The 2012 World Congress on Advances in Civil, Environmental, and Materials Research (ACEM’ 12)Seoul, Korea, August 26-30, 2012

Synthesis and Characterization of Zn-doped and un-doped SnO2 nanostructures for PV Applications

Gul Jabeen1), Saira Riaz1,2)*, and Shahzad Naseem1) 1) Centre of Excellence in Solid State Physics, Punjab University, Lahore 54590,

Pakistan 2) saira_cssp@yahoo.com

ABSTRACT

Tin oxide nanostructures, un-doped and Zn-doped, have been prepared by conventional sol-gel technique. Research grade tin chloride (SnCl2.2H2O) and zinc acetate (Zn(O2C2H3)2) were used as tin and zinc sources. Ethylene glycol and alcohol were used as solvents under different synthesis conditions. Transparent un-doped and Zn-doped SnO2 sol was prepared after continuous stirring and refluxing at 40oC. Thin films, prepared from SnO2 sol, have also been deposited on glass substrate by spin coating at 500 rpm for 10 seconds and then at 3000rpm for 30 seconds. Scanning electron microscopy results show the formation of nanodiamonds (~40nm) and nanorods (~30nm) after annealing at 300oC for 60 minutes. It is important to mention here that we have prepared SnO2 nanostructures without the use of any surfactant and ligand during sol synthesis. X-ray diffraction (XRD) patterns show that the preferred orientation was (111) with orthorhombic crystal structure. Crystallite size was calculated to be ~8 nm from XRD results. Optical properties measured by using spectroscopic ellipsometer have shown a band gap of 3.87 eV, which is larger than the band gap of SnO2 in bulk form (3.6eV). More than 90% transmission was observed in these sol-gel deposited SnO2 thin films. Electrical properties were studied using four-probe method and Hall effect. Thin films show room temperature conductivity of the order of 103 Ω-1cm-1. These optical and electrical values are comparable to values reported in literature. Zn-doped SnO2 films show conductivity slightly better than the undoped one. 1. INTRODUCTION

Due to their unique electronic, optical and mechanical properties (Zheng 2001) one dimension (1D) nanoscale materials have gained many scientist’s attention. At this structural level, nanosized semi-conducting oxide materials show the properties such as super-conductivity, ferro-electricity and magnetism etc. covering all the aspects of material science. Conductive oxides are promising materials for the manufacture of transparent electronics and optoelectronic devices in the ultraviolet (UV) region due to their wide band gap and high mobility. In the fast growing field of nano-technology, there is an urgent need of specially designed semiconductors to better match the properties of promising materials. Thus, it has led to a renewed interest in ternary oxide semiconductors (II–IV–VI oxides) of the form A2BO4, such as cadmium stannate

(Cd2SnO4), generally known as cadmium tin oxide (CTO), and zinc stannate (Zn2SnO4), which is mostly called zinc tin oxide (ZTO). Zn2SnO4 is recognized for having high electron mobility (Tsaroucha 2011, Erslev 2008), high electrical conductivity, and desirable optical properties that make it suitable for an extensive range of applications such as in solar cells (Li 2012, Lana-Villarreal 2007), negative electrode material for Li-ion battery (Kim 2012, Srivastava 2007), sensors for the detection of humidity and various combustible gases (Yadav 2012), and as a photocatalyst for the degradation of organic pollutants (Cun 2002). Compared with binary oxides, the complex ternary oxides like Zn2SnO4 are chemically more stable making them ideal for applications involving extreme conditions (Zhang 2007). Zn2SnO4 has been successfully used as a flame retardant and smoke suppressant (Horrocks 2012). The control of sizes and shapes of nanostructures is crucial as it may affect their electrical and optical properties (Lim 2011, Baruah 2009). For controlling the size of particles we have used sol-gel method. There are many advantages of sol-gel method such as it allows the production of new hybrid organic-inorganic materials, which do not exist in nature. Furthermore, it is cost-effective and large-area scalable technique. In this paper, we report on the sol-gel prepared un-doped and ZnO doped SnO2. Stannous chloride hydrate (SnCl2.2H2O) was used as a precursor. The structural and morphological characterization was carried out by XRD, SEM and AFM. Effect of doping and synthesis parameters on the optical properties were also studied.

