In-situ XRD study of crystallization of amorphous TiO2 thin films of different thickness

 

Lea Nichtová1, Radomír Kužel1, Zdeněk Matěj1, Jan Šícha2

 

1Department of Condensed Matter Physics, Faculty of Mathematics and Physics, Charles University in Prague, Ke Karlovu 5, 121 16 Praha 2

2Department of Physics, Faculty of Applied Sciences, University of West Bohemia in Pilsen, Czech Republic

Titanium dioxide films have found many different applications because of several excellent properties. First of all, this is photocatalytic activity and hydrophilicity after irradiation by UV radiation. Moreover, they are chemically stable and can have reasonably high hardness.

However, these properties depend significantly on the crystallinity, phase composition and microstructure of the films. In this study, crystallization of amorphous films with different thickness (50-2000 nm) deposited on silicon substrates was investigated by in-situ isochronal and isothermal annealing at different temperatures and compared with the post-annealing of both amorphous and nanocrystalline films.

The X’Pert Pro diffractometer with MRI high-temperature chamber and parallel beam geometry with Goebel mirror, for texture and stress measurements, the Eulerian cradle and polycapillary were used, respectively.

In earlier experiments, crystallization temperature of about 250 °C was found for thicker films while it was somewhat higher for very thin films (below 200 nm) [1, 2].

Therefore in-situ measurements were performed at slightly lower temperatures (180 °C, 220 °C) and time dependences of  selected XRD profiles were investigated. It was found, that the process can well be described by the modified Avrami equation (see Fig. 1) that is applied to integrated intensities of the diffraction peaks,

I = 1-exp[-b(t t0)n)],

 

where the exponent n was in the range 2−2.5 and it was slightly increasing with the film thickness. This lower value may indicate two dimensional character of the crystallite growth. The initial time t0 of crystallization (non-zero intensity) increases nearly exponentially with the decreasing thickness while the slope b increases significantly for thicker films. Typical time necessary for the crystallization of the whole film volume varied from several hours for thicker layers to about ten days for the thinnest films, for the used temperatures. Mesurements confirmed that the crystallization of very thin films is rather slow (Fig. 2). 

Figure 1. Normalized integrated intensity of anatase diffraction peak 101 in dependence on annealing time for the 630 nm thick film, dots - experimental data, line - fitted modified Avrami equation.

 

Figure 2. Normalized integrated intensity of anatase diffraction peak 101 in dependence on annealing time (annealing temperature 180 °C) for films of different thickness. Values were calculated by modified Avrami equation with the parameters fitted on experimental curves (Fig. 1). The thinnest film (48 nm) was heated to 220 °C.

 

 

Fast crystallization of the order of minutes appeared at 230 °C for thicker films and was higher (290 °C) for the thin films with the thickness below 100 nm. This only confirmed the results obtained on post-annealed films.

Weak texture was changing during the crystallization since the intensity ratii of different peaks were varied with annealing time. At the beginning, the crystallites with the (00l) orientation were developed. However, after complete crystallization, the texture was weak except the very thin films (below 100 nm).

Significant shifts of diffraction peaks with the temperature were observed and tensile residual stresses were confirmed by the sin2q method for different diffraction peaks. They decrease with the increasing film thickness. Line profile analysis indicated the growth of relatively large crystallites (100 nm) already at the beginning of crystallization unlike the films which were deposited as nanocrystalline with the crystallite size of 5−10 nm which remained nanocrystalline to relatively high temperatures (600 °C).

 

The work is supported by the Grant Agency of the Czech Republic (no. 106/06/0327) and Grant Agency of Charles University.

 

1.     R. Kužel, L. Nichtová, D. Heøman, J. Šícha, J. Musil

Zeitschrift fuer Kristallographie. Suppl. 26 (2007) 241-246.

2.     R. Kužel, L. Nichtová, Z. Matěj, D. Heřman, J. Šícha, J. Musil, Zeitschrift fuer Kristallographie. Suppl. 26 (2007) 247-252.