Thin Solid Films 518 (2010) 4615–4618 Contents lists available at ScienceDirect Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f On the structural and electrical characteristics of zinc oxide thin films Petronela Prepelita a,1, R. Medianu a, F. Garoi a,⁎, N. Stefan a, Felicia Iacomi b a b National Institute for Lasers, Plasma and Radiation Physics, Atomistilor 409, PO Box MG-36, Magurele 077125, Ilfov, Romania Al.I. Cuza University, 11 Carol I Bd., 700506 Iasi, Romania a r t i c l e i n f o Available online 7 January 2010 Keywords: Zinc oxide Thin films Rf magnetron sputtering Electrical properties a b s t r a c t ZnO thin films with thickness d = 100 nm were deposited by radio frequency magnetron sputtering onto glass substrate from different targets. The structural analyses of the films indicate they are polycrystalline and have a wurtzite (hexagonal) structure. Crystallites are preferentially oriented with (002) plane parallel to the substrate surface and the samples have low values for surface roughness, between 1.7 nm and 2.7 nm. The mechanism of electrical conduction in the studied films is strongly influenced by this polycrystalline structure and we used Van der Pauw method to analyze these properties. Electrical studies indicate that the ZnO thin films are n-type. For the cooling process, thermal activation energy of electrical conduction of the samples can vary from 1.22 eV to 1.07 eV (for the ZnO layer obtained from for metallic Zn target) and from 0.90 eV to 0.63 eV (for the ZnO layer obtained from ZnO target), respectively. The influence of deposition arrangement and oxidation conditions on the structural and electrical properties of the ZnO films was investigated in detail. © 2009 Elsevier B.V. All rights reserved. 1. Introduction ZnO thin films are intensively studied also due to their interesting properties such as high transmission coefficient, large energy band gap, high photoconductivity, etc [1–3]. An important factor that influences the properties of these films is the deviation from their stoichiometric composition. Because it has a large forbidden bandwidth (3.2 eV), ZnO is a multifunctional material shown in some papers [3–6]. Different physical and chemical techniques, such as magnetron sputtering [7,8], thermal evaporation [1,9], chemical vapor deposition [5], pulsed laser deposition [2], and sol–gel deposition [3] allow for ZnO thin films to be prepared. A safe, low cost and simple method for obtaining thin films of pure ZnO is the radio frequency (rf) magnetron sputtering technique. Because of these priorities, the researchers have paid attention to depositing the ZnO thin films obtained from ZnO [10] and Zn [11] targets (99.99% purity). The growing interest for the study of structural and electrical properties of ZnO films is due to the increased usage of these films in producing ultra-acoustic and acoustic–optical materials as well as a compound material [4,5]. In this paper we emphasized our analysis on the influence of the deposition conditions, respectively type of the used target (e.g. Zn or ⁎ Corresponding author. Atomistilor 409, PO Box MG-36, 077125, Ilfov, Romania. Tel.: +40 21 4574467; fax: +40 21 4574467. E-mail address: florin.garoi@inflpr.ro (F. Garoi). 1 Married as Petronela Garoi. 0040-6090/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2009.12.044 ZnO) on the surface morphology and the results of electrical characterization of ZnO thin films grown by magnetron sputtering technique. Physical properties of the samples were analyzed taking into account structural and electrical characteristics of the ZnO thin films. These characteristics were found by substrate surface analyses, by Hall effect measurements and respectively from the temperature dependence of the electrical conductivity of the ZnO films [7,12,13]. 2. Experimental The ZnO thin films were deposition by rf magnetron sputtering [7,8] onto 2.25 cm2 glass substrates. This modern technique under vacuum uses argon as the working gas in a VARIAN ER 3119 unit that allows controlling the thickness of the thin films in situ with a piezoceramic. ZnO thin films were obtained from high purity single crystals of metallic Zn and ZnO targets, due to oxygen pressure existing into the residual atmosphere. We used the following deposition conditions: disk targets of Zn and ZnO with 99.99% purity, power forward/power reflected: 71 W/ 64 W (for Zn target); power forward/power reflected: 70 W/4 W (for ZnO target); a deposition rate of about 0.5 Å/s (for Zn target) and a deposition rate of about 0.9 Å/s (for ZnO target). A Zeiss Microscope (Axio Imager model) was used to visualize and record images of our rf magnetron sputtering deposited thin films. The structure of the films was