Abstract— The purpose of this work is to understand, for Ce doped materials, how the effect of the local crystal environment of Ce compared to the character of the host cation, influences the scintillation and fluorescence of different materials. In the case of YAP (YAlO 3 ) and LAP (LaAlO 3 ) while very similar materials, YAP is a moderately bright Ce activated scintillator while LAP shows no Ce activation. It is known from experimental work that in the case of LAP the Ce 5d states lie in the conduction band of the host, quenching any scintillation. Since YAP and LAP have different crystal structures it is not possible to know from experiment if the change in the local environment of the Ce or the different host cation are responsible for the very different scintillation properties. To isolate the di fferent effects of local environment compared to host cation, two theoretically synthesized systems obtained by swapping La and Y in LAP and YAP geometries were created. In this way we artificially create a La version of YAP (and Y version of LAP) where the local environment of Ce is the same as YAP and only the next nearest neighbour cation differs. We then perform first principles density functional based calculations for all four systems to determine the positions of the 4f and 5d Ce states relative t o the conduction and valence bands of the host. Our preliminary results show that LAP in the YAP geometry still has the Ce 5d in the conduction band, quenching scintillation while YAP in the LAP geometry still has the 5d below the conduction band. This su ggests the cation character, which typically determines the conduction band character and position, is more important in determining Ce activation properties than the local dopant environment for these materials. I. BACKGROUND The subject of this work is the theoretical study of cerium - doped gamma -ray detector materials. Our goal is to gain new knowledge of general principles of scintillation through computational mode lling of the key scintillation step in synthetic materials. Scintillation in Ce -doped materia ls corresponds to a transition between the Ce 5d excited state, (usually referred to as (Ce3+)*) to the ground state 4f level. In previous work [1,2,3] we have performed density functional based theoretical calculations for the prediction of candidate scintillator materials based on the calculation of the Ce 5d Document received November 21, 2010. This work was supported by the U. S. Department of Homeland Security and carried out at the Lawrence Berkeley National Laboratory under U. S. Department of Energy Contract No. DE -AC02-05CH11231. The authors are with E.O. Lawren ce Berkeley National Laboratory, 1 Cyclotron Rd, MS55R0121, Berkeley, CA, 94720. (Phone: 510 -486-4858, fax: 510-486-4768, email: RBuchko@lbl.gov and 4f levels relative to the valence and conduction band of the host material. A necessary condition for Ce activated luminescence and scintillation is that the 4f and 5d levels must lie in the gap of the host material. An important question that needs to be answered within this paradigm is what is the main factor needed to satisfy this condition in a specific compound: the chemical composition of the compound, or the crystalline structure and neares t neighbour geometry of the cerium dopant ion. In order to answer this question, we consider two cerium -doped compounds: lanthanum and aluminium Perovskites ( commonly called LAP and YAP). YAP is a known cerium activated scintillation material, while LAP ha s been showen not to exhibit cerium -activated scintillation, unlike many other compounds that contain lanthanum. In order to check the relative importance of geometry and chemical composition in determining the position of cerium 4f and 5d levels within the host crystal, we consider two theoretically synthesized systems obtained by swapping La and Y in the two compared compound s. If the new synthetic YAlO 3 and LaAlO 3 in LAP and YAP geometries exhibit the same scintillation properties as real compounds it wo uld indicate that the chemical composition is the dominant factor. However, if change in the crystalline structure forced change in scintillation properties, it would indicate at the relative importance of local geometry of the Ce 3+ dopant. This study takes advantage of the unique ability of computational modelling to examine properties of compounds that may be unstable or cannot be sythesized experimentally II. METHODS Density functional theory calculations are used to construct wavefunctions of the ground st ate and the lowest excited state modelled by constraining band occupancy as described in [1 -3]. Wavefunction spatial localization within individual ion Wigner -Seitz cells is analysed to determine which states are ionic 4f and 5d states of the cerium dopant ion. Potential scintillators are defined as systems with Ce 4f above the valence band maximum of the host crystal and Ce 5d below the conduction band minimum of the host. Electronic structure calculations for the true LAP and YAP compounds were performed using the known geometry from ICSD database. Supercells of 120 -160 atoms were constructed from the unit cells a single Ce ion was inserted into the supercell replacing a lanthanum or an yttrium ion, and the atomic positions within the cell were relaxed t o accommodate the dopant. In the synthetic compounds, we first relaxed the host crystal geometry (both atomic positions A theoretical study of the relative importance of chemical and geometric effects for Ce -based scintillation in La and Y aluminum perovskites Rostyslav Boutchko, Andrew Canning, Anurag Chaudhry and Stephen Derenzo 275 978-1-4244-9105-6/10/$26.00 ©2010 IEEE