![]() ![]() The model system has a nominal Ti concentration of 0.04 at.%, which falls into a regime that is no longer accessible with STEM-based imaging techniques and is about three orders of magnitude lower compared to previous APT work performed on highly doped classical semiconductors 17, 18. Here, we overcome this fundamental limitation by utilizing the unique chemical accuracy and sensitivity of atom probe tomography (APT) to resolve the 3D lattice position of individual dopant atoms in the lightly doped narrow bandgap semiconductor Er(Mn,Ti)O 3 ( E g ≈1.6 eV 16). This additional degree of complexity is not captured by DFT, reflecting the need for an experimental probe that can resolve the individual dopant atoms and, hence, clarify the atomic-scale structure. The DFT calculations, however, are usually performed for the ground state structure without addressing effects that can arise during high-temperature crystal growth where both anions and cations are highly mobile and configurational entropy may favor, e.g., cation anti-sites and vacancies, dopant clustering and non-stoichiometry. Density functional theory (DFT) calculations are often applied to fill this gap and the progress in large-scale DFT modeling is continuously easing size limitations 15 so that lower and lower doping levels can be calculated. Furthermore, it is inherently restricted to 2D projections along specific zone axes, prohibiting the full three-dimensional (3D) characterization of dopant atoms. This correlated approach represents a breakthrough in the atomic-scale characterization of doped oxides, but it is limited to doping levels higher than a few at.%. To image single dopant atoms within the lattice and quantify their concentration, scanning transmission electron microscopy (STEM) has been applied in combination with energy-dispersive X-ray spectroscopy (EDX) 13, 14. Despite their high sensitivity, these measurements cannot be applied to probe small volumes, let alone the lattice position of individual dopants and their interactions at the local scale, as they lack the necessary spatial resolution. For this purpose, different experimental techniques, such as impedance spectroscopy 10, Hall measurements 11, and secondary ion mass spectrometry 12 have been applied, sensing average doping levels down to parts per billion. In order to master this level of complexity and understand emergent composition-driven phenomena and opportunities in oxide materials, a careful characterization of the dopant atoms is crucial. Furthermore, dopants may occupy different regular lattice or interstitial sites with drastically different consequences for the physical properties of the host material 8, 9. For example, dopants can induce local strain and strain gradients, electrostatic fields, orbital reconstruction, and novel magnetic phases 7. Importantly, because of the symmetry reduction and strong electronic correlations, individual dopants can do much more than only control the type and concentration of mobile charge carriers. In contrast to more than 70 years of research on conventional semiconductors, however, the incorporation of dopant atoms in complex oxides is much less explored. Furthermore, in complex oxides, strong correlations between charge, spin, and lattice degrees of freedom arise, promoting a wide variety of additional doping-induced effects, including insulator-metal transitions 4, interfacial magnetism 5, and superconductivity 6. The latter is reflected by doping-dependent studies on hexagonal manganites, where doping with aliovalent cations below 0.05 atomic percent (at.%) resulted in an order of magnitude lower electrical conductivity 3. Analogous to conventional semiconductors 2, very low concentrations of dopant atoms can lead to pronounced changes in the electronic properties of oxide materials. In a more recent development, oxide-based semiconductors moved into focus as a particularly promising class of tunable systems for device applications 1. Despite their substantial impact on the conductivity, the number of dopant atoms is usually small, and even highly doped silicon contains just 1 dopant atom per ∼10 3 Si atoms. The functionality of diodes and transistors, for example, relies on semiconductors where dopant atoms generate the free holes (p-type) or electrons (n-type) that define the transport properties. The engineering of electronic responses with dopant atoms is essential for modern technology. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |