, the success to the preparation of
PhSPN2DPV and PhFpy2DPV hinges on the availability of
dibromo-substituted species, 4,5-diaza-20,70-dibromo-9,90
-spi-robifluorene and 2,7-dibromo-9,90-di(2,200-pyridyl) fluorene.
Whereas 4,5-diaza-20,70-dibromo-9,90-spirobifluorene was
previ-ously known,11 2,7-dibromo-9,9-di(2,20-pyridyl)fluorene was
synthesized through a four-step reaction. First, 2-iodo-4,40
-dibromobiphenyl was obtained by the iodination of
4,40-dibromobiphenyl via Sandmeyer reaction of 2-amino-4,40
-dibromobiphenyl,12 Second, 2-iodo-4,40-dibromobiphenyl was
mono-lithiated with a stoichiometric amount of n-butyllithium at
78C, followed by the reaction with di(2,20-pyridyl)ketone to
afford the key intermediate tertiary alcohol (carbinol). Unlike
the one in the preparation of 4,5-diaza-9,90-spirobifluorene, such
carbinol failed to undergo ring-closure reaction with acids in the
formation of 2,7-dibromo-9,90-di(2,200-pyridyl)fluorene due to
the electron deficiency of di(2,20-pyridyl) substituents. Such
adverse situation was not found for the carbinol of
4,5-diaza-9,90-spirobifluorene because of the non-resonanced
meta-posi-tion of nitrogen atoms to the tertiary carbon of carbinol. The
carbinol precursor of PhFpy2DPV was converted to tertiary
halide by the treatment of thionyl chloride and then followed by the ring-closure via Friedel–Crafts alkylation in the presence of Lewis acid (aluminium trichloride in this case). In the synthesis of
PhSPN2DPV and PhFpy2DPV, the following three steps,
mono-formylation, Pd-catalyzed aromatic amination, and Horner– Wadsworth–Emmons reaction (d, e, and f in Scheme 2), were very similar to those in the preparation of PhSPDPV and
others.10,13Three blue emitters were fully characterized by1H and
13C NMR, fast atom bombardment or electron impact-mass
spectrometry (FAB-MS or EI-MS), and elemental analysis, and were consistent with proposed structures.
The ultimate structural evidence is from single crystal X-ray
diffraction analysis.14 A distinct structural difference of
PhSPN2DPV and PhFpy2DPV can be clearly seen in Fig. 1,
namely the coplanar geometry of 4,5-diazafluorene moiety of the
Scheme 1 Chemical Structures of PhSPDPV, PhSPN2DPV, and
PhFpy2DPV.
Scheme 2 Synthetic routes to PhFN2DPV (top) and PhFpy2DPV
(bottom). Reagents and conditions: (a) i. Mg, THF; ii. 4,5-diaza-9-fluo-renone, THF, reflux; (b) H2SO4, HOAc, reflux; (c) Br2, FeCl3, CH2Cl2,
rt.; (d) i. n-BuLi, THF,78 C; ii. DMF, rt.; (e) Ph
2NH, Cs2CO3,
Pd(OAc)2, P(t-Bu)3, toluene, 120C; (f) NaH, Ph2CHPO(OEt)2, THF,
reflux; (g) i. NaNO2, 8 N HCl(aq), 0C; ii. KI, rt.; (h) i. n-BuLi, THF,
78 C; ii. di-2-pyridyl ketone, THF, rt.; (i) i. NaH, THF, rt., then
SOCl2, 0C; ii. AlCl3, CH3NO2, reflux. See the experimental section for
full details.
ª The Royal Society of Chemistry 2009
former and the twisted non-planar conformation of di(2,20 -pyr-idyl) substituent of the latter. These structural features play an important role in affecting electron affinity, fluorescence
quantum yield (Ff), charge carrier mobility, and eventually EL
efficiency of OLEDs.
Theoretical estimation of molecular orbitals
The molecular structures of PhSPDPV, PhSPN2DPV, and
PhFpy2DPV were optimized by applying density functional
theory (DFT) with the hybrid B3LYP functional and 6–31G*
basis set, although single crystal X-ray structure of PhSPN2DPV
and PhFpy2DPV were the beginning structure in their structural
optimization processes. With the optimized structure,
calcula-tions on the electronic ground states of PhSPDPV, PhSPN2DPV,
and PhFpy2DPV were processed using DFT with the hybrid
B3LYP functional and 6–31G* basis set.15 The singlet excited
states of the three molecules were studied with time-dependent density functional theory (TDDFT) by using the hybrid B3LYP
functional.16 All calculations were preformed with a
develop-mental version of Q-Chem.17 We display a comparison of the
corresponding HOMO1, HOMO, LUMO, and LUMO +1 of
the three blue emitters in S0 state (Fig. 2). Because electronic
excitation from the HOMO to the LUMO produces the first
singlet excited state S1 (a Frank–Condon excited state), the
orbital features presented in Fig. 2 might provide important clues towards understanding the nature of optically accessible first excited state. The contour plots show that the HOMOs mainly consist of the p-orbitals of fluorene ring p-conjugated to the vinyl group, although a large extent of p-orbitals of the elec-tronic donor, N,N-diphenylamino group, is involved as well. For the LUMOs, the orbitals mainly locate on diphenylvinylfluorene
part of the molecule. p-Orbitals of N,N-diphenylamino elec-tronic donor do not constitute LUMOs any more, although there remains certain p-orbital of the nitrogen atom of N,N-dipheny-lamino group. Such orbital involvement of HOMO and LUMO is nearly same in all three blue emitters. One can conclude that the photo-induced excitation (i.e., absorption) or photo-induced relaxation (i.e., emission) is most likely p–p* and moderate intramolecular charge-transfer (ICT) in nature. From our theo-retical estimation, one should particularly note that the orbitals of the other molecular halve, i.e., unsubstituted fluorenyl of
PhSPDPV, 4,5-diazafluorenyl of PhSPN2DPV, and di(2,20
-pyr-idyl) of PhFpy2DPV, little involve in the first singlet excited state
(S1) or Frank–Condon excited state, which is predominantly
composed of LUMO in vacuum condition without the consid-eration of solvation effect.
Photophysical and electrochemical properties
The absorption and emission spectra of these blue emitters were studied in different organic solvents. As shown by data in Table 1, the absorption peak wavelength of three blue emitters exhibits insignificant solvent polarity dependence. Furthermore,
as shown in Fig. 3, each of PhSPDPV, PhSPN2DPV and
PhFpy2DPV has very similar energy in Frank–Condon excitation
when they are under the same solvent environment (polarity). The results can lead us to conclude a similar solvation effect between the ground and Frank–Condon excited states. In other words, the Frank–Condon excited state is subject to a rather small dipolar change with respect to the ground state. Such analysis is valid for all three blue emitters herein. This will not be
Fig. 1 The molecular structure of PhSPN2DPV (top) and PhFpy2DPV
(bottom) determined by X-ray diffraction analysis. Solvent molecule dichloromethane is included in the top figure and hydrogen atoms are removed for clarity in the bottom figure.
Fig. 2 Calculated frontier molecular orbital diagrams of PhSPDPV, PhSPN2DPV, and PhFpy2DPV (from left to right) from DFT-B3LYP
with 6–31G* basis set calculation: (a) LUMO +1; (b) LUMO; (c) HOMO; (d) HOMO1.
ª The Royal Society of Chemistry 2009
conceivable unless similar molecular structure (i.e., 2-dipheny-lamino-7-diphenylvinylfluorene) constituting both HOMO and LUMO in all three fluorophores.
