ARTICLE IN PRESS

Solar Energy Materials & Solar Cells 93 (2009) 1248–1255

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells

0927-02

doi:10.1

� Corr

E-m

journal homepage: www.elsevier.com/locate/solmat

Characteristics of flexible indium tin oxide electrode grown by continuousroll-to-roll sputtering process for flexible organic solar cells

Kwang-Hyuk Choi a,c, Jin-A Jeong b, Jae-Wook Kang c, Do-Guen Kim c, Jong Kuk Kim c, Seok-In Na d,Dong-Yu Kim d, Seok-Soon Kim e, Han-Ki Kim b,�

a School of Advanced Materials and Systems Engineering, Kumoh National Institute of Technology (KIT), Gumi 730-701, South Koreab Department of Display Materials Engineering, Kyung Hee University, 1 Seocheon-dong, Yongin-si, Gyeonggi-do 446-701, South Koreac Department of Material Processing, Korea Institute of Material Science (KIMS), 66 Sangnam-dong, Changwon-si, Gyeongnam 641-831, South Koread Heeger Center for Advanced Materials, Department of Materials Science and Engineering, Gwangju Institute of Science and Technology, 1 Oryoung-dong, Gwangju 500-712,

South Koreae School of Materials Science & Chemical Engineering, Kunsan Natioanl Unviersity, Kunsan, Chonbuk 753-701, South Korea

a r t i c l e i n f o

Article history:

Received 3 July 2008

Received in revised form

17 January 2009

Accepted 23 January 2009Available online 23 February 2009

Keywords:

Indium tin oxide

Roll-to-roll sputter

PET

Organic solar cells

Electrode

48/$ - see front matter & 2009 Elsevier B.V. A

016/j.solmat.2009.01.015

esponding author. Tel.: +82 31 201 2419; fax:

ail address: imdlhkkim@khu.ac.kr (H.-K. Kim)

a b s t r a c t

The preparation and characteristics of flexible indium tin oxide (ITO) electrodes grown on polyethylene

terephthalate (PET) substrates using a specially designed roll-to-roll sputtering system for use in

flexible organic solar cells are described. It was found that both electrical and optical properties of the

flexible ITO electrode were critically dependent on the Ar/O2 flow ratio in the continuous roll-to-roll

sputter process. In spite of the low substrate temperature (o50 1C), we can obtain the flexible ITO

electrode with a sheet resistance of 47.4O/square and an average optical transmittance of 83.46% in the

green region of 500–550 nm wavelength. Both X-ray diffraction and field emission scanning electron

microscopy analysis results showed that all flexible ITO electrodes grown on the PET substrate were

amorphous with a very smooth and featureless surface, regardless of the Ar/O2 flow ratio due to the low

substrate temperature, which is maintained by a cooling drum. In addition, the flexible ITO electrode

grown on the Ar ion-beam-treated PET substrates showed more stable mechanical properties than the

flexible ITO electrode grown on the wet-cleaned PET substrates, due to an increased adhesion between

the flexible ITO and the PET substrates. Furthermore, the flexible organic solar cell fabricated on the roll-

to-roll sputter-grown flexible ITO electrode at an optimized condition exhibited a power conversion

efficiency of 1.88%. This indicates that the roll-to-roll sputtering technique is a promising continuous

sputtering process in preparing flexible transparent electrodes for flexible solar cells or displays.

& 2009 Elsevier B.V. All rights reserved.

1. Introduction

There is currently considerable interest in flexible organic solarcells for use as a new generation of energy storage devices due totheir light weight, robust profile, ability to flex, curve, roll and foldfor portability, as well as their easy fabrication using roll-to-roll-based thin film technologies [1–7]. These flexible organic solarcells could offer the possibility for the low cost fabrication of largearea solar cells for harvesting energy from sunlight. To fabricatehigh performance and low cost flexible organic solar cells, it isnecessary to prepare transparent conducting oxide (TCO) with lowresistance, high transmittance, superior flexibility, and a smoothsurface on polymer substrates, such as polyethylene terephthalate(PET), polycarbonate (PC), polyimide polyethersulfone (PES), orpolyethylene naphthalate (PEN) for effective extraction of holecarriers from active organic layers [8–12]. In addition, the roll-to-

ll rights reserved.

+82 31204 8114.

.

roll-based continuous TCO deposition process should be em-ployed to deposit the flexible TCO electrode for low cost and themass production of flexible solar cells [13,14]. Until now,amorphous indium tin oxide (ITO) films grown by batch-typedirect current (DC) or radio-frequency (RF) sputtering have beenwidely used as the anode layer in flexible organic solar cells[2–4,15,16]. Although ITO electrode is widely used in OPVs as aTCO material, the high cost of ITO materials makes it difficult toenvisage very high volume OPV production. Furthermore, con-sidering roll-type polymer substrates, the batch-type DC or RFsputtering process is unsuitable due to difficulty of the continuousroll-to-roll process. For those reasons, the roll-to-roll sputteringtechnique has been the subject of considerable attention as acontinuous TCO deposition process in the mass production offlexible organic solar cells. Although the electrical and opticalproperties of an ITO electrode deposited on various polymersubstrates using the batch-type DC or RF sputtering have beenreported, the characteristics of a flexible ITO electrode grown bythe roll-to-roll sputtering process have not been investigated indetail [6,16–18].