2. EXPERIMENTAL DETAILS

a. Synthesis of undoped and Zn-doped SnO2 sol: In this technique tin chloride hydrate SnCl2.2H2O was used as starting material. A solution of 0.1 mol % SnCl2 in ethylene glycol was prepared by dissolving appropriate amounts of SnCl2 under vigorous stirring at 60⁰C until colorless and transparent sol was obtained. Before gelation process thin films were also prepared on glass substrate by spin coating. Samples were dried at room temperature for 24h and then annealed at 300⁰C for 1h. For doping purposes ZnO sol was synthesized by following the route mentioned in our previous paper (Riaz 2011). ZnO doped SnO2 sol was prepared under the same conditions used for un-doped SnO2 sol with Zn/Sn/O molar ratio of 2:1:2. Thin films of SnO2 and ZnO doped SnO2 were deposited with varying thicknesses. Doped films were also annealed at 300˚C to obtain stable Zn2SnO4 structure.

b. Characterization:

Crystallographic structure along with the phase variation and crystallite size of undoped and doped SnO2 samples was studied by Rigaku D-MAX/IIA X-ray diffractometer (XRD). CuKα (Ni filtered) radiations (λ=1.5405 Å) were used to obtain the XRD pattern. Surface morphology was observed using Hitachi S-3400N scanning electron microscope (SEM). Bruker CP-II atomic force microscopy (AFM) was used to characterize the surface nanostructure. Optical properties in the UV-VIS-IR range were obtained by using JA Woollam’s variable angle spectroscopic ellipsometer (VASE).

3. RESULTS AND DISCUSSION

Fig. 1 shows XRD pattern of undoped SnO2 thin films. The films were deposited with varying thickness of 80-200nm in order to study the effect of variation in thickness on structural and optical properties at nanoscale. All the peaks are well consistent with JCPDS card no. (29-1484), which confirmed the samples as pure orthorombic crystalline SnO2. Moreover, it can be seen that with increased thickness the FWHM of the (1 1 1) peak increased indicating a decrease in crystallite size.

Fig. 1 XRD pattern of undoped SnO2 thin films within the thickness range of 80-200nm

XRD plot of as deposited samples confirm the crystalline structure of SnO2 without any heat treatment. To obtain more quantitative information, the XRD pattern was analyzed with Gaussian function where full widths at half maxima [FWHM] were determined. The grain size of SnO2 thin film (D) can be estimated by the Debye-Scherrer formula.

D = 0.9 λ / B cosθ (1)

Where, D = Crystalline grain size, B = FWHM of the diffracted peak, λ = wavelength of the X-rays, θ = Angle of diffraction. Using Scherrer formula the average crystallite size of the deposited films are calculated to be 2.4nm, 3.9nm, 5.6nm, 6nm, 4nm. Fig. 2 shows the relation between thickness and crystallite size.

Fig. 2 Variation in crystallite size of SnO2 samples

In the polycrystalline thin film structures, generally the major structure or orientation mainly depends on the processing parameters. During the coalescence of two differently oriented nuclei the structural changes occur by surface diffusion and migration of grain boundaries. In such cases, smaller nuclei may easily rotate on coalescence that causes the structural changes. The lowest surface energy, the grain boundary energy and the diffusion of surface atoms influence the stable polycrystalline state of a material.

The reports by Lee (2008) reveal that the preferred orientation of grain growth and the final texture of thin films depend on the strain energy minimization and surface energy minimization. The level of these energy states depends on the thickness of the films. The minimization of strain energy supports one type of texture while the minimization of surface energy promotes another. Furthermore, surface energy dominates at a lower film thickness, whereas at the higher film thickness strain energy will be significant. The surface structure of the substrate forces the nuclei to grow along a specific orientation. It can be concluded from the above discussion that the interaction at the film and substrate interface could be high at lower thickness values. So with increasing thickness of the thin films, strain energy dominates which results in the broadening of peaks. In our XRD patterns, peak broadening is observed with increase in thickness. The relationship between the broadening produced and stress/strain can be found by differentiating the Bragg’s law as given in Fig. 3 (Cullity 1978)

b = ∆2θ = -2∆d/d tanθ

Where, b = extra broadening and ∆d/d = variation in strain

Fig. 3 Variation in Stress / Strain of SnO2 samples

Fig. 4 shows XRD pattern of Zn2SnO4 thin films. All the peaks are well consistent with JCPDS card no. (29-1484) for pure SnO2 and with JCPDS card no. (24-1470) for doped SnO2.

Fig. 4 XRD pattern of Zn2SnO4 thin films

Appearance of (3 1 1) characteristic peak of Zn2SnO4, under as-deposited condition, shows that ZnO has been successfully incorporated into the SnO2 matrix. Further investigations of doped SnO2 samples are under progress. Fig. 5 shows SEM micrographs of SnO2 and Zn2SnO4 thin films. Free large-area growth of nanorods and nanodiamonds of diameter 30nm and 40nm respectively is seen in case of SnO2 thin films (Fig. 5 (a,b)). While on doping different structure was observed due to difference in ionic radii of Zn and Sn. Spherically shaped nanoparticles and long tubular structure is observed with diameter of 20nm and 50nm respectively (Fig 5 (c,d)). Pal (2009) reported SnO2 nanostructures with use of cationic surfactant but it is worth mentioning here that these nanostructures are formed without the use of any surfactant during sol synthesis.

