Inter-defect interactions, oxygen-vacancy distribution, and oxidation in acceptor-doped ABO3 perovskites

This study employs statistical theory and Monte Carlo simulations to demonstrate that inter-defect interactions and non-uniform impurity distributions significantly govern defect thermodynamics, oxidation behavior, and hole conductivity in acceptor-doped ABO3 perovskites, with oxygen vacancy-impurity interactions proving more influential than inter-vacancy correlations.

The Gist

Imagine a crystal lattice as a massive, perfectly organized city made of atoms. In this city, the "buildings" are positive ions, and the "streets" are oxygen atoms. Sometimes, to make this city useful for things like clean energy devices, scientists add a few "foreign" residents (impurities) to the mix. These new residents have a different charge, so to keep the city balanced, some of the oxygen atoms have to leave their posts, creating empty spaces called oxygen vacancies.

This paper is like a detailed traffic study of that city. It asks: How do these empty spots (vacancies) and the new foreign residents (impurities) interact with each other? Do they hang out together, avoid each other, or get stuck in traffic jams?

Here is a simple breakdown of what the researchers found, using everyday analogies:

1. The "Best Friend" Effect (Vacancy-Impurity Interaction)

The most important discovery is that the empty spots (vacancies) really like to hang out near the foreign residents (impurities).

• The Analogy: Think of the impurities as popular celebrities and the vacancies as fans. The fans (vacancies) naturally want to sit as close to the celebrities (impurities) as possible.
• The Finding: The paper shows that this "hanging out" is the strongest force in the city. It matters much more than the fans arguing with each other. When a vacancy sits next to one celebrity, it's happy. When it sits between two celebrities, it's even happier. This "hugging" behavior changes how the whole city functions.

2. The "No Double-Booking" Rule (On-Site Correlations)

The researchers looked at what happens when a vacancy tries to sit in a spot that is already crowded.

• The Analogy: Imagine a specific seat in a theater (an oxygen site) that is surrounded by two celebrities. If the vacancy is a "super-fan," it really wants that seat. But there's a rule: Only one fan can sit in that specific seat at a time. You can't have two fans in one chair.
• The Finding: When the city is moderately crowded with fans (moderate doping), this "one seat, one fan" rule becomes very important. It forces the fans to spread out in a specific way, creating a unique pattern that wouldn't exist if they could pile on top of each other.

3. The "Personal Space" Rule (Inter-Site Repulsion)

The study also looked at what happens when two vacancies are neighbors.

• The Analogy: Imagine two fans trying to sit in the two seats right next to each other. Because they are both empty spots (missing oxygen), they repel each other like magnets with the same pole. They refuse to sit side-by-side.
• The Finding: This "personal space" rule becomes very important when the city gets very crowded (high doping). If the city is packed with fans, they can't all hug the celebrities; they have to spread out to avoid bumping into each other. This changes the overall layout of the city.

4. The "Bad Map" Problem (Non-Uniform Distribution)

Sometimes, when the city is built (during sample preparation), the celebrities aren't spread out evenly. They might clump together in one neighborhood and leave another empty.

• The Analogy: Imagine a city where all the celebrities live in the North District, and the South District has none.
• The Finding: The researchers found that this uneven distribution changes where the fans (vacancies) sit. However, it doesn't change the rules of how they interact or the overall "mood" of the city (oxidation) very much. The fans still find the celebrities, even if the map is a bit messy.

5. The "Energy Cost" of Oxidation

Finally, the paper looks at how the city reacts to fresh air (oxygen). This is called "oxidation."

• The Analogy: Imagine the city needs to let new oxygen people in. If the fans (vacancies) are too busy hugging the celebrities (impurities), it becomes harder and more expensive (energetically) to bring new oxygen in.
• The Finding: Because the vacancies are so busy interacting with the impurities, the process of adding oxygen changes. It becomes harder to do, and the amount of "electricity carriers" (holes) the city produces changes in a surprising, non-linear way depending on how many celebrities are in the city.

Why Does This Matter?

The paper concludes that if you want to build better materials for clean energy (like fuel cells), you can't just count the number of ingredients. You have to understand the social dynamics of the atoms:

• Who likes to hang out with whom?
• Who needs personal space?
• How does the crowd size change the rules?

By understanding these "social rules," scientists can better predict how these materials will behave and design them to work more efficiently. The paper confirms that the "hugging" between vacancies and impurities is the main driver of these behaviors, while the "personal space" rules only kick in when the city gets very crowded.

Technical Summary

Technical Summary: Inter-defect interactions, oxygen-vacancy distribution, and oxidation in acceptor-doped ABO3 perovskites

Problem Statement
Acceptor-doped wide-gap perovskites ($AB_{1-x}R_xO_{3-\delta}$) are critical materials for clean energy applications, serving as oxygen, hydrogen, or ionic-electronic conductors in solid oxide electrochemical cells. While the formation of oxygen vacancies via acceptor doping is well-established, the fundamental properties of these materials are significantly influenced by non-ideal behaviors, specifically the interactions between charge carriers (vacancies and holes) and acceptor impurities.

Previous studies have addressed vacancy trapping and hydration, yet fundamental issues regarding heavily doped oxides (where dopant content $x$ can exceed 20%) remain unresolved. Specifically, the implications of inter-defect interactions (vacancy-impurity and vacancy-vacancy) on oxygen-vacancy distribution, local ion coordination, and oxidation thermodynamics are not fully understood. Furthermore, the impact of non-uniform (quenched) impurity distributions, often resulting from sample preparation procedures, on these properties has not been systematically investigated.

