Synthesis of all-deuterated tris(2-phenylpyridine)iridium for highly stable electrophosphorescence: The "deuterium effect"

Article abstract of DOI:10.1039/c3tc30547c

Device stability and life-time rank the key issues for PhOLEDs. We synthesized deuterated Ir(ppy)₃-D₂₄. A device based on it has a current density twenty times higher than and a life-time six times longer than devices based on Ir(ppy)₃. The more stable C-D bond is found to be the main contributing factor, called the "deuterium effect".

Full text of DOI:10.1039/c3tc30547c

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
COMMUNICATION
Synthesis of all-deuterated tris(2-phenylpyridine)
iridium for highly stable electrophosphorescence: the
Cite this: J. Mater. Chem. C, 2013, 1,
4821
“deuterium effect”†
Ping Wang,‡a Fei-Fei Wang,‡b Yi Chen,a Qiang Niu,a Lei Lu,a Hong-Ming Wang,a
Xi-Cun Gao, Bin Wei, Hong-Wei Wu, Xin Cai and De-Chun Zou
Received 24th March 2013
Accepted 2nd July 2013
a
b
c
c
c
*
*
*
DOI: 10.1039/c3tc30547c
www.rsc.org/MaterialsC
Device stability and life-time rank the key issues for PhOLEDs. We
In PhOLEDs, the phosphorescent materials are usually
synthesized deuterated Ir(ppy)3-D24. A device based on it has a doped as emitting guests into charge-transporting and energy
current density twenty times higher than and a life-time six times harvesting host materials to obtain higher efficiency and longer
longer than devices based on Ir(ppy)3. The more stable C–D bond is life-time. The presence of a heavy-metal atom in the phospho-
found to be the main contributing factor, called the “deuterium rescent matrix provides a signicant spin-orbit interaction that
effect”.
allows the decay of a previously spin-forbidden radiative triplet
to happen. Due to both singlet and triplet excited states
participating in the emission, the internal quantum efficiency
can potentially reach as high as 100%.12 This paves the way for
high-efficiency organic color displays and white-lighting.
Iridium complexes are the most frequently used electro-
phosphorescent materials based on the remarkable “heavy
atom effect” of the Ir3+ nucleus. From this, early in 2009, we
came up with the idea of “all deuterating” the ligand to obtain a
heavier ligand for iridium for both green and blue electro-
phosphorescence. Close to the end of the synthesis, we
searched to nd a reference and discovered that someone else
had already executed the plan.13 However, our deuterating
method is different to theirs. Our direct synthesis from simple
reagents creates a higher deuteration rate (more than 99%
compared to the 95% reported). More importantly, we have
fabricated and tested electrophosphorescent devices. In the
following paragraphs, we will show that our device test results
have contributed to understanding the most important issues:
the device stability and operating lifetime can be greatly
increased or extended and the factors for these improvements
are identied.
Device lifetime and stability at high current densities are the key
issues for phosphorescent organic light-emitting diodes (PhO-
LEDs) because large scale manufacturing of real-world organic
at-panel color displays and white-lighting need thousands of
hours of operating time and are oen used at a high luminance.
Studies have revealed some chemical and physical degradation
mechanisms for PhOLEDs, such as cleavage of substituents of
organic materials,1 degradation in the HTL (hole-transport
layer),2 chemical degradation or decomposition of emitter/
host,3 concentration quenching,4 charge-trapping problems at
the sites of the dopant molecules owing to the large band gap
difference between the host and dopant molecules,5 triplet–
triplet annihilation,6 inefficient host to guest energy transfer,7
unbalanced carrier injection/transport,8 exciton diffusion from
the emitting layer,9 electrochemically instability of the host10
and electric eld induced quenching.11 However, the universal
key factors detrimental to PhOLED life-time remain unclear,
especially why the devices suffer “roll-off” under high current
densities.
By starting from the very simple and inexpensive reagent,
deuterated benzene, we obtained deuterated bromobenzene,
and then by reacting deuterated bromobenzene with another
simple and inexpensive regent, deuterated pyridine, we
obtained deuterated 2-phenylpyridine. Aer that, we obtained
deuterated tris(2-phenylpyridine)iridium by using this ligand.
