- Email: [email protected]
Surface modifications of ITO electrodes for polymer light-emitting devices
Applied Surface Science 253 (2006) 2102–2107 www.elsevier.com/locate/apsusc
Surface modifications of ITO electrodes for polymer light-emitting devices Zhong Zhi You a,*, Jiang Ya Dong b a
College of Electronic Information Engineering, South-Central University for Nationalities, Hubei, Wuhan 430074, People’s Republic of China b School of Optoelectronic Information, University of Electronic Science and Technology of China, Sichuan, Chengdu 610054, People’s Republic of China Received 25 November 2005; received in revised form 9 March 2006; accepted 3 April 2006 Available online 8 May 2006
Abstract Surface modifications were performed on the indium tin oxide (ITO) substrates for polymer light-emitting devices, using the different treatment methods including solvent cleaning, hydrochloric acid treatment and oxygen plasma. The influence of modifications on the surface properties of ITO electrodes were investigated by X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), contact angle, and four-point probe. The surface energies of the ITO substrates were also calculated from the measured contact angles. Experimental results demonstrate that the surface properties of the ITO substrates strongly depend on the modification methods, and oxygen plasma more effectively improves the ITO surface properties compared with the other treatments. Furthermore, the polymer light-emitting electrochemical cells (LECs) with the differently treated ITO substrates as device electrodes were fabricated and characterized. It is observed that the surface modifications on ITO electrodes have a certain degree of influence upon the injection current, luminance and efficiency, but hardly upon the turn-on voltages of current injection and light emission which are close to the measured energy gap of electroluminescent polymer. Oxygen plasma treatment on the ITO electrode yields the better performance of the LECs, due to the improvement of interface formation and electrical contact of the ITO electrode with the polymer blend in the LECs. # 2006 Elsevier B.V. All rights reserved. PACS : 81.65.-b; 42.70.Nq; 82.47.Tp; 85.60.Jb Keywords: Surface modification; ITO electrode; Polymer light-emitting electrochemical cells; Polymer light-emitting device
1. Introduction As a new type of polymer light-emitting devices, polymer light-emitting electrochemical cells (LECs) have attracted increasing scientific and industrial interest due to their high electroluminescence efficiency, low operating voltage and ease of fabrication since its discorvey by Pei et al. in 1995 [1]. It is well known that the polymer LECs usually consist of a lightemitting active blend layer of conjugated luminescent polymer and solid electrolyte with supporting salt sandwiched between an indium tin oxide (ITO) electrode and a metal electrode [1–9]. When a sufficiently high bias is applied onto the electrodes, electrochemical doping takes place accompanied
* Corresponding author. Tel.: +86 27 67843890; fax: +86 27 67842854. E-mail address: [email protected] (Z.Z. You). 0169-4332/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2006.04.009
by the redistribution of the ions in the ploymer blend film. As a result, the contact at the ploymer electrode interface is ohmic, and the conjugated polymer is doped p-type near the anode and n-type near the cathode respectively, and a dynamic p–i–n junction is created in situ. Hence, light is emitted from the intrinsic region near the center. Since the polymer blend layer is in direct contact with the ITO electrode, the surface characteristics of ITO and the interfacial properties between the polymer blend layer and the ITO electrode play an important role in the operation of ploymer LECs. In this study, the effect of different modification methods upon the surface properties of ITO electrodes were investigated through the measurements of X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), sheet resistance, contact angle and surface energy. In addition, we studied the influence of surface modifications of the ITO electrodes on the
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
performance of ploymer LECs, in terms of the electrical and optical characteristics of the devices. 2. Experimental details
2103
Table 1 Surface tension parameters (in mJ/m2) of test liquids Liquid
g pl
g dl
gl
H2O CH2I2
51.0 2.3
21.8 48.5
72.8 50.8
2.1. ITO surface modification Two different sets of processing techniques, wet and dry cleanings were used to modify the surface of ITO substrates. They are: a. Solvent cleaning (S1)—The substrates were successive washed in an ultrasonic bath of acetone and alcohol each for 20 min, then rinsed in deionized water, and dried with nitrogen. b. Hydrochloric acid (HCl) treatment (S2)—The substrates were dipped into 10% HCl solution at room temperature for 15 min, and then rinsed in deionized water and finally dried in a flow of nitrogen. c. Oxygen plasma treatment (S3)—The substrates were exposed to the oxygen plasma in a home-made parallel plate plasma generator for 5 min under pressure 20 Pa and voltage 250 V at room temperature. The ITO-coated glass substrates (11 V/sq) used in this study were purchased from CSG Holding Co. Ltd. and cut into 32 mm 32 mm plates. All the ITO substrates were cleaned prior to surface treatment, first by rubbing with detergent, then in ultrasonic bath for 10 min in deionized water, and finally dried in a flow of nitrogen. In this paper, the untreated ITO is referred to as S0.
