C A R B O N 5 0 ( 2 0 1 2 ) 1 7 0 – 1 7 4

.sc iencedi rect .com

Avai lab le at www

journal homepage: www.elsev ier .com/ locate /carbon

Effect of field enhancement on inorganic powderelectroluminescence using short carbon nanotubes

Jin-Young Kim a, Min Jong Bae b,c, Shang Hyeun Park b,*, Taewon Jeong b, Sunjin Song b,Jeonghee Lee b, Intaek Han b, Ji Beom Yoo c, Donggeun Jung a, SeGi Yu d,*

a Department of Physics, BK 21 Physics Research Division, Sungkyunkwan University, Suwon 440-746, Republic of Koreab Electronics Material Lab., SAIT Samsung Electronics, Yongin 446-712, Republic of Koreac SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University, Suwon 440-746, Republic of Koread Department of Physics, Hankuk University of Foreign Studies, Yongin 449-791, Republic of Korea

A R T I C L E I N F O

Article history:

Received 4 June 2010

Accepted 11 August 2011

Available online 7 September 2011

0008-6223/$ - see front matter � 2011 Elsevidoi:10.1016/j.carbon.2011.08.016

* Corresponding authors: Fax: +82 31 280 816E-mail addresses: sh100.park@samsung.c

A B S T R A C T

Inorganic powder electroluminescence (IPEL) devices with the insertion of a carbon nanotube

(CNT) layer were investigated to verify the effect of the increased local field produced by CNTs

on electroluminescence (EL). To increase the field strength effectively, the CNTs were short-

ened using the cryogenic crushing method. IPEL devices with the insertion of a short CNT

layer exhibited an increase in brightness and efficiency with increasing amount of CNTs.

The local field enhancement by CNTs, further enlarged by the triple-junction, could increase

the field strength applied to the phosphor, resulting in improved EL performance. In addition,

short CNTs in an EL device can lead to field enhancement without an unintentional current

flowing into the device.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Carbon nanotubes (CNTs) are considered to be attractive field

emitters, because of their high electrical conductivity and

high aspect ratio with nanometer sizes [1–3]. Studies on their

applications as electronic devices have produced field emis-

sion displays [4] and backlight units for liquid crystal displays

[5]. Recently, another type of field enhancement effect of

CNTs was reported in the field of inorganic powder electrolu-

minescence (IPEL), where a mixture of CNTs and cathodolu-

minescent (CL) phosphor can produce electroluminescence

(EL) along with the local field enhancement of CNTs [6]. On

the other hand, the device using the CL phosphor operated

at relatively high driving voltages (over 250 V) and exhibited

a low intensity of EL compared to conventional IPEL devices,

indicating the importance of field enhancement in IPEL

devices.

er Ltd. All rights reserved

7 (S.H. Park), +82 31 330 4om (S.H. Park), segiyu@h

In general, EL is a phenomenon of light emission of a mate-

rial under the influence of an electric field. The first inorganic

EL was observed from oil suspended ZnS powder under an

alternating current (AC) voltage by Destriau [7]. An EL device

is usually a sandwiched structure of a light-emitting layer be-

tween two electrodes. Among the many EL device structures,

AC-type thick-film IPEL structures have attracted considerable

attention as a potential candidate for illumination and display

applications, due primarily to the advantages in the substrate

flexibility and process, such as the simple fabrication of

large-sized panel, low cost manufacturing and no requirement

of vacuum processing [8–11]. In general, the structure of an IPEL

device is composed of an electrode/dielectric/phosphor/elec-

trode/substrate [12,13], and each component was fabricated

using the screen-printing method [14]. Despite the simple fab-

rication processes in IPEL devices, the actual application of IPEL

devices have been confined to backlight units in cellular

phones due mainly to their insufficient performance, such as

.

566 (S. Yu).ufs.ac.kr (S. Yu).