ARTICLE IN PRESS

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–1255 1249

In this work, we investigated electrical, optical, structural,mechanical, and the surface properties of flexible ITO electrodesgrown on a PET substrate using a specially designed roll-to-rollsputtering system for flexible solar cells. It was found that theresistivity and transmittance of flexible ITO electrodes arecritically influenced by the Ar/O2 flow ratio during continuousroll-to-roll sputtering. At an optimized Ar/O2 flow ratio, DC power,working pressure, and rolling speed, we can obtain a flexible ITOelectrode with a resistivity of 9.5�10�4O cm, average transmit-tance of �83.46%, and figure of merit value of about 3.4�10�3

O�1, despite its preparation at room temperature. In addition, theeffective surface treatment of the PET substrate using an Ar ionbeam enables us to make mechanically stable ITO electrode withgood robustness, due to an improved adhesion between the ITOand the PET substrate. Furthermore, we show the integration of aflexible ITO electrode with flexible organic solar cells, whichdemonstrate the possibility of the ITO films grown by continuousroll-to-roll sputtering as a flexible anode layer in the flexibleorganic solar cells.

2. Experimental

The flexible ITO electrode with a thickness of 200 nm wasdeposited on a flexible PET substrate using a specially designedroll-to-roll sputtering system. Fig. 1 shows the schematics of theroll-to-roll sputtering system equipped with a rewind roller,unwind roller, cooling drum, and a cold cathode-type ion gunsystem for the deposition of high-quality flexible ITO electrodes.The 200 mm wide PET substrate with a thickness of 188mm waspassed repeatedly over the cooling drum by motion of unwind andrewind roller for the deposition of ITO electrodes during the roll-to-roll sputtering process. The rolling speed of the PET substratecan be exactly controlled by the motor speed for unwind andrewind roller. The ITO (10 wt% SnO2+90 wt% In2O3) target and iongun were placed at a distance of 100 mm from a PET substrate.Before sputtering of the ITO electrode on the PET substrate, thesurface of the PET substrate was pretreated by the irradiation of anAr ion beam at a DC pulsed power of 200 W to remove surfacecontaminations and improve adhesion between the ITO electrodeand PET substrate. Subsequently, the ITO film was sputtered onthe rolling PET substrate, which is mechanically attached oncooling drum to maintain the substrate temperature below 50 1C,using DC power as a function of the Ar/O2 flow ratio. The thickness

Fig. 1. Schematic diagram of a specially de

of the flexible ITO electrode was measured by a profilometer. Theelectrical properties of flexible ITO were measured by means of afour-point probe and Hall measurement at room temperature. Theoptical transmittance was measured in the wavelength range250–800 nm by a UV–visible spectrometer. The surface morphol-ogy of the flexible ITO electrode was analyzed using a field-emission scanning electron microscope (FESEM) as a function ofthe Ar/O2 ratio. To investigate the structural properties of flexiblethe ITO electrode, both X-ray diffraction (XRD) and high resolutionelectron microscope (HREM) examinations were performed. Thedependence of binding energy for the elements of a flexible ITOelectrode on the Ar/O2 flow ratio was examined by an XPS(PHI5200) carried out by an Al Ka source in an ultra-high-vacuumsystem with a base pressure of �10�10 Torr. In addition, theinterfacial reaction between the flexible ITO and PET substratewas examined using an Auger electron spectroscopy (AES) depthprofile with 10 keV energy and 0.0236mA current of electronbeam. Furthermore, the flexibility of the flexible ITO electrode wasanalyzed using a laboratory-made bending test system. Theflexible ITO/PET samples were clamped between two semicircleparallel plates. One plate was mounted to the shaft of a motor,while the other was fixed to a rigid support. The distance of thestretched mode was 80 mm and that of the bent position was30 mm. The bending radius was approximated to 8 mm and thebending frequency was 1 Hz. During the bending test, theresistance of the flexible ITO/PET samples was measured using amulti-meter. Finally, we prepared the flexible organic solar cellson the roll-to-roll sputter-grown ITO/PET sample, which wasprepared at optimized roll-to-roll sputtering conditions. Aftercleaning the ITO electrode with various organic solvents andoxygen plasma treatment, poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS, Baytron P VPAI 4083) wasspin-coated onto the flexible ITO electrode with a thickness of�20 nm, followed by drying at 80 1C for 30 min in air. For thefabrication of the photoactive layers composed of interconnectednetworks of electron donor and acceptor, blend of poly(3-hexylthiophere)(P3HT, Rieke Metals) and 1-(3-methoxycarbo-nyl)-propyl-1-phenyl-(6,6) C61(PCBM, Nano-C) in chlorobenzenewith a ratio 1:0.5 was spin-coated on top of the PEDOT:PSS layer.Then annealed at 80 1C for 10 min in N2 atmosphere to enhancethe degree of P3HT ordering [19]. Finally, a Ca (20 nm)/Al (100 nm)cathode layer was patterned on P3HT-PCBM active layer using ashadow metal mask. The photocurrent density–voltage (J–V)curves were measured with a Keithley 4200 source measurement

signed roll-to-roll sputtering system.

