Fluorine cleavage of the light blue heteroleptic triplet emitter FIrpic

Article abstract of DOI:10.1016/j.jfluchem.2009.04.009

The lifetime stability of devices containing FIrpic as emitter has been a major concern for organic blue light emitting devices (OLEDs). To gain a deeper knowledge about the purity of FIrpic (bis[2-(4,6-difluorophenyl)pyridyl-N,C2′]iridium (III)) emitters

Full text of DOI:10.1016/j.jfluchem.2009.04.009

Journal of Fluorine Chemistry 130 (2009) 640–649
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Fluorine cleavage of the light blue heteroleptic triplet emitter FIrpic
a
a
b
Varatharajan Sivasubramaniam a, , Florian Brodkorb , Stephanie Hanning , Hans Peter Loebl ,
*
Volker van Elsbergen b, Herbert Boerner b, Ullrich Scherf c, Martin Kreyenschmidt a
a Department Chemical Engineering, University of Applied Sciences FH Muenster, Stegerwaldstrasse 39, 48565 Steinfurt, Germany
b Philips Technologie GmbH-Philips Research, Weisshausstr. 2, 52066 Aachen, Germany
c FB C-Makromolekulare Chemie und Institut fuer Polymertechnologie, U-10.24, Bergische Universitaet Wuppertal, 42097 Wuppertal, Germany
A R T I C L E I N F O
A B S T R A C T
Article history:
The lifetime stability of devices containing FIrpic as emitter has been a major concern for organic blue
light emitting devices (OLEDs). To gain a deeper knowledge about the purity of FIrpic (bis[2-(4,6-
difluorophenyl)pyridyl-N,C20]iridium (III)) emitters and how the purity is influenced by sublimation
steps, non-sublimated and sublimated FIrpic material was analyzed via liquid chromatography coupled
with electron spray ionization mass spectrometry (LC/ESI/MS). Cleavage of an electron-withdrawing
group from one of the ligands of the heteroleptic phosphorescent emitter could be identified in
sublimated FIrpic material via LC/ESI/MS. A detailed chemical analysis using LC/ESI/MS was carried out
Received 31 January 2009
Received in revised form 18 April 2009
Accepted 26 April 2009
Available online 5 May 2009
Keywords:
a
-NPD (a
-4,40-bis[(1-
for complete blue emitting devices of the following structure: indium–tin-oxide (ITO)/50 nm (
bis[(1-naphthyl)phenylamino]-1,10-biphenyl) -NPD)/10 nm 4,40,400-tris(carbazol-9-yl)triphenyla-
a
-4,40-
naphthyl)phenylamino]-1,10-biphenyl)
Organic light emitting devices (OLEDs)
Triplet emitter
(
a
mine (TCTA)/100 nm TCTA:8% FIrpic/50 nm 1,10-biphenyl-40-oxy)-bis(8-hydroxy-2-methylquinoli-
nato)-aluminum (BAlq)/1 nm LiF/100 nm Al. Two isomers of (FIrpic-1F) could be detected in an aged
OLED. Changes in the ligand systems of FIrpic, especially the loss of fluorine during the deposition
process can alter the emissive properties of the blue phosphorescent emitter. Beside isomer formation
and chemical degradation of FIrpic, substantial degradation was observed for the hole transport material
FIrpic (bis[2-(4,6-difluorophenyl)pyridyl-
N,C20]iridium (III))
Chemical degradation
a
-NPD in driven OLEDs.
ß 2009 Elsevier B.V. All rights reserved.
1. Introduction
efficiency can reach up to 100%. Classical examples of phosphor-
escent emitting materials are the green emitter fac Ir(ppy)3 and the
red emitter PtOEP [8,9]. In order to attain blue emission one
approach is to change the substituents of the ligands of the Ir-
complexes. In FIrpic, a prototypical blue-green emitter, the
heteroleptic iridium complex contains two fluoro-substituted
phenylpyridine ligands and an anionic 2-picolinic acid.
Since the report of this light blue phosphorescent material from
Adachi et al. [7], intense research was performed on many similar
six-membered complex molecules like FIrpic. Unfortunately, the
achieved lifetimes of blue emitter materials are still insufficient,
with the exception of a blue emitting material from Konica
Minolta [10].
In some cases, devices degradation was attributed to lumines-
cence quench induced by metal migration. Therefore different
analytical techniques were employed to investigate surface,
interfacial electronic structures and morphology of such emitting
devices, such as X-ray photoelectron spectroscopy (XPS) [11],
inverse photoemission spectroscopy (IPES), ultraviolet photoelec-
tron spectroscopy (UPS) and scanning electron microscopy (SEM)
[12]. Other techniques for thin-film analysis of organic devices,
such as dynamic secondary mass spectrometry (SIMS) [13,14]
were used as well. Similar to the organic light emitting diodes,
In 1983, Tang and Van Slyke reported the first thin-film green
emitting OLED based on tris(8-hydroxyquinoline) aluminium
(Alq3) [1]. Organic light emitting diodes have attracted great
attention in both fundamental research and device fabrication
because of their potential application as display components [2–7].
Usually, OLEDs consist of organic layer systems sandwiched
between two electrodes. Therefore, different organic layers with
multiple functions are used. The various layer thicknesses are
typically varied between 10 and 100 nm. The core of an OLED is the
emitting layer. Electrons will be delivered from the cathode and
the holes from the indium–tin-oxide (ITO) anode on the glass
substrate. In the emitting layer electrons and holes will recombine
exciting a molecule that can relax radiatively to the ground state.
Two modes of light emission have to be considered: fluorescence
and phosphorescence. Due to spin statistics, OLEDs using
fluorescent materials are limited to an internal quantum efficiency
of 25%, whereas in phosphorescent materials the internal quantum
* Corresponding author. Tel.: +49 2551962 291; fax: +49 2551962 711.
E-mail address: sivasubramaniam@fh-muenster.de (V. Sivasubramaniam).
0022-1139/$ – see front matter ß 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2009.04.009