C A R B O N 5 0 ( 2 0 1 2 ) 1 7 0 – 1 7 4 171

low brightness and efficiency [15,16]. Considerable effort has

been made to improve the properties of IPEL devices – using or-

ganic dye-phosphor composites [17] and adoption of a multi-

emitting-layer structure [18]. However, high brightness with a

simple structure operated at relatively low voltage and low fre-

quency is still required for EL devices as one of future flat panel

displays.

This paper reports the results of the local electric field ef-

fect, which might be enlarged by the existence of the triple-

junction [19,20], in IPEL devices by inserting a separate layer

of CNTs between the dielectric and phosphor layers. To pre-

vent an unintentional current by the CNT networks, the

length of the CNTs was shortened using a cryogenic crushing

method [21]. This study clarifies the brightness improvement

by the local electric field from the CNTs, and proposes a new

structure to construct highly efficient IPEL devices.

2. Experimental

2.1. Preparation of component layers

The top-emission type structure was used for the IPEL devices

(Fig. 1(a)) [22]. Fig. 1 (b) shows a cross-section scanning elec-

tron microscopy (SEM) image. The materials of the compo-

nent layers were prepared as follows. Single wall carbon

nanotubes (SWCNTs, Hanhwa Nanotec, ASP-100F) were

shortened using the cryogenic crushing method for 20 or

30 min [21]. An ethanolic solution of shortened SWCNTs

was dispersed further in dichlorobenzene and treated by ul-

tra-sonication for 2 h. The dielectric paste was prepared by

mixing the BaTiO3 powder (Samsung Fine Chemicals, SBT03)

and a fluoro-resin-based organic binder (ELK Corp., ELPR530)

using ethylene glycol mono-butyl ether acetate as the solvent.

The light-emitting paste was prepared using the same meth-

od of a dielectric paste, except that ZnS:Cu,Cl phosphor

(Sylvania, GG44, encapsulated by alumina) was used instead

of BaTiO3.

Fig. 1 – (a) Schematic diagram of the top-emission typed IPEL

structure with a layer of SWCNTs. (b) SEM image of the

cross-sectional view of an experimentally fabricated IPEL

device.

2.2. Formation of EL devices

Device fabrication was performed using step-by-step coatings

of the layers. First, a dielectric layer was deposited on an alu-

minum-coated glass substrate by screen-printing and dried at

130 �C for 30 min. A shortened SWCNT suspension was then

deposited on the dielectric layer by spin-coating at various

coating-speeds (300, 400, and 500 rpm) and dried at 100 �Cfor 30 min. The phosphor layer was then deposited on the

SWCNT layer by screen-printing and dried at 130 �C for

30 min. Finally, indium tin oxide (ITO) as a top electrode was

deposited on the emitting layer by dc sputtering using an

ITO target (DASOM RMS, xIn2O3–ySnO2).

2.3. Characteristic of EL devices

The characteristics of the devices were measured by applying

a sinusoidal AC voltage under an atmospheric environment.

All the AC readings in this paper refer to the peak-to-peak val-

ues. Combined equipments composed of a conventional func-

tion generator (Agilent, 33250A) and a power amplifier (Trek,

P0610B-K) were used to generate the AC voltages. An oscillo-

scope (Hewlett–Packard 400D) was connected to an IPEL de-

vice to monitor the actual value of the applied high voltage.

The brightness of the IPEL devices was measured using a

luminance colorimeter (Topcon, BM-7). The field emission

characteristics of the shortened SWCNTs were measured in

a vacuum chamber with a diode type configuration under

1 · 10�7 torr, where the CNTs were deposited on the ITO-

coated glass substrate as a cathode plate and the distance be-

tween the cathode and anode (a bare ITO substrate) was

maintained using 400 lm-thick glass spacers.

3. Results and discussion

The untreated SWCNTs (roughly 3 lm long) were shortened to

less than the sub-micrometer range after a cryo-crushing time

of 20 min [21]. The Raman spectral peak at 1325 cm�1 repre-

sents the disorder state of carbon atoms in the graphitic struc-

ture, known as the D-band. According to Fig. 2, the D-band

intensities of the SWCNTs increased with increasing crushing

Fig. 2 – Raman spectra of the untreated, 20 and 30 min-

crushed SWCNTs.

