Phosphazene-based host materials for the use in blue phosphorescent organic light-emitting diodes

Article abstract of DOI:10.1021/cm201940f

We present a series of low-molecular-weight materials based on cyclic phosphazenes for the use as host materials in blue phosphorescent organic light-emitting diodes. Substituted phenyl rings are attached to the central phosphazene ring either via phosphorus-oxygen bonds to yield phenoxy-substituted derivatives or via direct phosphorus-carbon bonds to yield phenyl-substituted derivatives. The phenoxy substituted cyclic phosphazenes were prepared by nucleophilic substitution of the six chlorine atoms in hexachlorocyclotriphosphazene with phenoxy groups, whereas the phenyl substituted cyclic phosphazenes were formed in a cyclocondensation reaction of three equivalents of substituted phosphinic amides. The phenyl substitution leads to materials with superior thermal properties compared to the phenoxy substitution. Because of the nonconjugated linkage to the phosphazene core, the host materials have very high triplet energies of more than 3 eV. In an OLED device using one compound as host for the saturated blue phosphorescent emitter Ir(dbfmi), a peak power efficiency of 7.6 lm W^-1 and a peak luminance of 5000 cd m^-2 were achieved.

Full text of DOI:10.1021/cm201940f

Article
pubs.acs.org/cm
Phosphazene-Based Host Materials for the Use in Blue
Phosphorescent Organic Light-Emitting Diodes
Pamela Schrogel,† Matthias Hoping,‡ Wolfgang Kowalsky,‡ Arvid Hunze,§ Gerhard Wagenblast,⊥
̈
Christian Lennartz,⊥ and Peter Strohriegl*,†
†Lehrstuhl Makromolekulare Chemie I, Universitat Bayreuth, 95440 Bayreuth, Germany,
̈
‡Institut fur Hochfrequenztechnik, TU Braunschweig, D-38023 Braunschweig,
̈
§Siemens AG, D-91050 Erlangen, Germany,
⊥BASF SE, D-67056 Ludwigshafen, Germany
ABSTRACT: We present a series of low-molecular-weight materials based on cyclic
phosphazenes for the use as host materials in blue phosphorescent organic light-emitting
diodes. Substituted phenyl rings are attached to the central phosphazene ring either via
phosphorus−oxygen bonds to yield phenoxy-substituted derivatives or via direct phosphorus−
carbon bonds to yield phenyl-substituted derivatives. The phenoxy substituted cyclic
phosphazenes were prepared by nucleophilic substitution of the six chlorine atoms in
hexachlorocyclotriphosphazene with phenoxy groups, whereas the phenyl substituted cyclic
phosphazenes were formed in a cyclocondensation reaction of three equivalents of substituted
phosphinic amides. The phenyl substitution leads to materials with superior thermal properties
compared to the phenoxy substitution. Because of the nonconjugated linkage to the phosphazene core, the host materials have
very high triplet energies of more than 3 eV. In an OLED device using one compound as host for the saturated blue
phosphorescent emitter Ir(dbfmi), a peak power efficiency of 7.6 lm W−1 and a peak luminance of 5000 cd m−2 were achieved.
KEYWORDS: OLED, blue phosphorescence, ultra high band gap host (UGH), cyclic phosphazenes
extremely confined which limits the choice of possible building
blocks. Organosilane compounds have been presented as
materials with ultra high bandgaps (UGH) and successfully
employed in OLEDs as host materials for blue phosphors.9−12
Besides organosilane compounds, several phosphorus con-
taining host materials, such as phosphine oxides,13−15 have
been introduced as promising candidates with high triplet
energies. Another class of phosphorus containing materials
interesting for the use in organic electroluminescent devices are
cyclic phosphazenes. The planar nonconjugated 6-membered
ring consisting of alternating P- and N-atoms reveals high
chemical and thermal stability and exhibits two bonding sites
for substituents on each P-atom. Solution-processable den-
drimers with various emitters attached to a cyclic phosphazene
core have already been reported. Via oxygen−phosphorus
bonds emitting fluorescent pyrene units or phosphorescent
iridium complexes were attached to the phosphazene core and
rigid spherical molecules with high glass transition temperatures
were obtained.16−18 Very recently, the same research group
presented phenoxy-carbazole and pyridin-oxy-carbazole func-
tionalized cyclic phosphazene materials and their use as host
materials for the greenish-blue triplet emitter FIrpic and the
green triplet emitter Ir(mppy)3 in solution processable OLED
devices.19 However, because of their high molecular weight,
INTRODUCTION
■
Since the first efficient multilayer organic light-emitting diode
(OLED) was invented by Tang and van Slyke in 1987, great
advances have been made by extensive development of new
OLED materials and sophisticated device setups.1−3 Nowadays,
OLEDs are attracting attention because of their capability to
realize full color displays and large area lighting with high
power efficiency and freedom of design. The key to high
efficiencies was the discovery of phosphorescent emitters based
on organo-transition metal complexes that are able to harvest