ARTICLE IN PRESS

300

400

[ohm

/sq] 8

10

hm-c

m]

DC power: 800 WWorking pressure: 3 mTorrRolling speed: 0.2 cm/sec

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–12551250

unit. Cell performance was measured under 100 mW/cm2 illumi-nation intensity from 1 kW Oriel solar simulator with an AM 1.5Gfilter in N2-filled glove box. For accurate measurement, lightintensity was calibrated by a radiant power meter and a referencesilicon solar cell certificated from national renewable energylaboratory (NREL).

0

100

200

Sheet resistance Resistivity

50:140:130:1Ar/O2 flow rate [sccm]

She

et re

sist

ance

20:10

2

4

6

Res

istiv

ity [1

0-3o

0

10

20

30

40 Mobility Carrier concentration

50:140:130:1Ar/O2 flow rate [sccm]

Mob

ility

[cm

2 /V

-s]

20:11x1020

2x1020

3x1020

4x1020

5x1020

Car

rior c

once

ntra

tion

[cm

-3]

Fig. 3. Variation of resistivity, sheet resistance, Hall mobility, and carrier

concentration as a function of Ar/O2 flow ratio from 20/1 to 50/1 sccm for the

flexible ITO film grown on a PET substrate using continuous roll-to-roll sputtering

at constant DC power of 800 W, working pressure of 3 mTorr and rolling speed of

0.2 cm/s: (a) sheet resistance and resistivity and (b) Hall mobility and carrier

concentration.

3. Results and discussion

Fig. 2 shows the thickness of a flexible ITO electrode grown ona PET substrate as a function of the DC power and rolling speed. Itwas shown that the thickness of flexible ITO electrode mono-tonically increased with increasing DC power. The DC powerdependence on the thickness of flexible ITO electrode could beexplained by following equation [20,21]:

Sr ¼ Sions þ Sneutrals

Theoretically, the total sputtering rate Sr in a roll-to-roll sputteringcan be determined by the sputtering rates due to ion (Sions) andneutral bombardment (Sneutrals). Therefore, the increased DCpower during the roll-to-roll sputtering of ITO electrodes couldresult in an increase of Sions, which critically affected thedeposition rate of the ITO electrode. However, the increase ofrolling speed lead to a decrease in thickness of the ITO electrode,because the exposure time of the PET substrate at the sputteringregion of ITO target linearly decreased with the increasingrolling speed. The sample grown at the optimized condition (DCpower of 800 W, working pressure of 3 mTorr, and rolling speed of0.2 cm/s) shows a thickness of 202 nm and a sputtering rate of105.3 nm/min.

Fig. 3 shows the electrical properties of flexible ITO electrodesgrown on a PET substrate at constant DC power of 800 W, workingpressure of 3 mTorr, and rolling speed of 0.2 cm/s as a function ofthe Ar/O2 flow ratio. Both the sheet resistance and resistivity ofthe flexible ITO electrode grown at the Ar/O2 flow ratio of 30/1 sccm are lower than those of flexible ITO electrodes grown at theAr/O2 flow ratio of 20/1 sccm as shown in Fig. 3(a). The minimumsheet resistance of 47.4O/square and resistivity of 9.5�10�4O cmwere obtained at the Ar/O2 flow ratio of 30/1 sccm. However, theincrease of Ar flow rate in the mixture of the Ar/O2 gas above30 sccm resulted in an abrupt increase of sheet resistance andresistivity. Fig. 3(b) shows a dependence of carrier mobility and aconcentration in the flexible ITO electrode on the Ar/O2 flow ratio.It was shown that the carrier concentration of the flexible ITO

200 400 600 800 10000

100

200

300

400

500

Thic

knes

s [n

m]

DC power [W]

Rolling speed1.0 cm/sec0.5 cm/sec0.3 cm/sec0.2 cm/sec0.1 cm/sec

Fig. 2. The thickness of the flexible ITO electrode grown on a PET substrate using

roll-to-roll sputtering system as a function of a DC power and rolling speed.

electrode increased, while the mobility of the flexible ITOelectrode decreased with the increasing Ar flow rate in themixture of Ar/O2 gas. Tahar et al. [15] reported that increasingoxygen partial pressure resulted in decrease in carrier concentra-tion and enhanced Hall mobility due to the dissipation of oxygenvacancies. Therefore, the low resistivity of flexible ITO films grownat the Ar/O2 flow ratio of 30/1 sccm can be explained by enhancedcarrier mobility resulting from the dissipation of oxygen vacancy.However, the flexible ITO films grown at the Ar/O2 flow ratio of20/1 sccm exhibited little increased resistivity due to decreasedoxygen vacancies at high oxygen partial pressure. Fig. 4 showsoptical transmittance spectra for the flexible ITO films grown onPET substrate as a function of the Ar/O2 flow ratio. All flexible ITOfilms show similar optical transparency regardless of the Ar/O2

flow ratio. It was believed that the transparency of flexible ITOelectrodes is not sensitive to the Ar/O2 flow ratio in the rangebetween 20/1 and 50/1 sccm. It is noteworthy that the transmit-tance of flexible ITO films in the green region (500–550 nm) ishigher than in the UV region (400–450 nm). The sharp absorptionedges in the transmittance spectra are caused by an extrinsicbandgap of ITO film in the range 3.8–4.0 eV [22].

