Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2: A First-Principles Study

arXiv:2606.25408 · cond-mat.mtrl-sci · Submitted 2026-06-24 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2".

Mira: Here's the gist of the paper. Lithium carbonate is thermodynamically stable and exhibits poor ionic and electronic conductivities, making its stability and decomposition behavior closely related to defect formation and charge transport properties.

Kai: This study first-principles calculations were performed to systematically analyze the intrinsic point-defect energetics of Li2CO3 and to evaluate the effects of fluorine doping on vacancy formation energies in Li2CO3 and Li2O2.

Mira: Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects.

Kai: Intrinsic Defect Analysis The intrinsic defect energetics of pristine Li2CO3 were evaluated by considering both neutral and charged defects.

Title and authors: Kai: So, focusing on the title and authors of this paper, "Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2: A First-Principles Study," we see a very specific focus on how those defects interact with fluorine.

Mira: The authors are taking the stability of lithium carbonate, which is already known to be stable, and they are testing what happens when you dope it with fluorine to see how that changes the defect energetics for both Li two CO three and Li two O two <ref:2606.25408#pg1>.

Lev: From an engineering standpoint, the point is that if we can predict how doping affects vacancy formation energy, we get a handle on whether adding fluorine helps us control the decomposition pathway.

Kai: It’s about moving past just saying Li two CO three decomposes to understanding exactly how much easier or harder it becomes to break apart when you introduce fluorine at the oxygen sites <ref:2606.25408#pg1>.

Mira: The key thing is that they are systematically analyzing intrinsic point defects and evaluating the effects of fluorine doping on vacancy formation energies in both Li two CO three and Li two O two to see how stability changes <ref:2606.25408#pg1>.

Lev: It’s a targeted study because it zeroes in on lithium-related defects as the main drivers of this behavior, which is more useful for designing actual batteries.

The paper's summary: Kai: So, what's the high-level summary they give us? They start by saying that Li two CO three is thermodynamically stable, but its stability and how it breaks apart are really linked to those defect formation and charge transport properties <ref:2606.25408#pg1>.

Mira: The study first does intrinsic defect analysis on pristine Li two CO three finding that the behavior is predominantly governed by lithium-related defects across the whole system <ref:2606.25408#pg1>.

Lev: Then they move to fluorine doping, and what they found is that fluorine selectively lowers the formation energies for lithium and carbon vacancies in Li two CO three but oxygen vacancy formation energy stays pretty much unchanged <ref:2606.25408#pg1>.

Kai: That selective lowering of lithium and carbon vacancies suggests that the carbonate framework itself gets partially destabilized when you substitute fluorine, which is a really important detail to hear.

Mira: And they quantify this by showing that the lithium vacancy formation energy drops from one point six three electron volts to negative one point seven five electron volts when fluorine substitutes at the oxygen site and we are at three point eight two volt.

Lev: That drop of about three point four eight electron volts is significant because it shows a substantial thermodynamic cost reduction for removing lithium, which is what you want to avoid in a battery system.

The paper's improvements: Kai: The paper points out some important improvements they made to the analysis, especially concerning how they modeled these defects and the voltage dependence of them.

Mira: They introduced neutral vacancy defects into both pristine and fluorine-doped structures for each system, but they also used complex vacancy defects like VLi two CO three zero using direct supercell calculations <ref:2606.25408#pg1>.

Lev: That’s good because it shows how the system behaves when you have clustered vacancies, which is a much more realistic scenario than just looking at single empty spots.

Kai: And they also noted that lithium and carbon vacancy formation energies show voltage dependence, while oxygen vacancy formation energies remain voltage independent across the applied cell voltage.

Mira: That distinction is key because it suggests that lithium extraction accompanied by electronic compensation might be thermodynamically favored during charging, whereas oxygen vacancies don't behave that way.

Lev: If we are building real hardware, knowing this voltage dependence means we have to account for the charge state of the defects when calculating the actual kinetic barriers for decomposition.

Conclusion: Kai: So to wrap up on "Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2: A First-Principles Study," they conclude that fluorine doping modifies the defect energetics through a richer set of mechanisms than in Li two O two <ref:2606.25408#pg1,Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2: A>.

Mira: They emphasize that this is not just about lowering lithium vacancy formation, but it also reduces the carbon vacancy formation energy, which is something you don't see in Li two O two <ref:2606.25408#pg1>.

Lev: The big implication here for the world is that fluorine doping seems to destabilize both discharge products through a common mechanism of destabilizing the cation sublattice, and it actually weakens the carbonate framework in Li two CO three too <ref:2606.25408#pg1>.

Kai: It means we might be able to use these same fluorine doping strategies that work for lithium peroxide to also control lithium carbonate decomposition pathways.

Mira: They suggest that this could help reduce the thermodynamic cost associated with forming those lithium deficient Li two-x O two phases, which is really relevant for performance degradation in LOBs <ref:2606.25408#pg1>.