both electrogenerated singlet and triplet excitons for light
emission. Therefore, the theoretical limit of the internal
quantum efficiency reaches 100%.4 In general, the triplet
emitters have to be embedded into a host to avoid
concentration quenching.5 It is essential that the triplet energy
of the host is larger compared to the triplet emitter to prevent
energy back transfer from the emitter to the host and to confine
triplet excitons on the emitter molecules. This restriction
becomes particularly challenging when host materials for deep
blue phosphors are in the focus of interest, where host triplet
energies larger than 3.0 eV are required. Among the commonly
used class of carbazole based host materials, some examples like
the substituted N-phenylcarbazole 9-(4-tert-butylphenyl)-3,6-
ditriphenylsilyl)-carbazole (CzSi, E(T1 − S0) = 3.02 eV)6 and
1,4-bis(4-(carbazol-9-yl)phenyl)cyclohexane (CBPCH, E(T1 −
S0) = 3.01 eV)7 show triplet energies close to their intrinsic
limit of 3 eV.8 To provide host materials with even larger triplet
band gaps the conjugation of the host molecules must be
Received: July 7, 2011
Revised: October 13, 2011
Published: October 28, 2011
© 2011 American Chemical Society
4947
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953

Chemistry of Materials
Article
calorimetry measurements (DSC) a Diamond DSC apparatus from
Perkin-Elmer was used (heating/cooling rate 10 K/min). The purity of
the target compounds was checked with a Waters size exclusion
chromatography system (SEC) for oligomers (analytical columns:
cross-linked polystyrene gel (Polymer Laboratories), length: 2 × 60
cm, width: 0.8 cm, particle size: 5 μm, pore size 100 Å, eluent THF
(0.5 mL/min, 80 bar), polystyrene standard). For the fabrication of
OLEDs the organic layers were deposited by thermal evaporation in
high vacuum (<1 × 10−6 mbar) onto indium−tin-oxide (ITO, 10
ohm/square) precoated glass substrates. The organic layers and the
metal cathode were evaporated without breaking the vacuum.
Computational Methods. The transport levels of the materials
used were determined via density functional calculations. For the
ionization potential and the electron affinity first the geometry of the
neutral as well as the charged states were optimized using the BP86-
functional20,21 in combination with a split-valence basis set (SV(P))
including polarization functions on all heavy atoms.22 For iridium an
effective core potential was employed.23 For the energetics we
performed single point calculations at the optimized geometries using
the same functional in combination with a TZVP-basis set.24 To
account for dielectric solid state effects, we used a UPS/IEPS-
calibrated version of the conductor like screening model (COSMO)25
in conjunction with these single-point calculations. All calculations
were carried out with the turbomole program package.26
these dendrimers are not suitable for an OLED-fabrication by
thermal evaporation.
In this study, we present a series of vacuum processable new
phosphazene based host materials for saturated blue phosphor-
escent emitters. In these materials, substituted phenyl rings
were either attached via phosphorus−oxygen bonds (phenoxy
substituted cyclic phosphazenes) or via phosphorus−carbon
bonds (phenyl substituted cyclic phosphazenes). We present
the synthesis of symmetrically substituted organocyclotriphos-
phazenes together with their thermal and optical properties.
The energy levels of the compounds were calculated by density
functional theory (DFT). To demonstrate the potential as host
material for blue phosphorescent emitters, we have made an
OLED device employing an organocyclotriphosphazene host.
In Figure 1, the molecular structures of the organo-
cyclotriphosphazenes are depicted.
Synthetic Procedures. The synthesis of tris[(3-phenyl-1H-
benzimidazol-1-yl2(3H)-yliden-1,2-phenylene]-irdidium (DPBIC),35
2,8-bis(triphenylsilyl)-dibenzofuran (DBFSi),36 and mer-tris(N-diben-
zofuranyl-N′-methylimidazole)iridium(III) (Ir(dbfmi))27,36 is de-
scribed in literature.
General Procedure for the Synthesis of Phenoxy-Substituted
Phosphazenes. The cyclotriphosphazenes OP1−OP3 were prepared
similar to a procedure described in ref 28.
Hexakisphenoxycyclotriphosphazene (OP1). A suspension of
sodium phenolate was prepared by adding sodium hydride (0.55 g,
23 mmol) to phenole (2.16 g, 23 mmol) dissolved in 40 mL of dry
THF under argon. Hexachlorocyclotriphosphazene (1.0 g, 2.88 mmol)
dissolved in 10 mL of dry THF was added dropwise via syringe and
the mixture was stirred under reflux for two days. After removal of the
solvent under reduced pressure, the residue was stirred in water for 1
h. The white solid was washed with methanol to remove the excess of
phenole. Yield: 1.69 g (85%). EI-MS m/z: 693 (M•+, 71), 599 (100),
Figure 1. Molecular structures of the cyclic phosphazenes (top)
OP1−OP3 with phenoxy substitution and (bottom) CP1−CP3 with
phenyl substitution.
1
505 (34). H NMR (250 MHz, CDCl3), δ (ppm): 7.21−7.08 (m,
18H), 6.93 (d, 12 H).