172 C A R B O N 5 0 ( 2 0 1 2 ) 1 7 0 – 1 7 4

time, where the lowest intensity was observed for the non-

treated SWCNTs. The higher D-band intensity of the shortened

SWCNTs originated from a relatively large amount of amor-

phous carbon through the cryo-crushing process [23]. In all

IPEL devices in this study, the crushing time was fixed to

20 min because short SWCNTs for a crushing time of 20 min

could induce effective field enhancement among various

crushing times. The amount of deposited SWCNTs was con-

trolled by the spin coating speed.

The measured luminance of each IPEL device with a layer of

SWCNTs coated at 300, 400, 500 rpm was 192, 182, and 174,

respectively, whereas the luminance for the no CNT was

160 cd/m2 (Fig. 3). This indicates that the luminescence of IPEL

was improved by the presence of SWCNTs, and the high con-

centration of CNT, i.e., obtained using a slow coating speed, re-

sulted in high EL performance. The 300 rpm sample was 20%

brighter than the sample without the SWCNTs [24]. The 20%

improvement in the brightness is still not sufficient to solve

the long-lasting brightness problem for IPEL devices, however,

the adoption of a nanomaterial might hint a new direction for

this problem, whereas the previous researches on EL rather fo-

cused on other issues, i.e., the crystallinity of the phosphor par-

ticle [16,25], the encapsulation of phosphor particle [26],

doubling the emitting layer [9], and incorporation of organic

dye [17]. Considering the recent intensive research activities

in various fields to utilizing nanotechnology [27,28], this ap-

proach, i.e., using a carbon based nanomaterials, at least,

might not lead to a wrong direction.

In general, an electric field in an AC IPEL device is the key

governing factor for the luminance because the intensity of

the luminance can be determined mainly by the intensity of

an electric field [14,29]. ZnS:Cu,Cl phosphor particles have

ZnS–CuxS heterojunctions, where ZnS (CuxS) is a n-type

(p-type) semiconductor. When an electric field of 106–107 V/

m is applied to the phosphor, the electric field induces tunnel-

ing of electrons and holes into ZnS lattice at opposite ends of

each CuxS, then electrons are trapped in the Cl donor site,

while holes are trapped by Cu recombination centers. At the

reversal field, the emitted electrons recombine with the

trapped hole to produce light [14,29]. In the present study,

the SWCNT layer on the dielectric layer, and at the same time

Fig. 3 – Luminance and efficiency as a function of the driving

voltage for the IPEL devices with no SWCNTs and with a

layer of shortened SWCNTs prepared for three coating

speeds (300, 400, and 500 rpm).

beneath the phosphor particle layer, could supply a high local

electric field to the phosphor particles, which might be ex-

plained by a triple-junction type enhancement between the

SWCNT, the dielectric material, and the phosphor [19]. This

field enhancement could not be simply explained by consid-

ering the high geometrical aspect ratio of SWCNTs as ex-

plained in Refs. [19,20].

The existence of the triple-junction between the SWCNT,

the dielectric material, and phosphor particles could be one

of main reasons for this improvement. The abrupt potential

change around SWCNTs, assisted by the work function differ-

ence of the surrounding dielectric and phosphor materials,

could provide a considerable high electric field [20]. The recom-

bination of holes trapped in the Cu recombination centers to-

gether with the increased electrons at the enhanced field at

the reverse voltage will produce more light [14]. Therefore,

the improved EL performance from the IPEL devices with a

layer of SWCNTs might be explained by the enlarged electric

field at the triple-junction.

