Covalently bound hole-injecting nanostructures. Systematics of molecular architecture, thickness, saturation, and electron-blocking characteristics on organic light-emitting diode luminance, turn-on voltage, and quantum efficiency

Published on Web 06/30/2005

Covalently Bound Hole-Injecting Nanostructures. Systematics

of Molecular Architecture, Thickness, Saturation, and

Electron-Blocking Characteristics on Organic Light-Emitting

Diode Luminance, Turn-on Voltage, and Quantum Efficiency

Qinglan Huang,^†Guennadi A. Evmenenko,^‡Pulak Dutta,^‡Paul Lee,^§

Neal R. Armstrong,^§and Tobin J. Marks*^,†

Contribution from the Department of Chemistry and the Materials Research Center, and

Department of Physics and Astronomy and the Materials Research Center,

Northwestern UniVersity, EVanston, Illinois 60208, and Department of Chemistry and

Optical Science Center, UniVersity of Arizona, Tucson, Arizona 85721

Received February 19, 2005 ; E-mail: t-marks@northwestern.edu

Abstract: Hole transporting materials are widely used in multilayer organic and polymer light-emitting diodes

(OLEDs, PLEDs, respectively) and are indispensable if device electroluminescent response and durability

are to be truly optimized. This contribution analyzes the relative effects of tin-doped indium oxide (ITO)

anode-hole transporting layer (HTL) contact versus the intrinsic HTL materials properties on OLED

performance. Two siloxane-based HTL materials, N,N′-bis(p-trichlorosilylpropyl)-naphthalen-1-yl)-N,N′-

diphenyl-biphenyl-4,4′-diamine (NPB-Si₂) and 4,4′-bis[(p-trichlorosilylpropylphenyl)phenylamino]biphenyl

(TPD-Si₂), are designed and synthesized. They have the same hole transporting triarylamine cores as

conventional HTL materials such as 1,4-bis(1-naphthylphenylamino)biphenyl (NPB) and N,N-diphenyl-N,N-

bis(3-methylphenyl)-1,1-biphenyl)-4,4-diamine (TPD), respectively. However, they covalently bind to the

ITO anode, forming anode-HTL contacts that are intrinsically different from those of the anode to TPD and

NPB. Applied to archetypical tris(8-hydroxyquinolato)aluminum(III) (Alq)-based OLEDs as (1) the sole HTLs

or (2) anode-NPB HTL interlayers, NPB-Si₂and TPD-Si₂enhance device electroluminescent response

significantly versus comparable devices based on NPB alone. In the first case, OLEDs with 36 000 cd/m²

luminance, 1.6% forward external quantum efficiency (η_ext), and 5 V turn-on voltages are achieved, affording

a 250% increase in luminance and ∼50% reduction in turn-on voltage, as compared to NPB-based devices.

In the second case, even more dramatic enhancement is observed (64 000 cd/m²luminance; 2.3% η_ext

;

turn-on voltages as low as 3.5 V). The importance of the anode-HTL material contact is further explored

by replacing NPB with saturated hydrocarbon siloxane monolayers that covalently bind to the anode, without

sacrificing device performance (30 000 cd/m²luminance; 2.0% η_ext; 4.0 V turn-on voltage). These results

suggest new strategies for developing OLED hole transporting structures.

1. Introduction

led to appreciable fundamental understanding of the physical

and chemical aspects of molecular solid-state electrolumines-

cence (EL) and prototype devices with impressive perform-

ance.^6-13The mechanism of emission in OLEDs is now

reasonably well understood and involves injection of electrons

and holes from the cathode and anode, respectively, under

application of an electric field. Some fraction of the electron-

hole pairs recombine to form excitons, which in turn radiatively

The past decade has witnessed the emergence of organic light-

emitting diodes (OLEDs) as a practical display technology due

to distinctive attractions such as low materials costs, self-

emission, efficient and broad color tunability, compatibility with

CMOS technology, and amenability to large-scale production.^1-5

Improving OLED electroluminescent characteristics has been

the focus of an extensive worldwide research effort that has

(6) Niu, Y. H.; Hou, Q.; Cao, Y. Appl. Phys. Lett. 2003, 82, 2163-2165.

(7) Duan, J. P.; Sun, P. P.; Cheng, C. H. AdV. Mater. 2003, 15, 224-227.

(8) Mller, C. D.; Falcou, A.; Reckefuss, N.; Rojahn, M.; Wiederhirn, V.; Rudati,

P.; Frohne, H.; Nuyken, O.; Becker, H.; Meerholz, K. Nature 2003, 421,

829-833.

^†Department of Chemistry and the Materials Research Center, North-

western University.

^‡Department of Physics and Astronomy and the Materials Research

Center, Northwestern University.

^§University of Arizona.

(1) For recent reviews of this subject, see: Veinot, J. G. C.; Marks, T. J. Acc.

Chem. Res., in press.

(2) Chen, C.-T. Chem. Mater. 2004, 16, 4389-4400.

(3) Nalwa, H. S. In Handbook of AdVanced Electronic and Photonic Materials

and DeVices; Nalwa, H. S., Ed.; Academic: San Diego, CA, 2001; Vol.

10, pp 1-51.

(4) Mitschke, U.; Bauerle, P. J. Mater. Chem. 2000, 10, 1471-1507.

(5) Shim, H.-K.; Jin, J.-I. AdV. Polym. Sci. 2002, 158, 193-243.

(9) Niu, Y. H.; Yang, W.; Cao, Y. Appl. Phys. Lett. 2002, 81, 2884-2886.

(10) Herguch, P.; Jiang, X. Z.; Liu, M. S.; Jen, A. K. Y. Macromolecules 2002,

35, 6094-6100.

(11) Lo, S. C.; Male, N. A. H.; Markham, J. P. J.; Magennis, S. W.; Burn, P.

L.; Salata, O. V.; Samuel, I. D. W. AdV. Mater. 2002, 14, 975-977.

(12) Huang, Q. L.; Cui, J.; Veinot, J. G. C.; Yan, H.; Marks, T. J. Appl. Phys.

Lett. 2003, 82, 331-333.

(13) D’Andrade, B. W.; Baldo, M. A.; Adachi, C.; Brooks, J.; Thompson, M.

E.; Forrest, S. R. Appl. Phys. Lett. 2001, 79, 1045-1047.

9

J. AM. CHEM. SOC. 2005, 127, 10227-10242

10227

A R T I C L E S

Huang et al.

indium oxide, ITO)-HTL interface is crucial to hole injection

in a common scenario where ITO-HTL contact rather than HTL

bulk mobility limits charge injection.^27-31A variety of interfacial

engineering approaches have been applied to the anode-HTL

junction,^32-37including introduction of π-conjugated polymers,³⁸

copper phthalocyanine,^26,39,40organic acids,⁴¹a thin layer of

platinum,²⁸self-assembled polar molecules^30,42or siloxanes,^31,43

and plasma treatment of the ITO surface.⁴⁴All of these

approaches have effects on hole injection and yield varying

degrees of improved device performance in terms of turn-on

voltage, luminance, stability, and/or quantum efficiency. How-

ever, in general, the interlayers employed in these approaches

cannot replace conventional triarylamine HTLs to yield high-

performance OLEDs. Note that designing new HTL materials

specifically targeting anode-HTL interfacial contact has largely

been ignored.^14,15,45

In this contribution, we present a full discussion of our efforts

to design, synthesize, implement, and understand the conse-

quences of hole transporting materials as OLED HTLs which

achieve covalent ITO anode-HTL bonding (Scheme 2).⁴⁶

Followed by deposition of a conventional EML/ETL (tris(8-

hydroxyquinolato)aluminum(III) (Alq, Scheme 1), this approach

affords OLEDs with superior performance versus devices relying

on simple ITO-NPB interfaces. For TPD-Si₂functionalized

anodes, luminances as high as 36 000 cd/m²with turn-on

voltages as low as 5.0 V are achieved, 2.5× brighter and 4 V

lower, respectively, than for analogous NPB-based devices.

Also, a current efficiency of ∼6 cd/A, equivalent to ∼1.6%

external forward quantum efficiency, is achieved at 12 V. We

also show that the same approach can be applied as covalently

bonded ITO anode-NPB interlayers, resulting in OLEDs with

more impressive performance (64 000 cd/m²luminance; 8 cd/A

Figure 1. Cross-section of a typical multilayer OLED structure.

decay to ground states and emit light.⁴Sophisticated multilayer

structures incorporating specifically tailored hole transport layers

(HTLs), emissive layers (EMLs), and electron transport layers

(ETLs) sandwiched between the two electrodes have been

developed and generally exhibit superior device performance

versus single-layer counterparts (Figure 1). To date, much effort

has been devoted to developing new HTL materials to fulfill

criteria such as substantial hole mobility, good energy level

matching with anodes and EMLs (low hole injection barriers

from the anode to the HTL and from the HTL to the EML;

large electron injection barrier from the EML to the HTL), good

thermal properties (stability; T_g), low optical absorption in the

visible region, and smooth, amorphous film-forming mor-

phology.^14-22

To date, triarylamines have proven to be one of the most

efficient classes of HTL molecules according to the above

criteria, with 1,4-bis(phenyl-m-tolylamino)biphenyl (TPD, Scheme

1) an archetypical example. However, TPD has a low glass

transition temperature (T_g) 65 °C) and tends to crystallize at

elevated temperatures, disrupting the amorphous nature of the

HTL and causing device degradation.²³Therefore, one major

focus in developing HTL materials is to improve the poor

thermal stability of TPD by synthesizing small molecule TPD

analogues with higher T_gparameters and/or by incorporating a

triarylamine hole transport motif into a polymer chain.^20,24,25

In this regard, 1,4-bis(1-naphthylphenylamino)biphenyl (NPB,

Scheme 1) is a TPD analogue with a higher T_gand has been

widely used as a TPD replacement.²⁶However, most of the

effort in this direction has not led to significant device

performance improvements over TPD-based OLEDs. In contrast,

increasing evidence indicates that the anode (usually tin-doped

(27) Forsythe, E. W.; Abkowitz, M. A.; Gao, Y. J. Phys. Chem. B 2000, 104,

3948-3952.

(28) Shen, Y.; Jacobs, D. B.; Malliaras, G. G.; Koley, G.; Spencer, M. G.;

Ioannidis, A. AdV. Mater. 2001, 13, 1234-1238.

(29) Nesch, F.; Forsythe, E. W.; Le, Q. T.; Gao, Y.; Rothberg, L. J. J. Appl.

Phys. 2000, 87, 7973-7980.

(30) Appleyard, S. F. J.; Day, S. R.; Pickford, R. D.; Willis, M. R. J. Mater.

Chem. 2000, 10, 169-173.

(31) Malinsky, J. E.; Jabbour, G. E.; Shaheen, S. E.; Anderson, J. D.; Richter,

A. G.; Marks, T. J.; Armstrong, N. R.; Kippelen, B.; Dutta, P.; Peygham-

barian, N. AdV. Mater. 1999, 11, 227-231.

(32) Zuppiroli, L.; Si-Ahmed, L.; Kamars, K.; Nesch, F.; Bussac, M. N.; Ades,

D.; Siove, A.; Moons, E.; Grtzel, M. Eur. Phys. J. B 1999, 11, 505-512.

(33) Ho, P. K. H.; Granstrom, M.; Friend, R. H.; Greenham, N. C. AdV. Mater.

1998, 10, 769-771.

(34) Nesch, F.; Rtzinger, F.; Si-Ahmed, L.; Zuppiroli, L. Chem. Phys. Lett. 1998,

288, 861-867.

(14) Ren, X.; Alleyne, B. D.; Djurovich, P. I.; Adachi, C.; Tsyba, I.; Bau, R.;

Thompson, M. E. Inorg. Chem. 2004, 43, 1697-1707.