EXPERIMENTAL SECTION
■
Hexakis(4-methoxyphenoxy)cyclotriphosphazene (OP2). OP2
was synthesized according to the route described above. The reaction
time was 24 h. Yield: 0.62 g (49%). EI-MS m/z: 873 (M•+, 100), 749
Materials. All chemicals and reagents were used as received from
commercial sources without further purification. All solvents for
reactions and purifications were distilled once, except tetrahydrofuran
(THF), which was additionally distilled over potassium.
1
(78), 626 (8). H NMR (250 MHz, CDCl3), δ (ppm): 6.83 (m, 12
H), 6.67 (m, 12Hs), 3.79 (s, 18H).
Characterization. 1H and 13C NMR spectra were recorded with a
Bruker AC 250 (250 MHz) and CDCl3 as solvent. All data are given as
chemical shifts δ (ppm) downfield from Si(CH3)4. Mass spectra were
taken on a Finnigan Mat 8500, Mat 112 S Varian machine using EI-
ionization. For optical measurements, 1 × 10−5 M cyclohexane
(Uvasol, Merck) solutions of the materials as well as spin coated thin
films on quartz substrates were prepared. The UV/vis spectra were
measured in solution and on neat films with a Hitachi U-3000
spectrometer. Fluorescence spectra were measured on a Shimadzu
spectrofluorophotometer RF-5301PC using excitation at 220 nm.
Phosphorescence spectra of thin films have been measured at 5 K
(Helium cryostat Optistat CF from Oxford Instruments) applying the
technique of time-gated spectroscopy (with fluorescence and
phosphorescence lifetime spectrometer FLSP 920 from Edinburgh
Instruments). The quantum yield of photoluminescence (η PL) of the
8% Ir(dbfmi)-doped films of CP1-CP3 was measured under N2 flow
using the absolute quantum yield measurement system, model
C9920−02 (Hamamatsu), which is based on an integrating sphere,
monochromatic excitation at 355 nm and a fibreoptic detector with a
multichannel spectrometer. Thermogravimetric analysis (TGA) was
performed on a Mettler Toledo TGA/SDTA815e machine at a heating
rate of 10 K/min in a nitrogen atmosphere. For differential scanning
Tris(phenylenedioxy)cyclotriphosphazene (OP3). Fifteen millili-
ters of dry THF were added to sodium hydride (1 g, 41.4 mmol)
under argon. Catechol (1.9 g, 17.3 mmol) dissolved in 15 mL of dry
THF was added dropwise to the NaH solution. After being stirred for
90 min at room temperature, the blue mixture was transferred into a
dropping funnel and added dropwise to a solution of hexachlor-
ocyclotriphosphazene (2.0 g, 5.75 mmol) dissolved in 10 mL of dry
THF. After stirring at room temperature for 24 h, the same
purification procedure as described above was used. Yield: 1.88 g
1
(71%). EI-MS m/z: 459 (M•+, 100), 351 (12). H NMR (250 MHz,
CDCl3), δ (ppm): 7.12−7.00 (m, 24 Hs).
General Procedure for the Synthesis of Phenyl Substituted
Phosphazenes. Bis(3,5-dimethylphenyl)phosphinic Acid (PAc2).
1-Bromo-3,5-dimethylbenzene (13.63 g, 71.44 mmol) was added
dropwise to magnesium chips (1.74 g, 71.44 mmol) in absolute THF
(80 mL) at room temperature. Once the magnesium was completely
dissolved (2 h), a solution of N,N-dimethyl-phosphoryl dichloride
N(Me)2P(O)Cl2 (4.33 mL, 35.72 mmol) in absolute THF (10 mL)
was added dropwise by syringe. After 2 h the mixture was poured into
an ice-cooled solution of NH4Cl (30 g) in water (500 mL) for the
separation of the formed salt. In a distillation setup with Vigreux
column, the THF was distilled off and the residue was treated with
4948
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953

Chemistry of Materials
Article
conc. HCl (75 mL) at 80 °C. The resulting white solid was dissolved
in aqueous NaOH (5.4 g, 300 mL H2O), and the aqueous layer was
twice extracted with ether. The clear aqueous layer was again acidified
by the addition of conc. HCl to precipitate the phosphinic acid PAc2
as white solid. Yield: 6.81 g (70%). EI-MS m/z: 274 (M•+, 100), 259
1
(39), 136 (5). H NMR (250 MHz, CDCl3), δ (ppm): 11.35 (s, 1H,
OH), 7.35 (d, 4H), 7.11 (s, 2H), 2.28 (m, 12H, CH3).
Bis(4-methoxyphenyl)phosphinic Acid (PAc3). The same syn-
thetic procedure as described above was used. Yield: 54% (5.36 g). EI-
MS m/z: 278 (M•+, 100), 263 (19), 247 (8), 107 (15).1H NMR (250
MHz, CDCl3), δ (ppm): 7.77 (br s, 1H, OH), 7.62 (dd, 4H), 6.83
(dd, 4H), 3.79 (m, 6H).
Bis(3,5-dimethylphenyl)phosphinic Amide. (PAm2). Distilled thi-
onyl chloride (0.87 mL, 12 mmol) was added dropwise to a
suspension of bis(3,5-dimethylphenyl)phosphinic acid (1 g, 3.65
mmol) in dry toluene (25 mL) at 55−60 °C and stirred for 20 min.