We initially hoped that through deuteration of 2-phenyl-
pyridine, we could greatly increase the phosphorescence effi-
ciency of the iridium complex as a whole by introducing twenty
four heavier deuterium atoms instead of the hydrogen atom and
then greatly improve the device efficiency. However, contrary to
aDepartment of Chemistry, School of Science, Nanchang University, Xue-Fu Road 999,
Hong-Gu-Tan New District, Nanchang 330031, China. E-mail: xcgao@ncu.edu.cn
bKey Laboratory of Advanced Display and System Applications, Ministry of Education,
Shanghai University, Shanghai 200072, China. E-mail: bwei@shu.edu.cn
cBeijing National Laboratory for Molecular Sciences, Key Laboratory of Polymer
Chemistry and Physics of Ministry of Education, College of Chemistry and Molecular
Engineering, Peking University, Beijing 100871, China. E-mail: dczou@pku.edu.cn
† Electronic supplementary information (ESI) available. CCDC 917827. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/c3tc30547c
‡ These authors contributed equally to this work.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4821–4825 | 4821

View Article Online
Journal of Materials Chemistry C
Communication
our expectation, neither the material's nor the device's effi-
ciency has been greatly improved, but the device stability under
high current densities has been greatly improved and the device
lifetime been greatly extended. This is marvellous! It is better
than our initial expectation! The main factor inuencing the
device stability and life-time never previously reported was also
identied. It can provide new insights into solving the stability
issue. The deuteration can be used as a prototype for all elec-
trophosphorescent devices. The following is the story in detail.
Scheme 1 demonstrates the synthetic routes to the deuter-
ated ligand and the iridium complex. The detailed synthetic
procedure and characterization including the absorption and
emission spectra, the electrochemistry, the thermal properties,
and phosphorescent life-time can be found in the ESI.†
We prepared single crystals of Ir(ppy)3-D24. The atomic
arrangement in space for both Ir(ppy)3-D24 and Ir(ppy)3 is facial.
(The facial isomer can be easily identied from Fig. 1. The
eleven 13C NMR signals also provided evidence for the facial
isomer, rather than the meridional isomer, in which many more
than eleven 13C NMR signals would be found.) The distance
between phenylpyridine fragments and the Ir–Ir separation
distance of the two proximal complexes are calculated to be too
far to create probable inter-molecular interactions. However,
the topology of the molecular packing is different with that of
Ir(ppy)3. The cell setting for Ir(ppy)3-D24 is tetragonal in contrast
˚
Fig. 1 ORTEP diagram of Ir(ppy)3-D24. Selected bond distances (A): Ir–C(11) ¼
2.017(5), Ir–C(22) ¼ 2.011(5), Ir–C(33) ¼ 2.018(5), Ir–N(1) ¼ 2.133(4), Ir–N(2) ¼
2.120(4), Ir–N(3)–2.138(4) and angles (deg): C(22)–Ir–N(2) ¼ 79.58(18), C(11)–Ir–
N(1) ¼ 78.96(19), C(33)–Ir–N(3) ¼ 79.08(19).
bending at ꢁ1600 cmꢀ1, ꢁ1470 cmꢀ1, and 755 cmꢀ1 for Ir(ppy)3
obviously moves to the lower energy of ꢁ1566 cmꢀ1
,
ꢁ1400 cmꢀ1 and 611 cmꢀ1 for Ir(ppy)3-D24 instead. The reduced
energy for C–H bending is 155 cmꢀ1, equalling 0.45 kcal molꢀ1
.
to trigonal for Ir(ppy)3.14 The space group is P4 for Ir(ppy)3-D24
ꢀ
Therefore, we conclude that the deuteration decreases the
stretching and bending energy of the entire Ir(ppy)3-D24 mole-
cule, especially the C–H stretching and bending. In other words,
the C-D stretching and bending in Ir(ppy)3-D24 possess much
lower energy than the C–H stretching and bending in Ir(ppy)3.
Fig. 2(b) compares the total internal energy between Ir(ppy)3-D24
and Ir(ppy)3 by density functional theory (DFT) calculation. We
can see that the former has a lower internal energy of 25.4 kcal
molꢀ1 than the latter.
We compared the photophysical properties of the two
complexes in detail. The absorption and emission spectra, the
thermal and electrochemical values for deuterated Ir(ppy)3-D24
and un-deuterated Ir(ppy)3 are almost identical (see ESI†).
