Advances in Dielectric Materials and Electronic Devices: Dielectrics and Piezoelectrics: Session II Modeling and Applications
Sponsored by: ACerS Electronics Division
Program Organizers: Amar Bhalla, University of Texas; Ruyan Guo, University of Texas at San Antonio; Rick Ubic, Boise State University; Danilo Suvorov, Jožef Stefan Institute
Wednesday 2:00 PM
November 4, 2020
Room: Virtual Meeting Room 9
Location: MS&T Virtual
Session Chair: Amar Bhalla, University of Texas; Ruyan Guo, University of Texas at San Antonio
2:00 PM Invited
Integrating Material Fabrication, Characterization and Modeling to Maximize ROI: Steven Tidrow1; 1Alfred University
Material properties for “ideal” systems, simple gases, elemental metals and semi-conductors as well as for some simply mixed “ideal” systems, like doping of semi-conductors are reasonably well understood. Less understood are ceramic systems that afford significant opportunities for an enormous range of devices and environments. Atomic positions in time and space define material properties and are related to boundary conditions: precursors, processing, kinetics, thermodynamics, etc. The Laboratories for Electro-ceramics (LEC) Group at Alfred University is integrating material design, fabrication, characterization and modeling to reduce the time for discovery and development of materials that outperform materials used in present energy storage and conversion technologies. Some of our ongoing efforts are discussed to illustrate our ability to a priori predict material properties using NSMM, mapping, and our ability to integrate material fabrication, characterization and modeling to maximize return on investment (ROI) in support of “sustainability”.
2:40 PM
Dielectric Capacitance for Chemical and Biological Sensing: Ian Emge1; Charmain Su1; Fow-Sen Choa1; Bradley Arnold1; Lisa Kelly1; Manish Verma1; Kamdeo Mandal1; Himagowri Prasad1; Narsingh Singh1; 1University of Maryland, Baltimore County
The perovskites are very important class of materials for variety of applications. To meet the demands for the tunable devices and high dielectric parallel plate capacitors, several perovskites have been studied extensively. In spite of great progress in processing, low resistivity and process driven variables in properties remain a big hurdle for its wide range of applications as a dielectric capacitor. We observed that effect of chemicals used in wet and semi wet affected dielectric and other semiconducting properties. With these goals, we used the parallel plate capacitors as chemical and biological sensors. The data indicated huge difference in the dielectric and resistivity of the exposed samples. This indicates that perovskites can be used for chemical and biological sensors at very low cost. Preliminary data indicates that after exposing in atmosphere, there materials can recover to original characteristics.
3:00 PM
Identifying Dielectric Breakdown Micromechanisms in Solid Oxides with In Situ TEM: Xinchun Tian1; Xiaoli Tan1; Geoff Brennecka2; Gabriel Caruntu3; 1Iowa State University; 2Colorado School of Mines; 3Central Michigan University
In this talk, we show our in-situ TEM results on the newly observed micromechanisms of dielectric breakdown in oxide nanocrystals using the advanced Hysitron PI95 TEM specimen holder and newly developed specimen preparation approach. First, we discuss the different breakdown mechanisms in TiO2 (i.e., rutile-to-anatase transition, amorphization/melting, and ablation) which depend on the field strength of applied voltages. Then, we discuss the electron-illumination-induced amorphization in electrically stressed BaTiO3, which is anisotropic, inhomogeneous and closely related with the field strength of applied voltage. Last, we show the effects of doping level on the dielectric response of La-doped or Cr-doped BaTiO3 nanocrystals. We expect our results to provide the bases for the overall understanding of the dielectric breakdown phenomenon and pave the way for the design and manufacture of future dielectric materials that operate reliably at higher voltages and higher efficiencies to meet the increasing demands for energy generation, storage, and transport.
3:20 PM Invited
Correlative Models of Some Structural Aspects of Perovskites: Evan Smith1; Kevin Tolman2; Srečo Škapin3; Rick Ubic1; 1Boise State University; 2Idaho National Laboratory; 3Jožef Stefan Institute
Predictive models for composition-structure-property relationships are essential to realizing the full potential of electroceramics; however, industry has largely had to rely upon simple rules of thumb or expensive trial-and-error methods. Empirically derived predictive models have the potential to significantly improve and guide future research in a more cost-effective and timely manner, yet few exist for perovskites containing extrinsic defects or cation ordering. In this work, such models have been derived for the prediction of the effective size of vacancies (oxygen vacancies as well as A-site cation vacancies) and the effects of ordering (A-site or B-site) on cell volume in complex perovskites. In addition, the magnitude of non-cubic distortions (tetragonal or trigonal) and the resultant polarization have been related to ionic sizes and charge. These models give very accurate results and allow the exploration of trends as a function of composition that would be difficult to realize by other methods.
4:00 PM Invited
Investigation of Intergranular Dielectric Properties within the Relation between Fractal, Graph and Neural Networks Theories: Vojislav Mitic1; Goran Lazovic2; Chun-An Lu3; Ivana Radovic4; Vesna Paunovic5; Aleksandar Stajcic2; Branislav Randjelovic6; Srdjan Ribar2; Branislav Vlahovic7; 1University of Nis; University of Belgrade; 2University of Belgrade; 3Industrial Technology Research Institute; 4University of Belgrade; VINČA" Institute of Nuclear Sciences - National Institute of the Republic of Serbia; 5University of Nis; 6University of Nis; University of K. Mitrovica; 7North Carolina Central University
This research has been focused on Yttrium salt modified nanoparticles BaTiO3, sintered at 1350 ˚C. Grain boundary investigations and the influence on dielectric properties were performed from different perspectives and scientific fields. Fractal corrections based on surface (αs) and pore size (αp) corrections, resulted in the relation between the capacitance and Curie temperature, which was important for future miniaturization of electronic devices. Second approach was to develop artificial neural networks for electronic parameters between the grains, giving relation between voltage and relative capacitance change, from the level of the bulk sample down to the grains boundaries. Finally, the third approach included employment of graph theory for networking of electronic parameters between the neighboring grains, which led to a calculation of capacitance change between grains. Downsizing the bulk results to an intergranular level showed that named three approaches open the new perspectives in the future research of miniaturization.
4:40 PM
Determining Complex Dielectric Properties from Coaxial Transmission Line Data Using a Machine Learning Approach: Robert Tempke1; Liam Thomas1; Christina Wildfire2; Dushyant Shekhawat2; Terence Musho1; 1West Virginia University; 2National Energy Technology Laboratory - US Department of Energy
This study investigated and developed an artificial neural network to predict the dielectric properties materials between 0.1-13.5 GHz. The approach utilized a two-dimensional convolutional neural network (CNN) in conjunction with a finite element electromagnetic model to generate a large solution space of different dielectric property combinations. This CNN was trained using a common back-propagation algorithm. The network is taught using supervised learning with a training, validation and test set. The dielectric material within the FE model was described using a complex description with the real part ranging from 1-100 and the imaginary part ranging from 0-0.2. Once convergence had been reached the network was double validated using experimental data collected in a coaxial airline. The same loss metrics were used to show that the network worked on experimental data and not just idealized computational data.