Nano-sized Ag-inserted amorphous ZnSnO3 multilayer ...

Viewer
Transcript

Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat

Nano-sized Ag-inserted amorphous ZnSnO3 multilayer electrodes for cost-efﬁcient inverted organic solar cells Yoon-Young Choi a, Kwang-Hyuk Choi a, Hosun Lee b, Hosuk Lee b, Jae-Wook Kang c, Han-Ki Kim a,n a

Department of Advanced Materials Engineering for Information and Electronics, Kyung Hee University, 1 Seocheon-dong, Yongin-si, Gyeonggi-do 446-701, South Korea Department of Applied Physics, Kyung Hee University, 1 Seocheon, Yongin, Gyeonggi-do 446-701, South Korea c Department of Material Processing, Korea Institute of Material Science (KIMS), 66 Sangnam-dong, Changwon-si, Gyeongnam 641-831, South Korea b

a r t i c l e i n f o

abstract

Article history: Received 13 November 2010 Received in revised form 4 January 2011 Accepted 11 January 2011 Available online 18 February 2011

A sheet resistance- and optical transmittance-tunable amorphous ZnSnO3 (ZTO) multilayer electrode created through the insertion of a nano-scale Ag layer is demonstrated as an indium-free transparent conducting electrode for cost-efﬁcient inverted organic solar cells (IOSCs). Due to the antireﬂection effect, the ZTO/Ag/ZTO/glass exhibited a high transmittance of 86.29% in the absorption wavelength region of the organic active layer and a low resistivity of 3.24 10 5 O cm, even though the ZTO/Ag/ZTO electrode was prepared at room temperature. The metallic conductivity of the electrode indicates that its electrical conductivity is dominated by the nano-scale Ag metal layer. In addition, optimization and control of the thickness of the nano-scale Ag layer are important in obtaining highly transparent ZTO/Ag/ZTO electrodes, because antireﬂection is strongly inﬂuenced by Ag thickness. Moreover, IOSCs fabricated on optimized ZTO/Ag/ZTO electrodes with Ag thicknesses of 12 nm showed power conversion efﬁciencies (2.55%) comparable to that of an IOSC prepared on a crystalline ITO electrode (2.45%), due to the low sheet resistance and high optical transmittance in the range of 400–600 nm. The performances of ZTO/Ag/ZTO multilayer electrodes indicate that ZTO/Ag/ZTO multilayers are promising as indium-free, transparent electrode substitutes for conventional ITO electrodes in cost-efﬁcient IOSCs. & 2011 Elsevier B.V. All rights reserved.

Keywords: Amorphous ZnSnO3 Ag Multilayer electrode Inverted organic solar cells

1. Introduction In spite of the potential of organic solar cells (OSCs) as low-cost and eco-friendly energy-harvesting devices [1–5], the interface instability between poly(3,4-ethylenedioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) and ITO in conventional OSCs has been known as critical drawbacks. To address these problems, the architecture of an inverted OSCs (IOSCs), which uses an Ag top anode and an ITO cathode with an oxide (ZnO, TiOx, WO3, or MoO3) buffer layer, has been suggested as a solution for interface instability in conventional OSCs [6–9]. To fabricate cost-efﬁcient IOSCs, it is important to develop a low-cost transparent electrode with low resistivity and high transparency because the key parameters of IOSCs, such as ﬁll factor, power conversion efﬁciency, and short-circuit current density, are critically dependent on the characteristics of the TCO electrodes [3,10,11]. Although crystalline Sn-doped In2O3 (ITO) has been widely used as a transparent cathode in IOSCs, the ITO electrode is not a desirable electrode material for use in cost-efﬁcient IOSCs, due to the scarcity and high cost of indium. For this reason, various

n

Corresponding author. Fax: + 82 31 204 8114. E-mail address: [email protected] (H.-K. Kim).

0927-0248/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.solmat.2011.01.013

indium-free transparent electrodes such as PEDOT:PSS, carbon nanotube (CNT) electrodes, graphene electrodes, and several oxide electrodes (Ga–ZnO, Al–ZnO, Nb–TiO2) have been extensively investigated as replacements for the ITO electrode [12–19]. In addition, hybrid electrodes with oxide and metal layers such as ITO/Ag/ITO, IZTO/Ag/IZTO, or IZO/Ag/IZO have been reported as promising cost-efﬁcient electrodes for use in OSCs because of their low resistances and room temperature processes [20–22]. However, these materials still contain the high-cost element indium and therefore suffer from all of the critical problems of indium-based TCO materials, even though they show much lower sheet resistances and high transparencies. Therefore, indium-free multilayer electrodes with low sheet resistance and high transparency comparable to those of crystalline ITO electrodes are required to realize cost-efﬁcient IOSCs. In this study, we report on the sheet resistance- and optical transparency-tunable amorphous ZnSnO3 (ZTO)/Ag/ZTO multilayer electrodes for use in cost-efﬁcient IOSCs. The control of the thickness and morphology of the nano-size Ag layer is critical to enhancing the transparency and conductivity of the ZTO/Ag/ ZTO multilayer. The inﬂuences of the Ag thickness and morphology on the electrical, structural, and optical properties of indiumfree ZTO/Ag/ZTO multilayer electrodes were investigated in detail. In particular, the antireﬂection effect was investigated using

1616

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

ellipsometry analysis to explain the high transparency of the electrodes. The comparable performance of the IOSC fabricated on the optimized ZTO/Ag/ZTO electrodes to that of the IOSC on crystalline ITO electrode reveals that the ZTO/Ag/ZTO electrode is a promising indium-free transparent electrode substitute for the crystalline ITO electrode in cost-efﬁcient IOSCs or ﬂexible IOSCs due to its low sheet resistance, high transparency, and low process temperature.