Unlike the lack of clear solvatochromism in absorption spectra, solvent polarity dependence is somewhat discernable in
the emission spectra of PhSPDPV, PhSPN2DPV and
PhFpy2DPV (Fig. 3), although there lacks quantitative linear
relationship in the plot of the fluorescence peak frequency as a function of solvent polarity. This indicates that the origin of the emission of the three blue emitters has limited character of intramolecular charge-transfer (ICT) or electron-transfer. Thus we can conclude that it is mainly non-directional p–p* transi-tions as the major feature in the emission of three blue emitters. However, considering the bathochromic emission in acetonitrile solution, we can not rule out that these blue fluorophores may adopt more polarized ICT state as the lowest excited state (involving LUMO +1 shown in Fig. 2 perhaps) under the influ-ence of highly polar environment.
One should note that the magnitude of Stokes shift (the energy gap between the first absorption band and emission spectrum) is
rather similar for PhSPDPV and PhSPN2DPV. Stokes shift is
4274 and 3948 cm1in toluene, 4539 and 4314 cm1in
chloro-benzene, and 6089 and 6425 cm1in acetonitrile, respectively for
PhSPDPV and PhSPN2DPV. From such results, we can derive
that the dipole moment of the emission excited state of these two blue emitters should be rather similar. Once again, such deriva-tion can not be justified unless the emission process does not
involve 4,5-diazafluorene of PhSPN2DPV or the unsubstituted
fluorene moiety of PhSPDPV. Within this context, di(2,20
-pyr-idyl) substituent of PhFpy2DPV may be involved to a certain
Table 1 Photop hysical and electroch emical propertie s a s w ell as energ y levels of thre e blu e emi tters Solutio n S o lid stat e D E (eV) c LUMO, HOM O (eV ) d E1/2 red,1 , E1/2 oxd,1 (V) e LUMO, HOM O (eV) f lmax ab (nm ) a lmax fl, (nm ) a Ff (%) b lmax ab , lmax fl (nm) Ff (%) b PhSPD PV 383 458 60 383, 467 51 2.90 2.58, 5.48 2.72, 0.46 2.08, 5.26 386 468 378 491 385 454 PhSP N2 DP V 388 466 66 390, 478 43 2.79 2.70, 5.49 2.63, 0.48 2.17, 5.28 379 501 383 462 PhF py2 DPV 386 480 48 389, 480 42 2.78 2.66, 5.44 2.69, 0.43 2.11, 5.23 379 505 a In toluene, chlorob enze ne, and ace tonit rile, respec tively. b F f ’s were determ ined by the inte grating-sphere metho d and the solv ent for solution Ff ’s was chlorob enze ne. c D E is the band -gap energ y estimated from the low energ y edge of the absorp tion spe ctra of the thin film mate rials. dHO MO energy level was determined by low-energy photoele ctron spectro meter (R iken-Keik i AC-2) and LU MO ¼ HO MO + D E. e Electro chemica l first reduct ion and first oxidatio n poten tials w ith respec t to the ferro cene /ferrocen ium red ox potent ial, whic h is 0.63 V vs. satur ated Ag/AgNO 3 , d etermined by differen tial pulsed volta mmetry in a 0.1 M solution o f (Bu 4 N)ClO 4 in tetrahyd rofuran . fLUMO ¼ 4.8 eV E1/2 red,1 and HO MO ¼ 4.8 eV E1/2 oxd,1 , w here 4.8 eV is the energ y leve l o f ferrocene /ferrocen ium below the vac uum leve l. 18
Fig. 3 Absorption (left) and emission (right) spectra of three blue emitters in three different solvents, toluene (top), chlorobenzene (center), and acetonitrile (bottom).
ª The Royal Society of Chemistry 2009
extent in the emission process because of its larger Stokes shifts
(4464, 5073, and 6584 cm1 in toluene, chlorobenzene, and
acetonitrile, respectively) when compared with those of
PhSPDPV and PhSPN2DPV. A lager Stokes shift indicates
a higher degree of change about the electron density distribution between the ground state and lowest photoexcited state. The larger Stokes shifts also reflect in the emission energy of
PhFpy2DPV, of which emission wavelength is always the longest
among three blue emitters (Fig. 3).
PhSPN2DPV has a relatively high fluorescence quantum yield
(Ffs) of 66% in chlorobenzene (Table 1), which is slightly better
than 60% of PhSPDPV. On the other hand, PhFpy2DPV has
a significantly lower Ffof 48% in chlorobenzene, probable due to
the vibrational quenching process of di(2,20-pyridyl) substituent.
Because of rotational vibration, di(2,20-dipyridyl) moiety is
relatively less rigid than fluorenyl of PhSPDPV or
4,5-diaza-fluorenyl of PhSPN2DPV. In the solid state, three blue emitters
all suffer from fluorescence concentration quenching, more or
less resulting in a reduction of Ff (see data in Table 1). From
solution to solid state, PhSPN2DPV has the highest Ffreduction
[(66 43)/66 ¼ 35%] among three blue emitters and it can be
attributed to its ground state dipole moment of 4.4 Debye,
which is relatively substantial when compared with virtually zero ground-state dipole moment (<1 Debye) of PhSPDPV or
PhFpy2DPV. We have previously demonstrated that solid state
fluorescence can be impaired by the molecular dipole.10
In order to probe the electron-accepting state of three blue emitters, electrochemical study was conducted by cyclic vol-tammetry (CV) and differential pulsed volvol-tammetry (DPV). Due to the ill-profile of CV voltammograms, the redox potential values of three blue emitters were based on voltammograms of
DPV (Fig. 4). As data listed in Table 1, the oxidation potential of three blue fluorophores was observed at a narrow potential range
of +0.99 +1.01 V vs.Ag/AgNO3(not shown in Fig. 4), which is
common to most diphenylamine-substituted spirobifluorene
compounds.11In the electrochemical measurement, we excluded
the near constant reduction potential around -0.9 V because of the residual oxygen in electrolyte solution. We particularly pay attention to the electrochemical reduction potential of the three blue emitters because it is parallel to electron affinity. The elec-trochemical measurement recorded the first reduction potential
of three blue fluorophores at 2.00 2.09 V vs.Ag/AgNO3
(Fig. 4), which agrees quite well with the reduction potential of
2.00 V reported for 4,5-diaza-9,90-spirobifluorene in
tetrahy-drofuran.11a Reduction potential is expected to be higher for
PhSPDPV than PhSPN2DPV or PhFpy2DPV due to the built-in
electron deficient structure of the latter two fluorophores.
According to the experimental data (E1/2red,1or the energy level of
LUMO derived from E1/2red,1) listed in Table 1, electron affinity is
strongest for PhSPN2DPV, followed by PhFpy2DPV and then
PhSPDPV. Even though such order is the same as the energy levels of LUMO (Frank–Condon excited state) determined by spectroscopic method (Table 1), we tend to believe that the electron reduction (and hence electron transporting) happens via
4,5-diazafluorene of PhSPN2DPV and di-2-pyridylmethylene of
PhFpy2DPV, which is the molecular halve that is irrelevant to the
emission process (luminescence).
Therefore, we seem to have two unusual cases (PhSPN2DPV
and PhFpy2DPV) that charge (electron) transporting and
light-emitting primarily involve different molecular halves.
4,5-Diazafluorenyl or di(2,20-diptridyl) molecular halve is
responsible for charge-transporting and diphenylamino-diphe-nylvinyl-substituted molecular halve is the structural moiety that is luminescent. Such divider of the charge (electron)-transporting and light-emitting function has been identified before on
4,5-diazafluorene-incorporated ter(9,90-di(p-tolyl)fluorene).11a
However, we have found, for the first time, electrochemical
reduction potentials (and hence electron affinity) of PhSPN2DPV
and PhFpy2DPV are closely related to their downward hole
mobilities, instead of the electron mobilities (see following section for details). The new finding has profound influence on their OLED performance.
