Origin of Oxygen Partial Pressure-Dependent Conductivity in SrTiO3

Chunlan Ma; Zenghua Cai

arxiv: 2410.20946 · v2 · submitted 2024-10-28 · ❄️ cond-mat.mtrl-sci · physics.comp-ph

Origin of Oxygen Partial Pressure-Dependent Conductivity in SrTiO3

Zenghua Cai , Chunlan Ma This is my paper

Pith reviewed 2026-05-23 18:59 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci physics.comp-ph

keywords SrTiO3intrinsic defectsoxygen vacanciesconductivityoxygen partial pressurefirst-principles calculationsFermi level

0 comments

The pith

Intrinsic defects explain oxygen pressure dependence of conductivity in SrTiO3

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper uses first-principles calculations to identify the intrinsic defects that control how SrTiO3 conductivity varies with oxygen partial pressure. Oxygen vacancies serve as the main donor under oxygen-poor conditions while strontium vacancies act as the primary acceptor. Titanium antisites take over as a key donor under oxygen-rich conditions. The work also accounts for titanium vacancies showing unexpected donor behavior through formation of oxygen trimers. These mechanisms shift the Fermi level to produce metallic, then n-type, then p-type conductivity and thereby resolve the observed pressure dependence.

Core claim

The results reveal that VO, VSr, and TiSr are the dominant intrinsic defects influencing STO's conductivity under varying O chemical potentials. Under O-poor condition, VO is the predominant donor while VSr is the main acceptor. As the oxygen pressure increases, TiSr emerges as a critical donor defect under O-rich condition. VTi typically an acceptor exhibits donor-like behavior due to the formation of O-trimer. This tunes the Fermi level altering STO's conductivity from metallic to n-type and eventually to p-type.

What carries the argument

Formation energies and charge transition levels of intrinsic point defects (oxygen vacancy VO, strontium vacancy VSr, titanium antisite TiSr, titanium vacancy VTi) computed as a function of oxygen chemical potential to determine dominant species and Fermi level position.

If this is right

Oxygen vacancies produce metallic or n-type conductivity under oxygen-poor conditions.
Strontium vacancies compensate donors and enable p-type behavior as oxygen pressure rises.
Titanium antisites sustain n-type character under oxygen-rich conditions.
Oxygen trimer formation allows titanium vacancies to contribute electrons instead of accepting them.
The balance among these four defects accounts for the full sequence of conductivity types.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same defect analysis could be applied to predict conductivity in doped or interface-engineered SrTiO3.
Annealing protocols in controlled oxygen atmospheres could be designed to target specific conductivity regimes.
Detection of titanium antisite concentrations in oxygen-rich samples would provide a direct test.
The oxygen trimer donor effect may appear in other perovskite titanates under similar conditions.

Load-bearing premise

The first-principles calculations correctly predict the relative formation energies and stability of these defects as oxygen chemical potential varies without dominant contributions from impurities or kinetic barriers.

What would settle it

Experimental absence of titanium antisite defects acting as donors in oxygen-rich pure SrTiO3 samples or failure to observe p-type conductivity at high oxygen pressures would falsify the proposed defect mechanism.

Figures

Figures reproduced from arXiv: 2410.20946 by Chunlan Ma, Zenghua Cai.

**Figure 4.** Figure 4: Based on the results of EF, the range of O chemical potential can be classified into three typical regions. The first region (R1) is where the EF is located above the band gap (Eg). In this region, EF goes beyond the CBM and STO shows metallic conductivity when the O chemical potential is poor (oxygen partial pressure is low). At this moment, the density of intrinsic electron carrier is extremely high, cl… view at source ↗

**Figure 3.** Figure 3: (a) The ground-state structure of neutral VTi; (b) and (c) are the bonding situation before and after forming neutral VTi; (d) The calculated transition levels of intrinsic defects in the band gap of STO [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗

**Figure 4.** Figure 4: The calculated Fermi level (EF), intrinsic carrier densities, and density of low-energy intrinsic defects in STO as function of O chemical potential (The specific chemical potential values of O-rich and O-poor are the same as that in [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗

read the original abstract

SrTiO3 (STO) displays a broad spectrum of physical properties, including superconductivity, ferroelectricity, and photoconductivity, making it a standout semiconductor material. Despite extensive researches, the oxygen partial pressure-dependent conductivity in STO has re-mained elusive. This study leverages first-principles calculations, and systematically investigates the intrinsic defect properties of STO. The results reveal that VO, VSr, and TiSr are the dominant intrinsic defects, influencing STO's conductivity under varying O chemical potentials (oxygen partial pressures). Under O-poor condition, VO is the predominant donor, while VSr is the main acceptor. As the oxygen pressure increases, TiSr emerges as a critical donor defect under O-rich condition, significantly affecting conductivity. Additionally, the study elucidates the abnormal phenomenon where VTi, typically an acceptor, exhibits donor-like behavior due to the formation of O-trimer. This work offers a comprehensive understanding of how intrinsic defects tune the Fermi level, thereby altering STO's conductivity from metallic to n-type, and eventually to p-type across different O chemical potentials. These insights resolve the long-standing issue of oxygen partial pressure-dependent conductivity and explain the observed metallic conductivity in oxygen-deficient STO.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper assigns VO, VSr, TiSr, and an O-trimer at VTi to explain STO's conductivity shift with oxygen pressure, but supplies no DFT details so the assignments cannot be checked.

read the letter

The central result is that first-principles defect calculations identify VO as the dominant donor under O-poor conditions, VSr as the main acceptor, TiSr as the key donor under O-rich conditions, and VTi forming an O-trimer that flips it to donor-like behavior. This sequence is used to map the metallic-to-n-type-to-p-type transition as oxygen partial pressure rises. The O-trimer mechanism and the prominence of TiSr under O-rich conditions are the two concrete assignments that go beyond generic defect lists in the literature. The work is otherwise a standard application of formation-energy calculations to a well-studied material. The abstract gives no functional, supercell size, finite-size correction scheme, or band-gap treatment, so it is impossible to tell whether the reported dominance order survives the usual sensitivities of oxide defect calculations. The stress-test concern about functional dependence therefore lands directly: a different hybrid or a different correction can reorder the formation energies and move the Fermi-level pinning. No comparison of calculated defect concentrations to measured conductivities appears, and extrinsic impurities or kinetics are not discussed. The paper is aimed at researchers who need a defect map for STO processing. A reader already working on titanate defect chemistry could extract the proposed defect ordering for discussion, but the lack of methodological transparency limits how far the claims can be taken without the full text. It is worth sending to referees so the methods and any validation can be examined.

Referee Report

3 major / 2 minor

Summary. The manuscript uses first-principles calculations to investigate intrinsic point defects in SrTiO3 and concludes that VO, VSr, and TiSr are the dominant defects controlling conductivity as a function of oxygen chemical potential. Under O-poor conditions VO acts as the primary donor while VSr is the main acceptor; under O-rich conditions TiSr becomes the key donor; VTi is reported to show donor-like behavior via O-trimer formation. These defects are said to tune the Fermi level, producing a metallic-to-n-type-to-p-type conductivity sequence with increasing pO2 and thereby resolving the long-standing experimental puzzle.