V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
641
indium diffusion from indium–tin-oxide has also been identified as
one of the failure mechanisms in polymer devices [15]. Never-
theless these techniques all detect responses related to specific
atoms. However, the first chemical analysis of devices using HPLC/
MS, in regarding degradation processes was performed by
Kondakov et al. [16]. Dissolving the organic layers of the devices
and direct analysis using high pressure liquid chromatography
(HPLC), matrix assisted laser desorption ionization time-of-flight
mass spectrometry (MALDI/TOF/MS), nuclear magnetic resonance
(NMR) were done to get information about the degradation
mechanism [16–18]. Recently Scholz et al. analyzed fully
processed organic devices via LDI/TOF/MS technique [19].
In this paper, we report on an investigation of the degradation
pathways of FIrpic, using HPLC coupled with mass spectrometry
(MS). It has been demonstrated that LC/MS is an efficient analytical
tool for the analysis of OLED materials and devices [16]. To analyze
the molecular changes of the polar analyte FIrpic, an ion trap MS
with electron spray ionization (ESI) was used as interface tool
between HPLC and MS.
In order to investigate the steps in which degradation of FIrpic
occurs, non-sublimated and sublimated materials were compared.
In addition two diodes with identical architecture – one pristine
and one stressed – were compared in order to examine emitter
degradation during device operation.
Fig. 1. (a) EIC m/z 718 and (b) EIC m/z 573 of unprocessed material FIrpic. (c) MS spectra of the FIrpic isomers (I + II). Cleavage of picolinate from FIrpic (d) at tR = 7 min
(tR = time elapsed between the injection point and the peak maximum).