Lev: I just want to say that while this study shows a mechanism, the authors themselves flag that they are still looking at intrinsic defects and haven't fully incorporated polaron dynamics or kinetic barriers yet to explain the full decomposition process under realistic electrochemical conditions.

Department of Materials Engineering, The University of Tokyo · Institute of Industrial Science, the University of Tokyo · National Institute for Fusion Science, Gifu 509-5292 · Research Center for Solar Energy Chemistry, Graduate School of Engineering Science, The University of Osaka · Innovative Catalysis Science Division, Institute for Open Transdisciplinary Research Initiatives

cond-mat.mtrl-sci

Submitted: 2026-06-24

Updated: 2026-11-13

Comments: 41 pages, 10 figures

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 86/100

The gist: Lithium carbonate is thermodynamically stable and exhibits poor ionic and electronic conductivities, making its stability and decomposition behavior closely related to defect formation and charge…

Key concepts

Defect Formation Energy
This measures the energy cost required to create a specific defect (like a missing atom) within the crystal structure. Lower formation energies mean that defects are more likely to form under certain conditions, which dictates the material's stability and decomposition behavior.
Fluorine Doping Effects
Introducing fluorine into Li2CO3 selectively lowers the energy needed to create lithium and carbon vacancies, but has little effect on oxygen vacancies. This indicates that fluorine strongly perturbs the local environment around lithium sites, affecting cation stability more than anion stability.
Voltage Dependence of Defects
The formation energies of lithium and carbon vacancies change depending on the cell voltage (U). This means that during battery operation, the thermodynamic favorability of removing lithium is linked to the applied voltage, allowing AI to predict which defect mechanism dominates under specific electrochemical states.

Terminology

Summary

Lithium carbonate is thermodynamically stable and exhibits poor ionic and electronic conductivities, making its stability and decomposition behavior closely related to defect formation and charge transport properties. p. 2

Where the paper says it

1–9 Due to the robust bonding network of carbonate units, Li2CO3 is thermodynamically stable and exhibits poor ionic and electronic conductivities. 7 These characteristics make Li2CO3 difficult to decompose electrochemically once formed, and its stability and decomposition behavior are closely related to defect formation and charge transport properties.

This study first-principles calculations were performed to systematically analyze the intrinsic point-defect energetics of Li2CO3 and to evaluate the effects of fluorine doping on vacancy formation energies in Li2CO3 and Li2O2. p. 1
Where the paper says it

However, compared with Li2O2, the intrinsic defect thermodynamics of Li2CO3 and how chemical doping modifies its defect energetics remain insufficiently understood. In this study, first-principles calculations were performed to systematically analyze the intrinsic point-defect energetics of Li2CO3 and to evaluate the effects of fluorine doping on vacancy formation energies in Li2CO3 and Li2O2.

Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. p. 1
Where the paper says it

In this study, first-principles calculations were performed to systematically analyze the intrinsic point-defect energetics of Li2CO3 and to evaluate the effects of fluorine doping on vacancy formation energies in Li2CO3 and Li2O2. Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects.

Intrinsic Defect Analysis

The intrinsic defect energetics of pristine Li2CO3 were evaluated by considering both neutral and charged defects. The formation energies of intrinsic point defects in bulk Li2CO3 were calculated as a function of the Fermi level across its band gap (7. p. 9

Where the paper says it

(b) is a carbonate unit. The formation energies of intrinsic point defects in bulk Li2CO3 were calculated as a function of the Fermi level across its band gap (~7.

98 eV) as shown in Figure 2. p. 9
Where the paper says it

The formation energies of intrinsic point defects in bulk Li2CO3 were calculated as a function of the Fermi level across its band gap (~7. 98 eV) as shown in Figure 2.

Under Li-rich conditions at 0 V, Lii+ and VCO20 exhibit relatively low formation energies. p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

At 3.82 V, corresponding to the theoretical equilibrium voltage of Li2CO3, VLi− becomes the dominant defect species. p. 7
Where the paper says it

This voltage-dependent definition of 𝜇𝐿𝑖 was applied to both discharge products. The equilibrium voltages of Li2O2 and Li2CO3 were used as reference electrochemical conditions for evaluating voltage-dependent vacancy formation energies.

Fluorine Doping Effects on Stability

Fluorine doping selectively decreases the lithium and carbon vacancies formation energies in Li2CO3, whereas oxygen vacancy formation is only weakly affected. The reduction in lithium vacancy formation energy for fluorine substitution at the O(f) site was from 1. p. 1

Where the paper says it

Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2.

63 eV to −1.75 eV (ΔE ≈ −3.38 eV) at 3.82 V. Simultaneously, the carbon vacancy formation energy dropped from 5. p. 1
Where the paper says it

Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2.

18 eV to −2.67 eV (ΔE ≈ −7.85 eV) for fluorine substitution at the O(f) site. p. 3
Where the paper says it

29 In this respect, fluorine differs from many other dopants that would require deliberate synthetic incorporation. Second, owing to its similar ionic radius to oxygen and its exceptional electronegativity, fluorine substitution at oxygen sites is expected to selectively perturb the local bonding environment around lithium ions, potentially modifying cation sublattice stability without wholesale structural disruption.