(15) Hreha, R. D.; George, C. P.; Haldi, A.; Domercq, B.; Malagoli, M.; Barlow,

S.; Bredas, J.-l.; Kippelen, B.; Marder, S. R. AdV. Funct. Mater. 2003, 13,

967-973.

(35) Morgado, J.; Charas, A.; Barbagallo, N.; Alcacer, L.; Matos, M.; Cacialli,

F. Macromol. Symp. 2004, 212, 381-386.

(36) Hatton, R. A.; Willis, M. R.; Chesters, M. A.; Rutten, F. J. M.; Briggs, D.

J. Mater. Chem. 2003, 13, 38-43.

(37) Campbell, I. H.; Kress, J. D.; Martin, R. L.; Smith, D. L.; Barashkov, N.

N.; Ferraris, J. P. Appl. Phys. Lett. 1997, 71, 3528-3530.

(38) Gross, M.; Muller, D.; Nothofer, H.; Scherf, U.; Neher, D.; Brauchle, C.;

Meerholz, K. Nature 2000, 405, 661-665.

(16) Gong, X.; Moses, D.; Heeger, A. J.; Liu, S.; Jen, A. K. Y. Appl. Phys.

Lett. 2003, 83, 183-185.

(17) Ma, C.-Q.; Zhang, L.-Q.; Zhou, J.-H.; Wang, X.-S.; Zhang, B.-W.; Cao,

Y.; Bugnon, P.; Schaer, M.; Nuesch, F.; Zhang, D.-Q.; Qiu, Y. Chin. J.

Chem. 2002, 20, 929-932.

(39) Tadayyon, S. M.; Grandin, H. M.; Griffiths, K.; Norton, P. R.; Aziz, H.;

Popovic, Z. D. Org. Electron. 2004, 5, 157-166.

(18) Fujikawa, H.; Ishii, M.; Tokito, I.; Taga, Y. Mater. Res. Soc. Symp. Proc.

2001, 621, Q3.4.1-Q3.4.11.

(40) Mori, T.; Mitsuoka, T.; Ishii, M.; Fujikawa, H.; Taga, Y. Appl. Phys. Lett.

2002, 80, 3895-3897.

(19) Liu, S.; Jiang, X.; Ma, H.; Liu, M. S.; Jen, A. K. Y. Macromolecules 2000,

33, 3514-3517.

(41) Nesch, F.; Forsythe, E. W.; Le, Q. T.; Gao, Y.; Rothberg, L. J. J. Appl.

Phys. 2000, 87, 7973-7976.

(20) Koene, B. E.; Loy, D. E.; Thompson, M. E. Chem. Mater. 1998, 10, 2235-

(42) Ganzorig, C.; Kwak, K. J.; Yagi, K.; Fujihira, M. Appl. Phys. Lett. 2001,

79, 272-274.

2250.

(21) Thayumanavan, S.; Barlow, S.; Marder, S. R. Chem. Mater. 1997, 9, 3231-

3235.

(43) Cui, J.; Huang, Q.; Wang, Q.; Marks, T. J. Langmuir 2001, 17, 2051-

2054.

(22) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913-915.

(23) Adachi, C.; Tsutsui, T.; Saito, S. Appl. Phys. Lett. 1990, 56, 799-801.

(24) Santerre, F.; Bedja, I.; Dodelet, J. P.; Sun, Y.; Lu, J.; Hay, A. S.; D’Iorio,

M. Chem. Mater. 2001, 13, 1739-1745.

(44) Milliron, D. J.; Hill, I. G.; Shen, C.; Kahn, A.; Schwartz, J. J. Appl. Phys.

2000, 87, 572-576.

(45) Adachi, C.; Nagai, K.; Tamoto, N. Appl. Phys. Lett. 1995, 66, 2679-2681.

(46) Portions of this work were previously communicated in part. Huang, Q.;

Evmenenko, G.; Dutta, P.; Marks, T. J. J. Am. Chem. Soc. 2003, 125,

14704-14705. Cui, J.; Huang, Q.; Veinot, J. C. G.; Yan, H.; Wang, Q.;

Hutchison, G. R.; Richter, A. G.; Evmenenko, G.; Dutta, P.; Marks, T. J.

Langmuir 2002, 18, 9958-9970.

(25) Shirota, Y.; Okumoto, K.; Inada, H. Synth. Met. 2000, 111-112, 387-

391.

(26) Van Slyke, S. A.; Chen, C. H.; Tang, C. W. Appl. Phys. Lett. 1996, 69,

2160-2162.

9

10228 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

Scheme 1. Structures of Multilayer OLED Constituent Materials: TPD, NPB, and PEDOT(HTL), Alq (EML/ETL), DIQA(EML Dopant), and

BCP (ETL)

Scheme 2. Chemical Structures of the Organosiloxanes Used as

HTLs and ITO Anode Modification Layers

from Sigma-Aldrich and purified via vacuum gradient sublimation. The

OLED component, 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline (BCP,

Scheme 1), was purchased from Fluka and purified via vacuum gradient

sublimation. NPB was synthesized according to the literature²⁰and was

purified by recrystallization followed by vacuum gradient sublimation.

N,N′-Di(3-heptyl)quinacridone (DIQA, Scheme 1) was synthesized and

purified according to the literature.⁴⁸NMR spectra were obtained on

Varian VXR-400 or 500 MHz NMR instruments. MS analyses were

conducted on a Micromass Quattro II Triple Quadrupole HPLC/MS/

MS mass spectrometer. Elemental analyses were carried out by Midwest

Microlabs. Cyclic voltammetry was performed with a BAS 100

electrochemical workstation (SAM-coated ITO with ∼1 cm²area

working electrodes, Ag wire pseudo-reference electrode, Pt wire counter

electrode, 0.1 M TBAHFP in anhydrous MeCN supporting electrolyte,

and 0.001 M ferrocene as the internal pinhole probe; scan rate ) 0.1

V/s). TBAHFP was recrystallized from an ethyl acetate/hexanes mixture

and dried in vacuo at 100 °C for 10 h. Ferrocene was purchased from

Sigma-Aldrich and purified via vacuum gradient sublimation. IR spectra

were obtained on a Bio-Rad FTS-40 FTIR spectrometer. AFM images

were obtained on a Nanoscope III AFM under ambient conditions in

the contact mode with Si₃N₄cantilevers. Specular X-ray reflectivity

experiments on coated single-crystal Si (111) or Si (100) substrates

were performed on the Naval Research Laboratory X23B beamline at

the National Synchrotron Light Source. Data were acquired and

analyzed as described previously.⁴⁶The thickness of spin-cast films

was measured with a Tencor P-10 profilometer. Ultraviolet photoelec-

tron spectroscopy (UPS) experiments on TPD-Si₂and NPB-Si₂SAM-

coated ITO substrates were carried out at the University of Arizona

using the 21.2 eV He (I) source (Omicron H15-13) of a Kratos Axis-

165 Ultra photoelectron spectrometer. HOMO values were estimated

from the median energy of the clearly defined photoionization peak,

while IP values were estimated from the extrapolation of the high kinetic

energy edge of that peak to zero intensity.

current efficiency at 10 V; turn-on voltages as low as 3.5 V).

These anode functionalization processes are expected to sig-

nificantly modify anode-HTL interfacial properties,⁴⁷and here

we discuss how these modifications translate into modified

OLED EL device response. We also present here evidence that

simple saturated hydrocarbon self-assembled monolayers (SAMs;

Scheme 2) can function as HTLs and ITO-NPB interlayers,

affording a better understanding of silane-anode modification

effects. Devices with luminances as high as 30 000 cd/m²and

current efficiencies of ∼7 cd/A, equivalent to ∼2% external

forward quantum efficiency, at 7.0 V can be achieved. Taken

together, these results require some reassessment of current HTL

structure-function models.

Synthesis of 4,4′-Bis[(p-bromophenyl)phenylamino)]biphenyl (1).

To a toluene solution (50 mL) of tris(dibenzyldeneacetone)dipalladium

(0.55 g, 0.60 mmol) and bis(diphenylphosphino)ferrocene (0.50 g, 0.90

mmol) was added 1,4-dibromobenzene (18.9 g, 0.0800 mol) at 25 °C.

Following stirring under an N₂atmosphere for 10 min, sodium tert-

butoxide (4.8 g, 0.050 mol) and N,N′-diphenylbenzidine (6.8 g, 0.020

mol) were added. The reaction mixture was then stirred at 90 °C for

12 h, followed by cooling to 25 °C. The reaction mixture was then

poured into water, and the organic and aqueous layers were separated.

The aqueous layer was extracted with toluene (3 × 100 mL), and the

2. Experimental Section

Materials and Methods. ITO glass sheets (20 Ω/0, rms roughness

) 2.5 nm) were purchased from Colorado Concept Coating. All

chemical reagents were used as received unless otherwise indicated.

All manipulations of air/moisture-sensitive materials were carried out

on a dual-manifold Schlenk line or in a nitrogen-filled glovebox. Ether

and THF were distilled before use from sodium/benzophenone ketyl.

Methylene chloride was distilled before use from calcium hydride.

Toluene was dried using activated alumina and Q5 columns and tested

with benzyphenone ketyl in ether solution. TPD and Alq were purchased

(47) Cahen, D.; Hodes, G. AdV. Mater. 2002, 14, 789-798.

(48) Shaheen, S. E.; Kippelen, B.; Peyghambarian, N.; Wang, J. F.; Anderson,

J. D.; Mash, E. A.; Lee, P. A.; Armstrong, N. R.; Kawabe, Y. J. Appl.

Phys. 1999, 85, 7939-7945.

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10229

A R T I C L E S

Huang et al.

resulting extracts were combined with the original organic layer. The

solvent was removed in vacuo giving a crude product which was

purified by chromatography on a silica gel column (6:1 hexane:ethylene

chloride eluent) to yield pure 1 as a colorless solid (6.9 g) in 50%

yield. H NMR (CDCl₃): δ 6.99 (d, J ) 8.8 Hz, 4H), 7.02-7.16 (m,

10H), 7.28 (t, J ) 7.6 Hz, 4H), 7.34 (d, J ) 8.8 Hz, 4H), 7.45 (d, J )

8.4 Hz, 4H).

ether extracts were washed with water (2 × 100 mL) and brine (2 ×

100 mL), and dried over anhydrous Na₂SO₄. Following filtration, the

solvent was removed in a vacuum to yield an oil. Chromatography on

silica gel with hexane:methylene chloride (10:1) afforded 0.070 g of 5

as a colorless solid. Yield, 80%. ¹H NMR (CDCl₃): δ 3.90 (d, J ) 4.5

Hz, 4H), 5.17-5.20 (m, 4H), 6.20 (m, 2H), 6.95 (t, J ) 7.5 Hz, 2H),

7.06-7.10 (m, 10H), 7.21-7.24 (m, 4H), 7.30 (d, J ) 7.5 Hz, 2H),

7.37-7.40 (m, 8H), 7.50-7.53 (m, 2H), 8.04 (d, J ) 8.0 Hz, 2H),

8.09 (d, J ) 9.0 Hz, 2H). MS (m/z): 669.2 [M, 100]. Anal. Calcd for

C₅₀H₄₀N₂: C, 89.77; H, 6.04; N, 4.19. Found: C, 89.54; H, 6.04; N,

4.21.