642
V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
2. Results and discussion
ligands were detected in non-sublimated material FIrpic.
Additionally in non-sublimed material a derivative of FIrpic
molecule in which one of the ligands has only one fluorine atom
was observed. Fig. 2(a) and (b) shows the EIC of m/z 700 and m/z
555. EIC of m/z 700 corresponds to (FIrpic-1F + Na+) and m/z 555
to (FIrpic-picolinate-1F). The corresponding mass spectrum is
given in Fig. 2(c).
These results demonstrate that in non-sublimated FIrpic two of
the four FIrpic isomers, (FIrpic-picolinate) and especially a FIrpic
derivative with only three fluorine atoms (FIrpic-1F) can be
identified.
Next, the analysis of the sublimated FIrpic was carried out
applying the same LC/ESI/MS conditions. A LC/ESI/MS chromato-
gram using EIC mode with the mass of m/z 718 and m/z 573 of
sublimated FIrpic is shown in Fig. 3(a) and (b).
EIC mode with the masses m/z 718 and m/z 573 indicates that
the sublimated FIrpic material only contains one of the isomers
with the retention time tR = 12 min. The chromatogram does not
display any signals before and after the retention time tR = 12 min
no signals with the masses of m/z 573 and m/z 718 could be
observed. Therefore the sublimation process is suitable to
separate the two isomers formed during the synthesis of the
material FIrpic.
Homoleptic and heteroleptic iridium complexes have differ-
ent isomers. Homoleptic d6 transitions Ir(III) complexes have
only the fac and mer isomers, whereas in heteroleptic systems
four different isomers are possible [20]. At first the non-
sublimated and sublimated FIrpic materials were analyzed using
a specially developed HPLC method, which is able to separate
the different isomers. The analytes were detected by ESI/MS
coupled to the HPLC. Moreover in order to determine if the
vapor deposition process and the electrical driving of the OLEDS
posses an influence on the FIrpic emitter material, pristine and
stressed device doped with 8% FIrpic were also studied. As the
first step of these investigations, the purity of the non-
sublimated FIrpic complex was investigated using the men-
tioned HPLC/ESI/MS method. Fig. 1(a) and (b) shows the
extracted ion current (EIC) signals of the non-sublimated
material FIrpic with the corresponding masses of m/z 718 and
m/z 573. m/z 718 corresponds to the molecule (FIrpic + Na+) and
m/z 573 to (FIrpic-picolinate).
Two signals were obtained at a mass of m/z 718 for the non-
sublimated FIrpic, which is shown in Fig. 1(a). One signal occurs
at tR = 9 min and the second one at tR = 12 min. Using EIC m/z
573 delivers also the two signals at tR = 9 min and at tR = 12 min.
These findings lead to the conclusion that at least two FIrpic
isomers were produced in higher concentrations during the
synthesis of the material. For both isomers (tR = 9 min and
tR = 12 min) identical mass spectra were obtained which is
illustrated in Fig. 1(c). Outlining from Fig. 1(a) and (b) two
isomers of FIrpic and FIrpic decorated only with the two dfp
In order to investigate if FIrpic derivatives with three fluorine
atoms are also separated in the sublimation process, EIC-MS
investigations for the mass of m/z 700 and m/z 555 were
performed on the cleaned materials. Fig. 3(a) and (b) clearly
reveals that one sublimation step alone is not suitable to
eliminate the derivative (FIrpic-1F) completely from the
mixture. Comparing the concentrations of sublimated and
Fig. 2. (a) EIC of m/z 700 of unprocessed material FIrpic. (b) EIC of m/z 555 of unprocessed material FIrpic which corresponds to (FIrpic-picolinate-1F). (c) Mass spectra of the
peak at tR = 13.5 min.