Surface smoothness and morphology of the flexible ITOelectrodes grown on PET substrates as a function of the Ar/O2

flow ratio were analyzed by the FESEM examination. Fig. 5 showsthat the surface images of all flexible ITO electrodes are verysmooth and featureless without defects such as pinholes, cracks,

ARTICLE IN PRESS

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–1255 1251

and protrusion regardless of the Ar/O2 flow ratio due to the lowsubstrate temperature during continuous roll-to-roll sputtering.In general, the temperature of PET substrate depends on howmuch thermal energy is absorbed per unit area and how muchheat is transferred or lost to the cooling drum during the rolling.Therefore, the smooth surface of the flexible ITO electrodeindicates that the temperature of PET substrate is maintainedbelow 50 1C due to the effective heat transfer of a cooling drum.The smooth surface of the anode layer is very important forflexible solar cells, because anode spikes can cause breakdown orshorting of flexible solar cells.

300 400 500 600 700 8000

20

40

60

80

100

Tran

smitt

ance

[%]

Wavelength [nm]

20:1 30:1 40:1 50:1 PET

Fig. 4. Optical transmittance of the flexible ITO electrodes grown on a PET

substrate using roll-to-roll sputtering as a function of Ar/O2 flow ratio from 20/1 to

50/1 sccm.

Fig. 5. Surface FESEM images of the flexible ITO electrodes grown on a PET substrate us

1 sccm.

To obtain an optimized flexible ITO electrode with lowerresistance and higher transmittance, we carried out roll-to-rollsputtering of flexible ITO electrodes at the narrow Ar/O2 flow ratioregion between 30/0 and 30/3 sccm. Fig. 6 shows electricalproperties of the flexible ITO electrode film grown on the PETsubstrate at constant DC power of 800 W, working pressure of3 mTorr, and rolling speed of 0.2 cm/s as a function of the Ar/O2

flow ratio. It was shown that the increase of the Ar/O2 flow ratioabove 30/2 sccm led to a significant increase in sheet resistanceand resistivity. In particular, both sheet resistance and resistivityof flexible ITO electrode grown at the Ar/O2 flow ratio of 30/3 sccmcannot be measured using the Hall measurement and four-pointprobe due to the high-resistivity-like insulating film. It is believedthat the incorporation of lots of oxygen atoms into the ITOelectrode during roll-to-roll sputtering led to the change of theITO electrode from conductor to insulators. Fig. 6(b) shows thedependence of carrier concentration and Hall mobility on the Ar/O2 flow ratio from 30/0 to 30/3 sccm. As expected from Fig. 3(b),the increase of partial oxygen flow ratio from 0 to 2 sccm resultedin a decrease in carrier concentration and an increase in carriermobility due to the incorporation of oxygen atoms into oxygenvacancies. Fig. 7 shows optical transmittance spectra for theflexible ITO film grown at the narrow Ar/O2 flow ratio region of30/0–30/3 sccm. It was noteworthy that the introduction ofoxygen gas into the Ar ambient leads to significant improvementof optical transmittance. Compared to the transmittance offlexible ITO electrodes grown at pure Ar ambient (30/0 sccm),those of flexible ITO electrodes grown at oxygen flow ratios above1 sccm showed much higher optical transmittance due to theoptimization of stoichiometry in the flexible ITO electrodes.Although the electrical properties of the flexible ITO electrodegrown at the Ar/O2 flow ratio of 30/3 sccm cannot be measureddue to the high resistivity of the ITO film, it shows high opticaltransmittance of 88.99% at a wavelength of 550 nm. Thisdemonstrates that the resistivity of roll-to-roll sputter-grown

ing roll-to-roll sputtering system as a function of Ar/O2 flow ratio from 20/1 to 50/

ARTICLE IN PRESS

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–12551252

flexible ITO electrodes is more sensitive to the Ar/O2 flow ratiothan optical transmittance. The surface FESEM images (not shownhere) of all flexible ITO electrodes grown at the Ar/O2 flow ratio of30/0–30/3 sccm exhibit very smooth and featureless surface of ITO

15

20

25

30

35

40

30:0 30:1 30:2 30:3Ar/O2 flow rate [sccm]

Mob

ility

[cm

2 /V-

s]

0.0

5.0x1019

1.0x1020

1.5x1020

2.0x1020

2.5x1020

3.0x1020

3.5x1020

Car

rier c

once

ntra

tion

[cm

-3]

0

100

200

300

400

30:330:230:1Ar/O2 flow rate [sccm]

She

et re

sist

ance

[ohm

/sq.

]

30:00

2

4

6

8

10

DC power 800 WWorking pressure 3 mTorrRolling speed 0.2 cm/sec

Res

istiv

ity [1

0-3 o

hm-c

m]

Fig. 6. Variation of resistivity, sheet resistance, Hall mobility, and carrier

concentration as a function of Ar/O2 flow ratio from 30/0 to 30/3 sccm for the

flexible ITO film grown at constant DC power of 800 W, working pressure of

3 mTorr and rolling speed of 0.2 cm/s: (a) sheet resistance and resistivity and (b)

hall mobility and carrier concentration.