Significance. If the computed formation energies and charge transition levels are robust, the work supplies a microscopic defect-physics account of pO2-dependent conductivity in STO, including the metallic regime in oxygen-deficient samples. It explicitly ranks multiple intrinsic defects under varying μ_O and identifies an unusual donor-like role for VTi, which could guide future defect-engineering studies.

major comments (3)

[Methods] Methods section: the exchange-correlation functional, supercell size, k-point mesh, and finite-size correction scheme (e.g., Freysoldt or Kumagai) applied to charged-defect formation energies E_f(q, μ_O) are not reported. These choices directly determine the relative stabilities of VO^{2+}, VSr^{2-}, and TiSr that are required for the claimed dominance ordering and the metallic → n-type → p-type sequence.
[Results] Results section on formation-energy diagrams: no sensitivity test or comparison against hybrid functionals or alternative band-gap corrections is presented. Because the Fermi-level pinning position and the crossover between VO and TiSr dominance are known to shift with these parameters, the central claim that the calculated energies produce the observed conductivity transitions cannot be assessed.
[Discussion] Discussion of VTi + O-trimer: the structural model, charge-state sequence, and explicit formation-energy values that establish the donor-like behavior of VTi are not shown in sufficient detail (no equation or figure reference is given for the O-trimer configuration or its transition levels). This mechanism is load-bearing for the p-type regime under O-rich conditions.

minor comments (2)

[Abstract] Abstract contains the typo 're-mained' (should be 'remained').
[Throughout] Notation for defect species (VO, VSr, TiSr, VTi) is used without an initial definition table; a compact summary table of formation energies versus μ_O would improve clarity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed and constructive comments, which highlight areas where additional clarity will strengthen the manuscript. We address each major point below and will revise the manuscript to incorporate the requested information and details.

read point-by-point responses

Referee: [Methods] Methods section: the exchange-correlation functional, supercell size, k-point mesh, and finite-size correction scheme (e.g., Freysoldt or Kumagai) applied to charged-defect formation energies E_f(q, μ_O) are not reported. These choices directly determine the relative stabilities of VO^{2+}, VSr^{2-}, and TiSr that are required for the claimed dominance ordering and the metallic → n-type → p-type sequence.

Authors: We agree that these methodological details are necessary for full reproducibility and evaluation of the results. The original submission omitted an expanded Methods section; in the revision we will add explicit statements of the PBE functional, 3×3×3 supercell, Γ-centered k-mesh, and the finite-size correction scheme (Freysoldt) used for all charged-defect formation energies. revision: yes
Referee: [Results] Results section on formation-energy diagrams: no sensitivity test or comparison against hybrid functionals or alternative band-gap corrections is presented. Because the Fermi-level pinning position and the crossover between VO and TiSr dominance are known to shift with these parameters, the central claim that the calculated energies produce the observed conductivity transitions cannot be assessed.

Authors: We recognize that quantitative pinning positions can shift with functional choice. Our PBE results already reproduce the experimental metallic-to-n-type-to-p-type sequence and the dominance ordering of VO, VSr, and TiSr; however, we will add a short paragraph in the revised Results/Discussion discussing the expected robustness of the qualitative ordering under hybrid functionals and note that the experimental trends are captured without adjustable parameters. revision: partial
Referee: [Discussion] Discussion of VTi + O-trimer: the structural model, charge-state sequence, and explicit formation-energy values that establish the donor-like behavior of VTi are not shown in sufficient detail (no equation or figure reference is given for the O-trimer configuration or its transition levels). This mechanism is load-bearing for the p-type regime under O-rich conditions.

Authors: We thank the referee for this observation. The O-trimer configuration and its effect on VTi charge states are presented in the Results, but the discussion will be expanded to include an explicit structural description, the relevant charge-transition levels, and a direct reference to the supporting figure and formation-energy equation so that the donor-like behavior under O-rich conditions is fully documented. revision: yes

Circularity Check

0 steps flagged

No circularity: conclusions follow from direct first-principles defect calculations

full rationale

The paper computes formation energies E_f(q, μ_O), charge transition levels, and defect concentrations via supercell DFT as explicit functions of oxygen chemical potential. Dominant defects (VO under O-poor, VSr, TiSr under O-rich, VTi+O-trimer) and the resulting metallic-to-n-type-to-p-type conductivity sequence are read off from these computed quantities without any fitting to measured conductivity data or self-citation chains that would reduce the claims to inputs by construction. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the accuracy of standard density-functional-theory approximations for defect formation energies in an oxide; no free parameters, invented entities, or ad-hoc axioms beyond routine DFT assumptions are stated in the abstract.

