(Invited) Sub-Nm Quantification of Point Defects in Ceramics and Semiconductors Using Atom Probe Tomography

Abstract

Many material properties are dependent upon local defect chemistry and associated charge compensations. Properties such as electronic and ionic transport, intergranular fracture, dielectric constant, and electrical breakdown are dependent upon the accumulation of point defects at internal interfaces such as grain boundaries and heterojunctions. Quantification of these point defect accumulations has been difficult, either due to light element sensitivity, detectability limits, or the 3-D nature of the internal interfaces. Recent work has shown that Atom Probe Tomography (APT) has the requisite counting statistics, 3-dimensionality, detectability limits, and spatial resolution to quantify point defect accumulations at grain boundaries and heterojunctions. Utilizing the 3-D atom by atom nature of APT data has enabled the conversion of analytical characterization data directly into space charge voltages and band alignments, ultimately enabling the direct relationship between materials characterization and materials or device properties. The ability of APT to quantify point defect distributions in 3-D has limitations that are dependent upon the type of defect. Individual vacancies on either oxygen or metal cation sites (Vo ** or Vm ’) have been quantified with APT at grain boundaries and heterojunctions. Quantification of vacancies are heavily dependent upon the detection efficiency of the APT experiment and upon the volume being sampled. Interstitial oxygen or metal cations (Oi ” or Mi *) can be observed but again is limited to the counting statistics of the volume being analyzed. However, their exact location can not be determined at this point as APT is limited in the necessary spatial resolution. Substitutional cations and defect pairs (MN ’) can be quantified as long as the cations are distinguishable in the mass spectrum. Substitutional defects coupled with vacancies (common ordered defect pairs) have been quantified at individual grain boundaries and, in some cases, allow for the calculation of electronic energy levels at interfaces. Schottky defects (Vo ** + Vm ’’) may be quantified using APT, as these defect pairs result in a change in atomic density, however, their quantification is limited by the detection efficiency of the APT instrument. Frenkel pairs on either the oxygen or metal cation sites (Vo ** + Oi ”) do not result in a change in atomic density and are only dependent upon site occupancies. Similar to interstitials, APT does not have the spatial resolution or detection efficiency necessary to identify individual Frenkel pairs. Such analyses will have to wait for Atomic Scale Tomography, of which APT may play a part. Ultimately, the combination of APT’s 3-D spatial and chemical resolution combined with analytical techniques such as EELS that allows for cation valence and free carrier energy levels to be determined will be the most complete solution to defect chemistry quantification.