Synthesis of N,N′-Bis(p-trichlorosilylpropyl)-naphthalen-1-yl)-

N,N′-diphenyl-biphenyl-4,4′-diamine (NPB-Si₂, 6). To a solution of

5 (0.040 g, 0.060 mmol) in 25 mL of dry toluene at 25 °C under inert

atmosphere was added H₂PtCl₆‚nH₂O (0.001 g), followed by trichlo-

rosilane (0.060 mL, 0.60 mmol). The reaction solution was warmed to

50 °C and monitored by NMR until the completion of reaction. Re-

moval of the solvent in a vacuum yielded an oil. Next, 20 mL of dry

toluene was added to the residue and the resulting solution was filtered

into a Schlenk flask by cannula. The filtrate was concentrated under

vacuum to give 6 as a pale-yellow oil. Yield, 98%. ¹H NMR (CDCl₃):

δ 1.70 (t, J ) 6.0 Hz, 4H), 2.10 (t, J ) 7 Hz, 4H), 3.21 (t, J ) 7 Hz,

4H), 6.94 (m, 2H), 7.03-7.07 (m, 10H), 7.19-7.22 (m, 4H), 7.33-

7.40 (m, 8H), 7.50-7.53 (m, 2H), 7.54 (d, J ) 8.0 Hz, 2H), 7.72 (d,

J ) 9.0 Hz, 2H), 8.04 (d, J ) 9.0 Hz, 2H). Anal. Calcd for

C₅₀H₄₂Cl₆N₂Si₂: C, 63.90; H, 4.50; N, 2.98. Found: C, 63.54; H, 4.04;

N, 2.90.

Self-Assembly of TPD-Si₂and NPB-Si₂on ITO Substrates. ITO

substrates were cleaned in an ultrasonic detergent bath, followed by

methanol, 2-propanol, and finally acetone. The substrates were subse-

quently treated in an oxygen plasma cleaner for 1 min to remove any

residual organic contaminants. Following strict Schlenk protocol, clean

ITO substrates were immersed in a 1.0 mM dry toluene solution of

TPD-Si₂or NPB-Si₂, respectively. After heating at ∼80 °C for 1 h, the

toluene solution was removed by cannula and the substrates were rinsed

with dry toluene (2 × 50 mL) and wet acetone, followed by transferring

to a 120 °C oven for 1 h to expedite cross-linking. Longer thermal

curing yields films with similar properties, and the coating procedure

has negligible effects on the measured sheet resistance of the underlying

ITO.

Self-Assembly of TPD-Si₂and NPB-Si₂on Silicon Substrates.

Silicon (111) or (100) substrates (Semiconductor Processing Co.) were

subjected to a cleaning procedure as follows: immersion in “piranha”

solution (concentrated H₂SO₄:30% H₂O₂; 70:30 v/v) at 80 °C for 1 h.

After being cooled to room temperature, substrates were rinsed

repeatedly with deionized (DI) water followed by an RCA-type cleaning

protocol (H₂O:30% H₂O₂:NH₃; 5:1:1 v/v/v; sonicated at room temper-

ature for 40 min). The substrates were finally rinsed with copious

amounts of DI water, heated to 125 °C for 15 min, and dried in vacuo.

TPD-Si₂and NPB-Si₂were then self-assembled onto the clean silicon

substrates following the procedure described above for ITO substrates.

Spin-Casting TPD-Si₂and NPB-Si₂on ITO Substrates. Dry

toluene solutions of TPD-Si₂or NPB-Si₂(10 mg/mL) were spin-coated

onto clean ITO surfaces at 2 krpm in air, followed by curing in a

vacuum oven at 110 °C for 1 h.

1

Synthesis of 4,4′-Bis[(p-allylphenyl)phenylamino]biphenyl (2).

Using standard Schlenk techniques, 1.6 mL (3.5 mmol) of n-butyl-

lithium(2.5 M in hexanes) was added dropwise under inert atmosphere

to an ether solution (10 mL) of 1 (1.02 g, 1.58 mmol) while maintaining

the temperature at 25 °C. The mixture was stirred for 2 h, after which

time CuI (0.76 g, 4.0 mmol) was added. Upon cooling the reaction

mixture to 0 °C, allyl bromide (0.60 g, 5.0 mmol) was added in one

portion, and the mixture was stirred for 14 h, followed by quenching

with saturated aqueous NH₄Cl solution (100 mL) and extraction with

ether (3 × 100 mL). The combined ether extracts were washed with

water (2 × 100 mL) and brine (2 × 100 mL), and dried over anhydrous

Na₂SO₄. Filtration and removal of solvent in vacuo afforded a yellow

oil, which was further purified by chromatography on a silica gel

column (4:1 hexane:methylene chloride) to yield 0.63 g of pure 2 as a

colorless solid. Yield, 70%. ¹H NMR (CDCl₃): δ 3.40 (d, J ) 10 Hz,

4H), 5.10-5.20 (m, 4H), 6.02 (m, 2H), 6.99-7.10 (m, 2H), 7.10-

7.20 (m, 16H), 7.28 (t, J ) 7.6 Hz, 4H), 7.46 (d, J ) 8.8 Hz, 4H).

Anal. Calcd for C₄₂H₃₆N₂: C, 88.68; H, 6.39; N, 5.23. Found: C, 87.50;

H, 6.35; N, 4.93.

Synthesis of 4,4′-Bis[(p-trichlorosilylpropylphenyl)phenylamino]-

biphenyl (TPD-Si₂, 3). Under inert atmosphere at 25 °C, a grain of

H₂PtCl₆‚nH₂O, followed by HSiCl₃(0.73 g, 5.5 mmol), was added to

a CH₂Cl₂solution (30 mL) of 2 (0.32 g, 0.55 mol), and the reaction

mixture was stirred at 30 °C for 4 h. Removal of the solvent in vacuo

yielded a dark-yellow oil, which was triturated with a mixture of 50

mL of pentane and 10 mL of toluene to yield a solid that was removed

by filtration. The filtrate was concentrated in vacuo to yield 3 as a

1

viscous, pale-yellow oil. Yield, 98%. H NMR (CDCl₃): δ 1.45 (t, J

) 7 Hz, 4H), 1.90 (t, J ) 7 Hz, 4H), 2.70 (brs, 4H), 6.80-7.80 (m,

26H). Anal. Calcd for C₄₂H₃₈Cl₆N₂Si₂: C, 60.07; H, 4.57. Found: C,

60.52; H, 4.87.

Synthesis of N,N′-Bis(p-bromonaphthalen-1-yl)-N,N′-diphenyl-

biphenyl-4,4′-diamine (4). To a toluene solution (50 mL) of tris-

(dibenzylideneacetone)dipalladium (0.048 g, 0.052 mmol) and bis-

(diphenylphosphino)ferrocene (0.044 g, 0.079 mmol) was added 1,4-

dibromonaphathelene (1.0 g, 3.5 mmol) at 25 °C. Following stirring

under an N₂atmosphere for 10 min, sodium tert-butoxide (0.34 g, 3.5

mmol) and N,N′-diphenylbenzidine (0.47 g, 1.4 mmol, purified by

sublimation) were added. The reaction mixture was stirred at 90 °C

for 22 h, followed by cooling to 25 °C. The reaction mixture was then

poured into water, and the organic and aqueous layers were separated.

The aqueous layer was extracted with toluene (3 × 100 mL), and the

resulting extracts were combined with the original organic layer. The

solvent was removed in vacuo giving a crude product that was purified

by chromatography on a silica gel column (6:1 hexane:methylene

chloride eluent) to yield pure 4 as a colorless solid (0.20 g) in 20%

yield. ¹H NMR (CDCl₃): δ 6.96-7.05 (m, 10H), 7.19-7.23 (m, 6H),

7.36-7.42 (m, 4H), 7.57 (t, J ) 8.5 Hz, 4H), 7.77 (d, J ) 9.5 Hz,

2H), 7.98 (d, J ) 10.5 Hz, 2H), 8.27 (d, J ) 10.5 Hz, 2H).

Synthesis of N,N′-Bis(p-allylnaphthalen-1-yl)-N,N′-diphenyl-bi-

phenyl-4,4′-diamine (5). To a stirring, anhydrous THF solution (10

mL) of 4 (0.10 g, 0.13 mmol) under inert atmosphere was added

dropwise at -78 °C n-butyllithium (1.6 M in hexanes, 0.17 mL, 0.27

mmol), and the mixture was stirred for 15 min. Copper iodide(I) (0.051

g, 0.27 mmol) was then added. After the mixture was stirred for 5

min, allyl bromide (0.07 g, 0.54 mmol) was added in one portion. The

solution was gradually warmed to 25 °C and stirred for 2 h, after which

time it was quenched with 100 mL of saturated aqueous NH₄⁺Cl^-

solution, followed by extraction with ether (3 × 100 mL). The combined

Fabrication of OLED Devices. All OLED devices were fabricated

on bare ITO substrates that were treated with solvents and an oxygen

plasma, as described previously, or on TPD-Si₂/NPB-Si₂modified ITO

substrates. The substrates were loaded into a bell jar deposition chamber

housed in a nitrogen-filled glovebox. A typical deposition procedure

is as follows: at 1 × 10^-6Torr, a 20 nm layer of NPB was first de-

posited, followed by 60 nm of Alq doped with 1% DIQA. Both organic

layers were grown at a deposition rate of 2-3 Å/s. Another 20 nm

thick layer of BCP was then deposited, followed by thermally evap-

orating 1 nm thick Li. Finally, a 100 nm thick Al cathode was deposited

through a shadow mask. This metallic layer was patterned via a shadow

9

10230 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

Scheme 4. Scheme for ITO Surface Modified by Covalently

Scheme 3. Synthesis of TPD-Si₂and NPB-Si₂

Bound HTL Materials

silanes to the surface.^49-52Further exposure to air and moisture

in the following wet acetone rinse in air hydrolyzes any

unreacted trichlorosilyl groups. Thermal curing facilitates the

formation of cross-linked siloxane networks, resulting in a thin

layer consisting of hole transporting motifs covalently anchored

to the ITO or Si surface. At completion of the silane condensa-

tion, chlorine concentrations in the film are below XPS detection

limits.^53,54The microstructural characterization of these layers

is presented below. Spin-casting TPD-Si₂and NPB-Si₂from

hydrocarbon solutions onto ITO or Si substrates affords thicker

(∼40 nm) films on these substrates, following chemistry similar

to that described above, given that the spin-coating parameters

such as solution concentration and spinning speed are carefully

controlled (see Experimental Section for details).

Characterization of TPD-Si₂and NPB-Si₂Siloxane Films

by Specular X-ray Reflectivity Measurements. Specular X-ray

reflectivity (XRR) measurements were performed upon the films

deposited on cleaned single-crystal Si(111) or Si (100) surfaces,

using the self-assembly procedure discussed above. Fitting of

the X-ray reflectivity data, normalized to the Fresnel reflectivity,

to a physically reasonable model provides details of the film

thickness, smoothness, interfacial abruptness, and density. In

general, the reflectivity can be expressed in terms of the average

electron density by eqs 1 and 2,⁵⁵

mask to give four devices, each with an area of 0.10 cm². OLED device

characterization was carried out with a computer-controlled Keithley

2400 source meter and IL 1700 research radiometer equipped with a

calibrated silicon photodetector at 25 °C under ambient atmosphere.

External quantum efficiency was estimated from current density versus

voltage and luminance versus current density characteristics.

Fabrication of Hole-Only Devices. All hole-only devices were

fabricated on bare ITO substrates that were treated with solvents and

an oxygen plasma, as described previously, or on TPD-Si₂/NPB-Si₂

modified ITO substrates. The substrates were loaded into a bell jar

deposition chamber housed within a nitrogen-filled glovebox. At a base

pressure of 1 × 10^-6Torr, NPB (400 nm) was vapor deposited at a

rate of ∼3 Å/s, followed by sputter-coating of a 6 nm of Au layer

through the same shadow mask employed for OLED fabrication de-

scribed above. Single-carrier device assembly was subsequently com-

pleted with the masked thermal vapor deposition of Al (150 nm). Device

behavior was evaluated using the computer-controlled Keithley 2400

sourcemeter.

2

R(k_z) ) R_F(k_z)|Φ(k_z)|

(1)

(2)

d F

1

Φ(k ) )

e

^{ik z}dz

z

∫

z

F_∞dz

where R_Fis the theoretical Fresnel reflectivity for a smooth

interface, k_zis the momentum transferred (k_z) 4π/λ

sin θ), d F /dz is the derivative of the electron density along

the surface normal direction, averaged over the in-plane

coherence length of the X-rays, and F_∞is the electron density

of the substrate (Si). By fitting the XRR data for the self-

assembled TPD-Si₂and NPB-Si₂films on Si, the thickness,

electron density, and roughness were determined. These data

are compiled in Table 1.