axioms (1)

domain assumption Density functional theory with chosen exchange-correlation functional and pseudopotentials yields reliable formation energies and charge states for intrinsic point defects in SrTiO3 as a function of oxygen chemical potential.
The entire analysis is built on first-principles defect calculations whose validity is presupposed.

pith-pipeline@v0.9.0 · 5741 in / 1450 out tokens · 33594 ms · 2026-05-23T18:59:29.449001+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

7 extracted references · 7 canonical work pages

[1]

1J. F. Schooley, W. R. Hosler, and M. L. Cohen, Superconductivity in Semiconducting SrTiO3. Phys. Rev. Lett. 12, 474 (1964). 2J. H. Haeni, P. Irvin, W. Chang, R. Uecker, P. Reiche, Y. L. Li, S. Choudhury, W. Tian, M. E. Hawley, B. Craigo, A. K. Tagantsev, X. Q. Pan, S. K. Streiffer, L. Q. Chen, S. W. Kirchoefer, J. Levy, and D. G. Schlom, Room-temperature...

work page 1964
[2]

J. Appl. Phys. 127, 245702 (2020). 18M. Siebenhofer, F. Baiutti, J. de Dios Sirvent, T. M. Huber, A. Viernstein, S. Smetaczek, C. Herzig, M. O. Liedke, M. Butterling, A. Wagner, E. Hirschmann, A. Limbeck, A. Tarancon, J. Fleig, and M. Kubicek, Exploring point defects and trap states in undoped SrTiO3 single crystals. J. Eur. Ceram. Soc. 42, 1510 (2022). 1...

work page 2020
[3]

Appl. Phys. Lett. 110, 263902 (2017). 22Y. Jin, F. Zhang, K. Zhou, C. H. Suen, X. Y. Zhou, and J. -Y. Dai, Oxygen vacancy and photoelectron enhanced flexoelectricity in perovskite SrTiO 3 crystal. Appl. Phys. Lett. 118, 164101 (2021). 23A. Stashans and L. Villamagua, Schottky defects in cubic lattice of SrTiO

work page 2017
[4]

J. Phys. Chem. Solids 70, 417 (2009). 24J. N. Baker, P. C. Bowes, J. S. Harris, and D. L. Irving, Mechanisms governing metal vacancy formation in BaTiO3 and SrTiO3. J. Appl. Phys. 124, 114101 (2018). 25A. Janotti, J. B. Varley, M. Choi, and C. G. Van de Walle, Vacancies and small polarons in SrTiO3. Phys. Rev. B 90, 085202 (2014). 26R. A. Mackie, S. Singh...

work page 2009
[5]

Appl. Phys. Lett. 112, 022902 (2018). 40Z. Zhang and A. Janotti, Cause of Extremely Long -Lasting Room -Temperature Persistent Photoconductivity in SrTiO3 and Related Materials. Phys. Rev. Lett. 125, 126404 (2020). 41M. Choi, F. Oba, and I. Tanaka, Role of Ti antisite like defects in SrTiO

work page 2018
[6]

Phys. Rev. Lett. 103, 185502 (2009). 42O. O. Brovko and E. Tosatti, Controlling the magnetism of oxygen surface vacancies in SrTiO3 through charging. Phys. Rev. Mater. 1, 044405 (2017). 43K. Klyukin and V. Alexandrov, Effect of intrinsic point defects on ferroelectric polarization behavior of SrTiO3. Phys. Rev. B 95, 035301 (2017). 44P. Reunchan, N. Umeza...

work page 2009
[7]