This decrease suggests a partial weakening of the framework’s thermodynamic stability. p. 1
Where the paper says it

Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium– oxygen batteries.

Comparative Doping Mechanisms

The magnitude of the fluorine doping effect on lithium vacancy formation was approximately twice as large in Li2CO3 (ΔE ≈ −3. p. 1

Where the paper says it

In this study, first-principles calculations were performed to systematically analyze the intrinsic point-defect energetics of Li2CO3 and to evaluate the effects of fluorine doping on vacancy formation energies in Li2CO3 and Li2O2. Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects.

4 eV to −3.9 eV) as in Li2O2 (ΔE ≈ −1. p. 0
Where the paper says it

1 Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2: A First-Principles Study Youjeong Choi1 , Tasuku Sugiura2 , Keisuke Mukai3 , Nanako Ishihara4 , Shuji Nakanishi4, 5, and Teruyasu Mizoguchi1, 2, * 1 Department of Materials Engineering, The University of Tokyo, Tokyo 113-8656, Japan 2 Institute of Industrial Science, The University of Tokyo, Tokyo 153-8505, Japan 3 National Institute for Fusion Science, Gifu 509-5292, Japan 4 Research Center for Solar Energy Chemistry, Graduate School of Engineering Science, The University of Osaka, Osaka 560-8531, Japan 5 Innovative Catalysis Science Division, Institute for Open Transdisciplinary Research Initiatives, The University of Osaka, Osaka 565-0871, Japan *Corresponding author. E-mail: teru@iis.

p. 0
Where the paper says it

1 Intrinsic Defect Energetics and Fluorine Doping Effects in Li2CO3 and Li2O2: A First-Principles Study Youjeong Choi1 , Tasuku Sugiura2 , Keisuke Mukai3 , Nanako Ishihara4 , Shuji Nakanishi4, 5, and Teruyasu Mizoguchi1, 2, * 1 Department of Materials Engineering, The University of Tokyo, Tokyo 113-8656, Japan 2 Institute of Industrial Science, The University of Tokyo, Tokyo 153-8505, Japan 3 National Institute for Fusion Science, Gifu 509-5292, Japan 4 Research Center for Solar Energy Chemistry, Graduate School of Engineering Science, The University of Osaka, Osaka 560-8531, Japan 5 Innovative Catalysis Science Division, Institute for Open Transdisciplinary Research Initiatives, The University of Osaka, Osaka 565-0871, Japan *Corresponding author. E-mail: teru@iis.

4 eV to −1.7 eV). This difference likely reflects the stronger perturbation induced by fluorine substitution within the rigid carbonate framework. p. 5
Where the paper says it

In contrast, carbon vacancy formation leads to a slight increase in magnetization, suggesting further perturbation of the local electronic structure within the carbonate framework. Oxygen vacancy formation results in only a minor change in magnetization, corroborating the selective nature of fluorine doping toward cation vacancy formation rather than anion vacancy formation.

Furthermore, Li2CO3 exhibits an additional fluorine doping effect that has no counterpart in Li2O2: a dramatic reduction in carbon vacancy formation energy (ΔE ≈ −7. p. 1
Where the paper says it

Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2.

9 eV).

Conclusion and Implications

Fluorine doping modifies the defect energetics of Li2CO3 through a richer set of mechanisms than in Li2O2, including a reduction in carbon vacancy formation energy that is absent in Li2O2. These results suggest that fluorine doping not only facilitates lithium removal but may also partially destabilize the carbonate network itself, potentially providing an additional decomposition pathway. p. 1

Where the paper says it

Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium– oxygen batteries.

Together with the results for Li2O2, these findings suggest that fluorine doping modulates the thermodynamic stability of both discharge products through a common mechanism of cation sublattice destabilization, while additionally destabilizing the carbonate framework in Li2CO3. p. 1
Where the paper says it

Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium– oxygen batteries.

These results are consistent with the possibility that fluorine doping strategies explored for Li2O2 may also be applicable to Li2CO3. p. 3
Where the paper says it

Third, while halogen doping effects including fluorine have been theoretically explored in Li2O2. 19,20 a systematic investigation of fluorine doping in Li2CO3 remains absent.

The paper concludes that lithium-related defects are the dominant intrinsic defects in Li2CO3, and decomposition is more closely related to lithium removal than to direct carbonate framework dissociation through bulk point defects. p. 1
Where the paper says it

Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2.

The future work will incorporate charge-state-dependent defect behavior, polaron dynamics, and kinetic barriers to fully elucidate the decomposition mechanism under realistic electrochemical conditions. p. 2
Where the paper says it

7 These characteristics make Li2CO3 difficult to decompose electrochemically once formed, and its stability and decomposition behavior are closely related to defect formation and charge transport properties. Therefore, understanding the intrinsic defect energetics of Li2CO3 is crucial for clarifying the thermodynamic factors governing its stability across lithium-based electrochemical systems.