Characterization of Self-Assembled TPD-Si₂and

NPB-Si₂Monolayers by Ultraviolet Photoelectron Spec-

troscopy (UPS). UPS measurements were conducted on ITO

surfaces covered with TPD-Si₂and NPB-Si₂SAMs. The bare

ITO surface, which had been Ar⁺-sputtered in vacuo, is used

3. Results

In Section 3.1, we report the deposition and characterization

of TPD-Si₂and NPB-Si₂on ITO or Si substrates. In Section

3.2, spin-coated or self-assembled TPD-Si₂and NPB-Si₂layers

replace the conventional vapor-deposited NPB layer in

Alq-based OLEDs. Their current/luminous functions as new

types of covalently bonded HTLs are described. Alternatively,

TPD-Si₂and NPB-Si₂are used as ITO anode-NPB interlayers,

and the effect on OLED EL response is analyzed here. In Section

3.3, alkylsiloxane SAMs are employed as HTLs and ITO-NPB

interlayers, and the effect on device performance is compared

to that of the above triarylaminesiloxanes.

3.1. Deposition and Characterization of TPD-Si₂and

NPB-Si₂on ITO or Silicon Substrates. The synthetic pathways

to TPD-Si₂and NPB-Si₂are summarized in Scheme 3, while

detailed procedures and characterization data are presented in

the Experimental Section. Utilizing a self-limiting, solution-

based chemisorption process, TPD-Si₂and NPB-Si₂are self-

assembled onto hydrophilic ITO substrate surfaces with nano-

precise control in thickness. As illustrated in Scheme 4, clean

ITO-coated glass or single-crystal silicon surfaces possess

hydroxyl functionalities and adsorbed water, which are reactive

toward chlorosilanes, thereby affording covalent binding of the

(49) Ulman, A. Chem. ReV. 1996, 96, 1533-1554.

(50) Jeon, N. L.; Nuzzo, R. G.; Xia, Y.; Mrksich, M.; Whitesides, G. M.

Langmuir 1995, 11, 3024-3026.

(51) Ulman, A. Thin Solid Films 1996, 273, 48-53.

(52) Woodward, J. T.; Ulman, A.; Schwartz, D. K. Langmuir 1996, 12, 3626-

3629.

(53) Donley, C.; Dunphy, D.; Paine, D.; Carter, C.; Nebesny, K.; Lee, P.;

Alloway, D.; Armstrong, N. R. Langmuir 2002, 18, 450-457.

(54) Malinsky, J. E.; Veinot, J. C. G.; Jabbour, G. E.; Shaheen, S. E.; Anderson,

J. D.; Lee, P.; Richter, A. G.; Burin, A. L.; Ratner, M. A.; Marks, T. J.;

Armstrong, N. R.; Kippelen, B.; Dutta, P.; Peyghambarian, N. Chem. Mater.

2001, 14, 3054-3065.

(55) Richter, A. G.; Yu, C. J.; Datta, A.; Kmetko, J.; Dutta, P. Phys. ReV. E

2000, 61, 607-615.

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10231

A R T I C L E S

Huang et al.

Table 1. XRR-Derived Electron Density, Roughness, and

Thickness Data for Self-Assembled TPD-Si₂and NPB-Si₂

Monolayers on Si (100) Substrates

electron density

roughness

(Å)

thickness

(Å)

3

monolayer

(e Å^-

)

TPD-Si₂

NPB-Si₂

0.32-0.35

0.45-0.47

7.8 ( 0.2

4.1 ( 0.2

17.8 ( 0.2

14.1 ( 0.2

Table 2. UPS Measured Ionization Potential, Highest Occupied

Molecular Orbital (HOMO) Energy, and Vacuum Level Shift for

TPDSi₂and NPBSi₂Self-Assembled Monolayers on ITO

Substrates

ionization

HOMO

(eV)

vacuum level

shift (eV)

potential (eV)

Figure 3. Cyclic voltammetry characteristics of self-assembled TPD-Si₂

and NPB-Si₂monolayer films on an ITO electrode immersed in a ferrocene

solution. Electrolyte: 0.1 M TBAHFP in anhydrous MeCN. Sweep speed:

0.1 V/s.

TPD-Si₂

NPB-Si₂

5.2

5.4

6.1

(0.1

-0.1

as reference for the low kinetic energy edge of the photoemission

spectrum in the measurement. A 0.1 eV shift in the low kinetic

energy edge of the photoemission spectrum obtained from

the analysis of the SAM-covered ITO surface is observed for

both TPD-Si₂and NPB-Si₂, indicating interfacial dipole for-

mation.⁵⁶The sign of the vacuum level shift for the TPD-Si₂

SAMs is uncertain due to experimental errors. Estimates of the

effective surface work function can be obtained from the

measured width of the UV-photoemission spectrum, subtracted

from the source energy, using the high and low kinetic energy

edges of the photoemission spectra. The analysis shows nearly

identical HOMO and IP values for TPD-Si₂and NPB-Si₂

SAM modified ITO surfaces (5.2 and 5.4 eV, respectively,

Table 2).

Characterization of TPD-Si₂and NPB-Si₂Siloxane Films

by Atomic Force Microscopy. Contact mode AFM imaging

was carried out on three spots randomly chosen on self-

assembled or spin-cast TPD-Si₂or NPB-Si₂covered ITO

substrates, respectively. Uniform films with no evidence of

cracks or pinholes were observed over a 5 × 5 µm scan area

(Figure 2). The rms roughness is 1.39 and 3.18 nm, respectively,

for TPD-Si₂and NPB-Si₂SAMs, and 0.79 and 3.43 nm,

respectively, for spin-coated TPD-Si₂and NPB-Si₂films, as

compared to bare ITO substrates with an rms roughness of

2.5 nm.

Characterization of TPD-Si₂and NPB-Si₂Siloxane Films

by Cyclic Voltammetry. TPD-Si₂and NPB-Si₂SAM-coated

ITO, silver wire, and Pt wire were used as the working elec-

trode, reference electrode, and counter electrode, respectively,

in 0.1 M acetonitrile solution of tetrabutylammonium hexa-

fluorophosphate as the electrolyte without the ferrocene. Inte-

gration of the oxidation peak areas and assuming two elec-

tron oxidation/reduction events per molecular unit yields

surface coverages of 2.5 × 10^-10mol/cm²(TPD-Si₂SAM) and

2.2 × 10^-10mol/cm²(NPB-Si₂SAM), respectively.⁵⁷Analo-

gously, a close-packed monolayer of ferrocene dicarboxylic

acid absorbed on ITO yields 4.0 × 10^-10mol/cm²surface

coverage.⁵³

To examine the present HTL SAMs for pinholes over the

entire ITO substrate area, cyclic voltammetry experiments

using ferrocene as a redox probe were then carried out.⁵⁸Here,

TPD-Si₂or NPB-Si₂SAM-coated ITO substrates were used as

working electrodes. Bare ITO substrates were used as references

to calibrate the ferrocene redox potential for each measurement.

It was observed that the ferrocene oxidation potential shifts from

0.59 V, as shown in Figure 3, when bare ITO substrates are

used as working electrodes, to 0.84 and 0.77 V, respectively,

for the TPD-Si₂and NPB-Si₂SAM-covered ITO substrates as

working electrodes. It is expected that the pinhole-free films of

TPD-Si₂and NPB-Si₂SAMs on ITO electrodes will block the

oxidation of solution ferrocene, until potentials are reached

where the bis-trarylamine groups are oxidized, providing sites

for mediated oxidation of solution phase ferrocene.^59,60For both

TPD-Si₂and NPB-Si₂SAMs on ITO, the peak potential for

oxidation of ferrocene is indeed shifted positively by ca.

0.2 V. The reduction of the ferricenium cation on the return

sweep is partially blocked by the NPB-Si₂layer, and more

completely blocked by the TPD-Si₂layer, suggesting that the

latter SAM layer is somewhat more pinhole free.⁶¹The larger

separations of ferrocene oxidative and reductive peak potentials

at the SAM-coated ITO electrodes versus on the bare ITO ones

(56) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. AdV. Mater. 1999, 11, 605-625.

(57) Murray, R. W. In Molecular Design of Electrode Surfaces, Techniques of

Chemistry; Murray, R. W., Ed.; Wiley: New York, 1992; Vol. 22, pp

1-158.

(58) DuBois, C. J., Jr.; McCarley, R. L. J. Electroanal. Chem. 1998, 454, 99-

105.

Figure 2. Tapping mode AFM images of self-assembled and spin-cast

TPD-Si₂and NPB-Si₂films on ITO substrates. (A) TPD-Si₂SAM on ITO.

RMS roughness ) 1.39 nm. (B) TPD-Si₂spin coated film on ITO. RMS

roughness ) 0.79 nm. (C) NPB-Si₂SAM on ITO. RMS roughness ) 3.18

nm. (D) NPB-Si₂spin coated film on ITO. RMS roughness ) 3.43 nm.

(59) Immoos, C. E.; Chou, J.; Bayachou, M.; Blair, E.; Greaves, J.; Farmer, P.

J. J. Am. Chem. Soc. 2004, 126, 4934-4942.

(60) Nowall, W. B.; Kuhr, W. G. Anal. Chem. 1995, 67, 3583-3588.

(61) Inzelt, G. In Electroanalytical Chemistry; Bard, A. J., Rubenstein, I., Eds.;

Marcel Dekker: New York, 1994; Vol. 18, pp 90-134.

9

10232 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

(1.68 V, TPD-Si₂SAM; 0.86 V, NPB-Si₂SAM; 0.30 V, bare

ITO) may also indicate that in the redox processes, counterion

penetration and diffusion in the SAMs are retarded, presumably

due to the densely packed, cross-linked SAM networks.^62-65It

is likely that TPD-Si₂, due to its smaller core triarylamine

molecular volume, encounters less steric hindrance in the film-

forming intermolecular cross-linking process as compared to

NPB-Si₂, resulting in more densely packed films. The more

distorted curve shape for the TPD-Si₂SAM is presumably

related to its more densely packed, cross-linked coverage and

different π-system redox kinetics versus NPB-Si₂.

3.2. Self-Assembled and Spin-Cast TPD-Si₂and NPB-Si₂

Films in OLEDs. Replacing NPB with TPD-Si₂and

NPB-Si₂SAMs in OLEDs. Alq-based OLEDs having the

structure ITO anode/HTL/Alq doped with DIQA/BCP doped

with Li/Al were fabricated to study the function of TPD-Si₂

and NPB-Si₂films as HTLs. Replacing vapor-deposited con-

ventional 20 nm thick NPB HTLs with nanometer-scale thick

SAMs of TPD-Si₂or NPB-Si₂in the above OLEDs was first

investigated, and EL response data are shown in Figure 4. Note

that the OLEDs exhibit greater luminance and, in some regimes,

higher external quantum efficiencies when the conventional 20

nm thick HTL (NPB) is replaced with ∼2 nm thick TPD-Si₂

and NPB-Si₂SAMs. The maximum luminance is ∼18 000 cd/

m²(NPB-Si₂), ∼17 000 cd/m²(TPD-Si₂), and ∼14 000 cd/m²

(NPB-only). Also, the NPB-Si₂and TPD-Si₂SAM-based

OLEDs have lower turn-on and operating voltages. For instance,

the operating voltage at 300 cd/m², a standard brightness for

displays, is 8.0, 8.5, and 10.5 V for NPB-Si₂, TPD-Si₂, and

NPB-only, respectively. However, the maximum external

forward quantum efficiencies are 0.8% (NPB-Si₂), 0.75%

(TPD-Si₂), and 1.87% (NPB-only). OLEDs without HTLs were

also fabricated here as a comparison, and the dramatic effects

on device performance induced by the SAMs are clearly evident

(Figure 4).