Phys. Rev. Mater. 2, 060403 (2018). 51K. Szot, W. Speier, R. Carius, U. Zastrow, and W. Beyer, Localized Metallic Conductivity and Self-Healing during Thermal Reduction of SrTiO3. Phys. Rev. Lett. 88, 075508 (2002). 52C. Nath, C. Y. Chueh, Y. K. Kuo, and J. P. Singh, Thermoelectric properties of p -type SrTiO3/graphene layers nanohybrids. J. Appl. Phys. 1...

work page 2018

[1] [1]

1J. F. Schooley, W. R. Hosler, and M. L. Cohen, Superconductivity in Semiconducting SrTiO3. Phys. Rev. Lett. 12, 474 (1964). 2J. H. Haeni, P. Irvin, W. Chang, R. Uecker, P. Reiche, Y. L. Li, S. Choudhury, W. Tian, M. E. Hawley, B. Craigo, A. K. Tagantsev, X. Q. Pan, S. K. Streiffer, L. Q. Chen, S. W. Kirchoefer, J. Levy, and D. G. Schlom, Room-temperature...

work page 1964

[2] [2]

J. Appl. Phys. 127, 245702 (2020). 18M. Siebenhofer, F. Baiutti, J. de Dios Sirvent, T. M. Huber, A. Viernstein, S. Smetaczek, C. Herzig, M. O. Liedke, M. Butterling, A. Wagner, E. Hirschmann, A. Limbeck, A. Tarancon, J. Fleig, and M. Kubicek, Exploring point defects and trap states in undoped SrTiO3 single crystals. J. Eur. Ceram. Soc. 42, 1510 (2022). 1...

work page 2020

[3] [3]

Appl. Phys. Lett. 110, 263902 (2017). 22Y. Jin, F. Zhang, K. Zhou, C. H. Suen, X. Y. Zhou, and J. -Y. Dai, Oxygen vacancy and photoelectron enhanced flexoelectricity in perovskite SrTiO 3 crystal. Appl. Phys. Lett. 118, 164101 (2021). 23A. Stashans and L. Villamagua, Schottky defects in cubic lattice of SrTiO

work page 2017

[4] [4]

J. Phys. Chem. Solids 70, 417 (2009). 24J. N. Baker, P. C. Bowes, J. S. Harris, and D. L. Irving, Mechanisms governing metal vacancy formation in BaTiO3 and SrTiO3. J. Appl. Phys. 124, 114101 (2018). 25A. Janotti, J. B. Varley, M. Choi, and C. G. Van de Walle, Vacancies and small polarons in SrTiO3. Phys. Rev. B 90, 085202 (2014). 26R. A. Mackie, S. Singh...

work page 2009

[5] [5]

Appl. Phys. Lett. 112, 022902 (2018). 40Z. Zhang and A. Janotti, Cause of Extremely Long -Lasting Room -Temperature Persistent Photoconductivity in SrTiO3 and Related Materials. Phys. Rev. Lett. 125, 126404 (2020). 41M. Choi, F. Oba, and I. Tanaka, Role of Ti antisite like defects in SrTiO

work page 2018

[6] [6]

Phys. Rev. Lett. 103, 185502 (2009). 42O. O. Brovko and E. Tosatti, Controlling the magnetism of oxygen surface vacancies in SrTiO3 through charging. Phys. Rev. Mater. 1, 044405 (2017). 43K. Klyukin and V. Alexandrov, Effect of intrinsic point defects on ferroelectric polarization behavior of SrTiO3. Phys. Rev. B 95, 035301 (2017). 44P. Reunchan, N. Umeza...

work page 2009

[7] [7]

Phys. Rev. Mater. 2, 060403 (2018). 51K. Szot, W. Speier, R. Carius, U. Zastrow, and W. Beyer, Localized Metallic Conductivity and Self-Healing during Thermal Reduction of SrTiO3. Phys. Rev. Lett. 88, 075508 (2002). 52C. Nath, C. Y. Chueh, Y. K. Kuo, and J. P. Singh, Thermoelectric properties of p -type SrTiO3/graphene layers nanohybrids. J. Appl. Phys. 1...

work page 2018