V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
643
Fig. 3. (a) EIC of m/z 718 and (b) EIC of m/z 573 which show the signals for the FIrpic isomer (II) in sublimed material FIrpic. (c) and (d) Corresponds to (FIrpic-1F + Na+) and
(FIrpic-picolinate-1F) in sublimed material FIrpic.
non-sublimated material, it is nevertheless evident that the
amount of (FIrpic-1F) is decreased substantially. The relative
decrease in the intensity of the formed analyte (FIrpic-1F) is
likely due to the one step sublimation process. Because it is
necessary to decrease the molar amount of the analyte (FIrpic-
1F) this would suggest and suppose that for application of pure
FIrpic, many sublimation cycles have to be performed in order to
get a pure material.
FIrpic isomer (III). Isotopic patterns and mass spectra confirmed
the identity of the formation of FIrpic isomer (III).
Beside the identification of FIrpic isomers (II) and (III) in
unstressed devices, no indication of the FIrpic isomer (I) could be
found (Fig. 4(c)).
Overall, this experiment confirms that during the deposition
process isomerization of the FIrpic emitter is taking place and the
newly formed isomers can be already detected in pristine devices.
Beside the formation of the two isomers of the heteroleptic
emitter FIrpic no evidence of chemical cleavage of picolinate from
FIrpic molecule during the vapor deposition process could be
observed. This compound would appear at tR = 7 min.
From the chromatographic analysis of the unprocessed and
sublimated material FIrpic the loss of one fluorine atom from the
emitter molecule FIrpic could be detected. In fact using EIC of m/z
700, the presence of the molecule (FIrpic-1F) can also be shown in
pristine devices, which is illustrated in Fig. 5.
The sublimation process is therefore
a suitable cleaning
procedure for FIrpic. To look for similar effects, analyses of pristine
and aged blue emitting FIrpic devices were carried out. Fig. 4(a)
and (b) shows the EIC of m/z 718 and m/z 573.
The peaks at tR = 12 min indicate the presence FIrpic isomer (II)
(Fig. 4(a) and (b)). Interestingly the extracted ion chromatograms
of m/z 718 and m/z 573 show dramatic changes, comparing
sublimated FIrpic with doped FIrpic devices. An additional signal
was observed at tR = 12 min, which indicates the formation of

644
V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
Fig. 4. (a) EIC of m/z 718, (b) EIC of m/z 573 of pristine device and (c) zoomed chromatogram of EIC of m/z 573.
However, it is an important result that during the manufacture
processes not only the two FIrpic isomers were generated but also
the molecule (FIrpic-1F).
The LC/ESI/MS spectra utilizing EIC and focusing on the masses
of m/z 718 and m/z 573 are shown in Fig. 6(a) and (b) for a driven
device.
To get further information into the role of the emitter
degradation during device operation, LC/ESI/MS measurements
were carried out on aged devices.
Fig. 6(a) and (b) indicates that FIrpic isomers (II) and (III) are
present in 24 h aged devices and moreover the FIrpic isomer (I)
(Fig. 6(c) at tR = 9 min) could be detected as well, which was not the
Fig. 5. EIC of m/z 700 of unstressed device.

V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
645
Fig. 6. (a) EIC of m/z 718, (b) EIC of m/z 573 of stressed device and (c) zoomed region of EIC of m/z 573.
Fig. 7. EIC of m/z 700 of aged device.

646
V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
case in unstressed devices. The presence of FIrpic isomer (I) could
be due to the thermal heat during device operation. The detailed
analysis of Fig. 6(c) further reveals that beside the formation of the
FIrpic (I) isomer an addition signal with m/z 573 using EIC at a
retention time tR = 7 min could be detected. This compound is a
cleavage product of the emitter molecule FIrpic, in which the
weakest ligand, the anionic picolinate is separated.
This dissociation of the picolinate was only observed in
aged devices (24 h). Therefore the fragmentation of FIrpic is a
direct result of the operation of the OLED. To further explore
the degradation process of the aged device, EIC profiles with m/z
700 were carried out. Fig. 7 shows the EIC of m/z 700 of
stressed device.
The relative intensity of the molecule (FIrpic-1F) increased
during device operation. This indicated that more (FIrpic-1F) are
formed during the aging process. The most significant observation
is the appearance of an additional signal at tR = 15 min using EIC m/
z 700. This signal was confirmed by several injections of different
extracts of stressed devices. It is another further isomer of the
molecule (FIrpic-1F). Theoretically sixteen different isomers of
the (FIrpic-1F) molecule are possible. All sixteen isomers of the
cleavage of one electron-withdrawing group fluorine atom of the
emitter FIrpic are given in Fig. 8.
These results were verified repeating the experiments several
times, always leading to the same results. The isotopic patterns
analyses revealed that the compound with a molecular mass of m/z
Fig. 8. Sixteen possible theoretical isomer products of loss of one fluorine atom from the emitting material FIrpic.