Methodology
The authors employ a dual approach combining a developed statistical theory and Monte Carlo (MC) simulations to model acceptor-doped perovskites under dry oxidizing conditions.

1. Statistical Model:

   • The system is modeled with a "frozen" impurity distribution (low cation mobility) and a reduced three-level energy spectrum for oxygen vacancies based on their nearest-neighbor cation environments: $V_f$ (B-V-B), $V_R$ (B-V-R), and $V_{2R}$ (R-V-R).
   • The model incorporates on-site Fermi-type correlations (prohibiting multiple vacancies on a single site) and inter-site vacancy repulsion (infinite repulsion between nearest-neighbor vacancies).
   • Defect thermodynamics are analyzed using the grand partition function, accounting for the distribution of impurity configurations ($\omega$) to calculate averaged observables.
   • The oxidation reaction is treated by deriving the mass action law, incorporating an activity coefficient ($\gamma_V$) to account for non-ideality, and calculating the oxidation enthalpy ($\Delta H_{ox}$).

2. Monte Carlo Simulations:

   • Simulations were conducted on a $30 \times 30 \times 30$ supercell with periodic boundary conditions.
   • Parameters varied included dopant content ($x = 0.05$ to $0.5$) and temperature ($600$ to $1500$ K).
   • The Metropolis algorithm was used to equilibrate vacancy ensembles, considering both weak (on-site) and strong (on-site + inter-site) correlation models.
   • Results were averaged over multiple initial configurations to ensure global thermodynamic equilibrium.

Key Contributions and Results

• Vacancy Distribution and Complex Formation:
  The study demonstrates that under realistic energy parameters, the interaction between oxygen vacancies and impurities generally exerts a greater influence on defect distribution than inter-vacancy correlations.

  • Trapping: Vacancies preferentially occupy sites near impurities. As dopant content ($x$) increases, the fraction of free vacancies decreases rapidly, and bound vacancies ($V_R$ and $V_{2R}$) become dominant.
  • Complexes: The concentration of vacancies surrounded by two impurities ($V_{2R}$) increases monotonically with $x$, becoming the dominant defect type at moderate to high doping levels. Conversely, the concentration of vacancies near a single impurity ($V_R$) exhibits a non-monotonic dependence, peaking at low $x$.

• Role of Correlations:

  • On-site (Fermi-type): These correlations become significant within a narrow doping range at moderate $x$ values, particularly when the ratio of trapping energies ($\Delta E_{V2R}/\Delta E_{VR}$) is high. They lead to a reduction in $V_{2R}$ concentration and an increase in $V_R$.
  • Inter-site (Repulsion): The impact of inter-vacancy repulsion increases with $x$. At sufficiently high doping levels, this repulsion significantly alters the distribution, reducing the formation of $V_{2R}$ complexes and increasing $V_R$. This effect is most pronounced at low temperatures.

• Local Structure and Coordination:

  • Inter-defect interactions lead to non-uniform local coordination. The average coordination number of host B-cations ($Z_{B-O}$) remains close to 6, while impurities ($R$) trap vacancies, reducing their coordination ($Z_{R-O}$) toward 5.
  • Short-range order parameters ($\alpha$) show that at low $x$, deviations from random distribution are driven by Fermi-type correlations, whereas at high $x$, inter-site repulsion dominates.
  • The study validates these findings against experimental $^{45}$Sc NMR data for BaZr$_{1-x}$Sc$_x$O$_{3-\delta}$, showing excellent agreement in the fractions of 5- and 6-coordinated Sc ions.

• Impact of Non-Uniform Impurity Distribution:
  Using a frozen high-temperature dopant configuration (derived from DFT/MC data on Y-doped BaZrO$_3$), the authors found that non-uniform impurity allocation significantly alters the distribution of vacancy types (decreasing $V_{2R}$ and increasing $V_R$). However, this non-uniformity has a weaker effect on short-range order parameters and oxidation behavior compared to the uniform distribution case.

• Oxidation Thermodynamics and Hole Conductivity:

  • Inter-defect interactions reduce the hole concentration and increase the oxidation enthalpy ($\Delta H_{ox}$).
  • The trapping of vacancies lowers the oxidation constant ($K_{ox}$).
  • Crucially, the oxidation enthalpy and hole concentration exhibit a non-trivial dependence on dopant content $x$. Depending on the trapping energy ratios, the hole concentration can show a maximum at low $x$ or increase monotonically.
  • These findings help explain the behavior of hole conductivity in Y- and In-doped BaSnO$_3$, where the activation energy is directly linked to the non-ideal oxidation enthalpy.

Significance and Claims
The paper claims to provide a fundamental understanding of defect interactions in acceptor-doped oxides, moving beyond ideal solution approximations. The authors assert that:

1. Vacancy-impurity interactions are the primary driver of defect thermodynamics and local structure, outweighing inter-vacancy correlations in most realistic scenarios.
2. Non-ideality (specifically Fermi-type correlations and inter-site repulsion) is essential for accurately describing heavily doped systems, particularly at low temperatures.
3. The developed statistical theory and MC simulations successfully reproduce experimental NMR data and explain trends in hole conductivity without invoking complex long-range ordering mechanisms.
4. The model offers a quantitative framework for optimizing dopant type and content to tailor the properties of perovskites for electrochemical applications, such as proton-conducting membranes and solid oxide cells.

The study concludes that while the model has limitations regarding long-range deformation interactions and specific compound-dependent effects, it successfully elucidates the general physical mechanisms governing defect distribution and oxidation in a wide class of acceptor-doped perovskites.