Perovskite oxynitrides of tantalum, titanium and niobium and their solid solutions as self-cleaning coatings.

Abstract

Metal oxynitrides adopting the perovskite structure have shown to be active photocatalysts. In this work we established a route for the synthesis of CaTaO2N, SrTaO2N, BaTaO2N, LaTaON2, EuTaO2N, SrNbO2N, LaNbON2 and LaTiO2N as powders as well as thin films, and an assessment on their photocatalytic activities. Their synthesis was achieved using the polymeric precursor method (Pechini method), which makes use of citric acid and propylene glycol to form a polymeric resin containing the metal cations homogeneously distributed. For the thin film deposition, alumina and quartz substrates were dip-coated into the polymeric gel to form an amorphous oxide precursor film, followed by ammonolysis. Prior to ammonolysis, both, powder and thin film precursors were annealed in air at 800˚C to obtain the oxide precursor. Perovskite oxynitride phases were synthesised in the temperature range of 850-1000 ˚C, a flowing rate of 250 ml min-1 , a heating ramp of 3˚C min-1 and reaction time of 10-54 hours. Phase purity was confirmed by XRD and Rietveld/Le Bail analysis and diffuse reflectance spectra were recorded for each sample. Optical band gaps were calculated from the Tauc plot derived from the Kubelka-Munk function and were found in the range of 1.7-2.4 eV. A cobalt oxide co-catalyst (CoOx) was deposited onto each film by drop casting and the photocatalytic activity was assessed under visible light using dichlorophenolindophenol (DCIP) dye degradation in the presence of a sacrificial oxidant. The light source used was a solar simulator equipped with a 400 nm cut-off filter. The dye degradation test demonstrated photocatalytic activity in all samples except EuTaO2N and BaTaO2N. The three most active samples SrTaO2N, CaTaO2N and SrNbO2N showed initial rate constants of 5.0(1) × 10−4 min−1, 24.3(5) × 10−4 min−1 and 64(5) × 10−4 min−1, respectively. The cocatalyst loading was investigated at nominal surface concentrations of 0.04-0.46 µg cm-2 , however, for all samples, a co-catalyst loading of 0.3 µg cm-2 resulted in being the optimal one, providing an ii equilibrium between sufficient active sites for the degradation to occur without blocking the light from reaching the underlying photocatalyst. The three most active samples SrTaO2N, CaTaO2N and SrNbO2N were assessed on their selfcleaning abilities using the stearic acid test, demonstrating full degradation of an organic contaminant under a visible light source. The protocol for the mineralisation of stearic acid required transparent substrates. In this context, quartz substrates were protected by a layer of Al2O3 deposited via Aerosol-Assisted Chemical Vapour Deposition (AACVD). This method allowed to dip-coat the polymeric gels onto the quartz substrates, preventing side reactions on the perovskite phase with the quartz during the ammonolysis. The pure oxynitride phase was obtained using the conditions optimised for the perovskite oxynitrides deposited onto alumina tiles. For the photocatalytic test, the three samples SrTaO2N, CaTaO2N and SrNbO2N were decorated with 0.3 µg cm-2 cobalt atoms, and the degradation of stearic acid was monitored by FTIR. The rate constants for the mineralisation of the stearic acid were determined and gave values of 14(2) × 10−4 min−1 for SrNbO2N, 8(1) × 10−4 min−1 for SrTaO2 N, and 7.8(6) × 10−4 min−1 for CaTaO2 N. In line with the results from the DCIP dye testing, the SrNbO2N thin film was the most active sample. Finally, we reported a method for the preparation of a materials library using manual inkjet printing and subsequent screening of the photocatalytic activities of the different constituting compositions. This method was applied for the synthesis of a materials library containing nine different compositions of alkaline (Na, K) and alkaline earth (Mg) doped SrNbO2N.

More information

Identifiers

ORCID for Geoffrey Hyett: orcid.org/0000-0001-9302-9723

Catalogue record

Export record