recorded within 30 s after the formation of the sessile drop. Five readings were taken, and the average was reported unless otherwise indicated. The typical error is around 28 in our present experiment. Distilled water (H2O) and diiodomethane (CH2I2) were chosen as the test liquids [11–14]. Both of them were reagent grade, and their surface tension and surface tension components are listed in Table 1. Based on the measured contact angles, the surface energy was calculated using the Owens–Wendt (OW) approach [11–13]. According to the OW method, the surface energy (gs) of a solid is equal to a sum of the polar (g ps ) and dispersion (g ds ) components: g s ¼ g ps þ g ds
(1)
and the following relation was used: g sl ¼ g s þ g l 2ð
qffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffi g ps g pl þ g ds g dl Þ
where gsl is the interfacial energy between the solid and the test liquid; gl is the surface tension of the liquid; and g pl and g dl are the polar and dispersion components of the surface tension of the liquid, respectively. When combining Eqs. (1) and (2) with the Young’s equation, the following expression can be obtained:
2.2. ITO characterization ð1 þ cos uÞg l ¼ 2ð The XPS measurements of the ITO substrates were carried out in a XSAM-800 spectrometer, using a monochromatic Al Ka (hn = 1486.60 eV) as X-ray source. Typically the operating pressure in the analysis chamber was about 107 Pa. All binding energies were referenced to the binding energy of the carbon C 1s peak at 285.0 eV. For the calculation of the atomic concentrations, a linear background correction was done, the peak areas were corrected with empirical sensitivity factors, the instruments transmission function and the specific mean free path lengths [10]. The ITO surfaces were structurally characterized by AFM with a Seiko Instruments SPA-400 in the contact mode. Measurements were achieved at room temperature in the air, using the same pyramidal Si3N4 tip. The surface roughness of the ITO samples were calculated with the AFM software. The sheet resistance of the ITO substrates were measured by a four-point probe (SZ-82) at room temperature in the air. The contact angle measurements were performed using a JY82 contact angle goniometer by the sessile drop technique [11–14] at 20 8C in an environmental chamber. A substrate was placed on the sample stage of the goniometer, and a microsyringe was used to deposit a liquid drop of 2–3 ml on the surface of the substrate. The steady-state contact angle was
(2)
qffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffi g ps g pl þ g ds g dl Þ
(3)
Here, u is the contact angle between the test liquid and the solid surface. Eq. (3), known as the OW method, is one of the approaches of estimating the solid surface energy. From Eq. (3), surface energy parameters g ps , g ds , gs and polarity xp (xp ¼ g ps =g s ) can be readily achieved from the values of u measured with a pair of test liquids whose surface tension components are known. 2.3. Polymer LECs fabrication Polymer LECs were fabricated using the polymer blend of poly [2-methoxy-5-(20 -ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) and poly(ethylene oxide) (PEO) complexed with lithium triflate (LiTf) as the active light-emitting layer, and the differently treated ITO substrates as device electrodes. The weight ratio of the MEH-PPV:PEO:LiTf in the solution was 12:5:2. The polymer blend film of thickness of about 200 nm was prepared by spin-coating cyclohexanone solution onto the ITO substrate. The silver layer (150 nm) was then deposited on the top of the blend film by the vacuum sublimation technique (about 104 Pa) to form ITO/MEH-PPV + PEO(LiTf)/Ag structure. The active area of each LEC device was 36 mm2.
2104
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
The organic materials were purchased from Aldrich Chemical Company and used without further purification. The polymer LECs fabricated with ITO samples S0, S1, S2, and S3 were referred to as devices D0, D1, D2, and D3, respectively. The bias voltage was applied with the ITO electrode as anode and the Ag electrode as cathode. The current–voltage– luminance (I–V–L) characteristics of the non-encapsulated devices were measured at a scan rate of 0.1 V/s with a KEITHLEY-4200 sourcemeter and a ST-86LA luminance meter. All measurements were performed at room temperature in ambient air. 3. Results and discussion 3.1. Modification effect on the ITO surface properties Table 2 summarizes the chemical composition, sheet resistance and surface roughness of the untreated and treated ITO substrates. From Table 2, We first note that the presence of
Table 2 Chemical composition, sheet resistance and surface roughness of the untreated and treated ITO substrates Substrate
S0 S1 S2 S3
Chemical composition (at.%) [C]
[O]
[In]
[Sn]
Sheet resistance Rs (V/sq)
5.6 9.2 4.9 4.6
15.1 14.9 13.6 16.5
68.8 67.5 72.5 68.5
10.5 8.4 9.0 10.4
11.0 10.9 14.7 10.5
Surface roughness Ra (nm) 3.2 3.1 6.4 2.3
carbon element, besides containing the elements of O, In and Sn on each ITO sample, indicating that carbon is the only major contaminant. Note also that the [C]:[O] ratios of substrates S0, S1, S2, and S3 are 5.6%:15.1%, 9.2%:14.9%, 4.9%:13.6%, and 4.6%:16.5%, respectively. Apparently, the oxygen plasma treatment largely decreases the carbon content and increases the oxygen content, and thereby more effectively enhances the work function of the ITO, since the ITO work function is
Fig. 1. AFM images of the untreated (a); solvent cleaned (b); hydrochloric acid treated (c); and oxygen plasma treated ITO substrates (d).