fully characterized using X-ray diffraction technique (XRD), X-ray photon spectroscopy (XPS), and atomic force microscopy. For surface morphology studies by AFM both a XE-100 Park and a Quesant Instrument Corporation-type Nomad atomic force microscope 4616 P. Prepelita et al. / Thin Solid Films 518 (2010) 4615–4618 were used. A Bruker D8 X-ray Grazing Incidence Diffractometer was used for XRD measurements. Transport properties of the samples were studied by Hall effect measurements using the Van der Pauw configuration. Four silver paste contacts were deposited in the corresponding four corners of each sample. With the Hall measuring system (Model 7604, Lake Shore) we were able to determine Hall coefficient values (RH), Hall mobility (µH) and the concentration of charge carriers (n). Transmission spectra were recorded (in the range 300–1300 nm) using an UV–VIS spectrophotometer model Cintra 10e. The effect of using different targets and the influence of deposition conditions on surface morphology and electrical properties of ZnO thin films was investigated. 3. Results Images of the films are shown in Fig. 1. We imaged the area between the substrate and the deposited film. The microscope was set to work in reflection mode, bright field regime and 50× magnification. Please note (Fig. 1b) that the ZnO film deposited from metallic Zn target is characterized by an irregular surface, clearly showing some defects in the foil at the micrometer scale. The structure of ZnO thin films was investigated by XRD, in the angular range 30°–70°, using CuKα1 radiation. These measurements offered us the information about the type of crystalline structure (e.g. amorphous or polycrystalline) of the films, the crystalline phases, the crystallinity layer, the crystalline orientation planes, etc. Diffractograms for two samples of ZnO representative layers are shown in Fig. 2a and b. They have the same thickness (100 nm) and are obtained from different targets (Zn and ZnO). Diffractogram of the ZnO thin film obtained from metallic Zn target underlines the idea that the presence of oxygen in the residual atmosphere determined the full oxidation of metallic Zn resulting in a transparent ZnO film with polycrystalline structure. The presence of two peaks of diffraction in each diffractogram can be observed (Fig. 2a and b). With the use of ASTM files [14], these peaks were identified to have maximum values corresponding to the planes (002) and (103). Thus, these ZnO thin films have a polycrystalline wurtzite type structure, with the lattice parameters: a = 3.248 Å and c = 5.211 Å. The existence of a diffraction peak at 34.41° shows that, in these films, the crystallites exhibit a preferential orientation with (002) planes parallel with the substrate (e.g. the hexagonal c-axis is normal to the substrate). Another diffraction maximum is at 62.84°, corresponding to the reflection on planes (103) of the ZnO hexagonal phase. Fig. 2. X-ray diffraction pattern for ZnO thin films prepared from: a) ZnO target, b) Zn target. The ZnO sample, obtained from metallic Zn target, shows a significant decrease of the intensity peaks corresponding to (002) and (103) planes as compared to the peaks of the sample obtained from ZnO target (Fig. 2a). We believe that this behaviour is also explained by diminution of crystallites orientation and by the crystallinity degree of the ZnO film obtained from metallic Zn target. The interplanar distance for hexagonal lattice is given by [15] " # 1 4 h2 + hk + k2 l2 ⋅ = + ; 3 d2hkl a2 c2 ð1Þ where h, k, l are the Miller indices, a and c are the hexagonal lattice parameters (see Table 1). Experimentally, with the help of XRD diffractograms, we established that both ZnO layers have a hexagonal structure. However, a ZnO thin film with a stable structure and strong crystallite orientation can be achieved if a ZnO target is used. The conclusion that the Zn phase is not detected in any XRD pattern for ZnO thin film obtained from Zn target is underlined by XPS [16], another modern technique of investigation of the thin films structure. This analysis also allows determining the elemental composition as well as chemical and electronic state of the elements that exist within a material. In these measurements we studied the Zn 2p3 and O 1s levels from ZnO sample. Fig. 1. ZnO thin films deposited from two different targets (a) from ZnO target; (b) from metallic Zn target. Images taken with a Zeiss Axio Imager optical microscope at 50× magnification. P. Prepelita et al. / Thin Solid Films 518 (2010) 4615–4618 Table 1 The unit cell parameters of ZnO thin films calculated from XRD patterns. Sample Target da (nm) (hkl)b 2θc (deg.) dhkld (Å) ae (Å) ce (Å) ZnO/glass Zn 100 ZnO/glass ZnO 100 a b c d e f 002 