Charge carrier mobility and single-carrier devices
The charge-transporting characteristics of the blue emitters can be best understood by their charge carrier mobility via the
transient photocurrent recorded by time-of-flight (TOF)
method.19The field dependence of the charge carrier mobility of
PhSPDPV, PhSPN2DPV, and PhFpy2DPV is shown in Fig. 5. By
extrapolating the data to the electric field at4.9 105V cm1
by the Poole–Frenkel relation, electron mobility is 8 104, 5
105, and 5 104cm2V1s1for PhSPDPV, PhSP
N2DPV, and
PhFpy2DPV, respectively. Apparently, the trend of electron
mobility does not follow the order of electron affinity of three emitters. Electron deficient structure moiety does not necessary enhance the electron mobility. This is nothing to be surprised because electron mobility depends on internal reorganization energies for electron transfer, in addition to the electron
affinity.20 Furthermore, charge carrier mobility is highly
Fig. 4 Cyclic voltammograms (top) and differential pulsed voltammo-grams (bottom) of three blue emitters in tetrahydrofuran containing 0.1 M of (Bu4N)ClO4as the supporting electrolyte.
ª The Royal Society of Chemistry 2009
molecular interaction dependent, which is closely related to the molecular aggregation in solid state, i.e., solid state order. Nevertheless, we do see the effect of those electron deficient moieties in lowering the hole mobility of the materials. Whereas
the electron mobility of PhSPN2DPV is still lower than that of
PhSPDPV, at an electric field of 2.5 105 V cm1 diaza
substituent dramatically reduces the hole mobility from 6
103cm2V1s1of PhSPDPV down to2 106cm2V1s1of
PhSPN2DPV. Similarly, a decreased hole mobility (2 104
cm2V1s1at2.5 105V cm1) was observed for PhF
py2DPV,
although it is to a smaller extent when compared with that of
PhSPN2DPV. Interestingly, for PhSPN2DPV and PhFpy2DPV,
the decrease of the hole mobility is parallel to the increase of the electron affinity.
We also fabricated the single-carrier devices for the evaluation of the charge carrier mobility in terms of current density of
PhSPDPV, PhSPN2DPV, and PhFpy2DPV. As shown in the inset
of Fig. 6, we used high-LUMO NPB to limit the electron carrier in hole-only devices and low-HOMO BCP
(2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline) and TPBI (2,20,200
-(1,2,5-phenyl-ene)tris(1-phenyl-1H-benzimidazole) to limit the hole carrier in electron-only devices. For the hole-only devices, the magnitude of current density clearly follows the order of hole mobility of
three blue emitters, which is PhSPDPV > PhFpy2DPV >
PhSPN2DPV. One should note that the current density of a
hole-only device is determined by several factors, hole injection barrier between the electrode and organic layer, the difference of the HOMO energy levels between the adjacent organic layers, and the hole mobility of each organic layers. In the case studied herein, HOMO energy levels of three blue emitters are relatively invariant, the biggest difference among blue emitters is only 0.05 eV. Therefore, the hole mobility of the blue emitters becomes the sole factor in determining current density of the hole-only devices. The situation of the current density of electron-only devices is more complicated. First, it is highly possible that the molecular part of the molecule involving electron transporting of
PhSPDPV is different from that of PhFpy2DPV or PhSPN2DPV.
Second, the variation of LUMO energy levels of the three blue emitters is not negligible and the variation can be as high as
0.15 eV. According to the estimated LUMO energy levels,
PhSPN2DPV has the lowest LUMO energy level and it is the
nearest one to the LUMO energy level of BCP or TPBI.
There-fore, among three blue emitters, it is PhSPN2DPV to show
highest current density in electron-only device, even though its electron mobility is not the highest among three blue emitters. On
the other hand, PhFpy2DPV shows lower current density than
does PhSPDPV and it can be attributed to its lower electron mobility. However, such current density analysis of electron-only devices seems to be over simplified. For hole-only device, NPB
has a hole mobility of 5.5 103 cm2 V1 s1,21 which is
comparable with that of PhSPDPV and significantly higher than
that of PhSPN2DPV or PhSPpy2DPV. This is not quite the same
for electron-only devices. Unlike three blue fluorophores domi-nating the current density in hole-only devices, it is BCP or TPBI instead of blue emitters as the limiting factor of the current density in electron-only devices. Literature reported value of the
electron mobility is about 5 107and 5 106cm2V1s1(at
an electric field of 4.9 105 V cm1) for BCP and TPBI,
respectively.22 This further complicates the assessment of the
electron mobility via the current density of electron-only device.
OLED characterization
Table 2 and Fig. 7 summarize the EL characteristics of three non-doped OLEDs based on three blue emitters studied herein. For
Fig. 6 Current density-voltage (I–V) characteristics of carrier-only devices: hole-only device (top) and electron-only device (bottom). Fig. 5 Field dependent carrier mobilities of PhSPDPV (top left),
PhSPN2DPV (top right), and PhFpy2DPV (bottom).
ª The Royal Society of Chemistry 2009
a fair comparison, three non-doped OLEDs have a same simple three-layer configuration: ITO/NPB (10 nm)/blue emitter (40 nm)/TPBI (50 nm)/LiF (1 nm)/Al (150 nm), where NPB is
a common hole transporting layer, TPBI is a common electron transporting layer. No exotic or fancy hole/electron-injecting layer or hole-blocking layer is necessary herein. Whereas
PhFpy2DPV OLED is slightly better than PhSPDPV one, bipolar
emitter PhSPN2DPV significantly outperforms previously known
PhSPDPV. PhSPN2DPV OLED has a remarkable EL
perfor-mance: maximum efficiency of 5% (or 10 ml W1), peak
brightness of 60,000 cd m2(1,800 cd m2 at 20 mA cm2).
More significantly, within a range of brightness (100 1,000 cd
m2) required for SSL, PhSP
N2DPV exhibits only miner
effi-ciency roll-off from 1 to 20 mA cm2with h
EXTof 5.0% to
4.6%. Such performance ranks PhSPN2DPV on top of most
electrofluorescence-based blue OLEDs and rivals those
super-efficient ones.7However, compared with thin-film emission peak
wavelength (Table 1), 6–11 nm of red-shifting were observed for EL of three blue emitters. Whereas PhSPDPV OLED is blue
CIEx,y ¼ 0.14, 0.22, PhSPN2DPV and PhFpy2DPV OLEDs
exhibit sky blue color CIEx,y¼ 0.16, 0.32 and CIEx,y¼ 0.15, 0.31,
respectively (Fig. 7a). In most cases, a sky blue OLED often performs better than a true blue OLED. However,
high-perfor-mance PhSPN2DPV OLED is not simply a matter of color purity
because the performance of similarly sky blue PhFpy2DPV
OLED is not much better than that of PhSPDPV OLED
(Table 2). Moreover, PhSPN2DPV has Ff around 43% in the
solid state, which is substantially lower than Ff 51% of
PhSPDPV but very similar to 42% of PhFpy2DPV.
The superior performance of PhSPN2DPV OLED compared
with PhSPDPV or PhFpy2DPV OLED can be mainly attributed
to the issue of charge balancing in their OLEDs. PhSPDPV behaves like a typical p-conjugated organic material: hole mobility is an order of magnitude higher than electron mobility.