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
determined by the oxygen content and the carbon content [15–18]. Secondly, we observe that the sheet resistance (Rs) of substrates S0, S1, S2, and S3 are 11.0, 10.9, 14.7, and 10.5 V/sq, respectively. The oxygen palsma treatment yields the lowest value of Rs (10.5 V/sq), while HCl treatment gives the highest value of Rs (14.7 V/sq). Thirdly, the surface roughness of ITO substrates can be observed to decrease in the sequence of S2, S0, S1, S3. The average roughness (Ra) values of substrates S2, S0, S1, and S3 are 6.4, 3.2, 3.1, and 2.3 nm, respectively. Meanwhile, the sample S3 exhibits more smooth and homogenous surface, compared with the other ITO substrates (see Fig. 1). Fig. 2(a) shows the measured contact angle of the untreated and treated ITO substrates. Using the OW approach, the surface energy and polarity were derived from the measured contact angles and are plotted in Fig. 2(b). The errors in the surface energy and polarity were calculated out to be less than 1.9 mJ/m2 and 0.03, respectively. From Fig. 2(a), we find that the measured contact angle is subjected to the treatment methods of ITO substrates. Note that the water contact angle uH2 O of substrates S0, S1, S2, and S3 are 85.18, 64.58, 82.08, and 30.58 respectively, reducing in the order of S0, S2, S1, S3. Obviously, oxygen plasma brings about a large decrease in the measured water contact angle uH2 O compared to the unterated ITO, from 85.18 to 30.58, indicating that the oxygen plasma treated ITO surface appears to be highly polar. Note also that the substrate S3 exhibits the smallest diiodomethane contact angle uH2 O , followed by the substrates S1, S0, and S2. The oxygen plasma treatment gives the minimum diiodomethane contact angle uCH2 I2 (32.18). Fig. 2(b) indicates that the surface energies gs and the corresponding components (g ps , g ds ), and polarities xp are strongly dependent on the surface modifications. For the treated substrates S1, S2 and S3, both gs and xp more or less increase from 36.5 mJ/m2 and 0.08 to 36.8– 64.7 mJ/m2 and 0.12–0.54 respectively, in comparison with the untreated substrate S0. In particular, we note that oxygen plasma yields the highest surface energy (64.7 mJ/m2) and the highest surface polarity (0.54), mainly due to the significantly enhanced polar component g ps (35.1 mJ/m2). The increase of the surface energy and polarity may be attributed to two reasons. One is that oxygen plasma can remove effectively the hydrocarbon contaminant on the ITO surface. Since hydrocarbons have low surface energy, the removal of hydrocarbons from the surface can lead to an increase in the surface energy. In addition, to the removal of hydrocarbon contaminant from the surface, oxide bonds can form on the substrate surface during oxygen plasma treatment, which leads the surface to be more hydrophilic and shows higher surface energy. Another is that the molecular water monolayer chemisorbed exists on the ITO surface for the oxygen plasma treatment [19]. And it is the presence of monolayer adsorbate coverage of water that increases the water concentration. Therefore, accordingly the polarity increases significantly. These results indicate that oxygen plasma treatment not only decreases the sheet resistance and surface roughness, but also enhances the work funtion, surface energy and polarity, and hence improves the the surface properties of the ITO substrate.
2105
Fig. 2. (a) Contact angle and (b) surface energy and polarity of the untreated and treated ITO substrates.
3.2. Modification effect on the LECs optoelectrical properties Fig. 3(a) gives the absorption spectrum of a 100 nm MEHPPV film measured by a Perkin-Elmer Lambda 900 UV7VIS/ NIR spectrophotometer at room temperature. As can be seen, the absorption spectrum of the thin film has a relatively sharp peak at 500 nm, with an onset of absorption at 570 nm. Based on the measured absorbance (A) spectrum, the absorption coefficient (a) was calculated using the following formula [20–22]: a ¼ 2:303
A t
(4)
where t is the thickness of the thin film. The energy gap (Eg) can be obtained by fitting the optical absorption coefficient a to Tauc’s relation [23–26]: ðahnÞ2 ¼ Cðhn Eg Þ
(5)
Here, C is an energy-independent constant, hn is the photon energy. Fig. 3(b) shows the Tauc plot: (ahn)2 versus hn for the
2106
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
Fig. 3. (a) Absorption spectrum and (b) Tauc plot of electroluminescent polymer MEH-PPV thin film. Dotted line indicates the linear part of the curve.
Fig. 4. (a) Current–voltage and (b) luminance–voltage curves of LECs fabriI L cated with the untreated and treated ITO substrates. Von and Von present the turnon voltages of current injection and light emission, respectively.
MEH-PPV film. The value of Eg can be readily obtained by extrapolation of the linear portion of the graph to (ahn)2 = 0, which is about 2.18 eV. Fig. 4 illustrates the semilogarithmic plots of the measured V–I and L–V characteristics for the devices fabricated with the untreated and treated ITO electrodes. As shown in Fig. 4(a), the I turn-on voltage of current injection (Von ) for all the devices share almost the same value (about 2.2 V), close to the measured Eg l value of MEH-PPV (i.e., Von =Eg/e, e is the electronic charge). Note from Fig. 4(b) that the turn-on voltage of light emission L (Von ) for the devices D0, D1, D2, and D3 are 2.7, 2.6, 2.9, and L I 2.4 V, respectively, and the difference between Von and Von exists L among all the devices. Each Von is observed to be slightly higher I than its corresponding Von , indicating that the turn-on voltage of light emission lags behind that of current injection. Similar results have been reported by Yang and Pei [27], Ko et al. [28], Chuai et al. [29]. There may be two reasons for this voltage lag of light emission. One is related to the sensitivity of the light intensity detector which is not sensitive enough for the weak
L value. Another reason is luminance at the voltage lower than Von the relatively slow response of the LECs, which also make the light emission lag behind the current injection. Note also from Fig. 4 that the injection current (I), luminance (L) and efficiency (h) of the LECs depend on the ITO electrode modifications. At V = 7 V for devices D0, D1, D2, and D3, the values of I are 27.1, 29.2, 21.4, and 48.8 mA; and L are 25.5, 40.3, 6.9, and 98.2 cd/ m2, respectively. Then the h values of devices D0, D1, D2, and D3
Table 3 Device performance of LECs fabricated with the untreated and treated ITO substrates Device
D0 D1 D2 D3
I (V) Von
2.2 2.2 2.2 2.2
L Von (V)
2.7 2.6 2.9 2.4
V=7V I (mA)
L (cd/m2)
h (cd/A)
27.1 29.2 21.4 48.8
25.5 40.3 6.9 98.2
0.034 0.050 0.012 0.073
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
2107
can be readily calculated out to be about 0.034, 0.050, 0.012, and 0.073 cd/A, respectively (see Table 3). Clearly, the oxygen plasma-treated device in our present work yields the higher injection current, luminance and efficiency, exhibiting the relatively better performance, while the hydrochloric acidtreated device the poorer. Such an experimental result suggests that the device performance of LECs is closely related to the characteristics of electrical contact at the ITO/polymer interface. Oxygen plasma treatment results in not only higher work function, lower sheet resistance, smoother surface, but also higher surface energy and polarity of the ITO electrode which is highly important to improve the surface morphological properties of polymer thin film and to optimize the interface characteristics of ITO/polymer. Firstly, the increase of surface polarity of ITO electrode leads to a more uniform spreading of the polymer solution on the ITO surface, reduces the voids and pinholes, and thereby obtains the homogeneous and compact polymer thin film. Secondly, the increase of the surface energy of ITO electrode provides a better adhesion of the polymer film and reduces the interfacial energy between the polymer film and ITO electrode, and therefore increases the effective area available between these two materials. These two effects above-stated can significantly improve the interface formation and electrical contact of the ITO electrode with the polymer blend in the LECs, so as to facilitate the electrochemical doping of the electroluminescent polymer when sufficiently large bias was applied, and therefore yield the relatively better performance of the LECs. Contrary to the situation in the oxygen plasma treatment, yet, hydrochloric acid treatment results in a poor electrical contact to the emissive layer in the LECs and hinders charge injection through the electrochemical doping, and thereby produces poorer performance of the LECs.