Replacing Conventional HTLs with Spin-Cast TPD-Si₂

and NPB-Si₂Films in OLEDs. TPD-Si₂and NPB-Si₂were

spin-coated onto ITO substrates and thermally cured in a vacuum

oven at 110 °C for 1 h, yielding films with a thickness of ∼40

nm, as measured by profilometry, followed by vapor-deposition

of Alq doped with DIQA, then BCP, Li, and Al. Again,

replacing NPB with the cross-linked siloxane films results in

higher luminance and external quantum efficiencies at identical

bias (Figure 5). The maximum luminance achieved is ∼23 000

cd/m²(NPB-Si₂), ∼36 000 cd/m²(TPD-Si₂), and ∼14 000

cd/m²(NPB-only). Maximum forward external quantum ef-

ficiencies are 0.8% (NPB-Si₂), 1.6% (TPD-Si₂), and 1.87%

(NPB-only). The operating bias for 300 cd/m²is 7.0, 7.0, and

10.5 V for NPB-Si₂, TPD-Si₂, and NPB-only, respectively. In

Table 3, the EL response data are compared among devices

using NPB-Si₂and TPD-Si₂SAMs, spin-cast NPB-Si₂and TPD-

Si₂films, and NPB-only as the HTL, respectively. OLEDs with

spin-cast TPD-Si₂films as the HTL exhibit more than a 2-fold

increase of maximum luminance, ∼50% lower turn-on voltage,

and comparable external forward quantum efficiency versus

those of devices having NPB only as the HTL.

Figure 4. Responses of OLEDs having structures ITO/HTL/Alq: 1% DIQA

(60 nm)/BCP:Li (20 nm)/Al (100 nm), HTL ) NPB-Si₂SAM (1.4 nm),

TPD-Si₂SAM (1.7 nm), NPB (20 nm) only. (A) Current density versus

voltage; (B) luminance versus voltage; (C) external forward quantum

efficiency versus voltage.

Applying TPD-Si₂and NPB-Si₂SAMs as ITO Anode-

NPB Interlayers. OLED EL Response. Self-assembled

TPD-Si₂and NPB-Si₂films can also be applied as interlayers

between the ITO anode and the conventional NPB HTL in

OLEDs having structures ITO anode/SAM/NPB/Alq doped with

DIQA/BCP/Li/Al. Insertion of these interlayers between the

anode and the NPB HTL dramatically enhances OLED EL

response as compared to devices without the interlayers (Figure

6). As compared to OLEDs using TPD-Si₂or NPB-Si₂SAMs

as the HTLs, which are directly in contact with the EML (Figure

4), the maximum light output is enhanced by a factor of ∼3×

both for TPD-Si₂- and for NPB-Si₂-based devices: ∼18 000

cd/m²f 50 000 cd/m²(NPB-Si₂/NPB), ∼17 000 cd/m²

f

45 000 cd/m²(TPD-Si₂/NPB). Likewise, the maximum forward

external quantum efficiency is increased to 1.8% (NPB-Si₂/NPB)

and 2.3% (TPD-Si₂/NPB) versus 0.8% (NPB-Si₂), 0.75%

(TPD-Si₂), and 1.87% (NPB alone).

(62) Napper, A. M.; Liu, H. Y.; Waldeck, D. H. J. Phys. Chem. B 2001, 105,

7699-7707.

(63) Nahir, T. M.; Clark, R. A.; Bowden, E. F. Anal. Chem. 1994, 66, 2595-

2598.

(65) Richardson, J. N.; Rowe, G. K.; Carter, M. T.; Tender, L. M.; Curtin, L.

(64) Forster, R. J. Inorg. Chem. 1996, 35, 3394-3403.

S.; Peck, S. R.; Murray, R. W. Electrochim. Acta 1995, 40, 1331-1338.

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10233

A R T I C L E S

Huang et al.

Figure 5. Responses of OLEDs having structures ITO/HTL/Alq: 1% DIQA

(60 nm)/BCP:Li (20 nm)/Al (100 nm), HTL ) spin-cast NPB-Si₂film (40

nm), spin-cast TPD-Si₂film (40 nm), and NPB (20 nm) only. (A) Current

density versus voltage; (B) luminance versus voltage; (C) external forward

quantum efficiency versus voltage.

Figure 6. Responses of OLEDs having structures ITO/HTL SAM/NPB

(20 nm)/Alq: 1% DIQA (60 nm)/BCP:Li (20 nm)/Al (100 nm), HTL SAM

) NPB-Si₂SAM and TPD-Si₂SAM. (A) Current density versus voltage;

(B) luminance versus voltage; (C) external forward quantum efficiency

versus voltage.

Spin-Cast TPD-Si₂and NPB-Si₂Films as ITO Anode-

NPB Interlayers. OLED EL Response. Spin-casting TPD-Si₂

or NPB-Si₂layers onto ITO anodes, followed by vapor-depo-

sition of NPB, was also investigated as an alternative approach

to applying these silanes as the anode-NPB interlayers. Again,

improved EL response is observed versus that of OLEDs with-

out interlayers and versus devices fabricated with only spin-

coated films as HTLs (Figure 7): maximum light output,

∼46 000 cd/m²(NPB-Si₂/NPB), ∼64 000 cd/m²(TPD-Si₂/NPB)

versus ∼23 000 cd/m²(NPB-Si₂), ∼36 000 cd/m²(TPD-Si₂),

∼14 000 cd/m²(NPB-only); maximum forward external quan-

tum efficiencies, 2.4% (NPB-Si₂/NPB), 2.3% (TPD-Si₂/NPB)

versus 0.8% (NPB-Si₂), 1.6% (TPD-Si₂), 1.87% (NPB-only);

turn-on voltages, 4.0 V (NPB-Si₂/NPB), 6 V (TPD-Si₂/NPB)

versus 5.0 V (NPB-Si₂), 5.0 V (TPD-Si₂), 9.0 V (NPB-only).

In Table 4, the EL response data are compared for devices using

NPB-Si₂and TPD-Si₂SAMs, as well as spin-cast NPB-Si₂

and TPD-Si₂films as ITO anode-NPB interlayers, respectively.

Hole-Only Devices. Hole-only devices were fabricated to

investigate the effect of NPB-Si₂and TPD-Si₂SAM coatings

on hole injection efficiency from the ITO anode into the NPB

HTL.⁶⁵I-V characteristics were measured on devices having

structure ITO/siloxane/NPB(400 nm)/Au (6 nm)/Al (Figure 8).

Orders of magnitude larger hole currents are observed for

devices with the siloxane SAMs as ITO anode-NPB interlayers,

indicating strong hole injection enhancement by modifying the

anode with NPB-Si₂and TPD-Si₂, respectively. Attempting to

measure hole fluence through either the self-assembled or the

spin-coated siloxane films in ITO/siloxane/Au/Al structures

yielded irreproducible results, presumably due to the leakage

currents in such device configurations.

3.3. Saturated Alkyl Siloxane SAMs as OLED Anode

Functionalization Layers. Saturated Alkyl Siloxane SAM

HTLs in OLEDs. To further understand the function of

TPD-Si₂, NPB-Si₂, and similar interlayers in OLEDs, saturated

alkylsilanes having varied alkyl chain length (Scheme 2) were

9

10234 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

Table 3. OLED EL Response Data for Devices Having the Structure ITO Anode/HTL/Alq Doped with DIQA/BCP Doped with Li/Al^a

no

NPB-Si₂

SAM

TPD-Si₂

SAM

NPB-Si₂

TPD-Si₂

HTL

(spin-coated)

NPB-only

maximum luminance (cd/m²)

maximum external forward

quantum efficiency (%)

∼1500

∼18 000

∼17 000

∼23 000

∼36 000

∼14 000

0.6

0.8

0.75

0.8

1.6

1.87

turn-on voltage (V)

9.0

10.5

4.6

8.0

5.5

8.5

5.0

7.0

5.0

7.0

9.0

10.5

operating voltage at 300 cd/m²(V)

^aThe HTL is NPB-Si₂SAMs, TPD-Si₂SAMs, spin-cast NPB-Si₂films, spin-cast TPD-Si₂films, and vapor deposited NPB alone, respectively.

Figure 8. Evaluation of hole injection properties of anode SAM function-

alization layers NPB-Si₂and TPD-Si, comparing I-V response for hole-

only devices having the structure ITO/HTL SAM/NPB(400 nm)/Au(6 nm)/

Al (120 nm).

TPD-Si₂SAM-based OLEDs as well as to devices having no

HTL (Figure 9, NPB-Si₂SAM data omitted for clarity; refer to

Figure 4 for comparison between TPD-Si₂and NPB-Si₂SAMs).

Note that the ITO/n-butylsiloxane SAM/Alq OLED exhibits

efficiency and luminance greater than those of the TPD-Si₂

SAM-based devices, with a maximum luminance of ∼26 000

cd/m²and a maximum forward external quantum efficiency

∼1.5% at 15 V. Also, the device performance is competitive

with that of NPB-based OLEDs. Interestingly, pronounced alkyl

chain length effects on the EL response are observed. For in-

stance, the maximum light output is 5000 cd/m²(methyl), 26 000

cd/m²(n-butyl), 13 000 cd/m²(n-octyl), and 60 cd/m²(n-octa-

decyl), and forward external quantum efficiency is 0.4% (meth-

yl), 1.5% (n-butyl), 1.1% (n-octyl), and 0.025% (n-octadecyl).

In Table 5, the EL response is compared among devices using

alkylsiloxane SAMs, a TPD-Si₂SAM, and NPB alone as HTLs.

Saturated Alkyl Siloxane SAMs as ITO Anode-NPB

Interlayers. OLED EL Response. Similar to the experiments

with TPD-Si₂and NPB-Si₂SAMs, alkyl siloxane SAMs were

self-assembled onto ITO anodes, followed by vapor deposition

Figure 7. Responses of OLEDs having structures ITO/spin-cast HTL/NPB

(20 nm)/Alq:1% DIQA (60 nm)/BCP:Li (20 nm)/Al (100 nm), spin-cast

HTL ) spin-cast NPB-Si₂film and TPD-Si₂film. (A) Current density versus

voltage; (B) luminance versus voltage; (C) external forward quantum

efficiency versus voltage.

of 20 nm thick NPB, so that the SAMs function as anode-

NPB interlayers in OLEDs having structures ITO anode/SAM/

NPB/Alq doped with DIQA/BCP/Li/Al. I-V-L characteristics

were measured and are shown in Figure 10. As compared to

OLEDs without anode-HTL interlayers, orders of magnitude

increased (methyl, n-butyl, and n-octyl siloxane SAMs) or

decreased (n-octadecyl siloxane SAM) levels of current density,

light output, and quantum efficiency at identical bias are

observed, indicating dramatic effects on hole injection efficiency

are effected by depositing the nanometer-thick alkylsiloxane

layers on the ITO anode surface. For example, at 10 V, the

light output is 7500 cd/m²(methyl SAM/NPB), 6500 cd/m²(n-

butyl SAM/NPB), 7100 cd/m²(n-octyl SAM/NPB), <1 cd/m²

(n-octadecyl SAM/NPB), and 110 cd/m²(NPB-only). Further-

more, the forward external quantum efficiency is 1.8% (methyl

self-assembled onto ITO surfaces, forming nanometer-thick

SAMs. In contrast and not unexpectedly, smooth and uniform

films could not be fabricated by spin-casting due to inefficient

cross-linking: the silanes have only a single trichlorosilyl group

per molecule. These silanes do not have hole transporting triaryl-

amine motifs as TPD-Si₂and NPB-Si₂, but covalently modify

the ITO anode via silane condensation in a phenomenologically

manner similar to TPD-Si₂and NPB-Si₂. I-V-L characteristics

of OLEDs having the structure ITO anode/alkylsiloxane

SAM/Alq doped with DIQA/BCP/Li/Al were compared to

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10235

A R T I C L E S

Huang et al.