After the removal of the excess of thionyl chloride by distillation
ammonia gas was passed into the clear solution for 10 min at 0 °C. In
order to separate the NH4Cl the organic layer was washed with water.
Before the extraction with THF the density of the water layer was
increased by the addition of sodium chloride. The combined organic
layers were dried over Na2SO4 and the solvent was removed under
reduced pressure. The phosphinic amide PAm2 was used for the
cyclization reaction without further purification. Yield: 0.98 g (98%).
EI-MS m/z: 273 (M•+, 100), 257 (19).1H NMR (250 MHz, CDCl3),
δ (ppm): 7.53 (d, 4H), 7.11 (s, 2H), 3.01 (br s, 2H, NH2), 2.32 (m,
12H).
Bis(3,5-dimethylphenyl)phosphinic Amide (PAm3). The same
synthetic pathway as described above was used. Yield: 90% (0.89 g).
EI-MS m/z: 277(M+, 100), 170 (45). 1H NMR (250 MHz, CDCl3), δ
(ppm): 7.83 (m, 4H), 6.92 (m, 4H), 3.81 (m, 3H), 3.08 (br s, 2 H,
NH2).
Hexaphenylcyclotriphosphazene (CP1).29 A suspension of diphe-
nylphosphinic amide (2.0 g, 9.22 mmol), triphenylphosphine (2.9 g,
11.1 mmol), carbon tetrachloride (0.89 mL, 9.22 mmol) and
triethylamine (1.3 mL, 9.22 mmol) in dry dichloromethane (25 mL)
was refluxed for 5 h. The solvent was removed under reduced pressure.
Liquid chromatography on silica gel with hexane:THF mixtures as
eluent gave the product as yellowish solid. Yield: 65% (1.2 g). EI-MS
1
m/z: 597 (M•+, 100), 520 (74), 299 (12), 260 (18), 77 (5). H NMR
Figure 2. Synthetic route to (a) phenoxy-substituted cyclic
phosphazenes OP1−OP3 and (b) phenyl-substituted cyclic phospha-
zenes CP1−CP3. Reagents and conditions: (i) NaH, abs. THF, 70 °C,
48 h (catechol: RT, 24 h); (ii) RMgBr, abs. THF, RT, 2 h; (iii) conc.
HCl, 80 °C, 1 h; (iv) thionyl chloride, abs. toluene, 55 °C, 20 min; (v)
NH3 (g), 0 °C, 10 min; (vi) PPh3, CCl4, NEt3, benzene,
dichloromethane, 40 °C, 5−22 h.
(250 MHz, CDCl3), δ (ppm): 7.76 (m, 12H), 7.30 (m, 18H). 13C
NMR (CDCl3), δ (ppm): 138.55 (dt, CP‑bound), 130.86 (ps d, Cmeta),
130.44 (m, Cpara), 127.98 (ps d, Cortho). Elemental anal. Calcd (%): C,
72.36; H, 5.06; N, 7.03; P, 15.55. Found: C, 72.38; H, 5.08; N, 7.07; P,
15.49.
Hexa-(3,5-dimethylphenyl)-cyclotriphosphazene (CP2). The
same procedure for the synthesis as described for CP1 was applied
except the reaction time was extended to 22 h. Yield: 65% (1.2 g). EI-
MS m/z: 765 (M•+, 100), 660 (45), 382 (10), 105 (10).1H NMR (250
MHz, CDCl3), δ (ppm): 7.40 (d, 12H), 6.98 (s, 6H), 2.21 (m, 36H).
13C NMR (CDCl3), δ (ppm): 138.35 (dt, CP‑bound), 137.16 (ps d,
are formed which replace the chlorine atoms by nucleophilic
substitution. The synthesis of the phenyl-substituted derivatives
by direct replacement of the six chlorine atoms in
hexachlorocyclotriphosphazene is rather difficult. Both an
approach by Grignard reaction and by Friedel−Crafts arylation
are known to yield mixtures of several substitution products
with only small amounts of the hexa-substituted com-
pounds.30,31 Therefore, we devised the synthetic route to
phenyl substituted organocyclotriphosphazenes as sketched in
Figure 2b. In a Grignard reaction two chlorine atoms of
dichloro-N,N-dimethylphosphinic amide are replaced by the
substituted phenyl rings. Acidic workup yields the phenyl
substituted phosphinic acids.32 After treatment with thionyl
chloride the intermediately formed phosphinic acid chlorides
are directly converted into the substituted phosphinic amides in
an amination reaction with gaseous ammonia. The organo-
phosphazenes CP1−CP3 are formed in a cyclocondensation
reaction described by Appel et al.29 from 3 equiv. of a phenyl
substituted phosphinic amide.
Cmeta), 131.97 (d, Cpara), 128.90 (d, Cortho) 21.34 (s, methyl).
Elemental anal. Calcd (%): C, 75.27; H, 7.11; N, 5.49; P, 12.13.
Found: C, 75.14; H, 7.04; N, 5.54; P, 12.18.