REFERENCE

B3pg3

B5pg5

B7pg7

B11pg11

B13pg13

F2pg8

F9pg8

Sufficient Data for Analysis

The study utilized density functional theory (DFT) as implemented in the Vienna Ab initio Simulation Package (VASP) for all calculations. The electron–ion interactions were described by the projector augmented-wave (PAW) method. p. 4

Where the paper says it

METHODS DFT calculations and structure formation All calculations were performed within the framework of density functional theory (DFT) as implemented in the Vienna Ab initio Simulation Package (VASP). 36–38 The electron–ion interactions were described by the projector augmented-wave (PAW) method.

For structural relaxations, the exchange-correlation energy was treated using the Perdew–Burke–Ernzerhof (PBE) functional within the generalized gradient approximation (GGA). p. 4
Where the paper says it

36–38 The electron–ion interactions were described by the projector augmented-wave (PAW) method. 39 For structural relaxations, the exchange-correlation energy was treated using the Perdew–Burke–Ernzerhof (PBE) functional within the generalized gradient approximation (GGA).

The plane-wave cutoff energy was set to 520 eV, and Brillouin zone integrations were performed using Γ-centered k-point grids generated with a reciprocal-space resolution of 0. p. 4
Where the paper says it

40 For Li, the Li_sv PAW potential including both 1s and 2s states in the valence was employed, whereas default PAW datasets were used for O, C, and F. The plane-wave cutoff energy was set to 520 eV, and Brillouin zone integrations were performed using Γ-centered k-point grids generated with a reciprocal-space resolution of 0.

025 Å−1. p. 4
Where the paper says it

The plane-wave cutoff energy was set to 520 eV, and Brillouin zone integrations were performed using Γ-centered k-point grids generated with a reciprocal-space resolution of 0. 025 Å-1 .

Spin polarization was included in all calculations, as its role becomes evident upon fluorine substitution in Li2CO3. p. 4
Where the paper says it

005 eV/Å for the residual ionic forces. Spin polarization was included in all calculations, as its role becomes evident upon fluorine substitution in Li2CO3.

Chemical Potential Definitions

The lithium chemical potential was defined to account for the electrochemical conditions of LOBs environment. p. 7

Where the paper says it

𝜇𝑖 is the chemical potential of species i. The lithium chemical potential was defined to account for the electrochemical conditions of LOBs environment.

The lithium chemical potential depends on U and was expressed as: muLi(U) = muLi bulk − eU (2). p. 7
Where the paper says it

In this environment, the applied cell voltage U exists between the lithium metal anode and the discharge products formed at the cathode. Accordingly, the lithium chemical potential depends on U and was expressed as:48 𝜇Li(𝑈) = 𝜇Li bulk − e𝑈 (2) where 𝜇Li bulk is the energy per atom of bulk lithium metal at 𝑈 = 0, and e is the elementary charge.

The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations. p. 7
Where the paper says it

96 V vs Li+/Li (3) 4Li+ + 4e − + 2CO2 + O2 ⇌ 2Li2CO3, 𝑈eq = 3. 82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1.

The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3. p. 7
Where the paper says it

82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1. The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3.

p. 7
Where the paper says it

82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1. The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3.

The fluorine chemical potential was primarily defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF. p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

Phase Diagram Constraints

The chemical potentials used for the voltage-dependent analysis and those obtained from the Li–C–O phase diagram are summarized in Table S1(a) and Table S1(b). The Li–C–O phase diagram is shown in Figure S1. p. 9

Where the paper says it

The formation energies of intrinsic point defects in bulk Li2CO3 were calculated as a function of the Fermi level across its band gap (~7. 98 eV) as shown in Figure 2.

p. 9
Where the paper says it

The formation energies of intrinsic point defects in bulk Li2CO3 were calculated as a function of the Fermi level across its band gap (~7. 98 eV) as shown in Figure 2.

Condition E denotes the equilibrium phases Li2CO3–Li2O–Li2O2. p. 7
Where the paper says it

82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1. The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3.

p. 7
Where the paper says it

82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1. The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3.

Defect Formation Energy Definition

The defect formation energy of a defect X in charge state q was defined as follows:Ef[Xq] = Etot[Xq] − Etot[ref] −∑nimui + qEF + Ecorr (1). The reference supercell is the pristine supercell for intrinsic defects and the fluorinesubstituted supercell for vacancy defects in doped systems. p. 7

Where the paper says it

8 𝐸 𝑓 [𝑋 𝑞 ] = 𝐸tot [𝑋 𝑞 ] − 𝐸tot[ref] −∑𝑛𝑖𝜇𝑖 𝑖 + 𝑞𝐸F + 𝐸corr (1) Where 𝐸tot[𝑋 𝑞 ] and 𝐸tot[ref] denote the total energies of the defective and reference supercells, respectively. The reference supercell is the pristine supercell for intrinsic defects and the fluorinesubstituted supercell for vacancy defects in doped systems.