However, for deuterated Ir(ppy)3-D24, the quantum efficiency at
77 K/in degassed solution and the phosphorescent life-time at
77 K/solid powder is slightly higher than those of un-deuterated
ꢀ
in contrast to P3 for Ir(ppy)3. Considering that the crystal
formation conditions are different in that Ir(ppy)3-D24 crystals
were prepared by sublimation while Ir(ppy)3 crystals were
prepared in solution, these differences are not so surprising. In
addition, in OLEDs, materials are fabricated into devices in the
morphological state not in the crystal state. We thus exclude the
possible contribution of crystal packing to device performances.
We turn our attention to the structural difference between
the two materials. Fig. 2(a) is a comparison of the IR spectra
between Ir(ppy)3 and Ir(ppy)3-D24. The most evident is that the
highest energy for C–H stretching has been shied from 3040 to
2274 cmꢀ1 by deuteration (the small peak at ꢁ3020 cmꢀ1 may
be due to some kind of O–H stretching from a residual crystal
water). This reduced 766 cmꢀ1 energy equals to 2.2 kcal molꢀ1
In addition, the C]C stretching, C]N stretching, and C–H
.
Scheme 1 Synthetic route for Ir(ppy)3-D24
.
4822 | J. Mater. Chem. C, 2013, 1, 4821–4825
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Communication
Journal of Materials Chemistry C
Fig. 2 (a) IR spectra of the two complexes. (b) DFT calculation of the enthalpy of
Ir(ppy)3-D24 from Ir(ppy)3.
Fig. 3 (a) Current efficiency (cd Aꢀ1) vs. current density (mA cmꢀ2) curve. (b)
Luminance (cd mꢀ2) vs. current density (mA cmꢀ2). The device architecture is
glass/ITO (150 nm)/2T-NATA (25 nm)/NPB (5 nm)/TCTA (10 nm)/iridium complex
6% doped CBP (20 nm)/TPBi (30 nm)/LiF (0.3 nm)/Al (100 nm), where 2T-NATA is
4,40,400-tris-N-naphthyl-N-phenylamino-triphenylamine, NPB is N,N0-bis-(1-naph-
thyl)-N,N0-diphenyl-1,10-biphenyl-4,40-diamine, TCTA is 4,40,400-tri(N-carbazolyl)
triphenylamine CBP is 4,40-N,N0-dicarbazolyl-1,10-biphenyl, TPBi is 1,3,5-tris(N-
phenylbenzimidazol-2-yl)benzene.
Ir(ppy)3. There are reports on the quenching effects of C–H
oscillators and substitution of these quenching C–H modes for
C–D bonds in lanthanide complexes leads to signicant
enhancement of the luminescent lifetime and quantum
yields.16 However, from the data above, it can be concluded that
deuteration has only a minor effect on the photophysical
properties of the materials. This might be understood in terms
of a “heaviest atom effect”, that is, an electron will search the
heaviest nucleus nearby (in this case, the heaviest nucleus is Ir3+
on deuterated Ir(ppy)3-D24 and un-deuterated Ir(ppy)3 reaches
rather than D) and completes its spin-ip to facilitate the as high as 74 and 68 cd Aꢀ1 respectively. At the initial current
coupling of the spin angular moment to the orbital angular densities (0.6–3 mA cm2, corresponding to ꢁ200 cd mꢀ2
@
momentum. In addition, the mechanism to generate phos- 8.5 V), efficiency of both device types decays very fast. This is
phorescence in heavy-metal complexes is different from that to common for many OLEDs because at the beginning, the lumi-
generate luminescence in lanthanide complexes in that the nance is very low but corresponding to even lower a current
former stems mainly from 3MLCT and the latter comes from the density and as a result, the efficiency turns out to be quite high.
low-lying 4f atomic orbital of the lanthanide ion. Table 1 This continues until the device reaches ꢁ50 cd Aꢀ1. Then the
device based on Ir(ppy)3-D24 experiences a slower decay while
the device based on Ir(ppy)3 still decays very fast. In other words,
summarizes the main photophysical comparison between
Ir(ppy)3 and Ir(ppy)3-D24.
