Composition and Temperature Dependent Properties of the Lithium Niobate Tantalate Crystal Family
Lade...
Datum
Autor:innen
Betreuer/Gutachter
Weitere Beteiligte
Beteiligte Institutionen
Herausgeber
Zeitschriftentitel
ISSN der Zeitschrift
Bandtitel
Verlag
Lizenz
Zitierlink
Zusammenfassung
Ferroelectrics garner a lot of attention due to their inherent properties which allow for widespread usage, ranging from memory devices (utilizing ferroelectricity itself), over actuators and frequency filters (utilizing piezoelectricity), to sensors or potential energy harvesting applications (utilizing either piezo- or pyroelectricity or photo-electrical properties). In contemporary society, ferroelectrics are crucial for telecommunication technology and within ever-downscaled consumer electronics. Among ferroelectrics, lithium niobate (LN) and its isomorphic counterpart lithium tantalate (LT) feature a prominent role due to their large piezo- and pyroelectric coefficients, as well as their high photoelasticity and non-linear optical properties. Crucially, both materials exhibit a Curie temperature amongst the highest for all known ferroelectrics, motivating their usage at harsh temperature environments. By combining both materials to lithium niobate tantalate solid solutions (LNT), their properties can be fine-tuned, allowing for a wider range of applications of the LN-LT material family.In this work, LN, LT, and LNT are investigated using an ab initio theoretical framework based on density functional theory (DFT).
The thermal properties of LN and LT are simulated by first determining the thermal expansion coefficients within the quasiharmonic approximation for phonons. The ferro- to paraelectric transition is examined by employing a fully anharmonic phononic approach within the stochastic self-consistent harmonic approximation. Further temperature-dependent material properties, such as the elastic- and piezoelectric constants, are determined by a specifically trained machine-learned force field, which is based on DFT training data.
The LNT solid solutions are modeled using special quasirandom structures. Their structural, electronical, optical, and thermal properties are computed within DFT and compared with respect to their niobium/tantalum composition. For most of the extracted properties, deviations from Vegard's law can be observed. Raman and infrared spectra are simulated and interpreted with respect to the selection rules of the end compounds LN and LT.
All results are carefully and extensively compared to measurements from experimental groups within the FOR5044 research unit, and the models used throughout this work are critically discussed.