The lithium chemical potential was defined to account for the electrochemical conditions of LOBs environment. The applied cell voltage U exists between the lithium metal anode and the discharge products formed at the cathode. p. 7
Where the paper says it

The lithium chemical potential was defined to account for the electrochemical conditions of LOBs environment. In this environment, the applied cell voltage U exists between the lithium metal anode and the discharge products formed at the cathode.

p. 7
Where the paper says it

The lithium chemical potential was defined to account for the electrochemical conditions of LOBs environment. In this environment, the applied cell voltage U exists between the lithium metal anode and the discharge products formed at the cathode.

Chemical Potential Calculations

The O2 correction procedure was expressed as follows: ∆Hfcal(Li2O2) = EDFT(Li2O2) − 2 EDFT(Libulk) − EDFT (O2). The DFT energy of O2 was increased by 0. p. 4

Where the paper says it

This work offers thermodynamic insight into how fluorine doping modifies the defect energetics of Li2CO3 and Li2O2, which may inform doping design strategies for controlling the stability of lithium-based discharge products. METHODS DFT calculations and structure formation All calculations were performed within the framework of density functional theory (DFT) as implemented in the Vienna Ab initio Simulation Package (VASP).

64 eV and the oxygen chemical potential was obtained as muO = 1/2 Ecorr(O2). p. 7
Where the paper says it

96 V vs Li+/Li (3) 4Li+ + 4e − + 2CO2 + O2 ⇌ 2Li2CO3, 𝑈eq = 3. 82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1.

The carbon chemical potential was treated as a dependent quantity determined from the stability condition of Li2CO3 for given lithium and oxygen chemical potentials. p. 7
Where the paper says it

82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1. The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3.

Fluorine Substitution Energies

The fluorine substitution energies calculated using different fluorine reference reservoirs are summarized in Table S3. The LiF-based values therefore represent conservative upper-bound estimates under equilibrium bulk conditions. p. 8

Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

Under fluorine-rich reference conditions using F2 and OF2, the substitution energies decreased substantially to 1. p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

76–1.81 eV for the O(f) site and 2. p. 9
Where the paper says it

10 The oxygen atoms in the carbonate unit can be classified into two inequivalent sites, denoted as O(f) and O(e). According to the optimized structure, the C–O bond length at the O(f) site is 1.

02–2.08 eV for the O(e) site in Li2CO3. p. 9
Where the paper says it

10 The oxygen atoms in the carbonate unit can be classified into two inequivalent sites, denoted as O(f) and O(e). According to the optimized structure, the C–O bond length at the O(f) site is 1.

The corresponding values for Li2O2 become negative, ranging from −0. p. 2
Where the paper says it

Therefore, understanding the intrinsic defect energetics of Li2CO3 is crucial for clarifying the thermodynamic factors governing its stability across lithium-based electrochemical systems. The stability and decomposition behavior of Li2CO3 has become especially important in LOBs, where Li2CO3 forms alongside Li2O2 as discharge products.

89 to −0.84 eV (Table S3). p. 8
Where the paper says it

49–51 RESULTS and DISCUSSION Structural Properties and intrinsic defect energetics of Li2CO3 The structural model used in this study is a monoclinic Li2CO3 supercell with space group C2/c, as presented in Figure 1 together with the local structure of the carbonate unit and the distinct atomic sites considered for defect modeling. The Wyckoff positions and fractional coordinates of all inequivalent atomic sites in the optimized structure are summarized in Table S2.

Structural Details

Li2CO3 consists of ionic interactions between Li+ cations and CO3 2− anions. Within each CO3 2− unit, the carbon atom is coordinated by three oxygen atoms in a trigonal planar geometry, reflecting the covalent character of the C–O bonds. p. 8

Where the paper says it

Li2CO3 consists of ionic interactions between Li+ cations and CO3 2− anions. Within each CO3 2− unit, the carbon atom is coordinated by three oxygen atoms in a trigonal planar geometry, reflecting the covalent character of the C–O bonds.

The oxygen atoms in the carbonate unit can be classified into two inequivalent sites, denoted as O(f) and O(e). p. 9
Where the paper says it

10 The oxygen atoms in the carbonate unit can be classified into two inequivalent sites, denoted as O(f) and O(e). According to the optimized structure, the C–O bond length at the O(f) site is 1.

The C–O bond length at the O(f) site is 1. p. 9
Where the paper says it

10 The oxygen atoms in the carbonate unit can be classified into two inequivalent sites, denoted as O(f) and O(e). According to the optimized structure, the C–O bond length at the O(f) site is 1.

31 Å and is coordinated by three neighboring lithium ions. p. 9
Where the paper says it

According to the optimized structure, the C–O bond length at the O(f) site is 1. 31 Å and is coordinated by three neighboring lithium ions.