2. Experimental details The ZTO/Ag/ZTO multilayer electrodes were prepared using a tilted dual-target RF and DC magnetron sputtering system at room temperature under optimized growth conditions for the amorphous ZTO electrode with varying Ag thicknesses. The substrate temperature during RF and DC sputtering processes was maintained at room temperature ( 30 1C) by a substrate cooling system. Using tilted cathode guns, the bottom ZTO, Ag, and top ZTO layers were continuously deposited at room temperature without breaking the vacuum. A ZTO target (35 wt% ZnO–65 wt% SnO2) was employed to deposit amorphous bottom and top ZTO layers. A 35-nm-thick bottom ZTO layer was sputtered at optimized conditions at a constant RF power of 100 W, an Ar ﬂow rate of 20 sccm, and a working pressure of 2 mTorr onto a 25 25 mm2 glass substrate at room temperature for 556 s. After sputtering the bottom ZTO layer, an Ag layer was continuously sputtered with varying Ag thicknesses using an Ag target at a constant DC power of 100 W for 9–42 s depending on its thickness. Finally, the top ZTO layer was sputtered onto the Ag layer under sputtering conditions identical to those used for the bottom ZTO layer. Note that all sputtering processes were performed at room temperature without heating substrate or post-annealing. The electrical properties of the ZTO/Ag/ZTO layer were characterized using Hall measurements (HL5500PC, Accent Optical Technology) as a function of Ag thickness. Using a UV/visible spectrometer (UV 540, Unicam), the optical transmittances of the ZTO/Ag/ZTO and Ag/ZTO multilayer electrodes were measured at wavelengths between 200–1200 nm as a function of the Ag thickness. The ellipsometric angles (C, D) of the thin ﬁlms were measured using spectroscopic ellipsometry (VASE model, J.A. Woollam Inc.) at room temperature and at various angles of incidence (651, 701, and 751). The surface morphology of the Ag layer grown on the bottom ZTO ﬁlm was examined using a ﬁeld emission scanning electron microscope (FESEM JSM-6500F) with an operating voltage of 15 keV as a function of Ag thickness. The structure of the ZTO/Ag/ZTO multilayer electrode was investigated using X-ray diffraction (XRD, D/Max2500, Rigaku). The microstructures and interface structures of the optimized ZTO/Ag/ZTO multilayer electrodes were examined using high resolution electron microscopy (HREM). Selected area electron diffraction (SAED) patterns and bright ﬁeld (BF) TEM images were obtained from a cross-sectional HREM specimen prepared via

focus ion beam (FIB) milling. In addition, the interfacial properties of the optimized ZTO/Ag (12 nm)/ZTO electrodes were analyzed using X-ray photoelectron spectroscopy (XPS) depth proﬁling. After the electrodes were cleaned, zinc acetate [Zn(ac)2] solution (157 g/l) in 96% 2-methoxyethanol and 4% ethanolamine were spin-coated at 2000 rpm onto the transparent electrode. Subsequent annealing at 300 1C for 10 min in air converted the [Zn(ac)2] to a ZnO layer with a thickness of 40 nm. The photoactive layer was then deposited by spin-coating a 1,2-dichlorobenzene solution containing 20 mg/ml P3HT and 20 mg/ml PCBM to a thickness of 250 nm (at a spin speed of 600 rpm). A solvent-annealing treatment was then performed by keeping the active ﬁlms for 2 h immediately after spin-coating, followed by an additional thermal annealing at 150 1C for 20 min on a hot plate. A PEDOT-PSS (Baytron P):IPA (PEDOT-PSS:IPA¼ 1:2) buffer layer was deposited onto a photoactive layer using a spin-coater after being ﬁltered through a 0.45 mm membrane ﬁlter with a thickness of 40 nm. The coated PEDOT-PSS ﬁlm was dried at 150 1C for 1 min. The Ag metal electrode was thermally deposited at a thickness of 120 nm through a shadow mask with a device area of 0.38 cm2. The current density–voltage (J–V) characteristics of the OSC devices were measured under AM1.5 simulated illumination with an intensity of 100 mW/cm2 (Pecell Technologies Inc., PEC-L11 model) [10]. The intensity of sunlight illumination was calibrated using a standard Si photodiode detector with a KG-5 ﬁlter. The J–V curves were recorded automatically with a Keithley SMU 2400 source meter via illumination of the OSCs.

3. Results and discussion Fig. 1 presents the continuous fabrication process of the ZTO/Ag/ZTO multilayer electrode, which occurred under vacuum to prevent the formation of the interfacial layer between the ZTO and Ag layer. After deposition of the bottom ZTO (35 nm) layer onto the glass substrate, the Ag layer was sputtered onto the bottom ZTO layer with varying Ag thicknesses using tilted cathode guns to control the electrical and optical properties of the multilayer electrodes. The symmetric ZTO/Ag/ZTO structure was completed with deposition of the top ZTO layer under sputtering conditions identical to those of the bottom ZTO layer at room temperature in contrast to the conventional crystalline ITO electrode, which was deposited at 250–300 1C [23,24]. Fig. 2 shows the dependences of sheet resistance and resistivity of the ZTO/Ag/ZTO multilayer electrode on the thickness of the Ag layer; the inset picture shows the measured sheet resistance of the optimized ZTO/Ag/ZTO multilayer electrode. Compared to the single ZTO layer (Rsh: 3.24 105 O/square and r: 6.48 O cm), the Ag-inserted multilayer showed remarkably reduced sheet resistance and resistivity, much lower than those of the conventional crystalline ITO electrode [24]. The ZTO/Ag/ZTO multilayer showed decreased sheet resistance and resistivity with increasing Ag thickness. Note that the resistivity of the ZTO/Ag/ZTO multilayer

Fig. 1. Continuous bottom ZTO, Ag, and top ZTO sputtering process for fabricating ZTO/Ag/ZTO multilayer electrode structure without breaking the vacuum.

