Electronic structure of cubic ErxGa1−xN using the LSDA+U approach

First-principles study of electronic structure and magnetism of cubic Al1−xErxN using the LSDA+U approach

“…The same effect has been reported by theoretical studies for Er in GaN [26,27]; in our earlier calculations [27], DE VBM and Nb have been estimated to be 0.057 and 0.11 eV, respectively, for Ga 0.75 Er 0.25 N. The hybridization of the e orbitals with the host states at the VBM is significantly weaker than that of Mn, hence we conclude that for AlN host there will be a much weaker spin polarization effect induced on the hole carriers by erbium than when Mn is used.…”

“…A number of calculations of the electronic and magnetic properties have been reported on RE-doped GaN [23][24][25][26][27], and the trivalent configuration of the RE impurities is shown to constitute the ground state [25,26], the most stable site when the RE is located at a Ga substitution site [23,25], and the magnetic properties were different from Mn-doped GaN [24,26,27]. Relatively weak magnetic interaction of the rare-earth ions with the host states at the valence and conduction band edges has been reported [24,26,27], in particular for GaN:Gd [24,26].…”

“…Relatively weak magnetic interaction of the rare-earth ions with the host states at the valence and conduction band edges has been reported [24,26,27], in particular for GaN:Gd [24,26]. Dalpian and Wei [24], using the general potential linearized augmented plane wave (LAPW) method, reported that the ferromagnetic (FM) phases can exist in n-type GaN:Gd samples originating from the coupling between the Gd e and the host s states.…”

Structure and energetics of Er defects inLiNbO3from first-principles and thermodynamic calculations

“…Unfortunately, no parameters have been determined for Er in either LiNbO 3 or Er 2 O 3 . However, there are several sets of parameters available for Er in GaN or in Er x Ga 1−x N. 66,67 For Er in GaN using the density functional based tight-banding ͑DFT-TB͒ method, 66 U − J = 10.3 eV was used. However, it is known that the DFT-TB method yields higher estimates for U and J than does regular DFT.…”

“…66 For Er x Ga 1−x N using LSDA+ U, 67 U = 8.6 eV and J = 0.75 eV has been used. Since there are no accepted values for U and J for LiNbO 3 , we have explored a range in order to map out the possible magnitude of the effects of electron localization: ͑i͒ U − J = 4 eV; ͑ii͒ U − J = 7.85 eV, 67 and ͑iii͒ U − J = 10.3 eV. 66 Although none of these parameters may actually be the best physical representation of electron localization for Er in LiNbO 3 , they do span the range of reasonable values and should thus enable us to assess the importance of the localization of electrons on its effects on the defect energetics.…”