Table 4. OLED EL Response Data for Devices Having the Structure ITO Anode/Interlayer/NPB/Alq Doped with DIQA/BCP Doped with

Li/Al^a

NPB-Si₂

SAM

TPD-Si₂

SAM

NPB-Si₂

TPD-Si₂

(spin-coated)

NPB-only

maximum luminance (cd/m²)

maximum external forward

quantum efficiency (%)

∼50 000

∼45 000

∼46 000

∼64 000

∼14 000

1.8

2.3

2.4

2.3

1.87

turn-on voltage (V)

3.5

5.5

5.0

7.2

4.0

7.2

6.0

7.0

9.0

10.5

operating voltage at 300 cd/m²(V)

^aThe interlayer is an NPB-Si₂SAM, a TPD-Si₂SAM, a spin-cast NPB-Si₂film, or a spin-cast TPD-Si₂film.

4. Discussion

Replacing Conventional HTLs with TPD-Si₂- and

NPB-Si₂-Based Materials. Relationships between ITO Anode-

HTL Contact and OLED Hole Injection Efficiency and EL

Response (Turn-on Voltage, Luminance). It has been dem-

onstrated here that TPD-Si₂and NPB-Si₂, with hole transporting

structural motifs similar to those of traditional triarylamine HTLs

such as TPD and NPB, respectively, can be self-assembled or

spin-cast onto the ITO anode surface, utilizing the reactive

trichlorosilyl functionalities to covalently graft the structure to

the surface. OLEDs having TPD-Si₂- and NPB-Si₂-based

materials as the HTLs exhibit good EL response (Figures 4 and

5, Table 3). As new HTL materials, TPD-Si₂and NPB-Si₂fulfill

the generally accepted requirements for efficacious HTLs: (1)

they possess electronic levels energetically comparable to those

of TPD and NPB according to UPS measurements (Scheme 5);

(2) they form smooth, conformal, uniform, and electroactive

films on ITO surfaces; (3) the films are strongly adherent and

thermally stable,⁶⁷and (4) the degree of physical contact

achieved between these HTLs and the ITO surface is not avail-

able to conventional HTLs. To date, in all reported polymeric

and in many reported small molecule OLEDs, HTLs such as

TPD, NPB, and poly(3,4-ethylenedioxythiophene)-poly(sty-

renesulfonate) (PEDOT-PSS) (Scheme 1) are deposited directly

onto the ITO anode by either thermal evaporation or spin-

coating, resulting in anode-HTL contact that is largely physical

in character. This interface is prone to large contact resistance,

which is detrimental to OLED EL response.⁶⁸Furthermore,

thermally induced delamination/dewetting is also observed at

ITO anode-TPD or NPB interfaces due to severe surface energy

mismatches, ^69-71and in some cases, traditional HTL materials

cause extensive ITO corrosion.^72,73Thermally induced interfacial

morphology changes are also observed on copper phthalocyanine

buffer layers.^69,74In contrast, TPD-Si₂and NPB-Si₂form

covalent bonds to the ITO surface, ensuring strong adhesion

and, via close surface energy matching, intimate ITO anode-

HTL physical and electrical contact.⁴⁷The improved light output

and hole injection efficiency, and reduced turn-on voltages

versus conventional ITO/NPB interfaces (Figures 4, 5, and 8),

demonstrate that such contact is crucial to good device

performance and should be considered in designing new HTL

materials. Interestingly, the magnitudes of hole current across

Figure 9. Responses of OLEDs having structures ITO/HTL SAM/Alq:

1% DIQA (60 nm)/BCP:Li (20 nm)/Al (100 nm); HTL ) methylsiloxane

SAM (C1), n-butylsiloxane SAM (C4), n-octylsiloxane SAM (C8), n-octa-

decylsiloxane SAM (C18), TPD-Si₂, and NPB. (A) Current density versus

voltage; (B) luminance versus voltage; (C) external forward quantum

efficiency versus voltage. Lines through the data points are drawn as a guide

to the eye.

SAM/NPB), 1.5% (n-butyl SAM/NPB), 1.5% (n-octyl SAM/

NPB), 0.24% (n-octadecyl SAM/NPB), and 0.5% (NPB-only).

Similar to using a TPD-Si₂SAM as the anode-NPB interlayer,

the alkylsiloxane SAMs with short alkyl chains result in

significant OLED performance enhancement as ITO anode-

NPB interlayers. In Table 6, EL response is compared among

devices using the alkylsiloxane SAMs and a TPD-Si₂SAM as

ITO anode-NPB interlayers.

(66) Crone, B. K.; Campbell, I. H.; Davids, P. S.; Smith, D. L. Appl. Phys.

Lett. 1998, 73, 3162-3164.

(67) Huang, Q. L.; Cui, J.; Marks, T. J. Abstr. Pap. Am. Chem. Soc. 2002, 224,

POLY 032-033.

(68) Shen, Y.; Klein, M. W.; Jacobs, D. B.; Scott, J. C.; Malliaras, G. G. Phys.

ReV. Lett. 2001, 86, 3867-3869.

(69) Cui, J.; Huang, Q.; Veinot, J. C. G.; Yan, H.; Wang, Q.; Hutchison, G. R.;

Richter, A. G.; Evmenenko, G.; Dutta, P.; Marks, T. J. Langmuir 2002,

18, 9958-9970.

(70) Cui, J.; Huang, Q.; Veinot, J. G. C.; Yan, H.; Marks, T. AdV. Mater. 2002,

14, 565-567.

9

10236 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

Table 5. Comparative OLED EL Response Data for Devices Having the Structure ITO Anode/HTL/Alq Doped with DIQA/BCP Doped with

Li/Al^a

methylsiloxane

SAM

n-butylsiloxane

SAM

n-octylsiloxane

SAM

n-octadecylsiloxane

SAM

TPD-Si₂

SAM

HTL

NPB-only

maximum luminance (cd/m²)

maximum external forward

quantum efficiency (%)

∼5000

∼26 000

∼13 000

∼60

∼17 000

∼14 000

0.4

1.5

1.1

0.025

0.75

1.87

turn-on voltage (V)

7.5

12.5

7.5

9.5

6

9.5

12

5.5

8.5

9

10.5

operating voltage at 300 cd/m²(V)

^aHTL ) a methylsiloxane SAM, an n-butylsiloxane SAM, an n-octylsiloxane SAM, an n-octadecylsiloxane SAM, a TPD-Si₂SAM, and vapor-deposited

NPB alone.

Scheme 5. Energy Diagram for the Electrodes and Organic

Layers Employed in This Study (Taken from Literature Data)

Efficiency. Note that replacing a conventional HTL (NPB) with

very thin siloxane SAMs results in devices with significantly

diminished external forward quantum efficiency. This is at-

tributed to a possible efficiency loss mechanism: as compared

to ∼20 nm thick NPB HTL, TPD-Si₂and NPB-Si₂SAMs are

likely to have reduced electron-blocking properties due to their

nanometer scale thickness. It has been recently argued that a

minimum of ∼20 nm NPB is required to effectively block

electrons from unproductively leaking to the ITO anode.⁷⁵

Possibly the ∼1.5 nm thick TPD-Si₂and NPB-Si₂SAMs are

not sufficiently thick to prevent all electrons from escaping the

emissive HTL/EML (Alq) interfacial zone. This hypothesis is

supported by the device efficiency recovery in the case of ∼40

nm thick spin-coated TPD-Si₂as the HTL, due to increased

HTL thickness and, within this explanation, more efficient

electron-blocking properties that outweigh thickness effects.

Also, the dramatic efficiency increase in OLEDs having the

SAMs as ITO anode-HTL interlayers versus as the only HTLs

(Figure 6 vs Figure 4) can be explained on the basis of the

superior electron-blocking properties of the combined NPB +

SAM layer. Here, the enhanced ITO-NPB physical/electrical

contact provided by the SAM is combined with the far more

favorable electron-blocking characteristics. However, the above

explanation does not completely explain the modest external

forward quantum efficiency gain when ∼1.5 nm NPB-Si₂SAM

is replaced with a ∼40 nm spin-coated NPB-Si₂film (Figures

4, 5 and Table 3). We speculate there are other factors

influencing the efficiency here. One reasonable explanation is

Figure 10. Responses of OLEDs having the structures ITO/HTL SAM/

NPB (20 nm)/Alq: 1% DIQA (60 nm)//BCP:Li (20 nm)/Al (100 nm), HTL

SAM ) methylsiloxane SAM (C1), n-butylsiloxane SAM (C4), n-octyl-

siloxane SAM (C8), n-octadecylsiloxane SAM (C18), and TPD-Si₂. (A)

Current density versus voltage; (B) luminance versus voltage; (C) external

forward quantum efficiency versus voltage.

the ITO anode-TPD-Si₂or NPB-Si₂SAM interfaces are very

similar, indicating that the contact rather than the details of hole

transporting motif must play a significant role. However, the

ITO/siloxane SAM OLEDs have somewhat lower forward

external quantum efficiencies than conventional ITO/NPB

devices (Figure 4 and Table 3), which is explained below.

Replacing Conventional HTLs with TPD-Si₂- and NPB-

Si₂-Based Materials. Effects on External Forward Quantum

(71) Cui, J.; Wang, Q.; Marks, T. J. Polym. Mater. Sci. Eng. 2000, 83, 239-

240.

(72) de Jong, M. P.; de Voigt, M. J. A. Appl. Phys. Lett. 2000, 77, 2255-2257.

(73) Ni, J.; Wang, A.; Yang, Y.; Stern, C. L.; Metz, A. W.; Jin, S.; Wang, L.;

Marks, T. J.; Kannewurf, C. R. J. Am. Chem. Soc. 2005, 127, 5613-5624.

(74) Xu, M. S.; Xu, J. B. J. Phys. D: Appl. Phys. 2004, 37, 1603-1608.

(75) Zhang, S. T.; Wang, Z. J.; Zhao, J. M.; Zhan, Y. Q.; Wu, Y.; Zhou, Y. C.;

Ding, X. M.; Hou, X. Y. Appl. Phys. Lett. 2004, 84, 2916-2918.

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10237

A R T I C L E S

Huang et al.

Table 6. Comparative OLED EL Response Data for Devices Having the Structure ITO Anode/Interlayer/NPB/Alq Doped with DIQA/BCP

Doped with Li/Al^a

methylsiloxane

SAM

n-butylsiloxane

SAM

n-octylsiloxane

SAM

n-octadecylsiloxane

SAM

TPD-Si₂

SAM

interlayer

NPB-only

maximum luminance (cd/m²)

maximum external forward

quantum efficiency (%)

∼30 000

∼25 000

∼28 000

∼3500

∼45 000

∼14 000

1.8

1.5

0.5

2.3

1.87

turn-on voltage (V)

4.2

6.2

4.0

6.7

4.5

7.0

11

16.8

5.0

7.2

9

10.5

operating voltage at 300 cd/m²(V)

^aThe interlayer is a methylsiloxane SAM, an n-butylsiloxane SAM, an n-octylsiloxane SAM, an n-octadecylsiloxane SAM, and a TPD-Si₂SAM.

concerned with the mechanism of siloxane film formation on

hydroxylated surfaces,^49,76in which incomplete thermal cross-

linking may result in residual, uncondensed silanol functional-

ities.⁷⁷Such Si-OH groups are known to induce the chemical

degradation of Alq^78,79and are likely to react with mobile

electrons and holes. Furthermore, it has been reported that

radical ions generated under high electric fields in operating

OLEDs can quench fluorescent dye dopant emission via direct

resonant energy transfer from excited states to quenching

centers.⁸⁰It is conceivable that Si-OH functionalities act as

quenchers.⁸¹The energy transfer rate is related to the excited

Figure 11. FTIR spectra of 40 nm TPD-Si₂and NPB-Si₂films spin-cast

and cured on glass substrates under the same conditions (bare glass substrate

absorption baseline subtracted).

state-quenching center distance by eq 3, where r is the energy

transfer rate, d is the separation between excited states and

quenching centers, and x depends on the nature of the interaction

between excited states and quenching centers.⁸⁰

state quenching. In this case, the effects of increased electron-

blocking with greater TPD-Si₂film thickness are dominant,

r

d^-x

(3)

affording enhanced external forward quantum efficiency via

replacing a TPD-Si₂SAM HTL with a spin-coated one (Table

3). In this scenario, the excitons must be spatially removed from

the silanol functionalities to minimize reaction and quenching,

which can be realized by using the siloxanes as ITO anode-

NPB interlayers, as discussed below.

