Optical absorption by design in a ferroelectric : co-doping in BaTiO3
- Authors
- Publication Date
- Jan 18, 2022
- Source
- HAL-Descartes
- Keywords
- Language
- English
- License
- Unknown
- External links
Abstract
Photovoltaic devices based on ferroelectrics have aroused vast interest due to their promising prospective in solar energy harvesting, photocatalysts, or optic-electro devices. Ferroelectrics have spontaneous polarization, which is helpful to separate the photogenerated carriers. The challenge is that most ferroelectrics have a typical bandgap larger than 3 eV, limiting the absorption of the solar spectrum. Generally, the bandgap ABO3 perovskite oxides can be tuned by chemical substitution. In the present work, firstly, The effect of (Mn3+, Nb5+) co-doping BaTiO3 on the B sites (BaTi1-x(Mn1/2Nb1/2)xO3) were investigated. The bandgap decreases from 3.4 eV for pure BaTiO3 to 2.5 eV for doped samples. Besides, the onset absorption drops down to 1.5 eV, which is promising for solar energy harvesting. The ferroelectric deterioration increases with the increase of doping levels. There is no ferroelectric properties with doping higher than 0.075. The influence of doping concentration on the structure was also investigated. Secondly, we have investigated the effects of the synthesis parameters on the microstructure, electrical and optical properties of BaTi1-x(Mn1/2Nb1/2)xO3 (x = 0.075) materials. The Curie temperature Tc shows strong variations on the synthesis conditions, specifically it depends on the ball-milling rate. Nevertheless, while the ferroelectric properties are affected, the absorption properties are weakly, if any, changes. We also studied the optical absorption of co-doped BaTi1-x(X1/2Y1/2)xO3, X = Sc, Mn, Fe, Co; Y= Nb, Ta) systems. It is found that (Co3+, Nb5+) co-doping show higher remnant polarization than the others, and the onset of absorption of this doping can decrease down to 1.5 eV. We have confirmed this by Density Functional Theory (DFT) calculations, demonstrating that the co-doping with 3d≠0 ions inserts intra-gap levels responsible for reducing “artificially” the optical bandgap, fully in line with our experimental observations. The tetragonality and the polar coherence length decrease after co-doping, impacting the ferroelectric features. The measured photoconductivity confirms that (Co3+, Nb5+) doping has superior absorption than others. Finally, we also considered (X2+, Y6+) co-doping of BaTiO3 by substitution on B sites or A-B sites. In contrast to our DFT calculations, the optical absorption shows a slight change whatever the dopants investigated, and the bandgap has only a tiny change. However, the ferroelectric behavior degenerates after doping making the hysteresis loops slimmer which permits (Zn2+, Mo6+) and (Zn2+, W6+) co-doping to show high energy storage responses, bigger than pure BaTiO3. Finally, this research work shows that co-doping in BaTiO3 is a good strategy for narrowing the bandgap and how the ferroelectric properties are impacted providing key ingredients for the application of such materials in optoelectronic devices.