V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
647
Table 1
Components of FIrpic in non-sublimated, sublimated, pristine and aged devices.
FIrpic isomer (I)
FIrpic isomer (II)
FIrpic isomer (III)
FIrpic-picolinate
FIrpic-1F
FIrpic-1F
isomer
Non-sublimated FIrpic
Sublimated FIrpic
Pristine OLED
+
–
–
+
+
–
–
+
+
+
–
–
+
++
+
–
–
–
+
+
+
+
24 h aged OLED
++
++
(ꢀ): not detected.
(+): detected.
(++): observed in higher amounts.
700 contains the metal atom iridium further confirming the
interpretation. As the intensity increases, it becomes clear that
formation of isomerization species during device operation will
play an important role in the investigation of degradation
mechanisms of the PhOLEDs.
The signal at tR = 10.5 min corresponding to the molecule (a-
NPD-naphthyl) is significantly increased in stressed devices. The
additional signal with m/z 463 is detected at a retention times
tR = 9 min and is only found in aged devices. Comparing the
mass spectra of the signal (a-NPD-naphthyl) at tR = 10.5 min
In summary, comparing the pristine with the aged devices,
three different degradation products of the emitter FIrpic were
detected using LC/ESI/MS. One important indication is the
formation of the dissociation product (FIrpic-picolinate). Further-
more the formation of FIrpic isomer (I) could only be detected in
aged devices. Finally another isomer of (FIrpic-1F) was identified.
Table 1 summarizes the results.
with the mass spectra of the new signal at tR = 9 min identical
spectra were obtained. This is a strong indication that another
isomer of the molecule (
operation.
a-NPD-naphthyl) was generated during
However the amounts of both signals with the mass of m/z 463
are too low to isolate them for determining the absolute
configuration via NMR.
In summary, it was shown that in using an appropriate LC/MS
method, the identification of the different species from the light
blue emitter material FIrpic. Direct chromatographic separation of
enantiomers on chiral stationary phases by liquid chromatography
has been extensively developed. With ordinary chromatographic
method, it has been shown that the different species from FIrpic
could be separated on a non-chiral column. This method will be
useful not only for the determination of the purity of the
sublimated material of the iridium complexes FIrpic but also for
obtaining the possible degradation products from pristine and
stressed devices.
No signs of chemical degradation of the host material TCTA and
of the electron transport material BAlq were found via LC/ESI/MS.
LC/ESI/MS is a powerful analytical technique to characterize
the structural transformations of the d6 transition heteroleptic
Ir(III) complex FIrpic during device operation. In conclusion, one
isomer was formed during the deposition process and the other
one during device operation. Chemical transformations were
observed after device operation, resulting in (FIrpic-1F) and
(FIrpic-picolinate).
Two different aspects have to be considered regarding the
degradation of the emitter material. Cleavage of one fluorine atom
from FIrpic may cause a significant spectral change in the emitted
light. Secondly, the cleavage product can undergo further chemical
reactions with other organic materials nearby. Among the various
factors that may contribute to the degradation phenomena in
OLED, isomerization and the loss of an electron-withdrawing
group may have a considerable influence on the electrolumines-
cent lifetime of blue OLEDs based on FIrpic.
To estimate the concentration of the materials was
a
challenging task. Degradation products from the cyclometalated
iridium (III) complex FIrpic are presently unknown. Several
repetitions of the measurements and comparing the signal ratio of
the analytes leads to the results that the amount of isomerization
products of the cyclometalated iridium (III) complex FIrpic
increases especially during the device operation. Both degrada-
tion and isomerization products increase in their intensity
gradually if pristine and aged devices were compared. This
similar observation and information was not only found for the
emitting material, but also for the degraded product from the hole
As in the report of the dissociation process of the molecule CBP
from Kondakov et al. [16], cleavage of N–C bonds was shown for
NPD in stressed devices.
a-
transport material
a
-NPD.
3. Experimental
These results support the hypothesis that isomerization is
taking place during device operation which could have a significant
impact on the lifetime of the OLED. Similar observations were
made in other phosphorescent emitter materials as well.
3.1. Materials
Unprocessed material FIrpic and pure sublimed standard
substances 4,40,400-tris(carbazol-9-yl)triphenylamine (TCTA),
4,40-bis[(1-naphthyl)phenylamino]-1,10-biphenyl -NPD), 1,10-
Furthermore the characteristics of the hole transport material
a-
a
-NPD in pristine and aged devices was investigated. First of all
(a
HPLC/ESI/MS analysis of the sublimed material
performed. No fragmentations of the substituents of
be detected in these investigations.
However in vacuum-deposited devices one signal with the
mass of m/z 469 at tR = 10.5 min (Fig. 9(a)) could be detected via LC/
ESI/MS. This corresponds to the loss of one naphthyl group. Such a
dissociation process during the evaporation process has not been
a
-NPD was
biphenyl-40-oxy)-bis(8-hydroxy-2-methylquinolinato)-aluminum
(BAlq) and bis(2-[4,6-difluorophenyl)pyridyl-N,C20]iridium (III)
(FIrpic) were kindly delivered from Philips Research Aachen.
Solvent dioxane (HPLC grade) was purchased from Merck.
a
-NPD could
3.2. Devices
reported for the case of
a
-NPD. The N–C bond of
a
-NPD seems to be
OLED devices with FIrpic were prepared by Philips Research
Aachen. The organic layers were deposited using thermal
evaporation on indium–tin-oxide (ITO) covered glass substrates.
prone to rupture during the heating process. Interestingly the EIC
experiments delivered two signals at different retention times
with a mass of m/z 463. Fig. 9(b) shows the EIC with the mass of m/z
463 of 24 h aged device.
The light blue emitting device consists of a 50-nm
a-NPD layer,
followed by 10 nm TCTA. As host 100 nm TCTA was used that