This is quite different from the electron deficient PhFpy2DPV or
PhSPN2DPV. Whereas PhFpy2DPV seems to be ambipolar (hole
and electron nobilities are quite similar to each other around 5
104 cm2 V1 s1 at an electric field of 4.9 105 V cm1),
PhSPN2DPV has a very low hole mobility of 3 106cm2V1
s1, which is even lower than its electron mobility 5 105cm2
V1s1at the same electric field (Fig. 5). With such characteristics
in charge carrier mobility (a significantly reduced hole mobility),
PhSPN2DPV OLED can be expected to be superb in EL
effi-ciency (maximum hEXT5% and 4.6% at 20 mA cm2) because
of the satisfactory situation of charge balancing. Normally, an
excess amount of hole carrier is expected if PhSPN2DPV (or
PhFpy2DPV) is absent in the device, i.e., ITO/NPB/TPBI/LiF/Al
is a hole-excess device. Within this context, charge balancing is not optimized in PhSPDPV OLED and its EL efficiency is expected to be less satisfactory because of very high hole mobility
(5 103cm2V1s1at an electric field of2.5 105V cm1)
and relatively low electron mobility (5 104cm2V1s1at an
Table 2 Electroluminescence characteristics of non-doped blue OLEDs of PhSPDPV, PhSPN2DPV, and PhFPy2DPV a
Max. luminance and voltage (cd m2, V)
Luminance, Efficiency,
Voltage (cd m2, %, V)b Max. Efficiency (%, cd
A1, lm W1) l maxel(nm) 1931 CIE Chromaticity (x, y) PhSPDPV 33020, 15 910, 2.9, 4.7 3.4, 5.4, 5.7 478 0.14, 0.22 PhSPN2DPV 60510, 15 1810, 4.4, 5.2 4.9, 10.2, 10.5 488 0.16, 0.32 PhFPy2DPV 15070, 15 1140, 2.8, 5.1 3.2, 6.4, 5.7 486 0.15, 0.31
aDevices have the configuration of ITO/NPB (10 nm)/blue emitter (40 nm)/TPBI (50 nm)/LiF (1 nm)/Al (150 nm).bAt current density of 20 mA cm2.
Fig. 7 Electroluminescence (EL) characteristics of non-doped blue OLEDs: PhSPDPV, PhSPN2DPV, and PhFpy2PV. EL spectra and CIE
chromaticity diagram (a), J–V–L (b), and hEXT–J–hPW(c) characteristics
of devices.
ª The Royal Society of Chemistry 2009
electric field of2.5 105V cm1) (see Fig. 5). However, it is the
relatively high Ff(51%) of PhSPDPV in the solid state rendering
hEXT(maximum 3.4% and 2.9% at 20 mA cm2) to be
compa-rable with or slightly better than that (hEXTmaximum 3.2% and
2.8% at 20 mA cm2) of PhF
py2DPV OLED.
Conclusions
In this study, we have synthesized and characterized two unusual
bipolar fluorophores, PhSPN2DPV and PhFpy2DPV, which
exhibit high performance when applied for OLEDs. Particularly,
non-doped PhSPN2DPV OLED display sky-blue EL very bright
(peak intensity of 60510 cd m2or 1810 cd m2at 20 mA cm2) and
efficient (maximum external quantum efficiency 4.9% or 4.4% at
20 mA cm2), having minimum efficiency roll-off, 4.9% at 100 cd
cm2(1 mA cm2) and 4.6% at 1,000 cd cm2(19 mA cm2).
PhSPN2DPV with such EL performance rivals those
super-effi-cient electrofluorescence blue OLEDs and is potential alternative for troublesome electrophosphorescence blue OLEDs. Funda-mentally, we have demonstrated that the reduction of the hole mobility is crucial for charge balancing in high efficiency OLEDs.
Moreover, we have verified that both PhSPN2DPV and
PhFpy2DPV have one molecular halve responsible for
electron-transporting and the other molecular halve for light-emitting. Such a structural separation in charge-transporting and light-emitting processes renders a potent molecular design in the opti-mization of fluorescent materials for EL application.
Experimental
General methodsBoth solution and solid-state fluorescence quantum yields (Ffs)
of the blue emitters were determined by the integrating-sphere
method.23The ionization potentials (or HOMO energy levels) of
three compounds were determined by low energy photo-electron spectrometer (Riken-Keiki AC-2). From our previous data, we
found that the precision of AC-2 measurement is about 0.05
eV. LUMO energy levels were estimated by subtracting the energy gap (DE) from HOMO energy levels. DE was determined by the on-set absorption energy from the absorption spectra of the materials as thin films. UV-visible electronic absorption spectra were recorded on a Hewlett-Packard 8453 Diode Array Spectrophotometer. Cyclic voltammetry (CV) and differential pulsed voltammetry (DPV) were performed using an Electro-chemical Analyzer BAS C3. The measurement, using a glassy carbon rod as the working electrode and a platinum wire as the counter electrode, was performed in a 0.1 M solution of
(Bu4N)ClO4in anhydrous tetrahydrofuran. The reported redox
potentials were estimated from DPV voltammograms in volts vs
saturated Ag/AgNO3 reference potential. We used ferrocene/
ferrocenium (0.63 V vs Ag/AgNO3) as reversibility criteria and
the calibration for energy levels (4.8 eV below the vacuum
level).18The method of time-of-flight (TOF) in measuring charge
carrier mobility has been described before.19
Synthesis of materials
4,5-Diaza-20,70-dibromo-9,90-spirobifluorene was prepared
according to a procedure previously reported by Wong et al.11
2-Amino-4,40-dibromobiphenyl was prepared in two steps
from 4,40-dibromobiphenyl according to published procedures.12
For the materials used in device fabrication, BCP (2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline) was commercially available
material. NPB (4,40
-bis[N-(1-naphthyl)-N-phenylamino]-biphenyl)24a and TPBI (2,20,200
-(1,2,5-phenylene)-tris(1-phenyl-1H-benzimidazole)24bwere prepared via published methods and
were subjected to gradient sublimation prior to use.
4,5-Diaza-20-bromo-9,90-spirobifluorene-70-carboxaldehyde. To
a THF solution (650 mL) of 4,5-diaza-20,70-dibromo-9,90
-spi-robifluorene (4.76 g, 10.0 mmol) was added n-BuLi (4.8 mL, 12.0
mmol, 1.6 M in hexane) slowly at78 C. The mixture was
stirred for 1 h under nitrogen atmosphere. After the slow addi-tion of DMF (2.33 mL, 30.0 mmol), the reacaddi-tion soluaddi-tion was gradually warmed up to room temperature and kept at this temperature overnight. The reaction was quenched with a 2 N HCl aqueous solution, extracted with ethyl acetate, and dried
over MgSO4. The solution was concentrated under reduced
pressure and subjected to flash column chromatography (silica gel, ethyl acetate/toluene: 2/3). A white solid was obtained with
a yield of 40% (1.7 g).1H NMR (400 MHz, CDCl 3): d 9.80 (s, 1H), 8.77 (dd, 2H, J¼ 4.6, 1.6 Hz), 7.97 (d, 1H, J ¼ 7.8 Hz), 7.94 (dd, 1H, J¼ 7.9, 1.3 Hz), 7.79 (d, 1H, J ¼ 8.2 Hz), 7.58 (dd, 1H, J ¼ 8.2, 1.8 Hz), 7.22 (d, 1H, J ¼ 1.3 Hz), 7.16 (d, 1H, J ¼ 4.8 Hz), 7.14 (d, 1H, J¼ 4.7 Hz), 7.10 (dd, 2H, J ¼ 7.7, 1.6 Hz), 6.89 (d, 1H, J¼ 1.7 Hz).13C NMR (100 MHz, CDCl 3): d 190.1, 158.8, 150.8, 149.2, 146.7, 146.5, 141.8, 139.1, 136.4, 132.1, 131.6, 130.9, 127.3, 124.8, 123.8, 123.7, 122.7, 120.8, 61.1. FAB-MS: calcd. 424.02, m/z¼ 425.03/427.03 (M + H+).
