STUDY OF THE Xe VI SPECTRUM

C.J.B. PAGAN; R. A. ABRAH~AO; J.G. REYNA ALMANDOS; M. RAINERI; M. GALLARDO

Costa do Sauípe, Bahía

Congreso; XXXVII Encontro Nacional de Física da Matéria Condensada; 2014

In present work, we calculated weighted transition probability rates (gA) for Xe VI from a semi-empirical method, considering the effects of core polarizability, and compared our results with the literature. The ground configuration of Xe VI is [Kr] 4d105s25p; the core configurations, used to calculate the electric diplole matrixelements, are [Kr] 4d105s2and [Kr] 4d10, that last considered when the 5s electron is active. Previous values for core polarizability were alfa = 4:2 +- 0:8a30for the [Kr] 4d105s2core, from the experimental work of Wang et al.(1996), and the theoretical value of alfa = 4:52a30from the Fraga&Muszy¶nska´s table (1981); for the [Kr] 4d10core, F&M obtained alfa = 0:88 a30. The semi-empirical method used here consists in recalculating the atomic structurethrough the Hartree-Fock method using Slater parameters adjusted to ¯t the levels obtained experimentally. This allows us to reproduce the atomic structure obtained in experiment, and to predict unknown levels of the Rydberg series needed to calculate the polarizability by the usual oscillator strength formula. In the past year, we presented the value for [Kr] 4d105s2 core polarizability as alfa = 5:8 +- 0:1 a30. Now, we calculated the core polarizability for [Kr] 4d10as alfa = 0:99 +-0:10 a30, and after that, we used that values to calculate the transition probabilities for Xe VI by the semi-empirical method described herein, considering the core polarization effects in our calculations. In order to verify our predictions, we calculated the statistical correlation between the log gA values obtained with this method with experimental line intensities using different values of alfa. For 5s25p ¡5s5p2transitions we did not found significant differences between the correlations obtained using our value for ® andthat from F&M table. Otherwise, for transitions that do not involve the 5s electron (5s25p ¡ 5s2(5d; 6s)), the experimental intensities are statistically correlated to the log gA value by 0:30 when using alfa = 4:52 a30and by 0:47 when alfa = 5:8 a30, what is an evidence that our value produced a much better result than that presentedpreviously in the literature.