To clarify the direct relationship of the IPEL performance to

the field enhancement effect asserted from SWCNTs, the field

emission characteristics of the shortened SWCNTs were mea-

sured for three coating speeds. Since the field emission of

SWCNTs on a dielectric layer is difficult due to the insulating

nature of the dielectric layer, the measurements were per-

formed directly on an ITO-coated glass substrate. Although

the magnitude of field enhancement of the shortened SWCNTs

on the dielectric layer might be different from that on the ITO

substrate, the tendency of the enhancing effect to field

enhancement could be matched between CNTs on the dielec-

tric layer and ITO layer. Therefore, the field emission of

SWCNTs on the ITO substrate for the three coating speeds

was measured and the corresponding Fowler–Nordheim (F–N)

plots [30], are shown in Fig. 4. The turn-on electric fields (de-

fined as the electric field at a current density of 10 lA/cm2) were

1.6, 1.8, and 2.1 V/lm, for coating speeds of 300, 400, and

500 rpm, respectively. The dimensionless field enhancement

factors (b) concerning the geometrical contribution of CNT

were 4750, 4500, and 3000, respectively, and the area factors

(a), related to the number of emitting sites were 3.2 · 10�8,

Fig. 4 – Field emission current density as a function of the

applied field for the shortened SWCNTs deposited at

various coating speeds on an ITO substrate. The inset

shows the corresponding Fowler–Nordheim plots.

Table 1 – List of the area factor (a) and field enhancement factor (b) of the shortened SWCNTs on the ITO substrate, and theluminance of the IPEL devices with a layer of shortened SWCNTs.

300 rpm 400 rpm 500 rpm No CNTs

a (10�8 cm2) 3.2 2.6 0.79 –b 4750 4500 3000 –Luminance (cd/m2) 192 182 174 160Luminance increment 20%" 14%" 8.4%" –

Fig. 5 – Luminance and the efficiency as a function of the

driving voltage of the IPEL devices with no SWCNTs and

with a layer of shortened and untreated SWCNTs. The spin

coating speed was set to 300 rpm for all devices. The

difference in the luminance for shortened and untreated

SWCNTs is too small, and it is difficult to distinguish the

two symbols.

C A R B O N 5 0 ( 2 0 1 2 ) 1 7 0 – 1 7 4 173

2.6 · 10�8 and 7.9 · 10�9 cm2, respectively for a 300, 400, and 500

spin speed. Here, the work function of SWCNTs was assumed

to be 5 eV [31]. Among these three samples, the one prepared

at 300 rpm exhibited the highest field emission characteristics,

which is due to the relatively high density of SWCNTs com-

pared to the other speeds. Table 1 summarizes the geometry

factor (b) and area factor (a) of the shortened SWCNTs on the

ITO substrate as well as the luminance of IPEL devices for three

coating speeds. The luminance improvement of IPEL devices

showed a similar tendency of field enhancement obtained

from the ITO substrates. High density of short SWCNTs by

low spin speed could improve the electric field in the IPEL de-

vice because of the triple-junction emission at the interface be-

tween the SWCNT, dielectric materials, and vacuum [19].

In general, the local electric field is increased in proportion

to the field enhancement factor, b, based on the equation,

E ¼ bV: A layer of SWCNTs with a high b value would enhance

the electric field with a help of the triple-junction, facilitating

electron tunneling and acceleration in the phosphor layer. In

addition, the higher a value, representing a larger number of

enhancing sites in field emission, would provide more recom-

bination sites in EL, producing brighter luminescence.

To investigate the length effect of the SWCNTs, three dif-

ferent IPEL devices, i.e., with no SWCNTs, with shortened

SWCNTs, and with untreated long SWCNTs, were prepared

at a fixed spin coating speed of 300 rpm. Their luminescence

characteristics were measured at driving voltages up to 150 V

and 400 Hz. According to Fig. 5, the luminance of the device

with a layer of SWCNTs, whether shortened or untreated,

was higher than that with no SWCNTs.

On the other hand, the EL efficiency of the device with the

untreated SWCNTs was similar to that without SWCNTs, and

lower than the one with the shortened SWCNTs. The effi-

ciency, g, was obtained from the equation, g ¼ p� LV�I (l m/W)

[14], where L is the luminance. The lower efficiency of the un-

treated SWCNT device was caused by the relatively high cur-

rent flow during the EL operation. This might be caused by the

interconnection of long SWCNTs producing a waste current

with no contribution to EL [32]. On the other hand, the device

with the shortened SWCNTs showed similar current flow

with respect to the one without SWCNTs, indicating much

fewer conducting paths in the shortened SWCNT layer.