Fig. 3 compares the efficiency (a) and luminance (b) vs. the device based on Ir(ppy)3-D24 endures high current densities
current density curves based on deuterated Ir(ppy)3-D24 and un- much better. The efficiency decays to half of its initial value
deuterated Ir(ppy)3. The highest efficiency for the device based until the current density reaches 140 mA cmꢀ2 for the device
Table 1 Photophysical properties of the deuterated Ir(ppy)3-D24 and un-deuterated Ir(ppy)3
Fc %
Fd
%
Fe %
sf ms
sg ms
sh ms
Td ꢂC
Vox
V
Isomeric
a
Entry unit
labs nm
lemb nm
Ir(ppy)3
Ir(ppy)3-D24
380
380
509/540
511/541
33
40
81
100
88
85
7.0/3.1
11/4.3
3.5/0.87
4.0/0.93
11
11
410
400
0.68
0.66
Facial
Facial
a
In CH2Cl2 room temperature. b In CH2Cl2 degassed, 77 K, excited by 365 nm. c In CH2Cl2 degassed, room temperature, excited by 365 nm, contrast
d
to quinine sulfate which is reported to have a F value of 54%.15 In CH2Cl2, 77 K, no reference can be used, so the value for Ir(ppy)3 is relative to
Ir(ppy)3-D24 which is set as 100%. e 6%(wt) doped in CBP thin solid lm. f Double-exponential decay, in CH2Cl2, 77 K. g Double-exponential decay,
solid sample, room temperature. h Single-exponential decay, 6%(wt) doped in CBP thin solid lm.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4821–4825 | 4823

View Article Online
Journal of Materials Chemistry C
Communication
based on Ir(ppy)3-D24 while the efficiency decays to half of its partially take part in the excited state electro-generated phos-
initial value earlier at ꢁ7 mA cmꢀ2 for the device based phorescence as evidenced by the slightly improved photo-
on Ir(ppy)3. In this sense, we can say that the device based on luminescence efficiency and narrowed electrophosphorescent
Ir(ppy)3-D24 is 20 times more stable than the device based on spectrum, and as supported by the reports in lanthanide
Ir(ppy)3! Fig. 3b indicates that the device based on Ir(ppy)3-D24 luminescent materials of ref. 16. Adachi et al. recently identied
moves toward high luminance more steadily. Fig. 4a is the the interface between the HTL and the emissive layer Ir(ppy)3 as
device luminance vs. time curve. At an initial driven luminance the most susceptive region to degration.18 In this interface, the
of 1000 cd mꢀ2, both devices experience accelerating decay. C–H/D bonds are just in close contact with the HTL and might
For the device based on Ir(ppy)3-D24, the half life-time (t1/2
)
inuence the degradation. We herein term the inuential
was ꢁ38.5 hours and for the device based on Ir(ppy)3, t1/2 is ꢁ6.5 factors in which the C–D bond takes part in as “the deuterium
hours, so the former has about 6 times longer life-time than the effect”. The detailed mechanism needs further investigation.
latter in the test. We can reasonably expect a much longer life
In conclusion, we have synthesized deuterated Ir(ppy)3-D24
time for the former than the latter at a lower driven luminance, by starting from simple deuterated reagents. So, the cost of
ex. 100 cd mꢀ2, because the decay curve for the former is much deuteration is reasonably low. Deuteration of Ir(ppy)3 creates
atter at the latter part of the curve.17 Fig. 4(b) outlines the only minor improvements in its photophysical properties but
electrophosphorescent spectra of the two devices. The emis- has signicant effects on the device stability and life-time. The
sions are almost identical to Ir(ppy)3-D24 but with a narrower device based on Ir(ppy)3-D24 has high current density 20 times
peak.
better and a life-time 6 times longer than the device based on
These results are very important. Many reports have already Ir(ppy)3 at an initial luminance of 1000 cd mꢀ2. From the IR
conrmed the “roll-off” degeneracy of PhOLEDs.1–11 our nd- spectra comparison and the DFT calculation, we can prove that
ings here with improved device stability and prolonged device Ir(ppy)3-D24 possesses much lower internal energy than Ir(ppy)3,
lifetime can be helpful to solve the “roll off” problem. Through especially the C–D stretching and bending. This is the main
the analysis above, the improvements are closely related to the contributing factor toward a higher device stability and longer
reduced internal energy of the C–D bond, which may be helpful device life-time and we call it a “deuterium effect”. Our ndings
to prevent the degradation of the emitter. The C–D bond also provide a prototype for improving the device stability and life-
time, which are the most important issues regarding the fate of
PhOLEDs in the future: maybe we should deuterate all phos-
phorescent complexes from now on.