degree of influence upon the injection current, luminance and efficiency, but hardly upon the turn-on voltages of current injection and light emission. Oxygen plasma treatment on the ITO electrode gives the better performance of the LECs, due to the improvement of interface formation and electrical contact between the ITO electrode and the polymer blend in the LECs.
4. Conclusion
[18] [19]
The effect of different modification methods on the surface properties of ITO electrodes were investigated by XPS, AFM, sheet resistance, contact angle and surface energy measurements. We observe that the surface properties of the ITO substrates are closely relevant to the modification methods, and the oxygen plasma is more efficient compared with the other treatments since it brings about smoother surface, lower sheet resistance, higher work function, higher surface energy and polarity of the ITO electrode. Furthermore, the influence of surface modifications of the ITO electrodes on the characteristics of polymer LECs were studied by the electrical and optical properties of the devices. Experimental results indicate that the surface modifications on ITO electrodes have a certain
Acknowledgment This work was financially supported by the National Natural Science Foundation of China (No. 60372002). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
[20] [21] [22] [23] [24] [25] [26] [27] [28] [29]
Q. Pei, G. Yu, C. Zhang, Y. Yang, A.J. Heeger, Science 269 (1995) 1086. Y. Yang, Q. Pei, Appl. Phys. Lett. 70 (1997) 1926. G. Yu, Q. Pei, A.J. Heeger, Appl. Phys. Lett. 70 (1997) 934. Y. Li, J. Gao, G. Yu, Y. Cao, A.J. Heeger, Chem. Phys. Lett. 287 (1998) 83. Y. Li, Y. Cao, J. Gao, D. Wang, G. Yu, A.J. Heeger, Synth. Met. 99 (1999) 243. F.-C. Chen, Y. Yang, Appl. Phys. Lett. 81 (2002) 4278. Q. Sun, H. Wang, C. Yang, G. He, Y. Li, Synth. Met. 128 (2002) 161. J.-C. Lepreˆtre, A. Deronzier, O. Ste´phan, Synth. Met. 131 (2002) 175. Q. Sun, X. Zhan, B. Zhang, C. Yang, Y. Liu, Y. Li, D. Zhu, Polym. Adv. Technol. 15 (2004) 70. D. Briggs, M.P. Seah, Practical Surface Analysis by Auger and X-ray Photoelectron Spectroscopy, Wiley, New York, 1983. D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (1969) 1741. R.J. Good, J. Adhes. Tech. 6 (1992) 1269. S. Wu, Polymer Interface and Adhesion, Marcel Dekker, New York, 1982. B. Janczuk, A. Zdziennicka, W. Wo´jcik, Appl. Surf. Sci. 120 (1997) 35. C.C. Wu, C.I. Wu, J.C. Sturm, A. Kahn, Appl. Phys. Lett. 70 (1997) 1348. J.S. Kim, M. Granstro¨m, R.H. Friend, N. Johansson, W.R. Salaneck, R. Daik, W.J. Feast, F. Cacialli, J. Appl. Phys. 84 (1998) 6859. M. Ishii, T. Mori, H. Fujikawa, S. Tokito, Y. Taga, J. Lumin. 87–89 (2000) 1165. Z.Z. You, J.Y. Dong, Appl. Surf. Sci. 249 (2005) 271. J.S. Kim, P.K.H. Ho, D.S. Thomas, R.H. Friend, F. Cacialli, G.W. Bao, S.F.Y. Li, Chem. Phys. Lett. 315 (1999) 307. M.-B. Park, N.-H. Cho, Appl. Surf. Sci. 190 (2002) 151. M.J. Alam, D.C. Cameron, Surf. Coat. Technol. 142–144 (2001) 776. A. Abu EL-Fadl, Galal A. Mohamad, A.B. Abd El-Moiz, M. Rashad, Phys. B: Condens. Matter. 366 (2005) 44. J. Tauc, R. Grigorovici, A. Vancu, Phys. Status Solidi 15 (1966) 627. Y. Ramadin, M. Al-Haj Abdallah, M. Ahmad, A. Zihlif, S.K.J. Al-Ani, S.G.K. Al-Ani, Opt. Mater. 5 (1996) 69. C. Baban, G.I. Rusu, Appl. Surf. Sci. 211 (2003) 6. T.S. Moss, G.J. Burrell, B. Ellis, Semiconductor Opto-Electronics, Butterworths, London, 1973. Y. Yang, Q. Pei, Appl. Phys. Lett. 68 (1996) 2708. H.C. Ko, D.K. Lim, S.-H. Kim, W. Choi, H. Lee, Synth. Met. 144 (2004) 177. Y. Chuai, D.N. Lee, C. Zhen, J.H. Min, B.H. Kim, D. Zou, Synth. Met. 145 (2004) 259.