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

equation [19]:

101

1 1 1 1 2 1 2dZTO dAg ¼ þ þ ¼ þ ¼ þ RT RZTO,bottom RAg RZTO,top RZTO RAg rZTO rAg

100 10-1

103

10-2 102 10-3

ρ [ohm-cm]

Sheet Resistance [ohm/sq.]

106 5x105

101 10-4 100

10-5 0

4

6

8 10 12 14 Ag Thickness [nm]

16

18

Fig. 2. Sheet resistance and resistivity of ZTO/Ag/ZTO multilayer electrodes as a function of Ag thickness for constant top ZTO and bottom ZTO layers; the inset picture shows the low sheet resistance of the ZTO/Ag(12 nm)/ZTO multilayer electrode.

Resistivity [10-5 ohm-cm]

3.0

2.5

2.0

1.5

1.0 1

10

100

1617

1000

1000/T [K] Fig. 3. Temperature dependence of resistivities for the ZTO/Ag(12 nm)/ZTO multilayer electrode deposited on glass substrate.

electrode was nearly saturated at 10 5 O cm, due to the connection of Ag islands, which act as electron transfer channel. Although the ZTO/Ag/ZTO electrode with an 18-nm-thick Ag layer showed the lowest resistivity, the Ag thickness was expected to be optimized at less than 18 nm, considering the low transparency of the thick Ag layer. Fig. 3 shows the resistivity variation in optimized ZTO/Ag (12 nm)/ZTO multilayer electrode as a function of temperature from 350 to 2 K. The ﬁgure also shows that the temperature coefﬁcient of resistivity (dr/dT) was positive from 350 to 2 K, indicating that the ZTO/Ag/ZTO electrode possessed typical metallic characteristics in electrical transport properties. This indicates that the dominant current ﬂow path in the ZTO/Ag/ZTO multilayer electrode was the inserted Ag layer, which controlled the overall electrical properties of the ZTO/Ag/ZTO multilayer electrode. Because transparent electrodes with metallic characteristics (dr/dT 40) were interesting due to their high conductivities, ZTO/Ag/ZTO electrodes with positive dr/dT values are expected to possess much greater conductivity than are conventional ITO electrodes with a negative dr/dT value. Fig. 4(a) shows the calculated resistivity of the Ag layer inserted between the top and bottom ZTO using the following

Assuming that the total resistance (RT) of the ZTO/Ag/ZTO multilayer results from the resistance of the single bottom ZTO (RZTO,bottom), Ag (RAg), and top ZTO (RZTO,top) layers coupled in parallel as in the inset of Fig. 4(a), it is possible to calculate the resistivity of the Ag layer. If the resistivity (rZTO ¼6.48 O cm) and thickness (dZTO) of the single ZTO layer and the thickness (dAg) of the Ag layer are known, the resistivity of the Ag layer can be calculated. Like the resistivity of the ZTO/Ag/ZTO electrode, the resistivity of the Ag layer decreased as the thickness of the Ag layer increased as shown in Fig. 4(a). In particular, the resistivity of the Ag layer with a thickness of 4 nm was fairly high because the Ag layer at that thickness existed as isolated islands, as conﬁrmed by the FESEM image in Fig. 4(b). However, an increase in Ag thickness to greater than 6 nm led to a decrease in the resistivity of the Ag layer, making it comparable to that of bulk Ag (1.587 10 6 O cm at 293 K) [25]. The resistivity dependence of the Ag layer on its thickness may be explained by the morphology and shape of the Ag layer, as shown in Fig. 4(b). At an Ag thickness of 4 nm, disconnected Ag islands appeared. Due to the presence of these islands, the ZTO/Ag/ZTO with a 4 nm Ag layer showed higher resistivity. The Ag islands began to merge at a thickness of 6 nm; however, due to the agglomeration and merging of the Ag layer, some of the bottom ZTO layer was not fully covered by the Ag layer. Due to the merging of the Ag islands, the resistivity of the Ag layer signiﬁcantly decreased, inﬂuencing the resistivity of the ZTO/Ag/ZTO multilayer electrode. In the cases of the 8–14-nm-thick Ag layers, a continuous Ag ﬁlm was formed on the bottom ZTO layer, although some Ag islands formed on the thin continuous Ag layer. As discussed by Wang et al. [26], an Ag layer grown on a crystal surface or interface under special conditions, follows the Stranski–Krastanov growth mechanism, a layer-plusisland growth model. For thicknesses greater than 16 nm, the Ag layer covered most of the surface of the bottom ZTO layer. Fig. 5 shows the optical transmittance of symmetric ZTO/Ag/ ZTO and asymmetric Ag/ZTO multilayers as a function of Ag thickness, with the pictures showing their optical transparencies. The optical transmittance of the symmetric ZTO/Ag/ZTO multilayer electrodes was critically inﬂuenced by the thickness of the inserted Ag layer. The ZTO/Ag/ZTO with an Ag thickness of 4 nm had low optical transparencies at wavelengths between 400 and 650 nm due to the scattering of light by the randomly distributed Ag islands shown in the surface FESEM image (Fig. 4(b)). It is noteworthy that the insertion of an 8–12-nm-thick Ag layer led to a remarkable increase in the optical transmittance in the wavelength region, corresponding to the absorption wavelength of the active layer in the IOSCs. However, further increases in Ag thickness resulted in decreases in the optical transmittances of the ZTO/Ag/ZTO multilayer electrodes. In particular, the transmittance in the infrared wavelength region decreased linearly with increasing Ag layer thickness because the plasma resonance frequency depends on the carrier concentration of the ZTO/Ag/ ZTO multilayer electrodes. However, the shift of the plasma edge did not affect the performances of the inverted OSCs because the main absorption region of the organic active layer (P3HT:PCBM) was between 400 and 600 nm [11]. The pictures also show the transparency dependence of the ZTO/Ag/ZTO multilayer electrode on the thickness of the inserted Ag. As expected from the transparency of the ZTO/Ag/ZTO electrodes, the ZTO/Ag/ZTO electrode with 4-nm-thick Ag layer was bluish in color due to light scattering. However, in speciﬁc Ag thickness regions (6–12 nm), the ZTO/Ag/ZTO samples showed very high transparencies in spite of the Ag layer insertion. For Ag thicknesses greater

1618

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

Ag Resistivity [10-5 ohm-cm]

2.0

1.0

4

6

8

10 12 14 Ag Thickness [nm]

16

18

Fig. 4. (a) Calculated resistivities of Ag inserted between ZTO layers and (b) morphology of the Ag layer on the bottom ZTO layer with increasing thickness.

than 14 nm, the sample again was a bluish color due to reﬂection from the Ag layer. Fig. 5(b) shows the optical transmittances of the asymmetric Ag/ZTO samples without the top ZTO layer as a function of Ag thickness with the pictures showing the transparencies. The Ag/ZTO layer with the same thickness as that of the ZTO/Ag/ZTO multilayer electrode exhibited a much lower optical transmittance than did the ZTO/Ag/ZTO electrodes due to the absence of the antireﬂection effect. Unlike the ZTO/Ag/ZTO multilayer electrodes, the asymmetric Ag/ZTO electrode shows linearly decreasing optical transmittance with increasing Ag thickness. The picture also shows that all asymmetric Ag/ZTO samples were dark blue, regardless of Ag thickness. The increased optical transmittance observed in the ZTO/Ag/ZTO multilayer electrode at special Ag thicknesses could be attributed to antireﬂection effects. As suggested by Fan et al. [27], when an Ag mirror layer is embedded between TiO2 layers, the resulting symmetric TiO2/Ag/ TiO2 multilayer structure can suppress reﬂections from the Ag layer and obtain a high optical transmittance in the visible wavelength region. To conﬁrm the antireﬂection effect in the ZTO/Ag/ZTO multilayer electrode, ellipsometry analysis was performed assuming ﬂat interfaces. The complex refractive indices (n + ik) and the thicknesses of the Ag layers were estimated using a combination of the parametric optical constant (POC) and Drude models through a multilayer model analysis of the ellipsometric angles (C, D). We employed the POC model developed by Johs et al. [28], which provides a Kramers–Kronig-consistent model dielectric function. The surface roughness was included as a parameter in the multilayer modeling by assuming a 50% volume fraction for both the ambient and the ZTO. Fig. 6 shows the estimated complex refractive indices (N ¼n + ik) of the ZTO ﬁlm and the Ag thin ﬁlms as a function of the Ag thickness. The values of the

refractive index (n) and the extinction coefﬁcient (k) of the Ag layer were almost the same except for that of the thinnest Ag ﬁlm at t ¼4 nm. The optical constants of the Ag ﬁlm were similar to that of bulk Ag for ﬁlm thicknesses equal to and greater than 6 nm. Fig. 7 shows the simulated transmittances of ZTO (35 nm)/Ag (t nm)/ZTO (35 nm)/glass and Ag(t nm)/ZTO(35 nm)/glass as functions of the Ag ﬁlm thickness (t). We found that the transmittance of ZTO/Ag/ZTO/glass was greater than that of ZTO (70 nm)/glass in the visible spectral range for all of the ﬁlms except at t ¼4 nm, whereas we observed a reduced transmittance for Ag(t nm)/ ZTO(35 nm)/glass. The increase in the transmittance was the greatest for t ¼6 nm and covered the widest band between 350 and 850 nm (Fig. 7(a)). The multilayer model simulation assuming ﬂat interfaces showed that the enhanced transmittance of the ZTO/Ag/ZTO/glass was due to the antireﬂection behavior of the symmetric ZTO/Ag/ZTO ﬁlm structure. Using the estimated n and k values, we simulated the normal transmittances of various ZTO/Ag/ZTO/glass thin ﬁlms (Fig. 7(a)) and found qualitative agreement with the measured transmittance data (Fig. 5(a)). Similar increase in the transmittance in the visible spectra range were reported for ITO/Ag/ITO and IZO/Ag/IZO/glass systems [21,23] and an antireﬂective coating obtained using a combination of dielectric and metallic layers was reported by Berning and Turner [29]. The phenomenon of enhanced transmittance caused by a dielectric–metal–dielectric (D–M–D) optical coating layer is called induced transmittance [29–31]. According to Refs. [29,30], symmetric designs consisting of dielectric layer stacks around a central layer of a metal thin ﬁlm can produce very high transmittances in the visible spectral range if the optical constants of the metal are characterized by high values of the ratio k/n. This condition is satisﬁed for Ag, which has k/n values greater than

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

1619

2.5 100 ZTO/Ag/ZTO

2.0

80

n

Transmittance [%]

1.5 60

1.0 40

ZTO 4 nm 6 nm 8 nm 10 nm

0.5 4 nm 6 nm 8 nm 10 nm

20

12 nm 14 nm 16 nm 18 nm

0.0 1

0 200

400

600 800 Wavelength [nm]

1000

2

1200

3 4 Photon Energy [eV]

12 nm 14 nm 16 nm 18 nm 5

6

10 ZTO 4 nm 6 nm 8 nm 10 nm

8

100 Ag/ZTO

6 k

80 Transmittance [%]

12 nm 14 nm 16 nm 18 nm

60

4

40

2

20

0 200

4 nm 6 nm 8 nm 10 nm 400

12 nm 14 nm 16 nm 18 nm 600

0 1

2

3

4

5

6

Photon Energy [eV] 800

1000

1200

Wavelength [nm] Fig. 5. Optical transmittances of (a) symmetric ZTO/Ag/ZTO and (b) asymmetric Ag/ZTO multilayer electrodes as a function of Ag thickness with pictures showing the transparencies of the ZTO/Ag/ZTO and Ag/ZTO samples. (For interpretation of the references to color in this ﬁgure, the reader is referred to the web version of this article.)

50 over most of the visible spectral range. Analysis of symmetric D–M–D three-layer systems, where D is a high-index (e.g., n ¼2.3) dielectric material, and M is a sufﬁciently thin metal layer of Ag, showed that such systems behave as broadband low-reﬂecting coatings. This results from the equivalent indexes of the D–M–D symmetric multilayer systems, which are nearly real and similar to the index of air or glass over the visible spectral range, and from the equivalent complex phase thicknesses with minimal imaginary parts over the same visible region. As a consequence, broadband high-transmittance coatings can be achieved. To simultaneously optimize the electrical and optical properties of ZTO/Ag/ZTO multilayer electrodes and to determine the thickness of the Ag layer, we calculated the ﬁgure of merit value (T10/Rsh) as a function of Ag thickness [32]. The ﬁgure of merit value was calculated using the average transmittance (T) and sheet resistance (Rsh) of the ZTO/Ag/ZTO multilayer electrodes. Fig. 8 shows the average transmittance (400–650 nm) of the electrodes and the ﬁgure of merit value

Fig. 6. Plots of (a) the estimated refractive indices (n) and (b) the extinction coefﬁcients (k) of the complex refractive indices (N ¼n+ ik) of the ZTO ﬁlm and the Ag thin ﬁlms as functions of wavelength for various Ag ﬁlm thicknesses.

calculated using the average transmittance from Fig. 5(a) and sheet resistance from Fig. 2; the inset picture shows comparison of the optimized ZTO/Ag/ZTO and crystalline ITO samples. For an Ag thickness of 4 nm, the ZTO/Ag/ZTO electrode had a fairly low ﬁgure of merit due to the high sheet resistance and low optical transmittance caused by light scattering of the unconnected Ag islands. With increasing Ag thickness, the ﬁgure of merit value also increased due to decreased sheet resistance caused by the Ag island connections and increased optical transmittance caused by antireﬂection. Despite the high transmittance of the ZTO/Ag (10 nm)/ZTO electrode, the ZTO/Ag (12 nm)/ZTO electrode showed a slightly higher ﬁgure of merit value due to its lower sheet resistance. Note that the highest fTC value (45.87 10 3 O 1) obtained from the ZTO/Ag (12 nm)/ ZTO electrode was much higher than that of a commercial crystalline ITO electrode (11.78 10 3 O 1) or the ITO/Ag/ITO electrodes ( 24.7 10 3 O 1), indicating that the ZTO/Ag/ZTO multilayer electrode is a promising substitute for conventional ITO electrodes [33]. However, further increases in the Ag thickness reduced the ﬁgure of merit value because the optical transmittances of the ZTO/Ag/ZTO electrodes decreased dramatically when the Ag layer completely covered the bottom ZTO layer at thicknesses greater than 14 nm. The optimized ZTO/Ag/ZTO sample shown in the inset

1620

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

100 100

40

Transmittance [%]

80

Transmittance [%]

80

60

60 20 40

40 4 nm 6 nm 8 nm 10 nm ZTO

20

0 200

400

20

12 nm 14 nm 16 nm 18 nm

ZTO/Ag/ZTO

ITO

Figure of merit [10-3 ohm-1]

ZTO/Ag/ZTO

0

0 ITO

4

6

8

10

12

14

16

18

Ag Thickness [nm] 600 800 Wavelength [nm]

100

1000

1200

Ag/ZTO

Fig. 8. Calculated ﬁgure of merit values based on transmittances and sheet resistances of ZTO/Ag/ZTO multilayer electrodes with increasing Ag thickness. The inset picture shows a comparison of the optimized ZTO/Ag/ZTO multilayer electrode (the highest ﬁgure of merit) with a conventional crystalline ITO electrode.

100

60 Ag

80 Atomic percent [%]

Transmittance [%]

80

40

20

0 200

4 nm 6 nm 8 nm 10 nm ZTO 400

12 nm 14 nm 16 nm 18 nm

600 800 Wavelength [nm]

1000

1200

60 O 40 Sn 20

Fig. 7. Plots of the simulated transmittances of (a) ZTO(35 nm)/Ag(t nm)/ ZTO(35 nm)/glass and (b) Ag(t nm)/ZTO(35 nm)/glass as functions of wavelength for various Ag ﬁlm thicknesses. For comparison, we also plotted the simulated transmittance of ZTO(70 nm)/glass (blue line). (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

Zn

0 200

400

600

800

1000

1200

Sputter time [sec] 7.0 Ag 3d

367.48eV

6.0 Intensity [105 Counts/sec]

of Fig. 8 exhibited a very high transparency comparable to those of commercial crystalline ITO electrodes. Fig. 9(a) shows the XPS depth proﬁles result obtained from the optimized ZTO/Ag(12 nm)/ZTO electrode. The individual layers of top ZTO, Ag, and bottom ZTO existed symmetrically without interfacial reaction. The constant atomic compositions of Zn, Sn, and O atoms in the bottom and top ZTO layers showed that both layers were identically deposited onto the glass substrate and Ag layer. In addition, the symmetrical features of the layers demonstrate that the layers were identical, having the same composition and thickness. Furthermore, there was neither out-diffusion of the Ag layer nor interfacial reaction between the Ag and ZTO layers, since our continuous ZTO, Ag, and ZTO deposition processes were carried out at room temperature. The constant Ag 3d (367.48, 373.48 eV) peak positions in Fig. 9(b), obtained from the interfacial region, conﬁrmed that there was no interfacial reaction between the Ag and ZTO layers within the XPS’s detection limit, consistent with the XPS depth proﬁle results in Fig. 9(a). Note that the dissociation of the ZTO layer in the interface region was difﬁcult because the enthalpies of formation for ZnO ( 350.5 kJ/mol) and

5.0

373.48eV

4.0 3.0 2.0 1.0 0.0 380 378 376 374 372 370 368 366 364 362 360 Binding energy [eV]

Fig. 9. (a) XPS depth proﬁle of optimized ZTO/Ag (12 nm)/ZTO multilayer electrode and (b) Ag 3d peak obtained from the interfacial region between the Ag and top ZTO layer showing no interfacial reactions.

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

Ag (111)

SnO2 ( 577.6 kJ/mol) were much higher than that of Ag2O ( 31.1 kJ/mol) [25]. The stable interface was also conﬁrmed via the high resolution electron microscopy (HREM) results in Fig. 11, as described in the next paragraphs. To investigate the microstructural properties of ZTO/Ag/ZTO multilayer electrodes, both XRD and HREM examinations were carried out. Fig. 10 shows XRD plots of the ZTO/Ag/ZTO multilayers with increasing Ag thickness. Up to a thickness of 8 nm, there was no peak related to crystalline Ag or ZTO. The XRD plots of the ZTO/Ag/ZTO multilayers with Ag thicknesses of 6 and 8 nm are close to that of the single ZTO sample, showing only a broad halo pattern indicating the amorphous structure of the Ag and ZTO layers. A crystalline Ag (1 1 1) peak started to appear at an Ag thickness of 10 nm. Increasing the Ag thickness leads to the increase of the Ag (1 1 1) peak from 10 to 16 nm. Fig. 11(a) shows the cross-sectional HREM image of an optimized ZTO/Ag(12 nm)/ZTO multilayer electrode, as well as TED patterns of the bottom ZTO (B-ZTO), Ag, and top ZTO (T-ZTO)

Intensity [a.u.]

Ag 16nm Ag 14nm Ag 12nm Ag 10nm Ag 8nm Ag 6nm Single ZTO 20

30

40 2theta [deg.]

50

60

Fig. 10. XRD plots of the ZTO/Ag/ZTO multilayer electrode as a function of Ag thickness.

1621

layers. The cross-sectional images clearly demonstrate welldeﬁned bottom B-ZTO, Ag, and T-ZTO layers without interface layers. These sharp interfaces indicate that there were no interface reactions and no formation of interfacial oxide layers between the ZTO and Ag layers due to the use of a continuous and low temperature sputtering process without a break in the vacuum. In addition, the uniform contrast (Fig. 11(b)) of the ZTO layer indicates that the structures of the ZTO layers on the glass substrate and the Ag layer were completely amorphous, which was conﬁrmed by the TED patterns. However, the inserted Ag layer existed in a crystalline form as shown in the enlarged images and TED pattern of the Ag layer in Fig. 11(c), consistent with the XRD results. Even though the Ag layer was sputtered at room temperature, it had a crystalline structure as reported previously [23]. We compared the performances of IOSCs fabricated on an optimized ZTO/Ag/ZTO multilayer electrode and on a crystalline ITO electrode (reference sample). Fig. 12(a) shows the schematic structure of an IOSC fabricated on the ZTO/Ag/ZTO multilayer electrode with a sheet resistance of 3.96 O/square and an optical transmittance of 86.29%. Fig. 12(b) shows the photocurrent density–voltage (J–V) curves of the IOSCs fabricated on the ZTO/ Ag/ZTO and c-ITO electrode measured under an illumination of 100 mW/cm2 at AM 1.5 G conditions with an inset picture of the IOSC. The IOSC fabricated on the ZTO/Ag/ZTO electrode exhibited a short circuit current (JSC) of 7.95 mA/cm2, a ﬁll factor (FF) of 0.588%, an open circuit voltage (VOC) of 0.546 V, and a calculated power conversion efﬁciency (PCE) of 2.55%.These results were comparable with those of the IOSCs with the c-ITO electrode. Table 1 summarizes the performances of the IOSCs fabricated on the ZTO/Ag/ZTO and c-ITO electrodes. Note that the PCE obtained from the IOSCs on the ZTO/Ag/ZTO electrode was similar to that of the IOSC fabricated on the c-ITO electrode. Despite the lower resistivity of the ZTO/Ag/ZTO electrode, both IOSCs performed similarly, which could be attributed to the small active areas ( 0.38 cm2) of the IOSCs. However, it is thought that the low sheet resistance of the ZTO/Ag/ZTO electrode had a signiﬁcant effect on the performances of the IOSCs with large active areas. As a result, the overall photovoltaic characteristics of the IOSCs on the ZTO/Ag/ZTO electrode were comparable to those of the cells on the c-ITO electrode. In addition, the qualitative similarity of

Fig. 11. (a) Cross-sectional HREM images of optimized ZTO/Ag/ZTO electrode and TED patterns of each top ZTO, Ag, and bottom ZTO layer. Enlarged images of the (b) ZTO layer and (c) interfacial region between the Ag and ZTO layer.