When siloxanes are used as HTLs, the EML (Alq) is

deposited directly on top of siloxane films, forming a HTL/

EML interface where hydroxyl groups may be present. Because

the Alq electron mobility is orders of magnitude greater than

the hole mobility, holes and electrons should primarily recom-

bine at or near HTL/EML interface, forming emissive excitons.⁸²

Some of these may be quenched/decomposed by hydroxyl

groups in this region, via the aforementioned mechanisms. The

observation that thicker spin-coated NPB-Si₂HTL does not

significantly improve external forward quantum efficiency

(Table 3) may be attributable to the presence of uncondensed

hydroxyl functionalities functioning as impurities/quenching

centers.^78,83The quenching effect appears to work against the

increased electron-blocking properties of the thicker NPB-Si₂

films and results in the overall similar external forward quantum

efficiencies for SAM and spin-coated NPB-Si₂HTL-based

OLEDs. TPD-Si₂may undergo more complete cross-linking as

indicated by the more distorted cyclic voltammetric curve shape

than that of NPB-Si₂(Figure 3) and the weaker intensity of the

silanol OH vibration transitions at 3150-3700 cm^-1in FTIR

spectra (Figure 11), suggesting more complete siloxane forma-

tion and correspondingly diminished reactivity and emissive

Self-Assembled and Spin-Coated TPD-Si₂and NPB-Si₂

Materials as ITO Anode-NPB Interlayers. As shown in

Figures 6, 7 and Table 4, significantly greater EL response,

including forward external quantum efficiency, is achieved by

using the TPD-Si₂or NPB-Si₂materials as the anode-NPB

interlayers versus applying them as the only HTL. This supports

the above discussion because (1) a ∼20 nm thick NPB layer

deposited on the interlayers enhances the electron-blocking

properties; (2) the NPB/Alq interface replaces the TPD-Si₂/Alq

or NPB-Si₂/Alq interface when the siloxanes are used as the

ITO anode-NPB interlayer, and thus the exciton recombination

region is at or very near the NPB/Alq interface, tens of

nanometers away from the siloxane structure. Any reactivity

or exciton quenching by silanol functionalities should be greatly

reduced according to eq 3.

In addition, the substantial OLED EL response enhancement

obtained by inserting the various siloxane structures between

the ITO anode and the NPB HTL, as compared to ITO/NPB,

can be plausibly explained on the basis of the current under-

standing of ITO anode-HTL interfacial function. To date, a

variety of explanations for the effects of ITO anode-HTL

interlayers have been proposed, including increased hole injec-

tion via dipole moment-induced additional electric fields,³⁰

reduced injected charge backscattering,³²decreased anode work

function-HTL HOMO energetic barriers,⁸⁴increased recombina-

tion efficiency via balancing electron and hole injection,^31,85and/

(76) Silberzan, P.; Leger, L.; Ausserre, D.; Benattar, J. J. Langmuir 1991, 7,

1647-1651.

(77) Zhao, X.; Kopelman, R. J. Phys. Chem. 1996, 100, 11014-11018.

(78) Liao, L. S.; He, J.; Zhou, X.; Lu, M.; Xiong, Z. H.; Deng, Z. B.; Hou, X.

Y.; Lee, S. T. J. Appl. Phys. 2000, 88, 2386-2390.

(79) Xu, M. S.; Xu, J. B.; Chen, H. Z.; Wang, M. J. Phys. D: Appl. Phys. 2004,

37, 2618-2622.

(80) Young, R. H.; Tang, C. W.; Marchetti, A. P. Appl. Phys. Lett. 2002, 80,

874-876.

(81) Schaer, M.; Nesche, F.; Berner, D.; Leo, W.; Zuppiroli, L. AdV. Funct.

Mater. 2001, 11, 116-120.

(82) Kondakov, D. Y.; Sandifer, J. R.; Tang, C. W.; Young, R. H. J. Appl.

Phys. 2003, 93, 1108-1119.

(83) Aziz, H.; Popovic, Z.; Tripp, C. P.; Hu, N.; Hor, A.; Xu, G. Appl. Phys.

Lett. 1998, 72, 2642-2644.

(84) Lee, S. T.; Wang, Y. M.; Hou, X. Y.; Tang, C. W. Appl. Phys. Lett. 1998,

74, 670-672.

9

10238 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

or confining electrons to the emissive layer.³³Surface energy-

promoted wetting/cohesion at the ITO-organic layer interface

is also known to be a significant factor.^1,43,70,74According to

these arguments, we propose that the TPD-Si₂and NPB-Si₂

materials, when employed as ITO anode-NPB interlayers,

increase hole injection principally by: (1) reducing the hole

injection barrier from the ITO anode to NPB (HOMO ∼ 5.5

eV) by providing an energy-mediating step (Table 2); (2)

enhancing ITO anode-NPB interfacial cohesion via eliminating

the surface energy mismatch;⁶⁹and (3) contributing to the

aforementioned electron confinement effects in conjunction with

NPB. Any interfacial dipole effects would appear to be relatively

small as indicated by the small UPS-determined (0.1 eV

vacuum level shift (Table 2), and cannot be fully responsible

for the dramatic EL response enhancement evident in Figures

6, 7 and in Table 4. The effect of TPD-Si₂and NPB-Si₂

interlayers on balancing electron and hole injection will be

discussed below. Overall, these factors contribute to improved

hole injection and device performance.

Figure 12. Comparative cyclic voltammetry characteristics of self-

assembled alkylsiloxane-coated ITO electrodes in ferrocene solution. The

SAMs are methylsiloxane SAM (C1), n-butylsiloxane SAM (C4), n-

octylsiloxane SAM (C8), and n-octadecylsiloxane SAM (C18). SAM-coated

ITO, silver wire, and Pt wire were used as the working electrode, reference

electrode, and counter electrode, respectively. All experiments were carried

out in 0.1 M acetonitrile solution of tetrabutylammonium hexafluorophos-

phate as the electrolyte and 0.001 M ferrocene as the internal pinhole probe

at scan rate 0.1 V/s.

Alkylsiloxane SAMs as HTLs in OLEDs. In a conventional

picture, OLED response should be eroded in the absence of a

triarylamine functionality interposed between the anode and the

EML/ETL. Interestingly, however, we find that replacing NPB

with n-butylsiloxane SAMs (Scheme 2) results in appreciable

increase in EL response (maximum light output ) 26 000

cd/m²(n-butyl) vs 16 000 cd/m²(NPB-only); forward external

quantum efficiency 1.5% (n-butyl) vs 1.9% (NPB-only), Figure

9). In contrast, replacing NPB with methylsiloxane or n-octa-

decylsiloxane leads to significantly diminished EL response

(Figure 9, Table 5). The poor EL response of OLEDs having

the n-octadecylsiloxane SAM can be readily explained from

established relationships between distance-dependent charge

carrier tunneling efficiency and alkylsiloxane SAM chain

length.^86-88Here, tunneling is believed to be the primary charge

transport mechanism through alkylsiloxane SAMs and follows

eq 4:^88,89

for effective hole tunneling from the ITO anode through the

n-octadecylsiloxane SAM into the Alq emissive layer.

Further discussion is appropriate to explain the lower forward

external quantum efficiencies of OLEDs using the very thin

methylsiloxane SAM as the HTL versus n-butyl- and n-octyl-

siloxane SAMs (Figure 9), even though all of the SAMs should

support efficient charge carrier tunneling according to the above

arguments. We suggest that methylsiloxane SAM has poorer

electron-blocking properties than n-butyl- and n-octylsiloxane

SAMs due to, among other factors, greater defect densities. It

is reported that SAM defects are correlated with the alkyl chain

length, with longer chains leading to SAMs with less defects.⁹⁴

In ITO/SAM/Alq OLEDs, the SAM defects may provide

pathways for electrons to escape the SAM/Alq interfacial region,

and reduce the SAM electron-blocking function, which is a

critical requirement for high-performance HTLs.⁷⁵Greater defect

densities for the methylsiloxane SAM than for the n-butyl- or

n-octylsiloxane SAM would result in less efficient electron-

blocking and hole-electron recombination, consequently lower-

ing device efficiency.

i ) i₀exp(-âd)

(4)

where d is the thickness of barrier through which the electron

tunneling takes place, and â is the tunneling coefficient (0.64-

1.16^-1for alkanethiol monolayers on gold⁹⁰). The thickness of

the methylsiloxane, n-butylsilxoane, n-octyl-, and n-octadecyl-

siloxane SAMs is reported to be 2.8, 6.2, 8.0, and 23 Å,⁹¹

respectively. In the case of the SAMs with short alkyl chains,

it is reasonable that significant hole injection from the ITO anode

into the EML can occur via tunneling through the very thin

SAMs, while in the case of the n-octadecylsiloxane SAM, the

tunneling becomes inefficient and the insulating properties

dominate.^88,92,93This can explain the dramatically decreased

current density, light output, and forward external quantum

efficiency seen in Figure 9, due to the significant energy barrier

To probe the presence of SAM defects, cyclic voltammetry

experiments using ferrocene as a redox probe were carried out

using the above SAM-coated ITO substrates as working

electrodes (Figure 12). The results of these experiments are

consistent with the above hypothesis. In the case of the

methylsiloxane SAM, ferrocene redox peaks are observed, which

are largely unchanged from those observed with bare ITO,

suggesting significant defect densities are present in the SAM

layer, which provide uninhibited ferrocene oxidation at the ITO

electrode surface. In the cases of n-butyl and n-octyl siloxane

SAMs, the apparent defect densities vary significantly with film

deposition conditions. Presumably, trace amounts of water have

large effects on SAM quality.⁴⁹SAMs with lower pinhole

densities, indicated by the shifted ferrocene oxidation peak

potential, lead to higher OLED efficiency seen in Figure 9.

Finally, the n-octadecylsiloxane SAM extensively passivates the

ITO surface, as judged both by the large separation between

(85) Forsythe, E. W.; Abkowitz, M. A.; Gao, Y. J. Phys. Chem. B 2000, 104,

3948-3952.

(86) Chidsey, C. E. D.; Bertozzi, C. R.; Putvinski, T. M.; Mujsce, A. M. J. Am.

Chem. Soc. 1990, 112, 4301-4306.

(87) Smalley, J. F.; Feldberg, S. W.; Chidsey, C. E. D.; Linford, M. R.; Newton,

M. D.; Liu, Y. P. J. Phys. Chem. 1995, 99, 13141-13149.

(88) Chidsey, C. E. Science 1991, 251, 919-922.

(89) Parker, I. D. J. Appl. Phys. 1994, 75, 1656-1666.

(90) Zhao, J.; Uosaki, K. J. Phys. Chem. B 2004, 108, 17129-17135 and

references therein.

(92) Wang, B.; Luo, J.; Wang, X.; Wang, H.; Hou, J. G. Langmuir 2004, 20,

5007-5012.

(93) Lercel, M. J.; Whelan, C. s.; Craighead, H. G.; Seshadri, K.; Allara, D. L.