300 400 500 600 700 8000

20

40

60

80

100

30:0 30:1 30:2 30:3 PET

Wavelength [nm]

Tran

smitt

ance

[%]

Fig. 7. Optical transmittance of the flexible ITO electrodes grown on a PET

substrate using roll-to-roll sputtering system at narrowAr/O2 flow ratio range from

30/0 to 30/3 sccm.

electrode regardless of the Ar/O2 flow ratio similar to the surfaceimages shown in Fig. 5, due to effective cooling of PET substrate.

From the electrical and optical properties of the flexible ITOelectrodes grown at an optimized condition, we can calculate thefigure of merit (fTC) suggested by Haacke [23]

fTC ¼T10

Rsh

where T is the transmittance and Rsh is the sheet resistance of thetransparent conducting oxide. A maximum fTC value of3.46�10�3O�1 could be obtained from the flexible ITO electrodegrown at DC power of 800 W, with a rolling speed of 0.2 scm/s,working pressure of 3 mTorr, and the Ar/O2 flow ratio of 30/1 sccm, which is an optimized condition in our roll-to-roll sputtersystem. Bender et al. [24] showed that a maximum fTC value of anITO–Ag–ITO layer with 10–12 nm thick metal layer on a glasssubstrate is 24.7�10�3O�1. The comparable fTC value of theflexible ITO electrode indicates the possibility of the roll-to-rollsputter-grown flexible ITO electrode as an electrode for flexibleorganic solar cells or organic light-emitting diodes.

Fig. 8 shows an XPS wide scan data obtained from the surfaceof the flexible ITO electrodes grown at an optimized condition.The XPS wide scan data clearly shows O 1s, In 3d, and Sn 3d peaks,which are indicative of bare ITO electrodes. Compared O 1s and In3d peaks, the Sn 3d peak shows weaker peak intensity due to lowa doping concentration of SnO2 (10 wt%) in In2O3. The detailedbinding energy of flexible ITO electrodes grown on PET substrateas a function of the Ar/O2 flow ratio was analyzed by a core-levelspectra of O 1s, In 3d, and Sn 3d, as shown in Fig. 9. It was foundthat there is no change in the binding energy of both In 3d and Sn3d peaks with increasing oxygen flow rate. However, the intensityof the two binding energies, 530.85 and 531.35 eV, in O 1s core-level spectra, which is caused by the two types of O2� ions in theflexible ITO electrode, is changed with increasing oxygen flowratio as shown in Fig. 9(c). The lower binding energy peak (OII) isfrom the O2� ions, which have neighboring indium atoms withtheir full complement, and the higher binding energy peak (OI)corresponds to oxygen-deficient regions [25]. Because the in-crease of oxygen flow ratio leads to the incorporation of theoxygen atom into the ITO electrode, the flexible ITO electrodegrown at the Ar/O2 flow ratio of 30/3 shows lowest OII peakintensity, which is related to oxygen vacancies [26].

1000 800 600 400 200 0

DC power: 800 WWorking pressure: 3 mTorrRolling speed: 0.2 cm/secAr:O2 = 30:1 sccm

In3p O

1sS

n3d

In3d

C1s

Inte

nsity

[arb

. uni

t]

Binding energy [eV]

Fig. 8. XPS wide scan data obtained from the flexible ITO electrode grown at

optimized roll-to-roll sputtering condition (DC power of 800 W, working pressure

of 3 mTorr, rolling speed of 0.2 cm/s, and Ar/O2 flow ratio of 30/1 sccm).

ARTICLE IN PRESS

458

In3d3/2

In3d5/2

Inte

nsity

[arb

. uni

ts]

Binding energy [eV]

In3d

30:3 sccm

30:2 sccm

30:1 sccm

498

Sn3d3/2

Inte

nsity

[arb

. uni

ts]

Binding energy [eV]

Sn3d5/2

Sn3d

30:3 sccm

30:2 sccm

30:1 sccm

540

531.35 eV

Inte

nsity

[arb

.uni

ts]

Binding energy [eV]

O1s1s

30:3 sccm

30:2 sccm

30:1 sccm

531.1 eV

530.85 eV

456 454 452 450 448 446 444 442 440

496 494 492 490 488 486 484 482 480

538 536 534 532 530 528 526 524

Fig. 9. XPS spectra of (a) In 3d, (b) Sn 3d, and (c) O 1s obtained from the flexible

ITO electrode grown on a PET substrate as a function of Ar/O2 flow ratio from 30/1

to 30/3 sccm.