Hexa-(4-methoxyphenyl)-cyclotriphosphazene (CP3). The same
procedure for the synthesis as described for CP1 was applied but the
reaction time was extended to 14 h. Yield: 50% (520 mg). EI-MS m/z:
1
777 (M•+, 99), 670 (45), 388 (22). H NMR (250 MHz, CDCl3), δ
(ppm): 7.64 (m, 12H), 6.80 (m, 12H), 3.78 (m, 18H). 13C NMR
(CDCl3), δ (ppm): 161.24 (m, Cpara), 132.71 (ps d, Cmeta), 130.84 (dt,
CP‑bound), 113.39 (ps d, Cortho), 55.31 (s, methoxy). Elemental anal.
Calcd (%): C, 64.86; H, 5.44; N, 5.40; O, 12.34; P, 11.95. Found: C,
65.28; H, 5.44; N, 5.76; O, 12.69; P, 10.81.
RESULTS AND DISCUSSION
■
Synthesis. The phenoxy-substituted phosphazenes OP1−
OP3 have been prepared by nucleophilic substitution of the six
chlorine atoms in hexachlorocyclotriphosphazene (Figure 2a).
By treatment of the phenols with sodium hydride phenoxylates
Thermal Analysis. We examined the thermal properties of
the organophosphazenes OP1-OP3 and CP1-CP3 by thermal
4949
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953

Chemistry of Materials
Article
gravimetric analysis (TGA) and differential scanning calorim-
etry (DSC) in a nitrogen atmosphere at a scanning rate of 10 K
min−1.
low glass transition temperatures of OP1 and OP2, no
morphologically stable films can be prepared. In contrast,
OP3 reveals higher transition temperatures; however, OP3
cannot be sublimed without decomposition. Because of their
poor thermal properties, the phenoxy substituted cyclic
phosphazenes OP1−OP3 are not further considered in this
study.
Optical Analysis. For the optical characterization, we
measured the UV/vis and photoluminescence spectra of
CP1−CP3 in thin films and in cyclohexane solutions (see
Figure 3). All materials have an absorption edge in the UV
For the application in small molecule OLEDs the materials
have to be processable by thermal evaporation. Therefore, the
molecular weight of the compounds should not be too high to
exclude decomposition during evaporation. As the cyclic
phosphazene core bears six substituents it is particularly
important to choose confined aromatic units as substituents.
In this study, we concentrate on phenyl substituents on the
inorganic phosphazene core. All results of the thermal
characterization are listed in Table 1.
a
Table 1. Thermal Properties of OP1-3 and CP1-3
b
Tm (°C)
Tcr (°C)
Tg (°C)
TID (°C)
c
OP1
OP2
OP3
CP1
CP2
CP3
116
107
251
235
214
281
38
−12
1
260
314
271
285
280
337
173
191
138
221
d
68
d
73
a
Tm, melting temperature; Tcr, crystallization temperature; Tg, glass
transition temperature; and TID, initial decomposition temperature.
b
Figure 3. Absorption spectra of thin films of CP1−CP3.
TID is the temperature at which an initial loss of mass was observed in
a thermogravimetric experiment with a heating rate of 10 K min−1
c
d
region between 277 and 295 nm. While the absorption of CP1
and CP2 are very similar, the absorption of CP3 is of a different
shape and is observed at higher wavelengths. We assign this to
the electron donating effect of the methoxy substituents at the
para-positions of the phenyl rings. At room temperature, we did
not detect any photoluminescence neither in solution nor in
film samples. From the wavelength of the absorption edge the
optical band gap ΔE(S0 − S1) can be calculated. All cyclic
phosphazenes reveal very large band gaps in the range of 4.2 to
4.5 eV. Because the exchange energy of many small molecules is
typically in the range of 0.7 to 1 eV,33 a large triplet energy is
expected for CP1−CP3.
under nitrogen. Observed only during the heating scan. Observed
only during heating scan after quenching from melt.
The phenoxy substituted cyclic phosphazenes OP1−OP3 are
thermally stable up to at least 260 °C. However, the high
rotational freedom given by the oxygen-linkage leads to very
low glass transition temperatures of −12 and 1 °C for
compounds OP1 and OP2. Thus, films of OP1 and OP2
crystallize readily. In the case of OP3, the bifunctional
substituents introduce more structural rigidity leading to a
much higher melting temperature of 251 °C. Yet, the material
does not sublime without decomposition.
At low temperatures (5 K), we were able to detect
phosphorescence. The phosphorescence spectra are shown in
Figure 4. The emission maxima are centered around 395 nm for
We prepared a second series of cyclic phosphazenes with the
substituents being attached via direct phosphorus−carbon
bonds (CP1−CP3). All materials show high thermal stability
up to at least 280 °C in the TGA measurements and much
higher melting temperatures than their phenoxy substituted
analogues. CP1 melts at 235 °C and crystallizes at 191 °C upon
cooling. The observed low supercooling of only 44 °C
demonstrates the high tendency of CP1 to crystallize. CP2
reveals a melting peak at 214 °C and crystallizes upon cooling
at 138 °C. Compared to CP1 the crystallization tendency could
be slightly decreased by introducing additional methyl groups
as evident from the higher supercooling of 76 °C. CP3 melts at
a temperature of 281 °C and crystallizes at 221 °C upon
cooling. Contrary to the twelve methyl groups in CP2, the six
methoxy groups in CP3 cause only a slight hindrance of
crystallization compared to hexaphenylcyclotriphosphazene
CP1 which is indicated by a supercooling of only 54 °C. In a
second DSC experiment the isotropic melts of the samples
were quenched with liquid nitrogen before measuring the glass
transition temperatures (Tg). The pretreated samples of CP2
and CP3 revealed Tg values of 68 and 73 °C, respectively.