103 002 103 34.41 62.84 34.40 62.83 2.604 1.478 2.606 1.478 3.250 3.246 Df (nm) 5.209 12 23 5.212 4 9 Film thickness. Miller indices of the planes (h denotes the hexagonal lattice). Bragg angle. Interplanar spacing. Unit cell parameters. Average crystallite size calculated from Debye–Scherrer formula [15]. The binding energies of Zn 2p3 and O 1 s were found to be between 531.9–532.1 eV for O 1s and between 1021.7–1021.9 eV for Zn 2p3. The experimental data revealed that there is a good stoichiometry for this film. We determined that these films, deposited both directly from metallic Zn and ZnO targets, are characterized by low surface roughness: Rrms = 1.7 nm for metallic Zn target and Rrms = 2.7 nm for ZnO target. AFM images in Fig. 3a and b revealed distinct characteristics for these ZnO thin films, though. The morphology of ZnO thin film obtained from ZnO target shows a better uniform granulation arrangement, along or interlaid into the crystalline lattice, than ZnO thin film obtained from Zn target. Also AFM analyses of the surface morphology indicate the surface of ZnO thin film obtained from ZnO target is smoother and more homogeneous than ZnO thin film obtained from Zn target. Clearly, the mechanism of electrical conduction in the studied film is strongly influenced by its polycrystalline structure [5]. It is well known that measurements of resistivity and Hall coefficient at different magnetic fields have an important role in the research of electric properties of semiconductor compounds. Ohmic contact was subsequently used for Hall measurements using the Van der Pauw [17,18] approach to investigate the electrical properties of the deposited ZnO sample. For studied samples the Hall coefficient was found to be negative, indicating the predominance of the electrons as majority charge carriers [5]. We noticed that RH is negative under positive applied magnetic field while it is positive otherwise. This means that the carriers in ZnO are electrons, which in agreement with several reports on ZnO [2,5]. By identification of the directions of longitudinal electric field, magnetic field (normal to the sample) and Hall electric field (transversal), we have come to the conclusion that the major electrical charge carriers are electrons and, consequently, the studied samples are type-n semiconductors [3,6,9]. The n-type carriers concentrations calculated for the ZnO samples using values of the Hall coefficient do not depend on the magnetic flux density but they increase with the electric current intensity used 4617 along the samples. The values of carrier concentrations are in the range 0.10–4.05 × 1017 cm− 2 for ZnO film obtained from Zn target and 0.25–3.35 × 1017 cm− 2 for ZnO film obtained from ZnO target (Fig. 4). The study of heating–cooling processes between 295 and 425 K can be observed in the conductivity/temperature dependencies of the analyzed ZnO samples. From Fig. 5 it can be observed that during heating the electrical conductivity is increasing, following a typical semiconductor behaviour. In this case and most precisely in the 400–425 K temperature range, the activation energy determined from conductivity/temperature dependency, is equal with the potential barrier heights between the crystallites [19]. From the slope of ln σ = f(103/T) dependency we were able to determine the values for thermic activation energy of the electrical conductivity Ea. The decrease of ZnO electrical resistivity with the increase of temperature is due to the improvement of crystallinity degree of the ZnO layers. This behaviour in the 295–425 K temperature range indicates a semiconductor-like behaviour. During the cooling process from 425 K to 295 K the potential barrier decreases from 1.22 eV to 1.07 eV and from 0.90 eV to 0.63 eV for the ZnO layer obtained from metallic Zn target and ZnO target, respectively. Thus, the levels of defects at the boundaries of grains decrease and the layers' crystallinity is improved. A better crystallinity may increase the mobility of carriers within ZnO layers due to oxygen desorbtion at grain's boundary in the case of high temperature. This leads to the disappearance of the acceptor states of oxygen at grain's boundary, which are in fact traps for electrons. By studying the optical characteristics (e.g. transmission and absorption spectra) of ZnO thin films we could obtain information on the energy band structure, energy levels of impurities, characteristics of interband transitions, etc. The shape of the absorption edge is determined by the characteristics of optical interband transitions. We know [20] that the fundamental absorption is due to allowed direct band-to-band transitions and it is described by the expression [21]: 1= α⋅hν = Aa ðhν−Eg Þ 2 ; ð2Þ where hν is the incident photon energy, Eg denotes the optical band gap and Aa is a characteristic parameter (independent of photon energy) for respective transitions. The analysis of our experimental data shows that the studied samples are characterized by allowed direct band-to-band transitions. This fact is confirmed by plotting (αhν)2 versus hν (Fig. 6). According to Eq. (2), (αhν)2 = f(hν) dependence is linear indeed [22]. The values for the optical band gap, Eg, were determined by extrapolating the linear portions of (αhν)2 =f(hν) curves to (αhν) = 0. Fig. 3. 