Experimental
Details of the synthetic procedures and characterization,
absorption spectra, emission spectra, electrochemistry, thermal
analysis, computational studies and device fabrication are
described in the ESI.†
Acknowledgements
This work was nancially supported by NSFC 60868001 and
61275041.
Notes and references
1 (a) V. Jankus, C. Winscom and A. P. Monkman, J. Chem.
Phys., 2009, 130, 074501; (b) N. Lin, J. Qiao, L. Duan,
H. F. Li, L. D. Wang and Y. Qiu, J. Phys. Chem., 2012, 116,
19451.
2 V. Sivasubramaniam, F. Brodkorb, S. Hanning, O. Buttler,
H. P. Loebl, V. V. Elsbergen, H. Boerner, U. Scherf and
M. Kreyenschmidt, Solid State Sci., 2009, 11, 1933.
3 (a) V. Sivasubramaniam, F. Brodkorb, S. Hanning,
H. P. Loebl, V. van Elsbergen, H. Boerner, U. Scherf and
M. Kreyenschmidt, Cent. Eur. J. Chem., 2009, 7, 836; (b)
S. Scholz, R. Meerheim, K. Walzer and K. Leo, in Organic
Optoelectronics and Photonics III, ed. P. L. Heremans, M.
Muccini and E. A. Meulenkamp, Proceedings of SPIE Vol.
6999, SPIE, Bellingham, Washington, 2008, ch. 1B; (c)
Fig. 4 (a) Luminance (L)/initial luminance (L0) for glass/ITO (150 nm)/2T-NATA
(25 nm)/NPB (5 nm)/TCTA (10 nm)/iridium complex 6% doped CBP (20 nm)/TPBi
(30 nm)/LiF (0.3 nm)/Al (100 nm). Both devices were operated at 2.2 mA cmꢀ2 and
the initial luminance L0 for both devices was 1000 cd mꢀ2. (b) Electrophosphorescent
spectraofthedevicesbasedontheiridiumcomplexes.Inset:top,pictureofthesealed
device turning-off; bottom, picture of the sealed device turning-on.
4824 | J. Mater. Chem. C, 2013, 1, 4821–4825
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Communication
Journal of Materials Chemistry C
D. Y. Kondakov, W. C. Lenhart and W. F. Nichols, J. Appl.
Phys., 2007, 101, 197; (d) L. J. Soltzberg, J. D. Slinker,
S. Flores-Torres, D. A. Bernards, G. G. Malliaras,
V. Adamovich, M. A. Baldo, M. E. Thompson and
S. R. Forrest, Appl. Phys. Lett., 2001, 79, 2082.
8 (a) J. X. Cai, T. L. Ye, X. F. Fan, C. M. Han, H. Xu, L. L. Wang,
D. G. Ma, Y. Lin and P. F. Yan, J. Mater. Chem., 2011, 21,
15405; (b) N. C. Giebink and S. R. Forrest, Phys. Rev. B:
Condens. Matter Mater. Phys., 2008, 77, 235215.