In contrast, the C–O bond length at the O(e) site is 1. p. 9
Where the paper says it

31 Å and is coordinated by three neighboring lithium ions. In contrast, the C–O bond length at the O(e) site is 1.

29 Å and is coordinated by two neighboring lithium sites. p. 9
Where the paper says it

In contrast, the C–O bond length at the O(e) site is 1. 29 Å and is coordinated by two neighboring lithium sites.

Defect Modeling Approach

Neutral vacancy defects were introduced into both pristine and fluorine-doped structures for each system. Complex vacancy defects, such as VLi2CO3 0, were evaluated using direct supercell calculations by removing one Li2CO3 formula unit from the lattice as a clustered vacancy. p. 6

Where the paper says it

In the fluorine-doped structures, vacancy sites were selected in the vicinity of the substituted fluorine atom to evaluate the local influence of fluorine substitution on vacancy formation energies. Complex vacancy defects, such as VLi2CO3 0 , were evaluated using direct supercell calculations by removing one Li2CO3 formula unit from the lattice as a clustered vacancy.

The supercells were optimized at the PBE level with fixed lattice parameters, allowing only internal atomic coordinates to relax. p. 6
Where the paper says it

This direct approach explicitly includes local structural relaxation and interactions among the constituent vacancies, instead of approximating the complex defect energy by summing the formation energies of individual charged vacancies. The supercells were optimized at the PBE level with fixed lattice parameters, allowing only internal atomic coordinates to relax.

The PBE-relaxed structures were used for single-point total energy calculations using the Heyd–Scuseria–Ernzerhof (HSE) functional to evaluate defect formation energies. p. 6
Where the paper says it

The supercells were optimized at the PBE level with fixed lattice parameters, allowing only internal atomic coordinates to relax. The PBE-relaxed structures were used for single-point total energy calculations using the Heyd–Scuseria–Ernzerhof (HSE) functional to evaluate defect formation energies.

The HSE mixing parameter was set to α = 0. p. 6
Where the paper says it

The PBE-relaxed structures were used for single-point total energy calculations using the Heyd–Scuseria–Ernzerhof (HSE) functional to evaluate defect formation energies. 43–45 The HSE mixing parameter was set to α = 0.

35, following previous work. p. 6
Where the paper says it

43–45 The HSE mixing parameter was set to α = 0. 35, following previous work,46 in which this value was adopted as an intermediate choice within the range of reported GW band gap estimates for Li2O2.

Voltage Dependence of Defects

The voltage-dependent defect behavior of Li2CO3 was generally governed by lithium-related defects, while carbon- and oxygen-related intrinsic defects remained thermodynamically unfavorable. In particular, the dominant formation of VLi− near and above the equilibrium voltage of Li2O2 suggests that Li+ extraction accompanied by electronic compensation may be thermodynamically favored during charging. p. 7

Where the paper says it

This voltage-dependent definition of 𝜇𝐿𝑖 was applied to both discharge products. The equilibrium voltages of Li2O2 and Li2CO3 were used as reference electrochemical conditions for evaluating voltage-dependent vacancy formation energies.

Lithium and carbon vacancy formation energies exhibit voltage dependence, while oxygen vacancy formation energies remain voltage independent. p. 1
Where the paper says it

Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium– oxygen batteries.

Selectivity of Fluorine Effect

In the fluorine-doped structure, the lithium and carbon vacancies formation energies were generally lowered compared to pristine state, whereas oxygen vacancy exhibits no significant change. This cation-selective destabilization, attributable to the strong electronegativity of fluorine perturbing the local electrostatic environment around lithium sites, not only lowers the thermodynamic cost of lithium removal but also partially weakens the carbonate framework itself. p. 3

Where the paper says it

29 In this respect, fluorine differs from many other dopants that would require deliberate synthetic incorporation. Second, owing to its similar ionic radius to oxygen and its exceptional electronegativity, fluorine substitution at oxygen sites is expected to selectively perturb the local bonding environment around lithium ions, potentially modifying cation sublattice stability without wholesale structural disruption.

The magnitude of the fluorine doping effect on lithium vacancy formation was approximately twice as large in Li2CO3 (ΔE ≈ −3.4 eV to −3.9 eV) as in Li2O2 (ΔE ≈ −1.4 eV to −1.7 eV). This suggests that fluorine doping modifies the defect energetics of Li2CO3 through a richer set of mechanisms than in Li2O2, including a reduction in carbon vacancy formation energy that is absent in Li2O2. p. 1
Where the paper says it

Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium– oxygen batteries.

Comparison to Li2O2

Fluorine doping reduced the lithium vacancy formation energies compared with the pristine state for both systems. In Li2O2, fluorine doping reduced the lithium vacancy formation energy from 1. p. 1

Where the paper says it

Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2.