Although CP1−CP3 show crystalline behavior in the DSC
measurements amorphous films can be prepared from solution
and evaporation which are morphologically stable over several
months at room temperature. To summarize, because of the
Figure 4. Phosphorescence spectra of CP1−CP3 measured on neat
films at 5 K; the arrows indicate the shoulder in the emission spectra
taken for the determination of the triplet energy of the compounds.
CP1 and 410 nm for CP2 and CP3. The highest energy peak is
visible as a small shoulder at 363 nm for CP1 and 382 nm for
CP2 and CP3 corresponding to very high triplet energies of 3.4
and 3.2 eV, respectively. All results from the photo physical
investigations are summarized in Table 2.
4950
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953

Chemistry of Materials
Article
Table 2. Optical Properties of the Host Materials CP1-CP3
a
b
c
d
d
e
λ
(nm)
ΔE(S0-S1) (eV)
λ5K (nm)
ΔE(T1-S0) (eV)
HOMO (eV)
LUMO (eV)
ηPL (%)
EA
CP1
CP2
CP3
277
288
295
4.5
4.3
4.2
363
382
382
3.4
3.2
3.2
6.47
5.98
5.71
1.53
1.35
1.14
71
73
63
a
b
Absorption edge measured on neat films at room temperature. The optical band gap was estimated from the onset of the UV/vis absorption of
c
d
neat films. Wavelength of the highest energy shoulder in the phosphorescence spectra of 100% films at 5 K. Energy levels obtained from DFT-
calculations. Quantum yield of photoluminescence measured on film samples of 8% Ir(dbfmi) doped into CP1−CP3.
e
Considering the spatial molecular structure of CP1 helps to
explain the experimental findings. As evident from the
geometry optimized molecular structure shown in Figure 5
To test the ability of the host materials CP1−CP3 to confine
triplet excitons on the emitter molecules we measured the
quantum yield of photoluminescence (η PL) on film samples of
the saturated blue phosphorescent emitter Ir(dbfmi)27 (ΔE(T1
− S0) = 2.75 eV, λem= 450 nm) doped into the host materials
CP1−CP3. All doped films exhibit high PL quantum yields in
the range of 63−73% (cf. Table 2), which demonstrates the
ability of CP1−CP3 to act as efficient host material for the
saturated blue emitter Ir(dbfmi).
Single-Carrier Devices. To get a more comprehensive
understanding about the charge-carrier transport properties of
the novel class of host materials, we prepared single-carrier
devices. Because of its high morphological stability, we chose
CP2 for the single-carrier-device tests. Here, the material is
sandwiched in varying thicknesses between two chemically
doped hole or electron transport layers with high mobilities for
holes or electrons. The J−V characteristics and the device setup
are shown in Figure 6.
CP2 is a good hole conductor and a current density of 1 mA
cm−2 is achieved at 1.6 V for holes. For electrons the current
density of 1 mA cm−2 is achieved at an even lower voltage of 1
V. The J−V characteristics show that CP2 has balanced
transport properties for both types of charge carriers.
Phosphorescent Organic Light-Emitting Diodes.
While common blue phosphors as for example FIrpic and
FIr6 show blue emission tailing to the green region with
emission maxima at 472 nm (CIE color coordinates x = 0.17, y
= 0.34)35 and 458 nm (CIE color coordinates x = 0.16, y =
0.26),10 respectively, phosphors with a pure blue emission
would be beneficial for solid-state lighting. Because of the high
triplet energy of more than 3 eV, the phosphazene-based host
materials should be suitable for such saturated blue emitters.
Thus, we employed CP2 in an OLED device as host material
Figure 5. Geometry-optimized molecular structure of CP1.
the pendant phenyl rings are aligned above and below the
central plane formed by the P3N3 ring. The calculated dihedral
angles between the outer phenyl rings and the central ring
plane are 48 to 56°. The calculated dihedral angles and bond
lengths are consistent with experimental crystallographic data of
CP1.34 As no conjugation over the central ring takes place, all
phenyl rings are electronically isolated from each other. Thus,
in the experiments, the absorption and emission properties of
isolated substituted phenyl rings are observed with nearly no
influence of the P3N3 core.
In Table 2, the energy levels of CP1-CP3 obtained from
DFT-calculations are included. By introducing electron-
donating methyl or methoxy groups the HOMO level shifts
from 6.47 eV for CP1 to 5.98 eV (CP2) and 5.71 eV (CP3),
respectively.
Figure 6. J−V characteristics of CP2 in e-only device (left) and h-only device (right).
4951
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953

Chemistry of Materials
Article
for the pure blue carbene emitter Ir(dbfmi)27 with a λem = 450
nm.