Surface modifications of ITO electrodes for polymer light-emitting devices Zhong Zhi You a,*, Jiang Ya Dong b a
College of Electronic Information Engineering, South-Central University for Nationalities, Hubei, Wuhan 430074, People’s Republic of China b School of Optoelectronic Information, University of Electronic Science and Technology of China, Sichuan, Chengdu 610054, People’s Republic of China Received 25 November 2005; received in revised form 9 March 2006; accepted 3 April 2006 Available online 8 May 2006
Abstract Surface modifications were performed on the indium tin oxide (ITO) substrates for polymer light-emitting devices, using the different treatment methods including solvent cleaning, hydrochloric acid treatment and oxygen plasma. The influence of modifications on the surface properties of ITO electrodes were investigated by X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), contact angle, and four-point probe. The surface energies of the ITO substrates were also calculated from the measured contact angles. Experimental results demonstrate that the surface properties of the ITO substrates strongly depend on the modification methods, and oxygen plasma more effectively improves the ITO surface properties compared with the other treatments. Furthermore, the polymer light-emitting electrochemical cells (LECs) with the differently treated ITO substrates as device electrodes were fabricated and characterized. It is observed that the surface modifications on ITO electrodes have a certain degree of influence upon the injection current, luminance and efficiency, but hardly upon the turn-on voltages of current injection and light emission which are close to the measured energy gap of electroluminescent polymer. Oxygen plasma treatment on the ITO electrode yields the better performance of the LECs, due to the improvement of interface formation and electrical contact of the ITO electrode with the polymer blend in the LECs. # 2006 Elsevier B.V. All rights reserved. PACS : 81.65.-b; 42.70.Nq; 82.47.Tp; 85.60.Jb Keywords: Surface modification; ITO electrode; Polymer light-emitting electrochemical cells; Polymer light-emitting device
1. Introduction As a new type of polymer light-emitting devices, polymer light-emitting electrochemical cells (LECs) have attracted increasing scientific and industrial interest due to their high electroluminescence efficiency, low operating voltage and ease of fabrication since its discorvey by Pei et al. in 1995 [1]. It is well known that the polymer LECs usually consist of a lightemitting active blend layer of conjugated luminescent polymer and solid electrolyte with supporting salt sandwiched between an indium tin oxide (ITO) electrode and a metal electrode [1–9]. When a sufficiently high bias is applied onto the electrodes, electrochemical doping takes place accompanied
* Corresponding author. Tel.: +86 27 67843890; fax: +86 27 67842854. E-mail address: [email protected] (Z.Z. You). 0169-4332/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2006.04.009
by the redistribution of the ions in the ploymer blend film. As a result, the contact at the ploymer electrode interface is ohmic, and the conjugated polymer is doped p-type near the anode and n-type near the cathode respectively, and a dynamic p–i–n junction is created in situ. Hence, light is emitted from the intrinsic region near the center. Since the polymer blend layer is in direct contact with the ITO electrode, the surface characteristics of ITO and the interfacial properties between the polymer blend layer and the ITO electrode play an important role in the operation of ploymer LECs. In this study, the effect of different modification methods upon the surface properties of ITO electrodes were investigated through the measurements of X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), sheet resistance, contact angle and surface energy. In addition, we studied the influence of surface modifications of the ITO electrodes on the
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
performance of ploymer LECs, in terms of the electrical and optical characteristics of the devices. 2. Experimental details
2103
Table 1 Surface tension parameters (in mJ/m2) of test liquids Liquid
g pl
g dl
gl
H2O CH2I2
51.0 2.3
21.8 48.5
72.8 50.8
2.1. ITO surface modification Two different sets of processing techniques, wet and dry cleanings were used to modify the surface of ITO substrates. They are: a. Solvent cleaning (S1)—The substrates were successive washed in an ultrasonic bath of acetone and alcohol each for 20 min, then rinsed in deionized water, and dried with nitrogen. b. Hydrochloric acid (HCl) treatment (S2)—The substrates were dipped into 10% HCl solution at room temperature for 15 min, and then rinsed in deionized water and finally dried in a flow of nitrogen. c. Oxygen plasma treatment (S3)—The substrates were exposed to the oxygen plasma in a home-made parallel plate plasma generator for 5 min under pressure 20 Pa and voltage 250 V at room temperature. The ITO-coated glass substrates (11 V/sq) used in this study were purchased from CSG Holding Co. Ltd. and cut into 32 mm 32 mm plates. All the ITO substrates were cleaned prior to surface treatment, first by rubbing with detergent, then in ultrasonic bath for 10 min in deionized water, and finally dried in a flow of nitrogen. In this paper, the untreated ITO is referred to as S0.
