Highly efficient top-emitting organic light-emitting diodes with self-assembed monolayer-modified Ag as anodes

Article abstract of DOI:10.1063/1.2388880

A series of p -substituted benzylmercaptans was used to form self-assembled monolayers (SAMs) on silver surface to serve as the anode in the fabrication of top-emitting organic light-emitting diodes. SAMs with electron-withdrawing substituents such as cyano and trifluoromethyl groups led to enhanced hole injection and electroluminescent performance, whereas the opposite is true for electron-donating groups. Direct quantitative correlation between the molecular dipole moment/work function or work function/hole injection does not exist. A high current efficiency of ～12 cdA was obtained, which is two and half times that of the conventional bottom-emitting device of the same layer structure.

Full text of DOI:10.1063/1.2388880

Highly efficient top-emitting organic light-emitting diodes with self-assembed
monolayer-modified Ag as anodes
Ming-Chin Hung, Kun-Yang Wu, Yu-Tai Tao, and Hung-Wei Huang
Citation: Applied Physics Letters 89, 203106 (2006); doi: 10.1063/1.2388880
View online: http://dx.doi.org/10.1063/1.2388880
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/89/20?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Tuning hole injection and charge recombination with self-assembled monolayer on silver anode in top-emitting
organic light-emitting diodes
Appl. Phys. Lett. 90, 241104 (2007); 10.1063/1.2747667
Self-assembled monolayer-modified Ag anode for top-emitting polymer light-emitting diodes
Appl. Phys. Lett. 89, 233513 (2006); 10.1063/1.2404589
High-efficiency microcavity top-emitting organic light-emitting diodes using silver anode
Appl. Phys. Lett. 88, 073517 (2006); 10.1063/1.2172734
Highly efficient top-emitting white organic electroluminescent devices
Appl. Phys. Lett. 86, 253508 (2005); 10.1063/1.1953883
Self-assembled monolayer modification of indium tin oxide anode surface for polymer light-emitting diodes with
poly[2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylene vinylene] for high performance
Appl. Phys. Lett. 86, 223505 (2005); 10.1063/1.1942644
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
138.251.14.35 On: Wed, 24 Dec 2014 22:54:59