02 eV to −0.66 eV (ΔE ≈ −1.68 eV) at the equilibrium voltage of Li2O2. p. 7
Where the paper says it

This voltage-dependent definition of 𝜇𝐿𝑖 was applied to both discharge products. The equilibrium voltages of Li2O2 and Li2CO3 were used as reference electrochemical conditions for evaluating voltage-dependent vacancy formation energies.

In contrast, the oxygen vacancy formation energy changed only marginally by 0. p. 5
Where the paper says it

In contrast, carbon vacancy formation leads to a slight increase in magnetization, suggesting further perturbation of the local electronic structure within the carbonate framework. Oxygen vacancy formation results in only a minor change in magnetization, corroborating the selective nature of fluorine doping toward cation vacancy formation rather than anion vacancy formation.

09 eV after fluorine doping and remained relatively high. p. 2
Where the paper says it

10–16 In LOBs, the accumulation of Li2CO3 on the cathode impedes charge transport and requires high charging overpotentials for decomposition, leading to severe losses in the energy efficiency and cyclability of LOBs. 6,8,17,18 To enhance charge transport and decomposability within the discharge products of LOBs, doping strategies have been actively investigated, particularly for Li2O2.

This indicates that the peroxide sublattice in Li2O2 maintains its structural integrity. p. 3
Where the paper says it

Second, owing to its similar ionic radius to oxygen and its exceptional electronegativity, fluorine substitution at oxygen sites is expected to selectively perturb the local bonding environment around lithium ions, potentially modifying cation sublattice stability without wholesale structural disruption. Third, while halogen doping effects including fluorine have been theoretically explored in Li2O2.

Summary of Findings

The study reveals that fluorine doping selectively reduces the lithium and carbon vacancies formation energies in both systems while exerting limited influence on oxygen vacancies. This cation-selective destabilization demonstrates that fluorine at oxygen sites has a stronger effect on cation sublattices rather than the anion sublattices. p. 5

Where the paper says it

In contrast, carbon vacancy formation leads to a slight increase in magnetization, suggesting further perturbation of the local electronic structure within the carbonate framework. Oxygen vacancy formation results in only a minor change in magnetization, corroborating the selective nature of fluorine doping toward cation vacancy formation rather than anion vacancy formation.

These findings imply that the experimentally reported influence of fluorinated electrolytes on Li2O2 discharge products extends to Li2CO3 by reducing the thermodynamic stability of its bulk structure and facilitating vacancy formation. p. 2
Where the paper says it

19–22 For example, dopants such as Ba, Na+ , and K+ have been reported to promote lithium vacancy formation and lower the charging overpotential. 23–28 Meanwhile, electrolyte additives have been used to modify the structural characteristics of Li2O2; fluorinated amide-based electrolytes have been experimentally shown to induce the formation of lithium-deficient Li2−xO2 and maintain a low charging voltage.

The results suggest that fluorine doping modulates the thermodynamic stability of both discharge products through a common mechanism of cation sublattice destabilization, while additionally destabilizing the carbonate framework in Li2CO3. p. 1
Where the paper says it

Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium– oxygen batteries.

These findings indicate that fluorine doping may contribute to reducing the thermodynamic cost associated with the formation of lithium-deficient Li2−xO2 phases. p. 3
Where the paper says it

First, from a practical standpoint, fluorinated electrolytes may serve as a source of fluorine-containing species under realistic battery operating conditions. Although the effects of fluorinated electrolytes on Li2CO3 have not been directly examined, their reported role in promoting the formation of lithium-deficient Li2−xO2 discharge products provides practical motivation for investigating fluorine as a dopant candidate.

The results are consistent with the possibility that fluorine doping strategies explored for Li2O2 may also be applicable to Li2CO3. p. 3
Where the paper says it

Third, while halogen doping effects including fluorine have been theoretically explored in Li2O2. 19,20 a systematic investigation of fluorine doping in Li2CO3 remains absent.

Supporting Information Details

The supporting information is available at the end of this preprint. Chemical potential calculations, structural information, and additional defect formation energy results (PDF). p. 8

Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

The calculation of chemical potential was determined from the stability condition of Li2CO3 for given lithium and oxygen chemical potentials. p. 7
Where the paper says it

82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1. The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3.

The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1. p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

The Li–C–O phase diagram is shown in Figure S1. Table S3 summarizes the fluorine substitution energies calculated using different fluorine reference reservoirs. p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

The chemical potential conditions used for defect formation energy calculations are summarized in Table S1(b). p. 8
Where the paper says it

49–51 RESULTS and DISCUSSION Structural Properties and intrinsic defect energetics of Li2CO3 The structural model used in this study is a monoclinic Li2CO3 supercell with space group C2/c, as presented in Figure 1 together with the local structure of the carbonate unit and the distinct atomic sites considered for defect modeling. The Wyckoff positions and fractional coordinates of all inequivalent atomic sites in the optimized structure are summarized in Table S2.

