Fermi level pinning in Co‐doped BaTiO₃: Part II. Defect chemistry models

Abstract

A first-principles informed grand canonical defect chemistry model capable of accounting for non-stoichiometry and partial equilibration of different sub-lattices is developed and used to study Mg and Mn doped, and (Mg+Y) and (Mn+Y) co-doped BaTiO 3 to elucidate the role of Mn and Y in improving the resistivity and resistance degradation of BaTiO 3 as observed by Ryu et al. in Part I of this series of papers. The model qualitatively captures the behavior of the samples in all conditions, reproducing the observed carrier plateau and increased resistivity of (Mn+Y) co-doped BaTiO 3 , and expected trends in the concentrations of free oxygen vacancies with doping. These trends reflect the observed differences in degradation characteristics, and help explain the substantially improved degradation resistance of the (Mn+Y) co-doped samples. Our model adds to the mechanism proposed by Yeoh et al. that the Fermi level is pinned by the multivalent character of Mn Ti in (Mn+Y) co-doped BaTiO 3 by giving insight into the role of barium vacancies, the site preferences of the dopants, and defect complexes in this mechanism. These insights provide a set of criteria in the search for sets of co-dopants with similar behaviors.

Fermi level pinning in Co‐doped BaTiO3: Part I. DC and AC electrical conductivities and degradation behavior

“…These issues can be circumvented by density-functional theory (DFT) calculations with grand canonical defect model in Part II of this series. 15 The defect chemistry simulations explain the observed electrical conductivity as a function of pO 2 and temperature in all compositions. Moreover, the Fermi level pinning mechanism in (Mn+Y) co-doped BaTiO 3 is elucidated and discussed with respect to the improved electrical resistivity and time-dependent degradation characteristics.…”

“…[23][24][25] While the grain-boundary behavior is certainly important to the overall conductivity, 23 the present A survey of the published literature shows that process modeling of CMCs has been studied by a number of researchers. Processes that have been modeled include RMI, 15,16 CVD, 17,18 CVI, [19][20][21][22] and Sol-Gel Infiltration. 23 Impregnation of alumina matrix in eight-harness satin Nextel 610 fabric has been studied numerically in Ref.…”

“…Although several canonical defect chemistry models developed in literature can predict the equilibrium defect structure, we find that our experimental results cannot be simply explained by canonical models due to the following issues: (1) defect complexes known to be important in BaTiO 3 27 are not considered, and (2) the cation dopants, in particular Y can substitute either on the A‐ or B‐cation sublattices, 28 and there is a lack of empirically derived thermodynamic parameters for these effects. These issues can be circumvented by density‐functional theory (DFT) calculations with grand canonical defect model in Part II of this series 15 …”

“…In addition, we measure the time‐dependent leakage current to understand the effects of the doping chemistries on the degradation kinetics. In Part II, 15 density functional theory (DFT) calculations are utilized to predict the defect chemistries of all compositions. Defect concentrations as a function of p O 2 are calculated over a wide range of temperatures with grand canonical defect models.…”

Defect mechanisms, oxygen vacancy trapping ability in rare‐earth doped BaTiO3 from first‐principles and thermodynamics

“…29 Hybrid functionals, for example, HSE06, or extrapolation schemes have been employed to correct the band gap error in defects calculation. 6,18,19,21 However, the extrapolation of band edges needs to be fitted with experiment data and is lack physical meaning, while the computational cost of hybrid functionals increases dramatically compared with traditional functionals. In this work, PBE functional was selected and confirmed to successfully reproduce the relative defect transition level, and site occupancy compared with the previous theoretical and experimental works (see the Supporting information S4 for the comparison of PBE and HSE functionals).…”