648
V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
Fig. 9. Chemical reaction (I) shows the possibilities of the cleavage of naphthyl from hole transport material
a-NPD. One of those occurs already in a tiny amount in unstressed
device (a). In stressed device both isomers occurs in greater amount which is shown in chemical reaction (II).
was doped with 8% of the blue turquoise phosphorescent
emitter FIrpic. As electron transport layer (ETL) BAlq was
employed in layer thickness of 50 nm. Electrons were injected
from 100 nm Al cathode which was deposited on the top of the
BAlq. As cathode, 1 nm LiF and 100 nm Al was deposited on top
of the organic layers. These devices were aged under glove box
conditions (<0.1 ppm H2O and O2 each) for 24 h at 10 V and
100 mA/cm2.

V. Sivasubramaniam et al. / Journal of Fluorine Chemistry 130 (2009) 640–649
649
References
3.3. Sample preparation
[1] S.A. Van Slyke, C.H. Chen, C.W. Tang, Applied Physics Letters 69 (1996) 2160–
2162.
[2] B.K. Crone, P.S. Davids, I.H. Campbell, D.L. Smith, Journal of Applied Physics 87
(2000) 1974–1982.
[3] S. Tokito, Y. Taga, Applied Physics Letters 66 (1995) 673–675.
[4] R.H. Jordan, L.J. Rothberg, A. Dodabalapur, R.E. Slusher, Applied Physics Letters 69
(1996) 1997–1999.
[5] V. Bulovic, P. Tian, P.E. Burrows, M.R. Gokhale, S.R. Forrest, M.E. Thompson,
Applied Physics Letters 70 (1997) 2954–2956.
For LC/ESI/MS analysis devices were prepared in the following
way. Active areas of the devices were cut out with a diamond
cutting tool. Using the cut-off technique, the organic thin films of
the devices were carefully separated from the non-active areas. For
the analysis 1 cm ꢁ 1 cm area of the device were used for the
further process. Unstressed and aged active area pieces were
extracted separately in 5 ml dioxane. The cathode and anode
materials were removed from the organic extract, by filtration
[6] R. Antony, A. Moliton, B. Lucas, Applied Physics A: Materials Science & Processing
70 (2000) 185–195.
[7] C. Adachi, R.C. Kwong, P. Djurovich, V. Adamovich, M.A. Baldo, M.E. Thompson, S.R.
Forrest, Applied Physics Letters 79 (2001) 2082–2084.
[8] M.A. Baldo, D.F. O’Brien, Y. You, A. Shoustikov, S. Sibley, M.E. Thompson, S.R.
Forrest, Nature (London) 395 (1998) 151–154.
through a 4-mm nylon microdisc filter. Afterwards the filtrate was
evaporated with a N2 stream and dissolved with 500
ml dioxane.
[9] M.A. Baldo, S. Lamansky, P.E. Burrows, M.E. Thompson, S.R. Forrest, Applied
Physics Letters 75 (1999) 4–6.
3.4. LC/ESI/MS measurements