Chemical Potential Tables

Table S1 shows the chemical potential conditions used for defect formation energy calculations. Table S3 presents the fluorine substitution energies calculated using different fluorine reference reservoirs. p. 8

Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

Condition E denotes the equilibrium phases Li2CO3–Li2O–Li2O2.

Structural Data

Table S2 summarizes the Wyckoff positions and fractional coordinates of the optimized Li2CO3 structure. Table S1(a) summarizes the voltage-dependent chemical potential conditions used in the main analysis. p. 7

Where the paper says it

This voltage-dependent definition of 𝜇𝐿𝑖 was applied to both discharge products. The equilibrium voltages of Li2O2 and Li2CO3 were used as reference electrochemical conditions for evaluating voltage-dependent vacancy formation energies.

Table S1(b) summarizes the phase-diagram based Li–C–O chemical potential conditions for additional calculations. p. 8
Where the paper says it

9 defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF:20 𝜇F = 𝐸(LiF) − 1 2 𝐸(Li2O2 ) + 1 2 𝐸(O2 ) (5) This represents a conservative upper-bound estimate of the substitution cost under equilibrium bulk conditions; additional fluorine chemical potentials were calculated using F2 and OF2 as fluorine-rich reference states. The detailed chemical potential calculations, phase stability boundaries, and definitions of the alternative fluorine reference states are described in Supporting Information S1.

Experimental Context

The experimental value was taken as ∆Hfexp(Li2O2, 300 K) = −6.57 eV. The applied cell voltage U exists between the lithium metal anode and the discharge products formed at the cathode. The oxygen chemical potential was corrected to account for the well-known overbinding error of gas-phase O2 in DFT calculations. p. 7

Where the paper says it

96 V vs Li+/Li (3) 4Li+ + 4e − + 2CO2 + O2 ⇌ 2Li2CO3, 𝑈eq = 3. 82 V vs Li+/Li (4) The oxygen chemical potential was fixed by equilibrium with molecular O2 in the atmosphere, with a correction applied to account for the well-known overbinding error of gas-phase O2 in DFT calculations; the correction was derived from the experimental formation enthalpy of Li2O2 as described in Supporting Information S1.

The carbon chemical potential was determined by the thermodynamic stability condition of Li2CO3. The fluorine chemical potential was primarily defined using LiF as the thermodynamic reference reservoir, assuming equilibrium among Li2O2, O2, and LiF.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved defect energetics modeling for lithium carbonate (Li2CO3) by incorporating fluorine doping effects specifically on cation sublattice stability, which lowers lithium and carbon vacancy formation energies while weakening the carbonate framework itself. This allows AI systems to predict how fluorine doping modifies the thermodynamic stability of Li2CO3 and Li2O2 under battery-relevant conditions.

  2. Enhanced predictive capability for parasitic discharge product control by using voltage-dependent chemical potentials derived from Eq. (2) to model lithium and carbon vacancy formation energies in LOBs, as the paper shows that lithium and carbon vacancy formation energies exhibit voltage dependence. This enables AI to assess the thermodynamic favorability of Li+ extraction during charging versus direct carbonate framework dissociation.

  3. Tailored doping strategy design by quantifying the magnitude of fluorine doping effects, noting that the magnitude of the fluorine doping effect on lithium vacancy formation was approximately twice as large in Li2CO3 (Delta E approximately-3.4 eV to-3.9 eV) as in Li2O2. This allows AI to prioritize dopants based on their specific, system-dependent impact on defect energetics.

  4. Selective sublattice perturbation analysis by leveraging the finding that fluorine doping selectively reduces the lithium and carbon vacancies formation energies in Li2CO3, whereas oxygen vacancy formation is only weakly affected. This enables AI models to predict that fluorine at oxygen sites has a stronger effect on cation sublattices than anion sublattices.

Abstract

Lithium carbonate, Li2CO3, is a thermodynamically stable carbonate phase whose defect energetics are closely related to its stability and decomposition behavior in various lithium-based electrochemical systems. These properties of Li2CO3 are particularly important in lithium-oxygen battery environments. In these systems, Li2CO3 can form as a parasitic discharge product alongside Li2O2, the primary discharge product, leading to performance degradation. However, compared with Li2O2, the intrinsic defect thermodynamics of Li2CO3 and how chemical doping modifies its defect energetics remain insufficiently understood. In this study, first-principles calculations were performed to systematically analyze the intrinsic point-defect energetics of Li2CO3 and to evaluate the effects of fluorine doping on vacancy formation energies in Li2CO3 and Li2O2. Intrinsic defect analysis reveals that defect behavior is predominantly governed by lithium-related defects. Upon fluorine doping, lithium and carbon vacancy formation energies decrease selectively in Li2CO3, partially destabilizing the carbonate framework, while a reduction in lithium vacancy formation energy is also observed in Li2O2. These results suggest that fluorine doping modulates the defect energetics of both discharge products, potentially providing a thermodynamic basis for controlling the stability of Li2CO3 and Li2O2 under thermodynamic conditions representative of lithium-oxygen batteries.

Related papers