The device structure is shown in Figure 7. On top of the
indium−tin-oxide (ITO) anode 120 nm of α-NPD p-doped
Figure 8. (Top) Current density−voltage characteristics (open
squares) and luminance−voltage plot (filled squares). (Bottom)
Power efficiency−luminance plot. Inset: EL spectrum of the device.
Table 3. Peak performance data and at 100 cd m−2 and 1000
cd m−2 using CP2 as host materials for the blue emitter
Ir(dbfmi)
turn on
voltage
(V)
luminous efficiency (cd A−1
power efficiency (lm
max, at 100 cd m−2, at 1000
W
−1) max, at 100 cd
CIE x,
Figure 7. Energy level diagram of the device with CP2 as host for
Ir(dbfmi) and the molecular structures of all materials. Ionization
potentials and electron affinity levels of the materials are indicated.
The dotted lines represent the levels of the emitter Ir(dbfmi). All
HOMO and LUMO levels were taken from density functional theory
(DFT) calculations.
cd m−2
)
m−2, at 1000 cd m−2
7.6, 4.9, 2.3
y
5.3
13.4, 11.4, 6.7
0.15,
0.18
CONCLUSION
■
We have described a novel series of host materials based on
cyclic phosphazenes. Substituted phenyl rings were attached to
the central phosphazene core either via phosphorus−oxygen
bonds (OP1−OP3) or via phosphorus−carbon bonds (CP1−
CP3). The phenoxy substituted phosphazenes OP1-OP3
revealed high crystallization tendency and poor thermal
stability. In contrast, the phenyl substituted phosphazenes
CP1−CP3 form morphologically stable films. Because of the
nonconjugated linkage via the central phosphazene rings CP1-
CP3 exhibit very high triplet energies of more than 3 eV. Thus,
energy transfer to even deep blue emitters is possible. The
phenyl-substituted phosphazenes reveal balanced transport
characteristic for both holes and electrons. In an OLED device,
CP2 was successfully tested as host material for the emitter
Ir(dbfmi) with saturated blue emission.
with 10 wt-% molybdenum(VI) oxide were deposited as hole-
injection and hole-transporting layer. The additional 10 nm
thick hole transporting layer of DPBIC36 prevents quenching of
triplet excitons at the doped NPD-layer. On top a double
emission layer of 10 nm of DPBIC doped with 8% Ir(dbfmi)
and 40 nm of CP2 doped with 8% Ir(dbfmi) was deposited.
The electron transport within the emission layer takes place
over the host CP2. The thin emitter doped DPBIC interlayer
allows residual electrons to recombine and emit efficiently on
the emitter. Ten nanometers of the hole and exciton blocking
material 2,8-bis(triphenylsilyl)-dibenzofuran (DBFSi)37 and 40
nm of the electron-transporting material 1,3,5-tris(2-N-
phenylbenzimidazolyl)benzene (TPBI) were deposited. As
cathode LiF (0.5 nm)/ Al (100 nm) was used.
Figure 8 (top) shows the current density−voltage−
luminance characteristics of the device. A peak luminance of
5000 cd m−2 was reached at 13.5 V. The peak power efficiency
of 7.6 lm W−1 is achieved at 5.5 V. At 100 and 1000 cd m−2, the
power efficiencies are 4.9 lm W−1 and 2.3 lm W−1, respectively
(see Figure 8, bottom). Table 3 lists the device results at
different luminances. To the best of our knowledge, this work
represents the first report about low-molecular-weight vacuum
processable cyclic phosphazene compounds as host materials
for an emitter with saturated blue phosphorescence.
AUTHOR INFORMATION
Corresponding Author
*E-mail: peter.strohriegl@uni-bayreuth.de. Fax: +49-921-
553206. Tel: +49-921-553296.
■
ACKNOWLEDGMENTS
We thank Dr. Evelyn Fuchs, Dr. Hans-Hermann Johannes, Dr.
Nicolle Langer, Dr. Oliver Molt, Dr. Ingo Munster, and Dr. Ralf
Krause for stimulating discussions. Financial support from the
■
̈
4952
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953

Chemistry of Materials
Article
(34) Dietrich, A.; Neumuller, B.; Dehnicke, K. Z. Anorg. Allg. Chem.
2000, 626, 2035.
(35) Tokito, S.; Iijima, T.; Suzuri, Y.; Kita, H.; Tsuzuki, T.; Sato, F.
Appl. Phys. Lett. 2003, 82, 569.
̈
BMBF projects OPAL 2008 (FKZ 13N 8992) and TOPAS
2012 (FKZ 13N 10447) is gratefully acknowledged. P.S. thanks
the Universitat Bayern e.V. for a grant.
̈
(36) Bold, M.; Lennartz, C.; Prinz, M.; Schmidt, H.-W.; Thelakkat,
REFERENCES
M.; Bate, M.; Neuber, C.; Kowalsky, W.; Schildnkecht, C.; Johannes,
̈
■
H.-H.; WO Patent 2005, 019373.
(1) Tang, C. W.; van Slyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(2) Yersin, H., Ed. Highly Efficient OLED’s with Phosphorescent
Materials; Wiley-VCH: Weinheim, Germany, 2008.