Fig. 5 SEM micrographs of (a-b) SnO2 (c-d) Zn2SnO4

Fig. 6 shows surface morphologies of un-doped SnO2 thin films with thickness of 80-200nm. It can be seen that films are extremely smooth with RMS surface roughness of

(a)

(c) (d)

(b)

less than 3nm for all the films. Grain sizes along with the average and root mean square roughness values, obtained from AFM, are shown in Fig. 7.

Fig. 6 AFM images of SnO2 (a) 80nm (b) 110nm (c) 140nm (d) 150nm (e) 200nm

Fig. 7 RMS and Average roughness of SnO2 Samples

The transmittances of doped and undoped films with different thicknesses are shown in Fig. 8, where (a) shows the results of undoped thin films with different thicknesses. The average transmittance of thin films at 339 nm is above 85% for all films regardless of film thickness, which is in good agreement with the reported values (Shamala, 2004). Generally the required transmittance of transparent conductive thin film for solar cells is over 85% and as such these results indicate that SnO2 thin films are a good candidate to be used as a window layer in solar cells. When zinc is incorporated into the structure a slight increase in average transmission of the films is observed as shown in Fig. 8 (b).

Fig.8 Transmission spectra of (a) SnO2 and (b) Zn2SnO4 samples

Figs. 9 and 10 show refractive indices and extinction coefficient values for SnO2 and Zn2SnO4 thin films with different thicknesses. The peak of refractive index at ~350nm corresponds to direct band gap transition. Variations in refractive index are observed with changes in thickness. For undoped thin films the refractive index increases till a thickness of 100nm after which the refractive index decreases. Refractive indices are in the range of 1.55-1.75 for undoped thin films while refractive index increases when Zn is incorporated into the structure. Refractive index of doped films is greater than the undoped ZnO and SnO2. Generally, the refractive index of zinc stannate thin films is higher than those of ZnO and SnO2 (Cetinorgu 2007).

Fig. 9 Variation in refractive index ‘n’ of (a) SnO2 (b) Zn2SnO4 thin films

Fig. 10 Variation in extinction coefficient ‘k’ of (a) SnO2 (b) Zn2SnO4 thin films

The direct band gap is extracted by linear extrapolation of α2 vs. E (eV) plot to zero on the energy axis. Fig. 11 shows the band gap of doped and undoped SnO2. Band gap for undoped thin films is in the range of 3.66-3.68eV while with incorporation of Zn into SnO2 structure causes an increase in band gap as expected due to Burstein Moss shift. The band gap of Zn2SnO4 thin films are within the range of 3.75-3.78eV. Wang (2007) reported band gap of 3.9-4.06eV. Our results are in good agreement with those reported in literature.

Fig. 11 Band gap variation of (a) SnO2 (b) Zn2SnO4 thin films

Thin films showed room temperature conductivity of the order of 103 Ω-1cm-1. These optical and electrical values are comparable to the ones reported in literature. Zn-doped SnO2 films showed conductivity slightly better than the undoped ones as shown in Fig. 12.

Fig. 12 Conductivity of doped and un-doped SnO2 thin films

CONCLUSION

Undoped and zinc doped tin oxide nanostructures are prepared via cost effective and application oriented sol-gel route. Thin films with different thicknesses are deposited on glass substrate in order to study the structural and optical properties. SEM images show the formation of nanorod (30nm) and nanodiamond (40nm) on the surface of thin films. AFM results show the formation of granular films with average grain sizes of less than 8nm and with RMS surface roughness of less than 3nm. The most important thing is that these uniformly distributed nanostructures are formed without the use of any surfactant. Band gap values are in the range of 3.65-3.68 eV for un-doped films, whereas 3.75 – 3.78 eV was observed for doped SnO2 thin films. Burstein Moss shift of band gap to high energies is observed in case of zinc doped SnO2 thin films. REFERENCE