recorded within 30 s after the formation of the sessile drop. Five readings were taken, and the average was reported unless otherwise indicated. The typical error is around 28 in our present experiment. Distilled water (H2O) and diiodomethane (CH2I2) were chosen as the test liquids [11–14]. Both of them were reagent grade, and their surface tension and surface tension components are listed in Table 1. Based on the measured contact angles, the surface energy was calculated using the Owens–Wendt (OW) approach [11–13]. According to the OW method, the surface energy (gs) of a solid is equal to a sum of the polar (g ps ) and dispersion (g ds ) components: g s ¼ g ps þ g ds
(1)
and the following relation was used: g sl ¼ g s þ g l 2ð
qffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffi g ps g pl þ g ds g dl Þ
where gsl is the interfacial energy between the solid and the test liquid; gl is the surface tension of the liquid; and g pl and g dl are the polar and dispersion components of the surface tension of the liquid, respectively. When combining Eqs. (1) and (2) with the Young’s equation, the following expression can be obtained:
2.2. ITO characterization ð1 þ cos uÞg l ¼ 2ð The XPS measurements of the ITO substrates were carried out in a XSAM-800 spectrometer, using a monochromatic Al Ka (hn = 1486.60 eV) as X-ray source. Typically the operating pressure in the analysis chamber was about 107 Pa. All binding energies were referenced to the binding energy of the carbon C 1s peak at 285.0 eV. For the calculation of the atomic concentrations, a linear background correction was done, the peak areas were corrected with empirical sensitivity factors, the instruments transmission function and the specific mean free path lengths [10]. The ITO surfaces were structurally characterized by AFM with a Seiko Instruments SPA-400 in the contact mode. Measurements were achieved at room temperature in the air, using the same pyramidal Si3N4 tip. The surface roughness of the ITO samples were calculated with the AFM software. The sheet resistance of the ITO substrates were measured by a four-point probe (SZ-82) at room temperature in the air. The contact angle measurements were performed using a JY82 contact angle goniometer by the sessile drop technique [11–14] at 20 8C in an environmental chamber. A substrate was placed on the sample stage of the goniometer, and a microsyringe was used to deposit a liquid drop of 2–3 ml on the surface of the substrate. The steady-state contact angle was
(2)
qffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffi g ps g pl þ g ds g dl Þ
(3)
Here, u is the contact angle between the test liquid and the solid surface. Eq. (3), known as the OW method, is one of the approaches of estimating the solid surface energy. From Eq. (3), surface energy parameters g ps , g ds , gs and polarity xp (xp ¼ g ps =g s ) can be readily achieved from the values of u measured with a pair of test liquids whose surface tension components are known. 2.3. Polymer LECs fabrication Polymer LECs were fabricated using the polymer blend of poly [2-methoxy-5-(20 -ethylhexyloxy)-1,4-phenylenevinylene] (MEH-PPV) and poly(ethylene oxide) (PEO) complexed with lithium triflate (LiTf) as the active light-emitting layer, and the differently treated ITO substrates as device electrodes. The weight ratio of the MEH-PPV:PEO:LiTf in the solution was 12:5:2. The polymer blend film of thickness of about 200 nm was prepared by spin-coating cyclohexanone solution onto the ITO substrate. The silver layer (150 nm) was then deposited on the top of the blend film by the vacuum sublimation technique (about 104 Pa) to form ITO/MEH-PPV + PEO(LiTf)/Ag structure. The active area of each LEC device was 36 mm2.
2104
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
The organic materials were purchased from Aldrich Chemical Company and used without further purification. The polymer LECs fabricated with ITO samples S0, S1, S2, and S3 were referred to as devices D0, D1, D2, and D3, respectively. The bias voltage was applied with the ITO electrode as anode and the Ag electrode as cathode. The current–voltage– luminance (I–V–L) characteristics of the non-encapsulated devices were measured at a scan rate of 0.1 V/s with a KEITHLEY-4200 sourcemeter and a ST-86LA luminance meter. All measurements were performed at room temperature in ambient air. 3. Results and discussion 3.1. Modification effect on the ITO surface properties Table 2 summarizes the chemical composition, sheet resistance and surface roughness of the untreated and treated ITO substrates. From Table 2, We first note that the presence of
Table 2 Chemical composition, sheet resistance and surface roughness of the untreated and treated ITO substrates Substrate
S0 S1 S2 S3
Chemical composition (at.%) [C]
[O]
[In]
[Sn]
Sheet resistance Rs (V/sq)
5.6 9.2 4.9 4.6
15.1 14.9 13.6 16.5
68.8 67.5 72.5 68.5
10.5 8.4 9.0 10.4
11.0 10.9 14.7 10.5
Surface roughness Ra (nm) 3.2 3.1 6.4 2.3
carbon element, besides containing the elements of O, In and Sn on each ITO sample, indicating that carbon is the only major contaminant. Note also that the [C]:[O] ratios of substrates S0, S1, S2, and S3 are 5.6%:15.1%, 9.2%:14.9%, 4.9%:13.6%, and 4.6%:16.5%, respectively. Apparently, the oxygen plasma treatment largely decreases the carbon content and increases the oxygen content, and thereby more effectively enhances the work function of the ITO, since the ITO work function is
Fig. 1. AFM images of the untreated (a); solvent cleaned (b); hydrochloric acid treated (c); and oxygen plasma treated ITO substrates (d).