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–1255 1253

Fig. 10 shows the XRD plots obtained from the flexible ITOelectrodes as a function of Ar/O2 flow ratio. All XRD plots of theflexible ITO electrodes show only the intense PET substrate peakat the region �25.841. Due to the low substrate temperatureduring the roll-to-roll sputtering process, all flexible ITO electro-des show amorphous structure regardless of the Ar/O2 flow ratio.This indicates that the substrate temperature was effectivelymaintained at low temperature, below 50 1C, by a cooling drum asexpected from FESEM results in Fig. 5. To investigate themicrostructure of the flexible ITO electrode grown at an optimizedcondition in detail, the HREM examination was employed. Fig. 11shows a cross-sectional HREM image obtained from the ITOelectrode grown on a Si substrate. Due to difficulty in the HREMsample preparation of flexible ITO electrodes grown on a PETsubstrate, the ITO electrode was intentionally grown on a Sisubstrate at identical conditions with an optimized flexible ITOgrowth condition. The uniform contrast of the ITO film in Fig. 11indicates that the structure of the ITO film on a Si substrate isamorphous, as expected from the XRD results. However, somenanocrystallines, as indicated by arrows are embedded in theamorphous ITO matrix, due to the low amorphous/polycrystallinetransformation temperature. Even if an ITO electrode is preparedat room temperature, nanocrystallines could be easily formed atthe amorphous ITO matrix at a low homogeneous temperatures(T/Tmo0.19–150 1C) [26,27].

20 30 40 50 60

Ar:O2 = 30:0 sccm

Ar:O2 = 50:1 sccm

Ar:O2 = 40:1 sccm

Ar:O2 = 30:1 sccm

Ar:O2 = 20:1 sccm

PET substrate

Ar:O2 = 30:3 sccm

Ar:O2 = 30:2 sccm

Inte

nsity

[arb

. uni

ts]

25.84

2θ

Fig. 10. XRD plots obtained from the flexible ITO electrodes grown on PET

substrates as a function of Ar/O2 flow ratio at constant DC power of 800 W,

working pressure of 3 mTorr, and rolling speed of 0.2 cm/s.

ARTICLE IN PRESS

0 5 10 15 20 25 30 35 40

0

20

40

60

80

100

Sn

O

In

C

PET substrate

Ato

mic

con

cent

ratio

n [%

]

Thickness [nm]

ITO

Fig. 12. AES depth profiles of a roll-to-roll sputtering grown flexible ITO/PET

substrate grown at optimized condition.

0 200 400 600 800 1000

1

10

100

ΔR/R

0

Bending cycle

Ar+ Ion treated 200nm ITO/PET untreated 200nm ITO/PET

Fig. 13. Normalized resistance change after repeated bending as a function of the

number of cycles for flexible ITO electrode grown on untreated and Ar+ ion-treated

PET substrate before ITO sputtering.

Fig. 11. Cross-sectional HREM image obtained from the flexible ITO electrode

grown on a Si substrate at optimized roll-to-roll sputtering condition.

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–12551254

Fig. 12 shows the AES depth profile obtained from the flexibleITO electrode grown on a PET substrate at an optimized condition.It was shown that the flexible ITO layer is well defined on a PETsubstrate, indicating the absence of significant interfacial reac-tions between the ITO and the PET substrates due to the lowsubstrate temperature. In addition, the uniform atomic concen-trations of In, O, and Sn elements throughout the ITO electrodeshows constant deposition of ITO films during continuous roll-to-roll sputtering.

To investigate the flexibility of the optimized flexible ITOelectrode prepared on a PET substrate, a laboratory-made bendingtest system was employed. For comparison, flexible ITO electrodeswere grown on both wet-treated and the Ar ion-treated PETsubstrates at the same optimized growth condition. The change inresistance was expressed as DR ¼ R�R0, where R0 is the initialresistance and R is the measured resistance after bending. Fig. 13shows changes in the resistance of the flexible ITO electrodegrown on wet-treated and the Ar ion-treated PET substrate at a DCpulsed power of 200 W. It was noteworthy that the DR/R0 value ofthe flexible ITO electrode grown on wet-treated PET substratesincreased remarkably at initial bending cycles, due to thegeneration and propagation of cracks. However, the flexible ITOelectrode grown on the Ar ion-treated PET substrates showed afairly constant DR/R0 value throughout the bending test. Com-pared to the flexible ITO electrode grown on the untreated PET

substrate, the flexible ITO electrode grown on the Ar ion-treatedPET substrate showed much more stable resistance. The robust-ness of the flexible ITO electrode grown on the Ar ion-treated PETsubstrate is attributed to the improvement of adhesion betweenflexible ITO films and the PET substrate. Ratchev et al. [28]reported that the ion-beam treatment of PC substrates causedsubstantial improvement in adhesion strength between Al filmand PC substrates. In addition, Zaporojtchenko et al. [29]suggested that the improvement in Cu film–PC substrate adhe-sion, by low-energy ion irradiation, is attributed to the creation ofa large density of new adsorption sites. Therefore, the Ar ion beamtreatment of the PET substrate during roll-to-roll sputteringprocess could result in the improvement of the adhesion betweenthe ITO electrode and PET substrate and creation of newadsorption sites.