3D AFM image (5 µm × 5 µm) for a ZnO thin film (d = 100 nm) deposited: a) from ZnO target, Rrms = 2.7 nm; b) from metallic Zn target, Rrms = 1.7 nm. 4618 P. Prepelita et al. / Thin Solid Films 518 (2010) 4615–4618 Fig. 4. Concentration of charge carriers dependence on the magnetic inductance in ZnO samples, obtained from two different targets. For ZnO thin film obtained from Zn target the value of optical band gap (Eg = 3.25 eV) is slightly lower and can be attributed to a greater density of the donor states near a conduction band determined by the small grain size of the film. 4. Conclusions and discussion In this manuscript we studied the structural and electrical properties of rf magnetron sputtering ZnO thin films. ZnO and metallic Zn were used as targets to obtain our ZnO thin films. It was found that the films are polycrystalline and have a wurtzite (hexagonal) structure. The film crystallites are oriented with the (002) planes parallel to the substrate. Modern methods to investigate the structure of the obtained ZnO thin films were applied. It was found that for ZnO target the morphology of ZnO thin film shows a uniform granulation arrangement, along or interlaid into the crystalline lattice, compared with ZnO thin film obtained from Zn target. The polycrystalline structure of the films plays in important role in the mechanism of electrical conduction. We have noticed that when Fig. 5. The temperature dependence of electrical conductivity during heating–cooling cycle for ZnO samples obtained by different targets. Fig. 6. Spectral dependence of the absorption coefficient, (αhν)2 = f(hν) for ZnO samples deposited from two different targets: a) ZnO target; b) metallic Zn target. increasing the temperature of both ZnO layers (e.g. obtained from metallic Zn and from ZnO), the decreasing resistivity determines an increase of both oxygen vacancies and charge carriers concentration. The experimental data of Hall analysis, using the Van der Pauw method showed that the ZnO thin films were n-type. The optical band gap determined from absorption spectra (supposing allowed direct band-to-band transitions) varied between 3.28 eV and 3.25 eV. References [1] M. Rusu, G.G. Rusu, M. Girtan, S. Dabos Seignon, J. Non-Cryst. Solids 354 (2008) 4461. [2] C. Ristoscu, D. Caiteanu, G. Prodan, G. Socol, S. Grigorescu, E. Axente, N. Stefan, V. Ciupina, G. Aldica, I.N. Mihailescu, Appl. Surf. Sci. 253 (2007) 6499. [3] K. Yoshino, T. Hata, T. Kakeno, H. Komaki, M. Yoneta, Y. Akaki, T. Ikari, Phys. Stat. Sol. (c) 0 (2) (2003) 626. [4] Fawzy A. Mahnoud, G. Kiriakidis, J. Ovonic Res. 5 (1) (2009) 15. [5] L. Olvera, A. Maldonado, R. Asomoza, S. Tirado-Guerra, Thin Solid Films 411 (2002) 198. [6] P. Prepelita, C. Baban, F. Iacomi, J. Optoelectron. Adv. Mater. 9 (2007) 2166. [7] N. Ekem, S. Korkmaz, S. Pat, M.Z. Balbag, E.N. Cetin, M. Ozmumcu, Int. J. Hydrogen Energy 34 (12) (2009) 5218. [8] M.K. Jayaraj, A. Antony, M. Ramachandran, Bull. Mater. Sci. 25 (3) (2002) 227. [9] E. Budianu, M. Purica, F. Iacomi, C. Baban, P. Prepelita, E. Manea, Thin Solid Films 516 (2008) 1629. [10] T. Yamamoto, T. Shiosaki, A. Kawabta, J. Appl. Phys. 51 (6) (1980) 3113. [11] B.T. Khuri Yakub, J.G. Smits, T. Barbee, J. Appl. Phys. (1981) 4772. [12] K.T. Roro, G.H. Kassier, J.K. Dangbegnon, S. Sivaraya, J.E. Westraadt, J.H. Neethling, A.W.R. Leitch, J.R. Botha, Semicond. Sci. Technol. 23 (2008) 055021. [13] Sol E. Harrison, Phys. Rev. 93 (1) (1954) 52. [14] ASTM X-ray powder data file, Card no. 5-0664. [15] B.D. Cullity, Elements of X-ray Difraction, 2nd Edition, Addison-Wesley, Reading, MA, 1978. [16] J. Daugėla, S. Joneliūnas, A. Jotautis, R. Naujokaitis, E. Klovaitė, Elektron Elecktrotech. 7 (63) (2005) 1215. [17] L.J. Van der Pauw, Philips Res. Rep. 13 (1958) 1. [18] Oliver Bierwagen, Tommy Ive, Chris G. Van de Walle, James S. Speck, Appl. Phys. Lett. 93 (2008) 242108. [19] H.F. Wolf, Semiconductors, Wiley-Interscience, New York, 1971. [20] M.R. Islam, J. Podder, Cryst. Res. Technol. 44 (3) (2009) 286. [21] J.N. Hodgson, Optical Absorbtion and Dispersion in Solids, Chapman & Hall, London, 1970. [22] V. Srikant, D.R. Clarke, J. Appl. Phys. 81 (9) (1997) 6357.