Al-doped SrTiO3: Part I, anomalous oxygen nonstoichiometry

Electronic transport in dopedSrTiO3: Conduction mechanisms and potential applications

“…The material is a semiconductor with a 3.2 eV band gap 1, 32 and can be doped n type [32][33][34][35][36][37][38][39][40][41][42][43][44] or p type. [45][46][47][48][49][50][51] n-type conduction has been most commonly achieved by substitution of La 3+ for Sr 2+ ͑i.e., Sr 1−x La x TiO 3 ͒, 1,33,34 Nb 5+ for Ti 4+ ͑i.e., SrTi 1−y Nb y O 3 ͒, [35][36][37] or by reduction to SrTiO 3−␦ , 32,35,[38][39][40][41][42][43][44] where, in a simple picture, each O vacancy generates two doped electrons. Similarly, p-type doping has been achieved by substituting trivalent metal ions ͑e.g., In 3+ , Al 3+ , Fe 3+ , and Sc 3+ ͒ for Ti 4+ .…”

“…Similarly, p-type doping has been achieved by substituting trivalent metal ions ͑e.g., In 3+ , Al 3+ , Fe 3+ , and Sc 3+ ͒ for Ti 4+ . [45][46][47][48][49][50][51] Electronic-transport measurements were reported as early as 1964, 38 with detailed studies of reduced and Nb-substituted samples appearing from 1967, 32,35,41,42 although they reported significant discrepancies between groups. Even at this early stage, enough was known about the dielectric response to relate certain features of the transport to the large, temperature-dependent, r .…”

Survey of acceptor dopants in SrTiO3: Factors limiting room temperature hole concentration

“…Here, μ e varies gradually from ~2 eV in reducing conditions to ∼1 eV in oxidizing conditions for each dopant. Coincidentally, activation energies extracted from high temperature bulk measurements for Al, Ni, and Fe were all estimated to be near 1 eV with 1.35 eV for Al, 0.96‐1.01 eV for Ni, and 0.94‐1.18 eV for Fe . In the case of Al, Ni, and Fe, each level was thought to correspond to a single impurity related defect with a single ionization level, but the influence of background impurities or multiple ionization levels were not explored.…”

“…Unless significant efforts are made to control the purity of source materials in the synthesis of SrTiO 3 , the nominally undoped bulk hosts a variety of unintentional impurities and native defects that together effectively act as an acceptor‐type dopant . Thus, the electrical response of doped SrTiO 3 includes contributions from these intentional impurities as well as others unintentionally added with the dopant . Despite many attempts, p ‐type SrTiO 3 conductive enough to measure a considerable concentration of holes near room temperature with conventional methods like Hall effect has eluded investigators .…”

“…Such models ignore the complexes thought to be responsible for some observed EPR signatures, the possibility of changing site substitution in different processing conditions, and any unintentional impurities present in the material. Recent studies using a similar approach have attempted to address unintentional impurities parametrically, but alternate site preferences and defect complexes were not considered …”