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
determined by the oxygen content and the carbon content [15–18]. Secondly, we observe that the sheet resistance (Rs) of substrates S0, S1, S2, and S3 are 11.0, 10.9, 14.7, and 10.5 V/sq, respectively. The oxygen palsma treatment yields the lowest value of Rs (10.5 V/sq), while HCl treatment gives the highest value of Rs (14.7 V/sq). Thirdly, the surface roughness of ITO substrates can be observed to decrease in the sequence of S2, S0, S1, S3. The average roughness (Ra) values of substrates S2, S0, S1, and S3 are 6.4, 3.2, 3.1, and 2.3 nm, respectively. Meanwhile, the sample S3 exhibits more smooth and homogenous surface, compared with the other ITO substrates (see Fig. 1). Fig. 2(a) shows the measured contact angle of the untreated and treated ITO substrates. Using the OW approach, the surface energy and polarity were derived from the measured contact angles and are plotted in Fig. 2(b). The errors in the surface energy and polarity were calculated out to be less than 1.9 mJ/m2 and 0.03, respectively. From Fig. 2(a), we find that the measured contact angle is subjected to the treatment methods of ITO substrates. Note that the water contact angle uH2 O of substrates S0, S1, S2, and S3 are 85.18, 64.58, 82.08, and 30.58 respectively, reducing in the order of S0, S2, S1, S3. Obviously, oxygen plasma brings about a large decrease in the measured water contact angle uH2 O compared to the unterated ITO, from 85.18 to 30.58, indicating that the oxygen plasma treated ITO surface appears to be highly polar. Note also that the substrate S3 exhibits the smallest diiodomethane contact angle uH2 O , followed by the substrates S1, S0, and S2. The oxygen plasma treatment gives the minimum diiodomethane contact angle uCH2 I2 (32.18). Fig. 2(b) indicates that the surface energies gs and the corresponding components (g ps , g ds ), and polarities xp are strongly dependent on the surface modifications. For the treated substrates S1, S2 and S3, both gs and xp more or less increase from 36.5 mJ/m2 and 0.08 to 36.8– 64.7 mJ/m2 and 0.12–0.54 respectively, in comparison with the untreated substrate S0. In particular, we note that oxygen plasma yields the highest surface energy (64.7 mJ/m2) and the highest surface polarity (0.54), mainly due to the significantly enhanced polar component g ps (35.1 mJ/m2). The increase of the surface energy and polarity may be attributed to two reasons. One is that oxygen plasma can remove effectively the hydrocarbon contaminant on the ITO surface. Since hydrocarbons have low surface energy, the removal of hydrocarbons from the surface can lead to an increase in the surface energy. In addition, to the removal of hydrocarbon contaminant from the surface, oxide bonds can form on the substrate surface during oxygen plasma treatment, which leads the surface to be more hydrophilic and shows higher surface energy. Another is that the molecular water monolayer chemisorbed exists on the ITO surface for the oxygen plasma treatment [19]. And it is the presence of monolayer adsorbate coverage of water that increases the water concentration. Therefore, accordingly the polarity increases significantly. These results indicate that oxygen plasma treatment not only decreases the sheet resistance and surface roughness, but also enhances the work funtion, surface energy and polarity, and hence improves the the surface properties of the ITO substrate.
2105
Fig. 2. (a) Contact angle and (b) surface energy and polarity of the untreated and treated ITO substrates.
3.2. Modification effect on the LECs optoelectrical properties Fig. 3(a) gives the absorption spectrum of a 100 nm MEHPPV film measured by a Perkin-Elmer Lambda 900 UV7VIS/ NIR spectrophotometer at room temperature. As can be seen, the absorption spectrum of the thin film has a relatively sharp peak at 500 nm, with an onset of absorption at 570 nm. Based on the measured absorbance (A) spectrum, the absorption coefficient (a) was calculated using the following formula [20–22]: a ¼ 2:303
A t
(4)
where t is the thickness of the thin film. The energy gap (Eg) can be obtained by fitting the optical absorption coefficient a to Tauc’s relation [23–26]: ðahnÞ2 ¼ Cðhn Eg Þ
(5)
Here, C is an energy-independent constant, hn is the photon energy. Fig. 3(b) shows the Tauc plot: (ahn)2 versus hn for the
2106
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
Fig. 3. (a) Absorption spectrum and (b) Tauc plot of electroluminescent polymer MEH-PPV thin film. Dotted line indicates the linear part of the curve.
Fig. 4. (a) Current–voltage and (b) luminance–voltage curves of LECs fabriI L cated with the untreated and treated ITO substrates. Von and Von present the turnon voltages of current injection and light emission, respectively.
MEH-PPV film. The value of Eg can be readily obtained by extrapolation of the linear portion of the graph to (ahn)2 = 0, which is about 2.18 eV. Fig. 4 illustrates the semilogarithmic plots of the measured V–I and L–V characteristics for the devices fabricated with the untreated and treated ITO electrodes. As shown in Fig. 4(a), the I turn-on voltage of current injection (Von ) for all the devices share almost the same value (about 2.2 V), close to the measured Eg l value of MEH-PPV (i.e., Von =Eg/e, e is the electronic charge). Note from Fig. 4(b) that the turn-on voltage of light emission L (Von ) for the devices D0, D1, D2, and D3 are 2.7, 2.6, 2.9, and L I 2.4 V, respectively, and the difference between Von and Von exists L among all the devices. Each Von is observed to be slightly higher I than its corresponding Von , indicating that the turn-on voltage of light emission lags behind that of current injection. Similar results have been reported by Yang and Pei [27], Ko et al. [28], Chuai et al. [29]. There may be two reasons for this voltage lag of light emission. One is related to the sensitivity of the light intensity detector which is not sensitive enough for the weak
L value. Another reason is luminance at the voltage lower than Von the relatively slow response of the LECs, which also make the light emission lag behind the current injection. Note also from Fig. 4 that the injection current (I), luminance (L) and efficiency (h) of the LECs depend on the ITO electrode modifications. At V = 7 V for devices D0, D1, D2, and D3, the values of I are 27.1, 29.2, 21.4, and 48.8 mA; and L are 25.5, 40.3, 6.9, and 98.2 cd/ m2, respectively. Then the h values of devices D0, D1, D2, and D3
Table 3 Device performance of LECs fabricated with the untreated and treated ITO substrates Device
D0 D1 D2 D3
I (V) Von
2.2 2.2 2.2 2.2
L Von (V)
2.7 2.6 2.9 2.4
V=7V I (mA)
L (cd/m2)
h (cd/A)
27.1 29.2 21.4 48.8
25.5 40.3 6.9 98.2
0.034 0.050 0.012 0.073
Z.Z. You, J.Y. Dong / Applied Surface Science 253 (2006) 2102–2107
2107
can be readily calculated out to be about 0.034, 0.050, 0.012, and 0.073 cd/A, respectively (see Table 3). Clearly, the oxygen plasma-treated device in our present work yields the higher injection current, luminance and efficiency, exhibiting the relatively better performance, while the hydrochloric acidtreated device the poorer. Such an experimental result suggests that the device performance of LECs is closely related to the characteristics of electrical contact at the ITO/polymer interface. Oxygen plasma treatment results in not only higher work function, lower sheet resistance, smoother surface, but also higher surface energy and polarity of the ITO electrode which is highly important to improve the surface morphological properties of polymer thin film and to optimize the interface characteristics of ITO/polymer. Firstly, the increase of surface polarity of ITO electrode leads to a more uniform spreading of the polymer solution on the ITO surface, reduces the voids and pinholes, and thereby obtains the homogeneous and compact polymer thin film. Secondly, the increase of the surface energy of ITO electrode provides a better adhesion of the polymer film and reduces the interfacial energy between the polymer film and ITO electrode, and therefore increases the effective area available between these two materials. These two effects above-stated can significantly improve the interface formation and electrical contact of the ITO electrode with the polymer blend in the LECs, so as to facilitate the electrochemical doping of the electroluminescent polymer when sufficiently large bias was applied, and therefore yield the relatively better performance of the LECs. Contrary to the situation in the oxygen plasma treatment, yet, hydrochloric acid treatment results in a poor electrical contact to the emissive layer in the LECs and hinders charge injection through the electrochemical doping, and thereby produces poorer performance of the LECs.