9 D. Y. Zhang, W. L. Li, B. Chu, J. Z. Zhu, T. L. Li, L. L. Han,
D. F. Bi, X. Li, D. F. Yang, F. Yan, H. H. Liu, D. Wang and
T. J. Tsuboi, Appl. Phys. Lett., 2007, 91, 183516.
~
H. D. Abruna, J. S. Kim, R. H. Friend, M. D. Kaplan and
V. Goldberg, J. Am. Chem. Soc., 2006, 128, 7761; (e) I. R de
Moraes, S. Scholz, B. Lussem and K. Leo, Org. Electron.,
2011, 12, 341; (f) R. Seifert, I. R. de Moraes, S. Scholz,
M. C. Gather, B. Lussem and K. Leo, Org. Electron., 2013,
14, 115; (g) I. R. de Moraes, S. Scholz, B. Lussem and
K. Leo, Appl. Phys. Lett., 2011, 99, 053302; (h) 10 Y. T. Tao, S. L. Gong, Q. Wang, C. Zhong, C. L. Yang, J. G. Qin
V. Sivasubramaniam, F. Brodkorb, S. Hanning, H. P. Loebl, and D. G. Ma, Phys. Chem. Chem. Phys., 2010, 12, 2438.
V. van Elsbergen, H. Boerner, U. Scherf and 11 S. Haneder, E. D. Como, J. Feldmann, M. M. Rothmann,
M. Kreyenschmidt, J. Fluorine Chem., 2009, 130, 640.
4 V. I. Adamovich, M. S. Weaver and J. J. Brown, in Organic
Light Emitting Materials and Devices XII, ed. F. So and C.
P. Strohriegl, C. Lennartz, O. Molt, I. Munster,
C. Schildknecht and G. Wagenblast, Adv. Funct. Mater.,
2009, 19, 2416.
Adachi, Proceedings of SPIE, vol. 7051, SPIE, Bellingham, 12 M. A. Baldo, D. F. O'Brien, Y. You, A. Shoustikov, S. Sibley,
Washington, 2008, ch. 0D. M. E. Thompson and S. R. Forrest, Nature, 1998, 395, 151.
5 (a) T. Tsuzuki and S. Tokito, Adv. Mater., 2007, 19, 276; (b) 13 T. Abe, A. Miyazawa, H. Konno and Y. Kawanishi, Chem.
M. A. Baldo and S. R. Forrest, Phys. Rev. B: Condens. Matter
Mater. Phys., 2000, 62, 10958.
6 (a) M. A. Baldo, C. Adachi and S. R. Forrest, Phys. Rev. B:
Phys. Lett., 2010, 491, 199.
14 J. Breu, P. Stossel, S. Schrader, A. Starukhin, W. J. Finkenzeller
and H. Yersin, Chem. Mater., 2005, 17, 1745.
Condens. Matter Mater. Phys., 2000, 62, 10967; (b) 15 W. R. Dawson and M. W. Windsor, J. Phys. Chem., 1968, 72,
J. Kalinowski, W. Stampor, J. Mezyk, M. Cocchi, D. Virgili, 3251.
V. Fattori and P. D. Marco, Phys. Rev. B: Condens. Matter 16 (a) C. Doffek, N. Alzakhem, C. Bischof, J. Wahsner, T. Guden-
¨
¨
Mater. Phys., 2002, 66, 235402; (c) K. S. Son, M. Yahiro,
T. Imai, H. Yoshizaki, J. Nishide, H. Sasabe and C. Adachi,
Jpn. J. Appl. Phys., 2008, 47, 7363; (d) A. Chaskar,
H. F. Chen and K. T. Wong, Adv. Mater., 2011, 23, 3876; (e)
M. H. Tsai, T. H. Ke, H. W. Lin, C. C. Wu, S. F. Chiu,
F. C. Fang, Y. L. Liao, K. T. Wong, Y. H. Chen and
C. I. Wu, ACS Appl. Mater. Interfaces, 2009, 1, 567.
Silber, J. Lugger, C. Platas-Iglesias and M. Seitz, J. Am. Chem.
Soc., 2012, 134, 16413; (b) G. A. Hebbink, D. N. Reinhoudt
and F. C. J. M. van Veggel, Eur. J. Org. Chem., 2001, 4101–
4106; (c) M. P. Oude Wolbers, F. C. J. M. van Veggel,
B. H. M. Snellink-Rue1l, J. W. Hofstraat, F. A. J. Geurts and
D. N. Reinhoudt, J. Am. Chem. Soc., 1997, 119, 138.
17 C. Fery, B. Racine, D. Vaufrey, H. Doyeux and S. Cina, Appl.
Phys. Lett., 2005, 87, 213502.
7 (a) R. J. Holmes, S. R. Forrest, Y. J. Tung, R. C. Kwong,
J. J. Brown, S. Garon and M. E. Thompson, Appl. Phys. Lett., 18 H. Yamamoto, C. Adachi, M. S. Weaver and J. J. Brown, Appl.
2003, 82, 2422; (b) C. Adachi, R. C. Kwong, P. Djurovich, Phys. Lett., 2012, 100, 183306.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4821–4825 | 4825