Electrocaloric Effect on Lead-Free Ferroelectrics: Challenges in Identifying Trends and Evaluating Predictive Models.

The electrocaloric effect (ECE) has become one of the most intensively studied topics in ferroelectrics, with dozens of new papers that report experimental results and provide increasingly extensive data compilations every year. However, the heterogeneity of the literature, arising from differences in compositions, dopants, preparation routes, measurement protocols, and analysis methods, makes direct comparison between studies highly problematic. In this work, we focus on barium titanate (BaTiO3) as a representative lead-free ferroelectric system. BaTiO3 was chosen because, within this class of materials, it offers by far the largest body of reported ECE results, obtained under a wide range of experimental conditions, thus allowing for the most comprehensive characterization. Using this example, we explore whether meaningful patterns related to the influence of chemical substitution on the magnitude and temperature dependence of the ECE can be discerned. In addition, we critically examine why certain comparisons reported in the literature may be misleading or inherently unreliable. Finally, we discuss predictive approaches, including those employing artificial intelligence algorithms, and evaluate their applicability and limitations in modeling the electrocaloric response.