degree of influence upon the injection current, luminance and efficiency, but hardly upon the turn-on voltages of current injection and light emission. Oxygen plasma treatment on the ITO electrode gives the better performance of the LECs, due to the improvement of interface formation and electrical contact between the ITO electrode and the polymer blend in the LECs.
4. Conclusion
[18] [19]
The effect of different modification methods on the surface properties of ITO electrodes were investigated by XPS, AFM, sheet resistance, contact angle and surface energy measurements. We observe that the surface properties of the ITO substrates are closely relevant to the modification methods, and the oxygen plasma is more efficient compared with the other treatments since it brings about smoother surface, lower sheet resistance, higher work function, higher surface energy and polarity of the ITO electrode. Furthermore, the influence of surface modifications of the ITO electrodes on the characteristics of polymer LECs were studied by the electrical and optical properties of the devices. Experimental results indicate that the surface modifications on ITO electrodes have a certain
Acknowledgment This work was financially supported by the National Natural Science Foundation of China (No. 60372002). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
[20] [21] [22] [23] [24] [25] [26] [27] [28] [29]
Q. Pei, G. Yu, C. Zhang, Y. Yang, A.J. Heeger, Science 269 (1995) 1086. Y. Yang, Q. Pei, Appl. Phys. Lett. 70 (1997) 1926. G. Yu, Q. Pei, A.J. Heeger, Appl. Phys. Lett. 70 (1997) 934. Y. Li, J. Gao, G. Yu, Y. Cao, A.J. Heeger, Chem. Phys. Lett. 287 (1998) 83. Y. Li, Y. Cao, J. Gao, D. Wang, G. Yu, A.J. Heeger, Synth. Met. 99 (1999) 243. F.-C. Chen, Y. Yang, Appl. Phys. Lett. 81 (2002) 4278. Q. Sun, H. Wang, C. Yang, G. He, Y. Li, Synth. Met. 128 (2002) 161. J.-C. Lepreˆtre, A. Deronzier, O. Ste´phan, Synth. Met. 131 (2002) 175. Q. Sun, X. Zhan, B. Zhang, C. Yang, Y. Liu, Y. Li, D. Zhu, Polym. Adv. Technol. 15 (2004) 70. D. Briggs, M.P. Seah, Practical Surface Analysis by Auger and X-ray Photoelectron Spectroscopy, Wiley, New York, 1983. D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (1969) 1741. R.J. Good, J. Adhes. Tech. 6 (1992) 1269. S. Wu, Polymer Interface and Adhesion, Marcel Dekker, New York, 1982. B. Janczuk, A. Zdziennicka, W. Wo´jcik, Appl. Surf. Sci. 120 (1997) 35. C.C. Wu, C.I. Wu, J.C. Sturm, A. Kahn, Appl. Phys. Lett. 70 (1997) 1348. J.S. Kim, M. Granstro¨m, R.H. Friend, N. Johansson, W.R. Salaneck, R. Daik, W.J. Feast, F. Cacialli, J. Appl. Phys. 84 (1998) 6859. M. Ishii, T. Mori, H. Fujikawa, S. Tokito, Y. Taga, J. Lumin. 87–89 (2000) 1165. Z.Z. You, J.Y. Dong, Appl. Surf. Sci. 249 (2005) 271. J.S. Kim, P.K.H. Ho, D.S. Thomas, R.H. Friend, F. Cacialli, G.W. Bao, S.F.Y. Li, Chem. Phys. Lett. 315 (1999) 307. M.-B. Park, N.-H. Cho, Appl. Surf. Sci. 190 (2002) 151. M.J. Alam, D.C. Cameron, Surf. Coat. Technol. 142–144 (2001) 776. A. Abu EL-Fadl, Galal A. Mohamad, A.B. Abd El-Moiz, M. Rashad, Phys. B: Condens. Matter. 366 (2005) 44. J. Tauc, R. Grigorovici, A. Vancu, Phys. Status Solidi 15 (1966) 627. Y. Ramadin, M. Al-Haj Abdallah, M. Ahmad, A. Zihlif, S.K.J. Al-Ani, S.G.K. Al-Ani, Opt. Mater. 5 (1996) 69. C. Baban, G.I. Rusu, Appl. Surf. Sci. 211 (2003) 6. T.S. Moss, G.J. Burrell, B. Ellis, Semiconductor Opto-Electronics, Butterworths, London, 1973. Y. Yang, Q. Pei, Appl. Phys. Lett. 68 (1996) 2708. H.C. Ko, D.K. Lim, S.-H. Kim, W. Choi, H. Lee, Synth. Met. 144 (2004) 177. Y. Chuai, D.N. Lee, C. Zhen, J.H. Min, B.H. Kim, D. Zou, Synth. Met. 145 (2004) 259.