APPLIED PHYSICS LETTERS 89, 203106 ͑2006͒
Highly efficient top-emitting organic light-emitting diodes
with self-assembed monolayer-modified Ag as anodes
Ming-Chin Hung, Kun-Yang Wu, Yu-Tai Tao,a͒ and Hung-Wei Huang
Institute of Chemistry, Academia Sinica, Taipei, Taiwan 115, Republic of China
͑Received 16 July 2006; accepted 9 October 2006; published online 14 November 2006͒
A series of p-substituted benzylmercaptans was used to form self-assembled monolayers ͑SAMs͒ on
silver surface to serve as the anode in the fabrication of top-emitting organic light-emitting diodes.
SAMs with electron-withdrawing substituents such as cyano and trifluoromethyl groups led to
enhanced hole injection and electroluminescent performance, whereas the opposite is true for
electron-donating groups. Direct quantitative correlation between the molecular dipole moment/
work function or work function/hole injection does not exist. A high current efficiency of ϳ12 cd/A
was obtained, which is two and half times that of the conventional bottom-emitting device of the
same layer structure. © 2006 American Institute of Physics. ͓DOI: 10.1063/1.2388880͔
Top-emitting organic light-emitting diodes ͑TOLEDs͒
have drawn much interest recently due to their superior ap-
plication potential ͑e.g., for better color saturation and high
aperture ratio͒ in active-matrix OLEDs.1 One of the require-
ments for TOLEDs is a highly reflective anode over the vis-
ible spectral range. Silver ͑Ag͒ and aluminum ͑Al͒ are suit-
able candidates due to their highest reflectivity among
various metals.1 However, the work function of Ag or Al is
low ͑4.3–4.6 eV͒, so that a large hole-injection barrier be-
tween these metals and common organic hole injection/
transport layers is present.1 Considerable efforts have been
devoted to overcome the problem. For example, a Ag2O
overlayer, a CFx coating, or a C60 layer has been reported to
improve the hole injection from Ag or Al anode.2 Alterna-
tively, a self-assembled monolayer ͑SAM͒ appears to be an
effective way of tuning the work function of metal
electrodes.3–5 Thus, the work function of silver can be raised
by more than 1 eV by a perfluorinated alkanethiol monolayer
or lowered by 0.6 eV by a alkanethiol SAM.5 The variation
of work function was modeled with two dipole sheets, one
associated with the metal-S bond and the other with the or-
ganic fragment. While perfluorinated alkyl chain is interest-
ing, use of other organic molecules as modifier on Ag for
top-emitting OLED fabrication is not reported. Aromatic thi-
ols such as benzylmercaptans are known to form well-
ordered monolayer on Au as well as Ag surface.6 The mol-
ecules were suggested to bind to the Au͑111͒ surface through
a ͑ͱ3ϫ ͱ3͒R30° binding scheme and the phenyl moieties
pack in a herringbone arrangement with near upright orien-
tation. The monolayer structure on Ag is similar to that on
Au due to very similar atomic lattice constants and the her-
ringbone packing of the aromatic rings.6 By introducing a
polar substituent at the paraposition, the dipole moment of
the molecule can be systematically tuned. In this work, we
examined TOLEDs using silver substrates modified with
SAMs of various benzylmercaptan derivatives. Indeed and in
general, electron-withdrawing substituents are found to raise
the work function and give enhanced device performance
while electron-donating groups gave the opposite effect.
However, a “quantitative” correlation of the work function or
the hole-injection efficiency with the molecular dipole mo-
ment is not observed. Other factors such as different dielec-
tric constants, different orientation of dipoles, or tunneling
barrier invoked by the organic molecule are also involved.
The anodes of the top-emitting devices were prepared by
thermally evaporating 150 nm of silver on the soda glass
with the area ͑0.625 mm2͒ defined by a patterned shadow
mask. The Ag substrates were then immersed in a 1 mM
ethanolic solution of various benzylmercaptans for 2 h. The
structures of thiol derivatives used are shown in Fig. 1. For
comparison, the as-deposited Ag and Ag2O-covered Ag were
also prepared. The Ag2O layer was prepared by exposure of
the bare Ag to ozone in a UV-ozone generator for 1 min. For
device fabrication, all the substrates were placed in a custom-
designed rotating substrate holder in a vacuum cham-
ber. Layers of 4,4 ,4 -tris͑3-methylphenylphenylamino͒
Ј Љ
triphenylamine ͑m-MTDATA, 30 nm͒, ␣-naphthylphenyl-
biphenyl diamine ͑NPB, 20 nm͒, and tris-͑8-hydroxy-
quinoline͒ aluminum ͑Alq, 50 nm͒ were deposited sequen-
tially. The cathode consists of triple layers of 1 nm LiF, 2 nm
Al, and 20 nm Ag to achieve both optical transmission and
effective electron injection.7 As a control, the bottom-
emitting OLED was prepared on indium-tin-oxide-coated
glass. Identical organic layers were deposited and 1 nm LiF
and 150 nm Al were used as the cathode. After evaporation,
the devices were encapsulated with transparent cover glasses
using UV-cured epoxy glue. The current-voltage-brightness
͑I-V-B͒ and electroluminescent spectra of devices were mea-
sured with a Photo Research PR650 spectroradiometer inter-
faced with a Keithley 2400 source meter.
As a first assumption, SAMs of various substituted ben-
zylmercaptans on Ag have the similar structure, that is, a
͑ͱ3ϫ ͱ3͒R30° binding scheme with the axis along C1-C4
pointing normal to the surface, the interfacial dipole created
by the SAM is related to the molecular dipole moment,
which consists of two independent dipoles: the Ag+–S− di-
a͒
Electronic mail: ytt@chem.sinica.edu.tw
FIG. 1. Structure of SAM-forming molecules.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
0003-6951/2006/89͑20͒/203106/3/$23.00 89, 203106-1 © 2006 American Institute of Physics
138.251.14.35 On: Wed, 24 Dec 2014 22:54:59