Therefore, the IPEL device with shortened SWCNTs exhibits

higher efficiency and high luminescence.

4. Conclusions

A series of IPEL devices was fabricated with insertion of a

shortened SWCNT layer between the phosphor and dielectric

layers. The shortened SWCNTs were prepared using the cryo-

genic crushing method to effectively enhance the electric

field but prevent possible conducting paths in the SWCNT

layer. The EL performance improved with increasing amount

of SWCNTs, reflecting the local field enhancement effect by

the shortened CNTs. The evaluation of the EL efficiency for

the IPEL device with long or short SWCNTs highlighted the

importance of the shortened character of the SWCNTs, which

increase the EL efficiency through the high luminance with

low current flow.

Acknowledgements

This work was supported in part by the National Research

Foundation of Korea (NRF 2010-0011894). D. Jung was sup-

ported by grants NRF-2010-0029699 (Priority Research Centers

Program). J.-Y. Kim is a recipient of the Seoul Science Fellow-

ships of the Seoul Metropolitan Government.

R E F E R E N C E S

[1] Wang Y, Wu J, Wei F. A treatment method to give separatedmulti-walled carbon nanotubes with high purity, highcrystallization and a large aspect ratio. Carbon2003;41:2939–48.

174 C A R B O N 5 0 ( 2 0 1 2 ) 1 7 0 – 1 7 4

[2] Zhu L, Xu J, Xiu Y, Sun Y, Hess DW, Wong CP. Growth andelectrical characterization of high-aspect-ratio carbonnanotube arrays. Carbon 2006;44:253–8.

[3] Chhowalla M, Teo KBK, Ducati C, Rupesinghe NL,Amaratunga GAJ, Ferrari AC, et al. Growth process conditionsof vertically aligned carbon nanotubes using plasmaenhanced chemical vapor deposition. J Appl Phys2001;90:5308–17.

[4] Choi JH, Choi SH, Han JH, Yoo JB, Park CY, Jung T, et al.Enhanced electron emission from carbon nanotubes thoughdensity control using in situ plasma treatment of catalystmetal. J Appl Phys 2003;94:487–90.

[5] Kim YC, Yoo EH. Printed carbon nanotube field emitters forbacklight applications. Jpn J Appl Phys 2005;44(15):L454–6.

[6] Bae MJ, Park SH, Jeong TW, Lee JH, Han IT, Jin YW, et al. ApplPhys Lett 2009;95:071901-1–3.

[7] Destriau G. Scintillation of zinc sulphide with u-rays. J ChemPhys 1936;33:587–625.

[8] Kim JY, Park SH, Jeong T, Bae MJ, Song S, Lee J, et al. Paper as asubstrate for inorganic powder electroluminescence devices.IEEE Trans Electron Dev 2010;57(6):1470–4.

[9] Satoh T, Nakatsuta T, Tsuruya K, Tabata Y, Tamura T,Ichikawa Y, et al. Electrical properties of two-sidedluminescence powder-distributed inorganicelectroluminescence panels. J Mater Sci Mater Electron2007;18:S239–42.

[10] Miura N. New trial of inorganic EL devices. Proceeding, 14thinternational workshop on inorganic and organicelectroluminescence. Tivoli: Rome, Italy; 2008. p. 427–9.

[11] Warkentin M, Bridges F, Carter SA, Anderson M.Electroluminescence materials ZnS:Cu, Cl and ZnS:Cu, Mn, Clstudied by EXAFS spectroscopy. Phys Rev B 2007;75:075301-1–9.

[12] Shionoya S, Yen WM. Phosphor handbook. Boca Raton: TheCRS Press; 1999. p. 581–612.

[13] Wang W, Huang F, Xia Y, Wang A. Improvementelectroluminescence performance of ZnS:Cu, Cl phosphorsby ultrasonic treatment. J Lumin 2008;128:199–204.

[14] Ono YA. Electroluminescent displays. Singapore: WorldScientific; 1995. p. 7–25.

