| About this Abstract |
| Meeting |
Materials Science & Technology 2019
|
| Symposium
|
Ceramics and Glasses Simulations and Machine Learning
|
| Presentation Title |
Physics-Based Machine Learning Models for High Throughput Screening of Novel Scintillator Chemistries |
| Author(s) |
Blas P. Uberuaga, Ghanshyam Pilania, Christopher R Stanek, Kenneth J McClellan |
| On-Site Speaker (Planned) |
Blas P. Uberuaga |
| Abstract Scope |
Applications of inorganic scintillators—activated with lanthanides such as Ce—are found in diverse fields. As a strict requirement for scintillation, the 4f ground state and 5d lowest excited state levels induced by the activator must lie within the host bandgap. This talk will discuss a new machine learning based screening strategy that relies on high throughput predictions of the lanthanide dopants’ energy levels with respect to the host valance and conduction band edges for efficient chemical space explorations to discover novel inorganic scintillators. Using a set of perovskites as examples, we demonstrate that the developed approach is able to capture systematic chemical trends across host chemistries and effectively screen promising compounds in a high throughput manner. While other performance requirements need to be considered for a viable scintillator, the present scheme can be a practical tool to systematically down-select the most promising candidates for a subsequent in-depth investigation. |