Baruah, S. and Dutta, J. (2009), “Effect of seeded substrates on hydrothermally grown ZnO nanorods.” J. Sol-Gel Sci. Technol., Vol. 50, 456–464. Cetinorgu, E. and Goldsmith, S. (2007), “Chemical and thermal stability of the characteristics of filtered vacuum arc deposited ZnO, SnO2 and zinc stannate thin films.” J. Phys. D: - Appl. Phys., Vol. 40, 5220–5226. Cun, W., Xinming, W., Jincai, Z., Bixian, M., Guoying, S., Ping’an, P., Jiamo, F. (2002), “Synthesis, characterization and photocatalytic property of nano-sized Zn2SnO4.” J. Mater. Sci., Vol. 37, 2989 – 2996. Erslev, P.T., Chiang, H.Q., Hong, D., Wager, J.F. and Cohen J.D. (2008), “Electronic properties of amorphous zinc tin oxide films by junction capacitance methods.” J. Non-Cryst. Solids, Vol. 354, 2801–2804. Horrocks, A.R., Smart, G., Kandola, B. and Price, D. (2012), “Zinc stannate interactions with flame retardants in polyamides; Part 2: Potential synergies with non-halogen-containing flame retardants in polyamide 6 (PA6).” J. Polym. Degrad. Stab., Vol. 97, 645-652. Ilican, S., Caglar, M. and Caglar, Y. (2010), “Sn doping effects on the electro-optical properties of sol gel derived transparent ZnO films.” J. Appl. Surf. Sci., Vol. 256, 7204–7210. Kim, K., Annamalai, A., Park, S.H., Kwon, T.H., Pyeon, M.W. and Lee, M. (2012), “Preparation and electrochemical properties of surface-charge-modified Zn2SnO4 nanoparticles as anodes for lithium-ion batteries.” Proceeding of ’12 J. Electrochimica Acta - in Press. Li, Z., Zhou, Y., Zhang, J., Tu, W., Liu, Q., Yu, T. and Zou, Z. (2012), “Hexagonal nanoplate-textured micro-octahedron Zn2SnO4: combined effects toward enhanced efficiencies of dye-sensitized solar cell and photoreduction of CO2 into hydrocarbon fuels.” J. Cryst. Growth Des., Vol. 12, 1476− 1481. Lim, H.N., Nurzulaikh, R., Harrison, I., Lim, S.S., Tan, W.T. and Yeo, M.C. (2011), “Spherical tin oxide, SnO2 particles fabricated via facile hydrothermal method for detection of mercury (II) ions.” Int. J. Electrochem. Sci., Vol. 6, 4329 – 4340.

Pal, J. and Chauhan, P. (2009), “Structural and optical characterization of tin dioxide nanoparticles prepared by a surfactant mediated method”, Mat. Charact., Vol. 60, 1512-1516. Pan, Z.W., Dai, Z.R. and Wang, Z.L. (2001), “Nanobelts of semiconducting oxides.” Sci., Vol. 291, 1947-1949. Riaz, S., Naseem, S. and Xu, Y.B. (2011), “Room temperature ferromagnetism in sol-gel deposited un-dpoed ZnO thin films.” J. Sol-Gel Sci. Technol., Vol. 59, 584-590. Shamala, K.S., Murthy, L.C.S and Narasimha, K., (2004), “Studies on tin oxide films prepared by electron beam evaporation and spray pyrolysis methods.” Bull. Mater. Sci., Vol. 27, 295–301. Srivastava, A., Rashmi, B. and Jain, K. (2007), “Study on ZnO-doped tin oxide thick film gas sensors.” J. Mater. Chem. Phys., Vol. 105, 385-390. Tsaroucha, M., Aksu, Y., Irran, E. and Driess, M. (2011), “Synthesis of Stannyl-substituted Zn4O4 cubanes as single-source precursors for amorphous tin-doped ZnO and Zn2SnO4 nanocrystals and their potential for thin film field effect transistor applications.” J. Chem. Mater., Vol. 23, 2428–2438. Villarreal, T.L., Boschloo, G. and Hagfeldt, A. (2007), “Nanostructured zinc stannate as semiconductor working electrodes for dye-sensitized solar cells.” J. Phys. Chem. C, Vol. 111, 5549 -5556. Wang, Y., Ramos, I., Santiago-Avilés, J.J., (2007), “Optical band gap and photoconductance of electrospun tin oxide nanofibers.” J. Appl. Phys., Vol. 102, 093517-1. Yadav, B.C., Singh, R., Singh, S. and Dwivedi, K.P. (2012), “Humidity sensing investigations on nanostructured zinc stannate synthesized via chemical precipitation method”, Int. J. Green Nanotech., Vol. 4(1), 37-45. Zhang, Y., Guo, M., Zhang, M., Yang, C., Ma, T. and Wang, X. (2007), “Hydrothermal synthesis and characterization of single-crystalline zinc hydroxystannate microcubes”, J. Cryst. Growth, Vol. 308, 99-104.