4,5-Diaza-20-diphenylamino-9,90-spirobifluorene-70
-carboxalde-hyde. A mixture of 4,5-diaza-20-bromo-9,90-spirobifluorene-70
-carboxaldehyde (1.47 g, 3.45 mmol), diphenylamine (0.64 g, 3.8
mmol), Pd(OAc)2(70 mg, 0.3 mmol), Cs2CO3(1.24 g, 3.8 mmol),
and P(tBu)3(0.12 g, 0.6 mmol) in toluene (35 mL) was stirred at
120C for 6 h under nitrogen atmosphere. After cooling to room
temperature, a saturated ammonium chloride solution was added to the reaction solution. The solution was extracted with
dichloromethane and dried over MgSO4. The solution was
concentrated under reduced pressure and subjected to flash column chromatography (silica gel, ethyl acetate/hexanes: 3/2).
A yellow solid was obtained with a yield of 55% (0.98 g).1H
NMR (400 MHz, CDCl3): d 9.77 (s, 1H), 8.70 (dd, 2H, J¼ 4.6, 1.7 Hz), 7.89 (dd, 1H, J¼ 7.9, 1.4 Hz), 7.84 (d, 1H, J ¼ 7.9 Hz), 7.10–7.20 (m, 9H), 7.74 (d, 1H, J¼ 8.4 Hz), 7.05 (dd, 1H, J ¼ 8.4, 2.1 Hz), 6.90–6.98 (m, 6H), 6.42 (d, 1H, J¼ 2.0 Hz).13C NMR (100 MHz, CDCl3): d 191.2, 158.9, 150.6, 149.8, 149.0, 147.9, 146.8, 142.7, 135.2, 133.7, 131.6, 129.3, 131.1, 124.7, 123.9, 123.7, 122.9, 122.2, 119.7, 117.4, 61.2. FAB-MS: calcd. 513.18, m/z¼ 514.19 (M + H+).
4,5-Diaza-20-diphenylamino-70-(2,200-diphenylvinyl)-9,90
-spirobi-fluorene (PhSPN2DPV). To diethoxydiphenylmethylphosphonate
(6.8 g, 22.0 mmol) in dry THF (10 mL) was added NaH (0.96 g, 60% w/w dispersion in mineral oil, 24.0 mmol), and stirred for 1 h
at 55 C under nitrogen atmosphere. After cooling to room
temperature, 4,5-diaza-20-diphenylamino-9,90-spirobifluorene-70
-carboxaldehyde (1.21 g, 2.36 mmol) was added, the reaction
ª The Royal Society of Chemistry 2009
solution was heated at reflux temperature overnight. After cool-ing to room temperature, the reaction mixture was added to
water, extracted with ethyl acetate, and dried over MgSO4. The
solution was concentrated under reduced pressure and subjected to flash column chromatography (silica gel, ethyl acetate/ hexanes: 2/3). A yellow solid was obtained with a yield of 68%
(1.06 g).1H NMR (400 MHz, CDCl 3): d 8.65 (dd, 2H, J¼ 4.4, 1.9 Hz), 7.57 (d, 1H, J¼ 8.3 Hz), 7.54 (d, 1H, J ¼ 7.9 Hz), 7.16–7.23 (m, 5H), 7.07–7.16 (m, 9H), 6.96–7.07 (m, 4H), 6.86–6.96 (m, 8H), 6.82 (s, 1H), 6.34 (d, 1H, J¼ 2.0 Hz), 5.94 (s, 1H).13C NMR (100 MHz, CDCl3): d 158.6, 150.0, 148.0, 147.7, 147.2, 145.6, 143.3, 142.6 (2), 140.1, 139.7, 136.5, 135.8, 131.5, 130.4, 129.6 (2), 129.1, 128.4, 128.1, 127.5, 127.4, 127.2, 124.2, 124.1, 123.6, 123.5, 123.1, 120.8, 119.1, 118.7, 61.1. FAB-MS: calcd. 663.27, m/
z¼ 664.27 (M + H+). Anal. calcd for C
49H33N3: C 88.66, H 5.01,
N 6.33; found: C 88.49, H 5.14, N 6.23.
2-Iodo-4,40-dibromobiphenyl. A solution of sodium nitrite (1.16
g, 16.8 mmol) in water (5 mL) was added dropwise at 0C to
a mixture of 2-amino-4,40-dibromobiphenyl (5.0 g, 15.3 mmol)
and 8 N HCl aqueous solution (100 mL). After 30 min of stirring, an aqueous solution of potassium iodide (2.8 g, 16.8 mmol) was added during 5 min and the mixture was warmed up to room temperature for 1 h. The solution was extracted with diethyl ether. The organic phase was washed with an aqueous solution of
NaHSO3and water, and dried over MgSO4. The solution was
concentrated under reduced pressure and subjected to flash column chromatography (silica gel, hexanes). A white solid was
obtained with a yield of 67% (4.5 g).1H NMR (400 MHz, CDCl
3):
d8.08 (d, 1H, J¼ 2.0 Hz), 7.53 (dm, 2H, J ¼ 8.5 Hz), 7.16 (dm, 2H,
J ¼ 8.5 Hz), 7.11 (d, 1H, J ¼ 8.2 Hz).13C NMR (100 MHz,
CDCl3): d 144.4, 141.9, 141.5, 131.4, 131.3, 130.8, 130.7, 122.3,
121.9, 98.7. EI-MS: calcd. 435.8, m/z 435.8/437.9 (M+).
2-[Di(2,200-pyridyl)hydroxymethyl]-4,40-dibromobiphenyl. To
a THF solution (100 mL) of 2-iodo-4,40-dibromodiphenyl (10.0 g,
23 mmol) was added n-BuLi (15.6 mL, 25 mmol, 1.6 M in hexane)
slowly at78C. The mixture was stirred for 1 h under nitrogen
atmosphere. Di(2-pyridyl) ketone (4.6 g, 9.3 mmol) in THF (20
mL) was added to the reaction solution at78C. The mixture
was continuously stirred at78C for 1 h and then warmed up to
room temperature for 1 h. Saturated ammonium chloride solution was added to the reaction solution. The solution was extracted
with ethyl acetate and dried over MgSO4. The solution was
concentrated under reduced pressure and subjected to flash column chromatography (silica gel, ethyl acetate/hexanes: 1/10).
A white solid was obtained with a yield of 86% (9.7 g).1H NMR
(400 MHz, CDCl3): d 8.39 (m, 2H), 7.61 (d, 2H, J¼ 8.0 Hz), 7.49 (td, 2H, J¼ 7.8, 1.7 Hz), 7.39 (dd, 1H, J ¼ 8.1, 2.0 Hz), 7.10–7.05 (m, 5H), 6.93 (d, 1H, J¼ 8.1 Hz), 6.88 (d, 1H, J ¼ 2.0 Hz), 6.82 (d, 2H, J¼ 8.4 Hz).13C NMR (100 MHz, CDCl 3): d 162.1, 147.1, 147.0, 141.1, 140.4, 136.3, 133.6, 131.8 (2), 130.3, 129.5, 123.0, 122.2, 121.1, 120.3, 80.5. FAB-MS: calcd. 494.0, m/z ¼ 495.0/
497.0 (M + H+). Anal. calcd for C
23H16Br2N2O: C 55.67, H 3.25,
N 5.65; found: C 55.67, H 3.13, N 5.70.
2-[Di(2,200-pyridyl)chloromethyl]-4,40-dibromobiphenyl. The
functional group transformation from carbinol to chloride
was performed by pertinent literature method.25To the carbinol
(4.5 g, 9.18 mmol) in dry THF (50 mL) was added NaH (0.48 g, 60% w/w dispersion in mineral oil, 12 mmol) and stirred for 1 h at
room temperature. After cooling the reaction solution to 0C,
thionyl chloride (0.9 mL, 12 mmol) was added dropwise to the mixture and stirred for 1 h. The reaction was quenched with
saturated NaHCO3 solution, extracted with dichloromethane
and dried over Na2SO4. The solution was concentrated under
reduced pressure. Chloro-substituted product was obtained as a pale yellow solid. It was used directly in the next step of
synthesis without further purification. 1H NMR (400 MHz,
CDCl3): d 8.5 (d, 2H, J ¼ 4.4 Hz), 7.52–7.40 (m, 6H), 7.10
(dd, 2H, J¼ 7.3, 4.8 Hz), 7.01 (d, 2H, J ¼ 8.4 Hz), 6.90 (d, 1H, J
¼ 8.0 Hz), 6.66 (d, 2H, J ¼ 8.3 Hz).