1622

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

(2.55%) comparable to that of the IOSC prepared on the crystalline ITO electrode (2.45%), indicating that the ZTO/Ag/ZTO multilayer is a promising indium-free transparent electrode substitute for the conventional ITO electrode for use in cost-efﬁcient IOSCs.

ZnO ZTO Ag ZTO

Ag

Acknowledgment This work was ﬁnancially supported by a grant from the Gyeonggi-do International Collaborative Research Program.

PEDOT:PSS P3HT:PCBM ZnO

References

Glass

Current density [mA/cm2]

10 c-ITO ZTO/Ag/ZTO

8 6 4 2 0 -2 -0.1

0.0

0.1

0.2 0.3 Voltage [V]

0.4

0.5

0.6

Fig. 12. (a) Schematic structure of the IOSC fabricated on the ZTO/Ag/ZTO multilayer electrode. (b) Current density–voltage characteristics (AM 1.5 G with an incident light power intensity of 100 mW cm 2) of IOSCs fabricated on the ZTO/ Ag/ZTO multilayer electrode and reference ITO electrode with an inset picture of the ZTO/Ag/ZTO-based IOSC. Table 1 Comparison of IOSC performance fabricated on optimized ZTO/Ag/ZTO multilayer electrode and reference crystalline ITO electrode.

ITO ZTO/Ag/ZTO

Voc (V)

Jsc (mA/cm2)

FF (%)

PCE (%)

0.55 0.55

7.79 7.95

0.58 0.59

2.46 2.55

the photovoltaic characteristics shown in Fig. 12(b) indicates the possibility of substituting the c-ITO with the ZTO/Ag/ZTO multilayer electrode. More importantly, considering that it was prepared at room temperature, it is believed that the ZTO/Ag/ZTO electrode is better suited to the fabrication of low-cost IOSCs.

4. Conclusion Indium-free ZTO/Ag/ZTO multilayer electrodes with low resistivity and high transmittance were studied as substitutes for conventional ITO electrodes for cost-efﬁcient IOSCs. By using the low resistivity of the Ag layer and the antireﬂection effects of the oxide–Ag–oxide structure, we obtained transparent and low resistance ZTO/Ag/ZTO multilayer electrodes that were prepared at room temperature. It was found that the tunable electrical and optical properties of the ZTO/Ag/ZTO multilayer electrodes were critically affected by the thickness and morphology of the Ag layer. In addition, the IOSCs fabricated on the optimized ZTO/ Ag(12 nm)/ZTO electrode showed a power conversion efﬁciency