J. Vac. Sci. Technol., B 1996, 14, 4085-4090.

(91) Jin, Z. H.; Vezenov, D. V.; Lee, Y. W.; Zull, J. E.; Sukenik, C. N.; Savinell,

R. F. Langmuir 1994, 10, 2662-2671.

(94) Ohtake, T.; Mino, N.; Kazufumi, O. Langmuir 1992, 8, 2081-2083.

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10239

A R T I C L E S

Huang et al.

ferrocene anodic and cathodic peaks and by the overall

magnitude of current flowing at any potential.⁹⁵Note that the

aqueous advancing contact angles for the above SAMs on ITO

glass substrates are all similar (90 ( 5°), making the correlation

between SAM defects and contact angle indexed surface

energies inconclusive.⁷⁶

The SAM alkyl chain length effects on OLED response are

also consistent with image force effects. It is generally accepted

that an image force model can be applied to charge injection

from electrodes into organic solids.⁹⁶Here, injected charges

induce counter charges on the electrode surface, which in turn

exert a force on charges being injected away from the electrodes.

This so-called image force effect has both negative and positive

characteristics with respect to charge injection. At close prox-

imities to the electrode (e1 nm), the charge backscattering

region, charges experience strong Columbic forces attracting

them back to the electrode, thereby retarding injection. However,

if the charges escape this region, they experience a lower barrier

and injection is facilitated. The thickness of the present n-butyl-

and n-octylsiloxane SAMs are in the dimensional length scale

of typical charge backscattering regions (∼1 nm).⁹⁶We suggest

that holes can therefore be injected with a low barrier substan-

tially through or beyond the backscattering region of the image

force field. In this way, hole injection is enhanced as compared

to that of bare ITO. Thus, the n-butylsiloxane and n-octyl SAMs

reduce the negative effect of the image force in terms of hole

injection. In the case of the thinner (∼2.8 Å) methylsiloxane

SAM deposited on the ITO surface, the negative side of the

image force effect should dominate and should retard hole

injection. The total current observed in the I-V characteristics

of Figure 9 is the sum of the injection current and of the leakage

current near either the hole or the electron injection contacts.⁹⁷

Based on the above discussion of SAM defects and image force

effects, n-butyl- and n-octylsiloxane SAMs would have larger

injection currents and smaller leakage currents versus those of

methylsiloxane SAM, and larger total currents, which explains

relative the I-V, L-V, and quantum efficiency characteristics

in Figure 9.

Comparison of Alkylsiloxane SAMs to TPD-Si₂and NPB-

Si₂SAMs as HTLs. Evidence of Significant ITO Anode-

HTL Contact Effects on OLED EL Response. The simple

treatment of the ITO surface with a saturated hydrocarbon

n-butyl siloxane monolayer results in a dramatic OLED EL

response enhancement (ITO/SAM/Alq vs ITO/Alq OLEDs), and

device performance roughly comparable to NPB-only HTL-

based OLEDs (ITO/SAM/Alq vs ITO/NPB/Alq OLEDs). These

results demonstrate that conventional hole transport layers may

be deleted from small molecule OLED structures without major

degradative effects on quantum efficiency and luminance,

conveying fundamental implications for OLED and HTL

design.⁹⁸The favorable performance data for OLEDs fabricated

with the alkylsiloxane or triarylamine siloxane-modified ITO

anodes and efficient electron injection systems indicate that ITO

anode-HTL chemical/physical/electrical contact is a significant

factor contributing to effective HTL function, considering the

Figure 13. Comparison of responses of OLEDs having the structures ITO/

HTL/Alq: 1% DIQA (60 nm)/Al to ITO/(HTL)/Alq: 1% DIQA (60 nm)/

BCP:Li (20 nm)/Al, HTL ) n-butylsiloxane SAM (C4), TPD-Si₂SAM

(TPD-Si₂), spin-cast TPD-Si₂(SC-TPD-Si₂), and NPB alone. (A) Current

density versus voltage; (B) luminance versus voltage; (C) external forward

quantum efficiency versus voltage.

very different molecular and electronic structures of n-butylsi-

loxane and TPD-Si₂, excepting the trichlorosilyl tethers, which

covalently graft them to the ITO anode. By carefully controlling

the chemisorption process, a simple alkylsiloxane monolayer

of appropriate thickness and defect density can replace a

conventional HTL constituent with a much more complex

structure. This provides motivation to shift research focus from

developing hole-transporting materials with new hole transport

motifs, to modifying currently available molecules to form

stable, robust contacts with ITO anodes.

Alkylsiloxane SAMs as ITO Anode-NPB Interlayers in

OLEDs. As noted above, introducing alkylsiloxane SAMs with

short alkyl chain lengths as ITO anode-NPB interlayers results

in dramatic device performance enhancements versus applying

them as the only HTL (Figure 10), which can be explained by

a function similar to that of TPD-Si₂and NPB-Si₂as anode-

NPB interlayers because (1) direct ITO surface-EML contact

in ITO/SAM/Alq OLEDs via SAM defects is avoided by the

(95) Finklea, H. O. In Electroanalytical Chemistry; Bard, A. J., Rubenstein, I.,

Eds.; Marcel Dekker: New York, 1996; Vol. 19, pp 177-194.

(96) Tutis, E.; Bussac, M. N.; Zuppiroli, L. Appl. Phys. Lett. 1999, 75, 3880-

3882.

(97) Khramtchenkov, D. V.; Bssler, H.; Arkhipov, V. I. J. Appl. Phys. 1996,

79, 9283-9290.

(98) Chan, I. M.; Hong, F. C. Thin Solid Films 2004, 450, 304-311

9

10240 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Covalently Bound Hole-Injecting Nanostructures

A R T I C L E S

Figure 14. Comparison of responses of OLEDs having structures ITO/(siloxane)/NPB (20 nm)/Alq: 1% DIQA (60 nm)/Al to ITO/(siloxane)/NPB (20

nm)/Alq: 1% DIQA (60 nm)//BCP:Li (20 nm)/Al, siloxane ) n-butylsiloxane SAM (C4), TPD-Si₂SAM (TPD-Si₂), spin-cast TPD-Si₂(SC-TPD-Si₂). (A)

Current density versus voltage; (B) luminance versus voltage; (C) external forward quantum efficiency versus voltage; (D) external forward quantum efficiency

versus voltage for OLEDs having structures ITO/(siloxane)/NPB/Alq: 1% DIQA/Al.

20 nm thick NPB layer, and electrons should be largely confined

to the NPB/Alq interface, increasing the probability of hole-

electron recombination, hence EL response; and (2) exciton

quenching and/or Alq degradation associated with any remaining

silanol functionalities is also avoided by depositing NPB on

top of the SAMs. The present device performance improvement

versus ITO/NPB devices is therefore primarily attributable to

improved ITO anode-NPB contact via surface energy match-

ing.¹⁵In contrast, employing the thicker n-octadecylsiloxane

SAM as an ITO anode-NPB interlayer blocks hole injection,

thereby reducing hole-electron recombination efficiency and

diminishing device performance.

Balancing Electron and Hole Injection in OLEDs Having

Siloxanes as HTLs or ITO Anode-NPB Interlayers. In the

Figure 15. External forward quantum efficiency versus voltage for

OLEDs fabricated with self-assembled or spin-coated triaryl-

aminosiloxanes or alkylsiloxanes as HTLs, the effects of elec-

tron injection/transport and hole-electron recombination on

device performance are very significant (Figure 13). Replacing

Alq/Li-doped BCP/Al with simple Alq/Al results in a larger

electron injection barrier and lower electron density in the charge

carrier recombination region, which diminishes device efficiency

and light output. A similar trend is observed when siloxanes

are applied as ITO anode-NPB interlayers (Figure 14). Also,

note that inserting siloxane interlayers enhances external forward

quantum efficiency in the electron-limited Alq/Al OLEDs

(Figures 14d and 15), requiring further examination of the

OLEDs having structures ITO/n-octadecylsiloxane SAM/NPB/Alq: 1%

DIQA/Al.

relationship between hole injection and quantum efficiency,

particularly in an electron injection-limited scenario.^{28,31,70,85,99}

Suppression of hole injection is known to ameliorate the

imbalance of major (hole) and minor (electron) charge carrier

densities near the HTL/EML interface, improving quantum

efficiency.^27,31However, a number of research groups have

experimentally observed and theoretically predicted regimes

(99) Zhou, X.; Pfeiffer, M.; Blochwitz, J.; Werner, A.; Nollau, A.; Fritz, T.;

Leo, K. Appl. Phys. Lett. 2001, 78, 410-412.

9

J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005 10241

A R T I C L E S

Huang et al.

having increased quantum efficiencies with enhanced hole

injection in electron-limited devices.^70,100-103The former regime

is observed when n-octadecylsiloxane SAM is inserted as

ITO anode-NPB interlayer, and the latter is observed when

TPD-Si₂and n-butylsiloxane are used as the interlayers, arguing

that holes are injected and accumulated at the HTL/EML inter-

face (aided by the hole/exciton-blocking characteristics of BCP),

thus altering the internal electrical field, enhancing the electric

field drop across the ETL, and thereby enhancing electron

injection.¹⁰⁰

robust covalent bonding. This unique contact effect distinguishes

the present HTL materials from traditional ones and results in

dramatic OLED device performance enhancement. It is also

demonstrated here that alkylsiloxane SAMs can function as

HTLs, even though they lack classical hole transport motifs,

provided that suitable alkyl chain lengths are chosen. This further

evidences the important role of ITO anode-HTL contact in HTL

design. The new HTL materials developed here can also be

applied as ITO anode-NPB interlayers, further enhancing

device performance and representing an effective approach to

fabricating OLEDs with high brightness (maximum ∼64 000

cd/m²), low operating voltage (∼7 V at 300 cd/m²), and large

current efficiencies (∼8 cd/A).

5. Conclusions

ITO anode-HTL contact plays a significant role in OLED

EL response. A new strategy in HTL materials development is

presented here, focusing on chemical/physical/electrical contact

rather than on HTL molecular architecture. In this contribution,

traditional HTL molecules such as TPD and NPB are modified

with trichlorosilyl groups and can be self-assembled or spin-

coated onto the ITO surface, enhancing ITO-HTL contact via

Acknowledgment. We thank the USDC and NSF through

the Northwestern MRSEC (DMR-0076097) for support of this

research. We thank Prof. M. Ratner, Dr. A. Burin, and Dr. He

Yan for stimulating discussions.

Supporting Information Available: X-ray reflectivity data

normalized to the Fresnel reflectivity, for a film formed by self-

assembly of TPD-Si₂and NPB-Si₂on Si, respectively. This

material is available free of charge via the Internet at

http://pubs.acs.org.

(100) Khramtchenkov, D. V.; Bssler, H.; Arkhipov, V. I. J. Appl. Phys. 1996,

79, 9283-9290.

(101) Burrows, P. E.; Shen, Z.; Bulovic, V.; McCarty, D. M.; Forrest, S. R.;

Cronin, J. A.; Thompson, M. E. J. Appl. Phys. 1996, 79, 7991-8006.

(102) Yang, S.; Wang, Z.; Xu, Z.; Hou, Y.; Xu, X. Chem. Phys. 2001, 274,

267-273.

(103) Giebeler, C.; Antoniadis, H.; Bradley, D. D. C.; Shirota, Y. Appl. Phys.

Lett. 1998, 72, 2448.

JA051077W

9

10242 J. AM. CHEM. SOC. VOL. 127, NO. 29, 2005

Article Doi

Covalently bound hole-injecting nanostructures. Systematics of molecular architecture, thickness, saturation, and electron-blocking characteristics on organic light-emitting diode luminance, turn-on voltage, and quantum efficiency

DOI: 10.1021/ja051077w

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja051077w

Full text of DOI:10.1021/ja051077w

Products guided by the article

R&D Labs maybe for 863090-19-1

Relevant to this article

Hot Product