2,7-Dibromo-9,90-di(2,200-pyridyl)fluorene. The crude
chloro-substituted intermediate was dissolved in nitromethane (50 mL). Powdered aluminium chloride (5.0 g, 37 mmol) was added to the solution at room temperature under nitrogen atmosphere and then heated under reflux for overnight. After cooling to room
temperature, saturated NaHCO3solution was carefully added to
the reaction mixture and then extracted with dichloromethane for several times. The organic solution was concentrated under reduced pressure and a large amount diethyl ether was poured into the residue. A light yellow solid was obtained after filtration. The filtrate was purified from flash column chromatography (silica gel, ethyl acetate/hexanes: 1/4) and a white solid was
obtained. In combining two solid samples, 2,7-dibromo-9,90
-di(2,200-pyridyl)fluorene was obtained with a yield of 75% (3.3 g).
1H NMR (400 MHz, CDCl 3): d 8.62–8.57 (m, 2H), 7.95 (d, 2H, J ¼ 1.7 Hz), 7.59 (d, 2H, J ¼ 8.1 Hz), 7.55–7.47 (m, 4H), 7.15–7.10 (m, 2H), 7.0 (d, 2H, J¼ 8.0 Hz).13C NMR (100 MHz, CDCl 3): d 162.9, 150.0, 149.8, 138.7, 136.6, 131.4, 130.4, 122.0, 121.8, 121.5, 120.9, 68.8. EI-MS: calcd. 476.0, m/z¼ 476.0/478.0 (M+).
Anal. calcd for C23H14Br2N2: C 57.77, H 2.95, N 5.86; found: C
57.34, H 2.81, N 5.90.
7-Bromo-9,90-di(2,200-pyridyl)-2-fluorenecarboxaldehyde. This
compound was prepared similarly to that of 4,5-diaza-20
-bromo-9,90-spirobifluorene-70-carboxaldehyde, except that
2,7-dibromo-9,90-di(2,200-pyridyl)fluorene (2.0 g, 4.18 mmol) was
used in the reaction, and ethyl acetate/dichloromethane/hexanes (1/1/2) was used as eluent for flash column chromatography. A
milk-white solid was obtained with a yield of 58% (1.03 g).1H
NMR (400 MHz, CDCl3): d 9.99 (s, 1H), 8.62–8.57 (m, 2H), 8.29 (d, 1H, J¼ 0.8 Hz), 8.01 (d, 1H, J ¼ 1.7 Hz), 7.95 (dd, 1H, J ¼ 7.9, 1.4 Hz), 7.87 (d, 1H, J¼ 7.9 Hz), 7.70 (d, 1H, J ¼ 8.1 Hz), 7.57 (dd, 1H, J¼ 8.2, 1.8 Hz), 7.51 (td, 2H, J ¼ 7.8, 1.9 Hz), 7.13 (ddd, 2H, J¼ 7.5, 4.8, 1.0 Hz), 7.01 (d, 2H, J ¼ 7.2 Hz).13C NMR (100 MHz, CDCl3): d 191.8, 162.6, 151.4, 150.0, 148.7, 145.8, 138.3, 136.6, 136.2, 131.7, 130.7, 129.8, 129.2, 123.3, 122.5, 122.1, 120.7, 68.8. FAB-MS: calcd. 426.0, m/z¼ 427.1 (M + H+).
7-(Diphenylamino)-9,90-di(2,200
-pyridyl)-2-fluorenecarboxalde-hyde. A mixture of 7-bromo-9,90-di(2,200
-pyridyl)-2-fluo-renecarboxaldehyde (1.92 g, 4.5 mmol), diphenylamine (1.14 g,
6.75 mmol), Pd(OAc)2(25 mg, 0.113 mmol), P(tBu)3(47 mg, 0.23
mmol), and Cs2CO3(3.3 g, 10.1 mmol) in toluene (300 mL) was
reacted in a similar fashion to that of preparing 4,5-diaza-20
-diphenylamino-9,90-spirobifluorene-70-carboxaldehyde, except
ª The Royal Society of Chemistry 2009
the reaction time was extended to 3 days, and ethyl acetate/ dichloromethane (1/4) was used as the eluent for flash column chromatography. A yellow solid was obtained with a yield of
77% (1.78 g).1H NMR (400 MHz, CDCl 3): d 9.95 (s, 1H), 8.53– 8.50 (m, 2H), 7.24 (d, 1H, J¼ 1.0 Hz), 7.91 (dd, 1H, J ¼ 7.9, 1.4 Hz), 7.77 (d, 1H, J¼ 7.9 Hz), 7.65 (d, 1H, J ¼ 8.4 Hz), 7.57 (d, 1H, J¼ 1.8 Hz), 7.48 (td, 2H, J ¼ 7.8, 1.9 Hz), 7.23–7.17 (m, 4H), 7.12–6.98 (m, 11H).13C NMR (100 MHz, CDCl 3): d 191.9, 163.2, 151.0, 149.8, 149.2, 148.6, 147.3, 146.9, 136.4, 135.0, 133.0, 129.8, 129.3, 129.2, 124.7, 123.4, 123.0, 121.9, 121.8 (2), 121.0, 119.7, 68.6. FAB-MS: calcd. 515.6, m/z¼ 516.2 (M + H+).
N-[7-(2,2-Diphenylvinyl)-9,90-di(2,200
-pyridyl)-2-fluorenyl]-N,N-diphenylamine (PhFpy2DPV). A mixture of
7-(diphenylamino)-9,90-di(2,200-pyridyl)-2-fluorenecarboxaldehyde (0.92 g, 1.6
mmol), diethoxydiphenylmethylphosphonate (1.95 g, 6.4 mmol), and NaH (0.28 g, 60% w/w dispersion in mineral oil, 7.0 mmol) in dry THF (40 mL) was reacted in a similar fashion to that of
preparing 4,5-diaza-20-diphenylamino-70-(2,20
-diphenylvinyl)-9,90-spirobifluorene, except the reaction time was extended to
overnight, and ethyl acetate/hexanes (1/4) was used as the eluent for flash column chromatography. The remaining
diethoxy-diphenylmethylphosphonate can be removed after
sub-limination. A yellow-green solid was obtained with a yield of 67%
(0.71g).1H NMR (400 MHz, CDCl 3): d 8.43 (dd, 2H, J¼ 4.7, 0.8 Hz), 7.59 (d, 1H, J¼ 1.3 Hz), 7.50 (d, 1H, J ¼ 8.2 Hz), 7.47–7.40 (m, 4H), 7.30–7.20 (m, 8H), 7.19–7.13 (m, 6H), 7.06–6.98 (m, 9H), 6.94 (t, 2H, J¼ 7.3 Hz), 6.87 (d, 2H, J ¼ 8.0 Hz).13C NMR (100 MHz, CDCl3): d 163.8, 150.1, 149.5, 147.7, 147.6, 147.5, 143.5, 142.1, 140.4, 139.3, 136.3, 136.1, 134.9, 130.5, 129.3, 129.1, 128.7, 128.5 (2), 128.1, 127.5, 127.4, 127.3, 124.1, 123.7, 123.1, 122.7, 121.4, 121.0, 120.4, 119.1, 68.6. FAB-MS: calcd. 665.3, m/z
¼ 666.2 (M + H+). Anal. calcd for C
49H35N3: C 88.39, H 5.30, N
6.31; found: C 88.13, H 5.40, N 6.01.