(37) Langer, N.; Kahle, K.; Lennartz, C.; Molt, O.; Fuchs, E.;
Rudolph, J.; Schildknecht, C.; Watanabe, S.; Wagenblast, G.; WO
Patent 2009, 003898.
(3) Reinecke, S.; Lindner, F.; Schwartz, G.; Seidler, N.; Walzer, K.;
Lussem, B.; Leo, K. Nature 2009, 459, 234.
̈
(4) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Silbey, S.;
Thompson, M. E.; Forrest, S. R. Nature 1998, 395, 151.
(5) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl.
Phys. 2001, 90, 5048.
(6) Tsai, M.-H.; Lin, H.-W.; Su, H.-C.; Ke, T.-H.; Wu, C.-C.; Fang,
F.-C.; Liao, Y.-L.; Wong, K.-T.; Wu, C.-I. Adv. Mater. 2006, 18, 1216.
(7) He, J.; Liu, H.; Dai, Y.; Ou, X.; Wang, J.; Tao, S.; Zhang, X.;
Wang, P.; Ma, D. J. Phys. Chem. C 2009, 113, 6761.
(8) Adams, J. E.; Mantulin, W. W.; Huber, J. R. J. Am. Chem. Soc.
1973, 95, 5477.
(9) Ren, X.; Li, J.; Holmes, R. J.; Djurovich, P. I.; Forrest, S. R.;
Thompson, M. E. Chem. Mater. 2004, 16, 4743.
(10) Holmes, R. J.; D’Andrade, B. W.; Forrest, S. R.; Ren, X.; Li, J.;
Thompson, M. E. Appl. Phys. Lett. 2003, 83, 3818.
(11) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Tsyba, I.; Ho, N. N.; Bau,
R.; Thompson, M. E. Polyhedron 2004, 23, 419.
(12) Eom, S.-H.; Zheng, Y.; Chopra, N.; Lee, J.; So, F.; Xue, J. Appl.
Phys. Lett. 2008, 93, 133309.
(13) Burrows, P. E.; Padmaperuma, A. B.; Sapochak, L. S.; Djurovich,
P.; Thompson, M. E. Appl. Phys. Lett. 2006, 88, 183503.
(14) Padmaperuma, A. B.; Sapochak, L. S.; Burrows, P. E. Chem.
Mater. 2006, 18, 2389.
́
́
ez-Bolivar, C.; Anzenbacher, P. Jr.
(15) Mamada, M.; Ergun, S.; Per
Appl. Phys. Lett. 2011, 98, 073305.
(16) Sudhakar, S.; Sellinger, A. Macromol. Rapid Commun. 2006, 27,
247.
(17) Bolink, H. J.; Barea, E.; Costa, R. D.; Coronado, E.; Sudhakar,
S.; Zhen, C.; Sellinger, A. Org. Electr. 2008, 9, 155.
(18) Bolink, H. J.; Santamaria, S. G.; Sudhakar, S.; Zhen, C.;
Sellinger, A. Chem. Commun. 2008, 618.
(19) Soh, M. S.; Santamaria, S. A. G.; Williams, E. L.; Perez-Morales,
M.; Bolink, H. J.; Sellinger, A. J. Polym. Sci. B: Polym. Phys. 2011, 49,
531.
(20) Perdew, J. P.; Yue, W. Phys. Rev. B: Condens. Matter Mater. Phys.
1986, 33, 8800.
(21) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 36, 3098.
(22) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
̈
(23) Andrae, D.; Haußermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
̈
Theor. Chim. Acta 1990, 77, 123.
(24) Schafer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
̈
5829.
(25) Klamt, A. J. Phys. Chem. 1995, 99, 2224.
(26) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Chem.
̈
̈
̈
Phys. Lett. 1989, 162, 165.
(27) Sasabe, H.; Takamatsu, J.; Motoyama, T.; Watanabe, S.;
Wagenblast, G.; Langer, N.; Molt, O.; Fuchs, E.; Lennartz, C.; Kido, J.
Adv. Mater. 2010, 22, 5003.
(28) Allcock, H. R.; Sunderland, N. J.; Primrose, A. P.; Rheingold, A.
L.; Guzei, I. A.; Parvez, M. Chem. Mater. 1999, 11, 2478.
(29) Appel, R.; Einig, H. Chem. Ber. 1975, 108, 914.
(30) Bode, H.; Bach, H. Ber. Dt. Chem. Ges. 1942, 75, 215.
(31) Acock, K. G.; Shaw, R. A.; Wells, F. B. G. J. Chem. Soc. 1964,
121.
(32) Mashima, K.; Kusano, K.; Sato, N.; Matsumura, Y.; Nozaki, K.;
Kumobayashi, H.; Sayo, N.; Hori, Y.; Ishizaki, T.; Akutagawa, S.;
Takaya, H J. Org. Chem. 1994, 59, 3064.
(33) Bassler, H.; Arkhipov, V. I.; Emelianova, E. V.; Gerhard, A.;
̈
Hayer, A.; Im, C.; Rissler, J. Synth. Met. 2003, 135−136, 377.
4953
dx.doi.org/10.1021/cm201940f|Chem. Mater. 2011, 23, 4947−4953