Structural and morphological properties of Fe₂O₃/TiO₂ nanocrystals in silica matrix

Václav Valeš¹, Václav Holý¹, Maja Buljan², Vesna Janicki², and Sigrid Bernstorff³

¹ Department of Condensed Matter Physics, Charles University in Prague, Czech Republic

² Rudjer Boskovic Institute, Zagreb, Croatia

³ ELETTRA Sincrotrone Trieste, Basovizza, Italy

Titania (TiO₂)-based systems have been very intensively studied in last decades because of their photocatalytic activity, which found broad commercial applications. Functionalized titania composites, especially Fe₂O₃/TiO₂ systems attracted a lot of attention recently (see [1, 2], among others), since they make it possible to improve the photocatalytic performance of titania. The ε-phase of Fe₂O₃ exhibits a very large magnetic coercivity at room temperature [3] so that Fe₂O₃/TiO₂ in solutions can easily be manipulated by external magnetic field. Fe₂O₃/TiO₂ compact thin layer composites as a photocatalyst can respond to visible light due to the narrow band-gap of Fe₂O₃.

We have studied two types of structures, namely (Fe₂O₃+SiO₂)/SiO₂ and (Fe₂O₃+SiO₂)/ (TiO₂+SiO₂)/SiO₂ periodic multilayers. Both types of samples were grown by a co-deposition of the actual material together with SiO₂ and pure SiO₂ as an interlayer spacer. The samples have been subsequently annealed for one hour at various temperatures in air or forming gas (FG – Ar + 4% H₂). We prepared a large series of samples with various thicknesses, annealing temperatures and annealing atmospheres in order to find the optimal conditions of the preparation; here we report only several characteristic examples, their basic parameters are summarized in Tab. 1. Samples A – F consist of 10 periods, samples A – C were annealed in forming gas atmosphere, while samples D – F at the air. Samples that did not contain titania were prepared under various conditions – thickness of Fe2O3+SiO2 layer from 0.6 nm to 2.0 nm, annealing temperature from 300 °C to 900 °C at air, forming gas, or vacuum. These samples consisted of 20 periods.

X-ray diffraction curves of the samples have been measured by laboratory diffractometer with a standard x-ray tube (CuKa radiation, 1.5 kW) using a parallel-beam setup. During the measurement the angle of incidence of the primary beam was kept constant at 1 deg to suppress the substrate signal.

Small-angle x-ray scattering in grazing incidence geometry (GISAXS method) has been carried out at ELETTRA synchrotron source (SAXS beamline) using the photon energy of 8 keV. The incidence angle of the x-ray beam was chosen 0.25 deg, i.e. just above the critical angle of total external reflection.

Figure 1. X-ray diffraction curves of samples A, B, C (a) and D, E, F (b). The insets display the details of the diffraction peaks along with their fits to theoretical curves (lines). The vertical lines indicate the theoretical positions of the diffraction peaks for various phases: black full line – rutile, red dashed line – hematite (a-Fe₂O₃), blue dash-dotted line – maghemite (g-Fe₂O₃), and green dotted - ε-Fe₂O₃.

The x-ray diffraction data on samples without titania exhibit diffraction maxima neither before, nor after post-growth annealing, i.e., the Fe₂O₃ component did not crystallize under the annealing conditions used in this work. The diffraction curves of samples A-F are presented in Figure 1. From the data it follows that TiO₂ nanoparticles grow during the annealing, from the positions of their diffraction peaks we identified the tetragonal rutile phase. More interestingly, in samples E and F we detected also Fe₂O₃ nanoparticles, however their diffraction maxima are rather weak, which indicates that the density of these particles is very small. The positions of the Fe₂O₃ maxima roughly correspond to the hexagonal hematite a-Fe₂O₃ or ε-Fe₂O₃ phase. We have compared the diffraction maxima with the simulations based on the Debye-formula approach [2] assuming spherical particles and from this comparison we estimated the mean particles sizes (Tab. 1).

Table 1. Parameters of the TiO₂-containing multilayers determined from x-ray diffraction and GISAXS methods, all values are in nm. d is the nominal thickness of layers, T is the annealing temperature, R_L, R_V are lateral and vertical diameters of particles resp., a_L, a_V are mean base lateral and vertical vectors and σ_L, σ_V are their rms.

Sample	d	T (°C)	x-ray diffraction		GISAXS
			RTiO2	RFe2O3	RL	RV	aL	aV	sL	sV
A	0.6	700			1.8 ± 1.0	1.8 ± 1.0	3 ± 3	10.0 ± 0.5	2 ± 1	5 ± 1
B	1.0	700			2.5 ± 1.0	1.5 ± 1.0	3 ± 3	11.5 ± 0.5	2 ± 1	4 ± 1
C	2.0	700			3.5 ± 1.0	1.5 ± 1.0	8 ± 3	12.5 ± 0.5	2 ± 1	4 ± 1
D	0.6	900	1.6 ± 1.0
E	1.0	900	2.1 ± 1.0		3.0 ± 0.5	2.0 ± 0.5	16 ± 5	10.0 ± 0.5	5 ± 1	5 ± 1
F	2.0	900	3.2 ± 1.0	2.7 ± 1.0	5.0 ± 0.5	3.0 ± 0.5	16 ± 5	11.5 ± 0.5	5 ± 1	10 ± 1

From the fitting of experimental GISAXS data we found that the (Fe₂O₃+SiO₂)/SiO₂ multilayers exhibit no diffraction peaks and no side maxima in GISAXS intensity maps. Therefore, they remain amorphous and do not form any ordered structure of amorphous or crystalline particles during post-growth annealing at temperatures up to 900°C. On the other hand, the multilayers (Fe₂O₃+SiO₂)/(TiO₂+SiO₂)/ SiO₂ exhibit both diffraction peaks and GISAXS side maxima after annealing, thus they contain ordered array of crystalline particles. In these samples, both rutile TiO₂ and hematite a-Fe₂O₃ are present, the latter with much smaller occurrence. Therefore, the presence of the TiO₂ phase facilitates the crystallization of Fe₂O₃ during post-growth annealing. Increasing the thicknesses d of the Fe₂O₃ and TiO₂ layers, the mean lateral size of the particles increases; the vertical size however remains almost constant so that the particles are disc-shaped for larger d. A similar trend of increasing the lateral particle size is observed for increasing annealing temperature. The number of the observed diffraction peaks did not make it possible to determine the lateral and vertical sizes. The particle size determined from diffraction is affected by possible deformation of the particle lattice and/or by structural defects, therefore a detailed comparison of the particle sizes determined from both methods is not possible. Nevertheless, the particle sizes obtained from x-ray diffraction and GISAXS data are not in contradiction.

References

1. H. Cui, W. Ren, W. Wang, J. Sol-Gel Sci. Technol. 58, 476 (2011).

2. V. Valeš, J. Poltierová Vejpravová, V. Holý, V. Tyrpekl, P. Brázda, and S. Doyle, phys. stat. solidi C 7, 1399 (2010).

3. J. Jin, K. Hashimoto, and S. Ohkoshi, J. Magn. Magn. Mater. 15, 1067 (2005).

Acknowledgement.

The work was supported by the Czech Science Foundation (project P204-11-0785) and by the Grant Agency of Charles University in Prague (project SVV 263307).