203106-2
Hung et al.
Appl. Phys. Lett. 89, 203106 ͑2006͒
FIG. 2. I-V characteristics of devices using various Ag substrates as anode.
FIG. 3. B-V characteristics of various devices.
␬ is the dielectric constant of the SAM. The discrepancy in
the order of work function may be due to different dielectric
constant, reorientation of the terminal dipole/molecular
framework as a result of the strong dipole-dipole repulsion,8
which increases with stronger dipoles. In contrast, the Ag2O
surface gave a work function of 4.9 eV. These results indi-
cate that SAMs with polar substituents offer a wide range of
work functions through the formation of an interfacial dipole
layer and change systematically the interfacial barrier for
charge injection from the Ag surface.
The coating of SAM has little effect on the reflectance
͑97%–98%͒ of the silver surface, except for NO2-substituted
SAM surface. The Ag substrate with NO2-substituted SAM
appeared brown and heterogeneous. In contrast, the reflec-
tance of Ag2O/Ag decreases sharply in the short-wavelength
range when the UV-ozone exposure time increases.2
Figure 2 compares the current-voltage ͑I-V͒ charac-
teristics for the otherwise identical top-emitting device
configuration of Ag/m-MTDATA͑30 nm͒/NPB͑20 nm͒/
Alq͑50 nm͒/LiF/Al/Ag with differently modified Ag as an-
ode. Under the same bias, devices with electron-withdrawing
groups such as CN and CF3 group gave higher currents than
that with as-deposited Ag as anode. Those with unsubstituted
benzylmercaptan and CH3O-substituted benzylmercaptan
gave significantly smaller currents. A rough correlation be-
tween the current and the work function of respective Ag
pole and the dipole of the aromatic moiety X-C6H4CH2. De-
pending on the substituent X, the two dipoles are along the
same direction or in the opposite direction. Calculation by
semiempirical method shows that the dipole moments of the
phenyl ring carrying electron-withdrawing groups ͑–NO2,
–CN, –CF3͒ point away from the surface, as the Ag+–S−
dipole does, with –NO2 giving the largest ͑4.37, 2.34, and
2.38 D, respectively͒. The unsubstituted phenyl ring and
phenyl ring carrying electron-donating groups ͑–H, –CH3,
and –OCH3͒ have the dipole moments pointing toward the
surface, opposite to the Ag+–S− dipole, with –OCH3 giving
the largest ͑1.88, 2.16, and 2.84 D, respectively͒. The work
functions of these SAM-modified surfaces were measured
by AC-2 ͑Hitachi, Japan͒. Compare to the as-prepared
Ag surface, which has a work function of 4.65 eV, the
CN-substituted and CF3-substituted benzylmercaptans in-
crease the work function of Ag by 0.40 and 0.70 eV, respec-
tively. However, the NO2-substituted benzylmercaptan raises
the work function by 0.6 eV, less than that by a CF3 groups.
The unsubstituted benzylmercaptan and CH3-substituted
benzylmercaptans decrease the work function by 0.13 and
0.22 eV, respectively. The strongest donor CH3O group low-
ers the work function by 0.20 eV, similar to that resulted
from a CH3 substituent. The change in work function is
related3 to N␮/␧ ␬, where the N is the density of dipoles, ␮
0
is the dipole moment, ␧0 is the permittivity of free space, and
FIG. 4. Current efficiency of various
devices.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
138.251.14.35 On: Wed, 24 Dec 2014 22:54:59