X-Ray crystallography studies†
Data collection was carried out on a Brucker X8APEX CCD
diffractometer at 100 K for PhSPN2DPV and PhFpy2DPV single
crystals. The radiation of Mo radiation (l¼ 0.71073 A˚ ) was used
for both crystals. The unit cell parameters were obtained by a least-square fit to the automatically centered settings for reflections. Intensity data were collected by using the u/2q scan mode. Corrections were made for Lorentz and polarization effects. The structures were solved by direct methods
SHELX-97.26All non-hydrogen atoms were located from the difference
Fourier maps and were refined by full-matrix least-squares procedures. The position of hydrogen atoms was calculated and located. Calculations and full-matrix least-squares refinements
were performed utilizing the WINGX program package27in the
evaluation of values of R (Fo) for reflections with I > 2s(I) and
Rw (Fo), where R ¼ S||Fo| |Fc||/S|Fo| and Rw¼ [S{w(Fo2
Fc2)2}/S{w(Fo2)2}]1/2. Intensities were corrected for absorption.
OLED fabrication and electroluminescence (EL) characterization
The fabrication of OLEDs and their EL characterization have
been described elsewhere.28 The current density-voltage-light
intensity (J–V–L) measurements were made simultaneously using a Keithley 2400 programmable source meter and a New-port 1835C Optical meter equipped with a NewNew-port 818-ST silicon photodiode. The device was placed close to the photo-diode such that all the forward light went to the photophoto-diode. The
effective size of the emitting diode was 3.14 mm2, which is
significantly smaller than the active area of the photodiode detector, a condition known as ‘‘under filling’’ to satisfy the
measurement protocol.29Only light emitting from the front face
of the devices was collected and used in subsequent calculations
of external quantum efficiency (hEXT) according to the method
described earlier.29 The luminous flux (lm) has been defined
previously,30and we adopted it to estimate the power efficiency
(hP) of OLEDs.
Acknowledgements
This research was supported in part by the National Science Council of Taiwan, National Chiao Tung University, and Academia Sinica. The authors also thank Prof. Chao-Ping Hsu for helpful instruction and discussion in theoretical calculation.
References
1 (a) A. K€ohler, J. S. Wilson and R. H. Friend, Adv. Mater., 2002, 14, 701; (b) S. R. Forrest, Org. Electron., 2003, 4, 45.
2 (a) Y. Sun, N. C. Giebink, H. Kanno, B. Ma, M. E. Thompson and S. R. Forrest, Nature, 2006, 440, 908; (b) P.-I. Shih, C.-F. Shu, Y.-L. Tung and Y. Chi, Appl. Phys. Lett., 2006, 88, 251110; (c) G. Schwartz, K. Fehse, M. Pfeiffer, K. Walzer and K. Leo, Appl. Phys. Lett., 2006, 89, 083509; (d) H. Kanno, Y. Sun and S. R. Forrest, Appl. Phys. Lett., 2006, 89, 143516; (e) B.-P. Yan, C. C. C. Cheung, S. C. F. Kui, H.-F. Xiang, V. A. L. Roy, S.-J. Xu and C.-M. Che, Adv. Mater., 2007, 19, 3599; (f) G. Schwartz, M. Pfeiffer, S. Reineke, K. Walzer and K. Leo, Adv. Mater., 2007, 19, 3672.
3 (a) A. B. Chwang, M. Hack and J. L. Brown, J. Soc. Inform. Display, 2005, 13, 481; (b) M. S. Weaver, R. C. Kwong, V. A. Adamovic, M. Kack and J. J. Brown, J. Soc. Inform. Display, 2006, 14, 449. 4 (a) R. J. Holmes, B. W. D’Andrade, S. R. Forrest, X. Ren, J. Li and
M. E. Thompson, Appl. Phys. Letter., 2003, 83, 3818; (b) S.-J. Yeh, M.-F. Wu, C.-T. Chen, Y.-H. Song, Y. Chi, M.-H. Ho, S.-F. Hsu and C. H. Chen, Adv. Mater., 2005, 17, 285; (c) C. S. K. Mak, A. Hayer, S. I. Pascu, S. E. Watkins, A. B. Holmes, A. Kohler and R. Friend, Chem. Commun., 2002, 4708; (d) X. Zhang, C. Jiang, Y. Mo, Y. Xu, H. Shi and Y. Cao, Appl. Phys. Letter., 2006, 88, 051116; (e) L.-L. Wu, C.-H. Yang, I.-W. Sun, S.-Y. Chu, P.-C. Kao and H.-H. Huang, Organometallic, 2007, 26, 2017; (f) P.-I. Shih, C.-H. Chien, C.-Y. Chuang, C.-F. Shu, C.-C.-H. Yang, J.-C.-H. Chen and Y. Chi, J. Mater. Chem., 2007, 17, 1692; (g) L. Chen, H. You, C. Yang, D. Ma and J. Qin, Chem. Commun., 2007, 1352; (h) E. Orselli, G. S. Kottas, A. E. Konradsson, P. Coppo, R. Fr€ohlich, L. De Cola, A. van Dijken, M. B€uchel and H. B€orner, Inorg. Chem., 2007, 46, 11082.
5 (a) S.-J. Yeh, C.-T. Chen, Y.-H. Song, Y. Chi and M.-H. Ho, J. Inform Display, 2005, 13, 857; (b) R. J. Holmes, S. R. Forrest, T. Sajoto, A. Tamayo, P. I. Djurovich, M. E. Thompson, J. Books, Y.-J. Tung, B. W. D’Andrade, M. S. Weaver, R. C. Kwong and J. J. Brown, Appl. Phys. Lett., 2005, 87, 243507; (c) C.-H. Yang, Y.-M. Cheng, Y. Chi, C.-J. Hsu, F.-C. Fang, K.-T. Wong, P.-T. Chou, C.-H. Chang, M.-H. Tsai and C.-C. Wu, Angew. Chem. Int. Ed., 2007, 46, 2418; (d) C.-F. Chang, Y.-M. C heng, Y. Chi, Y.-C. Chiu, C.-C. Lin, G.-H. Lee, P.-T. Chou, C.-C. Chen, C.-H. Chang and C.-C. Wu, Angew. Chem. Int. Ed., 2008, 47, 4542.
6 H. Y. Chen, W. Y. Lam, J. D. Luo, Y. L. Ho, B. Z. Tang, D. B. Zhu, M. Wong and H. S. Kwok, Appl. Phys. Letter., 2002, 81, 574. 7 (a) C.-C. Wu, Y.-T. Lin, K.-T. Wong, R.-T. Chen and Y.-Y. Chien,
Adv. Mater., 2004, 16, 61; (b) M.-T. Lee, C.-H. Liao, C.-H. Tsai
ª The Royal Society of Chemistry 2009
and C. H. Chen, Adv. Mater., 2005, 17, 2493; (c) M.-F. Li, L. Wang, W.-K. Wong, K.-W. Cheah, H.-L. Tam, M.-T. Lee and C. H. Chen, Appl. Phys. Lett., 2006, 89, 121913; (d) H.-C. Li, Y.-P. Lin, P.-T. Chou, Y.-M. Cheng and R.-S. Liu, Adv. Funct. Mater., 2007, 17, 520; (e) C. J. Tonzola, A. P. Kulkarni, A. P. Gifford, W. Kaminsky and S. A. Jenekhe, Adv. Funct. Mater., 2007, 17, 863; (f) Y. Wai and C.-T. Chen, J. Am. Chem. Soc., 2007, 129, 7478; (g) P.-I. Shih, C.-Y. Chuang, C.-H. Chien, E. W.-G. Diau and C.-F. Shu, Adv. Funct. Mater., 2007, 17, 3141; (h) K.-C. Wu, P.-J. Ku, C.-S. Lin, H.-T. Shih, F.-I. Wu, M.-J. Huang, J.-J. Lin, I.-C. Chen and C.-H. Cheng, Adv. Funct. Mater., 2008, 18, 67; (i) S.-K. Kim, B. Yang, Y. Ma, J.-H. Lee and J.-W. Park, J. Mater. Chem, 2008, 18, 3376; (j) Y.-Y. Lyu, J. Kwak, O. Kwon, S.-H. Lee, D. Kim, C. Lee and K. Char, Adv. Mater., 2008, 20, 2720; (k) S.-L. Lin, L.-H. Chan, R.-H. Lee, M.-Y. Yen, W.-J. Kuo, C.-T. Chen and R.-J. Jeng, Adv. Mater., 2008, 20, 3947.