[1] F.C. Frebs, T. Tromholt, M. Jorgensen, Upscaling of polymer solar cell fabrication using full roll-to-roll processing, Nanoscale 2 (2010) 873–886. [2] F.C. Krebs, All solution roll-to-roll processed polymer solar cells free from indium-tin-oxide and vacuum coating steps, Organic Electron. 10 (2009) 761–768. [3] F.C. Krebs, Fabrication and processing of polymer solar cells: a review of printing and coating technique, Sol. Energy. Mater. Sol. Cells. 93 (2009) 394–412. [4] C. Brabec, V. Dyakonov, U. Scherf, Organic Photovoltaics, 1st ed., Wiley-VCH Verlag, GmbH&Co, Weinheim, 2008. [5] B.C. Thompson, J.M.J. Fre´chet, Polymer–fullerene composite solar cells, AngewChem. Int. Ed. 47 (2007) 58–77. [6] C.S. Kim, L.L. Tinker, B.F. DiSalle, E.D. Gomez, S. Lee, S. Bernhard, Y.-L. Loo, Altering the thermodynamics of phase separation in inverted bulk-heterojunction organic solar cells, Adv. Mater. 21 (2009) 3110–3115. [7] D.W. Zhao, P. Liu, X.W. Sun, S.T. Tan, A.K.K. Kyaw, An inverted organic solar cell with an ultrathin Ca electron-transporting layer and MoO3 hole-transporting layer, Appl. Phys. Lett. 95 (2009) 153304-1–153304-3. ¨ ¨ [8] H. Schmidt, H. Flugge, T. Winkler, T. Bulow, T. Riedl, W. Kowalsky, Efﬁcient semitransparent inverted organic solar cells with indium tin oxide top electrode, Appl. Phys. Lett. 94 (2009) 243302-1–243302-3. [9] T. Ameri, G. Dennler, C. Waldauf, H. Azimi, A. Seemann, K. Forberich, J. Hauch, M. Scharber, K. Hingerl, C.J. Brabec, Fabrication, optical modeling, and color characterization of semitransparent bulk-heterojunction organic solar cells in an inverted structure, Adv. Funct. Mater. 20 (2010) 1592–1598. [10] S.-Y. Park, W.-I. Jeong, D.-G. Kim, J.-K. Kim, D.-C. Lim, J.H. Kim, J.-J. Kim, J.-W. Kang, Large-area organic solar cells with metal subelectrode on indium tin oxide anode, Appl. Phys. Lett. 96 (2010) 173301-1–173301-3. [11] K.-H. Choi, J.-A. Jeong, H.-K. Kim, Dependence of electrical, optical, and structural properties on the thickness of IZTO thin ﬁlms grown by linear facing target sputtering for organic solar cells, Sol. Energy Mater. Sol. Cells. 94 (2009) 1822–1830. [12] S.-I. Na, S.-S. Kim, J. Jo, D.-Y. Kim, Efﬁcient and ﬂexible ITO-free organic solar cells using highly conductive polymer anodes, Adv. Mater. 20 (2008) 4061–4067. [13] Y. Wang, X. Chen, Y. Zhong, F. Zhu, K.P. Loh, Large area, continuous, fewlayered graphene as anodes in organic photovoltaic devices, Appl. Phys. Lett. 95 (2009) 063302-1–063302-3. [14] J. Wu, H.A. Becerril, Z. Bao, Z. Liu, Y. Chen, P. Peumans, Organic solar cells with solution-processed graphene transparent electrodes, Appl. Phys. Lett. 92 (2008) 263302-1–263302-3. [15] T.M. Barnes, J.D. Bergeson, R.C. Tenent, B.A. Larsen, G. Teeter, K.M. Jones, J.L. Blackburn, J. Lagemaat, Carbon nanotube network electrodes enabling efﬁcient organic solar cells without a hole transport layer, Appl. Phys. Lett. 96 (2010) 243309-1–243309-3. [16] G. Fanchini, S. Miller, B.B. Parekh, M. Chhowalla, Optical anisotropy in singlewalled carbon nanotube thin ﬁlms: implications for transparent and conducting electrodes in organic photovoltaics, Nano Lett. 8 (2008) 2176–2179. [17] J.-H. Park, K.-J. Ahn, K.-I. Park, S.-I. Na, H.-K. Kim, An Al-doped ZnO electrode grown by highly efﬁcient cylindrical rotating magnetron sputtering for low cost organic photovoltaics, J. Phys. D: Appl. Phys. 43 (2010) 1151011–115101-6. [18] H.-K. Park, J.-W. Kang, S.-I. Na, D.-Y. Kim, H.-K. Kim, Characteristics of indium-free GZO/Ag/GZO and AZO/Ag/AZO multilayer electrode grown by dual target DC sputtering at room temperature for low-cost organic photovoltaics, Sol. Energy Mater. Sol. Cells 93 (2009) 1994–2002. [19] J.-H. Park, Han-Ki Kim, H. Lee, H. Lee, S. Yoon, C.-D. Kim, Highly transparent, low resistance, and cost-efﬁcient Nb:TiO2/Ag/Nb:TiO2 multilayer electrode prepared at room temperature using black Nb:TiO2 target, Electrochem. Solid-State Lett. 13 (2010) J53–J56. [20] J.-A. Jeong, H.-K. Kim, Low resistance and highly transparent ITO–Ag–ITO multilayer electrode using surface plasmon resonance of Ag layer for bulkheterojunction organic solar cells, Sol. Energy Mater. Sol. Cells 93 (2009) 1801–1809. [21] K.-H. Choi, H.-J. Nam, J.-A. Jeong, S.-W. Cho, H.-K. Kim, J.-W. Kang, D.-G. Kim, W.-J. Cho, Highly ﬂexible and transparent InZnSnOx/Ag/InZnSnOx multilayer electrode for ﬂexible organic light emitting diodes, Appl. Phys. Lett. 92 (2008) 223302-1–223302-3.

Y.-Y. Choi et al. / Solar Energy Materials & Solar Cells 95 (2011) 1615–1623

[22] J.-A. Jeong, Y.-S. Park, H.-K. Kim, Comparison of electrical, optical, structural, and interface properties of IZO–Ag–IZO and IZO–Au–IZO multilayer electrodes for organic photovoltaics, J. Appl. Phys. 107 (2010) 023111-1–023111-8. [23] H.-K. Kim, K.-S. Lee, Effect of introducing H2O vapor on properties of RF sputter-grown ITO anode layer for OLEDs, Electrochem. Solid-State Lett. 11 (2008) J57–J59. [24] 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. [25] D.R. Lide, CRC Handbook of Chemistry and Physics, 76th ed., CRC Press, 1995, E-93. [26] P. Wang, D. Zhang, D.H. Kim, Z. Qiu, L. Gao, R. Murakami, X. Song, Enhancement of light transmission by coupling to surface plasmon polaritons of a layer-plus-islands silver layer, J. Appl. Phys. 106 (2009) 103104-1–103104-5. [27] J.C. Fan, F.J. Bachner, G.H. Foley, P.M. Zavracky, Transparent heat-mirror ﬁlms of TiO2/Ag/TiO2 for solar energy collection and radiation insulation, Appl. Phys. Lett. 25 (1974) 693–695.

1623

[28] B. Johs, C.M. Herzinger, J.H. Dinan, A. Cornfeld, J.D. Benson, Development of a parametric optical constant model for Hg1 xCdxTe for control of composition by spectroscopic ellipsometry during MBE growth, Thin Solid Films 313 (1998) 137–142. [29] P.H. Berning, A.F. Turner, Induced transmission in absorbing ﬁlms applied to band pass ﬁlter design, J. Opt. Soc. Am. 47 (1957) 230–239. [30] P.H. Berning, Principles of design of architectural coatings, Appl. Opt. 22 (1983) 4127–4141. [31] H.A. Macleod, Thin-Film Optical Filters, 2nd ed., Taylor & Francis, New York, 2001, pp. 328–345. [32] C. Haacke, New ﬁgure of merit for transparent conductors, J. Appl. Phys. 47 (1976) 4086–4089. [33] M. Bender, W. Seeling, C. Daube, H. Frankenberger, B. Ocker, J. Stollenwork, Dependence of ﬁlm composition and thicknesses on optical and electrical properties of ITO–metal–ITO multilayers, Thin Solid Films 326 (1998) 67–71.