To investigate the possibility of roll-to-roll sputter-grownflexible ITO film, as a flexible anode in flexible solar cells, wefabricated flexible organic solar cells on the roll-to-roll grown ITOelectrode. Fig. 14 shows current density–voltage (J–V) curves offlexible organic solar cells fabricated on a roll-to-roll sputter-grown flexible ITO electrode, at optimized conditions. Forfabrication of the flexible organic solar cells, we cut the ITO/PETsample size of 1.5�1.5 cm2 from large-area ITO/PET sample(200 mm width�2000 mm length). Sample 1 and 2 are flexiblesolar cells fabricated from identical flexible ITO electrodes.Samples 1 shows VOC ¼ 0.54 V, ISC ¼ 6.31 mA cm�2, FF ¼ 55.2%,and a power conversion efficiency of ZAM1.5 ¼ 1.88%. In addition,similar performance of sample 2 as sample 1 (VOC ¼ 0.54 V,ISC ¼ 6.18 mA cm�2, FF ¼ 55.3%, and ZAM1.5 ¼ 1.84%) indicates thatour large area ITO/PET electrodes prepared by roll-to-roll sputter-ing system can be applied to the mass production of solar cellswith reproducible efficiency values. Although the samples showrelatively low power-conversion efficiency compared to theprevious reports (1.5–3%), the performance might be improvedafter further optimization of fabrication process including theannealing process, thickness of active layers, and surface treat-ment of flexible ITO electrodes [2,30,31].

4. Conclusion

In summary, this study has demonstrated the applicability of aroll-to-roll sputtering technique as an alternative to conventionalRF or DC sputtering for continuous sputtering of flexible ITO

ARTICLE IN PRESS

0.0 0.1 0.2 0.3 0.4 0.5 0.60

1

2

3

4

5

6

7 Sample1 : 1.88% (Jsc:6.31, Voc: 0.54, FF:55.2%) Sample2 : 1.84% (Jsc:6.18, Voc: 0.54, FF:55.3%)

Cur

rent

den

sity

[mA

/cm

2 ]

Voltage [V]

Fig. 14. Current density–voltage characteristics (air mass 1.5 G condition with

incident light power intensity of 100 mW cm�2) of flexible organic solar cell

fabricated on the roll-to-roll sputter-grown flexible ITO electrode prepared at

optimized condition.

K.-H. Choi et al. / Solar Energy Materials & Solar Cells 93 (2009) 1248–1255 1255

electrode on a PET substrate. It was found that both the electricaland optical properties of the flexible ITO electrode were criticallydependent on the Ar/O2 flow ratio during continuous roll-to-rollsputtering. In addition, all flexible ITO electrodes show amor-phous structure and a very smooth surface area regardless the ofAr/O2 flow ratio, due to the low substrate temperature which ismaintained by the cooling drum. Even though the flexible ITOelectrode was prepared at room temperature, we can obtain theflexible ITO electrode with a sheet resistance of 47.4O/square andaverage optical transmittance of 83.46% in the green regionbetween 500 and 550 nm wavelengths. In addition, it was foundthat the Ar ion treatment of the PET substrate could improve theflexibility of the ITO electrode due to effective removal of surfacecontamination and increase in adhesion of the ITO film with thePET substrate. Furthermore, the J–V characteristics of the flexibleorganic solar cell prepared on a flexible ITO electrode grown atoptimized conditions show typical current density–voltage char-acteristics with a conversion power efficiency of 1.84–1.88%. Thisindicates that the roll-to-roll sputtering technique is a promisingcontinuous sputtering process for continuous production of lowcost flexible solar cells.

Acknowledgement

This work was supported by the Korea Foundation forInternational Cooperation of Science & Technology (KICOS)through a grant provided by the Korean Ministry of Education,Science & Technology (MEST) in K20701010289-07B0100-07511.

References

[1] N.S. Sariciftci, L. Smilowitz, A.J. Heeger, F. Wudl, Photoinduced electrontransfer from a conducting polymer to buckminsterfullerene, Science 258(1992) 1474–1476.

[2] S.E. Shaheen, S.J. Brabec, N.S. Sariciftci, F. Padinger, T. Fromherz, J.C.Hummelen, 2.5% efficient organic plastic solar cells, Appl. Phys. Lett. 78(2001) 841–843.

[3] H. Hoppe, N.S. sariciftci, Organic solar cells: an overview, J. Mater. Res. 19(2004) 1924–1945.

[4] C.J. Brabec, N.S. Sariciftci, J.C. Hummelen, Plastic solar cells, Adv. Funct. Mater.11 (2001) 15–26.

[5] H. Spanggaard, F.C. Krebs, A brief history of the development of organic andpolymeric photovoltaics, Sol. Energy Mater. Sol. Cells 83 (2004) 125–146.

[6] E. Bundgaard, F.C. Krebs, Low band gap polymers for organic photovoltaics,Sol. Energy Mater. Sol. Cells 91 (2007) 954–985.

[7] M. Jorgensen, K. Norrman, F.C. Krebs, Stability/degradation of polymer solarcells, Sol. Energy Mater. Sol. Cells 92 (2008) 686–714.

[8] J.-W. Kang, W.-I. Jeong, J.-J. Kim, H.-K. Kim, D.-G. Kim, G.-H. Lee, High-performance flexible organic light-emitting diodes using amorphous indiumzinc oxide anode, Electrochem. Solid-State Lett. 10 (2007) J75–J78.

[9] H. Han, D. Adams, J.W. Mayer, T.L. Alford, Characterization of the physical andelectrical properties of indium tin oxide on polyethylene napthalate, J. Appl.Phys. 98 (2005) 083705-1–083705-8.