[15] Sawada M, Oobayashi S, Yamaguchi K, Takemura H,Nakamura M, Momose K, et al. Characteristics of lightemission lifetime of electroluminescent phosphorencapsulated by titanium–silicon–oxide film. Jpn J Appl Phys2002;41(6A):3885–9.

[16] Sharama G, Han SD, Kim JD, Khatkar SP, Rhee YW.Electroluminescent efficiency of alternating current thickfilm devices using ZnS:Cu,Cl phosphor. Mat Sci Eng B2006;131:271–6.

[17] Satoh T, Kobayashi M, Kawamura S, Uchida T. High-luminosity organic-dye-dispersed inorganicelectroluminescent panel. Electron Lett 2007;43(19):1047–8.

[18] Kim JY, Bae MJ, Park SH, Jeong T, Song S, Lee J, et al.Electroluminescence enhancement of the phosphordispersed in a polymer matrix using the tandem structure.Org Electron 2011;12(3):529–33.

[19] Alexandrou I, Kymakis E, Amaratunga GAJ. Polymer-nanotube composites: burying nanotubes improves theirfield emission properties. Appl Phys Lett 2002;80(8):1435–7.

[20] Chhowalla M, Ducati C, Rupesinghe NL, Teo KBK,Amaratunga GAJ. Field emission from short and stubbyvertically aligned carbon nanotubes. Appl Phys Lett2001;79(13):2079–81.

[21] Lee JH, Jeong TW, Heo JN, Park SH, Lee DH, Park JB, et al.Short carbon nanotubes produced by cryogenic crushing.Carbon 2006;44:2984–9.

[22] Kim JY, Jeong T, Park SH, Bae MJ, Song S, Lee J. Enhancedoptical and electrical properties of inorganicelectroluminescent devices using the top-emission structure.Jpn J Appl Phys 2011;50:06GF06-1–3.

[23] Liu Y, Liu L, Liu P, Sheng L, Fan S. Plasma etching carbonnanotubes arrays and the field emission properties. DiamondRelat Mater 2004;13:1609–13.

[24] To compare SWCNT layer incorporated IPEL devices withother group’s devices gives much clear interpretation on theimportance of a SWCNT layer. However, the fact that IPELdevices reported elsewhere [9,16,17,25,26] have differentstructures, materials, and operation conditions, yields todifficult in direct comparison with other devices unlikephotovoltaic solar cells. Instead, the reference device, i.e., theIPEL device without SWCNTs, was fabricated and comparedwith SWCNT IPEL devices, which could explain the meaningsof SWCNT’s contribution to EL.

[25] Nien YT, Chen IG. Raman scattering and electroluminescenceof ZnS:Cu,Cl phosphor powder. Appl Phys Lett2006;89:261906-1–3.

[26] Han SD, Singh I, Singh D, Lee YH, Sharma G, Han CH. Crystalgrowth of electroluminescent ZnS:Cu,Cl phosphor and itsTiO2 coating by sol–gel method for thick-film EL device. JLuminescence 2005;115:97–103.

[27] Kymakis E, Amaratunga GAJ. Single-wall carbon nanotube/conjugated polymer photovoltaic devices. Appl Phys Lett2002;80(1):112–4.

[28] Huynh WU, Dittmer JJ, Alivisatos AP. Hybrid nanorod-polymer solar cells. Science 2002;295:2425–7.

[29] Fischer AG. Electroluminescent lines in ZnS powder particles.J Electrochem Soc 1963;110(7):733–48.

[30] Spindt CA, Brodie I, Humphrey L, Westerberg EL. Physicalproperties of thin-film field emission cathodes withmolybdenum cones. J Appl Phys 1976;47:5248–63.

[31] Baik CW, Lee J, Choi JH, Jung I, Choi HR, Jin YW, et al. ApplPhys Lett 2010;96:023105-1–3.

[32] Dang ZM, Yao SH, Xu HP. Effect of tensile strain onmorphology and dielectric property in nanotube polymernanocomposite. Appl Phys Lett 2007;90(1):012907-1–3.