[10] K.H. Tomoyuki Nakayama, K. Furukawa, H. Ohtani, Konica Minolta Technology
Report, 2006, 5, 115–120.
For LC/ESI/MS analysis the injection volume was 10
ml extract.
The analytes and the device extracts were analyzed using a Merck
Hitachi HPLC system (Darmstadt, Germany). Consisting of a smash
pump (Model L-2100) including a degaser, an oven (Model L-
2300), an automated injector and an UV (Model L-2400) detector.
MS analyses were carried using a Bruker Daltonics esquire 4000
(Bremen, Germany) instrument. The ion trap octapole mass
spectrometer with electron spray ionization (ESI), which was
operated in positive ionization mode.
[11] L. Yan, Y. Gao, Thin Solid Films 417 (2002) 101–106.
[12] M. Carrard, S. Goncalves-Conto, L. Si-Ahmed, D. Ade, A. Siove, Thin Solid Films 352
(1999) 189–194.
[13] A. Delcorte, S. Hermans, M. Devillers, N. Lourette, F. Aubriet, J.F. Muller, P.
Bertrand, Applied Surface Science 231–232 (2004) 131–135.
[14] S.T. Lee, Z.Q. Gao, L.S. Hung, Applied Physics Letters 75 (1999) 1404–1406.
[15] A.R. Schlatmann, D.W. Floet, A. Hilberer, F. Garten, J.M. Smulders, T.M. Klapwijk, G.
Hadziioannou, Applied Physics Letters 69 (1996) 1764–1766.
[16] D.Y. Kondakov, W.C. Lenhart, W.F. Nichols, Journal of Applied Physics 101 (2007),
024512/024511-024512/024517.
[17] D.Y. Kondakov, W.C. Lenhart, W.F. Nichols, Abstracts of Papers, 232nd ACS
National Meeting, San Francisco, CA, United States, September 10–14, 2006
Journal Title: Abstracts of Papers, 232nd ACS National Meeting, San Francisco,
CA, United States, September 10–14, 2006, 2006, ORGN-095.
[18] V. Goldberg, M. Kaplan, L. Soltzberg, J. Genevich, R. Berry, A. Bukhari, S. Chan, M.
Damour, L. Friguglietti, E. Gunn, K. Ho, A. Johnson, Y.Y. Lin, A. Lowenthal, S. Suth, R.
To, R. Yopak, J.D. Slinker, G.G. Malliaras, S. Flores-Torres, H.D. Abruna, Materials
Research Society Symposium Proceedings 846 (2005) 301–306.
[19] S. Scholz, K. Walzer, K. Leo, Advanced Functional Materials 18 (2008) 2541–
2547.
All HPLC separations were carried out using a polar Chromolith
Si 100 column with 100% acetonitrile as eluent and a flow rate of
0.3 ml/min.
Acknowledgment
This work has been supported by BMBF (Bundesministerium fu¨r
Bildung und Forschung)-funded project 13N8669 (OPAL). The
authors wish to thank Philips Technologie GmbH-Philips Research
(Aachen, Germany) for their financial support.
[20] E. Baranoff, S. Suarez, P. Bugnon, C. Barolo, R. Buscaino, R. Scopelliti, L. Zuppiroli,
M. Graetzel, M.K. Nazeeruddin, Inorganic Chemistry, vol. 47, American Chemical
Society (ACS), Washington, DC, United States, 2008, pp. 6575–6577.