203106-3
Hung et al.
Appl. Phys. Lett. 89, 203106 ͑2006͒
anode was observed, indicating the injection barrier is the
determining factor. The current density for device with
Ag2O/Ag as the anode nevertheless is the highest, even
though the work function is lower than that of CN- and
CF3-substituted SAM surfaces. This may imply that the hole
injection barrier is not the only factor affecting the hole in-
jection efficiency. The organic molecules also impose a tun-
neling barrier for the charge injection.
All the top-emitting devices have a much narrower elec-
troluminescence ͑EL͒ spectrum, peaking at ϳ550 nm with a
full width at half maximum ͑FWHM͒ of 40 nm, presumably
due to the microcavity effect.9 In contrast, the bottom-
emitting device gave an EL peaking at 524 nm with a
FWHM of 100 nm. Although the microcavity effect would
lead to large variation of colors with viewing angles, it can
increase the outcoupling of light to a large extent in the for-
ward direction. While the device efficiency is highly depen-
dent on the exact thickness of the top cathode, the brightness
of current TOLEDs using CF3-SAM/Ag, CN-SAM/Ag, and
Ag2O/Ag as anodes are three, four, and five times, respec-
tively, higher than that of the bottom OLED at 9 V ͑Fig. 3͒.
Furthermore, the devices using CN-SAM/Ag and CF3-
SAM/Ag as the anode have efficiencies about 11–12 cd/A,
comparable or higher than that based on Ag2O as the anode
and are two and half times more efficient than the corre-
sponding bottom OLED ͑Fig. 4͒.
whereas electron-donating groups decrease it. No simple
quantitative correlation between the dipole of the molecule
and the work function was found. The variation in work
functions manifests itself in the hole-injection efficiency. In
comparison with the bottom OLED, TOLEDs using
CN-SAM/Ag and CF3-SAM/Ag as electrodes can give
similar I-V characteristics but higher current efficiency due
to strong microcavity effect. These SAM-modified electrodes
offer comparable or better performance than Ag2O/Ag,
which is more vulnerable to light and heat.
Financial support from the Ministry of Economics, Tai-
wan, the Republic of China is gratefully appreciated.
1C. C. Wu, C. W. Chen, C. L. Lin, and C. J. Yang, J. Disp. Technol. 01,
248 ͑2005͒.
2C. W. Chen, P. Y. Hsieh, H. H. Chiang, C. L. Lin, H. M. Wu, and C. C.
Wu, Appl. Phys. Lett. 83, 5127 ͑2003͒; H. W. Choi, S. Y. Kim, K. B. Kim,
Y. H. Tak, and J. L. Lee, ibid. 86, 012104 ͑2005͒; J. Y. Lee, ibid. 88,
073512 ͑2006͒; H. J. Peng, J. X. Sun, X. L. Zhu, X. M. Yu, M. Wong, and
H. S. Kwok, ibid. 88, 073517 ͑2006͒.
3R. W. Zehner, B. F. Parsons, R. P. Hsung, and L. R. Sita, Langmuir 15,
1121 ͑1999͒.
4I. H. Campbell, S. Rubin, T. A. Zawodzinski, J. D. Kress, R. L. Martin,
and D. L. Smith, Phys. Rev. B 54, R14321 ͑1996͒; I. H. Campbell, J. D.
Kress, R. L. Martin, and D. L. Smith, Appl. Phys. Lett. 71, 3528 ͑1997͒.
5B. D. Boer, A. Hadipour, M. M. Mandoc, T. V. Woudenbergh, and P. W.
M. Blom, Adv. Mater. ͑Weinheim, Ger.͒ 17, 621 ͑2005͒.
6S. C. Chang, I. Chao, and Y. T. Tao, J. Am. Chem. Soc. 116, 6792 ͑1994͒;
Y. T. Tao, C. C. Wu, J. Y. Eu, and W. L. Lin, Langmuir 13, 4018 ͑1997͒.
7L. S. Hung, C. W. Tang, M. G. Mason, P. Raychaudhuri, and J. Madathil,
Appl. Phys. Lett. 78, 544 ͑2001͒.
In summary, fine-tuning of work function of silver sub-
strate by a series of substituted benzylmercaptan through the
formation of a self-assembled monolayer was achieved.
These monolayer-covered surfaces retain the high reflectivity
of the silver metal. Benzylmercaptans carrying electron-
withdrawing groups increase the work function of silver
8S. Frey, A. Shaporenko, M. Zharnikov, P. Harder, and D. L. Allara,
J. Phys. Chem. B 107, 7716 ͑2003͒.
9N. Takada, T. Tsutsui, and S. Saito, Appl. Phys. Lett. 63, 2032 ͑1993͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
138.251.14.35 On: Wed, 24 Dec 2014 22:54:59