8 (a) S.-C. Chang, G. He, F.-C. Chen, T.-F. Guo and Y. Yang, Appl. Phys. Lett., 2001, 79, 2088; (b) C.-C. Wu, T.-L. Liu, W.-Y. Hung, Y.-T. Lin, K.-T. Wong, R.-T. Chen, Y.-M. Chen and Y. Y. Chien, J. Am. Chem. Soc., 2003, 125, 3710.
9 S. C. Tse, K. C. Kwok and S. K. So, Appl. Phys. Lett., 2006, 89, 262102.
10 C.-L. Chiang, S.-M. Tseng, C.-T. Chen, C.-P. Hsu and C.-F. Shu, Adv. Func. Mater., 2008, 18, 248.
11 (a) K.-T. Wong, R.-T. Chen, F.-C. Fang, C.-C. Wu and Y.-T. Lin, Org. Lett., 2005, 7, 1979; (b) K.-T. Wong, R.-T. Chen, F.-C. Fang, C.-C. Wu and Y.-T. Lin, Org. Lett., 2005, 7, 5925; (c) K.-T. Wong, H.-F. Chen and F.-C. Fang, Org. Lett., 2006, 8, 3501.
12 (a) F. D. Libman and R. Slack, J. Chem. Soc., 1951, 2588; (b) F. Dierschke, A. C. Grimsdale and K. M€ullen, Synthesis, 2003, 2470. 13 C.-L. Chiang, C.-F. Shu and C.-T. Chen, Org. Lett., 2005, 7, 3717. 14 Crystal data for PhSPN2DPV$CH2Cl2: C50H35Cl2N3: Fw¼ 748.71,
monoclinoc, P21/c, Z¼ 4, F(000) ¼ 1560. Cell dimendions: a ¼
12.5133(3) A˚ , b ¼ 34.5307(8) A˚, c ¼ 8.9169(2) A˚, a ¼ 90,
b¼ 100.493(2), g¼ 90, V¼ 3788.50(15) A˚3, 2q
max¼ 50.0, rcacld
¼ 1.313 mg m3. Of 27836 reflections, 6695 were independence, 497
parameters, R (Fo)¼ 0.0392 (for reflections with I > 2s(I)), Rw(Fo)
¼ 0.0801 (for reflections with I > 2s(I)). The GoF on F2was equal
0.776. Crystal data for PhFpy2DPV: C49H35N3: Fw ¼ 665.80,
monoclinoc, P21/c, Z¼ 4, F(000) ¼ 1400. Cell dimendions: a ¼
13.1622(6) A˚ ;, b ¼ 25.6042(11) A˚;, c ¼ 10.9596(5) A˚;, a ¼ 90,
b¼ 107.046(2), g¼ 90, V¼ 3531.2(3) A˚3, 2q
max¼ 50.1, rcacld¼
1.252 mg m3. Of 27399 reflections, 6169 were independence, 470
parameters, R (Fo)¼ 0.0395 (for reflections with I > 2s(I)), Rw(Fo)
¼ 0.0735 (for reflections with I > 2s(I)). The GoF on F2was equal
0.685. CCDC-682902 and CCDC-682903 contain the supplementary
crystallographic data of PhSPN2DPV$CH2Cl2 and PhFpy2DPV,
respectively. These data can be obtained free of charge via www.ccd.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic Data Centre, 12, Union Road, Cambridge CB21EZ, UK; fax: (+44)1223-336-033; or [email protected]). 15 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
16 M. E. Casida, In Recent Advances in Density Functional Methods, Part 1; D. P. Chong, Ed.; World Scientific: Singapore, 1995; p 155. 17 J. Kong, C. A. White, A. I. Krylov, C. D. Sherrill, R. D. Adamson,
T. R. Furlani, M. S. Lee, A. M. Lee, S. R. Gwaltney, T. R. Adams, C. Ochsenfeld, A. T. B. Gilbert, G. S. Kedziora, V. A. Rassolov, D. R. Maurice, N. Nair, Y. Shao, N. A. Besley, P. E. Maslen, J. P. Dombroski, H. Daschel, W. Zhang, P. P. Korambath, J. Baker, E. F. C. Byrd, T. Van Voorhis, M. Oumi, S. Hirata, C.-P. Hsu, N. Ishikawa, J. Florian, A. Warshel, B. G. Johnson, P. M. W. Gill, M. Head-Gordon and J. A. Pople, J. Comput. Chem., 2000, 21, 1532.
18 K. Araki, L. Angnes and H. E. Toma, Adv. Mater., 1995, 7, 554. 19 M.-F. Wu, S.-J. Yeh, C.-T. Chen, H. Murayama, T. Tsuboi, W.-S. Li,
I. Chao, S.-W. Liu and J.-K. Wang, Adv. Funct. Mater., 2007, 17, 1887.
20 M.-Y. Kuo, H.-Y. Chen and I. Chao, Chem. Eur. J., 2007, 13, 4750. 21 S. C. Tse, K. C. Kwok and S. K. So, Appl. Phys. Lett., 2006, 89,
262102.
22 (a) Z.-Y. Xie, T.-C. Wong, L.-S. Hung and S.-T. Lee, Appl. Phys. Letter., 2002, 80, 1477; (b) Y. Li, M. K. Fung, Z. Xie, S.-T. Lee, L.-S. Hung and J. Shi, Adv. Mater., 2002, 14, 1317.
23 (a) J. C. de Mello, H. F. Wittmann and R. H. Friend, Adv. Mater., 1997, 9, 230; (b) C.-L. Chiang, M.-F. Wu, D.-C. Dai, Y.-S. Wen, J.-K. Wang and C.-T. Chen, Adv. Funct. Mater., 2005, 15, 231. 24 (a) T. Yamamoto, M. Nishiyama and Y. Koie, Tetragedron Lett.,
1998, 39, 2367; (b) J. Shi, C. W. Tang and C. H. Chen, U.S. Patent No. 5645948, 1997.
25 (a) D. L. White and J. W. Faller, Inorg. Chem., 1982, 21, 3119; (b) A. J. Canty, N. J. Minchin, L. M. Engelhardt, B. W. Skelton and A. H. White, Aust. J. Chem., 1988, 41, 651.
26 G. M. Sheldrick, SHELXL-97, University of G€ottingen, Germany, 1997.
27 L. J. Farrugia, J. Appl. Cryst., 1999, 32, 837.
28 L.-H. Chan, R.-H. Lee, C.-F. Hsieh, H.-C. Yeh and C.-T. Chen, J. Am. Chem. Soc., 2002, 124, 6469.
29 S. R. Forrest, D. D. C. Bradley and M. E. Thompson, Adv. Mater., 2003, 15, 1043.
30 G. Wyszecki, W. S. Stiles, In Color Science: Concepts and Methods, Quantitative Data and Formulae; John Wiley & Sons: New York, 1982; p 259.
ª The Royal Society of Chemistry 2009