[10] C. Marcel, N. Naghavi, G. Couturier, J. Salardenne, J.M. Tarascon, Scatteringmechanisms and electronic behavior in transparent conducting ZnxIn2Ox�3

indium–zinc oxide thin films, J. Appl. Phys. 91 (2002) 4291.[11] J. Lewis, S. Grego, B. Chalamala, E. Vick, D. Temple, Highly flexible transparent

electrodes for organic light-emitting diode-based displays, Appl. Phys. Lett. 85(2004) 3450–3452.

[12] D.R. Cairns, R.P. Witte II, D.K. Sparacin, S.M. Sachsman, D.C. Paine, G.P.Crawford, Strain-dependent electrical resistance of tin-doped indium oxideon polymer substrates, Appl. Phys. Lett. 76 (2000) 1425–1427.

[13] G.P. Crawford, Flexible Flat Panel Displays, first ed., Wiley, England, 2005,pp. 409–445 (Chapter 21).

[14] M. Izu, T. Ellison, Roll-to-roll manufacturing of amorphous silicon alloy solarcells with in situ cell performance diagnostics, Sol. Energy Mater. Sol. Cells 78(2003) 613–626.

[15] R.B.H. Tahar, T. Ban, Y. Ohya, Y. Takahashi, Tin doped indium oxide thin films:electrical properties, J. Appl. Phys. 83 (1998) 2631–2645.

[16] M. Al-lbrahim, H.K. Roth, S. Sensfuss, Efficient large-area polymer solar cellson flexible substrates, Appl. Phys. Lett. 85 (2004) 1481–1483.

[17] P.F. Carcia, R.S. Mclean, M.H. Reilly, Z.G. Li, L.J. Pillione, R.f. Messier, Low-stressindium-tin-oxide thin films rf magnetron sputtered on polyester substrates,Appl. Phys. Lett. 81 (2002) 1800–1802.

[18] K. Noda, H. Sato, H. Itaya, M. Yamada, Characterization of Sn-doped In2O3 filmon roll-to-roll flexible plastic substrate prepared by DC magnetron sputtering,Jpn. J. Appl. Phys. 42 (2003) 217–222.

[19] Y. Kim, S.A. Choulis, J. Nelson, D.D.C. Bradley, S. Cook, J.R. Durrant, Deviceannealing effect in organic solar cells with blends of regioregular poly(3-hexylthiophene) and soluble fullerene, Appl. Phys. Lett. 86 (2005) 063502-1–063502-3.

[20] J.-H. Hsieh, C. Li, Calculation of sputtering rate during a plasma-assistedprocess, Jpn. J. Appl. Phys. 42 (2003) 5295–5298.

[21] J.-H. Bae, J.-M. Moon, S.W. Jeong, J.-J. Kim, J.-W. Kang, D.-G. Kim, J.-W. Park,H.-K. Kim, Transparent conducting indium zinc tin oxide anode for highlyefficient phosphorescent organic light emitting diodes, J. Electrochem. Soc.155 (2008) J1–J6.

[22] C.G. Choi, K. No, W.-J. Lee, H.-G. Kim, S.O. Jung, W.J. Lee, W.S. Kim, S.J. Kim, C.Yoon, Effect of oxygen partial pressure on the microstructure and electricalproperties of indium tin oxide film prepared by d.c. magnetron sputtering,Thin Solid Films 258 (1995) 274–278.

[23] C. Haacke, New figure of merit for transparent conductors, J. Appl. Phys. 47(1976) 4086–4089.

[24] M. Bender, W. Seelig, C. Daube, H. Frankenberger, B. Ocker, J. Stollenwerk,Dependence of film composition and thicknesses on optical and electricalproperties of ITO-metal-ITO multilayers, Thin Solid Films 326 (1998) 67–71.

[25] J.C.C. Fan, J.B. Goodenough, X-ray photoemission spectroscopy studies ofSn-doped indium-oxide films, J. Appl. Phys. 48 (1977) 3524–3531.

[26] J. -H. Bae, J. -M. Moon, J. -W. Kang, H. -D. Park, J. -J. Kim, W.J. Cho, H. -K. Kim,Transparent, low resistance, and flexible amorphous ZnO-doped In2O3 anodegrown on a PES substrate, J. Electrochem. Soc. 154 (2007) J81–J85.

[27] Y. Shigesato, S. Takaki, T. Haranoh, Electrical and structural properties of lowresistivity tin-doped indium oxide films, J. Appl. Phys. 71 (1992) 3356–3364.

[28] B.A. Ratchev, G.S. Was, J.H. Bookes, Ion beam modification of metal-polymerinterfaces for improved adhesion, Nucl. Instrum. Methods Phys. Res. B 106(1995) 68–73.

[29] V. Zaporojtchenko, J. Zekonyte, F. Faupel, Effects of ion beam treatment onatomic and macroscopic adhesion of copper to different polymer materials,Nucl. Instrum. Methods Phys. Res. B 265 (2007) 139–145.

[30] S.-S. Kim, S.-I. Na, J. Jo, G. Tae, D.-Y. Kim, Efficient polymer solar cellsfabricated by simple brush painting, Adv. Mater. 19 (2007) 4410–4415.

[31] M. Al-Ibrahim, H.K. Roth, U. Zhokhavets, G. Gobsch, S. Sensfuss, Flexible largearea polymer solar cells based on poly(3-hexylthiophene)/fullerene, Sol.Energy Mater. Sol. Cells 85 (2005) 13–20.